gamma-Aminobutyric acid and somatostatin immunoreactivity in the visual cortex of normal and dark-reared rats.

Our previous single unit and ultrastructural studies of visual cortex of dark-reared rats revealed an impairment of intracortical inhibitory mechanisms [2,3,5]. Neurochemical changes in inhibitory neurotransmitter and/or neuropeptides, such as gamma-aminobutyric acid (GABA) and somatostatin (SS), respectively, may contribute to the observed alterations. The present study was designed to measure GABA and SS alterations in the visual cortex of the same dark-reared preparation, as possible neurochemical correlates of the changes seen both physiologically and anatomically in previous companion studies. In the present investigation the mean densities of GABA- and SS-immunoreactive neurons in area 17 of dark-reared rats were determined and compared to the density of those of rats reared in normal lighting conditions. Dark-rearing resulted in a significant decrease in the density of GABA-immunoreactive neurons in all cell layers of area 17 of the rat visual cortex; not limited to the thalamorecipient layer(s). There was also a higher mean density of total cortical cells in dark-reared animals. No differences, however, were seen in the density of SS-immunoreactive neurons. The alterations of GABA-immunoreactive neurons in all cortical layers agree with the altered synaptic ultrastructure and physiological responses seen in all cortical layers as reported in our previous companion studies. Taken together, these studies further support the notion of a deficit in intracortical inhibitory mechanisms in the visual cortex of dark-reared adult rats.

Single neurons with both form/color differential responses and saccade-related responses in the nonretinotopic pulvinar of the behaving macaque monkey.

The nonretinotopic portion of the macaque pulvinar complex is interconnected with the occipitoparietal and occipitotemporal transcortical visual systems where information about the location and motion of a visual object or its form and color are modulated by eye movements and attention. We recorded from single cells in and about the border of the dorsal portion of the lateral pulvinar and the adjacent medial pulvinar of awake behaving Macaca mulatta in order to determine how the properties of these two functionally dichotomous cortical systems were represented. We found a class of pulvinar neurons that responded differentially to ten different patterns or broadband wavelengths (colors). Thirty-four percent of cells tested responded to the presentation of at least one of the pattern or color stimuli. These cells often discharged to several of the patterns or colors, but responded best to only one or two of them, and 86% were found to have statistically significant pattern and/or color preferences. Pattern/color preferential cells had an average latency of 79.1 +/- 46.0 ms (range 31-186 ms), responding well before most inferotemporal cortical cell responses. Visually guided and memory-guided saccade tasks showed that 58% of pattern/color preferential cells also had saccade-related properties, e.g. directional presaccadic and postsaccadic discharges, and inhibition of activity during the saccade. In the pulvinar, the mean presacadic response latency was earlier, and the mean postsaccadic response latency was later, than those reported for parietal cortex. We also discovered that the strength of response to patterns or colors changed depending upon the behavioral setting. In comparison to trials in which the monkey fixated dead ahead during passive presentations of pattern and color stimuli, 92% of the cells showed attenuated responses to the same passive presentation of patterns and colors during fixation when these trials were interleaved with trials which also required active saccades to pattern and color targets in the periphery. We conclude that properties which represent the functionally dichotomous object and spatial visual systems are found together in single pulvinar neurons and that the responses of these cells to pattern or color stimuli are influenced by the focus of spatial attention. The pulvinar is the first structure in the brain shown to have neurons which integrate both object and spatial properties and the response latencies indicate that this information is processed before that in cortex. These results are discussed in terms of role of the pulvinar in visual attention as well as its unique role in providing both object feature and spatial location information to the inferotemporal cortex.

Transient global amnesia and thalamic infarction.

We describe the clinical and neuroradiologic features of a patient with two episodes of transient amnesia who later developed persistent amnesia and an acute infarction in the left thalamus. The neurobehavioral manifestations were strikingly similar in all three episodes. Cranial computed tomography was normal following the first two episodes. Thalamic ischemia could explain some cases of transient global amnesia.

An investigation of collateral projections of the dorsal lateral geniculate nucleus and other subcortical structures to cortical areas V1 and V4 in the macaque monkey: a double label retrograde tracer study.

Previous anterograde studies in the macaque monkey have shown that, in addition to the projection to striate cortex (V1), the dorsal lateral geniculate nucleus (DLG) has a sparse, horizontally segregated projection to layers IV and V of prestriate cortex (V4). However, the distribution and degree of axon collateralization of DLG cells which give rise to these projections are unknown. This study was designed to answer these questions. The DLG (along with the pulvinar and other subcortical regions) was examined for the presence of single- or double-labeled cells after injections of two different (fluorescent or HRP) retrograde tracers into corresponding retinotopic points in visual cortical areas V1 and V4. In the DLG, it was found that cells projecting to V4, which reside in or near the tectorecipient interlaminar zones of the DLG, do not project to V1 and thus represent a separate population of cells. The organization of the macaque geniculo-prestriate projection thus seems quite different from that of carnivores. Both single- and double-labeled cells were found in other subcortical areas, e.g., single-labeled cells were found in the claustrum, hypothalamus and lateral pulvinar, and a double-labeled cell population was found in the inferior pulvinar.

Histochemical and architectonic differentiation of zones of pretectal and collicular inputs to the pulvinar and dorsal lateral geniculate nuclei in the macaque.

The pattern of acetylcholinesterase (AChE) reactivity was studied in the pulvinar and dorsal lateral geniculate nucleus (DLG) of the adult macaque monkey. Discrete islands of AChE reactivity were found that correlated precisely in location with the pattern of projections from the superior colliculus and pretectum. Specifically, AChE overlies terminal fields of superior colliculus projections in the DLG, in four foci in the medial and lateral pulvinar, and in several foci in the inferior pulvinar. All of these tectal projection areas have very high AChE reactivity such that they are easily distinguished. In addition, the pretectum projects to a specific focus in the lateral pulvinar that also has a very dense AChE histochemical reaction. A number of these AChE foci could be further distinguished from other areas in the pulvinar by myeloarchitectonic characteristics. Some of the foci in the lateral and inferior pulvinars could also be distinguished by unique cytoarchitectonic features (as seen with both Nissl and Golgi stains). In an attempt to determine the possible origin of a cholinergic input to the pulvinar, horseradish peroxidase (HRP) injections and choline acetyltransferase immunohistochemistry were also done. The results of this experiment indicate that the AChE reactivity seen in the midbrain projection zones to the thalamus may be due to the precise overlap of terminal projections from the brainstem cholinergic cell groups, Ch5, Ch6, and Ch8. These results, taken together, imply that there are several anatomically and histochemically distinct zones related to extrageniculate pathways located within classically defined thalamic boundaries.

Direct retinal pathways to the limbic thalamus of the monkey.

Tritiated proline, horseradish peroxidase (HRP), and wheat germ agglutinin conjugated to HRP (WGA-HRP) were used as anterograde tracers in the monkey to reveal visual pathways. After intravitreal injections, three separate, direct routes of labeled retinal axons were followed to the thalamus. These routes eventually converged to innervate the lateral dorsal and anterodorsal thalamic nuclei. Thus, retinal input may reach the posterior cingulate cortex after a single synapse in lateral and anterior thalamic nuclei.

Visual responses of single neurons in the caudal lateral pulvinar of the macaque monkey.

Single unit recordings were made in the portion of the lateral pulvinar which forms the lateral aspect of the caudal pole of the thalamus, i.e., PL gamma (Rezak, M., and L. A. Benevento (1977) Soc. Neurosci. Abstr. 3:573; Rezak, M. (1978) Soc. Neurosci. Abstr. 4: 642), of macaque monkeys. PL gamma receives convergent inputs from the occipital cortex and has strong reciprocal interconnections with the visual association cortex, including the inferotemporal cortex (areas 20 and 21). It was found the that PL gamma has a poor or nonexistent retinotopic organization. Many of the neurons had large, unflanked, overlapping receptive fields which often included the fovea. A few neurons could be influenced by a visual stimulus placed anywhere in the visual field described by a tangent screen. The receptive fields could be bilateral or located entirely within the contralateral or ipsilateral hemifields. The majority of units were binocular and exhibited various types of binocular interaction which could be quite complex. The binocular response was not predictable from the algebraic sum of the monocular responses and could be of the opposite sign (e.g., excitatory when the monocular response was inhibitory). Neurons which were also sensitive to the direction of movement of stimuli projected upon the tangent screen formed a major group. Of the units sensitive to tangentially moving stimuli, two special subgroups were found. One group of neurons gave sustained responses to static levels of luminance, while the other group was sensitive to simuli which moved toward or away from the eyes. The nonlinear rate of change of the apparent size of approaching or receding stimuli was described by a mathematical function which also describes the response of the neurons to the same stimuli. For many of these units which were sensitive to tangentially moving stimuli and one other class of stimuli, such as luminance levels of movement in depth, the responses to one class were seemingly unrelated to the responses to the other class. The same statement may be made for monocular and binocular responses. It may be, then, that different wiring diagrams describe these different types of inputs. These physiological results are discussed in terms of the inputs to PL gamma as well as its cortical targets.

The projection from the dorsal lateral geniculate nucleus of the thalamus to extrastriate visual association cortex in the macaque monkey.

Both autoradiographic and horseradish peroxidase tracing techniques were used to characterize a projection from the dorsal lateral geniculate (DLG) nucleus of the thalamus to visual association cortex (extrastriate cortex) in the macaque monkey. The results show that medium to large caliber DLG axons end to discontinuous terminal "patches" in layers V and lower IV of extrastriate cortex. There is a topographical organization to these projections which are mainly to area 19 and anterior 18 located on the lateral and medial surfaces of the hemisphere. Apparently there are no DLG projections to any of the cortical subdivision located within the lunate and superior temporal sulci.

An autoradiographic study of the projections of the pretectum in the rhesus monkey (Macaca mulatta): evidence for sensorimotor links to the thalamus and oculomotor nuclei.

Autordiographic tracing methods were used to determine the differential projections of the pretectal nuclei, in the rhesus monkey, in relation to their inputs. The sublentiform (SL) and olivary (ON) nuclei receive projections from the visual cortex, superior colliculus (SC) and equal bilateral projection from the retina. The nucleus of the posterior commissure (NPC) and its subdivisions do not receive any of these inputs. The projections of the pretectum involve a number of structures within the thalamus and brain stem and there are differences in the projection targets of the pretectal region which receives direct visual input (i.e., SL and ON) and the region which does not (i.e., nucleus of the posterior commissure, NPC). For example, while all pretectal regions project within the pretectum and to the SC, accessory oculomotor nuclei, reticular formation, intralaminar nuclei and hypothalamus, it is only the retinorecipient zone which projects to rostral regions such as the visceral oculomotor nuclei, the lateral pulvinar, the border between the lateral pulvinar and medial pulvinar, the oral pulvinar as well as to the thalamic reticular nucleus, ventral lateral geniculate nucleus, zona incerta and other structures. It is concluded that the retina, SC and cortex which influence the visceral oculomotor nuclei can only do so by virtue of their projections to the pretectum, and that any consideration of accommodative and pupillary reflexes must view the pretectum as an obligatory link through which various structures can influence the intrinsic musculature of the eye. In contrast to the SC, the pretectum does not project to any of the visual relay nuclei of the thalamus, such as the inferior pulvinar, which project to the visual cortices. Instead, the pretectum projects directly to visuomotor, visceromotor and arousal systems.

The ascending projections of the superior colliculus in the rhesus monkey (Macaca mulatta).

We studied and compared the ipsilateral efferents of the superficial and deep layers of the superior colliculus of the rhesus monkey. Using a stereotaxic method, microelectrodes were inserted through the contralateral hemisphere in order to make electrolytic lesions of the superior colliculus. Large lesions involved all layers of the superior colliculus, while smaller lesions involved either the superficial or the deep layers of the superior colliculus. Following various survival times, the brains were prepared with the Fink-Heimer technique ('67). Following lesions of the superficial layers of the superior colliculus, definite degenerated axonal endings were found in the dorsal and ventral lateral geniculate nuclei, inferior pulvinar, centrointermediate nucleus, magnocellular dorsomedial nucleus, anterior pretectal nucleus and pretectal region. Sparse degenerated axonal endings were found in the limitans nucleus, lateral posterior nucleus and some intralaminar nuclei following lesions of the superficial layers in the rostral portion of the superior colliculus. Following lesions of the deep layers of the superior colliculus, degenerated axonal endings were found in the central gray, magnocellular medial geniculate nucleus, suprageniculate nucleus, limitans nucleus, lateral posterior nucleus, medial and oral pulvinar, nucleus of the accessory optic tract, zona incerta, subdivisions of the ventral lateral and ventral posterior lateral nuclei, ventral posterior inferior nucleus, denosocellular and multiform dorsomedial nuclei, all intralaminar nuclei, inferior colliculus, parabigeminal nucleus, olivary nucleus, reunions nucleus, Forel's Field H and an undefined midbrain nucleus. In general the projections were topographically organized in that the caudal portion of the superior colliculus projected to the rostral portions of thalamic nuclei and the rostral portion of the superior colliculus projected to the caudal portions of thalamic nuclei. All the degeneration patterns seen after lesions of the superficial and deep layers were accounted for by large lesions which involved all layers of the superior colliculus. It is concluded that the superficial and deep layers of the rehesus monkey superior colliculus have different ascending projections. The finding, are related to the organization of visual and multimodal thalamocortical systems in primates and other mammals.

Stereotaxic coordinates for the Rhesus monkey thalamus and mesencephalon referencing visual afferents and cytoarchitecture.

When using a stereotaxic instrument for visual field stimulation we found that electrode placements in the thalamus and mesencephalon of prone rhesus monkeys with the aid of avaiable atlases showed considerable errors. As these animals are valuable for primate visual system reseach an atlas was constructed with methods that have not been used before for rhesus. In addition, the specific connections from the visual cortices, superior colliculus and retina to the thalamus and mesencephalon are also shown. Anesthetized monkeys of specific body dimensions had a matrix of pins inserted into the brain before fixation. A matrix was used so that the penetrations seen in the sectioned brain could be cross related as a control for accurate measurements of the stereotaxic planes throughout the brain. The surface of the whole brain frozen blocks were photographed on the microtome just before a cut section was taken. These calibrated pictures formed the "floor plan" of the atlas as they represent more accurately the brain geometry than individual sections which are distorted by cutting, staining and mounting. Cytoarchitectural (Nissl stain) and axonal connectional (Fink-Heimer stain) information was transferred and adjusted onto the block pictures from their corresponding stained sections. Follow up experiments showed that the present coordinates are accurate for these monkeys of restricted body dimensions. In addition, referencing visual axonal projections onto the same cytoarchitectural map in stereotaxic coordinates provides an atlas for localizing areas of the thalamus, on a basis other than cytoarchitecture, which receive combinations of visual inputs for further anatomical and physiological studies of the rhesus monkey visual system. The atlas further demonstrates that projections do not necessarily follow the cytoarchitectural definition of an area, but rather redefine the thalamus on the basis of specific axonal connections.