Postmitotic Retinal Cells Research Articles

The perplexed and confused Mutations Affect Distinct Stages during the Transition from Proliferating to Post-Mitotic Cells within the Zebrafish Retina

To identify and study genes essential for vertebrate retinal development, we are screening zebrafish embryos for mutations that disrupt retinal histogenesis. Key steps in retinogenesis include withdrawal from mitosis by multipotent neuroepithelial cells, specification to particular cell types, migration to the appropriate laminar positions, and molecular and morphological differentiation. In this study, we have identified two recessive mutations that affect the transition of proliferating neuroepithelial cells to postmitotic retinal cells. Both the perplexed and confused mutant phenotypes were initially detectable when the first retinal neuroepithelial cells began to leave the cell cycle. At this time, each mutant retina showed increased cell death and a lack of morphological differentiation. Cell death was found to be apoptotic in both perplexed and confused retinas based on TUNEL analysis and activation of caspase-3. TUNEL-phosphoRb-BrdU colocalization studies indicated that the perplexed mutation caused death in cells transitioning from a proliferative to postmitotic state. For the confused mutation, TUNEL-phosphoRb-BrdU analysis revealed that only a subset of postmitotic cells were induced to activate apoptosis. Mosaic analysis demonstrated that within the retina the perplexed mutation functions noncell-autonomously. Furthermore, whole lens or eye cup transplantations indicated that the retinal defect was intrinsic to the retina. Mosaic analysis with confused embryos showed this mutation acts cell-autonomously. From these studies, we conclude that the perplexed and confused genes are essential at distinct stages during the transition from proliferating to postmitotic cells within the zebrafish retina.

Lipid-mediated gene transfection into chick embryo retinal cells in ovo and in vitro

Several lipofection reagents were tested on chick embryo retinal cultures using green fluorescent protein (GFP) as a reporter gene; best results were obtained with the GenePORTER™ (GP) reagent, which yielded approximately 4.4% of the cells with intense GFP fluorescence. Cell survival and structural differentiation appeared normal, but one of the immunocytochemical markers studied (visinin) was less frequently observed in GP-treated cultures. When similar plasmid-GP mixtures were injected into chick embryo eyes in ovo, bright GFP-fluorescent cells were observed in different retinal layers, without detectable detrimental effects on retinal morphology. Particularly extensive reporter gene expression was obtained upon intraocular injection of GP plus naked DNA from a RCAS retrovirus, which resulted in the development of abundant radial columns of alkaline phosphatase-positive cells, separated by columns of negative cells. We conclude that lipid-based transfection offers a quick, simple and fairly innocuous means for gene delivery into proliferating and postmitotic retinal cells, in vitro as well as in the developing eye in ovo, and that transfection of naked retroviral DNA can lead to extensive expression of foreign genes by retinal cells, bypassing the time-consuming steps required for the generation of high-titer virion stocks.

Selective sensitivity of early postmitotic retinal cells to apoptosis induced by inhibition of protein synthesis

In previous work we showed that apoptosis in retinal tissue from developing rats can be induced by inhibition of protein synthesis (Rehen et al. 1996, Development, 122, 1439-1448). Here we show that recent postmitotic cells are the cells sensitive to apoptosis triggered by blockade of protein synthesis. To label all proliferating cells in the retina, a series of injections of the nucleotide analogue, bromo-deoxy-uridine (BrdU, 60 mg/kg b.w.), was given in rat pups. Then, explants of the retina were incubated in vitro with the inhibitor of protein synthesis anisomycin (1.0-3.2 microg/mL) for 1 day to induce apoptosis. Detection of apoptotic bodies under differential interference contrast microscopy was combined with immunocytochemistry for BrdU, proliferating cell nuclear antigen (PCNA) or for various markers of retinal cell differentiation. Despite the large number of BrdU- and PCNA-labelled cells in the tissue, the vast majority of the cells that underwent apoptosis were postmitotic cells which have left the mitotic cycle 3-4 days before. However, these cells were not labelled with antibodies to calretinin, calbindin, rhodopsin or to a Muller glial cell marker, suggesting that these are early postmitotic neurons. We suggest that during migration and initial differentiation, the apoptotic machinery is blocked by suppressor proteins, thus allowing recent postmitotic cells to find their final positions and differentiate while protected from apoptosis.

Clonal expansion and cell dispersion in the developing mouse retina.

The present study has used two different approaches for labelling progenitor cells at the optic vesicle stage in order to examine patterns of clonal expansion and cellular dispersion within the developing retina. X-inactivation transgenic mice and chimeric mice expressing the lacZ reporter transgene were examined during development and in adulthood to study the radial and tangential dispersion of proliferating neuroepithelial cells and postmitotic retinal cells of known identities. Chimeric retinas were used to measure tangential dispersion distances, while transgenic retinas were used to assess the frequency of tangential dispersion for individual populations of retinal neurons. Tangential dispersion is shown to be a universal feature of particular retinal cell types, being contrasted with the strictly radial dispersion of other cells. Tangential dispersion is a relatively short-distance phenomenon, with distinct dispersion distances characteristic for cone, horizontal, amacrine and ganglion cells. Embryonic and postnatal retinas show that tangential dispersion occurs at different times for these distinct cell types, associated with their times of differentiation rather than their neurogenetic periods. These developmental results rule out the possibility that tangential dispersion is due to a passive displacement produced by the proliferation of later-born cells, or to the lateral dispersion of a dividing sibling; rather, they are consistent with the hypothesis that tangential dispersion plays a role in the establishment of the orderly spatial distribution of retinal mosaics.

Cell birthdays in Xenopus laevis retina

Using an in vitro differentiation system, we reevaluated the stages of development during which retinal neurons become postmitotic in Xenopus laevis. We also examined whether retinal detachment and removal of the pigment epithelium stimulated proliferation of previously postmitotic retinal cells. Retinas with and without an adherent pigment epithelium (stages 24–40 and stage 33/34, respectively) were removed from X. laevis embryos, placed in culture and allowed to differentiate in the presence of 3H-thymidine. After 2 days, eyes were fixed and processed for autoradiography. The proportion of labeled to unlabeled nuclei in the posterior pole of the retina was determined for each of the three cell layers. Early in development, most unlabeled cells were found in the ganglion cell layer; at stage 24, 53% of the cells showed no labeling, but by stage 32–33/34 all of the cells present were unlabeled. Within the outer nuclear layer, 17% of the cells failed to incorporate the 3H-thymidine at stage 24 and by stage 33/34, 100% of the cells were unlabeled. Within the inner nuclear layer, 13.5% of the cells failed to show labeling at stage 24, whereas at stage 40, none of the cells were labeled. There was no difference in the proportion of labeled to unlabeled nuclei in any of the cell layers when retinas were allowed to differentiate either with or without an adherent pigment epithelium. These results indicate that as early as stage 24, some cells that will become positioned in each of the nuclear strata fail to incorporate 3H-thymidine, suggesting that these cells become postmitotic very early in neurogenesis. In addition, removal of the pigment epithelium does not stimulate mitosis in retinal cells that have previously left the mitotic cycle.

Emergence and development of immunoreactive cells in teleostean retinas during the perinatal period

We have used light-microscopical immunohistochemistry to investigate developmental changes of several neurochemical indicators in retinas of perinatal killifish and goldfish. Immunoreactive proliferating cell nuclear antigen (ir-PCNA/cyclin, a marker for replicating cells) was present in nuclei of all neuroblasts in the early monolayer stage, but was lost progressively in central-to-peripheral and proximal-to-distal order as the layers and cells of the mature retina appeared. The loss of ir-PCNA was slightly prior to the appearance of ir-TH (tyrosine hydroxylase), GAD (glutamic acid decarboxylase) and GS (glutamine synthetase) at the 4th embryonic day (E4) in both fish. Since hatching was earlier in goldfish (E5) than in killifish (E7), neurochemical maturation was evident at 2–3 days before hatching in killifish but not until around hatching in goldfish. Two markers, ir-somatostatin and protein kinase C, were detected by the 1st postnatal day (H1) in goldfish, but not in perinatal or adult killifish retinas. Thus the course of development of killifish and goldfish retinas is similar, but not identical. The validity of ir-PCNA as a marker for proliferating cells is confirmed by the coincidence of its disappearance with the appearance of neurochemical markers for mature, postmitotic retinal cells.