Patch-clamp recordings were used to investigate the properties of the regenerative activity in acutely isolated bipolar cells from the rat retina. Spontaneous, pacemaker-like membrane potential oscillations were observed in all rod bipolar cells and the majority of cone bipolar cells. The waveform of the regenerative potential was stereotypical but distinct among different bipolar cell groups, especially between rod and cone bipolar cells. The spontaneous activity was completely blocked by Co2+, suggesting that Ca2+ influx through voltage-dependent Ca2+ channels was required for initiating such activity. Ca2+-induced Ca2+ release, however, was not found to be involved. The spontaneous activity was also blocked by mibefradil, a T-type Ca2+ channel antagonist. In contrast, application of nimodipine, an L-type Ca2+ current antagonist, affected mainly the waveform of the regenerative potential. This study shows that mammalian retinal bipolar cells in isolation are also capable of generating Ca2+-dependent spontaneous regenerative potential. However, T-type Ca2+ channels appear to be essential for the initiation of the spontaneous activity in mammalian bipolar cells.

A fast-scanning dichroic microspectrophotometer was used to trace products of rhodopsin photolysis (metarhodopsins I/II/III and later) in structurally intact amphibian rod outer segments (ROSs) and metabolically active rods. The instrument allows the recording of absorbance spectra with a time resolution better than 1 s, and to discriminate between products with similar absorbance spectra that differ with respect to the orientation of their chromophore in the photoreceptor membrane. We demonstrate that metarhodopsin III is in a pH-reversible equilibrium with metarhodopsin II and that the metarhodopsin III chromophore is orientated with respect to the membrane plane even more strictly than the 11-cis retinal in “dark” rhodopsin. This indicates that all-trans retinal in metarhodopsin III is still attached to its native binding site on opsin. The kinetic scheme of the decay of metarhodopsins is presented in which metarhodopsin III lies in a shunt pathway from metarhodopsin II to retinal. Formation of metarhodopsin III was detected at bleaches as low as ≈ 3%, contrary to previous reports that it is not formed at below 10% bleaches. Another product that is spectrally similar to metarhodopsin III, termed P440, appears at later stages of photolysis as the result of the decay of metarhodopsin II and metarhodopsin III. The chromophoric group in P440 is orientated preferentially across the disk membrane. The final product(s) in isolated ROS, where the reduction of retinal to retinol is blocked, consists of a mixture of a free retinal and retinal possibly attached to different binding sites in the membrane. In metabolically active rods the later products are quickly converted to retinol. We conclude that metarhodopsin III represents a specific conformational state of metarhodopsin where the chromophoric binding site is still occupied by all-trans retinal. Hence, the formation and decay of metarhodopsin III may be limiting for the rate of rhodopsin regeneration and photoreceptor dark adaptation.

Light-evoked synaptic currents were recorded from on-cone bipolar cells in the mouse retina. Fluctuations in the synaptic current observed during maintained light steps were analyzed in order to estimate the amplitude of the underlying unitary event. The maximal synaptic current variance was 5-fold larger than the maximum expected from fluctuations in the number of active postsynaptic channels. Due to uncertainty in the contribution from channel variance, we calculated a range of values for the unitary event amplitude. The observed variance could be accounted for if 30–39 synaptic sites randomly generated unitary events with a waveform identical to the flash-response, and an amplitude of −3.1 to −2.4 pA. The amplitude is consistent with gating about five mGluR6 channels. The shape of the variance–mean relation suggests that in bright light transmitter release approaches zero, while in darkness transmitter release saturates the postsynaptic response. Thus the on-cone bipolar cell synapse is operating over its entire possible range. If it is assumed that the postsynaptic response saturates when one unitary event occurs per integration time, then a lower bound for the unitary event rate is 18 events/s/synaptic site. If the unitary event is generated by a single synaptic vesicle, the results suggest the total vesicle cycling rate available for encoding the on-cone bipolar cell signal is about 540–700 s−1.

Light adaptation in rod photoreceptors is thought to involve down-regulation of the signaling activity of photoactivated rhodopsin (R*). However, electrophysiological evidence in support of this notion has come largely from studies of truncated, perfused rod outer segments and of rods genetically engineered to perturb known steps in R* deactivation. To test this hypothesis within intact native rods, we examined the effect of a fixed conditioning flash on rods prepared to contain 9-demethyl rhodopsin (9dR) in addition to residual rhodopsin. 9dR, an opsin-based photopigment containing 11-cis 9-demethylretinal as its chromophore, exhibits a blue-shifted excitation spectrum and sluggish deactivation kinetics, properties that distinguish the signaling activities of photoactivated 9dR (9dR*) from those of R*. Saturating photocurrent responses mediated preferentially by R* and 9dR* were obtained with test flash stimulation at 640 and 440 nm, respectively, under dark-adapted conditions (unconditioned response) and at a fixed time after a 640-nm conditioning flash of fixed high intensity. At each test wavelength, the decrease in photocurrent saturation period induced by the conditioning flash was analyzed to determine ψ, the sensitivity of the conditioned response relative that of the unconditioned response; ψ640 /ψ440, the ratio of relative sensitivities, was then obtained. Data obtained from 12 rods yielded ψ640 /ψ440 = 0.60 ± 0.13 (mean ± SD). As common pools of transducin and other downstream components mediate transduction initiated by both R* and 9dR*, the finding that ψ640 < ψ440 provides direct evidence for the down-regulation specifically of R*'s signaling activity during rod light adaptation.

We have evaluated the spatial distribution of low-voltage-activated calcium currents in ganglion cells of the tiger salamander retina. Whole-cell recordings were obtained from ganglion cells in a retinal slice preparation and from acutely dissociated ganglion cells that were identified through retrograde dye injection. In single dissociated cells, we estimated the magnitude (pA) and current density (pA/pF) of LVA currents in ganglion cells, both with and without dendritic processes. Ganglion cells that retained a portion of their dendritic arbor had larger LVA calcium currents and higher LVA current densities than those which lacked processes. When cell capacitance measurements were used to derive the surface area of the soma and dendritic processes, we concluded that a higher LVA current density was present in the dendrites; we estimate that, on average, the current density in the dendrites is approximately five times that of the soma. The presence of a significant density of LVA calcium channels in the dendrites of ganglion cells suggests that they could be involved in a number of cellular functions, including dendritic integration of synaptic currents, impulse generation, and homeostatic functions related to changes in the intradendritic calcium concentration.

This work studies the afferent connectivity to different functionally identified tectal zones in goldfish. The sources of afferents contributed to different degrees to the functionally defined zones. The dorsocentral area of the telencephalon was connected mainly with the ipsilateral anteromedial tectal zone. At diencephalic levels, neurons were found in three different regions: preoptic, thalamic, and pretectal. Preoptic structures (suprachiasmatic and preoptic nuclei) projected mainly to the anteromedial tectal zone, whereas thalamic (ventral and dorsal) and pretectal (central, superficial, and posterior commissure) nuclei projected to all divisions of the tectum. In the mesencephalon, the mesencephalic reticular formation, torus longitudinalis, torus semicircularis, and nucleus isthmi were, in the anteroposterior axis, topographically connected with the tectum. In addition, neurons in the contralateral tectum projected to the injected zones in a symmetrical point-to-point correspondence. At rhombencephalic levels, the superior reticular formation was connected to all studied tectal zones, whereas medial and inferior reticular formations were connected with medial and posterior tectal zones. The present results support a different quantitative afferent connectivity to each tectal zone, possibly based on the sensorimotor transformations that the optic tectum carries out to generate orienting responses.

The nitric oxide (NO)/cyclic guanosine monophosphate (cGMP) signal transduction pathway plays a role in every retinal cell type. Previous studies have shown that excitatory glutamatergic synaptic pathways can increase cGMP-like immunoreactivity (cGMP-LI) in retina through stimulation of NO production, but little is known about the role of synaptic inhibition in the modulation of cGMP-LI. Gamma-amino-n-butyric acid (GABA) plays critical roles in modulating excitatory synaptic pathways in the retina. Therefore, we used GABA receptor antagonists to explore the role of GABAergic inhibitory synaptic pathways on the modulation of the NO/cGMP signal-transduction system. Cyclic GMP immunocytochemistry was used to investigate the effects of the GABA receptor antagonists bicuculline, picrotoxin, and (1,2,5,6-tetrahyropyridin-4-yl) methylphosphinic acid (TPMPA) on levels of cGMP-LI. Cyclic GMP-LI was strongly increased in response to the GABAA receptor antagonist bicuculline, while the GABAC receptor antagonist TPMPA had little effect on cGMP-LI. The GABAA/GABAC receptor antagonist, picrotoxin, caused a moderate increase in cGMP-LI, which was mimicked by the combination of bicuculline and TPMPA. The nitric oxide synthase inhibitor, S-methyl-L-thiocitrulline (SMTC), blocked the increased cGMP-LI in response to stimulation with either bicuculline or picrotoxin. Treatments with either of the glutamate receptor antagonists (5R,10S)-(+)-5-methyl-10,11-dihydro-5H-dibenzo[a,d]cyclohepten-5,10-imine (MK-801) or 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) partially blocked the increases in cGMP-LI seen in response to bicuculline, but a combination of MK-801 and CNQX completely eliminated these increases. These results suggest that inhibitory synaptic pathways involving both types of GABA receptors work through excitatory glutamatergic receptors to regulate the NO/cGMP signal-transduction pathway in retina.

Synaptotagmin I is the leading candidate for the calcium sensor that triggers exocytosis at conventional synapses. However, physiological characterization of the calcium sensor for phasic release at the ribbon-style synapses of the goldfish Mb1 bipolar cell demonstrates a lower than predicted affinity for calcium, suggesting that a modified or different sensor triggers exocytosis at this synapse. We examined synaptotagmin immunolabeling in goldfish retina using two different antibodies directed against synaptotagmin epitopes that specifically labeled the expected 65-kDa protein on western blots of goldfish and mouse retinal membranes. The first antiserum strongly labeled conventional synapses in the inner plexiform layer (IPL), but did not label the ribbon-style synapse-containing synaptic terminals of goldfish Mb1 bipolar cells or photoreceptors. The second antibody also specifically labeled the expected 65-kDa protein on western blots but did not label any synapses in the goldfish retina. A third synaptotagmin antibody that performed poorly on western blots selectively labeled goldfish photoreceptor terminals. These results suggest that synaptotagmin may exist in at least three distinct “forms” in goldfish retinal synapses. These forms, which are differentially localized to conventional synapses, bipolar cell, and photoreceptor terminals, may represent differences in isoform, posttranslational modifications, epitope availability, and protein-binding partners. Labeling with these antibodies in the salamander and mouse retina revealed species-specific differences, indicating that synaptotagmin epitopes can vary across species as well as among synapses.

The peripheral visual field is marked by a deterioration in color sensitivity, sometimes attributed to the random wiring of midget bipolar cells to cone photoreceptors in the peripheral retina (Mullen, 1991; Mullen & Kingdom, 1996). Using psychophysical methods, we explored differences in the sensitivity of peripheral color mechanisms with detection and discrimination of 2-deg spots at 18-deg eccentricity, and find evidence for a postreceptoral locus for the observed loss in sensitivity. As shown before, observers' sensitivity to green was lower than to red in the periphery, although the magnitude of this effect differed across observers. These results suggest that the asymmetry in peripheral sensitivity occurs at a postreceptoral site, possibly a cortical one. In addition, noise masking was used to determine the cone inputs to the peripheral color mechanisms. The masked detection contours indicate that the red and green mechanisms in the periphery respond to the linear difference of approximately equally weighted L- and M-cone contrasts, just as they do in the fovea. Thus, if the midget retinal ganglion system is responsible for red/green color perception in the fovea, it is likely to be responsible at 18-deg eccentricity as well.

In the nob mouse, a mutation in nyctalopin results in a loss of signal transmission from photoreceptors to depolarizing bipolar cells (DBCs). We used immunohistochemical techniques to assess the expression pattern of proteins found at either the photoreceptor terminal or bipolar cell dendrites within the outer plexiform layer. We labeled normal and nob retinas with antibodies against mGluR6, PKC, G0α, bassoon, PSD-95, the α1F subunit of voltage-gated calcium channels, trkB, and dystrophin. All labeling patterns in nob and normal retinas were comparable to those previously reported in mouse retina. Our results indicate that the absence of nyctalopin does not disrupt the expression pattern of other proteins known to be required for synaptic transmission.

Much has been learned from studies of Limulus photoreceptors about the role of the circadian clock and light in the removal of photosensitive membrane. However, little is known in this animal about mechanisms regulating photosensitive membrane renewal, including the synthesis of proteins in, and associated with, the photosensitive membrane. To begin to understand renewal, this study examines diurnal changes in the levels of mRNAs encoding opsin, the integral membrane protein component of visual pigment, and the relative roles of light and the circadian clock in producing these changes. We show that at least two distinct opsin genes encoding very similar proteins are expressed in both the lateral and ventral eyes, and that during the day and night in the lateral eye, the average level of mRNA encoding opsin1 is consistently higher than that encoding opsin2. Northern blot assays showed further that total opsin mRNA in the lateral eyes of animals maintained under natural illumination increases during the afternoon (9 & 12 h after sunrise) in the light and falls at night in the dark. This diurnal change occurs whether or not the eyes receive input from the circadian clock, but it is eliminated in eyes maintained in the dark. Thus, it is regulated by light and darkness, not by the circadian clock, with light stimulating an increase in opsin mRNA levels. The rise in opsin mRNA levels observed under natural illumination was seasonal; it occurred during the summer but not the spring and fall. However, a significant increase in opsin mRNA levels could be achieved in the fall by exposing lateral eyes to 3 h of natural illumination followed by 9 h of artificial light. The diurnal regulation of opsin mRNA levels contrasts sharply with the circadian regulation of visual arrestin mRNA levels (Battelle et al., 2000). Thus, in Limulus, distinctly different mechanisms regulate the levels of mRNA encoding two proteins critical for the photoresponse.

Retinal ganglion cells (RGCs) usually increase their dendritic field area with postnatal retinal growth. The mechanisms that regulate the postnatal shape of dendritic arbors in the growing retina are not well understood. Quantitative studies suffer from the difficulty of labeling specific subpopulations of RGCs selectively including their dendritic processes. In this study, we labeled displaced retinal ganglion cells (DGC) that are known to project to the accessory optic system (AOS) in juvenile and adult chameleons by retrograde transport of dextran amines. The complete population of DGCs was quantitatively screened for the effects of postnatal retinal growth on cell morphology, dendritic field coverage, and dendritic arbor size. The adult eye contained 2000 DGCs/retina. This number was already present at birth. The smaller size of the hatchling eye (approximately 1/3 of the adult size) led to higher densities of DGCs. The greatest accumulation of juvenile DGCs (two-fold higher compared to the adult) was found in the periphery of the retina where the greatest surface expansion was observed. DGC dendritic field areas were adjusted proportionally to this expansion in order to maintain a constant dendritic coverage. The increase of dendritic fields was mediated by two putative passive mechanisms: First, an elongation of individual dendrites similar to previous reports of postnatal RGC development in the retina of goldfish and chicks. Second, and more prominent, we observed that neighboring dendrites were pulled apart from each other. This resulted in a looser spacing of the initially tightly packed dendrites of each dendritic arbor. This dispersal of dendrites over a larger area was, due to its passive nature, proportional to the increase of the retinal surface and preserved a constant dendritic coverage irrespective of the animal's age and eye size.

We quantified and compared the effect of element spacing on contour integration between the achromatic (Ach), red–green (RG), and blue–yellow (BY) mechanisms. The task requires the linking of orientation across space to detect a contour in a stimulus composed of randomly oriented Gabor elements (1.5 cpd, σ = 0.17 deg), measured using a temporal 2AFC method. A contour of ten elements was pasted into a 10 × 10 cells array, and background elements were randomly positioned within the available cells. The effect of element spacing was investigated by varying the mean interelement distance between two and six times the period of the Gabor elements (λ = 0.66 deg) while the total number of elements was fixed. Contour detection was measured as a function of its curvature for jagged contours and for closed contours. At all curvatures, we found that performance for chromatic mechanisms declines more steeply with the increase in element separation than does performance for the achromatic mechanism. Averaged critical element separations were 4.6 ± 0.7, 3.6 ± 0.4, and 2.9 ± 0.2 deg for Ach, BY, and RG mechanisms, respectively. These results suggest that contour integration by the chromatic mechanisms relies more on short-range interactions in comparison to the achromatic mechanism. In a further experiment, we looked at the combined effect of element size and element separation in contour integration for the Ach mechanism.

The expression of glutamate decarboxylase forms, GAD-65 and GAD-67, in GABAergic cells was studied by immunocytochemistry in the retina of the New World monkey, Cebus apella. In the innermost rows of the inner nuclear layer (INL), somata that express GABA correspond to about 45% of the total number of cells in the central retina and about 25% on the periphery. Three subsets of GABAergic amacrine cells were identified along the horizontal meridian: about 5% express only GAD-65 and 40% GAD-67, and approximately 50% contain both forms of GAD. In the INL, GAD-65 immunoreactivity was detected in broad bands around strata 1, 3, and 5. GAD-67 immunoreactivity was observed throughout all strata. Somata that expressed GAD-67 exclusively stratified only in narrow bands around strata 2 and 4 of the INL and colocalized with β2 and β3 subunits of GABA-A receptor. Interplexiform and amacrine cells that express GABA also express tyrosine hydroxylase (TH) or nitric oxide synthase (NOS). GAD-67 colocalized with TH or NOS in presumed amacrine cells of the inner plexiform layer (IPL) and ganglion cell layer (GCL). GAD-65 was expressed in the TH- and NOS-immunoreactive interplexiform and amacrine cells, respectively. Different from what has been described in other mammals, TH and NOS were coexpressed in some neurons, indicating a partial overlap in retinal cell populations containing dopamine or nitric oxide in this primate.

Effects of temperature on the spectral properties of visual pigments were measured in the physiological range (5–28°C) in photoreceptor cells of bullfrog (Rana catesbeiana) and crucian carp (Carassius carassius). Absorbance spectra recorded by microspectrophotometry (MSP) in single cells and sensitivity spectra recorded by electroretinography (ERG) across the isolated retina were combined to yield accurate composite spectra from ca. 400 nm to 800 nm. The four photoreceptor types selected for study allowed three comparisons illuminating the properties of pigments using the dehydroretinal (A2) chromophore: (1) the two members of an A1/A2 pigment pair with the same opsin (porphyropsin vs. rhodopsin in bullfrog “red” rods); (2) two A2 pigments with similar spectra (porphyropsin rods of bullfrog and crucian carp); and (3) two A2 pigments with different spectra (rods vs. long-wavelength-sensitive (L-) cones of crucian carp). Qualitatively, the temperature effects on A2 pigments were similar to those described previously for the A1 pigment of toad “red” rods. Warming caused an increase in relative sensitivities at very long wavelengths but additionally a small shift of λmax toward shorter wavelengths. The former effect was used for estimating the minimum energy required for photoactivation (Ea) of the pigment. Bullfrog rod opsin with A2 chromophore had Ea = 44.2 ± 0.9 kcal/mol, significantly lower (one-tailed P < 0.05) than the value Ea = 46.5 ± 0.8 kcal/mol for the same opsin coupled to A1. The A2 rod pigment of crucian carp had Ea = 42.3 ± 0.6 kcal/mol, which is significantly higher (one-tailed P < 0.01) than that of the L-cones in the same retina (Ea = 38.3 ± 0.4 kcal/mol), whereas the difference compared with the bullfrog A2 rod pigment is not statistically significant (two-tailed P = 0.13). No strict connection between λmax and Ea appears to exist among A2 pigments any more than among A1 pigments. Still, the A1 → A2 chromophore substitution in bullfrog opsin causes three changes correlated as originally hypothesized by Barlow (1957): a red-shift of λmax, a decrease in Ea, and an increase in thermal noise.

Neural signals of the moving visual world are detected by a subclass of retinal ganglion cells that project to the accessory optic system in the vertebrate brainstem. We studied the dendritic morphologies and direction tuning of these brainstem neurons in turtle (Pseudemys scripta elegans) to understand their role in visual processing. Full-field checkerboard patterns were drifted on the contralateral retina while whole-cell recordings were made in the basal optic nucleus in an intact brainstem preparation in vitro. Neurobiotin diffused into the neurons during the recording and was subsequently localized in brain sections. Neuronal morphologies were traced using appropriate computer software to analyze their position in the brainstem. Most labeled neurons were fusiform in shape and had numerous varicosities along their processes. The majority of dendritic trees spread out in a transverse plane perpendicular to the rostrocaudal axis of the nucleus. Neurons near the brainstem surface were often oriented tangential to that surface, whereas more cells at the dorsal side of the nucleus were oriented radial to the brainstem surface. Further analysis of Nissl-stained neurons revealed the largest neurons are located in the rostral and medial portions of the nucleus although neurons are most densely packed in the middle of the nucleus. The preferred directions of the visual responses of the neurons in this sample did not correlate with their morphology and position in the nucleus. Therefore, the morphology of the cells in the turtle accessory optic system appears dependent on its position within the nucleus while its visual responses may depend on the synaptic inputs that contact each cell.

For over 30 years there has been a controversy over whether color-defined motion can be perceived by the human visual system. Some results suggest that there is no chromatic motion mechanism at all, whereas others do find evidence for a purely chromatic motion mechanism. Here we examine the chromatic input to global motion processing for a range of color directions in the photopic luminance range. We measure contrast thresholds for global motion identification and simple detection using sparse random-dot kinematograms. The results show a discrepancy between the two chromatic axes: whereas it is possible for observers to perform the global motion task for stimuli modulated along the red–green axis, we could not assess the contrast threshold required for stimuli modulated along the yellowish-violet axis. The contrast required for detection for both axes, however, are well below the contrasts required for global motion identification. We conclude that there is a significant red–green input to global motion processing providing further evidence for the involvement of the parvocellular pathway. The lack of S-cone input to global motion processing suggests that the koniocellular pathway mediates the detection but not the processing of complex motion for our parameter range.

The ability of the visual system to detect stimuli that vary along dimensions other than luminance or color— “second-order” stimuli—has been of considerable interest in recent years. An important unresolved issue is whether different types of second-order stimuli are detected by a single, all purpose, mechanism, or by mechanisms that are specific to stimulus type. Using a conventional psychophysical paradigm, we show that for a class of second-order stimuli—textures sinusoidally modulated in orientation (OM), spatial frequency (FM), and contrast (CM)—the human visual system employs mechanisms that are selective to stimulus type. Whereas the addition of a subthreshold mask to a test pattern of the same stimulus type was found to facilitate the detection of the test, no facilitation was observed when mask and test were of different types, suggesting mechanism independence for the different types of stimulus. This finding raises the important question of whether mechanism independence is compatible with the well-known filter-rectify-filter (FRF) model of second-order stimulus detection, since FRF mechanisms, in principle, do not discriminate between stimulus types. We show that for all mask/test combinations except those with CM masks, the FRF mechanism giving the largest response to the test modulation is largely unaffected by subthreshold levels of a different stimulus-type mask. For this reason, we cannot rule out the possibility that FRF mechanisms mediate the detection of our stimuli. For combinations involving CM masks, however, we propose that a process of contrast normalization renders the test stimulus insensitive to the mask stimulus.

Inhibitory synaptic transmission via GABA and glycine receptors plays a crucial role in shaping the excitatory response of neurons in the retina. Whole-cell recordings were obtained from ganglion cells in the intact rabbit eyecup preparation to correlate GABA- and glycine-activated currents with the presence of their specific receptors on morphologically identified α ganglion cells. Alpha ganglion cells were chosen based upon their large somata when viewing the retinal surface, and responses to light and dark spots were used to identify OFF-alpha ganglion cells. Light responses were abolished by superfusion of Ringer's containing cobalt to synaptically isolate the cell by blocking all Ca2+-mediated transmitter release. Pressure pulses of GABA and glycine were delivered to an area that encompassed the dendritic field while receptor antagonists were applied through superfusion to characterize the direct inhibition onto the ganglion cell. Physiological results indicated that OFF-α cells did not have any GABAC receptor-activated currents, but did express currents mediated by ionotropic GABAA receptors and metabotropic GABAB receptors that were blocked by their specific antagonists bicuculline and CGP55845, respectively. The amplitudes of strychnine-sensitive glycine-activated currents were always larger than the currents elicited by GABA. Confocal optical sections of physiologically identified, sulforhodamine B-stained cells displayed the localization of glycine and GABAA receptor subunit labeling dispersed over the stained dendrites.

Acetylcholine is one of the primary excitatory neurotransmitters/neuromodulators in the retina, but little is known about the downstream signaling pathways it can activate. The present study immunocytochemically examines the potential sources of acetylcholine and the location of the nicotinic cholinergic receptors in the turtle retina. It also examines how activation of these receptors can influence the nitric oxide (NO)/cyclic guanosine monophosphate (cGMP) signal-transduction pathways. Photoreceptors, amacrine cells, and potentially ganglion cells contain choline acetyltransferase-like immunoreactivity (LI). Nicotinic acetylcholine receptors are immunocytochemically localized on photoreceptors, horizontal, bipolar, and ganglion cells. Nitric oxide imaging indicates that stimulation with nicotine increases NO production primarily in photoreceptors, horizontal, Müller, bipolar, and ganglion cells. In turn, very select populations of amacrine cells respond to this NO with increased levels of cGMP-LI. Selective inhibitors reveal that nitric oxide synthase is involved in most, but not all, of these increases in cGMP-LI. These results show that acetylcholine can activate the NO/cGMP signal-transduction pathways in both the inner and outer retina. This indicates that both of the major excitatory retinal transmitters, glutamate and acetylcholine, can stimulate NO production that increases levels of cGMP-LI in overlapping populations of retinal cells.