Synthetic triterpenoids activate a pathway for apoptosis in ... - Nature

Leukemia (2003) 17, 2122–2129 & 2003 Nature Publishing Group All rights reserved 0887-6924/03 $25.00 www.nature.com/leu

Synthetic triterpenoids activate a pathway for apoptosis in AML cells involving downregulation of FLIP and sensitization to TRAIL W-S Suh1, YS Kim1, AD Schimmer1, S Kitada1, M Minden2, M Andreeff3, N Suh4, M Sporn4 and JC Reed1 1 The Burnham Institute, La Jolla, CA, USA; 2Princess Margaret Hospital, Toronto, Ont., Canada; 3MD Anderson Cancer Cntr, Houston, TX, USA; and 4Dartmouth Medical School, Hanover, NH, USA

Acute myelogenous leukemia (AML) remains a deadly disease for most adult patients, due primarily to the emergence of chemoresistant cells. Defects in apoptosis pathways make important contributions to chemoresistance, suggesting a need to restore apoptosis sensitivity or to identify alternative pathways for apoptosis induction. Triterpenoids represent a class of naturally occurring and synthetic compounds with demonstrated antitumor activity, including 2-cyano-3,12-dioxoolean1,9-dien-28-oic acid (CDDO) and its methyl ester (CDDO-m). We explored the effects of CDDO and CDDO-m in vitro on established AML cell lines (HL-60, U937, AML-2) and on freshly isolated AML blasts. CDDO and CDDO-m reduced the viability of all AML cell lines tested in a dose-dependent manner, with effective doses for killing 50% of cells (ED50) within 48 h of B1 and 0.5 lM, respectively. CDDO or CDDO-m also induced substantial increases in cell death in five out of 10 samples of primary AML blasts. Cell death induced by CDDO and CDDO-m was attributed to apoptosis, based on characteristic cell morphology and evidence of caspase activation. Immunoblot analysis demonstrated proteolytic processing of caspase-3, -7, and -8, but not caspase-9, suggesting the involvement of the ‘extrinsic’ pathway, linked to apoptosis induction by TNF-family death receptors. Accordingly, CDDO and CDDO-m induced concentration-dependent reductions in the levels of FLIP protein, an endogenous antagonist of caspase-8, without altering the levels of several other apoptosis-relevant proteins. Reductions in FLIP were rapid, detectable within 3 h after exposure of AML cell lines to CDDO or CDDO-m. CDDO and CDDO-m also sensitized two of four leukemia lines to TRAIL, a TNF-family death ligand. The findings suggest that synthetic triterpenoids warrant further investigation in the treatment of AML, alone or in combination with TRAIL or other immunebased therapies. Leukemia (2003) 17, 2122–2129. doi:10.1038/sj.leu.2403112 Published online 21 August 2003 Keywords: triterpenoid; CDDO; CDDO-m; AML; apoptosis; FLIP

Introduction Though often exhibiting initial responses to chemotherapy, acute myelogenous leukemia (AML) remains a deadly disease for most adult patients, due primarily to the emergence of chemoresistant cells.1–11 Defects in apoptosis pathways make important contributions to chemoresistance,1,2,10 suggesting a need to restore apoptosis sensitivity in AML or to identify alternative pathways for apoptosis induction. Most chemotherapeutic drugs currently available induce apoptosis via a pathway involving mitochondria, which release apoptogenic proteins into the cytosol, leading to activation of members of the caspase family of cell death proteases.12 Consequently, acquired resistance to cytotoxic anticancer drugs is commonly accompanied by defects in the mitochondria pathway for apoptosis,13 also known as the ‘intrinsic’ pathway.7 Agents that trigger leukemia cell death through alternative pathways are therefore desired. Correspondence: Dr JC Reed, The Burnham Institute, 10901 N. Torrey Pines Rd, La Jolla, CA 92037, USA; Fax: þ 1 858 646 3194 Received 21 March 2003; accepted 14 May 2003; Published online 21 August 2003

Members of the tumor necrosis factor (TNF) family of cytokine receptors can trigger a mitochondria-independent pathway for apoptosis, whereby ligation of death domain-containing receptors results in intracellular recruitment of adapter proteins and certain caspases to the receptor complex.14 The cell death pathway activated by TNF-family death receptors is sometimes referred to as the ‘extrinsic’ pathway for apoptosis,7 and it involves caspase-8 as the apical protease.15,16 In contrast, the apical protease in the mitochondrial pathway is caspase-9.17,18 Both the intrinsic and extrinsic pathways converge on downstream effector caspases, which become proteolytically processed and activated by upstream initiator caspases such as caspase-8 and -9.7 Triterpenoids represent a class of naturally occurring and synthetic compounds with demonstrated antitumor activity.19 Recent studies of the synthetic triterpenoids 2-cyano-3, 12-dioxoolean-1,9-dien-28-oic acid (CDDO) and its methyl ester (CDDO-m) have revealed an ability of these chemical compounds to induce apoptosis of malignant cells in vitro, at doses that are well tolerated in rodents.6,20–23 The principal mechanism of apoptosis induction by CDDO and CDDO-m appears to involve activation of caspase-8, thus mimicking the effects of TNF-family death receptors.20,22,23 In this regard, apoptosis induction by these compounds in the AML cell line U937 and in freshly isolated chronic lymphocytic leukemia (CLL) B cells is reportedly blocked by introduction or expression of the caspase-8 inhibitory viral protein, CrmA.20,22,23 In contrast, blockers of the intrinsic pathway, such as Bcl-2, Bcl-XL or the BIR3-fragment of XIAP, fail to protect leukemia cells and solid tumor cell lines from CDDO or CDDO-m.20,22,23 Furthermore, CDDO and CDDO-m have been shown by our laboratory to cause reductions in the levels of FLIP, an antiapoptotic protein that operates as an endogenous antagonist of caspase-8.22–24 To further explore the potential of synthetic triterpenoids as possible novel agents for the treatment of AML, we determined the effects of CDDO and CDDO-m on apoptosis induction in vitro, using established AML cell lines and primary cultured AML blasts. The effects of CDDO and CDDO-m on the expression of various apoptosis-relevant proteins were also evaluated using AML cell lines, and sensitization to TNF-family death ligand TRAIL was also examined. The findings suggest that CDDO and CDDO-m activate an apoptotic pathway in some AMLs, which is characterized by downregulation of FLIP and increased sensitivity to TRAIL. As such, these compounds provide a novel mechanism for triggering the apoptotic demise of leukemia cells, and warrant further evaluation for their potential use in human clinical trials of patients with chemorefractory AML.

Materials and methods

Reagents CDDO and CDDO-m were prepared as described previously.3,4 The compounds were dissolved in DMSO and stored at 201C.

Triterpenoids reduce FLIP and induce apoptosis of AML cells W-S Suh et al

Benzoylcarbonyl-valine-alanine-aspartate-fluor-methyl-ketone (zVAD-fmk), a broad-spectrum caspase inhibitor,25 was purchased from Calbiochem-Novabiochem (San Diego, CA, USA), dissolved in DMSO, and stored at 201C. Recombinant TRAIL was purchased from Alexis Inc., and stored in small aliquots at 801C.

Cell lines HL-60 and U937 cells were purchased from ATCC (Rockville, MD, USA), and cultured in RPMI supplemented with 10% (v:v) heat-inactivated fetal calf serum (FCS), 1 mM L-glutamine, and streptomycin/penicillin. The human leukemia cell line OCI/AML-2 was established in the laboratory of EA Mc Culloch, the Ontario Cancer Institute, Toronto, Canada,26,27 and was grown in a-minimal-essential medium (a-MEM) containing the same supplements. BC-CML is a new cell line established from a female patient with chronic myelogenous leukemia (CML) in blast crisis. Cells were incubated at 371C in an atmosphere of 5% CO2:95% air. Cells were grown to 2 106/ml, and subcultured approximately every 3 days at a density of 2 105/ml.

AML /BC-CML patient samples Prior to the start of this study, informed consent was obtained and the study protocol was approved by IRBs of MD Anderson Cancer Center (Houston, TX, USA) and the Princess Margaret Hospital (Toronto, Canada). Patients with AML (n ¼ 7) or CML in blast crisis (n ¼ 3) were studied. Peripheral blood samples were obtained from heparinized whole blood of both untreated patients (n ¼ 7) and from (n ¼ 3) patients diagnosed with CD34positive AML or BC-CML, according to standard criteria.28 Blasts were separated by Ficoll density-gradient centrifugation, and verified by immunofluorescence flow cytometry to be composed of 480% CD-34-positive cells. Primary blast cells were cultured at 6.25 105 cells/ml in Iscove’s modified Dulbecco’s medium (IMDM) containing 20% FCS, 1 mM L-glutamine, and streptomycin/penicillin.

Apoptosis assays CDDO and CDDO-m were added at the initiation of cultures at final concentrations of 0.1–3 mM, with or without 100 mmol/l zVAD-fmk. In some cases, TRAIL was added to cultures 6 h after CDDO or CDDO-m, at a final concentration of 100 ng/ml. Cells were recovered by centrifugation at various times for evaluation of the percentage of apoptotic cells by the Annexin V method using flow-cytometry analysis, essentially as described previously.23, 29

antibodies, essentially as described,30 including polyclonal rabbit antisera specific for FADD, DR4, DR5, Bcl-2, Bcl-XL, Mcl-1, Bax,31 and murine monoclonal antibodies specific for FLIP (gift of Dr Marcus Peter (U. Chicago)),32 b-actin (Sigma Immunochemicals), poly-(adenosine 50 -diphosphate-ribose) polymerase PARP (gift of NA Berger),33 and X-linked inhibitor of apoptosis (XIAP) (Transduction Laboratories, San Diego, CA, USA). Immunodetection was accomplished using horseradishperoxidase-conjugated secondary antibodies and an enhanced chemiluminescence (ECL) method (Amersham) involving exposure to X-ray film (Kodak XAR, Rochester, NY, USA). Multiple reprobings of blots were achieved by the MAD method.30

2123

Statistical analysis Data were compared for untreated and drug-treated specimens by paired t-test. Correlations with prior therapy and response were accomplished by the Fisher exact test. Comparisons of potentiation of CDDO and CDDO-m toxicity by TRAIL were performed by the analysis of variables (ANOVA) method.

Results

CDDO and CDDO-m exhibit cytotoxic effects on cultured AML cells We tested the effects of CDDO and CDDO-m on three established AML cell lines, including U937, HL-60, and AML-2. When used individually, CDDO and CDDO-m reduced the viability of all AML lines tested in a dose-dependent manner, with effective doses for killing 50% of cells (ED50) within 48 h of B1 and 0.5 mM, respectively (Figure 1a). Consistent with prior reports,6,20–23 cell death in cultures of CDDO- and CDDO-mtreated cells was attributed to apoptosis, based on characteristic cell morphology (cell shrinkage, membrane blebbing) and nuclear fragmentation, and chromatin condensation (data not shown). The cytotoxic effects of CDDO and CDDO-m were also tested using primary cultured AML and BC-CML blasts. Leukemia cells were derived from seven previously untreated AML and three relapsed CML/blast crisis patients. Among the 10 patient specimens tested, half demonstrated substantial increases in cell death, with 425% of the cells dying within 20 h of exposure to 1 mM of either CDDO or CDDO-m (Figure 1b). Cytotoxic responses to CDDO-m were superior to CDDO in all patients specimens tested (P ¼ 0.03 by t-test). No correlation was found between in vitro responsiveness to CDDO or CDDO-m, and patient response to induction chemotherapy cytogenetics, WBC at diagnosis, gender or FAB category (not shown).

Immunoblot analysis Cell lysates were prepared in RIPA buffer (10 mM Tris (pH 7.4), 150 mM NaCl, 1% Triton X-100, 1% deoxycholate, 0.1% sodium dodecyl sulfate (SDS), 5 mM EDTA, plus protease inhibitors, including phenyl-methyl sulfonyl fluoride, aprotinin, leupeptin, benzamide, and pepstatin), normalized for total protein content (25 mg), and subjected to SDS-polyacrylamide gel electrophoresis (SDS-PAGE) (4–20% gradient gels), followed by transfer to nitrocellulose filters (0.45 mM) (BioRAD, Inc., Richmond, CA, USA). Blots were incubated with primary

CDDO induces proteolytic processing of caspases in AML cells An immunoblotting approach was used to assess the effects of CDDO on processing of caspase-family cell death proteases. Consistent with apoptosis-associated proteolytic activation, CDDO induced consumption of the proforms of the upstream protease, caspase-8, and the downstream effector proteases, caspase-3 and -7 in AML cell lines, whereas procaspase-9 levels remained unchanged (Figure 2). Among these cell death Leukemia


2124

a

a Viability (%)

Viability (%)

80 60 40

80

0

1

2

5 (µM)

C8 C7

40

Pro-C3

0.1

0.5

1

0

2

0.1

CDDO (µM)

0.5

1

2

CDDO-me (µM)

AML-2 HL-60

Active-C3

b

U937

100

b CDDO (1µM) CDDO-me (1µM) 75

Viability(%)

80

100

CELL DEATH (%)

2

0

0

60 40 20

50

0 0

5

CDDO(µM)

25

*

NT

0 1

2

3

4

5

6

7

8

9

10

PATIENT SAMPLE Figure 1 Cytotoxic effects of CDDO and CDDO-m on AML cell lines. (a) The AML cell lines U937, HL-60, and AML-2 were cultured at 5 104 cells/well with various concentrations of CDDO or CDDO-m. Percentage cell viability was measured 48 h later by MTT assay. Data represent mean7s.d. (n ¼ 3), and are representative of several experiments. (b) Leukemia blasts from 10 patients were cultured as above, with 1 mM CDDO (black bars) or CDDO-m (gray bars), measuring percentage cell viability after 20 h. NT ¼ not tested.

proteases, the processed (active) form accumulated to detectable levels only for caspase-3, where the B20 kDa large subunit was observed in AML cells treated with the highest dose of CDDO. The concentrations of CDDO required to induce processing of downstream effector caspases (caspase-3, -7) correlated with the concentrations required for inducing cell death, based on side-by-side comparisons. In contrast, processing of the upstream protease caspase-8 tended to occur at slightly lower concentrations of CDDO than those that produced substantial loss of cell viability (Figure 2).

CDDO reduces FLIP protein levels in AML cell lines We explored the effects of CDDO treatment of the steady-state levels of several apoptosis-relevant proteins in the HL-60 and U937 cell lines, using immunoblotting methods. CDDO at 1 mM induced rapid decreases in the levels of antiapoptotic FLIP protein in both HL-60 and U937 cells, which were detectable within 3 h of treatment (Figure 3). Reductions in the antiapoptotic Bcl-2-family member Mcl-1 were also observed in HL-60 Leukemia

1

60

20

20

0

CDDO: C9

100

100

Figure 2 CDDO induces caspase processing in AML cells. AML cells were cultured at 0.5 106 cells/ml with various concentrations of CDDO for 48 h. (a) Cell lysates were prepared from 2 106 cells, normalized for total protein content (25 mg), and analyzed by immunoblotting, using antibodies specific for caspase-3, -7, -8, and -9. Bands corresponding to the inactive proforms of caspase-7, -8, and -9 are shown. After activation, the processed forms of caspase-7 and -8 were not detectable with these antibodies (not shown). In contrast, for caspase-3, both the proform and a B20 kDa band corresponding to processed (active) caspase-3 were detectable. (b) Percentage cell viability was determined by the Annexin V method at 48 h (mean7s.d. (n ¼ 3)). (Differences in the relative amounts of CDDO required to kill AML cells in Figures 1 and 2 are probably attributable to compound age.)

cells, but occurred later and were less pronounced. In contrast, no changes were observed in the levels of related proteins, Bcl-2, Bcl-XL, and Bax. CDDO also induced slight increases in TRAIL receptors (DR4 and DR5) in U937 cells but not HL-60 cells, and a transient increase in FADD levels in HL-60 but not U937 cells (Figure 3). Insufficient AML blasts were available for extending these studies from established cell lines to freshly isolated leukemia cells. The concentration dependence of CDDO- and CDDO-minduced reductions in FLIP protein levels was also examined in HL-60 and U937 cells. CDDO and CDDO-m were effective at reducing FLIP protein levels at concentrations X0.1 mM (Figures 4a, b). Though the specific dose required for achieving essentially complete disappearance of FLIP varied among cell lines, generally B1 mM was adequate. To determine whether CDDO-mediated reductions in FLIP occur as a result of caspase activation induced by this compound, we treated cells with the broad-spectrum caspase inhibitor zVAD-fmk.34 In cells treated with zVAD-fmk, CDDO still reduced FLIP protein levels (Figure 4b). In contrast, CDDO-induced proteolytic processing of caspase-3 was blocked by zVAD-fmk, serving as a control to


2125

a

a

b HL-60 0

3

6

HL-60

U937 CDDO (1 µM)

CDDO (1µM) 9

0

12 (hr)

FLIP

FLIP

Mcl-1

DR 4

FADD

DR 5

α -Tubulin

FADD

3

6

9

CDDO

12 (hr)

0

0.1

0.3

0.5

1.00 (µM)

FLIP α -Tubulin CDDO-me 0

Bcl-2

0.1

0.5 1.0

2.0 (µ M)

FLIP

Bcl-XL

α -Tubulin

Bax α-Tubulin

b

Figure 3 CDDO induces reductions in FLIP protein levels in AML cells. HL-60 or U937 cells were cultured at 0.5 106 cells/ml in the presence of 1 mM CDDO for various times, as indicated. Cell lysates were prepared, normalized for total protein content, and analyzed by immunoblotting using antibodies specific for FLIP, Mcl-1, FADD, a-Tubulin, DR4, DR5, Bcl-2, Bcl-XL, and Bax. Only those proteins that displayed a change after CDDO treatment (FLIP; Mcl-1; FADD) are shown for HL-60 cells.

U937 CDDO 0.1

0.5

1.0

2.0

(µM)

FLIP α -Tubulin CDDO-me 0

0.1

0.5 1.0

2.0 (µM)

FLIP

Since FLIP functions as an endogenous antagonist of caspase-8 activation in the ‘extrinsic’ pathway for apoptosis,8 we explored the possibility that these synthetic triterpenoids might be capable of sensitizing AML cells to TNF-family death ligands. We chose to study the TNF-family member TRAIL for these studies, because it is poised for human clinical trials, and because agonistic antibodies targeting TRAIL receptors have already entered clinical testing for treatment of cancer.35 Consistent with prior reports,36 treatment of HL-60 or U937 cells with TRAIL alone failed to induce cell death, with essentially 100% of the leukemia cells still surviving after 24 h (Figure 5). In contrast, combined treatment with TRAIL and either CDDO or CDDO-m resulted in enhanced apoptosis, compared to either agent used alone. For both HL-60 and U937 cells, addition of TRAIL shifted the kill curves for CDDO and CDDO-m to the left, lowering the concentrations of the triterpenoids required to induce cell death (Figure 5) (Po0.001 for all experiments, as determined by two-way ANOVA). The combination of TRAIL and CDDO also produced synergistic effects in terms of caspase-3 processing (measured by immunoblot analysis) and apoptosis induction (measured by Annexin V/propidium iodide staining) (Figure 6). The ability of CDDO or CDDO-m to collaborate with TRAIL in apoptosis induction, however, was not uniformly observed for leukemia cell lines. For example, while AML-2 cells were sensitive to CDDO and CDDO-m, addition of TRAIL did not modulate the dose–response, despite reducing FLIP protein levels (Figure 7). Also, the CML blast crisis cell line BC-CML was resistant to apoptosis induced by TRAIL, CDDO, CDDO-m, or combinations of these reagents (not shown).

c

FLIP zVAD

CDDO

U937 CDDO

CDDO and CDDO-m sensitize some leukemia cell lines to TRAIL-induced apoptosis

α -Tubulin

Control

verify effective protease suppression. We conclude therefore that CDDO reduces FLIP protein levels through a caspaseindependent mechanism.

+

+

-

Pro-C-3

Active C-3 α -Tubulin

Figure 4 Concentration-dependent, caspase-independent downregulation of FLIP in AML cells treated with CDDO or CDDO-m. (a) HL-60 and (b) U937 cells were cultured at 1 106 cells/ml with various concentrations of CDDO (top) or CDDO-m (bottom) for 12 h. Cell lysates were prepared, normalized for total protein content, and analyzed by immunoblotting using antibodies specific for FLIP or a-tubulin. (c) U937 cells were cultured without ‘control’ or with 0.5 mM CDDO, in the presence ( þ ) or absence () of 100 mM zVADfmk. Cell lysates were prepared 12 h later, normalized for total protein, and analyzed by immunoblotting using antibodies specific for FLIP (top), caspase-3 (middle), or a-tubulin (bottom). The bands corresponding to the proform and active large subunit of processed caspase3 (C-3) are indicated.

Since analysis of TRAIL knockout mice has demonstrated an important role for this cytokine in eliminating autoreactive thymocytes and T cells,37,38 we also tested the effects of the combination of TRAIL and CDDO or CDDO-m on apoptosis of peripheral blood lymphocytes (PBL), which are comprised mostly of T cells. Unlike leukemia cells, PBL cell death was not induced by individually TRAIL alone, CDDO, CDDOm, or by the combination of TRAIL and either CDDO or CDDOm (Figure 8). Thus, while preliminary, these findings, together Leukemia


2126

a HL-60

HL-60 100

100

Viability(%)

- TRAIL

Viability(%)

80 60 40

+ TRAIL

20 0

- TRAIL

80 60 40

+ TRAIL

20 0

0

0.1

0.2

0.5

1

2

5

0

0.05

CDDO (µM)

0.1

b

0.5

1

2

U937

U937 100

100

- TRAIL

Viability(%)

Viability(%)

0.2

CDDO-me (µM)

80 60

+ TRAIL

40 20

- TRAIL 80 60

+ TRAIL 40 20

0

0.1

0.2

0.3

0.5

1

2

0

0.1

CDDO (µM)

0.2

0.3

0.5

1

2

CDDO-me (µM)

Figure 5 CDDO and CDDO-m sensitize AML cells to TRAIL. HL-60 (top) and U937 (bottom) cells were cultured at 0.5 106 cells/ml for 12 h with various concentrations of CDDO (left) or CDDO-m (right), with ( þ ) or without () 100 ng/ml TRAIL. Relative numbers of viability cells were measured by the MTT method, expressing data as a percentage relative to control untreated cells (mean7s.d.; (n ¼ 3)).

CDDO: TRAIL:

− −

HL-60

+ −

+ +

− + %Viability

TRAIL (+)

Active C-3 α-Tubulin 80 Apoptosis (%)

TRAIL (−)

80

Pro-C-3

60 40 20

60

0 0

40

0

1

2

3

4

CDDO (µM)

20

CDDO TRAIL

0

− −

+ −

+ +

− +

Figure 6 Combination of TRAIL and CDDO induces caspase activation and apoptosis in AML cells. HL-60 cells were cultured for 24 h at 5 104 cells/well with or without 1 mM CDDO, 0.1 mg/ml TRAIL, or combinations of these reagents. (top) Cell lysates were prepared, normalized for total protein content, and analyzed by immunoblotting using antibodies specific for caspase-3 and a-tubulin. Bands corresponding to the proform (inactive) and processed (active) forms of caspase-3 are indicated. Note that the combination of TRAIL and CDDO was more effective than either agent individually at inducing consumption of procaspase-3 and production of the active form of the protease. (bottom) The percentage of apoptotic cells was determined by Annexin V/propidium iodide staining (mean7s.d. (n ¼ 3)) after 24 h culture with TRAIL, CDDO, or the combination of these agents. Leukemia

AML-2

100

1

2

4

(µM)

FLIP β-Actin Figure 7 CDDO fails to sensitize AML-2 cells to TRAIL. AML-2 cells were cultured with various concentrations of CDDO, with ( þ ) or without () 100 ng/ml TRAIL. (Top) Percentage cell viability was measured after 18 h by the Annexin V method (mean7std. dev; [n ¼ 3]). (Bottom) Cell lysates were prepared, normalized for total protein content, and analyzed by immunoblotting using antibodies specific for FLIP or b-actin.

with prior reports,22 suggest that these synthetic triterpenoids display selectivity for malignant over normal hematopoietic cells, even when used in combination with TRAIL.


2127

a Viable Cells (%Control)

120 100 80 60 40

CDDO CDDO+TRAIL (0.1) CDDO+TRAIL (0.2)

20 0 0

0.1 0.2

0.3

0.5 1.0

2.0

CDDO (µM) b Viable Cells (%Control)

120 100 80 60 40

CDDO-me CDDO-me+TRAIL (0.1) CDDO-me+TRAIL (0.2)

20 0 0

0.1

0.2

0.3

0.5

1.0

2.0

CDDO-me (µM) Figure 8 Combination of TRAIL and CDDO or CDDO-me is not toxic to PBLs. PBLs were cultured for 24 h at 5 104 cells/well with or without TRAIL at 0.1 mg/ml (open circles) and 0.2 mg/ml (squares), in combination with various concentrations of CDDO (left) or CDDOme (right). Cell viability was measured by the MTT assay, expressing data a percentage relative to untreated control cells (mean7s.d.; (n ¼ 3)).

Discussion Though use of cytosine arabinoside (Ara-C), anthracycline-type anticancer drugs (doxorubicin), and toposiomerase inhibitors (etoposide) is often successful at inducing remissions in patients with AML, most adult patients relapse with chemoresistant disease.39,40 Interest in supplementing or supplanting cytotoxic anticancer drugs with biological response modifiers has emerged from early successes with all-trans-retinoid acid (ATRA) in AML M3-type, but relapses following treatment with this agent in non-M3-type AMLs are common.41 Consequently, overall survival has not significantly improved for more than two decades for adult patients suffering from most types of AML. Triterpenoids represent a class of naturally occurring and synthetic compounds with demonstrated antitumor activity in rodent models of chemoprevention.19 These biological response modifiers can have pleiotrophic effects, modulating the activity or expression of a variety of signaling molecules, including PPARg, IkB kinases, Cox2, and nitric oxidase synthase (NOS).4,42 Attempts to generate more potent and selective

analogues have resulted in CDDO as the prototype compound for consideration for use in the treatment of cancer. CDDO and some of its derivatives are known to induce differentiation and apoptosis of leukemia cell lines.20,21 Prior studies of the AML cell line U937, the osteosarcoma Saos2, and freshly isolated CLL B-cells have suggested that CDDO triggers an apoptosis pathway characterized by activation of caspase-8, thereby emulating the effects of TNF-family death ligands.20,23 In contrast, caspase-9 activation is not generally observed in CDDO-treated cells, and proteins that interrupt the mitochondrial pathway for apoptosis do not routinely confer resistance to this synthetic triterpenoid or its methyl ester derivative.20,22,23 Since most anticancer drugs predominantly induce apoptosis through the intrinsic (mitochondrial) pathway, CDDO and related compounds may provide an alternative route to eradicating chemorefractory malignant cells. During attempts to delineate the apoptotic mechanism of CDDO and CDDO-m, we have discovered that these compounds reproducibly trigger dose-dependent reductions in the levels of the antiapoptotic protein FLIP.22,23 Since FLIP operates principally as an endogenous antagonist of caspase-8 in cells,8 the loss of FLIP expression seen in CDDO-treated cells may facilitate activation of the extrinsic apoptosis pathway. Thus, FLIP presumably represents one of the critical targets of CDDO and CDDO-m, though changes in the expression of other genes are also likely and may contribute to the cytotoxic activity of these compounds on malignant cells. In this report, we extended preclinical studies of CDDO and CDDO-m in the preparation for potential clinical trials of these agents in adult AML patients, which are planned to begin this year in collaboration with the National Cancer Institute’s developmental therapeutics branch. When tested as single agents, CDDO and CDDO-m displayed cytotoxic activity in cultures of three AML cell lines, and about half of the primary AML /CML blast specimens tested. With both established cell lines and primary leukemia blasts, CDDO-m was more potent than CDDO, by as much as a half-log. In contrast, preclinical comparisons of CDDO and CDDO-m on freshly isolated CLL specimens have suggested that CDDO may be more potent than CDDO-m at inducing apoptosis in vitro.23 These differences in the relative potencies of CDDO and CDDO-m on AML vs CLL cells suggest that the signal transduction and transcriptional pathways modulated by these compounds, or the cellular responses to these pathways, differ depending on cell context. The molecular determinants of responsiveness or lack thereof to apoptosis induced by CDDO and CDDO-m among AML specimens remain to be determined, but, if known, could assist with patient selection for clinical trials. In addition to exhibiting single-agent cytoxicity in vitro against AML cells, CDDO and CDDO-m also were capable of sensitizing some AMLs to the biological agent TRAIL. This observation is similar to our prior results obtained with solidtumor cell lines,22 and is consistent with the hypothesis that the triterpenoids reduce resistance to signaling through the extrinsic pathway by virtue of their ability to downregulate FLIP. TRAIL is a member of the TNF family that induces apoptosis of cancer cells in vitro and in xenograft models, while having little activity against normal cells.9 While recombinant TRAIL may soon enter human clinical trials, an agonistic antibody that targets the apoptosis-inducing TRAIL-receptor, TRAIL-R1 (DR4), has recently entered Phase I trials for cancer (http://hgs.com) and preclinical evaluations of a monoclonal antibody that triggers the TRAIL-receptor TRAIL-R2(DR5) have been reported.43 Previous studies have documented intrinsic resistance of AML Leukemia


2128

cells to TRAIL-induced apoptosis,44 raising concerns about the efficacy of TRAIL-based biological therapies. Indeed, among the 10 freshly isolated samples of AML that we evaluated here for sensitivity, none was responsive to apoptosis induction by TRAIL (not shown). When used in combination with CDDO or CDDOm, however, sensitivity to TRAIL was restored in two of four leukemia lines tested in vitro. But, the failure of some AML cell lines to respond to TRAIL, even though FLIP was successfully downregulated, indicates that combination therapies that include TRAIL and a synthetic triterpenoid are likely to be suitable only for a subset of AML patients. Similar observations were made for CLL, where CDDO failed to sensitize some CLL B-cell samples to TRAIL, despite downregulating FLIP.23 Thus, the conditions required for achieving sensitization to TRAIL are likely to be complex, extending beyond FLIP, and probably including differences in the relative levels of apoptosis-inducing TRAIL receptors, decoy receptors, and various components of the intracellular signaling machinery that link TRAIL receptors to caspase activation pathways.9 Also, it remains to be determined whether CDDO and CDDO-m will have deleterious effects on sensitizing normal cells to TRAIL. Thus far, however, we have not seen apoptosis induction of normal hepatocytes, endothelial cells, bone marrow cells, lymphocytes, or fibroblasts when treated in vitro with the combination of TRAIL and either CDDO or CDDO-m (see above).22 The ability of CDDO and CDDO-m to reproducibly induce marked decreases in FLIP protein in a rapid and concentrationdependent manner in AML cells and other types of malignant cells suggests that FLIP could be exploited as a pharmacodynamic end-point for adjusting doses of these agents in human clinical trials. In contrast to traditional cytotoxic anticancer drugs, biological response modifiers such as synthetic triterpenoids may not produce the rate-limiting toxicities traditionally used for setting dosing schedules. A pharmacodynamic measurement such as FLIP may allow one to sample peripheral blood or bone marrow leukemia cells, determining when doses sufficient for downregulation of FLIP have been achieved. Alternative pharmacodynamic end-points can be considered once the primary molecular target of CDDO and CDDO-m is known.

References 1 Reed JC. Bcl-2 family proteins: regulators of apoptosis and chemoresistance in hematologic malignancies. Semin Hematol 1997; 34: 9–19. 2 Makin G, Hickmann JA. Apoptosis and cancer chemotherapy. Cell Tissue Res 2000; 301: 143. 3 Honda T, Rounds BV, Gribble GW, Suh N, Wang Y, Sporn MB. Design and synthesis of 2-cyano-3,12-dioxoolean-1,9-dien-28-oic acid, a novel and highly active inhibitor of nitric oxide production in mouse macrophages. Bioorg Med Chem Lett 1998; 8: 2711–2714. 4 Suh N, Honda T, Finlay HJ, Barchowsky A, Williams C, Benoit NE et al. Novel triterpenoids suppress inducible nitric oxide synthase (iNOS) and inducible cyclooxygenase (COX-2) in mouse macrophages. Cancer Res 1998; 58: 717–723. 5 Suh N, Wang Y, Honda T, Gribble GW, Dmitrovsky E, Hickey WF et al. A novel synthetic oleanane triterpenoid, 2-cyano-3,12dioxoolean-1,9-dien-28-oic acid, with potent differentiating, antiproliferative, and anti-inflammatory activity. Cancer Res 1999; 59: 336–341. 6 Ito Y, Pandey P, Place A, Sporn MB, Gribble GW, Honda T et al. The novel triterpenoid 2-cyano-3,12-dioxoolean-1,9-dien-28-oic acid induces apoptosis of human myeloid leukemia cells by a caspase-8-dependent mechanism. Cell Growth Differ 2000; 11: 261–267. 7 Salvesen GS. Caspases: opening the boxes and interpreting the arrows. Cell Death Differ 2002; 9: 3–5. Leukemia

8 Tschopp J, Irmler M, Thome M. Inhibition of Fas death signals by FLIPs. Curr Opin Immunol 1998; 10: 552–558. 9 Ashkenazi A, Dixit VM. Apoptosis control by death and decoy receptors. Curr Opin Cell Biol 1999; 11: 255–260. 10 Schimmer A, Hedley DW, Penn LZ, Minden MD. Receptor- and mitochondrial-mediated apoptosis in acute leukemia: a translational view. Blood 2001; 98: 3541–3553. 11 Andreeff M, Tabe Y, Milella M, Carter BZ, Tsa T, McQueen T et al. The role of apoptosis and cell signaling in myeloid leukemia. Cytometry Res 2002; 12: 20–21. 12 Green DR, Reed JC. Mitochondria and apoptosis. Science 1998; 281: 1309–1312. 13 Reed J. Dysregulation of apoptosis in cancer. J Clin Oncol 1999; 17: 2941. 14 Wallach D, Kovalenko AV, Varfolomeev EE, Boldin MP. Death-inducing functions of ligands of the tumor necrosis factor family: a Sanhedrin verdict. Curr Opin Immunol 1998; 10: 279–288. 15 Muzio M, Chinnaiyan AM, Kischkel FC, O’Rourke K, Shevchenko A, Ni J et al. FLICE, a novel FADD-homologous ICE/CED-3-like protease, is recruited to the CD95 (Fas/APO-1) death-inducing signaling complex. CELL 1996; 85: 817–827. 16 Boldin MP, Goncharov TM, Goltsev YV, Wallach D. Involvement of MACH, a novel MORT1/FADD-interacting protease, in Fas/APO-1- and TNF receptor-induced cell death. Cell 1996; 85: 803–815. 17 Li P, Nijhawan D, Budihardjo I, Srinivasula SM, Ahmad M, Alnemri ES et al. Cytochrome c and dATP-dependent formation of Apaf-1/caspase-9 complex initiates an apoptotic protease cascade. Cell 1997; 91: 479–489. 18 Hakem R, Hakem A, Duncan GS, Henderson JT, Woo M, Soengas MS et al. Differential requirement for caspase 9 in apoptotic pathways in vivo. Cell 1998; 94: 339–352. 19 Place AE, Suh N, Williams CR, Risingsong R, Honda T, Honda Y, Gribble GW, Leesnitzer LM, Stimmel JB, Willson TM, Rosen E, Sporn MB. The novel synthetic triterpenoid, CDDO-Imidazolide, inhibits inflammatory response and tumour growth in vivo. Clin Cancer Res 2003; 9: 2798–2806. 20 Ito Y, Pandey P, Sporn MB, Datta R, Kharbanda S, Kufe D. The novel triterpenoid CDDO induced apoptosis and differentiation of human osteosarcoma cells by a caspase-8 dependent mechanism. Mol Pharmacol 2001; 59: 1094–1099. 21 Konopleva M, Tsao T, Ruvolo P, Stiouf I, Estrov Z, Leysath CE et al. Novel triterpenoid CDDO-Me is a potent inducer of apoptosis and differentiation in acute myelogenous leukemia. Blood 2002; 99: 326–335. 22 Kim Y, Suh N, Sporn M, Reed JC. An inducible pathway for degradation of FLIP protein sensitizes tumor cells to TRAILinduced apoptosis. J Biol Chem 2002; 277: 22320–22329. 23 Pedersen IM, Kitada S, Schimmer A, Kim Y, Zapata JM, Charboneau L et al. The triterpenoid CDDO induces apoptosis in refractory CLL B-cells. Blood 2002; 100: 2965–2972. 24 Chu P, Deforce D, Pedersen IM, Kim Y, Kitada S, Reed JC et al. Latent sensitivity to Fas-mediated apoptosis after CD40 ligation may explain activity of CD154 gene therapy in chronic lymphocytic leukemia. Proc Natl Acad Sci USA 2002; 99: 2854–3859. 25 Armstrong RC, Aja T, Xiang J, Gaur S, Krebs JF, Hoang K et al. Fasinduced activation of the cell death related protease CPP32 is inhibited by Bcl-2 and by ICE family protease inhibitors. J Biol Chem 1996; 271: 16850–16855. 26 Rayappa C, McCulloch EA. A cell culture model for the treatment of acute myeloblastic leukemia with fludarabine and cytosine arabinoside. Leukemia 1993; 7: 992–999. 27 Hu Z-B, Minden MD, McCulloch EA. Regulation of the synthesis of bcl-2 protein by growth factors. Leukemia 1996; 10: 1925–1929. 28 Cheson BD, Cassileth PA, Head DR, Schiffer CA, Bennett JM, Bloomfield CD et al. Report of the National Cancer Institute-sponsored workshop on definitions of diagnosis and response in acute myeloid leukemia. J Clin Oncol 1990; 8: 813–819. 29 Schimmer AD, Munk-Pedersen I, Kitada S, Demiralp E, Minden MD, Pinto R et al. Functional blocks in caspase activation pathways are common in leukemia and predict


2129 30 31 32 33

34

35 36

37 38

patient-response to induction chemotherapy. Cancer Res 2003; 63: 1242–1248. Krajewski S, Zapata JM, Reed JC. Detection of multiple antigens on Western blots. Anal Biochem 1996; 236: 221–228. Krajewska M, Krajewski S, Epstein JI, Shabaik A, Sauvageot J, Song K et al. Immunohistochemical analysis of Bcl-2, Bax, Bcl-X and Mcl-1 expression in prostate cancers. Am J Pathol 1996; 148: 1567–1576. Stegh AH, Barnhart BC, Volkland J, Ke N, Reed J, Peter ME. Inactivation of caspase-8 on mitochondria of Bcl-xL expressing MCF7-Fas cells. J Biol Chem 2002; 277: 4351–4360. Whitacre CM, Berger NA. Factors affecting topotecan-induced programmed cell death: adhesion protects cells from apoptosis and impairs cleavage of poly(ADP-ribose)polymerase. Cancer Res 1997; 57: 2157–2163. Thornberry NA, Rano TA, Peterson EP, Rasper DM, Timkey T, Garcia-Calvo M et al. A combinatorial approach defines specificities of members of the caspase family and granzyme B. J Biol Chem 1997; 272: 17907–17911. Reed JC. Apoptosis-targeted therapies for cancer. Cancer Cell 2003; 3: 17–22. Bortul R, Tazzari PL, Cappellini A, Tabellini G, Billi AM, Bareggi R et al. Constitutively active Akt1 protects HL60 leukemia cells from TRAIL-induced apoptosis through a mechanism involving NF-kappaB activation and cFLIP(L) up-regulation. Leukemia 2003; 17: 379–389. Lamhamedi-Cherradi SE, Zheng SJ, Maguschak KA, Peschon J, Chen YH. Defective thymocyte apoptosis and accelerated autoimmune diseases in TRAIL(/) mice. Nat Immunol 2003; 4: 255–261. Cretney E, Takeda K, Yagita H, Glaccum M, Peschon JJ, Smyth MJ. Increased susceptibility to tumor initiation and metastasis in

39

40

41

42

43

44

TNF-related apoptosis-inducing ligand-deficient mice. J Immunol 2002; 168: 1356–1361. Leith CP, Kopecky KJ, Godwin J, McConnell T, Slovak ML, Chen IM et al. Acute myeloid leukemia in the elderly: assessment of multidrug resistance (MDR1) and cytogenetics distinguishes biologic subgroups with remarkably distinct responses to standard chemotherapy. A Southwest Oncology Group Study. Blood 1997; 89: 3323–3329. Slovak ML, Kopecky KJ, Cassileth PA, Harrington DH, Theil KS, Mohammed A et al. Karyotypic analysis predicts outcome of preremission and postremission therapy in adult acute myeloid leukemia: A Southwest Oncology Group/Eastern Cooperative Oncology Group Study. Blood 2000; 96: 4075–4083. Bassan R, Lerede T, Di Bona E, Rambaldi A, Rossi G, Pogliani E et al. Induction–consolidation with idarubicin-containing regimen, unpurged marrow autograft, and post-graft chemotherapy in adult acute lymphoblastic leukaemia. Br J Hematol 1999; 104: 755–762. Wang Y, Porter WW, Suh N, Honda T, Gribble GW, Leesnitzer LM et al. A synthetic triterpenoid, 2-cyano-3,12-dioxooleana-1,9-dien28-oic acid (CDDO), is a ligand for the peroxisome proliferatoractivated receptor g. Mol Endocrinol 2000; 14: 1550–1556. Ichikawa K, Liu W, Wang Z, Liu D, Zhao L, Ohtsuka T et al. Tumoricidal activity in the absence of hepatocyte cytoxicity of a novel anti-human DR5 monoclonal antibody. Nat Med 2001; 7: 954–960. Wuchter C, Krappmann D, Cai Z, Ruppert V, Scheidereit C, Dorken B et al. In vitro susceptibility to TRAIL-induced apoptosis of acute leukemia cells in the context of TRAIL receptor gene expression and constitutive NF-kappaB activity. Leukemia 2001; 15: 921–928.

Leukemia