stichoposidecinducesapoptosisthroughthegenerationof ...was evaluated by annexin v-ﬂuorescein...

Cancer Therapy: Preclinical

Stichoposide C Induces Apoptosis through the Generation ofCeramide in Leukemia and Colorectal Cancer Cells andShows In Vivo Antitumor Activity

Seong-Hoon Yun1, Eun-Seon Park1, Sung-Won Shin1, Yong-Woo Na1, Jin-Yeong Han2, Jin-Sook Jeong3,Valeria V. Shastina4, Valentin A. Stonik4, Joo-In Park1, and Jong-Young Kwak1

AbstractPurpose: Marine triterpene glycosides that are physiologically active natural compounds isolated from

sea cucumbers (holothurians) and sponges have antifungal, cytotoxic, and antitumor activities, whose

specific molecular mechanisms remain to be elucidated. In this study, we examined if and through which

mechanisms stichoposide C (STC) from Thelenota anax (family Stichopodidae) induces apoptosis in

leukemia and colorectal cancer cells.

Experimental Design: We examined STC-induced apoptosis in human leukemia and colorectal cancer

cells in the context of mitochondrial injury and signaling pathway disturbances, and investigated the

antitumor effect of STC in mouse CT-26 subcutaneous tumor and HL-60 leukemia xenograft models.

Results:We found that STC induces apoptosis in these cells in a dose-dependentmanner and leads to the

activation of Fas and caspase-8, cleavage of Bid, mitochondrial damage, and activation of caspase-3. STC

activates acid sphingomyelinase (SMase) and neutral SMase, which resulted in the generation of ceramide.

Specific inhibition of acid SMase or neutral SMase and siRNA knockdown experiments partially blocked

STC-induced apoptosis. Moreover, STC markedly reduced tumor growth of HL-60 xenograft and CT-26

subcutaneous tumors and increased ceramide generation in vivo.

Conclusions: Ceramide generation by STC, through activation of acid and neutral SMase, may in part

contribute to STC-induced apoptosis and antitumor activity. Thus, STCmay have therapeutic relevance for

human leukemia and colorectal cancer. Clin Cancer Res; 18(21); 5934–48. �2012 AACR.

IntroductionImprovement of cure rates of leukemia and colorectal

cancer require the identification of potentially new thera-peutic agents. Acute myeloid leukemia (AML), whichaccounts for 80% of all acute leukemias in adults, is grad-ually increasing in incidence (1), and although 60% to 80%of all treated AML patients achieve initial remission, the 5-year survival rate is only 15% in adults (2). StandardAML therapy requires intensive combination chemotherapy(1), which leads to significant treatment-related toxicity

especially in elderly patients. Colorectal cancer is the thirdmost common cancer in the world (3) and although newchemotherapeutic agents have improved prognosis (4),complete remission after surgery or chemotherapy remainsdifficult to achieve in advanced disease.

Marine triterpene glycosides are physiologically activenatural compounds isolated from sea cucumbers(holothurians) and some sponges, which show a widespectrum of biologic activities, including antifungal, anti-tumor, hemolytic, and cytostatic against various tumor cells(5, 6). In particular, stichoposides from sea cucumbers haveantifungal, cytotoxic, and antitumor activities (7) throughundefined molecular mechanisms. Stichoposide C (alsoknown as stichloroside C1; STC; Fig. 1A) is a quinovose-containing hexaoside first isolated from the holothurianStichopus chloronotus (8, 9) and later found in other Sticho-podidae such as Thelenota ananas (10). It was suggested thatSTC antitumor activity was mediated by membranotropiceffects (7).

Ceramide, which is involved in regulating many diversecellular processes, including cell-cycle arrest, apoptosis,senescence, and stress responses (11), is generated eitherby de novo synthesis or by sphingomyelin hydrolysis(12, 13). There are several types, namely acid, neutral, oralkaline, of sphingomyelinases (SMase) that hydrolyze

Authors' Affliations: 1Department of Biochemistry, Departments of 2Lab-oratory Medicine and 3Pathology, Dong-A University College of Medicine,Busan, South Korea; and 4Pacific Institute of Bioorganic Chemistry, FarEast Division, The Russian Academy of Sciences, Vladivostok, Russia

Note: Supplementary data for this article are available at Clinical CancerResearch Online (http://clincancerres.aacrjournals.org/)

Seong-Hoon Yun and Eun-Seon Park contributed equally to this work.

CorrespondingAuthors: Joo-In Park, Department of Biochemistry, Dong-A University College of Medicine, 3 Ga 1, Dongdaesin-Dong, Seo-Gu, 602-714, South Korea. Phone: 82-51-240-2881; Fax: 82-51-241-6940; E-mail:[email protected]; and Jong-Young Kwak, E-mail: [email protected]

doi: 10.1158/1078-0432.CCR-12-0655

�2012 American Association for Cancer Research.

ClinicalCancer

Research

Clin Cancer Res; 18(21) November 1, 20125934

sphingomyelin in biologic membranes to release cer-amide (11, 14) depending on their maximum activity atacid, neutral, or alkaline pHs, respectively (15). AcidSMase and neutral SMase are involved in ceramide gen-eration in response to apoptotic stimuli (16–18). Anti-cancer agents increase ceramide levels, to variable extents,in all types of cancer cells (19). In this context, thepharmacologic modulation of sphingolipid metabolismto enhance tumor cell ceramide represents a novelapproach to cancer chemotherapy. Moreover, under con-ditions where the classical apoptotic pathway fails, intra-cellular generation of ceramide may function as part of anancient backup system that enables caspase-independentprogrammed cell death (19).In this study, we explore the effects of Thelenota anax-

derived STC on human leukemia and colorectal cancercell proliferation, and the relationship to perturbations insignaling pathways. We show that STC induces apoptosisand ceramide production in leukemic cells and colorectalcancer cells. Our results suggest that ceramide generation,through activation of acid and neutral SMase, may play arole in STC-induced apoptosis and that STC might be apromising agent for the treatment of leukemia and colo-rectal cancer.

Materials and MethodsCell preparationsHuman leukemic HL-60, K562, THP-1, NB4 cell lines,

human colorectal cancer cell lines (SNU-C4, HT-29), andmurine colorectal cancer CT-26 cell lines were obtainedfrom the Korean Cell Line Bank (Seoul National University,Seoul, Korea) and cultured in RPMI1640 or Dulbecco’sModified Eagle’s Medium supplemented with 10% FBS,100U/mL penicillin, and 100 mg/mL streptomycin. Humanhematopoietic progenitor CD34þ cells were purchasedfrom STEM CELL Technologies and were cultured in

Hematopoietic Progenitor Expansion Medium DXF withcytokine mix E (PromoCell). Human primary leukemiccells were obtained from 5 patients with AML at theDong-A University Hospital, Busan, Korea. Informed con-sent was obtained from all patients before sample collec-tion, according to the Institutional guidelines. Bonemarrowmononuclear cells were isolated by Ficoll-Paque (Amer-sham Biosciences) density-gradient centrifugation. Thepercentage of blasts in the samples ranged between 75%and 91%. The purity of these cells in the samples wasgreater than 95%, which was confirmed by flow cytometricimmunophenotyping usingCD45dim (þ), lineage-specific(myeloid) markers, and side scatter (SSC). These cells werecultured in RPMI1640 medium supplemented with 10%FBS, 100 U/mL penicillin, and 100 mg/mL streptomycin.Cultures were maintained in a humidified atmosphere of95% air/5% CO2 at 37

�C.

ReagentsSTCwas isolated according to the procedure published by

Stonik and colleagues (8). Briefly, specimens of the seacucumber T. anax were collected at the Great Barrier Reefarea, Australia, in September 1988, and were cut andextracted with ethanol at room temperature more than 1week. The extract was evaporated at low pressure to dryness,dissolved in water, and loaded on a column with Teflonpowder (Polychrom-1, Biolar, Latvia) fitted in alcohol andwashed with water before loading. The salts and polarimpurities were washed off with water and then the crudeglycoside fraction was eluted with a mixture of acetone:water (1:1). The obtained eluate was concentrated by vac-uum drying. The crude glycoside fraction was separated byrepeated chromatography on a column with Si gel usingchloroform:methanol:water (75:25:1) to yield STC. Thepurity of STC was confirmed by direct comparison of 1Hand 13C NMR spectra and physical constants as well as byTLC on silica gel plates against a standard sample of theglycoside isolated from S. chloronotus. For experiments, STCwas dissolved in dimethyl sulfoxide (DMSO). Annexin Vwas from BD Biosciences Clontech (Palo Alto). Antiprocas-pase 8, antiprocaspase-3, antiprocaspase 9, and anticyto-chrome c antibodies were purchased from Santa CruzBiotechnology. Antibody against PARPwas purchased fromCell Signaling Technology. The anti-b-actin antibodywas from Sigma. Z-Ile-Glu-Thr-Asp-fluoromethyl ketone(Z-IETD-FMK), Z-Val-Ala-Asp-fluoromethyl ketone (Z-VAD-FMK), Z-Leu-Glu-His-Asp-fluoromethyl ketone (Z-LEHD-FMK), and Z-Asp (OCH3)-Glu (OCH3)-Val-Asp(OCH3)-fluoromethyl ketone (Z-DEVD-FMK) were fromSigma. L-buthionine-(S,R)-sulfoximine (BSO) was fromSigma. Unless stated otherwise, all other chemicals werepurchased from Sigma.

Cell-cycle and apoptosis analysesCell-cycle analysis was carried out on propidium iodide

(PI)-stained samples (20); whereas the extent of apoptosiswas evaluated by annexin V-fluorescein isothiocyanate(FITC) and flow cytometry, as previously described (21).

Translational RelevanceTriterpene glycosides from Stichopus chloronotus and

Thelenota ananas are known to inhibit development oflymphocytic leukemia P388 through poorly character-ized molecular mechanisms. This study found thatstichoposide C (STC) from Thelenota anax induces apo-ptosis of leukemic and colorectal cancer cells via extrinsicand intrinsic pathways. The effect includes ceramidegeneration by activation of acid and neutral SMase bycapase-8 activation and by glutathione depletion andreactive oxygen species production, respectively. STCalso significantly reduced tumor size and upregulatedceramide expression without toxicity in a HL-60 xeno-graft and a CT-26 subcutaneous tumor models. Thesefindings suggested consistency between in vitro and invivo data and potential for STC as an anticancer drug forleukemia and colorectal cancer patients.

In Vitro and In Vivo Antitumor Activity of STC

www.aacrjournals.org Clin Cancer Res; 18(21) November 1, 2012 5935

Yun et al.

Clin Cancer Res; 18(21) November 1, 2012 Clinical Cancer Research5936

Measurement of mitochondrial membrane potentialVariations in mitochondrial membrane potential (Djm)

during the induction of apoptosis were examined withDiOC6 (Molecular Probes), as previously described (21).

Separationof the cytosolic andmitochondrial proteinsCytosolic and mitochondrial proteins were separated as

previously described (20, 22). In brief, cells treated with0.1%DMSO or with STC for indicated times were collectedand resuspended in mitochondrial isolation buffer andprotease inhibitor cocktail (BoehringerMannheim) supple-mented with 10 mmol/L digitonin. Suspensions were incu-bated at 37�C for 5 minutes and centrifuged at 12,000 g for15 minutes. The supernatant (cytosolic fraction) was col-lected forWestern blotting. The pellets were resuspended inthe radioimmunoprecipitation assay buffer with proteaseinhibitor cocktail, incubated at 4�C for 30 minutes withvortexing, and centrifuged at 15,000 rpm for 30 minutes.The supernatant (membrane fraction) was collected forWestern blotting.

Separation of the cytosolic and membrane proteinsCytosolic and membrane proteins were isolated using

ProteoJET membrane protein extraction kit (Fermentas)according to manufacturer’s instructions. Briefly, 5 � 106

cells were pelleted by centrifugation for 5 minutes at 250 g,resuspended in 3 mL of ice-cold cell wash solution andrecentrifuged. Ice-cold cell permeabilization buffer (1.5mL) was added and incubated for 10 minutes at 4�C whilecontinuously rocking. The permeabilized cells were centri-fuged at 16,000 g for 15 minutes at 4�C. The supernatant(cytoplasmic protein extract) was collected for Westernblotting. The pellets were resuspended in ice-cold mem-brane extraction buffer, incubated for 30 minutes at 4�C inthe thermomixer, shaking at 1,400 rpm, and centrifuged at16,000 g for 15 minutes at 4�C. The supernatant (mem-brane protein fraction) was used for Western blotting.

Western blot analysisCell lysis and Western blot analysis were conducted as

described previously (21). A total of 30 mg of protein wasused for immunoblotting. b-actin was used as the loadingcontrol.

Immunofluorescence stainingCells were fixed and permeabilized with 1% formalde-

hyde/methanol in PBS for 10minutes at room temperature.After cells were washed, a series of antibodies was used as

indicated, followed by FITC- or PE-conjugated goat anti-mouse and anti-rabbit IgG (Calbiochem) staining. Thesamples were then mounted using glycerol and analyzedby confocal microscope (Carl Zeiss LSM 510; Carl Zeiss)using a �40 C-Apochromat objective. Negative controlstaining was conducted with secondary antibodies alone.

siRNA transfectionPredesigned siRNA targeted to human acid SMasemRNA

(catalog number SI00011557; ID 6609), neutral SMasemRNA (catalog number SI02655114; ID 6610), and All-Stars negative control siRNA (catalog number 1027280)were obtained from Qiagen. The other siRNA sequencesused for the targeted silencing of acid SMase (siRNAsequence for acid SMase 50GGUCUAUUCACCGCCAU-CAtt30 and 50UGAUGGCGGUGAAUAGACCtt30) or neutralSMase (siRNA sequence for neutral SMase 50UGCUACUG-GCUGGUGGACCtt30 and 50GGUCCACCAGCCAGUAG-CAtt30) were designed by Qiagen. The Fas siRNA (siRNAsequences for Fas siRNA 50GAAGCGUAUGACACAUU-GAtt30 and 50UCAUGUGUCAUACGCUUCtt30, 50CCCAA-ACAUGGAAAUAUCAtt30 and 50UGAUAUUUCCAUGU-UUGGGtt30, and 50GAACCCAUGUUUGCAAUCAtt30 and50UGAUUGCAAAACAUGGGUUCtt30) was obtained fromSantaCruzBiotechnology. For transfection, cellswere resus-pended at 1.3 � 107/0.5 mL in PBS and mixed with 200nmol/L siRNA for acid SMase, neutral SMase, or Fas or with200 nmol/L nonsilencing siRNA. The mixture was thenadded to an electroporation cuvette (0.4-cm electrode gap)and subjected to 300 V and 950 mF (Gene Pulser Xcell Elec-troporation System; Bio-Rad). After electroporation, cellswere cultured in 10% FBS-supplemented RPMI1640 for48hours, then treatedwith 0.1%DMSOor STC for indicatedtimes. These cells were then used for annexin-V staining,immunofluorescence staining, and Western blot analysis.

Assay of intracellular glutathione levelTotal GSH levels in cells treated with 0.1%DMSO or STC

for indicated timesweremeasured by a glutathione assay kit(Sigma) according to manufacturer’s instructions. In brief,the harvested cells were suspended in 3 volumes of the 5%sulfosalicylic acid solution of the packed cell pellet, frozen,and thawed twice. The precipitated proteins were pelletedby centrifugation at 10,000 g for 10minutes at 4�C.Aliquotsof the soluble supernatant were mixed with a workingmixture [100 mmol/L potassium phosphate buffer, pH7.0, with 1 mmol/L EDTA, glutathione reductase, and5,50-dithiobis-(2-nitrobenzoic acid) (DTNB) stock

Figure 1. Stichoposide C (STC) induces apoptosis in leukemic and colorectal cancer cells. A, structure of STC. B, top: HL-60 cells were seeded 4 hours beforetreatment with various concentrations of STC (0, 0.3, 0.5, 1.0, or 1.5 mmol/L) for 6 hours. Bottom: HL-60 cells were seeded 4 hours before treatmentwith STC (0.3 mmol/L) for the indicated times. C, NB4, THP-1, and K562 cells were seeded 4 hours before treatment with STC (0.3, 0.3, or 0.5 mmol/L,respectively) for 6 hours. CT-26, HT-29, and SNU-C4 cells were seeded 24 hours before treatment with STC (0.5, 1.0, 1.5 mmol/L, respectively) for 24 hours.The percentage of apoptotic cells was determined by annexin V-FITC/PI staining as described in theMaterials andMethods. Each top panel is representativeof 3 separate experiments. Each lower panel represents the mean � SD of 3 independent experiments. ��, P < 0.01; and ��, P < 0.001 versus controlcells. D, human primary leukemia cells isolated from 5 patients with acutemyelogenous leukemia, and human normal hematopoietic progenitor (CD34þ) cellswere both treated with various STC concentrations (0, 0.3, 0.5, 1.0, or 1.5 mmol/L) for 6 hours. The percentage of apoptotic cells was determined byannexin V-FITC/PI staining asdescribed in theMaterials andMethods. Left panels show representatives of 3 separate experiments.Right panels represent themean � SD of 3 independent experiments. ��, P < 0.01; and ��, P < 0.001 versus control cells.



Figure 2. Treatment of HL-60 and K562 cells with STC leads to activation of extrinsic and intrinsic pathways. A, HL-60 and K562 cells were treated with0.3 mmol/L STC or 0.5 mmol/L STC for indicated times, respectively. Protein lysates were prepared and subjected to Western blot analysis as described inMaterials and Methods using corresponding antibodies. Equal protein loading was ensured by demonstrating uniform b-actin expression. The blot isrepresentative of 3 separate experiments. B, functional involvement of caspases in STC-induced apoptosis of HL-60 and K562 cells. Cells were pretreatedwith Z-VAD-FMK (pan-caspase inhibitor; 25 mmol/L), Z-IETD-FMK (inhibitor of caspase-8; 20 mmol/L), Z-LEHD-FMK (inhibitor of caspase-9; 20 mmol/L),or Z-DEVD-FMK (inhibitor of caspase-3; 50mmol/L) for 1 hour before treatmentwith STC (0.3 or 0.5mmol/L) for 6 hours. The extent of apoptosiswasmeasuredby flow cytometry after annexin V staining. These data represent the mean� SD of 3 independent experiments. �, P < 0.05; ��, P < 0.01; ��, P < 0.001 versusSTC-treated cells. C, left; HL-60 and K562 cells were treated with 0.3 mmol/L STC or 0.5 mmol/L, respectively for 6 hours. The extent of apoptosis wasmeasured by flow cytometry after annexin V staining. These data represent the mean� SD of 3 independent experiments. ��, P < 0.01; ��, P < 0.001 versus

Yun et al.


solution] for 5 minutes and the diluted NADPH solutionwas added. The increase in absorbance at 412 nm wasmonitored to determine the amount of GSH in the sample.A reference curve was generated with 1 nmol of GSHstandards. The concentration of GSH in the samples wascalculated with the following equation:

nmole GSH/mL of sample ¼ DA412/min ðsampleÞ� dilution factor/DA412/min ð1 nmolÞ � volume:

Assessment of ROS productionReactive oxygen species (ROS) production was moni-

tored by flow cytometry using 20,70-Dichlorodihydrofluor-escein diacetate (DCFH-DA) as described previously (21).In the present study, cells were treated with STC for theindicated times and were washed twice with PBS to removeextracellular compounds. DCFH-DA (100 mmol/L) wasadded for an additional hour. Green fluorescence wasexcited using an argon laser and was detected using a525-nm band-pass filter by flow cytometric analysis.

Establishment of HL-60 leukemia xenograft andmouseCT-26 subcutaneous tumors and experimental designAll animal procedures and care were approved by the

Institutional Animal Care andUsage Committee of Dong-AUniversity. To determine the in vivo activity of STC, viableHL-60 cells (2�107/100mLPBSpermouse) andCT-26 cells(2 � 106/100 mL PBS per mouse), as confirmed by trypanblue staining, were injected into the right flank of 6- to 7-week-old female Balb/c nudemice and Balb/c mice, respec-tively (Orient Bio Inc.) as previously described (20). Whenaverage subcutaneous tumor volume reached 60 to 100mm3, mice were assigned into 2 treatment groups: (i)control and (ii) STC, given at a dose of 7.19 mg/kg via tailvein every 3 days. Control groups were treated with vehicleonly. Tumor size was measured daily with a caliper (calcu-lated volume ¼ shortest diameter2 � longest diameter/2).Mice were followed for tumor size and body weight andwere sacrificed on the 14th or 21st day. Tumors wereresected, weighed, and frozen or fixed in formalin andparaffin embedded for immunohistochemical studies.

Histology and immunohistochemical analysisAt the termination of experiments, tumor tissues were

removed for histologic examination. Sections of tumortissues were stained with hematoxylin and eosin, andimmunohistochemistry was conducted with the DAKOEnVision Kit (DakoCytomation) as described previously(21). Primary antibody against ceramide (1:10 dilution)was applied for 1 hour at room temperature. Next, the

sections were incubated with EnVision reagent (DakoCy-tomation), which is a peroxidase-conjugated polymer back-bone, carrying secondary antibody molecules. The sectionswere lightly counterstained with hematoxylin. After wash-ing with PBS, VectorShield (Vector Laboratories, Burlin-game, CA, USA) mounting medium was applied, and sec-tions were cover slipped and imaged by ScanScope (AperioTechnologies, Inc.).

Statistical analysisStatistical analyses were done with the SPSS 14.0 statis-

tical package for Windows (SPSS). Data are expressed asmean values � SD. One-way ANOVA was applied to deter-mine whether there were significant differences in cellviability between STC-treated and control cells. Differencesin tumor volumes between treated and control groups wereevaluated using the unpaired Student’s t test. Statisticalsignificance was defined as P < 0.05.

ResultsSTC induces apoptosis of leukemic and colorectalcancer cells

To investigate the effect of STC on the growth of HL-60cells, we used fluorescence-activated cell sorting analysis toanalyze the cell-cycle profile. The flow cytometric analysisrevealed no change except for an increase in representationof sub-G1 phase (data not shown). These results suggest thatSTC inhibits proliferation of HL-60 cells by inducing apo-ptosis. To confirm an induction of apoptosis, HL-60 cellswere treatedwith various concentrations of STCover a rangeof times and costained with PI and annexin V conjugatedwith FITC. STC treatment resulted in dose- and time-depen-dent increases in apoptotic cell proportions (Fig. 1B). Toevaluate whether the induction of apoptosis by STC isspecific to HL-60 cells, or a more general effect, the sameexperiment was conducted in other leukemia cell lines(THP-1, K562, and NB4 cells) and colorectal cancer celllines (CT-26, HT-29, and SNU-C4 cells). We also observedthe induction of apoptosis by STC in additional cell linesand human primary leukemia cells, even though the effec-tive concentration of STC used in each cell linewas different(Fig. 1C and D). In contrast, the concentrations of STC thatwere used in this study (0.3–1.5 mmol/L) did not increaseapoptosis of normal human hematopoietic progenitor cellscompared with control, as further confirmed by annexin-V/PI staining (Fig. 1D).

STC activates both extrinsic and intrinsic pathwaysThere are 2 primary pathways for the activation of cas-

pases—one triggered by receptor activation on the cell

control cells. Right; HL-60 andK562 cells were treatedwith 0.3 mmol/L STCor 0.5 mmol/L for 2 or 4 hours, respectively. The cells were stainedwith DiOC6, andreduction in Djm was determined by monitoring the uptake of DiOC6 using flow cytometry, as described in Materials and Methods. Low Djm values areexpressed as the percentage of cells exhibiting a diminished mitochondrial potential. The values obtained from the DiOC6 assays represent the mean � SDof 3 independent experiments. ��, P < 0.01; ��, P < 0.001 versus control cells. D, Western blot for mitochondrial proteins (AIF, Smac/DIABLO, cytochromeoxidase IV, and cytochrome c). Cytochrome oxidase IV (COX IV) was used as a marker for mitochondria. Protein lysates were prepared and subjected toWestern blot analysis as described inMaterials andMethods using corresponding antibodies. Equal protein loading was ensured by showing uniform b-actinexpression. The blot is representative of 3 separate experiments.



surface and the other by mitochondrial damage (23–25).The results described in the previous section prompted us tofocus on the apoptotic signaling induced by STC in HL-60and K562 cells and specifically on the cascade of caspaseactivation. Activation of caspases by STC was suggested bythe cleavage of the caspase-3 substrate, PARP, and con-firmed by the presence of cleaved caspase-3, caspase-8, andcaspase-9 (Fig. 2A). To evaluate the functional involvementof caspases in STC-induced apoptosis, we used the pan-caspase inhibitor (Z-VAD-FMK) and specific inhibitors ofcaspase-3 (Z-DEVD-FMK), caspase-9 (Z-LEHD-FMK), orcaspase-8 (Z-IETD-FMK). Pretreatment with Z-VAD-FMKpartially abolished STC-induced apoptosis (Fig. 2B). Inaddition, pretreatment with Z-DEVD-FMK, Z-IETD-FMK,or Z-LEHD-FMK partly reversed STC-induced apoptosis(Fig. 2B). These data suggest that a caspase-dependentmechanism may partially contribute to STC-induced apo-ptosis in HL-60 and K562 cells. To assess activation of themitochondrial pathway by treatment with STC, we mea-sured the mitochondrial membrane potential (Djm) andexamined the expression of mitochondrial proteins in thecytosol by Western blot analysis. A marked loss of Djm wasobserved in STC-treated HL-60 and K562 cells (Fig. 2C).Loss of Djm was accompanied by cytoplasmic release ofcytochrome c, Smac/DIABLO, and AIF after treatment withSTC (Fig. 2D). These findings indicate that treatment of HL-60 and K562 cells with STC activates both the intrinsic andextrinsic apoptotic pathways.

STC generates ceramide through activation of acid andneutral SMase in leukemic and colorectal cancer cells

To investigate the molecules upstream of STC-inducedapoptosis, we assessed ceramide production using immu-nofluorescence microscopy. TNF-a increased the produc-tion of ceramide by activation of acid SMase and neutralSMase (26) and was used as a positive control. STCincreased ceramide generation (Supplementary Fig. S1A).To assess if acid SMase, neutral SMase, or ceramide synthasemediate STC-induced apoptosis, cells were incubated withdesipramine (an inhibitor of acid SMase), GW4869 (aninhibitor of neutral SMase), or fumonisin B1 (an inhibitorof ceramide synthase) for 1 hour before theywere incubatedwith STC. STC-induced apoptosis was partially blocked bypretreatment with desipramine or GW4869 (Fig. 3A); how-ever, it was not blocked by pretreatment with fumonisin B1(data not shown). The generation of ceramide by STC wasalso partially blocked by pretreatment with desipramine orGW4869 inHL-60andK562cells (Fig. 3B andC). Itwas alsoobserved in CT-26 and SNU-C4 cells (Supplementary Fig.S2A–S2D). It is well established that activation of acidSMase is coupled with its translocation from intracellularcompartments to the plasma membrane during cellularstress responses (27–30). Especially, the translocation ofacid SMase by apoptin was observed by immunofluores-cence microscopy (31). Therefore, we investigated whetherSTC treatment caused a change in the subcellular location ofacid SMase or neutral SMase by immunofluorescencemicroscopy using FITC-anticeramide antibody and PE-anti-

acid SMase or PE-antineutral SMase antibodies. Weobserved activation of acid SMase or neutral SMase by STC,which resulted in translocation of the enzymes to theplasma membrane (Fig. 3B and C). In addition, the cer-amide generation was observed as was in human primaryleukemia cells (Supplementary Fig. S1B and S1C). Weobserved the translocation of acid SMase or neutral SMaseinto membrane using the membrane fractions and cytosolby Western blot analysis. Acid SMase or neutral SMase wasincreased in the membrane factions and decreased in thecytosol by STC treatment, as seen in TNFa-treated HL-60cells (Fig. 3D). In addition, the translocation of acid SMaseor neutral SMase from cytosol to membrane fractions bySTCwas inhibited by desipramine or GW4869, respectively(Fig. 3D). Taken together, these results suggest the gener-ation of ceramide by STC, through activation of acid SMaseor neutral SMase, may partly contribute to STC-inducedapoptosis in leukemic and colorectal cancer cells.

To further confirm the essential role of acid SMase orneutral SMase activation in STC-mediated cell death, HL-60, K562, HT-29, and SNU-C4 cells were transiently trans-fectedwith 2different siRNAs for acid SMase (ASMsiRNA-1,ASM siRNA-2) or a neutral SMase (NSM siRNA-1, NSMsiRNA-2) and compared with those transfected with thenonspecific control siRNA. Knockdown of acid SMase orneutral SMase was confirmed by immunofluorescencestaining and Western blot analysis using the respectiveantibodies (Fig. 4A–C). The extent of apoptosis was mon-itored in transfected cells exposed to STC. Knockdown ofacid SMase or neutral SMase by 2 different siRNAs partiallyprotects cells from apoptosis induced by STC (Fig. 4D).These findings indicate that activation of acid SMase orneutral SMase plays a functional role in the death of humanleukemic and colorectal cancer cells treated with STC.

Activation of acid and neutral SMase by STC isindependent from activation of Fas and caspase-8

To investigate the hierarchy of events accompanyingSTC–induced cell death, Fas, and caspase activation by STCwas monitored in desipramine-pretreated or GW4869-pre-treated HL-60 and K562 cells. Fas activation was notreversed by desipramine or GW4869 pretreatment (Supple-mentary Fig. S3A). In addition, activation of caspase-9 or -3was reversed by treatment with desipramine or GW4869;however, activation of caspase-8 was not blocked by eithertreatment (Supplementary Fig. S3B). The same results wereobserved in cells transfected with acid SMase siRNA andneutral SMase siRNA (Supplementary Fig. S3C). Theseresults indicate that the activation of neutral SMase andacid SMase, in response to STC, occurs independently fromFas and caspase-8 activation or occurs downstream fromFasand caspase-8 activation.

Knockdown of Fas inhibits STC-induced apoptosis andacid SMase activation, but not neutral SMase activation

Our aforementioned data suggest that STC treatmentactivates Fas. To evaluate the functional significance of Fasactivation in STC-induced apoptosis, cells were transiently

Yun et al.


Figure 3. Treatment of HL-60, K562, and colorectal cancer cells with STC leads to generation of ceramide through the activation of acid SMase and neutralSMase. A,HL-60, K562 cells (1�105 cells/well) and colorectal cancer cells (HT-29, SNU-C4, CT-26 cells) were incubatedwith STC in the presence or absenceof desipramine or GW4869 for 6 and 24 hours, respectively. After treatment, the percentage of apoptotic cells was determined by annexin V-FITC/PI stainingas described in the Materials and Methods. The top panel shows the mean � SD of 5 independent experiments. The lower panel is representativeof 5 experiments in each cell line. �, P < 0.05; ��, P < 0.01; ��, P < 0.001 versus STC-treated cells. B and C, in separate experiments, cells were fixed andpermeabilized. Then, sampleswere stainedwith FITC-anticeramide and (B) PE-antiacid SMase or (C) PE-antineutral SMase antibodies. Each left panel showsthe highmagnifications of HL-60 cells treated as in A. Each right panel shows highmagnifications of K562 cells treated as in A. D, left: HL-60 cells were treatedwith 0.3 mmol/L STC for indicated times or 20 ng/mL TNF-a for 4 hours. Membrane and cytosolic fractions were prepared using ProteoJET membraneprotein extraction kit and subjected to Western blot analysis as described in "Materials andMethods" using corresponding antibodies. Equal protein loadingwas ensured by showing uniform b-actin expression. The blot is representative of 5 separate experiments. TNF-a was used as a positive control. HL-60cells (1� 105 cells/well) were incubated with STC in the presence or absence of desipramine or GW4869 for 2 hours. Membrane and cytosolic fractions wereprepared using ProteoJET membrane protein extraction kit and subjected to Western blot analysis as described in Materials and Methods usingcorresponding antibodies. Equal protein loading was ensured by showing uniform b-actin expression. The blot is representative of 5 separate experiments.TNF-awasusedasapositive control. Right: K562cellswere treatedwith 0.5mmol/LSTC for indicated times.Membraneandcytosolic fractionswerepreparedusing ProteoJET membrane protein extraction kits, and were subjected to Western blot analysis using corresponding antibodies, as described inMaterials and Methods section. Equal protein loading was ensured by demonstrating uniform b-actin expression. The blot is representative of 5 separateexperiments. K562 cells (1 � 105 cells/well) were incubated with STC in the presence or absence of desipramine or GW4869 for 2 hours. Membrane andcytosolic fractions were prepared using ProteoJET membrane protein extraction kits and were subjected to Western blot analysis using correspondingantibodies, as described in Materials and Methods section. Equal protein loading was ensured by showing uniform b-actin expression. The blot isrepresentative of 5 separate experiments.



Figure 4. Transfection of siRNA for acid SMase or neutral SMase can inhibit STC-induced apoptosis in human leukemic and colorectal cancer cells. HL-60,K562, HT-29, and SNU-C4 cells were transiently transfected by electroporation with none (shock), nonspecific control (NC) siRNA, siRNA for acid SMase(ASM), or siRNA for neutral SMase (NSM) for 48 hours. A, the transfected HL-60 and K562 cells were treated with STC for 2 hours and fixed. Followingpermeabilization, samples were stained with FITC-anticeramide and (left) PE-antiacid SMase or (right) PE-antineutral SMase antibodies as described in the"Materials and Methods." The pictures are representative of 5 separate experiments. B, the transfected HT-29 and SNU-C4 cells were treated with STC for24 hours and fixed. Following permeabilization, samples were stained with FITC-anticeramide and (left) PE-antiacid SMase or (right) PE-antineutral SMaseantibodies as described in the Materials and Methods. The pictures are representative of 5 separate experiments.

Yun et al.


transfected with a Fas siRNA, and compared with thosetransfectedwith the nonspecific control siRNA. Knockdownof Fas was confirmed by Western blot analysis and immu-nofluorescence staining using antibodies against Fas (datanot shown, Fig. 5A) and the extent of apoptosis was mon-itored in transfected cells exposed to STC. Knockdown ofFas partially protected cells from apoptosis induced by STC(Fig. 5D). These findings indicate that Fas activation plays afunctional role in the death of human leukemic and colo-rectal cancer cells treated with STC. In addition, to inves-tigate the hierarchy of events accompanying STC-inducedcell death, activation of acid SMase or neutral SMase wasmonitored in Fas siRNA-transfected cells. Knockdown ofFas, which protects cells from STC-induced apoptosis, haslittle effect on STC-induced neutral SMase activation, butreduces STC-induced acid SMase activation (Fig. 5B and C).Activation of caspase-8 was also evaluated in the siRNAexperiments and similar results were found (data notshown). These findings indicate that the activation of acidSMase, in response to STC, occurs downstream from acti-vation of Fas and caspase-8.

GSH depletion and ROS production may contribute toSTC-inducedneutral SMase activation and apoptosis inleukemic and colorectal cancer cellsA previous study showed that GSH inhibited in vitro

neutral SMase from Molt-4 leukemic cells (32). To inves-tigate the effects of GSH on STC-induced apoptosis and

neutral SMase activation, cells were treated with STCand then the levels of GSH and apoptosis were measured.STC treatment resulted in decreased GSH levels andSTC-induced apoptosis was decreased by pretreatmentwith GSH and N-acetylcysteine (NAC), a precursor ofGSH biosynthesis (Supplementary Fig. S4A and S4B). Inaddition, GSH and NAC pretreatment inhibited the acti-vation of neutral SMase, but not the activation of Fas andacid SMase (Supplementary Figs. S5A–S5C and S6A–S6C). GSH pretreatment inhibited the activation ofcaspase-9 and caspase-3; however, it did not inhibit theactivation of caspase-8 (Supplementary Fig. S5D). Theseresults suggest that the depletion of GSH by STC occursbefore activation of neutral SMase. We also examined theeffect of BSO, an inhibitor of GSH biosynthesis, onSTC-induced apoptosis and neutral SMase activation. Asexpected, BSO significantly increased STC-induced apo-ptosis and activation of neutral SMase in leukemicand colorectal cancer cells (Supplementary Figs. S4B andS6B). Thus, these data are similar to previously reportedones showing that GSH inhibits the activation of neutralSMase (26).

The depletion of GSH by STC raises the possibility that1 or more oxidative injury-related mechanism of lethalitymay be involved. Thus, we measured ROS levels in HL-60,K562, HT-29, or SNU-C4 cells exposed to STC. STCinduces the production of ROS in these cells (Supplemen-tary Fig. S4C). To investigate the functional relationship of

Figure 4. (Continued ) C, HL-60 and K562 cells were transiently transfected by electroporation with none (shock), NC siRNA, siRNAs for ASM(ASM siRNA-1, ASM siRNA-2), or siRNAs for NSM (NSM siRNA-1, NSM siRNA-2) for 48 hours. Protein lysates were prepared and subjected toWestern blot analysis as described in Materials and Methods using corresponding antibodies. Equal protein loading was ensured by showing uniformb-actin expression. The blot is representative of 5 separate experiments. D, the culture medium was changed, and cells were treated with orwithout STC (0.3, 0.5, 1.0, or 1.5 mmol/L) for 6, 6, 24, or 24 hours, respectively. The percentage of apoptotic cells was determined by annexin V-FITC/PIstaining as described in "Materials and Methods." The top panel shows the mean � SD of 5 independent experiments. �, P < 0.05; ��, P < 0.01;��, P < 0.001 (HL-60 and K562 cells treated with STC vs. HL-60 and K562 cells transfected with siRNA for acid SMase or siRNA for neutral SMase andtreated with STC, respectively). ��, P < 0.01; ��, P < 0.001 (HT-29 and SNU-C4 cells treated with STC vs. HT-29 and SNU-C4 cells transfected withsiRNA for acid SMase or siRNA for neutral SMase and treated with STC, respectively). Bottom panel results are representative of 5 independentexperiments in HL-60, K562, and HT-29 cells.



Figure 5. Knockdown of Fas caninhibit STC-induced apoptosis inhuman leukemic and colorectalcancer cells. HL-60, K562, HT-29,and SNU-C4 cells were transientlytransfected by electroporation withnone (shock), nonspecific control(NC) siRNA or siRNA for Fas for 48hours. A, left: the transfected HL-60 andK562 cellswere treatedwithSTC for 2 hours and fixed.Following permeabilization,sampleswere stainedwithPE-anti-Fas and FITC-anticeramideantibodies as described in"Materials and Methods." Thepicture is representative of 5separate experiments. Right: thetransfected HT-29 and SNU-C4cells were treated with STC for 24hours and fixed. Followingpermeabilization, samples werestainedwith PE-anti-Fas and FITC-anticeramide antibodies asdescribed in "Materials andMethods." The picture isrepresentative of 5 separateexperiments. B and C, left: thetransfected HL-60 and K562 cellswere treated with STC for 2 hoursand fixed. Followingpermeabilization, samples werestained with FITC-anticeramideand (B) PE-antiacid SMase or (C)PE-antineutral SMase antibodiesas described in Materials andMethods. The picture isrepresentative of 5 separateexperiments. Right: the transfectedHT-29 and SNU-C4 cells weretreated with STC for 24 hours andwere then fixed. Followingpermeabilization, samples werestained with FITC-anticeramideand (B) PE-antiacid SMase or (C)PE-antineutral SMase antibodiesas described in the Materials andMethods section. The image is arepresentative image of 5 separateexperiments.

Yun et al.


ROS production by STC to cell death, cells were exposed toSTC in the absence or presence of NAC or catalase. Coex-posure of cells to NAC or catalase blocked STC-inducedROS production and essentially abolished apoptosis (Sup-plementary Fig. S4B and S4C). Moreover, cells treated withSTC plus NAC or catalase displayed a marked inhibition ofneutral SMase activation and a partial recovery of GSHdepletion (Supplementary Figs. S4A and S6A–S6C). Thus,these data suggest that the depletion of GSH and theproduction of ROS by STC may interact reciprocally andthat both occur upstream of neutral SMase activation.

Antitumor activity and generation of ceramide by STCin vivoWe observed the STC-induced apoptosis and the upre-

gulation of ceramide in CT-26 cells as well as in humanleukemia cells. Thus, we evaluated the ability of STC toinhibit tumor growth of cancer cells in a mouse HL-60xenograft tumor model and CT-26 subcutaneous tumormodel. As shown in Figs. 6A and C, STC significantlydecreased tumor growth. The tumors from control miceshowed the typical histologic appearance of leukemic andcolorectal cancer cells (Fig. 6B and C). After 21 days, themean volume of tumors inmice treated with STC was morethan 70%, or 60% smaller than the tumors in the vehicle-treated mice (control group mean ¼ 4,158.75 � 469.44mm3; STC group mean ¼ 1,068.23 � 181.51 mm3; P <0.001, or the control group mean ¼ 2,950.22 � 281.58mm3; STC groupmean¼ 1,134.70� 358.26mm3; P < 0.01,respectively). The sections from STC-treated HL-60 xeno-graft tumors and CT-26 tumors showed cancer cells weremarkedly decreased, with apoptotic regions with infiltra-tion of inflammatory cells. As expected, tumors from vehi-cle-treated control mice stained weakly for ceramide (Fig.6B and D). In contrast, immunohistochemical analysis oftumors from STC-treated mice revealed upregulation ofceramide (Fig. 6B and D). These data are consistent withthe in vitro findings. Mice treated with STC did not showsigns of wasting (data not shown). We also assessed the

toxicity of STC in Balb/c mice by checking the level ofglutamate-oxaloacetate transaminase and glutamate-pyru-vate transaminase. The dosage that we used in this exper-iment was safe based on the blood chemistry results (datanot shown). These data suggest upregulation of ceramideparallels in part the antineoplastic activity of STC and thatthese effects on a major signal transduction pathway couldbe related to the mechanism of STC.

DiscussionNatural products have been recognized as an important

source of therapeutically effective drugs, and their role in theprevention and treatment of cancer is becoming increas-ingly evident. Previous studies have shown that triterpeneglycosides from S. chloronotus and T. ananas inhibit thedevelopment of lymphocytic leukemia P388 (33). In thisstudy, we showed that STC from T. anax negatively regulatescell proliferation by inducing apoptosis in leukemic andcolorectal cancer cells. Until now, the molecular mechan-isms underlying STC-induced apoptosis in these cell lineshave been unknown. The present study shows that STCtreatment activates extrinsic and intrinsic pathways of apo-ptosis. To confirm whether STC-induced apoptosis occursthrough a caspase-dependent pathway, cells were treatedwith caspase inhibitors before the addition of STC. Caspaseinhibitors partially reversed the induction of apoptosis bySTC. These data suggest that STC-induced apoptosis occursin part through a caspase-dependent pathway.

Ceramide has recently emerged as a novel second mes-senger to exert potent proapoptotic effects in a variety of celltypes (34).However, little is knownof the functional role ofceramide in mediating STC-induced lethality in leukemiaand colorectal cancer cells. Our data suggest that generationof ceramide, through activation of acid SMase and neutralSMase, may contribute to the lethality of leukemic andcolorectal cancer cells. Similar to the findings in TNFa-induced apoptosis (26), we found that STC-induced apo-ptosis requires activation of acid SMase and neutral SMase.Thus, we suggest that the action of STC may be related to

Figure 5. (Continued ) D, the culture medium was changed, and the cells were treated with or without STC (0.3, 0.5, 1.0, or 1.5 mmol/L) for 6, 6, 24, or24 hours, respectively. The percentage of apoptotic cells was determined by annexin V-FITC/PI staining as described in the Materials and Methods section.The left panel represents the mean � SD of 5 independent experiments. ��, P < 0.001 (HL-60, K562, and HT-29 cells treated with STC vs. siRNA forFas-transfected HL-60, K562, and HT-29 cells treated with STC, respectively). ��, P < 0.01 (SNU-C4 cells treated with STC vs. siRNA for Fas-transfectedSNU-C4 cells treated with STC). The right panel is the representative of 5 independent experiments in HL-60, K562, and SNU-C4 cells.



TNFa production. However, we did not check the level ofTNFa in this study. Future studies will be necessary tounravel the role of TNFa in STC-induced apoptosis.

Some reports have shown that ceramide molecules formceramide-enriched membrane platforms, a prerequisite forsignal amplification and induction of death via CD95(27, 35) whereas other reports suggested that acid SMaseactivation results from that of caspase-8 (36). To clarifywhether caspase-8 activation by STC occurs through acidSMase activation, we conducted several experiments usingsiRNA transfection and chemical inhibitors. The activationof acid SMase by STCwas found to be reversed by Fas siRNAknockdown. In addition, pretreatment with desipramine

did not block the activation of Fas and caspase-8. On thebasis of our observations, we suggest that STC leads to Fasactivation followed by caspase-8 activation, finally resultingin acid SMase activation. Thus, we suggest that the primaryaction point of STC is a lipid raft of plasma membrane, asSTC has strong membranotropic activity. However, theinvolvement of a lipid raft in STC-induced apoptosis wasnot examined in this study. Further studies are required toinvestigate whether lipid raft clustering is involved in STC-induced apoptosis.

Depletion ofGSHhas been observed in response tomanyinducers of apoptosis (37, 38). Especially, TNFa causes adramatic depletion of GSH, which is closely related to

Figure 6. STC inhibits the growth of mouse tumors and increases the production of ceramide in vivo. A, HL-60 cells (2 � 107 cells/100 mL) were injectedsubcutaneously intoBalb/c nudemice. After HL-60 cells formedpalpable tumors,micewere randomized into 2 groups (n¼ 5), and treatmentwas initiated.MiceweretreatedwithvehiclecontrolandSTC(7.19mg/kg)asdescribed in theMaterialsandMethods.Tumorsizewasmeasureddailywithacaliper (calculatedvolume¼ shortestdiameter2 � longest diameter/2). ��, P < 0.01; ��, P < 0.001, significantly different from respective control. B, top: hematoxylin-eosin staining. Stained sectionswere examined and photographed with a ScanScope (Aperio Technologies, Inc.). HL-60 leukemia xenografts from mice treated with STC showed apoptosis andextensive necrosis with phagocytes (�200). Bottom: tumor tissues obtained from the above experiment on day 14 were subjected to immunohistochemistry usingantibodies against ceramide. The sections were lightly counterstained with hematoxylin and photographed with a Scan Scope. HL-60 xenograft tumors from micetreatedwith STC showed a dramatic increase of ceramide (�200). C, CT-26 cells (2� 106 cells/100 mL) were injected subcutaneously into the right flank. After CT-26cells formedpalpable tumors,micewere randomized into 2 groups (n¼ 5), and treatmentwas initiated.Micewere treatedwith vehicle control andSTC (7.19mg/kg) asdescribed in the Materials and Methods. Tumor size was measured daily with a caliper (calculated volume¼ shortest diameter2 � longest diameter/2). ��, P < 0.01;��, P < 0.001, significantly different from respective control. D, top: hematoxylin-eosin staining. Stained sections were examined and photographed with aScanScope (Aperio Technologies, Inc.). CT-26 tumors frommice treatedwith STC showed apoptosis and extensive necrosiswith phagocytes (�200). Bottom: tumortissues obtained from the above experiment on day 14 were subjected to immunohistochemistry using antibodies against ceramide. The sections were lightlycounterstained with hematoxylin and photographed with a Scan Scope. CT-26 tumors frommice treated with STC showed a dramatic increase of ceramide (�200).

Yun et al.


activation of neutral SMase (37). In this study, STC leads toGSH depletion, increased ROS production, neutral SMaseactivation, ceramide generation, and apoptosis. Replenish-ment ofGSHprevents activationof neutral SMase andofferspartial protection against STC-induced ROS production,ceramide generation, and eventual cell death. It wasobserved that BSO treatment increased STC-induced apo-ptosis, as well as ROS generation, and neutral SMase acti-vation. It is usually assumed that GSH depletion reflects anintracellular oxidation, similar to our observations. In con-trast, a previous study showed that GSH is directly exportedfrom a cell undergoing apoptosis via activation of a trans-membrane channel (38). Our results are consistent with aprevious report that GSH depletion functions in the acti-vation of neutral SMase (37).We also observed the antitumor effect of STC in vivo and

the upregulation of ceramide in a mouse HL-60 leukemicxenograft model and CT-26 subcutaneous tumor model.These results are consistent with the in vitro data. In addi-tion, the lack of toxicity in amousemodel points to STC as apotential candidate for therapeutic use. These data suggestSTC can be a possible anticancer agent for the treatment ofleukemia and colorectal cancer.In summary, our study provides the first evidence that

STC induces apoptosis of leukemic and colorectal cancercells via ceramide generation by activation of acid SMasefollowing activation of caspase-8 and activation of neutralSMase from depletion of GSH and increased production ofROS. In addition, STC has antitumor activity in vivowithoutany toxicity. These results suggest STCmay be of therapeutic

importance in the treatment of human leukemia and colo-rectal cancer.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: S.-H. Yun, E.-S. Park, J.-Y. Han, V.A. Stonik, J.I.Park, J.-Y. KwakDevelopment of methodology: S.-H. Yun, E.-S. Park, J.-Y. Han, J.I. ParkAcquisitionofdata (provided animals, acquired andmanagedpatients,provided facilities, etc.): S.-W. Shin, Y.-W. Na, J.-Y. Han, J.-S. Jeong, V.V.ShastinaAnalysis and interpretation of data (e.g., statistical analysis, biosta-tistics, computational analysis): S.-H. Yun, E.-S. Park, S.-W. Shin, J.-Y.Han, J.-S. Jeong, V.V. Shastina, V.A. Stonik, J.I. Park, J.-Y. KwakWriting, review, and/or revision of themanuscript: S.-H. Yun, E.-S. Park,J.-Y. Han, J.I. ParkAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): S.-W. Shin, J.-Y. Han, J.I. ParkStudy supervision: J.-Y. Han, J.I. Park

AcknowledgmentsThe authors thank Dr. Kalinin and Prof. Kim for providing a sample of

STC and human primary leukemia cells, respectively.

Grant SupportThis study was supported by the National Research Foundation of Korea

(NRF) grant funded by the Korea Ministry Education, Science and Technol-ogy (MEST; R13-2002-044-04001-0 and R13-2002-044-05002-0).

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received February 24, 2012; revised August 20, 2012; accepted August 27,2012; published online November 1, 2012.

References1. SmithM,BarnettM, BassanR,GattaG, Tondini C, KernW. Adult acute

myeloid leukemia. Crit Rev Oncol Hematol 2004;50:197–222.2. Jemal A, Tiwari RC, Murray T, Ghafoor A, Samuels A, Ward E, et al.

Cancer statistics. CA Cancer J Clin 2004;54:8–29.3. ParkinDM,BrayF, Ferlay J, Pisani P.Global cancer statistics, 2002.CA

Cancer J Clin 2005;55:74–108.4. Water J, Cunningham D. The changing face of chemotherapy in

colorectal cancer. Br J Cancer 2000;84:1–7.5. Stonik VA, Kalinin VI, Avilov SA. Toxins from sea cucumbers (holothur-

oids): chemical structures, properties, taxonomic distribution, biosyn-thesis and evolution. J Nat Toxins 1999;8:235–48.

6. Stonik VA. Some terpenoid and steroid derivatives from echinodermsand sponges. Pure Appl Chem 1986;58:423–36.

7. Kalinin VI, Aminin DL, Avilov SA, Silchenko AS, Stonik VA. Triterpeneglycosides from sea cucumbers (Holothuriodea, Echinodermata). Bio-logical activities and functions. In: Atta-ur-Rahman, editor. Studies innatural products chemistry. Vol. 35, London, UK: Elsevier Science BV;2008. p. 135–96.

8. Stonik VA, Maltsev II, Kalinovsky AI, Conde K, Elyakov GB. Glyco-sides of marine- invertebrates. XI. The two novel triterpene glyco-sides from holothorian of Stichopodidae family. Chem Nat Prod1982;194–9.

9. Kitagawa I, Kobayashi M, Inamoto T, Yosuzawa T, Kyogoku Y. Thestructures of six antifungal oligoglycosides, Stichlorosides A1, A2, B1,B2, C1 and C2 from the sea cucumber Stichopus chloronotus (Brandt).Chem Phar Bull 1981;29:2387–91.

10. Stonik VA, Maltsev II, Elyakov GB. Structures of thelenotoside-A andthelenoside-B from the sea cucumber Thelenota ananas. Chem NatProd 1982;624–7.

11. Ogretmen B, Hannun YA. Biologically active sphingolipids in cancerpathogenesis and treatment. Nat Rev Cancer 2004;4:604–16.

12. Brown DA, London E. Functions of lipid rafts in biological membranes.Annu Rev Cell Dev Biol 1998;14:111–36.

13. Kolesnick RN, Goni FM, Alonso A. Compartmentalization of ceramidesignaling: physical foundations and biological effects. J Cell Physiol2000;184:285–300.

14. Hannun YA, Obeid LM. Principles of bioactive lipid signaling: lessonsfrom sphingolipids. Nat Rev Mol Cell Biol 2008;9:139–50.

15. Strum JC, Ghosh S, Bell RM. Lipid second messengers. A role in cellgrowth regulation andcell cycle regulation. AdvExpMolBiol 1997;407:421–31.

16. Goni FM, Alonso A. Sphingomyelinases: enzymology and membraneactivity. FEBS Lett 2002;531:38–46.

17. Gulbins E, Kolesnick R. Acid sphingomyelinase-derived ceramidesignaling in apoptosis. Subcell Biochem 2002;36:229–44.

18. Levade T, Jaffrezou JP. Signalling sphingomyelinases: which, where,how and why?Biochim Biophys Acta 1999;1438:1–17.

19. Taha TA, Mullen TD, Obeid LM. A house divided: ceramide, sphingo-sine, and sphingosine-1-phosphate in programmed cell death. Bio-chim Biophys Acta 2006;1758:2027–36.

20. Lee CJ, Han JS, Seo CY, Park TH, Kwon HC, Jeong JS, et al.Pioglitazone, a synthetic ligand for PPARg, induces apoptosis inRB-deficient human colorectal cancer cells. Apoptosis 2006;11:401–11.

21. Shin SW, Seo CY, Han H, Han JY, Jeong JS, Kwak JY, et al. 15d-PGJ2induces apoptosis by reactive oxygen species-mediated inactivationof Akt in leukemia and colorectal cancer cells and shows in vivoantitumor activity. Clin Cancer Res 2009;15:5414–25.



22. Han H, Shin SW, Seo CY, Kwon HC, Han JY, Kim IH, et al. 15-Deoxy-D12,14-prostaglandin J2 (15d-PGJ2) sensitizes human leukemic HL-60cells to tumor necrosis factor-related apoptosis-inducing ligand(TRAIL)-induced apoptosis through Akt downregulation. Apoptosis2007;12:2101–14.

23. Fulda S, Debatin KM. Extrinsic versus intrinsic apoptosis pathways inanticancer chemotherapy. Oncogene 2006;25:4798–811.

24. Green D, Kroemer G. The central executioners of apoptosis: caspasesor mitochondria? Trends Cell Biol 1998;8:267–71.

25. Green DR, Reed JC. Mitochondria and apoptosis. Science 1998;281:1309–12.

26. Dbaibo GS, El-Assaad W, Krikorian A, Liu B, Diab K, Idriss NZ, et al.Ceramide generation by twodistinct pathways in tumor necrosis factora-induced cell death. FEBS Lett 2001;503:7–12.

27. Grassme H, Cremesti A, Kolesnick R, Gulbins E. Ceramide-mediatedclustering is required for CD95-DISC formation. Oncogene 2003;22:5457–70.

28. Grassme H, Jekle A, Riehle A, Schwartz H, Berger J, Sandhoff K, et al.CD95 signaling via ceramide-rich membrane rafts. J Biol Chem2001;276:20589–96.

29. Grassme H, Jendrossek V, Bock J, Riehle A, Gulbins E. Ceramide-richmembrane rafts mediate CD40 clustering. J Immunol 2002;168:298–307.

30. Lacour S, Hammann A, Grazide S, Lagadic-Gossmann D, Athias A,Sergent O, et al. Cisplatin-induced CD95 redistribution intomembrane

lipid rafts of HT29 human colon cancer cells. Cancer Res 2004;64:3593–8.

31. LiuX,ZeidanY, ElojeimyS,HolmanDH,El-ZawahryAM,GuoGW,et al.Involvement of sphingolipids in apoptin-induced cell killing. Mol Ther2006;14:627–36.

32. Liu B, Hannun YA. Inhibition of the neutral magnesium-dependentsphingomyelinase by glutathione. J Biol Chem 1997;272:16281–7.

33. Pettit GR, Herald CL, Herald DL. Antineoplastic agents XLV: seacucumber cytotoxic saponins. J Pharm Sci 1976;65:1558–9.

34. Hannun YA. Functions of ceramide coordinating cellular responses tostress. Science 1996;274:1855–9.

35. Grassme H, Schwarz H, Gulbins E. Molecular mechanisms of cer-amide-mediated CD95 clustering. Biochem Biophys Res Commun2001;284:1016–30.

36. Sawada M, Nakashima S, Kiyono T, Yamada J, Hara S, Nakagawa M,et al. Acid sphingomyelinase activation requires caspase-8but not p53nor reactive oxygen species during Fas-induced apoptosis in humanglioma cells. Exp Cell Res 2002;273:157–68.

37. Liu B, Andrieu-Abadie N, Levade T, Zhang P, Obeid LM, Hannun YA.Glutathione regulation of neutral sphingomyelinase in tumor necrosisfactor-a-induced cell death. J Biol Chem 1998;273:11313–20.

38. vandenDobbelsteenDJ,NobelCS,Schlegel J, Cotgreave IA,OrreniusS, Slater AF. Rapid and specific efflux of reduced glutathione duringapoptosis induced by anti-Fas/APO-1 antibody. J Biol Chem 1996;271:15420–7.

Yun et al.


stichoposidecinducesapoptosisthroughthegenerationof ...was evaluated by annexin v-ﬂuorescein...

Documents