JBC Origene Your Gene Company

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M304793200 on September 8, 2003

J. Biol. Chem., Vol. 278, Issue 47, 47326-47339, November 21, 2003
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
278/47/47326    most recent
M304793200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Meng, X. W.
Right arrow Articles by Kaufmann, S. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Meng, X. W.
Right arrow Articles by Kaufmann, S. H.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Central Role of Fas-associated Death Domain Protein in Apoptosis Induction by the Mitogen-activated Protein Kinase Kinase Inhibitor CI-1040 (PD184352) in Acute Lymphocytic Leukemia Cells in Vitro*

Xue Wei Meng, Joya Chandra{ddagger}, David Loegering, Keri Van Becelaere§, Timothy J. Kottke, Steven D. Gore¶, Judith E. Karp¶, Judy Sebolt-Leopold§, and Scott H. Kaufmann||

From the Division of Oncology Research, Mayo Clinic, and Department of Molecular Pharmacology, Mayo Graduate School, Rochester, Minnesota 55905, the §Cancer Molecular Sciences Department, Pfizer Global Research and Development, Ann Arbor, Michigan 48105, and The Sidney Kimmel Johns Hopkins Oncology Center, Baltimore, Maryland 21287

Received for publication, May 7, 2003 , and in revised form, August 15, 2003.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Because the MAPK pathway plays important roles in cell proliferation and inhibition of apoptosis, this pathway has emerged as a potential therapeutic target for solid tumors and leukemia. At the present time there is little information about activation of this pathway and the consequences of its inhibition in acute lymphocytic leukemia cells (ALL). In the present study, constitutive MAPK pathway activation, as evidenced by phosphorylation of ERK1 and ERK2, was observed in 8 of 8 human lymphoid cell lines and 33% (8:24) of pretreatment ALL bone marrows. Inhibition of this pathway by the MEK inhibitors CI-1040 and PD098059 induced apoptosis through a unique pathway involving dephosphorylation and aggregation of Fas-associated death domain protein followed by death receptor-independent caspase-8 activation. Jurkat cell variants lacking Fas-associated death domain protein or procaspase-8 were resistant to CI-1040-induced apoptosis, as were Jurkat or Molt3 cells treated with the O-methyl ester of the caspase-8 inhibitor N-(N{alpha}-benzyloxycarbonylisoleucylglutamyl) aspartate fluoromethyl ketone. In contrast, CI-1040-induced apoptosis was unaffected by blocking anti-Fas antibody, soluble tumor necrosis factor-{alpha}-related apoptosis-inducing ligand decoy receptor, or transfection with cDNA encoding the anti-apoptotic Bcl-2 family member Mcl-1 or dominant negative caspase-9. Collectively, these results identify the MAPK pathway as a potential therapeutic target in ALL and delineate a mechanism by which MEK inhibition triggers apoptosis in ALL cells.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
ERK11 and ERK2 are the final kinases in a signal transduction pathway that involves binding of ligands to cell surface receptors, activation of Ras isoforms, recruitment and activation of Raf kinases, and phosphorylation of the MAPK kinases MEK1 and MEK2 (1–3). A number of cytokines, including steel factor, granulocyte-macrophage colony-stimulating factor, interleukin-3, and interleukin-5, signal through this pathway in hematopoietic cells (4, 5). Once activated by dual phosphorylation on a threonine and nearby tyrosine, the ERKs in turn translocate to the nucleus, where they phosphorylate and activate a number of transcription factors, including Elk-1, c-Jun, and c-Myc (6–8). In addition, activated ERKs phosphorylate and enhance the activity of the anti-apoptotic Bcl-2 family members Bcl-2 and Mcl-1 (9, 10). Collectively, phosphorylation of these substrates enhances proliferation and inhibits apoptosis, providing two alterations that are thought to be critical for neoplastic transformation (11). Consistent with these observations, expression of a constitutively active MEK1 allele is sufficient to transform cells (12, 13).

In view of the involvement of this pathway in both proliferative and anti-apoptotic signaling, there has been considerable interest in exploring the effects of MEK inhibitors in a variety of cell types (1, 2, 14). The aminoflavone derivative PD098059 has been shown to stabilize an inactive MEK conformation, thereby inhibiting the activity of MEK1 and MEK2 in vitro and in vivo (15, 16). As a presumed consequence of this inhibition, PD098059 inhibits proliferation and induces apoptosis in solid tumor cell lines in vitro (1, 17, 18). Additional studies have demonstrated that CI-1040 (previously designated PD184352), an orally bioavailable MEK inhibitor derived from a chemical series structurally distinct from PD098059, inhibits ERK phosphorylation in colon cancer xenografts for up to 12 h after administration to tumor-bearing mice (19). Studies with CI-1040 (19), as well as PD098059 (18), have suggested that the antiproliferative effects of MEK inhibitors are greater in cells with higher levels of MAPK pathway activation.

Previous studies (20–22) have demonstrated constitutive ERK activation in 50–70% of clinical AML specimens. Importantly, ERK phosphorylation correlates with the ability of CI-1040 to induce apoptosis in AML in vitro (22). In AML cell lines, CI-1040-induced apoptosis is preceded by a p27Kip1-mediated G1 arrest and by down-regulation of the anti-apoptotic proteins Mcl-1 and Bcl-xL (22). The latter changes suggest that CI-1040 might induce apoptosis through the mitochondrial pathway, a series of reactions involving cytochrome c translocation to the cytoplasm, ATP-dependent assembly of the procaspase-9/Apaf-1 holoenzyme, and subsequent proteolytic activation of caspases-3, -6, and -7 (reviewed in Refs. 23 and 24). Potential involvement of other caspases in MEK inhibitor-induced killing has not been examined previously.

Because the number of ALL specimens examined previously was limited, less is known about MAPK pathway activation in this disorder. ALL occurs about 1/3 as often as AML in adults and remains a difficult malignancy to treat. Most adult ALL patients respond to initial therapy but then relapse and die of this disease (25–27). Accordingly, there is considerable interest in identifying mechanisms of drug resistance and developing new therapies for this disorder.

In the present study, we examined the biochemical consequences of MAPK inhibition in ALL. In particular, in view of data presented below showing constitutive ERK phosphorylation in clinical ALL samples, we examined the effect of the MEK inhibitor CI-1040 on cell cycle progression and survival in lymphoid cell lines. Results of these studies differed substantially from previously reported effects of MEK inhibition in solid tumor and AML cell lines. In particular, CI-1040 induced S phase arrest in 6 of 8 lymphoid cell lines examined. This was followed by FADD dephosphorylation and aggregation, which led to receptor-independent caspase-8 activation. Collectively, these observations indicate that the MAPK pathway is frequently activated in ALL cells and identify a novel mechanism by which MEK inhibitors can kill these cells.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Materials—The broad spectrum caspase inhibitor ZVAD(OMe)-fmk (28) and caspase-8-selective inhibitor IETD(OMe)-fmk were from Enzyme Systems Products (Dublin, CA). The caspase substrate DEVDAFC and the MEK1/2 inhibitor PD098059 were purchased from Biomol (Plymouth Meeting, PA) and Alexis (San Diego, CA), respectively. Other reagents were obtained as described (29).

Rabbit sera that specifically recognize the phosphorylated threonine and tyrosine residues of ERK as well as separate sera that recognize ERK irrespective of phosphorylation state were from Promega (Madison, WI) or Cell Signaling Technology (Beverly, MA). Additional antibodies were obtained from the following suppliers: murine monoclonal antibodies that recognize XIAP, Mcl-1, Bcl-xL, FADD, and caspases-2, -3, and -8 from BD Biosciences; monoclonal anti-Fas (Apo-1–1) from Alexis (San Diego, CA); monoclonal anti-Bcl-2 from Dako (Carpenteria, CA); rabbit anti-Bax and monoclonal anti-Cdc25a from Santa Cruz Biotechnology (Santa Cruz, CA); monoclonal anti-Myc from Covance (Richmond, CA); rabbit anti-Bak and murine anti-Fas (CH-11 agonistic antibody and ZB4 blocking antibody) from Upstate Biotechnology (Lake Placid, NY); and rabbit anti-cleaved poly(ADP-ribose) polymerase (PARP) from Promega (Madison, WI). Murine monoclonal antibodies that recognize PARP and HSP90 were gifts from Dr. G. Poirier (Laval University, Ste-Foy, Quebec, Canada) and David Toft (Mayo Clinic, Rochester, MN), respectively. Rabbit antisera that recognize neoepitopes at the C termini of the caspase-3, -8, and -9 large subunits were characterized previously (29, 30). Peroxidase-coupled and phycoerythrin-coupled secondary antibodies were obtained from Kirkegaard & Perry (Gaithersburg, MD) and Southern Biotechnology Associates (Birmingham, AL), respectively.

Tissue Culture, Drug Treatment, and Analysis—Cell lines were obtained from the following sources: Jurkat (T cell ALL) and JM14A5 (a Jurkat clone lacking Fas expression (31)) from Paul Leibson (Mayo Clinic); SKW6.4 (B lineage), CEM (T cell ALL), H9 (Sezary syndrome), Molt3 (T cell ALL), I2.1 (a Jurkat clone lacking FADD), and I9.2 (a Jurkat clone lacking procaspase-8) from American Type Culture Collection (Manassas, VA); BJAB (Burkitt's lymphoma) from Marcus Peter (University of Chicago, Chicago); JB-6 (a Jurkat derivative lacking procaspase-8 (32)) from Shigekazu Nagata (Kyoto University, Kyoto, Japan); and Daudi and Raji (both Burkitt's lymphoma) from Adele Fielding (Mayo Clinic). All lines were maintained at <=1.0 x 106 cells/ml in RPMI 1640 medium containing 100 units/ml penicillin G, 100 µg/ml streptomycin, 2 mM glutamine, and 15 (H9, I2.1, I9.2, JB-6) or 10% heat-inactivated fetal bovine serum.

Unless otherwise indicated, log phase cells were treated for the indicated lengths of time with 10 µM CI-1040, a concentration chosen to approximate serum levels of the drug and its active metabolite observed in a phase I clinical trial of CI-1040 in solid tumor patients (33). After treatment, cells were harvested for assessment of cell cycle distribution and apoptosis.

To determine cell cycle distribution, cells were sedimented at 200 x g for 10 min, washed in ice-cold calcium- and magnesium-free Dulbecco's phosphate-buffered saline, fixed in 50% ethanol, treated with RNase A, and stained with 50 µg/ml propidium iodide in 0.1% sodium citrate (34). After 20,000 cells were analyzed on a FACScan flow cytometer using an excitation wavelength of 488 nm and an emission wavelength of 617 nm, data were analyzed using ModFit software (BD Biosciences).

Apoptosis was assessed by using one of two techniques. For morphological analysis, cells were sedimented at 200 x g for 10 min, fixed in 3:1 (v/v) methanol/acetic acid, stained with 1 µg/ml Hoechst 33258 in 50% (v/v) glycerol, and examined under epi-illumination using a Zeiss Axioplan microscope. At least 300 cells/sample were scored for apoptotic changes (peripheral chromatin condensation or nuclear fragmentation) as illustrated previously (35). Alternatively, the sedimented cells were subjected to flow microfluorimetry essentially as described above. After 10,000 events were collected, data were analyzed for the percentage of subdiploid cells using CellQuest software (Verity Software House, Topsham, ME). In experiments where both methods of quantitating apoptotic cells were employed, the results agreed within 3–5%.

Samples for immunoblotting were treated with CI-1040 for the indicated length of time. For whole cell lysates, cells were sedimented at 200 x g for 10 min, washed in buffer A (RPMI 1640 medium containing 10 mM HEPES (pH 7.4 at 4 °C)), lysed in alkylation buffer (6 M guanidine hydrochloride, 250 mM Tris-HCl (pH 8.5 at 21 °C), and 10 mM EDTA supplemented before use with 1 mM {alpha}-phenylmethylsulfonyl fluoride and 150 mM {beta}-mercaptoethanol), and prepared for SDS-PAGE as described previously (36). To evaluate FADD solubility, washed cells were extracted at 4 °C for 15 min with DISC buffer (1% (w/v) Triton X-100, 30 mM Tris-HCl (pH 7.5 at 4 °C), 150 mM NaCl, 1% (v/v) glycerol, 1 mM {alpha}-phenylmethylsulfonyl fluoride, 100 mM NaF, 1 mM Na2VO4, 20 nM microcystin, and 10 µg/ml leupeptin and pepstatin A) and sedimented at 15,000 x g for 15 min. Aliquots containing 50 µg of total cellular protein or the Triton-soluble versus -insoluble fractions were separated on SDS-polyacrylamide gels and probed with antibodies as described above. To assess caspase activation, cytosol (100,000 x g supernatant) was prepared and incubated with DEVD-AFC exactly as described (31).

Analysis of Cell Surface Fas Expression—1 x 106 cells were stained with mouse Apo-1-1 anti-Fas on ice for 45 min. After washing, cells were incubated with phycoerythrin-conjugated anti-mouse IgG for an additional 30 min on ice. Following washing, cells were fixed in 1% paraformaldehyde and stored in the dark at 4 °C until analyzed by flow cytometry.

Transient Transfections—Plasmids encoding EGFP alone (pEGFPN1), a dominant negative caspase-9/EGFP fusion protein, a dominant negative FADD/EGFP fusion protein, or Mcl-1 were obtained from Clontech (Palo Alto, CA), Emad Alnemri (Thomas Jefferson University, Philadelphia), Greg Gores (Mayo Clinic), and Ruth Craig (Dartmouth Medical School. Hanover, NH), respectively. cDNAs encoding full-length FADD or procaspase-8 were, respectively, cloned into pEGFP-N1 or pIRES-Stag-Rad9-EGFP (from Larry Karnitz, Mayo Clinic) (37) after removing Rad9. Plasmids were sequenced to confirm their integrity. Log phase Jurkat, JB-6, or I2.1 cells were transfected in the buffer described by van den Hoff et al. (38) using a T820 square wave electroporator (BTX, San Diego, CA) delivering a 240-V pulse for 10 ms. After a 24-h incubation, 30–45% of the cells displayed green fluorescence. The brightest 10–12% of the total cell population was isolated by fluorescence-activated cell sorting, exposed to drug or diluent as indicated in each figure legend, fixed, and examined for apoptotic morphological changes.

RT-PCR—After log phase Jurkat cells were treated with 10 µM CI-1040 for the indicated length of time, total RNA was isolated (RNeasyTM mini kit, Qiagen, Valencia, CA). cDNAs were synthesized using a SuperscriptTM First-strand Synthesis kit (Invitrogen). One-twentieth of the cDNA product was used for each amplification reaction.

Using the primers described in Table I, PCRs were performed using ExpandTM high fidelity PCR reagents from Roche Applied Science following the supplier's instructions. Following amplification, products were electrophoresed on a 1% agarose gel containing 0.5 µg/ml ethidium bromide in 1x TAE buffer (30.7 mM Tris, 20 mM sodium acetate, and 1 mM EDTA), visualized on a UV transilluminator, and sequenced to confirm their identity.


View this table:
[in this window]
[in a new window]
  		TABLE I
Primers used for RT-PCR

 
Clinical Samples—In conjunction with an institutional review board-approved treatment protocol (39), heparinized bone marrow aspirates were obtained from the posterior iliac crests of newly diagnosed ALL patients prior to the initiation of induction chemotherapy. Within 2 h of aspiration, samples were sedimented on ficoll-Hypaque step gradients (density = 1.077 and 1.119 g/cm3). Cells collected from the upper interface were diluted with buffer A, sedimented at 200 x g for 10 min, and resuspended in buffer A. After aliquots were removed for counting and cytospins, samples were sedimented at 200 x g for 10 min, solubilized in alkylation buffer, prospectively prepared for electrophoresis as described above, and lyophilized in multiple single-use vials (36). Aliquots containing 5 x 105 marrow mononuclear cells were subjected to electrophoresis on gels containing a 5–15% acrylamide gradient. To provide a positive control, HL-60 cells, which are known to have constitutively phosphorylated ERK1 and ERK2 (40), were prepared for electrophoresis in an identical fashion. After SDS-PAGE, samples were transferred to nitrocellulose and probed with primary antibodies followed by horse-radish peroxidase-coupled secondary antibodies as described previously (41). Blots were quantified on a FluorS Max Multimaging System (Bio-Rad) and considered positive for ERK phosphorylation if the integrated phospho-ERK signal was >=20% of the total ERK signal.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
MAPK Pathway Activation in Lymphoid Cell Lines and Clinical ALL—To assess MAPK pathway activation in lymphoid cells, whole cell lysates were subjected to immunoblotting with a rabbit antiserum that specifically recognizes dually phosphorylated ERK1 and ERK2. In contrast to solid tumor cell lines, some of which fail to exhibit constitutive MAPK activation (19), immunoblotting readily demonstrated phosphorylated ERK1 and ERK2 in all lymphoid cell lines examined (Fig. 1A and data not shown). Collectively, these lymphoid lines provide a model system to examine the effect of MEK inhibition in the studies described below.



View larger version (27K):
[in this window]
[in a new window]
  		FIG. 1.
Constitutive MAPK pathway activation in human leukemia cell lines and clinical leukemia specimens. A, after SDS-PAGE and transfer to nitrocellulose, aliquots containing 50 µg of protein from the indicated cell lines were probed with antibodies that recognize phospho-ERK (upper panel). After erasure, blots were probed with antibodies to total ERK (bottom panel). B, samples containing 5 x 105 bone marrow mononuclear cells from patients with newly diagnosed ALL (lanes 6–15) were subjected to SDS-PAGE followed by transfer to nitrocellulose and blotting with antibodies that recognize phospho-ERK (upper panel). After erasure, blots were probed with antibodies to total ERK (bottom panel) and histone H1, which served as a loading control. Samples containing 2.5 x 104, 5 x 104, 1.25 x 104, 2.5 x 105, and 5 x 105 HL-60 cells (lanes 1–5, respectively) served as a positive control for phospho-ERK blotting. Asterisks indicate samples considered positive in this analysis using criteria described under "Experimental Procedures." C, survival of the ALL patients who achieved a complete remission and whose leukemic cells lacked (solid line) or displayed (dotted line) constitutive ERK phosphorylation. Data are plotted by the method of Kaplan and Meier.

 
To determine whether the MAPK pathway is also constitutively activated in the clinical setting, 24 pretreatment ALL bone marrow samples that contained >80% blasts + lymphocytes (median 88% blasts) and additional samples from patients with other leukemias were examined using identical techniques. Results of this analysis are summarized in Table II. Constitutive MAPK activation was not detectable in specimens of chronic lymphocytic leukemia (Table II), an indolent lymphoid malignancy with an extremely low proliferative index. In contrast, 12 of 33 ALL marrows displayed constitutive ERK phosphorylation, as illustrated in Fig. 1B. Reprobing of the same blots with antiserum that recognized ERK regardless of phosphorylation state confirmed the presence of ERK1 and ERK2 in all samples at similar levels (Fig. 1B, middle panel). Collectively, these results not only provide evidence for MAPK activation in 1/3 of pretreatment ALL samples but also demonstrate that this activation is independent of ERK overexpression, a factor that has been implicated in MAPK activation in some AML samples (21).


View this table:
[in this window]
[in a new window]
  		TABLE II
Frequency of constitutive ERK phosphorylation in pretreatment marrow samples

 
When possible relationships between constitutive ERK phosphorylation and various clinical parameters were examined, there was no correlation between ERK phosphorylation and lineage (T versus B cell, Table II) or peripheral leukocyte count at diagnosis (not shown). ERK phosphorylation was undetectable in all Ph+ ALL samples as well as all samples from patients with chronic myelogenous leukemia in lymphoid blast crisis (Table II). Interestingly, among ALL patients who achieved a CR, there was a trend (Fig. 1C) toward poorer survival with phospho-ERK-positive leukemia (median CR duration 7 months, range 1–14 months) compared with phospho-ERK-negative blasts (median CR duration 34 months, range 6 to >l48 months). These observations provided impetus for studying the effect of MEK inhibition in lymphoid cells.

CI-1040 Induces S Phase Arrest In Lymphoid Cell Lines— Because clinical leukemia samples undergo apoptosis when cultured for prolonged periods in vitro, we utilized leukemia cell lines to examine the biological effects of MEK inhibition. Previous studies demonstrated that MEK1/2 inhibition results in p27Kip-associated G1 arrest in human carcinoma (18, 19) and AML cell lines (22). To determine whether lymphoid cells respond similarly, lymphoid cell lines were treated with 10 µM CI-1040, a concentration that approximates steady state concentrations of CI-1040 and its biologically active metabolite PD0184264 observed in clinical trials (19, 33). CI-1040 completely inhibited ERK phosphorylation within 1 h (Fig. 2A). Within 24 h, Jurkat T cell leukemia cells accumulated in S phase (Fig. 2B). This S phase slowing was observed at CI-1040 concentrations as low as 2.5 µM. At 1.25 µM, CI-1040 had no detectable effect on the Jurkat cell cycle. This S phase arrest was clearly different from the CI-1040-induced G1 arrest observed in HL-60 (Fig. 2B) and ML-1 (not shown) myeloid leukemia cells.



View larger version (34K):
[in this window]
[in a new window]
  		FIG. 2.
Cell cycle effects of CI-1040. A, log phase Jurkat cells were treated with 10 µM CI-1040 for the indicated length of time (lanes 1–6). Alternatively, cells were treated with 50 nM phorbol 12-myristate 13-acetate, a known MAPK pathway activator (44), for 30 min prior to lysis (lane 7). Samples were subjected to SDS-PAGE followed by immunoblotting with antibodies that recognize phospho-ERK and total ERK as described in the legend to Fig. 1. B, log phase HL-60 or Jurkat cells were treated with diluent (left panel)or10 µM CI-1040 (right panel) for 24 h prior to fixation in 50% ethanol and staining with propidium iodide. Numbers represent cell cycle distribution (mean ± S.D.) inferred from Modfit analysis of 4–6 separate experiments. C, RT-PCR analysis of mRNA encoding p21Waf1, p27Kip1, or {beta}-actin after treatment of Jurkat cells with 10 µM CI-1040 for the indicated length of time. D, log phase Jurkat cells were treated with 10 µM CI-1040 for the indicated length of time followed by immunoblotting of whole cell lysates with monoclonal antibody to c-Myc or cdc25a. Histone H1 served as a loading control.

 
Additional experiments were performed to evaluate potential explanations for the S phase arrest in Jurkat cells. Increased expression of p21Waf1 or p27Kip1 was not detected at the mRNA or protein level in these cells (Fig. 2C and data not shown), providing an explanation for the lack of G1 arrest and clearly distinguishing their response from that of carcinoma and AML cell lines (18, 19, 22). Activation of checkpoint kinase 1 and degradation of its substrate Cdc25a are prominent features of S phase arrest after DNA damage (42, 43). Further studies failed to demonstrate either an activating phosphorylation of checkpoint kinase 1 (not shown) or an appreciable decrease in Cdc25a protein at the time of the S phase arrest (Fig. 2D). Instead, down-regulation of c-Myc (Fig. 2D) and cyclin E (not shown) was detected within 24 h of CI-1040 addition. Further experiments revealed that these changes occurred even in the presence of the broad spectrum caspase inhibitor ZVAD(OMe)-fmk.2

Analysis of seven additional cell lines representing both T (CEM, Molt3, and H9) and B (SKW6.4, Daudi, Raji, and BJAB) lineages showed that CI-1040 also induced S phase arrest in CEM, Molt3, H9, Daudi, and Raji cells. In contrast, CI-1040 induced a marked G1 arrest in SKW6.2 and BJAB cells, with a concomitant decrease in S and G2 populations. These results not only demonstrated that the cell cycle effects of CI-1040 in 6 of 8 lymphoid cell lines differ from the effects reported previously in myeloid cells but also indicated that the effects in lymphoid lines are heterogeneous.

CI-1040 Induces Apoptosis in Lymphoid Cell Lines—Because the cell cycle effects of CI-1040 in most of the lymphoid cells were unexpectedly different from those reported in AML lines, we asked whether more prolonged CI-1040 treatment induces apoptosis in lymphoid cell lines. As illustrated in Fig. 3A, analysis by flow cytometry after propidium iodide staining or morphological examination demonstrated apoptosis in 40–50% of Jurkat cells 72 h after the addition of 10 µM CI-1040. Although all eight of the lymphoid cell lines examined underwent CI-1040-induced apoptosis, their sensitivity varied (Fig. 3B). In some, exemplified by Molt3, CI-1040 concentrations as low as 0.5 µM induced readily detectable apoptosis. In others, e.g. CEM cells, higher CI-1040 concentrations were required before apoptosis was observed.



View larger version (31K):
[in this window]
[in a new window]
  		FIG. 3.
Induction of apoptosis in lymphoid cell lines. A, log phase Jurkat were treated with the indicated CI-1040 concentration for 72 h, fixed in ethanol, and examined by flow cytometry for the presence of cells with extractable DNA ("subdiploid cells"). Inset, cells treated with diluent or 10 µM CI-1040 for 72 h were fixed, stained with Hoechst 33258, and examined by fluorescence microscopy. White circle, fragmented nucleus of an apoptotic cell. B, Molt3 and CEM cells were treated with the indicated CI-1040 concentration for 72 h and examined by flow cytometry as described in A. C, Jurkat cells were treated with 10 µM CI-1040 for the indicated length of time, fixed, and examined for apoptotic morphological changes as indicated in the inset to A. D, Jurkat cells treated with 10 µM CI-1040 for the indicated length of time were lysed under hypotonic conditions. Activity that was capable of cleaving DEVD-AFC in cytosol (280,000 x g supernatant) was assayed as described previously (82). Similar results were obtained when the caspase-6-selective substrate VEID-AFC or the caspase-9-selective substrate LEHD-AFC was substituted for DEVD-AFC. Results in each panel are representative of 3–4 independent experiments.

 
Jurkat cells were chosen for more detailed analysis because of their ease of transfection, sensitivity to a variety of death-inducing stimuli, and well characterized apoptotic pathways (31, 44). Time course studies (Fig. 3C) indicated that 10 µM CI-1040 induced relatively little apoptosis in Jurkat cells over the first 24 h. With continuous treatment for 48–72 h, increasing numbers of cells displayed chromatin condensation and nuclear fragmentation. These apoptotic morphological changes were accompanied by caspase activation, as indicated by the appearance of activity that cleaves DEVD-AFC (Fig. 3D) and by cleavage of intracellular caspase substrates, including PARP, topoisomerase I, and protein kinase C{delta} (Fig. 4A). In addition, HSP90, which is not traditionally viewed as caspase substrate, was also cleaved (Fig. 4A).



View larger version (50K):
[in this window]
[in a new window]
  		FIG. 4.
Caspase activation occurs during CI-1040-induced apoptosis. A, after log phase Jurkat cells were treated with 10 µM CI-1040 for the indicated length of time; whole cell lysates (50 µg of protein) were subjected to SDS-PAGE followed by immunoblotting with antibodies that recognize the caspase substrates described previously. Cells treated for 6 h with 68 µM etoposide served as a positive control for caspase activation (31). Arrowheads indicate cleavage products. Dotted line indicates juxtaposition of two widely spaced areas on the same piece of x-ray film. Topo, topoisomerase; PKC, protein kinase C. B, log phase Jurkat cells were treated with 10 µM CI-1040 for the indicated length of time. Whole cell lysates were subjected to SDS-PAGE followed by immunoblotting with antibodies that recognize the indicated procaspases or epitope-specific antisera that recognize the indicated cleaved caspases (30). Results are representative of four independent experiments.

 
To determine which caspases were activated, whole cell lysates were probed with antibodies that recognize caspase zymogens and/or cleaved caspases. This analysis demonstrated cleaved fragments of two initiator caspases, caspases-8 and -9, as well as caspase-3 at 36 h (Fig. 4B). By 48–60 h, decreases in procaspases-2, -6, and -7 were also evident, with cleavage products consistent with active caspases being demonstrable in some cases.

CI-1040 Induces Apoptosis Despite Inhibition of the Mitochondrial Pathway—Previous studies have demonstrated that ERK pathway activation affects Bcl-2 family members in several ways. Bcl-2 and BAD are phosphorylated, enhancing the anti-apoptotic capability of the former (9) and decreasing the proapoptotic properties of the latter (45, 46). In addition, ERK pathway activation enhances synthesis of the anti-apoptotic family members Bcl-2, Bcl-xL, and Mcl-1 in some cell lines (17, 47). These observations raised the possibility that CI-1040 might induce apoptosis by affecting Bcl-2 family member-mediated regulation of the mitochondrial pathway (48, 49).

To assess this possibility, we next examined the effect of CI-1040 on levels of Bcl-2 family polypeptides. Immunoblotting demonstrated that Mcl-1, Bcl-2, Bax, and Bak were all down-regulated 72 h after CI-1040 addition (Fig. 5A, lane 6). Importantly, Mcl-1, which has a relatively short half-life, diminished markedly within 36 h, whereas the other polypeptides decreased only at 60–72 h. In the case of BAK, a discrete polypeptide with greater mobility was detected (Fig. 5A, lanes 5 and 6, arrowhead), raising the possibility that down-regulation of Bcl-2 family members might result from caspase activation. Indeed, previous studies have shown that Bcl-2, Bcl-xL, Bax, and Bad can all be proteolytically cleaved by caspases (50–54).



View larger version (47K):
[in this window]
[in a new window]
  		FIG. 5.
CI-1040-induced changed in Bcl-2 family members. A, down-regulation of Mcl-1, Bcl-2, Bax, and Bak after CI-1040 treatment. After log phase Jurkat cells were treated with 10 µM CI-1040 for the indicated length of time, whole cell lysates were subjected to immunoblotting using reagents specific for the indicated antigen. B, Mcl-1 down-regulation does not require caspase activity. Log phase Jurkat cells were treated with diluent (-)or10 µM CI-1040 (+) for 48 h. Immediately prior to CI-1040 addition and 24 h later, ZVAD(OMe)-fmk was added at a final concentration of 50 µM to the indicated samples. At the completion of the incubation, whole cell lysates (50 µg of protein) were subjected to SDS-PAGE followed by immunoblotting. Blotting for the cleavage fragment of PARP, which can be produced by caspases-3, -7, or -9, confirmed the inhibition of caspases by ZVAD(OMe)-fmk. C, RT-PCR for Bcl-2, Bcl-xL, Mcl-1, and actin mRNA.

 
To evaluate the possible role of caspases in Mcl-1 down-regulation, Jurkat cells were treated with CI-1040 in the absence or presence of ZVAD(OMe)-fmk. Cleavage of PARP and Bak was inhibited by ZVAD(OMe)-fmk (Fig. 5B, lane 3), confirming the efficacy of caspase inhibition. In contrast, ZVAD(OMe)-fmk had no effect on Mcl-1 down-regulation (Fig. 5B, lane 3), suggesting that the latter process is caspase-independent. Additional experiments indicated that the Mcl-1 down-regulation also occurred independent of changes in Mcl-1 mRNA (Fig. 5C).

To assess the importance of Mcl-1 down-regulation in CI-1040-induced apoptosis, Jurkat cells were transfected with cDNA encoding Mcl-1 behind a strong constitutive promoter. This forced overexpression was sufficient to markedly inhibit etoposide-induced apoptosis (Fig. 6A). In contrast, the effect on CI-1040-induced apoptosis was more limited, raising the possibility that events other than Mcl-1 down-regulation contribute to CI-1040-induced cytotoxicity.



View larger version (33K):
[in this window]
[in a new window]
  		FIG. 6.
Dependence of CI-1040-induced apoptosis on caspase-8 rather than the mitochondrial pathway. A, Jurkat cells were transfected with a 4:1 ratio of plasmids encoding Mcl-1 and EGFP or EGFP alone. 24 h after electroporation, cells expressing EGFP were isolated by flow cytometry and treated with diluent, 10 µM CI-1040, or 1.7 µM etoposide. After incubation for 48 h, cells were harvested, stained with Hoechst 33258, and examined by fluorescence microscopy as illustrated in Fig. 3A. B, Jurkat cells were transfected with a 4:1 ratio of plasmids encoding dominant negative caspase-9 (dn-Casp9) (56, 83) and EGFP or EGFP alone. 24 h after electroporation, cells expressing EGFP were isolated by flow cytometry and treated with diluent, 10 µM CI-1040, or 100 µM paclitaxel for 48 h. Apoptotic cells were quantitated as described in A. C, Jurkat cells were treated with 10 µM IETD(OMe)-fmk or diluent (0.1% Me2SO) for 5 min followed by addition of 10 µM CI-1040 (left panel) or 1 µM etoposide (right panel). At the indicated time after drug addition, cells were examined for apoptotic morphological changes. D, JB-6 cells were transfected with plasmids encoding full-length wild-type procaspase-8 and EGFP or EGFP alone. 24 h after electroporation, EGFP-expressing cells were isolated by flow cytometry and treated with diluent or 10 µM CI-1040 for 48 h. Apoptotic morphological changes were assessed as described in A. Inset, whole cell lysates containing 50 µg of protein were subjected to SDS-PAGE followed by immunoblotting with antibodies that recognized the indicated polypeptides. E, after Jurkat, Molt3, HL-60, or ML-1 cells were treated with 10 µM CI-1040 for 72 h in the absence or presence of 10 µM IETD(OMe)-fmk, apoptotic morphological changes were assessed as described in A.

 
Because CI-1040 could potentially affect the levels or post-translational modification of additional Bcl-2 family members, all of which participate in regulation of the mitochondrial pathway of apoptosis (48), the effect of inhibiting the mitochondrial pathway was assessed next. Transfection with dominant negative caspase-9, a well established inhibitor of the mitochondrial pathway (55, 56), diminished paclitaxel-induced apoptosis but had little effect on CI-1040-induced apoptosis (Fig. 6B). These results are difficult to reconcile with any model that implicates the mitochondrial pathway in CI-1040-induced apoptosis in these cells.

CI-1040 Activates a Caspase-8-dependent Pathway—In view of data suggesting that caspase-8 might also be activated early during CI-1040-induced apoptosis (Fig. 4B), further experiments examined the effect of the caspase-8 inhibitor IETD(OMe)-fmk. This inhibitor caused a transient delay in etoposide-induced apoptosis in Jurkat cells (Fig. 6C, right panel), confirming previous reports that a Fas-dependent process might accelerate etoposide-induced apoptosis under certain conditions but is not essential for this process (31, 57). In contrast, IETD(OMe)-fmk completely abrogated CI-1040-induced apoptosis (Fig. 6C, left panel), raising the possibility that caspase-8 might play a critical role in CI-1040-induced death. Consistent with this conclusion, the Jurkat cell variant JB-6, which lacks procaspase-8, was resistant to CI-1040-induced apoptosis; and restoration of procaspase-8, which was itself slightly toxic, enhanced CI-1040 sensitivity (Fig. 6D). Although caspase-8-deficient variants of other hematopoietic cell lines are not available, further experiments indicated that caspase-8 inhibition had a similar effect on Molt3 cells but not in either myeloid cell line examined (Fig. 6E).

FADD Dephosphorylation and Aggregation in CI-1040-treated Cells—Most studies implicating caspase-8 as an initiator caspase have involved DRs and their ligands (58, 59). FasL up-regulation has been observed after treatment with a variety of DNA-damaging agents (60, 61). In addition, TRAIL has been implicated in retinoid-induced apoptosis in progranulocytic leukemia cells (62), and TRAIL receptor up-regulation has been implicated in the cytotoxicity of sulindac sulfide (63) and 2-methoxyestradiol (64). Accordingly, the possibility that CI-1040 might be activating caspase-8 through these DRs was examined.

RT-PCR failed to detect increased expression of any death ligand examined, including FasL, TRAIL, tumor necrosis factor-{alpha}, or their receptors (Fig. 7A). Because this experiment could not rule out the possibility that CI-1040 might signal through DRs without altering mRNA expression, e.g. by inducing altered ligand or receptor trafficking (65), we also examined the effect of receptor blockade or deletion. ZB4, which inhibits induction of apoptosis by FasL (31) or the agonistic anti-Fas antibody CH-11 (Fig. 7B), did not block CI-1040-induced apoptosis. Consistent with these results, JM14A5 cells, a Jurkat variant lacking the Fas receptor (Fig. 7C, inset), remained sensitive to CI-1040 despite resistance to CH-11 (Fig. 7C). Collectively, these results argue against involvement of the Fas/FasL pathway in CI-1040-induced apoptosis. Moreover, a soluble DR5:Fc fusion construct, which binds TRAIL and inhibits its interaction with both DR4 and DR5, failed to affect CI-1040-induced apoptosis even though it completely blocked TRAIL-induced apoptosis (Fig. 7D). These results suggest that the most widely studied DRs and their ligands are not involved in CI-1040-induced apoptosis.



View larger version (29K):
[in this window]
[in a new window]
  		FIG. 7.
Lack of evidence for DR signaling in CI-1040-induced apoptosis. A, Jurkat cells were treated with 10 µM CI-1040 for the indicated length of time. RT-PCR was performed to examine steady state message levels for tumor necrosis factor-{alpha} (TNF-{alpha}), FasL, TRAIL, and their receptors TNFR1, Fas, DR4, and DR5, respectively. Message encoding the housekeeping protein glyceraldehyde-3-phosphate dehydrogenase (GAPDH) served as a loading control. Previous studies (84) have demonstrated the lack of DR4 expression in Jurkat cells. B, Jurkat cells were treated for 72 h with diluent or 10 µM CI-1040 or for 24 h with 10 ng/ml agonistic anti-Fas antibody CH-11 in the absence or presence of the blocking anti-Fas antibody ZB4 (31, 85). Note that ZB4 inhibited CH-11-induced apoptosis but not CI-1040-induced apoptosis. C, Jurkat cells or JM14A5 cells (31) were treated for 72 h with diluent or 10 µM CI-1040 or for 24 h with 10 ng/ml antibody CH-11. Inset, Jurkat or JM14A5 cells were subjected to flow cytometry after staining with APO-1–1 anti-Fas (heavy line) or isotype-matched control (fine line) followed by phycoerythrinconjugated anti-mouse IgG. Note that JM14A5 cells, which lack Fas expression, are resistant to CH-11 but not CI-1040. D, Jurkat cells were treated for 72 h with diluent or 10 µM CI-1040 or for 24 h with 1 ng/ml recombinant human TRAIL (R & D Systems, Minneapolis, MN) in the absence or presence of 1 µM soluble artificial decoy receptor DR5:Fc (Alexis, San Diego, CA), which competes for TRAIL binding to its receptors. Note that DR5:Fc prevents TRAIL-induced apoptosis but not CI-1040-induced apoptosis.

 
To determine whether FADD was involved in CI-1040-induced caspase-8 activation, we examined I2.1 cells (66), a Jurkat variant lacking FADD (Fig. 8A, inset). CI-1040 inhibited ERK phosphorylation in these cells (Fig. 8B) but failed to activate caspases (Fig. 8B) or induce apoptotic morphological changes (Fig. 8A). These results indicated that FADD is essential for CI-1040-induced caspase-8 activation.



View larger version (47K):
[in this window]
[in a new window]
  		FIG. 8.
FADD dephosphorylation and aggregation during CI-1040-induced apoptosis. A, Jurkat cells or the FADD-deficient variant I2.1 (66) was treated with 10 µM CI-1040 for 72 h or 10 ng/ml CH-11 agonistic anti-Fas antibody for 24 h. Inset, whole cell lysates containing 50 µg of protein were subjected to SDS-PAGE followed by immunoblotting with antibodies that recognize the indicated polypeptides. B, Jurkat cells or I2.1 cells were treated 10 µM CI-1040 for the indicated length of time. Whole cell lysates were subjected to SDS-PAGE followed by blotting with reagents that recognize the indicated polypeptides. C, Jurkat cells were treated 10 µM CI-1040 for the indicated length of time. Whole cell lysates were subjected to SDS-PAGE followed by blotting with reagents that recognize FADD or, as a loading control, histone H1. D, left panel, after Jurkat cells were treated for 48 h with 10 µM CI-1040 (+) or diluent (-) in the absence (-) or presence (+) of 50 µM ZVAD(OMe)-fmk, whole cell lysates were subjected to SDS-PAGE followed by blotting with anti-FADD antibody. Right panel, Jurkat cells were lysed in DISC buffer and treated with 54 units of calf intestine alkaline phosphatase (CIAP) (or diluent) at 30 °C for 30 min. After SDS-PAGE, samples were probed with anti-FADD antibody. E, after treatment with 10 µM CI-1040 for the indicated length of time, Jurkat cells were fractionated into Triton-soluble and Triton-insoluble components. Aliquots containing 50 µg of protein were subjected to SDS-PAGE and probed with anti-FADD antibody. F, cells treated with 250 ng/ml CH-11 agonistic anti-Fas antibody for 90 (lanes 2 and 6) or 180 min (lanes 3 and 7) or with 50 µM etoposide for 4 h were fractionated as described in D. G, I2.1 cells transfected with plasmid encoding free EGFP (upper row) or full-length FADD fused at its C terminus to EGFP (bottom row) were treated with diluent (control) or 10 µM CI-1040 for 30 h or with 250 ng/ml CH-11 for 90 min. Bar graph, the percentage of cells displaying visible aggregates was determined after 30 h of treatment with 0.1% Me2SO (diluent) or 10 µM CI-1040. Bars, mean and range of two separate experiments examining 100–250 EGFP-expressing cells in each treatment group.

 
In wild-type Jurkat cells, further examination revealed that FADD underwent a mobility shift between 24 and 72 h after CI-1040 (Fig. 8C). Because this mobility shift occurred in the presence of the broad spectrum caspase inhibitor ZVAD(OMe)fmk (Fig. 8D, left panel), caspase-mediated proteolysis did not appear to be involved. Instead, as reported previously (67), this mobility shift was indistinguishable from the mobility shift induced by phosphatase treatment (Fig. 8D, right panel). After CI-1040 treatment, the mobility shift was accompanied by progressive redistribution of FADD from cytosol to a fraction that resisted extraction in nonionic detergent (Fig. 8E). Interestingly, only dephosphorylated FADD was recovered in the insoluble compartment even at the earliest time points (Fig. 8E, lanes 5–8). This redistribution was unique to CI-1040 treatment and was not observed after other apoptotic stimuli such as Fas receptor ligation or etoposide treatment (Fig. 8F).

In further experiments, the fate of FADD was followed by introducing a full-length FADD-GFP fusion protein into FADD-deficient I2.1 cells and examining its distribution by confocal microscopy (Fig. 8G). Prior to any apoptotic stimulus, FADD-GFP was uniformly distributed throughout the cytoplasm. Fas ligand treatment resulted in small submembranous FADD aggregates (Fig. 8G) as reported previously (68). In contrast, CI-1040 caused the formation of large curved or linear FADD aggregates that circumscribed or transected the cells. Importantly, the large CI-1040-induced FADD-GFP aggregates shown in Fig. 8G first became visible 24–30 h after CI-1040 addition, i.e. just before cells were starting to become apoptotic (Fig. 3C).

FADD and Caspase-8 Are Also Required for PD098059-induced Apoptosis in Jurkat Cells—In a final series of experiments, the major findings in Figs. 6, 7, 8 were reproduced by using the more widely studied MEK1/2 inhibitor PD098059. Results of this analysis demonstrated that two separate caspase-8-deficient Jurkat variants, JB-6 and I9.2 (69), were resistant to induction of apoptosis by PD098059 (Fig. 9A). Likewise, FADD-deficient I2.1 cells were resistant to PD098059 (Fig. 9B). As was the case with CI-1040, treatment with PD098059 was accompanied by progressive accumulation of dephosphorylated FADD in insoluble aggregates (Fig. 9C, lanes 5–8) as wild-type Jurkat cells underwent apoptosis. Collectively, these results argue that the dependence on FADD and caspase-8 for induction of apoptosis is not due to some unique property or spectrum of activity of CI-1040.



View larger version (34K):
[in this window]
[in a new window]
  		FIG. 9.
Comparison of apoptosis induced by CI-1040 and PD098059 in Jurkat cells. A, Jurkat cells and the caspase-8-deficient variants I9.2 and JB-6 were treated with diluent, 10 µM CI-1040, or 30 µM PD098059 for 72 h. Apoptosis was quantitated as illustrated in Fig. 3A. B, Jurkat cells and the FADD-deficient variant I2.1 were treated as described in A. C, after treatment with 30 µM PD098059 for the indicated length of time, Jurkat cells were fractionated into Triton-soluble and Triton-insoluble components. Aliquots containing 50 µg of protein were subjected to SDS-PAGE and probed with anti-FADD antibody.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
Results of the present study demonstrate constitutive MAPK pathway activation in ALL cell lines and a substantial fraction of clinical ALL specimens. In view of this observation, the effects of MEK1/2 inhibition were examined in lymphoid cells. Results of this analysis demonstrated that CI-1040 induces a previously undescribed S phase arrest in 6 of 8 lymphoid lines examined, and that this S phase arrest is followed by induction of apoptosis through a unique pathway that involves FADD dephosphorylation and aggregation followed by caspase-8 activation. These observations, which support the model depicted in Fig. 10, suggest previously unrecognized processes that might be regulated by ERKs in lymphoid cells.



View larger version (18K):
[in this window]
[in a new window]
  		FIG. 10.
Diagram showing proposed pathway of CI-1040-induced cell cycle effects and apoptosis in Jurkat cells. As described in the text, ERK1/2 activation appears to contribute to stabilization of c-Myc and Mcl-1. Down-regulation of these polypeptides, however, does not appear to be directly involved in the cytotoxic effects of CI-1040. Instead, CI-1040-induced apoptosis appears to be triggered by FADD aggregation followed by caspase-8 activation. As illustrated in Figs. 6, 7, 8, 9, this apoptosis is diminished in cells that lack FADD (I2.1 cells) or procaspase-8 (JB-6 cells, I9.2 cells) or by treatment of Jurkat cells with the caspase-8 inhibitor (28) IETD(OMe)-fmk.

 
Earlier studies in carcinoma and AML cell lines indicated that CI-1040 and other MEK inhibitors induce a p27Kip1-mediated G1 arrest (18, 19, 22). In contrast, the present study failed to demonstrate p27Kip1 up-regulation (Fig. 2C) or G1 arrest (Fig. 2B) in Jurkat cells. In fact, 6 of 8 lymphoid lines examined lacked a CI-1040-induced G1 arrest. Instead, CI-1040 caused S phase slowing in these lines (Fig. 2B). Down-regulation of Cdc25a, one of the features of DNA damage-induced S phase slowing (42, 70), was not seen in these cells (Fig. 2D). These results not only distinguish the CI-1040-induced S phase accumulation from a DNA damage response but also suggest that CI-1040-treated Jurkat cells might be useful for future studies to identify additional events that contribute to the poorly understood S phase checkpoint (70, 71).

In view of the unexpected cell cycle response to CI-1040, we also examined the consequences of more prolonged drug exposure. CI-1040 induced apoptosis in all of the lymphoid examined (Fig. 3 and data not shown). The mechanism, however, was unanticipated.

Previous studies had demonstrated that transcription of Bcl-2, Bcl-xL, and Mcl-1 is regulated by ERK in some cell lines (17, 47). In the present study, Bcl-2, Bax, and Bak decreased after a 72-h CI-1040 treatment (Fig. 5A). RT-PCR, however, failed to demonstrate decreased Bcl-2 or Bcl-xL mRNA (Fig. 5C). Instead, time course studies indicated that Bcl-2, Bax, and Bak down-regulation (Fig. 5A) occurred after caspase activation (Figs. 3D and 4). Moreover, in the case of Bak, a discrete cleavage product accumulated during apoptosis (Fig. 5A); and appearance of this fragment was inhibited by ZVAD(OMe)fmk (Fig. 5B). Whereas caspase-mediated degradation of a number of Bcl-2 family members, including Bcl-2, Bcl-xL, Bax and Bad has been observed previously (50–54), to our knowledge the present observations provide the first report of caspase-dependent Bak cleavage during apoptosis.

Compared with Bak, Mcl-1 down-regulation occurred at earlier times (Fig. 5A) and did not depend on caspase activation (Fig. 5B). When combined with previous data showing ERK-mediated phosphorylation of Mcl-1 (10), our results suggested that Mcl-1 phosphorylation might stabilize this anti-apoptotic polypeptide, much as has been postulated for ERK-mediated phosphorylation of c-Myc (72). Despite the fact that CI-1040 induced Mcl-1 down-regulation, however, Mcl-1 transfection failed to protect the cells from CI-1040 (Fig. 6A). It is possible that the failure of forced Mcl-1 overexpression to protect cells might result from inability to up-regulate Mcl-1 sufficiently. On the other hand, the extensive toxicity of CI-1040 in cells transfected with dominant negative caspase-9, which diminishes the cytotoxicity of multiple agents that activate the mitochondrial pathway (Fig. 6B) (55, 56), suggested that CI-1040 induces apoptosis in Jurkat cells through a different apoptotic pathway.

Consistent with this possibility, the caspase-8 inhibitor IETD(OMe)-fmk abrogated CI-1040-induced apoptosis in Jurkat cells (Fig. 6C). Moreover, caspase-8 deficiency diminished CI-1040 sensitivity, and caspase-8 transfection restored sensitivity (Fig. 6D). Collectively, these observations suggested a critical role for caspase-8 in CI-1040-induced apoptosis.

Although caspase-8 can sometimes be activated downstream of caspase-3 during the course of apoptosis triggered through the mitochondrial pathway (29, 73, 74), it is important to emphasize that the involvement of caspase-8 in CI-1040-induced apoptosis is different. First, caspase-8 cleavage is observed at the earliest time when caspase activation is detected during CI-1040 exposure (Fig. 4B). Second, elimination of caspase-8 activity renders cells resistant to CI-1040 (Fig. 6, C and D), whereas caspase-8 inhibition has no effect when apoptosis is triggered through the mitochondrial pathway (73).

Even though most caspase-8-initiated apoptotic processes involve DR ligation, CI-1040-induced apoptosis does not appear to follow this paradigm. In contrast to 5-fluorouracil and retinoids, which reportedly trigger apoptosis through up-regulation of FasL and TRAIL (62, 75), respectively, CI-1040 did not induce up-regulation of FasL, TRAIL, or their receptors (Fig. 7A). Instead, CI-1040 induced gradual FADD dephosphorylation (Fig. 8C), and the dephosphorylated FADD accumulated in insoluble aggregates (Fig. 8, E and G) that were not seen with other apoptosis-inducing agents (Fig. 8F). Because of the ability of FADD to bind procaspase-8 (76, 77), the formation of these aggregates presumably is sufficient to juxtapose procaspase-8 molecules and induce their activation. These observations provide at least one explanation for DR-independent FADD-dependent apoptosis, a phenomenon that has also been implicated recently in the cytotoxic action of a granulocyte/macrophage-colony-stimulating factor/diphtheria toxin conjugate (78).

The mechanism proposed in Fig. 10 is supported by a variety observations, including the demonstration that CI-1040 induces FADD aggregation (Fig. 8, E and G) as well as the lack of cytotoxicity of CI-1040 in Jurkat variants that lack procaspase-8 or FADD (Figs. 6D and 8A, respectively). Similar results were obtained with PD098059 (Fig. 9). As indicated in Fig. 6E, however, IETD(OMe)-fmk inhibits CI-1040-induced apoptosis in the lymphoid lines Jurkat and Molt3 but not the myeloid lines HL-60 and ML-1. This variation might reflect diminished inhibitor uptake into the myeloid cell lines or the involvement of caspase-10, which is 10-fold resistant to IETD-fmk3 but is lacking from Jurkat cells (79). Alternatively, the inability of IETD(OMe)-fmk to protect HL-60 and ML-1 cells might reflect utilization of a different pathway for triggering apoptosis in myeloid versus lymphoid lines. Further experiments are required to distinguish among these possibilities. Moreover, additional experiments are required to determine whether activated ERKs phosphorylate FADD directly or regulate its phosphorylation indirectly.

In addition to providing evidence that ERK activation regulates S phase progression and FADD solubility in lymphoid cells, results of the present study have potential implications for the development of MEK inhibitors in lymphoid malignancies. CI-1040, the first MEK inhibitor to be tested in the clinic, is well tolerated at doses that inhibit ERK phosphorylation in situ (33). Although earlier reports demonstrated constitutive MAPK pathway activation in 50–70% of AML specimens (20–22), only a limited number of ALL specimens had been examined previously. The present study demonstrated constitutive ERK phosphorylation in 8 of 24 pretreatment ALL samples (Fig. 1B). Interestingly, constitutive phosphorylation was not detected in samples of Ph+ ALL or other Bcr/Abl-driven leukemias (Table II). This observation confirms a previous study showing Raf activation without ERK activation downstream of Bcr/Abl in tissue culture cells (80) and supports the hypothesis that the anti-apoptotic effects of Bcr/Abl overexpression result from activation of other pathways (81). On the other hand, the observation that ERK phosphorylation is associated with shorter survival in Ph-negative ALL (Fig. 1C), if confirmed in larger studies, raises the possibility that MEK inhibitors might be worthy of further study, alone or in combination with other agents, in poor prognosis adult ALL.


   FOOTNOTES
 
* This work was supported in part by National Institutes of Health Grants R01 CA69008 and F32 CA93055. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

{ddagger} Present address: Dept. of Pediatrics, M. D. Anderson Cancer Center, Houston, TX 77030. Back

|| To whom correspondence and reprints should be addressed: Division of Oncology Research, Guggenheim 1342 C, Mayo Clinic, 200 First St., S.W., Rochester, MN 55905. Tel.: 507-284-8950; Fax: 507-284-3906; E-mail: Kaufmann.Scott{at}Mayo.edu.

1 The abbreviations used are: ERK, extracellular signal-regulated kinase; ALL, acute lymphocytic leukemia; AML, acute myelogenous leukemia; CR, complete response; DEVD-AFC, N{alpha}-acetylaspartylglutamylvalinyl-aspartyl-7-amino-4-trifluoromethylcoumarin; DR, death receptor; EGFP, enhanced green fluorescent protein; FADD, Fas-associated death domain protein; FasL, Fas ligand; IETD(OMe)-fmk; N-(N{alpha}-benzyloxycarbonylisoleucylglutamylthreonyl)aspartic acid fluoromethyl ketone; MAPK, mitogen-activated protein kinase; MEK, mitogen-activated kinase/extracellular signal-regulated kinase kinase; Ph+, Philadelphia chromosome-positive; TRAIL, tumor necrosis factor-{alpha}-related apoptosis-inducing ligand; ZVAD(OMe)-fmk, the O-methyl ester of N-(N{alpha}-benzyloxycarbonylvalinylalanyl) aspartic acid fluoromethyl ketone; RT, reverse transcriptase; PARP, poly(ADP-ribose) polymerase; HSP90, heat shock protein 90. Back

2 D. Loegering and S. H. Kaufmann, unpublished observations. Back

3 T. J. Kottke and S. H. Kaufmann, unpublished observations. Back


   ACKNOWLEDGMENTS
 
We thank the Flow Cytometry and Optical Morphology Core for support of cell cycle analysis, cell sorting, and confocal microscopy; David Toft, Guy Poirier, Emad Alnemri, Paul Leibson, Adele Fielding, Marcus Peter, Shigekazu Nagata, Greg Gores, and Larry Karnitz for reagents; Greg Gores, Dan Billadeau, Larry Karnitz, and Junjie Chen for helpful discussions; William Slichenmyer for encouraging this collaboration; and Deb Strauss for secretarial assistance.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Sebolt-Leopold, J. S. (2000) Oncogene 19, 6594-6599[CrossRef][Medline] [Order article via Infotrieve]
  2. Dent, P., and Grant, S. (2001) Clin. Cancer Res. 7, 775-783[Free Full Text]
  3. Johnson, G. L., and Lapadat, R. (2002) Science 298, 1911-1912[Abstract/Free Full Text]
  4. Okuda, K., Sanghera, J. S., Pelech, S. L., Kanakura, Y., Hallek, M., Griffin, J. D., and Druker, B. J. (1992) Blood 79, 2880-2887[Abstract/Free Full Text]
  5. Welham, M. J., Duronio, V., Sanghera, J. S., Pelech, S. L., and Schrader, J. W. (1992) J. Immunol. 149, 1683-1693[Abstract]
  6. Pulverer, B. J., Kyriakis, J. M., Avruch, J., Nikolakaki, E., and Woodgett, J. R. (1991) Nature 353, 670-674[CrossRef][Medline] [Order article via Infotrieve]
  7. Marais, R., Wynne, J., and Treisman, R. (1993) Cell 73, 381-393[CrossRef][Medline] [Order article via Infotrieve]
  8. Seth, A., Alvarez, E., Gupta, S., and Davis, R. J. (1991) J. Biol. Chem. 266, 23521-23524[Abstract/Free Full Text]
  9. Deng, X., Ruvolo, P., Carr, B., and May, W. S., Jr. (2000) Proc. Natl. Acad. Sci. U. S. A. 97, 1578-1583[Abstract/Free Full Text]
  10. Domina, A. M., Smith, J. H., and Craig, R. W. (2000) J. Biol. Chem. 275, 21688-21694[Abstract/Free Full Text]
  11. Green, D. R., and Evan, G. I. (2002) Cancer Cell 1, 19-30[CrossRef][Medline] [Order article via Infotrieve]
  12. Cowley, S., Paterson, H., Kemp, P., and Marshall, C. J. (1994) Cell 77, 841-852[CrossRef][Medline] [Order article via Infotrieve]
  13. Mansouer, S. J., Mathen, W. T., Hermann, A. S., Candida, J. M., Rong, S., Fukasawa, K., Vande Woude, G. F., and Ahn, N. G. (1994) Science 265, 966-970[Abstract/Free Full Text]
  14. English, J. M., and Cobb, M. H. (2002) Trends Pharmacol. Sci. 23, 40-45[CrossRef][Medline] [Order article via Infotrieve]
  15. Dudley, D. T., Pang, L., Decker, S. J., Bridges, A. J., and Saltiel, A. R. (1995) Proc. Natl. Acad. Sci. U. S. A. 91, 7686-7689
  16. Alessi, D. R., Cuenda, A., Cohen, P., Dudley, D. T., and Saltiel, A. R. (1995) J. Biol. Chem. 270, 27489-27494[Abstract/Free Full Text]
  17. Boucher, M. J., Morisset, J., Vachon, P. H., Reed, J. C., Laine, J., and Rivard, N. (2000) J. Cell. Biochem. 79, 355-369[CrossRef][Medline] [Order article via Infotrieve]
  18. Hoshino, R., Tanimura, S., Watanabe, K., Kataoka, T., and Kohno, M. (2001) J. Biol. Chem. 276, 2686-2692[Abstract/Free Full Text]