Systemic T Cells Immunosuppression of Glioma Stem Cell-Derived Exosomes Is Mediated by Monocytic Myeloid-Derived Suppressor Cells

Abstract

A major contributing factor to glioma development and progression is its ability to evade the immune system. Nano-meter sized vesicles, exosomes, secreted by glioma-stem cells (GSC) can act as mediators of intercellular communication to promote tumor immune escape. Here, we investigated the immunomodulatory properties of GCS-derived exosomes on different peripheral immune cell populations. Healthy donor peripheral blood mononuclear cells (PBMCs) stimulated with anti-CD3, anti-CD28 and IL-2, were treated with GSC-derived exosomes. Phenotypic characterization, cell proliferation, Th1/Th2 cytokine secretion and intracellular cytokine production were analysed by distinguishing among effector T cells, regulatory T cells and monocytes. In unfractionated PBMCs, GSC-derived exosomes inhibited T cell activation (CD25 and CD69 expression), proliferation and Th1 cytokine production, and did not affect cell viability or regulatory T-cell suppression ability. Furthermore, exosomes were able to enhance proliferation of purified CD4+ T cells. In PBMCs culture, glioma-derived exosomes directly promoted IL-10 and arginase-1 production and downregulation of HLA-DR by unstimulated CD14+ monocytic cells, that displayed an immunophenotype resembling that of monocytic myeloid-derived suppressor cells (Mo-MDSCs). Importantly, the removal of CD14+ monocytic cell fraction from PBMCs restored T-cell proliferation. The same results were observed with exosomes purified from plasma of glioblastoma patients. Our results indicate that glioma-derived exosomes suppress T-cell immune response by acting on monocyte maturation rather than on direct interaction with T cells. Selective targeting of Mo-MDSC to treat glioma should be considered with regard to how immune cells allow the acquirement of effector functions and therefore counteracting tumor progression.

Fig 1. Characterization of exosomes-enriched preparation obtained from GSC.

The NTA was performed on GSC-derived exosomes samples in order to quantify particle concentration normalized for the number of producing cells or millilitre of supernatants. (A) A representative graph of NTA is shown. (B) The data show the amount of exosomes produced by different GSC samples considering either the number of cells counted at the end of the 48 hours culture or the volume of cell supernatants. The data are presented as mean ± SD; n = 7. (C) Immunoblotting of the Jurkat whole cell lysate (positive control), GSC, GSC-derived ExoQuick pellet and supernatant for exosomal surface protein TSG101 (Molecular Weight, 43kDa). (D) Representative FACS histograms of CD9, CD81 and CD63 exosome-specific markers are shown.

Fig 2. GSC-derived exosomes inhibit T-cell proliferation and expression of activation markers and modulate cytokine production of PBMCs.

CFSE-labeled PBMCs isolated from healthy donors were pretreated for 24 hours without (white column, CTRL) or with GSC-derived exosomes (black column, GSC-EXO) and stimulated for 4 days with anti-CD3 and anti-CD28. (A) Representative microscope images and respective cytometry CFSE histograms, showing the fraction of proliferative CD3+ T cells, in unstimulated PBMCs (i), stimulated PBMCs (ii) and exosomes-treated stimulated PBMCs (iii). (B-D) Histograms showing, within the PBMCs, the fraction of proliferating CD3+ (B), CD4+ (C) and CD8+ (D) T cells. (E-F) CD3+ T-cell expression of CD25 and CD69 was measured by flow cytometry on day 2. (G-L) PBMC-derived supernatants were harvested after 48 hours and used for ELISA with the Bio-plex cytokine assay system. Cytokines that showed statistically significant differences with the exosome treatment are reported. Concentration of IL-2 (G), INF-γ (H), TNF-α (I) and IL-5 (J) are expressed as pg/ml. In B-L, the data are presented as mean ± SD (n = 6). *, p<0.05 versus control.

Fig 3. GSC-derived exosomes stimulate CD25 expression and proliferation of isolated CD4+ T cells but do not affect differentiation and suppressive activity of Treg cells.

CD4+ T cells, isolated from PBMCs by negative selection, were stimulated with anti-CD3, anti-CD28 and IL-2 in the absence (white column, CTRL) or presence (black column, GSC-EXO) of GSC-derived exosomes. Expression of CD25 (A), percentage of proliferative CFSE-labelled cells in the presence of indicated stimuli (B) and frequency of CD4+/CD25+/FoxP3+ (C) was determined by flow cytometry on day 4. Columns, mean (n = 6); bars, SD; *, significantly different from the control; p<0.05. (D) CFSE-labelled purified CD4+ T cells, stimulated with anti-CD3, were co-cultured for 4 days with mitomycin-treated PBMCs and CD4+/CD25+/CD127^dim T-reg cells pre-incubated without (white column, CTRL) or with (black column, GSC-EXO) GSC-derived exosomes. Percentage of proliferative CFSE-labelled CD4+ T cells was measured by flow cytometry. Columns, mean (n = 4); bars, SD.

Fig 4. GSC-derived exosomes are internalized by monocytes and stimulate proliferation of CD14 negatively-sorted PBMCs.

Fluorescent exosomes (A) were incubated with PBMCs for 5 hours and the uptake by CD14+ monocytes, CD4+ or CD8+ T cells was measured by flow cytometric analysis in the bulk population. In representative cytometry histograms, the isotype control is in black and in grey are PBMCs cells incubated with labelled-exosomes and gated on CD14+ monocytes (left, top), on CD4+ (left, bottom) or on CD8+ T cells (right, bottom). (B) Gating and sorting strategies of PBMCs and CD14-depleted PBMC. (Top, left) physical parameters, i.e. forward scatter (FSC) and side scatter (SSC), were used to select PBMCs (gate R1). Monocytes were recognized by evaluating, in PBMCs, the expression of CD14 (gate R2, top, right panel). A PE-isotype matched antibody was used to define R2 (top, central panel). PBMCs-depleted cells were identified as cells included in R1 but not in R2. (C) Representative dot-plots showing the reanalysis of the FACS-sorted PBMCs (left panel) and CD14-depleted PBMCs (right panel). As expected, CD14-positive cells were present in the sorted PBMCs- but not in the CD14-depleted- samples. (D) Proliferation of both unfractioned PBMCs and PBMCs depleted of the CD14+ population (CD14-) was measured, after CFSE labelling assay, by flow cytometry. Cells were pre-incubated without (white column, CTRL) or with (black column, GSC-EXO) GSC-derived exosomes. Columns, mean (n = 6); bars, SD; *, significantly different from the control; P<0.05. Representative cytometry CFSE histograms of PBMCs (C-E) and CD14- cells (D-F) are shown with the percentage of proliferative cells indicated.

Fig 5. GSC-derived exosomes stimulate IL-1β, IL-6 and IL-10 production in unstimulated CD14+ monocytes within PBMC population.

Unstimulated PBMCs were incubated in the absence (white column, CTRL) or presence (black column, GSC-EXO) of GSC-derived exosomes. Incubation with LPS (square column) was used as monocyte stimulation positive control. Cells were surface stained with anti-CD14 and then stained to detect an intracellular level of IL-1β, IL-6 and IL-10 by flow cytometry. (A-C) Representative FACS plot of the intracellular staining is shown by the indicated percentage of CD14+/IL-1β+, CD14+/IL-6+ and CD14+/IL-10+ positive cells, respectively. (D-F) The mean of the experiments is shown (n = 6); bars, SD; *, significantly different from the control; P<0.05. Supernatants of unsorted unstimulated PBMCs incubated in the absence (white column) or presence of GSC-derived exosomes were harvested after 48 hours and used for a cytokine assay with the Bio-plex cytokine assay system. (G-I). As a control, cytokine concentration was also tested on a medium added with isolated exosomes (J). Concentration of IL-1β, IL-6 and IL-10, respectively, are expressed as pg/ml. Columns, mean (n = 6); bars; SD; *, significantly different from the control; P<0.05.

Fig 6. Within PBMCs population, GSC-derived exosomes promote an immunosuppressive phenotype in monocytes and stimulate the production of arginase-1 and IL-10 by Mo-MDSCs.

Unstimulated PBMCs were incubated in absence (CTRL) or presence (GSC-EXO) of GSC-derived exosomes. Cells were surface stained with anti-CD14, anti CD33, anti CD11b and HLA-DR and then stained to detect intracellular level of IL-10 and arginase-1 by flow cytometry. (A) Gating strategy: physical parameters, i.e. forward scatter (FSC) and side scatter (SSC), were used to select monocytes (gate R3, left panel). Monocytes were recognized evaluating the expression of CD11b/CD33 (gate R4, middle panel) and CD14/HLA-DR (gate R5, right panel). (B) Representative FACS histograms of the intracellular staining of IL-10 and arginase-1 and of HLA-DR staining of CD14+/CD11b/CD33+ cells are shown. (C) The percentage of cells expressing IL-10 and arginase-1 and the MFI ratio of HLA-DR expression are shown (n = 6); bars, SD;*, significantly different from the control; P<0.05.

Fig 7. Exosomes isolated from plasma of high grade glioma patients inhibit CD3+T cells proliferation through an effect mediated by CD14+monocyte.

(A) PBMCs were stimulated with anti-CD3 and anti-CD28 in the absence (white column, CTRL) or presence (black column, GBM-EXO) of different dilutions of exosomes isolated from plasma of high grade glioma patients (GBM). Healthy donor plasma-derived exosomes were used as control (striped column, EXO-HEALTHY). (B) PBMCs or CD14 negatively-sorted PBMCs (CD14-) were stimulated with anti-CD3 and anti-CD28 in the absence (white column, CTRL) or presence (black column, GBM-EXO) of exosomes isolated from plasma of GBM, 1:10 dilution. Proliferation of CD3+ was measured by CFSE assay on day 4. Columns, mean (n = 6); bars, SD;*, significantly different from the control; P<0.05.

Fig 8. Ultracentrifugated (UC) GSC-derived exosomes promote an immunosuppressive phenotype in monocytes similarly to ExoQuick (EQ) purified GSC- or GBM-derived exosomes.

(A) CFSE-labelled PBMCs isolated from healthy donors (left) or CD14 negatively-sorted PBMCs (CD14-) (right) were pre-treated for 24 hours without (white column, CTRL) or with EQ or UC isolated GSC-derived exosomes (black column, EQ GSC-EXO; grey column, UC GSC-EXO, respectively) and stimulated for 4 days with anti-CD3 and anti-CD28. Histograms show a significant difference in the percentage of proliferating CD3+ (n = 4); bars, SD;*; from the control; P<0.05. (B-D) Induction of a Mo-MDSC phenotype on monocytes. Unstimulated PBMCs were incubated in the absence (CTRL) or presence of GSC-derived exosomes purified by EQ (black column, EQ EXO-GSC) or UC(grey column, UC EXO-GSC). Cells were surface stained with (B) anti-CD14, anti CD33, anti CD11b and (C) HLA-DR and then stained to detect intracellular level of IL-10 and arginase-1 by flow cytometry. (B) Gating strategy: physical parameters, i.e. forward scatter (FSC) and side scatter (SSC), were used to select monocytes (gate R1, left panel). Monocytes were recognized by evaluating the expression of CD11b/CD33 (gate R2, middle panel) and CD14/HLA-DR (gate R3, right panel). (C) Representative FACS histograms of the intracellular staining of IL-10 and arginase-1 and of HLA-DR staining of CD14+/CD11b/CD33+ cells are shown. (D) The percentage of cells expressing IL-10 and arginase-1 and the MFI ratio of HLA-DR expression are shown (n = 3); bars, SD;*, significantly different from the control; *, P<0.05; **,P<0.01. (E) PBMCs were stimulated with anti-CD3 and anti-CD28 in the absence (white column, CTRL) or presence (black column, GBM-EXO) of exosomes isolated from plasma of glioblastoma patients (GBM) by either EQ or UC and used at 1:10 dilution. Healthy donor plasma-derived exosomes were used as a control (striped column, EXO-HEALTHY). Proliferation of CD3+ was measured by CFSE assay at day 4. Proliferation in the presence of exosomes was normalized to CD3+ cell proliferation in the absence of exosomes (CTRL set to 100%). (F) Proliferation of both unfractioned PBMCs and PBMCs depleted of the CD14+ population (CD14-) was measured, after CFSE labelling assay, by flow cytometry. Cells were pre-incubated without (white column, CTRL) or with EQGBM-derived exosomes (black column, EQ GBM-EXO) or UC GBM-derived exosomes (grey column, UC GBM-EXO). Columns, mean (n = 4); bars, SD; *, significantly different from the control; P<0.05.