A survey of the signaling pathways involved in ... - Nature

Oncogene (2006) 25, 781–794

& 2006 Nature Publishing Group All rights reserved 0950-9232/06 $30.00 www.nature.com/onc

ORIGINAL ARTICLE

A survey of the signaling pathways involved in megakaryocytic differentiation of the human K562 leukemia cell line by molecular and c-DNA array analysis A Jacquel1, M Herrant1, V Defamie2, N Belhacene1, P Colosetti1, S Marchetti1, L Legros3, M Deckert4, B Mari2, J-P Cassuto3, P Hofman5 and P Auberger1 1

INSERM U526, Physiopathologie de la Survie et de la Mort Cellulaires, Equipe Labelliseé par la Ligue Nationale contre le Cancer, Universite´ de Nice Sophia-Antipolis, IFR50, Faculte´ de Me´decine, Avenue de Valombrose, 06107 Nice Cedex 2, France; 2 Institut de Pharmacologie Moleculaire et Cellulaire UMR 6097 Centre National de la Recherche Scientifique, 660 Route des Lucioles, 06560 Valbonne, France; 3Service d’He´matologie, Hoˆpital de l’Archet, 151 route de Saint-Antoine de Ginestiere, BP3079, 06202 Nice Cedex 2, France; 4INSERM U576, Hopital de l’Archet, 151 route de Saint-Antoine de Ginestie`re, 06202 Nice, France and 5INSERM U721 Universite´ de Nice Sophia-Antipolis, IFR50, Faculte´ de Me´decine, Avenue de Valombrose, 06107 Nice Cedex 2, France

The K562 cell line serves as a model to study the molecular mechanisms associated with leukemia differentiation. We show here that cotreatment of K562 cells with PMA and low doses of SB202190 (SB), an inhibitor of the p38 MAPK pathway, induced a majority of cells to differentiate towards the megakaryocytic lineage. Electronic microscopy analysis showed that K562 cells treated with PMA þ SB exhibited characteristic features of physiological megakaryocytic differentiation including the presence of vacuoles and demarcation membranes. Differentiation was also accompanied by a net increase in megakaryocytic markers and a reduction of erythroid markers, especially when both effectors were present. PMA effect was selectively mediated by new PKC isoforms. Differentiation of K562 cells by the combination of PMA and SB required Erk1/2 activation, a threshold of JNK activation and p38 MAPK inhibition. Interestingly, higher concentrations of SB, which drastically activated JNK, blocked megakaryocytic differentiation, and considerably increased cell death in the presence of PMA. c-DNA microarray membranes and PCR analysis allow us to identify a set of genes modulated during PMAinduced K562 cell differentiation. Several gene families identified in our screening, including ephrins receptors and some angiogenic factors, had never been reported so far to be regulated during megakaryocytic differentiation. Oncogene (2006) 25, 781–794. doi:10.1038/sj.onc.1209119; published online 26 September 2005 Keywords: megakaryocyte differentiation; PKC; MAPK; c-DNA array; leukemia

Correspondence: P Auberger, INSERM U526, Faculte´ de Me´decine Pasteur, Avenue de Valombrose, Nice 06107, France. E-mail: [email protected] Received 3 June 2005; revised 4 August 2005; accepted 15 August 2005; published online 26 September 2005

Introduction Hematopoiesis in vertebrates is a complex, multistep and highly ordered process that takes place in specific organs during development. It requires cell-specific gene expression that directs growth and differentiation of a pluripotent cell to acquire the phenotype changes of mature blood cells. The human chronic myelogenous (CML) cell line K562 has been used extensively as a model for the study of leukemia differentiation (Lam et al., 2000; Matsumura et al., 2000; Racke et al., 2001). K562 behaves as pluripotent hematopoietic precursor, expressing in undifferentiated conditions markers for both erythroid and megakaryocytic lineages. K562 cells can undergo further differentiation in both megakaryocytic and erythroid lineages depending on the stimulus. Phorbol esters such as PMA-stimulated megakaryocytic differentiation (Shelly et al., 1998; Kim et al., 2001; Dorsey et al., 2002; Pettiford and Herbst, 2003), whereas hemin, hydroxyurea, Ara-C or as reported more recently imatinib mesylate (gleevec) induced erythroid differentiation of this cell line (Belhacene et al., 1998; Park et al., 2001; Jacquel et al., 2003; Kawano et al., 2004a, b). Megakaryocytic differentiation of K562 cells induced by PMA mimics, in part, the physiologic process that takes place in the bone marrow in response to a variety of stimuli (Long et al., 1990). This differentiation process is accompanied by changes in cell morphology, adhesive properties, cell growth arrest, endomitosis, and acquisition of specific markers of megakaryocytes such as CD41, CD61, and CD10 (Belhacene et al., 1998; Herrera et al., 1998; Dorsey et al., 2002). The signaling cascade that leads to PMA-induced differentiation of K562 cells has been partially described. and a general role of PKC in the megakaryocyte differentiation process has been previously established (Murray et al., 1993; Goldfarb et al., 2001; Racke et al., 2001). The PKC family comprises at least 11 isotypes classified into three subgroups according to their structure and cofactor

Signaling pathways in leukemia cell differentiation A Jacquel et al

782

requirements: Conventionnal PKCs (PKC: PKCa, PKCb1, PKCb2, and PKCg), novel PKCs (PKCd, PKCe, PKCZ, PKCy , and PKCm) and atypical PKCs (PKCz, PKCi/l) (Ron and Kazanietz, 1999; Nalefski and Newton, 2001; Parker and Murray-Rust, 2004). Even though a general role for PKC in megakaryocytic differentiation of K562 cells has been established, the nature of the PKC isotypes, the signaling pathways downstream of PKC activation, and the modulation of gene expression that follows from this process are currently poorly determined. More recently, several reports have also highlighted the role of the MEK/MAPK pathway in the modulation of megakaryocytic differentiation downstream from PKC activation (Racke et al., 1997; Rouyez et al., 1997). As previously mentioned, Erk1/2 activation is required for PMA-induced differentiation of K562 cells towards the megakaryocytic lineage, while both inhibition of this pathway and activation of the p38 MAPK are likely to play a role in signal transduction mechanisms leading to erythroid differentiation (Park et al., 2001; Huang et al., 2004; Uddin et al., 2004). Although Erk1/2 activation is thought to be important for differentiation of K562 cells towards the megakaryocytic lineage, the role of other pathways such as p38 MAPK and JNK is far less documented. In the present study, using K562 cells as a cellular model of CML differentiation towards different lineages, we established the main requirements for PMAmediated megakaryocytic differentiation. We show that PMA induced a PKC-dependent, Erk1/2, JNK, p38 MAPK-dependent and PI3K-independent differentiation program in K562 cells. More precisely, it results from our data that upon PMA stimulation, new PKC isotypes trigger Erk1/2 activation, and moderate JNK stimulation and p38 MAPK inhibition. The role of p38 MAPK inhibition in the process of megakaryocytic differentiation seems particularly important since two inhibitors of this pathway were shown to drastically increase differentiation towards the megakaryocytic lineage. Moreover, we reported that inhibition of the p38 MAPK pathway is necessary for both a quantitative and qualitative megakaryocytic differentiation (i.e. nearly 100% of cells differentiate and many of them exhibit features of terminal megakaryocyte differentiation). By contrast, higher concentrations of p38 MAPK inhibitors markedly reduced differentiation and induced cell death in PMA-treated cells. Finally, c-DNA array analysis allows us to characterize a panel of genes whose expression was strikingly modulated by PMA. Some of these genes have never been reported so far to be expressed or involved during megakaryocytic differentiation. The potential role of these new identified genes in the regulation of the differentiation programs of K562 cells are discussed. Results Low concentrations of the p38 MAPK inhibitor SB202190 favor megakaryocytic differentiation of K562 cells while higher doses inhibit it Exposure of K562 cells to PMA induces megakaryocytic differentiation. This differentiation program is characOncogene

terized by morphological modifications, easily detectable by phase contrast microscopy or after May– Grunwald–Giemsa staining (Figure 1A). After 48 h of treatment, a net increase in cell size and the presence of some vacuoles were observed (Figure 1A, a–e). In a first series of experiments, based on the use of pharmacological inhibitors, we established that GF109203X (GFX), an inhibitor of new and conventional PKCs (Figure 1A), and to a lesser extent Uo126 and PD98059, two MEK1 inhibitors (not shown), inhibited PMAmediated megakaryocytic differentiation of K562 cells. By contrast, LY294002 and wortmannin, two inhibitors of PI3 K pathway, and Go¨6976, an inhibitor of conventional PKCs, had no effect (not shown). Interestingly, the morphological modifications characteristic of megakaryocytic differentiation appeared earlier when 5 mM of SB202190 (SB), an inhibitor of p38 MAPK, was added concomitantly to PMA (Figure 1A, b–e). A more differentiated phenotype was also observed in MEG-01 cells incubated in the presence of the combination PMA þ SB (not shown). Neither SB nor GFX alone induced differentiation of K562 cells. Figure 1B illustrates a quantitative analysis of megakaryocyte differentiation as a function of time in the presence of PMA and different inhibitors of various signaling pathways. At 24 h no differentiation was detected in the presence of PMA, while approximately 20% of the cells already exhibited morphological evidence of differentiation in the presence of PMA þ SB. More than 90% of K562 cells differentiated towards the megakaryocytic lineage after 72 h in the presence of the combination of PMA and SB. Electronic microscopy analysis allowed us to better analyse the ultrastructural modifications in differentiated K562 cells. In the presence of PMA and SB, giant cells (30–50 mm) containing huge vacuoles, as compared to cells treated with PMA alone, were visualized (Figure 1C, a–c). Larger magnifications allow us in certain circumstances to evidence some structures inside these vacuoles (not shown). Numerous cellular structures resembling demarcation membranes were also visible inside the cytosol of differentiating megakaryocytes, more particularly in cells that received the PMA þ SB combination (Figure 1C, c). These membranes have many things in common with the demarcation membranes found in the course of physiological megakaryocytic differentiation (Cramer et al., 1997). Finally, higher concentrations of SB (20–40 mM) were found to inhibit PMA-mediated differentiation and to induce K562 cell apoptosis in the presence of the phorbol ester as judged by both morphological analysis and caspase assay (Figure 2a and b). Induction and repression of specific differentiation markers during differentation of K562 cells by PMA A representative flow cytometric analysis for the expression of CD44, CD41 and CD10, is shown in Figure 3. CD44 expression was used here as a positive control of PMA stimulation. CD41 and CD10 have been previously reported to represent specific markers of megakaryocytic differentiation (Belhacene et al., 1998;


783

A

PMA

Control

20 µm

PMA + GFX

GFX

PMA + SB

SB

24 h

a

48 h

b

72 h

c

48 h

d

48 h

e

Control

10 µm

B

C

100

48 h 90 a Cells with megakaryocyte like features (%)

80 Control PMA PMA+SB SB

70 60

2 µm

b PMA

50

N

n

Control

40

v

n

N

N v

2 µm

2 µm

30

dm

v 20

c

10

PMA + SB

v m

v v v

0 0

24

48

72 96 120 144 168 Time (h)

n 2 µm

N 2 µm

Figure 1 SB202190 favors megakaryocytic differentiation of K562 cells. K562 cells were cultured for the times indicated in the presence of 10 ng/ml PMA, 1 mM GF109203X (GFX), 5 mM SB202190 (SB), or the combinations of PMA þ SB or PMA þ GFX. (A) Light microscopic analysis: the cells were photographed either directly ( 100, 200) or after staining with May–Grunwald–Giemsa stain ( 200). (B) Quantification: the percentage of differentiated cells corresponds to the ratio of the number of large vacuolated and/ or adherent cells on the total number of cells present in the field considered. Calculation was performed at various times and under the different conditions described above by three different examiners in a blind manner. (C) Electronic microscopy: cells were incubated for 48 h in the presence of PMA or the combination of PMA þ SB. They were prepared for electronic microscopy and finally photographed ( 2000 to 4000). N: nucleus, n: nucleolus, v: vacuole, dm: demarcation membranes.

Herrera et al., 1998; Dorsey et al., 2002). CD44 expression drastically increased after 48 h in the presence of PMA, an effect abrogated by GFX. PMA treatment resulted in a significant augmentation of both CD41 and CD10 expression, which was further increased by SB

(Figure 3a). As expected, GFX abrogated the PMAmediated increase in CD41 and CD10 expression. We also investigated by RT–PCR, the expression of mRNA specific for megakaryocytic or erythroid differentiation. As shown in Figure 3b, PMA induced CD10 and Oncogene


784 Apoptotic cells %

a

b

10 µm

80 70 60 50 40 30 20 10 0

control PMA

0

10

20 30 SB202190 (µM)

40

Control

PMA

PMA + SB 5µM

PMA + SB 20µM

PMA + SB 40µM

SB 40µM

Figure 2 High concentrations of SB202190 inhibit PMA-mediated differentiation and induce K562 cell apoptosis. K562 cells were cultured for 48 h in the presence of 10 ng/ml PMA and/or different concentrations of SB202190 (1–40 mM). (a) Facs analysis of apoptosis: Cells were stained using the antiactive caspase-3-PE Mab Apoptosis Kit. Fluorescence was measured by using the FL1 channel of a FACScan. (b) Light microscopic analysis: cells were photographed either directly (magnification 200) or following incubation for 48 h in the presence of PMA and/or various concentrations of SB202190 (SB).

GATA-2 mRNA expression, two markers of megakaryocytic differentiation, while it reduced the expression of several erythroid markers including glycophorin A, g-globin, and CD36. Interestingly, in the presence of SB, both induction of megakaryocytic and repression of erythroid markers by PMA were significantly increased (Figure 3b), supporting our finding of a rise in megakaryocytic differentiation in the presence of both effectors (Figure 1). Of note, SB, independently of PMA, increased c-mpl mRNA indicating that TPO receptor expression could be negatively regulated by the p38 MAPK pathway. GFX, as expected, was also shown to abrogate the positive and negative effects of PMA on megakaryocytic and erythroid differentiation, respectively. Signaling pathways controlling megakaryocytic differentiation of K562 cells It is currently admitted that the effects of phorbol esters are mediated by conventional PKCs a, b1, b2, and g and new PKCs d, e, y, and Z, whose activation requires a relocation from the cytoplasm to the plasma membrane (Schaefer et al., 2001; Tanimura et al., 2002). We used the latter feature to analyse the implication of new and conventional PKCs during megakaryocytic differentiation. Expression of the different PKC isoforms was therefore analysed by Western blot in cytoplasmic and Oncogene

microsomal fractions following treatment of K562 cells with PMA or the combination of PMA þ SB. Globally, 15 min following PMA stimulation, n and cPKCs disappeared from the cytoplasmic fraction concomitantly to their redistribution in the microsomal fraction (Figure 4a). SB by itself had no effect on PMA-induced PKCs relocation. Finally, using a pan phospho-PKC antibody, we also checked that relocated PKCs were active in the microsomal fraction. However, using tet-on K562 cells inducible for the expression of wild type or constitutively active PKC a, b, e, and y, we were unable to demonstrate that any of these PKC isotypes used alone can induce megakaryocytic differentiation (not shown). It is thus likely that more than one PKC isoform is required for the differentiation process. Finally, the lack of redistribution of PKCi, a PMAinsensitive isoform, is indicative of the specificity of the redistribution of novel and conventional PKCs by phorbol esters. To further characterize the molecular events involved during megakaryocytic differentiation of K562 cells by PMA, we investigated the main signaling pathways potentially required for this process and more particularly Erk1/2, JNK, p38 MAPK and Akt activation. The phosphorylation status of these kinases is thought to represent a reliable index of their activity (Herrant et al., 2002). In the absence of PMA, the significant activation of Erk1/2 can be explained by the presence in these cells of the constitutively activated Bcr-Abl chimaeric protein (Jacquel et al., 2003) (Figure 4b). PMA was found to stimulate phosphorylation of Erk1/2, JNK and Akt at 30 min, an effect maintained for approximately 6 h, but inhibit p38 MAPK phosphorylation after 2 h of treatment (Figure 4b). In K562 cells stimulated with the combination of PMA and SB, there was a further increase in both Erk1/2 and JNK phosphorylation and no modification of the Akt phosphorylation status. Interestingly, GFX was found to abolish both the stimulatory effect of PMA on Erk1/2, JNK and Akt phosphorylation, and the inhibitory effect of the phorbol ester on p38 MAPK. Identical results were obtained using the MEG-01 CML cell line (not shown). Of note, the effect of SB was even more pronounced at the c-Jun phosphorylation level (Figure 4b). As higher concentrations of SB induced apoptosis and repressed megakaryocytic differentiation in the presence of PMA (Figure 2), we analysed the effect of increasing concentrations of SB on Erk1/2, p38 MAPK, JNK activation, and c-Jun phosphorylation. Low doses of SB (1–5 mM) moderately increased JNK activation and c-Jun phosphorylation, while higher concentrations (20–40 mM) drastically induced phosphorylation of both proteins, an effect that correlates with increased cell death (Figures 2a and 4c). Interestingly, potentialization of the PMA-mediated Erk1/2 activation by SB was reproduced in cells from a patient with CML (Figure 4d). Moreover, in primary CML cells, phosphorylation of Erk1/2 was very long lasting and still detected 120 h following the onset of PMA treatment. Altogether our results strongly suggest that megakaryocytic differentiation of K562 cells (i) involved new PKC


785

a

Control

Cell number

PMA + GFX

GFX

PMA + SB

SB

100 100 80 80 87.8% 9% 60 60 40 40 20 20 0 0 0 1 2 3 4 10 10 10 10 10 100 101 102 103 104

100 80 6.4% 60 40 20 0 100 101 102 103 104

100 80 90.2% 60 40 20 0 100 101 102 103 104

100 80 60 40 20 0

100 80 1.6% 60 40 20 0 100 101 102 103 104

100 100 80 80 5% 57.2% 60 60 40 40 20 20 0 0 0 1 2 3 4 10 10 10 10 10 100 101 102 103 104

100 80 4.4% 60 40 20 0 100 101 102 103 104

100 71.7% 80 60 40 20 0 100 101 102 103 104

100 80 60 40 20 0

100

100

100 80 8.7% 60 40 20 0 100 101 102 103 104

CD44

CD41

PMA

100

48.1%

3.2%

9%

100 101 102 103 104

1.6%

100 101 102 103 104 100

54.9%

1.8%

CD10 0 100 101 102 103 104

0 100 101 102 103 104

0 100 101 102 103 104

0 100 101 102 103 104

Fluorescence intensity

b

C T PM A PM A SB + S 2 B PM 02 202 A G +G FX F X

48 h

CD 10 c-mpl

Megakaryocytic specific genes

GATA-2 Glycophorin A γ - globin

Erythroid specific genes

CD36 CD44 Bcr-abl Actin

PMA inducible gene Unregulated genes

Figure 3 Induction of megakaryocytic markers and repression of erythroid markers during differentiation of K562 cells by PMA. K562 cells were cultured 48 h in the presence of 10 ng/ml PMA, 1 mM GF109203X (GFX), 5 mM SB202190 (SB), or the combination PMA þ SB or PMA þ GFX. (a) Facs analysis of different differentiation markers: undifferentiated or 48 h differentiated K562 cells were incubated with saturating amount of either anti-CD44-FITC, anti-CD10-PE, or anti-CD41-FITC monoclonal antibodies for 30 min at 41C. Fluorescence was measured by using the FL1/FL2 channels of a FACScan. (b) RT–PCR: RT–PCR analysis of genes modulated in cells treated with 10 ng/ml PMA, 1 mM GF109203X (GFX), 5 mM SB202190 (SB), or the combination PMA þ SB or PMA þ GFX. Amplification method was described in Materials and methods. Actin and Bcr-Abl were used as invariant controls in the experiment.

isoforms, (ii) required activation of Erk1/2 and JNK, and inhibition of p38 MAPK and (iii) is independent of Akt activation. This conclusion is also supported by the two following observations: (i) GFX, U0126 (not shown) and high concentrations of SB (20–40 mM) that induced drastic JNK activation and apoptosis in the presence of PMA blocked megakaryocytic differentiation and (ii) low concentrations of SB which increased Erk1/2 and JNK activity and c-Jun phosphorylation and inhibited p38 MAPK without significantly affecting cell death favor this process. Finally, we checked that the PI3K inhibitors LY294002 and wortmannin (not shown) failed to affect megakaryocytic differentiation ruling out the possible involvement of the PI3K/Akt pathway in this process.

c-DNA array analysis of the effects of PMA on megakaryocytic differentiation To assess the effect of PMA on gene expression, we used a c-DNA array approach. We focused our attention on a limited number of selected genes (847), mainly involved in apoptosis, differentiation, angiogenesis, transcription, cell cycle, and signal transduction, all cellular processes involved in the regulation of cell fate. In total, 55 genes were found to be modulated by PMA either alone or in the presence of SB. Among them, eight were downregulated and 47 were upregulated (Table 1). SB was found to increase the effect of PMA on the expression of most of these genes, either positively or negatively (not shown). Our c-DNA array approach was validated by the fact that, as expected, PMA increased Oncogene


786 CYTOPLASMIC MICROSOMAL FRACTION FRACTION PMA - + + - + + SB 202190 - - + + - - + + Mw (kDa)

a

PKC α / β

82

PKC γ

80

PKC ε

90

PKC δ

78

PKC ι

74

Pan-phospho PKC

80

P Erk 1/2

44 42

Conventional PKCs Novel PKCs Atypical PKC

b

CTSBGFX

PMA

PMA + SB PMA + GFX

Mw (kDa)

Time (h) 0 6 6 0.5 2 6 24 0.5 2 6 24 0.5 2 6 24 P Erk1/2

44 42

P p38

38 54

P JNK

P Jun

37

P AKT

60

HSP 60

60

c SB202190 Time (6h) 0 11 2 5 0 20 40 P JNK P Jun

46

54

d PMA SB202190 (5 µM) Uo126 (10 µM)

46

48 h

Mw (kDa)

37

P

-

+ -

+ + -

+ +

Mw (kDa) 44 42

Erk1/2 120 h

44 42

Figure 4 Signaling pathways controlling megakaryocytic differentiation of K562 cells. (a) K562 cells were cultured for 15 min in the presence of 10 ng/ml PMA, 5 mM SB202190 or the combination of both effectors (PMA þ SB). Cell lysates were subfractionated in cytoplasmic and microsomal fractions, and proteins contained in each fraction were separated by electrophoresis on 10% polyacrylamide gels. Proteins were then blotted on to PVDF membranes with either anti-PKC a/b, g, e, d, i, anti-pan phospho PKC, or antiphospho Erk1/2 antibodies. (b) K562 cells were incubated for the times indicated in the presence of 10 ng/ml PMA, 1 mM GF109203X (GFX), 5 mM SB202190 (SB), or the combination PMA þ SB or PMA þ GFX. Cells lysates were separated by electrophoresis on 10% polyacrylamide gels. Proteins were then blotted on to PVDF membranes with either antiphospho Erk1/2, antiphospho p38 MAPK, antiphospho JNK, antiphospho Jun, antiphospho AKT and HSP60 antibodies. (c) K562 cells cultured for 6 h in the presence of 10 ng/ml PMA and/or various amounts of SB202190 (1–40 mM). Cell lysates were separated by electrophoresis on 10% polyacrylamide gels. Proteins were then blotted on to PVDF membranes with either antiphospho JNK or antiphospho Jun. (d) Primary cells from a patient with chronic myeloid leukemia (Ph þ ) were treated for 48 or 120 h in the presence of 10 ng/ml PMA, 5 mM SB 202190, 10 mM Uo126, or the combination PMA þ SB or PMA þ Uo. After electrophoresis and transfer, proteins were then blotted on to PVDF membranes using an antiphospho Erk1/2.

the expression of megakaryocytic markers such as CD10, CD41 and, CD61 and decreased that of erythroid markers such as GPA, g-globin, and CD36 (Figures 3a and b, 5a and Table 1). The genes whose expression was increased or induced in this screening belong to different protein families including adhesion molecules (ICAM-1) and integrins such as a4, a5, aM, b1, b3, b5, b6, and b7, Oncogene

angiogenic factors (VEGF, VEGF-R1, and VEGF-R2, TIE-1, VRAP and angiopoietin), tyrosine kinase receptors (ephrin A4, B2, B6, AXL, and MPL-R), proteases and inhibitors (granzyme B, MMP-1 and 3, PAI, urokinase receptor, TIMP-1), chemokines and receptors (CCR7 and MCP-1) and transciption factors such as c-fos, stat-1, egr-1, RARe, notch 3 and 4, and Jun-B.


787 Table 1 List of PMA-modulated genes Genes

Untreated

PMA

Proteases, Protease receptors and inhibitors

MMP-1 Reelina Urokinase R TIMP-1a MMP-3 TIMP-3 PAI-1a

1 1 1 1 1 1 1

7.8 0.25 30 150 5 22 200

Angiogenic factors

VEGF R2 VRAP TIE-1a VEGF R1 VEGF

1 1 1 1

40 8 55 90 3

Transcription factors

RAR epsilon c-fosa EGR1 Notch-3 JUN B Notch-4

1 1 1 1 1 1

8 11 8 7 10 8

Signal transduction

PTX3 CAD-12 ED1 SOCS1 GAB1

1 1 1 1 1

0.4 100 112 5 0.15

Integrin family

Alpha 4 (CD49d) Alpha x (CD11c) Beta 3 (CD61) Beta 7 ICAM-1 Alpha 3 (CD49c) Beta 6 Aplha 2 (CD49b) Alpha 5 (CD49e) Alpha M (CD11b) Beta 1 (CD29)

1 1 1 1 1 1 1 1 1 1 1

7 8 17 10 12 10 6 3 8 99 4

Growth factors, cytokines and receptors

Activine Aa MCP-1 Oncostatine M Transferrine R Alox 5 IGF R1 CCR-7 SCGF IL-1 alpha MPL R AXLa PDGF-A chain PREF-1a

1 1 1 1 1 1 1 1 1 1 1 1 1

204 260 7 0.3 0.3 93 50 6 17 4 10 30 10

Cell cycle

Cycline D Cycline D2 Cycline A2

1 1 1

7 0.1 0.15

Ephrin family

EphB6 EphB2 EphA4

1 1 1

40 10 24

Apoptosis

Granzyme Ba Cytochrome c

1 1

100 0.25

The list contains genes that showed over three-fold induction or repression in three different experiments as compared to control conditions (ratio of one). The ratio of gene expression in cells treated for 24 h with 10 ng/ml PMA is shown. aGenes also indicated in Figure 5.

While most of the modulated genes were found to be upregulated by PMA, expression of eight genes, namely, cyt-c, pentraxin 3, reelin, 5-lipooxygenase, cyclin A2 and D2, GAB-1 and transferin receptor was decreased. Among these latter genes, Reelin and 5-lipooxygenase have been previously shown to be increased during erythroid differentiation of K562 cells. Finally, many of the regulated genes detected in our screening have not been reported before to be modulated by PMA or during megakaryocytic differentiation. This is particularly true for ephrin receptors and angiogenic factors other than VEGF receptors. The regulation of some of these genes by PMA was confirmed by RT–PCR analysis (Figure 5a). Finally, we also performed kinetic analysis of CD10, tie-1, glycophorin A, and actin mRNA expression upon PMA and PMA þ SB treatments. Interestingly, we confirmed that SB further increased the positive effect of PMA on CD10 and tie-1, and its negative effect on glycophorin A mRNA expression, respectively (Figures 3b and 5b). As expected, we found no difference in actin mRNA expression whatever the conditions used to stimulate K562 cells. RQ-PCR analysis of the expression of various transcription factors during megakaryocytic differentiation of K562 cells The differentiation of the K562 leukemia cell line is under the control of transcription factors that can direct these cells specifically towards the erythroid or the megakaryocytic lineage. To analyse more precisely the effects of PMA and PMA þ SB on megakaryocytic differentiation, we evaluated by RQ-PCR the level of expression of more than 80 transcription factors. PMA was found to modulate the expression of at least ten transcription factors. Among them, nine (LKLF, c-Fos, Fos-B, Rel-B, Egr-1, Egr-2, c-Jun, Jun-B, and Blimp-1) were upregulated. Interestingly, SB was found to potentiate PMA effect on the expression of at least six of them (LKLF, Rel-B, C-Fos, Egr-1 and 2, and c-Jun), suggesting a role for these genes in favoring megakaryocytic differentiation in K562 cells. The expression of the three latter genes (Fos-B, Blimp-1, and jun-B) was not further modulated by SB. The potentiating effect of SB was even more pronounced on the expression of early immediate genes such as Egr-1 and 2 at 6 h (not shown). The expression of some other genes present in the assay including Cyclin D2 and c-Myc (not shown) was decreased by PMA and PMA þ SB in agreement with the c-DNA array analysis. Finally, we found that IEX-1 expression, a gene recently reported to be important for megakaryocytic differentiation (Garcia et al., 2002), was induced by PMA, an effect clearly potentiated by SB. Altogether, our results linked the potentiating effect of SB on PMA-mediated megakaryocytic differentiation to the expression of several transcription factors such as LKLF, Rel-B, C-Fos, Egr-1 and 2, and c-Jun. Finally, none of the effectors significantly affected the expression of different housekeeping genes whatever the conditions used (Figure 6 for HPRT and not shown). Oncogene


788

a

b

24 h

CT PMA

CT PMA

CD 10

AXL

CD 61

PAI

GPA

Granzyme B

VEGF-R1

c-fos

TIE-1

Reelin

IGF-R1

Activin A

CT Time (h)

24 h

PMA 8

16

24

PMA + SB 48

72

8

16

24

48

SB 72

8

16

24

48

72

CD10 Glycophorin A Tie-1 Actin Figure 5 c-DNA array and RT–PCR analysis of genes modulated in response to PMA and PMA þ SB. (a) Validation of PMAmodulated genes by RT–PCR. RNA was prepared from cells left untreated or treated for 24 h with 10 ng/ml PMA. Amplification method was described in the Materials and methods section. Actin was used as an invariant control in the experiment. (b) RT–PCR analysis CD10, GPA, and Tie-1 mRNA. RNA was prepared from cells left untreated or treated for the indicated times with 10 ng/ml PMA, 5 mM SB202190, or the combination of both effectors (PMA þ SB). Amplification method was described in the Materials and methods section. Actin was used as an invariant control in the experiment.

Discussion While it has been extensively reported that human CML K562 cells can be induced to differentiate towards the megakaryocytic lineage by PMA, the molecular mechanisms underlying such differentiation induction remain not completely deciphered. In the present study using both pharmacological, biochemical and c-DNA array analysis approaches, we explored in detail the pathways involved in PMA-mediated differentiation of K562 cells towards the megakaryocytic lineage. It has been shown previously that PMA-induced megakaryocytic differentiation of K562 cells is dependent on the activity of the MEK/MAPK pathway (Melemed et al., 1997; Whalen et al., 1997; Herrera et al., 1998). The data presented here confirm an important role of the MEK/Erk1/2 pathway in PMAinduced differentiation of K562 and also establish that the concomitant inhibition of the p38 MAPK pathway by PMA and a moderate activation of JNK also represent important determinants for complete (i.e. quantitative and qualitative) megakaryocytic differentiation of this cell line. This observation is supported by the finding that the p38 MAPK inhibitors SB202190 and SB203580 (not shown) considerably accelerated the kinetic and the extent of PMA-induced K562 cell differentiation allowing nearly 100% of cells to acquire a more pronounced megakaryocytic phenotype (Figure 1). This potentiating effect of the p38 MAPK Oncogene

inhibitors seems to be due to both p38 MAPK inhibition and a significant activation of JNK and increased phosphorylation of c-Jun by these inhibitors (Figure 4). Surprisingly, higher doses of SB202190, which drastically increased JNK activity, were nevertheless found to inhibit megakaryocytic differentiation of K562 and to induce cell death more especially in the presence of PMA. The effect of the combination of PMA and high doses of SB may appear somewhat surprising since PMA has been consistently shown to protect various cell lines against apoptosis (Bertolotto et al., 2000; Villalba et al., 2001; Herrant et al., 2002). However, we have previously established that inhibition of the NFkB pathway, for example, can convert PMA from a death-protecting to a death-inducing function (Busuttil et al., 2002). The drastic JNK activation achieved in the presence of high doses of SB may thus explain the considerable increase in caspase activation and apoptosis observed in the presence of PMA þ SB. p38 MAPK activation is thought to be a prerequisite for erythroid differentiation of K562 cells (Park et al., 2001). As megakaryocytic and erythroid differentiation are mutually exclusive, it is noteworthy that PMA-mediated differentiation of K562 cells was accompanied by an inhibition of p38 MAPK activation and that p38 MAPK inhibitors can potentiate megakaryocytic differentiation of K562 cells. In line with our results, it has been recently reported that inhibition of p38 MAPK potentiates the JNK/SAPK pathway and AP-1 activity


789 LKLF

FOS B 59 165

247

372

40 000 30 000 20 000 349

364

300 000 250 000 200 000 150 000

PMA

SB

PMA SB

42 347

39 170

100 000 50 000

0

CT

271 388

350 000

50 000

10 000

400 000

Arbitrary unit

4 877

358 098

49 491

60 000

Arbitrary unit

Arbitrary unit

9 570

10 000 9 000 8 000 7 000 6 000 5 000 4 000 3 000 2 000 1 000 0

Cyclin D2

0

CT

PMA

IEX1

SB

PMA SB

CT

PMA

Rel B

SB

PMA SB

c-Fos 48 472 96 949

1 000 000

Arbitrary unit

Arbitrary unit

1 200 000

800 000 600 000

299 561

400 000 60 513

39 102

200 000 0

CT

PMA

SB

100 000 90 000 80 000 70 000 60 000 50 000 40 000 30 000 20 000 10 000 0

PMA SB

Arbitrary unit

1 044944

35 423

2 599

2 477

CT

PMA

EGR1

SB

50 000 45 000 40 000 35 000 30 000 25 000 20 000 15 000 10 000 5 000 0

PMA SB

31 925

1 354

1 107

CT

PMA

Blimp-1

SB

PMA SB

c-Jun

8 435

19 114

Arbitrary unit

20 000 15 000 10 000 1 402

658

5 000

700 000

4 000 3 000 2 000 1 000

442

207

CT

PMA

SB

PMA SB

PMA

SB

Arbitrary unit

Arbitrary unit

20 000 15 000 10 000 8

0

SB

200 000

56 070

18 884

CT

PMA

PMA SB

2 000 000 1 800 000 1 600 000 1 400 000 1 200 000 1 000 000 800 000 600 000 400 000 200 000 0

PMA SB

HPRT 728 713 683 459

800 000 1 570 179

112 143

SB

786 451

1 857 585

25 000

PMA

300 000

Jun B

22 889

CT

400 000

PMA SB

29 630

147

420 718

500 000

0

CT

EGR2

30 000

600 000

100 000

0

0

5 000

800 000

700 000

71 466

Arbitrary unit

Arbitrary unit

25 000

723 679 7 302

9 000 8 000 7 000 6 000 5 000

Arbitrary unit

26 154

30 000

504 675

600 000 500 000 400 000 300 000 200 000 100 000 0

CT

PMA

SB

PMA SB

CT

PMA

SB

PMA SB

Figure 6 Real-time quantitative PCR. RNA was prepared from cells left untreated or treated for 24 h with 10 ng/ml PMA, 5 mM SB202190 or the combination of both effectors (PMA þ SB). The amplification method was described in the Materials and methods section.

in monocytic differentiation of HL60 cells (Wang and Studzinski, 2001). The same authors previously demonstrated that 1,25 dihydroxyvitamin D3 was able to block Ara-C-induced erythroid differentiation (Moore et al., 1991). Therefore, it would be of interest to determine whether vitamin D3 does increase megakaryocyte differentiation of K562 cells in the presence of PMA

and whether inhibition of p38 MAPK is involved in this process. A general role for PKCs in megakaryocytic differentiation induced by PMA has previously been established (Hong et al., 1996). However, the specific PKC isotypes involved in this process, that is, conventional PKCs (cPKC) or new PKCs (nPKC), have not been Oncogene


790

a

Uninduced K562 cells

Erk

JNK

PMA-induced K562 cells

p38

Cell proliferation Low Hb synthesis Undifferentiated status

Erk

JNK

PMA+SB202190-induced K562 cells

p38

Cell proliferation Megakaryocyte differentiation Very low Hb synthesis Erythroid differentiation

Erk

JNK

p38

Cell proliferation Megakaryocyte differentiation Very low Hb synthesis Erythroid differentiation

PMA

b

GF109203X PKCs

Erk

JNK

PRAK

p38

SB202190

MAPKAP

Figure 7 (a) A model of PMA-mediated pathways in induction of K562 cell differentiation. (b) A possible model for SB202190mediated increase in megakaryocytic differentiation of K562 cells.

characterized yet. Using a pharmacological approach, we demonstrated that new but not conventional PKCs are involved in this process since (i) GFX (an inhibitor of both cPKC and nPKC) abrogated PMA-induced differentiation of K562 cells as shown by both morphological assay and analysis of specific megakaryocytic markers, while (ii) Go6973 (an inhibitor of cPKC) had no effect (Figure 7). To address the role of specific PKC isotypes in the differentiation process, we generated stable and inducible K562 cells (K562-tet-on) expressing dominant negative or constitutively active PKC a, b, e, and y. None of these isoforms alone were able to induce megakaryocytic differentiation of K562 cells or to potentiate PMA-induced differentiation (not shown). Thus, PMA-induced differentiation of K562 cells might depend on other PKC isotypes (not tested in the present study such as PKCg or PKCd) or more likely on a combination of two or more of these isotypes. Nevertheless, our results demonstrated that new PKCs rather than conventional PKCs mediated Oncogene

megakaryocytic differentiation of K562 cells. Megakaryocytic differentiation of K562 cells by PMA was only partly dependent on ERK/MAPK activity since it was significantly but not fully blocked by Uo126 and PD98059, two inhibitors of MEK1 (not shown). Accordingly, Shelly et al. (1998) reported that PMA triggered phenotypic changes in K562 cells by both MAPK-dependent and -independent events. In agreement with the involvement of other signaling pathways, we established that complete differentiation (i.e. quantitative and qualitative) of K562 cells towards the megakaryocytic lineage by PMA requires not only an inhibition of the p38 MAPK pathway but also a threshold of JNK and c-Jun activation (Figure 6a). This is particularly evident from the experiments with the p38 MAPK inhibitors, which favor megakaryocytic differentiation in the low concentration ranges. It has been recently reported that the p38 MAPK pathway may provide negative feedback for Ras signaling (Chen et al., 2000). More specifically, it was shown that the


791

p38 MAPK pathway inhibited Ras activity by blocking activation of JNK, without however a direct effect upon Erk1/2. This effect was ultimately mediated by the p38 MAPK substrate, p38-related/activated protein kinase (PRAK). Such a negative feedback mechanism could provide a good explanation for the increase in differentiation observed in our study in the presence of SB, since blockade of p38 MAPK would remove the inhibitory effect on JNK, thus favoring megakaryocytic differentiation of K562 cells in the presence of PMA (Figure 7b). PMA-induced differentiation of K562 cells is intimately associated with the expression of several genes (Alitalo, 1990; Cheng et al., 1994; Namciu et al., 1994; Belhacene et al., 1998; Jacquel et al., 2003). To date, very few studies have analysed at a large scale (c-DNA array or DNA biochips) gene expression in K562 cells undergoing differentiation towards the megakaryocytic or erythroid lineage (Jacquel et al., 2003; Pettiford and Herbst, 2003; Huang et al., 2004). It is well established that in vitro differentiation of K562 cells is characterized by morphological changes such as an increase in cell size and adhesion properties of the differentiating cells. In agreement with the latter finding, we detected by microarray analysis a drastic increase in several integrin subunits (Table 1). Enhanced expression of CD41 and CD61 at the cell surface is commonly used as a hallmark of megakaryocytic differentiation, but besides these two markers, we identified several other integrins whose expression is induced by PMA and that have not yet been associated with this process. It seems likely that enhanced expression of these integrins is intimately linked to the increased adhesion observed in PMAtreated K562 cell cultures (not shown). Megakaryocytic differentiation of K562 cells also requires the activation of transcription factors that control the expression of specific genes in this lineage. In agreement with this statement, we found that PMA induced expression of different transcription factors known to be involved in megakaryocytic differentiation or associated with PMA stimulation such as c-Fos, B-Fos, Egr-1 and 2, Stat-1, Gata-2, c-jun, jun B, and Rel B (Table 1, Figures 5 and 6). Interestingly, PMA also represses the expression of erythroid-specific transcription factors including Gfi-1B (Jacquel et al., 2003). This profile of transcription factor and integrin expression is typical of megakaryocytic differentiation and thereby confirms the validity of our c-DNA array approach. Interestingly, we also detected in this screening the modulation of two other gene families that had never been described before to be induced or involved in megakaryocytic differentiation. First, we identified an increased expression of several angiogenic factors upon PMA treatment of K562 cells such as VEGF, VEGFR-1 and VEGFR-2, VRAP, TIE-1, and angiopoietin. Consistently, we have recently shown that K562 and CML cells from patients secrete a high amount of VEGF (Legros et al., 2004). It is not known at present whether this specific profile of angiogenic gene expression is involved in the differentiation process or is rather a consequence of it. However, recent data from the literature have pointed out a role of

VEGF as an autocrine factor for CML cell proliferation (Ruan et al., 2004). Neoangiogenesis has also been shown to occur in the course of normal megakaryocytic differentiation and in this context increased expression of angiogenic factors in PMA-treated K562 cells makes sense. It is noteworthy that differentiation of K562 cells towards the megakaryocytic lineage was also accompanied by induction or increased expression of Ephrin receptors. The Eph family of receptors constitutes the largest known family of tyrosine kinase receptors comprising not less than 14 members (Kullander and Klein, 2002; Poliakov et al., 2004). Expression of Eph receptors is highly tissue specific and modulated in various pathological situations including cancer (Brantley-Sieders et al., 2004; Surawska et al., 2004). To the best of our knowledge, a role of Eph receptors in leukemia cell differentiation has not yet been established. In a set of preliminary experiments, we have not been able to demonstrate a role of different recombinant Ephrins on megakaryocytic differentiation, but further studies are needed to confirm the potential role of Eph receptors in the differentiation program of K562 cells. Finally, the increase in AXL mRNA expression, which has also been confirmed at the protein level (not shown) during megakaryocytic differentiation of K562 cells is particularly interesting since members of the UFO/AXL receptor tyrosine kinase family are altered in a great variety of malignant phenotypes and Axl is widely associated with chronic myeloproliferative disorders and chronic myeloid leukemia (Neubauer et al., 1994). Further studies are thus needed to assess more precisely the role of AXL and its ligand Gas 6 in megakaryocytic differentiation. In conclusion, in the present study we have determined at the molecular level the requirements for PMA-induced differentiation of the K562 cell line towards the megakaryocytic lineage and evidenced by c-DNA microarray a set of new genes linked to this process. We showed that full differentiation of K562 cells by PMA towards the megakaryocytic phenotype required new PKC-dependent activation of Erk1/2 and JNK and inhibition of p38 MAPK. Next, we identified at least two gene families that had never been described before to be modulated or involved in this process. Notably, the observation of the induction of angiogenic factors, Ephrin, and AXL/UFO receptor genes during the course of megakaryocytic differentiation is of great interest since these two families of tyrosine kinase receptors seem intimately linked to differentiation processes and are also associated with leukomogenesis. In the near future, the involvement of these kinase families should be more globally addressed by selective silencing of the corresponding mRNA using RNA interference. In this line, stable K562 clones expressing inducible shRNA under the control of the tet promoter, which have been successively established in our laboratory for Bcr-Abl and Src kinases (not shown), should also be generated for the silencing of Eph and AXL/UFO receptor family members. Oncogene


792 Materials and methods Reagents and antibodies RPMI and fetal calf serum (FCS) were purchased from GIBCO BRL (Paisley, UK). Sodium fluoride, sodium orthovanadate, phenylmethylsulfonyl fluoride (PMSF), aprotinin, leupeptin, and Phorbol 12-myristate 13-acetate (PMA) were purchased from Sigma (Saint-Louis, MO, USA). Bisindolylmaleimide-l (GF109203X), Go¨6976, Uo126, SB203190, SB202190, LY294002, and wortmannin were from Calbiochem (La Jolla, CA, USA). Anti-PKC a/b, g, e, d, and i were purchased from Transduction Laboratory (Lexington, KY, USA). Anti-pan phospho PKC, antiphospho Erk1/2, antiphospho p38 MAPK, antiphospho JNK, antiphospho c-Jun, antiphospho Akt, and peroxydase-conjugated anti-rabbit antibody were from Cell Signaling Technology (Beverly, MA, USA). Peroxidase-conjugated anti-mouse was from Dakopatts (Glostrup, Denmark). Cell lines and human primary cells The human cell line K562 and MEG-01 has been described elsewhere (Belhacene et al., 1998) and was cultured in RPMI 1640-glutamax supplemented with 5% FCS, 50 U/ml penicillin, 50 mg/ml streptomycin, and 1 mM pyruvate under 5% CO2 in a humidified incubator. CML samples were obtained after approval by the institutional review board and appropriate informed consent from the patients. CML cells were cultured in RPMI 1640-glutamax supplemented with 10% FCS, 50 U/ml penicillin, 50 mg/ml streptomycin, and 1 mM pyruvate under 5% CO2 in a humidified incubator. Phase contrast microscopy K562 cells were treated for various times with PMA and different signaling inhibitors or the combination of both effectors. Morphologic changes characteristic of megakaryocytic differentiation were visualized using standard optics (Zeiss, Oberkochen, Germany). In some experiments, quantification of cells with megakaryocytic features was realized. To avoid potential errors in statistical sampling, fields were randomly selected and counts were performed by three different investigators in a blind manner. Cytospin, May–Grunwald–Giemsa staining Approximatily 5 104 cells were spun onto a microscope slide for 4 min at 800 g under medium acceleration in a cytospin 3. After air drying, slides were stained with May–Grunwald– Giemsa stain (Sigma) according to the manufacturer’s instruction. Electronic microscopy of control and differentiated cells Cells were fixed with 2% formaldehyde, in 0.1 M Na cacodylate, pH 7.4 for 1 h at 41C. Pellets were rinsed in cacodylate buffer, postfixed in 1% OsO4 for 1 h, dehydrated through graded alcohols, and embedded in epoxy resin. Oriented 1 mm sections were obtained with diamond knives and multiple areas were thin sectioned, mounted on copper mesh grids, and stained with uranyl acetate and lead citrate. Ultrathin sections were examined on a Jeol 1200 XII electron microscope. Flow cytometry To determine apoptosis, cells were treated and stained using the Antiactive-caspase-3-PE MAb Apoptosis Kit (BD Biosciences; San Diego, CA, USA). To determine megakaryocytic differentiation, cells were incubated to 41C for 30 min in 100 ml of Oncogene

PBS 0.1% bovine serum albumin (BSA) with monoclonal antibodies anti-CD44-FITC, and/or anti-CD41-FITC, and/or anti-CD10-PE (BD Biosciences; San Diego, CA, USA). After three washes with PBS, cells were fixed in 1% formaldehyde for 30 min at 41C. Fluorescence was measured by using the FL1 channel of a FACScan (Becton Dickinson; Cowley, UK). RT–PCR Analysis Isolation of RNA and RT–PCR has been described in detail elsewhere (Herrant et al., 2002). To confirm differential mRNA expression, a panel of genes was quantified by RT– PCR using the following primers. CD10-s:

GGTCATAGGACACGAAATCA/ as: AGATCACCAAACCCGGCACT c-mpl-s: GCACTGTGATGCTTTATGCAAC/ as: TGAACGGTTTAGAGGATGAGGA GATA-2-s: CGTCTTCTATCACCTCG/ as: CGTCTTGGAGAAGGGC Glycophorin A-s: GGAATTCCAGCTCATGATCTCAGG ATG Glycophorin A-as: TCCACATTTGGTTTGGTGAACAGA TTC g-Globin-s: ACTCGCTTCTGGAACGTCTGA/ as: GTATCTGGAGGACAGGGCACT CD36-s: CTGGCTGTGTTTGGAGGTATTCT/ as: AGCGTCCTGGGTTACATTTTCC CD44-s: GATCCACCCCAATTCCATCTGTGC CD44-as: AACCGCGAGAATCAAAGCCAAGG CC Bcr-Abl-s: GAAGAAGTGTTTCAGAAGCTTCTCC Bcr-Abl-as: GACCCGGAGCTTTTCACCTTTAGTT Actin-s: ACCCACACTGTGCCCATCTTA/ as: CTAGAAGCATTTGCGGTGGA CD61-s: TATAGCATTGGACGGAAGGC/ as: GACCTCATTGTTGAGGCAGG Activin A-s: GGCTTGGAGTGTGATGGC/ as: GCAGCCACACTCCTCCACAAT Granzyme B-s: GGGGAAGCTCCATAAATGTCACCT Granzyme B-as : TACACACAAGAGGGCCTCCAGAGT c-Fos-s : AAGGAGAATCCGAAGGGAAAGGA ATAAGATGGCT c-Fos-as: AGACGAAGGAAGACGTGTAAGCA GTGCAGCT TIE-1-s: GCAGACAGACGTGATCTGGA/ as: CACCTCCATGTAGGCAACCT Reelin-s: GCCACAATGGAACAGGTCAT/ as: CAATACTGCCACTGTAACTG IGF-R1-s: TCCAACAGCTGGAACATGG/ as: ATTCAGCCTCCTCCTTCTCG VEGF R1-s: GTAGCTGGCAAGCGGTCTTACCGG CTC VEGF R1-as: GGATTTGTCTGCTGCCCAGTGGGT AGAGA AXL-s: GGTGGCTGTGAAGACGATGA/ as: CTCAGATACTCCATGCCACT PAI-1-s: GCTGAATTCCTGGAGCTCAG/ as: CTGCGCCACCTGCTGAAACA Western blot assays Cells were incubated with different effectors for the times indicated in the figure legends, and then lysed in buffer B containing 50 mM HEPES (pH 7.4), 150 mM NaCl, 20 mM EDTA, 100 mM NaF, 10 mM Na3VO4, 1 mM PMSF, 1 mM leupeptin, 20 mg/ml aprotinin, and 1% Triton X-100. A total of 100 mg of protein were separated on 10–12% polyacrylamide


793 gel and transferred onto PVDF membrane (Immobilon-P, Millipore; Bedford, MA, USA). After blocking nonspecific binding sites, the membranes were incubated with specific antibodies. The membranes were washed three times with TNA-1% NP-40 (Tris 50 mM, NaCl 150 mM, pH 7.5) and incubated further with horseradish peroxidase conjugated antibody for 60 min at room temperature. Immunoblots were revealed by autoradiography using the enhanced chemiluminescence detection kit (Amersham Biosciences, Uppsala, Sweden).

Probe synthesis was performed as previously described (Herrant et al., 2002). Probes were hybridized to each array membrane for approximately 15 h in the Hybridization Solution at 651C. The array membranes were washed in decreasing concentrations of SSC. The damp arrays were exposed to a phosphor-imaging screen for 3–5 days. Exposed screens were scanned using a phosphor-imager STORM 840 (Molecular Dynamics). The images were analysed by the Image Quant 5.0 software essentially as previously described.

Subcellular fractionnements After stimulation, K562 cells (4 106) were resuspended in hypotonic buffer containing 100 mM Tris-HCl pH 7.4, 5 mM EGTA, 1 mM PMSF, 1 mM leupeptin, and 20 mg/ml aprotinin, and rapidly sonicated at 41C. Cell extracts were first centrifuged at 2000 g for 10 min. Supernatants were then centrifuged at 100 000 g for 45 min. Soluble proteins (cytosol) were collected. Pellets (microsomal fraction) were lysed in buffer B. In each lane the amount of total, cytosolic, and microsomal proteins loaded on the gel corresponds to 4 106 cells.

Real-time quantitative PCR Total RNA from cells was isolated from cells incubated for 24 h with the different effectors using the TRIzol reagent following the instructions of the manufacturer (Invitrogen). After treatment with DNaseI, 1 mg of total RNA was reverse transcribed using random priming and Superscript II reverse transcriptase (Invitrogen). Real-time PCR was performed in an ABI PRISM 5700 Sequence Detector System (Applied Biosystems) using the SYBR Green detection protocol as outlined by the manufacturer. Gene-specific primers were designed using the Primer Express software (Applied Biosystems). Relative expression level of target genes was normalized for RNA concentrations with four different house-keeping genes (GAPDH, b-actin, HPRT and ubiquitin). mRNA values are expressed as arbitrary units and represent the mean7s.d. of duplicates and are representative of two independent experiments.

Expression array R&D Systems human Apoptosis Expression Array represents a comprehensive collection of genes implicated in apoptosis, cell cycle regulation, adhesion, differentiation, and signal transduction. The array consists of 847 different cloned cDNAs, printed as PCR products, on a positively charged nylon membrane. Total RNA was isolated using TriPure reagent (Boehringer Mannheim, Germany) from cells incubated for 24 h with the different effectors. PMA was used at 10 ng/ml. After purification, the total RNA pellets were resuspended in water and quantified using OD260 spectrophotometry (Perkin-Elmer MBA2000). Ribosomal RNA integrity was verified on a 1% agarose gel electrophoresis.

Acknowledgements This work was supported by the Institut National de la Sante´ et de la Recherche Me´dicale, the Ligue Nationale contre le Cancer (Equipe labelliseé), and the Canceropole PACA. SM is a recipient from the LNC.

References Alitalo R. (1990). Leuk Res 14: 501–514. Belhacene N, Maulon L, Guerin S, Ricci JE, Mari B, Colin Y et al. (1998). FASEB J 12: 531–539. Bertolotto C, Maulon L, Filippa N, Baier G, Auberger P. (2000). J Biol Chem 275: 37246–37250. Brantley-Sieders D, Parker M, Chen J. (2004). Curr Pharm Des 10: 3431–3442. Busuttil V, Bottero V, Frelin C, Imbert V, Ricci JE, Auberger P et al. (2002). Oncogene 21: 3213–3224. Chen G, Hitomi M, Han J, Stacey DW. (2000). J Biol Chem 275: 38973–38980. Cheng T, Wang Y, Dai W. (1994). J Biol Chem 269: 30848–30853. Cramer EM, Norol F, Guichard J, Breton-Gorius J, Vainchenker W, Masse JM et al. (1997). Blood 89: 2336–2346. Dorsey JF, Cunnick JM, Mane SM, Wu J. (2002). Blood 99: 1388–1397. Garcia J, Ye Y, Arranz V, Letourneux C, Pezeron G, Porteu F. (2002). EMBO J 21: 5151–5163. Goldfarb AN, Delehanty LL, Wang D, Racke FK, Hussaini IM. (2001). J Biol Chem 276: 29526–29530. Herrant M, Luciano F, Loubat A, Auberger P. (2002). Oncogene 21: 4957–4968. Herrera R, Hubbell S, Decker S, Petruzzelli L. (1998). Exp Cell Res 238: 407–414. Hong Y, Martin JF, Vainchenker W, Erusalimsky JD. (1996). Blood 87: 123–131.

Huang M, Wang Y, Collins M, Graves LM. (2004). Leukemia 18: 1857–1863. Jacquel A, Herrant M, Legros L, Belhacene N, Luciano F, Pages G et al. (2003). FASEB J 17: 2160–2162. Kawano T, Horiguchi-Yamada J, Iwase S, Furukawa Y, Kano Y, Yamada H. (2004a). Mol Cell Biochem 258: 25–33. Kawano T, Horiguchi-Yamada J, Saito S, Iwase S, Furukawa Y, Kano Y et al. (2004b). Leuk Res 28: 623–629. Kim KW, Kim SH, Lee EY, Kim ND, Kang HS, Kim HD et al. (2001). J Biol Chem 276: 13186–13191. Kullander K, Klein R. (2002). Nat Rev Mol Cell Biol 3: 475–486. Lam LT, Ronchini C, Norton J, Capobianco AJ, Bresnick EH. (2000). J Biol Chem 275: 19676–19684. Legros L, Bourcier C, Jacquel A, Mahon FX, Cassuto JP, Auberger P et al. (2004). Blood 104: 495–501. Long MW, Heffner CH, Williams JL, Peters C, Prochownik EV. (1990). J Clin Invest 85: 1072–1084. Matsumura I, Kawasaki A, Tanaka H, Sonoyama J, Ezoe S, Minegishi N et al. (2000). Blood 96: 2440–2450. Melemed AS, Ryder JW, Vik TA. (1997). Blood 90: 3462–3470. Moore DC, Carter DL, Bhandal AK, Studzinski GP. (1991). Blood 77: 1452–1461. Murray NR, Baumgardner GP, Burns DJ, Fields AP. (1993). J Biol Chem 268: 15847–15853. Nalefski EA, Newton AC. (2001). Biochemistry 40: 13216– 13229. Oncogene


794 Namciu S, Lieberman MA, Stavnezer E. (1994). Oncogene 9: 1407–1416. Neubauer A, Fiebeler A, Graham DK, O’Bryan JP, Schmidt CA, Barckow P et al. (1994). Blood 84: 1931–1941. Park JI, Choi HS, Jeong JS, Han JY, Kim IH. (2001). Cell Growth Differ 12: 481–486. Parker PJ, Murray-Rust J. (2004). J Cell Sci 117: 131–132. Pettiford SM, Herbst R. (2003). Leukemia 17: 366–378. Poliakov A, Cotrina M, Wilkinson DG. (2004). Dev Cell 7: 465–480. Racke FK, Lewandowska K, Goueli S, Goldfarb AN. (1997). J Biol Chem 272: 23366–23370. Racke FK, Wang D, Zaidi Z, Kelley J, Visvader J, Soh JW et al. (2001). J Biol Chem 276: 522–528. Ron D, Kazanietz MG. (1999). FASEB J 13: 1658–1676. Rouyez MC, Boucheron C, Gisselbrecht S, Dusanter-Fourt I, Porteu F. (1997). Mol Cell Biol 17: 4991–5000.

Oncogene

Ruan GR, Liu YR, Chen SS, Fu JY, Chang Y, Qin YZ et al. (2004). Leuk Res 28: 763–769. Schaefer M, Albrecht N, Hofmann T, Gudermann T, Schultz G. (2001). FASEB J 15: 1634–1636. Shelly C, Petruzzelli L, Herrera R. (1998). Leukemia 12: 1951–1961. Surawska H, Ma PC, Salgia R. (2004). Cytokine Growth Factor Rev 15: 419–433. Tanimura A, Nezu A, Morita T, Hashimoto N, Tojyo Y. (2002). J Biol Chem 277: 29054–29062. Uddin S, Ah-Kang J, Ulaszek J, Mahmud D, Wickrema A. (2004). Proc Natl Acad Sci USA 101: 147–152. Villalba M, Bushway P, Altman A. (2001). J Immunol 166: 5955–5963. Wang X, Studzinski GP. (2001). J Cell Biochem 82: 68–77. Whalen AM, Galasinski SC, Shapiro PS, Nahreini TS, Ahn NG. (1997). Mol Cell Biol 17: 1947–1958.