CHARACTERIZATION OF RETINAL PROGENITOR CELLS

Jeff Trimarchi (Postdoctoral Fellow)

Timothy Cherry (Graduate Student)

Karolina Mizeracka (Graduate Student)

We have been addressing the question of whether extrinsic or intrinsic cues are important for retinal cell fate determination. We found that both types of information feed into the cell fate determination process. To better understand how much of the heterogeneity in the cell types produced by retinal progenitor cells might be due to intrinsic differences in the progenitor cells themselves, we have taken genomics approaches to determine the gene expression programs, and in particular, the differences, among retinal progenitor cells. Individual cells taken from the retina at different times are made into cDNA probes. The probes are hybridized to Affymetrix microarrays. Each single cell probe is compared to many other single cell probes made from retinal tissue at the same and at different times in development (Trimarchi et al. 2008; Figures 1 and 2). In addition to profiling progenitor cells, we are profiling single cells that are at various, early stages in the differentiation process. This allows us to identify how a cell changes its expression profile as it commits to a cell fate and initiates its differentiation.

Figure 1. Retinal progenitor cells are hypothesized to undergo changes in their competence to make different retinal cell fates. The changes in competence are likely produced by changes in gene expression within retinal progenitor cells. To determine these gene expression differences, single progenitor cells, as well as single cells just entering their differentiation program, are used to prepare cDNA probes. These probes are then applied to an Affymetrix microarray.

Figure 2. Expression levels of retinal progenitor genes. The signals on a microarray generated by a collection of single retinal cells are shown. The signal for each cell are shown in a column, and the genes depicted are represented as rows. The degree of signal is represented such that >10,000 is bright red, <1,000 is black and graded is 1,00-10,000. Signals from retinal ganglion cells (RGCs), amacrine cells (ACs), and photoreceptor cells (PRs) are shown for comparison. Known retinal progenitor genes expressed by the majority of progenitor cells are shown at the top, and a sample of genes expressed early or late in progenitor cells is shown in the bottom. We looked for consistent expression of genes only in early, or only in late, progenitor cells. Only sFRP2 was seen in the majority of early progenitor cells, while a cluster of genes was seen in the majority of late progenitor cells (e.g. Crym). To confirm these data, the location and frequency of cells expressing these genes was investigated by in situ hybridization on retinal tissue sections (Figure 3, 4) or by performing in situ hybridization on dissociated cells (Figure 5,6). From Trimarchi et al. 2008.

Figure 3. Retinal tissue from different ages, as indicated, were probed for the expression of genes expressed primarily early or late in retinal development. These data confirim the microarray data, e.g. sfrp2 is expressed in the majority of early progenitor cells, but few late progenitor cells. Crym has the opposite pattern. From Trimarchi et al. 2008.

For further quantification and confirmation of the microarray data, retinal cells are dissociated and plated on a slide, then hybridized with one or two probes that are detected fluorescently (Figures 4 and 5). This allows quantification of the % of cells that express a gene, or coexpress two genes, at different times in retinal development. Many pair wise analyses allow confirmation of the microarray data. In addition, the phase of the cell cycle in which a gene is expressed can be determined by this method. A retina is pulsed with 3H-thymidine for 1 hour and then either harvested immediately, or chased in cold thymidine for various time periods. By performing autoradiography in combination with in situ hybridization on dissociated cells, one can determine in which phase of the cell cycle a progenitor gene is expressed. One can also do this in conjunction with long chase periods to determine the order in which genes are expressed in postmitotic cells as they enter their differentiation programs.

Figure 4. FGF15 is expressed in cycling retinal progenitor cells. To further determine that FGF15 is expressed in cycling retinal cells, the retina was pulsed with 3H-thymidine for one hour and then the tissue was dissociated and placed on a slide. The cells were fixed and probed for FGF15 (red) and processed for autoradiography (black dots). It can be seen that approximately 50% of cells in S phase express FGF15 (ex. Red arrow), and that not all FGF15 cells are in S phase. Systematic analyses such as these allow alignment of gene expression patterns with the cell cycle. From Trimarchi et al. 2008.

Figure 5. Genes expressed in postmitotic cells. The suggestion by the microarray data and from the literature is that GAP43 would not be expressed in cycling retinal cells. This was confirmed using the method outlined in Figure 3. Red signal marks cells that express GAP43 RNA (red arrow) and the black dots mark cells in S phase or early G2 (Green arrow). From Trimarchi et al. 2008.

The expression of transcription factors was examined to see if there were consistent patterns of transcription factors that correlated temporally (Figure 6).

Figure 6.The relative signal levels for a sampling of transcription factors are shown for RPC and a few other cell types for comparison. The signal for each cell are shown in a column, and the genes depicted are represented as rows. The degree of signal is represented such that >10,000 is bright red, <1,000 is black and graded is 1,00-10,000. From Trimarchi et al. 2008.

Again, a great deal of heterogeneity was seen, though there were temporal trends. Overall, we interpret these data to mean that there is a great deal of heterogeneity among retinal progenitor cells. These differences likely reflect differences in competencies to make different retinal cell types, different locations in the retina, as well as differences in cell cycle phase, and/or mitotic potential differences. As there are at least 60 different retinal cell types, lack of reproducible patterns might be due to the fact that we did not sample enough cells. We are now sampling cells with expression of a particular gene in order to search for more reproducibility.