Download Capture of plasma membrane fragments from target

From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
IMMUNOBIOLOGY
Capture of plasma membrane fragments from target cells by trogocytosis requires
signaling in T cells but not in B cells
Anne Aucher,1 Eddy Magdeleine,1 Etienne Joly,1 and Denis Hudrisier1,2
1Institut
de Pharmacologie et de Biologie Structurale (IPBS), Centre National de Recherche Scientifique (CNRS), Unité Mixte de Recherche 5089, Toulouse; and
Paul Sabatier, Toulouse, France
2Université
Upon recognition of their respective cellular partners, T and B cells acquire their
antigens by a process of membrane capture called trogocytosis. Here, we report
that various inhibitors of actin polymerization or of kinases involved in intracellular
signaling partially or fully inhibited trogocytosis by CD8ⴙ and CD4ⴙ T cells,
whereas they had no effect on trogocytosis by B cells. Similarly, trogocytosis by
T cells was inhibited at 4°C, whereas in
B cells it was independent of temperature, indicating that trogocytosis by
B cells does not rely on active processes.
By contrast, most inhibitors we tested
impaired both T-cell and B-cell activation.
The differential effect of inhibitors on
T-cell and B-cell trogocytosis was not due
to the higher affinity of the B-cell receptor
for its cognate antigen compared with the
affinity of the T-cell receptor for its own
antigen, but it correlated tightly with the
abilities of T cells and B cells to form
conjugates with their target cells in the
presence of inhibitors. Trogocytosis thus
has different requirements in different
cell types. Moreover, the capture of membrane antigen by B cells is identified as a
novel signaling-independent event of Bcell biology. (Blood. 2008;111:5621-5628)
© 2008 by The American Society of Hematology
Introduction
T lymphocytes and B lymphocytes are the 2 main cell types
responsible for the adaptive immune response in vertebrates.
Whereas B cells recognize native, unprocessed antigens using their
B-cell receptor (BCR), T cells recognize antigenic peptides bound
to major histocompatibility complex (MHC) molecules on the
surface of antigen-presenting cells (APCs) using their T-cell
receptor (TCR). Antigen recognition results in activation of the
lymphocytes, the acquisition of their effector functions, and their
cooperation with other cell types in the course of the adaptive
immune response.
Like many receptors on the cell surface, the antigen receptors
on the surface of lymphocytes are taken up into the cell by
endocytosis together with the antigens they bind.1,2 This is surprising because the antigens recognized by the TCR, the peptide-MHC
complexes, are integral membrane proteins, which do not normally
pass from one cell membrane to another. This observation, first
reported for CD8⫹ cytotoxic T lymphocytes (CTLs),2 was confirmed by several other studies of the 2 major classes of T cell:
CD8⫹ (CTL) and CD4⫹ (helper) T cells.3 Likewise, in an elegant
system developed by Batista et al, B cells have also been reported
to acquire antigens that are membrane-bound and to be able to
introduce them, like soluble antigens, in the presentation pathway.4
Our group has demonstrated that peptides bound to MHC
complexes translocate from the APC to the T cell in membrane
fragments that contain both lipids and many other membranebound proteins.5 We coined the term trogocytosis to describe this
process of unidirectional transfer of plasma membrane material
from target cells to effector cells of the immune system.6 Initially,
using well-characterized murine models of antigen-specific lymphocytes, we made this observation in CD8⫹ CTLs but, later, we
showed that CD4⫹ T cells and B cells also perform trogocytosis (ie,
they acquire membrane-anchored antigen in fragments of membrane).7,8 Trogocytosis has since also been reported for most other
hematopoietic cells including natural killer (NK) cells (see RodaNavarro and Reyburn9 for a review), dendritic cells,10 monocytes,11,12 and neutrophils,13 indicating that antigen recognition by
antigen receptors is not the only molecular trigger for trogocytosis.
Worthy of note, activated but not resting CD4⫹ and CD8⫹ T cells
were shown to acquire membrane patches from target cells in the
absence of antigen and independently of the TCR.14-16
Trogocytosis is now a well-recognized feature of T- and B-cell
biology, and numerous hypotheses propose that the process is
involved in the control of immune responses or in the spreading of
pathogens.3,6,17,18 The lack of information about the molecular
players involved in trogocytosis is a major obstacle to understanding the mechanism and the roles that the process may play in
different cell types. In comparison with T cells5,8,15,19-21 and with
NK cells (see Roda-Navarro and Reyburn9 for a review), much less
is known on the parameters governing B-cell trogocytosis.
The acquisition of antigen by B cells is a central process of
adaptive immunity that has been known for decades. Upon antigen
recognition, the B cell internalizes the antigen, processes it into
protein fragments, and presents these peptides bound to MHC class
II molecules on its own surface. This peptide-MHC complex is then
recognized by CD4⫹ helper T cells, which stimulate the B cell to
secrete antibodies (Abs) of higher affinity for their antigens and of
diversified biologic functions.1,22 To date, the acquisition of
membrane-bound antigens by B cells constitutes the sole unequivocal role for trogocytosis. In the course of a previous study
exploiting redirected trogocytosis to characterize what molecules
Submitted January 15, 2008; accepted March 21, 2008. Prepublished online as
Blood First Edition paper, April 1, 2008; DOI 10.1182/blood-2008-01-134155.
The publication costs of this article were defrayed in part by page charge
payment. Therefore, and solely to indicate this fact, this article is hereby
marked ‘‘advertisement’’ in accordance with 18 USC section 1734.
The online version of this article contains a data supplement.
© 2008 by The American Society of Hematology
BLOOD, 15 JUNE 2008 䡠 VOLUME 111, NUMBER 12
5621
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
5622
BLOOD, 15 JUNE 2008 䡠 VOLUME 111, NUMBER 12
AUCHER et al
Table 1. Cellular systems and stimuli used in this study
Redirected trogocytosis,
Target: P815 cells (FcRⴙ)
Antigen-specific trogocytosis
Target cells
Effector cells
Antigen recognized
Figures in
current paper
Triggering antibody
Figures in
current paper
OT-I CD8⫹ T cells
RMA or EL4 cells (H-2Kb⫹)
SIINFEKL peptide (OVA257-264)
1,2,4,S1
2C11 Anti-CD3 mAb
4A,B
DO11.10 CD4⫹ T cells
A20 cells (I-Ad⫹)
OVA323-339
1,4,S1
2C11 Anti-CD3 mAb
4A
MD4 B cells
J558L cells or HEK cells
Membrane-bound hen egg lysozyme
1-5,S1,S2
187.1 Anti-k mAb
4A,B
(mHEL) expressed by transfection
MD4 B cells
HEK transfected with mutated
Mutant K97A mHEL (low affinity)
4C,S2
mHELK97A
3-83 B cells
C3H splenocytes
H-2k (high affinity)
4D,E
3-83 B cells
B6 splenocytes
H-2b (low affinity)
4D,E
3-83 B cells
Balb/c splenocytes
H-2d (no affinity)
4D,E
1H11-34 T cell hybridoma
MD4 B cells (I-Ed⫹)
HEL107-116
5
could trigger the phenomenon, we observed certain differences
between T and B lymphocytes. Here, we compared antigentriggered trogocytosis in T and B lymphocytes with the goal to
understand whether differences between these 2 cell types may
help us to decipher the molecular mechanisms whereby fragments
of plasma membrane can transfer from one cell to another one.
Methods
Cell lines and mice
Effector cells originated from OT-I mice (CD8⫹ T cells specific for
OVA257-264 presented by H-2Kb), DO11.10 mice (CD4⫹ T cells specific
for OVA323-339 presented by I-Ad), or MD4 mice (B cells specific for hen
egg lysozyme [HEL]). Spleens from the 3-83 strain of transgenic mice
(containing B cells specific for H-2Kk) were shipped to us by Dr
Tybulewicz (London, United Kingdom). T cells (preactivated in vitro with
the specific antigen) or naive B cells were exposed to target cells in various
combinations as summarized in Table 1. The origin, characteristics, and
culture of the target cells and effector cells are detailed in Document S1
(available on the Blood website; see the Supplemental Materials link at the
top of the online article.)
Reagents and antibodies
Reagents and antibodies used in this study are reported in Document S1.
Target cell staining
For their subsequent use in trogocytosis assays, target cells were either
stained with a lipophilic dye (PKH67, PKH26, or DiI) or were cell surface
biotinylated, as described previously in detail.23 For their subsequent use in
conjugate formation assay, cells were sometimes labeled with carboxyfluorescein succinimidyl ester (CFSE). For this, cells (10 ⫻ 106/mL) in
complete culture medium were incubated for 5 minutes at 37°C with
2.5 ␮M CFSE and then washed 3 times in complete culture medium.
Trogocytosis assays
Trogocytosis experiments were performed as described previously.23 In
brief, after staining with a membrane dye, target cells were placed in
U-bottomed 96-well plates (0.5 ⫻ 106 cells/well in 200 ␮L final volume).
When T cells were used as effectors, target cells were incubated with the
indicated concentration of the appropriate peptide. After 1 hour at 37°C,
cells were washed 3 times with 200 ␮L culture medium. After the last wash,
cell pellets were resuspended in 100 ␮L medium containing 105 T or
B cells, centrifuged for 30 seconds at 160g to promote conjugate formation,
and then left at 37°C or, in some experiments, at 4°C for 1 hour. Conjugates
were then dissociated by washing cells twice in phosphate-buffered saline
containing 0.5 mM ethylenediaminetetraacetic acid (EDTA) and pipetting
them up and down thoroughly, before staining with a monoclonal antibody
(mAb) against CD8, CD4, or B220. Cells were then analyzed by flow
cytometry on a FACSCalibur (Becton Dickinson, Mountain View, CA).
Effector cells were gated positively according to their staining with
lineage-specific markers (CD8 for CTL, CD4 for T helper cells, and B220
for B cells). When PKH- or DiI-labeled target cells were used, trogocytosis
was measured directly by detecting these fluorescent markers on the
effectors. When biotinylated target cells were used, flow cytometric
detection of trogocytosis was performed by staining with fluorescent
streptavidin together with CD8⫹, CD4⫹, or B-cell markers. In some cases,
trogocytosis with MD4 B cells was performed in the presence of the
indicated concentrations of soluble hen egg lysozyme (sHEL). In experiments involving inhibitory drugs, T and B cells were pretreated at 37°C
before coincubation with their targets. All these drugs were left throughout
the assays. Conditions for the use of the different drugs are summarized in
Table 2. Redirected trogocytosis16 was performed similarly except that
P815 cells were used as targets and were mixed with effector cells that had
been preincubated with 5 ␮g/mL of the different unlabeled mAbs for
30 minutes at 4°C. Values used for our analyses were median fluorescence
Table 2. Drug treatments used in this study
Drug
Provider
Concentration used
(range)
Latrunculin B
Calbiochem
25 ␮M (1-300 ␮M)
PP2
Calbiochem
10 ␮M (0.1-100 ␮M)
Cytochalasin D
Sigma-Aldrich
10 ␮M
Pretreatment of
effector cells
2h
20 min
Figures in
current paper
1,2,4,5
1,2,4,S1
2h
1,4
1
Nocodazole
Sigma-Aldrich
100 nM
2h
Piceatannol
Calbiochem
100 ␮M
20 min
1
Wortmannin
Sigma-Aldrich
100 nM
20 min
1
EDTA
Sigma-Aldrich
2 mM
Calbiochem (San Diego, CA); Sigma-Aldrich (St Louis, MO).
None
2,4
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
BLOOD, 15 JUNE 2008 䡠 VOLUME 111, NUMBER 12
DIFFERERENTIAL TROGOCYTOSIS BY T AND B CELLS
5623
Figure 1. The differential effect of inhibitors and
temperature on T-cell and B-cell trogocytosis.
(A) DO11.10 CD4⫹ T cells (left panels), OT-I CD8⫹
T cells (middle panels), and MD4 B cells (right panels)
were exposed in the presence or absence of 25 ␮M
latrunculin B (Lat. B) to their respective biotinylated
target cells that either expressed, on their surface, the
appropriate antigen (open histograms) or not (closed
histograms). After 1 hour at 37°C, conjugates were
dissociated and lymphocytes were stained with fluorescent streptavidin to detect capture of biotinylated components, and with lineage-specific mAbs. Graphs show
the level of biotin staining on gated effector cells.
(B) Quantification of trogocytosis efficiency performed
by lymphocytes in the presence or absence of 25 ␮M
latrunculin B, 10 ␮M cytochalasin D (Cyt. D), 100 nM
nocodazole (Noco.), 10 ␮M PP2, 100 ␮M piceatannol
(Pice.), and 100 nM wortmannin (Wort.). Residual trogocytosis in the presence of the indicated inhibitors was
calculated using the formula described in Document
S1. DO11.10 CD4⫹ T cells were not tested in presence
of cytochalasin D and nocodazole. Bars represent the
standard deviations; n ⫽ 5 for T CD4⫹ cells and n ⬎ 6
for T CD8⫹ cells and B cells. Levels of statistical
significance were calculated using the Student t test;
**P ⬍ .01. (C) As in panel A, except that 10 ␮M PP2
was used instead of latrunculin B. The data are from
experiments in which all cells were treated in parallel
with the same aliquots of the indicated inhibitors.
(D) OT-I CD8⫹ T cells (squares) and MD4 B cells
(diamonds) were analyzed as in panel A except that a
range of concentrations of latrunculin B (top panel) or
PP2 (bottom panel) was used. Residual trogocytosis
was calculated as in panel B. Similar results were
obtained in 4 independent experiments. (E) As in panel
A except that cocultures were incubated at 37°C or at
4°C rather than with or without inhibitor. Note that for
experiments performed at 4°C, we used an effector to
target ratio of 15:1 to ensure that all B cells were in
contact with a target cell.
intensities (mfi). Residual trogocytosis in the presence of the indicated
inhibitors was calculated using the following formula: 100 ⫻ (mfi with
antigen ⫺ mfi without antigen, with drug)/(mfi with antigen ⫺ mfi
without antigen, without drug), where mfi is the median fluorescence
intensity.
phycoerythrin and selected magnetically using antiphycoerythrin microbeads (Miltenyi, Auburn, CA). Sorted cells (105) were then incubated at
37°C for 24 hours with 105 1H11-34 cells (T-cell hybridoma cells specific
for HEL107-116 presented by I-Ed). The levels of interleukin-2 (IL-2) in the
harvested culture supernatants were measured using an enzyme-linked
immunosorbent assay (ELISA) kit (BD Bioscience, San Jose, CA).
Conjugate formation
Conjugate formation was analyzed as described previously.24 Briefly,
5 ⫻ 105 CFSE-labeled target cells were incubated with 105 PKH26-labeled
T cells, or with anti-B220 mAb-stained B cells, in 200 ␮L culture medium
in 96-well U-bottom plates for 1 hour at 37°C. Cells were then fixed with
1% paraformaldehyde overnight at 4°C and analyzed by flow cytometry.
Conjugates were identified as being CFSE⫹ PKH26⫹ for T cells and CFSE⫹
B220⫹ for B cells. Formation of conjugates by redirected trogocytosis was
performed similarly except that CFSE-labeled P815 cells were used as
targets and were incubated with effector cells in the presence of 5 ␮g/mL of
the various unlabeled mAbs for 30 minutes at 4°C. In some assays, drugs
were used as described in Table 2. Percentages of conjugated lymphocytes
were calculated as 100 ⫻ (lymphocytes in conjugate/total number of
lymphocytes). For B cells, the total number corresponds to the number of
cells staining positively for B220, whereas for T cells, which are an almost
pure population after in vitro stimulation, we simply used the number of
PKH26-labeled cells.
Antigen presentation assays
In an initial step of antigen capture, 106 MD4 B cells, expressing an
HEL-specific IgM BCR, were coincubated for 1 hour at 37°C with 5 ⫻ 106
J558L target cells (either stable J558L transfectants expressing membranebound HEL [mHEL], or untransfected J558L as negative control, or J558L
in the presence of sHEL). All conditions were tested either with or without
latrunculin B (25 ␮M). B cells were then labeled with anti–B220-
Results
T cells require an active actin cytoskeleton and enzymatic
activities to perform trogocytosis but B cells do not
By making use of a novel approach in which trogocytosis is
triggered by Abs, we obtained data suggesting that, unlike T cells,
B cells undergo trogocytosis in the presence of drugs that block the
actin cytoskeleton.16 This approach, however, used nonphysiologic
stimulation of T and B cells with Abs and therefore did not provide
information regarding antigen-mediated trogocytosis. To explore
possible differences in trogocytosis triggered by antigen recognition by T cells and B cells, we investigated the effect of various
inhibitors on trogocytosis by murine T or B lymphocytes from
well-characterized transgenic strains. Ovalbumin (OVA)–specific
OT-I CD8⫹ T cells, DO11.10 CD4⫹ T cells, and hen egg lysozyme
(HEL)–specific MD4 B cells efficiently trogocytosed membrane
fragments from respective target cells (EL4, A20, and J558L cells)
when these cells expressed their cognate antigen at the plasma
membrane, as shown by the detection on lymphocytes of biotinylated determinants initially present in the membrane of target cells
(Figure 1A). In the presence of 25 ␮M latrunculin B, an inhibitor of
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
5624
AUCHER et al
BLOOD, 15 JUNE 2008 䡠 VOLUME 111, NUMBER 12
Figure 2. Formation of stable conjugates with target cells. (A) OT-I CD8⫹ T cells (labeled in red with the
fluorescent lipophilic dye PKH26) were mixed at 37°C
in the presence or absence of 25 ␮M latrunculin B with
EL4 target cells (labeled in green with CFSE) that had
either been pulsed (⫹pOVA) or not (⫺pOVA) with
OVA257-264 peptide. After 1 hour, cells were fixed and
analyzed by flow cytometry. The percentages of T cells
found conjugated to target cells indicated in the dot
plots were calculated as follows: 100 ⫻ [(PKH26⫹
CFSE⫹ events)/(total number of PKH26⫹ events)].
(B) Percentages of conjugates formed by OT-I cells
with their targets in the presence of the indicated
inhibitors, calculated as in panel A. Gray represents the
“background” level of conjugates formed with target
cells in the absence of antigen, and black represents
the level of conjugates formed with target cells in the
presence of antigen. Levels of statistical significance
were calculated using the Student t test; ***P ⬍ .001.
(C) As in panel A except that MD4 B cells were stained
with the B cell–specific B220 mAb before mixing them
with CFSE-labeled J558L or J558LmHEL cells. The
percentages of B cells found in conjugates with target
cells were calculated using the following formula:
100 ⫻ [(B220⫹ CFSE⫹ events)/(total number of B220⫹
events)]. (D) As in panel B but for MD4 B cells. Bars
represent standard deviation; n ⫽ 4.
actin polymerization, trogocytosis by CD8⫹ and CD4⫹ T cells was
fully inhibited, whereas trogocytosis by B cells was not (Figure
1A). Similar results were obtained with another actin-depolymerizing agent, cytochalasin D (Figure 1B). In most of our experiments,
the extent of trogocytosis by B cells was in fact increased by the
presence of these 2 inhibitors, although this effect was not
statistically significant for cytochalasin D (Figure 1B). Nocodazole, an inhibitor of microtubule polymerization, had only a
marginal effect on the extent of trogocytosis by CD8⫹ T cells
(which was not statistically significant), and had no detectable
effect on B-cell trogocytosis (Figure 1B).
We extended our study to inhibitors of the Src-, Syk-, and
phosphatidylinositol-3-kinase (PI3K) pathways (PP2, piceatannol,
and wortmannin, respectively), which target important enzymes in
TCR and BCR signaling and are all frequently used to block
activation of T and B cells. Confirming our previous studies,5,8,19
we found that PP2 markedly inhibited trogocytosis in both CD4⫹
and CD8⫹ T cells (Figure 1C). Wortmannin also inhibited trogocytosis by CD4 ⫹ T cells as well as by CD8⫹ T cells, albeit to a much
lesser extent (Figure 1B). Piceatannol inhibited trogocytosis by
CD4⫹ T cells and, in some but not all experiments, by CD8⫹ T cells
(Figure 1B). When we tested B cells, however, we again found that
none of these inhibitors had a significant effect on trogocytosis
(Figure 1B,C). No inhibition of trogocytosis by B cells was
observed even at concentrations of inhibitors that were much
higher than those described commonly in the literature (Figure 1D).
We can be sure that all the inhibitors were active because they all,
except for nocodazole, had an effect on at least one type of cell
among the CD4 T, CD8 T, and B cells we used; in the case of
nocodazole, we checked the activity of the drug by its effect on the
microtubule cytoskeleton on fibroblasts (not shown). In addition,
when treated with PP2 (Figure S1) or latrunculin B (not shown),
both T-cell activation and B-cell activation were inhibited, showing
that B cells are not intrinsically resistant to these inhibitors. These
data indicate that trogocytosis by T cells and by B cells has
different requirements for signaling: T cells require the actin
cytoskeleton and kinases such as Src-kinase, Syk-kinase, and PI3K
(with some differences between CD8⫹ and CD4⫹ T cells), whereas
these activities are all dispensable for B-cell trogocytosis. To
explore whether B-cell trogocytosis requires any enzymatic or
active processes, we performed the incubation between target and
effector cells at 4°C. As shown in Figure 1E, we found that, at 4°C,
trogocytosis by T cells was completely inhibited whereas, in
B cells, it was not. Our observation that the capture of membrane
fragments from APCs by B cells is not affected by a broad range of
inhibitors was quite unexpected and suggests that B-cell trogocytosis can occur independently of signaling enzymes or the cytoskeleton. For T cells, our study also broadens the panel of inhibitors of
trogocytosis described in previous studies.5,15,19,21
T cells, but not B cells, require active actin cytoskeleton and
enzymatic activities to form conjugates with target cells
To investigate the underlying mechanisms accounting for the
difference observed in T-cell and B-cell trogocytosis and since
transfer of membrane material has been shown to require direct
cell-cell contact, we measured the formation of cellular conjugates
between target cells and lymphocytes in the absence or presence of
inhibitors. We found that the percentage of OT-I CD8⫹ T cells
conjugated to EL4 target cells increased from 8.5% to 87.2% of
total OT-I cells when the antigen was added (Figure 2A). This
increase was abolished by latrunculin B (Figure 2A,B) and
markedly reduced by PP2 (Figure 2B). Similar results were
obtained with CD4⫹ T cells (not shown). In marked contrast, the
formation of conjugates between MD4 B cells and their target cells
was not inhibited by latrunculin B or PP2 (Figure 2C,D); on the
contrary, these inhibitors tended to increase conjugate formation,
although the differences were not statistically significant. Somewhat surprisingly to us, even when we treated lymphocytes with
2 mM EDTA, the B cells continued to form conjugates with their
target cells, whereas the formation of conjugates between T cells
and their targets, which is known to depend critically upon divalent
cations, was completely inhibited. The extent of the inhibitory
effect of these various treatments on the formation of conjugates
therefore correlates strongly with the degree of inhibition of
trogocytosis documented in Figure 1, indicating that conjugate
formation is a key step preceding trogocytosis and reinforcing the
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
BLOOD, 15 JUNE 2008 䡠 VOLUME 111, NUMBER 12
DIFFERERENTIAL TROGOCYTOSIS BY T AND B CELLS
5625
Figure 3. The role of the B-cell receptor–antigen interaction in B-cell trogocytosis. (A) MD4 B cells were incubated for 1 hour at 37°C with the following target cells labeled
with the fluorescent lipophilic dye PKH26:J558L (filled histogram), J558LmHEL (black line histogram), or J558LmHEL in the presence of 100 ␮g/mL sHEL (gray line histogram).
The B cells were then analyzed for PKH26 acquisition. Histograms show the levels of PKH26 fluorescence on B220⫹ B cells. (B) Coincubation of MD4 B cells with
PKH26-labeled J558LmHEL cells in the presence of increasing concentrations of sHEL for 1 hour at 37°C. Ordinate values correspond to the percentage of staining due to the
capture of PKH26 relative to the level of staining on the PKH26-labeled J558LmHEL target cells.
notion that trogocytosis occurs at the immunologic synapse formed
between a lymphocyte and its cellular partner.
The differential sensitivity of trogocytosis in T cells and B cells
to inhibitors is independent of the affinity of the stimulus
received by their antigen receptors
The presence of membrane antigen is the only critical requirement
we have thus far identified for trogocytosis by B cells. This
requirement for antigen was confirmed by the observation that
sHEL inhibited, in a dose-dependent manner, the trogocytosis
of membrane from J558LmHEL target cells by MD4 B cells
(Figure 3A,B).
One major difference between antigen recognition by T cells
and B cells is the affinity of binding between the antigen and its
receptor, which is typically much higher for the BCR (10⫺910⫺11 M) than for the TCR (10⫺6-10⫺4 M). It is possible, therefore,
that the higher interaction strength of B cells with their target cells,
compared with T cells, allows B cells to capture membrane
fragments even in the presence of inhibitors. To address the issue of
the strength of interaction between BCR and antigen on the
sensitivity of trogocytosis to signaling inhibitors, we used
3 different experimental systems.
First, we used a “redirected trogocytosis” assay,16 in which
trogocytosis is triggered by a mAb either against the BCR or
against the TCR/CD3 complex in the presence of target cells
expressing Ab Fc domain receptors (FcRs). The mAb bridges
between the T cell or B cell and the target cell by binding on the one
hand to the BCR or TCR and on the other hand to the target cell
FcRs. In this setup, the stimuli for both types of cells are conveyed
by molecular scaffolds of similar stability involving a high-affinity
interaction (that of either the anti-BCR or anti-TCR/CD3 mAb with
their respective antigens) and one of weaker affinity (between the
mAb and the FcR). Under these “redirected” conditions, trogocytosis by MD4 B cells was unaffected by latrunculin B, whereas
trogocytosis by both DO11.10 CD4⫹ T cells and OT-I CD8⫹ T cells
was inhibited by the drug (Figure 4A). Likewise, conjugate
formation between T cells and target cells was inhibited by
latrunculin B, PP2, or EDTA under these conditions, whereas these
inhibitors had no effect on the formation of conjugates between
B cells and target cells (Figure 4B). This reinforces our previous
finding that, when the same anti–MHC class I mAb was used to
trigger redirected trogocytosis in naive T and B cells, latrunculin
B and cytochalasin D also inhibited trogocytosis only by T cells
and not by B cells.16 A further advantage of using redirected
trogocytosis is that it rules out a possible role of the APC in
determining the passive or active nature of trogocytosis since the
presenting cell is the same in both T-cell and B-cell assays.
Second, we mutated the antigen recognized by MD4 B cells, in
order to decrease its affinity for the BCR, by introducing a Lys97
for Ala (mHELK97A) mutation in the mHEL antigen, which is
known to decrease of the affinity of HEL for the MD4 BCR from
Kd ⫽ 50 pM to Kd ⫽ 0.1 ␮M.26 Membrane capture triggered by
either the WT or mutant HEL, expressed at similar levels at the
surface of J558L transfectants (Figure S2), was unaffected by
treatment with PP2, by performing the incubation at 4°C (Figure
4C) or by treatment with latrunculin B (not shown).
Third, we used the 3-83 strain of mice, in which B cells express
a transgenic BCR that recognizes various MHC class I molecules
on target cells with either high affinity (H-2k), very low affinity
(H-2b), or no detectable reactivity (H-2d).25 As with mutated HEL,
the 3-83 B cells trogocytosed membrane from both low-affinity
H-2b and high-affinity H-2k target splenocytes, and this capture was
insensitive to PP2, latrunculin B, or cytochalasin D (Figure 4D,E).
Experiments using HEK cells transiently transfected with plasmids
driving the expression of either H-2Kk or H-2Kb gave results
similar to those obtained with splenocytes (not shown). This and
the preceding 2 experiments all support the conclusion that the
differential sensitivity of T-cell and B-cell trogocytosis to inhibitors
is not due to the difference in affinity of their respective antigen
receptors for their ligands.
Of note, B-cell early activation, as indicated by the upregulation of CD69 on the cell surface, was triggered with similar
efficiency by both forms of HEL, but fully inhibited by PP2 or
incubation at 4°C (not shown).
In both the MD4 and 3-83 systems, the amount of membrane
capture by B cells (assessed by capture of the fluorescent lipophilic
dye DiI) was noticeably decreased when target cells expressed the
low-affinity antigen compared with the high-affinity one (Figure
4C-E), in line with our previous study on T cells showing that
affinity does impact on the efficiency of trogocytosis.5
Antigen captured by B-cell trogocytosis in the presence of
enzyme inhibitors is available for subsequent processing and
T-cell activation
Finally, we evaluated the functional consequences of antigen
capture by trogocytosis realized in the presence of signaling
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
5626
AUCHER et al
BLOOD, 15 JUNE 2008 䡠 VOLUME 111, NUMBER 12
Figure 4. The differential sensitivity of T- and B-cell trogocytosis to inhibitors is not due to differences in affinities of the TCR and BCR for their respective ligands.
(A) DO11.10 CD4⫹ T cells (left panels), OT-I CD8⫹ T cells (middle panels), and MD4 B cells (right panels) were used in a redirected trogocytosis assay against PKH67-labeled
P815 target cells triggered by either the 2C11 anti-CD3 mAb (open histograms for T cells, and closed for B cells) or the anti-BCR ␬ chain mAb (open histograms for B cells, and
closed for T cells). The effector cells were treated with 25 ␮M latrunculin B before and during coculture with the P815 target cells (bottom panels) or left untreated (top panels).
Graphs show the levels of PKH67 fluorescence on gated effector cells. Similar results were obtained in 3 independent experiments. (B) The percentages of OT-I CD8⫹ T cells
(top panel) or MD4 B cells (bottom panel) found conjugated to P815 target cells in the absence of mAb (gray) or in the presence (black) of 2C11 anti-CD3 mAb (for OT-I T cells)
or anti-BCR ␬ chain mAb (for MD4 B cells) was calculated as in Figure 2. Bars represent standard deviations; n ⫽ 4. Levels of statistical significance were calculated using the
Student t test; **P ⬍ .01 and ***P ⬍ .001. (C) MD4 B cells were exposed to the following target cells labeled with the fluorescent lipophilic dye DiI: HEK (closed histograms),
HEKmHELWT (open histograms), and HEKmHELK97A (gray line histograms). Effector cells were either left untreated (top panels), or treated for 20 minutes at 37°C with 10 ␮M
PP2 (bottom panels) or placed at 4°C (middle, horizontal panels) before coculture with target cells. After 1 hour of coculture at 37°C (top and bottom panels) or at 4°C (middle
panels), we measured the capture of membrane components (DiI; left panels) and of mHEL (biotinylated F10.6.6 mAb ⫹ fluorescent streptavidin; right, vertical panels). Similar
results were obtained in 4 independent experiments. (D) Splenocytes from 3.83 mice were exposed for 1 hour at 37°C to PKH67-labeled splenocytes from Balb/c (no affinity,
left panel), C57/BL6 (weak affinity, middle panel), or C3H/He (high affinity, right panel) before analysis by flow cytometry. These respective affinities for the indicated H-2
antigen were reported in.25 B220⫹ and B220⫺ splenocytes (donor cells) (PKH67bright) fall within the right quadrants, while the B220⫹ and B220⫺ 3.83 splenocytes (recipient
cells) occupy the left quadrants. Numbers represent trogocytosis indexes (TIs) calculated as indicated below. (E) As in panel D except that the indicated inhibitors were added
during trogocytosis. Trogocytosis indexes were calculated as follows: mfi on recipient B220⫹ cell in the upper left quadrant (B cells)/mfi on recipient B220⫺ cells in the lower left
quadrant (non-B cells). Similar results were obtained in 3 independent experiments.
inhibitors. Indeed, although B cells captured similar quantities of
membrane material in the presence and in the absence of signaling
inhibitors, the mechanisms of capture were not necessarily the
same. If different mechanisms operate under the different conditions, we might expect that the acquired materials had different
fates within the cell. The capture of antigen by B cells is usually
followed by internalization, proteolytic processing, and presentation of the antigen, in association with MHC class II molecules, to
T cells.22 We used this assay to test whether the antigen-processing
pathway was also followed by antigen acquired in the presence of
the actin inhibitor latrunculin B (Figure 5). We first cocultured
MD4 B cells with target cells expressing (J558LmHEL) or not
expressing (J558L) membrane-bound HEL (mHEL) to trigger the
capture of membrane. The B cells were then incubated with a T-cell
hybridoma line 1H11-34, which produces IL-2 upon recognition of
the HEL107-116 peptide processed from mHEL and presented by
the I-Ed MHC class II molecule on the surface of the MD4 B cells.
Although the J558LmHEL target cells, which express MHC
molecules of the H-2b haplotype and lack MHC class II, could not
directly activate 1H11-34 T cells (not shown), the MD4 B cells
were separated magnetically from the J558LmHEL target cells
before incubation with the 1H11-34 T cells. This prevented antigen
acquisition from the target cells by B cells and its presentation to
T cells during the secondary coculture. In negative controls, no
IL-2 production was detected when T cells were cultured with MD4
B cells that had been exposed to J588L cells that did not express
mHEL (Figure 5). In positive controls, T cells recognized the
antigen and produced IL-2 when cultured with MD4 B cells that
were incubated with sHEL or that were exposed to J558LmHEL.
When latrunculin B was present during the coculture of MD4
B cells with J558LmHEL, IL-2 was produced with comparable
efficiency by T cells in the secondary coculture, indicating that
antigenic material captured by MD4 B cells in the presence of
latrunculin B was processed and presented to T cells. As a control
for the action of latrunculin B, we found that IL-2 secretion by
T cells was fully inhibited when the drug was present during both
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
BLOOD, 15 JUNE 2008 䡠 VOLUME 111, NUMBER 12
DIFFERERENTIAL TROGOCYTOSIS BY T AND B CELLS
5627
Figure 5. Antigenic material captured by B cells in the presence of enzyme inhibitors is available for subsequent processing and T-cell activation. MD4 B cells were
incubated at 37°C for 1 hour with either culture medium, sHEL, J558L, or J558LmHEL target cells, and in the presence or absence of 25 ␮M latrunculin B. MD4 B cells were
then selected magnetically (purity ⬎ 85%) before an overnight incubation at 37°C with 1H11-34 T-cell hybridoma. T-cell activation was measured by determining
concentrations of IL-2 in the supernatants by ELISA. Black and white bars correspond to 2 independent experiments.
the first and second cocultures (not shown). Thus, material captured
by B-cell trogocytosis in the presence of latrunculin B remains
available for subsequent processing and presentation to T cells,
suggesting that the mechanisms involved in the capture of the
antigen in the presence of this inhibitor are not dramatically
different from those that occur in untreated cells.
Discussion
The main conclusion we draw from the data presented here is that
trogocytosis has different requirements in T cells and in B cells.
Trogocytosis occurs in both cell types in response to stimulation by
antigen, yet it relies on an active process in T cells but not in
B cells. A secondary conclusion is that in B cells the initial
BCR-dependent events, such as capture of membrane-bound
antigen, do not depend upon signaling, whereas later events, such
as CD69 up-regulation, do require signaling.
We found that, in the case of B cells, trogocytosis clearly
depends upon the presence of antigen on the APC, but it came as a
surprise to find that trogocytosis occurs in the presence of inhibitors
of all the biologic pathways we tested (Figure 1). Perhaps our most
surprising observation was that, unlike T cells, B cells perform
trogocytosis even at 4°C. B-cell activation clearly requires
signaling,27 as we confirmed by our finding that some of the
inhibitors (and low temperature) markedly inhibited or blocked
CD69 up-regulation while not affecting trogocytosis (Figure S1).
Therefore, in B cells, antigen capture and activation can be
dissociated, whereas, in T cells, both events are sensitive to the
same inhibitory agents.
A potential explanation to this difference lies with our observation that the formation of conjugates was dramatically affected by
signaling inhibitors in T cells but not in B cells (Figure 2),
indicating a marked difference in the way B cells and T cells
interact with their targets. This conclusion is supported by the fact
that B cells seem much less dependent than T cells upon key actors
of the actin cytoskeleton mobilization28 or of the adhesion process
such as LFA-1.29 Although the BCR/antigen interaction is of much
higher strength than between a TCR and a peptide-MHC complex,
our results in 3 separate experimental systems clearly ruled out
that the differential sensitivity of trogocytosis by T cells and
B cells to signaling inhibitors could be simply explained by this
factor (Figure 4).
Our observation that the capture of membrane fragments from
APCs by B cells is not affected by several different enzyme
inhibitors was quite unexpected, yet, a careful review of the
literature identifies BCR signaling and antigen internalization as
mutually exclusive,30 a notion supported by the dissociation we
observed between trogocytosis and activation in B cells. In
addition, the literature provides several examples of signalingindependent events of B-cell biology including BCR endocytosis
with antigen (see Caballero et al31 and references therein), BCR
recruitment into lipid rafts,32 antigen capture by B cells,33 and, in
one study,33 but not in another one,26 the formation of immunologic
synapses between APCs and B cells. Our data now extend this list
of signaling-independent events in B-cell biology to include
membrane capture by trogocytosis. Thus, BCR signaling, although
critical for many events of B-cell biology, might not be necessary
for some of the earliest steps, including trogocytosis.
B cells acquire membrane-bound antigen by trogocytosis; they
process the antigen and present it to CD4⫹ T cells to subsequently
benefit from T-cell help.1,22 We demonstrated clearly that antigenic
materials acquired by B cells in the presence of latrunculin B are
available for processing and presentation to T cells (Figure 5), strongly
suggesting that the molecular mechanisms involved in trogocytosis are
not dramatically altered by the presence of this inhibitor.
How can we explain the occurrence of trogocytosis in the
absence of signaling in B cells? Recent evidence indicates that
dramatic morphologic and dynamic changes in the plasma membrane can be triggered in the absence of any signaling following
protein-lipid interactions.34 Similarly, in a system where no source
of energy is involved, lipid-lipid interactions between vesicles (a
potential vector of membrane fragments during trogocytosis) and
supported bilayers induce membrane lipid exchange that impacts
on the lipid symmetry in the recipient membranes as well as on the
adhesion/migration properties of the vesicle.35 Conceivably, recognition of either soluble or membrane-bound antigens (a unique
property of B cells) might trigger morphologic changes similar to
those reported in the in vitro systems mentioned above, thus
accounting for the few early, signaling-insensitive events of B-cell
biology. Conceptually, this property of B cells could relate to the
B cell–like system recently identified in lamprey where lymphocytes capture particulate materials using glycophosphatidyl inositol–
anchored VLRB receptors that are not connected to any known
signaling pathway.36
Our study provides a foundation from which to explore the
mechanism(s) of trogocytosis further. Because trogocytosis is
performed not only by lymphocytes but also by most hematopoietic
cells9-13 and is proposed to have important physiopathologic6,17 and
possibly therapeutic12 consequences, exploring the mechanisms of
trogocytosis is of a crucial importance and may open new ways to
manipulate the immune response.
Note added in proof: In a very recent paper, Quah et al report
that activated B cells can share their BCR with other B cells via
membrane transfer, and that this phenomenon is also insensitive to
a broad range of inhibitors, including incubation at 4°C (Quah BJ,
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
5628
BLOOD, 15 JUNE 2008 䡠 VOLUME 111, NUMBER 12
AUCHER et al
Barlow VP, McPhun V, Matthaei KI, Hulett MD, Parish CR.
Bystander B cells rapidly acquire antigen receptors from activated
B cells by membrane transfer. Proc Natl Acad Sci U S A.
2008;105:4259-4264).
Acknowledgments
We thank members of the I3 club at IPBS for their kind gift of
reagents and helpful suggestions on our work. We also thank
Facundo Batista for the kind gift of the J558L cells and the mHEL
transfectant, Victor Tybulewicz for providing us with 3-83 mouse
spleens, Gunter Haemmerling for the H-2Kk cDNA, and JeanCharles Guery for the 1H11-34 hybridoma. We thank Carol
Featherstone for scientific editing of our paper.
This work was supported by funds from the Agence Nationale
de Recherche/Recherche et Innovation en Biotechnologies
(TAAVAC project; E.J.), Institut des Technologies Avancées en Sciences
du Vivant (E.J.), Région Midi-Pyrénées (D.H.), and CNRS.
Authorship
Contribution: A.A. designed and performed most experiments,
analyzed the data, made the figures, and wrote the paper; E.M.
prepared molecular biology reagents; E.J. and D.H. coordinated the
project, participated in the design of the experiments, and revised
the paper.
Conflict-of-interest disclosure: The authors declare no competing financial interests.
Correspondence: Denis Hudrisier, Institut de Pharmacologie et
de Biologie Structurale, Centre National de Recherche Scientifique, Unité Mixe de Recherche 5089, 205 route de Narbonne,
31077 Toulouse cedex 4; e-mail: [email protected]; or Etienne Joly, Institut de Pharmacologie et de Biologie Structurale,
Centre National de Recherche Scientifique, Unité Mixe de Recherche 5089, 205 route de Narbonne, 31077 Toulouse cedex 4; e-mail:
[email protected].
References
1. Lanzavecchia A. Receptor-mediated antigen uptake and its effect on antigen presentation to
class II-restricted T lymphocytes. Ann Rev Immunol. 1990;8:773-793.
2. Huang JF, Yang Y, Sepulveda H, et al. TCRmediated internalization of peptide-MHC complexes acquired by T cells. Science. 1999;286:
952-954.
3. Wetzel S, Parker D. MHC transfer from APC to
T cells following antigen recognition. Crit Rev Immunol. 2006;26:1-21.
4. Batista FD, Iber D, Neuberger MS. B cells acquire
antigen from target cells after synapse formation.
Nature. 2001;411:489-494.
5. Hudrisier D, Riond J, Mazarguil H, Gairin JE, Joly
E. Cutting edge: CTLs rapidly capture membrane
fragments from target cells in a TCR signalingdependent manner. J Immunol. 2001;166:36453649.
6. Joly E, Hudrisier D. What is trogocytosis and
what is its purpose [letter]. Nat Immunol. 2003;4:
815.
7. Daubeuf S, Aucher A, Sampathkumar S, Preville
X, Yarema K, Hudrisier D. Chemical labels metabolically installed into the glycoconjugates of
the target cell surface can be used to track
lymphocyte/target cell interplay via trogocytosis:
comparisons with lipophilic dyes and biotin. Immunol Invest. 2007;36:687-712.
lymphoma and chronic lymphocytic leukemia
cells by THP-1 monocytes. J Immunol. 2006;176:
2600-2609.
13. Whale T, Beskorwayne T, Babiuk L, Griebel P.
Bovine polymorphonuclear cells passively acquire membrane lipids and integral membrane
proteins from apoptotic and necrotic cells. J Leukoc Biol. 2006;79:1226-1233.
14. Lemaoult J, Caumartin J, Daouya M, et al. Immune regulation by pretenders: cell-to-cell transfers of HLA-G make effector T cells act as regulatory cells. Blood. 2006;109:2040-2048.
15. Pardigon N, Takeda K, Saunier B, et al. CD8mediated intraepithelial lymphocyte snatching of
thymic leukemia MHC class Ib molecules in vitro
and in vivo. J Immunol. 2006;177:1590-1598.
16. Hudrisier D, Aucher A, Puaux AL, Bordier C, Joly
E. Capture of target cell membrane components
via trogocytosis is triggered by a selected set of
surface molecules on T or B cells. J Immunol.
2007;178:3637-3647.
27. Kurosaki T. Genetic analysis of B cell antigen receptor signaling. Annu Rev Immunol. 1999;17:
555-592.
28. Snapper SB, Rosen FS, Mizoguchi E, et al.
Wiskott-Aldrich syndrome protein-deficient mice
reveal a role for WASP in T but not B cell activation. Immunity. 1998;9:81-91.
18. Hudrisier D, Bongrand P. Intercellular transfer of
antigen-presenting cell determinants onto T cells:
molecular mechanisms and biological significance. FASEB J. 2002;16:477-486.
30. Hou P, Araujo E, Zhao T, et al. B Cell Antigen Receptor Signaling and Internalization Are Mutually
Exclusive Events. PLoS Biol. 2006;4:1147-1158.
9. Roda-Navarro P, Reyburn HT. Intercellular protein
transfer at the NK cell immune synapse: mechanisms and physiological significance. FASEB J.
2007;21:1636-1646.
20. Hwang I, Huang JF, Kishimoto H, et al. T cells can
use either T cell receptor or CD28 receptors to
absorb and internalize cell surface molecules derived from antigen-presenting cells. J Exp Med.
2000;191:1137-1148.
12. Beum PV, Kennedy AD, Williams ME, Lindorfer
MA, Taylor RP. The shaving reaction: rituximab/
CD20 complexes are removed from mantle cell
26. Fleire SJ, Goldman JP, Carrasco YR, Weber M,
Bray D, Batista FD. B cell ligand discrimination
through a spreading and contraction response.
Science. 2006;312:738-741.
29. Berlin-Rufenach C, Otto F, Mathies M, et al. Lymphocyte migration in lymphocyte functionassociated antigen (LFA)-1-deficient mice. J Exp
Med. 1999;189:1467-1478.
19. Hudrisier D, Riond J, Garidou L, Duthoit C, Joly
E. T cell activation correlates with an increased
proportion of antigen among the materials acquired from target cells. Eur J Immunol. 2005;35:
2284-2294.
11. Xu H, Dhanireddy KK, Kirk AD. Human monocytes as intermediaries between allogeneic endothelial cells and allospecific T cells: a role for direct scavenger receptor-mediated endothelial
membrane uptake in the initiation of alloimmunity.
J Immunol. 2006;176:750-761.
25. Lang J, Jackson M, Teyton L, Brunmark A, Kane
K, Nemazee D. B cells are exquisitely sensitive to
central tolerance and receptor editing induced by
ultralow affinity, membrane-bound antigen. J Exp
Med. 1996;184:1685-1697.
17. Davis DM. Intercellular transfer of cell-surface
proteins is common and can affect many stages
of an immune response. Nat Rev Immunol. 2007;
7:238-243.
8. Puaux A, Campanaud J, Salles A, et al. A very
rapid and simple assay based on trogocytosis to
detect and measure specific T and B cells reactivity by flow cytometry. Eur J Immunol. 2006;36:
779-788.
10. Harshyne L, Watkins S, Gambotto A, BarrattBoyes S. Dendritic cells acquire antigens from
live cells for cross-presentation to CTL. J Immunol. 2001;166:3717-3723.
D. The activating NKG2D ligand MHC class Irelated chain A transfers from target cells to NK
cells in a manner that allows functional consequences. J Immunol. 2007;178:3418-3426.
21. Hwang I, Sprent J. Role of actin cytoskeleton in
T cell absorption and internalization of ligands
from APC. J Immunol. 2001;166:5099-5107.
22. Lanzavecchia A. Antigen-specific interaction between T and B cells. Nature. 1985;314:537-539.
23. Daubeuf S, Puaux A, Joly E, Hudrisier D. A
simple trogocytosis-based method to detect,
quantify, characterize and purify antigen-specific
live lymphocytes by flow cytometry, via their capture of membrane fragments from antigenpresenting cells. Nat Protoc. 2007;1:2536-2542.
24. McCann F, Eissmann P, Onfelt B, Leung R, Davis
31. Caballero A, Katkere B, Wen XY, Drake L, Nashar
TO, Drake JR. Functional and structural requirements for the internalization of distinct BCRligand complexes. Eur J Immunol. 2006;36:31313145.
32. Cheng PC, Brown BK, Song W, Pierce SK. Translocation of the B cell antigen receptor into lipid
rafts reveals a novel step in signaling. J Immunol.
2001;166:3693-3701.
33. Iber D. Formation of the B cell synapse: retention
or recruitment? Cell Mol Life Sci. 2005;62:206213.
34. Romer W, Berland L, Chambon V, et al. Shiga
toxin induces tubular membrane invaginations for
its uptake into cells. Nature. 2007;450:670-675.
35. Solon J, Streicher P, Richter R, Brochard-Wyart F,
Bassereau P. Vesicles surfing on a lipid bilayer:
self-induced haptotactic motion. Proc Natl Acad
Sci U S A. 2006;103:12382-12387.
36. Alder M, Herrin B, Sadlonova A, et al. Antibody
responses of variable lymphocyte receptors in the
lamprey. Nat Immunol. 2008;9:319-327.
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
2008 111: 5621-5628
doi:10.1182/blood-2008-01-134155 originally published
online April 1, 2008
Capture of plasma membrane fragments from target cells by
trogocytosis requires signaling in T cells but not in B cells
Anne Aucher, Eddy Magdeleine, Etienne Joly and Denis Hudrisier
Updated information and services can be found at:
http://www.bloodjournal.org/content/111/12/5621.full.html
Articles on similar topics can be found in the following Blood collections
Immunobiology (5489 articles)
Information about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests
Information about ordering reprints may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#reprints
Information about subscriptions and ASH membership may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtml
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American Society
of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.