The role of Mcl-1 downregulation in the proapoptotic activity ... - Nature

3 downloads 24 Views 255KB Size Report
Jun 27, 2005 - We are grateful to Keith C Bible, Neil E Kay, Paul Haluska,. Crescent ... Boucher MJ, Morisset J, Vachon PH, Reed JC, Laine J and. Rivard N.
Oncogene (2005) 24, 6861–6869

& 2005 Nature Publishing Group All rights reserved 0950-9232/05 $30.00 www.nature.com/onc

ORIGINAL PAPERS

The role of Mcl-1 downregulation in the proapoptotic activity of the multikinase inhibitor BAY 43-9006 Chunrong Yu1, Laura M Bruzek1, Xue Wei Meng1, Gregory J Gores2, Christopher A Carter3, Scott H Kaufmann1,2 and Alex A Adjei*,1 1 Division of Medical Oncology, Mayo Clinic, 200 First Street. SW, Rochester, MN 55905, USA; 2Department of Medicine, Mayo Clinic, Rochester, MN 55905, USA; 3Bayer HealthCare, 400 Morgan Lane, West Haven, CT 06516, USA

BAY 43-9006, a multikinase inhibitor that targets Raf, prevents tumor cell proliferation in vitro and inhibits diverse human tumor xenografts in vivo. The mechanism of action of BAY 43-9006 remains incompletely defined. In the present study, the effects of BAY 43-9006 on the antiapoptotic Bcl-2 family member Mcl-1 were examined. Treatment of A549 lung cancer cells with BAY 43-9006 diminished Mcl-1 levels in a time- and dose-dependent manner without affecting other Bcl-2 family members. Similar BAY 43-9006-induced Mcl-1 downregulation was observed in ACHN (renal cell), HT-29 (colon), MDAMB-231 (breast), KMCH (cholangiocarcinoma), Jurkat (acute T-cell leukemia), K562 (chronic myelogenous leukemia) and MEC-2 (chronic lymphocytic leukemia) cells. Mcl-1 mRNA levels did not change in BAY 43-9006treated cells. Instead, BAY 43-9006 enhanced proteasome-mediated Mcl-1 degradation. This Mcl-1 downregulation was followed by mitochondrial cytochrome c release and caspase activation as well as enhanced sensitivity to other proapoptotic agents. The caspase inhibitor Boc-D-fmk inhibited BAY 43-9006-induced caspase activation but not cytochrome c release. In contrast, Mcl-1 overexpression inhibited cytochrome c release and other features of BAY 43-9006-induced apoptosis. Conversely, Mcl-1 downregulation by short hairpin RNA enhanced BAY 43-9006-induced apoptosis. Collectively, these findings demonstrate that drug-induced Mcl-1 downregulation contributes to the proapoptotic effects of BAY 43-9006. Oncogene (2005) 24, 6861–6869. doi:10.1038/sj.onc.1208841; published online 27 June 2005 Keywords: Mcl-1; apoptosis; BAY 43-9006 (sorafenib); MEK; Raf; adaphostin

Introduction The Raf family of serine/threonine kinases, which consists of A-Raf, B-Raf and C-Raf, plays an important role in proliferative signaling in mammalian cells. *Correspondence: AA Adjei; E-mail: [email protected] Received 28 March 2005; revised 3 May 2005; accepted 4 May 2005; published online 27 June 2005

Recently, activating mutations of B-Raf have been reported in approximately 70% of malignant melanomas and in a smaller fraction of other cancer types (ovarian, thyroid, colon, lung) (Davies et al., 2002; Pollock and Meltzer, 2002; Cohen et al., 2003; Mercer and Pritchard, 2003; Singer et al., 2003; Tannapfel et al., 2003). In melanoma cells expressing mutant B-Raf, constitutive extracellular signal-regulated kinase (ERK) activation is required for proliferation. In these cells, depletion of B-Raf by small inhibitory RNA (siRNA) inhibits ERK phosphorylation, DNA synthesis and continued cell survival, whereas depletion of A-RAF or C-RAF does not. Coupled with additional data (Wan et al., 2004), these results suggest that B-Raf may be an important cancer therapeutic target (Dhillon et al., 2003; Karasarides et al., 2004). BAY 43-9006 is an orally bioavailable bisarylurea that has been developed as a multikinase inhibitor. It is a potent inhibitor of Raf kinases, with IC50 values of 2, 25, and 38 nM against C-Raf, wild-type B-Raf, and mutant B-Raf in vitro under cell-free conditions (Wilhelm et al., 2004). Anticancer activity has been observed in cancers that harbor Ras and B-raf mutations, as well as in cancers in which the Ras-ERK pathway is activated through overexpression of growth factor receptors, suggesting the potential use of this compound in a wide spectrum of cancers. Under cellfree conditions, BAY 43-9006 has also been shown to be a potent inhibitor of receptor tyrosine kinases that bind vascular endothelial growth factor (VEGFR-2, IC50 ¼ 90 nM) and platelet-derived growth factor b (IC50 ¼ 38 nM) (Wilhelm et al., 2004), two molecules that are important mediators of tumor angiogenesis, as well as c-Kit, Flt-3, and p38. Thus, the anticancer activity of BAY 43-9006 is currently thought to be due to a dual mechanism of action of inhibiting tumor cell proliferation and tumor angiogenesis (Wilhelm et al., 2004). In clinical trials, the recommended phase II dose of 400 mg BAY 43-9006 twice daily results in a steady-state mean plasma concentration of approximately 10 mM (Strumberg et al., 2005). Phase II trials of BAY 439006 have indicated clinical antitumor activity. In renal cell carcinoma, a X25% tumor shrinkage was documented in 40% of patients, with a median survival of 12.3 months (Ratain et al., 2004). In spite of the

BAY 43-9006-induced Mcl-1 downregulation and apoptosis C Yu et al

6862

identification of the targets of BAY 43-9006, the molecular events following inhibition of these targets in cancer cells remain unknown. The antiapoptotic Bcl-2 family member Mcl-1 has been implicated in resistance to anticancer drugs in a number of settings (Craig, 2002). For example, Mcl-1 upregulation occurs in 50% of acute leukemias at the time of recurrence (Kaufmann et al., 1998). Likewise, Mcl-1 overexpression has been associated with a poor clinical response in chronic lymphocytic leukemia (CLL) (Kitada et al., 1998). Additional studies have shown that Mcl-1 confers resistance to apoptosis induced by cytotoxic agents, c-myc overexpression and growth factor withdrawal in CLL cells (Moshynska et al., 2004). More recently, Mcl-1 overexpression has been shown to mediate resistance to TRAIL-induced apoptosis in cholangiocarcinoma cell lines as well (Taniai et al., 2004). Recent data have suggested that Mcl-1 stability depends on ERK-mediated phosphorylation (Domina et al., 2000, 2004; Milella et al., 2001; Meng et al., 2003). In view of these results, the objective of this study was to determine whether the apoptotic effects of BAY 43-9006 were mediated through Mcl-1 downregulation. We not only report here that Mcl-1 is selectively downregulated after BAY 43-9006 treatment in a wide variety of cancer cell lines, but also demonstrate that forced Mcl-1 overexpression protects cells from BAY 43-9006-induced apoptosis, whereas Mcl-1 downregulation by shRNA enhances sensitivity to BAY 43-9006. Based on these results, we describe strategies employing BAY

a

43-9006 to enhance the cytotoxicity of other proapoptotic agents.

Results Effects of BAY 43-9006 on Mcl-1 expression Based on previous studies demonstrating that inhibition of MEK1 can affect expression of Bcl-2 family members (Boucher et al., 2000; Milella et al., 2001; Meng et al., 2003), the effect of BAY 43-9006 on expression levels of Bcl-2 family members and other antiapoptotic proteins in A549 cells was examined by immunoblotting. After a 24 h drug exposure, levels of Bcl-2, Bcl-xL, Bax, Bak, Bid and Bim did not change appreciably. Only Mcl-1 levels were decreased (Figure 1a). To further explore BAY 439006-induced Mcl-1 downregulation, A549 cells were exposed for varying lengths of time as indicated in Figure 1b. Mcl-1 downregulation was noted as early as 3 h after addition of 5.0 mM BAY 43-9006 or 12 h after addition of 2.5 mM BAY 43-9006, indicating that Mcl-1 is selectively downregulated by BAY 43-9006 in a timeand dose-dependent manner. To rule out the possibility that BAY 43-9006-induced Mcl-1 downregulation was cell line-specific, parallel studies were performed in additional transformed human cell lines. As was the case in A549 cells, exposure to 5.0 mM BAY 43-9006 for 6 h induced Mcl-1 downregulation in the ACHN renal cell carcinoma, HT-29 colon cancer, MDA-MB-231 breast cancer, KMCH

b

BAY (µM)

0

1.25 2.5

5.0

BAY (µM)

0

1.25 2.5

5.0

15 s

Bcl-2 Bcl-xL

30 m

Mcl-1

1h

XIAP

3h

Bid 6h Bax 12 h Bak 24 h

Bim

actin

Actin

c

A549

BAY (µM) 0

5.0

ACHN

HT-29 MDA-MB-231

KMCH

Jurkat

K562

MEC-2

0

0

0

0

0

0

5.0

5.0

0

5.0

5.0

5.0

5.0

5.0 Mcl-1 Actin

Figure 1 Effects of BAY 43-9006 on Mcl-1 and other Bcl-2 members. (a) After A549 cells were treated with the indicated concentration of BAY 43-9006 for 24 h, levels of Bcl-2, Bcl-xL, Mcl-1, Bid, Bax, Bak, Bim and XIAP in whole-cell extracts were examined by immunoblotting. (b) Following treatment of A549 cells with BAY 43-9006 for the indicated length of time, Mcl-1 levels were examined in whole-cell lysates by immunoblotting. (c) After the indicated cell lines were exposed to 5 mM BAY 43-9006 for 6 h, Mcl-1 content of whole cell lysates was examined by immunoblotting. In each panel, actin served as a loading control. BAY ¼ BAY 439006 Oncogene

BAY 43-9006-induced Mcl-1 downregulation and apoptosis C Yu et al

6863

BAY 43-9006 was developed as a Raf kinase inhibitor to block signaling through the mitogen-activated protein kinase (MAPK) pathway (Wilhelm et al., 2004). Previously published data also indicated that the MAPK pathway regulates Mcl-1 levels post-translationally (Domina et al., 2000, 2004; Milella et al., 2001; Meng et al., 2003). To assess the role of MAPK signaling in Mcl-1 downregulation, A549 cells were treated for 12 h with 5.0 mM BAY 43-9006 in the presence or absence of U0126, a selective MEK1/2 inhibitor. Even though U0126 inhibited ERK1/2 phosphorylation (Figure 3a), it did not affect Mcl-1 levels over the time course of the experiment. A similar dissociation between downregulation of Mcl-1 levels by BAY 43-9006 and inhibition of ERK phosphorylation was observed in MEC-2 cells (Figure 3b). After BAY 43-9006 treatment for 24 h, dose-dependent downregulation of Mcl-1 was observed, with modest Mcl-1 downregulation after exposure to 2.5 mM BAY 43-9006 and almost complete downregulation at 5.0 mM. Under these conditions, however, phospho-Erk1/2 levels were not altered appreciably (Figure 3b). These findings are consistent with a

0 1.25 2.5 5.0

0 1.25 2.5 5.0 GAPDH

Mcl-1

b

A549

MDA

HT29

K562

M GAPDH

Mcl-1

c

CHX Time (h)

0

1

CHX+BAY

2

3

0

1

2

3 Mcl-1 Actin

d

100

CHX CHX+BAY

75 50 25 0

0

1

2

3

e

MG+BAY

Time (h)

MG-132

BAY 43-9006-induced Mcl-1 downregulation is independent of MAPK signaling in A549 cells

MDA-MB-231

BAY

To determine whether Mcl-1 was downregulated at the transcriptional level, as is the case with the cyclindependent kinase inhibitor flavopiridol (Gojo et al., 2002), Mcl-1 mRNA levels were examined by RT–PCR in a number of cell lines exposed to BAY 43-9006. Our results indicate that Mcl-1 mRNA levels were not altered in two cell lines treated with various concentrations of BAY 43-9006 for 3 h (Figure 2a) or in four additional cell lines treated with 5.0 mM BAY 43-9006 for 6 h (Figure 2b). To further examine the effects of BAY 43-9006 on Mcl-1 downregulation, protein synthesis in K562 cells was blocked by cycloheximide; and the levels of Mcl-1 were examined after 1, 2 or 3 h in the absence or presence of BAY 43-9006. This analysis demonstrated that BAY 43-9006 significantly accelerated Mcl-1 degradation (Po0.01) (Figure 2c, d). Taken together, these results suggest that BAY 43-9006induced Mcl-1 downregulation is a post-translational event. Additional experiments demonstrated that BAY 43-9006 failed to downregulate Mcl-1 in the presence of the proteasome inhibitor MG-132 (Figure 2e), providing further support for the view that BAY 43-9006 is affecting Mcl-1 turnover.

BAY (µM) M

A549

con

BAY 43-9006-induced Mcl-1 downregulation involves protein destabilization

a

Signal (% of Control)

cholangiocarcinoma, Jurkat acute T-cell leukemia, K562 chronic myelogenous leukemia, and MEC-2 chronic lymphocytic leukemia cell lines (Figure 1c). Equivalent results were obtained as early as 3 h after drug addition and persisted for at least 24 h (data not shown). These data indicate that BAY 43-9006-induced Mcl-1 downregulation is not restricted to a few select cell lines.

Mcl-1 Actin Figure 2 Mcl-1 downregulation is a post-translational event. (a) After the indicated cell lines were exposed to increasing concentrations of BAY 43-9006 for 3 h, RNA was isolated and subjected to RT–PCR using Mcl-1 and GAPDH primers. M ¼ DNA size markers. (b) Following BAY 43-9006 treatment for 6 h, RT–PCR analysis was conducted using Mcl-1 and GAPDH primers. (c) After K562 cells were treated with 100 mM cycloheximide in the absence or presence of 5 mM BAY 43-9006 for the indicated length of time, whole-cell lysates were probed for Mcl-1. (d) Results of densitometric analysis of immunoblot bands from Figure 2c and duplicate experiments. Untreated samples were set at 100%. (e) After K562 cells were treated with 5 mM BAY 43-9006 in the absence or presence of 250 nM MG-132 for 24 h, Western analysis was performed to measure Mcl-1 levels

previous report that BAY 43-9006 did not impact the RAF/MEK/ERK pathway in these cells (Wilhelm et al., 2004). To further address the relationship between MEK inhibition and Mcl-1 downregulation, A549 cells were transfected with GFP-tagged constitutively active MEK1. Although this construct enhanced ERK phosphorylation, Oncogene

BAY 43-9006-induced Mcl-1 downregulation and apoptosis C Yu et al

Mcl-1

cyto c actin

50

S-100

75

BOC+BAY

100

BAY

% Apoptotic Cells

p-Erk1/2

b control

a

U0+BAY

U0126

BAY

a

control

6864

cleaved caspase 3

25 caspase 8

Actin

0

BAY (µM) 0

0.05 0.10 0.25 0.50 1.0

2.5

5.0 p-Erk1/2 Mcl-1 Actin

Y +B A C

Mcl-1 Actin

Figure 4 Effect of BOC-D-fmk on BAY 43-9006-induced Mcl-1 downregulation and apoptosis. (a) K562 cells were exposed to 5 mM BAY 43-9006 in the presence or absence of 50 mM BOC-D-fmk for 24 h, after which the percentage of cells exhibiting apoptotic morphology was evaluated using Wright-Giemsa staining. (b) After cells were treated with BAY 43-9006 in the absence or presence of BOC-D-fmk as described in panel a, 100 000g supernatants (S-100 fractions) or whole cell lysates were prepared for immunoblotting with reagents that recognize the indicated antigens. cyto c ¼ cytochrome c. CF ¼ cleaved fragment

BAY

GFP/MEK1 con

BAY

GFP con

c

Y BO

co

b

BA

nt

ro

l

CF

p-Erk1/2 Mcl-1 Actin Figure 3 BAY 43-9006-induced Mcl-1 downregulation occurs independent of MAPK inhibition. (a) A549 cells were exposed to 5 mM BAY 43-9006 in the presence or absence of 20 mM U0126 for 12 h. Whole-cell lysates were then probed with antibodies that recognize phospho-Erk1/2 and Mcl-1. (b) After MEC-2 cells were exposed to the indicated concentrations of BAY 43-9006 for 24 h, whole cell lysates were blotted was described in panel a. (c) A549 cells were transiently transfected with plasmids encoding GFP or a constitutively active MEK1 construct fused to GFP, incubated overnight to allow transgene expression, and then sorted based on GFP fluorescence. GFP-positive cells were treated with 5 mM BAY 43-9006 for 12 h and harvested for analysis of phospho-Erk1/2 and Mcl-1 by immunoblotting

it did not appreciably inhibit BAY 43-9006-induced Mcl-1 downregulation (Figure 3c). Collectively, these results suggest that BAY 43-9006 downregulates Mcl-1 through a post-translational process that involves a pathway distinct from MEK/ERK signaling. BAY 43-9006-induced Mcl-1 downregulation is independent of caspase activation Previous studies have also indicated that Mcl-1 can be cleaved by activated caspases (Karasarides et al., 2004). To rule out the possibility that Mcl-1 downregulation was caspase-mediated, K562 cells were treated with 5.0 mM BAY 43-9006 for 24 h in the presence or absence of the broad-spectrum caspase inhibitor BOC-D-fmk. Even though BAY 43-9006-induced apoptosis was completely blocked by BOC-D-fmk (Figure 4a), Mcl-1 levels were not appreciably restored (Figure 4b), ruling out the possibility that Mcl-1 downregulation merely reflects caspase activation. Oncogene

BAY 43-9006-induced Mcl-1 downregulation correlates with subsequent induction of apoptosis Based on recent experiments showing that Mcl-1 plays a critical role in the survival of Bcr/abl-expressing cells (Aichberger et al., 2005), the induction of apoptosis in K562 cells after Mcl-1 downregulation shown in Figure 4a is perhaps not surprising. In an effort to determine whether Mcl-1 downregulation is accompanied by induction of apoptosis in the other cell lines, various cell lines were exposed to BAY 43-9006 and examined for morphologic evidence of apoptosis. In A549 cells, apoptotic morphological changes first became evident after a 4- to 5-day exposure to 5.0 mM BAY 43-9006 (Figure 5a). Importantly, these changes were not evident at earlier time points even though Mcl1 downregulation had occurred (Figure 1b), again arguing against the possibility that caspase activation contributes to Mcl-1 downregulation. In addition, apoptosis became prominent only after treatment with 5.0 mM BAY 43-9006 (Figure 5a) and not after exposure to lower doses that were less effective in downregulating Mcl-1 (Figure 1a, b), suggesting a correlation between Mcl-1 downregulation and subsequent induction of apoptosis. Parallel experiments performed in MDAMB-231 (Figure 5b) and KMCH cells (Figure 5c) likewise demonstrated that BAY 43-9006 concentrations capable of downregulating Mcl-1 (Figure 1c) were also required for the induction of apoptosis (Figure 5b, c), which was again delayed by several days relative to the time course of Mcl-1 downregulation. Mcl-1 protects cells from BAY 43-9006-induced apoptosis The preceding results suggest a model in which BAY 439006 induces apoptosis by causing Mcl-1 downregulation. If this model is correct, it leads to a number of predictions.

BAY 43-9006-induced Mcl-1 downregulation and apoptosis C Yu et al

6865

a

A549 day 1

100 % Apoptotic Cells

day 2 day 3

75

day 4 day 5 50

25

0 0.0

b

10.0

MDA-MB-231

100

% Apoptotic Cells

2.5 5.0 7.5 [BAY 43-9006] (µM)

BAY 43-9006 synergistically enhances adaphostin-induced apoptosis

75

50

25

0 0.0

% Apoptotic Cells

c

expressing low Mcl-1 levels (Po0.001). In addition, immunoblotting revealed that Mcl-1 overexpression attenuated BAY 43-9006-induced cytochrome c release into the cytosol and caspase 9 cleavage (Figure 6b). Similar results were obtained in a second Mcl-1overexpressing clone (C Yu, unpublished observations). Conversely, this model predicts that Mcl-1 downregulation, by providing a lower baseline of Mcl-1, should sensitize cells to BAY 43-9006. To test this hypothesis, the effects of BAY 43-9006 on parental KMCH cells and a clone stably transfected with Mcl-1 shRNA (Taniai et al., 2004) were compared. Consistent with the prediction, Mcl-1 downregulation enhanced BAY 43-9006-induced apoptosis in this model system (Figure 6c, d), a finding also reflected in the increased cleavage of caspase 9 and PARP (Figure 6e).

100

2.5 5.0 7.5 [BAY 43-9006] (µM)

10.0

KMCH

75

50

Since Mcl-1 is thought to protect cells from a variety of proapoptotic stimuli that activate the mitochondrial apoptotic pathway (Craig, 2002), the observation that BAY 43-9006 downregulates Mcl-1 suggests that BAY 43-9006 should sensitize cells to other stimuli that activate this pathway. To test this possibility, we chose adaphostin, an agent that triggers apoptosis through the mitochondrial pathway (Svingen et al., 2000) by upregulating ROS (Chandra et al., 2003; Yu et al., 2004), as a prototypic proapoptotic stimulus. K562 cells were exposed to BAY 43-9006 alone, adaphostin alone, or the combination. As indicated in Figure 7a and b, BAY 43-9006 dramatically enhanced the ability of adaphostin to induce apoptosis. When the data were analysed by the median effect method, the combination index was o1 over a wide concentration range, indicating a synergistic interaction between BAY 439006 and adaphostin.

25

0 0.0

Discussion 2.5 5.0 7.5 [BAY 43-9006] (µM)

10.0

Figure 5 BAY 43-9006 induces apoptosis in multiple cell lines. MDA-MB-231 (a), A549 (b) or KMCH cells (c) were exposed to the indicated concentrations of BAY 43-9006 for 1, 2, 3, 4 and 5 days. At each time point, nonadherent and adherent cells were pooled, stained with DAPI, and evaluated for apoptotic morphological changes

First, treatments that upregulate Mcl-1 should protect against BAY 43-9006-induced apoptosis. To test this prediction, K562 cells were stably transfected with Mcl1; and multiple clones expressing various Mcl-1 levels were isolated. As shown in Figure 6a, Mcl-1 overexpression protected cells from BAY 43-9006-induced apoptosis. Only 20–30% of the K562 cells expressing high levels of Mcl-1 became apoptotic after BAY 439006 treatment, compared to 60–70% of the cells

BAY 43-9006 is the first orally bioavailable multikinase inhibitor that targets Raf kinase to undergo clinical testing. The beneficial clinical effect of treatment with BAY 43-9006 has predominantly been disease stabilization in a wide variety of tumor types. This effect is consistent with the preclinical data that BAY 43-9006 exerts antitumor effects by both directly inhibiting tumor cell proliferation and by inhibiting tumor angiogenesis. Objective evidence of tumor shrinkage has been seen in renal cell carcinoma (Ahmad and Eisen, 2004), melanoma (Flaherty et al., 2004), hepatocellular (Strumberg et al., 2004), pancreatic, and papillary thyroid carcinoma (Ratain et al., 2004). The mechanism of tumor response to BAY 43-9006 is undefined but may include a proapoptotic component. Results of the present study not only highlight the importance of Mcl-1 downregulation in the proapoptotic effects of BAY 43-9006, but also suggest that BAY 43-9006 might Oncogene

BAY 43-9006-induced Mcl-1 downregulation and apoptosis C Yu et al

6866 2

b

Mcl-1-low Mcl-1-high

Mcl-1

Mcl-1 p-Erk1/2

50

caspase 9 CF

*

25

Actin cyto c

c

1

Y

e

d % Apoptotic Cells

Actin

50

KMCH Mcl-1-sh-RNA-T3

25

0 0

Actin

BA

ol co

nt r

Y

2 Mcl-1

75 % Apoptotic Cells

BA

co

nt r

ol

0

1 2 3 5µM BAY (day)

4

S-100

% Apoptosis Cells

con BAY con BAY

Actin

75

Mcl-1-low Mcl-1-high

cell lysate

1

a

100

KMCH

KMCH Mcl-1 sh-RNA

Mcl-1 shRNA

con BAY con BAY 75

caspase 9 CF

50

PARP CF

25

Actin 0 0.0

2.5

5.0

7.5

10.0

[BAY] (µM)

Figure 6 Effect of Mcl-1 up- or downregulation on BAY 43-9006-induced apoptosis. (a) K562 cells were stably transfected with pcDNA3.1 encoding human Mcl-1 cDNA. Clones with high Mcl-1 expression, designated Mcl-1-high (panel a inset, lane 2) and clones with Mcl-1 expression indistinguishable from parental cells, designated Mcl-1-low (lane 1), were exposed to 5.0 mM BAY 43-9006 for 50 h, stained with Wright-Giemsa, and examined for apoptotic morphological changes. * indicates Po0.001. (b) After Mcl-1-low and Mcl-1-high cells were treated with BAY 43-9006 as described in panel a , control (con) ; and BAY 43-9006(BAY) – treated whole cell lysates were probed for Mcl-1, phospho-Erk1/2 and caspase 9. In addition, 100 000 g supernatants were probed for cytochrome c (cyto c) and actin. CF ¼ cleaved fragment. (c, d) a KMCH clone stably transfected with an shRNA expression vector targeting Mcl-1 (Mcl-1shRNA), named Mcl-1-shRNA-T3 (panel c inset, lane 2), and parental cells (panel c inset, lane 1) were exposed to 5.0 mM BAY 43-9006 for the indicated length of time (panel c) or to the indicated concentration of BAY 43-9006 for 4 days (panel d), stained with DAPI, and examined for apoptotic morphological changes. (e) After KMCH and Mcl-1 sh-RNA cells were treated with 5 mM BAY 43-9006 for 4 days, Western analysis of caspase 9 and PARP from whole-cell lysates was performed

be exerting this effect through a pathway that is distinct from the canonical MEK/ERK pathway. Previous studies have demonstrated that MEK inhibitors can affect the expression and stability of a number of antiapoptotic Bcl-2 family members (Boucher et al., 2000; Milella et al., 2001; Meng et al., 2003; Domina et al., 2004). Based on these results, we have examined the effects of BAY 43-9006 on a number of Bcl-2 family members and identified Mcl-1 as the only Bcl-2 family member that is downregulated. BAY 439006-induced downregulation of Mcl-1 is detectable in a wide range of cultured human cell lines of different histological origins and with different molecular characteristics (Figure 1). The observation that Mcl-1 upregulation inhibits BAY 43-9006-induced apoptosis (Figure 6a) and Mcl-1 downregulation facilitates it (Figure 6c, d) provides further evidence for the importance of Mcl-1 downregulation in the proapoptotic effects of BAY 43-9006. At the present time, the mechanism underlying BAY 43-9006-induced Mcl-1 downregulation is incompletely understood. This downregulation occurs without any change in Mcl-1 message levels (Figure 2a, b), a feature that distinguishes the effect of BAY 43-9006 from that of flavopiridol (Gojo et al., 2002). Instead, BAY Oncogene

43-9006-induces a decrease in Mcl-1 stability (Figure 2c, d) reminiscent of that reported after treatment with MEK inhibitors (Milella et al., 2001; Meng et al., 2003; Domina et al., 2004). Several additional observations, however, distinguish the effects of BAY 43-9006 from those of the MEK inhibitors. First, the effect of BAY 43-9006 on Mcl-1 levels is more rapid than that observed previously with MEK inhibitors (e.g., effects in Jurkat cells at 6 h in Figure 1c vs 24–48 h in Meng et al., 2003). Second, the effect of BAY 43-9006 at these early time points is not reproduced by the MEK1/2 inhibitor U0126 (Figure 3a). Third, BAY 43-9006 is able to diminish Mcl-1 under conditions that do not affect ERK phosphorylation (Figure 3a, c). While these observations seem to rule out the dependence of Mcl-1 downregulation on the MEK/ERK pathway, it is still possible that a Raf-dependent and MAPK-independent pathway might play a role. Alternatively, because BAY 43-9006 has significant inhibitory activity against several receptor tyrosine kinases involved in neo-vascularization and tumor progression, including VEGFR-2, VEGFR-3, PDGFRb, Flt-3, and c-Kit, it is possible that interruption of signaling downstream of one of these targets, but independent of Raf, contributes to the Mcl-1

BAY 43-9006-induced Mcl-1 downregulation and apoptosis C Yu et al

6867

a

BAY BAY+AD-1.25µM

% Apoptotic Cells

100 75 50 25 0 0

1

2 3 [BAY] (µM)

0

0.2

0.4 0.6 Fractional Effect

4

5

b

CI

1.5 1.0 0.5 0 0.8

1.0

Figure 7 BAY 43-9006 sensitizes cells to the proapoptotic effects of adaphostin. (a) After exposure for 24 h to the indicated concentration of BAY 43-9006 alone, 1.25 mM adaphostin alone, or BAY 43-9006 þ 1.25 mM adaphostin, K562 cells were stained with Wright-Giemsa and examined for apoptotic morphological changes. BAY ¼ BAY 43-9006, AD ¼ adaphostin. (b), K562 cells were treated with various concentrations of BAY 43-9006 (1.2– 2.4 mM) and/or adaphostin (0.75–1.5 mM) at a fixed ratio of 1.6 : 1. After the induction of apoptosis was quantitated, the combination index was calculated as described in the Materials and methods. A combination index o1 represents synergism

downregulation observed in the present study. Consistent with this latter possibility, a recent study has demonstrated that Mcl-1 levels diminish due to enhanced protein turnover when Akt signaling is disrupted downstream of the interleukin-6 receptor (Kobayashi et al., 2005). Since Mcl-1 is phosphorylated on a number of different residues in response to a variety of stimuli (Domina et al., 2000, 2004; Inoshita et al., 2002), turnover of this short-lived antiapoptotic protein is likely subject to regulation by a number of different signaling pathways. To the extent that BAY 43-9006 inhibits multiple kinases, Mcl-1 downregulation might even reflect the ability of BAY 43-9006 to target more than one pathway affecting Mcl-1 stability. The present studies also demonstrated that BAY 439006 induces apoptosis in a wide variety of transformed cell lines. This result is consistent with that reported in melanoma cell lines (Karasarides et al., 2004). Among the apoptotic changes observed after BAY 43-9006 treatment were mitochondrial release of cytochrome c and cleavage of caspases to their active fragments (Figure 4b and 6b) as well as apoptotic morphological changes. The observation that apoptotic changes only occur after treatment with BAY 43-9006 concentrations

that downregulate Mcl-1 (cf. Figures 1 and 5) further highlights the importance of Mcl-1 downregulation in this process. Interestingly, this Mcl-1 downregulation occurs very rapidly (evident at 3 h), but apoptosis only becomes evident after 1 (Figures 4 and 7) to 4 days (Figure 5a). A similar delay between Mcl-1 downregulation and subsequent cell death has previously been reported in murine lymphohematopoietic tissues following conditional deletion of the Mcl-1 gene in vivo (Opferman et al., 2003, 2005). Collectively, these observations raise the possibility that Mcl-1 downregulation contributes to apoptosis but might not be sufficient for apoptosis induction. Instead, it is possible that additional events (e.g., damage from endogenous ROS) serve as a trigger for apoptosis in cells in which Mcl-1 has been downregulated. To test this hypothesis, cells were treated with adaphostin, an agent that has recently been shown to upregulate ROS (Chandra et al., 2003; Yu et al., 2004). The observation that these two agents synergize (Figure 7) is certainly consistent with the possibility that BAY 43-9006-induced Mcl-1 downregulation might sensitize cells to endogenous triggers of apoptosis. Moreover, these findings suggest that BAY 43-9006 might also sensitize cells to other agents that trigger apoptosis through the mitochondrial pathway. In summary, the present studies have demonstrated that BAY 43-9006 induces apoptosis through selective destabilization and downregulation of Mcl-1. These observations simultaneously provide a potentially useful biomarker of BAY 43-9006 that can be explored in future clinical trials as well as a plausible explanation for the ability of this agent to shrink some tumors. Further preclinical and possible clinical studies are required to determine whether BAY 43-9006 would be efficacious in tumors in which Mcl-1 accumulation causes resistance to therapy, such as CLL, relapsed acute leukemia and cholangiocarcinoma. Likewise, further studies are required to determine whether the Mcl-1 downregulation translates into an ability of BAY 43-9006 to synergize with a broader range of chemotherapeutic agents.

Materials and methods Materials BAY 43-9006 and adaphostin were kindly provided by Bayer Pharmaceutical Corporation (West Haven, CT, USA) and the Drug Synthesis and Chemistry Branch, Division of Cancer Treatment and Diagnosis, National Cancer Institute (Bethesda, MD, USA), respectively. BOC-D-fmk, U0126 and cycloheximide were purchased from Enzyme System Products (Livermore, CA, USA), Calbiochem (San Diego, CA, USA), and Sigma Chemical Co. (St Louis, MO, USA). Cells A549, ACHN, HT-29, MDA-MB-231, Jurkat and K562 cell lines were purchased from American Type Culture Collection, Rockville, MD, USA. MEC-2 cells were kindly supplied by Neil Kay (Mayo Clinic, Rochester, MN, USA). All cells were grown in the following media containing 100 mg/ml penicillin Oncogene

BAY 43-9006-induced Mcl-1 downregulation and apoptosis C Yu et al

6868 G, 100 mg/ml streptomycin, and 2 mM glutamine: A549 in RPMI 1640 supplemented with 5% (v/v) FBS; ACHN in minimum essential medium containing Earl’s salts with 10% FBS, 1.0 mM sodium pyruvate, 0.1 mM nonessential amino acids, and 1.5 g/l sodium bicarbonate; HT-29 in McCoy’s 5A with 10% FBS; Jurkat and K562 in RPMI 1640 with 10% FBS; MDA-MB-231 in IMEM with 10% FBS; KMCH in DMEM with 10% FBS and (g) MEC-2 in IMDM with 10% FBS. KMCH cells and their derivative Mcl-1-sh-RNA-T3, which was stably transfected with Mcl-1 shRNA, were propagated as previously described (Taniai et al., 2004).

gel containing 0.5 mg/ml EB (ethidium bromide) in 1  TAE buffer (30.7 mM Tris, 20 mM sodium acetate, and 1 mM EDTA), and visualized on a UV transilluminator. Transient transfection

Subconfluent, logarithmically growing cells were placed in sterile plastic T-flasks, allowed to adhere, supplemented with the designated drugs, and incubated in a humidified incubator with 5% CO2 at 371C for various intervals as indicated.

Constitutively active MEK1 (S218D/S222D) in pUSEamp (Clontech Laboratories) was subcloned in-frame into the pEGFP plasmid (Upstate Biotechnology) using standard techniques. The entire MEK1/GFP cDNA in the fusion construct was sequenced, and the reading frame was confirmed. Log-phase A549 cells were transfected with a pEGFP empty vector or a pEGFP/MEK1 fusion plasmid using Lipofectamine Plus (Invitrogen) following the supplier’s instructions. After 24-h incubation, 50–60% of the cells displayed green fluorescence. The brightest 30–40% of the total cell population was isolated by fluorescence-activated cell sorting and analysed as described in the text.

Immunoblot analysis

Generation of a stably transfected cell line

Following drug treatment, cell lysates were prepared and subjected to immunoblot analysis using 30 mg of cellular protein as described previously (Yu et al., 2003). The following primary antibodies were utilized at 1 : 1000 dilution unless otherwise indicated: rabbit polyclonal antibodies to caspase 9, caspase 8, cleaved caspase 3, XIAP, Bax and phospho-ERK1/2 (Thr202/Tyr204), from Cell Signaling Technology (Beverly, MA, USA); murine monoclonal anti-Mcl-1 from PharMingen (San Diego, CA, USA); and rabbit anti-Bcl-xL (1 : 2000) or Bid, goat anti-Bim, and murine monoclonal antibodies to Bcl2 (1 : 2000), cytochrome c (1 : 2000) and Bak from Santa Cruz Biotechnology (Santa Cruz, CA, USA). To ensure equivalent loading and transfer, blots were stripped and reprobed with antiactin antiserum. Western blots are representative of three independent experiments.

K562 cells were transfected with pcDNA3.1 containing human full-length Mcl-1 cDNA (kindly provided by Ruth Craig, Dartmouth Medical School, Hanover, NH, USA) as described previously (Meng et al., 2003). Stably transfected Mcl-1 clones were selected with 800 mg/L G418, cloned by limiting dilution, screened by immunoblotting, and propagated in medium supplemented with 500 mg/L G418.

Detection of apoptotic changes

Statistical analysis

Following drug treatment, the release of cytochrome c from the mitochondria was analysed by a selective digitonin permeabilization method as reported previously (Yu et al., 2003). Changes in nuclear morphology were assessed after Wright-Giemsa or DAPI staining as described previously (Yu et al., 2003; Taniai et al., 2004).

All of the data were expressed as mean7s.d. from three individual experiments. Differences between groups were determined by using the Student’s t-test for unpaired observations. Po0.05 was considered significant.

Cell culture

RT–PCR After drug treatment, RNA was isolated using an RNeasyt mini kit (Qiagen, Valencia, CA, USA). One-twentieth of the cDNA product was used for each amplification reaction. Primers used were as follows. Mcl-1 forward primer 50 CGGTAATCGGACTCAACCTC-30 , reverse primer 50 CCTCCTTCTCCGTAGCCAA-30 . GAPDH forward primer 50 -GGCAAATTCCATGGCACCGTCAGG-30 , and reverse primer 50 -GGAGGCATTGCTGATGA-TCCTGAGG-30 . PCRs were performed using Expandt high-fidelity PCR reagents from Roche Applied Science following the supplier’s instructions. The amplification was followed by 30 cycles with 951C hot start for 4 min, 951C for 30 s, 551C for 45 s, and 721C for 1 min. PCR products were electrophoresed on a 1% agarose

Analysis of combined drug effects The effect of combining BAY 43-9006 and adaphostin was analysed using the median effect method of Chou and Talalay (1984) using Calcusyn software (Biosoft, Oxford, UK) as reported previously (Adjei et al., 2001).

Abbreviations ERK, extracellular signal-regulated kinase; CLL, chronic lymphocytic leukemia; FBS, heat-inactivated fetal bovine serum; GFP, green fluorescent protein; MEK, mitogenactivated and extracellular signal-regulated kinase; ROS, reactive oxygen species; shRNA, short hairpin RNA. Acknowledgements We are grateful to Keith C Bible, Neil E Kay, Paul Haluska, Crescent Isham, Andrea McCollum and Edward Sausville for reagents and/or advice. We thank Mrs Raquel Ostby for expert secretarial assistance. This work was supported in part by a Research Scholar Award from the American Cancer Society (AAA) and grants from the National Institutes of Health to GJG (R01 DK59427) and SHK (R01 CA69008).

References Adjei AA, Davis JN, Bruzak L, Erlichman C and Kaufmann SH. (2001). Clin. Cancer Res., 7, 1438–1445. Ahmad T and Eisen T. (2004). Clin. Cancer Res., 10, 6388S–6392S. Oncogene

Aichberger KJ, Mayerhofer M, Krauth MT, Skvara H, Florian S, Sonneck K, Akgul C, Derdak S, Pickl WF, Wacheck V, Selzer E, Monia BP, Moriggl R, Valent P and Sillaber C. (2005). Blood, 105, 3303–3311.

BAY 43-9006-induced Mcl-1 downregulation and apoptosis C Yu et al

6869 Boucher MJ, Morisset J, Vachon PH, Reed JC, Laine J and Rivard N. (2000). J. Cell. Biochem., 79, 355–369. Chandra J, Hackbarth J, Le S, Loegering D, Bone N, Bruzek LM, Narayanan VL, Kay NE, Tefferi A, Karp JE, Sausville EA and Kaufmann SH. (2003). Blood, 102, 4512–4519. Chou T-C and Talalay P. (1984). Adv. Enzyme Reg., 22, 27–55. Cohen Y, Xing M, Mambo E, Guo Z, Wu G, Trink B, Beller U, Westra WH, Ladenson PW and Sidransky D. (2003). J. Natl. Cancer Inst., 95, 625–627. Craig RW. (2002). Leukemia, 16, 444–454. Davies H, Bignell GR, Cox C, Stephens P, Edkins S, Clegg S, Teague J, Woffendin H, Garnett MJ, Bottomley W, Davis N, Dicks E, Ewing R, Floyd Y, Gray K, Hall S, Hawes R, Hughes J, Kosmidou V, Menzies A, Mould C, Parker A, Stevens C, Watt S, Hooper S, Wilson R, Jayatilake H, Gusterson BA, Cooper C, Shipley J, Hargrave D, PritchardJones K, Maitland N, Chenevix-Trench G, Riggins GJ, Bigner DD, Palmieri G, Cossu A, Flanagan A, Nicholson A, Ho JW, Leung SY, Yuen ST, Weber BL, Seigler HF, Darrow TL, Paterson H, Marais R, Marshall CJ, Wooster R, Stratton MR and Futreal PA. (2002). Nature, 417, 949–954. Dhillon AS, Meikle S, Peyssonnaux C, Grindlay J, Kaiser C, Steen H, Shaw PE, Mischak H, Eychene A and Kolch W. (2003). Mol. Cell. Biol., 23, 1983–1993. Domina AM, Smith JH and Craig RW. (2000). J. Biol. Chem., 275, 21688–21694. Domina AM, Vrana JA, Gregory MA, Hann SR and Craig RW. (2004). Oncogene, 23, 5301–5315. Flaherty KT, Brose M, Schucter L, Tuveson D, Lee R, Schwartz B, Lathia C, Weber B and O’Dwyer P. (2004). Proc. Am. Soc. Clin. Oncol., 23. Gojo I, Zhang B and Fenton RG. (2002). Clin. Cancer Res., 8, 3527–3538. Inoshita S, Takeda K, Hatai T, Terada Y, Sano M, Hata J, Umezawa A and Ichijo H. (2002). J. Biol. Chem., 277, 43730–43734. Karasarides M, Chiloeches A, Hayward R, Niculescu-Duvaz D, Scanlon I, Friedlos F, Ogilvie L, Hedley D, Martin J, Marshall CJ, Springer CJ and Marais R. (2004). Oncogene, 23, 6292–6298. Kaufmann SH, Karp JE, Svingen PA, Krajewski S, Burke PJ, Gore SD and Reed JC. (1998). Blood, 91, 991–1000. Kitada S, Andersen J, Akar S, Zapata JM, Takayama S, Krajewski S, Wang HG, Zhang X, Bullrich F, Croce CM, Rai K, Hines J and Reed JC. (1998). Blood, 91, 3379–3389. Kobayashi S, Werneburg NW, Bronk SF, Kaufmann SH and Gores GJ. (2005). Gastroenterology, in press. Meng XW, Chandra J, Loegering D, Van Becelaere K, Kottke TJ, Gore SD, Karp JE, Sebolt-Leopold J and Kaufmann SH. (2003). J. Biol. Chem., 278, 47326–47339.

Mercer KE and Pritchard CA. (2003). Biochim. Biophys. Acta, 1653, 25–40. Milella M, Kornblau SM, Estrov Z, Carter BZ, Lapillonne H, Harris D, Konopleva M, Zhao S, Estey E and Andreeff M. (2001). J. Clin. Invest., 108, 851–859. Moshynska O, Sankaran K, Pahwa P and Saxena A. (2004). J. Natl. Cancer Inst., 96, 673–682. Opferman JT, Iwasaki H, Ong CC, Suh H, Mizuno S, Akashi K and Korsmeyer SJ. (2005). Science, 307, 1101–1104. Opferman JT, Letai A, Beard C, Sorcinelli MD, Ong CC and Korsmeyer SJ. (2003). Nature, 426, 671–676. Pollock PM and Meltzer PS. (2002). Cancer Cell, 2, 5–7. Ratain MJ, Flaherty KT, Stadler WM, O’Dwyer P, Kaye S, Xiong H, Patnaik A, Gore M, Lee RJ and Eisen T. (2004). Proc. Am. Soc. Clin. Oncol., 22, 382s. Singer G, Oldt III R, Cohen Y, Wang BG, Sidransky D, Kurman RJ and Shih Ie M. (2003). J. Natl. Cancer Inst., 95, 484–486. Strumberg D, Richly H, Hilger RA, Schleucher N, Korfee S, Tewes M, Faghih M, Brendel E, Voliotis D, Haase CG, Schwartz B, Awada A, Voigtmann R, Scheulen ME and Seeber S. (2004). J. Clin. Oncol., 21, 21. Strumberg D, Richly H, Hilger RA, Schleucher N, Korfee S, Tewes M, Faghih M, Brendel E, Voliotis D, Haase CG, Schwartz B, Awada A, Voigtmann R, Scheulen ME and Seeber S. (2005). J. Clin. Oncol., 23, 965–972. Svingen PA, Tefferi A, Kottke TJ, Kaur G, Narayanan VL, Sausville EA and Kaufmann SH. (2000). Clin. Cancer Res., 6, 237–249. Taniai M, Grambihler A, Higuchi H, Werneburg N, Bronk SF, Farrugia DJ, Kaufmann SH and Gores GJ. (2004). Cancer Res., 64, 3517–3524. Tannapfel A, Sommerer F, Benicke M, Katalinic A, Uhlmann D, Witzigmann H, Hauss J and Wittekind C. (2003). Gut, 52, 706–712. Wan PT, Garnett MJ, Roe SM, Lee S, Niculescu-Duvaz D, Good VM, Jones CM, Marshall CJ, Springer CJ, Barford D and Marais R. (2004). Cell, 116, 855–867. Wilhelm SM, Carter C, Tang L, Wilkie D, McNabola A, Rong H, Chen C, Zhang X, Vincent P, McHugh M, Cao Y, Shujath J, Gawlak S, Eveleigh D, Rowley B, Liu L, Adnane L, Lynch M, Auclair D, Taylor I, Gedrich R, Voznesensky A, Riedl B, Post LE, Bollag G and Trail PA. (2004). Cancer Res., 64, 7099–7109. Yu C, Rahmani M, Almenara J, Sausville EA, Dent P and Grant S. (2004). Oncogene, 23, 1364–1376. Yu C, Rahmani M, Conrad D, Subler M, Dent P and Grant S. (2003). Blood, 102, 3765–3774.

Oncogene