Skip to Main Content

Biomedical Research

The retinal pigment epithelium (RPE) plays a central role in retinal physiology by forming the outer blood-retinal barrier and supporting the function of the photoreceptors. Many retinopathies involve a disruption of the epithelium's interactions with the neural retina or its uncontrolled proliferation. Surgical interventions limit the progression of disease, but fail to restore function. Although encouraging progress has been made with RPE transplantation, its effectiveness is limited to the earliest stages of disease when patients would be reluctant to have surgery. Our goal is to expand that window of opportunity by understanding the interactions of the RPE with its neighbors, the choroid and the neural portion of the retina. Our early studies with chick RPE demonstrated that:

  1. As the neural retina matures, it secretes factors that induce the RPE to form the outer blood-retinal barrier by decreasing the permeability of RPE junctions.
  2. At the RPE/neural retina interface, extracellular matrix or cell-cell interactions regulate the distribution of certain integrins. These integrins are redistributed when the neural retina and its extracellular matrix mature.
  3. Initially, diffusible factors produced by the neural retina maintain the apical polarity of the Na,K-ATPase. These retinal factors differ from those that decrease the permeability of the monolayer, and may act indirectly through effects on the structure of the apical microvilli.
  4. Gene array studies demonstrate that 40% of the RPE transcriptome changes in parallel with retinal development. Retinal secretions regulate many of these changes.

Our current research asks whether the chick studies are relevant to human biology. Surprisingly, we found that the tight junctions of human RPE differ significantly from those of non-primate vertebrates -- surprising because tight junctions serve a conserved function. Tight junctions are an integral part of any blood-tissue barrier, because they regulate diffusion across the paracellular spaces of an epithelial monolayer.

Tight junctions form a network of anastomosing strands that encircles each cell and binds it to its neighbors in the monolayer. They regulate the permeability and selectivity of the paracellular path and are matched with the ion channels and transporters that regulate the transcellular movement of solutes. Claudins are a family of at least 24 proteins that determine the properties of tight junctions and each epithelium expresses a subset that reflects the physiology of the organ. Human RPE expresses a different set of claudins that non-primate vertebrates which implies differences in retinal physiology.

In both cultures of human fetal RPE (hfRPE) and human embryonic stem cell (hESC)-derived RPE, we found that we could make the barrier function more in vivo-like by using a serum free medium that we call SFM-1. Studies of the transcriptome demonstrate that SFM-1 affects many genes and that adaptation of the cultures to SFM-1 furthers the maturation of hESC-RPE. This was demonstrated in two cells, H1 and H9. Examining function and the transcriptome leads us to believe that even after SFM-1, hESC-RPE remains less mature than hfRPE isolated from 16 week-gestation fetal eyes. The study identifies 25 marker genes that can be used to monitor the maturation of RPE that we predict will occur when hESC-RPE is co-cultured with hESC-derived retinal precursors (RPC).

To test this hypothesis, RPE and RPC need to be culture in the same media. It appears that SFM-1 furthers the maturation of RPC so that co-culture is feasible. We have found that a scaffold of gelatin decorated with glycoproteins provides a way to culture RPC as a flat sheet that can be layered atop a sheet of RPE. Preliminary data indicate that gene expression is altered in both the RPE and RPC layers following co-culture.

I believe this co-culture model will provide a superior platform to explore the efficacy of drugs that might treat retinal degenerations or improve the efficacy of transplantation. Further, the model itself may prove to be a suitable tissue for transplantation into patients with advanced retinal degeneration.