Visual receptive fields of neurons in inferotemporal cortex of the monkey.

Neurons in inferotemporal cortex (area TE) of the monkey had visual receptive fields which were very large (greater than 10 by 10 degrees) and almost always included the fovea. Some extended well into both halves of the visual field, while others were confined to the ipsilateral or contralateral side. These neurons were differentially sensitive to several of the following dimensions of the stimulus: size and shape, color, orientation, and direction of movement.

Patterns of corticocortical, corticotectal, and commissural connections in the opossum visual cortex.

Patterns of connections of the visual cortex of the South American opossum, Didelphis aurita, were revealed by using neuronal tracers to identify and characterize visual specializations of the peristriate cortex (PS). The visuotopy of corticotectal connections of the anterolateral portion of PS (PSal) is symmetrical to that of the striate cortex (ST or primary visual area [V1]). Three consecutive bands of commissural connections coincide, respectively, with the ST-PS border, the limit between the caudal and rostral PSal halves (PSc and PSr), and the border of PS with the parietal and temporal cortices. PSc and PSr contain regular commissural rings similar to those present in the peristriate cortex of eutherian mammals. ST projections define in PSc two strings of periodical foci consecutively concentric to V1 and a single focus in PSr. Although they were organized topographically, ascending, descending, and commissural connections between ST and PSal showed a high degree of convergence and divergence. These results conform to the model of a single area homologous to the second visual area (V2) bordering V1. Moreover, they suggest the possibility that PSal includes either one or two additional belt-like areas successively anterior to V2. Along with the finding of alternating bands of high and low cytochrome oxidase activity in PSal, the data further suggest that this region contains modular specializations similar to those of the peristriate cortex of primates and other eutherian mammals. The posterolateral peristriate cortex (PSpl) constitutes another visual area, since it consists of a distinct focus of reciprocal corticocortical and interhemispheric connections and a separate source of corticotectal projections. Finally, a visuomotor function for the orbital cortex is proposed based on its direct projections to optical tectal layers. The close cladistic relationship of opossums to mammalian ancestral forms suggests that the PSal parcelation into belt-like areas that contain modules reflects the primitive organization of the visual cortex. Moreover, a highly diffuse pattern of corticocortical connections may represent a requirement for a brain with few visual areas to perform global processing.

Cytochrome oxidase and NADPH-diaphorase on the afferent relay branch of the optokinetic reflex in the opossum.

In the present study, histochemical techniques combined with more conventional anatomical methods were used to refine the identification of the nucleus of the optic tract and the nuclei of the accessory optic system in the opossum. The distribution of the enzyme cytochrome oxidase (CO) was examined in the cells and the neuropil of the opossum's mesodiencephalic region. Strong CO labeling was present in the nucleus of the optic tract (NOT)-dorsal terminal nucleus (DTN). Alternate sections, taken from animals that had received bilateral injections of horseradish peroxidase centered in the region of the inferior olive, were subjected to assays for CO and horseradish peroxidase. The region occupied by CO-labeled cells in the NOT-DTN superimposed with the one defined by retrogradely labeled cells. Cell counts along the NOT-DTN anteroposterior axis revealed that although the olivary and CO-positive cells were confined within similar boundaries, the latter are up to twofold more numerous than the former. As revealed by cytochrome oxidase histochemistry, the outlines of the NOT-DTN, the other pretectal nuclei and the nuclei belonging to the accessory optic system coincided with those revealed by the histochemistry for nicotinamide dinucleotide phosphate diaphorase (NADPH-d). After an intraocular injection of cholera toxin beta subunit and alternate sections processing for NADPH-d and CO, the distribution of labeled retinal terminal fields in the mesodiencephalic region was shown to be coincident with regions of high levels of histochemical labeling. These results are discussed in the light of previous anatomofunctional assessments of the pretectum and accessory optic system.

Cortical and subcortical influences on the nucleus of the optic tract of the opossum.

In the present work we propose a new phylogenetic hypothesis for the role played by cortical and subcortical afferents to the nucleus of the optical tract, the main visual relay station of the horizontal optokinetic reflex in mammals. The hypothesis is supported by anatomical and physiological data obtained in the South American opossum (Didelphis aurita) using the following experimental approaches: (i) single-unit recordings in the nucleus of the optic tract and simultaneous electrical stimulation of the contralateral nucleus of the optic tract; (ii) single-unit recordings in the nucleus of the optic tract and simultaneous electrical stimulation of the ipsilateral striate cortex; (iii) injection of cholera toxin subunit B into the striate cortex and subsequent immunohistochemical reaction to reveal the presence of the marker in the thalamus and mesencephalon; and (iv) single-unit recordings in the nucleus of the optic tract both before and after ablation of the ipsilateral visual cortex. The main results are: (i) there is a strong inhibitory reciprocal effect upon the nucleus of the optic tract following stimulation of its contralateral counterpart; (ii) electrophysiological and anatomical data imply that the visual cortex does not project directly to the nucleus of the optic tract. Rather, cortical terminals seem to target the nearby anterior and posterior pretectal nuclei and orthodromic latencies in the nucleus of the optic tract following stimulation of the visual cortex were twice as large as in the superior colicullus; and (iii) ablation of the entire visual cortex did not have any effect upon binocularity of cells in the nucleus of the optic tract. These results strengthen the model proposed here for the role of the interactions between the nuclei of the optic tract under optokinetic stimulation. The hypothesis in the present work is that the cortical influences upon the nucleus of the optical tract, in addition to the subcortical ones, appeared only recently in phylogenesis. In more primitive mammals, such as the opossum, subcortical interactions are thought to play a relatively important role. With the emergence of retinal specializations, such as the fovea, one might suppose that there followed the appearance of new ocular movements, such as the smooth pursuit and certain types of saccades, that came to join the pre-existent optokinetic reflex.

On the functional anatomy of the nucleus of the optic tract-dorsal terminal nucleus commissural connection in the opossum (Didelphis marsupialis aurita).

Immunocytochemical methods revealed the presence of GABA in cell bodies and terminals in the nucleus of the optic tract-dorsal terminal nucleus, the medial terminal nucleus, the lateral terminal nucleus and the interstitial nucleus of the superior fasciculus of the opossum (Didelphis marsupialis aurita). Moreover, after unilateral injections of rhodamine beads in the nucleus of the optic tract-dorsal terminal nucleus complex and processing for GABA, double-labelled cells were detected in the ipsilateral complex, up to 400 microns from the injected site, but not in the opposite. Analysis of the distributions of GABAergic and retrogradely-labelled cells throughout the contralateral nucleus of the optic tract-dorsal terminal nucleus showed that the highest density of GABAergic and rhodamine-labelled cells overlapped at the middle third of the complex. Previous electrophysiological data obtained in the opossum had suggested the existence, under certain conditions, of an inhibitory action between the nucleus of the optic tract-dorsal terminal nucleus of one side over the other. The absence of GABAergic commissural neurons may imply that this inhibition is mediated by an excitatory commissural pathway that activates GABAergic interneurons.

The pretectal complex in the opossum: projections from the striate cortex and correlation with retinal terminal fields.

Two terminal fields were revealed in the pretectal complex of the opossum by the Fink-Heimer method after striate cortical lesions. A rostral field is located within a rostrolateral strip of the compact part of the anterior pretectal nucleus, where a partial topographic arrangement of this projection is present. A caudal field is located within the sub-brachial nucleus of the optic tract, located between the brachium of the superior colliculus and the posterior pretectal nucleus. The corticotopic projection to this field is mirror-symmetric to that found in the superior colliculus and overlaps a bilateral projection from the retina. Based on neural pathway evidence, it is concluded that the nucleus of the optic tract in the opossum can be subdivided in (a) an intrabrachial nucleus receiving a direct projection from the contralateral retina and (b) a sub-brachial nucleus receiving projections from both retinae and from the striate cortex. The pretectal complex, as the superior colliculus, can be anatomically subdivided in a superficial region receiving visual input (the optic pretectum) and a deep region only remotely connected to the visual system. The optic pretectum, however, differs from the superior colliculus in displaying a multiple-map arrangement within its constituent nuclei, instead of a single continuous representation of the visual field.

Postnatal development of retinogeniculate, retinopretectal and retinotectal projections in the opossum.

The postnatal development of retinogeniculate, retinopretectal and retinotectal projections has been studied by the anterograde transport of proline-labeled proteins in 20 pouch young opossums aged from 10 to 60 days. Radioautographical findings suggest delayed development of uncrossed as compared to crossed projections. There is a phase of overlapping projections from both eyes in thalamic and tectal target sites. Partial segregation of projections to the dorsal lateral geniculate nucleus (GLD) is preceded by differential distribution of crossed and uncrossed terminal fields along its dorsoventral axis (at age 23 days). The quasilaminar pattern of projections in the dorsocaudal region of GLD pars alpha is incipient by 42 days and is fully established at 60 days of age, as eye opening starts. The mature pattern of projections to the ventral lateral geniculate nucleus (GLV) is established much earlier, at 23 days of age. The development of retinopretectal projections is assessed mostly from the analysis of the olivary pretectal nucleus (PO). Distribution of silver grains into discrete areas coextensive with PO is relatively delayed (by 23 days of age) as compared to the nuclei of the lateral geniculate body. Soon after, however, the mature pattern of projections to PO is established (at 33 days of age). The early development of retinotectal projections (from 10 to 23 days) is compatible with an initial tangential course of crossed optic fibers in the superior colliculus (CS) but other alternatives remain open, such as a sequential outside-in arrangement of terminal fields of deeply coursing fibers. Arborization of uncrossed fibers is delayed at extreme rostromedial and caudolateral portion of the territory of the main uncrossed retinotectal projection. Segregation of uncrossed projections at different depths of CS is nearly complete by 42 days. Differences in the development of terminal fields in different target nuclei or in regions of a given target site are discussed in relation to retinal and local factors.

Observations on postnatal neurogenesis in the superior colliculus and the pretectum in the opossum.

Postnatal neurogenesis has been detected in the superior colliculus (CS) and caudal pretectum of the opossum in the period ranging from 2 to 13 days ( PND2 to PND13 ) of life in the pouch. Examination of the pattern of labeling in specimens exposed to a pulse of tritiated thymidine ( [3H]T) in PND4 or PND7 and allowed 1.5 h survival reveals that postnatal cell proliferation for the CS is virtually confined to the ventricular zone with no evidence for in situ[3H]T uptake in the collicular plate. Semi-quantitative analysis in long survival animals shows that postnatal neurogenesis peaks later in the CS ( PND7 ) than in the caudal pretectum ( PND4 ) and also persists longer in the former than in the latter. Comparisons of the numerical density of heavily labeled neurons suggest the occurrence of ventro-dorsal and rostro-caudal gradients of neurogenesis in the CS. Separate analysis of superficial, intermediate and deep layers shows, in addition, a combined rostrolateral-to-caudomedial gradient of neurogenesis in the superficial layers. Comparisons of the time schedules of neurogenesis for the superficial layers and of the deployment of optic fibers suggest that migration of neurons to their eventual destination is completed at or after the arrival of afferents.

Effects of monocular enucleation at different stages of development on the uncrossed retinocollicular projection in the opossum.

Pouch young opossums aged from 5 to 34 days underwent unilateral eye enucleation and the projection from the remaining eye to the superior colliculus (SC) was explored at maturity by anatomical and physiological methods. The results were compared to data from normal or adult enucleated animals. Analysis of the experimental group showed that early enucleation resulted in an increased ipsilateral pathway which varied in terminal distribution with the timing of the lesion. When performed in the very young (first and second postnatal week) the uncrossed retinocollicular projection covered the entire area of the superficial layers of the superior colliculus but formed mediolateral bands of high and low label density at the rostral aspects of the SC. In the remainder of the early enucleated group the labeling also covered the entire area of the optic layers but the density of labeling was relatively homogeneous throughout the rostrocaudal extension of the SC. The distribution pattern of the uncrossed projection in adult enucleated opossums was similar to that described previously in normal animals. The electrophysiological results showed that, at all ages examined, early enucleation resulted in an orderly but expanded map in the SC restricted to the sector of the visual field seen by the temporal retina. The different anatomical patterns found in animals enucleated at these two early stages of development could not be distinguished by multiunit recordings.

Receptive field properties of single units in the opossum striate cortex.

On the basis of their trigger-features, 98 units out of 127 recorded in striate cortex of immobilized opossums, under forced breathing of a nitrous oxide/oxygen mixture, were classified into 5 receptive field groups. Group 1 units (20/127) responding to small stationary spots were shown to be made up of regions of opposite response type and mutual antagonism, separate by linear boundaries. The optimal discharge was elicited by a stimulus configuration consisting of rectilinear regions of opposite contrast positioned and oriented in the visual field so as not to elicit antagonism while maximizing the overlap with regions responsive to that contrast. To edges in motion these units were shown to be made up of light and dark discharge centers, the locations of which could not be predicted from the map to stationary spots. In addition to position and orientation, direction was another important stimulus parameter. Group 2 units (34/127) had uniform requirements of stimulus orientation, direction of motion or both, througout the receptive field. Width was rarely a significant variable. Three subgroups were detected: orientation selective, directional selective and orientation-direction sensitive. Group 3 units (18/127) required stopped stimuli. In most instances (14/18) this property was attributed to a suppressive surround with relatively non-specific stimulus requirements. Oriented and non-oriented responsive receptive fields were observed. Group 3 units with no surround (4/18) responded best to properly positioned and oriented wedges, usually of 90 degrees. Group 4 units (24/127) had uniform fields with little stimulus specificity and were often responsive to diffuse light. Although not sensitive to stimulus orientation and direction, motion was frequently a requisite for optimal responses. Group 5 receptive fields (2/127) had concentrically arranged regions of distinct response type which displayed mutual antagonism. No sensitivity to orientation or direction was detected. Twenty-nine units remained unclassified. Other group distinctions were the relatively higher spontaneous activity of group 4 units and the large field sizes encountered among groups 1 and 4 when compared to group 2. Based on their properties and receptive field type distribution, we propose that striate receptive fields in the opossum have a similar organization to those of other mammals.

Growth and restriction of the ipsilateral retinocollicular projection in the opossum.

The distribution of optic nerve fibers and terminals in the superior colliculus (SC) was followed throughout its development in pouch young opossums in order to establish the normal sequence of events leading to the formation of mature patterns. Up to 7 days of life in the pouch, labeled fibers can be followed only as far as the rostral aspect of the optic tract. The earliest evidence for crossed retinal projections in the SC is found at 10 days of age. In parasagittal sections, the label extends along the rostrocaudal tectal axis from the rostral border to the presumptive caudal pole of the SC. Unequivocal evidence for ipsilateral retinocollicular projection is found at 15 days extending to all but the caudal 5th of the rostrocaudal extent of the SC. The projections from both eyes overlap extensively in the SC at 22 days and after this age significant changes occur, mostly at the ipsilateral side: a sub-pial tier of fine label develops excluding both rostral and caudal collicular poles; a deeper tier of coarse label extends from the rostral to the caudal pole and a third, patchy tier of label is found at the prospective strata griseum superficiale and griseum intermediate. By 47 and 60 days the tangential distribution of the projections is virtually indistinguishable from the adult pattern although laminar segregation does not seem as sharp as in the adult. Comparisons of the changeable patterns of ipsilateral retinocollicular projections from 22 to 34 days with the invariant, aberrant pattern in adult animals submitted to uniocular enucleation at either age suggests that the preservation of a juvenile pattern does not provide a comprehensive explanation for the formation of aberrant projections.

Distribution of NADPH-diaphorase cells in visual and somatosensory cortex in four mammalian species.

The distribution of the well-labeled nicotinamide adenine dinucleotide phosphate diaphorase (NADPHd) Type I neurons was evaluated in the isocortex of four mammalian species: the Didelphis opossum, the Monodelphis opossum, the rat and the marmoset. In Didelphis opossum, laminar distribution was examined in tangential and non-tangential sections. The density increases from superficial to deep layers of the gray matter. In rats' tangential sections, infragranular and supragranular layers have higher density than layer IV. Cell density measurements in the visual and the somatosensory cortices were compared in tangential sections from flattened hemispheres of the four species. Somatosensory areas were identified histochemically in rat (barrel fields) and marmoset (S1 and S2/PV). In the opossums, areas S1 and S2/PV were identified by multiunit recording. Except in the rat, primary visual cortex (V1) was labeled histochemically by NADPHd and/or cytochrome oxidase. In the four species, cell density in somatosensory cortex was significantly higher than in visual cortex. Taken together these results demonstrate that NADPHd Type I neurons are not homogeneously distributed in the isocortex of these mammals. In conclusion, the tangential distribution of Type I neurons in the sensory areas examined, but not its laminar distribution, was similar in the four species. Given that rats, marmosets and opossums are distantly related species, and that the latter are considered to have more 'generalized' brains, it is conceivable that this pattern of tangential distribution of Type I neurons is a general feature of mammalian isocortex.

The horizontal optokinetic reflex of the opossum (Didelphis marsupialis aurita): physiological and anatomical studies in normal and early monoenucleated specimens.

In the opossum the symmetrical binocular horizontal optokinetic nystagmus gives way to an asymmetrical monocular reflex: the nasotemporal (NT) stimulation yielding lower gain than the temporonasal (TN). In adults, monocularly enucleated at postnatal days 21-25 (pnd21-25), the gain of NT responses is markedly increased, approaching that of TN. Severe cell loss was detected in the nucleus of the optic tract (NOT) on the deafferented side in early monoenucleated specimens. In normal animals retinal afferents to the NOT are all crossed, while in animals enucleated at pnd21-25 sparse uncrossed retinal elements were observed. Although this abnormal projection might influence the increased NT response in this subgroup, it is argued that the increased symmetry in monoenucleated opossums may be the result of changes mediated by the commissural connection between both NOTs.

Organization of the visual thalamus: corticothalamic projections from the primary visual area in the opossum.

The projections from the visual cortex to thalamic nuclei in the opossum were investigated by degeneration and radioautographic methods. Efferent axons from the striate cortex course from the internal capsule to the thalamus in two bundles, one of which innervates the reticular and dorsal lateral geniculate nuclei, while the other bundle innervates the external layer of the ventral lateral geniculate nucleus. Both bundles course along and through terminal fields found in the lateral posterior nucleus. The visuotopic organization inferred from the corticofugal pathway shows that projection lines in the dorsal lateral geniculate nucleus (DLGN) run along medio-lateral strips including both the beta and alpha segments, across the quasi-laminae of retino-geniculate projection, and that naso-temporal axes in the visual field run in a dorso-ventral direction through the DLGN. The external layer of the ventral lateral geniculate nucleus receives a projection from the striate cortex and bilateral projections from the retina, and naso-temporal axes in the visual field are represented along a dorso-ventral direction in this layer. The striate cortical projections cover approximately the lateral two-thirds of the lateral posterior nucleus, overlapping a small retinal terminal field, and naso-temporal axes in the visual field are represented onto the cortico-recipient zone in a mirror-symmetric direction to that of the adjoining DLGN. In all these three cortico-recipient zones, dorso-ventral axes in the visual field are represented in a rostro-caudal direction.

Patterns of cytochrome oxidase activity in the visual cortex of a South American opossum (Didelphis marsupialis aurita).

The normal pattern of cytochrome oxidase (CO) activity in the posterior cortical areas of the South American opossum (Didelphis marsupialis aurita) was assessed both in horizontal sections of flattened cortices and in transversal cortical sections. The tangential distribution of CO activity was uniformly high in the striate cortex. In the peristriate region alternating bands of dense and weak staining occupied all the cortical layers with the exception of layer I. This observation suggests the existence of a functional segregation of visual processing in the peristriate cortex of the opossum similar to that present in phylogenetically more recent groups.

Horizontal optokinetic reflex in the opossum Didelphis marsupialis aurita.

Electro-oculographic recordings were performed in 10 opossums. The optokinetic reflex was elicited by projecting a random dot stimulus on a cylindrical screen moving horizontally from left to right or right to left at various constant speeds. Binocular stimulation yielded the same response as the temporal to nasal monocular condition. The nasal to temporal monocular response was always less than that to the opposite direction: 50% at 3 degrees/s and 15% at 18 degrees/s. These results are discussed in a comparative context.

Evidence for recrossing of the pathway from the retina to the ipsilateral nucleus of the optic tract.

Single-unit recordings of the nucleus of the optic tract (NOT) under visual stimulation were performed in 5 opossums. Most of the units were directionally selective. Receptive fields for the contralateral eye lie mainly in the contralateral field while those for the ipsilateral eye were mainly in the ipsilateral field. As the nasal retina does not project ipsilaterally, recrossing must occur in the pathway from the retina to the ipsilateral NOT. Possible sites for this recrossing are discussed.