Mediation of visual responses in the nucleus of the optic tract in cats and rats by excitatory amino acid receptors

Question of Reference Frames: Visual Direction-Selective Neurons in the Accessory Optic System of Goldfish

We investigated if visual direction-selective neurons in the pretectal area (APT) of goldfish ( Carassius auratus auratus) preferred visual stimuli resulting from rotations around axes corresponding to the best responsive axes of the semicircular canals [optic flow that is consistent to a maximal activation of the horizontal canal pair (yaw), to a maximal activation of the right anterior/left posterior semicircular canal pair (RALP), and to a maximal activation of the left anterior/right posterior semicircular canal pair (LARP)]. Our sample of neurons recorded in the left pretectum had two preferred axes of rotation: first, rotation around the yaw axis and second, rotation around the RALP axis. Both axes of rotation correspond to best responsive axes of the semicircular canals. For this reason, coding in a reference frame defined by the vestibular system or the pulling direction of the eye muscles is suggested. In our population of recorded APT neurons, we did not find segregation of different preferred axes of rotation into different anatomical structures. Furthermore in all axes no bias for clockwise or counterclockwise rotations was obvious. This is particularly noteworthy for the yaw axis because preference for temporo-nasal and naso-temporal rotations was found at the same recording side. Hence we conclude that in fish the accessory optic system may consist of one nucleus on each side of the midbrain only, the APT. Segregation into different nuclei coding for different axes and different senses of rotation probably first developed in amphibians.

The commissural transfer of the horizontal optokinetic signal in the rat: a c-Fos study

We applied the Fos method in rats subjected to horizontal optokinetic stimulation (OKS) to study whether optokinetic information is transferred through the commissural pretectal fibres from one optic tract nucleus (NOT) to the opposite. In binocular as well as in monocular nasalward OKS, the highest Fos immunoreactivity was found in the NOT contralateral to the nasalward stimulation, as expression of the activation either of direction-selective cells and of commissural neurons. Even the opposite NOT showed many Fos-positive cells activated by the opposite nucleus throughout the commissural pretectal pathway. They might be the GABA positive cells, which are thought to allow the activation in one nucleus to be transformed into inhibition of the opposite side. In monocular temporalward OKS, the inhibition on direction-selective cells and the consequent silencing of commissural neurons caused the faint immunoreactivity in the NOT contralateral to eye stimulated. In the opposite nucleus the few Fos-positive cells emerged as a consequence of the lack of the normal tonic commissurally mediated inhibition.

Synaptic circuitry in the retinorecipient layers of the optic tectum of the lamprey (Lampetra fluviatilis). A combined hodological, GABA and glutamate immunocytochemical study

The ultrastructure of the retinorecipient layers of the lamprey optic tectum was analysed using tract tracing techniques combined with GABA and glutamate immunocytochemistry. Two types of neurons were identified; a population of large GABA-immunonegative cells, and a population of smaller, highly GABA-immunoreactive interneurons, some of whose dendrites contain synaptic vesicles (DCSV). Five types of axon terminals were identified and divided into two major categories. The first of these are GABA-immunonegative, highly glutamate-immunoreactive, contain round synaptic vesicles, make asymmetrical synaptic contacts, and can in turn be divided into AT1 and AT2 terminals. The AT1 terminals are those of the retinotectal projection. The origin of the nonretinal AT2 terminals could not be determined. AT1 and AT2 terminals establish synaptic contacts with DCSV, with dendrites of the retinopetal neurons (DRN), and with conventional dendritic (D) profiles. The terminals of the second category are GABA-immunoreactive and can similarly be divided into AT3 and AT4 terminals. The AT3 terminals contain pleiomorphic synaptic vesicles and make symmetrical synaptic contacts for the most part with glutamate-immunoreactive D profiles. The AT4 terminals contain rounded synaptic vesicles and make asymmetrical synaptic contacts with DRN, with DCSV, and with D profiles. A fifth, rarely observed category of terminals (AT5) contain both clear synaptic vesicles and a large number of dense-core vesicles. Synaptic triads involving AT1, AT2 or AT4 terminals are rare. Our findings are compared to these of previous studies of the fine structure and immunochemical properties of the retinorecipient layers of the optic tectum or superior colliculus of Gnathostomes.

The Effects of Selective Serotonin-Reuptake Inhibitor on Visual Evoked Potential in Rats

The present study was undertaken to clarify the effects of selective serotonin-reuptake inhibitors (SSRIs) on visual evoked potential (VEP) in rats. To elucidate the mechanism of action of SSRIs, some serotonin (5-HT) agonists were used. SSRIs, fluvoxamine and paroxetine, caused a reduction in the amplitudes of P1-N1, P3-N3, and N3-P4 components of VEP. The amplitude of the P1-N1 component was also reduced by the 5-HT1A agonist 8-OH-DPAT and 5-HT1B agonist anpirtoline. On the other hand, amplitudes of P3-N3 and N3-P4 components were reduced by anpirtoline and the 5-HT2 agonist DOI. These results indicate that the reduction in the amplitude of the P1-N1 component of VEP induced by SSRIs may participate in 5-HT1A and 5-HT1B receptors, and those of P3-N3 and N3-P4 components induced by SSRIs may be closely related with 5-HT1B and 5-HT2 receptors.

The synaptic pharmacology underlying sensory processing in the superior colliculus

The superior colliculus (SC) is one of the most ancient regions of the vertebrate central sensory system. In this hub afferents from several sensory pathways converge, and an extensive range of neural circuits enable primary sensory processing, multi-sensory integration and the generation of motor commands for orientation behaviours. The SC has a laminar structure and is usually considered in two parts; the superficial visual layers and the deep multi-modal/motor layers. Neurones in the superficial layers integrate visual information from the retina, cortex and other sources, while the deep layers draw together data from many cortical and sub-cortical sensory areas, including the superficial layers, to generate motor commands. Functional studies in anaesthetized subjects and in slice preparations have used pharmacological tools to probe some of the SC's interacting circuits. The studies reviewed here reveal important roles for ionotropic glutamate receptors in the mediation of sensory inputs to the SC and in transmission between the superficial and deep layers. N-methyl-D-aspartate receptors appear to have special responsibility for the temporal matching of retinal and cortical activity in the superficial layers and for the integration of multiple sensory data-streams in the deep layers. Sensory responses are shaped by intrinsic inhibitory mechanisms mediated by GABA(A) and GABA(B) receptors and influenced by nicotinic acetylcholine receptors. These sensory and motor-command activities of SC neurones are modulated by levels of arousal through extrinsic connections containing GABA, serotonin and other transmitters. It is possible to naturally stimulate many of the SC's sensory and non-sensory inputs either independently or simultaneously and this brain area is an ideal location in which to study: (a) interactions between inputs from the same sensory system; (b) the integration of inputs from several sensory systems; and (c) the influence of non-sensory systems on sensory processing.

Enrichment of glutamate-like immunoreactivity in the retinotectal terminals of the viper Vipera aspis: an electron microscope quantitative immunogold study

A post-embedding immunogold study was carried out to estimate the immunoreactivity to glutamate in retinal terminals, P axon terminals and dendrites containing synaptic vesicles in the superficial layers of the optic tectum of Vipera. Retinal terminals, identified following either intraocular injection of tritiated proline, horseradish peroxidase (HRP) or short-term survivals after retinal ablation, were observed to be highly glutamate-immunoreactive. A detailed quantitative analysis showed that about 50% of glutamate immunoreactivity was localized over the synaptic vesicles, 35.8% over mitochondria and 14.2% over the axoplasmic matrix. The close association of immunoreactivity with the synaptic vesicles could indicate that Vipera retino-tectal terminals may use glutamate as their neurotransmitter. P axon terminals and dendrites containing synaptic vesicles, strongly gamma-aminobutyric (GABA)-immunoreactive, were shown to be also moderately glutamate-immunoreactive, but two to three times less than retinal terminals. Moreover, in P axon terminals, the glutamate immunoreactivity was denser over mitochondria than over synaptic vesicles, possibly reflecting the 'metabolic' pool of glutamate, which serves as a precursor in the formation of GABA.

The contribution of GABA-mediated inhibition to response properties of neurons in the nucleus of the optic tract in the rat

The contribution of GABA-mediated inhibition to the generation of directional selectivity of neurons in the nucleus of the optic tract (NOT) and the dorsal terminal nucleus of the accessory optic system (DTN) was examined in anaesthetized rats by iontophoretic application of the GABAA receptor antagonist bicuculline methiodide. Spontaneous and visually evoked NOT-DTN cell activities were always increased by bicuculline application. The directional selectivity of NOT-DTN cells to slowly moving whole-field stimuli, expressed as the direction index, was reduced for most neurons. However, the difference between firing rates during stimulus movements in the preferred and in the non-preferred direction did not change systematically. On average, this difference was not significantly affected in the majority of the neurons, although bicuculline more strongly increased the activity during movement in the preferred or non-preferred direction in some of the neurons. These results indicate that directionally selective neurons in the rat NOT-DTN receive GABAergic inhibition which is most likely tonic and independent of the stimulus direction.

Excitatory amino acid receptors participate in synaptic transmission of visual responses in the superficial layers of the cat superior colliculus

The contribution of NMDA and non-NMDA receptors to visual synaptic transmission in the superficial layers of the superior colliculus of the cat has been studied using extracellular recording and iontophoretic techniques. Neuronal responses to natural visual stimulation and the ejection of NMDA and AMPA were observed in the absence and presence of the antagonists CNQX, CPP and AP5. CNQX routinely reduced the responses to visual stimulation at ejection currents which selectively blocked the responses to AMPA but not those to NMDA. Agonist selective ejection currents of CPP and AP5 also reduced visual responses of most SC neurons, but there was a substantial majority whose visual responses were resistant to these antagonists. Neurons with CPP/AP5 resistant visual responses were more commonly found 750-1000 microns from the dorsal surface of the SC. The data indicate that, while non-NMDA receptors are heavily involved in visual synaptic transmission in the superficial SC, the involvement of NMDA receptors varies with recording depth.

Segregation of direction selective neurons and synaptic organization of inhibitory intranuclear connections in the medial terminal nucleus of the rat: an electrophysiological and immunoelectron microscopical study

A combined electrophysiological and morphological investigation of the medial terminal nucleus (MTN) in the rat was undertaken, aimed at a better understanding of the relationship between structure and function in this nucleus. The locations of upward and downward direction selective units in the MTN were documented with extracellular electrophysiological recording. By means of tracer experiments, with Phaseolus vulgaris-leucoagglutinin, biocytin, and cholera toxin subunit B-horseradish peroxidase, the internal connections of the MTN, its retinal afferents, and the projection neurons to the inferior olive were visualized. Terminals originating from the retina and from internal connections were characterized at the ultrastructural level. Their termination pattern on cells in the MTN, including identified inferior olive projection neurons, were determined. Additionally, postembedding GABA immunocytochemistry was performed to identify GABAergic elements. From reconstructions of the positions of electrophysiologically recorded units in the MTN, a local segregation between upward and downward direction selective units was revealed. Upward direction selective units were found in the dorsal part and ventromedially, whereas downward direction selective units were found ventral and laterally in the MTN. The MTN receives optic fibers via two separate routes which, based on their trajectory, presumably terminate in different parts of the MTN: the inferior fascicle of the accessory optic tract in the dorsal part, and the posterior fiber bundle of the superior fascicle in the ventral part of the MTN. A correspondence has been found between the segregation of direction selective units and the areas in the MTN where the retinal fibers from the two pathways distribute. It is, therefore, proposed that the inferior fasciculus conveys upward direction selectivity and the posterior fiber bundle downward direction selectivity, and that the two fiber bundles terminate segregated in the MTN. After anterograde tracing from the eye, retinal terminals were found evenly distributed throughout the MTN. They are characterized as GABA negative R-type terminals. After retrograde tracing from the inferior olive, identified MTN-inferior olive projection neurons were found in the dorsal MTN and medially in the ventral MTN. Their location in the MTN suggests that MTN-inferior olive projection neurons are upward direction selective. MTN-inferior olive projection neurons are large non-GABAergic cells, with a variable form. A majority of both F- and R-type terminals were found to make synaptic contacts on the dendrites of MTN cells. MTN-inferior olive projection neurons did not differ from other neurons in this respect.

OKN-related neurons in the rat nucleus of the optic tract and dorsal terminal nucleus of the accessory optic system receive a direct cortical input

It has been previously assumed that the asymmetry of the monocular optokinetic nystagmus (OKN) of lateral-eyed mammals is caused by an absence of visual cortex projections to directional selective neurons in the pretectal nucleus of the optic tract and dorsal terminal nucleus of the accessory optic system (NOT-DTN). In contrast to this generally accepted hypothesis, we present multiple evidence that OKN-related neurons in the rat NOT-DTN in fact do receive input from the visual cortex. We studied the corticofugal projection to NOT-DTN physiologically, with extracellular single unit recording and electrical stimulation of the optic chiasma and the visual cortex, and anatomically, using retrograde and anterograde tracing techniques. In particular we focussed our attention on the NOT-DTN neurons, which control eye movements during OKN. All OKN-related NOT-DTN cells were activated after optic chiasma stimulation. Forty-five percent of these neurons were also activated after stimulation of the visual cortex (VC). The majority of neurons activated from VC (80%) also responded to monocular stimulation of either eye. On the contrary, most of the neurons that responded to stimulation of the contralateral eye only were not activated from VC. After injection of fluorescent latex microspheres into the NOT-DTN, retrogradely labeled neurons were found in areas 17, 18, and 18A of the visual cortex. Phaseolus vulgaris leucoagglutinin injected into the visual cortex anterogradely labeled fibres and terminals throughout the NOT-DTN complex. Labeled boutons were found in close proximity to OKN-related NOT-DTN cells, selectively stained after horseradish peroxidase (HRP) injections into the inferior olive. Our results demonstrate that NOT-DTN cells in the rat, which are involved in the generation of horizontal OKN, receive a direct input from the ipsilateral visual cortex.