Library

Notes: The results of electrical stimulation experiments [Bullier et al., (1988) Exp. Brain Res., 70, 90–98] demonstrated that afferents from areas 18 and 19 contact different functional types of neurons in area 17. We were therefore interested in examining whether these results could be explained by differences in the morphology of the terminals of these two groups of afferent connections to area 17. We also wanted to confirm, by a direct method, our earlier results [Salin et al. (1989) J. Comp. Neurol., 283, 486–512] that cortical afferents to area 17 in the cat present extensive divergences. We therefore placed small injections of anterograde tracers in areas 18 and 19 and examined the laminar distributions of terminals thus revealed and the extent of the surface of area 17 contacted by these terminals. Three tracers were used: wheat germ agglutinin – horseradish peroxidase (WGA–HRP), Phaseolus vulgaris leucoagglutinin (Pha-L) and biocytin. The results show that the divergence of these afferent connections are very extensive: 7–8 mm in the rotrocaudal direction and 3.5–6 mm in the mediolateral direction. In other words, neurons located in a region a few hundreds micron wide in areas 18 or 19 contact a region of area 17 covering several millimeters. Corticocortical connections are therefore not organized in a point-to-point fashion but are strongly divergent. The laminar distributions of terminals from areas 18 and 19 displayed a specific pattern. Area 19 projects most heavily to layers 5 and 6, also terminates in layers 1–3 and very little is present in layer 4. In contrast, the afferent terminals from area 18 are heaviest in layers 1, 2, 3, 4A and 5 and are rare in layer 6. Injections placed at different depths in area 18 revealed that upper layer neurons in that area mostly project to layers 1, 2, 3 and 5 in area 17, whereas lower layer neurons send their heaviest projections to layers 4A, 5 and 6 and hardly project to layers 1, 2 and 3.

Notes: A fluorescent dye (usually fast blue or rhodamine tagged latex microspheres) was injected into cortical area 17 (or area 17 and the lateral part of area 18b) of adult and juvenile (15–22 day old) Sprague-Dawley albino rats. Another fluorescent dye (usually diamidino yellow) was injected into cortical areas 17, 18a and 18b of the opposite hemisphere. The injections involved only the cortical grey matter. After postinjection survival of 2–14 days the distribution of retrogradely labelled mesencephalic and prosencephalic cells was analysed. Both small and large injections labelled retrogradely a substantial number of cells in specific and nonspecific dorsal thalamic nuclei (lateral geniculate, lateral posterior, ventromedial, several intralaminar nuclei and nucleus Reuniens) as well as a small number of cells in the preoptic area of the hypothalamus and the mesencephalic ventral tagmental area (VTA). While labelled thalamic cells contained only the dye injected into the ipsilateral cortex, a small proportion of hypothalamic and VTA cells was labelled with the dye injected into the contralateral cortex. Virtually none of the cells in these areas were double labelled with both dyes. Both small and large injections labelled cells in the ipsilateral telencephalic magnocellular nuclei of the basal forebrain and the caudal claustrum. A substantial minority of labelled cells in these structures was labelled by the dye injected into the contralateral cortex. Furthermore, a small proportion (about 1%) of claustral cells projecting to the ipsilateral cortex were double labelled with both dyes. In several cortical areas ipsilateral to the injected area 17, associational neurons were intermingled with commissural neurons projecting to the contralateral visual cortex. A substantial proportion of associational neurons projecting to ipsilateral area 17 also projected to the contralateral visual cortex (associational-commissural neurons). Thus, in visual area 18a, the associational-commissural neurons were located in all laminae, with the exception of lamina 1 and the bottom of lamina 6, and constituted about 30% of the neurons projecting to ipsilateral area 17. In paralimbic association area 35/13, associational-commissural neurons were located in lamina 5 and constituted about 20% of neurons projecting to ipsilateral area 17. In the limbic area 29d, the associational-commissural neurons were located in laminae 4, 5 and the upper part of lamina 6 and constituted about 10% of the associational-commissural neurons projecting to ipsilateral area 17. In oculomotor area 8, double-labelled neurons were located in lamina 5 and constituted about 10% of the neurons projecting to ipsilateral area 17. Thus, it appears that the axons of mesencephalic and diencephalic neurons projecting to the visual cortex do not send collaterals into both hemispheres. The bihemispheric projection to the rat's visual cortex originates almost exclusively in the retinotopically organized cortical area 18a and in integrative cortical areas 35/13, 29d and 8.

Notes: Numerous cortical neurons in the juvenile and adult rat project to visual areas of both hemispheres whereas the vast majority of subcortical structures projecting to the visual cortex send strictly ipsilateral projections (Dreher et al., 1990). In the present study, the authors have sought to determine whether this pattern of axonal bifurcation in the connectivity of the visual areas undergoes a change during postnatal development. Two retrograde fluorescent dyes were used, fast blue (FB) and diamidino yellow (DY). Large multiple injections of one of the dyes were placed in all visual areas of one hemisphere and a small injection of the other dye was placed in area 17 of the opposite hemisphere. Labelled neurons were observed in subcortical and cortical structures on the side of the small injection. The experiments were performed on ten neonatal albino rat pups aged between 3 and 12 postnatal days (p.n.d.) at the time of injection and the results were compared with those obtained in the juvenile and adult animals, as reported in the preceding paper. In the thalamus of newborn animals, neurons belonging to nuclei located away from the midline send strictly ipsilateral cortical projections. However, in the midline nuclei of the intralaminar thalamic complex, a small region of overlap was observed between neurons projecting ipsilaterally and neurons projecting contralaterally in animals aged less than 9 postnatal days. In addition, in these neonatal animals a small number of bilaterally projecting neurons was detected in this region of overlap. In all other subcortical structures examined (ventral tegmental area, diagonal band of Broca, claustrum), the laterality of the projection was the same in the newborn and the adult animals. In particular, in the claustrum of neonatal animals, as in adult animals, there was a large contingent of contralaterally projecting neurons and only a very small number of bilaterally projecting neurons. The results in the cortex contrast with those observed in subcortical structures. Whereas ipsilaterally projecting neurons were distributed in a broadly similar way in newborn and adult animals, the laminar and areal distribution of contralaterally projecting neurons in newborn animals clearly differed from those observed in the adult animals. Furthermore, double labelled neurons were more numerous in animals aged less than 12 days than in adults. The proportions of such bilaterally projecting neurons were computed with respect to the numbers of neurons sending ipsilateral projections to area 17. These proportions are constant at all ages in the claustrum and cortical area 8. In areas 18a, 29 and 35 on the other hand, the proportions of bilaterally projecting neurons increase after 5 days and reach a peak in the period extending from 9 to 11 days of age when more than half of the neurons projecting ipsilaterally also send an axonal branch to the contralateral cortex. In cortical areas 29 and 35, this peak is followed by a sudden drop to the adult level at 12 postnatal days, whereas the return to the adult level is gradual in area 18a. These results demonstrate that, in subcortical structures and in cortical area 8, the laterality of the afferent connections to the visual cortex does not change during postnatal development. By contrast, cortical areas 18a, 29 and 35 go through a stage when numerous cells send bifurcating connections to both hemispheres. The timing of the decrease in proportions of bilaterally projecting neurons in these areas suggests that numerous neurons retract their callosal axonal branch when the adult pattern of callosal connectivity is established at 9–11 days of age.

Notes: [Auszug] A single visual stimulus activates neurons in many different cortical areas. A major challenge in cortical physiology is to understand how the neural activity in these numerous active zones leads to a unified percept of the visual scene. The anatomical basis for these interactions is the dense ...

Notes: Summary Using the retrograde tracers, fast blue and horseradish peroxidase we have shown the presence of projections from extensive regions of the frontoparietal and temporal cortex to areas 17, 18 and 19 in the newborn kitten. These projections are transitory as they do not exist in the adult cat. The anterograde transport of horseradish peroxidase conjugated with wheat germ agglutinin after injections in frontoparietal and temporal cortex revealed that these transitory projections terminate in the gray matter and that they could therefore play a functional role in the development of the visual cortex.

Notes: Abstract  The results presented in the companion paper showed that extracellular electrical stimulation of the gray matter directly activates axons, but not cell bodies. The second set of experiments presented here was designed to separate the contribution of the axon initial segments and cell bodies from that of the axonal branches to the pool of presynaptic neuronal elements activated by electrical stimulation. For that purpose, N-methyl-d-aspartate (NMDA) iontophoresis was used to induce a selective inactivation of the cell body and of the adjoining portion of the axon by depolarization block, without affecting axonal branches that lack NMDA receptors. After NMDA iontophoresis, the neurons located near the iontophoresis electrode became unable to generate action potentials in an irreversible manner. When the NMDA-induced depolarization block was performed at the site of electrical stimulation, an unexpected increase in the amplitude of the orthodromic responses was observed. Several control experiments suggested that the field potential increase was due to changes of the local environment in the vicinity of the iontophoresis pipette, which led to an increased excitability of the axons. After the period of superexcitability, the orthodromic responses displayed an amplitude that was 15—20% lower than that observed before the NMDA-induced depolarization block, even though cell bodies and axon initial segment at the site of stimulation could not be activated by electrical stimulation. This result shows a low contribution for axon initial segments to the pool of neuronal elements activated by the electrical stimulation. Altogether, these experiments demonstrate that the postsynaptic responses obtained after electrical stimulation of the cortical gray matter result almost exclusively from the activation of axonal branches. Since the neocortex is organised as a network of local and long-range reciprocal connections, great attention must be paid to the interpretation of data obtained with electrical stimualtion.

Notes: Summary Cells with uniform receptive fields were selected for extra cellular recording in the striate cortex of anaesthetised cats. From their responses to electrical stimulation at three sites in the primary visual pathway the cells were grouped according to their ordinal position and whether their afferent drive came from the brisk sustained or brisk transient type of LGN neuron. From differences in laminar distribution and afferent stream the population was divided into 4 subgroups. Within these 4 subgroups there were two basic visual response patterns, which had been identified previously, and attributed to B and C cells. The B cells, which have a smaller receptive field, a lower spontaneous activity and cut-off velocity than C cells, were found to receive their input from slowly conducting afferents while the afferents to C cells arose from the fast stream. A high proportion of both B and C cells received a monosynaptic or direct drive from the optic radiations and responded with multiple spiking to a single electrical shock. Multiple spiking was viewed as evidence of secondary pathways travelling via intermediate cortical neurons to contribute to the cell's input. An examination of the visual properties of all subclasses showed that the more obvious differences in receptive field properties were associated with the type of afferent coming from the LGN rather than with the ordinal or the laminar position of the cell. In this respect the cells in the C/B family resemble S cells, whose receptive field properties also show a dependence on the type of LGN input.

Notes: Summary The spike trains of X and Y retinal ganglion cell axons and neurons in the lateral geniculate nucleus (LGN) of cats were compared to determine if the visual signal could be better discriminated from the maintained activity in the LGN relative to the retina. Curves for relative or receiver operating characteristics (ROC) were derived from the interspike interval data of the spike trains using maintained activity as “noise” and visually-driven activity as the “signal”. Analyses were also made using spike densities and more restricted time intervals. Although it was expected that neurons in the LGN might better distinguish the signal from the noise, the results of the ROC curve and spike density analyses did not bear out this expectation; that is, neither the X or Y cells in the LGN provided better discrimination of the visually-driven activities from the maintained activities compared to the incoming retinal information. Thus, at least in the anesthetized preparation, the LGN does not play a role in increasing the signal-to-noise ratio.