Parainflammation associated with advanced glycation endproduct stimulation of RPE in vitro: implications for age-related degenerative diseases of the eye

Abstract

Age related macular degeneration (AMD) is one of the leading causes of blindness in Western society. A hallmark of early stage AMD are drusen, extracellular deposits that accumulate in the outer retina. Advanced glycation endproducts (AGE) accumulate with aging and are linked to several age-related diseases such as Alzheimer's disease, osteoarthritis, atherosclerosis and AMD. AGE deposits are found in drusen and in Bruch's membrane of the eye and several studies have suggested its role in promoting oxidative stress, apoptosis and lipofuscin accumulation. Recently, complement activation and chronic inflammation have been implicated in the pathogenesis of AMD. While AGEs have been shown to promote inflammation in other diseases, whether it plays a similar role in AMD is not known. This study investigates the effects of AGE stimulation on pro- and anti-inflammatory pathways in primary culture of human retinal pigment epithelial cells (RPE). Differential gene expression studies revealed a total of 41 up- and 18 down-regulated RPE genes in response to AGE stimulation. These genes fell into three categories as assessed by gene set enrichment analysis (GSEA). The main categories were inflammation (interferon-induced, immune response) and proteasome degradation, followed by caspase signaling. Using suspension array technology, protein levels of secreted cytokines and growth factors were also examined. Anti-inflammatory cytokines including IL10, IL1ra and IL9 were all overexpressed. Pro-inflammatory cytokines including IL4, IL15 and IFN-γ were overexpressed, while other pro-inflammatory cytokines including IL8, MCP1, IP10 were underexpressed after AGE stimulation, suggesting a para-inflammation state of the RPE under these conditions. Levels of mRNA of chemokine, CXCL11, and viperin, RSAD2, were up-regulated and may play a role in driving the inflammatory response via the NF-kB and JAK-STAT pathways. CXCL11 was strongly immunoreactive and associated with drusen in the AMD eye. The pathways and novel genes identified here highlight inflammation as a key response to AGE stimulation in primary culture of human RPE, and identify chemokine CXCL11 as putative novel agent associated with the pathogenesis of AMD.

Fig. 1

Human postmortem eye with AGE immunoreactivity. Intensity of AGE immunoreactivity in the outer retina of the AMD eye with drusen (>70 yrs of age) compared to a young, normal eye without drusen (<55 yrs of age). Intense immunoreactivity (purple) is seen in Bruch’s membrane (BM) and choroid (Ch) in the older donor eye with drusen (A). In the younger eye (<55 yrs), less immunoreactivity is seen in choroid and BM (B). Replacement of the primary antibody with a non-specific IgG serum served as control and demonstrated non-specific background immunoreactivity levels (C). Scale bar is (C) represents magnification for A–C.

Fig. 2

Phase contrast photomicrograph of primary culture of RPE cells at passage 1–2. Representative examples of the morphology of cells used in this study. Note the hexagonal cell profiles and pigment granules (black) within the RPE cells. Scale bar = 20 μm.

Fig. 3

RPE cell viability after AGE stimulation. MTT assay was performed at 24 h post AGE stimulation and the percentage of RPE cell survival after AGE stimulation was calculated. A dose dependent decrease in cell viability is seen. A dosage of 10 μg/ml was chosen for stimulation studies based on this dose response curve as well as the literature that reported physiological levels of AGE in human sera between values of at 2–10 μg/ml [20,33]. Cell viability grown under several control conditions such as DMEM only, unmodified BSA at 10 μg/ml, and DMEM and 1% FBS are also shown.

Fig. 4

Differential gene expression of RPE cells after 10 μg/ml AGE stimulation for 24 h. Fold changes (normalized to GADPH household gene) are compared between Agilent OligoMicroarray (black bars) and qRT-PCR (gray bars). Values are expressed as relative fold changes with respect to their corresponding (untreated) controls. Primers used for each gene product are shown in Table 3. All data shown demonstrated a >2.0-fold change and reached significance (p < 0.05). The largest upregulated fold changes were seen in CXCL11, RSAD2 and IFIT2. The greatest downregulated fold changes were seen in C8B, CALN1 and GARBG2. Data represent the mean ± SD of three independent experiments for both microarray and qRT-PCR. The Pearson’s correlation coefficient of the microarray data and the PCR for the genes shown in red and blue is 0.89 (p < 0.05).

Fig. 5

CXCL11 protein levels demonstrated by Western Blot. CXCL11 protein expression was determined by Western blot analysis and band intensities were analyzed using Image-J program. CXCL11 protein increased by 1.24-fold (±0.07 SD) in cultured human RPE cells stimulated with 10 μg/ml AGE for 24 h. The result is expressed as the mean ± SD and represents four experiments. Statistical comparisons were determined by homoscedastic one-tailed Student’s t-test. *p < 0.001 was considered significant.

Fig. 6

CXCL11 immunoreactivity in postmortem eye with AMD. Immunoreactivity displayed strong labeling (AEC, red/pink) of the cytoplasmic compartment of RPE cells overlying and surrounding the drusen site (arrows in A), but not in RPE cells distant to drusen sites in the AMD eye (B) or in age-matched control (non-diseased) eyes (C) Micrographs demonstrate endogenous brown pigment in RPE cells in all sections. All sections were counterstained with hematoxylin resulting in blue nuclei. Scale bar = 10 μm for A–C.

Fig. 7

GSEA leading edge gene sets analysis. The GSEA analysis selects the genes that appear across multiple enriched gene sets (y-axis) from the public databases. These genes (x-axis) are considered to have higher biological significance. The analysis of our microarray data identified genes highly correlated with interferons, proteasome degradation, caspase pathways, DNA replication, toll-like receptor signaling and neurodegeneration.

Fig. 8

Potential cellular response pathways of RPE cells upon AGE stimulation. The two main signaling pathways suggested by our data are the NF-κB pathway and the JAK-STAT pathway. Extracellular receptors that bind AGE known to be present on RPE cells include toll-like receptors (TLRs), receptors for advanced glycation endproducts (RAGEs), and advanced glycation endproducts receptors (AGERs). RAGEs and TLRs are mainly responsible for inflammatory processes within the cell and AGERs are important inhibitors of these signals. Supernatant data that support the two pathways are shown in purple retangles. mRNA data that support the pathways are shown in orange ovals. Studies from the literature that support the pathways shown include: 1, Vlassara and Striker [43]; 2, Chen et al. [62]; 3, Yang et al. [51]; 4, Kim et al. [34]; 5, Oeckinghaus et al. [67]; 6, Kato et al. [68].