From www.bloodjournal.org by guest on June 15, 2017. For personal use only.
High Levels of the Shed Form of L-Selectin Are Present in Patients With
Acute Leukemia and Inhibit Blast Cell Adhesion to Activated Endothelium
By Olivier Spertini, Patrizia Callegari, Anne-Sophie Cordey, Jacques Hauert, Jean Joggi,
Vladimir von Fliedner, and Marc Schapira
L-selectin is expressed by mostleukocytes and mediates the
initial step of adhesion t o vascular endothelium. A feature
of thisadhesion receptor is t o be shed from thecell surface.
We report here the presence of high levels of theshed form
of L-selectin (sL-selectin) in plasma from patients with acute
leukemia. We also show thatsL-selectin purified fromacute
leukemia plasma exhibits functional activity. The mean ( + l
SD) plasma level of sL-selectin among 100 healthy individuals was 2.1 0.7 pg/mL. This value was increased (>2 SD
above the mean) in 63% of 58 patients with acute lymphoblastic leukemia (ALL) and 59% of 93 patients with acute
myelogenous leukemia ([AMLI P < .001).Repeated measurements in 24 patients showed normal-range levels in 16 of
16 patients in complete remission and high levels in eight
of eight patients with therapy-resistant acute leukemia or
leukemia relapse. Furthermore, elevated sL-selectin levels
were detected in cerebrospinal fluid of three patients with
ALL suffering from a relapse limited t o the central nervous
system. Epitope mappingwith monoclonal antibodiesdemonstrated that L-selectin shedding from leukemic blasts was
accompanied by conformational changes of its epidermal
growth factor-like domain. A functional role forsL-selectin
purified from leukemic plasma was supported by its ability
t o completely inhibit L-selectin-dependent adhesion of
blast cells t o tumor necrosis factor-a (TNF-&activated endothelium in vitro. These results suggest that sL-selectin
may have an important role in the regulation of leukemic
cell adhesion t o endothelium. In addition, monitoring of the
sL-selectin level may be useful for evaluating leukemia activity, in particular for the detection of leukemia relapse.
0 1994 b y The American Society of Hematology.
C
of patients with acute leukemia and to evaluate its role in
modulating the adhesion of leukemic cells to endothelium.
IRCULATING LEUKOCYTES use several adhesion
molecules to bind to the vascular endothelium, leave
the bloodstream, and migrate into tissue^."^ L-selectin plays
a critical role in the initiation of normal leukocyte attachment
to activated endothelium:"' whereas other receptors (including integrins and immunoglobulin-like adhesion molecules)
are involved in leukocyte subsequent firm adhesion and
transmigration into tissues.'.''
The primary structures of L-selectin and other members
of the selectin family have been determined. The various
selectins have in common the presence of an N-terminal end
lectin domain, an epidermal growth factor-like domain, and
short consensus repeats similar to those found in complement
regulatory proteins.'3"6 L-selectin is expressed on most normal l e u k o ~ y t e s , ' ~
and
~ ' ~is also detectable on blast cells from
certain patients with acute lymphoblastic leukemia (ALL)
or acute myelogenous leukemia (AML)." The lectin domain
of L-selectin interacts with glycoconjugates expressed by
high endothelial venules of peripheral lymph node^,'^^''-'^
endothelium of mesenteric venules," endothelium adjacent
to inflammatory lesions,6,",28endothelial cells that have been
activated in vitro by inflammatory c y t ~ k i n e s , ~and
, ~ . 'myelin~
ated tracts of the central nervous ~ystem.~'.~'
Recently, heparin-like molecules have been identified as endothelial cell
ligands for L-~electin.~'
The binding of L-selectin to its ligands is inhibited by monoclonal antibodies against the receptor lectin d ~ m a i n , ~ ,certain
~ . ' ~ ,sulfated
~~
carb0hydrates,34-~~
and the shed form of L-selectin (~L-selectin).~~
sL-selectin,
which circulates in normal plasma at a concentration of approximately 2 /.~g/mL,~~"is formed by proteolytic cleavage
of the extracellular region of L-selectin followed by shedding
of the cleaved domain from the cell ~~Tface.'~.'~.~'.~~.~',~'
Different cleavage products are generated from different cell
types. A fragment with an M , of 62,000 is released from
lymphocytes, whereas fragments with M,s ranging from
80,000 to 105,000 are shed from neutrophils.2'~28~38~4'
Little information is yet available on the diagnostic and
physiologic significance of the circulating shed form of Lselectin, sL-selectin.3840The present study was undertaken
to determine sL-selectin levels during the clinical course
Blood, Vol 84,
No 4 (August 15). 1994:
pp 1249-1256
MATERIALS AND METHODS
Patients, plasma, and cell samples. Heparinized plasma samples
were obtained from 151 patients with newly diagnosed acute leukemia, including 93 patients with AML and 58 with ALL. The diagnosis and classification of AML or ALL were based on the criteria of
the French-American-British Cooperative
and immunophenotypic studies (see below). For certain patients, additional samples were obtained during induction therapy, in remission, or after
relapse. Control plasma samples were obtained from 100 healthy
blood donors. Cerebrospinal fluid was withdrawn from patients with
ALL before intrathecal chemotherapy. A reference value for the
level of sL-selectin in cerebrospinal fluid wasobtained by evaluating
samples from 41 patients with various neurologic conditions including ischemic cerebrovascular disease, intervertebral disk protrusion,
multiple sclerosis, and aseptic meningoencephalitis. After collection,
plasma and cerebrospinal fluid samples were immediately centrifuged at 13,000 X g and kept frozen at -80°C until further use.
Leukemic patients were treated according to standard protocols
for AML or ALL. Most patients with AML received an induction
therapy of cytosine arabinoside and daunorubicin, whereas consolidation included a course of amacrine and etoposide followed by a
course of high-dose cytosine arabinoside and daunorubicin. Induction therapy for ALL usually included viacristine, prednisone, daunorubicin, methotrexate, and L-asparaginase, as well as central ner-
From the Division of Hematology, University of Lausanne, Lausanne; and the Ludwig Institute for Cancer Research, Lausanne,
Switzerland.
Submitted November 18, 1993; accepted April 23, 1994.
Supported by Grant No. 31-33792.92 from the Swiss National
Foundation for Scient$c Research.
Address reprint requests to Olivier Spertini, MD, Division of Hematology, University of Lausanne, 1011-CHUV Lausanne, Switzerland.
The publication costsof 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 1994 by The American Society of Hematology.
0006-497I/94/8404-0027$3.00/0
1249
From www.bloodjournal.org by guest on June 15, 2017. For personal use only.
1250
vous system intrathecal prophylaxis, whereas intensification therapy
included a first course of cytosine arabinoside and daunorubicin
followed by a second course of cyclophosphamide and mitoxantrone.
Patients with relapsed AML or ALL were treatedusingrelevant
protocols.
Imrnunophenotyping. Mononuclear cells were prepared from
heparinized peripheral blood or bone marrow samples by centrifugation on Ficoll-Hypaque (Pharmacia, Uppsala, Sweden). Cell surface
antigens were detected by standard immunofluorescence methods
using a large panel of directly fluorescein isothiocyanate- or phycoerythrin-conjugated monoclonal antibodies. The expression of surface antigens was studied in double immunofluorescence using antibodies that react with CD1, -2, -3, -4, -5, -7, -10, -13, -14, -15,
-19, -20, -22, -25, -33, -34, -41, -61, and -71, glycophorin, surface
immunoglobulins, and HLA-DR (Becton Dickinson, Mountain
View, CA; Coulter, Hialeah, FL; and Dako, Glostrup, Denmark).
The expression of L-selectin on blast cells was assessed with the
anti-LAM1-3 monoclonal antibody.’2 Double immunofluorescence
analysis was performed with an Epics flow cytometer (Coulter Electronics, Hialeah, FL). A total of 5,000 cells were analyzed for each
sample.
Acetone-methanol-fixed blast cells were examined by light microscopy for the presence of terminal deoxynucleotidyl transferase,
intracytoplasmic immunoglobulins, and CD3. Rabbit polyclonal antibody against terminal deoxynucleotidyl transferase (Supertechs,
Bethesda, MD), monoclonal antibody against human immunoglobulin (Dako), anti-CD3 Leu-4 monoclonal antibody (Becton Dickinson), and alkaline phosphatase-labeled antimouse or antirabbit antibodies (Dako) were obtained from the designated suppliers.
sL-selectin enzyme-linked immunosorbent assay. sL-selectin
was assayed using a sandwich enzyme-linked immunosorbent assay
The anti-LAMI-5 monoclonal antibody was used as the
capture antibody. The presence of sl-selectin was revealed with
biotinylated anti-LAMI-3 monoclonal antibody and avidin-horseradish peroxidase, with 0-phenylene-diamine (0.125%, wt/vol) in
citrate buffer, pH 4.5, as the substrate. Plasma samples were run in
duplicate or triplicate and diluted at 11100 to112,000to obtain a
measure in the linear range of our assay. Absorbances at 405 nm
were measured using an MR 700 Microplate ELISA reader (Dynatech, Embrach, Switzerland). The concentration of sL-selectin was
determined using a standard curve constructed for each ELISA plate
with a reference plasma. The sL-selectin level inthe reference
plasma was measured using a recombinant L-selectin/IgG heavychain y l chimeric molecule as the standard.?’ This chimeric protein
was produced in COS-l cells transfected with pL-selectidy 1 cDNA
subcloned into the Ap’M8 expression vector (the vector was provided
by Lloyd Klickstein, Center for Blood Research, Boston, MA). After
transfection, the cells were cultured in Dulbecco’s minimal essential
medium containing 5% fetal calf serum. The fusion protein was
purified from culture medium by salt fractionation and immunoadsorption to immobilized anti-LAM1-3 monoclonal antibody. After
extensive washing with phosphate-buffered saline containing 0.5
molL NaCland 0.05% Tween 20, the fusion proteinwas eluted
with 4 mol/L imidazole, pH 7.5. Sodium dodecyl sulfate (SDS)
polyacrylamide gel (10%) electrophoresis and silver staining of the
eluate showed a large band with an M , of 180,000 and two additional
bands with Mrs of 70,000 and 45,000. By Western blot analysis, the
band with an M , of 180,000 was identified as theL-selectidyl heavy
chain of IgG chimera. Analysis of the gel with a Cliniscan 2 Scanner
(Helena Laboratories, Beaumont, TX) indicated that the band with
an M , of 180,OOO represented 50% of the sample total protein. The
protein concentration was determined with the Micro BCA Protein
Assay Reagent (Pierce, BA oud Beijerland, Holland) with purified
mouse immunoglobulin used as the standard.
With this ELISA, sL-selectin levels were linearly correlated with
the mean absorbance of triplicate dilutions of the standard plasma
SPERTlNl ET AL
over a concentration range between 0.005 and 0.060 pg/mL (n =
63, r 2 > .98). Samples anticoagulated with heparin or EDTAyielded
identical results. Since sL-selectin concentration is stable in frozen
plasma,” most determinations were performed in samples kept at
-80°C.
Shedding of L-selectin by blast cells and lymphocytes. Mononuclear cells were isolated by Ficoll-Hypaque density-gradient centrifugation from three normal blood donors and three patients with Lselectin’ acute leukemia, two with ALL and one with AML. The
six isolates were incubated inRPM1 1640 supplemented with 2%
fetal calf serum for 0, 2, 5, and 18 hours. At these various times,
the suspensions were centrifuged for 30 minutes at13,000 x g .
Mononuclear cells obtained from patients with leukemia contained
more than 95% of L-selectin’ blast cells. The cell suspensions isolated from normal donors were depleted of monocytes by adherence
on plastic. Cell viability was greater than 90% after 8 and 18 hours
of incubation, as determined by trypan blue exclusion. The concentration of sL-selectin and L-selectin in cell supernatants and cell
lysates was determined by ELISA.
Detection of sL-selectin by Western blot analysis. Plasma samples (1 mL) from eight patients with AMLand six patients with
ALL were fractionated with Na2S04(18% wt/vol) and centrifuged
for 20 minutes at 10,OOO X g at room temperature. Supernatants
were dialyzed against IO mmol/L. Tris hydrochloride, pH 8.0. The
samples were then incubated with constant rotation for 12 hours
at 4°C with 100 pL anti-LAMI-3-Sepharose 4B. After extensive
washing, sL-selectin was eluted with 4 mol& imidazole, pH 7.5.
After dialysis against phosphate-buffered saline, sL-selectin samples
were electrophoresed on 7.5% SDS-polyacrylamide gels according
to the method of Laemmli,&and transferred onto 0.2-mm nitrocellulose filters (Bio-Rad Laboratories, Glattbrugg, Switzerland). The
presence of sL-selectin was detected with the anti-LAMI-14 monoclonal antibody:’ peroxidase-conjugated goat antimouse immunoglobulin antibody (Dako), and enhanced chemiluminescence detection reagents (Amersham, Zurich, Switzerland). Mrs were determined
using molecular weight marker standards (Bio-Rad Laboratories).
Endothelial-leukemic blast cell attachment assay. The attachment of leukemic blasts to endothelial cells was determined using
previously published procedure^.^^^^ Human umbilical vein endothelial cells (passage 2 to 3) were grown to confluence in 0.5% gelatincoated Petri dishes (Becton Dickinson). After an 8-hour stimulation
Boehringer,
with 100 U1mL tumor necrosis factor-a([TNF-a]
Mannheim, Germany), endothelial cell monolayers were washed and
incubated for 30 minutes with medium (RPM1 1640/5% fetal calf
serum) alone or medium containing purified sL-selectin (5 pg1mL)
or the purified F(ab’), fragment of the W6132 anti-HLA class 1
monoclonal antibody (5 pg1mL). sL-selectin waspurified by saltfractionation andaffinity chromatography from plasma obtained
from three patients withALLand
three patients with AML, as
described above. Leukemic blast cells were isolated by Ficoll-Hypaque density gradient centrifugation from six patients with Lselectin’ acute leukemia, two with ALL and four with AML. The
six isolates contained greater than 95% of L-selectin’ blast cells.
Leukemic blasts (4 X 10’) were preincubated for 15 minutes on ice
in 100 pL of medium alone or containing 5 pg1mL of the purified
F(ab’), fragment of the W6132 anti-HLA class I monoclonal antibody or 5 pg1mL of the purifiedanti-LAM1-3 monoclonal antibody.
After 15 minutes of preincubation at 4”C, leukemic blasts were added
to the endothelial monolayers kept under rotation at 72 rpm. After
30 minutes at 4°Cor 15 minutes at 37°C nonadherent cells were
disgarded and Petri dishes were placed vertically in 2% glutaraldehyde andfixed overnight. After washing, the number of adherent
cells was determined by counting eight to 12 microscopic fields
(0.08 mm’ per field). Results were expressed as the mean t 1 SD.
Statistical analysis. The relationship between parameters was
evaluated by determination of the Spearman correlation coefficient.
From www.bloodjournal.org by guest on June 15, 2017. For personal use only.
1251
sL-SELECTINLEVELS IN ACUTE LEUKEMIA
Normal
donors
blast cells by inhibiting their attachment to activated endothelium, the clinical characteristics (ie, presence of splenoE
megaly, hepatomegaly, enlargement of lymph nodes, skin or
mucosal infiltration) of 41 patients with AML were related
to the plasma concentration of sL-selectin at diagnosis. The
15 patients with splenomegaly had significantly higher
plasma levels of the shed receptor than the 26 patients who
10
did not have an enlarged spleen ( P < .Ol). It was notpossible
to assess the impact of sL-selectin on the development of
hepatomegaly and/or lymphadenopathies because only 7 patients were found to have enlarged liver or lymph nodes.
0 0
A weak correlation was noted between plasma sL-selectin
0 0
levels in patients with acute leukemia and their venous blood
blast cell counts (rs = .47, n = 69, P < .01). Furthermore,
observations made with 3 patients with aleukemic acute leuJ
0
kemia indicated that plasma sL-selectin reflected the total
J
mass of blast cells in the body. Two patients with ALL who
v)
had no detectable blast cells in their peripheral blood had
Fig 1. sL-selectin levels in plasma of 100 normal blood donors(0). increased plasma levels of sL-selectin (5.2 and 17.4 pg/mL,
93 patients with AML (01,and 58 patients with ALL (0).Median
respectively) and more than 50% lymphoblasts in the bone
values are indicated by horizontal bars.
marrow. In addition, a patient with therapy-resistant aleukemic M1 AML was evaluated for 42 days. During the entire
follow-up period, he had continuing marrow infiltration by
Differences between groups were assessed using the Kruskal-Wallis
leukemic blasts, almost no blast cells in the peripheral blood,
statistic, Mann-Whitney rank sum tests, and Student's t test.
and elevated sL-selectin levels (data not shown).
RESULTS
sL-selectin levels and the clinical course of acute leukemia. Thirteen patients with ALL and three with AML, all
sL-selectin levels in plasma of patients with acute leukeof whom had increased sL-selectin levels at diagnosis (from
mia. Plasma sL-selectin levels ranged from 0.65 to 3.6 pg/
4.5 to 73.5 pg/mL), were reevaluated after having entered a
mL (mean ? 1 SD, 2.1 ? 0.7 pg/mL) among 100 healthy
complete remission. In all the remission samples (16 of 16),
blood donors, from 0.7 to 68 pg/mL (median, 4.1 pg/mL)
the concentration of sL-selectin was within the normal range
among 93 patients with AML, and from 0.11 to 98.3 pg/mL
(<3.5 pg/mL; Fig 2). Additional samples were obtained
(median, 4.6 pg/mL) among 58 patients with ALL (Fig 1).
from 5 of these 16 patients after their leukemia had relapsed;
Thus, both AML and ALL patients had increased sL-selectin
levels ( P < .001). Moreover, 59% of the patients with AML
(55 of 93) and 66% of those with ALL (38 of 58) had sL100,
selectin levels exceeding 3.5 pg/mL (mean + 2 SD among
healthy blood donors). The following French-American-British subtype distribution was observed among the 93 patients
with AML: 15, M1; 46, M2; 6, M3; 16, M4; 9, M5; and 1,
M6. Increased sL-selectin concentrations were seen with the
M1, M2, M4, and M5 subtypes, whereas levels within the
normal range were observed in all the patients (6 of 6 ) with
acute promyelocytic leukemia (M3). An L-selectin level
within the normal range was also found in the only patient
with erytholeukemia (M6). Among the patients with ALL, 14
had T-ALL and 44 had lymphoblasts expressing B-lineage
markers. Similar sL-selectin levels were measured in the two
groups of patients with ALL.
The level of expression of L-selectin by blast cells was
compared with the plasma concentration of sL-selectin in 56
patients with acute leukemia (44 patients with AML and 12
patients with ALL). Statistical analysis did not disclose a
significative correlation between these two parameters (rs =
. I J
.06). As previously reported, blast cells from most patients
diagnosis
complete
with AML expressed low levels of the receptor. Thus, the
median level of L-selectin expression in 35 patients with
remission
AML was 20% (range, 2.5% to 78%). Significantly higher
Fig 2. sL-selectin levels in plasma of 16 patients with acute leukelevels of L-selectin were observed in T-ALL (median value,
mia at diagnosis and at remission. The area under the broken hori70%; n = 7; P < .01).
zontal line indicatesthe 95% confidence interval of sl-selectin conBecause sL-selectin could influence the final homing of
centration in normal plasma.
i
AML
ALL
From www.bloodjournal.org by guest on June 15, 2017. For personal use only.
1252
SPERTlNl ET AL
increased sL-selectin levels (from 4.8 to 16.2 pg/mL) were
observed in all of them ( 5 of 5). sL-selectin levels and circulating blast cell counts of a patient receiving chemotherapy
for relapsed pre-B-ALL are shown in Fig 3. Leukemia relapse was established on a bone marrow aspirate obtained 6
months after the patient had entered a complete remission.
At relapse, no circulating leukemic blasts were detectable,
but sL-selectin was increased to 9.4 pg/mL (Fig 3, day l).
A second remission, as determined by marrow examination
at day 42, was observed after reinduction chemotherapy.
Between day 2 and day 10, sL-selectin decreased from 17.9
to less than 3.5 ,ug/mL, and concentrations within the normal
range were observed untilday 80. The patient relapsed a
second time at day 90, and complete remission could not
be observed despite aggressive chemotherapy. sL-selectin
remained elevated (>3.5 pg/mL) throughout the second relapse, although blast cells were not detectable in the peripheral blood from day 105 to day 128 (Fig 3).
High concentrations of sL-selectin in the cerebrospinal
fluid frompatients with meningeal leukemia. sL-selectin
levels in the cerebrospinal fluid of 41 patients with various
neurologic conditions (see Materials and Methods) ranged
from 0.003 to 0.044 pg/mL (mean ? 1 SD, 0.020 +- 0.010
pg/mL). Similar sL-selectin concentrations were found in the
cerebrospinal fluid of 23 patients with ALL without central
nervous system involvement (mean ? 1 SD, 0.018 ? 0.010
pg/mL). Thirteen to 34 months after the initial diagnosis, 3
patients withALL suffered from a relapse limited to the
central nervous system, as assessed by the presence of leukemic lymphoblasts on cerebrospinal fluid examination. Cerebrospinal fluid samples obtained at that time contained increased levels of sL-selectin (0.066 to 0.202 pg/mL). In l
patient, a cerebrospinal Auid sL-selectin increase (0.2 pg/
mL) was observed 8 weeks before the appearance of clinical
and cytologic signs of meningeal involvement. It should be
emphasized that these 3 patients with relapses confined to
the central nervous system had plasma sL-selectin levels
within the normal range.
L-selectin is shed ,from leukemic blast cells. Additional
1 10
30
50
70
90
110
130
Days after Chemotherapy
Fig 3. (0)
sL-selectin levels and (H)peripheral blood blast cell
counts in a patient with ALL in first relapse. Two courses of chemotherapy were administered, the first on day 1 and the second on day
100 (arrows).
n
100
Y
80
60
40
20
m
0
0
2
5
18
Time (h)
Fig 4. Kinetics of L-selectin shedding from leukemic cells and normal peripheral blood lymphocytes. Cells (30x lo6)were cultured for
various times at 37°C. sL-selectin levels in cell supernatants were
measured by ELISA. Values represent the mean f 1 SD of duplicate
samples. At the indicated times, lymphocyte lysates contained 133,
129, 118,
and 1 4 0 ng, whereas blast lysates contained 120, 88, 52,
and 50 ng of sL-selectin. Results are representative of three similar
Lymphocytes; (
experiments. (0)
experiments were undertaken to study the shedding of Lselectin from leukemic blasts and to characterize the structural properties of sL-selectin in acute leukemia. The kinetics
of L-selectin shedding were examined using L-selectin+
blasts obtained from three patients with AML or ALL. After
5 to 18 hours of incubation, 50% to 70% of the L-selectin
initially expressed at the surface of blast cells was released
in the medium in the form of sL-selectin (Fig 4). For comparison, normal peripheral blood lymphocytes cultured under
the same conditions released 30% to 40% of cell surface
L-selectin. The molecular characteristics of sL-selectin in
plasma samples collected from leukemic patients were evaluated by SDS-polyacrylamide gel electrophoresis and Western blotting using the anti-LAM1-14 monoclonal antibody
directed against the sL-selectin short consensus repeat domain. Samples were obtained from 14 patients with acute
leukemia, 6 with ALL and 8 with AML. In contrast to the
pattern seen with normal plasma (which contains two sLselectin isoforms mainly derived from neutrophils and lymphocytes), the various leukemic plasma samples contained
only a single form of sL-selectin (Fig 5). The ALL samples
contained an sL-selectin form withan M , ranging from
60,000 to 90,000, whereas sL-selectin from AML plasma
had an M, ranging from 70,000 to 95,000 (Fig 5). These
results are consistent with earlier data on the sL-selectin
isoforms released from normal peripheral blood leukocytes,
which indicated that sL-selectin from lymphocytes had an
M,of 62,000, whereas fragments shed from neutrophils had
M,s ranging from 80,000 to 105,000.2L~28~38~41
Additional information on the structure of sL-selectin was
provided by epitope mapping. A series of binding studies
used the anti-LAMl-I'* and anti-LAM1-5 monoclonal an-
From www.bloodjournal.org by guest on June 15, 2017. For personal use only.
SL-SELECTINLEVELS IN ACUTELEUKEMIA
Mr x I O 3
205
-
116
-
80
-
49
-
a a a
32 Fig 5. sl-selectin isoforms in AML and ALL. After immunoprecipitation from salt-fractionated plasma and SDS-polyacrylamide (7.5%)
gel electrophoresis, sL-selectin was revealed by Western blot analysis using the anti-LAM1-14 monoclonal antibody and horseradish
peroxidase-conjugated goat antimouse antibody. M, ( x lo-') values
are indicated at left.
tibodies"'thatreactwith
sites presentonthe
epidermal
growth factor domain of cell surface L-selectin. Aspreviously described for sL-selectin present in normal
plasma,3x,3'the shed form of L-selectin isolated from ALL
and AML plasma exhibited the site for anti-LAMI-5 but
not the one for anti-LAMI-l monoclonal antibody. In contrast, the recombinant L-selectin/IgG I heavy-chain chimera
expressed sites for both anti-LAM 1 - 1 and anti-LAMI-5.
This pattern demonstrates a conformational difference between the epidermal growth factor domains of leukemic cell
sL-selectin andcell surface L-selectin. Additional experiments using monoclonalantibodies against the short consensus repeat domain (anti-LAMI-14). the lectin domain (antiLAM 1-6. -7, -8, and - 1 I ) , and the epidermal growth factor
domain (anti-LAMI-IS) of sL-selectin compared the properties of sL-selectin in AML and ALL samples. sL-selectins
fromthe two lineages had similar reactivities with these
various monoclonal antibodies (not shown).
sL-selectin inhibition of the attachment of leukemic blasts
to activated human vascular endothelium in vitro. The
functional properties of sL-selectin isolated from leukemic
plasma were evaluated by measuring its ability to inhibit the
1253
attachment of L-selectin'blast cells to TNF-a-activated
endothelium. The attachment of leukemic blasts was determined under rotation, since the L-selectin-mediated component of this reaction is better detected under nonstatic condition~.'.~'Only a few leukemic blast cells were attached to
unactivated endothelium at 4°C (9 5 7 blasts per field, n =
3) andat37°C (21 2 12 blasts perfield, n = 3). On the
other hand, endothelial cells activation with TNF-a for 8
hours induced a significant increase in leukemic blast cell
attachment at 4°C (21- to 59-foId. n = 3) and at 37°C (eightfold to 39-fold, n = 3). In three experiments performed at
4°C where the shedding of L-selectin is minimal4' and the
family of &integrins are not active, the preincubation of
blast cells with the anti-LAMI-3 monoclonal antibody inhibited the cytokine-induced increase in blast cell adhesion
by 42% 2 7% ( P < ,005. n = 3). At 37°C. the anti-LAMI3 monoclonal antibody inhibited it by 5 1 % 2 5% ( P < .005,
n = 3). Under the same conditions. thepreincubation of
endothelium with 5 pg/mL purified sL-selectin inhibited
52% 2 5% of blast cell adhesion at 4°C (P < .OOS, n = 3)
and 43% 2 IO% at 37°C ( P < .005, n = 3). indicating that
the L-selectin-dependent adhesion of leukemic blasts was
entirely prevented by the binding of the shed form of Lselectin to endothelium (Fig 6). Furthermore, these results
indicate that L-selectin acts in conjunction with other adhesion receptors to mediate blast cell attachment to cytokineactivated endothelium. sL-selectin was purified from plasma
obtained from three patients with AML and three
patients
withALL. The different sL-selectin forms had the same
capacity to inhibit blast cell adhesion to endothelium. Furthermore, the lymphoblastic and myeloid shedforms had the
same inhibitory effect on the lymphoblast andmyeloblast
attachment to activated endothelium without respect to blast
cell origin (data not shown). Additional experiments examinedthe effect of substituting L-selectin- leukemic blasts
for the L-selectin' cells. The adhesion of L-selectin- blast
cells to the activated endothelium was not influenced by
the presence of S pg/mL sL-selectin (data not shown). This
observation further demonstrated that sL-selectin is a specific
inhibitor for the attachment of L-selectin+ blasts to activated
endothelial monolayers.
DISCUSSION
In this study we show that sL-selectin, the cleavedand
shed form of the cell adhesion receptor L-selectin, is present
at high concentrations in plasma samples of a majority of
patients with AML or ALL. We also provide evidence that
adhesion of L-selectin+ leukemic blasts to the cytokine-activated endothelium canbeinhibited by sL-selectin. These
observations support the conclusion that cleavage of cell
surface L-selectin and inhibition of leukemic cell adhesion
to the endothelium by the shed receptor provide opportunities for regulating the traffic of blast cells out of the bloodstream of patients with acute leukemia.
The majority (62%) of the 151 patients with acute leukemiawhowere studied hadsL-selectin levels morethan 2
SD above the meanvalue observed amonghealthyblood
donors (Fig l ) . sL-selectin concentrations are likely to reflect
the total massof leukemic cells, as indicated by observations
showing that high plasma sL-selectin levels can be seen in
From www.bloodjournal.org by guest on June 15, 2017. For personal use only.
SPERTlNl ET AL
1254
A
without TNF
control
anti-HLA class I mAb
anti-LAM1-3 mAb
sL-selectin (5 pglml)
0
B
100
200
300
400
without TNF
control
anti-HLA class I mAb
anti-LAM1-3 mAb
sL-selectin (5 Fg/ml)
0
l00
200
300
400
500
Attached BlastslField
Fig 6. L-selectin-mediated attachment of leukemic blast cells to
cytokine-activated endothelium. Mononuclear cells containing more
than 95% of blast cells were isolated from the peripheral blood
of a
patient with an M 1 (A) and an M 4 (B) AML. Assays were performed
at 4°C for 30 minutes (A) and at 37°C for 10 minutes (B). Confluent
endothelial monolayers were cultured for 8 hours at 37°C with medium or TNF-a. sL-selectin was purified by salt-fractionation and affinity chromatography from plasma obtained fromt w o patients with
AML, as described in Materials and Methods. The figure shows the
number of myeloblasts attached per field. Results are expressed as
the mean 2 1 SD and are representative of six similar experiments.
mAb, monoclonal antibody.
aleukemic leukemia (Fig 3, days 105 to 128). The concentrations of sL-selectin may also be dependent on other factors
such as the rate of synthesis of the receptor by blast cells,
its rate of shedding from leukemic cells, and its capacity to
bind ligands expressed on vascular endothelium. Furthermore, the rates of shedding and synthesis of L-selectin may
be variable among leukemic cells and could be upregulated
by cytokines produced by blast cells or present in their microenvironment. Therefore, since the plasma concentration
of sl-selectin does not depend only on the level of expression of L-selectin,the absence of correlationbetween the
concentration of sL-selectin and the level of expression of
the cell surface receptor is not surprising.
Although it is possible thata cellular reaction to blast
overload may participate in the increase of sL-selectin concentrations in patients with acuteleukemia, severalevidences support the conclusion that the major portion of the
circulating receptor is shed from blast cells. Thus, normal
plasma containstwo sL-selectinisoforms; oneform ( M , ,
62,000) is shed from lymphocytes and the other ( M , , 80,000
to lOS,000) from neutrophils.” In coRtrast, only one isoform
is detectable in leukemicplasma,mostcertainly
because
normal hematopoietic cells are replaced by monoclonal leukemic cells (Fig 5 ) . Furthermore, a decrease of L-selectin
expression should be demonstrated at the surface of peripheral blood lymphocytes if the increase of sl-selectin plasma
concentration resulted from a lymphocytic reaction to blast
~ v e r l o a d . ” , ’This
~ was not the case, since lymphocytes and
neutrophilsisolated from peripheralblood of two patients
with aleukemic leukemia expressedthe same levels of Lselectin as normal leukocytes (70% to 80% and 85% to 95%,
respectively).
Impressively, a perfect correlation was found in patients
with &selectin+ acute leukemia between sL-selectin levels
and the clinical courseof acute leukemia, with normal-range
levels being observed in 16 of 16 patients in complete remission and high levels in eight of eight patients with therapyresistant acute leukemia or leukemia relapse (Fig 2). Thus,
monitoring sL-selectin levels was helpful for the early diagnosis of leukemia relapse in patients with sL-selectin‘ acute
leukemia. Furthermore, an absence of normalization of the
sL-selectin level after induction chemotherapy was associated with a failure to achieve a complete remission (Fig 3,
days 105 to 128). It should be emphasized that apart from
L-selectin’ acuteleukemia,elevatedconcentrations
of sLselectin can also be seen
in patients with othermalignant
(0. Spertini, M.Schapira,unpubhematologicdisorders
lished observations, June 1993). Slightly increased levels are
also observed in acute inflammation, sepsis, or the acquired
immunodeficiency syndrome.”
Lymphoblastsand monoblastsfrequentlyinfiltratethe
central nervous system.48In symptomatic individuals, leukemic infiltration of the meninges can often be demonstrated
by the presence of blast cells in the cerebrospinal fluid.However, in asymptomatic patients, cerebrospinal fluid examination may not disclose any leukemic cells ormay yield nondi41 patients with
agnostic results.4’.”’ We havefoundthat
variousvascular,inflammatory,
or mechanicalneurologic
disorders had a mean cerebrospinal fluid sL-selectin level of
in 23 ALL
0.020 5 10 pg/mL.Similar valueswereseen
not havemeningeal leukemia(0.00sto
patientswhodid
0.045 pg/mL). In contrast, higher values (0.066, 0.1 SO, and
0.202 pg/mL) were found in cerebrospinal fluid samples of
three patients with arelapsed ALL limited tothecentral
nervous system. The onlyotherconditioninwhichanincreased sL-selectin level was observed in the cerebrospinal
(0. Spertini,A.Stucki,M.
fluid is bacterialmeningitis
Schapira, unpublished observation). Thus, elevated concentrations of sL-selectin in the cerebrospinal fluid of ALL patients may facilitate the diagnosis of central nervous system
involvement.
Whereas normal plasmacontainstwo
sL-selectinisoforms, only one form of the shed receptor is detectable in
leukemic plasma. sL-selectin from patients with ALL has an
M, of 60,000to90,000,whereas
sL-selectin fromAML
plasma has an M , between 70,000 and 95,000 (Fig 5). This
difference in M,s is likely to result from posttranslational
modifications of the L-selectin molecule. Asingle andidenticalL-selectin mRNA species,which encodes aprotein of
From www.bloodjournal.org by guest on June 15, 2017. For personal use only.
SL-SELECTIN LEVELS IN ACUTELEUKEMIA
1255
receptor in acute leukemia may contribute in vivo to maintain
approximately 37,000 daltons, is found in cells that differenthese cells in the bloodstream. However, further in vivo
tiate along the myeloid and lymphocytic pathway^.^' Bestudies are needed to establish definitively the role of sLcause AML and ALL plasmas contain distinct sL-selectin
selectin in the regulation of blast cell migration into tissues.
isoforms, the M , value of the shed receptor could be considWatson et all' previously reported that soluble recombinant
ered, in addition to conventional criteria, as a marker of
L-selectin can prevent in vivo the migration of neutrophils
myeloid or lymphoid differentiation.
into sites of inflammation. The findings made in our study
sL-selectin isolated from plasma of patients with acute
suggest that sL-selectin may have a similar role in vivo
leukemia expressed the same epitopes as sL-selectin from
and regulates, in concert with other adhesion molecules, Lnormal plasma, but conformational differences were obselectin+ blast cell circulation.
served between the shed form of L-selectin and cell surface
L-selectin. The epidermal growth factor-like domains of cell
surface L-selectin and sL-selectin have an identical primary
ACKNOWLEDGMENT
structure. However, the epitope definedon the epidermal
The authors thank Dr P. Schneider, Dr J.-P. Grob, J. Durand, C.
growth factor-like domain of L-selectin by the anti-LAM1Cretegny, B. Brocard, S. Quarroz, M. Barclay, V. Joliquin, and J.1 monoclonal antibody is not detectable on sL-selectin. As
C. de Flaugergues and theparticipants of the Swiss group for clinical
previously reported for normal leukocytes, the loss of this
cancer research (SAKK) for providing plasma and cerebrospinal
fluid samples and performing immunophenotypic studies, andN.
epitope upon L-selectin shedding from the cell surface demMonai for technical assistance.
onstrates that this reaction is associated with a change in the
conformation of the epidermal growth factor-like domain.3*
REFERENCES
Since the LAMI-l epitope seems to be involved in the regulationof L-selectin affinity for its ligand,22,47the loss of
1. Osborn L: Leukocyte adhesion to endothelium in inflammation.
Cell 62:3, 1990
this epitope may reduce the affinity of L-selectin for its
2. Springer TA: Adhesion receptors of the immune system. Nacounterreceptor. If such a scenario is operative, one might
ture 346:42S, 1990
expect that the affinity between sL-selectin and its endothe3. Stoolman LM: Adhesion molecules controlling lymphocyte
lial cell ligand would be weaker that the affinity between
migration. Cell 56:907, 1989
the ligand and cell surface L-selectin. This would provide a
4. Ley K, Gaehtgens P, Fennie C, Singer MS, Lasky LA, Rosen
mechanism for detaching sL-selectin from its endothelial
SD: Lectin-like cell adhesion molecule 1 mediates leukocyte rolling
ligand and replacing it by higher-affinity L-selectin molein mesenteric venules in vivo. Blood 77:2SS3, 1991
cules present at the surface of normal activated leukocytes.
S . von Andrian UH, Chambers JD, MC Evoy LM, Bargatze RF,
Since the number of L-selectin receptors expressed by norArfors K-E, Butcher EC: Two step model of leukocyte-endothelial
mal leukocytes is lower than the number of sL-selectin molecell interaction in inflammation: Distinct roles for LECAM-l and
the leukocyte p 2 integrins in vivo. Proc Natl Acad Sci USA 88:7538,
cules present in normal plasma, this mechanism maybe
1991
important in regulating the attachment of normal leukocytes
6. Jutila MA, Rott L, Berg EL, Butcher EC: Function and regulato the vascular endothelium." Different conditions are obtion of the neutrophil MEL-14 antigen in vivo: Comparison with
served in acute leukemia since leukemic blast cells express
LFA-1 and MAC-l. J Immunol 143:3318, 1989
much less L-selectin than normal leukocytes,21and we show
7. Lewinsohn DM, Bargatze RF, Butcher EC: Leukocyte-endohere that leukemic plasma contains much higher concentrathelia1cell recognition: Evidence of a common molecular mechanism
tions of sL-selectin than normal plasma. Under these condishared by neutrophils, lymphocytes, and other leukocytes. J Immunol
tions, the inhibitory effect of sL-selectin on the attachment
138:4313,1987
of blast cells to their endothelial ligands is likely to exceed
8. Smith CW, Kishimoto TK, Abbass 0, Hughes B, Rothlein R,
the capacity of blast cells to attach via the L-selectin mechaMcIntire LV, Butcher E, Anderson DC: Chemotactic factors regulate
lectin adhesion molecule 1 (LECAM-1)-dependent neutrophil adhenism. Direct support for this hypothesis was obtained by
sion to cytokine-stimulated endothelial cells in vitro. J Clin Invest
experiments examining the effect of sL-selectin on blast cell
87:609,1991
attachment to cytokine-activated endothelium under non9. Spertini 0, Luscinskas FW, Munro JM, Kansas GS, Griffin
static conditions. Preincubation of endothelial cell monoJD, Gimbrone MA, Tedder T F Leukocyte adhesion molecule-l
layers with 5 pg/mL sL-selectin purified from plasma sam(LAM-l, L-selectin) interacts with an inducible endothelial cell liples of patients with ALL or AML almost completely
gand to support leukocyte adhesion. J Immunol 147:256S, 1991
inhibited L-selectin-mediated attachment of leukemic blast
10. Picker LJ, Warnock RA, Bums AR, Doerschuk CM, Berg
cells to the activated endothelium (Fig 6).
EL, Butcher EC: The neutrophil selectin LECAM- 1 presents carboWe have shown here that in acute leukemia, interactions
hydrate ligands to the vascular selectins ELAM-l and GMP-140.
between leukemic blasts and endothelial cells via the LCell 66:921,1991
11. Watson SR, Fennie C, Lasky LA: Neutrophil influx into an
selectin mechanism can be inhibited in vitro by high levels
inflammatory site inhibited by a soluble homing receptor-IgG chiof sL-selectin. Since in vivo experiments have shown that
mera. Nature 349: 164, 1991
leukocyte rolling, mediated through L-selectin, is a prerequisite for subsequent steps of migration into t i s s ~ e s , 4 . ~high
. ' ~ * ~ ~ 12. Butcher EC: Leukocyte-endothelial cell recognition: Three
(or more) steps to specificity and diversity. Cell 67:1033, 1991
plasma sL-selectin levels in patients with acute leukemia
13. Tedder TF, Ernst TJ, Demetri CD, Isaacs CM, AdlerDA,
may have an important role in regulating the initiation of
Disteche CM: Isolation and chromosomal localization of cDNAs
blast cell adhesion to endothelium. The capacity of sL-selecencoding a novel human lymphocyte cell-surface molecule, LAMI :
tin to inhibit leukemic cell attachment to endothelium in vitro
Homology with the mouse lymphocyte homing receptor and other
suggests that the elevated concentrations of this circulating
human adhesion proteins. J Exp Med 170: 123, 1989
From www.bloodjournal.org by guest on June 15, 2017. For personal use only.
1256
14. Bowen B, Nguyen T, Lasky LA: Characterization of a human
homologue of the murine peripheral lymph node homing receptor.
J Cell Biol 109:42I , 1989
15. Siegelman MH, van de Rijn M, Weissman IL: Mouse lymph
node homing receptor cDNA clone encodes a glycoprotein revealing
tandem interaction domains. Science 243:1165, 1989
16. Lasky LA: Selectins: Interpreters of cell-specific carbohydrate
information during inflammation. Science 258:964, 1992
17. Griffin JD, Spertini 0, Ernst TJ, BelvinMP, Levine HB,
Kanakura Y, Tedder T F GM-CSF and other cytokines regulate surface expression of the leukocyte adhesion molecule-l on human
neutrophils, monocytes, and their precursors. J Immunol 145576,
1990
18. Tedder TF, Penta AC, Levine HB, Freedman AS: Expression
ofthehuman leukocyte adhesion molecule, LAMl. Identity with
the TQI and Leu-8 differentiation antigens. J Immunol 144532,
1990
19. Kishimoto TK, Jutila MA, Butcher EC: Identification of a
human peripheral lymph node homing receptor: A rapidly downregulated adhesion molecule. Proc NatlAcad Sci USA 87:2244,
1990
20. Kansas GS, Muirhead MJ, DaileyMO: Expression of the
CD I I/CD 18, LAM- 1, and CD44 adhesion molecules during myeloid
and erythroid differentiation in man. Blood 76:2483, 1990
21. Spertini 0, Freedman AS, Belvin MP, Penta AC, Griffin JD,
Tedder TF: Regulation of leukocyte adhesion molecule-l (TQ 1, Leu8) expression and shedding by normal andmalignant cells. Leukemia
5:300, 1991
22. Spertini 0, Kansas GS, ReimannKA, Mackay CR, Tedder
TF: Function and evolutionary conservation of distinct epitopes on
the leukocyte adhesion molecule- 1 (TQ- I, Leu-8) that regulate leukocyte migration. J Immunol 147:942, 1991
23. Stoolman LM, Yednock TA, Rosen SD: Homing receptors
on human and rodent lymphocytes-Evidence for a conserved carbohydrate-binding specificity. Blood 70:1842, 1987
24. Imai Y, Singer MS, Fennie C, Lasky LA, Rosen SD: Identification of a carbohydrate-based endothelial ligand for a lymphocyte
homing receptor. J Cell Biol 113:1213, 1991
25. Stamper HB Jr, Woodruff JJ: Lymphocyte homing into lymph
nodes: In vitro demonstration of the selective affinity of recirculating
lymphocytes for high-endothelial venules. J Exp Med 144:828, 1976
26. Baumhueter S, Singer MS, Henzel W, Hemmerich S, Renz
M,Rosen SD, Lasky LA: Binding of L-selectin tothe vascular
sialomucin CD34. Science 262:436, 1993
27. Ley K, Tedder T, Kansas G: L-selectin can mediate leukocyte
rolling in untreated mesenteric venules in vivo independent of E- or
P-selectin. Blood 82:1632, 1993
28. Kishimoto TK, Julita MA, Berg EL, Butcher EC: Neutrophil
Mac-l and MEL-14 adhesion proteins inversely regulated by chemotactic factors. Science 245:1238, 1989
29. Spertini 0, Luscinskas F W , Gimbrone MA Jr, Tedder TF:
Monocyte attachment to activated human vascular endothelium in
vitro is mediated by leukocyte adhesion molecule-l (L-selectin) under non-static conditions. J Exp Med 175:1789, 1992
30. Huang K, Geoffroy JS, Singer MS, Rosen S: A lymphocyte
homing receptor (L-selectin) mediates the in vitro attachment of
lymphocytes to myelinated tracts of the central nervous system. J
Clin Invest 88:1778, 1991
3 I . Needham LK, Schnaar RL: The HNK- 1 reactive sulfoglucuronyl glycolipids are ligands for L-selectin and P-selectin but not Eselectin. Proc Natl Acad Sci USA 1993:1359, 1993
SPERTlNl ET AL
32. Norgard-Sumnicht ICE, Varki NM, Varki A: Calcium-dependent heparin-like ligands for L-selectin in nonlymphoid endothelial
cells. Science 261:480, 1993
33. Gallatin WM, Weissman IL, Butcher EC: A cell-surface molecule involved in organ-specific homing of lymphocytes. Nature
304:30, 1983
34. Ley K, Linnemann G, Meinen M, Stoolman LM, Gaehtgens
P: Fucoidin, but not yeast polyphosphomannan PPME, inhibits leukocyte rolling in venules of the rat mesentery. Blood 8 I :177, 1993
35. Stoolman LM, Rosen SD: Possible role of cell-surface carbohydrate-binding molecules in lymphocyte recirculation. J Cell Biol
96:722, 1983
36. Stoolman LM, Tenforde TS, Rosen SD: Phosphomannosyl
receptors may participate in the adhesive interactions between lymphocytes and high endothelial venules. J Cell Biol 99:1535, 1984
37. Yednock TA, Butcher EC, Stoolman LM, Rosen SD: Receptors involved in lymphocyte homing: Relationship between a carbohydrate-binding receptor andthe
MEL-l4 antigen. J CellBiol
104:725, 1987
38. Schleiffenbaum B, Spertini 0, Tedder TF: Soluble L-selectin
is present in human plasma athigh levels and retains functional
activity. J Cell Biol 119:229, 1992
39. Spertini 0, Schleiffenbaum B, White-Owen C, Ruiz P Jr,
Tedder TF: ELISA for quantitation of L-selectin shed from leukocytes in vivo. J Immunol Methods 156:115, 1992
40. Gearing J, NewmanW: Circulating adhesion molecules in
disease. Immunol Today 14506, 1993
4 I . Jung TM, Dailey MO: Rapid modulation of homing receptors
(gp90Me'"') induced by activators of protein kinase C. Receptor shedding due to accelerated proteolytic cleavage atthe cell surface. J
Immunol 144:3130, 1990
42. Jutila MA, Kishimoto TK, Butcher EC: Regulation and lectin
activity of the humanneutrophil lymph node homing receptor. Blood
76: 178, I990
43. Benett JM, Catovsky D, Daniel MT, Flandrin G, Galton DAG,
Gralnick HR, Sultan C: Proposals for the classification of the acute
leukemias. Br J Haematol 33:451, 1976
44. Benett JM, Catovsky D, Daniel MT, Flandrin G, Galton DAG,
Gralnick HR, Sultan C: Proposed revised criteria for the classification of acute myeloid leukemia: A report of the French-AmericanBritish Cooperative Group. Ann Intern Med 103:620, 1985
45. Benett JM, Catovsky D, Daniel MT. Flandrin G, Galton DAG,
Gralnick HR, Sultan C: Criteria for the diagnosis of acute leukemia
of megakaryocytic lineage (M7): A report of the French-AmericanBritish Cooperative Group. Ann Intern Med 103:460, 1985
46. Laemmli UK: Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680, 1970
47. Spertini 0, Kansas GS, MUNOJM, Griffin JD, Tedder TF:
Regulation of leukocyte migration by activation ofthe leukocyte
adhesion molecule-l (LAM-l) selectin. Nature 349:691, 1991
48. Wolk RN, Masse SR. Conklin R, Freireich EJ: Incidence of
central nervous system leukemia in adults. Cancer 33:863. 1974
49. Bigner SH, Johnston NW: The cytopathology of cerebrospinal
fluid. I. Nonneoplastic conditions, lymphoma and leukemia. Acta
Cytol 25:335, 1981
50. Nies B, Thomas L, Freireich E: Meningeal leukemia: A follow up study. Cancer 18:546, 1965
5 I . Ord DC, Ernst TJ, Zhou LJ, Rambaldi A, Spertini 0, Griffin
JD, Tedder TF: Structure of the gene encoding the human leukocyte
adhesion molecule-l (TQI, Leu-8) of lymphocytes and neutrophils.
J Biol Chem 265:7760, 1990
From www.bloodjournal.org by guest on June 15, 2017. For personal use only.
1994 84: 1249-1256
High levels of the shed form of L-selectin are present in patients with
acute leukemia and inhibit blast cell adhesion to activated
endothelium
O Spertini, P Callegari, AS Cordey, J Hauert, J Joggi, V von Fliedner and M Schapira
Updated information and services can be found at:
http://www.bloodjournal.org/content/84/4/1249.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.