1. health and fitness
2. disease
3. cancer

Expression of the c-mpl Proto-oncogene in Human Hematologic Malignancies
By 1. Vigon, F. Dreyfus, J. Melle, F. ViguiB, V. Ribrag, L. Cocault, M. Souyri, and
Similar to two other hematopoietic growth factor receptors, the c-fms (macrophage colony-stimulatingfactor receptor) and the c-kit genes, c-mpl has been discovered
through the study of oncogenic retroviruses. Unlike c-fms
and c-kif, which both belong to a subgroup of tyrosine kinase receptors, the c-mpl proto-oncogeneencodes a new
member of the cytokine receptor superfamily. W e have
studied the expression of c-mpl in a series of 105 patients
with hematologic malignancies using Northern blot analysis. The levels of c-mpl transcripts in lymphoid malignancies and in chronic myeloproliferative disorders were not
significantly different from those found in normal bone
marrow cells, in which c-mpl was barely detectable. In
S.Gisselbrecht
contrast, c-mpl expression was increased in 26 of 51 patients with acute myeloblastic leukemia (AML) and in 5 of
16 patients with myelodysplastic syndromes. Amplification of the c-mplgene was detected in genomic DNA of one
M4 AML patient. There was no significant correlation between c-mpl expression and the French-American-British
classificationof AML. Patients with high c-mpl expression
appeared to belong to a subgroup of AML with a low rate of
complete remission and a poor prognosis, including secondary leukemia and AML with unfavorable cytogenetic abnormalities.
0 1993 by The American Society of Hematology.
T
HE PROLIFERATION and maturation of hematopoitive or myelodysplastic syndromes and in AML patients
etic cells are tightly regulated by factors that positively
with poor prognosis because of chemotherapy-resistant disor negatively modulate pluripotential stem cell proliferation
ease, including most AML patient’s with deletions of chroand multilineage differentiation. These effects are mediated
mosomes 5 and/or 7.
through the high-affinity binding of extracellular factors to
Although c-mpl expression was detected only in maligspecific cell surface receptors. Human leukemic cells have
nancies with immature blast cell proliferation, its expresbeen shown to express functional receptors for a variety of
sion did not uniformly correlate with the presence of immahematopoietic growth factors including interleukin-3
ture blast cells. Our results suggest that c-mpl expression in
(IL-3), granulocyte-macrophage colony-stimulating factor
AML patients could be of prognostic significance in predict(GM-CSF) granulocyte-CSF (G-CSF), macrophage-CSF
ing response to induction therapy.
(M-CSF), and stem cell factor.’-7Several investigators have
shown that deregulated expression of growth factors and/or
MATERIALS AND METHODS
growth factor receptors could be involved in the pathogenePatients and cell line. Peripheral blood (PB) or bone marrow
sis of human leukemias and myeloproliferative syn(BM) samples were obtained from 105 patients (104 adults and 1
dromes.8-12We have recently cloned complete cDNAs of a
child) who had been referred to the Hematology Department at the
new member of the cytokine receptor superfamily designed
Hapita1 Cochin between 1988 and 1992. All specimens were colas c-mpl.I3This family includes not only cytokine receptors
lected before the initiation of therapy and were obtained from 5 1
such as the IL-2 receptor 0 and y chains and the receptors
patients with AML, 11 with acute lymphoblastic leukemia (ALL),
for IL-3, IL-4, IL-5, IL-6, IL-7, IL-9, GM-CSF, G-CSF,
19 with myeloproliferative disorders (MPD) in chronic phase, 8
with non-Hodgkin’s lymphoma (NHL), and 16 with myelodysplaserythropoietin, LIF, and oncostatin M, but also the growth
tic syndromes (MDS) (Table 1). The diagnosis of each AML was
hormone and prolactin receptors.’416c-mpl is the cellular
established on the basis of morphologic and cytochemical staining,
homologue of the viral oncogene transduced in an acute
and patients were classified according to the criteria of the Frenchdefective murine retrovirus, the myeloproliferative leukeAmerican-British (FAB) committee.*’Two patients were classified
mia virus (MPLV).I7 Within a few weeks, mice infected
asFAB MO, 2 as MI, 13 asM2,5 as M3,8 asM4, I6 as M5,5 as M6,
with MPLV develop an hepatosplenomegaly and their he7 as L2, and 4 as L3. The diagnosis of MDS was made according to
matopoietic progenitors yield terminally differentiating
the FAB criteria, and 1 I patients were classified as refractory anecells in vitro in the absence of any added growth f a c t ~ r . ’ ~ * ’ ~
Infection of bone marrow cells in vitro with MPLV generates autonomous growth factor-independent hematopoietic
From the U363 INSERM, ICGM, and the Hematology Departcell lines belonging to various lineages.” Previously, we
ment, H6pital Cochin, Paris: and the Laboratoire de Cytogknktique,
showed that expression of the c-mpl proto-oncogene was
H6tel Dieu, Paris, France.
restricted to nonlymphoid hematopoietic tissues.” ThereSubmitted November 30, 1992; accepted March 15. 1993.
fore, it seemed to be of interest to study the expression of
Supported by grants from the Institut National de la Santk et de la
this gene in patients suffering from different hematopoietic
Recherche Mkdicale and Comitk de Paris de la Ligue Nationale
contre le Cancer. I. V. received afellowshipfrom the Ligue Nationale
malignancies. Very low or undetectable levels of c-mpl trancontre le Cancer.
scripts were found in cells from patients with lymphoid maAddress reprint requests to F. Dreyfus, MD, Hematology Departlignancies or with myeloproliferative and myelodysplastic
ment,
H6pital Cochin, 27 rue du Faubourg, Saint-Jacques, 75014,
syndromes in chronic phase. However, c-mpl expression
Paris, France.
was markedly detectable in 26 of 5 1 cases of acute myeloThe publication costs of this article were defrayed in part by page
blastic leukemia (AML). In one case of AML, high exprescharge payment. This article must therefore be hereby marked
sion was associated with a genomic amplification of the
“advertisement” in accordance with 18 U.S.C. section I734 solely to
lp34 region encompassing the c-mpl locus.2oc-mpl expresindicate this fact.
sion was frequently observed in patients with secondary
0 1993 by The American Society of Hematology.
AML developed during the progression of myeloprolifera0006-4971/93/8203-001I $3.00/0
Blood, Vol82, No 3 (August 1). 1993: pp 877-883
877
VIGON ET AL
Table 1. c-mpl Expression in Human Hematologic Malignancies
Diagnosis
Increase Over
Normal c-mpl
Expression
AML (51 patients)
MO
MI
M2
012
012
9/13
M3
M4
M5
215
318
7/16
M6
MDS (16 patients)
515
RSA
RAE6
MPD (19 patients)
015
511 1
CML
PV
017
011 1
ET
ALL (11 patients)
Q/1
L2
L3
NHL (8 patients)
017
014
018
mia with excess of blasts (RAEB), 5 as sideroblasticanemia (RSA).
In the group of patients with MPD, 7 had chronic myelogenous
leukemia (CML), 1 1 had polycythemia vera (PV), and 1 had an
essential thrombocytemia (ET). As a control, PB and normal BM
from volunteerswere obtained after an informed consent. A human
erythroleukemia cell line (HEL), provided by W. Vainchenker
(U362-IGR Villejuif, Villejuif, France), was cultured in (Y medium
containing 10% fetal calf serum.
AML patient population and treatment. The clinical characteristics ofthe 5 1 patients with AML are shown in Table 2. Of 5 1 AML
patients, 47 were treated with doxorubicin and cytosinearabinoside
(35 mg/m2/d for 3 days and 100 mg/m2/dfor 10 days, respectively).
The 4 untreated patients included 2 elderly patients (unique patient
number [UPN] 42 and 50: Table 2) and 1 patient with apoor performance status (UPN 51). One patient (UPN 49) refused therapy.
Complete remission (CR) after induction chemotherapy was defined by a normocellular marrow with less than 5% blasts. We considered induction failure patients as those with resistant disease and
those who died of infection or bleeding during induction therapy.
Patients with resistant AML and with a good performance status
received a second course of chemotherapywith mitoxantrone at 12
mg/m2/d for 3 days, etoposide at 200 mg/mz/d on days 8 through
10, and cytosine arabinoside at 500 mg/m2/d on days 1 through 3
and 8 through 10."
mRNA preparation and Northern blot analysis. PB or BM cells
were isolated after Ficoll-Hypaque density gradient centrifugation
and washed three times in Hanks' balanced salt solution. RNA was
extracted by the guanidium thiocyanate procedure.23After precipitation in ethanol at -2O"C, RNA pellets were resuspended in water.
RNA concentrations were determined by optical density, and samples were frozen at -80°C until use.
RNA samples ( 10 pg of total RNA) were denatured in glyoxal,
dimethyl sulfoxide, and phosphate buffer at 65°C for 10 minutes,
size-fractionated by electrophoresis through 1% agarose gel, and
transferred to nylon membrane. Membranes were hybridized with
random '2P-labeled probes and washed under stringent conditions
as previously described." The c-mpl probe used in this study was a
280-bp-long fragment located in the human c-mpl extracellular
domain.
RNA loading was verified by hybridization of each filter with a
murine 0 actin probe. The level of c-mpl expression was compared
with that observed in HEL cells and in normal BM cells. The presence of megakaryocytes or platelets in BM and PB samples was
detected with a glycoprotein IIb (GPIIb) cDNA
given by Dr
G. Uzan (U 2 I7-Grenoble, Grenoble, France).
DNA preparation and Southern blot analysis. High molecular
weight DNA was extracted by conventional methods. DNAs (10 pg)
were digested with restriction enzymes under reaction conditions
recommended by the manufacturers. Agarose gel electrophoresis,
Southern blot transfer, and hybridization were performed as previously described.17Probes used as markers of the 1p32-34 chromosomal region were tal- 1, L-myc, c-jun (provided by Dr D. MathieuMahul, U 301, Pans, France), lck (provided by Dr R. Benarous, U
332, Paris, France), Rhesus (from Dr J.P. Cartron, CNTS, Pans,
France), and a G-CSF receptor probe (given by Dr S. Gillis, Immunex, Seattle, WA).
Cytogenetic analyses. The cytogenetic study was performed on
37 patients with AML at diagnosis on BM samples. Cells were cultured for 24 or 48 hours before chromosome preparation. When
possible, 20 metaphases were photographed and analyzed after reverse heat G (RHG) banding of the chromosomes?5 Karyotypes
were described according to the International System for Human
Cytogenetic Nomenclature.26
Statistical analysis. The x2 test was used to determine significance of qualitative variable differences.Probabilities of c.05were
considered statistically significant. The log-rank test was used to
compare the groups of patients with respect to achieving CR and
disease-free survival.27
RESULTS
c-mpl is expressed in patients with AML or RAEB.
RNAs isolated from PB or BM from 105 patients with hematologic malignancies described in Materials and Methods
and Table 1 were studied for c-mpl expression by Northern
blot analysis. We previously reported that c-mpl was detected as a major 3.7-kb and a minor 2.7-kb mRNA species
in the human HEL cell line. Both transcripts have been
cloned, and they potentially code for polypeptides that
share common extracellular and transmembrane domains
but differ in their cytoplasmic regions. These two mRNA
probably derive from a unique c-mpl gene by alternative
~plicing.'~
To study c-mpl expression, we used as a probe a
fragment located in the mpl extracellular domain that detects the two c-mpl transcripts in HEL cells (Fig I , lane a).
As shown in Fig 1, no signal was detected in mRNA isolated from normal PB cells (Fig 1, lane b), and c-mpl was
barely detectable as a faint 3.7-kb transcript in BM samples
from healthy individuals (Fig 1, lane c). Very low c-mpl
expression was detected in the 19 patients with lymphoproliferative disorders and in the 19 patients with MPD in
chronic phase. In contrast, c-mpl mRNA levels appeared to
be elevated in 26 of 5 1 ( 5 1%) AML patients and in 5 of 16
(31%) MDS patients. Typical results are shown in Fig 1.
Variability in intensity was seen among patients despite
equal loading (as verified by 6-actin hybridization), but cmpl expression in patient samples was usually lower than in
HEL cells, except for 1 patient (UPN 23 in Table 2) who
presented with a diagnosis of AML, subtype M4 (Fig 1, lane
g). Although the 3.7-kb c-mpl transcript was consistently
expressed more than the 2.7-kb c-mpl mRNA species, as
c-mpl EXPRESSION IN LEUKEMIAS
879
Table 2. AML Patient Characteristics
Remission
Outcome
(mo)
Cell
Blasts
UPN
Sex/Age
FAB
Sources
(%)
Karyotype
Status
1
2
3
4
5
6
7'
8'
9
10
11
12'
13
14'
15'
16
17'
18
19
MO
MO
M1
M1
M2
M2
M2
M2
M2
M2
M2
M2
M2
M2
M2
M2
M2
M3
M3
M3
M3
M3
M4
M4
M4
M4
M4
M4
M4
M4
M5
M5
M5
M5
M5
M5
M5
M5
M5
M5
M5
M5
M5
M5
M5
M5
M6
M6
PB
BM
100
93
85
93
65
67
90
95
45
84
93
80
61
64
43
83
38
82
80
80
90
65
87
72
90
93
92
60
65
75
90
61
94
92
43
66
100
81
62
87
48
58
90
80
80
91
60
34
CR
CR
CR
F
F
CR
F
F
F
CR
F
F
CR
F
CR
CR
F
CR
CR
CR
F
CR
F
CR
F
CR
F
CR
CR
CR
CR
F
F
CR
F
F
F
F
CR
CR
CR
NT
F
F
CR
CR
F
F
49'
50'
51'
MI49
F/81
F/56
M6
M6
M6
BM
46XX
46XY
?
50XX,+4,+6,+19,+21
?
46XX,t(8;2 l)(q22;q22)
?
46XY.-5.-8,1 Zp+,+derl3,-17,-20,+marl
,+mar2,+mar3
43XY,de15(q31),- 11 ,-13,+14,-18,-22
46XX
46XX
44XY,-2,-3,-5,t(5;?)(qll
-ql2,?),-7,-9,t(9;19)(~22.?),16,-18
46XX
46XX
46XX
?
48XY,1 p-,3q+,+8,t(9;22)(q34;ql l),lnvl7q,+l9
46XX.t(15;17)(q22;qll)
46XY,t(l5;17)(q22;qll)
46XY,t(l5;17)(q22;qll)
46XY,t(l5;17)(q22;qll)
46XY,t(l5;17)(q22;qll)
?
?
46XY.dell p.dell(q1 1),-5,de16,11 p+,t( 1 1,2 l)(q21,q?)- 17,-21
46XY
?
46XY
?
?
46XX.de17.lnvl6
49XX,-1 ,t(1 ,?)(p36?),de15(q13,q33),-8,+de18(q42),-21,
46XX,lnv3(q22;q26)
?
?
?
47XX,-7,+i(7p),+l4
46XY,t( 1;3)(p36;q21)
46XY
46XX
46XY
?
?
46XY
46XY
46XX.5q45XY.-7
42XY ,del 1(q1 1),-5,6p+,7p+,der(9),+ 1O,+der( 1 1)- 15,- 17,- 18,
-19,-20
44XY.-3.de1(5)(ql2.q32).-7,de17,+i( 16p),-21,-22,+marl
45X.-X,del3(q22),-5,deI(5),del(6)ql4,- 18,-2 1 ,+marl ,+mar2
46XX,t(l0:21)(pl3,qll)
A 24
A9
R 12
D
D
A9
D
D
D
R9
D
D
A6
20
21
22
23
24'
25'
26
27
28'
29
30'
31
32'
33
34
35"
36
37
38
39
40
41
42
43
44
45
46
47
48
F/40
MI60
F/40
F/20
MI46
F/62
MI57
MI62
MI58
F/32
F/54
MI67
F/33
F/65
F/26
F/76
MI40
F/41
MI34
F/74
MI58
MI76
F/68
MI11
MI73
MI54
MI49
MI61
F/37
F/67
F/29
F/63
F/48
MI44
MI34
F/87
F/71
MI57
MI40
F/44
MI23
F/90
MI49
MI63
MI25
MI42
MI32
MI53
NT
NT
NT
D
D
D
BM
BM
PB
BM
PB
PB
BM
PB
BM
PB
BM
BM
BM
BM
BM
BM
BM
BM
BM
BM
PB
PB
PB
BM
BM
BM
BM
PB
BM
BM
PB
PB
BM
BM
PB
PB
BM
BM
BM
BM
PB
PB
P8
BM
BM
BM
BM
BM
34
61
39
c-mpl
Expression
D
A 54
A 36
D
A 18
A 10
A 50
D
A 27
D
A9
D
A 12
D
A 11
A8
A 12
A 74
D
D
R 19
D
D
D
D
A 24
A 26
A 18
D
D
D
R 9
A9
D
D
Abbreviations: FAB. French-American-British classification; BM, bone marrow; PB, peripheral blood; CR, complete remission; F, failure; NT, not
treated; R, relapse; D, dead; A, alive.
Secondary leukemia.
observed in normal BM and in HEL cells (in BM, the level
of mpl expression is not high enough to detect the 2.7-kb
transcript), the 2.7-kb c-mpl transcript was found to be quite
abundant in the two cases shown in Fig 1, lanes g and i.
In AML patients, it appeared that blast cell populations
were heterogeneous with respect to the expression of c-mpl.
No significant correlation was found between c-mpl expression and the FAB classification because c-mpl transcripts
VIGON ET AL
880
a b c d e f g h i
j k
Fig 1. Expression of c-mpl mRNA in human hematologic malignancies. Total RNA (10 pg) was denatured by glyoxal, size-fractionated by electrophoresis, and analyzed by blot hybridization using the
c-mplprobe (A) or the@actin probe (B). Lane a, HEL
cell line; lanes band c, PB and BM from healthy donors, respectively; lane d, patient with MPD; lane e,
patient with ALL; lane f, patient with NHL; lane g,
UPN 23; lane h, patient with RAEB; lane i, UPN 25;
lane j, UPN 34; lane k, UPN 50.
A
B
were found in patients with M2, M3, M4, M5, and M6
AML (Table I). However, it is noticeable that, although the
percentage of blast cells were lower in M6 AML samples
than in other AML subtypes, all patients with M6 leukemias
were c-mpl positive.
Of the 16 MDS patients, none of the 5 patients with RSA
had increased c-mplexpression, whereas 5 ofthe I I patients
with a diagnosis of RAEB showed increased c-mpl transcripts (Table I).
The c-mpl gene is amplified in one patient with an M4
AML. To determine whether genomic alterations might
be responsible for high expression, genomic DNAs of leukemic cells from patients expressing the highest levels of c-mpl
mRNA were subjected to Southern blot analysis. No gross
alteration of the c-mplgene was detected in these DNA samples, except in DNA from UPN 23, in which a c-mpl gene
amplification was found (Fig 2A, lanes a and d). Although
cytogenetic analysis was not available for this patient, these
EcoRl
a
A
b c
Sac1
d e f
results could be explained by either an amplification encompassing the c-mpl locus or by chromosome duplication.
The c-mpl gene was previously located on human chromosome 1 ~ 3 4 . ~Therefore,
'
probes for 6 genes located in the
small arm of chromosome I (tal-1, L-mvc, c-jim. Ick, Rhesus, and G-CSF receptor)28were hybridized to DNA from
UPN 23. No amplification was shown with any other probe
tested (see, for example, hybridization with a L-myc probe
Fig 2B), suggesting that the c-mpl locus was most likely amplified in the leukemic cells from this patient. Dot blot analysis performed on DNA dilutions indicated that the c-mpl
gene was eightfold more abundant in DNA from UPN23
than in control DNA (data not shown).
Correlation between c-mpl expression andclinicaland bioIogicSeatirres in AML. The characteristics of the AML
population we studied are listed in Table 2. We found no
significant correlation between c-mpl expression and characteristics such as sex, age, white blood cells, and platelets
counts or the origin of sampling (PB or BM) (TaLes 2 and
3). The HEL cell line which has megakaryocytic and erythroid characteristics is the only human hematopoietic cell
line we tested in which we detected c-mpl transcripts by
Northern blot analy~is.'~
Consequently, we investigated the
possibility of a correlation between GPIIb, glycophorin, and
c-mpl expression in AML. By immunophenotyping, only
M6 AML expressed the early erythroid glycophorin marker.
Northern blots from AML patients were rehybridized with a
GPIIb probe. No correlation was found between GPIlb and
c-mpl expression (data not shown).
Of the 51 AML cases we studied, 16 were presumably
secondary leukemias; 12 were blast crises arising during the
progression of either chronic MDS (UPN 7, 14, 15, 30, 50,
and 5 1) or MPD (UPN 8, 12, 17,24, 28, and 35); I patient
(UPN 49) had been professionally exposed to irradiation;
and 3 patients (UPN 3,25, and 32) had received chemotherapy after diagnosis of a solid tumor. Sixty-eight percent ( 1 1
of 16) of these 16 secondary leukemias were c-mpl positive
versus 43% (15 of 35) of the 35 AML patients presenting
with a diagnosis of de novo AML (Table 3).
Prognostic value of e-mpl expression in AML. Thirty
seven patients with AML could be classified into prognostic
groups based on the results of cytogenetic studies according
to the prognostic cytogenetic criteria of Keating et a129and
em--
B
Fig 2. Southern blot analysis of DNA from patients with high
mpl expression. Ten micrograms of high molecular weight DNA
from UPN 2 3 (lanes a and d) and UPN 2 5 (lanes band e) or control
PB DNA (lanes c and f) were digested with the restriction endonuclease FcoRl or Sad, electrophoresed in 0.8%agarose gel, blotted,
and hybridized to the mpl probe (A) or to the L-myc probe (B).
c-mpl EXPRESSION IN LEUKEMIAS
881
Table 3. Relationship of c-mpl Expression
to AML Patient Characteristics
DISCUSSION
Increase Over Normal Expression
(+I
(-J
Age (vr)
Sex ratio (M/F)
White blood cell count
(X 1 0 7 ~ )
Platelet count (X 1 Os&)
Median
Range
Median
Range
50.9
1.07
20-90
51.7
1.18
11-87
37.7
95.8
1.5-200 42.3
16-344 91.2
3.5-180
15-600
De novo AML (35patients)
Secondary AML (1 6 patients)
20 (57%)
5 (31%)
15 (43%)
1 1 (68%)
Good prognostic karyotype
(20patients)
14 (70%)
6 (30%)
Poor prognostic karyotype
(1 7 patients)
4 (23%)
13 (76%)
Schouten et al3' [good prognostic: normal karyotype, t(8,2 1)
(q22q22), t( 15,17)(q22ql I), or inv( 16)(q22);poor prognostic: all others]. Among the 18 c-mpl-negative AML, 14 had
a good prognostic karyotype [ I 1 had a normal karyotype, 3
had a translocation t( 15,17)(q22ql l)], and 4 had a poor risk
karyotype. In contrast, among the 19 c-mpl-positive AML,
only 6 had good prognostic characteristics: 3 had no cytogenetic abnormalities, 2 had a translocation t( 15,17)(q22q1 l),
and 1 had a t(8,2 l)(q22q22). However, 13 patients had chromosomal abnormalities affecting chromosome 5 and/or 7
or chromosome deletions or additions. As summarized in
Table 3, 70% (14 of 20) of AML with a good prognostic
karyotype were c-mpl negative, whereas 76% (1 3 of 17) of
AML with a high-risk karyotype were classified as c-mpl
positive. The correlation between high c-mpl expression and
the presence of a poor prognosis karyotype was highly significant (P< .001).
Of 5 1 AML patients, 47 were treated with doxorubicin
and cytosine arabinoside as described in Materials and
Methods. Of the 4 untreated AML patients, 1 was c-mpl
negative (UPN 42) and 3 were c-mpl positive (UPN 49, 50,
and 51), presenting with a diagnosis of M6 leukemia that
evolved from a myelodysplastic disorder (UPN 49 and 50)
or secondary exposure to irradiation (UPN 51). CR was
obtained in 24 of 47 patients treated (51%). Eight of 23
c-mpl-positive AML (35%) achieved CR, whereas 16 of 24
c-mpl-negative AML (67%)reached CR (P= .O 1). Resistant
disease was the main cause of induction failure in c-mplpositive AML (1 3 of 15) and was also observed in 3 of the 8
c-mpl-negative AML who did not achieve CR. Of the 13
c-mpl-positive patients who failed to enter CR, 5 patients
with a good performance status were treated with a second
course of chemotherapy, as indicated in Materials and
Methods. None of these 5 patients reached CR. However, in
our experience, this intensive therapy leads to CR in 45% of
refractory AML, suggesting a primary resistance in c-mplpositive AML.22Because too few patients in the c-mpl-positive group achieved CR, a meaningful comparison of the
duration of remission could not be performed.
In this report, the expression of the c-mpl gene that encodes a new member of the cytokine receptor superfamily
was examined by Northern blot analysis in cells from patients with various hematologic neoplasms. RNA blot analysis show that c-mpl transcripts were not detectable in normal PB and were expressed at very low levels in normal BM
cells. We found c-mpl mRNA to be expressed above normal
BM cell levels in 5 1% of AML and in 5 of 1 1 RAEB samples.
No increase in expression was observed in acute leukemias
of lymphoid origin, in lymphoma cells, or in samples of
myelodysplastic or myeloproliferative disorders in chronic
phase. This result contrasts with the biologic properties of
MPLV, a murine retrovirus that has transduced a constitutively activated form of this receptor chain and induces the
overproliferation of differentiating hematopoietic progenitors. Therefore, it seems very unlikely that c-mpl expression
in AML is responsible for the impaired differentiation of
AML blasts; but, because c-mpl is a component of a growth
factor signal transduction pathway, expression of the c-mpl
product could play a role in the proliferation of these cells.
c-mpl expression in AML may be caused by an alteration of
the c-mpl gene or may reflect the expansion of a cell population that normally expresses the c-mpl gene and represents
only a small proportion of normal BM cells.
Amplification of the c-mpl gene evidenced in AML DNA
was probably responsible for high c-mpl expression in 1 M4
AML (UPN 23). Two AML had translocations involving
the lp36 region, but 1 (UPN 38) did not express c-mpl.
However, UPN 32 expressed a normal-sized c-mp13.7-kb
transcript at relatively low levels (not shown). No gross alteration or amplification of the c-mpl gene was detected in the
DNA of other c-mpl-positive AML.
c-mpl expression in AML was associated with poor prognosis because, compared with c-mpl-negative AML, the cmpl-positive AML had a significantly lower rate of CR in
response to chemotherapy. Chemotherapy resistance disease was the main cause of induction failure in c-mpl-positive AML. Therefore, we wondered whether c-mpl expression would correlate with other factors known to be of
prognostic significance in AML. c-mpl expression did not
significantly correlate with age, leucocyte count, or platelet
count. Characteristics of the 26 patients with c-mpl-positive
AML and of the 25 patients with c-mpl-negative AML were
compared and showed that c-mpl was more frequently expressed in secondary than in de novo AML. Moreover, our
results showed that c-mpl expression correlated with the
presence of cytogenetic abnormalities known to be of major
prognostic value in the outcome of AML in terms of both
CR rate and long-term disease-free s u r ~ i v a l . ~We
~ ~noted
~"
that 1 1 of the 12 AML patients with abnormalities leading
to loss of the whole chromosome or part of chromosome 5
and/or chromosome 7 were c-mpl positive. AML with deletion of chromosomes 5 and/or 7 represent a subset of AML
characterized by poor prognosis usually because of chemotherapy-resistant disease.
Although c-mpl did not correlate with the FAB classification, the fact that 5 erythroleukemias (FAB M6) were all
VIGON ET AL
882
c-mplpositive raised questions concerning the possible influence of the phenotype of these leukemias o n c-mpl expression. Alternatively, because 3 M6 AML patients had secondary AML and 4 of 5 had abnormalities of chromosomes 5
and /or 7 frequently associated with c-mpl-positive AML,
these M6 leukemias could represent a subset of AML in
which c-mpl is frequently expressed.
Another significant prognostic indicator of response to
therapy is the expression of the glycoprotein CD34 antigen
( M Y ~ O ) . ~High
' - ~ ~incidence of secondary AML versus de
novo AML and chromosomal abnormalities covariated
~'
with CD34 expression in a large multivariate a n a l y ~ i s .Our
preliminary results indicate that, although CD34 and c-mpl
expression both correlate with the same poor prognostic factors, their expression does not completely overlap because
some CD34+ AML did not overexpress c-mpl, and, conversely, some c-mpl-positive AML were CD34- (data not
shown). Analysis of a larger series of patients would indicate
to what extent these two markers covanate.
Interestingly, our study o n MDS patients shows that, similar to AML, elevated c-mpl expression could also be associated with a poor prognosis. Indeed, 5 of 1 1 RAEB patients
had increased c-mpl expression versus none of the 5 RSA
patients. A survey of c-mpl expression in a larger panel of
MDS patients is currently being performed in an attempt to
define the precise relationship between c-mpl expression
and the emergence of a blastic cell population.
Further studies on the c-mpl protein will show whether
c-mpl is a ligand binding chain or a part of a multimeric
complex. Because c-mpl has the overall structure of cytokine receptor, one could suppose that expression of the cmpl gene product may promote the proliferation of myeloid
blast cells in vivo.
ACKNOWLEDGMENT
We thank F. Ajchenbaum, F. Picard, M. Sitbon and A. Turhan
for critical reading of the manuscript and V. Chauvin for expert
secretarial support.
REFERENCES
1. Budel LM, Touw IP, Delwel R, Clark S, Lowenberg B: IL-3
and GM-CSF receptors on human acute myelocytic leukemia cells
and the relationship to the proliferative response. Blood 74:565,
1989
2. Park LS, Waldron PE, Friend D, Sassenfeld HM, Price V,
Anderson D, Cosman D, Andrews RG, Bernstein ID, Urdal DL:
Interleukin-3, GM-CSF, and G-CSF receptor expression on cell
lines and primary leukemia cells: Receptor heterogeneity and relationship to growth factor responsiveness.Blood 74:56, 1989
3. Begley CG, Metcalf D, Nicola NA: Primary human myeloid
leukemia cells: Comparativeresponsivenessto proliferativestimulation by GM-CSF or G-CSF and membrane expression of CSF receptors. Leukemia I:l,1987
4. Dubreuil P, Torres H, Courcoul MA, Birg F, Mannoni P:
c-fms expression is a molecular marker of human acute myeloid
leukemias. Blood 72:1081, 1988
5. Ikeda H, Kanakura Y, Tamaki T, Kuriu A, Kitayoma H,
Ishikawa J, Kanayama Y, Yonezawa T, Tarui S, Griffin JD: Expression and functional role of the proto-oncogene c-kit in acute
myeloblastic leukemia cells. Blood 78:2962, 1991
6. Lerner NB, Nocka KH, Cole SR, Qiu F, Strife A, Ashman LK,
Besmer P Monoclonal antibody YB5.B8 identifiesthe human c-kit
protein product. Blood 77:1876, 1991
7. Broudy VC, Smith FO, Lin N, Zsebo KM, Egrie J, Bemstein
I D Blasts from patients with acute myelogenous leukemia express
functional receptors for stem cell factor. Blood 80:60, 1992
8. Depper JM, Leonard WJ, Kranke M, Waldmann TA, Greene
WC: Augmented T cell growth factor receptor expression in
HTLV-1 infected human leukemic T cells. J Immunol 133:169I ,
1984
9. Rambaldi A, Wakamiya N, Vellenga E, Horiguchi J, Warren
MK, Kufe D, Griffin J D Expression of the macrophage colony
stimulating factor and c-fms genes in human acute myeloblastic
leukemic cells. J Clin Invest 81:103O, 1988
IO. Mitjavila MT, le Couedic JP, Casadevall N, Navarro S, VileVal JL, Dubart A, Vainchenker W Autocrine stimulation by erythropoietin and autonomous growth of human erythroid leukemic
cells in vitro. J Clin Invest 88:789, 1991
1 I. Eaves AC, Eaves CJ: Erythropoiesis in culture. Clin Haemato1 13:371, 1984
12. Prchal JF, Axelrad AA: Bone marrow responsesin polycythemia vera. N Engl J Med 290: 1382, 1974
13. Vigon I, Mornon JP, Cocault L, Mitjavila MT, Tambourin
P, Gisselbrecht S, Souyri M: Molecular cloning and characterization of h-mpl, the human homolog of the v-mpl oncogene: Identification of a new member of the hematopoietic growth factor receptor superfamily. Proc Natl Acad Sci USA 89:5640, 1992
14. Cosman D, Lyman SD, Idzerda RL, Beckmann MP, Park
LS, Goodwin RG, March CJ: A new cytokinereceptor superfamily.
Trends Biochem Sci 15:265, 1990
15. Miyajima A, Hara T, Kitamura T Common subunits of
cytokine receptors and the functional redundancy of cytokines.
Trends Biochem Sci 17:378, 1992
16. Renauld JC, Druez C, Kermouni A, Houssiau F, Uyttenhove C, Van Roost E, Van Snick J: Expression cloning of the murine and human interleukin 9 receptor cDNAs. Proc Natl Acad Sci
USA 895690, 1992
17. Souyri M, Vigon I, Penciolelli JF, Heard JM, Tambourin P,
Wendling F A putative truncated cytokine receptor gene transduced by the myeloproliferativeleukemia virus immortalizeshematopoietic progenitors. Cell 63: l 137, 1990
18. Wendling F, Vigon I. Souyri M, Tambourin P Myeloid progenitor cells transformed by the myeloproliferative leukemia virus
proliferate and differentiate in vitro without the addition of growth
factors. Leukemia 3:475, 1989
19. Wendling F, Charon M, Penciolelli JF, Tambourin P Factor-independent erythropoietic precursor cells in leukemia induced
by the myeloproliferativeleukemia virus. Blood 73:1161, 1989
20. Le Conniat M, Souyri M, Vigon I, Wendling F, Tambourin
P, Berger R The human homolog of the myeloproliferative virus
maps to chromosome band lp34. Hum Genet 83:194, 1989
21. Bennet JM, Catowsky D, Daniel MT, Flandrin G, Galton
DAG, Granlick H, Sultan C: Proposals for the classificationof the
acute leukemia. Br J Haematol 33:451, 1976
22. Archimbaud E, Leblond V, Michallet M, Cordonnier C,
Fenaux P, Travade P, Dreyfus F, Jaubert J, Devaux Y, Fiere D:
Intensive sequentialchemotherapywith mitoxantrone and continuous infusion etoposide and cytarabine for previously treated acute
myelogenous leukemia. Blood 77: 1894, 1991
23. Chirgwin JM, Przybyla AE, MacDonald RJ, Cowan NJ,
Rutter WJ, Kirschner MW: Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry
18:5294, 1979
24. Frachet P, Uzan G, Thevenon D, Denarier E, Prandini MH,
c-mpl EXPRESSION IN LEUKEMIAS
Marguerie G: GpIIb and GpIIIa amino acid sequences deduced
from human megakaryocytescDNAs. Mol Biol Res 14:27, 1990
25. Dutrillaux B, Lejeune J: Sur une nouvelle technique d'analyse du caryotype humain. C R Acad Sci 111 272:2638, 1971
26. ISCN: An international system for human cytogenetic nomenclature, in Harnden DG, Klinger HP (eds). Published in collaboration with Cytogenet Cell Genet, Basel, Switzerland, Karger,
1985
27. Kaplan EL, Meier P Nonparametric estimation for incomplete observation. J Am Stat Assoc 53:457, 1958
28. Collins A, Keats BJ, Dracopoli N, Shields DC,Morton NE:
Integration of gene maps: Chromosome I . Proc Natl Acad Sci USA
89:4598, 1992
29. Keating MJ, Cork A, Broach Y , Smith T, Walters RS,
McCredie KB, Trujillo J, Freireich EJ: Toward a clinically relevant
cytogenetic classification of acute myelogenous leukemia. Leuk
Res 11:119, 1987
30. Schouten HC, van Putten WW, Hagemeijer A, Blijham GH,
883
Sonnenveld P, Willemze R, Slater R, van de Lely J, Verdonck LF,
Lowenberg B The prognostic significanceof chromosomal findings
in patients with acute myeloid leukemia in a study comparing the
efficacy of autologous and allogeneicbone marrow transplantation.
Bone Marrow Transplant 8:377, 1991
31. Vaughan WP, Civin CI, Weisenburger DD, Karp JE, Graham ML, Sanger WG, Grierson HL, Joshi SS, Burke PJ: Acute
leukemia expressing the normal human hematopoietic stem cell
membrane glycoprotein CD34 (My 10). Leukemia 2:661, 1988
32. Borowitz MJ, Gockerman JP, Moore JO, Civin CI, Page SO,
Robertson J, Bigner SH: Clinicopathologicand cytogenetic features
of CD34 (My 10)-positiveacute nonlymphocytic leukemia. Am J
Clin Path01 91:265, 1989
33. Geller RB, Zahurak M, Hunvitz CA, Burke PJ, Karp JE,
Piantadosi S, Civin CI: Prognostic importance of immunophenotyping in adults with acute myelocytic leukaemia: The significance
of the stem-cell glycoprotein CD34 (My 10). Br J Haematol76:340,
I990