Journal of Integrative Neuroscience

Interactions between color channels (long-wave (L), middle-wave (M) and short-wave (S)) in the receptive field of direction-selective (DS) and orientation-selective (OS) ganglion cells (GCs) were investigated with combined selective stimulation of pairs of cone types (L and M, L and S, M and S). In the experiments with DS GCs of both ON and OFF types, it was shown that: (1) M and S channels were synergistic relative to each other and opponent to L channel. (2) Three-parameter signal (from L, M and S cones) is transformed to one-parameter signal at the output of DS GC, thus illustrating the principle of univariance. (3) In the experiments with OS GCs, it was shown that L and M channels were synergistic in the OFF-pathway, while the S channel was opponent to them. Our results suggested that photoreceptor synaptic connectivity of the bipolar cells hypothetically involved in the goldfish OS circuitry substantially differs from connectivity of bipolar cells presumably targeting DS GC. (4) To sum up, the results obtained on DS GCs confirmed the plausibility of proposed DS GC wiring diagrams; as to the OS circuitry of fish retina it still remains unclear and needs further investigation.

Extracellular recordings were performed from 69 units at different depths between 50 and $400\mu$m below the surface of tectum opticum in goldfish. Using large field stimuli (86${}^{\circ }$ visual angle) of 21 colored HKS-papers we were able to record from 54 color-sensitive units. The colored papers were presented for 5s each. They were arranged in the sequence of the color circle in humans separated by gray of medium brightness. We found 22 units with best responses between orange, red and pink. About 12 of these red-sensitive units were of the opponent “red-ON/blue-green-OFF” type as found in retinal bipolar- and ganglion cells as well. Most of them were also activated or inhibited by black and/or white. Some units responded specifically to red either with activation or inhibition. 18 units were sensitive to blue and/or green, 10 of them to both colors and most of them to black as well. They were inhibited by red, and belonged to the opponent “blue-green-ON/red-OFF” type. Other units responded more selectively either to blue, to green or to purple. Two units were selectively sensitive to yellow. A total of 15 units were sensitive to motion, stimulated by an excentrically rotating black and white random dot pattern. Activity of these units was also large when a red-green random dot pattern of high L-cone contrast was used. Activity dropped to zero when the red-green pattern did not modulate the L-cones. Neither of these motion selective units responded to any color. The results directly show color-blindness of motion vision, and confirm the hypothesis of separate and parallel processing of “color” and “motion”.

Direction selective cells have been found in the retina, the first level of the visual system, in mammals and recently also in the archer fish. These cells are involved in a variety of fast neural computation processes, from the control of eye movements to the detection of prey by the archer fish. The standard model for this mechanism in mammalian retina is well understood and is based on the asymmetry of inhibitory and excitatory inputs to the retinal ganglion cells. However, it remains unclear whether the mechanism that underlies direction selectivity is similar across animal classes. This study reports a pharmacological investigation designed to elucidate the mechanism that underlies motion detection in the archer fish retina. Direction selectivity in the retina was characterized under the influence of specific channel blockers that are known to be present in the different types of neurons of the retina. The results show that the direction-selective mechanism in the archer fish retina is modified only when the inhibitory channels of GABA and Glycine are manipulated. This suggests that the mechanism of direction selectivity in the archer fish retina is fundamentally different from the mechanism of direction selectivity in the mammalian retina.

Electrophysiological responses from 91 direction-selective (DS) units located in the basal optic root (nBOR) and the adjacent dorsomedial area have been recorded extracellularly in the frog (Rana temporaria L.). Based on characteristics of the recorded responses, 23 of the units are considered to be retinal ganglion cells’ (RGC; axon terminals). The majority of the remaining 68 units were considered to be DS neurons of the nBOR. The results of our study suggest that in the nBOR area there might be four subtypes of the tegmental and retinal DS units that respond selectively to stimuli moving in the dorso-ventral, ventro-dorsal, caudo-rostral and rostro-caudal directions. The receptive field (RF) sizes of the nBOR DS neurons were estimated to be about 30–$60^{\circ }$, while those for the retinal units were significantly smaller — just 6–$8^{\circ }$. In response to abrupt darkening within the units’ RFs, both the nBOR DS neurons and the RGC DS units respond by a weak discharge. Our results indicate that the frog nBOR DS neurons integrate the inputs from the retinal OFF-type DS units over relatively large segments of the visual field.

The starburst amacrine cell (SAC) plays a fundamental role in retinal motion perception. In the vertebrate retina, SAC dendrites have been shown to be directionally selective in terms of their Ca${}^{2+}$ responses for stimuli that move centrifugally from the soma. The mechanism by which SACs show Ca${}^{2+}$ bias for centrifugal motion is yet to be determined with precision. Recent morphological studies support a presynaptic delay in glutamate receptor activation induced Ca${}^{2+}$ release from bipolar cells preferentially contacting SACs. However, bipolar cells are known to be electrotonically coupled so time delays between the bipolar cells that provide input to SACs seem unlikely. Using fluorescent microscopy and imunnostaining, we found that the endoplasmic reticulum (ER) is omnipresent in the soma extending to the distal processes of SACs. Consequently, a working hypothesis on heterogeneity of intracellular Ca${}^{2+}$ dynamics from ER is proposed as a possible explanation for the cause of speed tuning of direction-selective Ca${}^{2+}$ responses in dendrites of SACs.

The vertebrate visual system is determined by two main factors, a species’ lifestyle and phylogenetic legacy. Studying the visual system in outgroup lineages may shed some light on the balance of these factors within a certain radiation. We studied the topography of retinal ganglion cells (RGCs) in the retina of the oriental fire-bellied toad Bombina orientalis. These toads belong to the ancient superfamily Discoglossoidea, a sister group to all extant Anura except for two small families. RGCs were retrogradely labeled with tetramethylrhodamine– dextran amine (TMR-DA) and examined in retinal wholemounts. RGCs occurred all over the retina except for the far periphery. Their total number was $214.42\pm 14.49\times 10^3$ ($\mathrm{\text{mean}}\pm \mathrm{\text{SEM}}$, $N=5$). They comprised 73–77% of all cells in the ganglion cell layer. The spatial density of GCs increased gradually from the dorsal and ventral retinal periphery toward the equator to form a weak visual streak and a moderately pronounced area centralis. The minimum density was $5.42\pm 0.15$, and the maximum, $14.92\pm 0.52\mathrm{\text{cells}}∕{\mathrm{\text{mm}}}^2$. The maximum density gradient was $2.76\pm 0.14$. The spatial resolution was minimum in the dorsal and ventral periphery ($3.68\pm 0.05$ and $2.77\pm 0.04$ cycles per degree in water and air, respectively). Intermediate values of spatial resolving power were found within the visual streak ($5.38\pm 0.06$ and $4.05\pm 0.05$ cycles per degree) and reached a peak in area centralis ($6.11\pm 0.11$ and $4.59\pm 0.08$ cycles per degree). This is sufficient for efficient prey location and capture. The relatively high RGC density and the presence of specialized retinal regions in oriental fire-bellied toads are consistent with their highly visual behavior. A brief review comparing the phylogeny and ecology of this with other anuran species suggests that the main factor shaping the RGC distribution in Anura is phylogenetic legacy; the environmental pressure results mainly in adjusting the maximum spatial density of RGCs (and hence the visual acuity) to meet the species’ needs.

The output units of fish retina, i.e., the retinal ganglion cells (detectors), send highly processed information to the primary visual centers of the brain, settled in the midbrain formation tectum opticum (TO). Axons of different fish motion detectors terminate in different tectal levels. In the superficial layer of TO, axons of direction-selective ganglion cells (DS GCs) are terminated. Single unit responses of the DS GCs were recorded in intact fish from their axon terminals in TO. Goldfish DS GCs projecting to TO were shown to comprise six physiological types according to their selectivity to sign of stimulus contrast (ON and OFF units) and their preferred directions: three directions separated by 120${}^{\circ }$. These units, characterized by relatively small receptive fields and remarkable spatial resolution should be classified as local motion detectors. In addition to the retinal DS GCs, other kinds of DS units were extracellularly recorded in the superficial and deep sublaminae of tectum. Some features of their responses suggested that they originated from tectal neurons (TNs). Contrary to DS GCs which are characterized by small RFs and use separate ON and OFF channels, DS TNs have extra-large RFs and ON-OFF type responses. DS TNs were shown to select four preferred directions. Three of them are compatible with those already selected on the retinal level. Complementary to them, the fourth DS TN type with rostro-caudal preference (lacking in the retina) has been revealed. Possible functional interrelations between DS GCs and DS TNs are discussed.