Apoptosis and erythroid differentiation triggered by Bcr-Abl ... - Nature

Oncogene (2007) 26, 2445–2458

& 2007 Nature Publishing Group All rights reserved 0950-9232/07 $30.00 www.nature.com/onc

ORIGINAL ARTICLE

Apoptosis and erythroid differentiation triggered by Bcr-Abl inhibitors in CML cell lines are fully distinguishable processes that exhibit different sensitivity to caspase inhibition A Jacquel1,2, P Colosetti1,2, S Grosso1,2, N Belhacene1,2, A Puissant1,2, S Marchetti1,2, J-P Breittmayer2,3 and P Auberger1,2 1

INSERM, U526, Cell Death Differentiation and Cancer Team, Equipe labelliseé par la Ligue Nationale contre le Cancer, Nice, France; 2Universite´ de Nice Sophia-Antipolis, Nice, France and 3INSERM, U576, CHU de Nice, Hoˆpital de l’Archet, Nice, France

Imatinib targets the Bcr-Abl oncogene that causes chronic myelogenous leukemia (CML) in humans. Recently, we demonstrated that besides triggering apoptosis in K562 cells, imatinib also mediated their erythroid differentiation. Although both events appear to proceed concomitantly, it is not known at present whether or not imatinib-induced apoptosis and differentiation are interdependent processes. Hence, we investigated the requirements for BcrAbl inhibitor-mediated apoptosis and erythroid differentiation in several established and engineered CML cell lines. Imatinib triggered apoptosis and erythroid differentiation of different CML cell lines, but only apoptosis exhibited sensitivity to ZVAD-fmk inhibition. Conversely, the p38 mitogen-activated protein (MAP) kinase inhibitor, SB202190, significantly slowed down erythroid differentiation without affecting caspase activation. Furthermore, imatinib and PD166326, another Bcr-Abl inhibitory molecule, triggered erythroid differentiation of K562 cell clones, nevertheless resistant to Bcr-Abl inhibitor-induced apoptosis. Finally, short hairpin RNA inhibitor (shRNAi) silencing of caspase 3 efficiently inhibited caspase activity but had no effect on erythroid differentiation, whereas silencing of Bcr-Abl mimicked imatinib or PD166326 treatment, leading to increased apoptosis and erythroid differentiation of K562 cells. Taken together, our findings not only demonstrate that Bcr-Abl inhibitor-mediated apoptosis and differentiation are fully distinguishable events, but also that caspases are dispensable for erythroid differentiation of established CML cell lines. Oncogene (2007) 26, 2445–2458. doi:10.1038/sj.onc.1210034; published online 9 October 2006 Keywords: imatinib; PD166326; apoptosis; erythroid differentiation; caspases; CML

Correspondence: Dr P Auberger, INSERM, U526, Cell Death Differentiation and Cancer Team, Equipe labelliseé par la Ligue Nationale contre le Cancer, 28 Avenue de Valombrose, 06107, Nice Cedex 02, France. E-mail: [email protected] Received 3 February 2006; revised 27 April 2006; accepted 21 June 2006; published online 9 October 2006

Introduction Chronic myelogenous leukemia (CML) is caused by the Philadelphia chromosome translocation t(9;22) (q34;q11), which fuses the bcr and the c-abl genes (Wong and Witte, 2004). The resulting p210 Bcr-Abl protein is a constitutively active tyrosine kinase, phosphorylating a variety of substrates that activate multiple signaling pathways, including among others phosphatidylinositol 3-kinase (PI3K), Janus kinase (JAK)/signal transducers and activator of transcription (STAT) and the Ras/MAP kinase/ERK kinase (MEK)/ extracellular signal-regulated protein kinase (Erk)1/2 pathways, thus conferring growth factor-independent proliferation and survival of myeloid progenitor cells (Deininger et al., 2000; Steelman et al., 2004). Imatinib mesylate (Gleevec, STI571) has emerged as the lead compound to treat the chronic phase of CML (O’Brien et al., 2003). It is well established that the drug targets the ATP-binding site of Bcr-Abl, thereby inhibiting its tyrosine kinase activity (Roskoski, 2003). Other molecules that bind to Bcr-Abl such as PD166326 also act as efficient inhibitors of the oncogenic tyrosine kinase (Wolff et al., 2005). Following drug treatment, the Bcr-Abl protein is rapidly dephosphorylated and inactivated, leading to the blocking of the above-mentioned signaling cascades. Subsequently, the cell cycle progression is impaired and cells undergo an apoptotic programme characterized by activation of caspases 9, 8 and 3 (Jacquel et al., 2003). Furthermore, it has been recently shown that inhibition of caspases in imatinib-treated K562 cells shifted the cells from a caspase-dependent to a necrosis-like cell death process, suggesting that imatinib also triggered caspase-independent cell death (CID) (Okada et al., 2004). Increasing evidence also indicates that besides their recognized role in the initiation and completion of apoptosis, caspases also play a regulatory function in hematopoietic cell proliferation and differentiation. In this line, erythroid differentiation in the bone marrow has been shown to involve caspase-dependent processes (Zermati et al., 2001; Kolbus et al., 2002; Carlile et al., 2004).

Bcr-Abl, apoptosis and erythroid differentiation A Jacquel et al

2446

Results Bcr-Abl inhibitors induced caspase activation and cell death in K562 cells Imatinib and PD166326 induced a very similar profile of caspase activation and inhibition of cell viability after a 48 h treatment of the K562 CML cell line. However, PD166326 was found to be at least 300 times more effective than imatinib (Figure 1a) with a half-maximal modulation for both effects in the nanomolar range in accordance with previous results from the literature (Huron et al., 2003; Wolff et al., 2005). As imatinibmediated cell death is thought to be a caspase-dependent event, we analysed the effect of ZVAD-fmk on imatinib and PD166326-mediated caspase activation. As illustrated on Figure 1b, ZVAD-fmk was found to abolish Bcr-Abl inhibitor-induced caspase activation. Externalization of phosphatidylserine was next followed by annexin V binding. The number of annexin V-positive cells reached from 10% in control cells to 60% in imatinib or PD166326-treated cells, and ZVAD-fmk decreased the number of annexin V cells to approximatively 30% in the presence of the Bcr-Abl inhibitors (Figure 1c). Oncogene

XTT metabolism Caspase activity 100

% response

80 60 40 20 0

0 10-4

10-3

10-2

10-1

PD166326

b

Caspase 3 activity (A.U.)

100

101 Log (µM)

Imatinib

2.0 1.5 1.0 0.5 0 0 0.1 0.3 1 Ima (µM)

1

3 10 0 0.1 0.3 1 PD (nM) Ima (µM)

1

3 10 PD (nM)

ZVAD-fmk

c

Annexin V positive cells (%)

The human K562 cell line is widely used as a model for leukemia differentiation (Belhacene et al., 1998; Racke et al., 2001; Jacquel et al., 2006). Interestingly, it expresses the p210 Bcr-Abl protein and can also be induced to differentiate towards the megakaryocytic or erythroid lineages depending on the stimuli and, therefore, represents an appropriate cellular model to study the relationships between apoptotic and differentiation program in leukemia (Dorsey et al., 2002; Jacquel et al., 2003; Pettiford and Herbst, 2003; Kawano et al., 2004). We and others have recently shown that besides inducing apoptosis in K562 cells, imatinib also triggered erythroid differentiation of this cell line increasing its hemoglobin content and the expression of glycophorin A (GPA), and other specific erythroid markers at the cell surface (Kohmura et al., 2004; Kuzelova et al., 2005). However, despite extensive studies, the types of cell death used by imatinib and other Bcr-Abl-inhibiting compounds to kill Ph þ cells remains only partially elucidated even though caspase-dependent and -independent processes have been evidenced. Another interesting and unresolved question concerns the influence that differentiation and cell death may have on each other in imatinib-treated CML cells. In the present study, we demonstrated that imatinib and PD166326, two Bcr-Abl inhibitors, induced caspase-dependent cell death in CML cell lines and that apoptosis and erythroid differentiation are two independent processes that can be easily dissociated. Finally, our results also indicated that both processes can be distinguished on the basis of their caspase dependence. The notion that Bcr-Abl-mediated cell death and erythroid differentiation are independent events may have interesting applications in the future for the treatment of CML.

100 80 60 40 20 0 Imatinib 1 µM PD166326 10 nM ZVAD-fmk 50 µM

-

+ -

+ -

+

+ +

+ +

Figure 1 Imatinib mesylate and PD166326 induce apoptosis and decreased cell metabolism in K562 cells. (a) Caspase activity was assessed on lysates prepared from cells stimulated for 48 h with increasing concentrations of imatinib or PD166326 by using 0.2 mM Ac-DEVD-pNa as substrate. Substrate hydrolysis was determined at 405 nm. To assess specific caspase activity, hydrolysis was followed at different times in the presence or absence of 10 mM Ac-DEVD-CHO. The surviving cells were quantitated by XTT assay as described in the Materials and methods section. Results are expressed as percentage of response and represent the mean7s.d. of four different determinations. (b) Caspase activity was assessed on lysates prepared from cells stimulated for 48 h with increasing concentrations of imatinib or PD166326 in the presence or absence of 50 mM ZVAD-fmk using 0.2 mM Ac-DEVD-AMC as substrate. Ac-DEVD-AMC hydrolysis was determined following emission at 460 nm after excitation at 390 nm. To assess specific caspase activity, hydrolysis was followed at different times in the presence or absence of 10 mM Ac-DEVD-CHO. Results are expressed as arbitrary units and represent the mean7s.d. of four different determinations. (c) Cells were treated for 48 h with imatinib or PD166326 in the presence or absence of 50 mM ZVADfmk. Cells were then stained with Annexin-V-fluos staining kit according to the manufacturer’s indications. A representative experiment is shown.

Cell metabolism was next assessed by the sodium 30 (l-(phenylaminocarbonnyl)-3,4-tetrazolium)-bis-acid(4methoxy-6-nitro)benzen sulfonic acid hydrat (XTT) assay at different times and various drug concentrations. A 24 h treatment with the Bcr-Abl inhibitor induced a


2447

dose-dependent loss of cell metabolism which reached 30% of the control value at the highest drug concentration and was not affected by caspase inhibition (Figure 2a). After 48 and 72 h in the presence of the drugs, XTT metabolism dropped to only 30 and 20%, respectively, and ZVAD-fmk was found to increase cell metabolism significantly to approximatively 50%

100

XTT metabolism (%)

XTT metabolism (%)

a

80 60 40

-ZVAD

20

+ZVAD

0 0

0.1 0.3

100 80 60 40 20 0 0

1

Ima (µM) 24h 100

XTT metabolism (%)

XTT metabolism (%)

b

80 60 40

-ZVAD

20

+ZVAD

0 0

0.1 0.3

80 60 40 20 0

1

0

-ZVAD +ZVAD

0 0

XTT metabolism (%)

d

XTT metabolism (%)

60

0.1 0.3 Ima (µM) 72h

100 80 60 40 20 0 0 24h -ZVAD Ima 1µM

3

10

100 80 60 40 20 0 0

1

XTT metabolism (%)

XTT metabolism (%)

80

20

1

PD (nM) 48h

100

40

1 3 10 PD (nM) 24h

100

Ima (µM) 48h

c

48h 72h +ZVAD Ima+ZVAD

(Figure 2b–d). Finally, as next shown on Figure 2d, cell metabolism decreased progressively as a function of time in the presence of the Bcr-Abl inhibitors and ZVAD-fmk only partially prevented it. Globally, these findings show that imatinib and PD166326-treated K562 cells can be only partly rescued from the loss of cell metabolism in the presence of the caspase inhibitor, strongly suggesting that apoptosis is only one of the mechanisms used by these drugs to affect cell viability. Therefore, the drug-induced drop of cell metabolism occurred by both caspase-dependent and -independent mechanisms.

1 3 10 PD (nM) 72h

100 80 60 40 20 0 0

24h

48h

72h

-ZVAD

+ZVAD

PD 10nM

PD+ZVAD

Figure 2 Effect of Bcr-Abl inhibitors on K562 cell metabolism. K562 cells were grown in the presence of increasing concentrations of imatinib or PD166326 for 24 h (a), 48 h (b) and 72 h (c). Cell viability was measured by the XTT assay as described in the Materials and methods section. Results represent the mean7s.d. of four different determinations. (d) K562 cells were stimulated for various times with 1 mM imatinib or 10 nM PD166326 in the presence or not of 50 mM ZVAD-fmk. Then, the cells were quantitated by the XTT assay as described in the Materials and methods section. Results represent the mean7s.d. of four different determinations. Error bars are under the limit of detection.

Caspase inhibitors prevented Bcr-Abl inhibitor-mediated apoptosis but not erythroid differentiation of CML cell lines We have previously shown that imatinib induced K562 cells to differentiate towards the erythroid lineage (Jacquel et al., 2003). Figures 3a and 4a illustrate the effect of various concentrations of imatinib and PD166326 on erythroid differentiation as assessed by specific benzidine staining of hemoglobin. Imatinib and PD166326 both induced a dose-dependent increase in erythroid differentiation and apoptosis (Figures 3a and b, 4a and b). Apoptosis was efficiently inhibited in the presence of ZVAD-fmk and cells that failed to die in this condition differentiated towards the erythroid lineage. At the maximal dose of Bcr-Abl inhibitors, 35–40% of cells were benzidine positive and the number of differentiated cells reached 60–65% in the presence of ZVAD-fmk. To follow concomitantly apoptosis and differentiation, we analysed both active caspase 3 and GPA expression by flow cytometry in K562 cells treated for 48 h with imatinib or PD166326 (Figures 3c and 4c). Drugs induced both a dose-dependent increase in GPA expression and caspase 3 activity. At maximally effective concentrations of imatinib and PD166326, there was a conflict between differentiation and apoptosis, with approximatively as much cells that differentiate and die (Figures 3c and 4c). Interestingly, caspase 3-negative cells expressed high level of GPA, suggesting differentiation rather than death. Inhibition of caspases by ZVADfmk was found to abrogate caspase 3 activation and apoptosis and allowed all the cells protected from apoptosis to differentiate towards the erythroid lineage as shown by GPA staining. Finally, the fact that BcrAbl inhibitor-mediated differentiation proceeded normally in the absence of caspase activation was confirmed by reverse transcription–polymerase chain reaction (RT–PCR) analysis of zeta and alpha globin gene expression (Figures 3d and 4d). All together, our findings suggested that although concomitant events, apoptosis and differentiation are two unrelated processes in imatinib and PD166326-treated CML cell lines. A p38 MAP kinase inhibitor prevented erythroid differentiation but not apoptosis in CML cell lines We have previously reported that SB202190, a potent p38 mitogen activated protein (MAP) kinase inhibitor, repressed erythroid differentiation of K562 cells as Oncogene


2448

a 0.1

0.3

Benzidine positive cells (%)

0

70

b

Ima (µM) 1

untreated

+ ZVAD

60 50 40 30 20 10 0 Control

Control

c

104 62%

103

103

102

102 52%

GPA

100 100

10

4

10

3

101

2%

102

103

100

104

4

10

67%

100

16%

101

8%

102

Ima 1 µM

103

104

Ima 1 µM ZVAD

10

87%

3

10

10

102

102 18%

100 100

104

102

4

42%

101

103

4%

101

3

102

100 100

31%

101

Ima 0.3 µM

101

Ima ZVAD

Ima 0.1 µM

104 44%

101

Ima

35%

101

102

6%

101

103

100

104

100

101

4%

102

103

104

Active caspase 3

Imatinib

d µM

0

0.1

0.3

Ima + ZVAD-fmk 1

0

0.1

0.3

1

Globin zeta

Globin alpha

Actin

Figure 3 Imatinib induces caspase-independent erythrocytic differentiation of K562 cells. K562 cells were grown for 48 h with increasing concentrations of imatinib in the presence or absence of 50 mM ZVAD-fmk. (a) Cells were directly photographed or following staining with benzidine stain ( 100). (b) The percentage of benzidine-positive cells was determined after 48 h of treatment. Results represent the mean7s.d. of three different determinations made in quadruplicate. (c) Undifferentiated or 48 h differentiated K562 cells were labeled with anti-active-caspase 3-fluorescein iso thiocyanate (FITC) and anti-glycophorinA phycoerythrin (GPA-PE) monoclonal antibodies. Fluorescence was analysed by using the fluorescence channel 1/fluroscence channel 2 (FL1/FL2) channels of a FACScan. (d) RT–PCR analysis of the expression of the alpha and zeta globin isoforms was realized after 24 h of treatment. Amplification method was described in the Materials and methods section. Actin was used as an invariant control in the experiment.

judged by RT–PCR analysis of erythroid markers (Jacquel et al., 2003; Jacquel et al., 2006). We took advantage of this property to analyse the effect of SB202190 on imatinib-mediated apoptosis and erythroid Oncogene

differentiation using flow cytometry. K562 cells were incubated with or without 10 mM SB202190 and then treated with imatinib for 48 h in either the presence or the absence of ZVAD-fmk. SB202190 alone or in


2449

a 1

3

Benzidine positive cells (%)

0

70

b

PD (µM) 10

untreated

+ ZVAD

60 50 40 30 20 10 0 Control

c

Control 104

103

102

102

100 100 GPA

PD 1 nM

44%

52%

101

2%

102

101

103

100 100

104

PD 3 nM 104

29%

101

PD 10 nM 104

68%

4%

102

104

32%

103

103

103

102

102

102

101 100 100

12%

101

17%

102

21%

101

103

PD ZVAD

104 44%

103

101

PD

100 100

104

44%

101

102

103

PD 10 nM ZVAD 88%

101

103

100 100

104

104

7%

101

3%

102

103

104

Active caspase 3

d

PD166326 nM

0

1

3

PD + ZVAD-fmk 10

0

1

3

10

Globin zeta

Globin alpha

Actin

Figure 4 PD166326 induces caspase-independent erythrocytic differentiation of K562 cells. K562 cells were grown during 48 h in the presence of increasing concentrations of PD166326 in the presence or absence of 50 mM ZVAD-fmk. (a) Cells were photographed directly or after staining with benzidine stain ( 100). (b) The percentage of benzidine-positive cells was determined after 48 h of treatment. Results are the mean7s.d. of three independent experiments. (c) Undifferentiated or 48 h differentiated K562 cells were labeled with anti-active-caspase 3-FITC and anti-GPA-PE monoclonal antibodies. Fluorescence was analysed by using the FL1/FL2 channels of a FACScan. (d) RT–PCR analysis of the expression of the alpha and zeta globin isoforms was realized after 24 h of treatment as previously described. Actin was used as an invariant control in the experiment.

combination with imatinib reduced GPA expression (mean fluorescence ¼ 900 vs 300 in untreated cells and 2200 vs 1000 in the presence of imatinib), but failed to alter caspase 3 activation (Figure 5a and b). Finally,

SB202190 also reduced imatinib-induced differentiation in the presence of ZVAD-fmk (2300 vs 1000 in ima þ ZVAD-treated K562 cells). All together, these findings demonstrate that SB202190 prevents erythroid Oncogene


2450

a

104

K562

104

49%

103

103

103

102

102

102

65%

101

2%

100 100 GPA

Ima1 µM

104

29%

104

101

102

103

16%

101 100 104 100

K562

104

3%

23%

101

102

Ima + ZVAD 71%

untreated 12%

101

103

104

Ima1 µM

100 100

104

34%

1%

103

103

103

102

102

102

101

102

103

104

Ima + ZVAD 43%

+ SB202190 10 µM

10

1

2%

90%

10

100

1

35%

25%

10

100 100

101

102

103

104

49%

1

1%

100 100

101

102

103

104

100

101

102

103

104

Active caspase 3

b 100 80 60 40 20 Control

0 Number of cells

100

101

102

103

104

Imatinib 1µM Ima + ZVAD

100

SB202190 10 µM

80

Ima + SB

60 40 20 0 100

101

102

103

104

GPA

Figure 5 SB202190 prevents erythrocytic differentiation of K562 cells induced by Bcr-Abl inhibitors. K562 cells were cultured for 48 h in the presence of 1 mM imatinib, 10 mM SB202190 or the combination of ima þ ZVAD-fmk (50 mM) or ima þ ZVAD-fmk þ SB202190. (a) Undifferentiated or 48 h differentiated K562 cells were labeled with anti-active-caspase 3-FITC and anti-GPA-PE monoclonal antibodies. Fluorescence was analysed by using the FL1/FL2 channels of a FACScan. (b) Undifferentiated or 48 h differentiated K562 cells were labeled with the anti-GPA-PE monoclonal antibody. Fluorescence was analysed by using the FL1 channel of a FACScan.

Oncogene


2451

Figure 6 Imatinib-resistant K562 cells differentiate towards the erythroid lineage. Parental K562 and imatinib-resistant K562 cells (imatinib R) were cultured in the presence of increasing concentrations of imatinib or PD166326 for 48 h. (a) Cell metabolism was measured by the XTT assay as previously described. Results represent the mean7s.d. of four different determinations. (b) Caspase activity was assessed on lysates prepared from cells stimulated for 48 h by using 0.2 mM Ac-DEVD-pNa. Substrate hydrolysis was determined at 405 nm. To assess specific caspase activity, hydrolysis was followed at different times in the presence or absence of 10 mM Ac-DEVD-CHO. Results are expressed as nanomoles of substrate hydrolysed per minute and per milligram of protein and represent the mean7s.d. of four different determinations. (c) K562 and imatinib-resistant K562 cells were cultured in the presence of 10 nM PD166326 for 48 h. Then, cells were labeled with anti-GPA-PE monoclonal antibody. Fluorescence was analysed by using the FL1 channel of a FACScan. (d) RT–PCR analysis of GPA and alpha and zeta globin isoforms was realized after 24 h of treatment as previously described. Actin was used as an invariant control in the experiments.

a

K562

Imatinib R

XTT metabolism (%)

100 80 60 40 20 0

0 10-3

10-2

10-1

100

PD166326

b

0.25

Caspase 3 activity nmoles/min/mg

Bcr-Abl inhibitor-induced erythroid differentiation occurs normally in different apoptosis-resistant CML cell clones To analyse further the potential dependence of druginduced apoptosis and differentiation, we derived several imatinib and PD166326-resistant clones by long-term incubation (up to 6 months) of parental K562 cells with iterative addition (every 15 days) of increasing concentrations of the Bcr-Abl inhibitors. Figures 6 and 7 depict the results obtained for two representative clones among several others that gave identical results, an imatinib (IIIF5) and a PD166326 (IVH2)-resistant one. Cell metabolism was significantly maintained in resistant clones in the presence of imatinib or PD166326, both clones exhibiting a 10–30-fold decrease in drug sensitivity as compared to parental K562 cells (Figures 6a and 7a). A very moderate caspase 3 activation was detected in IIIF5 and IVH2 clones but only at high concentrations of inhibitors (Figures 6b and 7b). By contrast, imatinib-resistant cells remained sensitive to some other proapoptotic stimuli including staurosporin, arsenic trioxide or resveratrol (not shown). Although drastically resistant to imatinib in term of caspase 3 activation, the IIIF5 and IVH2 clones nevertheless exhibited significant loss of cell metabolism in the presence of the Bcr-Abl inhibitors (Figures 6a and 7a), likely reflecting caspase-independent cell death. Conversely, to the observation of a blockade of apoptosis in imatinib and PD166326-resistant clones, erythroid differentiation proceeded nearly identically in Bcr-Abl inhibitor-treated K562 parental cells and the IIIF5 and IVH2 variants as shown by Facs analysis of GPA staining and RT–PCR analysis of specific globin isoforms (Figures 6c and d, 7c and d), even though expression of GPA was found to be more heterogeneous in resistant cells than in the parental ones. Identical results were obtained for several independent resistant clones (not shown).

shRNAi silencing of caspase 3 blocked imatinib and PD166326-triggered caspase activation but had no effect on erythroid differentiation Caspase 3 has been reported to play a role in erythropoiesis, more particularly in the early stages of the differentiation process in the bone marrow (Carlile et al., 2004). Thus, we thought to determine whether

0.20

101 Log (µM)

Imatinib Imatinib R

K562

0.15 0.10 0.05 0

0 10-3

10-2

10-1

PD166326

c

100

101 Log (µM)

Imatinib

Control PD166326 10 nM 100 K562

80 60 40 20 Number of cells

differentiation, but not apoptosis in CML cell lines. Identical results were obtained with PD166326 (not shown).

0 100

60 50 40 30 20 10

101

102

103

104

102

103

104

Imatinib R

0 100

101

GPA

d

Ima (µM) 0

1

PD (nM) 3

3

10

GPA Globin alpha Globin zeta Actin Imatinib R Oncogene


2452

a

K562

PD166326 R

XTT metabolism (%)

100 80 60 40 20 0

0 10-3

10-2

10-1

100

PD166326

b Caspase 3 activity nmoles/min/mg

Imatinib

K562

0.25

101 Log (µM)

PD166326 R

0.20 0.15 0.10 0.05 0 0 10-3

10-2

10-1

100

PD166326

101 Log (µM)

Imatinib

Control PD166326 10 nM

c 100 K562

80 60 40

Number of cells

20 0 100

60 50 40 30 20 10

101

102

103

104

102

103

104

PD166326 R

0 100

101

GPA Ima (µM)

d 0

1

PD (nM) 3

3

10

GPA Globin alpha Globin zeta Actin PD166326 R

Figure 7 PD166326-resistant K562 cells differentiate towards the erythroid lineage. The conditions used were exactly those of Figure 6. Cell metabolism (a), caspase activity (b), erythoid differentiation (c) and RT–PCR analysis (d) were performed exactly as described above.

caspase 3 was involved in the Bcr-Abl-mediated cell death and differentiation programs in K562 cells. To this aim, we generated several K562 clones inducible for the expression of shRNAi directed against caspase 3. Following stimulation with tetracycline for 5 days, caspase 3 expression could be specifically abrogated (Figure 8a). Silencing of caspase 3 was found to drastically inhibit imatinib-induced Ac-DEVD-ase activity as compared to control K562 cells (Figure 8a). The very low level of residual activity likely reflected a low rate of cleavage of the substrate by other executionary caspases. Caspase 3 silencing only partially increased cell metabolism in cells treated by Bcr-Abl inhibitors (Figure 8b), suggesting the existence of a substantial caspase-independent component in the effect of Bcr-Abl inhibitors on cell metabolism. By contrast, caspase 3 silencing did not affect imatinib and PD166326-induced GPA expression and hemoglobin accumulation (Figure 8c and data not shown), confirming that BcrAbl inhibitor-mediated apoptosis and differentiation are independent processes. Taken together, our results also demonstrate that caspase 3 activity is required for BcrAbl-mediated cell death but dispensable for erythroid differentiation of the K562 cell line. shRNAi silencing of Bcr-Abl mimicked the effects of imatinib and PD166326 on cell death and differentiation Besides their well-documented effects as Bcr-Abl inhibitors, imatinib and PD166326 may also affect other kinases such as c-Kit, the platelet derived growth factor (PDGF) receptor and, to a less extent, Src kinases (Buchdunger et al., 2000; Heinrich et al., 2000; Huron et al., 2003). To specifically assess the role of Bcr-Abl inhibition in the cell death and differentiation programs of K562 cells, we generated several clones inducible for the expression of a shRNAi directed against Bcr-Abl. Stimulation with tetracycline for 7 days efficiently inhibited Bcr-Abl expression (Figure 9a). Bcr-Abl silencing resulted in the induction of apoptosis, which culminates in caspase 3 activation and could be abrogated by ZVAD-fmk. BcrAbl silencing drastically decreased cell metabolism, to 30 and 5%, respectively, at day 5 and 7 following tetracycline addition. As previously shown for caspase 3 silencing, ZVAD-fmk was found to only partially rescue K562 cells from loss of cell metabolism (Figure 9b). Interestingly, reduced Bcr-Abl expression correlated with increased erythroid differentiation as assessed by GPA expression (Figure 9c and d). ZVAD-fmk efficiently blocked caspase activation and apoptosis and allowed most of K562 cells to differentiate towards the erythroid lineage. Identical results were obtained using several other K562 cell clones (not shown). These findings demonstrated that Bcr-Abl silencing reproduced the effect of Bcr-Abl inhibitors on cell death and differentiation of CML cell lines. Discussion The present study reports that Bcr-Abl inhibitormediated apoptosis and erythroid differentiation, which proceed concomitantly in several CML cell lines, are

Oncogene


2453 Caspase 3 activity nmoles/min/mg

a

0.30 0.25

K562 casp.3 shRNAi

0.20 0.15 0.10 0.05 0

MW(kDa)

Caspase 3

32

Hsp60

60

Imatinib 1 µM

-

+

-

-

+

-

-

+

-

PD166326 10 nM

-

-

+

-

-

+

-

-

+

-

-

+

Tetracycline 1 µg/ml

-

-

-

+

+

+

-

-

-

+

+

+

b

K562

-

+

-

casp.3 shRNAi

XTT metabolism (%)

100

c

80 60 40 20

0 Imatinib 1 µM

-

+

-

-

+

-

-

+

-

-

+

-

PD166326 10 nM

-

-

+

-

-

+

-

-

+

-

-

+

Tetracycline 1 µg/ml

-

-

-

+

+

+

-

-

-

+

+

+

K562

Casp. 3 shRNAi

100

80

80

60

60

40

40 Number of cells

20 0 100

Control

0 101

102

103

104

100

100

101

102

103

104

80

80

Imatinib 1 µM PD166326 10 nM

60

60

40

40

tetracycline

20

20 0 100

untreated

20

0 101

102

103

104

100

101

102

103

104

GPA

Figure 8 Inhibition of caspase 3 expression by shRNAi does not modify the capacity of K562 cells to differentiate towards the erythroid lineage. After a 5-day pretreatment with 1 mg/ml tetracycline to allow optimal extinction of caspase 3, K562 and caspase 3 shRNAi K562 (tet-on) cells were cultured for 48 h in the presence of 1 mM imatinib, 10 nM PD166326, 1 mg/ml tetracycline or the combination of imatinib þ tetracycline or PD166326 þ tetracycline. (a) Caspase activity was assessed as previously described. Results are expressed as nanomoles of substrates per minute per milligram of protein and represent the mean7s.d. of four different determinations. Cell lysates were also analysed by SDS–PAGE on 10% polyacrylamide gels. Proteins were then transferred onto PVDF membranes and blotted with either anti-caspase 3 or anti-Hsp60 antibodies. (b) Cell metabolism was determined by the XTT assay as described previously. Results are the mean7s.d. of three independent experiments made in quadruplicate. (c) After 48 h, K562 and caspase 3 shRNAi K562 cells were labeled with an anti-GPA-PE monoclonal antibody and fluorescence was analysed by using the FL1 channel of a FACScan.

Oncogene


2454 Caspase 3 activity nmoles/min/mg

a

b

0.12 0.10 0.08 0.06 0.04 0.02 0

MW(kDa) 100 210 80 XTT metabolism (%)

Bcr-Abl

145 115 PARP 85

40

20

60

Hsp60 Tetracycline

-

ZVAD-fmk

5d

-

7d

-

c Number of cells

60

5d

-

0 Tetracycline

-

5d

7d

5d

7d

ZVAD-fmk

-

-

-

+

+

7d

+

+

untreated

+ ZVAD-fmk

50

50

40

40

30

30

20

20

10

10

0

Tetra 5 days Tetra 7 days Control

0 100

101

102

103

104

100

101

102

103

104

GPA

d

Control 104

Tetra 7 days 104

11%

31%

103

103

GPA

Tetra 7 days + ZVAD 104

102

103

102

5%

1

10

10

5%

1

1

10

22%

19%

0

10

10

101

102

28%

67% 100

52%

102

103

104

0

0

10 100

101

102

103

104

100

101

102

103

104

Active caspase 3

Figure 9 Inhibition of Bcr-Abl expression by shRNAi mimics the proapototic and differentiating effects of Bcr-Abl inhibitors. BcrAbl shRNAi K562 (tet-on) was cultured for 5 or 7days in the presence of 1 mg/ml tetracycline or the combination tetracycline þ 50 mM ZVAD-fmk. (a) Caspase activity was assessed as previously. Results are expressed as nanomoles of substrates per minute per milligram of protein and represent the mean7s.d. of four different determinations. Cells lysates were also analysed by SDS–PAGE on 10% polyacrylamide gels. Proteins were then transferred onto PVDF membranes and blotted with either anti-Bcr-Abl, anti-PARP or antiHsp60 antibodies. (b) Cell metabolism was determined as described in Figure 8b. Results are the mean7s.d. of three independent determinations. (c) Bcr-Abl shRNAi K562 cells were labeled with anti-GPA-PE monoclonal antibody. Fluorescence was analysed by using the FL1 channel of a FACScan. (d) Bcr-Abl shRNAi K562 cells were labeled with anti-active-caspase 3-FITC and anti-GPA-PE monoclonal antibodies. Fluorescence was analysed by using the FL1/FL2 channels of a FACScan.

fully distinguishable processes. This conclusion relies on the following observations: (i) suppression of apoptosis by ZVAD-fmk, a pancaspase inhibitor, does not prevent imatinib or PD166326-induced erythroid differentiation of K562 cells; (ii) SB202190, an inhibitor of the p38 Oncogene

MAP kinase pathway, efficiently blocks Bcr-Abl inhibitor-mediated erythroid differentiation without altering apoptosis; (iii) several K562 cell clones generated for their resistance to the induction of apoptosis by imatinib and PD166326 exhibited normal erythroid


2455

differentiation as compared to their parental counterparts; (iv) caspase 3 silencing in K562 cells using stably transfected and inducible shRNAi that abrogates caspase 3 expression and activity has no effect on Bcr-Abl inhibitor-mediated erythroid differentiation and finally (v) inducible Bcr-Abl silencing reproduced accurately the effects of Bcr-Abl inhibitors on cell death and differentiation of K562 cells, demonstrating the involvement of Bcr-Abl in both processes. Taken together, our results also show that erythroid differentiation does not require caspase activation at least in some CML lines. The pathway by which chemotherapeutic agents such as imatinib trigger erythroid differentiation of CML cell lines remains unclear. A possible mechanism would be that stress conditions favor erythroid differentiation. However, this is unlikely the case in the present study, as stressing K562 cells with a large panel of effectors does not consistently lead to erythroid differentiation (not shown). Moreover, we failed to find any correlation between stress-induced caspase activation and differentiation in most of cases. Finally, imatinib-induced caspase activation and differentiation were unaffected by antioxidants, suggesting that reactive oxygen species production is not involved in these processes (not shown). A recent study by Okada et al. (2004) reported that imatinib-induced cell death in CML cells occurs by both caspase-dependent and -independent mechanisms. The partial caspase independence of Bcr-Abl inhibitors is confirmed in our study by the observation that even in the presence of ZVAD-fmk, a significant loss of cell metabolism (around 50%) could be still detected upon imatinib and PD166326 treatment of CML cell lines. These results strongly suggest that other types of cell death and/or growth arrest may contribute to the effect of Bcr-Abl inhibitors on cell metabolism. This finding is also reinforced by the observation of a diminished cell rate proliferation in K562 clones, nevertheless resistant in term of apoptosis to Bcr-Abl inhibitors. The caspaseindependent counterpart of the imatinib-induced cell death could be likely owing to activation of HtrA serine peptidase 2 (HTRA2) as described recently (Okada et al., 2004) or to other currently unrevealed mechanisms. However, in our hands, neither tosyl-L-lysine chloromethylketone, a general trypsin-like inhibitor, nor the more specific OMI/HTRA2 inhibitor, UFC101, were able to further increase cell metabolism in the presence of Bcr-Abl inhibitors. To our knowledge, this is the first report that studies specifically the relationship between cell death and differentiation in CML cell lines. This led us to develop several K562 cell clones that exhibited variable levels of imatinib and PD166326 resistance. Globally, these K562 cell variants are highly resistant to Bcr-Abl inhibitorinduced caspase activation and apoptosis, but are perfectly able to differentiate towards the erythroid lineage in response to these drugs, confirming that BcrAbl inhibitor-mediated apoptosis and differentiation are unrelated processes. To definitely eliminate a possible implication of caspases in Bcr-Abl inhibitor-induced erythroid differ-

entiation of K562 cells, we derived several stable K562 clones in which caspase 3 or caspase 9 (not shown) silencing can be achieved by inducible shRNAi. Although caspase 3 and 9 expression and Bcr-Abl inhibitor-induced caspase activity can be selectively abrogated in these cell lines upon tetracycline treatment, erythroid differentiation does occur in the presence of the drugs ruling out the involvement of these two caspases in this process. Several recent studies have demonstrated the ability of small interfering RNAs to reduce Bcr-Abl expression in CD34 þ cells from CML patients (Scherr et al., 2003; Rangatia and Bonnet, 2005) and different established cell lines including the K562 CML cell line (Wohlbold et al., 2003; Rangatia and Bonnet, 2005; Withey et al., 2005), conferring increased sensitivity to apoptosis. So we thought to determine, using inducible shRNA interference, whether Bcr-Abl silencing could mimic the effect of imatinib and PD166326 on K562 cell death and differentiation. This was of importance as it was not known whether Bcr-Abl inhibitor-mediated apoptosis and differentiation both relied on Bcr-Abl inhibition. We found that incubation of K562 cells for 5–7 days in the presence of tetracycline efficiently inhibited Bcr-Abl expression, an event accompanied by caspase activation and erythroid differentiation as shown by fluorescenceactivated cell sorting (FACS) analysis of GPA expression and specific caspase assays. Interestingly, as previously shown for Bcr-Abl inhibitors, blockade of caspase activity with ZVAD-fmk abrogates tetracyclineinduced caspase 3 activation and allows all the surviving cells to differentiate towards the erythroid lineage. Identical results were obtained with three different K562 cell clones and an excellent correlation between residual level of Bcr-Abl expression and sensitivity to apoptosis was evidenced (not shown). Very recently, Rangatia and Bonnet (2005) reached the same conclusions using transient or long-term silencing of Bcr-Abl in K562 cells and samples from CML patients. Our present study provides thus the first example of stable and inducible CML cell lines in which Bcr-Abl silencing leads to both massive apoptosis and erythroid differentiation. Interestingly, we also show that silencing BcrAbl reproduced accurately the effects of Bcr-Abl inhibitors such as imatinib or PD166326 on cell death and differentiation. Finally, this observation also indicates that although different signaling pathways are involved in both processes, Bcr-Abl inhibition is critical for both apoptosis and erythroid differentiation at least concerning the K562 cell line. As previously mentioned, imatinib and PD166326 targeted the Bcr-Abl protein that is responsible for CML in humans. As a consequence, the main survival pathways including Erk1/2, PI3K, JAK/STAT and the NF-kB cascade are all altered by the inhibitor. By contrast, several pieces of evidence indicate that imatinib induced activation of the p38 MAP kinase pathway and that could be important for the differentiation of CML cells towards the erythroid lineage (Kohmura et al., 2004; Parmar et al., 2004). In a recent study, Parmar et al. reported that imatinib inhibited the Oncogene


2456

growth of leukemic myeloid progenitors from CML patients in a p38 MAP kinase-dependent manner. The fact that SB202190 fails to inhibit the apoptotic effect of imatinib on K562 cells despite a strong inhibition of erythroid differentiation is another argument demonstrating that imatinib-induced differentiation and apoptosis of CML cell lines are two independent processes. At present, it is not known whether Bcr-Abl inhibition by imatinib interferes with p38 MAP kinase but the different K562 cell clones generated in the present study offer an excellent opportunity to analyse how Bcr-Abl silencing leads to p38 MAP kinase activation. Increasing evidence indicates that besides their wellestablished function in apoptosis, caspases could play important regulatory function in proliferation and differentiation of hematopoietic cells and cells from other origin (Perfettini and Kroemer, 2003; Carlile et al., 2004; Launay et al., 2005). More specifically, caspases have been shown to be required for early-stage erythropoiesis. Of note, Carlile et al. reported that inhibition of caspase by RNA interference blocked cells at the proerythroblast stage, suggesting that caspases are involved in the stages of erythropoiesis encompassing burst forming unit erythroid differentiation to proerythroblast. Clearly, at least in the CML models used in our study, erythroid differentiation can perfectly occur in the presence of caspase inhibitors. Moreover, inhibiting apoptosis by ZVAD-fmk allows the surviving cells to differentiate towards the erythroid lineage. Thus, to some extent, caspase inhibition favors erythroid differentiation. Finally, in agreement with our present findings, Krauss et al. recently reported that late-stage erythropoiesis does not involve caspase-dependent mechanisms (Krauss et al., 2005). In conclusion, we demonstrate for the first time that Bcr-Abl-mediated apoptosis and erythroid differentiation are two independent processes which can be fully distinguished notably regarding their sensitivity to caspase and p38 MAP kinase inhibition. The K562 cell lines, which are now at our disposal, will represent invaluable tools to dissect at the molecular and cellular level the role of individual caspases and of Bcr-Abl on the cell death and differentiation programs of established CML cell lines. Finally, the results presented here might have interesting therapeutic implication in the treatment of CML. Indeed, induction of erythroid differentiation may be useful for those CML patients who are resistant to imatinib. Moreover, drugs that modulate the p38 MAP kinase pathway should exhibit some therapeutic benefit in such patients. Materials and methods Reagents and antibodies Imatinib (STI571, Gleevec) and PD166326 were kindly provided by Novartis Pharma (Basel, Switzerland) and Pfizer (Groton, CT, USA), respectively. Roswell Park Memorial Institute 1640 (RPMI 1640) medium and fetal calf serum (FCS) were purchased from Gibco BRL (Paisley, UK). Sodium fluoride, sodium orthovanadate, phenylmethylsulfonyl Oncogene

fluoride (PMSF), aprotinin and leupeptin were purchased from Sigma (Saint Louis, MO, USA). SB202190 was from Calbiochem (La Jolla, CA, USA). Ac-Asp-Glu-Val-Asp-pNA (para-nitroaniline), Ac-DEVD-AMC (7-amino-4methylcoumarin), Ac-DEVD-CHO and ZVAD-fmk (ZVAD) were from Alexis Biochemicals (Lausan, Switzerland). Anti-caspase 3 and anti-Hsp60 antibodies were purchased from Transduction Laboratories (Lexington, KY, USA) and Santa Cruz Biotechnology (Santa Cruz, CA, USA), respectively. HRP-conjugated anti-mouse and anti-goat antibodies were from Dakopatts (Glostrup, Denmark). Cell lines The human chronic myelogenous leukaemia cells K562 and JURL-MK1 were grown at 371C under 5% CO2 in RPMI 1640 medium (Gibco BRL, Paisley, UK) supplemented with 5 or 10% FCS (Gibco BRL, Paisley, UK), 50 U/ml penicillin, 50 mg/ml streptomycin and 1 mM sodium pyruvate. From the K562 cell line, we established resistant clones by addition in the culture medium of increasing concentrations of imatinib (up to 8 mM) and PD166326 (up to 10 nM). Ten clones either resistant to imatinib or to PD166326 were selected by limiting dilutions. For convenience, only the results obtained with two of these clones resistant, respectively, to imatinib (IIIF5 or imatinib R) and PD166326 (IVH2 or P166326 R) are presented. Caspase activity measurement After stimulation, cells were lysed for 30 min at 41C in lysis buffer (50 mM HEPES pH 8, 150 mM NaCl, 20 mM ethylene diaminetetraacetic acid (EDTA), 1 mM PMSF, 10 mg/ml leupeptin, 10 mg/ml aprotinin and 0.2% Triton X-100), and lysates were cleared at 10 000 g for 15 min at 41C. Each assay (in triplicate) was performed with 50 mg of protein prepared from control or stimulated cells. Briefly, cellular extracts were then incubated in a 96-well plate, with 0.2 mM of Ac-DEVDpNA or Ac-DEVD-AMC as substrates for various times at 371C as previously described (Herrant et al., 2004). Caspase activity was measured either following absorbance at 405 nm or emission at 460 nm (excitation at 390 nm) in the presence or not of 10 mM of Ac-DEVD-CHO. Enzyme activities were determined as initial velocities expressed as nanomoles of pNa released per minute and per milligram of protein or relative intensity per minute and per milligram of protein. Cell viability (XTT) Cells (15 103 cells/100 ml) were incubated in a 96-well plate with different effectors for the times indicated in the figure legends. In total, 50 ml of XTT reagent (sodium 30 -[1(phenylaminocarbonyl)-3,4-tetrazolium]-bis(4-methoxy-6-nitro) benzene sulfonic acid hydrate) was added to each well. The assay is based on the cleavage of the yellow tetrazolium salt XTT to form an orange formazan dye by metabolically active cells. The absorbance of the formazan product, reflecting cell viability, was measured at 490 nm. Each assay was performed in quadruplicate. Benzidine staining and phase contrast microscopy Cell hemoglobinization was analysed by benzidine staining. In total, 1000 ml of (0.5–1 106 cells/ml) was mixed with 200 ml benzidine dihydrochloride reagent (Sigma, Saint Louis, MO, USA). Morphological changes characteristic of erythroid differentiation were visualized using standard phase optics (Zeiss, Oberkochen, Germany).


2457 Flow cytometry After stimulation, cells were washed with ice-cold phosphatebuffered saline (PBS) and incubated at 41C for 30 min in 100 ml of PBS containing 0.1% bovine serum albumin (BSA) with anti-GPA-PE monoclonal antibody. After one wash with PBS, cells were fixed and permeabilized with Cytofix/Cytopermt solution. After two washes with Perm/Washt solution at room temperature, cells were incubated for 30 min with anti-activecaspase 3-FITC monoclonal antibody. Finally, cells were washed and resuspended in Perm/Washt solution. All reagents were from BD Bioscience (San Diego, CA, USA). In some experiments, control cells or Bcr-Abl inhibitorstreated cells were incubated for 48 h in the presence or absence of ZVAD-fmk and stained with the annexin-V-fluos staining kit (Roche, Meylan, France) according to the manufacturer’s procedure. Fluorescence was measured by using the FL1 and FL2 channels of a FACS apparatus (FACScan, Becton Dickinson, Cowley, UK). RT–PCR Analysis After cell treatment, total RNAs were isolated using Trizol Reagent (Invitrogen, Cergy Pontoise, France). The supernatant was cleared by centrifugation, precipitated with isopropanol, and resuspended in RNAse and DNAsefree water. Total RNA (5 mg) was reverse transcribed using the SuperScript II reverse transcriptase (Invitrogen, Cergy Pontoise, France) following the manufacturer’s instructions in a 40 ml final volume. cDNAs (2 ml) were amplified in a final volume of 10 ml using 1 U of Taq polymerase (New England Biolabs, Ipswich, MA, USA), containing buffer with 1.5 mM MgCl2, 0.2 mM deoxynucleoside triphosphate (dNTP) and 0.5 mM of the forward and reverse following primers: Globin alpha-s: CTGGAGAGGATGTTCCTGTCCTTG Globin alpha-as: CAGCTTAACGGTATTTGGAGGTCAT Globin zeta-s: ACCAAGACTGAGAGGACCATCATTA Globin zeta-as: TCAGGACAGAGGATACGACCGATAC GPA-s: GGAATTCCAGCTCATGATCTCAGGATG GPA-as: TCCACATTTGGTTTGGTGAACAGATTC Actin-s: ACCCACACTGTGCCCATCTTA Actin-as: CTAGAAGCATTTGCGGTGGA.

Caspase 3 and Bcr-Abl shRNAi siRNA-expressing plasmids were constructed targeting a specific region of human caspase 3 (50 -AGTGAAGCAAATCAGAAAC-30 ) (Carlile et al., 2004) or human BCR-ABL (50 AGCAGAGTTCAAAAGCCCT-30 ) (Scherr et al., 2003). The oligonucleotides purchased from Eurogentec s.a. (Seraing, Belgium) were annealed and cloned into the pTER vector (a generous gift from Dr Hans Clevers at the Hubrecht Laboratory, Utrecht, The Netherlands) using the BglII and HindIII restriction sites (van de Wetering et al., 2003). This result in a tet-inducible shRNA expression system when stably co-transfected into K562 (tet-on) cells with the pcDNA6/TR vector (Invitrogen Ltd, Paisley, UK). Western blot assays After stimulation, cells were lysed for 30 min at 41C in lysis buffer (50 mM HEPES pH 7.4, 150 mM NaCl, 20 mM EDTA, 100 mM NaF, 10 mM Na3VO4, 1 mM PMSF, 10 mg/ml leupeptin, 10 mg/ml aprotinin, and 1% Triton X-100). Lysates were centrifuged at 10 000 g for 10 min at 41C and supernatants were supplemented with concentrated sodium dodecyl sulfate (SDS) sample buffer. A total of 100 mg of protein were separated on 12% polyacrylamide gel and transferred onto polyvinylidine fluoride (PVDF) membrane (Immobilon-P, Millipore, Bedford, MA, USA) in a 20 mM Tris, 150 mM glycine and 20% methanol buffer at 500 mA during 4 h at 41C. After blocking nonspecific binding sites in saturation buffer (50 mM Tris pH 7.5, 50 mM NaCl, 0.15% Tween, 5% BSA), the membranes were incubated with specific antibodies. The membranes were washed three times using Tris 50 mM pH 7.5, NaCl 150 mM, 1% NP-40 incubated further with HRP-conjugated antibody for 1 h at room temperature. Immunoblots were revealed using the enhanced chemiluminescence detection kit (Amersham Biosciences, Uppsala, Sweden). Acknowledgements We are indebted to Dr Hans Clever for the kind gift of the pTER vector for inducible expression of shRNAi. We thank Novartis for the kind gift of imatinib mesylate and Pfizer for that of PD166326. SM and AJ are recipients of fellowships from the Fondation contre la Leuce´mie and the Ligue Nationale contre le Cancer, respectively.

References Belhacene N, Maulon L, Guerin S, Ricci JE, Mari B, Colin Y et al. (1998). Differential expression of the Kell blood group and CD10 antigens: two related membrane metallopeptidases during differentiation of K562 cells by phorbol ester and hemin. FASEB J 12: 531–539. Buchdunger E, Cioffi CL, Law N, Stover D, Ohno-Jones S, Druker BJ et al. (2000). Abl protein-tyrosine kinase inhibitor STI571 inhibits in vitro signal transduction mediated by c-kit and platelet-derived growth factor receptors. J Pharmacol Exp Ther 295: 139–145. Carlile GW, Smith DH, Wiedmann M. (2004). Caspase-3 has a nonapoptotic function in erythroid maturation. Blood 103: 4310–4316. Deininger MW, Goldman JM, Melo JV. (2000). The molecular biology of chronic myeloid leukemia. Blood 96: 3343–3356. Dorsey JF, Cunnick JM, Mane SM, Wu J. (2002). Regulation of the Erk2-Elk1 signaling pathway and megakaryocytic differentiation of Bcr-Abl(+) K562 leukemic cells by Gab2. Blood 99: 1388–1397. Heinrich MC, Griffith DJ, Druker BJ, Wait CL, Ott KA, Zigler AJ. (2000). Inhibition of c-kit receptor tyrosine kinase

activity by STI 571, a selective tyrosine kinase inhibitor. Blood 96: 925–932. Herrant M, Jacquel A, Marchetti S, Belhacene N, Colosetti P, Luciano F et al. (2004). Cleavage of Mcl-1 by caspases impaired its ability to counteract Bim-induced apoptosis. Oncogene 23: 7863–7873. Huron DR, Gorre ME, Kraker AJ, Sawyers CL, Rosen N, Moasser MM. (2003). A novel pyridopyrimidine inhibitor of abl kinase is a picomolar inhibitor of Bcr-abl-driven K562 cells and is effective against STI571-resistant Bcr-abl mutants. Clin Cancer Res 9: 1267–1273. Jacquel A, Herrant M, Defamie V, Belhacene N, Colosetti P, Marchetti S et al. (2006). A survey of the signaling pathways involved in megakaryocytic differentiation of the human K562 leukemia cell line by molecular and c-DNA array analysis. Oncogene 25: 781–794. Jacquel A, Herrant M, Legros L, Belhacene N, Luciano F, Pages G et al. (2003). Imatinib induces mitochondriadependent apoptosis of the Bcr-Abl-positive K562 cell line and its differentiation toward the erythroid lineage. FASEB J 17: 2160–2162. Oncogene


2458 Kawano T, Horiguchi-Yamada J, Iwase S, Furukawa Y, Kano Y, Yamada H. (2004). Inactivation of ERK accelerates erythroid differentiation of K562 cells induced by herbimycin A and STI571 while activation of MEK1 interferes with it. Mol Cell Biochem 258: 25–33. Kohmura K, Miyakawa Y, Kawai Y, Ikeda Y, Kizaki M. (2004). Different roles of p38 MAPK and ERK in STI571induced multi-lineage differentiation of K562 cells. J Cell Physiol 198: 370–376. Kolbus A, Pilat S, Husak Z, Deiner EM, Stengl G, Beug H et al. (2002). Raf-1 antagonizes erythroid differentiation by restraining caspase activation. J Exp Med 196: 1347–1353. Krauss SW, Lo AJ, Short SA, Koury MJ, Mohandas N, Chasis JA. (2005). Nuclear substructure reorganization during late-stage erythropoiesis is selective and does not involve caspase cleavage of major nuclear substructural proteins. Blood 106: 2200–2205. Kuzelova K, Grebenova D, Marinov I, Hrkal Z. (2005). Fast apoptosis and erythroid differentiation induced by imatinib mesylate in JURL-MK1 cells. J Cell Biochem 95: 268–280. Launay S, Hermine O, Fontenay M, Kroemer G, Solary E, Garrido C. (2005). Vital functions for lethal caspases. Oncogene 24: 5137–5148. O’Brien SG, Guilhot F, Larson RA, Gathmann I, Baccarani M, Cervantes F et al. (2003). Imatinib compared with interferon and low-dose cytarabine for newly diagnosed chronic-phase chronic myeloid leukemia. N Engl J Med 348: 994–1004. Okada M, Adachi S, Imai T, Watanabe K, Toyokuni SY, Ueno M et al. (2004). A novel mechanism for imatinib mesylate-induced cell death of BCR-ABL-positive human leukemic cells: caspase-independent, necrosis-like programmed cell death mediated by serine protease activity. Blood 103: 2299–2307. Parmar S, Katsoulidis E, Verma A, Li Y, Sassano A, Lal L et al. (2004). Role of the p38 mitogen-activated protein kinase pathway in the generation of the effects of imatinib mesylate (STI571) in BCR-ABL-expressing cells. J Biol Chem 279: 25345–25352. Perfettini JL, Kroemer G. (2003). Caspase activation is not death. Nat Immunol 4: 308–310. Pettiford SM, Herbst R. (2003). The protein tyrosine phosphatase HePTP regulates nuclear translocation of

Oncogene

ERK2 and can modulate megakaryocytic differentiation of K562 cells. Leukemia 17: 366–378. Racke FK, Wang D, Zaidi Z, Kelley J, Visvader J, Soh JW et al. (2001). A potential role for protein kinase C-epsilon in regulating megakaryocytic lineage commitment. J Biol Chem 276: 522–528. Rangatia J, Bonnet D. (2005). Transient or long-term silencing of BCR-ABL alone induces cell cycle and proliferation arrest, apoptosis and differentiation. Leukemia 20: 68–76. Roskoski Jr R. (2003). STI-571: an anticancer protein-tyrosine kinase inhibitor. Biochem Biophys Res Commun 309: 709–717. Scherr M, Battmer K, Winkler T, Heidenreich O, Ganser A, Eder M. (2003). Specific inhibition of bcr-abl gene expression by small interfering RNA. Blood 101: 1566–1569. Steelman LS, Pohnert SC, Shelton JG, Franklin RA, Bertrand FE, McCubrey JA. (2004). JAK/STAT, Raf/MEK/ERK, PI3K/Akt and BCR-ABL in cell cycle progression and leukemogenesis. Leukemia 18: 189–218. van de Wetering M, Oving I, Muncan V, Pon Fong MT, Brantjes H, van Leenen D et al. (2003). Specific inhibition of gene expression using a stably integrated, inducible smallinterfering-RNA vector. EMBO Rep 4: 609–615. Withey JM, Harvey AJ, Crompton MR. (2005). RNA interference targeting of Bcr-Abl increases chronic myeloid leukemia cell killing by 17-allylamino-17-demethoxygeldanamycin. Leuk Res 30: 553–560. Wohlbold L, Van der Kuip H, Miething C, Vornlocher HP, Knabbe C, Duyster J et al. (2003). Inhibition of bcr-abl gene expression by small interfering RNA sensitizes for imatinib mesylate (STI571). Blood 102: 2236–2239. Wolff NC, Veach DR, Tong WP, Bornmann WG, Clarkson B, Ilaria Jr RL. (2005). PD166326, a novel tyrosine kinase inhibitor, has greater antileukemic activity than imatinib mesylate in a murine model of chronic myeloid leukemia. Blood 105: 3995–4003. Wong S, Witte ON. (2004). The BCR-ABL story: bench to bedside and back. Annu Rev Immunol 22: 247–306. Zermati Y, Garrido C, Amsellem S, Fishelson S, Bouscary D, Valensi F et al. (2001). Caspase activation is required for terminal erythroid differentiation. J Exp Med 193: 247–254.