Molecular and morphological changes in the cat lateral geniculate nucleus and visual cortex induced by visual deprivation are revealed by monoclonal antibodies Cat-304 and Cat-301.

Monoclonal antibody Cat-301 recognizes a surface-associated proteoglycan on subsets of neurons in the mammalian CNS (Hockfield and McKay, 1983). The expression of Cat-301 immunoreactivity on Y cells in the cat LGN is sharply reduced by early visual deprivation (Sur et al., 1988). We employed an immunosuppression strategy (Hockfield, 1987) to further study alterations in the expression of experience-dependent molecules. Newborn BALB/c mice were injected with LGN from dark-reared cats to induce a suppression of the immune response to antigens expressed in visually deprived animals. These mice were then immunized with LGN from normal cats to elicit an immune response to antigens with an expression dependent on normal early visual experience. This strategy permitted the generation of monoclonal antibody Cat-304, which recognizes a surface-associated antigen on neuronal cell bodies and proximal dendrites, and which appears histologically identical to Cat-301. Further analyses show that Cat-304 and Cat-301 recognize different epitopes on the same 680-kDa chondroitin sulfate proteoglycan. We examined the effects of early visual deprivation on Cat-304 immunoreactivity in the LGN and visual cortex of cats. In LGN from normal cats, Cat-304 labels neurons in layers A, A1, and C, in interlaminar zones, and in the medial interlaminar nucleus. In LGN from dark-reared cats, the number of antibody-positive neurons is markedly reduced, and the cross-sectional area of the remaining positive neurons is smaller than normal. In cortical area 17 of normally reared cats, Cat 304-positive neurons are densely distributed in 2 bands, in layers IV and V/VI. Labeled neurons are also present in layers II and III. In area 17 of dark-reared cats, the number of antibody-positive neurons is reduced. The reduction in the number of labeled neurons is most pronounced in layers II/III and V/VI. Antibody-positive neurons are smaller in all cortical layers of dark-reared cats. The changes in the expression of Cat-301 immunoreactivity in dark-reared visual cortex and LGN are identical to those of Cat-304. The laminar differences in the effect of dark rearing on Cat-301 and Cat-304 expression in the visual cortex provides support for the suggestion that layer IV of cortical area 17 may be less susceptible to prolongation of plasticity by dark rearing than layers II/III and V/VI. Further, the biochemical and histological studies reported provide evidence that early visual experience regulates protein expression in the cat LGN and visual cortex.

GAP-43 in adult visual cortex.

GAP-43 was purified from cat brain by a rapid isolation procedure and was used to raise highly specific polyclonal antibodies in rabbits. Immunoblots of proteins from adult cat, monkey and human visual cortex as well as bovine cortex also showed specific staining of a single protein that was present in both soluble and membrane fractions. Immunocytochemistry of both cat and human adult visual cortex showed that GAP-43 has a laminar distribution.

Ammonia and monoamine concentrations in two brain areas in rats after one hyperoxic seizure.

Monoamines (catecholamines, serotonin, and metabolites) and ammonia were studied within two areas of the rat brain--the frontal cortex (FC) and the striatum (SA)--after exposure to hyperbaric oxygen (HBO) at 6 ATA up to the first seizure. An increase of norepinephrine (NE), dopamine (DA), and metabolites (HVA, DOPAC) measured by the HPLC/EC method were found in SA with a parallel increase of ammonia at variance with the FC where no monoamine changes, but a slight increase of ammonia, were found. Blood ammonia did not change with HBO. So, 20 min after one HBO seizure, there are regional differences in the brain, which are consistent with the previous findings of an SA start of electrocortical abnormalities at the onset of a seizure. Elevated DA, and possibly NE, levels may contribute to the accumulation of ammonia in the brain. During prolonged HBO exposure, this rise of ammonia could be one of the mechanisms involved in the relapse of seizures. It might also be implicated in initiation of the first seizure. By their situations and contents, SA glial cells could play an important role in brain HBO susceptibility.

Postnatal development of vimentin-immunoreactive radial glial cells in the primary visual cortex of the cat.

In kitten area 17 vimentin-like immunoreactivity is expressed in radial glial fibres up to one month postnatally, i.e. the time for which neuronal migration continues. During this time fibre density gradually decreases. A subpopulation of these fibres also contains S-100 protein and glial fibrillary acidic protein. However, these latter antigens disappear earlier than vimentin. In addition, vimentin immunoreactivity can be observed in astroglial cells of the white matter between the second and fifth postnatal week. Many of these cells resemble mature astrocytes but partially they have an intermediate morphology suggesting the possibility that they originated from radial glia. Such displaced radial glial cells' are also positive for S-100 protein both in the cortex and white matter. Thus it is conceivable that at least part of the astrocytes of mature cat visual cortex originate from vimentin- or S-100-immunoreactive radial glia.

Light and electron microscopic immunocytochemical analysis of the dopamine innervation of the rat visual cortex.

The dopaminergic innervation of the rat primary (area 17) and secondary (areas 18 and 18a) visual cortical areas was examined immunocytochemically using an antiserum directed against dopamine. This innervation was characterized by the differential density of the respective afferents within individual visual areas. Area 18, especially its rostral part, was observed to receive a considerable amount of dopaminergic axons, whereas areas 17 and 18a were sparsely innervated. The innervation of all layers of area 18 seemed to consist to a considerable extent of axonal branches of radial fibres ascending from layer VI to layer I. At the ultrastructural level, dopamine profiles were found to display similar characteristics in all visual areas. Dopamine labelled axon-terminals and axonal varicosities, examined in single and serial ultrathin sections, were seen to form primarily asymmetrical synaptic contacts with dendritic profiles. These observations suggest a 'specific' innervation of cytoarchitectonically distinct cortical regions by dopaminergic axons.

Physiological, morphological, and cytochemical characteristics of a layer 1 neuron in cat striate cortex.

We have recorded from a small neuron in layer 1 of the striate visual cortex in a 34-day-old kitten. It had a simple, orientation-selective receptive field that was nondirectional and showed length summation. The neuron was injected intracellularly with horseradish peroxidase. Computer-aided reconstruction revealed that it had a dense axonal plexus confined to layer 1, elongated in the anteroposterior dimension. By means of an antibody directed against a GABA-like antigen, and postembedding immunocytohemistry, the neuron was found to be strongly immunoreactive. The main input to soma and dendrites of the neuron was from synapses that were not GABA-L-immunoreactive, and probably originated from pyramidal cells. The axon of the cell formed synapses on dendritic shafts and spines, whose most likely sources were the apical tufts of pyramidal cell dendrites. These data suggest that such neurons may be involved in local circuits that contribute to the formation of pyramidal cell receptive fields.

Age-related changes of catecholamines and their metabolites content in the visual system of the rat.

The changes in the content of the catecholamines in each structure of the geniculate and extrageniculate visual system of the rat during the aging period (6-30 months) have been studied. Dopamine was found at lower levels than noradrenaline in all the structures. The dopamine and noradrenaline showed different developmental profiles. Dopamine and its metabolite levels decreased in the lateral geniculate and visual cortex and increased in superior colliculus and posterior thalamus. Noradrenaline and its metabolites increased in all structures during the aging period. However, 3-methoxy-4-hydroxyphenylglycol/noradrenaline and normetanephrine/noradrenaline ratios decreased in all structures except in superior colliculus. These results suggest age-related changes in the catecholamines in the visual system of the rat.

Quantitative autoradiography of major neurotransmitter receptors in the monkey striate and extrastriate cortex.

In vitro autoradiography was used to determine the binding properties and distribution of 9 major neurotransmitter receptors and their subtypes in the striate (area 17 of Brodmann) and extrastriate (areas 18 and 19) cortex of 1 infant and 3 adult rhesus monkeys. Differences in total labeling and nonspecific labeling, as well as Kd and Bmax values, were determined for all cortical layers and sublayers in both cytoarchitectonic areas by Scatchard analysis of autoradiograms. Area 17 differed from area 18 in the laminar pattern and density of virtually every ligand examined, i.e., 3H-clonidine, 3H-prazosin, 125I-iodopindolol, 3H-quinuclidinyl benzilate, 3H-5-hydroxytryptamine, 3H-ketanserin, 3H-muscimol, 3H-flunitrazepam, and 3H-spiperone. Kd and Bmax values for each ligand were remarkably consistent across the 3 adult monkeys analyzed quantitatively. Particularly dramatic contrasts were observed with clonidine, 5-hydroxytryptamine, and ketanserin, which have high affinity for alpha 2-adrenergic, 5-HT1-, and 5-HT2-receptors, respectively. The differences in distribution of these ligands, especially clonidine and 5-hydroxytryptamine, correlated well with specific laminae and hence exhibited distinctly different patterns in areas 17 and 18. Other ligands, such as flunitrazepam and quinuclidinyl benzilate that bind to GABAergic and cholinergic receptors, were visually less discriminating both among layers and between regions. However, layer for layer, the Bmax values for quinuclidinyl benzilate were higher in area 17 than 18, indicating the subtle differences between areas may be revealed only by quantitative measures. Some ligands were particularly dense in layer I (iodopindolol in areas 17 and 18; 5-hydroxytryptamine in area 18), and others subdivided cortical layers that are otherwise cytoarchitectonically uniform (e.g., flunitrazepam and clonidine in layer VI of area 17), indicating that areal differences in ligand binding are not a simple read-out of cell-packing density but most likely reflect a genuine difference related to the neurotransmitters of intrinsic and extrinsic afferents in each area. The presence of binding sites in every layer of both areas for all ligands examined indicates that their distribution across laminae is quantitative and not all-or-none. No layer contained less than 50% of binding sites present in any other layer. These findings reveal that visual cortical areas differ in density and lamination of neurotransmitter receptors and presumably in their sensitivity to circulating levels of endogenous neurotransmitters and pharmacologically active compounds.

Synaptic organization of serotonin-immunoreactive fibers in primary visual cortex of the macaque monkey.

The macaque neocortex is very densely innervated by serotonin-containing fibers. The highest density of these fibers is in primary sensory regions such as the primary visual cortex. By using an antibody against serotonin, we analyzed the distribution and morphology of serotonin-immunoreactive fibers and synapses in the primary visual cortex of the adult cynomolgus monkey. In addition, we quantified the laminar distribution of labeled varicosities and the distances between varicosities in single fibers. While serotonin-immunoreactive fibers are found in all cortical layers, at least three bands of heightened density of innervation were readily recognized that were coincident with 1) layer IIIB to IVC alpha, 2) layer VA, and 3) layer VIB. Layer IVC alpha of area 17 contained more varicosities per unit area than any other sublayer. There was a high degree of variability in the intervaricosity distances along single fibers; more than half were longer than 10 microns. At the electron microscopic level, synaptic contacts were also observed throughout the entire thickness of area 17, with the highest frequency in layer IV. The labeled varicosities were packed with electron-lucent synaptic vesicles and formed synaptic complexes with small, but conspicuous, post-synaptic densities. Dendritic shafts were the most common postsynaptic target of the labeled synapses. Among these characteristically slender post-synaptic shafts, profiles with structural features of both spiny and smooth dendrites were observed. The small diameter of most of the postsynaptic dendrites indicated that distal dendrites were preferentially contacted by serotonin-immunoreactive varicosities. Although direct identification of the postsynaptic neurons will be required for complete characterization of this circuitry, the distribution of serotonin-immunoreactive varicosities suggests that serotoninergic interactions in the primary visual cortex of the cynomolgus monkey are directed predominantly at the distal dendrites of granular and infragranular neurons rather than at targets in the supragranular layers.

The distribution and ontogenesis of [3H]nicotine binding sites in cat visual cortex.

In vitro autoradiographic techniques using [3H]nicotine were used to characterise nicotine binding sites in developing kitten visual cortex. These binding sites in adult animals have a Bmax of 3.91 fmol/mg protein and a Kd of 4.40 nM. Displacement experiments indicate that [3H]nicotine binds to a nicotinic receptor site that is similar to central nicotinic sites described by investigators in other mammals. The number of binding sites increases during postnatal development, peaking near 60 days of age and levelling-off thereafter. There is no evidence for large changes in affinity during postnatal development for this binding site. [3H]Nicotine binding sites are densely concentrated in layer IV in the visual cortex of adult animals, with sharply reduced binding outside of cortical areas 17 and 18. This laminar pattern does not change during postnatal development, but an increase in the number of binding sites in layer IV as well as in layers I and VI occurs during early postnatal life. These binding sites disappear when extrinsic cortical inputs are severed. However, they survive when neurons in the visual cortex are selectively destroyed with a cell-specific neurotoxin. Unilateral destruction of the lateral geniculate nucleus eliminates [3H]nicotine binding sites in the visual cortex ipsilateral to the lesion, suggesting that they are located presynaptically on the terminals of lateral geniculate nucleus afferent fibres. The laminar pattern of binding of [3H]nicotine during early development of the visual cortex is complimentary to that for muscarinic acetylcholine receptors. These latter receptors redistribute during postnatal development becoming less prominent in layer IV at the same time as the [3H]nicotine binding sites are increasing in number in this layer. For a short period of time at the height of the critical period for cortical plasticity, both populations of binding sites are located in layer IV.

Heterogeneity of GABAergic cells in cat visual cortex.

Antibodies against neuropeptides and against a vitamin D-dependent calcium-binding protein (CaBP) label small cells with nonpyramidal-like morphology in the cat visual cortex (areas 17, 18, and 19). Since GABAergic cells are interneurons, a double-staining procedure was used to test for the coexistence of cholecystokinin (CCK), somatostatin (SRIF), neuropeptide Y (NPY), corticotropin-releasing factor (CRF), vasoactive intestinal polypeptide (VIP), and CaBP with glutamic acid decarboxylase (GAD). Our results show that CRF and VIP do not coexist with GAD, while the 3 other peptides and CaBP do. Hence GAD-positive cells can be subdivided into 4 broad groups: (1) cells that are only GAD-positive, (2) cells that are GAD- and CaBP-positive, (3) GAD-positive neurons also containing CCK, and (4) GAD-positive cells that also contain SRIF. A small subset of class 2 also contains SRIF and most cells of class 4 also contain NPY. The 4 classes of GAD-positive cells differ in laminar position: class 1 predominates in layers IV and V, classes 2 and 3 in the upper laminae (II and III), and class 4 in the deepest layer (VI).

Quantitative immunohistochemical analysis of VIP-neurons in the rat visual cortex.

A critical appraisal of quantitative immunohistochemistry of neuropeptides is presented defining the main criteria of selecting the type of immune-staining and preparation suitable for these investigations. As an example of meeting the established criteria, the immunohistochemical demonstration of vasoactive intestinal polypeptide (VIP)-containing neurons in the rat brain and the processing of VIP-immunostained preparations for computer-controlled image analysis are described.

Ramification patterns of vasoactive intestinal polypeptide (VIP)-cells in the rat primary visual cortex. An immunohistochemical study.

Vasoactive intestinal polypeptide (VIP)-immunoreactive cells in the primary visual cortex of the rat were classified on the basis of ramification pattern of cell processes. The distribution of cells over cortical layers, and proportions of cell classes relative to total cell numbers were evaluated by means of quantitative methods. Two main types of VIP-positive neurons, the bipolar and the multipolar were distinguished constituting 76% and 24% of the VIP populations, respectively. The axons of vertically oriented bipolars were observed to ramify within a column around the descending dendrite. By contrast, multipolar cells have a non-oriented ramification pattern. The two overlapping axonal systems form the VIP-innervation of the rat visual cortex.

Types and spatial distribution of vasoactive intestinal polypeptide (VIP)-containing synapses in the rat visual cortex.

In the rat visual cortex vasoactive intestinal polypeptide (VIP)-containing structures were studied by means of light and electron microscopy and image analysis. VIP-immunoreactive axon terminals were found to form symmetric synapses with small dendritic shafts, dendritic spines and somata of pyramidal cells and interneurons. VIP-terminals often occurred in pairs with VIP-negative, asymmetric synapses on the same postsynaptic structure. VIP-immunostained dendrites and perikarya were contacted by a purely asymmetric and a mixed population of VIP-negative terminals, respectively. Synaptic connections between two VIP-neurons are seldom as compared to the other types of VIP-synapses. Quantitative studies obtained by the image analysis of VIP-stained boutons and dendritic particles in light microscopic preparations suggest a distinct laminar distribution. Dendritic particles are most frequent in layers I-II, whereas axonal boutons have three laminar accumulations: at the border of layers I-II, in layer IV and layer VI. Together with previous results, the present findings argue for a non-random spatial distribution of VIP-boutons.

Enigmatic bipolar cell of rat visual cortex.

Our earlier Golgi-electron microscopic study of bipolar cells in the rat visual cortex showed the axons of these neurons as forming asymmetric synapses (Peters and Kimerer; J. Neurocytol, 10:921-946, '81) in which the most common postsynaptic elements were dendritic spines. This result was unexpected, since Parnavelas (Parnavelas, Sullivan, Lieberman, and Webster: Cell Tissue Res. 183:499-517, '77) had earlier shown a bipolar cell from the same cortex to have an axon forming symmetric synapses with dendritic shafts. Here then was an enigma, strengthened by examination of neuronal components labelled by antibodies to two compounds in particular--namely, vasoactive intestinal polypeptide (VIP) and choline acetyltransferase (ChAT). Antibodies to these compounds preferentially label bipolar cells in the rat cerebral cortex, and the labelled axon terminals form symmetric synapses. Against this background the present study was performed, and it has been shown that the resolution to the enigma is that there are two different populations of bipolar cells in the rat visual cortex. Thus some Golgi-impregnated bipolar cells examined by electron microscopy after gold toning have been found to possess axons forming asymmetric synapses, and others have been found to have axons forming symmetric synapses. The axons of the bipolar cells forming asymmetric synapses most commonly synapse with dendritic spines (67%), although other terminals synapse with dendritic shafts (33%). In contrast, the bipolar cells with axons forming symmetric synapses preferentially synapse with dendritic shafts (100%). The population of bipolar cells that form symmetric synapses includes the ones that label with antibodies to vasoactive intestinal polypeptide (VIP), for the axons of VIP-labelled bipolar cells have been traced to labelled terminals forming symmetric synapses. However, examination of the population of VIP-labelled axon terminals shows that in addition to dendritic shafts, some of the labelled terminals synapse with the cell bodies of pyramidal and nonpyramidal cells. This includes bipolar cells, some of which receive large numbers of VIP-labelled axon terminals. It is also shown that some VIP-positive bipolar cells have myelinated axons. Analysis of tissue labelled with VIP antibody reveals that about 50% of the total population of bipolar cells in the rat visual cortex is VIP positive. These results are discussed in the light of information about labelling of bipolar cells with antibodies to gamma-aminobutyric acid (GABA) and to other peptides, and it is suggested that most VIP-positive bipolar cells also contain GABA.

Distribution of catecholamines, serotonin, and their major metabolites in the rat cingulate, piriform-entorhinal, somatosensory, and visual cortex: a biochemical survey using high-performance liquid chromatography.

The catecholamines noradrenaline (NA), dopamine (DA), adrenaline (AD), the indoleamine 5-hydroxytryptamine (5-HT; serotonin), as well as some of their major metabolites were assayed by high-performance liquid chromatography (HPLC) with electrochemical detection, in four well-defined areas of the rat cerebral cortex: anterior cingulate (CIN;Cg1 and Cg3), piriform and entorhinal (PiEn), hind-limb primary somatosensory (SSC;HL) and primary visual (VIS; Oc1M and Oc1B). The concentrations of NA and that of its main metabolite 3-methoxy-4-hydroxyphenylglycol were highest in PiEn, had intermediate values in CIN and were lowest for SSC and VIS cortices. The DA levels were also highest in PiEn, intermediate in CIN, while the lowest values were in SSC and VIS cortices. The different DA/NA ratios support the hypothesis that they are indeed independent neurotransmitters. In addition, the levels of 3,4-dihydroxyphenylacetic acid, homovanillic acid and 3-methoxytyramine paralleled the distribution of DA, thus confirming the presence of release sites, even in regions in which the low levels of this catecholamine could be interpreted simply as the precursor of NA. Traces of AD were detected in all the regions examined. The 5-HT contents, as well as that of its precursor 5-hydroxy-1-tryptophan and that of its metabolite 5-hydroxyindole-3-acetic acid were also found to be non-homogenous, with the highest levels measured in the PiEn and CIN regions.

Light and electron microscopic immunocytochemical analysis of the serotonin innervation of the rat visual cortex.

The serotonin afferents of the rat visual cortex were examined immunocytochemically at the light and electron microscopic levels. Immunoreactive fibres were typically thin, tortuous and varicose. Occasionally, some thicker fibres were found. The orientation of labelled axons varied according to laminar position, with fibres running parallel to the pial surface present mainly in layers I and VI, and radially oriented fibres prominent in layers II and III. Branches arising from horizontal or radially oriented fibres were seen to form irregularly shaped loops particularly in layers IV and V. The density of innervation and the prevailing axonal orientation in each cortical layer were similar in both coronal and parasagittal planes. The ultrastructural features of serotonin-labelled axon terminals were examined in single and serial ultrathin sections. While in single sections the majority did not exhibit synaptic specializations, extensive serial section analysis showed that virtually all of these terminals were engaged in junctional complexes. Postsynaptic elements were spines and dendritic shafts, including pyramidal cell apical dendrites, with both symmetrical and asymmetrical membrane specializations. In axospinous synapses, the labelled terminals were usually adjacent to unstained axon terminals contacting the same postsynaptic element.

Distribution of neuropeptide Y immunoreactivity in human visual cortex and underlying white matter.

Immunocytochemical techniques have been used to study neuropeptide Y (NPY) distribution in the human visual cortex (Brodman's areas 17, 18 and 19) NPY cell bodies belong mostly to inhibitory (multipolar and bitufted) but also to excitatory (bipolar and some pyramidal) neuronal types. Their distribution is similar in the three cortical areas studied: 20 to 40% of the NPY perikarya are located in the cortical gray matter, mostly in the deep layers, while the remaining 60 to 80% are located in the underlying white matter. Immunoreactive NPY processes form a rich network of intersecting fibers throughout the entire visual cortex. A superficial plexus (layers I and II) and a deep plexus (deep layer V and layer VI) of NPY fibers are present in areas 17, 18 and 19. In area 17, an additional well developed plexus is present in layers IVb and IVc. These plexuses receive branches from long parallel fibers arising from deep cortical layers or underlying white matter and terminating in superficial layers. Local or extrinsic NPY terminals wind around vessels in the cortex as well as in the white matter, and either penetrate them or form clusters of club endings on their walls. Our results suggest a role for NPY in human visual circuitry and in cortical blood flow regulation.

Extensive co-existence of neuropeptides in the rat visual cortex.

The peroxidase-antiperoxidase immunohistochemical technique has been used to examine the co-existence of peptides within individual neurons of the rat visual cortex. Pairs of consecutive paraffin sections were stained alternately for 2 of the 4 peptides: somatostatin (SRIF), vasoactive intestinal polypeptide (VIP), cholecystokinin (CCK) and neuropeptide Y (NPY). Analysis revealed the co-existence of SRIF with VIP, CCK and NPY and between VIP and CCK. These results show that the co-localization of neuropeptides in cortical neurons is more widespread than previously demonstrated.