Docstoc

BCR-bound antigen is targeted to exosomes in human follicular

Document Sample
BCR-bound antigen is targeted to exosomes in human follicular Powered By Docstoc
					Biol. Cell (2006) 98, 491–501 (Printed in Great Britain)   doi:10.1042/BC20060027
                                                                                                               Research article


BCR-bound antigen is targeted to
exosomes in human follicular
lymphoma B-cells1
Pascale Rialland*, Danielle Lankar*, Graca Raposo†, Christian Bonnerot* and Pascale Hubert*2
                                        ¸
*INSERM U520, Institut Curie, 26, rue d’Ulm, 75 248 Paris cedex 05, France, and †CNRS UMR 144, Institut Curie, 26, rue d’Ulm,
75 248 Paris cedex 05, France


Background information. Exosomes are small membrane vesicles secreted by several cell types during exocytic
fusion of multivesicular bodies with the plasma membrane. Exosomes from tumour cells can transfer antigens
from cell to cell, a property favouring antigen-specific immune responses in vitro and in vivo, and are thus an
interesting putative therapeutic tool in human cancers. Exosomes have been well studied in EBV (Epstein–Barr
virus)-transformed human B-cell lines; however, biological stimuli regulating exosome secretion quantitatively
and/or qualitatively still remain poorly defined.
Results. We analysed the effect of the BCR stimulation on exosome release in the human follicular lymphoma B-cell
line DOHH2. We found that BCR (B-cell receptor) triggering of DOHH2 cells induced the polarization of CD63+
MHC class II compartments. Moreover, BCR stimulation increased the release of exosome-associated proteins in
the extracellular space. Finally, we found that the BCR was expressed at the surface of exosomes, and could target
a bound anti-human IgG to these vesicles.
Conclusions. BCR can modulate the protein content of exosomes upon stimulation, and can target its bound
antigen to these vesicles.




Introduction                                                                        LAMP-1/2 (lysosome-associated membrane protein
Exosomes are small 50–90 nm diameter vesicles                                       1/2) and the invariant chain Ii (Escola et al., 1998;
secreted by several haematopoietic and non-haemato-                                 Denzer et al., 2000), which invaginate to form in-
poietic cells (Denzer et al., 2000; Th´ ry et al., 2002a;
                                      e                                             ternal vesicles. Exosome composition has been ex-
F´ vrier and Raposo, 2004). They are of endosomal
 e                                                                                  tensively studied in several cell types including APCs
origin and are released upon fusion of MVBs (multi-                                     e
                                                                                    (Th´ ry et al., 2002a). On their surface, exosomes bear
vesicular bodies) with the cell surface (Raposo et al.,                             mature MHC class II and MHC class I molecules,
1996). In APCs (antigen-presenting cells), MVBs                                     tetraspanins, such as CD63, as well as the co-stimu-
correspond to late endocytic MIICs (MHC class II-                                   latory molecule CD86. Cytosolic proteins are found
enriched compartments) formed by an outer mem-                                      inside the vesicles, such as cytoskeletal proteins (actin
brane expressing class II molecules together with                                   and tubulin), or proteins involved in the endocytic
                                                                                    traffic (annexins, Tsg101 and Alix). Raft-associated
1 Thiswork is dedicated to the memory of Christian Bonnerot, head of the
                                                                                    proteins (flotillin and Lyn) are also present in exo-
INSERM U520.                                                                        somes, consistent with the cholesterol/sphingolipids
2 To whom correspondence should be addressed (email
                                                                                    composition of the exosomal membrane (de Gassart
pascale.hubert@curie.fr).
Key words: B-cell, B-cell receptor (BCR), exosome, MHC class II molecule.           et al., 2003; Wubbolts et al., 2003). TfR (transferrin
Abbreviations used: Ag, antigen(s); APC, antigen-presenting cell; BCR, B-cell       receptor) and acetylcholinesterase are also present at
receptor; EBV, Epstein–Barr virus; GaH IgG, goat anti-human IgG antibody
F(ab )2 fragment; GaM IgM, goat anti-mouse IgM antibody F(ab )2 fragment;           high levels in exosomes from reticulocytes and from
gp96, glycoprotein 96; HLA DR, human leucocyte antigen DR; HRP,                     the K562 erythroleukemia cell line (Vidal et al.,
horseradish peroxidase; mAb, monoclonal antibody; MIIC, MHC class II-
enriched compartment; MVB, multivesicular body; PE, phycoerythrin; TfR,
                                                                                    1997). Moreover, endogenous Ag (antigens) such as
transferrin receptor.                                                               melanA or PrPc (cellular isoform of the prion protein)


                                                                                                    www.biolcell.org | Volume 98 (8) | Pages 491–501   491
                                                                                                P. Rialland and others


      and PrPsc (scrapie prion-related protein) can be found      omas, which result from the malignant transformation
      in exosomes from melanoma cells and prion-infected          of a germinal centre B-cell bearing the chromosomal
                                                   e
      cells respectively (Wolfers et al., 2001; F´ vrier and      translocation t(14;18) (Gandhi and Marcus, 2005).
      Raposo, 2004).                                              In these cancers, besides the Ig idiotype of the malig-
         MVBs/exosomes are at the meeting point between           nant B-cell clone which is specific for each patient,
      the endocytic and the exocytic pathways, but the            no shared tumour rejection Ag has been identified,
      mechanisms regulating their formation have not              compromising Ag-specific immunotherapies (Xie
      yet been fully elucidated. Recently, Hrs (hepatocyte        et al., 2003).
      growth factor-regulated tyrosine kinase substrate),            In B-cells, the BCR (B-cell receptor) binds its Ag
      Alix and the ESCRT (endosomal sorting complexes             specifically, and this complex is then internalized, de-
      required for transport) machinery have been involved        graded into peptides in the endocytic pathway and
      in the targeting of ubiquitinated cargo proteins to the     loaded on to newly synthesized MHC class II mol-
      internal vesicles of MVBs (de Gassart et al., 2004). In     ecules in the so-called MHC–peptide loading com-
      particular, membrane receptors, such as TfR (Vidal          partments (West et al., 1994). These compartments
      et al., 1997), or proteins endocytosed from the extra-      are thought to be part of the MIICs, and have a similar
      cellular medium (Raposo et al., 1996) are targeted          morphology to that of MVBs with immuno-electron
      to MVBs, after which they are processed to lyso-            microscopy (Peters et al., 1991; Lankar et al., 2002).
      somes for degradation or to exosomes for excretion          Afterwards, the MHC class II–peptide complexes are
      outside the cell (de Gassart et al., 2004). Interest-       delivered to the cell surface, allowing Ag presenta-
      ingly, cross-linking of TfR or acetylcholinesterase by      tion to CD4+ T-cells (Tulp et al., 1994; West et al.,
      specific mAbs (monoclonal antibodies) not only res-          1994). Interestingly, it has been demonstrated that
      ults in enhanced internalization and degradation of         BCR stimulation promotes de novo MVB formation
      the receptor (Weissman et al., 1986), but also in           in murine lymphoma B-cells (Lankar et al., 2002),
      an increased release of exosomes containing this re-        which contain components of the BCR (West et al.,
      ceptor (Vidal et al., 1997; Savina et al., 2003). In        1994; Lankar et al., 2002). In addition, EBV
      the same way, stimulation of RBL (rat basophilic            (Epstein–Barr virus)-transformed B-cell lines display
      leukaemia) cells by an anti-IgE mAb or ionomycin            large numbers of MVBs which spontaneously fuse
      have also been shown to trigger exosome exocyto-            with the plasma membrane and release numerous
      sis (Vincent-Schneider et al., 2002), suggesting that       exosomes (Raposo et al., 1996). However, the ques-
      this phenomenon might be tightly regulated. In fact,        tion of whether BCR triggering can modulate the
      Rab11 and calcium flux positively regulate exosome           exosome release quantitatively and/or qualitatively
      secretion in K562 cells (Savina et al., 2002, 2003).        has not been addressed.
         Although the physiological role of exosomes re-             Because of the putative clinical applications of exo-
      mains yet unclear, their characteristics suggest that       somes in tumour immunotherapy, it has become im-
      they could serve as intercellular communication             portant to determine the biological stimuli that are
      vehicles, particularly for transferring MHC–peptide         able to regulate their production by both normal and
      complexes or whole cellular Ag from cell to cell            transformed cells. In the present study, we used the
           e
      (Th´ ry et al., 2002b). Indeed, exosomes produced           B-cell line DOHH2 (Kluin-Nelemans et al., 1991),
      by autologous or allogeneic tumour cells can inhibit        which is a model for a human follicular lymphoma.
      tumour growth in mice by the transfer of shared             We asked whether BCR triggering might influence
      tumour-associated Ag to host dendritic cells, sub-          exosome secretion and affect the protein profile of exo-
      sequently eliciting Ag-specific anti-tumour immune           somes, as well as if the BCR could target its bound
      responses (Wolfers et al., 2001). Thus the use of exo-      Ag to exosomes through the MIIC.
      somes produced by allogeneic tumour cell lines or
      the stimulation of exosome production by endogen-           Results
      ous tumour cells could be useful in the treatment of        DOHH2 phenotype allows dynamic studies of
      human cancers for which no shared tumour-associated         MIICs and exosome secretion
      Ag is yet available. This is the case of human follicular   Exosomes have been demonstrated to transfer Ag
      B-cell lymphomas, a subset of non-Hodgkin’s lymph-          from tumour cells to APCs, therefore inducing


492   C   Portland Press 2006 | www.biolcell.org
BCR targets antigen to lymphoma B-cell exosomes                                                Research article
Figure 1 Phenotype of DOHH2 cells
(A) Cells were washed, labelled with the indicated mAb and analysed by flow cytometry. Broken lines, isotype control; continuous
lines, indicated specific mAb. (B) Cells were fixed, permeabilized, labelled with the anti-(MHC class II) mAb L243 followed by
goat anti-(mouse IgG) coupled to Alexa Fluor® 488, and analysed by confocal microscopy.




Ag-specific immune responses and inhibiting tumour                lar lymphoma, and bears the chromosomal trans-
growth (Wolfers et al., 2001). Allogeneic tumour ex-             location t(14;18) characteristic of this pathology
osomes thus represent an attractive therapeutic tool             (Kluin-Nelemans et al., 1991). These DOHH2 cells
for human cancers for which no shared tumour-asso-               were CD20+ CD38high as expected for germinal
ciated Ag is known (Wolfers et al., 2001), such as               centre B-cells, and expressed a BCR composed of
follicular B lymphomas (Xie et al., 2003). However,              an IgG heavy chain and a λ light chain (IgGλ)
biological stimuli regulating exosome release are yet            (Figure 1A). They also expressed markers commonly
poorly defined. Thus, in the present study, we have               found in B-cells, such as the Ag-presenting MHC
investigated a cellular model to allow the analysis of           class I and MHC class II molecules, CD40, CD71 and
the effect of BCR triggering on exosome secretion and            the co-stimulatory molecule CD86. Interestingly,
protein composition in human B-cell lymphomas.                   they did not express the activation marker CD25,
   EBV-transformed human B-cell lines have been ex-              and only 15% of them displayed low levels of CD80,
tensively used to study MIICs and exosomes, since                thus suggesting that DOHH2 cells display a rest-
they constitutively display high levels of intracellular         ing activation status. Moreover, immunofluorescence
MHC class II molecules and numerous MVBs (Peters                 labelling followed by confocal microscopy revealed
et al., 1991; West et al, 1994; Raposo et al., 1996;             that MHC class II molecules were mainly localized
Escola et al., 1998). However, such characteristics              at the plasma membrane of DOHH2 cells (Fig-
impede the study of the dynamic of MIIC formation                ure 1B), therefore this cell line is a suitable model to
and exosome secretion upon receptor-mediated cell                study the redistribution of MHC class II molecules
stimulation. Indeed, cross-linking the surface Ig in             inside the cell and exosome secretion after cell activ-
a human EBV-transformed B-cell line does not alter               ation.
intracellular distribution of MHC class II molecules
(Peters et al., 1995). In contrast, mouse B-cell lines,          BCR-bound Ag induces the polarization of CD63+
such as A20 derived from a spontaneous lymphoma,                 MIICs, and then is targeted to these
display, in the basal state, few intracellular MHC               compartments in DOHH2 cells
class II molecules and MVBs, which are triggered to              It has been demonstrated in murine lymphoma B-
form upon BCR stimulation (Siemasko et al., 1998;                cells that BCR triggering induces the rapid aggreg-
Lankar et al., 2002).                                            ation of vesicular MIICs (Siemasko et al., 1998)
   We used the human B-cell line DOHH2,                          and MVBs (Lankar et al., 2002). This vesicular co-
which was derived from a patient with a follicu-                 alescence occurs in response to biochemical signals


                                                                                    www.biolcell.org | Volume 98 (8) | Pages 491–501   493
                                                                                                         P. Rialland and others


      Figure 2 BCR-bound Ag induces the polarization of CD63+ MIICs, and then BCR–Ag is processed to these
      compartments in DOHH2 cells
      Cells were stimulated with GaH IgG (50 µg/ml) (A) or with Alexa Fluor® 488-conjugated GaH IgG (50 µg/ml) (B and C) for the
      indicated period of time, then fixed on to glass coverslips and permeabilized. Cells were then labelled with the following anti-
      bodies: rabbit anti-(MHC class II) and goat anti-(rabbit Alexa Fluor® 488), together with anti-CD63 mAb and donkey anti-(mouse
      Cy3) (A); anti-CD63 mAb (B) or anti-(MHC class II) mAb L243 (C) followed by donkey anti-(mouse Cy3). Arrows indicate examples
      of co-localization of proteins.




      produced by the BCR, such as tyrosine kinase activ-              ilar to murine B-cells (Siemasko et al., 1998; Lankar
      ation (Siemasko et al., 1998). First, we therefore               et al., 2002), and occurred in 80% of activated
      determined whether stimulation of the human B-                   DOHH2 cells.
      cells DOHH2 through the BCR might also affect the                   Moreover, a similar percentage of cells was able to
      distribution of MHC class II molecules and MIICs.                internalize the BCR-specific ligand GaH IgG coupled
         We verified that BCR triggering with a satur-                  to Alexa Fluor® 488. The fluorescence could be detec-
      ating amount of GaH IgG [goat anti-human IgG                     ted in the cytoplasm of DOHH2 cells at 10 min after
      antibody F(ab )2 fragment] was efficient at activating            the onset of the stimulation. However, the fluoresence
      DOHH2 cells, inducing calcium flux, early protein                 staining only co-localized after 30 min and only par-
      tyrosine phosphorylation and CD69 expression (data               tially with CD63+ compartments (Figure 2B) or with
      not shown). Confocal microscopy observations were                MHC class II molecules (Figure 2C). The same results
      then made using DOHH2 cells incubated with GaH                   were obtained where the cells were labelled with an
      IgG at various chase times. After staining with a rab-           anti-LAMP-1 (lysosome-associated membrane pro-
      bit anti-(MHC class II molecule) antibody, cells were            tein 1) mAb, another marker of MIICs and MVBs
      co-labelled with a mAb specific for CD63, a marker                (data not shown). Taken together, these results sug-
      of MIICs and MVBs (Peters et al., 1991; Escola et al.,           gested that the BCR-bound Ag was internalized rap-
      1998; Wubbolts et al., 2003). In unstimulated cells,             idly in DOHH2 cells, but reached the MVBs after
      CD63 was detected in small intracytoplasmic vesi-                they were polarized.
      cles, and did not co-localize with MHC class II mol-
      ecules found at the plasma membrane (Figure 2A).                 Phenotypic characterization of exosomes from
      However, BCR stimulation induced a rapid accu-                   DOHH2 cells
      mulation of intracellular MHC class II molecules                 Exosomes are spontaneously released from MVB upon
      which co-localized with vesicular CD63+ compart-                 fusion with the plasma membrane, and display vari-
      ments that were redistributed in large aggregates                ous surface and intra-vesicular proteins depending
      polarized at a cellular pole. This coalescence began to          on the cell from which they originate (Raposo et al.,
      appear at 10 min and reached a maximum after 1 h                          e                   e
                                                                       1996; Th´ ry et al., 2002a; F´ vrier and Raposo, 2004).
      of BCR stimulation, a time-scale of coalescence sim-             To characterize the exosomes spontaneously secreted


494   C   Portland Press 2006 | www.biolcell.org
BCR targets antigen to lymphoma B-cell exosomes                                             Research article
Figure 3 Phenotype of DOHH2 cell exosomes                        cells produced low amounts of exosomes compared
(A) An electron micrograph of DOHH2 cell exosomes. (B) Exo-      with EBV-transformed B-cell lines and dendritic cells
somes produced by unstimulated DOHH2 cells for 6 h at 37◦ C                                  e
                                                                 (Raposo et al., 1996; Th´ ry et al., 2002a, 2002b;
were coated on to latex beads, labelled with the indicated mAb     e
                                                                 F´ vrier and Raposo, 2004), which was confirmed
and analysed by flow cytometry. Broken lines, isotype control;    by protein quantification in exosome preparations by
continuous lines, specific mAbs. Broken lines are not visible     the Lowry assay (on average, 5 µg of protein/1 × 108
where lines overlapp.                                            DOHH2 cells over 6 h).
                                                                    Flow cytometry analysis of these exosomes coated
                                                                 on to latex beads (Figure 3B) revealed, at their sur-
                                                                 face, a strong expression of MHC class I and a lower
                                                                 expression of MHC class II molecules, as well as
                                                                 the presence of CD86, similar to that observed at the
                                                                 plasma membrane of DOHH2 cells (see Figure 1A).
                                                                 In contrast, CD40 and CD71, although highly ex-
                                                                 pressed at the cell surface, were not found on exo-
                                                                 somes, confirming that exosome preparations were
                                                                 not contaminated with fragments of broken plasma
                                                                 membrane. Moreover, the tetraspanin CD63, which
                                                                 is known to be associated with detergent-resistant do-
                                                                 mains of exosomes (Wubbolts et al., 2003), was also
                                                                 found by FACS, together with the CD20, another
                                                                 member of this family of proteins which is specifi-
                                                                 cally expressed in B-cells. Interestingly, the endogen-
                                                                 ous BCR detected by an anti-(IgGλ light chain) mAb
                                                                 or an anti-human IgG serum was present on the sur-
                                                                 face of exosomes from unstimulated DOHH2 cells.
                                                                    Moreover, Western blotting experiments showed
                                                                 that exosomes from unstimulated DOHH2 cells dis-
                                                                 played high levels of MHC class I and low levels
                                                                 of MHC class II molecules in accordance with the
                                                                 flow cytometry data (Figure 4, lane 2). The lipid
                                                                 raft marker flotillin-1, which has been found in exo-
                                                                 somes from the Daudi B-cell line (de Gassart et al.,
                                                                 2003), was also found, but Tsg101, which is present
                                                                                                              e
                                                                 in dendritic cell-derived exosomes (F´ vrier and
                                                                 Raposo, 2004), could not be detected in exosomes
                                                                 from DOHH2 cells (data not shown). However, only
                                                                 small amounts of TfR, actin, endogenous BCR and
                                                                 CD63 were seen, probably because of the weak pro-
                                                                 duction of exosomes by DOHH2 cells in the basal
                                                                 state. The lack of contamination of exosome prepara-
                                                                 tions by endoplasmic reticulum membranes was also
                                                                 verified by the absence of calnexin and gp96 (gly-
                                                                 coprotein 96) labelling.
by DOHH2 cells, the cells were cultured for 6 h in
depleted medium and exosomes were purified as de-                 BCR triggering increases the release of
scribed previously (Wubbolts et al., 2003). The pur-             exosome-associated proteins by DOHH2 cells
ity of the preparations was assessed by electron mi-             BCR triggering induces MVB formation in mur-
croscopy (Figure 3A). It was noted that DOHH2                    ine (Siemasko et al., 1998; Lankar et al., 2002) and


                                                                                 www.biolcell.org | Volume 98 (8) | Pages 491–501   495
                                                                                                        P. Rialland and others


      Figure 4 Western blotting analysis of DOHH2 cell                    is dependent on calcium flux, a critical biochemical
      exosomes                                                            event of the BCR signalling pathway (Savina et al.,
      Equal numbers of DOHH2 cells (1 × 108 ) were cultured for           2003).
      6 h at 37◦ C in the presence (BCR) or absence (NS) of GaH              FACS analysis was not sensitive enough to reveal
      IgG. Exosomes were purified from the culture medium, and             any quantitative variation in exosomes produced after
      samples (4 µl) of each exosome preparation (corresponding           BCR stimulation (data not shown). Thus we used
      to exosomes produced by 2 × 107 unstimulated or stimu-              Western blotting to analyse exosomes from BCR-
      lated cells) were lysed, subjected to SDS/PAGE (10% gel) and        stimulated cells. Interestingly, all proteins expressed
      Western blotting with the indicated antibody. Note that the TfR     in exosome preparations from unstimulated DOHH2
      appears as a doublet, with the lower band probably corres-          cells were increased in exosomes recovered from the
      ponding to a truncated form of the molecule (Vidal et al., 1997).   same number of BCR-stimulated cells (Figure 4,
      For CD63 analysis, samples were run under non-reducing              lane 3). This suggested that BCR triggering could en-
      conditions. Whole unstimulated DOHH2 cell lysate (10 µg)            hance exosome secretion and/or the protein content
      was run in parallel. Proteins were revealed using the ECL®
                                                                          of exosomes. This result was confirmed by the in-
      (exposure time, 30 s).
                                                                          creased protein concentration found in exosome pre-
                                                                          parations from BCR-stimulated cells compared with
                                                                          unstimulated cells, as determined by the Lowry assay
                                                                          and analysed by paired Student’s t test (mean + S.D.:
                                                                                                                          −
                                                                          0.43 µg/µl + 0.15 versus 0.26 µg/µl + 0.12 respect-
                                                                                       −                          −
                                                                          ively; n = 4; P = 0.024). Moreover, although MHC
                                                                          class I molecules and flotillin were enhanced 1.6-fold
                                                                          in exosome preparations from stimulated cells, MHC
                                                                          class II molecules appeared particularly enriched,
                                                                          with a 13-fold increase as determined by densitomet-
                                                                          ric quantification of the bands (Figure 4). We verified
                                                                          that the increase in protein content seen in exosome
                                                                          preparations from BCR-stimulated cells was not an
                                                                          artifact due to the aggregation of spontaneously ex-
                                                                          creted exosomes (which expressed endogenous IgG,
                                                                          see Figure 3) by the stimulating GaH IgG that re-
                                                                          mained in the culture medium (data not shown). Thus
                                                                          these data suggested that BCR triggering could in-
                                                                          crease the excretion of exosomes in DOHH2 cells,
                                                                          and/or could induce a selective recruitment of pro-
                                                                          teins, such as MHC class II molecules, to these micro-
                                                                          vesicles.

                                                                          BCR-bound Ag is targeted to exosomes
                                                                          In unstimulated B-cells, the BCR spontaneously in-
                                                                          ternalizes and recycles back to the plasma membrane,
                                                                          while a small fraction is targeted to late endosomal
                                                                          compartments (Davidson et al., 1991; Song et al.,
                                                                          1995; Brown et al., 1999). BCR cross-linking ac-
                                                                          celerates its endocytosis and enhances its delivery
                                                                          to MIICs, which undergo morphological changes
      human lymphoma B-cells (Figure 2), however, no                      and acquire newly synthesized MHC class II mol-
      data are available describing a putative effect of BCR              ecules (Brown et al., 1999; Clark et al., 2004). Thus,
      stimulation on exosome production. This idea is sug-                BCR–Ag complexes rapidly enter the MIICs and the
      gested by the fact that, in K562 cells, exosome release             MVBs in an intact conformation (Davidson et al.,


496   C   Portland Press 2006 | www.biolcell.org
BCR targets antigen to lymphoma B-cell exosomes                                                  Research article
Figure 5 The stimulating GaH IgG is targeted to exosomes
(A) DOHH2 cells (1 × 108 ) were left unstimulated or were stimulated with GaH IgG or with control GaM IgM for 6 h at 37◦ C, and
then exosomes were purified from the supernatants. Exosomes produced by 35 × 108 cells (7 µl of each exosome preparation)
were then lysed, and samples were loaded on to a 10% gel for SDS/PAGE and blotted using a donkey anti-goat antibody
coupled to HRP. Note that the stimulating GaH IgG is not present in DOHH2 whole-cell lysate (10 µg) which was run in parallel.
(B) Cells (1 × 108 ) were coated with GaH IgG–FITC or GaM IgM–FITC or left in AIMV medium alone for 15 min on ice, then
washed twice and chased in fresh AIMV medium for 6 h at 37◦ C. Exosomes were purified from the supernatants, coated on
to latex beads (3 µl/sample) and analysed by FACS. Upper panels: Fl1 fluorescence intensity of exosomes from unstimulated
cells (broken lines) and from activated cells (continuous lines). Lower panels: exosomes from stimulated cells were labelled with
an isotype control (broken lines) or with anti-(HLA DR) mAb (continuous lines). (C) Left-hand panel: samples of the exosome
preparations used in (B) (4 µl/lane) were blotted with anti-(MHC class I) mAb. DOHH2 whole-cell lysate (10 µg) was run in parallel.
The exposure time for the blot after the ECL® reaction was 5 min. Right-hand panel: to verify the efficiency of BCR triggering in
(B), a sample of DOHH2 cells was taken from each culture flask after 5 min of chase at 37◦ C, and the levels of phosphotyrosine
proteins was analysed by Western blotting.




1990; West et al., 1994; Siemasko et al., 1998; Lankar                DOHH2 cells were stimulated with GaH IgG, and
et al., 2002), and subsequently are degraded to allow              exosomes were purified as above, lysed and the pro-
the loading of antigenic peptides on to MHC class II               teins subjected to Western blotting using a donkey
for presentation to T-cells (Davidson et al., 1990).               anti-goat serum conjugated to HRP (horseradish per-
However, previously published data showed that a                   oxidase). A goat light chain, at its native molecular
small fraction of internalized BCR-bound Ag was                    mass of 25 kDa, was only present in exosome pre-
not degraded in late endosomes at later times after                parations obtained from GaH IgG-stimulated cells.
the onset of endocytosis, and could be excreted in the             Control exosomes obtained from cells activated with
extracellular medium (Watts and Davidson, 1988;                    a GaM IgM [goat anti-mouse IgM antibody F(ab )2
Davidson et al., 1990). These data, together with the              fragment] did not display any goat light chain, show-
fact that the BCR was expressed in exosomes from                   ing that antibodies present freely in the medium did
unstimulated DOHH2 cells (Figure 3), prompted us                   not co-purify non-specifically with exosomes (Fig-
to address the question of whether the BCR could                   ure 5A). Moreover, the same result was obtained when
target its bound Ag to exosomes.                                   DOHH2 cells were first coated with GaH IgG or


                                                                                      www.biolcell.org | Volume 98 (8) | Pages 491–501   497
                                                                                                  P. Rialland and others


      control GaM IgM on ice, washed and chased in fresh           similar levels of both molecules (Clayton et al., 2001).
      medium for 6 h at 37◦ C (data not shown). We were            This could be a consequence of the low level of
      unable to detect degraded fragments of the stimulat-         intracellular MHC class II expression in DOHH2
      ing goat antiserum by Western blotting in exosome            cells. Other proteins expressed at the plasma mem-
      preparations, suggesting that only intact BCR and            brane, such as CD71 and CD40, were not detected
      Ag might be targeted to these vesicles. Alternatively,       on DOHH2 exosomes, supporting the idea that sur-
      Western blotting would not be sensitive enough to            face transmembrane molecules could be differentially
      detect degraded fragments, because of their small            sorted to exosomes through a yet poorly understood
      molecular masses.                                            mechanism (de Gassart et al., 2004). Interestingly,
         In another set of experiments, DOHH2 cells were           we demonstrated, in the present study, the presence
      incubated with FITC-conjugated GaH IgG or GaM                of the endogenous membrane IgG in DOHH2 exo-
      IgM or with the same volume of PBS on ice for                somes. This could be explained by the fact that un-
      15 min, washed twice and chased in fresh AIMV                ligated BCR is constitutively internalized at a very
      medium. Exosomes were then recovered and coated              low rate, and then either recycles back to the plasma
      on to latex beads. Flow cytometry analysis (Fig-             membrane or reaches, to a lesser extent, the late endo-
      ure 5B, upper panels) showed that exosomes from              somal compartments (Song et al., 1995; Brown
      (GaH IgG)–FITC-coated cells displayed a signifi-              et al., 1999), a fraction of which would be then sorted
      cant increase in Fl1 fluorescence compared with exo-          to exosomes.
      somes from unstimulated cells, which was not the                BCR is responsible for efficient Ag capture, which
      case of exosomes produced by control cells incubated         is followed by B-cell activation, and by Ag presenta-
      with (GaM IgM)–FITC. The efficiency of BCR activ-             tion after targeting to peptide-loading compartments
      ation was verified by the increase in tyrosine phos-          (Clark et al., 2004). We observed in the present study
      phorylated proteins in lysates from DOHH2 cells              that another consequence of BCR stimulation was to
      coated with (GaH IgG)–FITC and chased for 5 min              increase the release of exosome-associated proteins by
      at 37◦ C (Figure 5C, right-hand panel). In addition,         DOHH2 lymphoma cells (Figure 4). This could be
      MHC class I molecules were increased 1.6-fold in             due to an enhanced formation of MVBs by the BCR,
      exosome preparation from (GaH IgG)–FITC-coated               as shown in murine B-cells (Lankar et al., 2002), and
      DOHH2 cells compared with unstimulated or with               suggested by immunofluorescence studies (Figure 2),
      (GaM IgM)–FITC-incubated cells, confirming the ef-            which could then lead to the release of increased num-
      ficiency of this stimulus (Figure 5C, left-hand panel).       bers of MVB internal vesicles. The mechanism under-
      However, we could not determine if all of the FITC           lying this increase in exosome secretion might in-
      was still associated with intact GaH IgG, or if part         volve a rise in intracellular calcium flux (Savina et al.,
      of the FITC was disrupted from the IgG or was asso-          2003), as well as PI3K (phosphoinositide 3-kinase)
      ciated with degraded IgG fragments.                          activation (Siemasko et al., 1998; Granboulan et al.,
         Nevertheless, these results suggested that the GaH        2003; Clark et al., 2004), which are two events of the
      IgG ligand internalized through the BCR (see Fig-            BCR activation pathway thought to be implicated
      ure 2) could be processed, at least partially in an intact   in MVB biogenesis. Alternatively, BCR stimulation
      form, to the exosomes.                                       could enhance the recruitment of some proteins such
                                                                   as MHC class II molecules to exosomes (Figure 4).
      Discussion                                                   The yet unresolved question is the origin of these
      DOHH2 cell exosomes resemble those secreted by               MHC class II molecules seen in the coalescent MIICs
      dendritic cells and EBV-transformed B-cell lines in          after BCR stimulation (Figure 2) and highly targeted
      terms of protein composition (Raposo et al., 1996;           to exosomes. We did not notice any down-modulation
                                                   e
      Escola et al., 1998; Clayton et al., 2001; Th´ ry et al.,    of MHC class II molecules expressed at the plasma
      2002a; Wubbolts et al., 2003) (and see Figures 3             membrane by FACS after BCR triggering (data not
      and 4). Nevertheless, in exosomes from DOHH2                 shown), suggesting that the massive accumulation of
      cells, MHC class II were weakly expressed compared           intracellular MHC class II molecules did not come
      with MHC class I molecules, contrary to exosomes             from the cell surface (Pinet et al., 1995). Another
      from EBV-transformed B-cell lines which harbour              possibility might be that the redistribution of MHC


498   C   Portland Press 2006 | www.biolcell.org
BCR targets antigen to lymphoma B-cell exosomes                                               Research article
class II molecules could be associated with their rapid         Antibodies and reagents
neosynthesis upon BCR stimulation (Davidson et al.,             GaH IgG and GaM IgM sera [F(ab )2 fragments], either un-
                                                                coupled or coupled to FITC, were purchased from Jackson Im-
1991; West et al., 1994).                                       munoResearch Laboratories (West Grove, PA, U.S.A.). In some
   In addition, we observed that, after incubation              experiments, we conjugated the GaH IgG to Alexa Fluor® 488
with DOHH2 cells, the BCR ligand GaH IgG could                  (Molecular Probes, Leiden, The Netherlands).
be found in exosome preparations (Figure 5). This                  The following mAbs were used for flow cytometry: anti-(HLA
probably occurred through endocytosis of BCR–Ag                 DR)–FITC, anti-CD86–FITC, anti-CD80–PE and anti-CD38–
                                                                PE were from Becton Dickinson (where HLA DR is human
complexes (see Figure 2) and targeting to exosomes,             leucocyte antigen DR and PE is phycoerythrin); anti-CD71–
rather than by shedding of these complexes to the cul-          FITC and PE-conjugated anti-(human λ light chain) were
ture medium and then co-purification with exosomes.              from Pharmingen (San Diego, CA, U.S.A.); anti-CD20–FITC,
Indeed, unbound GaM IgM did not co-purify with                  anti-CD63–FITC, anti-CD40–PE and anti-(HLA DR)–PE,
                                                                were from Immunotech (Marseille, France); the anti-(MHC I)
exosomes (Figure 5A). This targeting of BCR–Ag                  mAb W6/32 was purified from hybridoma supernatants and
complexes to exosomes would thus explain the data               coupled to FITC.
reported in previous studies (Watts and Davidson,                  For confocal microscopy, we used as primary antibodies the
1988; Mitchell et al., 1995), where Ag – after endo-            anti-(MHC class II) mAb L243 produced in our laboratory,
                                                                and the anti-CD63 mAb (clone MOF11, kindly given by P.
cytosis through the BCR – could be recovered in the                                                                     e
                                                                Vincendeau, Laboratoire de Parasitologie, Universit´ Victor
extracellular medium in an intact TCA (trichloro-               Segalen Bordeaux II, Bordeaux, France), all diluted at 10 µg/ml.
acetic acid)-precipitable form, which might corres-             The polyclonal rabbit anti-(MHC class II) antibody kindly given
pond, in fact, to Ag being associated with exosomes             by H. Ploegh (Whitehead Institute for Biomedical Research,
                                                                Cambridge, MA, U.S.A.) was used at 1:1000. The secondary
produced by the cells. The presence of undegraded               antibodies were goat anti-(mouse Alexa Fluor® 488), goat anti-
BCR-targeted Ag in DOHH2 cells exosomes could                   (rabbit Alexa Fluor® 488) (Molecular Probes), and donkey
be explained by the recent finding that B-cells con-             anti-(mouse Cy3) (Jackson ImmunoResearch Laboratories).
tain only low levels of lysosomal proteases, therefore             For Western blots, the HC10 mAb specific for MHC class I,
                                                                the 1B5 mAb that recognizes MHC class II molecules and the
avoiding extensive Ag degradation, thus preserving              H68.4 mAb against the TfR were produced in our laboratory.
T-cell epitopes and promoting Ag presentation                   Anti-β-actin mAb was from Sigma (Saint-Quentin Fallavier,
(Delamarre et al., 2005). The recruitment of the BCR            France), anti-flotillin-1 mAb was from Pharmingen, polyclonal
and its bound ligand to exosomes could be due to their          rabbit anti-calnexin(N-Terminal) and rat mAb anti-gp96 were
preferential targeting by signals originating from the          from Stressgen Biotechnologies (Victoria, BC, Canada). The anti-
                                                                phosphotyrosine mAb 4G10 was from U.B.I. (Charlottesville,
BCR complex itself. Indeed, upon cross-linking,                 VA, U.S.A.), and the HRP-conjugated donkey anti-goat poly-
the BCR is recruited to plasma membrane lipid rafts             clonal antibody was from Santa Cruz Biotechnology (Santa Cruz,
containing the protein tyrosine kinase Lyn (Cheng               CA, U.S.A.).
et al., 1999). This initial step could explain the              Flow cytometry
delayed appearance of the BCR complexes in CD63-                Cells were washed and labelled according to standard proce-
positive compartments compared with MHC class II                dures. Cells were then analysed on a FACSCalibur® cytofluoro-
molecules (Figure 2). Moreover, since lipid micro-              meter (Becton Dickinson) using the CellQuest® software.
                                                                   Exosomes (3 µl of a preparation from unstimulated cells, see
domains containing Lyn are also present in B-cell               below) were incubated with aldehyde/sulphate latex beads (1 µl
exosomes (de Gassart et al., 2003), it is thus tempt-           per sample; Interfacial Dynamics, Portland, OR, U.S.A.) for
ing to speculate that lipid rafts might play a role in          15 min at room temperature in a final volume of 10 µl. BSA
targeting the BCR and its bound Ag to exosomes.                 (final concentration, 0.5 mg/ml) was added for a further 15 min.
                                                                Exosome-coated beads were then gently shaken in 1 ml of PBS
                                                                for 30 min, and the reaction was stopped by the addition of
                                                                100 mM glycine in PBS for another 30 min. After washing,
                                                                beads were labelled with the indicated mAb and analysed by
Materials and methods                                           FACS as above.
Cells
The human follicular lymphoma B-cell line DOHH2                 Immunofluorescence and confocal microscopy
(Kluin-Nelemans et al., 1991) obtained from DSMZ                DOHH2 cells (1 × 106 /sample) were coated in AIMV medium
(Braunschweig, Germany) was cultured in RPMI 1640 me-           (Gibco) with unlabelled- or Alexa Fluor® 488-conjugated GaH
dium (Gibco BRL, Cergy-Pontoise, France) supplemented with      IgG at saturating concentration (50 µg/ml) for 15 min on ice.
5% heat-inactivated pooled human serum, antibiotics (penicil-   Control cells were incubated with medium alone. After washing
lin 50 units/ml and streptomycin 50 µg/ml), 4 mM L-glutamine    in cold AIMV medium, cells were resuspended in 250 µl of this
and 1 mM sodium pyruvate.                                       medium and chased for various times in a water bath maintained



                                                                                   www.biolcell.org | Volume 98 (8) | Pages 491–501   499
                                                                                                                   P. Rialland and others


      at 37◦ C. Stimulation was stopped by the addition of cold PBS        Lowry assay (Bio-Rad, Richmond, CA, U.S.A.). Lysates were
      followed by centrifugation for 5 min at 350 g in a micro-            subjected to SDS/PAGE (10% gel) and the proteins were trans-
      fuge at 4◦ C. Cells were washed and allowed to adhere on to          ferred on to a PVDF membrane. After immunoblotting with
      poly(L-lysine)-coated glass coverslips for 1 h on ice. Cells were    the indicated antibodies, the reaction was revealed with an en-
      then fixed with 4% paraformaldehyde for 10 min at room tem-           hanced chemiluminescence system (ECL® ; Amersham Interna-
      perature, quenched with 100 mM glycine in PBS, and permeab-          tional). Densitometric analysis of gel bands was performed using
      ilized with PBS containing 0.05% saponin and 0.2% BSA. Cells         a laser densitometer and the ImageQuant® software (Amersham
      were incubated with the indicated primary antibody for 30 min,       Biosciences, Orsay, France).
      washed and labelled with the conjugated secondary antibody.
      After mounting on to glass slides with Mowiol, confocal sec-
      tions of 0.4 µm were acquired using a confocal laser-scanning        Acknowledgements
      microscope Leica TCS SP2 (Leica Lasertechnik, Heidelberg,            This work was supported by grants from the Institut
      Germany). Single 0.4 µm confocal sections are shown in the
      Figures.                                                                                e                     e
                                                                           National de la Sant´ et de la Recherche M´ dicale. We
                                                                                                             e
                                                                           thank Claire Hivroz, Clotilde Th´ ry and Ana-Maria
                                                                           Lennon-Dumenil for critical reading of the manu-
      Production and purification of exosomes
      DOHH2 cells (108 per sample) were washed twice with PBS              script and helpful discussions.
      and resuspended in AIMV medium at 5 × 106 cells/ml in 75 cm2
      culture flasks. The GaH IgG or GaM IgM antibodies (50 µg/ml)
      or the same volume of PBS vehicle were then added, and cells         References
      were incubated for 6 h at 37◦ C. In Figures 5(B) and 5(C), PBS       Brown, B.K., Li, C., Cheng, P.C. and Song, W. (1999) Trafficking of
      washed cells (108 cells per sample) were first incubated with the        the Igα/Igβ heterodimer with membrane Ig and bound antigen
                                                                              to the major histocompatibility complex class II peptide-loading
      indicated FITC-conjugated antibodies or with the same volume
                                                                              compartment. J. Biol. Chem. 274, 11439–11446
      of PBS in AIMV on ice for 15 min, then washed twice in cold PBS      Cheng, P.C., Dykstra, M.L., Mitchell, R.N. and Pierce, S.K. (1999)
      and chased for 6 h at 37◦ C in AIMV medium at 5 × 106 cells/            A role for lipid rafts in B cell antigen receptor signaling and antigen
      ml. To avoid contamination with protein aggregates present in           targeting. J. Exp. Med. 190, 1549–1560
      the medium, AIMV was ultracentrifuged for 18 h at 100 000 g          Clark, M.R., Massenburg, D., Siemasko, K., Hou, P. and Zhang, M.
      (depleted medium).                                                      (2004) B-cell antigen receptor signaling requirements for
         Exosomes were purified from the culture supernatant by dif-           targeting antigen to the MHC class II presentation pathway.
      ferential centrifugation as described previously (Wubbolts et al.,      Curr. Opin. Immunol. 16, 382–387
      2003). Cells were removed by centrifugation at 200 g for 10 min.     Clayton, A., Court, J., Navabi, H., Adams, M., Mason, M.D., Hobot,
      Supernatants were collected and sequentially centrifuged twice          J.A., Newman, G.R. and Jasani, B. (2001) Analysis of antigen
                                                                              presenting cell derived exosomes, based on immuno-magnetic
      for 10 min at 500 g, once for 15 min at 2000 g and once for
                                                                              isolation and flow cytometry. J. Immunol. Methods 247, 163–174
      30 min at 10 000 g to remove cellular debris. Exosomes were          Davidson, H.W., West, M.A. and Watts, C. (1990) Endocytosis,
      then pelleted by ultracentrifugation for 60 min at 70 000 g with        intracellular trafficking, and processing of membrane IgG and
      a SW28 rotor (Beckman Instruments, Palo Alto, CA, U.S.A.),              monovalent antigen/membrane IgG complexes in B lymphocytes.
      washed with PBS and re-pelleted at 70 000 g for 1 h. Finally,           J. Immunol. 144, 4101–4109
      exosome pellet was washed with 4 ml of PBS in a TLA 110 rotor        Davidson, H.W., Reid, P.A., Lanzavecchia, A. and Watts, C. (1991)
      (Beckman Instruments) for 1 h at 100 000 g, and resuspended in          Processed antigen binds to newly synthesized MHC class II
      20 µl of PBS which was considered as exosome preparation. To            molecules in antigen-specific B lymphocytes. Cell 67, 105–116
      assess the purity of exosome preparations, aliquots were fixed in     de Gassart, A., Geminard, C., Fevrier, B., Raposo, G. and Vidal, M.
                                                                                                                ´
                                                                              (2003) Lipid raft-associated protein sorting in exosomes.
      phosphate buffer containing 2% paraformaldehyde and 0.125%
                                                                              Blood 102, 4336–4344
      glutaraldehyde, placed on to formvar carbon-coated grids and         de Gassart, A., Geminard, C., Hoekstra, D. and Vidal, M. (2004)
      examined by electron microscopy using a CM120 Twin Phillips             Exosome secretion: the art of reutilizing nonrecycled proteins?
      electron microscope (Phillips Electronic Instruments, Mathway,          Traffic 5, 896–903
      NJ, U.S.A.).                                                         Delamarre, L., Pack, M., Chang, H., Mellman, I. and Trombetta, E.S.
                                                                              (2005) Differential lysosomal proteolysis in antigen-presenting cells
                                                                              determines antigen fate. Science 307, 1630–1634
      Western blot                                                         Denzer, K., Kleijmeer, M.J., Heijnen, H.F., Stoorvogel, W. and Geuze,
      Cells (2.5 × 106 /sample) were lysed in lysis buffer (20 mM             H.J. (2000) Exosome: from internal vesicle of the multivesicular
      Tris/HCl, pH 7.5, 1 mM EDTA, 140 mM NaCl, 1% Nonidet                    body to intercellular signaling device. J. Cell Sci. 113, 3365–3374
      P40, 50 units/ml aprotinin, 1 mM sodium orthovanadate and            Escola, J.M., Kleijmeer, M.J., Stoorvogel, W., Griffith, J.M., Yoshie, O.
      1 mM PMSF) at 4◦ C for 30 min. Post-nuclear supernatants were           and Geuze, H.J. (1998) Selective enrichment of tetraspan proteins
      then diluted in Laemmli sample buffer (500 mM Tris/HCl,                 on the internal vesicles of multivesicular endosomes and on
                                                                              exosomes secreted by human B-lymphocytes. J. Biol. Chem. 273,
      pH 6.8, 10% SDS, 10% glycerol, 5% 2-mercaptoethanol and
                                                                              20121–20127
      10% Bromophenol Blue) and boiled for 5 min. Exosomes                 Fevrier, B. and Raposo, G. (2004) Exosomes: endosomal-derived
                                                                            ´
      (4 µl/sample, taken from a 20 µl exosome preparation produced           vesicles shipping extracellular messages. Curr. Opin. Cell Biol. 16,
      from 1 × 108 cells as described above, and thus consisting of exo-      415–421
      somes produced by 2 × 107 cells) were lysed with 4 µl of lysis       Gandhi, M.K. and Marcus, R.E. (2005) Follicular lymphoma: time for a
      buffer and boiled. In both cases, proteins were quantified by            re-think? Blood Rev. 19, 165–178




500   C   Portland Press 2006 | www.biolcell.org
BCR targets antigen to lymphoma B-cell exosomes                                                            Research article
Granboulan, M., Lankar, D., Raposo, G., Bonnerot, C. and Hivroz, C.       Thery, C., Zitvogel, L. and Amigorena, S. (2002a) Exosomes:
                                                                             ´
  (2003) Phosphoinositide 3-kinase activation by Igβ controls               composition, biogenesis and function. Nat. Rev. Immunol. 2,
  de novo formation of an antigen-processing compartment.                   569–579
  J. Biol. Chem. 278, 4331–4338                                           Thery, C., Duban, L., Segura, E., Veron, P., Lantz, O. and
                                                                             ´
Kluin-Nelemans, H.C., Limpens, J., Meerabux, J., Beverstock, G.C.,          Amigorena, S. (2002b) Indirect activation of naive CD4+ T cells by
  Jansen, J.H., de Jong, D. and Kluin, P.M. (1991) A new                    dendritic cell-derived exosomes. Nat. Immunol. 3, 1156–1162
  non-Hodgkin’s B-cell line (DoHH2) with a chromosomal                    Tulp, A., Verwoerd, D., Dobberstein, B., Ploegh, H.L. and Pieters, J.
  translocation t(14;18)(q32;q21). Leukemia 5, 221–224                      (1994) Isolation and characterization of the intracellular MHC
Lankar, D., Vincent-Schneider, H., Briken, V., Yokozeki, T., Raposo, G.     class II compartment. Nature (London) 369, 120–126
  and Bonnerot, C. (2002) Dynamics of major histocompatibility            Vidal, M., Mangeat, P. and Hoekstra, D. (1997) Aggregation reroutes
  complex class II compartments during B cell receptor-mediated             molecules from a recycling to a vesicle-mediated secretion
  cell activation. J. Exp. Med. 195, 461–472                                pathway during reticulocyte maturation. J. Cell Sci. 110,
Mitchell, R.N., Barnes, K.A., Grupp, S.A., Sanchez, M., Misulovin, Z.,      1867–1877
  Nussenzweig, M.C. and Abbas, A.K. (1995) Intracellular targeting        Vincent-Schneider, H., Stumptner-Cuvelette, P., Lankar, D., Pain, S.,
  of antigens internalized by membrane immunoglobulin in B                  Raposo, G., Benaroch, P. and Bonnerot, C. (2002) Exosomes
  lymphocytes. J. Exp. Med. 181, 1705–1714                                  bearing HLA-DR1 molecules need dendritic cells to efficiently
Peters, P.J., Neefjes, J.J., Oorschot, V., Ploegh, H.L. and Geuze, H.J.     stimulate specific T cells. Int. Immunol. 14, 713–722
  (1991) Segregation of MHC class II molecules from MHC class I           Watts, C. and Davidson, H.W. (1988) Endocytosis and recycling
  molecules in the Golgi complex for transport to lysosomal                 of specific antigen by human B cell lines. EMBO J. 7,
  compartments. Nature (London) 349, 669–676                                1937–1945
Peters, P.J., Raposo, G., Neefjes, J.J., Oorschot, V., Leijendekker,      Weissman, A.M., Klausner, R.D., Rao, K. and Harford, J.B. (1986)
  R.L., Geuze, H.J. and Ploegh, H.L. (1995) Major histocompatibility        Exposure of K562 cells to anti-receptor monoclonal antibody
  complex class II compartments in human B lymphoblastoid cells             OKT9 results in rapid redistribution and enhanced degradation of
  are distinct from early endosomes. J. Exp. Med. 182, 325–334              the transferrin receptor. J. Cell Biol. 102, 951–958
Pinet, V., Vergelli, M., Martin, R., Bakke, O. and Long, E.O. (1995)      West, M.A., Lucocq, J.M. and Watts, C. (1994) Antigen processing
  Antigen presentation mediated by recycling of surface HLA-DR              and class II MHC peptide-loading compartments in human
  molecules. Nature (London) 375, 603–606                                   B-lymphoblastoid cells. Nature (London) 369, 147–151
Raposo, G., Nijman, H.W., Stoorvogel, W., Liejendekker, R., Harding,      Wolfers, J., Lozier, A., Raposo, G., Regnault, A., Thery, C.,
                                                                                                                                ´
  C.V., Melief, C.J. and Geuze, H.J. (1996) B lymphocytes secrete           Masurier, C., Flament, C., Pouzieux, S., Faure, F., Tursz, T.,
  antigen-presenting vesicles. J. Exp. Med. 183, 1161–1172                  Angevin, E., Amigorena, S. and Zitvogel, L. (2001) Tumor-derived
Savina, A., Vidal, M. and Colombo, M.I. (2002) The exosome pathway          exosomes are a source of shared tumor rejection antigens for CTL
  in K562 cells is regulated by Rab11. J. Cell Sci. 115, 2505–2515          cross-priming. Nat. Med. 7, 297–303
Savina, A., Furlan, M., Vidal, M. and Colombo, M.I. (2003) Exosome        Wubbolts, R., Leckie, R.S., Veenhuizen, P.T., Schwarzmann, G.,
  release is regulated by a calcium-dependent mechanism in K562             Mobius, W., Hoernschemeyer, J., Slot, J.W., Geuze, H.J. and
  cells. J. Biol. Chem. 278, 20083–20090                                    Stoorvogel, W. (2003) Proteomic and biochemical analyses of
Siemasko, K., Eisfelder, B.J., Williamson, E., Kabak, S. and Clark,         human B cell-derived exosomes. Potential implications for their
  M.R. (1998) Cutting edge: signals from the B lymphocyte antigen           function and multivesicular body formation. J. Biol. Chem. 278,
  receptor regulate MHC class II containing late endosomes.                 10963–10972
  J. Immunol. 160, 5203–5208                                              Xie, X., Wacker, H.H., Huang, S., Regitz, E., Preuss, K.D.,
Song, W., Cho, H., Cheng, P. and Pierce, S.K. (1995) Entry of B cell        Romeike, B., Parwaresch, R., Tiemann, M. and Pfreundschuh, M.
  antigen receptor and antigen into class II peptide-loading                (2003) Differential expression of cancer testis genes in histological
  compartment is independent of receptor cross-linking. J. Immunol.         subtypes of non-Hodgkin’s lymphomas. Clin. Cancer Res. 9,
  155, 4255–4263                                                            167–173



Received 1 March 2006/2 May 2006; accepted 5 May 2006
Published as Immediate Publication 5 May 2006, doi:10.1042/BC20060027




                                                                                               www.biolcell.org | Volume 98 (8) | Pages 491–501     501

				
DOCUMENT INFO
Shared By:
Categories:
Stats:
views:22
posted:2/7/2011
language:English
pages:11