Selective visual experience fails to modify receptive field properties of rabbit striate cortex neurons.

During development, rabbits were exposed only to vertical or horizontal lines to determine if the receptive field characteristics of visual cortex cells would be altered as they are in the cat. Motion and directional selectivity were preserved, and orientation specificity remained unaffected by the restricted experience, which suggests that the rabbit may lack the neural plasticity seen in some other mammals.

Synaptic regulation of L-type Ca(2+) channel activity and long-term depression during refinement of the retinocollicular pathway in developing rodent superior colliculus.

The retinocollicular pathway undergoes activity-dependent refinement during postnatal development, which results in the precise retinotopic order seen in adults. This process is NMDA- and nitric oxide-dependent. Recent studies have shown that L-type Ca2+ channels may also play a role in synaptic plasticity, but such channel activity has not previously been reported in the developing superior colliculus (SC). Here we report the presence of a postsynaptic plateau potential mediated by L-type Ca2+ channels using whole-cell current clamp of the SC in an isolated brainstem preparation of rats. Seventy percent of SC neurons showed these potentials as early as postnatal day 0 (P0)-P2. The potential was blocked by nitrendipine and/or APV and facilitated by bicuculline, showing that the channel is activated by NMDA receptor-mediated EPSPs and deactivated by GABAA receptor-mediated IPSPs. Blockade of L-type Ca2+ channels also diminished long-term depression, which we could induce in the retinocollicular pathway in neonatal animals. The incidence of plateau potentials decreased to 39% of neurons by P10-P14, suggesting that L-type calcium channels may contribute to retinocollicular pathway refinement in the developing SC.

Immunocytochemical localization of gamma-aminobutyric acid (GABA) in the cat superior colliculus.

This paper reports the pattern of labeling in the cat superior colliculus produced by an antiserum raised against BSA-conjugated gamma aminobutyric acid (GABA) and visualized by light and electron microscope immunocytochemistry. Neuropil labeling was densest within the zonal and superficial gray layers but was also found in the deep layers. Neurons labeled by the GABA antibody were also most dense within the zonal and superficial gray layers, although many labeled neurons were also found in the deeper layers. The ratio of labeled to unlabeled cells varied from an average of 45% in the superficial subdivision and the intermediate gray layer to less than 30% in the deeper laminae. Almost all intensely labeled cells were small (mean area = 127 micron 2) and had varied morphologies. Several types of labeled cell were observed with the electron microscope. One type had a horizontal, fusiform cell body and a deeply invaginated nucleus. Another type had a small round or ovoid cell body with cytoplasm clumped at one end. Labeled cells with other morphologies were also occasionally seen. No labeled glial cells were found. Two types of vesicle-containing dendrite were stained by the GABA antibody. One type had loose accumulations of small synaptic vesicles and often received input from retinal terminals. Another type had spines also containing small synaptic vesicles. Labeled dendrites without synaptic vesicles were also seen frequently. Putative axon terminals labeled by the GABA antibody had densely packed synaptic vesicles and formed symmetric synaptic contacts. Labeled myelinated axons were also commonly found. These results confirm those using uptake of tritiated GABA (Mize et al.: J. Comp. Neurol. 202:385-396, '81, J. Comp. Neurol, 206:180-192, '82) in that two of the same classes of GABA neuron, horizontal I and granule I cells, were identified in the superficial laminae. However, the GABA antiserum used in this study also revealed a third class of GABA neuron with vesicle-containing spines. The antiserum also labeled a significant number of putative GABAergic neurons located in the deep subdivision of the cat superior colliculus which were not previously recognized by using transmitter autoradiography.

Enkephalin-like immunoreactivity in the cat superior colliculus: distribution, ultrastructure, and colocalization with GABA.

The distribution of enkephalin (ENK) immunoreactivity has been examined in the cat superior colliculus (SC) by means of light and electron microscope immunocytochemistry. The antisera were directed against leucine enkephalin but also recognized methionine enkephalin. Colocalization of ENK with gamma aminobutyric acid (GABA) was studied with a two-chromagen double-labeling technique. Enkephalin antiserum labeling was highly specific. Dense neuropil labeling was found only in a thin band 75-100 microns wide within the upper superficial gray layer of SC. Negligible neuropil labeling was seen deeper, except for patches of label within the intermediate gray layer. Intensely labeled neurons also had a specific distribution. Forty-seven percent were located within the upper 200 microns of SC, 40% within the deep superficial gray layer, 11% in the optic layer, and only 2% below that layer. Almost all ENK-labeled cells were small (mean area of 117 microns2). Some of these had horizontal fusiform cell bodies and horizontally oriented dendrites. Others had small round somata and thin, obliquely oriented dendrites. In double-labeling experiments, 18% of anti-ENK-labeled cells were also immunoreactive for GABA. Four distinct types of ENK-labeled profile were identified with the electron microscope. Presynaptic dendrites (PSD) with loose accumulations of synaptic vesicles were densely labeled with the antiserum. Conventional dendrites were also labeled. Both types of labeled profile received input from unlabeled synaptic terminals, including those from the retina that contained pale mitochondria and round synaptic vesicles and formed asymmetric synaptic contacts. Retinal terminals were never labeled with the antisera. However, some axon terminals with round synaptic vesicles, dark mitochondria, and symmetric synaptic densities were labeled by the antisera, as were some thinly myelinated axons. These results show that there is a small population of enkephalinergic neurons in the cat SC, some of which also contain GABA. Because not all cells with identical morphologies were double labeled, it appears that neurons of like morphology are chemically heterogeneous.

Choline acetyltransferase-immunoreactive patches overlap specific efferent cell groups in the cat superior colliculus.

Fibers containing acetylcholine (ACh) form distinct patches in the dorsal intermediate gray layer (IGL) of the cat superior colliculus (SC). Although these patches are known to overlap several afferent projections to SC, it is not known whether they are associated with specific postsynaptic cell groups. We have examined the relationship of these ACh fiber patches to specific efferent cell groups by combining retrograde transport of horseradish peroxidase (HRP) with choline acetyltransferase (ChAT) immunocytochemistry. Successful HRP injections were made into the predorsal bundle (PB), the tecto-pontine-bulbar pathway (TPB) and the cuneiform region (CFR), the inferior olive (IO), the dorsolateral pontine gray nucleus (PGD), and the pedunculopontine tegmental nucleus (PPTN). The distribution of HRP-labeled neurons which project to these targets was mapped by a computer-based microscope plotter. Distinct clusters of HRP-labeled neurons in the IGL were seen after three injections into the mesencephalic reticular formation that involved the caudal TPB and cuneiform region (CFR), and after one injection into the medial accessory nucleus of IO. As many as seven clusters of labeled neurons were found in some sections through the caudal one-half of SC after the TPB/CFR injections. Each cluster consisted of 3-20 cells, all of which were small to medium in size. In sections also tested for ChAT, the cell clusters in the TPB/CFR cases were found to overlap precisely the ACh patches in the IGL. In addition, SC neurons projecting to the IO formed clusters above the ChAT patches and in the intermediate white layer (IWL) of SC. None of the other HRP injections produced any obvious cell clusters in the deep layers of SC. These results are the first to show that specific cell groups, distinguished by size and projection site, form clusters that match the patch-like innervation of cholinergic afferents to SC. This modular organization may correspond to saccade-related cells that have also been reported to be organized into clusters in the cat SC.

Superior colliculus neurons which project to the cat lateral posterior nucleus have varying morphologies.

We have studied the laminar position, morphology, and synaptic relationships of neurons in the cat superior colliculus which project to the interadjacent division of the lateral posterior nucleus (LPi), using the retrograde transport of horseradish peroxidase. The neurons which project to LPi are remarkably varied in depth, size, morphology, and synaptic density and appear to consist of at least four cell types. Labeled cells were found laminae. Forty-seven percent were found in the superficial gray layer (50-550 micrometer), all but a few within its deep subdivision. Forty-seven percent were located in the optic layer (550-1,200 micrometer), the majority of these being within the upper one-half of the layer Seven percent were found in the intermediate and deep gray layers (below 1,200 micrometer). Cell body area varied widely, ranging from 37 to 768 micrometer 2 (mean of 243 micrometer2). Based on cell size, shape, and dendritic field orientation, we identified four distinct cell morphologies which were labeled. Thirty-five percent were stellate, 32% were vertical fusiform 19% were granule, and 12% were horizontal cells. Electron microscope analysis confirmed that neurons projecting to the lateral posterior nucleus are a morphologically diverse group. A sample of 71 labeled cells varied significantly in density of synaptic input as well as in size, shape, depth, dendritic distribution, and cytology.

Postembedding immunocytochemistry demonstrates directly that both retinal and cortical terminals in the cat superior colliculus are glutamate immunoreactive.

Although the excitatory neurotransmitter glutamate is known to be present in the cat superior colliculus (SC), the types of synapses that contain glutamate have not been examined. We, therefore, studied the ultrastructure of synaptic profiles labeled by a glutamate antibody by using electron microscopic postembedding immunocytochemistry. In addition, unilateral aspiration lesions of areas 17-18 were made at 5-28 days before death in order to determine whether degenerating terminals from visual cortex were glutamate immunoreactive (Glu-ir). Three types of axon terminal were glu-ir: 1) those containing large, round synaptic vesicles and pale mitochondria, characteristic of retinal terminals (RT profiles); 2) those containing small, round synaptic vesicles and dark mitochondria (RSD profiles); and 3) those containing large, round synaptic vesicles and dark mitochondria (RLD profiles). Measures of mean gold particle density revealed that RT, RSD, and RLD profiles had similar average grain densities (11.3-12.7 particles/unit area). Other labeled profile types included cell bodies, large-calibre dendrites, and myelinated axons. Axon terminals containing flattened synaptic vesicles and vesicle-containing presynaptic dendrites, both of which contain gamma-aminobutyric acid (GABA), had many fewer gold particles (3.6 and 4.8 mean particles/unit area, respectively). Following unilateral removal of visual cortex, normal RSD terminals were observed infrequently in the SC ipsilateral to the lesion. Synaptic terminals in the initial stages of degeneration were heavily labeled by the glutamate antibody, as were axon terminals and myelinated axons undergoing hypertrophied or neurofilamentous degeneration. These results show that both major sensory afferents to the superficial layers of cat SC contain glutamate--RT terminals from the retina and RSD terminals from visual cortex. The origin of RLD terminals is unknown.

Alterations in receptive field properties of superior colliculus cells produced by visual cortex ablation in infant and adult cats.

To determine if functional alterations in the superior colliculus might account for recovery of visual behaviors following visual cortex removal in infant cats, the receptive field characteristics of single units in the superior colliculus of cats whose visual cortex was removed within the first week of life were compared with those of cats which sustained visual cortex lesions in adulthood and with those of normal cats. In the normal superior colliculus, 90% of all cells responded to moving stimuli irrespective of shape or orientation. Sixty-four percent of these units were directionally selective, responding well to movement in one direction but poorly or not at all to movement in the opposite direction. Ninety percent of units were binocular, the vast majority of these responding equally to stimulation of either eye or showing only slight preference for stimulation of the contralateral eye. Responses to stationary flashes of light were observed in only 33% of all visually activated cells in the normal superior colliculus. After visual cortex ablation in adult cats, only six percent of movement sensitive cells were directionally selective. Binocular preference was shifted following adult visual cortex lesions such that sixty percent of all cells responded exclusively or predominantly to stimulation of the contralateral eye. Seventy-one percent of all visually responsive units responded to stationary lights flashed on or off within their receptive field boundaries. Lesions limited primarily to area 17 had the same effect as larger lesions of visual cortex. Infant visual cortex lesions resulted in receptive field alterations similar to those observed after adult ablation. Only fifteen percent of motion sensitive units were directionally selective. Seventy-one percent responded exclusively or predominantly to stimulation of the contralateral eye. Seventy-six percent of visually responsive cells were activated by stationary light. Lesions largely confined to area 17 produced the same alterations as more extensive lesions of visual cortex. Thus, no evidence was found that the superior colliculus is involved in the functional reorganization presumed to occur following visual cortex ablation in infant cats. Recovery of visual behaviors following neonatal injury may therefore not involve alterations in the receptive fields of single cells.

Retinal synapses of the cat medial interlaminar nucleus and ventral lateral geniculate nucleus differ in size and synaptic organization.

The retinal terminals of the medial interlaminar nucleus (MIN) and ventral lateral geniculate nucleus ( VLG ) have been examined quantitatively to determine if there are morphological differences in their synaptic ultrastructure which reflect their distinctive physiologies . The cross-sectional area and density (number per unit area) of synaptic contact zones with conventional and presynaptic dendrites (F2 profiles) were measured for each retinal terminal. The densities of F2 presynaptic dendrites and F1 flattened vesicle axon terminals were also measured. Retinal terminals in MIN were often large (mean size = 2.7 micron2 area) and had a high density of synaptic contacts (0.14 per micron surface area) with conventional dendrites, presynaptic dendrites, and dendritic spines. A high density of F2 presynaptic dendrites (0.08 per micron2 area) was found in MIN. F1 axon terminals were also found frequently (0.04 per micron2). MIN retinal terminals were often organized in glomeruli like those of the dorsal lateral geniculate nucleus. The retinal terminals in VLG were almost always small (mean size = 0.94 micron2 area), although they also had a high density of synaptic contacts (0.17 per micron surface area). They frequently synapsed on small dendrites and dendritic spines and less frequently on large dendrites. Unlike MIN, retinal terminals in VLG rarely contacted F2 presynaptic dendrites which were much less frequent in VLG (0.01 per micron2 area). Like MIN, VLG contained numerous F1 axon terminals (0.06 per micron2 area). No typical retinal glomeruli were found in VLG . These results show that MIN, which contains many Y cells, has a population of large retinal terminals and many F2 presynaptic dendrites. VLG , which apparently has only W cells, contains only small retinal terminals and has fewer F2 presynaptic dendrites. Both have a high density of F1 flat vesicle axon terminals.

Neurons and glia in cat superior colliculus accumulate [3H]gamma-aminobutyric acid (GABA).

We have examined by autoradiography the labeling pattern in the cat superior colliculus following injection of tritiated gamma-aminobutyric acid (GABA). Silver grains were heavily distributed within the zonal layer and the upper 200 micrometer of the superficial gray. Fewer grains were observed deeper within the superficial gray, and still fewer were found within the optic and intermediate gray layers. The accumulation of label was restricted to certain classes of neuron and glia. Densely labeled neurons were small (8-12 micrometer in diameter) and located primarily within the upper 200 micrometer. Dark oligodendrocytes and astrocytes showed a moderate accumulation of label while pale oligodendrocytes and microglia were unlabeled. Label was also selectively accumulated over several other types of profile within the neuropil, including presynaptic dendrites, axons, and axon terminals.

Two types of GABA-accumulating neurons in the superficial gray layer of the cat superior colliculus.

Two types of neuron in the upper superficial gray layer of the cat superior colliculus accumulated exogenous 3H-gamma-aminobutyric acid intensely. The first type was a horizontal cell with a fusiform cell body, horizontal dendrites, a low synaptic density, but a high percentage of cortical synaptic contacts. This cell had presynaptic dendrites. The second type was a granule cell (type A) with a small round cell body, thin and obliquely oriented dendrites, a moderate synaptic density, and few cortical synaptic contacts. These two types differed in size, shape, dendritic morphology, and patterns of synaptic input. They likely participate in different inhibitory mechanisms. Four types of unlabeled neurons were also identified. Type B granule cells were found only within the upper subdivision of the superficial gray layer. They had moderate-sized cell bodies, a high synaptic density, and numerous somatic spines. A third type of granule cell (type C) was found only in the deep subdivision of the superficial gray. This type had a low synaptic density and spines that contained synaptic vesicles. Vertical fusiform and stellate forms were also found. We conclude that at least six types of neurons populate the upper superficial gray layer of the cat superior colliculus.

GABAergic and non-GABAergic neurons in the nucleus of the optic tract project to the superior colliculus: an ultrastructural retrograde tracer and immunocytochemical study in the rabbit.

Both the nucleus of the optic tract (NOT) and the superior colliculus (SC) are thought to play important roles in the regulation of eye movements. The superior colliculus contributes to visual orientation and saccades, and the nucleus of the optic tract contributes to the detection of slow movements of the visual surround. Recently, a GABAergic projection has been described between these two nuclei in the cat, a species with frontal vision. The present study aimed at determining whether a similar GABAergic pathway exists in the rabbit, a species with lateral vision. To study this pathway we used the retrograde tracer cholera-toxin (CTB) to identify NOT neurons projecting to the SC and GABA-antibody immunostaining to identify GABA-containing neurons and processes. CTB injections into the superficial laminae of the SC showed that GABAergic and non-GABAergic neurons in the NOT project to the SC. Both types of neurons have structural characteristics similar to other projection neurons in the NOT. In contrast to the NOT neurons projecting to the inferior olive (IO) which are mainly located in the rostral NOT, the GABAergic and non-GABAergic NOT-SC neurons are situated throughout the nucleus. The somata and principal dendrites of both neuron types receive numerous synaptic contacts from GABAergic terminals and only a few from retinals. The NOT projection neurons to the SC thus establish prominent excitatory and inhibitory links between the two structures, suggesting the existence of separate circuits that could interact through a GABAergic and non-GABAergic NOT-SC projection. It is further suggested that these circuits may be involved in the regulation of saccades in the SC during optokinetic nystagmus.

Quantitative comparison of retinal synapses in the dorsal and ventral (parvicellular) C laminae of the cat dorsal lateral geniculate nucleus.

Three physiological classes of retinal ganglion cell project to the cat dorsal lateral geniculate nucleus (DLGN). The dorsal laminae A, A1, and magnocellular C receive X and Y retinal input, whereas the ventral parvicellular laminae C1 and C2 receive predominantly W input. We have compared quantitatively the retinal synaptic terminals of the dorsal and ventral laminae to determine whether there are morphological differences in the terminals that correspond to their different response properties. Anterogradely labeled retinal synaptic terminals in all laminae contained pale mitochondria and large, round synaptic vesicles. However, retinal terminals with pale mitochondria varied in size and synaptic organization in different laminae. The terminals in the A laminae were, on average, quite large and made numerous contacts with conventional dendritic profiles and with profiles that themselves contained synaptic vesicles (F2 profiles). The terminals in lamina C that contained pale mitochondria had a smaller overall mean area. Terminals with pale mitochondria in C1 and C2 were almost all small and synapsed with F2 profiles less frequently than did terminals in the A laminae or in lamina C. These results provide quantitative evidence that visual areas receiving W-type retinal input contain smaller retinal terminals and have a different synaptic organization from that of laminae receiving X and Y input.

Organization and synaptic connections of cholinergic fibers in the cat superior colliculus.

The cat superior colliculus (SC) receives a dense cholinergic input from three brainstem nuclei, the pedunculopontine tegmental nucleus, the lateral dorsal tegmental nucleus, and the parabigeminal nucleus (PBG). The tegmental inputs project densely to the intermediate gray layer (IGL) and sparsely to the superficial layers. The PBG input probably projects only to the superficial layers. In the present study, the morphology of choline acetyltransferase (ChAT)-immunoreactive axons and synaptic endings in the superficial and deep layers of the SC was examined by light and electron microscopy to determine whether these cholinergic afferents form different types of synapses in the superficial and deep layers. Two types of fibers were found within the zonal (ZL) and upper superficial gray layers (SGL): small diameter fibers with few varicosities and larger diameter fibers with numerous varicosities. Quantitative analysis demonstrated a bimodal distribution of axon diameters, with one peak at approximately 0.3-0.5 micron and the other at 0.9-1.0 micron. On the other hand, ChAT-immunoreactive fibers in the IGL were almost all small and formed discrete patches within the IGL. Two types of ChAT-immunoreactive synaptic profiles were observed within the ZL and upper SGL using the electron microscope. The first type consisted of small terminals containing predominantly round synaptic vesicles and forming asymmetric synaptic contacts, mostly on dendrites. The second type was comprised of varicose profiles that also contained round synaptic vesicles. Their synaptic contacts were always symmetric in profile. ChAT-immunoreactive terminals in the IGL patches contained round or pleomorphic synaptic vesicles, and the postsynaptic densities varied from symmetric to asymmetric, including intermediate forms. However, no large varicose profiles were observed. This study suggests that cholinergic fibers include at least two different synaptic morphologies: small terminals with asymmetric thickenings and large varicose profiles with symmetric terminals. The large varicose profile in the superficial layers is absent in the IGL. This result suggests that the cholinergic inputs that innervate the superficial layers and the patches in the IGL of the cat SC differ in their synaptic organization and possibly also in their physiological actions.

Development of the visual pathway is disrupted in mice with a targeted disruption of the calcium channel beta(3)-subunit gene.

Refinement of the retinal pathways to the superior colliculus (SC) and dorsal lateral geniculate nucleus (dLGN) is mediated by nitric oxide (NO). Long-term depression (LTD) can also be induced in SC and LGN during the time at which these pathways are refined, and this LTD is partially dependent on NO and L-type Ca(2+) channel function. In an effort to determine whether NO-mediated pathway refinement is also mediated by Ca(2+) channel function, we have examined the refinement of the retinocollicular and retinogeniculate pathways in mice which lack the gene for the Ca(2+) channel beta(3) subunit (CCKO) and which have significantly reduced L-type Ca(2+) currents. Injections of the anterograde tracer cholera toxin subunit B/HRP were made into one eye of these knockout animals and in wild-type mice ages postnatal day (P) 13, P19, and P26. After 48 hours, mice were perfused and sections processed by using tetramethylbenzidine histochemistry. Labeling distribution in some animals was analyzed quantitatively. Obvious differences in the distribution of the ipsilateral retinocollicular pathway were observed at P15, with the pathway being more exuberant in CCKO mice. This difference was statistically significant. More subtle differences were seen at P21 and P28. Obvious differences were also seen in the contralateral retinogeniculate pathway which in CCKO mice filled most of the domain normally occupied by ipsilateral eye fibers. This difference was also statistically significant. We conclude that reduction in L-type Ca(2+) currents has an effect on axonal refinement similar to that which occurs in NO knockout mice, which supports the possibility that L-type Ca(2+) channel-dependent LTD mediates NO-dependent axonal refinement.

CalbindinD28k- and parvalbumin-immunoreactive neurons form complementary sublaminae in the rat superior colliculus.

By using light microscopic immunocytochemistry and computer analysis, we have mapped the distributions of two calcium-binding proteins (CaBPs), calbindinD28k (CB) and parvalbumin (PV), in the rat superior colliculus (SC). The patterns of CaBP expression were complementary. A band of heavily labeled, medium-sized CB-immunoreactive cells (CB-cells) was centered in the optic layer (OL), whereas PV-immunoreactive cells (PV-cells) were found predominantly in the intermediate gray layer (IGL), where they were clustered within patches of PV-labeled fibers. The superficial gray layer (SGL) could be divided into two sublaminae. CB-cells were found mostly in the dorsal half of the SGL, whereas PV-cells were scattered throughout the ventral SGL and the dorsal OL. Most of the CaBP-immunoreactive cells in the SGL were small bipolar cells with vertically oriented dendrites; however, there were also some PV-cells with horizontally oriented dendrites. Quantitative analysis of the CaBP distributions reinforced our observations that these cells are distributed in complementary tiers that are not restricted to the traditional laminae. The size and shape of some of these tiers were determined from a three-dimensional reconstruction of serial sections. The complementarity of the CaBP-immunoreactive tiers was also confirmed by fluorescence microscopy of double-labeled sections, in which few if any double-labeled neurons were observed. Complementary tiers of CB-cells and PV-cells have been observed previously in the SC of the cat. The present results demonstrate them in another species and further suggest that there are functional sublaminae in the SC that can be distinguished by CaBP content.

Normal development of the ipsilateral retinocollicular pathway and its disruption in double endothelial and neuronal nitric oxide synthase gene knockout mice.

The development of the ipsilateral retinocollicular pathway involves activity-dependent refinement in which misdirected axons retract to form a precise retinotopic map in adults. This refinement is altered by disruption of genes for the endothelial and neuronal isoforms of nitric oxide synthase (e,nNOS), but the extent of disruption during early development is not known. Therefore, we studied the refinement of this pathway in normal C57/BL6 and e,nNOS double knockouts from P4 to P21 and in adults. Anterograde tracers were injected into one eye to localize the ipsilateral retinal projection (IRP) within the superior colliculus (SC). At P4, the IRP in normal mice was distributed throughout the dorsoventral extent of the superficial gray layer (SGL) across most of the rostrocaudal axis of SC. Between P4 and P9, the pathway retracted to the rostromedial SC, and retracted further between P15 and P21, such that multiple patches of label were seen only in the rostral 200-300 microm. Refinement also began to occur between P4 and P9 in e,nNOS double knockout mice, but labeling was more extensive in P9, P15, and P21 knockout animals. This delay in refinement was confirmed quantitatively at P15 where differences in the area occupied by the pathway were statistically significant. The refinement process is therefore in progress in both normal and e,nNOS knockout mice before eye opening but is significantly delayed in the double knockouts. The IRP in normal mice is also more exuberant at early ages, and the process of refinement more protracted than has been previously reported, suggesting that there is a prolonged critical period of synaptic plasticity.

Postnatal development of nitric oxide synthase expression in the mouse superior colliculus.

Since nitric oxide has a role in the refinement of the retinal projection to the superior colliculus (SC), we studied the onset of neuronal nitric oxide synthase (nNOS) expression in the mouse SC in order to compare its development with that of the refinement process. Sections from animals at ages P1, P5, P8, P11, P15, and P21 and adults were examined with nicotinamide adenine dinucleotide phosphate diaphorase (NADPHd) histochemistry or immunocytochemistry using an antibody directed against nNOS. At all ages there was a wedge of labeled neurons in the dorsolateral periaqueductal gray extending into the deep layers of the SC. At P1 there was also a single superficial band of labeled neurons within the region that will become the intermediate gray layer (IGL). By P5, labeled neurons were also seen in what will become the superficial gray layer. There was a ventral to dorsal progression in nNOS expression with substantial changes in the numbers of labeled neurons in the different laminae between P5 and adulthood. The number of labeled neurons in the IGL peaked at P15, whereas in the superficial layers the numbers continued to increase through P21 and then declined in adults. At all ages these neurons represented a variety of morphological cell types. The onset of nNOS expression in the different laminae is earlier than has been reported in studies using NADPHd as a marker for nNOS. The temporal and spatial patterns of nNOS expression reported here match more closely the time course of pathway refinement in the SC, providing additional evidence for the involvement of nitric oxide in this process.

The calcium binding protein calbindin-D 28K reveals subpopulations of projection and interneurons in the cat superior colliculus.

The calcium binding protein calbindin-D 28K (CaBP) has been localized in the cat superior colliculus (SC). Four important features of SC organization have been revealed by using CaBP immunocytochemistry. 1) CaBP neurons formed three laminar tiers in SC, one within the upper one half of the superficial gray layer (SGL), the second bridging the deep optic (OL) and intermediate gray layers (IGL), and the third within the deep gray layer (DGL). 2) CaBP labeled several classes of interneuron in SC. In the upper CaBP tier, the labeled neurons were all small, but they varied in morphology and included horizontal, pyriform, and stellate neurons. A unique class of interneuron was labeled by anti-CaBP in the OL-IGL tier. This cell was stellate-like with highly varicose dendrites and broad dendritic trees. Other labeled neurons in the intermediate and deep tiers included nonvaricose stellate neurons and rare large neurons in the DGL. 3) A few anti-CaBP neurons were projection neurons. Virtually no CaBP neurons were retrogradely labeled after injections of HRP into the predorsal bundle and dorsolateral midbrain tegmentum or into the lateral posterior nucleus. However, 2.4% of anti-CaBP neurons were retrogradely labeled after HRP injections into the dorsal and ventral lateral geniculate nuclei. These represented 14.7% of all neurons projecting to the LGN complex. 4) A small percentage of CaBP neurons co-localized GABA. A two-chromagen double-labeling technique showed that about 4.0% of labeled neurons were labeled by both antibodies. In summary, antibodies to CaBP densely labeled subpopulations of neurons in the cat SC, most of which were interneurons, some of which projected to the LGN, and a few of which co-localized GABA.

Ultrastructural organization of GABA in the rabbit superior colliculus revealed by quantitative postembedding immunocytochemistry.

We have studied the organization of gamma-aminobutyric acid (GABA)ergic profiles in the superior colliculus of the rabbit to determine whether the synaptic types found in cat and monkey also exist in a mammalian species whose visual system has a different organization. Ultrastructure of GABAergic profiles was examined by use of a polyclonal antibody to GABA and quantitative postembedding immunocytochemistry. Three distinct types of vesicle-containing profiles were labeled by the GABA antibody in the rabbit superior colliculus. One type was a putative presynaptic dendrite (PSD profile) that received synaptic input from other profiles and contained pleomorphic synaptic vesicles scattered throughout the profile. These PSD profiles frequently received retinal input and formed dendrodendritic synapses. A second type of profile was a large caliber dendrite, often horizontal in orientation (H profile), that had one or more discrete clusters of pleomorphic synaptic vesicles at sites of synaptic contact with conventional dendrites. These H profiles received few synaptic contacts. A third profile type was a putative axon terminal (F profile) with smaller, more flattened synaptic vesicles that densely and uniformly filled the profile. Quantitative analysis of gold particle density revealed that F profiles had a significantly higher gold particle density (14.3/microns 2) than did PSD or H profiles (10.4 and 10.2/microns 2), suggesting that GABAergic profile types contain different concentrations of GABA. The vesicle density of these profile types also differed, but no obvious relationship between vesicle and particle distributions was observed. We conclude that the profiles labeled by GABA in rabbit superior colliculus are similar to those in cat and monkey and must represent a phylogenetically conserved organization common to many mammals, and that particle density analysis of postembedding immunocytochemistry can distinguish different GABAergic profile types.