Blood Cells, Molecules, and Diseases 35 (2005) 11 – 16 www.elsevier.com/locate/ybcmd
Phenotypic and functional characterization of intestinal epithelial exosomes J. Mallegol, G. van Niel, M. Heyman* INSERM EMI 0212, Faculte´ Necker-Enfants Malades, 156 rue de Vaugirard, 75730 Paris, France Submitted 31 March 2005 Available online 11 May 2005 (Communicated by M. Lichtman, M.D., 4 April 2005)
Abstract Intestinal epithelial cells (IEC) are located at a strategic position between the external environment and the most extended lymphoid tissue in the body. Besides their central role in the absorption of nutrients, IEC also provide antigenic information to the immune system and are involved in the balance tolerance/allergy to food antigens. Like professional antigen presenting cells, IEC have been shown to secrete 30- to 90-nm diameter vesicles named exosomes, in a polarized way, either from their apical or basolateral side. These vesicles carry molecules involved in adhesion and antigen presentation, comprising major histocompatibility complex (MHC) class I and class II molecules, tetraspan proteins, CD26/dipeptidyl-peptidase IV, and A33 antigen, a molecule essentially restricted to the intestinal epithelium. Invariant chain, transferrin receptor, and Na-K-ATPase are not expressed on epithelial exosomes. In vivo, in mice, epithelial exosomes carrying MHC/ ovalbumin peptide complexes induce specific immune responses when injected intraperitoneally. A33 antigen, an Ig-like molecule highly specific for intestinal epithelial cells and enriched in epithelial exosomes, is found at the surface of cells entering mesenteric lymph nodes suggesting exosome migration from the epithelial layer to the gut associated lymphoid system. Taken together, intestinal epithelial exosomes released at the basolateral surface of enterocytes could be antigen-carrying structures constituting a link between luminal antigens and the local immune system and acting as sensors of the antigenic information present in the intestinal lumen. D 2005 Elsevier Inc. All rights reserved. Keywords: Mucosal immunity; Intestinal epithelial cells; Exosomes
Introduction The intestinal epithelial layer represents a large interface between the external milieu (intestinal lumen) and the most important immune system spread along the intestinal mucosa. It is not only involved in the absorption of nutrients and ions but also constitutes an efficient barrier against the penetration of intact food antigens and infectious agents. Although the majority of ingested proteins are degraded by digestive enzymes, a small fraction escapes hydrolysis and can be endocytosed by intestinal epithelial cells (IEC) and transported to the lamina propria. In physiological conditions, the main transport pathway of food antigens across
* Corresponding author. Fax: +33 1 40 61 56 38. E-mail address:
[email protected] (M. Heyman). 1079-9796/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.bcmd.2005.04.001
the digestive epithelium is transcellular [1]. Paracellular diffusion of proteins between epithelial cells is negligible due to the fine control of tight junction permeability preventing diffusion of molecules larger than 600 Da towards the internal compartment [2]. The antigen sampling by IEC is necessary to the induction of oral tolerance to food antigens and the non-responsiveness of the immune system, but it may lead in certain environmental conditions (stress, infection, inflammation) to sensitization of the host and food allergy. Thus, epithelial –lymphoid interactions are probably crucial in the regulation of immune responses. Over the last 10 years, professional antigen presenting cells (APC) have been shown to release major histocompatibility complex (MHC)-bearing vesicles called exosomes, capable of stimulating T cell responses. Exosomes may represent a source of ‘‘pre-processed‘‘ antigens, as they bear MHC II/peptide complexes and can be internalized by
12
J. Mallegol et al. / Blood Cells, Molecules, and Diseases 35 (2005) 11 – 16
immature dendritic cells capable of loading exosomederived peptides for presentation to CD4+ T cells [3]. Exosomes may thus be considered as soluble messengers between cells, disseminating information at distance. The possibility that exosomes released by IEC could be part of the intestinal antigen presenting pathway has been investigated both in vitro using intestinal epithelial cell lines and in vivo in a murine model of tolerance/sensitization to ovalbumin.
Role of intestinal epithelial cells in antigen transport, processing, and presentation Exogenous antigens can be transported across the intestinal epithelium towards the underlying tissues along different pathways (Fig. 1). The epithelium overlying Peyer’s patches contains various amounts of M cells, recognized as a portal of entry of bacteria, viruses, and perhaps food antigens [4]. The absence of basement membrane under the epithelial layer covering Peyer’s patches and the presence of M cells lacking lysosomal system favor the transmission of intact antigenic material to the underlying immune system. A new pathway allowing transmission of luminal antigens or particles has been recently described. It involves the outstanding capacity of dendritic cells to extend dendrites into the intestinal lumen through tight junctions between epithelial cells and to catch bacteria directly within the gut lumen, especially in inflammatory conditions, when chemokines drive DC into
the epithelium [5,6]. Finally, and this pathway should not be neglected, minute amounts of luminal antigens are endocytosed constitutively by the apical membrane of epithelial cells lining the digestive tract and transported to the basal side of the epithelium. The vast majority of the endocytosed material (90%) is degraded during transport in the endosomo-lysosomal system of enterocytes before being released by exocytosis in the lamina propria. Among the degradation products, amino acids account for 50% (total degradation) and peptides for 40%. Interestingly, such peptides have a length compatible with the binding to MHC class II restriction molecules and could thus interfere with immune cells spread in the lamina propria or in more distal compartments [1,7]. Besides their role in the sampling of luminal antigens, IEC, like endothelial cells and fibroblasts, can also play a role of non-professional APC (Fig. 2). IEC express accessory molecules involved in antigen presentation, including MHC class I and class II, invariant chain, CD54, CD58, CD1d, and gp180. In vitro, the antigen presenting capacities of IEC involve two different pathways, depending on the presence or absence of mucosal inflammation and expression of MHC class II molecules with different immunomodulatory properties. Due to the polarized nature of IEC, antigen uptake occurs at the apical surface, whereas antigen presentation is only possible at the basolateral surface [8]. However, in vivo, direct contacts between IEC and CD4+ T cells are very limited due to the presence of a basement membrane, even though T cells and IEC can occasionally develop projections through the pores
Fig. 1. How are mucosal immune cells informed of luminal antigens? The three main transport pathways involved in the transepithelial passage of luminal antigens or macromolecules include (1) capture by M cells overlying Peyer’s patches, (2) direct capture by dendritic cells extending dendrites into the intestinal lumen through tight junctions between enterocytes, and (3) constitutive endocytosis of antigens at the apical membrane of epithelial cells lining the digestive tract followed by partial intracellular hydrolysis, transcytosis to the basal side of the epithelium, and release of intact antigen (<10%), peptides (40%), and amino acids (50%). Peptides generated during epithelial transcytosis could be protected from total degradation by their binding to MHC class II molecules and their secretion via exosomes. The relative importance of such antigen transport pathways remains to be established.
J. Mallegol et al. / Blood Cells, Molecules, and Diseases 35 (2005) 11 – 16
13
Fig. 2. Intestinal epithelial cells (IEC) secrete exosomes. IEC express accessory molecules (MHC class II, invariant chain, HLA-DM) and are considered as non-professional antigen presenting cells. The lack of direct contact between IEC and CD4+ T cells limits direct antigen presentation in vivo. However, IEC secrete exosomes which are small membrane vesicles originating from the MHC class II-enriched compartment (MIIC) and are released by exocytosis of these compartments in the external medium. Such epithelial exosomes bear class II/peptide complexes and molecules potentially involved in cell – cell or cell – matrix interactions (adapted from [9], van Niel et al., 2001).
(0.3 to 3 Am) in this membrane. Some years ago, we raised the hypothesis that IEC may participate to intestinal antigen presentation by delivering exosomes as vehicles of luminal antigens.
Intestinal epithelial cells, like immune cells, can release exosomes We initially showed, using HT29-19A and T84DRB1*0401/CIITA intestinal cell lines, that IEC secrete exosomes (membrane vesicles of 30- to 90-nm diameter) both in basal and inflammatory conditions [9]. This secretion is polarized since exosomes released from the apical and basolateral sides of the enterocytes have distinct molecular composition. When released from the basolateral side of the cell, epithelial exosomes express molecules including membrane proteins such as A33 antigen [10] and the epithelial cell surface antigen [11], which are exclusively found in the basolateral exosomes and could be implicated in cell –cell or cell – matrix interaction. The enrichment of tetraspanins, like CD63, a marker of late endosomal compartment, indicates that exosomes are not plasma membrane fragments but originate from an endosomal compartment, probably the MIIC compartment, where loading of MHC class II molecules with antigenic peptides occurs. In basal conditions, epithelial exosomes express low levels of MHC class I and class II molecules which are upregulated in inflammatory conditions. The external position of MHC molecules on basolaterally released
epithelial exosomes indicates an appropriate orientation for antigen presentation, and the presence of stable MHC class II a/h dimers suggests that exosomes carry functional MHC class II/peptide complexes (Fig. 2). Exosomes released from the apical pole of the enterocytes have a distinct composition. They express syntaxin 3 and microsomal dipeptidase, molecules probably involved in their addressing to the apical side of the cells. CD26, an exopeptidase highly expressed at the apical membrane of enterocytes, is enriched on apically released epithelial exosomes. This is explained by the fact that this molecule undergoes a complex intracellular traffic, since it is first addressed both to the basolateral and apical membranes before basolaterally bound molecules are transcytosed back to the apical surface where they concentrate. In addition, other molecules are present in both types of exosomes, i.e., proteins of the cytoskeleton (actin, tubulin), cytosolic proteins, and enzymes involved in intracellular metabolism or exosome biogenesis (kinases, dehydrogenases, enolase) which may be trapped into exosomes during their formation.
Function of epithelial-derived exosomes? The intestinal epithelium is a relatively static structure in the body, despite its rapid renewal, and epithelial exosomes may constitute a link between luminal antigens and immune cells spread in the lamina propria. The expression of A33 antigen, a receptor-like molecule of the immunoglobulin
14
J. Mallegol et al. / Blood Cells, Molecules, and Diseases 35 (2005) 11 – 16
superfamily, on basolaterally released exosomes has been used to localize epithelial exosome migration once released by IEC. In mice, immunohistochemistry indicated A33 antigen staining at the external membrane of immune cells entering mesenteric lymph nodes, suggesting that epithelial exosomes may reach secondary lymphoid organs bound to immune cells. Indeed, such immune cells do not express A33 antigen whose expression is highly specific to enterocytes [12]. Basolaterally released epithelial exosomes may therefore be taken up by professional APC, as shown for dendritic cell-derived exosomes [13,14]. The latter possibility is also supported by the observation that MHC class II positive B cell-derived exosomes are found attached to the cell surface of follicular dendritic cells [15]. An important question concerns the outcome of antigen presentation via epithelial exosomes. The intestine is generally recognized as tolerogenic as far as soluble food antigens are concerned. It is important to stress that, in the intestinal environment, dendritic cells induce tolerance due to the downregulation of co-stimulatory molecules and induction of regulatory T cells. The concept that exosomes might be tolerogenic is strengthened by studies on liposomes with incorporated MHC II/peptide complexes. In the absence of co-stimulatory molecules, MHC II-bearing liposomes lead to the anergy of CD4+ T lymphocytes, but in the presence of professional APC providing co-stimulation, MHC II/liposomes gain the capacity to activate T cells [16]. It is therefore possible that epithelial exosomes, according to their composition (presence of MHC class II molecules but apparent lack of co-stimulatory molecules), may be
involved in this tolerogenic process. Indeed, exosome-like structures named tolerosomes isolated from rat IEC were shown to induce antigen-specific tolerance when administered intraperitoneally to naive recipient rats [17]. These results were not confirmed by our group, since intestinal epithelial cell (MODE K)-derived exosomes, pre-loaded with ovalbumin peptides, were not able to tolerize mice when administered intraperitoneally. However, exosomes bearing MHC class II/OVA peptide complexes were capable of transmitting immune information since very low amount was capable of stimulating ovalbumin-specific humoral (IgG and IgE) immune response [18] (Fig. 3). This stimulatory effect was observed when exosomes were produced in inflammatory conditions (IFNg), when MHC class II molecules were upregulated on epithelial cells, but not in basal conditions. That epithelial exosomes can induce a specific immune response to OVA despite their apparent lack of costimulatory molecules suggests a need to interact with professional APC expressing these molecules, as previously demonstrated with dendritic cell-derived exosomes [19]. Indeed, our results are in agreement with those reported in mice on dendritic cell-derived exosomes which, after injection, can generate an antigen-specific, MHC class IIrestricted, CD4+ T cell proliferation and an upregulation of activation markers. This activation mainly depends on costimulatory molecules expressed by dendritic cells [19]. The involvement of dendritic cells in antigen presentation of peptide-loaded exosomes was also confirmed recently [14]. Similarly, mast cell-derived exosomes also need dendritic
Fig. 3. Epithelial exosomes display immunostimulatory properties in the presence of IFNg. The murine epithelial cell line MODE K is cultured in the presence of ovalbumin (OVA) T IFNg, and exosomes are purified from culture conditioned media after 48 h. Western blot analysis shows that exosomes prepared in the presence of IFNg (EXO-OVA-IFN) bear high amount of MHC class II molecules (probably MHC II/OVA peptide complexes) as compared to exosomes obtained in basal conditions (EXO-OVA) or cell lysate. Exosomes (10 Ag) are injected intraperitoneally to C3H/HeN mice. The OVA boost shows that EXOOVA-IFN have stimulated OVA-specific humoral immune response (IgG and IgE), whereas exosomes obtained in the absence of IFNg or free OVA peptides at similar concentration fail to stimulate the immune response (adapted from reference [18], van Niel et al., 2003).
J. Mallegol et al. / Blood Cells, Molecules, and Diseases 35 (2005) 11 – 16
cells to efficiently activate T cells in vitro [20], suggesting that MHC class II/peptide complexes are transferred from exosomes to dendritic cells. Thus, epithelial exosomes would transfer luminal antigens from the intestinal lumen to local dendritic cells ‘‘sensing’’ the gut lumen and to mesenteric lymph nodes, the immune response probably depending on the activation status of local DCs. In the intestinal suppressive microenvironment (TGFh), the immune recognition of exosomal MHC II/peptide complexes by mucosal lymphocyte might drive tolerogenic responses, at least in physiological conditions. Whereas basolaterally released exosomes interfere with immune cells in the lamina propria, the role of apically released exosomes is more puzzling [21]. One possibility could be that apical exosomes serve a pathway of recycling or release of infective viral particles from the IEC back to the intestinal lumen. Such a possibility does exist, since EBV-encoded latent membrane protein 1 found on EBVpositive cell lines is released in exosomes [22], and most viruses are released from the apical side of epithelial cells [23]. This hypothesis is sustained by the presence of syntaxin 3, a molecule involved in the delivery or secretion of proteins to the apical surface of epithelial cells on apical exosomes. In addition, a recent report indicates that HIV budding from infected macrophages supports the Trojan exosome hypothesis predicting that retroviral budding may represent the exploitation of a pre-existing pathway of intercellular vesicle trafficking, namely exosomes [24]. In conclusion, although epithelial cells release exosomes bearing luminal antigen-derived peptides, the relative significance of such pathway in the transepithelial transport of luminal antigens and the information of the immune system is not known. Future studies will have to delineate the actual role of epithelial exosomes in the modulation of mucosal immune response.
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
Acknowledgments [14]
This paper is based on a presentation at a Focused Workshop sponsored by The Leukemia and Lymphoma Society on ‘‘Exosomes: Biological Significance’’ held in Montreal, Canada, May 20– 21, 2005. The authors also thank the Nutricia Research Foundation for its financial support.
[15]
[16]
References [17] [1] M. Heyman, R. Ducroc, J.F. Desjeux, J.L. Morgat, Horseradish peroxidase transport across adult rabbit jejunum in vitro, Am. J. Physiol. 242 (1982) G558 – G564. [2] C.J. Watson, M. Rowland, G. Warhurst, Functional modeling of tight junctions in intestinal cell monolayers using polyethylene glycol oligomers, Am. J. Physiol.: Cell Physiol. 281 (2001) C388 – C397. [3] A.E. Morelli, A.T. Larregina, W.J. Shufesky, M.L. Sullivan, D.B.
[18]
[19]
15
Stolz, G.D. Papworth, A.F. Zahorchak, A.J. Logar, Z. Wang, S.C. Watkins, L.D. Falo Jr., A.W. Thomson, Endocytosis, intracellular sorting, and processing of exosomes by dendritic cells, Blood 104 (2004) 3257 – 3266. M.R. Neutra, Current concepts in mucosal immunity. V Role of M cells in transepithelial transport of antigens and pathogens to the mucosal immune system, Am. J. Physiol. 274 (1998) G785 – G791. M. Rescigno, M. Urbano, B. Valzasina, M. Francolini, G. Rotta, R. Bonasio, F. Granucci, J.P. Kraehenbuhl, P. Ricciardi-Castagnoli, Dendritic cells express tight junction proteins and penetrate gut epithelial monolayers to sample bacteria, Nat. Immun. 2 (2001) 361 – 367. J.H. Niess, S. Brand, X. Gu, L. Landsman, S. Jung, B.A. McCormick, J.M. Vyas, M. Boes, H.L. Ploegh, J.G. Fox, D.R. Littman, H.C. Reinecker, CX3CR1-mediated dendritic cell access to the intestinal lumen and bacterial clearance, Science 307 (2005) 254 – 258. K. Terpend, F. Boisgerault, M.A. Blaton, J.F. Desjeux, M. Heyman, Protein transport and processing by human HT29-19A intestinal cells: effect of interferon gamma, Gut 42 (1998) 538 – 545. R.M. Hershberg, D.H. Cho, A. Youakim, M.B. Bradley, J.S. Lee, P.E. Framson, G.T. Nepom, Highly polarized HLA class II antigen processing and presentation by human intestinal epithelial cells, J. Clin. Invest. 102 (1998) 792 – 803. G. van Niel, G. Raposo, C. Candalh, M. Boussac, R. Hershberg, N. Cerf-Bensussan, M. Heyman, Intestinal epithelial cells secrete exosome-like vesicles, Gastroenterology 121 (2001) 337 – 349. C.N. Johnstone, N.C. Tebbutt, H.E. Abud, S.J. White, K.L. Stenvers, N.E. Hall, S.H. Cody, R.H. Whitehead, B. Catimel, E.C. Nice, A.W. Burgess, J.K. Heath, Characterization of mouse A33 antigen, a definitive marker for basolateral surfaces of intestinal epithelial cells, Am. J. Physiol.: Gastrointest. Liver. Physiol. 279 (2000) G500 – G510. B. Simon, D.K. Podolsky, G. Moldenhauer, K.J. Isselbacher, S. Gattoni-Celli, S.J. Brand, Epithelial glycoprotein is a member of a family of epithelial cell surface antigens homologous to nidogen, a matrix adhesion protein, Proc. Natl. Acad. Sci. U. S. A. 87 (1990) 2755 – 2759. J.K. Heath, S.J. White, C.N. Johnstone, B. Catimel, R.J. Simpson, R.L. Moritz, G.F. Tu, H. Ji, R.H. Whitehead, L.C. Groenen, A.M. Scott, G. Ritter, L. Cohen, S. Welt, L.J. Old, E.C. Nice, A.W. Burgess, The human A33 antigen is a transmembrane glycoprotein and a novel member of the immunoglobulin superfamily, Proc. Natl. Acad. Sci. U. S. A. 94 (1997) 469 – 474. L. Zitvogel, A. Regnault, A. Lozier, J. Wolfers, C. Flament, D. Tenza, P. Ricciardi Castagnoli, G. Raposo, S. Amigorena, Eradication of established murine tumors using a novel cell-free vaccine: dendritic cell-derived exosomes, Nat. Med. 4 (1998) 594 – 600. A.E. Morelli, A.T. Larregina, W.J. Shufesky, M.L. Sullivan, D.B. Stolz, G.D. Papworth, A.F. Zahorchak, A.J. Logar, Z. Wang, S.C. Watkins, L.D. Falo Jr., A.W. Thomson, Endocytosis, intracellular sorting and processing of exosomes by dendritic cells, Blood 104 (2004) 3258 – 3266. K. Denzer, M. van Eijk, M.J. Kleijmeer, E. Jakobson, C. de Groot, H.J. Geuze, Follicular dendritic cells carry MHC class II-expressing microvesicles at their surface, J. Immunol. 165 (2000) 1259 – 1265. A.J. van Rensen, L.S. Taams, M.C. Grosfeld-Stulemeyer, W. van Eden, D.J. Crommelin, M.H. Wauben, Induction of T cell anergy by liposomes with incorporated major histocompatibility complex (MHC) II/peptide complexes, Pharm. Res. 17 (2000) 720 – 726. M. Karlsson, S. Lundin, U. Dahlgren, H. Kahu, I. Pettersson, E. Telemo, ‘‘Tolerosomes’’ are produced by intestinal epithelial cells, Eur. J. Immunol. 31 (2001) 2892 – 2900. G. van Niel, J. Mallegol, C. Bevilacqua, C. Candalh, S. Brugiere, E. Tomaskovic-Crook, J.K. Heath, N. Cerf-Bensussan, M. Heyman, Intestinal epithelial exosomes carry MHC class II/peptides able to inform the immune system in mice, Gut 52 (2003) 1690 – 1697. C. Thery, L. Duban, E. Segura, P. Veron, O. Lantz, S. Amigorena,
16
J. Mallegol et al. / Blood Cells, Molecules, and Diseases 35 (2005) 11 – 16
Indirect activation of naive CD4+ T cells by dendritic cell-derived exosomes, Nat. Immun. 3 (2002) 1156 – 1162. [20] H. Vincent-Schneider, P. Stumptner-Cuvelette, D. Lankar, S. Pain, G. Raposo, P. Benaroch, C. Bonnerot, Exosomes bearing HLA-DR1 molecules need dendritic cells to efficiently stimulate specific T cells, Int. Immunol. 14 (2002) 713 – 722. [21] A. Simonsen, E. Stang, B. Bremnes, M. Roe, K. Prydz, O. Bakke, Sorting of MHC class II molecules and the associated invariant chain (Ii) in polarized MDCK cells, J. Cell Sci. 110 (Pt. 5) (1997) 597 – 609.
[22] D.F. Dukers, P. Meij, M.B. Vervoort, W. Vos, R.J. Scheper, C.J. Meijer, E. Bloemena, J.M. Middeldorp, Direct immunosuppressive effects of EBV-encoded latent membrane protein 1, J. Immunol. 165 (2000) 663 – 670. [23] E.R. Boulan, D.D. Sabatini, Asymmetric budding of viruses in epithelial monolayers: a model system for study of epithelial polarity, Proc. Natl. Acad. Sci. U. S. A. 75 (1978) 5071 – 5075. [24] D.G. Nguyen, A. Booth, S.J. Gould, J.E. Hildreth, Evidence that HIV budding in primary macrophages occurs through the exosome release pathway, J. Biol. Chem. 278 (2003) 52347 – 52354.