B

rpictttfa

M

Lnawwtfw

N1ntspmdwCb

B

F

aA

SB

5

Phosphatidylinositol 3-Kinase Inhibition Down-Regulates Survivinand Facilitates TRAIL-Mediated Apoptosis in NeuroblastomasBy Sunghoon Kim, Junghee Kang, Jingbo Qiao, Robert P. Thomas, B. Mark Evers, and Dai H. Chung

Galveston, Texas

cf

R

wtcaii

C

dmpttJr

I

ackground/Purpose: Recent findings have correlated neu-oblastoma development with aberration of apoptosis. Inarticular, increased levels of survivin, a member of the

nhibitor of apoptosis protein (IAP) family, have been asso-iated with increased resistance to apoptosis in neuroblas-omas. The purpose of this study was to determine whetherhe phosphatidylinositol 3-kinase (PI3K)/Akt pathway can al-er the expression of survivin and facilitate tumor necrosisactor (TNF)-related apoptosis-inducing ligand (TRAIL)-medi-ted apoptosis in neuroblastoma cells.

ethods: Human neuroblastoma cells (SK-N-SH, BE[2]C,AN-1 and IMR-32) were treated with LY294002 or wortman-in, inhibitors of PI3K, for 24 hours. Transient transfection ofdominant negative PI3K expression plasmid, pCGNN-�p85,as performed to inhibit PI3K activation. Cells were treatedith TRAIL, caspase-3, or pan-caspase inhibitors. RNase pro-

ection assay was performed to assess mRNA changes in IAPamily members, including survivin. Western blot analysis

as performed to measure changes in the levels of pro- s

x-foreemo

optot tosto-ellpasors.theof

ct

actcfhsscllSttstta

caBdoi:10.1016/j.jpedsurg.2003.12.008

16

aspases and survivin. Apoptosis was assessed by a DNAragmentation ELISA assay.

esults: The authors found that survivin and cIAP1 mRNA asell as protein expression were decreased after PI3K inhibi-

ion. Combination treatment with LY294002 and TRAIL in-reased apoptosis of SK-N-SH cells compared with TRAILlone; these results were further corroborated by completenhibition of apoptosis by caspase-3 or pan-caspasenhibitor.

onclusions: The authors report that PI3K pathway inhibitionown-regulates survivin expression and enhances TRAIL-ediated apoptosis in neuroblastomas. PI3K pathway may

lay a crucial role in neuroblastoma cell survival; therefore,reatment with inhibitors of PI3K may provide potential novelherapeutic options.

Pediatr Surg 39:516-521. © 2004 Elsevier Inc. All rightseserved.

NDEX WORDS: Neuroblastoma, survivin, PI3K, tumor necro-

is factor–related apoptosis-inducing ligand.

ro-

ellsn offe inin

cers;tis-es-

asedhere.

by

inin

ctiverapy

pli-ofh as

EUROBLASTOMA is the most common solid etracranial tumor in children and is responsible

5% of pediatric cancer deaths.1 Patients with late-stageuroblastomas do poorly, because resistance to ch

herapy is common; altered cellular responses to apis and induction of antiapoptotic proteins are thoughlay an important role in drug resistance in neuroblaas.2,3 Central to the execution of programmed ceath or apoptosis are endoproteases known as cashich are expressed as constitutive inactive precurs4

aspases interact with other proteins, which includecl-2 family of apoptosis regulators and a family

From the Department of Surgery, The University of Texas Medicalranch, Galveston, TX.Presented, in part, at the American College of Surgeons, San

rancisco, California, October 8, 2002.Supported by grants RO1DK61470, RO1DK48498, PO1DK35608,

nd T32DK07639 from the National Institutes of Health and by anmerican Cancer Society Institutional Grant (#2002-01).Address reprint requests to Dai H. Chung, MD, Department of

urgery, The University of Texas Medical Branch, 301 Universityoulevard, Galveston, TX 77555-0536.© 2004 Elsevier Inc. All rights reserved.0022-3468/04/3904-0002$30.00/0

--

es,

ytoplasmic proteins called inhibitors of apoptosis peins (IAP),5 to modulate apoptosis.

IAPs are overexpressed in numerous malignant c6

nd are thought to suppress apoptosis by inhibitioaspases, primarily caspase-3 and -7.6-8 Moreover, one ohe IAP proteins, survivin, may play an oncogenic rolertain cancers.9 Interestingly, survivin is expressedetal tissues, most transformed cell lines, and canowever, it is not observed in adult differentiatedues.10 In neuroblastomas, high levels of survivin exprion correlate with more advanced-stage cancer.11 Re-urrent primary neuroblastomas also express increevels of survivin, and cell lines that express a higevel of survivin display a more malignant phenotyp12

urvivin expression is thought to be up-regulatedumor necrosis factor (TNF)-induced NF-�B activa-ion.13 However, the importance of other pathwaysurvivin regulation is unknown. Based on its roleumor progression, survivin may represent an attraherapeutic target for apoptosis-based chemothegainst neuroblastomas.Phosphatidylinositol 3-kinase (PI3K) has been im

ated not only in cell survival signaling but inhibitionpoptosis by inactivation of cell death proteins suc

14 15

AD and a death effector protein, caspase-9.PI3K

Journal of Pediatric Surgery, Vol 39, No 4 (April), 2004: pp 516-521

a itoll ,p on-t anc celm

re-s ondi -in-d tsd ought atedc fectoc sua ap-o isu uts asl ins asn

bi-t lls.G inn ationo i-a gest era-p

M

romC (StL ide( -c wasp MK)w diesf inedf antaC mS GE4 bad,C oRad( s purc inedf set,h gen( nt-nD o-f hers-bR weref

C

MR-3 tion( as ag LosA fetalb ng of5 witht of thei aceda ftc tionu dedp forW

D

tiona rw foretc ithL ne-a e de-t er

R

w rer’sp pres-s 2)a gA toh fterR re re-s posedt d.

W

b e withp paredu to1 Gela ane.M tion( 5%T oomt ris-b 5%T atedw our atr utionc ris-b visu-a

S

toriale ests

517PI3K INHIBITION DOWN-REGULATES SURVIVIN IN NEUROBLASTOMA

ctivates Akt by phosphorylation via phosphoinosipid and PDK1 intermediates.16 Activated Akt, in turnhosphorylates multiple proteins implicated in the c

rol of cell survival. These cell survival signals counterbalance apoptosis propagation initiated by aembrane death receptor–mediated pathway.Members of the TNF receptor (TNF-R) family rep

ent the death receptors, which bind to their correspng ligands (eg, FasL, TNF, TNF-related apoptosisucing ligand [TRAIL]). Upon ligand binding to ieath receptor, programmed cell death occurs thr

he recruitment and activation of procaspase-8. Activaspase-8 cleaves and activates downstream efaspases (caspase-3, -6, and -7) and other proteinss BID, which participates in mitochondria-initiatedptosis. Among the death receptor ligands, TRAILnique in that it displays antitumor activity withoignificant systemic toxicity compared with TNF or Figand.17 Whether the PI3K/Akt pathway is involvedurvivin regulation or in TRAIL-mediated apoptosis hot been delineated in neuroblastomas.In this study, we examined the effect of PI3K inhi

ion on survivin expression in neuroblastoma ceiven the role of survivin as an inhibitor of apoptosiseuroblastoma, we evaluated whether down-regulf survivin by PI3K inhibition augments TRAIL-medted apoptosis in neuroblastoma. Our findings sug

hat targeted suppression of survivin is a viable theutic strategy for neuroblastoma treatment.

MATERIALS AND METHODS

aterials

PI3K inhibitors, LY294002 and wortmannin, were purchased fell Signaling (Beverly, MA) and Sigma Chemical Companyouis, MO), respectively, and dissolved in dimethyl sulfoxDMSO). Killer TRAIL (rhsTRAIL) was obtained from Alexis Biohemicals (San Diego, CA). Caspase-3 inhibitor (Z-DEVD-FMK)urchased from Alexis Corp, and pan-caspase inhibitor (Z-VAD-Fas purchased from MD Biosciences (Montreal, Quebec). Antibo

or survivin, phos-Akt (Ser473), Akt, and caspase-9 were obtarom Cell Signaling. Anti-cIAP1 antibody was purchased from Sruz (Santa Cruz, CA). Anti-�-actin antibody was purchased froigma. Cell Lysis Buffer was purchased from Cell Signaling. NuPA% to 12% Bis-Tris Gel was purchased from Invitrogen (CarlsA), and Immunoblot PVDF membranes were purchased from Bi

Hercules, CA). Enhanced chemiluminescence (ECL) system wahased from Amersham (Arlington, IL). TRIZOL reagent was obtarom LIFE Technology (Grand Island, NY). Multiprobe templateAPO-5c and T7 MAXIscript kit were purchased from PharMinSan Diego, CA) and Ambion (Austin, TX), respectively. A dominaegative PI3K mutant plasmid of p85, pCGNN-�p85, was a gift fromr Warren G. King (Harvard Medical School, Boston, MA). Lip

ectamin-plus was purchased from Life-Technologies Inc. (Gaiturg, MD). Cell Death Detection ElisaPlus kit was purchased fromoche Applied Science (Indianapolis, IN). Cell culture reagents

rom Cellgro (Herndon, VA). w

l

-

rch

t

-

ell Culture

The human neuroblastoma cell lines, SK-N-SH, BE(2)C and I2, were purchased from the American Type Culture CollecATCC; Manassas, VA). The neuroblastoma cell line, LAN-1, wift from Dr Robert C. Seeger (University of Southern California,ngeles, CA). Cells were maintained in RPMI 1640 and 10%ovine serum (FBS) at 37°C in a humidified atmosphere consisti% CO2 and 95% air. Cells were plated 24 hours before treatment

he specified reagents in fresh media with 10% FBS. Becausenstability of wortmannin in aqueous solution, the media was replnd cells retreated with wortmannin every 4 to 5 hours.18 The effect o

he dominant negative PI3K expression plasmid, pCGNN-�p85, wasompared with its empty vector, pCGNN, by transient transfecsing Lipofectamin-plus following the manufacturer’s recommenrotocol. After 48 hours of incubation, the cells were harvestedestern blotting.

NA Fragmentation Assay

Apoptosis induction was measured using a DNA fragmentassay as previously described.19 Briefly, cells (100�L; 5,000 cells peell in 96-well plates) were plated in quadruplicate 24 hours be

reatment. Cells were pretreated with caspase-3 (40�mol/L) or pan-aspase (20�mol/L) inhibitors for 30 minutes before treatment wY294002 and TRAIL. After 24-hour treatment, cytoplasmic histossociated DNA fragments (mono- and oligonucleosomes) wer

ected using a Cell Death Detection ElisaPluskit. The experiments werepeated on at least 3 separate occasions.

ibonuclease Protection Assay

Cells were treated with LY294002 (20�mol/L) for 24 hours. RNAas extracted using TRIZOL reagent following the manufacturotocol. A multiprobe template set (hAPO-5c), which assesses exion of IAP family members (XIAP, survivin, NIAP, cIAP1, cIAPnd the housekeeping genesGAPDH and L32, was labeled usinmbion T7 MAXIscript. The Ambion RPA III kit was then utilizedybridize the RNA from the samples with the labeled probes. ANase treatment of hybridized samples, protected products weolved on a 5% acrylamide gel, adsorbed onto filter paper, and exo Kodak BioMax MR film in a cassette at�70°C and later develope

estern Blot Analysis

Western blots were performed as described previously.20 Briefly,oth floating and attached cells were collected and washed onchosphate-buffered saline (PBS). Whole-cell lysates were presing Cell Lysis Buffer containing 1 mmol/L PMSF. Proteins (5000�g per lane) were separated by NuPAGE 4% to 12% Bis-Trisnd electrophoretically transferred to Immuno-Blot PVDF Membrembranes were blocked overnight at 4°C with blocking solu

Tris-buffered saline solution containing 5% nonfat dry milk and 0.0ween 20). Blots then were incubated for 2 to 3 hours at r

emperature with primary antibodies in 1% buffer solution (Tuffered saline solution containing 1% nonfat dry milk and 0.0ween 20). washed 3 times with 1% buffer solution, and incubith a horseradish peroxidase-labeled secondary antibody for 1 h

oom temperature. After 3 washes with Tris-buffered saline solontaining 0.05% Tween 20 followed by a final wash with Tuffered saline without Tween 20, the immune complexes werelized using the ECL system.

tatistical Analysis

Data were analyzed using analysis of variance for 2-factor facxperiment; the 2 factors were TRAIL and LY294002 treatment. T

ere assessed at the .05 level of significance.

P

cwsSt(sdLcTsns

c(dLn

PaawmNsduc

w

W

d

e

n

L

s

518 KIM ET AL

RESULTS

I3K Inhibition Decreases Survivin and cIAP1

The PI3K/Akt pathway has been shown to protectells from apoptosis and promote survival.21 To evaluatehether PI3K/Akt inhibition affects IAP mRNA levels, a

pecific PI3K inhibitor (LY294002) was used to treatK-N-SH cells. After 24-hour treatment, RNA was ex-

racted and RPA performed to evaluate the levels of IAPXIAP, survivin, NAIP, cIAP1, cIAP2) mRNAs. Ashown in Fig 1A, treatment with LY294002 (20 �mol/L)ecreased mRNA expression for survivin and cIAP1.ittle or no difference was noted for XIAP, and NAIP;IAP2 expression was not detected. No changes inRPM-2, also known as clusterin and described as aurvival gene which counteracts cytotoxic stress,22 wasoted. Western blot analysis of similarly treated cells

Fig 1. PI3K inhibition decreases survivin and cIAP1 expression. (A)

as extracted, and multiprobe RPA was performed. Decreased survi

estern blot analysis performed to evaluate the protein expression

ecreased survivin and cIAP1 protein expression with LY294002 treat

xpression. (C) Cells were transiently transfected with a dominant n

mol/L) for 24 hours as described in the Methods section. Western blo

Y294002. (D) BE(2)C, LAN-1, and IMR-32 cells were treated with LY2

imilar to SK-N-SH results. Equal �-actin levels indicate even loading

howed decreased protein expression for survivin and s

IAP1, consistent with the decreases in mRNA levelsFig 1B). Specificity of LY294002 was confirmed byecreased phos-Akt, the active form of Akt, withY294002 treatment; total Akt protein expression wasot altered (Fig 1B).These results were further confirmed with another

I3K inhibitor, wortmannin, and a PI3K dominant neg-tive expression plasmid pCGNN-�p85 (Fig 1C). Threedditional cell lines, BE(2)C, LAN-1, and IMR-32,hich have MYCN amplification and considered highlyalignant, were treated with LY294002. Similar to SK--SH cells, LY294002 also decreased survivin expres-

ion in these cell lines (Fig 1D). Taken together, theseata suggest that the PI3K/Akt pathway positively reg-lates survivin and cIAP1 expressions in neuroblastomaells and is consistent with the role of PI3K/Akt as a cell

-SH cells were treated with LY294002 (20 �mol/L) for 24 hours RNA

d cIAP1 mRNA levels are noted. (B) Cells were treated as in (A) and

rvivin and cIAP1. Consistent with the RPA, Western blotting shows

. LY294002 also decreased phos-Akt level without affecting total Akt

e PI3K expression plasmid, �p85, or treated with wortmannin (100

lysis shows decreased survivin expression similar to treatment with

2 as in (A). Western blotting showed decreased survivin expression

SK-N

vin an

for su

ment

egativ

t ana

9400

.

urvival pathway in neuroblastomas.

P

eanm(tLw2sgcS

vn

CcLscLhm

P

teWldlbafcsstc

iPmAlcaTlssrt

S

n

a

T

s

w

�

a

t

b

(

w

N

p

a

519PI3K INHIBITION DOWN-REGULATES SURVIVIN IN NEUROBLASTOMA

I3K Inhibition Augments TRAIL-Mediated Apoptosis

We next examined whether PI3K inhibition wouldnhance TRAIL-mediated apoptosis. Cells were platednd treated with LY294002 (20 �mol/L), TRAIL (50g/mL) or combination for 24 hours. Apoptosis waseasured by an enzyme-linked immunosorbent assay

ELISA) method that detects the level of DNA fragmen-ation, a hallmark of apoptosis.23 We found thatY294002 treatment alone did not induce apoptosis,hereas TRAIL alone induced moderate apoptosis (FigA). A significant augmentation in apoptosis was ob-erved with the combination treatment. These data sug-est that down-regulation of survivin by PI3K inhibitionontributes to increased TRAIL-mediated apoptosis inK-N-SH cells.Because survivin is known to inhibit caspase-3 acti-

ation, we examined whether caspase-3 activation was

Fig 2. PI3K inhibition augments TRAIL-mediated apoptosis. (A)

K-N-SH cells were treated with LY294002 (20 �mol/L), TRAIL (50

g/mL) or combination for 24 hours. Apoptosis was evaluated using

n ELISA method. Significant apoptosis induction was noted for

RAIL treatment compared with control. Combination treatment

howed an enhanced effect compared with TRAIL alone. (B) Cells

ere pretreated with caspase-3 (40 �mol/L) or pan-caspase (20

mol/L) inhibitors for 30 minutes before treatment with LY294002

nd TRAIL. Apoptosis was evaluated as in (A) after 24-hour incuba-

ion. Both caspase-3 and pan-caspase-3 inhibitors completely

locked apoptosis induction with LY294002/TRAIL treatment.

Mean � SEM for quadruplicate determinations; *P < .05 compared

ith control, †P < .05 compared with TRAIL alone).

ecessary for LY294002/TRAIL-induced apoptosis. b

ells were pretreated with either caspase-3 or pan-aspase inhibitor before the combination treatment withY294002 and TRAIL. DNA fragmentation ELISAhowed that either caspase-3 or pan-caspase inhibitorompletely blocked apoptosis induction mediated byY294002/TRAIL (Fig 2B). These data corroborate theypothesis that reduction in survivin facilitates TRAIL-ediated caspase-3 activation and subsequent apoptosis.

I3K Inhibition Does Not Affect Caspase-9 Activation

To further evaluate the mechanism of apoptosis induc-ion with LY294002 and TRAIL treatment, we nextxamined the level of caspase-9 in SK-N-SH cells byestern blot analysis. As shown in Fig 3, procaspase-9

evels were unchanged, and cleaved products were notetected. Extrinsic (ligand-induced) apoptosis pathway isinked to the intrinsic mitochondrial apoptosis pathwayy BID.24 Cleaved BID can initiate the mitochondrialpoptotic pathway, which results in cytochrome c releaserom mitochondrion and subsequent cleavage of pro-aspase-9.25 Unchanged levels of procaspase-9 expres-ion in SK-N-SH cells with LY294002/TRAIL treatmentuggest that apoptosis induction is mainly mediatedhrough the extrinsic apoptosis pathway mediated byaspase-8 and caspase-3 activation.

DISCUSSION

In the current study, we examined the effect of PI3Knhibition on the expression levels of IAP. We found thatI3K inhibition results in decreased survivin and cIAP1RNA and protein levels. Consistent with the report byzuhata et al3 in which higher levels of survivin corre-

ated with greater resistance to TRAIL in neuroblastomaells, we found that down-regulation of survivin, medi-ted by PI3K/Akt inhibition, resulted in enhancedRAIL-induced apoptosis. Previous studies have estab-

ished the role of the PI3K/Akt pathway to promote cellurvival.26,27 In addition, the PI3K/Akt pathway has beenhown to protect cells from apoptotic stimuli.21,28 In thisegard, results from our study add survivin and cIAP1 tohe list of antiapoptotic gene targets. These studies pro-

Fig 3. PI3K inhibition does not induce caspase-9 cleavage. SK-

-SH cells were treated as described in Fig 2. Western blotting was

erformed to assess caspase-9 levels. Procaspase-9 levels were un-

ffected with LY294002/TRAIL treatments, and cleaved caspase-9

ands were not detected. Equal �-actin levels indicate even loading.

vA

cmshfnaTteafiasgs

aamsavtf

toeaile(td

idtidstBtat

MSU

Hm

b

nH

p

m3

a3

a

s1

m

s

s

520 KIM ET AL

ide additional evidence regarding the role of the PI3K/kt pathway in the regulation of apoptosis.It has been suggested that inhibitors of apoptosis

ontribute to cancer promotion, persistence of uncheckedutations, and resistance to chemotherapy.29 Previous

tudies have reported that survivin, although absent inuman adult differentiated tissues, is expressed in trans-ormed cell lines, cancers in vivo, and fetal tissue.10 Ineuroblastomas, survivin expression is associated withn unfavorable histology and disseminated disease.11

he elevated levels of survivin expression in neuroblas-omas do not necessarily prove survival advantage; how-ver, based on its known mechanism of action to blockpoptosis,7 such a role is likely, particularly given ourndings that decreased levels of survivin and cIAP1ugmented TRAIL-mediated apoptosis. Hence, targetingurvivin may result in the selective inhibition of tumorrowth and increased neuroblastoma apoptosis whileparing normal cells.

Li et al9 have suggested that survivin may counteractdefault induction of apoptosis in the G2/M phase by

ssociating with the mitotic spindle at the beginning ofitosis. In cervical cancer cell lines, interference with

urvivin function causes multiple cell division defectsnd apoptosis induction.30 Therefore, a decrease in sur-ivin levels may be sufficient for apoptosis induction inumor cells undergoing mitosis. In our current study, we

ound that a decrease in survivin levels by LY294002 f

REFEREN

12. Azuhata T, Scott D, Takamizawa S, et al: The inhibitor of

an

as

iS

d1

e1

tM

Pp

A1

rPo

reatment (for 24 hours) alone did not result in apoptosisf SK-N-SH cells. Although we did not examine theffect of prolonged PI3K inhibition on SK-N-SH cellpoptosis, treatment with LY294002 alone for 24 hoursncreased apoptosis in another human neuroblastoma celline, LAN-1, which does not have basal procaspase-8xpression and is resistant to TRAIL-induced apoptosisdata not shown). These results suggest that sensitivityo PI3K inhibition-mediated apoptosis is cell lineependent.Antiapoptotic factors, such as survivin, clearly play an

mportant role in the resistance to activation-induced celleath and prognosis of neuroblastoma. Given the findinghat decreased survivin expression facilitated apoptosisnduction in neuroblastoma, it will be of interest toetermine which physiologic growth factors up-regulateurvivin in neuroblastoma and evaluate whether blockinghese growth factors will decrease survivin expression.ased on our understanding of molecular mechanisms

hat are unique to neuroblastoma, targeted therapy tontiapoptotic factors may advance and improve futurereatment of this devastating childhood malignancy.

ACKNOWLEDGMENT

The authors thank Dr W. G. King (Harvard Medical School, Boston,A) for plasmid pCGNN-�p85, Dr Robert C. Seeger (University of

outhern California, Los Angeles, CA) for the LAN-1 cell line, Tatsuochida for statistical analysis, and Eileen Figueroa and Karen Martin

or manuscript preparation.

CES

1. Gurney JG, Ross JA, Wall DA, et al: Infant cancer in the U.S.:istology-specific incidence and trends, 1973 to 1992. J Pediatr He-atol Oncol 19:428-432, 19972. Brodeur GM: Molecular basis for heterogeneity in human neuro-

lastomas. Eur J Cancer 31A:505-510, 19953. Nakagawara A: Molecular basis of spontaneous regression of

euroblastoma: Role of neurotrophic signals and genetic abnormalities.um Cell 11:115-124, 19984. Salvesen GS, Dixit VM: Caspases: Intracellular signaling by

roteolysis. Cell 91:443-446, 19975. Liston P, Roy N, Tamai K, et al: Suppression of apoptosis inammalian cells by NAIP and a related family of IAP genes. Nature

79:349-353, 19966. LaCasse EC, Baird S, Korneluk RG, et al: The inhibitors of

poptosis (IAPs) and their emerging role in cancer. Oncogene 17:3247-259, 19987. Deveraux QL, Reed JC: IAP family proteins—Suppressors of

poptosis. Genes Dev 13:239-252, 19998. Shin S, Sung BJ, Cho YS, et al: An anti-apoptotic protein human

urvivin is a direct inhibitor of caspase-3 and -7. Biochemistry 40:117-1123, 20019. Li F, Ambrosini G, Chu EY, et al: Control of apoptosis anditotic spindle checkpoint by survivin. Nature 396:580-584, 199810. Ambrosini G, Adida C, Altieri DC: A novel anti-apoptosis gene,

urvivin, expressed in cancer and lymphoma. Nat Med 3:917-921, 199711. Adida C, Berrebi D, Peuchmaur M, et al: Anti-apoptosis gene,

urvivin, and prognosis of neuroblastoma. Lancet 351:882-883, 1998

poptosis protein survivin is associated with high-risk behavior ofeuroblastoma. J Pediatr Surg 36:1785-1791, 200113. Wang CY, Mayo MW, Korneluk RG, et al: NF-kappaB anti-

poptosis: Induction of TRAF1 and TRAF2 and c-IAP1 and c-IAP2 touppress caspase-8 activation. Science 281:1680-1683, 1998

14. del Peso L, Gonzalez-Garcia M, Page C, et al: Interleukin-3-nduced phosphorylation of BAD through the protein kinase Akt.cience 278:687-689, 199715. Cardone MH, Roy N, Stennicke HR, et al: Regulation of cell

eath protease caspase-9 by phosphorylation. Science 282:1318-1321,99816. Duronio V, Scheid MP, Ettinger S: Downstream signalling

vents regulated by phosphatidylinositol 3-kinase activity. Cell Signal0:233-239, 199817. Walczak H, Miller RE, Ariail K, et al: Tumoricidal activity of

umor necrosis factor-related apoptosis-inducing ligand in vivo. Nated 5:157-163, 199918. Kimura K, Hattori S, Kabuyama Y, et al: Neurite outgrowth of

C12 cells is suppressed by wortmannin, a specific inhibitor of phos-hatidylinositol 3-kinase. J Biol Chem 269:18961-18967, 199419. Kim S, Kang J, Hu W, et al: Geldanamycin decreases Raf-1 and

kt levels and induces apoptosis in neuroblastomas. Int J Cancer03:352-359, 200320. Kim S, Domon-Dell C, Wang Q, et al: PTEN and TNF-alpha

egulation of the intestinal-specific Cdx-2 homeobox gene through aI3K, PKB/Akt, and NF-kappaB-dependent pathway. Gastroenterol-gy 123:1163-1178, 2002

21. Datta SR, Dudek H, Tao X, et al: Akt phosphorylation of BAD

c9

oa

a

pt

1

s6

p

mp

d

dN

521PI3K INHIBITION DOWN-REGULATES SURVIVIN IN NEUROBLASTOMA

ouples survival signals to the cell-intrinsic death machinery. Cell1:231-241, 199722. Wong P, Taillefer D, Lakins J, et al: Molecular characterization

f human TRPM-2/clusterin, a gene associated with sperm maturation,poptosis and neurodegeneration. Eur J Biochem 221:917-925, 1994

23. Wyllie AH, Kerr JF, Currie AR: Cell death: The significance ofpoptosis. Int Rev Cytol 68:251-306, 1980

24. Luo X, Budihardjo I, Zou H, et al: Bid, a Bcl2 interactingrotein, mediates cytochrome c release from mitochondria in responseo activation of cell surface death receptors. Cell 94:481-490, 1998

25. Green DR, Reed JC: Mitochondria and apoptosis. Science 281:309-1312, 199826. Dudek H, Datta SR, Franke TF, et al: Regulation of neuronal

urvival by the serine-threonine protein kinase Akt. Science 275:661-65, 1997

27. Sato T, Irie S, Kitada S, et al: FAP-1: A protein tyrosinehosphatase that associates with Fas. Science 268:411-415, 1995

28. Hausler P, Papoff G, Eramo A, et al: Protection of CD95-ediated apoptosis by activation of phosphatidylinositide 3-kinase and

rotein kinase B. Eur J Immunol 28:57-69, 1998

29. Thompson CB: Apoptosis in the pathogenesis and treatment ofisease. Science 267:1456-1462, 1995

30. Li F, Ackermann EJ, Bennett CF, et al: Pleiotropic cell-divisionefects and apoptosis induced by interference with survivin function.at Cell Biol 1:461-466, 1999