Informed consent was obtained from the relatives of all donors. The study was conducted in accordance with The Declaration of Helsinki and was approved by the institutional review board of The First Affiliated Hospital of Chongqing Medical University (Chongqing, China). Human ocular posterior segments free of any known ocular disease were obtained from Chongqing Eye Bank (Chongqing, China). The retina was torn from RPE with forceps, and the RPE was rubbed off of the choroid with a cotton swab. Choroids were torn from sclera, and endothelial cells were isolated from 2 choroids of 1 donor by treatment with 0.2% type II collagenase in minimum essential medium (Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) at 37°C and 5% CO₂ until the tissue was visibly digested. The cell suspension was filtered through sterile 70 and 40-mm filters (BD Biosciences, Franklin Lakes, NJ, USA). The filtered suspension was centrifuged at 300 × g and suspended in medium containing anti-CD31 magnetic beads (MiltenyiBiotec, GmbH, Bergisch Gladbach, Germany) according to the manufacturer's protocol. CECs were isolated from human choroid and cultured in endothelial growth medium (EGM-2 MV; Lonza Group Ltd., Basel, Switzerland) supplemented with 5% fetal bovine serum (FBS; Invitrogen; Thermo Fisher Scientific, Inc.) at 37°C in a humidified atmosphere with 5% CO₂, as previously described (17). CECs between passages three and five were used for experiments. ARPE-19 cells were obtained from the American Type Culture Collection and maintained in Dulbecco's modified Eagle's medium/F12 (DMEM/F12; 1:1; Invitrogen; Thermo Fisher Scientific, Inc.) supplemented with 10% FBS (Invitrogen; Thermo Fisher Scientific, Inc.), 100 U/ml penicillin and 100 µg/ml streptomycin in a humidified incubator at 37°C in 5% CO₂.

ARPE-19 monolayer cultures were seeded at 2×10⁵ cells/cm² onto Transwell filters (diameter, 12 mm; pore size, 0.4 µm) which had been coated with 160 ml Matrigel (dilution, 1:30; BD Biosciences, Franklin Lakes, NJ) in DMEM/F12 (Invitrogen; Thermo Fisher Scientific, Inc.) and air-dried overnight, as previously described (18). DMEM/F12 (Invitrogen; Thermo Fisher Scientific, Inc.) medium supplemented with 0.5% FBS (Invitrogen; Thermo Fisher Scientific, Inc.) was used to culture the ARPE-19 cells. The medium was replaced every 4 days and cells were maintained for 8 weeks.

The ARPE-19 monolayer was subsequently cultured with or without 20 ng/ml IL-17 (R&D Systems, Inc., Minneapolis, MN, USA) in EGM-2 MV medium (Lonza Group Ltd.) for 48 h, and conditioned medium with 20 ng/ml IL-17 (IL-17-CM) or without IL-17(CM) was obtained from the lower chamber of the ARPE-19 monolayer. To avoid the stimulation of CECs by redundant IL-17 in IL-17-CM and to reasonably match the control conditions, both CM and IL-17-CM were supplemented with 0.2 ug/ml neutralizing antibodies (cat. no. AF-317-NA; R&D Systems, Inc.) against IL-17for 10 min, at room temperature, prior to use.

CECs were seeded at 2×10³ cells per well in 96-well plates (Corning Incorporated, Corning, NY, USA) and starved in FBS free EGM-2 MV medium (Lonza Group Ltd) for 12 h. Cells were subsequently treated with CM or IL-17-CM for 24, 48 or 72 h. Cell proliferation was measured according to the Cell Proliferation Reagent Water Soluble Tetrazolium 1 (WST-1) kit manufacturer's protocol (C0035; Beyotime Institute of Biotechnology, Shanghai, China). Cells were incubated in WST-1 solution for 2 h at 37°C and the absorbance at a wavelength of 450 nm was measured using a spectrophotometer (SpectraMax; Molecular Devices, LLC, Sunnyvale, CA, USA). All measurements were taken in quadruplicate.

CECs were cultured to confluence in 6-well plates containing EGM-2 MV (Lonza Group Ltd) supplemented with 5% FBS (Invitrogen; Thermo Fisher Scientific, Inc.) and subsequently starved in FBS free EGM-2 MV for 12 h. The monolayer was injured with a sterile pipette tip and washed with PBS to remove debris. The injured monolayer was treated with CM, IL-17-CM or IL-17-CM with 0.1 µg/ml neutralizing antibody against CXCL8 (R&D Systems, Inc., cat. no. AF-208-NA) and/or 0.2 µg/ml neutralizing antibody against CCL2 (R&D Systems, Inc. cat. no. AF-279-NA), for 10 min at room temperature. Images were taken at 0, 4, 8 and 12 h with a fluorescence microscope (DM6000; Leica, Wentzler, Germany). Wound closure rate was measured using Image Pro Plus 6.0 software (Media Cybernetics, Inc., Rockville, MD, USA) and measurements were taken in triplicate.

Ice-cold growth factor reduced basement membrane matrixgel (Matrigel; BD Biosciences) was added at 300 µl per well to pre-cooled 24-well plates at 4°C and allowed to polymerize at 37°C for 30 min. CECs (1×10⁵) were plated onto the gel surface and incubated at 37°C for 16 h with 200 µl CM, IL-17-CM or IL-17-CM with anti-CXCL8 and/or anti-CCL2 antibodies. Cell rearrangement and tube formation was visualized using phase-contrast microscopy. Images of five random fields at ×5 magnification per well were taken and endothelial tube length was quantified using Image Pro Plus 6.0 software (Media Cybernetics, Inc.).

The ARPE-19 monolayer was maintained in DMEM/F12 (Invitrogen; Thermo Fisher Scientific, Inc.) medium supplemented with 0.5% FBS (Invitrogen; Thermo Fisher Scientific, Inc.) for 8 weeks to allow polarization. The cells were then treated with or without IL-17 (50 ng/ml) for 24 h. The supernatant of the lower chamber was centrifuged at 300 × g, 37°C for 5 min to remove particulates and stored at −70°C until further use. Total protein for each sample was determined using a protein assay kit (Bio-Rad Laboratories, Inc. Hercules, CA, USA). VEGF levels were measured using a human ELISA development kit (R&D Systems, Inc. cat. no. SVE00) according to the manufacturer's protocol. Each experiment was repeated four times.

Rac1 and RhoA activity was analyzed using Rac1 and RhoA pull-down activation assay kits (Cytoskeleton, Inc., Denver, CO, USA), according to the manufacturer's protocols. The CECs were cultured in conditions identical to those previously described in the cell migration assay. The cells were washed with ice-cold PBS and lysed in ice-cold lysis buffer with the addition of a 1:100 final dilution of a protease inhibitor cocktail (Cytoskeleton, Inc., Denver, CO, USA). Lysed cells were harvested with a cell scraper and centrifuged at 10,000 × g at 4°C for 1 min. The supernatant was aliquoted, snap frozen in liquid nitrogen and stored at −70°C until further use. An equal volume of supernatant was incubated for 1 h at 4°C in the presence of glutathione S-transferase fused p21 activated kinase 1-p21 binding domain beads for GTP-Rac1 adsorption, or Rhotekin-Rho binding domain bound to glutathione-sepharose beads for GTP-Rho adsorption. The supernatant was subsequently washed three times with the wash buffer. Bound proteins were eluted in sample buffer and analyzed by western blot analysis using antibodies against Rac1 and RhoA. All materials used including those for the western blot analysis, were included in the pull-down assay kits.

Data are presented as the mean ± standard deviation, using SPSS software v11.0 (SPSS, Chicago, IL, USA) the cell proliferation assay and ELISA results between groups were statistically compared using the Student's t-test. Tube formation assay and migration assays results between CM and IL-17-CM were analyzed using the Student's t-test. Tube formation, migration and the Rac1 and RhoA pull-down assays were analyzed using one-way analysis of variance, followed by a Bonferroni correction test as a post-hoc test. P<0.05 was considered to indicate a statistically significant difference.

CM and IL-17-CM induced the proliferation of CECs, however there was no significant difference in the relative absorbance of the supernatant between the two groups at 24, 48 and 72 h (Fig. 1A). IL-17-CM significantly increased tube formation in CECs. The data was normalized to a CM value of 1, with the relative tube length of CECs stimulated by IL-17-CM at 1.25±0.08 (P=0.001; Fig. 1B). The migration distance of the CECs increased with duration in both the IL-17-CM and the CM treated groups. The migration distance of CECs stimulated with IL-17-CM was significantly increased at 12 h in comparison with CM (Fig. 1C; CM, 185±26 µm; IL-17-CM, 285±30 µm; P=0.012).

To investigate whether migration and tube formation of CECs was induced by VEGF secreted by the RPE, the expression of VEGF was detected in the supernatant of the ARPE-19 monolayer with or without IL-17 stimulation. The expression levels of VEGF in the supernatant of ARPE-19 with or without IL-17 stimulation were not significantly different (Fig. 2; IL-17-CM, 0.19±0.12 pg/µg total protein; CM, 0.16±0.09 pg/µg total protein).

In the author's previous study, the levels of CXCL8 and CCL2 increased markedly in the supernatant from RPE cells treated with IL-17 (16). To investigate the potential role of CXCL8 and CCL2 in the migration of CECs in IL-17-CM, CXCL8 and/or CCL2 expression was neutralized in IL-17-CM prior to migration and tube formation assays. The migration distance and tube formation ability of CECs in CM, IL-17-CM, IL-17-CM with anti-CXCL8 and anti-CCL2, IL-17-CM with anti-CCL2 only or IL-17-CM withanti-CXCL8 only was investigated (Fig. 3). The data was normalized to a CM value of 1.CEC migration was significantly inhibited by IL-17-CM with the combination ofanti-CXCL8 and anti-CCL2 (P=0.024), however the inhibitory effect was not present in treatment with either antibody alone (Fig. 3A). Similarly, the addition of bothanti-CXCL8 and anti-CCL2 in IL-17-CM significantly reduced tube length compared with IL-17-CM (IL-17-CM, 1.25±0.08; IL-17-CM + anti-CCL2 + anti-CXCL8, 1.06±0.07; P=0.04), however addition of either antibody alone had no significant effect on tube length (Fig. 3B). The values were normalized to a CM value of 1. This data suggested that the combination of CXCL8 and CCL2 expression in IL-17-CM was essential to induce the migration and tube formation of CECs.

The expression levels of Rac1 and RhoA in CECs was significantly increased in IL-17-CM treated groups in comparison with the CM group (Fig. 4; Rac1, P=0.008; RhoA, P=0.042). Neutralizing antibodies to both CXCL8 and CCL2 significantly inhibited Rac1 and RhoA activity in the IL-17-CM group following 10 min of treatment (Rac1, P=0.037; RhoA, P=0.038). A significant inhibition of Rac1 and RhoA activity was not present when CXCL8 or CCL2 alone were neutralized with the respective antibodies. Collectively, these results suggested that IL-17-CM induced the upregulation of Rac1 and RhoA activity in CECs, which may be in part mediated by CXCL8 and CCL2.