Reprogramming chick RPE progeny cells to differentiate towards retinal neurons by ash1

Abstract

Purpose: Harnessing a cell culture of retinal pigment epithelium (RPE) to give rise to retinal neurons may offer a source of developing neurons for cell-replacement studies. This study explores the possibility of reprogramming RPE progeny cells to differentiate toward retinal neurons with achaete-scute homolog 1 (ash1), a proneural gene that is expressed in progenitor cells in the developing retina and promotes amacrine cell production when overexpressed in the chick retina.

Methods: Replication Competent Avian Splice (RCAS) retrovirus was used to drive the ectopic expression of ash1 in cell cultures of dissociated RPE isolated from day 6 chick embryos. RCAS expressing green fluorescent protein (RCAS-GFP) was used as control. The cultures were examined for de novo generation of neuron-like cells by molecular, cellular, and physiologic criteria.

Results: In control cultures infected with RCAS-GFP, RPE cells appeared cobblestone-like and often darkly pigmented. In cultures infected with RCAS-ash1, however, cells remained de-pigmented and frequently formed clusters. Further examination at the morphological and molecular levels showed the development of elaborate processes characteristic of neurons and the expression of genes/markers that identify different types of retinal neurons. The most prevalently expressed neural marker was calretinin, which in the chick retina identifies amacrine, ganglion, and horizontal cells. As an assay for functional maturation, the reprogrammed cells were analyzed for the presence of functional, ionotropic glutamate receptors that lead to a rise in the cytosolic free calcium (Ca(2+)) concentration. Calcium imaging showed that reprogrammed cells responded to glutamate and N-methyl-D-aspartate (NMDA) by increasing their Ca(2+) concentrations, which, after reaching a peak level, returned to the basal level. The response curves of reprogrammed cells resembled those of cultured retinal neurons.

Conclusions: These results suggest that RPE progeny cells can be reprogrammed by ash1 to develop molecular, morphological, and physiologic properties that are characteristic of retinal neurons.

Figure 1

The appearance of RPE cell culture infected with RCAS-ash1 differed from that of the control. A: The control retinal pigment epithelium (RPE) cell culture infected with Replication Competent Avian Splice (RCAS)-green fluorescent protein (GFP; left) showed dark pigmentation, but the experimental culture infected with RCAS-achaete-scute homolog 1 (ash1; right) remained un-pigmented, except at a few places. B: The control culture displayed a monolayer-appearance. C: The experimental culture contained clusters or aggregates of cells. Arrows point to cell clusters. Scale bars represents 100 μm.

Figure 2

Neural markers were detected in RPE cell cultures infected with RCAS-ash1 (or Replication Competent Avian Splice (RCAS)-ash1ΔC_rb) with immunocytochemistry. A: Immunostaining for calretinin showed no positive cells in a control culture infected by RCAS-green fluorescent protein (GFP). B: A large of number of calretinin+ cells were present in a culture infected by RCAS-ash1. C is a higher magnification view of a section in B. Arrows point to clusters of calretinin⁺ cells. D: MAP2⁺ cells were present in a culture infected by RCAS-ash1ΔC_rb. E: MAP2⁺ cells in RCAS-ash1ΔC_rb-infected culture displayed neuron-like morphologies. F-H: Double-labeling for calretinin (F) and visinin (G) showed that a small number of calretinin⁺ cells were also visinin⁺. H is a merge of F and G. Arrowheads in F, G, and H each point to a visinin⁺ cell that lacked calretinin immunostaining. Scale bars represents 50 μm.

Figure 3

Neuron-like morphologies were formed by reprogrammed cells in RPE cell cultures infected with RCAS-ash1 (or RCAS-ash1ΔC_rb). A: Under bright-field, a control retinal pigment epithelium (RPE) cell culture infected with RCAS contained cells densely organized into a mono-layer. B: After fluo-4 AM labeling, these cells were invisible with epifluorescence, due to their low Ca²⁺ levels. C: A calretinin⁺ cell in an RPE cell culture infected with Replication Competent Avian Splice (RCAS)-achaete-scute homolog 1 (ash1) exhibited elaborate cellular processes, reminiscent of neural processes. D: Reprogrammed cells in a RPE cell cultures infected with RCAS-ash1 displayed neuron-like morphologies, as revealed with fluo-4 AM labeling. E: Reprogrammed cells in RCAS-ash1ΔC_rb-infected culture also displayed neuron-like morphologies, as revealed with transfection of with AAV-GFP DNA. Scale bars represents 50 μm.

Figure 4

Reprogrammed cultures lacked vimentin⁺ cells that were Müller glia-like. A, B: Immunostaining for vimentin (B) showed no cells that displayed spindle-like processes (i.e., Müller glia-like) in the control culture infected with Replication Competent Avian Splice (RCAS)-green fluorescent protein (GFP), and A is a bright-field view of the culture. C, D: Reprogrammed cultures (infected with RCAS-ash1) also lacked vimentin⁺ cells with spindle-like processes (D); C is a bright-field view of the culture. Arrows point to cells with compact cell bodies discernible under Hoffman modulation optics. E-G: A lack of vimentin⁺ cells that displayed spindle-like processes in reprogrammed culture was not due to a lack of reprogramming, because double-labeling showed that no vimentin⁺ cells spindle-like processes were detected (F) at places where a large number of calretinin⁺ cells were present (G); E is a bright-field view of the culture. Arrows point to calretinin⁺ cells. H and I are higher magnifications of F and G, respectively. Scale bars represents 50 μm.

Figure 5

Reprogrammed cells responded to NDMA+glycine. A: Montage of fluorescence micrographs is shown to illustrate responses by two reprogrammed cells (arrows) in an retinal pigment epithelium (RPE) cell culture infected with RCAS-ash1. The image in the first panel was captured with longer exposure time for a clear view of the cells. Time (in seconds) after the application of the neurotransmitter is shown at the bottom of each panel. B, C: The ΔF/Fo of each of the two cells identified in A was plotted against time (in seconds) to produce a response curve. D: The ΔF/Fo a retinal cell in an E16 chick retinal cell culture was plotted against time to produce a response curve as a reference for the reprogrammed cells. E: The ΔF/Fo of a cell in the control RPE culture infected with RCAS was plotted against time to produce a response curve as a negative control. While no response was observed with RPE cells, reprogrammed cells responded to NDMA+glycine by transiently increasing Ca²⁺ concentrations, similar to the responses by retinal cells. The arrow in B-E points to the time at which the neurotransmitter was applied. Scale bars represents 50 μm.

Figure 6

Reprogrammed cells showed responses to exogenous glutamate. A, B: The ΔF/Fo of each of two reprogrammed cells in an retinal pigment epithelium (RPE) cell culture infected with Replication Competent Avian Splice (RCAS)-achaete-scute homolog 1 (ash1) was plotted against time (in seconds) to produce response curves. C, D: The ΔF/Fo of each of two retinal cells in an E16 chick retinal cell culture was plotted against time (in seconds) to produce response curves as reference to those of reprogrammed cells. E, F: The ΔF/Fo of each of two RPE cells in the control culture infected with RCAS was plotted against time to produce response curves as negative controls. While no responses were observed with RPE cells, reprogrammed cells responded to glutamate and the response curves were similar to that of retinal cells. The arrow in each plot points to the time at which the neurotransmitter was applied.