Repeat exposure to polyinosinic:polycytidylic acid induces TLR3 expression via JAK-STAT signaling and synergistically potentiates NFκB-RelA signaling in ARPE-19 cells.

Dry age-related macular degeneration (AMD), accounting for approximately 90% of AMD cases, is characterized by photoreceptor death, retinal pigment epithelium (RPE) dysfunction and, ultimately, geographic atrophy - the localized death of RPE leading to loss of the center of the visual field. The pathological etiology of AMD is multifactorial, but innate immune signaling and inflammation are involved in early stages of the disease. Although numerous single-nucleotide polymorphisms in innate immune genes are associated with dry AMD, no single gene appears to cause dry AMD. Here, we hypothesized that activation of TLR3 potentiates expression of TLR3 itself and the NFκB-p65 (RelA) subunit as part of pro-inflammatory RPE signaling. Furthermore, we hypothesized that TLR3 activation can 'prime' cells to future RelA stimulation, leading to enhanced, persistent RelA expression and signaling following a second TLR3 activation. We used the human RPE-derived cell line ARPE-19 as a model system for RPE signaling and measured NFκB expression and activity in response to TLR3 stimulation with its ligand, polyinosinic:polycytidylic acid (pI:C). Activation of TLR3 with pI:C led to increased TLR3 and RelA expression that was sustained for at least 24 h. Cells exposed for a second time to pI:C after an initial pI:C exposure displayed elevated RelA expression and RelA nuclear translocation above the level generated by individual primary or secondary exposures alone. Such an elevated response could also not be generated by a single application of higher concentrations of the agonist pI:C. Additionally, we determined the mechanism for TLR3 mediated TLR3 and RelA expression by using inhibitors of canonical TLR3-TBK1-IKKε and JAK-STAT signaling pathways. These data suggest that initial exposure of ARPE-19 cells to pI:C upregulates TLR3 and RelA signaling, leading to potentiated and persistent RelA signaling potentially generated by a positive feedback loop that may cause exacerbated inflammation in AMD. Furthermore, inhibition of JAK-STAT signaling may be a possible therapeutic treatment to prevent induction of TLR3 expression subsequent to pI:C exposure. Our results identify possible therapeutic targets to reduce the TLR3 positive feedback loop and subsequent overproduction of pro-inflammatory cytokines in RPE cells.