Involvement of integrins alpha v beta 3 and alpha v beta 5 in ocular neovascular diseases.

Angiogenesis underlies the majority of eye diseases that result in catastrophic loss of vision. Recent evidence has implicated the integrins alpha v beta 3 and alpha v beta 5 in the angiogenic process. We examined the expression of alpha v beta 3 and alpha v beta 5 in neovascular ocular tissue from patients with subretinal neovascularization from age-related macular degeneration or the presumed ocular histoplasmosis syndrome or retinal neovascularization from proliferative diabetic retinopathy (PDR). Only alpha v beta 3 was observed on blood vessels in ocular tissues with active neovascularization from patients with age-related macular degeneration or presumed ocular histoplasmosis, whereas both alpha v beta 3 and alpha v beta 5 were present on vascular cells in tissues from patients with PDR. Since we observed both integrins on vascular cells from tissues of patients with retinal neovascularization from PDR, we examined the effects of a systemically administered cyclic peptide antagonist of alpha v beta 3 and alpha v beta 5 on retinal angiogenesis in a murine model. This antagonist specifically blocked new blood vessel formation with no effect on established vessels. These results not only reinforce the concept that retinal and subretinal neovascular diseases are distinct pathological processes, but that antagonists of alpha v beta 3 and/or alpha v beta 5 may be effective in treating individuals with blinding eye disease associated with angiogenesis.

Starburst amacrine cells change from spiking to nonspiking neurons during retinal development.

The membrane excitability of cholinergic (starburst) amacrine cells was studied in the rabbit retina during postnatal development. Whole-cell patch-clamp recordings were made from 110 displaced starburst cells in a thin retina] slice preparation of rabbits between postnatal days P1 and P56 old. We report that displaced starburst cells undergo a dramatic transition from spiking to nonspiking, caused by a loss of voltage-gated Na currents. This change in membrane excitability occurred just after eye opening (P10), such that all of the starburst cells tested before eye opening had conspicuous tetrodotoxin-sensitive Na currents and action potentials, but none tested after the first 3 postnatal weeks had detectable Na currents or spikes. Our results suggest that starburst cells use action potentials transiently during development and probably play a functional role in visual development. These cells then cease to spike as the retina matures, presumably consistent with their role in visual processing in the mature retina.

Retinal engineering: engrafted neural cell lines locate in appropriate layers.

A major question in central nervous system development, including the neuroretina, is whether migrating cells express cues to find their way and settle at specific locations. We have transplanted quail neuroretinal cell lines QNR/D, a putative amacrine or ganglion cell, and QNR/K2, a putative Müller cell into chicken embryo eyes. Implanted QNR/D cells migrate only to the retinal ganglion and amacrine cell layers and project neurites in the plane of retina; in contrast, QNR/K2 cells migrate through the ganglion and amacrine layers, locate in the inner nuclear layer, and project processes across the retina. These data show that QNR/D and QNR/K2 cell lines represent distinct neural cell types, suggesting that migrating neural cells express distinct address cues. Furthermore, our results raise the possibility that immortalized cell lines can be used for replacement of specific cell types and for the transport of genes to given locations in neuroretina.

A circadian clock regulates rod and cone input to fish retinal cone horizontal cells.

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. In some of the experiments, fish were kept in constant darkness for 3-48 hr prior to surgery. After prolonged darkness during the night, but not during the day, the light responses of L-type cone horizontal cells resembled those of rod horizontal cells with respect to threshold, waveform, intensity-response functions, and spectral sensitivity. Following light sensitization during the night and day, the light responses of rod and cone horizontal cells were clearly different with respect to threshold, waveform, intensity-response functions, and spectral sensitivity. Under conditions of constant darkness for two full light/dark cycles, average responses of cone horizontal cells to a bright light stimulus during the subjective day were greater than during the subjective night. Prior reversal of the light/dark cycle reversed the 24-hr rhythm of cone horizontal cell responses to bright lights. In addition, following one full cycle of constant darkness, average cone horizontal cell spectral sensitivity during the subjective night closely matched that of rod horizontal cells, whereas average cone horizontal cell spectral sensitivity during the subjective day was similar to that of red (625 nm) cones. These results indicate that the effects of dark adaptation depend on the time of day and are regulated by a circadian clock so that cone input to cone horizontal cells predominates in the day and rod input predominates in the night.

Oligodeoxynucleotides inhibit retinal neovascularization in a murine model of proliferative retinopathy.

Diseases characterized by retinal neovascularization are among the principal causes of visual loss worldwide. The hypoxia-stimulated expression of vascular endothelial growth factor (VEGF) has been implicated in the proliferation of new blood vessels. We have investigated the use of antisense phosphorothioate oligodeoxynucleotides against murine VEGF to inhibit retinal neovascularization and VEGF synthesis in a murine model of proliferative retinopathy. Intravitreal injections of two different antisense phosphorothioate oligodeoxynucleotides prior to the onset of proliferative retinopathy reduced new blood vessel growth a mean of 25 and 31% compared with controls. This inhibition was dependent on the concentration of antisense phosphorothioate oligodeoxynucleotides and resulted in a 40-66% reduction in the level of VEGF protein, as determined by Western blot analysis. Control (sense, nonspecific) phosphorothioate oligodeoxynucleotides did not cause a significant reduction in retinal neovascularization or VEGF protein levels. These data further establish a fundamental role for VEGF expression in ischemia-induced proliferative retinopathies and a potential therapeutic use for antisense phosphorothioate oligodeoxynucleotides.

POU domain factor Brn-3b is required for the development of a large set of retinal ganglion cells.

The three members of the Brn-3 family of POU domain transcription factors are found in highly restricted sets of central nervous system neurons. Within the retina, these factors are present only within subsets of ganglion cells. We show here that in the developing mouse retina, Brn-3b protein is first observed in presumptive ganglion cell precursors as they begin to migrate from the zone of dividing neuroblasts to the future ganglion cell layer, and that targeted disruption of the Brn-3b gene leads in the homozygous state to a selective loss of 70% of retinal ganglion cells. In Brn-3b (-/-) mice other neurons within the retina and brain are minimally or not at all affected. These experiments indicate that Brn-3b plays an essential role in the development of specific ganglion cell types.

Retinal glial cell glutamate transporter is coupled to an anionic conductance.

Application of L-glutamate to retinal glial (Müller) cells results in an inwardly rectifying current due to the net influx of one positive charge per molecule of glutamate transported into the cell. However, at positive potentials an outward current can be elicited by glutamate. This outward current is eliminated by removal of external chloride ions. Substitution of external chloride with the anions thiocyanate, perchlorate, nitrate, and iodide, which are known to be more permeant at other chloride channels, results in a considerably larger glutamate-elicited outward current at positive potentials. The large outward current in external nitrate has the same ionic dependence, apparent affinity for L-glutamate, and pharmacology as the glutamate transporter previously reported to exist in these cells. Varying the concentration of external nitrate shifts the reversal potential in a manner consistent with a conductance permeable to nitrate. Together, these results suggest that the glutamate transporter in retinal glial cells is associated with an anionic conductance. This anionic conductance may be important for preventing a reduction in the rate of transport due the depolarization that would otherwise occur as a result of electrogenic glutamate uptake.

Definition of the HLA-A29 peptide ligand motif allows prediction of potential T-cell epitopes from the retinal soluble antigen, a candidate autoantigen in birdshot retinopathy.

The peptide-binding motif of HLA-A29, the predisposing allele for birdshot retinopathy, was determined after acid-elution of endogenous peptides from purified HLA-A29 molecules. Individual and pooled HPLC fractions were sequenced by Edman degradation. Major anchor residues could be defined as glutamate at the second position of the peptide and as tyrosine at the carboxyl terminus. In vitro binding of polyglycine synthetic peptides to purified HLA-A29 molecules also revealed the need for an auxiliary anchor residue at the third position, preferably phenylalanine. By using this motif, we synthesized six peptides from the retinal soluble antigen, a candidate autoantigen in autoimmune uveoretinitis. Their in vitro binding was tested on HLA-A29 and also on HLA-B44 and HLA-B61, two alleles sharing close peptide-binding motifs. Two peptides derived from the carboxyl-terminal sequence of the human retinal soluble antigen bound efficiently to HLA-A29. This study could contribute to the prediction of T-cell epitopes from retinal autoantigens implicated in birdshot retinopathy.

Regulation of phosducin phosphorylation in retinal rods by Ca2+/calmodulin-dependent adenylyl cyclase.

The phosphoprotein phosducin (Pd) regulates many guanine nucleotide binding protein (G protein)-linked signaling pathways. In visual signal transduction, unphosphorylated Pd blocks the interaction of light-activated rhodopsin with its G protein (Gt) by binding to the beta gamma subunits of Gt and preventing their association with the Gt alpha subunit. When Pd is phosphorylated by cAMP-dependent protein kinase, it no longer inhibits Gt subunit interactions. Thus, factors that determine the phosphorylation state of Pd in rod outer segments are important in controlling the number of Gts available for activation by rhodopsin. The cyclic nucleotide dependencies of the rate of Pd phosphorylation by endogenous cAMP-dependent protein kinase suggest that cAMP, and not cGMP, controls Pd phosphorylation. The synthesis of cAMP by adenylyl cyclase in rod outer segment preparations was found to be dependent on Ca2+ and calmodulin. The Ca2+ dependence was within the physiological range of Ca2+ concentrations in rods (K1/2 = 230 +/- 9 nM) and was highly cooperative (n app = 3.6 +/- 0.5). Through its effect on adenylyl cyclase and cAMP-dependent protein kinase, physiologically high Ca2+ (1100 nM) was found to increase the rate of Pd phosphorylation 3-fold compared to the rate of phosphorylation at physiologically low Ca2+ (8 nM). No evidence for Pd phosphorylation by other (Ca2+)-dependent kinases was found. These results suggest that Ca2+ can regulate the light response at the level of Gt activation through its effect on the phosphorylation state of Pd.

Determination of the transiently lowered pKa of the retinal Schiff base during the photocycle of bacteriorhodopsin.

Reprotonation of the transiently deprotonated retinal Schiff base in the bacteriorhodopsin photocycle is greatly slowed when the proton donor Asp-96 is removed with site-specific mutagenesis, but its rate is restored upon adding azide or other weak acids such as formate and cyanate. As expected, between pH 3 and 7 the rate of Schiff base protonation in the photocycle of the D96N mutant correlates with the concentrations of the acid forms of these agents. Dissection of the rates in the biexponential reprotonation kinetics of the Schiff base between pH 7 and 9 yielded calculated rate constants for the protonation equilibrium. Their dependencies on pH and azide or cyanate concentrations are consistent with both earlier suggested mechanisms: (i) azide and other weak acids may function as proton carriers in the protonation equilibrium of the Schiff base, or (ii) the binding of their anionic forms may catalyze proton conduction to and from the Schiff base. The measured rate constants allow the calculation of the pKa of the Schiff base during its reprotonation in the photocycle of D96N. It is 8.2-8.3, a value much below the pKa determined earlier in unphotolyzed bacteriorhodopsin.

Molecular biology of retinal ganglion cells.

Retinal ganglion cells are the output neurons that encode and transmit information from the eye to the brain. Their diverse physiologic and anatomic properties have been intensively studied and appear to account well for a number of psychophysical phenomena such as lateral inhibition and chromatic opponency. In this paper, we summarize our current view of retinal ganglion cell properties and pose a number of questions regarding underlying molecular mechanisms. As an example of one approach to understanding molecular mechanisms, we describe recent work on several POU domain transcription factors that are expressed in subsets of retinal ganglion cells and that appear to be involved in ganglion cell development.

Multineuronal codes in retinal signaling.

The visual world is presented to the brain through patterns of action potentials in the population of optic nerve fibers. Single-neuron recordings show that each retinal ganglion cell has a spatially restricted receptive field, a limited integration time, and a characteristic spectral sensitivity. Collectively, these response properties define the visual message conveyed by that neuron's action potentials. Since the size of the optic nerve is strictly constrained, one expects the retina to generate a highly efficient representation of the visual scene. By contrast, the receptive fields of nearby ganglion cells often overlap, suggesting great redundancy among the retinal output signals. Recent multineuron recordings may help resolve this paradox. They reveal concerted firing patterns among ganglion cells, in which small groups of nearby neurons fire synchronously with delays of only a few milliseconds. As there are many more such firing patterns than ganglion cells, such a distributed code might allow the retina to compress a large number of distinct visual messages into a small number of optic nerve fibers. This paper will review the evidence for a distributed coding scheme in the retinal output. The performance limits of such codes are analyzed with simple examples, illustrating that they allow a powerful trade-off between spatial and temporal resolution.

Molecular mechanism of protein-retinal coupling in bacteriorhodopsin.

Bacteriorhodopsin is a membrane protein that functions as a light-driven proton pump. Each cycle of proton transport is initiated by the light-induced isomerization of retinal from the all-trans to 13-cis configuration and is completed by the protein-driven reisomerization of retinal to the all-trans configuration. Previous studies have shown that replacement of Leu-93, a residue in close proximity to the 13-methyl group of retinal, by alanine, resulted in a 250-fold increase in the time required to complete each photocycle. Here, we show that the kinetic defect in the photocycle of the Leu-93-->Ala mutant occurs at a stage after the completion of proton transport and can be overcome in the presence of strong background illumination. Time-resolved retinal-extraction experiments demonstrate the continued presence of a 13-cis intermediate in the photocycle of the Leu-93-->Ala mutant well after the completion of proton release and uptake. These results indicate that retinal reisomerization is kinetically the rate-limiting step in the photocycle of this mutant and that the slow thermal reisomerization can be bypassed by the absorption of a second photon. The effects observed for the Leu-93-->Ala mutant are not observed upon replacement of any other residue in van der Waals contact with retinal or upon replacement of Leu-93 by valine. We conclude that the contact between Leu-93 and the 13-methyl group of retinal plays a key role in controlling the rate of protein conformational changes associated with retinal reisomerization and return of the protein to the initial state.

Suppression of retinal neovascularization in vivo by inhibition of vascular endothelial growth factor (VEGF) using soluble VEGF-receptor chimeric proteins.

The majority of severe visual loss in the United States results from complications associated with retinal neovascularization in patients with ischemic ocular diseases such as diabetic retinopathy, retinal vein occlusion, and retinopathy of prematurity. Intraocular expression of the angiogenic protein vascular endothelial growth factor (VEGF) is closely correlated with neovascularization in these human disorders and with ischemia-induced retinal neovascularization in mice. In this study, we evaluated whether in vivo inhibition of VEGF action could suppress retinal neovascularization in a murine model of ischemic retinopathy. VEGF-neutralizing chimeric proteins were constructed by joining the extracellular domain of either human (Flt) or mouse (Flk) high-affinity VEGF receptors with IgG. Control chimeric proteins that did not bind VEGF were also used. VEGF-receptor chimeric proteins eliminated in vitro retinal endothelial cell growth stimulation by either VEGF (P < 0.006) or hypoxic conditioned medium (P < 0.005) without affecting growth under nonstimulated conditions. Control proteins had no effect. To assess in vivo response, animals with bilateral retinal ischemia received intravitreal injections of VEGF antagonist in one eye and control protein in the contralateral eye. Retinal neovascularization was quantitated histologically by a masked protocol. Retinal neovascularization in the eye injected with human Flt or murine Flk chimeric protein was reduced in 100% (25/25; P < 0.0001) and 95% (21/22; P < 0.0001) 0.0001) of animals, respectively, compared to the control treated eye. This response was evident after only a single intravitreal injection and was dose dependent with suppression of neovascularization noted after total delivery of 200 ng of protein (P < 0.002). Reduction of histologically evident neovascular nuclei per 6-microns section averaged 47% +/- 4% (P < 0.001) and 37% +/- 2% (P < 0.001) for Flt and Flk chimeric proteins with maximal inhibitory effects of 77% and 66%, respectively. No retinal toxicity was observed by light microscopy. These data demonstrate VEGF's causal role in retinal angiogenesis and prove the potential of VEGF inhibition as a specific therapy for ischemic retinal disease.

Glial cell line-derived neurotrophic factor but not transforming growth factor beta 3 prevents delayed degeneration of nigral dopaminergic neurons following striatal 6-hydroxydopamine lesion.

Glial cell line-derived neurotrophic factor (GDNF) and transforming growth factor beta 3 (TGF-beta 3) are members of the TGF-beta superfamily with high neurotrophic activity on cultured nigral dopamine neurons. We investigated the effects of intracerebral administration of GDNF and TGF-beta 3 on the delayed cell death of the dopamine neurons in the rat substantia nigra following 6-hydroxydopamine lesions of dopaminergic terminals in the striatum. Fluorescent retrograde tracer injections and tyrosine hydroxylase immunocytochemistry demonstrated nigral degeneration with an onset 1 week after lesion, leading to extensive death of nigral neurons 4 weeks postlesion. Administration of recombinant human GDNF for 4 weeks over the substantia nigra at a cumulative dose of 140 micrograms, starting on the day of lesion, completely prevented nigral cell death and atrophy, while a single injection of 10 micrograms 1 week postlesion had a partially protective effect. Continuous administration of TGF-beta 3, starting on the day of lesion surgery, did not affect nigral cell death or atrophy. These findings support the notion that GDNF, but not TGF-beta 3, is a potent neurotrophic factor for nigral dopamine neurons in vivo.