**1. Introduction**

One of the most oxidative environments in the body is the retinal pigment epithelium (RPE) [1], which is in close contact with the photoreceptors and maintains visual function [2]. Low levels of reactive oxygen species are required to maintain physiological functions [3], but the combination of visible light exposure, elevated metabolic activity, the accumulation of oxidized lipoproteins, and decreased antioxidant functions during aging make the retinal tissue vulnerable to oxidative stress [4,5]. Oxidative damage to the RPE has therefore been identified as a contributing factor to di fferent retinal degenerative diseases, such as age-related macular degeneration or Stargardt disease [6–11].

Consistent with these observations, constant oxidative stress can trigger chronic inflammation, subsequently leading to cellular damage in the RPE/retina [6,12]. On the basis of genetic polymorphisms in genes of the complement system, systemic complement activation, and local complement deposition in degenerative retinal tissue, a contribution of the complement system to oxidative stress-related retinal degeneration has been hypothesized [7,13,14]. The complement system is composed of over 40 proteins, which bridge the innate and adaptive immune defense [15]. The main functions are (I) the removal of damaged cells, (II) protection against invading pathogens, and (III) the attraction of immune cells.

Besides the traditional view, evidence is accumulating that complement components are also involved in physiological processes, such as responses to oxidative stress and cellular maintenance [6]. The complement system comprises several soluble and membrane-bound proteins and receptors, which can be produced by a number of cells (including nonimmune cells and extrahepatic tissue) and contribute to autocrine cell physiology [16]. The role of endogenous complement-dependent regulation of cellular homeostasis has recently been extensively studied [17]. T-cells, B-cells, and human airway epithelial cells contain intracellular stores of C3, which is endogenously cleaved into its active forms, C3a and C3b, by intracellular proteases [18–20]. Activated C3 is correlated with the activation of the NLR family pyrin domain containing 3 (NLRP3) inflammasome in T-e ffector cells [18], which leads to proinflammatory cytokine release. An antagonizing complement modulation has been described for regulatory T-cells, where C3aR and C5aR1 activation resulted in the activation of the forkhead box P3 (FOXP3) transcription factor [18,21]. The FOXP3 transcription factor acts in multimodal fashion and stimulates the release of anti-inflammatory cytokines and proangiogenic factors [21–23].

Oxidative stress and inflammasome activation have previously been correlated with external complement activity in RPE cells [6,24]. FOXP3 activation in RPE cells also depends on extracellularly added complement components [25]. However, an RPE-derived complement has not been discussed as a source for NLRP3 or FOXP3 modulation. Complement components can be produced by RPE cells [26], and their expression is changed under oxidative stress [27–31]. Further activated forms of C3 (C3a), independently of extracellular complement sources, are also secreted by RPE cells, suggesting a similar function of the complement system in RPE cells compared to T-cells [32–35].

In this study, we report that H2O2 stimulated the accumulation of complement proteins C3, complement factor H (CFH), and properdin in RPE cells and increased the expression of complement receptors C5aR1 and CR3 (subunit CD11b). This was accompanied by increased *nlrp3* inflammasome expression and the FOXP3-associated release of proangiogenic factors. Our results indicate a cell homeostatic function of cell-derived complement components that is independent of external complement receptor ligands.
