Pandyra Aleksandra A 201311 PhD thesis · Thesis – Aleksandra Pandyra iv Acknowledgements First, I would like to thank my supervisor, Dr. Linda Penn, for the opportunity to work

Preclinical potentiators of statin-induced anticancer effects

By

Aleksandra Anna Pandyra

A thesis submitted in conformity with the requirements

for the Degree of Doctor of Philosophy

Department of Medical Biophysics

University of Toronto

© copyright by Aleksandra A. Pandyra 2014

Thesis – Aleksandra Pandyra

ii

Preclinical potentiators of statin induced anticancer effects

Aleksandra Anna Pandyra

Doctor of Philosophy

Department of Medical Biophysics

University of Toronto

2014

Abstract

Statins have been used for decades in the treatment of hypercholesterolemic patients to

decrease the incidence of adverse cardiovascular events. Statins target the rate-limiting

enzyme of the mevalonate (MVA) pathway, 3-hydroxy-3-methylglutaryl coenzyme A

reductase (HMGCR). Statin-mediated HMGCR inhibition in hepatocytes leads to depletion of

MVA derived sterols, including cholesterol, and a subsequent uptake of low-density

lipoprotein (LDL) cholesterol from the plasma. Aside from the intended cholesterol-lowering

properties, statins exert many pleiotropic effects including anti-tumour activity. The well

characterized pharmacokinetic profiles, safety and the fact that many statins are off patent has

ensured their prompt entry into clinic for evaluation of anti-cancer efficacy. We hypothesize

that the therapeutic window of statins can be increased through (i) the use of a

pharmacological screen to identify compound/s that synergize with statins to induce tumour-

specific apoptosis and (ii) the use of a genome-wide small hairpin ribonucleic acid (shRNA)

screen to identify novel pathways/targets whose knockdown enhances the vulnerability of

tumour cells to statin-inhibition. Firstly we carried out a screen in a multiple myeloma (MM)


iii

cell line where we combined atorvastatin with other off-patent FDA approved drugs. We

identified dipyridamole, a commonly used anti-platelet agent as synergizing with and

potentiating statin-induced apoptosis in a variety of MM and acute meylogenous leukemia

(AML) cell lines and primary patient samples. The efficacy of the combination was

demonstrated in vivo. Secondly, through genome-wide shRNA screen, we identified several

putative targets whose knockdown potentiated statin induced anti-cancer effects in the A549

lung cancer cell line. Amongst the most promising hits, sterol regulatory element binding

transcription factor 2 (SREBF2) was validated in lung and breast cancer cell lines. The

knockdown of SREBF2 was found to dramatically potentiate statin-induced apoptosis and

block the upregulation sterol-responsive gene targets. Taken together, this research has

uncovered a novel combination of clinically translatable drugs with strong preclinical efficacy

in hematological malignancies. Furthermore we have identified several novel validated targets

that sensitize the MVA pathway to statin-induced apoptosis.


iv

Acknowledgements

First, I would like to thank my supervisor, Dr. Linda Penn, for the opportunity to work

in her lab, in one of the best research centers in the world. I have learned so much from her

and I will never forget my experience in this incredibly supportive lab. Along with Dr. Linda

Penn’s unwavering support, I would like to thank the members of my supervisory committee,

Drs Aaron Schimmer and Robert Kerbel for their scientific guidance. I felt like I could always

go to them for advice and this has been instrumental in helping me through graduate school.

Furthermore, I would like to thank my family for their never-ending support and love,

most important Mom, Tat, and Magda and certainly every other member of the Polish gang,

Kate, Aga, all the Piotr’s, Ciocia, Wujek, Grandma, Grandpa and the little ones, and the new

ones (Jon and Allison). I would like to thank Mark, my best friend, who has always pushed

me to be the best I can be in every aspect of my life and go through life screaming and

fighting. And of course my other friends, I can’t even imagine going through life without

knowing you: Tracy, Amanda, Sara, Pete, Cory, Manpreet, Carolyn, Paul B., Janice, Laura,

Cynthia, Ric, Lisa, Trudy, Larry, Jamie, Kevin, Stepho, Tracey, Valery. A special thanks to

Maria K., I wish I could have been a better friend.

Moreover, I would like to thank all the collaborators, Drs. Jason Moffat, Corey Nislow,

Mark Minden, Paul Boutros and Suzanne Trudel, who have all been instrumental in

contributing to the work presented in this thesis.


v

Also, I would like to thank Dr. Karl Lang for giving me the great opportunity to work

in his oustanding, excellent and awesome lab, his everlasting support and guidance on present

and future research endeavours.

Finally I would like to thank Philipp Lang, for everything. There’s too much to say

than a few sentences could cover.


vi

Table of contents

Abstract ........................................................................................................................................................ ii

Acknowledgements ................................................................................................................................. iv

Table of contents ...................................................................................................................................... vi

List of figures ............................................................................................................................................. ix

Table of tables ........................................................................................................................................... xi

Abbreviations ........................................................................................................................................... xii

Chapter 1 : Introduction ................................................................................................................ 1

1.1 Treatment of Cancer ......................................................................................................................... 2

1.1.1 Cytotoxic Chemotherapy ............................................................................................................................ 3

1.1.2 Targeted Molecular Therapies ................................................................................................................. 3

1.1.2 Drug Repositioning ....................................................................................................................................... 6

1.2 Repositioning Statins as anti-‐cancer agents ............................................................................. 7

1.2.1 Statins in the cardiovascular setting ..................................................................................................... 7

1.2.2 The pharmacology of statins .................................................................................................................... 8

1.2.3 The mevalonate pathway and cancer ................................................................................................ 11

1.2.4 Statins as preventative anti-‐cancer agents ...................................................................................... 15

1.2.5 Statins in the clinic ..................................................................................................................................... 17

1.3 Combination therapy ...................................................................................................................... 21

1.4 Research Objectives and thesis outline .................................................................................... 23

1.5 References .......................................................................................................................................... 25

Chapter 2 : Combining statins and dipyridamole effectively targets acute

myelogenous leukemia and multiple myeloma ................................................................. 35

2.1. Abstract .............................................................................................................................................. 36

2.2. Introduction ...................................................................................................................................... 37


vii

2.3. Results ................................................................................................................................................ 40

2.3.1 A screen of pharmacologically active drugs identifies dipyridamole as a potentiator of

the anticancer effects of atorvastatin ............................................................................................................ 40

2.3.2. The combination of statins and dipyridamole is synergistically antiproliferative and

induces apoptosis in AML and MM cell lines ............................................................................................. 44

2.3. 3. The combination of statins and dipyridamole induces apoptosis in primary AML and

MM cells ..................................................................................................................................................................... 53

2.3.4. The combination of statins and dipyridamole delays tumour growth in leukemia

xenografts ................................................................................................................................................................. 60

2.3.5. Mediators of cAMP signaling in combination with statins induce apoptosis and raise

intracellular cAMP levels in leukemia cells ................................................................................................ 65

2.4. Materials and Methods .................................................................................................................. 74

2.5. Discussion .......................................................................................................................................... 79

2.6. References ......................................................................................................................................... 84

Chapter 3 : Genome wide shRNA screen reveals novel sensitizers of statin-‐induced

apoptosis ......................................................................................................................................... 89

3.1 Abstract ............................................................................................................................................... 90

3.2 Introduction ....................................................................................................................................... 91

3.3 Results ................................................................................................................................................. 94

3.2.1 A genome-‐wide screen identifies shRNA drop-‐outs after fluvastatin exposure ........ 94

3.2.2 Validation of shRNA hits in the A549 cells uncovers potentiators of statin-‐induced

apoptosis and anti-‐proliferative effects. .................................................................................................... 101

3.2.3 Concomitant targeting of PI4KB and HMGCR showed anti-‐proliferative effects in lung

and breast cell lines. ........................................................................................................................................... 115

3.2.4 Down-‐regulating the SREBF2 mediated sterol-‐feedback loop in combination with

statins is an effective anti-‐cancer strategy to induce lung and breast tumour cell kill. ........ 122


viii

3.2.5 Fluvastatin delays tumour growth in MDAMB231 xenografts. .......................................... 129

3.4 Methods ............................................................................................................................................ 131

3.5 Discussion ........................................................................................................................................ 135

3.6. References: ..................................................................................................................................... 143

Chapter 4 : Discussion ............................................................................................................... 149

4.1 Statins as anti-‐cancer agents – How translational is the cell culture work? ............. 150

4.2 Potentiating statin-‐induced apoptosis – combinatorial approaches .......................... 154

4.2.1 Statins and rationally crafted combinations ................................................................................. 154

4.2.2 Statins and dipyridamole ...................................................................................................................... 156

4.2.3 Limitations and future directions of this work ........................................................................... 158

4.3 Targeting the MVA pathway – unbiased knockdown approaches ................................ 161

4.3.1 The MVA pathway’s other targets ..................................................................................................... 161

4.3.2 Limitations and future directions of this work ........................................................................... 162

4.4 Concluding remarks and anticipated impact ...................................................................... 164

4.5 References: ...................................................................................................................................... 165

Publications (2008-‐2013): ................................................................................................................ 170


ix

List of figures

Figure 1-1: The MVA pathway. ............................................................................................... 12

Figure 2-1: A drug screen reveals dipyridamole as a potentiator of the anti-proliferative

effects of atorvastatin. ...................................................................................................... 43

Figure 2-2: The statin-dipyridamole combination is synergistic in AML and MM cell lines. 46

Figure 2-3: Dipyridamole potentiates the anti-proliferative effects of fluvastatin and the

combination is synergistic. ............................................................................................... 49

Figure 2-4: The statin-dipyridamole combination induces apoptosis in AML and MM cell

lines. ................................................................................................................................. 52

Figure 2-5: The statin-dipyridamole combination induces apoptosis in the OCI-AML3 cell

line. ................................................................................................................................... 54

Figure 2-6: The statin-dipyridamole combination induces apoptosis in primary AML and MM

patient cells. ...................................................................................................................... 59

Figure 2-7: Dipyridamole plasma concentrations are higher following i.p. administration than

p.o. .................................................................................................................................... 61

Figure 2-8: The statin-dipyridamole combination delays tumour growth in leukemia

xenografts. ........................................................................................................................ 64

Figure 2-9: The relative sensitivity to doxorubicin in 8226DOX and 8226 parental cells is not

significantly altered in response to dipyridamole exposure ............................................. 67

Figure 2-10: Cilostazol, a PDE3 inhibitor, potentiates the anti-proliferative activity of statins.

.......................................................................................................................................... 70

Figure 2-11: Modulators of cAMP induce apoptosis and increase intracellular cAMP levels in

AML cells. ........................................................................................................................ 73


x

Figure 3-1: A genome wide dropout screen uncovers putative shRNAs that potentiate

fluvastatin-induced cell death. .......................................................................................... 97

Figure 3-2: HMGCR inhibition by statins triggers a restorative feedback loop in hepatocytes

........................................................................................................................................ 103

Figure 3-3: Line plots of the two best hairpins for each hit chosen for validation show the

dropout over time. .......................................................................................................... 105

Figure 3-4: Generation of A549 sublines stably expressing shRNA constructs demonstrates

knockdown of protein or mRNA target. ......................................................................... 109

Figure 3-5: shRNAs targeting gene hits identified in the screen potentiate fluvastatin-induced

apoptosis and anti-proliferative effects. ......................................................................... 114

Figure 3-6: Transient knockdown of PI4KB sensitizes lung and breast cancer cells to the anti-

proliferative effects of fluvastatin. ................................................................................. 118

Figure 3-7: The combination of fluvastatin and a pharmacological inhibitor of PI4KB is anti-

proliferative and induces apoptosis in lung and breast cancer cells. .............................. 121

Figure 3-8: Stable knockdown of SREBF2 sensitizes breast cancer MCF7 and MDAMB231

cells to the pro-apoptotic and anti-proliferative effects of fluvastatin. .......................... 125

Figure 3-9: Knockdown of SREBF2 abrogates the upregulation of HMGCR and HMGCS1

upon fluvastatin exposure. .............................................................................................. 128

Figure 3-10: Fluvastatin delays tumour growth in MDAMB231 xenografts. ........................ 130

Figure 3-11: Targeting the MVA pathway in tumour cells. ................................................... 140


xi

Table of tables

Table 1-1: The pharmacokinetic properties of the different statins ......................................... 10

Table 2-1: The statin dipyridamole combination induces apoptosis in primary AML cells .... 56

Table 3-1: Putative gene hits targeted by shRNAs that potentiate the anti-cancer effects of

fluvastatin. ...................................................................................................................... 100

Table 3-2: TRC Clone ID's of shRNA hits chosen for validation. ......................................... 107


xii

Abbreviations

Abbreviation Full name

3D 3 dimensional

4-MD 4-{[3′,4′-(Methylenedioxy)benzyl]amino}-6-methoxyquinazoline

4S Scandinavian Simvastatin Survival Study

5-FU 5-fluorouracil

AACR American Association of Cancer Research

ALK anaplastic lymphoma kinase

AML acute myelogenous leukemia

ANOVA Analysis of variance

ATP adenosine triphosphate

AV annexin V

b.i.d. bis in die

BRAF B Rapidly Accelerated Fibrosarcoma

C concentration

cAMP Cyclic adenosine monophosphate

CD cluster of differentiation

cGMP cyclic guanosine monophosphate

CHD coronary heart disease

CI combination index

CML chronic myelogenous leukemia

CoA Coenzyme A

CTLA-4 cytotoxic T lymphocytes antigen 4


xiii

CYP3A4 cytochrome p450 A4

db-cAMP dibutyryl cAMP

DCIS ductal carcinoma in-situ

DMEM H21 Dulbecco's modified eagle's medium H21

DMSO dimethyl sulfoxide

DNA Deoxyribonucleic acid

dUTP deoxyuridine-triphosphat

EC effective concentration

eEF1A2 eukaryotic elongation factor 1 alpha 2

EGFR epidermal growth factor receptor

EML4 Echinoderm microtubule-associated protein-like 4

ENT1 equilibrative nucleoside transporter

EORTC European Organisation for Research and Treatment of Cancer

ER estrogen receptors

Erk Extracellular signal-regulated kinases

esiRNA endoribonuclease-prepared siRNA

FBS fetal bovine serum

FDA Food and Drug Administration

FDPS farnesyl pyrophosphate synthase

FITC fluorescein isothiocyanate

FTIs farnesyltransferase inhibitors

g gram

GARP gene activity ranking profile


xiv

GCSF Granulocyte colony-stimulating factor

GGPP gernaylgernayl pyrophosphate

GGPS1 geranylgeranyl diphosphate synthase 1

GGTase-I gernaylgeranyltransferases I

GLUT glucose transporters

HCC hepatocellular carcinoma

HDL high-density lipoprotein

Her2 Human Epidermal Growth Factor Receptor 2

HMGCR hydroxy-3-methylglutaryl coenzyme A reductase

HMGCS1 hydroxymethylglutaryl coenzyme A synthase 1

HMGCS1 3-hydroxy-3-methylglutaryl-Coenzyme A synthase 1

i.p. intraperitoneal

IC inhibitory concentration

IMDM Iscove modified Dulbecco medium

JNK c-Jun N-terminal kinases

kg kilogramm

LDL low-density lipoprotein

LDLr low-density lipoprotein receptor

m meter

MAPK mitogen-activated protein kinase

MAP2K4 mitogen-activated protein kinase kinase 4

MEM modified Eagle's medium

mg milligramm


xv

MM multiple myeloma

mRNA messenger ribonucleic acid

mTOR mammalian target of rapamycin

MTS

3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium

MTT 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide

MVA mevalonate

NCI National Cancer Institute nM nanomolar NSCLC non–small cell lung cancer

OATB organic anion transporting polypeptide

P-gp P-glycoprotein

p.o. per os

p53 protein 53

PARP Poly (adenosinediphosphate-ribose) polymerase

PBS phosphate buffered saline

PBSC peripheral blood stem cells

PCR polymerase chain reaction

PD-1 Programmed cell death protein 1

PDEs phosphodiesterases

PEGylated polyethylene glycol modified

PI propidium iodide

PIK4B phosphatidylinositol 4-kinase, catalytic beta

PR progesterone receptor


xvi

RNAi Ribonucleic acid interference

RPMI Roswell Park Memorial Institute

SCAP SREBP cleavage activating protein

SCID Severe combined immunodeficiency

SD standard deviation

SDS sodium dodecyl sulfate polyacrylamide

SHARP shRNA Activity Ranking Profile

shRNA small hairpin ribonucleic acid

siRNA small interfering RNA

SRE sterol responsive elements

SREBF2 sterol regulatory element binding transcription factor 2

TACE transarterial chemoembolization

TAE transcatheter arterial embolization

TD thalidomide and dexamethasone

TRC1 RNAi Consortium

TUNEL Terminal deoxynucleotidyl transferase dUTP nick end labeling

VAD vincristine, doxorubicin and dexamethasone

VEGF vascular endothelial growth factor

zGARP Z normalized GARP

µL microliter

µM micromolar


1

Chapter 1 : Introduction


2

1.1 Treatment of Cancer

Cancer poses a significant global health problem and accounts for 13% of deaths

worldwide1. Advances in molecular biology have revealed cancer to be an extremely

heterogenous and innately complex disease. High-throughput genomic and transcriptomic

analysis has demonstrated that there exist multiple unique molecular subsets not only within a

tumour type such as breast2-4 and lung5, but complexity can exist within the context of a

patient‘s individual tumour6-9. As summarized by Hanahan and Weinberg in their seminal

review, tumour cells are characterized by replicative immortality, resistance to apoptosis and

anti-growth signals, induction of angiogenesis, as well as the capacity to invade and

metastasize10. Amongst these tumour-defining characteristics, there are the novel emergent

hallmarks namely altered metabolism and immune evasion.

Several mechanisms of triggering cell death have been described. There is the classical

non-inflammatory programmed apoptosis molecularly mediated by caspases and

morphologically characterized by nucleus condensation, membrane blebbing and eventual

phagocytosis of apoptotic cells11. Cells can also die through necrosis, a process mediated by

the activation of RIP1 and RIP3 that causes a local inflammatory response11. Autophagic cell

death entails a caspase-independent vacuolization of cellular organelles11. Although not strictly

leading to cell death, aberrations during mitosis termed mitotic catastrophe, can also lead to

oncosupressive cell senescence11. Resistance to cell death is a pivotal hallmark of cancer that

can contribute to tumorigenesis and many anti-cancer agents target this hallmark by triggering

tumour cell apoptosis. Increasingly detailed elucidation of the cell death machinery as well as

different forms of cell death has afforded novel and more specific ways to target cancer cells12.

The arsenal of current drug therapies available to treat cancer broadly falls into two

general categories: cytotoxic chemotherapy and targeted molecular therapies.


3

1.1.1 Cytotoxic Chemotherapy

Ever since their inception in the early 1970s for the curative treatment of advanced

Hodgkin’s disease13, cytotoxic chemotherapeutic drugs have been used for decades in the

treatment of solid tumours and hematological malignancies at various stages of therapeutic

intervention. These drugs include: the alkylating agents14 such as temozolomide used in

treating gliomas15 ; carboplatin of the platinum compounds16 which is the mainstay of ovarian

cancer chemotherapy17; the antimetabolites18 of which 5-fluorouracil is used in first-line

therapy of colorectal cancer; the topoisomerase inhibitors19 an example of which is idarubicin

used in consolidation and maintenance therapy of acute myelogenous leukemia (AML)

patients20; and, the tubulin-binding drugs21 such as docetaxel recommended for adjuvant

therapy in luminal B and triple negative breast cancer22. These drugs reduce tumour burden by

targeting rapidly proliferating cells, and as a consequence, normal tissues are also collaterally

affected leading to general systemic side effects such as central and peripheral neurotoxicity,

cardiotoxicity, gastrointestinal symptoms and immune suppression23-26. Despite problems of

wide-ranging toxicities, intrinsic and acquired drug resistance27, and controversies over

clinical benefits in many cancers28, these cytotoxic drugs are still the mainstay of most

treatment regimens.

1.1.2 Targeted Molecular Therapies

An increased understanding of the aberrant molecular pathways that drive

tumorgenesis has subsequently fuelled the rational design of agents to inhibit them. A target

required for aberrant tumour growth, survival and metastasis is usually mutated or over-

expressed in the tumour but not normal cells thereby providing the rationale for a tumour-


4

normal index. The two primary types of targeted therapy are monoclonal antibodies or small

molecule inhibitors.

The development of monoclonal antibodies to relieve immune suppression and

enhance anti-tumour immunity in immunogenic tumours has begun to fill an important gap in

the treatment of metastatic melanoma, a cancer with few treatment options and among the

highest mortality rates. Ipilimumab, a monoclonal antibody targeting cytotoxic T lymphocytes

antigen 4 (CTLA-4) was approved by the European Union in 2010 and the Food and Drug

Administration (FDA) in March 2011 for the treatment of metastatic melanoma patients.

CTLA-4 is an inhibitory molecule expressed on T cells during activation. There are currently

eight monoclonal antibodies targeting other immune targets such as Programmed cell death

protein 1 (PD-1) in clinical trials29. Bevacizumab, another monoclonal antibody, blocks

vascular endothelial growth factor (VEGF), and stops tumour growth by preventing the

formation of new blood vessels. It has been approved for the treatment of metastatic

colorectal30, breast31, kidney32 and ovarian cancer33.

Over the last decade, many small molecule inhibitors have been approved for the

treatment of various tumour types. Recent approvals include vemurafenib in the treatment of

melanoma with the BRAF V600E mutation34, and crizotinib an agent that targets the EML4–

ALK (Echinoderm microtubule-associated protein-like 4 - anaplastic lymphoma kinase)

fusion protein in non-small-cell lung cancer35. The successful drug imatinib, which targets the

BCR-ABL fusion protein in Philadelphia chromosome-positive chronic myelogenous

leukemia (CML), has been used in the clinic for years 36 and has improved the outcomes of

many CML patients37. Imatinib also targets c-KIT and has been approved for treated patients

with gastrointestinal stromal tumors38. The list of other targeted agents, currently approved

and in development, is extensive. However, what was once considered the best new method

of specifically targeting tumours is fraught with unexpected challenges.


5

Due to the specificity of the targeted agents they were thought to be less toxic than the

traditional cytotoxic agents. However, it is now apparent that the safety of the targeted agents

has been over-estimated. Targeted agents often cause organ-specific toxicities. The EGFR

inhibitors can cause dermatologic toxicity39, the angiogenesis inhibitors and tyrosine kinase

inhibitors cardiovascular toxicity40, the mTOR (mammalian target of rapamycin) inhibitors

metabolic toxicities23. These toxicities were not evident during clinical trials as the positive

selection of responsive patients masked the effects apparent in a broader patient population

characterized by other comorbidities not evaluated in clinical trials. A recent meta-analysis of

38 randomized clinical trials testing novel agents approved by the FDA for the treatment of

sold tumours between 2000 and 2010 reported that despite modest improvements in outcomes,

these new drugs, were generally more toxic than the old drugs41. Despite initial claims of

safety, targeted agents also have side effects that have a significant impact on a patient’s

quality of life.

At the recent 24th EORTC-NCI-AACR (European Organisation for Research and

Treatment of Cancer-National Cancer Institute-American Association of Cancer Research)

Symposium on ‘Molecular Targets and Cancer Therapeutics’ in Ireland, the progress and

effective incorporation of targeted agents into the clinic were discussed42. Redundancies of

signaling pathways, innate or acquired resistance, presence of feedback loops that dampen

efficacy are all obstacles facing the clinical utility of the targeted agents. The use of targeted

agents in anti-cancer therapeutic regimens incurs high costs. For example, cetuximab, a drug

targeting the epidermal growth factor receptor (EGFR), approved for the treatment of non–

small cell lung cancer (NSCLC) costs $80,000 for an 18-week treatment in the United States

and has an extended survival benefit of 1.2 months43. On average, bevacizumab costs $90,000

to treat a patient with an overall survival benefit of 1.5 months43. Future implementation of

cost-effectiveness analyses as well as companion studies to identify biomarkers of response


6

and consequently sub-populations of cancer patients most likely to benefit from these

treatments is necessary for the realistic implementation of these therapies into the standard of

care.

While the traditional cytotoxic agents and targeted therapies have made great impact

in cancer patient care, their toxicities, lack of efficacies in treating certain tumours and

prohibitive costs highlights an urgent need for the development of novel therapeutic strategies.

1.1.2 Drug Repositioning

Exploiting drugs already approved for non-cancerous diseases is an attractive

alternative for development of novel anti-cancer therapies44, 45. Since the pharmacokinetic,

pharmacodynamic and toxicity profiles of these drugs are already known, their translation into

clinical trials testing for their anti-cancer efficacy has the potential to be rapid. As many of

these compounds have been on the market for many years in the treatment of other diseases,

they are often off-patent and readily accessible form the economic perspective.

Identification of drug candidates for repositioning can occur through chemical

screening where vulnerabilities of cancer cell lines exposed to FDA-approved compounds is

tested. Using this approach, several such candidates have been recently identified such as the

antimicrobial tigecycline for the treatment of AML46. Similarly, the antiparastistic agent

ivermectin was found to have anti-leukemic activity47. Tumour cell kill can occur through a

mechanisms related to the compound’s known mode of action (“on-target repositioning) or a

combination of on-target effects and several other pleotropic effects. The oral antiprarastic

agent, clioquinol, whose mechanism of action related to its antiparastistic efficacy is unknown,

was found anti-leukemia activity through inhibition of the proteasome48.


7

Some drugs have been used for decades in the treatment of conditions affecting large

portions of the population. This can lead to epidemiological evidence suggesting that these

drugs could affect the incidence of other conditions such as cancer. Two such examples

include metformin and statins. The former has been used for the treatment of non-insulin

required diabetes for over 50 years. Numerous case-control and cohort studies have linked

metformin usage with reduced cancer incidence49, 50. Metformin, whose anticancer effects are

attributable to both direct (insulin-independent) and indirect (insulin-dependent) actions of the

drug, is currently being evaluated in prospective clinical trials where efficacy is being tested

in breast cancer patients51, 52.

Statins are a family of drugs that have been used to lower cholesterol for decades in

patients with cardiovascular and coronary heart disease, also have anti-cancer effects. Their

application as anti-cancer therapeutics is the focal point of this body of work.

1.2 Repositioning Statins as anti-cancer agents

1.2.1 Statins in the cardiovascular setting

The observation that patients dying from occlusive vascular disease had thick and

irregular artery walls eventually led to the development of the lipid hypothesis linking

coronary heart disease (CHD) death with high levels of LDL-cholesterol. In 1973, the

pioneering work of Goldstein and Brown demonstrated that 3-hydroxy-3-methylglutaryl

coenzyme A reductase (HMGCR) activity, already known to be the rate-limiting enzyme of

cholesterol biosynthesis, was correlated to the extracellular levels of lipoproteins in cultured

fibroblasts53 It was correctly rationalized that by inhibiting HMGCR, plasma LDL-cholesterol

could be lowered. Efforts to find such a compound culminated in the discovery of mevastatin,

a microbial metabolite capable of potently inhibiting HMGCR54, 55. Mevastatin’s in vivo


8

efficacy was subsequently demonstrated in humans56. Despite demonstrated efficacy in

lowering serum cholesterol levels, mevastatin was not marketed and the first commercially

available statin was lovastatin, a natural product isolated from Aspergillus terreus.

In the early 1990’s other synthetic lovastatin analogues soon followed with

simvastatin and pravastatin becoming available. However, it wasn’t until the seminal

Scandinavian Simvastatin Survival Study (4S) that established the role of statins in the

prevention of adverse cardiovascular events57 including mortality. The study randomized

4444 patients with angina pectoris or previous myocardial infarctions into two-arms, with one

arm receiving simvastatin and the other a placebo. After a five year follow up, it was found

that patients in the simvastatin arm suffered coronary deaths, less major coronary events and a

decreased risk of undergoing myocardial revascularisation procedures. These reductions were

correlated to decreased in total plasma cholesterol, LDL-cholesterol and increases in high-

density lipoprotein (HDL) cholesterol. Similar large studies58, 59 demonstrating the efficacy in

the treatment of hypercholesterolemia followed, cementing their role in the prevention of

cardiovascular disease and treatment of hyperlipidemia.

1.2.2 The pharmacology of statins

There are currently seven statins approved in North America for cardiovascular

indications. Structurally, synthetic and naturally occurring statins are composed of a core

pharmacophore, common to all statins and an attached moiety, a ring (reduced naphthalene,

indole, pyrrole or quinolone) with different substituents is the distinguishing feature between

statins. The pharmacophore binds HMGCR in a competitive, reversible way and the

substituents on the ring define a statin’s solubility and other pharmacological properties listed

in Table 1-160, 61. Typical cholesterol-lowering dose range from 20-80 mg and are orally


9

administered daily in tablet form. Statins undergo extensive first-pass uptake and metabolism

by the liver. The lipophilic statins passively diffuse into hepatocytes and peripheral tissues

whereas hydrophilic statins are transported into the liver by organic anion transporting

polypeptide (OATB) family of drug transporters and as such their systemic penetration into

extra-hepatic sites is limited. Potential adverse drug interactions between statins and other

drugs occur with the statins metabolized by the cytochrome p450 A4 (CYP3A4) enzyme,

mainly simvastatin, lovastatin and atorvastatin. Statins are generally well tolerated with mild

side effects. Hepatotoxicity62 and myotoxicity which may progress to rhabdomyolysis63 are

amongst the most serious side-effects but do not occur in the majority of patients. High statin

doses (Table 1-1), up to 25 times the normal cholesterol-lowering doses, have been

administered to humans and were tolerated. Taken together, statins are safe, well tolerated and

their pharmacology and toxicology has been well investigated.


10

Tabl

e 1-

1: T

he p

harm

acok

inet

ic p

rope

rtie

s of

the

diffe

rent

sta

tins*

Bio

-av

aila

bilit

yPl

asm

a Pr

otei

n B

indi

ngPl

asm

a H

alf-

life

(%)

(%)

(h)

Lova

stat

inye

sye

s5

>95

310

-20

(25-

49)

yes

25 m

g/kg

(ref

97)

Fluv

asta

tinye

sno

19-2

9>9

80.

5-2.

344

8 (1

090)

no (C

YP2C

9)na

Ato

rvas

tatin

yes

no12

80-9

814

-30

27-6

6 (4

8-18

8)ye

sna

Sim

vast

atin

yes

yes

594

-98

1.9-

310

-34

(24-

81)

yes

15m

g/kg

(ref

99)

Pita

vast

atin

yes

no80

9611

nano

(CYP

2C9)

na

Prav

asta

tinno

no18

43-5

50.

8-3

45-5

5 (1

06-1

29)

no16

80 m

g/da

y (r

ef 1

04)

Ros

uvas

tatin

nono

2088

-90

2037

(77)

nona

* Tab

le w

as a

dapt

ed a

nd m

odifi

ed fr

om G

azze

rro

et a

l. (5

7) a

nd S

hita

ra e

t al.

(58)

** F

ollo

win

g a

typi

cal c

hols

tero

l-low

erin

g do

se (2

0-80

mg/

day)

Hig

hest

kno

wn

tole

rate

d do

se in

ca

ncer

pat

ient

sSt

atin

Lipo

phili

cR

equi

res

Met

abol

ic

Act

ivat

ion

Peak

Pla

sma

Con

cent

ratio

n (4

0 m

g)

ng/m

l (nM

)**

CYP

3A4

Subs

trat

e

Table 1-1: The pharmacokinetic

properties of the different statins.


11

1.2.3 The mevalonate pathway and cancer

The mevalonate (MVA) pathway is an essential metabolic pathway whose end

products provide the cell with bioactive molecules (Figure 1-1). These include: sterols such as

cholesterol involved in membrane integrity and steroid production; farnesyl and

geranylgeranyl isoprenoids which are substrates for the post-translational modification of

proteins such as those of the Ras family; dolichol; ubiquinone; and isopentenyladenine. In

hepatocytes, inhibition of HMGCR, the rate-limiting enzyme of the MVA pathway, and

subsequent depletion of sterols leads to a restorative feedback response mediated by the

transcription factor sterol regulatory element binding transcription factor 2 (SREBP2). When

deprived of sterols, the latent SREBF2 is activated, transported to the golgi, cleaved by golgi-

resident site-1 and -2 proteases and translocated to the nucleus where it binds to sterol

responsive elements (SRE) initiating transcription of sterol responsive genes such as HMGCR

and low-density lipoprotein receptor (LDLr). Increased LDLr transcription leads to higher

LDLr protein at the membrane surface and LDL-uptake leading to decreased plasma

cholesterol levels53, 64-66. Drs Goldstein and Brown revealed this elegant regulation of

cholesterol homeostasis for which they were awarded the Nobel prize67. Studies in other

normal cells, limited mostly to circulating mononuclear cells have suggested that normal cells

are subject to similar regulation. A recent report demonstrated that following treatment of

healthy subjects with atorvastatin, LDLr mRNA levels were increased in circulating

mononuclear 68 and others have reported elevated HMGCR activity in mononuclear

leukocytes of healthy subject following statin administration69.


12

Figure 1-1: The MVA pathway.

Statins block the mevalonate pathway by inhibiting the rate-limiting enzyme, HMG-

CoA reductase (HMGCR). Statins compete with the natural substrate, HMG-CoA, for the

active site of HMG-CoA reductase. Mevalonate (MVA), the product of the reaction, is the

precursor to many critical cellular end-products essential for cell proliferation and survival.


13

Several lines of evidence suggest that the MVA pathway is dysregulated in cancer at

several levels and not subject to normal feedback control. Acute leukemia patients are often

hypocholesterolemic and their leukemic cells were found to have higher LDL-receptor (LDLr)

activity inversely correlated with their plasma cholesterol levels70 and as a consequence

higher LDL uptake71. Furthermore, there was elevated HMGCR activity in mononuclear cells

from leukemic patients relative to mononuclear cells from healthy subjects72 and leukemic

cells were not as responsive to sterol changes as normal cells73. Others groups have found that

AML cells isolated from patients increase intracellular cholesterol in response to treatment

with chemotherapeutic agents although this was not correlated with increased LDL

accumulation indicating that de novo cholesterol synthesis was responsible for this increase74.

Taken together, aberrant cholesterol homeostasis and elevated HMGCR activity has been

found in AML patient cells indicating that a window of therapeutic intervention by agents that

target the MVA pathway such as statins exists.

Aberrant MVA pathway regulation has also been documented in other hematological

malignancies. Our laboratory has surveyed 17 multiple myeloma (MM) cell lines and

stratified them according to their susceptibility to undergo lovastatin-induced apoptosis75.

Yielding approximately equal sized cohorts defined by clear differences in statin-sensitivity,

enabled the further characterization of the panel and find molecular markers of statin

sensitivity. The MM cell lines had a wide range of genetic abnormalities and some

abnormalities differed and were common to both cohorts of MM cell lines. However, no clear

cohort-defining abnormality that could be correlated with statin-sensitivity was found

although this lack of a genetic biomarker could be a consequence of the underpowered study.

Interestingly, in subsequent studies, our group found that the MM cell lines that were

sensitive to lovastatin-induced apoptosis were deficient in sterol-feedback control and unable

to upregulate HMGCR and hydroxymethylglutaryl coenzyme A synthase 1 (HMGCS1)76. The


14

conferral of sensitivity in MM cell lines with abnormal sterol-feedback responses further

strengthens the concept that tumour cells with dysregulated MVA pathway are targetable by

statins.

Our group found that over-expression of HMGCR mRNA, and other members of the

MVA pathway including HMGCS1 and farnesyl pyrophosphate synthase (FDPS), was

correlated with poor prognosis of and decreased survival of breast cancer patients77. In the

same study we found that ectopic introduction of the catalytic portion of HMGCR induced

transformation in non-transformed cells and increased oncogenic potential in transformed

cells cell culture and in vivo. A recent study investigating mutant protein 53 (p53)’s

disruptive effect on mammary acinar morphogenesis showed that the mutant’s effects are

mediated by the MVA pathway78 which was over-expressed in breast cancer cells harbouring

the mutant and grown in 3D (3 dimensional) culture. Following analysis of five breast cancer

patient data sets, it was determined that the samples harboring p53 mutations were also

characterized by elevated expression of sterol biosynthesis genes compared to patient tumours

bearing wild-type p53. Furthermore, clustering analysis demonstrated that the cluster with the

highest MVA gene expression was associated with the poorest prognosis. Examined

individually, over-expression of MVA pathway genes, including HMGCR and HMGCS1,

was also correlated with worse prognosis. An earlier report noted that exogenous addition of

MVA promoted in vivo tumour growth of subcutaneous xenografts79. Although it is not clear

whether the dysregulation of the MVA pathway plays a causal role in oncogenesis, MVA

pathway activity is important in cancer. MVA pathway dysregulation has been demonstrated

at multiple levels in cancer cell lines and primary patient samples. A dysregulated MVA

pathway presents a tumour vulnerability with the potential of positive therapeutic

consequences when targeted by agents, such as statins, that inhibit it.


15

1.2.4 Statins as preventative anti-cancer agents

Millions of patients are prescribed statins on a yearly basis enabling retrospective

case-control and prospective cohort studies to uncover potential protective non-cardiovascular

effects. Accumulating epidemiological evidence accrued over the last two decades suggests

statin use may reduce cancer incidence but conclusive evidence recommending statin use for

primary prevention is lacking. Ample epidemiological analysis exists for the most common

cancers such as breast and lung cancer. In breast cancer, several large cohort studies amongst

them the Cancer Prevention Study II Nutrition Cohort and the Kaiser Permanente80-82 and

case-control studies have demonstrated no relationship between statin use and incidence of

breast cancer. Some of these studies were criticized for lack of information on potential

confounders such as physical activity and diet, short follow-up times, lack of consideration

for duration and type of statins used. When considering key variables such as the nature of

the statin used, duration of usage, and tumour sub-type positive associations were evident. A

large cohort study found that there was an 18 % reduction in breast cancer risk amongst

lipophilic statin users83. A case-control study found that long term statin use of greater than 5

years was associated with a decreased breast cancer risk84. A retrospective cohort study

analyzing the occurrence of ER/PR-negative (estrogen receptors/progesterone receptors)

tumours amongst statin users found that breast cancer patients taking statins had fewer

ER/PR-negative tumours, which were of lower grade and stage85.

Although studies examining the use of statins for the secondary prevention of breast

cancer have not been prevalent the few published studies have been extremely encouraging of

statin use to prevent breast cancer recurrence. A prospective cohort study found that lipophilic

statin use amongst early stage breast cancer patients, post-cancer diagnosis, was associated

with decreased risk of breast cancer recurrence86. The risk of recurrence in this study was

furthermore inversely correlated with duration of statin use. A recently published, much larger


16

and better-powered Danish study found a significant decrease in breast cancer recurrence

amongst lipophilic statin users87. Of twenty-five epidemiologic studies composed of case-

control, cohort and examining statin use and incidence or recurrence of breast cancer, seven

found that there was a reduction in breast cancer incidence or recurrence associated with

statin use.

Amongst lung cancer patients, many studies reported no association between statin use

and lung cancer risk88, 89. However, these studies were characterized by relatively low

numbers of statin users. A large cohort study of 2000 veteran lung cancer patients, on the

other hand found that statin use was associated with a 45% decreased risk of lung cancer90.

Another retrospective cohort study composed of veterans reported a 30% reduction in the risk

of lung cancer amongst statin users91. Taken together, there is supporting epidemiological

evidence to indicate that statins may play a role in the prevention of lung cancer amongst

male users. Of fourteen epidemiologic studies composed of case-control, cohort and

examining statin use and incidence of lung cancer, three found that there was a reduction in

lung cancer incidence associated with statin use. However, unlike in breast cancer, few

attempts have been made to focus on effects specific to lung cancer sub-types and types of

statin used.

In hematological malignancies there has been some evaluation of preventive statin use

but this has been largely limited by relatively low numbers of patients suffering from these

cancers. The largest study examining lymphoid neoplasms, the European case-control EPI-

LYMPH study, found that there was a significant reduction in lymphoma risk amongst statin-

users. The same risk reduction were maintained when the 2,362 patients were stratified

according to histological subtype including multiple myeloma (MM) although that group

contained only 288 MM patients92. Mention of the role of statins as preventative agents in

other tumour types not presented in this thesis is beyond the scope of this introduction and


17

this topic comprehensively reviewed elsewhere93. There is supportive evidence that statin use

decreases the incidence and recurrence of breast cancer, lung cancer and hematological

malignancies, especially when the type of statin used, tumour sub-type and duration of statin

use is carefully considered. However, overall evidence supporting the chemopreventive

effects of statins is inconsistent and reflects the limited nature of observational studies. Many

of these observational studies did not properly evaluate links between prevention and statin

dose and duration of statin use. Confounding by indication cannot be ruled out. Statin-users

may differ from non-statin users in a way that might positively impact cancer risk factors

independent of any specific effects of statins. Adherence to drug therapy especially for

preventative purposes has been linked to individuals that are generally more health-conscious

and have more education. It has been suggested that low levels of cholesterol are associated

with increased cancer incidence94, 95, and it may be that hypercholesterolemic patients are

protected for years by their increased cholesterol levels before they start to use statins. Taken

together, to ultimately address whether statin use is associated with a positive anti-cancer

effect, prospective clinical trials directly assessing statin benefit in cancer patients need to

carried out.

1.2.5 Statins in the clinic

Early cell culture studies in the nineties exploring the anti-cancer effects of statins

primarily used lovastatin in a variety of solid and hematological tumour cell lines. Lovastatin

was shown induce cell cycle arrest in a dose and time-dependent manner in the solid tumour

cell lines 96-98 and in hematological cells99-101 at concentrations ranging from 2-10 µM. The

statin induced anti-cancer effects were determined to be a result of direct HMGCR inhibition

as they were reversible with the addition of MVA. As the cell culture lovastatin


18

concentrations used to inhibit growth or induce apoptosis were clearly above what was

clinically achievable with the typical 10-80 mg (~0.15-1mg/kg/day) oral daily dose, initial

prospective clinical trials involving the evaluation of statin use in cancer patients have sought

to address key issues such elevated dosing and potential dose limiting toxicities. Early dose-

finding studies established that lovastatin can be tolerated at high concentrations greatly

exceeding cholesterol-lowering doses. In two separate studies, short-term oral lovastatin

administration at doses ranging from 1 to 45 mg/kg/day to patients with advanced pre-treated

malignancies was tolerated102,103. In the earliest study, lovastatin was administered 88

patients with a variety of solid tumours for seven consecutive days in monthly cycles and was

well tolerated at doses up to 25 mg/kg; above that dosing, side effects such as myopathy,

fatigue and nausea occurred but could be partly controlled with the concomitant

administration of ubiquinone102. In a similarly dosed phase I study, 18 patients with glioma

and glioblastoma multiforme were administered lovastatin of 30 mg/kg/day and did not

experience adverse reactions during treatment103. High doses of other statins such as

simvastatin are also well tolerated. Simvastatin at doses of 15mg/kg/day was administered in

patients with relapsed or refractory myeloma or lymphoma104. Taken together, statins are well

tolerated at levels greatly exceeding cholesterol-lowering doses.

The pharmacodynamics of statins administered at higher doses have been also been

addressed albeit in fewer studies. A trial where lovastatin was administered at a dose of 10

mg/m2 to 415 mg/m2. every 6 hours for 96 hours measured lovastatin peak plasma bioactivity

levels of 0.06-12.3 µM105. However, the measured concentrations were not dose-dependent

and any dose-dependent relationship is likely complicated by patient variability to hydrolyze

lovastatin to the active form of the drug. The above-mentioned early lovastatin trial also

detected micromolar concentrations of lovastatin peaking at 3.9 µM105. Although

concentrations of lovastatin equivalent to doses needed in tissue culture studies to achieve an


19

anti-cancer effect were detected in some patients, very few objective responses were evident

in those trials were lovastatin was tested in a monotherapy setting.

Statin efficacy, when combined with other agents has been demonstrated. A few

studies have documented statin therapy success in hepatocellular carcinoma (HCC). HCC

patients receiving pravastatin (20-40 mg/day) and with the standard of care transarterial

chemoembolization (TACE) experienced a significantly longer median survival when

compared to patients receiving TACE alone106. Another group demonstrated pravastatin

efficacy in HCC as an adjuvant treatment following transcatheter arterial embolization (TAE)

and oral 5-FU treatment107. Recently, a trial combining lovastatin with thalidomide and

dexamethasone (TD) in patients with relapsed or refractory MM illustrated lovastatin

prolonged overall survival and progression-free survival108. Encouraging response rates in an

AML phase I trial where pravastatin of doses of up to 1680mg/day (24mg/kg/day) was safely

combined with idaraubicin and high dose cytarabine, in newly diagnosed and salvage AML

patients109 has warranted an ongoing follow-up multi-site phase II trial (NCT00840177). In a

neo-adjuvant trial, women with stage I breast cancer and ductal carcinoma in-situ (DCIS)

were administered 80 mg or 20 mg of fluvastatin 3-6 weeks before surgery. Fluvastatin

treatment reduced proliferation and increased apoptosis in the high-grade tumours110.

According to ClinicalTrials.gov, there are 35 clinical trials involving statins being

evaluated in the clinic as anti-cancer agents: 6 trials in breast cancer patients, 5 in lung cancer

patients and 6 in patients with hematological malignancies. Out of the 35 studies, 4 are statin

prevention studies involving women with either high-inherited risk for breast cancer or who

have previously undergone surgery for ductal carcinoma. Monotherapeutic use of statins is a

feature of 11 of the 35 clinical trials; 2 are early phase safety studies. Interestingly, another 2

of the clinical trials where statin is used as a monotherapy will attempt to elucidate a

biomarker of statin response, assess intracellular statin levels (NCT00828282), and test the


20

hypothesis of statin-selective effects on basal subtype breast cancer (NCT00807950). The

majority of the ongoing studies are testing statins in a combination setting. Twenty clinical

studies are evaluating statins in combination with other agents at various points of therapeutic

intervention. Most of these agents are the standard of care chemotherapies. Interestingly, there

are two active studies also combining statins with non-standard agents. One single group

assessment trial is combining simvastatin with metformin in treating men with recurrent

prostate carcinoma (NCT01561482). Another study, also currently recruiting, will treat

chemotherapy resistant, refractory MM patients with simvastatin and zoledronic acid in

addition to the chemotherapy (NCT01772719). The results from the large number of ongoing

studies are eagerly awaited and will be instructive in how best to apply statins in clinical use..

These prospective trials have been and will continue to be inexorably far more informative

than the non-interventional retrospective analyses.

Statins are tolerated at high doses but it remains unclear whether they will need to be

administered at high doses as clinical anti-cancer efficacy has been demonstrated at both

cholesterol lowering doses106, 107, 110-113 and higher statin doses108, 109. Statin-responsive

tumours include hematological malignancies, AML109, 111 and MM108, 113 and breast110, 112.

Although there are clinical trials testing statins as single agents, many of these studies are not

only evaluating patient response but are also seeking to establish the safety of statins in a

particular clinical setting or attempting to establish a biomarker of statin response. Statins are

unlikely to be used as a monotherapy with the possible exception in neo-adjuvant setting.

This is not only a consequence of limited statin anti-cancer efficacy when used alone102, 103, 114,

115, but a consequence of trying to eradicate an inherently complex and heterogeneous disease

necessitating the need for multi-modal combinatorial approaches.


21

1.3 Combination therapy

There are many obstacles to effective drug therapy when treating cancer patients. Not

only are tumour cells heterogeneous and dependent on many altered molecular pathways

subject to complex intertwining feedback loops, but they can quickly develop resistance to

single agents. Resistance, intrinsic or acquired is equally a problem for the classically used

cytotoxics and novel targeted therapies. Single agents are rarely able to provide sustained

long-term response and there are numerous benefits to combinatorial approaches. Drugs with

different mechanisms of action share non over-lapping toxicities. Their co-administration

minimizes individual resistance. For targeted agents, combination therapy can combat

feedback loops and a redundancy in signaling, factors that often times limit the efficacy of a

single agent.

The therapy-limiting presence of feedback loops was elegantly demonstrated is a

recent screen that sought to understand the unresponsiveness of colon cancer harboring the B

Rapidly Accelerated Fibrosarcoma (BRAF V600E) oncoprotein. BRAF (V600E) is targeted

by vemurafenib, a small molecule inhibitor approved for the treatment of melanoma. Through

an RNA interference screen in colon tumour lines, it was shown that knockdown of the

epidermal growth factor receptor (EGFR) displayed strong synergy with vemurafenib. Colon

cancer cells rapidly activate EGFR in response to vemurafenib treatment, in contrast to

melanoma cells that express low EGFR levels, thereby negating any anti-tumoural

vemurafenib effects116. Therefore, using vemurafenib in combination with an EGFR targeted

agent such as erlotinib in the treatment BRAF (V600E) bearing-oncoprotein colon tumours

would circumvent this therapy-limiting feedback activation. These rationally crafted

combinations of targeted agents are currently being evaluated in the clinic117 but require a

thorough molecular understanding of each individual targeted agent and the signaling


22

pathways they perturb. Furthermore, the costs involved in combining two targeted agents in

chemotherapeutic regimen are likely cost-prohibitive.

Recent efforts to identify novel combinations of anti-cancer agents, discussed at the

24th EORTC-NCI-AACR Symposium on ‘Molecular Targets and Cancer Therapeutics’42

have explored unbiased approaches. Led by S. Holbeck, 100 FDA-approved small-molecule

anti-cancer agents are being screen in combination on the NCI’s panel of 60 cell lines and

thousands of unexpected combinations are evaluated for their anti-cancer efficacy.

Preclinically successful combinations, that include agents like statins, can be fast-tracked into

clinical trials because their pharmacokinetic, dosing and toxicity have already been

established. The quest for novel, efficacious combinations is paramount to the successful

treatment of patients that do not respond to the standard of care treatments currently available.


23

1.4 Research Objectives and thesis outline

My global hypothesis is that targeting the mevalonate (MVA) pathway in tumour cells

is an effective strategy to induce tumour cell apoptosis. My research objectives are to explore

how to maximize statin induced apoptosis: (i) through the use of a pharmacological screen to

identify compound/s that synergize with statins in inducing apoptosis in AML and MM, (ii)

through the use of a genome-wide shRNA screen to identify novel pathways/targets whose

knockdown enhances the vulnerability to statin-inhibition in lung and breast cancer.

Chapter 2 of this thesis uncovers a novel drug combination with strong antimyeloma

and antileukemic activity. Our focus in hematological malignancies stems from preclinical

and clinical evidence generated by our group and others, demonstrating dysregulation of the

MVA pathway in both AML and MM. Furthermore, positive evidence hinting at statin

efficacy AML and MM patients following statin treatment suggested to us that statins will be

clinically useful in these tumour types. Through the use of a pharmacological screen

composed of 100 off-patent, FDA-approved drugs, dipyridamole was found to potentiate the

anti-cancer effects of atorvastatin in multiple myeloma. Subsequent validation in a panel of

MM and AML cell lines demonstrated that the combination was synergistic and capable of

inducing apoptosis. Primary MM and AML patient samples were also vulnerable to the

combinations’ apoptosis-inducing effects and effectiveness in a leukemia xenograft model

was demonstrated. The chapter is concluded by explorations into the mechanism of drug

synergy.

Chapter 3 of this thesis details the execution of a genome-wide shRNA screen in the

A549 lung cancer cell line stably transduced with the RNAi Consortium (TRC1) shRNA

library. Sublethal fluvastatin exposure over several passages uncovered drop-out hits

identified through microarray hybridization of amplified genomic DNA. Several hits were


24

validated in lung and breast cancer cells, the most promising of which, the transcription factor

SREBF2, was also characterized in vivo. Other promising kinases were also identified which

are subject to further exploration.

Chapter 4 of this thesis discuses the overall feasibility of effectively repurposing

statins as anti-cancer agents and focuses on discussing the gaps between the cell culture

molecular mechanisms postulated to be responsible for the anti-cancer effects and the in vivo

translational relevance of these findings. Furthermore, a discussion on the importance of

finding statin-responsive patients and a biomarker of statin response or intrinsic sensitivity

will be discussed. This is paramount to extending the benefit of statins to as many patients

without incurring unwarranted side effects. Overall, this thesis extends our understanding of

how to maximize the anti-tumour effects of statins by successfully combining them with

novel, unexpected agents and shRNAs targeting kinases and enzymes of the of the MVA

pathway.


25

1.5 References

1. Globocan 2008. IARC 2010. 2. Perou CM, Sorlie T, Eisen MB, van de Rijn M, Jeffrey SS, Rees CA et al. Molecular

portraits of human breast tumours. Nature 2000; 406(6797): 747-52. 3. Foulkes WD, Smith IE, Reis-Filho JS. Triple-negative breast cancer. The New

England journal of medicine 2010; 363(20): 1938-48. 4. Curtis C, Shah SP, Chin SF, Turashvili G, Rueda OM, Dunning MJ et al. The

genomic and transcriptomic architecture of 2,000 breast tumours reveals novel subgroups. Nature 2012; 486(7403): 346-52.

5. Chang HH, Dreyfuss JM, Ramoni MF. A transcriptional network signature

characterizes lung cancer subtypes. Cancer 2011; 117(2): 353-60. 6. Shah SP, Roth A, Goya R, Oloumi A, Ha G, Zhao Y et al. The clonal and mutational

evolution spectrum of primary triple-negative breast cancers. Nature 2012; 486(7403): 395-9.

7. Nik-Zainal S, Van Loo P, Wedge DC, Alexandrov LB, Greenman CD, Lau KW et al.

The life history of 21 breast cancers. Cell 2012; 149(5): 994-1007. 8. Gerlinger M, Rowan AJ, Horswell S, Larkin J, Endesfelder D, Gronroos E et al.

Intratumor heterogeneity and branched evolution revealed by multiregion sequencing. The New England journal of medicine 2012; 366(10): 883-92.

9. Hernandez L, Wilkerson PM, Lambros MB, Campion-Flora A, Rodrigues DN,

Gauthier A et al. Genomic and mutational profiling of ductal carcinomas in situ and matched adjacent invasive breast cancers reveals intra-tumour genetic heterogeneity and clonal selection. The Journal of pathology 2012; 227(1): 42-52.

10. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell 2011;

144(5): 646-74. 11. 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 and differentiation 2012; 19(1): 107-20.

12. Ocker M, Hopfner M. Apoptosis-modulating drugs for improved cancer therapy.

European surgical research. Europaische chirurgische Forschung. Recherches chirurgicales europeennes 2012; 48(3): 111-20.

13. Devita VT, Jr., Serpick AA, Carbone PP. Combination chemotherapy in the treatment

of advanced Hodgkin's disease. Annals of internal medicine 1970; 73(6): 881-95.


26

14. Lind MJ, Ardiet C. Pharmacokinetics of alkylating agents. Cancer surveys 1993; 17:

157-88. 15. Momota H, Narita Y, Miyakita Y, Shibui S. Secondary hematological malignancies

associated with temozolomide in patients with glioma. Neuro-oncology 2013. 16. Belani CP. Recent updates in the clinical use of platinum compounds for the treatment

of lung, breast, and genitourinary tumors and myeloma. Seminars in oncology 2004; 31(6 Suppl 14): 25-33.

17. Ledermann JA, Kristeleit RS. Optimal treatment for relapsing ovarian cancer. Annals

of oncology : official journal of the European Society for Medical Oncology / ESMO 2010; 21 Suppl 7: vii218-22.

18. Walling J. From methotrexate to pemetrexed and beyond. A review of the

pharmacodynamic and clinical properties of antifolates. Investigational new drugs 2006; 24(1): 37-77.

19. Pommier Y. Topoisomerase I inhibitors: camptothecins and beyond. Nature reviews.

Cancer 2006; 6(10): 789-802. 20. Petersdorf SH, Rankin C, Head DR, Terebelo HR, Willman CL, Balcerzak SP et al.

Phase II evaluation of an intensified induction therapy with standard daunomycin and cytarabine followed by high dose cytarabine for adults with previously untreated acute myeloid leukemia: a Southwest Oncology Group study (SWOG-9500). American journal of hematology 2007; 82(12): 1056-62.

21. Morris PG, Fornier MN. Microtubule active agents: beyond the taxane frontier.

Clinical cancer research : an official journal of the American Association for Cancer Research 2008; 14(22): 7167-72.

22. Joerger M, Thurlimann B. Chemotherapy regimens in early breast cancer: major

controversies and future outlook. Expert review of anticancer therapy 2013; 13(2): 165-78.

23. Cleeland CS, Allen JD, Roberts SA, Brell JM, Giralt SA, Khakoo AY et al. Reducing

the toxicity of cancer therapy: recognizing needs, taking action. Nature reviews. Clinical oncology 2012; 9(8): 471-8.

24. Sonis ST. Regimen-related gastrointestinal toxicities in cancer patients. Current

opinion in supportive and palliative care 2010; 4(1): 26-30. 25. Sul JK, Deangelis LM. Neurologic complications of cancer chemotherapy. Seminars

in oncology 2006; 33(3): 324-32. 26. Albini A, Pennesi G, Donatelli F, Cammarota R, De Flora S, Noonan DM.

Cardiotoxicity of anticancer drugs: the need for cardio-oncology and cardio-oncological prevention. Journal of the National Cancer Institute 2010; 102(1): 14-25.


27

27. Raguz S, Yague E. Resistance to chemotherapy: new treatments and novel insights

into an old problem. British journal of cancer 2008; 99(3): 387-91. 28. Morgan G, Ward R, Barton M. The contribution of cytotoxic chemotherapy to 5-year

survival in adult malignancies. Clin Oncol (R Coll Radiol) 2004; 16(8): 549-60. 29. Sapoznik S, Hammer O, Ortenberg R, Besser MJ, Ben-Moshe T, Schachter J et al.

Novel anti-melanoma immunotherapies: disarming tumor escape mechanisms. Clinical & developmental immunology 2012; 2012: 818214.

30. Beretta GD, Petrelli F, Stinco S, Cabiddu M, Ghilardi M, Squadroni M et al. FOLFIRI

+ bevacizumab as second-line therapy for metastatic colorectal cancer pretreated with oxaliplatin: a pooled analysis of published trials. Med Oncol 2013; 30(1): 486.

31. Montagna E, Cancello G, Bagnardi V, Pastrello D, Dellapasqua S, Perri G et al.

Metronomic chemotherapy combined with bevacizumab and erlotinib in patients with metastatic HER2-negative breast cancer: clinical and biological activity. Clinical breast cancer 2012; 12(3): 207-14.

32. Melichar B, Prochazkova-Studentova H, Vitaskova D. Bevacizumab in combination

with IFN-alpha in metastatic renal cell carcinoma: the AVOREN trial. Expert review of anticancer therapy 2012; 12(10): 1253-61.

33. Stark D, Nankivell M, Pujade-Lauraine E, Kristensen G, Elit L, Stockler M et al.

Standard chemotherapy with or without bevacizumab in advanced ovarian cancer: quality-of-life outcomes from the International Collaboration on Ovarian Neoplasms (ICON7) phase 3 randomised trial. The lancet oncology 2013; 14(3): 236-43.

34. Chapman PB, Hauschild A, Robert C, Haanen JB, Ascierto P, Larkin J et al. Improved

survival with vemurafenib in melanoma with BRAF V600E mutation. The New England journal of medicine 2011; 364(26): 2507-16.

35. Kwak EL, Bang YJ, Camidge DR, Shaw AT, Solomon B, Maki RG et al. Anaplastic

lymphoma kinase inhibition in non-small-cell lung cancer. The New England journal of medicine 2010; 363(18): 1693-703.

36. Kantarjian H, Sawyers C, Hochhaus A, Guilhot F, Schiffer C, Gambacorti-Passerini C

et al. Hematologic and cytogenetic responses to imatinib mesylate in chronic myelogenous leukemia. The New England journal of medicine 2002; 346(9): 645-52.

37. Bjorkholm M, Ohm L, Eloranta S, Derolf A, Hultcrantz M, Sjoberg J et al. Success

story of targeted therapy in chronic myeloid leukemia: a population-based study of patients diagnosed in Sweden from 1973 to 2008. J Clin Oncol 2011; 29(18): 2514-20.

38. Demetri GD, von Mehren M, Blanke CD, Van den Abbeele AD, Eisenberg B, Roberts

PJ et al. Efficacy and safety of imatinib mesylate in advanced gastrointestinal stromal tumors. The New England journal of medicine 2002; 347(7): 472-80.


28

39. Fakih M, Vincent M. Adverse events associated with anti-EGFR therapies for the treatment of metastatic colorectal cancer. Curr Oncol 2010; 17 Suppl 1: S18-30.

40. Force T, Kerkela R. Cardiotoxicity of the new cancer therapeutics--mechanisms of,

and approaches to, the problem. Drug discovery today 2008; 13(17-18): 778-84. 41. Niraula S, Seruga B, Ocana A, Shao T, Goldstein R, Tannock IF et al. The price we

pay for progress: a meta-analysis of harms of newly approved anticancer drugs. Journal of clinical oncology : official journal of the American Society of Clinical Oncology 2012; 30(24): 3012-9.

42. Kummar S, Doroshow JH. Molecular targets in cancer therapy. Expert review of

anticancer therapy 2013; 13(3): 267-9. 43. Fojo T, Grady C. How much is life worth: cetuximab, non-small cell lung cancer, and

the $440 billion question. Journal of the National Cancer Institute 2009; 101(15): 1044-8.

44. Ashburn TT, Thor KB. Drug repositioning: identifying and developing new uses for

existing drugs. Nature reviews. Drug discovery 2004; 3(8): 673-83. 45. Duenas-Gonzalez A, Garcia-Lopez P, Herrera LA, Medina-Franco JL, Gonzalez-

Fierro A, Candelaria M. The prince and the pauper. A tale of anticancer targeted agents. Molecular cancer 2008; 7: 82.

46. Skrtic M, Sriskanthadevan S, Jhas B, Gebbia M, Wang X, Wang Z et al. Inhibition of

mitochondrial translation as a therapeutic strategy for human acute myeloid leukemia. Cancer cell 2011; 20(5): 674-88.

47. Sharmeen S, Skrtic M, Sukhai MA, Hurren R, Gronda M, Wang X et al. The

antiparasitic agent ivermectin induces chloride-dependent membrane hyperpolarization and cell death in leukemia cells. Blood 2010; 116(18): 3593-603.

48. Mao X, Li X, Sprangers R, Wang X, Venugopal A, Wood T et al. Clioquinol inhibits

the proteasome and displays preclinical activity in leukemia and myeloma. Leukemia 2009; 23(3): 585-90.

49. Evans JM, Donnelly LA, Emslie-Smith AM, Alessi DR, Morris AD. Metformin and

reduced risk of cancer in diabetic patients. BMJ 2005; 330(7503): 1304-5. 50. Decensi A, Puntoni M, Goodwin P, Cazzaniga M, Gennari A, Bonanni B et al.

Metformin and cancer risk in diabetic patients: a systematic review and meta-analysis. Cancer Prev Res (Phila) 2010; 3(11): 1451-61.

51. Goodwin PJ, Pritchard KI, Ennis M, Clemons M, Graham M, Fantus IG. Insulin-

lowering effects of metformin in women with early breast cancer. Clinical breast cancer 2008; 8(6): 501-5.


29

52. Hadad S, Iwamoto T, Jordan L, Purdie C, Bray S, Baker L et al. Evidence for biological effects of metformin in operable breast cancer: a pre-operative, window-of-opportunity, randomized trial. Breast cancer research and treatment 2011; 128(3): 783-94.

53. Brown MS, Dana SE, Goldstein JL. Regulation of 3-hydroxy-3-methylglutaryl

coenzyme A reductase activity in human fibroblasts by lipoproteins. Proceedings of the National Academy of Sciences of the United States of America 1973; 70(7): 2162-6.

54. Endo A, Kuroda M, Tanzawa K. Competitive inhibition of 3-hydroxy-3-

methylglutaryl coenzyme A reductase by ML-236A and ML-236B fungal metabolites, having hypocholesterolemic activity. FEBS letters 1976; 72(2): 323-6.

55. Endo A, Kuroda M, Tsujita Y. ML-236A, ML-236B, and ML-236C, new inhibitors of

cholesterogenesis produced by Penicillium citrinium. The Journal of antibiotics 1976; 29(12): 1346-8.

56. Yamamoto A, Sudo H, Endo A. Therapeutic effects of ML-236B in primary

hypercholesterolemia. Atherosclerosis 1980; 35(3): 259-66. 57. Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease:

the Scandinavian Simvastatin Survival Study (4S). Lancet 1994; 344(8934): 1383-9. 58. Prevention of cardiovascular events and death with pravastatin in patients with

coronary heart disease and a broad range of initial cholesterol levels. The Long-Term Intervention with Pravastatin in Ischaemic Disease (LIPID) Study Group. The New England journal of medicine 1998; 339(19): 1349-57.

59. Downs JR, Clearfield M, Weis S, Whitney E, Shapiro DR, Beere PA et al. Primary

prevention of acute coronary events with lovastatin in men and women with average cholesterol levels: results of AFCAPS/TexCAPS. Air Force/Texas Coronary Atherosclerosis Prevention Study. JAMA : the journal of the American Medical Association 1998; 279(20): 1615-22.

60. Shitara Y, Sugiyama Y. Pharmacokinetic and pharmacodynamic alterations of 3-

hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors: drug-drug interactions and interindividual differences in transporter and metabolic enzyme functions. Pharmacology & therapeutics 2006; 112(1): 71-105.

61. Gazzerro P, Proto MC, Gangemi G, Malfitano AM, Ciaglia E, Pisanti S et al.

Pharmacological actions of statins: a critical appraisal in the management of cancer. Pharmacological reviews 2012; 64(1): 102-46.

62. Liu Y, Cheng Z, Ding L, Fang F, Cheng KA, Fang Q et al. Atorvastatin-induced acute

elevation of hepatic enzymes and the absence of cross-toxicity of pravastatin. International journal of clinical pharmacology and therapeutics 2010; 48(12): 798-802.


30

63. Williams D, Feely J. Pharmacokinetic-pharmacodynamic drug interactions with HMG-CoA reductase inhibitors. Clinical pharmacokinetics 2002; 41(5): 343-70.

64. Brown MS, Goldstein JL. The SREBP pathway: regulation of cholesterol metabolism

by proteolysis of a membrane-bound transcription factor. Cell 1997; 89(3): 331-40. 65. Horton JD, Goldstein JL, Brown MS. SREBPs: activators of the complete program of

cholesterol and fatty acid synthesis in the liver. J Clin Invest 2002; 109(9): 1125-31. 66. Brown MS, Goldstein JL. Multivalent feedback regulation of HMG CoA reductase, a

control mechanism coordinating isoprenoid synthesis and cell growth. Journal of lipid research 1980; 21(5): 505-17.

67. Goldstein JL, Brown MS. History of science. A golden era of Nobel laureates. Science

2012; 338(6110): 1033-4. 68. Pocathikorn A, Taylor RR, Mamotte CD. Atorvastatin increases expression of low-

density lipoprotein receptor mRNA in human circulating mononuclear cells. Clinical and experimental pharmacology & physiology 2010; 37(4): 471-6.

69. Stone BG, Evans CD, Prigge WF, Duane WC, Gebhard RL. Lovastatin treatment

inhibits sterol synthesis and induces HMG-CoA reductase activity in mononuclear leukocytes of normal subjects. Journal of lipid research 1989; 30(12): 1943-52.

70. Vitols S, Gahrton G, Bjorkholm M, Peterson C. Hypocholesterolaemia in malignancy

due to elevated low-density-lipoprotein-receptor activity in tumour cells: evidence from studies in patients with leukaemia. Lancet 1985; 2(8465): 1150-4.

71. Vitols S, Angelin B, Ericsson S, Gahrton G, Juliusson G, Masquelier M et al. Uptake

of low density lipoproteins by human leukemic cells in vivo: relation to plasma lipoprotein levels and possible relevance for selective chemotherapy. Proceedings of the National Academy of Sciences of the United States of America 1990; 87(7): 2598-602.

72. Vitols S, Norgren S, Juliusson G, Tatidis L, Luthman H. Multilevel regulation of low-

density lipoprotein receptor and 3-hydroxy-3-methylglutaryl coenzyme A reductase gene expression in normal and leukemic cells. Blood 1994; 84(8): 2689-98.

73. Tatidis L, Gruber A, Vitols S. Decreased feedback regulation of low density

lipoprotein receptor activity by sterols in leukemic cells from patients with acute myelogenous leukemia. Journal of lipid research 1997; 38(12): 2436-45.

74. Banker DE, Mayer SJ, Li HY, Willman CL, Appelbaum FR, Zager RA. Cholesterol

synthesis and import contribute to protective cholesterol increments in acute myeloid leukemia cells. Blood 2004; 104(6): 1816-24.

75. Wong WW, Clendening JW, Martirosyan A, Boutros PC, Bros C, Khosravi F et al.

Determinants of sensitivity to lovastatin-induced apoptosis in multiple myeloma. Mol Cancer Ther 2007; 6(6): 1886-97.


31

76. Clendening JW, Pandyra A, Li Z, Boutros PC, Martirosyan A, Lehner R et al.

Exploiting the mevalonate pathway to distinguish statin-sensitive multiple myeloma. Blood 2010; 115(23): 4787-97.

77. Clendening JW, Pandyra A, Boutros PC, El Ghamrasni S, Khosravi F, Trentin GA et

al. Dysregulation of the mevalonate pathway promotes transformation. Proceedings of the National Academy of Sciences of the United States of America 2010; 107(34): 15051-6.

78. Freed-Pastor WA, Mizuno H, Zhao X, Langerod A, Moon SH, Rodriguez-Barrueco R

et al. Mutant p53 disrupts mammary tissue architecture via the mevalonate pathway. Cell 2012; 148(1-2): 244-58.

79. Duncan RE, El-Sohemy A, Archer MC. Mevalonate promotes the growth of tumors

derived from human cancer cells in vivo and stimulates proliferation in vitro with enhanced cyclin-dependent kinase-2 activity. The Journal of biological chemistry 2004; 279(32): 33079-84.

80. Boudreau DM, Yu O, Miglioretti DL, Buist DS, Heckbert SR, Daling JR. Statin use

and breast cancer risk in a large population-based setting. Cancer epidemiology, biomarkers & prevention : a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology 2007; 16(3): 416-21.

81. Jacobs EJ, Newton CC, Thun MJ, Gapstur SM. Long-term use of cholesterol-lowering

drugs and cancer incidence in a large United States cohort. Cancer research 2011; 71(5): 1763-71.

82. Friedman GD, Flick ED, Udaltsova N, Chan J, Quesenberry CP, Jr., Habel LA.

Screening statins for possible carcinogenic risk: up to 9 years of follow-up of 361,859 recipients. Pharmacoepidemiology and drug safety 2008; 17(1): 27-36.

83. Cauley JA, McTiernan A, Rodabough RJ, LaCroix A, Bauer DC, Margolis KL et al.

Statin use and breast cancer: prospective results from the Women's Health Initiative. J Natl Cancer Inst 2006; 98(10): 700-7.

84. Boudreau DM, Gardner JS, Malone KE, Heckbert SR, Blough DK, Daling JR. The

association between 3-hydroxy-3-methylglutaryl conenzyme A inhibitor use and breast carcinoma risk among postmenopausal women: a case-control study. Cancer 2004; 100(11): 2308-16.

85. Kumar AS, Benz CC, Shim V, Minami CA, Moore DH, Esserman LJ. Estrogen

receptor-negative breast cancer is less likely to arise among lipophilic statin users. Cancer Epidemiol Biomarkers Prev 2008; 17(5): 1028-33.

86. Kwan ML, Habel LA, Flick ED, Quesenberry CP, Caan B. Post-diagnosis statin use

and breast cancer recurrence in a prospective cohort study of early stage breast cancer survivors. Breast Cancer Res Treat 2008; 109(3): 573-9.


32

87. Ahern TP, Pedersen L, Tarp M, Cronin-Fenton DP, Garne JP, Silliman RA et al.

Statin prescriptions and breast cancer recurrence risk: a Danish nationwide prospective cohort study. Journal of the National Cancer Institute 2011; 103(19): 1461-8.

88. Graaf MR, Beiderbeck AB, Egberts AC, Richel DJ, Guchelaar HJ. The risk of cancer

in users of statins. J Clin Oncol 2004; 22(12): 2388-94. 89. Blais L, Desgagne A, LeLorier J. 3-Hydroxy-3-methylglutaryl coenzyme A reductase

inhibitors and the risk of cancer: a nested case-control study. Archives of internal medicine 2000; 160(15): 2363-8.

90. Khurana V, Bejjanki HR, Caldito G, Owens MW. Statins reduce the risk of lung

cancer in humans: a large case-control study of US veterans. Chest 2007; 131(5): 1282-8.

91. Farwell WR, Scranton RE, Lawler EV, Lew RA, Brophy MT, Fiore LD et al. The

association between statins and cancer incidence in a veterans population. Journal of the National Cancer Institute 2008; 100(2): 134-9.

92. Fortuny J, de Sanjose S, Becker N, Maynadie M, Cocco PL, Staines A et al. Statin use

and risk of lymphoid neoplasms: results from the European Case-Control Study EPILYMPH. Cancer epidemiology, biomarkers & prevention : a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology 2006; 15(5): 921-5.

93. Boudreau DM, Yu O, Johnson J. Statin use and cancer risk: a comprehensive review.

Expert opinion on drug safety 2010; 9(4): 603-21. 94. Jacobs EJ, Gapstur SM. Cholesterol and cancer: answers and new questions. Cancer

epidemiology, biomarkers & prevention : a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology 2009; 18(11): 2805-6.

95. Williams RR, Sorlie PD, Feinleib M, McNamara PM, Kannel WB, Dawber TR.

Cancer incidence by levels of cholesterol. JAMA : the journal of the American Medical Association 1981; 245(3): 247-52.

96. Jakobisiak M, Bruno S, Skierski JS, Darzynkiewicz Z. Cell cycle-specific effects of

lovastatin. Proceedings of the National Academy of Sciences of the United States of America 1991; 88(9): 3628-32.

97. Sumi S, Beauchamp RD, Townsend CM, Jr., Uchida T, Murakami M, Rajaraman S et

al. Inhibition of pancreatic adenocarcinoma cell growth by lovastatin. Gastroenterology 1992; 103(3): 982-9.

98. Girgert R, Marini P, Janessa A, Bruchelt G, Treuner J, Schweizer P. Inhibition of the

membrane localization of p21 ras proteins by lovastatin in tumor cells possessing a mutated N-ras gene. Oncology 1994; 51(4): 320-2.


33

99. Dimitroulakos J, Nohynek D, Backway KL, Hedley DW, Yeger H, Freedman MH et

al. Increased sensitivity of acute myeloid leukemias to lovastatin-induced apoptosis: A potential therapeutic approach. Blood 1999; 93(4): 1308-18.

100. Newman A, Clutterbuck RD, Powles RL, Catovsky D, Millar JL. A comparison of the

effect of the 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors simvastatin, lovastatin and pravastatin on leukaemic and normal bone marrow progenitors. Leukemia & lymphoma 1997; 24(5-6): 533-7.

101. Newman A, Clutterbuck RD, Powles RL, Millar JL. Selective inhibition of primary

acute myeloid leukaemia cell growth by lovastatin. Leukemia 1994; 8(2): 274-80. 102. Thibault A, Samid D, Tompkins AC, Figg WD, Cooper MR, Hohl RJ et al. Phase I

study of lovastatin, an inhibitor of the mevalonate pathway, in patients with cancer. Clinical cancer research : an official journal of the American Association for Cancer Research 1996; 2(3): 483-91.

103. Larner J, Jane J, Laws E, Packer R, Myers C, Shaffrey M. A phase I-II trial of

lovastatin for anaplastic astrocytoma and glioblastoma multiforme. American journal of clinical oncology 1998; 21(6): 579-83.

104. van der Spek E, Bloem AC, van de Donk NW, Bogers LH, van der Griend R, Kramer

MH et al. Dose-finding study of high-dose simvastatin combined with standard chemotherapy in patients with relapsed or refractory myeloma or lymphoma. Haematologica 2006; 91(4): 542-5.

105. Holstein SA, Knapp HR, Clamon GH, Murry DJ, Hohl RJ. Pharmacodynamic effects

of high dose lovastatin in subjects with advanced malignancies. Cancer Chemother Pharmacol 2006; 57(2): 155-64.

106. Graf H, Jungst C, Straub G, Dogan S, Hoffmann RT, Jakobs T et al.

Chemoembolization combined with pravastatin improves survival in patients with hepatocellular carcinoma. Digestion 2008; 78(1): 34-8.

107. Kawata S, Yamasaki E, Nagase T, Inui Y, Ito N, Matsuda Y et al. Effect of pravastatin

on survival in patients with advanced hepatocellular carcinoma. A randomized controlled trial. Br J Cancer 2001; 84(7): 886-91.

108. Hus M, Grzasko N, Szostek M, Pluta A, Helbig G, Woszczyk D et al. Thalidomide,

dexamethasone and lovastatin with autologous stem cell transplantation as a salvage immunomodulatory therapy in patients with relapsed and refractory multiple myeloma. Annals of hematology 2011; 90(10): 1161-6.

109. Kornblau SM, Banker DE, Stirewalt D, Shen D, Lemker E, Verstovsek S et al.

Blockade of adaptive defensive changes in cholesterol uptake and synthesis in AML by the addition of pravastatin to idarubicin + high-dose Ara-C: a phase 1 study. Blood 2007; 109(7): 2999-3006.


34

110. Garwood ER, Kumar AS, Baehner FL, Moore DH, Au A, Hylton N et al. Fluvastatin reduces proliferation and increases apoptosis in women with high grade breast cancer. Breast Cancer Res Treat 2010; 119(1): 137-44.

111. Minden MD, Dimitroulakos J, Nohynek D, Penn LZ. Lovastatin induced control of

blast cell growth in an elderly patient with acute myeloblastic leukemia. Leuk Lymphoma 2001; 40(5-6): 659-62.

112. Bjarnadottir O, Romero Q, Bendahl PO, Jirstrom K, Ryden L, Loman N et al.

Targeting HMG-CoA reductase with statins in a window-of-opportunity breast cancer trial. Breast Cancer Res Treat 2013.

113. Schmidmaier R, Baumann P, Bumeder I, Meinhardt G, Straka C, Emmerich B. First

clinical experience with simvastatin to overcome drug resistance in refractory multiple myeloma. Eur J Haematol 2007; 79(3): 240-3.

114. Kim WS, Kim MM, Choi HJ, Yoon SS, Lee MH, Park K et al. Phase II study of high-

dose lovastatin in patients with advanced gastric adenocarcinoma. Investigational new drugs 2001; 19(1): 81-3.

115. Knox JJ, Siu LL, Chen E, Dimitroulakos J, Kamel-Reid S, Moore MJ et al. A Phase I

trial of prolonged administration of lovastatin in patients with recurrent or metastatic squamous cell carcinoma of the head and neck or of the cervix. Eur J Cancer 2005; 41(4): 523-30.

116. Prahallad A, Sun C, Huang S, Di Nicolantonio F, Salazar R, Zecchin D et al.

Unresponsiveness of colon cancer to BRAF(V600E) inhibition through feedback activation of EGFR. Nature 2012; 483(7387): 100-3.

117. Kummar S, Chen HX, Wright J, Holbeck S, Millin MD, Tomaszewski J et al.

Utilizing targeted cancer therapeutic agents in combination: novel approaches and urgent requirements. Nature reviews. Drug discovery 2010; 9(11): 843-56.


35

Chapter 2 : Combining statins and dipyridamole effectively targets

acute myelogenous leukemia and multiple myeloma

Contributions:

The majority of work presented in this chapter was completed by the author of this thesis.

Additional contributions are as follow:

Zhihua Li carried out expeirments for Figure 2.6 D

Dr. Peter J Mullen and Janice Pong carried out experiments for Figure 2.9


36

2.1. Abstract

Statins have been successfully utilized for decades to treat patients with

hypercholesterolemia. Promising preclinical evidence indicates that statins may be useful as

anti-cancer agents and that their anti-tumour efficacy is increased when combined with other

agents. To identify novel therapeutic combinations with elevated anti-cancer activity, we

screened atorvastatin in combination with FDA approved drugs and identified dipyridamole, a

drug used for the prevention of cerebral ischemic events, as a potentiator of the anti-myeloma

activity of statins. The statin-dipyridamole combination was synergistic and induced

apoptosis in cell lines and primary cells from myeloma and acute myelogenous leukemia

patients. Additionally, this novel drug combination decreased tumour growth of leukemia

xenografts. To decipher which activities of dipyridamole contribute to the synergy, we

combined statins with drugs mimicking each of the many pharmacological actions of

dipyridamole. We determined that modulators of cAMP (Cyclic adenosine monophosphate),

in combination with statins, also have anti-cancer activity. This suggests that the mechanism

of synergy between statins and dipyridamole may be attributable to the inhibitory effect of

dipyridamole on phosphodiesterases (PDEs). This work provides a strong rationale to

evaluate the therapeutic efficacy of a novel combination of the FDA approved drugs, statins

and dipyridamole, in further clinical investigations.


37

2.2. Introduction

There is an urgent need for novel therapeutic strategies in treating both acute

myelogenous leukemia (AML) and multiple myeloma (MM) especially in heavily pre-treated

and relapsed patients. Despite recent advances in MM treatment with the introduction of

thalidomide, landolidomide and bortezomib, it is difficult to achieve progression free survival

beyond 36 months1. In AML, survival is poor following relapse and 40-50% of older AML

patients and 20-30% of younger AML patients will experience primary inductive failure2.

Statins, potent inhibitors of the rate-limiting enzyme in the mevalonate (MVA) pathway, 3-

hydroxy-3-methylglutaryl coenzyme A reductase (HMGCR)3 have been used for decades in

the treatment of patients with hypercholesterolemia. Their frequent use in the prevention of

adverse cardiovascular events has led to accumulating epidemiological evidence that suggests

statin use may reduce cancer incidence4-6. In hematological malignancies, we and others have

shown that statins can effectively trigger tumour-specific apoptosis7-12,13 The apoptotic effects

of statins have been attributed to direct inhibition of HMGCR in tumour cells followed by

depletion of fundamental MVA-derived end-products such as isoprenoids and cholesterol10, 14,

15. In tumour cells, dysregulation of the MVA pathway at several levels has been postulated to

be responsible for the observed therapeutic index. For example, higher tumour levels of

HMGCR and other MVA pathway enzyme mRNA are associated with poor prognosis and

reduced survival in cancer patients16, 17. Dysregulation of the MVA pathway at the level of the

restorative sterol-feedback coordination is present in both MM9 and AML18, 19. In AML,

dysregulation of the MVA pathway can also cause abnormal cholesterol homeostasis leading

to overactive production and import of cholesterol contributing to decreased therapeutic

efficacy 11, 20. Taken together, dyregulation of the MVA pathway in hematological

malignancies provides a strong rationale for statin therapy.


38

Initial prospective clinical trials involving the evaluation of statin use in cancer

patients have sought to address key issues such as dose limiting toxicities. Early dose-finding

studies established that statins can be tolerated at high concentrations greatly exceeding

cholesterol-lowering doses which range form 20 to 80 mg/day (~0.3-1mg/kg/day). In two

separate studies, short-term oral lovastatin administration at doses ranging from 1 to 45

mg/kg/day to patients with advanced pre-treated malignancies was well tolerated21,22.

Simvastatin was tolerated at doses of 15mg/kg/day in patients with relapsed or refractory

myeloma or lymphoma23. In a phase I trial combining pravastatin with idaraubicin and high

dose cytarabine, in newly diagnosed and salvage AML patients, pravastatin doses of up to

1680mg/day (24mg/kg/day) were safely administered24. Therefore, high doses of statins can

be safely administered in the clinical cancer care setting but the ideal dosing regimen remains

unclear as anti-cancer efficacy has been observed with both high24 and low25, 26 cholesterol-

lowering doses.

Statins have also been safely combined with the standard of care therapy regimens in

MM and AML. Recently, a trial combining lovastatin with thalidomide and dexamethasone

(TD) in patients with relapsed or refractory MM illustrated lovastatin prolonged overall

survival and progression-free survival27. Encouraging response rates as well as the well-

tolerated addition of pravastatin in the abovementioned AML phase I trial24 has warranted an

ongoing follow-up phase II trial (NCT00840177). However, in a trial of twelve refractory or

relapsed MM patients simvastatin was administered with vincristine, doxorubicin and

dexamethasone (VAD chemotherapy) and only one partial response was observed28 Statins

can be combined with the standard of care in AML and MM patients without serious side

effects in inductive, consolidation and maintenance therapy settings. While this approach has

shown some promise, there remain many non-responsive patients, highlighting an urgent need

to develop new additional synergistic combinatorial approaches utilizing statin chemotherapy.


39

Building on promising results of statins as anti-cancer agents in AML and MM, we conducted

a pharmacological screen of FDA approved drugs in combination with statins to identify

novel combinations with anti-cancer efficacy in hematological malignancies that could be

immediately used for patient care. The screen identified dipyridamole, a commonly used anti-

platelet agent for the prevention of secondary stroke and ischemic heart attack, as a

potentiator of the anti-proliferative effects of statins in AML and MM cell lines. The

combination, synergistic and capable of inducing apoptosis at low micromolar doses in AML

and MM cells, effectively slowed tumour growth in a leukemia xenograft model and induced

apoptosis in primary AML and MM patient samples. Dipyridamole is known to elicit

numerous effects including phosphodiesterase (PDE) inhibition. Modulators of cAMP

signaling including other PDE inhibitors, also induced apoptosis in combination with statins.

Taken together, these findings not only postulate a role of the cAMP pathway in MVA

pathway regulation, but provide a strong rationale for developing statin-dipyridamole therapy

for AML and MM patients.


40

2.3. Results

2.3.1 A screen of pharmacologically active drugs identifies dipyridamole as a potentiator

of the anticancer effects of atorvastatin

To identify a therapeutically effective combination of drugs with novel anticancer

activities using an unbiased approach, we screened atorvastatin in combination with a library

of 100 on-and off-patent drugs composed of antimicrobials and metabolic regulators.29 Well

known pharmacokinetic profiles as well as high achievable plasma concentrations were

amongst the characteristics of the drugs in the library. The KMS11 MM cell line was treated

for 72 hours with either a singularly sublethal dose of atorvastatin (0-20% effect on viability),

each of the 100 drugs alone or in combination with atorvastatin . The combination of

dipyridamole, a well-known anti-platelet agent and atorvastatin was found to decrease cell

viability as assessed by MTS (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-

(4-sulfophenyl)-2H-tetrazolium)) activity (Figure 2-1A). Validation in a panel of malignant

AML and MM cell lines showed that dipyridamole was capable of significantly potentiating

the antiproliferative effects of atorvastatin at multiple doses (Figure 2-1B) following 48 hours

of incubation and as measured by cellular metabolic activity, an indicator of cell viability,

assessed by the MTT (3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay.

Dipyridamole alone, used at a concentration of 5 µM, did not have significant effects on cell

viability. A significant increase in the anti-cancer effect of atorvastatin was also observed

when combined with lower micromolar doses (2.5 µM) of dipyridamole in the AML and MM

cell lines (data not shown). Statins, of which there are six approved in North America, are

often used interchangeably. However, structural differences of each statin governing key

facets such as metabolism, drug interactions and lipophilicity, not only impact their

cholesterol-lowering efficacies but also pleiotropic effects. We therefore extended our


41

investigations to include fluvastatin, another lipophilic statin and found that dipyridamole was

also able to potentiate its anti-cancer effects (Supplementary Figure 2-1A).


42


43

Figure 2-1: A drug screen reveals dipyridamole as a potentiator of the anti-

proliferative effects of atorvastatin.

(A) The KMS11 MM cell line was exposed to 100 drugs (at concentrations of 5-50 µM) alone

and in combination with a sublethal dose of atorvastatin (3.5 µM) for 72 hours. Dipyridamole

(open black square) was identified as a compound decreasing MTS activity in combination

with atorvastatin. Results are representative of two independent screens. (B) Dipyridamole

potentiated the cytotoxic effects of atorvastatin at various doses, in AML (OCI-AML2, OCI-

AML3, HL-60) and MM (LP1, KMS11, 8226) cell lines as assessed by MTT assay following

48 hours of compound exposure, either atorvastatin (closed circle) or atorvastatin and 5 µM

dipyridamole (closed square). *p < 0.05 (t-test, unpaired, two-tailed). Data represent the mean

± SD of three independent experiments.


44

2.3.2. The combination of statins and dipyridamole is synergistically antiproliferative

and induces apoptosis in AML and MM cell lines

We next evaluated whether the statin-dipyridamole combination was synergistic in the

panel of AML and MM cell lines. We treated cells with increasing concentrations of statin

and dipyridamole alone and in combination (using a constant ratio at all doses) for 48 hours

and used the MTT assay to assess effects on viability. Synergy at multiple effect levels was

evaluated using the combination index (CI), where CI values of less than 1 indicate synergy30.

The combination of atorvastatin and dipyridamole synergistically decreased cell viability at

multiple effective concentrations as indicated by the CI values at the EC50 (50% effective

concentration), EC25, EC75 and EC90, which were significantly lower than 1 in all AML and

MM cell lines (Figure 2-2).


45


46

Figure 2-2: The statin-dipyridamole combination is synergistic in AML and

MM cell lines.

AML and MM cell lines were exposed to increasing concentrations of atorvastatin,

dipyridamole, or atorvastatin-dipyridamole combination in a fixed ratio and dose-response

curves were generated. Results were analyzed to evaluate synergy using the combination

index (CI). CI < 1 indicates synergy, CI = 1 indicates additivity and CI > 1 indicates

antagonism. Representative CI-fractions effect (fa) plots show the atorvastatin-dipyridamole

combination is synergistic at multiple effective concentrations in (C) AML and (D) MM cell

lines. The EC50 (50% effective concentration) EC25, EC75 and EC90 from at least three

independent experiments are shown for (C) AML (D) MM cell lines. *p < 0.05 (one sample

t-test, comparing EC values to 1.0) Data are the mean ± SD.


47

Similar remarkable effects were observed with the fluvastatin-dipyridamole

combination (Figure 2-3).


48


49

Figure 2-3: Dipyridamole potentiates the anti-proliferative effects of

fluvastatin and the combination is synergistic.

(A) Dipyridamole potentiated the anti-proliferative effects of fluvastatin at various doses, in

AML (OCI-AML2, OCI-AML3, HL-60) and MM (LP1, KMS11, 8226) cell lines as assessed

by MTT assay following 48 hours of compound exposure. *p < 0.05 (t-test, unpaired, two-

tailed). (B) AML and (C) MM cell lines were exposed to increasing concentrations of

fluvastatin, dipyridamole, or fluvastatin- dipyridamole combination in a fixed ratio and dose-

response curves were generated. Results were analyzed to evaluate synergy using the

combination index (CI). CI < 1 indicates synergy, CI = 1 indicates additivity and CI > 1

indicates antagonism. The EC50 (50% effective concentration) EC25, EC75 and EC90 from

at least three independent experiments are shown. *p < 0.05 (one sample t-test, comparing

EC values to 1.0) Data represent the mean ± SD.


50

The limitation of ectrophotometric cell viability assays such as the MTT and MTS

assays is reliance on mitochondrial enzymes whose rates of conversion of the MTT and MTS

substrates do not directly assess apoptosis31. We therefore chose representative cell lines from

the AML and MM panel, OCI-AML3 and KMS11, respectively, and assessed for apoptosis

using TUNEL (Terminal deoxynucleotidyl transferase dUTP nick end labeling) and PARP

(Poly (adenosinediphosphate-ribose) polymerase) cleavage. We treated OCI-AML3 (Figure

2-4A, left panel) and KMS11 (Figure 2-4B, left panel) cells with atorvastatin and

dipyridamole and found that there is a dramatic induction of apoptosis when atorvastatin is

combined with dipyridamole with no effect of either drug alone. The apoptotic effect was

abrogated with the co-administration of mevalonate (MVA) and therefore was deemed to

result specifically from the inhibition of HMGCR, the target of statins. The degree of

apoptosis was proportional to the concentration of dipyridamole with a 1 µM dose of

dipyridamole resulting in significant induction of apoptosis when combined with statins in

OCI-AML3 cells. Cleavage of PARP also occurred in OCI-AML3 and KMS11 cells (Figure

2-4A-B, right panels respectively) following exposure to the atorvastatin-dipyridamole

combination for 48 hours. Thus, the statin-dipyridamole combination is synergistic and

induces apoptosis in AML and MM cells.


51


52

Figure 2-4: The statin-dipyridamole combination induces apoptosis in AML

and MM cell lines.

Treatment of the OCI-AML3 (A) or the KMS11 (B) cell lines with sublethal low micromolar

doses of atorvastatin (Ator.) and dipyridamole (DP) induces apoptosis following 48 hours of

compound exposure as assessed by TUNEL (left panels). The apoptosis was abrogated in the

presence of exogenous mevalonate (MVA). The atorvastatin-dipyridamole combination also

caused PARP cleavage in both cell lines (right panels). *p < 0.05 (one-way ANOVA with a

Tukey post test, the statin-dipyridamole group being significantly different than all other

groups). TUNEL data represent the mean ± SD of at least three independent experiments. All

immunoblots are representative of a minimum of three independent experiments.


53

2.3. 3. The combination of statins and dipyridamole induces apoptosis in primary AML

and MM cells

To determine whether primary AML and MM cells are sensitive to the statin-

dipyridamole combination, we exposed patient samples to statins and dipyridamole for 48

hours. In AML samples, cell death was measured using annexin V/ propidium iodide (AV/PI)

staining and then summing the AV+/PI- and AV+/PI+ cell populations. Alternative to

TUNEL and PARP cleavage, the Annexin V/PI stain requires fewer cells and is suitable for

primary samples. The assays are comparable and interchangeable as similar levels of death

were measured using TUNEL (Figure 2-4A) and AV/PI (Figure 2-5) in OCI-AML3 cells.


54

Figure 2-5: The statin-dipyridamole combination induces apoptosis in the

OCI-AML3 cell line.

OCI-AML3 cells were treated for 48 hrs with vehicle, 2 µM atorvastatin, 5 µM dipyridamole

or the atorvastatin-dipyridamole combination and cells were assessed for Annexin V (AV)-

propidium iodide (PI) stating flow cytometry. Cell death following atorvastatin-dipyridamole

treatment was comparable to TUNEL (Figure 3A, left panel). *p < 0.05 (one-way ANOVA

with a Tukey post test, the statin- dipyridamole group being significantly different than all

other groups). Data represent mean ± SD of four independent experiments.


55

Primary cells from AML patients with intermediate-risk cytogenetics (n = 2) and

adverse cytogenetics (n = 3) were treated with ranges of doses of statins and dipyridamole

(Table 1).


56

Table 2-1: The statin dipyridamole combination induces apoptosis in

primary AML cells


57

The statin-dipyridamole combination significantly induced apoptosis in primary AML

cells (Figure 2-6B), and a representative dot blot is shown in Figure 2-6A. Death was

dipyridamole and statin dose-dependent and observed at similar doses used in cell lines. The

effects of the combination were minimal in PBSCs (peripheral blood stem cells) obtained

from GCSF (Granulocyte colony-stimulating factor)-mobilized peripheral blood to be utilized

for allotransplantation (Figure 2-6C). The PBSCs were treated at two-times higher statin

micromolar doses than the primary AML cells to thoroughly evaluate potential cytotoxicity to

non-transformed cells. Even at these higher doses, cytotoxicity was not significant. Primary

MM samples were similarly stained with AV to determine apoptosis induced by the statin-

dipyridamole combination and CD138 (cluster of differentiation 138) to identify MM cells.

As shown in Figure 2-6D, patient 1 was sensitive to statin-dipyridamole induced apoptosis as

assessed by the decrease in the viable upper left quadrant CD138+/AV- cell population

without a concomitant decrease in the CD138-/AV- normal population. Upon undergoing

apoptosis, MM cells rapidly lose CD138 (syndecan-1)32, and therefore, apoptotic MM cells

can possibly be found in both CD138+/AV+ and CD138-/AV+ quadrants. Therefore, when

examining effects of drugs on the viability of MM cells, changes in the CD138+/AV-

population were considered. MM patient 2 was not sensitive to the statin-dipyridamole

combination and this is consistent with our previous results showing that not all MM patients

were responsive to statins9. Taken together, these data underscore the therapeutic utility of the

statin-dipyridamole combination in AML and MM patient samples.


58


59

Figure 2-6: The statin-dipyridamole combination induces apoptosis in

primary AML and MM patient cells.

Primary AML patient cells (B, n = 5) and primary MM patient cells (D, n = 2) were treated

for 48 hrs with vehicle, 5 µM fluvastatin, 10 µM dipyridamole and the fluvastatin-

dipyridamole combination. Primary normal hematopoietic cells (PBSCs) (C, n = 4) were

treated for 48 hrs with vehicle, 10 µM atorvastatin, 10 µM dipyridamole or the atorvastatin-

dipyridamole combination. Primary AML and PBSCs were assessed for Annexin V (AV)-

propidium iodide (PI) staining by flow cytometry. MM cells were assessed for AV and

CD138 staining by flow cytometry. A representative primary AML patient sample is shown in

(A). For PBSCs and AML cells, percent apoptosis was evaluated by summing the AV+/PI-

and AV+/PI+ quadrants. For MM primary samples, a decrease in the viable CD138 positive

cells marks apoptosis. *p < 0.05 (one-way ANOVA with a Tukey post test, the statin-

dipyridamole group being significantly different than all other groups).


60

2.3.4. The combination of statins and dipyridamole delays tumour growth in leukemia

xenografts

To evaluate the statin-dipyridamole combination in vivo, we conducted a treatment

trial in SCID (Severe combined immunodeficiency) mice harboring established xenografts of

OCI-AML2 cells. We chose to orally administer atorvastatin because this is the route of

delivery for humans, a longer serum half-life is evident orally compared to other routes of

delivery33 and previously demonstrated in vivo efficacy9. Although oral administration of

dipyridamole, like of atorvastatin is the conventional route of delivery in patients, oral gavage

of dipyridamole resulted in significantly lower dipyridamole serum concentrations when

compared to intraperitoneal (i.p.) administration (Figure 2-7).


61

Figure 2-7: Dipyridamole plasma concentrations are higher following i.p.

administration than p.o.

Higher dipyridamole plasma concentrations were achieved with i.p. administration as

compared to oral dosing three hours post administration. *p < 0.05 (t-test, unpaired, two-

tailed).


62

As differences in gastric pH levels between mice and humans have been reported to

impair the oral bioavailability of dipyridamole in mice34, we reasoned that i.p. administration

would be suitable for evaluating the in vivo efficacy of combining statins and dipyridamole.

The dipyridamole concentration in serum of mice treated with dipyridamole reached the

micromolar concentration range (Figure 2-8A) three hours post administration and was

comparable to the doses used in our cell culture studies. Mice were injected with OCI-AML2

cells subcutaneously and when xenografts were palpable, mice were orally treated daily with

vehicle alone, 50 mg/kg atorvastatin, 120 mg/kg dipyridamole, or the combination of

atorvastatin and dipyridamole. The combination of atorvastatin and dipyridamole significantly

decreased tumour volume (Figure 2-8B) and the tumour weight at the time of sacrifice (Figure

2-8C).


63


64

Figure 2-8: The statin-dipyridamole combination delays tumour growth in

leukemia xenografts.

(A) Plasma concentrations of dipyridamole 3 hours post-administration of 120 mg/kg

dipyridamole and vehicle intraperitoneally (i.p.) (B) SCID mice were injected subcutaneously

with 106 OCI-AML2 cells. After tumours became palpable, mice were randomized into

groups and treated daily with 50 mg/kg atorvastatin orally (p.o.), 120 mg/kg dipyridamole

(i.p.), a combination of dipyridamole and atorvastatin or PBS (phosphate buffered saline) (p.o.

= per os) and vehicle (i.p.) (Con.). Tumour volume was measured every two days. After 14

days of treatment, mice were sacrificed and tumours were resected and weighed (B). *p <

0.05 (one-way ANOVA (Analysis of variance) with a Tukey post test, the statin-dipyridamole

group being significantly different than all other groups. For tumour weights the statin-

dipyridamole group was significantly different than the PBS and the atorvastatin groups. (C)

Plasma concentrations of dipyridamole 3 hours post-administration as administered in (A).

Data represent the mean ± SD, and are representative of two independent in vivo experiments,

both showing similar results.


65

2.3.5. Mediators of cAMP signaling in combination with statins induce apoptosis and

raise intracellular cAMP levels in leukemia cells

The observed anti-cancer effects of statins occur through on-target inhibition of

HMGCR as concomitant addition of MVA abrogated apoptosis (Figure 2-4). Identifying the

mechanism of how dipyridamole contributes to the observed synergy is more difficult as

dipyridamole is known to elicit many molecular effects. Our strategy to uncouple which of

these effects contributed to the observed synergy was to systematically evaluate

pharmacological inhibitors of each known activity of dipyridamole that could plausibly

contribute to the anti-tumour effects for their ability to phenocopy dipyridamole in

potentiating statin induced anti-cancer effects in AML and MM cells.

Dipyridamole is a reported P-glycoprotein (P-gp) inhibitor35. P-gp is an ATP (adenosine

triphosphate) -binding cassette transporter; its overexpression in cancer cells can contribute to

efflux of chemotherapeutic drugs leading to treatment resistance. P-gp modulatory

interactions of statins either as a substrate or inhibitor have consistently been reported for

lovastatin33, 36, and the majority of evidence shows atorvastatin and fluvastatin are not

significantly transported by transported by P-gp36. We first determined whether dipyridamole

modulating P-gp could be contributing to the observed statin-dipyridamole synergistic effect.

We used a system of MM 8226 paired cells lines, one parental and one over-expressing P-gp

(Figure 2-9A). The P-gp over-expression was induced by long-term exposure to doxorubicin a

common P-gp substrate, leading to these 8226 cells (8226DOX) being resistant to doxorubicin.

Dose-response curves were obtained by treating the parental 8226 and 8226DOX cells to

increasing concentrations of doxorubicin with and without 5 µM dipyridamole. Dipyridamole

did not reverse doxorubicin resistance in 8226DOX cells (Figure 2-9B). The doxorubicin IC50


66

(inhibitory concentration 50) values in the 8226DOX cells were still in the high micromolar

range upon addition of dipyridamole compared to the parental 8226 cells in which the

doxorubicin IC50 values were in the nanomolar range. There was a slight decrease in IC50

values upon addition of dipyridamole, however this was evident in both parental and 8226DOX

cells. Thus, evidence shows that dipyridamole inhibition of P-gp is unlikely involved in the

synergistic efficacy of the statin-dipyridamole combination.


67

Figure 2-9: The relative sensitivity to doxorubicin in 8226DOX and 8226

parental cells is not significantly altered in response to dipyridamole

exposure

(A) P-gp expression histogram of parental 8226 cells and P-gp overexpressing doxorubicin

resistant 8226 cells (8226DOX). Histogram is representative of three independent experiments.

(B) Parental 8226 and 8226DOX were treated with increasing concentrations of doxorubicin for

48 hours generating dose-response curves from which IC50 values were computed as

described in methods. The addition of dipyridamole (5 µM) to the 8226DOX cells did not

reverse doxorubicin resistance. *p < 0.05 (t-test, unpaired, two-tailed). Data represent the

mean ± SD of three independent experiments.


68

We further investigated dipyridamole’s other pharmacological activities to determine

whether they could contribute to the anti-tumour effects observed in combination with statins

(Figure 2-10A). Dipyridamole inhibits the equilibrative nucleoside transporter (ENT1) which

transports nucleosides such as adenosine into cells37. In addition, inhibition of glucose uptake

through glucose transporters (GLUT1, GLUT4) is another reported dipyridamole activity38.

Also, dipyridamole is a multi-isoform phosphodiesterase (PDE) inhibitor with varying

degrees of inhibition reported for PDE 5, 6, 7, 8, 10, 1139 and an ability to increase both

cAMP and cGMP (cyclic guanosine monophosphate) levels in cell culture and in vivo40. To

investigate which, if any, of the above reported effects of dipyridamole were responsible for

the potentiation of the anti-tumour effects of statins, we chose representative compounds of

each of these pathways, and then evaluated response of AML and MM cells upon exposure to

sublethal doses of these compounds alone and in combination with statins.

There were no significant effects when statins were combined with the glucose

transport inhibitor fasentin41 (Figure 2-10B) or the equilibrative nucleoside transport inhibitor

NMBPR (Figure 2-10C). Similarly, the cGMP substrate-specific PDE inhibitor 4-{[3′,4′-

(Methylenedioxy)benzyl]amino}-6-methoxyquinazoline (4-MD)42 also failed to potentiate

statin induced anti-proliferative effects (Figure 2-10D). Interestingly, the cAMP substrate-

specific PDE3 inhibitor cilostazol43, another anti-platelet drug currently in clinical use for the

treatment of intermittent claudication did promote statin-induced anti-proliferative effects in

AML and MM cells (Figure 2-10E).


69


70

Figure 2-10: Cilostazol, a PDE3 inhibitor, potentiates the anti-proliferative

activity of statins.

(A) A schematic displaying some of dipyridamole’s known pharmacological effects (black)

and representative drugs (red) of each class. The glucose transport inhibitor fasentin (at

concentrations of 12.5 µM, 6.3 µM and 12.5 µM in OCI-AML-2, OCI-AML-3 and KMS11

cells respectively) (B), equilibrative nucleoside transport inhibitor NMBPR (at concentrations

of 20 µM for all cell lines) (C) and the PDE5-specific inhibitor 4-MD (at concentrations of 10

µM for all cell lines) (D), did not potentiate the anti-proliferative effects of atorvastatin (at

concentrations of 4 µM, 2 µM and 4 µM in OCI-AML2, OCI-AML3 and KMS11 cells

respectively) following 48 hrs of treatment while cilostazol (at concentrations of 25 µM, 12.5

µM and 25 µM in OCI-AML2, OCI-AML3 and KMS11 respectively) (E), a cAMP elevating

PDE3 inhibitor did potentiate atorvastatin-mediated anti-proliferative effects. *p < 0.05 (one-

way ANOVA with a Tukey post test, the statin-dipyridamole group being significantly

different than all other groups) Data represent the mean ± SD of at least three independent

experiments.


71

To evaluate dipyridamole’s functionality as a PDE inhibitor, we tested whether

dipyridamole could raise intracellular cAMP levels. We measured intracellular cAMP levels

and detected a three-fold induction in AML cells following exposure to dipyridamole (Figure

2-11A). Statin and dipyridamole co-treatment also resulted in raised cAMP levels in AML

cells similar to that of singular dipyridamole treatment (Figure 2-11A). To ensure that the

observations extended to other statins, we assayed fluvastatin in subsequent experiments. We

next tested whether this observation was limited to the single agent cilostazol or could be

broadened to include other modulators of the cAMP signaling arm. We therefore combined

sublethal doses of forskolin, an adenylate cyclase activator which is known to raise

intracellular cAMP levels, as well as a cell permeable analog of cAMP, dibutyryl cAMP (db-

cAMP), with fluvastatin and assessed apoptosis using AV/PI staining after treatment for 48

hours. When combined with fluvastatin, all three modulators of the cAMP signaling arm,

cilostazol, forskolin and db-cAMP, resulted in a dramatic increase in apoptosis in AML cells

similar to that observed with the statin-dipyridamole combination (Figure 2-11B). Indeed, this

greater apoptotic effect of statins in combination with cilostazol, forskolin and db-cAMP was

also observed in primary AML patient cells (Figure 2-11C). These data suggest that the

potentiation of statin-induced apoptosis by dipyridamole can be mediated through the

modulation of cAMP signaling as three different agents used at concentrations known to raise

cAMP levels, when combined with statins, also induced apoptosis in AML cells and primary

AML patient cells. This observation is further strengthened by the ability of dipyridamole to

raise cAMP levels in AML cells.


72


73

Figure 2-11: Modulators of cAMP induce apoptosis and increase

intracellular cAMP levels in AML cells.

(A) Intracellular cAMP levels were raised in OCI-AML3 cells following 15 min treatment

with 5 µM dipyridamole, and the fluvastatin (2 µM)-diyridamole combination relative to

control. The PDE3 inhibitor cilostazol (20 µM), the adenylate cyclase activator forskolin (10

µM), and a cell permeable analog of cAMP, dibutyryl cAMP (0.1 mM, db-cAMP) in

combination with fluvastatin (2 µM, 4 µM and 5 µM for OCI-AML3, OCI-AML2 and

primary cells respectively) induced apoptosis in AML cells (B) and primary patient samples

(C). *p < 0.05 (one-way ANOVA with a Tukey post test, the statin-dipyridamole group being

significantly different than all other groups) Data represent the mean ± SD of at least three

independent experiments.


74

2.4. Materials and Methods

Cell Culture and Compounds

Human MM cell lines (LP1, KMS11, 8226) were maintained in RPMI (Roswell Park

Memorial Institute) 1640 medium. Human AML cell lines (OCI-AML3, HL-60 andOCI-

AML2) were maintained in alpha modified Eagle's medium (αMEM) and Iscove modified

Dulbecco medium (IMDM) respectively. 8226DOX cells were additionally cultured in 0.3 µM

doxorubicin. Media was supplemented with 10% fetal bovine serum (FBS, GIBCO) and

penicillin-streptomycin. Cells were incubated at 37°C in 5% CO2 and all cell lines were

routinely confirmed to be mycoplasma-free (MycoAlert mycoplasma detection kit, Lonza).

Atorvastatin calcium (21 CEC Pharmaceuticals LTD) and fluvastatin (US Biologicals) were

dissolved in ethanol. Dipyridamole (Sigma), Cilostazol (Tocris Bioscience), 6-S-heacep22-6-

thioinosine (NBMPR, Tocris Bioscience), 4-{[3′,4′-(Methylenedioxy)benzyl]amino}-6-

methoxyquinazoline (4-MD) (Calbiochem), Fasentin (Sigma) were dissolved in dimethyl

sulfoxide (DMSO).

Primary Cells

Primary AML and MM patient samples were obtained from consenting AML and MM

patients respectively. Peripheral blood mononuclear cells (PBSCs) were obtained from

healthy volunteers donating cells for allotransplantation. The PBSCs were granulocyte

colony-stimulating factor (GCSF)-mobilized. Mononuclear cells were fractioned by Ficoll-

Hypaque gradient sedimentation. AML and PBSC primary cells were cultured in IMDM

medium supplemented with 20% FBS and 5% 5367-conditioned medium. MM primary cells

were cultured in IMDM medium supplemented with 10% fetal bovine serum, 1% glutamine,

and penicillin-streptomycin. Primary AML cells were obtained from frozen stocks stored in


75

liquid nitrogen, thawed, and within 2-10 hours, treated for 48 hours. PBSC and primary MM

cells were obtained fresh and treated as indicated and as previously reported9. Use and

collection of human tissue for this study was approved by the University Health Network

Institutional Review Board (Toronto, ON).

Chemical Screen for Cytotoxic Drugs

96 well non-tissue culture treated plates (BD Falcon) of KMS11 cells (20000 cells/well) were

treated with aliquots of a chemical library29 of 100 drugs dissolved in DMSO (3-50 µM) using

a Biomek FX Laboratory Automated Workstation (Beckman Coulter). One plate had been

pre-treated with 3.5 uM of atorvastatin. Following 72 hours of incubation, cell viability was

assessed using the (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-

sulfophenyl)-2H-tetrazolium) (MTS) reduction assay (Promega) as per the manufacturer’s

protocol and as previously described29.

MTT, TUNEL and Annexin V apoptosis assays

(3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT, Sigma-Aldrich) assays

were carried out as previously described44. Briefly, 2 x 105 to 3 x 105 cells/ml (MM and AML

cell lines) were plated in 96 well plates and after 24 hours, treated with the indicated

compounds and concentration ranges for 48 hours. Half-maximal inhibitory concentrations

(IC50) values were computed from dose-response curves using Prism (v5.0, GraphPad

Software). For TUNEL assays, 2.5 x 105 cells/ml were seeded in 6 well plates and treated for

48 hours as indicated. Cells were fixed in ethanol and staining was performed using terminal

deoxynucleotide transferase dUTP nick end labeling (TUNEL) according to the

manufacturer’s instructions (APO-BRDU Apoptosis Kit, Phoenix Flow Systems). Annexin V

(AV) apoptosis assays (Biovision) were carried out as per the manufacturer’s protocol.


76

Briefly, 2.5 x 105 cells/ml (cell lines) and 5 x 105 cells/ml (primary cells) were treated for 48

hours and stained with AV-fluorescein isothiocyanate–conjugated annexin (FITC) and

propidium iodide (PI). Primary MM cells were labeled with anti-CD138-phycoerythrin

(Immunotech) and AV-FITC (R&D Systems). Cells were analyzed for apoptosis by flow

cytometry; ten thousand events were acquired using a FACSCalibur cytometer (BD

Biosciences). CD138-positive and AV negative cells constitute viable myeloma cells and

apoptotic myeloma cells fall within the CD138-negative AV-positive population.

Drug Combination Studies

Synergy between statins and dipyridamole was evaluated using the combination index (CI)30.

Dose response curves were generated for statins and dipyridamole alone and in combination

at a constant ratio following compound exposure for 48 hours. Viability was assessed by

MTT assay. Calcusyn software (biosoft) was used to evaluate synergy using the median effect

model.

Immunoblotting

2.5 x 105 cells/ml were seeded in 6 well tissue-culture plates and treated as indicated for 48

hours. For PARP detection, cells were washed with PBS and lysed using boiling hot SDS

(sodium dodecyl sulfate polyacrylamide) lysis buffer (1.1% SDS, 11% glycerol, 0.1M Tris pH

6.8) with 10% β-mercaptoethanol. Blots were probed with anti-tubulin (Santa Cruz

Biotechnology) and anti-PARP (Cell Signaling).

Intracellular cAMP assays

Intracellular levels of cAMP were measured using the Cyclic AMP Chemiluminescent

Immunoassay Kit (Cell Technology, INC) as per the manufacturer’s protocol. Briefly, 1.5 x


77

106 cells per well of a 6 well plate were incubated with compounds as indicated, washed with

PBS and lysed using 150 µL (microliter) of the provided lysis buffer. Samples were stored at -

80 °C if not immediately used.

Leukemia xenograft models

Severe Combined Immunodeficiency (SCID) male mice (7-9 week old) were subcutaneously

injected with 106 OCI-AML2 cells. When tumours became palpable (15 mm3), mice were

randomized and treated daily with 120 mg/kg dipyridamole administered intraperitoneally

(i.p.) (from a solution of 5 mg/ml dipyridamole, 50mg/ml polyethylene glycol 600, and 2

mg/ml tartaric acid), 50 mg/kg atorvastatin administered by oral gavage (p.o.), a combination

of dipyridamole and atorvastatin, or vehicle (PBS orally and vehicle i.p.). Tumours were

measured every two days using digital calipers and tumour volume was calculated using the

following formula: (tumour length x width2)/2. Animal work was carried out with the

approval of the Princess Margaret Hospital ethics review board in accordance to the

regulations of the Canadian Council on Animal Care.

Assessment of dipyridamole levels in serum

Levels of dipyridamole in the serum of mice were determined as previously described by

spectrofluorometry using differences in the fluorescence of dipyridamole between acidic and

basic conditions 45. Briefly 52 µL of serum from one mouse was equally divided into two

solutions containing glycine buffer (186 µL, 67.8 mM) and ethanol (37.5 µL) pH 2.6 and pH

9.8 respectively. For standard curves, accompanying dipyridamole control solutions dissolved

in ethanol were prepared using serum from untreated mice. Fluorescence spectra (450-600 nm,

maximum fluorescence occurring at 490 nM, excitation at 420 nM) were measured using a

SpectraMax M5 plate reader (Molecular Devices).


78

P-glycoprotein (P-gp) cell surface expression

Briefly, 1.0 x 106 cells/ml were stained with 10 µL of FITC-tagged anti-P-gp antibody (BD

Biosciences) and expression was detected by flow cytometry using a FACSCalibur cytometer

(BD Biosciences).


79

2.5. Discussion

From a screen of FDA approved drugs, we identified an effective statin-dipyridamole

combination as having previously unrecognized anti-cancer activity against AML and MM

cells. The combination was synergistic and significantly induced apoptosis not only in cell

lines but also in primary myeloma and leukemia patient samples. A leukemia xenograft model

confirmed the in vivo efficacy of the combination in its ability to slow tumour growth

following treatment of pre-established tumours. The mechanism by which dipyridamole

potentiated statin-induced apoptosis was explored by combining pharmacological inhibitors

representative of each of the previously reported molecular effects of dipyridamole with

statins. Using this approach, we eliminated the possibility that the synergy was mediated

through dipyridamole’s inhibitory effects on P-glycoprotein, equilibrative nucleoside

transporters and glucose transporters. Cilostazol, a cAMP specific PDE inhibitor was able to

potentiate statin-induced apoptosis suggesting that dipyridamole’s PDE inhibitory activities

could be responsible for the observed synergy. Indeed, dipyridamole was able to raise

intracellular levels of cAMP in AML cells and other modulators of cAMP signaling were also

able to induce apoptosis when combined with statins in AML cells and primary patient cells.

Taken together, we identify a novel combination of two FDA-approved agents for the

treatment of hematological cancers and identify the critical mechanism of action.

Statins are well tolerated in the clinical cancer setting including hematological malignancies.

In dose-escalating studies where cancer patients were administered lovastatin doses of up to

45 mg/kg/day, average low micromolar peak plasma bioactivity statin levels were achieved 21.

Atorvastatin and fluvastatin, used in the current study, have pharmacokinetic properties that

are systemically more favorable than lovastatin, including a longer half-life and higher peak

concentration (Cmax), respectively. The cholesterol lowering 40 mg oral dose of fluvastatin


80

has a Cmax 200-440 ng/ml33, which is equivalent to 0.49-1.0 µM. Although the administration

of higher than cholesterol-lowering fluvastatin or atorvastatin doses have not yet been

evaluated in cancer patients, the tolerability observed with other statins suggests that elevated

doses of fluvastatin and atorvastatin will be similarly tolerated and that low micromolar range

(2–5 µM) doses used in our cell culture studies are likely readily achievable in humans.

Importantly, the anti-proliferative activity of statins is dose and time dependent, and evidence

shows that even cholesterol lowering-doses can decrease tumor burden in cancer patients26,25.

Thus, the optimal dose of statins to use for cancer patient treatment remains unclear, yet

evidence strongly suggests effective dosing can be achieved in vivo.

Like all anti-cancer agents, the optimal treatment strategy is to administer in multimodal and

combinatorial treatment strategies to increase tumor-specific anti-proliferative activities.

Statins can be successfully combined preclinically with number of diverse agents46. The

addition of statins to the standard of care regimen in the clinical cancer setting has also been

used to successfully treat some AML24 and MM27 patients. Here, we provide a complimentary

approach of combining statins with an already FDA-approved agent in the treatment of

hematological malignancies.

Dipyridamole has been used as part of antithrombotic therapy for decades and its

pharmacology has been thoroughly investigated. Recently, the extended-release formulation

of dipyridamole in combination with aspirin has been widely accepted for use in the

prevention of cerebral ischemic events in patients with stroke. Dipyridamole is constantly

being re-formulated to maximize systemic exposure and extended release formulations of

dipyridamole have a reported half-life of 13.6 hours following typical 200 mg twice daily

(b.i.d.) dosing with steady state peak plasma concentrations of 1.0-4.0 µg/ml (2.0 – 7.9 µM)47.

Dipyridamole is tightly protein bound to α1-acid-glycoprotein (AGP) in serum 48, however

this does not preclude dipyridamole activity as an antithrombotic, suggesting that the steady


81

state micromolar-range concentrations of dipyridamole achieved with antithrombotic doses of

200 mg b.i.d. (bis in die) will be sufficient for effective anti-cancer therapy when combined

with statins. Interestingly, much higher dipyridamole doses have been tolerated in humans as

reported from overdose case reports49, suggesting dosing could potentially be elevated.

Taken together, evidence suggests the combination of statins and dipyridamole will be a well-

tolerated and efficacious strategy for the clinical treatment of leukemia and myeloma.

Our apoptosis assays in primary cells demonstrated that the combination of statin and

dipyridamole was capable of inducing apoptosis in primary AML and MM patient samples.

Moreover, even when used at elevated concentrations, primary normal PBSCs did not

demonstrate significant toxicity in response to dipyridamole and statin exposure. This further

attests to the potential therapeutic window of the statin-dipyridamole combination. Indeed, the

statin-dipyridamole combination was shown to be safely combined and well-tolerated in

humans in studies of cardiovascular protection50 and renal function51. Therefore given that we

have shown that the statin-dipyridamole combination minimally affects normal hematopoietic

cells and this combination has been safely administered to humans, we predict that a

substantial therapeutic window exists in the treatment of AML and MM patients.

Dipyridamole has been associated with multiple pharmacological actions including the

inhibition of P-gp, nucleoside transport, glucose uptake and PDEs. These attributes have led

to explorations of the potential benefits of dipyridamole in diseases other than cardiovascular

conditions, including cancer. Dipyridamole has been combined with antimetabolite agents

such as 5-fluorouracil (5-FU) in an effort to prevent salvage metabolism through

dipyridamole’s nucleoside transport inhibitory activities. Although addition of dipyridamole

at 300 mg daily to the regimen of 5-FU, leucovirin and mitomycin-C to patients with

advanced pancreatic cancer was well tolerated, these studies lacked encouraging therapeutic

responses when compared to the standard of care52. This is consistent with our data showing


82

the combination of statins with other nucleoside transport inhibitors, or inhibitors of glucose

transport, was ineffective at triggering cell death of AML and MM cells.

Interestingly, the combination of statins and the PDE3 inhibitor cilostazol did have anti-

leukemia and anti-myeloma activity. Cilostazol is a selective PDE3 inhibitor whose reported

effects include cAMP elevation and like dipyridamole, is also currently in clinical use. As the

specific cGMP elevating agent, 4-MD, showed no anti-leukemia and myeloma activity in

combination with statins, we further explored other modulators of cAMP signaling and tested

their potential to induce apoptosis in combination with statins. The well-characterized cAMP

elevating agent forskolin, cilostazol, and the cell permeable cAMP analog db-CAMP all

induced apoptosis in leukemia cells and higher levels of cAMP were observed in cells treated

with the statin-DP combination. There is some evidence to indicate that pro-apoptotic

responses to cAMP signaling occur. In one report, albeit limited to one cell line, IPC-81,

treatment with a cAMP analog not only induced apoptosis as a single agent but, when

combined at sublethal doses with a chemotherapeutic drug daunorubicin, the two agents

synergized in promoting apoptosis53. Another report identified cAMP selective PDE 3, 4 and

7 inhibitors, including cilostazol, in combination with adenosine receptor agonists as having

strongly synergistic anti-proliferative effects in MM cells and increases in intracellular cAMP

levels were also observed with the combination though in vivo preclinical efficacy was not

demonstrated54. Here we show that the inhibition of the MVA pathway using statins and

dipyridamole’s effects on intracellular cAMP levels may contribute to the observed synergy

in inducing tumour cell kill.

Interestingly, the statin-dipyridamole combination effectively reduced proliferation in

MM cell lines previously characterized as being statin-insensitive9. The division of MM cells

into statin-sensitive and statin-insensitive cohorts, has been linked to differences in the sterol-

feedback loop. Whereas in the statin-sensitive KMS11 MM cells the upregulation of sterol


83

responsive genes such as HMGCR in response to statin exposure is limited, the statin-

insensitive LP1 have a robust feedback loop. The anti-myeloma effects of the statin-

dipyridamole combination observed in difficult to treat MM cells further demonstrates its

therapeutic potential.

In summary, we have identified a synergistic combination of statins and dipyridamole that is

preclinically effective in treating AML and MM. Our mechanistic investigations have

uncovered other FDA-approved drugs, such as cilostazol that, in combination with statins,

could also have applications in the hematological cancer therapeutic setting. These studies

may serve as a foundation for developing a phase I clinical trial involving the combination of

statins and dipyridamole for the effective treatment of AML and MM.


84

2.6. References

1. Ludwig H, Durie BG, McCarthy P, Palumbo A, San Miguel J, Barlogie B et al. IMWG consensus on maintenance therapy in multiple myeloma. Blood 2012; 119(13): 3003-15.

2. Ofran Y, Rowe JM. Treatment for relapsed acute myeloid leukemia: what is new?

Current opinion in hematology 2012; 19(2): 89-94. 3. Goldstein JL, Brown MS. Regulation of the mevalonate pathway. Nature 1990;

343(6257): 425-30. 4. Ahern TP, Pedersen L, Tarp M, Cronin-Fenton DP, Garne JP, Silliman RA et al.

Statin prescriptions and breast cancer recurrence risk: a Danish nationwide prospective cohort study. J Natl Cancer Inst 2011; 103(19): 1461-8.



6. Nielsen SF, Nordestgaard BG, Bojesen SE. Statin use and reduced cancer-related

mortality. The New England journal of medicine 2012; 367(19): 1792-802. 7. Dimitroulakos J, Nohynek D, Backway KL, Hedley DW, Yeger H, Freedman MH et


8. Wu J, Wong WW, Khosravi F, Minden MD, Penn LZ. Blocking the Raf/MEK/ERK

pathway sensitizes acute myelogenous leukemia cells to lovastatin-induced apoptosis. Cancer Res 2004; 64(18): 6461-8.



10. Schmidmaier R, Baumann P, Simsek M, Dayyani F, Emmerich B, Meinhardt G. The

HMG-CoA reductase inhibitor simvastatin overcomes cell adhesion-mediated drug resistance in multiple myeloma by geranylgeranylation of Rho protein and activation of Rho kinase. Blood 2004; 104(6): 1825-32.

11. Stirewalt DL, Appelbaum FR, Willman CL, Zager RA, Banker DE. Mevastatin can

increase toxicity in primary AMLs exposed to standard therapeutic agents, but statin efficacy is not simply associated with ras hotspot mutations or overexpression. Leuk Res 2003; 27(2): 133-45.


85

12. Newman A, Clutterbuck RD, Powles RL, Catovsky D, Millar JL. A comparison of the effect of the 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors simvastatin, lovastatin and pravastatin on leukaemic and normal bone marrow progenitors. Leukemia & lymphoma 1997; 24(5-6): 533-7.

13. Clutterbuck RD, Millar BC, Powles RL, Newman A, Catovsky D, Jarman M et al.

Inhibitory effect of simvastatin on the proliferation of human myeloid leukaemia cells in severe combined immunodeficient (SCID) mice. British journal of haematology 1998; 102(2): 522-7.

14. Li HY, Appelbaum FR, Willman CL, Zager RA, Banker DE. Cholesterol-modulating

agents kill acute myeloid leukemia cells and sensitize them to therapeutics by blocking adaptive cholesterol responses. Blood 2003; 101(9): 3628-34.

15. Xia Z, Tan MM, Wong WW, Dimitroulakos J, Minden MD, Penn LZ. Blocking

protein geranylgeranylation is essential for lovastatin-induced apoptosis of human acute myeloid leukemia cells. Leukemia 2001; 15(9): 1398-407.


al. Dysregulation of the mevalonate pathway promotes transformation. Proc Natl Acad Sci U S A; 107(34): 15051-6.

17. Freed-Pastor WA, Mizuno H, Zhao X, Langerod A, Moon SH, Rodriguez-Barrueco R

et al. Mutant p53 disrupts mammary tissue architecture via the mevalonate pathway. Cell 2012; 148(1-2): 244-58.

18. Tatidis L, Gruber A, Vitols S. Decreased feedback regulation of low density

lipoprotein receptor activity by sterols in leukemic cells from patients with acute myelogenous leukemia. Journal of lipid research 1997; 38(12): 2436-45.

19. Vitols S, Norgren S, Juliusson G, Tatidis L, Luthman H. Multilevel regulation of low-

density lipoprotein receptor and 3-hydroxy-3-methylglutaryl coenzyme A reductase gene expression in normal and leukemic cells. Blood 1994; 84(8): 2689-98.



21. Thibault A, Samid D, Tompkins AC, Figg WD, Cooper MR, Hohl RJ et al. Phase I




23. van der Spek E, Bloem AC, van de Donk NW, Bogers LH, van der Griend R, Kramer

MH et al. Dose-finding study of high-dose simvastatin combined with standard


86

chemotherapy in patients with relapsed or refractory myeloma or lymphoma. Haematologica 2006; 91(4): 542-5.



25. Garwood ER, Kumar AS, Baehner FL, Moore DH, Au A, Hylton N et al. Fluvastatin

reduces proliferation and increases apoptosis in women with high grade breast cancer. Breast Cancer Res Treat 2009.



27. Hus M, Grzasko N, Szostek M, Pluta A, Helbig G, Woszczyk D et al. Thalidomide,

dexamethasone and lovastatin with autologous stem cell transplantation as a salvage immunomodulatory therapy in patients with relapsed and refractory multiple myeloma. Annals of hematology 2011; 90(10): 1161-6.

28. van der Spek E, Bloem AC, Sinnige HA, Lokhorst HM. High dose simvastatin does

not reverse resistance to vincristine, adriamycin, and dexamethasone (VAD) in myeloma. Haematologica 2007; 92(12): e130-1.

29. Sharmeen S, Skrtic M, Sukhai MA, Hurren R, Gronda M, Wang X et al. The

antiparasitic agent ivermectin induces chloride-dependent membrane hyperpolarization and cell death in leukemia cells. Blood 2010; 116(18): 3593-603.

30. Chou TC, Talalay P. Quantitative analysis of dose-effect relationships: the combined

effects of multiple drugs or enzyme inhibitors. Adv Enzyme Regul 1984; 22: 27-55. 31. Galluzzi L, Aaronson SA, Abrams J, Alnemri ES, Andrews DW, Baehrecke EH et al.

Guidelines for the use and interpretation of assays for monitoring cell death in higher eukaryotes. Cell death and differentiation 2009; 16(8): 1093-107.

32. Jourdan M, Ferlin M, Legouffe E, Horvathova M, Liautard J, Rossi JF et al. The

myeloma cell antigen syndecan-1 is lost by apoptotic myeloma cells. British journal of haematology 1998; 100(4): 637-46.

33. Shitara Y, Sugiyama Y. Pharmacokinetic and pharmacodynamic alterations of 3-

hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors: drug-drug interactions and interindividual differences in transporter and metabolic enzyme functions. Pharmacol Ther 2006; 112(1): 71-105.

34. Kim HH, Sawada N, Soydan G, Lee HS, Zhou Z, Hwang SK et al. Additive effects of

statin and dipyridamole on cerebral blood flow and stroke protection. J Cereb Blood Flow Metab 2008; 28(7): 1285-93.


87

35. Verstuyft C, Strabach S, El-Morabet H, Kerb R, Brinkmann U, Dubert L et al. Dipyridamole enhances digoxin bioavailability via P-glycoprotein inhibition. Clin Pharmacol Ther 2003; 73(1): 51-60.

36. Goard CA, Mather RG, Vinepal B, Clendening JW, Martirosyan A, Boutros PC et al.

Differential interactions between statins and P-glycoprotein: implications for exploiting statins as anticancer agents. International journal of cancer. Journal international du cancer 2010; 127(12): 2936-48.

37. Hammond JR. Interaction of a series of draflazine analogues with equilibrative

nucleoside transporters: species differences and transporter subtype selectivity. Naunyn-Schmiedeberg's archives of pharmacology 2000; 361(4): 373-82.

38. Hellwig B, Joost HG. Differentiation of erythrocyte-(GLUT1), liver-(GLUT2), and

adipocyte-type (GLUT4) glucose transporters by binding of the inhibitory ligands cytochalasin B, forskolin, dipyridamole, and isobutylmethylxanthine. Molecular pharmacology 1991; 40(3): 383-9.

39. Bender AT, Beavo JA. Cyclic nucleotide phosphodiesterases: molecular regulation to

clinical use. Pharmacological reviews 2006; 58(3): 488-520. 40. Schoeffter P, Lugnier C, Demesy-Waeldele F, Stoclet JC. Role of cyclic AMP- and

cyclic GMP-phosphodiesterases in the control of cyclic nucleotide levels and smooth muscle tone in rat isolated aorta. A study with selective inhibitors. Biochem Pharmacol 1987; 36(22): 3965-72.

41. Wood TE, Dalili S, Simpson CD, Hurren R, Mao X, Saiz FS et al. A novel inhibitor of

glucose uptake sensitizes cells to FAS-induced cell death. Molecular cancer therapeutics 2008; 7(11): 3546-55.

42. Tokumura T, Takase Y, Horie T. Determination of a novel and potent cyclic GMP

phosphodiesterase inhibitor, 4-([3,4-(methylenedioxy)-benzyl]amino)-6,7,8-trimethoxyquinazoline, in dog plasma by high-performance liquid chromatography. Journal of chromatography. B, Biomedical applications 1994; 658(1): 202-6.

43. Schror K. The pharmacology of cilostazol. Diabetes, obesity & metabolism 2002; 4

Suppl 2: S14-9. 44. Dimitroulakos J, Ye LY, Benzaquen M, Moore MJ, Kamel-Reid S, Freedman MH et

al. Differential sensitivity of various pediatric cancers and squamous cell carcinomas to lovastatin-induced apoptosis: therapeutic implications. Clin Cancer Res 2001; 7(1): 158-67.

45. Oshrine B, Malinin A, Pokov A, Dragan A, Hanley D, Serebruany V. Criticality of pH

for accurate fluorometric measurements of dipyridamole levels in biological fluids. Methods Find Exp Clin Pharmacol 2005; 27(2): 95-100.

46. Jakobisiak M, Golab J. Statins can modulate effectiveness of antitumor therapeutic

modalities. Medicinal research reviews 2010; 30(1): 102-35.


88

47. Derendorf H, VanderMaelen CP, Brickl RS, MacGregor TR, Eisert W. Dipyridamole

bioavailability in subjects with reduced gastric acidity. J Clin Pharmacol 2005; 45(7): 845-50.

48. Fischer PH, Willson JK, Risueno C, Tutsch K, Bruggink J, Ranhosky A et al.

Biochemical assessment of the effects of acivicin and dipyridamole given as a continuous 72-hour intravenous infusion. Cancer Res 1988; 48(19): 5591-6.

49. Lagas JS, Wilhelm AJ, Vos RM, van den Dool EJ, van der Heide Y, Huissoon S et al.

Toxicokinetics of a dipyridamole (Persantin) intoxication: case report. Human & experimental toxicology 2011; 30(1): 74-8.

50. Meijer P, Wouters CW, van den Broek PH, Scheffer GJ, Riksen NP, Smits P et al.

Dipyridamole enhances ischaemia-induced reactive hyperaemia by increased adenosine receptor stimulation. Br J Pharmacol 2008; 153(6): 1169-76.

51. Kano K, Nishikura K, Yamada Y, Arisaka O. Effect of fluvastatin and dipyridamole

on proteinuria and renal function in childhood IgA nephropathy with mild histological findings and moderate proteinuria. Clinical nephrology 2003; 60(2): 85-9.

52. Burch PA, Ghosh C, Schroeder G, Allmer C, Woodhouse CL, Goldberg RM et al.

Phase II evaluation of continuous-infusion 5-fluorouracil, leucovorin, mitomycin-C, and oral dipyridamole in advanced measurable pancreatic cancer: a North Central Cancer Treatment Group Trial. Am J Clin Oncol 2000; 23(5): 534-7.

53. Huseby S, Gausdal G, Keen TJ, Kjaerland E, Krakstad C, Myhren L et al. Cyclic

AMP induces IPC leukemia cell apoptosis via CRE-and CDK-dependent Bim transcription. Cell death & disease 2011; 2: e237.

54. Rickles RJ, Pierce LT, Giordano TP, 3rd, Tam WF, McMillin DW, Delmore J et al.

Adenosine A2A receptor agonists and PDE inhibitors: a synergistic multitarget mechanism discovered through systematic combination screening in B-cell malignancies. Blood 2010; 116(4): 593-602.


89

Chapter 3 : Genome wide shRNA screen reveals novel sensitizers

of statin-induced apoptosis

Contributions:

The majority of work presented in this chapter was completed by the author of this thesis.

Additional contributions are as follow:

Screen design, execution and interpretation: Aleksandra Pandyra, Dr. Carolyn A. Goard and

Dr. Elke Ericson.

Genomic DNA extraction, hairpin amplification and array hybridization: Dr. Elke Ericson and

Marinella Gebbia.

All bioinformatics analysis was performed by Dr. Kevin Brown

Figure 3.3. Dr. Carolyn A. Goard and Janice Pong contributed to harvesting and

immunoblotting replicates.

Figure 3.4 A and B. Dr. Carolyn A. Goard and Janice Pong contributed to the generation of

the cells and some replicates of the apoptosis assays.

Figure 3.8 C. Dr. Peter J. Mullen carried out the proliferation assays


90

3.1 Abstract

Statins have been used for decades to treat hypercholesterolemic patients and their

widespread use in the clinic has uncovered some pleiotropic effects including anti-tumour

efficacy. Statins target the rate-limiting enzyme of the mevalonate (MVA) pathway, 3-

hydroxy-3-methylglutaryl coenzyme A reductase (HMGCR), a pathway that supplies the cell

with essential sterol end products such as cholesterol. The MVA pathway is dysregulated in

many cancers and statins are promising novel agents currently being evaluated in the clinic

for efficacy in treating cancer. To uncover novel targets that potenitate statin induced

apotosis, we carried out an unbiased genome wide shRNA screen in the A549 lung cancer cell

line stably transduced with the RNAi Consortium (TRC1) shRNA library. A549 cells were

exposed to sublethal doses of fluvastatin over 12 days and significant shRNAs under

represented in the fluvastatin treated group relative to control yielded a list of putative

potentiators of statin induced anti cancer-effects. Among 151 hits, the sterol regulatory

element binding transcription factor 2 (SREBF2) was validated and found to dramatically

sensitize A549, MDAMB231 and MCF7 breast cancer cells to statin-induced apoptosis.

Stable SREBF2 knockdown in lung and breast cancer cells completely abrogated the statin-

induced upregulation of sterol responsive genes HMGCR and 3-hydroxy-3-methylglutaryl-

CoA synthase 1 (HMGCS1). Subcutaneous xenografts of stably expressing SREBF2 sublines

in NOD SCID mice were more sensitive to oral fluvastatin treatment than the control sublines.

Taken together, we have identified and validated an important novel potentiator of statin

induced apoptosis in lung and breast cancer cells.


91

3.2 Introduction

Statins, potent inhibitors of the rate-limiting enzyme in the mevalonate (MVA)

pathway, 3-hydroxy-3-methylglutaryl coenzyme A reductase (HMGCR)1 have been used for

decades in the treatment of patients with hypercholesterolemia. Their wide spread use for the

prevention of adverse cardiovascular events has led to the accumulation of retrospective

epidemiological evidence that indicates statin use can reduce cancer incidence 2-5. Beyond

lowering cholesterol, statins exert pleiotropic effects including direct anti-proliferative and

pro-apoptotic effects on tumour cells.

The anti-cancer effects of statins have been attributed to direct target HMGCR in

tumour cells leading to depletion of fundamental MVA-derived end-products such as

isoprenoids and cholesterol6-8. The consequential downstream effects of this depletion affect

oncogenic pathways such as the MAPK/Erk (Mitogen-activated protein kinases/Extracellular

signal-regulated kinases)9, 10, JNK (c-Jun N-terminal kinases) 11, 12 pathways and the

epidermal growth factor receptor (EGFR)13. Statin induced anti-proliferative effects and

apoptosis have been shown to be mainly p53 independent11, 14, 15 and abrogated by the extopic

expression of BLC216, 17. Statins are currently being evaluated as anti-cancer agents in

patients with a broad range of tumour types. Data from completed clinical trials data

indicates that statins are well tolerated in the clinical cancer care setting at cholesterol

lowering18, 19 and higher doses20-22, alone or when co-administered with the standard of care23-

26. Taken together, preclinical, epidemiological and clinical evidence has demonstrated that

repurposing statins for the treatment of cancer has potential clinical utility.

To further extend the benefit of statins and maximize anti-tumour efficacy, we sought

to identify novel targets capable of potentiating statin-induced anti-cancer effects through a

genome wide shRNA screen. The advent of RNA interference (RNAi) technology27 has


92

revolutionized the experimental landscape and consequently augmented our understanding of

cancer genetics. Knockdown approaches, encompass small interfering RNAs (siRNAs)28,

endoribonuclease-prepared siRNAs (esiRNA)29 and short hairpin RNAs (shRNA)30, coupled

with completely sequenced human and mouse genomes, have enabled the creation of multiple

RNAi libraries which are of great utility in functional screens. The advantages of using a

functional screen to identify products of genes which regulate a particular biological activity

following a perturbation lie in its unbiased and genome-wide global nature31. Such screens

have been successfully used to identify essential genes in tumour cell lines32, oncogene-

associated synthetic lethal interactions33, determinants of response to therapies34, 35 and

sensitizers to anti-cancer drugs36. Although all RNAi methods can be used for transient short-

term knockdown approaches, shRNA libraries when combined with viral mediated

transduction integrate into the genome of a target cell and are advantageous for long-term

experiments particularly when employing nontransfectable cells.

We conducted an shRNA screen using the A549 non-small-cell lung cancer (NSCLC)

stably transduced with the RNAi Consortium (TRC1) shRNA library (shA549) and treated the

cells with sublethal doses of fluvastatin for twelve days. Significantly under-represented

shRNAs in the fuvastatin-treated group were identified as potential gene targets whose

knockdown enhanced the anti-proliferative effects of fluvastatin. Subsequent validation of

four gene hits demonstrated that knockdown of geranylgeranyl diphosphate synthase 1

(GGPS1), sterol regulatory element-binding protein 2 (SREBF2), mitogen-activated protein

kinase kinase 4 (MAP2K4) and phosphatidylinositol 4-kinase, catalytic beta (PIK4B) in

combination with fluvastatin possessed pro-apoptotic and anti-proliferative effects. Extending

the validation to breast cancer, revealed that, SREBF2 and PI4KB dramatically enhanced

fluvastatin’s pro-apoptotic and anti-proliferative effects in MCF7 and MDAMB 231 breast

cancer cells, effects which were corroborated with a PI4KB pharmacological inhibitor.


93

Expression analysis in the lung and breast cells stably expressing shRNAs targeting SREBF2

revealed that these sublines lacked the ability to upregulate HMGCR in response to

fluvastatin exposure, a phenomenon likely responsible for their remarkable fluvastatin-

sensitivity. Therefore concomitant inhibition of the MVA pathway through targeting

HMGCR using statins and the transcription factor SREBF2 is an effective therapeutic strategy

that induces apoptosis in lung and breast cancer cells.

Using a genome wide shRNA screen we have identified novel targets whose inhibition,

combined with fluvastatin, possesses anti-tumour therapeutic efficacy in lung and breast

cancer cells. The development of novel therapeutic cocktails in the treatment of these tumours

is imperative especially in patients unsuccessfully treated with several front-line treatments.


94

3.3 Results

To identify novel sensitizers of statin-induced apoptosis, we designed a genome-wide

shRNA screen in which some shRNA-stably expressing cells when combined with sublethal

doses of fluvastatin, were outcompeted over time. The genes targeted by these dropout

shRNAs were identified as potenitiators of statin-induced anti-cancer effects and further

validated in attempts to identify novel targets and or pathways which will increase statin anti-

tumour efficacy.

3.2.1 A genome-wide screen identifies shRNA drop-outs after fluvastatin exposure

The A549 adenocarcinoma lung cancer cell line, stably transduced with the RNAi

Consortium (TRC1) shRNA library (shA549), was utilized for screen execution. The TRC1

library is composed of lentiviral shRNAs targeting nearly 16,000 human genes. Five or four

hairpins exist per gene and nearly 7,000 genes have been validated. The A549 cells are

characterized by elevated HMGCR activity37. Elevated HMGCR activity relative to normal

cells not only establishes an active target but is indicative of a potential therapeutic index

making the A549 cells a suitable choice for the screen. Furthermore, the A549 cells are

relatively statin insensitive and uncovering potentiators of statin anti-cancer activity will

broaden the applicability of statins to tumours that would otherwise remain unresponsive to

statins. shA549 cells were exposed to sublethal doses of fluvastatin over twelve days, with

cell-splitting, harvesting cells for genomic DNA and re-seeding of remaining cells occurring

every 3 days (Figure 3-1 A). Cell viability, relative to ethanol, was maintained constant

throughout the duration of the experiment at 60-70% (Figure 3-1 B). Genomic DNA from

shA549 cells on days 0, 3, 6 and 12 was, PCR amplified and hybridized onto custom


95

Affymetrix Gene Modulation Array Platform (GMAP) arrays. Following extraction and

normalization, each shRNA was assigned an shRNA Activity Ranking Profile score (SHARP)

which, within each condition, incorporates not only the abundance of the shRNA relative to

time 0 but shA549 doubling time, enabling shRNA drop-out assessment at multiple time

points. On average, 5 hairpins form the TRC1 library target one gene. SHARP scores were

converted to the gene activity ranking profile (GARP) scores by averaging the two lowest

SHARP scores for each gene in order to quantify the gene’s shRNA dropout rate. A

scatterplot of Z normalized GARP scores (zGARP) in the fluvastatin-treated shA549’s plotted

against the ethanol-treated shA549’s (Figure 3-1 C) shows the gene drop-out distribution. A

relatively negative zGARP score attests to the general essentiality of a gene as related to

cellular proliferation.


96

A B

C

ï� ï� ï� ï� 0

ï�ï�

ï�ï�

�

Control (EtOH) Treatment (zGARP)

Fluv

asta

tin T

reat

men

t (zG

ARP)

3,�.%

0$3�.�

65(%)�

GGPS1

A

3 6 120

20

40

60

80

100 replicate 1replicate 2replicate 3

Days

% v

iabi

lity

rela

tive

to e

than

ol c

ontr

ol

D

ZFluva - ZEthanolzFuva - zEthanol

Den

sity

151

D


97

Figure 3-1: A genome wide dropout screen uncovers putative shRNAs that

potentiate fluvastatin-induced cell death.

(A) A schematic representing the steps of the screen. A549 cells were treated with sublethal

doses of fluvastatin (4-5 µM) every 3 days, over 12 days. Viability relative to ethanol in the

fluvastatin-treated replicates was consistently in the 60-70% range on days 3, 6 and 12 (B).

gDNA was amplified and shRNA populations hybridized onto custom Affymetrix Gene

Modulation Array Platform (GMAP) arrays. Differences between shRNA abundances over

the 12 days were incorporated into a SHARP score, collapsed into a GARP score and z

transformed. A scatterplot of fluvastatin treated zGARP scores plotted against the ethanol

(EtOH) treated ones show the comparative shRNA zGARP score distribution. Significant hits

chosen for follow-up are indicated on the scatterplot (yellow diamonds) (C). Hits were

defined as genes where the ΔZfluvastatin was more than three standard deviations lower than the

mean ΔZfluvastatin of all genes (D).


98

In order to uncover fluvastatin-potentiating drop-outs, zGARP score differences

between the fluvastatin and ethanol treated groups (Zfluvastatin – Zethanol = ΔZfluvastatin) were

computed. Genes were the ΔZfluvastatin was more than three standard deviations lower than the

mean ΔZfluvastatin of all genes were defined as hits (Figure 3-1 D, Table 3.1). Taken together,

we have identified 151 potential putative genes whose knockdown could potentiate the anti-

cancer effects of fluvastatin.


99

Gene symbol Gene name Zethanol Zfluvastatin ΔZ(fluvastatin-ethanol)

1 TEX101 testis expressed 101 -0.126 -3.332 -3.2072 ZNF174 zinc finger protein 174 -0.643 -3.762 -3.1193 PSMB2 proteasome (prosome, macropain) subunit, beta type, 2 -4.042 -7.102 -3.0604 UNK similar to LIM homeo domain transcription factor -0.680 -3.678 -2.9975 GPR113 G protein-coupled receptor 113 -2.446 -5.371 -2.9256 HMGCS1 3-hydroxy-3-methylglutaryl-Coenzyme A synthase 1 (soluble) -0.268 -3.016 -2.7487 GMFG glia maturation factor, gamma -1.153 -3.851 -2.6988 RPS19 ribosomal protein S19 -4.700 -7.392 -2.6929 SF3B3 splicing factor 3b, subunit 3, 130kDa -1.299 -3.984 -2.685

10 DCK deoxycytidine kinase -2.010 -4.686 -2.67711 PRPF38A PRP38 pre-mRNA processing factor 38 (yeast) domain containing A -1.830 -4.495 -2.66512 SREBF2 sterol regulatory element binding transcription factor 2 -0.387 -3.048 -2.66113 UNK similar to class II basic helix-loop-helix protein TCF23 -0.882 -3.481 -2.59914 MEN1 multiple endocrine neoplasia I -2.556 -5.066 -2.51015 TBK1 TANK-binding kinase 1 -0.843 -3.273 -2.43116 PTCH2 patched homolog 2 (Drosophila) -2.942 -5.345 -2.40317 GGPS1 geranylgeranyl diphosphate synthase 1 -3.952 -6.342 -2.38918 SIK2 salt-inducible kinase 2 -1.672 -3.982 -2.31019 SFRS1 splicing factor, arginine/serine-rich 1 -2.578 -4.821 -2.24420 AFG3L2 AFG3 ATPase family gene 3-like 2 (yeast) -5.350 -7.590 -2.24021 GRM1 glutamate receptor, metabotropic 1 -0.923 -3.148 -2.22522 KIAA1012 KIAA1012 -0.114 -2.285 -2.17123 IL6R interleukin 6 receptor -3.090 -5.231 -2.14124 RAD18 RAD18 homolog (S. cerevisiae) -1.207 -3.342 -2.13525 MMAA methylmalonic aciduria (cobalamin deficiency) cblA type -0.950 -3.043 -2.09426 GMIP GEM interacting protein -0.395 -2.485 -2.09027 IFNA17 interferon, alpha 17 -0.080 -2.168 -2.08928 TRIM50 tripartite motif-containing 50 -1.780 -3.849 -2.07029 KRT20 keratin 20 -1.673 -3.741 -2.06830 EDF1 endothelial differentiation-related factor 1 -2.104 -4.158 -2.05431 PFKFB1 6-phosphofructo-2-kinase/fructose-2,6-biphosphatase 1 -0.028 -2.067 -2.04032 HSD3B2 hydroxy-delta-5-steroid dehydrogenase, 3 beta- and steroid delta-isomerase 2 -2.611 -4.649 -2.03833 G3BP2 GTPase activating protein (SH3 domain) binding protein 2 -1.605 -3.630 -2.02634 EP300 E1A binding protein p300 -2.030 -4.052 -2.02235 ETFDH electron-transferring-flavoprotein dehydrogenase -2.001 -4.019 -2.01836 ZNF642 zinc finger protein 642 -1.342 -3.341 -1.99937 AGPAT4 1-acylglycerol-3-phosphate O-acyltransferase 4 (lysophosphatidic acid acyltransferase, delta) -2.165 -4.161 -1.99638 MTMR11 myotubularin related protein 11 -1.054 -3.048 -1.99439 MS4A6E membrane-spanning 4-domains, subfamily A, member 6E -2.376 -4.363 -1.98740 GPR32 G protein-coupled receptor 32 -1.470 -3.453 -1.98341 ADCK5 aarF domain containing kinase 5 -2.156 -4.136 -1.98042 HMGN3 high mobility group nucleosomal binding domain 3 0.328 -1.637 -1.96543 DACH2 dachshund homolog 2 (Drosophila) -2.082 -4.008 -1.92644 ALG1L asparagine-linked glycosylation 1-like -0.339 -2.261 -1.92245 NFIA nuclear factor I/A -1.468 -3.386 -1.91846 CDC42BPG CDC42 binding protein kinase gamma (DMPK-like) -0.957 -2.873 -1.91647 WWP2 WW domain containing E3 ubiquitin protein ligase 2 -2.423 -4.330 -1.90648 ALDH7A1 aldehyde dehydrogenase 7 family, member A1 -1.772 -3.665 -1.89349 USP44 ubiquitin specific peptidase 44 -3.219 -5.099 -1.88050 RPS13 ribosomal protein S13 -1.990 -3.866 -1.87751 EXOSC9 exosome component 9 -2.901 -4.762 -1.86252 EPHB4 EPH receptor B4 -2.311 -4.163 -1.85253 UXS1 UDP-glucuronate decarboxylase 1 -0.583 -2.432 -1.84954 PKN2 protein kinase N2 -1.077 -2.911 -1.83455 CAMK2G calcium/calmodulin-dependent protein kinase II gamma -3.947 -5.776 -1.83056 ATP5I ATP synthase, H+ transporting, mitochondrial F0 complex, subunit E -1.971 -3.800 -1.82957 ZFY zinc finger protein, Y-linked -1.749 -3.575 -1.82558 CASP2 caspase 2, apoptosis-related cysteine peptidase -3.944 -5.764 -1.81959 SLC39A8 solute carrier family 39 (zinc transporter), member 8 -0.665 -2.481 -1.81660 ZFP82 zinc finger protein 82 homolog (mouse) -0.535 -2.349 -1.81461 PTPN22 protein tyrosine phosphatase, non-receptor type 22 (lymphoid) -2.779 -4.591 -1.81162 UNK similar to CDC-like kinase 3; cdc2/CDC28-like protein kinase 3 -1.313 -3.123 -1.81063 DOCK2 dedicator of cytokinesis 2 -1.035 -2.822 -1.78764 UPF1 UPF1 regulator of nonsense transcripts homolog (yeast) -1.222 -3.002 -1.77965 KCNN3 potassium intermediate/small conductance calcium-activated channel, subfamily N, member 3-0.608 -2.384 -1.77666 ZNF230 zinc finger protein 230 -1.558 -3.327 -1.76967 GLRA1 glycine receptor, alpha 1 -0.708 -2.471 -1.76368 UBQLNL ubiquilin-like -1.934 -3.694 -1.76069 RGS13 regulator of G-protein signaling 13 -1.223 -2.982 -1.76070 ST6GALNAC3 ST6 (alpha-N-acetyl-neuraminyl-2,3-beta-galactosyl-1,3)-N-acetylgalactosaminide alpha-2,6-sialyltransferase 3-2.400 -4.152 -1.75271 CS citrate synthase -1.198 -2.943 -1.74572 SRGN serglycin -1.778 -3.521 -1.743


100

Table 3-1: Putative gene hits targeted by shRNAs that potentiate the anti-

cancer effects of fluvastatin.

Gene symbol Gene name Zethanol Zfluvastatin ΔZ(fluvastatin-ethanol)

73 CSNK2B casein kinase 2, beta polypeptide -0.358 -2.097 -1.73874 RPS6 ribosomal protein S6 -4.126 -5.856 -1.73075 STYK1 serine/threonine/tyrosine kinase 1 -2.857 -4.583 -1.72576 SAE1 SUMO1 activating enzyme subunit 1 -1.483 -3.198 -1.71577 XYLB xylulokinase homolog (H. influenzae) -0.599 -2.296 -1.69878 PRKAA1 protein kinase, AMP-activated, alpha 1 catalytic subunit -1.008 -2.693 -1.68479 DDX21 DEAD (Asp-Glu-Ala-Asp) box polypeptide 21 -2.028 -3.708 -1.68080 UNK sorbitol dehydrogenase-like -0.688 -2.366 -1.67881 FLVCR1 feline leukemia virus subgroup C cellular receptor 1 -1.061 -2.737 -1.67682 ULK4 unc-51-like kinase 4 (C. elegans) -0.702 -2.378 -1.67683 KLK14 kallikrein-related peptidase 14 -1.267 -2.940 -1.67484 NLN neurolysin (metallopeptidase M3 family) -2.308 -3.961 -1.65385 RASAL3 RAS protein activator like 3 -0.573 -2.217 -1.64486 DDX3Y DEAD (Asp-Glu-Ala-Asp) box polypeptide 3, Y-linked -0.790 -2.434 -1.64387 TNF tumor necrosis factor (TNF superfamily, member 2) 0.403 -1.236 -1.64088 CBR3 carbonyl reductase 3 -1.377 -3.016 -1.63989 NCOA2 nuclear receptor coactivator 2 -2.190 -3.817 -1.62890 RTF1 Rtf1, Paf1/RNA polymerase II complex component, homolog (S. cerevisiae) -1.927 -3.547 -1.61991 RBM5 RNA binding motif protein 5 -1.075 -2.693 -1.61892 ATP2A3 ATPase, Ca++ transporting, ubiquitous -1.009 -2.627 -1.61893 ANKK1 ankyrin repeat and kinase domain containing 1 -0.181 -1.797 -1.61694 MITF microphthalmia-associated transcription factor -1.198 -2.804 -1.60695 CASR calcium-sensing receptor -1.125 -2.730 -1.60596 HAL histidine ammonia-lyase -1.424 -3.026 -1.60297 UNK similar to TRIMCyp -0.955 -2.556 -1.60198 PPIL1 peptidylprolyl isomerase (cyclophilin)-like 1 0.303 -1.297 -1.60099 SERPINC1 serpin peptidase inhibitor, clade C (antithrombin), member 1 -1.247 -2.841 -1.594

100 GAPDHS glyceraldehyde-3-phosphate dehydrogenase, spermatogenic -0.906 -2.494 -1.588101 RNF39 ring finger protein 39 -1.750 -3.337 -1.587102 ADAMTS13 ADAM metallopeptidase with thrombospondin type 1 motif, 13 -1.690 -3.277 -1.586103 CNGB1 cyclic nucleotide gated channel beta 1 -2.157 -3.742 -1.585104 PDPK1 3-phosphoinositide dependent protein kinase-1 -2.173 -3.751 -1.578105 IQCD IQ motif containing D -1.300 -2.877 -1.577106 RNF167 ring finger protein 167 -1.262 -2.839 -1.576107 RPL23A ribosomal protein L23a -3.661 -5.234 -1.573108 PLCG1 phospholipase C, gamma 1 -1.317 -2.885 -1.568109 DNAJB9 DnaJ (Hsp40) homolog, subfamily B, member 9 -2.281 -3.848 -1.568110 PTPN23 protein tyrosine phosphatase, non-receptor type 23 -1.072 -2.631 -1.559111 RASSF6 Ras association (RalGDS/AF-6) domain family member 6 -0.967 -2.526 -1.559112 MLLT6 myeloid/lymphoid or mixed-lineage leukemia (trithorax homolog, Drosophila); translocated to, 6-0.916 -2.472 -1.555113 BRCA1 breast cancer 1, early onset 0.212 -1.342 -1.554114 PI4KB phosphatidylinositol 4-kinase, catalytic, beta -1.723 -3.265 -1.542115 B3GNT6 UDP-GlcNAc:betaGal beta-1,3-N-acetylglucosaminyltransferase 6 (core 3 synthase) -1.314 -2.838 -1.525116 RPS14 ribosomal protein S14 -6.974 -8.491 -1.517117 RAPGEFL1 Rap guanine nucleotide exchange factor (GEF)-like 1 -1.223 -2.733 -1.511118 TTR transthyretin -1.252 -2.759 -1.507119 NMUR2 neuromedin U receptor 2 -0.458 -1.964 -1.506120 CCNL1 cyclin L1 -0.406 -1.911 -1.504121 LCT lactase -0.065 -1.563 -1.498122 MLH1 mutL homolog 1, colon cancer, nonpolyposis type 2 (E. coli) -0.903 -2.393 -1.491123 PRKAG1 protein kinase, AMP-activated, gamma 1 non-catalytic subunit -2.317 -3.803 -1.485124 DYSF dysferlin, limb girdle muscular dystrophy 2B (autosomal recessive) -1.334 -2.819 -1.485125 KCTD4 potassium channel tetramerisation domain containing 4 0.136 -1.347 -1.482126 H6PD hexose-6-phosphate dehydrogenase (glucose 1-dehydrogenase) -0.534 -2.010 -1.475127 DDR2 discoidin domain receptor tyrosine kinase 2 -0.430 -1.902 -1.472128 PPP1R15B protein phosphatase 1, regulatory (inhibitor) subunit 15B -2.764 -4.233 -1.470129 ZNF317 zinc finger protein 317 -0.932 -2.402 -1.470130 STK32A serine/threonine kinase 32A -2.760 -4.228 -1.467131 MYO3B myosin IIIB -0.662 -2.126 -1.463132 MAP2K4 mitogen-activated protein kinase kinase 4 -0.220 -1.680 -1.460133 TTK TTK protein kinase -0.822 -2.279 -1.457134 UNK similar to Phosphorylase B kinase gamma catalytic chain, skeletal muscle isoform (Phosphorylase kinase gamma subunit 1)-2.559 -4.015 -1.456135 NHLH2 nescient helix loop helix 2 -0.893 -2.348 -1.455136 OVOL2 ovo-like 2 (Drosophila) -4.077 -5.531 -1.454137 TGM1 transglutaminase 1 (K polypeptide epidermal type I, protein-glutamine-gamma-glutamyltransferase)-1.030 -2.478 -1.448138 CDH9 cadherin 9, type 2 (T1-cadherin) -3.663 -5.109 -1.446139 GRM4 glutamate receptor, metabotropic 4 -0.297 -1.743 -1.446140 NRP2 neuropilin 2 -1.049 -2.494 -1.444141 ATP5J ATP synthase, H+ transporting, mitochondrial F0 complex, subunit F6 -1.640 -3.082 -1.443142 RYR3 ryanodine receptor 3 -0.375 -1.816 -1.441143 DNTTIP2 deoxynucleotidyltransferase, terminal, interacting protein 2 -2.014 -3.454 -1.441144 TAS2R45 taste receptor, type 2, member 45 -1.354 -2.791 -1.438145 RPA2 replication protein A2, 32kDa -0.902 -2.336 -1.433146 GTF2E2 general transcription factor IIE, polypeptide 2, beta 34kDa -1.633 -3.064 -1.431147 ANUBL1 AN1, ubiquitin-like, homolog (Xenopus laevis) -0.395 -1.814 -1.419148 PLK1 polo-like kinase 1 (Drosophila) 0.517 -0.896 -1.414149 DUSP22 dual specificity phosphatase 22 -1.553 -2.965 -1.413150 NUDT21 nudix (nucleoside diphosphate linked moiety X)-type motif 21 -1.271 -2.680 -1.409151 GRM5 glutamate receptor, metabotropic 5 -1.602 -3.009 -1.407


101

3.2.2 Validation of shRNA hits in the A549 cells uncovers potentiators of statin-induced

apoptosis and anti-proliferative effects.

The genome-wide shRNA screen yielded a list of 151 potential hits (Table 3-1).

Amongst the hits, ranked based on negativity of the zGARP scores, there were several MVA

pathway related genes namely 3-hydroxy-3-methylglutaryl-Coenzyme A synthase 1

(HMGCS1), geranylgeranyl diphosphate synthase 1 (GGPS1), sterol regulatory element-

binding protein 2 (SREBF2). Immediately preceding the conversion of 3-hydroxy-3-

methylglutaryl-coenzyme A (HMG-CoA) to MVA catalyzed by HMGCR, HMGCS1 is

responsible for the production of HMG-CoA. Downstream of HMGCR, GGPS1 catalyzes the

synthesis of gernaylgernayl pyrophosphate (GGPP) from farnesyl diphosphate and

isopentenyl diphosphate. GGPP is integral to the post-translation modification of many

proteins and GGPP incorporation to their C-terminus results in protein prenylation. SREBF2

is a key transcription factor involved in the maintenance of cholesterol homeostasis through a

sterol-mediated feedback loop. Upon sterol-depletion, the latent SREBF2 is activated and

induces transcription of sterol-responsive genes including HMGCR, HMGCS1 and low-

density lipoprotein receptor (LDLr) (Figure 3-2)38, 39.


102


103

Figure 3-2: HMGCR inhibition by statins triggers a restorative feedback

loop in hepatocytes

(A) HMGCR maintains normal levels of MVA derived end-products in hepatocytes.

(B) Upon statin-mediated HMGCR inhibition, the cell is depleted of sterols such as

cholesterol (1). In sterol-depleted cells, the latent transcription factor, SREBF2, is

translocated from the endoplasmic reticulum to the Golgi apparatus where is it cleaved by

proteases (2) and its mature, active form translocates to the nucleus (3). (C) Once in the

nucleus, SREBF2 binds to promoter regions containing sterol response elements (SREs)

inducing the transcription of HMGCR, HMGCS1 and LDLr. (D) Resulting increased LDLr at

the surface of the membrane internalizes and thereby lowers serum LDL-cholesterol levels.


104

Targeting the MVA through inhibition of HMGCR using statins has been suggested

for decades as being an effective anti-cancer therapeutic strategy. Dysregulation of the MVA

pathway at the level of increased activity and expression of HMGCR has been implicated in a

variety of cancers. Targeting the MVA pathway at more than one nodal point has the potential

to be an effective therapeutic strategy and we therefore chose two MVA pathway related hits,

GGPS1 and SREBF2 for further validation. Subsequent putative hits chosen for validation

were mitogen-activated protein kinase kinase 4 (MAP2K4) and phosphatidylinositol 4-kinase,

catalytic beta (PIK4B). Targeting these genes as an anti-cancer therapeutic strategy, either

alone or in combination with statins has not been previously explored but is attractive since

pharmacological compounds targeting the MAP2K4 and PIK4B proteins directly or indirectly

are available.

Distribution of gene hits chosen for validation are highlighted in the scatterplot of

fluvastatin and ethanol zGARP scores (Figure 3-1C). The individual drop-out profiles of the

two best shRNAs for each gene over time are shown Figure 3-3.


105

Figure 3-3: Line plots of the two best hairpins for each hit chosen for

validation show the dropout over time.

Line plots of the two best hairpins for each gene (GGPS1 MAP2K4, PI4KB and SREBF2)

from ethanol (solid line) and fluvastatin (dashed line) treated shA549 cells. Matching hairpins

in each condition are color coded in red and blue.

0 5 10 15

02

46

810

1214

GGPS1

Time (doublings)

GM

AP In

tens

ity (m

ean

log2

)

TRCN0000045788

TRCN0000045791

EtOH ControlFluvastatin

0 5 10 15

02

46

810

1214

MAP2K4

Time (doublings)

GM

AP In

tens

ity (m

ean

log2

)

TRCN0000001389TRCN0000001390


0 5 10 15

02

46

810

1214

PI4KB

Time (doublings)

GM

AP In

tens

ity (m

ean

log2

)

TRCN0000005694

TRCN0000005696


0 5 10 15

02

46

810

1214

SREBF2

Time (doublings)

GM

AP In

tens

ity (m

ean

log2

)

TRCN0000020664

TRCN0000020667



106

Both hairpins targeting the MVA pathway related GGPS1 and SREBF2 show a robust

decrease in the fluvastatin treated group relative to ethanol. In order to validate the four hits,

we chose two independent shRNAs from the original TRC1 library targeting each gene and

generated A549 sublines stably expressing each shRNA (Table 3-2). Two control sublines,

transduced with shRNAs targeting a non-human gene, LacZ, were concurrently generated.


107

Table 3-2: TRC Clone ID's of shRNA hits chosen for validation.

shRNA TRC Clone IDshSREBF2 #1 TRCN0000020664shSREBF2 #2 TRCN0000020667shGGPS1 #1 TRCN0000045788shGGPS1 #2 TRCN0000045790shPI4KB #1 TRCN0000005695shPI4KB #2 TRCN0000005696

shMAP2K #1 TRCN0000001393shMAP2K #2 TRCN0000039916


108

Protein expression of each target was assessed for knockdown following selection of

cell lines for the stably expressing shRNA construct (Figure 3-4).


109

IB: MAP2K4

IB: Tubulin

IB: PI4KB

IB: Tubulin

1 2 3 4

1 2 3 4

IB: GGPS1

IB: Actin 52

38

38

31

52

52

38

102

76

150

102

76

IB: SREBF2

52 IB: Tubulin

shGGPS1 #1 shGGPS2 #2shLacZ #2

1 = shLacZ #1

2 = shLacZ #2

3 = shHit #1

4 = shHit #2

1 2 3 4

0

20

40

60

80

100

* * *

GG

PS

1 r

ela

tiv

e e

xp

re

ss

ion

1 2 3 4

Figure 3-4: Generation of A549 sublines stably expressing shRNA

constructs demonstrates knockdown of protein or mRNA target.

Following selection, protein lysates harvested from asynchronously growing A549 cells were

immunoblotted with the appropriate primary and secondary antibodies. Knockdown of

GGPS1 was also validated using real-time PCR (polymerase chain reaction) relative to

GAPDH and expression was normalized to the shLacZ #2 control. All immunoblots are

representative of a minimum of three independent experiments. Real-time data represent the

mean ± SD of three independent experiments *p < 0.05 (one-way ANOVA with a Dunnet

post test, the shGGPS1 groups being significantly different from the control shLacZ.


110

Stable knockdown of SREBF2 was evident in both sublines when compared to the

LacZ A549 controls and growth immediately following selection of each subline was

comparable to the LacZ subline controls. Following exposure to sublethal concentrations of

fluvastatin, both shSREBF2 A549 sublines showed a remarkable increase in apoptosis as

measured by assessing cellular pre-G1 DNA content (Figure 3-5 A). The fluvastatin-induced

apoptosis was dose dependent and minimal in the shLacZ A549 sublines, as expected.

Furthermore, the percent MTT activity and the half-maximal fluvastatin inhibitory

concentrations (IC50) values generated by treating A549 sublines with a range fluvastatin

doses were significantly decreased in both shSREBF2 A549 sublines, respectively (Figure 3-5

C and D). The IC50 values for both LacZ control A549 sublines, ranging from 30-45 µM, were

reduced to 5 µM in the shSREBF2 A549 sublines.

GGPS1 knockdown at the protein level (Figure 3-4) was not strong and GGPS1

mRNA levels were reduced by only 50% in both GGPS1 A549 sublines. However, there was

sufficient knockdown to observe a dramatic induction fluvastatin-mediated apoptosis (Figure

3-5A). Again, as with the SREBF2 A549 sublines, the apoptosis was dose-dependent. The

concomitant addition of GGPP with fluvastatin, the immediate downstream product formed

by GGPS1 mediated catalysis, abrogated apoptosis (Figure 3-5 B) indicating that the

knockdown of GGPS1 was on-target.

A dramatic reduction in fluvastatin IC50 in the GGPS1 A549 sublines when compared

to LacZ controls was also apparent (Figure 3-4 C). Variations in cellular growth between the

different sublines could impact fluvastatin sensitivity irrespective of any potentiating effects

caused by the specific knockdown the target. All cell lines were equally seeded at the

commencement of an experiment and but there were no significant differences between the

final net absorbance in the control cells (Figure 3-4 E). Although the MTT assay measures the

activity of mitochondrial succinate dehydrogenase and is therefore only a proxy for cellular


111

proliferation, the lack of observable differences in the final absorbance between the untreated

controls in the assessed A549 sublines suggests that growth between the cells was similar.

At least one of the A549 sublines expressing shRNA targeting MAP2K4 and PIK4B,

shMAP2K4 #2 and shPI4KB #2 respectively, showed increased sensitivity to fluvastatin

exposure when compared to the A549 control sublines (Figure 3-5A). Although a small

degree of potentiation upon treatment with fluvastatin was observed in the shMAP2K4 #2

subline, the drop-out over time was minimal (Figure 3.3). Furthermore, as both shMAP2K4

sublines were characterized by similar levels of knockdown at the protein level (Figure 3.4)

and only one demonstrated an appreciable effect, it is possible that this is an off target effect.

Further validation using different shRNA constructs or siRNA’s would need to be conducted

before concluding that knockdown of MAP2K4 potenitates statin-induced apoptosis.

In the case of PI4KB, degree of shRNA knockdown was not correlated to levels of

protein expression. Although the shPI4KB #1 A549 subline consistently expressed lower

amounts of PI4KB protein following multiple passages, this subline was not sensitive to

fluvastatin-induced apoptosis (Figure 3-5 A). The generation of A549 sublines stably

expressing shRNAs targeted against GGPS1, PI4KB and MAP2K4 were difficult to grow

following selection and sublines needed several passages before cells could be utilized for

experiments. Adaptation of a subline to stable knockdown of a target essential for growth is a

possibility, potentially confounding assessments of sensitivity due to compensatory

upregulation of other proteins or pathways. A subline such as A549 shPI4K #1, expressing

lower amounts of PI4KB protein relative to the A549 shPI4K #2 cells took longer to adapt

following selection and did not show increased fluvastatin-induced apoptosis. In both of the

MAP2K4 sublines showed that the knockdown of the kinase targets was decreased over time

(data not shown) indicating selection of subpopulations with higher protein levels of

MAP2K4. Furthermore, the shMAP2K4 #2 A549 subline had higher levels of basal apoptosis


112

when compared to all other sublines (Figure 3-5A). In the shGGPS1 A549 sublines, a 50%

mRNA target reduction was sufficient to potentiate the apoptosis-inducing effects of

fluvastatin. Taken together, validation of hits using the approach of generating sublines stably

expressing shRNA constructs is not appropriate for every target and highly dependent on

whether the target is essential for growth.

Whereas stable knockdown of SREBF2 was well tolerated and differences between

the shSREBF2 A549 sublines and control cells were apparent only when challenged with

statins, probable adaptation upon knockdown of MAP2K4 and PI4KB in the A549 cells led to

a masking of their ability to behave as potenitators of fluvastatin-induced apoptosis and anti-

proliferative effects. Therefore, for further validation of non-MVA related targets we chose

the approach of transiently knocking down the target using small interfering RNAs (siRNA).

The availability of a direct pharmacological inhibitor of PI4KB led us to further validate this

candidate hit.


113

shLa

cZ #

1

shLa

cZ #

2

shM

AP2

K4

#1

shM

AP2

K4

#2

shPI

4KB

#1

shPI

4KB

#2

shSR

EBF2

#1

shSR

EBF2

#2

shG

GPS

1 #1

shG

GPS

1 #2

0

20

40

60

80ethanol 5 uM fluvastatin 10 uM fluvastatin

** *

**

*

**

* **

*

% P

re-G

1 Po

pula

tion

A

B C

shla

cZ #

1

shla

cZ #

2

shPI

4KB

#2

shSR

EBF2

#1

shSR

EBF2

#2

shG

GPS

1 #1

shG

GPS

1 #2

0

20

40

60

80

* * * *IC50

Flu

vast

atin

(uM

)

D

0 50 100 150 2000

20

40

60

80

100

120

Fluvastatin (uM)

Nor

mal

ized

%

MTT

Act

ivity

shlacZ #1

shSREBF2 #1shSREBF2 #2

shlacZ #2

E

shla

cZ #

1

shla

cZ #

2

shPI

4KB

#2

shSR

EBF2

#1

shSR

EBF2

#2

shG

GPS

1 #1

shG

GPS

1 #2

0.0

0.5

1.0

1.5

2.0

* * * *

Net

Abs

orba

nce

shLa

cZ #

2

shG

GPS

1 #1

0

20

40

60

80

% P

re-G

1 Po

pula

tion

+ 1

0 uM

GG

PP


114

Figure 3-5: shRNAs targeting gene hits identified in the screen potentiate

fluvastatin-induced apoptosis and anti-proliferative effects.

(A) Exposure of A549 sublines stably expressing the indicated shRNA construct to sub-lethal

doses of fluvastatin for 72 hours significantly induced apoptosis in at least one of the two cell

lines expressing an shRNA targeting a single gene hit. Apoptosis was assessed by fixed

propidium iodide staining and dead cells were characterized by pre-G1 DNA content. Bars

represent the mean ± SD of at least three independent experiments *p < 0.05 (one-way

ANOVA with Tukey post test within each A549 subline comparing all groups, the fluvastatin

treated groups being significantly different from the ethanol control). (B) Fluvastatin-induced

apoptosis in the shGGPS1 #1 subline was reversed by the concomitant addition of GGPP.

Bars represent the mean ± SD of at least 3 independent experiments *p < 0.05 (one-way

ANOVA with Tukey post test within each A549 subline comparing all groups, the fluvastatin

treated group being significantly different from the others). The A549 sublines were

independently generated a second time and similar results were obtained. (C) Knockdown of

SREBF2 and GGPS1 significantly decreased the fluvastatin IC50 values computed as

described in methods. Sublines were exposed to a range of fluvastatin doses for 72 hours and

viability was assessed using the MTT assay. *p < 0.05 (one-way ANOVA with Tukey post

test comparing all groups, all SREBF2 and GGPS1 sublines being significantly different from

the LacZ controls). Representative dose-response curves for the SREBF2 sublines are shown

in (D) and corresponding final net absorbances (following subtraction of blank) at 590 nm for

the untreated controls 96 hours post-growth for each subline are shown in (E) Bars represent

the mean ± SD of at least 3 independent experiments.


115

3.2.3 Concomitant targeting of PI4KB and HMGCR showed anti-proliferative effects in

lung and breast cell lines.

The reduction of target mRNA using siRNA’s is short–term and therefore

advantageous if adaptive responses to knockdown of a specific target occur. In order to clarify

whether the inconclusive results obtained with the use of the shRNAs targeting PI4KB were

off-target or a result of adaptive responses, we employed transient knockdown approaches.

We therefore used siRNA duplexes at low nanomolar concentrations and transiently

transfected parental A549 cells. Reduction of PI4KB protein was evident 72 hours post

transfection as compared to A549 cells transfected with a negative scrambled siRNA control

and an untransfected control (Figure 3-6 A). As cells were treated with fluvastatin 48 hours

post transfection, reductions in PI4KB protein levels coincided with fluvastatin exposure

allowing for potentiation of anti-proliferative activity to be assessed. In A549 cells transfected

with siRNA #1, which resulted in the greatest decrease of PI4KB protein, there was a

significant reduction in fluvastatin IC50 values compared to cells transfected with the siRNA

negative control (Figure 3-6 B).

To exclude the possibility that the identified hit shRNA hits that potentiated

fluvastatin-mediated anti-cancer effects are limited to one cell line, we extended our

experiments to a different cell line of another tumour type, the triple negative (ERα-, PR-, and

HER2-negative) basal-like MDAMB231 breast cancer cells. Breast cancers characterized by

triple negative tumours have fewer effective therapeutic options upon recurrence 40. As

lipophilic statins such as fluvastatin have already shown clinical promise in the treatment of

breast cancer18, and additional clinical trials are underway, uncovering potenitiators of

fluvastatin’s anti-cancer effects in this tumour type has high potential for immediate clinical

utility.


116

Similar to the A549 cells, reduction of PI4KB protein also occurred 72 hours post

transfection relative to the negative scrambled siRNA and untransfected controls (Figure 3-

6C). Consequently, transfection with all three siRNA duplexes targeting PI4KB significantly

decreased fluvastatin IC50 values relative to negative siRNA control (Figure 3-6D). In the

breast cancer cells, all three siRNA duplexes consistently knocked down the PI4KB target and

this was accompanied with concomitant decreases in the fluvastatin IC50 values whereas in

the A549 cells, only one duplex, which demonstrated the greatest knockdown resulted in

sensitization to fluvastatin’s anti-cancer effects. Higher concentrations of the other two

duplxes could be used to achieve greater knockdown and further verify PI4KB as a

potenitator of statin induced anti-cancer effects.


117

DB

A549 MDAMB231

1 2 3 4 5 1 2 3 4 5170

130

100

170

130

100

1 = Control2 = siRNA Control3 = PI4KB siRNA #14 = PI4KB siRNA #2 5 = PI4KB siRNA #3

A C

siR

NA

Con

trol

siR

NA

#1

siR

NA

#2

siR

NA

#30.0

0.5

1.0

1.5

2.0

2.5

* ** *

IC50

Flu

vast

atin

(uM

)

siR

NA

con

trol

siR

NA

#1

siR

NA

#2

siR

NA

#30

510152025303540

* *

IC50

Flu

vast

atin

(uM

)


118

Figure 3-6: Transient knockdown of PI4KB sensitizes lung and breast

cancer cells to the anti-proliferative effects of fluvastatin.

PI4KB protein expression in A549 lung and MDAMB231 breast cancer cells was decreased

72 hours following transient transfection with 10 nM of three different siRNAs targeting

PI4KB mRNA relative to a negative scrambled siRNA control and an untransfected control

(A and C respectively). Fluvastatin IC50 values in A549 and MDAMB231 cells were

significantly lower following siRNA transfection targeting PI4KB. 24 hours post-transfection,

cells were re-seeded in 96 well plates, allowed to adhere overnight and treated with a range of

fluvastatin doses for 72 hours (B and D respectively). Bars represent the mean ± SD of least 3

independent experiments *p < 0.05 (one-way ANOVA with Tukey post test comparing all

groups, the indicated groups being significantly different from the siRNA control).

Immunoblots are representative of a minimum of three independent experiments for the

MDAMB231 cells and two independent experiments for the A549 cells at the 72 hour time-

point.


119

To corroborate the observations using a pharmacological approach we used a PI4KB

inhibitor, PIK-9341, 42 and co-treated the A549 and MDAMB 231 cells. PIK-93 when used at a

concentration of 1 µM is reported to specifically inhibit PI4KB and not the type III

counterpart phosphatidylinositol 4-kinase, catalytic alpha (PI4K-alpha) which requires PIK-

93 concentrations greater than 10 µM to achieve inhibition42. Furthermore, this dose was

sublethal in our cell systems allowing for the assessment of potentiation. Remarkably, the

fluvastatin-PIK-93 combination not only significantly lowered fluvastatin IC50 values in A549

(Figure 3-7A and B) and MDAMB231 cells (Figure 3-7 C) but also induced apoptosis in the

MDAMB231 cells (Figure 3-7 D and E). In combination with PIK-93, the fluvastatin

concentrations needed to induce apoptosis were in the sub micromolar range (0.25 µM).

Using genetic and pharmacological approaches we have validated a novel target that

sensitizes breast and lung cancer cells to fluvastatin induced apoptosis.


120

0

10

20

30

40

*

IC50

Flu

vast

atin

(uM

)

A

0 50 100 150 2000

20

40

60

80

100

% N

orm

aliz

ed

MTT

Act

ivity

Fluvastatin (uM)

BSt

atin

0.0

0.5

1.0

1.5

IC50

Flu

vast

atin

(uM

)

P = 0.053

C

0

10

20

30

40

*%

Pre

-G1

Popu

latio

nD

- + - - + - Fluvastatin (0.25 M) µ Fluvastatin (0.5 M) µ PIK93 (1 M) µ

- - + - - + - - - + + +

*

0 200 400 600 800 1000

0

100

200

300

4.87

0 200 400 600 800 1000

0

100

200

300

400

500

10.6

0 200 400 600 800 1000

0

100

200

300

400

500

4.05

0 200 400 600 800 1000

0

100

200

300

400

30.1

PI

Fluvastatin (0.5 M) µ PIK93 (1 M) µ Control Fluvastatin + PIK93 E

A549

MDAMB231

A549

MDAMB231

MDAMB231

Fluvastatin Fluvastatin + PIK93 (1 uM)




121

Figure 3-7: The combination of fluvastatin and a pharmacological inhibitor

of PI4KB is anti-proliferative and induces apoptosis in lung and breast

cancer cells.

Fluvastatin IC50 values in A549 and MDAMB231 cells were decreased following

concomitant treatment with 1 µM of a PI4KB inhibitor PIK93 for 72 hours (A and C

respectively). Representative fluvastatin dose-response curves in the presence and absence of

PIK93 for A549 cells are shown in B, viability was assessed using the MTT assay. Bars

represent the mean ± SD of least 3 independent experiments. *p < 0.05 (t-test, unpaired, two-

tailed). The combination of sublethal doses of PIK93 and fluvastatin induce apoptosis in

MDAMB231 cells following 96 hours of treatment (D), representative histograms are shown

in (E). Apoptosis was assessed by fixed propidium iodide staining and dead cells were

characterized by pre-G1 DNA content. Bars represent the mean ± SD of 3 independent

experiments.*p < 0.05 (one-way ANOVA with Tukey post test comparing all groups, the

indicated groups being significantly different from control, fluvastatin 0.25 µM, fluvastatin

0.5 µM and PIK93 1 µM groups).


122

3.2.4 Down-regulating the SREBF2 mediated sterol-feedback loop in combination with

statins is an effective anti-cancer strategy to induce lung and breast tumour cell kill.

The remarkable sensitization to the anti-cancer effects of fluvastatin upon stable

SREBF2 knockdown in A549 cells (Figure 3-5) prompted a more thorough investigation into

the generalizability of this phenomenon in another solid tumour type. Due to recent positive

clinical results in the treatment of breast cancer patients with fluvastatin18, we chose two

breast cancer cell lines, the triple negative basal-like MDAMB231 and the luminal MCF7

(ERα+, PR+, and HER2- (Human Epidermal Growth Factor Receptor 2), both previously

characterized to have some molecular fidelity to the primary tumours whose subtypes they

represent43. The MDAMB231 and MCF7 cells display a differential range of statin

sensitivities; while the MDAMB231 are characterized by fluvastatin IC50 values in the low

micromolar range, those of the MCF7 cells are ten fold higher. As there is no universal

biomarker of statin-sensitivity, identifying a target that sensitizes cells displaying a wide

range of statin sensitivities to statin-mediated anti-cancer effects is therapeutically attractive.

SREBF2 knockdown in the A549 shSREBF2 sublines was well tolerated; we therefore

generated MDAMB231 and MCF7 sublines stably expressing shRNAs targeting SREBF2.

SREBF2 protein levels in the MDAMB231 sublines were decreased relative to shLacZ

sublines (Figure 3-8A) and consequently this resulted in a significant reduction of fluvastatin

IC50 values (Figure 3-8B). Although in both MDAMB231 and MCF7 shSREBF2 sublines

there was a reduction of fluvastatin IC50 relative to controls, the magnitude of difference was

greater in the MCF7 cells (Figure 3-8D). Control fluvastatin IC50 values in the MCF7 control

sublines ranged from 30 to 50 µM and were in the lower micromolar range (5 to 10 µM) in

the shSREBF2 sublines.


123

Exposure of representative shSREBF2 and control shLacZ MDAMB231 and MCF7

sublines to sublethal doses of fluvastatin resulted in a dramatic increase in apoptosis in the

shSREBF2 sublines (Figure 3-8 C and E respectively).


124

A

IB: Tubulin

IB: SREBF2

Uncleaved SREBF2

Cleaved SREBF2

1 2 3 4

1 2 3 40.0

0.5

1.0

1.5

2.0

2.5

**IC

50 F

luva

stat

in (u

M)

*

1 2 4 30

10

20

30

40

50

60

* *IC50

Flu

vast

atin

(uM

)

E

1 = shLacZ #12 = shLacZ #23 = shSREBF2 #14 = shSREBF2 #2

2 30

10

20

30ethanol1 uM fluvastatin

% P

re-G

1 Po

pula

tion

*

2 40

10

20

30

40

50

60 ethanol 10 uM fluvastatin

% P

re-G

1 Po

pula

tion

*

B C

F

130

100

70

55

MDAMB231 MDAMB231 MDAMB231

MCF7MCF7D1 2 4

130

100

70

55 IB: Tubulin

Uncleaved SREBF2

Cleaved SREBF2

IB: SREBF2

MCF7


125

Figure 3-8: Stable knockdown of SREBF2 sensitizes breast cancer MCF7

and MDAMB231 cells to the pro-apoptotic and anti-proliferative effects of

fluvastatin.

Parental MDAMD231 and MCF7 cells were infected with lentiviral particles and puromycin-

selected for 48 hours. Following puromycin-selection, protein lysates harvested from

asynchronously growing MDAMB231 and MCF7 cells were immunoblotted and showed

decreased SREBF2 protein expression relative to cells transduced with the LacZ controls (A

and D respectively). Fluvastatin IC50 values in the MDAMB231 and MCF7 sublines

expressing the SREBF2 shRNAs were significantly lower when compared to the control

sublines. Dose response curves were generated following exposure to fluvastatin for 72 hours

and viability was assessed using the MTT assay (B and E respectively). Exposure of

representative MDAMB231 and MCF7 sublines to sublethal doses of fluvastatin caused a

significant increase in apoptosis relative to a control LacZ subline (C and F respectively).

Apoptosis was assessed by fixed propidium iodide staining and dead cells were characterized

by pre-G1 DNA content. Bars represent the mean ± SD of least 3 independent experiments *p

< 0.05 (one-way ANOVA with Tukey post test comparing all groups, the indicated groups

being significantly different from control sublines in C and F and from all other groups in B

and E). Immunoblots are representative of a minimum of three independent experiments.


126

SREBF2 is a key transcriptional regulator of cholesterol homeostasis44. SREBF2’s

role in inducing the expression of sterol responsive genes upon sterol-depletion for instance

following statin treatment prompted us to assess the levels of HMGCR and other sterol-

responsive genes in the shSREBF2 sublines.

Remarkably, the knockdown of SREBF2 by two independent shRNAs nearly

completely abolished the upregulation of sterol-responsive genes HMGCR and HMGCS1

following treatment with fluvastatin. While in the shLacZ A549 and MCF7 sublines there was

a robust upregulation of HMGCR and HMGCS1 following 24 hours of fluvastatin treatment,

this was largely abrogated in the shSREBF2 sublines (Figure 3-9 A and B respectively).

Levels of SREBF2 itself did not increase upon fluvastatin treatment and robust knockdown

was confirmed in both cell lines. As sensitivity to drugs may be affected by proliferation45, we

conducted proliferation assays using all shSREBF2 and shLACZ sublines. There were no

significant differences in doubling times in all four sublines in MCF7, MDAMB231 and

A549 cells (Figure 3-9 C).

The sensitization to fluvastatin-induced apoptosis upon knockdown of SREBF2 was

not only evident in A549 lung cancer cells but observable in breast cancer cells characterized

by a wide range of statin sensitivities. Next we investigated whether targeting HMGCR using

statins in combination with SREBF2 knockdown is an effective therapeutic strategy in vivo.


127

A B

A549

1 2 1 2 1 2 1 20

50

100

150Ethanol

Fluvastatin

Rela

tive H

MG

CR

levels

1 2 1 2 1 2 1 20

200

400

600

800

1000

Ethanol

Fluvastatin

Rela

tive H

MG

CS

1 levels

1 2 1 2 1 2 1 20

20

40

60

80

Ethanol

Fluvastatin

Rela

tive S

RE

BF

2 levels

1 2 1 2 1 2 1 20

50

100

150

200

250

Ethanol

Fluvastatin

Rela

tive H

MG

CR

levels

1 2 1 2 1 2 1 20

200

400

600

800

1000

Ethanol

Fluvastatin

Rela

tive H

MG

CS

1 levels

1 2 1 2 1 2 1 20

50

100

150

200

250

Ethanol

Fluvastatin

Rela

tive S

RE

BF

2 levels

* ** *

* *

* *

MCF7

**

C

0

10

20

30

40

Do

ub

lin

g tim

e (h

ou

rs

)

shLacZ #1

shLacZ #2

shSREBF2 #1

shSREBF2 #2

A549MCF7MDAMB231

shSREBF2 #2shSREBF2 #1 shLacZ #1 shLacZ #2



shSREBF2 #2shSREBF2 #1 shLacZ #1 shLacZ #2 shSREBF2 #2shSREBF2 #1 shLacZ #1 shLacZ #2



128

Figure 3-9: Knockdown of SREBF2 abrogates the upregulation of HMGCR

and HMGCS1 upon fluvastatin exposure.

Parental A549 and MCF7 cells were infected with lentiviral particles and puromycin-selected

for 48 hours. Four sublines were generated from each cell line. 2 sublines stably expressing

shRNAs targeting SREBF2 and 2 control sublines stably expressing shRNAs targeting LacZ.

Following 48 hours of puromycin-selection of infected cells, RNA was harvested from

subconfluent cells grown over several passages and treated for 24 hours with 10 µM of

fluvastatin or ethanol control and assessed for expression of HMGCR, HMGCS1 and

SREBF2 relative to GAPDH in shA549 (A) and shMCF7 (B) sublines (shSREBF #1,

shSREBF #2, shLacZ #1, shLacZ #2). Data represent the mean ± SD of at least three

independent experiments. *p < 0.05 (t-test, unpaired, two-tailed, comparison made within

each subline). For basal comparisons across four sublines, *p < 0.05 (one-way ANOVA with

a Tukey post test, the LacZ sublines being significantly different than both SREBF2

sublines).There were no significant differences in the growth rates between the shSREBF2

and control shLacZ sublines in MDAMB231, MCF7 and A549 cells as assessed using the

CyQuant cell proliferation assay kit (C).


129

3.2.5 Fluvastatin delays tumour growth in MDAMB231 xenografts.

To evaluate whether SREBF2 potentiated the anti-cancer effects of fluvastatin in vivo,

we treated NOD-SCID mice harboring established xenografts of MDAMB231 shSREBF2 #1

and shLacZ #2 sublines. We chose to orally administer fluvastatin because this is the route of

delivery for humans. Mice were subcutaneously injected with MDAMB231 sublines and

when xenografts were palpable, mice were orally treated 3 times a week with PBS, or 25

mg/kg fluvastatin for ten days. Fluvastatin significantly decreased final tumour weight (Figure

3-10 A) and final tumour volume (Figure 3-10 B) in the MDAMB231 shSREBF2 xenograft

group. Although not reaching statistical significance likely due to limited number of mice in a

group, fluvastatin also had an effect on final tumour volume and weight within the

MDAMB231 shLacZ group.

A similar experiment was conducted using the shSREBF2 #2 and shLacZ #2 sublines.

Mice were treated with lower oral doses of fluvastatin (10 mg/kg), and tumours were allowed

to reach a larger volume. Treatment of the mice harboring the shSREBF2 xenografts with

fluvastatin resulted in a small but significant reduction in tumour volume following 21 days of

treatment. There were no differences between the PBS and fluvastatin treated groups in the

mice harboring the shLaCZ #2 controls. Experiments are underway with the remaining the

cell lines.


130

Figure 3-10: Fluvastatin delays tumour growth in MDAMB231 xenografts.

NOD-SCID mice were injected subcutaneously with 4 million MDAMB231 shSREBF2 or

shLACZ cells. After tumours became palpable (2-3 weeks), mice within each subline

xenograft were randomized into groups and treated three times a week with 25 mg/kg

fluvastatin orally (p.o.) or PBS (p.o.) and vehicle (i.p.) (Con.). Tumour volume was measured

every two-three days. After 10 days of treatment, mice were sacrificed and tumours were

resected and weighed (B). n= 3-5 mice per group. *p < 0.05 (t-test, unpaired, two-tailed,

comparison made within each subline).


131

3.4 Methods

Compounds.

Fluvastatin, dissolved in ethanol was purchased from US Biologicals. PIK-93, dissolved in

DMSO was purchased from SelleckBio and geranylgeranyl pyrophosphate (GGPP) was

obtained as a solution from Sigma. DNAse free RNAse was purchased from Roche.

Cell Culture.

All cell lines were cultured in Dulbecco's modified eagle's medium H21 (DMEM H21).

Media was supplemented with 10% fetal bovine serum and penicillin-streptomycin. Cells

were grown asynchronously as monolayers at 37°C in 5% CO2 and cell lines were routinely

confirmed to be mycoplasma-free (MycoAlert mycoplasma detection kit, Lonza).

shRNA dropout screen.

A549 adenocarcinoma lung cancer cell line stably transduced with a TRC1 shRNA library

(sh-A549) containing over 80,000 shRNAs encoded in the pLKO.1 vector targeting 16,000

human genes, approximately 7,000 of which have been validated,30, 46 were seeded in factory

tissue culture vessels (Nunc, 16.5 million sh-A549 cells/vessel). 2 hours post-seeding and

upon adherence, cells were treated with sublethal doses fluvastatin (4-5 µM) or ethanol

control. Fluvastatin dosing was chosen and accordingly adjusted throughout the duration of

the experiment such that treatment led to decrease cell viability of 20-40% relative to ethanol

control. Every 3 days cells were re-seeded and excess cells were pelleted and stored at -80 °C

for subsequent genomic DNA extraction (QIAmp Blod Maxi Kit, Qiagen). PCR amplified

shRNA populations were hybridized to the custom Affymetrix Gene Modulation Array

Platform (GMAP) arrays as detailed elsewhere47.


132

shRNA screen analysis and scoring.

Data was analyzed as previously described32. Briefly, data was extracted and normalized

using the CCBR-OICR Lentiviral Technology (COLT) database48. Affymetrix Power Tools

v1.12.0 were used for extraction and the Bioconductor affy package (v1.26.1) for

normalization. Within each treatment condition and time point, the fold changes relative to

time zero (t0), with cell doublings being accounted for, were calculated and an shRNA

Activity Ranking Profile score (SHARP) was assigned. For each gene, two hairpins with the

most negative SHARP scores were averaged and collapsed to the gene activity ranking profile

(GARP)32 and Z-score normalized. Z score differences between the fluvastatin and ethanol

treated groups (Zfluvastatin – Zethanol = ΔZfluvastatin) were computed. A negative Z score is

indicative of an shRNA that was depleted over time. A lower ΔZfluvastatin , is indicative of an

shRNA that was more rapidly depleted in the fluvastatin treatment group and therefore the

absence of the gene targeted by the shRNA, could be a potentiator of statin-induced anti-

cancer effects. Genes were the ΔZfluvastatin was more than three standard deviations lower than

the mean ΔZfluvastatin of all genes were defined as hits.

MTT, fixed propidium iodide (PI) assays.

(3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT, Sigma-Aldrich) assays

were done as previously described49. In brief 1.5 x 104 to 3 x 105 cells/ml were plated in 96

well plates and after 24 hours, treated with the indicated compounds and concentration ranges

for 72 hours. Half-maximal inhibitory concentrations (IC50) values were computed from

dose-response curves using Prism (v5.0, GraphPad Software). For fixed PI assays, 3-3.5 x 105

cells/dish were seeded in 10 cm dishes, after 24 hours, treated for 72 hours as indicated. Cells

were fixed in 70% ethanol, treated with DNAse free RNAse and stained with 50 µg/mL PI


133

(Sigma-Aldrich). Cells were analyzed for the dying pre-GI population by flow cytometry; ten

thousand events were acquired using a FACSCalibur cytometer (BD Biosciences).

Immunoblotting.

Sub-confluent growing cells were washed with PBS and lysed using boiling hot SDS lysis

buffer (1.1% SDS, 11% glycerol, 0.1M Tris pH 6.8) with 10% β-mercaptoethanol. Blots were

probed with anti-SREBF2 (BD Pharmingen), anti-MAP2K4 (Cell Signaling Technology),

anti-PI4KB (BD Pharmingen), anti-tubulin (Santa Cruz Biotechnology), anti-actin (Sigma).

Primary antibodies were detected using the Odyssey infrared imaging system with IRDye-

labeled secondary antibodies (LI-COR Biosciences)

Real-Time PCR.

RNA was harvested from sub-confluent growing cells using TRIzol Reagent (Invitrogen).

cDNA was synthesized from 0.5 µg using SuperScript III (Invitrogen). HMGCR mRNA

levels relative to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were assayed as

previously described using SYBR Green master mix (Applied Biosystems)50. LDLR mRNA

levels were assayed with TaqMan Gene Expression Assays, SREBF2 (Hs01081784_m1),

GGPS1 (Hs00191442_m1), GAPDH (Hs99999905_m1) using TaqMan master mix (Applied

Biosystems). ABI Prism 7900 Sequence Detection System was used for the acquisition and

analysis of relative transcript levels.

Generation of shRNA validation cell lines.

DNA from bacterial glycerol stocks of shRNAs from the TRC1 library (Table 3.) was

generated (Qiagen Plasmid Maxiprep kit). Lentiviral particles were generated by calcium

phosphate transfection of sub-confluent (50-60%) 293TV cells with 10 µg of TRC pLKO.1


134

puro shRNA construct, 5 µg each of pMDG1.vsvg, pRSV-Rev and pMDLg/pRRE constructs.

Virus was collected 24 and 48 hours later, filtered though a 0.45 µm filter and stored at -80 °C.

Stably expressing shRNA A549, MCF7 and MDAMB231 cell lines were generated following

infection of cells (60-70% confluent) with 2 ml of the lentiviral particles combined with 1 ml

of medium containing 8 µg/ml of polybrene. After 48 hours cells were split and selected using

2 µg/ml of puromycin and maintained as pooled populations.

siRNA transient transfections.

Cells were seeded in (1.2-1.5 x 105 cells/well) and 24 hours later transfected with individual

siRNA duplexes (Origene). Briefly, duplexes at a of concentration of 250 nM were incubated

in 100 µL of serum and antibiotic free DMEM medum with 20 µL of HiPerfect (Qiagen) for

10 minutes following which they were added to the cells. Cells were either harvested for

protein expression analysis or re-seeded in 96 wells for MTT assays.

Xenografts models.

Logarithmically growing MDAMB231 cells stably transduced with shRNA SREBF2 and

control constructs (4 million) were subcutaneously injected into the flanks of NOD SCID

mice. When tumours became palpable, 2-3 weeks post-injection, treatment commenced. Mice

were orally treated with 25 mg/kg of fluvastatin suspended in PBS 5 times a week. Following

ten days of treatment, tumour weight and volume were measured. Final tumour volume was

calculated using the following formula: (tumour length x width2)/2. Animal work was carried

out with the approval of the Princess Margaret Hospital ethics review board in accordance to

the regulations of the Canadian Council on Animal Care.


135

3.5 Discussion

Statins are currently being evaluated in the clinic as anti-cancer agents. To increase

their utility in a combination setting, we used a genome-wide shRNA screen to identify

potentiators of statin-induced anti-tumour effects. In this screen we treated the

adenocarcinoma A549 cell line stably transduced with the TRC1 shRNA library with

sublethal doses of fluvastatin and identified significant shRNAs that dropped out in response

to fluvastatin exposure over time. The A549 cell line is an adenocarcinoma NSCLC line and

represents a tumour type faced with considerable treatment challenges. Lung cancer is a

principal cause of cancer-related deaths and the histologically complex NSCLC comprises the

majority of lung cancer. Patients with the best prognosis are those with surgically resectable

early stage NSCLC but even those patients have a 70% 5-year survival and a high recurrence

rate51, 52. Although chemotherapy is used at various stages of intervention in NSCLC patients,

there are few effective treatment options and novel anti-cancer agents are urgently needed.

From the 151 putative genes whose knockdown could potentiate the anti-cancer effects of

statins in A549 cells, we validated four, SREBF2, GGPS1 MAP2K4 and PI4KB.

Indicative of the robustness of our screen was the identification of GGPS1 as a

potentiator of fluvastatin’s anti-proliferative effects. Targeting isoprenoid biosynthesis with

agents specifically developed to inhibit the isoprenoid transferase enzymes downstream of

HMGCR, has been investigated for its ability to perturb essential pathways in cancer cells

such as Ras. The most clinically advanced class of the isoprenoid transferase-inhibiting

compounds, are the farnesyltransferase inhibitors (FTIs), but they have yet to demonstrate real

clinical efficacy53. It has been suggested that the limited clinical anti-tumour activity of the

FTIs results from alternative prenylation stemming from the activity

gernaylgeranyltransferases I (GGTase-I). This has launched the development of dual FTI and


136

GGTase-I inhibitor, however these compounds have dose-limited toxicities precluding

clinical use54. To date, four drugs have been tested in more than seventy clinical trials:

tipifarnib, lonafarnib, BMS-214662, and L-778123. The first three drugs are FTIs and L-

778123 is the abovementioned dual FTI and GGTase-I inhibitor which is no longer in

development. The anti-cancer efficacy of these drugs was tested in solid and haematological

malignancies, alone or in combination with other agents and the results have been largely

negative55. Only one specific GGTase-I inhibitor has been tested in clinical trials and the

GGTase-II inhibitors have not yet reached clinical testing. Given the overall disappointing

clinical results of targeting the FTIs and GGTase’s, alternative therapeutic strategies of

disrupting protein prenylation in cancer cells, are needed. Here we demonstrate that targeting

the MVA pathway using statins and by knockdown of GGPS1, resulted in the dramatic

induction of apoptosis in A549 cells, reversible by the concomitant addition of GGPP (Figure

3-4 A and B). Our finding is supported another group who have found that inhibiting GGPS1

with newly synthesized GGPS1 inhibitors and statins has synergistic anti-cancer activity56.

This strategy of targeting GGPS1 and HMGCR is advantageous as it precludes the use of

clinically limited high doses of a single compound to induce tumour kill. The difficulty in

generating A549 sublines with highly sustained GGPS1 knockdown (Figure 3-3) is supported

by high negative Zethanol values (Table 3-1) indicting that GGPS1 is essential for cancer cell

growth and potentially a singularly useful target. Taken together, we have shown that

inhibition of the MVA pathway using statins and knockdown of another enzyme in the MVA

pathway, GGPS1 results in a robust induction of apoptosis reversible by concomitant addition

of GGPP and therefore likely caused by isoprenoid depletion and disruption of protein

prenylation.

HMGCR expression is in part controlled by SREBF2 mediated transcriptional

regulation. Sensing sterol depletion as for instance upon treatment with a statin, latent


137

SREBF2 is activated and induces transcription of sterol-responsive genes including HMGCR.

SREBF2 was identified as a drop-out hit in our genome-wide shRNA screen.

The generation shSREBF2 A549 sublines with stable suppression of SREBF2 protein and

mRNA levels was remarkably well tolerated and shSREBF2 A549 sublines were

characterized by the same growth rates as control sublines. The dramatic induction of

apoptosis and remarkable reduction in IC50 values upon fluvastatin exposure in the shSREBF2

sublines prompted further validation in breast cancer where statins have shown tremendous

therapeutic promise18. SREBF2 knockdown resulted in a dramatic sensitization to the pro-

apoptotic and anti-proliferative effects of fluvastatin in MCF7 and MDAMB231 cells (Figure

3-7) without any changes to the proliferation of the shSREBF2 sublines. Remarkably, despite

residual levels of SREBF2 mRNA (~10 %) in all shSREBF2 sublines, there was no

upregulation of HMGCR upon fluvastatin exposure.

Recently, an independent group published a second window-of-opportunity trial in

which patients with primary invasive breast cancer were treated with the lipophilic

atorvastatin 2 weeks prior to surgery57. In addition to evaluating tumour proliferation (Ki67

index), HMGCR protein expression prior to and following atorvastatin treatment in tumour

tissue was assessed. HMGCR-positive patient tumours not only experienced a significant

decrease in Ki67 following atorvastatin treatment but a concomitant increase in HMGCR

protein levels. These findings are of importance because they suggests that statins directly

target the tumour cell and that the tumour cell then experiences a feedback loop mediating

HMGCR upregulation following statin exposure. The anti-proliferative response in HMGCR-

positive tumours after statin treatment suggests a potential predictive biomarker for statin

response. We have previously shown that elevated HMGCR mRNA was correlated with poor

prognosis and reduced survival of breast cancer patients58. Taken together, HMGCR is an

important anti-cancer target whose over-expression has been linked to poor patient prognosis


138

and direct HMGCR inhibition in a patient tumour cell has demonstrated anti-proliferative

efficacy. The consequential implications of concomitant HMGCR upregulation within a

tumour following statin exposure in a clinical setting have not been addressed but as

demonstrated by our results, blocking the upregulation via SREBF2 knockdown could have

profound anti-tumour consequences. In cell culture, the implications of HMGCR upregulation

following statin exposure as it relates to statin-sensitivity, has been explored in the context of

multiple myeloma (MM)50. In MM cell lines, statin-sensitive MM cell lines did not

demonstrate a robust feedback loop and the expected upregulation of sterol-responsive genes

such as HMGCR. In contrast, statin-insensitive MM cell lines exhibited a robust feedback

loop. In the current study, we have demonstrated the functional consequences of blocking the

feedback loop in breast and lung cancer cells in the context of the remarkable sensitization to

statin-induced apoptosis (Figure 3-11).

Targeting SREBF2 with an shRNA was an effective cell culture strategy to

demonstrate anti-cancer efficacy in combination with statins. In vivo however, the clinical

utility of RNAi technology has yet to be demonstrated. The chief obstacle to systemic RNAi

non-viral delivery has been the appropriate delivery vehicle. While advances have been made

particularly by Alnylam Pharmaceuticals and Tekmira Pharmaceuticals in their use of

polyethylene glycol modified (PEGylated) lipid nanoparticle technology coupled with

stabilizing nucleic acid modifications, their lead compounds are still only in early phase

clinical trials59. Although there are no specific direct SREBF2 inhibitors, recently, compounds

such as betulin that indirectly inhibit SREBP activity by preventing its maturation through

inducing interaction of SREBP cleavage activating protein (SCAP) and Insig have been

identified60. Our work supports future evaluation of statins in combination with inhibitors of

SREBP activity for their anti-cancer efficacy.


139

LDLr$LDLr$

HMGCR$

LDLr$LDLr$

HMGCR$ sta-ns$

Sterol'deple*on,

A.$

SREBF2$

B.$

LDLr$LDLr$

HMGCR$ sta-ns$

Sterol'deple*on,

SRE$

HMGCR$

LDLr$

HMGCS1$

LDLr$LDLr$

Restora-on$of$$cholesterol$and$other$MVA$derived$endAproducts$$$

SRE$

HMGCR$

LDLr$

HMGCS1$

LDLr$LDLr$

Serum$LDLAcholesterol$

1),

2), 3),

Normal$cell$

LDLr$LDLr$

SRE$

HMGCR$

LDLr$

HMGCS1$

LDLr$LDLr$ LDLr$

LDLr$

HMGCR$

HMGCR$HMGCR$

Cholesterol,$FFP,$GGPP$

1)$

2)$

3)$ 4)$

Tumour$cell$LDLr$LDLr$

SRE$

LDLr$LDLr$ LDLr$

LDLr$HMGCR$

HMGCR$HMGCR$

$FFP,$GGPP$

FTIs,$GGTIs$$Protein$$prenyla-on$

SREBF2$

shRNA$

GGPS1$shRNA$

C.$


140

Figure 3-11: Targeting the MVA pathway in tumour cells.

(A, upper left panel) In normal hepatocytes, HMGCR maintains levels of MVA

derived end-products. (upper right panel) Upon statin-mediated HMGCR inhibition, the cell is

depleted of sterols such as cholesterol (1). In sterol-depleted cells, the latent transcription

factor, SREBF2, is translocated from the endoplasmic reticulum to the nucleus (3). (lower

right panel) Once in the nucleus, SREBF2 binds to promoter regions containing sterol

response elements (SREs) inducing the transcription of HMGCR, HMGCS1 and LDLr.

(lower left panel) Resulting increased LDLr at the surface of the membrane internalizes LDL-

choslterol from the serum thereby lowering LDL-cholesterol levels.

(B) In tumour cells, the MVA pathway can be dysregulated in several ways. (1) LDLr

activity and expression can be elevated in cancer cells as can HMGCR expression and activity

(2). Tumour cells are vulnerable to the depletion of sterol derived MVA endproducts that are

used for protein prenylation of several key oncogenic proteins such as Ras (3). Dysregulation

at the level of the restorative sterol feedback loop mediated by the SREBF2 can also occur.

Some cancer cells are not able to upregulate sterol-responsive genes following statin

inhibition while others are characterized by abnormally high cholesterol import despite

already high endogenous cholesterol levels (4). Points of potential therapeutic intervention of

the MVA pathway in cancer cells are illustrated in (C). Statins target HMGCR, the rate-

limiting enzyme of the MVA pathway. FTIs and GGTIs prevent protein prenylation of key

oncogenic molecules such as Ras. Newly defined mechanisms by which the MVA pathway

can be targeted include the knockdown of GGPS1, an enzyme responsible for the production

of GGPP and one level upstream of the gernaylgernayl transferase enzymes, and knockdown

of the transcription SREBF2 which prevents the upregulation of sterol-responsive genes in

cancer cells where the feedback loop is intact.


141

In addition to SREBF2 and GGPS1, we have identified non-MVA related targets

PI4KB and MAPK4, whose knockdown potentiated the anti-cancer effects of statins. PI4KB

regulates transport of ceramide, required for the synthesis of sphingomyelin, between the

endoplasmic reticulum (ER) and Golgi42, and phosphorylates phosphatidylinositol (PI) to

phosphatidylinositol 4-phosphate (PI(4)P). Recent reports implicate PI4KB’s oncogenic

involvement in mammary neoplasia and ability to cooperate with eukaryotic elongation factor

1 alpha 2 (eEF1A2) in disrupting acinar morphogenesis61. Recently, a genomic and

transcriptomic analysis of 2,000 breast tumours identified the PI4KB gene as being located in

a genomic region prone to amplification and characterized by putative driver aberrations62.

This may support a role for targeting breast tumours characterized by both this genomic

abnormality and HMGCR-positivity using PI4KB inhibitors and statins. Indeed, we have

already shown in our cell culture studies that combining fluvastatin with the PI4KB inhibitor

PIK-93 induces apoptosis in the MDAMB231 breast cancer cell line. The same study

identified MAP2K4 as being located in a genomic region of recurrent deletions in breast

tumours62. As suggested by our results where knockdown of MAP2K4 potentiated the anti-

cancer effects of fluvastatin, targeting tumours with deletions encompassing MAP2K4 could

potentially widen the therapeutic index of statins.

Through a genome-wide shRNA screen we have identified and validated several

potentially clinically relevant sensitizers of statin-mediated anti-cancer effects. Recent clinical

evidence showing statin efficacy in breast cancer patients treated with the same doses used to

treat hypercholesterolemic patients underscores their importance in the clinical cancer care

setting. We have shown that targeting the MVA pathway at multiple points maximizes tumour

cell kill. Targeting SREBF2 effectively suppresses the restorative feedback loop that causes

upregulation of HMGCR in response to statin treatment, effectively widening statins’


142

therapeutic window. Statins are safe, FDA-approved and, many are off patent making them

cost-effective anti-cancer agents. Our study has identified targets whose inhibition has the

potential to extend the benefit of statins and maximize tumour cell kill.


143

3.6. References:

1. Goldstein JL, Brown MS. Regulation of the mevalonate pathway. Nature 1990; 343(6257): 425-30.

2. Ahern TP, Pedersen L, Tarp M, Cronin-Fenton DP, Garne JP, Silliman RA et al.

Statin prescriptions and breast cancer recurrence risk: a Danish nationwide prospective cohort study. Journal of the National Cancer Institute 2011; 103(19): 1461-8.



4. Nielsen SF, Nordestgaard BG, Bojesen SE. Statin use and reduced cancer-related

mortality. The New England journal of medicine 2012; 367(19): 1792-802. 5. Khurana V, Bejjanki HR, Caldito G, Owens MW. Statins reduce the risk of lung

cancer in humans: a large case-control study of US veterans. Chest 2007; 131(5): 1282-8.



7. Li HY, Appelbaum FR, Willman CL, Zager RA, Banker DE. Cholesterol-modulating

agents kill acute myeloid leukemia cells and sensitize them to therapeutics by blocking adaptive cholesterol responses. Blood 2003; 101(9): 3628-34.

8. Xia Z, Tan MM, Wong WW, Dimitroulakos J, Minden MD, Penn LZ. Blocking

protein geranylgeranylation is essential for lovastatin-induced apoptosis of human acute myeloid leukemia cells. Leukemia 2001; 15(9): 1398-407.

9. Campbell MJ, Esserman LJ, Zhou Y, Shoemaker M, Lobo M, Borman E et al. Breast

cancer growth prevention by statins. Cancer Res 2006; 66(17): 8707-14. 10. Pelaia G, Gallelli L, Renda T, Fratto D, Falcone D, Caraglia M et al. Effects of statins

and farnesyl transferase inhibitors on ERK phosphorylation, apoptosis and cell viability in non-small lung cancer cells. Cell proliferation 2012; 45(6): 557-65.

11. Koyuturk M, Ersoz M, Altiok N. Simvastatin induces apoptosis in human breast

cancer cells: p53 and estrogen receptor independent pathway requiring signalling through JNK. Cancer letters 2007; 250(2): 220-8.


144

12. Liu H, Liang SL, Kumar S, Weyman CM, Liu W, Zhou A. Statins induce apoptosis in ovarian cancer cells through activation of JNK and enhancement of Bim expression. Cancer chemotherapy and pharmacology 2009; 63(6): 997-1005.

13. Zhao TT, Le Francois BG, Goss G, Ding K, Bradbury PA, Dimitroulakos J.

Lovastatin inhibits EGFR dimerization and AKT activation in squamous cell carcinoma cells: potential regulation by targeting rho proteins. Oncogene 2010; 29(33): 4682-92.

14. Rao S, Lowe M, Herliczek TW, Keyomarsi K. Lovastatin mediated G1 arrest in

normal and tumor breast cells is through inhibition of CDK2 activity and redistribution of p21 and p27, independent of p53. Oncogene 1998; 17(18): 2393-402.

15. Shibata MA, Ito Y, Morimoto J, Otsuki Y. Lovastatin inhibits tumor growth and lung

metastasis in mouse mammary carcinoma model: a p53-independent mitochondrial-mediated apoptotic mechanism. Carcinogenesis 2004; 25(10): 1887-98.

16. Dimitroulakos J, Thai S, Wasfy GH, Hedley DW, Minden MD, Penn LZ. Lovastatin

induces a pronounced differentiation response in acute myeloid leukemias. Leukemia & lymphoma 2000; 40(1-2): 167-78.

17. Schmidt F, Groscurth P, Kermer M, Dichgans J, Weller M. Lovastatin and

phenylacetate induce apoptosis, but not differentiation, in human malignant glioma cells. Acta neuropathologica 2001; 101(3): 217-24.

18. Garwood ER, Kumar AS, Baehner FL, Moore DH, Au A, Hylton N et al. Fluvastatin

reduces proliferation and increases apoptosis in women with high grade breast cancer. Breast Cancer Res Treat 2010; 119(1): 137-44.



20. van der Spek E, Bloem AC, Sinnige HA, Lokhorst HM. High dose simvastatin does

not reverse resistance to vincristine, adriamycin, and dexamethasone (VAD) in myeloma. Haematologica 2007; 92(12): e130-1.

21. Thibault A, Samid D, Tompkins AC, Figg WD, Cooper MR, Hohl RJ et al. Phase I





Blockade of adaptive defensive changes in cholesterol uptake and synthesis in AML


145

by the addition of pravastatin to idarubicin + high-dose Ara-C: a phase 1 study. Blood 2007; 109(7): 2999-3006.

24. Kawata S, Yamasaki E, Nagase T, Inui Y, Ito N, Matsuda Y et al. Effect of pravastatin

on survival in patients with advanced hepatocellular carcinoma. A randomized controlled trial. Br J Cancer 2001; 84(7): 886-91.

25. Graf H, Jungst C, Straub G, Dogan S, Hoffmann RT, Jakobs T et al.

Chemoembolization combined with pravastatin improves survival in patients with hepatocellular carcinoma. Digestion 2008; 78(1): 34-8.

26. Han JY, Lee SH, Yoo NJ, Hyung LS, Moon YJ, Yun T et al. A randomized phase II

study of gefitinib plus simvastatin versus gefitinib alone in previously treated patients with advanced non-small cell lung cancer. Clinical cancer research : an official journal of the American Association for Cancer Research 2011; 17(6): 1553-60.

27. Fire A, Xu S, Montgomery MK, Kostas SA, Driver SE, Mello CC. Potent and specific

genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature 1998; 391(6669): 806-11.

28. Huesken D, Lange J, Mickanin C, Weiler J, Asselbergs F, Warner J et al. Design of a

genome-wide siRNA library using an artificial neural network. Nature biotechnology 2005; 23(8): 995-1001.

29. Buchholz F, Kittler R, Slabicki M, Theis M. Enzymatically prepared RNAi libraries.

Nature methods 2006; 3(9): 696-700. 30. Moffat J, Grueneberg DA, Yang X, Kim SY, Kloepfer AM, Hinkle G et al. A

lentiviral RNAi library for human and mouse genes applied to an arrayed viral high-content screen. Cell 2006; 124(6): 1283-98.

31. Tochitani S, Hayashizaki Y. Functional screening revisited in the postgenomic era.

Mol Biosyst 2007; 3(3): 195-207. 32. Marcotte R, Brown KR, Suarez F, Sayad A, Karamboulas K, Krzyzanowski PM et al.

Essential gene profiles in breast, pancreatic, and ovarian cancer cells. Cancer discovery 2012; 2(2): 172-89.

33. Luo J, Emanuele MJ, Li D, Creighton CJ, Schlabach MR, Westbrook TF et al. A

genome-wide RNAi screen identifies multiple synthetic lethal interactions with the Ras oncogene. Cell 2009; 137(5): 835-48.

34. Brummelkamp TR, Fabius AW, Mullenders J, Madiredjo M, Velds A, Kerkhoven RM

et al. An shRNA barcode screen provides insight into cancer cell vulnerability to MDM2 inhibitors. Nat Chem Biol 2006; 2(4): 202-6.

35. Huang S, Holzel M, Knijnenburg T, Schlicker A, Roepman P, McDermott U et al.

MED12 controls the response to multiple cancer drugs through regulation of TGF-beta receptor signaling. Cell 2012; 151(5): 937-50.


146

36. Zhu YX, Tiedemann R, Shi CX, Yin H, Schmidt JE, Bruins LA et al. RNAi screen of

the druggable genome identifies modulators of proteasome inhibitor sensitivity in myeloma including CDK5. Blood 2011; 117(14): 3847-57.

37. Bennis F, Favre G, Le Gaillard F, Soula G. Importance of mevalonate-derived

products in the control of HMG-CoA reductase activity and growth of human lung adenocarcinoma cell line A549. International journal of cancer. Journal international du cancer 1993; 55(4): 640-5.

38. Horton JD, Goldstein JL, Brown MS. SREBPs: activators of the complete program of

cholesterol and fatty acid synthesis in the liver. J Clin Invest 2002; 109(9): 1125-31. 39. Clendening JW, Penn LZ. Targeting tumor cell metabolism with statins. Oncogene

2012; 31(48): 4967-78. 40. Metzger-Filho O, Tutt A, de Azambuja E, Saini KS, Viale G, Loi S et al. Dissecting

the heterogeneity of triple-negative breast cancer. J Clin Oncol 2012; 30(15): 1879-87. 41. Knight ZA, Gonzalez B, Feldman ME, Zunder ER, Goldenberg DD, Williams O et al.

A pharmacological map of the PI3-K family defines a role for p110alpha in insulin signaling. Cell 2006; 125(4): 733-47.

42. Toth B, Balla A, Ma H, Knight ZA, Shokat KM, Balla T. Phosphatidylinositol 4-

kinase IIIbeta regulates the transport of ceramide between the endoplasmic reticulum and Golgi. The Journal of biological chemistry 2006; 281(47): 36369-77.

43. Neve RM, Chin K, Fridlyand J, Yeh J, Baehner FL, Fevr T et al. A collection of breast

cancer cell lines for the study of functionally distinct cancer subtypes. Cancer Cell 2006; 10(6): 515-27.

44. Brown MS, Goldstein JL. Cholesterol feedback: from Schoenheimer's bottle to Scap's

MELADL. Journal of lipid research 2009; 50 Suppl: S15-27. 45. Valeriote F, van Putten L. Proliferation-dependent cytotoxicity of anticancer agents: a

review. Cancer research 1975; 35(10): 2619-30. 46. Blakely K, Ketela T, Moffat J. Pooled lentiviral shRNA screening for functional

genomics in mammalian cells. Methods Mol Biol 2011; 781: 161-82. 47. Ketela T, Heisler LE, Brown KR, Ammar R, Kasimer D, Surendra A et al. A

comprehensive platform for highly multiplexed mammalian functional genetic screens. BMC genomics 2011; 12: 213.

48. Koh JL, Brown KR, Sayad A, Kasimer D, Ketela T, Moffat J. COLT-Cancer:

functional genetic screening resource for essential genes in human cancer cell lines. Nucleic acids research 2012; 40(Database issue): D957-63.


147

49. Dimitroulakos J, Ye LY, Benzaquen M, Moore MJ, Kamel-Reid S, Freedman MH et al. Differential sensitivity of various pediatric cancers and squamous cell carcinomas to lovastatin-induced apoptosis: therapeutic implications. Clin Cancer Res 2001; 7(1): 158-67.

50. Clendening JW, Pandyra A., Li, Z., Boutros P.C., Martirosyan A., Jurisica I., Trudel,

S., Penn L.Z. Exploiting the mevalonate pathway to distinguish statin-sensitive multiple myeloma. Blood 2009: Under Revision.

51. Mountain CF. Revisions in the International System for Staging Lung Cancer. Chest

1997; 111(6): 1710-7. 52. Paoletti L, Pastis NJ, Denlinger CE, Silvestri GA. A decade of advances in treatment

of early-stage lung cancer. Clinics in chest medicine 2011; 32(4): 827-38. 53. Tsimberidou AM, Chandhasin C, Kurzrock R. Farnesyltransferase inhibitors: where

are we now? Expert opinion on investigational drugs 2010; 19(12): 1569-80. 54. Martin NE, Brunner TB, Kiel KD, DeLaney TF, Regine WF, Mohiuddin M et al. A

phase I trial of the dual farnesyltransferase and geranylgeranyltransferase inhibitor L-778,123 and radiotherapy for locally advanced pancreatic cancer. Clinical cancer research : an official journal of the American Association for Cancer Research 2004; 10(16): 5447-54.

55. Holstein SA, Hohl RJ. Is there a future for prenyltransferase inhibitors in cancer

therapy? Current opinion in pharmacology 2012; 12(6): 704-9. 56. Dudakovic A, Wiemer AJ, Lamb KM, Vonnahme LA, Dietz SE, Hohl RJ. Inhibition

of geranylgeranyl diphosphate synthase induces apoptosis through multiple mechanisms and displays synergy with inhibition of other isoprenoid biosynthetic enzymes. The Journal of pharmacology and experimental therapeutics 2008; 324(3): 1028-36.




al. Dysregulation of the mevalonate pathway promotes transformation. Proceedings of the National Academy of Sciences of the United States of America 2010; 107(34): 15051-6.

59. Snead NM, Rossi JJ. RNA interference trigger variants: getting the most out of RNA

for RNA interference-based therapeutics. Nucleic acid therapeutics 2012; 22(3): 139-46.

60. Tang JJ, Li JG, Qi W, Qiu WW, Li PS, Li BL et al. Inhibition of SREBP by a small

molecule, betulin, improves hyperlipidemia and insulin resistance and reduces atherosclerotic plaques. Cell metabolism 2011; 13(1): 44-56.


148

61. Pinke DE, Lee JM. The lipid kinase PI4KIIIbeta and the eEF1A2 oncogene co-operate

to disrupt three-dimensional in vitro acinar morphogenesis. Experimental cell research 2011; 317(17): 2503-11.

62. Curtis C, Shah SP, Chin SF, Turashvili G, Rueda OM, Dunning MJ et al. The

genomic and transcriptomic architecture of 2,000 breast tumours reveals novel subgroups. Nature 2012; 486(7403): 346-52.


149

Chapter 4 : Discussion


150

4.1 Statins as anti-cancer agents – How translational is the cell culture work?

Studies exploring the anti-cancer properties of statins in cell culture studies of cell

lines or primary patient samples have been numerous and span a wide variety of solid and

liquid tumours. Establishment of a statin-induced anti-proliferative or apoptotic effect is

usually followed by an assessment of statin-induced perturbation of an activated oncogenic

pathway or effects on an over-expressed protein known to contribute to oncogenic

progression in that particular tumour model. Work in our laboratory has shown that lovastatin

suppresses the Raf/MEK/ERK pathways in primary AML patient samples responsive to

statin-induced apoptosis1 and others have reported similar effects in breast cancer

concomitant with decreases in NF-κB activity and adapter protein 1 DNA binding activities2.

Anti-proliferative simvastatin effects on breast cancer cell lines have also been attributed to

activation of JNK and c-Jun phosphorylation3, 4. In MM, Rho proteins, whose isoprenyation

by the addition of GGPP is critical for proper membrane localization, have been shown to

mediate statin-induced apoptosis and reversal of cell-adhesion mediated drug resistance

through statin mediated inhibitory effects on Rho kinase5. Suppressive effects on the AKT

pathway have been reported in lung cancer 6,7 and breast cancer cells8.

Nearly all the cell culture studies showing statin anti-cancer efficacy were

accompanied by effect reversal following the addition of MVA. Most, though not all of the

effects have been attributed to depletion of the isoprenylation arm of the MVA pathway as

most commonly the addition of GGPP blocked statin-mediated effects. Amidst this plethora

of statin reported molecular effects in cell lines and ex vivo cultured primary patient samples,

it is difficult to identify a specific biomarker of statin response since many of these effects

following HMGCR inhibition appear to be indirect consequences of perturbations on

commonly active pathways. While it is clear that inhibition of geranylgeranylation of Rho


151

GTPases is a critical mediator of statins’ anti-cancer activity, the identification of a specific

proteins responsible for these effects has been elusive although some potential candidates

such as Ras-related C3 botulinum toxin substrate 1 (Rac1) and cell division cycle 42 (Cdc42)9

have been identified.

So far there has been a general lack of a translational link between cell culture studies

and clinical investigations. Most of these cell culture studies address effects of statins directly

on a tumour cell but there is little clinical evidence to indicate that a statins directly target a

tumour cell bringing up the issue of how much statin actually reaches the extra-hepatic sites.

Usually, the concentrations used in cell culture are not clinically achievable. Although the

early dose-finding studies did show that lovastatin and simvastatin can be tolerated at very

high doses, these studies did not actually demonstrate any real efficacy and micromolar doses

of around 10 were detected in only a few patients. Furthermore, statins are not formulated for

administration at such high doses as the maximum dose of a single pill is 80 mg, and this

limits long-term administration. Oftentimes, the rationale of the statin used in a clinical trial

does not reflect the cell culture hypothesis upon which efficacy was based . Recently,

lovastatin was found to inhibit EGFR dimerization and synergize with the EGFR inhibitor

gefitinib in inducing cytotoxicity in various cell lines10. Based on this preclinical data, a

clinical trial, currently ongoing, testing the efficacy of a statin combined with an EGFR

inhibitor, erlotinib, in patients with squamous cell carcinomas and NSCLC was initiated

(NCT00966472). The statin being used in this clinical trial is rosuvastaitn, which is a

hydrophilic statins that does not passively diffuse through a plasma membrane but requires

carrier-mediated active transport for entry into hepatocytes. Hepatocytes express organic

anion transporting polypeptides (OATP) which facilitates the entry of hydrophilic statins such

as rosuvastatin and pravastatin into hepatocytes from the portal blood11. Rosuvastaitn is

therefore unlikely to reach cells of the extra-hepatic tissues including the solid tumour and


152

rarely used in cell culture studies to demonstrate anti-cancer efficacy. Although this fact does

not preclude anti-cancer efficacy of the rosuvastatin-erlotinib combination, as other clinical

pleiotropic effects of statins might contribute to anti-cancer efficacy, it might complicate

interpretation of results and confound subsequent biomarker discovery. Clinically, in addition

to their cholesterol lowering abilities, statins are known to exert anti-inflammatory12 and

immunomodulatory effects13. Taken together, there are still many obstacles in unifying cell

culture anti-cancer statin activity with clinical relevance.

On the other hand, a few clinical trials have sought to unify the preclinical studies with

relevant clinical correlates. The phase I study in AML patients where pravastatin was

administered with idaraubicin and cytarabine at pravastatin doses 40-1680mg/day14 was based

on the cell culture findings showing that inhibition of cholesterol synthesis sensitized AML

cells to chemotherapeutic treatment. AML cells upon chemotherapeutic treatment mounted

adaptive responses and increased their intracellular cholesterol15. The newly diagnosed

patients with unfavorable cytogenetics fared better than historical controls when evaluating

complete remission. Amongst the responders, LDL/cholesterol levels were lower than

baseline throughout therapy. In the cell culture study that found simvastatin reversed cell-

adhesion mediated drug resistance5, simvastatin was used at clinically achievable low

micromolar doses to treat 6 refractory multiple myeloma patients saw encouraging initial

responses as ascertained by decreases in paraprotein levels compared to historical controls 16.

A recent study found evidence of inhibition of geranylgeranylation in the bone marrow

mononuclear cells of one AML patient out of 4 following seven days of treatment with

simvastatin17. The above trials are encouraging in their demonstration, albeit limited, of

clinical correlates with efficacy, future efforts should be directed at uncovering molecular

indicators of statin activity within a tumour.


153

A recent window-of-opportunity trial in patients with primary invasive breast cancer

compared HMGCR expression in paired tumour samples before and after treatment with

standard doses of atorvastatin (80 mg/day)18. This pivotal study presented some key findings,

chiefly that post statin-treatment, increases in HMGCR levels in 68% of the samples were

observed. Interestingly, significant decreases in Ki67 expression following atorvastatin

treatment in the tumours with initial detectable HMGCR levels were observed. HMGCR

upregulation in response to atorvastatin treatment is quite suggestive of atorvastatin exerting a

direct anti-tumour effect in response to depletion of intracellular sterol pools as occurs upon

statin-mediated HMGCR inhibition19, 20. The clinical study did not indicate whether

cholesterol levels in patients were lowered following administration but presumably this

would have been the case. It is therefore also possible that HMGCR upregulation within the

tumour occurred due to decreased cholesterol levels in the plasma. However, the tumours

with the significant decreases in the Ki67 index were the ones with targetable HMGCR prior

to statin-treatment. Taken together, this study suggests that HMGCR-positive breast tumours

can be successfully targeted by statins. Other recent clinical trials have specifically started to

hone in on patient subpopulations and specific tumour types predicted to respond to statin

therapy. An ongoing trial (NCT00807950) where simvastatin will be pre-operatively

administered to patients with primary breast cancer is testing the hypothesis that simvastatin

will selectively affect tumours of the basal subtype21. The clinical trial intends to examine

genomic changes in the tumour following simvastatin treatment and correlate them with

biological effects. These clinical exploration into the availability of biomarkers to predict

statin response and assess statin-directed tumour effects are key to selecting statin responsive

patients in future clinical trials and increasing the chances of demonstrating statin clinical

efficacy.


154

Despite the rapid translation of statins into the clinic, there are gaps between cell

culture anti-cancer activity and clinical relevance. Many studies employ high statin

concentrations not clinically attainable or assume that statins are interchangeable despite their

clearly different pharmacological profiles. Furthermore studies using statins with more

favorable extra-hepatic and capable of reaching the tumour pharmacokinetic properties such

as fluvastatin have been extremely encouraging in demonstrating the benefits of using statins

in cancer patient care.

4.2 Potentiating statin-induced apoptosis – combinatorial approaches

In Chapter 2, we carried out a pharmacological screen intended to identify compounds

that potentiate the anti-cancer effects of statins. Challenges to successful chemotherapeutic

and targeted agent drug treatments, include innate and acquired resistance, redundancy of

signaling pathways, tumour heterogeneity, presence of feedback loops that activate other

pathways, and toxicity.

As elegantly summarized by Dr. Chou, a founder of multiple synergy algorithms extensively

used in pre-clinical biology, combining drugs has multiple potential benefits. Combining

drugs might increase therapeutic efficacy, decrease toxicity by being able to decrease the

dosage of one or two drugs while maintain the same efficacy and potentially retard drug

resistance22.

4.2.1 Statins and rationally crafted combinations

Statins have been combined with a multitude of currently available chemotherapeutics

as well as targeted agents. The combination of statins with reagents that are currently already


155

in the clinic and comprise the standard of care is a logical progression to broadening the

benefits of statins. In addition to potentiating the activity of already available anti-cancer

agents it has been important to evaluate the interaction of statins with many drugs as many

elderly caner patients receive statins for their cardiovascular problems. Although statins have

well-established safety profiles, the potential adverse reactions with other drugs administered

to already compromised patients necessitates their combinatorial evaluation.

Statins have been combined with anthracyclines, mainly doxorubicin, and synergism

and potentiation have been demonstrated in tissue culture23, 24 and murine cancer murine

xenograft models25, 26. The addition of lovastatin attenuated doxorubicin’s cardiotoxicity25 and

protected endothelial cells from doxorubicin’s cytotoxic effects27. The exact mechanisms of

synergy have been elusive and often confounded by contrasting actions of the individual

drugs on the same mediator. For example, it has been suggested that inhibition of the

activation of nuclear factors κβ (NF- κβ) by lovastatin contributed to lovastatin’s sensitization

to the cytotoxic effects of doxorubicin28, 29. However, it has also been reported that NF- κβ

activation by doxorubicin is essential for its cytotoxic effects30. Lovastatin’s inhibitory effects

on P-glycoprotein31 have also been implicated as being responsible for the synergy observed

when combining lovastatin with doxorubicin, a substrate for P-glycoprotein. Statins have

been similarly combined with an arsenal of other classical chemotherapeutic such as platinum

compounds24, 32, antimetabolites particularly 5-fluorouracil24 and tubulin-binding drugs such

as paclitaxel33, with synergy and potentiation being cell-dependent and mechanism of action

not fully elucidated.

Statins have also been combined with other drugs that target the MVA pathway,

namely the yet to be approved farnesyltransferase inhibitors (FTIs) and

geranygeranyltransferase inhibitors (GGTIs). Synergy between lovastatin and multiple FTIs

and GGTIs in inducing anti-cancer effects against multiple myeloma was previously


156

demonstrated34. Synergy with inhibitors of geranylgeranyl diphosphate synthase, the enzyme

upstream of geranylgernayltransferase which synthesizes geranygernaylpyrosphosphate

(GGPP), and lovastatin were similarly observed in leukemia cells35. Statins have also been,

combined with some targeted agents, and have shown to enhance the activity EGFR inhibitors,

gefitinib and erlotinib in glioblastoma36, vascular endothelial growth factor receptor (VEGFR)

inhibitors37, sorafenib in melanoma38, trastuzumab in breast cancer39, the m-TOR inhibitors in

leukemia40, imatinib in chronic myelogenous leukemia41.

Statins can successfully be combined with many agents that are currently approved for

clinical use in the treatment of cancer patients and some of them are even being evaluated in

the clinic. However, many patients will and have not responded to the classic

chemotherapeutic agents, alone or in combination with statins and the targeted agents are still

cost prohibitive. We therefore pursued an alternative approach of discovering novel

unexpected combinations. In chapter 2 we used of a pharmacological screen composed of

FDA-approved drugs for non-cancer indications and combined it with statins to uncover novel

combinations with anti-cancer activity. In Chapter 3, we undertook a genomic approach and

carried out an shRNA screen to discover novel targets/pathways whose knockdown combined

with inhibition of the MVA pathway maximized anti-tumour effects.

4.2.2 Statins and dipyridamole

The screen in chapter 2 carried out the in the KMS11 MM cell line uncovered

dipyridamole, a drug that has been used for decades in the clinic as an antithrombotic agent.

The MM KMS11 cell line was chosen for the screen because it was a relatively statin

sensitive cell line42, 43. The rationale in choosing this cell line was that enhancing tumour kill

in a cell line that already responds to statins will further maximize tumour kill.


157

Strong preclinical activity was demonstrated using multiple MM and AML cell lines.

Interestingly, in the LP1 MM cell line, the statin-dipyridamole combination demonstrated

robust synergy and dipyridamole-mediated potentiation of the statins’ anti-cancer effects

indicating that the combination has the potential to target tumour cell intrinsically insensitive

to statin-induced apoptosis.

Primary AML and MM patients were also susceptible to the combination and in vivo

tumour delaying activity was demonstrated in OCI-AML2 xenografts. In attempts to

phenocopy dipyridamole’s many molecular effects we utilized pharmacological inhibitors

representative of each class and also combined them with statins. With the exception of

cilostazol, the other compounds representing nucleoside and glucose transport inhibitors, and

the cyclic guanosine monophosphate (cGMP) elevating phopshodiesterase (PDE) inhibitors

did not the potentiate statin induced anti-cancer effects. As other PKA activating, cyclic

adenosine monophosphate (cAMP) elevating agents such forskolin also significantly

potentiated statin-induced apoptosis in AML cells and primary patient samples, that activation

of that signaling arm is implicated in contributing to the synergistic effect observed. We thus

identified another class of drugs, the cAMP elevating PDE inhibitors such as cilostazol also

capable of inducing tumour apoptosis when combined with statins. Cilostazol is also an

antiplatelet agent currently in clinical use.

As discussed in Chapter 2, dipyridamole’s inhibitory activities exerted upon P-

glycoprotein and the nucleoside transporter inhibitors have been of interest when combined

with standard chemotherapeutic drugs whose actions are dampened by the expression of P-

glycoprotein and nucleoside salvage metabolism. The early clinical trials based on these

principles of dipyridamole activity in combination chemotherapy in solid tumours were not

successful in demonstrating efficacy44-47 and some of the failure was attributed to

dipyridamole’s binding to α1-acid-glycoprotein (AGP) in serum48. However, as novel and


158

better dipyridamole formulations have since entered the market49,50 there is likely to be a

resurgence of this drug for used other than those in the cardiovascular setting.

Indeed in recent years there has been a resurgence of dipyridamole being used as an

anti-cancer agent particularly in breast cancer. One study treated MMTV-PyMT transgenic

mice, which develop a highly aggressive breast cancer, with 10 mg/kg of dipyridamole for

almost three months and found that dipyridamole exhibited chemopreventive activities against

tumorigenesis and metastasis51. Another group which administered dipyridamole to breast

cancer tumour bearing xenografted mice found that the drug decreased primary tumour

growth, metastasis, macrophage tumour infiltration and resulted in decreased activation of the

Wnt, ERK1/2-MAPK and NF-kB pathways52. The mechanism by how dipyridamole

perturbed those pathways was not elucidated.

4.2.3 Limitations and future directions of this work

Although we attributed the involvement of cAMP elevation and PKA activation as

being partly responsible for the mechanism by which dipyridamole potentiates statin-induced

apoptosis, we did not demonstrate the functional consequences of the reverse. Over-

expression of PDE’s, knockdown of PKA or abrogation of its activity would have

strengthened the significance of our findings and this work is underway. We have also

demonstrated that the statin-dipyridamole combination abrogates the statin-induced

upregulation of HMGCR and this is also being further explored especially since we have

demonstrated in Chapter 3 the functional consequences of this abrogation via knockdown of

SREBF2.

The frequent utilization of the MTS and MTT assays to assess cell viability limits

interpretation and does not measure apoptosis. The calorimetric assays, such as the MTS and


159

MTT, although cost-effective and relatively quick, should only represent a first pass attempt

at discovering potential anti-cancer effects. Although in the current work, validation of hits

was accompanied by apoptosis assays, utilizing these assays for cancer drug screening

purposes should be approached with caution as it can potentially result in the pursuit of hits

that are anti-proliferative and do not cause tumour cell apoptosis.

The use of pharmacological modulators of cAMP signaling to mimic the statin-

dipyridamole combination and prove its significance in modulating the synergistic interaction

is also problematic. Although the pharmacological inhibitors were used at concentration

shown in the literature to elicit the intended effects, they are likely not specific and like

dipyridamole have other modes of action. It is very difficult to accurately ascertain by which

mechanism two drugs interact to yield the observed net effect especially when drugs have

multiple mechanisms of action. As cautioned by Dr. Chou, it is rather improbable to ever fully

ascertain the mechanism of synergy and especially to demonstrate it clinically53, especially

since rarely do we ever fully know the mechanism of a single drug at the biochemical,

molecular, cellular and pharmacological level22, 54. When two drugs are combined, there may

be several hundred steps occurring from time of treatment to death or cycle arrest of a cell.

However, the manner in which well-characterized pathways are perturbed by either drug or

their concomitant addition is a good method to exclude certain possibilities and narrow down

through which signaling cascades effects might be occurring.

There are limitations with the in vivo model used to assess drug efficacy. The statin-

dipyridamole combination merely delayed tumour growth and did not regress or keep the

tumour volume constant. In a clinical situation, it might therefore be difficult to translate this

result into therapeutic efficacy. On the other hand, the dosing regimen could be further

optimized. Animals were treated once daily, and as the half-life of dipyridamole is less than

six hours, more frequent administration might result in improved in vivo efficacy. The in vivo


160

studies involved the use of xenografts implanted in immune-deficient animals. There are

limitations to using this murine model to predict in vivo efficacy. Firstly, statins exert many

immunomodulatory effects, most of them immunosuppressive and the NOD SCID mice have

impaired T and B lymphocyte development, a feature which prevents them from rejecting

human cell line xenografts but therefore does not take into account key statin-mediated

clinical responses. Furthermore, it is difficult to modulate cholesterol levels in these mice and

special transgenic models are required55 to effectively lower cholesterol levels using statins.

Again, as this is a primary mode of statin activity, anti-tumour contribution resulting from

lowering cholesterol may be missed when using immunodeficient mice. Furthermore, AML is

not a disease where solid tumours are formed and therefore more sophisticated models are

required to recapitulate the disease. Primarily future work will focus on evaluating whether

the statin-dipyridamole combination can effectively target the leukemic cancer stem (LCS)

cell. Others56, 57 and we58 have demonstrated that statins can effectively target the LCS.

Engraftment potential of CD34 purified primary AML samples injected into the femur of

sublethally irradiated NOD SCID mice, followed by weeks of treatment of the mice with

statin, dipyridamole or the combination will be evaluated.

More broadly, another general concern in utilizing xenografts and even some of the

more sophisticated transgenic spontaneous tumour models is that drug efficacy is being tested

against primary tumours. However, many cancer patients, especially in clinical trials, present

advanced metastatic disease that has already failed primary therapy. Many of the experimental

drugs that have fared well preclinically, fail in clinical trials because of inadequate preclinical

models that do not recapitulate metastatic disease. Researchers have sought to addres this

question by resecting the primary tumours in mice to enable the progression of metastatic

disease and have successfully tested drugs in this setting59, 60.


161

4.3 Targeting the MVA pathway – unbiased knockdown approaches

In Chapter 3, pan genome-wide shRNA screen identified several hits that were further

pursued as putative potentiators of statin induced anti-cancer effects. Although the screen was

carried out over several passages and doubling times, putative drop-out shRNA hits were

successfully verified using shorter term assays.

4.3.1 The MVA pathway’s other targets

The presence of several hits from the MVA pathway emphasized its importance in

being targeted as an anti-cancer therapeutic strategy. HMGCR is the rate-limiting enzyme of

the MVA pathway and its inhibition will lead to a depletion of all downstream end products.

Inhibition of other nodal points of the MVA pathway will lead to a complex interplay of

accumulation and inhibition of its various intermediate products and potentiation is not

always obvious. For instance treating with a statin will lead to depletion of farnesyl

pyrophosphate (FPP) but this might be negated if GGPS1 is simultaneously blocked because

it will lead to an accumulation of FPP. Therefore if the statin induced anti-cancer effect is

mediated by blocking ras farnesylation for instance then synergy is unlikely. We

demonstrated that knockdown of GGPS1 dramatically potentiated statin-induced apoptosis.

However, the cells lines stably expressing the shRNA targeting GGPS1 were difficult to

generate, highlighting the overall importance of this target for maintaining cancer cell growth.

Inhibiting GGPS1 will affect a largest amount of proteins when targeting just one of the

downstream geranygernayltransferases. An independent group that has synthesized

pharmacological inhibitors to GGPS1 has also noted dramatic synergy of the GGPS1

inhibitors when combined with statins in inducing tumour cell kill in leukemia cells35.


162

The identification and validation of the transcription factor sterol regulatory element-

binding protein 2 (SREBF2) as potentiating stain-induced apoptosis in several tumour types

provides strong evidence for the functional consequences of the restorative sterol feedback

loop. The stably expressing shSREBF2 lung and breast cancer sublines were characterized by

similar growth rates but underwent a dramatic induction of apoptosis upon treatment with

sublethal doses of fluvastatin when compared to the control sublines. Despite residual

SREBF2 levels (mRNA knockdown was approximately 85% effective), the shSREBF2

sublines lacked the ability to upregulate HMGCR and HMGCS1 levels in response to

fluvastatin exposure within 24 hours, an effect which likely blunts the effects of the statins in

the control sublines.

4.3.2 Limitations and future directions of this work

We have identified numerous other novel hits such as phosphatidylinositol 4-kinase,

catalytic beta (PIK4B) that were validated in both breast and lung cancer cells using genomic

and pharmacological approaches. The sensitizing effects upon targeting PI4KB to statin

induced apoptosis should be further mechanistically explored. Net effects of the combination

of pathways known to be perturbed by inhibiting either targets such as the Raf/MEK/ERK and

Akt pathways will be examined. Furthermore, as pharmacological inhibitors to PIK4B are

available in vivo efficacy and safety of the combination can be evaluated.

Although we have shown that SREBF2 knockdown strongly potentiates statin induced

apoptosis, the effects and potential toxicity to normal cells has not been explored. There are

no known specific inhibitors of SREBF2 and as such, this question cannot be readily


163

answered using murine in vivo models. The question can perhaps be addressed by cell culture

studies using normal primary cells.

HMGCR is the rate-limiting enzymes of the MVA pathway and its inhibition, by

definition, should render inhibition of enzymes downstream of it such as GGPS1

inconsequential. However, as we clearly observe strong potentiation upon knockdown of

GGPS1 and inhibition of HMGCR using statins, this suggests that the degree of HMGCR

inhibition was sub-optimal and further inhibition at the level of GGPS1 was able to cooperate

with statins to further block this arm of the pathway. This should be directly investigated. For

example, the level of mevalonate and/or other end products such as GGPP before and after

statin inhibition could be measured using high-perfomarnce liquid chromatography. By

assessing the mevalonate flux, a clearer idea about the kinetics and degree of HMGCR

inhibition will emerge and answer key questions pertaining to whether knockdown of the

GGPS1, HMGCS1 and SREBF2 in combination with statins is due to MVA pathway

inhibition and/or potential involvement of other molecular pathways perturbed in response to

GGPS1, HMGCS1 and SREBF2 knockdown.


164

4.4 Concluding remarks and anticipated impact

In this body of work we have identified a translatable combination of safe and readily

available FDA approved drugs, statins and dipyridamole which have the potential to directly

impact patient care. We have demonstrated strong preclinical antimyeloma and antileukemia

efficacy of the combination. Through our investigations into uncovering the mechanism by

which dipyridamole potentiates statin-induced apoptosis, we have identified an additional

class of drugs, the cAMP elevating PDE inhibitors, some of which are also clinically

approved, with antileukemia activity. The targets uncovered through the shRNA screen have

not only increased our understanding of the key branches of the MVA pathway that maximize

statin induced anti-tumour effects but have also uncovered a plethora of novel target kinases

such as PIK4B and MAP2K4 that can be further assessed in the future for preclinical efficacy.


165

4.5 References:

1. Wu J, Wong WW, Khosravi F, Minden MD, Penn LZ. Blocking the Raf/MEK/ERK pathway sensitizes acute myelogenous leukemia cells to lovastatin-induced apoptosis. Cancer Res 2004; 64(18): 6461-8.

2. Campbell MJ, Esserman LJ, Zhou Y, Shoemaker M, Lobo M, Borman E et al. Breast

cancer growth prevention by statins. Cancer Res 2006; 66(17): 8707-14. 3. Koyuturk M, Ersoz M, Altiok N. Simvastatin induces apoptosis in human breast

cancer cells: p53 and estrogen receptor independent pathway requiring signalling through JNK. Cancer letters 2007; 250(2): 220-8.

4. Gopalan A, Yu W, Sanders BG, Kline K. Simvastatin inhibition of mevalonate

pathway induces apoptosis in human breast cancer cells via activation of JNK/CHOP/DR5 signaling pathway. Cancer Lett 2013; 329(1): 9-16.



6. Chen J, Lan T, Hou J, Zhang J, An Y, Tie L et al. Atorvastatin sensitizes human non-

small cell lung carcinomas to carboplatin via suppression of AKT activation and upregulation of TIMP-1. The international journal of biochemistry & cell biology 2012; 44(5): 759-69.

7. Hwang KE, Na KS, Park DS, Choi KH, Kim BR, Shim H et al. Apoptotic induction

by simvastatin in human lung cancer A549 cells via Akt signaling dependent down-regulation of survivin. Investigational new drugs 2011; 29(5): 945-52.

8. Klawitter J, Shokati T, Moll V, Christians U, Klawitter J. Effects of lovastatin on

breast cancer cells: a proteo-metabonomic study. Breast cancer research : BCR 2010; 12(2): R16.

9. Zhu Y, Casey PJ, Kumar AP, Pervaiz S. Deciphering the signaling networks

underlying simvastatin-induced apoptosis in human cancer cells: evidence for non-canonical activation of RhoA and Rac1 GTPases. Cell death & disease 2013; 4: e568.

10. Zhao TT, Le Francois BG, Goss G, Ding K, Bradbury PA, Dimitroulakos J.

Lovastatin inhibits EGFR dimerization and AKT activation in squamous cell carcinoma cells: potential regulation by targeting rho proteins. Oncogene 2010; 29(33): 4682-92.

11. Gazzerro P, Proto MC, Gangemi G, Malfitano AM, Ciaglia E, Pisanti S et al.

Pharmacological actions of statins: a critical appraisal in the management of cancer. Pharmacological reviews 2012; 64(1): 102-46.


166

12. Greenwood J, Mason JC. Statins and the vascular endothelial inflammatory response.

Trends in immunology 2007; 28(2): 88-98. 13. Khattri S, Zandman-Goddard G. Statins and autoimmunity. Immunologic research

2013. 14. Kornblau SM, Banker DE, Stirewalt D, Shen D, Lemker E, Verstovsek S et al.




16. Schmidmaier R, Baumann P, Bumeder I, Meinhardt G, Straka C, Emmerich B. First

clinical experience with simvastatin to overcome drug resistance in refractory multiple myeloma. Eur J Haematol 2007; 79(3): 240-3.

17. van der Weide K, de Jonge-Peeters S, Huls G, Fehrmann RS, Schuringa JJ, Kuipers F

et al. Treatment with high-dose simvastatin inhibits geranylgeranylation in AML blast cells in a subset of AML patients. Experimental hematology 2012; 40(3): 177-186 e6.



19. Shimomura I, Bashmakov Y, Shimano H, Horton JD, Goldstein JL, Brown MS.

Cholesterol feeding reduces nuclear forms of sterol regulatory element binding proteins in hamster liver. Proceedings of the National Academy of Sciences of the United States of America 1997; 94(23): 12354-9.

20. Brown MS, Goldstein JL. The SREBP pathway: regulation of cholesterol metabolism

by proteolysis of a membrane-bound transcription factor. Cell 1997; 89(3): 331-40. 21. Kumar AS, Benz CC, Shim V, Minami CA, Moore DH, Esserman LJ. Estrogen

receptor-negative breast cancer is less likely to arise among lipophilic statin users. Cancer Epidemiol Biomarkers Prev 2008; 17(5): 1028-33.

22. Chou TC. Theoretical basis, experimental design, and computerized simulation of

synergism and antagonism in drug combination studies. Pharmacological reviews 2006; 58(3): 621-81.

23. Goard CA, Mather RG, Vinepal B, Clendening JW, Martirosyan A, Boutros PC et al.

Differential interactions between statins and P-glycoprotein: implications for exploiting statins as anticancer agents. International journal of cancer. Journal international du cancer 2010; 127(12): 2936-48.


167

24. Roudier E, Mistafa O, Stenius U. Statins induce mammalian target of rapamycin (mTOR)-mediated inhibition of Akt signaling and sensitize p53-deficient cells to cytostatic drugs. Mol Cancer Ther 2006; 5(11): 2706-15.

25. Feleszko W, Mlynarczuk I, Balkowiec-Iskra EZ, Czajka A, Switaj T, Stoklosa T et al.

Lovastatin potentiates antitumor activity and attenuates cardiotoxicity of doxorubicin in three tumor models in mice. Clin Cancer Res 2000; 6(5): 2044-52.

26. Feleszko W, Mlynarczuk I, Olszewska D, Jalili A, Grzela T, Lasek W et al. Lovastatin

potentiates antitumor activity of doxorubicin in murine melanoma via an apoptosis-dependent mechanism. Int J Cancer 2002; 100(1): 111-8.

27. Damrot J, Nubel T, Epe B, Roos WP, Kaina B, Fritz G. Lovastatin protects human

endothelial cells from the genotoxic and cytotoxic effects of the anticancer drugs doxorubicin and etoposide. British journal of pharmacology 2006; 149(8): 988-97.

28. Arlt A, Vorndamm J, Breitenbroich M, Folsch UR, Kalthoff H, Schmidt WE et al.

Inhibition of NF-kappaB sensitizes human pancreatic carcinoma cells to apoptosis induced by etoposide (VP16) or doxorubicin. Oncogene 2001; 20(7): 859-68.

29. Das KC, White CW. Activation of NF-kappaB by antineoplastic agents. Role of

protein kinase C. J Biol Chem 1997; 272(23): 14914-20. 30. Ashikawa K, Shishodia S, Fokt I, Priebe W, Aggarwal BB. Evidence that activation of

nuclear factor-kappaB is essential for the cytotoxic effects of doxorubicin and its analogues. Biochem Pharmacol 2004; 67(2): 353-64.

31. Wang E, Casciano CN, Clement RP, Johnson WW. HMG-CoA reductase inhibitors

(statins) characterized as direct inhibitors of P-glycoprotein. Pharmaceutical research 2001; 18(6): 800-6.

32. Agarwal B, Bhendwal S, Halmos B, Moss SF, Ramey WG, Holt PR. Lovastatin

augments apoptosis induced by chemotherapeutic agents in colon cancer cells. Clin Cancer Res 1999; 5(8): 2223-9.

33. Holstein SA, Hohl RJ. Synergistic interaction of lovastatin and paclitaxel in human

cancer cells. Mol Cancer Ther 2001; 1(2): 141-9. 34. Morgan MA, Sebil T, Aydilek E, Peest D, Ganser A, Reuter CW. Combining

prenylation inhibitors causes synergistic cytotoxicity, apoptosis and disruption of RAS-to-MAP kinase signalling in multiple myeloma cells. British journal of haematology 2005; 130(6): 912-25.

35. Dudakovic A, Wiemer AJ, Lamb KM, Vonnahme LA, Dietz SE, Hohl RJ. Inhibition

of geranylgeranyl diphosphate synthase induces apoptosis through multiple mechanisms and displays synergy with inhibition of other isoprenoid biosynthetic enzymes. The Journal of pharmacology and experimental therapeutics 2008; 324(3): 1028-36.


168

36. Cemeus C, Zhao TT, Barrett GM, Lorimer IA, Dimitroulakos J. Lovastatin enhances gefitinib activity in glioblastoma cells irrespective of EGFRvIII and PTEN status. J Neurooncol 2008; 90(1): 9-17.

37. Zhao TT, Trinh D, Addison CL, Dimitroulakos J. Lovastatin inhibits VEGFR and

AKT activation: synergistic cytotoxicity in combination with VEGFR inhibitors. PloS one 2010; 5(9): e12563.

38. Zhang S, Doudican NA, Quay E, Orlow SJ. Fluvastatin enhances sorafenib

cytotoxicity in melanoma cells via modulation of AKT and JNK signaling pathways. Anticancer research 2011; 31(10): 3259-65.

39. Budman DR, Tai J, Calabro A. Fluvastatin enhancement of trastuzumab and classical

cytotoxic agents in defined breast cancer cell lines in vitro. Breast Cancer Res Treat 2007; 104(1): 93-101.

40. Calabro A, Tai J, Allen SL, Budman DR. In-vitro synergism of m-TOR inhibitors,

statins, and classical chemotherapy: potential implications in acute leukemia. Anticancer Drugs 2008; 19(7): 705-12.

41. Oh B, Kim TY, Min HJ, Kim M, Kang MS, Huh JY et al. Synergistic killing effect of

imatinib and simvastatin on imatinib-resistant chronic myelogenous leukemia cells. Anticancer Drugs 2013; 24(1): 20-31.

42. Wong WW, Clendening JW, Martirosyan A, Boutros PC, Bros C, Khosravi F et al.

Determinants of sensitivity to lovastatin-induced apoptosis in multiple myeloma. Mol Cancer Ther 2007; 6(6): 1886-97.



44. Zaniboni A, Simoncini E, Marpicati P, Montini E, Palazzi M, Pipino G et al. 5-

Fluorouracil and dipyridamole in metastatic breast cancer: a pilot study. J Chemother 1989; 1(4): 266-8.

45. Buzaid AC, Alberts DS, Einspahr J, Mosley K, Peng YM, Tutsch K et al. Effect of

dipyridamole on fluorodeoxyuridine cytotoxicity in vitro and in cancer patients. Cancer Chemother Pharmacol 1989; 25(2): 124-30.

46. Tsavaris N, Kosmas C, Polyzos A, Genatas K, Vadiaka M, Paliaros P et al.

Leucovorin + 5-fluorouracil plus dipyridamole in leucovorin + 5-fluorouracil-pretreated patients with advanced colorectal cancer: a pilot study of three different dipyridamole regimens. Tumori 2001; 87(5): 303-7.

47. Remick SC, Grem JL, Fischer PH, Tutsch KD, Alberti DB, Nieting LM et al. Phase I

trial of 5-fluorouracil and dipyridamole administered by seventy-two-hour concurrent continuous infusion. Cancer Res 1990; 50(9): 2667-72.


169

48. Fischer PH, Willson JK, Risueno C, Tutsch K, Bruggink J, Ranhosky A et al. Biochemical assessment of the effects of acivicin and dipyridamole given as a continuous 72-hour intravenous infusion. Cancer Res 1988; 48(19): 5591-6.

49. Aggrenox (aspirin/extended-release dipyridamole) 25 mg/200mg capsules

[Prescribing information]. Biberach Germany: Boehringer-Ingelheim; 2004. . In. 50. Derendorf H, VanderMaelen CP, Brickl RS, MacGregor TR, Eisert W. Dipyridamole

bioavailability in subjects with reduced gastric acidity. J Clin Pharmacol 2005; 45(7): 845-50.

51. Wang C, Schwab LP, Fan M, Seagroves TN, Buolamwini JK. Chemoprevention

Activity of Dipyridamole in the MMTV-PyMT Transgenic Mouse Model of Breast Cancer. Cancer Prev Res (Phila) 2013; 6(5): 437-47.

52. Spano D, Marshall JC, Marino N, De Martino D, Romano A, Scoppettuolo MN et al.

Dipyridamole prevents triple-negative breast-cancer progression. Clinical & experimental metastasis 2013; 30(1): 47-68.

53. Chou TC. Preclinical versus clinical drug combination studies. Leuk Lymphoma 2008;

49(11): 2059-80. 54. Chou TC, Talalay P. Quantitative analysis of dose-effect relationships: the combined

effects of multiple drugs or enzyme inhibitors. Adv Enzyme Regul 1984; 22: 27-55. 55. Zaragoza C, Gomez-Guerrero C, Martin-Ventura JL, Blanco-Colio L, Lavin B,

Mallavia B et al. Animal models of cardiovascular diseases. Journal of biomedicine & biotechnology 2011; 2011: 497841.

56. Sachlos E, Risueno RM, Laronde S, Shapovalova Z, Lee JH, Russell J et al.

Identification of drugs including a dopamine receptor antagonist that selectively target cancer stem cells. Cell 2012; 149(6): 1284-97.

57. Misaghian N, Ligresti G, Steelman LS, Bertrand FE, Basecke J, Libra M et al.

Targeting the leukemic stem cell: the Holy Grail of leukemia therapy. Leukemia 2009; 23(1): 25-42.

58. Dimitroulakos J, Nohynek D, Backway KL, Hedley DW, Yeger H, Freedman MH et


59. Guerin E, Man S, Xu P, Kerbel RS. A model of postsurgical advanced metastatic

breast cancer more accurately replicates the clinical efficacy of antiangiogenic drugs. Cancer research 2013; 73(9): 2743-8.

60. Hackl C, Man S, Francia G, Milsom C, Xu P, Kerbel RS. Metronomic oral topotecan

prolongs survival and reduces liver metastasis in improved preclinical orthotopic and adjuvant therapy colon cancer models. Gut 2013; 62(2): 259-71.


170

Publications (2008-2013):

1. Clendening JW, Pandyra A, Li Z, Boutros PC, Martirosyan A, Lehner R, Jurisica I, Trudel

S, Penn LZ. Exploiting the mevalonate pathway to distinguish statin-sensitive multiple

myeloma. Blood. 2010 Jun 10;115(23):4787-97.

2. Clendening JW, Pandyra A, Boutros PC, El Ghamrasni S, Khosravi F, Trentin GA,

Martirosyan A, Hakem A, Hakem R, Jurisica I, Penn LZ. Dysregulation of the mevalonate

pathway promotes transformation. Proc Natl Acad Sci U S A. 2010 Aug 24;107(34):15051-

6.

3. Lang PA*, Xu HC*, Grusdat M, McIlwain DR, Pandyra AA, Harris IS, Shaabani N, Honke

N, Maney SK, Lang E, Pozdeev VI, Recher M, Odermatt B, Brenner D, Häussinger D,

Ohashi PS, Hengartner H, Zinkernagel RM, Mak TW, Lang KS. Reactive oxygen species

delay control of lymphocytic choriomeningitis virus. Cell Death Differ. 2013 Apr;20(4):649-

58.

4. Lang KS*, Lang PA*, Meryk A*, Pandyra AA*, Boucher LM, Pozdeev VI, Tusche MW,

Göthert JR, Haight J, Wakeham A, You-Ten AJ, McIlwain DR, Merches K, Khairnar V,

Recher M, Nolan GP, Hitoshi Y, Funkner P, Navarini AA, Verschoor A, Shaabani N, Honke

N, Penn LZ, Ohashi PS, Häussinger D, Lee KH, Mak TW. Involvement of Toso in activation

of monocytes, macrophages, and granulocytes. Proc Natl Acad Sci U S A. 2013 Feb

12;110(7):2593-8.


171

5. Wasylishen AR, Kalkat M, Kim SS, Pandyra A, Chan PK, Oliveri S, Sedivy E, Konforte D,

Bros C, Raught B, Penn LZ. MYC activity is negatively regulated by a C-terminal lysine

cluster. Oncogene. 2013 Feb 25. doi: 10.1038/onc.2013.36.

6. Pandyra AA, Mullen PJ, Pong JT, Li Z, Trudel S, Minden MD, Schimmer AD, and Penn

LZ. Combining statins and dipyridamole effectively targets acute myelogenous leukemia and

multiple myeloma (submitted).

7. Pandyra AA, Goard CA, Pong JP, Mullen PJ, Ericson E, Gebbia M, Brown K, Nieslow C,

Moffat J, Penn LZ. Genome wide shRNA screen reveals novel sensitizers of statin-induced

apoptosis. (in preparation).

Documents

Pandyra Aleksandra A 201311 PhD thesis · Thesis – Aleksandra Pandyra iv Acknowledgements First, I would like to thank my supervisor, Dr. Linda Penn, for the opportunity to work