Early embryonic interaction of retinal pigment epithelium and mesenchymal tissue induces conversion of pigment epithelium to neural retinal fate in the silver mutation of the Japanese quail

The neural retina and retinal pigment epithelium (RPE) diverge from the optic vesicle during early embryonic development. They originate from different portions of the optic vesicle, the more distal part developing as the neural retina and the proximal part as RPE. As the distal part appears to make contact with the epidermis and the proximal part faces mesenchymal tissues, these two portions would encounter different environmental signals. In the present study, an attempt has been made to investigate the significance of interactions between the RPE and mesenchymal tissues that derive from neural crest cells, using a unique quail mutant silver (B/B) as the experimental model. The silver mutation is considered to affect neural crest-derived tissues, including the epidermal melanocytes. The homozygotes of the silver mutation have abnormal eyes, with double neural retinal layers, as a result of aberrant differentation of RPE to form a new neural retina. Retinal pigment epithelium was removed from early embryonic eyes (before the process began) and cultured to see whether it expressed any phenotype characteristic of neural retinal cells. When RPE of the B/B mutant was cultured with surrounding mesenchymal tissue, neural retinal cells were differentiated that expressed markers of amacrine, cone or rod cells. When isolated RPE of the B/B mutant was cultured alone, it acquired pigmentation and did not show any property characteristic of neural retinal cells. The RPE of wild type quail always differentiated to pigment epithelial cells. In the presence of either acidic fibroblast growth factor (aFGF) or basic FGF (bFGF), the RPE of the B/B mutant differentiated to neural retinal cells in the absence of mesenchymal tissue, but the RPE of wild type embryos only did so in the presence of 10–40 times as much aFGF or bFGF. These observations indicate that genes responsible for the B/B mutation are expressed in the RPE as well as in those cells that have a role in the differentiation of neural crest cells. They further suggest that development of the neural retina and RPE is regulated by some soluble factor(s) that is derived from or localized in the surrounding embryonic mesenchyme and other ocular tissues, and that FGF may be among possible candidates.     

Tissue interaction between the retinal pigment epithelium and the choroid triggers retinal regeneration of the newt Cynops pyrrhogaster

Complete retinal regeneration in adult animals occurs only in certain urodele amphibians, in which the retinal pigmented epithelial cells (RPE) undergo transdifferentiation to produce all cell types constituting the neural retina. A similar mechanism also appears to be involved in retinal regeneration in the embryonic stage of some other species, but the nature of this mechanism has not yet been elucidated. The organ culture model of retinal regeneration is a useful experimental system and we previously reported RPE transdifferentiation of the newt under this condition. Here, we show that cultured RPE cells proliferate and differentiate into neurons when cultured with the choroid attached to the RPE, but they did not exhibit any morphological changes when cultured alone following removal of the choroid. This finding indicates that the tissue interactions between the RPE and the choroid are essential for the former to proliferate. This tissue interaction appears to be mediated by diffusible factors, because the choroid could affect RPE cells even when the two tissues were separated by a membrane filter. RPE transdifferentiation under the organotypic culture condition was abolished by a MEK (ERK kinase) inhibitor, U0126, but was partially suppressed by an FGF receptor inhibitor, SU5402, suggesting that FGF signaling pathway has a central role in the transdifferentiation. While IGF-1 alone had no effect on isolated RPE, combination of FGF-2 and IGF-1 stimulated RPE cell transdifferentiation similar to the results obtained in organ-cultured RPE and choroid. RT-PCR revealed that gene expression of both FGF-2 and IGF-1 is up-regulated following removal of the retina. Thus, we show for the first time that the choroid plays an essential role in newt retinal regeneration, opening a new avenue for understanding the molecular mechanisms underlying retinal regeneration.     

Activin signaling limits the competence for retinal regeneration from the pigmented epithelium

Regeneration of the retina in amphibians is initiated by the transdifferentiation of the retinal pigmented epithelium (RPE) into neural progenitors. A similar process occurs in the early embryonic chick, but the RPE soon loses this ability. The factors that limit the competence of RPE cells to regenerate neural retina are not understood; however, factors normally involved in the development of the eye (i.e. FGF and Pax6) have also been implicated in transdifferentiation. Therefore, we tested whether activin, a TGFbeta family signaling protein shown to be important in RPE development, contributes to the loss in competence of the RPE to regenerate retina. We have found that addition of activin blocks regeneration from the RPE, even during stages when the cells are competent. Conversely, a small molecule inhibitor of the activin/TGFbeta/nodal receptors can delay, and even reverse, the developmental restriction in FGF-stimulated neural retinal regeneration.     

Tyrosinase-related protein 2 promoter targets transgene expression to ocular and neural crest-derived tissues

In an effort to identify a promoter suitable for studying early ocular development, we generated transgenic mice carrying the lacZ reporter gene linked to the tyrosinase-related protein 2 (TRP2) promoter. TRP2-lacZ was expressed in early retinal pigment epithelium (RPE) and early neural crest cells in embryos. The promoter activity was robust and consistent in independent transgenic lines. The transgene was also expressed in the optic nerve and neural crest-derived neuronal cells in which the endogenous TRP2 gene is not expressed. This suggests that repressor elements may be missing in the promoter used in this study. To test whether this promoter can be used to study melanocyte development, we cross-mated TRP2-lacZ transgenic mice with mice heterozygous for the Patch (Ph) mutation. The pattern of beta-galactosidase activity in the embryos correlates well with the pigmentation phenotype in postnatal and adult Ph/+ mice. We also generated transgenic mice expressing fibroblast growth factor 9 (FGF9) directed by the TRP2 promoter and examined the effect on ocular development. Ectopic expression of FGF9 in the early embryonic RPE switched its differentiation pathway to a neuronal fate, resulting in formation of a duplicated neural retina in transgenic mice. These studies demonstrate that the TRP2 promoter is valuable for transgenic studies of ocular differentiation and development of neural crest cells.     

MEK mediates in vitro neural transdifferentiation of the adult newt retinal pigment epithelium cells: Is FGF2 an induction factor?

Adult newts can regenerate their entire retinas through transdifferentiation of the retinal pigment epithelium (RPE) cells. As yet, however, underlying molecular mechanisms remain virtually unknown. On the other hand, in embryonic/larval vertebrates, an MEK [mitogen‐activated protein kinase (MAPK)/extracellular signal‐regulated kinase (ERK) kinase] pathway activated by fibroblast growth factor‐2 (FGF2) is suggested to be involved in the induction of transdifferentiation of the RPE into a neural retina. Therefore, we examined using culture systems whether the FGF2/MEK pathway is also involved in the adult newt RPE transdifferentiation. Here we show that the adult newt RPE cells can switch to neural cells expressing pan‐retinal‐neuron (PRN) markers such as acetylated tubulin, and that an MEK pathway is essential for the induction of this process, whereas FGF2 seems an unlikely primary induction factor. In addition, we show by immunohistochemistry that the PRN markers are not expressed until the 1–3 cells thick regenerating retina, which contains retinal progenitor cells, appears. Our current results suggest that the activation of an MEK pathway in RPE cells might be involved in the induction process of retinal regeneration in the adult newt, however if this is the case, we must assume complementary mechanisms that repress the MEK‐mediated misexpression of PRN markers in the initial process of transdifferentiation.     

MEK mediates in vitro neural transdifferentiation of the adult newt retinal pigment epithelium cells: Is FGF2 an induction factor?

Adult newts can regenerate their entire retinas through transdifferentiation of the retinal pigment epithelium (RPE) cells. As yet, however, underlying molecular mechanisms remain virtually unknown. On the other hand, in embryonic/larval vertebrates, an MEK [mitogen-activated protein kinase (MAPK)/extracellular signal-regulated kinase (ERK) kinase] pathway activated by fibroblast growth factor-2 (FGF2) is suggested to be involved in the induction of transdifferentiation of the RPE into a neural retina. Therefore, we examined using culture systems whether the FGF2/MEK pathway is also involved in the adult newt RPE transdifferentiation. Here we show that the adult newt RPE cells can switch to neural cells expressing pan-retinal-neuron (PRN) markers such as acetylated tubulin, and that an MEK pathway is essential for the induction of this process, whereas FGF2 seems an unlikely primary induction factor. In addition, we show by immunohistochemistry that the PRN markers are not expressed until the 1-3 cells thick regenerating retina, which contains retinal progenitor cells, appears. Our current results suggest that the activation of an MEK pathway in RPE cells might be involved in the induction process of retinal regeneration in the adult newt, however if this is the case, we must assume complementary mechanisms that repress the MEK-mediated misexpression of PRN markers in the initial process of transdifferentiation.     

Trunk lateral cells are neural crest-like cells in the ascidian Ciona intestinalis: insights into the ancestry and evolution of the neural crest

Neural crest-like cells (NCLC) that express the HNK-1 antigen and form body pigment cells were previously identified in diverse ascidian species. Here we investigate the embryonic origin, migratory activity, and neural crest related gene expression patterns of NCLC in the ascidian Ciona intestinalis. HNK-1 expression first appeared at about the time of larval hatching in dorsal cells of the posterior trunk. In swimming tadpoles, HNK-1 positive cells began to migrate, and after metamorphosis they were localized in the oral and atrial siphons, branchial gill slits, endostyle, and gut. Cleavage arrest experiments showed that NCLC are derived from the A7.6 cells, the precursors of trunk lateral cells (TLC), one of the three types of migratory mesenchymal cells in ascidian embryos. In cleavage arrested embryos, HNK-1 positive TLC were present on the lateral margins of the neural plate and later became localized adjacent to the posterior sensory vesicle, a staging zone for their migration after larval hatching. The Ciona orthologues of seven of sixteen genes that function in the vertebrate neural crest gene regulatory network are expressed in the A7.6/TLC lineage. The vertebrate counterparts of these genes function downstream of neural plate border specification in the regulatory network leading to neural crest development. The results suggest that NCLC and neural crest cells may be homologous cell types originating in the common ancestor of tunicates and vertebrates and support the possibility that a putative regulatory network governing NCLC development was co-opted to produce neural crest cells during vertebrate evolution.     

The distribution of tenascin coincides with pathways of neural crest cell migration

The distribution of the extracellular matrix (ECM) glycoprotein, tenascin, has been compared with that of fibronectin in neural crest migration pathways of Xenopus laevis, quail and rat embryos. In all species studied, the distribution of tenascin, examined by immunohistochemistry, was more closely correlated with pathways of migration than that of fibronectin, which is known to be important for neural crest migration. In Xenopus laevis embryos, anti-tenascin stained the dorsal fin matrix and ECM along the ventral route of migration, but not the ECM found laterally between the ectoderma and somites where neural crest cells do not migrate. In quail embryos, the appearance of tenascin in neural crest pathways was well correlated with the anterior-to-posterior wave of migration. The distribution of tenascin within somites was compared with that of the neural crest marker, HNK-1, in quail embryos. In the dorsal halves of quail somites which contained migrating neural crest cells, the predominant tenascin staining was in the anterior halves of the somites, codistributed with the migrating cells. In rat embryos, tenascin was detectable in the somites only in the anterior halves. Tenascin was not detectable in the matrix of cultured quail neural crest cells, but was in the matrix surrounding somite and notochord cells in vitro. Neural crest cells cultured on a substratum of tenascin did not spread and were rounded. We propose that tenascin is an important factor controlling neural crest morphogenesis, perhaps by modifying the interaction of neural crest cells with fibronectin.     

Neural retinal regeneration in the anuran amphibian Xenopus laevis post-metamorphosis: transdifferentiation of retinal pigmented epithelium regenerates the neural retina

In urodele amphibians like the newt, complete retina and lens regeneration occurs throughout their lives. In contrast, anuran amphibians retain this capacity only in the larval stage and quickly lose it during metamorphosis. It is believed that they are unable to regenerate these tissues after metamorphosis. However, contrary to this generally accepted notion, here we report that both the neural retina (NR) and lens regenerate following the surgical removal of these tissues in the anuran amphibian, Xenopus laevis, even in the mature animal. The NR regenerated both from the retinal pigment epithelial (RPE) cells by transdifferentiation and from the stem cells in the ciliary marginal zone (CMZ) by differentiation. In the early stage of NR regeneration (5-10 days post operation), RPE cells appeared to delaminate from the RPE layer and adhere to the remaining retinal vascular membrane. Thereafter, they underwent transdifferentiation to regenerate the NR layer. An in vitro culture study also revealed that RPE cells differentiated into neurons and that this was accelerated by the presence of FGF-2 and IGF-1. The source of the regenerating lens appeared to be remaining lens epithelium, suggesting that this is a kind of repair process rather than regeneration. Thus, we show for the first time that anuran amphibians retain the capacity for retinal regeneration after metamorphosis, similarly to urodeles, but that the mode of regeneration differs between the two orders. Our study provides a new tool for the molecular analysis of regulatory mechanisms involved in retinal and lens regeneration by providing an alternative animal model to the newt, the only other experimental model.     

Complete reconstruction of the retinal laminar structure from a cultured retinal pigment epithelium is triggered by altered tissue interaction and promoted by overlaid extracellular matrices

The retina regenerates from retinal pigment epithelial (RPE) cells by transdifferentiation in the adult newt and Xenopus laevis when it is surgically removed. This was studied under a novel culture condition, and we succeeded, for the first time, in developing a complete retinal laminar structure from a single epithelial sheet of RPE. We cultured a Xenopus RPE monolayer sheet isolated from the choroid on a filter cup with gels overlaid and found that the retinal tissue structure differentiated with all retinal layers present. In the culture, RPE cells isolated themselves from the culture substratum (filter membrane), migrated, and reattached to the overlaid gel, on which they initiated transdifferentiation. This was exactly the same as observed during in vivo retina regeneration of X. laevis. In contrast, when RPE monolayers were cultured similarly without isolation from the choroid, RPE cells proliferated, but remained pigmented instead of transdifferentiating, indicating that alteration in tissue interaction triggers transdifferentiation. We then examined under the conventional tissue culture condition whether altered RPE‐choroid interaction induces Pax6 expression. Pax6 was upregulated in RPE cells soon after they were removed from the choroid, and this expression was not dependent of FGF2. FGF2 administration was needed for RPE cells to maintain Pax6 expression. From the present results, in addition to our previous ones, we propose a two‐step mechanism of transdifferentiation: the first step is a reversible process and is initiated by the alteration of the cell‐extracellular matrix and/or cell–cell interaction followed by Pax6 upregulation. FGF2 plays a key role in driving RPE cells into the second step, during which they differentiate into retinal stem cells. © 2009 Wiley Periodicals, Inc. Develop Neurobiol, 2009     

Retinoids control anterior and dorsal properties in the developing forebrain

We have previously shown that retinoic acid (RA) synthesized by the retinaldehyde dehydrogenase 2 (RALDH2) is required in forebrain development. Deficiency in RA due to inactivation of the mouse Raldh2 gene or to complete absence of retinoids in vitamin-A-deficient (VAD) quails, leads to abnormal morphogenesis of various forebrain derivatives. In this study we show that double Raldh2/Raldh3 mouse mutants have a more severe phenotype in the craniofacial region than single null mutants. In particular, the nasal processes are truncated and the eye abnormalities are exacerbated. It has been previously shown that retinoids act mainly on cell proliferation and survival in the ventral forebrain by regulating SHH and FGF8 signaling. Using the VAD quail model, which survives longer than the Raldh-deficient mouse embryos, we found that retinoids act in maintaining the correct position of anterior and dorsal boundaries in the forebrain by modulating FGF8 anteriorly and WNT signaling dorsally. Furthermore, BMP4 and FGF8 signaling are affected in the nasal region and BMP4 is ventrally expanded in the optic vesicle. At the optic cup stage, Pax6, Tbx5 and Bmp4 are ectopically expressed in the presumptive retinal pigmented epithelium (RPE), while Otx2 and Mitf are not induced, leading to a dorsal transdifferentiation of RPE to neural retina. Therefore, besides being required for survival of ventral structures, retinoids are involved in restricting anterior identity in the telencephalon and dorsal identity in the diencephalon and the retina.     

A novel antigen is shared by retinal pigment epithelium and pigmented neural crest

 Pigment cells in vertebrate embryos are formed in both the central and peripheral nervous system. The neural crest, a largely pluripotent population of precursor cells derived from the embryonic neural tube, gives rise to pigment cells which migrate widely in head and trunk.The retinal pigment epithelium is derived from the optic cup, which arises from ectoderm of the neural tube. We have generated an antibody, ips6, which stains an antigen common to pigment cells of retinal pigment epithelium and neural crest. Ips6 stains retinal pigment epithelium and choroid as well as a subset of crest cells that migrate in pathways typical of melanoblasts. Immunoreactivity is seen first in the eye and later in a subset of migrating crest cells. Crest cells in the amphibian embryo migrate along specific, stereotyped routes; ips6 immunoreactive cells are found in some but not all of these pathways. In older wild-type embryos, cells expressing ips6 appear coincident with pigment-containing cells in the flank, head, eye and embryonic gut. In older animals, staining in the eye extends to the intraretinal segment of optic nerve and interstices between photoreceptors and cells at the retinal periphery. We suggest that the ips6 antibody defines an antigen common to pigment cells of central and peripheral origin. Received: 22 January 1996/Accepted: 15 July 1996     

bHLH genes cath5 and cNSCL1 promote bFGF-stimulated RPE cells to transdifferentiate toward retinal ganglion cells

The molecular mechanism of retinal ganglion cell (RGC) genesis and development is not well understood. Published data suggest that the process may involve two bHLH genes, ath5 and NSCL1. Gain-of-function studies show that ath5 increases RGC production in the developing retina. We examined whether two chick genes, cath5 and cNSCL1, can guide retinal pigment epithelial (RPE) cells to transdifferentiate toward RGCs. Ectopic expression of cath5 and cNSCL1 in cultured chick RPE cells was achieved through retroviral transduction. cath5 alone was unable to induce de novo expression of early RGC markers, such as RA4 antigen, neurofilament (160 kDa), and a neurofilament-associated antigen. However, cath5 induced the expression of these proteins when the RPE cells were cultured with medium supplemented with bFGF. Since bFGF alone can induce only RA4 antigen, the expression of the additional RGC markers reflects a synergism between cath5 and bFGF in promoting RPE transdifferentiation toward RGCs. Morphologically, the RA4(+) cells in bFGF + cath5 cultures appeared more neuron-like than those generated by bFGF alone. cNSCL1 also promoted bFGF-stimulated RPE cells to transdifferentiate toward RGCs that expressed RA4 antigen, N-CAM, Islet-1, neurofilament, and neurofilament-associated antigen. We found that cath5 induced cNSCL1 expression, but not vice versa. Our data suggest that cath5 or cNSCL1 alone was insufficient to induce RPE transdifferentiation into RGCs, but could further neural differentiation initiated by bFGF. We propose that intrinsic factors act synergistically with extrinsic factors during RGC genesis and development.     

Posteriorization by FGF, Wnt, and retinoic acid is required for neural crest induction.

The neural crest is a unique cell population induced at the lateral border of the neural plate. Neural crest is not produced at the anterior border of the neural plate, which is fated to become forebrain. Here, the roles of BMPs, FGFs, Wnts, and retinoic acid signaling in neural crest induction were analyzed by using an assay developed for investigating the posteriorization of the neural plate. Using specific markers for the anterior neural plate border and the neural crest, the posterior end of early neurula embryos was shown to be able to transform the anterior neural plate border into neural crest cells. In addition, tissue expressing anterior neural plate markers, induced by an intermediate level of BMP activity, was transformed into neural crest by posteriorizing signals. This transformation was mimicked by bFGF, Wnt-8, or retinoic acid treatment and was also inhibited by expression of the dominant negative forms of the FGF receptor, the retinoic acid receptor, and Wnt signaling molecules. The transformation of the anterior neural plate border into neural crest cells was also achieved in whole embryos, by retinoic acid treatment or by use of a constitutively active form of the retinoic acid receptor. By analyzing the expression of mesodermal markers and various graft experiments, the expression of the mutant retinoic acid receptor was shown to directly affect the ectoderm. We thereby propose a two-step model for neural crest induction. Initially, BMP levels intermediate to those required for neural plate and epidermal specification induce neural folds with an anterior character along the entire neural plate border. Subsequently, the most posterior region of this anterior neural plate border is transformed into the neural crest by the posteriorizing activity of FGFs, Wnts, and retinoic acid signals. We discuss a unifying model where lateralizing and posteriorizing signals are presented as two stages of the same inductive process required for neural crest induction.     

Basic fibroblast growth factor induces retinal regeneration in vivo

In the present study, we have investigated the effect of basic fibroblast growth factor (bFGF) on retinal regeneration in the stage 22-24 chick embryo. The neural retina was surgically removed in ovo leaving the retinal pigment epithelium (RPE) intact and then slow-release, plastic implants containing bFGF were inserted into the eye. Light microscopic examination of eyes 7 days later revealed that bFGF induced retinal regeneration in a dose-dependent manner. The absence of the RPE in these eyes and the reversed polarity of the regenerated neural retina is consistent with the hypothesis that this process occurs by transdifferentiation of the RPE. This represents the first time that a known molecule has been shown to induce retinal regeneration in vivo.     

Perturbation of cranial neural crest migration by the HNK-1 antibody

The HNK-1 antibody recognizes a carbohydrate moiety that is shared by a family of cell adhesion molecules and is also present on the surface of migrating neural crest cells. Here, the effects of the HNK-1 antibody on neural crest cells were examined in vitro and in vivo. When the HNK-1 antibody was added to neural tube explants in tissue culture, neural crest cells detached from laminin substrates but were unaffected on fibronectin substrates. In order to examine the effects of the HNK-1 antibody in vivo, antibody was injected lateral to the mesencephalic neural tube at the onset of cranial neural crest migration. The injected antibody persisted for approximately 16 hr on the injected side of the embryo and appeared to be most prevalent on the surface of neural crest cells. Embryos fixed within the first 24 hr after injection of HNK-1 antibodies (either whole IgMs or small IgM fragments) showed one or more of the following abnormalities: (1) ectopic neural crest cells external to the neural tube, (2) an accumulation of neural crest cell volume on the lumen of the neural tube, (3) some neural tube anomalies, or (4) a reduction in the neural crest cell volume on the injected side. The ectopic cells and neural tube anomalies persisted in embryos fixed 2 days postinjection. Only embryos having 10 or less somites at the time of injection were affected, suggesting a limited period of sensitivity to the HNK-1 antibody. Control embryos injected with a nonspecific antibody or with a nonblocking antibody against the neural cell adhesion molecule (N-CAM) were unaffected. Previous experiments from this laboratory have demonstrated than an antibody against integrin, a fibronectin and laminin receptor caused defects qualitatively similar to those resulting from HNK-1 antibody injection (M. Bronner-Fraser, J. Cell Biol., 101, 610, 1985). Coinjection of the HNK-1 and integrin antibodies resulted in a greater percentage of affected embryos than with either antibody alone. The additive nature of the effects of the two antibodies suggests that they act at different sites. These results demonstrate that the HNK-1 antibody causes abnormalities in cranial neural crest migration, perhaps by perturbing interactions between neural crest cells and laminin substrates.