BAD as a Prototype for a Novel Therapeutic Agent for the Treatment of Acute Leukemia

Aaron D. Schimmer

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy

Graduate Department of the Institute of Medical Sciences University of T oronto

@Copyright by Aaron D. Schimrner (2001)

National Library 1+1 of Canada Bibliothèque nationale du Canada

Acquisitions and Acquisitions et Bibliographic Services services bibliographiques 395 Wellington Street 395, rue Wellington Oüawa ON K I A ON4 OttawaON K1AONs Canada Canada

The author has granted a non- exclusive licence allowing the National Libmy of Canada to reproduce, loan, distribute or sell copies of this thesis in microform, paper or electronic formats.

The author retains ownership of the copy~@~t in ths thesis. Neither the thesis nor substantial extracts fiom it may be printed or otherwise reproduced without the author's permission.

L'auteur a accordé une licence non exclusive permettant à la Bibliothèque nationale du Canada de reproduire, prêter, distribuer ou vendre des copies de cette thèse sous la forme de microfiche/film, de reproduction sur papier ou sur format électronique.

L'auteur conserve la propriété du droit d'auteur qui protège cette thèse. Ni la thèse ni des extraits substantiels de celle-ci ne doivent être imprimés ou autrement reproduits sans son autorisation.

BAD as a Prototype for a NoveI Therapeutic Agent for the Treatment of Acute Leukemia

Aaron D. Schimmer, PhD Thesis Institute of Medical Sciences. University of Toronto

200 1

Abstract

Inhibitors of Bcl-2 may be useful therapeutic agents for the treatment of a wide

variety of malignancies including leukemia. A potential prototype of such a compound is

the endogenous Bcl-2 and Bcl-xL binding protein BAD. Previous reports indicate that

BAD can overcome the anti-apoptotic effect of Bcl-xL but not Bcl-2. If BAD c m o t

induce apoptosis in cells over-expressing Bcl-2, it would limit the application of

molecules like BAD as novel anti-tumor agents. We reported that transient transfection

of BAD induced cell death in cells with and without over-expression of Bcl-2 or Bcl-XL

and thereby validated the punuit of molecules like BAD as novel therapeutic agents.

Small cytotoxic peptides that mimic BAD and overcome Bcl-2 mediated dnig

resistance may be useful in the treatment of malignant disease. One example is the BH3

domain of BAD that binds and inhibits Bcl-2 and Bcl-xL. A cell-permeable version of

the peptide waç created by hsing it to cpm, a decanoic fatty acid intemaiization

sequence. This fusion peptide induced rapid ce11 death and apoptosis. However, its

cytotoxicity was significantly impaired by the over-expression of Bcl-xL, which could

limit the usefulness of this compound in neating malignancies with increased levels of

Bcl-2 and Bcl-xL. Therefore, enhancing the ability of the BH3 domain of BAD to kill

cells over-expressing these swival factors might be therapeutically useful. One strategy

to enhance its toxicity is to capitalize on the inherent alpha helical shape of the peptide,

since some amphrpathic alpha hetices are potent naturally occhng antibiotics capable

disrupting negativeiy charged membranes.

We harnessed the aipha helical propenies of the BH3 domain of BAD by fusing it

to the Ante~apedia intemalkation sequence. We created a 37 amino acid fusion peptide

corresponding to the 21 amino acids of the human BH3 domain of BAD fused to the C-

terminus of the 16 amino acids of the ANT peptide (ANTBH3BA.û). The ANTBH3BAD

fusion peptide induced rapid ce11 death and apoptosis. Its alpha helical secondary

structure disrupted mitochondnal function and allowed the peptide to overcome Bcl-2

over-expression and caspase inhibiton.

Tabte of Contents . -

Abstract 11

Table of Contents i v Introduction

Surnmary - 7

Apoptotic pathways 5 Extnnsic pathway 8 lntrinsic pathway 1 O

Cythochrome c and the mithochondria 10 Bcl-2 family proteins 14

B &Y 18 Bd-2 18 BAD a BH3-only protein 19

Cross talk between the pathways 20 Regulation of apoptotic proteins through phosphorylation 20

Defects in apoptosis are important prognostically -- 77

Extrinsic pathway -- 7?

Intrinsic pathway 23 Bcl-2 24 Bax 25 Downstream Effectors 25

The apoptosis pathway as a target for novel therapeutic agents 27 Extrinsic pathway 3 1

TRAIL 3 1 Caspase-8 activators 3 1

Intrinsic pathway 32 Bcl-2 inhibitors 33 Second Messengers 35 Mi tochondria 37'

Conclusion 37 Re ferences 38

Bad ioduces apoptosis in cells over-expression Bcl-2 or Bcl-xl without loss of mitochondrial membrane potential 56

Abstract 57 Introduction 58 Materials and Methods 59

Cells 59 PIasmids 59 Transfection 60 Western Blot 61 Ce11 Death Assays 62

GFP expression 62 Floating cells 63 Clonogenic colony assay 63 Luci ferase Assay 64

Apoptosis assay Mitochondrial membrane potential

Results Selection of Bcl-2 and BcI-xL expressing clones BAD induces ce11 death in COS cells

GFP expression Floating celIs Clonogenic assay BAD induces ce11 death in hematopoetic ce11 lines B A D induces apoptosis but does not alter AYsi

Discussion References

The BH3 domain of BAD fused to the Antennapedia peptide induces apoptosis via its alpha helical structure and independent of Bcl-2

Abstract Introduction Expenmental Procedures

Peptides Real time surface plasmon resonance Ce11 lines Internalization of the peptides Cionogenic colony assay Annexin V assay Circular Dichroism Measurement of mitochondrial membrane potential,

reactive oxygen intemediates and cytosolic calcium Caspase-3 activation

Results Bcl-2 binds ANTBH3BAD and BHSBAD, but not ANT Intemalization of the peptides A i i H 3 B A D induces a rapid cell death ANTBH3BAD induces apoptosis ANTBH3BAD induces ce11 death independent of Bcl-2 Toxicity of the firsion peptide is due

to its alpha helical shape ANTHB3 BAD disrupts mitochondrial hnction ANTBH3BAD induces caspse-3 activation, but caspase inhibitors do not pmtect against ce11 death

P-glycoprotein protects against ANTBH3BAD toxicity Discussion Re ferences

Discussion Future Directions

List of Tables and Figures

Introduction Table 1. Mechanisms of mulitcimg resistance Figure 1. Pathways of apoptosis Figure 2. Activation of apoptosis through the TNF farnily

of death receptors Figure 3a. Electron transport chah Figure 3b. ATP and H+ flux across the mitochronàrial membranes Figure 4. Family of Bcl-2 proteins Table 2. The BH3 domains of the Bcl-2 family of proteins Table 3. Prototypic thenpeutic agents for treatment of acute leukemia Table 4a. Site of action of prototypic therapeutic agents for the

treatment of acute leukernia (intrinsic pathway) Table 4b. Site of action of prototypic therapeutic agents for the

treatment of acute leukemia (extnnsic pathway)

Bad induces apoptosis in cells over-expression Bcl-2 or Bcl-xl without loss of mi tohcondrial membrane po ten tiai

Figure la. Western blot dernonstrating expression of Bcl-2, Bcl-xL in COS cells

Figure1 b. Westem blot demonstrating expression of BAD in COS cells Figure 2. BAD induced ce11 death as measured by

decrease in GFP expression Figure 3. Dose response curve for BAD-induced apoptosis Figure 4. BAD induced ce11 death as measured by an increase

in the nurnber of fioating cells Figure 5. Westem blot demonstnting expression of BcI-2

and Bcl-xL in 3SD cells and KS62 cefls Figure 6. BAD induced cell death in K562 cells

in a dose dependent manner Figure 7. BAD induced apoptosis as demonstrated

by staining with H33342 Figure 8. BAD does not alter AYM Figure 9. BAD induced cytochrome c release

The BH3 domain of BAD fused to the Antennapedia peptide induces apoptosis via its alpha helicai structure and independent of Bcl-2

Table 1. Peptides Figure 1. Real time surface plasmon resonance Figure 2. Intemalization of the peptides Figure 3. Morphologie evidence of ceil death Figure 4. ANTBH3BAD induces ceIl death

Figure 5. Dose response m e IO8 Figure 6. Bcl-2 does not abrogate the toxicity of the fusion peptide 110 Figure 7. Circular Dichroism 112 Figure 8. ï h e non-helical mutant is non toxic 114 Figure 9. ANTBH3BAD induces loss of AYh1 115 Figure 10. zVAD Mls to prevent death 117 Figure 1 1. P-glycoprotein protects against ANTBH3BAD toxicity 119

Discussion Figure 1. Proposed mechanisrn of transduction of ANT internalizntion

sequence across ce11 membranes 132 Figure 2. Alpha helical stnicture 134 Table 1 .Cornparison of cytotoxic moiecules 136

vii

Apoptosis in acute leukemia - a translational view

Aaron D. Schimmer, David W. Hedley, Linda 2. Penn, and Mark D. Minden

Rincess Margaret Hospital, University Health Network, Toronto. ON, Canada

Running Title: Apoptosis in acute leukemia - a translational view

SUMMARY Acute myeloid leukemia and adult-onset acute lyrnphocytic leukernia are aggressive

diseases with generally poor prognoses. The primary cause of treatment failures in these

patients is the emergence of multi-drug resistance. Common mechanisms of resistant

disease are defects in apoptotic pathways. Apoptosis is a process by which cells undergo

organized self-destruction. Apoptosis is ultimately executed by caspases, a farnily of

cysteine proteases that cleave critical intracellular proteins and execute the apoptotic

program. At least two major pathways for caspase activation have been identified: a) the

extrinsic pathway which involves members of the TNF family of death receptors and b)

the intrinsic pathway involving cytochrome c release from rnitochondria. Defects in the

extrinsic or inuinsic pathway may confer prognostic information for adult patients with

acute leukemia suggesting ways that such information can be used in risk stratification.

In addition, emerging knowledge about apoptosis mechanisms has laid the foundation for

the eventual development of new therapies capable of overcoming blocks in apoptosis.

Here 1 review some of the recent advances in translating information about apoptosis

mechanisms into promising new stntegies for improved treatment of adult leukernia.

Acute leukemi-a is a neoplastic disease characterized by a rapid accmnufation of

primitive hematopoietic cells. Acute leukemia is sub-classified as myeloid or lymphoid

depending on the origin of the malignant cell. Acute lymphocyuc leukernia (AU) is

predominantly a disease of childhood with 75% of ail cases occumng in patients less than

15 years of agel. in contrast, acute myeloid leukemia (AML) is more common in adults

with an incidence that increases with as2. Both AML and adult AU. are aggressive

diseases with generally poor prognoses. For exarnple, patients with AML who are less

than 55 years of age and who have no adverse prognostic features have a cornplete

remission rate of 70-85% with standard induction chemotherapy, but their long-term

survival is only approximately 50%. The outcome in elderly patients is wone. Patients

older than 65 yean of age with AML treated with standard chemotherapy have a median

survivai of 6 months and only 7% rernain in remission at 3 years3.

The primary cause of marnent failures in adult patients with leukemia is the

emergence of resistant disease. Multiple mechanisms may account for why leukemic

cells become resistant to chemotherapy (Table 1). One common cause includes defects in

the cell's inherent programmed ce11 death (apoptosis) pathways. These defects

theoretically may impair the ability to achieve remission and cure with chemotherapy. In

an atternpt to overcome drug resistance and improve clinical outcornes, attention is

tuming to developing therapeutic agents that overcome defects in the apoptosis pathways.

Though mostly in preclinical stages of development, these efforts offer a glimpse into the

future armamentatium of agents that may eventually be available for treating leukemia.

In this review, I focus on translationai approaches to apoptosis research in adult

acute leukemia. More basic details of the apoptosis pathway have been reviewed

Table 1: Mechanisms of multidmg resistance

Decreased dmg delivery Decreased dmg uptake Increased drug export Decreased metabolism of dnig to active form Resistance of target to drug-action Increased repair of hg-induce DNA damage Increased d m g metabolisrn Increased ceIl survival Decreased apoptosis

etsewhere ' with s o m t articles m m t ~ t i n g on hdividud pehway components such as

the mitochon~iria~*~, the TNF superfamily of re~e~tors"~ , ~c1-2~"'~and c a ~ p a s e s ' ~ ~ ~ ~ . Here.

1 will discuss the apoptosis pathways concentrating on members with important

prognostic or therapeutic potential. 1 will then illustrate how defects in apoptosis confer

prognostic information in patients with acute leukemia. By identifying patients at highest

nsk of resistant disease or relapse, one rnay be able to offer them more aggressive or

altemate therapy at disease onset. Finally, 1 will discuss small molecules that overcome

blocks in apoptosis and that will likely have a major impact in the treatment of acute

leukemia.

Apoptotic pathways

Apoptosis is a process by which cells undergo organized self-destruction without

eliciting an inflarnmatory response. Morphologically, cells undergoing apoptosis display

membrane blebbing and nuclear and cytoplasrnic condensation. Molecularly, classical

apoptosis is caused by the activation of caspases. Caspases are a farnily of intracellular

cysteine proteases which lie in a latent (zyrnogen) state in cells but become activated in

response to a wide variety of ce11 death stimuli.

Growing evidence indicates that ce11 death with apoptotic features can occur

without caspase activation. The mechanisms of caspase-independent ce11 death still

requires clarification, however. Here, 1 will lirnit my discussion to caspase-rnediated ce11

death as understanding these pathways has lead to the development of agents that

enhance apoptosis and may be of therapeutic value.

Caspases are organized in a cascade with upstream (initiator) caspases responsible

for activating the downstream (effector) caspases. Upsnearn caspases contain long

prodomains that hteract with ttieir spMfic activators. The downstream caspases contairi

smaller prodomains and are activated by proteolytic cleavage by upstrearn caspases.

Once activated, the downstream caspases cleave specific protein substrates leading to the

execution of apoptosis,

At present, at least two major pathways of caspase activation have been revealed:

a) the extrinsic apoptosis pathway where the TNF- family of death receptors activate

upstream casPase-812'" and b) the inuinsic apoptosis pathway where cytochrome c is

released from the mi tochondna and activates upstrearn casPase-9 '5-18(Figure 1 ).

Both pathways culminate in the activation of the downstream effector caspase - caspase-

3 l9.?O . A third minor pathway of caspase activation involving Granzyme B has also been

revealed. Granzyme B is a senne protease synthesized in cytotoxic T lymphocytes. Via

the pore forming protein perfonn. cytotoxic T lymphocytes inject Granzyme B into target

cells. Granz yme B direct1 y cleaves and activates several caspases incl uding pro-caspase-

321-23 . As such, it bypasses both the intrinsic and extrinsic pathways of caspase

activation. While over fourteen caspases have been identified. 1 will restnct my

cornments to caspases 3,8 and 9 as exarnples of downstream effector, extrinsic pathway

initiator and intrinsic pathway initiator caspases respectiveiy.

Extrinsic Pathwav

TNF farnily of death receptors

1 Faddlmort- 1

Granzyme B - Caspase-3

Intrinsic Pathwav

mitochondria

cytochrome c

Apaf- 1 I

Figure 1: Pathways of apoptosis Two major pathways of apoptosis have been discemed -the intrinsic and extrinsic. Both pathways culminate in the activation of caspase-3 - the downstrearn effector caspase. A third minor pathway has also been identified whereby Granzyme B directly activates caspase-3.

Exrnnsic Path way:

The extrinsic pathway of apoptosis begins with the TNF family of cytokine

recepton that includes Fas (CD95). DR4 (Trail-R 1) and TNF-R 1 (CD UOa), and TNFRII

(Figure 2). These receptors differ in their ligand specificity, activating binding partners

and downstrearn effectors. Upon ligand binding. activated TNFRl and Fas receptors

recruit and bind the death effector protein ~addl~ort - lx -> . Bound Fadd/Mort- 1 recruits

procaspase-8 ? Pro-caspase-8 is converted to its active fonn and is released back into

the cytosol where it cleaves and activates the downstrearn effector caspase-3 19*20.

This receptor family also induces apoptosis through an alternative signal

transduction pathway. According to one proposed mode1, Fas receptors bind the adapter

protein Daxx that in tum activates the MAP3 kinase, ASK1. When activated, ASKl

launches a phosphorylation cascade that culminates in the activation of JNKZ4.

Activation of JNK leads to the phosphorylation of c-Jun that potentiates apoptosis

through an unknown mechanism. The TNFRI and TNFRII receptors also activate the

MAPK/JNK pathway by binding the adapter proteins TRADD and TRAF that activate

A S K - ~ ~ ~ ? Disputing this mode1 Michaelson et al. 28 reponed that Daxx knockout mice

Fas l

TNF-R 1

MEK cascade MEK cascade \

\ +

Figure 2:Activation of apoptosis through the TNF family of death recepton. Activated Fas and TNFR-1 launch the exuinsic pathway of apoptosis by recruiting the adapter protein Fadd/Mort- 1. In tum, caspase-8 is activated which cleaves procaspase-3. These receptors also induce apoptosis through an altemate pathway. Through the adapter pmteins Dam or TRADD and TRAF, a phosphoryiation cascade is initiated leading CO the activation of IMC Activated JNK phosphorylates the transcription factor c-Jun that Ieads to apoptosis through unknown mechanisms. The role of Daxx in this model, however is controversial (see tex t).

display widespread apoptosis and do not snMve beyond day IO of gestation. if Daxx

functions as a pro-apoptotic protein, then one would expect to see hyperproliferation in

these knock out mice rather than apoptosis. Furthexmore, Tom et al. 29 demonstrated that

human Daxx dues not bind Fas, but, rather. promotes Fas-mediated apoptosis through its

role as a transcriptional modulator in the nucleus. Therefore, activation of Fas can induce

apoptosis through multiple pathways, but some of the mechanisms still require

clarification.

Interestingly, activation of TNFR 1 can also inhibit apoptosis. Activated TNF-R 1

recmits adapter proteins such as, TRADD, and TRAF. These adapter proteins launch a

phosphorylation cascade that frees NFKB from its inhibitor MB. Once free, NFKB

rnigrates to the nucleus where it mediates the transcription of survival genes2'30.

lntrinsic Pathway

The intrinsic pathway for caspase activation is initiated by the release of

cytochrome c from the mitochondria. Cytochrome c is normal1 y sequestered between the

inner and outer membranes of the mitochonckia, In response to a variety of pro-apoptotic

stimuli, cytochrome c is released into the ~ ~ t o s o l ' ~ . Cytochrome c then binds and

activates Apaf- 1. Apaf- 1 activates procaspase-9, which in tum cleaves procaspase-316-18.

This pathway is potentiated by BaklBax and inhibited by Bcl-2 and its anti-apoptotic

family members.

Cytochrome c and the mitachondria

Cytochrome c release from the mitochondria can occur with and without loss of

the mitoc fionciriai membrane potential (AYM) and ph ysical mitochondrid swelling .

These differences relate to the site of mitochondrial damage. A description of the

mechanisms of mitochondiiai damage is important to understand the action of key

apoptotic regdators such as Bcl-2 and Bax. Furthemon, these mechanisms serve as the

basis for developing thetapeutic agents that enhance apoptosis by targeting the

mi tochondria.

Mitochondria contain two membranes-the inner and outer membranes -

sepsrated by the inter membrane space (Figure 3). The inner membrane sunounds the

mitochondrial matrix and is the site of electron transport and ATP synthesis. Under

normal conditions, respiratory chah complexes pump H+ ions across the relatively

impermeable inner mitochondnaf membrane. Through this process, a negative electncal

gradient, AYM, is established across this membrane. The AYM drives Fo-Fi ATP

synthase to produce ATP from ADP and Pi with the resultant influx of into the matrix.

The adenine nucleotide transporter (Ant) exchanges ATP for ADP across the inner

membrane. Dmage to the Ant creates a conforenational change leading to the formation

of a permeable pore in the inner membrane. When the inner membrane becomes

permeable, AYM is lost. the rnitochondnal matrix swells, and cytochrome c is

3132 released .

The outer mitochondnai membrane contains the voltage-dependent anion

channel (VDAC). VDAC is involved in rnitochondrial respiration and controls ATP

efflux From the mitochondrial outer membrane33". Disruptions of VDAC lead to

increased pcrmeability of the outer membrane and cytochrome c release. In this scenario,

NADH NAD+

Fig 3a

a) Electron transport chah The conversion of glucose to energy in the Krebs cycle requires the reduction of NAD+ to NADH. The reoxidization of NADH occurs through the electron transport chain in the inner mitochondrial membrane. The reaction can be summarized as: NADH + WC + 1/202 NAD' + H20

Mitochondria

ATP ADP

Figure 3b: ATP and flux across the mitochondnd membranes

The reoxidization of NADH to NAD+ in the eIectron transport chah produces H&J and releases energy. This energy is used to drive the production of ATP in the rnitochondrial mauix. The adenine nucleotide transporter (Ant) controls the flux of ADP, ATP across the inner mi tochondriaî membrane. The Voltage Dependent Anion Channel (VD AC) regulates the flux of ATP across the outer mitocho~al membrane. Darnage to VDAC or ANT can induce cytochrome c release and initiate the intrinsic pathway of apoptosis. Darnage to ANT is also accompanied by physicd rnitochondrial swelling and loss of ATM.

the inner membrane remains intact, so there is no ross of &YM or mitochondrid

swelling3'. Likewise, pore forming pro-apoptotic proteins such as Bax or Bak form

channels in the outer membrane independent of VDAC that increase the permeabilizaiton

of the outer membrane and induce cytochrome c release without damaging the inner

6.36.37 membrane .

Modulation of VDAC, which is reported to occur following growth factor

withdrawal, can also close this channel. When VDAC closes, it impairs the exchange of

mitochondrial ATP for cytosolic ADP. ATP then accumulates in the inter membrane

space dong with H+ ions which darnage the mitochondnal membranes. The increased

positive charge in the inter membrane space leads to a transient increase in ATM. AS

damage to the inner membrane ensues, however. there is swelling of the rnitochondrial

matrix. cytochrome c release and ultimately loss of A Y ~ ~ ~ .

BCL-2 f m i l y proteins:

The intrinsic pathway of apoptosis is regulated by the Bcl-2 family of anti- and

pro-apoptotic proteins (Figure 4). Membership in the Bcl-2 family is bved on shared

Bcl-2 homology (BH) domains BH1, BH2, BH3, and BH4. Currently, therapeutic agents

that either block or mimic these domains are being developed. The BHL and BH2

domains are present in al1 of the suMval factors as well as the death agonists Bax, Bak

and mtd These domains are Iikely involved in pore formation as their t h e dimensional

structure resembles the pore fonning region of the diphtheria toxin 39vj0. The ciifference

in the charge and amino acid sequences between the BHI and BH2 domains of the pro-

and anti-apoptotic proteins may lead to different pore properties and account for the

differences in function between these molecuIes.

Bcl-2, Bcl-xL, MCL-1 LOOP BH3 BHl BEI2 TM

BH4 BH3 BHI BH2 TM

Bax, Bak, mtd

Bcl-xs

Bk, Hrk, Bim, BIk

BAD. Bid

Figure 4: Family of Bcl-2 proteins

The Bcl-2 family of proteins are grouped based on their pro- or anti-apoptotic functions and their common Bcl-2 homology (BH) domains. Some members also share cornmon transmembrane domains (TM) and loop domains.

The BH.3 domain is a death-promoting region common to d l members of the Bcl-

2 family (Table 2). The BH3 domain contains a core sequence of 8 amino acids of which

the leucine at position 1 and the aspartic acid at position 6 are critical for

heterodimerization. Mutation of these residues prevents heterodimerization and

abrogates the toxicity of the death agonists4'43. In the anti-apoptotic membee, the BH3

domain foms a pocket with the BH1, and BH2 domains. This pocket serves two

purposes. Fint, it buries the BH3 domain so it cannot exert its death-promoting activity.

Second, the pocket is the docking site for the BH3 domains of the death agonists 39v40.

Heterodimerization with the death agonists inhibits the function of the survival factors, as

discussed below".

The BH4 domain is found primarily in the pro-survival factors and appears to be

involved in protein-protein interactions with regulatory proteins outside the Bcl-2 family.

This domain is involved in controlling progression of the ceIl cycle. It also appears to

stabilize the VDAC in the outer mitochondrial membraned5.

A regulatory loop domain between the EH3 and BH4 dornains is found in the

anti-apoptotic proteins BcI-2, Bcl-xL and MCl-1. Interestingly, cleavage of Bcl-2 at the

loop domain by caspase-3 releases the BH4 domain and tums Bcl-2 into a potent pro-

apoptotic protein '? Perhaps by cleaving the loop domain, the BHI, BH2, BH3 pocket is

disrupted leading to exposure of the death-promoting BH3 domain. The importance of

this cleaved isoform in regulating the apoptosis pathways is unknown.

Table 2: The BH3 domdms of the Bct-2 famiiy of prottms

Bd-2 = A L S P V P V V H L T L R Q A G D F S R R

B d - x L ~ E V I P M A A V K Q A L R E A G D E F E L

BAK 6 7 P S S T M G Q V G R Q L A I I G D D I N R

BAX 5 2 Q D A S T K K L S E C L K R I G D E L D S

BIK W C M E G S D A L A L R L A C I G D E M D V

BAD ~ ~ Y J N L W A A Q R Y G R E L R R M S D E F V D

EID , t J Q E D I 1 R N 1 A R H L A Q V G D S M D R

Core Homology Region

Bax

The Bax family of death agonists act at the mitochondria to induce apoptosis.

Three mechanisrns have been identified by which Bax induces apoptosis. Fint. at the

outer mitochondrial membrane, Bax accelerates the opening of VDAC with subsequent

release of cytochrome c3'. Second, at the inner mitochondrial membrane, Bax interacts

with Ant and enhances the pemeability of the inner mitochondriai membrane"3. Finally,

when expressed at high levels, Bax is capable of forming pores by direct oligornerization

mechanisms that disrupt the permeability of the mitochondnal membranes and induce

c ytoc home c release 3637.47

Bci-2

The Bcl-2 class of anti-apoptotic proteins are important inhibitors of the intrinsic

pathway of apoptosis and are cornmon targets of novel therapeutic agents. These proteins

inhibit apoptosis by interacting with BaxlBak, and the mitochondria. Bcl-2 c m inhibit

the action of Bax/Bak by forming inactivating heterodimersJ4. Early reports suggested

that BcI-2 functioned as an anti-oxidant by preventing reactive oxygen species

generation48'49. In retrospect, this function likely relates to Bcl-2's role at the

mitochondria. At the mitochondria, BcI-2 binds VDAC and stabilizes it, thereby

preventing permeabilization of the outer mitochondrial membranes0. These anti-

apoptotic molecules also fom pores that stabilize the outer mitochondrial membrane

preventing loss of AYM cytochrome c release and ultimately reactive oxygen species

production. In addition to its role as an anti-apoptotic protein, Bcl-2 has a separate

funetion in arresting ceth at aie GO p k of the cctt cyde. The BH4 domain of the

protein is responsible for this property, but the mechanism is poorly u n d e n t ~ o d ~ ' ~ ~ .

Bcl-2 also functions at the endoplasmic reticulum (ER), where it may have

distinct actions in preventing a p ~ ~ t o s i s ~ ~ . For example, in a rat-fibroblast ce11 line

expressing an activated c-myc allele, Bcl-2 targeted to the ER is as potent an inhibitor of

senun-withdrawal induced apoptosis as Bcl-2 confined to the rnitochon~iria~~. In

addition. BcI-2 reguiates calcium flux from the ER, although the effect of this function on

apoptosis is uncleds.

BAD a BH3-only protein:

The BW-only class of proteins are endogenous inhibitors of Bcl-2 and Bcl-xL

and are attractive prototypes for therapeutic agents. This class of compounds inhibits

Bcl-UBcl-xL through heterodimer formation. Among the BH3-only class of proteins,

BAD was the first identified and is the best characterized. BAD was first cloned from a

murine cDNA library through its ability to bind Bcl-2 in a yeast-two-hybrid assaYs6.

BAD heterodimerizes with Bcl-xL and Bcl-2 but not the pro-apoptotic members Bcl-xs,

Bax or ~ a k ? To date, the only known function of BAD is to bind Bc12 and Bcl-xL, and

block the anti-apoptotic action of these proteins"J7. BAD inhibits these anti-apoptotic

proteins by preventing them from binding ~ax/~ak'"" . BAD may also inhibit the ability

of Bcl-2 and Bcl-xL to bind Apaf-1. form pores in the mitochondrial membrane, or bind

and stabilize VDAC.

The BH3 domain of BAD is necessary for its ability to bind Bc12 and Bcl-xL and

induce apoptosis. Mutations within the BH3 domain of full Iength BAD inhibit the

binding of BAD to Bc12 and Bcl-xL and block BAD-induced apoptosis in ceI1s over-

txprtssing BCCXL~~. Frathemiore, peptides eomponding te the BH3 domein of BAD

competitively inhibit the binding of Bcl-xL to Bax. Mutations within these peptides

permit the binding of Bcl-xL to ~ a x ?

Cross ta lk between pathways

While the intrinsic and extrinsic pathways are presented as separate entities, cross

talk exists between them. For example, the two pathways cooperate to enhance apoptosis

through Bid. Bid, a member of the BH3-only family of pro-apoptotic proteins, is cleaved

and activated by caspase-8. Active Bid induces cytochrome c release and thereby

initiates the intrinsic pathway58sg. Conversely, BAR inhibits apoptosis in both pathways.

BAR binds and inhibits procaspase-8 and thereby blocks the extrinsic pathway. BAR

also inhibits the intrinsic pathway through an unknown rnechanism that involves

interactions with Bcl-2 and Bax. Physically, BAR bridges the two pathways by forming

a complex with procapsase-8 and Bcl-2, aithough the significance of this cornplex is

unknownm.

Regularion of apoptotic proteins through phosphorylation.

The apoptosis pathways are positive1 y and negativel y regulated through signal

transduction. Teleologically, this regulation allows tight control of apoptotic signals

avoiding unwanted ce11 death or proliferation. Therapeutically, these regdatory points

are valuable targets for the development of compounds that enhance apoptosis. Bcl-2 and

BAD are examples of proteins that are regulated by phosphorylation and whose

regulation is king exploited for dnig developrnent.

Three known sites of phosphorylation regdate Bcl-2; ser-17, ser-70 and thr-56.

Taxol induces phosphorylation of ser-17 and inhibits the anti-apoptotic function of Bcl-

261". Under Wfic conditions, phosphorylalion of ser-70 within the loop domain by

protein kinase C (PKC) activates the anti-apoptotic function of BCI-263. This finding is

controversial. however as other reports indicate that phosphorylation on ser-70 down-

regulates the action of Bel-2. Phosphorylation of Bcl-2 on thr-56 in the hop domain by

the ce11 cycle regulator CDC2 kinase is required for activation of Bcl-2's role as a ce11

cycle inhibitor. Mutation of thr-56 abrogates the role of Bcl-2 as a cell cycle inhibitor,

but does not alter the anti-apoptotic action of the protein; indicating that the function of

Bcl-2 as an inhibitor of apoptosis and a ce11 cycle inhibitor cm be ~i i s soc ia t ed~~~~.

Other sites of phosphorylation also appear important for the regulation Bcl-2 and

are currently king characterized. For example, ATRA inhibits Bcl-2 by inducing

phosphorylation on a senne distinct fiom the sites phosphorylated by taxol and PKC?

BAD is an example of a pro-apoptotic protein that is regulated by

phosphorylation. Phosphorylation of ser 112, ser 136 or ser-155 inhibits the function of

BAD. BAD is phosphorylated on ser-136 by Akt in the AktPUK pathway6s"6 and on

ser-112 by protein kinase C in the RasRaflMAPK p t h ~ a ~ ~ ' " ~ . Phosphorylation at these

sites promotes the binding of BAD to 14-3-3 proteins57. When bound to 14-3-3 proteins,

BAD is sequestered in the cytosol and unable to interact with Bcl-2 or Bcl-xL at the outer

mitochondrial membrane. The phosphorylation of ser-136 or ser-112 also induces a

conformational change in BAD that increases the accessibility of ser-155 within the BH3

domain to phosphorylation by protein kinase A~~'' . Phosphorylation of ser-155 directly

impairs the ability of BAD to bind Bcl-xL and BC~-Z"*".

Pathways that dephosphorylate BAD increase its activity and promote apoptosis.

CalcineUrin, a calcium dependent phosphatase, dephosphorylates BAD at both S112 and

5136 an& promotes BAD-hdnrrd apc~ptosis~~. The s t c a i c t mtssnrger cnamide piornotes

the accumulation of dephosphorylated forms of BAD by inhibiting

Defects in apoptosis are important prognostically

Defects in the apoptosis pathway render leukemic blasts resistant to multiagent

chemotherapy. These blocks translate clinically into reduced rates of remission, higher

rates of relapse, and reduced overall survival. Studies have identified specific blocks in

the extrinsic and intrinsic pathways that confer prognostic information in adult patients

with acute leukemia.

Ejltrinsic pathway .-

There is evidence to implicate defects in the extrinsic pathway of caspase

activation with chemoresistance in leukemia. Exposure of the T-ALL ce11 line CEM to

doxombicin induces apoptosis and up regulates Fas expression. A mutant CEM cell line

resistant to Fas ligand fails to up-regulate Fas or undergo apoptosis in response to

doxorubicin treat~nent~~. In contmt to this report, a Iurkat ce11 Iine resistant to Fas ligand

retained sensitivity to doxorubicin-induced apoptosis75. The discrepancy between these

two studies may reflect the numerous sites of blockage in the extrinsic pathway that can

induce resistance to Fas ligand. A defect at the level of the Fas receptor rnay render a ce11

resistant to Fas ligand, but if doxombicin triggers the pathway downstream of the

receptor, then the ce11 would rernain sensitive to doxorubicin. On the other hand, defects

in the extrinsic pathway downstream of where doxorubicin acts will render cells resistant

to both Fas ligand and doxombicin. Altematively, the discrepancy rnay result because

chemotherapeutic agents effect multiple pathways simultaneously. As cells acquire

defects in one pattrway they mmiask the dfem of the chemttierapeutic agent on the

other pathway.

Fas expression confea important prognostic information in patients with

leukemia. The Fas gene encodes two isoforms, full length Fas with a transmembrane

domain and a soluble fonn of Fas that lacks this transmembrane domain. Full length Fas

is the initiator of the extrinsic pathway of apoptosis. In contrast, soluble Fas is excreted

into the extracellular environment where it acts as a decoy and binds Fas ligand. As such,

soluble Fas is an inhibitor of apoptosis. In a study of 59 patients with adult T-cell

leukemia, levels of full length Fas expression did not influence patient survival.

However, increased serum levels of soluble Fas was associated with reduced survivai and

remained an important predictor in multivariate anal ysis even after accounting for clinicd

risk facto^-s7!

The prognostic importance of defects in the extrinsic pathway has not received

much attention to date. Perhaps, in the future, additional prognostic markers within this

pathway will be identified. For example, caspase-8 is inhibited by the dominant negative

inhibitor 1-FLICE. 1 - N C E over-expression has k e n documented in the

erythroleukemia ce11 line K562, but its incidence and prognostic importance in patients

with leukemia is unknownn.

lntnnsic Path way

Defects in the inninsic pathways of apoptosis also induce multidnig resistance

and confer important prognostic information. Most of the work on this pathway has

focussed on the Bcl-2 family of proteins.

BCL-2

Over-expression of Bcl-2 is an important cause of multi-drug resistance. As a

result, patients with AML and high b e l s of Bcl-2 have lower complete remission rates

and a lower overall survival compared to patients without high levels of Bcl-2. For

exarnple. in a study of 82 consecutive patients with AML treated with standard induction

chemotherapy, patients with high levels of BcI-2 expression by irnrnunofluroescent

staining had a complete response rate of 29% venus an 85% complete response rate in

patients without high levels of Bcl-2. Likewise. high level of Bcl-2 translated into

decreased overail survival. The predicted survival at 2 years in patients with increased

Bcl-2 was 15% venus 40% in patients without increased Bcl-2. Bcl-2 remained an

important predictor of sumival in multivariate analysis over other known prognostic

marken including age and WBC >30. The influence of cytogenetics was not reported,

however7*.

Surprisingly, some studies have indicated that increased levels of Bcl-2 are

associated wi th improved survival. In patients wi th breast cancer, increased Bcl-2 by

immunostaining comlated with improved disease-free sur~ival'~ and in childhood AU,

increased BcI-2 was associated with improved event-free survivalsO. These results may

reflect the importance of post-translational modification of Bcl-2. For example, PKC

activates B C I - ~ ~ ~ , so in the absence of PKC, the Bcl-2 that accumulates is non-hinctional,

confemng a better outcorne. In support of this hypothesis, lower levels of PKC

improved the prognosis in patients with AML8'.

Bax

In a study of 165 patients with newly diagnosed AMLal, levels of Bax expression

by irnmunoblotting did not comlate with response to induction chemotherapy or

survival. However, high ratios of Bcl2:Bax protein confemd a poor prognosis with

decreased rates of complete remission and overall survival. The prognostic

discrimination from the ratio of the proteins was greater than Bcl-2 levels aione. The

ratio also provided prognosîic information above cytogenetics. The study is flawed,

however, in that the patients selected were not consecutive and inclusion cnteria were not

predefined. In the future, by using more sensitive techniques to quantitate protein

expression, Bax may prove to be an independent prognostic marker.

Do wnstream Efectors

Levels of caspase inhibitors have been correlated with suMval in acute myeloid

leukemia (AML). Inhibitors of apoptosis proteins (IAP's) are a family of caspase

inhibiton that bind and inactivate caspases including caspases 3 and 9. As such, they

block apoptotic stimuli in both the intrinsic and extnnsic pathway. In a snidy of 78

patients with AML, 76 had detectable levels of XLAP. Patients with the lowest levels of

XIAP expression had the longest median swivala2. The importance of XIAP in the

context of other clinical prognostic factors is unknown, however.

In another study of patients with newly diagnosed AML, increased levels of

procaspase-3 in peripheral blood mononuclear cells correlated wiîh a poor prognosis.

However, patients with spontaneously increased levels of active caspase-3 at the time of

diagnosis showed improved swivals3. Likewise, in 45 consecutive adult patients with

ALL, incnased levels of active caspase-3 correlated with an improved rate of complete

remission? ni mttivariate andysts, tevets of caspase-3 w m no longer predictive of

survival, but the small sample size may have obscured the predictive effect. The poor

pmgnosis arnong patients with high levels of procaspase-3 may represent defects in

caspase activation and the subsequent accumulation of uncleaved caspase-3. The better

prognosis arnong patients with increased levels of cleaved caspase-3 may indicate

competent caspase activation pathways.

In contrast to these findings, Svingen et al.'' found no correlation between levels

of Apaf-1, prwaspase 3, 8, and 9 and response to induction chemotherapy in adult

patients with AML (n=42) or AIL (n=l8). A number of explanations may account for

the lack of correlation. First, the study had small sample size and thus, limited power to

detect a difference. However, the wide variation in the ievels of these apoptotoic

members suggests that simply increasing the number of patients would not reveal a

correlation. Altematively, this study used response to chemotherapy as their endpoint.

The overall response rate for patients with AML after the first induction was 53%, which

is lower than expected. Therefore, this group of patients may have had particularly

aggressive disease due to other molecular abnormalities that ovenhadow the importance

of procaspases or Apaf- 1.

The apoptosts padmay as a EInget lm uovet tnerapeuüc agents

Classical chemotherapeutic agents induce ce11 death through a variety of

mechanisms including directly or indirectiy affecting regulators of the apoptosis pathway.

Rather than review the mechanism of action of chemotherapeutic agents. however, 1 will

focus on small molecules that specificaily overcome blocks in the apoptosis pathway.

Anempts to develop these novel therapeutic compounds have targeted both the extrinsic

and intnnsic pathways of apoptosis (Table 3, figure 4).

Table 3: Prototypic therapeutic agents for treatment of leukemi a

Extrinsic Pathway

TNF receptor ligands TRALL

Caspase-8 activators CDDO

Intrinsic Pathway

Mitochondrial Disruption Alpha helix (KLALAKKLALAK)

Bcl-2 inhibitors Ant-BH3BAK cpm-BH3BAD HA- 14- 1 Antisense Bcl-2 oligonucleotides

Second Messengers PKC inhibitors

chelerythnne calphostin

PI3 kinase inhibitors W ortmanin Ly294002

AKT inhibitors

Intrinsic Pathw av

Cytochrome c

+ Apaf- 1

Figure 4a: Site of action of prototypic therapeutic agents for the treatment of acute leukemia (intrinsic pathway)

Figure 4b: Site of action of prototypic therapeutic agents for the treatment of acute leukemia (Extrinsic pathway)

Srnall molecules that target the extrinsic pathway cm induce caspase-3 activation

and overcome blocks confined to the intnnsic pathway, such as the over-expression of

Bcl-2 or Bcl-xL

TRAIL:

Treatment with TNF can induce apoptosis through the extrinsic pathway, but dso

produces severe toxicity to normal tissue8688. In contrast, TRAIL (AP02L), which is

sirnilar to Fas ligand and activates the DR3 and DR4 receptors, shows selectivity for

tumour cells despite the presence of the DR4 receptor on normal tissues89. In a study by

Ashkenazi et alw, soluble TRAIL induced ce11 death and apoptosis in a variety of

leukernic and solid tumor ce11 lines. In contrast, normal tissue including prostate,

fibroblasts, colon and smooth muscle cells were unaffected by exposure to TRAIL. In

nude rnice. intraperitoneal treatment with TRAIL induced apoptosis and shrinkage of

colon cancer xenografts. No observable toxicity from TRAIL was observed in these mice

or in studies of non-human primates. A m e n t report, however, has cautioned that

humans rnay experience significant hepatotoxicity after treatment with TRALL~~. TRAIL

induced apoptosis in normal human hepatocytes in culture, but was not toxic to rat,

mouse or monkey hepatocytes. Given these findings, TRAIL may have to be selectively

targeted to malignant cells to be a useful therapeutic agent.

Cmpase-8 activators

Compounds that activate the extrinsic pathway at the level of caspase-8 may

overcome blocks to apoptosis upstream of this level of regulation, such as at the level of

the TNF receptor family. Such compounds also can act independent of the level of

expression d the TNF receptor. One swh agent is 2-cyano-3,12-dioxodean-1,P-dien-2&

oic acid (CDDO). CDDO is a synthetic compound that is synthesized in an 1 1-step

reaction from oleanolic acid. This compound is a potent analogue of the naturally

occurring triterpenoids which are used as herbal remedies used to treat inflammation and

cancer. CDDO induces apoptosis in leukemic ceIl lines by activating procaspase-8. As

the compound targets the extrinsic pathway, over-expression of Bcl-xL has little effect on

the toxicity of this cornpound9'~92. Currently, it is unknown whether CDDO is a direct or

indirect activator of caspase-8. Potentially, direct caspase-8 activators would be very

useful clinically, because they could bypass both defects in both the inûinsic pathway and

blocks upstrewn of caspase-8 in the extrinsic pathway. The effects of CDDO on normal

tissue and bone marrow remain unknown, however, and this information would be

important before advancing this agent into clinical development.

lntrinsic Pathway

Most of the work in developing small molecules that target defects in apoptosis

have concentrated on the intrinsic pathway and Bcl-2 in particular.

Bcl-2 inhibitors

Bcl-2 is an attractive therapeutic target. It is over-expressed in a wide variety of

maiignancies including acute leukemia and its over-expression is often associated with

clinical treatment failures. Most intriguing, the temporary inhibition of Bcl-2 rnay not be

toxic to normal cells. In support of this hypothesis, Bcl-2 knockout mice survive until

birth with no histological abnormaiities in the brain, skin, intestines, or bone marrow

hematopoietic precursors. Beginning one week after birth, however, these mice begin to

develop rend dysfunction similar to polycystic kidney disease. By four weeks after birth,

the mice become lymphopnric prharify dat to the deah of mature T cellsg3.

Furthermore, inhibition of Bcl-2 does not automatically induce a compensatory

upregulation of other anti-apoptotic proteins. However, compounds that predominantly

inhibit Bcl-xL may be less useful therapeutically due to their potential toxicity to normal

tissue. In contrast to the Bcl-2 knockout mice, Bcl-xL knockout mice do not survive

beyond day 13 of gestation. These mice sustain massive neuronai ce11 death as well as

apoptosis of the hepatic haematopoietic and immature Furthermore. in

the most primitive hematopoietic precurson (CD34*,lin',CD38-, small size), Bcl-2 is

expressed in only 1-446 of cells. In contrast, Bcl-xL is expressed in greater than 95% of

these cells. As these primitive cells proliferate and differentiate, Bcl-2 expression is up

reg~la ted~~.

One strategy to create Bcl-2 inhibiton has focussed on developing small

molecules that mimic the action of the endogenous Bcl-2-binding death agonists. An

early prototype was the BH3 death domain of Bak. Hollinger et alg6.created a ce11

permeable version of BH3Bak by fusing it to the C terminus of the 21 amino acids

Antennapedia sequence that allows for entry into cells with high efficiency and low

toxicity. They demonstrated that this fusion peptide inhibited Bcl-2 and induced ce11

death and apoptosis in HeLa cells. Its effects on leukemic cells and its tumor specificity

were not reported. One might speculate, however that the BH3 domain on Bak might

show preferentiai killing of malignant cells and not induce apoptosis in normal tissues.

Bak induces apoptosis directly through its effects on the mitochondria, so the full-length

protein Uely would be toxic to normal cells. However, the role of its BH3 domain

seems confined to bindmg and intribitmg Bcf-2. and as sucit, rnight not induce apoptasis

in normal cells that do not have concurrent active death signals.

Compounds that rnimic the BH3-only class of death agonists may aiso be useful

novel therapeutic agents. BH3-only death agonists, such as BAD, inhibit the survival

proteins Bcl-2 and Bcl-xL, but do not appear to have independent death-promoting

activity. As such, they may show selectivity for malignant cells whose active death

signais are blocked by Bcl-Del-xL over-expression. Recently, the effects of a peptide

corresponding to the BH3 domain of BAD coupled to cpm, a fatty acid moiety

intemalization sequence have been repocted9'. The decanoic fatty acid cpm

(CH3(CH2)8CO) can shuttle peptides across ce11 membranes without untoward toxicity.

Like the peptide-based intemaiization sequences such as A N T ~ ' * ~ ~ , and TATI"

represents a novel vehicle for delivenng therapeutic peptides into celis.

The cpm-BH3BAD fusion peptide mimicked the action of its parent protein BAD.

It bound Bcl-2 in a ce11 free system and induced apoptosis in HL60 cells that over-

expressed Bcl-2. In contrast, cpm-BIUBAD had minimal effect on resting and

stimulatecl peripheral blood lymphocytes. When administered intraperitoneally into

SCID rnice, this peptide suppressed the growth of HL60 tumour xenografts without

inducing untoward toxicity. Based on the understanding of BAD. one rnight have

predicted that this compound would have been toxic to nomal tissue. BAD binds Bcl-xL

with higher aff5nity than BCI-~~ ' , and Bcl-xL is critical for the development of

hematopoietic precursoa and neural ce119*. So, one might have expected to see toxicity

to nonnal tissues as a result of its preferential inhibition of Bcl-xL.

fn a novet stratcgy to devetop a Bd-2 fibitor, Wang et d ' O 1 . screened a vimial

library of compounds. They modeled the binding of Bak's BH3 domain to the BHI,

BH2, BH3 pocket of Bcl-2. Then. using a cornputer simulation of binding, they screened

a virtuai library of 193,833 compounds for their potential ability to bind Bcl-2. From this

virtual database, 28 compounds were selected for physical testing. Of these 28, one

synthetic compound, HA-14-1 displayed an ability to bind Bcl-2 and induce apoptosis in

HL60 cells. The selectivity of this compound has not been reported, yet. Since the

binding pockets of Bcl-2 and Bcl-xL differ in sequence and structure. this synthetic

compound may show preferential toxicity to cells with high levels of Bcl-2 over those

with increased Bcl-xL.

A different strategy to inhibit Bcl-2 involves antisense oligonucleotides.

Liposomal antisense Bcl-2 oligonucleotides cm induce apoptosis in HL60 cells and

pnmary AMI, blasts. In addition, they c m sensitize the cells to matment with ~ r a - ~ ' " .

Antisense oligonucleotides have been used clinically in a phase I trial. In 21 heavily

pretreated patients with B-ce11 NHL, continuous subcutaneous infusion of an &mer Bcl-

2 antisense oligonucleotide yielded encouraging results. There was one complete

response, two partial responses and eight patients with stable disease. The only notable

toxicity was thrombocytopenia that was likely secondary to the phosphorothioate

backbone of the antisense rn01ecule~~~.

Second Messengers:

Targeting the proteins that regulate the activity of the Bcl-2 family of proteins

may be another useful strategy to developing novel therapeutic agents. Activation of BcI-

2 may require phosphorylation on ser-70 by PKC under certain àrcurnstances.

Thercfm, cumpwnQ that inhibit PKC may mhence apoptosis by blocking the

activation of Bcl-2. In HL60 cells, the PKC inhibitors chelerythnne and calphostin

induced apoptosis without changing levels of Bcl-2 ~ R N A ' ~ . However, the

phosphorylation status of Bcl-2 was not measured. Since PKC has many targets, it is

unclear if these inhibitoa functioned by blocking Bcl-2. Interestingl y, PKC activators.

such as bryostatin, also sensitized leukernic cells to ara-c induced apoptosis, which

highlights the importance of other pathways dependent of the activity of PKC"'.

Targeting the phosphorylation of BAD is another therapeutic strategy. The pro-

apoptotic action of BAD is inhibited by phosphorylation on ser-136 through the PDIAKT

pathway. Preven ting BAD-phosphoryiation may increase the acti vi ty of endogenous

BAD and promote apoptosis. Wortmanin and Ly294002 are P13K inhibitors and

sensi tize HL60 cells to undergo apoptosis after treatment with c hemotherapeutic agents.

While these agents block Akt activation, they do not alter phosphorylation of BAD or the

ability of BAD to heterodimerize with Bcl-2 or cl-dm. Therefore, these kinase

inhibitoa induce apoptosis but act through proteins other than BAD. Perhaps, these PU

kinase inhibitors act by enhancing the activity of caspase-9. Phosphorylation of caspase-

9 by Akt leads to inactivation1? Therefore, inhibition of AKT may prevent caspase-9

inactivation and enhance apoptosis.

In another attempt to prevent BAD-phosphorylation, AKT was targeted directty.

The C terminai fragment fiom the Protein Kinase C-Related Kinase-2 was identified as

an AICT inhibitor. Treatment with this protein inhibited AKT, blocked phosphorylation

of BAD, and sensitized 293 cells to apoptosislo8.

On-ect mirochmdnat t0Xms.-

Compounds that directly disrupt mitochondrial function and induce cytochrome c

release may serve as prototypes for therapeutic agents for the treatment of leukemia.

Alpha helical peptides are one such cornpound. Some amphipathic alpha helices are

potent naturally occurring antibiotics that disnipt the negatively charged bacterial

membranes. These antibiotics are not toxic to the mammalian ce11 outer membranes as

the membranes are only weakly negatively charged and stabilized by cholesterol. When

intemalized into mammalian cells, alpha helixes are exposed to the negatively charged

inner mitochondria that lacks cholesterol 109*110. Ellerby et al ' ". intemalized a 14 amino

acid alpha helical peptide (KLALAK)2 and demonstrated that it induced mitochondrial

swelling, caspase activation, and ce11 death. Using a tumour specific intemalization

sequence, they demonstrated that the alpha helix could cause regression of h a s t cancer

xenografis in nude mice without untoward toxicity to normal tissue. Potentially, alpha

helicai peptides could have value in the treatment of leukemia if targeted selectively.

CONCLUSION

Through basic investigation, the apoptosis pathway has ken pieced together.

Over tirne, M e r details will be elucidated, but the major players are likely in place.

Now, the goal is to translate this basic work into clinical practice. Prognostic marken are

king identified and prototypes for novel therapeutic agents are king developed. The

challenge ahead is to furthet the knowledge of prognostic markers and incorporate them

into current treatment protocols. On the therapeutic front, small molecules that overcome

defects in apoptosis hold promise for reversing multidrug resistance. Here the challenge

relates to enhancing tumour specificity, improving drug delivery, and moving the agents

intu c M c d me. The proof of principle has bem estabtished, however, and the future

looks brighter for the treatment of acute leukemia.

ACKNO WLEDGEMENT

We would like to thank JC Reed (Bumham Institute, La lolla, CA) for his heIpfuI advice

and discussion.

Authorization to include this material in the thesis has k e n granted by the CO-authors.

Contributions to manuscript: ADS conceived of the article, and wrote the paper. DWH and LZP reviewed the manuscript and provided critical feedback. MDM supervised the work and provided cri tical feedback

REFERENCES

1. Wang PP, Haines CS. Childhood and adolescent leukaemia in a North American

population. Int J Epiderniol. l995;6: 1 100-1 109.

2. Bricker H. Estimate of overall treatment results in acute nonlymphocytic leukemia

based on age-specific rates of incidence and of complete remission. Cancer Treat

Rep. 1985;69:5-11.

3. Leith CP, Kopecky KJ, Godwin J, et al. Acute rnyeloid leukemia in the elderly:

assessrnent of multidmg resistance (mdrl) and cytogenetics distinguishes biologic

subgroups with remarkably distinct responses to standard chemotherapy. A

southwest oncology group study. Blood. l997;89:3323-3329.

4. Hengartner MO. The biochemistry of apoptosis. Nanire.2000;407:770-776.

5. Krocmcr G, R d . MitochmQiat conmi of cet1 dtath. Nature

Medicine.2000;6:5 13-5 19.

6. Crompton M. The mitochondrial permiablity transition pore and its role in ceIl

death. Biochem J. 1999; 34 1 :233-249.

7. Sheikh MS, Fomance M. Death and decoy recepton and p53-rnediated apoptosis.

Leukernia.2000; 14: 150% 15 13.

8. Baker S, Reddy EP. Modulation of life and death by the TNF receptor superfamily.

On~ogene. 1998; U:326 1-3270.

9. Wmocha K, Salles G. The tumor necrosis factor signaling complex: choosing a

path toward ce11 death or cell proliferation. Leuk Lymphorna. 1998;29:8 1-92.

10. Tsumimoto Y, Shimizu S. Bci-2 family: life or death switch. FEBS.2000;466:6-IO.

II. Kelekar A, Thornpson CB. Bcl-2 family of proteins: the role of the BH3 domain in

apoptosis. Trends Ce1 Biol. 1998;8:324-330.

12. Fadeel B. Omenius S, Zhivotovsky B. The most unkindest cut of dl: the multiple

d e s of mammalian caspases. Leukemia.2000; 14: 15 14- 1525.

13. Nunez G, Benedict MA, Hu Y, Inohara N. Caspases: the proteases of the apoptotic

pathway. Oncogene. 1998; l7:3237-3245.

14. Roktrh OW, Glaver RA, Cohen NB. Fas mediaed iqoptosis in human prostate

carcinoma ce11 lines occun via activation of caspase-8 and caspase-7. Cancer

Res. 1998;58:5870-5875.

15. Kluck RM, Bossy-Wetzel E, Green DR. Newmeyer DD. The release of cytochrome

c from the mitochondria: a primary site for BcI-2 regulation of apoptosis.

Science. l997;27S: 1 132- 1 136.

16. Saleh A, Srinivasula SM, Acharay S. Fishel R, Alnemri ES. Cytochrome c and

dATP-mediated oligomerization of Apaf-1 is a prerequisite for procaspase-9

activation. J Biol Chern.1999;274:17941-17945.

17. Zou H, Li Y, Liu X, Wang X. An Apaf-1 cytochrome c multimeric complex is a

functional apoptosome that activates procaspase-9. J Biol Chem. l999;274: 1 1549-

11556.

18. Hu Y, Ding L, Spencer DM, Nunez G. WD-40 repeat region regulates Apaf-1 self-

associateci and procaspase-9 activation. f Bid Ctterrr. 1998;273:33489-33494.

19. Kang JJ, Schaber MD, Srinivasula SM, et al. Cascades of rnammalian caspases

activation in the yeast Saccharomyces cervesisiae. J Bi01 Chem.1999;274:3189-

3 198.

20. Stennicke HR, Snrgensmeier M, S b W. et at. Roc--3 is a major

physiologie target of caspase-8. J Biol Chem. 1998;273:27084-27090.

21. Zapata JM, Takahashi R, Salvesen GS, Reed JC. Granzyme release and caspase

activation in activated huma. T-lymphocytes. J Bi01 Chem. 1998;273:69 16-6920.

22. Barry M. Heibein JA. Pinkoski MJ, et al. Granzyme b shortsircuits the need for

caspase 8 activity dunng granule-mediated cytotoxic T-lymphocyte killing by

directly cleaving Bid. Mol Ce11 Bio.2000;20:378 1-3794.

23. Atkinson EA, Barry M. Darmon AJ, et al. Cytotoxic T lymphocyte-assisted

suicide. J Bi01 Chcm. l998;273:2 1261-21266.

24. Chang HY, Yang X, Baltimore D. Dissecting fas signaling with an altered-

specificity death-domain mutant: requirement of FADD binding for apoptosis but

not Jun N-terminal kinase activation. Proc.Natl.Acad.Sci.USA. lW9;96: 1252- 1256.

25. Berglund & Olerenshaw D, Sankar A, Fedemisch M, McDondd NQ, Driscall PC.

The the-dimensional solution structure and dynamic properties of the human

FADD death domain, J Mol Bio.2OOO;302: 17 1- 188.

26. Tsao DH, McDonagh T, Telliez JB, et al. Solution structure of N-TRADD and

characterization of the interaction of N-TRADD and C-TRAF=I, a key stop in the

TNFRl signaling pathway. Mol Cell.2ûûû;S: 1051-5057.

27. b u H, Ximg f , Gotddet DV. Tht TNF rmptor 1 -associecd protein TRADD

signals ceII death and NF-kappa B activation. Ce11.1995;81:495-504.

28. Michaelson JA, Bader D, Kuo F, Kozak C, Leder P. Loss of D a x , a promiscuousIy

interacting protein, results in extensive apoptosis in early mouse development.

Genes Devel. l999;13: 1918-1923.

29. Torii S, Egan DA, Evans R, Reed JC. Human Daxx regulates Fas-induced apoptosis

from nuclear PML oncogenic domains (PODs). EMBO J. 1999; 18:6037-6Wg.

30. Han Y, Brasier AR. Mechanism for biphasic Re1 A- NF kB 1 nuclear translocation

in tumor necrosis factor a -stimulated hepatocytes. J Bi01 Chem. 1997;272:9825-

9832.

31. Narita MT Shimizu S, Ito T, et al. Bax interacts with the penniability transition pore

to induce permeability transition and cytochrorne c release in isolated rnitochondria.

Proc.Nat1.Acad.Sci.USA. 1998;95: 1468 1- 14686.

32. Brenner C, Cadiou H, LaViera H, et al. Bcl-2 and Bax regulate the channel activity

of the rnitochondnal adenine nucleotide translocator. Oncogene.2000; 19:329-336.

33. Rostovtseva TT Colombini M. ATP flux is controlled by a voltage-gated channel

fiom the mitochondrid outer membrane. J Bi01 Chem. 1996;27 1 :28ûû6-28008.

M. Liu MY, Calombini M. Regularioxr of m i t o c h d a l respiration &y controlling the

permeability of the outer membrane through the mitochondrial channel, VDAC.

Biochim Biophys Acta. 1992; 1098:255-260.

35. Pastorino JG, Tafani M, Rothman RJ, Marcineviciute A, Hoek JB, Farber JL.

Functional consequences of the sustained or transient activation by Bax of the

mitochondnal penneability transition pore. J Bi01 Chem.1999;274:3 1734-3 1739.

36. Basanez G, Nechushtan A, Drozhinin O, et al. Bax, but not Bcl-xL decreases the

lifetime of planar phospholipid bilayer membranes at subnanomolar concetrations.

Proc.Nat1. Acad.Sci .USA. L 999;96:5492-5497.

37. Antonsson B, Conti F, Ciavatta A, et al. Inhibition of Bax channel-forming activity

by Bcl-2. Science. 1997;277:370-372.

38. Vander Heider MG, Chandel NS, Schumacker Pï, Thompson CB. Bcl-xL prevents

ce11 death following growth factor withdrawal by facilitating mitochondrial

ATPfADP exchmge. Mol CeIl. 1999;3: 159-167.

39. Muchmore SW, Sattler M, Liang H, et al. X-ray and NMR structure of human Bcl-

xL an inhibitor of programrned ce11 death. Nature. 1996;38 1 :335-341.

40. Liang H, Fesik SW. ThRe-dimensional structures of proteins involved in

pmgrammed ce11 death. J Mol Bio. MV;S74:29 1-302.

41. Ottifie S. Diaz L, Home W, et d. Dmiematim pqnt ies of )iman BAD. The 3

Bi01 Chem. 1997;272:30866-30872.

42. Zha J. Harada H. Osipov K. Jockel J. Waksman G, Konmeyer K. BH3 domain of

BAD is required for heterodimerization with Bcl-xL and pro-apoptotic activity. J

Bi01 Chem. l997;272:24 10 1-24 104.

43. Marzo 1, Brenner C, Zarnzami N, et al. Bax and adenine nucieotide translocator

cooperate in the rnitochondriai control of apoptosis. Science. l998;28 1 :2027-203 1.

44. Oltvai ZN. Milliman CL, Korsmeyer K. Bcl-2 heterodimerizes in vivo with a

conserved homolog. Bax, that accelerates programmed ce11 death.

Cell. 1993;74:609-6 19.

45. Schimizu S, Konishi A. Kodama T, Tsujimoto Y. BH4 domain of antiapoptotic Bcl-

2 family members closes voltage-dependent anion channel and inhibits apoptotic

mitochondrid changes and ce11 deaih. Proc.Natl.Acad.Sci.USA.2000;97:3 100-

3 105.

46. Cheng EHY, Kirsch DG, Clem RJ, et al. Conversion of Bcl-2 io a Bax-like death

effector by caspases. Science. 1997;278: 1966- 1968.

47. forgensmeier Slvt, Xie 2, Deveranx Q, Elterby LM, Bredesen DE, Reed JC. Bax

directly induces release of cytochrome c from isolated rnitochondna.

Proc.Natl.Acad.Sci.USA. 1998;95:4997-5002.

48. Kane DJ, Safafian TA, Anton R, et al. Bcl-2 inhibition of neural death: decreased

genention of reactive oxygen species. Science. l993;262: 1274- 1277.

49. Hockenberry DM, Oltvai ZN, Yin X, Milliman CL, Konmeyer K. Bcl-2 functions

in an antioxidant pathway to prevent apoptosis. Ce11.1993;75:24 1-25 1.

50. Shimizu S, Shinohara Y, Tsujimoto Y. Bax and Bcixl independentiy regulate

apoptotic changes of yeast mitochondria that require VDAC but not adenine

nucleotide translocator. Oncogene.2000; 19:4309-43 18.

51. Furukawa Y, Iwase S, Kikuchi JT et al. Phosphorylation of Bcl-2 protein by cdc2

kinase during G2/M phases and its role in ceIl cycle regulation. J Bi01

Chern.2000;275:2 166 1-2 1667.

52. Huang DCS, OReiIly LA, Strasser A, Cory S. The anti-apoptosis function of Bcl-2

can be genetically separated h m its inhibitory effect on ce11 cycle entry. EMBO

J. 1997; 16~4628-4638.

53. Krajewski S. Tanak S. Takayama S, Schibler UT, Fenton W, Reed K. Investigation

of the subcellular distribution of the bcl-2 oncoprotein: residence in the nuclear

envelope, endoprasmic reticuhmi. and onter mitochondnat membrane. Cancer

Re. l993;4îS:47O 147 14.

54. Lee ST, Hoeflich KP, Wasfy GW, Woodgett IR, Leber B, Andrews DW, Hedley

DW, Penn LZ. Bcl-2 targeted to the endoplasmic reticulum can inhibit apoptosis

induced by Myc but not etoposide in Rat-1 fibroblasts. Oncogene. 1999;18:3520-

3528.

55. Foyouzi-Youssefi R, Amaudeau S, Borner C, et al. Bcl-2 decreases the free ca2*

concentration within the endoplasmic reticulum.

Proc.Nati.Acad.Sci.USA.2000;97:5723-5728.

56. Yang E, Zha J, Jockel I, Boise LEI, Thompson CB, Korsmeyer K. Bad, a

heterodimeric partner for Bcl-xL and Bcl-2 displaces Bac and promotes ce11 death.

Cell. l995;80:28%?9 1.

57. Zha J, Haracla H, Yang E, Jockel J, Korsmeyer K. Senne phosphorylation of death

agonist BAD in response to sarvivat factorresab in binding to 14-3-3 not BCL-

XL. Cell. 1996;87:628

58. Wang K, Yin X, Chao DT, Milliman CL, Korsmeyer K. Bid: a novel BH3 domain-

only death agonist. Genes Devel. 1996; 10:2859-2869.

59. Tang D, Mti fM, Ki& V. Caspase-8 activation m d bid cl~avagt contributes to

MCFV cellular execution in a casapse-3-dependent manner during staurosponne-

mediated apoptosis. J Bi01 Chern.2000;275:9303-9307.

60. Zhang H, Xu Q. Krajewski S, et al. BAR: an apoptosis regulator at the intersection

of caspases and Bcl-2 fûmily proteins. Proc.Natl.Acad.Sci.USA.2000;97:2597-

2602.

61. HaIdar S, Jena N, Croce CM. Inactivation of Bcl-2 by phosphorylation.

Proc.Nat1.Acad.Sci.USA. l995;92:450745 1 1.

62. Haldar S, Chintapalli J, Croce CM. Taxol induces Bcl-2 phoshorylation and death

of prostate cancer cells. Cancer Res. 1996;56: 1253- 1255.

63. Ito T, Deng X, Carr B, May WS. BcI-2 phosphorylation required for anti-apoptosis

function. J Bi01 Chem. l997;272: 1 167 1- 1 1673,

64. Hu 233, Minden MD, McCulloch EA. Phosphorylation of Bcl-2 after exposure of

human leukernic celis to retinoic acid. B lood. 1998;92: 1768- 1775.

65. Scheid MP, Duronio V. Dissociation of cytokine-induced phosphorylation of Bad

and activation of PKBIakt: involvement of MEK upstream of Bad phosphorylaiton.

Proc.Nati.Acad.Sci.USA. 1998;95:7439-7444.

der Peso t. Gonzalez-Garcia M, Page C, Herrera R, NmiezG. fnterieukirr-3-

induced phosphorylation of BAD through the protein kinase Akt.

Science. 1997;278:687-689.

Harada H. Beckneil B. Wilm M, et al. Phosphorylation and inactivation of BAD by

mitochondria-anchored protein kinase A. Mol Cell. l999;3:4 13-422.

Tan Y, Ruan H, Demeter MW, Comb MJ. p9ûrsk blocks bad-mediated ceIl death

via a protein kinase C-dependent pathway. J Biol Chem.1999:274:34859-34867.

Lizcano JM, Momce N, Cohen P. Reguiation of BAD by CAMP-dependent protein

kinase is mediated via phosphorylation of a novel site, ser155. Biochem

J.200;349:547-557.

Zhou X, Liu Y, Payne G, Lutz RI, Chittenden T. Growth factors inactivate the ce11

death promoter BAD by phosphorylation of its BH3 domain on ser155. I Bi01

Chern.2000;275:2506-2505 1.

Tan Y, Demeter W , Ruan H, Comb UT. BAD ser-155 phosphorylation regulates

BADIBcl-xL interaction and ce11 survival. J Bi01 Chern.2000;275:25865-25869.

Wang HG, Pathan N, Ethell IM, et al. ca2+-induced apoptosis through calcineuin

dephosphorylation of BAD. Science. l999;284:339-343.

73. Basu S. Bayoumy S, Zhang Y, tozano I. KoLesrrick R. Bad enables cermnide to

signal apoptosis via Ras and Raf- 1. J Bi01 Chern. N98;273:304 19-3Wî6.

74. Friesen C, Fulda S, Debatin KM. Deficient activation of the CD95 (apo-llfas)

system in hg-resistant cells. Leukemia. 1997; 1 1: 1833- 184 1.

75. Eischen CM, Kottke TJ, Basi GS, et al. Cornparison of apoptosis in wild-type and

fas-resistant cells: chemotherapy-induced apoptosis is not dependent on fas/fas

Iigand interactions. Blood. l997;90:935-943.

76. Kamihira S, Yarnada Y, Tomonaga M, Sugahara K, Tsunida K. Discrepant

expression of membrane and soluble isoforms of Fas (CD95fApo- 1) in adult T-cell

leukaemia: soluble Fas isoform is an independent risk factor for prognosis. Br J

Hematol. 1999; 10784:85 1-860.

77. Hu S, Vincenz C, Ni J, Gentz R, Dixit VM. 1-Flice, a novel inhibitor of tumor

necrosis factor receptor-1- and CD-95 induced apoptosis. I Biol.

Chem. 19973272: 17255-17257.

78. Campos L, Rouault J, Sabido O, et ai. High expression of hl-2 protein in acute

myeloid leukemia cells is associated with poor response to chemotherapy.

B h d . 1993;8 1 :3O9 1-3096.

79. Charpin C, Garcia S. Bormier P, a at. Bct-2 automatcd and quantitative

immunocytochemical assays in breast carcinomas: correlations with 10-year follow-

up. Journal of Clinical Oncology. 1998; 16:2025-203 1.

80. Couston-Smith E, Kitanaka A, Pui C, et al. Clinical devance of Bcl-2

overexpression in childhood acute lymphoblastic leukemia. Blood. l9%;87: 1 140-

1146.

8 1. Kornblau SM, Vu HT, Ruvolo P, et al. BAX and PKCa Modulate the prognostic

impact of Bc1-2 expression in acute myelogenous leukemia. Clin Cancer

Res.2000;6: 140 1 - 1409.

82. Tamm 1, Kornblau SM. Segdl H. et ai. Expression and prognostic significance of

IAP-farnily genes in human cancers and myeloid leukemias. Clin Cancer

Res.2000;6: 1796- 1803.

83. Estrov 2, Thall PF, Talpaz M, et al. Caspase 2 and caspase 3 protein levels as

predictors of survivai in acute myetogenous leukemia. Btwd. 1998;92:3090-3097.

84. Farde1 S, Thall PF, Kantarjian HM, et al. Caspase 2 andcaspase 3 as predictors of

complete remission and survival in adults with acute lymphoblastic leukemia. Clin

Cancer Res. 1999;5:404 1-4047.

85. Smnge~~ PA, Karp W, KraJew~ki S. et at Evalaatim of A@- 1 mid procspasees-2.-

3,-7.43. and -9 as potentiai prognostic markers in acute leukemia.

Bl00d.2000;96:3922-393 1.

86. JO M, Kim T, Seol D, et al. Apoptosis induced in nomai human hepatocytes by

tumor necrosis factor-related apoptosis-inducing ligand. Nanire.2000;6:564-567.

87. Leist M, Gantner F, M g S. Wende1.A. Activation of the SSkDa TNF receptor is

necessary and sufficient for TNF-induced liver failure. hepaiocyte apoptosis and

nitri te release. .J Irnmunol. 1995; 154: 130% 13 16.

88. Robaye B. Mosselmans R, Fiers W, Dumont JE, Galand P. Turnor necrosis factor

induces apoptosis (programmed ce11 death) in nomai endothelid cells in vitro. Am

J Pathol. 199 1 ; l38:447-453.

89. Wiley SR. Schwley K. Davis Smith T, Rauch C, Smith A, Goodwin RG.

Identification and charactenzation of a new member of the TNF family that induces

apoptosis. Immmity. l995;3:673

90. Ashkenazi A, Pai RC, Fong S. et al. Safety and antiturnor activity of recombinant

soluble Apo2 ligand. J Clin Invest. 1999; 104: 155-162.

91. Ito Y, Pandey P, Place A, et al. The nove1 triterpenoid 2<yano3,12-dioxoolean-

1,9-dien-28-oic acid induces apoptosis of human myeloid leukemia cells by a

caspasedilependent mechanism. Ce11 Growth Differ.2000; 11:261-267.

92. Tumrnino PJ, Gafni A. Adenine nucleotide translocates greatly increases the

partition of trinitrophenyl-ATP into reduced Tri ton X- 100 micelles. Bioph ys

J. 1 W2;63: 1071-1080.

93. Nakayama K, Negishi 1, Kuika K, Sawa H, Loh D. Targeted disruption of Bcl-2a$

in mice: occurrence of gray hair, polycystic disease, and lymphocytopenia.

Proc.Nat1.Acad.Sci.USA. 1994;9 1 :3700-3407.

94. Motoyama N, Wang F, Roth KA, et al. Massive ce11 death of immature

hematopoietic cells and neurons in bcl-x-deficient mice. Science. 1995;267: 1506-

1510.

95. Park IR, Bernstein ID, Hockenbery. Primitive human hematopoietic precurson

express Bcl-x but not Bci-2. Blwd. I995;86: 868-876.

96. Holinger EP, Chittenden T, Lutz RJ. Bak BH3 peptides antagonize Bcl-xL function

and induce apoptosis through cytochrome c-independent activation of caspases. J

Biol. Chem. f999;274: 13298- l3304.

97. Wang J, Zhang 2. Choksi S, et al. CeIl permiabb BctZ bindmg peptides: a

chemical approach to apoptosis induction in tumor cells. Cancer

Res.2000;60: 1498- 1502.

98. Fenton M. Bone N, Sinclain AJ. The efficient and rapid import of a peptide into

pnmary B and T lymphocytes and a lymphoblasroid ce11 line. J Immunol

Methods. l998;212:4 148.

99. Derossi D, loliot AH, Chassaing G, Prochiantz A. The third helix of the

Antennapedia homeodomain translocates through biological membranes. J Biol.

Chem. 1994;269: lO444- 10450.

10. Schwarze SR, Ho A, Vocero-Akbani A, Dowdy SF. In vivo protein transduction

delivery of a biologically active protein into the mouse. Science. l999;285: 1%- 159.

101. Wang J, Liu D, Zhang 2, et al. Structure-based discovery of an organic compound

that binds Bcl-2 protein and induces apoptosis of tumor cells.

ProcNatl. Acad.Sci. USA.2000;9?:7 124-7 129.

102. Konopleva M. Tari AM, Estrov 2, et al. Liposomal Bcl-2 antisense

oligonucleotides enhance proliferation, sensitize acute myeloid leukemia to

cytosine-arabinoside, and induce apoptosis independent of other antiapoptotic

p e i n s . Blood200();95:3929-3938.

103. Waters JS, Webb A, CtmMngha~ D, et at. Phase 1 Ciinid and Phmacokinetic

snidy of blc-2 antisense oligonucleotide therapy in patients with non-Hodgkin's

lymphoma. J Clin Oncol. 2000; 18: 18 12- 1823.

104. Jarvis D, Turner AJ, Povirk LF, Traylor RS, Grant S. Induction of apoptotic DNA

fragmentation and ceIl death in HL40 human promyelocytic leukemia cells by

pharmacolgical inhibitors of protein kinase C.

Proc.Nat1.Acad.Sci.USA. 1994;54: 1707-17 14.

105. Wang S, Vrana JA, Bartimore TM, et al. Agents that down-regulate or inhibit

protein kinase C circumvent mistance to 1 -p-D-arabinofuranosylacytosine-induced

apoptosis in human leukemic cells that overexpress Bcl-2. Mol

Pharmacol. 199732: 1000- lûû9.

106. Oriorman DM, McKenn SL, McGahon AJ, Knox KA, Cotter TG. Sensitisation of

HL60 human leukaemic cells to cytotoxic dnig-induced apoptosis by inhibition of

PU-kinase survival signals. Leukemia.2000; 14:602-6 1 1.

107. Cardone MH, Roy NT Stennicke HR, et al. Regulation of ce11 death by caspase-9 by

phosphorylation. Science. 1998;282: 13 18- 132 1.

108. Koh H, Lee KH, Kim D, Kim S, Kim J, Chung J. Inhibition of Akt and its anti-

apoptotic activities by tumor necrosis factor-induced PRK2 cleavage. I Biol.

Chem.2000;in press

iû9. BlonûeHe S, Houghten RA. Design of model amphipathic peptides having potent

antimicrobid activi ties. Biochemistry. 1 W2;3 1 : 12688- 12694.

110. Bessdle R, Kapitovsky A, Gorea A. Shalit 1. Fridkin M. Ail-D-magainin: chirality,

antimicrobial activity and proteolytic resistance. FEBS. 1990;274: 15 1 - 155.

1 1 1. Ellerby HM, Arap W. EIlerby LM. et ai. Anti-cancer activity of targeted pro-

apoptotic peptides. Nat Med.. 1999;s: 1032- 1038.

BAD Induces Apoptosis in Cells Over-Expressing Bcl-2 or Bcl-xL without Loss of Mitochondrial

Membrane Potential

Aaron D. Schimmer, David W. Hedley, Nhu-An Pham, Sue Chow, Mark D.

Minden

Pnncess Margaret Hospital, University Hedth Network. Toronto. ON. Canada

Running TitIe: BAD induces apoptosis

ABSTRACT Inhibitors of BcI-2 may be useful therapeutic agents for the treatment of a wide variety of

malignancies including leukernia. A potential prototype of such a compound is the

endogenous Bcl-2 and Bcl-xL binding protein BAD. Previous reports indicate that BAD

can overcome the anti-apoptotic effect of Bcl-xL but not Bcl-2. If BAD cannot induce

apoptosis in cells over-expressing Bcl-2, it would limit the application of molecules like

BAD as novel anti-tumor agents. In this study, transient transfection of BAD induced cell

death in cells with and without over-expression of Bcl-2 or Bcl-xL. Forty-eight houn

after transfection, BAD increased cc11 death in COS, COS Bcl-2, and COS Bcl-XL cells.

In addition, BAD induced ce11 death in leukemic cell lines over-expressing Bcl-2 and Bcl-

xL. BAD-induced apoptosis was not accompanied by loss of mitochondrial membrane

potential. Therefore, in conclusion mnsient transfection of BAD directly induces

apoptosis in cells over-expressing Bcl-2 or Bcl-xL and validates the pursuit of molecules

like BAD as novei thenpeutic agents.

Key words: BAD, apoptosis, Bcl-2, Bcl-xL, rnitochondria, mitochondrial membrane

potential

INTRODUCTION The over-expression of Bcl-2, a common finding in rnalignant cells. is an

important mechanism by which they become resistant to a variety of foms of

chernotherapy"'. For example, leukernic blasts over-expressing Bcl-2 are resistant to

cytosine mbinoside. etoposide. dexamethasone. and ta~ol '*~. In addition, patients with

acute leukernia and high levels of Bcl-2 have lower complete remission rates and a lower

ovenll survival compared to patients without high levels of BCI-z718. Thus, compounds

that inhibit Bcl-2 may overcome multi-drug resistance and be useful in the treatment of a

wide variety of malignancies including acute leukemia. A prototype of such a molecule is

the endogenous Bcl-2 inhibitor. BAD (Bcl-2 Associated Death protein).

A family of prûteins including BAD^'", ~ i k " . ~ l k " , ~ i m l ~ , ~ i d ' ' and ~ r k ' ~ share

a cornmon BH3 domain and are endognous inhibiton of Bcl-2 . Of these family

mcmbers, BAD was the first identified and is the best characterized. BAD is a 29kDa

phosphoprotein that heterodimenzes with Bcl-xL and Bcl-2 but not the proapoptotic

members Bcl-xs, Bax or Bak 10.1 1.14.17.18 . To date the only known function of BAD is to

bind Bc12 and Bcl-xL, and block the anti-apoptotic action of these proteins. According to

the proposeci mechanism, blocking B c l - and Bcl-xL pemirs Bax or Bak

homodimerization and potentiates apoptosis.lO.l'.l'

Based on this rnechanism of action, BAD should induce apoptosis in cells over-

expressing Bcl-2 or Bcl-xL. Despite this widely held belief, the literature indicates that

BAD does not sensitize rnalignant ceils with high levels of Bcl-2 to apoptosis after

cytokine withdrawal. Yet, the same cells over-expressing Bcl-xL rire sensitized by BAD

over-expressionlO. if BAD is unobk to indoce apoptosis in cetk over-expressing Bct-2. it

would limit the application of molecules like BAD as novel therapeutic agents.

Here 1 report that transient transfection of BAD directly induces apoptosis in cells

with and without over-expression BcI-2 and Bcl-xL. These findings demonstrate that

BAD can overcome the anti-apoptotic action of Bcl-2 and validate the potential role of

molecules that mimic the action of BAD as broadly acting anti-tumor agents.

MATERIALS AND METHODS

Cells - Monkey kidney COS cells were naintained in Dulbeco's Modified Eagle's medium with

450mgL glucose (DMEM H21). Erythmleukemica K562 cells were grown in Minimal

Essential Medium alpha (alpha MEM). Murine myeloid precursor 32D cells over-

expressing Bcl-2 (32D Bcl-2) (a gift frorn S. Berger) were created by infection of 33D

cells with a retrovims containing munne Bcl-7. They were rnaintained in RPMI 1610

with 2 8 X63 conditioned medium containing IL-3. Al1 cell lines were supplemented

with 10% fetal calf serum (Wisent), penicillin and streptomycin. The cells were grown at

37'C in 3 5% C a incubator

PIasmids

Murine BAD with a hemaglutinin tag (gift from S.J. Korsmeyer) was subcloned into the

EcoRl sites of pcDNA3 (Invitrogen) and pTracer CMV (Invitrogen). The vector pTncer

CMV is a dual promoter plasrnid where the CMV promoter drives the gene of interest

and the SV40 promoter drives green fluorescent protein (GFP). Restriction enzyme

digestion and DNA sequencing conffrmed the correct sequences and orientations. f m m

Bcl-2 (gift from S.J. Korsmeyer) was obtained in prcCMV and human Bcl-xL with a

hemaglutinin tag (a gift from G. Nunez) was obtained in pcDNA3 (Invitrogen). In both

of these vecton, the CMV promoter drives expression of the gene of interest.

Transfec tion

COS cells were rnnsfecred with the desired plûsrnids using Lipofectarnine

(Gibco-BRL) according to the recommended protocol. COS cells (2x IO.') were seeded

onto six well plates i n ZmL of DMEM H2 1 with IO% fetd cdf serum. Twenty-four

houn later. they were transfected with Lipofectamine at a ratio of 2ug plasmid to lOul of

Lipofectarnine. The transfection medium was incubated with the cells for fÏve hours and

then replaced with fresh medium. The medium was changed again the next moming.

Using this transfection protocol, 1 could routine1 y achieve a transfection efficienc y of

85% as mesured by GFP expression after transfection of GFP in a pcDNA3 vector. GFP

expression after transfection of G W in pTracer was approxirnately 35%. The lower tevel

of GFP expression after transfection of pTncer is due to the weaker SV40 promoter

dnving the expression of GFP and the presence of two competing promoters in this

vector.

COS cells over-expressing BcI-2 (COS Bcl-2) and Bcl-xL (COS Bcl-xL) were

isolated by selecting cells with 0.3mg/ml of G418 after transfection. Stable clones were

selected 14 days after transfection with cloning rings. Over-expression of Bcl-2 and BcI-

xL in individual clones was confirmed by a Western blor assay. Single clones stably

expressirrg Bel-2 or Bci-fi were maint- in d m with 0.3rng/ml G4 18 unti1

needed for the experiments.

K562 cells were transfected with the desired plarnids using Lipofectamine. K562

cells (2x 106) were seeded onto six well plates in 0.8mL DMEM H I 1 without serum or

antibiotics. The cells were immediately tnnsfected with a ratio of Zug plasmid to 5uL of

Lipofectarnine. After a five hour incubation, 4mL of fresh alpha MEM +10% fetal calf

serum was added to the cells. With this protocol. I could rourinely achieve a transfection

efficiency of 5% as measured by GFP expression after transfection of GFP in a pcDNA3

vector.

32D cells were transfected by electropontion. 32D cells ( 5 . 5 ~ 106) were

electroporated (Gene Puiser II, Biorad. MA) with a capacitance of 975uF. 300 volts, and

20ug of plasmid in 1mL of PBS. After electroporation, the cells were left on ice for 30

minutes and then transferred to lOmL of complete medium. Using this transfection

protocol. 1 could routinely achieve a tnnsfection efficiency of 4% as measured by GFP

expression after transfection of GFP in a pcDNA3 vector.

Western Blot

Protein extracts were prepared from COS cells 24 hours after transfeciion. The cells were

washed with PBS and resuspended in lysis buffer (50mM Tris pH 8, 15OmM NaCL, O. 1%

SDS, 1% NP-40, 0.5% DOC, irnM pMSE lOug/mt aprotinin, and 5uM leupeptin).

Protein concentrations were determined by spectrophotometry. Equd protein

concentrations were electrophoresed on a standard SDS- 15% polyacrylamide gel and

transferred to a nitrocellulose membrane (Mandel Scientific, Ontario). Membranes were

washed twice in PBS with 1% non-fat milk and 0.1% Tween (PBS-T). They were

blockeb for one hour in PBS witk &O% non-far milk. Membranes were then probeci witk

polyclonal rabbit-BAD (1:300 dilution) ( a gift from S.J. Konmeyer), polyclonal mouse

anti-human Bcl-2 (1: 1000 dilution) (Dako), polyclonal nbbit anti-mouse Bcl-2 (1: 1000

dilution) (Pharmingen), monoclonal mouse anti-HA to detect HA-Bcl-xL

(Sug/rnL)(Boehnnger Mannheim), pol yclonal mouse an ti-Bcl-xL( 1 2 5 0 dilution)

(Pharmingen), polyclonal mouse anti pigeon cytochrome c ( 1 : 1000 dilution) (Sigma) or

polyclonal nbbit anti-actin (1:300 dilution) (Sigma) as a loading control. The

membranes were probed with the primary antibody for iwo houn at room temperature

and ihen washed twice in PBS-T. Secondary anti-mouse (L:SOOOdilution) or anti-nbbit

(1:5000 dilution) antibodies were applied for one hour. The membranes were washed

twice in PBS-T. The membranes were then developed using an enhanced

cherniluminescence (ECL) detection kit (Amersharn, Arlingon Heights, IL) according to

the manufacturen directions. The intensity of the autoradiognph bands were quanti tated

with a densitometer (Molecular D ynamics, Sunnyvale, CA)

Cell death assavs

GFP expression

GFP expression was assayed as a measure of cell ~iabilit~'~. COS ceUs were transfected

with pTracer-CMV BAD. the empty pTracer-CMV vector or BAD in the antisense

direction in pTracer-CMV. in pTracer, GFP and the gene of interest are in the same

plasmid. Thus, there is no need to adjust ratios of the marker plasmid to the plasrnid of

interest. Forty-eight hours after tnnsfection, GFP expression was measured by flow

cytornetry. Adherent cells were harvested, washed once in PBS, and then resuspended in

500ul of PBS. The cells were stained for five minutes at room tempenture with the

DNA-binding dye 7-runino-actinomy.Cin (7AAD) (fin& concentration 2ng/ut) (Sigma) as

a viability marker. After staining, GFP and 7AAD fluorescence were measured by flow

cytometry (FACScan; Becton, Dickinson, Immunocytometry system, CA, USA). The

Lysis II program was used to analyze these data. Viable cells were gated and the

percentage of gated cells expressing GFP was deterrnined. As transfected cells died, the

percentage of cells expressing GFP decreased. The number of cells expressing GFP after

transfecting pTracer-CMV BAD was expressed as a percentage of the number of cells

expressing GFP after transfection of the empty pTracer-CMV vector. GFP expression is

the mean of three transfections.

Floating cells

Ce11 death was measured by determining the number of celis that had detached from the

tissue culture plate and were floating in the rnedium19. As COS cells die. they detach

from the tissue culture plate and float in the medium above. Using a hematocytometer,

ce11 death was measured by counting the number of cells floating in the medium 48 houn

after transfection. The number of floating cells after transfection of pTrücer-CMV BAD

was expressed as a fold increase over the number of floating cells after transfection of the

ernpty uector,

Clonogenic colony assay

A clonogenic colony assay was performed after transfecting equal number of COS Bcl-2

cells with pTracer-CMV BAD or empty vector. Forty-eight hours after transfection

equal volumes of cells were plated into three sepante 100mm3 tissue culture plates. No

antibiotic selection was performed. One week after seeding, the medium was removed,

and the plates stained with methylene blue. The number of colonies formed was counted.

With rhis protocot, I coukt routinefy achieve a 50% ptating efficiency ;ifter mnsfectim of

cells with the empty vector.

Luciferase Assay

Luci ferase mtivity was measured as a marker of ce11 v i ab i~ i t~ '~ . K562 cells and 32D Bcl-

2 cells were CO-transfected with equal concentrations of BAD in pcDNA3 or empty

vector and the luci ferase reporter gene under control of the RSV promoter. Forty-eight

houn after tnnsfection, cells were harvested, and washed in PBS. Cytoplasmic protein

was extr~cted with lysis buffer (Promega, Madison, WI, USA). Luciferase assay

subsuate (Promega, WI, USA) was added to an aliquot of lysate and luciferase activity

was measured with a Lumat LB9507 luminometer (Berthold Systems Inc, Manyville, TN.

USA). As transfected cells died, the luci ferase activity decreased. The luci ferase activi ty

after transfection of B A D was expressed as a fold decrease compared to tnnsfection of

the empty vector. Luciferase activity represents the mean of three transfections.

Apo~tosis assay

We measured apoptosis by staining for fragmenad genomic DNA with the DNA-binding

dye Hoechst 33341 (H33342) stain. COS, COS Bcl-2 and COS Bcl-xL were transfected

with pTncer-CMV B A D or empty vector as described above. Forty-eight houn after

transfection, the adherent and fioating cells were harvested sepantely and stained with

H33342 for a final concentration of 20n.M for 30 minutes at 37 '~ . After staining, the

cells were mounted on slides and were examined under ultraviolet Iight for the presence

of apoptotic bodies. The percentage of apoptotic cells represents the mean of the at least

three transfections.

Mitoc hondnaf Membrane Potentiaf

Changes in Mitochondrial Membrane Potential (A y%*) were detected by staining the cells

with DiIC 1 (5 ) ( 1,1 '3.3,3',3'-hexamethylindodicarbocyanine)"." at a final concentration

of 4OnM and propidiurn iodide (final concentration Sug/rnL). DiIC l(5) concentntes in

the negatively charged mitochondrial rnatrix. As mitochondria lose their membrane

potential, the dye is no longer bound in the mitochondria and diffuses out of the cell.

Cells stained with DiICI(5) were incubated at 3 7 ' ~ for 30 minutes and then analyzed by

flow cytometry (Coulter Epics Elite, Coulter, Flonda) by exciting at 633nm and

rneasuring through a 675 _+ 20nm bandpass filter.

RESULTS

Selection of Bcl-2 and Bcl-xL ex~ressing clones

COS cells over-expressing Bcl-2 (COS Bcl-2) or Bcl-xL (COS Bcl-xL) were

selected two weeks after transfecting them with human BcI-2 or Bcl-xL expression

vectors. Bcl-2 and Bcl-xL over-expression were confirmed by Western blot (Figure la).

Bci-2 was detected with a polyclonal anti-human Bcl-2 antibody and Bcl-xL was detected

with a monoclonal anti-HA antibody against the HA tag on Bcl-xL. COS cells had

detectable basal levek of Bel-2 prior ta trmsfectim. The stable dones h3d 5 and 1 1 fotd

higher Bcl-2 and Bcl-xL expression respectively than the parent COS line as detemined

by densitometry of the Western blot.

COS COS + BCL- 2

BCL- 2 -

ACTlN - BCLXL

COS COS +BCLXL

ACTlN r

Fi pure 1 q Western blot demonstmting expression of Bcl-2, Bcl-xL in COS cells. COS cells were transfected with human Bcl-2 or Bcl-xL and stable clones were selected with G418. Protein lysates were separated on a 15% polyacrylmide gel and transferred to a membrane filter. Bcl-2 was detected with a polyclonai rabbit anti-human Bcl-2 antibody and Bcl-xL was detected with a monoclonal mouse anti-HA antibody directed against the HA tag on Bcl-xL. Membranes were pmbed with polycional rabbit anti-actin to demonstrate equal protein loading. Secondary anti-mouse or anti-rabbit antibodies were applied. The membranes were developed using an ECL detection kit.

BAD inchces cett death TI COS cetts

The effects of BAD on ce11 death and apoptosis were studied after transient tnnsfection.

BAD expression was detected 24 houn after tnnsfection by Western blot using a

polyclonal anti-BAD antibody (Figure I b). No BAD expression was detected in COS

cells pnor to transfection.

GFP expression

Cell death was measured by dernonstrating a decrease in the percentage of cells

expressing GFP by flow cytornetry. GFP expression peaked 24 hours after transfection of

pTracer-CMV and ce11 death due to BAD was maximal at 48 hours. In contrast. ceIl

death was minimal before 34 hours after transfection. For example, in a representative

experiment. 24 houn after tnnsfection of pTraccr-CMV BAD. the number of cells

expressing GFP was 83% of the empty vector. Therefore, I perfonned al1 measurements

of ce11 viability 48 houn after transfection. pTncer-CMV BAD or the empty vector were

tnnsfected into COS, COS Bcl-2 or COS Bcl-xL . Forty-eight houn after transfection,

the cells were analyzed by flow cytometry. The viable cells (7AAD negative) were gated,

and GFP expression among the gated cells was measured. The number of cells

expressing GFP after transfecting pTracer CMV- BAD was expressed as a pementage of

the number of cells expressing GFP after tnnsfection of the empty vector (Figure 2). The

percentage of GFP expression is the mean of the three transfections. Transfection of

pTncer-CMV BAD into COS, COS Bcl-2, or COS Bcl-xL reduced the number of cells

expressing GFP to 9.5+1.2%, 1021.8 and 8.621.5 % of the empty vector respectively.

These reductions were significant by the t-test (p = O.Oû3,0.0001. and 0.009

COS COS + BAD

Fieure 1 b Western blot demonstrating expression of BAD in COS cells. COS cells were transfected with murine BAD. Twenty-four hours after transfection, cellular protein was extracted. Protein lysates were separated on a 15% polyacryiamide gel and transfened to a membrane filter. BAD was detected with a polyclonal nbbit anti-BAD antibody. Membranes were probed with polyclonal nbbit anti-actin to demonstrate equal protein loading. Secondary anti-mouse antibodies were applied. The membranes were developed using an ECL detection kit.

Untreated

O empty vector

COS COS Bcl-2 COS Bcl-xL

cell line

Empty Vector BAD

Figure 2 BAD induced ce11 death as measured by a decrease in GFP expression. COS, COS Bcl-2 and COS Bcl-xL were transfected with pTracer-CMV BAD or empty vector. Forty-eight hours after transfection, adherent cells were harvested and GFP expression arnong viable celb was measured by flow cytometry. a) The number of cells expressing GFP d e r transfecting BAD was expressed as a percentage of the number of cells expressing GFP after transfecting the empty vector. The transfections were performed in ûiplicate. The results shown are the mean of three tram fec tions. Bars represent 1 standard error (SE) b) A representative flow cytometry experiment of COS cells transfected with pTracer- CMV BAD or empty vector.

respectively). Transfection of antisense BAD in pTncer CMV yielded GFP expression

equivalent to the empty vector (data not shown). The reduction in the percentage of cells

expressing GFP after BAD tnnsfection coincided with a reduction in the mean channel

fluorescence for GFP demonstnting that only cells weakly expressing GFP remained

after BAD transfection.

A dose response curve for BAD-induced ce11 death was constructed. COS cells were

tnnsfected with increasing amounts of pTracer-CMV BAD and GFP expression was

measured 48 hours after transfection. Empty vector was supplernented to maintain an

equivaleni DNA concentration of 2ug for the tnnsfections. Transfection of COS cells

with increasing amounts of pTncer-CMV BAD caused a corresponding decrease in GFP

expression (Figure 3).

I 1 I I

0.5 1 1.5 2

BAD cDNA (ug)

Figure 3

Dose response curve for BAD-induced apoptosis. COS cells were tmnsfected with increasing amounts of pTracer-CMV BAD. Empty vector was supplemented to maintain an equivalent DNA concentration of 2ug. Forty-eight houn after transfection, adherent cells were harvested the percentage of viable cells expressing GFP was determined by flow cytometry. The trûnsfections were performed in tnplicate and the results shown are the mean of three transfections. Bars represent 1 SD

Floating cells COS cells normally adhere to the tissue culture plate. but they round up. detach from the

plate, and float in the medium above when they lose viability. 1 measured ce11 death by

determining the number of ceils that were floating in the medium, An increase in tloating

cells has been reported as a marker of ceIl death after exposing ridherent cells to

chemotherapy2', heat shock2', c ytokines'4 and ionizing radiation? Desjardins et al".

determined that the detachment of HT29 cells from the tissue culture plate wlrs a late

event in the apoptosis cascade and correlated with fragmentation of the genornic DNA

into 180- 120bp oligomen.

Transfection of pTracer CMV -BAD into COS, COS Bcl-2, or COS Bcl-xL increased the

number of floating cells 8.7 2 1.7 fold, 5.2 t 1 and 5.2 21.3 fold respectively compared to

transfection of the empty vector (Figure 4). These increases were significant by the t-test

(p = 0.001.0.02, and 0.02 respectively). Transfection of antisense BAD did not increase

the number of floating cells compared to transfection of the empty vector. The floating

celIs were replated in fresh medium to confirm thrit they were nonviable. No cells

adhered ro the plate and no colonies grew after one week of observation. In contrast.

viable COS cells adhere to the tissue culture plate within 18 hours of plating and fom

colonies within a week.

L

L cn O L œ - a O $15 1 0 -

7 - .E 02 4 - 3 . 5 , O , € I 1 t - - QS O COS COS Bcl-xL COS Bcl-2

cell line

Figure 4

BAD induced ceIl death as rneasured by an increase in the number of floating cells. COS, COS Bcl-2, and COS Bcl-xL were tmnsfected with pTracer-CMV BAD or empty vector. Forty-eight hours after transfection, the number of floating cells were counted. The number of floating cells after transfection of BAD are expressed as ii fold increase over the number of floating ceils after transfection of the empty vector. The results shown are the mem of three transfections. Bars represent I SD.

Cionooerric assay

To funher study BAD-induced cell death in cells over-expressing Bcl-2, I performed a

clonogenic colony formation assay. Forty-eight hours after transfection of pTmcer-CMV

BAD or the empty vector into COS Bcl-2, the cells were harvested and equal volumes

were plated into 100mm3 plates. This volume was intended to represent 400 cells pnor to

transfection. One week later, the number of colonies was counted. Transfection with

BAD yielded 1 15 14 coionies whereris tnnsfection wi th the empty vector yielded 192 &

3 colonies. This difference was significant by ri t-test (p=0.0006). The reduction in the

number of colonies formed is less than the reduction in GFP expression. The reduction is

less, because tnnsfection efficiencies were approxirnately 35% as measured by GFP

expression after transfection of the empty pTracer CMV vector. Therefore, in the colony

formation assay, the majority of cells plated were not transfected. In contnst, only

successfull y transfected cells were assayed in the GFP expression experirnent.

BAD induces ce11 death in hematopoietic ceIl Iines

Next, 1 wished to determine whether BAD could induce death in hematopoietic ceIl lines.

K562 is a leukemic ce11 line with high levels of BCI-XL'~ and 32D BCI-2 is a murine

mye loid precunor ce l l line over-expressing ~ c l - ~ " ( ~ i ~ u r e 5). These cells were co-

transfected with equal concentrations of BAD in pcDNA3 or empty vector and the

luciferase reporter gene under control of the RSV promoter. Forty-eight houa after

tnnsfection, lucifemse activity was measured. Transfection of BAD into K562 and 32D

Bcl-2 cells reduced the luciferase activity 4.29.8 and 5S+L -9 fold respectively compared

to transfection of the empty vector. Transfection of BAD in the anti-sense direction

produced similar results as the empty vector.

Actin - Actin -

m Western blot demonstrating expression of %cl-2 and Bcl-xL in 32D Bcl-2 cells and K562 cells. Pro- tein lysates from 32D Bcl-2 cells and K562 cells were separated on a L5% polyacrylamide pl and transfened to a membrane filter. Bcl-2 was detected with a polyclonal nbbit anti-mouse BcI-2 anti- body and Bcl-xL was detected with a monoclonal rabbit anti-Bcl-x antibody. Membranes were probed with polyclonal rabbit anti-actin to demonstrate equd protein loading. Secondary anti-rabbit antibodies were applied. The membranes were developed using an ECL detection kit.

O ! I 1 1

O 0.25 0.5 1

BAD cDNA (ug)

Figure 6

BAD induced ce11 death in K562 cells in a dose dependent rnanner. K562 celis were transfected with increasing doses of BAD in pcDNA3 and the tuciferase reporter gene. Empty vector was supplemented to maintain an equal DNA concentration. Forty eight hours after trmsfection, the cells were harvested and luciferase activity was measured. Results are expressed as a percentage of the luciferase activity after trmsfection of empty vector. Each point represents the mean of three transfections. Bars represent 1 SD.

A dose response curve was constrrrcted for the ef fe ts of BAD an K562 cetls. As t he

dose of BAD cDNA was increased, there was a corresponding decrease in luciferase

activity (Fig 6).

BAD induces apo~tosis but does not alter 4 Y

WC wished to determine whether the increased toxicity observed after transient

transfection of BAD was due to an increase in apoptosis. Apoptosis was mecisured by

staining for fragmented genomic DNA with the H33342 stain (Figure 7). Forty-eight

hours after transfection, the adherent and floating cells were harvested sepuately and

examined for apoptotic bodies using H33342. Transfection of BAD into COS, COS Bcl-

2, and COS Bcl-xL increased the percentage of apoptotic floating cells compared to

transfection of the empty vector. In addition, transfection of BAD increased the number

of floating cells as stated above. BAD over-expression increased the number of apoptotic

cells still adherent to the tissue culture plate. The absolute percentage of apoptotic ceIls

still adherent to the tissue culture plate was relatively low, because the majority of the

cells on the plate were not successfully transfected and many of the apoptotic cells had

detached from the plate and were floating in the medium. Transfection of the empty

vector produced a basal level of apoptosis ihat was attributed to the ~ransfection

procedure and the Lipofectamine reagent.

Mitochondna play an important role in the apoptosis pathway. Feiitures of

mitochonàrial dysfunction in apoptosis are loss of AThi and release of cytochrome c into

the cytosol. Previous reports have indicated that apoptosis produced by Bax and Bak is

accompanied by loss of AY 51 and cytochrome c r e l e a ~ e ' ~ ~ ~ ~ . In addition,

unphosphorylated BAD translocates from the cytosol to the mitochondria where it

C

mmpty vector l BAD

I

COS COS Bcl-2 COS Bcl-xL

cell Iine

O empty vector

BAD

COS COS Bcl-2 COS Bcl-XL

cell line

Figure 7: BAD induced apoptosis as demonstr~ted by staining with H33342. COS, COS Bcl-2, and COS Bcl-xL were transfected with pTracer-CMV BAD or empty vector. Forty-eight hours after transfection, floating and adherent cells were harvested sepmtely. The cells were stained with H33342 and the cells observed under ultraviolet light for the presence of apoptotic bodies. The percentage of apoptotic cells (A) Roating cells and (B) adherent cells was determined. Results show represent the mean of at least t h e transfections. The di fference between the number of apoptotic adherent cells after transfection of BAD or the empty vector into COS, COS Bcl-2 or COS Bel-xL is significant by the t-test (p = 0.0003,0.003,0.022 respectively). The ciifference between the number of apoptotic floating cells after transfection of BAD or the empty vector into COS, COS Bcl-2 or COS Bcl-xL is significant by the t-test (p = 0.02,0.0001.0.032 respectively). Bars represent 1 SD

interacts with Bcl-2 and induces apoptosis30. The impact of BAD on A Y M and

cytochrome c release wüs not measured. Therefore, I wished to determine whether the

apoptosis induced by BAD is associated with these features.

Thirty hours after transfection of BAD cDNA . while the majority of the cells still

remained adherent to the tissue culture plate, adherent and floating COS cells were

harvested and stained with DiIC l(5). Loss of A Yzl was rneasured by fIow cytornetry.

No increase in the percentage of cells displaying loss of A Y,\! was observed after

transfection of BAD compared to transfection of the empty vector despite an increase in

the number of cells staining positive for propidium iodide (Figure 8). A subpopulation of

cells with loss of A was also not detected among the cells expressing GFP after

transfecting BAD in pTracer. Similar results were obtained when the cells were

harvested 48 hours after transfection. At this time point. more cells stained positive for

propidi um iodide, but no loss of AYM was observed. Li kewise, no increased loss of A <YAv

was observed in K562 cells transfected with BAD compared to empty vector.

Cytochrome c release was measured frorn COS cells after transfection with BAD

or empty vector. Thirty hours after transfection, the cells were harvested and the proteins

extracted. Cytochrome c wüs measured in the protein lysate by immunoblotting (Figure

9). Transfection of BAD increased cytochrome c release 2.2 fold compared to

transfection of the empty vector as measured by densitometry. The increase in

cytochrome c release after transfection of the empty vector compared to untreated cells

likely reflects the basal level of apoptosis produced by the transfection procedure.

untreated empty vector BAD

Figure 8 BAD does not alter A PM. Thirty hours after transfection of BAD in pcDNA3 or empty vector, the cells were harvested and stained with propidium iodine and DiICl(5). Loss of A YM was measured by flow cytometry. A representative experiment is shown.

cytochrome c

Actin

untreated empty vector

Fimue 9. BAD induces cy toc~ome c release. COS cells were transfected with BAD in pcDNA3 or empty vector. fhirty hours afler transfection, the floating and adherent ceils were harvested and the proteins extracted. Equal concentrations of SDS protein lysates were separated on a 15% polyacrylamide gel and transfened to a membrane filter. Cytochrorne c was detected with a polyclonal mouse anti-pigeon cytochrome c antibody. Membranes were probed with polyclonaI rabbit anti-actin to demonstrate equal protein loading. Secondary anti-mouse or anti- rabbit antibodies were applied. The membranes were developed using an ECL detection kit.

We adopted a transient transfection mode1 to assess the impact of BAD on cells over-

expressing Bcl-2 and Bcl-xL. 1 began my studies with COS cells because they can be

transfected with high efficiency and thus, they serve as a useful system in which to assess

ceIl viability after transient tnnsfection. Using this system. 1 demonstrated that BAD

induced apoptosis in ce1 1s over-ex pressing Bcl-2 and Bcl-xL. Having refuted the notion

that Bcl-2 over-expressing ceils are nsistant to BAD-induced apoptosis, I extended my

work to dernonstrate that BAD induced cell death in malignant hematopoietic ce11 lines

over-expressing these survival factors.

The ability of BAD to induce apoptosis in cells over-expressing Bcl-2 is

predicted, but kas not been shown previously. In fact, just the opposite is sugpsted in the

first description of BAD by Yang et al1'., who reported that BAD could modulate

apoptosis in cells over-expressing Bcl-xL but not Bcl-2. E U . 13 cells stably over-

expressing BAD and Bcl-xL underwent apoptosis efter IL-3 withdnwal. In contrast, the

same ce11 line with stable over-expression of BAD and Bcl-2 remained viable after

cytokine withdnwal. The protective effects of Bcl-2 were supported by Ottilie et al9. In

a vansient transfection assay, BAD induced cell death in 293 cells over-expressing Bcl-

xL, but could not induce ce11 death in the same cells over-expressing Bcl-2. In the

experiments with cells over-expressing Bcl-2, there was a high background of ce11 death

due to the transfection method and this high background may have obscured the effects of

BAD. As a consequence of these studies, other reports of BAD and apoptosis have only

evaluated the effects of BAD in cells over-expressing B C I - X L " + ~ ~ .

A few possibiIities rnay exphin why I was able to demonstrate BAD-indnced

apoptosis in cells over-expressing Bcl-2 when others have not. First. the high levels of

BAD expression in my transient transfection mode1 rnay have overcome the relative

lower affinity of BAD for cl-z9. Second the use of a transient transfection assay may

have allowed us to see IeveIs of apoptosis not apprecirtted in stable transfection

expenments used by others. Dunng the selection of stable ce11 lines. BAD may have

induced apoptosis in a portion of cells. such that, at the end of selection, only weak

expressen or resistant cells remained. Finally, it is possible that B A D may induce

apoptosis through a Bcl-2 independent pathway and this pathway becomes more evident

when BAD is expressed at high leve1s.

Mitochondrial dysfunction is a well described phenornenon in the apoptosis

pathway. An element of mitochondrial dysfunction is loss of the inner mitochondrial

membrane potential-termed loss of A Y,%[. Bax and Bak- induced apoptosis is associated

with loss of A Wang et al3' demonstnted that calcineurin dephosphorylates

BAD and permits its translocation to the mitochondnal outer membrane where it interacü

with Bcl-2. Translocation of BAD was associated with apoptosis. However, they did not

mesure A Yibf or cytochrome c release in this report. fn my study, BAD induced

cytochrome c release without loss of A Y1bf thiny houn or more after transfection.

Apoptosis and cyrochrome c release have been reporred in the absence OF loss of A

For example, the BH3-only family memben Bid and Bik induce cytochrome c release and

induce the formation of apoptotic bodies without dtentions in A y,vI3'. It would appear,

therefore, that the interaction of BH3-only family mernbers with BcI-2 at the

mitochondria induces apoptosis rhrough a pathway that does not invotve dismption of the

mitochondnal membrane electrical gradient.

Whether mitochondnal disruption leads to loss of AY likely depends on the site

of damage within the mitochondna. For example, damage to the adenine nucleotide

trmsporter that pumps protons and ATP across the relatively impermeable inner

mitochondnal membrane and maintains AYhf, leads to permeabilization of this

-7-34 membrane, mitochondnal swelling, loss of AYu, and ultimately cytochrome c release3-

In contrast, damage to the voltage-dependent anion channel at the outer mitochondnal

membrane c m lead to cytochrome c release wi thout disrupting the inner membrane

remiuns or inducing loss of A Y ~ ~ ' . Through a similar mechanism. the pore forming pro-

apoptotic proteins Bax or Bak can form channels that increase the permeabilization of the

outer membrane and induce cytochrome c release without damaging the inner

36-38 membrane .

There are a number of limitations to this study. First. 1 did not demonstrate that

Bcl-2 or Bcl-xL protected COS cells from other apoptotic stimuii. However, in a

previous report. COS cells over-expressing Bcl-2 or Bcl-xL were protected from

etoposide-induced apoptosis3g. Furthetmore. in my study, BcI-2 and Bcl-xL offered some

protection against BAD-induced ce11 death as measured by enumerating the increase in

floating cells after transfection of BAD. Second, cytochrome c release was measured in

cytosolic extracts of cells lysed with RIPA buffer. Theoretically. the detergents in the

RIPA buffer may have solubilized the mitochondria and released cytochrome c.

However, untransfected ceIIs and cells transfected with empty vector of BAD cDNA were

treated equally and no cytochrome c release was detected in the untreated controls. To

could have been lysed by mechanicd rneans and a tontrol antibody directed against COX

TV (an inner mitochondnal membrane protein) could have been used. Third. this study

Iacked a positive control such as cyanide rn-chlorophenylhydrazone (CCCP) to

demonstrate that COS celIs cm undergo loss of AYbi.

From my study. 1 conclude that BAD directly induces apoptosis in cells over-

expressing Bcl-2 and Bcl-xL. Furthermore, BAD-induced apoptosis is associated with

cytochrome c release but not loss of AYM suggesting that i t facilitates damage to the

mitochondria confined to the outer membrane. Since B A D is capable of inducing

apoptosis in cells over-expressing Bcl-2 it furthen the goal of using BAD as a prototype

to develop broadly acting therapeutic agents to induce apoptosis in malignant cells.

Attempts to develop therapeutic compounds that mimic BAD are currently underway.

Small synthetic molecules capable of transversing ce11 membranes that act like BAD by

binding and inhibiting Bcl-2 could be useful therapeutically40. Altematively, peptides

that correspond to the short death domains of pro-apoptotic proteins could be delivered

into cells using intemalization sequences such as ~ n t e n n a ~ e d i a ~ ' or fatty acid moieties

ACKNOMZEDGEMENTS

We thank Trude Nicklee for assistance photographing the cells stained with H33342.

Authorization to include this material in the thesis has been granted by the CO-authors and the publishen of Leukemia and Lymphoma. Contribution to manuscript: ADS conceived of the study, executed and analyzed the experiments and wrote the paper. DWH, NP and SC assisted in the design, execution and interpretation of the experiments pertaining to flow cytometry. They also reviewed the manuscript and provided cntical feedback. MDM supervised the work and provided cntical feedback that enhanced the design and execution of the snidy.

REFERENCES

1. Pepper, C, Bentley, P. and Hoy, T. (1996) Regulation of clinical chemoresistance

by bcl-2 and bax oncoproteins in B-ceIl chronic lymphocytic leukaemia, British

Journal of Haeriratology, 95, 5 13-5 17.

1. Sangflet, O, Osterborg, A, Grander, D, Anderbnng, E, Ost. A. Mellstedt. H.

Einhom, S. (1995) Response to interferon thenpy in patients with multiple

myeloma correlates with expression of Bcl-2 oncoprotein. Internatzonal Joitmul

of Cancer, 63, 190- 192.

3. Kitada. S. T~kayama S, De Riel. K. Tanaka. S, Reed K. (1994) Reversa1 of

chemoresistance of 1 ymphoma cells by antisense-rnediated reduction of bcl-2 gene

expression. Antisense Research and Development. 4.7 1-79.

4. Piche, A, Grim, J, Rancourt, C, Gomez-Navarro, J, Reed JC, Curiel, DT, (1998)

Modulation of Bcl-2 protein levels by intracellular anti-Bcl-2 singlechain

rntibody increases drug-induced cytotoxicity in the breast cancer ceIl line MCF-7.

Cancer Research, 58,2 134-2 140.

5. Miyashita, T and Rccd, 3. f 1993) BcI-2 oncopmiein blocks chemotherapy-ioduced

apoptosis in a human leukemia ce11 line. Blood, 91, 15 1-157.

6. Tang, C, Willingham, MC, Reed, IC, Miyashita, T, Ray, S. Ponnnathpur, V,

Huang, Y. Mahoney, ME, Bullock, G, Bhalla, K. (1994) High levels of p26BCL-

2 oncoprotein retard tacol-induced apoptosis in hurnan pre-B leukemia cells.

Leukemiu, 8, 1960- 1969.

7. Maung. ZI: Maclean, FR, Reid, MM. Pearson, RD, Proctor, SJ, Hmilton, W.

Hall, AG. (1994) The relationship between bcl-2 expression and response to

chemotherapy in acute leukaemia. British Journal of Haeniatology, 88, 105- 109.

8. Campos, L, Rouault, JP. Sabido. O, Pascale. O, R0ub.i N, Vasselon, C,

Archimbaud, E, Magoud, JP. Gutotat. D. (1993) High expression of bcl-2 protein

in acute myeloid leukemia cells is associated with poor response to chemotherapy.

Blood, 81,309 1-3096.

9. Ottilie, S, Diaz, J, Home, W, Chang, J, Yan. W. Wilson G, Chang, S. Weeks, S.

Fritz, LC, Oltersdoff, T. ( 1997) Dimerization propenies of human BAD. Journal

of Biological Chemistry, 272,30866-30873.

10. Yang, E, Zha, I, Jockel, J, Boise, LH. Thompson CB, Konmeyer, S. (1995) Bad.

a heterodimenc partner for Bcl-xL and Bcl-2 displaces Bax and promotes ceIl

death. Ceil, 80,285-29 1.

I l . Zha, J, Harada, H, Yang, E, Jockel, J, Konmeyer. S. (1996) Senne

phosphorylation of death agonist BAD in response to survival factor resuits in

binding to 14-3-3 not Bcl-xL. Cell, 87,619-628.

12. Etangovan, 0 and Chinnaduni, G. (1997) Funetional dissection of the pro-

apopto tic protein Bi k. Joiirnal of Biological Chemistry, 272,24494-24498.

13. Hegde, R, Srinivasula, S, Ahmad, M, Fernandes-Alnemri. T, Alnernri ES. ( 1998)

Blk, a BMcontainin mouse protein that intencts with Bcl-2 and Bcl-xL. is a

potent death agonist. Journal of Biological Chemistry, 7783-7786.

14. O'Cmo-rL, StasserA, O'Reitkyt, H a m , G, Adrrms, JM, Corry, S, Huang,

DC (1998). Bim: a novel member of the Bcl-2 family that promotes apoptosis.

EMBO J, 17,384-395.

15. Wang K, Yin X, Chao D, Milliman, CL, Konmeyer, SJ. (1996) BID: a novel BH3

domain-only death agonist. Genes and Developmnt, 10. 2859-2869.

16. Inohara. N, Ding. L, Chen, S, Nunez, G. (1997) Harakiri, a novel regulûtor of ce11

death, encodes a protein that activates apoptosis and intencts selectively wi th

survi val-promoting proteins Bcl-2 and Bcl-xL. EMBO Journal, 16, 1686- 1694.

17. Kelekar, A, Chang, B, Harlan, J, Fesik SW, Thompson, CB. (1997) Bad is a BH3

domaincontaining protein that fonns an inactivating dimer with Bcl-xL.

Molecular and Cellltlar Biology, 17,7040-7046.

18. Zha, J, Harada, H, Konstantin, O, Jockel, J, Waksman, G, Korsmeyer, S. (1997)

BH3 domain of BAD is required for heterodirnerization with Bcl-xL and pro-

apoptotic activity. Jmmal of Biological Chemistry, 272, 24101-24104.

19. Desjardins, L, and MacManus, 1. (1995) An adherent ce11 mode1 to study different

stages of apoptosis. Erperimenr Ce11 Research, 216,380-3.

20. Sheng-Tanner, X, Bump, EA, Hedley, DW. ( 1998) An oxidarive stress mediated

death pathway in imadiated human leukemia cells mapped using multilaser flow

cytometry. Radiation Research, 150,636-647.

21. Hakem, R, Hakem, A, Duncan, GS, Henderson, JT, Woo, MT Soengas, MS, Elia,

A, Luis el ai Pompa, J, Kagi, D, Khoo, W. Potter, J, Yoshida, R, Kaufman, SA,

Lowe, SW, Penninger, JM, Mak, TW. (1998) Differential requirement for

caspase-9 in apoptotic pathways in vivo. Cell, 94,339-352.

22. Ra;iphurst,G, Mao,I, Yang, RGoet, R, N i h f s , B, Shiim, FH, Yazd& HM,

Rippstein, P, Ng, C. (1998) Evaluation of apoptosis in four human tumour ce11

lines with differing sensitivities to cisplatin. Anticancer Research, 18, 2945-295 1.

23. Chow, A. and Zhang, R. (1998) Glutamine reduces heat shock-induced ce11 death

in rat intestinal epithelial cells. Journal of Nzitrition, 128, 1296- 130 1.

24. Khosla, S, Oursler, M, Schroeder M, Eberhardt, NJ. (1994) Transforming growth

factor-beta 1 induces growth inhibition of a human medullary thyroid cminoma

cell line despite an increase in steady state c-myc messenger ribonucleic acid

levels. Endocrinology, 135, 1887- 1893.

25. Soldritenkov, V, Pmsad, S, Notario, V, Dritschilo. A. (1995) Radiation-induced

apoptosis of Ewing's sarcoma cells: DNA fragmentation and proteolysis of

poly(ADP-ri bose) polyrnense. Cancer Research, 55,42404242.

26. Benito, A, Silva, M, GriIlot, D, Nunez, G, Fernandez-Luna, JL. (1996) Apoptosis

induced by erthroid differentiation of human leukemia cell lines is inhibited by

Bcl-xL. Blood, 87,3837-3843.

27. Gommerman, JL and Berger, SA. (1998) Protection from apoptosis by Steel factor

but not interleukin-3 is revend ihrough blockage of calcium influx. Bkwd, 91,

1891-1900.

28. Jurgensmeierr, JM, Xie, 2, Devaraux, Q, Ellerby, L, Bredesen, D, Reed, J. (1998)

Bax directly induces release of cytochrome c from isolated mitochondria.

Proccedings of the National Academy of Science, USA, 9,4997-5002.

39. Shimizu S, Narita M, and Tsujinioto Y. (1 999r Bcl-2 farrrity proteins regtitate the

release of apoptogenic cytochrome c by the mitochondrial channel VDAC.

Nature, 6735,483-487.

30. Wang, H, Pathan, N, Ethell, IM, Krajewski, S, Yamaguchi, Y, Shibasaki, F,

McKeon, F, Bobo, T, Franke, T, Reed, K. (1999) ca2'-induced apoptosis through

cdcineurin dephosphorylation of BAD. Science, ZM, 339-343.

3 1. Shimizu, S. Tsujimoto. Y. (2000) Proapoptotic BH3-only Bcl-2 famiiy rnembers

induce cytochrome c release, but not mitochondnai membrane potential loss. and

do not directly rnoduIate voltage-dependent anion channel activity. Procceclings of

the Nutional Acaderny of Science, USA, 97,577-582.

32. Naita, M, Shimizu, S, Ito, T, Chittenden, T, Lutz, EU, Matsuda, H, Tsujimoto, Y.

(1998) Bax interacts with the permeability transition pore to induce permeability

transition and cytochrome c release in isolated mitochondria. Proccedings of the

National Academy of Science, USA, 95, 1468 1 - 14686.

33. Brenner, C, Cadiou, H, Vieira, HLA, Zarnzami, N, Marzo, 1, Xie, 2, Leber, B,

Andrews, D, Duclohier, H, Reed, JC, Kroemer, G. (2000) Bcl-2 and Bax regulate

the channe1 rrctivity of the mitochondna1 adenine nuclmtide tnnstocriror.

Oncogene, 19,329-336.

34. Li, YZ, Li, CJ, Pinto, AV, Pardee, AB. (1999) Release of mitochondrial

cytochrorne c in both apoptosis and necrosis induced by B-Lapachone in human

carcinoma cells. Moiecrtlar Medicine, 5,232-239.

35. Pastorino JG, Tafani, M, Rothman RJ, Marcineviciute A, Hoek, JB, Farber, JL.

(1999) Functional consequences of the sustained or transient activation by Bax of

the mïtochond'riat pemeability transition pore. T h e J m a / of Biotogi~l

Chemistry, 44, 3 1734-3 1739.

36. Antonsson, B, Conti, F, Ciavatta, AM, Montessuit, S, Lewis, S, Martinou, 1,

Bernasconi, L, Bernard, A, Merrnod, JJ, Mazzei,G, Maundrell, K, Gmbale, F,

Sadoul, R, Martinou, JC (1997) Inhibition of Bax channeI-forming activity by

BcI-2. Science, 277,370-373.

37. Eskes, R, Antonsson, B, Osen-Sand, A, Montessuit, S, Richter, C, Sadoul, R,

Mazzei, G, Nichois, A, Martinou, JC (1998) Bu-induced cytochrome c release

from mitochondria is independent of the permeability transition pore but highly

dependent on M~'' ions. n i e Jmmal of Ce11 Biology, 143,2 17-224.

38. Basanez, G, Nechushtan, A, Drozhinin, O, Chanturiya, A, Choe, E, Tutt, S.

Wood, KA, Hsu, YT, Zimrnerberg, J, Youle, R.J (1999) Bax, but not Bcl-xL,

decreases the lifetime of planar phospholipid bilayer membranes at subnanornolar

concentration. Proccedings of the National Academy of Science, USA ,96,5492-

5497.

39. Kitanaka, C, Narniki, T, Noguchi, K, Mochizuki, T, Kagaya. S, Chi. S. Hayashi.

A, Asai, A, Tsujimoto, Y, Kuchino, Y ( 1947) Caspase-dependent apoptosis of

COS-7 cells induced by Bax overexpression: differential effects of Bcl-2 and Bcl-

xL on Bax-induced caspase activation and apoptosis. Oncogene. 15, 1763-7772.

40. Wang, JL, Liu, D, Zhang ZJ, Shan, S, Han. X, Srinivasula, SM, Croce, CM,

Alnernri, ES, Huang, 2. (2000) Structure based discovery of an organic compound

that binds Bcl-2 protein and induces apoptosis of tumor cells. . Proccedings of the

Natirna! Academy of Science, USA. P r o c - c e ~ g s of the Narionu! Acadmy of

Science, USA, 97.7 1 24-7 1 29.

41. Hollinger. EP. Chittenden. T, Lutz. RI. (1999) Bak BH3 peptides antagonize Bcl-

xL function and induce apoptosis through cytochrome c-independent activation of

caspases. The Journal of Biological Chemistry, 274. 13298- 13304.

42. Wang JL, Zhang ZJ, Choksi S, Shan S, Lu X, Croce CM, Alnermri ES, Korngold

R. Huang Z. (2000) Cell permeable Bci-2 binding peptides: a chernical approach

to apoptosis induction in tumor cells. Cancer Research 60, 1498- 1502.

THE BH3 DOMAIN OF BAD FUSED TO THE ANTENNAPEDIA PEPTIDE INDUCES APOPTOSIS VIA ITS ALPHA HELICAL

STRUCTURE AND INDEPENDENT OF BCL-2

Aaron D ~chimmer'. David W ~ e d l e ~ l , Sue chowl, Nhu-An Pham, ' Avijit chaknbarttyl, Denis ~ouchard'~'. Tak W. ~ a k l * ' , Michael R TI-US', Mark D ind den'

1. Princess Margaret Hospital, University Health Network, Toronto, ON, Canada 2. Arngen Institute, Toronto. ON, Canada

Running Title: The BE33 domain of BAD induces apoptosis via its alpha helical structure

ABSTIEACT Since the over-expression of Bcl-2 is a common cause of multi-drug resistance,

cytotoxic peptides that overcome the effects of Bcl-2 may be clinically useful. 1

haniessed the death-promoting alpha heiicd properties of the BH3 domain of BAD by

fusing it to Antennapedia (ANT) domain, which allows for cell entry (ANTBH3BAD).

Treatment of 32D cells with the ANTBH3BAD peptide results in a 99% inhibition of

colony formation. No significant toxicity is observed after treatment wi th ANT or

BH3BAD alone. A mutant fusion peptide unable to bind Bcl-2 induces cell denth as

effectively as the wild type ANTBH3BAD. Furthemore, 32D cells over-expressing Bcl-

2 show no resistance to the ANTBH3BAD peptide. Therefore. the toxicity of the peptide

was independent of the Bcl-2 pathway. I demonstrated that the toxicity of the peptide is

due to its alpha helicity that disrupts mitochondrial function. Since this peptide

overcomes major forms of dnig resistance. it may be therapeutically useful if

appropriatel y targeted to rnalignant cells.

Key words: apoptosis, alpha helix, rnitochondria, Bcl-2, caspases

INTRODUCTION

Since the over-expression of BcI-7 is a comrnon mechanisrn by which malignant

cells become resistant to multiple chemothenpeutic agents1", cytotoxic peptides that

overcorne BcI-3 mediated dmg resistwce may be useful in the treatment of malignanr

disease. One example is the alpha helical BH3 domain of BAD that binds and inhibits

Bcl-2 and Bcl-xL. Wang et aL8 created a cell-permeable version of the peptide by fusing

it to cpm, a decanoic fatty acid intemalization sequence. This fusion peptide induced

npid cell death and apoptosis. However, its cytotoxicity was significantly impaired by

the over-expression of Bcl-xL, which could limit the usefulness of this compound in

treating malignancies with increased levels of Bcl-2 and Bcl-xL. Therefore. enhancing

the ability of the BH3 domain of BAD to kill cells over-expressing these survivai factors

might be therapeutically useful. One strategy to enhance its toxicity is to capitalize on

the inherent alpha helical shape of the peptide, since some amphipathic alpha helices are

potent naturally occurring antibiotics. These compounds dismpt negatively charged

bacterial membranes, but are not toxic to the marnmalian cells, because their membranes

are only weakly negatively charged and stabilized by cholestero19-". When internalized

into mamrnalian cells. hoivever, the alpha helical peptide (KLALAK)2 induced ceil death

and apoptosis by disrupting the negatively charged mitochondria".

Here, 1 have harnessed the alpha helicd properties of the BH3 domain of BAD by

fusing it to the Antennapedia intemalization sequence. 1 have created a 37 amino acid

fusion peptide comesponding to the 21 amino acids of the human BH3 domain of BAD

fused to the C-terminus of the 16 arnino acids of the ANT peptide (ANTBH3BA-D).

ANT is an intemalization sequence that successfully translocates across a variety of ce11

membrimes with high efficimcy md h w toxi~it~'~~~~. f report bat the ANTBH3BAD

fusion peptide induces rapid ceIl death and apoptosis. Its alpha helical secondary

structure disrupts mitochondrial function and allows the peptide to overcome caspase

inhibitors and Bcl-2 over-expression.

EXPERIMENTAL PROCEDURES

Peptides

Peptides (Table 1) were synthesized comrnercially to my specifications (Amgen,

Colorado) and were purified to greater than 95% punty by reverse phase HPLC. Peptides

were resuspended in DMSO and stored at -20'~.

Real Tirne Surface Plasmon Resonance

Binding of Bcl-2 to the peptides was assessed using a BIAcore sensor system

(BIAcore, Uppsala, Sweden). ANTBH3BAD' ANT, BH3BAD. the LI LSA D 1 19R

mutant peptide, and BSA were diluted in t OrnM Na acetate pH 5 and their amine groups

were coupled to a CM5 sensor surface activated with 37.5mg/mL N-Ethyl-N'-

(3'dimethylaminopropyi) carbodiimide hydrochloride (EDC) and 7.5mg/rnL N-

hydroxysuccinimide (NHS). Peptides were added to saturate the chip. Remaining

reactive groups on the chip were then inactivated with LM ethanolamine pH 8.5. GST-

Bcl-2 (Santa Cruz) was diluted in running buffer (0.011M Hepes pH 7.4,O. 15 NaCI, 3mM

EDTA, and 0.005% polysorbate 20 (vlv)) and injected ont0 the sensor chip at varying

concentrations. All experirnents were performed at 2 5 ' ~ . Between injections of GST-

BCL-2, the chip was washed with running buffer and regenerated with LM NaCI and

50miui NaOH. Data were anaiyzed with tbe BIAevaluation software

Peptides

1 Intemîlization sequence

BH3 BAD

1 Non-helix mutant 1

BH3 BAD sequence - -

NLWAAQRYGRELRRMSDEFVD

NLWAAQRYGRELRRMSDEFVD

NLWAAQRYGREARRMSREFVD

NLWPPQRYGREPRRMSDEFVD

Cet/ tirres

HeLa cells were grown in Dulbeco's Modified Eagle Medium wi th high glucose

(DMEM H2 1). The T-ALL ceII line CEM and the vinblastine resistant CEM ce11 line

over-expressing p-glycooprotein (CEM VBL) were grown in RPMI 1640. A 32D ce11

line over-expressing Bcl-2 (a gift from S. Berger) was created by infecting murine

hematopoetic 32D cells with a retrovirus containing murine Bcl-2. Over-expression of

Bcl-2 was confirmed by Western blot. 32D cells were maintained in WMI 1640 and 2%

WeHi-3B conditioned medium as a source of IL-3. AI1 cell lines were supplemented

with 101 fetal calf serum, penicillin and streptomycin. Cells were grown at 3 7 ' ~ with

5% CO2 in a humid atmosphere.

Intemalkation of the Peptides

Biotinylated ANT or ANTBH3BAD peptides were added to 1x10~ 32D cells at a

final concentration of 50pM in semm free DMEM H11 (SF-DMEM). After 20 minutes,

t he cells were washed twice in PBS and fixed in 4% paraformaldehyde in PBS (pH 7.0)

for 10 minutes at roorn temperature. After fixation. celis were washed in PBS and

permeabilized with LI Triton X- 100 in PBS for 10 minutes at room temperature. Cells

were again washed in PBS and incubated with 3% bovine serum albumin in PBS for 1

hour. Streptavidin-FITC (5uL) (Pierce) in 3% bovine serum albumin in PBS was added

to the cells in a totd volume of 500uL and incubated at room tempenture for 30 minutes.

Cells were washed in PBS and andyzed by flow cytometry (FACScan: Becton Dickinson

Immunocytometry S ystems, CA) or mounted on glass slides and viewed with a

fluorescence (Leica DMLB) or a confocai fluorescence (Zeiss LSM 5 IO) microscope.

Ctmogenic Cutorry Assay

Suspension cells (5x IO*) in SF-DMEM H21 were treated with the peptides for 3

houn. Equal volumes of cells were plated in quadmplicate into 96 weli plates in

complete medium, with 0.85% methylcellulose. One week after seeding, the number of

colonies formed was counted. HeLa cells ( t x l d ) in DMEM H21 with 10% FCS were

seeded into 6 well plates. Twenty four houn Iater, they were treated with the peptides for

3 houn in SF-GhEM H21. Following treatment with the peptides, equal volumes of

cells were plated in triplicate into 100mm3 tissue culture plates. One week after seeding,

the medium was removed, the plates stained with methylene blue, and the number of

colonies fonned were counted. Yeast (S. cerevisiae) were grown overnight i n YPD broth

at 3 0 ' ~ . The next morning, 3 r d of culture was dded to 7mL YPD broth and grown at

3 0 ' ~ to i n OD 595 of 0.3-0.35. One ml of culture wûs diluted in PBS. Peptides were

added and incubated with the yeast for 2 hours at room temperature. Equal volumes of

cells were plated in triplicate onto YPD plates and incubated at 30 '~. Four days later, the

number of colonies formed were counted.

Annexin V Assay

Peptides were added to lx 106 32D cells in SF-DMEM at a final concentntion of

50pM for 30 minutes. ETïC conjugated Annexin V (5uL) (Pharmingen) and propidium

iodide (PI) (Sigma) (final concentntion SugImL) were added. The cells were incubated

for 20 minutes at room temperature in the dark and fluorescence was measured by flow

cytometzy (FACScan; Becton, Dickinson, hmunocytometry system, CA, USA). The

Lysis II program was used to analyze these data. Cells staining positive for Annexin V

and negative for PI were deemed apoptotic.

Circrltar Dicfrroim

ANT and ANTBH3BAD peptides were dissolved in O. 1M Na211P04 buffer, 0.1M

NarHP04 buffer and 1% SDS or O. IM Na2HP04 buffer and 40% (vlv) TFE at

approxirnately lrng/mL. Concentrations of the peptides were detemüned by tryptophan

absorbance16. Peptides were dispensed into a 1 cm path length cuvette. CD spectra were

recorded at room temperature on an AVIV CD spectrometer mode1 62A DS. Spectn

were collected with an averriging tirne of 10 seconds per point. Each point represents the

averaged e 1 lipticity value recorded over a 1 nm spectral interval.

Measlu-enlent of Mitochondrial Membrane Potential, Reactive 0-rygen In temediates and

Cytosolic Calcirim

Changes in Mitochondrial Membrane Potential (AThl) were detected by staining

the cells with DiIC l(5) (I,1'3,3,3',3'-hexiunethylindodicarbocyanine) at a final

concentration of 40nM and propidium iodide (final concentration 5ug/mL) as described

previously'7'19. Cells stained with DiICI(5) were incubated at 3 7 ' ~ for 20 minutes and

then iinalyzed by flow cytometry (Coulter Epics Elite, Coulter, Florida) by exciting at

633nm and measuring through a 675 t 20nm bandpass filter. While measunng changes

in AYbf over tirne. stained cells were maintained at 37'~.

Changes in reactive oxygen intermediate production, and cytosolic calcium were

detected by staining the cells with 5pM carboxy-DCFDA (carboxy-dichlorofIuroescin

diacetate), and 3pM indo- 1 AM, respectively. Carboxy-DCF excitation was at 488 nm

and fluorescence was measured using a 525 nm t lOnm bandpass filter. Ionized calcium

was obtained by electronic ratio rneasurement of indo-l AM ernission at 405nm and

5Xnm at an excitation of 360nm.

To rissess changes in h !&in the preseirce of a pmeabiked ceil membrane, 32D

cells were treated according to the rnethod of Pham et alf7. Bnefly, cells were suspended

in respiratory buffer (0.25M sucrose, 2rnM KH2P04, 5mM MgCl*, ImM EDTA, O. 1 %

fatty acid free BSA, 1 mM ADP, and 70mM MOPS (pH 7.4)). Digitonin (4 pg/mL final

concentration) was added to the cells for 5 minutes. Successful permeabilization was

documented by detecting entry of propidium iodine into the cells. Mitochondrial

function was driven by adding 5rnM succinate, a cornplex II subswte. BHSBAD

peptides were added at a final concentration of 50 ph4 and cells were incubated at 3 7 ' ~

for 15 minutes. A Kbtwas then measured by flow cytometnc analysis after staining with

40 nM DiIC l(5).

Caspase-3 Activation

Caspase-3 activation was rneasured with the Apo-Alert Caspase-3 nuorescent

assay kit (Clontech. CA) according to the manufacturers instructions. Cells ( l x lo6) were

treated with the peptides (50pM) for 2 houn and lysed. The caspase-3 substnte DEVD-

AMC was added to the lysate and incubated for one hour at 37'~. Release of free AMC

was measured in a fluorometer (Photon Techonology International) with a 400 nm

excitation filter and a 505 nm emission filter,

RESULTS

Bcl-2 binds ANTBH3BAD and BH3BAD, bris not ANT

To dernonstrate that .4NTBH3BAD binds Bcl-2, the interaction between Bcl-2

and the peptides was analyzed with teal time surface plasmon resonance detection.

ANTBWBAD, BH3BA.D and ANT peptides and BSA were coupled to a biosensor chip

through their amine groups. GST-Bcl-2 was injected at increasing concentrations onto

the sasor ch-@ and bmdirrg ro the peptides aba i r the BSA mntrot was m d . GST-

Bcl-2 bound avidly to ANTBH3BAD (kD = l x l ~ ' ~ ~ , by BiAevaluation software,

(BIAcore. Uppsala, Sweden)) and BH3BAD in a dose-dependent manner and did not

bind significantly to ANT (Figure 1). Bcl-2 remained tightly bound to the peptides with

minimal dissociation. GST done did not bind to the peptides.

Internali~atiori 0 f th e peptkles

ANT is an internalization sequence that facilitates the transport of peptides across

ce11 membranes with high efficiency. To study the intemalization and localization of the

peptides, 32D cells were treated with biotinytated venions of ANT and ANTBH3BAD.

In a representative experiment, 89% and 100% of 32D cells treated with N-terminally

biotinyated ANT or BH3BAD respectively intemalized the peptides as measured by flow

cytometry. The peptides were distributed diffusely throughout the nucleus and cytoplasm

as detennined by confocal fluorescent microscopy (Figure 2).

=INTBH3Brl D Indtces u Rapid Ce&/ Death

To assess the impact of the peptides on ce11 viability, HeLa cells were treated with

50pM of ANT, BH3BA.D or ANTBH3BAD for 3 hours. Cells treated with

ANTBH3BAD displayed rnorphologic evidence of ce11 death characterized by ce11

shrinkage, membrane blebbing, and detachment from the tissue culture plate. In contrast.

no significant morphological changes were seen atter treatment with 50pM ANT or

ANTBH3BAD. Cells treated with ANT or BH3BAD showed no evidence of toxicity

(Figure 3).

O 100 200 300 400 500 600 700

Time (seconds)

-- -

O 100 -200 300 400 500 600 700

Time (seconds)

Figure 1: Red time surface plasmon resonance BWBAD, ANT, ANTBH3BA.D and BSA were coupled to a CM5 BIAcore sensor chip. GST-Bcl-2 was injected ont0 the chip at increasing concentrations. The sensor surface was regenerated with 1M NaCl and 50mM NaOH between each injection. Binding to the peptides above the BSA control was measured as an increase in Response Units above the BSA control. A representative sensogram &ter the injection of GST-Bcl-2 (2pM) is shown (a). A dose response curve for the binding of GST-Bcl-3 to ANTBH3BAD is presented (b)

ANT

Figure 2: Intemaiization of the peptides Biotinylated versions of ANT and ANTBH3BAD were obtained. 33D cells (1x106 cells) were treated with the peptides a i a final concentration of 50pM for 20 minutes. The cells were washed, fixed in 4% parafonnaldehyde, permeabilized with 1 % Triton-X 100 and probed with Streptavidin- FITC to detect the biotinylated peptides. Cells were viewed under a confocal microscope with fluo- rescent Iight to determine the distribution of the peptide.

ANT BH3BAD

Figure 3: Morphologie evidence of ce11 death. HeLa cells (1x10~ cells) were seeded into 12 well plates in DMEM H21 + 10% fetal cdf serum. Twenty-four hours later, BEUBAD, ANT and ANTBH3BAD were added to the cells in SF-DMEM HX at a final concentration of SOW. Three hours after incubation, morphologie evidence of ce11 death was observed among the cells treated with ANTBH3BA.D. Cells treated with ANT or BH3BAD showed no evidence of toxicity.

To quantitate the effects of the peptides m cet1 snrvivat, H e h , 32D wd yeast were

treated with the peptides at a final concentration of 50ph4 and cell survival was measured

by a colony formation assay. Drnmatic decreases in cell survival were seen in al1 three

cell types after treatment with ANTBH3BAD. No significant toxicity was observed in

cells treated with ANT or BH3BAD alone (Figure 4). Therefore. ANTBH3BAD induces

ce11 death, but the toxicity is not related to the ANT domain. The BH3BAD peptide

alone is not toxic because it is not internalized.

Dose response curves for ANTBH3BAD-induced ce11 death were constructed.

32D cells were treated with increasing concentrations of ANTBH3BAD peptide for 3

hours and ceIl survival wris rneasured in a colony formation assay. As the peptide

concentration increased, there was a corresponding decrease in ce11 survival wiis observed

(Figure 5). The dose response curve was steep. as minimal toxicity was observed below

a peptide concentration of 1OpM and nearly complete growth inhibition was seen at a

concentration of 50pM.

ANTBH3BAD Induces Apoptosis

Next, 1 determined whether the cell death induced by ANTBH3BAD occumd via

apoptosis. 32D cells were treated with the peptides for 30 minutes and stained with FITC

labeled anti-Annexin V and PI. The percentage of apoptotic cells was determined by

flow cytomeuic analysis. After treatment with ANTBH3BAD 97+1%, of cells were

apoptotic based on Annexin V staining. In contrat, only 1522% of cells treated with

ANT and 1312% of cells treated with BH3BAD were apoptotic by Annexin V staining.

untreated BH3BAD ANT ANTBHSBAO

peptide (5OpM)

untreated BH38AD ANT ANTBH3BAD

peptide (50pM)

HeLa

S. cerevisia e

untreated BH3BAD ANT ANTBH3BAD

Figure 4: ANTBH3BAD induces celi death HeLa cells, 32D cells and S. cerevisiae were treated with BBBAD, ANT, or ANTBH3BA.D at a final concentration of 50p.M for three hours. After the incubation, equal volumes of cells were plated in a colony formation assay as descrïbed in "Experimental Procedures". The nurnber of colonies formed was counted. The results are a mean of at least three expenments. Bars equal SD

Figure 5: Dose response c w e A dose response curves for ANTBWBAD-induced ce11 death was consmicted. 32D cells were incubated with increasing concentrations of ANTBK3BAD for 3 hours. After incubation, equal volumes of cells were plated in a colony formation assay as described in "Experimental Frocedures". One week later, the number of colonies formed was counted.

ANTBH3BAD hdr~ces Ceif Deuth ttrdepmdent of Bct-2

We studied the effects of Bcl-2 over-expression on the toxicity of ANTBH3BAD.

32D cells with and without over-expression of rnurine Bcl-2 (5x10~ cells) were treated

with ANTBH3BAD and ce11 viability was measured by trypan blue exclusion. Over-

expression of Bcl-2 did not abrogate the toxicity of rny peptide (Figure 6). In contrast,

32D cells over-expressing Bcl-2 were protected from IL-3 withdrawal (data not shown),

indicating that Bcl-2 can protect this ce11 line from other apoptotic stimuli.

To confirrn the lack of impact of Bcl-2 on the toxicity of ANTBH3BAD. a mutant

fusion peptide was prepared in which residues LI 14 and Dl 19 in the BH3 domain of

BAD were substituted with A 1 14 and R 1 19. In a BIAcore binding assay. GST-Bcl-2 did

not bind to this mutant peptide. Although incapable of binding Bcl-2. the mutant peptide

was as toxic as the wild type indicating that the toxici ty of ANTBH3BAD is independent

of the Bcl-2 pathway.

Figure 6: BcI-2 does not abrogate the toxicity of the fusion peptide 32D cells and 32D cells over-expressing Bcl-2 were treated with ANTBMBAD at a final concentration of 10pM, 30pM and SOpM. Six hours after treatment, an equal volume of trypan blue (0.4% in 0.9% normal saline) was added for 10 minutes. At least 100 cells were scored. The percentage of cells staining positive for trypan blue was determined. The results represent the mean of three experiments. Bars equai SD

Tmicitp of the Fusion Pepnde is chte ta its Atphct Helicul Shape

Certain alpha helical peptides are potent antirnicrobial agents that induce ce11

death b y disrupting the negativel y charged bacterial ce11 membranegg10. When

intemalized into eukaryotic cells, afpha helical peptides induce apoptosis by disrupting

the negatively charged mi tochondna". ANTBH3BAD is predicted to have an alpha

helical conformation. To evaluate the contribution of the helical shape to its toxicity. 1

obtained, a non-helicai mutant of AiVTBH3BAD by substituting A106, A107, and Li 14

in BH3BAD with prolines. 1 tested the impact of this mutation on helical structure by

circula. dichroisrn (CD) (Figure 7). In 1OOmM phosphate buffer (pH 7). both the wild

type and non-helical mutant ANTBH3BAD peptides displayed a CD spectrum consistent

with a random coil. However, the wild type peptide adopted an alpha helical

conformation in 1% SDS and 40% (v/v) TFE, but the non-helical mutant mainly

remained as a random coil in these solutions. In addition, the ANT and BH3BA.D

peptides alone adopted alpha helical secondary structure in these solutions.

-30000 I l 1 1 1 1

200 210 220 230 240 250 260

Wavelength (nm)

Figure 7: Circular D i c b i s m secondary structure of ANTBH3BAD and the non-helical mutant were determined

by circular dichmism. The peptides were dissolved in 0.1M Na&P04 buffer (A), 1% SDS (B), or 40%(v/v) TFE (C). CD spectra were recorded at room temperature. Each point represents the averaged ellipticity value recorded over Inm spectrai intervais.

= ANTBH3BAD, - = non-helical mutant

A cotony formation a s s q was performedwittr the nm-hetic;it mutant to detemrine the

contribution of the alpha helicity to the toxicity of ANTBH3BAD (Figure 8). The non-

helical mutant was not toxic to 32D cells demonstrating that the toxicity of

ANTBH3BAD depends on i ts he 1 ica1 configuration.

ANTBHjBAD Disnlpts Mitochondriaf Funcrion

Alpha helical peptides induce swelling of isolated mitochondna". Therefore, to

confirm that the cytotoxcity of ANTBWBAD wûs related to its helical configurütion. 1

investigated the effects of the peptides on intracellular mitochondrial function.

HaIlmarks of mitochondrial dysfunction include Ioss of mitochondnal membrane

potential (A Yb/) and increased production of reactive oxygen intermediates (ROI).

Chariges in A !&and ROI production were measured over time by flow cytometry after

treating 32D cells with the peptides. ANTBH3BAD rapidly decreased A Tt1, whereas no

significant change was observed after treatment with ANT or BH3BAD (Figure 9). The

non-helical mutant also did not induce loss of A y\,. The population with decreased A q t l

also showed increased ROI production. The increased ROI and decreased A were not

accompanied by significant increases in cytosolic calcium. Therefore. the loss of A

mt secondary to large increases in cytosolic calcium from disruption of the endoplasmic

reticulum.

To ensure that the toxicity of ANTBMBAD is due to the alpha helical properties

of BH3BAD and not an artifact of the fusion peptide, 1 investigated the effects of

BH3BAD on A YArin pemeabilized cells. 32D cetls were suspended in respiratory buffer

to mimic the environment of the cytoplasm. The plasma membrane was permeabilized

with digitonin and mitochondrial activity was driven with succinate. BH3BAD (50 pM)

control ANTBH3BAD non-helica wild type mutant

Peptide (50pM)

Figure 8: The non-helical mutant is non toxic 32D cells (5 x 104 cells) were treated with ANTBH3BAD or the non-helical mutant at a final concentration of 50w. ARer a three hour incubation, equai volumes of cells were plated into complete medium with 0.85% methyl cellulose. One week after seeding, the nurnbcr of colonies formed was counted. Bars equal SD

O I! *--+--+--*

I i I 1

20 30 40 50

Time (min)

CCCP BH3BAD r

ANT

- ANTBH3BAD - -+-- ANT - BH3BAD

Figure 9: PLNTBH3BAD induces loss of A FM 32D cells (5xl0~cells) were ueated with BWBAD, ANT, ANTBH3BAD, at a final concentration of 50pM. Cells were also treated with the uncoupling agent carbonyl cyanide m-chlorophenylhydrazone (CCCP) (final concentration 100phQ to induce complete Ioss of ATM and thereb y serve as a positive control. A YM was detected b y staining with DiIC1 (40nm). Changes in A YM were m e a s u ~ d at various times after the addition of the peptides by flow cytometry. A representative time course for the change in A YMis shown (a). A representative flow cytomeaic expriment after the addition of the peptides is presented (b).

was addecl to the pmeabikzed cetts for 15 mhtes mdcharrges in b Yhfw.,werr measured

by flow cytornetry. BH3BAD induced a 4.6 & 0.3 fold loss of A compared to

treatment with digitonin and succinate alone. Therefore, when intemalized into cells and

exposed to the mitochondria, BH3BA.D also displays an ability to induce loss of A %j.

ANTBH3BAD Induces Caspase-3 Activation, but Caspase Inh ibitors do no f Protect

Against Ce21 Death

Mitochondnal darnage can activate caspases. To determine i f the cytotoxici ty of

ANTBH3 B AD is caspase-dependent, 1 s tudied the e ffect of ANTB H3B AD on caspase

activation. Treatment with ANTBH3BAD induced a 5fl.9 fold increase in caspase-3

activation compared to untreated cells. In contrat, no increase in caspase-3 activity was

seen after treatment with ANT or BH3BAD. Next, 1 investigated the effects of the

genenl caspase inhibitor zVAD-fmk on ANTBHfBAD-induced caspase activation and

cytotoxicity. Pretreatment with LOO pM zVAD-fmk for one hour prevented

ANTBH3BAD-induced caspase-3 activation. Retreûtment with zVAD-frnk also

prevented the death of 32D cells after withdrawal of L-3 indicating that zVAD-fmk is

capable of inhibiting apoptotic stimuli in this ce11 line. Interestingly, pretreatment with

zVAD-fmk did not inhibit ANTBH3BA.D-induced ce11 death (Figure 10). Therefore,

ANTBH3BA.D induces caspase activation, but in the presence of caspase inhibitors. it

rernains cytotoxic due to caspase-independent mechanisrns.

Figure 10: zVAD-fmk fails to prevent cell death

32D cells (5xldcells) were pretreated with 1ûûp.M zVAD-mik for one hour followed by the addition of ANTBH3BA.D (30pM and 50pM). After three hours, ce11 viability was determined by nypan blue exclusion. The results represent the mean of three experiments. Bars equal SD

Over-expression of p-glycoprotein is another important mechanism of multidmg

resistance and is a poor prognostic marker in patients with ~eukernia'~~". Peptides and

proteins can act as substntes of p-g~ycoprotcin2'-". Therefore, I studied the effects of p-

glycoprotein over-expression on the toxicity of the peptides. CEM cells and CEM VBL

cells over-expressing p-glycoprotein were treated with varying concentrations of

ANTBH3BAD. Compared to the CEM parent ce11 line. CEM VBL cells displayed

resistance to ANTBH3BAD that was most apparent at a final concentration of lOpM

ANTBH3BAD. Pretreatment of CEM VBL with the p-glycoprotein inhibitor

cyclosporine (final concentration I O p M ) for one hour restored the toxicity of

ANTBH3BAD to levels seen in the wild type CEM Iine (Figure I 1). Therefore, p-

glycoprotein limits the toxicity of this peptide, but the effect can be reversed with p-

glycoprotein inhibitors.

DISCUSSION As a prototype for a novel therapeutic agent, 1 studied the effects of the BH3

domain of human BAD fused to the ANT intemalization sequence. This fusion peptide

induced a rapid ce11 death in al1 cells tested including yeast. The peptide was toxic

because its alpha helical shape dismpted mitochondriai Function.

In a ceIl free system. the ANTBH3BAD peptide bound BcI-3 with higher affinity

than the BH3BAD peptide alone. A similar finding was reported with BH3BAD fused to

the fatty acid moiety cpms. The increased binding likely reflects the additionai

interaction of the positively charged ANT sequence with Bcl-2.

1 0 0 - V1 Q1 - O Cern wt

8 1 0 - - I cem vbl O O

1 i- I 1 i

ANTBHBBAD peptide (PM)

untreated CYa ANTBHJBA-D cya + ANTBH3BAD

treatment (cya I O u M , ANTBH3BAD 1 O p M )

Figure 1 1 : P-glyocprotein protects against ANTBHîB AD toxicity

a) CEM and the p-glycoprotein over-expressing ce11 line CEM VBL were treated with AN?', BWBAD, or ANTBH3BAD for three hours at various concentrations. Cells were then seeded in complete medium with 0.85% methylcellulose. One week after seeding the nurnber of colonies formed was counted. b) CEM VBL cells were pretreated with cyclosporine (cya) at a final concentration of i0p.M for one hour. ANTBH3BAD was added at a final concentration of 10m. Three hours after incubation, the cells were seeded in complete medium with 0.85% methylcellulose. One week d e r seeding the number of colonies fomed was counted.

Despite being capable of binding Bcl-2, the cytotoxicity of ANTBH3BAD was

independent of the Bcl-2 pathway. Cells over-expressing Bcl-2 showed no resistance to

the peptide and a mutant fusion peptide unable to bind Bcl-2 was as toxic as the wild

type. 1 could not demonstrate Bcl-2-dependence because the alpha helicai propenies of

the peptide overcarne the protective role of Bcl-2.

The cytotoxicity of ANTBH3BAD was not an artifact of the fusion peptide. but

nther related to the helical properties of BH3BAD. As evidence to this point. BH3BAD

was capable of inducing loss of A !&in permeablized cells.

Previous studies of full iength BAD^*'^ or the BH3 domain of BAD^^'.' have not

reported toxicity attributable to the alpha helical configuration. These studies

demonsüated that over-expression of Bcl-2 md Bcl-xL inhibits apoptosis induced by the

full-length protein and peptide. Furthemore, mutants that do noi bind BcI-2 are not

ic6-8 J- ' A number of explanations rnay account for my ability to detect toxicity due

to the alpha helicity of BH3BAD. First. fusing the BH3 domain of BAD to the A1NT

intemalization sequence rnay have targeted the peptide more effectively to the

mitochondria where its alpha helical shape could dismpt the inner membrane. BH3BAD

peptides coupied to the cpm fatty acid rnoiety were confinecf to the cytoptasm. In

contrast, my peptide was distributed diffusely throughout the cell. Second. the

concentration of BAD used in the previous studies rnay have been too low to detect the

effects of the alpha helix. If BH3 BAD-mutants unable to bind Bcl-2 or Bcl-xL were

used at higher concentrations, the cytotoxicity from the alpha helical shape rnight have

becorne apparent. Finally, the full-length protein BAD rnay constrain the death

promoting attributes of the alpha helicai BH3 domain.

In a previous report by ~ t f e rbye t at" the 14 arrrirro acid alpha hetix (KIALAK)2

induced ce11 death and apoptosis. The peptide induced mitochondrial swelling and

caspase-3 activation in isolated mitochondria added to cytosolic extracts but the effect on

A cbIwas not reported. In addition, the effects of Bcl-2 and p-glycoprotein on the toxicity

of (KLALAK)? were not studied. Here, 1 have demonstrated that ANTBH3BAD disrupts

rnitochondria and induces loss of A q t f in intact and permeabilized cells. Unlike the

(KLALAK)? peptide, the toxicity of ANïBH3BAD was not inhibited by the general

caspase inhibitor zVAD-fmk. ZVAD-frnk c m inhi bi t mi tochondnal induced caspüse

activation. but i t does not prevent the peptide from darnaging the mitochondrin. As

mitochondriai damage continues, the ability of the mitochondria to synthesize ATP is

lost. and, as a consequence, cells undergo a metabolic and caspase independent death2'-19.

The finding of toxicity not affected by Bcl-2 suggests that alpha helical molecules

such as ANTBH3BAD can be potent mti-cancer agents capable of overcoming important

mechanisms of multidrug resistance. To see if other mechanisms of multidnig resistance

impact the toxicity of the peptide. I studied the effects of p-glycoprotein. Over-

expression of p-glycoprotein protected against the toxicity of ANTBH3BA.D and this

protection was revened by adding the p-glycoprotein inhibitor cyclosporine. P-

giycoprotein is best known for its role in extruding chemotherapeutic agents, but it also

7' 24.303 1 exports peptides and large proteins such as IL-2 and gamma interferon-'

Therefore, p-glycoprotein may protect the cells from the toxicity of ANTBH3BA.D by

exporting it out of the cells. Alternatively. it may be abrogating the peptide's toxicity

through its d e in inhibiting apoptosis. P-glycoprotein inhibits apoptosis in response to

Fas ligand, ultraviolet irradiation, and serum starvation, and the addition of venpamil can

restore the apoptotic signab ~fttieçestirnut?~~~. Cumntly, it is unknown how it directly

protects cells from these apoptotic stimuli.

There are a number of limitations to this study. Fint. 1 did not demonstrate that

the nonhelical mutant was intemalized into cells. Theoreticaily, the mutations in the BH3

domain of BAD could have produced a fusion peptide that could not be internalized.

However, ANT-mediated internalization does not appear dependent on the fusion partner.

Multiple peptides with a variety of secondary structures have been intemalized with the

ANT intemalization sequence and there are no reported cases where the fusion partner

inhibits this intemalization. Future experirnents would include obtaining a biotinylated

version of the non-helical mutant and studying its ability to transverse cell membranes.

Second. the circular dichroism data indicate that the peptide is a random coi1 in the

aqueous buffer. Only in the presence of SDS or TFE does the peptide acquire a helicd

shape. The degree of helicity in the presence of ce11 membranes has not been tested.

Therefore. it is unknown what conformation the peptide has when inside the cell. Third. 1

did not demonstrate that the ANTBH3BAD peptide directly interacts with the

mitochondriril membrane. Although the peptide is distributed di ffusel y throughout the

cytoplasm and nucleus, we did not isolate rnitochondria membranes to determine if they

contained the peptide. Therefore the toxicity of the peptide may be due to novel

mechanisms and the effects on the rnitochondria are a secondary phenomenon. In the

future, 1 propose to isolate mitochondria from peptide treated cells to determine whether

the peptide is partitioned in this cellular fraction.

In conclusion, I describe a cytotoxic effect of the BH3 domain of BAD is likely

due to its alpha helical properties and have harnessed this cytotoxic action by fusing

two major foms of dmg resistance - Bcl-2 and caspase inhibitors. As such this

compound may have clinical utility if it cm be selcctively targeled. This work also

furthers the understanding of how aipha helical peptides induce apoptosis in mrunmalian

cells and investigates factors that impact their toxicity.

ACKNO WLEGMENT

We would like to thank R Lutz and E. Hollinger (Immunogn, MA) for their helpful

advice and discussion.

Authorization to inciude this material in the thesis has been granted by the CO-authors and the publishers of Ce11 Death and Differentiation.

Contribution to manuscript: ADS conceived of the study, executed and anaiyzed the expenments and wrote the paper. DWH. NP and SC assisied in the design, execution and interpretation of the experiments pertaining to flow cytometry. AC assisted in the design. execution and interpretation of the expenments pertaining to circular dichroism. DB assisted in the design. execution and interpretation of the experiments pertaining to surface plasmon resonûnce. Al1 authon reviewed the manuscript and provided cntical feedback. MDM supervised the work and provided criticai feedback that enhanced the design and execution of the study.

REFERENCES

1. Pepper C, Bentley P, and Hoy T (1996) Regulation of clinical chemoresistance by

bcl-2 and bax oncoproteins in B-cell chronic lymphocytic leukaemia. Br J

Haematol. 95: 5 13-5 17.

2. Smgflet O. Osterborg A. Grander D. Anderbring E, Ost A, Mellstedt H. and

Einhom. S (1995) Response to interferon therapy in patients with multiple

myelomn correlates with expression of Bcl-2 oncoprotein. Int J Cancer. 63: 190-

192.

3. Kitada S. Takayama S. De Riel, K, Tanaka. S. and Reed IC (1994) Revend of

chemoresistance of lymphorna cells by antisense-mediated reduction of bcl-2

gene expression. Antisense Res Dev. 4: 7 1-79.

4. Piche A, Grim J, Rancourt C, Gomez-Navarro J, Reed JC, and Curie1 DT (1998)

Modulation of Bci-2 protein levels by intracellular anti-Bcl-2 single-chah

antibody increases cimg-induced cytotoxicity in the breast cancer ceIl line MCF-7.

Cancer Res. 58: 2134-2140.

5. Miyashita T and Reed JC (1993) Bcl-2 oncoprotein blocks chemotherapy-induced

apoptosis in a human leukemia ceIl line. Blood. 91: 151457.

6. Kelekar A. Chang B, Harlan J. Fesik SW. and Thompson. C B ( 1997) Bad is a

BH3 domain-containing protein that fonns an inactivating dimer with Bcl-xL.

Mol Ce11 Biol. 17: 7040-7046.

7. Zha J, Harada H, Konstantin O, Jockel J, Waksman G, and Korsmeyer S (1997)

BH3 domain of BAD is required for heterodimerization with Bcl-xL and pro-

apoptotic activity. J Biol Chern. 272: 24101-24104.

8. Wang IL, Zhang ZJ. Choksi S, Shan S, Lu X, Croce CM, Alne& ES, Kurrrgokd

R, and Huang 2. (2000) Cell permeable Bcl-2 binding peptides: a chernical

approach to apoptosis induction in tumor cells. Cancer Res. 60: W 8 - 1502.

9. Blondelle SE, and Houghten RA. (1992) Design of mode1 arnphipathic peptides

having potent antirnicrobial action. Biochemistry. 3 1: 12688-12694.

[O. Bessale R, Kapitkovsky A, Gorea A, Shalit 1, Fridkin M. (1990) All-D-magainin:

chirality. antimicrobial activity and proteolytic resistance. FEBS Let. 274: 15 1-

155.

I l . Saleh MT, Ferguson J, Boggs JM, and Gariepy J. (1996) Insertion and orientation

of a synthetic peptide representing the C-terminus of the A l domain of Shiga

toxin into phospholipid membranes. Biochemistry. 35: 9325-9334.

12. Ellerby HM, Anp W, Elierby LM. Kain R, Andrusiak R, Del Rio G, Krajewski S.

Lombardo CR, Rao R. Ruoslahti E, Bredesen DE, and Pasqualini R. (1999) Anti-

cancer activity of targeted pro-apoptotic peptide. Nat Med. 5: 1032- 1038

13. Derossi D, Calvet S, Trembleau A, Brunissen A, Chassaing G. and Prochiantz A.

(1996) Ce11 intemalization of the third helix of the Antennapedia homeodomain is

receptor-independent. J. Biol. Chem. 27 1: 18 188- 18 193.

14. Derossi D, Jolio A, Chassaing G, and Prochiantz A. (1994) The third helix of the

Antennapedia homeodomain tram locates through biologicd membranes. J. Biol.

Chem. 269: 1û4i4-10450.

15. Fenton M, Bone N, and Sinclair A (1998) The efficient and rapid irnport of a

peptide into primary B and T lymphocytes and a lymphoblastoid ce11 line. J

Immun01 Methods. 212: 41-48.

Id. EdeIhoch W (1967) Spemscupic detennhtiort of tryptophan and t m n e in

proteins. Biochernistry. 6: 1948-1954.

17. Pham N, Robinson BH, and Hedley DW (2000) Simultaneous detection of

mitochondrial respiratory chah activity and reactive oxygen in digitonin

permeabilized cells using flow cytomeuy. Cytometry. 41 : 245-25 1.

18. Sheng-Tanner X, Bump EA, Hedley DW (1998) An oxidative stress mediated

derith pathway in irradiated human leukernia cells mapped using multilaser flow

c ytometry. Radiation Research, 150: 636-647.

19. Hakern R, Hakem, A, Duncan GS, Hendenon JT, Woo M, Soenga, MS, Eli, A.

Luis, Pompa J. Kagi D, Khoo W. Potter J, Yoshida R, Kaufman SA. Lowe SW,

Penninger JM, Mak TW. (1998) Differential requirement for caspase-9 in

apoptotic pathways in vivo. Cell, 94: 339-352.

20. Goasguen JE, Dossot .JM, Fardel O, LeMee F, Le Gall, E, Leblay R, LePrise PY.

Chapenon J, and Fauchet R. (1993) Expression of the multidrug resistance-

associated P-glycoprotein (P-170) in 59 cases of de novo acute lymphoblastic

leukernia: prognostic implications. Blood. 8 1: 2394-2398.

21. Campos L, Guyotat D, Archimbaud E, Calmard-On01 P. Tsumso T. Troncy J,

Treille D, and Fiere D. (1992) Clinicai significance of multidrug resistance P-

glycoprotein expression on acute nonlymphoblastic leukemia ceils at diagnosis.

Blood. 79: 473-476.

22. Drach J, Gsur A, Hamilton G, Zhao S, Angerler J, Fiegl M, Zojer N, Raderer M,

Haber1 1, Andreeff M, and Huber H. (1996) Involvement of P-glycoprotein in the

trsuismerxtbrrme ûmsport of interleukirr-2 f LI), E l - 4 , arid interferon-gamma in

normal human T lymphocytes. Blood. 88: 1747-1754.

23. Sharom FJ, Lu P, Liu R, and Yu X. (1998) Linear and cyclic peptides as

substrates and modulators of P-glycoprotein: peptide binding and effects on drug

transport and accumulation. Biochem J. 333: 61 1-630.

24. Sharorn FJ, DiDiodrito G, Yu X, and Ashbourne KJD (1995) Interaction of the P-

gl ycoprotein multidrug transponer with peptides and ionophores. J. Biol. Chem.

270: 10334- 1 O34 1.

25. Yang E, Zha J, Jockel J, Boise LH, Thompson CB, and Korsmeyer S (1995) Bad,

a heterodirneric partner for BcI-xL and Bcl-2 displaces B a and promotes ce11

death. Cell. 80: 285-39 1.

26. Zha J, Harada H, Yang E, Jockel J, and Korsmeyer S (1996) Serine

phosphorylation of death agonist B A D in response to survival Factor results in

binding to 14-3-3 not Bcl-xL. Cell. 87: 619-628.

27. Qian T, Herman B, and Lemasters JJ (1999) The mitochondnal permeability

transition mediates both necrotic and apoptotic death of hepatocytes exposed to

&-A23 187. Toxicol Appl Phmacol. 154: 1 17-125.

28. Lemasters JJ, Nierninen AL, Qian T, Trost LC, Elmore SP, Nishimura Y, Crowe

RA, Casciao W, Bradharn CA, Brenner DA, and Herrnan B (1998) The

mitochondnai pemeability transition in ce11 death; a comrnon mechanism in

necrosis, apoptosis and autophagy Biochim Biophys Acta. 1366: 177-196.

29. Eguchi Y, Shimizu S, and Tsujimoto Y (1997) Inhacellular ATP levels determine

ce11 death fate by apoptosis or necrosis Cancer Res. 57: 1835- 1 840.

30. Saritadi B, MuHaM, Homoiya t, HoNoZ, Sep& J, Genmann UA, Gottesmen

MM, Price EM, and Boucher RC. (1994) Interaction of bioactive hydrophobic

peptides with the human multidmg transporter. FASEB J. 8:. 766-770.

3 1. Sharom FJ, Yu X, DiDonato G. and Chu JWK ( 1996) S ynthetic hydrophobic

peptides are substrates for P-glycoprotein and stimulate drug transport. Biochem

J. 320: 421-428.

32. Robinson U, Roberts WK, Ling TT', Lamming D, Sternberg SS, and Roepe PD

(1997). Human MDRl protein overexpression delays the apoptotic cascade in

Hinese hamster ovary fibroblasts. Biochemistry. 36: 116% 1 1 178.

33. Johnstone RW, Cretney E, and Smyth MJ (1999) P-glycoprotein protects

leukernic ce1 1s against caspase-dependent, but not caspase-independent ce11 death.

Blood 93, 1075-1085.

In an attempt to develop a novel therapeutic agent for the treatment of acute

leukemia. I tumed to the endogenous Bcl-2 and Bcl-xL-binding protein BAD. BAD

induces apoptosis by binding and inhibiting Bcl-2 and cl-XL'.'. In the proposed model,

BAD inhibits these survival factors and permits ce11 death signals to act u n ~ ~ ~ o s e d . ~ ~ .

Thus, BAD acts passively to induce cell death and does not have direct death-promoting

activity like Bax, Bak or caspase-3. Therefore, i t was hypothesized that BAD might

show selective toxicity for malignant cells because these cells have active death stimuli

that are blocked by the over-expression of the anti-apoptotic proteins Bcl-2 and Bcl-xL.

In contrat, normal cells may not have active death signals and may not undergo

apoptosis after temporaril y inhibiting these survival factors.

According to the widely accepted model, BAD induces apoptosis in cells over-

expressing Bcl-2 or Bcl-xL. Despite this widely held betief, the literature indicates that

BAD does not sensitize malignant cells with high levels of Bcl-2 to apoptosis after

cytokine withdnwal. Yet, the s m e cells over-expressing Bcl-xL are sensitized by BAD

over-expression1.'. If BAD is unable to induce apoptosis in cells over-expressing Bcl-2.

it would limit the appiication of molecuIes like BAD as novel therapeutic agents as Bcl-2

often propagates neoplastic growth, and Bcl-xL is more important than Bcl-2 for normal

tissue d e v e ~ o ~ r n e n t ~ ~ .

In my study, using a transient transfection model, 1 demonstrated that BAD

induced apoptosis in cells over-expressing Bcl-2 and Bcl-xL. 1 also extended the results

to leukemic ceIl lines with high levels of these survivai factors. While these effects of

BAD rnay be predicted and confirm the mainstream d o p a , they had not been previously

repmed. This worir validates t h pursoit of d L molecules that mirnie BAD as

potential novel thenpeutic agents.

A peptide corresponding to the 71 amino acid BH3 domain of BAD is an exarnple

of a small molecule that mimics its parent protein and binds and inhibits Bcl-2 and Bcl-

xL. Wang et aL6 created a cell-permeable version of this peptide by fusing it to cpm, a

decanoic fatty acid internalization sequence. This fusion peptide induced rapid ce11 death

and apoptosis. However. its cytotoxicity was significantly impaired by the over-

expression of Bcl-xL, which could limit the usefulness of this compound in treating

malignancies with increased levels of Bcl-2 and Bcl-xL. Therefore, enhancing the ability

of the BH3 domain of BAD to kil1 cells over-expressing these survival factors rnight be

thenpeutically useful. One stntegy to enhance its toxicity is to capitalize on the inherent

alpha helical shape of the peptide, since some amphipathic alpha helices are potent

naturally occumng antibiotics.

We harnessed the alpha helical propenies of the BH3 domain of BAD by fusing i t

to the Antennapedia intemalization sequence. I created a 37 amino acid fusion peptide

corresponding to the 2 1 amino acids of the human BH3 dornain of BAD fused to the C-

terminus of the 16 arnino acids of the ANT peptide (PLNTBH3BAD). The ANTBH3BA.D

fusion peptide induced rapid ce11 death and apoptosis. Its alpha helical secondary

structure disrupted mi tochondrial function and ailowed the peptide to overcome caspase

inhibitors and Bcl-2 over-expression.

Previous studies of full length BAD'*' or the BH3 domain of BAD^.^ have not

reported toxicity attributabIe to the alpha helical configuration. A number of

explmations may account for my ability to detect toxicity due to the alpha helicity of

BH3BAD. Fmt, fmmg the BH3 domain of BAD to tht ANT intmdization sequence

rnay have targeted the peptide more effeLuvely to the mitochondna where its alpha

helical shape could dismpt the inner membrane. BH3BAD peptides coupled to the cpm

fatty acid moiety were confined to the cytoplasm6. In contrast, my peptide was

distributed diffusely throughout the cell. Second, the concentration of BAD used in the

previous studies rnay have been too low to detect the effects of the alpha helix. If BH3

BAD-mutants unable to bind Bcl-2 or Bcl-xL were used at higher concentrations, the

cytotoxicity from the alpha helical shape might have become apparent. Finally, the full-

length protein BAD rnay constrain the death promoting attributes of the alpha helical

BH3 domain.

In a previous study, the BH3 domain of B A fused to ANT behaved like full-

length Bak and induced apoptosis that was inhibited by Bcl-xL over-expression and

caspase inhibitoa9. Unlike ANTBH3BAD, no toxicity attnbutable to the alpha helical

configuration was observed. The difference in action between ANTBH3BAD and

ANTBH3Bak rnay be due to the greater predicted alpha helicity of the BH3 domain of

BAD than the BH3 domain of Bak.

We dernonstrated that transient transfection of BAD cDNA induced apoptosis and

cytochrome c release without affecting AYu. In contrast, the BH3BA.û peptide induced

loss of A\YM in permeabilized cells. A similar discrepancy was reponed with the BH3-

only farnily members Bik and id". The full-length proteins induced cytochrome c

release and apoptosis without altering AYM, but peptides corresponding to their BH3

domains, induced loss of ATM and cytochrome c release in isolated mitochondna. A

number of explanations rnay account for the difference between the full-length protein

and the peptide. Fint. the &Il-tength protein may constrain the effea of the BH3

domain. Loss of AYM is related to disruption of the inner mitochondnai membrane and

the full- length protein may prevent the BH3 domain from acting at this site1'.".

Alternatively. the concentration of the full-length protein may have been too low to

produce an effect on ATM.

In rny study. the ANT intemalization sequence was used to transport the

BH3BAD peptide across ceil membranes. The ANT intemalization sequence is derived

from the 42.8kDa Antennapeida transcription factor that controls the development of the

Drosophila embryo. In 199 1, Joliot et a1 made the serendipitous discovery that

Antennapedia can transverse ce11 membranes with high eficiency and low toxicity13.

Later, a 16 amino acid region of the third alpha helix was reported to be sufficient for ce11

transduction1". The mechanism by which the ANT transverses ce11 membranes is not

entirely clear. Intemalization occun both at 4*C and 37OC indicating that the mechanism

is not energy dependent. Revening the sequence of ANT and using a peptide with D-

enantiomers did not effect intemalization, demonstrating that the mechanism is not

dependent on a specific receptor. The alpha helical stnicture of the peptide also does not

account for its ability to translocate, because dismpting the alpha helix does not hamper

intemalization. Currently, i t is believed that the positive charge on the ANT peptide

facilitates its translocation by interacting with the negative charged phospholipids in the

ce11 membrane. The peptide forrns inverted micelles with the plasma membrane and

thereby passes into the cytosol (Figure l)15.

1 peptide 1

T t T T T l T

Figure 1 : Proposed mechanism of transduction of ANT internakation sequence across ceII rnem branes. The positive charge on the ANT peptide interacts with the negatively charged phospholipids in the cell membrane. The peptide forms an inverteci micelle and thereby passes through into the cytosol

We enhanced the toxicity of the BH3BAD by hmessing Its alpha helical properties.

Alpha helixes are corkscrew-shaped secondary structures that contain 3.6 amino acid

residues per tum and nse 1.5 A per residue. Alpha helixes have phi and psi angles of

approximately -60" and -50' respectively where the phi angle refers ro the mgle between

the WCa bond and the psi angle refen to the C C' bond (Figure 2). The helical

structure is stabilized by hydrogen bonds that form between c'=O of residue n and the

NH of residue n 4 . The amino acids Ala (A), Glu (E), Leu (L) and Met (M) favour at

helical structure, whereas the nnged form of proline disrupts the a helical shape. In the

helical configuration, the amino acid side chains face outwards. When one side of the

alpha helix contains hyârophobic side chains and the other side of the helix contains

hydrophilic (polar or charged) side chains. the alpha helix is deemed amphipathic.

Some amphipathic alpha helices are toxic because they can insert thernselves into

negatively charged membranes and disrupt their f u n ~ t i o n ' ~ ' ~ . Two models have been

proposed to explain this ability. In the "barrel-stave model", the alpha helix forms pores

by inserting perpendicuiariy into the membrane with the hydrophobie side chains facing

outwards. The pores facilitate leakage across the membrane and ultimatel y membrane

disruption. In the "carpet model", large arnounts of peptide bind transvenely to the

surface of the membrane and act like a detergent to form holes and soiubilize the

membrane2'.

Figure 2: Alpha Helical structure An alpha helix is a secondary peptide structure charactenzed by phi and psi angles of - 60° and -50' respectively.

The ANTBH3BAD peptide can be compared to other cytotoxic molecules

including dipthena toxin". Shiga t o ~ i n ~ ' ~ calicheami~in~~, gelonin toxin3 and the alpha

helix (KLALAK)2 able 1). These molecules differ in the mechanism by which they

induce ce11 death. Diphtheria toxin2'. Shiga toxin2' and gelonin?s induce ce11 death by

inhibiting protein synthesis at the level of the ribosome, but differ in their specific

molecular targets. Calichamicin cleaves DNA and thereby induces cell death'". In

contrast, (KLALAK)~'~ and ANTBH3BAD induce ce11 death by disrupting mitochondnal

function. ANTBH3BAD is unique in that it is unaffected by Bcl-2 over-expression and

caspase inhibitors. ZVAD-fmk inhibits the toxicity of (KLALAIQ and Bcl-2 over-

expression abrogates the effects of dipthena toxin. The influence of caspase inhibitors

and Bcl-2 on the other cytotoxic compounds has not been reported. Of this group of

compounds, only Shiga toxin displays selective toxicity for malignant cells. Shiga toxin

requires the CD77 receptor for intemalization. CD77 is found on the surface of most

myeloma and low-grade lymphoma cells, but is not present on normal hematopoietic

7 7 precursors--. In contrast. the other cytotoxic molecules in this group are equally toxic to

n m a i cetts anci need ta be setectiveiy targeted to be therapeuticaiiy useful.

Table 1: Cornparison of cytotoxic molecules

Inhibited by P-

gl ycoprotein

Nme

Diptheria Toxin

Gelonin Toxin

Shiga Toxin

Calicheamicin

(KLALAIQ2

AIWBH3BA.D

Inhibited by BcI-2

Size

58kDa

30kDa

58kDA

1 SkDa

2.2 D a 4.8 kDa

Inhibited by

caspase inhi bitors

Mechanism

Inhibit elongation factor 2 Inhibit nbosornal subunit 60s Inhibit elongation factor 1 Cleaves DNA Disrupts rnitochondria Disrupts mitochondna

Specific for

malignant cel1s

N

N

Y

N

N

N

To extend the work on ANTBH3BAD and continue the pursuit of this rnolecule

as a novei therapeutic agent, it would have to be selectively targeted to leukemic cells.

Selective targeting may spare normal celis from the toxicity of the peptide. In addition,

an anti-tumor effect rnay be achieved with Iower concentrations of the peptide because i t

would not have to equilibrate into al1 cells.

One approach to targeting the peptide to AML cells is to conjugate it to anti-

CD33. The CD33 antigen is found on the surface of AML blasts in over 80% of cases. It

is not expressed normally on non-hematopoietic cells, but is found on the surface of early

myeloid progeniton. monocytes. and dendritic cells. Other toxins, such as calichamicin"

and gelenonin=, have been conjugated to anti-CD33. These conjugated immunotoxins

show specific anti-tumor properties without untoward toxicity in mice or humans. For

example, CMA-676, a humanized anti-CD33 monoclonal antibody linked to caiichamicin

was evaluated in a phase II study of 23 patients with AML in first relapse. In this study.

10 patients responded to the immunotoxin and 3 achieved a complete response. Toxicity

was manageable and consisted of fever, chills, nausea, and tnnsient pancytopenia. There

was no evidence of mucositis or allopecia".

Sorne potentiai probiems may exist in using anti-CD33 to target ANTBH3BAD.

First. due to the ANT intemalization sequence, the irnrnunoconjugate may still transverse

al1 cells and not be selective. This problem might be overcome by deleting the two N

terminal amino acids from the ANT sequence that are required for intemalization. This

mutant would retain its alpha helical shape, but would not be intemdized. Internalization

of the imrnunotoxins occurs through endocytosis, so an ANTBH3BAD would have to be

abie to exit the nidocytotic vesicles to be effective. It is unclew, ~ O W ~ V ~ F , whether an

ANTBH3BA.D with a mutated intemalization sequence would be able to exit from the

endocytotic vesicles into the cytoplasm.

Additional future work will involve studying the apoptosis cascade in patients

with AML and in particular the pathways of caspase activation. At its most basic level.

treatment failures in AML represent a failure of the leukemic cells to undergo apoptosis

in response to chemothenpy. Failure to activate caspases may account for some of this

resistance to apoptosis. Defects in caspase activation and caspase inhibiton have been

described in AML ceIl lines and patient samples, but. the evidence, in many cases. is

circumstantial and a systematic evaluation of the caspase cascade has not been performed

for AML. Furthemore, the functional status of the caspase cascade in patients with

AML is unknown.

This future work will determine whether functional blocks in the caspase

activation pathways exist in patients with AML and delineate the defects responsible for

these blocks. In addition, it will attempt to modulate defective pathways to overcorne

sites of blockage and restore competency to the activation cascade. The hypothesis is that

defects in caspase activation will be common in patients with AML. Since inhibition of

caspase activation renden cells resistant to chemotherapy. one cm also hypothesize that

patients with chemoresistant disease will display defects in one or more of the pathways

whereas patients with chernosensitive disease and subsequent durable remissions are

more Ii kel y to have intact activation pathways. Through these expenments. new

prognostic markea for AML may be developed. By modulating the pathways and

restomig competency to the cascade stnitegia for novel h g development €0 crovercome

drug resistance in AML may become evident.

References

1. Yang E, Zha J, Jockel J, Boise L.H. Thompson CB, Korsmeyer K. Bad, a

heterodirnenc partner for Bcl-xL and Bcl-2 displaces Bax and promotes cell death.

Cell. 1995;8O:XS-Z 1.

2. Zha J. Harada H, Yang E, Jockel J, Korsmeyer K. Senne phosphorylation of death

agonist BAD in response to survival factor results in binding to 14-3-3 not BCL-

XL. Ce11.1996;87:628

3. Ottilie S. Diaz .K. Home W. et al. Dimerization properties of human BAD. J Bi01

Chem. 1997;272:30866-30872.

4. Nakayama K. Negishi 1, Kuika K, Sawa H, Loh D. Targeted disruption of Bcl-2ap

in rnice: occurrence of gray hair, polycystic disease, and lymphocytopenia.

Proc.Natl.Acad.Sci .USA. 1994;9 1 :3700-3407.

5. Motoyarna N, Wang F, Roth KA, et ai. Massive ce11 death of immature

hematopoietic cells and neurons in bcl-x-deficient mice. Science. 1995;267: 1506-

1510.

6. Wang J, Zhang 2, Choksi S, et d. CeII permeable &I-2 binding peptides: a

chernical approach to apoptosis induction in tumor cells. Cancer

Res.2000;60: 1498- 1502.

7. Strobel T, Tai Y, Korsrneyer S, Cannistra SA. BAD partly reverses paclitaxel

resistance in human ovarian cancer cells. Oncogene. 1998; 17: 24 19-2427.

8. Zha J, Harada H, Osipov K, Jockel J, Waksman G, Korsmeyer K. BH3 domain of

BAD is required for heterodimerization with Bcl-xL and pro-apoptotic activity. J

Bi01 Chem. l997:272:24 101-24104.

9. Holinger EP, Chittenden T, Lutz RI. Bak BH3 peptides antagonize Bcl-xL function

and induce apoptosis through cytochrome c-independent activation of caspases. J

Bi01 Chem. l9W;Dt: 13298- 13304.

t O. Shimizu S, Tsujimoto Y. Proapoptotic BH3-only Bcl-2 family memben induce

cytochrome c release, but not mitochondnal membrane potential los , and do not

directly modulate voltagedependent anion channel actiuity.

Proc.Natl.Acad.Sci.USA.2000;97:577-582.

1 1. Narita M, Shimizu S, Ito T, et al. Bax interacts with the permeability transition

pore to induce permeability transition and cytochrome c release in isolated

mitochondria. Proc.Natl.Acad.Sci.USA. l998;95: 1468 1-14686.

12. Brnmer C, Cadioa ff. LaVien H, et al. Bct-2 anr( B a regtriate t he channel aetivity

of the mitochondrial adenine nucleotide translocator. Oncogene.2000; 19:329-336.

13. Joliot AH. Pemelle C, Deagostini-Bazin H, Prochiantz A. Antennapedia homeobox

peptide regulates neural rnorphogenesis. Roc.Natl.Acad.Sci.USA. 199 1 ;88: 1864-

1868.

14. Derossi D, Joliot AH, Chassaing G, Prochiantz A. The third helix of the

Antennapedia homeodomain translocates through biological membranes. J Bi01

Chern. 1994;269: 10444- 10450.

15. Derossi D, Calvet S, Trembleau A, Brunissen A, Chassaing G, Prochiantz A. Ce11

Internalization of the third helix of the Antennapedia homeodomain is receptor-

independent. J Bi01 Chem. 1996271: 18 188-18 193.

16. Blondelle S, Houghten RA. Design of mode1 arnphipathic peptides having potent

antimicrobial activities. Biochernistry. 1992;3 1 : 12688- 12694.

17. Bessalle R, Kapitovsky A, Gorea A, S halit 1, Fridkin M. All-D-magainin: chirality.

antimicrobial activity and proteolytic resistance. FEBS. 1990;274: 15 1 - 155.

18. Ellerby HM. Arap W, Ellerby LM, et ai. Anti-cancer activity of targeted pro-

apoptotic peptides. Nat Med. 1999;5: 1032- 1038.

19. Tsrr~mioto Y. Fmger tR, Ymris f, Noweil Pe, C m CM. Ctorriirg of the

chromosome breakpoint of neoplastic B cells with the ((14; 18) chromosome

translocation. Science. 1984226: lW7- 1099.

20. Shai Y. Mechanisms of the binding, insertion and destabilization of phospholipid

bilayer membranes by alpha-helical antimicrosomal and ce11 non-selective

membrane 1 ytic peptides. Biochim Biophys Acta. 1999; l462:SMO.

21. Mitamura T, Umata T, Nakano F, et al. Structure-function analysis of the diptheria

toxin receptor toxin binding site by site-directed mutagenesis. J Bi01

Chem. 1997;272:27084-27090.

22. LaCasse E, Bray MR. Pattenon B, et al. Shiga-iike toxin-1 receptor on human

breast cancer, lymphoma, and myeloma and absence from CD34+ hematopoietic

stem cells: implications for ex vivo tumor purging and autologous stem ce11

transplantation. Blood. 1999;94:290 1-29 10.

13. Obrig E, Moran T, Brown E. The mode of aclon of Shiga toxin on peptide

elongation of eukaryotic protein synthesis. Biochem J.1987;244:287-294.

24. Hinman LM, Hamann PR, Wallace R, Menendez AT, Dun FE, Upeslacis J.

Preparation and characterization of monoclonal antibody conjugates of the

caiichearnicinis: a novel and potent farnily of antitumor antibiotics. Cancer

Res. 1993;53:3336-3342.

25. Xu Y, Xu Q, Rosenbtum M, Schhberg D. Amifeukeniic acitivty of recombinant

hurnanized M195-gelonin immunotoxin in nude rnice. Leukemia. 1996; 10:321-326.