SUMMARY Although rod and cone photoreceptor cells in the vertebrate retina are anatomically connected or coupled by gap junctions, rod-cone coupling is thought to be weak. By using a combination of tracer labeling and electrical recording in the goldfish retina and tracer labeling in the mouse retina, we show that the retinal circadian clock, a local endogenous process, and not the retinal response to the visual environment, controls the extent and strength of rod-cone coupling by activating dopamine D2-like receptors in the day, so that rod-cone coupling is weak during the day but remarkably robust at night. The results demonstrate that circadian control of rod-cone coupling serves as a synaptic switch between the rod and cone pathways, so that cones and neurons postsynaptic to cones receive very dim light signals from rods at night, but not in the day. The increase in the strength and extent of rod-cone coupling at night may enhance the reliability of the rod light response and facilitate the detection of large dim objects.

Lateral interactions in the outer retina, particularly negative feedback from horizontal cells to cones and direct feed-forward input from horizontal cells to bipolar cells, play a number of important roles in early visual processing, such as generating center-surround receptive fields that enhance spatial discrimination. These circuits may also contribute to post-receptoral light adaptation and the generation of color opponency. In this review, we examine the contributions of horizontal cell feedback and feed-forward pathways to early visual processing. We begin by reviewing the properties of bipolar cell receptive fields, especially with respect to modulation of the bipolar receptive field surround by the ambient light level and to the contribution of horizontal cells to the surround. We then review evidence for and against three proposed mechanisms for negative feedback from horizontal cells to cones: 1) GABA release by horizontal cells, 2) ephaptic modulation of the cone pedicle membrane potential generated by currents flowing through hemigap junctions in horizontal cell dendrites, and 3) modulation of cone calcium currents (ICa) by changes in synaptic cleft proton levels. We also consider evidence for the presence of direct horizontal cell feed-forward input to bipolar cells and discuss a possible role for GABA at this synapse. We summarize proposed functions of horizontal cell feedback and feed-forward pathways. Finally, we examine the mechanisms and functions of two other forms of lateral interaction in the outer retina: negative feedback from horizontal cells to rods and positive feedback from horizontal cells to cones.

In the vertebrate retina, the light responses of post-receptor neurons depend on the ambient or background illumination. Using intracellular recording, we have found that a circadian clock regulates the light responses of dark-adapted fish cone horizontal cells. Goldfish were maintained on a 12-hr light/12-hr dark cycle. At different times of the day or night, retinas were superfused in darkness for 90 min ("prolonged darkness"), following which horizontal cells were impaled without the aid of any light flashes. The vertebrate retina is able to respond to visual images in starlight, in the midday sun, and at all times in between, during which the ambient or background illumination changes by 6 to 12 orders of magnitude (1, 2). This ability derives from the presence of two kinds of photoreceptor cells, rods and cones, which subserve nighttime and daytime vision, respectively, and from at least two kinds of adaptive mechanisms. The first kind of adaptive mechanism, photoreceptor adaptation, is determined by the relative degree of visual pigment bleached or regenerated in photoreceptors (3) and by the regulatory role of Ca2+ in the phototransduction process (4). The second adaptive mechanism, network or neural adaptation, is determined by post-receptor cellular and synaptic mechanisms of retinal networks or circuits (3).Network adaptation likely underlies various reported changes in the light responses of post-receptor neurons that depend on the ambient illumination, including changes in the center-surround receptive field organization of ganglion cells (5, 6) and in the light responsiveness of horizontal cells, a type of second order cell (7). In nonmammalian vertebrates dark adaptation increases rod, and decreases cone, input to horizontal cells and light adaptation has the opposite effects (8, 9).A circadian clock is a type of biological oscillator that has persistent rhythmicity with a period of approximately 24 hr in the absence of external timing cues (e.g., constant darkness). In addition, a circadian clock can be entrained by cyclic environmental stimuli, such as light (10). In vertebrate retinas, a variety of cellular phenomena are regulated by circadian rhythms, including melatonin production and release (11, 12), tyrosine hydroxylase activity (13) Intact, isolated retinas were superfused at 0.5 ml/min with a Ringer's solution that contained 130 mM NaCl, 2.5 mM KCl, 20 mM NaHCO3, 0.7 mM CaCl2, 1.0 mM MgCl2, and 20 mM glucose, as described (22,23). Oxygenation of the superfusate with a mixture of 95% 02/5% CO2 maintained the superfusate at a pH of 7.4 in the retinal chamber. After surgery, the retinas were superfused in darkness for 90 min ("prolonged darkness"), following which a horizontal cell was impaled without Abbreviation: ZT, Zeitgeber.

A circadian (24-hour) clock regulates the light responses of fish cone horizontal cells, second order neurones in the retina that receive synaptic contact from cones and not from rods. Due to the action of the clock, cone horizontal cells are driven by cones in the day, but primarily driven by rods at night. We show here that dopamine, a retinal neurotransmitter, acts as a clock signal for the day by increasing cone input and decreasing rod input to cone horizontal cells. The amount of endogenous dopamine released from in vitro retinae was greater during the subjective day than the subjective night. Application of dopamine or quinpirole, a dopamine D 2 -like agonist, during the subjective night increased cone input and eliminated rod input to the cells, a state usually observed during the subjective day. In contrast, application of spiperone, a D 2 -like antagonist, or forskolin, an activator of adenylyl cyclase, during the subjective day reduced cone input and increased rod input. SCH23390, a D 1 antagonist, had no effect. Application of R p -cAMPS, an inhibitor of cAMPdependent protein kinase, or octanol, an alcohol that uncouples gap junctions, during the night increased cone input and decreased rod input. Because D 2 -like receptors are on photoreceptor cells, but not horizontal cells, the results suggest that the clock-induced increase in dopamine release during the day activates D 2 -like receptors on photoreceptor cells. The resultant decrease in intracellular cyclic AMP and protein kinase A activation then mediates the increase in cone input and decrease in rod input.

The mechanisms in the retina that generate light responses selective for the direction of image motion remain unresolved. Recent evidence indicates that directionally selective light responses occur first in the retina in the dendrites of an interneuron, i.e., the starburst amacrine cell, and that these responses are highly sensitive to the activity of Na-K-2Cl (NKCC) and K-Cl (KCC), two types of chloride cotransporter that determine whether the neurotransmitter GABA depolarizes or hyperpolarizes neurons, respectively. We show here that selective blockade of the NKCC2 and KCC2 cotransporters located on starburst dendrites consistently hyperpolarized and depolarized the starburst cells, respectively, and greatly reduced or eliminated their directionally selective light responses. By mapping NKCC2 and KCC2 antibody staining on these dendrites, we further show that NKCC2 and KCC2 are preferentially located in the proximal and distal dendritic compartments, respectively. Finally, measurements of the GABA reversal potential in different starburst dendritic compartments indicate that the GABA reversal potential at the distal dendrite is more hyperpolarized than at the proximal dendrite due to KCC2 activity. These results thus demonstrate that the differential distribution of NKCC2 on the proximal dendrites and KCC2 on the distal dendrites of starburst cells results in a GABA-evoked depolarization and hyperpolarization at the NKCC2 and KCC2 compartments, respectively, and underlies the directionally selective light responses of the dendrites. The functional compartmentalization of interneuron dendrites may be an important means by which the nervous system encodes complex information at the subcellular level.direction-selective ͉ GABAergic excitation ͉ interneuron

No abstract

1. The influence of horizontal cells on ganglion cells, the output neuron of the retina, was examined in an in vitro rabbit eyecup preparation. The extracellular spike activity of ganglion cells was monitored while pulsatile DC or sinusoidally modulated current was injected intracellularly into nearby horizontal cells. Interactions between the effects of light stimulation and horizontal cell current injections on ganglion cell responses were also examined. 2. Horizontal cells were found to contribute to the receptive field surround of ganglion cells. In particular, horizontal cells contributed to surround excitability and to surround antagonism of the centre light response. 3. Brisk, sluggish and direction‐selective ganglion cells were all affected by current injections into horizontal cells. However, brisk ganglion cells responded to lower amplitude currents than did sluggish or direction‐selective cells. 4. Horizontal cells with receptive fields that overlap those of ganglion cells were able to affect ganglion cell discharge. Moreover, the closer a horizontal cell was to the receptive field centre of a ganglion cell, the more effective were current injections in modulating ganglion cell discharge rate. The length constant of the horizontal cell contribution to the ganglion cell receptive field was approximately 200 microns. These results indicate that horizontal cells which are located within or outside of a ganglion cell's receptive field centre can influence that ganglion cell's activity. 5. The influence of horizontal cells on ganglion cell discharges was relatively weak at low temporal frequencies of sinusoidally modulated current. 6. Application of 2‐amino‐4‐phosphonobutyrate (APB), a glutamate analogue, blocked the modulation of spike activity of on‐centre ganglion cells that was induced by sinusoidally modulated current injected into nearby horizontal cells. The spike activity of off‐centre ganglion cells was not blocked. 7. These findings suggest that horizontal cells contribute to the surround of ganglion cells and bipolar cells primarily through a feedback pathway onto cone photoreceptor cells.

No abstract