Molecular Vision 2007; 13:1842-1850 <http://www.molvis.org/molvis/v13/a205/> Received 7 March 2007 | Accepted 23 September 2007 | Published 2 October 2007

Download Reprint

Fibroblast growth factor and epidermal growth factor differently affect differentiation of murine retinal stem cells in vitro

Francesca Giordano,¹ Anna De Marzo,² Francesco Vetrini,¹ Valeria Marigo²
 
 

¹Telethon Institute of Genetics and Medicine, Naples; ²Department of Biomedical Sciences, University of Modena and Reggio Emilia, Modena, Italy

Correspondence to: Valeria Marigo, Department of Biomedical Sciences, University of Modena and Reggio Emilia, via Campi 287, 41100 Modena, Italy; Phone: +390592055392; FAX: +390592055410; email: valeria.marigo@unimore.it
 
Dr. Giordano is now at the Institute Curie, UMR144 CNRS, Structure et Compartiment membranaires, Paris, France. Dr. Vetrini is now at Baylor College of Medicine, Department of Molecular and Human Genetics, Houston, TX.

Abstract

Purpose: The developmental processes that mediate differentiation from retinal stem cells (RSC) to different retinal neuronal types remain unclear. During retinal development, progenitor cells modify expression of growth factor (GF) receptors and their differentiation potentials. Similarly, RSC in culture may exhibit alternative molecular characteristics in response to different GF stimuli.

Methods: RSC were purified from the adult ciliary margin and exposed to fibroblast growth factor (FGF), epidermal growth factor (EGF), or FGF+EGF. Proliferation was analyzed by bromodeoxyuridine (BrdU) labeling. Differentiation was evaluated by immunofluorescence with antibodies recognizing specific markers of different retinal cell types.

Results: In the absence of GF stimuli, RSC in culture expressed FGFR1, similar to early progenitors in vivo. Treatment with GFs up-regulated the expression of both fibroblast growth factor receptor 1 (FGFR1) and epidermal growth factor receptor (EGFR). Exposure to either FGF, EGF, or FGF+EGF strongly affected retinal stem cell-renewal and differentiation. Specifically, expression of progenitor/stem cell markers and stem cell-renewal was higher in the presence of FGF than in that of EGF. FGF favored differentiation of RSC into photoreceptor-like cells. Finally, we showed that the treatment of the primary culture with FGF+EGF imprinted the cells and limited plasticity in subsequent differentiation.

Conclusions: We provide evidence that conditions of the primary culture have a strong influence on cell-renewal and differentiation potentials of RSC.

Introduction

The retina is a highly organized laminar structure in which cells are born in a sequential, temporally defined order. It has been suggested that characteristics of retinal progenitors change during differentiation. Early progenitors are able to differentiate into all retinal neuronal and glial cell types while later progenitors are unable to give rise to early born retinal neurons such as ganglion cells and cones, but generate late born cells [1]. This competence appears to be due both to intrinsic characteristics and external stimuli. Among the intrinsic features are relative expressions of fibroblast growth factor receptor 1 (FGFR1) and EGFR. Retinal progenitor cells in early embryonic retina normally express higher levels of FGFR1 and lower levels of EGFR than progenitor cells in later retina. The differential expression of EGFR in retinal progenitor cells at distinct developmental stages has been proposed to be associated to signaling thresholds affecting differentiation [2].

Retinal progenitors can be purified from the embryonic retina and induced to differentiate into photoreceptors and other retinal neuronal cell types [3]. Stem cells have also been identified in the marginal region of the adult eye, and retinal stem cells (RSC) can be derived and cultured in vitro from the adult rodent and human ciliary body [4-6] and iris [7]. RSC are purified by enzymatic and mechanical dissociation and grown as pigmented neurospheres in suspension in minimal medium without serum but supplemented with growth factors, such as fibroblast growth factor (FGF) and epidermal growth factor (EGF). These cells are able to self-renew because they clonally generate new retinal spheres. When retinal spheres are allowed to grow on a laminin substrate in the presence of fetal bovine serum (FBS) they undergo differentiation into different retinal neuronal and glial cell types [4-6]. It is still unclear how much primary culture conditions have an effect on RSC self-renewal and differentiation potential. Das et al. suggested that RSC have an early progenitor receptor code with higher levels of FGFR1 than EGFR [8]. It is therefore reasonable that preparation and culture of retinal spheres in the presence of either FGF or EGF may affect their differentiation competence.

We assessed proliferation and differentiation potentials of RSC treated with either FGF, EGF, or FGF+EGF. Here we show that the different growth factors strongly affect retinal stem cell proliferation kinetics, self-renewal, and differentiation potential. Interestingly, GF treatment of a primary retinal sphere also influences reprogramming during generation of secondary retinal spheres and limits their differentiation potential.

Methods

Retinal stem cell culture

All procedures on mice (including their euthanasia) were performed in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and with institutional guidelines for animal research. Mice in this study were C57Bl6, purchased from Charles River Italy (Calco, Italy) and housed under standard conditions with a 12-hour light/dark cycle. We dissected eyes from 12-week-old C57Bl6 mice in artificial cerebral spinal fluid (ACSF) that containing 124 mM NaCl, 5 mM KCl, 100 nM CaCl₂, 1.3 mM MgCl₂, 26 mM NaHCO₃, and 10 mM D-glucose. Eyes were halved, and the lens and the neural retina were carefully removed. The ciliary margin was separated from the retinal pigment epithelium (RPE), and the cornea and the ciliary margin cells were scraped from the sclera after incubation in dispase (BD Biosciences, Milan, Italy) for 10 min at 37 °C and in enzyme mix (1.33 mg/ml trypsin, 0.67 mg/ml hyaluronidase, and 0.13 mg/ml kynurenic acid in ACSF) for 10 min at 37 °C. Cells were mechanically separated and centrifuged at 1500 rpm for 5 min. The enzyme solution was removed and replaced with serum free media containing 1 mg/ml trypsin inhibitor (Invitrogen, San Giuliano Milanese, Italy). Cells were further dissociated until single cell suspension and then centrifuged again. The supernatant was replaced with serum free medium (0.6% glucose and N2 hormone mix in DMEM-F12) containing either 20 ng/ml basic FGF supplemented with 2 μg/ml heparin (Sigma, Milan, Italy), or 20 ng/ml EGF, or both FGF and EGF. The cells were seeded at a concentration of 40,000 cells/ml and incubated for 3-7 days until floating spheres formed.

In differentiation experiments, retinal floating spheres were plated on eight well glass slide that was coated with extracellular matrix (ECM, Sigma) in DMEM-F12 supplemented with either 20 ng/ml FGF or 20 ng/ml EGF or FGF+EGF. The cells were allowed to proliferate and migrate over the course of four days. The medium was then replaced with 1% FBS (Gibco, San Giuliano Milanese, Italy) containing medium.

In vitro passaging

Single retinal floating spheres treated with either FGF, EGF, or EGF+FGF were collected and incubated in enzyme solution (in ACSF containing 1.33 mg/ml trypsin, 0.67 mg/ml hyaluronidase, 0.5 mg/ml collagenase type I-A, 0.5 mg/ml collagenase XI, 0.13 mg/ml kynurenic acid) for 1 h at 37 °C. After centrifugation at 400xg for 5 min, the supernatant was replaced with 1 mg/ml of trypsin inhibitor in medium and the spheres mechanically dissociated. Dissociated cells were seeded at the concentration of 40,000 cells/ml in serum free medium containing FGF, EGF, or EGF+FGF and incubated for six days.

In reprogramming experiments, EGF+FGF treated primary spheres after the first passage were incubated with FGF, EGF, or EGF+FGF.

Immunofluorescence

Cells were fixed in 4% paraformaldehyde for 15 min at room temperature. They were then permeabilized with 0.2% TritonX-100 in 10% goat serum and incubated with primary antibodies in (PBS) overnight at 4 °C. Cells underwent five washes with PBS, then were incubated with fluorescent-conjugated secondary antibodies for 1 h at room temperature. Primary antibodies used were as follows: 1:400 anti-nestin mouse monoclonal (Chemicon, Chandlers Ford, UK), 1:1500 anti-Pkc-α rabbit polyclonal (Sigma), 1:100 anti-Pde6b rabbit polyclonal (ABCAM, Cambridge, UK), 1:10,000 anti-rhodopsin mouse monoclonal RETP1 (Sigma), 1:400 anti-GFAP rabbit polyclonal (Sigma), 1: 400 anti-GS6 mouse monoclonal (Chemicon), 1:400 anti-G0α mouse monoclonal (Chemicon). Secondary antibodies were as follows: 1:1000 Oregon Green® 488 goat anti-mouse (Molecular Probes, San Giuliano Milanese, Italy) and 1:1000 Alexa Fluor® 568 goat anti-rabbit (Molecular Probes). Fluorescent cells in single spheres were counted on 20 image stacks (5 μm) generated by a Leica laser confocal microscope system.

Bromodeoxyuridine labeling

Retina floating spheres at three, five, and seven days of culture and adherent progenitors at four days of culture were treated with 10 mM bromodeoxyuridine (BrdU) for 3 h and then washed with PBS and fixed in 4% paraformaldehyde (PFA). Cells were treated with 2N HCl at 30 °C for 30 min, placed in a 0.1 M borate buffer pH 8.5 for 15 min, and then washed with PBS. Blocking was performed in 10% goat serum, 3% bovine serum albumin (BSA), 1% glycine and 0.3% Triton-X 100 for 30 min at room temperature followed by incubation with 1:8000 anti-BrdU monoclonal antibody (Developmental Hybridoma, Iowa City, IA) overnight at 4 °C. Slides were washed with PBS, then incubated with 1:1000 Oregon Green® 488 goat anti-mouse secondary antibody (Molecular Probes) for 1 h, washed and nuclei were labeled with 50 mg/ml propidium iodide and 2.5 mg/ml RNAse A at 37 °C for 1 h. Slides were mounted with Vectashield (Vector Laboratories, Segrate, Italy), and BrdU positive cells were counted in a stack of 20 images (5 μm) at a Leica laser confocal microscope system.

Real-time polymerase chain reaction

Total RNA was extracted from floating neurospheres and adherent cells using RNeasy MiniKit (Qiagen, Milan, Italy) according to the manufacturer's instructions. The same amount of cDNA for each treatment was synthesized using Superscript II (Invitrogen) and random primers. Real-time PCR was carried out with the GeneAmp 7000 Sequence Detection System (Applied Biosystems). The PCR reaction was performed using cDNA, 12.5 ml SYBR green master mix (Applied Biosystems) and 400 nM primer for each gene (see Table 1). Water was added to a final reaction volume of 25 ml. The PCR conditions were as follows: preheating at 50 °C for 2 min and 95 °C for 10 min; 40 cycles of 15 s at 95 °C and 1 min at 60 °C. Quantification results were expressed in terms of the cycle threshold (Ct). All real-time quantitative PCR reactions were run in triplicate and the Ct values were averaged from three independent samples. The S26 gene was used as an endogenous control (reference marker). Differences between the mean Ct values of each gene and those of the reference gene were calculated as ΔCt=Ct_gene-Ct_S26 and represented as 2^-ΔCt values.

Results

Progenitor markers and self-renewal of retinal stem cells in culture

Previous studies have shown different responsiveness of RSC to EGF and FGF [8]. In order to better characterize these observations at the molecular level, we tested different in vitro culture conditions of retinal spheres derived from the murine adult ciliary margin. We generated primary spheres with minimal medium in the presence of either FGF, EGF, or FGF+EGF. We compared growth competence of cells exposed to different growth factors and found that treatment with both growth factors gave rise to spheres larger in diameter than spheres treated with either FGF or EGF (Figure 1A-C) When we evaluated proliferation by using BrdU labeling, we found that proliferation was strongly stimulated in the first days of culture by the presence of EGF with 15% of BrdU positive cells at the fifth day of culture. This probably accounts for the larger diameter of the spheres at D7 (Figure 1D). However, growth kinetics were delayed when cells were treated with FGF alone, showing significantly lower numbers of proliferating cells at the fifth day of culture (9%) but 20% of BrdU positive cells after seven days of culture (Figure 1A,D). It is possible that spheres would increase in size with longer time in culture with FGF. Interestingly, cells exposed to either EGF or FGF+EGF have a sharp drop in proliferation after five days in culture. Proliferation was also evaluated by Cyclin D1 expression at day seven of culture. Spheres treated with GF expressed higher levels of Cyclin D1 compared to spheres not exposed to GF (Figure 1E).

We then analyzed expression of markers of neuronal stem cells and progenitors. Seven days of culture in the presence of FGF alone or together with EGF allowed higher expression of nestin and GFAP than with EGF alone (Figure 2D,F). Most cells positive to nestin and GFAP also expressed Pkc-α (Figure 2A-C; data not shown), a marker previously reported, and confirmed by our immunofluorescence analysis, to label only bipolar cells in the adult retina [9]. Here we found that Pkc-α was also expressed in retinal spheres and showed partial co-localization with nestin (Figure 2A-C). Further, we confirmed that Pkc-α labels proliferating progenitors by double staining with BrdU and anti- Pkc-α (Figure 2G). Otherwise, exposure to only EGF limited the number of cells expressing stem cell/progenitor markers.

Given the differential expression of EGFR and FGFR1 by retinal progenitors at different developmental stages [10], we considered whether different treatments affected expression of the two receptors. As previously reported, we found that in the absence of GF, RSC express mostly FGFR1 (Figure 2H-I blue bars) [8]. After seven days in culture FGFR1 mRNA appeared to be induced by treatment with GFs and relatively more by the presence of FGF in the culture medium (Figure 2H). A similar result was observed when EGFR expression was analyzed (Figure 2I).

Self-renewal was assessed by dissociation of the primary retinal spheres and re-plating in GF medium. Retinal spheres treated with EGF alone showed a poor ability to give rise to secondary spheres. Cultures with FGF alone or FGF+EGF doubled the number of retinal spheres at the first passage, but this characteristic was lost in the following passages (Figure 3A). In order to determine if the decrease in self-renewal ability was associated with a reduction in cell proliferation of secondary spheres, we labeled secondary retinal spheres with BrdU. The percentage of proliferating cells was only slightly reduced compared to primary retinal spheres (Figure 3B) and confirmed that the presence of EGF in the culture medium negatively affected cell proliferation.

Altogether, these data showed that EGF does not favor expression of stem cell markers and limits RSC renewal and proliferation.

Differentiation potentials of retinal stem cells in culture

The interest in the evaluation of RSC culture treatments was derived from the search of favorable conditions to obtain rod photoreceptor differentiation. We therefore allowed primary retinal spheres to attach to an ECM substrate and differentiate without using specific treatments other than 1% FBS. Cells were treated with the different GFs during the first four days in culture to facilitate their growth out of the spheres. At this time point both FGF and EGF similarly induced proliferation as evaluated by BrdU incorporation or cyclin D1 expression (data not shown).

Differentiation of cells derived from RSC treated with the different GFs was evaluated by immunofluorescence analysis at different times during differentiation. We decided to analyze in detail differentiation of rods and compare this to differentiation of bipolars, because these two retinal cell types differentiate in vivo during the same time window. Under proper condition, progenitors have also been shown to be able to change their fate from rods to bipolars [9]. Our culture conditions were appropriate for differentiation into cells expressing several retinal neuronal markers such as horizontal-like cells (positive to Calbindin, Figure 4I), amacrine-like cells (positive to Syntaxin, Figure 4J), and Müller glia-like (positive to GFAP and GS6, Figure 4K). Rod photoreceptor-like cells were identified by co-expression of two cell-specific markers: rhodopsin (Rho) and Pde6b (Figure 4A-D). An elongated shape of the cells with accumulation of Rho at the peripheral region was observed in prolonged cultures (Figure 4B,D, arrow). Bipolar-like cells were defined as positive to Pkc-α and to G0α, a marker of ON bipolars [11], but nestin negative (Figure 4E-H). As previously shown, in our hands Pkc-α is co-expressed with stem cell/progenitor markers in retinal spheres. G0α was never observed after four days in culture but turned on in prolonged cultures (Figure 4H, arrow). Finally, co-expression of nestin and Pkc-α characterized retinal progenitors (Figure 4E-F, arrows). Progenitors decreased over differentiation time with all treatments, and EGF appeared to favor this phenomenon as assessed by nestin and Pkc-α expression (Figure 5A). Generation of retinal spheres in the presence of FGF increased the number of cells differentiating into rod-like cells and, interestingly, four days in differentiating condition allowed 25% of cells to express rhodopsin and Pde6b (Figure 5B). The percentage of rod-like cells slightly increased over time, reaching 30% after 18 days of culture (Figure 5B). Concomitantly, cells acquired an elongated shape (Figure 4C-D). Bipolar-like cell fate was favored by preparation of retinal spheres with EGF. In this condition we obtained 24% of cells expressing Pkc-α and turning off nestin within four days of differentiation (Figure 5D). We also observed that over time in culture, the percentage of cells expressing Pkc-α decreased. To better characterize cells after 18 days of differentiation, we analyzed expression of FGFR1 and EGFR. FGFR1 mRNA was only slightly lower in cultures derived from spheres treated with EGF than with FGF alone, while EGFR mRNA was significantly higher in cultures exposed to FGF alone (Figure 4L).

We finally tested whether RSC were able to reprogram themselves during in vitro passaging. Toward this aim we set primary cultures of retinal spheres in the presence of FGF+EGF. Cells were left in culture for six days to allow spheres to form, then were dissociated and plated in stem cell conditions with either FGF, EGF, or FGF+EGF. Newly formed retinal spheres (secondary retinal spheres) were seeded on an ECM substrate and permitted to differentiate as done for primary cultures. The initial treatment with both GFs had a strong effect on differentiation potentials of the cells (Figure 5C,E). Specific differentiation effects of FGF and EGF stimuli were attenuated in the secondary cultures. Treatment of the secondary culture with FGF gave rise to less photoreceptor-like cells (21.7±2.5 versus 31.0±2.3) and more bipolar-like cells (8.3±1.5 versus 4±0.85) than cells that were exposed to FGF from the primary culture. On the contrary, exposure of the primary culture to both GFs did not affect EGF induction of rod photoreceptor-like cells but reduced its ability to activate bipolar-like cell markers (7.3±1.5 versus 11.5±1.6; Figure 5C,E). Of note is the observation that primary and secondary cultures exposed to FGF+EGF showed similar differentiation potentials with any treatment. The average number of cells expressing rod cell markers is similar in all conditions and not significantly different from primary cultures exposed to FGF+EGF (about 20% in Figure 5C). Similarly, we observed that the primary treatment with FGF+EGF gave rise to a similar percentage of cells choosing the bipolar-like fate when treated with the three different conditions (about 8% in Figure 5E).

Discussion

The aim of this work was to better characterize RSC derived from the adult ciliary margin and define their differentiation potential. We primarily wanted to analyze how FGF and EGF affect proliferation, self-renewal, and differentiation of RSC. We started by characterizing molecular features and self-renewal of RSC in culture. We report that treatments with distinctive GFs differently affect cell proliferation of the retinal spheres. Proliferation in the first days of culture is favored by the presence of EGF in the culture medium but then drops off after seven days. This may affect self-renewal as suggested by our data showing that cells treated with EGF alone have minimal ability to form new retinal spheres after passaging. The drop in recovery of retinal spheres after the second passage is not easily explained. The percentage of proliferating cells in secondary retinal spheres remains similar to primary spheres, but we probably reduce the percentage of stem cells compared to progenitors that are not able to self-renew in vitro. Our data are in agreement with previous reports that showed a similar low recovery after the second passage [12].

Our analysis points to an influencing effect of the primary culture conditions on the differentiation potential of retinal progenitors. Treatment with FGF favors rod-like differentiation. Time course experiments showed that within eight days the number of cells expressing Rhodopsin and Pde6b reached 30% and was maintained during the subsequent 10 days in culture. In addition, a longer time in culture allowed the cells to acquire an elongated shape and to accumulate rhodopsin at the tip of the cell. This suggests that FGF positively acted on rod-like differentiation while EGF had a negative effect by limiting the number of cells expressing rod markers. It is not likely that EGF reduced cell viability because the number of cells turning on rod cell markers continuously increased over time in culture (see Figure 5B). The inducing activity of FGF is also supported by the experiment in the presence of FGF+EGF in which the percentage of cells that differentiated into rod-like cells after four to eight days of differentiation is higher than in EGF treated cells. Previous reports suggested a positive role of FGF on rod differentiation and survival, but not the negative effect of EGF [13].

The choice to become bipolar-like cells is favored by EGF that, however, may not be sufficient for the maintenance of this cell type. Cells expressing bipolar cell markers decreased over time in culture. Because of the late expression of the G0α marker, it is possible that not all cells positive to Pkc-α but negative to nestin will eventually complete their bipolar-like differentiation. Our data also suggested that expression of Pkc-α in in vitro cultured retinal cells is not strictly related to expression of this marker in bipolar cells but also labels progenitor cells. The number of nestin-Pkc-α double labeled cells is higher in FGF-treated cells than in EGF-treated cells, while the positive effect of EGF on bipolar-like differentiation is shown by the high percentage of G0α-Pkc-α double labeled cells. FGF appeared to oppose bipolar-like differentiation as can be seen by the low percentage of cells expressing G0α-Pkc-α in cultures treated with both FGF and EGF or FGF alone.

The differential role of EGF and FGF in differentiation may be due to their effect on progenitors and not directly on stem cells. We should keep in mind that only a small number of cells in each retinal sphere are real stem cells (after dissociation, not all cells form new retinal spheres as shown by our self-renewal test), and progenitors may have chosen alternative commitments in response to different GFs. Cells within a few days on ECM substrate start to turn off progenitor markers like nestin and undertake their fate. We cannot exclude that these cells are still malleable and able to change their choice when properly instructed.

Our study provides evidence that exposure of freshly dissociated cells from the ciliary margin to different GFs affected their differentiation potentials in subsequent passages. Cells that were treated in the primary culture with FGF+EGF gave rise to a defined percentage of rods and bipolar cells even if only one of the GFs was present in the secondary culture. This was an unexpected result. Even if GFs can induce progenitor pre-commitment, cells after dissociation and growth, as secondary spheres with either FGF or EGF, should behave like freshly harvested stem cells. Our observations suggest that stem cells in culture are affected by different GFs. Further studies are needed to define the molecular events underlying this phenomenon.

Our study highlights the importance of GFs as favoring conditions to in vitro differentiate retinal stem cells. This is the first step toward the definition of optimal cell culture conditions to in vitro generate rod photoreceptors that may be employed in transplants of degenerating retinas. As recently suggested [14], a proper differentiation of RSCs into photoreceptor precursors will be fundamental to obtain graft integration and proper differentiation of transplanted cells. Future studies will focus on the identification of more specific inducing factors to be used in differentiating cultures and to allow a higher percentage of cells to undertake the rod pathway.

Acknowledgements

The authors thank Dr. Derek van der Kooy, Brenda Coles, and the animal house facility. This work was supported by research grants EVI-GENORET: LSHG-CT-2005-512036 from the European Community, research grant GGP06096 from Fondazione Telethon, PRIN 2006053302_003 and EMBL short-term fellowship ASTF 233-2002 (V.M).

References

1. Cepko CL, Austin CP, Yang X, Alexiades M, Ezzeddine D. Cell fate determination in the vertebrate retina. Proc Natl Acad Sci U S A 1996; 93:589-95.

2. Lillien L, Wancio D. Changes in Epidermal Growth Factor Receptor Expression and Competence to Generate Glia Regulate Timing and Choice of Differentiation in the Retina. Mol Cell Neurosci 1998; 10:296-308.

3. Livesey FJ, Cepko CL. Vertebrate neural cell-fate determination: lessons from the retina. Nat Rev Neurosci 2001; 2:109-18.

4. Ahmad I, Tang L, Pham H. Identification of neural progenitors in the adult mammalian eye. Biochem Biophys Res Commun 2000; 270:517-21.

5. Coles BL, Angenieux B, Inoue T, Del Rio-Tsonis K, Spence JR, McInnes RR, Arsenijevic Y, van der Kooy D. Facile isolation and the characterization of human retinal stem cells. Proc Natl Acad Sci U S A 2004; 101:15772-7.

6. Tropepe V, Coles BL, Chiasson BJ, Horsford DJ, Elia AJ, McInnes RR, van der Kooy D. Retinal stem cells in the adult mammalian eye. Science 2000; 287:2032-6.

7. Haruta M, Kosaka M, Kanegae Y, Saito I, Inoue T, Kageyama R, Nishida A, Honda Y, Takahashi M. Induction of photoreceptor-specific phenotypes in adult mammalian iris tissue. Nat Neurosci 2001; 4:1163-4.

8. Das AV, James J, Rahnenfuhrer J, Thoreson WB, Bhattacharya S, Zhao X, Ahmad I. Retinal properties and potential of the adult mammalian ciliary epithelium stem cells. Vision Res 2005; 45:1653-66.

9. Ezzeddine ZD, Yang X, DeChiara T, Yancopoulos G, Cepko CL. Postmitotic cells fated to become rod photoreceptors can be respecified by CNTF treatment of the retina. Development 1997; 124:1055-67.

10. James J, Das AV, Rahnenfuhrer J, Ahmad I. Cellular and molecular characterization of early and late retinal stem cells/progenitors: differential regulation of proliferation and context dependent role of Notch signaling. J Neurobiol 2004; 61:359-76.

11. Dhingra A, Jiang M, Wang TL, Lyubarsky A, Savchenko A, Bar-Yehuda T, Sterling P, Birnbaumer L, Vardi N. Light response of retinal ON bipolar cells requires a specific splice variant of Galpha(o). J Neurosci 2002; 22:4878-84.

12. Inoue Y, Yanagi Y, Tamaki Y, Uchida S, Kawase Y, Araie M, Okochi H. Clonogenic analysis of ciliary epithelial derived retinal progenitor cells in rabbits. Exp Eye Res 2005; 81:437-45.

13. Gargini C, Belfiore MS, Bisti S, Cervetto L, Valter K, Stone J. The impact of basic fibroblast growth factor on photoreceptor function and morphology. Invest Ophthalmol Vis Sci 1999; 40:2088-99.

14. MacLaren RE, Pearson RA, MacNeil A, Douglas RH, Salt TE, Akimoto M, Swaroop A, Sowden JC, Ali RR. Retinal repair by transplantation of photoreceptor precursors. Nature 2006; 444:203-7.