Chapter 63

Sensory Epithelium of the Eye and Ear

Constance Cepko* and Donna M. Fekete***Department of Genetics, Howard Hughes Medical Institute, Harvard Medical School, Boston, MA, USA, **Department of Neurobiology, Harvard

Medical School, Boston, MA, USA

Ha

Co

Chapter Outline

Introduction 739

Introduction to Progenitor and Stem Cells in the Retina 739

The Optic Vesicle Generates Diverse Cell Types that

can Undergo Transdifferentiation 740

In Vivo Neurogenesis in the Posthatch Chicken 741

Growth of Retinal Neurospheres from the Ciliary Margin

of Mammals 742

Prospects for Stem Cell Therapy in the Retina 744

Development and Regeneration of Tissues Derived

from the Inner Ear 744

ndbook of Stem Cells, Two-Volume Set. DOI: http://dx.doi.org/10.1016/B978-0-12-385942-6.00

pyright � 2013 Elsevier Inc. All rights reserved.

In Vivo Neurogenesis in Postembryonic Animals 745

Proliferation in Normals (or After Growth

Factor Treatment) 745

Proliferation after Destruction of Cells 746

Transcription Factor Requirements 747

In Vitro Expansion of Otic Progenitors 747

Prospects for Therapy 748

Acknowledgments 748

References 748

INTRODUCTION

Humans rely heavily on both vision and hearing. Unfor-tunately, both deteriorate with age, partly because of thedeath of cells in the primary sensory organs, the eye andthe ear. In addition, the frequency of disease genes thataffect one or both of these modalities is relatively high(Retnet; Li and Friedman, 2002; Kramer and Hoefsloot,2002). Stem cells able to replace some of the dying cells,either in situ or through engraftment, have been a hope forsome time. In part, this is because there are few effectivetherapies for diseases of these tissues. Work aimed atidentifying retinal and otic stem cells has been undertakenwith more energy in the last several years because of theexciting findings for stem cells elsewhere in the body.Here, we review these findings in the context of normaldevelopment.

INTRODUCTION TO PROGENITORAND STEM CELLS IN THE RETINA

The retina has served as a model of central nervoussystem (CNS) anatomy, physiology, and development

(Dowling, 1987; Rodieck, 1998). Most studies aimed atunderstanding its development have concerned theproduction of the retinal neurons and glia from retinalprogenitor cells. These cells were originally shown bylineage analysis to be multipotent throughout develop-ment, capable of generating both neurons and glia, evenin a single, terminal cell division (reviewed by Cepkoet al., 1996). Retinal progenitor cells do not appear to betotipotent except for the earliest progenitor cells, whenclones can comprise all retinal cell types. Moreover,retinal progenitor cells do not appear to be able toproliferate extensively in vivo (Turner et al., 1990) orfollowing explantation and exposure to different cultureconditions (e.g., see Lillien and Cepko, 1992 and Jensenand Raff, 1997). More recent studies have been aimed atfinding retinal stem cells. These studies have been con-ducted along the two lines established in the search forstem cells elsewhere in the CNS. One approach has beento search for mitotic cells capable of generating retinalneurons in the adult in vivo. The other approach hasbeen to culture cells in growth factors. Both types ofexperiments have begun to yield promising answers, butmuch more needs to be done.

063-9 739

740 VOLUME | 2 Adult and Fetal Stem Cells

THE OPTIC VESICLE GENERATES DIVERSECELL TYPES THAT CAN UNDERGOTRANSDIFFERENTIATION

To appreciate some of the intriguing observations concern-ing retinal stem cells, a review of the tissues that derive fromthe optic vesicle is needed (Mann, 1950; Barishak, 2001).The optic vesicle is an evagination of the neural tube wherethe diencephalon and telencephalon meet. The vesicle at firstprotrudes as a simple evagination when the neural tubeforms. Soon thereafter, the vesicle undergoes an invaginationto form a two-layered optic cup. The outer cup will forma non-neural structure, the retinal pigmented epithelium(RPE), as well as with other support structures of the eye(Figure 63.1). The RPE is a single layer of epithelial cellsheavily pigmented to capture stray light that passes throughthe retina. It performs several support functions, includingsuch highly specialized functions as the isomerization oftrans to cis retinal to allow the photopigments, the opsins, tocontinue to capture light. The RPE expresses many specificgene products. A transcriptome analysis conducted usingserial analysis of gene expression (SAGE) showed that 40%of the RPE SAGE tags did not have a corresponding cDNAin GenBank (Sharon et al., 2002). This is a much higher rateof unknowns than seen in other tissues (e.g., the humanretina).

The inner wall of the optic cup forms the neural retina(Figure 63.1). The primary sensory cells are the photore-ceptors (PRs), which comprise two types: the rods and the

Lens

IrisCiliaryBody

Retina

RPE

FIGURE 63.1 The eye is a complex tissue, developing from cells

originating from the neural tube, the neural crest, the surface ectoderm,

and the mesoderm. The retina is the neurosensory tissue that originates

from the inner layer of the optic cup, and the RPE originates from the outer

optic cup layer. Both the inner and outer optic cup layers contribute to the

formation of the ciliary body and iris. The ciliary body also comprises cells

from the neural crest, which form the ciliary muscle. The iris muscles

derive from the outer layer of the optic cup. Stem cells have been isolated

from the ciliary body (Ahmad et al., 2000; Tropepe et al., 2000) and iris

(Haruta et al., 2001), and Muller glia in the retina have been found to

divide and generate neurons in the early posthatch chick (Fischer and Reh,

2001).

cones. The rods are active under dim light, and the conesare active under daylight conditions. In addition, there areseveral types of interneurons, horizontal cells, amacrinecells, and bipolar cells and an output neuron, the retinalganglion cell. The retina also has one glial cell that spansthe retinal layers, the Muller glial cell. During the earlyphase of retinal neurogenesis, retinal progenitor cellsproduce the various retinal neurons in a conserved fashion,typically beginning with production of ganglion cells andfinishing with production of rod PRs, bipolar interneurons,and Muller glia (reviewed by Altshuler et al., 1991). Theproduction of these cells begins in the center of the retinaand proceeds to the retinal periphery, or margin. Inamphibians and teleost fish, there is continual growth at themargin throughout the life of the animal in a region termedthe ciliary marginal zone, or CMZ (Hollyfield, 1968;Straznicky and Gaze, 1971; Johns, 1977). In addition, infish, there is a late wave of production of rod PRs as theretina expands (Johns and Fernald, 1981).

The developmental sequence at the periphery of theretina is complex and not at all understood at a molecularlevel, although it is an important region for stem cells. Themargin is the fold that develops following invagination ofthe primary optic vesicle. Following the initial formation ofthe morphologically simple folds at the periphery, where thepresumptive RPE and retina meet, some rather unusualmorphogenetic events, including transdifferentiation, takeplace to form several anterior support structures for the eye.The ciliary body, with the associated pars plicata and parsplana, as well as the iris, form from this area (Figure 63.1).The pars plana and plicata each comprise two epitheliallayers, one pigmented and one unpigmented. They are thesite of attachment of the zonules, or suspensory ligaments ofthe lens. The unpigmented epithelial layer of the pars plicataand plana is continuous with the retina, and the pigmentedlayer is continuous with the RPE. The tight apposition ofthese two epithelial layers allows regulation of secretionfrom the ciliary body, as it is highly vascularized witha rather leaky type of blood vessel. Beyond secretion ofaqueous (through the pars plicata) and vitreous (throughthe pars plana) humor, the ciliary body also controls theshape of the lens. Neural-crest-derived muscles form withinthe ciliary body and contract and relax the ligamentssurrounding the lens during lens accommodation.

The iris is the shutter that opens and closes to allowmore or less light to penetrate the eye. It includesa pigmented epithelial layer derived from the margin ofthe optic cup that is continuous with the RPE. It also hasan initially unpigmented epithelial layer, the inner orposterior layer of epithelium, which is continuous with theretina. This layer gradually becomes pigmented, however,and additional pigmentation of the iris is contributed byneural-crest-derived melanocytes. Remarkably, the pupilis opened and closed by muscles that derive from the

741Chapter | 63 Sensory Epithelium of the Eye and Ear

margin of the optic cup, the only ectodermally derivedmuscles in the body. This occurs because of trans-differentiation as initially pigmented cells separate fromthe epithelial sheet, proliferate, and form muscles. Thus,the retinal margin develops to serve several functions, andmost of these diverse cell types derive from both the outerand the inner walls of the early optic cup.

Classical embryological experiments with birds, fish,amphibians, and mammals revealed a great deal of plas-ticity among the ocular tissues. For example, extirpation ofmost e but importantly, not all e of the retina leads totransdifferentiation of the RPE into the retina (Coulombreand Coulombre, 1965). This capacity exists only untilembryonic day (E) 4 in chickens (Coulombre andCoulombre, 1965) and E14 in mice (Detwiler and VanDyke, 1954; Zhao et al., 1997). In urodeles, it can occurthroughout life (Stone, 1950; Sologub, 1968; Okada, 1980).In chicks, it was found to be induced by fibroblast growthfactor (FGF) in RPE cultures (Park and Hollenberg, 1989;Guillemot and Cepko, 1992). In vivo, it is not clear if FGFis involved in the initial distinction between the retinaand RPE, but delivery of FGF8 in vivo can trigger trans-differentiation (Vogel-Hopker et al., 2000). Wolffianregeneration in newts is a remarkable process by which thedorsal iris can regenerate a lens (Reyer, 1954), originallynot derived from the optic vesicle but from the surfaceectoderm. This type of regeneration has not been seenin birds or mammals. All of these examples reveal thatend-stage differentiated cells are not necessarily committedor irreversibly differentiated. Perhaps most remarkably, asexplained later in this chapter, the pigmented cells derivedfrom this area of the optic cup display broad developmentalpotential in adult mammals in that they are the source ofretinal stem cells.

IN VIVO NEUROGENESIS IN THEPOSTHATCH CHICKEN

A search for stem cells in the developed retina wasconducted by Fischer and Reh in the chicken (Fischer andReh, 2000, 2001). Retinal neurogenesis in most of the chickretina is complete by E12 (Prada et al., 1991). Fischer andReh (2000) examined the posthatch chick (i.e., >E21) forthe incorporation of bromodeoxyuridine (BrdU). Theyfound that two areas could be labeled. In a normal retinawithout injury, the P7 retina was labeled in the ciliarymargin, reminiscent of the aforementioned findings foramphibians and fish (Fischer and Reh, 2000). These cellswere followed using their BrdU label and were found toincorporate into the inner nuclear layer (INL), generatingbipolar and amacrine neurons. No cells were found in theouter nuclear layer (ONL), the layer containing PRs.Antigens consistent with the INL fates were also observed.

The newly generated cells appeared progressively morecentrally as harvests were made later and later. Thesefindings suggest that the CMZ cells were generating moreretinal neurons to accommodate additional growth of theeye, previously thought to occur only through an expansionof the volume of the vitreous cavity, and stretching of theretinal tissue. However, this does not occur throughout thelife of the chicken as it does in amphibians and fish, sincethe growth stops a few weeks posthatch. The number ofmitotic cells within the CMZ was not increased followinginjection of a toxin, N-methyleD-aspartate (NMDA),unlike the response seen in the Xenopus eye (Reh, 1987).However, injection of 100-ng doses of epidermal growthfactor (EGF), insulin, or insulin-like growth factor 1 (IGF-1), but not FGF, did increase the mitotic activity in thisarea. In addition, if insulin was applied with FGF2, cellswith gene expression profiles and processes consistent withthe ganglion cell fate were observed (Fischer et al., 2002).

Another site of BrdU incorporation in the posthatchchick could be identified following application of the toxinNMDA, which primarily targets amacrine cells. If BrdUwas applied 2 days after injection of NMDA at P7, incor-poration of BrdU into Muller glial cells in the central retinawas observed. Labeling with BrdU at 1 day or 3 daysfollowing toxin administration led to few BrdU-labeledcells. It thus appears that a process triggered by toxinadministration requires approximately 2 days to stimulateone or two rounds of cell division. This response is alsodevelopmentally limited, as few BrdU-labeled cells wereobserved after P14, and the response was lost in a central-to-peripheral manner, similar to the initial wave of neuro-genesis in the retina. Muller glial cells do possess the abilityto undergo reactive gliosis, a phenomenon associated withvarious types of retinal damage in adult mammals and birds(MacLaren, 1996). Reactive gliosis occurs in astrocytesthroughout the CNS and is characterized by: limited celldivision; expression of intermediate filament proteins, suchas vimentin and glial fibrillary acidic protein (GFAP); andincreased process outgrowth (Hatten et al., 1991). It has notbeen established that this type of cell division leads toproduction of neurons elsewhere in the CNS or in the retinaof adult birds.

The BrdU-labeled Muller glial cells induced by toxintreatment were found in the ONL and INL, whereas Mullerglial nuclei are typically only found in the INL. Some ofthese BrdU-labeled cells coexpressed two markers ofretinal progenitor cells, Chx10 and Pax6, and someexpressed a bHLH gene, CASH-1, a marker of early retinalprogenitor cells. A small percentage (<10%) of the BrdUþ

cells subsequently were found to have neuronalmorphology and to express markers of amacrine cells andbipolar cells. Transiently, however, many cells expresseda neurofilament (NF) marker, normally expressed onhorizontal and ganglion cells in the retina. This appeared to

742 VOLUME | 2 Adult and Fetal Stem Cells

be transient as the number of NFþ cells decreased, andmarkers and morphology consistent with mature ganglionand horizontal cells were not seen. Researchers did notfind markers of PR cells. Many of the BrdU-labeled cellspersisted in what appeared to be an arrested state for at least12 days. They explored the possibility that the in vivoenvironment was limited for production of PR cells byculturing the toxin-treated retinas, but they were similarlyunable to observe the genesis of PR cells, though they didfind that toxin-treated retinas proliferated in vitromore thanthe untreated controls.

The division of Muller glia, following by genesis ofneurons in the chick, can also be stimulated by growthfactors in the absence of added toxins (Fischer et al., 2002).Application of both insulin and FGF2 by intraocularinjection, starting at P7 and continuing for 3 days, led to theproduction of many mitotic Muller glial cells. Fourteendays after the last injection, some BrdUþ cells showedmarkers of amacrine and perhaps ganglion cells, and othersshowed markers of Muller glia. Similar to the findingsfollowing NMDA injections, no markers of PR cells wereobserved. To explore whether the type of toxin, and thus thetarget cell killed by the toxin, combined with growth factorinjection might give a more specific replacement of thetargeted cells, researchers injected several types of toxinswith insulin and FGF2 (Fischer and Reh, 2002).When ganglion cells were targeted by the toxins, kainicacid, or colchicine, more cells with ganglion cell markerswere observed. These studies lead to the hope that thespecific cells that die in various retinal diseases, such asganglion cells in glaucoma, might be effectively replacedfollowing stimulation of stem cells with the right cocktailof factors.

Ahmad et al. (2000) as well as Kubota et al. (2002) havesearched for proliferation in the uninjured retina ofmammals. Ahmad et al. injected 4-week-old rats intraper-itoneally with BrdU for 5 days. The only incorporation thatthey report was in the ciliary margin e that is, no labelingof Muller glial cells was observed. Kubota et al. (2002)examined opossum and mouse (as well as quail) and foundno incorporation in centrally located Muller glia. Kubotaet al. did find some incorporation of BrdU in the ciliarymargin of quail, though less than in the chicken, and a fewlabeled cells in the ciliary margin of the opossum, but theyfound no labeled cells in the ciliary margin in the mouse.

Radial glial cells, astrocytes, and Muller glia have beenshown to share some antigens with progenitor cells(e.g., see Linser et al., 1997; Doetsch et al., 1999; andLendahl et al., 1990). This resemblance is more extensivethat previously appreciated. We systematically analyzedgene expression in the developing and mature murine retina(Blackshaw et al., 2004). We found that there were 85 genespreferentially expressed in Muller glia in the mature retina.Of these genes, the majority were also found in retinal

progenitor cells. Some of these genes, such as cyclin D3,undoubtedly reflect that Muller cells retain the ability todivide but other genes are more enigmatic. Nonetheless, thestudies from Fisher and Reh mentioned previously and theSAGE data argue that Muller glia should be exploredfurther as a source of cells that might replace dyingneurons. This notion is in keeping with the idea that radialglial cells elsewhere in the developing CNS (Kriegstein andGotz, 2003), as well as astrocytes in the mature forebrain(Reynolds and Weiss, 1992), can serve as neuronalprogenitor cells and stem cells, respectively.

GROWTH OF RETINAL NEUROSPHERESFROM THE CILIARY MARGIN OFMAMMALS

Reynolds and Weiss (1992) demonstrated that cultures ofCNS tissue in the presence of EGF, FGF, or both would leadto balls of cells with indefinite proliferation capacity, orneurospheres. Neurospheres were subsequently shown toproduce neurons, astrocytes, and oligodendrocytes andwere thus identified as originating from neural stem cells.Several groups have applied these protocols to the retina.Cultures from ocular tissues of mouse, rat, human, and cowhave been made using FGF and EGF (Tropepe et al., 2000;Haruta et al., 2001). Retinal neurospheres have beenrecovered and appear to have cells with indefinite prolif-eration potential, multipotency, and possibly totipotency.

Tropepe et al. (2000) reported that the ciliary body fromadult mice had the most enriched source of retinal stemcells. They examined E14 RPE and retina and adult retina,RPE, iris, and ciliary body for production of neurospheresin the presence of FGF, EGF, or both. They were unable torecover any neurospheres with the proliferative capacity ofstem cells from the embryonic or adult retina or from theadult iris, RPE, and ciliary muscle. A few neurosphereswere recovered from the embryonic RPE, which includedthe peripheral margin, the precursor to the ciliary body.In the adult, neurospheres were only recovered from thepigmented cells of the ciliary body (Figure 63.1), termedthe pigmented ciliary margin (PCM). Although there wassome recovery from the E14 RPE, the number per eyeincreased 10-fold in the adult PCM compared to the entireRPE of the E14 retina. This curious finding suggests thatthese cells are formed at the end of development and not thebeginning, the opposite of what one might have predicted.Alternatively, during the maturation of the PCM, there is anexpansion of a few early stem cells. Tropepe et al. foundthat the stem cells were either rare or hard to culture, asonly 0.6% of the adult PCM-plated cells would producea neurosphere in the presence of FGF2. They did notrequire exogenous FGF, as they could arise, albeit ata reduced frequency, in its absence. This was presumably

743Chapter | 63 Sensory Epithelium of the Eye and Ear

because of endogenous FGF, as addition of anti-FGFantibody reduced the formation of neurospheres.

As the PCM neurospheres originated with thepigmented cells of the ciliary body, it was of interest todetermine if pigment was necessary for their ability to bestem cells. This was examined in albinos, where they wereisolated at a comparable frequency to that of pigmentedanimals (Tropepe et al., 2000). The RPE and ciliarybody normally do not express Chx10. However, uponthe genesis of neurospheres, cells began to express thisretinal marker. If cultured under conditions that favordifferentiation, the cells turned on markers of PRs, bipolarcells, and glia. However, no markers of amacrine andhorizontal cell interneurons, or of ganglion cells, wereseen. It is significant that rhodopsin, a definitive markerof a PR cell, can be expressed. Rhodopsin is not expressedby other CNS neurospheres or neural cell linesderived from other CNS locations (Takahashi et al., 1998).Retinal progenitor cells normally do not make oligoden-drocytes, and the oligodendrocyte marker, O4, was notobserved.

There is a naturally occurring null allele of the paired-type homeobox gene, Chx10, in mice (orj). This mutant hasa retina and RPE approximately 10-fold smaller than wild-type mice, with an expanded ciliary margin (Burmeisteret al., 1996). When cultures were made from the ciliarymargin of orj, approximately fivefold more neurosphereswere recovered (Tropepe et al., 2000). These spheres wereapproximately one third of the size of the wild-typespheres, in keeping with the finding that Chx10 is requiredfor the full proliferation of retinal progenitor cells. Thus,Chx10must be involved with the allocation or regulation ofthe number of stem cells in the PCM.

Ahmad et al. (2000) also report the isolation of retinalneurospheres from the ciliary body of rats, and Tropepeet al. (2000) reported such results from postmortemhuman and bovine eyes. The rat neurospheres weredependent upon FGF2 and could not be recovered fromthe retina, the RPE, or the nonpigmented portion of theciliary epithelium. Haruta et al. (2001) also isolated retinalstem cells from the rat eye, but they used iris tissue ratherthan ciliary margin tissue. As shown in Figure 63.1, theiris and ciliary body are adjacent to each other at themargin of the eye. The iris cells were cultured as anexplant in the presence of FGF, rather than as dissociatedcells. Tropepe et al. (2000) had reported previously that,unlike the ciliary body cells, dissociated cells from the irisdid not generate neurospheres. Cells that migrated out ofthe iris explants were able to express some neural markers,such as NF200, but not PR-specific markers. However, ifthe cells were transduced with crx, a homeobox geneimportant in PR differentiation (Chen et al., 1997; Fur-ukawa et al., 1997), approximately 10% of the cellsexpressed rhodopsin and recoverin, two markers of PR

cells. The ciliary-body-derived neurospheres can generatecells bearing the same retinal markers, includingrhodopsin, without transduction of crx. The differencemay be caused by culture conditions. The ciliary-body-derived cells can be grown as spheres, and this environ-ment may support PR development without a need for crxtransduction. When Haruta et al. (2001) cultured ciliary-body-derived cells as monolayers, as they culturediris-derived cells, the ciliary-body-derived cells did notexpress PR markers.

Haruta et al. (2001) point out that tissue from the iriscan be readily obtained for autologous grafts. It is far moredifficult to obtain tissue from the ciliary body, with anaccompanying risk of damage to the ciliary body. However,as Tropepe et al. (2000) reported that postmortem humanand bovine PCM could produce spheres at low frequency, itis possible that humans could be used as a source of donorcells. This would not be as immunologically compatible asan autologous graft, but it nonetheless might suffice, as it isstill not clear how well tolerated retinal grafts might be(Streilein et al., 2002).

That the pigmented ciliary body and iris cells are thesource of retinal stem cells is rather surprising. It wasexpected that the periphery of the eye might be the areawhere stem cells would reside, as this is the location ofstem cells in amphibians and fish. In the CMZ ofamphibians and fish, the nonpigmented cells contiguouswith the retinal epithelium, located nearest to the retina,make more retinal cells. However, stem cells have notbeen recovered from this area in mammals. An alternativeprediction might have been that retinal stem cells wouldbe the RPE cells. As explained previously, the RPE isquite plastic early in development in mammals and chicksand throughout life in urodeles, maintaining the ability tomake retinal cells in response to certain conditions.However, the adult mammalian RPE has not been shownto generate neurospheres when cultured under the condi-tions described previously. Moreover, even the embryonicRPE does not supply many neurospheres in the neuro-sphere culture conditions. In contrast to the precedingpredictions, the most robust source of retinal stem cells inthe adult mammal is the ciliary body pigmented cells.These cells are immediately adjacent to, and contiguouswith, the RPE, but they are not RPE cells, at least in termsof function, as described previously. We do not havemarkers that would help to further define them. Theyderive from the outer walls of the optic cup, and the innerwalls normally produce the retina in normal development.It should be noted, however, that whether the irisstem cells are derived from the inner or outer walls of theoptic cup is not clear at this time. Although the iris stemcells are pigmented, both the inner and outer walls ofthe optic cup normally develop pigmentation in theiris. Both iris- and ciliary-body-derived stem cells are

744 VOLUME | 2 Adult and Fetal Stem Cells

pigmented, which may provide a useful marker for theirprospective isolation.

PROSPECTS FOR STEM CELL THERAPYIN THE RETINA

A large number of the diseases of the retina are caused bydegeneration of PR cells. Approximately 40% of the genesidentified as human disease genes that lead to blindness arerod specific (Retnet). Many of these diseases nonethelesslead to the loss of cone PR cells. This nonautonomous deathof cones is the reason for loss of daylight, high-acuityvision. Thus, replacement of dying rods, or the retardationof the death of rods, might prevent or slow the death ofcones. Replacement of dying cones themselves is anotherpotential therapeutic approach. This is particularly appro-priate when the etiology of the disease is not clear, as is thecase in the most prevalent disease, age-related maculardegeneration. The source of either rod or cone PR cellscould be the endogenous stem cells themselves. The bestscenario would be the stimulation of the division of Mullerglia, which are distributed through the retina, followed bythe induction of PR differentiation. Unfortunately, as notedpreviously, Muller glia have not been observed to generatePR cells in the chick or any mammal that has been inves-tigated. Nonetheless, future studies might lead to a manip-ulation that would stimulate PR production by Muller glia.A second source of PR cells might be the endogenous stemcells in the ciliary body or iris, left in situ. Although thesecells can produce PR cells when cultured, they have notbeen shown to generate PR cells in situ. In addition, unlessthe cells could be made to migrate and cover the centralretina, where most of our high-acuity vision occurs, theywould not lead to retention or recovery of high-acuityvision. Still, if they could lead to retention of peripheralvision, some therapeutic benefit would be realized. Finally,engraftment of stem cells or of PR cells generated in vitroby stem cells can be attempted. The problems of graftrejection, if not an autologous graft, would have to beconfronted, but they might not be as difficult to overcomeas engraftment to sites in the periphery (Streilein et al.,2002). Furthermore, iris-derived cells might be used as anautologous graft. Preliminary data from Tropepe et al. ofengraftment of the neurospheres derived from the PCM arepromising. Injection of such cells into the vitreous body ofpostnatal day 0 rats led to the formation of many PR cells,expressing rhodopsin, in the ONL (Van der Kooy, personalcommunication). If such cells could be formed ina diseased retina, then two possible benefits might berealized. One would be to simply prevent further degener-ation of endogenous PR cells, which, as mentioned previ-ously, can die by nonautonomous processes. The secondbenefit might be that the engrafted PR cells synapse with

second-order neurons and provide vision themselves.To date, this has not been achieved (Radner et al., 2002).The concern here is that the site of engraftment might notsupport synaptogenesise particularly at the advanced stagewhen much of the retina has degenerated, likely the stagewhen such therapies would be attempted. Nonetheless,such strategies are worth pursuing, particularly nowthat there are stem cells that can be manipulated togenerate retinal cells; our understanding of the processes ofthe development of retinal cells has similarly beenadvancing.

DEVELOPMENT AND REGENERATIONOF TISSUES DERIVED FROM THEINNER EAR

The entire vertebrate inner ear derives from the oticplacode, a thickening of the dorsolateral surface ectodermimmediately lateral to the hindbrain. Like the lens andolfactory placodes, the otic placode invaginates andpinches off to form a single-layered ball of cells, now calledthe otic vesicle. From this simple epithelium, a large varietyof tissues and cells types arise. The otic ectoderm isneurogenic for the first-order neurons of the eighth cranialganglion, the statoacoustic ganglion. The ganglion neuro-blasts are the earliest recognizable cell type; they delami-nate from the otic ectoderm at the otic cup stage evenbefore it completes vesicle formation. The otic vesicle isalso sensorigenic, generating 6e8 different sensorypatches. Inner ear sensory organs subserve hearing andbalance and are differentiated according to their function.There are three major classes of sensory organs: macula,crista, and acoustic. Acoustic organs vary substantially instructure and sensitivity across the vertebrates, reachingtheir highest complexity and frequency selectivity in themammalian organ of Corti. In all inner ear sensory organs,the mechanosensory hair cells are interspersed amonga field of supporting cells essential for hair cell survival andfunction. Finally, beyond the sensory patches, several typesof nonsensory tissues derive from the otic epithelium.The most highly differentiated is the tissue that secretes theextracellular fluid, called endolymph, which bathes theapical surfaces of all otic epithelial cells. Endolymphcontains an unusually high concentration of potassium ions.The tissue responsible for endolymph production isanatomically complex, highly vascularized, and endowedwith many ion pumps and channels. Other nonsensoryepithelia flank the sensory organs and may contribute, withthe supporting cells, to the secretion of the specializedextracellular matrices perched above the hair cells toenhance their mechanosensitivity.

Lineage studies have yet to reveal all possiblerelationships among the constellation of inner ear cells, so

745Chapter | 63 Sensory Epithelium of the Eye and Ear

we do not yet know if individual otic placode cells aretruly pluripotent for all inner ear cell types. We are confi-dent that mechanoreceptors and their supporting cells sharea common progenitor in the bird and zebra fish inner ear(Fekete et al., 1998; Haddon et al., 1998; Lang and Fekete,2001) and in the regenerating salamander lateral line(Balak et al., 1990). There is also evidence that sensory andnonsensory cells can be clonally related in the chicken ear(Fekete et al., 1998; Lang and Fekete, 2001). Preliminarystudies in mice include a single clone with members in boththe sensory saccule and nonsensory utricle (Brigande andFekete, unpublished observations). Furthermore, oticneurons and sensory cells can be related in the bird (Satohand Fekete, 2003). To date, there is no direct evidence forthe existence of a true otic stem cell e that is, one thatdivides asymmetrically to replicate itself while generatinga daughter with an alternative cell fate or fates. Nonethe-less, lineage studies indicate that multipotent progenitorcells constitute a normal feature of inner ear development,leaving open the possibility that similar cells may lurkin mature ears, where they might be poised to expandwith appropriate signals or culture conditions.

A variety of growth factors and growth factor receptorshave been associated with developing inner ears (reviewedby Oesterle and Hume, 1999). However, it is importantto distinguish between growth factors that may regulate cellproliferation and those that may influence cell fatespecification in other ways, such as the role of FGFs in oticinduction (Noramly and Grainger, 2002). Furthermore,many of the growth factor assays are performed in vitro,with the inherent risk that cells and tissues can change theirgrowth factor responsiveness depending upon cultureconditions (described in Oesterle and Hume, 1999). Forexample, culturing of neonatal sensory epithelia leads toupregulation of FGFR1/2 and IGF-1R in macular sup-porting cells (Zheng et al., 1997) and downregulation ofEFGR in the organ of Corti (Zine and de Ribaupierre,1999). This caveat notwithstanding, members of severalgrowth factor families can enhance cell proliferation ofdeveloping otocyst or neonatal inner ear tissues, eitheralone or in combinations, when presented in culture. Theseinclude bombesin, EGF, FGFs, GGF2, heregulin, insulin,IGFs, PDGF, and TGFa (referenced by Oesterle and Hume,1999; see also Zheng et al., 1999a).

IN VIVO NEUROGENESIS INPOSTEMBRYONIC ANIMALS

Proliferation in Normals (or After GrowthFactor Treatment)

The vestibular maculae are involved in sensing gravity, andin fishes and amphibians these organs continue to increasein size throughout life (reviewed by Corwin and Warchol,

1991). Here, we focus on the sensory organs of warm-blooded vertebrates, where there is a marked contrastbetween birds and mammals in the timing of inner earorganogenesis (reviewed by Corwin and Oberholtzer,1997). In both classes, sensory organs stop generating newcells approximately midway through embryogenesis(Ruben, 1967; Katayama and Corwin, 1989). A notableexception is the vestibular maculae of birds, where there isongoing addition and death of cells well beyond hatching.Cell counts or BrdU labeling have led to estimates that haircells turn over with half-lives of 20 days (Kil et al., 1997),30 days (Goodyear et al., 1999), or 52 days (Stone et al.,1999) in the chicken utricular macula. Within 2 weeks ofhatching, nearly 500 cells in the saccular macula and 1400cells in the utricular macula may be added per day(Kil et al., 1997) with the steady-state addition of 850 haircells per utricle per day reached 60 days after hatching(Goodyear et al., 1999). However, these numbers areapparently achieved in the absence of an amplifyingprogenitor. Rather, most progenitors divide once to producea sibling pair consisting of one hair cell and one nonhaircell (presumed to be the supporting cell). BrdUþ clustersexceeding three cells are rare 2e4 months after labeling(Stone et al., 1999). Thus, if there is a self-renewing stemcell pool in the mature avian macula, it consists of cells thatdivide on an extremely slow timescale. Interestingly, theneuronal colony-forming cells of the olfactory epithelium,thought to be the true stem cell of this sensory organ, arealso rare (1 in 3600 purified progenitors) and divide ata very slow rate (reviewed by Calof et al., 2002). Ongoingreceptor cell turnover (and regeneration) in the olfactoryepithelium utilizes a transient amplifying progenitor poolthat in turn generates a population of immediate neuronalprecursors; the latter will divide symmetrically to makedifferentiated olfactory receptor cells.

In contrast to lower vertebrates, in mammals thevestibular macula is quiescent from birth. So, too, are theauditory organs of both birds and mammals. For example,BrdU injections failed to label cells in the adult mouseorgan of Corti (Lowenheim et al., 1999). However, inmice lacking the cyclin-dependent kinase inhibitors p27kip1

or p19Ink4d, cell turnover continues in postnatal animals(Chen and Segil, 1999; Lowenheim et al., 1999; Chen et al.,2003). Dividing cells are observed several weeks afterbirth among hair cell (p19Ink4d�/�) or supporting cell(p27kip1�/�) layers. Prolonged mitosis in the p27kip1-null isaccompanied by the differentiation of supernumerary haircells and supporting cells. In both mutants, many hair cellseventually undergo apoptosis, leading to hearing loss.Nonetheless, these data suggest that the differentiatedorgan of Corti, which has never been shown to regeneratenaturally, has the potential to harbor cells that can divideand differentiate under appropriate circumstances. In thiscontext, it is intriguing that a subset of cells in both auditory

FIGURE 63.2 Nestin-GFP in organ of Corti

(left) and utricular macula (right) of P5 mouse.

In the cochlea, GFP is seen in border cells (bc)

surrounding inner hair cells (ihc), Deiters cells

(Dc) surrounding outer hair cells (outer hair cell

rows 1, 2, and 3 are indicated), and greater

epithelial ridge cells (ger). In the utricular macula,

GFP is seen in stromal cells (st), supporting cells

(sc) and hair cells (hc). Figure courtesy of Ivan

Lopez (UCLA).

746 VOLUME | 2 Adult and Fetal Stem Cells

and vestibular organs of the neonatal mouse express theneural stem cell marker, nestin (Figure 63.2). Nestin israpidly downregulated about one week after birth, althoughit persists in nonsensory cochlear cells to day 15 (Mal-grange et al., 2002).

Several growth factors and cytokines can affect cellproliferation in undamaged, mature inner ear sensoryorgans or within sheets of sensory epithelial supportingcells (reviewed by Oesterle and Hume, 1999). Insulin,IGF-1, or IL-1b enhanced proliferation of normal chickenutricular cells in vitro, and FGF-2 reduced proliferation.Rodent utricular cells proliferated in response to GGF2,EGFþinsulin, neu differentiation factor, TGFa, orTGFaþinsulin (see references by Oesterle and Hume,1999 and Zheng et al., 1999a) but lost response to here-gulin by adulthood (Hume et al., 2003).

Proliferation after Destruction of Cells

The strongest evidence that progenitor cells reside inquiescent sensory organs of warm-blooded vertebratescomes from regeneration studies. Beginning with ground-breaking work in the late 1980s, many studies have shown

that damaged sensory organs will regenerate new hair cellsin chickens, largely through a proliferative mechanism(reviewed by Cotanche and Lee, 1994; Corwin andOberholtzer, 1997; and Stone and Rubel, 2000). Thus,despite the absence of ongoing turnover, supporting cells ofthe auditory organ (the basilar papilla) mount a vigorousmitotic response to damaging conditions: Up to 15% ofthem enter the cell cycle to generate both hair cells andnew supporting cells. Whether supporting cells coexistalongside self-renewing stem cells remains an open ques-tion. Data suggest that only 1e4% of cycling cells in theregenerating basilar papilla will divide more than oncewithin a 3-day window after ototoxic drug treatment(Stone et al., 1999). Even among this pool, ongoingproliferation appears to be extremely modest, althoughlineage analysis has not yet been performed to provide anunequivocal measure of clonal expansion. Like the basilarpapilla, chicken macular cells also divide and differentiatein response to hair cell loss. The proliferation index of thedrug-damaged macula rises in the presence of TGFa orTNFa (Warchol, 1999).

Once again, we contrast mammals to lower vertebrates:The mammalian macula has only a weak proliferative

747Chapter | 63 Sensory Epithelium of the Eye and Ear

response to hair cell loss, with scant evidence that hair cellscan be regenerated through a cycling intermediate. Instead,the maculae primarily restore their hair cells through directtransdifferentiation of supporting cells (reviewed byCorwin and Oberholtzer, 1997) or through self-repair ofsublethally damaged hair cells (Zheng et al., 1999b). Thelimited proliferation that accompanies recovery may serveto replace transdifferentiated supporting cells rather thanhair cells (see the description by Forge and Nevill, 1998).Hair cell recovery is promoted by the addition of TGFa,IGF-1, retinoic acid, and brain-derived neurotrophicfactor in vivo (Kopke et al., 2001) or in vitro (Oesterle andHume, 1999). The weak proliferative response of thecultured, drug-damaged macula averaged 26 BrdUþ cellsper sensory organ. This proliferation was enhanced 10-foldby addition of heregulin (a member of the neuregulinfamily) and to a lesser extent by EGF or TGFa alone orwith insulin (Zheng et al., 1999a). Neither hair cells norsupporting cells of the mammalian organ of Corti respon-ded to heregulin, although more remote nonsensoryepithelial cells of the cochlea did.

Transcription Factor Requirements

Although many cell types are generated from oticepithelium, relatively few transcription factors have beendefinitively associated with cell fate specification in theear (reviewed by Fekete and Wu, 2002; and Rubel andFritzsch, 2002). NeuroD, Neurogenin-1, and Eya1 areessential for otic ganglion cell fate. Brn3a/Brn3.0 isneeded for ganglion cell survival and differentiation.Math-1 is required for hair cell development and survival,and Pou4f3/Brn3c/Brn3.1 is needed for subsequent haircell differentiation. Ectopic delivery of Math1 leads toectopic production of partially differentiated hair cellsin the mammalian ear, both in the adult guinea pig(Kawamoto et al., 2003) and in cultures of postnatal ratsensory organs (Zheng and Gao, 2000). Thus, some cellsretain the capacity to switch to a hair cell fate even intoadulthood.

IN VITRO EXPANSION OF OTICPROGENITORS

Several groups have used immortalizing oncogenes toisolate cell lines from the developing inner ear and exploretheir differentiation potential (reviewed by Rivolta andHolley, 2002). Efforts to expand unadulterated oticprogenitor pools are just beginning, using either immatureor differentiated otic epithelium as starting material. Thework is presented in order based on age of the startingtissue. Several of the studies have been reported only inabstract form to date.

Work in the laboratories of Segil and Groves definedculture conditions (EGFþ periotic mesenchyme) thatpermit E13.5 cochlear progenitors to persist in culture longafter they would normally become postmitotic(Deotzlhofer et al., 2003). Mitotic progenitors generateislands of Math-1þ hair cells, with numbers of hair cellscontinuing to increase 2 weeks after plating (Segil, personalcommunication).

Malgrange et al. (2002) dissociated cells from thenewborn rat organ of Corti, approximately 5e6 days aftersensory progenitors become postmitotic in vivo. Filteringthrough fine (15 mm) nylon mesh resulted in the isolation ofa population of small cells, 98% of which expressed nestin.Beginning with cell suspensions, spherical colonies calledotospheres developed when cultured in the presence ofEGF, FGF2, or both factors. After only 2 days of culturing,BrdUþ progenitors expressing the hair cell marker myosinVIIAwere observed, albeit in very small numbers (one percolony on average). About two cells per colony wereimmunopositive for the cell cycle inhibitor, p27kip1, whichnormally labels the supporting cells of the organ of Corti.The number of myosin VIIAþ cells increased to five cellsper colony by 14 days, often appearing as a coherent islet.Some ultrastructural features of hair cells were evident inrare cells after both 2 and 14 days in vitro. Hair celldifferentiation was not enhanced by switching the oto-spheres to “adherent” culture conditions that had beenreported to induce neuronal differentiation in neurospheres.It is important to note that the method used in this study didnot ensure that each otosphere originated from a singleprogenitor cell.

A preliminary study from another group reportedsuccessful generation of spherical cultures from mouseutricle or cochlea but only until postnatal day 6. Cells from14- and 21-day-old sensory organs failed under similarculture conditions (Licht et al., 2003).

Zhao (2001, 2003) reports culturing spheres from theisolated organ of Corti of adult guinea pig. Initially thespheres did not express nestin or the supporting cell marker,cytokeratin. Serum or EGF supplements, or long-termculture, induced nestin expression and allowed differenti-ation of a small number of cells as hair cells (calretininþ,myosin VIIAþ, prestinþ, or Brn3.1þ), supporting cells(cytokeratinþ or connexinþ), neurons (NFþ or bIII-tubulinþ), or astrocytes (GFAPþ). The appearance ofdifferentiated hair cells was extremely rare, at less than 1%of the cells in long-term cultures (Zhao, personalcommunication).

Most promising to date is the generation and differen-tiation of spherical cultures from single cells isolated fromthe utricular macula of the adult mouse. Taking theircue from the growth factor responsiveness of damagedvestibular sensory organs (Zheng et al., 1997; Kopke et al.,2001), Heller’s laboratory used EGF, IGF-1, and bFGF

748 VOLUME | 2 Adult and Fetal Stem Cells

to enhance sphere formation (Li et al., 2003a). Thecombination of EGFþIGF-1 was most effective. Nestinþ,sphere-forming cells could be dissociated into single cellsand then expanded into new spheres through severalrounds. Approximately 2.5 spheres could be formed at eachpassage, suggesting that a small number of cells retainsphere-renewal capability. To induce differentiation,spheres were moved to adherent culture conditions in thepresence of serum, but then grown for 14 days in serum-free conditions. The cells downregulated nestin and othermarkers of the early otic vesicle and upregulated markers ofseveral differentiated cell types. Cells expressing hair cellmarkers were present in up to 15% of differentiated cells.Many showed features consistent with rudimentarystereociliary bundle formation and were surrounded bycells with an expression profile consistent with supportingcells. Math-1þ cells colabeled with BrdU, indicating thatthese hair cells arose from a proliferative progenitor.Significantly, neurospheres grown from the subventricularzone of the mouse forebrain, cultured under identicalconditions, did not generate hair cells. This suggests thatthe utricular stem cells have a special capacity to form innerear mechanoreceptors. Macular-derived spheres alsoproduced a significant percentage of cells with neuronal(6%) or astrocytic (35%) phenotypes, cell types normallyabsent from the macular epithelium. Spheres could alsogenerate an array of ectoderm, mesoderm, or endodermderivatives when the cells were delivered into the amnioticcavity of stage 4 chicken embryos. The incidence ofsphere-forming stem cells was rare even under optimalgrowth conditions: 0.07% of plated cells. This is consistentwith the absence of BrdU-labeling of adult sensory organsin the mouse and suggests that stem cells may be both rareand quiescent in vivo.

Heller’s group also defined culture conditions thatinduced ES cells to form spheres containing many BrdUþ,nestinþ cells (Li et al., 2003b). Under growth conditions,the cells expressed nestin and markers of early otic vesicle(e.g., Pax2, BMP4, and BMP7). Under differentiatingconditions, early otic markers plummeted, and markers ofhair cells and supporting cells rose. This is extremelyencouraging as it suggests that it may be unnecessary tostart with endogenous ear tissue to generate otic progeni-tors for therapeutic purposes.

PROSPECTS FOR THERAPY

As methods for culturing otic stem cells become estab-lished, one can ask whether the addition of different tran-scription factors (such as Math1) can induce differentiationof one cell type over another. Stem cells or differentiatedcells derived from various sources could then be implantedback into the animal (Iguchi et al., 2003; Jurney, 2003;Tateya et al., 2003) to ask whether the cells will integrate

and provide restoration of function in animal models ofinner ear cell loss (reviewed by Duan et al., 2002; andHolley, 2002). The ear has some definite advantagesfor delivery of cells or gene transfer vectors, such asviruses. Surgical approaches to the fluid compartment ofthe inner ear provide access to the inner ear hair cellswithout requiring systemic delivery. For example, it ispossible to inject through the round window, deliveringsubstances, such as neurotrophins, that can influencesurvival of sensory tissues or ganglion cells (Miller et al.,1997; Duan et al., 2002; Shinohara et al., 2002). Delivery ofcells that release soluble molecules, such as growth factors,could potentially provide functional restoration withoutnecessarily restoring structural integrity. On the other hand,structural integration of replacement mechanoreceptorswill probably be essential, even if extremely difficult, inview of the precision with which hair cell stereocilia mustinteract with the nonsensory matrices overlying them.Replacement of functional ganglion neurons, rather thansensory receptor cells, may be less problematic (Oliviuset al., 2003). We anticipate considerable progress in theseand related therapeutic approaches over the next decade,although substantial technical hurdles remain.

ACKNOWLEDGMENTS

Donna Fekete thanks S. Heller, H.B. Zhao, and T. Nakagawa for

sharing unpublished data and I. Lopez for Figure 63.2.

REFERENCES

Ahmad, I., Tang, L., Pham, H., 2000. Identification of neural progenitors

in the adult mammalian eye. Biochem. Biophys. Res. Commun. 270,

517e521.

Altshuler, D.M., Turner, D.L., Cepko, C.L., 1991. Specification of cell

type in the vertebrate retina. In: Lam, D.M.K., et al. (Eds.),

Development of the Visual System. MIT Press, Cambridge. MA,

pp. 37e58.

Balak, K.J., Corwin, J.T., Jones, J.E., 1990. Regenerated hair cells can

originate from supporting cell progeny: evidence from phototoxicity

and laser ablation experiments in the lateral line system. J. Neurosci.

10, 2502e2512.

Barishak, Y.R., 2001. Embryology of the Eye and Its Adnexa, 2nd ed..

Karger, Basel.

Blackshaw, S., Harpavat, S., Trimarchi, J., Cai L., Huang, H., Kuo, W.P.,

et al. Genomic analysis of mouse retinal development. PLoS Biol. 2,

E247.

Burmeister, M., Novak, J., Liang, M.Y., Basu, S., Ploder, L., Hawes, N.L.,

et al., 1996. Ocular retardation mouse caused by Chx10 homeobox

null allele: impaired retinal progenitor proliferation and bipolar cell

differentiation. Nature 12, 376e383.

Calof, A.L., Bonnin, A., Crocker, C., Kawauchi, S., Murray, R.C.,

Shou, J., et al., 2002. Progenitor cells of the olfactory receptor neuron

lineage. Microsci. Res. Tech. 58, 176e188.

749Chapter | 63 Sensory Epithelium of the Eye and Ear

Cepko, C.L., Austin, C.P., Yang, X., Alexiades, M., Ezzeddine, D., 1996.

Cell fate determination in the vertebrate retina. Proc. Natl. Acad. Sci.

U.S.A. 93, 589e595.

Chen, P., Segil, N., 1999. p27(Kip1) links cell proliferation to morpho-

genesis in the developing organ of Corti. Development 126,

1581e1590.

Chen, P., Zindy, F., Abdala, C., Liu, F., Li, X., Roussel, M.F., et al., 2003.

Progressive hearing loss in mice lacking the cyclin-dependent kinase

inhibitor Ink4d. Nat. Cell Biol. 5, 422e426.

Chen, S., Wang, Q.L., Nie, Z., Sun, H., Lennon, G., Copeland, N.G., et al.,

1997. Crx, a novel Otx-like paired-homeodomain protein, binds to and

transactivates photoreceptor cell-specific genes. Neuron 19,

1017e1030.

Corwin, J.T., Oberholtzer, J.C., 1997. Fish n’ chicks: model recipes for

hair-cell regeneration? Neuron 19, 951e954.

Corwin, J.T., Warchol, M.E., 1991. Auditory hair cells: structure, function,

development, and regeneration. Annu. Rev. Neurosci. 14, 301e333.

Cotanche, D.A., Lee, K.H., 1994. Regeneration of hair cells in the ves-

tibulocochlear system of birds and mammals. Curr. Opin. Neurobiol.

4, 509e514.

Coulombre, J.L., Coulombre, A.J., 1965. Regeneration of neural retina

from the pigmented epithelium in the chick embryo. Dev. Biol. 12,

79e92.

Deotzlhofer, A., et al., 2003. In vitro growth and differentiation of sensory

hair cell progenitors from the embryonic mouse inner ear. Assoc. Res.

Otolaryngol. Abstr. 26, 204.

Detwiler, S.R., Van Dyke, R.H., 1954. The induction of neural retina from

the pigment epithelial layer of the eye. J. Exp. Zool. 126, 135e150.

Doetsch, F., Caille, I., Lim, D.A., Garcia-Verdugo, J.M., Alvarez-

Buylla, A., 1999. Subventricular zone astrocytes are neural stem cells

in the adult mammalian brain. Cell 97, 703e716.

Dowling, J.E., 1987. The RetinadAn Approachable Part of the Brain.

Harvard University Press, Cambridge, MA.

Duan, M.L., Ulfendahl, M., Laurell, G., Counter, A.S., Pyykko, I.,

Borg, E., et al., 2002. Protection and treatment of sensorineural

hearing disorders caused by exogenous factors: experimental findings

and potential clinical application. Hear. Res. 169, 169e178.

Fekete, D.M., Wu, D.K., 2002. Revisiting cell fate specification in the

inner ear. Curr. Opin. Neurobiol. 12, 35e42.

Fekete, D.M., Muthukumar, S., Karagogeos, D., 1998. Hair cells and

supporting cells share a common progenitor in the avian inner ear. J.

Neurosci. 18, 7811e7821.

Fischer, A.J., Reh, T.A., 2000. Identification of a proliferating marginal zone

of retinal progenitors in postnatal chickens. Dev. Biol. 220, 197e210.

Fischer, A.J., Reh, T.A., 2001. Muller glia are a potential source of neural

regeneration in the postnatal chicken retina. Nat. Neurosci. 4, 247e252.

Fischer, A.J., Reh, T.A., 2002. Exogenous growth factors stimulate the

regeneration of ganglion cells in the chicken retina. Dev. Biol. 251,

367e379.

Fischer, A.J., Dierks, B.D., Reh, T.A., 2002. Exogenous growth factors

induce the production of ganglion cells at the retinal margin.

Development 129, 2283e2291.

Fischer, A.J., McGuire, C.R., Dierks, B.D., Reh, T.A., 2002. Insulin and

fibroblast growth factor 2 activate a neurogenic program in Muller

glia of the chicken retina. J. Neurosci. 22, 9387e9398.

Forge, A., Li, L., Nevill, G., 1998. Hair cell recovery in the vestibular

sensory epithelia of mature guinea pigs. J. Comp. Neurol. 397,

69e88.

Furukawa, T., Morrow, E.M., Cepko, C.L., 1997. Crx, a novel otx-like

homeobox gene, shows photoreceptor-specific expression and regu-

lates photoreceptor differentiation. Cell 91, 531e541.

Goodyear, R.J., Gates, R., Lukashkin, A.N., Richardson, G.P., 1999. Hair-

cell numbers continue to increase in the utricular macula of the early

posthatch chick. J. Neurocytol. 28, 851e861.

Guillemot, F., Cepko, C., 1992. Retinal fate and ganglion cell differen-

tiation are potentiated by acidic FGF in an in vitro assay of early

retinal development. Development 114, 743e754.

Haddon, C., Jiang, Y.J., Smithers, L., Lewis, J., 1998. DeltaeNotch

signaling and the patterning of sensory cell differentiation in the

zebra fish ear: evidence from the mind bomb mutant. Development

125, 4637e4644.

Haruta, M., Kosaka, M., Kanegae, Y., Saito, I., Inoue, T., Takahashi, M.,

et al., 2001. Induction of photoreceptor-specific phenotypes in adult

mammalian iris tissue. Nat. Neurosci. 4, 1163e1164.

Hatten, M.E., Liem, R.K., Shelanski, M.L., Mason, C.A., 1991. Astroglia

in CNS injury. Glia. 16, 779e789.

Holley, M., 2002. Application of new biological approaches to stimulate

sensory repair and protection. Br. Med. Bull. 63, 157e169.

Hollyfield, J.G., 1968. Differential addition of cells to the retina in Rana

pipiens tadpoles. Dev. Biol. 18, 163e179.

Hume, C.R., Kirkegaard, M., Oesterle, E.C., 2003. ErbB expression: the

mouse inner ear and maturation of the mitogenic response to here-

gulin. J. Assoc. Res. Otolaryngol. 4, 422e443.

Iguchi, F., Nakagawa, T., Tateya, I., Kim,T.S., Endo, T., Taniguchi, Z., et al.,

2003. Trophic support of mouse inner ear by neural stem cell trans-

plantation. Neuroreport 14, 77e80.

Jensen, A.M., Raff, M.C., 1997. Continuous observation of multipotential

retinal progenitor cells in clonal density culture. Dev. Biol. 188,

267e279.

Johns, P.R., 1977. Growth of the adult goldfish eyedIII: Source of the

new retinal cells. J. Comp. Neurol. 176, 343e357.

Johns, P.R., Fernald, R.D., 1981. Genesis of rods in teleost fish retina.

Nature 293, 141e142.

Jurney, W.M., 2003. Survival and distribution of adult-derived stem cells

transplanted into the adult mouse inner ear. Assoc. Res. Otolaryngol.

Abstr. 26, 262.

Katayama, A., Corwin, J.T., 1989. Cell production in the chicken cochlea.

J. Comp. Neurol. 281, 129e135.

Kawamoto, K., Ishimoto, S., Minoda, R., Brough, D.E., Raphael, Y.,

2003. Math1 gene transfer generates new cochlear hair cells in

mature guinea pigs in vivo. J. Neurosci. 23, 4395e4400.

Kil, J., Warchol, M.E., Corwin, J.T., 1997. Cell death, cell proliferation,

and estimates of hair cell life spans in the vestibular organs of chicks.

Hear. Res. 114, 117e126.

Kopke, R.D., Jackson, R.L., Li, G., Rasmussen, M.D., Hoffer, M.E.,

Frenz, D.A., et al., 2001. Growth factor treatment enhances vestibular

hair cell renewal and results in improved vestibular function. Proc.

Natl. Acad. Sci. U.S.A. 98, 5886e5891.

Kramer, H.H., Hoefsloot, L.H., 2002. Molecular diagnosis of hereditary

hearing impairment. Adv. Otorhinolaryngol. 61, 11e27.

Kriegstein, A., Gotz, M., 2003. Radial glia diversity: a matter of cell fate.

Glia. 43, 37e43.

Kubota, R., Hokoc, J.N., Moshiri, A., McGuire, C., Reh, T.A., 2002.

A comparative study of neurogenesis in the retinal ciliary marginal

zone of homeothermic vertebrates. Brain Res. Dev. Brain Res. 134,

31e41.

750 VOLUME | 2 Adult and Fetal Stem Cells

Lang, H., Fekete, D.M., 2001. Lineage analysis in the chicken inner ear

shows differences in clonal dispersion for epithelial, neuronal, and

mesenchymal cells. Dev. Biol. 234, 120e137.

Lendahl, U., Zimmerman, L.B., McKay, R.D., 1990. CNS stem cells

express a new class of intermediate filament protein. Cell 60,

585e595.

Li, H., Roblin, G., Heller, S., 2003a. Pluripotent stem cells from the adult

mouse inner ear. Nat. Med. 9, 1293e1299.

Li, H., Roblin, G., Heller, S., 2003b. Generation of hair cells by stepwise

differentiation of embryonic stem cells. Proc. Natl. Acad. Sci. U.S.A.

100, 13495e13500.

Li, X.C., Friedman, R.A., 2002. Nonsyndromic hereditary hearing loss.

Otolaryngol. Clin. North Am. 35, 275e285.

Licht, K., Wachs, F.P., Strutz, J., 2003. Cultures of inner ear tissue reveal

potential stem cells only within two weeks after birth in NMRI mice.

Assoc. Res. Otolaryngol. Abstr. 26, 259.

Lillien, L., Cepko, C., 1992. Control of proliferation in the retina:

temporal changes in responsiveness to FGF and TGF-a. Development

115, 253e266.

Linser, P.J., Schlosshauer, B., Galileo, D.S., Buzzi, W.R., Lewis, R.C.,

1997. Late proliferation of retinal Muller cell progenitors facilitates

preferential targeting with retroviral vectors in vitro. Dev. Genet. 20,

186e196.

Lowenheim, H., Furness, D.N., Kil, J., Zinn, C., Gultig, K., Fero, M.L.,

et al., 1999. Gene disruption of p27(Kip1) allows cell proliferation

in the postnatal and adult organ of Corti. Proc. Natl. Acad. Sci.

U.S.A. 96, 4084e4088.

MacLaren, R.E., 1996. Development and role of retinal glia in regener-

ation of ganglion cells following retinal injury. Br. J. Ophthalmol. 80,

458e464.

Malgrange, B., Belachew, S., Thiry, M., Nguyen, L., Rogister, B.,

Alvarez, M.L., et al., 2002. Proliferative generation of mammalian

auditory hair cells in culture. Mech. Dev. 112, 79e88.

Mann, I., 1950. The Development of the Human Eye. Grune and Stratton,

New York.

Miller, J., Chi, D.H., O’Keefe, L.J., Kruszka, P., Raphael, Y.,

Altschuler, R.A., 1997. Neurotrophins can enhance spiral ganglion

cell survival after inner hair loss. Int. J. Dev. Neurosci. 15, 631e643.

Noramly, S., Grainger, R.M., 2002. Determination of the embryonic inner

ear. J. Neurobiol. 53, 100e128.

Oesterle, E.C., Hume, C.R., 1999. Growth factor regulation of the cell

cycle in developing and mature inner ear sensory epithelia. J.

Neurocytol. 28, 877e887.

Okada, T.S., 1980. Cellular metaplasia or transdifferentiation as

a model for retinal cell differentiation. Curr. Top. Dev. Biol. 16,

349e380.

Olivius, P., Alexandrov, L., Miller, J., Ulfendahl, M., Bagger-

Sjoback, D., Kozlova, E.N., 2003. Allografted fetal dorsal root

ganglion neuronal survival in the guinea pig cochlea. Brain Res.

979, 1e6.

Park, C.M., Hollenberg, M.J., 1989. Basic fibroblast growth

factor induces retinal regeneration in vivo. Dev. Biol. 134,

201e205.

Prada, C., Puga, J., Perez-Mendez, L., Lopez, R., Ramirez, G., 1991.

Spatial and temporal patterns of neurogenesis in the chick retina. Eur.

J. Neurosci. 3, 559e569.

Radner, W., Sadda, S.R., Humayun, M.S., Suzuki, S., de Juan Jr., E.,

2002. Increased spontaneous retinal ganglion cell activity in rd mice

after neural retinal transplantation. Invest. Ophthalmol. Vis. Sci. 43,

3053e3058.

Reh, T.A., 1987. Cell-specific regulation of neuronal production in the

larval frog retina. J. Neurosci. 7, 3317e3324.

Retnet. http://www.sph.uth.tmc.edu/RetNet/.

Reyer, R.W., 1954. Regeneration in the lens in the amphibian eye.

Q. Rev. Biol. 29, 1e46.

Reynolds, B.A., Weiss, S., 1992. Generation of neurons and astrocytes

from isolated cells of the adult mammalian central nervous system.

Science 255, 1707e1710.

Rivolta, M.N., Holley, M.C., 2002. Cell lines in inner ear research. J.

Neurobiol. 53, 306e318.

Rodieck, R.W., 1998. The First Steps in Seeing. Sinauer, Sunderland, MA.

Rubel, E.W., Fritzsch, B., 2002. Auditory system development: primary

auditory neurons and their targets. Annu. Rev. Neurosci. 25, 51e101.

Ruben, R.J., 1967. Development of the inner ear of the mouse: a radio-

autographic study of terminal mitosis. Acta. Oto-Laryngologica. 220

(Suppl.), 4e44.

Satoh, T., Fekete, D.M., 2003. Mechanosensory epithelial cells and

ganglion cells are clonally related. Assoc. Res. Otolaryngol. Abstr.

26, 118.

Sharon, D., Blackshaw, S., Cepko, C.L., Dryja, T.P., 2002. Profile of

the genes expressed in the human peripheral retina, macula, and

retinal pigment epithelium determined through serial analysis of

gene expression (SAGE). Proc. Natl. Acad. Sci. U.S.A. 99,

315e320.

Shinohara, T., Bredberg, G., Ulfendahl, M., Pyykko, I., Olivius, N.P.,

Kaksonen, R., et al., 2002. Neurotropic factor intervention restores

auditory function in deafened animals. Proc. Natl. Acad. Sci. U.S.A.

99, 1657e1660.

Sologub, A.A., 1968. On the capacity of eye pigmented epithelium for

transformation into retina in anuran amphibian tadpoles. Tsitologiya.

10, 1526e1532.

Stone, J.S., Rubel, E.W., 2000. Cellular studies of auditory hair cell

regeneration in birds. Proc. Natl. Acad. Sci. U.S.A. 97,

11714e11721.

Stone, J.S., Choi, Y.S., Woolley, S.M., Yamashita, H., Rubel, E.W.,

1999. Progenitor cell cycling during hair cell regeneration in the

vestibular and auditory epithelia of the chick. J. Neurocytol. 28,

863e876.

Stone, L.S., 1950. The role of retinal pigmented cells in regenerating

neural retinae of adult salamander eyes. J. Exp. Zool. 113, 9e31.

Straznicky, K., Gaze, R.M., 1971. The growth of the retina in Xenopus

laevis: an autoradiographic study. J. Embryol. Exp. Morphol. 26,

67e79.

Streilein, J.W., Ma, N., Wenkel, H., Ng, T.F., Zamiri, P., 2002. Immu-

nobiology and privilege of neuronal retina and pigment epithelium

transplants. Vision Res. 42, 487e495.

Takahashi, M., Palmer, T.D., Takahashi, J., Gage, F.H., 1998. Wide-

spread integration and survival of adult-derived neural progenitor

cells in the developing optic retina. Mol. Cell Neurosci. 12,

340e348.

Tateya, I., Nakagawa, T., Iguehi, F., Kim, T.S., Endo, T., Yamada, S.,

et al., 2003. Fate of neural stem cells grafted into injured inner

ears of mice. Neuroreport. 14, 1677e1681.

Tropepe, V., Coles, B.L., Chiasson, B.J., Horsford, O.J., Elia, A.J.,

McInnes, R.R., et al., 2000. Retinal stem cells in the adult

mammalian eye. Science 287, 2032e2036.

751Chapter | 63 Sensory Epithelium of the Eye and Ear

Turner, D.L., Snyder, E.Y., Cepko, C.L., 1990. Lineage-independent

determination of cell type in the embryonic mouse retina. Neuron 4,

833e845.

Vogel-Hopker, A., Momose, T., Rohrer, H., Yasuda, K., Ishihara, L.,

Rapaport, D.H., 2000. Multiple functions of fibroblast

growth factor 8 (FGF-8) in chick eye development. Mech. Dev. 94,

25e36.

Warchol, M.E., 1999. Immune cytokines and dexamethasone influence

sensory regeneration in the avian vestibular periphery. J. Neurocytol.

28, 889e900.

Zhao, H.B., 2001. Long-term natural culture of cochlear sensory epithelia

of guinea pigs. Neurosci. Lett. 315, 73e76.

Zhao, H.B., 2003. Multipotent differentiability of adult mammalian

cochlear cells. Assoc. Res. Otolaryngol. Abstr. 26, 260.

Zhao, S., Rizzolo, L.J., Barnstable, C.J., 1997. Differentiation and

transdifferentiation of the retinal pigment epithelium. Int. Rev. Cytol.

171, 225e266.

Zheng, J.L., Gao, W.Q., 2000. Overexpression of Math1 induces robust

production of extra hair cells in postnatal rat inner ears.

Nat. Neurosci. 3, 580e586.

Zheng, J.L., Helbig, C., Gao, W.Q., 1997. Induction of cell proliferation

by fibroblast and insulin-like growth factors in pure rat inner ear

epithelial cell cultures. J. Neurosci. 17, 216e226.

Zheng, J.L., Frantz, G., Lewis, A.K., Sliwkowski, M., Gao, W.Q., 1999a.

Heregulin enhances regenerative proliferation in postnatal rat utric-

ular sensory epithelium after ototoxic damage. J. Neurocytol. 28,

901e912.

Zheng, J.L., Keller, G., Gao, W.Q., 1999b. Immunocytochemical and

morphological evidence for intracellular self-repair as an important

contributor tomammalianhair cell recovery. J.Neurosci. 19, 2161e2170.

Zine, A., de Ribaupierre, F., 1999. Tissue-specific levels and cellular

distribution of epidermal growth factor receptors within control and

neomycin-damaged neonatal rat organ of Corti. J. Neurobiol. 38,

313e322.