Dry eye disease (DED) is the most common disease affecting vision and quality of life. PM_2.5 was a potential risk of DED. Herein, we conducted animal exposure and cell-based studies to evaluate the pathogenic effect of PM_2.5 exposure on the ocular surface and DED etiological mechanisms. C57 mice were exposed to filtered air and PM_2.5 aerosol. We assessed health conditions and inflammation of the ocular surface by corneal fluorescein staining and immunohistochemistry. In parallel, cultured human corneal epithelial cells (HCETs) were treated with PM_2.5, followed by characterization of cell viability, intracellular ATP level, mitochondrial activities, and expression level of DED relevant mRNA and proteins. In mice, PM_2.5 exposure induced severe superficial punctate keratopathy and inflammation in their cornea. In HCETs, cell proliferation and ROS generation followed dose-response and time-dependent manner; meanwhile, mitochondrial ROS (mtROS) level increased and mitochondrial membrane potential (MMP) level decreased. Inflammation cascade was triggered even after short-term exposure. The reduction of ATP production was alleviated with Nrf2 overexpression, NF-κB P65 knockdown, or ROS clearance. Nrf2 overexpression and P65 knockdown reduced inflammatory reaction through decreasing expression of P65 and increasing of Nrf2, respectively. They partly alleviated changes of ROS/mtROS/MMP. This research proved that PM_2.5 would cause DED-related inflammation reaction on corneal epithelial cells and further explored its mechanism: ROS from mitochondrial dysfunctions of corneal epithelial cells after PM_2.5 exposure partly inhibited the expression of anti-inflammatory protein Nrf2 led the activation of inflammatory protein P65 and its downstream molecules, which finally caused inflammation reaction. Graphical abstract

Purpose: To determine the effect of decorin on the barrier function of human retinal pigment epithelial (RPE) cells under high glucose (HG) plus hypoxia conditions. Methods: Human RPE cells (ARPE-19) were cultured for 18 days in normal glucose (5.5 mM) or HG (25 mM) medium. In addition, to mimic the hypoxic impact which occurs in diabetic retinopathy, cells were treated with 100 µM cobalt chloride (CoCl2) during the last two days of the experiment. Decorin 100 nM was applied 1 h before the addition of CoCl2. The RPE barrier function was evaluated by measuring transepithelial electrical resistance (TER) and apical-basolateral permeability of fluorescein isothiocyanate (FITC)-dextran. The content and distribution of tight junction proteins (claudin-1, occludin and zonula occludens-1 [ZO-1]) were examined by western blot and immunofluorescence. The p38 mitogen-activated protein kinase (MAPK) phosphorylation was evaluated by western blot and small interfering RNA transfection to p38 MAPK was also performed in ARPE-19 monolayers. Results: HG + hypoxia significantly increased FITC-dextran permeability, paralleled by decreased TER. Decorin reversed both of these effects. HG + hypoxia-induced reduction and disorganization of occludin and ZO-1 were also reversed by decorin. Decorin could prevent the activation of p38 MAPK induced by hypoxia. Silence of p38 MAPK by RNA interference could also inhibit the breakdown of ARPE-19 cell monolayer induced by HG + hypoxia. Conclusion: RPE barrier disruption induced by HG plus hypoxia is prevented by decorin via suppression of p38 MAPK activation, which could present a new therapeutic strategy for inhibition of diabetic macular edema.

Vascular endothelial growth factor (VEGF) is one of the major regulatory molecules in diabetic retinopathy (DR). In our previous study, we demonstrated that succinate levels were elevated in the retinas of diabetic rats and that the knockdown of the succinate receptor, G-protein-coupled receptor 91 (GPR91), inhibited the release of VEGF and attenuated retinal vascular disorder in the early stages of DR. In the present study, we examined the signaling pathways involved in the GPR91-dependent release of VEGF in the retinal ganglion cell line, RGC-5. The cells were infected with a lentiviral small hairpin RNA (shRNA) expression vector targeting GPR91 (LV.shGPR91). Immunofluorescence staining revealed that GPR91 was predominantly localized in the cell bodies of the RGC-5 cells. RT-qPCR, western blot analysis and ELISA indicated that succinate exposure upregulated VEGF expression, activated the extracellular signal-regulated protein kinase (ERK)1/2, c-Jun N-terminal kinase (JNK) and p38 mitogen-activated protein kinase (MAPK) signaling pathways and led to the release of cyclooxygenase-2 (COX-2) and prostaglandin E2 (PGE2). The knockdown of GPR91 inhibited ERK1/2 and JNK activity, but did not inhibit the activation of the p38 MAPK pathway. The increase in COX-2 expression and the release of PGE2 were inhibited by transduction with LV.shGPR91 and ERK1/2, JNK and COX-2 inhibitors. The expression and release of VEGF showed similar results. Cell Counting Kit-8 (CCK-8) assays revealed that the shRNA-mediated knockdown of GPR91 decreased the proliferation of RF/6A cells cultured in succinate-conditioned medium. Our data suggest that GPR91 modulates the succinate-induced release of VEGF through the MAPK/COX-2/PGE2 signaling pathway.

MicroRNAs (miRNAs) - as negative regulators of target genes - are associated with various human diseases, but their precise role(s) in diabetic retinopathy (DR) remains to be elucidated. The aim of this study was to elucidate the involvement of miRNAs in early DR using in silico analysis to explore their gene expression patterns. We used the streptozotocin (STZ)-induced diabetic rat to investigate the roles of miRNAs in early DR. Retinal miRNA expression profiles from diabetic versus healthy control rats were examined by miRNA array analysis. Based on several bioinformatic systems, specifically, gene ontology (GO) and the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways, we identified signatures of the potential pathological processes, gene functions, and signaling pathways that are influenced by dysregulated miRNAs. We used quantitative real-time polymerase chain reaction (qRT-PCR) to validate six (i.e. those with significant changes in expression levels) of the 17 miRNAs that were detected in the miRNA array. We also describe the significant role of the miRNA-gene network, which is based on the interactions between miRNAs and target genes. GO analysis of the 17 miRNAs detected in the miRNA array analysis revealed the most prevalent miRNAs to be those related to biological processes, olfactory bulb development and axonogenesis. These miRNAs also exert significant influence on additional pathways, including the mitogen-activated protein and calcium signaling pathways. Six of the seventeen miRNAs were chosen for qRT-PCR validation. With the exception of a slight difference in miRNA-350, our results are in close agreement with the differential expressions detected by array analysis. This study, which describes miRNA expression during the early developmental phases of DR, revealed extensive miRNA interactions. Based on both their target genes and signaling pathways, we suggest that miRNAs perform critical regulatory functions during the early stages of DR evolution.