Monica L Vetter and Edward M Levine

1. introduction

Degenerative diseases of the retina result in the loss of specific populations of retinal neurons. For example, retinitis pigmentosa is characterized by progressive loss of rod photoreceptors, macular degeneration is a common disease of the elderly in which rod and cone photoreceptors degenerate, and glaucoma is marked by a loss of retinal ganglion cells. Thus, there is considerable interest in identifying retinal stem cells with the capacity to repopulate the retina in response to disease or injury. Although this has not yet been achieved in the mammalian eye, recent results hold promise for future success in this area.

One source of evidence for a resident stem cell population in the retina is the capacity of retinal tissue to regenerate in response to injury or damage. There are considerable species differences in the potential for retinal regeneration. It has been known for some time that retinal stemlike cells exist in the eyes of fish and amphibians because classic studies demonstrated a capacity for retinal regeneration in these species (1-5). More recently, there is evidence in chick and even mammals for cell populations that have potential to differentiate into retinal cells (4,5). There are a number of different sources for cells that can undergo neural retinal differentiation; however, one common feature found thus far is that all cells capable of giving rise to retinal neurons originate from the neural ectoderm-derived structures of the optic vesicle.

Here, we review the understanding of stem cells in the adult vertebrate retina. We first describe the organization of the vertebrate retina, briefly discuss how cells of the retina are generated during development, and then present the evidence for stem cells both in the neural retina and in ocular tissues outside the retina. Finally, we discuss the prospects for therapeutic treatment of retinal disease using retinal stem cell technology.

From: Adult Stem Cells Edited by: K. Turksen © Humana Press Inc., Totowa, NJ

Eye Cell Populations

Fig. 1. Structures of the eye, neural retina, and developing optic vesicle. (A) The main components of the mature vertebrate eye. Ocular tissues discussed in the text and that contain cell populations with retinal stem cell properties are indicated with bold labels. (B) The neural retina contains six major neuronal cell types (rod and cone photoreceptors, horizontal cells, bipolar cells, amacrine cells, and ganglion cells) and one major glial cell type (Müller glia). These cell types are organized

Fig. 1. Structures of the eye, neural retina, and developing optic vesicle. (A) The main components of the mature vertebrate eye. Ocular tissues discussed in the text and that contain cell populations with retinal stem cell properties are indicated with bold labels. (B) The neural retina contains six major neuronal cell types (rod and cone photoreceptors, horizontal cells, bipolar cells, amacrine cells, and ganglion cells) and one major glial cell type (Müller glia). These cell types are organized

2. organization of the vertebrate ocular tissues

Several regions of the eye have been implicated in housing cell populations with stem cell properties or that have the potential to become retinal stem cells. These include the neural retina itself, the retinal pigmented epithelium (RPE), the pigmented ciliary epithelium, and the iris epithelium (Fig. 1A). In this section, we briefly describe the basic cellular organization of these tissues in the mammalian eye. For a comprehensive description of visual system anatomy, physiology, and function, see the Webvision Web site at

2.1. Neural Retina

The neural retina processes visual information and is subdivided into three distinct cellular layers with cellular composition that is correlated with the flow of visuosensory input (Fig. 1B). The outer nuclear layer (ONL) is composed of the rod and cone photoreceptors; the inner nuclear layer (INL) is composed of three classes of interneurons, horizontal cells, bipolar cells, and amacrine cells; and the ganglion cell layer (GCL) is composed of retinal ganglion cells and displaced amacrine cells. The Müller glia nuclei are resident in the INL, but the full extent of these cells traverses all layers and terminates to form two membranes: the external limiting membrane and the inner limiting membrane. Two synaptic layers are sandwiched between the nuclear layers. Rod and cones synapse onto the dendrites of horizontal cells and bipolar cells in the outer plexiform layer, and bipolar cells and amacrine cells synapse onto the retinal ganglion cell dendrites in the inner plexiform layer. Finally, the nerve fiber layer is composed of retinal ganglion cell axons and astrocytes. Importantly, many of these major cell classes, most notably

into three distinct layers: the ganglion cell layer, the inner nuclear layer, and the outer nuclear layer. The nonneural retinal pigment epithelium (RPE) is in close contact with the outer segments of the rod and cone photoreceptors. (C) Eye development begins when the optic vesicles evaginate from the walls of the forebrain at very early stages of neural development. As development proceeds, the optic vesicles invaginate to form an optic cup. The outer layer of the optic cup will become the retinal pigment epithelium (RPE) and more distally will give rise to the pigmented ciliary epithelium (PCE) and the pigmented iris epithelium (PIE). The inner layer of the optic cup will differentiate into neural retina and more distally will give rise to the nonpigmented ciliary epithelium (CE) and the nonpigmented iris epithelium (IE). The optic stalk will become the optic nerve (ON). (Panels A and B modified from Webvision,, with the expert assistance of Mary Scriven and kind permission of Dr. Helga Kolb.)

the interneurons and ganglion cells, are further divided into many distinct subtypes.

2.2. Retinal Pigmented Epithelium

The RPE is a monolayer of epithelial cells in close contact with the outer segments of the rod and cone photoreceptors in the neural retina. RPE cells are highly differentiated in that they have an apical (toward the photoreceptors) and basal polarity, and all of the RPE cells are joined near the apical side by tight junctions (zonula occludens), which functions in part to form the blood-brain barrier. RPE cells are important for maintaining the integrity of photoreceptors by providing trophic support and by phagocytosing outer segment disks shed by the photoreceptors. RPE cells are highly enriched in the pigment melanin.

2.3. Ciliary Epithelium

The ciliary epithelium is a bilayered epithelium that extends from the periphery of the neural retina and RPE. The monolayer that is continuous with the neural retina is not pigmented, whereas the monolayer continuous with the RPE is pigmented. This tissue is an important region for attachment of several ocular structures, including the retina and lens. In addition, the nonpigmented ciliary epithelial cells secrete aqueous fluid and maintain intraocular pressure.

2.4. Iris Epithelium

The iris epithelium is similar in structure to the ciliary epithelium, and it forms the outermost margin of tissue continuous with the neural retina, RPE, and ciliary epithelium and forms the boundaries of the pupil. Like the ciliary epithelium, the iris epithelium is home to several ocular muscles that are interspersed through the epithelium.

3. development of ocular tissues 3.1. Optic Vesicle Development

The neural ectoderm structures of the eye, including the neural retina, RPE, ciliary epithelium, and iris epithelium, are derived from the optic vesicles (Fig. 1C). These consist of bilateral evaginations from the walls of the forebrain during early nervous system development. As development proceeds, the optic vesicles expand toward the head ectoderm before invagi-nating to form a bilayered optic cup. The outer layer of the optic cup (proximal) will give rise to the RPE, and the inner layer of the optic cup (distal) thickens, undergoes a proliferative expansion, and will differentiate into the neural retina. The iris and ciliary epithelia arise primarily from the neu-roepithelium.

3.2. Retinal Cell Fate Specification

Within the neural retina domain of the optic cup are proliferating retinal progenitors that will give rise to six major neuronal cell types (rod and cone photoreceptors, bipolar cells, amacrine cells, horizontal cells, ganglion cells), as well as to Müller glia. These retinal cell types are born in an overlapping sequence that is largely conserved across species (6). Lineage analysis in a number of species has demonstrated that retinal progenitors are multipotent rather than dedicated to the generation of specific retinal cell types (7-9). However, there is evidence that the competence of retinal progenitors changes over developmental time, so that at any one time, they give rise to a limited subset of retinal cell types (6). In addition, there is progressive restriction in developmental potential of retinal progenitors. Thus, the ultimate fate of differentiating retinal cells depends on the intrinsic competence of the retinal progenitors and extrinsic signals that provide specific differentiation cues (6,10,11).

4. definition of a retinal stem cell

As any reader of the stem cell literature can attest, it is essentially impossible to assign a precise and all-inclusive definition of what constitutes a stem cell or a stem cell population. A major reason for this is that the application of the term stem cell has become context dependent. This has occurred in large part because of the experimental interventions often used to identify a stem cell, and because different tissues have distinct requirements in maintaining cell numbers and cell-type diversity. Moreover, what is considered a primary characteristic of a stem cell in one tissue may be considered a minor characteristic in another. Thus, to evaluate the fast-growing field of stem cell research critically, the context in which the term stem cell is applied must be understood.

Potten and Loeffler (1990) proposed an adult stem cell lineage model that is based on the robust and continuous renewal of the crypt epithelium in the small intestine (12). Primary characteristics of the adult stem cell are that they are undifferentiated, have a long cell cycle time, and are self-renewing. A subset of stem cell progeny rapidly proliferate (transit-amplifying cells) and produce all of the differentiated progeny that make up the complement of cell types in a given structure. Although the transit-amplifying cell can generate many progeny, it is not capable of long-term self-renewal under normal conditions. Finally, with appropriate stimuli such as tissue damage, an adult stem cell population may have the propensity to regenerate the damaged tissue. The remainder of this chapter considers putative sources of adult retinal stem cells with these characteristics in mind.

In the vertebrate neural retina, two predominant precursor cell populations are recognized: stem cells and progenitor cells (2,4,5). During embryonic development, as yet there is no evidence for a self-renewing retinal stem cell population. Rather, the cells of the mature retina born during the developmental period are derived from multipotent progenitors. These progenitors are essentially a transit-amplifying cell population because they proliferate rapidly, give rise to the complement of neural cell types, and are ultimately depleted. As described in the following sections, there are now several putative sources of adult retinal stem cells, some originating in the neural retina and some originating from other ocular tissues. These candidate stem cell populations were discovered in vivo as well as by in vitro approaches.

5. sources of stem cells in the retina

There are several cell populations in the neural retina that have the potential to contribute to retinal regeneration, although considerable species differences exist. Fish, amphibians, and chicks have a population of stem cells at the margins of the retina that contribute to the normal growth of the eye and can repopulate cells in response to damage or injury (4,5). In adult fish, there is also a more specialized progenitor population in the retina that generates rod photoreceptors selectively (2). Evidence also suggests that, in response to neurotoxic damage, Müller glia in the chick retina can reenter the cell cycle and give rise to subsets of retinal neurons (13). These different cell populations are reviewed below.

5.1. Ciliary Marginal Zone of the Retina

In fish and amphibians, growth of the retina and RPE after the initial embryonic period is achieved by the addition of cells from a proliferative zone at the retinal margins known as the ciliary marginal zone (CMZ). This was demonstrated by showing that a pulse of [3H]-thymidine in the postem-bryonic period labels cells at the margins of the retina, and these labeled cells ultimately differentiate into retinal neurons and glia (14-17). Thus, new cells are added in rings as the retina grows. In the CMZ, the most peripheral cells are slowly dividing and only become labeled with prolonged [3H]-thymidine labeling (2). This is consistent with the low rate of proliferation characteristic of stem cells. Cells that are more centrally located divide more rapidly and can be labeled with short pulses of [3H]-thymidine, suggesting that these are rapidly cycling progenitors.

Injury or damage to the retina can stimulate proliferation of cells in the CMZ. Elegant experiments showed that selective ablation of amacrine and bipolar cells by intraocular injections of kainate in Rana pipiens tadpoles can stimulate increased cell production from the margins, which results in selective replacement of the ablated cells (18). Thus, the cells of the CMZ not only contribute to the normal growth of the retina, but also can be a source for regeneration of retinal cells.

To determine whether cells in the CMZ are multipotent or instead represent committed precursors of specific retinal cell types, the lineage tracer rhodamine dextran was injected into individual precursor cells at the margins of the Xenopus laevis retina (19). It was shown that cells in the CMZ are multipotent and can give rise to all major retinal cell types as well as to nonneural pigment epithelial cells. Analysis of clone size suggested that, in the margins, there are both self-renewing stem cells and progenitors with more limited proliferative potential (19). This has been confirmed by in situ analysis of gene expression at the margins of fish and Xenopus retinas (2024). These studies showed that there is a peripheral-to-central sequence of gene expression that recapitulates the temporal sequence of gene expression during retinal histogenesis (25). In fish, the cells of the CMZ give rise to all retinal cell types except rods, which are instead derived from a dedicated rod precursor population distributed throughout the INL of the retina (see Section 5.2.).

Until recently, it was believed that the CMZ was a feature unique to fish and amphibians. However, it has now been shown that in chicks a proliferative population of cells exists at the margins of the retina for up to 3 wk after hatching (26). These dividing cells differentiate and give rise to amacrine, bipolar, and Müller cells, but were not found to differentiate into ganglion cells, horizontal cells, or photoreceptors. However, injection of insulin and fibroblast growth factor 2 (FGF-2) into the vitreous chamber of posthatch chickens could stimulate the production of retinal ganglion cells, suggesting that it is either the absence of appropriate signals or the presence of inhibitory signals that normally prevents these cells from differentiating into all retinal cell types (27). Additional experiments will be required to determine whether cells in the CMZ of the chick are fully multipotent. Interestingly, unlike fish and amphibians, neurotoxic lesions to the retina do not provoke an increase in proliferation of cells in the chicken CMZ (26).

There is no evidence yet for CMZ-like cells in the mammalian retina, although it has been suggested that cells in the ciliary body of the eye are analogous to cells of the CMZ found in other vertebrates (see Section 6.1.). It will be interesting to determine why there are such intriguing species differences in the development and maintenance of a CMZ in the vertebrate retina.

5.2. Rod Precursor Lineage of Fish

In addition to the CMZ, another adult neurogenic progenitor cell population was initially discovered in retinas of teleost fish over 20 yr ago (28-30). These cells were termed rod precursor cells because they were observed to give rise exclusively to rod photoreceptors. In contrast to the CMZ cells, rod precursors are distributed along the entire central-to-peripheral plane of the mature retina and are found in the ONL interspersed with mature photo-receptors.

Several characteristics of the rod precursor cell suggest that it is a transit-amplifying cell rather than a true stem cell. First, its rate of cell division is quite rapid. Second, pulse labeling with BrdU or [3H]-thymidine followed by long survival times showed that the rod precursor cells do not self-renew, but rather differentiate into rods. Third, the cell output appears lineage restricted. Thus, even by minimal criteria, rod precursor cells do not qualify as bona fide adult stem cells. However, because this cell population persists throughout most of the life of the fish, it has been postulated that, in the retina, a stem cell population exists with an output that is the rod precursor population.

The identification of the putative rod precursor stem cell population has been elusive (1). For some time, the location of the rod precursor stem cell has been postulated to reside in the INL. This was demonstrated by [3H]-thymidine injections followed by successively longer survival times to trace the fate of the labeled cells. These proliferative INL cells divide and form clusters of labeled cells termed neurogenic clusters. Subsequently, individual cells migrate into the ONL, where they give rise to the rod precursor cell population.

A study by Otteson and colleagues (31) suggests that the neurogenic clusters may be maintained by yet another population of INL cells, and they propose that this population may be the stem cells. As opposed to a relatively short pulse of BrdU or [3H]-thymidine, goldfish were exposed to BrdU continuously for 9 d. This long labeling period would allow for the detection of a slow-dividing cell population. Using this approach, they observed two morphologically and spatially distinct populations of BrdU cells in the INL. One population had cell bodies of a fusiform morphology and was positioned toward the outer portion of the INL; this population is most likely the neurogenic clusters. The other population had a spherical morphology, was much fewer in number, and was positioned immediately adjacent to the neu-rogenic cluster at the inner boundary of the INL. In experiments in which the fish were allowed to survive for varying periods following BrdU treatment, it was observed that the BrdU-labeled spherical cells were maintained, the BrdU-labeled fusiform cells decreased, and the BrdU-labeled cells in the ONL increased. These observations suggest that the spherical cells self-renew and give rise to the neurogenic clusters, which in turn give rise to the rod precursor cells. Consistent with this, these different cell populations were organized into radial arrays.

Several issues still need to be resolved, however, before it can be concluded that the spherical cells are stem cells. First, it needs to be determined that the neurogenic clusters are lineal descendants of the spherical cells. Second, can the spherical cells repopulate neurogenic clusters in a manner similar to that shown in depletion-replacement experiments done in the hematopoietic lineage and more recently in the subventricular zone of adult rodent brains?

Another interesting feature of the rod precursor lineage is its potential to become multipotent following injury to the retina. On physical ablation or severe neurotoxic damage, retinal neurons and glia are replaced, and it is well established that the source of new cells arises from the mature retina (1-3). Interestingly, it is suggested that the rod precursors in the ONL and the neurogenic clusters and spherical cells in the INL all contribute to the regenerative process (31). Thus, it appears that the healthy retinal environment restricts the rod precursor lineage.

It is now well established that adult neurogenesis occurs in the subventricular zones lining the lateral ventricles in the cerebrum (32). Two cell populations have been suggested as the neural stem cell: the astrocytes of the subependyma and the ependymal cells that line the ventricles (33,34). Although the precise identity of the stem cell is still controversial (35,36), in either case, the candidate populations are both glial cells. This is intriguing because it suggests that adult neural stem cells may be glia that dedifferentiate and then transdifferentiate into neurons.

As described in Section 2.1., the Müller glia are derived from the retinal progenitor population (7). Because they are lineally related to retinal neurons, it is tempting to speculate that the Müller glia may have the potential to behave as retinal stem cells. To date, there is no evidence that Müller glia are an intrinsic source of retinal neurons under normal conditions.

A study by Fischer and Reh (13), however, suggested that, in response to neurotoxic damage, Müller glia reenter the cell cycle, and new neurons are generated in the avian retina. N-Methyl-d-aspartate was administered to the retinas of juvenile chicks at a dose sufficient to kill amacrine cells and possibly other retinal cell types. After 2 d, a robust induction of proliferation was observed by BrdU incorporation in the INL. This proliferative response was transient, however, suggesting that the proliferating cells may have exited the cell cycle and differentiated.

To address this possibility, the fate of the BrdU-labeled cells were followed by examining expression of cell class-restricted proteins over a period of days to weeks. Initially, the overwhelming majority of BrdU-labeled cells expressed glutamine synthetase (GS), a marker of Müller glia. In days, however, many of the BrdU-labeled cells were GS negative, but expressed the homeodomain proteins Pax6, Chx10, and the basic helix-loop-helix protein CASH1, all of which are transcription factors coexpressed exclusively in retinal progenitors. Furthermore, some BrdU-positive cells also expressed Hu, a marker of differentiated ganglion cells and amacrine cells, or cellular retinoic acid binding protein (CRABP), a marker of differentiated amacrine and bipolar cells. These observations suggest that, given the appropriate stimulus, Müller glia have the potential to dedifferentiate into a progenitorlike state and then differentiate into retinal neurons.

Although the above study is certainly provocative, several unanswered questions still remain and warrant consideration. For instance, because clonal analysis of the damaged-induced proliferative cells was not done, it remains to be demonstrated that the BrdU-positive retinal neurons arose from dedifferentiated Müller glia. Second, the authors did not observe expression of neurotransmitters or of a photoreceptor phenotype in the BrdU-positive neurons. It is therefore important to determine whether these neurons have the capacity to mature fully or adopt a photoreceptor fate and what the signals are that can promote these fates. Finally, it is not yet known whether adult avian Müller glia retain this neurogenic potential, or whether mammalian Müller glia have the propensity to dedifferentiate into a progenitor-like cell that can generate retinal neurons.

6. sources of stem cells outside the retina

Outside the neural retina itself, there exist cell populations that have the potential to contribute to regeneration of retinal neurons. In a number of species, including chicks and amphibians, the nonneural RPE can transdifferentiate into retinal tissue (37). In mammals, there are cells from both the pigmented epithelium of the ciliary body and the iris epithelium that, under certain conditions, can differentiate into retinal neurons in culture (38-40).

6.1. Pigmented Ciliary Epithelium in Mammals

In mammals, neither the neural retina nor the RPE show any evidence for regenerative potential in adults. This raises the question of whether there are any cells in the adult eye with retinal stem cell-like properties. Long-term labeling of 4-wk-old rats with BrdU revealed a small population of proliferative cells in the pigmented ciliary body (38). The number of cells that incorporated BrdU could be stimulated in explant culture by treatment with FGF-2, a known mitogen for neural stem cells. Under no conditions was BrdU incorporation detected in the retina, RPE, or nonpigmented ciliary body. The presence of proliferative cells in the ciliary body and the fact that this structure shares embryological origins with the neural retina raised the possibility that these proliferative cells could represent stemlike cells in the adult mammalian eye. In addition, the ciliary body may be related to the ciliary marginal zone of the fish, amphibian, and chick neural retina, which contains resident retinal stem cells that contribute to retinal growth.

Dissociated cells from the pigmented ciliary epithelium of both mouse and rat could be grown in culture, in which, at very low frequency, they formed neurospheres, a proliferative colony of neural stemlike cells (38,39). Proliferative cells in the neurospheres were positive for nestin, a marker expressed by neural stem cells. The neurospheres had the capacity for self-renewal because a subset of single pigmented cells from dissociated neurospheres gave rise to new neurosphere colonies when recultured (38,39), and this could be repeated for at least six generations (39). Treatment with FGF-2 in serum-free culture enhanced neurosphere formation; however, there appeared to be production of endogenous FGF-2, permitting neurospheres to grow in the absence of exogenous FGF-2 (39). Neurosphere colonies did not arise from cultures of adult neural retina, RPE, iris epithelium, or ciliary muscle or from nonpigmented ciliary process cells. Pigmentation was not required for neurosphere formation because colonies could be generated from albino tissue. Neurospheres could not be cultured from adult neural retina or from adult RPE. Even neurospheres derived from E14 neural retina did not show a capacity for self-renewal. Thus, cells derived from the ciliary epithelium were unique in having stem celllike properties. Interestingly, the ability of ciliary margin tissue to give rise to neurospheres is conserved across species because colonies could also be derived from postmortem adult bovine and human ciliary margin tissue (39).

Neurospheres appeared to arise from single pigmented ciliary body cells that proliferated in culture and gave rise to mixtures of pigmented and non-

pigmented cells. Nonpigmented cells in the neurosphere expressed markers of retinal progenitors such as the homeodomain transcription factor Chx10. This suggests that pigmented cells of the ciliary body may be transdifferentiating into nonpigmented neural retinal progenitors (38,39).

To determine whether neurosphere colonies could differentiate into retinal neurons or glia, the cells were grown under conditions that promote differentiation. Cells from both primary and secondary neurosphere colonies differentiated and expressed markers for rods, bipolar cells, and Müller glia; however, markers for retinal ganglion cells, horizontal cells, and amacrine cells were either not detected or were extremely rare (38,39). This may be because of the culture conditions under which the assay was performed because amacrine cell differentiation could be enhanced by high-density pellet culture conditions (39). Thus, neurospheres derived from the pigmented ciliary epithelium not only self-renew, but are multipotential and give rise to cells with retinal progenitorlike properties that can differentiate into neurons and glia that express markers of differentiated retinal cells.

This work has obviously generated real excitement over the promise for regenerating retinal neurons lost to disease or injury; however, a number of questions remain. For example, can the retinal stemlike cells from the pig-mented ciliary epithelium give rise to all classes of retinal neurons? Will these cells (or progenitors derived from these cells) survive and differentiate when transplanted in vivo? And, can we define conditions to promote the differentiation of selected retinal cell types for replacement of retinal neurons lost to diseases such as retinitis pigmentosa or glaucoma? In addition, the role of these stemlike cells in vivo is not yet clear. Will it be possible to stimulate these cells in vivo to transdifferentiate to retinal progenitors and differentiate into retinal neurons in reponse to disease or injury? Much work remains to be done before the therapeutic potential of this work can be realized.

6.2. Iris Epithelium

The iris epithelium is contiguous with the pigmented ciliary epithelium, raising the possibility that it may share some of the stem cell-like properties of its neighbor (see Chapter 13). Traditionally, the iris epithelium has been associated with lens regeneration, but the potential for these cells to generate retinal tissue had not been carefully examined. Iris tissue does not give rise to neurosphere colonies (38,39); however, iris cells grown in monolayer culture in the presence of FGF proliferate and could be stimulated to differentiate into neurofilament 200-positive cells (40). However, unlike differ entiated cells derived from pigmented ciliary epithelium, the differentiated cells derived from the iris epithelium did not express rhodopsin, a marker of differentiated rod photoreceptors. No other retinal markers were examined in these experiments.

To determine whether iris-derived cells could be induced to differentiate into rodlike cells, the photoreceptor-specific homeobox gene Crx was overexpressed. Differentiated cells expressing Crx expressed rhodopsin as well as recoverin, another gene expressed in photoreceptors and a subset of bipolar neurons, suggesting that Crx expression in iris-derived cells was sufficient to promote photoreceptor differentiation (40). Similar results were obtained by overexpression of Crx in pigmented ciliary epithelium cells grown in monolayer culture (which does not normally result in photoreceptor differentiation). Interestingly, overexpression of Crx in a neural stem cell line derived from the adult rat hippocampus did not promote expression of photoreceptor-specific markers, suggesting that tissues derived from the optic vesicle may have a unique ability to differentiate into neurons with retinal properties.

In summary, these experiments demonstrated that cells derived from the iris epithelium have the potential to differentiate into photoreceptors, but only when Crx is expressed. However, because the self-renewing potential of these cells was not examined and their multipotential properties were not tested, there is no evidence yet that true retinal stemlike cells exist in the iris epithelium.

6.3. RPE Transdifferentiation

Urodele amphibians (newts and salamanders) have a remarkable capacity for adult regeneration, and this is also true in the eye. Removal of the neural retina results in complete retinal regeneration from cells of the RPE. After removal of the retina, the cells of the RPE proliferate, dedifferentiate, and lose their pigmentation, then form a second layer of cells that differentiates into neural retina (37). Similar transdifferentiation of RPE to neural retina is possible in anuran amphibians until metamorphosis, but does not occur in adults. In other vertebrate species, such as chicks and rodents, RPE transdifferentiation is restricted to embryonic periods (up to E4.5 in chick and E13 in rats) and diminishes as development proceeds (41,42).

Fibroblast growth factors, such as FGF-1, FGF-2, or FGF-8, can stimulate transdifferentiation of embryonic RPE to neural retina in vitro in multiple species (43-47). In addition, RPE transdifferentiation can also be promoted by overexpression of cell intrinsic factors. For example, expression of the basic helix-loop-helix transcription factor NeuroD in chick RPE

cells isolated at E6 promoted neuronal differentiation and expression of the photoreceptor markers, including visinin (48,49). Similarly, adult human RPE cells expressing the oncogenic form of the ras signaling molecule adopted a neuronal phenotype and expressed several neuronal markers, including neurofilament and neuron-specific enolase, although retinal-specific markers were not examined (50). Thus, RPE cells in homeothermic vertebrates may retain some capacity to transdifferentiate beyond the embryonic period, but the full potential of these cells remains to be examined. Again, clues may lie with the intriguing species differences in the ability of RPE cells to transdifferentiate to neural retina.

7. conclusions and future prospects

Tissues derived from the optic vesicle have a unique ability to generate retinal neurons. This neurogenic capacity, however, diminishes in most vertebrate organisms as they reach adulthood, with the exception of teleost fishes and urodele amphibians. The continued neurogenic capacity found in these organisms is due in large part to the presence of active adult stem cells. Interestingly, recent studies demonstrated that most, if not all, vertebrate classes have cell populations in the eye that retain neurogenic potential and, if given the appropriate stimulus, can actively differentiate into retinal neurons. Whether these cell populations can be coaxed into becoming productive retinal stem cells and used for therapeutic purposes is still an open question.

To utilize a stem cell therapy to replace dying retinal neurons, many significant hurdles need to be overcome. With respect to the findings presented in this chapter, it has not been demonstrated that the newly identified neuro-genic cell populations produce fully differentiated and functional neurons. Thus, further studies are needed to determine how these cell populations can be manipulated into producing sufficient progeny without introducing deleterious changes in their genomes and then differentiating into functional neurons of the cell class desired (i.e., photoreceptors for retinitis pigmentosa and macular degeneration, ganglion cells for glaucoma). An important approach in this regard is to continue to identify and understand the regulatory pathways that promote retinal development and regeneration and develop strategies to activate these pathways in cell populations with neuro-genic potential.


1. Raymond, P. A., and Hitchcock, P. F. (1997). Retinal regeneration: common principles but a diversity of mechanisms. Adv Neurol 72, 171-184.

2. Reh, T. A., and Levine, E. M. (1998). Multipotential stem cells and progenitors in the vertebrate retina. J Neurobiol 36, 206-220.

3. Raymond, P. A., and Hitchcock, P. F. (2000). How the neural retina regenerates. Results Probl Cell Differ 31, 197-218.

4. Perron, M., and Harris, W. A. (2000). Retinal stem cells in vertebrates. Bioessays 22, 685-688.

5. Reh, T. A., and Fischer, A. J. (2001). Stem cells in the vertebrate retina. Brain Behav Evol 58, 296-305.

6. Cepko, C. L., Austin, C. P., Yang, X., Alexiades, M., and Ezzeddine, D. (1996). Cell fate determination in the vertebrate retina. Proc Natl Acad Sci U S A 93, 589-595.

7. Turner, D. L., and Cepko, C. L. (1987). A common progenitor for neurons and glia persists in rat retina late in development. Nature 328, 131-136.

8. Holt, C. E., Bertsch, T. W., Ellis, H. M., and Harris, W. A. (1988). Cellular determination in the Xenopus retina is independent of lineage and birth date. Neuron 1, 15-26.

9. Wetts, R., and Fraser, S. E. (1988). Multipotent precursors can give rise to all major cell types of the frog retina. Science 239, 1142-1145.

10. Fuhrmann, S., Chow, L., and Reh, T. A. (2000). Molecular control of cell diversification in the vertebrate retina. Results Probl Cell Differ 31, 69-91.

11. Livesey, F. J., and Cepko, C. L. (2001). Vertebrate neural cell-fate determination: lessons from the retina. Nat Rev Neurosci 2, 109-118.

12. Potten, C. S. and Loeffler, M. (1990). Stem cells: attributes, cycles, spirals, pitfalls, and uncertainties. Lessons for crypt. Development 110, 1001-1020.

13. Fischer, A. J., and Reh, T. A. (2001). Muller glia are a potential source of neural regeneration in the postnatal chicken retina. Nat Neurosci 4, 247-252.

14. Straznicky, K., and Gaze, R. (1971). The growth of the retina in Xenopus laevis: an autoradiographic study. J Embryol Exp Morphol 26, 67-79.

15. Hollyfield, J. (1971). Differential growth of the neural retina in Xenopus laevis larvae. Dev Biol 24, 264-286.

16. Johns, P. (1977). Growth of the adult goldfish eye. III. Sources of the new retinal cells. J Comp Neurol 176, 343-357.

17. Meyer, R. (1978). Evidence from thymidine labeling for continuing growth of retina and tectum in juvenile goldfish. Exp Neurol 59, 99-111.

18. Reh, T. A. (1987). Cell-specific regulation of neuronal production in the larval frog retina. J Neurosci 7, 3317-3324.

19. Wetts, R., Serbedzija, G. N., and Fraser, S. E. (1989). Cell lineage analysis reveals multipotent precursors in the ciliary margin of the frog retina. Dev Biol 136, 254-263.

20. Levine, E. M., Hitchcock, P. F., Glasgow, E., and Schechter, N. (1994). Restricted expression of a new paired-class homeobox gene in normal and regenerating adult goldfish retina. J Comp Neurol 348, 596-606.

21. Hitchcock, P. F., Macdonald, R. E., VanDeRyt, J. T., and Wilson, S. W. (1996). Antibodies against Pax6 immunostain amacrine and ganglion cells and neuronal progenitors, but not rod precursors, in the normal and regenerating retina of the goldfish. J Neurobiol 29, 399-413.

22. Levine, E. M., Passini, M., Hitchcock, P. F., Glasgow, E., and Schechter, N.

(1997). Vsx-1 and Vsx-2: two Chxl0-like homeobox genes expressed in overlapping domains in the adult goldfish retina. J Comp Neurol 387, 439-448.

23. Sullivan, S. A., Barthel, L. K., Largent, B. L., and Raymond, P. A. (1997). A goldfish Notch-3 homologue is expressed in neurogenic regions of embryonic, adult, and regenerating brain and retina. Dev Genet 20, 208-223.

24. Perron, M., Kanekar, S., Vetter, M. L., and Harris, W. A. (1998). The genetic sequence of retinal development in the ciliary margin of the Xenopus eye. Dev Biol 199, 185-200.

25. Harris, W. A., and Perron, M. (1998). Molecular recapitulation: the growth of the vertebrate retina. Int J Dev Biol 42, 299-304.

26. Fischer, A. J., and Reh, T. A. (2000). Identification of a proliferating marginal zone of retinal progenitors in postnatal chickens. Dev Biol 220, 197-210.

27. Fischer, A. J., Dierks, B. D., and Reh, T. A. (2002). Exogenous growth factors induce the production of ganglion cells at the retinal margin. Development 129, 2283-2291.

28. Johns, P. R., and Fernald, R. D. (1981). Genesis of rods in teleost fish retina. Nature 293, 141-142.

29. Johns, P. R. (1982). Formation of photoreceptors in larval and adult goldfish. J Neurosci 2, 178-198.

30. Julian, D., Ennis, K., and Korenbrot, J. I. (1998). Birth and fate of proliferative cells in the inner nuclear layer of the mature fish retina. J Comp Neurol 394, 271-282.

31. Otteson, D. C., D'Costa, A. R., and Hitchcock, P. F. (2001). Putative stem cells and the lineage of rod photoreceptors in the mature retina of the goldfish. Dev Biol 232, 62-76.

32. Alvarez-Buylla, A., and Garcia-Verdugo, J. M. (2002). Neurogenesis in adult subventricular zone. J Neurosci 22, 629-634.

33. Doetsch, F., Caille, I., Lim, D. A., Garcia-Verdugo, J. M., and Alvarez-Buylla, A. (1999). Subventricular zone astrocytes are neural stem cells in the adult mammalian brain. Cell 97, 703-716.

34. Johansson, C. B., Momma, S., Clarke, D. L., Risling, M., Lendahl, U., and Frisen, J. (1999). Identification of a neural stem cell in the adult mammalian central nervous system. Cell 96, 25-34.

35. Barres, B. A. (1999). A new role for glia: generation of neurons! Cell 97, 667-670.

36. Morshead, C. M., and van der Kooy, D. (2001). A new "spin" on neural stem cells? Curr Opin Neurobiol 11, 59-65.

37. Zhao, S., Rizzolo, L., and Barnstable, C. (1997). Differentiation and transdifferentiation of the retinal pigment epithelium. Int Rev Cytol 171, 225-266.

38. Ahmad, I., Tang, L., and Pham, H. (2000). Identification of neural progenitors in the adult mammalian eye. Biochem Biophys Res Commun 270, 517-521.

39. Tropepe, V., Coles, B. L., Chiasson, B. J., et al. (2000). Retinal stem cells in the adult mammalian eye. Science 287, 2032-2036.

40. Haruta, M., Kosaka, M., Kanegae, Y., et al. (2001). Induction of photore-ceptor-specific phenotypes in adult mammalian iris tissue. Nat Neurosci 4, 1163-1164.

41. Coulombre, J., and Coulombre, A. (1965). Regeneration of neural retina from the pigmented epithelium in the chick embryo. Dev Biol 12, 79-92.

42. Zhao, S., Thornquist, S., and Barnstable, C. (1995). In vitro transdifferentiation of embryonic rat pigment epithelium to neural retina. Brain Res 677, 300-310.

43. Park, C., and Hollenberg, M. (1989). Basic fibroblast growth factor induces retinal regeneration in vivo. Dev Biol 134, 201-205.

44. Pittack, C., Jones, M., and Reh, T. A. (1991). Basic fibroblast growth factor induces retinal pigment epithelium to generate neural retina in vitro. Development 113,577-588.

45. Guillemot, F., and Cepko, C. (1992). Retinal cell fate and ganglion cell differentiation are potentiated by acidic FGF in an in vivo assay of early retinal development. Development 114, 743-754.

46. Pittack, C., Grunwald, G. B., and Reh, T. A. (1997). Fibroblast growth factors are necessary for neural retina but not pigmented epithelium differentiation in chick embryos. Development 124, 805-816.

47. Vogel-Hopker, A., Momose, T., Rohrer, H., Yasuda, K., Ishihara, L., and Rapaport, D. H. (2000). Multiple functions of fibroblast growth factor-8 (FGF-8) in chick eye development. Mech Dev 94, 25-36.

48. Yan, R. T., and Wang, S. Z. (2000). Differential induction of gene expression by basic fibroblast growth factor and NeuroD in cultured retinal pigment epithelial cells. Vis Neurosci 17, 157-164.

49. Yan, R. T., and Wang, S. Z. (2000). Expression of an array of photoreceptor genes in chick embryonic retinal pigment epithelium cell cultures under the induction of neuroD. Neurosci Lett 280, 83-86.

50. Dutt, K., Scott, M., Sternberg, P. P., Linser, P. J., and Srinivasan, A. (1993). Transdifferentiation of adult human pigment epithelium into retinal cells by trans-fection with an activated H-ras proto-oncogene. DNA Cell Biol 12, 667-673.

0 0

Post a comment