RPE-Immune System Interactions in and around the SRS

The integrated effect of proinflammatory molecules on RPE function depends on the polarized location of the cognate receptors and the access of their ligands (cytokines and chemokines) to the apical and basolateral membranes, and the interactions of downstream signaling pathways. For the experiments summarized in Table 1, we used a mixture of three proinflammatory cytokines, interleukin 1 beta (IL-1p), interferon gamma (IFNg), and tumor necrosis factor-alpha (TNF-a) to stimulate confluent monolayers of hfRPE. These proinflammatory cytokines are elevated in patients with uveitis and are detected in the vitreous and blood of patients with proliferative diabetic retinopathy (PDR) and AMD with choroidal neovascularization (CNV).

As a first step in understanding how the RPE in vivo can actively control the inflammatory environment in the SRS and choroid, Shi and colleagues used confluent monolayers of human fetal RPE primary cultures to (1) measure the constitutive and polarized secretion of angio-genic/angiostatic cytokines by the RPE; (2) determine how this pattern of polarized secretion changes in the inflammatory state; and (3) demonstrate that the inflammatory state alters RPE physiology. Constitutively, the human RPE secretes massive amounts of monocyte chemoattractant protein 1 (MCP-1) to the SRS and lesser amounts of IL-6 and IL-8 (Table 1), all of which contribute to the ongoing downregulation of the immune environment of the retina. RPE activation was achieved using a cocktail of IL-1p, TNF-a, and INFg with similar concentrations as that detected in the diseased eye. We showed that IL-1 p receptors are mainly localized to the apical membrane and TNF-a and INFg (subunit 1) receptors are mainly localized at the basolateral membrane. This cocktail significantly increased the secretion of various cytokines/chemokines to both baths, but significantly more to the apical bath. The increase in angiogenic cytokine secretion exceeds the increase in an-giostatic cytokine secretion. However, two chemokines generally thought to be angiostatic, interferon-inducible T-cell a-chemoattractant (I-Tac) and monokine induced by g interferon (MIG), were secreted to the apical bath in significant quantities. The mechanisms by which these chemokines exert their effects and their role in eye physiology are not yet known. Similarly intriguing and not understood are the secretions into the apical bath of interferon-inducible protein 10 (IP-10), monocyte chemoattractant protein 3 (MCP-3), and the Rantes chemo-kine. In animal model experiments from Charlotte Reme’s group, blue-light-induced oxidative damage induces the invasion of blood-borne monocytes and activation of retinal microglia, thus stimulating the secretion of cytokines to induce an inflammatory response. Our experiments strongly suggest that the RPE is a significant source of cytokines and chemokines. Thus both retinal microglia and RPE can contribute to the inflammatory response in a diseased eye. Our further demonstration that basolateral addition of the cocktail acutely increases fluid absorption across the RPE (Figure 6), from the apical to basal baths (retina to choroidal side of tissue), and significantly decreases transepithelial resistance after a 24-h treatment is important because of the possibility that with age or accumulated oxidative stress these changes can alter chemo-kine/cytokine gradients across the RPE. These gradients regulate the attraction of monocytes to the RPE basement membrane and, thus, play a role in the accumulation of drusen with age. We believe that this concept is important for understanding early events that underlie chronic disease processes, such as AMD, a notion revisited below.

  • Contact
  • Category: Eye diseases