Avoiding the Pitfalls of siRNA Delivery to the Retinal Pigment Epithelium with Physiologically Relevant Cell Models

Inflammation is involved in the pathogenesis of several age-related ocular diseases, such as macular degeneration (AMD), diabetic retinopathy, and glaucoma. The delivery of anti-inflammatory siRNA to the retinal pigment epithelium (RPE) may become a promising therapeutic option for the treatment of inflammation, if the efficient delivery of siRNA to target cells is accomplished. Unfortunately, so far, the siRNA delivery system selection performed in dividing RPE cells in vitro has been a poor predictor of the in vivo efficacy. Our study evaluates the silencing efficiency of polyplexes, lipoplexes, and lipidoid-siRNA complexes in dividing RPE cells as well as in physiologically relevant RPE cell models. We find that RPE cell differentiation alters their endocytic activity and causes a decrease in the uptake of siRNA complexes. In addition, we determine that melanosomal sequestration is another significant and previously unexplored barrier to gene silencing in pigmented cells. In summary, this study highlights the importance of choosing a physiologically relevant RPE cell model for the selection of siRNA delivery systems. Such cell models are expected to enable the identification of carriers with a high probability of success in vivo, and thus propel the development of siRNA therapeutics for ocular disease.