UvA-DARE (Digital Academic Repository) Mechanisms of ... · Mechanisms of Betulinic acid-induced...

UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl)

UvA-DARE (Digital Academic Repository)

Mechanisms of Betulinic acid‐induced cell death

Potze, L.

Link to publication

Citation for published version (APA):Potze, L. (2015). Mechanisms of Betulinic acid‐induced cell death.

General rightsIt is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s),other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).

Disclaimer/Complaints regulationsIf you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, statingyour reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Askthe Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam,The Netherlands. You will be contacted as soon as possible.

Download date: 03 Jan 2020

https://dare.uva.nl/personal/pure/en/publications/mechanisms-of-betulinic-acidinduced-cell-death(9134926e-7576-483f-aee5-fba0b0a44c59).html

Mechanisms of Betulinic acid-induced cell death

Lisette Potze

2015

Mechanism

s of Betulinic acid-induced cell death

Lisette Potze

Uitnodiging

Voor het bijwonen van de openbare verdediging van

het proefschrift:

Mechansisms of Betulinic acid-induced cell death

doorLisette Potze

Op dinsdag 24 novemberom 10.00 uur

in de AgnietenkapelOudezijds Voorburgwal 231

1012 EZ Amsterdam

Aansluitend bent u van harte welkom bij de receptie

in de Agnietenkapel

ParanimfenDaisy Picavet06-24920886

Danique ten Pas06-11283870

Lisette PotzeWeth. Insingerstraat 72

1107 XD [email protected]


Lisette Potze


© 2015 Lisette Potze. Thesis University of Amsterdam, The Netherlands.

All rights reserved. No part of this thesis may be reproduced, stored or transmitted in any form or by any means, without the prior permission of the author, or, when applicable, of the publishers of the scientific papers.

ISBN: 978‐94‐6299‐201‐6

Printing: Ridderprint BV, the Netherlands

Lay‐out: Lisette Potze

About the cover: The author got inspired by the painting La Gerbe (1953) by H. Matisse. On the front

cover a caricature of the painting where instead of leaves the inner membrane of the mitochondria

are schematic represented. The different colors used, represent the different mechanisms of Betulinic

acid and the chapters in this thesis. Cover design by Lisette Potze and René Scriwanek.

The research described in this thesis was performed at the Center for Experimental and Molecular

Medicine (CEMM), Laboratory of Experimental Oncology and Radiobiology (LEXOR) at the Academic

Medical Center (AMC), University of Amsterdam (UVA), The Netherlands.

The research described in this thesis was supported by a grant of Stichting Nationaal Fonds tegen

Kanker. This support is gratefully acknowledged.

Printing of this thesis was financially supported by:

The Academic Medical Center Amsterdam, ABN Amro Bank, Rob & Gerda Potze.


ACADEMISCH PROEFSCHRIFT

ter verkrijging van de graad van doctor

aan de Universiteit van Amsterdam

op gezag van de Rector Magnificus

prof. dr. D.C. van den Boom

ten overstaan van een door het College voor Promoties ingestelde commissie,

in het openbaar te verdedigen in de Agnietenkapel

op dinsdag 24 november 2015, te 10.00 uur

door

Lisette Potze

geboren te Alkmaar

Promotiecommissie

Promotor:

Prof. dr. J.P. Medema Universiteit van Amsterdam

Copromotores:

dr. J.H. Kessler Universiteit van Amsterdam

dr. F.M. Vaz Universiteit van Amsterdam

Overige leden:

Prof. dr. C.J.F. van Noorden Universiteit van Amsterdam

Prof. dr. E.F. Eldering Universiteit van Amsterdam

Prof. dr. R.J.A. Wanders Universiteit van Amsterdam

Prof. dr. A.J. Verhoeven Universiteit van Amsterdam

Prof. dr. S. de Jong Rijksuniversiteit Groningen

dr. S.W.G. Tait University of Glasgow

Faculteit der Geneeskunde

“An expert is someone who has made all the mistakes which can be made, in a very

narrow field.”

Niels Bohr

aan mijn ouders

Table of contents Chapter 1

General introduction 9

Chapter 2

Betulinic acid‐induced mitochrondria‐dependent cell death 55

is counter balanced by an autophagic salvage response

Chapter 3

Betulinic acid induces a novel cell death pathway that 77

depends on cardiolipin modification

Chapter 4

Betulinic acid induces a rapid form of cell death in colon 107

cancer stem cells

Chapter 5

Improved identification of lipids using physico‐chemical properties: 127

application to lipidomics analysis of Betulinic acid treatment

Chapter 6

General discussion 147

Annexes

Summary 165

Nederlandse samenvatting 171

Dankwoord/Words of gratitude 179

Curriculum Vitae 189

Portfolio 193

Publication list 197

Chapter 1

Introduction

"Begin at the beginning," the King said gravely, "and go on till you come to the end; then

stop."

Lewis Carroll, Alice in Wonderland

Introduction

11

1. Cancer

Cancer is a group of diseases characterized by unregulated cell growth and the invasion and

spread of cells from the site of origin, or primary site, to other sites in the body. Cancer has

defects in regulatory circuits that control normal cell proliferation and homeostasis,

resulting in uncontrolled expansion of cells. Processes that contribute to the overall net cell

number are cell proliferation and elimination of cells by programmed cell death (discussed

below). Cancer is a genetic disease, where mutation in oncogenes (genes that have

dominant gain of function due to mutation, amplification and/or overexpression) and tumor

suppressor genes (genes with recessive loss of function mutations, deletion and/or

epigenetic silencing of these genes) dictate the defects in cellular expansion.(1, 2) In 2000,

Hanahan and Weinberg proposed that cancer arises through a multistep, mutagenic process

in which oncogenes and tumor suppressor genes play an important role, whereby cancer

cells acquire essential alterations (hallmarks) in the physiology of the cell that combined

result in malignant growth.(3) These hallmarks include self‐sufficiency in growth signals,

insensitivity to anti‐growth signals, tissue invasion and metastasis, limitless replicative

potential, sustained angiogenesis and evasion of apoptosis. Conceptual progress has been

made after the description of the first hallmarks and therefore a decade after these

descriptions were proposed, Hanahan and Weinberg added two additional characteristics

and two emerging hallmarks. These include: deregulated cellular energetics, evasion of anti‐

cancer immunity, tumor promoting‐inflammation and genome instability and mutation.(4)

The hallmarks of cancer have attributed to our understanding of the disease and are a

potential target for the treatment of cancer.

In this introduction we will focus on the following hallmarks: evading apoptosis and

deregulated cellular energetics. We provide an overview of cell death mechanisms and how

these are altered in cancer, followed by an overview of cell metabolism and its alterations

occurring in cancer.

2. Programmed cell death

Programmed cell death (PCD) is an evolutionary highly conserved fundamental biological

process. PCD removes old needless cells during tissue formation thereby maintaining tissue

homeostasis. Homeostasis is needed to keep internal conditions stable and relatively

constant. To maintain homeostasis, PCD also removes damaged or abnormal cells. Several

forms of PCD exist and the last few years more accurate definitions of these cell death

pathways have been postulated by the Nomenclature Committee on Cell Death (NCCD).

These definitions are based on molecular characteristics instead of the classical

morphological traits and include extrinsic apoptosis, caspase‐dependent or caspase‐

independent intrinsic apoptosis, autophagic cell death, regulated necrosis and mitotic

catastrophe.(5) Apoptosis is a cell autonomous program of damaged or stressed cells

Chapter 1

12

resulting in ‘organized’ cell death which can be executed by two distinguishable pathways:

the extrinsic pathway and the intrinsic pathway.

2.1. Extrinsic apoptosis

The extrinsic pathway, also known as death receptor pathway, is activated by binding of a

so‐called death ligand to its receptor belonging to the tumor necrosis factor (TNF)

superfamily. Members of the TNF family share similar cysteine‐rich extracellular domains

and have a cytoplasmic domain about 80 amino acids in size called the death domain.(6) The

best characterized ligands are FasL/CD95L, TNF‐alpha and TNF‐related apoptosis inducing

ligand (TRAIL). The corresponding receptors are Fas/CD95, TNFR1‐2 and TRAILR 1‐4,

respectively.(5, 7‐9) Binding of a death receptor ligand to its receptor results in the

recruitment of adaptor protein Fas‐associated death domain (FADD) and in case of TNFR‐1

like proteins, TNFR‐associated death domain (TRADD) is first recruited followed by FADD.(10)

Besides FADD and TRADD, cellular inhibitor of apoptosis proteins (cIAPs), cellular FADD‐like

interleukin‐1β‐converting enzyme (FLICE)‐like inhibiting proteins (cFLIPs), receptor‐

interacting protein kinase 1 (RIPK1 also known as RIP1 ), E3 ubiquitin ligases and pro‐

caspase‐8 and pro‐caspase‐10 are recruited to the death domain.(5) The resulting complex,

named ‘death‐inducing signaling complex (DISC), constitutes a platform that regulates the

activation of caspase‐8 and caspase‐10.(5, 11, 12) Depending on cell type (type 1 or type 2 cells)

two different routes of further downstream pathways are known.(13, 14) In type 1 cells, active

caspase‐8 directly activates caspase‐3 and thereby trigger the executioner phase. In type 2

cells, caspase‐8 mediates the cleavage of BH3 interacting domain death agonist (BID) this

results in the generation of truncated BID, wich translocates to the mitochondria and

releases cytochrome c and results in mitochondrial outer membrane permeabilization

(MOMP).(5) These pathways are tightly regulated. At the level of DISC inhibitory proteins like

cFLIPs and cIAPs can prevent the recruitment of procaspase‐8 by competitive binding to the

death domain of FADD or by inhibition of caspase‐8 activation within the cFLIP‐containing

DISC.(10) In type 2 signaling, apoptosis can, in addition to regulation at the DISC, be regulated

by anti‐apoptotic BCL2 proteins like BCL2 or BCL‐XL to block the truncated BID‐mediated

apoptosis. Another route of inhibition is by inhibitor of apoptosis proteins (IAPs), which

prevents caspase‐9 activation.(10) These inhibitory proteins themselves are counteracted by

mitochondrial proteins such as SMAC/DIABLO. More information about mitochondrial

involvement in apoptosis and the proteins involved is described below.

2.2. Caspase‐dependent and caspase‐independent intrinsic apoptosis

Intrinsic apoptosis is triggered by intracellular stress signals such as growth factor

withdrawal, DNA damage, oxidative stress or oncogene activation.(15) Intrinsic apoptosis is

defined as a cell death process that is mediated by mitochondrial outer membrane

permeabilization (MOMP) and is consequently associated with loss of mitochondrial

Introduction

13

transmembrane potential, release of mitochondrial intermembrane space proteins (i.e

cytochrome c) into the cytosol and loss of mitochondrial respitory chain activity.(5)

Currently it is still debated how cytochrome c is released from the mitochondria, however

it is known that the release is tightly regulated by the B cell CLL/lymphoma‐2 (BCL‐2) family

proteins which consists of pro‐apoptotic proteins and pro‐survival proteins.(15, 16) The BCL‐2

family is a large set of proteins which al contain at least one conserved BCL‐2 homology (BH)

domain. The anti‐apoptotic proteins (BCL‐2, BCL‐XL, BCL‐W, Mcl1 and A1) share four BCL‐2

homology domains (BH1–BH4) among each other, with the exception of Mcl‐1 that contains

only three BH domains.(17) The anti‐apoptotic proteins are generally integrated within the

outer mitochondrial membrane (OMM) but can also found in the cytosol and endoplasmic

reticulum.(18)

The pro‐apoptotic BCL2 family members can be divided into two groups: BAX like molecules

(BAX, BAK, BOK, BCL‐G and BFK, also known as effector proteins) which contains multiple

(at least two) BH domains (BH1, BH2, BH3 and BH4).(16, 18) The second group of proteins are

the BH3‐only proteins (amongst others: BID, BIM, BAD, PUMA (p53 upregulated modulator

of apoptosis), Noxa, BIK, HRK and BMF) which share only the BH3 domain.(17)

In initiation of intrinsic apoptosis, BAK/BAX activation is crucial. Oligomerization of BAK/BAX

results in proteolipid pores within the OMM and thereby promote MOMP with subsequent

cytochrome c release. Combined deletion of BAK/BAX prevents release of cytochrome c and

results in resistance to all death stimuli of the intrinsic apoptosis pathway.(19) BH3 only

proteins are cell death initiators, which are activated by transcriptional or posttranslational

mechanisms in a tissue‐restricted and signal specific manner.(20‐22) They act upstream of

BAK/BAX and can be subdivided based on their ability to interact with the anti‐apoptotic

proteins or both anti‐apoptotic as well as BAK like molecules (the effectors). BH3‐only

proteins that only bind and inhibit the anti‐apoptotic proteins are called “sensitizers” or

“de‐repressors” (BAD, BIK, Noxa and HRK), while BIM, BID and PUMA, known as

“activators”, can bind the anti‐apoptotic proteins as well as directly bind to BAK/BAX and

inducing oligomerization and MOMP.(18, 23) Each sensitizer/de‐repressor protein has a

unique binding profile for the anti‐apoptotic proteins. Sensitization lowers the threshold for

the activation of BAK/BAX and MOMP, but does not cause apoptosis itself. In de‐repression,

a direct activator is bound by an anti‐apoptotic protein, and a subsequent BH3‐only protein

releases the activator to promote MOMP.(18)

Upon MOMP, the released cytochrome c binds apoptotic protease activating factor 1 (APAF‐

1), inducing its conformational change and oligomerization leading to the formation of a

caspase activation platform called the apoptosome. The apoptosome recruits, dimerizes

and activates caspase 9 (an initiator caspase), which can cleave and activate caspase 3 and

caspase 7 and results in apoptosis.(24) MOMP results also in the release of mitochondrial

release of second mitochondria‐derived activator of caspase (SMAC, also known as DIABLO)

Chapter 1

14

and OMI (also known as HTRA2) which block X‐linked inhibitor of apoptosis protein (XIAP)‐

mediated inhibition of caspase activity.(24‐27)

Tumor suppressor protein p53

Upon cell stress and DNA damage, a tumor suppressor gene called p53 is activated, and has

several downstream effects including cell cycle arrest, DNA repair, inhibition of

angiogenesis and apoptosis. With these effects the cells get a chance to repair the damage.

p53 induces the expression of genes that encode for death receptors and pro‐apoptotic

proteins (i.e. Fas receptor, BAK and BAX)(28‐31) while on the other hand it represses the

expression of anti‐apoptotic proteins (i.e. BCL‐2, BCL‐XL and IAPs) The pro‐apoptotic protein

PUMA is a direct target of p53.(32) It has been proposed that p53 can also induce apoptosis

independently of its transcriptional activity by activating BAX in the cytoplasm and

subsequent cytochrome c release.(33) Taken together one can conclude that p53 is a key

regulator in apoptosis signaling.

Figure 1: Apoptotic signaling. Induction of apoptosis by the extrinsic pathway depends on the binding of

ligand to its receptor. A death‐inducing signaling complex is formed and active caspase 8 generated, which

can be inhibited by high levels of c‐FLIP. Crosstalk between extrinsic and intrinsic pathway can occur through

BID cleavage, which leads to activation of the pro‐apoptotic BCL2 family proteins BAX and BAK, resulting in

MOMP. The intrinsic apoptosis pathway depends on MOMP and apoptosome formation which results in

caspase 3 activation and apoptosis. Intrinsic apoptosis is tightly regulated by the balance between the anti‐

apoptotic and pro‐apoptotic members of the BCL‐2 family. Abbreviations: APAF1, apoptotic protease‐

activating factor 1; DISC, death‐inducing signaling complex; FADD, Fas‐associated protein with death

domain; c‐FLIP, cellular FLICE‐like inhibitory protein; FasL, Fas ligand; MOMP, mitochondrial outer

membrane permeabilization; SMAC, second mitochondria‐derived activator of caspase; XIAP, X‐linked

inhibitor of apoptosis protein.

Extrinsic pathway

Intrinsic pathway

Procaspase 8 Caspase 8

c-FLIP

BID

tBID BAX BAK

BCL2 BCL-XL

Procaspase 3

Caspase 3

Apoptosis

MOMP

Procaspase 9 APAF-1

Cytochrome cApoptosome

Caspase 9

SMACXIAP

FasL

Fas

FADD

Introduction

15

2.3. Autophagy

Autophagy is a highly conserved cellular proteolytic (lysosomal) degradation process in

which cytosolic components are first encapsulated by a membrane and subsequently

degraded in the lysosome, thereby providing new building blocks for the cell and is typically

induced upon nutrient starvation. In cells without stress, basal levels of autophagy take

place to perform homeostatic functions such as protein and organelle turnover.

There are three forms of autophagy which are mediated by distinct mechanisms. In

chaperone‐mediated autophagy, targeted proteins are translocated across the lysosomal

membrane in a complex with chaperone proteins (i.e. Hsc‐70) that are recognized by the

lysosomal membrane receptor lysosomal‐associated membrane protein 2A (LAMP‐2A),

resulting in their unfolding and degradation.(34, 35) In contrast, micro‐ and macroautophagy

involves the sequestration of cargo (both selective and non‐selective mechanisms) via a

sequestering membrane. In microautophagy the sequestration of cytosolic components

occurs directly by the lysosomes, while in macroautophagy the cargo is sequestered within

a unique double membrane cytosolic vesicle (an autophagosome) which later on fuses with

a lysosome for degradation of the cargo.(36, 37) Selective degradation involves targeting

specific cargoes (i.e. organelles (i.e. damaged mitochondria) or invasive microbes).

Macro autophagy starts with an isolated membrane (phagophore), which is suggested to be

derived from lipid bilayer contributed by endoplasmic reticulum, trans‐Golgi system and

endosomes.(38) The phagophore expands so it can engulf cytosolic cargo (i.e organelles,

ribosomes, protein aggregates) to form a double‐membraned autophagosome.(39) The

autophagosome fuses with a lysosome, promoting degradation of the cargo by lysosomal

proteases. The resulting amino acids and other by‐products of degradation are released to

the cytosol via lysosomal permeases and lysosomal transporters where they can be used as

building blocks.(39) Canonical autophagy involves five steps where autophagy related

proteins play an important role: (1) initiation (which depends on ULK1 complex (ULK1,

ATG13 and ATG17) and receives stress signals from mTOR complex 1), (2) nucleation (which

depends on the BECLIN1–PtdIns3KC3–ATG14L complex), (3) elongation and closure of

autophagosome (which depend on ATG12–ATG5 (including ATG7) and LC3–PE conjugation

systems), (4) recycling (which depends on ATG9) and (5) degradation by fusion to the

lysosome.(40) The formation of functional autophagosomes can bypass some of these steps,

although these alternative mechanisms are currently under debate and are named non‐

canonical autophagy. There are some forms of non‐canonical autophagy pathways that

have been identified under certain cellular circumstances: one that bypasses the proteins

involved in elongation and closure (ATG7, ATG5 and LC3) and others that bypass proteins

that are important for initiation (ULK1) and nucleation (BECLIN1). (reviewed in (40) )

The capacity for large scale degradation carries a certain risk for cells, as unregulated

degradation of cytoplasm or organelles is likely to be lethal.(37) Therefore induction of

Chapter 1

16

autophagy is tightly regulated by several proteins. Two evolutionarily conserved nutrient

sensors play a role in autophagy regulation: (1) the mammalian target of rapamycin (mTOR)

kinase which under normal conditions inhibits autophagy). When mTORC1 kinase activity is

inhibited autophagosome formation occurs.(41, 42) Second the eukaryotic initiation factor 2α

(eIF2α) kinase Gcn2 and its downstream target Gcn4 (a transcriptional transactivator of

autophagy genes) turn on autophagy during nutrient depletion.(43) Also p53 plays a role in

regulating autophagy. As a transcription factor in the nucleus it can activate proteins (i.e

DRAM) which activates autophagy.(44)

Figure 2: An overview of several forms of autophagy. Macroautophagy, microautophagy and chaperone

mediated autophagy are depicted. See text for detailed information.

2.4. Autophagic cell death

The definition of autophagic cell death (according to NCCD) is cell death that is mediated by

autophagy and as such can be suppressed by the inhibition of the autophagic pathway by

genetic means (gene knockout or RNA interference for i.e. ATG5, ATG12)) and/or chemicals

(i.e., agents that target VPS34).(5) Certain dying cells display the morphological hallmarks of

autophagy, however the question whether autophagy has a causative role in cell death has

been debated. It has been suggested that under specific circumstances that depend on the

nature of the stimulus, the duration and its amplitude, extensive autophagy may cause cell

death.(45) The discovery of chemical compounds (i.e 3‐ methyladenine (3‐MA) and

wortmannin) that can inhibit autophagy resulted in several reports about autophagic cell

death. These reports describe the existence of a caspase‐independent cell death which

Cytosolic chaperone complex

Cytosolic protein

CMA

LAMP-2A

Microautophagy

Macroautophagy

Phagophore

Autophagosome

Lysosome

Introduction

17

proceeded with an accumulation of autophagosomes and increased lysosomal activity, and

using the inhibitors of autophagy a reduction in cell death was observed. For example, a

study using oncogenic Ras‐induced death of glioma or gastric cells showed cell death in the

absence of caspase activation, which was not inhibited by the overexpression of anti‐

apoptotic BCL‐2 protein.(46) Another study showed reduced cell death using 3‐MA for anti‐

estrogen induced cell death in MCF‐7 cells.(47) However, in these studies, autophagy

occurred in cells thought to die by apoptosis, and it was presumed that autophagy triggered

apoptosis, instead of playing a causative role. Moreover, 3‐MA is not specific for class III

PI3K kinases and can inhibit other kinases as well as inhibiting the permeability transition in

mitochondria.(48) Thus, it is not possible to directly implicate autophagy in death execution

from these 3‐MA inhibitor studies. A different approach was taken with RNA interference

(RNAi) of two ATG genes, ATG7 and BECLIN1, this resulted in cell death in mouse L929 cells

and in macrophages when caspase inhibitor zVAD was used.(49, 50) RNAi against ATG5 and

BECLIN1 resulted in an inhibition in cell death in BAK/BAX knock out murine embryonic

fibroblasts (MEFs) treated with etopside or staurosporin.(51) Notably, in these studies,

apoptosis was blocked and the ATG gene RNAi blocked the cell death. These findings

exclude that autophagy is triggering apoptosis, nonetheless they raise the question if

autophagy could be a cell death mechanism is cells whose apoptotic machinery is intact.(52)

In etopside‐treated wild type MEFs (which die via apoptosis) only small amounts of

autophagy were detected and inhibition of autophagy by 3‐MA did not reduce the etopside‐

induced cell death. This indicates that autophagy cannot perform cell death induction

unless apoptosis is blocked.(51) It could be that cells prefer to die via apoptosis but will die

by an alternative route if the stimuli is harsh enough.(53) A different possibility is that

apoptotic cell death is faster than autophagic cell death and therefore the last one is

observed only in apoptotic deficient cells.(52) So up to now it is still debatable if autophagic

cell death occurs in cells which have intact apoptosis.

2.5. Regulated necrosis

Necrosis is a form of passive cell death that is induced upon strong insults such as

mechanical injury of cells. Typical features include swelling, rupture of organelle

membranes as well as the outer cell membrane and as a result the cell contents are

released, often causing inflammation in vivo.(54) In the past, necrosis has been considered

as an accidental cell death mechanism defined by the absence of morphological

characteristics of apoptosis and autophagy.(55) However, several laboratories have now

shown that necrosis can occur in a regulated manner.(56) Under selected circumstances

regulated necrosis can be induced by several triggers like excitotoxins, ligation of death

receptors and alkylating DNA damage.(57‐60) Regulated necrosis was first described that it

include caspase inhibition (especially caspase‐8) by pharmacological agents (i.e., chemical

caspase inhibitors like Z‐VAD‐fmk) or genetic manipulations (i.e., RNA interference or gene

knockout), followed by a block in degradation of receptor interacting protein 1 (RIP1) and

Chapter 1

18

RIP3. The latter two proteins in turn form a multiprotein complex called the nescrosome,

which further activates players for execution of necrotic cell death.(56) However later

research showed that regulated necrosis does not occur only in caspase‐incompetent cells

upon the activation of the receptor interacting protein 1 (RIPK1) homolog RIPK3. There are

multiple molecular circuits that can drive regulated necrosis including (but not limited to)

necroptosis (RIP1 dependent, inhibitable by necrostatin 1 form of TNFR1 induced regulated

necrosis), mitochondrial permeability transition (MPT)‐dependent regulated necrosis (relies

on cyclophylin D, reviewed in (61)), and parthanatos (a non‐apoptotic cell death subroutine

that critically relies on the (hyper)activation of poly(ADP‐ribose) (PAR) polymerase 1

(PARP1).(62, 63) Over the past years it has become clear that regulated necrosis plays a role in

physiological scenarios (e.g., embryonic development) and pathological settings (e.g.,

ischaemic injury, neurodegeneration and viral infection). This led to the suggestion that

patients might benefit from pharmacological modulation of regulated necrosis.(11, 56, 64)

2.6. Lysosomal cell death

Lysosomes are the main compartment for intracellular degradation and subsequent

recycling of cellular constituents. The degradation is carried out by a number of acid

hydrolases (i.e phosphatases, nucleases, glycosidases, proteases, peptidases, sulfatases and

lipases) capable of digesting all major cellular macromolecules.(65, 66) The best‐studied

lysosomal hydrolases are the cathepsin proteases. Based on their active site amino acid, i.e.

cysteine (B, C, H), aspartate (D and E) and serine (G) cathepsins they are divided into three

subgroups.(67) Lysosome membrane permeabilization and the consequent leaking of

hydrolases (mainly cathepsin proteases) into the cytosol, can initiate the intrinsic apoptosis

pathway (68) but also can trigger caspase‐independent non‐apoptotic cell death pathway,

indicating that the lysosomal hydrolases can act as initiators as well as effectors of PCD.(66, 69, 70) This form of cell death where lysosomal hydrolases are the key activation step is called

lysosomal cell death.(66, 71) Although the induction of lysosomal cell death is clearly distinct

from necrosis and apoptosis, the execution phase is similar.

The function of lysosomal cathepsin proteases is not limited to intralysosomal protein

turnover and degradation of extracellular matrix when secreted. The last years many

specific functions of the cathepsin proteases have been discovered. These specific functions

include roles in cell death, bone remodeling, antigen presentation, epidermal homeostasis,

angiogenesis and cancer cell invasion.(66, 72‐75)

In cancer, cysteine cathepsins (i.e. B, C, H) are upregulated by a variety of mechanisms like

gene amplification, transcription regulation, post‐transcriptional modifications and

epigenetic regulation.(76) Trafficking and subcellular localization of these cysteine cathepsins

changes during neoplastic progression, resulting in active and inactive forms which have

pro‐oncogenic effects. The enhanced secretion of these cathepsins initiate tumor growth,

migration, invasion, angiogenesis and metastasis that increase neoplastic progression. This

role of cathepsins in cancer has been well studied in vitro and in vivo. (reviewed in (76)) At

Introduction

19

the same time the upregulation of these cathepsins may sensitize cells towards the

lysosomal cell death pathway. Especially cell death could occur in tumor cells with several

defects in the classical apoptosis pathways.(77)

Cancer cells have developed several strategies to protect them from lysosomal membrane

permeabilization and the acid hydrolases leaking into the cytosol. One strategy is

upregulation of lysosomal protease inhibitors, the best characterized inhibitors are the

squamous cell carcinoma antigens (SCCA) 1 and 2. SCCA 1 and 2 are tumor‐associated

proteins which can inhibit serine proteases and whose levels are used in diagnosis of

squamous cell carcinoma.(78) Cysteine cathepsins can be inhibited by cystatin A and B and

serine protease inhibitor 2A.(66, 79, 80) In cystatin B‐deficient mice an increase in apoptosis

was observed, demonstrating the importance of these inhibitors in preventing cell death.(81)

A second strategy is upregulation of heat shock protein 70 (Hsp70), heat shock proteins are

molecular chaperones that interact with several proteins to assist in their folding, stability

and function. Hsp70 stabilizes lysosomes by binding to phospholipid bis‐(monoacylglycero)‐

phosphate (BMP), which is highly present in intraluminal membranes of the lysosomes.(70,

82, 83) BMP is a well described docking lipid for enzymes and cofactors involved in lysosomal

degradation of sphingolipids.(82) One of these enzymes is acid sphingomyelinase (ASM),

which activity depends on its recruitment to the intralysosmal membranes by BMP (84), and

plays a crucial role in the Hsp70‐mediated stabilization of lysosomes.(83) Hsp70 mediates an

increase in ASM activity, which leads to higher lysosomal ceramide content and increased

lysosomal stability.(70) How Hsp70 affects the stability of the lysosome is still unclear but the

hypothesis is that the change in lipid composition of the intralysosomal membranes has an

influence on the stability of the entire lysosome. It could be that an ASM dependent

increase of ceramide alters the properties of membranes in such a way that it has a positive

effect on the stability of the outer membrane.(70)

In light of the dependence on these survival strategies further elucidation of these

mechanisms controlling lysosomal membrane stability could lead to potential cancer drug

targets.

3. Cell death in cancer

3.1. Apoptosis and cell death

The transforming effects of proto‐oncogenes (i.e. Myc) that mediate unrestrained cell

proliferation are countered by “intrinsic tumor suppressor mechanisms” that most often

trigger apoptosis.(85) Defects in the apoptotic machinery play an important role in

oncogenesis, allowing neoplastic cells to survive over intended lifespans, subverting the

need for exogenous survival factors and providing protection from oxidative stress and

hypoxia as the tumor mass expands.(86) Apoptosis defects are considered an important

complement of pro‐oncogene activation, as many deregulated oncoproteins (i.e. Myc,

Cyclin‐D1) that drive cell division also trigger apoptosis.(87) For example, Myc constantly

Chapter 1

20

induces cell proliferation and apoptosis. To facilitate myc‐driven tumorigenesis, inhibition

of apoptosis (e.g., loss of function of p53 or overexpression of BCL2) is required.(85)

Mutations that affect the extrinsic pathway of apoptosis are for example mutations in death

receptor genes. Epigenetic silencing of death receptors occur like epigenetic silencing of

TRAIL‐R1 results in apoptosis resistance in gliomas(88) and ovarian cancer.(89) Furthermore,

cFLIP is overexpressed in many tumors and thereby block caspase 8.(90, 91) Beside blocking

of caspase‐8, loss of caspase‐8 expression (i.e. by epigenetic silencing or loss of

heterozygosity (LOH)) has been observed.(92) Downregulation/loss of Fas expression has

been also found in a variety of tumors including melanoma, lung adenocarcinomas and

esophageal cancer.(93‐95)

A well‐known alteration in the resistance to intrinsic apoptosis are mutations in the tumor

suppressor gene p53. This results in loss of induction of BAX and PUMA and to some extent

NOXA and as a consequence a block in apoptosis.(96)

Next to loss of p53‐dependent apoptosis, overexpression of anti‐apoptotic BCL‐2 or BCL‐XL

probably occurs in more than half of all cancers(97), rendering tumor cells resistant to

apoptotic stimuli, including most cytotoxic anticancer drugs. Cancer cells can acquire

resistance to apoptosis by down regulation or mutation of pro‐apoptotic proteins such as

BAK and BAX. For example, haploid loss of BAX results in accelerated tumor progression in

breast cancer mice models (98) and in myc‐mediated lymphogenesis.(99) In tumors with

microsatellite instability, several mutation in BAX are observed leading to decreased

apoptosis and thus increased tumor progression.(100, 101) In general, cancer cells have

acquired many different mutations that result in a diminished or complete block in

apoptosis, thereby promoting tumor progression.

3.2. Autophagy and cancer

The regulation of autophagy overlaps closely with signaling pathways that regulate

tumorigenesis. Tumor suppressor genes like p53, PTEN, TSC1 and TSC2 involved in the

upstream inhibition of mTOR signaling, can stimulate autophagy. In contrast oncogene

products like class I PI3K, AKT and BCL‐2 can activate mTOR signaling and inhibit

autophagy.(102, 103) In cancer, anti‐apoptotic proteins BCL‐2 and BCL‐XL are upregulated in

many tumors and they inhibit autophagy by binding to the BECLIN‐1 autophagy protein.(104,

105) In 40‐75% of the cases of human ovarian, prostate and breast cancer one allele of

BECLIN‐1 is deleted. This is important in oncogenesis because mice with heterozygous

disruption of BECLIN‐1 have a diminished autophagy and were more prone to develop

tumors.(106, 107) These data suggest that autophagy acts as a tumor suppressive mechanisms,

however the mechanism by which autophagy functions in tumor suppression is still unclear.

Autophagy also plays a role genomic stability, mono‐allelic or bi‐allelic loss of BECLIN‐1 or

ATG5 results in increased DNA damage, gene amplification and aneuploidy in parallel with

increased oncogenesis.(108) For the mechanism behind this more research is needed.

Introduction

21

3.3. Necrosis and cancer

In (regulated) necrosis cells release many pro‐inflammatory signals to the surrounding,

resulting in recruitment of immune inflammatory cells(109), which have been shown to

actively promote tumorigenesis by fostering angiogenesis and cancer cell proliferation and

invasiveness.(4) On top of that, necrotic cells release bioactive regulatory factors (i.e IL‐1α)

which can directly stimulate the microenvironment to proliferate facilitating neoplastic

progression.(110)

Thus in general, cancer cells can make evasive adaptations in all different cell death

mechanisms or use them to their benefit. These alterations render the cells to resistance to

conventional therapies like chemotherapy and radiotherapy.

4. Metabolism

Activation of oncogenes and the loss of tumor suppressor genes promote metabolic

reprogramming in cancer. This hallmark of cancer cells is mirrored in enhanced nutrient

uptake to supply energy and building blocks for biosynthetic pathways. Moreover, cancer

cells are typified by their ability to exhibit metabolic flexibility to sustain their growth and

survival when nutrients are limited.(111) Over the last decade, researchers started to unravel

the precise mechanisms behind the multitude of metabolic changes observed in cancer

cells. This has been and still is a formidable task as cellular metabolism in eukaryotes is a

complex system with more than 5000 metabolic reactions catalyzed by more than 10.000

enzymes.(112) and www.genome.jp/kegg) Metabolism pathways are divided into two main

categories, which are energy metabolism pathways (such as glycolysis, Krebs cycle, fatty

acid oxidation and oxidative phosphorylation) and biosynthesis pathways (such as lipid and

fatty acid synthesis and biosynthesis of nucleotides via the pentose phosphate pathway).

4.1. Energy metabolism

Adenosine triphosphate (ATP) is an energy carrier in all mammalian cells. Cells produce ATP

via multiple pathways. Exogenous glucose is via glycolysis converted to pyruvate, resulting

in a net gain of two ATP molecules. In the absence of oxygen, pyruvate is further

metabolized to lactate without a further gain in ATP molecules. In the presence of oxygen,

pyruvate is converted into acetyl CoA and this product can enter the citric acid cycle (also

known as Krebs cycle), which is a continuous cycle where acetyl CoA and oxaloacetate are

processed by 8 different enzymes. The first conversion is to citrate followed by processing

to cis‐Aconitate, isocitrate, which is further oxidized to oxalosuccinate and converted to α‐

ketoglutarate. α‐ketoglutarate is processed to succinyl‐CoA, then to succinate. Succinate

dehydrogenase converts succinate into fumarate, which is then processed into l‐malate.

The final step involves the oxidation of l‐malate into the starting product oxaloacetate,

which can then with a new molecule acetyl CoA restart the cycle. The crucial product of the

Chapter 1

22

Krebs cycle is NADH and FADH2 which are used to generate ATP via oxidative

phosphorylation. Via glycolysis and the Krebs cycle the breakdown of one glucose molecule

results in a total of 36 ATP molecules.

Next to glucose other biomolecules can enter the Krebs cycle and tumor cells utilize another

source for energy production via the so called glutaminolysis pathway. This pathway shunts

glutamine via glutamate into the Krebs cycle at the level of α‐ketoglutarate and can

therefore also generate NADH and FADH2.

4.2. Biosynthesis

Dividing cells need nucleotides, lipids and amino acids to create new DNA, membranes and

proteins. The pentose phosphate pathway (PPP) produces ribose‐5‐phosphate, which can

be used to create nucleic acids and nucleotides. Fatty acids from exogenous sources can be

taken up by cells via fatty acids transporters(113) or are synthesized de novo, which under

physiological conditions occurs mainly in the liver and to a lesser extent in adipose tissue

and in lactating breast.(114) Acetyl‐CoA gets carboxylated by acetyl‐CoA carboxylase (ACC) to

form malonyl‐CoA. The malonyl‐CoA is further converted by fatty acid synthase (FASN) to

saturated long chain fatty acids.(115) The maximum length of de novo synthesized fatty acids

via this pathway contains 16 carbon atoms (palmitate C16:0). These fatty acids can undergo

modifications to be elongated into longer fatty acids by fatty acid elongases or can be

desaturated by stearoyl CoA desaturase 1 (SCD‐1) to create unsaturated fatty acids.

Unsaturated fatty acids are essential components of cell membranes and desaturation of

fatty acids occurs to maintain a healthy pool of fatty acids. These processes are regulated

by sterol regulatory element binding proteins (SREBPs), a family of transcription factors that

regulate lipid homeostasis by controlling the expression of a range of enzymes (ATP‐citrate

lyase (ACL), ACC and FASN) required for endogenous cholesterol, fatty acid (FA),

triacylglycerol and phospholipid synthesis.(116) SREBPs are regulated by SREBP cleavage

activating protein (Scap) and ER‐resident protein Insig. In a sterol rich environment, Scap

binds to cholesterol in the ER membrane and to Insig. This retains the SREBP‐Scap complex

in the ER. In sterol low conditions, Scap no longer binds Insig and the SREBP‐Scap complex

is sorted into COPII‐coated vesicles to the Golgi, where SREBP is proteolytic cleaved by site1

(S1P) and site 2 (S2P) proteases which results in the release of the N‐terminal transcription

factor.(117)

Fatty acids can also be a source of energy, when fatty acids are degraded via β‐oxidation to

generate acetyl‐CoA. When acetyl‐CoA enters the Krebs cycle energy is produced as

described above.

Introduction

23

Figure 3: Overview of cell metabolism. The main metabolic pathways that contribute to the production of

macromolecules in mammalian cells are nucleotide synthesis, the pentose phosphate pathway,

glutaminolysis, cholesterol synthesis, fatty‐acid synthesis and desaturation and elongation. The enzymes

involved in these pathways are shown in bold. See text for more detail. Abbreviations: ACC, acetyl‐CoA

carboxylase; ACL, ATP citrate lyase; ACO, aconitase CoA; coenzyme A; CS, citrate synthase; ELOVL6, fatty

acid elongase 6; FASN, fatty‐acid synthase; F1,6BP, fructose‐1,6‐ bisphosphate; F6P, fructose‐6‐phosphate;

FH, fumarate dysratase; GLS, glutaminase; G6P, glucose‐6‐phosphate; G6PD, G6P dehydrogenase; HK,

hexokinase; HMGCR, 3‐hydroxy‐3‐methylglutaryl‐CoA reductase; IDH, isocitrate dehygronease; MDH,

malate dehydrogenase; PDHK1, pyruvate dehydrogenase kinase; PEP, phosphoenolpyruvate; PFK,

phosphofructokinase; PKM2, pyruvate kinase M2; PL, phospholipids; SCD‐1, stearoyl‐CoA desaturase 1;

SDH, succinate dehygrogenase TAG, triacylglycerides.

4.3. Energy metabolism and cancer

Over time, more evidence appeared showing the importance of a shift in cellular

metabolism in cancer cells. The first discovery of a necessary metabolic alteration has been

made > 80 years ago by Otto Warburg. He observed that glycolysis is upregulated in tumor

cells even if enough oxygen in the surrounding of the cells is available to create ATP. This

phenomenon was coined the Warburg effect.(118, 119) This very fast but inefficient way to

produce energy might constitutes an advantage for tumor growth. Firstly, cells that rely on

oxidative phosphorylation need a constant flux of oxygen. In tumors hypoxic circumstances

with fluctuating oxygen levels are omnipresent. By upregulating glycolysis tumor cells are

not harmed by these oxygen fluctuations.(120‐122) Secondly, the Warburg effect leads to the

Glucose

G6P

F6P

F1,6BP

Glyceraldehyde-3-P

PEP

3-Phosphoglycerate

Pyruvate

Ribulose-5-P

Pentose phosphate pathway

HK

Glycerol-3-P Phosphatidic acid

Acetyl-CoA

Citrate

Isocitrate

Succinate

Fumarate

Oxaloacetate

Malate Krebscycle

α-KG

Citrate Acetyl-CoAACL

Malonyl-CoAACC

HMGCRMevalonate

Cholesterol

Cholesterol synthesis

FASN

Palmitate

Stearate

ELOVL6

SCD-1

G6PD

PKM2

Glutamate

Glutamine

GlutaminolysisGLS

Fatty-acid synthesis

Elongation

Desaturation

PFK1

SDH

FH

MDHCS

ACO

IDH2/3

TAG PL

Palmitoleic acid Oleic acid

SCD-1

Desaturation

Chapter 1

24

production of a high level of lactate acids. These acids are suggested to condition the

microenvironment of the tumor, favor tumor invasion and suppress anti‐cancer immune

effectors.(120, 123‐125) Third, cancer cells use intermediates of the glycolytic pathway i.e. ribose

sugars for nucleotides, glycerol and citrate for lipids, nonessential amino acids and via the

pentose phosphate pathway NADPH.(120, 126, 127)

In glycolysis, hexokinases catalyzes the essentially irreversible first step in this pathway

where glucose is rapidly phosphorylated to glucose‐6‐phosphate. Hexokinases are strongly

upregulated in cancer.(128) Hexokinase (HK)‐II is located in the mitochondria and bound to

the outer membrane via the voltage‐dependent anion channel (VDAC). HK‐II binds ATP

(brought to HK‐II by VDAC) and the glucose to produce glucose‐6‐phosphate. Tumor cells

have multiple genetic, epigenetic, transcriptional and post‐translational strategies to

enhance expression and function of hexokinase (HK) II. (reviewed in (128)) Besides playing a

role in glycolysis HK‐II via its mitochondrial location also suppresses death of cancer cells.

Binding of HK‐II to VDAC diminish the availability of free VDAC sites that can interact with

activated pro‐apoptotic molecules.(129, 130)

Glucose provides the cell with acetyl CoA, which can be converted to citrate which is

shuttled out of the mitochondria for de novo fatty acid synthesis, however continuous

export of citrate introduces a deficiency to the Krebs cycle which must be replaced by an

anaplerotic flux to maintain fatty acid synthesis to take place.(131) In glutaminolysis,

glutamine is shuttled into the Krebs cycle where it is converted by glutaminases to

glutamate. This route is typically bidirectional (glutamine synthetase to create glutamate)

but in cancer glutaminases are overexpressed and/or glutamine synthetase are suppressed,

favoring the reaction to create glutamate.(132‐134) Glutamate is converted into α‐

ketoglutarate and thereby replenishing the intermediates in the Krebs cycle. The citrate

produced can be cleaved to generate oxaloacetate which is converted into malate and

ultimately lactate. Glutamine can also be converted into lactate when mitochondrial malate

is exported to the cytoplasm. In tumor cells this pathway is a major source of NADPH.(131)

Under normal conditions the transcription factor hypoxia‐inducible factor 1α HIF‐1α (active

subunit of HIF‐1) is posttranslational modified by prolyl hydroxylation, which promotes

association with the von Hippel‐Lindau (VHL) tumor suppressor resulting in degradation of

HIF‐1α by ubiquitination.(126) During hypoxia a process involving reactive oxygen species

(ROS) generated in the mitochondria, inhibits the prolyl hydroxylation resulting in

stabilization of HIF‐1α(135) which stimulates cells to consume glucose and produce

lactate.(136) Stabilization of HIF‐1α during normoxia occurs in tumors by mutations in VHL,

or mutations in succinate dehygrogenase and fumarate hydratase (Krebs cycle intermediate

enzymes) Active HIF‐1α can enter the nucleus and activates transcription of glucose

transporters, glycolytic enzymes (i.e. pyruvate dehygrogenase kinase 1 (PDK1)) and lactate

dehygrogenase A (LDH‐A).(126)

Introduction

25

4.4. Biosynthesis and cancer

In order to keep up with the rapid cell proliferation, tumor cells increase the rate of

metabolic reactions to generate amino acids, nucleotides and lipids that are needed to

create new biomass.(122) In tumor cells the PPP pathway activity is increased. The higher

activity of this pathway results not only in increased nucleotide biosynthesis but also

enhances the antioxidant capacity of the cell, due to generation of NADPH, and thus

protects cells against chemotherapeutics and a harmful microenvironment.(120, 122, 127)

Cancer requires an enormous supply of lipids for membrane biogenesis and protein

modifications. To meet this requirement cancer cells undergo major changes in their lipid

metabolism and shift towards de novo lipid synthesis.(112, 114, 137, 138) Increased fatty acid

synthesis is found in 20% to 90% of tumors and is reflected in the up‐regulation of key

enzymes (FASN, ACC and ACL) involved in this pathway.(139‐143) In cancer multiple oncogenic

mutations leads to a rapid fatty acid synthesis i.e. mutations in the phosphatidylinositol 3

kinase (PI3K)/Akt/mTOR pathway. Activating mutations of PI3K or elimination of negative

regulator PTEN results in the overexpression of SREBs.(114, 139, 144) Overexpression of mTOR

results in an increase in surface expression of glucose transporters.(145, 146) The breakdown

of lipids by β‐oxidation is suppressed by PI3K/AKT.(147) In hematopeoic cells PI3K/AKT

suppresses carnitine palmitoyltransferase‐1A (CPT1A). CPT1A is the rate‐limiting enzyme for

β‐oxidation.(148) To fulfill the need of unsaturated fatty acids SCD‐1 is overexpressed in

several tumors.(149) These enzymes and genes mentioned above, have been proposed as

novel targets for tumor treatment.(138, 150)

Despite the growing evidence demonstrating deregulated fatty acid metabolism as a

feature of cancer, the exact role of these metabolic alterations in the development and

maintenance of the disease is not fully understood. Besides being needed for membrane

biosynthesis, energy rich lipids could serve as a source of energy when nutrients are low. It

has also been suggested that lipid biogenesis could play a more active role in cell

transformation and cancer development. (reviewed in (151)) Lipids acts as signaling

molecules in cancer, i.e. phosphatidylinositol (3,4,5)‐trisphosphate (PIP3).PIP3 is a produced

by PI3K pathway in response to growth factor signaling and mediates the recruitment and

activation of Akt.(151) Lipids also play a role in the tumor microenviroment. Activation of

SREBP and induction of enzymes of the mevalonate pathway are involved in the disruption

of normal tissue architecture. Lipids are also involved in the interaction of cancer cells with

components of the tumor stroma. For example, cancer‐associated fibroblasts (CAFs) show

increased expression of FASN.(152) A study in ovarian cancer cells showed that they

preferentially metastise to an abdominal fat pad (omentum). In this study it was found that

omental adipocytes promote homing of these ovarian cancer cells through the induction of

specific adipocyte derived cytokines.(153) Taken together, it is clear that altered lipid

metabolism has beneficial effects for tumors.

Chapter 1

26

5. Cardiolipin

Cardiolipin (CL) is a unique phospholipid, which is predominantly localized in the inner

mitochondrial membrane (IMM) where it is synthesized from phosphatidylglycerol and

cytidinediphosphate‐diacylglycerol with the last step being catalyzed by the enzyme

cardiolipin synthase.(154, 155), CL contains a glycerol backbone and four acyl chains which are

highly unsaturated. These unsaturated acyl chains are necessary for the normal cellular

function.(156) The precise mechanism of remodeling of the acyl chains is unclear, however it

is known that newly synthesized CL is deacylated by a CL‐specific deacylase (Cld1).(156, 157)

One saturated fatty acyl chain is removed from CL by Cld1 to form monolysocardiolipin.(157)

To form mature CL, an unsaturated fatty acid is reacylated into monolysocardiolipin by the

transacylase tafazzin (Taz1) (see figure 4).(158‐160) Mutations in the tafazzin gene results in a

disease called Barth syndrome, an X‐linked recessive childhood disorder. (MIM 302060) The

characteristics of the disease are cardioskeletal myopathy, neutropenia and abnormal

growth.(161, 162) Biochemically, Barth syndrome is characterized by decreased levels of CL and

increases in monolysocardiolipin(163, 164) and a shift is observed in the level of unsaturation

of the acyl side chains.(164) In addition, the CL species in Barth syndrome cells are more

saturated than the CL species in control cells.(164)

Phospholipids are the main component of membranes and are important for maintaining

the structural integrity of membranes. CL has several functions in the mitochondria. CL is

required for optimal activity of complex I (NADH: ubiquinone oxido‐reductase), complex III

(ubiquinone: cytochrome c oxido‐reductase), complex IV (cytochrome c oxidase) and

complex V (ATP synthase).(165) CL is associated with and modulates the activity of several

enzymes involved in the respitory chain, including cytochrome c oxidase, carnitine

palmitoyltransferase, creatinephospokinase, the pyruvate translocator, the phosphate

transporter, mono‐, di‐, and tricarboxylate carriers, glycerol‐3‐phosphate dehydrogenase,

ATP/ADP translocase and ATP synthase.(154, 166, 167 )

Next to its role in normal mitochondrial physiology, CL has also been shown to associate

with members of the apoptotic machinery, including cytochrome c, Bid and caspase 8.(165)

Cytochrome c interacts with CL in the outer leaflet of the IMM through two independent

binding sites.(168‐171) Kagan et al. found that CL‐bound cytochrome c acts as a peroxidase

capable of catalyzing H2O2‐dependent peroxidation of CL and that this CL oxidation is an

essential step in the release of cytochrome c during apoptosis.(172) During apoptosis, Bid is

cleaved by caspase 8 to produce tBid. tBid translocates to the mitochondria and binds to

CL‐enriched contact sites and induces the translocation of BAK and BAX to the

mitochondrial outer membrane.(154, 173) In Fas receptor activated apoptosis, CL provides an

anchor and activating platform for caspase 8 translocation to the mitochondria.(174) CL is

also necessary for the function and stabilization of the caspase‐8/BID complex.(175) The

translocation of CL from IMM to the contact sites is mediated by phospholipid scramblase

Introduction

27

3 and is important for the apoptotic signal.(176) Mitofusin‐1 interacts with OPA1, which

controls the tight structure of the mitochondrial cristae, keeping most cytochrome c in the

cristae. tBid formation also results in the disruption of OPA1‐mediated tight structure of the

cristae resulting in the release of cytochrome c into the intermembrane space and thereby

facilitates its release from mitochondria when these are permeabilized.(154)

Figure 4: Cardiolipin synthesis and remodeling in eukaryotes. In cardiolipin synthesis, phosphatidic is

converted into cytidinediphosphate‐diacylglycerol (CDP‐DAG) by CDP‐DAG synthase (CDS). CDP‐DAG is

converted by phosphatidylglycerolphosphate (PG‐P) synthase (PGPS) to PG‐P. Dephosphorylation of PG‐P

by a phosphatase results in phosphatidylglycerol (PG). In the final step of CL biosynthesis, condensation of

one molecule of PG and one molecule of CDP‐DAG by CL synthase (CS) results in the formation of immature

CL. Remodeling occurs via Taffazin where phospholipase A2 (PLA2) hydrolysis an acyl chain from immature

cardiolipin to generate monolyso‐cardiolipin (MLCL). Taffazin reacylates MLCL to generate mature and

remodeled cardiolipin. Cardiolipin is degraded by hydrolysation by PLA2 via MLCL and dilyso‐cardiolipin

(DLCL).

6. Cancer therapeutics: history and trends

Cancer therapy was initially mainly focused on targeting DNA integrity and/or replication of

DNA, or on blocking mitosis by interfering with microtubule dynamics of the mitotic

spindle.(177) An example of these types of drugs are the platinum based derivatives (i.e

cisplatin), topoisomerease inhibitors (i.e irinotecan), vinca alkaloids (i.e vinblastine) and

taxanes (i.e taxol). These chemotherapeutics are still widely used in the clinic today.

Unfortunately, in a lot of instances resistance occurs.(178) For instance, in the case of cisplatin

DNA damage is followed by apoptosis, which can be counteracted by a multitude of

resistance mechanisms that can for instance arise as a consequence of intracellular changes

P

Phosphatidic acid

CDS

CMP

CDP-DAG

PGPS

P P

OH

P

PGPP

OH

P OH

PG-P PG

CS

P

OH

P

Immature CL

P

OH

PP

OH

P

OH

TAZ

MLCL Remodeled CL

P

OH

P

OHOH

DLCL

PLA2

Degradation

PLA2

C18:1 C18:1 C18:1

C16:0

C18:2 C18:2 C18:2C18:2

C18:1 C18:1 C18:1

Chapter 1

28

that either effectively repair cisplatin‐induced DNA‐adduct, prevent DNA damage signals

from activating the apoptotic machinery, or prevent cisplatin uptake.(179) Mechanisms that

interfere with DNA damage signal to the apoptotic machinery include loss of damage

recognition, loss of p53 function, overexpression of HER‐2/neu, activation of the PI3‐K/Akt

pathway, overexpression of anti‐apoptotic proteins, and interference in caspase

activation.(179) Another downside of these agents are that they block DNA replication and

cell division, thereby killing all cells that are rapidly dividing, not only cancer cells.

The growing insight into the hallmarks of cancer and several important technical

developments like the next generation DNA sequencing techniques and advances in

pharmaceutical drug discovery (180) have contributed to the search for therapies specifically

targeting these hallmarks, the so‐called targeted therapies. Some cellular targets are

genetically altered in cancer cells and are essential to tumor development and survival. This

phenomenon has been coined oncogene addiction.(181) Other targets are not genetically

altered, but still their expression is more important in cancer cells than in normal cells,

which is known as non‐oncogene addiction.(181) For an overview of different targeted

therapies see table 1.

A well‐known prototype and the first clinically approved targeted drug is imatinib (Gleevec,

Novartis), a small molecule inhibiting the Abelson tyrosine kinase (ABL) that is translocated

from chromosome 9 to the breaking point cluster region (BCR) gene on chromosome 22,

forming an oncogenic BCR‐ABL gene fusion, which is found in chronic myeloid leukemia

(CML). Imatinib is well tolerated as chronic therapy and induces molecular remission in

chronic CML.(182) Initially, the dramatic molecular and clinical effects of imatinib raised the

hope that this drug would be curative for most treated patients, however soon it became

apparent that therapy resistance to this inhibitor is frequent owing to mutations in the BCR‐

ABL kinase domain.(183)

There is growing evidence that tumors can escape hallmark‐targeting therapy. For example,

a targeted therapeutic agent inhibiting one key pathway in a tumor may not completely

block a hallmark, allowing some tumor cells to survive with residual function. These cells or

their progeny will eventually adapt to the selective pressure established by the therapy

applied. The adaptations can be accomplished by mutation, epigenetic changes or

remodeling of the microenvironment resulting in renewed tumor growth and clinical

relapse.(4) In response to targeted therapy, cancer cells can reduce their dependence on one

hallmark, becoming more dependent on another. This phenomenon was for instance shown

by the unexpected responses to antiangiogenic therapies. Folkman and Kalluri anticipated

that effective inhibition of angiogenesis would lead to dormant tumors and maybe lead to

their dissolution.(184) However, the clinical responses to antiangiogenic drugs have been

found to be transient.(185‐187) Clinical validation for this evasive resistance is shown by the

increased invasion and local metastasis observed in glioblastomas treated with

Introduction

29

antiangiogenic therapies.(188, 189) These acquired resistance to therapies targeting only one

hallmark are an obstacle in cancer therapy therefore cotargeting of multiple core and

emerging hallmarks will result in more effective therapies.(4)

6.1. Synthetic lethality

To overcome resistance and to specifically target cancer cells but not normal cells, the

exploitation of the concept of synthetic lethality has shown to be of great potential.(203, 204)

Synthetic lethality is derived from classical genetic studies and is based on the interaction

of two genes that both contribute to an essential process. When either gene is mutated

alone, the cell is viable, however the combination of mutations in both genes results in cell

death. This process is referred to as synthetic lethality as it is not possible to directly isolate

such cells as they die instantly when they have both mutations.(205) In cancer therapy this

synthetic lethality is being explored by developing therapies inhibiting a specific gene

product that is the synthetic lethal partner of the cancer related mutation. Not all

oncogenes are targetable by pharmacological intervention, thus synthetic lethality is an

option for these undrugable targets. On top of that, synthetic lethal therapeutic agents can

be applied alone by monotherapy, or in combination with more conventional

treatments.(206‐208) Therapies that target synthetic lethal partners of mutations in cancers

cells are cancer specific, resulting in less off‐target side effects. Examples of this method is

PARP1 inhibition in BRCA1/2 mutant tumors. BRCA1/2 are important in DNA repair of

double strand breaks, PARP1 is an enzyme involved in single strand break repair. Inhibition

of PARP1 results in an accumulation of DNA breaks which are not repaired resulting in cell

death.(209, 210)

An example that looks like synthetic lethality is targeting the RAF‐MEK pathway in KRAS

mutant tumors. Combination of RAF inhibition together with MEK inhibition results in

synthetic lethality of colorectal cancer and lung cancer cells that have a KRAS mutation.(211)

They found that inhibition of MEK increases the GTP‐bound fraction of KRAS, promotes the

formation of RAF1‐BRAF heterodimers and drives constitutive phosphorylation of ERK.(211)

Inhibition of RAF1 therefore results in a block in the phosphorylation of ERK. These results

are in accordance with a report showing that RAF1 knockdown enhanced MEK inhibition in

a KRAS mutant model.(212)

Synthetic lethality is a promising treatment mechanism, which could target tumors very

effectively with less off‐target effects. Currently with all these targeted therapies and

synthetic lethality approaches researchers are doing their best to find a cure for cancer.

However, still a lot of therapies fail to eradicate the tumor. This is suggested to be a cause

of so‐called cancer stem cells. This will be discussed below

Chapter 1

30

Table 1: Therapeutic Targeting of the Hallmarks of Cancer.

Cancer Hallmark

Targeted therapy (example)

Cancer type Drug name Resistance

Sustaining proliferative signaling

Tyrosine kinase inhibitors

Colorectal cancer, Lung cancer, Chronic myeloid leukemia

Cetuximab, Erlotinib, Imatinib

Yes (183, 190‐192)

Evading apoptosis

Cyclin‐dependent kinase inhibitors

Chronic lymphocytic leukemia, lung cancer

Alvocidib, Seliciclib n.o (193)

Avoiding immune destruction

Immune activating mAb

Melanoma, non‐Hodgkin lymphoma

Tremelimumab, Rituximab

n.o, yes (Rituximab)(194, 195)

Enabling replicative immorality

Telomerase inhibitors

Multiple myeloma, myelofibrosis

Imetalstat n.o (Clinical trials.gov)

Tumor promoting inflammation

Selective anti‐inflammatory drugs

Familial adenomatous polpolis

Celecoxib n.o (196, 197)

Activating invasion & metastasis

Inhibitors of HGF/c‐Met

Lung cancer, medullary thyroid cancer

Cabozantinib, Foretinib, Tivatinib

n.o ((198) and Clinical trial.gov)

Inducing angiogenesis

Inhibitors of VEGF signaling

Metastatic cancers (colon, lung,ovarian)

Bevacizumab Yes (199)

Genome instability &mutation

PARP inhibitors

Ovarian, breast, prostate

Olaparib n.o (200)

Resisting cell death

Proaoptotic BH3 mimetics

Acute myeloid leukemia, chronic lymphocytic leukemia

ABT‐199, ABT‐263 n.o ( (201) and Clinical trial.gov)

Deregulating cellular energetics

Aerobic glycolysis inhibitors

Pre‐clinical 3‐bromopyruvate n.o (202)

Abbreviations: mAb, monoclonal antibody; n.o, not observed. This lists only represents a small part of the

therapeutics currently used or under investigation.

Introduction

31

7. Tumor heterogeneity and cancer stem cells

Intra‐tumor heterogeneity arise among cancer cells within the same tumor as a

consequence of genetic changes, environmental differences, epigenetic changes and

reversible changes in cell properties.(213) Two main conceptual frameworks have been

elaborated to conceptualize the link between intra‐tumor heterogeneity and therapy

resistance. The first, and most established idea, is clonal evolution, where the tumor arises

from a single mutated cell and over time acquires additional mutations resulting in

subpopulations with evolutionary advantages.(214) The other one is the cancer stem‐cell

model. This model hypothesizes that cancers are organized into a hierarchy of

subpopulations of tumorigenic cancer stem cells (CSCs) and their non‐tumorigenic

progeny.(213, 215‐217) CSCs are defined as a subset of tumor cells which possess self‐renewal

and multi‐lineage differentiation potential(218‐220) CSCs can be identified by various

markers(219), and in vivo experiments have shown that CSCs very efficiently form tumors that

resemble the original tumor from which the CSCs were derived.(221) It has been suggested

that CSCs are responsible for recurrences and metastasis(219), because they are thought to

be therapy resistant (both chemo‐ and radiotherapy) and to effectively repopulate the

tumor.(4, 221, 222)

Currently, a lot of research is being done on targeting CSCs and deplete the tumor. However,

up to date no specific CSCs targeted therapy is available. Therefore, drugs attacking cancer

cells more broadly may be usable across multiple cancers and CSCs. This possible new

approach consists of compounds that are not per se cancer‐specific but work selective

against cancer due to alterations in metabolism and phenotype of cancer cells as compared

to normal cells. Currently, several natural compounds are intensively investigated as

potential cancer drugs in this perspective.

8. Natural compounds

A significant percentage of golden standard chemotherapeutic agents are based on

compounds from nature. Currently there are several classes of natural compounds or their

semi‐synthetic derivatives used in the clinic like vinca alkaloids, taxanes and

anthracyclines.(223‐228) Vinca alkaloids were introduced into the clinics in late 1950s as

Velban® and Oncovin®, and later on Eldisine® and Navelbine® followed.(223) The mechanism

of action of these compounds has been identified as the destabilization of microtubules,

which leads to G2/M cell cycle arrest and apoptosis.(229) Taxanes belong to the terpenoids

and one well known compound is Taxol (from Taxus)(225), which was shown in 1979 to

stabilize microtubule assembly.(226) It is used in the clinic for the treatment of various cancer

types, including ovarian, lung and breast cancer.(227) Anthracyclines are DNA‐intercalating

agents that block cell division and they are derived from bacteria (Streptomyces).

Doxorubicin is one of the best known members of this family and is used in the clinic for

treatment of several cancer types.(228, 230)

Chapter 1

32

Terpenoids (containing five‐carbon isoprene units) of which the triterpenoids (containing

six‐carbon isoprene units; e.g. plant sterols) are a subclass(231) can be isolated from many

different plant sources and occur in different variations.(232) Several of these variations or

their synthetic derivatives are investigated as potential medicinal products against various

diseases including cancer.(233) One of the most promising triterpenoids with anti‐cancer

activity is Betulinic acid (BetA, 3b‐hydroxy‐lup‐20(29)‐en‐28‐oic acid, C30H48O3), which is

discussed extensively below.

8.1. Betulinic acid

BetA is a pentacyclic triterpene of the lupane type that can be isolated from various plant

sources in small amounts. Its precursor betuline, which can be easily converted by oxidation

into BetA, is available in high abundance from the bark of the white birch (Betula

pubescens). Betulinic acid was for the first time isolated in 1948 from the bark of the London

plane tree (Platanus acerifolia) by Bruckner.(234)

8.2. Betulinic acid: pharmaceutical properties

Different pharmaceutical properties have been attributed to BetA. In particular its capacity

to suppress HIV infection has attracted much attention. In 1994 it was shown for the first

time that BetA has an potent anti‐HIV activity.(235) The wish to still optimize its anti‐HIV

activity together with intellectual property issues, has led to the synthesis of various BetA

derivatives. Improved activity against HIV‐1(236, 237), and HIV‐2(238) of several BetA derivatives

has been documented. Especially BetA derivative PA‐457 has been preclinically developed

as HIV‐1 maturation inhibitor that specifically inhibit the last step in processing of Gag. (237)

Under the name Bevirimat this compound has been tested as HIV‐1 drug in the clinic in

phase I/II clinical trials as single dose administration with no Bevirimat resistance mutations

occurring in this study.(239) Unfortunately, other studies showed that mutations in Gag occur

and result in resistance to the compound.(240‐243) Optimization of BetA derivatives against

HIV‐1 and HIV‐2 and Bevirimat is still ongoing with reported positive results.(244‐246)

Besides its anti‐ HIV activity, BetA and its derivatives have been described to have

therapeutic activity against inflammation (247, 248) and malaria.(249‐252) Anti‐angiogenic(253‐256),

and anti‐fibrotic activity by BetA has also been reported.(257)

Introduction

33

Figure 5: Chemical structure of BetA

8.3 Anti‐cancer activity of Betulinic acid

In 1995, BetA was selected out of 2500 plant extracts in a screen by the National Cancer

Institute as it displayed potent in vitro cytotoxic effects against human melanoma cells.

Follow‐up studies showed the first in vivo efficacy of BetA against melanoma.(258)

Subsequently, it was shown that BetA is also effective against other tumor types like brain

tumors, including the neuroectodermal glioma, medulloblastoma and glioblastoma and

showed no toxicity in murine non‐malignant neuronal cells.(259‐262) In 2001, anti‐proliferative

effects of BetA in non‐small and small cell lung cancer, ovarian carcinoma, cervix carcinoma

were observed, while in contrast, peripheral blood lymphocytes and human normal dermal

fibroblast were not affected by BetA, suggesting that BetA has a tumor specific effect.(263)

Two years later, head and neck squamous cellular carcinoma cells were also shown to be

sensitive to BetA(264) Until 2007 no research had been performed on the efficacy of BetA in

prevalent solid tumor types such as breast‐, lung‐, colon‐ and prostate cancer. In a study

conducted by our group, using three different assays, all tested cell lines from solid tumors

were sensitive to BetA: 10 lung cancer cell lines, 10 colorectal cancer cell lines, 9 breast

cancer cell lines, 4 prostate cancer cell lines and 3 cervix carcinoma cell lines were strongly

affected in their growth potential and viability after BetA treatment.(265) BetA was shown to

be a very potent agent against tumor with EC50 values around 10µg/ml in lung cancer cell

lines, prostate cancer cell lines, cervical cancer cell lines and values ranging from 4‐16 µg/ml

in colorectal cancer cell lines and breast cancer cell lines.(265) In this study it was also shown

that BetA even in high doses has no effect on healthy cells such as human blood‐derived

PBMC, cytotoxic T lymphocyte clones and activated B cells.(265) The efficacy of BetA was also

shown in hepatoblastoma(266), rhabdomyosarcoma(267) and nasopharyngeal carcinoma(268)

using several concentrations of BetA ranging up to 20 µg/ml. BetA has also been shown to

be effective in hematological malignancies i.e BetA induced cell death in murine leukemia

OH

O

HO

Chapter 1

34

cell line L1210(269) and in leukemia cells form patients and cell lines.(270) The last study

showed cell death inducing capacities of BetA independent of risk stratification, age and sex

of the patient and leukemia type.(270) BetA also induced cell death in human CML cell line K‐

562.(271) Combined these studies indicate that BetA is a promising anti‐cancer compound

because of its potent and broad tumor selective activity.

8.4. Mechanism of Betulinic acid

Thus far, the precise mechanism of anti‐cancer action of BetA has not been identified.

Although various and broad intra cellular anti‐cancer effects of the compound have been

revealed until now, these effects were difficult to be linked with unique and defined

molecular target(s). However, accumulating research of various research groups has

revealed important characteristics of BetA‐induced cancer cell death.

8.4.1. Betulinic acid and apoptosis

In 1997 Fulda et al. showed that in SHEP neuroblastoma cells apoptosis was induced after

BetA treatment and that overexpression of BCL‐2 and BCL‐XL blocked BetA‐induced loss of

mitochondrial membrane potential, PARP cleavage, caspase processing and ROS

production.(259) Subsequently, in isolated mitochondria of SHEP cells, BetA triggered

permeability transition (PT) and cytochrome c release, while mitochondria derived from

SHEP cells that overexpressed BCL‐2 or BCL‐XL were resistant to BetA.(272) Consistent with

these data, BetA induced ROS production, which was inhibited by BCL‐2 overexpression or

antioxidants.(262)

Although BCL‐2 overexpression showed protection against BetA‐induced apoptosis in SHEP

and glioma cells, it only partially protected Jurkat cells(270, 273), MCF‐7 cells(273) and melanoma

cells(274) against BetA. These results suggest that the protective effects of anti‐apoptotic

BCL‐2 family members are possibly cell type dependent. In 2009, our group showed that

BetA induced cytochrome c release and apoptosis in BAK/BAX double deficient mouse

embryonic fibroblasts (MEFs) and in HCT116 double knock out BAK/BAX cells. The levels of

apoptosis induced in the double deficient cells were comparable with the wild‐type control

cells, suggesting that BetA induces cytochrome c release and apoptosis independent of

BAK/BAX.(273) Our study also showed that BetA‐induced cell death could be blocked by the

addition of cyclosporine A, a PT‐pore inhibitor suggesting an important role for this pore in

BetA‐induced cell death.(273)

Interestingly, our group showed that BetA induced cell death in the presence of pan‐caspase

inhibitor zVAD.fmk in Jurkat cells, and at the same time PARP processing and DNA

fragmentation were completely blocked.(265) These results suggest that the cytotoxic effects

of BetA are not completely caspase dependent and that other cell death pathways are likely

involved.

Introduction

35

8.4.2. Betulinic acid and autophagic cell death

The focus in unraveling the mechanism of BetA has been mainly on apoptosis. However in

2012 a study of Gonzalez et al. showed that treatment with a glycosylated derivative of

BetA, named B10, leads to autophagic cell death. In this work it was shown that

downregulation of autophagy genes ATG7, ATG5 and BECN1 by RNAi‐mediated suppression

significantly decreased B10‐induced cell death. B10 induces autophagy and disrupts

autophagic flux as measured by LC3 lipidation.(275) Subsequently, by another research group

it was shown that BetA induces autophagy in multiple myeloma cells by an induction of LC3

lipidation. However a block in autophagic flux was observed as P62 levels were also

increased after BetA treatment.(276) The main conclusion of the latter study was that BetA

inhibits autophagic flux and the compound induced cell death via apoptosis. These two

studies, one using a derivative of BetA, are the only studies described.

With these contradicting results and as above mentioned the role of autophagy in the

induction of cell death remains unclear. Therefore more research about the role of BetA in

autophagic cell death is needed. In this thesis we investigated the role of autophagy as a

cell death mechanism (chapter 2)

8.5. Betulinic acid in combination therapy

Several studies have described the effect of BetA in combination with another therapeutic

intervention. In 2000, it was shown that the combination of BetA with radiotherapy had an

additive effect in human melanoma cells.(277) BetA improved vincristine‐induced cytoxicity

in mouse melanoma cells and reduced the number of metastases in vivo.(278) Fulda et al.

showed that combination of BetA and anticancer drugs cisplatin or doxorubicin cooperated

to induce apoptosis and to inhibit clonogenic survival of tumor cells.(279) It was also shown

that BetA sensitizes drug‐resistant colon cancer cells and esophageal squamous carcinoma

cells to oxaliplatin, irinotecan and 5‐fluorouacil.(280, 281) The combination of BetA and

docetaxel or 2‐methoxyestradiol resulted in increased apoptosis.(282) A study by the group

of Vordermark showed that an additive effect of BetA and radiotherapy was also observed

in glioma cells. This effect was most pronounced when cells were hypoxic.(283) On top of that

it was shown that in pancreatic cancer the combination of BetA and mithramycin A (a DNA‐

binding, anti‐tumor and neuroprotective antibiotic originally isolated from S. grieseus)

resulted in an inhibitory effect on cell proliferation, invasion and angiogenesis.(284) This

combination also resulted in less discernible side effects than gemcitabine alone, which was

used as a standard reference drug.(284) Together these data indicate that BetA is a very

potent compounds for combination treatments.

Chapter 1

36

8.6. Betulinic acid in vivo

Besides many in vitro studies, BetA has also been shown to be effective in vivo. Because of

the lipophilic characteristics of BetA and its consequently poor solubility in vivo studies has

been limited. The first study describing in vivo application of BetA used athymic mice with

human melanomas xenografted subcutaneously that were treated by intraperitoneal

injections of BetA in a formulation with polyvinylpyrrolidone (PVP).(258) A dose of 50 mg/kg

body weight was injected every four days for six times resulting in tumor regression. In this

study no toxicity as measured by weight loss was observed up to 500 mg/kg body weight.(258)

As described before BetA and in combination with vincristine reduced the number of

metastases. The treatment dose of BetA (dissolved in DMSO) was 10 mg/kg divided over 10

days out of 17 days.(278) BetA dissolved in DMSO was also used in breast cancer xenograft

model where mice were treated intraperitoneal with 50 and 100 mg/kg bodyweight BetA

every 3‐4 days for 6 doses resulting in delayed tumor growth.(285) Another study, using

human ovarian carcinoma xenograft model showed a clear survival benefit for mice treated

with intraperitoneal injections of a formulation of ethanol, Tween‐80 and water (10%, 10%

and 80%). Mice were injected every 3‐4 days with 100 mg/kg bodyweight for a total of six

injections.(263) Oral treatment with 10 or 20 mg/kg bodyweight BetA in corn oil in mice

bearing subcutaneously grafted prostate cancer cell line resulted in inhibition of tumor

outgrowth as compared to corn oil alone.(286) Also intravenous injection of BetA (10 mg/kg

bodyweight, every day for 14 days) has been shown to reduce adenomacarcinoma

xenograft size.(287) All these studies showed no systemic signs of toxicity, however the

formulations used are either not suitable for human application or are not precisely defined

and therefore cannot be standardized. Therefore our group investigated liposomes as a

delivery system of BetA. Liposomes are small vesicles consisting of one or more concentric

phospholipid bilayers with an aqueous core.(288) Both water soluble as well as lipophilic

compounds can be incorporated in the liposomes. This study showed that intravenous

injection of liposomes containing BetA both reduced tumor growth and prolongs survival of

colon carcinoma and lung carcinoma bearing mice. Mice were treated three times a week

with 50mg/kg bodyweight for three months (lung carcinoma) and two months (colon

carcinoma).(288) This study also showed that oral administration of BetA liposomes reduces

tumor growth and prolongs survival of colon carcinoma bearing mice, although these data

were less pronounced as intravenous injections.(288) In this study no systemic toxicity was

observed, suggesting that BetA‐liposome formulation is a possible formulation for BetA to

be used in preclinical studies.

Introduction

37

Outline of this thesis

The scope of this thesis was to investigate the mechanisms by which BetA induces cell death

in cancer cells in more detail. At the start of the studies described in this thesis several

questions urgently needed an answer. Although BetA induces cell death via apoptosis, when

blocking this form of cell death cancer cells still die.(265) In tumor therapy it is still unclear

whether activation of autophagy contributes to cell death or rather represents a

resistance/survival mechanism. We therefore investigated the roles of other cell death

mechanisms including necroptosis and autophagy in BetA‐induced cell death. The results of

this investigation are described in chapter 2. BetA induces cell death independently of

BAK/BAX but in a mitochondrial dependent fashion, but no reports exist revealing how this

form of cell death occurs, we set off to study the mechanism by which BetA induces cell

death in more detail. We found that BetA interferes with lipid cell metabolism. The results

of this research are described in chapter 3. Cancer stem cells are tumor resistant and

currently no effective therapy has been shown for these cells. BetA is a very potent and

broad acting compound therefore we treated colon cancer stem cells with BetA. Results of

these experiments are laid out in chapter 4. Finally, we were also interested in the effect of

BetA on several lipids in the cell. We set‐up a lipidomics pipeline and used a BetA dataset

to test the functionality of this pipeline as well as to investigate the difference in lipids after

BetA treatment. Chapter 5 contains these results. In chapter 6 all data of this thesis are

discussed in relation to the literature.

Chapter 1

38

References

1. Weinberg RA. Oncogenes and tumor suppressor genes. CA Cancer J Clin. 1994;44(3):160‐70. 2. Weinberg RA. Coming full circle‐from endless complexity to simplicity and back again. Cell. 2014;157(1):267‐71. 3. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell. 2000;100(1):57‐70. 4. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell. 2011;144(5):646‐74. 5. Galluzzi L, Vitale I, Abrams JM, Alnemri ES, Baehrecke EH, Blagosklonny MV, et al. Molecular definitions of cell death subroutines: recommendations of the Nomenclature Committee on Cell Death 2012. Cell Death Differ. 2012;19(1):107‐20. 6. Ashkenazi A, Dixit VM. Death receptors: signaling and modulation. Science. 1998;281(5381):1305‐8. 7. Ashkenazi A, Holland P, Eckhardt SG. Ligand‐based targeting of apoptosis in cancer: the potential of recombinant human apoptosis ligand 2/Tumor necrosis factor‐related apoptosis‐inducing ligand (rhApo2L/TRAIL). J Clin Oncol. 2008;26(21):3621‐30. 8. Elmore S. Apoptosis: a review of programmed cell death. Toxicol Pathol. 2007;35(4):495‐516. 9. Roy S, Nicholson DW. Cross‐talk in cell death signaling. J Exp Med. 2000;192(8):F21‐5. 10. Schutze S, Tchikov V, Schneider‐Brachert W. Regulation of TNFR1 and CD95 signalling by receptor compartmentalization. Nat Rev Mol Cell Biol. 2008;9(8):655‐62. 11. Weinlich R, Dillon CP, Green DR. Ripped to death. Trends Cell Biol. 2011;21(11):630‐7.

12. Medema JP, Scaffidi C, Kischkel FC, Shevchenko A, Mann M, Krammer PH, et al. FLICE is activated by association with the CD95 death‐inducing signaling complex (DISC). EMBO J. 1997;16(10):2794‐804. 13. Scaffidi C, Schmitz I, Zha J, Korsmeyer SJ, Krammer PH, Peter ME. Differential modulation of apoptosis sensitivity in CD95 type I and type II cells. J Biol Chem. 1999;274(32):22532‐8. 14. Barnhart BC, Alappat EC, Peter ME. The CD95 type I/type II model. Semin Immunol. 2003;15(3):185‐93. 15. de Bruin EC, Medema JP. Apoptosis and non‐apoptotic deaths in cancer development and treatment response. Cancer TreatRev. 2008;34(8):737‐49. 16. Cory S, Adams JM. The Bcl2 family: regulators of the cellular life‐or‐death switch. NatRevCancer. 2002;2(9):647‐56. 17. Labi V, Erlacher M, Kiessling S, Villunger A. BH3‐only proteins in cell death initiation, malignant disease and anticancer therapy. Cell Death Differ. 2006;13(8):1325‐38. 18. Chipuk JE, Moldoveanu T, Llambi F, Parsons MJ, Green DR. The BCL‐2 family reunion. Mol Cell. 2010;37(3):299‐310. 19. Wei MC, Zong WX, Cheng EH, Lindsten T, Panoutsakopoulou V, Ross AJ, et al. Proapoptotic BAX and BAK: a requisite gateway to mitochondrial dysfunction and death. Science. 2001;292(5517):727‐30. 20. Czabotar PE, Lessene G, Strasser A, Adams JM. Control of apoptosis by the BCL‐2 protein family: implications for physiology and therapy. Nat Rev Mol Cell Biol. 2014;15(1):49‐63. 21. Danial NN, Korsmeyer SJ. Cell death: critical control points. Cell. 2004;116(2):205‐19.

Introduction

39

22. Youle RJ, Strasser A. The BCL‐2 protein family: opposing activities that mediate cell death. Nat Rev Mol Cell Biol. 2008;9(1):47‐59. 23. Gimenez‐Cassina A, Danial NN. Regulation of mitochondrial nutrient and energy metabolism by BCL‐2 family proteins. Trends Endocrinol Metab. 2015;26(4):165‐75. 24. Tait SW, Green DR. Mitochondria and cell death: outer membrane permeabilization and beyond. Nat Rev Mol Cell Biol. 2010;11(9):621‐32. 25. Yang QH, Church‐Hajduk R, Ren J, Newton ML, Du C. Omi/HtrA2 catalytic cleavage of inhibitor of apoptosis (IAP) irreversibly inactivates IAPs and facilitates caspase activity in apoptosis. Genes Dev. 2003;17(12):1487‐96. 26. Chai J, Du C, Wu JW, Kyin S, Wang X, Shi Y. Structural and biochemical basis of apoptotic activation by Smac/DIABLO. Nature. 2000;406(6798):855‐62. 27. Srinivasula SM, Gupta S, Datta P, Zhang Z, Hegde R, Cheong N, et al. Inhibitor of apoptosis proteins are substrates for the mitochondrial serine protease Omi/HtrA2. J Biol Chem. 2003;278(34):31469‐72. 28. Owen‐Schaub LB, Zhang W, Cusack JC, Angelo LS, Santee SM, Fujiwara T, et al. Wild‐type human p53 and a temperature‐sensitive mutant induce Fas/APO‐1 expression. Mol Cell Biol. 1995;15(6):3032‐40. 29. Muller M, Wilder S, Bannasch D, Israeli D, Lehlbach K, Li‐Weber M, et al. p53 activates the CD95 (APO‐1/Fas) gene in response to DNA damage by anticancer drugs. J Exp Med. 1998;188(11):2033‐45. 30. Miyashita T, Reed JC. Tumor suppressor p53 is a direct transcriptional activator of the human bax gene. Cell. 1995;80(2):293‐9. 31. Oda E, Ohki R, Murasawa H, Nemoto J, Shibue T, Yamashita T, et al. Noxa, a BH3‐only member of the Bcl‐2 family and candidate

mediator of p53‐induced apoptosis. Science. 2000;288(5468):1053‐8. 32. Yu J, Zhang L. PUMA, a potent killer with or without p53. Oncogene. 2008;27 Suppl 1:S71‐83. 33. Chipuk JE, Kuwana T, Bouchier‐Hayes L, Droin NM, Newmeyer DD, Schuler M, et al. Direct activation of Bax by p53 mediates mitochondrial membrane permeabilization and apoptosis. Science. 2004;303(5660):1010‐4. 34. Dice JF. Chaperone‐mediated autophagy. Autophagy. 2007;3(4):295‐9. 35. Rubinsztein DC, Gestwicki JE, Murphy LO, Klionsky DJ. Potential therapeutic applications of autophagy. Nat Rev Drug Discov. 2007;6(4):304‐12. 36. Shintani T, Klionsky DJ. Autophagy in health and disease: a double‐edged sword. Science. 2004;306(5698):990‐5. 37. Mizushima N, Levine B, Cuervo AM, Klionsky DJ. Autophagy fights disease through cellular self‐digestion. Nature. 2008;451(7182):1069‐75. 38. Axe EL, Walker SA, Manifava M, Chandra P, Roderick HL, Habermann A, et al. Autophagosome formation from membrane compartments enriched in phosphatidylinositol 3‐phosphate and dynamically connected to the endoplasmic reticulum. J Cell Biol. 2008;182(4):685‐701. 39. Mizushima N. Autophagy: process and function. Genes Dev. 2007;21(22):2861‐73. 40. Codogno P, Mehrpour M, Proikas‐Cezanne T. Canonical and non‐canonical autophagy: variations on a common theme of self‐eating? Nat Rev Mol Cell Biol. 2012;13(1):7‐12. 41. Mizushima N. The role of the Atg1/ULK1 complex in autophagy regulation. Curr Opin Cell Biol. 2010;22(2):132‐9.

Chapter 1

40

42. Jung CH, Jun CB, Ro SH, Kim YM, Otto NM, Cao J, et al. ULK‐Atg13‐FIP200 complexes mediate mTOR signaling to the autophagy machinery. Mol Biol Cell. 2009;20(7):1992‐2003. 43. Talloczy Z, Jiang W, Virgin HWt, Leib DA, Scheuner D, Kaufman RJ, et al. Regulation of starvation‐ and virus‐induced autophagy by the eIF2alpha kinase signaling pathway. Proc Natl Acad Sci U S A. 2002;99(1):190‐5. 44. Crighton D, Wilkinson S, Ryan KM. DRAM links autophagy to p53 and programmed cell death. Autophagy. 2007;3(1):72‐4. 45. Gozuacik D, Kimchi A. Autophagy and cell death. Curr Top Dev Biol. 2007;78:217‐45. 46. Chi S, Kitanaka C, Noguchi K, Mochizuki T, Nagashima Y, Shirouzu M, et al. Oncogenic Ras triggers cell suicide through the activation of a caspase‐independent cell death program in human cancer cells. Oncogene. 1999;18(13):2281‐90. 47. Bursch W, Ellinger A, Kienzl H, Torok L, Pandey S, Sikorska M, et al. Active cell death induced by the anti‐estrogens tamoxifen and ICI 164 384 in human mammary carcinoma cells (MCF‐7) in culture: the role of autophagy. Carcinogenesis. 1996;17(8):1595‐607. 48. Xue L, Borutaite V, Tolkovsky AM. Inhibition of mitochondrial permeability transition and release of cytochrome c by anti‐apoptotic nucleoside analogues. Biochem Pharmacol. 2002;64(3):441‐9. 49. Yu L, Alva A, Su H, Dutt P, Freundt E, Welsh S, et al. Regulation of an ATG7‐beclin 1 program of autophagic cell death by caspase‐8. Science. 2004;304(5676):1500‐2. 50. Xu Y, Kim SO, Li Y, Han J. Autophagy contributes to caspase‐independent macrophage cell death. J Biol Chem. 2006;281(28):19179‐87. 51. Shimizu S, Kanaseki T, Mizushima N, Mizuta T, Arakawa‐Kobayashi S, Thompson

CB, et al. Role of Bcl‐2 family proteins in a non‐apoptotic programmed cell death dependent on autophagy genes. Nat Cell Biol. 2004;6(12):1221‐8. 52. Levine B, Yuan J. Autophagy in cell death: an innocent convict? J Clin Invest. 2005;115(10):2679‐88. 53. Lockshin RA, Zakeri Z. Apoptosis, autophagy, and more. Int J Biochem Cell Biol. 2004;36(12):2405‐19. 54. Vandenabeele P, Vanden Berghe T, Festjens N. Caspase inhibitors promote alternative cell death pathways. SciSTKE. 2006;2006(358):e44. 55. Kroemer G, Galluzzi L, Vandenabeele P, Abrams J, Alnemri ES, Baehrecke EH, et al. Classification of cell death: recommendations of the Nomenclature Committee on Cell Death 2009. Cell Death Differ. 2009;16(1):3‐11. 56. Vandenabeele P, Galluzzi L, Vanden Berghe T, Kroemer G. Molecular mechanisms of necroptosis: an ordered cellular explosion. NatRevMolCell Biol. 2010;11(10):700‐14. 57. He S, Wang L, Miao L, Wang T, Du F, Zhao L, et al. Receptor interacting protein kinase‐3 determines cellular necrotic response to TNF‐alpha. Cell. 2009;137(6):1100‐11. 58. Cho YS, Challa S, Moquin D, Genga R, Ray TD, Guildford M, et al. Phosphorylation‐driven assembly of the RIP1‐RIP3 complex regulates programmed necrosis and virus‐induced inflammation. Cell. 2009;137(6):1112‐23. 59. Zong WX, Ditsworth D, Bauer DE, Wang ZQ, Thompson CB. Alkylating DNA damage stimulates a regulated form of necrotic cell death. Genes Dev. 2004;18(11):1272‐82. 60. Bano D, Young KW, Guerin CJ, Lefeuvre R, Rothwell NJ, Naldini L, et al. Cleavage of the plasma membrane Na+/Ca2+ exchanger in excitotoxicity. Cell. 2005;120(2):275‐85.

Introduction

41

61. Galluzzi L, Kepp O, Krautwald S, Kroemer G, Linkermann A. Molecular mechanisms of regulated necrosis. Semin Cell Dev Biol. 2014;35:24‐32. 62. Andrabi SA, Dawson TM, Dawson VL. Mitochondrial and nuclear cross talk in cell death: parthanatos. Ann N Y Acad Sci. 2008;1147:233‐41. 63. David KK, Andrabi SA, Dawson TM, Dawson VL. Parthanatos, a messenger of death. Front Biosci (Landmark Ed). 2009;14:1116‐28. 64. Galluzzi L, Vanden Berghe T, Vanlangenakker N, Buettner S, Eisenberg T, Vandenabeele P, et al. Programmed necrosis from molecules to health and disease. Int Rev Cell Mol Biol. 2011;289:1‐35. 65. Bainton DF. The discovery of lysosomes. J Cell Biol. 1981;91(3 Pt 2):66s‐76s. 66. Kirkegaard T, Jaattela M. Lysosomal involvement in cell death and cancer. Biochim Biophys Acta. 2009;1793(4):746‐54. 67. Rawlings ND, Barrett AJ. MEROPS: the peptidase database. Nucleic Acids Res. 1999;27(1):325‐31. 68. Boya P, Kroemer G. Lysosomal membrane permeabilization in cell death. Oncogene. 2008;27(50):6434‐51. 69. Kroemer G, Jaattela M. Lysosomes and autophagy in cell death control. Nat Rev Cancer. 2005;5(11):886‐97. 70. Petersen NH, Kirkegaard T, Olsen OD, Jaattela M. Connecting Hsp70, sphingolipid metabolism and lysosomal stability. Cell Cycle. 2010;9(12):2305‐9. 71. Aits S, Jaattela M. Lysosomal cell death at a glance. JCell Sci. 2013;126(Pt 9):1905‐12. 72. Turk V, Turk B, Turk D. Lysosomal cysteine proteases: facts and opportunities. EMBO J. 2001;20(17):4629‐33.

73. Reinheckel T, Deussing J, Roth W, Peters C. Towards specific functions of lysosomal cysteine peptidases: phenotypes of mice deficient for cathepsin B or cathepsin L. Biol Chem. 2001;382(5):735‐41. 74. Balaji KN, Schaschke N, Machleidt W, Catalfamo M, Henkart PA. Surface cathepsin B protects cytotoxic lymphocytes from self‐destruction after degranulation. J Exp Med. 2002;196(4):493‐503. 75. Guicciardi ME, Leist M, Gores GJ. Lysosomes in cell death. Oncogene. 2004;23(16):2881‐90. 76. Mohamed MM, Sloane BF. Cysteine cathepsins: multifunctional enzymes in cancer. Nat Rev Cancer. 2006;6(10):764‐75. 77. Jaattela M. Multiple cell death pathways as regulators of tumour initiation and progression. Oncogene. 2004;23(16):2746‐56. 78. Suminami Y, Nawata S, Kato H. Biological role of SCC antigen. Tumour Biol. 1998;19(6):488‐93. 79. Abrahamson M, Alvarez‐Fernandez M, Nathanson CM. Cystatins. Biochem Soc Symp. 2003(70):179‐99. 80. Turk B, Turk V, Turk D. Structural and functional aspects of papain‐like cysteine proteinases and their protein inhibitors. Biol Chem. 1997;378(3‐4):141‐50. 81. Pennacchio LA, Bouley DM, Higgins KM, Scott MP, Noebels JL, Myers RM. Progressive ataxia, myoclonic epilepsy and cerebellar apoptosis in cystatin B‐deficient mice. Nat Genet. 1998;20(3):251‐8. 82. Kolter T, Sandhoff K. Lysosomal degradation of membrane lipids. FEBS Lett. 2010;584(9):1700‐12. 83. Kirkegaard T, Roth AG, Petersen NH, Mahalka AK, Olsen OD, Moilanen I, et al. Hsp70 stabilizes lysosomes and reverts

Chapter 1

42

Niemann‐Pick disease‐associated lysosomal pathology. Nature. 2010;463(7280):549‐53. 84. Jenkins RW, Canals D, Hannun YA. Roles and regulation of secretory and lysosomal acid sphingomyelinase. Cell Signal. 2009;21(6):836‐46. 85. Shortt J, Johnstone RW. Oncogenes in cell survival and cell death. Cold Spring Harb Perspect Biol. 2012;4(12). 86. Hassan M, Watari H, AbuAlmaaty A, Ohba Y, Sakuragi N. Apoptosis and molecular targeting therapy in cancer. Biomed Res Int. 2014;2014:150845. 87. Green DR, Evan GI. A matter of life and death. Cancer Cell. 2002;1(1):19‐30. 88. Elias A, Siegelin MD, Steinmuller A, von Deimling A, Lass U, Korn B, et al. Epigenetic silencing of death receptor 4 mediates tumor necrosis factor‐related apoptosis‐inducing ligand resistance in gliomas. Clin Cancer Res. 2009;15(17):5457‐65. 89. Horak P, Pils D, Haller G, Pribill I, Roessler M, Tomek S, et al. Contribution of epigenetic silencing of tumor necrosis factor‐related apoptosis inducing ligand receptor 1 (DR4) to TRAIL resistance and ovarian cancer. Mol Cancer Res. 2005;3(6):335‐43. 90. Bagnoli M, Canevari S, Mezzanzanica D. Cellular FLICE‐inhibitory protein (c‐FLIP) signalling: a key regulator of receptor‐mediated apoptosis in physiologic context and in cancer. Int J Biochem Cell Biol. 2010;42(2):210‐3. 91. French LE, Tschopp J. Defective death receptor signaling as a cause of tumor immune escape. Semin Cancer Biol. 2002;12(1):51‐5. 92. Shivapurkar N, Toyooka S, Eby MT, Huang CX, Sathyanarayana UG, Cunningham HT, et al. Differential inactivation of caspase‐8 in lung cancers. Cancer Biol Ther. 2002;1(1):65‐9.

93. Hahne M, Rimoldi D, Schroter M, Romero P, Schreier M, French LE, et al. Melanoma cell expression of Fas(Apo‐1/CD95) ligand: implications for tumor immune escape. Science. 1996;274(5291):1363‐6. 94. Sprecher E, Bergman R, Meilick A, Kerner H, Manov L, Reiter I, et al. Apoptosis, Fas and Fas‐ligand expression in melanocytic tumors. J Cutan Pathol. 1999;26(2):72‐7. 95. Leithauser F, Dhein J, Mechtersheimer G, Koretz K, Bruderlein S, Henne C, et al. Constitutive and induced expression of APO‐1, a new member of the nerve growth factor/tumor necrosis factor receptor superfamily, in normal and neoplastic cells. Lab Invest. 1993;69(4):415‐29. 96. Yu J, Zhang L. The transcriptional targets of p53 in apoptosis control. Biochem Biophys Res Commun. 2005;331(3):851‐8. 97. Amundson SA, Myers TG, Scudiero D, Kitada S, Reed JC, Fornace AJ, Jr. An informatics approach identifying markers of chemosensitivity in human cancer cell lines. Cancer Res. 2000;60(21):6101‐10. 98. Shibata MA, Liu ML, Knudson MC, Shibata E, Yoshidome K, Bandey T, et al. Haploid loss of bax leads to accelerated mammary tumor development in C3(1)/SV40‐TAg transgenic mice: reduction in protective apoptotic response at the preneoplastic stage. EMBO J. 1999;18(10):2692‐701. 99. Eischen CM, Roussel MF, Korsmeyer SJ, Cleveland JL. Bax loss impairs Myc‐induced apoptosis and circumvents the selection of p53 mutations during Myc‐mediated lymphomagenesis. Mol Cell Biol. 2001;21(22):7653‐62. 100. Inoue K, Kohno T, Takakura S, Hayashi Y, Mizoguchi H, Yokota J. Frequent microsatellite instability and BAX mutations in T cell acute lymphoblastic leukemia cell lines. Leuk Res. 2000;24(3):255‐62.

Introduction

43

101. Miquel C, Borrini F, Grandjouan S, Auperin A, Viguier J, Velasco V, et al. Role of bax mutations in apoptosis in colorectal cancers with microsatellite instability. Am J Clin Pathol. 2005;123(4):562‐70. 102. Levine B, Kroemer G. Autophagy in the pathogenesis of disease. Cell. 2008;132(1):27‐42. 103. Botti J, Djavaheri‐Mergny M, Pilatte Y, Codogno P. Autophagy signaling and the cogwheels of cancer. Autophagy. 2006;2(2):67‐73. 104. Pattingre S, Tassa A, Qu X, Garuti R, Liang XH, Mizushima N, et al. Bcl‐2 antiapoptotic proteins inhibit Beclin 1‐dependent autophagy. Cell. 2005;122(6):927‐39. 105. Maiuri MC, Le Toumelin G, Criollo A, Rain JC, Gautier F, Juin P, et al. Functional and physical interaction between Bcl‐X(L) and a BH3‐like domain in Beclin‐1. EMBO J. 2007;26(10):2527‐39. 106. Qu X, Yu J, Bhagat G, Furuya N, Hibshoosh H, Troxel A, et al. Promotion of tumorigenesis by heterozygous disruption of the beclin 1 autophagy gene. J Clin Invest. 2003;112(12):1809‐20. 107. Yue Z, Jin S, Yang C, Levine AJ, Heintz N. Beclin 1, an autophagy gene essential for early embryonic development, is a haploinsufficient tumor suppressor. Proc Natl Acad Sci U S A. 2003;100(25):15077‐82. 108. Mathew R, Karantza‐Wadsworth V, White E. Role of autophagy in cancer. Nat Rev Cancer. 2007;7(12):961‐7. 109. Galluzzi L, Kroemer G. Necroptosis: a specialized pathway of programmed necrosis. Cell. 2008;135(7):1161‐3. 110. Grivennikov SI, Greten FR, Karin M. Immunity, inflammation, and cancer. Cell. 2010;140(6):883‐99.

111. Boroughs LK, DeBerardinis RJ. Metabolic pathways promoting cancer cell survival and growth. Nat Cell Biol. 2015;17(4):351‐9. 112. DeBerardinis RJ, Thompson CB. Cellular metabolism and disease: what do metabolic outliers teach us? Cell. 2012;148(6):1132‐44. 113. Glatz JF, Luiken JJ, Bonen A. Membrane fatty acid transporters as regulators of lipid metabolism: implications for metabolic disease. Physiol Rev. 2010;90(1):367‐417. 114. Santos CR, Schulze A. Lipid metabolism in cancer. FEBS J. 2012;279(15):2610‐23. 115. Mashima T, Seimiya H, Tsuruo T. De novo fatty‐acid synthesis and related pathways as molecular targets for cancer therapy. Br J Cancer. 2009;100(9):1369‐72. 116. Eberle D, Hegarty B, Bossard P, Ferre P, Foufelle F. SREBP transcription factors: master regulators of lipid homeostasis. Biochimie. 2004;86(11):839‐48. 117. Espenshade PJ. SREBPs: sterol‐regulated transcription factors. J Cell Sci. 2006;119(Pt 6):973‐6. 118. Warburg O. On respiratory impairment in cancer cells. Science. 1956;124(3215):269‐70. 119. Warburg O. On the origin of cancer cells. Science. 1956;123(3191):309‐14. 120. Kroemer G, Pouyssegur J. Tumor cell metabolism: cancer's Achilles' heel. Cancer Cell. 2008;13(6):472‐82. 121. Pouyssegur J, Dayan F, Mazure NM. Hypoxia signalling in cancer and approaches to enforce tumour regression. Nature. 2006;441(7092):437‐43. 122. Tennant DA, Duran RV, Boulahbel H, Gottlieb E. Metabolic transformation in cancer. Carcinogenesis. 2009;30(8):1269‐80. 123. Koukourakis MI, Giatromanolaki A, Harris AL, Sivridis E. Comparison of metabolic

Chapter 1

44

pathways between cancer cells and stromal cells in colorectal carcinomas: a metabolic survival role for tumor‐associated stroma. Cancer Res. 2006;66(2):632‐7. 124. Swietach P, Vaughan‐Jones RD, Harris AL. Regulation of tumor pH and the role of carbonic anhydrase 9. Cancer Metastasis Rev. 2007;26(2):299‐310. 125. Fischer K, Hoffmann P, Voelkl S, Meidenbauer N, Ammer J, Edinger M, et al. Inhibitory effect of tumor cell‐derived lactic acid on human T cells. Blood. 2007;109(9):3812‐9. 126. DeBerardinis RJ, Lum JJ, Hatzivassiliou G, Thompson CB. The biology of cancer: metabolic reprogramming fuels cell growth and proliferation. Cell Metab. 2008;7(1):11‐20. 127. Gatenby RA, Gillies RJ. Why do cancers have high aerobic glycolysis? Nat Rev Cancer. 2004;4(11):891‐9. 128. Mathupala SP, Ko YH, Pedersen PL. Hexokinase II: cancer's double‐edged sword acting as both facilitator and gatekeeper of malignancy when bound to mitochondria. Oncogene. 2006;25(34):4777‐86. 129. Capano M, Crompton M. Biphasic translocation of Bax to mitochondria. Biochem J. 2002;367(Pt 1):169‐78. 130. Pastorino JG, Shulga N, Hoek JB. Mitochondrial binding of hexokinase II inhibits Bax‐induced cytochrome c release and apoptosis. J Biol Chem. 2002;277(9):7610‐8. 131. Deberardinis RJ, Sayed N, Ditsworth D, Thompson CB. Brick by brick: metabolism and tumor cell growth. Curr Opin Genet Dev. 2008;18(1):54‐61. 132. Turner A, McGivan JD. Glutaminase isoform expression in cell lines derived from human colorectal adenomas and carcinomas. Biochem J. 2003;370(Pt 2):403‐8.

133. Perez‐Gomez C, Campos‐Sandoval JA, Alonso FJ, Segura JA, Manzanares E, Ruiz‐Sanchez P, et al. Co‐expression of glutaminase K and L isoenzymes in human tumour cells. Biochem J. 2005;386(Pt 3):535‐42. 134. Matsuno T, Goto I. Glutaminase and glutamine synthetase activities in human cirrhotic liver and hepatocellular carcinoma. Cancer Res. 1992;52(5):1192‐4. 135. Mansfield KD, Guzy RD, Pan Y, Young RM, Cash TP, Schumacker PT, et al. Mitochondrial dysfunction resulting from loss of cytochrome c impairs cellular oxygen sensing and hypoxic HIF‐alpha activation. Cell Metab. 2005;1(6):393‐9. 136. Gordan JD, Simon MC. Hypoxia‐inducible factors: central regulators of the tumor phenotype. Curr Opin Genet Dev. 2007;17(1):71‐7. 137. Medes G, Thomas A, Weinhouse S. Metabolism of neoplastic tissue. IV. A study of lipid synthesis in neoplastic tissue slices in vitro. Cancer Res. 1953;13(1):27‐9. 138. Swinnen JV, Brusselmans K, Verhoeven G. Increased lipogenesis in cancer cells: new players, novel targets. CurrOpinClinNutrMetab Care. 2006;9(4):358‐65. 139. Menendez JA, Lupu R. Fatty acid synthase and the lipogenic phenotype in cancer pathogenesis. Nat Rev Cancer. 2007;7(10):763‐77. 140. Li JN, Mahmoud MA, Han WF, Ripple M, Pizer ES. Sterol regulatory element‐binding protein‐1 participates in the regulation of fatty acid synthase expression in colorectal neoplasia. ExpCell Res. 2000;261(1):159‐65. 141. Swinnen JV, Vanderhoydonc F, Elgamal AA, Eelen M, Vercaeren I, Joniau S, et al. Selective activation of the fatty acid synthesis pathway in human prostate cancer. IntJCancer. 2000;88(2):176‐9.

Introduction

45

142. Bauer DE, Hatzivassiliou G, Zhao F, Andreadis C, Thompson CB. ATP citrate lyase is an important component of cell growth and transformation. Oncogene. 2005;24(41):6314‐22. 143. Hatzivassiliou G, Zhao F, Bauer DE, Andreadis C, Shaw AN, Dhanak D, et al. ATP citrate lyase inhibition can suppress tumor cell growth. Cancer Cell. 2005;8(4):311‐21. 144. Porstmann T, Griffiths B, Chung YL, Delpuech O, Griffiths JR, Downward J, et al. PKB/Akt induces transcription of enzymes involved in cholesterol and fatty acid biosynthesis via activation of SREBP. Oncogene. 2005;24(43):6465‐81. 145. Edinger AL, Thompson CB. Akt maintains cell size and survival by increasing mTOR‐dependent nutrient uptake. Mol Biol Cell. 2002;13(7):2276‐88. 146. Wieman HL, Wofford JA, Rathmell JC. Cytokine stimulation promotes glucose uptake via phosphatidylinositol‐3 kinase/Akt regulation of Glut1 activity and trafficking. Mol Biol Cell. 2007;18(4):1437‐46. 147. Buzzai M, Bauer DE, Jones RG, Deberardinis RJ, Hatzivassiliou G, Elstrom RL, et al. The glucose dependence of Akt‐transformed cells can be reversed by pharmacologic activation of fatty acid beta‐oxidation. Oncogene. 2005;24(26):4165‐73. 148. Deberardinis RJ, Lum JJ, Thompson CB. Phosphatidylinositol 3‐kinase‐dependent modulation of carnitine palmitoyltransferase 1A expression regulates lipid metabolism during hematopoietic cell growth. J Biol Chem. 2006;281(49):37372‐80. 149. Scaglia N, Igal RA. Stearoyl‐CoA desaturase is involved in the control of proliferation, anchorage‐independent growth, and survival in human transformed cells. J Biol Chem. 2005;280(27):25339‐49. 150. Currie E, Schulze A, Zechner R, Walther TC, Farese RV, Jr. Cellular fatty acid

metabolism and cancer. Cell Metab. 2013;18(2):153‐61. 151. Baenke F, Peck B, Miess H, Schulze A. Hooked on fat: the role of lipid synthesis in cancer metabolism and tumour development. Dis Model Mech. 2013;6(6):1353‐63. 152. Santolla MF, Lappano R, De Marco P, Pupo M, Vivacqua A, Sisci D, et al. G protein‐coupled estrogen receptor mediates the up‐regulation of fatty acid synthase induced by 17beta‐estradiol in cancer cells and cancer‐associated fibroblasts. J Biol Chem. 2012;287(52):43234‐45. 153. Nieman KM, Kenny HA, Penicka CV, Ladanyi A, Buell‐Gutbrod R, Zillhardt MR, et al. Adipocytes promote ovarian cancer metastasis and provide energy for rapid tumor growth. Nat Med. 2011;17(11):1498‐503. 154. Houtkooper RH, Vaz FM. Cardiolipin, the heart of mitochondrial metabolism. Cell Mol Life Sci. 2008;65(16):2493‐506. 155. Krebs JJ, Hauser H, Carafoli E. Asymmetric distribution of phospholipids in the inner membrane of beef heart mitochondria. J Biol Chem. 1979;254(12):5308‐16. 156. Patil VA, Greenberg ML. Cardiolipin‐mediated cellular signaling. AdvExpMedBiol. 2013;991:195‐213. 157. Beranek A, Rechberger G, Knauer H, Wolinski H, Kohlwein SD, Leber R. Identification of a cardiolipin‐specific phospholipase encoded by the gene CLD1 (YGR110W) in yeast. J Biol Chem. 2009;284(17):11572‐8. 158. Gu Z, Valianpour F, Chen S, Vaz FM, Hakkaart GA, Wanders RJ, et al. Aberrant cardiolipin metabolism in the yeast taz1 mutant: a model for Barth syndrome. Mol Microbiol. 2004;51(1):149‐58.

Chapter 1

46

159. Xu Y, Malhotra A, Ren M, Schlame M. The enzymatic function of tafazzin. J Biol Chem. 2006;281(51):39217‐24. 160. Xu Y, Kelley RI, Blanck TJ, Schlame M. Remodeling of cardiolipin by phospholipid transacylation. J Biol Chem. 2003;278(51):51380‐5. 161. Hauff KD, Hatch GM. Cardiolipin metabolism and Barth Syndrome. Prog Lipid Res. 2006;45(2):91‐101. 162. Schlame M, Ren M. Barth syndrome, a human disorder of cardiolipin metabolism. FEBS Lett. 2006;580(23):5450‐5. 163. Vreken P, Valianpour F, Nijtmans LG, Grivell LA, Plecko B, Wanders RJ, et al. Defective remodeling of cardiolipin and phosphatidylglycerol in Barth syndrome. Biochem Biophys Res Commun. 2000;279(2):378‐82. 164. Valianpour F, Mitsakos V, Schlemmer D, Towbin JA, Taylor JM, Ekert PG, et al. Monolysocardiolipins accumulate in Barth syndrome but do not lead to enhanced apoptosis. J Lipid Res. 2005;46(6):1182‐95. 165. Gonzalvez F, Gottlieb E. Cardiolipin: setting the beat of apoptosis. Apoptosis. 2007;12(5):877‐85. 166. Gonzalvez F, D'Aurelio M, Boutant M, Moustapha A, Puech JP, Landes T, et al. Barth syndrome: cellular compensation of mitochondrial dysfunction and apoptosis inhibition due to changes in cardiolipin remodeling linked to tafazzin (TAZ) gene mutation. Biochim Biophys Acta. 2013;1832(8):1194‐206. 167. Schlame M, Rua D, Greenberg ML. The biosynthesis and functional role of cardiolipin. Prog Lipid Res. 2000;39(3):257‐88. 168. Paradies G, Petrosillo G, Pistolese M, Ruggiero FM. Reactive oxygen species generated by the mitochondrial respiratory chain affect the complex III activity via

cardiolipin peroxidation in beef‐heart submitochondrial particles. Mitochondrion. 2001;1(2):151‐9. 169. Schug ZT, Gottlieb E. Cardiolipin acts as a mitochondrial signalling platform to launch apoptosis. BiochimBiophysActa. 2009;1788(10):2022‐31. 170. Gorbenko GP. Structure of cytochrome c complexes with phospholipids as revealed by resonance energy transfer. Biochim Biophys Acta. 1999;1420(1‐2):1‐13. 171. Iverson SL, Orrenius S. The cardiolipin‐cytochrome c interaction and the mitochondrial regulation of apoptosis. Arch Biochem Biophys. 2004;423(1):37‐46. 172. Kagan VE, Tyurin VA, Jiang J, Tyurina YY, Ritov VB, Amoscato AA, et al. Cytochrome c acts as a cardiolipin oxygenase required for release of proapoptotic factors. Nat Chem Biol. 2005;1(4):223‐32. 173. Esposti MD, Cristea IM, Gaskell SJ, Nakao Y, Dive C. Proapoptotic Bid binds to monolysocardiolipin, a new molecular connection between mitochondrial membranes and cell death. Cell Death Differ. 2003;10(12):1300‐9. 174. Gonzalvez F, Schug ZT, Houtkooper RH, MacKenzie ED, Brooks DG, Wanders RJ, et al. Cardiolipin provides an essential activating platform for caspase‐8 on mitochondria. JCell Biol. 2008;183(4):681‐96. 175. Schug ZT, Gonzalvez F, Houtkooper RH, Vaz FM, Gottlieb E. BID is cleaved by caspase‐8 within a native complex on the mitochondrial membrane. Cell Death Differ. 2011;18(3):538‐48. 176. Liu J, Dai Q, Chen J, Durrant D, Freeman A, Liu T, et al. Phospholipid scramblase 3 controls mitochondrial structure, function, and apoptotic response. Mol Cancer Res. 2003;1(12):892‐902.

Introduction

47

177. Jackson JR, Patrick DR, Dar MM, Huang PS. Targeted anti‐mitotic therapies: can we improve on tubulin agents? Nat Rev Cancer. 2007;7(2):107‐17. 178. Dobbelstein M, Moll U. Targeting tumour‐supportive cellular machineries in anticancer drug development. Nat Rev Drug Discov. 2014;13(3):179‐96. 179. Siddik ZH. Cisplatin: mode of cytotoxic action and molecular basis of resistance. Oncogene. 2003;22(47):7265‐79. 180. Bernards R. A missing link in genotype‐directed cancer therapy. Cell. 2012;151(3):465‐8. 181. Luo J, Solimini NL, Elledge SJ. Principles of cancer therapy: oncogene and non‐oncogene addiction. Cell. 2009;136(5):823‐37. 182. Kantarjian HM, Cortes JE, O'Brien S, Giles F, Garcia‐Manero G, Faderl S, et al. Imatinib mesylate therapy in newly diagnosed patients with Philadelphia chromosome‐positive chronic myelogenous leukemia: high incidence of early complete and major cytogenetic responses. Blood. 2003;101(1):97‐100. 183. Shah NP, Nicoll JM, Nagar B, Gorre ME, Paquette RL, Kuriyan J, et al. Multiple BCR‐ABL kinase domain mutations confer polyclonal resistance to the tyrosine kinase inhibitor imatinib (STI571) in chronic phase and blast crisis chronic myeloid leukemia. Cancer Cell. 2002;2(2):117‐25. 184. Folkman J, Kalluri R. Cancer without disease. Nature. 2004;427(6977):787. 185. Azam F, Mehta S, Harris AL. Mechanisms of resistance to antiangiogenesis therapy. Eur J Cancer. 2010;46(8):1323‐32. 186. Ebos JM, Lee CR, Kerbel RS. Tumor and host‐mediated pathways of resistance and disease progression in response to antiangiogenic therapy. Clin Cancer Res. 2009;15(16):5020‐5.

187. Bergers G, Hanahan D. Modes of resistance to anti‐angiogenic therapy. Nat Rev Cancer. 2008;8(8):592‐603. 188. Ellis LM, Reardon DA. Cancer: The nuances of therapy. Nature. 2009;458(7236):290‐2. 189. Verhoeff JJ, van Tellingen O, Claes A, Stalpers LJ, van Linde ME, Richel DJ, et al. Concerns about anti‐angiogenic treatment in patients with glioblastoma multiforme. BMC Cancer. 2009;9:444. 190. Yonesaka K, Zejnullahu K, Okamoto I, Satoh T, Cappuzzo F, Souglakos J, et al. Activation of ERBB2 signaling causes resistance to the EGFR‐directed therapeutic antibody cetuximab. Sci Transl Med. 2011;3(99):99ra86. 191. Engelman JA, Zejnullahu K, Mitsudomi T, Song Y, Hyland C, Park JO, et al. MET amplification leads to gefitinib resistance in lung cancer by activating ERBB3 signaling. Science. 2007;316(5827):1039‐43. 192. Bean J, Brennan C, Shih JY, Riely G, Viale A, Wang L, et al. MET amplification occurs with or without T790M mutations in EGFR mutant lung tumors with acquired resistance to gefitinib or erlotinib. Proc Natl Acad Sci U S A. 2007;104(52):20932‐7. 193. Gallorini M, Cataldi A, di Giacomo V. Cyclin‐dependent kinase modulators and cancer therapy. BioDrugs. 2012;26(6):377‐91. 194. Reuben JM, Lee BN, Li C, Gomez‐Navarro J, Bozon VA, Parker CA, et al. Biologic and immunomodulatory events after CTLA‐4 blockade with ticilimumab in patients with advanced malignant melanoma. Cancer. 2006;106(11):2437‐44. 195. Rezvani AR, Maloney DG. Rituximab resistance. Best Pract Res Clin Haematol. 2011;24(2):203‐16. 196. Gill S, Sinicrope FA. Colorectal cancer prevention: is an ounce of prevention worth a

Chapter 1

48

pound of cure? Semin Oncol. 2005;32(1):24‐34. 197. Rayburn ER, Ezell SJ, Zhang R. Anti‐Inflammatory Agents for Cancer Therapy. Mol Cell Pharmacol. 2009;1(1):29‐43. 198. Colombo JR, Wein RO. Cabozantinib for progressive metastatic medullary thyroid cancer: a review. Ther Clin Risk Manag. 2014;10:395‐404. 199. Mesange P, Poindessous V, Sabbah M, Escargueil AE, de Gramont A, Larsen AK. Intrinsic bevacizumab resistance is associated with prolonged activation of autocrine VEGF signaling and hypoxia tolerance in colorectal cancer cells and can be overcome by nintedanib, a small molecule angiokinase inhibitor. Oncotarget. 2014;5(13):4709‐21. 200. Lord CJ, Ashworth A. Mechanisms of resistance to therapies targeting BRCA‐mutant cancers. Nat Med. 2013;19(11):1381‐8. 201. Anderson MA, Huang D, Roberts A. Targeting BCL2 for the treatment of lymphoid malignancies. Semin Hematol. 2014;51(3):219‐27. 202. Ganapathy‐Kanniappan S, Kunjithapatham R, Geschwind JF. Anticancer efficacy of the metabolic blocker 3‐bromopyruvate: specific molecular targeting. Anticancer Res. 2013;33(1):13‐20. 203. Kaelin WG, Jr. Synthetic lethality: a framework for the development of wiser cancer therapeutics. Genome Med. 2009;1(10):99. 204. Kaelin WG, Jr. The concept of synthetic lethality in the context of anticancer therapy. Nat Rev Cancer. 2005;5(9):689‐98. 205. Chan DA, Giaccia AJ. Harnessing synthetic lethal interactions in anticancer drug discovery. Nat Rev Drug Discov. 2011;10(5):351‐64.

206. Gien LT, Mackay HJ. The Emerging Role of PARP Inhibitors in the Treatment of Epithelial Ovarian Cancer. J Oncol. 2010;2010:151750. 207. Audeh MW, Carmichael J, Penson RT, Friedlander M, Powell B, Bell‐McGuinn KM, et al. Oral poly(ADP‐ribose) polymerase inhibitor olaparib in patients with BRCA1 or BRCA2 mutations and recurrent ovarian cancer: a proof‐of‐concept trial. Lancet. 2010;376(9737):245‐51. 208. Tutt A, Robson M, Garber JE, Domchek SM, Audeh MW, Weitzel JN, et al. Oral poly(ADP‐ribose) polymerase inhibitor olaparib in patients with BRCA1 or BRCA2 mutations and advanced breast cancer: a proof‐of‐concept trial. Lancet. 2010;376(9737):235‐44. 209. Bryant HE, Schultz N, Thomas HD, Parker KM, Flower D, Lopez E, et al. Specific killing of BRCA2‐deficient tumours with inhibitors of poly(ADP‐ribose) polymerase. Nature. 2005;434(7035):913‐7. 210. Farmer H, McCabe N, Lord CJ, Tutt AN, Johnson DA, Richardson TB, et al. Targeting the DNA repair defect in BRCA mutant cells as a therapeutic strategy. Nature. 2005;434(7035):917‐21. 211. Lamba S, Russo M, Sun C, Lazzari L, Cancelliere C, Grernrum W, et al. RAF suppression synergizes with MEK inhibition in KRAS mutant cancer cells. Cell Rep. 2014;8(5):1475‐83. 212. Lito P, Saborowski A, Yue J, Solomon M, Joseph E, Gadal S, et al. Disruption of CRAF‐mediated MEK activation is required for effective MEK inhibition in KRAS mutant tumors. Cancer Cell. 2014;25(5):697‐710. 213. Meacham CE, Morrison SJ. Tumour heterogeneity and cancer cell plasticity. Nature. 2013;501(7467):328‐37. 214. Greaves M, Maley CC. Clonal evolution in cancer. Nature. 2012;481(7381):306‐13.

Introduction

49

215. Clevers H. The cancer stem cell: premises, promises and challenges. Nat Med. 2011;17(3):313‐9. 216. Dick JE. Stem cell concepts renew cancer research. Blood. 2008;112(13):4793‐807. 217. Reya T, Morrison SJ, Clarke MF, Weissman IL. Stem cells, cancer, and cancer stem cells. Nature. 2001;414(6859):105‐11. 218. Vermeulen L, Sprick MR, Kemper K, Stassi G, Medema JP. Cancer stem cells‐‐old concepts, new insights. Cell Death Differ. 2008;15(6):947‐58. 219. Medema JP. Cancer stem cells: the challenges ahead. Nat Cell Biol. 2013;15(4):338‐44. 220. Clarke MF, Dick JE, Dirks PB, Eaves CJ, Jamieson CH, Jones DL, et al. Cancer stem cells‐‐perspectives on current status and future directions: AACR Workshop on cancer stem cells. Cancer Res. 2006;66(19):9339‐44. 221. Colak S, Medema JP. Cancer stem cells‐‐important players in tumor therapy resistance. FEBS J. 2014;281(21):4779‐91. 222. Colak S, Zimberlin CD, Fessler E, Hogdal L, Prasetyanti PR, Grandela CM, et al. Decreased mitochondrial priming determines chemoresistance of colon cancer stem cells. Cell Death Differ. 2014;21(7):1170‐7. 223. Risinger AL, Giles FJ, Mooberry SL. Microtubule dynamics as a target in oncology. Cancer Treat Rev. 2009;35(3):255‐61. 224. Johnson IS, Armstrong JG, Gorman M, Burnett JP, Jr. The Vinca Alkaloids: A New Class of Oncolytic Agents. Cancer Res. 1963;23:1390‐427. 225. Chabner BA, Roberts TG, Jr. Timeline: Chemotherapy and the war on cancer. Nat Rev Cancer. 2005;5(1):65‐72.

226. Schiff PB, Fant J, Horwitz SB. Promotion of microtubule assembly in vitro by taxol. Nature. 1979;277(5698):665‐7. 227. McChesney JD, Venkataraman SK, Henri JT. Plant natural products: back to the future or into extinction? Phytochemistry. 2007;68(14):2015‐22. 228. Hortobagyi GN. Anthracyclines in the treatment of cancer. An overview. Drugs. 1997;54 Suppl 4:1‐7. 229. Pasquier E, Kavallaris M. Microtubules: a dynamic target in cancer therapy. IUBMB Life. 2008;60(3):165‐70. 230. Tacar O, Sriamornsak P, Dass CR. Doxorubicin: an update on anticancer molecular action, toxicity and novel drug delivery systems. J Pharm Pharmacol. 2013;65(2):157‐70. 231. Gershenzon J, Dudareva N. The function of terpene natural products in the natural world. Nat Chem Biol. 2007;3(7):408‐14. 232. Connolly JD, Hill RA. Triterpenoids. Nat Prod Rep. 2008;25(4):794‐830. 233. Petronelli A, Pannitteri G, Testa U. Triterpenoids as new promising anticancer drugs. Anticancer Drugs. 2009;20(10):880‐92. 234. Bruckner V, Kovacs J, Koczka I. Occurrence of betulinic acid in the bark of the plane tree. J Chem Soc. 1948;1:948‐51. 235. Fujioka T, Kashiwada Y, Kilkuskie RE, Cosentino LM, Ballas LM, Jiang JB, et al. Anti‐AIDS agents, 11. Betulinic acid and platanic acid as anti‐HIV principles from Syzigium claviflorum, and the anti‐HIV activity of structurally related triterpenoids. J Nat Prod. 1994;57(2):243‐7. 236. Mayaux JF, Bousseau A, Pauwels R, Huet T, Henin Y, Dereu N, et al. Triterpene derivatives that block entry of human immunodeficiency virus type 1 into cells. Proc Natl Acad Sci U S A. 1994;91(9):3564‐8.

Chapter 1

50

237. Li F, Goila‐Gaur R, Salzwedel K, Kilgore NR, Reddick M, Matallana C, et al. PA‐457: a potent HIV inhibitor that disrupts core condensation by targeting a late step in Gag processing. Proc Natl Acad Sci U S A. 2003;100(23):13555‐60. 238. Dang Z, Lai W, Qian K, Ho P, Lee KH, Chen CH, et al. Betulinic acid derivatives as human immunodeficiency virus type 2 (HIV‐2) inhibitors. J Med Chem. 2009;52(23):7887‐91. 239. Smith PF, Ogundele A, Forrest A, Wilton J, Salzwedel K, Doto J, et al. Phase I and II study of the safety, virologic effect, and pharmacokinetics/pharmacodynamics of single‐dose 3‐o‐(3',3'‐dimethylsuccinyl)betulinic acid (bevirimat) against human immunodeficiency virus infection. Antimicrob Agents Chemother. 2007;51(10):3574‐81. 240. Adamson CS, Waki K, Ablan SD, Salzwedel K, Freed EO. Impact of human immunodeficiency virus type 1 resistance to protease inhibitors on evolution of resistance to the maturation inhibitor bevirimat (PA‐457). J Virol. 2009;83(10):4884‐94. 241. Zhou J, Chen CH, Aiken C. Human immunodeficiency virus type 1 resistance to the small molecule maturation inhibitor 3‐O‐(3',3'‐dimethylsuccinyl)‐betulinic acid is conferred by a variety of single amino acid substitutions at the CA‐SP1 cleavage site in Gag. J Virol. 2006;80(24):12095‐101. 242. Verheyen J, Verhofstede C, Knops E, Vandekerckhove L, Fun A, Brunen D, et al. High prevalence of bevirimat resistance mutations in protease inhibitor‐resistant HIV isolates. AIDS. 2010;24(5):669‐73. 243. Seclen E, Gonzalez Mdel M, Corral A, de Mendoza C, Soriano V, Poveda E. High prevalence of natural polymorphisms in Gag (CA‐SP1) associated with reduced response to Bevirimat, an HIV‐1 maturation inhibitor. AIDS. 2010;24(3):467‐9.

244. Dang Z, Qian K, Ho P, Zhu L, Lee KH, Huang L, et al. Synthesis of betulinic acid derivatives as entry inhibitors against HIV‐1 and bevirimat‐resistant HIV‐1 variants. Bioorg Med Chem Lett. 2012;22(16):5190‐4. 245. Qian K, Bori ID, Chen CH, Huang L, Lee KH. Anti‐AIDS agents 90. novel C‐28 modified bevirimat analogues as potent HIV maturation inhibitors. J Med Chem. 2012;55(18):8128‐36. 246. Dang Z, Ho P, Zhu L, Qian K, Lee KH, Huang L, et al. New betulinic acid derivatives for bevirimat‐resistant human immunodeficiency virus type‐1. J Med Chem. 2013;56(5):2029‐37. 247. Gautam R, Jachak SM. Recent developments in anti‐inflammatory natural products. Med Res Rev. 2009;29(5):767‐820. 248. Enwerem NM, Okogun JI, Wambebe CO, Okorie DA, Akah PA. Anthelmintic activity of the stem bark extracts of Berlina grandiflora and one of its active principles, Betulinic acid. Phytomedicine. 2001;8(2):112‐4. 249. Steele JC, Warhurst DC, Kirby GC, Simmonds MS. In vitro and in vivo evaluation of betulinic acid as an antimalarial. Phytother Res. 1999;13(2):115‐9. 250. de Sa MS, Costa JF, Krettli AU, Zalis MG, Maia GL, Sette IM, et al. Antimalarial activity of betulinic acid and derivatives in vitro against Plasmodium falciparum and in vivo in P. berghei‐infected mice. Parasitol Res. 2009;105(1):275‐9. 251. Innocente AM, Silva GN, Cruz LN, Moraes MS, Nakabashi M, Sonnet P, et al. Synthesis and antiplasmodial activity of betulinic acid and ursolic acid analogues. Molecules. 2012;17(10):12003‐14. 252. da Silva GN, Maria NR, Schuck DC, Cruz LN, de Moraes MS, Nakabashi M, et al. Two series of new semisynthetic triterpene derivatives: differences in anti‐malarial activity, cytotoxicity and mechanism of action. Malar J. 2013;12:89.

Introduction

51

253. Dehelean CA, Feflea S, Ganta S, Amiji M. Anti‐angiogenic effects of betulinic acid administered in nanoemulsion formulation using chorioallantoic membrane assay. J Biomed Nanotechnol. 2011;7(2):317‐24. 254. Shin J, Lee HJ, Jung DB, Jung JH, Lee HJ, Lee EO, et al. Suppression of STAT3 and HIF‐1 alpha mediates anti‐angiogenic activity of betulinic acid in hypoxic PC‐3 prostate cancer cells. PLoS One. 2011;6(6):e21492. 255. Yun Y, Han S, Park E, Yim D, Lee S, Lee CK, et al. Immunomodulatory activity of betulinic acid by producing pro‐inflammatory cytokines and activation of macrophages. Arch Pharm Res. 2003;26(12):1087‐95. 256. Saaby L, Jager AK, Moesby L, Hansen EW, Christensen SB. Isolation of immunomodulatory triterpene acids from a standardized rose hip powder (Rosa canina L.). Phytother Res. 2011;25(2):195‐201. 257. Wan Y, Wu YL, Lian LH, Xie WX, Li X, Ouyang BQ, et al. The anti‐fibrotic effect of betulinic acid is mediated through the inhibition of NF‐kappaB nuclear protein translocation. Chem Biol Interact. 2012;195(3):215‐23. 258. Pisha E, Chai H, Lee IS, Chagwedera TE, Farnsworth NR, Cordell GA, et al. Discovery of betulinic acid as a selective inhibitor of human melanoma that functions by induction of apoptosis. Nat Med. 1995;1(10):1046‐51. 259. Fulda S, Friesen C, Los M, Scaffidi C, Mier W, Benedict M, et al. Betulinic acid triggers CD95 (APO‐1/Fas)‐ and p53‐independent apoptosis via activation of caspases in neuroectodermal tumors. Cancer Res. 1997;57(21):4956‐64. 260. Schmidt ML, Kuzmanoff KL, Ling‐Indeck L, Pezzuto JM. Betulinic acid induces apoptosis in human neuroblastoma cell lines. Eur J Cancer. 1997;33(12):2007‐10. 261. Wick W, Grimmel C, Wagenknecht B, Dichgans J, Weller M. Betulinic acid‐induced

apoptosis in glioma cells: A sequential requirement for new protein synthesis, formation of reactive oxygen species, and caspase processing. J Pharmacol Exp Ther. 1999;289(3):1306‐12. 262. Fulda S, Jeremias I, Steiner HH, Pietsch T, Debatin KM. Betulinic acid: a new cytotoxic agent against malignant brain‐tumor cells. IntJ Cancer. 1999;82(3):435‐41. 263. Zuco V, Supino R, Righetti SC, Cleris L, Marchesi E, Gambacorti‐Passerini C, et al. Selective cytotoxicity of betulinic acid on tumor cell lines, but not on normal cells. Cancer Lett. 2002;175(1):17‐25. 264. Thurnher D, Turhani D, Pelzmann M, Wannemacher B, Knerer B, Formanek M, et al. Betulinic acid: a new cytotoxic compound against malignant head and neck cancer cells. Head Neck. 2003;25(9):732‐40. 265. Kessler JH, Mullauer FB, de Roo GM, Medema JP. Broad in vitro efficacy of plant‐derived betulinic acid against cell lines derived from the most prevalent human cancer types. Cancer Lett. 2007;251(1):132‐45. 266. Eichenmuller M, von Schweinitz D, Kappler R. Betulinic acid treatment promotes apoptosis in hepatoblastoma cells. Int J Oncol. 2009;35(4):873‐9. 267. Eichenmuller M, Hemmerlein B, von Schweinitz D, Kappler R. Betulinic acid induces apoptosis and inhibits hedgehog signalling in rhabdomyosarcoma. Br J Cancer. 2010;103(1):43‐51. 268. Liu Y, Luo W. Betulinic acid induces Bax/Bak‐independent cytochrome c release in human nasopharyngeal carcinoma cells. Mol Cells. 2012;33(5):517‐24. 269. Noda Y, Kaiya T, Kohda K, Kawazoe Y. Enhanced cytotoxicity of some triterpenes toward leukemia L1210 cells cultured in low pH media: possibility of a new mode of cell

Chapter 1

52

killing. Chem Pharm Bull (Tokyo). 1997;45(10):1665‐70. 270. Ehrhardt H, Fulda S, Fuhrer M, Debatin KM, Jeremias I. Betulinic acid‐induced apoptosis in leukemia cells. Leukemia. 2004;18(8):1406‐12. 271. Raghuvar Gopal DV, Narkar AA, Badrinath Y, Mishra KP, Joshi DS. Betulinic acid induces apoptosis in human chronic myelogenous leukemia (CML) cell line K‐562 without altering the levels of Bcr‐Abl. Toxicol Lett. 2005;155(3):343‐51. 272. Fulda S, Scaffidi C, Susin SA, Krammer PH, Kroemer G, Peter ME, et al. Activation of mitochondria and release of mitochondrial apoptogenic factors by betulinic acid. J BiolChem. 1998;273(51):33942‐8. 273. Mullauer FB, Kessler JH, Medema JP. Betulinic acid induces cytochrome c release and apoptosis in a Bax/Bak‐independent, permeability transition pore dependent fashion. Apoptosis. 2009;14(2):191‐202. 274. Raisova M, Hossini AM, Eberle J, Riebeling C, Wieder T, Sturm I, et al. The Bax/Bcl‐2 ratio determines the susceptibility of human melanoma cells to CD95/Fas‐mediated apoptosis. J Invest Dermatol. 2001;117(2):333‐40. 275. Gonzalez P, Mader I, Tchoghandjian A, Enzenmuller S, Cristofanon S, Basit F, et al. Impairment of lysosomal integrity by B10, a glycosylated derivative of betulinic acid, leads to lysosomal cell death and converts autophagy into a detrimental process. Cell DeathDiffer. 2012;19(8):1337‐46. 276. Yang LJ, Chen Y, He J, Yi S, Wen L, Zhao J, et al. Betulinic acid inhibits autophagic flux and induces apoptosis in human multiple myeloma cells in vitro. Acta PharmacolSin. 2012;33(12):1542‐8. 277. Selzer E, Pimentel E, Wacheck V, Schlegel W, Pehamberger H, Jansen B, et al. Effects of betulinic acid alone and in combination with

irradiation in human melanoma cells. J Invest Dermatol. 2000;114(5):935‐40. 278. Sawada N, Kataoka K, Kondo K, Arimochi H, Fujino H, Takahashi Y, et al. Betulinic acid augments the inhibitory effects of vincristine on growth and lung metastasis of B16F10 melanoma cells in mice. BrJCancer. 2004;90(8):1672‐8. 279. Fulda S, Debatin KM. Sensitization for anticancer drug‐induced apoptosis by betulinic Acid. Neoplasia. 2005;7(2):162‐70. 280. Jung GR, Kim KJ, Choi CH, Lee TB, Han SI, Han HK, et al. Effect of betulinic acid on anticancer drug‐resistant colon cancer cells. Basic ClinPharmacolToxicol. 2007;101(4):277‐85. 281. Yamai H, Sawada N, Yoshida T, Seike J, Takizawa H, Kenzaki K, et al. Triterpenes augment the inhibitory effects of anticancer drugs on growth of human esophageal carcinoma cells in vitro and suppress experimental metastasis in vivo. IntJCancer. 2009;125(4):952‐60. 282. Parrondo R, de las Pozas A, Reiner T, Rai P, Perez‐Stable C. NF‐kappaB activation enhances cell death by antimitotic drugs in human prostate cancer cells. Mol Cancer. 2010;9:182. 283. Bache M, Zschornak MP, Passin S, Kessler J, Wichmann H, Kappler M, et al. Increased betulinic acid induced cytotoxicity and radiosensitivity in glioma cells under hypoxic conditions. Radiat Oncol. 2011;6:111. 284. Gao Y, Jia Z, Kong X, Li Q, Chang DZ, Wei D, et al. Combining betulinic acid and mithramycin a effectively suppresses pancreatic cancer by inhibiting proliferation, invasion, and angiogenesis. Cancer Res. 2011;71(15):5182‐93. 285. Damle AA, Pawar YP, Narkar AA. Anticancer activity of betulinic acid on MCF‐7 tumors in nude mice. Indian J Exp Biol. 2013;51(7):485‐91.

Introduction

53

286. Chintharlapalli S, Papineni S, Ramaiah SK, Safe S. Betulinic acid inhibits prostate cancer growth through inhibition of specificity protein transcription factors. Cancer Res. 2007;67(6):2816‐23. 287. Rajendran P, Jaggi M, Singh MK, Mukherjee R, Burman AC. Pharmacological evaluation of C‐3 modified Betulinic acid derivatives with potent anticancer activity. Invest New Drugs. 2008;26(1):25‐34. 288. Mullauer FB, van BL, Daalhuisen JB, Ten Brink MS, Storm G, Medema JP, et al. Betulinic acid delivered in liposomes reduces growth of human lung and colon cancers in mice without causing systemic toxicity. Anticancer Drugs. 2011;22(3):223‐33.

Chapter 2

Betulinic acid‐induced mitochrondria‐

dependent cell death is counter

balanced by an autophagic salvage

response

Lisette Potze, Franziska B. Mullauer, Selcuk Colak, Jan H. Kessler,

Jan Paul Medema

Cell Death and Disease 2014

"Tut, tut, child!" said the Duchess. "Every thing's got a moral, if only you can find it."


Chapter 2

56

Abstract

Betulinic acid (BetA) is a plant‐derived pentacyclic triterpenoid that exerts potent anti‐

cancer effects in vitro and in vivo. It was shown to induce apoptosis via a direct effect on

mitochondria. This is largely independent of proapoptotic BAK and BAX, but can be inhibited

by cyclosporin A (CsA), an inhibitor of the permeability transition (PT) pore. Here we show

that blocking apoptosis with general caspase inhibitors did not prevent cell death, indicating

that alternative, caspase‐independent cell death pathways were activated. BetA did not

induce necroptosis, but we observed a strong induction of autophagy in several cancer cell

lines. Autophagy was functional as shown by enhanced flux and degradation of long‐lived

proteins. BetA‐induced autophagy could be blocked, just like apoptosis, with CsA,

suggesting that autophagy is activated as a response to the mitochondrial damage inflicted

by BetA. As both a survival and cell death role have been attributed to autophagy,

autophagy‐deficient tumor cells and mouse embryo fibroblasts were analyzed to determine

the role of autophagy in BetA‐induced cell death. This clearly established BetA‐induced

autophagy as a survival mechanism and indicates that BetA utilizes an as yet‐undefined

mechanism to kill cancer cells.

Betulinic acid induces a rescue autophagy response

57

Introduction

Betulinic acid (BetA) is a naturally occurring triterpenoid with potent cytotoxic effects on

cancer cells.1–4 It was initially proposed to have a direct effect on the mitochondria and to

induce apoptosis in a BCL‐2‐dependent fashion.5–7 However, BCL‐2 overexpression can only

provide short‐term protection and eventually these cells do succumb to apoptosis. In

addition, BAK/BAX double deficiency does not protect against BetA‐induced cell death,8

indicating that cell death ensues independently of the BCL‐2 family. Mitochondrial outer

membrane permeabilization (MOMP), however, is crucial as cyclosporin A (CsA) and

bongkrekic acid, both inhibitors of the permeability transition pore (PT‐pore), and as a

consequence MOMP were able to protect BetA‐treated cells from releasing cytochrome c

and subsequent apoptosis.7,8 Downstream of the mitochondria caspases are activated, but

direct inhibition of caspase activity with the pan‐caspase inhibitor zVAD.fmk (N‐

benzyloxycarbonyl‐Val‐Ala‐Asp‐fluoromethylketone) did not offer protection to BetA‐

induced cell death.4 Combined this indicates that BetA does induce apoptosis, but that

additional mechanisms must exist by which cell death is induced by the compound.

Alternative cell death pathways include necrosis, necroptosis, lysosomal membrane

permeabilization (LMP) and autophagy. Necrosis is a form of passive cell death that is

induced upon strong insults such as mechanical injury of cells. Typical features include

swelling, rupture of organelle membranes as well as the outer cell membrane and as a result

the cell contents are released, often causing inflammation in vivo.9 Necroptosis, a highly

regulated form of necrosis, can be induced by ligation of death receptors in the presence of

caspase inhibitors.10 Necroptosis is regulated by distinct complexes called the necrosome

or ripoptosome that include receptor interacting protein (RIP) kinases, FAS‐associated

death domain and caspase‐8.10,11 Necroptosis can be blocked by small molecules, such as

necrostatin‐1, that allosterically block the kinase activity of RIP1.12,13

Lysosomal cell death is induced by a destabilization of the lysosomal membrane.14 The cell

death depends on the leakage of lysosomal components into the cytoplasm, and the activity

of cathepsins that can induce a necrotic‐like or apoptotic cell death depending on the extent

of the leakage. As such the induction of cell death is clearly distinct from necrosis and

apoptosis; nevertheless, the execution phase is similar.

Autophagy is a highly conserved cellular proteolytic degradation process in which cytosolic

components are first encapsulated by a membrane and subsequently degraded in the

lysosome, thereby providing new building blocks for the cell. Autophagy is typically induced

upon nutrient starvation. However, autophagy also involves sequestration of damaged

cytoplasmic components and organelles, via induction of de novo double‐membrane

vesicles (autophagosomes) that surround the cargo. The autophagosome transports the

Chapter 2

58

cargo to the lysosome where fusion of the autophagosome and lysosome leads to

degradation of the cargo (autophagic flux).15 Autophagy, even though initially regarded as

a cell survival mechanism, has also been suggested to serve a role in inducing cell death. It

may therefore represent a balancing mechanism between cell survival and cell death.16–18

Autophagy can target cytosolic components specific for degradation, known as specific

autophagy. Specific autophagy can include ubiquitinated proteins, peroxisomes and

mitochondria.19–21 Mitophagy involves the selective degradation of mitochondria and is

among others used for clearance of damaged mitochondria.17,22 Previously, BetA and a

derivative of BetA, B10, have been shown to induce autophagosome formation in multiple

myeloma cells and glioblastoma cells.23,24 In these studies it was suggested that the

autophagic flux was prevented, leading to an accumulation of undigested autophagosomes.

Although both studies observed autophagy, the role of autophagy as a cell death

mechanism was not addressed for BetA.23,24

We hypothesized that autophagy/mitophagy is induced upon BetA treatment to clear the

damaged mitochondria. We show that autophagy is massively induced in various BetA‐

treated tumor cells, but is prevented by CsA, suggesting that autophagy occurs downstream

of the BetA‐induced mitochondrial damage. With the use of knockout and knockdown

studies of key regulators of the autophagy pathway, we demonstrate that autophagy serves

as a rescue pathway and is not responsible for the cell death induced by BetA.

Results

BetA‐induced cell death is independent of apoptosis.

BetA induced a very potent form of cell death in HeLa cells that displayed features like cell

membrane rupture (Figure 1a), apoptosis (Figures 1b and c) and mitochondrial

depolarization (Figures 1d and e). The effect of BetA was, as shown before, concentration

dependent, starting at 7.5 µg/ml and reaching a plateau at 10 µg/ml (Supplementary Figure

1). Previously we have shown, in Jurkat cells, that BetA‐induced apoptosis is blocked by co‐

treatment with zVAD.fmk, but that this pan‐caspase inhibitor does not prevent cell death

as measured by propidium iodide (PI) exclusion.4 Similarly, cell death induced by BetA in

HeLa cells was only partly blocked with a more potent caspase inhibitor Q‐VD‐OPh (Figure

1f), which did block apoptotic features like DNA fragmentation (results not shown). This

indicates that caspase inhibition was effective, but not sufficient to block the induction of

cell death. Thus, while apoptosis is induced by BetA, these findings indicate that alternative

cell death pathways are activated as well. To study whether necroptosis serves as an

alternative cell death mechanism induced by BetA, we made use of necrostatin, a specific

inhibitor of this cell death mechanism. Either necrostatin alone or in combination with

caspase inhibition failed to prevent cell death (Figure 1f). As a control for the action of


59

necrostatin, U937 cells were used, which have been reported to undergo necroptosis upon

tumor necrosis factor‐α (TNF‐ α) treatment12 in the presence caspase inhibitors. In

accordance with literature12 this indeed led to a significant induction necroptosis, which was

prevented by necrostatin (Supplementary Figure 2). As this combination did not inhibit

BetA‐induced cell death, our data point to a cell death mechanism that is independent of

caspases and necroptosis.

Figure 1. BetA induces cell death independent of apoptosis and necroptosis. HeLa cells were subjected to

10 µg/ml BetA for 48 h after which cell death by PI exclusion (a), DNA fragmentation (b), phosphatidylserine

(PS) exposure (c), and loss of mitochondrial activity by mitotracker orange (d) or JC‐1 (e) were assessed (f)

HeLa cells were pre‐treated for 1 h with or without 20 µM Q‐VD‐Oph and with or without 25 µM necrostatin‐

1 and subjected to 10 µg/ml BetA for 48 h after which cell death was assessed via PI exclusion. Mean ± S.D.

of three independent experiments are shown, *P< 0.05; **P< 0.01; ***P<0.001; NS, not significant.

BetA induces autophagy

To determine whether autophagy is responsible for the cell death observed, LC3

conjugation was studied. Conjugated LC3 (LC3‐II) is associated with autophagic membranes

and therefore fusion of LC3 to green fluorescent protein (GFP) can be used to detect

autophagosomes.25 In untreated cells, LC3–GFP is evenly dispersed in the cells, with a slight

preference for the nucleus, whereas in cells where autophagy is induced, LC3 translocation

is evident (Figure 2a).25 Both HeLa\LC3–GFP and MCF‐7\LC3–GFP cells treated with the

classical autophagy inducer rapamycin displayed a clear induction of punctuated LC3–GFP

structures. Interestingly, BetA was even more potent in inducing autophagy, showing

ba c d e

f

BetAQ-VD-OPh

--

-+

+- -

--+ +-

necrostatin - - - -+ ++

++

+++

***** *** *** ***

**

*ns

% P

Ipo

s

%A

nne

xin

Vp

os

% D

epo

lariz

atio

n

% M

itotr

acke

rn

eg

0

20

40

60

80

BetADMSO BetADMSO BetADMSO BetADMSO0

20

40

60

80

0

20

40

60

80

0

20

40

60

80

BetADMSO0

20

40

60

80

0

20

40

60

80

% D

NA

fra

gm

enta

tion

% P

Ipo

s

Chapter 2

60

enlarged punctae (Figure 2a). To validate that BetA indeed induced autophagy, a separate

measure to monitor the induction of autophagy was used, namely the processing of LC3‐I

to its PE‐conjugated LC3‐II using immunoblotting.26 Rapamycin treatment resulted in

enhanced LC3‐II levels in HeLa\LC3–GFP, A549, MCF‐7\LC3–GFP and SW480\LC3‐GFP cells

as compared with their DMSO‐treated counterparts (Figure 2b). Analyzing multiple

independent assays consistently revealed this induction, but due to the variation in

background levels of LC3‐II it only reached significance in HeLa and A549 cells (Figure 2c). In

contrast, BetA‐induced LC3‐II levels are significantly elevated in all cell lines and were clearly

more pronounced as compared with rapamycin (Figures 2b and c), similar to the

observations with LC3–GFP staining. The effect of BetA on autophagy occurs relatively rapid

upon BetA treatment and required concentrations that are also exerting a toxic effect (7.5–

10 µg/ml) (Supplementary Figures 3a and b).

Both detection of LC3‐II via immunoblotting and LC3–GFP translocation are measures for

the amount of autophagosomes present at a certain time point, but do not provide

information about the cause of this phenotype. It is possible that an increased number of

autophagosomes is caused by an inhibition of basal autophagic flux rather than by induction

of autophagy.27,28 Previously, it was reported that BetA in KM3 cells dose‐dependently

induces the expression of LC3‐II, but also of p62, a protein that is normally degraded during

autophagy. This study suggested that BetA rather inhibited autophagic flux instead of

inducing autophagy.24 To analyze the effect of BetA on p62 we studied its expression in HeLa

cells. This confirmed a strong induction of p62 upon BetA treatment (Supplementary Figure

3c). However, this is due to induction of de novo synthesis and could be blocked with the

addition of cycloheximide. The combination of cycloheximide and BetA did reveal

degradation of p62, pointing to a functional flux (Supplementary Figure 3c). To better

evaluate this autophagic flux, degradation of long‐lived proteins was measured, which

reportedly are at least in part degraded by autophagy.27–29 In agreement, rapamycin clearly

induced degradation of these long‐lived proteins after 14 h (Figure 2d). Moreover,

degradation continued after this time point as was evident from analyzing the release of

labeled amino acids in the time period between 14 and 20 h (Figure 2e). In contrast to the

finding of Yang et al.24 That BetA prevents autophagic flux, BetA clearly increased the

degradation of long‐lived proteins after 14 h and the following 6 h after medium change.

This was also more pronounced as compared to rapamycin‐induced degradation (Figures

2d and e). Because degradation of long‐lived proteins is not solely specific for autophagy

and cannot discriminate between proteasomal‐ and autophagosomal degradation more

specific assays using tandem RFP/eGFP‐tagged LC3 as well as a tandem mCherry/GFP‐

tagged p62 were used.30 As RFP and mCherry are pH stable, they remain fluorescent after

fusion of the autophagosomes with the lysosomal compartment, while eGFP fluorescence

is lost. This allows detection of autophagic flux by simply observing the formation of red

fluorescent lysosomes from green/red fluorescent autophagosomes. Using either LC3


61

(Figure 2f) or p62 (Supplementary Figure 3d) we observed that rapamycin as well as BetA

induced a functional autophagic flux using. This confirms that BetA is a potent inducer of

autophagy resulting in an enhanced autophagic flux.

Figure 2. BetA induces autophagy (a) HeLa\LC3‐GFP cells and MCF‐7\LC3‐GFP cells were treated with 10

µg/ml BetA or 10 µM rapamycin and after 18 h cells were analyzed by confocal microscopy. Quantification

of LC3 punta was perfomed using Image J. Mean of positive pixels/total pixel cell (%) ± S.D. of three fields of

view per sample are shown. (b) HeLa\LC3‐GFP, A549, MCF‐7\LC3‐GFP and SW480\LC3‐GFP were treated

with 10 µg/ml BetA or 10 µM rapamycin for 18 h. LC3 processing was assessed via immunoblotting. Results

for endogenous LC3 are shown. LC3‐I (18 kDa) and LC3‐II (16 kDa). Tubulin is used as a control. (c) Band

intensity of western blot was quantified using Image J software. DMSO control is set to 1. Mean ± S.D. of

three indendent experiments are shown, *P < 0.05; **P< 0.01; ***P<0.001; n.s., not significant. (d) MCF7

cells were treated with the indicated concentrations of BetA or 1 µM rapamycin and after 14 h degradation

of long‐lived proteins was measured or (e) after 14 h medium was refreshed and degradation in the following

6 h was measured. Mean ± S.D. of triplicates are shown (f) HeLa and A549 cells with tandem fluorescent LC3

(mRFP‐eGFP‐LC3) were treated with 10 µg/ml BetA or 10 µM rapamycin and after 18 h cells were analyzed

by confocal microscopy. Quantification of red LC3 puncta was performed using Image J. Mean of positive

pixels/total pixel cell (%) ± S.D. of three fields of view per sample are shown.

DMSO rapamycinBetA

MC

F-7

\LC

3-G

FP

HeL

a\LC

3-G

FP

5.795 ±0.86

10.36 ±1.8 3.707 ±0.61

DMSO rapamycinBetA

1.452 ±2.1

a

d

0.5

1.0

1.5

2.0

Degra

dation (

ratio)

0 5 7.5 10 1µM

BetA (µg/ml) Rapa

0.5

1.0

1.5

2.0

0 5 7.5 10 1µMBetA (µg/ml) Rapa

De

gra

da

tio

n (

ratio

)

e

LC

3 I

I /

LC

3 I

0

2

4

6 ***

**

**ns

*ns

A549HeLa SW480MCF-7

0.042 ±0.04 8.415 ±2.8

0.16 ±0.2

c

He

La

0 µm 10 0 µm 10 0 µm 10

f

A5

49

0 µm 25 0 µm 25 0 µm 25

BetArapamycin

+- - - - - - - -+ + ++ + + + +-------

2.539 ±2.7 10.99 ±9.0 5.218 ±4.4

8.36 ±6.9 6.637 ±3.1

b

LC3 I

LC3 II

Tubulin

A549HeLa SW480MCF-7

BetArapamycin

+- - - - - - - -+ + ++ + + + +------

Chapter 2

62

Cyclosporin A blocks BetA‐induced autophagy.

Permeability transition pore (PT‐pore) opening results in membrane depolarization and

leads to cytochrome c release and subsequent caspase activation.31 The proposed structure

of the pore is formed by a voltage‐dependent anion channel, adenine nucleotide

translocator and cyclophilin D complex,32 and opening of the PT‐pore can be blocked by

inhibition of cyclophilin D using CsA.33 Previously we have shown that BetA‐induced

apoptosis and cytochrome c release proceeded in a PT‐pore‐dependent fashion.8 As CsA

inhibits BetA‐induced apoptotic features,8 the effect of CsA on BetA‐induced autophagy was

investigated. HeLa\LC3–GFP cells were treated with BetA alone or in combination with CsA

and analyzed via confocal microscopy. Autophagosome formation was clearly inhibited in

the presence of CsA (Figure 3a). A similar CsA dependency was observed in MCF‐7\LC3 cells

(Figure 3a). The effect of CsA was confirmed by immunoblotting for LC3 in BetA‐treated

HeLa, but also in other cancer lines (A549, MCF‐7 and SW480). In all tested cell lines, BetA‐

induced formation of lipidated LC3‐II was reduced when cells were pretreated with CsA

(Figures 3b and c). These data suggest that BetA‐induced autophagy is a consequence of

mitochondrial damage triggered by BetA and can be prevented by inhibition of PT‐pore

opening.

Autophagy serves as a rescue pathway, not as an alternative cell death pathway

in BetA‐treated cells.

Autophagy, even though initially regarded as a cell survival pathway, can play a role in dying

cells as well and has been suggested to serve as a balancing mechanism between cell

survival and cell death.16,17 Autophagy is detectable already at low BetA concentrations (7.5

µg/ml) and we reasoned that it is primarily induced as a rescue mechanism. However, it is

feasible that at higher BetA concentrations autophagy is induced beyond a certain

threshold, which shifts the balance from cell survival to cell death.17,34 To test this

hypothesis, we used cell lines with a deficiency in ATG5 or ATG7, which are crucial regulators

in the induction of autophagy.35,36 First we used retroviral shRNA against ATG5 in HeLa and

HeLa\LC3–GFP cells. ATG5 knockdown levels of 77 and 86% were obtained with shRNA

against ATG5 in HeLa and HeLa\LC3–GFP cells, respectively (Supplementary Figure 4a). The

abrogation of autophagy in these cells was confirmed by confocal microscopy (Figure 4a).

Importantly, when analyzing BetA‐induced cell death in HeLa ATG5 knockdown cells, this

was found, if anything, to be enhanced as compared to the cell death induced in control

knockdown cells (Figure 4b). The increase in BetA‐induced cell death in autophagy‐impaired

cells was also observed in MCF‐7\LC3–GFP cells (Supplementary Figures 4a and b). To

formally confirm that impaired autophagy enhances BetA‐induced cell death, MEFs derived

from Atg5 or Atg7 knockout mice were used. While these cells are completely autophagy

deficient, they showed increased levels of BetA‐induced cell death (Figure 4c) and Annexin


63

V exposure (Supplementary Figure 4c). Interestingly, caspase inhibition in these autophagy‐

deficient cells did not prevent cell death either (Figure 4d). However, both autophagy‐

proficient and –deficient cells displayed a similar increase in mitochondrial depolarization

(Figure 4e), suggesting that the initial mitochondrial insults are independent of the

induction of autophagy. Combined these data indicate that autophagy serves primarily as a

survival mechanism in BetA‐treated cells and that BetA‐induced cell death is independent

of autophagy, necroptosis and caspases and follows an as yet‐undefined pathway.

Figure 3. CsA inhibits BetA‐induced autophagy (a) HeLa\LC3‐GFP cells and MCF‐7\LC3‐GFP cells were pre‐

treated with 10 µg/ml CsA for one hour and subjected to 10 µg/ml BetA for 18 h and analyzed by confocal

microscopy. Quantification of LC3 puncta was performed using Image J. Mean of positive pixels/total pixel

cell (%) + S.D. of three fields of view per sample are shown. (b) HeLa\LC3‐GFP, A549, MCF‐7\LC3‐GFP and

SW480\LC3‐GFP were pre‐treated with 10 µg/ml CsA for one hour and subjected to 10 µg/ml BetA for 18 h.

LC3 processing was assessed via immunoblotting. Results for endogenous LC3 are shown; LC3‐I (18 kDa) and

LC3‐II (16 kDa). Tubulin is used as a control. (c) Band intensity of LC3 western blot was quantified using

Image J software. DMSO control is set to 1. Mean + S.D. of three independent experiments are shown, *P <

0.05; **P< 0.01; ***P<0.001; N.S., not significant.

a

b

LC3 ILC3 II

A549HeLa SW480MCF-7

BetACsA

+- - - - - - - -+ + + + + + ++ + + + + + + +- - - - - - - -

LC3

II / L

C3

I

Tubulin

0.0

2.5

5.0

7.5

10.0

A549HeLa SW480MCF-7

BetACsA

+- - - - - - - -+ + + + + + ++ + + + + + + +- - - - - - - -

c

DMSO BetA CsA BetA + CsA

MC

F-7

\LC

3-G

FP

He

La\L

C3

-GF

P

*** * * ns

1.09 ±1.1 6.152 ±0.48 0.664 ±0.53 1.192 ±1.5

1.152 ±0.99 4.611 ±1.39 0.869 ±1.1 1.56 ±0.69

Chapter 2

64

Figure 4. Autophagy serves as a rescue pathway (a) HeLa\LC3‐GFP cells with pSuper empty (control) or

ATG5 #2 knockdown were treated with 10 µM rapamycin for 6 h and analyzed with confocal microscopy.

Representative pictures of two experiments are shown. Quantification of LC3 puncta was performed using

Image J software. Mean of positive pixels/total pixel cell (%) ± S.D. of three fields of view per sample are

shown. (b) HeLa cells with ATG5 #2 knockdown and the control were subjected to different concentrations

of BetA for 48 h after which cell death was measured via PI exclusion and DNA fragmentation was assessed.

Mean ± S.D. of triplicates are shown. (c) Mouse embryonic fibroblasts lacking either ATG5 or ATG7 and their

respective control cells were subjected to different concentrations of BetA and cell death (24 h) using (c) PI

exclusion was measured. (d) Mouse embryonic fibroblasts lacking either ATG5 or ATG7 and their respective

control cells were subjected to different concentrations of BetA for 24 h in the presence or absence of 20

µM zVAD.fmk and cell death using PI exclusion was measured. Mean ± S.D. of three independent

experiments are shown. (e) Loss of mitochondrial activity measured by JC‐1 after 18 h was assessed. Mean

± S.D. of three independent experiments are shown.

Discussion

BetA is a promising anti‐cancer agent with apoptosis‐inducing effects that acts on the PT‐

pore in a BAX/BAK‐independent fashion.8 Here we show that BetA highly efficiently induced

autophagy in several cancer cells. Autophagic flux was found to be functional and led to

a

c

DMSOco

ntro

lsh

RN

AA

TG

5 2

rapamycin

d

0

10

20

30

40

50

% D

NA

frag

men

tatio

n0 5 10 15

controlsh RNAATG5 #2

0

20

40

60

80

100

% P

Ipo

s

0 5 10 15

bcontrolsh RNAATG5 #2

e

0.088 ±0.07 0.2178 ±0.18

1.245 ±1.02 0.339 ±0.5

+ ZVAD-fmk

+ ZVAD-fmk0 5 10 15 20

0

20

40

60

80

100

% P

Ipo

s

0 5 10 15 200

20

40

60

80

100

% P

Ipo

s

BetA (µg/ml)

wtAtg5 -/-

BetA (µg/ml)

wtAtg5 -/-

Atg7 -/-Atg7 +/-

0 5 10 15 200

20

40

60

80

100

% P

Ipo

s

BetA (µg/ml)0 5 10 15 20

0

20

40

60

80

100

% P

Ipo

s

Atg7 +/-Atg7 -/-

wtAtg5 -/-

Atg7 +/-Atg7 -/-

0

20

40

60

80

100

% D

epol

ariz

atio

n0 5 10 15 20

0

20

40

60

80

100%

Dep

olar

izat

ion

0 5 10 15 20


65

enhanced degradation of long‐lived proteins and degradation of GFP in tandem fluorescent

LC3 and p62 proteins (Figure 2 and Supplementary Figure 3). Our observations are in

contrast to previous findings by Yang et al.24 reporting that BetA inhibits autophagic flux in

human multiple myeloma cells as measured by accumulation of p62 protein. Similar to BetA,

B10, a semi‐synthetic glycosylated derivative of BetA, induced cell death in both an

apoptosis‐dependent and apoptosis‐independent fashion. In response to B10 treatment,

LMP and subsequent release of lysosomal enzymes was observed in a human glioblastoma

cell line, impairing autophagy in later stages.23 LMP results in the release of lysosomal

hydrolases and cathepsins into the cytosol.37 Massive lysosomal leakage leads to

uncontrolled necrosis, whereas minor LMP can activate either the intrinsic caspase‐

dependent apoptosis pathway or caspase‐independent alternative cell death programs.38–

43 LMP could therefore also be a possible mechanism by which BetA induces cell death.

However, in contrast to B10, which disrupts the autophagic flux, our data show that

autophagy is enhanced after BetA treatment. It could well be that at later time points the

flux is halted, potentially due to an overload of cargo. This could explain the relatively large

LC3–GFP structures that are observed at later time points. Nevertheless, we show that the

initial enhanced autophagic flux also serves a protective function as it delays the execution

of cell death, while it does not appear to limit the mitochondrial depolarization. Blocking

autophagy completely is detrimental to the cells confirming that the observed increase in

autophagosomes and LC3‐II is not simply due to a block in the flux.

An interesting connection between BetA‐induced apoptosis and autophagy is the fact that

both are efficiently inhibited by CsA, an inhibitor of the PT‐pore8 (Figure 3) CsA has been

reported to block mitophagy induced by loss of MOMP in rat hepatocytes.44,45 Blocking of

apoptosis and autophagy by CsA hints to a scenario in which both pathways are triggered

by a common upstream event that is related to PT‐pore opening. How this is achieved by

BetA remains yet to be established. It is consistent though with the observations in

autophagy‐deficient cells, which do show a decrease in cell death, but not in mitochondrial

defects. This suggests that the primary insult is independent of autophagy, but that the

execution of cell death is dampened by an initial salvage response through the induction of

autophagy. The fact that autophagy occurs relatively rapid upon BetA treatment is also in

line with this notion.

Even though our results suggest that autophagy does not primarily serve as a cell death

mechanism, it is conceivable that it plays an indirect role in the toxicity exerted by BetA. At

higher concentration of BetA, autophagy may keep the balance up for a period of time, after

which either the mitochondrial damage is too massive to be counteracted by autophagy or

that lysosomal exhaustion occurs, which as a consequence will prevent further autophagic

flux. This may then result in release of cytochrome c, activation of caspases and subsequent

apoptosis. This is supported by the observations that BetA‐induced cell death is higher in

Chapter 2

66

autophagy‐deficient cells compared to control cells (Figure 4 and Supplementary Figure 4).

Nevertheless, even under these conditions caspase inactivation is not sufficient to rescue

the cells either, suggesting that the mitochondrial damage inflicted is not compatible with

cellular survival.

Taken together our findings show that BetA‐induced cell death is independent of caspases,

necroptosis and autophagy and, thus, alternative cell death mechanisms must be involved.

This potentially explains the high tumoricidal efficacy of BetA, as it induces cell death that

is not prevented by classical antiapoptotic mechanisms.

Materials and Methods

Chemicals/antibodies.

Betulinic acid (BioSolution Halle, Halle, Germany, (>99% purity) was dissolved in DMSO at 4

mg/ml and stored at ‐80 °C. PI and CsA were purchased from Sigma‐Aldrich (St. Louis, MO,

USA). Rapamycin was purchased from VWR (Amsterdam, The Netherlands) and C‐14 valine

(CFB75‐50 µCi) from Amersham Biosciences (Amersham, UK). Q‐VD‐OPh was purchased

from R&D Systems (Minneapolis, MN, USA). Necrostatin‐1 and TNF‐α were obtained from

Enzo Life Sciences (Farmingdale, NY, USA). Anti‐LC3 antibody (51520) was obtained from

Abcam (Cambridge, UK). Anti‐ATG5 antibody and anti‐ SQSTM1/p62 were from Cell

Signaling Technology (Danvers, MA, USA). Anti α Tubulin antibody was from Santa Cruz

Biotechnology (Dallas, TX, USA). Annexin V‐APC and 7‐AAD were from BD Biosciences

(Franklin Lakes, NJ, USA). Mitotracker Orange CMTMRos and JC‐1 were obtained from Life

Technologies (Carlsbad, CA, USA).

Cell culture.

HeLa cells, A549 cells, MCF‐7 cells, SW480 and U937 cells were obtained from the ATCC

(Manassas, VA, USA). MCF‐7 cells overexpressing LC3– GFP were generated as described.46

HeLa cells and SW480 cells overexpressing LC3‐GFP were made by stable transfection of

plasmid pEGFP‐C1‐LC3, which was a gift from G. Kroemer (Institut Gustave Roussy, Villejuif,

France). All LC3–GFP overexpressing cell lines were cultured with 500 µg/ml G418 in order

to select for LC3–GFP. ATG5 knockout and the corresponding control MEFs were kindly

provided by N Mizushima (Department of Cell Biology, Okazaki, Japan)25 and the ATG7

knockout and corresponding control MEFs were obtained from M Komatsu (Department of

Molecular Oncology, Tokyo, Japan).35

All cells were cultured in IMDM supplemented with 8% FCS, 2mM L‐glutamine, 40 U/ml

penicillin and 40 µg/ml streptomycin.

RNA interference.

Retroviral vectors pSuper ATG5 #2, pSuper ATG5 #3 and pSuper empty, a gift from S.W. Tait

(Beatson Institute, Institute of Cancer Sciences, Glasgow, UK),47 were transfected into


67

Phoenix Amphothrophic cells using the standard Ca²+‐phosphate procedure. After 48 h, the

virus containing medium was centrifuged at 1200 r.p.m. to purify virus from cell debris and

supernatant was supplemented with 10 µg/ml polybrene. Target cells were plated at 70%

confluence and allowed to attach overnight in standard medium. For infections, the culture

medium was replaced with the virus containing medium and incubated for 8 h at 37 °C.

Transduced cell populations were subsequently selected with 1–2 µg/ml puromycin,

depending on cell type (2 µg/ml for HeLa cells and 1 µg/ml for MCF‐7 cells). After 1 week of

selection, expression of the targeted proteins was determined by western blot.

Cell death analysis.

Cell death was determined by PI exclusion. In short, 50 000 cells were plated in a 24‐well

plate and allowed to attach overnight. Cells were treated for 48 h and harvested. The cell

pellet was resuspended in 200 µl medium and stained with PI at 1 µg/ml just before

measuring by flow cytometry. Samples were analyzed using FlowJo software (Tree Star Inc.,

Ashland, OR, USA). For quantification of apoptotic DNA fragmentation (Nicoletti assay) cells

were resuspended in Nicoletti buffer48 supplemented with 50 µg/ml PI for 12 h, subsequent

flow cytometric measurement of PI‐stained nuclei was performed. Phosphatidylserine

exposure during apoptosis was detected by annexin V assay. In short, 50 000 cells were

plated in a 24‐well plate and allowed to attach overnight. Cells were treated for 48 h and

harvested. The cell pellet was stained with annexin V‐APC and 7‐AAD for 15 min at RT,

subsequently flow cytometric measurement of annexin V and 7‐AAD was performed.

Mitochondrial membrane potential analysis.

Mitochondrial activity was measured with Mitotracker Orange CMTMRos. Treated cells

were incubated for 30 min with 25 nM Mitotracker Orange at 37 °C. After staining, cells

were washed and 7‐AAD was added followed by cytometric analysis.

As a second mitochondrial membrane potential assay, JC‐1 staining was used. Cells were

stained for 30 min at RT with 4 µM JC‐1. Depolarization was measured in FITC and PE

channel by flow cytometry.

LC3 fluorescence microscopy.

HeLa and MCF‐7 cells overexpressing LC3–GFP were cultured on Poly‐D‐lysine‐coated 24

mm diameter glass coverslips in six‐well plates and treated with BetA, rapamycin and CsA

for indicated time points and concentrations. The glass coverslips were mounted into a

chamber of the microscope for live‐cell imaging and Z‐stack measurements and placed in a

Leica DMI 6000 (TCS SP8) microscope (with adaptive focus, Motorized XY‐Stage and SuperZ

Galvo) and a case incubator at 37 °C. Samples were analyzed using Leica Las AF software.

Quantification was performed with Image J49 generating a cut‐off for the basal dispersed

LC3 fluorescence, allowing for the quantification of fluorescence in the concentrated areas

(LC3–GFP‐associated autophagosomes).

Chapter 2

68

Tandem fluorescence microscopy.

HeLa and A549 cells were transiently transfected with tandem fluorescent LC3 (mRFP–

eGFP–LC3, kindly provided by T Yoshimori30 using polyethylenimine (PEI) transfection

(Brunschwig, The Netherlands). 24 h after transfection cells were treated with 10 µg/ml

BetA or 10 µM rapamycin for 18 h. HeLa cells were transiently transfected with tandem

fluoresecent p62 (pDest‐mCherry‐EGFP‐p62, kindly provided by E Reits, AMC, The

Netherlands), using PEI transfection. Twenty‐four hours after transfection, cells were

treated with 10 µg/ml BetA or 10 µM rapamycin for 12 h.

Degradation of long‐lived proteins.

MCF‐7 cells were seeded in 12‐well plates. The next day cells were labeled with C‐14 valine

(0.2 µCi/ml). After 24 h, a cold chase using cell culture medium only was performed before

adding BetA or rapamycin for various time points. Supernatants and cells were collected

separately and both were precipitated for 30 min on ice in 10% TCA (trichloroacetic acid).

Precipitates were collected via centrifugation for 10 min at 10 000 r.p.m. and subsequently

dissolved in 0.5M NaOH. Radioactivity in supernatant and cell samples was measured and

the ratio determined. The relative degradation ratio in the control cells was set to ‘one’ and

compared with the degradation ratios in BetA and rapamycin‐treated samples.

Western blot analysis.

Cells were lysed in 1x RIPA lysis buffer (Thermo Fisher Scientific, Waltham, MA, USA)

containing complete proteinase inhibitor (Roche, Penzberg, Germany) and subjected to

protein quantification using a BCA kit (Pierce, Thermo Fisher Scientific, Waltham, MA, USA).

A quantity of 25 µg protein per lane was applied for SDS‐PAGE. Subsequent blotting was

performed using a PVDF membrane (Amersham Biosciences, Amersham, UK). Membranes

were blocked in 5% milk (blocking buffer) in TRIS phosphate‐buffered saline solution

containing TWEEN 20 (0.2%) (TBST‐T) for 1 h at room temperature. Primary antibody

incubations were performed in 1% milk/TBS‐T (LC3 antibody, tubulin antibody) or 5%

BSA/TBS‐T (ATG5 antibody, p62 antibody and ERK1/2 antibody) overnight at 4°C.

Membranes were washed three times and incubated in blocking buffer for 1 h with a

secondary HRP (horseradish peroxidase)‐labeled antibody (anti‐rabbit IgG (H+L) or anti‐

mouse IgG (H+L); Southern Biotech, Birmingham, AL, USA). For chemiluminescent

visualization, Lumi‐Light plus substrate (Roche, Penzberg, Germany) was used. Blots were

analyzed by ImageQuant LAS4000. Band intensities were quantified using Image J software

(NIH, Bethesda, MD, USA).

Statistical analysis.

Statistical analyses were performed with Prism 5 (GraphPad Software, La Jolla, CA, USA)

applying Student’s t‐test. Differences were considered significant with P‐values <0.05 (*),

<0.01 (**) and <0.001 (***).


69

Conflict of Interest

The authors declare no conflict of interest.

Acknowledgements

We would like to thank G Kroemer (Institut Gustave Roussy, Villejuif, France) for the LC3–

GFP construct, N Mizushima (Department of Cell Biology, Okazaki, Japan) and M Komatsu

(Department of Molecular Oncology, Tokyo, Japan) for providing MEFs, SW Tait (Beatson

Institute, Institute of Cancer Sciences Glasgow, UK) for the plasmids. The microscopy images

were taken at the core facility Cellular Imaging/LCAM‐AMC. We would like to thank DI

Picavet for technical assistance, furthermore MF Bijlsma for his valuable scientific input and

helpful discussions. This work was supported by a grant of the Stichting Nationaal Fonds

tegen Kanker (SNFK), Amsterdam, The Netherlands.

Chapter 2

70

References 1. Ehrhardt H, Fulda S, Fuhrer M, Debatin KM, Jeremias I. Betulinic acid‐induced apoptosis in leukemia cells. Leukemia 2004; 18: 1406–1412. 2. Rzeski W, Stepulak A, Szymanski M, Sifringer M, Kaczor J, Wejksza K et al. Betulinic acid decreases expression of bcl‐2 and cyclin D1, inhibits proliferation, migration and induces apoptosis in cancer cells. Naunyn Schmiedebergs Arch Pharmacol 2006; 374: 11–20. 3. Jung GR, Kim KJ, Choi CH, Lee TB, Han SI, Han HK et al. Effect of betulinic acid on anticancer drug‐resistant colon cancer cells. Basic Clin Pharmacol Toxicol 2007; 101: 277–285. 4. Kessler JH, Mullauer FB, de Roo GM, Medema JP. Broad in vitro efficacy of plant‐derived betulinic acid against cell lines derived from the most prevalent human cancer types. Cancer Lett 2007; 251: 132–145. 5. Fulda S, Friesen C, Los M, Scaffidi C, Mier W, Benedict M et al. Betulinic acid triggers CD95 (APO‐1/Fas)‐ and p53‐independent apoptosis via activation of caspases in neuroectodermal tumors. Cancer Res 1997; 57: 4956–4964. 6. Fulda S, Susin SA, Kroemer G, Debatin KM. Molecular ordering of apoptosis induced by anticancer drugs in neuroblastoma cells. Cancer Res 1998; 58: 4453–4460. 7. Fulda S, Scaffidi C, Susin SA, Krammer PH, Kroemer G, Peter ME et al. Activation of mitochondria and release of mitochondrial apoptogenic factors by betulinic acid. J Biol Chem 1998; 273: 33942–33948. 8. Mullauer FB, Kessler JH, Medema JP. Betulinic acid induces cytochrome c release and apoptosis in a Bax/Bak‐independent, permeability transition pore dependent fashion. Apoptosis 2009; 14: 191–202.

9. Vandenabeele P, Vanden Berghe T, Festjens N. Caspase inhibitors promote alternative cell death pathways. Sci STKE 2006; 2006: e44. 10. Vandenabeele P, Galluzzi L, Vanden Berghe T, Kroemer G. Molecular mechanisms of necroptosis: an ordered cellular explosion. Nat Rev Mol Cell Biol 2010; 11: 700–714. 11. Tenev T, Bianchi K, Darding M, Broemer M, Langlais C, Wallberg F et al. The Ripoptosome, a signaling platform that assembles in response to genotoxic stress and loss of IAPs. Mol Cell 2011; 43: 432–448. 12. Degterev A, Huang Z, Boyce M, Li Y, Jagtap P, Mizushima N et al. Chemical inhibitor of nonapoptotic cell death with therapeutic potential for ischemic brain injury. Nat Chem Biol 2005; 1: 112–119. 13. Degterev A, Hitomi J, Germscheid M, Ch’en IL, Korkina O, Teng X et al. Identification of RIP1 kinase as a specific cellular target of necrostatins. Nat Chem Biol 2008; 4: 313–321. 14. Aits S, Jaattela M. Lysosomal cell death at a glance. J Cell Sci 2013; 126(Pt 9): 1905–1912. 15. Parzych KR, Klionsky DJ. An Overview of Autophagy: Morphology, Mechanism, and Regulation. Antioxid Redox Signal 2014; 20: 460–473. 16. Saeki K, Yuo A, Okuma E, Yazaki Y, Susin SA, Kroemer G et al. Bcl‐2 down‐regulation causes autophagy in a caspase‐independent manner in human leukemic HL60 cells. Cell Death Differ 2000; 7: 1263–1269. 17. Baehrecke EH. Autophagy: dual roles in life and death? Nat Rev Mol Cell Biol 2005; 6: 505–510. 18. Chen N, Karantza V. Autophagy as a therapeutic target in cancer. Cancer Biol Ther 2011; 11: 157–168. 19. Weidberg H, Shvets E, Elazar Z. Biogenesis and cargo selectivity of autophagosomes. Annu Rev Biochem 2011; 80: 125–156.


71

20. Lee J, Giordano S, Zhang J. Autophagy, mitochondria, and oxidative stress: cross‐talk and redox signalling. Biochem J 2012; 441: 523–540. 21. Till A, Lakhani R, Burnett SF, Subramani S. Pexophagy: the selective degradation of peroxisomes. Int J Cell Biol 2012; 2012: 512721. 22. Kim I, Rodriguez‐Enriquez S, Lemasters JJ. Selective degradation of mitochondria by mitophagy. Arch Biochem Biophys 2007; 462: 245–253. 23. Gonzalez P, Mader I, Tchoghandjian A, Enzenmuller S, Cristofanon S, Basit F et al. Impairment of lysosomal integrity by B10, a glycosylated derivative of betulinic acid, leads to lysosomal cell death and converts autophagy into a detrimental process. Cell Death Differ 2012; 19: 1337–1346. 24. Yang LJ, Chen Y, He J, Yi S, Wen L, Zhao J et al. Betulinic acid inhibits autophagic flux and induces apoptosis in human multiple myeloma cells in vitro. Acta Pharmacol Sin 2012; 33: 1542–1548. 25. Mizushima N. Methods for monitoring autophagy. Int J Biochem Cell Biol 2004; 36: 2491–2502. 26. Mizushima N, Yoshimori T. How to interpret LC3 immunoblotting. Autophagy 2007; 3: 542–545. 27. Klionsky DJ, Abeliovich H, Agostinis P, Agrawal DK, Aliev G, Askew DS et al. Guidelines for the use and interpretation of assays for monitoring autophagy in higher eukaryotes. Autophagy 2008; 4: 151–175. 28. Klionsky DJ, Abdalla FC, Abeliovich H, Abraham RT, Acevedo‐Arozena A, Adeli K et al. Guidelines for the use and interpretation of assays for monitoring autophagy. Autophagy 2012; 8: 445–544. 29. Mortimore GE, Poso AR. Intracellular protein catabolism and its control during

nutrient deprivation and supply. Annu Rev Nutr 1987; 7: 539–564. 30. Kimura S, Noda T, Yoshimori T. Dissection of the autophagosome maturation process by a novel reporter protein, tandem fluorescent‐tagged LC3. Autophagy 2007; 3: 452–460. 31. Scorrano L, Ashiya M, Buttle K, Weiler S, Oakes SA, Mannella CA et al. A distinct pathway remodels mitochondrial cristae and mobilizes cytochrome c during apoptosis. Dev Cel 2002; 2: 55–67. 32. Newmeyer DD, Ferguson‐Miller S. Mitochondria: releasing power for life and unleashing the machineries of death. Cell 2003; 112: 481–490. 33. Tsujimoto Y, Shimizu S. Role of the mitochondrial membrane permeability transition in cell death. Apoptosis 2007; 12: 835–840. 34. Galluzzi L, Vicencio JM, Kepp O, Tasdemir E, Maiuri MC, Kroemer G. To die or not to die:that is the autophagic question. Curr Mol Med 2008; 8: 78–91. 35. Komatsu M, Waguri S, Ueno T, Iwata J, Murata S, Tanida I et al. Impairment of starvationinduced and constitutive autophagy in Atg7‐deficient mice. J Cell Biol 2005; 169: 425–434. 36. Kuma A, Hatano M, Matsui M, Yamamoto A, Nakaya H, Yoshimori T et al. The role of autophagy during the early neonatal starvation period. Nature 2004; 432: 1032–1036. 37. Petersen NH, Olsen OD, Groth‐Pedersen L, Ellegaard AM, Bilgin M, Redmer S et al. Transformation‐associated changes in sphingolipid metabolism sensitize cells to lysosomal cell death induced by inhibitors of Acid sphingomyelinase. Cancer Cell 2013; 24: 379–393.

Chapter 2

72

38. Brunk UT, Neuzil J, Eaton JW. Lysosomal involvement in apoptosis. Redox Rep 2001; 6: 91–97. 39. Leist M, Jaattela M. Four deaths and a funeral: from caspases to alternative mechanisms. Nat Rev Mol Cell Biol 2001; 2: 589–598. 40. Guicciardi ME, Leist M, Gores GJ. Lysosomes in cell death. Oncogene 2004; 23: 2881–2890. 41. Kroemer G, Jaattela M. Lysosomes and autophagy in cell death control. Nat Rev Cancer 2005; 5: 886–897. 42. Boya P, Kroemer G. Lysosomal membrane permeabilization in cell death. Oncogene 2008; 27: 6434–6451. 43. Groth‐Pedersen L, Jaattela M. Combating apoptosis and multidrug resistant cancers by targeting lysosomes. Cancer Lett 2013; 332: 265–274. 44. Elmore SP, Qian T, Grissom SF, Lemasters JJ. The mitochondrial permeability transition initiates autophagy in rat hepatocytes. FASEB J 2001; 15: 2286–2287. 45. Rodriguez‐Enriquez S, He L, Lemasters JJ. Role of mitochondrial permeability transition pores in mitochondrial autophagy. Int J Biochem Cell Biol 2004; 36: 2463–2472. 46. Hoyer‐Hansen M, Bastholm L, Szyniarowski P, Campanella M, Szabadkai G, Farkas T et al Control ofmacroautophagy by calcium, calmodulin‐dependent kinase kinase‐beta, and Bcl‐2. Mol Cell 2007; 25: 193–205. 47. Colell A, Ricci JE, Tait S, Milasta S, Maurer U, Bouchier‐Hayes L et al. GAPDH and autophagy preserve survival after apoptotic cytochrome c release in the absence of caspase activation. Cell 2007; 129: 983–997. 48. Nicoletti I, Migliorati G, Pagliacci MC, Grignani F, Riccardi C. A rapid and simple

method for measuring thymocyte apoptosis by propidium iodide staining and flow cytometry. J Immunol Methods 1991; 139: 271–279. 49. Hoogendijk AJ, Roelofs JJ, Duitman J, van Lieshout MH, Blok DC, van der Poll T et al. R‐roscovitine reduces lung inflammation induced by lipoteichoic acid and Streptococcus pneumoniae. Mol Med 2012; 18: 1086–1095.


73

Supplementary figures

Supplementary Figure 1. BetA induces cell death in a concentration dependent fashion. HeLa cells were

treated for 48 hours with indicated concentrations of BetA after which cell death was assessed by PI

exclusion.

Supplementary Figure 2. Necrostatin blocks TNF‐induced necroptosis. U937 cells were pre‐treated with or

without 10 µM zVAD.fmk and with or without 25 µM necrostatin‐1 for one hour and subjected to 40 ng/ml

TNF‐α for 48 h after which cell death was assessed through PI exclusion. Mean + S.D. of three indendent

experiments are shown, ***P<0.001.

0

20

40

60

80

100

% P

Ipos

0 5 7.5 10

μg/ml BetA

***

***%

PIpo

s

Q-VD-OPh--

-+

+

- --

-

necrostatin - - ++

++

++

TNF-α

***

0

20

40

60

+

***

Chapter 2

74

Supplementary Figure 3. BetA induces autophagy and autophagic flux. HeLa cells were treated with (a)

indicated concentrations of BetA for 18 hours or (b) indicated time points. LC3 processing was assessed via

immunoblotting. Results for endogenous LC3 are shown. LC3 I (18 kDa) and LC3 II (16 kDa). Tubulin is used

as a control. Band intensity of western blot was quantified using Image J software. Mean + S.D. of three

independent experiments are shown, *P < 0.05; **P< 0.01; ***P<0.001; n.s., not significant. (c) HeLa cells

were pre‐treated for 1 hour with 10 µg/ml cycloheximide followed by 48 h of BetA treatment. P62 levels

were determined via western blot. ERK1/2 is used as a control. Band intensity was quantified using Image J

software. Single experiment is shown. (d) HeLa cells and A549 cells with tandem fluorescent P62 (pDest‐

mCherry‐EGFP‐p62) were treated with 10 µg/ml BetA or 10 µM rapamycin and after 18 h cells were analyzed

by confocal microscopy. Representative pictures are shown of three experiments.

a

LC3 I

LC3 II

Tubulin

0 5 7.5 10 μg/ml BetA2.5 - + - + - + μg/ml BetA6h 12h 18h

b

LC3 I

LC3 II

Tubulin

0

1

2

3

LC

3 I

I /

LC

3 I

μg/ml BetA0 5 7.5 102.50

2.5

5

LC

3 I

I /

LC

3 I

- + - + - + μg/ml BetA6h 12h 18h

nsns

*****

ns * *

c

DMSO rapamycinBetA

0 µm 10 0 µm 10 0 µm 10

0 µm 10 0 µm 10 0 µm 10

0 25µm 0 25µm 0 25µm

HeL

aA

549

d

0

1

2

rel. p

62

p62

Erk1/2

rapamycin-- -

++

- --

-cycloheximide - -+

+ +

+BetA

+- -


75

Supplementary Figure 4. ATG deficiency results in increased BetA‐induced cell death (a) Knockdown of

ATG5 in HeLa, HeLa\LC3‐GFP and MCF‐7\LC3‐GFP cells with shRNA control, shRNA ATG5 #1 and shRNA ATG5

#3 was assessed via immunoblotting. (b) MCF‐7\LC3‐GFP cells with ATG5 knockdown #1 and #2 and control

cells were subjected to different concentrations of BetA. After 48 h after cell death was measured via PI

exclusion and DNA fragmentation was assessed. Mean + S.D. of three independent experiments are shown.

(c) Mouse embryonic fibroblasts lacking either AtTG5 or ATG7 and their respective control cells were

subjected to different concentrations of BetA for 18 h and phosphatidylserine (PS) exposure was measured.

Mean + S.D. of three independent experiments are shown.

a

+

+-

-control

ATG5 #2

Atg5

level ATG5

ATG5 #1

+

++

+

--

- --

- -

- --

HeLaMCF-7LC3-GFP

HeLaLC3-GFP

+

++

-- -

- --

1.00 0.28 0.22 0.251.00 1.000.14 0.42 0.29

controlshRNAATG5 #1shRNAATG5 #2

0 5 10 150

20

40

60

80

100

BetA (µg/ml) BetA (µg/ml)0 5 10 15

0

10

20

30

40

50

% D

NA

fra

gmen

tatio

nb

c

% P

Ipos

BetA (µg/ml) BetA (µg/ml)

controlshRNAATG5 #1shRNAATG5 #2

0

20

40

60

80

100

% A

nnex

in V

pos

0 5 10 15 20 0 5 10 15 200

20

40

60

80

100

% A

nnex

in V

pos

wtAtg5 -/-

Atg7 +/-Atg7 -/-

Chapter 3

Betulinic acid induces a novel cell death

pathway that depends on cardiolipin

modification

Lisette Potze, Simone di Franco, Mia Pras‐Raves, Daisy I. Picavet,

Henk A. van Veen, Henk van Lenthe, Franziska B. Mullauer,

Nicole N. van der Wel, Angela Luyf, Antoine H. C. van Kampen,

Stephen Kemp, Vincent Everts, Jan H. Kessler, Frédéric M. Vaz,

Jan Paul Medema

Oncogene 2015

“The more there is of mine, the less there is of yours”


Chapter 3

78

Abstract

Cancer is associated with strong changes in lipid metabolism. For instance, normal cells take

up fatty acids (FAs) from the circulation, while tumour cells generate their own and become

dependent on de novo FA synthesis, which could provide a vulnerability to target tumour

cells. Betulinic acid (BetA) is a natural compound that selectively kills tumour cells through

an ill‐defined mechanism that is independent of BAX and BAK, but depends on

mitochondrial permeability transition‐pore opening. Here we unravel this pathway and

show that BetA inhibits the activity of stearoyl‐CoA‐desaturase (SCD‐1). This enzyme is

overexpressed in tumour cells and critically important for cells that utilize de novo FA

synthesis as it converts newly synthesized saturated FAs to unsaturated FAs. Intriguingly,

we find that inhibition of SCD‐1 by BetA or, alternatively, with a specific SCD‐1 inhibitor

directly and rapidly impacts on the saturation level of cardiolipin (CL), a mitochondrial lipid

that has important structural and metabolic functions and at the same time regulates

mitochondria‐dependent cell death. As a result of the enhanced CL saturation mitochondria

of cancer cells, but not normal cells that do not depend on de novo FA synthesis, undergo

ultrastructural changes, release cytochrome c and quickly induce cell death. Importantly,

addition of unsaturated FAs circumvented the need for SCD‐1 activity and thereby

prevented BetA‐induced CL saturation and subsequent cytotoxicity, supporting the

importance of this novel pathway in the cytotoxicity induced by BetA.

BetA‐induced cell death by cardiolipin saturation

79

Introduction

Cancer is characterized by cell growth and proliferation, which requires an enormous surge

in novel building blocks, such as nucleic acids, lipids and amino acids. To meet these

requirements, cancer cells undergo major changes in their metabolism. For instance, lipid

metabolism undergoes a dramatic shift towards lipid synthesis.1–3 For cellular lipid

production, fatty acids (FAs) are needed as building blocks, which are either derived from

exogenous sources or from de novo FA synthesis. Whereas normal cells prefer exogenous

sources, tumour cells favour de novo synthesis.4–6 This differential usage of FA source is

supported by observations that enzymes involved in FA synthesis, such as ATP‐citrate lyase

and FA synthase, are required for transformation of cells and upregulated in several tumour

types.7–10 The main products of FA synthesis are palmitic acid (PA) and stearic acid,

saturated FAs that are toxic for the cell at high levels and therefore converted to their

monounsaturated forms (palmitoleic acid (POA) and oleic acid) by stearoyl‐CoA‐desaturase

(SCD‐1). This enzyme, located in the endoplasmatic reticulum, catalyses the introduction of

a double bond at the ω‐9 position of saturated FAs and is overexpressed in cancer cells.11–

16 Unsaturated FAs are abundantly present in mammalian cells, as their structural properties

are needed to maintain optimal fluidity of cellular membranes. Due to its tumour‐selective

activation, de novo FA synthesis has been suggested as an attractive target of anticancer

therapy.8

Betulinic acid (BetA) is a cytotoxic plant‐derived compound that is tumour selective and

does not kill normal cells.17–19 It induces apoptosis in a wide variety of tumour types through

a not completely understood, but mitochondria‐dependent mechanism.20–23 Importantly,

this involves mitochondrial permeability transition (PT)‐pore opening, but is independent

of BAX and BAK, pointing to a distinct mechanism as compared with classical mitochondrial‐

dependent apoptosis. The general anti‐cancer applicability of BetA and its selectivity

suggests that it targets an essential and common pathway in tumour cells. Here we show

that BetA affects mitochondrial morphology by changing the FA composition of cardiolipin

(CL). CL is a unique mitochondrial phospholipid that contains four FA side chains. It is

associated with several mitochondrial proteins and respiratory chain complexes and is

essential for mitochondrial function and structural integrity of the organelle.24–27 In addition,

CL is known to have an important role in apoptosis due to its interaction with core apoptotic

machinery components, such as cytochrome c, caspase‐ 8, BCL‐2 family members and PT‐

pore components.25, 28–31

Our study demonstrates that BetA treatment increases the incorporation of saturated FAs

in CL through inhibition of SCD‐1. This is detrimental for mitochondrial morphology and

leads to apoptotic cancer cell death.

Chapter 3

80

Results

Mitochondrial morphology is changed and CL saturation increased upon BetA

treatment

In untreated cancer cells and normal cells, mitochondria appeared as a tubular network

evenly divided throughout the cell (Figure 1a and Supplementary Figure 1). In contrast, in

cancer cells treated with BetA the mitochondrial network changed in a concentration

dependent manner and became more fragmented (Figure 1a and Supplementary Figure 1).

Strikingly, this fragmentation is not observed in non‐transformed human fibroblasts (Co18;

Supplementary Figure 1). To gain more insight into these mitochondrial changes at the

ultrastructural level, transmission electron microscopy was employed. Untreated cells

contained mitochondria with a typical longitudinal cristae. Upon BetA treatment,

mitochondria were more electron dense and displayed significant ultrastructural changes,

in particular, concentric (circular shaped) cristae were evident (Figure 1b). Interestingly,

such cristae abnormalities have previously been observed in tissues of Barth syndrome

(BTHS) patients,32 a disease that is caused by a germline mutation in the Tafazzin gene,

which encodes a transacylase involved in CL remodelling and maturation. Cells of BTHS

patients display characteristic changes in CL, which cause mitochondrial ultrastructural as

well as functional abnormalities.33

To determine whether the observed BetA‐induced mitochondrial changes were also due to

changes in the composition of CL, analysis of CL in several cancer cell lines was performed

using high‐performance liquid chromatography–mass spectrometry. A large variety of CL

molecules is normally detected, which is due to a difference in the length and saturation of

the four FA side chains in CL. In untreated tumour cells, CL appeared as six major clusters

containing FA side chains of increasing length (from C64 to C74). The individual clusters are

then further subdivided based on the level of saturation of their FA side chains (indicated

by CL(64:X) to CL(74:X), where X denotes the number of unsaturations, that is, double bonds

in the FA side chains; Figure 2a). The major CL cluster detected contained 68 side chain

carbons (C68) and one to five double bonds (C68:1 to C68:5; Figure 2a and c). Importantly,

BetA treatment of several cancer cell lines resulted in significant changes in the composition

of CL that is evident from a shift in CL clusters to higher m/z values and hence less

unsaturated CL species (Figure 2b). Zooming in on the major C68 cluster in these cell lines

revealed a strong decrease in CL species with four or five double bonds and a concomitant

increase of more saturated C68 CL species with only two or three unsaturations (Figure 2c

and e, Supplementary Figure 2A, B and Supplementary Figure 6B and C). This profile is very

similar to the one observed in fibroblasts from BTHS patients (Figure 2f and h), which as a

consequence have major defects in their mitochondria, suggesting that the BetA‐induced

CL modification is instrumental in the mitochondrial changes.


81

The changes in FA composition of CL can be brought about by either de novo synthesis of

complete CL molecules or by direct modification of the FA side chains by specific

remodelling enzymes.34 To monitor the effect of BetA on de novo FA incorporation in CL,

cells were incubated with heptadecanoic acid (C17:0), which is a low‐abundant FA in normal

physiology. Within a day, C17‐containing CL clusters (odd numbered CL species) made up a

significant proportion of the CL present in cells, indicating that this odd‐chain FA was rapidly

incorporated into CL (Figure 2i and k). These observations indicate that CL is relatively

rapidly metabolized in cancer cells. In addition, the degree of unsaturation of C17‐

containing CL clusters was comparable to the existing (even numbered) clusters (Figure 2m),

which suggested that exogenous C17:0 was taken up into the FA pool and also converted to

its monounsaturated form, C17:1. Importantly, in BetA‐treated cells rapid CL side chain

turnover was also observed. However, in the presence of BetA the C17‐containing clusters

were more saturated (Figure 2j and l), a shift that was similar to the even‐numbered

counterparts (Figure 2e and n). This indicated that BetA‐induced CL modification is a direct

result from incorporation of novel FA side chains that in the presence of BetA are more

saturated. Importantly, phosphaditylglycerol, which is the precursor of CL, also

accumulated in a relatively saturated form in BetA‐treated cells (Supplementary Figure 3A

and B). Modification of CL was therefore dependent on a step further upstream in the

biosynthesis of CL.

SCD‐1 activity is instrumental in BetA‐induced mitochondrial damage

In cells, de novo synthesized and exogenous saturated FAs are converted to unsaturated

FAs via the enzyme SCD‐1 (Figure 3a). In cancer cells SCD‐1 RNA levels are higher as

compared with normal fibroblasts (Supplementary Figure 4A). This is consistent with

literature reports and suggests that cancer cells are more dependent on desaturase

activity13,16 due to the activation of de novo FA synthesis in favour of FA uptake.

Accumulation of saturated FAs in phosphaditylglycerol and CL could therefore be a direct

result of dysfunctional FA conversion by SCD‐1. To measure SCD‐1 activity, stable isotope

labelled 2H3‐C16:0 and 2H5‐C18:0 were added to live cells and conversion rates to

unsaturated FAs were measured. Untreated cancer cells displayed a rapid conversion to the

monounsaturated FAs, which was blocked by a specific SCD‐1 inhibitor or genetic

knockdown of SCD‐1, confirming that SCD‐1 was responsible for this conversion (Figure 3b

and d and Supplementary Figure 4B and C). In line with the RNA levels, the SCD‐1 activity is

significantly lower in Co18 as compared with cancer cells.

Importantly, generation of unsaturated FAs by SCD‐1 was strongly reduced by addition of

BetA in all lines tested. This indicates that BetA inhibited SCD‐1 activity (Figure 3b and c and

Supplementary Figure 4C) and directly affects the pool of intracellular FAs, which then

impacts on the saturation level of CL. In agreement, inhibition of SCD‐1 with a commercial

inhibitor also resulted in a rapid change in the composition of CL (Figure 3g) as well as in a

Chapter 3

82

significant change in the size, ultrastructure and network of mitochondria (Figure 3h and

Supplementary Figure 5). These data suggest that blocking FA desaturation is detrimental

for the mitochondria through a rapid modification of CL side chains.

Figure 1. BetA‐induced mitochondrial morphology changes (a) HeLa cells with MitoRFP were treated with

vehicle (DMSO), 5 µg/ml or 10 µg/ml BetA and after 18h cells were analysed by confocal microscopy. (b)

HeLa cells were treated with vehicle (DMSO) or 10 µg/ml BetA for 18 h after which cells were fixed and

processed for TEM analysis. Pictures representative for 3 experiments are shown.

Unsaturated FAs revert CL modification and mitochondrial changes

High levels of saturated FAs are known to be cytotoxic and our current data point to a direct

impact on the mitochondria.35,36 If this model is correct then addition of a surplus of

saturated FAs would aggravate, while addition of unsaturated FAs would rather revert BetA‐

induced mitochondrial damage. Cells were therefore first treated with either PA (C16:0) or

POA (C16:1). Addition of PA had relatively little impact on the composition of CL, as

exemplified by the C68 cluster (Figure 3e and f and Supplementary Figure 6A and C),

indicating that the added FA did not change normal CL metabolism. Single addition of POA

slightly enhanced the level of unsaturation in CL as compared with untreated tumour cells

(Figure 3e and f and Supplementary Figure 6A and C). However, the most striking effect was

A DMSO 5 μg/ml BetA 10 μg/ml BetA

ZO

OM

B

ZO

OM

ZO

OM

DM

SO

Bet

A

0 μm 25 0 μm 25 0 μm 25

500nm 500nm


83

observed when POA was added together with BetA as this prevented the induction of CL

saturation significantly (Figure 3e and f and Supplementary Figure 6A and C). Similar effects

were observed when CL saturation was induced by SCD‐1 inhibitor (Figure 3g and

Supplementary Figure 6D), indicating that the desaturase inhibition indeed directly impacts

on CL due to a decrease in the production of unsaturated FAs. In contrast, PA addition rather

enhanced the effect that BetA and SCD‐1 inhibition exerted on CL saturation (Figure 3e and

g and Supplementary Figure 6A and D). This confirmed that blocking SCD‐1 activity is

instrumental in BetA‐induced CL changes as saturation can be prevented by the addition of

unsaturated FAs, which bypasses the need for SCD‐1 activity. The causal link between SCD‐

1 activity and CL saturation prompted us to investigate whether BetA‐induced

mitochondrial morphology changes were a direct result of CL saturation or whether they

are induced through alternative pathways. Addition of either saturated (PA) or unsaturated

(POA) FAs had no effect on the tubular mitochondrial network and cristae structure of

mitochondria (Supplementary Figure 7A and B). However, the BetA‐induced mitochondrial

changes were clearly potentiated by addition of saturated FAs (PA) and more importantly

prevented by unsaturated FAs (POA; Figure 3i and j), establishing a causal relationship

between SCD‐1 activity, CL FA composition and mitochondrial morphology.

CL modification evokes cell death and decreases clonogenic survival

Previously, we provided evidence that BetA induces mitochondrial leakage, cytochrome c

release in a PT‐pore dependent fashion and subsequent cell death.23 Cytochrome c release

was indeed observed upon BetA treatment and with SCD‐1 inhibitor treatment (Figure 4a

and c), but interestingly, this release was significantly reduced when treatments were

combined with POA. Vice versa, addition of PA enhanced the amount of cells showing

cytochrome c release (Figure 4a and c), indicating that the inhibitory effects of BetA on SCD‐

1 activity directly impact on mitochondrial damage and lead to cytochrome c release.

Moreover, these observations extend to the actual induction of cell death as measured by

propidium iodide exclusion (Figure 4d and e). The validity of this novel cell death pathway

was further substantiated with the use of SCD‐1 inhibitor or genetic knockdown, which, next

to mitochondrial defects, also induced cell death that could be blocked by exogenous

addition of POA (Figure 4f and Supplementary Figure 8A and B). These data suggest that

BetA‐induced cell death by inhibiting SCD‐1. It is important to realize though that BetA‐

induced death is more rapid than SCD‐1‐induced cell death, whereas the effects on SCD‐1

activity are if anything more effective with SCD‐1 inhibitors. This suggests that BetA may

have additional effects that enhance the dependency on SCD‐1 activity. In agreement,

combining BetA with SCD‐1 inhibitor resulted in enhanced cell death as compared with SCD‐

1 inhibitor alone, confirming that BetA has a stronger impact on cells. We hypothesize that

BetA enhances the need for unsaturated FA by increasing the turnover of lipids and as

Chapter 3

84

m/z

m/z

670 680 690 700 710 720 730 740

20

100

0

40

60

80

rela

tive

inte

nsity

(%

)

672.4673.4

674.4

686.5685.5

687.5

698.5

699.5

700.5

701.5

711.5

712.5

713.5

714.5

715.5724.5

725.5

726.5 727.5

728.5

737.5

738.5

739.5

C74

C72PG(32:x)

C70C68

20

100

0

40

60

80

rela

tive

inte

nsity

(%

)

672.4

673.4674.4

675.4

668.5

686.5

687.5

688.5

689.5

698.5

699.5

701.5

702.5

700.5

711.5

712.5

713.5 714.5

715.5

716.5

PG(32:x)

724.5

725.5

726.5727.5

728.5

737.5

738.5739.5

698 699 700 701 702 703

100

80

60

40

20

0

Rel

ativ

e in

tens

ity (

%)

698 699 700 701 702 703

100

80

60

40

20

0

Rel

ativ

e in

tens

ity (

%)

DMSO BetA

698 699 700 701 702 703

100

80

60

40

20

0

Re

lativ

e in

tens

ity (

%)

704 698 699 700 701 702 703

100

80

60

40

20

0

Rel

ativ

e in

tens

ity (

%)

704m/z

m/z m/z

AAA

B

C D E

F G H

fo

ld in

crea

se/d

ecre

ase

Bet

A v

s. a

vg.

DM

SO

( 2lo

g)

f

old

incr

eas

e/d

ecre

ase

BH

TS

vs.

avg

. co

ntro

l ( 2lo

g)

1.0

1.5

0.0

0.5

-1.0

-1.5

-0.5

0.0

2.0

4.0

-2.0

-4.0

Control BTHS

DMSO

BetA

C64

C66

C74

C72

C70

C68

C64

C66

C68:5

C68:4

C68:3C68:2

C68:1

C68:5

C68:4

C68:3

C68:2

C68:1

C68:5

C68:4

C68:3

C68:2

C68:1C68:5

C68:4

C68:2

C68:1

C68:5C68:4

C68:3

C68:2

C68:1

CL(68:2)

CL(68:3)

CL(68:4)

CL(68:5)

CL(68:2)

CL(68:3)

CL(68:4)

CL(68:5)

C68:3


85

Figure 2. BetA‐induced cardiolipin saturation (a) HeLa cells were treated for 18 h with vehicle (DMSO)

followed by CL analysis by HPLC mass spectrometry. (b) HeLa cells were treated for 18 h with 10 µg/ml BetA

followed by CL analysis by HPLC mass spectrometry. (c) Zoom in of CL C68 cluster (containing 2 x C16 and 2

x C18 fatty acids) of DMSO treated cells 18 h (d) Zoom in of CL C68 cluster of 10 µg/ml BetA treated cells.

(e) Quantification of fold change of C68:X species in BetA treated (18 h) HeLa cells (n=8) compared to vehicle

treated HeLa cells (n=8). (f) Zoom in of CL C68 cluster of control fibroblasts. (g) Zoom in of CL C68 cluster of

BTHS patients. (h) Quantification of fold change of C68:X species in BTHS patients (n=3) compared to

controls (n=3). HeLa cells were pre‐treated for 4 hours with vehicle (DMSO) (i, j) or 50 µM heptadecanoic

acid (C17:0) (k, l) after which vehicle (DMSO) i, k) or 10 µg/ml BetA (j, l) was applied for 18h. Zoom in of CL

clusters of C68 and C69 (containing C17) are shown. (m) Quantitative analysis of CL content in C68 and C69

CL cluster of DMSO C17:0 treated cells (n=2). (n) Quantification of change of C68:X species in BetA treated

HeLa cells compared to vehicle (DMSO) treated HeLa cells (n=2)

100

80

60

40

20

0

5

4

3

2

1

54 3

21

698 700 702 704 706 708 710

5

43

2

1

5

4 32

1

698 700 702 704 706 708 710

5

4

3

2

1

5

4

3

2

1

100

80

60

40

20

0698 700 702 704 706 708 710

5

43

2

15

4

3

2

1

698 700 702 704 706 708 710

DMSO BetA

DMSO + C17 BetA + C17

Rel

ativ

e in

tens

ity (

%)

Rel

ativ

e in

tens

ity (

%)

m/zm/z

m/zm/z

nmol

CL/

mg

pro

tein

0.0

0.2

0.4

0.6

0.8

C68 C69

4

5

3

2

fold

incr

ease

/decr

ease

BetA

vsavg

.D

MS

O (

2lo

g)

C68 C69 C68 C69

C68 C69C68 C69

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

I J

K L

M N

100

80

60

40

20

0

100

80

60

40

20

0

4

5

3

2

Chapter 3

86

0.00

0.05

0.10

0.15

ratio

2H

3-

C16

:1/2

H3-

C16

:0 ***

***

0.0

0.2

0.4

0.6***

***

SCD-1 inhibBetA

+-- -

-+

SCD-1 inhibBetA

+-- -

-+

CO

S

CoA

CO

S

CoA

C9 C10

SCD-1

Saturated fatty acids

Unsaturated fatty acids

A B C

DMSO

PO

AD

MS

O

100

80

60

40

20

0

C68:1

C68:2

C68:3

C68:4

C68:5

698 699 700 701 702 703

100

80

60

40

20

0

Re

lativ

e in

ten

sity

(%

)

C68:1

C68:2

C68:3

C68:4

C68:5

698 699 700 701 702 703

C68:1

C68:2

C68:3

C68:4

C68:5

698 699 700 701 702 703

m/z

C68:1

C68:2

C68:3

C68:4

C68:5

698 699 700 701 702 703

C68:1

C68:2

C68:3C68:4

C68:5

698 699 700 701 702 703

C68:1

C68:2

C68:3C68:4

C68:5

698 699 700 701 702 703

PA

BetAD

0

1

2

3

rela

tive

CL

satu

ratio

nPOA

PABetA

+- - - -- -- -- - -

+

+ +++

+

*

nsE

ratio

2H

5-

C18

:1/2

H5-

C18

:0

100

80

60

40

20

0

m/z

1.5

0.0

0.5

1.0

rela

tive

CL

satu

ratio

n

**

***

POAPA

+- - - -- -- -- - -

+

+ +++

+SCD-1 inhib

F

500nm

DMSO SCD-1 inhibitor

500nm

G

Re

lativ

e in

ten

sity

(%

)R

ela

tive

inte

nsi

ty (

%)


87

Figure 3. Desaturase activity is instrumental in BetA‐induced mitochondrial damage (a) SCD‐1 enzyme,

introduces a double bond in saturated fatty acids at the n‐9 position producing unsaturated fatty acids. (b)

HeLa cells were treated with vehicle (DMSO), 150 nM SCD‐1 inhibitor or 10 µg/ml BetA for 18 h after which 2H3‐C16:0 was applied for additional 6 h. Conversion to 2H3‐C16:1 was measured. Mean ± SD of 3

experiments are shown. ***P<0.001. (c) HeLa cells were treated with vehicle, 150 nm SCD‐1 inhibitor or 10

µg/ml BetA for 18 h after which 2H5‐C18:0 was applied for additional 6 h. Conversion to 2H5‐C18:1 was

measured. Mean ± SD of 3 experiments are shown. ***P<0.001. (d) Zoom in of CL C68 clusters. HeLa cells

were incubated with vehicle (DMSO), 10 µg/ml BetA , 50 µM POA, 50 µM PA or a combination of 10 µg/ml

BetA with 50 µM POA or 50 µM PA for 18 h after which CL analysis was performed using HPLC mass

spectrometry (e) Quantification of CL saturation as measured in (d). Relative C68 CL saturation is shown

which is calculated as (C68:2 + C68:3) / (C68:4 + C68:5). Mean ± SD of 2 experiments are shown. * P< 0.05;

ns, not significant. (f) HeLa cells were subjected to vehicle (DMSO) or 200 nM SCD‐1 inhibitor with or without

50 µM POA or 50 µM PA for 40 h after which CLs were analysed by HPLC mass spectrometry. Relative C68

CL saturation is shown which was calculated as (C68:2 + C68:3) / (C68:4 + C68:5). Mean ± SD of 3 experiments

are shown. **P< 0.01; ***P<0.001. (g) HeLa cells were treated with vehicle (DMSO) or 200 nM SCD‐1

inhibitor for 40h after which cells were analysed via TEM. (h) HeLa cells with MitoRFP were subjected to

vehicle (DMSO) or 10 µg/ml BetA with or without 50 µM POA or 50 µM PA and after 18 h cells were analysed

via confocal microscopy. Pictures representative for 3 experiments are shown. (i) HeLa cells were subjected

to vehicle (DMSO), 10 µg/ml BetA with or without 50 µM POA or 50 µM PA and after 18 h cells were analysed

via TEM. Pictures representative for 3 experiments are shown.

DMSO BetA BetA + POA BetA + PAZ

OO

M

ZO

OM

ZO

OM

ZO

OM

0 μm 25 0 μm 25 0 μm 25 0 μm 25

DMSO BetA BetA + POA BetA + PA

500nm 500nm 500nm 500nm

H

I

Chapter 3

88

such increases the incorporation of toxic saturated FA into lipids. Sustaining a stable pool of

unsaturated FAs, by exogenous addition, therefore maintains the unsaturation grade of CL,

mitochondrial integrity and thereby allows for cell survival. The maintenance of the

unsaturated FA pool exerted long‐term protection as it even resulted in the recurrence of

clonogenic potential of BetA or SCD‐1 inhibitor‐treated cells. Single addition of these

compounds resulted in a complete loss of clonogenic growth (Figure 4g and h), which was

regained by the simple addition of POA. This effect was selective for this pathway and for

instance not observed when cells were treated with classical chemotherapy such as

etoposide (Supplementary Figure 8D). Taken together these data indicate that tumour cells

carry an intrinsic mitochondrial vulnerability that can be effectively targeted by inhibition

of SCD‐1, using either specific inhibitors or the natural compound BetA and is not observed

in normal fibroblasts. This vulnerability depends on enhancing the saturation of CL leading

to altered mitochondrial structure, membrane permeability and as a consequence cell

death and loss of clonogenic growth occur.

Discussion

Altered lipid metabolism is a key feature of tumour cells and required to sustain the

constant need for new biomass. Our current study provides strong evidence for an

unexpected mitochondrial vulnerability of tumour cells that depends on this de novo lipid

synthesis. De novo FA synthesis generates intrinsically toxic FAs, which need to be converted

by SCD‐1 into monounsaturated FAs. Here we show that inhibition of this final step in the

generation of preferred FAs results in cell death and loss of clonogenic growth. Interestingly,

we confirmed that normal cells do not contain high levels of SCD‐1 expression or activity

consistent with the idea that these cells do not utilize de novo FA synthesis but instead

acquire unsaturated FA from the surrounding. Although this effect of SCD‐1 inhibition on

tumour cells has been documented before, our data now provides an unexpected

mechanistic explanation of this observed toxicity. Firstly, we show that SCD‐1 inhibition

quickly impacts on the composition of CL FA side chains, resulting in mitochondrial

abnormalities, cytochrome c release and subsequent cell death. Secondly, we demonstrate

that BetA, which is a potent anti‐cancer agent, does so by inhibiting SCD‐1 and therefore

utilizes this tumour cell vulnerability. Together these data point to an upstream effect of

BetA on CL biosynthesis. The fact that SCD‐1 inhibitors mimic the effect of BetA and the

observation that both can be reverted by the simple addition of unsaturated FAs all point

to SCD‐1 as the vulnerability in the CL biosynthesis pathway of tumour cells. Previously, we

observed that BetA‐induced cell death is tumour selective and that normal cells are not

affected at doses that kill tumour cells both in vitro and in vivo. 22,37 In agreement, the BetA

or SCD‐1‐induced mitochondrial fragmentation in cancer cells is not detected in normal

cells. The here proposed mechanism of action could


89

Figure 4. CL modification evokes cell death and decreases clonogenic survival (a) HeLa cells were pre‐treated for 24h with 50 µM POA after which vehicle (DMSO) or 10 µg/ml BetA was applied for 40 h. Other

samples were left untreated for 24 h after which 50 µM PA with or without 10 µg/ml BetA for 40 h was

applied. After treatment cytochrome c release was assessed via FACS. Data of 3 independent experiments

are shown. * P< 0.05, ns not significant. (b, c) HeLa cells were pre‐treated with 50 µM POA after which

vehicle (DMSO) or 10 µg/ml BetA was applied for 48 h. Other samples were left untreated for 24 h after

which 50 µM PA with or without 10 µg/ml BetA for 48h was applied. Cell death was assessed via PI exclusion.

Mean ± SD of 3 experiments are shown. * P< 0.05, ns not significant. (d) HeLa cells were subjected to vehicle

or 75 nM SCD‐1 inhibitor with or without 50 µM POA or 50 µM PA for 72 h after which cell death was

assessed via PI exclusion. Mean ± SD of 3 experiments are shown. **P< 0.01; ***P<0.001 (e) HeLa cells were

treated for 48 h with vehicle (DMSO) or 10 µg/ml BetA with or without 50 µM POA or 50 µM PA after which

fresh medium was applied for 72h. Cells were stained with crystal violet blue for 6h. Mean ± SD of 3

experiments are shown. **P< 0.01. (f) HeLa cells were treated for 72 h with 75 nM SCD‐1 inhibitor with or

without 50 µM POA or 50 µM PA after which fresh medium was applied for 48h. Cells were stained with

crystal violet blue for 6h. Mean ± SD of 2 experiments are shown. ***P<0.001.

POAPA

BetA

+- - - -- -- -- - -

+

+ +++

+

0

20

40

60

80

100

% P

Ipos

*

ns

FCS-H

DMSO POA PA

PIpo

s

BetA BetA + POA BetA + PA

3.1% 2.6% 3.3%

53.8% 23.9% 72.6%

0

20

40

60

80

% C

ytoc

hro

me

c re

lea

se

POAPA

BetA

+- - - -- -- -- - -

+

+ +++

+

1

0

2

3

POAPA

BetA

+- - - -- -- -- - -

+

+ +++

+

**

**

*

ns

A B

C D

E

**

***

0

20

40

60

80

100%

PIpo

s

POAPA

+- - - -- -- -- - -

+

+ +++

+SCD-1 inhib

abs

590

- 6

30

0

0.5

1.0

1.5

2.0

******

F

POAPA

+- - - -- -- -- - -

+

+ +++

+SCD-1 inhib

abs

590

- 6

30

Chapter 3

90

provide an explanation to this tumour selectivity. First, tumour cells appear addicted to de

novo FA synthesis2–5 and as such continuously generate saturated FAs that need to be

converted to unsaturated FAs. It is not completely clear why tumour cells prefer this

synthetic pathway even in the presence of exogenous FAs. Except for adipocytes, this

pathway is not utilized by normal cells, which acquire their FAs from exogenous sources.38

The latter provides a stable source of unsaturated FAs, and as such alleviates the need for

desaturase activity in normal cells. Second, tumour cells have previously been shown to

already contain CL species that are relatively more saturated as compared with normal

cells.39 This could be a consequence of the apparently higher turnover of CL in tumour cells

allowing for a less intense side chain modification or, alternatively, may be caused by the

fact that mainly monounsaturated FAs are generated by de novo synthesis. Indeed,

saturated FAs are abundantly found in cancer cells and also an increased content of

monounsaturated at the expense of poly‐unsaturated FAs is observed in transformed and

cancer cells.40–42

The higher saturation of CL suggests that tumour cells are closer to an unacceptable level

of saturation than normal cells and therefore appear to be nearer to the tipping point of

detrimental mitochondrial ultrastructural modifications and subsequent mitochondrial

permeability. In agreement, BTHS patients contain a similarly low level of unsaturated FA

side chains in CL and contain the same ultrastructural changes. However, this modification

by itself is not sufficient for the actual mitochondrial leakage to occur as most cells in BTHS

patients do not undergo cell death, suggesting that additional changes are needed. These

could involve increased ROS production, which is often observed in tumour cells43,44 or a

change in the way mitochondria are utilized in tumour cells, in particular in relation to the

so‐called Warburg effect.45,46 In agreement, our data show that BetA exerts cell death more

rapidly and more vigorously than SCD‐1 inhibition, which supports the idea that BetA is

more potent in inducing cell death despite the fact that we observe more potent SCD‐1

activity inhibition with SCD‐1 inhibitors. We believe this is due to the fact that BetA

enhances the need for new FA production. In essence we think that BetA not only stops the

production of unsaturated FA, but on the other end increases the turnover of lipids and

thereby the incorporation of saturated FA, which in the case of CL results in the toxic effects

on mitochondria. This could explain the tumour selectivity as normal cells are not as

dependent on SCD‐1 and can maintain a relatively stable pool of unsaturated FA by uptake

from the environment. The nature of this enhanced lipid turnover is not completely clear,

but could for instance be a result of the massive induction of autophagy observed with

BetA,67 which generates an enormous need for de novo membrane formation and thus lipid

synthesis. Further insight into these additional changes may reveal another vulnerability

that converges on the same pathway. Clues for these changes may come from neutrophils

in BTHS patients as these display a relatively high level of apoptosis and indeed these

patients frequently suffer from neutropenia.32,47


91

The direct link between BetA‐induced cell death and mitochondria is consistent with

previous data which indicate that cell death ensues in a BAX and BAK‐independent

manner,23 but requires mitochondria and opening of the PT‐pore. In agreement,

cyclosporine A, an inhibitor of the PT‐pore, blocks cell death induced by BetA23 and data not

shown. Interestingly, recent data suggested that the FO c‐subunit ring of F1FO ATP synthase

can form a channel, which can lead to uncontrolled depolarisation and MOMP when opened

persistently.48 Whether this represents the PT‐pore remains to be confirmed, but it is

important to mention that F1FO ATP synthase as well as several of the other proposed

constituents of the PT‐pore associate with CL49 and CL can regulate PT‐pore opening.31

Similarly, cytochrome c, which is released upon BetA treatment, interacts directly with CL

through two independent binding sites.29,50–53 Moreover, Kagan et al.54 found that CL‐bound

cytochrome c acts as a peroxidase capable of catalysing H2O2‐dependent peroxidation of CL

and that this CL oxidation is an essential step in the release of cytochrome c during

apoptosis. Whether the increased CL saturation has similar effects on the release of

cytochrome c needs to be determined, but is a reasonable possibility. However, this initial

release of cytochrome c from the inner membrane is not sufficient for apoptosis to occur.

This requires mitochondrial outer membrane permeabilization either through BAX‐

dependent pore formation or through opening of the PT‐pore.31 As BetA‐induced death

depends on the latter it seems likely to assume that CL saturation also affects the opening

of the PT‐pore.

In addition to the release of cytochrome c, mitochondrial fragmentation was rapidly

observed in cells that have impaired SCD‐1 activity and therefore modified CL FA side chains.

The mitochondrial network is maintained by a continuous process of fusion (elongation of

mitochondria) and fission (fragmentation of mitochondria). It is debated whether

fusion/fission is important for cytochrome c release, but during apoptosis mitochondria are

remodelled via activation of the fission machinery and synchronal neutralization of the

fusion machinery.55,56 Importantly, this favoured mitochondrial fission occurs at the same

time as BAX activation and permeabilization of the outer mitochondrial membrane, leading

to cytochrome c release,57 suggesting that the events could be coupled. Mitochondrial

fusion/fission itself is orchestrated by proteins such as Drp1 and Opa1, but also the

structural, flexible, properties of CL.26,56,58 The observed mitochondrial fragmentation could

therefore also be a direct effect of increased CL saturation, but how this would favour

fragmentation is unclear and will require further analysis. Combined, our data point to a

rapid effect of SCD‐1 inhibition on the structural flexibility of CL and subsequently the

mitochondrial integrity. Previous observations highlighted the importance of SCD‐1 in

tumours. Neoplastic cells contain higher levels of SCD‐1 as compared with normal human

skin fibroblasts13 and SCD‐1 expression and activity are increased in several types of

cancer.16 Moreover, normal cells appear to be resistant to the cytotoxic effect of SCD‐1

inhibition.15 These observations suggests that a constantly active SCD‐1 must be present in

Chapter 3

92

cancer cells in order to maintain the level of unsaturated FAs to be incorporated in major

cell lipids and thereby protect the tumour cells from excess saturated FAs.13,15,59–62 In

agreement, knockdown of SCD‐1 or inhibition with a specific small molecule inhibitor led to

impaired biosynthesis of FAs, cholesteryl ester, triacylglycerol and phospholipid

synthesis12,15,63,64 and induced death, confirming a pivotal role for SCD‐1 in cancer.63 Several

studies have reported the use of pharmacologic inhibitors of SCD‐1 to reduce tumour

growth in preclinical cancer models.65,66 In line with our data, toxicity induced by impaired

SCD‐1 expression sensitized cancer cells to the pro‐apoptotic effects of PA.12,64 Our data

now provide mechanistic insight and indicate that inhibition of SCD‐1 rapidly changes the

side chain saturation of CL and impairs normal mitochondrial structure. We conclude that

CL is a crucial player in mitochondrial integrity that is strongly dependent on the level of

saturation of its side chains. Targeting of this saturation or potentially even CL synthase in

cancer cells provides a tumour‐selective cell death mechanism that is induced by BetA.

MATERIALS AND METHODS

Chemicals/antibodies

BetA (BioSolution, Halle, Germany, >99% purity) was dissolved in dimethylsulfoxide at 4

mg/ml and stored at − 80 °C. POA, PA, heptadecanoic acid and etoposide were purchased

from Sigma‐Aldrich (St Louis, MO, USA). SCD‐1 inhibitor (cat# 1716) was from Biovision

Chemscene (Milpitas, CA, USA). Primary antibody to cytochrome c (6H2B4) was obtained

from BD Pharmingen (Franklin Lakes, NJ, USA), secondary antibody fluorescein

isothiocyanate conjugated antibody was obtained from Beckman Coulter (Fullerton, CA,

USA). Crystal violet blue and glutaraldehyde were purchased from Merck (Darmstadt,

Germany).

Cell culture

HeLa cells, A549 cells, MCF‐7 cells, RKO cells and Co18 were obtained from the ATCC

(Manassas, VA, USA). Cells were cultured in Iscove's modified Dulbecco's medium

supplemented with 8% fetal calf serum, 2 mM L‐glutamine, 40 U/ml penicillin and 40 μg/ml

streptomycin.

RNA interference

Lentiviral vectors TRIPZ SCD small hairpin RNAs (GE Healthcare, cod. RHS4740‐EG6319,

Buckinghamshire, UK), or the non‐silencing small hairpin RNA (GE Healthcare, cod.

RHS4743) were transfected into HEK‐293 T cells using polyethylenimine (Brunschwig, The

Netherlands) according to vendor’s instructions. After 48 h and 72 h, the virus containing

medium was collected and filtered using Amicon Ultra‐15 Centrifugal Filter Unit with

Ultracel‐100 membrane (Millipore, Billerica, MA, USA). The virus were resuspended in

medium and HeLa cells were infected 24–48 h with virus containing medium supplemented

with 10 μg/ml polybrene. Transduced cell populations were subsequently selected with 1


93

μg/ml puromycin and knockdown was induced by 72 h 2 μg/ml doxycycline treatment after

which expression of the targeted proteins were determined via quantitative real‐time PCR.

RNA isolation and quantitative PCR

RNA was isolated using the Superscript III Reverse Transcriptase Kit (Invitrogen, Waltham,

MA, USA). Total RNA was retrotranscribed using Nucleospin RNA Kit (Macherey‐Nagel,

Düren, Germany). Quantitative PCR analysis was performed using the following primers:

SCD‐1‐F (5′‐TCTAG CTCCTATACCACCACCA‐3′), SCD‐1‐R (5′‐TCGTCTCCAACTTATCTCCTCC‐3′),

GAPDH‐F (5′‐AATCCCATCACCATCTTCCA‐3′) and GAPDH‐R (5′‐TGGACTCC ACGACGTACTCA‐

3′).

Cell death analysis and cytochrome c release

Cell death was determined by propidium iodide exclusion as described before.67

Cytochrome c release was performed according to the protocol of N. Waterhouse.68 Briefly,

cells were permeabilized with 50 μg/ml digitonin/phosphate‐buffered saline/KCl buffer,

fixed for 30 min with paraformaldehyde and incubated in blocking buffer (3% BSA, 0.05%

saponin, 0.02% azide and 1:200 normal goat serum in phosphate‐buffered saline). Anti‐

cytochrome c incubation was performed overnight at 4 °C, after washing, fluorescein

isothiocyanate conjugated secondary antibody was applied for 1 h at room temperature.

Electron microscope

After the indicated treatment, cells were fixed in 1% glutaraldehyde and 4%

paraformaldehyde in 0.1 M sodium cacodylate buffer (McDowell fixative) and postfixed

with 1% osmiumtetroxide (OsO4, Electron microscopy sciences, Hatfield, PA, USA) and

kaliumhexacyanoferrate (K3FE(CN)6), VWR, Amsterdam, The Netherlands) in cacodylate

buffer. Subsequently, the cells were dehydrated in an alcohol series and embedded into

Epon (LX‐112 resin Ladd Research, Williston, VT, USA). Ultrathin sections were collected on

formvar‐coated grids, counterstained with uranil acetate and lead citrate and visualized

with transmission electron microscope (FEI technai 12).

Confocal fluorescence microscopy

HeLa cells transiently transfected using polyethylenimine according to vendor’s

instructions, with MitoRFP were grown on Poly‐D‐lysine coated 24 mm diameter glass cover

slips in six‐well plates. Cells were treated with BetA alone or in combination with the

indicated concentrations of PA or POA for 18 h. A549, MCF‐7, RKO and Co18 cells were

grown on Poly‐Dlysine coated 24 mm diameter glass cover slips in six‐well plates. Cells were

treated with vehicle or 10 μg/ml BetA for 18 h and stained 30 min with 100 nM mitotracker

red CMXROS. The glass cover slips were mounted and placed in a Leica DMI 6000 (TCS SP8)

microscope (with adaptive focus, Motorized XY‐Stage and SuperZ Galvo) and a case

Chapter 3

94

incubator at 37 °C for Z‐stack measurements. Samples were analysed using Leica Las AF

software (Leica Microsystems, Wetzlar, Germany).

Mass spectrometry measurements of CL

HeLa and A549 cells were harvested by trypsin and centrifugation at 300 g for 5 min at room

temperature, washed once with phosphate‐buffered saline, pelleted by centrifugation at

300 g for 5 min at room temperature. CL was analysed essentially as described previously.33

The relative abundances of the species in the sample‐extracts were determined by high‐

performance liquid chromatography‐mass spectrometry using a Surveyor high‐

performance liquid chromatography system hyphenated to a TSQ Quantum AM tandem

mass spectrometer (Thermo Finnigan Corporation, San Jose, CA, USA). The MS was

operated in the negative ion electrospray ionisation mode. Data were analysed using

Xcalibur software (Thermo Fisher Scientific, Waltham, MA, USA).

Desaturase activity measurements

Cells were treated for 18 h with 10 μg/ml BetA or 150 nM SCD‐1 inhibitor after which 50

μM 2H3‐palmitate (2H3‐C16:0; Cambridge Isotope Laboratories) or for HeLa cells, 50 μM 2H5‐

stearate (2H5‐C18:0; Cambridge Isotope Laboratories) was added for an additional 6 h.

Conversion of these saturated deuterium‐containing FAs to unsaturated 2H3‐C16:1 or 2H5‐

C18:1 was measured as described before.69,70

Clonogenic growth

HeLa cells were seeded in a 24‐well plate and treated with 10 μg/ml BetA, indicated

concentrations of SCD‐1 inhibitor or with 1 μg/ml etoposide, with or without 50 μM POA or

50 μM PA for 48 h (BetA) or 72 h (SCD‐1 inhibitor). After treatment, medium was changed

and cells were left to grow for an additional 2 days. Cells were fixed in 6% glutaraldehyde

and stained with crystal violet blue. Staining was dissolved in 0.5 ml 99% ethanol for 1 h

with mild shaking. 100 μl of dissolved crystal violet staining was transferred to a 96‐well

plate and absorbance (590–630 nm) was measured (BioTek, Winooski, VT, USA).

Statistical analysis

Statistical analyses were performed with Prism 5 (GraphPad software, La Jolla, CA, USA)

applying a 2‐tailed unpaired Student t‐test. Differences were considered significant when

P‐values, *P<0.05; **P<0.01; ***P<0.001.


95

Conflict of interest


Acknowledgements

We thank Inge Dijkstra, Rob Ofman, Femke Beers‐Stet for technical assistance and Louis

Vermeulen for his valuable scientific input and helpful discussions. This work was supported

by a grant of the Stichting Nationaal Fonds tegen Kanker (SNFK), Amsterdam, The

Netherlands (LP).

Chapter 3

96

References

1. Swinnen JV, Brusselmans K, Verhoeven G. Increased lipogenesis in cancer cells: new players,novel targets. Curr Opin Clin Nutr Metab Care 2006 Jul;9(4):358‐65.

2. DeBerardinis RJ, Thompson CB. Cellular metabolism and disease: what do metabolic outliers teach us? Cell 2012 Mar 16;148(6):1132‐44. 3. Santos CR, Schulze A. Lipid metabolism in cancer. FEBS J 2012 Aug;279(15):2610‐23. 4. Medes G, Thomas A, Weinhouse S. Metabolism of neoplastic tissue. IV. A study of lipid synthesis in neoplastic tissue slices in vitro. Cancer Res 1953 Jan;13(1):27‐9. 5. Ookhtens M, Kannan R, Lyon I, Baker N. Liver and adipose tissue contributions to newly formed fatty acids in an ascites tumor. Am J Physiol 1984 Jul;247(1 Pt 2):R146‐R153. 6. Menendez JA, Lupu R. Fatty acid synthase and the lipogenic phenotype in cancer pathogenesis. Nat Rev Cancer 2007 Oct;7(10):763‐77. 7. Li JN, Mahmoud MA, Han WF, Ripple M, Pizer ES. Sterol regulatory element‐binding protein‐1 participates in the regulation of fatty acid synthase expression in colorectal neoplasia. Exp Cell Res 2000 Nov 25;261(1):159‐65. 8. Swinnen JV, Vanderhoydonc F, Elgamal AA, Eelen M, Vercaeren I, Joniau S, et al. Selective activation of the fatty acid synthesis pathway in human prostate cancer. Int J Cancer 2000 Oct 15;88(2):176‐9. 9. Bauer DE, Hatzivassiliou G, Zhao F, Andreadis C, Thompson CB. ATP citrate lyase is an important component of cell growth and transformation. Oncogene 2005 Sep 15;24(41):6314‐22. 10. Hatzivassiliou G, Zhao F, Bauer DE, Andreadis C, Shaw AN, Dhanak D, et al. ATP

citrate lyase inhibition can suppress tumor cell growth. Cancer Cell 2005 Oct;8(4):311‐21. 11. Kumar‐Sinha C, Ignatoski KW, Lippman ME, Ethier SP, Chinnaiyan AM. Transcriptome analysis of HER2 reveals a molecular connection to fatty acid synthesis. Cancer Res 2003 Jan 1;63(1):132‐9. 12. Scaglia N, Igal RA. Stearoyl‐CoA desaturase is involved in the control of proliferation, anchorage‐independent growth, and survival in human transformed cells. J Biol Chem 2005 Jul 8;280(27):25339‐49. 13. Scaglia N, Caviglia JM, Igal RA. High stearoyl‐CoA desaturase protein and activity levels in simian virus 40 transformed‐human lung fibroblasts. Biochim Biophys Acta 2005 Feb 21;1687(1‐3):141‐51. 14. Paton CM, Ntambi JM. Biochemical and physiological function of stearoyl‐CoA desaturase. Am J Physiol Endocrinol Metab 2009 Jul;297(1):E28‐E37. 15. Scaglia N, Chisholm JW, Igal RA. Inhibition of stearoylCoA desaturase‐1 inactivates acetyl‐CoA carboxylase and impairs proliferation in cancer cells: role of AMPK. PLoS One 2009;4(8):e6812. 16. Igal RA. Stearoyl‐CoA desaturase‐1: a novel key player in the mechanisms of cell proliferation, programmed cell death and transformation to cancer. Carcinogenesis 2010 Sep;31(9):1509‐15. 17. Zuco V, Supino R, Righetti SC, Cleris L, Marchesi E, Gambacorti‐Passerini C, et al. Selective cytotoxicity of betulinic acid on tumor cell lines, but not on normal cells. Cancer Lett 2002 Jan 10;175(1):17‐25. 18. Rzeski W, Stepulak A, Szymanski M, Sifringer M, Kaczor J, Wejksza K, et al. Betulinic acid decreases expression of bcl‐2 and cyclin D1, inhibits proliferation, migration and induces apoptosis in cancer cells. Naunyn


97

Schmiedebergs Arch Pharmacol 2006 Oct;374(1):11‐20. 19. Mullauer FB, Kessler JH, Medema JP. Betulinic acid, a natural compound with potent anticancer effects. Anticancer Drugs 2010 Mar;21(3):215‐27. 20. Fulda S, Friesen C, Los M, Scaffidi C, Mier W, Benedict M, et al. Betulinic acid triggers CD95 (APO‐1/Fas)‐ and p53‐independent apoptosis via activation of caspases in neuroectodermal tumors. Cancer Res 1997 Nov 1;57(21):4956‐64. 21. Fulda S, Susin SA, Kroemer G, Debatin KM. Molecular ordering of apoptosis induced by anticancer drugs in neuroblastoma cells. Cancer Res 1998 Oct 1;58(19):4453‐60. 22. Kessler JH, Mullauer FB, de Roo GM, Medema JP. Broad in vitro efficacy of plant‐derived betulinic acid against cell lines derived from the most prevalent human cancer types. Cancer Lett 2007 Jun 18;251(1):132‐45. 23. Mullauer FB, Kessler JH, Medema JP. Betulinic acid induces cytochrome c release and apoptosis in a Bax/Bak‐independent, permeability transition pore dependent fashion. Apoptosis 2009 Feb;14(2):191‐202. 24. Fry M, Green DE. Cardiolipin requirement for electron transfer in complex I and III of the mitochondrial respiratory chain. J Biol Chem 1981 Feb 25;256(4):1874‐80. 25. Gonzalvez F, Gottlieb E. Cardiolipin: setting the beat of apoptosis. Apoptosis 2007 May;12(5):877‐85. 26. Houtkooper RH, Vaz FM. Cardiolipin, the heart of mitochondrial metabolism. Cell Mol Life Sci 2008 Aug;65(16):2493‐506. 27. Patil VA, Greenberg ML. Cardiolipin‐mediated cellular signaling. Adv Exp Med Biol 2013;991:195‐213.

28. Gonzalvez F, Schug ZT, Houtkooper RH, MacKenzie ED, Brooks DG, Wanders RJ, et al. Cardiolipin provides an essential activating platform for caspase‐8 on mitochondria. J Cell Biol 2008 Nov 17;183(4):681‐96. 29. Schug ZT, Gottlieb E. Cardiolipin acts as a mitochondrial signalling platform to launch apoptosis. Biochim Biophys Acta 2009 Oct;1788(10):2022‐31. 30. Zhang T, Saghatelian A. Emerging roles of lipids in BCL‐2 family‐regulated apoptosis. Biochim Biophys Acta 2013 Oct;1831(10):1542‐54. 31. Paradies G, Paradies V, De B, V, Ruggiero FM, Petrosillo G. Functional role of cardiolipin in mitochondrial bioenergetics. Biochim Biophys Acta 2014 Apr;1837(4):408‐17. 32. Barth PG, Scholte HR, Berden JA, Van der Klei‐Van Moorsel JM, Luyt‐Houwen IE, Van 't Veer‐Korthof ET, et al. An X‐linked mitochondrial disease affecting cardiac muscle, skeletal muscle and neutrophil leucocytes. J Neurol Sci 1983 Dec;62(1‐3):327‐55. 33. Houtkooper RH, Rodenburg RJ, Thiels C, van LH, Stet F, Poll‐The BT, et al. Cardiolipin and monolysocardiolipin analysis in fibroblasts, lymphocytes, and tissues using high‐performance liquid chromatography‐mass spectrometry as a diagnostic test for Barth syndrome. Anal Biochem 2009 Apr 15;387(2):230‐7. 34. Ren M, Phoon CK, Schlame M. Metabolism and function of mitochondrial cardiolipin. Prog Lipid Res 2014 Apr 24;55C:1‐16. 35. Shimabukuro M, Zhou YT, Levi M, Unger RH. Fatty acid‐induced beta cell apoptosis: a link between obesity and diabetes. Proc Natl Acad Sci U S A 1998 Mar 3;95(5):2498‐502. 36. Listenberger LL, Han X, Lewis SE, Cases S, Farese RV, Jr., Ory DS, et al. Triglyceride accumulation protects against fatty acid‐

Chapter 3

98

induced lipotoxicity. Proc Natl Acad Sci U S A 2003 Mar 18;100(6):3077‐82. 37. Mullauer FB, van BL, Daalhuisen JB, Ten Brink MS, Storm G, Medema JP, et al. Betulinic acid delivered in liposomes reduces growth of human lung and colon cancers in mice without causing systemic toxicity. Anticancer Drugs 2011 Mar;22(3):223‐33. 38. Coleman RA, Lee DP. Enzymes of triacylglycerol synthesis and their regulation. Prog Lipid Res 2004 Mar;43(2):134‐76. 39. Kiebish MA, Han X, Cheng H, Chuang JH, Seyfried TN. Cardiolipin and electron transport chain abnormalities in mouse brain tumor mitochondria: lipidomic evidence supporting the Warburg theory of cancer. J Lipid Res 2008 Dec;49(12):2545‐56. 40. Ruggieri S, Roblin R, Black PH. Lipids of whole cells and plasma membrane fractions from Balb/c3T3, SV3T3, and concanavalin A‐selected revertant cells. J Lipid Res 1979 Aug;20(6):760‐71. 41. Bougnoux P, Chajes V, Lanson M, Hacene K, Body G, Couet C, et al. Prognostic significance of tumor phosphatidylcholine stearic acid level in breast carcinoma. Breast Cancer Res Treat 1992 Mar;20(3):185‐94. 42. Horie Y, Suzuki A, Kataoka E, Sasaki T, Hamada K, Sasaki J, et al. Hepatocyte‐specific Pten deficiency results in steatohepatitis and hepatocellular carcinomas. J Clin Invest 2004 Jun;113(12):1774‐83. 43. Pelicano H, Carney D, Huang P. ROS stress in cancer cells and therapeutic implications. Drug Resist Updat 2004 Apr;7(2):97‐110. 44. Nogueira V, Hay N. Molecular pathways: reactive oxygen species homeostasis in cancer cells and implications for cancer therapy. Clin Cancer Res 2013 Aug 15;19(16):4309‐14.

45. Warburg O. On respiratory impairment in cancer cells. Science 1956 Aug 10;124(3215):269‐70. 46. Ward PS, Thompson CB. Metabolic reprogramming: a cancer hallmark even warburg did not anticipate. Cancer Cell 2012 Mar 20;21(3):297‐308. 47. Clarke SL, Bowron A, Gonzalez IL, Groves SJ, Newbury‐Ecob R, Clayton N, et al. Barth syndrome. Orphanet J Rare Dis 2013;8:23. 48. Alavian KN, Beutner G, Lazrove E, Sacchetti S, Park HA, Licznerski P, et al. An uncoupling channel within the c‐subunit ring of the F1FO ATP synthase is the mitochondrial permeability transition pore. Proc Natl Acad Sci U S A 2014 Jul 22;111(29):10580‐5. 49. Acehan D, Malhotra A, Xu Y, Ren M, Stokes DL, Schlame M. Cardiolipin affects the supramolecular organization of ATP synthase in mitochondria. Biophys J 2011 May 4;100(9):2184‐92. 50. Cory S, Adams JM. The Bcl2 family: regulators of the cellular life‐or‐death switch. Nat Rev Cancer 2002 Sep;2(9):647‐56. 51. Tuominen EK, Wallace CJ, Kinnunen PK. Phospholipid‐cytochrome c interaction: evidence for the extended lipid anchorage. J Biol Chem 2002 Mar 15;277(11):8822‐6. 52. Orrenius S, Zhivotovsky B, Nicotera P. Regulation of cell death: the calcium‐apoptosis link. Nat Rev Mol Cell Biol 2003 Jul;4(7):552‐65. 53. Huttemann M, Pecina P, Rainbolt et al. The multiple functions of cytochrome c and their regulation in life and death decisions of the mammalian cell: From respiration to apoptosis. Mitochondrion 2011 May;11(3):369‐81. 54. Kagan VE, Tyurin VA, Jiang J, Tyurina YY, Ritov VB, Amoscato AA, et al. Cytochrome c acts as a cardiolipin oxygenase required for


99

release of proapoptotic factors. Nat Chem Biol 2005 Sep;1(4):223‐32. 55. Jourdain A, Martinou JC. Mitochondrial outer‐membrane permeabilization and remodelling in apoptosis. Int J Biochem Cell Biol 2009 Oct;41(10):1884‐9. 56. Landes T, Martinou JC. Mitochondrial outer membrane permeabilization during apoptosis: the role of mitochondrial fission. Biochim Biophys Acta 2011 Apr;1813(4):540‐5. 57. Martinou JC, Youle RJ. Which came first, the cytochrome c release or the mitochondrial fission? Cell Death Differ 2006 Aug;13(8):1291‐5. 58. Pellegrini L, Scorrano L. A cut short to death: Parl and Opa1 in the regulation of mitochondrial morphology and apoptosis. Cell Death Differ 2007 Jul;14(7):1275‐84. 59. Li J, Ding SF, Habib NA, Fermor BF, Wood CB, Gilmour RS. Partial characterization of a cDNA for human stearoyl‐CoA desaturase and changes in its mRNA expression in some normal and malignant tissues. Int J Cancer 1994 May 1;57(3):348‐52. 60. Lu J, Pei H, Kaeck M, Thompson HJ. Gene expression changes associated with chemically induced rat mammary carcinogenesis. Mol Carcinog 1997 Oct;20(2):204‐15. 61. Thai SF, Allen JW, DeAngelo AB, George MH, Fuscoe JC. Detection of early gene expression changes by differential display in the livers of mice exposed to dichloroacetic acid. Carcinogenesis 2001 Aug;22(8):1317‐22. 62. Yahagi N, Shimano H, Hasegawa K, Ohashi K, Matsuzaka T, et al. Co‐ordinate activation of lipogenic enzymes in hepatocellular carcinoma. Eur J Cancer 2005 Jun;41(9):1316‐22. 63. Morgan‐Lappe SE, Tucker LA, Huang X, Zhang Q, Sarthy AV, Zakula D, et al.

Identification of Ras‐related nuclear protein, targeting protein for xenopus kinesin‐like protein 2, and stearoyl‐CoA desaturase 1 as promising cancer targets from an RNAi‐based screen. Cancer Res 2007 May 1;67(9):4390‐8. 64. Scaglia N, Igal RA. Inhibition of Stearoyl‐CoA Desaturase 1 expression in human lung adenocarcinoma cells impairs tumorigenesis. Int J Oncol 2008 Oct;33(4):839‐50. 65. Fritz V, Benfodda Z, Rodier G, Henriquet C, Iborra F, Avances C, et al. Abrogation of de novo lipogenesis by stearoyl‐CoA desaturase 1 inhibition interferes with oncogenic signaling and blocks prostate cancer progression in mice. Mol Cancer Ther 2010 Jun;9(6):1740‐54. 66. Roongta UV, Pabalan JG, Wang X, Ryseck RP, Fargnoli J, Henley BJ, et al. Cancer cell dependence on unsaturated fatty acids implicates stearoyl‐CoA desaturase as a target for cancer therapy. Mol Cancer Res 2011 Nov;9(11):1551‐61. 67. Potze L, Mullauer FB, Colak S, Kessler JH, Medema JP. Betulinic acid‐induced mitochondria‐dependent cell death is counterbalanced by an autophagic salvage response. Cell Death Dis 2014;5:e1169. 68. Waterhouse NJ, Trapani JA. A new quantitative assay for cytochrome c release in apoptotic cells. Cell Death Differ 2003 Jul;10(7):853‐5. 69. Valianpour F, Selhorst JJ, van Lint LE, van Gennip AH, Wanders RJ, Kemp S. Analysis of very long‐chain fatty acids using electrospray ionization mass spectrometry. Mol Genet Metab 2003 Jul;79(3):189‐96. 70. Engelen M, Ofman R, Mooijer PA, Poll‐The BT, Wanders RJ, Kemp S. Cholesteroldeprivation increases mono‐unsaturated very long‐chain fatty acids in skin fibroblasts from patients with X‐linked adrenoleukodystrophy. Biochim Biophys Acta 2008; 3: 105–111.

Chapter 3

100


Supplementary Figure 1: BetA‐induced mitochondrial morphology changes in cancer cells and not in

normal cells. Indicated cell lines were treated with vehicle (DMSO) or 10 µg/ml BetA for 18 h after which

cells were stained with 100 nM mitotracker red CMXROS and analysed by confocal microscopy.

Supplementary Figure 2: Induction of CL saturation upon BetA treatment. MCF‐7 cells were subjected to

vehicle (DMSO) or 10 µg/ml BetA for 24 h after which CL content was measured. Relative C68 CL saturation

is shown, which is calculated as (C68:2 + C68:3) / (C68:4). Mean ± SD of 2 experiments are shown. * P< 0.05;

ns, not significant. (B) MCF‐7 cells were subjected to vehicle (DMSO) or 10 µg/ml BetA for 24 h after which

CL content was measured. Quantification of CL C68 cluster is shown. Mean ± SD of 2 experiments are shown.

A549 MCF-7 RKO CO18

Bet

AD

MS

O

rela

tive

CL

satu

ratio

n

0.0

0.2

0.4

0.6

0.8

BetA - +0.0

0.5

1.0

1.5 CL (68:4)

CL (68:3)

CL (68:2)

DMSO BetA

nmol

/mg

prot

ein

BA*


101

Supplementary Figure 3: Heptadecanoic acid incorporates into CL. HeLa cells were pre‐treated with vehicle

(DMSO) or 50 µM heptadecanoic acid for 4 h after which vehicle (DMSO) (A) or 10 µg/ml BetA (B) was

applied for 18 h followed by CL analysis by HPLC mass spectrometry. Note the higher abundance of the C17

containing PG (33:0) compared to that of PG (33:1) in BetA treated cells.

Supplementary Figure 4:

Desaturase activity and levels in

cancer cells. (A) Indicated cell lines

were analysed for their relative

SCD‐1 mRNA levels by qPCR. (B)

HeLa cells with SCD‐1 shRNA were

treated with doxycycline for 72 h to

induce knockdown of SCD‐1. Levels

of SCD‐1 mRNA levels were

measured by qPCR. (C) Indicated

cell lines were treated with vehicle

(DMSO), 150 nM SCD‐1 inhibitor or

10 µg/ml BetA for 18 h after which 2H3‐C16:0 was applied for

additional 6 h. Conversion to 2H3‐

C16:1 was measured. Mean ± SD of

3 experiments are shown.

***P<0.001, **P<0.01.

670 680 690 700 710 720 730 740

20

100

0

40

60

80

rela

tive

inte

nsity

(%

)

686.5685.5 687.5

698.5694.4

693.4

699.5700.5

701.5

711.5

712.5

713.5

714.5

715.5

718.5

719.5

720.5721.5

722.5728.5

725.5

726.5

727.5

C74

C72

C70

C68

C69C71

C6620

100

0

40

60

80

rela

tive

inte

nsity

(%

)

DMSO + C17AAA

B BetA+ C17

C64

C67

C65

707.5706.5

708.5

705.5692.4

686.5685.5 687.5

698.5694.4

693.4

699.5

700.5

701.5 711.5

712.5

713.5

714.5

715.5

718.5

719.5

720.5

721.5

722.5728.5

725.5

726.5

727.5

C74

C72

C70

C68

C69 C71

C66

C64

C67

C65

707.5

706.5

708.5

705.5

692.4

PG(33:0)

PG(33:1)

PG(33:1)PG(33:0)

ratio

2 H3-

C16

:1/2 H

3- C

16:0

SCD-1 inhibBetA

+-- -

-+

0.20

0.00

0.05

0.10

0.15

0.25

+-- -

-+

+-- -

-+

+-- -

-+

A549 MCF-7 RKO CO18

******

******

*****

******

0.0

0.5

1.0

1.5

f

old

cha

nge

inS

CD

-1 m

RN

A le

vels

shSCD1H12dox

++-

+

non silencing

- +- +-

-- - - -

- -- -- -

++++

0.0

0.2

0.4

0.6

0.8

A54

9

HeL

a

MC

F-7

RK

O

CO

18

rela

tive

mR

NA

leve

ls

(SC

D-1

/GA

PD

H)

shSCD1D6

A B

C

Chapter 3

102

Supplementary Figure 5: SCD‐1 inhibitor induced mitochondrial morphology changes. HeLa cells with

MitoRFP were subjected to vehicle (DMSO), 200 nM SCD‐1 inhibitor with or without 50 µM POA or 50 µM

PA for 40 h after which cells were analysed by confocal microscopy. Representative pictures of 2

experiments are shown.

0 25µm 0 25µm 0 25µm

DMSO POA PA

SCD-1 inhib

0 μm 25 0 μm 25 0 μm 25

ZO

OM

ZO

OM

ZO

OM

SCD-1 inhib + POA SCD-1 inhib + PA

ZO

OM

ZO

OM

ZO

OM


103

Supplementary Figure 6: CL levels of BetA and SCD‐1 inhibitor treated cells (A) HeLa cells were subjected

to vehicle (DMSO) or 10 µg/ml BetA with or without 50 µM POA or 50 µM PA for 18 h after which CL content

was measured. Quantification of CL C68 cluster is shown. Mean ± SD of 5 experiments are shown (B) A549

cells were subjected to vehicle (DMSO) or 10 µg/ml BetA with or without 50 µM POA or 50 µM PA for 18 h

after which CL content was measured. Relative C68 CL saturation of A549 cells is shown, which is calculated

as (C68:2 + C68:3) / (C68:4 + C68:5). Mean ± SD of 3 experiments are shown. * P< 0.05; ns, not significant.

(C) A549 cells were subjected to vehicle (DMSO) or 10 µg/ml BetA with or without 50 µM POA or 50 µM PA

for 18 h after which CL content was measured. Quantification of CL C68 cluster is shown. Mean ± SD of 2

experiments are shown. (D) HeLa cells were subjected to vehicle (DMSO) or 200 nM SCD‐1 inhibitor with or

without 50 µM POA or 50 µM PA for 40 h after which CL content was measured. Quantification of the CL

C68 cluster is shown. Mean ± SD of 3 experiments are shown.

CL (68:5)

CL (68:4)

CL (68:3)

CL (68:2)

0.0

0.2

0.4

0.6

0.8

1.0

nm

ol/m

g pr

ote

in

BetA

POA

PA

- + ++ +

+ + +

- - -----

- - -

A

C

D

0

20

40

60CL (68:5)

CL (68:4)

CL (68:3)

CL (68:2)

%C

Lfr

omto

talC

L

POA

PA

- + ++ +

+ + +

- - -----

- - -SCD-1 inhib

nm

ol/m

g pr

ote

in

BetA

POA

PA

- + ++ +

+ + +

- - -----

- - -

CL (68:5)

CL (68:4)

CL (68:3)

CL (68:2)

0.0

0.5

1.0

1.5

B

0.0

0.5

1.0

1.5

rela

tive

CL

satu

ratio

n

POAPA

BetA

+- - - -- -- -- - -

+

+ +++

+

*

ns

Chapter 3

104

Supplementary Figure 7: Effect of POA and PA on mitochondria (A) HeLa cells with MitoRFP were subjected

to vehicle (DMSO), 50 µM POA or 50 µM PA for 18h after which cells were analysed via confocal microscopy.

Pictures of 3 individual experiments are shown. (B) HeLa cells were treated with 50 µM POA or 50 µM PA

for 18h after which cells were analysed via TEM.

0 25µm 0 25µm 0 25µm

DMSO POA PAZ

OO

M

ZO

OM

ZO

OM

DMSO POA PA

500nm500nm500nm

A

B


105

Supplementary Figure 8: CL modification evokes cell death and decreases clonogenic survival (A) HeLa

cells with induced knockdown of SCD‐1 D6 and H12 were left for 5 days, after which cell death was assessed

via PI exclusion. (B) HeLa cells were treated with indicated concentrations of SCD‐1 inhibitor for 96 h after

which cell death was assessed via PI exclusion. (C) HeLa cells were treated with vehicle (DMSO), 10 µg/ml

BetA, 75nM SCD‐1 inhibitor (+), 200 nM SCD‐1 inhibitor (++) or combinations of BetA and SCD‐1 inhibitor

for 96h after which cell death was assessed via PI exclusion. (D) HeLa cells were treated for 48 h with vehicle

(DMSO) or 1 µg/ml etoposide with or without 50 µM POA or 50 µM PA after which fresh medium was applied

for 72 h. Cells were stained with crystal violet blue for 6 h. Mean ± SD of 3 experiments are shown.

A

1

0

2

3

POAPA

+- - - -- -- -- - -

+

+ +++

+etoposide

abs

590

- 630

B

BetASCD-1 inhib

+- - - +- ++- +

++++

0

20

40

60

80

100

% P

Ipo

s

0

20

40

60

80

100

% P

Ipo

s

0 0.075 0.2 0.5 1μM SCD-1 inhibitor

CD

% P

Ipo

s

shSCD1H12dox

++

+

non silencing

- +-

- -- - ++

0

5

10

15

20

25

Chapter 4

Betulinic acid induces a rapid form of cell

death in colon cancer stem cells

Lisette Potze, Simone di Franco, Jan H. Kessler, Giorgio Stassi,

Jan Paul Medema

Under revision at Stem Cell International

Door: Why it’s simply impassible!

Alice: Why? Don’t you mean impossible?

Door: No, I do mean impassible (chuckles) Nothing’s impossible!

Lewis Carroll, Alice in Wonder land

Chapter 4

108

Abstract

Cancer stem cells (CSCs) are considered to be the origin of cancer and it is suggested that

they are resistant to chemotherapy. Current therapies fail to eradicate the CSCs and

therefore select a resistant cell subset that is able to induce tumor recurrence. Betulinic acid

(BetA) is a broad acting natural compound, shown to induce cell death via inhibition of the

stearoyl‐CoA‐ desaturase (SCD‐1). This enzyme converts saturated fatty acids into

unsaturated fatty acids and is over‐expressed in tumor cells. Here we show that BetA

induces rapid cell death in all colon CSCs tested and is able to affect the CSCs directly as

shown via the loss of clonogenic capacity. Similar results were observed with inhibition of

SCD‐1, suggesting that SCD‐1 activity is indeed a vulnerable link in colon CSCs and can be

considered an ideal target for therapy in colon cancer.

BetA‐induced cell death in colon cancer stem cells

109

Introduction

Colorectal cancer (CRC) is one of the major causes of death worldwide(1) and the third

leading cause of cancer death in the western world (WHO 2014, World Cancer Report 2014).

Despite different and intensive treatments, current therapies are inadequate in eradicating

the disease due to the occurrence of resistance. It has been suggested that such resistance

depends on the intra‐tumoral heterogeneity, in which the so‐called cancer stem cells (CSCs)

play a fundamental role.(2) This cell subset is characterized by self‐renewal and

differentiation capacity(3, 4) and it is regulated by a coordination of several pathways and

signals responsible for the maintenance of such phenotype, as WNT(5‐8), BMP(9‐12) and

Notch.(13‐15) CSCs presence was confirmed by several studies in different solid tumors,

including colon cancer.(16‐19)

The resistance of colon CSCs to conventional therapy(17, 20‐22) makes these cells an obvious

target to optimize current therapies. Strategies to identify and target specific cell surface

markers are currently under investigation, with a particular focus on a functional

characterization of the CSCs.(3, 23‐26) Despite more than two decades of research on the

genetics of CRC(27), it appears that we have only uncovered the tip of the iceberg.(28) The

extensive inter‐ and intra‐tumor heterogeneity both genetically and therapeutically still

forms an enormous challenge. We are only starting to understand the role and nature of

genetic and epigenetic changes, the tumor microenvironment and the effects of different

cell subsets within the tumor mass.(29‐32)

Current therapeutic approaches in colon cancer mainly utilize the proliferative capacity of

tumor cells as a means to target the cancer. As cells can escape this type of therapy through

multiple mechanisms, novel broad‐acting compounds that target a distinct, but essential

feature of the cancer cells are required. Such a therapy is able to target all the cells

belonging to the tumor bulk, including the CSCs. A promising compound is represented by

Betulinic Acid (BetA), which is a plant‐derived compound that shows selective cytotoxicity

for tumor cells sparing healthy cells.(33‐35) We have previously shown that BetA is very potent

and effective in a wide range of tumor cells.(24, 34, 36, 37) More importantly, we recently

showed that BetA induces cell death via the inhibition of stearoyl‐CoA desaturase (SCD‐1)

activity, an enzymatic activity that appears crucial to cancer but not normal cells.(38) This

enzyme converts saturated fatty acids derived from the de novo fatty acid synthesis

pathway into monounsaturated fatty acids and as such provides building blocks for new

lipid biomass. The saturation grade of fatty acids in cells is crucial to maintain optimal

fluidity of cellular membranes and it has been shown that inhibition of SCD‐1 induces cell

death in a variety of cells including lung CSCs.(39‐41) Due to its broad and potent activity we

investigated the role of BetA in colon CSCs and also evaluated the effect of SCD‐1 inhibition

on these cells. Here we show that BetA and SCD‐1 inhibition both induces cell death and

loss of clonogenicity in colon CSCs. BetA induces a more rapid cell death in these cells

Chapter 4

110

compared to SCD‐1 inhibitor, thus suggesting that BetA is a very potent compound for

cancer therapy.

Results

BetA induces a rapid cell death in colon cancer stem cells.

BetA has been shown to be very potent in killing several cancer cells after 48h with 60‐90 %

cell death in a p53 and BAX/BAK‐independent manner(36, 38) (see Table 1 for mutation

analysis of colon CSCs). To determine whether BetA was also effective against colon CSCs,

we used a panel of CSC cultures that were characterized for the most common stemness

markers CD166, CD133(16) and CD44v6(3), and CK20, a colonic differentiation marker(42) (Fig.

1A). We treated these spheroid cultures with BetA and observed a very rapid and potent

induction of cell death. Within 2 h about 60 % of cell death was detected in BetA‐treated

GTG7 cells, while this resulted in 30 % and 40 % cell death in CC09 and Co147 cells

respectively. Colon CSCs death induction is concentration‐ and time‐ dependent with cell

death increasing to 60‐80 % cell death after 8 h using the highest BetA concentration (Fig.

1B‐G). Importantly, the rapid cell death is not dependent on culture conditions, but seemed

to be selective for these spheroid cultured CSC cultures as normal colon cancer cell lines,

such as RKO cells, only showed around 20% cell death at these early time points

(Supplementary Fig. 1A‐B). According with previous findings(43), we observed BetA‐induced

cell death in both p53 mutant and p53 wild‐type cell lines (Table 1).

Colon cancer stem cells lose their clonogenic capacity upon BetA treatment

One of the hallmarks of CSCs is represented by their resistance to therapy. Most

conventional therapies induce cell death in differentiated tumor cells, but not in CSCs.(44) As

our spheroid cultures represent a mixture of CSCs and more differentiated tumor cells, we

set out to directly analyze the effect of BetA on CSCs using WNT signaling activity as a read‐

out for stemness.(5) Spheroid cells were therefore sorted for 10% low GFP, representing the

more differentiated tumor cells, 10% high GFP, representing CSCs, and the total population.

In all three cell populations BetA induced a rapid cell death suggesting that BetA targets

both the more differentiated cancer cells as well as the CSCs (Figure 2A‐D).

Table 1. Mutation status of colon CSCs

PI3KCA APC BRAF SMAD4 KRAS P53

GTG7c.1258T>C

(p.Cys420Arg)c.2413C>T

(p.Arg805Ter)wt

c.1607T>G(p.Leu536Arg)

c.38G>A(p.Gly13Asp)

wt

CC09c.344G>A

(p.Arg115Gln)c.2557G>T

(p.Glu853Ter)c.1799T>A

(p.Val600Glu)c.931C>T

(p.Gln311Ter)wt

c.818G>A(p.Arg273His)

CO147c.1090G>C

(p.Gly364Arg)c.2839T>C

(p.Cys947Arg)c.1799T>A

(p.Val600Glu)wt wt

c.733G>A(p.Gly245Ser)


111

To confirm the cytotoxic effect on CSCs we looked into the functional impact of BetA

treatment. One of the important features of CSCs is their ability to self‐renew and to give

arise to a heterogeneous population of CSCs and more differentiated cells. In other words,

a single CSC is capable of forming a new colony with all cell types present.(16) This capacity

is present in the TOP‐GFP high cells within the spheroids and defines the true CSCs. We

therefore analyzed three distinct colon CSC cultures for their clonogenic capacity after

treatment with different concentrations of BetA. BetA pre‐treatment (2 h, single hit) clearly

showed a strong reduction on stem cell frequency (60‐80 % of reduction) in all the colon

CSC cultures, in a dose‐dependent manner (Figure 2E‐G). In contrast, these short treatments

did not affect the clonogenicity of RKO cells, confirming that BetA indeed targeted colon

CSCs very effectively. (Supplementary Fig. 2) A separate measure of stemness is to

determine the effect of therapy at later time points. We therefore analyzed the long‐term

effect of conventional chemotherapy (oxaliplatin) in comparison with BetA treatment on

survival and regrowth of colon CSCs, using a CellTiter‐Blue® cell viability assay over a longer

period. Treatment of two independent colon CSC cultures with oxaliplatin showed that

despite an initial decrease in cell number, cultures eventually overcame the negative impact

of chemotherapy and expanded over time, indicating that CSCs did not succumb to therapy

even when higher doses of oxaliplatin were used (Figure 2H‐I and Supplementary Fig. 3A‐

D). In contrast, treatment with increasing doses of BetA for only two hours revealed that

long term expansion was completely annihilated at the highest concentrations of BetA

(Figure 2H‐I). Differential sensitivity was observed between the two CSC cultures, similar to

the short‐term toxicity assays with GTG7 being the most BetA sensitive cell line tested. In

these cultures BetA concentrations of 5 µg/ml completely prevented outgrowth of cells,

whereas in CC09, which were less sensitive to BetA, only the highest concentration of BetA

(10 µg/ml) prevented outgrowth completely (Figure 2H‐ I and Supplementary Fig. 3 E‐F).

These data suggest that BetA affects colon CSCs viability and consequently long‐term

clonogenicity, while conventional therapy, such as oxaliplatin, gave rise to CSC‐based

resistance phenomena and subsequent recurrences in terms of growth.

Inhibition of stearoyl CoA desaturase leads to cell death and loss of clonogenicity

in colon CSCs

We recently showed that BetA functions through inhibition of the enzymatic activity of SCD‐

1.(38) In agreement, BetA treatment, SCD‐1 inhibition or genetic knock down of SCD‐1 in

cancer cells, affect the saturation grade of fatty acids and specifically cardiolipin, resulting

in impairment of the mitochondrial structure with consequent cell death.(38) To validate

whether BetA‐induced death of colon CSCs is due to a similar SCD‐1‐dependent mechanism,

Chapter 4

112

Figure 1. BetA‐induced cell death in colon cancer stem cells. Flow cytometry analysis (A) of CD166, CD166,

CD44v6 and CK20 in GTG7, CC09 and CO147 colon CSC lines (black histograms represent isotype control).

GTG7 (B, C), CC09 (D, E) and CO147 (F, G) colon CSCs were treated for 2 h, 4 h or 8 h with indicated

concentrations of BetA after which cell death was assessed via PI exclusion. In B, D, F representative FACS

plots of GTG7, CC09 and CO147 colon CSCs are shown which were treated for 8 hours with DMSO or 10

µg/ml BetA. Mean ± SD of 3 experiments are shown. P < 0.05 (*); P < 0.01 (**); P< 0.001 (***).

100

0

20

40

60

80

100

% P

Ipos

2 h 4 h 8 h

0 µg/ml BetA2.5 µg/ml BetA5 µg/ml BetA7.5 µg/ml BetA10 µg/ml BetA

0

20

40

60

80

100

% P

Ipos

2 h 4 h 8 h


B C

D

GTG7

CC09

***

*****

**

********

0

20

40

60

80

% P

Ipos

2 h 4 h 8 h


CO147

**

% P

Ipos

FCS-H

7.8% 75.2%

6.6% 54.1%

8.3% 56.3%

DMSO BetA

% P

Ipos

% P

Ipos

E

F G

DMSO BetA

DMSO BetA

*********

******

***** ***

***

CD166

2.7% 17.9%

CD133 CD44v6

13.1%

CK20

3.7%

61.2% 5.81% 33.5% 2%

91.7% 67.5% 73.4% 1.7%

AC

C0

9C

O14

7G

TG

7


113

cells were treated with a commercially available SCD‐1 inhibitor. Interestingly, SCD‐1

inhibition led to a time‐ and concentration‐dependent cell death in both GTG7 and CC09.

The timing of cell death was different in these cells, as GTG7 reaches a maximum cell death

at 72 h, while this appears to take longer in CC09 (Figure 3A‐B). In agreement with the effect

observed with BetA treatment, SCD‐1 inhibition dramatically reduced clonogenic potential

of GTG7 and CC09 cells (Figure 3C‐D). Henceforth, we conclude that BetA is an effective

compound for the targeting of colon CSCs and this can be explained at least in part by the

effect of BetA on SCD‐1 activity in CSCs.

Discussion

CSCs are known to be more resistant to conventional treatment due to their (acquired)

resistance mechanisms, such as high levels of ATP‐binding cassette (ABC) transporters and

anti‐apoptotic molecules, active DNA‐repair, in some cases reduced proliferation and the

production of growth factors that confer refractoriness to anti‐neoplastic treatments.(44)

The escape mechanisms from standard therapies were shown both in vitro and in vivo.(17, 44)

In particular chemotherapy treatment of colon cancer results in an increase in CD133+

(oxaliplatin treatment), or ESA+/CD44+/CD166+ (irinotecan treatment) cell populations,

suggesting that these markers identify the CSC population, which are more resistant than

their differentiated counterpart.(21) CSC resistance mechanisms are frequently suggested to

provide an explanation for minimal residual disease in which an initial response is followed

by recurrences due to survival and subsequent expansion of colon CSC. Here we show that

BetA and SCD‐1 inhibition can both circumvent this CSC resistance and induced cell death

and loss of clonogenicity in colon CSCs. Our current study provides evidence for a

vulnerability in colon CSCs in the SCD‐1 pathway. Previously, we have also shown that this

BetA‐induced killing depends on the mitochondria, but occurs in a BCL‐2 family‐

independent manner.(38, 45) This may be essential for its capacity to kill CSCs as we and others

recently provided evidence that CSCs protect themselves from classical chemotherapy using

BCLXL‐dependent mechanisms. As BetA kills cells in a BCL‐2 family independent manner it

is expected to also target these CSCs. BetA has previously been shown to be tumor specific

and not targeting healthy cells. No toxicity to BetA was observed in mice treated with BetA

as well as vehicle liposomes, as both had the same amount of proliferating cells in the small

intestine(46). On top of this, we showed that non‐transformed fibroblasts are insensitive to

BetA as well as SCD‐1 inhibition. Cancer cells have reprogrammed their metabolism with

increased glycolysis and lipogenesis (de novo fatty acid synthesis) to maintain the need for

new biomaterials.(47) This reprogramming needs to already be present in the tumor initiating

cell, either pushing the cell fate toward transformation or creating an appropriate

“metabolic state” required for

Chapter 4

114

0

20

40

60

80

100

% P

Ipos

CC09

0

20

40

60

80 DMSOBetA

GTG7

% P

Ipos

low high total

6.0% 6.6% 6.0%

57.4% 51.9% 57.7%

FSC-A

low high total%

PIpo

s DM

SO

Bet

A

11.4% 7.6% 11.2%

74.4% 55.3% 76.3%

low high total

% P

Ipos D

MS

OB

etA

FSC-A

A B

C D

E F G

H I

low high total

DMSOBetA

*** ******

*** ******

****

***

***

0.0

0.5

1.0

1.5

fold

cha

nge

0 2.5 5 7.5 µg/ml BetA

100 2.5 5 7.5 µg/ml BetA

0.0

0.5

1.0

1.5

0.0

0.5

1.0

1.5

100 2.5 5 7.5 µg/ml BetA

GTG7 CC09 CO147

fold

cha

nge

fold

cha

nge

***

GTG7 CC09 OXA

0

2.5

5

7.5

10

cell

titer

blu

e (O

D)

5000

4000

3000

2000

1000

00 5 10 15 20

days0 5 10 15 20

days

10000

8000

6000

cell

titer

blu

e (O

D)

4000

2000

0


115

Figure 2. BetA‐induced loss of clonogenicity in colon cancer stem cells. (A, B) GTG7 colon CSCs were sorted

into three fractions, high GFP (indicating the stem cells), low GFP (indicating more differentiated cells) and

total cells. Overnight adhered cells were treated with vehicle (DMSO) or 10 µg/ml BetA for 2 h, after which

cell death was measured via PI exclusion. Representative FACS plots are shown in A. Mean ± SD of 3

experiments are shown. P< 0.001 (***). (C, D) CC09 colon CSCs were sorted into three fractions, high GFP

(indicating the stem cells), low GFP (indicating more differentiated cells) and total cells. Overnight adhered

cells were treated with vehicle (DMSO) or 10 µg/ml BetA for 2 h, after which cell death was measured via PI

exclusion. Representative FACS plots are shown in C. Mean ± SD of 3 experiments are shown. P< 0.001 (***).

Limiting dilutions experiments were performed and stem cell frequency was calculated in GTG7 (E), CC09

(F) and CO147 (G) colon CSCs that were pre‐treated for 2 h with indicated concentrations BetA. Mean ± SD

of 3 experiments are shown. P < 0.05 (*); P < 0.01 (**); P< 0.001 (***). GTG7 (H) and CC09 (I) colon CSCs

were treated with indicated concentrations BetA for 2h or with 0.5 µM oxaliplatin for 24 h after which cell

numbers were measured at different time points using cell titer blue.

Figure 3. SCD‐1 inhibitor‐induced cell death and loss of clonogenicity in colon cancer stem cells. GTG7 (A)

and CC09 (B) colon CSCs were treated for 24 h, 48 h, 72 h or 96 h with indicated concentrations of SCD‐1

inhibitor after which cell death was assessed via PI exclusion. Mean ± SD of 3 experiments are shown. P <

0.05 (*); P < 0.01 (**); P< 0.001 (***). Limiting dilutions experiments were performed and stem cell

frequency was calculated in GTG7 (C) and CC09 (D) colon CSCs that were pre‐treated for 48 h (GTG7) or 72

h (CC09) with 1µM SCD‐1 inhibitor. Mean ± SD of 3 experiments are shown. P< 0.001 (***).

0

20

40

60

80

100

% P

Ipos

24 h 48 h 72 h 96 h

00.20.51

0 µM inhib0.2 µM inhib0.5 µM inhib1 µM inhib

0

20

40

60

80

100

% P

Ipos

24 h 48 h 72 h 96 h

A

B

00.20.51

0 µM inhib0.2 µM inhib0.5 µM inhib1 µM inhib

0.0

0.4

1.0

0.2

0.6

0.8

fold

cha

nge

DM

SO

SC

D-1

0.0

0.4

1.0

0.2

0.6

0.8

fold

cha

nge

DM

SO

SC

D-1

C

D

GTG7GTG7

CC09

***

***

*

****

**

***

**

****

****

***CC09

***

***

Chapter 4

116

Tumorigenesis.(48) Although there is no direct proof, a growing body of evidence suggests

that metabolic reprogramming is at the core of cancer and it may provide a vulnerability of

cancer cells.(48‐50) To maintain a pool of healthy fatty acids needed for new biomaterials,

tumor cells start to depend on de novo fatty acid synthesis, while normal cells hardly utilize

this pathway as they acquire unsaturated fatty acids from their surroundings. Newly

generated fatty acids are fully saturated and are relatively toxic to the cells necessitating

their conversion to unsaturated fatty acids by introduction of a double bond at the ω‐9

position by the enzyme SCD‐1.(40, 51‐54) This points out the need for SCD‐1 activity in cancer

cells, but could also hint to a vulnerability of cancer cells to this pathway. In several types

of cancer, including colon cancer, SCD‐1 expression and activity are increased (38, 54, 55),

moreover SCD‐1 levels were found elevated in in vivo models that were genetically

predisposed to develop certain cancers.(56) These observations suggest that SCD‐1 activity

is needed in cancer cells in order to maintain the levels of unsaturated fatty acids and by

doing so protecting tumor cells from a surplus of toxic saturated fatty acids.(40, 55, 57) Indeed

when inhibiting SCD‐1 by small molecule inhibitors or genetic knockdown induction of cell

death was observed in cancer cells(38, 57) and this is a result of a strong switch in the

saturation grade of the mitochondrial lipid cardiolipin and as a consequence the induction

of mitochondrial fragmentation and induction of cell death.

Here we show that colon CSCs are also sensitive to inhibition of the SCD‐1 enzyme. The

vulnerability to inhibition of SCD‐1 activity was also recently observed in another study

where inhibition of SCD‐1 in lung CSCs resulted in a reduced stem cell marker expression

and induced cell death.(41) These data are in agreement with what we observed in our colon

CSCs model. One striking observation is the velocity of cell death induction, which appears

to be much higher as compared to classical colon cancer cell lines. This is specific for BetA

and not observed with SCD‐1 inhibition, which appears as effective in cancer cells and colon

CSC cultures.(38) We previously have hypothesized that the rapid effect of BetA is due to a

combined effect of BetA, on one hand blocking the production of unsaturated fatty acids,

but on the other hand increasing turnover of lipids and thereby the incorporation of

saturated fatty acids.(38) This concept therefore may point to an additional effect of BetA on

colon CSC cultures that strongly enhances the impact on these cells. At this point we have

no definitive proof as to this effect, but it is interesting to note that colon CSCs can be

identified by their high level of lipid droplets when compared with differentiated tumor and

even more so when compared to cancer cell lines.(58) Lipid droplets are fat storing

organelles, that are suggested to be involved in the storage, transport and metabolism of

lipids, in signaling, and as a specialized microenviroment for metabolism.(59, 60) SCD‐1 plays

an important role in lipid droplets formation and size. In SCD‐1 WT C. elegans, large‐sized

lipid droplets were observed, while in SCD‐1 mutant C. elegans small‐sized lipid droplets are

found.(61) Small‐sized lipid droplets are also found in patients with Berardinelli‐Seip

congenital lipodystrophy, which have decreased SCD‐1 activity.(62) These lipid droplets


117

appear to be crucial for the CSCs and we hypothesize that BetA might interfere with these

lipid droplets and therefore induces a very rapid induction of mitochondrial damage.(38)

We conclude that SCD‐1 is a crucial player in the maintenance of stemness of cancer cells

and inhibition is incompatible with long‐term survival. Targeting the activity of this enzyme

in CSCs provides an effective means to target these cells and can be achieved either with

direct chemical inhibitors or even more effectively with BetA.

Material &Methods

Chemicals/antibodies

Betulinic acid (BioSolution Halle, Germany, >99% purity) was dissolved in DMSO at 4 mg/ml

and stored at ‐80°C. Oxaliplatin was purchased from Sigma‐Aldrich (St. Louis, MO, USA).

Stearoyl‐CoA desaturase inhibitor (cat# 1716) was from Biovision (Milpitas, CA, USA).

Isolation, Culture and Characterization of Colon Cancer Cells

Colon cancer tissues were collected at the Department of Surgical and Oncological Sciences,

and CEMM Department, in accordance with the ethical standards of the University of

Palermo and AMC institutional committees. Colon spheroid cultures were derived from

colorectal cancer patients and maintained in stem cell medium (advanced DMEM/F12

(Gibco) supplemented with N2 Supplement (Gibco), 6 mg/ml glucose, 5 mM HEPES, 2 mM

L‐glutamine, 4 μg/ml heparin, 50 ng/ml epidermal growth factor and 10 ng/ml basic

fibroblast growth factor) as previously described (16). The colon CSC lines obtained were

subjected to genotypic characterization in order to validate each cell line’s individuality and

their mutational status (Table 1), and further tested for their ability to generate tumor

xenografts that replicated the parental histology. Colon CSCs were characterized by their

expression of the putative cancer stem cell markers CD166‐PE (3A6; mouse IgG1k; BD

Biosciences), CD133‐APC (293C3; mouse IgG2bk; Miltenyi) and CD44v6‐APC (2F10; mouse

IgG1; R&D systems), and CK20 (Ks20.8; mouse; IgG2ak), a colonic differentiation marker,

data were acquired by flow cytometry using a BD ACCURI C6 and analyzed using the FlowJo

software (Tree Star Inc., Ashland, OR, USA). CK20 was detected by using a goat anti‐Mouse

IgG (H+L) secondary antibody, Alexa Fluor®‐488 conjugate. RKO cells were obtained from

ATCC (Manassas, VA, USA) and maintained in IMDM supplemented with 8% FCS, 2 mM L‐

glutamine, 40 U/ml penicillin and 40 mg/ml streptomycin.GTG7 and CC09 cells bear the TCF

Optimal Promoter (TOP)‐green fluorescent protein (GFP) construct.(5) These cells, derived

from single‐cell cloned TOPGFP cultures, still show a big heterogeneity in Wnt signaling

levels, which positively correlates with cancer stemness in colon CSCs.(63)

Cell death analysis

Cell death was determined by PI exclusion. In short, 50 000 single colon CSCs were seeded

in a 24‐well plate and allowed to attach overnight. Cells were treated for indicated time‐

Chapter 4

118

points with BetA or SCD‐1 inhibitor and harvested. The cell pellet was resuspended in 200 μl

medium and stained with PI at 0.5 μg/ml just before measuring by flow cytometry

(FACSCanto, BD Biosciences). Samples were analyzed using FlowJo software (Tree Star Inc.,

Ashland, OR, USA).

Cell Proliferation Assay

For determination of cell viability, 1000 cells per well were plated in repellent surface for

low attachment 96‐well plates (CELLSTAR® Cell‐Repellent Surface, Greiner Bio‐ One) in

medium containing vehicle (DMSO), 0.5 µM oxaliplatin or increasing doses of BetA. Viable

cells were determined at indicated time‐points using Cell Titer Blue (Promega, Fitchburg,

WI, USA) by measurement of absorbance at 540 nm in a Synergy plate reader (Biotek). Cell

viability at each concentration was expressed as OD cells – OD background.

Limiting dilution assay

The self‐renewal capacity of colon CSCs was assayed by dissociation of colon cancer

spheroids. Colon CSCs were cultured in adhesion o/n and pre‐treated 2 h with indicated

concentrations BetA or pre‐treated 48 h or 72 h with SCD‐1 inhibitor after which cells were

collected and plated at serial dilution (1, 2, 4, 8, 16, 32, 64 and 128 cells/ well) in 96 well

microplates with flat bottom and repellent surface for low attachment (CELLSTAR® Cell‐

Repellent Surface, Greiner Bio‐ One), using a FACS Aria II. Results were statistically

evaluated after 3 weeks by using the Extreme Limiting Dilution Analysis (ELDA) software.(64)

Statistical analysis.

Results are shown as the mean ± SD for at least three repeated independent experiments

for each group. The mean and SD were obtained by analyzing three replicates using Prism

5 (GraphPad Software, La Jolla, CA, USA) applying a 2‐tailed unpaired Student t‐test.

Differences were considered significant when p values P < 0.05 (*); P < 0.01 (**); P< 0.001

(***).


119

Conflict of interest


Acknowledgements

This work was supported by a grant of the Stichting Nationaal Fonds tegen Kanker (SNFK),

Amsterdam, The Netherlands (L.P), by Associazione Italiana per la Ricerca sul Cancro (AIRC)

to G.S. (AIRC IG 8730 and AIRC 19 5x1000) and Dutch Cancer Society Grants (UvA2012‐5425)

to J.P.M. We would like to thank Louis Vermeulen for his valuable scientific input and helpful

discussions and Kate Cameron for her assistance with fluorescence‐activated cell sorting

experiments.

Chapter 4

120

References 1. Jemal, A., et al., Global cancer statistics. CA Cancer J Clin, 2011. 61(2): p. 69‐90. 2. Valent, P., et al., Cancer stem cell definitions and terminology: the devil is in the details. Nat Rev Cancer, 2012. 12(11): p. 767‐75. 3. Todaro, M., et al., CD44v6 is a marker of constitutive and reprogrammed cancer stem cells driving colon cancer metastasis. Cell Stem Cell, 2014. 14(3): p. 342‐56. 4. Kemper, K., C. Grandela, and J.P. Medema, Molecular identification and targeting of colorectal cancer stem cells. Oncotarget, 2010. 1(6): p. 387‐95. 5. Vermeulen, L., et al., Wnt activity defines colon cancer stem cells and is regulated by the microenvironment. Nat Cell Biol, 2010. 12(5): p. 468‐76. 6. Kinzler, K.W. and B. Vogelstein, Lessons from hereditary colorectal cancer. Cell, 1996. 87(2): p. 159‐70. 7. Fearon, E.R., Molecular genetics of colorectal cancer. Annu Rev Pathol, 2011. 6: p. 479‐507. 8. Clevers, H., Wnt/beta‐catenin signaling in development and disease. Cell, 2006. 127(3): p. 469‐80. 9. Whissell, G., et al., The transcription factor GATA6 enables self‐renewal of colon adenoma stem cells by repressing BMP gene expression. Nat Cell Biol, 2014. 16(7): p. 695‐707. 10. Lombardo, Y., et al., Bone morphogenetic protein 4 induces differentiation of colorectal cancer stem cells and increases their response to chemotherapy in mice. Gastroenterology, 2011. 140(1): p. 297‐309. 11. Kosinski, C., et al., Gene expression patterns of human colon tops and basal crypts

and BMP antagonists as intestinal stem cell niche factors. Proc Natl Acad Sci U S A, 2007. 104(39): p. 15418‐23. 12. He, X.C., et al., BMP signaling inhibits intestinal stem cell self‐renewal through suppression of Wnt‐beta‐catenin signaling. Nat Genet, 2004. 36(10): p. 1117‐21. 13. Sikandar, S.S., et al., NOTCH signaling is required for formation and self‐renewal of tumor‐initiating cells and for repression of secretory cell differentiation in colon cancer. Cancer Res, 2010. 70(4): p. 1469‐78. 14. Prasetyanti, P.R., et al., Regulation of stem cell self‐renewal and differentiation by Wnt and Notch are conserved throughout the adenoma‐carcinoma sequence in the colon. Mol Cancer, 2013. 12(1): p. 126. 15. Bu, P., et al., A microRNA miR‐34a‐regulated bimodal switch targets Notch in colon cancer stem cells. Cell Stem Cell, 2013. 12(5): p. 602‐15. 16. Vermeulen, L., et al., Single‐cell cloning of colon cancer stem cells reveals a multi‐lineage differentiation capacity. Proc Natl Acad Sci U S A, 2008. 105(36): p. 13427‐32. 17. Todaro, M., et al., Colon cancer stem cells dictate tumor growth and resist cell death by production of interleukin‐4. Cell Stem Cell, 2007. 1(4): p. 389‐402. 18. Ricci‐Vitiani, L., et al., Identification and expansion of human colon‐cancer‐initiating cells. Nature, 2007. 445(7123): p. 111‐5. 19. O'Brien, C.A., et al., A human colon cancer cell capable of initiating tumour growth in immunodeficient mice. Nature, 2007. 445(7123): p. 106‐10. 20. Vermeulen, L., et al., The developing cancer stem‐cell model: clinical challenges and opportunities. Lancet Oncol, 2012. 13(2): p. e83‐9.


121

21. Dylla, S.J., et al., Colorectal cancer stem cells are enriched in xenogeneic tumors following chemotherapy. PLoS One, 2008. 3(6): p. e2428. 22. Colak, S., et al., Decreased mitochondrial priming determines chemoresistance of colon cancer stem cells. Cell Death Differ, 2014. 21(7): p. 1170‐7. 23. Medema, J.P., Cancer stem cells: the challenges ahead. Nat Cell Biol, 2013. 15(4): p. 338‐44. 24. Jung, P., et al., Isolation and in vitro expansion of human colonic stem cells. Nat Med, 2011. 17(10): p. 1225‐7. 25. Huang, E.H., et al., Aldehyde dehydrogenase 1 is a marker for normal and malignant human colonic stem cells (SC) and tracks SC overpopulation during colon tumorigenesis. Cancer Res, 2009. 69(8): p. 3382‐9. 26. Dalerba, P., et al., Phenotypic characterization of human colorectal cancer stem cells. Proc Natl Acad Sci U S A, 2007. 104(24): p. 10158‐63. 27. Fearon, E.R. and B. Vogelstein, A genetic model for colorectal tumorigenesis. Cell, 1990. 61(5): p. 759‐67. 28. Zeuner, A., et al., Colorectal cancer stem cells: from the crypt to the clinic. Cell Stem Cell, 2014. 15(6): p. 692‐705. 29. Sadanandam, A., et al., A colorectal cancer classification system that associates cellular phenotype and responses to therapy. Nat Med, 2013. 19(5): p. 619‐25. 30. Kreso, A., et al., Variable clonal repopulation dynamics influence chemotherapy response in colorectal cancer. Science, 2013. 339(6119): p. 543‐8. 31. De Sousa, E.M.F., et al., Cancer heterogeneity‐‐a multifaceted view. EMBO Rep, 2013. 14(8): p. 686‐95.

32. De Sousa, E.M.F., et al., Poor‐prognosis colon cancer is defined by a molecularly distinct subtype and develops from serrated precursor lesions. Nat Med, 2013. 19(5): p. 614‐8. 33. Zuco, V., et al., Selective cytotoxicity of betulinic acid on tumor cell lines, but not on normal cells. Cancer Lett, 2002. 175(1): p. 17‐25. 34. Rzeski, W., et al., Betulinic acid decreases expression of bcl‐2 and cyclin D1, inhibits proliferation, migration and induces apoptosis in cancer cells. Naunyn Schmiedebergs Arch Pharmacol, 2006. 374(1): p. 11‐20. 35. Mullauer, F.B., J.H. Kessler, and J.P. Medema, Betulinic acid, a natural compound with potent anticancer effects. Anticancer Drugs, 2010. 21(3): p. 215‐27. 36. Kessler, J.H., et al., Broad in vitro efficacy of plant‐derived betulinic acid against cell lines derived from the most prevalent human cancer types. Cancer Lett., 2007. 251(1): p. 132‐145. 37. Ehrhardt, H., et al., Betulinic acid‐induced apoptosis in leukemia cells. Leukemia, 2004. 18(8): p. 1406‐12. 38. Potze, L., et al., Betulinic acid induces a novel cell death pathway that depends on cardiolipin modification. Oncogene, 2015. 39. Scaglia, N. and R.A. Igal, Inhibition of Stearoyl‐CoA Desaturase 1 expression in human lung adenocarcinoma cells impairs tumorigenesis. Int J Oncol, 2008. 33(4): p. 839‐50. 40. Scaglia, N., J.W. Chisholm, and R.A. Igal, Inhibition of stearoylCoA desaturase‐1 inactivates acetyl‐CoA carboxylase and impairs proliferation in cancer cells: role of AMPK. PLoS One, 2009. 4(8): p. e6812.

Chapter 4

122

41. Noto, A., et al., Stearoyl‐CoA desaturase‐1 is a key factor for lung cancer‐initiating cells. Cell Death Dis, 2013. 4: p. e947. 42. Chan, C.W., et al., Gastrointestinal differentiation marker Cytokeratin 20 is regulated by homeobox gene CDX1. Proc Natl Acad Sci U S A, 2009. 106(6): p. 1936‐41. 43. Fulda, S., et al., Betulinic acid triggers CD95 (APO‐1/Fas)‐ and p53‐independent apoptosis via activation of caspases in neuroectodermal tumors. Cancer Res, 1997. 57(21): p. 4956‐64. 44. Colak, S. and J.P. Medema, Cancer stem cells‐‐important players in tumor therapy resistance. FEBS J, 2014. 281(21): p. 4779‐91. 45. Mullauer, F.B., J.H. Kessler, and J.P. Medema, Betulinic acid induces cytochrome c release and apoptosis in a Bax/Bak‐independent, permeability transition pore dependent fashion. Apoptosis., 2009. 14(2): p. 191‐202. 46. Mullauer, F.B., et al., Betulinic acid delivered in liposomes reduces growth of human lung and colon cancers in mice without causing systemic toxicity. Anticancer Drugs, 2011. 22(3): p. 223‐33. 47. Menendez, J.A. and R. Lupu, Fatty acid synthase and the lipogenic phenotype in cancer pathogenesis. Nat Rev Cancer, 2007. 7(10): p. 763‐77. 48. Sebastian, C., Tracking down the origin of cancer: metabolic reprogramming as a driver of stemness and tumorigenesis. Crit Rev Oncog, 2014. 19(5): p. 363‐82. 49. Schulze, A. and A.L. Harris, How cancer metabolism is tuned for proliferation and vulnerable to disruption. Nature, 2012. 491(7424): p. 364‐73. 50. Green, D.R., L. Galluzzi, and G. Kroemer, Cell biology. Metabolic control of cell death. Science, 2014. 345(6203): p. 1250256.

51. Scaglia, N., J.M. Caviglia, and R.A. Igal, High stearoyl‐CoA desaturase protein and activity levels in simian virus 40 transformed‐human lung fibroblasts. Biochim Biophys Acta, 2005. 1687(1‐3): p. 141‐51. 52. Paton, C.M. and J.M. Ntambi, Biochemical and physiological function of stearoyl‐CoA desaturase. Am J Physiol Endocrinol Metab, 2009. 297(1): p. E28‐37. 53. Kumar‐Sinha, C., et al., Transcriptome analysis of HER2 reveals a molecular connection to fatty acid synthesis. Cancer Res, 2003. 63(1): p. 132‐9. 54. Igal, R.A., Stearoyl‐CoA desaturase‐1: a novel key player in the mechanisms of cell proliferation, programmed cell death and transformation to cancer. Carcinogenesis, 2010. 31(9): p. 1509‐15. 55. Li, J., et al., Partial characterization of a cDNA for human stearoyl‐CoA desaturase and changes in its mRNA expression in some normal and malignant tissues. Int.J.Cancer, 1994. 57(3): p. 348‐352. 56. Falvella, F.S., et al., Stearoyl‐CoA desaturase 1 (Scd1) gene overexpression is associated with genetic predisposition to hepatocarcinogenesis in mice and rats. Carcinogenesis, 2002. 23(11): p. 1933‐6. 57. Scaglia, N. and R.A. Igal, Stearoyl‐CoA desaturase is involved in the control of proliferation, anchorage‐independent growth, and survival in human transformed cells. J Biol Chem, 2005. 280(27): p. 25339‐49. 58. Tirinato, L., et al., Lipid droplets: a new player in colorectal cancer stem cells unveiled by spectroscopic imaging. Stem Cells, 2015. 33(1): p. 35‐44. 59. Yang, L., et al., The proteomics of lipid droplets: structure, dynamics, and functions of the organelle conserved from bacteria to humans. J Lipid Res, 2012. 53(7): p. 1245‐53.


123

60. Walther, T.C. and R.V. Farese, Jr., Lipid droplets and cellular lipid metabolism. Annu Rev Biochem, 2012. 81: p. 687‐714. 61. Shi, X., et al., Regulation of lipid droplet size and phospholipid composition by stearoyl‐CoA desaturase. J Lipid Res, 2013. 54(9): p. 2504‐14. 62. Boutet, E., et al., Seipin deficiency alters fatty acid Delta9 desaturation and lipid droplet formation in Berardinelli‐Seip congenital lipodystrophy. Biochimie, 2009. 91(6): p. 796‐803. 63. Ben‐David, U., et al., Selective elimination of human pluripotent stem cells by an oleate synthesis inhibitor discovered in a high‐throughput screen. Cell Stem Cell, 2013. 12(2): p. 167‐79. 64. Hu, Y. and G.K. Smyth, ELDA: extreme limiting dilution analysis for comparing depleted and enriched populations in stem cell and other assays. J Immunol Methods, 2009. 347(1‐2): p. 70‐8.

Chapter 4

124


Supplementary Figure 1. BetA‐induced cell death in colon cancer cells (A, B) RKO cells were treated for 2

h, 4 h, and 8 h with indicated concentrations of BetA after which cell death was assessed via PI exclusion.

Representative FACS plots of RKO cells treated for 8 h with DMSO or 10 µg/ml BetA are shown in A. Mean

± SD of 3 experiments are shown. P < 0.05 (*); P < 0.01 (**); P< 0.001 (***).

Supplementary Figure 2. Clonogenicity in colon cancer cells RKO cells were pre‐treated for 2 h with

indicated concentrations BetA after which limiting dilution experiments were performed and stem cell

frequency was calculated.

0

20

40

60

80

100

% P

Ipos

2 h 4 h 8 h


***

**

3.7%

DMSO

24.1%

BetA

FCS-H

% P

Ipos

A B

****

****

0.0

0.5

1.0

1.5

fold

cha

nge

in s

tem

c

ell f

requ

ency

0 2.5 5 7.5 µg/ml BetA

10


125

Supplementary Figure 3. Long term assay on CSCs. (A) GTG7 and CC09 (B) colon CSCs were treated with 0.5

µM oxaliplatin for 24 h or indicated concentrations BetA (µg/ml) for 2 h after which cells were left to grow.

Pictures (10x magnification) were taken at day 10 after which the number of cells were determined with

cell titer blue. (C) GTG7 and CC09 (D) colon CSCs were treated with indicated concentrations oxaliplatin (μM)

for 24 h after which cells were plated and cell numbers were measured at different time points using cell

titer blue. (E) GTG7 and CC09 (F) colon CSCs were treated with indicated concentration oxaliplatin (μM) for

24 h after which cells were left to grow. Pictures (4x magnification) were taken at day 15 after which the

number of cells were determined with cell titer blue.

5

OXA 0 2.5

7.5 10

A BOXA 0 2.5

5 7.5 105

C D

E F

0 5 10 15days

0

500

1000

1500

cell

titer

blu

e (

OD

)

GTG7 CC09

0

500

1000

1500ce

ll tit

er b

lue

(O

D)

0 5 10 15days

20

00.51.02.55.010

5

0 0.5 1.0

2.5 10 5

0 0.5 1.0

2.5 10

Chapter 5

Improved identification of lipids using

physico‐chemical properties: application

to lipidomics analysis of Betulinic acid

treatment

Lisette Potze, Mia L. Pras‐ Raves, Henk van Lenthe, Martin A.T

Vervaart, Jan H. Kessler, Angela C. Luyf, Jan Paul Medema,

Antoine H. C. van Kampen, Frédéric M. Vaz

Under review at Journal of Lipid Research

“Who in the world am I? Ah, that’s the great puzzle.”


Chapter 5

128

Abstract

The correct identification of all metabolites in LC‐MS lipidomic experiments is a time‐

consuming step in the data analysis process. Here we describe a method which not only

uses the mass and retention time of peaks, but also incorporates the elution pattern of the

phospholipid cluster to improve the identification of phospholipid species. Thus, even at

low resolution (<10,000), a clear distinction can be made between chromatographically

separated isobaric lipids such as phosphatidylglycerol and bis(monoacylglycero)phosphate,

between phosphatidylethanolamine (PE) and plasmalogen PE, and between

phosphatidylserine and phosphatidylinositol. The method is illustrated using a lipidomics

data set acquired on a low resolution triple quadrupole MS to investigate the effect of the

anti‐cancer drug Betulinic acid in HeLa cells.

List of abbreviations:

PL phospholipid

IS internal standard

BMP bis(monoacylglycero)phosphate

CL cardiolipin

mlCL monolyso‐cardiolipin

PE phosphatidylethanolamine

pPL PL plasmalogen

aPL O‐alkyl PL

PI phosphatidylinositol

PS phosphatidylserine

PG phosphatidylglycerol

PC phosphatidylcholine

LPL lyso‐PL

ST sulfatide

m/z mass over charge ratio

RT retention time

BetA Betulinic acid

SCD‐1 Stearoyl‐CoA desaturase 1

Improved identification of lipids using physico‐chemical properties

129

Introduction

Lipidomics is the comprehensive study of the lipid composition of cells, organelles or

biofluids (1). The combination of (ultra‐)high‐performance liquid chromatography ((U)HPLC)

and mass spectrometry (MS) is instrumental in the measurements of lipids in complex

samples (2). Despite the progress and technical advances in the field in the last decade, there

are still many challenges. First, the chromatographic separation of lipids is challenging due

to the complexity of the samples, matrix effects, and the sheer number of different, but

very similar, metabolites (3, 4). Inadequate separation will produce complex mass spectra

that are more difficult to analyze due to, for example, overlapping peaks in the

chromatographic and/or mass domain. Secondly, current data pre‐processing methods

produce peak tables containing monoisotopic and isotope peaks based on their retention

time and m/z values, often still requiring intervention and correction by experts

experienced with LC‐MS spectra to reduce the number of false positive and missing peaks.

This makes the pre‐processing of raw MS data very time consuming but, surprisingly, data

pre‐processing receives relatively little attention in literature (5). Finally, the assignment of

metabolites to all peaks in the peak lists is a major bottleneck. Frequently, only a small

fraction of the observed peaks can be reliably assigned to a compound. This complicates

further data analysis and comprehensive biological interpretation. The assignment of

compounds generally requires a combination of knowledge of expected metabolites,

internal standards, fragmentation patterns and spectral databases to which MS spectra can

be compared (6). There are several repositories and search engines to assist in the

identification of phospholipids (PLs), including Lipid Maps (Lipidomics Gateway), containing

over 30,000 lipid species (3, 7) and Lipid Search (8) consisting of more than 200,000 structures.

The LipidBlast database combines computer‐generated spectra and experimental

fragmentation data, describing about 120,000 structures for lipids (9) while online databases

such as VaLID (5) aid in the identification of glycerophospholipids based on their molecular

mass and provide visualization of the molecules. Database searches, however, are generally

conducted peak by peak, thus ignoring the pattern that is formed by the whole cluster of

PLs that belong to the same class, a pattern which is caused by similarities and subtle

differences in their physico‐chemical properties. In a normal‐phase LC system, PLs with the

same head group will have roughly the same RT, but within such a cluster, the inclusion of

an extra –CH2– group or the presence of an additional double bond in one of the side chains

will influence the observed RT of the PLs. The resolution of the MS measurements is another

important factor in the ease of the identification. Higher resolutions yield more accurate

m/z values, which enables the determination of the exact elemental composition of the

underlying metabolite, even though this may not always pinpoint a unique identifier (for

example, bis(monoacylglycero)phosphate (BMP) and phosphatidylglycerol (PG) have the

same elemental composition but different chemical structures). Querying m/z values from

low‐resolution MS (resolution < 10,000) in a reference database results in a long list of

Chapter 5

130

candidate metabolites for every peak, rendering the identification process time‐consuming

and error‐prone. Many clinical diagnostics laboratories currently make extensive use of low‐

resolution single‐ or tandem mass filter (quadrupole) setups (10), for which the usefulness of

brute force database searches, without regard for the context of the peak, is limited.

With the aim of reducing the amount of work that is required for the analysis and

interpretation of the results of a data set after pre‐processing, we describe a method which

not only uses the mass and retention time of peaks, but also incorporates the elution

pattern of the phospholipid cluster to improve the identification of phospholipid species.

We have integrated our identification method with XCMS (11), which is an R package widely

used for the analysis of metabolomics data, into a pipeline which includes functionalities for

isotope correction and normalization of intensities for assigned peaks (Figure 1).

To test our lipidomics pipeline we made use of a data set of HeLa cells treated with Betulinic

acid (BetA). BetA is a plant‐derived compound that shows selective cytotoxicity for tumor

cells while healthy cells remain unaffected (12, 13). Recently, we showed that BetA induces

apoptosis by inhibition of stearoyl‐CoA desaturase 1 (SCD‐1), resulting in changes of the

side‐chain saturation of cardiolipin (CL, a phospholipid mainly located in the inner

membrane of mitochondria) and impaired mitochondrial morphology. We concluded that

CL is a crucial player in mitochondrial integrity that is strongly dependent on the level of

saturation of its side chains. However, in the above‐mentioned study we only focused on

PG, the precursor of CL, and on CL itself (13). Therefore, we revisited this data set to perform

a complete lipidomics analysis to investigate the effect of BetA on the full lipidome and

demonstrate the usefulness of our pattern recognition algorithm.

Results

Data processing:

We developed a method that facilitates the assignment of lipid compound names to peak

groups originating from the same class of PLs. We have integrated our identification method

with XCMS (11), a software package in the R programming language (http://www.R‐

project.org) which is widely used for the analysis of metabolomics data.

A typical full scan analysis yields between 1000 and 10,000 peak groups. Each of these peak

groups is characterized by the median m/z and median RT of all the peaks from the different

samples. For pre‐processing packages such as XCMS (11), this is the final output. The

assignment of compound labels to each peak group is then in the hands of the user.


131

Figure 1: Schematic representation of the lipidomics pre‐processing pipeline. Raw data files are obtained for

every sample; the pre‐processing pipeline using XCMS, followed by identification, isotope correction and

normalization yields the final annotated peak list.

Plotting the peak groups (Figure 2) immediately allows the experienced eye to observe

clusters, which are the result of groups of phospholipids with the same head group but with

different side‐chain compositions. The addition of internal standards (IS) per class of PLs to

the samples not only serves for normalization and quantification purposes: identification of

the IS peaks in the peak group list enables the determination of the location of the entire

cluster of peaks originating from compounds from the same class, and aids in the

identification of all peaks in this region. Specific head group scans, such as neutral‐loss or

precursor‐ion scans, can also be helpful in locating clusters of PLs. Rather than relying only

on the match to a theoretical mass of a compound, our method makes use of the systematic

pattern that is formed by all the compounds in the same PL class upon normal‐phase LC

separation. This pattern is the result of differences in mobility on the LC column of PL

molecules with different side‐chain compositions. On a normal‐phase column, PLs with

longer side chains have a lower RT, due to a decreased polarity of the molecules. In addition,

as the number of double bonds in the combined side chains increases, the RT decreases.

These patterns can be used to aid in the assignment of PL names to peak groups. This is

illustrated in Figure 3A and B, which shows the CL region in close‐up, and described in the

methods section in detail. Resulting data files are available as supplemental files

(supplemental files 1‐4).

Peak finding

Matching peaks across samples

Fill in missing peak data

Retention time correction

Identification Isotope correction Normalization

Visualization Statisti cal Analysis

RawRaw RawRaw RawRaw

Peaklist

Chapter 5

132

Figure 2: All peaks in lipidomics data set. Clusters observed in PL peak groups in negative (lower panel) and

positive scan mode (top panel, only precursor ion scan m/z 184 (PR184) is shown). Every peak group that

was obtained using XCMS is plotted as a black dot. Several clusters of peak groups can be identified and are

labeled in the figure, with a line showing the direction of the peak groups in each cluster. Peak groups

observed in positive PR184 scan mode are shown above the horizontal line at RT=8.5 minutes; below the

line are the peak groups in negative mode.


133

Figure 3: Pattern observed in doubly charged cardiolipin. (A) The full CL cluster, starting in the upper left‐

hand corner with the IS peak, CL(56:0). (B) Close‐up of CL(64:x) up to CL(66:y). For chromatographic

separation with a normal‐phase column, the RT of the compounds decreases with increasing (combined)

chain length and number of double bonds in the chains. Dots represent the median location of all samples

in a peak group. The size of the circle indicates the relative average intensity of those peaks on an arbitrary

scale.

Chapter 5

134

For the data set described in this paper, which was measured at relatively low resolution

(approximately 2000 at m/z 500), we were able to identify metabolites from 9 classes in

negative scan mode (BMP, CL, mlCL, PG, PA, PE, pPE, PI, PS), and as many as 40% of all peak

groups could be identified as either belonging to monoisotopic peaks (e.g. metabolites) or

to corresponding isotope peaks. In positive scan mode (precursor‐ion scan for m/z 184.1

(PR184) and neutral‐loss scan for m/z 141.1 (NL141)), 10 classes (LPC, LpPC, PC, pPC, aPC,

SM in PR184 and LPE, PE, pPE, aPE in NL141) and more than half of all peak groups could be

identified in PR184 and NL141, (Table 1, supplemental tables I‐IV). Other regions in the m/z

versus RT space contain either noise, peaks from unknown PL classes, or peaks originating

from compounds other than PLs. In neutral‐loss or precursor‐ion spectra, where only one

class of PL is studied, the success rate of the identification is higher than for full‐ion scans

(supplemental tables II and III).This method works equally well for low‐resolution lipidomics

data as for high‐resolution measurements (data not shown).

Figure 4: Total phospholipid levels in BetA or vehicle treated cells. Highly abundant PLs (PE, PS, PC and SM)

(A) and less abundant PLs (BMP, CL, LPC, PG, PA) (B) for DMSO (vehicle) and BetA treated cells are depicted.

*=P<0.05, ***=P<0.001

Analysis of the BetA data set using the pattern recognition algorithm

Differences between the cells treated with vehicle (DMSO) and BetA treated cells were

found in different phospholipid classes (Figure 4A and B). The most significant changes

occurred in the total levels of BMP and PG: BetA reduces total BMP levels while PG levels

are increased after treatment with BetA. There were effects on PC and LPC levels (elevated

in BetA) as well, although they were less pronounced. Further analysis within the PL clusters

showed that BetA induced an increase in intensity for saturated CL (number of double bonds

less than 4) and a decrease for poly‐unsaturated CL (4 double bonds and up), compared to

the control (DMSO) samples (Figure 5). This effect is not observed in PG or BMP, even

though PG is the precursor for CL (Supplemental Figure IA and IB).


135

Figure 5: Heatmap of differences in cardiolipin composition. The cardiolipins have been sorted by total

number of double bonds in the combined side chains; only CL(64:x) through CL(70:y) are shown. A clear

distinction between mostly saturated CLs (number of double bonds <4) and poly‐unsaturated CLs (number

of double bonds > 3) can be seen.

Chapter 5

136

Discussion

Being able to automatically identify metabolites greatly speeds up the data analysis process

for metabolomics experiments. Querying m/z values in different (online) databases is not

always practical: it often yields multiple assignments for a metabolite, sometimes none. In

any case, the analysis of the data sets requires considerable attention and time from

metabolomics experts which in practice results in a bottleneck since individual datasets

contain hundreds to thousands of peak groups (and considerably more for high‐resolution

data sets). The pre‐processing lipidomics pipeline described in this paper is suitable both for

low‐resolution (the data set described here has a resolution of approximately 2000 at m/z

500) and for high‐resolution data (120,000 at m/z 500 for a Thermo Scientific Q‐Exactive

plus). For low‐resolution data, the greatest benefit of our method is in the disambiguation

of peaks with almost identical mass, such as PE and pPE. For high‐resolution data, the main

advantage is that a great number of peaks can be rapidly and reliably identified.

The pipeline yields peak tables and plots per class and per section of a class for all identified

PLs, such as illustrated for CL in Figure 3. This enables the user to check whether the

identification process has given the correct result. In many cases, analyzing the lipidome of

a biological system gives considerably more insight into the underlying biochemical

processes than one would get when focusing on only one PL class. One of our goals is

therefore to identify as many PL classes as we possibly can. When identifying PLs

incorporated in the current pipeline we essentially perform a targeted analysis. Yet, we also

spot unknowns which may represent relevant differences between the investigated groups

which can be further investigated. In the current method, it is fairly easy to add new

(phospho)lipid classes based on their chemical composition and physico‐chemical

properties.

The method works well for low‐resolution data as it addresses and aids to resolve

ambiguous phospholipid assignments. PG and BMP, for example, are two classes of PLs that

have the exact same chemical formula and therefore the same m/z. Using our algorithm in

combination with appropriate chromatographic separation and internal standards achieves

the correct identification of compounds from each class. Another example is PE and pPE,

which differ in chemical structure but at low resolution cannot be distinguished based on

m/z alone. In this case, the difference in polarity causes a distinct shift in the pattern of the

peaks observed for each of these clusters, so that the ambiguity can be resolved and PE and

pPE can be distinguished. A third example is PI and PS, which often co‐elute. Since PSs have

even masses and PI have odd masses in ionized state, the first isotope of one species often

overlaps with the monoisotopic peak of the other species. Identification followed by isotope

correction can then solve this problem and annotate species of both PL classes.

For high‐resolution data, the accuracy of the observed m/z values is higher and the peaks

are much narrower on the m/z‐axis, so that there is much less overlap between different


137

peaks. This results in a much greater number of peak groups, but the percentages of

identified metabolites are still comparable to what we find for low‐resolution data (data not

shown). At high resolution, the benefit of this method is in the larger number of peaks that

can be identified in an automated fashion, without (or with minimal) human intervention.

Material and Methods

Chemicals

Betulinic acid (BioSolution Halle, Germany, >99% purity) was dissolved in DMSO at 4 mg/ml

and stored at ‐80°C.

Cell culture

HeLa cells were obtained from the ATCC. Cells were cultured in IMDM supplemented with

8% FCS, 2mM L‐glutamine, 40 U/ml penicillin and 40 µg/ml streptomycin.

Mass spectrometry

Cell pellets were resuspended in water and sonicated for 30 s (8 W) using a tip sonicator.

Protein concentration of the homogenates was determined according to the BCA

protocol(14). Phospholipids were extracted using a single‐phase extraction. A defined

amount of internal standards (0.1 nmol of CL(14:0)4 , 0.2 nmol of BMP(14:0)2 , 2.0 nmol of

PC(14:0)2 , 0.1 nmol of PG(14:0)2 , 5.0 nmol of PS(14:0)2 , 0.5 nmol of PE(14:0)2 , 1.0 nmol

of PA(14:0)2 , 2.0 nmol of SM(14:0)2 , 0.02 nmol of LPG(14:0), 0.1 nmol of LPE(14:0), 0.5

nmol of LPC(14:0), 0.1 nmol of LPA(14:0) (purchased from Avanti Polar Lipids, Alabaster, AL,

USA), dissolved in 120 µL of chloroform/methanol (1:1, v/v)) and 1.5 mL of

chloroform/methanol (1:1, v/v) were added to cell homogenates that contained 1 mg of

protein. Subsequently, the mixture was sonicated in a water bath for 5 min, followed by

centrifugation at 15.000 x g for 5 min. The organic layer was transferred to a glass vial and

evaporated to dryness under a nitrogen steam at 60ᵒC. Subsequently, the residue was

dissolved in 200 µL of chloroform/methanol/water (50:45:5,v/v/v) containing 0.01% of

NH4OH, and 20 µL of the solution was injected into the liquid chromatography‐mass

spectrometry (LC/MS) system. LC/MS analysis was performed as described in (15). In

negative mode, mass spectra of phospholipid molecular species were obtained by

continuous scanning from m/z 380 to m/z 1100 with a scan time of 1 s. In positive mode,

mass spectra of PC, LPC and SM were obtained by continuous scanning for product ions with

m/z 184 from m/z 400 to 1000. Mass spectra for PE, LPE and PE plasmalogens were obtained

by scanning on neutral losses of 141 from m/z 400 to 1000. The raw LC/MS data were

converted to mzXML format using msConvert (16) for the negative scan data and ReAdW (17)

for the positive specific scan data.

Chapter 5

138

Data pre‐processing

All LC‐MS spectra of the samples were pre‐processed with XCMS (11). In brief, in the first

step all peaks in the individual LC‐MS spectra were found using the matched filtration

method. The XCMS function ‘matchedFilter’ was used with parameters fwhm=11, max=5,

snthresh=7, steps=2, profMethod=Intlin, step=0.2, mzdiff=0.2. Some spectra from the

negative scan mode contain neighboring or partly overlapping peaks (e.g., for PG and BMP)

that have identical m/z values and similar retention times. To detect such peaks we

modified the matched filtration method by recalculating the second‐order derivative

Gaussian filter after the integration of each peak within the same m/z slice. For the

precursor ion scans, we minimized the number of false positive peaks by retaining only

peaks with at least three consecutive points above the signal‐to‐noise threshold in the peak

list.

After peak finding, peaks from different samples were matched and grouped together into

a peak group, while accounting for small variations in m/z and retention time. The XCMS

‘group’ function was used with parameters family=symmetric and smooth=linear.

Subsequently, chromatograms from the samples were aligned with the function ‘retcor’

(minfrac=0.80, bw=2). In the last step of XCMS, a table was generated comprising the peak

groups, with median mass and retention time calculated from all samples.

Assignment of compound names

For the identification of metabolites, the assignment of compound labels from a certain

class of PL was limited to a subset of peak groups, contained within a box spanning a specific

m/z and retention time range, covering the whole PL cluster. Subsequently, a line was

defined (as shown in Figure 3a) that runs through the cluster of peak groups. Compound

names were then assigned to peak groups that are in close proximity to the line. Duplicate

assignments were resolved based on the likelihood of each separate assignment to occur,

taking into account the patterns of peaks. The steps in the assignment are explained in more

detail below:

Step 1. Generation of an m/z look‐up table for lipid compounds;

Step 2. Assignment of labels within a PL cluster;

Step 3. Collecting all peak groups;

Step 4. Manual correction of incorrect assignments.

Step 1. Generation of an m/z look‐up table for PL compounds. A look‐up table, containing

the chemical composition and corresponding masses (m/z) in positive and negative scan

modes for all theoretically possible combinations of chain length and number of double

bonds for every class of PL, was generated. For a single fatty acid side chain, the chain

lengths ranged from 14 to 28, and the number of double bonds ranged from 0 to 6; multiples

of these were used for PLs with more than one variable side chain. For each chemical


139

structure, the expected isotope percentages were calculated using the R package Rdisop (18)

and added to the table.

Step 2. Assignment of labels within a PL cluster. A plot of median m/z versus median RT for

all peak groups (Figure 2) was used to determine the location of the clusters of peak groups

that originated from the same class of PL. For each PL class, an m/z and RT box was defined

based on visual inspection of the plot and a line through two peak groups in the cluster was

used to define the location and direction (slope) of the peak groups in that particular cluster

(Figure 3). Assignments of PLs within a cluster were done by matching the m/z of each peak

group to the masses in the compound look‐up table, allowing for small deviations (±0.2

atomic units for the data set described in this paper, or ±3 ppm for high‐resolution data).

The context of the peaks was also incorporated in the assignment step: checking for

systematic patterns in lines of compounds with the same chain length or degree of

saturation (Figure 3B). Whenever more than one assignment within the same class was

possible, (e.g., PC(37:0) or PC(38:7), both with m/z 804.6), the most likely assignment was

chosen. For the assignment of peaks from doubly‐charged molecular species (e.g. CL and

mlCL), only peaks with isotope peaks at m/z ± 0.5 were considered.

Step 3. Collecting all peak groups. The lists of peak groups for each PL class, assigned in Step

2, were collated. Multiple assignments for one peak (e.g., PC(38:7) or pPC(39:6), m/z 804.6),

were resolved based on the patterns of the peak groups in the different clusters.

All assigned peak groups were then mapped back onto the original peak group list, so that

the full list contains a combination of assigned and non‐assigned peaks.

Step 4. Manual correction of incorrect assignments. By investigating the assignments in the

close‐up plots that were generated for every PL cluster (Figure 3B), erroneous or suspicious

assignments were edited and corrected. Generally, about 95% of the assigned peak groups

were identified correctly by the pipeline.

Data processing

After the identification of the monoisotopic peaks for different PLs, their isotope peaks were

marked in the peak group list. Intensities for observed isotope peaks were corrected using

the method described by (19) in order to obtain deconvoluted intensities for peak groups

which overlap with isotope peaks from other compounds. For normalization purposes,

intensities of peaks belonging to a certain PL class were scaled to the response of the IS of

that class. Thus, the intensity of the IS peak was set to the corresponding concentration of

that IS that was added to each sample, and the intensities of the other peaks of the same

PL class could be estimated semi‐quantitatively, relative to that of the IS peak. Lastly, the

intensities of all peaks were normalized to the protein concentration of the homogenate.

Chapter 5

140

The result of the pre‐processing pipeline as described, is a table in Excel format that contains

all measured peak groups, characterized by their median m/z and RT values and, where

possible, their identified compound name (see supplemental files 1‐4). Isotope‐corrected,

normalized intensities and per‐feature statistics based on the normalized and corrected

values are included in the table, as well as EIC plots for the peak areas and boxplots, to allow

for quick visual assessment of the various peaks.

Acknowledgement/grant support

L.P. was supported by a grant of Stichting Nationaal Fonds Tegen Kanker (SNFK) in the

Netherlands.


141

References

1. Vaz, F. M., M. Pras‐Raves, A. H. Bootsma, and A. H. van Kampen. 2015. Principles and practice of lipidomics. J Inherit Metab Dis 38: 41‐52. 2. Hermansson, M., A. Uphoff, R. Kakela, and P. Somerharju. 2005. Automated quantitative analysis of complex lipidomes by liquid chromatography/mass spectrometry. Anal Chem 77: 2166‐2175. 3. Fahy, E., D. Cotter, R. Byrnes, M. Sud, A. Maer, J. Li, D. Nadeau, Y. Zhau, and S. Subramaniam. 2007. Bioinformatics for lipidomics. Methods Enzymol 432: 247‐273. 4. Sparvero, L. J., A. A. Amoscato, C. E. Dixon, J. B. Long, P. M. Kochanek, B. R. Pitt, H. Bayir, and V. E. Kagan. 2012. Mapping of phospholipids by MALDI imaging (MALDI‐MSI): realities and expectations. Chem Phys Lipids 165: 545‐562. 5. Blanchard, A. P., G. S. McDowell, N. Valenzuela, H. Xu, S. Gelbard, M. Bertrand, G. W. Slater, D. Figeys, S. Fai, and S. A. Bennett. 2013. Visualization and Phospholipid Identification (VaLID): online integrated search engine capable of identifying and visualizing glycerophospholipids with given mass. Bioinformatics 29: 284‐285. 6. Boccard, J., J. L. Veuthey, and S. Rudaz. 2010. Knowledge discovery in metabolomics: an overview of MS data handling. J Sep Sci 33: 290‐304. 7. Sud, M., E. Fahy, D. Cotter, E. A. Dennis, and S. Subramaniam. 2012. LIPID MAPS‐Nature Lipidomics Gateway: An Online Resource for Students and Educators Interested in Lipids. J Chem Educ 89: 291‐292. 8. Taguchi, R., and M. Ishikawa. 2010. Precise and global identification of phospholipid molecular species by an Orbitrap mass spectrometer and automated search engine Lipid Search. J Chromatogr A 1217: 4229‐4239.

9. Kind, T., K. H. Liu, Y. Lee do, B. DeFelice, J. K. Meissen, and O. Fiehn. 2013. LipidBlast in silico tandem mass spectrometry database for lipid identification. Nat Methods 10: 755‐758. 10. Grebe, S. K., and R. J. Singh. 2011. LC‐MS/MS in the Clinical Laboratory ‐ Where to From Here? Clin Biochem Rev 32: 5‐31. 11. Smith, C. A., E. J. Want, G. O'Maille, R. Abagyan, and G. Siuzdak. 2006. XCMS: processing mass spectrometry data for metabolite profiling using nonlinear peak alignment, matching, and identification. Anal Chem 78: 779‐787. 12. Mullauer, F. B., L. van Bloois, J. B. Daalhuisen, M. S. Ten Brink, G. Storm, J. P. Medema, R. M. Schiffelers, and J. H. Kessler. 2011. Betulinic acid delivered in liposomes reduces growth of human lung and colon cancers in mice without causing systemic toxicity. Anticancer Drugs 22: 223‐233. 13. Potze, L., S. Di Franco, C. Grandela, M. L. Pras‐Raves, D. I. Picavet, H. A. van Veen, H. van Lenthe, et al. 2015. Betulinic acid induces a novel cell death pathway that depends on cardiolipin modification. Oncogene. 14. Smith, P. K., R. I. Krohn, G. T. Hermanson, A. K. Mallia, F. H. Gartner, M. D. Provenzano, E. K. Fujimoto, N. M. Goeke, B. J. Olson, and D. C. Klenk. 1985. Measurement of protein using bicinchoninic acid. Anal Biochem 150: 76‐85. 15. Houtkooper, R. H., R. J. Rodenburg, C. Thiels, H. van Lenthe, F. Stet, B. T. Poll‐The, J. E. Stone, C. G. Steward, R. J. Wanders, J. Smeitink, W. Kulik, and F. M. Vaz. 2009. Cardiolipin and monolysocardiolipin analysis in fibroblasts, lymphocytes, and tissues using high‐performance liquid chromatography‐mass spectrometry as a diagnostic test for Barth syndrome. Anal Biochem 387: 230‐237. 16. Chambers, M. C., B. Maclean, R. Burke, D. Amodei, D. L. Ruderman, S. Neumann, L. Gatto, et al.. 2012. A cross‐platform toolkit for mass spectrometry and proteomics. Nat Biotechnol 30: 918‐920.

Chapter 5

142

17. Pedrioli, P. G., J. K. Eng, R. Hubley, M. Vogelzang, E. W. Deutsch, B. Raught, B. Pratt, E. Nilsson, et al 2004. A common open representation of mass spectrometry data and its application to proteomics research. Nat Biotechnol 22: 1459‐1466. 18. Bocker, S., M. C. Letzel, Z. Liptak, and A. Pervukhin. 2009. SIRIUS: decomposing isotope patterns for metabolite identification. Bioinformatics 25: 218‐224. 19. Liebisch, G., B. Lieser, J. Rathenberg, W. Drobnik, and G. Schmitz. 2004. High‐throughput quantification of phosphatidylcholine and sphingomyelin by electrospray ionization tandem mass spectrometry coupled with isotope correction algorithm. Biochim Biophys Acta 1686: 108‐117.


143

Supplementary figures/tables

Supplementary figure 1: Heatmap of differences in BMP and PG composition. (A) The BMPs have been

sorted by total number of double bonds in the combined side chains. (B) The PGs have been sorted by

total number of double bonds in the combined side chains. In both heatmaps no clear distinction between

DMSO and BetA treated samples are observed.

Supplemental table I: Number of PLs identified per major class in Negative scan mode

Monoisotopic peaksa Monoisotopic + isotope peaksa

BMP 30 50

CL2 69 147

mlCL2 32 62

PA 12 33

PE 29 53

PG 23 43

PI 51 108

pPE 16 27

PS 23 48 aIn full negative mode, 285 out of 2163 (13%) peak groups can be identified as monoisotopic and 571 out of

2163 (26.4%) can be identified as monoisotopic plus isotope peaks. Numbers in the Results section refer to

a truncated list, where the weakest peak groups have been omitted to improve the clarity of Figure 2.

Chapter 5

144

Supplemental table II: Number of PLs per major class identified Positive scan mode (PR184)

Monoisotopic peaks Monoisotopic + isotope peaks

PC 66 151

pPC 23 45

aPC 3 11

LPC 25 58

LpPC 5 9

SM 22 42

Supplemental table III: Number of PLs identified per major class in Positive scan mode (NL141)

Monoisotopic peaksa Monoisotopic + isotope peaksa

PE 39 89

pPE 13 24

aPE 17 27

LPE 10 14 aIn full PR184 scan mode, 144 out of 785 (18.3%) peak groups can be identified as monoisotopic and 316

out of 785 (40.2%) can be identified as monoisotopic plus isotope peaks. In full NL141 scan mode, 74 out of

225 (32.9%) and 154 out of 225 (68.4%) can be identified as monoisotopic plus isotope peaks. Numbers in

the Results section refer to a truncated list, where the weakest peak groups have been omitted.

Supplemental table IV: Clusters of PLs that can be identified using the algorithma

One variable side

chain

Two variable side

chains

Three variable side

chains

Four variable side

chains

LPC, LPG, LPS, LPE,

LPI, LPA, SM, ST,

STOH, LpPE, LpPC,

LaPE, LlyPG, Cer, FA,

FAOH

PC, diLPC, diLpPC,

diSM, diSMCl, diLPE,

diLpPE, PG, PS, PE, PI,

PA, BMP, pPE, aPE,

pPC, aPC, pPA, aPA,

dlCL1, dlCL2,

mlNAPS1, mlNAPS2,

lyPG, PIP, diFA, DAG

mlCL1, mlCL2, NAPS2,

NAPS1, SLBPA, TAG,

triLPE

CL1, CL2, CL2OH,

diPG, diPE, diPA,

dipPE, diPCCl, diPS,

diPC , dipPC

aLPL; lyso‐PL, pPL; plasmalogen PL, aPL; O‐alkyl PL, diPL; dimeric PL, mlPL; monolyso‐PL, FA; fatty acid, NAPS;

N‐acetyl PS, suffix 1 (as in mlCL1); singly charged, suffix 2; doubly charged, suffix OH; hydroxy, suffix Cl;

chloride adduct. SLBPA; semilysobisphosphatidic acid (a.k.a. hemi‐BMP), DAG; diacylglycerol, TAG;

triacylglycerol.

Chapter 6

General Discussion

“Curiouser and curiouser!”


Discussion

149

The role of autophagy in BetA‐induced cell death

Since the discovery of BetA as an anti‐cancer compound, research has focused on finding

the mechanism of action of this natural compound. Research quickly showed that the

mitochondria were involved and apoptosis was induced, although independent of

BAK/BAX.(1) When a caspase inhibitor was used, these compounds did not prevent cell

death, indicating that alternative, caspase‐independent cell death pathways must be

activated. In chapter 2, we describe that necroptosis is not induced, however a massive and

rapid form of autophagy is induced. This induction of autophagy could be blocked by

cyclosporine A, an inhibitor of the permeability transition pore. Blocking the opening of this

pore by cyclosporine A resulted also in a block in apoptosis as previously described.(2) This

observation suggests that autophagy is a consequence of mitochondrial damage triggered

by BetA and can be prevented by inhibition of PT‐pore opening. We also observed that in

autophagy knock out cells, enhanced cell death was observed even when apoptosis was

blocked by caspase inhibitors. These data suggest that autophagy acts as a survival

mechanism rather than the cell death execution machinery and that there must be a yet

undefined other cell death mechanism involved that is induced by BetA. Alternative

pathways to death have been reported and could for instance involve lysosomal

permeabilization. Interestingly one of the compounds that is reported to induce lysosomal

destabilization, siramesine, shares some properties with BetA, as it also induces cell death

independent of caspases, BCL‐2 and P53,(3‐5) pointing to a potential mechanistic overlap.(4, 5)

Lysosomal membrane permeabilization can be measured using fluorescent labeled galectin‐

1 or galectin‐3 cells(6) and experiments to test the role of the lysosome in BetA‐induced cell

death should be conducted. It is important to note though that we have shown that

inhibition of cathepsin B, which is reported to be crucial for the lysosomal destabilization (7),

did not influence BetA‐induced cell death, arguing against lysosomal membrane

destabilization as the causal pathway to cell death. Therefore more work needs to be

performed and we decided to focus more in detail on how the mitochondrial BetA‐induced

cell death occurred.

Differential effects of BetA on cancer cells and normal cells

One characteristic of BetA is that its cytotoxic effects are observed on cancer cells and not

on healthy cells.(8, 9) For a long time it has remained enigmatic as to why this differential

effect is observed. In chapter 3, we described that interfering with the fatty acid metabolism

of a cancer cell, by inhibiting stearoyl CoA desaturase enzyme, the side chains of the

phospholipid cardiolipin become more saturated, which results in morphological changes

in the inner membrane of the mitochondria followed by cell death. We also confirmed that

the cytotoxic effect of BetA is only on cancer cells and not on healthy cells. We believe these

observations could explain the selectivity for cancer cells as a hallmark of cancer is altered

metabolism. That is, cancer cells have a different lipid metabolism as compared to healthy

Chapter 6

150

cells. Cancer cells increase their de novo fatty acid synthesis, to fulfill the need for new lipids

in order to create new biomass which is needed even under scarce conditions. It is energetic

very inefficient to generate de novo fatty acids while enough fatty acids can be taken up by

fatty acids transporters from the surrounding suggesting that cancer cells benefit from de

novo fatty acids for yet undefined reasons. The end product of de novo fatty acid synthesis

is palmitic acid. This saturated fatty acid is converted to unsaturated fatty acids by SCD‐1

and further elongated by several elongation enzymes to create longer fatty acids. In several

types of cancer, SCD‐1 is overexpressed.(10) Cancer cells have become addicted to the

constant rate of conversion of saturated fatty acids into unsaturated fatty acids and thereby

the levels of SCD‐1 they express. BetA and SCD‐1 inhibitors affect this balance and thereby

create a toxic pool of saturated fatty acids. Importantly, in healthy cells basal levels of

saturated fatty acids are lower as compared to tumor cells (11) as normal cells do not utilize

the de novo synthesis pathway. Instead, healthy cells take up unsaturated fatty acids from

the surrounding mainly provided by food intake. Thereby inhibiting the activity of SCD‐1

shows less impact and is therefore less harmful to healthy cells as they are not dependent

on this pathway (described in chapter 3).(12) Unsaturated fatty acids are important

precursors for various products in the cell, like phospholipids (cell membrane),

diacyglycerols (signaling) and triglycerides (energy storage).(13) Overexpression of SCD‐1 is a

result of the overall increase of de novo fatty acid synthesis where besides SCD‐1 also sterol

regulatory element binding proteins (SREBPs) are overexpressed.(14, 15) SREBPs control the

expression of several enzymes (ATP‐citrate lyase (ACL), ACC and FASN) required for

endogenous cholesterol, fatty acid (FA), triacylglycerol and phospholipid synthesis.

Upregulation of these enzymes demands the cell also to increase their SCD‐1 activity to

fulfill the need of an unsaturated fatty acid pool.

The de novo fatty acid synthesis pathway is a potential drug target due to the differential

usage of this pathway between normal cells and cancer cells. Several studies have shown

that inhibition of fatty acid synthesis enzymes by inhibiting FASN, ACC or ACL results in

cancer cell death in vitro and retards the growth of tumors in vivo.(16‐20)

Also shown in chapter 3 is that in cancer cells the saturation levels of the side chains of

cardiolipin are already more saturated as compared to healthy fibroblasts. BetA and SCD‐1

inhibitors treatment result in an even higher saturation grade of the side chains of

cardiolipin. Saturated fatty acid makes membranes more rigid and loss of membrane fluidity

results in loss of function and structure and we hypothesize that this results in the inner

cristae morphology change, which results in permeability pore opening and leakage of

cytochrome c into the cytosol. This initial higher saturation grade of membranes makes

cancer cells more vulnerable for inhibition of SCD‐1 activity as compared to normal cells.

Barth syndrome patients also have highly saturated fatty acid side chains of cardiolipin and

similar morphological changes in the mitochondria are observed.(21, 22) However these

modifications by itself are not sufficient for cytochrome c leakage to occur as most cells in

Discussion

151

Barth syndrome patients show no increase in cell death. This suggests that besides

saturation of cardiolipin side chains, additional changes are needed to induce cell death.

This could be reactive oxygen species (ROS), as inhibiting mitochondrial complex 1 results

in OPA1 (a GTPase in the inner mitochondrial membrane which plays a role in regulating

mitochondrial fusion and pro‐apoptotic remodeling of the mitochondria)

desoligomerization leading to ROS production.(23) This loss of oligomerization of OPA1

results in cytochrome c release and mitochondrial cristae remodeling. It is known that

cancer cells have higher levels of ROS,(24, 25) and therefore it could be that the combination

of higher saturation of cardiolipin and the high amounts of ROS induced in cancer cells

results in cell death.

It is likely that due to the increased need of lipids, cancer cells also has more lipid droplets

which they can use for their supply of fatty acids. During nutrient deprivation, triglycerides

in lipid droplets are hydrolyzed into fatty acids.(26) The breakdown of these stored lipid

droplet products has been shown to occur, besides cytosolic hydrolytic enzymes or lipases,

via a specialized form of autophagy, named lipophagy(27), which link autophagy to lipid

metabolism. Lipophagy may protect cells for apoptosis by breaking down lipids to supply

for energy and prevention of ATP depletion. Another protective role for lipophagy is by

preventing liver injury induced by increased hepatic lipid accumulation or steatosis.(28) Lipids

may also play a role in the regulation of levels of autophagy. It was shown that unsaturated

fatty acid, oleic acid, promoted the formation of triglyceride‐enriched lipid droplets and

induced autophagy, while saturated fatty acid, palmitic acid, was poorly incorporated into

lipid droplets and suppressed autophagy and rather increased apoptosis.(26) This suggest

that formation of lipid droplets and induction of lipophagy is protective for cells.

The differential effect of BetA on cancer cells and normal cells could also be due to massive

induction of autophagy as a response to the induced mitochondrial stress. The autophagy

induced by BetA could also be partially lipophagy, to increase the amount of unsaturated

fatty acids to prevent the toxic effect of accumulating saturated fatty acids by the inhibition

of SCD‐1. Due to the increase of saturated fatty acids, the protective role of lipid droplets

and autophagy are blocked resulting in increased cell death. As normal cells have higher

amounts of unsaturated fatty acids as compared to cancer cells it could be that BetA induces

more autophagy and lipophagy in cancer cells to overcome the toxic effects of saturated

fatty acids. Investigating the role of lipophagy and the blocking of lipophagy, more in detail

could result in potential new targets. Combination of blocking lipophagy and SCD‐1

inhibition might result in a faster more efficient cell death.

The role of cardiolipin in cell death

The toxic saturated fatty acids are incorporated in several phospholipids and one

phospholipid that shows dramatic changes in saturation levels is cardiolipin. Cardiolipin is a

phospholipid that is important for the structure of the inner membrane of the mitochondria

Chapter 6

152

and it is involved in apoptosis. Two main levels of apoptosis regulation by cardiolipin have

been described. The first involves the release of cytochrome c in a BAK/BAX dependent

fashion. Cytochrome c interacts with cardiolipin in the outer leaflet of the IMM through two

independent binding sites, this is needed for the function of cytochrome c as an electron

carrier in the mitochondrial respitory chain.(29‐32) Cardiolipin‐bound cytochrome c acts as a

peroxidase capable of catalyzing H2O2‐dependent peroxidation of cardiolipin. This

cardiolipin oxidation is an essential step in the release of cytochrome c during apoptosis.(33)

The second level of cardiolipin‐dependent apoptosis regulation that is suggested to occur,

involves Bid‐induced cytochrome c release. During apoptosis, Bid is cleaved by caspase 8 to

produce tBid which translocates to the mitochondria. tBid has been suggested to bind to

CL‐enriched contact sites and induces the translocation of BAK and BAX to the

mitochondrial outer membrane.(34‐36)

Cardiolipin is therefore suggested to serves as a platform to allow apoptosis via the BCL2

family to occur. Whether cardiolipin saturation levels change this platform function is not

clarified, but is a possibility. The level of saturation is however clearly of importance to the

physiological function of cardiolipin and lipids in general. In biological membranes one finds

unsaturated fatty acids of which the structural properties allow for membrane fluidity.

Cardiolipin has four acyl side chains that are highly unsaturated to maintain normal cellular

function.(34) These side chains can be remodeled via tafazzin. We hypothesized that BetA

besides SCD‐1 inhibition also increases the turnover of lipids and thereby the incorporation

of saturated fatty acids into cardiolipin. As cardiolipin regulates and is involved in

permeability pore opening (34, 37) we hypothesized that the saturation of cardiolipin results

in morphological and functional changes in the mitochondrial cristae and leads to

permeability pore opening and cytochrome c release.

The mitochondrial network is maintained by a continuous process of fusion (elongation of

mitochondria) and fission (fragmentation of mitochondria). Whether fusion/fission is

important for cytochrome c release is still debated, however during apoptosis mitochondria

are remodeled via activation of the fission machinery and synchronal neutralization of the

fusion machinery.(38, 39)

Mitochondrial fission occurs at the same time as the activation of BAX, mitochondrial outer

membrane permeabilization and the resulting cytochrome c release.(40) On top of that, it

was shown that disruption of OPA1 complexes results in cytochrome c release and

mitochondrial cristae remodeling.(23) These data indicate that mitochondrial fission/fusion

and cytochrome c release are linked to each other. Cardiolipin also plays a role in

mitochondrial fission/fusion(34) and therefore an increase in saturation grade of the

cardiolipin side chains could result in the observed mitochondrial fragmentation. How this

fragmentation is induced by cardiolipin needs further investigation. Cardiolipin side chain

Discussion

153

remodeling into more saturated is therefore a very interesting mechanism for further

investigation and possible therapeutic targets.

The role of SCD‐1 in cancer therapy

The differential effects of both BetA and SCD‐1 inhibition on healthy cells and tumor cells

makes both compounds very interesting to investigate further for therapeutic applications.

In the past pharmaceutical companies started to look for small molecule SCD‐1 inhibitors to

treat metabolic diseases like obesity, diabetes and fatty liver as these diseases are

associated with up‐regulation of SCD‐1. Since 2005 several patents have been granted and

many of these compounds have good in vitro activity and pharmacodynamics in rodent

models, unfortunately, only a few of these inhibitors have progressed to clinical trials and

no reports can be found of compounds beyond phase IIa studies for obesity, diabetes of

fatty liver disease treatment. The differences of lipid metabolism in rodents to humans is

given as the main explanation.(41) Research in the last years have shown that SCD‐1 plays a

role in the regulation of metabolism and growth signaling pathways in cancer. SCD‐1

contributes to maintain a shift in lipid metabolism and intracellular signaling (i.e. activation

of Akt signal and deactivation of AMPK pathway). This results in an accelerated rate of cell

proliferation, increased invasiveness and enhanced survival.(10) Several studies have shown

that inhibition of SCD‐1 (either genetically or pharmaceutical) leads to reduced cancer

formation in vivo and transformation in vitro.(10, 42) The mechanism by which SCD‐1 inhibition

leads to tumor reduction has been unclear. A hypothesis mentioned in literature, is that

changes in lipid composition result in ER stress and trigger unfolded protein response

(UPR).(43)

We show that inhibition of SCD‐1 by chemical inhibition (BetA and SCD‐1 inhibitor) leads to

increased saturation of CL. This saturation leads to morphological changes in the inner

membrane of the mitochondria and results in cell death. How exactly the changes in

saturation and morphology results in cell death remains unclear, however we hypothesize

that the change in the inner membrane results in opening of the PT‐pore and thereby

release of cytochrome c into the cytosol. (chapter 3) In our research we have not tested ER

stress and UPR. It could be that the ER stress inflicted by shifting the balance between

saturated fatty acids and unsaturated fatty acids precedes the mitochondrial stress we

observe.

Currently no clinical trials are ongoing for SCD‐1 inhibition (clinical trial.gov) in the

treatment of cancer. Inhibition of SCD‐1 as monotherapy could not be optimal as we

observed that inhibition of SCD‐1 results in a slower cell death as compared to BetA.

Therefore the focus should be more on combination therapy, where besides SCD‐1

inhibition another stressor is applied, resulting in synergy of both compounds. We have

tried to find this combination of SCD‐1 inhibitor and a compound that results in the same

efficacy and efficiency as BetA‐induced cell death. However several compounds that

Chapter 6

154

stressed the mitochondria did not result in cell death synergy when combined with SCD‐1

inhibition. Further research into understanding BetA mechanism and synthetic lethality

using SCD‐1 inhibitor could lead to a beneficial non‐toxic combination therapy.

Betulinic acid as cancer therapy

Besides looking into SCD‐1 inhibition for therapeutic approaches, BetA also has some

potential. Research by others and previously by our group has shown BetA has no toxic

effects in vivo.(8, 9) The development for BetA as a potential drug is not as advanced as SCD‐

1 inhibitors, however the higher efficacy of BetA makes it a better compound as compared

to SCD‐1 inhibitors. Unfortunately, BetA is a very lipophilic compound and therefore harder

to administer to patients. It is known that lipophilic drugs can be more toxic because they

bind (i.e to receptors, plasma, proteins or tissue) while unbound drugs most of the time are

responsible for the efficacy of the drug.(44, 45) For lipophilic compounds, there are usually

much higher concentrations of bound drug than unbound drug, therefore toxicity will be

more likely if it is determined by total drug concentrations as higher doses are needed to

get enough unbound drug in the blood system. However up to now, even when high doses

of BetA were used in vivo no toxic effects were observed in normal tissue although the

formulations used are not suited for the clinic.(8, 9) Accordingly there is a need for a non‐

toxic formulation which can deliver BetA, preferably specific, to the tumor.

Cancer stem cells targeted by BetA

The cancer stem cell model hypothesizes that cancers are organized into a hierarchy of

subpopulations of tumorigenic cancer stem cells and their non‐tumorigenic progeny.(46‐48)

Cancer stem cells are defined as a subset of tumor cells which possess self‐renewal and

multi‐lineage differentiation potential.(49, 50) Cancer stem cells are known to be more

resistant to conventional treatment due to their (acquired) resistance mechanisms.(51, 52) In

chapter 4 we show that BetA can induce a rapid cell death in colon cancer stem cells. We

hypothesize that this induction of cell death is SCD‐1 dependent as SCD‐1 inhibition also

kills colon cancer stem cells, although the induction takes longer than BetA. The precise

mechanism how BetA induces cell death in colon cancer stem cells is still unclear and further

research should reveal if indeed BetA results in SCD‐1 inhibition and a saturation of CL in

these cells leading to apoptosis. Due to the rapid induction of cell death in colon cancer

stem cells as compared to cancer cells a different mechanism might be involved. It could be

that colon cancer stem cells rely more on their fatty acids as they possess more lipid droplets

than regular cancer cells.(53) Interference in the unsaturation grade of these fatty acids could

be a weak spot for these stem cells. Cancer stem cells protect themselves from classical

chemotherapy by using BCL‐XL‐dependent mechanisms (54), as BetA‐induced killing is

independent of the BCL‐2 family this could be a reason why BetA targets the cancer stem

cells. How CSCs die exactly remains to be elucidated. However these data suggest that CSCs

Discussion

155

could be targeted very efficiently if the precise mechanism of action of BetA in these cells is

known.

Lipidomics as a tool for diagnostics and monitoring of patient.

Studying lipids is complex due to the diversity of many lipids. Analysis of lipids was

traditionally done by thin‐layer chromatography, gas chromatography and mass

spectrometry. Recent years more and more technical advances in mass spectrometry have

created the ability to study lipids in a metabolomics way the so‐called lipidomics. Lipidomics

aims to study the pathways and networks of lipids by the characterization and

quantification of all lipids in a sample.(55) In chapter 5 we described a new lipidomics pipeline

which we have tested for BetA treated samples. We show that using the lipidomics pipeline

we can identify and separate a lot of lipids based on their features i.e their saturation

grades. We show that we can distinguish samples that are DMSO or BetA treated based on

their CL saturation grade. This could be used in the clinic to discriminate healthy individuals

from Barth syndrome patients, but also if SCD‐1 inhibitor of BetA makes it into the clinic this

pipeline can be used to monitor treatment response. This method could be used for

monitoring treatment response of compounds that have BetA and/or SCD‐1 inhibitor mode

of action or monitor the effect of SCD‐1 inhibitors and/or BetA when it makes it into the

clinic. With further optimalization of the lipidomics (i.e improved high resolution mass

spectrometry which result in better detection of masses with a higher accuracy and thus

leading to more reliable identification of lipids) this pipeline could also be used for validation

of new compounds to see if they exert the mechanism to inhibit SCD‐1 and thereby

increasing the saturation of CL.

Besides playing a role in cancer, lipids and their metabolism are also involved in several

other diseases.(55, 56) This makes lipidomics a great new tool for diagnostics. The ultimate

goal of lipidomics would be to have one patient sample and measure the complete lipidome

in a fast and accurate fashion so the diagnosis of patient disease could be made. Besides

diagnosis, monitoring of disease progression, treatment response and biomarkers research

are possible.

Concluding remarks

There are still a lot of things we do not understand about BetA and how it induces cell death.

The observations that when apoptosis is blocked by caspase inhibitors, BetA still induces

cell death indicates that a different form of cell death or even a new pathway remains to be

elucidated. And up to now due to its lipophilic character no clinical trials are attempted. In

this thesis we provide data that contribute to the understanding of BetA‐induced apoptotic

cell death. The finding that inhibition of SCD‐1 by BetA results in saturation of cardiolipin

and cytochrome c release opens a new field for cancer research and potential new use for

SCD‐1 inhibitors in the clinic. Targeting cancer stem cells is a kind of Holy Grail in cancer

Chapter 6

156

research, as eradicating all tumor cells is wanted. Interestingly, applying BetA on colon

cancer stem cells resulted in a massive and fast cell death in these otherwise therapy

resistant cells. Discovering the precise mechanism of action of BetA in these cells will

provide the cancer field with potential new drug targets a possible therapy for hard to target

cancers. Monitoring drug response can be done by using the lipidomics pipeline described

in this thesis and this could result in faster diagnosis and specialized patient care. Taken

together, the new findings in this thesis can provide a basis for future studies for cancer

therapy based on cancer metabolism and especially SCD‐1 inhibition and cardiolipin

saturation and on the quest of the mechanism of action of BetA.

Discussion

157

References

1. Fulda S, Friesen C, Los M, Scaffidi C, Mier W, Benedict M, et al. Betulinic acid triggers CD95 (APO‐1/Fas)‐ and p53‐independent apoptosis via activation of caspases in neuroectodermal tumors. Cancer Res. 1997;57(21):4956‐64. 2. Mullauer FB, Kessler JH, Medema JP. Betulinic acid induces cytochrome c release and apoptosis in a Bax/Bak‐independent, permeability transition pore dependent fashion. Apoptosis. 2009;14(2):191‐202. 3. Ostenfeld MS, Fehrenbacher N, Hoyer‐Hansen M, Thomsen C, Farkas T, Jaattela M. Effective tumor cell death by sigma‐2 receptor ligand siramesine involves lysosomal leakage and oxidative stress. Cancer research. 2005;65(19):8975‐83. 4. Ostenfeld MS, Hoyer‐Hansen M, Bastholm L, Fehrenbacher N, Olsen OD, Groth‐Pedersen L, et al. Anti‐cancer agent siramesine is a lysosomotropic detergent that induces cytoprotective autophagosome accumulation. Autophagy. 2008;4(4):487‐99. 5. Groth‐Pedersen L, Ostenfeld MS, Hoyer‐Hansen M, Nylandsted J, Jaattela M. Vincristine induces dramatic lysosomal changes and sensitizes cancer cells to lysosome‐destabilizing siramesine. Cancer research. 2007;67(5):2217‐25. 6. Aits S, Kricker J, Liu B, Ellegaard AM, Hamalisto S, Tvingsholm S, et al. Sensitive detection of lysosomal membrane permeabilization by lysosomal galectin puncta assay. Autophagy. 2015:0. 7. Joy B, Sivadasan R, Abraham TE, John M, Sobhan PK, Seervi M, et al. Lysosomal destabilization and cathepsin B contributes for cytochrome c release and caspase activation in embelin‐induced apoptosis. Molecular carcinogenesis. 2010;49(4):324‐36. 8. Mullauer FB, van BL, Daalhuisen JB, Ten Brink MS, Storm G, Medema JP, et al. Betulinic acid delivered in liposomes reduces growth of

human lung and colon cancers in mice without causing systemic toxicity. Anticancer Drugs. 2011;22(3):223‐33. 9. Zuco V, Supino R, Righetti SC, Cleris L, Marchesi E, Gambacorti‐Passerini C, et al. Selective cytotoxicity of betulinic acid on tumor cell lines, but not on normal cells. Cancer Lett. 2002;175(1):17‐25. 10. Igal RA. Stearoyl‐CoA desaturase‐1: a novel key player in the mechanisms of cell proliferation, programmed cell death and transformation to cancer. Carcinogenesis. 2010;31(9):1509‐15. 11. Li J, Ding SF, Habib NA, Fermor BF, Wood CB, Gilmour RS. Partial characterization of a cDNA for human stearoyl‐CoA desaturase and changes in its mRNA expression in some normal and malignant tissues. IntJCancer. 1994;57(3):348‐52. 12. Scaglia N, Chisholm JW, Igal RA. Inhibition of stearoylCoA desaturase‐1 inactivates acetyl‐CoA carboxylase and impairs proliferation in cancer cells: role of AMPK. PLoSOne. 2009;4(8):e6812. 13. Roongta UV, Pabalan JG, Wang X, Ryseck RP, Fargnoli J, Henley BJ, et al. Cancer cell dependence on unsaturated fatty acids implicates stearoyl‐CoA desaturase as a target for cancer therapy. MolCancer Res. 2011;9(11):1551‐61. 14. Eberle D, Hegarty B, Bossard P, Ferre P, Foufelle F. SREBP transcription factors: master regulators of lipid homeostasis. Biochimie. 2004;86(11):839‐48. 15. Espenshade PJ. SREBPs: sterol‐regulated transcription factors. J Cell Sci. 2006;119(Pt 6):973‐6. 16. Kuhajda FP. Fatty‐acid synthase and human cancer: new perspectives on its role in tumor biology. Nutrition. 2000;16(3):202‐8. 17. Kuhajda FP, Jenner K, Wood FD, Hennigar RA, Jacobs LB, Dick JD, et al. Fatty acid

Chapter 6

158

synthesis: a potential selective target for antineoplastic therapy. Proceedings of the National Academy of Sciences of the United States of America. 1994;91(14):6379‐83. 18. Brusselmans K, De Schrijver E, Verhoeven G, Swinnen JV. RNA interference‐mediated silencing of the acetyl‐CoA‐carboxylase‐alpha gene induces growth inhibition and apoptosis of prostate cancer cells. Cancer research. 2005;65(15):6719‐25. 19. Menendez JA, Lupu R. Fatty acid synthase and the lipogenic phenotype in cancer pathogenesis. NatRevCancer. 2007;7(10):763‐77. 20. Mashima T, Seimiya H, Tsuruo T. De novo fatty‐acid synthesis and related pathways as molecular targets for cancer therapy. Br J Cancer. 2009;100(9):1369‐72. 21. Barth PG, Scholte HR, Berden JA, Van der Klei‐Van Moorsel JM, Luyt‐Houwen IE, Van 't Veer‐Korthof ET, et al. An X‐linked mitochondrial disease affecting cardiac muscle, skeletal muscle and neutrophil leucocytes. JNeurolSci. 1983;62(1‐3):327‐55. 22. Houtkooper RH, Rodenburg RJ, Thiels C, van LH, Stet F, Poll‐The BT, et al. Cardiolipin and monolysocardiolipin analysis in fibroblasts, lymphocytes, and tissues using high‐performance liquid chromatography‐mass spectrometry as a diagnostic test for Barth syndrome. AnalBiochem. 2009;387(2):230‐7. 23. Ramonet D, Perier C, Recasens A, Dehay B, Bove J, Costa V, et al. Optic atrophy 1 mediates mitochondria remodeling and dopaminergic neurodegeneration linked to complex I deficiency. Cell Death Differ. 2013;20(1):77‐85. 24. Pelicano H, Carney D, Huang P. ROS stress in cancer cells and therapeutic implications. Drug ResistUpdat. 2004;7(2):97‐110. 25. Nogueira V, Hay N. Molecular pathways: reactive oxygen species homeostasis in

cancer cells and implications for cancer therapy. ClinCancer Res. 2013;19(16):4309‐14. 26. Singh R, Kaushik S, Wang Y, Xiang Y, Novak I, Komatsu M, et al. Autophagy regulates lipid metabolism. Nature. 2009;458(7242):1131‐5. 27. Liu K, Czaja MJ. Regulation of lipid stores and metabolism by lipophagy. Cell death and differentiation. 2013;20(1):3‐11. 28. Reddy JK, Rao MS. Lipid metabolism and liver inflammation. II. Fatty liver disease and fatty acid oxidation. American journal of physiology Gastrointestinal and liver physiology. 2006;290(5):G852‐8. 29. Paradies G, Petrosillo G, Pistolese M, Ruggiero FM. Reactive oxygen species generated by the mitochondrial respiratory chain affect the complex III activity via cardiolipin peroxidation in beef‐heart submitochondrial particles. Mitochondrion. 2001;1(2):151‐9. 30. Schug ZT, Gottlieb E. Cardiolipin acts as a mitochondrial signalling platform to launch apoptosis. BiochimBiophysActa. 2009;1788(10):2022‐31. 31. Gorbenko GP. Structure of cytochrome c complexes with phospholipids as revealed by resonance energy transfer. Biochim Biophys Acta. 1999;1420(1‐2):1‐13. 32. Iverson SL, Orrenius S. The cardiolipin‐cytochrome c interaction and the mitochondrial regulation of apoptosis. Arch Biochem Biophys. 2004;423(1):37‐46. 33. Kagan VE, Tyurin VA, Jiang J, Tyurina YY, Ritov VB, Amoscato AA, et al. Cytochrome c acts as a cardiolipin oxygenase required for release of proapoptotic factors. Nat Chem Biol. 2005;1(4):223‐32. 34. Houtkooper RH, Vaz FM. Cardiolipin, the heart of mitochondrial metabolism. Cell MolLife Sci. 2008;65(16):2493‐506.

Discussion

159

35. Esposti MD, Cristea IM, Gaskell SJ, Nakao Y, Dive C. Proapoptotic Bid binds to monolysocardiolipin, a new molecular connection between mitochondrial membranes and cell death. Cell Death Differ. 2003;10(12):1300‐9. 36. Lutter M, Fang M, Luo X, Nishijima M, Xie X, Wang X. Cardiolipin provides specificity for targeting of tBid to mitochondria. Nat Cell Biol. 2000;2(10):754‐61. 37. Scorrano L, Ashiya M, Buttle K, Weiler S, Oakes SA, Mannella CA, et al. A distinct pathway remodels mitochondrial cristae and mobilizes cytochrome c during apoptosis. DevCell. 2002;2(1):55‐67. 38. Jourdain A, Martinou JC. Mitochondrial outer‐membrane permeabilization and remodelling in apoptosis. IntJBiochemCell Biol. 2009;41(10):1884‐9. 39. Landes T, Martinou JC. Mitochondrial outer membrane permeabilization during apoptosis: the role of mitochondrial fission. BiochimBiophysActa. 2011;1813(4):540‐5. 40. Martinou JC, Youle RJ. Which came first, the cytochrome c release or the mitochondrial fission? Cell DeathDiffer. 2006;13(8):1291‐5. 41. Zhang Z, Dales NA, Winther MD. Opportunities and challenges in developing stearoyl‐coenzyme A desaturase‐1 inhibitors as novel therapeutics for human disease. J Med Chem. 2014;57(12):5039‐56. 42. Scaglia N, Igal RA. Inhibition of Stearoyl‐CoA Desaturase 1 expression in human lung adenocarcinoma cells impairs tumorigenesis. IntJOncol. 2008;33(4):839‐50. 43. Ackerman D, Simon MC. Hypoxia, lipids, and cancer: surviving the harsh tumor microenvironment. Trends Cell Biol. 2014;24(8):472‐8. 44. Gonzalez D, Schmidt S, Derendorf H. Importance of relating efficacy measures to

unbound drug concentrations for anti‐infective agents. Clin Microbiol Rev. 2013;26(2):274‐88. 45. Lin JH, Lu AY. Role of pharmacokinetics and metabolism in drug discovery and development. Pharmacol Rev. 1997;49(4):403‐49. 46. Meacham CE, Morrison SJ. Tumour heterogeneity and cancer cell plasticity. Nature. 2013;501(7467):328‐37. 47. Dick JE. Stem cell concepts renew cancer research. Blood. 2008;112(13):4793‐807. 48. Reya T, Morrison SJ, Clarke MF, Weissman IL. Stem cells, cancer, and cancer stem cells. Nature. 2001;414(6859):105‐11. 49. Vermeulen L, Sprick MR, Kemper K, Stassi G, Medema JP. Cancer stem cells‐‐old concepts, new insights. Cell Death Differ. 2008;15(6):947‐58. 50. Medema JP. Cancer stem cells: the challenges ahead. Nat Cell Biol. 2013;15(4):338‐44. 51. Colak S, Medema JP. Cancer stem cells‐‐important players in tumor therapy resistance. FEBS J. 2014;281(21):4779‐91. 52. Todaro M, Alea MP, Di Stefano AB, Cammareri P, Vermeulen L, Iovino F, et al. Colon cancer stem cells dictate tumor growth and resist cell death by production of interleukin‐4. Cell Stem Cell. 2007;1(4):389‐402. 53. Tirinato L, Liberale C, Di Franco S, Candeloro P, Benfante A, La Rocca R, et al. Lipid droplets: a new player in colorectal cancer stem cells unveiled by spectroscopic imaging. Stem Cells. 2015;33(1):35‐44. 54. Colak S, Zimberlin CD, Fessler E, Hogdal L, Prasetyanti PR, Grandela CM, et al. Decreased mitochondrial priming determines chemoresistance of colon cancer stem cells. Cell Death Differ. 2014;21(7):1170‐7.

Chapter 6

160

55. Vaz FM, Pras‐Raves M, Bootsma AH, van Kampen AH. Principles and practice of lipidomics. J Inherit Metab Dis. 2015;38(1):41‐52. 56. Hyotylainen T, Oresic M. Systems biology strategies to study lipidomes in health and disease. Prog Lipid Res. 2014;55:43‐60.

Annexes

Summary

“Well that was the silliest tea party I ever went to! I am never going back there again!”


Summary

167

Mechanistic insight into BetA‐induced cell death

This thesis focuses on the mechanisms of BetA induced cancer cell death. In chapter 1, we

start with an introduction on cancer, cancer therapeutics, cell death mechanisms and BetA.

The mechanism by which BetA induces cell death has been investigated by many people

over the past years, however the exact mechanism remained unclear. In chapter 2 we

explore the possibility of other cell death mechanisms besides apoptosis. We observe that

BetA induces cell death in cells which are treated with a caspase inhibitor indicating that

another cell death mechanism must be causing the observed cell death. We therefore

investigated the role of necroptosis in BetA‐induced cell death. Cells treated with capasase

inhibitor and necrostatin‐1 (a necroptosis inhibitor) also showed induction of cell death. So

we proceeded to another possible cell death mechanism, autophagy. Upon BetA treatment

massive induction of autophagy, as shown by lipidated LC3II, was observed both by confocal

microscopy as well as western blot. Autophagosomes needs to be degraded by the

lysosomes in order to get full breakdown of the cargo. Autophagic flux was measured using

long lived proteins, a tandem RFP/eGFP‐tagged LC3 protein and a tandem mCherry/GFP‐

tagged p62 protein. Previously, we observed that cyclosporin A, which targets cyclophilin D

and effectively serves as an inhibitor of the mitochondrial permeability pore, prevents BetA‐

induced apoptotic features. We therefore also investigated the role of this compound in the

induction of autophagy and showed that autophagosome formation was significantly

reduced when cells were pre‐treated with cyclosporin A. These data suggest that BetA‐

induced autophagy is a consequence of mitochondrial damage triggered by BetA and can

be prevented by inhibition of PT‐pore opening. Autophagy, even though initially regarded

as a cell survival pathway, can play a role in dying cells as well and has been suggested to

serve as a balancing mechanism between cell survival and cell death. To test the role of

autophagy in BetA‐induced cell death we used autophagy‐deficient cells. Surprisingly, we

observed an increase in BetA‐induced cell death in these cells, which could not be prevented

by caspase inhibition. Both autophagy‐proficient and ‐deficient cells displayed a similar

increase in mitochondrial depolarization, suggesting that the initial mitochondrial insults

were independent of autophagy induction. Taken together these date indicate that

autophagy serves primarily as a survival mechanism in BetA treated cells and that BetA

induced cell‐death is independent of autophagy, necroptosis and caspases but requires the

mitochondria.

How the induction of apoptosis was realized was therefore further investigated in chapter

3. When analyzing the mitochondria via confocal or electron microscopy we observed

mitochondrial morphology changes upon BetA treatment. The mitochondrial tubular

network became fragmented and the mitochondria displayed a circular morphology by

electron microscopy. These circular cristae are also detected in patients with a X‐linked

syndrome called Barth syndrome, which suffer from a mutation in Taffazin gene. Taffazin

plays a role in cardiolipin remodeling and side chain saturation and cells of these patients

Summary

168

display enhanced cardiolipin saturation that causes mitochondrial ultrastructural as well as

functional abnormalities. We therefore investigated the saturation grade of cardiolipin in

cells treated with BetA and show that upon BetA treatment cardiolipin side chains get more

saturated as compared to vehicle treated cells. To investigate how this saturation occurs we

looked into the function of stearoyl CoA desaturase 1 (SCD‐1) a key enzyme that converts

saturated fatty acids into unsaturated fatty acids. We observed in several cancer cell lines

that the activity of SCD‐1 was blocked by BetA, resulting in an increase in saturated fatty

acids. The role of SCD‐1 inhibition in cardiolipin saturation was further corroborated by the

use of specific SCD‐1 inhibitors. In addition, exogenously added fatty acids (unsaturated or

saturated) strongly affected the saturation grade of cardiolipin. When we treated cells with

BetA or SCD‐1 inhibitor in combination with the saturated fatty acid, palmitate, we observed

an increase in saturation grade of cardiolipin. In contrast, when using the unsaturated fatty

acid, palmitoleic acid, cardiolipin saturation decreased. Importantly, the BetA or SCD‐1

inhibitor‐induced mitochondrial morphology, cytochrome c release, cell death and

clonogenic growth were all strongly affected by exogenously added fatty acids, confirming

that the inhibition of SCD‐1 activity is instrumental in inducing cell death.

Previously we showed that normal cells are less affected by BetA and this is consistent with

our current findings that revealed no morphology changes in mitochondria in normal

fibroblasts upon BetA treatment. As these cells have less SCD‐1 activity, but more

importantly do not appear to depend on this activity to acquire unsaturated fatty acids, the

inhibition of SCD‐1 by BetA or SCD‐1 inhibitor is not as detrimental as compared to tumor

cells. Taken together these data indicate that cancer cells carry an intrinsic mitochondrial

vulnerability that can be effectively targeted by inhibition of SCD‐1, using either specific

inhibitors or BetA, and is not observed in normal fibroblasts. This vulnerability depends on

enhancing the saturation of cardiolipin leading to altered mitochondrial structure and as a

consequence mitochondrial leakage and cell death.

Cancer stem cell sensitivity

Cancer stem cells are known to be very resistant to therapy. As BetA is acting in a broad

fashion to kill cells we investigated the effect of BetA on colon cancer stem cells. In chapter

4 we show that BetA induces a rapid cell death in three different colon cancer stem cell lines

in a time and concentration dependent fashion. Strikingly already after 2 hours an induction

of cell death is observed. As the spheroid cultures contain both cancer stem cells and more

differentiated cells, we set out to determine if BetA induces cell death in the cancer stem

cell population using FACS‐based sorting. Using this approach we revealed that BetA is

capable of inducing cell death in both differentiated tumor cells and cancer stem cells. This

was mirrored in an almost complete loss of clonogenic capacity of these cancer stem cell

cultures treated with different concentrations of BetA. Already after 2 hours of treatment

with a low concentration of BetA, clonogenic capacity is reduced in GTG7 cells, while the

Summary

169

other lines require slightly higher BetA concentrations. In the previous chapter we showed

that BetA induces cells death via inhibition of SCD‐1. We therefore also investigated the role

of SCD‐1 in inducing cell death in colon cancer stem cells. Although it takes more time to

induce cell death with SCD‐1 inhibitors in these cells as compared to BetA, the induction of

cell death and loss of clonogenic capacity is very efficient. We therefore conclude that SCD‐

1 is a crucial player in the maintenance of stemness of cancer cells and inhibition is

incompatible with long‐term survival. Targeting the activity of this enzyme in colon cancer

stem cells via BetA or chemical inhibitors could therefore be a therapeutic approach.

Lipidomics

In chapter 5, we describe the set‐up of a lipidomics pipeline to improve the characterization

and identification of phospholipids. To test this pipeline we used BetA treated samples.

Using the pipeline we found that we could identify several clusters of peaks belonging to

different phospholipids. On top of that we could identify also length and saturation grade

of these phospholipids. This resulted in the identification of total phospholipid levels of

vehicle and BetA treated samples. Differences in total phospholipid levels were mainly

observed in phosphatidylcholine, phosphatidylglycerol and bis‐(monoacylglycero)‐

phosphate. However in these phospholipids no differences in the saturation grade was

observed between samples. The only phospholipid that displayed a clear difference in

saturation grade was cardiolipin. In BetA treated samples it was clearly visible that the side

chains of cardiolipin are more saturated as compared to vehicle treated cells. This indicates

that BetA specifically alters the saturation grade of cardiolipin which eventually results in

cell death. With this pipeline we show that identification of phospholipids can be automated

and this method can be used in the clinic for diagnostic purposes.

Finally, in chapter 6, the data described in this thesis are put in perspective. We showed

that BetA‐induced cell death does not result of necroptosis or autophagy but likely via a yet‐

undefined pathway. We discuss what the role of autophagy in BetA‐induced cell death is

and what the possible cell death mechanism can be. Moreover we show in this thesis that

BetA‐induced cell death evolves due to an inhibition of SCD‐1 resulting in cardiolipin

saturation and cell death. We discuss what role BetA, SCD‐1 inhibition and cardiolipin have

on cancer and cancer stem cells and how this can play a role in cancer therapy. In this

chapter we give perspectives for further investigation to clarify the mechanism of BetA, as

well as future interventions and therapies are proposed for both BetA and SCD‐1 inhibitor.

For the investigation of lipids we set‐up a lipidomics pipeline and we give future

perspectives of a lipidomics pipeline and use of this pipeline in the clinic.

In conclusion, we believe our data provide compelling evidence that lipid metabolism,

specifically the saturation of cardiolipin, can serve as a reliable target in cancer therapy,

which is effectively exploited by BetA.

Nederlandse Samenvatting

“Well! I've often seen a cat without a grin,' thought Alice 'but a grin without a cat! It's the

most curious thing I ever saw in my life!”



173

Kanker is een veelvoorkomende ziekte met 100.000 nieuwe gevallen per jaar. De ziekte

wordt gekenmerkt door ongecontroleerde groei van weefsels vanwege een aanhoudende

celdeling. Gezonde cellen in het lichaam delen alleen wanneer dat nodig is (bijvoorbeeld

om wondjes te genezen). Tijdens celdeling krijgen de cellen een specifieke functie en

zodoende ook een specifieke vorm en grootte, dit proces heet differentiatie van de cel. Je

lichaam heeft stamcellen die ervoor zorgen dat vanuit 1 cel een nieuwe groep cellen ( die

ook gedifferentieerd zijn) kan groeien. Het delen en differentiëren van cellen gebeurd door

signalen die gegeven worden. Deze signalen kunnen van buiten de cel komen (zoals

chemische stofjes, virussen) maar ook van binnenuit (eiwitten). Zowel de uitwendige als de

inwendige signalen kunnen de cel laten delen of remmen de celdeling. Dit proces van groei

stimulatie en remming van groei staat onder strenge controle in de cel. Als cellen

ontsnappen aan deze controle en er een ontregeling plaatsvindt tussen de balans van groei

stimulatie en groei remming in het voordeel van de groei factoren dan kan er een

onbeheerste groei ontstaan. Deze onbeheerste groei kan goedaardig of kwaadaardig zijn.

Bij kwaadaardige groei spreken we van kanker, deze cellen zijn in staat om het orgaan

waarin ze zitten te vernietigen en kunnen zich ook verspreiden door het lichaam.

Goedaardige tumorcellen beschadigen alleen het orgaan waarin ze zitten en kunnen zich

niet verspreiden in het lichaam.

Tot op heden is kanker een ziekte die moeilijk te bestrijden is en zodoende wordt er veel

onderzoek gedaan naar hoe kankercellen werken en hoe deze anders zijn ten opzichte van

een gezonde cel. Momenteel zijn er vele soorten therapieën in de kliniek beschikbaar,

echter zijn deze behandelingen (bv chemotherapie en radiotherapie) niet specifiek voor de

kankercellen en werken ze ook op de gezonde cellen, waardoor de patiënt vaak last heeft

van bijwerkingen. Door te ontdekken hoe een kankercel werkt, kan men gericht medicijnen

zoeken die de kankercel dood maken, maar de gezonde cel geen schade berokkent.

Daarnaast wordt er ook onderzoek gedaan naar nieuwe stoffen die chemisch gemaakt zijn

of in de natuur zijn gevonden, deze nieuwe stoffen worden getest in een laboratorium op

verschillende soorten kankercellen om te onderzoeken wat voor effect deze stoffen hebben

op de vitaliteit van de cellen. Via zo’n soort test is Betulinezuur (BetA) ontdekt. Dit stofje

dat voorkomt in onder andere berkenbomen blijkt in staat te zijn om kankercellen dood te

maken terwijl het geen schadelijke effecten heeft op de gezonde cel. Dit proefschrift gaat

over BetA en hoe het kankercellen dood maakt, daarnaast is er ook een hoofdstuk wat een

bio‐informatisch algoritme beschrijft over het ontdekken en benoemen van lipiden (vetten)

In hoofdstuk 1 beschrijven we de literatuur van de afgelopen jaren over kanker, de

therapieën van kanker, de verschillende vormen van celdood die er zijn en het metabolisme

(stofwisseling) van de cel en hoe deze anders is in een kankercel. Daarnaast wordt BetA

geïntroduceerd, hoe het stofje ontdekt is, en wat er tot nu toe bekend is over onder andere

het werkingsmechanisme van BetA geïnduceerde celdood.


174

Het is bekend dat BetA een vorm van gereguleerde celdood induceert die apoptose wordt

genoemd. Echter als we deze vorm van celdood remmen met specifieke remmers (caspase

remmers) is er nog een groot deel van de kankercellen welke alsnog dood gaan. Dit is alleen

mogelijk als er een ander celdood mechanisme is die dit veroorzaakt. In hoofdstuk 2 is er

gekeken naar welke vormen van celdood er naast apoptose worden geïnduceerd door BetA.

We zijn begonnen met het kijken naar necroptose, een vorm van gereguleerde celdood

welke alleen kan optreden in een cel als er geen caspases zijn. Om te zien of necroptose

deel uitmaakt van BetA geïnduceerde celdood hebben we naast apoptose ook necroptose

geremd. De combinatie van deze remmers zouden als deze vormen van celdood door BetA

geïnduceerd worden, moeten leiden tot 100% levende cellen aangezien we beiden vormen

van celdood geremd hebben. Echter zagen we in deze proef dat de kankercellen nog steeds

dood gingen na behandeling met BetA, hieruit kunnen we concluderen dat een andere vorm

van celdood betrokken moet zijn.

De aandacht is vervolgens gevestigd op autofagie, een vorm van celdood waarbij de cel

zichzelf als het ware opeet om de opgegeten delen vrij te geven aan de omgeving om als

voeding te dienen. Vaak wordt autofagie ook gezien als een overlevingsmechanisme, omdat

de cel zichzelf opeet om andere cellen van bouwstenen te voorzien. Autofagie is een proces

waarbij onderdelen van de cel in een blaasje worden gestopt, ook wel autofagosoom (een

soort vuilniszakje) genoemd. Om het “afval” op te ruimen moet het blaasje samensmelten

met een lysosoom (de vuilverbranding), waarbij de inhoud van het blaasje wordt

afgebroken en gerecycled. Dit proces van complete afbraak wordt ook wel autofagische flux

genoemd.

In hoofdstuk 2 laten we zien dat BetA een massale vorm van autofagie induceert, door het

meten van een eiwitvorm (LC3II) wat voorkomt op de blaasjes (vuilniszakken). Met behulp

van de microscoop zagen we dikke groene puntjes van het eiwit ontstaan terwijl we met

eiwitmetingen (Western Blot) zien dat de LC3II vorm zichtbaar wordt na behandeling met

BetA. Soms gebeurt het wel eens dat het afval niet verwerkt wordt in de lysosomen en dat

je daardoor zo’n hoge hoeveelheid LC3II meet. Om te zien of autofagische flux (de

samensmelting van het blaasje met lysosoom en dus complete afbraak van het afval)

plaatsvindt, is dit gemeten door gebruik te maken van een LC3 eiwit wat is gekoppeld aan

een rode en groene kleur. Als rood en groen gemengd worden, ontstaat er een gele kleur,

tijdens het proces van autofagische flux wordt de groene kleur afgebroken in het lysosoom

terwijl de rode kleur blijft. Behandeling met BetA laat een rode kleur achter in de cellen wat

tot de conclusie leidt dat er volledige afbraak is. Dit is ook aangetoond door het meten van

afbraak van radioactieve lang levende eiwitten.

Mitochondriën zijn energiefabriekjes van de cel en ze zijn betrokken bij apoptose.

Mitochondriën hebben poortjes die open en dicht kunnen, waarbij er stofjes van binnen het

mitochondrion naar buiten kunnen (bv cytochroom c, wat celdood (apoptose) induceert).


175

Eerder onderzoek liet zien dat cyclosporine A (een stofje wat mitochondriale poortjes dicht

houdt) de kankercellen in leven houdt door remming van apoptose. We waren benieuwd

wat dit stofje doet op de autofagie inductie. We zagen dat er minder autofagie werd

geïnduceerd in combinatie van BetA met cyclosporine A dan met BetA alleen. Dit suggereert

dat geïnduceerde autofagie een gevolg is van de schade die BetA geeft aan de

mitochondriën.

Bij autofagie eet de cel zichzelf op om een andere cel van voeding te voorzien. Autofagie is

al bij lagere concentraties BetA te zien, mogelijk wijzend op dat autofagie vooral een

overlevingsmechanisme is na BetA behandeling. Het is echter ook denkbaar dat de massale

inductie van autofagie na BetA behandeling resulteert in een disbalans tussen overleving en

celdood en dat autofagie inductie een celdood mechanisme is. Om dit te onderzoeken is er

gebruik gemaakt van cellen die bepaalde eiwitten missen die belangrijk zijn voor autofagie

en zodoende kunnen deze cellen kunnen niet langer autofagie induceren. Als we deze cellen

behandelden met BetA zagen we meer celdood ontstaan en dus is autofagie een

overlevingsmechanisme in de cel. Het blijft echter onduidelijk welke vorm van celdood BetA

naast apoptose induceert.

In hoofdstuk 3 bestudeerden we de vorm van apoptose meer in diepgang. Apoptose is een

vorm van gereguleerde celdood die op 2 manieren plaats kan vinden. Via een receptor

pathway ( een sleutel en slot principe, je steekt een sleutel in een slot en het slot kan open,

waarbij de sleutel een proces in gang zet waarbij stofjes aangezet worden die leiden tot

celdood) of via een mitochondriale pathway (waarbij het energiefabriekje van de cel lek

raakt en er stofjes (oa. cytochroom c) vrij komen die celdood induceren). Omdat BetA de

mitochondriale route induceert hebben we gekeken wat er gebeurt met de mitochondriën

zelf. We zagen dat de structuur van de mitochondriën verandert na behandeling met BetA.

Het binnen membraan (cristae) van de mitochondriën wat normaal gesproken lange

strepen laat zien is na behandeling met BetA rond geworden. Dit is een zeer opmerkelijke

waarneming en deze ronde structuren van de cristae van mitochondriën zijn ook zichtbaar

in Barth syndroom patiënten. Deze patiënten hebben een defect in het Taffazin gen

waardoor zij een meer verzadigde vorm hebben van cardiolipine (CL).

CL is een lipide (vet) wat vier vetzuurketens heeft en voornamelijk voorkomt in de

mitochondriën waar het betrokken is bij de structuur van de cristae en tevens ook

cytochroom c vasthoudt in de mitochondriën. De zijketens van CL kunnen verschillen in

lengte en in verzadigingsgraad. Meer verzadigd vet (zoals roomboter) betekend dat deze

minder flexibel is als onverzadigd vet (zoals olie). Het is belangrijk voor de cel om een juiste

balans te hebben tussen verzadigd en onverzadigd vet, omdat veel verzadiging leid tot

starre vetten die niet meer flexibel zijn en aangezien vetten voorkomen in membranen kan

dit leiden tot structuur veranderingen en verandering in functionaliteit van het membraan.


176

Omdat de veranderingen in de mitochondriale cristae na BetA behandeling zo lijken op de

cristae in Barth syndroom patiënten hebben we de levels en verzadiging van CL gemeten.

We zagen dat CL meer verzadigd raakt na behandeling met BetA. We vroegen ons toen af

wat de oorzaak is van verzadiging. In het lichaam worden vetten opgenomen uit de voeding

of kunnen worden aangemaakt. In kankercellen maakt de cel liever vetten aan dan dat het

deze opneemt uit de voeding, waarschijnlijk omdat kankercellen zo snel delen en meer

vetten nodig hebben als bouwsteen. Nieuwe vetten maken gebeurt via een ingewikkeld

proces wat resulteert in een verzadigd vetzuur. Het lichaam maakt onverzadigde vetten via

een enzym stearoyl‐CoA desaturase 1 (SCD‐1) wat het verzadigd vetzuur omzet in een

onverzadigd vetzuur. Deze vetzuren kunnen dan in alle lipiden ingebouwd worden.

We bedachten dat BetA misschien invloed heeft op dit enzym, zodoende werd de activiteit

van dit enzym gemeten. We zagen dat na BetA behandeling of behandeling met een SCD‐1

remmer de activiteit van dit enzym was geremd en er dus meer verzadigd vetzuur aanwezig

was in de cellen. De vraag was nu of remming van SCD‐1 met een specifieke SCD‐1 remmer

ook dezelfde veranderingen in de mitochondriën opleverde, en inderdaad remmen van

SCD‐1 met de SCD‐1 remmer leidt ook tot mitochondriale veranderingen en CL verzadiging.

We hebben daarna gekeken of het toevoegen van extra onverzadigde vetzuren in

combinatie met BetA de morfologie van de mitochondriën hersteld en of ook CL levels weer

normaal werden. Het toevoegen van extra onverzadigde vetzuren bij BetA behandeling

resulteerde in vermindering van veranderingen in de mitochondriën en ook in CL

verzadigingsgraad. Het tegenovergestelde werd aangetoond met de combinatie van

verzadigd vetzuur en BetA, in deze combinatie werd de structuur van de mitochondriën nog

slechter dan met BetA alleen en ook waren de CL levels meer verzadigd dan na BetA

behandeling alleen. Daarnaast lieten we zien dat BetA of SCD‐1 beiden cytochroom c vrij

laten uit de mitochondriën en celdood induceren en dat de combinatie van SCD‐1 remming

(door BetA ofwel SCD‐1 remmer) en extra verzadigd vetzuur tot meer uitscheiding van

cytochroom c leiden en tot grotere inductie van celdood. Dit terwijl toevoeging van

onverzadigd vetzuur de schadelijke effecten van BetA of SCD‐1 tegengaan en er juist minder

cytochroom c vrijgelaten wordt en er minder celdood ontstaat. In dit hoofdstuk hebben we

een nieuwe celdood route ontdekt die specifiek kankercellen raakt. Gezonde cellen zijn

minder afhankelijk van SCD‐1 activiteit dan kankercellen en daardoor zijn kankercellen veel

gevoeliger voor BetA.

Kankerstamcellen zijn een onderdeel van de tumor die zelf vernieuwende eigenschappen

hebben en in verschillende andere cellen kunnen differentiëren. Er wordt van deze cellen

gedacht dat zij zorgen voor het terugkomen van de kanker na behandeling en voor de

verspreiding (metastasering) van de tumor door het lichaam. Kankerstamcellen kunnen

namelijk van 1 losse cel een hele nieuwe groep kankercellen maken en kankerstamcellen

zijn ook nog eens erg ongevoelig voor de verschillende vormen van therapie. Een grote

uitdaging in het kankeronderzoek is dan ook het begrijpen hoe deze cellen werken en het


177

vinden van therapieën die ook deze cellen dood maken. In hoofdstuk 4 hebben we de

werking van BetA op darmkankerstamcellen getest. Tot onze verbazing werkt BetA zeer

effectief op kankerstamcellen en zelfs sneller dan in kankercellen. Al na 2 uur zien we een

celdood inductie in de totale populatie van de cellen. Als we specifiek kijken naar het effect

op kankerstamcellen zien we dat BetA ook in 10% van de kankerstamcellen celdood

induceert en daarnaast werd ook een verlaging van zelfvernieuwing werd geobserveerd. De

huidige standaardtherapie voor darmkanker, oxaliplatin bleek in langdurige experimenten

minder effectief dan BetA in het verminderen van uitgroei van de darmkankerstamcellen.

Dit suggereert dat bij oxaliplatin resistentie tegen het middel op lijkt te treden terwijl dit bij

BetA niet gebeurd. Om te kijken of darmkankerstamcellen ook de beschreven nieuwe

celdood route uit hoofdstuk 2 volgen werden darmkankerstamcellen behandeld met SCD‐1

remmer. Ook met SCD‐1 remmer werd een celdood geïnduceerd, echter was deze niet zo

snel als bij BetA. Verder onderzoek naar het precieze mechanisme en de werking van BetA

in darmkankerstamcellen is nodig.

Omdat het remmen van SCD‐1 (via BetA of SCD‐1 remmer) leidt tot een ophoping van

verzadigde vetzuren die in elk lipide ingebouwd kunnen worden hebben we gekeken wat

de effecten van BetA op lipiden in het algemeen is (dus niet alleen CL). Het meten van

lipiden (vetten) is een langdurig en ingewikkeld proces. Zodoende hebben we in hoofdstuk

5 alle lipiden geanalyseerd met een nieuwe methode (lipide pipeline) die sneller en

nauwkeuriger lipiden kan herkennen. We beschrijven de opzet van deze pipeline en testen

we de werking van deze methode door gebruik te maken van BetA behandelde cellen. Met

deze nieuwe bio‐informatische methode kunnen we veel sneller en meer lipiden herkennen

en benoemen dan voorheen. We hebben verschillende fosfolipiden kunnen identificeren en

we hebben bekeken wat de verschillen waren tussen onbehandelde en behandelde cellen.

We zagen dat er significante verschillen waren in de totale waarden van sommige lipiden

(bv PG en BMP). In CL totaal levels is geen significant verschil gevonden terwijl we wel

verschil zien in verzadigingsgraad van de zijketens. In PG en BMP daarentegen zien we

verschillen in totale levels maar niet in de verzadiging. Waarom BetA alleen CL meer

verzadigd maakt moet verder worden onderzocht.

In hoofdstuk 6 worden de resultaten van de voorgaande hoofdstukken bediscussieerd in de

context van literatuur en wordt er besproken hoe BetA en SCD‐1 remming een rol kunnen

spelen in toekomstige therapie. Tevens bespreken we de rol van deze lipide pipeline in de

kliniek.

Samenvattend laten we zien dat vetzuurmetabolisme, in het specifiek CL verzadiging, een

zwakke schakel is in de kankercel en dat dit een mogelijk nieuw aangrijpingspunt is voor

therapie.

List of Publications

“Why, sometimes I’ve believed as many as six impossible things before breakfast.”


List of publications

199

Potze L*, Pras‐ Raves ML*, van Lenthe H, Vervaart MAT, Kessler J.H, Luyf AC, Medema JP, et al. Improved identification of lipids using physico‐chemical properties: application to lipidomics analysis of Betulinic acid treatment Manuscript submitted * Equal contribution

Potze L*, di Franco S*, Kessler JH, Stassi G Medema JP. Betulinic acid induces a rapid form of cell death in colon cancer stem cells. Manuscript submitted. *Equal contribution

Potze L, Di Franco S, Grandela C, Pras‐Raves ML, Picavet DI, van Veen HA, et al. Betulinic acid induces a novel cell death pathway that depends on cardiolipin modification. Oncogene. 2015.

Tirinato L, Liberale C, Di Franco S, Candeloro P, Benfante A, La Rocca R, Potze L, et al. Lipid droplets: a new player in colorectal cancer stem cells unveiled by spectroscopic imaging. Stem cells. 2015;33(1):35‐44.

Potze L, Mullauer FB, Colak S, Kessler JH, Medema JP. Betulinic acid‐induced mitochondria‐dependent cell death is counterbalanced by an autophagic salvage response. Cell DeathDis. 2014;5:e1169.

UvA-DARE (Digital Academic Repository) Mechanisms of ... · Mechanisms of Betulinic acid-induced...

Documents

Transcript of UvA-DARE (Digital Academic Repository) Mechanisms of ... · Mechanisms of Betulinic acid-induced...