Immunocytochemical localization of glutamic acid decarboxylase in monkey striate cortex

Distribution of cytochrome oxidase in rat brain: studies with diaminobenzidine histochemistry in vitro and [14C]cyanide tissue labeling in vivo

Studies in experimental animals and post-mortem studies in humans have indicated that the level of the mitochondrial enzyme cytochrome oxidase within brain anatomical pathways is regulated by the long-term functional use of those pathways. To study this relationship, we have measured cytochrome oxidase spectrophotometrically in punch biopsies from different brain regions of rat. We compared these assays against results from the diaminobenzidine histochemical technique. We found a high degree of correlation (r = 0.90) between the density of diaminobenzidine reaction product and enzyme activity. This validates the usefulness of the diaminobenzidine technique for anatomical localization and measurement of this enzyme. To study the feasibility of using radioactive cyanide as an in vivo ligand of cytochrome oxidase, we performed quantitative autoradiographic analysis of rat brains of animals given an intravenous bolus injection of [14C]cyanide. Analysis of the arterial blood curve indicated a complex redistribution of cyanide between red blood cells, plasma, and tissues. Brain labeling reached peak levels at 1 min and then fell despite rising concentrations of free plasma cyanide. Analysis of autoradiographic images revealed good anatomical resolution. The density of labeling in individual structures over time failed to show a strong correlation with cytochrome oxidase activity or diaminobenzidine reaction product.

Picture processing techniques applied to autoradiographic studies of visual cortex

Picture processing techniques are applied to 2-deoxyglucose autoradiographs of sections from striate cortex and to patterns resulting from staining these sections for cytochrome oxidase. This procedure allows computer identification of deoxyglucose active and inactive regions in the autoradiographs and cytochrome active and inactive regions in the stain patterns. Subsequently, the topographical relationship between these patterns can be quantitatively analyzed by means of overlap and density distribution measures and can be displayed using color enhanced graphics. The processing techniques have been applied in studies of the functional organization of visual cortex in primates. Computer graphic techniques have allowed implementation of split-field presentations of stimuli in deoxyglucose experiments. An application of this split-field technique for presenting multiple-stimuli to distinct parts of the visual field is described and an autoradiograph from a split-field experiment is shown.

A decrease in the number of GABAergic somata is associated with the preferential loss of GABAergic terminals at epileptic foci

Previous studies have indicated that a loss of GABAergic terminals occurs at epileptic foci. The present study was undertaken to investigate if this loss is associated with a loss of GABAergic neuronal somata. Seven juvenile monkeys (M. mulatta) received alumina gel injections to the pre-central gyrus of the left cerebral hemisphere to produce epileptic foci. Four of these monkeys were chosen for further quantitative study. One was sacrificed prior to seizure onset ('pre-seizure'), one had seizures for 3 days ('acute'), and two had a seizure record of one month ('chronic'). Sections of tissue from the epileptic cortex and from the contralateral, non-epileptic cortex were processed for glutamate decarboxylase (GAD) immunocytochemistry at the light microscopic level. Quantitative analysis revealed that a loss of GAD-positive neuronal somata ranging from 24 to 52% occurred at epileptic foci for all monkeys. This decrease was significant (P less than 0.01) for the two chronic monkeys. There was also a slight decrease in GAD-positive neurons 1 cm distal to the focus ('parafocus') in the chronic monkeys, but not in the acute or pre-seizure animals. In addition, small GAD-positive somata (50-150 micron2) were more severely decreased in number at epileptic foci than larger ones (200-250 micron2). As an experimental control, an additional monkey was given a surgical lesion in area 4 of one cerebral hemisphere. It did not display seizure activity prior to sacrifice and did not show a loss of GAD-positive neurons proximal to the control lesions. The results of this study indicate that a loss of GABAergic neuronal somata is associated with a loss of GABAergic terminals at epileptic foci, and that this loss may be more specific for the small GABAergic neurons.

Ultrastructural immunocytochemistry of gamma-aminobutyrate in the cerebral and cerebellar cortex of the rat

gamma-Aminobutyrate containing structures in the cerebral and cerebellar cortex of the rat were visualized by an immunocytochemical method using glutaraldehyde fixation and an antiserum developed against a gamma-aminobutyrate-glutaraldehyde-protein conjugate. Labelled elements (perikarya and cell processes) were observed to be distributed throughout the layers of the cerebral cortex in a pattern similar to that described using glutamate decarboxylase immunocytochemistry. The morphological features of many immunoreactive cell bodies were typical of stellate neurons. In the cerebellar cortex, Purkinje, basket, Golgi and stellate, cell bodies were found to be immunoreactive along with numerous labelled neuronal processes. At the ultrastructural level, the labelled processes in both areas corresponded to immunoreactive dendrites and fibres. Labelled synaptic boutons, generally of the symmetrical type, could also be seen in contact with positive or negative cell bodies and dendrites. In the cerebellum, glomeruli could be clearly identified including mossy fibres surrounded by unlabelled dendrites in contact with immunoreactive terminals. At the subcellular level in both brain regions, the areas occupied by the Golgi apparatus were never labelled, although the nuclei had varied reactions. The strong glutaraldehyde fixation that limits the diffusion of gamma-aminobutyrate limits also antibody diffusion. However, this fixation is compatible with a good morphological preservation and should enable immunocytochemistry studies to be compared to other methods such as autoradiography.

Differences between strychnine and penicillin epileptogenesis suggest a laminar organization of neocortical inhibition

Transient foci of epileptiform alteration in neuronal population activity were induced by microinjection of strychnine sulfate into different layers of cat striate cortex. Potentials evoked by visual field-specific photic stimulation were recorded from microelectrodes at the injection site and in adjacent laminae. Epileptogenesis, characterized by an enhancement of the normal primary response followed by the development of a large late potential, occurred only with strychnine injections into superficial pyramidal layers 2 and 3. By contrast, stellate layer 4 has been shown to be most susceptible to epileptogenic effects of penicillin and bicuculline. Since disinhibitory convulsants should be most effective where their actions antagonize the prevalent type of inhibition, these findings suggest that there may be a laminar segregation of neocortical inhibition, possibly glycine-mediated in layers 2-3 and probably gamma aminobutyric acid (GABA)-mediated in layer 4.

Correlation between cytochrome oxidase staining and the uptake and laminar distribution of tritiated aspartate, glutamate, gamma-aminobutyrate and glycine in the striate cortex of the squirrel monkey

The cellular uptake and laminar distribution of tritium-labeled gamma-aminobutyrate, aspartate, glutamate and glycine were examined in the primary visual cortex of squirrel monkeys. The purpose was to correlate the distribution of these labeled neurons with their level of cytochrome oxidase activity, particularly in laminae II-III (puffs) and adjacent non-puff regions. In general, tritium-labeled neurons that had either high or low levels of cytochrome oxidase activity were present in all laminae with each amino acid tested; however, their density varied between laminae and with the amino acid injected. Specifically, in laminae II-III, very few neurons were labelled with either of the putative excitatory amino acids (aspartate and glutamate). An increased uptake for both was observed in lamina IVC, with the greatest increase for each occurring in laminae V and VI. Significantly more neurons in each lamina were labeled with the putative inhibitory transmitters (gamma-aminobutyrate and glycine) than with either aspartate or glutamate. gamma-Aminobutyrate-labeled neurons were more prevalent in lamina II than III, and an increase in labeling was observed in laminae IV-VI, with the most prominent increase found in laminae V and VI. Glycine-labeled neurons were larger, more uniformly distributed and more abundant throughout all cortical laminae than those labeled with the other amino acids. Significantly more gamma-aminobutyrate- and glycine-labeled neurons were found in the puff regions than in the non-puff areas. No difference was found between puff and non-puff regions for the tritium-labeled leucine controls. Labeled neurons included stellate, fusiform and pyramidal-shaped cells of varying sizes; however, gamma-aminobutyrate-labeled pyramidal cells were not observed outside of the intense injection site. Large glycine-labeled cytochrome-oxidase-reactive pyramidal cells (24-32 micron in diameter) were present at the boundary between laminae V and VI. In addition, a row of large glycine-labeled, fusiform neurons were present in lamina IVB. With each amino acid injected, the tritium-labeled neurons that were darkly reactive for cytochrome oxidase were, on average, larger than the tritium-labeled neurons that were only lightly reactive for cytochrome oxidase. Thus, each of the four amino acids tested had its unique pattern of distribution in the primate striate cortex. Whether one or all of them served as neurotransmitter(s) for distinct neuronal groups is beyond the scope of this study. Glycine, in particular, might be used in part or in whole for metabolic purposes.(ABSTRACT TRUNCATED AT 400 WORDS)

Evidence for specific immunolysis of nerve terminals using antisera against choline acetyltransferase, glutamate decarboxylase and tyrosine hydroxylase

Synaptosomes prepared from striatum or cerebral cortex of rat brain were incubated with antibodies raised against three neurotransmitter biosynthetic enzymes, choline acetyltransferase, glutamate decarboxylase and tyrosine hydroxylase in the presence or absence of complement. Immunolysis was first assessed by measuring the release of lactic dehydrogenase or reduction in potassium from synaptosomes, and lysis of neurochemically specific subpopulation of synaptosomes was detected by measuring release of either transmitters, their biosynthetic enzymes or by blockade of sodium-dependent uptake of transmitter or precursor. In both striatum and cortex, antibodies to choline acetyltransferase lysed only cholinergic while those against glutamate decarboxylase only lysed GABAergic nerve terminals. Antibodies against tyrosine hydroxylase lysed only the dopaminergic terminals in striatum but not noradrenergic terminals in cortex. The lysis occurred only in the presence of complement, and was never observed in the absence of complement. The studies indicate that antibodies to the neurotransmitter biosynthetic enzymes recognize antigens in the synaptosomal membrane specific only to neurons harboring the transmitters. The results suggest that the antibody-positive peptides in the membrane and neurotransmitter biosynthetic enzyme share common antigenic sites, probably common peptides.

Functional organization of the cat's visual cortex after prenatal interruption of binocular interactions

The functional consequences of interrupting in utero binocular interactions were studied by recording from single cells in area 17 of adult cats that had one eye removed at least 2 wk before birth. In these animals all cortical neurons could be driven by the remaining eye, and in tangential microelectrode penetrations, sequences of neurons containing a full 180-degree cycle of preferred orientations were encountered. Other response properties of cortical neurons in the prenatally enucleated animals were also normal with the notable exception that the dimensions of receptive fields were significantly smaller when compared with those of control animals. Our results indicate that orientation columns in the visual cortex can develop independently of ocular dominance columns, and they suggest that interruption of binocular interactions during prenatal development of the visual pathways may enhance the resolving power of the remaining eye.

Complex-unoriented cells in a subregion of primate area 18

In primates, both the primary and secondary visual cortical areas can be subdivided histologically by staining for the mitochondrial enzyme cytochrome oxidase. In the primary visual cortex (area 17, the first cortical receiving area for visual information) these histological differences correspond to functional subdivisions, cytochrome-dark regions being concerned with information about colour and cytochrome-light regions concerned with form. Here we report that the second visual area, area 18, which receives its main cortical input from area 17 (refs 7,8), similarly has functional subdivisions that correspond to the cytochrome oxidase staining pattern. In area 18 the segregation between form and colour is maintained, reinforcing our notion that form and colour information follow parallel pathways. The specific differences between cells in areas 17 and 18 suggest that a possible step in hierarchical information processing is spatial generalization, analogous to the difference between simple and complex cells.

Alteration of putative amino acid levels and morphological findings in neural tissues of methylmercury-intoxicated mice

Methylmercury chloride was administered PO to male Kud: ddY mice at a dose of 5 mg/kg/day for 20 days. The contents of taurine, aspartate, glutamate, glycine, and gamma-aminobutyric acid were determined in tissue and crude synaptosomal (P2) fraction of cerebellum, cerebral cortex, and spinal cord of methylmercury-treated mice with or without ataxia. In the cerebellum of ataxic mice, increased levels of taurine and glycine were found in the tissue and P2 fraction, and increased levels of glutamate were found in the P2 fraction. In the cerebral cortex, the levels of gamma-aminobutylic acid decreased in the tissue and in the P2 fraction of ataxic mice, but increased levels were found in the tissue of non-ataxic mice. A decreased aspartate level in the cerebral cortex of ataxic mice and an increased taurine level in the cerebral cortex of non-ataxic mice were also found. In the spinal cord of ataxic mice, taurine increased in the tissue and in the P2 fraction. Glutamate level decreased in the spinal cord of ataxic mice, but increased in the P2 fraction of non-ataxic mice. Increased glycine levels in the P2 fraction of the spinal cord were also found in non-ataxic mice. Histologically, some degenerative changes were demonstrated in the cerebral and cerebellar cortices of ataxic mice. Such changes were also present to a mild degree in non-ataxic mice. In conclusion, methylmercury treatment altered the levels of putative neurotransmitter amino acids in neural tissue of mice. These alterations might be caused by specific neural cell dysfunction and could be related to the appearance of ataxia.

Modulatory and inhibitory processes in the visual cortex

The functional role of three putative neurotransmitter systems in the visual cortex is compared; the GABAergic inhibitory interneurons, the interneurones containing somatostatin and the cholinergic input originating from the nucleus basalis of Meynert (nbM). Evidence is presented to support the role of GABAergic processes in the generation of the functional structure of the visual cortex and the view that the cholinergic input exerts a neuromodulatory influence enhancing stimulus selective responses. Although the neuropeptide somatostatin produces facilitatory and inhibitory effects on visual cortical cells there is no clear functional pattern to its action. The possible significance of this data and the interaction of SSt with GABA is discussed in the light of evidence that they may coexist in some cells.

GABAergic inhibition and orientation selectivity of neurons in the kitten visual cortex at the time of eye opening

Effects of iontophoretic application of gamma-aminobutyric acid (GABA) and its antagonist, N-methyl-bicuculline (BIC), on visual responses of striate cortical neurons were studied in kittens 6-13 days old. Visually responsive cells were classified into three groups, i.e. orientation-selective, orientation-bias and nonoriented cells. In almost all of the orientation-selective cells, their responses were completely suppressed by GABA while the majority of the others were not significantly or only weakly suppressed. An application of BIC abolished or reduced the selectivity of all the orientation-selective cells but did not affect any of the nonoriented cells tested. These results suggest that GABAergic inhibition already operates on a group of cortical neurons to make them orientation-selective at the time of eye opening, but such an action of GABA on other groups of neurons develops later.

Inhibitory modulation of cat somatosensory cortex: a pharmacological study

In anesthetized preparations, GABA and taurine produced rapid, reversible inhibition of the negative component (N20) of the primary somatosensory evoked potential (SEP) without effect on the earlier positivity (P11). This effect was also produced by low doses of 4-aminopyridine. Neither bicuculline or picrotoxin antagonized these drug effects. A predominance of type B GABA receptors in the superficial layers of the somatosensory cortex is proposed.

Neuropeptide-containing neurons of the cerebral cortex are also GABAergic

Neurons in the cat and monkey cerebral cortex were stained immunocytochemically for glutamic acid decarboxylase (GluDCase; L-glutamate 1-carboxy-lyase, EC 4.1.1.15), somatostatin (SRIF), neuropeptide Y (NPY), and cholecystokinin octapeptide (CCK). In all areas of cortex examined (somatic sensory, motor, parietal and visual areas), neurons displaying immunoreactivity for each of these molecules were nonpyramidal cells. Co-localization of GluDCase immunoreactivity with peptide immunoreactivity in the same cells was demonstrated by (i) the antibody elution method, staining the same cells by immunofluorescence, first for a peptide and then for GluDCase; (ii) double staining of the same sections with sheep anti-GluDCase and rabbit anti-peptide antisera, the bound antibodies being localized by rhodamine-conjugated donkey anti-sheep and fluorescein-conjugated swine anti-rabbit secondary antisera. With both procedures, cell bodies immunoreactive for GluDCase and for each of the peptides were found in all areas of cortex examined. With double labeling on single sections, it was found that all CCK-, SRIF-, and NPY-immunoreactive cells in cat cortex and 90%-95% in monkey cortex are also GluDCase positive. Many more cells, however, are immunoreactive for GluDCase alone. GluDCase was co-localized with CCK, SRIF, or NPY not only in cell somata, but also in small punctate structures, which are likely to be axon terminals. From the data gained in previous electron microscopic studies, we postulate that neurons displaying GluDCase- and CCK-like immunoreactivity are a class separate from those displaying GluDCase- and SRIF-like immunoreactivity. NPY, however, is co-localized with SRIF immunoreactivity. These results imply that classes of cortical interneuron contain a conventional neurotransmitter (gamma-aminobutyric acid) and a neuromodulator (one of the peptides).

Characteristics and distribution of muscimol binding sites in cat visual cortex

In vitro receptor binding techniques were used to study the characteristics and distribution of [3H]muscimol binding sites in cat visual cortex. [3H]muscimol, a specific GABA agonist, labeled a single population of binding sites with a Kd of 18 nM. Specific binding was saturable, reversible, and was blocked by the addition of GABA or (+)-bicuculline. Autoradiograms revealed that the highest density of [3H]muscimol binding sites occurred in cortical layer IV. Little variation between the various visual cortical areas was noted in contrast to marked regional heterogeneity within subcortical structures.

Intralaminar neurochemical distributions in human midtemporal cortex: comparison between Alzheimer's disease and the normal

The intralaminar distributions of transmitter and nontransmitter enzyme activities and amino acid levels were determined in the midtemporal cortices from normal individuals and established cases of Alzheimer's disease. In the normal, choline acetyltransferase (CAT) and acetylcholinesterase (AChE) activities were relatively high in the outer cortical layers, particularly, for CAT, in the two granular layers (II and IV). Both activities were reduced in Alzheimer's disease at all, although generally most extensively in the outer and middle layers of the grey matter whereas activities were near normal in the white matter. Further, the enzyme distribution patterns of these cholinergic activities were also disrupted in Alzheimer's disease and the activity of CAT throughout the cortex was generally reduced to that found in the white matter. No such differences in distribution were found for two other enzymes, pseudocholinesterase and lactate dehydrogenase. Assessment of the gamma-aminobutyric acid (GABA) system in the normal revealed a much more extensive intralaminar variation in the enzyme, glutamate decarboxylase, compared with the level of GABA itself. In contrast with the cholinergic enzymes, neither the levels nor intralaminar patterns of GABA were altered in Alzheimer's disease. From an analysis of free amino acids at the different cortical levels, the cortical pattern of glutamic acid in the normal was different from that for GABA, aspartic acid, or nontransmitter amino acids such as alanine. Neither of the putative amino acids, glutamate or aspartate, was altered in Alzheimer's disease. These findings demonstrate the relatively selective nature of microchemical changes occurring in the cortex in Alzheimer's disease and suggest that a functional abnormality in cholinergic input to the outer neocortical layers (I-IV) with predominantly receptive and associative functions may be an important feature of the disease.

Interhemispheric connections of the visual cortex in the grey squirrel (Sciurus carolinensis)

The total pattern of visual callosal connections was studied in the grey squirrel by using the Fink-Heimer technique for axonal and terminal degeneration and the autoradiographic and horseradish peroxidase techniques for axonal transport. The pattern of terminations was correlated with architectonic landmarks. The results show that callosal terminations are distributed in a complex fashion within the visual cortical areas. The major terminations form a band in area 17 along its border with area 18. This band is contiguous rostrally with the callosal terminations in area L that extend caudomedially onto the medial wall of the hemisphere. Caudally the band in area 17 wraps around the ventral aspect of the occipital pole and ends medially at the level of the hippocampus. This band exhibits a distinct periodicity in the density of terminations. The callosal terminations in area 18 are usually found along the lateral and medial borders and are concentrated in discrete patches. The pattern in area 19 exhibits two or three primary patches and only loosely corresponds to the borders of the area. Few callosal terminations are found in area 19p and the posterior temporal area, Tp, while the intermediate temporal area, Ti, receives an extensive input. The laminar distribution of callosal terminations is different in each area studied. Characteristically, area 18 has dense terminations in layers III, II, and the inner one-half of layer I, with less dense terminations in layers V and VI, and sparse terminations in layer IV. Area 17 has a similar pattern in the supragranular and infragranular layers but also has dense terminations in layer IV. The patterns in area 19 are intermediate between these extremes but are more similar to those in area 17. The cells that give rise to the callosal projections were found primarily in layers III and V and occasionally in layers II, IV, and VI. The distribution of the callosal efferent neurons is more extensive than the areas of terminations. The distribution of callosal terminations suggests that the organization of visual cortical areas in the grey squirrel is more complex than had been previously recognized. This finding is discussed with reference to the general organization of the mammalian visual cortical areas, and a need for more extensive analyses of visual cortical areas in the grey squirrel, particularly with respect to extrastriate visual areas, is indicated.

Effect of impulse blockage on cytochrome oxidase activity in monkey visual system

Cytochrome oxidase (cytochrome c oxidase; ferrocytochrome c: oxygen oxidoreductase, EC 1.9.2.1) has been introduced as an oxidative metabolic marker for neurones in the central nervous system. Previous studies have shown that mature neurones remained sensitive to altered functional demands, and that both developing and adult neurones responded to sensory deprivation or deafferentation by reducing their cytochrome oxidase (Cyt. Ox.) activity. More recently, we showed that the blockage of retinal impulse transmission with tetrodotoxin led to a reversible reduction in Cyt. Ox. staining of affected lateral geniculate (LGN) and striate neurones in adult cats. The present study sought to extend these findings to adult monkeys, where Cyt. Ox. 'puffs' or 'blobs' are uniquely present in the visual cortex. We found that, while the retina remained histologically intact, with only moderate decreases in Cyt. Ox. staining of large ganglion cells and the two plexiform layers, subtle changes occurred in the LGN as early as 1 day post-tetrodotoxin injection, and clear reduction in enzyme levels was evident in both the LGN and the visual cortex by 3 days. Changes became progressively more severe up to 4 weeks post-injection. Within area 17, alternating bands of high and low Cyt. Ox. staining occurred in lamina IV, with alternating rows of dark and lightly reactive puffs superimposed in exact register. Thus, the mature visual neurones in the primate remain extremely sensitive to the cessation of retinal impulse transmission, and plastic metabolic changes occur through several synapses along the sensory pathway.

Monoclonal antibody that identifies subsets of neurones in the central visual system of monkey and cat

Striking correlations between structure and function are found in the visual cortex of Old World primates. These include the co-localization of glutamic acid decarboxylase (GAD, the biosynthetic enzyme of the inhibitory neurotransmitter, gamma-aminobutyric acid) with the mitochondrial enzyme, cytochrome oxidase (CO) in functionally distinct subcompartments of ocular dominance columns. We report here immunocytochemical studies with a monoclonal antibody (CAT 301) showing that the antibody recognizes an uncharacterized antigen on surfaces of some neurones in certain layers of the monkey striate cortex (area 17), and in certain parts of the cat and monkey dorsal lateral geniculate nuclei (LGN). Patches of immunocytochemically stained neurones and neuropil, apparent in layers III, IVB and VI of the striate cortex of normal monkeys, become even more clearly delineated in animals from which one eye has been removed. The antibody-stained patches in the three layers line up radially with one another in lines passing through the centres of ocular dominance columns (demonstrable by CO staining in layers IVA and IVC). In layers III and VI the patches coexist with CO-positive patches and, in the horizontal dimension, both antibody and CO-positive patches are aligned to form rows. Stained neurones in the monkey LGN are primarily in the magnocellular layers and in the cat LGN are confined to laminae A and A1, the inter-laminar plexuses, the perigeniculate nucleus and the medial inter-laminar nucleus. The antigen we have localized is associated with particular cell populations, some of which may correspond to a specific, physiological class.

Mapping of cytochrome oxidase patches and ocular dominance columns in human visual cortex

The cytochrome oxidase stain was applied to autopsy specimens of human brain. In primary visual cortex patches of darker enzyme staining were present in layers II, III, IV b, V, and VI. The patches were oval, about 400 by 250 µm, with a density of one patch per 0.6-0.8 mm 2 of cortex. They were organized into rows spaced about 1 mm apart, intersecting the 17-18 border at right angles. The patches also stained preferentially for AChE activity. The lateral geniculate body was examined in two patients who died many years after losing one eye as adults. In atrophied laminae cytochrome oxidase activity was severely reduced. In the visual cortex from three cases after monocular enucleation, regular alternating light and dark columns of cytochrome oxidase activity were visible in layer IV c, presumably corresponding to ocular dominance columns. In two cases their pattern was reconstructed over 200-400 mm 2 of striate cortex. The columns appeared as roughly parallel slabs about 1 mm wide, oriented perpendicular to the 17-18 border as in the macaque. In the upper layers light and dark rows of patches were present, which fit in register with the light and dark ocular dominance columns below. In layer IV the ocular dominance columns were also visible in Nissl stained sections as a consequence of secondary anterograde transneuronal degeneration. Darker Nissl stained columns matched lighter cytochrome oxidase stained columns corresponding to the missing eye. Quantitative measurements demonstrated a 10% loss of mean cell area and 35% increase in cell density in ocular dominance columns belonging to the missing eye, which accounts for their darker appearance in the Nissl stain. Patches were not present in a foetus at six months gestation. However, they were clearly formed in a six month old baby, although they appeared smaller and more closely spaced than in the adult. These results show that patches are present in man, in addition to other primates, although they appear proportionately larger. Ocular dominance columns are also present, in common with certain species of primates like the macaque, baboon and galago. Cytochrome oxidase histochemistry promises to be a useful technique for mapping anatomical features of the human brain post mortem .

The distribution of hexokinase compared to cytochrome oxidase and acetylcholinesterase in the somatosensory cortex and the superior colliculus of the rat

The distribution of hexokinase, a general glycolytic enzyme, was compared to that of cytochrome oxidase and acetylcholinesterase (AChE) in the somatosensory cortex and the superior colliculus of the rat. The vibrissal barrel fields of the adult rat contain high hexokinase and cytochrome oxidase activity and low AChE activity. In the superior colliculus, hexokinase activity was highest in cell layers and discrete foci of intense activity were observed in the deep grey layer. This distribution was different from that of both cytochrome oxidase and AChE in this structure.

A model for generalization and specification by single neurons

A rule for environmentally dependent modification of the neuronal state is examined. Under the rule, the neuron selects a trigger feature that matches either a particular pattern in the stimulus set, or the most common pattern component, depending on a certain parameter. Thus a neuron may evolve to respond to its stimulus environment in one of two capacities, namely specification or generalization. Neurons of the former variety are labelled "S-cells"; and those of the latter, "G-cells". In the model, synaptic modification is modulated by two postsynaptic mechanisms which act antagonistically to strengthen or weaken the synaptic connectivities. The functional dependence of these mechanisms on the postsynaptic activity is shown to determine whether the neuron acts as an S-cell or a G-cell. A circuit is proposed for a module that consists of a G-cell and several S-cells sharing a common set of inputs. By inhibiting the G-cells, the S-cell acts as a contrast-enhancing element, increasing their specificities for individual patterns in the stimulus set. The output from the module is a recoded representation of the environment with respect to its general and distinctive features.

Quantitative light and electron microscopic analysis of cytochrome oxidase-rich zones in the striate cortex of the squirrel monkey

Cytochrome oxidase activity was examined in the striate cortex (area 17) of squirrel monkeys at both the light and ultrastructural levels. Two prominent bands of reactivity were found in 4A and 4C with intermittent puffs of cytochrome oxidase reactivity in laminae 2 and 3. These puffs, spaced 0.5 mm apart, were in register with intermittent concentrations of activity in laminae 4B, 5, and 6. A thin band of reactivity was observed in lamina 1. The upper portion of 4C beta was less reactive than 4C alpha or the lower portion of 4C beta. Reactive neurons included stellate cells in all laminae and pyramidal cells in laminae 2 through 4B, 5, and 6. A row of large reactive pyramidal cells was observed in upper lamina 6. More reactive neurons were found in the puffs (laminae 2 and 3) than were observed in interpuff regions, and the reactive neurons were significantly larger than the nonreactive neurons. Reactive neurons contained two to three times as many reactive mitochondria as did the nonreactive neurons and often had indented nuclei. Based on the number of darkly or highly reactive, moderately reactive and lightly reactive mitochondria, puff regions were significantly different from nonpuff regions; there were approximately two times as many darkly reactive mitochondria in puff regions as compared to a similar nonpuff area. The majority of mitochondria (32% in puff; 44% in nonpuff) were found to reside in the dendritic profiles, which also contained the majority of highly reactive mitochondria. In a separate analysis, the total area of highly reactive mitochondria within puff regions was found to be twice the total area of highly reactive mitochondria in a comparable nonpuff region. An analysis of synapses showed that there were more asymmetrical synapses in both puff and nonpuff regions (55% and 54%, respectively) than symmetrical ones (45% in puff and 46% in nonpuff). There was an increase in mitochondrial reactivity in both asymmetrical and symmetrical synapses in the puff areas; however, the increased reactivity within asymmetrical terminals was significantly greater than that within symmetrical ones. Several somatodendritic synapses were observed and they were all of the symmetrical variety. Axospinous contacts were primarily of the asymmetrical type; however, symmetrical axospinous synapses were observed and were typically seen in association with an asymmetrical synapse. It was concluded that cytochrome oxidase activity is localized primarily within the dendritic profiles in both puff and nonpuff regions.(ABSTRACT TRUNCATED AT 400 WORDS)

Retinotopy and orientation columns in the monkey: a new model

A model is presented in which orientation columns arise directly out of retinotopy. According to the model, iso-orientation lines are arrayed radially around nodal centers which correspond to cytochrome oxidase patches. The nodal centers form a square matrix superimposed upon the map of ocular dominance stripes. In the supragranular layers horizontal iso-orientation lines run down the centers of ocular dominance stripes, with vertical iso-orientation lines crossing perpendicularly. Diagonal orientations (45 degrees and 135 degrees) are represented as alternating iso-orientation zones at the centers of the interstices in the matrix (internodal centers). Preferred orientations in the infragranular layers are reversed with respect to the supragranular layers. The model is consistent with new data concerning ocularity and preferred orientation in systematic penetrations through striate cortex, and helps to explain some previously puzzling features of the relationship between ocular dominance columns, orientation columns and retinotopy.

The visual field representation in striate cortex of the macaque monkey: asymmetries, anisotropies, and individual variability

The topographic organization of striate cortex in the macaque was studied using physiological recording techniques. Results were displayed on two-dimensional maps of the cortex, which facilitated the quantitative analysis of various features of the visual representation. The representation was found to be asymmetric with more cortex devoted to lower than to upper fields. Over much of striate cortex the representation is anisotropic, in that the magnification factor depends upon the direction along which it is measured. There is considerable individual variability in these features as well as in the overall size of striate cortex. Outside the fovea, the cortical representation shows only modest deviations from a logarithmic conformal mapping, in which the magnification factor is proportional to the inverse of eccentricity in the visual field. Comparison of receptive field size with cortical magnification was used to estimate the "point image size" in the cortex (i.e. the extent of cortex concerned with processing inputs from any given point in the visual field). Our evidence supports a previous report that point-image size varies significantly with eccentricity. This is of interest in relation to anatomical evidence that the dimensions of columnar systems in striate cortex are largely independent of eccentricity.

Quantitative light and electron microscopic analysis of cytochrome oxidase-rich zones in V II prestriate cortex of the squirrel monkey

Area 18 of V II of the prestriate cortex of the squirrel monkey was examined at both the light and electron microscopic (EM) levels for cytochrome oxidase (C.O.) activity. At the 17/18 border, the intense C.O. staining of lamina 4 abruptly ended and a new pattern continued for approximately 6 mm into the adjacent prestriate cortex. Here, periodic puffs of high C.O. activity appeared in laminae (lam.) 2 and 3, with the highest activity in lower 3 (3B) extending slightly into upper 4. There was a hint of a columnar pattern in that lam. 4 and especially 5 below the puffs were slightly more reactive than adjacent areas. A thin band of activity could also be seen in upper 5 (5A) and another one between 5 and 6. Tangential sections revealed that the puffs were arranged in alternating wide and narrow rows that radiated orthogonally from the 17/18 border. The puffs in the wider rows tended to be larger (700-1,100 micrometers in diameter) than those in the narrow rows (400-890 micrometers in diameter). The center-to- center spacing between the puffs was approximately 1,100 micrometers. Both C.O.-reactive and nonreactive stellate and pyramidal cells were found between lam. 2 and 6. Quantitatively analysis of the supragranular layers indicated that the mean area of reactive neurons was significantly larger than that of nonreactive neurons in both the puffs and interpuff (nonpuff) regions. The relative density of reactive neurons was also significantly greater than that of nonreactive neurons, and was highest within the puffs. At the EM level, reactive neurons were medium to large pyramidal cells as well as medium-sized stellates with mild to severely indented nuclei and darker cytoplasm filled with reactive mitochondria. The majority of small stellates with scanty cytoplasm and few mitochondria were nonreactive. Extensive quantitative analysis of mitochondria number and level of reactivity in different neuronal profiles indicated that the number and area of darkly reactive mitochondria was significantly higher in the puffs than in the nonpuffs, and that the majority of them resided in dentritic profiles. Between a third to half of the mitochondria in axonal profiles were darkly reactive, the frequency being slightly higher in profiles with flattened vesicles making symmetrical synapses than those with round vesicles making asymmetrical synapses. Mitochondria in axonal trunks and myelinated axons contributed to only a small percentage of the total population. Glial cells, in general, were not very reactive.(ABSTRACT TRUNCATED AT 400 WORDS)

Developmental changes in the relationship between type 2 synapses and spiny neurons in the monkey visual cortex

This study continues an exploration of synaptic development in the primary visual cortex of the monkey (Macaca nemestrina). In a prior study (Mates and Lund, '83a), we observed that type 2 synapses on the cell bodies of spiny stellate neurons of lamina 4C appeared not only to increase in number during early postnatal development but also subsequently decreased during maturation. Using quantitative, stereological electron microscopic methods, we examined the maturation of this synapse population from embryonic day 159 to adult, on spiny stellate neurons of 4C alpha and beta and, for comparison, on pyramidal neurons in upper and lower lamina 6. Tissue was also taken for comparison from two animals reared to 8 weeks of age with binocular eyelid closure from birth. We confirmed that a marked increase and subsequent decrease occurred in this somal type 2 synapse population on both neuron populations. However, due to the infrequency of the smooth dendritic neurons (approximately 5% of the neuron population) giving rise to the type 2 contacts, and due to expansion of the neuropil during maturation increasing intercell distances against constant volume of the type 2 axon arbors, it is concluded that the decrease in type 2 somal synapses may represent a redistribution to dendrites rather than loss from the neuropil. Cells of lamina 4C beta (receiving input from the parvocellular lateral geniculate nucleus-LGN) show a slower initial accumulation of type 2 contacts compared to neurons of lamina 4C alpha (receiving input from magnocellular LGN), or to pyramidal neurons of lamina 6.(ABSTRACT TRUNCATED AT 250 WORDS)

Columnar aggregation of prefrontal and anterior cingulate cortical cells projecting to the thalamic mediodorsal nucleus in the monkey

The columnar arrangement of the prefrontal and anterior cingulate cortex cells that project to the thalamic mediodorsal (MD) nucleus in rhesus and Formosan monkeys was studied by injecting horseradish peroxidase into the MD nucleus. Labelled cells appeared ipsilaterally in layers V and VI in the dorsolateral, medial, and orbital prefrontal cortex, and in layer VI in the anterior cingulate cortex. Labelled cells were aligned in the radially oriented cords of neurons and grouped into clusters, 25-200 microns wide, thus forming columns. Radial cellular cords with no labelled cells were interspersed between the labelled radial cords.

Synaptic connections of morphologically identified and physiologically characterized large basket cells in the striate cortex of cat

Neurons were studied in the striate cortex of the cat following intracellular recording and iontophoresis of horseradish peroxidase. The three selected neurons were identified as large basket cells on the basis that (i) the horizontal extent of their axonal arborization was three times or more than the extent of the dendritic arborization; (ii) some of their varicose terminal segments surrounded the perikarya of other neurons. The large elongated perikarya of the first two basket cells were located around the border of layers III and IV. The radially-elongated dendritic field, composed of beaded dendrites without spines, had a long axis of 300-350 microns, extending into layers III and IV, and a short axis of 200 microns. Only the axon, however, was recovered from the third basket cell. The lateral spread of the axons of the first two basket cells was 900 microns or more in layer III and, for the third cell, was over 1500 microns in the antero-posterior dimension, a value indicating that the latter neuron probably fulfills the first criterion above. The axon collaterals of all three cells often branched at approximately 90 degrees to the parent axon. The first two cells also had axon collaterals which descended to layers IV and V and had less extensive lateral spreads. The axons of all three cells formed clusters of boutons which could extend up a radial column of their target cells. Electron microscopic examination of the second basket cell showed a large lobulated nucleus and a high density of mitochondria in both the perikarya and dendrites. The soma and dendrites were densely covered by synaptic terminals. The axons of the second and third cells were myelinated up to the terminal segments. A total of 177 postsynaptic elements was analysed, involving 66 boutons of the second cell and 89 boutons of the third cell. The terminals contained pleomorphic vesicles and established symmetrical synapses with their postsynaptic targets. The basket cell axons formed synapses principally on pyramidal cell perikarya (approximately 33% of synapses), spines (20% of synapses) and the apical and basal dendrites of pyramidal cells (24% of synapses). Also contacted were the perikarya and dendrites of non-pyramidal cells, an axon, and an axon initial segment. A single pyramidal cell may receive input on its soma, apical and basal dendrites and spines from the same large basket cell.(ABSTRACT TRUNCATED AT 400 WORDS)

Entorhinal lesions result in shrinkage of the outer molecular layer of rat dentate gyrus leading subsequently to an apparent increase of glutamate decarboxylase and cytochrome oxidase activities

In intact dentate gyrus, glutamate decarboxylase immunoreactivity (GAD) and cytochrome oxidase activity (CyO) showed different distributions patterns. Entorhinal lesions caused increases of GAD and CyO in the outer molecular layer (OML) of the ipsilateral side. Submicroscopical localization of these enzymes did not change, except for CyO labeling more astrocytic mitochondria. The increase in numerical density of GAD puncta correlated quantitatively with shrinkage of OML, whereas in the whole molecular layer the number of GAD puncta remained unchanged. Hence, the localized increase of enzyme activities and lysosomes is apparently related to shrinkage of OML, but does not indicate plasticity of GABAergic neurons.

Specificity of cortico-cortical connections in monkey visual system

When the primate primary visual cortex, area 17, is stained for the mitochondrial enzyme cytochrome oxidase, it shows a striking polka-dot pattern (cytochrome oxidase blobs). Area 18, the second visual area, shows a cytochrome-oxidase pattern of coarse alternating thick and thin stripes running perpendicular to the 17-18 border and separated by lighter (interstripe) regions. Here we show that the thin cytochrome oxidase stripes, and possibly also the thick stripes, in area 18 receive projections specifically from the blobs in area 17, and that the interstripe regions of 18 receive projections from the interblob matrix of area 17. This indicates a specificity of cortico-cortical connections far exceeding the demands of topographical mapping. Together with our physiological results, it suggests that within the pathway from area 17 to area 18 different kinds of information may be handled separately and in parallel.

Morphological diversity of immunocytochemically identified GABA neurons in the monkey sensory-motor cortex

GABAergic neurons have been identified in monkey sensory-motor cerebral cortex by light microscopic, immunocytochemical localization of the GABA synthesizing enzyme, glutamic acid decarboxylase (GAD). All GAD-positive neurons are non-pyramidal cells. Their somata are present in all layers and are evenly distributed across layers II-VI of the motor cortex (area 4), but are found in greater concentrations in layers II, IV and VI of all areas of first somatic sensory cortex (SI; areas 3a, 3b and 1-2). GAD-positive puncta (putative axon terminals) are present throughout the sensory-motor cortex, and they are found immediately adjacent to the somata, dendrites and presumptive axon initial segments of GAD-negative pyramidal cells. In addition, they are observed in close approximation to the somata of both large and small GAD-positive neurons. In area 4, the density of puncta is highest in the superficial cortical layers (layers I-III) and gradually declines throughout the deeper layers. In SI, the highest densities of puncta are present in layer IV, while moderately high densities are found in layers I-III and VI. In areas 3a and 3b, the puncta in layers IV and VI are particularly numerous and form foci that exhibit greater density than adjacent regions. GAD-positive neurons with large somata, 15-33 micron in diameter, are present in layers IIIB-VI of all areas. Such cells have many primary dendrites that radiate in all directions. In addition they have axons that ascend either from the superficial aspect of the somata or from primary dendrites, and that exhibit horizontal collateral branches. These neurons closely resemble the large basket cells (Marin-Padilla, 1969; Jones, 1975), and they may give rise to many of the GAD-positive endings surrounding the somata and proximal dendrites of pyramidal cells in layers III-VI. In addition, small GAD-positive somata are present in all layers, but they are most numerous in layers II and IIIA of all areas and in layer IV of SI. The somata and proximal dendrites of these cells vary from a multipolar shape with small, beaded dendrites, found primarily in layer IV, to bitufted and multipolar shapes with larger, smooth dendrites. The diversity of somal sizes and locations, the variety of dendritic patterns, and the different distributions of GAD-positive puncta, all combine to suggest that several different morphological classes of intrinsic neurons comprise the GABA neurons of monkey cerebral cortex.

Synaptic organization of immunocytochemically identified GABA neurons in the monkey sensory-motor cortex

Neurons in the monkey somatic sensory and motor cortex were labelled immunocytochemically for the GABA synthesizing enzyme, glutamic acid decarboxylase (GAD), and examined with the electron microscope. The somata and dendrites of many large GAD-positive neurons of layers III-VI receive numerous asymmetric synapses from unlabelled terminals and symmetric synapses from GAD-positive terminals. Comparisons with light and electron microscopic studies of Golgi-impregnated neurons suggest that the large labelled neurons are basket cells. Small GAD-positive neurons generally receive few synapses on their somata and dendrites, and probably conform to several morphological types. GAD-positive axons from symmetric synapses on many neuronal elements including the somata, dendrites and initial segments of pyramidal cells, and the somata and dendrites of non-pyramidal cells. Synapses between GAD-positive terminals and GAD-positive cell bodies and dendrites are common in all layers. Many GAD-positive terminals in layers III-VI arise from myelinated axons. Some of the axons form pericellular terminal nests on pyramidal cell somata and are interpreted as originating from basket cells while other GAD-positive myelinated axons synapse with the somata and dendrites of non-pyramidal cells. These findings suggest either that the sites of basket cell terminations are more heterogeneous than previously believed or that there are other classes of GAD-positive neurons with myelinated axons. Unmyelinated GAD-positive axons synapse with the initial segments of pyramidal cell axons or form en passant synapses with dendritic spines and small dendritic shafts and are interpreted as arising from the population of small GAD-positive neurons which appears to include several morphological types.

Four types of neuron in layer IVab of cat cortical area 17 accumulate 3H-GABA

Roughly 10% of the neurons in layer IVab of cat area 17 accumulate exogenous 3H-gamma-aminobutyric acid (GABA) but how many types of neuron comprise this population was unknown. We characterized these neurons by partial reconstruction of their somas from serial electron microscope autoradiograms and distinguished four types. GABA 1 was large (greater than 16.5 micron) and dark with a dense distribution of synaptic terminals, substantial geniculate input to the soma, and a moderate accumulation of GABA. GABA 2 was small (less than 13 micron) and pale, also with a dense distribution of terminals but without evidence of somatic geniculate input, and a moderate accumulation of GABA. GABA 3 was radially fusiform (20 micron X 8 micron) with varicose dendrites, a sparse distribution of synaptic terminals, and a heavy accumulation of GABA. GABA 4 was medium in size (15 micron) with a moderate distribution of synaptic terminals and a heavy accumulation of GABA. Reasons are presented for believing that each of these four categories of GABA-accumulating neuron represents a fundamental cell type.

The section-Golgi impregnation procedure. 2. Immunocytochemical demonstration of glutamate decarboxylase in Golgi-impregnated neurons and in their afferent synaptic boutons in the visual cortex of the cat

Sections of the cat's visual cortex were stained by an antiserum to glutamate decarboxylase using the peroxidase-antiperoxidase method; they were then impregnated by the section Golgi procedure and finally the Golgi deposit was replaced by gold. Neurons containing glutamate decarboxylase immunoreactivity were found in all layers of the visual cortex, without any obvious pattern of distribution. Fifteen immunoreactive neurons were also Golgi-impregnated and gold-toned, which enabled us to study the morphology and synaptic input of identified GABAergic neurons. These neurons were found to be heterogeneous both with respect to the sizes and shapes of their perikarya and the branching patterns of their dendrites. All the immunoreactive, Golgi-impregnated neurons had smooth dendrites, with only occasional protrusions. The synaptic input of glutamate decarboxylase-immunoreactive neurons was studied in the electron microscope. Immunoreactive neurons received immunoreactive boutons forming symmetrical synapses on their cell bodies. The Golgi-impregnation made it possible to study the input along the dendrites of immunoreactive neurons. One of the large neurons in layer III whose soma was immunoreactive was also Golgi-impregnated: it received numerous non-immunoreactive asymmetrical synaptic contacts along its dendrites and occasional ones on its soma. The same neuron also received a few boutons forming symmetrical synaptic contacts along its Golgi-impregnated dendrites; most of these boutons were immunoreactive for glutamate decarboxylase. Glutamate decarboxylase-immunoreactive boutons were also found in symmetrical synaptic contact with non-immunoreactive neurons that were Golgi-impregnated. A small pyramidal cell in layer III was shown to receive several such boutons along its somatic membrane. It is concluded that the combination of immunoperoxidase staining and Golgi impregnation is technically feasible and that it can provide new information. The present study has shown that there are many morphologically distinct kinds of aspiny GABAergic neurons in the visual cortex; that the predominant type of synaptic input to the dendrites of such neurons is from boutons forming asymmetrical synapses, but that some of the GABAergic neurons also receive a dense symmetrical synaptic input on their cell bodies, and occasional synapses along their dendrites, from the boutons of other GABAergic neurons. These findings provide a morphological basis, firstly, for a presumed powerful excitatory input to GABAergic interneurons and, secondly, for the disinhibition which has been postulated from electrophysiological studies to occur in the cat's visual cortex.

Intrinsic laminar lattice connections in primate visual cortex

Intracortical injections of horseradish peroxidase (HRP) reveal a system of periodically organized intrinsic connections in primate striate cortex. In layers 2 and 3 these connections form a reticular or latticelike pattern, extending for about 1.5-2.0 mm around an injection. This connectional lattice is composed of HRP-labeled walls (350-450 microns apart Saimiri and about 500-600 microns in macaque) surrounding unlabeled central lacunae. Within the lattice walls there are regularly arranged punctate loci of particularly dense HRP label, appearing as isolated patches as the lattice wall labeling thins further from the injection site. A periodic organization has also been demonstrated for the intrinsic connections in layer 4B, which are apparently in register with the supragranular periodicities, although separated from these by a thin unlabeled region. The 4B lattice is particularly prominent in squirrel monkey, extending for 2-3 mm from an injection. In both layers, these intrinsic connections are demonstrated by orthogradely and retrogradely transported HRP and seem to reflect a system of neurons with long horizontal axon collaterals, presumably with arborizations at regularly spaced intervals. The intrinsic connectional lattice in layers 2 and 3 resembles the repetitive array of cytochrome oxidase activity in these layers; but despite similarities of dimension and pattern, the two systems do not appear identical. In primate, as previously described in tree shrews (Rockland et al., '82), the HRP-labeled anatomical connections resemble the pattern of 2-deoxyglucose accumulation resulting from stimulation with oriented lines, although the functional importance of these connections remains obscure.

Functional organization of the second cortical visual area in primates

The functional organization of the second cortical visual area was examined with three different anatomical markers: 2-[14C]deoxy-D-glucose, cytochrome oxidase, and various myelin stains. All three markers revealed strips running throughout the area, parallel to the cortical surface. The boundaries of these strips provide an anatomical criterion for defining the borders of this extrastriate region. Further, the demonstration of these strips allows a functional and anatomical analysis of modules in the area, just as the recent demonstration of spots in the primary visual cortex has allowed an analysis of modules there. The strips differ structurally and functionally from interstrip regions and these differences are similar to those seen between the spots and the interspot regions in the primary visual cortex. In the macaque the strips and spots differ with regard to binocular organization.

Retrograde transport of gamma-amino[3H]butyric acid reveals specific interlaminar connections in the striate cortex of monkey

Several lines of evidence suggest that gamma-aminobutyric acid is an inhibitory neurotransmitter in the cerebral cortex. To study the intracortical projection of neurons that selectively accumulate this amino acid, we injected radioactive gamma-aminobutyric acid into the upper layers of the striate cortex of monkeys along tracks at an oblique angle to the pia. Sections from the injected area were then processed by a combination of autoradiography and Golgi impregnation to reveal the distribution of labeled neurons and their morphological characteristics. Labeled neurons always occurred around the injection site in each layer. In addition, a consistent radial pattern of perikaryal labeling was observed in layers IVc-VI below the injection track in layers I-IVa. The closer the injection track was to the pia the deeper the peak density of labeled cells appeared. After injection in layers IVa and the lower part of III, the highest number of labeled neurons was in layer IVc; after injection in the upper part of layer III, most labeled neurons were in layer V; and, after injection in layers I and II, the proportion of labeled neurons increased in the lower part of layer V and in layer VI. All these neurons in the infragranular layers are presumably labeled by retrograde axonal transport via the labeled fiber bundles that extended from upper to lower layers. Thirty-four Golgi-stained neurons of various types were also examined for retrograde labeling. Two were labeled, and both were aspiny stellate cells in layer V. The arrangement of these putative GABAergic neurones, with axons that ascend from lower to upper layers in a regular pattern and arborize locally, would enable them to mediate inhibition within cortical columns and between neighboring columns.