Previous Section || Go to Beginning || Next Section

Report of the Retinal Diseases Panel

Program Overview and Goals

The retina is a specialized light-sensitive tissue that contains photoreceptor cells (rods and cones) and neurons connected to a neural network for the processing of visual information. This information is sent to the brain for decoding into a visual image. The accessibility of and diversity within the retina make it an ideal system to study. Such studies have a wealth of information on fundamental mechanisms that can be applied to the nervous system. The retina depends on cells of the adjacent retinal pigment epithelium (RPE) for support of its metabolic functions. Photoreceptors in the retina, perhaps because of their huge energy requirements and highly differentiated state, are sensitive to a variety of genetic and environmental insults. The retina is thus susceptible to an array of diseases that result in visual loss or complete blindness.

Two major systems of blood vessels nourish the retina. The retinal circulation feeds the inner layers of the retina (nearest the vitreous), while the choroidal circulation feeds the outer retina comprised of the RPE and photoreceptor cells. Vessels of the retinal circulation have endothelial tight junctions, like capillaries of the brain, serving as a barrier to the diffusion of large molecules. This is the cellular basis of the blood-retinal barrier. The retinal vasculature can be affected by three pathological processes: excessive permeability, vascular closure, and proliferation of newly formed blood vessels (neovascularization).

Diabetic retinopathy is a major cause of excessive permeability and is typically accompanied by neovascularization with ballooning of the retinal capillaries to form microaneurysms. The blood-retinal barrier may break down within these microaneurysms, causing leakage of blood proteins with subsequent hemorrhage into the retina and visual loss. Newly formed blood vessels tend to break through the retinal surface, which may result in hemorrhage into the vitreous and in traction retinal detachment, where the retina is pulled away from the underlying choroid. Because of the prevalence of diabetes, diabetic retinopathy is a major cause of blindness. Laser photocoagulation is a useful clinical tool for treating proliferative retinopathy. Unfortunately, it is estimated that as many as 50 percent of patients are not diagnosed at a stage early enough for this treatment to be effective.

Ocular inflammatory diseases (uveitis) represent another category of retinal vascular disorders. Uveitis can be categorized as infectious and noninfectious types. Infectious agents, such as viruses and bacteria, can be very destructive when they enter the eye. Noninfectious inflammation is caused by problems with immune regulation. Uveitis is usually associated with painless but rapid visual loss.

The acquired immunodeficiency syndrome (AIDS) has had a dramatic impact on the field of ophthalmology. The majority of AIDS patients have ocular manifestations, most commonly a noninfectious microangiopathy called AIDS retinitis or cytomegalovirus (CMV) retinitis. About one-quarter of AIDS patients have compromised vision due to CMV activation. Untreated, CMV retinitis is progressive, resulting in ultimate destruction of the entire retina. Drugs such as ganciclovir and foscarnet have been effective in treating CMV retinitis and have helped AIDS patients preserve their vision in spite of the serious nature of their illness.

The two most common forms of cancer that affect the eye are retinoblastoma (RB) and choroidal melanoma. RB is mainly a disease of childhood. Thanks to the efforts of many vision scientists, RB is now one of the best understood of all solid tumors. Ninety percent of individuals who inherit specific mutations in the RB gene will develop the tumor. Each year, 300 to 400 new cases of RB are diagnosed in the United States. Unfortunately, the most prevalent treatment for RB at this time is surgical removal of the affected eye. Choroidal melanoma primarily affects adults and its etiology is poorly understood. Nearly 1,500 new cases of choroidal melanoma are diagnosed annually in the United States, and the optimal therapy for this disorder is still unclear.

The inherited retinal degenerations are typified by retinitis pigmentosa (RP), which results in the destruction of photoreceptor cells, the RPE, and choroid. This group of debilitating conditions affects approximately 100,000 people in the United States. Knowledge gained from studies of the structure, function, and metabolism of the normal retina and RPE have had a large impact on scientists' ability to understand what goes wrong in diseases like RP. A great deal of the progress made in dealing with this important clinical problem has depended on advances in research on photoreceptor cell biology, molecular biology, molecular genetics, and biochemistry over the past two decades. Animal models of hereditary retinal disease have been vital in helping unravel the specific genetic and biochemical defects that underlie abnormalities in human retinal diseases. It now seems clear that both genetic and clinical heterogeneity underlie many hereditary retinal diseases.

The leading cause of visual loss in the elderly is macular degeneration (MD). The social and economic impact of this disease in the United States is increasing. The macula is a structure near the center of the retina that contains the fovea. This specialized portion of the retina is responsible for the high-resolution vision that permits activities such as reading. The loss of central vision in MD is devastating. Degenerative changes to the macula (maculopathy) can occur at almost any time in life but are much more prevalent with advancing age. With growth in the aged population, age-related macular degeneration (AMD) will become a more prevalent cause of blindness than both diabetic retinopathy and glaucoma combined. Laser treatment has been shown to reduce the risk of extensive macular scarring from the "wet" or neovascular form of the disease. The effects of this treatment are short-lived, however, due to recurrent choroidal neovascularization. Recently, important new strides have been made in understanding the molecular basis of several forms of MD. Sorsby's fundus dystrophy, a rare form of MD, has been found to respond to vitamin A supplementation. There are currently no effective treatments for the vast majority of MD patients.

Retinal detachments are important clinically and can be divided into three basic types: (1) rhegmatogenous detachment, in which a tear or hole occurs in the retina; (2) serous detachment, resulting from subretinal fluid collection; and (3) traction retinal detachment (TRD), in which the retina is pulled away from the RPE by contractile tissue in the vitreous body or on the retinal surface itself. Rhegmatogenous detachments are the most common and the greatest threat to vision. TRD usually occurs in proliferative retinopathies, most commonly proliferative diabetic retinopathy or proliferative vitreoretinopathy (PVR). Although significant advances have been made in the management of retinal detachments, they remain an important cause of visual morbidity.

One of the major achievements in all of biology has been defining cellular events involved in the process of visual transduction. This has become a classic model that has led the way toward researchers' current understanding of signal processing in other systems. Advances in understanding visual biochemistry have yielded important new insights into the causes of retinal diseases. The majority of these diseases affect photoreceptors and the RPE. For some inherited retinal diseases the affected gene and protein have been identified. Scientists are beginning to understand the early effects on photoreceptors of proteins that are abnormal or that have lost their function. However, they have yet to make the definitive connection between the abnormal function of individual proteins and the death of photoreceptor cells.

The visual images that fall on the retina are sent to the brain to be decoded and interpreted. The visual images that are perceived result from integration of electrical impulses generated within the retina that are in turn transmitted by ganglion cells via the optic nerve to the part of the brain called the visual cortex. Remarkably, the visual system can function over a hundred-millionfold range in illumination. Individual photoreceptor cells are responsive to light intensities varying between sunlight and candlelight. Yet when the lights are turned off, the perception of darkness is nearly instantaneous. What makes this possible? The tools of modern neurobiology offer the potential to understand the cellular mechanisms of both light adaptation (sensitivity to varying light levels) as well as inactivation (turning off the sensitivity to light). At birth, our eyes have their full complement of 300 million retinal cells, with all 10 billion synaptic contacts already established. Most of the known neurotransmitters and neuropeptides involved in cell-cell communications are represented in the retina. A central unanswered question in neurobiology is how this complex network permits the formation of images and the discrimination of colors.

In Fiscal Year 1997, the National Eye Institute (NEI) funded approximately 600 extramural research projects in the Retinal Diseases Program at a total cost of $128,316,000.

The goals for laboratory and clinical research conducted within the Retinal Diseases Program for the next 5 years are to:

• Understand the molecular and biochemical basis for the different forms of MD, improve early diagnosis, characterize environmental effects on the etiology of MD, and develop new treatments.
  
  
• Understand the pathogenesis of diabetic retinopathy and other vascular diseases of the retina and develop strategies for primary prevention and improved treatment.
  
  
• Identify the genes involved in retinal degenerative diseases, including RP, and determine the pathophysiological mechanisms underlying these mutations.
  
  
• Explore new potential therapeutic strategies for inherited retinal diseases, such as gene transfer, tissue and cell transplantation, growth factor therapy, and pharmacological intervention.
  
  
• Establish the causes and etiology of uveitis and improve methods for its diagnosis, therapy, and prevention.
  
  
• Use both molecular and physiological approaches to study light adaptation in photoreceptors, with particular emphasis on the visual cycle.
  
  
• Build on knowledge gained from retinal neuroscience to understand how retinal networks process visual images, a central unanswered question of modern neurobiology.
  
  

ASSESSMENT OF PROGRESS

Within the Retinal Diseases Program, significant progress has been made in the last 5 years in understanding the fundamental and pathogenic processes in the retina and in improving the diagnosis and treatment of a variety of retinal diseases.

Epidemiological, cell biological, and molecular genetic studies of MD. In the past decade, it has become increasingly apparent that the clinical entity known as AMD is actually a heterogeneous group of disorders with multiple pathophysiological mechanisms. Recent epidemiological, cell biological, and genetic studies of MD have incorporated standardized, detailed classification schemes into their study design. The recent development of an international classification and grading system for AMD should allow improved comparisons to be made across different studies.

Recent epidemiologic studies have identified factors, such as cigarette smoking, that increase the risk of AMD. Conversely, antioxidants have been suggested to decrease the risk, raising the possibility that a combination of treatments and behavioral changes could further decrease the risk of AMD (see Development of effective treatments for retinal diseases). Encouraging results have also emerged from studies that suggest a protective effect of antioxidants.

Progress has also been made in the area of the molecular genetics of various types of MD. Since 1992, the genes for a number of different forms of heritable macular disease have been mapped to specific chromosomes. These include: North Carolina macular dystrophy, Best's disease, Stargardt's disease (both dominant and recessive forms), pattern dystrophy, Sorsby's fundus dystrophy, bifocal chorioretinal atrophy, and autosomal dominant radial drusen (Malattia Leventinese). In three cases (pattern dystrophy, Sorsby's fundus dystrophy, and recessive Stargardt's disease), the mutated genes (RDS, TIMP-3, and ABCR, respectively) have actually been identified. An animal model for the RDS mutation is already available for experiments aimed at improved understanding of the associated pathophysiology, as well as experiments designed to evaluate novel treatment methods with potential application to human disease. Recently, it has been found that the ABCR gene mutated in Stargardt's disease appears to be altered in some patients with AMD. Studies are ongoing to determine the prevalence of ABCR gene mutations in large cohorts of patients with AMD. Detection of genes mutated in patients with AMD will permit the development of genetic tests that may identify individuals at risk for the disease.

Given the high prevalence of AMD in the older population, it is hoped that inexpensive, safe medications can be identified that will be efficacious for the prevention and treatment of common forms of MD. Severe visual loss in patients with MD is associated with the development of neovascularization. Thus, most treatment trials have been aimed at treating this complication. While the results of the macular photocoagulation study showed some promise regarding laser photocoagulation for treatment of certain patients with neovascularization, the benefits of this treatment are limited. Other treatments are also being explored, but to date none have been found to be effective for the majority of these patients.

Pathogenesis of vascular diseases of the retina. Vascular endothelial growth factor (VEGF) has become a leading candidate for the long-sought agent responsible for neovascularization in retinal diseases. Retinal neovascularization is often associated with retinal ischemia and hypoxia. Hypoxia induces VEGF production. VEGF is present at high concentrations in the vitreous fluid of patients with proliferative diabetic retinopathy and is low to absent in the vitreous of patients with nonvasoproliferative disease. VEGF levels are high in the retina and vitreous of animals with experimental retinal or iris neovascularization, and methods that block VEGF action (e.g., neutralizing antibodies, soluble receptors, or antisense DNA) prevent neovascularization. In human eyes with retinal and choroidal vascular diseases, and in experimental animals, VEGF is localized primarily in the glial cells of the retina and optic nerve, and in the RPE cells. Although hypoxia has not been identified in choroidal neovascular diseases, VEGF has been reported in the RPE cells of choroidal neovascular membranes. While macular edema and neovascularization apparently result when VEGF is upregulated during certain pathologic processes, the normal function of VEGF may be to stimulate blood vessel growth in fetal development. Mice with a targeted disruption of the VEGF gene die in embryo due to defective vascular development.

An enzyme that may be critical for the development of diabetic retinopathy is aldose reductase, the initial enzyme of the sorbitol pathway. Although extensive laboratory studies have shown that aldose reductase inhibitors (ARIs) can prevent the development of diabetic retinopathy in experimental animal models, clinical studies using ARIs have not shown significant impact on the development or progression of retinopathy. However, a recent result involving the effect of a potent new ARI in the retinas of experimental animals may be cause for renewed optimism. A new and highly potent ARI that inhibits aldose reductase by approximately 90 percent can prevent VEGF expression in long-term galactosemic rats. It appears that, to be effective, ARIs must penetrate the retina sufficiently to inhibit nearly all of the aldose reductase present. Currently available ARIs have not been able to achieve this level of penetration at safe doses, but newer drugs appear promising.

In diabetic retinopathy, glucose may exert its deleterious effects by directly modifying the expression of genes. Cultured retinal pericytes grown in high glucose show differences in gene expression when compared to cells grown in normal glucose. Basement membranes of blood vessels from diabetic or galactosemic animals contain a profile of collagens different than basement membranes of control animals, suggesting altered expression of genes. Similarly, when animals in poor diabetic "control" or those maintained on high galactose diets for a short time are switched to "tight" control or a normal diet, they develop retinopathy after a delay of several years. The diabetes control and complications trial and its followup showed that the delay of onset and possible prevention of diabetic eye disease was due to tight control of glycemic levels. This study has made a significant contribution to patient welfare and quality of life.

Data from the multicenter Cryotherapy for Retinopathy of Prematurity Study showed that the incidence of severe retinopathy of prematurity (ROP) in very low birth weight premature infants is about double in Caucasian children when compared to African-American children. The reasons for this disparity may include genetic differences and the effects of increased ocular pigmentation in preventing oxidative damage to the retina.

A role for pituitary-associated factor in diabetic retinopathy has been appreciated for many years. Several decades ago, retinal neovascularization was found to regress after pituitary ablation in diabetic patients that appeared to be related to postsurgical growth hormone (GH) deficiency. In addition, insulin-like growth factor-1 (IGF-1) appears to be associated with proliferative retinopathy. To study the role of GH and IGF-1 in ischemia-induced retinal neovascularization and its interaction with VEGF, transgenic mice were studied. It was found that systemic inhibition of GH, IGF-1, or both may have therapeutic potential in preventing some forms of retinopathy.

Morphological, cellular, and molecular events that accompany retinal wound healing. The retina is composed of a complex and elaborate network of neuronal and nonneuronal cells. Retinal tissue can be exposed to a range of insults from a variety of diseases, including PVR, diabetic retinopathy, maculopathy, AMD, and RP. Physical trauma, environmental insults, and even surgery itself can damage the retina. These insults can often result in a specific biological response to injury called wound healing, which in the case of PVR is characterized by macro-phage infiltration, RPE cell transformation, and the deposition of extracellular matrix components. The attempt by the retina to repair damage after injury can result in fibrosis, scar formation, neovascularization, and proliferation of glial cells(gliosis) within the retina. As such, retinal wound healing is often a fundamentally undesirable event with respect to the visual consequences.

Regardless of the cause of the initial damage, retinal wound healing shares important features with wound healing in other tissues. However, because of the retina's unique character as a highly differentiated and specialized tissue, it is not equipped to repair itself in such a way as to restore its own specialized properties. The result is activation of a wound repair process that is incompatible with the retina's function as a neurosensory tissue. Attempts at repair are an ordered process involving complex cell-cell interactions, but ultimately can result in partial or complete loss of vision.

A variety of overlapping biological processes, including abnormal cellular migration, cellularproliferation, cellular transformation, cellular death, invasion, inflammation, fibrosis, gliosis, scarring, and neovascularization, are associated with the retina's attempt to repair itself. Macrophages and other leukocytes are typically attracted to the site of initial damage, playing a crucial role as regulators of cellular migration and growth, stimulating additional cellular activities, and modulating local cellular activities. Macrophages also produce a variety of biologic response modifiers, most notably fibroblast growth factor (FGF), transforming growth factor (TGF), and platelet-derived growth factor (PDGF).

The exact role of retinal cells, RPE cells, and other cells recruited to the wound site is unclear. There is some indication that RPE cells are one of the sources of regulatory factors involved in the process of wound healing. Studies have shown that a vast number of growth factors, cytokines, and their receptors are secreted or expressed by the RPE, although the precise mechanisms that regulate and modulate their expression remains to be determined. A significant finding related to the potential role of the RPE in wound healing has been the demonstration that there are multiple, self-regulatory, or autocrine loops in RPE cells. Both PDGF and VEGF autocrine loops have been demonstrated. Upregulation of PDGF and its receptors in wounded RPE cells has been documented in vitro. The autocrine loop is also modulated in vivo and appears to be a potent stimulus for upregulation of PDGF by RPE cells following retinal detachment. Photoreceptor cells or other cells in the retina may play a role in the modulation of these autocrine loops, which is consistent with the finding that the RPE proliferates and becomes more responsive to growth factors following injuries such as retinal detachment. Perhaps similar modulation occurs in the VEGF autocrine loop since VEGF, PDGF, and their receptors can be detected when RPE cells participate in wound repair, as occurs during formation of epiretinal membranes. A better understanding of these regulatory pathways may be useful in developing therapies for controlling the proliferation of RPE and other cells during wound healing.

Genetic etiology of RP and allied diseases. Since 1990, 10 genes causing RP have been identified (rhodopsin, RDS, α- and β-subunits of rod cGMP-phosphodiesterase, ROM1, α-subunit of the rod cGMP-gated channel, RP GTPase regulator, cellular retinaldehydebinding protein, myosin VIIA, and PE65). Furthermore, at least 24 additional loci causing RP have been placed on the human genome map and are in varying stages of being identified through positional cloning strategies. Progress is being made in the elucidation of genes causing other hereditary diseases of the retina. These diseases include congenital stationary night blindness (some forms of which are caused by mutations in the genes encoding rhodopsin and the β-subunit of rod cGMP-phosphodiesterase), Oguchi disease (arrestin and rhodopsin kinase), and juvenile hereditary MDs (TIMP-3 and ABCR, a rod photoreceptor-specific protein). Finally, numerous candidate genes (narrowly defined here as genes with known function in the retina and specifically expressed by the retina) have been isolated and are available for chromosomal mapping (to determine if they map to chromosomal regions with known retinal degeneration loci) or for direct mutation analysis in large cohorts of patients with forms of hereditary retinal degeneration, dysfunction, or developmental anomalies.

With regard to RP, each identified gene accounts for a few percent of cases, and rhodopsin, which accounts for about 10 percent of cases. In total, all identified genes to date account only for about one-third of cases. It is uncertain at the present time what proportion of cases are due to defects in the 24 or more unidentified genes known only through linkage studies in one or a few families each. It is possible that there may be 50 to 100 or more human loci that can confer hereditary retinal disease phenotypes. This estimate is consistent with genetic mapping and cloning studies in Drosophila, which have provided firm evidence for well over 80 retinal degeneration and dysfunction loci. Therefore, there is little reason to doubt that the set of genes causing retinal degeneration in humans is of comparable size.

Gene identifications in humans have allowed scientists to identify or create animal models with retinal degeneration due to defects in the same gene homologs. In particular, the genes responsible for retinal degeneration in the naturally arising RD and RDS mouse strains (the β-subunit of phosphodiesterase and peripherin/RDS, respectively) are now known to be causes of RP and other retinal degenerations in humans. Transgenic mice, rats, and pigs expressing dominant rhodopsin mutations have been developed, as have transgenic mice with dominant and digenic RDS and ROM1 mutations. These animal models are already the subject of intensive study to determine the pathophysiological mechanisms whereby these gene defects lead to photoreceptor degeneration and hopefully will lead to pilot studies of novel therapies for retinal degenerative diseases.

Developing Drosophila as a model for studying hereditary human diseases that cause progressive retinal degeneration such as RP has provided exciting information. The availability of eye mutants in Drosophila provides models for dissecting the molecular bases of retinal degeneration. The relevance of this work for humans with RP and allied diseases is illustrated by the fact that mutations in the human homologs of some of the Drosophila genes, such as rhodopsin and arrestin, have been found to cause RP or stationary night blindness. It is likely that future evaluation of the human homologs of other Drosophila retinal degeneration genes will lead to the identification of novel causes of human hereditary retinal diseases.

The identification of the genes causing some forms of retinal degeneration and retinal dysfunction has permitted advances in understanding the corresponding visual dysfunctions. Studies of dominant rhodopsin mutations using in vitro, transgenic mouse, and transgenic Drosophila systems, for example, have provided evidence suggesting that, at least for some mutations, the transport of mutant rhodopsin to the rod outer segments is defective. Null mutations in the genes encoding the catalytic subunits of rod phosphodiesterase likely are deleterious because of a secondary increase in the cytoplasmic level of cGMP. The proteins RDS and ROM1 form heterotetramers that accumulate at the rim regions of outer segment discs and perhaps have a key structural role there; without this complex, outer segment discs do not form properly. Regardless of the underlying genetic defect and the early biochemical steps leading to photoreceptor degeneration, the final common pathway for cell death has been found to be apoptosis.

Development of effective treatments for retinal diseases. During the past decade, important insights have been gained into the molecular etiologies of several inherited retinal and macular dystrophies. Studies from many laboratories have defined several promising therapeutic strategies. Progress has been reported in the area of somatic gene therapy with recombinant adenoviral and adenoviral-associated viral vectors. Several laboratories have shown significant slowing of photoreceptor degeneration in several animal models with the administration of basic FGF and neurotrophic agents. Another promising strategy involves directly targeting the mutant gene or its mRNA product while leaving the normal allele unaltered. Current strategies include ribozymes, antisense nucleic acids, and triplex-forming oligonucleotides. Of the three, the mode of action for ribozymes is best understood. Specific cleavage of single-base substituted mutant mRNAs without degradation of the wild-type mRNA has been demonstrated in vitro.

Some advances have been made in the area of retinal transplantation. Cells of the RPE have been grafted into the subretinal space of Royal College of Surgeons (RCS) rat eyes, prolonging the survival of photoreceptors. Embryonic mouse retinal tissue has been grafted into brains of adult mice. These grafts sent out neurites that established synaptic contacts with neurons of the recipient's brain and developed recognizable retinal cell types.

New knowledge about the molecular defects underlying inherited blindness in animals and humans has shown that the retinal degenerations are a highly heterogeneous group of diseases. It is likely that photoreceptors die for different reasons in the different mutants. A rhodopsin mutation resulting in a trafficking defect, for example, may represent a completely distinct cellular lesion from a mutation that causes constitutive activation of the phototransduction cascade. Multiple animal models involving spontaneous mutations in retinal genes have been defined. In addition, several transgenic models of both recessive and dominant human inherited retinal dystrophies have been generated.

A double-masked clinical trial of about 600 patients with RP found that oral vitamin A supplementation slowed the course of retinal degeneration, as measured by electroretinogram (ERG), and that vitamin E hastened the course of degeneration. Reduction in light exposure appears to modulate the severity of disease in some but not all mouse models of RP; a large-scale clinical trial of the possible therapeutic benefit of reduction of light exposure (e.g., sunglasses) has not yet been carried out.

Unique properties of intraocular immunity and inflammation. The intraocular spaces in which immunity and inflammation can occur include the anterior (and posterior) chamber(s), the vitreous cavity, and the subretinal space. Recent evidence indicates that the phenomenon of immunologic privilege exists in each of these compartments because allogeneic grafts of tumors and of retinal tissues (neuronal retina, RPE) placed within these sites have been seen to display prolonged (even indefinite) survival. However, observations regarding the subretinal space are largely qualitative and require further refinement. For technical reasons, much more has been learned about modulation of immunity and inflammation within the anterior chamber than within the other compartments. In that regard, an ever-expanding list of immunosuppressive and anti-inflammatory factors has been generated by examining aqueous humor. These factors include TGF-β2, α-melanocyte stimulating hormone, vasoactive intestinal peptide, calcitonin gene related peptide, macrophage migration inhibitory factor (MIF), and soluble inhibitors of complement activation. In addition, a natural defensin that has been demonstrated to be an antibiotic for gram-positive cocci has been described in aqueous humor. Along with these soluble immunomodulatory factors, the cells that surround the anterior chamber and the subretinal space constitutively express membrane bound forms of Fas ligand and the complement inhibitors, CD59, DAF, and CD46. Moreover, cultured epithelial cells of the iris and ciliary body, as well as RPE, have been found constitutively to secrete immunosuppressive factors in vitro.

Anti-inflammatory and immunomodulatory factors in the ocular microenvironment affect two phenomena: the existence of intraocular immunologic privilege and the induction of systemic immune deviation or anterior chamber associated immune deviation (ACAID) to intraocular antigens, and the ability of the eye to resist blinding intraocular inflammation in response to injury and infection. The mechanisms by which specific factors in the ocular microenvironment induce ACAID are beginning to be elucidated.

The multiplicity of immunomodulatory factors in aqueous humor has important effects on immune and inflammatory effector mechanisms. TGF-β2, as well as other factors in the aqueous humor, inhibit macrophage activation and prevent these cells from acquiring certain effector function (cytotoxicity,generation of oxygen free radicals, and nitric oxide). MIF is a powerful inhibitor of natural killer cells in the eye. Moreover, factors in aqueous humor also have profound effects on certain T-lymphocytes. Primed T-cells are inhibited from proliferating in vitro in response to antigen or T-cell receptor ligand stimulation, and the cells are also prevented from secreting proinflammatory cytokines such as γ-interferon. But not all immunosuppression in the eye depends on soluble molecules in the fluid phase. The recent dramatic demonstration of constitutive expression of Fas ligand on numerous cell types within the eye indicates that a mechanism of peripheral deletion of Fas+T-lymphocytes (via apoptosis) operates to limit intraocular immune expression. Recent studies confirm that intraocular expression of Fas ligand accounts, at least in part, for the extraordinary success of orthotopic corneal allografts in mice, and there is evidence suggesting that Fas ligand also contributes to the generation of the ACAID-inducing signal.

Intraocular inflammation of the posterior segment. Posterior segment intraocular inflammation is an important cause of blindness. The inflammation can display one or more of four distinct clinical features: variably sized focal chorioretinal infiltrate of inflammatory cells; retinal vessel inflammation; vitreous cellular infiltrates; and edema of the macula, optic nerve head, or the entire retina, producing a subretinal posterior segment inflammation. When an infectious agent is identified, the cause is straight-forward and appropriate therapy can be instituted. But in many instances, no such agent can be found, and attempts to sort out unknown infectious causes from immunopathogenic processes, and these in turn from autoimmune mechanisms, are usually unsuccessful.

For many years, the laboratory model of choice has been experimental autoimmune uveoretinitis (EAU), evoked in laboratory animals by immunization with one of several autoantigens produced by the adult retina (arrestin, interphotoreceptor retinol binding protein [IRBP], rhodopsin). In certain genetically defined strains of mice (B10.A), immunization with retinal S-antigen (arrestin) produces a disease similar to human ocular histoplasmosis, whereas in rats, similar immunization leads to an exudative retinal detachment resembling Vogt-Koyanagi-Harada disease. Arrestin immunization of monkeys produces a retinal vasculitis. This set of observations makes the following points: the posterior segment of the eye responds to insults, whether infectious, autoimmune, or immunopathogenic, in a limited set of patterns, and these patterns are dictated in part by the host's genetic makeup, the nature of the insult, and the status of the host's immunologic history.

Immunodominant peptides derived from retinal autoantigens have been identified for arrestin andfor IRBP in genetically defined experimental animals. CD4 positive T-cells have emerged as the primary initiators of intraocular inflammation in EAU, and γ-interferon is believed to be the major proinflammatory cytokine involved. To that end, transgenic mice have been produced in which the γ-interferon gene has been inserted under the control of the rhodopsin promoter, and the eyes of these mice develop a blinding posterior segment inflammation. However, confusion about the role of γ-interferon has emerged as investigators have discovered that exogenous administration of γ-interferon reduces the incidence and severity of EAU in experimental animals, and neutralizing antibodies against this cytokine make the experimental disease more intense. Similarly, exogenous interleukin-12 (IL-12), the cytokine most critical to activation of γ-interferon-producing CD4 positive T-cells, also mitigates against intense uveoretinitis. At the same time, inducing oral tolerance by feeding retinal antigens to experimental animals has proven to prevent or greatly diminish the severity of EAU, and a recent report from a pilot human clinical trial suggests the same conclusion. Similarly, induction of ACAID with retinal antigens has been shown to prevent EAU and reduce the severity of extent disease.

By switching from retinal antigens to molecules associated with melanin pigment in the uveal tract, a new model of ocular inflammation has been produced, called experimental anterior autoimmune uveitis (EAAU). The pertinent antigen appears to be a protein or proteins associated with the insoluble fraction of melanin granules, and the pattern of disease caused by immunization with this antigen differs from EAU. The inflammation of EAAU is located primarily within the anterior segment of the eye, emphasizing the fact that the localization of autoimmune-based intraocular inflammation depends, in part, on the nature (and presumably the source) of the antigen within the eye.

Bacterial lipopolysaccharide has recently been exploited as an experimental inducer of endotoxin-induced uveitis in mice and rats, and this newer model has considerably enhanced researchers' understanding of ocular inflammation due to immunopathogenic, rather than autoimmune, processes. The vessels of the uveal tract are uniquely susceptible to mediators released upon endotoxin stimulation, especially IL-1, IL-6, and tumor necrosis factor-α. As a consequence, the uveal tract itself becomes inflamed and infiltrated with blood-borne cells; the blood-ocular barrier is breached.

One of the most severe forms of posteriorsegment inflammation is caused by CMV, a common opportunistic infection in patients with AIDS. CMV retinitis accounts for about 85 percent of CMV disease in patients with AIDS and is a significant cause of morbidity. CMV retinitis is a progressive and destructive infection that, left untreated, causes blindness. The NEI-sponsored clinical trial called Studies of the Ocular Complications of AIDS (SOCA) has demonstrated that for AIDS patients with CMV retinitis a combination therapy with foscarnet and ganciclovir is more effective than either of these drugs alone in controlling CMV retinitis. However, there can be complications of intravitreal therapy injection, including endophthalmitis and retinal detachment. A recent advance has been the development of a sustained-release ganciclovir device that is surgically implanted into the vitreous cavity and releases drug over several months. An NEI intramural clinical trial demonstrated a prolonged delay in progression of retinitis for patients receiving the implant when compared to controls.

The molecular basis of visual transduction pathway. Proteins in rod and cone photoreceptor cells that are responsible for the capture of light and its conversion into electrical signals have been identified and characterized in considerable detail. The genes for the photoreceptor proteins, rhodopsin and cone opsins, from a variety of species including human, have been cloned and sequenced. This information has been used with biochemical and biophysical studies to identify specific amino acids and structural regions of these proteins that play key roles in spectral sensitivity and color discrimination, initiation and termination of the photoresponse, and protein folding and stability. This has led to the development of mathematical models that accurately simulate the photoexcitation phase of the visual response in rod cells.

A high-resolution, three-dimensional structure of transducin obtained by X-ray crystallography has been used to understand the mechanism by which guanosine diphosphate bound to transducin is exchanged for guanosine triphosphate following photoactivation of rhodopsin; it has also provided a glimpse of how transducin subsequently interacts with phosphodiesterase to catalyze the hydrolysis of cGMP. The end target of the visual cascade pathway, the cGMP-gated channel, has also been characterized at molecular and physiological levels. These studies have provided important new insight into how the activity of this channel is controlled by cGMP and extracellular divalent cations and modulated by calmodulin. Considerable progress has also been made toward analyzing the molecular structure, function, and regulation of proteins that play central roles in the termination of the photoresponse and the recovery of the rod cell to its dark-adapted state. These proteins include rhodopsin kinase, arrestin, guanylate cyclase, and several key regulatory calcium-binding proteins. A particularly significant development has been the identification and characterization of the guanylate cyclase activating proteins. These proteins have been shown to regulate the activity of guanylate cyclase activity in a calcium-dependent manner and play a crucial role in photorecovery and light adaptation in photoreceptor cells. Studies have also led to a more complete understanding of the function of the Na/Ca exchanger in maintaining calcium homeostasis in rod cells and the role of glucose metabolism in the production of energy for phototransduction. Disruption of metabolic function and homeostasis in the retina may have serious consequences and contribute to photoreceptor degeneration.

Signal processing in the retina. The retina is a biological image processor. It receives light from images in the world, transforms each image into electrical signals, codes these signals by extracting the essential information from the image, and finally sends this code as patterns of spikes down the optic nerve to the brain. Normal vision requires the retina to operate over illumination conditions that span 10 orders of magnitude (10 billion to one), a feat well beyond the capabilities of any of the most sophisticated optical instruments. During the last 5 years, researchers have gained insight into many areas related to how the retina is "wired" to function and how it is able to adapt to whatever the ambient illumination might be. This progress has been made possible by applying techniques that look at the activity of many cells simultaneously, by using ever more sophisticated molecular and cellular approaches, and by adopting techniques perfected in nonmammalian retinal studies to study mammalian and primate retinal preparations.

By recording the spike activity of many ganglion cells simultaneously, using arrays of electrodes and powerful computer-based analysis, it has been possible to discover the correlated firing of many cells. This suggests that visual signals transmitted to the brain do not simply reflect the firing in single axons but in combinations of axons. These multielectrode studies have also shown that waves of electrical activity course across the retinas of mammals and reptiles during postnatal development. A great deal has been learned of the synaptic transmitter molecules and the network on cell-to-cell connections that generates these waves. These waves are believed to be important for generating synchronized signals to permit the proper "wiring" of higher visual centers in the brain. Numerous important insights into normal and pathophysiological function of the photoreceptors and other retinal cells have been gained from sophisticated analysis of ERG recordings, the massed electrical response of the retinal neurons that is routinely measured from the cornea of patients.

Over the last 5 years, it has become increasingly apparent that cells of all types in the retina can be electrically coupled to similar or different cell types. Studies of the coupling of fluorescent dyes between these cells show conclusively that not only do the size and nature of the gap junction proteins differ, but that these coupling junctions can be regulated by different "neuromodulator" chemicals that may diffuse through the retina in response to different lighting conditions. Although the synaptic neurotransmitters released by each cell class is close to being identified, it is becoming increasingly evident that many different isoforms of the receptors to these neurotransmitters are expressed on the postsynaptic cells. Each may serve a different function. Moreover, postsynaptic receptors may gate an ion channel or activate intracellular signaling pathways within the same cell. Thus, one neurotransmitter can "drive" a cell in many ways, some of them through rapid activating pathways and some through slower, modulatory pathways. These slower pathways probably serve to adapt cellular function to fit ambient light conditions.

The newly discovered information about the control of neurotransmitter from the ribbon synapses of bipolar and photoreceptor cells has become the exemplar model of how glutamate is released from most neurons in the central nervous system. These studies have been made feasible by newly developed techniques that measure the release of endogenous glutamate from either a specially prepared sheet of photoreceptors or from single cells and/or that measure exocytosis employing electrophysiologically based capacitance measurements. Researchers now know much more about how calcium influx controls glutamate release and how there are "pools" of readily releasable glutamate-containing vesicles that can meet the needs of very fast signaling and "pools" of more slowly releasable vesicles that supply transmitter over longer periods of time. Exquisite correlations have been made between the biophysically measured release rates and the quantity and positioning of neurotransmitter-laden vesicles at the ribbon synapses.

Microelectrode studies formerly restricted to retinas of cold-blooded species are now enjoying success in revealing fundamental concepts of how mammalian and primate retinas function. With these microelectrodes it is possible to record the responses of individual cells to light or neurotransmitters, and then inject dyes into the same cells to illuminate their morphology and connections to other cells. It is becoming increasingly apparent that many of the neurotransmitter receptors are expressed in similar cell types in all vertebrate species. Dye injections viewed with advanced microscopy reveals that the microcircuitry of rabbits, rats, or cats is very similar to that of monkeys and humans. Surprisingly, however, the processing of color information by primate retinas is very distinctive from that of reptilian and fish retinas.

The former promise of molecular biology has proven itself extremely useful in helping to dissect the "wiring" of the retina. Two of the most important retinaspecific neurotransmitter receptors, the mGluR6 and the GABAc, have been cloned. Structure function studies on these proteins are just beginning. A mouse line lacking the mGluR6 receptor has been made and used to verify the role of the rod ON bipolar cell in the conduction of low-light level, rod-driven responses to the ganglion cells. In other studies, transgenic mice have been created with cell type-specific markers. One use of this technique has been to use intense light to photoablate these specific cells and thereby "drop" these cells from a functioning retinal circuit. This has allowed investigators to unravel roles of individual cell types without having to deal with some of the complexities inherent in pharmacological methods.

Retinal development: establishing and specifying various retinal cell types. Cell lineage analysis has shown that retinal cells are generated from multipotent progenitors throughout development. The cell types generated in vitro can be influenced by the environment, and certain growth factors added to retinal cell cultures can lead to shifts in the types of cells produced. Growth factors can also influence the survival of retinal cells in vitro or in vivo. For example, members of the ciliary neurotrophic factor family of cytokines can reduce the number of rods formed early in development, but prevent their degeneration later in development. FGFs have a number of effects, including effects on cell division, rod development, and survival. In addition to the effect of extrinsic cues, intrinsic properties of progenitors contribute to the genesis of retinal cell types as well. Factors that affect the development of retinal cells may also affect their survival and function and have implications for developing effective treatments for retinal degenerative diseases such as RP.

One aspect of the progress in retinal development that should be noted is the value of model systems like Drosophila, Xenopus, and zebrafish to understanding vertebrate systems. The overall strategy of retinal cell fate determination and differentiation is the same in the vertebrate and Drosophila retina, as are many of the molecules employed. For example, the extracellular receptor Notch and its ligand Delta are crucial for the development of all retinal neurons in Drosophila and vertebrates. The basic helix-loop-helix genes are important for both, as are homeobox genes. The eyeless gene encodes a Drosophila Pax-6 and specifies eyes in Drosophila. In addition, vertebrate Pax-6 molecules can specify ectopic eyes in Drosophila, providing strong support for similarities in the development of visual systems. Although the true homology of the systems is still being debated, there is no doubt that the work that has been carried out in animal models has greatly contributed to and continues to enhance scientists' understanding of the development and function of many classes of genes.

Genes expressed in the retina and choroid. Major progress has been made in cataloging and mapping genes expressed in the retina and choroid with the advent of the field of genomics and the associated technology. Systematic sequencing of the entire genome of model organisms is underway, and several have been completed. Of particular relevance for understanding the human visual system is the human genome sequencing project. In addition to generating a genome sequence, a genetic map of various organisms has been made, using primarily polymorphic DNA markers and polymerase chain reaction technology. These databases will be invaluable in the work ahead and have already been a major force in changing the way that studies of disease and function have been conducted.

The goal of sequencing the mRNAs expressed in various tissues is also underway. These sequences, known as expressed sequence tags (ESTs), now allow recognition of which genes are expressed in various tissues and allow recognition of transcriptional units among genomic sequences. The "chip" technology that has recently been applied to molecular biology provides tools that will afford a more rapid accumulation of data concerning which cells express which ESTs and will provide for rapid sequencing of genes for diagnostic purposes.

With respect to progress on gene expression in the visual system, the accomplishments have been outstanding. Some of the genes specific to the visual system have been isolated (see Genetic etiology of RP and allied diseases). Biochemical isolation of proteins and the acquisition of their sequences, as well as the search for genes expressed in the visual system using homology to genes operative in other systems, have provided most of these genes. These genes have supplied a wealth of sequences for use in candidate gene approaches for the identification of mutations in various visual system diseases and for developing strategies for treatment.

Maintenance of a healthy neurosensory retina. The RPE performs a variety of transport functions that impact on the maintenance of a healthy neurosensory retina. Considerable amounts of lactate are formed by the photoreceptor cells as a product of anaerobic metabolism. The RPE removes this lactate from the neurosensory retina by transporting it into the choroidal circulation. Water movement is coupled to lactate transport via an H⁺/lactate cotransporter in the apical plasma membrane of the RPE. This system may account for a large fraction of the fluid absorption across the RPE. At light onset, the volume of the extracellular compartment surrounding the photoreceptors increases by 20 percent to 60 percent, much of which is due to changes in RPE transport. The subretinal K⁺ concentration decreases at light onset as a result of a change in photoreceptor activity, and this serves as the paracrine signal that triggers the transport of K⁺ and Cl^-, which are responsible for changes in subretinal volume. The K and Cl channels that mediate these changes have been characterized, as has their pharmacological regulation. The RPE secretes TIMP-3 into Brüch's membrane. Mutations in the TIMP-3 gene cause Sorsby's fundus dystrophy, an autosomal-dominant inherited disease with a phenotype similar to AMD.

The K⁺ channels of Müller cells have been shown to play important roles in regulating the concentration of K⁺ in the extracellular space. Müller cells possess neurotransmitter transporters that contribute to neurotransmitter reuptake, glutamate recycling, and protection of neurons against excitotoxicity. In addition, Müller cells have acid/base transport systems that play a role in regulating external pH within the retina.

The Xenopus (African clawed frog) neurosensory retina, reduced experimentally to a single neuronal population of rod and cone photoreceptor cells, displays a persistent rhythm in the synthesis and release of the circadian hormone melatonin and in the activity of N-acetyltransferase, the rate-limiting enzyme responsible for melatonin synthesis. In the isolated intact neurosensory retina, retinal melatonin rhythms are, in turn, controlled by dopamine, which is presumably synthesized and released by amacrine cells to interact with dopamine class 2 receptors located on the photoreceptors. Recently, it has been shown that a photoreceptor-specific protein called nocturnin is encoded by a circadian clock-regulated gene. Thus, with the Xenopus retina at least, a circadian clock resides within the photoreceptors. In addition, cyclic melatonin synthesis has been found in retinas of rats and hamsters.

Vision scientists continue to characterize the enzymes and transport proteins involved in the cycling and utilization of vitamin A (retinol) and its derivatives (retinoids) within the retina. This process is called the visual cycle. The retinoid 11-cis retinaldehyde is attached to a class of proteins called opsins. These opsins plus 11-cis retinaldehyde form the rod and cone photoreceptor visual pigments that trap light (photons) and initiate the visual process in the retina. The enzyme that produces 11-cis retinaldehyde from 11-cis retinol, called 11-cis retinol dehydrogenase, has now been characterized by recombinant DNA methods. This protein is found in the retinal pigment epithelium where 11-cis retinol is formed.

The characterization of proteins involved in the visual cycle has been useful for understanding the fundamental processes of retinol uptake, processing, and utilization in the retina, but it also serves the practical purpose of identifying mutations in the genes coding for these proteins. Very recently, mutations have been found in specific genes that encode for proteins involved in the visual cycle in retinal degenerative diseases.

PROGRAM OBJECTIVES

Program objectives for the next 5 years in the area of retinal diseases include both laboratory and applied research.

• Explore the pathophysiological heterogeneity of AMD to hasten development of the tools needed for improved diagnosis, prevention, and therapy.
  
  
• Investigate the pathogenesis of vasculardiseases of the retina and choroid, including diabetic retinopathy, AMD, and ROP; develop better methods of prevention and therapy.
  
  
• Identify novel causes of inherited retinal degenerations; further examine the cellular and molecular mechanisms whereby identified gene defects cause retinal degenerations.
  
  
• Further develop and critically evaluate therapies involving gene delivery, growth factors, and transplantation for the treatment of retinal disease.
  
  
• Explore the cellular and molecular basis of the response to retinal injury.
  
  
• Identify the factors that dictate the unique properties of intraocular immunity and inflammation and alter systemic immunity to intraocular antigens.
  
  
• Develop diagnostic methods and therapeutic approaches that distinguish among infectious, immunopathogenic, and autoimmune posterior segment inflammation.
  
  
• Analyze the mechanisms underlying light adaptation and recovery following phototransduction.
  
  
• Study how visual information is transformed by successive layers of the neural retina and the mechanisms involved.
  
  
• Identify and characterize factors important in retinal cell fate determination and differentiation.
  
  
• Catalog, map, and functionally characterize genes expressed in the retina and choroid and begin to determine the cellular sites of retinal gene expression in health and disease.
  
  
• Probe the control of the retina's microenvironment through studies of Brüch's membrane, the interphotoreceptor matrix, the RPE, glia, choroid, and vitreous.

The needs and opportunities related to each of these objectives and the strategies for accomplishing them will now be considered.

Objective 1: Explore the pathophysiological heterogeneity of AMD to hasten development of the tools needed for improved diagnosis,prevention, and therapy.

Research Needs and Opportunities

Progress has been made in the last 5 years in different areas of AMD. It will now be important to identify the cellular, molecular, and systemic factors that are involved in the pathophysiological cascade of AMD. This can best be accomplished with a combination of approaches including epidemiology, morphology, cell and molecular biology, and genetics. In the area of molecular genetics, studies need to be undertaken to identify genes responsible for late-onset AMD using a combination of linkage and candidate gene approaches. The availability of human donor eyes affected with AMD will be extremely valuable and will create a special opportunity to identify candidate molecules involved in the pathogenesis of AMD which, in turn, can be further evaluated by genetic, biochemical, and epidemiological approaches. Developing animal models of MD, including transgenic technology, of the more prevalent forms of human AMD, will provide the opportunity to elucidate the pathophysiological mechanisms involved in AMD and will allow researchers to develop methods for interrupting or mitigating the disease process. Therapies found to be most promising in the AMD animal models should be considered for human clinical trials. Some strategies may involve genetic detection of presymptomatic individuals followed by treatment designed to prevent or delay the onset of the disease. Additional epidemiologic studies are needed to investigate the association of AMD with other systemic diseases and potentially modifiable risk factors (e.g., diet and vitamin supplements).

Strategic Research Questions

What genes are responsible for significant proportions of typical late-onset AMD? There are a number of dystrophies that affect the macula, whose heritability is undisputed. Some of these are clinically similar enough to AMD that distinction from the latter is sometimes difficult in affected persons over age 50. But, as more families with a history of late-onset AMD are identified and analyzed for genetic mutations, the greater the likelihood that genes causing this disease will be discovered.

Can animal models of the more prevalent genetic forms of human AMD be identified or developed to test the efficacy of existing or novel therapies? Both naturally occurring and transgenic animals have played important roles in understanding many eye diseases. This will certainly be true for AMD.

Can systematic genotyping be accomplished to identify high-risk groups for early detection and treatment? A variety of pathogenic mechanisms are likely to be involved in AMD, and it is unlikely that any given treatment will be effective for all of them. It is essential to develop genotyping methods to reliably subdivide patients into pathophysiologically similar groups. This will allow identification of disease-causing mechanisms in presymptomatic individuals so that specific therapy can be administered at the earliest possible stages of the disease.

What is the interplay between genetic versus environmental factors in AMD development? AMD may be precipitated or exacerbated by cumulative damage from environmental factors such as light toxicity. But it seems clear that at some point in the process, genetic mechanisms come into play. At what stage this occurs and the precipitating events that are involved will not be clarified until more is known about the interaction between genetics and the environment.

Can effective treatment strategies be developed for the most common forms of AMD? Within families with AMD there are some patients who do relatively well clinically while others do poorly. This suggests that dietary, physical, or additional genetic factors are capable of modulating the effects of the disease gene. Such modulators have the potential to be the basis of effective therapy. If modulators of the disease can be discovered, they would be effective if they altered the rate of progression of the disease by only 15 percent, making a significant number of patients asymptomatic for their entire lives.

What factors or methods can retard the growth of choroidal neovascular membranes, and can this lead to effective treatment strategies? Growth factors such as VEGF are thought to be important modulators of neovascularization. If AMD leads to growth of new blood vessels, it is important to consider growth factors as targets for treatment strategies that can slow or prevent this form of the disease.

Are there modifiable risk factors for AMD to prevent or reduce the risk or progression? Some risk factors for AMD, such as smoking, are modifiable through changes in smoking habits. There may be other factors conferring risk, such as micronutrients, behavioral, and other environmental factors, which may be amenable to modification.

Objective 2: Investigate the pathogenesis of vascular diseases of the retina and choroid, including diabetic retinopathy, AMD, and ROP; develop better methods of prevention and therapy.

Research Needs and Opportunities

Recent advances have provided the identification and characterization of factors and proteins that may play a critical role in the management of diabetic retinopathy. There is a need to test new therapeutic approaches with potentially useful agents such as VEGF neutralizing agents, inhibitors of isoform of protein kinase C (PKC), aminoguanidine, and inhibitors of aldose reductase. Collaborations between National Institutes of Health investigators and the private sector should be encouraged. Since neuron and glial cells in the retina are primary sources of vasoactive compounds such as VEGF, it will be important to understand the metabolism of these cells in diabetes. To increase the pace of discovery of genetic factors involved in diabetic retinopathy, both molecular techniques and animals models need to be developed to allow study of genetic factors involved in the disease. It is important to identify key genes and as well as the mechanisms involved in hyperglycemia. Chronic hyperglycemia is the hallmark event for the development and progression of the disease, and hyperglycemia can act through its effect on genetically controlled mechanisms. The blood-retinal barrier is often compromised in the diabetic state; therefore, it is important to undertake molecular studies of the embryonic development of the blood-retinal barrier, the molecular mechanisms of its maintenance in adult life, and its breakdown in diseased states. Since oxidative processes may be involved in diabetic retinopathy, the measurement of toxic oxidation products in tissues and evaluation of antioxidant enzymes by direct enzyme assay of small tissue samples are needed. The preventive effects of antioxidant compounds on lesions putatively caused by toxic oxidation products need to be tested in experimental animals or in human clinical trials.

Strategic Research Questions

What pharmacological agents can be developed to prevent or cause the regression of retinal or choroidal neovascularization? Among the candidates that should be tested in the laboratory and, if appropriate, by controlled clinical trials are: VEGF, isoform of PKC, aminoguanidine, and ARIs.

What is the role of tissue hypoxia in VEGF upregulation and expression? The role of tissue oxygen could be tested by a new method using magnetic resonance imaging to evaluate quantitatively retinal oxygenation.

Do smaller amounts of VEGF lead to vascular leakage and macular edema, while a larger amount produces neovascularization? Is hypoxia involved in VEGF expression in choroidal neovascularization? VEGF may have different actions solely on the basis of the amount of factor present in the tissues.

Do toxic oxidation products play a role in the pathogenesis of retinal vascular diseases? If so, can antioxidants be used therapeutically to retard this pathogenesis? Measurements of toxic products of oxidation can be carried out in tissues, and levels of antioxidant enzymes can be evaluated by direct enzyme assay of small tissue samples.

Objective 3: Identify novel causes of inherited retinal degenerations; further examine the cellular and molecular mechanisms whereby identified gene defects cause retinal degenerations.

Research Needs and Opportunities

Great strides have been made in identifying genetic defects in inherited retinal degenerations. Efforts should continue toward identifying mutations that cause retinal degeneration or dysfunction in humans by evaluating genes encoding proteins in the phototransduction cascade and other retinal-specific pathways, including the visual cycle, positional cloning, and evaluating human homologs of genes found to cause retinal degeneration or dysfunction in animals. Since the molecular tools to localize and identify mutated genes for retinal degenerative diseases are available, the genetic defects in different forms of RP have been rapidly forthcoming. Research strategies must now actively be pursued to search for the molecular mechanisms of the pathophysiology of genetic mutations in human retinal degenerations. Specific animal models and in vitro systems with gene defects homologous to those known to cause human retinal degenerative diseases will be valuable in this pursuit. Since the final common pathway for cell death has been found to be apoptosis, continuing research of the mechanisms of cell death and their role in retinal degeneration may provide clues on whether the interference of the cell death pathway is therapeutic. In addition, implementing large-scale mutation-screening technologies to genotype large cohorts of patients will permit clinical studies aimed at finding shared clinical features of patients with similar genetic defects and potentially the evaluation of gene-specific treatments.

Strategic Research Questions

What is the pathophysiology of human retinal degenerative diseases? Many mutations have been discovered in photoreceptor genes, and these have been shown to be associated with RP. As a major component of rod photoreceptor cells, rhodopsin is easily the best studied retinal protein. There are over 90 mutations in rhodopsin, but information is needed on how these mutations actually cause cellular damage.

Can human diseases that are genetically or biochemically homologous to those found in animal models be identified? Animal models form the basis for understanding disease at the most basic level, but the animal model must mimic the human disease if it is to be useful.

Can novel methods be developed and evaluated to slow the progress of retinal degeneration in animal models, and can the promising therapies be evaluated in genetically characterized sets of patients? Neurotrophic factors have been shown to slow photoreceptor degeneration in animal studies, but the reason is unclear. Growth factor therapy may be one way of slowing the rate of degeneration in humans, but they will, in all probability, be most useful initially in genetically well-characterized patients.

What is the degree of clinical and genetic heterogeneity in different forms of RP and retinal degenerative diseases? Now that molecular genetics and molecular biology techniques are revealing the precise genetic mutations causing retinal degenerations, it is becoming clear that clinically distinct entities may overlap etiologically. Some diseases clinically categorized under a single heading are genetically heterogeneous. It is important for potential therapies to ascertain a patient's genetic status so that nutritional or pharmacologic therapy or transplantation will be most effective.

Objective 4: Further develop and critically evaluate therapies involving gene delivery, growth factors, and transplantation for the treatment of retinal disease.

Research Needs and Opportunities

Molecular etiologies of several inherited retinal and macular dystrophies have been discovered in the past decade. Studies from many laboratories have developed several promising therapeutic strategies. Neurotrophic factors are assuming increasing importance as potential therapeutic agents for retinal disease. Methods for their delivery and study of their efficacy in rescuing degenerative photoreceptor diseases need to be explored. Similarly, gene replacement therapy for retinal disease offers great potential for ameliorating the consequences of retinal degeneration. To make progress in this area, new methods and vector systems for gene delivery need to be developed. This will require delivery systems to have: reasonable efficiency, low cytotoxicity and immunogenicity, the ability to carry a DNA-insert of practical size, and long-term passenger-gene expression.

There are other cutting-edge approaches that may be potentially useful in treating retinal degeneration. These include studies of antiapoptotic proteins as potential therapeutic agents as well as ribozymes, antisense nucleic acids, and triplex-forming oligo-nucleotides. The latter should be studied for their in vivo efficacy using relevant animal models of human retinal disease.

Strategic Research Questions

Can combination treatments be explored that are effective in treating cell death in different forms of retinal degeneration? In several rodent animal models of RP, mutations in rhodopsin and other genes cause rapid rod cell death by apoptosis. Photoreceptors in animal models that overexpress bcl-2 have decreased apoptosis and survived environmental and genetic insults longer than cells with normal levels of bcl-2. Thus, bcl-2 appears to protect against retinal degeneration. FGF signaling appears to be associated with progressive photoreceptor degeneration, suggesting it may act as a survival factor.

What are the underlying molecular and cellular defects in retinal degenerative diseases, since understanding the basic mechanisms of retinal degeneration is critical to the development of effective therapy? Many mutations in many proteins of retinal cells have been identified and linked to photoreceptor degeneration. Most cases of RP are monotonic, but digenic inheritance of RP has been demonstrated. Apoptosis appears to be a final common pathway used by cells that are targeted to die.

Can researchers understand why the adult retina is incapable of regeneration following damage, injury, ischemia, or degenerative disease? Transplanted fetal rat retinal cells grafted into the subretinal space in the eye of light-blinded rats develop, differentiate as photoreceptors, and form synaptic contacts. In animals, when fetal RPE cells are transplanted into the subretinal space, they will survive and protect photoreceptors from degeneration.

How well do retinal transplants work (i.e., degree of visual function maintained or restored and length of survival)? Adult photoreceptors from mouse retinas can be isolated and injected into the subretinal space. These cells survive with normal-appearing synaptic terminals, but the outer segments degenerate. Other experiments have used slices of retinas or fetal retinal cells injected into the subretinal space in an effort to promote survival and rescue, all with mixed success, in part because the status of the outer segments is unclear. Fetal or neonatal transplants do lead to maintenance of photoreceptor outer segment structure, but they have an altered morphology, appearing as "rosettes" or centrally placed photoreceptors surrounded by the remaining retinal cells.

What immunologic issues govern transplant survival in the subretinal space (immune privilege and immunogenicity of transplantation antigens)? Virtually no effort has been made to understand the immune barriers to transplantation of allogeneic and xenogeneic tissues, especially those derived from the retina, into the posterior compartment of the eye. Animal studies aimed at understanding the immunogenetic and immunologic rules of transplantation into the posterior segment are important.

What are the effects of experimental and surgical manipulations and of disease on the immunological microenvironment of the graft site? The short- and long-term immunologic consequences of transplantation into the posterior segment of the eye are unknown.

Objective 5: Explore the cellular and molecular basis of the response to retinal injury.

Research Needs and Opportunities

The process of retinal wound healing has unique molecular and cellular properties. Retinal wound healing is an ordered, albeit undesirable, process involving complex cell-cell interactions. The cellular and molecular events associated with retinal wound healing need careful evaluation to identify the key processes that are involved. Extraretinal cells may influence the retinal wound response through an array of mitogenic, chemotactic, and trophic factors. It is important to know how they interact with their receptors, initiating the migration, proliferation, and phenotypic alteration of retinal cells. In this regard, researchers need to know more about the complex control mechanisms involved in the interaction of retinal growth factors with their receptors under normal conditions and during wound healing. Since mechanical wounding of retinal cells can promote the rescue and survival of photoreceptor and other retinal cells, this mechanism should be unraveled.

Because of the difficulty of studying the retinal wound healing process in patients, the development and availability of animal models will allow the determination of the sequence of molecular and cellular events. Animal models will permit the study of regulatory molecules and provide a means of evaluating new therapies to prevent the retinal wound healing response and neovascularization. Investigating possible genetic factors involved in these diseases, with particular attention to hereditary factors, may be useful.

Strategic Research Questions

What structural, functional, and molecular interactions exist between retinal cells, especially between glial cells and neurons, in the normal retina and in retinal wound healing response? In the ocular wound healing response, cells migrate and proliferate into the subretinal space, the retinal surface, and the vitreous cavity. These cells produce a collagen matrix and avascular membranes that contain a heterogeneous population of cells.

How is the normal retinal environment regulated, and how does this environment change following retinal injury and during retinal wound healing? Cytokines such as the interleukins, interferons, and the chemokines, as well as growth factors like basic FGF, TGF, and epidermal growth factor (EGF) are thought to play some role in wound healing events. There is increasing evidence that the normal process of wound healing, including cell proliferation, migration, collagen synthesis, arachidonic acid metabolism, and angiogenesis are coordinated by a complex array of cytokines.

Why and how do retinal injuries increase the survivability of photoreceptor cells and perhaps other cells in some instances? Studies have shown that injury itself is sufficient to increase the survival of retinal cells, but the basis for this phenomenon is unclear.

What molecular factors are associated with stages of retinal wound healing events, which include gliosis, edema, fibrosis, scar formation, and neovascularization? There is a constellation of events that surround retinal wound healing, and the RPE cell appears to be an important player as a source of biologically active molecules like cytokines, chemokines, and growth factors. RPE cells produce IL-1, M-CSF, and MGSA, and monocytes can influence the expression of multiple RPE-derived cytokines.

Can reliable animal models be developed that accurately mimic human retinal wound healing? The development of animal models that specifically mimic wound healing would hasten progress toward understanding the retinal wound response.

Can effective therapies be developed to arrest or prevent the retinal wound response in humans? Development of effective therapies will likely be dependent on a more indepth knowledge of the various biologically active molecules involved in the wound healing response. It remains to be determined which cytokines and growth factors are major players in the wound response or whether they all participate equally.

Objective 6: Identify the factors that dictate the unique properties of intraocular immunity and inflammation and alter systemic immunity to intraocular antigens.

Research Needs and Opportunities

Eye-derived immunosuppressive and anti-inflammatory factors play critical roles in ocular immunology, and there is a need to understand how they mediate intraocular immunity and inflammation. Naturally occurring molecules like defensins are important in intraocular immunity and inflammation. The intraocular cellular sources of natural defensins need to be identified, and the molecular mechanisms that enable these factors to be expressed constitutively need to be defined. The immunosuppressive, anti-inflammatory, and natural defensins that are expressed on ocular cells and are present within privileged ocular compartments need to be fully described and characterized.

Novel genetic programs that dictate the creation and maintenance of ocular immune privilege are expressed in the eye by a unique set of genes, including those encoding immunosuppressive and anti-inflammatory factors and their receptors. More information on these genes and their expression and control is needed. In this regard, transgenic animal technology should be employed and will provide an opportunity to elucidate the molecular mechanisms by which immune privilege operates.

Strategic Research Questions

What intraocular factors, soluble or membrane bound, can inhibit immunogenic inflammation within the eye and enable ocular antigens to create a systemic immune-deviant response? The eye is one of several specialized organs or tissues that display immune privilege (i.e., sites that permit foreign tissue grafts to enjoy prolonged survival). Immune privilege is an active rather than a passive process, and knowledge of its molecular basis will facilitate understanding of the underlying mechanism.

Which ocular cells produce intraocular factors, and what is the nature of gene regulation that enables their constitutive production? The microenvironment within the eye contains important immunomodulatory molecules that alter the manner in which ocular antigens are first perceived by the immune system. It also modifies the extent to which immune effectors can mediate protection in the eye.

How do intraocular factors interact with lymphoreticular cells and vascular endothelial cells to regulate the expression of immunity and inflammation within the eye? Immunomodulatory factors within the eye interact with target receptors on the cell surface to regulate immunity. One important mechanism by which immune privilege molds the systemic immune response is by promoting antigen-specific tolerance among peripheral T-cells.

What are the cellular and molecular mechanisms that enable antigenic signals arising from the eye to induce systemic immune deviation? Antigens injected intraocularly have a significant impact on the systemic immune response. One expression of that impact is the emergence of regulatory cells in the spleens of recipients.

Can the principles of immune privilege and immune deviation be used to influence the course of ocular inflammatory and immune diseases? One important mechanism by which immune privilege molds the systemic immune response is by inducing or prompting antigen-specific tolerance among peripheral T-cells. At least four mechanisms are considered relevant to privilege related tolerance: clonal selection, clonal anergy, immune deviation, and suppression.

Objective 7: Develop diagnostic methods and therapeutic approaches that distinguish among infectious, immunopathogenic, and autoimmune posterior segment intraocular inflammation.

Research Needs and Opportunities

Posterior segment intraocular inflammation is an important cause of blindness and can display distinct clinical features. Fundamental knowledge on the extent to which infectious, immunopathogenic, and autoimmune processes interact in producing ocular inflammation is needed. Experimental autoimmune ocular inflammation is a convenient model for the study of ocular inflammation, and the molecules that can induce this condition need to be characterized. The genes that contribute to the susceptibility and resistance to ocular inflammation and that are induced by immunization with ocular autoantigens also need to be identified.

The two major impediments to understanding the pathogenesis of posterior segment intraocular inflammation and for developing treatment and prevention strategies are: (1) lack of knowledge concerning the etiologic factors that cause the disease, and (2) inability to sample the inflamed eye for accurate diagnosis and evaluation of disease status. In many cases it is unclear whether the inflammation results from infectious, immunopathogenic, or autoimmune processes. Information is needed on the extent to which these three pathogenic events interact in producing ocular inflammation. Technical barriers to invasive procedures for acquisition of tissue and fluids for analysis should be overcome and will provide an opportunity to learn more about infectious agents and inflammation-producing immune effector cells and molecules. Opportunities to develop diagnostic approaches and criteria to distinguish among these infectious, autoimmune, and immunopathogenic mechanisms of ocular inflammation will result from these investigations.

Strategic Research Questions

What molecules uniquely expressed in the posterior segment of the eye can function as autoantigens in the pathogenesis of uveitis? Experimental autoimmune ocular inflammation has been an effective model for the study of ocular inflammation. Immunodominant peptides derived from retinal autoantigens have been identified and shown to induce uveitis. The function of these molecules in the pathogenesis of the disease needs to be elucidated.

What immune effectors (T-cell subsets, types of antibodies, effectors of innate immunity) trigger and participate in inflammation in the posterior segment of the eye? How is inflammation in the posterior segment of the eye triggered? To what extent do silent or unsuspected infections within the eye or extraocular processes act as triggers? What are the offending pathogens? The fundamental mechanisms on triggering the inflammation response in the eye are not known. Recent experimental evidence has shown the involvement of effectors and mediators. The role of T-cells as the primary initiator of inflammation needs to be further investigated.

Which polymorphic host genes confer susceptibility and resistance to inflammation of the posterior segment of the eye? Do these genes encode proteins involved in the adaptive immune response? In the innate immune response? The specific genes and genetic mechanisms involved in both susceptibility and resistance to inflammation need to be identified. Emerging molecular immunological technology will aid in this search.

Which proinflammatory mediators, cytokines, and chemokines mediate intraocular inflammation? The observations concerning mediators in the eye need to be expanded to identify both the mediators and the specific intraocular cells that release these mediators. Mediators can be identified by means of a variety of methods, including in situ hybridization.

Can samples of intraocular fluids or tissues and cells be obtained for evaluation of the etiology and pathogenesis of intraocular inflammation? The ability to obtain a sample of an ocular tissue or fluid will serve both to identify the infectious agent for accurate diagnosis and treatment and to evaluate the pathogenesis of the posterior segment inflammation. Advances in surgical procedures for invasive procedures, coupled with sensitive molecular probes, will allow infectious causes and immunopathogenic processes to be sorted out.

Objective 8: Analyze the mechanisms underlying light adaptation and recovery following phototransduction.

Research Needs and Opportunities

During the past decade, biochemical, molecular biological, physiological, and structure-function studies have produced a detailed understanding of the phototransduction cascade. Explorations have established the molecular description and key components of the pathway. Although the activation limb of phototransduction is now fairly well understood, inactivation is less so. There is a need to understand the detailed molecular mechanism by which both rod and cone photoreceptors recover to their dark state following the photoexcitation. The detailed mechanisms for the termination and recovery phase of the photoresponse in photoreceptors cells (rods and cones) need to be determined. In addition, the control of reactions responsible for light adaptation in photoreceptor cells and feedback pathways from horizontal cells need to be defined.

To solve this problem, the mechanisms by which photoreceptors adapt to different light intensities must be discovered. This will require a more indepth study of the role of calcium and regulatory reactions, such as protein phosphorylation and protein-protein interactions in individual steps in the visual transduction pathway. The relationship of adaptation mechanisms in photoreceptors to adaptation mechanisms in other retinal neurons also needs to be understood. New, emerging molecular and physiological technologies will provide unique opportunities to identify and characterize the function of novel proteins that are involved in recovery and adaptation reactions. The various proteins and reactions that mediate the photoresponse in cone cells should be identified.

Understanding the cone photoresponse at a molecular level will provide insights into the molecular basis for differences in light sensitivity, response, and recovery of rod and cone photoreceptors. Analyzing the components and mechanisms of phototransduction and light adaptation in cone cells should shed new light on the molecular basis for differences in the phototransduction and adaptation in rod and cone photoreceptors.

Strategic Research Questions

Can current and newly emerging techniques in molecular and cellular biology, biochemistry, physiology, and biophysics be used to enhance scientists' understanding of phototransduction, recovery, and light adaptation? The current task is to elucidate the mechanisms of recovery and light adaptation. The identification of components and mechanisms is mandatory to understanding the detailed molecular basis of visual transduction.

What novel proteins and mechanisms related to phototransduction and adaptation in rods and cone cells remain to be identified? Additional components are needed to account for the termination of the visual signal and for adaptation. Several components have been identified and set the stage for studies that will determine how they participate in the process. Highly sensitive molecular biology and biochemical techniques (i.e., protein expression systems) can be used to characterize these proteins and elucidate their role in the photoresponse.

Can high-resolution structural analysis be used in conjunction with mutagenesis and gene expression to provide detailed information about the important functional domains of proteins, sites of protein-protein interaction, and reaction mechanisms involved in phototransduction, recovery, and adaptation? Biophysical analysis coupled with molecular biological technology is now an exciting tool for exploring the molecular attributes of function. Such information should provide not only a detailed description of the mechanism of the photoresponse but also be valuable for assessing how mutations in these proteins can affect the function of photoreceptors and cause photoreceptor degeneration.

Objective 9: Study how visual information is transformed by successive layers of the neural retina and the mechanisms involved.

Research Needs and Opportunities

Understanding how visual information is processed as it transverses through successive neural layers of the retina is critical to elucidating retinal circuitry and identifying key levels at which therapeutic intervention may be possible. Work on processing of information by retinal neurons should exploit recent technical advances and move beyond descriptive studies. It is now possible to record from many mammalian retinas, including those of primates, under visual control. This allows selection of the neuron to be studied and identification of its shape and connectivity after characterization of the physiological behavior. New coding of visual information within ganglion cells needs to be determined. The temporal shaping of visual information of retinal ganglion cells should be elucidated. Further technical work should extend these in vitro capabilities, particularly in the arena of isolated and cultured cells and whole retinas. New methods for recording from the retina should be extended, including the use of multielectrode arrays and optical indicators of neuronal activity. Physiological, anatomical, and molecular analysis of the transmission of color information in higher vertebrates should be continued.

The biophysics of the retina's ribbon synapses and multisynaptic complexes needs to be understood. The pattern of neural connectivity among the cells should be determined. The activity of the different retinal synapses needs to be understood at a molecular level. In addition, their complements of neurotransmitter receptors and ion channels must be characterized, as it is now clear that each cell class has its own array. These should be studied by both molecular and physiological methods. The role of gap junctions needs further exploration. The search for genes expressed selectively in subclasses of retinal neurons should be pursued, as should methods of controlling the expression of specific genes in targeted retinal neurons.

Strategic Research Questions

What are the transformations of the visual input that occur within each of the retina's neural layers? With the cataloging of neurotransmitters nearly complete, newly emerging molecular and cellular biological techniques must be applied to discover the fundamentals of retinal circuitry and neurotransmitter function. In parallel with the emphasis on the molecular properties of retinal neurons, the way that these cells work together to process visual information must be discovered.

Can optical and electrophysiological methods of multineuronal recording be extended to study retinal interneurons? A multidisciplinary approach must be taken to reveal and understand the fundamental processes of normal retinal function. The multiple recording techniques are powerful because recording simultaneously from many cells is more efficient than studying them one at a time. Important information on waves of retinal activity is emerging that may help the brain "wire" itself correctly during early development.

How does the synaptic connectivity of the neurons control the visual transformations that are accomplished? The basic connectivity, pharmacology, physiology, and cell biology of both the rod and other specific synaptic pathways must be investigated. Determining the regulation of neuromodulators on functional and structural connectivity of cellular pathways in the retina is important.

What are the neurotransmitter receptors and ion channels on each of the retina's cell types? The diversity of retinal functions requires multiple types of neurotransmitter molecules and receptors. A great deal of information has been obtained on the identification and localization of neurotransmitters and their receptors. Additional emphasis must be placed on the investigations of how neurotransmitters and their receptors exert their specific actions. Identifying the ion channels and their specific role in transmitting visual information in the retina is of particular interest.

What role in visual processing is served by the retina's ribbon synapses, multisynaptic complexes, and gap junctions? The regulation of the number and function of gap junctions in retinal neurons and the role of ribbon synapses must be explored using multifaceted approaches, thereby extending the anatomical identification.

Can molecules and their function, specific to individual types of cells or synapses, be identified? Can cell-specific molecules (or whole classes of cells) be modified or manipulated by genetic techniques in a way useful for dissecting retinal function? Retinal neuroscience can capitalize on the development of molecular technology for probing cellular functions. Molecular probes will continue to be useful in identifying the presence of specific molecules in the neural retina and in localizing the molecules within cells.

Objective 10: Identify and characterize factors in retinal cell fate determination and differentiation.

Research Needs and Opportunities

Researchers currently do not know how many types of progenitor cells exist, or how their properties might contribute to the genesis of different cell types. They do not know if there are any "stem cells" in the retina that could be exploited for transplantation therapies. The different types of progenitor cells need to be identified and described.

Studies on signaling mechanisms in cell fate determination using both cell culture systems and analyses in vivo need to be emphasized. Culture systems that allow assay of factors on both cell fate determination and differentiation have been developed in the past, but they need to be expanded and made more sophisticated. For example, many studies have examined rod photoreceptor development, but very few have examined cone development. Potential extrinsic cues that affect development will be identified.

Systems amenable to genetic analysis are invaluable for in vivo studies of signaling pathways such as the EGF, Hedgehog, and Wingless signaling pathways. Important questions are how the signaling pathways are integrated and used in cell fate decisions in the eye. Animal models such as Drosophila, zebrafish, Xenopus, and mouse each offer a different set of advantages and disadvantages for studies of function in vivo. Knockouts, particularly conditional knockouts, and transgenic mice should prove to be critical for these studies. When researchers have made progress in understanding the progenitors and know more about the extrinsic cues that affect them, they will be in a position to use this information for transplantation therapies.

Very little is known about how retinal cells adopt their final cell shapes, structures, and polarity. Studies on the roles of membrane biogenesis, integrins, microtubules, actin, myosin, protein targeting, and transport in photoreceptor cell morphogenesis should be emphasized. In addition, mechanisms that regulate the proliferation of retinal precursor cells and subsequently lock them in a postmitotic state when they differentiate are poorly understood. Understanding these mechanisms is of obvious significance to developmental processes and to many human retinal diseases. The adult retina is incapable of regeneration following damage, injury, ischemia, or degenerative disease.

Strategic Research Questions

What are the properties of progenitors that contribute to the genesis of different cell types? Are there totipotent stem cells in the retina that could be exploited for transplantation therapies? Can mechanisms that regulate the proliferation of retinal precursor cells and subsequently lock them in a postmitotic state when they differentiate be understood? Visual system development involves the production and specification of individual neurons and glial cells that are necessary for assembling the retina. Intrinsic and extrinsic cues that affect development should be identified. Molecular and cellular biological techniques provide powerful tools for dissecting mechanisms of retinal developmental events.

Can tissue-culture systems be developed and expanded that allow assay of factors on both cell fate determination and differentiation, especially for cone development? By applying growth factors, small molecules, and other secreted gene products to cells grown in vitro, cell biological screens can be conducted. The types of effects assayed in vitro include changes in cell fate, effects on differentiation, and effects on survival. Perturbation of expression or function of molecules with effects in vitro can then be expanded to in vivo studies.

Can systems amenable to genetic analysis of development, such as Drosophila, zebrafish, and Xenopus, be pursued? The overall strategy of retinal cell fate determination and differentiation is the same in the vertebrate and Drosophila retina, as are many of the molecules employed. For example, the extracellular receptor Notch, and its ligand Delta, are crucial for the development of all retinal neurons in Drosophila and vertebrates.

How do retinal cells adopt their final cell shapes, structures, and polarity? Retinal cells have remarkable structure, shape, and polarity. Fundamental retinal processes that both create and maintain retinal organization are unknown. A detailed understanding of the components and mechanisms that play a role in retinal organization will lead to insights in both the normal and the degenerative retina.

Can cell-specific markers be developed and used to stage the differentiation pathways of each cell type? Antibodies and molecular probes specific for receptors and proteins that may be involved in retinal developmental stages have enormous potential for sorting out differentiation pathways. Genes that are expressed in specific cell types can be used as a basis for making antisera for rapid identification and manipulation of different cell types.

Can animal models (i.e., knockouts, particularly conditional knockouts and transgenic mice) be used specifically for providing critical information on the molecular mechanisms of retinal development? Animal models will provide an opportunity to study retinal development in vivo. For such studies, it would be helpful to have reagents for manipulating gene expression in the retina. For example, promoters that could be regulated to control the onset of expression of genes and strains of mice with reporter genes for identifying particular cell populations would be important tools for understanding the molecular mechanisms involved in retinal development.

How are the signaling pathways integrated and used in cell fate decisions? Cell culture and in vivo systems are available for studying signaling mechanisms. Studies on rod developmental pathways need to be expanded to include cone development and may include developing more molecular markers to identify cell types.

Objective 11: Catalog, map, and functionally characterize genes expressed in the retina and choroid and begin to determine the cellular sites of retinal gene expression in health and disease.

Research Needs and Opportunities

The opportunity to identify all genes expressed in the retina and choroid is now possible through resources developed by the Human Genome Project and related projects for other systems. Determining this complete set of genes and their cellular location and specificity is critical for understanding retinal structure and function and the cellular interactions during development. It will be possible to map retinal disease genes in specific animal models of human retinal diseases and map human retinal diseases for which there are families of a significant size. The mapped genes can be evaluated that are specific to the retina or choroid as candidate genes for disease.

The next step will include determining the cellular specificity of genes expressed in the retina or choroid and determining the changes in gene expression that occur as a result of disease. It is important to begin to functionally characterize these retinal and choroidal genes to further understand the relationship of their physiology to human disease.

Strategic Research Questions

Can genes expressed in the adult retina and at key developmental times be identified through analysis of normalized cDNA libraries? The retina and choroid offer an ideal set of tissues to identify genes expressed in development and in the adult. It is straightforward to isolate for the preparation of cDNA libraries from different stages. The opportunity to identify all genes expressed in the retina and choroid is now possible through resources developed by the Human Genome Project and other related projects.

Can cDNA libraries from critical retinal regions, such as the macula and ciliary margin, be prepared and used for identifying unique genes and their function? cDNA libraries are being prepared and sequenced in several laboratories. The cataloging of which genes are expressed in what cells will be challenging, and emerging technologies will be useful. Current technologies for mapping sites of gene expression, including immunohistochemistry and in situ hybridization, are expensive and time consuming to perform for the many genes that will need to be characterized.

Can ESTs that appear to be unique to the visual system or that have hallmarks of genes that are good candidates for visual system disease be mapped and used as candidate genes for retinal diseases? The availability of ESTs specific for the visual system may provide a valuable tool for identifying genes that are responsible for retinal degenerative diseases. The genetic mapping of the ESTs will be necessary.

Can genes responsible for human retinal disease and development be functionally characterized to elucidate molecular mechanisms of retinal diseases? Can emerging novel technologies, including DNA chips, be employed to provide the best methods for this goal? The mechanism by which a gene identified as the diseased gene results in retinal degeneration is critical to the application of this information to patients with retinal degenerative diseases. Taking advantage of recent technological advances to develop new research strategies should be incorporated in all molecular and cellular approaches, if applicable.

Objective 12: Probe the control of the retina's microenvironment through studies of Brüch's membrane, the interphotoreceptor matrix, the RPE, glia, choroid, and vitreous.

Research Needs and Opportunities

Studying the production, maintenance, and turnover of Brüch's membrane and the interphotoreceptor matrix is important to understanding the normal, healthy retina and its interaction with its microenvironment. The basic biology of the vitreous, including its development and changes in aging, need to be studied. Researchers also need to determine how proteases and protease inhibitors are involved in the aging process of Brüch's membrane, photoreceptor sheaths, and vitreous. The roles of paracrine factors in the function and survival of retinal neurons must be clarified, which includes the mechanisms by which light-induced, circadian, and pathological changes in the concentrations of ions, neurotrophic factors, neuromodulators, and enzymes in the extracellular spaces of the retina affect the physiology and survival of retinal neurons, RPE cells, and Müller cells. In addition, a better understanding of factors produced by the RPE and other retinal and choroidal cells that inhibit neovascularization must be gained. A thorough investigation of the role of Müller and RPE cells in retinoid uptake processing, release, and transport into and within the retina will be critical to understanding the maintenance of the normal retina. This should include the development of a better understanding of the morphogenesis and physiology of Müller cells and astrocytes, particularly the regulation of the ionic and neurotransmitter content of the retinal and subretinal extracellular space, establishment and maintenance of polarity, extracellular and intracellular signaling mechanisms, metabolic support of retinal neurons, intercellular coupling, and identification of secretory products. The mechanisms for the specific morphology and physiology of the RPE need to be better understood, particularly with regard to blood-retinal barrier selectivity, maintenance of cell polarity, phagocytosis, RPE phenotype plasticity, metabolism, and signaling mechanisms. In addition, a complete understanding of all unidentified or uncharacterized proteins involved in the uptake, processing, transport, and release of retinoids by cells of the retina should be undertaken.

Strategic Research Questions

What are the cellular sites for the synthesis of the molecular components of Brüch's membrane and the rod and cone outer segment sheaths? An intricate network of macromolecules and components completely surround the RPE/neural retina and may play a role in regulating the behavior of the surrounding cells. Once the components are identified, cell biological techniques could be employed to investigate their synthesis.

Can Müller cells and retinal astrocyte cocultures be developed for electrophysiological analysis of interactions between these cells? It will be necessary to develop cocultures that actually express the same characterization as cells in vivo. Well-defined in vivo systems may provide the first step in characterizing the transporters, receptors, and other genes expressed by Müller cells.

Can the roles of genes in the development and differentiated functions of RPE cells, Müller cells, and retinal astrocytes in their normal environment be examined? Emerging transgenic technologies can be exploited to sort out the specialized functions of the cells surrounding the retina, such as transepithelial flow of ions and nutrients in the RPE.

Can trophic factors and their receptors synthesized by Müller and RPE cells be identified and characterized? The viability of the retinal structures relies on the components secreted by the cells that establish the intercellular matrices. Cellular biological technology should be exploited to identify the trophic factors involved, and their role in the development and morphogenesis of neural retina should be determined.

Can the roles of Müller and RPE cells in paracrine signaling be characterized? The RPE's response to paracrine factors and details of the signaling pathway need to be understood. By applying electrophysiological and molecular methods, the identity of the specific factors and retinal targets involved may be discovered.

Can the remaining, unresolved components of the visual cycle be characterized so researchers can develop a complete understanding of this process and its role in inherited retinal degenerations? What is the role of Müller cells in the processing and recycling of retinoids? The enzymes and proteins of the visual cycle continue to be characterized. Molecular dissection of the pathway will be necessary to fill in the details of the visual cycle and may serve to provide molecular probes for identifying mutations in genes coding proteins of the visual cycle. The role of Müller cells needs further clarification.

Retinal Diseases Panel

Chairpersons

Dean Bok, Ph.D.
University of California at Los Angeles
Los Angeles, CA

Paul Sieving, M.D., Ph.D.
Kellogg Eye Center
Ann Arbor, MI

Panel Members

Mary Beth Burnside, Ph.D.
University of California, Berkeley
Berkeley, CA

Constance Cepko, Ph.D.
Harvard Medical School
Boston, MA

Nansi Colley, Ph.D.
University of Wisconsin
Madison, WI

Patricia D'Amore, Ph.D.
Children's Hospital
Boston, MA

Thaddeus Dryja, M.D.
Massachusetts Eye and Ear Infirmary
Boston, MA

Robert Frank, M.D.
Kresge Eye Institute
Detroit, MI

Ronald N. Germain, M.D., Ph.D.
National Institute of Allergy and Infectious Diseases, NIH
Bethesda, MD

Gregory Hageman, Ph.D.
University of Iowa
Iowa City, IA

Joseph Hammang, Ph.D.
CytoTherapeutics, Inc.
Lincoln, RI

William Hauswirth, Ph.D.
University of Florida College of Medicine
Gainesville, FL

Leslie Hyman, Ph.D.
SUNY at Stony Brook
Stony Brook, NY

Richard Masland, Ph.D.
Massachusetts General Hospital
Boston, MA

Robert Molday, Ph.D.
University of British Columbia
Vancouver, British Columbia
CANADA

Margaret Pericak-Vance, Ph.D.
Duke University Medical Center
Durham, NC

Edwin M. Stone, M.D., Ph.D.
University of Iowa College of Medicine
Iowa City, IA

J. Wayne Streilein, M.D.
Schepens Eye Research Institute
Boston, MA

Gabriel Travis, M.D.
University of Texas Southwestern Medical Center
Dallas, TX

NEI Staff

Peter A. Dudley, Ph.D.
National Eye Institute, NIH
Bethesda, MD

Maria Y. Giovanni, Ph.D.
National Eye Institute, NIH
Bethesda, MD