Neuroretinal Cells Research Articles

Ocular tissue interactions during the development of human fetal neuroretina grafted into nude mice.

To determine the roles of different ocular tissues in the development of the human fetal neuroretina, a study ethically and technically impossible in human subjects, human embryonic and fetal retinas were heterotopically implanted into nude mice. Ninety-five eyeballs were obtained from legally aborted 6- to 7-week-old embryos or 8- to 10-week-old fetuses. Ten isolated neuroretinas with vitreous but without pigment epithelium, 20 half-eyeballs and 70 intact eyeballs, of which 12 had a thick layer of periocular tissue, were microsurgically grafted. Five intact eyeballs were used for reference. Over a period of 1-245 days, all of the grafts were removed for light and electron microscopy observations. All of the isolated neuroretinas had disappeared by the second day after transplantation. Grafts of the posterior section of the eyeball contained only some clusters of pigment epithelium, occasionally covered with undifferentiated neuroretinal cells. Grafts of the retrolental section of the eyeball contained small areas of dysplasic neuroretina with folds and rosettes. Grafts of the 70 intact eyeballs were successful, but only 26 showed normal histological organization of the choriocapillaris, the retinal pigment epithelium and the neuroretina in the posterior part of the posterior chamber. Photoreceptor differentiation was evident in these retinas after approximately 80 days of transplantation and was complete after 166 days. Their anterior part was always dysplasic, with occasional ciliary differentiation. Twenty-three grafted eyeballs had a dysplasic neuroretina with folds, rosettes and necrotized areas. Twenty-one were atrophic, 12 of which were the eyeballs grafted with periocular tissue. These results demonstrate the role of the fetal mesenchyme and pigment epithelium in the rapid revascularization, and subsequent survival and tissue organization, of the neuroretina. The stratified development of the neuroretina required a thin mesenchymal environment for revascularization of the graft by human vasculogenesis or neoangiogenesis and a normal retinal pigment epithelium for normal neuroretinal differentiation. When these conditions were not satisfied, the neuroretina disappeared or was dysplasic, partly necrotized or atrophic. This model might prove useful for a number of therapeutic or clinical studies.

Identification of RanBP2- and kinesin-mediated transport pathways with restricted neuronal and subcellular localization.

Ran-binding proteins, karyopherins, and RanGTPase mediate and impart directionality to nucleocytoplasmic transport processes. This biological process remains elusive in neurons. RanBP2 has been localized at the nuclear pore complexes and is very abundant in the neuroretina. RanBP2 mediates the assembly of a large complex comprising RanGTPase, CRM1/exportin-1, importin-beta, KIF5-motor proteins, components of the 19S cap of the 26S proteasome, ubc9 and opsin. Here, we show RanBP2 is abundant in the ellipsoid compartment of photoreceptors and RanGTPase-positive particles in cytoplasmic tracks extending away from the nuclear envelope of subpopulations of ganglion cells, suggesting RanBP2's release from nuclear pore complexes. KIF5C and KIF5B are specifically expressed in a subset of neuroretinal cells and differentially localize with RanBP2 and importin-beta in distinct compartments. The C-terminal domains of KIF5B and KIF5C, but not KIF5A, associate directly with importin-beta in a RanGTPase-dependent fashion in vivo and in vitro, indicating importin-beta is an endogenous cargo for a subset of KIF5s in retinal neurons. The KIF5 transport pathway is absent from the myoid region of a topographically distinct subclass of blue cones and the distribution of kinesin-light chains is largely distinct from its KIF5 partners. Altogether, the results identify the existence of neuronal- and subtype-specific kinesin-mediated transport pathways of importin-beta-bound cargoes to and/or from RanBP2 and indicate RanBP2 itself may also constitute a scaffold carrier for some of its associated partners. The implications of these findings in protein kinesis and pathogenesis of degenerative neuropathies are discussed.

Cloning and characterization of a novel transcription factor involved in cellular proliferation arrest: PATF.

Cell cycle withdrawal involves several transcription factors such as E2Fs members that play a key role in cell growth control. Here we describe a novel putative bZIP transcription factor isolated from the retina and involved in neuronal proliferation arrest at the terminal differentiation: PATF (Proliferation Arrest Transcription Factor). We show that PATF associates with E2F4 protein and interacts with the E2F consensus site. PATF expression increases with establishment of quiescent state. Furthermore, the nuclear PATF localization like E2F4, depends on cell growth arrest. The decrease of PATF amount, using a retroviral antisense strategy, results in pursued neuroretina cell mitosis. Our results indicate that PATF could be a new molecular signal implicated in the final neuronal cell cycle withdrawal.

Induction of postmitotic neuroretina cell proliferation by distinct Ras downstream signaling pathways.

Ras-induced cell transformation is mediated through distinct downstream signaling pathways, including Raf, Ral-GEFs-, and phosphatidylinositol 3-kinase (PI 3-kinase)-dependent pathways. In some cell types, strong activation of the Ras-Raf-MEK-extracellular signal-regulated kinase (ERK) cascade leads to cell cycle arrest rather than cell division. We previously reported that constitutive activation of this pathway induces sustained proliferation of primary cultures of postmitotic chicken neuroretina (NR) cells. We used this model system to investigate the respective contributions of Ras downstream signaling pathways in Ras-induced cell proliferation. Three RasV12 mutants (S35, G37, and C40) which differ by their ability to bind to Ras effectors (Raf, Ral-GEFs, and the p110 subunit of PI 3-kinase, respectively) were able to induce sustained NR cell proliferation, although none of these mutants was reported to transform NIH 3T3 cells. Furthermore, they all repressed the promoter of QR1, a neuroretina growth arrest-specific gene. Overexpression of B-Raf or activated versions of Ras effectors Rlf-CAAX and p110-CAAX also induced NR cell division. The mitogenic effect of the RasC40-PI 3-kinase pathway appears to involve Rac and RhoA GTPases but not the antiapoptotic Akt (protein kinase B) signaling. Division induced by RasG37-Rlf appears to be independent of Ral GTPase activation and presumably requires an unidentified mechanism. Activation of either Ras downstream pathway resulted in ERK activation, and coexpression of a dominant negative MEK mutant or mKsr-1 kinase domain strongly inhibited proliferation induced by the three Ras mutants or by their effectors. Similar effects were observed with dominant negative mutants of Rac and Rho. Thus, both the Raf-MEK-ERK and Rac-Rho pathways are absolutely required for Ras-induced NR cell division. Activation of these two pathways by the three distinct Ras downstream effectors possibly relies on an autocrine or paracrine loop, implicating endogenous Ras, since the mitogenic effect of each Ras effector mutant was inhibited by RasN17.

Insulin as a Neurotrophin

Survival of neurons is dependent on neurotrophins both in vitro and in vivo. In the absence of these growth factors, the neurons undergo apoptosis. Apoptosis occurs early in neurogenesis before target-derived growth factors are present. Díaz et al. demonstrate that, in developing chick retina, survival of the neuroretinal cells is dependent on insulin. Blockade of insulin signaling with antibodies to proinsulin or the insulin receptor in ovo increased the number of apoptotic cells. Embryonic retinas in culture exhibited the same dependence on insulin for survival and survival correlated with insulin-stimulated phosphorylation of Akt. Thus, the neurotrophic actions of insulin may be mediated by the Akt survival pathway.Díaz, B., Serna, J., De Pablo, F., and de la Rosa, E.J. (2000) In vivo regulation of cell death by embryonic (pro)insulin and the insulin receptor during early retinal neurogenesis. Development 127: 1641-1649. [Online Journal]

Insulin as a Neurotrophin

Survival of neurons is dependent on neurotrophins both in vitro and in vivo. In the absence of these growth factors, the neurons undergo apoptosis. Apoptosis occurs early in neurogenesis before target-derived growth factors are present. Díaz et al. demonstrate that, in developing chick retina, survival of the neuroretinal cells is dependent on insulin. Blockade of insulin signaling with antibodies to proinsulin or the insulin receptor in ovo increased the number of apoptotic cells. Embryonic retinas in culture exhibited the same dependence on insulin for survival and survival correlated with insulin-stimulated phosphorylation of Akt. Thus, the neurotrophic actions of insulin may be mediated by the Akt survival pathway. Díaz, B., Serna, J., De Pablo, F., and de la Rosa, E.J. (2000) In vivo regulation of cell death by embryonic (pro)insulin and the insulin receptor during early retinal neurogenesis. Development 127 : 1641-1649. [Online Journal]

The Transplantation of Human Fetal Neuroretinal Cells in Advanced Retinitis Pigmentosa Patients: Results of a Long-Term Safety Study

The purpose of this study was to determine the long-term safety of transplanting human fetal neuroretinal cells (14 to 18 week gestational age) into a series of patients with advanced retinitis pigmentosa (RP). After obtaining informed consent, both hosts and mothers of donors were screened for transmissible diseases. Pre- and postoperative clinical exams, visual acuity, electroretinograms, and fluorescein angiograms were performed and visual field testing was attempted in each case. Surgically, an anterior approach through pars plana ciliaris was used. A retinotomy was performed in the paramacular area and a two-function cannula was introduced into the subretinal space to deliver a suspension of donor cells. The cell suspension carried approximately 4000 cells/μl; the volume injected did not exceed 150 μl. The patients were examined for periods ranging from 12 to 40 months posttransplantation. To date, no evidence of inflammation, infection, or overt rejection of the graft was noted in the host eye, neither was any change observed in the contralateral, unoperated eye. In conclusion, neuroretinal cells were injected into the subretinal space of 14 patients with advanced RP with no clinical appearance of detrimental effects at the time of surgery or up to 40 months postinjection except in 1 patient who developed retinal detachment. This sets the stage for a phase II clinical trial to determine the possible beneficial effects of this procedure in patients blinded by degenerative retinal disease.

Intraocular transplantation of E1A-immortalized retinal precursor cells

The purpose of this study was to examine the effect of the ocular environment on the survival, tumorigenicity, and phenotypic marker expression of immortalized retinal precursor cells transplanted into immunocompetent adult and neonatal Sprague–Dawley rats. E1A-NR.3, a rat immortalized retinal precursor cell culture, was used as an inexhaustible source of experimental graft material. These cells were prelabeled with the fluorescent marker diI (1,1′-dioctadecyl-3,3,3′,3′-tetramethylindocarbocyanine perchlorate) and transplanted intravitreally (50,000 cells per μL) into 11 adult and 31 neonatal Sprague–Dawley rat eyes. At 1 mo posttransplant, animals were sacrificed and retinal tissue sections examined histologically for the presence of grafted cells, signs of tumor formation, and retinal phenotypic marker expression. No obvious signs of tumor formation or rejection were seen in a total of 42 eyes in the immunocompetent hosts. Our results indicate that E1A-NR.3 cells survive at least 1 month in vivo, and can migrate from the vitreous into neuroretinal cell layers. Subpopulations of surviving grafted cells were seen to express photoreceptor markers rhodopsin and recoverin comparably between in vitro and in vivo conditions. However, the number of cells immunoreactive for vimentin and E1A decreased significantly under in vivo conditions. This report represents the first experimental intravitreal transplantation of E1A-immortalized retinal precursor cells into adult and neonatal rats. The intraocular location and environment appears to affect phenotypic expression of surviving grafted cells, especially with respect to vimentin and E1A expression. The fact that E1A-NR.3 cells survived intraocularly at least 1 mo without tumor formation suggests that the cells may continue to be useful for further in vivo studies of experimental retinal transplantation, and effects of histological location on retinal cell phenotype and histogenesis in immunocompetent hosts.

Intraocular transplantation of E1A-immortalized retinal precursor cells.

The purpose of this study was to examine the effect of the ocular environment on the survival, tumorigenicity, and phenotypic marker expression of immortalized retinal precursor cells transplanted into immunocompetent adult and neonatal Sprague-Dawley rats. EIA-NR.3, a rat immortalized retinal precursor cell culture, was used as an inexhaustible source of experimental graft material. These cells were prelabeled with the fluorescent marker dil (1,1'-dioctadecyl-3,3,3',3'-tetramethylindocarbocyanine perchlorate) and transplanted intravitreally (50,000 cells per microL) into 11 adult and 31 neonatal Sprague-Dawley rat eyes. At 1 mo posttransplant, animals were sacrificed and retinal tissue sections examined histologically for the presence of grafted cells, signs of tumor formation, and retinal phenotypic marker expression. No obvious signs of tumor formation or rejection were seen in a total of 42 eyes in the immunocompetent hosts. Our results indicate that EIA-NR.3 cells survive at least 1 month in vivo, and can migrate from the vitreous into neuroretinal cell layers. Subpopulations of surviving grafted cells were seen to express photoreceptor markers rhodopsin and recoverin comparably between in vitro and in vivo conditions. However, the number of cells immunoreactive for vimentin and E1A decreased significantly under in vivo conditions. This report represents the first experimental intravitreal transplantation of E1A-immortalized retinal precursor cells into adult and neonatal rats. The intraocular location and environment appears to affect phenotypic expression of surviving grafted cells, especially with respect to vimentin and E1A expression. The fact that E1A-NR.3 cells survived intraocularly at least 1 mo without tumor formation suggests that the cells may continue to be useful for further in vivo studies of experimental retinal transplantation, and effects of histological location on retinal cell phenotype and histogenesis in immunocompetent hosts.

Regulation of E2F-1 gene expression in avian cells.

E2F-1 is the prototype of a family of transcription factors playing a central role in the control of cell proliferation and apoptosis. E2F DNA binding activity is down-regulated during cellular differentiation, which is correlated with cell division arrest. We report here that the expression of E2F-1 itself is down-regulated in the developing quail neural retina between embryonic days E8-E10. This event occurs just after the massive arrest of the quail neuroretina cell division (E7-E8). To gain further insight into the regulatory mechanisms monitoring E2F-1 expression in differentiating neurons, we have cloned the quail E2F-1 promoter. In vivo DNA footprintings of this promoter have shown that a number of potential SP-1 and C/EBP response elements are constitutively occupied in the entire quail neuroretina of E5 and E14, whereas the two consensus palindromic E2F binding sites are only protected at E5. This suggests that these E2F elements participate in down-regulation of E2F-1 gene expression during avian neuroretina development. CAT reporter assays have shown that E2F-1 in association with its partner DP-1 transactivates its own promoter, whereas p105Rb inhibits the E2F-1 promoter. Both E2F-1/DP-1 and p105Rh require the presence of the E2F binding sites to mediate their effects. However, experiments performed with deletion mutants of the promoter strongly suggest that other regions located upstream of the E2F binding sites also mediate part of the E2F-1 transactivating effect on its own promoter. Altogether, these results suggest that the down-regulation of E2F-1 gene expression in differentiating neurons could be due, in part, to the E2F/Rb complexes binding to the E2F-1 promoter.

Toxicity of cholesterol oxides on cultured neuroretinal cells

PURPOSE. By using nerve growth factor-differentiated PC12 cells as a model for sympathetic neurons, we have recently shown that cholesterol oxides are toxic to cells of neural origin. Since lipid metabolism is known to be involved in some pathological conditions associated with the visual system, we sought to extend this line of research by studying the potential cytotoxicity of cholesterol oxides on primary cultures derived from neuroretinas. METHODS. Dissociated cultures derived from neuroretinas of 1-day-old Sprague-Dawley rats were used in this series of studies. Immunohistochemical staining was used to identify neuronal and glial cell types in these cultures. MTT assay was used to determine the cytotoxicity of cholesterol oxides, including 7-beta-OH-, 7-keto-, 19-OH-, 22(R)-OH-, 22(S)-OH- and 25-OH-cholesterol. RESULTS. Among the cholesterol oxides tested, 7-beta-OH- and 7-keto-cholesterol were the most effective in causing cell death, such that 20 µg/ml (50 µM) of these agents killed approximately 80% of cells in 3 days. A time-dependent experiment indicated that 10 µg/ml of 7-beta-cholesterol was able to kill 50% of cells in approximately 5 h. A number of pharmacological agents were tested for their ability to prevent cell death induced by cholesterol oxides. Among them, vitamin E and methyl-beta-cyclodextrin were able to prevent cholesterol oxide-induced cell death in a dose-dependent manner. CONCLUSIONS. These results suggest that, in addition to causing pathological changes in cells directly involved in atherosclerosis, cholesterol oxides may be toxic to cells derived from neuroretinas.

Retinal Engineering: Engrafted Neural Cell Lines Locate in Appropriate Layers.

Abstract A major question in central nervous system development, including the neuroretina, is whether migrating cells express cues to find their way and settle at specific locations. We have transplanted quail neuroretinal cell lines QNR/D, a putative amacrine or ganglion cell, and QNR/K2, a putative Muller cell into chicken embryo eyes. Implanted QNR/D cells migrate only to the retinal ganglion and amacrine cell layers and project neurites in the plane of retina; in contrast, QNR/K2 cells migrate through the ganglion and amacrine layers, locate in the inner nuclear layer, and project processes across the retina. These data show that QNR/D and QNR/K2 cell lines represent distinct neural cell types, suggesting that migrating neural cells express distinct address cues. Furthermore, our results raise the possibility that immortalized cell lines can be used for replacement of specific cell types and for the transport of genes to given locations in neuroretina.

P34(cdc2) and mitotic cyclin expression in the developing quail neuroretina.

After an initial proliferation phase, neurons of the central nervous system (CNS) of higher eukaryotes remain postmitotic during their entire lifespan. This requires that a very stringent control be exerted on the cell division apparatus, whose molecular mechanisms remain quite elusive. Here we have used quail neuroretina as a model to study the control of cell division in the developing CNS. In vertebrates, embryonic neuroretinal cells (NR cells) stop their proliferation at different times depending on the cell type. Most NR cells in the quail embryo become postmitotic between E7 and E8. To acquire a better understanding of the molecular events leading to quiescence in NR cells, we have analyzed the expression of cdc2 and of two activators of p34(cdc2): cyclin A and cyclin B2 in the developing neuroretina. We report that these three proteins are downregulated between E7 and E9, suggesting that a common mechanism could block their transcription in differentiating neurons. We also report, using an immunohistochemical approach, that p34(cdc2) downregulation is correlated with the appearance of the microtubule-associated protein tau. These results strongly suggest that inhibition of cdc2 gene expression is closely linked to the achievement of terminal differentiation in neurons. However, we also show that postmitotic ganglion cells precursors begin to synthesize the early neuronal differentiation marker beta3-tubulin while p34(cdc2) is still detectable in these cells, suggesting that p34(cdc2) or a closely related kinase could play a role in some "young" postmitotic neurons.

Retinal engineering: engrafted neural cell lines locate in appropriate layers.

A major question in central nervous system development, including the neuroretina, is whether migrating cells express cues to find their way and settle at specific locations. We have transplanted quail neuroretinal cell lines QNR/D, a putative amacrine or ganglion cell, and QNR/K2, a putative Müller cell into chicken embryo eyes. Implanted QNR/D cells migrate only to the retinal ganglion and amacrine cell layers and project neurites in the plane of retina; in contrast, QNR/K2 cells migrate through the ganglion and amacrine layers, locate in the inner nuclear layer, and project processes across the retina. These data show that QNR/D and QNR/K2 cell lines represent distinct neural cell types, suggesting that migrating neural cells express distinct address cues. Furthermore, our results raise the possibility that immortalized cell lines can be used for replacement of specific cell types and for the transport of genes to given locations in neuroretina.

A retinal pigment epithelium-derived cell line from transgenic mouse harboring temperature-sensitive simian virus 40 large T-antigen gene.

We established a retinal pigment epithelium-derived cell line from transgenic mouse harboring temperature-sensitive simian virus 40 T-antigen gene (tsSV40T) and examined its characteristics. We enucleated both eyes from a 2-month-old transgenic mouse and removed the retinal pigment epithelial (RPE) cells and neuroretinal cells. After cloning the RPE cells, we obtained a cell line (RPET). RPET cells grew well at 33 degrees C but not at 37 degrees C, expressing on the temperature-sensitive character of tsSV40T, and maintained characters of RPE cells such as T1-tyrosinase production, phagocytosis of rod outer segments, and presence of cytokeratin, microvilli on the cell surface and lysosome-like granules around the Golgi apparatus in the cytoplasm. Conditioned medium (CM) from a culture of neuroretinal cells harboring tsSV40T was essentially required for growth. The factor(s) in CM was heat-and acid labile, but was resistant to trypsin digestion. In the presence of 3% CM, epidermal growth factor (EGF) and basic fibroblast growth factor (bFGF) and strong effects on RPET cells, whereas insulin, insulin-like growth factor I (IGF-I), and IGF-II had moderate growth effects. Interestingly, none of these growth factors stimulated the RPET cells in the absence of CM. EHS-Matrix had growth effect, whereas laminin, collagen types I and IV, and fibronectin had no marked growth effects on RPET cells. RPET cells were morphologically changed on a laminin-coated dish. They could not spread on the coated dish, and the majority of the cells floated. But when the floating cells were transferred to non-coated dishes, they immediately attached themselves. These results suggest that RPET cells are a good model for for finding novel growth factor(s) and for investigating the mechanism of cell-laminin attachment.

A novel cDNA corresponding to transcripts expressed in retina post-mitotic neurons

The long term objective of this study is to isolate genes specifically expressed at the onset of neuronal cell cycle withdrawal. As an experimental paradigm we have used a quail neuroretinal cell clone (clone K2) immortalized by a thermosensitive mutant of Rous Sarcoma Virus. K2 cells proliferate at 36°C but stop synthesizing DNA after a shift to 41.5°C. We have constructed a cDNA library from K2 cells transferred to 41.5°C and autosubtracted with RNAs from K2 cells maintained at 36°C. This strategy has led to the isolation of cDNAs which recognize mRNAs expressed in quail neuroretina (NR) during development. We report here one of these cDNAs, cDNA QN1, that hybridizes with transcripts expressed in retina neurons, in parallel with their withdrawal from the cell cycle. QN1 ORF codes for a 138 kDa polypeptide corresponding to the protein observed in Western blot analysis. A role of QN1 product(s) on neuronal quiescence is suggested by the positive effect of an antisense oligonucleotide on DNA synthesis of K2 cells.

MODULATION OF Y-79 RETINOBLASTOMA CELL-DIFFERENTIATION AND IRBP EXPRESSION BY DIBUTYRYL-CYCLIC-AMP AND LAMININ

Y-79 retinoblastoma cells grown on a poly-D-lysine-coated substratum can assume a photoreceptor-like morphology and they express the retina-specific interphotoreceptor-retinoid-binding protein (IRBP) mRNA in abundance. In contrast, cells in suspension culture remain morphologically undifferentiated and exhibit a much lower IRBP message level. After treatment with the differentiating agent dibutyryl cyclic AMP (dbcAMP), monolayer cultures of Y-79 cells grown in serum-containing medium show a partial morphological and biochemical differentiation towards a Muller-glial-like phenotype. This is accompanied by a decrease in IRBP mRNA level with respect to that in Y-79 cells grown on poly-D-lysine. Attached cells which had been pretreated with laminin for three days do not demonstrate an altered IRBP mRNA level in response to dbcAMP as compared to cells not pretreated. Surprisingly, however, cells in suspension culture that were exposed to laminin and then to dbcAMP showed a 3-fold increase in IRBP mRNA level. Thus, expression of an important photoreceptor-specific marker is clearly modulated by both dbcAMP and laminin. Moreover, attachment itself may be a significant factor in determination of a neural or neuro-retinal cell phenotype.

V-erbA Cooperates with bFGF in Neuroretina Cell Transformation

We have studied the transforming ability of the oncogenes erbA and/or erbB in chicken neuroretina (CNR) cells and the effect of basic fibroblast growth factor (bFGF) on the transformed phenotype. The erbA oncogene alone was only transforming in the presence of bFGF. In contrast, cells expressing erbB as well as erbA + erbB were transformed in a bFGF-independent manner and were unresponsive to the growth factor. We studied whether other oncogenes could also block the cooperation between erbA and bFGF. Cytoplasmic or membrane-bound oncogenes ( src, ras, or millraf ) increased the transforming potential of erbA but rendered the cells unresponsive to bFGF. Conversely, the nuclear oncogenes tested (fos and myb-ets + myc) also cooperated with erbA in CNR cell transformation but the cells remained responsive to the growth factor. A likely explanation is that CNR cells carrying the cytoplasmic but not the nuclear oncogenes have already activated the bFGF signal transduction pathway.

Loss of transformed phenotype upon senescence of Rous sarcoma virus-infected chicken neuroretinal cells.

Success in obtaining permanent Rous sarcoma virus-infected chicken cell lines has been limited because of a senescence phenomenon. We show that a diminished, transformed phenotype, followed by dramatic morphological changes, precedes senescence. These changes are associated with continued expression of pp60v-src, as well as specific alterations in expression of two possible phosphorylated substrates of pp60v-src.

A choice between glycogen and delta-crystallin accumulation is made in glial cells and not influenced by overlying neurons.

Chick-embryo neuroretinal cells convert extensively into lens under low-glucose conditions, but this transdifferentiation process is blocked by high-glucose media. We have previously observed an inverse relationship between the levels of glycogen (a marker of normal retinoglial differentiation) and of delta-crystallin (a lens marker) in such cultures. However, most of the glycogen accumulated under high-glucose conditions is apparently localized in those glial (G) cells underlying clusters of neurons (N cells). We here show that glial-enriched cultures (largely depleted of N cells) both accumulate glycogen and fail to transdifferentiate in high-glucose media. Moreover, glycogen localization in groups of glial cells is unaffected by the absence of N cells. Thus the choice between normal and foreign differentiation pathways is made autonomously within the retinoglial-cell population and is not influenced significantly by the presence or absence of N cells.