There is much evidence for an endocannabinoid system in the retina. However, neither the distribution of endocannabinoid uptake, the regulation of endocannabinoid levels, nor the role of endocannabinoid metabolism have been investigated in the retina. Here we focused on one endocannabinoid, anandamide (AEA), and its major hydrolyzing enzyme, fatty acid amide hydrolase (FAAH), in the goldfish retina. Immunoblots of FAAH immunoreactivity (IR) in goldfish retina, brain and rat retina, and brain homogenates showed a single band at 61 kDa that was blocked by preadsorption with peptide antigen. Specific FAAH IR (blocked by preadsorption) was most prominent over Müller cells and cone inner segments. Weaker label was observed over some amacrine cells, rare cell bodies in the ganglion cell layer, and in four lamina in the inner plexiform layer. FAAH activity assays showed that goldfish-retinal and brain homogenates hydrolyzed AEA at rates comparable to rat brain homogenate, and the hydrolysis was inhibited by methyl arachidonyl fluorophosphonate (MAFP) and N-(4 hydroxyphenyl)-arachidonamide (AM404), with IC50s of 21 nM and 1.5 μM, respectively. Cellular 3H-AEA uptake in the intact retina was determined by in vitro autoradiography. Silver-grain accumulation at 20°C was most prominent over cone photoreceptors and Müller cells. Uptake was significantly reduced when retinas were incubated at 4°C, or preincubated with 100 nM MAFP or 10 μM AM404. There was no differential effect of blocking conditions on the distribution of silver grains over cones or Müller cells. The codistribution of FAAH IR and 3H-AEA uptake in cones and Müller cells suggests that the bulk clearance of AEA in the retina occurs as a consequence of a concentration gradient created by FAAH activity. We conclude that endocannabinoids are present in the goldfish retina and underlay the electrophysiological effects of cannabinoid ligands previously shown on goldfish cones and bipolar cells.

Myopia is increasing in prevalence world-wide, nearing epidemic proportions in some populations. This has led to expanded research efforts to understand how ocular growth and refractive errors are regulated. Eye growth is sensitive to visual experience, and is altered by both form deprivation and optical defocus. In these cases, the primary targets of growth regulation are the choroidal and scleral layers of the eye that demarcate the boundary of the posterior vitreous chamber. Of significance to this review are observations of local growth modulation that imply that the neural retina itself must be the source of growth-regulating signals. Thus the retinal pigment epithelium (RPE), interposed between the retina and the choroid, is likely to play a critical role in relaying retinal growth signals to the choroid and sclera. This review describes the ion transporters and signal receptors found in the chick RPE and their possible roles in visually driven changes in eye growth. We focus on the effects of four signaling molecules, otherwise implicated in eye growth changes (dopamine, acetylcholine, vasoactive intestinal peptide (VIP), and glucagon), on RPE physiology, including fluid transport. A model for RPE-mediated growth regulation is proposed.

In this special issue of Visual Neuroscience, we present a series of papers to honor the life and career of Robert William Rodieck, who passed away at his home in Seattle on September 30, 2003. Rodieck held the E.K. Bishop Professorship in Ophthalmology at the University of Washington Medical Center from 1978–1997. Known to everyone as “Bob,” he leaves behind an intellectual legacy often admired by his colleagues and friends for its scope, intensity, and empathy for what was beautiful in the object of his studies.

One clear, crisp November day, Grant returned to the lab after attending a high school sports event with his family. An hour later an arrhythmia stopped his heart. Those who knew Grant knew of his devotion to his family, his science, and soccer. Few of us knew how many lives he touched until over 1200 people gathered in the early afternoon of November 29, 2004 to remember their husband, father, brother, nephew, friend, colleague, coach, and mentor.

The macaque visual cortex is exquisitely organized into columns, modules, and streams, much of which can be correlated with its metabolic organization revealed by cytochrome oxidase (CO). Plasticity in the adult primate visual system has also been documented by changes in CO activity. Yet, the molecular mechanism of regulating this enzyme remains not well understood. Being one of only four bigenomic enzymes in mammalian cells, the transcriptional regulation of this enzyme necessitates a potential bigenomic coordinator. Nuclear respiratory factor 2 (NRF-2) or GA-binding protein is a transcription factor that may serve such a critical role. The goal of the present study was to determine if the two major subunits of NRF-2, 2α and 2β, had distinct subcellular distribution in neurons of the rat and monkey visual cortex, if major metabolic neuronal types in the macaque exhibited different levels of the two subunits, and if they would respond differently to monocular impulse blockade. Quantitative immuno-electron microscopy was used. In both rats and monkeys, nuclear labeling of α and β subunits was mainly over euchromatin rather than heterochromatin, consistent with their active participation in transcriptional activity. Cytoplasmic labeling was over free ribosomes, the Golgi apparatus, and occasionally the nuclear envelope, signifying sites of synthesis and possible posttranslational modifications. The density of both subunits was much higher in the nucleus than in the cytoplasm for all neurons examined, again indicating that their major sites of cellular action is in the nucleus. In both layer IVC and supragranular puffs of the macaque visual cortex, the expression of both NRF-2α and β was higher in medium-sized, non-pyramidal (type C and C-like) cells previously shown to have higher CO activity than small, type A and A-like cells with low CO activity. Pyramidal, type B cells in puffs had intermediate levels of CO as well as NRF-2α and β labeling. Monocular impulse blockade induced a greater reduction of NRF-2 labeling in type C/C-like than type A/A-like cells. These results substantiate and extend our previous findings that NRF-2 is constitutively active in adult primate and rat visual cortical neurons, that it is expressed more strongly in metabolically more active neurons, and that its level is directly regulated by neuronal activity, the blockade of which imposes a greater down-regulation of this transcription factor in metabolically more active than less active neurons.

Retinal bipolar cells show heterogeneous expression of voltage-dependent Na+ and K+ currents. We used whole-cell patch-clamp recordings to investigate the possible roles of these currents in the response properties of bipolar cells in rats. Isolated bipolar cells showed robust spontaneous regenerative activity, but the regenerative potential of rod bipolar cells reached a more depolarized level than that of cone bipolar cells. In both isolated cells and cells in retinal slices, the membrane depolarization evoked by current injection was apparently capped. The evoked membrane potential was again more depolarized in rod bipolar cells than in cone bipolar cells. Application of tetraethylammonium and 4-aminopyridine shifted the spontaneous regenerative potential as well as the evoked potential to a more depolarized level. In addition, a subclass of cone bipolar cells showed a prominent spike in the initial phase of the voltage response when the cells were depolarized from a relatively negative membrane potential. The spike was mediated mainly by tetrodotoxin-sensitive Na+ current. The presence of the spike sped up the response kinetics and enhanced the peak membrane potential. Results of this study raise the possibility that voltage-dependent K+ currents may play a role in defining different membrane operating ranges of rod and cone bipolar cells and that voltage-dependent Na+ currents may enhance the response kinetics and amplitude of certain cone bipolar cells.

Phosducin (Pd) is a 28-kD phosphoprotein whose expression in retina appears limited to photoreceptor cells. Pd binds to the β,γ subunits of transducin (Gt). Their binding affinity is markedly diminished by Pd phosphorylation. While Pd has long been regarded as a candidate for the regulation of Gt, the molecular details of Pd function remain unclear. This gap in understanding is due in part to a lack of precise information concerning the total amount and subcellular localization of rod Pd. While earlier studies suggested that Pd was a rod outer segment (ROS) protein, recent findings have demonstrated that Pd is distributed throughout the rod. In this report, the subcellular distribution and amounts of rat Pd are quantified with immunogold electron microscopy. After light or dark adaptation, retinal tissues were fixed in situ and prepared for ultrathin sectioning and immunogold labeling. Pd concentrations were analyzed over the entire length of the rod. The highest Pd labeling densities were found in the rod synapse. Less intense Pd staining was observed in the ellipsoid and myoid regions, while minimal labeling densities were found in the ROS and the rod nucleus. In contrast with rod Gt, no evidence was found for light-dependent movement of Pd between inner and outer segments. There is a relative paucity of Pd in the ROS as compared with the large amounts of Gt found there. This does not support the earlier idea that Pd could modulate Gt activity by controlling its concentration. On the other hand, the presence of Pd in the nucleus is consistent with its possible role as a regulator of transcription. The functions of Pd in the ellipsoid and myoid regions remain unclear. The highest concentration of Pd was found at the rod synapse, consistent with a suggested role for Pd in the regulation of synaptic function.

Calcium ion (Ca2+) signaling has been widely implicated in developmental events in the retina, but little is known about the specific mechanisms utilized by developing neurons to decrease intracellular Ca2+. Using immunocytochemistry, we determined the expression profiles of all known isoforms of a key Ca2+ transporter, the plasma membrane Ca2+ ATPase (PMCA), in the rat retina. During the first postnatal week, the four PMCA isoforms were expressed in patterns that differed from their expression in the adult retina. At birth, PMCA1 was found in the ventricular zone and nascent cell processes in the distal retina as well as in ganglion and amacrine cells. After the first postnatal week, PMCA1 became restricted to photoreceptors and cone bipolar cells. By P10 (by postnatal day 10), most inner retinal PMCA consisted of PMCA2 and PMCA3. Prominent PMCA4 expression appeared after the first postnatal week and was confined primarily to the ON sublamina of the inner plexiform layer (IPL). The four PMCA isoforms could play distinct functional roles in the development of the mammalian retina even before synaptic circuits are established. Their expression patterns are consistent with the hypothesis that inner and outer retinal neurons have different Ca2+ handling needs.

To describe the wide-field ganglion cells, they were injected intracellularly with Neurobiotin using an in vitro preparation of macaque retina and labeled with streptavidin-Cy3. The retinas were then labeled with antibodies to choline acetyltransferase and other markers to indicate the depth of the dendrites within the inner plexiform layer (IPL) and analyzed by confocal microscopy. There were eight different subtypes of narrowly unistratified cells that ramified in each of the 5 strata, S1–5, including narrow thorny, large sparse, large moderate, large dense, large radiate, narrow wavy, large very sparse, and fine very sparse. There were four types of broadly stratified cells with dendritic trees extending from S4 to S2. One type resembled the parvocellular giant cell and another the broad thorny type described previously in primates. Another broadly stratified cell was called multi-tufted based on its distinctive dendritic branching pattern. The fourth type had been described previously, but not named; we called it broad wavy. There was a bistratified type with its major arbor in S5, the same level as the blue cone bipolar cell; it resembled the large, bistratified cell with blue ON-yellow OFF responses described recently. Two wide-field ganglion cell types were classified as diffuse because they had dendrites throughout the IPL. One had many small branches and was named thorny diffuse. The second was named smooth diffuse because it had straighter dendrites that lacked these processes. Dendrites of the large moderate and multi-tufted cells cofasciculated with ON-starburst cell dendrites and were, therefore, candidates to be ON- and ON–OFF direction-selective ganglion cells, respectively. We concluded that there are at least 15 morphoplogical types of wide-field ganglion cells in macaque retinas.

The use of knock-out and transgenic mice has been instrumental for advancing our understanding of retinal development and disease. In this perspective, we review existing genetic approaches to studying retinal development and present a series of new genetic tools that complement the use of standard knock-out and transgenic mice. Particular emphasis is placed on elucidating cell-autonomous and non-cell-autonomous roles of genes important for retinal development and disease in vivo. In addition, a series of gene-swapping vectors can be used to elucidate the function of proteins that regulate key processes in retinal development and a wide variety of retinopathies.

The biochemical and morphological specializations of rod and cone photoreceptors reflect their roles in sight. The apoprotein opsin, which converts photons into chemical signals, functions at one end of these highly polarized cells, in the outer segment. Previous work has shown that the mRNA of rod opsin, the opsin specific to rods, is renewed in the outer segment with a diurnal rhythm in the retina of the teleost fish Haplochromis burtoni. Here we show that in the same species, all three cone opsin mRNAs (blue, green, and red) also have a diurnal rhythm of expression. Quantitative real-time polymerase chain reaction (PCR) with primer pairs specific for the cone photoreceptor opsin subtypes was used to detect opsin mRNA abundance in animals sacrificed at 3-h intervals around the clock. All three cone opsins were expressed with diurnal rhythms similar to each other but out of phase with the rod opsin rhythm. Specifically, cone opsin expression occurs at a higher level near the onset of the dark period, when cones are not used for vision. Finally, we found that the rhythm of cone opsin expression in fish appears to be light dependent, as prolonged darkness changes normal diurnal expression patterns.

The morphology and distribution of normally placed and displaced A horizontal cells were studied in the retina of a diurnal hystricomorph rodent, the agouti Dasyprocta aguti. Cells were labeled with anti-calbindin immunocytochemistry. Dendritic-field size reaches a minimum in the visual streak, of about 9000 μm2, and increases toward the retinal periphery both in the dorsal and ventral regions. There is a dorsoventral asymmetry, with dorsal cells being larger than ventral cells at equal distances from the streak. The peak value for cell density of 281 ± 28 cells/mm2 occurs in the center of the visual streak, decreasing toward the dorsal and ventral retinal periphery, paralleling the increase in dendritic-field size. Along the visual streak, the decline in cell density is less pronounced, remaining between 100–200 cells/mm2 in the temporal and nasal periphery. Displaced horizontal cells are rare and occur in the retinal periphery. They tend to be smaller than normally placed horizontal cells in the ventral region, whilst no systematic difference was observed between the two cell groups in the dorsal region. Mosaic regularity was studied using nearest-neighbor analysis and the Ripley function. When mosaic regularity was determined removing the displaced horizontal cells, there was a slight increase in the conformity ratio, but the bivariate Ripley function indicated some repulsive dependence between the two mosaics. Both results were near the level of significance. A similar analysis performed in the capybara retina, a closely related hystricomorph rodent bearing a higher density of displaced horizontal cells than found in the agouti, suggested spatial independence between the two mosaics, normally placed versus displaced horizontal cells.

We measured mosaic properties of midget and parasol ganglion cells in the retina of a New World monkey, the common marmoset Callithrix jacchus. We addressed the functional specialization of these populations for color and spatial vision, by comparing the mosaic of ganglion cells in dichromatic (“red–green color blind”) and trichromatic marmosets. Ganglion cells were labelled by photolytic amplification of retrograde marker (“photofilling”) following injections into the lateral geniculate nucleus, or by intracellular injection in an in vitro retinal preparation. The dendritic-field size, shape, and overlap of neighboring cells were measured. We show that in marmosets, both midget and parasol cells exhibit a radial bias, so that the long axis of the dendritic field points towards the fovea. The radial bias is similar for parasol cells and midget cells, despite the fact that midget cell dendritic fields are more elongated than are those of parasol cells. The dendritic fields of midget ganglion cells from the same (ON or OFF) response-type array show very little overlap, consistent with the low coverage of the midget mosaic in humans. No large differences in radial bias, or overlap, were seen on comparing retinae from dichromatic and trichromatic animals. These data suggest that radial bias in ganglion cell populations is a consistent feature of the primate retina. Furthermore, they suggest that the mosaic properties of the midget cell population are associated with high spatial resolution rather than being specifically associated with trichromatic color vision.

The genetic locus for incomplete congenital stationary night blindness (CSNB2) has been identified as the CACNA1f gene, encoding the α1F calcium channel subunit, a member of the L-type family of calcium channels. The electroretinogram associated with CSNB2 implicates α1F in synaptic transmission between retinal photoreceptors and bipolar cells. Using a recently developed monoclonal antibody to α1F, we localize the channel to ribbon active zones in rod photoreceptor terminals of the mouse retina, supporting a role for α1F in mediating glutamate release from rods. Detergent extraction experiments indicate that α1F is part of a detergent-resistant active zone complex, which also includes the synaptic ribbons. Comparison of native mouse rod calcium currents with recombinant α1F currents reveals that the current–voltage relationship for the native current is shifted approximately 30 mV to more hyperpolarized potentials than for the recombinant α1F current, suggesting modulation of the native channel by intracellular factors. Lastly, we present evidence for L-type α1D calcium channel subunits in cone terminals of the mouse retina. The presence of α1D channels in cones may explain the residual visual abilities of individuals with CSNB2.

Retinal neurons and Müller cells express amiloride-sensitive Na+ channels (ASSCs). Although all major subunits of these channels are expressed, their physiological role is relatively unknown in this system. In the present study, we used the electroretinogram (ERG) recorded from anesthetized rabbits and isolated rat and rabbit retina preparations to investigate the physiological significance of ASSCs in the retina. Based upon our previous study showing expression of α-ENaC and functional amiloride-sensitive currents in rabbit Müller cells, we expected changes in Müller cell components of the ERG. However, we observed changes in other components of the ERG as well. The presence of amiloride elicited changes in all major components of the ERG; the a-wave, b-wave, and d-wave (off response) were enhanced, while there was a reduction in the amplitude of the Müller cell response (slow PIII). These results suggest that ASSCs play an important role in retinal function including neuronal and Müller cell physiology.

Retinal development in 3 strains of rd-3/rd-3 mutant mice, previously shown to have different rates of degeneration, was studied using light, electron, and immunofluorescence microscopy. The time course and phenotype of the degeneration as well as details on the mechanism of massive photoreceptor cell loss are compared with other known retinal degenerations in mice. Up until postnatal day (P) 10, the retinas of all three strains (RBF, 4Bnr, In-30) develop similarly to those of pigmented and nonpigmented controls. TUNEL-positive cells appear in the outer nuclear layer (ONL) by P14, and reach a maximum in all three mutant strains around P21. Scattered rods and cones form a loose, monolayered ONL by 8 weeks in the albino RBF strain, by 10 weeks in the albino 4Bnr strain, and by 16 weeks in the pigmented In-30 strain. Though the initial degeneration begins in the central retina, there is no preferred gradient of cell death between central and peripheral photoreceptors. Rods and cones are present at all ages examined. During development, stacks of outer segments (OS) form in all three strains though they never achieve full adult lengths, and often have disorganized, atypical OS. Rod opsin is expressed in the developing OS but is redistributed into plasma membrane as OS degeneration proceeds. Retinal pigment epithelial (RPE) cells of all mutant strains contain packets of phagocytosed OS, and their apical processes associate with the distal ends of the OS. At their synaptic sites, photoreceptor terminals contain ribbons apposed to apparently normal postsynaptic triads. As photoreceptors are lost, Müller cells fill in space in the ONL but they do not appear to undergo significant hypertrophy or migration, though during the degeneration, glial fibrillary acidic protein (GFAP) expression is gradually upregulated. Macrophage-like cells are found frequently in the subretinal space after the onset of photoreceptor apoptosis. As OS disappear, the RPE apical processes revert to simple microvilli. Late in the degeneration, some RPE cells die and neighboring cells appear to flatten as if to maintain confluence. In regions of RPE cell loss that happen to lie above retina where the ONL is gone, cells of the inner nuclear layer (INL), wrapped by Müller cell processes, may front directly on Bruch's membrane.

Brain-derived neurotrophic factor (BDNF) is a preferred ligand for a member of the tropomyosin-related receptor family, trkB. Activation of trkB is implicated in various activity-independent as well as activity-dependent growth processes in many developing and mature neural systems. In the subcortical visual system, where electrical activity has been implicated in normal development, both differential survival, as well as remodeling of axonal arbors, have been suggested to contribute to eye-specific segregation of retinal ganglion cell inputs. Here, we tested whether BDNF is required for eye-specific segregation of visual inputs to the lateral geniculate nucleus and the superior colliculus, and two other major subcortical target fields in mice. We report that eye-specific patterning is normal in two mutants that lack BDNF expression during the segregation period: a germ-line knockout for BDNF, and a conditional mutant in which BDNF expression is absent or greatly reduced in the central nervous system. We conclude that the availability of BDNF is not necessary for eye-specific segregation in subcortical visual nuclei.

Visual systems break scenes down into individual features, processed in distinct channels, and then selectively recombine those features according to the demands of particular behavioral tasks. In primates, for example, there are distinct pathways for motion and form processing. While form vision utilizes color information, motion pathways receive input from only a subset of cone photoreceptors and are generally colorblind. To explore the link between early channeling of visual information and behavioral output across vertebrate species, we measured the chromatic inputs to the optomotor response of larval zebrafish. Using cone-isolating gratings, we found that there is a strong input from both red and green cones but not short-wavelength cones, which nevertheless do contribute to another behavior, phototaxis. Using a motion-nulling method, we measured precisely the input strength of gratings that stimulated cones in combination. The fish do not respond to gratings that stimulate different cone types out of phase, but have an enhanced response when the cones are stimulated together. This shows that red and green cone signals are pooled at a stage before motion detection. Since the two cone inputs are combined into a single ‘luminance’ channel, the response to sinusoidal gratings is colorblind. However, we also find that the relative contributions of the two cones at isoluminance varies with spatial frequency. Therefore, natural stimuli, which contain a mixture of spatial frequencies, are likely to be visible regardless of their chromatic composition.

Horizontal cell morphology was studied in the retina of the nocturnal owl-monkey, Aotus, and compared with that of its diurnal, close relative, the capuchin monkey, Cebus. Cells were initially labeled with DiI and the staining was later photoconverted in a stable precipitated using DAB as chromogen. The sizes of cell bodies, dendritic fields, and axon terminals, number of dendritic clusters, intercluster spacing, and intercone spacing were measured at increasing eccentricities. Two distinct morphological classes of horizontal cells were identified, which resembled those of H1 and H3 cells described in diurnal monkeys. A few examples of a third class, possibly corresponding to the H2 cells of diurnal monkeys, were labeled. Both H1 and H3 cells increased in size and had increasing numbers of dendritic clusters with eccentricity. H3 cells were larger and had a larger number of dendritic clusters than H1 cells. Owl-monkey H1 cells had larger dendritic fields than capuchin monkey H1 cells at all quadrants in the central and midperipheral retinal regions, but the difference disappeared in the far periphery. Owl-monkey and capuchin monkey H1 cells had about the same number of dendritic clusters across eccentricity. As owl-monkey H1 cells were larger than capuchin monkey H1 cells, the equal number of clusters in these two primates was due to the fact that they were more spaced in the owl-monkey cells. H1 intercluster distance closely matched intercone spacing for both the owl-monkey and capuchin monkey retinas. On the other hand, H3 intercluster distance was larger than intercone spacing in the retina of both primates. Owl-monkey H1 axon terminals had 2–3 times more knobs than capuchin monkey H1 axon terminals in spite of having about the same size and, consequently, knob density was 2–3 times higher for owl-monkey than capuchin monkey H1 axon terminals across all eccentricities. The differences observed between owl-monkey and capuchin monkey horizontal cells, regarding the morphology of their dendritic trees and axon terminals, may be related to the differences found in the cone-to-rod ratio in the retina of these two primates. They seem to represent retinal specializations to the nocturnal and diurnal life styles of the owl-monkey and capuchin monkey, respectively.

The present investigation has sought to determine whether the population of retinal horizontal cells undergoes an increase in the precision of its mosaic patterning during postnatal development, and if so, whether this increase is compatible with three different mechanistic accounts of retinal mosaic formation. Horizontal cells were labeled with antibodies to neurofilaments or calbindin at different developmental stages, and then visualized in retinal wholemounts. Multiple fields were sampled from each retina to determine horizontal cell density, while the X–Y coordinates of each cell in a field were determined. An estimate of total horizontal cell number was calculated for each retina, while the Voronoi domain regularity index and the packing factor were computed for each field. Two strains of mice showing a two-fold difference in the size of their horizontal cell population in maturity were sampled, C57BL/6J and A/J. Horizontal cell number in C57BL/6J was approximately twice that observed in A/J at all postnatal stages, with neither strain showing an effect of age on horizontal cell number. In both strains, however, the Voronoi domain regularity index and the packing factor were significantly lower at P-1 relative to later developmental stages. These results show that accounts of mosaic formation proposing the selective death of irregularly positioned cells, or the periodic occurrence of fate-determining events, are insufficient to establish the final patterning achieved by horizontal cells. Rather, they support the hypothesis that tangential dispersion enhances mosaic patterning during postnatal development.