From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
CD38-Mediated Growth Suppression of B-Cell Progenitors Requires
Activation of Phosphatidylinositol 3-Kinase and Involves Its Association
With the Protein Product of the c-cbl Proto-Oncogene
By Akira Kitanaka, Chikako Ito, Hikari Nishigaki, and Dario Campana
The signaling pathways that arrest the cell cycle and trigger
cell death are only partially known. Dimerization of CD38, a
45-kD transmembrane type II glycoprotein highly expressed
in immature B cells, inhibits cell growth and causes
apoptosis in normal and leukemic B-cell progenitors, but the
molecular mechanisms underlying these cellular responses
are unknown. In the present study, we found that CD38 ligation in the immature B-cell lines 380, REH, and RS4;ll
caused rapid tyrosine phosphorylation of the protein product of the proto-oncogene c-cbl. Dimerization of CD38 was
accompanied by the association of cb/with the p85 subunit
of phosphatidylinositol 3-kinase (PI 3-K), resulting in markedly increased PI 3-K activity in antiphosphotyrosine and
anti-cbl immunoprecipitates. Wortmannin and LY294002,
two potent inhibitors of PI 3-K, rescued immature B cells
from CD38-mediated growth suppression. This effect was
observed not only in model B a l l lines, but also in cultures
of leukemic lymphoblasts from patients, and in normal bone
marrow B-cell progenitors as well. Concentrations of inhibitors that reversed cellular responses t o CD38 significantly
decreased PI 3-K activity. By contrast, rapamycin, a p70 S6kinase inhibitor, did not rescue immature B cells from CD38mediated suppression. These results suggest that PI 3-K
activity is essential for CD38-mediated inhibition of lymphopoiesis and that cbl and PI 3-K are regulatory molecules
whose activation can result in suppression of cell proliferation and apoptosis in immature lymphoid cells.
0 1996b y The American Society of Hematology.
C
induced protein tyrosine kinase activity, including a prominent 120-kD tyrosine-phosphorylated protein, remained to
be identified. In addition, the link between biochemical
events and cellular responses triggered by CD38 dimerization had not been proven.
In the present study, we further investigated the signaling
pathways used by CD38 and attempted to establish a causeand-effect relationship between the biochemical and cellular
consequences of CD38 ligation. We first identified the 120kD protein phosphorylated after CD38 ligation as cbl, a
proto-oncogene product that, in mature lymphoid cells, is
phosphorylated after engagement of the B- and T-cell antigen receptor^.'^,'^ Second, we observed that CD38-mediated
phosphorylation of cbl is accompanied by its association
with PI 3-K. Third, we found that selective inhibitors of
mammalian PI 3-K activity, such as wortmannin and
LY294002, rescued normal and leukemic immature B
lymphoid cells from CD38-mediated growth suppression.
D38 IS A 45-kD transmembrane glycoprotein with the
N-terminus in its short cytoplasmic domain.’,’ CD38
is expressed by many cell types, including lymphoid progenitors and activated lymphocyte^.^.^ CD38 ligation has stimulatory effects on mature lymphocytes,h but inhibits cell growth
and induces apoptosis in B-cell precursor^.^ The molecular
mechanisms underlying CD38-mediated growth suppression
are still unclear. The structural homology between CD38
and adenosine-diphosphate ribosyl (ADPR)-cyclase had suggested that the function of CD38 might be related to the
production of cyclic ADPR, a calcium-mobilizing agent.8
However, CD38 appears to exert its inhibitory effects independently of its enzymatic activity in B-lymphoid progenitor~.~
We previously found that CD38 ligation with specific antibodies in immature B cells initiates protein tyrosine kinase
activity, which results in the rapid and transient tyrosine
phosphorylation of several intracellular subtrates, including
syk and PLC-Y.~
CD38 dimerization also promotes the association of tyrosine-phosphorylated proteins with the p85 regulatory subunit of phosphatidylinositol 3-kinase (PI 3-K),’ a
lipid kinase that phosphorylates phosphatidylinositols on the
D3 position of the inositol ring, and appears to regulate
diverse cellular functions such as proliferation, differentiation, and apopto~is.’~-’~
However, other substrates of CD38From the Department of Hematology-Oncology, St Jude Children’s Research Hospital, Memphis, TN: and the Universily of Tennessee College of Medicine, Memphis, TN.
Submitted January 26, 1996; accepted March 14, 1996.
Supported by Grants No. ROI-CA58297 and P30-CA21765 from
the National Cancer Institute and by the American Lebanese Syrian
Associated Charities (ALSAC}.
Address reprint requests to Akira Kitanaka, MD, PhD, Department of Hematology-Oncology, St Jude Children’s Research Hospital, 332 N Lauderdale, Memphis, TN 38105-2794.
The publication costs of this article were defrayed in part by page
charge payment. This article must therefore be hereby marked
“advertisement” in accordance with 18 U.S.C. section 1734 solely to
indicate this fact.
0 1996 by The American Society of Hematology.
0006-4971/96/8802-0O9$3.00/0
590
MATERIALS AND METHODS
Antibodies and reagents. Monoclonal anti-CD38 antibody T16
(IgG1) was purchased from Immunotech Inc (Westbrook, ME).’,’
Anti-CD38 antibody and unreactive control IgGl (Becton Dickinson,
San Jose, CA) were dialyzed in phosphate-buffered saline (PBS),
sterile-filtered, and used at 2 to 5 pg/mL. Monoclonal antibody
(MoAb) to phosphotyrosine (4G10) was purchased from UBI (Lake
Placid, NY). MoAb to phosphotyrosine (PY20) and polyclonal antiPI 3-K p85 were from Transduction Laboratories (Lexington, KY).
Polyclonal rabbit antiserum to cbl was purchased from Santa Cruz
Biotechnology (Santa Cruz, CA). Recombinant human interleukin3 (IL-3), IL-7, and stem cell factor (SCF) were from R & D Systems,
Inc (Minneapolis, MN). Wortmannin was purchased from Sigma (St
Louis, MO). 2-(4-morpholinyl)-8-phenyl-4H-l-benzopyran-4-one
(also known as LY294002) was synthesized as described.“ Rapamycin was from LC Laboratories (Wobum, MA). Wortmannin,
LY294002, and rapamycin were dissolved in dimethyl sulfoxide
(DMSO) and stored at -20°C. All other reagents were purchased
from Sigma.
Cells. Human immature B-cell lines 380, REH, and RS4; I I
were available in our institutional cell bank. All lines express CD19
and CD38 and lack surface Ig. 380 and REH also express CDIO,
whereas RS4; 1 I cells lack this antigen. Cell lines were maintained
Blood, VOI 88, N O 2 (July 15). 1996 pp 590-598
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
591
SIGNALING PATHWAYS OF CD38-MEDIATED B-CELL GROWTH SUPPRESSION
in RPMI-1640 (Whittaker Bioproducts Inc, Walkersville, MD) with
10%fetal calf serum (FCS; Whittaker), L-glutamine, and antibiotics.
Bone marrow samples were taken, with informed consent from
the patients and Institutional Review Board approval, from four
healthy bone marrow transplant donors (12 to 24 years of age; median, 14 years) and from five patients with newly diagnosed Blineage acute lymphoblastic leukemia (ALL; 3 to 15 years of age;
median, 6 years). In each case of ALL, greater than 80% of the blasts
were positive for CD19, CD22, CD38, terminal deoxynucleotidyl
transferase, and class I1 antigens and negative for surface Ig. Mononucleated cells were washed three times in PBS and once in AIMV medium (GIBCO, Grand Island, NY). Normal CD19+ bone marrow cells were purified by use of CD19-immunomagnetic beads
(Dynal, Oslo, Norway). Cells were detached from the beads using
a goat antiserum to mouse Ig Fab (DETACHaBEAD; Dynal). T
cells were removed from suspensions of leukemic lymphoblasts by
using a mixture of magnetic beads conjugated to CD4 and CD8
antibodies (Dynal). Cell sorting procedures yielded 90% to 99%
pure CD19’ cells. The cells’ viability consistently exceeded 90%
by trypan-blue dye exclusion.
To obtain bone marrow stromal cells, we collected mononucleated
cells from normal marrow donors. The cells were separated as described above and washed three times in RPMI-1640. Stromal layers
were prepared in flat-bottomed 96-well plates (Costar, Cambridge,
MA) and fed with RPMI-1640,10% FCS, and
mol& hydrocortisone, as previously de~cribed.’~’’~’~
Cell culture studies andflow cytometry. Before each experiment,
we removed the media from cultured stromal cells and washed the
adherent cells seven times with RPMI-I640 to fully remove hydrocortisone. Leukemic and normal B cells were resuspended in serumfree AIM-V medium and cell lines were resuspended in RPMI-1640
+ 10% FCS. Two hundred microliters of the cell suspension (0.1
to 1.5 cells X 106/mL)were then seeded onto marrow stromal cells.
In cultures with wortmannin, LY294002, or rapamycin, control wells
were prepared with their diluent DMSO at the maximum concentration used (0.05%). DMSO at this concentration had no discernible
effect on cell growth. All cell cultures were incubated at 37°C in
5% C 0 2 with 90% humidity. At the termination of cultures, cells
were harvested by vigorous pipetting, suspended in PBS, and passed
through a 19-gauge needle to disrupt clumps. Microscopic examination of the plates ensured that all cells in the wells (ie, adherent and
nonadherent) were recovered. Viable cells in culture were enumerated by flow cytometry, as previously d e ~ c r i b e d . ~ . ’ ~ . ’ ~
Immunoprecipitation, sodium dodecyl sulfate-polyacrylamide gel
electrophoresis (SDS-PAGE), and Western blotting. Immunoprecipitation was performed essentially as described in Silvennoinen et
Briefly, after exposure to anti-CD38 antibody or control Ig, cells
were lysed for 20 minutes in 1 mL of ice-cold lysis buffer (50 mmoV
L Tris [pH 7.51, 150 mmoVL NaCI, 1% [voUvol] Triton X-100, 5
pglmL aprotinin, 1 mmoVL phenylmethylsulfonyl fluoride, 1 mmoV
L EDTA, and 1 mmoVL Na3V04) and centrifuged at 20,OOOg for
20 minutes at 4°C. Supernatants were precleared with 1 hour of
protein-A-Sepharose treatment (20 pL of 50% slurry). Antibodies
were then added to the cleared lysates, which were incubated at 4°C
for 1 to 2 hours. The immune complexes were collected by using
protein A-Sepharose. In some experiments, supernatants precleared
with anti-cbl and protein A-Sepharose were used instead of the
immunoprecipitates.
For SDS-PAGE, immunoprecipitates were resuspended in
Laemmli’s sample buffer (10% [voUvol] glycerol, 1 mmoVL dithiothreitol, 1 % [wt/voll SDS, 50 mmoVL Tris-HC1 [pH 6.81, and
0.002% [wt/voll bromophenol blue) and subjected to SDS-PAGE
(7.5% acrylamide).’ After transfer, nitrocellulose filters were incubated first in 5% nonfat dry milk in TBS-T (20 mmoVL Tris [pH
7.61, 137 mmom NaCI, 0.1% Tween 20) for 2 hours and then with
primary antibodies for 1 hour. After washing in TBS-T, the filters
were incubated for 1 hour with horseradish peroxidase-conjugated
sheep antimouse Ig or donkey antirabbit Ig (Amersham Corp, Arlington Heights, IL). The filters were then washed, incubated with enhanced chemiluminescence detection reagents (Amersham), and exposed to Kodak BioMax MR film (Eastman Kodak, Rochester, NY).
For reprobing, the filters were stripped and then reblocked, washed,
and r e p r ~ b e dAll
. ~ experiments were repeated at least three times.
PI 3-K assay. PI 3-K activity was assessed as described by
Whitman et al,” with some modifications? Briefly, the immunoprecipitates were incubated for 15 minutes at 30°C in 50 pL of 20
mmoVL Tris, pH 7.5, 100 m o l & NaCI, 5 m o V L magnesium
chloride, 0.5 mmoVL EGTA, 0.5 mmoVL p-nitrophenylphosphate,
0.2 mmoVL adenosine, 0.5 mg/mL L-a-phosphatidylinositol, 0.5
mg/mL L-a-phosphatidylserine, and 50 pmoVL ATP containing 2
pCi of [y-’*P]ATP.Phospholipids were separated by thin-layer chromatography and autoradiographed. The radioactivity of phospholipids was quantitated by PhosphorImager (Molecular Dynamics, Sunnyvale, CA) with ImageQuant software. All experiments were
repeated at least three times.
RESULTS
CD38-mediated tyrosine phosphorylation of cbl and its
association with PI 3-K p85. CD38 ligation in immature
B cells induces tyrosine phosphorylation of several proteins,
including a prominent one with a mass of approximately 120
kD.9 A protein of similar molecular mass, phosphorylated
after cross-linking of B- and T-cell antigen receptors, has
recently been identified as the protein product of the protooncogene c - ~ b l . ’ ~To
, ’ ~determine whether the 120-kD protein phosphorylated by CD38 dimerization corresponded to
cbl, we ligated CD38 with the MoAb T16 in the immature
B-cell line RS4; 11 and assessed tyrosine phosphorylation in
Western blots after removing the cbl protein from the cell
lysates with a anti-cbl antibody. This procedure caused a
marked decrease in the 120-kD band, without differences in
the appearance of the remaining proteins tyrosine phosphorylated after CD38 ligation (Fig l). To confirm that cbl participated in the CD38-mediated signaling, we directly examined
cbl tyrosine phosphorylation in three immature B-cell lines,
380, REH, and RS4; 11. Exposure to anti-CD38, but not to
isotype-matched control antibody, caused marked tyrosine
phosphorylation of cbl in the three lines (Fig 2).
Previous studies had shown that cbl can associate with
several other intracellular molecules after phosphorylation.14,15,2021In both B and T cells, for example, engagement
of the antigen receptor induces an association between cbl
and the p85 subunit of PI 3-K.’4.2’.23.24
To assess whether
CD38 ligation in immature B cells that lack sIg induced PI
3-K association with cbl, we determined whether PI 3-K was
present in the cbl immunoprecipitates obtained from 380,
REH, and RS4; 11 cells exposed to anti-CD38 using an antiPI 3-K p85 antibody. In all three lines, the amount of PI 3K p85 that coprecipitated with cbl markedly increased after
CD38 ligation (Fig 2). These results indicate that CD38 dimerization causes tyrosine phosphorylation of cbl and its
association with PI 3-K p85.
PI 3-K activity afer CD38 ligation. We found previously that CD38 dimerization causes association of PI 3K p85 with other tyrosine phosphorylated proteins in the
RS4; 11 and 380 immature B-cell lines.’ As expected, CD38
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
KITANAKA ET AL
592
- + +
- - +
CD38
Cbl Preclear
- 205
Probe:
- 121
- 86
anti-pTyr
- 51
-121
anti-Cbl
Fig 1. Identification of the 120-kD protein tyrosine phosphorylated
by CD38-ligation as cbl. RS4;ll immature B cells were incubated
with or without anti-CD38 antibody for 5 minutes. Anti-cbl antibody
and protein A-Sepharose were used t o preclear the cell lysates; protein A-Sepharose alone was used as a control. Cell lysates were then
separated by SDS-PAGE and transferred t o a nitrocellulose membrane. The membrane was probed with antiphosphotyrosine antibody (pTyr; upper panel) and then stripped and reprobed with anticbl antibody (lower panel). Molecular mass markers (in kilodaltons)
are indicated. Preclearing with anti-cbl markedly reduced the 120-kD
band (upper panel). The efficient depletion of cbl by the preclearing
procedure is shown in the lower panel.
REH
380
I
II
Qo
cr)
m
ligation in 380, REH. and RS4; 1 I cells also caused a striking
increase of PI 3-K activity in antiphosphotyrosine immunoprecipitates (Fig 3A). whereas no increase was detected
when an isotype-matched nonreactive antibody was used
instead. The increase in precipitable PI 3-K activity was
detectable after 1 minute of CD38 ligation and appeared to
reach maximal levels within 5 minutes, decreasing to less
than 50% of maximum within 30 minutes of exposure to
anti-CD38 (Fig 3B). The observed production of phospholipids was most likely due to specific PI 3-K activity in the
immunoprecipitates. because the assays were performed in
the presence of 0.2 mmol/L adenosine, which inhibits PI
4-kina~e.'~
Moreover. the production of phospholipids was
completely inhibited by the addition of 0.1% Triton X-100
or 100 nmol/L wortmannin into the reaction mixture (data
not shown), conditions that inhibit PI 3-K but not PI 4-K.'X-"'
The association between chl and PI 3-K p85 induced by
CD38 ligation implied that this might be accompanied by
increased PI 3-K activity in the chl-p85 complexes. A
marked increase in PI 3-K activity was indeed detected in
the chl immunoprecipitates from 380, REH, and RS4: 1 I
cells exposed to anti-CD38 (Fig 4).
Ligation of CD38 also increased PI 3-K activity in immunoprecipitates
obtained with anti-PI 3-K p85 antibody, although
.
.
the increase was significantly lower than that observed in immunoprecipitates obtained with antiphosphotyrosine or anti-chl
antibodies. After exposure of 380 cells to anti-CD38. PI 3-K
activity increased by 30% to 70% in anti-p8S immunoprecipitates, as compared with the 1.500% increase seen in anti-chl
and the 2,200% increase seen in antiphosphotyrosine immunoprecipitates. Taken together, these data indicate that dimerization of CD38 activates PI 3-K activity and strongly promotes its
association with chl and other tyrosine phosphorylated proteins.
Although some surface receptor proteins have been shown
to associate with PI 3-K after ligand stimulation,"'." we
could detect neither increased PI 3-K activity nor the presence of tyrosine phosphorylated proteins in anti-CD38 antibody immunoprecipitates, even under mild detergent conditions (ie, using a lysis buffer containing 0.1% NP-40 or
0.5% digitonin; data not shown). Thus, the molecular links
between CD38 and PI 3-K are still unknown.
RS4:ll
II
m
I
Q
- 0
Probe:
- .---
- 121
anti-pTyr
anti-Cbl
anti-PI3-K (p85)
.
i
- 121
- 86
Fig 2. Ligation of CD38 induces tyrosine phosphorylation of cbland its association with PI 3-K p85.
Immature B cells (380, REH, and RS4;ll) were incubated with anti-CD38 antibody or control lgGl for 5
minutes. Cell lysates were immunoprecipitated with
anti-cbl antibody, separated by SDS-PAGE, and
transferred t o a nitrocellulose membrane. The membrane was probed with antiphosphotyrosine antibody (pTyr; upper panel) and then stripped and reprobed with anti-cbl antibody (middle panel). After
further stripping, the membrane was reprobed with
anti-PI 3-K p85 antibody (lower panel). Molecular
mass markers (in kilodaltons) are indicated.
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
SIGNALING PATHWAYS OF CD38-MEDIATED B-CELL GROWTH SUPPRESSION
IgG1
CD38
A
593
I
380
CD38
IgG1
380
REH
REH
RS4;11
RS4;11
B
2o
Fig 4. Ligation of CD38 increases PI 3-K activity in anti-cbl immunoprecipitates. 380, REH, and RS4; 11were incubated with antLCD38
antibody or control lgGl for 5 minutes. Cell lysates were immunoprecipitated with anti-cblantibody and PI 3-K activity in the immunoprecipitates was measured as described in the Materials and Methods.
Autoradiograms of thin-layer chromatography plates after separation of the resulting phosphatidylinositol 3-phosphate are shown.
1y
0
20
10
30
time (min)
Fig 3. Ligation of CD38 increases PI 3-K activity in antiphosphotyrosine immunoprecipitates. (A) Immature B cells 380, REH, and
RS4;ll were incubated with anti-CD38 antibody or control lgGl for
5 minutes. Cell lysates were immunoprecipitated with antiphosphotyrosine antibody and PI 3-K activity in the immunoprecipitates was
measured as described in the Materials and Methods. Autoradiograms of thin-layer chromatography plates after separation of the
resulting phosphatidylinositol 3-phosphate are shown. (B)380 and
REH cells were incubated with antLCD38 antibody for the indicated
times, and cell lysates were immunoprecipitated as described above.
PI 3-K activity at different time points is shown as a percentage of
the maximal level. Data are from one experiment representative of
three separate experiments.
Effect of PI 3-K inhibitors on CD38-induced growth slippression. To test whether activation of PI 3-K was related
to growth suppression after CD38 ligation, we took advantage of two compounds, wortmannin and LY294002, which
preferentially inhibit PI 3-K both in vitro and in
We previously showed that CD38-mediated suppression of
cell growth can be seen only in cultures supported by stroma
or stroma-derived cytokines.' Thus, their effects were tested
in stroma-supported 4-day cultures of immature I3 cells (380,
REH, and RS4; 1 1 ) exposed to anti-CD38 antibody. A single
addition of wortmannin (100 nmol/L) or LY294002 (2 pmol/
L) at the beginning of the cultures inhibited CD38-induced
growth suppression in the three cell lines (Table I). Similar
results were obtained when TI6 was replaced by another
antLCD38 antibody (IB4 of IgG2 class; gift of Dr F. Malavas, University of Ancona. Italy; data not shown). CD38
surface expression remained unchanged when measured
Table 1. Recovery of Immature B Cells After Culture in the
Presence of AntLCD38 and PI 3-K Inhibitors
Cell Recovery (96 of control).
Cell Type
Cell lines
380
REH
RS4; 11
B-lineage ALL
1
2
3
4
5
Normal immature B cellst
1
2
3
4
CD38
CD38 + Wortmannin
(100 nmolIL1
CD38 + LY294002
(2 pmol/L)
14 t 1
42 I
5
8 2 1
84 t 5
90 t 4
41 f 5
46 t 1
1
74 I
30 t 5
56 2 5
40 t 4
24 I5
11 2 1
66 I5
74
84
93
12
87
t 10
I
21
t4
I
3
NT
NT
NT
36 t 2
ia t2
23 I8
17 t 1
421
16 f 2
66 1 15
71 1 2
2
25 I
31 t 5
NT
NT
28 I2
36 I3
t2
Abbreviation: NT, not tested.
Percentage of cell recovery after 2 to 4 days of culture on stroma relative to
that of control cultures with isotype-matched unreactive antibody. Results are
expressed as the mean f SD of at least three measurements. The increased cell
recovery caused by PI 3-K inhibitors was statistically significant in all experiments
( P < .01 by t-test).
t CD19'. slg cells from normal bone marrow.
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
KITANAKA ET AL
594
L l + lgGl
0 + lgGl + wortmannin
120
+ anti-CD38
+ anti-CD38 + wortmannin
T -
-l
Stroma
IL-3, IL-7,SCF
Fig 5. Wortmannin rescues immature B cells from CD38-mediated
growth suppression in the presence of stroma or stroma-derived
cytokines. 380 cells were cultured in the presence of stroma or 11-3
(10 ng/mL), IL-7 (25 ng/mL), and SCF (20 ng/mL), with or without
l00 nmol/Lwortmannin, before anti-CD38 antibody or control lgGl
was added t o the cultures. Bars (mean +. SD of triplicate tests)represent thepercentage of cell recovery after 4 days with anti-CD38 antibody relative t o parallel cultures with control IgG1.
after 30 minutes and 24 hours of incubation of 380 and REH
cells with the inhibitors (data not shown). Thus, the effect
Wortrnannin
of PI 3-K inhibitors was not
due to downregulation of CD38
expression.
In dose-response experiments, the effects of wortmannin
were maximal at 100 nmoVL. At this concentration, recovery
of viable cells was similar (>80%) to that of control cultures
with 0.05% DMSO, whereas higher concentrations were
markedly cytotoxic (data not shown). The optimal concentration of LY294002 was 2 pnol/L, because higher concentrations were also cytotoxic.
In our experiments, a single addition of wortmannin to
cultures effectively rescued immature B cells from CD38mediated suppression. Wortmannin is known to be unstable,
and its anti-PI 3-K effect may weaken during culture.'" In
time-course experiments with the 380 cell line, wortmannin
(100 nmol/L) rescued cells exposed to anti-CD38 only when
added to cultures within 12 hours of CD38 ligation (data not
shown). These results suggest that early activation of PI
3-K after CD38 ligation is critical for subsequent growth
suppression.
The above experiments were performed with immature B
cells in coculture with stroma. To determine whether the
effect of wortmannin on CD38-induced growth suppression
could extend to stroma-free cultures, 380 cells were incubated in the presence of three stroma-derived cytokinesIL-3 (10 ng/mL), IL-7 (25 ng/mL), and SCF (20 ng/mL)culture conditions that also permit CD38-mediated suppression of cell growth.' After 4 days of culture, there was a
marked decrease in cell recovery in the presence of antiCD38, and the addition of wortmannin to the cultures reversed this suppressive effect (Fig 5). Similar results were
obtained when only IL-3 was used (data not shown). Thus,
wortmannin's reversal of CD38-mediated growth suppres-
sion reflects its direct effect on immature lymphoid cells,
rather than an indirect effect on stroma elements.
Finally, we determined whether the rescuing effects of PI
3-K inhibitors extended to cultures of normal B-cell progenitors and leukemic lymphoblasts from patients. In four experiments withpurifiednormal
bone marrow B cells, CD38
ligation induced marked loss of CD19' sIg- B-cell precursors, an effect that was antagonized by adding 1 0 0 nmol/L
wortmannin to the cultures (Table 1 and Fig 6). Wortmannin
and LY294002 had similar effects on CD38+ ALL cells from
patients (Table 1).
Inhibition of PI 3-K activity by wortmannin
and
LY294002. In the next set of experiments, wemeasured
the concentrations of wortmannin and LY294002 needed to
inhibit PI 3-K activity, for comparison with the concentrations needed to rescue cell growth in the same cells. At 100
nmol/L, wortmannin inhibited virtually all detectable PI 3K activity in immunoprecipitates produced with antibody to
PI 3-K p85 in 380 cells, whereas 2 ,umol/L LY294002 inhibited PI 3-K activity by approximately 50% (Fig 7A). The
addition of 100 nmol/L wortmannin to intact 380 and REH
cells also significantly inhibited the activity of PI 3-K immunoprecipitated by antiphosphotyrosine antibody after CD38
ligation (P < .02;Fig 7B), butno significant differences
were detectable after incubation with 2 ,umol/L LY294002
Control
CD38 CD38
Wortrnannin
CD38Fig 6. Wortmannin rescues normal Bcell progenitors from
mediated suppression. Highly enriched bone marrow CD19+ cells
were cultured for 3 days on bone marrow stroma with anti-CD38
antibody or control IgGl in the presence or absence of 100 nmollL
wortmannin. Cell numbers and immunophenotype were
assayed by
flow cytometry. Isometric contour plots show the relative
size of the
different cell populations
labeled with the indicatedantibodies. Mean
percentages of CD19'. slg- B-cell progenitors recovered, relative t o
control cultures, were 83% with wortmannin, 23% with anti-CD38,
and 66% with anti-CD38 plus wortmannin. Numbers of CD19'. slg'
B lymphocytes recovered remained essentially unchaged under all
culture conditions.
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
595
SIGNALING PATHWAYS OF CD38-MEDIATED B-CELL GROWTH SUPPRESSION
3
2
C
h
3c
40
20
h
0
LY294002 (M)
Wortmannin (M)
CD38
Wortmannin
+ CD38
CD38
Wortmannin
+ CD38
CD38
LY294002
+ CD38
CD38
LY294002
+ CD38
Fig 7. Concentrations of wortmannin and LY291002 that are effective in culture inhibit PI 3-K activity. (A) PI 3-K was immunoprecipitated
from 380 cell lysates with anti-PI 3-K p85 antibody and assayed for its activity in the presence of indicated concentrations of wortmannin
(left panel) or LY294002 (right panel). Results are presented as the percentage of activity compared with control samples not exposed to
inhibitors and are the averages of two independent determinations.(B) 380 and REH cells were first incubated for 20 minutes with or without
100 nmollL wortmannin (left panel) or 2 pmollL LY294002 (right panel) and then exposedto anti-CD38 antibodyfor 5 minutesbeforeextraction.
The PI 3-K activity precipitated by an antiphosphotyrosine antibody was measured as described in the Materials and Methods. Bars (mean 2
SD of triplicate tests) representthe ratio of PI 3-K activity in CD38-stimulated cells to that in unstimulated cells.
(Fig 7B). This finding was not unexpected, because binding
of LY294002 to PI 3-K is readily reversible and is unlikely
to survive immunoprecipitation.16.31
We also determined whether wortmannin or LY294002
influenced protein tyrosine phosphorylation induced by
CD38 dimerization. As shown in Fig 8, CD38 ligation in
380 and REH cells induced rapid tyrosine phosphorylation
of several intracellular proteins, and PI 3-K was present in
antiphosphotyrosine immunoprecipitates. Preincubation of
cells with 100 nmol/L wortmannin or 2 pmol/L, LY294002
did not inhibit CD38-induced protein tyrosine phosphorylation, and the amount of PI 3-K p85 protein coprecipitated
with tyrosine-phosphorylated proteins was not affected by
preincubation with the inhibitors. These data collectively
indicate that wortmannin and LY294002 rescue immature B
cells from CD38-mediated growth suppression by inhibiting
PI 3-K rather than protein tyrosine kinase activity.
Rapamycin does not rescue immature B cells from CD38mediated growth suppression. It has been reported that
wortmannin and LY294002 can also block p70 S6-kinase
activity caused by platelet-derived growth factor and insulin
stim~lation.~
~~
To~ .determine
whether p70 S6-kinase was
involved in the CD38-mediated suppression of cell growth,
we used rapamycin at a concentration (10 ng/mL) that abrogates p70 S6-kinase a ~ t i v i t y .In
~~
380
, ~and
~ REH cells, rapamycin decreased cell recovery after 4 days of culture on
stroma. Mean cell recovery rates were 3 1% in 380 and 34%
in REH cells (in quadruplicate measurements), as compared
with control cultures without rapamycin. Lower recovery
rates were attributed to cell cycle arrest, a known effect of
rapamycin, because cell viability remained greater than 90%
(data not shown). In parallel cultures, mean cell recovery in
the presence of antLCD38 was 17% in 380 cells and 31%
in REH cells. The addition of rapamycin to cultures containing anti-CD38 did not improve cell recovery. On the
contrary, cell recovery was 7% in 380 and 14% in REH in
cultures containing both anti-CD38 and rapamycin. Thus,
the rescuing effects of wortmannin and LY294002 are unlikely to be due to p70 S6-kinase inhibition.
DISCUSSION
Normal and malignant cells have signaling pathways that
can arrest the cell cycle and trigger programmed cell death.35
In immature B cells, one such regulatory pathway is initiated
by CD38 dimerization.’ In beginning to elucidate the biochemical changes accompanying CD38-mediated growth
suppression, we had previously shown that CD38 dimerization rapidly activates protein tyrosine kinase-dependent
phosphorylation, including that of syk and PLCy.’ In the
present study, we have identified cbl and PI 3-K as key
components of this signaling pathway and established a connection between biochemical events and cellular response.
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
596
- 205
Probe:
REH
380
I
- 205
- 121
&-
-86
-
Wortmanin
LY294002
- + - + - +
-- ++- -- --++
51
121
-86
anti-pyr
CD38
KITANAKA ET AL
U.
w
m
C
51
- + - + -+
--++ - ---- ++
Our results suggest that activation of PI 3-K is central to
CD38-mediated suppression of immature lymphoid cell
growth, because inhibitors of PI 3-K rescued immature B
cells from the effects of CD38 ligation. Importantly, the
critical role of PI 3-K is not confined to model cell lines,
but extends to immature B-lineage lymphoblasts freshly obtained from children with leukemia and to normal bone marrow B-cell progenitors.
chl is a 906 amino acid protein that is found in the cytosol
and cytoskeletal compartments of early B-lineage and myeloid cells and in nonhematopoietic
chl lacks kinase
activity but contains several proline-rich motifs and multiple
potential tyrosine phosphorylation sites that might mediate
its interaction with protein displaying SH3 and SH2 domain~..".'~Tyrosine phosphorylation of chl occurs after the
engagement of a variety of cell surface receptors, and its
binding to proteins involved in signal transduction, including
Src-family kinases, htk, ah/, nck, grh2, PLCy, and PI 3-K,
has been ~ h o w n . ~ ' After
) ~ ~ ~Fcy
~.'~
receptor engagement, chl
also binds to syk." However, under conditions that preserved
cbl/PI 3-K association, we could not detect cbllsyk association in REH and RS4:ll cells after CD38 ligation (data
not shown). Tyrosine phosphorylation of chl may also be
accompanied by its subcellular transl~cation.~~
Thus, chl
could promote the formation of transient signaling complexes containing different proteins that may regulate cellular functions by promoting localized increases in enzymatic
activity of the associated proteins." Our results indicate that
the ccllular consequcnces of chl activation includc cell cyclc
arrest and apoptosis. Involvement of cbl may be a common
feature of CD38-mediated signal transduction, because Kontani et al'9 have also recently found chl tyrosine phosphorylation after engagement of CD38 in retinoic acid-treated HL60
myeloid cells.
PI 3-K activity has been implicated in multiple cellular
Fig 8. Wortmannin and LY294002 do not inhibit
protein tyrosine phosphorylationtriggered by CD38
dimerization. 380 and REH cells were pretreatedwith
100 nmol/L wortmannin, 2 pmol/L LY294002, or vehicle only (0.05% DMSO) and then stimulated with
anti-CD38 antibody for 5 minutes. Cell lysates were
prepared and immunoprecipitated with antiphosphotyrosine antibody. The immunoprecipitatedproteins were separated by SDS-PAGE and transferred
to nitrocellulose membranes, which were probed
with antiphosphotyrosine antibody (pTyr; upper
panel) and then stripped and reprobed with antiPI 3-K p85 antibody (lower panel). Molecular mass
markers (in kilodaltonsl are indicated. The intense
band seen in all lanes near the 51-kD molecular
marker in the upper panel is the lg heavy chain of
the antibody used for immunoprecipitation.
functions.'".l3 Our observation that PI 3-K activation results
in inhibition of proliferation and induction of apoptosis contrasts with the involvement of this molecule in promoting
cell growth observed in other cell models."'." However, of
note, Beckwith et al'" recently reported that PI 3-K activity
is required for the anti-Ig-mediated growth inhibition of a
B-lymphoma cell line. The biochemical events leading to
the diverse cellular functions of PI 3-K activity remain to
be elucidated, although it has been recently suggested that
these could result from specific D3 phosphoinositides binding to distinct SH2 domain^.^' Because our initial experiments with PI 3-K inhibitors used stroma-supported cultures
of immature B cells, the effects of wortmannin and
LY294002 might have reflected suppression of the stroma
stimulus, rather than a direct effect on immature B cells. This
possibility was proved untrue by the ability of wortmannin to
rescue cells from the suppressive effects of CD38 ligation
in cultures in which stroma was replaced by IL-3, IL-7, and
SCF. Notably, similar results were seen with IL-3 alone, but
we failed to detect increased PI 3-K activity in antiphosphotyrosine immunoprecipitates from 380 cells exposed to IL3 (data not shown). Thus, we believe that changes on PI
3-K activity mediated by CD38 ligation, rather than by cytokines or contact with stroma, are central to the resulting
growth suppression. PI 3-K inhibitors did not completely
rescue cells from CD38-mediated inhibition. Although this
could be due to the instability of the compounds in culture,
we cannot rule out that PI 3-K-independent signaling could
also bc involved.
The human CD38 is a type I1 transmembrane protein with
a short intracytoplasmic domain that does not contain antigen
recognition activation motifs (ARAM) or tyrosine residues.'
Thus, it is likely that CD38-mediated signaling involves activation of other membrane-associated components. However,
these have not yet been identified. Funaro et al" found co-
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
SIGNALING PATHWAYS OF CD38-MEDIATED B-CELL GROWTH SUPPRESSION
capping of CD38 with surface Ig and CD19 in mature B
cells, and the latter may be involved in CD38-mediated signaling in sIg- immature B cells. Although CD38 expression
in immature B cells is not apparently modulated after incubation with specific antibodies, a similar interaction needs to
be tested in these cells. However, we have so far been unable
to detect CD38-associated proteins using a variety of lysing
conditions (data not shown).
In conclusion, we have shown that CD38-mediated signal
transduction in immature lymphoid cells involves tyrosine
phosphorylation of cbl and its association with PI 3-K; activation of PI 3-K results in cell cycle arrest and apoptosis.
CD38-mediated signal transduction may be crucial in the
homeostatic control of lymphopoiesis. Importantly, this control mechanism in normal B-cell progenitors remains intact
in their leukemic counterparts, offering a potentially effective avenue of therapeutic inter~ention.~~
Agonistic antibodies that activate CD38-derived growth inhibitory signals
could be exploited for treating leukemia and without the
need for toxin conjugation. Importantly, virtually all cases
of ALL express CD38,43 and CD38 dimerization is highly
effective in suppressing leukemic cell growth in vitro' and
in SCID mice engrafted with human ALL cells (K.F. Bradstock, University of Sydney, Sydney, Australia, personal
communication, October 1995). Humanized anti-CD38 antibodies for in vivo immunotherapy have already been engineered,44 based on the consideration that CD38 is only
weakly expressed or absent in primitive multipotent hematopoietic ~ e l l s . ~ ~ . ~ ~
ACKNOWLEDGMENT
We thank Drs Terukazu Tanaka, Yoshitsugu Kubota, and Charles
Rock for helpful discussions and Sharon Naron for editorial suggestions.
REFERENCES
1. Stamenkovic I, Stauton J, Seed B: Molecular cloning of CD38,
in Knapp W, Dorken B, Gilks WR, Rieber EP, Schmidt RE, Stein
H, Von dem Bome AEGKr (eds): Leucocyte Typing IV. Oxford,
UK, Oxford, 1989, p 87
2. Jackson DG, Bell JI: Isolation of a cDNA encoding the human
CD38 (T10) molecule, a cell surface glycoprotein with an unusual
discontinuous pattem of expression during lymphocyte differentiation. J Immunol 144:2811, 1990
3. Reinherz EL, Kung PC, Goldstein G, Levey RH, Schlossman
S F Discrete stages of human intrathymic differentiation: Analysis
of normal thymocytes and leukemic lymphoblasts of T-cell lineage.
Roc Natl Acad Sci USA 77:1588, 1980
4. Janossy G, Tidman N, Papageorgiou S, Kung PC, Goldstein
G: Distribution of T lymphocyte subsets in the human bone marrow
and thymus: An analysis with monoclonal antibodies. J Immunol
126:1608, 1981
5. Sieff C, Bicknell D, Caine G, Robinson J, Lam G, Greaves
M F Changes in cell surface antigen expression during hemopoietic
differentiation. Blood 60:703, 1982
6. Funaro A, Spagnoli GC, Ausiello CM, Alessio M, Roggero S,
Delia D, Zaccolo M, Malavasi F: Involvement of the multilineage
CD38 molecule in a unique pathway of cell activation and proliferation. J Immunol 145:2390, 1990
7. Kumagai M, Coustan-Smith E, Muny DJ, Silvennoinen 0,
Murti KG, Evans WE, Malavasi F, Campana D: Ligation of CD38
suppresses human B lymphopoiesis. J Exp Med 181:1101, 1995
597
8. Howard M, Grimaldi JC, Bazan JF, Lund FE, Santos-Argumedo L, Parkhouse RME, Walseth TF, Lee HC: Formation and
hydrolysis of cyclic ADP-ribose catalyzed by lymphocyte antigen
CD38. Science 262:1056, 1993
9. Silvennoinen 0, Nishigaki H, Kitanaka A, Kumagai M, Ito C,
Malavasi F, Lin Q, Conley ME, Campana D: CD38 signal transduction in human B-cell precursors: Rapid induction of tyrosine phosphorylation, activation of syk tyrosine kinase, and phosphorylation
of phospholipase C-y and phosphatydilinositol 3-kinase. J Immunol
156:100, 1996
10. Fry MJ: Structure, regulation and function of phosphoinositide 3-kinases. Biochem Biophys Acta 1226:237, 1994
11. Divecha N, Imine RF: Phospholipid signaling. Cell 80:269,
1995
12. Yao R, Cooper GM: Requirement for phosphatidylinositol-3
kinase in the prevention of apoptosis by nerve growth factor. Science
267:2003, 1995
13. Kimura K, Hattori S, Kabuyama Y, Shizawa Y, Takayanagi
J, Nakamura S, Toki S, Matsuda Y, Onodera K, Fukui Y: Neurite
outgrowth of PC12 cells is suppressed by wortmannin, a specific
inhibitor of phosphatidylinositol 3-kinase. J Biol Chem 269: 18961,
1994,
14. Kim TJ, YT Kim, Pillai S: Association of activated phosphatidylinositol 3-kinase with ~ 1 2 0 ' ~
in' antigen receptor-ligated B cells.
J Biol Chem 270:27504, 1995
15. Donovan JA, Wange RL, Langdon WY, Samelson LE: The
protein product of the c-cbl protooncogene is the 120-kDa tyrosinephosphorylated protein in Jurkat cells activated via the T cell antigen
receptor. J Biol Chem 269:22921, 1994
16. Vlahos CJ, Matter WF, Hui KY, Brown RF: A specific inhibitor of phosphatidylinositol 3-kinase 2-(4-morpholinyl)-8-phenyl-4HI-benzopyran-4-one (LY294002). J Biol Chem 269:5241, 1994
17. Manabe A, Coustan-Smith E, Kumagai M, Behm FG, Raimondi SC, Pui C-H, Campana D: Interleukin-4 induces programmed
cell death (apoptosis) in cases of high risk acute lymphoblastic leukemia. Blood 83:1731, 1994
18. Kumagai M, Manabe A, Pui C-H, Behm FG, Raimondi SC,
Hancock ML, Mahmoud H, Crist WM, Campana D: Growth of
childhood B-lineage acute lymphoblastic leukemia cells on bone
marrow stroma as a predictor of treatment outcome. J Clin Invest
97:755, 1996
19. Whitman M, Kaplan DR, Schaffhausen B, Cantley L, Roberts
TM: Association of phosphatidylinositol kinase activity with polyoma middle-T competent for transformation. Nature 315:239, 1985
20. Cory GOC, Lovering RC, Hinshelwood S, MacCarthy-Morrogh L, Levinsky RJ, Kinnon C: The protein product of the c-cbl
protooncogene is phosphorylated after B cell receptor stimulation
and binds the SH3 domain of Bruton's tyrosine kinase. J Exp Med
182:611, 1995
21. Meisner H, Conway BR, Hartley D, Czech MP: Interactions
of cbl with grb2 and phosphatidylinositol 3'-kinase in activated Jurkat cells. Mol Cell Biol 15:3571, 1995
22. Rivero-Lezcano OM, Sameshima JA, Marcilla A, Robbins
KC: Physical association between Src homology 3 elements and the
protein product of the c-cbl proto-oncogene. J Biol Chem 269: 17363,
1994
23. Hartley D, Meisner H, Corvera S: Specific association of the
B isoform of the p85 subunit of phosphatydilinositol-3 kinase with
the proto-oncogene c-cbl. J Biol Chem 27018260, 1995
24. Fukazawa T, Reedquist KA, Trub T, Soltoff S, Panchamoorthy G, Druker B, Cantley LC, Shoelson SE, Band H: The SH3
domain-binding T cell tyrosyl phosphoprotein pl20. Demonstration
of its identity with the'c-cbl protooncogene product and in vivo
complexes with Fyn, Grb2, and phosphatidylinositol-3-kinase.J Biol
Chem 270:19141, 1995
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
598
25. Meisner H. Czech MP: Coupling of the proto-oncogene product c-cbl to the epidermal growth factor receptor. J Biol Chem
270:25332, 1995
26. Marcilla A, Rivero-Lezcano OM, Agarwal A, Robbins KC:
Identification of the major tyrosine kinase substrate in signaling
complexes formed after engagement of the Fcy receptors. J Biol
Chem 270:91 15, 1995
27. Tanaka S, Neff L, Baron R, Levy JB: Tyrosine phosphorylation and translocation of the c-cbl protein after activation of tyrosine
kinase sigiialing pathways. J Biol Chem 270: 14347, 1995
28. Carpenter CL, Cantley LC: Phosphoinositide kinases. Biochemistry 29:l 1147, 1990
29. Yano H, Nakanishi S, Kimura K, Hanai N, Saitoh Y. Fukui
Y, Nonmura Y, Matsuda Y: Inhibition of histamine secretion by
wortmannin through the blockade of phosphatidylinositol 3-kinase
in RBL-2H3 cells. J Biol Chem 268:25846, 1993
30. Ui M, Okada T, Hazeki K, Hazeki 0:Wortmannin as a unique
probe for an intracellular signalling protein, phosphoinositide 3-kinase. Trends Biol Sci 20:303, 1995
31. Cbeatham B, Vlahos CJ, Cheatham L, Wang L, Blenis J,
Kahn CR: Phosphatidylinositol 3-kinase activation is required for
insulin stimulation of pp70 S6 kinase, DNA synthesis, and glucose
transporter translocation. Mol Cell Biol 14:4902, I994
32. Chung J, Grammer TC, Lemon KP, Kazlauskas A, Blenis J:
PDGF- and insulin-dependent ~ ~ 7 0 activation
'"~
mediated by phosphatidylinositol-3-OH kinase. Nature 370:7 1, 1994
33. Chung J, Kuo CJ, Crabtree GR, Blenis J: Rapamycin-FKBP
specifically blocks growth-dependent activation of and signaling by
the 70 kd S6 protein kinase. Cell 69: 1227, 1992
34. Price DJ, Grove JR, Calvo V, Avruch J, Bierer BE: Rapamycin-induced inhibition of the 70-kilodalton S6 protein kinase.
Science 257973. 1992
35. Scheuermann RH, Uhr JW: Connections between signal transduction components and cellular responses initiated by antigen receptor on B lymphocytes. J Exp Med 182:903, 1995
36. Blake TJ, Heath KG, Langdon WY: The truucation that gener-
KITANAKA ET AL
ated the v-cbl oncogene reveals an ability for nuclear transport, DNA
binding and acute transformation. EMBO J 12:2017, 1993
37. Blake TJ, Shapiro M, Morse HC 111, Langdon WY: The sequences of the human and mouse c-cbl proto-oncogenes show v-cbl
was generated by a large truncation encompassing a proline-rich
domain and a leucine zipper-like motif. Oncogene 6:653, 1991
38. Andoniou CE, Tien CBF, Langdon WY: Tumour induction
by activated ab1 involves tyrosine phosphorylation of the product of
the cbl oncogene. EMBO J 13:4515, 1994
39. Kontani K, Kukimoto I. Nishina H, Hoshino S. Hazeki 0,
Kanako Y, Hatada T: Tyrosine phosphorylation of the c-cbl protooncogene product mediated by cell surface antigen CD38 in HL-60
cells. J Biol Chem 271:1534, 1996
10. Beckwith M, Fenton RG, Katona IM, Longo DL: Phosphatidylinositol-3-kinase activity is required for the anti-lg-mediated
growth inhibition of a human 9-lymphoma cell line. Blood 87:2020,
1996
41. Rameh LE, Chen C-S. Cantley LC: Phosphatidylinositol
(3,4,5)P, interacts with SH2 domains and modulates PI 3-kinase
association with tyrosine-phosphorylated proteins. Cell 83:82 I , 1995
42. Funaro A, De Monte LB, Dianziani U, Forni M, Malavasi F:
Human CD38 is associated to distinct molecules which mediate
transmeinbrane signaling in different lineages. Eur J lminunol
23:2407, I993
43. Koehler M, Behm F, Hancock M, Pui C-H: Expression of
activation antigens CD38 and CD71 is not clinically important in
childhood acute lymphoblastic leukemia. Leukemia 7:41, I993
44. Ellis JH. Barber KA, Tutt A, Hale C, Lewis AP, Glennie
MJ, Stevenson GT. Crowe JS: Engineered antiLCD38 monoclonal
antibodies for immunotherapy of multiple myeloma. J lmmunol
155:925, 1995
45. Huang S, Terstappen LWMM: Lymphoid and myeloid differentiation of single human CD34'. HLA-DR' , CD38 heinatopoietic
stctii cells. Blood 83:1.515, 1994
46. Craig W, Kay R, Cutler KL, Lansdorp PM: Expression of
Thy-I on human hematopoietic progenitor cells. J Exp Med
177:1331, I993
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
1996 88: 590-598
CD38-mediated growth suppression of B-cell progenitors requires
activation of phosphatidylinositol 3-kinase and involves its
association with the protein product of the c-cbl proto-oncogene
A Kitanaka, C Ito, H Nishigaki and D Campana
Updated information and services can be found at:
http://www.bloodjournal.org/content/88/2/590.full.html
Articles on similar topics can be found in the following Blood collections
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.