Molecular Vision 2001; 7:154-163 <http://www.molvis.org/molvis/v7/a22/> Received 11 April 2001 | Accepted 18 June 2001 | Published 30 June 2001

Download Reprint

Localization of pigment epithelium derived factor (PEDF) in developing and adult human ocular tissues

Petros C. Karakousis,¹ Sinoj K. John,² Kathryn C. Behling,¹ Enrico M. Surace,¹ Julie E. Smith,² Anita Hendrickson,³ Wai-Xing Tang,² Jean Bennett,¹ Ann H. Milam²
 
 

¹F. M. Kirby Center for Molecular Ophthalmology, ²Scheie Eye Institute, University of Pennsylvania, Philadelphia, PA; ³Department of Biological Structure, University of Washington, Seattle, WA

Correspondence to: Ann H. Milam, Ph.D., Scheie Eye Institute, 51 North 39th Street, Philadelphia, PA, 19104; Phone: (215) 662-8102; FAX: (215) 662-5879; email: annmilam@mail.med.upenn.edu

Abstract

Purpose: To localize pigment epithelium-derived factor (PEDF) in developing and adult human ocular tissues.

Methods: PEDF was localized in fetal and adult eyes by immunofluorescence with a polyclonal antibody (pAb) against amino acids 327-343 of PEDF, or a monoclonal antibody (mAb) against the C-terminal 155 amino acids of PEDF. Specificity of the antibodies was documented by Western blotting. PEDF mRNA was localized in adult retina by in situ hybridization.

Results: In developing retinas (7.4 to 21.5 fetal weeks, Fwks), pAb anti-PEDF labeled retinal pigment epithelium (RPE) granules, developing cones, some neuroblasts and many cells in the ganglion cell layer (GCL). In adult retinas, pAb anti-PEDF labeled rod and cone cytoplasm and nuclei of rods but not cones. Cells in the INL and GCL, choroid, corneal epithelium and endothelium, and ciliary body were also pAb PEDF-positive. Preadsorption of pAb anti-PEDF with the immunizing peptide blocked specific labeling in retina and other tissues, except for photoreceptor outer segments. In agreement with the immunolocalization with pAb anti-PEDF, in situ hybridization revealed PEDF mRNA in the RPE, photoreceptors, inner nuclear layer cells and ganglion cells in adult retina. In developing retinas 18 Fwks and older, and in adult retinas, mAb anti-PEDF labeled the interphotoreceptor matrix (IPM). Western blots of retina, cornea, and ciliary body/iris with pAb anti-PEDF produced several bands at about 46 kDa. With mAb anti-PEDF, retina produced one band at about 46 kDa; cornea and ciliary body/iris had several bands at about 46 kDa.

Conclusions: PEDF, originally reported as a product of RPE cells, is present in photoreceptors and inner retinal cell types in developing and adult human eyes. Photoreceptors and RPE may secrete PEDF into the IPM.

Introduction

Pigment epithelium-derived factor (PEDF) was first identified as a 50 kDa secreted protein in conditioned medium from cultured fetal human retinal pigment epithelium (RPE) cells [1,2]. The first evidence that PEDF might be involved in neuronal differentiation in the retina came from the observation that addition of crude soluble interphotoreceptor matrix (IPM) extracts containing PEDF or purified PEDF to human Y-79 retinoblastoma cells induced a neuronal phenotype, evidenced morphologically by extension from the cells of long, neurite-like processes and biochemically by increased expression of neuron-specific proteins [1-4]. More recently, PEDF was shown to have neurotrophic activity in the retina in vivo, as intravitreal injections of PEDF delayed apoptotic photoreceptor loss in mouse models of inherited retinal degeneration [5]. PEDF is also a neurotrophic and neuroprotective factor in other systems, including cultured cerebellar granule cells [6-8], primary hippocampal neurons [9], spinal cord motor neurons [10,11] and amphibian photoreceptors cultured in the absence of RPE cells [12]. The binding of PEDF to retinoblastoma and cerebellar granule cells may be receptor-mediated [13]. Recently, PEDF was shown to be a potent inhibitor of angiogenesis in endothelial cell cultures and in rat cornea and retina in vivo [14]. In addition, systemic application of PEDF prevented inner retinal neovascularization in a mouse model of retinopathy of prematurity [15].

The gene for PEDF was localized to chromosome 17p13 [16,17], near the locus of the RP13 gene for autosomal dominant retinitis pigmentosa (RP) and Leber congenital amaurosis [18]. However, disease-causing PEDF mutations have not yet been identified. PEDF shares sequence and structural homology with the serine protease inhibitor (serpin) family [3,19]. However, PEDF does not inhibit proteases and does not require the serpin reactive loop for its neurotrophic activity [20]. Northern analysis of RNA from bovine RPE and retina revealed expression of PEDF only by the RPE [21]. In situ hybridization studies demonstrated PEDF mRNA in the RPE of a human fetal eye at 17 weeks gestation [22], and Western blot analyses identified PEDF in soluble washes of the IPM from several species [4,22]. The protein is also synthesized and secreted by human ciliary epithelium [23]. Finally, PEDF gene expression has been documented in a broad range of fetal and adult human tissues, including brain and various endocrine organs [24].

In order to gain insight into the role of PEDF as a neurotrophic and anti-angiogenic factor in the retina, we have used immunofluorescence to localize PEDF in developing and adult human ocular tissues.

Methods

Tissue preparation

Post mortem normal adult human eyes were obtained from the University of Washington Lions Eye Bank (Seattle, WA) and the Foundation Fighting Blindness (Owings Mill, MD). Normal fetal human eyes were obtained from the University of Washington Human Embryology Laboratory (Seattle, WA). Informed consent was obtained from all adult donors ante mortem. All research was conducted in accordance with the Declaration of Helsinki and was approved by the institutional human subjects review boards of the University of Pennsylvania and the University of Washington.

The following specimens were evaluated: seven retinas aged 7.4 to 21.5 fetal weeks (Fwks) and three adult retinas aged 53, 65 and 75 years (Table 1). The anterior segments were also studied from the fetal eyes and two additional normal adult globes (Table 1). All globes were fixed for several weeks to months in 4% paraformaldehyde, with or without 0.5% glutaraldehyde, in 0.1 M phosphate buffer (pH 7.3) and stored thereafter in 2% paraformaldehyde in the same buffer.

Immunocytochemistry

Retinal samples were cryosectioned at 12 mm and processed for immunofluorescence by published techniques [25]. Two antibodies against PEDF were used: an affinity-purified rabbit polyclonal antibody (pAb) to a synthetic peptide corresponding to amino acids 327-343 of the PEDF protein (1:200; Research Genetics, Inc., Huntsville, AL) and a mouse monoclonal antibody (mAb) to a recombinant GST fusion protein containing a 155 amino acid sequence at the C-terminal of human PEDF (1:2000; Chemicon International, Inc., Temecula, CA). The company, Chemicon, indicates that the mAb does not recognize GST. The epitope on PEDF recognized by this mAb is not shared with GST. Immunocytochemistry labeling with both the pAb and mAb were repeated over 10 times, always with the same results.

Double labeling was performed with antibodies to cell-specific proteins in rods, cones, the IPM, and neurons in the inner retina (Table 2). The secondary antibodies (goat anti-rabbit or anti-mouse IgG) were labeled with Alexa Fluor 488 (green; Molecular Probes, Eugene, OR), Cy-2 (green), or Cy-3 (red; Jackson ImmunoResearch Laboratories, Inc., West Grove, PA). Nuclei were stained with propidium iodide (red, 2 mg/ml) or 4',6'-diamidino-2-phenylindole (blue, DAPI, 1 mg/ml), both from Molecular Probes. Control sections were treated in the same way with omission of primary antibody or preincubation of the primary antibody overnight at 4 °C with PEDF protein at a concentration of 50 mg/ml.

The immunolabeled sections were examined with a Leica DMR epifluorescence microscope (Deerfield, IL) and photographed with Kodak (Rochester, NY) Elite Chrome film, ASA 400, or with a laser scanning confocal microscope (BioRad MRC-600, Richmond, CA). Images were digitized with a flat bed scanner (Saphir HiRes, Heidelberg CPS GmbH, Bad Homburg, Germany) using LinoColor Elite 5.1 software (Heidelberg CPS GmbH), imported into a graphics program (Photoshop 5.0, Adobe, San Jose, CA) and dye-sublimation prints were generated.

Western blot analysis

Two normal, unfixed globes from donors aged 74 and 78 were frozen at 4.5 and 20 h post mortem, respectively (Table 1). The globes were stored at -80 °C, thawed, and dissected immediately. The corneas, iris/ciliary bodies and retinas were processed for Western blot analysis using pAb or mAb anti-PEDF. Briefly, the tissue was homogenized in 0.09 M Tris-HCl (pH 7.4) containing 2.8% sodium dodecyl sulfate (SDS); 15 mg of pepstatin A, leupeptin, and aprotinin; 0.1 mM phenylmethylsulfonyl fluoride; and 2 mM EDTA. Protein concentration was determined using a BCA reagent kit (Pierce, Rockford, IL). Eight ml of 0.09 M Tris-HCl and 40 ml of a second buffer (0.4% bromophenol blue and 4.7% 2-mercaptoethanol in 9.4% glycerol) were added to 92 ml of each sample containing 15 mg cornea, 30 mg iris and ciliary body, or 30 mg retina homogenate [26]. Samples were boiled (5 min) and each sample (140 ml) was electrophoresed on a 10% SDS polyacrylamide gel for 4 h at 60 mA.

Positive control protein was isolated from conditioned media of 293 cells, 48 h post-transfection with pCMV-PEDF-6xHis using a Xpress purification kit (Invitrogen, Carlsbad, CA). Briefly, ten 150 mm plates of 293 cells were transiently transfected with pCMV-PEDF-6xHis using the CalPhos mammalian Transfection kit (Clonetech, Palo Alto, CA). Forty-eight h post-transfection, the media were collected and incubated with 2 ml ProBond Resin (Invitrogen, Carlsbad, CA) for 3 h with agitation. PEDF-6xHis was eluted from the resin using increasing concentrations of imidazole (50 mM, 200 mM, 350 mM, and 500 mM). The pCMV-PEDF-6xHis plasmid had been generated by cutting pCEP4-PEDF (kindly provided by N. Bouck) [14] with SalI, resulting in a cassette containing cDNA for 6xHis tagged human PEDF driven by a CMV promoter. This cassette was ligated (blunt ends) to psub201 [27] that had been digested with SnaB1 and EcoR1, followed by gel purification of the plasmid backbone. Accuracy of the cloning was confirmed by restriction digest and sequencing. Separated proteins were transferred to nitrocellulose membranes using a BioRad semi-dry transfer apparatus (1 h at 20 V), and were blocked for 10 min using 5% skim milk in incubation buffer (Tris-buffered saline, 0.1% Tween-20). Membranes were incubated overnight at 4 °C with rabbit pAb anti-PEDF (1:500) diluted in Tris buffer with 0.5% milk. The membranes were washed extensively with Tris buffer containing 0.1% Tween-20 and incubated for 2 h at room temperature with peroxidase-conjugated goat anti-rabbit IgG (Jackson ImmunoResearch Laboratories, Inc.) at a dilution of 1:5000 in Tris buffer containing 0.5% milk. The membranes were washed extensively with Tris buffer containing 0.1% Tween-20 and treated with ECL Western blotting detection reagents (Amersham Pharmacia Biotech UK Limited, Buckinghamshire, UK). Following ECL detection, the membranes were stripped with 0.2 M glycine, pH 2.2 in 0.1% SDS in 1% Tween-20 and exposed to mAb anti-PEDF at 1:500 dilution. The remaining procedures were identical to those described above with the exception that a peroxidase-conjugated goat anti-mouse secondary antibody (1:5000) was used.

In situ hybridization

The pCMV-PEDF-6xHis (see above) was cut with NheI, blunted and then cut with EcoRI. The resultant fragment was gel purified and ligated to pBluescript KS(+/-) (Stratagene, La Jolla, CA) cut with BamHI, blunted, and then cut with EcoRI. This plasmid was digested with KpnI and religated in order to remove the 3' end of the cDNA, leaving a 356 bp insert. The plasmid was linearized with SpeI for antisense probe generation and NcoI for sense probe generation and transcribed with T3 and T7, respectively, using a DIG RNA labeling kit (Roche Pharmaceuticals, Indianapolis, IN). In situ hybridization was performed on 12 mm cryosections of normal human retina (99-11-23, a surgical specimen), as previously described [28], except the sections were digested in proteinase K for 20 min and the hybridization temperature was 65 °C.

Results

Fetal human retinas

Fetal retina samples treated with secondary antibodies but no primary antibody showed little or no autofluorescence of the retina and RPE (Figure 1A). The antigen-adsorbed controls were also negative (not shown).

When possible, observations were made on the most central regions of the fetal retinas. Developing retinas aged 7.4 to 21.5 Fwks treated with pAb anti-PEDF showed strong labeling of a band in the outermost part of the retina (Figure 1B). Double labeling with pAb anti-PEDF and mAb anti-interphotoreceptor retinoid-binding protein (IRBP) demonstrated that the PEDF-positive band in the outermost retina was in the cytoplasm of the differentiating photoreceptors and not the IPM (Figure 1B). At ages 14.3 to 21.5 Fwks, the IRBP-positive band was thinner and external to the pAb PEDF-positive band (Figure 1B). Double labeling with the cone-specific marker, mAb 7G6, revealed that the PEDF-labeling was in the apical cytoplasm of the differentiating cones, including their developing inner segments at 21.5 Fwks (Figure 1C,D). Double labeling with pAb anti-PEDF and mAb anti-rhodopsin demonstrated small rods with rhodopsin-positive surface membranes at 21.5 Fwks; these rods were not labeled with pAb anti-PEDF (Figure 1E,F).

From 7.4 Fwks onward, pAb anti-PEDF also labeled fine granules in the RPE, scattered cells in the neuroblastic layer and numerous cells in the GCL (Figure 1B,C,D). From 18 Fwks onward, some pAb PEDF-labeled cells were present at the level of the horizontal cells in the outer part of the INL (Figure 1D). However, these PEDF-positive cells did not correspond to the differentiating horizontal cells that were labeled with anti-calbindin or anti-parvalbumin [29] (Figure 1G).

Developing retinas, ages 7.4 to 16.4 Fwks showed no labeling with mAb anti-PEDF. Treatment of fetal retinas at ages 18, 19 and 21.5 Fwks with mAb anti-PEDF produced a thin labeled band between the neural retina and RPE; this band was also positive with pAb anti-IRBP (Figure 1H). No labeling was seen with mAb anti-PEDF in the inner retina at these stages of development.

Adult human retinas

Control sections treated with secondary antibodies but no primary antibody had strongly autofluorescent lipofuscin granules in the RPE (Figure 2A). Because of these highly autofluorescent granules, it was not possible to determine if other cytoplasmic areas of the adult RPE cells were reactive with mAb or pAb anti-PEDF. The antigen-adsorbed pAb anti-PEDF controls were negative except weak labeling of photoreceptor outer segments (OS, Figure 2B).

In the adult human retinas, all cells showed the expected labeling patterns with the cell-specific antibodies used in this study (see [30]). All cones had cytoplasmic and OS labeling with mAb 7G6, and all rod OS were labeled with both mAb (4D2) and pAb anti-rhodopsin. The IPM was labeled with both mAb and pAb anti-IRBP.

The pAb anti-PEDF produced strong immunolabeling of the rod OS and a few cone OS (Figure 2C,D). Most cone OS, identified by mAb 7G6, were PEDF-negative (Figure 2D). Double labeling with anti-PEDF and anti-calbindin, which labels the cytoplasm of red/green but not blue cones [31], revealed that the PEDF-positive cone OS were those of the blue cones (Figure 2E). However, weak labeling of the rod and blue cone OS persisted after the pAb anti-PEDF was adsorbed with the immunizing antigen (Figure 2B), indicating that this label may not be authentic PEDF.

The rod and cone inner segments and synapses were positive with pAb anti-PEDF, and the nuclei of the rods but not the cones were PEDF-positive (Figure 2C). Some cells in the INL and GCL had PEDF-positive cytoplasm but negative nuclei (Figure 2C). Discrete laminae in the inner plexiform layer and cells throughout the choroid were also positive for PEDF (Figure 2C). The labeling of the rod nuclei, INL and GCL cells, inner plexiform layer and choroid were absent with adsorbed pAb anti-PEDF (Figure 2B).

Adult retinas fixed with glutaraldehyde showed no immunolabeling with mAb anti-PEDF. Treatment with mAb anti-PEDF of the retina fixed with only paraformaldehyde labeled the IPM, identified by double labeling with anti-IRBP (Figure 2F).

Human anterior segments

The corneal epithelium and endothelium, ciliary body nonpigmented epithelium and muscle were strongly positive with pAb anti-PEDF from age 8.4 Fwks onward (not shown). The adult corneal epithelium and endothelium (Figure 3A) and the ciliary body non-pigmented epithelium and muscle (Figure 3C) were strongly positive with pAb anti-PEDF but were negative with mAb anti-PEDF. Tissues of the anterior segments were negative when no primary antibodies were used or when the pAb anti-PEDF was pre-adsorbed with PEDF (Figure 3B). As noted above, labeling of the IPM with mAb anti-PEDF was fixation-sensitive and not detectable in glutaraldehyde-fixed tissues. The anterior segments used for immunocytochemistry had been fixed with only paraformaldehyde, but this may have masked reactivity for PEDF, which was clearly detected in these tissues by Western analysis.

Western blotting

The same results were obtained with tissues from the two adult human eyes used for western blotting. The pAb anti-PEDF (Figure 4, "poly") recognized a band at about 50 kDa corresponding to the positive control PEDF-6xHis fusion protein (Figure 4, "+"). The predominant band recognized by this antibody in cornea, retina and iris/ciliary body was about 46 kDa, although other bands close to this size were also present (Figure 4). The mAb anti-PEDF (Figure 4, "mono") produced similar results as the pAb except it recognized only one band at about 46 kDa in retina.

In situ hybridization

The antisense PEDF probe applied to sections of adult human retina produced labeling of the RPE, photoreceptors, INL cells and GCL cells (Figure 5A). Sections treated with the sense probe showed no specific labeling (Figure 5B).

Discussion

To provide normal vision, the human eye contains precisely ordered neurons and blood vessels. Neurons of the retina are highly differentiated and organized into regular layers from late fetal development throughout adult life. The vasculature of the eye is also established early in development and maintenance of this blood supply is essential for normal visual function. Certain adult anterior segment structures (cornea, anterior chamber and lens) and the vitreous are normally free of blood vessels, as their transparency is essential for light to reach the retina. In the retina, branches of the central retinal blood vessels nourish the inner layers of the retina but do not penetrate into the vitreous. The photoreceptor layer is not vascularized, deriving metabolic support from the underlying RPE and choriocapillaris.

A number of angiogenic and anti-angiogenic factors have been identified in the eye and several of these appear to play major roles in normal development and in diseases. Basic fibroblast growth factor (bFGF), for example, is an angiogenic factor in many tissues and expression of this gene in the retina is known to be developmentally regulated [32] and altered by local injury [33]. Vascular endothelial growth factor (VEGF) is also found in the eye, and its levels parallel neovascularization disease states [34]. Experimental delivery of either of these two factors to ocular tissues produces dramatic neovascularization [35].

Similarly, several inhibitors of angiogenesis are normally present in the eye. A factor that has received a great deal of attention is PEDF, a protein originally identified in human RPE-conditioned medium [1,2]. PEDF inhibits endothelial cell migration and corneal neovascularization and has stronger anti-angiogenic activity than any of the other known angiogenesis inhibitors, including thrombospondins-1 and 2, endostatin, proliferin-related protein, retinoic acid and angiostatin. Notably, PEDF has 27 times more anti-angiogenic activity than angiostatin [14]. PEDF is also neurotrophic and promotes neuronal differentiation [1-3,16,19].

As a first step in evaluating the potential anti-angiogenic and neurotrophic properties of PEDF, we used immunocytochemistry to localize this protein at different stages of human ocular development. We found high levels of PEDF in developing and mature corneas, supporting data [14] that PEDF is the main anti-angiogenic factor in this tissue, and confirmed a previous report that PEDF is present in human ciliary body [23].

We also localized PEDF in developing human retinas. We had expected to find high levels of PEDF in the RPE, as this protein was first discovered in RPE-conditioned medium (see above), and mRNA for PEDF was identified in the RPE of a fetal human eye [22]. Our immunocytochemical results with pAb anti-PEDF suggest that the factor is synthesized by human RPE and cone cells, as well as cells of the inner retina, during development. Our findings with mAb anti-PEDF, which recognizes the C-terminal of PEDF, are consistent with secretion of PEDF into the IPM in fetal and adult human eyes.

In adult retinas, immunolabeling with pAb anti-PEDF demonstrated the protein in rods and cones, as well as cells of the inner retina (possible PEDF-reactivity in the RPE could not be determined because of the high autofluorescence of the lipofuscin granules). In agreement with the immunofluorescence labeling, mRNA for PEDF was localized to the RPE, photoreceptors, inner nuclear layer cells and ganglion cells in adult human retina by in situ hybridization (Figure 5) [36]. Immunolocalization of PEDF in the IPM of adult retinas by means of the mAb, which presumably recognizes a different epitope, agrees with previous biochemical studies and is consistent with the putative anti-angiogenic activity of PEDF. Changes in PEDF levels in this region in human maculas with AMD may contribute to development of choroidal neovascularization.

PEDF also promotes neuronal differentiation in vitro and survival in vivo. In photoreceptors, PEDF may function to maintain the highly differentiated state and survival of these fragile cells. Similar effects may occur in the inner layers of developing and adult human retinas. Many questions remained unanswered about the role(s) of PEDF in the eye [37]. Further studies, in progress, are aimed at evaluating changes in PEDF expression in retinal degenerative and neovascular disease states.

Acknowledgements

The authors thank the scientists listed in the Methods section for providing antibodies; the Foundation Fighting Blindness, the University of Washington Lions Eye Bank, and the University of Washington Human Embryology Laboratory for providing human eyes; Drs. N. Syed and J. Tombran-Tink for scientific input, Dr. N. Dejneka for biochemistry advice, and Mr. J. Bouy for technical help. Supported by the Foundation Fighting Blindness (JB and AHM), Hunt Valley, MD; the Milton and Ruth Steinbach Foundation (JB), New York, NY; the Paul and Evanina Bell Mackall Foundation Trust (AHM), New York, NY; NIH Grant EY12156 (JB), Bethesda, MD; the Kayser Award (AH); Research to Prevent Blindness, Inc. (AH, JB, and AHM), New York, NY; and the F. M. Kirby Foundation (JB), Morristown, NJ.

References

1. Tombran-Tink J, Johnson LV. Neuronal differentiation of retinoblastoma cells induced by medium conditioned by human RPE cells. Invest Ophthalmol Vis Sci 1989; 30:1700-7.

2. Tombran-Tink J, Chader GJ, Johnson LV. PEDF: a pigment epithelium-derived factor with potent neuronal differentiative activity. Exp Eye Res 1991; 53:411-4.

3. Steele FR, Chader GJ, Johnson LV, Tombran-Tink J. Pigment epithelium-derived factor: neurotrophic activity and identification as a member of the serine protease inhibitor gene family. Proc Natl Acad Sci U S A 1993; 90:1526-30.

4. Wu YQ, Notario V, Chader GJ, Becerra SP. Identification of pigment epithelium-derived factor in the interphotoreceptor matrix of bovine eyes. Protein Expr Purif 1995; 6:447-56.

5. Cayouette M, Smith SB, Becerra SP, Gravel C. Pigment epithelium-derived factor delays the death of photoreceptors in mouse models of inherited retinal degenerations. Neurobiol Dis 1999; 6:523-32.

6. Taniwaki T, Becerra SP, Chader GJ, Schwartz JP. Pigment epithelium-derived factor is a survival factor for cerebellar granule cells in culture. J Neurochem 1995; 64:2509-17.

7. Taniwaki T, Hirashima N, Becerra SP, Chader GJ, Etcheberrigaray R, Schwartz JP. Pigment epithelium-derived factor protects cultured cerebellar granule cells against glutamate-induced neurotoxicity. J Neurochem 1997; 68:26-32.

8. Araki T, Taniwaki T, Becerra SP, Chader GJ, Schwartz JP. Pigment epithelium-derived factor (PEDF) differentially protects immature but not mature cerebellar granule cells against apoptotic cell death. J Neurosci Res 1998; 53:7-15.

9. DeCoster MA, Schabelman E, Tombran-Tink J, Bazan NG. Neuroprotection by pigment epithelial-derived factor against glutamate toxicity in developing primary hippocampal neurons. J Neurosci Res 1999; 56:604-10.

10. Houenou LJ, D'Costa AP, Li L, Turgeon VL, Enyadike C, Alberdi E, Becerra SP. Pigment epithelium-derived factor promotes the survival and differentiation of developing spinal motor neurons. J Comp Neurol 1999; 412:506-14.

11. Bilak MM, Corse AM, Bilak SR, Lehar M, Tombran-Tink J, Kuncl RW. Pigment epithelium-derived factor (PEDF) protects motor neurons from chronic glutamate-mediated neurodegeneration. J Neuropathol Exp Neurol 1999; 58:719-28.

12. Jablonski MM, Tombran-Tink J, Mrazek DA, Iannaccone A. Pigment epithelium-derived factor supports normal development of photoreceptor neurons and opsin expression after retinal pigment epithelium removal. J Neurosci 2000; 20:7149-57.

13. Alberdi E, Aymerich MS, Becerra SP. Binding of pigment epithelium-derived factor (PEDF) to retinoblastoma cells and cerebellar granule neurons. Evidence for a PEDF receptor. J Biol Chem 1999; 274:31605-12.

14. Dawson DW, Volpert OV, Gillis P, Crawford SE, Xu H, Benedict W, Bouck NP. Pigment epithelium-derived factor: a potent inhibitor of angiogenesis. Science 1999; 285:245-8.

15. Stellmach VV, Crawford SE, Zhou W, Bouck N. Prevention of ischemia-induced retinopathy by the natural ocular antiangiogenic agent pigment epithelium-derived factor. Proc Natl Acad Sci U S A 2001; 98:2593-7.

16. Tombran-Tink J, Pawar H, Swaroop A, Rodriguez I, Chader GJ. Localization of the gene for pigment epithelium-derived factor (PEDF) to chromosome 17p13.1 and expression in cultured human retinoblastoma cells. Genomics 1994; 19:266-72.

17. Goliath R, Tombran-Tink J, Rodriquez IR, Chader G, Ramesar R, Greenberg J. The gene for PEDF, a retinal growth factor, is a prime candidate for retinitis pigmentosa and is tightly linked to the RP13 locus on chromosome 17p13.3. Mol Vis 1996; 2:5 <http://www.molvis.org/molvis/v2/a5/>.

18. Koenekoop R, Pina AL, Loyer M, Davidson J, Robitaille J, Maumenee I, Tombran-Tink J. Four polymorphic variations in the PEDF gene identified during the mutation screening of patients with Leber congenital amaurosis. Mol Vis 1999; 5:10 <http://www.molvis.org/molvis/v5/a10/>.

19. Becerra SP. Structure-function studies on PEDF. A noninhibitory serpin with neurotrophic activity. Adv Exp Med Biol 1997; 425:223-37.

20. Becerra SP, Sagasti A, Spinella P, Notario V. Pigment epithelium-derived factor behaves like a noninhibitory serpin. Neurotrophic activity does not require the serpin reactive loop. J Biol Chem 1995; 270:25992-9.

21. Perez-Mediavilla LA, Chew C, Campochiaro PA, Nickells RW, Notario V, Zack DJ, Becerra SP. Sequence and expression analysis of bovine pigment epithelium-derived factor. Biochim Biophys Acta 1998; 1398:203-14.

22. Tombran-Tink J, Shivaram SM, Chader GJ, Johnson LV, Bok D. Expression, secretion, and age-related downregulation of pigment epithelium-derived factor, a serpin with neurotrophic activity. J Neurosci 1995; 15:4992-5003.

23. Ortego J, Escribano J, Becerra SP, Coca-Prados M. Gene expression of the neurotrophic pigment epithelium-derived factor in the human ciliary epithelium. Synthesis and secretion into the aqueous humor. Invest Ophthalmol Vis Sci 1996; 37:2759-67.

24. Tombran-Tink J, Mazuruk K, Rodriguez IR, Chung D, Linker T, Englander E, Chader GJ. Organization, evolutionary conservation, expression and unusual Alu density of the human gene for pigment epithelium-derived factor, a unique neurotrophic serpin. Mol Vis 1996; 2:11 <http://www.molvis.org/molvis/v2/a11/>.

25. Li ZY, Kljavin IJ, Milam AH. Rod photoreceptor neurite sprouting in retinitis pigmentosa. J Neurosci 1995; 15:5429-38.

26. Frederick JM, Krasnoperova NV, Hoffmann K, Church-Kopish J, Ruther K, Howes K, Lem J, Baehr W. Mutant rhodopsin transgene expression on a null background. Invest Ophthalmol Vis Sci 2001; 42:826-33.

27. Samulski RJ, Chang LS, Shenk T. A recombinant plasmid from which an infectious adeno-associated virus genome can be excised in vitro and its use to study viral replication. J Virol 1987; 61:3096-101.

28. Surace EM, Angeletti B, Ballabio A, Marigo V. Expression pattern of the ocular albinism type I (Oa1) gene in the murine retinal pigment epithelium. Invest Ophthalmol Vis Sci 2000; 41:4333-7.

29. Wassle H, Dacey DM, Haun T, Haverkamp S, Grunert U, Boycott BB. The mosaic of horizontal cells in the macaque monkey retina: with a comment on biplexiform ganglion cells. Vis Neurosci 2000; 17:591-608.

30. John SK, Smith JE, Aguirre GD, Milam AH. Loss of cone molecular markers in rhodopsin-mutant human retinas with retinitis pigmentosa. Mol Vis 2000; 6:204-15 <http://www.molvis.org/molvis/v6/a28/>.

31. Ahnelt PK, Kolb H. The mammalian photoreceptor mosaic-adaptive design. Prog Retin Eye Res 2000; 19:711-77.

32. Ishigooka H, Kitaoka T, Boutilier SB, Bost LM, Aotaki-Keen AE, Tablin F, Hjelmeland LM. Developmental expression of bFGF in the bovine retina. Invest Ophthalmol Vis Sci 1993; 34:2813-23.

33. Faktorovich EG, Steinberg RH, Yasumura D, Matthes MT, LaVail MM. Basic fibroblast growth factor and local injury protect photoreceptors from light damage in the rat. J Neurosci 1992; 12:3554-67.

34. Aiello LP, Avery RL, Arrigg PG, Keyt BA, Jampel HD, Shah ST, Pasquale LR, Thieme H, Iwamoto MA, Park JE, et al. Vascular endothelial growth factor in ocular fluid of patients with diabetic retinopathy and other retinal disorders. N Engl J Med 1994; 331:1480-7.

35. Chen C, Parangi S, Tolentino MJ, Folkman J. A strategy to discover circulating angiogenesis inhibitors generated by human tumors. Cancer Res 1995; 55:4230-3.

36. Behling KC, Surace EM, Milam AH, Bennett J. Endogenous and exogenous expression of pigment epithelium derived factor (PEDF) in the retina. Invest Ophthalmol Vis Sci 2001; 42:S771.

37. Chader GJ. PEDF: Raising both hopes and questions in controlling angiogenesis. Proc Natl Acad Sci U S A 2001; 98:2122-4.