Leukemia (2005) 19, 1248–1252
& 2005 Nature Publishing Group All rights reserved 0887-6924/05 $30.00
www.nature.com/leu
Mcl-1 is overexpressed in multiple myeloma and associated with relapse and shorter
survival
S Wuillème-Toumi1, N Robillard1, P Gomez2, P Moreau3, S Le Gouill3, H Avet-Loiseau1, J-L Harousseau3, M Amiot2 and
R Bataille2
1
Central Laboratory of Hematology, University Hospital of Nantes, France; 2INSERM UMR 601, Nantes, France; and 3Department
of Clinical Hematology, University Hospital of Nantes, France
We and others have shown that Mcl-1 was essential for the
survival of human myeloma cells in vitro. Furthermore, this
antiapoptotic protein is upregulated by interleukin-6, which
plays a critical role in multiple myeloma (MM). For these
reasons, we have evaluated the expression of Mcl-1 in vivo in
normal, reactive and malignant plasma cells (PC), that is,
myeloma cells from 51 patients with MM and 21 human
myeloma cell lines (HMCL) using flow cytometry. We show that
Mcl-1 is overexpressed in MM in comparison with normal bone
marrow PC. In total, 52% of patients with MM at diagnosis
(P ¼ 0.017) and 81% at relapse (P ¼ 0.014 for comparison with
diagnosis) overexpress Mcl-1. Of note, only HMCL but not
reactive plasmacytoses have abnormal Mcl-1 expression,
although both PC expansions share similar high proliferation
rates. Of interest, Bcl-2 as opposed to Mcl-1, does not
discriminate malignant from normal PC. Finally, the level of
Mcl-1 expression is related to disease severity, the highest
values at diagnosis being associated with the shortest eventfree survival (P ¼ 0.002). In conclusion, Mcl-1, which has been
shown to be essential for the survival of human myeloma cells
in vitro, is overexpressed in vivo in MM in relation with relapse
and shorter survival. Mcl-1 represents a potential therapeutical
target in MM.
Leukemia (2005) 19, 1248–1252. doi:10.1038/sj.leu.2403784
Published online 5 May 2005
Keywords: plasma cells; multiple myeloma; Mcl-1
Introduction
Multiple myeloma (MM) is a rapidly fatal plasma-cell malignancy that evolves mainly in the bone marrow. Myeloma cells
proliferate very slowly and display a weak apoptotic index in
vivo. Whether this is due to abnormalities of cell cycle or
apoptosis remains pending. In any case, myeloma cells turn out
to be resistant to chemotherapy and the prognosis of the disease
remains poor.1 Understanding the mechanisms of survival of
myeloma cells within the bone marrow and of escape
(resistance) of these cells to chemotherapy is a major challenge
in order to define new therapeutical targets and treatments.
Many studies have shown that both the survival and resistance
to chemotherapy of myeloma cells was dependent on (i) bone
marrow environment, (ii) capacity of myeloma cells to interact
with this environment and (iii) survival and growth factors
supplied by this environment (reviewed in Nefedova et al2). The
expression of the prosurvival members of the Bcl-2 family has
been shown to be a key process in the survival of myeloma cells.
More particularly, Mcl-1 expression turned out to be critical for
their survival.3,4 Indeed, knockdown of Mcl-1 by antisenses
induces apoptosis in myeloma cells whereas Bcl-2 or Bcl-xL
antisenses have few or no effect.4–6 Moreover, a critical
Correspondence: Professor R Bataille, INSERM UMR 601, Institut de
biologie, 9 quai moncousu, 44093 Nantes Cedex 01, France;
Fax: þ 33 2 40 08 47 78; E-mail: [email protected]
Received 24 November 2004; accepted 4 March 2005; Published
online 5 May 2005
threshold level of Mcl-1 is necessary for viability of myeloma
cells.3 Finally, Mcl-1 was found to be the only antiapoptotic Bcl2 family member, whose level of expression was modified by
cytokine treatment of myeloma cells.7–10 These data suggest that
increased levels of Mcl-1 during disease progression could be an
important mechanism in the acquisition of resistance to
chemotherapy in MM. Surprisingly, expression of Mcl-1 has
not been evaluated in MM in vivo at diagnosis and during
disease progression, data being available only on a small
number of patients.11 Thus, we used flow cytometry to evaluate
the expression of Mcl-1 in normal, reactive and malignant
plasma cells (PC), that is, myeloma cells from 51 patients with
MM, either at diagnosis or relapse and 21 human myeloma cell
lines (HMCL).
Materials and methods
Mononuclear bone marrow and peripheral blood PC,
freshly explanted human myeloma cells and HMCL
Seven bone marrow samples came from healthy bone marrow
donors during the process of bone marrow transplantation. In
total, 51 consecutive patients with MM were evaluated,
including 25 newly diagnosed patients and 26 at the time of
medullary relapse. No serial study of individual patients could
be performed. All the patients at diagnosis had symptomatic
disease and fulfilled the diagnostic criteria of the Southwest
Oncology Group of United States of America.12 Relapse was
defined by the occurrence of any new symptom relevant to MM
(mainly anemia and/or new lytic bone lesions) and/or increase
of the monoclonal component 425% above previous level. The
University Hospital of Nantes review board approved these
studies. Informed consent was provided according to the
Declaration of Helsinki. In total, 21 HMCL were included in
this study, commercially available (U266, LP1, L363, OPM2,
NCIH929, RPMI8226, JJN3), gifts (JIM3 from Leiff Bergsagel,
Cornell University, NY, USA and KMS12PE, KMS18, KMS11,
KMS12BM,13 or established by ourselves (NAN1, NAN2,
NAN3, XG1, XG2, XG6, MDN, SBN, BCN).
Eight cases of reactive plasmacytoses (RP) were studied. In all
these cases, transient polyclonal circulating PC were observed
in the peripheral blood. These RP were detected initially by
cytological identification of PC on smears and a final diagnosis
was established by immunophenotyping.14
Monoclonal antibodies (mAbs) and reagents
Anti-CD45 (clone J33), anti-CD138 (B-B4) were purchased from
Beckman Coulter (Miami, F, USA). Anti-CD38 (HB7) and
isotype control mAbs were from BD Bioscience (San Jose,
CA, USA) and anti-Bcl-2 from Dako (Glostrup, Denmark). The
Mcl-1 expression in normal and malignant human plasma cells
S Wuillème-Toumi et al
anti-Mcl-1mAb (clone 22), directly conjugated to PE, was
customized by BD Bioscience (San Jose, CA, USA).
Cell staining for immunophenotyping
Since Gojo et al15 have shown that Mcl-1 mRNA and protein
both have a short half-life in MM cells, cells were treated within
2 h following bone marrow aspiration. After ficoll density
separation, bone marrow mononuclear cells (5 105) were
stained in a four-color assay with optimal concentration of each
mAb per test. For staining of surface molecules, mononuclear
cells were incubated 20 min at room temperature with 10 ml
volumes of pretitrated antibody: anti-CD38-APC, anti-CD45
(FITC or PC5) and anti-CD138 (PC5 or Phyco-erythrin (PE))
mAbs, in PBS containing 20% AB serum. Then, the mononuclear cells were washed twice and permeabilized using Intra
prep permeabilization reagent (Beckman Coulter) recommended by the manufacturer and stained with anti-Mcl-1-PE
or anti-Bcl-2-FITC and with the control isotype-PE or FITC
mAbs. The cells were fixed in PBS 1% formaldehyde.
Flow cytometry analysis
Data were acquired by means of a Becton Dickinson FACS
Calibur with CellQuest Pro software (BD Bioscience). To
increase the sensitivity and accuracy of the analysis, acquisition
was performed in two consecutive steps: in the first step, a total
of 15 000 events/tube was acquired and in the second step
acquisition through a large ‘live-gate’ drawn on SSC log/CD38
strongly positive cells was performed. In this latter step, a
minimum of 1000 PC was acquired.16
For analysis, PC were identified using a sequential gating
strategy: an initial region (R1) was set around cells expressing a
high level of CD38 with their typical side scatter and then a
second region (R2) was set on the CD38 vs CD138. To exclude
contaminating events as well as apoptotic cells and cellular
debris, the third region (R3) was set on the side light scatter (SSC)
vs the forward light scatter (FSC).
The final PC gate used was a combination of R1, R2 and R3
(CD38, CD138, scatter characteristics). Analysis of CD38 vs
CD138 expression provides the best separation of PC from other
leucocytes. The mean fluorescence intensity ratio (MFIR) was
obtained by dividing the mean fluorescence intensity of each
antigen by the mean fluorescence intensity of its isotypic
control.
Survival analysis
In total, 25 patients (median age 65 years) with newly diagnosed
MM received treatments according to ongoing protocols of the
IFM group, and are included in the survival analysis. Overall
survival (OS) was calculated from the first day of treatment to the
date of death, and event-free survival (EFS) was calculated from
the first day of treatment to the date of relapse, progression or
death. Analysis of prognostic factors for survival was performed
including usual clinical (age, sex), biological (isotype, b2microglobulin, CRP, creatinine, hemoglobin, albumin, calcium)
and cytogenetic (chromosome 13 deletion and 14q32 rearrangement) characteristics at presentation, and the level of Mcl-1
expression.
Statistical analysis
1249
For statistical analysis, we used nonparametric tests, that is, the
Wilcoxon rank sum test, the Fisher exact test and the w2 method
when appropriate. Curves for OS and EFS were plotted
according to the method of Kaplan and Meier and were
compared by the log-rank test. Multivariate analysis was
performed using the Cox model.
Results
Mcl-1 is overexpressed in MM patients
The level of expression of Mcl-1 was determined by a four-color
immunofluorescence analysis in normal PC and in myeloma
cells from MM at diagnosis. All bone marrow PC from healthy
donors expressed Mcl-1. The intensity of protein expression is
defined by the MFIR. MM patients at diagnosis did also express
Mcl-1 on the whole malignant cell population. Myeloma cells at
diagnosis have a higher MFIR of Mcl-1 than normal bone
marrow PC (median value: 11.7, range 4–22 vs 8.7, range 5–9.7,
P ¼ 0.025) (Figure 1a). Furthermore, 52% of MM at diagnosis
presented abnormal Mcl-1 expression (P ¼ 0.025) (ie, Mcl-1
expression above the upper limit defined by 2 standard
deviations above the mean value of normal PC). Of note, when
Bcl-2 expression was considered, no difference was found
between normal bone marrow PC and myeloma cells at
diagnosis (7.2, range 3–16 vs 5.7 range 4–10) (Figure 1c).
Mcl-1 expression correlates with disease progression
To investigate the clinical significance of Mcl-1 in MM, we
investigated its expression in MM patients either at diagnosis
(n ¼ 25) or relapse (n ¼ 26). Analysis of the MFIR of Mcl-1
indicated a median value of 11.7 (range 4–22) for MM patients
at diagnosis vs 15 (range 6–30) for patients at relapse (P ¼ 0.04)
(Figure 1a). In total, 81% of MM patients at relapse had
abnormal Mcl-1 expression as opposed to 52% at diagnosis
(P ¼ 0.04). These results clearly indicated an increase of Mcl-1
expression at the time of relapse and thus during disease
progression. However, serial studies in individual patients will
be necessary to confirm this concept. Of note, when Bcl-2
expression was considered, no difference was found between
MM at diagnosis and MM in relapse (7.2, range 3–16 vs 9, range
3–20 (Figure 1c).
HMCL but not reactive plasmocytoses (RP) overexpress
Mcl-1
To see whether or not the overexpression of Mcl-1 in MM was
related to malignancy and not simply to excess of proliferation,
we compared its expression between HMCL and RP, both
situations in which the labeling index was 420%.14 Only
intrinsic malignancy discriminate HMCL from RP. Of note,
HMCL overexpress Mcl-1 (P ¼ 0.01) (median value: 13.5, range
7.4–29) whereas 3 RP had low Mcl-1 expression (median value:
5, range 2.5–5) (Figure 1b). On the other hand, when Bcl-2 was
considered, low values were observed in both HMCL and RP
(Figure 1d) as previously described.14 However, a larger number
of RP with Mcl-1 expression evaluation will be necessary.
Leukemia
Mcl-1 expression in normal and malignant human plasma cells
S Wuillème-Toumi et al
1250
35
35
MFIR
Mcl-1
Mcl-1
30
30
25
25
20
20
15
15
10
10
5
5
0
0
Normal bone
marrow PC
(n= 7)
MM
Diagnosis
(n= 25)
MM
Relapse
(n= 26)
c
RP
(n=21)
(n=3)
d
35
35
Bcl-2
MFIR
HMCL
Bcl-2
30
30
25
25
20
20
15
15
10
10
5
5
0
0
Normal bone
marrow PC
(n= 7)
MM
Diagnosis
(n= 24)
MM
Relapse
(n= 19)
HMCL
RP
(n=20)
(n=8)
Figure 1
Comparison of Mcl-1 and Bcl-2 expression in normal and malignant PC. Comparison of Mcl-1 (a) or Bcl-2 (c) expression in normal
bone marrow PC, malignant PC from MM patients either at diagnosis or relapse. Comparison of Mcl-1 (b) or Bcl-2 (d) expression in HMCL and RP.
Horizontal lines: mean of fluorescence intensity ratio in each group of patients. MFIR: mean fluorescence intensity ratio.
Mcl-1 expression correlates with disease severity and
clinical outcome
In this group of 25 newly diagnosed patients, two deaths were
observed due to progressive disease, with a median follow-up
for living patients of 12 months (range: 4–40). The OS was
90.7% at 40 months ( þ 6%). The median EFS was 26 months for
the whole group of patients (range: 4–40). The single factor
influencing EFS was the level of Mcl-1 expression (median
MFIR: 13, range: 5.8–22). The median duration of EFS was 12
months (range: 4–18) in the group of 11 patients with an
abnormal expression of Mcl-1 (two deaths and three relapses
Leukemia
were observed), as compared with an EFS of 66.7% at 40
months (median not reached) in the group of 14 patients with a
normal expression of Mcl-1 (1 relapse observed) (P ¼ 0.002)
(Figure 2). These two groups of patients (normal and abnormal
expression of Mcl-1) were identical regarding initial characteristics and regarding treatment regimen: the percentage of
patients treated with either conventional treatment, or single
autologous stem cells transplantation, or tandem transplant was
similar in both groups of patients. In this small cohort of patients,
chromosome 13 deletion, beta-2-microglobulin level or other
presenting features did not statistically influence survival as
single parameters.
Mcl-1 expression in normal and malignant human plasma cells
S Wuillème-Toumi et al
1251
The molecular mechanisms involved in this malignant overexpression turn out to be a major challenge for the next future to
clarify the pathogenesis of MM. It will be more particularly
important to know whether the overexpression of Mcl-1 is
related to an abnormal response to cytokines like interleukin-6
or to mutations of the promoter of the Mcl-1 gene, a mechanism
recently involved in chronic lymphocytic leukemia.19,20
1.0
0.9
Mcl-1 low
0.8
0.7
Mcl-1 high
0.6
0.5
0.4
Acknowledgements
0.3
This work was supported by The Ligue Nationale contre le Cancer
(équipe labelisée 2004).
0.2
0.1
P = .002
References
0
0
6
12
18
Event-free survival
24
30
months
Figure 2
Event-free survival (EFS) of MM patients at diagnosis
according to Mcl-1 level of expression. Abnormal level of expression:
Mcl-1413.2, 3 s.d. above the mean level of normal PC.
Discussion
A critical role of Mcl-1 has been established in the biology of
MM.4,17,18 However, its expression has never been evaluated in
vivo in patients with MM. For these reasons, we have evaluated
the expression of Mcl-1 in both normal (bone marrow), reactive
(RP) and malignant (MM and HMCL) PC. Of note, RP are
characterized by expansion of highly proliferative normal PC
progenitors and precursors (labeling index 420%) as previously
described by ourselves,14 whereas HMCL are expansions of
highly proliferative malignant PC.
By evaluating the level of Mcl-1 expression in myeloma cells
of patients either at diagnosis or relapse in comparison with that
of normal PC in the bone marrow, we have shown that about
half of MM patients at diagnosis and more than three quarters of
MM patients at the time of relapse overexpress Mcl-1. The
correlation of Mcl-1 expression with disease activity is
confirmed by the high expression of Mcl-1 we found in more
than 80% of HMCL. In comparison, Bcl-2 did not discriminate
MM from normal PC. Thus, although it is not (yet) possible to
evaluate any correlation between Mcl-1 expression and
sensitivity to chemotherapy (achievement of complete remission), we have already shown that overexpression of Mcl-1
directly correlates with disease status generally associated with
chemoresistance (relapses, terminal phase with immortalization). Finally, this is confirmed by the significant impact of Mcl-1
on EFS. However, further investigation will be necessary to
confirm these points, that is, (i) real upregulation of Mcl-1 during
disease progression in individual patients, and (ii) the final
impact of Mcl-1 on the clinical outcome of patients with MM.
Although the number of RP is too low to draw any conclusion,
the low levels of Mcl-1 detected in these three cases are of major
importance. RP have been already characterized by their high
labeling index and low expression of Bcl-2. With regard to Mcl1, their comparison with highly proliferative myeloma cells
(HMCL) is of major interest. Indeed, in these cases of HMCL (ie,
with a labeling index ranging from 25 to 30% as that of RP), Bcl2 is low but Mcl-1 is high. Thus, Mcl-1 but not Bcl-2
discriminates RP from HMCL, that is, normal from malignant
expansions of PC. This shows that the overexpression of Mcl-1 is
clearly related to malignancy rather than to simply proliferation.
1 Bataille R, Harousseau JL. Multiple myeloma. N Engl J Med 1997;
336: 1657–1664.
2 Nefedova Y, Landowski TH, Dalton WS. Bone marrow stromalderived soluble factors and direct cell contact contribute to de
novo drug resistance of myeloma cells by distinct mechanisms.
Leukemia 2003; 17: 1175–1182.
3 Zhang B, Gojo I, Fenton RG. Myeloid cell factor-1 is a critical
survival factor for multiple myeloma. Blood 2002; 99: 1885–1893.
4 Derenne S, Monia B, Dean NM, Taylor JK, Rapp MJ, Harousseau JL
et al. Antisense strategy shows that Mcl-1 rather than Bcl-2 or Bclx(L) is an essential survival protein of human myeloma cells. Blood
2002; 100: 194–199.
5 van de Donk NW, Kamphuis MM, van Dijk M, Borst HP, Bloem
AC, Lokhorst HM. Chemosensitization of myeloma plasma cells by
an antisense-mediated downregulation of Bcl-2 protein. Leukemia
2003; 17: 211–219.
6 Liu Q, Gazitt Y. Potentiation of dexamethasone-, paclitaxel-, and
Ad-p53-induced apoptosis by Bcl-2 antisense oligodeoxynucleotides in drug-resistant multiple myeloma cells. Blood 2003; 101:
4105–4114.
7 Puthier D, Thabard W, Rapp M, Etrillard M, Harousseau J, Bataille
R et al. Interferon alpha extends the survival of human myeloma
cells through an upregulation of the Mcl-1 anti-apoptotic
molecule. Br J Haematol 2001; 112: 358–363.
8 Puthier D, Derenne S, Barille S, Moreau P, Harousseau JL, Bataille
R et al. Mcl-1 and Bcl-xL are co-regulated by IL-6 in human
myeloma cells. Br J Haematol 1999; 107: 392–395.
9 Jourdan M, Veyrune JL, Vos JD, Redal N, Couderc G, Klein B.
A major role for Mcl-1 antiapoptotic protein in the IL-6-induced
survival of human myeloma cells. Oncogene 2003; 22:
2950–2959.
10 Jourdan M, De Vos J, Mechti N, Klein B. Regulation of Bcl-2-family
proteins in myeloma cells by three myeloma survival factors:
interleukin-6, interferon-alpha and insulin-like growth factor 1.
Cell Death Differ 2000; 7: 1244–1252.
11 Zhang B, Potyagaylo V, Fenton RG. IL-6-independent expression of Mcl-1 in human multiple myeloma. Oncogene 2003; 22:
1848–1859.
12 Durie BGM, Salmon SE. Multiple myeloma macroglobulinaemia
and monoclonal gammopathies. In: Hoffbrand AV, Brain MC, Hish
J (eds). Recent Advances in Haematology. Edinburgh Churchill
Livingstone, 1977, pp 243–261.
13 Ohtsuki T, Yawata Y, Wada H, Sugihara T, Mori M, Namba M.
Two human myeloma cell lines, amylase-producing KMS-12-PE
and amylase-non-producing KMS-12-BM, were established from a
patient, having the same chromosome marker, t(11;14)(q13;q32).
Br J Haematol 1989; 73: 199–204.
14 Jego G, Robillard N, Puthier D, Amiot M, Accard F, Pineau D et al.
Reactive plasmacytoses are expansions of plasmablasts retaining
the capacity to differentiate into plasma cells. Blood 1999; 94:
701–712.
15 Gojo I, Zhang B, Fenton RG. The cyclin-dependent kinase
inhibitor flavopiridol induces apoptosis in multiple myeloma cells
through transcriptional repression and down-regulation of Mcl-1.
Clin Cancer Res 2002; 8: 3527–3538.
16 Robillard N, Avet-Loiseau H, Garand R, Moreau P, Pineau D, Rapp
MJ et al. CD20 is associated with a small mature plasma cell
Leukemia
Mcl-1 expression in normal and malignant human plasma cells
S Wuillème-Toumi et al
1252
morphology and t(11;14) in multiple myeloma. Blood 2003; 102:
1070–1071.
17 Choi Y, Zhang J, Murga C, Yu H, Koller E, Monia BP et al. PTEN,
but not SHIP and SHIP2, suppresses the PI3K/Akt pathway and
induces growth inhibition and apoptosis of myeloma cells.
Oncogene 2002; 21: 5289–5300.
18 Munshi NC, Hideshima T, Carrasco D, Shammas M, Auclair D,
Davies F et al. Identification of genes modulated in multiple
Leukemia
myeloma using genetically identical twin samples. Blood 2004;
103: 1799–1806.
19 Moshynska O, Sankaran K, Pahwa P, Saxena A. Prognostic
significance of a short sequence insertion in the MCL-1 promoter
in chronic lymphocytic leukemia. J Natl Cancer Inst 2004; 96:
673–682.
20 Kitada S, Reed JC. MCL-1 promoter insertions dial-up aggressiveness of chronic leukemia. J Natl Cancer Inst 2004; 96: 642–643.