CD73 expression on extracellular vesicles ... - Wiley Online Library

2430

Lesley Ann Smyth et al.

Eur. J. Immunol. 2013. 43: 2430–2440

DOI: 10.1002/eji.201242909

CD73 expression on extracellular vesicles derived from CD4+CD25+Foxp3+ T cells contributes to their regulatory function Lesley Ann Smyth1 , Kulachelvy Ratnasothy1 , Julia Y. S. Tsang2 , Dominic Boardman1 , Alice Warley3 , Robert Lechler∗1 and Giovanna Lombardi∗1 1

MRC Centre for Transplantation, King’s College London, Guy’s Hospital, London, UK Department of Anatomical and Cellular Pathology, University of Hong Kong, Hong Kong SAR, China 3 Centre for Ultrastructural Imaging, King’s College London, Guy’s Campus, London, UK 2

CD4+ CD25+ Foxp3+ Treg cells maintain immunological tolerance. In this study, the possibility that Treg cells control immune responses via the production of secreted membrane vesicles, such as exosomes, was investigated. Exosomes are released by many cell types, including T cells, and have regulatory functions. Indeed, TCR activation of both freshly isolated Treg cells and an antigen-specific Treg-cell line resulted in the production of exosomes as defined morphologically by EM and by the presence of tetraspanin molecules LAMP-1/CD63 and CD81. Expression of the ecto-5-nucleotide enzyme CD73 by Treg cells has been shown to contribute to their suppressive function by converting extracellular adenosine-5-monophosphate to adenosine, which, following interaction with adenosine receptors expressed on target cells, leads to immune modulation. CD73 was evident on Treg cell derived exosomes, accordingly when these exosomes were incubated in the presence of adenosine-5-monophosphate production of adenosine was observed. Most importantly, CD73 present on Treg cell derived exosomes was essential for their suppressive function hitherto exosomes derived from a CD73-negative CD4+ T-cell line did not have such capabilities. Overall our findings demonstrate that CD73-expressing exosomes produced by Treg cells following activation contribute to their suppressive activity through the production of adenosine.

Keywords: CD4+ CD25+ Foxp3+ Treg cells

r

CD73

r

Exosomes

r

Immune regulation

r

T cells

Additional supporting information may be found in the online version of this article at the publisher’s web-site

Introduction CD4+ CD25+ Foxp3+ Treg cells control immune responses and maintain immunological tolerance [1]. These cells regulate immune responses via several mechanisms including suppressive

Correspondence: Dr. Lesley Ann Smyth e-mail: [email protected]

cytokine production (IL-10, IL-35, and TGF-β), modification of dendritic cell (DC) function (downregulation of co-receptors CD80/86), cytolysis of target cells (mediated by granzyme A, B, and perforin-dependent mechanisms), cytokine deprivation (through sequestering of IL-2 by CD25), and the intercellular

∗

These authors share senior co-authorship.

C 2013 The Authors. European Journal of Immunology published by Wiley-VCH Verlag GmbH & Co. KGaA

Weinheim. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

www.eji-journal.eu

Eur. J. Immunol. 2013. 43: 2430–2440

transfer of cAMP [2, 3]. Another described suppressive mechanism is metabolic disruption through the production of anti-inflammatory adenosine [4, 5]. Kobie et al. [5] showed that Treg cells express high levels of the cell surface enzyme ecto-5-nucleotide CD73 and this molecule converts extracellular adenosine-5-monophosphate (AMP) to adenosine. The ectoenzyme CD39, also found on Treg cells, provides the substrate for CD73, hydrolyzing extracellular adenosine-5-triphosphate (ATP) into the nucleotide AMP [4]. Adenosine binds to four G-protein-coupled adenosine receptors, A1 , A2a , A2b , and A3 receptors (R), all of which are expressed broadly by many cells including T cells, B cells, and DCs. T cells express both the high-affinity A2a R and the low-affinity A2b R [6] and following cell activation A2a R levels become elevated [7, 8]. Genetic depletion of the A2a R in mice resulted in exaggerated tissue damage highlighting the importance of this pathway in immune modulation [9]. In fact, adenosine produced by Treg cells has been shown to inhibit effector T-cell function through the interaction with the A2a R [4, 10]. Following the binding of adenosine to the A2a R, expressed on activated effector T cells, the intracellular levels of cAMP are increased leading to the inhibition of cytokine production, including IL-2 and IFN-γ [7, 10]. In addition, this interaction also leads to production of adaptive Treg cells through the inhibition of IL-6 and production of TGF-β [11]. TGF-β has recently been shown to play, in turn, a key role in controlling the expression of CD73 and the generation of adenosine by both adaptive Treg cells and DCs [12]. Exosomes are small, secreted membrane vesicles (50–100 nM in diameter) of endocytic origin that are formed by the fusion of multivesicular endosomes with the plasma membrane [13–17]. They are thought to play an important role in intercellular communication and are produced by many different cell types including CD4+ and CD8+ T cells [18–22]. Recently, Clayton et al. [23] described that exosomes derived from human cancer cells contribute to extracellular adenosine production through the expression of CD39 and CD73, leading to the inhibition of human T-cell proliferation and cytokine function. These authors speculated that the release of exosomes expressing CD39 and CD73 within the local environment increases the surface area by which these membrane-associated enzymes can function [23]. The discovery that activated T cells produced exosomes with immunemodulating functions [19–21, 24] raised the question of whether this applies to CD4+ CD25+ Foxp3+ Treg cells. As CD39 and CD73 are highly expressed on Treg cells it could be envisaged that a similar mechanism of immune modulation akin to cancer-derived exosomes, described above, could contribute to their suppressive function. Our study demonstrated that Treg cells indeed release “saucer-shaped” lipidic bilayer vesicles characteristic of exosomes as determined by size, 100 nm in diameter, and expression of CD81 and LAMP-1/CD63, following TCR activation. Treg cell derived exosomes display immune-modulating properties in vitro, which was attributed to adenosine production by CD73 detected on these microvesicles. Our data highlight an additional and novel mechanism by which Treg cells mediate immune suppression. C 2013 The Authors. European Journal of Immunology published by

Wiley-VCH Verlag GmbH & Co. KGaA Weinheim.

Immunomodulation

Results Foxp3 expressing CD4+ CD25+ T cells produce “exosome-like structures” following TCR activation Jurkat T cells, human CD4+ T cells as well as murine CD4+ and CD8+ T cells release extracellular vesicles following TCR engagement [18–21]. These observations raised the question of whether Treg cells also release similar structures following TCR activation. Foxp3 expressing CD4+ CD25+ T cells isolated from B6 mice (Fig. 1A) were stimulated with plate-bound antibodies to CD3ε and CD28. After 24 h, cell culture media was collected and cells, together with cell debris, were removed via centrifugation and filtration, as described in “Materials and methods” section. Extracellular vesicles were then isolated using ultracentrifugation. “Saucer-shaped” lipidic bilayer particles of 100 nm in diameter were observed using EM (Fig. 1B and Supporting Information Fig. 1). This morphology and size has previously been described for exosomes [18], suggesting that murine CD4+ CD25+ FoxP3+ cells release these extracellular vesicles following activation. As relatively few Treg cells could be isolated from the spleens of mice, to further characterize Treg-specific exosomes a murine Treg-cell line with self-specificity (Auto-Treg cells), well characterized in our laboratory, was used [25, 26]. This Treg-cell line expresses classical T-cell markers such as CD4, CD3, and CD2 as well as high levels of CD25, Foxp3, CTLA-4, CD39, and CD73 (Fig. 2A, black histogram lines). In addition, these cells also expressed high levels of the tetraspanin and T-cell co-stimulatory molecule, CD81 [27]. Functionally, this Treg-cell line was capable of suppressing a polyclonal CD4+ T-cell response (Fig. 2B, p < 0.001). Like freshly isolated CD4+ CD25+ T cells, the Auto-Treg line produced 100 nm “saucer-shaped” exosomes following TCR activation (Fig. 3A and Supporting Information Fig. 2). To characterize these further, we attached them to 4 μm beads as previously described [18] and stained with various T-cell-specific antibodies. Treg cell derived exosomes expressed the tetraspanin molecules CD81 and CD63 commonly found on exosomes [28] (Fig. 3B). As exosomes have the same membrane orientation as the plasma membrane of the cell from which they are derived [13, 29], we also stained Treg-cell exosomes with antibodies to CD4+ T-cell plasma membrane molecules, including CD4, CD2, and MHC class I. All of these proteins were detected on Treg cell derived exosomes (Fig. 3B). Additionally, Treg cells also express several cell surface and intracellular receptors capable of mediating suppression including CD25, CTLA-4, and CD73 [4, 30–32] (Fig. 2A). Analysis of the expression levels of these molecules on exosomes derived from the Auto-Treg line revealed that CD73 and CD25 were present at intermediate/highly levels, respectively, while CTLA-4 was detected at very low/negligible levels (Fig. 3B). As described in the Materials and methods, the Auto-Treg line was maintained in the presence of irradiated BM-DCs for 7 days. During this period the majority of BM-DCs underwent cell death (Supporting Information Fig. 3A). To exclude the possibility www.eji-journal.eu

2431

2432


Eur. J. Immunol. 2013. 43: 2430–2440

Figure 1. Freshly isolated CD4+ CD25+ T cells produce exosomes following TCR activation. CD4+ CD25+ T cells were isolated from the spleen and LNs of B6 mice. (A) Cells were assessed for intracellular expression of Foxp3 and cell surface expression of CD4 and CD25 by flow cytometry. Following gating on live cells (FSC versus SSC), CD4 and CD25 cell surface expression levels are represented as a dot plot and compared with that of isotype controls (n = 2). Foxp3 expression on CD4+ CD25+ cells is represented as a histogram (black line). Control isotype antibody expression on CD4+ CD25+ cells is represented as a histogram (gray line; n = 2). (B) Exosomes from activated CD4+ CD25+ T cells were isolated and visualized using EM (original magnification ×280 000, scale bar 100 nm) as described in the Materials and methods. An arrow indicates an individual exosome. Data are representative of two independent experiments performed.

that some of the exosomes identified were actually derived from remaining DCs, which are known to constitutively produce exosomes [33], the presence of MHC class II on exosomes derived from Treg cells was analyzed. It has previously been shown that activated CD4+ T cells lack expression of MHC class II [34] unlike BM-DCs, which expressed high levels of MHC class II (Supporting Information Fig. 3B). No MHC class II was evident on the Treg cell derived exosomes thereby excluding any contribution of DCsderived exosomes in our preparations (Supporting Information Fig. 3C). Taken together, the data suggest that exosomes are produced by Treg cells following TCR activation that carry molecules with regulatory function, such as CD25 and CD73.

dependent, with exosomes derived from 8 × 107 , 5 × 107 , and 3 × 107 Treg cells inhibiting CD4+ T-cell proliferation by 50, 30, and 25%, respectively (Fig. 4B). In addition, cytokine production was also affected; both IL-2 and IFN-γ production by CD4+ T cells were significantly reduced in the presence of Treg cell derived exosomes (Fig. 4C and D; p < 0.05), however, IL-2 production was only affected in the presence of high exosome concentrations (data not shown). In summary, we observed that exosomes derived from Auto-Treg cells inhibit polyclonal CD4+ T-cell proliferation and cytokine production in a dose-dependent manner.

Suppression by Treg cell derived exosomes is not due to the expression of CTLA-4 Suppression of CD4+ CD25− T-cell proliferation in the presence of Treg cell derived exosomes Having established that exosomes were isolated from activated Treg cells, the possibility that these particles modulated CD4+ T-cell responses in vitro was investigated. Exosomes, derived from 5 × 107 Treg cells, were isolated and added to a co-culture of CD4+ T cells (105 cells) and autologous B6 APCs (consisting of >95% B cells, Supporting Information Fig. 3D) in the presence of antiCD3ε antibody. When Treg cell derived exosomes were added, CD4+ T-cell proliferation was reduced as compared with that of CD4+ T cells stimulated in their absence (Fig. 4A). This inhibition was statistically significant (p < 0.05). Titrating the amount of exosomes added revealed that their suppressive capacity was dose C 2013 The Authors. European Journal of Immunology published by


To investigate the mechanisms behind the immune-modulatory effect of Treg cell derived exosomes, we analyzed the contribution of regulatory molecules present on these structures to their function (Fig. 2A). CTLA-4 is a key molecule involved in Tregcell suppression [30], however, CTLA-4 was detected at very low levels on exosomes derived from the Auto-Treg line (Fig. 3B) suggesting that this molecule was not involved. To further test whether CTLA-4 expression was contributing to the suppressive nature of exosomes, an anti-CTLA-4-neutralizing antibody was added to the co-culture. As expected, the suppressive capacity of Treg cell derived exosomes was unaffected by the addition of this antibody suggesting that the level of CTLA-4 detected was too low to have any significant regulatory effect (Fig. 5). In contrast, the www.eji-journal.eu

Eur. J. Immunol. 2013. 43: 2430–2440

Immunomodulation

Figure 2. In vitro expanded Treg cells express Foxp3, CD25, CTLA-4, CD39, and CD73 and are suppressive. (A) Auto-specific Treg cells and a NS CD4+ T-cell line were stained with antibodies to the following T-cell specific markers, CD2, CD3, CD81, CD4, CD25, CD73, CD39, CTLA-4, and Foxp3 using fluorescently labeled antibodies and appropriate isotype controls. Representative histogram staining profiles are shown indicating the expression of these markers on live cells (FSC and SSC gate). Black and dark gray lines represent the Auto-Treg-cell and NS CD4+ T-cell lines, respectively, while the light gray lines represent control isotype-specific antibodies staining. The data are representative of more than three individual experiments. (B) 1 × 105 Auto-Treg cells or NS CD4+ T-cell line were added to autologous CD4+ T and APCs in the presence of anti-CD3ε antibody. Control wells contained anti-CD3ε antibody stimulated CD4+ T cells co-cultured with APCs, as well as, NS CD4+ T cells or Auto-Treg cells cultured alone. 3 H-thymidine incorporation was measured as counts per minute (CPM) after 72 h stimulation and shown as mean +SD of triplicate conditions from one experiment representative of five performed. ***p < 0.001, unpaired t-test is shown. No significant difference is denoted as NS.

presence of the same antibody in culture significantly inhibited (p < 0.05) the ability of Treg cells to regulate polyclonal CD4+ T-cell responses (Fig. 5). These data suggest that the suppressive nature of the Treg cell derived exosomes is not mediated through CTLA-4.

CD73 on Treg cell derived exosomes may facilitate suppression through adenosine production CD73 carried on exosomes derived from cancer cells has been shown to modulate T-cell responses [23]. In the light of this observation, given that CD73 was detected on exosomes derived from Auto-Treg line, (Fig. 3), the contribution of this molecule to exosome immune modulation was investigated. First, the regulatory function of CD73-expressing exosomes derived from the Auto-Treg line was compared to exosomes derived from a CD73-negative CD4+ T-cell line. Although these cells expressed CD2, CD3, CD39, and CD25 molecules at similar or higher levels compared with the Auto-Treg line, they lacked expression of Foxp3 and CD81, and had lower CTLA-4 expression levels com C 2013 The Authors. European Journal of Immunology published by


pared to the Treg-cell line (Fig. 2A, dark gray histogram lines). In addition, they lacked suppressive capacity (Fig. 2B). Although the nonsuppressive (NS) CD4+ T-cell line produced exosomes following TCR activation, carrying CD2 and CD25 at similar levels to that detected on exosomes derived from the Auto-Treg line (Fig. 6A and Supporting Information Fig. 4), they lacked CD73 and did not inhibit a polyclonal T-cell response (Fig. 6A and B). Clayton et al. [23] have shown that CD39 and CD73, expressed by cancer exosomes, are involved in ATP and 5 -AMP-hydrolytic activity, respectively. As our Auto-Treg cells expressed CD73, we addressed whether expression of this molecule on both Treg cells and exosomes leads to 5 -AMP-hydrolytic activity and adenosine production. The ability of CD73 to produce inorganic phosphate from exogenous 5 -AMP substrate using a colorimetric malachite green assay was measured in the absence or presence of the CD73 inhibitor 5 -α,β-methylene ADP (APCP). In addition, and as a control, inorganic phosphate production in the presence of CD73-negative cells was measured. As expected, compared with cells cultured in the absence of 5 -AMP, phosphate production was observed in the presence of CD73-expressing Auto-Treg cells but not in the presence of the CD73-negative NS CD4+ T-cell www.eji-journal.eu

2433

2434


Eur. J. Immunol. 2013. 43: 2430–2440

Figure 3. Auto-Treg cells produce exosomes with detectable CD81, CD63, as well as CD73 expression following TCR activation. Auto-Treg cells were stimulated with anti-CD3ε and anti-CD28 antibodies. After 24 h, cell culture media was collected, filtered, and ultracentrifugation. Exosomes structures were detected using (A) EM (original magnification ×183 000, scale bar 0.2 μM) and (B) flow cytometry analysis. (B) Isolated exosomes were stained with antibodies to CD81, LAMP-1/CD63, CD4, CD2, MHC class I (H-2Kb ), CTLA-4, CD73, and CD25 (black lined histogram) as outlined in the Materials and methods. Controls include unstained beads (gray-filled histogram) and exosome-coated FACs beads stained with isotype control antibodies (gray-lined histogram). The geometric mean fluorescence intensity is indicated within individual histograms. Data are representative staining profiles of three individual experiments.

line following incubation with 50 μM of 5 -AMP. Phosphate production by CD73-expressing Treg cells was significantly inhibited following addition of the CD73 inhibitor APCP (Fig. 7A). Even more relevant, phosphate production was also observed in the presence of CD73 carrying Treg cell derived exosomes, which was also reduced in the presence of APCP (Fig. 7A). The capacity to generate adenosine from its AMP substrate was measured using a PathHunter ADORA2B assay, which has previously been used to measure adenosine production by murine CD4+ T cells [12]. As expected, the conversion of 5 AMP to adenosine was evident in the presence of CD73-expressing Auto-Treg cells as well as Treg cell derived exosomes (Fig. 7B, p < 0.01 as compared with media plus AMP). This was significantly inhibited following addition of APCP (Fig. 7B, p < 0.001) and in the presence of adenosine deaminase (Supporting Information Fig. 5). No adenosine production was observed using the NS CD4+ T-cell line reflecting the lack of CD73 expression by these cells. These observations suggest that CD73 present on Treg cell derived exosomes may play an important role in immune modulation through the production of adenosine. C 2013 The Authors. European Journal of Immunology published by


Discussion In this study, we demonstrate that Treg cells release exosomes following TCR activation. Exosomes were characterized morphologically, via EM, and by the expression of the tetraspanins LAMP-1/CD63 and CD81. Importantly, Treg cell derived exosomes displayed immunomodulatory properties. Although exosome release is a potential suppressive Treg mechanism, it is clearly not the main one. This is highlighted by the observation that exosomes from a large number of Treg cells (8 × 106 ) suppress 50% less CD4+ proliferation than 105 Treg cells when added to 105 CD4+ responder T cells. However, it is conceivable that upon recognition of antigen on an APC, the release of exosomes from Treg cells allows the dissemination of molecules/receptors capable of affecting the local microenvironment, and along with other mechanisms, maybe part of the overall suppressive process. Whether this represents a mechanism used by both adaptive and natural Treg cells needs to be further evaluated. Nolte-’t Hoen et al. [35] have shown that exosomes derived from anergic rat T cells inhibit effector T-cell functions following www.eji-journal.eu

Eur. J. Immunol. 2013. 43: 2430–2440

Immunomodulation

Figure 4. Exosomes derived from Auto-Treg lines have suppressive capacity. (A and B) Exosomes derived from (A) 5 × 107 Treg cells (TregEXO ) or (B) from 3 × 107 , 5 × 107 , or 8 × 107 Treg cells were added to autologous CD4+ T cells co-cultured with APCs in the presence of anti-CD3ε antibody. Control wells contained anti-CD3ε antibody stimulated CD4+ T cells cocultured with APCs in the presence or absence of Treg cells, Treg cells co-cultured with APCs, and Treg cells and CD4+ T cells cultured alone. (A) T-cell proliferation is expressed as 3 H-thymidine incorporation and shown as mean +SD of triplicate conditions from one experiment representative of four performed. (B) Each symbol represents an individual sample and the percent suppression data are from one experiment representative of four performed. (C and D) IL-2 and IFN-γ were measured using specific ELISAs. Data are shown as mean +SD of triplicate conditions from one experiment representative of four performed. *p < 0.05, **p < 0.01, and ***p < 0.001, unpaired t-test. No significant difference is denoted as NS.

coculture with B cells and DCs in vitro. Interestingly, these T-cellderived exosomes had high levels of CD25 compared with control T cells and the authors suggested that exosomes expressing CD25 molecules, by binding to the surface of an APC, bestows that cell with the ability to bind free IL-2 in the local environment leading to depletion of available cytokines and apoptosis of effector T cells. Although CD25 molecules were present on exosomes derived from Treg cells, they were also expressed by exosomes derived from a NS T-cell line suggesting a redundant role for CD25 molecules on exosomes in our system. In addition, we demonstrate that CTLA-4 did not contribute in exosome-mediated suppression, despite the key role of this molecule in Treg-cell function. C 2013 The Authors. European Journal of Immunology published by


CD73 present on Treg cell derived exosomes represents one potential immune-suppressive mechanism. Expression of both CD39 and CD73 molecules has been shown on natural and induced Treg cells and together they contribute to immune suppression through the production of anti-inflammatory mediator adenosine [4]. Recently, CD39 and CD73 expression by cancer-derived exosomes was associated, through adenosine production, with impaired T-cell function, proliferation, and cytokine (IL-2 and IFN-γ) release [23]. It could be envisaged that a similar mechanism might occur following activation of Treg cells in vivo. Indeed, we observed that CD73 on Treg cell derived exosomes hydrolyses exogenous 5 -AMP leading to the production of adenosine. www.eji-journal.eu

2435

2436


Eur. J. Immunol. 2013. 43: 2430–2440

supporting the idea that exosomes derived from Foxp3 negative CD4+ T cells inhibit immune responses [19, 21]. The lack of suppression by exosomes derived from the NS CD4+ T-cell line used in this study may reflect the fact that, in the aforementioned studies, freshly isolated TCR transgenic antigen-specific CD4+ T cells, rather than an autologous CD4+ T-cell line, were used. In summary, the results presented in our study suggest that the release of immune-modulatory exosomes by Auto-Treg cells following TCR activation is an additional and novel mechanism by which these cells induced suppression. It is conceivable that the release of CD73-expressing exosomes by Treg cells may lead to infectious tolerance whereby the production of adenosine leads to modulation of effector cell or antigen presenting cell function, as well as the induction of adaptive Treg cells. Although T-cellderived exosomes have been shown to function in vivo whether Treg cell derived exosomes do so is still unknown and warrants further investigation.

Figure 5. Suppressive capacity of Treg cell derived exosomes is not due to CTLA-4. Anti-CTLA-4 antibody was added to cocultures containing autologous B6 CD4+ T cells and APCs in the presence or absence of Auto Treg cell derived exosomes (TregEXO , isolated from 5 × 107 cells) or AutoTreg cells, plus anti-CD3ε antibody). Control wells lacked this antibody. 3 H-thymidine incorporation is shown as mean +SD of triplicate conditions from one experiment representative of two performed. *p < 0.05, ***p < 0.001, unpaired t-test. No significant difference is denoted as NS.

Adenosine production in the extracellular environment has been shown to play a key role in protecting cells from immune damage [9]. Binding of this molecule to adenosine receptors A2a R, expressed by activated effector T cells triggers intracellular cAMP production leading to the inhibition of cytokine production, thereby limiting T-cell responses [10]. This might explain the reduced proliferation and cytokine production of CD4+ T cells seen in the presence of Treg-specific exosomes. Therefore, the release of CD73 bearing exosomes from Treg cells during activation within the local environment may help increase the surface area by which this membrane-associated enzyme functions, leading to increased adenosine production and consequently suppression of effector T-cell responses as well as increased adaptive Treg-cell numbers [11]. Several regulatory mechanisms have been described for exosomes derived by conventional CD4+ and CD8+ T cells, including downregulation or masking of MHC:peptide molecules on the surface of APCs, induction of APC apoptosis through the Fas/FasL pathway and delivery of miRNA with gene regulation properties [19–21, 24]. It is feasible that one or more of these mechanisms also contributes to the suppressive nature of Treg cell derived exosomes together with CD73 and warrants future investigation. Lastly, it is conceivable that exosomes derived from Treg cells and Foxp3 negative CD4+ T cells may have different immunological functional purposes. In support of this idea recently, Wahlgren et al. [36] showed that exosomes derived from human T cells in the presence of IL-2 increased proliferation of autologous resting T cells. However, at the same time, there are other studies C 2013 The Authors. European Journal of Immunology published by


Materials and methods Mice C57BL/6 ((B6) H-2b ) mice were purchased from Harlan Olac Ltd. (Bicester, UK). Mice were maintained under specific pathogen free conditions, mouse handling and experimental procedures were conducted in accordance with national and institutional guidelines for animal care and use.

Treg-cell and NS CD4+ T-cell line maintenance Cell cultures were performed in complete media consisting of RPMI 1640 medium (Invitrogen, Paisley, UK) supplemented with 100 IU/mL penicillin, 100 μg/mL streptomycin, 2 mM L-glutamine, 0.01 M Hepes, 50 μM 2β-mercaptoethanol (Invitrogen), and 10% heat-inactivated fetal calf serum (FCS; SERAQ, Sussex, UK). Cells were maintained at 37◦ C in a humidified atmosphere with 5% CO2 . CD4+ CD25+ Foxp3+ T-cell lines and NS CD4+ T cells were maintained in the aforementioned culture media and were stimulated once per week with BM-DCs. To makes BM-DCs, BM derived from B6 mice was passed through a 70 μm cell strainer to obtain a single-cell suspension and erythrocytes were lysed using ACK buffer (0.15 M NH4 Cl/1 mM KHCO3 /0.1 mM Na2 -EDTA). BM cells were then incubated for 30 min, at 4◦ C, with supernatants from the following hybridoma cultures: YTS 191 (anti-CD4, M5/114 (anticlass II), RA3–3A1 (anti-B220), and YTS 169 (anti-CD8)). R The cells were washed twice then incubated with Dynabeads ◦ for 30 min (4 C) followed by magnet separation. After washing, BM cells were seeded in 24-well plates at 1 × 106 cells/well in complete medium supplemented with 20 ng/mL of murine recombinant GM-CSF (kind gift from GlaxoSmithKline R&D, UK). Media was changed on days 2 and 4 with fresh GM-CSF containing media. www.eji-journal.eu

Eur. J. Immunol. 2013. 43: 2430–2440

Immunomodulation

Figure 6. NS CD4+ T cells produce exosomes that lack immune modulation capacity. Exosomes isolated from both the Auto-Treg-cell and the NS CD4+ T-cell lines following TCR stimulation were visualized using (A) EM (original magnification ×224 000, scale bar 100 nm, NS CD4+ T-cell line only) and flow cytometry. Representative staining of CD2, CD25, and CD73 expression is shown for exosome-coated beads for NS CD4+ (CD4+ EXO , light gray lines) and Auto-Treg cells (TregEXO , black lines) as well as uncoated beads (filled gray lines; n = 3). (B) Exosomes derived from either 5 × 107 Auto-Treg (TregEXO ) or NS CD4+ T cells (CD4+ EXO ) were added to autologous CD4+ T cells and APCs in the presence of anti-CD3ε antibody. Control wells contained CD4+ T cells co-cultured with APCs plus anti-CD3ε antibody in the absence of exosomes as well as CD4+ T cells stimulated, or not with APCs, in the absence of anti-CD3ε antibody. Data are shown as mean +SD of triplicate conditions and are from one experiment representative of three performed. ***p < 0.001, unpaired t-test. No significant difference is denoted as NS.

The purity of CD11c+ BM-DCs was measured by flow cytometry analysis and was found to be greater than 90% (data not shown). Auto-Treg cells and NS CD4+ T cells were cultured in 24-well plates with irradiated B6 BM-DCs at a ratio of T:DCs of 4:1. IL-2 (10 IU/mL) supplemented complete media was added on days 1, 3, and 5 of culture.

RBC-depleted splenocytes from B6 mice were isolated and CD4+ 25+ T cells purified using a murine CD4+ CD25+ isolation kit (Invitrogen) following manufactures instructions.

at 100 000 × g (Sorvall Discovery 100 with a T890 rota) followed by filtration using a 0.22-μm filter (Millipore, Bedford, MA, USA). Exosomes were isolated as previously described [18]. To summarize, cells were cultured overnight in exosome-free media in the presence of plate-bound anti-CD3ε (5 μg/mL, clone 145.2C11, BD Bioscience, San Jose, USA) and anti-CD28 (5 μg/mL, clone 61109, R+D Systems, Minneapolis, USA). Cell culture media was collected and cells, plus cell debris, were removed via centrifugation (1800 rpm/7 min) and filtration using a 0.22 μm filter (Millipore), respectively. Exosomes were recovered after ultracentrifugation at 100 000 × g for 1 h at 4◦ C followed by washing with PBS supplemented with protease inhibitors (Sigma Aldrich, Dorset, UK).

Preparation of exosomes

Electron microscopic (EM) analysis

To deplete FCS of exosomes and other microvesicles, undiluted FCS was subjected to ultracentrifugation overnight (at least 15 h)

For EM observation, exosomes were prepared as described earlier and fixed in phosphate buffer containing 4% paraformaldehyde

Isolation of CD4+ CD25+ Foxp3+

C 2013 The Authors. European Journal of Immunology published by


www.eji-journal.eu

2437

2438


Eur. J. Immunol. 2013. 43: 2430–2440

Figure 7. CD73 present on Treg-cell exosomes hydrolyze exogenous 5 -AMP to adenosine. (A)105 Auto-Treg cells (Treg), NS CD4+ T cells (CD4+ ), and exosomes (TregEXO ) isolated from 3–4 × 107 Auto-Treg cells, were incubated alone (dotted bars) or in the presence of 50 μM 5 -AMP in the presence (black bars) or absence (white bars) of 100 μM APCP. The copper-based colorimetric reagent was added and absorbance measured at 630 nm. Values were compared with a standard curve to quantify phosphate levels. Data represent the amount of phosphate produced over background (cells or exosomes incubated in the absence of AMP) and are shown as mean +SEM of four samples pooled from four independent experiments. *p < 0.05, unpaired t-test. No significant difference is denoted as NS. (B) Supernatant levels of adenosine generated from AMP were measured using the DiscoveRx PathHunter β-Arrestin assay with the ADORA2B receptor-transfected reporter cell line. Levels of adenosine generated by 2 × 105 Auto-Treg cells (Treg), NS CD4+ T cells (CD4+ ), and CD73-expressing Treg cell derived exosomes (TregEXO ) isolated from 2 × 107 Auto-Treg cells were incubated alone (dotted bars) in the presence of 5 mM AMP (white bars) or in the presence of AMP and 1 mM APCP (black bars). Control wells contained media alone, media plus AMP and media plus AMP and APCP. Values were compared with a standard adenosine curve. Assays were performed in triplicate. Data are shown as mean +SEM of nine samples pooled from three independent experiments. Statistical significance was assessed using one-way ANOVA and each “group + AMP” was compared by Dunnett’s multiple comparison posttest to the reference “medium + AMP” background. An unpaired Student’s t-test was used to compare adenosine production by the Auto-Treg cells, NS CD4+ T cells, and CD73expressing Treg cell derived exosomes in the presence, or in the absence of the inhibitor APCP. *p < 0.05, **p < 0.01, and ***p < 0.001, whereas nonsignificant p-values are labeled “ns”.

before being loaded onto carbon-coated Formwar EM grids. After incubation for 20 min, the samples were washed twice in PBS then fixed for 5 min in 1% glutaraldehyde. After washing, the grids were stained for 10 min with saturated aqueous uranyl before viewing on a Tecnai T12 BioTWIN electron microscope (FEI, the Netherlands).

were analyzed on a BD FACSCalibur machine and subsequent analysis was performed with FlowJo software (Ashland, OR, USA). For analysis of expression of cell surface markers, data are shown for live cells and beads as defined by an forward scatter versus side scatter gate (Supporting Information Fig. 6).

MTT cell survival assays Phenotypic analysis by flow cytometry Fluorochrome-conjugated (fluorescein isothiocyanate, FITC; phycoerythrin, PE; allophycocyanin) mAbs against the following mouse cell-surface antigens, CD4, CD25, CD2, MHC class I, MHC class II, CD39, and CD73 were purchased from eBioscience (San Diego, CA, USA) with their relevant isotype controls. CD63 was purchased from The Medical and Biological Laboratories Company, Japan and used following manufactures instructions. For flow cytometry analysis, 1 × 105 cells were labeled with fluorochrome-conjugated mAbs for 20 min (4◦ C), washed twice in FACS buffer (PBS/2% FCS/0.1% EDTA), and analyzed on a TM TM FACSCalibur , using Cell Quest software (Becton Dickinson, Mountain View, CA, USA). Foxp3 staining was performed using a murine Foxp3 kit according to the manufacturer’s protocol (eBioscience). For flow cytometry analysis of exosomes, isolated exosomes were coated onto aldehyde/sulfate latex 4%w/v 4 μm microbeads (Invitrogen) as previously described [18]. Cells and exosomes C 2013 The Authors. European Journal of Immunology published by


In total, 1 × 105 BM-DCs either exposed to 3 000 Gy of irradiation 1 or 6 days previously or unirradiated DC controls were added to each well of a 96-well plate in 200 μL of complete media lacking phenol red. Each condition was tested in triplicate. MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay was performed following manufactures instructions (Invitrogen) at 37◦ C/5% CO2 .

In vitro suppression assay A single-cell suspension was obtained by passing spleens and pooled lymph nodes through a 70 μm cell strainer (BD Pharmingen, CA, USA). Erythrocytes were lysed and CD4+ cells were isolated by negative selection using a CD4 Dynabeads isolation kit (Dynal, Wirral, UK). The purity of the selected CD4+ population was >90% as assessed by flow cytometry (data not shown). Antigen presenting cells were purified following incubation with www.eji-journal.eu

Eur. J. Immunol. 2013. 43: 2430–2440

anti-CD4 and anti-CD8 antibodies, YTS-191 and YTS-169, respectively. T-depleted cells were isolated by negative selection using anti-mouse IgG-coated beads. These cells consist of >90% B cells (Supporting Information 3D). Suppression assays were set up as follows: 1 × 105 T cells and 2 × 105 APCs were co-cultured in the presence or absence of 1 μg/mL of anti-CD3ε antibody (eBioscience, UK). In some wells, 1 × 105 Treg cells or exosomes isolated from defined Treg numbers were added. In some experiments, anti-CTLA-4 antibody (100 μg/mL) was added to the co-cultures. On day 2 of culture, cells were pulsed with 1 μCi/well 3 H thymidine (Amersham Pharmacia, UK). Proliferation was measured by 3 H thymidine incorporation after 20 h by liquid scintillation counting using a beta plate counter.

5 -AMP hydrolysis assay The malachite green colorimetric assay was used to quantify the production of inorganic phosphate from 5 -AMP substrate (Sigma) using the SensoLyte kit (AnaSpec, Fremont, CA, USA). Exosomes were prepared as above, however, extracellular vesicles were washed and resuspended in phosphate-free buffer [37]. 5 -AMP diluted in phosphate-free buffer to give a final concentration of 50 μM before being added to extracellular vesicle and either 105 suppressive or NS cells for 30 min at room temperature. Controls included cells and exosomes incubated in phosphate-free buffer in the absence of AMP. The copper-based colorimetric reagent was added and absorbance measured at 630 nm after 5-min gentle shaking. Values were compared with a standard curve to quantify phosphate levels. The CD73 inhibitor APCP (Sigma Aldrich, 100 μM) was used in some assays.

Immunomodulation

Statistical analysis To determine statistical significance, we carried out one-way ANOVA test with Dunnett’s multiple comparison posttest or Student’s t-test (unpaired, two-tailed) using the GraphPad Prism software, http://www.graphpad.com/prism/prism.htm. In the figures, p-values