Article Figures & Data
Figures
- FIGURE 1.
Flow cytometric characterization of NKG2D ligands expressed by tumor and nontumor cells. Tumor cell lines (Meso; mesothelioma, PC3 and Du145; prostate, EBV-B, EBV-transformed B-lymphoblastoid cells) or nontumor cells (HFF; skin fibroblasts; PBLs) were stained with various Abs against NKG2D ligands, or as positive controls with anti MHC class I and/or CD81 (solid line). Staining was compared with irrelevant isotype control Ab (dashed line). Numerical values indicating MFI of the histogram are shown.
- FIGURE 2.
Characterization of tumor exosomes. Whole cell lysates of a mesothelioma cell line, or purified exosomes from this line were analyzed by Western blot (2 or 10 μg/well), demonstrating expression of typical exosome markers including MHC class I and TSG101. Expression of the MHC-like molecules MICA and MICB was also evident (A). These exosomes were coupled to the surface of aldehyde-sulfate latex beads, and after blocking were stained with primary Ab (as indicated) followed by goat anti-mouse IgG-FITC. Beads were analyzed by flow cytometry, demonstrating exosome surface expression of class I, MICA, MICB, and ULBP2. MFI values are shown (B). Similarly, 7 μg of these exosomes (Meso) were compared with 7 μg of exosomes purified from malignant mesothelioma PF (from two patients, designated PF1 and PF2) by Western blot (C). Exosomes from PF1 were further analyzed by flow cytometry, as above (D). TGFβ1 ELISA analysis of exosomes from cultured mesothelioma cell exosomes (Meso, mean of five preparations) or from three PF exosome preparations (PF1, PF2, and PF3, each measured in triplicate) by ELISA, indicating comparable levels of TGFβ1 (E).
- FIGURE 3.
Tumor exosomes trigger down-modulation of cell surface NKG2D. Schematic representation of the multicolor flow cytometry gating scheme used to analyze lymphocyte NKG2D and CD94 or CD69 expression (A). Exosomes purified from a mesothelioma cell line were added at doses (1–10 μg/105 PBL) to fresh healthy donor PBL and after 48-h incubation, flow cytometry was performed to determine levels of expression of these markers, by CD8+ T cells (B) or NK cells (C). Data show the proportion of positive cells and MFI as indicated. The bar graph (left) shows constitutive levels, and levels following IL-15 or TGFβ1 treatment (both at 50 ng/ml). The line graphs (right) show expression relative to untreated PBL (normalized to a value of 1). Graphs show mean ± SE, of triplicates. ∗, p < 0.05; ∗∗, p < 0.01; ∗∗∗, p < 0.001. The data are representative of over 10 experiments with various tumor cell type-derived exosomes, and different healthy donors.
- FIGURE 4.
Kinetics of exosome-mediated down-regulation of NKG2D. IL-15 or TGFβ1 (50 ng/ml each) or mesothelioma exosomes (1 or 5 μg) was added to 105 fresh healthy donor PBL, and changes in proportion of NKG2D-positive CD8+ T cells (A) and NK cells (B) cells was monitored by multicolor flow cytometry immediately, and thereafter at indicated times up to 96 h. Graph shows the percentage of NKG2D-positive cells relative to untreated (normalized to a value of 1), for each time point. The results were obtained from three experiments (mean ± SE, n = 3) conducted using different donors and different preparations of mesothelioma exosomes on separate occasions.
- FIGURE 5.
Exosome-mediated NKG2D down-regulation is dependent on the exosome phenotype. Fresh healthy donor PBL were left untreated, or were treated with TGFβ1 (n = 8) or 10 μg of exosomes isolated from various cell types, including mesothelioma (n = 8), prostate cancer PC3 (n = 2), and DU145 (n = 2), B-lymphoblastoid cell line IB4 (n = 2), HFF fibroblasts (n = 3), and from 72-h lymphocyte cultures (n = 2) as indicated, of n experiments performed on different occasions with different blood donors. The effect of exosomes isolated from cancer patients PF is shown as ▨, showing average response measured from the same blood donor, with exosomes from three patients. Following 48-h incubation, NKG2D expression was measured by flow cytometry. Graph shows the percentage of NKG2D-positive cells relative to untreated (mean + SE of n experiments: ∗, p < 0.05; ∗∗, p < 0.01; ∗∗∗, p < 0.01, n/s = not significant).
- FIGURE 6.
Down-regulation of NKG2D is driven by exosomal TGFβ1, and partially by MICA. Fresh healthy donor PBL were treated with TGFβ1 (10 ng/ml) or with mesothelioma exosomes (10 μg/105 cells), in the presence of neutralizing Abs against human MICA, anti-TGFβ1 Ab, or with isotype-matched control Ab (all at 5 μg/ml), as indicated. Expression of NKG2D by CD8+ T cells (A) and NK cells (B) was assessed by flow cytometry. Graphs show mean + SE, of triplicate measurements; ∗∗, p < 0.01), and is representative of two experiments performed with different donors.
- FIGURE 7.
IL-15 counters exosome down-regulation of NKG2D. PBL obtained from three healthy donors, were treated with cytokines (IL-15 or TGFβ1, 0–50 ng/ml) in the absence or presence of mesothelioma exosomes (10 μg/105 PBL), and at 48 h, expression of NKG2D in CD8+ T cells (A) and NK cells (B) was measured by flow cytometry. Graphs show the percentage of NKG2D-positive cells (left) or NKG2D expression levels (MFI) (right) relative to untreated PBL (normalized to a value of 1). (mean ± SE, n = 3; ∗, p < 0.05; ∗∗, p < 0.01).
- FIGURE 8.
Tumor exosomes impair NKG2D-mediated lymphocyte activation. Fresh healthy donor PBL, were treated with mesothelioma exosomes and/or IL-15 (10 ng/ml) in the presence of IL-2 (100U/ml). After 48 h, PBL were washed and labeled with CFSE, and added to NKG2D Ab, or isotype control Ab-coated plates (3.5 × 105 cells/well). After incubation for 30 min, wells were washed three times, and fluorescence was measured to determine number of adhering cells. Graph shows mean ± SE of triplicates, with isotype control subtracted) (A). In a similar manner, PBL pretreated in parallel were incubated on Ab-coated plates in the presence of brefeldin A, and following 18 h of incubation, the production of IFN-γ was determined by flow cytometry. Graphs of mean ± SE of triplicate measurements, with isotype control subtracted, show the proportion of positive cells (▪) or MFI (□), gating on CD3-positive cells (B) or CD3-negative cells (C). PBL pretreated similarly, but for 72 h, were tested for their ability to kill murine FcγR+ P815-mastocytoma cells, in the presence of 2.5 μg/ml NKG2D-agonist Ab (5C6) (D) or isotype-matched control Ab (E). Graphs of mean ± SE of triplicate measurements.
In this issue
