Amacrine cells constitute a diverse class of interneurons that contribute to visual signal processing in the inner retina, but surprisingly, little is known about the physiology of most amacrine cell subtypes. Here, we have taken advantage of the sparse expression of vesicular glutamate transporter 3 (VGLUT3) in the mammalian retina to target the expression of yellow fluorescent protein (YFP) to a unique population of amacrine cells using a new transgenic mouse line. Electrophysiological recordings made from YFP-positive (VGLUT3+) amacrine cells provide the first functional data regarding the active membrane properties and synaptic connections of this recently identified cell type. We found that VGLUT3+ amacrine cells receive direct synaptic input from bipolar cells via both N-methyl-d-aspartate receptors (NMDARs) and non-NMDARs. Voltage-gated sodium channels amplified these excitatory inputs but repetitive spiking was never observed. VGLUT3+ amacrine cells responded transiently to both light increments (ON response) and decrements (OFF response); ON responses consisted exclusively of inhibitory inputs, while OFF responses comprised both excitatory and inhibitory components, although the inhibitory conductance was larger in amplitude and longer in time course. The physiological properties and anatomical features of the VGLUT3+ amacrine cells suggest that this bistratified interneuron may play a role in disinhibitory signaling and/or crossover inhibition between parallel pathways in the retina.

The endocannabinoid (ECB) system has been found throughout the central nervous system and modulates cell excitability in various forms of short-term plasticity. ECBs and their receptors have also been localized to all retinal cells, and cannabinoid receptor activation has been shown to alter voltage-dependent conductances in several different retinal cell types, suggesting a possible role for cannabinoids in retinal processing. Their effects on synaptic transmission in the mammalian retina, however, have not been previously investigated. Here, we show that exogenous cannabinoids alter spontaneous synaptic transmission onto retinal ganglion cells (RGCs). Using whole-cell voltage-clamp recordings in whole-mount retinas, we measured spontaneous postsynaptic currents (SPSCs) in RGCs in adult and young (P14–P21) mice. We found that the addition of an exogenous cannabinoid agonist, WIN55212-2 (5 μM), caused a significant reversible reduction in the frequency of SPSCs. This change, however, did not alter the kinetics of the SPSCs, indicating a presynaptic locus of action. Using blockers to isolate inhibitory or excitatory currents, we found that cannabinoids significantly reduced the release probability of both GABA and glutamate, respectively. While the addition of cannabinoids reduced the frequency of both GABAergic and glutamatergic SPSCs in both young and adult mice, we found that the largest effect was on GABA-mediated currents in young mice. These results suggest that the ECB system may potentially be involved in the modulation of signal transmission in the retina. Furthermore, they suggest that it might play a role in the developmental maturation of synaptic circuits, and that exogenous cannabinoids are likely able to disrupt retinal processing and consequently alter vision.

The rat is a popular animal model for vision research, yet there is little quantitative information about the physiological properties of the cells that provide its brain with visual input, the retinal ganglion cells. It is not clear whether rats even possess the full complement of ganglion cell types found in other mammals. Since such information is important for evaluating rodent models of visual disease and elucidating the function of homologous and heterologous cells in different animals, we recorded from rat ganglion cells in vivo and systematically measured their spatial receptive field (RF) properties using spot, annulus, and grating patterns. Most of the recorded cells bore likeness to cat X and Y cells, exhibiting brisk responses, center-surround RFs, and linear or nonlinear spatial summation. The others resembled various types of mammalian W cell, including local-edge-detector cells, suppressed-by-contrast cells, and an unusual type with an ON–OFF surround. They generally exhibited sluggish responses, larger RFs, and lower responsiveness. The peak responsivity of brisk-nonlinear (Y-type) cells was around twice that of brisk-linear (X-type) cells and several fold that of sluggish cells. The RF size of brisk-linear and brisk-nonlinear cells was indistinguishable, with average center and surround diameters of 5.6 ± 1.3 and 26.4 ± 11.3 deg, respectively. In contrast, the center diameter of recorded sluggish cells averaged 12.8 ± 7.9 deg. The homogeneous RF size of rat brisk cells is unlike that of cat X and Y cells, and its implication regarding the putative roles of these two ganglion cell types in visual signaling is discussed.

Motion-sensitive neurons in the visual systems of many species, including humans, exhibit a depression of motion responses immediately after being exposed to rapidly moving images. This motion adaptation has been extensively studied in flies, but a neuronal mechanism that explains the most prominent component of adaptation, which occurs regardless of the direction of motion of the visual stimulus, has yet to be proposed. We identify a neuronal mechanism, namely frequency-dependent synaptic depression, which explains a number of the features of adaptation in mammalian motion-sensitive neurons and use it to model fly motion adaptation. While synaptic depression has been studied mainly in spiking cells, we use the same principles to develop a simple model for depression in a graded synapse. By incorporating this synaptic model into a neuronally based model for elementary motion detection, along with the implementation of a center-surround spatial band-pass filtering stage that mimics the interactions among a subset of visual neurons, we show that we can predict with remarkable success most of the qualitative features of adaptation observed in electrophysiological experiments. Our results support the idea that diverse species share common computational principles for processing visual motion and suggest that such principles could be neuronally implemented in very similar ways.

The cingulate cortex (CG) and the adjacent region designated as the splenial visual area (SVA) project to areas of the extrageniculate thalamic system that are concerned with processing visual information. En route to the thalamus, they pass through the thalamic reticular nucleus (TRN), an important source of thalamic inhibition. We wished to determine whether SVA axon collaterals projected to the previously defined visual sector of the TRN or a separate projection zone and did this differ from the projection zone of CG. We iontophoretically injected different neuroanatomical tracers into several locations within CG/SVA and traced the labeled axons through the TRN. The CG and SVA have a projection zone that only partially overlaps the dorsorostral regions of the visuocortical projection zone; there was no evidence to suggest separate SVA and CG zones or tiers of label within the TRN. The projection formed only a weak topographic map in the TRN, which is largely defined in the rostrocaudal axis and is similar to that of the area 7 projection; both projections have a high degree of overlap in the dorsal TRN. We postulate that CG/SVA may be involved in the initiation of orientation behaviors via stimulation of thalamic nuclei and attentional mechanisms of the TRN.

Visual masking can result from the interference of perceptual signals. According to the principle of functional specialization, interference should be greatest when signal and mask belong to the same visual attribute (e.g., color or motion) and least when they belong to different ones. We provide evidence to support this view and show that the time course of masking is visual attribute specific. First, we show that a color target is masked most effectively by color (homogeneous target-mask pair) and least effectively by motion (heterogeneous pair) and vice versa for a motion target. Second, we show that the time at which the mask is most effective depends strongly on the target-mask pairing. Heterogeneous masking is strongest when the mask is presented before the target (forward masking) but this is not true of homogeneous masking. This finding supports a delayed cross-feature interaction due to segregated processing sites. Third, lengthening the stimulus onset asynchrony between target and mask leads to a faster improvement in color than in motion detectability, lending support for a faster color processing system and consistent with reports of perceptual asynchrony in vision. In summary, we present three lines of psychophysical evidence, all of which support a segregated neural coding scheme for color and motion in the human brain.