The development of inducibility for glutamine synthetase in embryonic neural retina: inhibition by BrdU

The maturation of photoreceptors in the avian retina is stimulated by thyroid hormone

During retinal development, the cell-fate of photoreceptors is committed long before maturation, which entails the expression of opsins and functional transduction of light. The mechanisms that delay the maturation of photoreceptors remain unknown. We have recently reported that immature photoreceptors express the LIM domain transcription factors Islet2 and Lim3, as well as the cell-surface glycoprotein axonin1 [Fischer et al., (2008a) J Comp Neurol 506:584-603]. As the photoreceptors mature to form outer segments and express photopigments, the expression of the Islet2, Lim3 and axonin1 is diminished. The purpose of this study was to investigate whether thyroid hormone (TH) influences the maturation of photoreceptors. We studied the maturation of photoreceptors across the gradient of maturity that exists in far peripheral regions of the post-natal chicken retina [Ghai et al., (2008) Brain Res 1192:76-89]. We found that intraocular injections of TH down-regulated Islet2, Lim3 and axonin1 in photoreceptors in far peripheral regions of the retina. By contrast, TH stimulated the up-regulation of red-green opsin, violet opsin, rhodopsin and calbindin in photoreceptors. We found a correlation between the onset of RLIM (RING finger LIM-domain binding protein) and down-regulation of Islet2 and Lim3 in maturing photoreceptors; RLIM is known to interfere with the transcriptional activity of LIM-domain transcription factors. We conclude that TH stimulates the maturation of photoreceptors in the avian retina. We propose that TH inhibits the expression of Islet2 and Lim3, which thereby permits photoreceptor maturation and the onset of photopigment-expression.

Effects of taurine on glutamate uptake and degradation in Müller cells under diabetic conditions via antioxidant mechanism

Glutamate is the excitatory neurotransmitter in the retina, but it is neurotoxic in excessive amounts. A decrease in the ability of Müller cells to remove glutamate from the extracellular space may play a crucial role in the disruption of glutamate homeostasis that occurs in the diabetic retina. Previously we have shown that taurine has protective effects against diabetes-induced glutamate dysmetabolism in retinal Müller cells. The aim of this study is to examine the effects and underlying mechanism of taurine on high glucose-induced alterations of Müller cells glutamate uptake and degradation. Müller cells cultures were prepared from 5- to 7-day-old Sprague-Dawley rats. Glutamate uptake was measured as (3)H-glutamate content of the lysates. Glutamine synthetase (GS) activity was assessed by a spectrophotometric assay. The expressions of glutamate transporters (GLAST) and GS were examined by RT-PCR and western-blot. In 25 mmol/l high glucose-treated cultures, Müller cells glutamate uptake, GS activity and GLAST, GS expressions were decreased significantly compared with 5 mmol/l normal glucose cultures (p<0.05). Taurine (1 and 10 mmol/l) significantly inhibited the high glucose-induced decreases in glutamate uptake, GS activity and GLAST, GS expressions (p<0.05). The generation of TBARS, ROS and NO in Müller cells increased significantly after treatment with high glucose compared with normal glucose. However, treatment of 1 and 10 mmol/l taurine resulted in a significant decrease in TBARS, ROS and NO levels (p<0.05). The high glucose treatment decreased antioxidant enzyme (catalase, SOD and GSH-px) activities compared with normal glucose. Taurine treatment increased the catalase, SOD and GSH-px activity in a dose-dependent manner. These findings suggest that taurine may regulate Müller cells' glutamate uptake and degradation under diabetic conditions via its antioxidant mechanism.

Target-derived BDNF (brain-derived neurotrophic factor) is essential for the survival of developing neurons in the isthmo-optic nucleus

Neurons in the peripheral nervous system depend on single neurotrophic factors, whereas those in the brain are thought to utilize many different trophic factors. This study examined whether some neurons in the brain critically depend on a single trophic factor during development. Neurons in the isthmo-optic nucleus (ION) of chick embryos respond to exogenous brain-derived neurotrophic factor (BDNF). Relatively high concentrations of endogenous BDNF were present in the ION of 14-18-day-old chick embryos. ION target cells in the retina were immunolabeled for BDNF but showed surprisingly low levels of BDNF mRNA. These data suggest that ION target cells derive some BDNF from other retinal sources. No BDNF mRNA was detected in the ION itself. ION neurons had a very efficient retrograde transport system for BDNF and exogenous BDNF arrived in the ION intact. When the ION was deprived of endogenous trkB ligands by injection of trkB fusion proteins in the eye, cell death of ION neurons was enhanced, and this effect was mimicked by BDNF-specific blocking antibodies in the eye. TrkB fusion proteins in the retina induced cell death of ION neurons prior to visible effects on ION target cells in the retina. Immunolabel for endogenous BDNF was sparse in pyknotic ION neurons, suggesting that ION neurons with low BDNF content were eliminated by apoptosis. These data show that BDNF is an essential target-derived trophic factor for developing ION neurons and thereby validate the neurotrophic hypothesis for at least one neuronal population in the brain.

Purification of chick retinal ganglion cells for molecular analysis: combining retrograde labeling and immunopanning yields 100% purity

Retinal ganglion cells (RGCs) from embryonic and posthatch chickens were 100% purified by a novel combination of three steps: (1) Retrograde labeling by injection of the fluorescent carbocyanine tracer DiI into the optic nerve, (2) immunopanning of dissociated retinal cells with Thy1 antibodies, and (3) micro-aspiration of labeled RGCs into glass capillaries. The retina was dissected and dissociated with trypsin 12-15 h after the injection of DiI. DiI-labeled cells were identified on immunopanned dishes by fluorescence and collected for molecular analysis within 3 h after dissociation. This technique allowed the collection of up to 500 RGCs per capillary tube and 1500 labeled RGCs per retina. Extraction of RNA and molecular analysis by RT-PCR from 600 RGCs shows that expression of rare genes, such as those of neurotrophic factors, can be detected. This is the first description of a rapid and reliable technique for a 100% purification of RGCs with sufficient yield for molecular analysis of rare gene expression. The protocol can be modified for the purification of other cell types. The advantages and limitations of the three-step purification method are compared with previous RGC purification protocols.

A silencer element in the regulatory region of glutamine synthetase controls cell type-specific repression of gene induction by glucocorticoids

Glutamine synthetase is a key enzyme in the recycling of the neurotransmitter glutamate. Expression of this enzyme is regulated by glucocorticoids, which induce a high level of glutamine synthetase in neural but not in various non-neural tissues. This is despite the fact that non-neural cells express functional glucocorticoid receptor molecules capable of inducing other target genes. Sequencing and functional analysis of the upstream region of the glutamine synthetase gene identified, 5' to the glucocorticoid response element (GRE), a 21-base pair glutamine synthetase silencer element (GSSE), which showed considerable homology with the neural restrictive silencer element NRSE. The GSSE was able to markedly repress the induction of gene transcription by glucocorticoids in non-neural cells and in embryonic neural retina. The repressive activity of the GSSE could be conferred on a heterologous GRE promoter and was orientation- and position-independent with respect to the transcriptional start site, but appeared to depend on a location proximal to the GRE. Gel-shift assays revealed that non-neural cells and cells of early embryonic retina contain a high level of GSSE binding activity and that this level declines progressively with age. Our results suggest that the GSSE might be involved in the restriction of glutamine synthetase induction by glucocorticoids to differentiated neural tissues.

Cyclic adenosine-3',5'-monophosphate-mediated activation of a glutamine synthetase composite glucocorticoid response element

The glutamate synthetase gene (GS) contains a composite glucocorticoid response element (cGRE) comprised of a GRE and an adjacent element with features of both a cAMP-response element (CRE) and a 12-O-tetradecanoylphorbol 13-acetate (TPA) response element (TRE). The CRE/TRE element of the cGRE contributed to two modes of transcriptional activation: 1) enhancement of the response to cortisol and 2) a synergistic response to cortisol and increased cAMP. COS-7 cells transfected with a cGRE-luciferase construct show minimal expression under basal conditions or forskolin treatment. After cortisol treatment, luciferase activity from the cGRE is enhanced 4- to 8-fold greater than the GRE portion of the cGRE or a GRE from the tyrosine aminotransferase gene. Treatment with both forskolin and cortisol produced a 2- to 4-fold synergistic response over cortisol alone. Synergy is also seen with 8-bromo-cAMP, is specific for the cGRE, and occurs in a number of established cell lines. Elimination of the GRE or CRE/TRE reduces the synergy by 70-100%. Altering the CRE/TRE to GRE spacing changed both enhancement and synergy. Moving the elements 3 bp closer or extending 15 bp reduced enhancement. Synergy was markedly reduced when elements were one half of a helical turn out of phase. Western blots verified that CREB (cAMP-responsive binding protein) and ATF-1 (activating transcription factor-1) binds to the cGRE sequence. A specific dominant negative inhibitor of the CREB family, A-CREB, reduced synergy by 50%. These results suggest that the GS cGRE can potentially integrate signaling from both the cAMP and glucocorticoid receptor transduction pathways and that CREB/ATF-1 may play an important role in this process.

Regulation of the spatiotemporal pattern of expression of the glutamine synthetase gene

Glutamine synthetase, the enzyme that catalyzes the ATP-dependent conversion of glutamate and ammonia into glutamine, is expressed in a tissue-specific and developmentally controlled manner. The first part of this review focuses on its spatiotemporal pattern of expression, the factors that regulate its levels under (patho)physiological conditions, and its role in glutamine, glutamate, and ammonia metabolism in mammals. Glutamine synthetase protein stability is more than 10-fold reduced by its product glutamine and by covalent modifications. During late fetal development, translational efficiency increases more than 10-fold. Glutamine synthetase mRNA stability is negatively affected by cAMP, whereas glucocorticoids, growth hormone, insulin (all positive), and cAMP (negative) regulate its rate of transcription. The signal transduction pathways by which these factors may regulate the expression of glutamine synthetase are briefly discussed. The second part of the review focuses on the evolution, structure, and transcriptional regulation of the glutamine synthetase gene in rat and chicken. Two enhancers (at -6.5 and -2.5 kb) were identified in the upstream region and two enhancers (between +156 and +857 bp) in the first intron of the rat glutamine synthetase gene. In addition, sequence analysis suggests a regulatory role for regions in the 3' untranslated region of the gene. The immediate-upstream region of the chicken glutamine synthetase gene is responsible for its cell-specific expression, whereas the glucocorticoid-induced developmental appearance in the neural retina is governed by its far-upstream region.

Steroid hormone receptors activate transcription in glial cells of intact retina but not in primary cultures of retinal glial cells

We have compared the steroid responsiveness of Müller glial cells of intact embryonic chicken retina with that of primary cultures derived from Müller glia. Appropriately constructed fusion genes were found to be highly glucocorticoid inducible after their cotransfection with an expression vector encoding the human glucocorticoid receptor (GR) into intact embryonic d-10 (E10) or E5.5 retina. Dramatically attenuated inductions were obtained after contransfection of Müller cell primary cultures. The progesterone receptor (PR) was also demonstrated to function in intact retina, but not in Müller cell primary cultures. An immunochemical assay was utilized to confirm that a glucocorticoid-responsive, beta-galactosidase-encoding fusion gene was specifically induced in Müller cells after its transfection into intact retina. Thus, in contrast to Müller cells in intact retina, Müller cells in primary culture have lost the capacity to achieve transcriptional activation by steroid receptors. We postulate that coordinate expression of the GR, and other more general factors required for steroid inducibility, is lost by dispersion and primary culture of retinal Müller glial cells.

A transcriptional enhancer of the glutamine synthetase gene that is selective for retinal Müller glial cells

This article demonstrates that the chicken glutamine synthetase (GS) promoter contains cis-acting elements that direct transcription to retinal Müller glial cells. The transient assay system developed to identify these elements involved electroporation of intact retinal tissue with GS-beta-galactosidase fusion genes followed by preparation of primary cultures and histochemical assay of cells expressing beta-galactosidase. Plasmids containing beta-galactosidase under transcriptional control by two different viral promoters are expressed primarily in neuronal cells after transfection of intact embryonic d 12 retina. In sharp contrast, expression is primarily in Müller glia after transfection with a GS-beta-galactosidase fusion gene. Although GS is glucocorticoid inducible, steroid hormone is not required to achieve Müller cell-selective expression of the GS-beta-galactosidase fusion gene. Deletion studies indicate that multiple cis-acting elements located between nucleotides-436 and -61 relative to the GS transcription start site contribute to produce Müller cell selectivity. Moreover, these upstream elements enhance expression of a heterologous promoter in Müller cells not neurons. These results indicate that an enhancer located between 61 and 436 nucleotides upstream of the transcription start site contributes to Müller cell-selective expression of the GS gene in the retina.

Expression of the BDNF gene in the developing visual system of the chick

Using a sensitive and quantitative method, the mRNA levels of brain-derived neurotrophic factor (BDNF) were determined during the development of the chick visual system. Low copy numbers were detected, and BDNF was found to be expressed in the optic tectum already 2 days before the arrival of the first retinal ganglion cell axons, suggesting an early role of BDNF in tectal development. After the beginning of tectal innervation, BDNF mRNA levels markedly increased, and optic stalk transection at day 4 (which prevents subsequent tectal innervation) was found to reduce the contralateral tectal levels of BDNF mRNA. Comparable reductions were obtained after injection of tetrodotoxin into one eye, indicating that, already during the earliest stages of target encounter in the CNS, the degree of BDNF gene expression is influenced by activity-dependent mechanisms. BDNF mRNA was also detected in the retina itself and at levels comparable to those found in the tectum. Together with previous findings indicating that BDNF prevents the death of cultured chick retinal ganglion cells, these results support the idea that the tightly controlled expression of the BDNF gene might be important in the co-ordinated development of the visual system.

Molecular basis for differential expression of glutamine synthetase in retina glia and neurons

Glutamine synthetase (GS) is a differentiation marker of retina glial cell. It is expressed in the chicken neural retina at a particularly high level, is inducible by glucocorticoids and is always confined to Müller glia. This study investigated the molecular basis for tissue and cell-type specific expression of the GS gene. A high level of GS expression in the retina was found to coincide with the accumulation of a relatively high level of GS mRNA in this tissue. The gliatoxic agent alpha-aminoadipic acid, which can selectively destroy glia cells, was used to demonstrate that restriction of GS induction to Müller glia is controlled at a transcriptional level. Cortisol could induce accumulation of GS mRNA and transcription of the GS gene in Müller glia but not in retina neurons. Glia and neurons were also found to differ in their ability to express the glucocorticoid inducible CAT construct, p delta G46TCO, which is controlled by a 'simple GRE' promoter. When introduced into cells of retina tissue, this construct was cortisol-inducible in glia whereas in neurons it was only slightly inducible or not at all. Introduction of a glucocorticoid receptor expression vector into the cells facilitated induction of the CAT construct in neurons. Analysis by immunoblotting revealed that expression of the glucocorticoid receptor protein is predominantly restricted to Müller glia. These results suggest that differential levels of glucocorticoid receptor expression in glia and neurons might be the basis for cell-type specific induction of GS.

Protein kinase A activation of glucocorticoid-mediated signaling in the developing retina

This report establishes that increasing the activity of cyclic AMP-dependent protein kinase (protein kinase A; PKA) potentiates glucocorticoid-mediated signaling in embryonic day 5.5 (E5.5) chicken retina. Expression of a glutamine synthetase-chloramphenicol acetyltransferase (CAT) fusion gene is not induced by treatment with glucocorticoid hormone in transfected E5.5 retina. However, treatment of the retina with forskolin, an activator of adenyl cyclase, or cotransfection with an expression vector encoding PKA is sufficient to render the fusion gene hormonally responsive. Similar results are obtained after forskolin treatment of E5.5 retina that have been transfected with a plasmid that contains the CAT reporter gene under transcriptional control by the thymidine kinase promoter and a 46-nucleotide enhancer with two glucocorticoid response elements (GREs). In contrast, forskolin augments but is not required to achieve glucocorticoid-inducible CAT gene expression in E5.5 retina transfected with a plasmid that contains the reporter driven by a minimal promoter with six juxtaposed GREs. Based on these results, we postulate that E5.5 retina contain glucocorticoid receptors whose signal transduction properties are enhanced by PKA. Unlike the transiently expressed glutamine synthetase fusion gene, however, activation of PKA does not render the endogenous glutamine synthetase gene glucocorticoid-inducible. Thus, its expression appears to be subject to an additional level of control in the developing retina.

Molecular control of glutamine synthetase expression in the developing retina tissue

Glutamine synthetase is a differentiation marker of the neural retina, whose expression is restricted to Müller glia cells, is inducible by glucocorticoids and is dependent on tissue development. The retina tissue acquires the competence to express GS in response to glucocorticoids with development, although the level of hormone binding activity in the cells does not alter with age. Using CAT constructs that are controlled by "simple GRE" promoters we demonstrated that glucocorticoid receptor transcription activity in retina cells increases with development. The increase in receptor activity correlates directly with the increase in inducibility of the glutamine synthetase gene and inversely with the rate of retina cell proliferation. At early developmental ages, when retina cells are still proliferating, the glucocorticoid receptor is transcriptionally inactive and glutamine synthetase expression cannot be induced. Receptor activity increases progressively with development and by day 12, when cell proliferation ceases, competence for glutamine synthetase induction is high. This competence for glutamine synthetase induction can be repressed by overexpressing the oncogene v-src, which stimulates retina cell proliferation. We discuss possible mechanisms for developmental-dependent modulation of glucocorticoid receptor transcriptional activity.

Involvement of a C/EBP-like protein in the acquisition of responsiveness to glucocorticoid hormones during chick neural retina development

The glucocorticoid receptor in chicken embryonic neural retina is expressed early in ontogeny, yet the tissue's response to the glucocorticoid hormone, i.e., induction of glutamine synthetase (GS), develops later, only during week 2 of ontogeny. Transient transfection of embryonic day 7 (E7) retinal cells, which are nonresponsive to glucocorticoids, with chimeric plasmids containing the chloramphenicol acetyltransferase reporter gene under the control of glucocorticoid-responsive promoters demonstrated that GR in E7 cells is a functional transactivating factor. We show that the limiting transcription factor that controls the developmental acquisition of responsiveness to glucocorticoids is similar to a CCAAT enhancer-binding protein (C/EBP). This protein recognizes a sequence in the promoter of the chick GS gene, which is required for eliciting the glucocorticoid response. Retinal C/EBP-like protein was not detected in the glucocorticoid-nonresponsive (E7) proliferating glioblasts but was found to be present in the glucocorticoid-responsive (E12) postmitotic cells. Premature expression of C/EBP in the nonresponsive E7 cells by transfection was shown to enhance the developmental acquisition of responsiveness to the glucocorticoid hormone, as deduced from the level of GS inducibility.

Involvement of a C/EBP-like protein in the acquisition of responsiveness to glucocorticoid hormones during chick neural retina development

The glucocorticoid receptor in chicken embryonic neural retina is expressed early in ontogeny, yet the tissue's response to the glucocorticoid hormone, i.e., induction of glutamine synthetase (GS), develops later, only during week 2 of ontogeny. Transient transfection of embryonic day 7 (E7) retinal cells, which are nonresponsive to glucocorticoids, with chimeric plasmids containing the chloramphenicol acetyltransferase reporter gene under the control of glucocorticoid-responsive promoters demonstrated that GR in E7 cells is a functional transactivating factor. We show that the limiting transcription factor that controls the developmental acquisition of responsiveness to glucocorticoids is similar to a CCAAT enhancer-binding protein (C/EBP). This protein recognizes a sequence in the promoter of the chick GS gene, which is required for eliciting the glucocorticoid response. Retinal C/EBP-like protein was not detected in the glucocorticoid-nonresponsive (E7) proliferating glioblasts but was found to be present in the glucocorticoid-responsive (E12) postmitotic cells. Premature expression of C/EBP in the nonresponsive E7 cells by transfection was shown to enhance the developmental acquisition of responsiveness to the glucocorticoid hormone, as deduced from the level of GS inducibility.

Expression of v-src in embryonic neural retina alters cell adhesion, inhibits histogenesis, and prevents induction of glutamine synthetase

Using Rous sarcoma virus as the vector, v-src or c-src genes were introduced into 6-day chicken embryo retina tissue in organ culture and their effects on retina development were investigated. Overexpression of c-src in many of the cells had no noticeable effect on retina development. In contrast, infection with v-src resulted in abnormal histogenesis and inhibition of differentiation. Although only a portion of the cells in infected tissue expressed the oncogene and displayed the transformation phenotype, the other cells were also hindered from becoming normally positioned and organized. Therefore, presence of oncogene-transformed cells within the tissue hindered organization and development of adjacent nontransformed cells. Failure of normal cell relationships impeded induction by cortisol of glutamine synthetase in Muller glia, which requires contact associations of the glia cells with neurons. The transformed cells tended to assemble into chaotic clusters, suggesting that their adhesiveness and contact affinities had become altered. This was confirmed by aggregation experiments with dissociated cells which showed that adhesiveness of transformed cells was greatly reduced and that they had lost the ability to cohere with nontransformed cells. In binary mixtures of transformed and nontransformed cells, the two sorted out into separate aggregates. Transformed cells formed loose clusters devoid of tissue architecture; aggregates of nontransformed cells became organized into retinotypic structures, and glutamine synthetase was inducible. Our findings suggest that the mechanisms of cell adhesion and cell affinities are a key target of v-src activity in infected cells and that modification of the cell surface may be a leading factor in other cellular changes characteristic of the v-src transformation phenotype.

Expression of v-src in embryonic neural retina alters cell adhesion, inhibits histogenesis, and prevents induction of glutamine synthetase

Using Rous sarcoma virus as the vector, v-src or c-src genes were introduced into 6-day chicken embryo retina tissue in organ culture and their effects on retina development were investigated. Overexpression of c-src in many of the cells had no noticeable effect on retina development. In contrast, infection with v-src resulted in abnormal histogenesis and inhibition of differentiation. Although only a portion of the cells in infected tissue expressed the oncogene and displayed the transformation phenotype, the other cells were also hindered from becoming normally positioned and organized. Therefore, presence of oncogene-transformed cells within the tissue hindered organization and development of adjacent nontransformed cells. Failure of normal cell relationships impeded induction by cortisol of glutamine synthetase in Muller glia, which requires contact associations of the glia cells with neurons. The transformed cells tended to assemble into chaotic clusters, suggesting that their adhesiveness and contact affinities had become altered. This was confirmed by aggregation experiments with dissociated cells which showed that adhesiveness of transformed cells was greatly reduced and that they had lost the ability to cohere with nontransformed cells. In binary mixtures of transformed and nontransformed cells, the two sorted out into separate aggregates. Transformed cells formed loose clusters devoid of tissue architecture; aggregates of nontransformed cells became organized into retinotypic structures, and glutamine synthetase was inducible. Our findings suggest that the mechanisms of cell adhesion and cell affinities are a key target of v-src activity in infected cells and that modification of the cell surface may be a leading factor in other cellular changes characteristic of the v-src transformation phenotype.

Modulation of retinal differentiation by oncogenes: effect of the v-src gene on expression of choline acetyltransferase and glutamine synthetase

Expression of the protooncogene c-src in chick neural retina is developmentally regulated and associated with neural differentiation. In the present study, chick neural retina (NR) cell cultures from 7 day embryos were exposed to the exogenous src oncogene, the c-src counterpart, to establish the effect of expression of v-src on specific retinal cellular differentiation. NR cells from 7 day chick embryos were placed in monolayer or rotation culture and infected with Rous sarcoma virus (RSV) containing a single transforming gene. Other cultures were infected with a transforming defective mutant of RSV which still possesses mitogenic activity for NR cells. While control cultures showed typical neuronal and Muller cell morphologies at the light and electron microscopic level, NR cells infected with RSV exhibited dramatic morphological alterations in monolayer culture and cell aggregates. However, the mutant src gene induced mitosis without accompanying transforming properties. When aggregate cultures were treated with hydrocortisone to induce glutamine synthetase (GS) expression in Muller cells, control cultures showed the typical immunofluorescence pattern of GS staining, while RSV infected cultures showed no GS fluorescence. Cultures infected with mutant RSV showed some staining for GS. In contrast, choline acetyltransferase activity was shown to increase in both monolayer and aggregate cultures of retinal cells following v-src expression. These data indicate that the presence of excess v-src in differentiating cultures of NR inhibits the expression of some neural specific enzymes and enhances the presence of other specific proteins. Moreover, continually growing cultures of oncogene-altered retinal cells may be useful as models to study gene expression in development of the nervous system.

Pathogenesis of bromodeoxyuridine-induced cleft palate in hamster

In the present study, the morphological, histochemical, biochemical, and cellular aspects of the pathogenesis of bromodeoxyuridine (BrdU)-induced cleft palate in hamster fetuses were analyzed. Morphological observations indicated that BrdU interferes with the growth of the vertical shelves and thus induces cleft palate. At an ultrastructural level, BrdU-induced changes were first seen in the mesenchymal cells. Eighteen hours after drug administration, the initial alterations were characterized by swelling of the nuclear membrane and the appearance of lysosomes in the mesenchymal cells of the roof of the oronasal cavity. During the next 6 hr, as the palatal primordia developed, lysosomes were also seen in the overlying epithelial cells. The appearance of lysosomal activity, which was verified by acid phosphatase histochemistry, was temporally abnormal and was interpreted as a sublethal response to BrdU treatment. Later the cellular alterations subsided; 48 hr after BrdU treatment, they were absent in both the epithelial and mesenchymal cells of the vertically developing palatal shelves. Subsequently, unlike controls (in which the palatal shelves undergo reorientation and fusion), the BrdU-treated shelves remained vertical until term. Biochemical determination of DNA synthesis indicated that although there was an inhibition of DNA synthesis at the time of appearance of palatal primordia, a catch-up growth during the ensuing 12 hr may have restored the number of cells available for the formation of a vertical palatal shelf. It was suggested that BrdU affected cytodifferentiation in the palatal tissues during the critical phase of early vertical development to induce a cleft palate.

Glucocorticoid-inducible expression of a glutamine synthetase-CAT-encoding fusion plasmid after transfection of intact chicken retinal explant cultures

In this communication we demonstrate that gene transfer methodology can be applied to study gene expression in intact retinal explant cultures. The appropriate enzyme activity is observed in extracts obtained after electroporation of embryonic day-10 chicken retina with plasmids containing the chloramphenicol acetyltransferase-encoding or beta-galactosidase-encoding reporter genes under transcriptional control by the Rous sarcoma virus long terminal repeat. Similar results are obtained using Ca.phosphate-mediated gene transfer. Moreover, it has been previously established that glucocorticoid hormones stimulate transcription of glutamine synthetase (Glns) mRNA in embryonic retina. We report here that, based on the results of gene transfer experiments with chimeric plasmids containing 5'-flanking DNA derived from the cloned chicken Glns-encoding gene (Glns), essential glucocorticoid response elements reside between approx. 1.3 kb and 2.5 kb upstream from the Glns transcription start point. These data show that transfection of explant cultures can provide a useful approach to the study of gene expression in complex systems.

Embryonic cell recognition: uncoupling tissue-specific affinities from cell-type specificities

We have experimentally defined the two major aspects of embryonic cell recognition-adhesion (ReAd), tissue type-specific ReAd and cell type-specific ReAd; we showed that they arise consecutively during cell differentiation, and that the former can function in the absence of the latter. Embryonic chick cells (retina and chondroblasts) in which differentiation was arrested by BrdU at an early stage, failed to express cell-type ReAd, yet they continued to display tissue-type ReAd: they distinguished tissue-self from non-self and selectively cohered with self. Our results indicate that tissue-type and cell-type ReAd represent distinct, separately controlled mechanisms. BrdU appears to be useful as a probe for investigating the regulation of these mechanisms, and as an experimental effector of differentiation abnormalities associated with defects in cell recognition.

Developmentally regulated primary glucocorticoid hormone induction of chick retinal glutamine synthetase mRNA

We have characterized the glucocorticoid hormone induction of glutamine synthetase mRNA in embryonic chick retinal organ cultures by quantitative dot hybridization using a cDNA clone derived from chick retinal RNA. Hydrocortisone (Kapp = 3-4 nM) and dexamethasone (Kapp = 1-2 nM) produce an approximate 30-fold increase in glutamine synthetase mRNA after incubation of organ cultures derived from embryonic day 12 retinae with either hormone for 3 hr. Progesterone is a poor inducer. The glucocorticoid-mediated rise is rapid (t1/2 = 2-3 hr) and occurs in the presence of either of the protein synthesis inhibitors cycloheximide or puromycin, indicating that the induction is a primary or direct response to the hormone. However, the magnitude of the hormonal response observed in culture increases markedly during retinal development. These observations, coupled with the previously reported absence of a hormonal induction in embryonic liver, raise the possibility of a synergistic mechanism, involving tissue-specific regulatory molecules in addition to the glucocorticoid hormone receptor, to explain the retinal-specific primary glucocorticoid hormone induction of glutamine synthetase mRNA.

Epigenesis in developing avian scales. III. Stage-specific alterations of the developmental program caused by 5-bromodeoxyuridine

As an approach to the study of a developmental program, 5-bromodeoxyuridine (BrdU) was administered to chick embryos in ovo at various stages of avian scale formation. This brought about stage-specific alterations in morphogenesis in the anterior tarsometatarsus such as feathered scales, from Day 6 through Day 6 1/2; feathers only, from Day 6 3/4 through Day 7 1/4; scalelessness and rudimentary scales, from Day 7 7/8 through Day 8 1/8; and partial ridge scales, from Day 8 1/8 through Day 10. The effects of BrdU were completely nullified by an excess dose of thymidine which instantly suppressed BrdU incorporation into nuclear DNA. Effects of BrdU causing scalelessness were further examined. The percentage of BrdU labeled cells was immunohistochemically detected. It increased linearly in both the epidermis and dermis, reaching nearly 100% 24 hr following its injection on Day 8. However, scale forming potency, as assayed by the area of scale epidermis on Day 11, decreased with the duration of BrdU incorporation into the cells and disproportionately dropped at 15 hr when about 50% of the cells had incorporated BrdU. Scalelessness was also produced when the period of the incorporation of BrdU exceeded 15 hr. Time sequence observations demonstrated epidermal cell shape, polarity, alignment, and packing density to be remarkably disordered so that the placode and interplacode failed to develop on Day 9 1/4. Epidermal-dermal recombinations were carried out by exchanging normal tissues with those treated with BrdU in the anterior tarsometatarsus. The results clearly showed defects in the dermis at the time of reassociation, giving rise to scalelessness.

Immunohistochemical characterization of delta crystallin-containing retina/optic nerve "boundary" cells in the chick embryo

The term "transdifferentiation" has been used to describe the apparent phenotypic conversion of chick embryo neural retina Müller glial cells into lens-like cells in vitro. This phenotypic conversion is characterized by expression of such lens-specific proteins as delta crystallin and has been viewed as an example of cells transforming from the phenotype of a given tissue to that of another. We have identified a population of neuroglia-like cells in the embryonic chick retina which express high levels of delta crystallin as a function of normal development. The position and morphology of these cells is quite distinctive in that they form a loose meshwork which defines the boundary between the neural retina and the optic nerve head. These "boundary" cells are detectable as early as Day 5 of development through hatching. However, the meshwork structure formed by the cells is only readily observed between Days 8 and 9 of development. Double-immunolabeling procedures comparing delta crystallin staining to that of glial and neuronal markers suggest that these cells are a form of retinal Müller glial cell. The results show that under appropriate microenvironmental conditions, expression of delta crystallin falls into the normal repertoire of retinoblast cells. The results also demonstrate the presence of a cellular boundary defining the junction between the neural retina and the optic nerve, tissues that are ontogenetically and structurally continuous but functionally distinct.

Tissue-specific regulation of avian glutamine synthetase expression during development and in response to glucocorticoid hormones

We have isolated a glutamine synthetase cDNA clone derived from chicken retinal RNA. The clone detects a 3.2-kilobase RNA in chicken retina, liver, and brain, based on Northern blotting analysis. The dramatic developmental rise observed for the retinal enzyme, assayed as glutamyl transferase activity, is accompanied by a corresponding rise in this RNA. Injection of hydrocortisone 21-phosphate into the yolk sac of day 10 embryos produces an increase in retinal glutamine synthetase mRNA and glutamyl transferase activity, assayed 4 days after injection. An increase in glutamine synthetase mRNA is also observed within 2 h of incubation of retinal organ cultures with hydrocortisone. Moreover, incubation of these cultures with cycloheximide at a concentration that inhibits protein synthesis by 93% affects neither the basal level nor the hydrocortisone-mediated induction of glutamine synthetase mRNA. Although expression of this RNA is developmentally regulated in the brain, steroid hormone injection does not result in a substantial induction. Hepatic glutamine synthetase mRNA is expressed constitutively between embryonic day 10 and 6 days after hatching and is also not hormone inducible. Southern blotting data with chicken DNA digested with EcoRI, HindIII, and BamHI are best interpreted in terms of the cDNA clone detecting only one gene. If so, several cell-type-specific regulatory mechanisms must function to modulate expression of this gene during development.

Tissue-specific regulation of avian glutamine synthetase expression during development and in response to glucocorticoid hormones

We have isolated a glutamine synthetase cDNA clone derived from chicken retinal RNA. The clone detects a 3.2-kilobase RNA in chicken retina, liver, and brain, based on Northern blotting analysis. The dramatic developmental rise observed for the retinal enzyme, assayed as glutamyl transferase activity, is accompanied by a corresponding rise in this RNA. Injection of hydrocortisone 21-phosphate into the yolk sac of day 10 embryos produces an increase in retinal glutamine synthetase mRNA and glutamyl transferase activity, assayed 4 days after injection. An increase in glutamine synthetase mRNA is also observed within 2 h of incubation of retinal organ cultures with hydrocortisone. Moreover, incubation of these cultures with cycloheximide at a concentration that inhibits protein synthesis by 93% affects neither the basal level nor the hydrocortisone-mediated induction of glutamine synthetase mRNA. Although expression of this RNA is developmentally regulated in the brain, steroid hormone injection does not result in a substantial induction. Hepatic glutamine synthetase mRNA is expressed constitutively between embryonic day 10 and 6 days after hatching and is also not hormone inducible. Southern blotting data with chicken DNA digested with EcoRI, HindIII, and BamHI are best interpreted in terms of the cDNA clone detecting only one gene. If so, several cell-type-specific regulatory mechanisms must function to modulate expression of this gene during development.

Developmental regulation of glutamine synthetase and carbonic anhydrase II in neural retina

Glutamine synthetase (GS) is expressed in the neural retina only in Muller glia cells and is inducible with cortisol. A chicken genomic clone that contains at least part of the coding region for the GS enzyme was used to investigate developmental changes in the level of GS mRNA in embryonic chicken retina. A major GS transcript (approximately equal to 3 kilobases) detected by the probe begins to accumulate sharply on day 15 of embryonic development. When cortisol is prematurely supplied to early embryonic retina, it induces precocious accumulation of GS mRNA and of the GS enzyme. At later ages, these effects of cortisol are significantly greater, which suggests that competence to transcribe or stabilize GS mRNA in response to stimulation with cortisol increases with development. Carbonic anhydrase II (CA-II) is expressed in early retina in all the cells, but it becomes later restricted to Muller glia. Using cloned CA-II cDNA, we detected a high level of CA-II mRNA in early retina, followed by a decline due to arrest of CA-II mRNA accumulation in differentiated neurons. As glia cells mature, CA-II mRNA and the enzyme increase to a new high level. Therefore, changes in CA-II gene expression during retina development reflect differentiation-dependent cell-type-specific control of CA-II mRNA accumulation.

Formation of lentoids from retina gliocytes: ultrastructural study

In primary monolayer cultures of dispersed neural retina cells from 13-day chick embryo, gliocytes (Müller glia cells) multiply and rapidly change into a lentoidal (lens-like) phenotype. They express lens proteins, including MP26 (a lens plasma-membrane antigen) and ultra-structurally appear to resemble lens cells. A significant aspect of this modification is that the glia-derived lentoidal cells no longer display contact-affinity for neurons but become preferentially adhesive to each other; in aggregates, they assemble into compact lentoids. A likely explanation for this change in cell affinities is that the modified gliocytes express little or no R-cognin, a retinal cell-surface antigen implicated in mutual recognition and adhesion of retina cells. Although lentoidal cells express MP26, a gap-junction component in the lens, no gap junctions could be found in the lentoids.

Cortisol receptors and inducibility of glutamine synthetase in embryonic retina

Glutamine synthetase (GS) is a marker enzyme for Müller glia cells in neural retina. In chick embryo retina GS begins to increase sharply on the 16th day of development, but can be precociously induced by premature supply of the inducer, cortisol, already on the 8th day. At this stage GS inducibility is low, but it increases progressively with embryonic age. We investigated whether there was a corresponding age-dependent increase of cortisol-binding molecules (cortisol receptors) and found that their level is highest in the early retina and decreases with development. In light of this inverse relationship, we examined whether functional characteristics of these receptors change with age, but detected no differences. In in vitro tests, receptors from older retina translocated cortisol into nuclei from young retina, and vice versa, with similar effectiveness. Also, cortisol receptors from liver cells (which differ from retina receptors) can translocate the hormone into retina nuclei, and vice versa. These findings indicate that translocation of cortisol receptors is neither tissue-specific or age-dependent, nor is it conditional on the total amount of receptors normally present in cells. Therefore, the age-dependent increase of GS inducibility in embryonic retina cannot be directly related to quantitative or functional differences of cortisol receptors and is evidently controlled primarily at the gene level. The very large amount of cortisol-binding molecules in early embryonic retina raises the possibility that they play some role in early differentiation of retina cells unrelated to hormone binding.

Cell disorganization and malformation in neural retina caused by antibodies to R-cognin: ultrastructural study

Retina tissue from 6-day chick embryos was organ-cultured for 3 days in the presence of antibodies to R-cognin, a surface antigen of retina cells. The antibodies which are known to bind to this antigen caused a striking malformation: interruption of the outer limiting membrane and extensive cell disorganization resulting in exteriorization of many cells and forming of chaotic masses on the surface of the tissue. Controls did not show these effects. These results further confirm that R-cognin is involved in the mechanism of histotypic contacts and recognition of retina cells, and that it plays an essential role in cell organization and histogenesis in the retina.

Biosynthesis of gangliosides in cultured retina from chick embryos

Retina tissue from 7-day chick embryos was maintained in culture for up to 10 days. After 5 days in culture the incorporation of [3H]leucine into proteins and of [3H]glucosamine into gangliosides was similar to that found in retinas from 12-day embryos. The incorporation of [3H]thymidine into DNA decreased steadily with time in culture; after 5 days it was about 20% of the initial value and approximately twice that determined in retinas from 12-day embryos. The radioactivity pattern of gangliosides labeled with [3H]glucosamine showed a predominance of the label in disialosyllactosylceramide (GD3); up to the 3rd day of culture. From then on, there was a progressive increase in the labeling of disialosylgangliotetraosylceramide (GD1a); by day 7 of culture, labeling of GD1a predominated and the labeling pattern was indistinguishable from that found in retinas from 12-day-old embryos. The specific activities of the CMP-NeuAc:GM3 sialosyl- and UDP-GalNAc:GM3 N-acetylgalactosaminyl-transferases decreased to 15% and increased to 400%, respectively, of the values determined in the retinas of 7-day embryos. The cultured retinas progressed in their organization into layers with culture time. The labeling transition from GD3 to GD1a was also detected after inhibition of the histotypic organization by addition of 5-bromo 2-deoxyuridine to the culture medium. Results suggest that high activity of GM3:sialosyl transferase and high labeling of GD3 are associated with the proliferative state of retina cells, while high activity of GM3:N-acetylgalactosaminyltransferase and high labeling of GD1a are associated with the non-proliferative, differentiated state of these cells.

Transformation of retinal glia cells into lens phenotype: expression of MP26, a lens plasma membrane antigen

We describe experiments in which dissociated cells from differentiated, post-mitotic neural retina of late chicken embryos (13 and 16 days) rapidly and consistently transform (transdifferentiate) in vitro into lens-like phenotype and form spherical lentoids. Using immunohistochemical and other tests, we have established that the lentoids arise from the progeny of definitive retinal glia cells (Müller cells). An early event in their transformation is the appearance in the cell surface of MP26, a plasma membrane protein characteristic of lens but not found in the retina. The results support the hypothesis that the phenotype of definitive glia cells in the retina is stabilized by contact-mediated interactions with neurons; disruption of cell contacts and cell separation alter surface properties of the glia cells, decontrol their phenotype, and predispose them to phenotype transformation.

Hormonal induction of glutamine synthetase in cultures of embryonic retina cells: requirement for neuron-glia contact interactions

Cortisol induces glutamine synthetase (GS) in gliocytes of chick embryo neural retina. Using adherent cultures of retina cells we have demonstrated that responsiveness of the gliocytes to GS induction by the hormone requires contact with neurons. GS is not inducible in high-density cultures depleted of neurons and consisting only of gliocytes. In neuron-containing cultures, induced GS was detected immunohistochemically only in those gliocytes that were closely juxtaposed with clusters of neurons. Unlike the induction of GS, the expression of carbonic anhydrase-C (which does not require cortisol) persisted in these glia cells also in the absence of neurons. The nature and role of glia-neuron interactions in the hormonal induction of GS are briefly discussed.

Antimitotic drugs that enhance neuronal survival in olfactory bulb cell cultures

Cell cultures of olfactory bulb from 2-3-day-old rats were used to evaluate 6 antimitotic drugs for their effects on neuronal survival. In cultures grown in medium without drugs for 7 days, neurons died as a layer of non-neuronal cells proliferated. A series of experiments were performed to determine dilutions and exposure times that allow maximum neuronal survival for each of the antimitotic drugs. Use of bromodeoxyuridine (BUdR) at 10(-5) M with a 5 day exposure gave cultures with large aggregates of neurons and an extensive network of interconnecting neurites. Cytosine arabinoside (Ara-C) at 8 X 10(-6) M for 5 days of exposure was also effective but the aggregates contained cellular debris. Fluorodeoxyuridine (FUdR) at 10(-4) for 5 days was less effective in flowing neurons to survive. Methotrexate (MTX), high thymidine and hydroxyurea were all not effective in allowing neuronal survival. In all cases there was an inverse relationship between the survival of neurons and proliferation of non-neuronal cells. Additional experiments were performed to determine how proliferating non-neuronal cells can lead to the death of neurons. In a series of transfer experiments, cultures were exposed to conditioned medium. Cultured olfactory bulb neurons grown on small cover slips were exposed to BUdR under optimal conditions and after 7 days were transferred to cultures treated or not treated with BUdR. In cultures not treated with BUdR, most of the transferred BUdR treated neurons died, while in cultures treated with BUdR the BUdR treated neurons survived. These results suggest that antimitotics enhance olfactory bulb neuronal survival by reducing the number of non-neuronal cells. In addition, it appears that proliferating non-neuronal cells are responsible for neuronal cell death by a medium factor and not by contact with the dividing non-neuronal cells.

Glycoconjugates of the avian eye: the development and maturation of the neural retina as visualized by lectin binding

The distribution of binding sites for four rhodamine-conjugated lectins has been examined in paraffin-embedded tissue sections of the developing and mature avian neural retina. The method of fixation employed here (95% ethanol:glacial acetic acid) is shown to conserve substantially more in vivo labeled 3H-glycoprotein or 3H-ganglioside than fixation with either 3.7% formaldehyde or 2% glutaraldehyde. Each lectin was shown to be optimally active by direct hemagglutination and inhibitable by the appropriate hapten in the cytochemical assay. Staining with wheat germ agglutinin (leads to N-acetylglucosamine) occurs predominantly to each of the anuclear retinal layers and progressively disappears from each of the nuclear layers. By adulthood, the inner nuclear layer is stained in a gradient which is highest adjacent to the inner nuclear layer while the nuclear layers are virtually unstained. Although similar to wheat germ agglutinin, staining with concanavalin A (leads to mannose) is less intense and persists in the adult nuclear layers. Dolichos biflorus agglutinin (leads to N-acetylgalactosamine) predominantly stains the embryonic ganglion cell layer. With development, this lectin demonstrates profound differences between the major cell types of the inner nuclear layer and the inner (negative) and outer (positive) plexiform layers. Differences between the principal synaptic layers are further accentuated by Ulex europaeus agglutinin I (leads to fucose): Binding sites are virtually absent from the 12th day embryonic retina, but are abundant throughout the retina by the 18th day and persist after hatching. However, in the adult retina staining persists in the outer plexiform layer but is lost from the inner plexiform and nerve fiber layers. In contrast to Dolichos biflorus agglutinin, Ulex europaeus agglutinin I stains the inner nuclear layer uniformly. Temporal differences in the characteristic lectin binding patterns exhibited by each of the major nuclear and anuclear layers of the developing avian neural retina are thus documented.

Changes in patterns and synthesis of proteins in embryonic neural retina studied by 2-dimensional gel electrophoresis

Changes in protein patterns during early differentiation of embryonic neural retina (chick) were studied by 2-dimensional gel electrophoresis. The procedures employed here made it possible to visualize the overall population of proteins present in the tissue at a given time and, on the same gel to distinguish labeled from unlabeled proteins. 2-Dimensional gels were stained by a highly sensitive silver stain to visualize, map and quantitate proteins (and polypeptides) resolved by electrophoresis; the same gels were then autoradiographed in order to differentiate between actively synthesized and pre-existing proteins at each development stage. The effectiveness of this combinative analysis was first verified by identifying and localizing glutamine synthetase, an inducible enzyme marker of retina differentiation. Next, protein patterns in retina tissue at 2 embryonic ages were compared. Of the large number of spots visualized by the above methods approximately 10% showed distinct qualitative-quantitative developmental changes; these were grouped into 7 classes representative of major modes of alteration of protein patterns during cell differentiation.

Development of differential affinities and positional information in embryonic retina cells: inhibition by BrdU

Embryonic neural retina cells fail to develop surface properties for type-specific cell recognition if they are transiently exposed to BrdU during an early, critical age. Such cells do not proceed with histogenetic positioning, organization and differentiation and, instead, form a malformed cell mass. This effect of BrdU is correlated with BrdU incorporation into DNA, and it can be prevented by simultaneous treatment of the cells with cytosine arabinoside. The proposed working hypothesis is that, in this system, BrdU interferes with expression of genes controlling cognitive specification of the cell surface.