Modulation of Immune Responses by Extracellular Vesicles From Retinal Pigment Epithelium

Abstract

Purpose: Extracellular vesicles (EV), such as exosomes, are important mediators of intercellular communication and have been implicated in modulation of the immune system. We investigated if EV released from retinal pigment epithelium (RPE) modulate immune responses in vitro.

Methods: Extracellular vesicles were isolated from ARPE-19 cultures stimulated or not with the inflammatory cytokines IL-1β, IFN-γ, and TNF-α. Isolated EV were characterized by nanoparticle flow and Western blot analyses. Retinal pigment epithelium-derived EV were cultured with human peripheral blood mononuclear cells, which were assayed for T-cell proliferation by 3H-thymidine incorporation. Retinal pigment epithelium-derived EV were also independently cultured with enriched lymphocytes or monocytes. Cell phenotype and cell death were evaluated by flow cytometric analysis. Cytokine levels were assayed in culture supernatants by multiplex bead analysis.

Results: The concentration of ARPE-derived EV from cytokine-stimulated cultures was slightly higher than from nonstimulated cultures. The size of EV was approximately 100 nm in both groups. Extracellular vesicles from both nonstimulated and cytokine-stimulated ARPE-19 significantly inhibited T-cell proliferation without affecting T-cell viability. Culture of EV from nonstimulated ARPE-19 with undifferentiated human monocytes induced an immunoregulatory phenotype with a significantly higher percentage of CD14++CD16+ monocytes and upregulation of TGF-β1. Culture of EV from cytokine-stimulated ARPE-19 cells with human monocytes induced upregulation of several proinflammatory cytokines and monocyte death.

Conclusions: Retinal pigment epithelium cells constitutively secrete EV in the size range of exosomes, with increased release from RPE cells stimulated with inflammatory cytokines. Extracellular vesicles from both nonstimulated and cytokine-stimulated RPE inhibited T-cell stimulation. Extracellular vesicles from nonstimulated ARPE-19 cells promoted an immunoregulatory CD14++CD16+ phenotype in human monocytes, while EV from cytokine-stimulated ARPE-19 cells caused human monocyte death. These findings suggest that RPE cells use EV to induce a downregulatory immune environment under homeostatic conditions. In an inflammatory milieu, RPE-derived EV may mitigate a potentially harmful inflammatory response through killing of monocytes.

Figure 1

Characterization of EV released by RPE cells stimulated or not with inflammatory cytokines. Concentration (A) and size (B) of EV isolated with ExoQuick from ARPE-19 cells stimulated or not with 10 ng/mL each IL-1β, IFN-γ, and TNF-α for 48 hours as measured by NanoSight nanoparticle analysis. Presented are the mean ± SEM from three independent experiments (n = 5 per group). Student's t-test did not reveal significant differences between groups for concentration (P = 0.12) or size (P = 0.84 for mean, P = 0.98 for mode). (C) Extracellular vesicles isolated from ARPE-19 with ExoQuick were labeled with BODIPY-FITC and captured with anti-human CD81-coated magnetic nanoparticles prior to flow cytometric analysis. Counting beads were included to calculate EV concentration. Presented are the mean ± SEM from three independent experiments (n = 5 per group). (D) ARPE-19 cells stimulated or not with 10 ng/mL each IL-1β, IFN-γ, and TNF-α for 48 hours were stained for surface and intracellular CD81 prior to flow cytometric analysis (n = 2 per group). Data representative of three independent experiments. (E) Extracellular vesicles isolated from ARPE-19 with ExoQuick were digested and analyzed for the apolipoprotein markers ApoB100 and ApoA1, as well as the extracellular vesicle marker CD63 by Western blot.

Figure 2

Retinal pigment epithelium–derived EV inhibit T-cell proliferation but do not induce T-cell death. (A) Extracellular vesicles were isolated from ARPE-19 cells stimulated (EV- stimulated; 5.7e⁸ ± 1.1e particles/mL) or not (EV: nonstimulated; 3.0e⁸ ± 5.4e particles/mL) with IL-1β, IFN-γ, and TNF-α (10 ng/mL of each), and 5 to 50 μL (data from 50 μL are shown) of EV isolate were cultured with PBMC isolated from peripheral blood of uveitis patients (100,000 cells/well in RPMI with 10% FBS). T-cell stimulation was measured as described in the Methods section. Stimulation index (SI) was calculated by dividing the counts per minute (CPM) of experimental groups by that of the control nonstimulated group. P values calculated by ANOVA with Tukey's multiple comparisons test. (B) Elutriated human lymphocytes (0.5–1 × 10⁶ cells/well) were cultured alone or with 75 μL EV released by resting (1.2e⁹ ± 6.1e particles/mL) or cytokine-stimulated (1.2e⁹ ± 3.7e particles/mL) ARPE-19 cells. At 1, 2, 3, and 4 days of culture, cells were harvested and stained with CD3 and DAPI and analyzed with flow cytometry. Data from 4-day cultures are shown. All other time points showed similar low levels of T-cell DAPI inclusion.

Figure 3

Modulation of human monocyte phenotype by ARPE-19–derived EV. (A) Extracellular vesicles were isolated from ARPE-19 stimulated (EV: stimulated) or not (EV: nonstimulated) with IL-1β, IFN-γ, and TNF-α 10 ng/mL each for 48 hours, and 75 μL of EV isolate (1.1e⁹ ± 4.8e particles/mL for cytokine-stimulated ARPE-19 and 9.0e⁸ ± 1.6e particles/mL for nonstimulated ARPE-19) were cultured with elutriated human monocytes (0.5–1 × 10⁶ cells/well) for 24 hours. Cells were then stained with DAPI to assess viability and for CD14 and CD16 and analyzed by flow cytometry. Top row: dot plots showing forward (FSC) and side scatter (SSC); graph on the right shows the proportion of viable cells compared with those kept at 4° during the culture period. Bottom row: dot plots showing CD14 and CD16 expression; graph on the right shows the percentage of intermediate (CD14⁺⁺CD16⁺) monocytes of all CD14⁺ cells in the different groups. Data representative of three independent experiments. (B) Extracellular vesicles were isolated from ARPE-19 not exposed to cytokines, and the indicated volume of EV isolate (2.2e⁸ ± 4.0e particles/mL) was cultured with elutriated human monocytes for 24 hours. Cells were then stained for CD14 and CD16 and analyzed by flow cytometry. * P < 0.05; P values calculated by ANOVA with Tukey's multiple comparisons test.

Figure 4

Cytokine levels in cultures of human monocytes with ARPE-19–derived EV. Extracellular vesicles isolated from ARPE-19–stimulated (EV: stimulated) or not (EV: nonstimulated) with IL-1β, IFN-γ, and TNF-α were cultured with elutriated human monocytes for 24 hours as in Figure 3, and supernatants were analyzed for the indicated cytokines by multiplex bead analysis. Results from two independent experiments are shown. P values calculated by ANOVA with Tukey's multiple comparisons test or Student's t-test.