Targeting energy metabolism in colorectal cancer

Targeting energy metabolism in colorectal cancer

by

Nelson Ho

A Thesis presented to

The University of Guelph

In partial fulfillment of requirements for the degree of

Doctor of Philosophy in

Biomedical Sciences

Guelph, Ontario, Canada

© Nelson Ho, August, 2015

ABSTRACT

TARGETING ENERGY METABOLISM IN COLORECTAL CANCER

Nelson Ho Advisor: University of Guelph, 2015 Dr. B. L. Coomber

A switch to aerobic glycolysis regardless of oxygen tension is a metabolic

phenotype observed in a variety of tumors, known as the Warburg effect, and is a

possible target in cancer therapy. The abnormal tumor vasculature leads to a

heterogeneous microenvironment with areas of hypoxia, acidity, and low nutrient supply.

This thesis investigated the effects of the tumor microenvironment on the expression of

glycolytic enzymes and on the efficacy of compounds targeting them. The first set of

experiments examined the molecular and metabolic changes associated with

dichloroacetate-induced cytoprotection in different human colorectal cancer (CRC) cell

lines under anoxic conditions. There was evidence of differential regulation of pyruvate

dehydrogenase phosphorylation between different cells leading to altered mitochondrial

activity following dichloroacetate exposure.

Hexokinase II (HKII) is a glycolytic enzyme often upregulated in tumors. 3-

bromopyruvate (3BP) is an alkylating agent that shows promising anti-cancer effects in

various cancer models and HKII has been suggested to be one of its principal targets. The

efficacy of 3BP against CRC cells was assessed under different media glucose

concentrations, and it was determined that 3BP was effective in reducing cell growth

through alterations in AKT signaling and that media glucose availability played a role in

cytotoxic sensitivity to 3BP. However, knockdown in HKII expression did not alter 3BP

sensitivity in different CRC cell lines.

Finally, HKII expression was assessed in colorectal cancer tissues and correlated

with clinical parameters and patient outcome. Dual immunofluorescence was used to co-

localize HKII with a marker of ischemia, carbonic anhydrase IX in 60 cases of human

CRC. HKII expression was found only in non-ischemic areas of the tumors. While HKII

expression did not correlate with any measured clinicopathlogical characteristics, patients

with low HKII-expressing tumors had a worse prognosis than those with high HKII-

expressing tumors.

Overall, this thesis presents the effects of the tumor microenvironment on both

glycolytic enzyme expression and therapeutic efficacy. This research promises to

improve our understanding of tumor metabolism in the heterogeneous microenvironment

in order to better apply this field of cancer biology for therapeutic use.

iv

ACKNOWLEDGEMENTS

A dissertation of this magnitude cannot be completed without the help of many. With that in mind, there are a few things I would like to say and a few people I would like to thank before half the words don’t make sense to a general audience.

To my supervisor, Dr. Brenda Coomber. I recently looked back at the first e-mail I sent you back in November of 2009, asking to see if there would be a position in your laboratory 10 months down the road. To be quite honest, reading the e-mail was quite cringe-worthy and not my best work, yet you still took a chance on me. Here we are, over 5 years later. I can’t thank you enough for all the guidance, wisdom, and support that you have provided me over the years. Your door was always open, and you were more than willing to listen to your students, whether it be a good or bad idea rolling off our tongues. The wealth of knowledge you possess regardless of topic is astonishing.

I would also like to thank the members of my advisory committee: Dr. James

Kirkland, Dr. Roger Moorehead, and Dr. Ray Lu. You all provided valuable insight and suggestions that helped shape my research and dissertation. Beyond faculty members, I would like to thank Betty-Anne McBey who played a crucial role in the early stages of my flow cytometry analysis in her training and technical support.

I would like to thank the Department of Biomedical Sciences. It was a pleasure working in an environment filled with such inspiring minds. I would like to thank the incredible office staff: Kim, Frances, Sally, and Wendy. You all kept the ship afloat and were always available for a quick social break.

I would like to thank my officemates whom I have shared many laughs with both in and out of the office. In particular, I would like to thank Amanda and Emily for putting up with my shenanigans, especially during social outings. Graham, I never would have thought back in 1st year of undergrad when I met you that we would be sharing this PhD road together, in the same department no less. Your faith in my fantasy hockey knowledge to make me your assistant GM was appreciated. I also would like to thank you for not thinking I was too terrible in volleyball to put up with me as a teammate for all these years. Lindsay, without going into too much detail, you were my partner in crime and catalyzed me interpersonally by providing me with the platforms to highlight some of my more “artistic” abilities shall we say? I would like to thank all the 4th year project students and summer research students whom I had the pleasure of supervising – Rebecca Theal, Stacey Butler, Gaelan Melanson, and Andreza Silva. I thoroughly enjoyed being a mentor to you all and I hope you enjoyed your experience. I would like to thank all the past and current lab members of the Coomber, Viloria-Petit, and Mutsaers labs for providing such a friendly and engaging working environment. To Amanda, Kaya, Nathan, Leanne, Geordon, Amy, Sonja, Divya and Richard, the “old” crew. I will never forget the memories we shared in the confines of the

v

lab or the shenanigans that would ensue following a long week. The recent closing of Vinyl may be symbolic to an end of our reign. Barbs, I am not sure how better to thank you for your friendship than by saying “Ignition Remix”. Kaya, thank you for showing me the ropes when I first came into the lab and for passing down the PhD torch to me. Geo, you are the model of perseverance and will never be a duster in my book. Rich, you are an inspiration with what you have accomplished since I first met you and I am proud to have you as a friend. To my Lab Mom, Jodi Morrison. No words can describe the appreciation I have for you as a scientist and as a human being. For every bit of success or failure, you have been there for me. We were models of procaffination and I will never forget the stories we shared over those times. You were always there to answer any of my questions even though I tried my best to act like the oldest son and not bother you. I wish nothing but the best for you and your family moving forward. I would like to thank my family. To my mom, you have been an inspiration throughout my life and while you still don’t understand my research over these years, just know that your hard working attitude and perseverance has rubbed off on me. To my sister, Vanessa, thank you for taking care of mom while I have been away. Although we don’t see each other much anymore, just know that I will always be there for you. To Rico, I never thought I could love a dog as much as you, whatever mix you may truly be. Finally, I would like to thank Kate. You are my best friend, my confidant, and my biggest supporter. You never doubt my inspirations and have full faith in whatever decisions I make. I am truly blessed to have you in my life. I love you so much.

In a special tribute, I would like to thank my dad, Lawrence. You left us over 9 years ago, but I will be forever grateful for how you provided for your family even though times were never easy. You have been my inspiration over these last 5 years, and the reason for my choice to study cancer. I know you look down on me everyday and I hope that you are proud of the man that I have become. So dad, I dedicate this work to you.

Love, Son.

vi

DECLARATION OF WORK PERFORMED

I declare that with the exception of the items indicated below, all work reported in

this thesis was performed by me.

Under my supervision, Andreza Silva performed media-glucose reduction on

HCT116 and DLD-1 cell lines.

vii

TABLE OF CONTENTS

ACKNOWLEDGEMENTS ............................................................................................ iv

DECLARATION OF WORK PERFORMED .............................................................. vi LIST OF TABLES ............................................................................................................ x

LIST OF FIGURES ......................................................................................................... xi LIST OF ABBREVIATIONS ........................................................................................ xii

INTRODUCTION ............................................................................................................ 1 CHAPTER I – REVIEW OF THE LITERATURE ...................................................... 3

Glucose metabolism ...................................................................................................... 3 Warburg effect .............................................................................................................. 3

Advantages of the Warburg effect to cancer cells ...................................................... 6 Regulation of glycolytic enzymes ................................................................................. 9

Hexokinase ................................................................................................................ 12 Pyruvate dehydrogenase kinase ................................................................................ 14

Metabolic regulation by oncogenes and tumor suppressors ................................... 16 Warburg effect as a therapeutic target ..................................................................... 19

Dichloroacetate ......................................................................................................... 22 3-bromopyruvate ....................................................................................................... 25

Tumor microenvironment .......................................................................................... 28 Hypoxia ..................................................................................................................... 29 HIF1 Regulation ........................................................................................................ 31 Hypoglycemia ........................................................................................................... 32

RATIONALE .................................................................................................................. 35 CHAPTER II – EFFECTS OF DCA ON ANOXIC HUMAN COLORECTAL CANCER CELLS ........................................................................................................... 37

Introduction ................................................................................................................. 37 Materials and Methods ............................................................................................... 39

Cell culture ................................................................................................................ 39 In vitro exposure to anoxia and/or DCA ................................................................... 40 Measurement of cell growth ..................................................................................... 40 qRT-PCR for PDK mRNA expression ..................................................................... 40 Protein isolation and western immunoblot analysis .................................................. 41 FACS analysis of mitochondrial function ................................................................. 42 Glucose consumption and lactate production ........................................................... 43 Statistical analysis ..................................................................................................... 43

Results .......................................................................................................................... 44 DCA reduces both normoxic and anoxic cell growth ............................................... 44 PDKs 1 and 3 are highly expressed in human CRC cells ......................................... 44 PDKs 1 and 3 are upregulated in response to treatment ........................................... 45 PDH phosphorylation status varies in response to treatment between cell lines ...... 45 Mitochondrial activity varies in response to DCA between human CRC cells ........ 46

viii

DCA does not significantly alter metabolic status of human CRC cells .................. 47 Discussion .................................................................................................................... 48

CHAPTER III – METABOLIC DISRUPTION VIA 3-BROMOPYRUVATE IN HUMAN COLORECTAL CANCER CELLS ............................................................. 63


Cell culture and media-glucose reduction ................................................................. 65 Measurement of cell growth ..................................................................................... 66 Protein isolation and western immunoblot analysis .................................................. 66 Cell death assay ......................................................................................................... 67 DNA fragmentation .................................................................................................. 68 RNA interference ...................................................................................................... 68 Statistical analysis ..................................................................................................... 69

Results .......................................................................................................................... 69 HKII expression is correlated with 3BP sensitivity in human CRC cells ................. 69 3BP-induced cell death is dose dependent ................................................................ 70 3BP induces rapid AKT phosphorylation at residue Thr308 .................................... 71 Glucose availability affects 3BP sensitivity in human CRC cells ............................ 72 HKII knockdown does not lead to changes in 3BP sensitivity ................................. 72

Discussion .................................................................................................................... 73

CHAPTER IV – HEXOKINASE II AS A POTENTIAL PROGNOSTIC BIOMARKER FOR HUMAN COLORECTAL CANCER ....................................... 87


Samples ..................................................................................................................... 89 Cell lines and immunoblot analysis .......................................................................... 89 Immunofluorescent staining ...................................................................................... 90 Imaging and quantification of immunofluorescence ................................................ 91 Statistical analysis ..................................................................................................... 92

Results .......................................................................................................................... 92 Patient characteristics ................................................................................................ 92 Immunoblot and immunofluorescent detection of HKII and CAIX ......................... 92 Levels of CAIX and HKII staining ........................................................................... 93 Association of CAIX and HKII expression with patient characteristics .................. 93 Association of CAIX and HKII expression with patient outcome ........................... 94 Association of stromal HKII staining with patient characteristics and outcome ...... 94

Discussion .................................................................................................................... 95 GENERAL DISCUSSION ........................................................................................... 111

SUMMARY AND CONCLUSIONS ........................................................................... 122 REFERENCES .............................................................................................................. 125

APPENDIX I – CHEMICAL LIST AND SUPPLIERS ............................................ 163 APPENDIX II – PREPARATION OF MATERIALS ............................................... 165

APPENDIX III – AV/PI STAINING OF ETOPOSIDE-TREATED SAMPLES ... 168

ix

APPENDIX IV – DENSITOMETRY OF AKT PHOSPHORYLATION AT DIFFERENT RESIDUES FOLLOWING 4 H 3BP EXPOSURE ............................ 169

APPENDIX V – FLOW CYTOMETRY ANALYSIS OF AV AND PI STAINING FOLLOWING 48 H 3BP EXPOSURE ....................................................................... 170

x

LIST OF TABLES

Table I. Primer sequences for quantitative real-time PCR analysis of PDK isoforms ......56 Table II. Patient characteristics ........................................................................................101 Table III. Associations of CAIX and HKII staining with patient characteristics and tumor

pathology .....................................................................................................................102 Table IV. Associations of stromal HKII staining with patient characteristics and tumor

pathology .....................................................................................................................103

xi

LIST OF FIGURES

Figure 1. DCA reduces cell growth in a cell line-dependent manner ................................57 Figure 2. PDK1 and 3 are highly expressed in human CRC cells .....................................58 Figure 3. PDK1 and 3 levels are upregulated in response to treatment .............................59 Figure 4. Phosphorylation status of PDH at residues Ser293 and Ser300 varies in

response to treatment .....................................................................................................60 Figure 5. Mitochondrial response to DCA exposure is cell line-dependent ......................61 Figure 6. DCA does not significantly alter metabolic status of human CRC cells ............62 Figure 7. HKII expression is correlated to 3BP sensitivity ...............................................81 Figure 8. 3BP-induced cell death is cell-line dependent ....................................................82 Figure 9. Type of cell death induced by 3BP is dose-dependent .......................................83 Figure 10. 3BP treatment leads to rapid AKT phosphorylation at residue Thr308 ...........84 Figure 11. Glucose availability alter 3BP sensitivity in CRC cells ...................................85 Figure 12. HKII expression knockdown does not lead to changes in 3BP sensitivity ......86 Figure 13. HKII and CAIX immunoblotting in colorectal cancer cell lines ....................104 Figure 14. HKII and CAIX immunostaining in colorectal cancer ...................................105 Figure 15. Evaluation of CAIX immunostaining .............................................................106 Figure 16. Evaluation of HKII immunostaining ..............................................................107 Figure 17. Immunofluorescent staining of HKII and corresponding F-Scores ...............108 Figure 18. Progression-free survival and overall survival by CAIX and HKII staining .109 Figure 19. Progression-free survival and overall survival by the presence of stromal HKII

staining ........................................................................................................................110

xii

LIST OF ABBREVIATIONS

Abbreviation Full Name 3BP 3-bromopyruvate 5-FU 5-fluorouracil APS Ammonium persulfate AV Annexin V BSA Bovine serum albumin CAIX Carbonic anhydrase IX CRC Colorectal cancer DAPI 4’, 6’-Diamidino-2-phenylindole DCA Dichloroacetate DMEM Dulbecco’s Modified Eagle Media DMSO Dimethyl sulfoxide EDTA Ethylenediaminetetraacetic acid ETC Electron transport chain FBS Fetal bovine serum FFPE Formalin-fixed paraffin-embedded HIF Hypoxia-inducible factor HK Hexokinase HRP Horseradish peroxidase IF Immunofluorescence MA Mitochondrial activity PCR Polymerase chain reaction PBS Phosphate-buffered saline PDH Pyruvate dehydrogenase PI Propidium iodide PMSF Phenylmethylsulfonyl fluoride PVDF Polyvinylidine fluoride ROS Reactive oxygen species SDS Sodium dodecyl sulfate siRNA Small interfering RNA TAE Tris-acetate-EDTA buffer TBS Tris-buffered saline TCA Tricarboxylic acid TEMED Tetramethylethylenediamine VHL von Hippel-Lindau

1

INTRODUCTION

Cancer is a group of complex diseases defined by uncontrolled cell growth. In

2012, there were an estimated 14.1 million new cancer cases and 8.2 million cancer-

related deaths worldwide (Torre et al, 2015). In 2014, it was estimated that 191,300 new

cancer cases and 76,600 cancer-related deaths occurred in Canada (Canadian Cancer

Society). This thesis will focus on colorectal cancer, which is the third most commonly

diagnosed and has the second highest mortality rate of all cancer types in Canada.

Specifically, colorectal cancer accounted for 13.9% and 11.6% of new cases among men

and women, respectively, and 12.8% and 11.5% of cancer-related deaths among men and

women, respectively, in 2014 (Canadian Cancer Society). The 5-year survival rate for

cases that are diagnosed while localized to the primary site is 90.3%. This drops

substantially to 12.5% in cases where disease has spread to distant sites by diagnosis,

accounting for 20% of all cases (Siegel et al, 2014). Thus, researchers seek to better

understand colorectal tumorigenesis and develop better screening methods and novel

treatments.

In 1926, the German scientist Otto Warburg reported that cancer cells produced

most of their energy through glycolysis, even under aerobic conditions (Warburg et al,

1927). Today, the “Warburg effect” can be described as an adaptation of cancer cells to

preferentially utilize glycolysis for energy production, even in the presence of abundant

oxygen. This adaptation is caused by alterations in signaling pathways involved in energy

metabolism, including glucose uptake, utilization and regulation of mitochondrial activity

(Koppenol et al, 2011). In 2011, its importance to tumor progression was signified by its

addition as an emerging hallmark to the revised list of hallmarks of cancer (Hanahan &

2

Weinberg, 2011). Recent studies provided researchers with a further understanding of

cancer metabolism to better determine targets that may be suitable for development of

anti-cancer therapeutic strategies. Although there is a high degree of heterogeneity in

mutational status of various tumors and thus in the inability to specifically target tumors,

an underlying mechanism that most tumors possess is rewiring of metabolic pathways.

Metabolic modulators that have shown promise in their ability to target key components

of the Warburg effect and leave normal cells unharmed include dichloroacetate and 3-

bromopyruvate.

Due to the highly disorganized and chaotic nature of the tumor vasculature, cells

in solid tumors are transiently and/or chronically exposed to decreased blood flow, or

ischemia. These heterogeneous microenvironments expose cancer cells to hypoxia and

hypoglycemia (Raghunand et al, 2003). While numerous studies have documented the

effects of ischemia on metabolic signaling, the mechanisms by which it may affect the

drug efficacy of metabolic modulators in cells is not fully understood. Recent studies

have shown that hypoxic signaling may somehow play a role in protecting cancer cells

against drugs targeting metabolic signaling (Shahrzad et al, 2010).

In summary, aerobic glycolysis plays an important role in tumor progression.

While recent studies have provided further insight into the metabolic pathways essential

to the rewiring of tumor energetics, there remain many questions as to the impact of the

tumor microenvironment on targeting key metabolic pathways and targets.

3

CHAPTER I – REVIEW OF THE LITERATURE

Glucose metabolism

Normal differentiated cells readily depend on mitochondrial respiration via

oxidative phosphorylation for efficient ATP production to meet energy demands,

accounting for 90% of all energy generation (Dasu et al, 2003). Glucose is initially

metabolized to pyruvate and further processed to acetyl-CoA through the TCA cycle. The

NADH-reducing equivalents generated through this process provide the hydrogen ions

required to establish an electrochemical gradient to facilitate ATP production through the

electron transport chain, in which oxygen is the final electron acceptor. Under anaerobic

conditions, glucose is converted to lactate via glycolysis. This reaction generates only 2

ATP molecules per glucose molecule, compared to the 38 ATP molecules per glucose

molecule generated through aerobic respiration (Vander Heiden et al, 2009). Thus,

aerobic respiration maximizes ATP production through mitochondrial respiration

compared to anaerobic glycolysis in the cytoplasm.

Warburg effect

The metabolic properties of cancer cells differ markedly from those of normal

cells. Dr. Otto Warburg formulated his hypothesis on cancer cell metabolism in the 1920s

when he observed that under aerobic conditions, tumor tissues metabolized glucose to

lactate approximately ten fold more than normal tissues (Warburg et al, 1927). This

ability of cancer cells to metabolize glucose to lactate regardless of oxygen tension was

coined the “Warburg effect”, or aerobic glycolysis. For example, leukemic cells are

highly glycolytic despite residing in the oxygen rich bloodstream (Elf & Chen, 2014). In

4

essence, it is the reverse of the Pasteur effect, which is the inhibition of fermentation by

O2. Warburg hypothesized that this metabolic phenotype arose from mitochondrial

defects within tumor cells that inhibited their ability to effectively oxidize glucose to

carbon dioxide. While striking, his observations did not initially prompt additional

research, attributed to a poor understanding of tumor metabolism at the time. It was

understood that pyruvate metabolism through fermentation and lactate production was

energetically inefficient and would not contribute to the neoplastic phenotype. There was

also a lack of knowledge on the signaling pathways associated with the metabolic switch

from pyruvate conversion to acetyl-CoA through the mitochondria versus conversion to

lactate in the cytoplasm. It was originally thought that cancer was a disease of genetic and

epigenetic alterations in oncogenes and tumor suppressors, where altered metabolism was

considered an indirect secondary phenomenon (Levine & Puzio-Kuter, 2010).

Now in the 21st century, scientists have begun to thoroughly investigate and

accelerate our understanding of tumor metabolism. First and foremost, most tumor

mitochondria are not defective in their ability to carry out oxidative phosphorylation. It is

clear now that mitochondrial metabolism is reprogrammed to support the anabolic

products required for rapid cell growth and proliferation. This metabolic strategy

emphasizes the production of nucleotides, amino acids, and fatty acids which in turn fuel

tumor growth by maximally producing RNA/DNA, proteins, and lipids, respectively, all

necessary for rapid cell division (Vander Heiden et al, 2009). Similar alterations in

metabolic signaling are observed in normal human physiological events with the need for

rapidly proliferating cells, including embryonic development, wound healing, particularly

in liver regeneration, and in immune responses to specific antigens, where clonal

5

selection provides increased cell numbers with increased immune specificity (Levine &

Puzio-Kuter, 2010). These represent appropriate responses to physiological growth

signals as opposed to constitutive cell autonomous adaptations as seen with neoplastic

growth (Newsholme et al, 1985).

In essence, increased glucose consumption is devoted to lactate conversion and

biosynthesis. While a significant portion of glucose consumed by cancer cells is directed

toward the glycolytic pathway for synthesis of anabolic precursors, oxidative

phosphorylation is utilized in most cancer cells and still a major source for ATP

generation (Guppy et al, 2002; Marin-Valencia et al, 2012). Warburg observed the rate of

O2 consumption in cancer cells remained comparable to that of normal cells (Warburg et

al, 1927). ATP concentration in tumors changes only marginally when compared to

normal tissues (Sonveaux et al, 2008). Cancer cells upregulate glucose transporters to

increase glucose uptake to support both anabolic and catabolic processes required to

sustain their rapid growth and proliferation (Macheda et al, 2005). Clinicians measure

this unique metabolic trait using 2-fluoro-2-deoxy-D-glucose positron emission

tomography (FDG-PET) as a diagnostic tool to assess various cancers including

colorectal cancer (CRC) (Artiko et al, 2015; Kruse et al, 2013; Rigo et al, 1996).

Although mitochondrial function remains active in neoplastic cells, it has been

reported that β-F1-ATPase, the catalytic subunit of the mitochondrial H+-ATP synthase

and the rate-limiting component of oxidative phosphorylation, is significantly reduced in

human tumors compared to its expression in normal tissues (Cuezva et al, 2002). This

observation was accompanied by an increase in the expression of glyceraldehyde 3-

phosphate dehydrogenase (GAPDH), a marker of glycolysis. The β-F1-ATPase/GADPH

6

ratio was termed the cell’s bioenergetic signature (Cuezva et al, 2002), and its reduction

is a common trait in over 95% of the carcinomas analyzed in large cohorts of breast,

colon, and lung cancer patients (Cuezva et al, 2009). The expression of β-F1-ATPase and

the decreased bioenergetic signature of lung cancer were shown to inversely correlate

with the rate of glucose consumption as assessed by FDG-PET imaging (Lopez-Rios et

al, 2007).

Advantages of the Warburg effect to cancer cells

Hanahan and Weinberg initially proposed that six biological capabilities were

acquired during tumorigenesis and were termed the “Hallmarks of Cancer” (Hanahan &

Weinberg, 2000). In 2011, this list was expanded to include "reprogramming energy

metabolism" as an emerging hallmark of cancer, highlighting its evident importance but

also the unsolved issues surrounding its mechanisms (Hanahan & Weinberg, 2011).

However, regardless of mechanisms, the Warburg effect clearly contributes to the

malignant phenotype of cancer cells.

Recent progress in cancer research has offered insight into the benefits and

selective advantages of the Warburg effect. As a result of rapid growth and proliferation,

cancer cells are required to increase glucose consumption and direct glucose derivatives

to anabolic pathways such as acetyl-CoA for fatty acid synthesis, glycolytic intermediates

for nonessential amino acids, and ribose for nucleotide synthesis (Vander Heiden et al,

2009). To compensate for a reduction in the levels of pyruvate available to be exhausted

through the TCA cycle and oxidative phosphorylation for ATP production, cancer cells

also rely on glutamine metabolism as an additional source of fuel (Gao et al, 2009; Wise

7

et al, 2008). Glucose and glutamine account for most carbon and nitrogen metabolism in

mammalian cells (Wellen et al, 2010). Thus, cancer cells increase glutamine consumption

for use through the TCA cycle so glucose derivatives can be shunted into anabolic

pathways (DeBerardinis et al, 2007). Glutamine metabolism also contributes nitrogen for

synthesis of purines, pyrimidines, nonessential amino acids, and the reduced form of

nicotinamide adenine dinucleotide phosphate (NADPH) for synthesis of fatty acids

(Metallo et al, 2012). On top of fuel for lipid synthesis, NADPH also functions as an

antioxidant to quench ROS produced during rapid proliferation of cancer cells. Thus, it is

imperative in both anabolic processes and for the maintenance of cellular redox

homeostasis. It is suspected that NADPH production may be the rate-limiting factor for

cell proliferation (Vander Heiden et al, 2009).

Aggressive tumors acquire an ability to invade the local environment and

metastasize to distant sites. Cancer cells secrete lactate, leading to extracellular

acidification and increased motility of cells to break through the basement membrane and

metastasize (Gatenby & Gillies, 2004; Gillies et al, 2008; Martinez-Zaguilan et al, 1996).

For metastasis to occur, cells must acquire an ability to resist anoikis, a form of

programmed cell death induced by the loss of cell-matrix interaction (Frisch & Ruoslahti,

1997). Detachment from the extracellular matrix leads to upregulation and activation of

several BH3-only proteins of the BCL-2 family of proteins (Gilmore, 2005). This leads to

activation of pro-apoptotic BCL-2 family of proteins, resulting in mitochondrial outer

membrane permeabilization that facilitates the release of apoptotic factors, downstream

caspase activation, and eventual cell death via apoptosis (Llambi et al, 2011).

8

Mitochondrial respiration is a major source of cellular ROS. In low levels, ROS

are able to activate various signaling pathways to stimulate cell proliferation and survival

while irreversibly damaging cellular components at higher levels that can lead to cell

death (Orrenius et al, 2007). To counteract the effects of ROS, cells express anti-oxidant

enzymes that detoxify ROS and prevent them from reaching high concentrations

(Trachootham et al, 2009). Tumors generate high ROS levels compared to normal tissue

(Haklar et al, 2001). Many cellular anti-oxidants are constitutively overexpressed in

cancer cells to counteract these effects, including the principal mitochondrial antioxidant

enzyme, manganese superoxide dismutase (Landriscina et al, 2009; Pani et al, 2004).

Matrix detachment also leads to ROS generation that can cause cell death

(Kamarajugadda et al, 2013; Li et al, 1999). Anoikis is suppressed in detached cells when

treated with antioxidants, further supporting the role of ROS generation in anoikis-

mediated cell death (Kamarajugadda et al, 2012; Schafer et al, 2009). Thus, the

metastatic potential of cancer cells is attributed to the Warburg effect and their ability to

resist cell death caused by increased ROS production as a result of matrix detachment and

anoikis-mediated cytotoxicity.

Glycolytic cancer cells are also better adapted to resist cell death through

modulation of BCL-2 family proteins (MacFarlane M 2012 Cell Cycle). Glycolysis-

mediated AKT activation stabilizes the expression of the anti-apoptotic BCL-2 protein

MCL-1 by preventing its proteasomal degradation (Coloff et al, 2011; Pelicano et al,

2006). This leads to the inhibition of pro-apoptotic proteins PUMA, BID, and Noxa and

their ability to permeabilize the mitochondrial outer membrane that would otherwise lead

to caspase-mediated apoptosis (Youle & Strasser, 2008). Furthermore, glycolysis

9

maintains normal translation of MCL-1 through the AMPK/mTOR pathway (Pradelli et

al, 2010). The ability of AKT to inhibit PUMA expression is dependent upon glucose

metabolism (Coloff et al, 2011). Likewise, the BH3-only protein Noxa is regulated via

phosphorylation of a serine residue in the presence of cytoplasmic glucose (Lowman et

al, 2010).

Regulation of glycolytic enzymes

Cancer cells exhibiting the Warburg effect have been shown to upregulate several

key glycolytic enzymes to further drive aerobic glycolysis. Pyruvate kinase (PK)

catalyzes the conversion of phosphoenolpyruvate to pyruvate while concurrently

producing ATP. PKM2, an alternatively spliced variant of M1, is expressed during

embryonic development and more recently has been shown to play a key role in

promoting the Warburg effect in tumorigenesis (Christofk et al, 2008a). PKM2 is

allosterically regulated between a less active dimer and an active tetramer (Mazurek,

2011; Mazurek et al, 2005). The low-activity PKM2 dimer drives aerobic glycolysis,

while the high-activity PKM2 tetramer produces pyruvate for oxidative phosphorylation

(Christofk et al, 2008b; David et al, 2010). Regulation between the two states is dictated

by fructose-1,6-bisphosphate (FBP), which when bound to PKM2 activates

tetramerization through high affinity association (Deprez et al, 1997). Dissociation of

FBP from the PKM2 tetramer results in conversion to the PKM2 dimer (Christofk et al,

2008a). PKM2 is important for proliferation and migration of colon cancer cells (Kwon et

al, 2012; Zhou et al, 2012). PKM2 knockdown results in suppression of proliferation and

invasion in vitro and the formation of xenograft tumors in vivo (Goldberg & Sharp, 2012;

Kefas et al, 2010). Phosphoglycerate mutase 1 catalyzes the conversion of 3-

10

phosphoglycerate (3PG) to 2-phosphoglycerate (2PG) during glycolysis. In many

cancers, including CRC, PGAM1 activity is increased compared with normal tissue (Liu

et al, 2008).

The important conversion of pyruvate to lactate is catalyzed by lactate

dehydrogenase (LDH). Attenuation of LDH activity impairs tumor metabolism and

subsequently proliferation (Fantin et al, 2006). Homo- and heterotetramers consisting of

gene products of LDHA and LDHB form five different isoforms (Draoui & Feron, 2011).

LDHA is upregulated by both hypoxia-inducible factor 1α (HIF1α) and MYC oncogene

(Semenza et al, 1996; Shim et al, 1997). LDH isoforms with high LDHB gene product

content catalyze the reverse conversion of lactate to pyruvate. Aberrant methylation of

LDHB promoter region, leading to transcriptional silencing, has been reported in gastric

and prostate cancers (Leiblich et al, 2006; Maekawa et al, 2003). It has been proposed

that a switch to aerobic glycolysis in colorectal cancer is due to in large part to

suppression of LDHB transcription (Thorn et al, 2009).

Unlike the glycolytic enzymes previously discussed, isocitrate dehydrogenase

(IDH) isoforms have been identified as mutated in human cancers. Sequencing studies

revealed that 12% of patients with glioblastoma multiforme have the same point mutation

in cytosolic IDH1 while up to 90% of patients with secondary gliomas have the same

mutation or the analogous mutation in mitochondrial IDH2 (Mardis et al, 2009; Parsons

et al, 2008; Yan et al, 2009). IDH normally functions to catalyze the conversion of

isocitrate to α-ketoglutarate via NADPH-dependent reduction. Mutated IDH confers

neomorphic activity in which isocitrate is covered to 2-hydroxyglutarate. This in turn

reduces α-ketoglutarate availability as a substrate for enzymes that methylate DNA and

11

histones, leading to enhanced tumorigenesis through modification of the epigenome

(Dang et al, 2010; Figueroa et al, 2010; Lu et al, 2012; Ward et al, 2010).

As previously stated, Warburg’s initial hypothesis attributing aerobic glycolysis to

mitochondrial dysfunction was incorrect. It is now well established that the mitochondria

continue to play a critical role even though cancer cells heavily depend on glycolysis for

their anabolic needs. However, β-F1-ATPase is significantly reduced in human tumors

compared to its expression in normal tissues (Cuezva et al, 2002). This leads to

mitochondrial hyperpolarization and inhibits the release of cytochrome C and apoptosis

inducing factor to avoid the apoptotic cascade (Kroemer et al, 2007).

Oncogenic mutations in mitochondrial metabolic enzymes further highlight their

importance in tumorigenesis. Familial paraganglioma can be caused by mutations in any

of the 4 subunits of the succinate dehydrogenase (SDH) complex (Astuti et al, 2001;

Baysal et al, 2000; Michelakis et al, 2008; Niemann & Muller, 2000). It is the only

enzyme that participates in both the TCA cycle and electron transport chain, where it

catalyzes the oxidation of succinate to fumarate with the reduction of ubiquinone to

ubiquinol (Oyedotun & Lemire, 2004). Mutations in fumarate hydratase (FH), involved

downstream of SDH in the TCA cycle, result in uterine fibroids, leiomyoma, and

papillary renal cell cancer (Tomlinson et al, 2002).

This thesis will focus on two key glycolytic enzymes in cancer cell metabolism:

hexokinase and pyruvate dehydrogenase kinase.

12

Hexokinase

Hexokinase is responsible for the first irreversible catalytic reaction of glycolysis

in which glucose is phosphorylated to become glucose-6-phosphate in the following

reaction:

Glucose + ATP è Glucose-6-PO42- + ADP

Glucose-6-phosphate serves as a precursor for not only glycolysis, but also the pentose

phosphate pathway (NADPH and ribulose-5-phosphate), glycogenesis (glycogen), and

hexosamine biosynthetic pathway (uridine diphosphate N-acetylglucosamine) (Pedersen,

2007; Wilson, 1995). Therefore, HK plays important roles in the regulation of both

anabolic and catabolic processes. Glucose-6-phosphate subsequently is able to inhibit HK

activity thus providing a negative feedback mechanism (Ardehali et al, 1999). The

fundamental approach to FDG-PET imaging and diagnosis of metabolically active tumors

depends on HK function, as it phosphorylates FDG at the 6th carbon with ATP to become

FDG-6-G and detected by PET scanning.

To date, four HK isoforms have been identified and are all encoded at different

chromosomal loci. Isoforms I, II, III, and IV are located on chromosome arms 10q22,

2p13, 5q35, and 7p15, respectively. HK isoforms I-III have a molecular mass double that

of HKIV (~100 vs. 50 kDa) as a result of evolutionary duplication and fusion of a single

domain HK molecule (Wilson, 1995; Wilson, 1997; Wilson, 2003). As a result, HK

isoforms I-III have a 250-fold higher affinity for glucose compared to HKIV. HK

isoforms are differentially expressed in various tissues. Both HKI and III are ubiquitously

expressed, although HKI is the main isoform in the brain and HKIII is not predominant in

any tissue. HKIV is primarily restricted to the liver and pancreas. In normal tissue, HKII

13

is limited to insulin-sensitive tissues such as adipose, skeletal and cardiac muscle

(Wilson, 1995; Wilson, 2003).

HKII is upregulated in many tumors, and is considered a consequence of

metabolic re-programming in cancer. Its expression level is closely associated with tumor

grade and mortality in hepatocellular carcinoma (Kwee et al, 2012). It has also been

shown to be upregulated in glioblastoma multiforme and brain metastases of breast

cancer (Palmieri et al, 2009; Wolf et al, 2011). During tumor development in tissues

normally expressing HKIV, HKII gene expression is induced whereas HKIV is silenced

(Mathupala et al, 1997b; Mayer et al, 1997; Rempel et al, 1996). Furthermore, the

catalytic activity of both HK domains is preserved only in HKII while the carboxyl

terminal domain is the only functional domain in HKI and III. Thus, the switch to a

higher glucose affinity HK isoform may allow the neoplastic cell to meet its high-glucose

demand. HKII is able to bind to voltage dependent anion channel (VDAC), a protein

located on the outer mitochondrial membrane (Nakashima et al, 1986). This allows

neoplastic cells to preferentially utilize the rapidly producing ATP generated through

oxidative phosphorylation to drive glycolysis. In human gliomas, it was found that 69%

of HKs were mitochondrial bound, compared to 9% that were mitochondrial bound in the

normal brain (Oudard et al, 1997).

Besides its role in metabolism, HKII also functions as a protective molecule. Pro-

apoptotic BH3 only proteins BAX and BAD exert their effects through binding to free

VDAC sites on the mitochondrial membrane, leading to its oligomerization by truncated

BID. This in turn leads to the activation of the mitochondrial permeability transition pore

to facilitate the release of cytochrome c into the cytoplasm (Scorrano & Korsmeyer,

14

2003). Mitochondrial-bound HKII (mito-HKII) competitively inhibits BAX and BAD

binding to the mitochondria and prevent localization to the outer mitochondrial

membrane (Majewski et al, 2004b). Forced detachment of HKII from the mitochondria

increases the ability of BAX to induce apoptosis (Pastorino et al, 2002). Decreased HKII

level sensitizes cells to apoptotic or necrotic stimuli (Roberts & Miyamoto, 2015).

Pyruvate dehydrogenase kinase

A key branching point in the glycolytic pathway is the production of pyruvate,

which in anaerobic conditions is converted to lactate. Under normoxic conditions, the

pyruvate dehydrogenase (PDH) complex (PDC) catalyzes the oxidative decarboxylation

of pyruvate to acetyl-CoA for use in the TCA cycle for cellular respiration. PDC is

composed of three catalytic components: pyruvate dehydrogenase (E1),

dihydrolipoamide acetyl-transferase (E2), and dihydrolipoamide dehydrogenase (E3).

PDC activity is inhibited in response to site-specific phosphorylation at three residues on

the PDH (E1) subunit: Ser232, Ser293, and Ser300 (Sale & Randle, 1981).

These reactions are catalyzed by one of four pyruvate dehydrogenase kinases

(PDK) that exhibit tissue-specific expression (Kolobova et al, 2001). PDK2 is reported to

be ubiquitously expressed, while PDK1 is expressed in the heart, pancreatic islets, liver

and skeletal muscle (Bowker-Kinley et al, 1998). PDK3 expression is primarily limited to

the testis, but is also expressed at low levels in the lung, kidney, spleen, heart and brain

(Bowker-Kinley et al, 1998). PDK4 is highly expressed in skeletal muscle and heart

while weakly expressed in the kidney, liver and lung (Bowker-Kinley et al, 1998).

15

The expression of different PDK isoforms depends markedly on the cancer type.

PDK1 is upregulated in many cancer cells and has been proposed as a key regulator of

the Warburg effect (Kim et al, 2006). It has been suggested that PDK1 could be used as a

biomarker of poor prognosis in patients with gastric cancer and head and neck squamous

cell cancer (Hur et al, 2013; Wigfield et al, 2008). PDK2 is a strong, independent

prognostic factor in cervical cancers (Lyng et al, 2006). PDK3, like PDK1, is upregulated

in colon cancer and its expression level is positively associated with tumor severity and

poor disease-free survival (Lu et al, 2011). Conflicting reports on the PDK4 expression

have been published; PDK4 upregulation is observed in renal, ovarian, prostate, lung and

skin cancer cells, while its downregulation is observed in breast, ovarian, lung, and colon

tumors when compared to its corresponding normal tissues (Blouin et al, 2011; Grassian

et al, 2011; Ross et al, 2000). By extension of their function to switch cells over to

glycolysis, cells with high expression of PDK1 and 4 are more resistant to anoikis

(Kamarajugadda et al, 2012). This is presumably a result of attenuation of mitochondrial

function, thereby lowering ROS production and sensitivity.

PDK1, 3 and 4 are reported to be hypoxia-responsive proteins under the control of

transcription factor HIF-α and are often upregulated during tumorigenesis (Kim et al,

2006; Lee et al, 2012; Lu et al, 2011). Hitosugi and colleagues showed that post-

translational activation of PDK1 can occur by tyrosine phosphorylation (Y243) due to

oncogenic tyrosine kinases localized in the mitochondria, thereby increasing the

conversion of pyruvate to lactate (Hitosugi et al, 2011).

16

Metabolic regulation by oncogenes and tumor suppressors

Research over the past decade has not only further elucidated the complexity of

tumor metabolism, but also has led to an alternative view towards mutations in oncogenic

signaling pathways. While metabolism was initially thought to be an indirect secondary

effect to oncogenic mutations, there is growing evidence to suggest that the primary

functions of activated oncogenes and inactivated tumor suppressors are to reprogram

cellular metabolism. It is becoming more clear that many key oncogenic signaling

pathways converge to adapt tumor cell metabolism in order to support growth and

survival (Ward & Thompson, 2012).

Oncogenic transcription factor MYC plays several roles in cell metabolism (Dang

et al, 2009). When overexpressed, MYC can induce the glycolytic phenotype in

transgenic cardiomyocytes in vivo. The first mechanistic link reported between an

activated oncogene and altered glucose metabolism was transcriptional activation of

LDHA by MYC (Shim et al, 1997). Knockdown of LDHA expression in MYC-

dependent tumors inhibits their growth (Le et al, 2010). MYC is able to induce the

splicing factors that are involved with the processing of the PK transcript to PKM2

(David et al, 2010). Certain genes involved in glutamine metabolism are under direct

transcriptional control of MYC (Gao et al, 2009; Wise et al, 2008). It enhances

transcription of glutaminase 1 (GLS1), an enzyme that catalyzes the hydrolysis of

glutamine to glutamate (DeBerardinis et al, 2007). MYC is also able to regulate

glutaminolysis through repression of miR-23a and miR-23b, two microRNAs shown to

inhibit expression of their target protein, GLS1 (Gao et al, 2009). Furthermore, MYC

17

cooperates with HIF to activate several glucose transporters and glycolytic enzymes,

including LDHA and PDK1 (Dang et al, 2008; Kim et al, 2007).

The tumor suppressor TP53 gene is the most commonly mutated gene in human

cancer (Kandoth et al, 2013). While p53 is given the title ‘guardian of the genome’, it

also serves as a regulator of cell metabolism (Vousden & Ryan, 2009). Wild type p53

functions to stimulate mitochondrial respiration and suppress glycolysis. Cytochrome c

oxidase assembly factor SCO2 transcription is stimulated by p53, which along with

SCO1 play an integral role in biogenesis of cytochrome c oxidase subunit II of the

electron transport chain (Leary et al, 2009; Matoba et al, 2006). p53 is able to indirectly

inhibit various glycolytic pathways by upregulating expression of TP53-induced

glycolysis and apoptosis regulator (TIGAR). TIGAR causes reductions in fructose-2,6-

bisphosphate levels and blocks further glycolytic reactions (Bensaad et al, 2006). Two

functional p53 motifs exist within the HKII promoter, indicative of p53 regulation

(Mathupala et al, 1997a). While MYC is responsible for GLS1 regulation, p53 is

responsible for GLS2 regulation with interestingly opposite effects on the cell (Hu et al,

2010). GLS1 catalyzes the conversion of glutamine to glutamate while GLS2 upregulates

antioxidant glutathione levels upon oxidative stress (Lora et al, 2004; Mates et al, 2009).

p53 also suppresses glucose uptake through repression of transcription of glucose

transporters (GLUTs) 1 and 4 (Dang, 1999). Expression of phosphatase and tensin

homolog (PTEN), a negative regulator of the PI3K pathway through dephosphorylation

of phosphatidylinositol (3,4,5)-triphosphate (PIP3), is upregulated by p53, further

suppressing glycolysis (Stambolic et al, 1998). Thus, loss of p53 in cancers may be a

major driving force behind acquiring the Warburg effect.

18

The PI3K pathway is one of the most commonly altered signaling pathways in

human cancers. Its activation can be attributed to mutations to tumor suppressors, such as

PTEN, in components of the PI3K complex itself, or by aberrant signaling from receptor

tyrosine kinases (Wong et al, 2010). AKT, frequently activated downstream of PI3K,

stimulates glycolytic flux by increasing translocation of glucose transporters to the cell

surface and by phosphorylating and altering localization of glycolytic enzymes such as

HKII and phosphofrutokinase 1 and 2 (Deprez et al, 1997; Elstrom et al, 2004; Majewski

et al, 2004a; Robey & Hay, 2009). AKT overexpression has been reported as an early

event during sporadic colon carcinogenesis (Roy et al, 2002). Although mitochondrial

acetyl-CoA is committed for further processing through the TCA cycle, it can be

processed to citrate via citrate synthase and be exported back out to the cytosol, where it

can be processed back to acetyl-CoA by ATP-citrate lyase (ACL). AKT is able to

phosphorylate and activate ACL, thereby facilitating the diversion of mitochondrial

citrate to cytoplasmic acetyl-CoA production to be directed instead for anabolic pathways

(Bauer et al, 2005; Berwick et al, 2002; Hatzivassiliou et al, 2005). AKT also increases

HIF1 activity via activation of downstream kinase FKBP-rapamycin-associated protein

(FRAP), further enhancing induction of glycolysis (Laughner et al, 2001).

Approximately 30% to 50% of colorectal tumors have a mutated KRAS gene

(Wilson et al, 2010). Colon carcinoma cells under hypoglycemic conditions have been

shown to increase RAS mutation rate, which upon transcriptional activation, facilitates

glucose uptake through induction of GLUT1 (Yun et al, 2009). In a multistep, multigene

transformation study of human breast epithelial cells, it was documented that while initial

transformation by viral oncogenes and telomerase reverse transcriptase increased

19

mitochondrial function, it took KRAS activation as the final reaction step to transform

cells to exhibit a glycolytic phenotype (Ramanathan et al, 2005). It’s hypothesized that

activated RAS induces MYC activity and enhances non-hypoxic levels of HIF1, although

the mechanisms remain unclear (Kikuchi et al, 2009; Sears et al, 1999).

Taken together, alternations in regulation of oncogenes and tumor suppressors

concomitantly rewire and coordinate cellular metabolism to meet the neoplastic demands

for biosynthetic precursors for rapid growth and proliferation.

Warburg effect as a therapeutic target

The concept of exploiting cancer cell metabolism is nothing new to clinical

practice. For nearly 50 years since the introduction of chemotherapy, this strategy has

been in practice. In principle, current chemotherapeutics aim to target all rapidly dividing

cells by interfering with critical pathways that are all, on some level, driven by cellular

metabolism. Antimetabolites were among the first effective chemotherapeutic agents

further supporting the concept of targeting cancer metabolism as a treatment strategy

(Kaye, 1998).

Aminopterin, a 4-amino analog of folic acid, was the first antimetabolite used in

cancer treatment. In 1947, it was the first drug shown to induce remissions among

patients with acute lymphoblastic leukemia (Farber & Diamond, 1948). Methotrexate

soon replaced aminopterin and is still currently used in the treatment of several cancer

types (Meyer et al, 1950; Purcell & Ettinger, 2003).

5-fluoruracil (5-FU) has been in clinical use for 40 years, and is used to treat a

variety of cancer types including CRC (Longley et al, 2003; Seifert et al, 1975). It

20

inhibits thymidylate synthase and the methylation of deoxyuridine monophosphate into

deoxythymidine monophosphate, the latter of which is required for DNA replication

(Peters et al, 2002). It was the only effective drug against colorectal cancer for decades

until the late 1990s when other anti-cancer agents arrived on scene, including

capecitabine, oxaliplatin, bevacizumab, cetuximab and most recently, regorafenib

(Sridharan et al, 2014).

Over the past decade, advances in our understanding of cancer cell metabolism

have allowed scientists to better comprehend and appreciate its complexity. Development

of targeted therapies now aims to exploit metabolic vulnerabilities in cancer cells and

specifically target cancer cell-specific metabolic alterations. The Warburg effect is a

metabolic phenotype that neoplasms share with very few other cell types, as previously

discussed. Therefore, inhibition of aerobic glycolysis can reduce tumor malignancy by

decreasing proliferation and apoptotic resistance, and ultimately inducing cell death,

providing a useful therapeutic window. Furthermore, acidic extracellular environments

lead to immunosuppression (Fischer et al, 2007). Thus, switching aerobic glycolysis to

mitochondrial respiration may provide an additional benefit of improving immune

function in already immunosuppressed patients.

2-deoxyglucose (2DG) is a glucose analog initially shown to be phosphorylated

by hexokinase and cannot be further metabolized, thereby acting as a competitive

inhibitor. However, later studies revealed that the principal mechanisms of 2DG’s

anticancer effects vary (Kurtoglu et al, 2007; Zhong et al, 2008). Although studies have

not supported its anticancer potential as a single agent in vivo, 2DG synergizes with other

anti-cancer compounds such as adriamycin and cisplatin (Maschek et al, 2004; Simons et

21

al, 2007). Lonidamine is a derivative of indazole-3-carboxylic acid, and a specific

inhibitor of HKII. A phase III trial targeting prostate cancer has been completed, although

its clinical success was limited due to significant pancreatic and hepatic toxicities (Price

et al, 1996). FK11, a novel LDH inhibitor, inhibits growth of both human B-lymphoma

and pancreatic cancer xenografts (Le et al, 2010).

Imatinib (Gleevec), among its many targets, inhibits constitutively active tyrosine

kinase BCR-ABL, a fusion protein found in nearly all patients with chronic myeloid

leukemia (CML) (Cambier et al, 1998; Kantarjian et al, 2011). BRC-ABL induces

increased glucose uptake and utilization through upregulation of the PI3K/AKT signaling

pathway (Elstrom et al, 2004). Imatinib reverses the Warburg effect through decreased

glucose uptake, reduced HK and G6PDH activity and increased mitochondrial

respiration. It has been approved by the US Food and Drug Administration and is used as

an anticancer drug against CML and gastrointestinal stromal tumors (Druker et al, 2006;

Druker et al, 2001; Joensuu et al, 2001).

The above examples support the significant opportunity metabolic pathways

provide for cancer control strategies. Therefore, it is imperative to continue to further our

knowledge of cancer metabolism in order to develop more effective and exclusive cancer

therapies that exploit the Warburg effect. In this thesis, exclusive focus will be placed on

examining dichloroacetate and 3-bromopyruvate as possible anti-cancer agents against

colorectal cancer.

22

Dichloroacetate

Humans have long been exposed to dichloroacetate (DCA) through water

chlorination and the metabolism of various chlorine-containing drugs or chemicals

(Stacpoole et al, 1998). It is an off-patent small molecule that has been in use for many

years to treat diseases such as lactic acidosis and inherited defects in mitochondrial

metabolism (Stacpoole et al, 2008; Stacpoole et al, 2003). In 1962, diabetic rats treated

with diisopropylammonium dichloroacetate showed a reduction in hyperglycemia and

increased respiration (Lorini & Ciman, 1962). The ability of this compound to promote

glucose oxidation exclusively in diabetic rats was thought to be due to its acid moiety,

dichloroacetate (Stacpoole & Felts, 1970). In 1973, researchers further showed that DCA

treatment led to an increase in glucose and pyruvate oxidation associated with an increase

in PDH-catalyzed acetyl-CoA levels (McAllister et al, 1973). The link between DCA and

PDK was finally elucidated in 1975 (Halestrap, 1975).

Of the 4 PDK isoforms, PDK2 is most sensitive to DCA. PDK inhibition by DCA

is a result of direct drug-protein interaction, whereby the carboxylate group of DCA

forms a salt bridge with Arg154 of PDK2 (Knoechel et al, 2006). Inhibition of PDK2 by

siRNA mimicked the physiological effect of DCA on cancer cells, and addition of DCA

in combination with PDK2 siRNA had no additional effects (Bonnet et al, 2007).

DCA is able to promote preferential oxidation of carbohydrates over fatty acids in

cardiac tissue and increase efficient energy production, thereby improving myocardial

function in vitro and in vivo (Bersin & Stacpoole, 1997). DCA is able to enhance

myocardial efficiency in patients with coronary artery disease (Wargovich et al, 1988).

This disease can lead to cardiac hypertrophy, in which cells exhibit a glycolytic shift, and

23

treatment with DCA reversed the metabolic shift and improved cardiac output (Piao et al,

2010). DCA therapy can improve post-ischemic function and recovery in rat hearts by

coupling glycolysis with glucose oxidation following cardiac hypertrophy (Wambolt et

al, 2000).

DCA had shown promise as a therapy for lactic acidosis, a potentially fatal

disease caused by buildup of lactic acid in tissues and blood (Stacpoole et al, 1988).

However, it showed little effect on the clinical outcome of patients, and further revealed

significant changes in blood pH following treatment. Thus, results have not translated

well in clinical trials (Stacpoole et al, 2006; Stacpoole et al, 1992).

In 2007, Bonnet and colleagues showed that exposing rats to DCA in drinking

water caused a significant regression of their xenografted A549 lung carcinoma cells

(Bonnet et al, 2007). Furthermore, they showed that the metabolic agent was able to

induce apoptosis through a mitochondrial-dependent pathway in cancer cells without

affecting normal cells (Bonnet et al, 2007). This metabolic shift to oxidative

phosphorylation will increase ROS production and alter mitochondrial membrane

potential, leading to the release of sequestered cytochrome c and activation of the

apoptotic-signaling cascade. Other studies have also shown DCA-induced apoptosis in

different cancer models, including ovarian (Saed et al, 2011), prostate (Cao et al, 2008),

(Madhok et al, 2010), and malignant melanoma (Populo et al, 2015).

Although several studies highlight the ability of DCA to induce apoptosis in vitro,

in vivo results show that DCA treatment slows tumor growth, rather than causes tumor

eradication. Metastatic breast cancer in vivo showed 58% decrease in lung metastases

following DCA treatment (Sun et al, 2010). The same study showed cell-line dependent

24

reduction in proliferation by 20-80% without increase in apoptosis following DCA

exposure in rat mammary adenocarcinoma cells in vitro. In multiple myeloma, certain

cell lines exhibited reduced proliferation without apoptotic induction at low doses of

DCA, while high doses of DCA led to smaller reduction in proliferation but an increase

in apoptosis (Sanchez et al, 2013). This dose effect was also observed in a separate study

on colorectal cancer cell lines (Madhok et al, 2010). However, DCA treatment of SW480

CRC xenografts led to an increase in tumor growth (Shahrzad et al, 2010). Furthermore

in HCT116 CRC cells, DCA exposure led to a reduction in proliferation with no effects

on cell survival (Delaney et al, 2015). Therefore, the effects of DCA may be cancer cell

line specific and dose-dependent.

The Warburg effect increases HIF1α levels and stimulates angiogenesis in vitro

and in vivo (Lu et al, 2002). DCA counteracts this effect in non-small cell lung cancer

and mammary carcinoma, in which a downregulation in HIF1α and decreased ability to

initiate and sustain angiogenesis in vitro and in vivo was shown (Sutendra et al, 2013).

Glioma cells stabilize HIF1α through IDH mutations that decrease cellular α-

ketoglutarate levels (Zhao et al, 2009). DCA is able to inhibit angiogenesis in

glioblastoma in vivo through increasing cellular α-ketoglutarate levels and destabilizing

HIF1α (Michelakis et al, 2010). A decrease in HIF1α levels as a result of DCA treatment

has also been reported in melanoma and T cell lymphoma (Kluza et al, 2012; Kumar et

al, 2012).

DCA treatment has been particularly successful against brain tumors such as

gliomas and glioblastomas, possibly attributed to its ability to cross the blood-brain

barrier (Duan et al, 2013; Kumar et al, 2013). The first clinical trial in glioblastoma was

25

conducted in 2010 (Michelakis et al, 2010). In freshly isolated glioblastoma tumors from

49 patients, DCA treatment depolarized mitochondria, increased mitochondrial ROS,

increased apoptosis, decreased proliferation, and decreased HIF1α expression with a

corresponding decrease in tumor vasculature. Additional Phase I trials have since been

conducted to determine drug safety and tolerability in patients (Chu et al, 2015; Dunbar

et al, 2013). In a Phase II clinical trial, 5 glioblastoma patients were treated with DCA for

15 months at a therapeutic concentration of approximately 0.5 mM (Michelakis et al,

2010). All but one of those patients, who died after three months, were clinically stable

after this regimen. However, there are significant drawbacks to all these trials. First, no

DCA-free group was included in the respective study designs. Secondly, the patients

recruited for these studies had all previously been heavily treated with chemotherapy

and/or radiation, and some were receiving standard treatment along with DCA.

Therefore, the beneficial outcomes of DCA treatment as outlined by these studies cannot

be attributed to DCA alone. Currently, clinical trials to study DCA in combination with

cisplatin for head and neck cancers, as well as DCA treatment in glioblastomas and other

neoplasms are ongoing (U.S. National Institutes of Health, 2015).

3-bromopyruvate

The pyruvate/lactate analog 3-bromopyruvate (3BP) was initially reported to

regulate enzyme activity in 1969 (Baker & Rabin, 1969). Fonda demonstrated the ability

of 3BP to alkylate cysteine residues via 3BP-mediated inactivation of glutamate

apodecarboxylase (Fonda, 1976) and its alkylating properties were outlined in a 1978

publication (Meloche et al 1978). Other subsequent studies validated the ability of 3BP to

26

inhibit other metabolic enzymes such as malate dehydrogenase, SDH, and carbonic

anhydrase (Chang & Hsu, 1973; Conroy & Maren, 1985; Sanborn et al, 1971).

In the past decade, numerous studies have highlighted the effects of 3BP on

cancer cells and its potential as an effective anti-cancer agent. 3BP-induced cell death has

been shown to be tumor-specific with limited cytotoxic effects against normal cells. 3BP

was less effective in weakly malignant pancreatic ductal adenocarcinoma cells and had

no effect on non-malignant cells (Isayev et al, 2014). A reason for its tumor-specific

actions could be due to an upregulation of monocarboxylate transporters (MCTs) in

cancer cells (Pinheiro et al, 2009; Pinheiro et al, 2010). The family of MCT proteins

helps to deliver monocarboxylates such as pyruvate and lactate across the cellular plasma

membrane. MCT1 in particular appears to be the main transporter of 3BP to glycolytic

tumors (Birsoy et al, 2013). Pretreatment of some breast cancer cell lines with butyrate

enhances MCT1 expression and sensitizes cells to 3BP treatment (Queiros et al, 2012). In

this scenario, 3BP may act as a ‘Trojan horse’ metabolite by mimicking lactate/pyruvate

but with deleterious effects once it enters the cell.

In 2001, Ko et al. initially showed the potential of 3BP to specifically kill cancer

cells in the rabbit VX2 tumor model of hepatocellular carcinoma (Ko et al, 2001). It was

later shown that the compound was able to eradicate all the tumors induced in 19 rats

tested with no obvious toxic effects or recurrence (Ko et al, 2004). Furthermore, the first

human case report was published in 2012 highlighting promising anti-tumor results in a

young adult cancer patient with fibrolamellar hepatocellular carcinoma without apparent

cytotoxicity (Ko et al, 2012). Other studies have also shown the potential of 3BP as a

therapeutic agent in different cancer models, including colorectal (Ihrlund et al, 2008;

27

Lea et al, 2013; Sanchez-Arago & Cuezva, 2011), breast (Queiros et al, 2012), ovarian

(Wintzell et al, 2012), pancreatic (Isayev et al, 2014), leukemia (Chen et al, 2009b), brain

(Davidescu et al, 2012; El Sayed et al, 2012), malignant mesothelioma (Icard et al,

2012), and malignant myeloma (Nakano et al, 2012).

The alkylating properties of 3BP allow it to react with thiol (SH) and hydroxyl

(OH) groups of various enzymes. One of its principle targets is HKII, an enzyme

upregulated in cancer cells (Ko et al, 2001). While the compound 2DG acts as a direct

competitor against glucose via HK phosphorylation, 3BP is able to inhibit mito-HKII

activity through formation of a pyruvinylated-adduct with –SH group(s) of the enzyme

(Oronsky et al, 2012). This in turn promotes the release of HKII from the mitochondria,

leading to depletion of cellular ATP levels and allowing pro-apoptotic molecules to bind

onto free VDAC sites and promote the release of apoptotic inducing factor (AIF) and

eventual cell death (Chen et al, 2009b; Kim et al, 2008). In addition, direct dissociation

of HKII from the mitochondria using a peptide was sufficient to induce cell death, but

unlike 3BP it did so without ROS induction or loss of mitochondrial membrane potential

(Chen et al, 2009b).

It is speculated that 3BP cytotoxicity is mediated through several combined

regulatory mechanisms, including ROS overproduction, ATP depletion, and modulations

to survival signaling pathways (Kim 2008, El Myiyad 2011, Calvino 2014). Researchers

believe that the rapid induction of cell death by 3BP is too fast to be simply accounted for

by the alkylation of HKII (Pereira da Silva et al, 2009). As an alkylating agent, the

effects of 3BP would not be limited to HKII. In agreement, 3BP affects various enzymes

involved in both glycolysis and mitochondrial respiration. Pereira and colleagues

28

demonstrated that exposure of HepG2 cells to 150 µM 3BP for 30 min led to significant

enzyme modifications involved in the glycolytic pathway (Pereira da Silva et al, 2009).

The activities of GADPH and phosphoglycerate kinase (PGK) were also strongly

suppressed (90 and 70% of control, respectively) while PK activity was almost twice that

of control. Furthermore, 3BP was shown to inhibit mitochondrial respiratory complex II

SDH activity. 3BP inhibits the efflux of lactate in HepG2 cells, altering the internal pH of

cells (Pereira da Silva et al, 2009). It is speculated that this is a result of MCT4

alkylation, thereby preventing the shunting of lactate out of the cell.

Several studies have shown different responses to 3BP at different doses in

various cancer types. In human myeloid leukemic cells, 3BP induced both apoptosis and

necrosis at 20-30 µM and pure cell necrosis at 60 µM (Calvino et al, 2014). El Sayed and

colleagues showed caspase-dependent cell death in C6 glioma cells with 3BP treatments

as low as 15 µM while Davidescu and colleagues demonstrated 3BP-induced autophagy

in GL15 glioblastoma cells at 80 µM (Davidescu et al, 2012; El Sayed et al, 2012). Huh-

7 hepatocellular carcinoma cells treated with 100 µM 3BP led to a significant increase in

apoptosis (Yu et al, 2012). Therefore, cellular response to 3BP appears to be dose and

cell line-dependent.

Tumor microenvironment

Tumors, like normal tissues, require a blood supply to gain access to the required

oxygen, nutrients and method for removal of waste products (Papetti & Herman, 2002).

Tumor expansion progressively distances cells from the normal vasculature, and

therefore oxygen and nutrients. During tumorigenesis, a solid tumor cannot exceed 2-3

29

mm3 without an adequate blood supply (Folkman, 1971). A fundamental hallmark of

cancer is the release of pro-angiogenic factors to stimulate the formation of new blood

vessels in a process known as angiogenesis (Holash et al, 1999). Tumor vasculature is

often characterized as disorganized, tortuous, leaky, and dilated, with uneven diameter,

and excessive branching and shunts (Carmeliet & Jain, 2000). Therefore, the variability

in blood flow within the tumor leads to a heterogeneous microenvironment with areas of

hypoxia, acidity, and low nutrient supply (Raghunand et al, 2003).

Hypoxia

Hockel and colleagues initially showed that hypoxia could be a possible

prognostic factor for cancer patients in a study of cervical cancer (Hockel et al, 1996a).

In 2005, an international multi-centre study involving 397 patients with head and neck

tumors provided further evidence of the prognostic value in tumor hypoxia (Nordsmark et

al, 2005). Hypoxia influences the effectiveness of chemotherapy by slowing cell cycle

progression and limiting drug diffusion (Brown, 1999). The expression of multidrug

resistance (MDR) genes is upregulated under hypoxic conditions, further providing

hypoxic cells with a selective advantage during chemotherapy (Wartenberg et al, 2003).

Hypoxia has also been associated with poor response to radiochemotherapy (Brizel et al,

1997; Hockel et al, 1996b; Nordsmark et al, 1996). Due to its ability to affect both

radiation and chemotherapeutic protocols, hypoxia is problematic in CRC treatment. In

CRC patients, the percentage of hypoxic cells in tissue biopsies varied from 2.2 to 37.8%

(Goethals et al, 2006). As the availability of oxygen to cancer cells is a direct function of

cell distance from a nearby functional blood supply, this leads to a gradient of diffusion-

30

limited hypoxia (Secomb et al, 1993), with both hypoxic (pO2 < normal) and anoxic (pO2

= 0 mmHg) microenvironments existing in the tumor (Vaupel et al, 1989). Differences in

vascular dependence of cancer cells are in part mediated by HIF1α expression; cells

lacking HIF1α expression were far less dependent on blood vessel proximity for their

growth and survival in vivo (Yu et al, 2001).

Hypoxia can be broadly categorized into 2 types: chronic or diffusion-limited

hypoxia, and acute, intermittent, or transient hypoxia (Brown, 1979; Thomlinson & Gray,

1955). Chronic hypoxia occurs in cells that are more than 100 µm away from blood

vessels, thus making this type of condition a function of vasculature distance/vascular

density. Acute hypoxia occurs as a result of chaotic vasculature and high interstitial

pressures within tumors, leading to vascular stasis and local fluctuations in blood

perfusion and oxygen supply. Tissue re-oxygenation/reperfusion leads to cellular damage

mediated by increased oxidative stress and ROS production. Administration of

superoxide dismutase at the time of vascular reperfusion leads to signification protection

against cell death (Parkins et al, 1995).

The ability of cells to adapt to changes in oxygen tension is in large part attributed

to HIF1 function. In 1992, Semenza and colleagues identified HIF1 and its consensus

binding sequence in the erythropoietin gene of hypoxic cells (Semenza & Wang, 1992).

Studies have shown that the activation of HIF1 has a profound effect on tumorigenesis by

activating transcription of genes involved with apoptosis resistance, invasion, metastasis

and angiogenesis (Lopez-Lazaro, 2006; Semenza, 2003; Semenza, 2006).

HIF1 is a heterodimer consisting of a constitutively expressing HIF1β subunit and

the oxygen-responsive HIF1α subunit (Jiang et al, 1996). The functionality of HIF1 is

31

dependent on the α subunit, whose synthesis is regulated through an oxygen-independent

process, whereas its degradation is oxygen-dependent (Semenza, 2003). HIF1α is able to

recognize hypoxia-responsive elements (HREs) in the promoter region of target genes

(Hu et al, 2003).

HIF1 Regulation

High HIF1α expression correlates with poor survival in numerous cancer types,

including breast (Bos et al, 2003; Gruber et al, 2004; Trastour et al, 2007; Vleugel et al,

2005), head and neck (Aebersold et al, 2001), esophagus (Matsuyama et al, 2005),

stomach (Griffiths et al, 2007), and lung (Swinson et al, 2004). HIF1α displays one of the

shortest half-lives (<5 mins) among cellular proteins in well-oxygenated cells (Berra et

al, 2001). In normoxia, HIF1α is degraded through posttranslational modification by

prolyl hydroxylation. Once hydroxylated, HIF1α will be targeted by and bound to the von

Hippel-Lindau (VHL) tumor suppressor protein – the recognition component of an E3

ubiquitin-protein ligase (Kaelin, 2003). Following its ubiquitination, HIF1α will rapidly

be degraded by proteasomes (Maxwell et al, 1999). In hypoxia, HIF1α is not modified by

oxygen-sensitive prolyl hydroxylases and thus becomes stabilized (Stiehl et al, 2006).

Other tumor suppressor effectors have been implicated in the regulation of HIF1 activity.

For instance, loss of PTEN facilitates HIF1-mediated gene expression (Zundel et al,

2000), and p53 association with HIF1α inhibits HIF1-stimulated transcription

(Blagosklonny et al, 1998).

The net result of hypoxic HIF1 activation is a shift in energy production from

mitochondria-dependent oxidative phosphorylation to mitochondria-independent

32

glycolysis. Papendreou and colleagues were the first to show that HIF1 is both necessary

and sufficient for downregulation of mitochondrial function in hypoxia (Papandreou et al,

2006). It coordinately drives glycolysis; all 12 glycolytic enzymes are in various ways

directly under HIF1 regulation (Iyer et al, 1998; Seagroves et al, 2001).

Interestingly, high levels of HIF1α have been shown in well-oxygenated tumors,

indicating potential oxygen-independent mechanisms of HIF1α stabilization (Hagg &

Wennstrom, 2005). Thus, as a result of aberrant activation of oncogenic signaling

pathways in cancers and mutations in tumor suppressor genes, HIF1 is activated even

under normoxic conditions. AKT increases HIF1 activity independent of O2 tension

(Inoki et al, 2005; Plas & Thompson, 2005). VHL is lost in renal cell carcinoma,

resulting in elevated non-hypoxic expression of HIF1α (Gordan et al, 2008). Mutations to

both SDH and FH that lead to accumulation in succinate and fumarate, respectively, have

been shown to stabilize HIF1 in normoxic cells. HIF1α hydroxylation by prolyl

hydroxylase domains involves the conversion of α-ketoglutarate to succinate. Increased

levels of succinate resulting from mutations in SDH can slow this enzyme, leading to

HIFα stabilization (Isaacs et al, 2005; Selak et al, 2005). Thus, glycolytic genes under

HIF1 regulation are transcribed even in the presence of oxygen, increasing the glycolytic

flux observed in cancer cells (Semenza, 2010).

Hypoglycemia

Like hypoxia, there is a predicted concentration gradient with glucose availability

in the extracellular space to cancer cells (Kallinowskil et al, 1985). Unlike hypoxia

however, the effects of hypoglycemia on tumor progression are not well studied. Based

33

on tumor cells’ reliance on glucose metabolism for both anabolic and catabolic processes

to sustain its rapid progression, a limitation of glucose should be expected to hinder

proliferative signaling and proliferative potential.

Cancer cells have a preference for either glucose or lactate, depending on their

microenvironment. Cells at greater distances from blood vessels rely more heavily on

glycolysis for energy production and thus show an increase in lactate production. Cells

that are located near blood vessels, and are thus aerobic, preferentially utilize lactate

secreted by glycolytic tumor cells as their main source of fuel (Sonveaux et al, 2008).

This allows the diffusion of glucose over a greater distance to reach glycolytic cells that

would otherwise be deprived of glucose. In return, these glycolytic cells produce lactate

that is effluxed out to the extracellular space to fuel aerobic cells, generating a positive

feedback loop between tumor cells across great distances of different microenvironments.

Thus, a lactate gradient is observed that mirrors the oxygen gradient in the tumor. This is

in large part controlled by MCTs 1 and 4 (Sonveaux et al, 2008). Lactate uptake by

aerobic tumor cells occurs through high-affinity lactate transporter MCT1, whereas

lactate produced by glycolysis-driven tumor cells is released through low-affinity lactate

transporter MCT4. This substitution to utilize lactate for oxidation is also observed in

ischemic brain recovery (Tanaka et al, 2004).

Previous work in our laboratory found that DCA provided a cytoprotective effect

to different human colorectal cancer cell lines when cultured under anoxic conditions

(Shahrzad et al, 2010). Furthermore, DCA-treated immune deficient mice bearing

SW480 CRC cell xenografts exhibited significantly increased tumor growth than in

untreated mice. Conversely, hypoxic conditions increased the sensitivity of colorectal and

34

lymphoma cells to 3BP (Xu et al, 2005). Strese and colleagues screened 19 commercially

available anti-cancer drugs on various cancer cell lines under different oxygen tensions.

They found that sorafenib, irinotecan, and docetaxel were in general more effective in an

oxygenated environment, while cisplatin, mitocycin c, and tirapazamine were more

effective in low oxygen environments (Strese et al, 2013). These findings suggest that the

efficacy of compounds targeting the glycolytic pathway might be dependent on the tumor

microenvironment.

In summary, the Warburg effect plays a critical role in tumor development by re-

wiring cellular metabolism to support an increased demand for anabolic processes, while

inhibiting mitochondrial driven apoptosis. Due to this 'metabolic addiction', targeting

cancer cell metabolism may provide a promising therapeutic window leading to the

development of novel anti-cancer agents. This thesis aims to explore the effects of

targeting different glycolytic enzymes shown to be upregulated in cancer progression,

with interest on the impact of the tumor microenvironment on drug efficacy.

35

RATIONALE

The ability of neoplastic cells to adapt and rewire metabolic signaling to produce

the necessary precursors to sustain continuous proliferation is a key aspect to cancer

being a disease of uncontrolled growth. Ever since 1926 when Dr. Otto Warburg made

the initial observations on the unique metabolic phenotype of cancer cells, this

relationship has gained popularity and researchers have been attempting to exploit this

avenue and find alternative therapeutics against cancer. An important aspect that has been

overlooked when examining the effects of various drugs as possible anti-cancer agents is

the tumor microenvironment. It is of particular importance when targeting metabolism as

the microenvironment has direct implications upon how neoplastic cells may regulate

their metabolic pathways.

Previous results from our laboratory showed that DCA provided a cytoprotective

effect to some human colorectal cancer cells under anoxic conditions. This led to interest

in further understanding the mechanisms that may be regulating this cytoprotection and

how the effects of the tumor microenvironment may impact the efficacy of other

metabolic-targeting compounds on human CRC cells. The first objective of this thesis is a

direct follow-up of the work on DCA-induced cytoprotection in anoxic human colorectal

cancer cells. Specifically, I assessed the expression of key glycolytic enzymes that are

directly affected by the effects of DCA: PDH and PDK. Furthermore, I also examined

changes in mitochondrial activity and metabolic output of cells following anoxic DCA

exposure. I hypothesized that the DCA-induced cytoprotection observed could be

attributed to differences in expression and activation of these molecules between various

human colorectal cancer cells.

36

The second and third objectives of this thesis looked to answer whether

hexokinase II is a viable enzyme target in human colorectal cancer. More specifically, the

second objective examined this question by treating human colorectal cancer cells with 3-

bromopyruvate and assessing its efficacy. I took into consideration the regulation of the

enzyme under varying glucose concentrations and examined the effects of the compound

under hypoglycemia conditions. I hypothesized that while 3BP would be able to induce

cell death, its efficacy would be regulated by cellular glucose availability and subsequent

regulation of HKII expression.

Based on the results of the second objective, I decided to examine hexokinase II

as a possible prognostic biomarker for human colorectal cancer. The experimental design

for the third objective utilized immunofluorescence staining of formalin-fixed paraffin-

embedded samples of human colorectal cancer tumor. I determined HKII expression

levels in these samples and correlated staining patterns to both its relation in different

microenvironments and to patient outcomes. I hypothesized that strong HKII expression

would correlate with worse outcome for patients with colorectal cancer.

In conclusion, these studies aimed to investigate how an intrinsic property of solid

tumors impacts expression of glycolytic enzymes and cellular metabolism using human

CRC as a model. Furthermore, I sought to better understand the impact of the tumor

microenvironment when targeting different aspects of tumor metabolism and the

Warburg effect.

37

CHAPTER II – EFFECTS OF DCA ON ANOXIC HUMAN COLORECTAL CANCER CELLS

A modified version of this chapter is already published:

Ho, N., Coomber, B.L. 2015. Pyruvate dehydrogenase kinase expression and metabolic changes following dichloroacetate exposure in anoxic colorectal cancer cells.

Experimental Cell Research 331: 73-81

Introduction

Cancer cells undergo a metabolic transition into a state of aerobic glycolysis

known as the “Warburg effect”, characterized by a paradoxical reliance on glycolytic

pathways for pyruvate reduction under environments of high oxygen tension (Warburg et

al, 1927). This unique metabolic profile provides cancer cells with numerous selective

advantages, including adaptation to hypoxia, resistance to mitochondria-mediated

apoptosis, and acidification of the tumor microenvironment leading to increased tumor

invasion and metastasis (Gillies & Gatenby, 2007; Gogvadze et al, 2010). Studies have

correlated the glycolytic state of tumor cells with cancer aggressiveness, further

establishing a critical role for the Warburg Effect (Simonnet et al, 2002).

Due to the highly disorganized and chaotic nature of the tumor vasculature,

regions of solid tumors are transiently and/or chronically exposed to ischemia and

reperfusion, leading to hypoxic/anoxic conditions, which are also known to be mutagenic

(Raghunand et al, 2003). The impact of the various microenvironments found in solid

tumors on the effectiveness of anti-cancer therapies is not well understood. The Warburg

effect enables cancer cells to use glycolytic pathways regardless of the

microenvironment, preventing metabolic dictation and possibly less efficient switching

between different strategies.

38

As discussed in Chapter I, an important branching point in glucose metabolism is

the production of pyruvate. Its use as a precursor for either subsequent fermentation in

the cytosol or mitochondrial uptake for processing through the TCA cycle and oxidative

phosphorylation is dictated by the function of the pyruvate dehydrogenase complex. Its

activity is regulated by site-specific phosphorylation at three serine residues of PDH:

Ser232, Ser293 and Ser300 (Sale & Randle, 1981). PDH phosphorylation is catalyzed by

the four PDK isoforms that exhibit tissue-specific expression (Kolobova et al, 2001).

PDK1, 3 and 4 are reported to be hypoxia-responsive proteins under the control of

transcription factor HIF-α and are often upregulated during tumorigenesis, playing a

crucial role in the Warburg effect (Kim et al, 2006; Lee et al, 2012; Lu et al, 2011).

Furthermore, PDK1 has been proposed as a key regulator of the Warburg Effect (Kim et

al, 2006). It has been suggested that PDK1 could be used as a biomarker of poor

prognosis in patients with gastric cancer and head and neck squamous cell cancer (Hur et

al, 2013; Wigfield et al, 2008).

Dichloroacetate (DCA) is an inexpensive small molecule that is currently being

evaluated in phase I/II clinical trials for solid tumors, including glioblastoma and glioma

(Chu et al, 2015; Dunbar et al, 2014; Yeluri et al, 2012). DCA inhibits PDK activity,

thereby indirectly activating PDH and shifting metabolism to mitochondrial glucose

oxidation (Stacpoole, 1989). Bonnet et al. initially showed that DCA exposure led to a

significant regression of tumors derived from A549 lung carcinoma cells xenografted in

rats (Bonnet et al, 2007). Other studies have also demonstrated DCA-induced apoptosis

in different cancer models, including ovarian (Saed et al, 2011), endometrial (Wong et al,

2008), neuroblastoma (Michelakis et al, 2010) and colorectal cancer (CRC) (Madhok et

39

al, 2010). However, my laboratory previously determined that DCA provided a

cytoprotective effect to some CRC cells under anoxic conditions in vitro and in vivo

(Shahrzad et al, 2010). Colorectal cancer is the third most common cancer worldwide,

and is the second leading cause of cancer-related deaths in Canada (Canadian Cancer

Society). Due to its high prevalence, a better understanding of the responses of CRC to

DCA is needed. This chapter begins to explore some of the molecular mechanisms by

which cancer cells under different microenvironmental stimuli may respond to

chemotherapeutics. In this study, I further elucidate the mechanisms by which DCA may

act on CRC cells under different oxygen tensions with particular emphasis on PDK

expression and PDH regulation.

Materials and Methods

A list of all chemicals and suppliers, and details of the materials and solutions

used for the following experiments are detailed in Appendices I and II, respectively.

Cell culture

Human colorectal cancer cell lines SW480 and LS174T and rat intestinal

epithelial cell line IEC-18 were obtained from ATCC (Manassas, VA, USA). Cells were

maintained in DMEM supplemented with 10% FBS, 1 mM sodium pyruvate, and 50

µg/ml gentamicin. Normal human colon mucosal epithelial cell line NCM460 was

obtained from INCELL and maintained in M3:D culture media supplemented with 10%

FBS. All cell lines were incubated at 37oC in a humidified atmosphere containing 5%

CO2 in room air (“normoxia”).

40

In vitro exposure to anoxia and/or DCA

Cells were seeded into plates and incubated under standard cell culture conditions

overnight. Thereafter, plates were assigned to control and anoxia treatment, with or

without 10 mM DCA (Sigma-Aldrich) for the indicated time periods. Anoxic conditions

were achieved using a Modular Incubator Chamber (Billups-Rothenberg Inc., Del Mar,

CA, USA) modified to allow continuous flushing of the chamber with a humidified

mixture of 95% N2 and 5% CO2; the oxygen content in the chamber was less than 0.1%

in all anoxic experiments.

Measurement of cell growth

Cell number was approximated through crystal violet staining. Briefly, 5 x 103

cells were seeded into a 96-well plate and cultured in the presence or absence of DCA in

normoxia or anoxia for 72 h. Post-treatment, media was aspirated and 1% crystal violet

solution was added to each well and incubated for 10 min in room temperature. Plates

were then aspirated, rinsed and left to dry overnight. 10% acetic acid was used to dissolve

the crystals and the absorbance at 570 nm was recorded. Cell numbers were extrapolated

through standard curves previously generated for each cell line. For NCM460 cells, 1 x

104 cells were seeded for the experiments described above.

qRT-PCR for PDK mRNA expression

Following 24 h treatment, RiboZol was used to homogenize cells and isolate total

RNA according to the manufacturer’s protocol. Samples were further purified using the

41

Aurum Total RNA Fatty and Fibrous Kit. iScript Reverse Transcription Supermix reagent

was used for cDNA synthesis under the following thermal cycle conditions: 25oC for 5

min, 42oC for 30 min, 85oC for 5 min. qRT-PCR was performed using the SsoFast

EvaGreen Supermix reagent and reaction conditions were optimized according to the

MIQE guidelines (Bustin et al, 2009) to achieve an assay performance efficiency of 90-

110%. The following primers for qRT-PCR analysis were obtained using Primer3

software (Rozen & Skaletsky, 1998) and outlined in Table I. qRT-PCR was conducted in

a CFX-96 Real-Time System (BioRad) under the following conditions: 95oC for 30 sec

followed by 40 cycles of 95oC for 10 sec and 59oC for 10 sec. Melt curve analysis was

performed immediately following PCR to ensure only the desired product was amplified.

Amplicons were sequenced to further validate their specificity. Gene expression was

analyzed using the CFX Manager software (Bio-Rad).

Protein isolation and western immunoblot analysis

Briefly, 5 x 105 cells were seeded in 60 mm2 plates and cultured in the presence or

absence of 10 mM DCA in normoxia or anoxia for 24 h. Following treatment, cells were

lysed on ice with lysis buffer supplemented with 2 µg/ml aprotinin, 1 mM PMSF and

phosphatase inhibitor cocktail II. Samples were centrifuged at 12,000 x g for 15 min at

4oC, and supernatant was aliquoted and stored at -80oC for future use. Protein was

quantified using the Bio-Rad DC Protein Assay Kit. 25 µg of total protein was resolved

in a 10% polyacrylamide gel using SDS-PAGE then transferred onto a polyvinylidene

difluoride (PVDF) membrane with wet transfer buffer at 100 V for 2 hr. Gels used to

subsequently detect caspase-3 expression were transferred onto PVDF membranes using

42

a Trans-Blot SD Semi-Dry Transfer Cell apparatus (Bio-Rad) with semi-dry buffer at 16

V for 30 min. Following transfer, membranes were blocked for 1 h at room temperature

in 5% non-fat dry milk diluted in 0.1% Tween 20 in TBS (TBST) followed by an

overnight incubation in blocking solution with primary antibodies. After washing,

membranes were incubated for 1 h at room temperature with the appropriate peroxidase-

conjugated secondary antibody, washed, and subjected to Luminata Forte

chemiluminescent substrate. Membranes were imaged using the ChemiDoc XRS+ system

(Bio-Rad) and densitometry was performed using Image Lab software (Bio-Rad).

Primary antibodies used included rabbit anti-caspase 3 (1:1,000), anti-PDK1(1:1,000) ,

anti-PDK2 (1:1,000), anti-PDK3 (1:1,000), anti-PDK4 (1:1,000), anti-pPDH Ser293

(1:1,000), anti-pPDH Ser300 (1:1,000), mouse anti-PDH (1:1,000), mouse anti-α-tubulin

(1:100,000), and rabbit or mouse HRP-labeled secondary antibodies (all 1:10,000).

FACS analysis of mitochondrial function

Mitochondrial activity was quantified through dual-labeling of cells with

MitoTracker dyes. Briefly, 5 x 105 cells were seeded into 60 mm2 plates and incubated

overnight. Following 24 h treatment, cells were trypsinized, and passed through a 20-

gauge needle to create a single cell suspension. Cells were centrifuged at 350 x g for 4

min, resuspended in non-supplemented DMEM with MitoTracker Green FM (50 nM) and

Orange CM-H2TMRos (300 nM) probes and incubated for 30 min at 37oC, protected

from light. Cells were then subsequently analyzed using a FACScan ™ flow cytometer

(BD Biosciences). In total, 1 x 104 events were counted per sample.

43

Glucose consumption and lactate production

Metabolic output was examined by analyzing glucose consumption and lactate

production. Briefly, 2.5x105 cells were seeded into a 24-well plate and cultured in the

presence or absence of DCA in normoxia or anoxia for 24 h. Following treatment,

conditioned media was collected and assessed for glucose consumption and lactate

production using the Amplex Red Glucose Assay and Eton Bioscience L-Lactate Assay

kits, respectively, following the manufacturer’s protocols. After collection of conditioned

media for analysis, treatment wells were stained with crystal violet as previously

described to determine cell numbers and normalize output values. Cell numbers were

extrapolated through standard curves previously generated for each cell line.

Statistical analysis

Data are presented as means of independent measurements +/- SE. GraphPad

Prism v6.0 software (GraphPad Software Inc., La Jolla, CA, USA) was used to perform

statistical analysis. Kriskal-Wallis ANOVA followed by Dunn’s post-hoc tests were

performed to determine differences between means within each cell line for cell growth,

qRT-PCR and western blot densitometric analysis. One-way ANOVA followed by

Tukey’s post-hoc tests were performed to determine differences between means within

each cell line for flow cytometry and metabolic output analysis. At least three biological

replicates were used for each analysis, and treatments were considered significantly

different if a p-value ≤ 0.05 was achieved.

44

Results

DCA reduces both normoxic and anoxic cell growth

The effects of DCA on CRC cell growth under different atmospheric conditions

were determined using crystal violet staining. Cells were treated with 0 – 20 mM DCA.

Following 72 h treatment, cells responded to DCA in a dose-dependent manner.

Significant repression on cell number was observed in all four cell lines when treated

with 20 mM DCA (Figure 1A-D). Non-tumorigenic IEC-18 cells were most sensitive to

the cell proliferation effects of DCA (Figure 1B). Anoxic exposure also led to a

significant reduction in IEC-18 cell growth while no significant changes were observed in

human normal NCM460 and CRC cells. The repressive effects of DCA on cell number

were less profound for SW480 cells compared to LS174T cells (Figure 1C,D). No

additional reduction in cell growth following DCA treatment under anoxic conditions

was observed in all cell lines examined. Anoxic exposure increased caspase-3 cleavage in

IEC-18 and LS174T cells. DCA treatment reduced caspase-3 cleavage in CRC cells

following 48 h treatment, regardless of atmospheric conditions (Figure 1E). This

suggests the observed reduction in cell growth was not a result of apoptotic induction.

Human CRC cells predominantly express PDKs 1 and 3

mRNA and protein expression of PDK isoforms in CRC cells were determined

using both western blot and qRT-PCR analysis. Both SW480 and LS174T cells highly

expressed mRNA transcripts of PDKs 1 and 3 relative to isoforms 2 and 4. (Figure

2A,B). These results correlated with protein expression detected via western blot (Figure

2C); SW480 and LS174T cells expressed PDKs 1 and 3 but not 2 and 4. LS174T cells

45

appeared to show higher expression of PDK3 than SW480 at the protein level. Normal

human NCM460 cells showed protein expression of PDK isoforms 1 – 3. Normal rat

IEC-18 cells showed protein expression of PDK isoforms 1, 3 and 4 (Figure 2).

PDKs 1 and 3 are upregulated in response to treatment

Both CRC cell lines showed an increase in PDK1 but not PDK3 mRNA levels

following anoxic exposure (Figure 3). DCA treatment led to slight increases in mRNA

levels of both PDK isoforms examined, which was further increased in combination with

anoxia. PDK3 mRNA expression was significantly upregulated when anoxic cells were

exposed to 10 mM DCA. Western blot analysis confirmed changes observed in mRNA

levels in which anoxic exposure led to increased PDK1 protein expression which further

increased in combination with DCA in LS174T cells (Figure 3B).

PDH phosphorylation status varies in response to treatment between cell lines

Following 24 h anoxic DCA exposure, PDH phosphorylation was assessed

through western blotting (Figure 4). In the non-tumorigenic cell line IEC-18, as

expected, DCA reduced and abolished phosphorylation at residues Ser293 and Ser300,

respectively, while anoxic exposure increased phosphorylation at both residues. Human

CRC cells exhibited the same phosphorylation pattern as IEC-18 cells at residue Ser300

following treatment. However, phosphorylation at residue Ser293 varied between cell

lines. In SW480 cells, neither DCA, anoxia, nor combined exposure led to significant

changes in pSer293 phosphorylation. In LS174T cells, phosphorylation only differed

46

from control cells following combination treatment (Figure 4). Native PDH levels did

not significantly change between groups over the course of treatment.

Mitochondrial activity varies in response to DCA between human CRC cells

The effects of DCA on mitochondrial activity was determined utilizing

MitoTracker probes and flow cytometry analysis. The MitoTracker Green FM probe

accumulates in the lipid environment of the mitochondria while the MitoTracker Orange

CM-H2TMRos probe does not fluoresce until it enters an actively respiring cell and is

oxidized in the mitochondria (Johnson, 2010). Representative flow plots demonstrate that

the proportion of cells with highly active mitochondria varies between cell lines (Figure

5A). Human CRC cell lines displayed a higher mitochondrial activity (MA) under

control conditions (SW480 cells: 81.1 ± 2.3%, LS174T cells: 69.4 ± 2.2%) compared to

non-tumorigenic control IEC-18 cells (57.2 ± 1.4%) (Figure 5B). In IEC-18 cells, DCA

increased MA, anoxic exposure reduced MA, and DCA restored the MA of anoxic cells

as expected. LS174T cells followed the same pattern in their response to DCA and anoxic

exposure. Surprisingly, SW480 cells differed in their mitochondrial responses to various

treatments. Unlike the other two cell lines, the MA of SW480 cells decreased following

DCA exposure in normoxia and further decreased in combination with anoxia (Figure

5B). Changes from control following anoxic DCA exposure were most profound with

SW480 compared to the other cell lines.

47

DCA does not significantly alter metabolic status of human CRC cells

Conditioned media were collected following 24h DCA exposure under different

oxic conditions to analyze glucose consumption and lactate production as outputs for

cellular metabolism. Both metabolic outputs showed similar trends between treatment

groups for each cell line (Figure 6). Cells under anoxic conditions upregulated glycolysis

and thus consumed more glucose and produced more lactate. This metabolic activity was

notably greater in SW480 and LS174T cells compared to IEC-18 cells even under

normoxia, indicative of the Warburg effect. Due to the lack of oxygen and necessity to

undergo anaerobic glycolysis, anoxic IEC-18 cells consumed approximately 4.4 fold

more glucose and produced approximately 7 fold more lactate compared to the same cells

under normoxia. The changes in metabolic outputs under low oxygen tensions were not

as profound in CRC cells: anoxic SW480 cells consumed approximately 1.6 fold more

glucose and produced approximately 1.4 fold more lactate compared to when under

normoxia, and LS174T cells consumed approximately 2.5 fold more glucose and

produced approximately 2.6 fold more lactate compared to when under normoxia (Figure

6). As expected, DCA did not significantly alter either metabolic output regardless of

atmospheric conditions in IEC-18 cells. DCA did, however, lower both glucose

consumption and lactate production in both CRC cell lines. These changes were not

statistically significant, but were more profound under normoxia than in anoxia,

presumably due to the cell's further reliance on glycolysis under these conditions.

48

Discussion

Previous studies in my laboratory demonstrated a cytoprotective effect induced by

DCA in anoxic human CRC cells (Shahrzad et al, 2010). In this chapter, I sought to

investigate the molecular mechanisms by which this cytoprotection was conferred.

Anoxic exposure led to profound changes in PDH activation, mitochondrial activity and

metabolic output. Although DCA exposure led to differences in changes in PDH

activation and mitochondrial activity between different human CRC cells, it did not lead

to differences in metabolic and growth responses.

I initially sought to determine whether DCA exposure under different atmospheric

conditions would have an effect on cell growth. Due to the hypothesized actions of DCA

on the mitochondria, I avoided commonly used cell viability assays that relied on

mitochondrial activity (i.e. MTT, WST-1) and used crystal violet staining as an indirect

method to determine cell number and growth. Normal human colon mucosal epithelial

NCM460 cells and rat intestinal epithelial IEC-18 cells were used in conjunction with

human CRC cells SW480 and LS174T to evaluate the effects of DCA treatment under

different atmospheric conditions. All cell lines responded to DCA in a dose-dependent

manner, although significant reduction in cell growth was only observed at 20 mM DCA.

While it has been hypothesized that normal cells would be unaffected by DCA treatment,

I found that IEC-18 cells were most sensitive to DCA treatment (Bonnet et al, 2007). By

altering the metabolic status of cells to enhance glycolysis for energy production as a

result of anoxic exposure, I expected that DCA treatment would further reduce cell

growth as cells are 'forced' to enter into the TCA cycle under metabolic stress. However, I

found no additional reduction in cell growth following DCA treatment under anoxic

49

conditions in all cell lines examined. The lack of caspase-3 cleavage suggests reduced

cell growth was not a result of apoptotic induction. Other studies have shown that DCA

does not need to induce apoptosis to reduce cell number in different cancer types,

including prostate (Cao et al, 2008), pancreatic (Chen et al, 2009a), breast (Sun et al,

2010), colon (Sanchez-Arago et al, 2010), and lung (Stockwin et al, 2010).

Due to the differences observed in DCA sensitivity between SW480 and LS174T

cells, I determined whether this sensitivity could be attributed to differential expression

of PDK isoforms in these two cell lines. qRT-PCR and western blot results revealed that

these two cell lines primarily expressed PDK isoforms 1 and 3. Although it has been

suggested that PDK2 is the major isoform responsible for metabolic control over PDH

activity (Bowker-Kinley et al, 1998), both mRNA and protein levels were barely

detectable in the CRC cell lines examined. These results suggest that regulation of PDH

activity in these CRC cells is through PDK isoforms 1 and 3. PDK2 is most sensitive to

inhibition by DCA (Bowker-Kinley et al, 1998), perhaps explaining the minimal change

in cell growth following DCA exposure in the cell lines examined. The addition of

exogenous pyruvate when conducting these experiments could have altered PDK

expression, although this has not been confirmed. In Baker’s study, while PDK2 activity

is reduced to <10% residual activity as a result of elevated pyruvate, its expression was

not examined (Baker et al, 2000).

PDK isoforms 1 and 3 are upregulated in response to hypoxia in CRC

Colo320DM, HT29, and HCT116 cells (Lu et al, 2011). Contrary to what has been

reported in the literature, anoxic exposure did not have a significant effect on PDK3

transcript levels in either SW480 or LS174T cells. The addition of DCA to anoxic CRC

50

cells led to significant upregulation in PDK3 mRNA expression. Furthermore, DCA

treatment led to an increase in both normoxic and anoxic PDK1 expression in LS174T

cells. The ability of cells to upregulate both PDK isoforms under anoxia coincides with

DCA-induced cytoprotection under the same conditions, leading to speculation that these

enzymes may some how play a role in this process.

Due to differences observed in PDK expression following treatment, I next sought

to determine changes in PDH expression and regulation. PDH phosphorylation by PDK

leads to enzymatic inhibition and prevents pyruvate conversion to acetyl-CoA and entry

into the mitochondria where it would enter into the TCA cycle to further produce

reducing intermediates. PDH phosphorylation occurs at 3 residues: Ser293, Ser300, and

Ser232. Each PDK isoform displays a different site-specificity for PDH phosphorylation.

While all 4 PDK isoforms are able to phosphorylate PDH at residues Ser293 and Ser300,

only PDK1 is able to phosphorylate Ser232. It has been suggested that pSer232 may have

a limited function in PDC regulation in comparison to residues Ser293 and Ser300

(Rardin et al, 2009).

Thus, I focused my efforts on examining changes to PDH phosphorylation at

residues Ser293 and Ser300. Following 24 h anoxic DCA exposure, PDH

phosphorylation was assessed through western blot analysis. As DCA is an inhibitor of

the enzyme responsible for PDH phosphorylation, it would be expected that there would

be a decrease in PDH phosphorylation. This was demonstrated with the non-tumorigenic

cell line IEC-18; DCA drastically reduced and abolished phosphorylation at residues

Ser293 and Ser300, respectively.

51

Under anoxic conditions, increases in PDK expression would subsequently lead to

increased PDH phosphorylation and inactivation, and a shift in metabolism towards

glycolytic pathways and away from mitochondrial respiration. While this pattern was

observed in IEC-18 cells, phosphorylation at residue Ser293 varied in the two human

CRC cells in response to experimental treatment. These results suggest that differences in

sensitivities to DCA between different human CRC cell lines could be attributed to PDH

phosphorylation status at residue Ser293 but not Ser300. The PDH phosphorylation

patterns at Ser293 in DCA treated anoxic CRC cells could be attributed to the differences

in PDK1 regulation at the protein level. While LS174T cells are able to upregulate PDK1

expression following anoxic DCA exposure, this upregulation was not observed in

SW480 cells. Despite these differences, responses in cell growth following various

treatments appeared the same between SW480 and LS174T cells, suggesting PDH

phosphorylation at residue Ser300 still plays a role in maintaining PDH activity. Residue

Ser293 is the predominant PDH inactivation site while residues Ser300 and Ser232 may

play a limited role (Sale & Randle, 1981). Pharmacological manipulation of the

phosphorylation status of PDH with various known inhibitors of oxidative

phosphorylation or signaling pathways showed no changes at residue Ser232 (Rardin et

al, 2009).

Recent studies have revealed emerging roles of alternate post-translational

modifications to PDH that would further our understanding of its molecular regulation

beyond its activation via dephosphorylation. Individual global acetyl-proteomic and

lysine desuccinylation analyses revealed several lysine acetylation and succinylation sites

on PDH (Fan et al, 2014; Park et al, 2013). Therefore, although some differences were

52

observed in PDH phosphorylation, the response of CRC cells to DCA exposure via PDH

activation is unclear unless more experiments are conducted to examine these novel post-

translational modifications.

The effects of DCA treatment under different atmospheric conditions on

mitochondrial activity and the cell metabolic status were also examined. I opted to

determine changes in mitochondrial activity through mitochondrial staining and

subsequent flow cytometry analysis. Results showed that control CRC cells displayed a

higher MA compared to non-tumorigenic control IEC-18. This could be attributed to the

presence of hyperpolarized mitochondria in the cancer cells as a result of the Warburg

effect (Bonnet et al, 2007). SW480 cells differed in their mitochondrial responses to

various treatments. This coincides with the unique PDH phosphorylation pattern

observed at residue Ser293 following DCA exposure in SW480 cells. These results

suggest that the differences in regulation of PDH phosphorylation at residue Ser293

between different cell lines could be affecting mitochondrial activity.

I then investigated whether the changes observed in mitochondrial activity

correlated with the metabolic output of treated cells. As expected, CRC cells consumed

more glucose and produced more lactate as a result of the Warburg effect. Under anoxic

conditions, IEC-18 cells had a greater increase in glucose consumption and lactate

production compared to CRC SW480 and LS174T cells. These numbers suggest that as a

result of the Warburg effect, the CRC cells have no need to heavily re-route their

metabolic pathways to compensate for lack of oxygen in order to undergo further

glycolysis. DCA was shown to reduce glucose consumption and lactate production in

normoxic CRC cells with no apparent effect on IEC-18 cells, indicative of a re-wiring of

53

the Warburg effect (Stacpoole et al, 1998; Warburg et al, 1927). However, this change

was diminished in anoxic CRC cells, presumably due to an increased reliance on

glycolysis under these conditions.

An issue with interpreting results from DCA studies conducted in vitro is the

super-physiological doses that are used to induce cellular responses. This is required due

to a lack of effective cellular uptake and localization inside mitochondria (Stockwin et al,

2010). SLC5A8 is the plasma membrane transporter responsible for DCA uptake. This

transporter is expressed in normal cells but limited in tumor cells, leading to an increased

resistance to the antitumor effects of DCA and the detrimental side effects involving the

nervous system associated with high doses in clinical studies (Babu et al, 2011; U.S.

National Institutes of Health, 2015). Derivatives of DCA have been shown to be more

potent in cancer cells. Mito-DCA (Pathak et al, 2014) and mitaplatin (Dhar & Lippard,

2009) are two new DCA-derived molecules that are now under investigation. Mito-DCA

is more potent and selective towards cancer cells than DCA. The incorporation of a

lipophilic triphenylphosphonium cation through a biodegradable linker allows

mitochondrial targeting in Mito-DCA. Pathak and colleagues showed that this compound

was able to re-route metabolism from lactate production to oxidative phosphorylation,

which subsequently led to induction of apoptosis in affected cells (Pathak et al, 2014).

These effects were found to be cancer specific and had no significant effect on normal

cells. Mitaplatin in essence is a complex formed by combining cisplatin and DCA in one

molecule (Dhar & Lippard, 2009). Its potency lies in its ability to attack both nuclear

DNA via cisplatin and aerobic glycolysis via DCA. The authors showed that mitaplatin

was able to significantly increase apoptosis vs. either cisplatin or DCA alone (55, 172 and

54

1.2%), while also significantly reducing the cytotoxic effects observed in normal lung

fibroblasts in response to cisplatin when treated with mitaplatin. (2.8 down to 73.6%)

(Dhar & Lippard, 2009).

In summary, human CRC cells SW480 and LS174T express predominantly PDK

isoforms 1 and 3, which are upregulated in response to DCA and anoxia. I found

evidence for differential phosphorylation of PDH at residue Ser293 that was not reflected

in changes in cell proliferation. Activation of PDK on Ser293 could possibly lead to

differences in mitochondrial activity, however, the metabolic changes following anoxic

DCA exposure did not differ between cell lines. This study confirms previous findings

that the effects of DCA on CRC cells are heterogeneous, which may limit the usefulness

of this putative anti-cancer agent.

The results presented here suggest there is still potential for DCA to be used as a

chemosensitizing agent. DCA was able to alter mitochondrial activity and mitochondrial

membrane potential. Since reducing the polarization of the mitochondria of cancer cells

could lead to facilitated release of pro-apoptotic signaling molecules, this could potentiate

current chemotherapeutics. For instance, Sanchez et al. (Sanchez et al, 2013) showed an

increase in sensitivity to bortezomib in combination with DCA in myeloma-bearing mice,

and Kumar et al. (Kumar et al, 2013) reported that DCA was able to amplify the effects

of bevacizumab in a glioblastoma-derived xenograft mouse model. Activation of

mitochondrial respiration by DCA can sensitize can cells to sorafenib in hepatocellular

carcinoma cells (Shen et al, 2013). Reduction in lactic acid secretion and

immunosuppression in thymoma and melanoma tumors following DCA exposure in vivo

highlights another pathway towards chemosensitization (Ohashi et al, 2013). The results

55

presented suggest different colorectal cancer cell lines have diverse responses to the

presence of DCA under different atmospheric conditions. Since tumors are under

constant changes in oxygen levels due to the tumor microenvironment, it is important to

take into account the atmospheric conditions which cells are being treated with possible

anti-cancer agents.

56

Table I. Primer sequences for quantitative real-time PCR analysis of PDK isoforms.

Gene Primer Sequence

PDK1 Forward 5'-CTATGAAAATGCTAGGCGTCT-3'

Reverse 5'-AACCACTTGTATTGGCTGTCC-3'

PDK2 Forward 5'-AGGACACCTACGGCGATGA-3'

Reverse 5'-TGCCGATGTGTTTGGGATGG-3'

PDK3 Forward 5'-GCCAAAGCGCCAGACAAAC-3'

Reverse 5'-CAACTGTCGCTCTCATTGAGT-3'

PDK4 Forward 5'-TTATCATACTCCACTGCACCA-3'

Reverse 5'-ATAGACTCAGAAGACAAAGCCT-3'

β-actin Forward 5'-AAGATCAAGATCATTGCTCCTC-3'

Reverse 5'-CAACTAAGTCATAGTCCGCC-3'

57

Figure 1. DCA reduces cell growth in a cell line-dependent manner. Crystal violet staining assay was used to examine cell growth following 72 h 0 – 20 mM DCA and/or anoxic (< 0.1% O2) treatment in non-tumorigenic NCM460 (A), IEC-18 (B) and human CRC SW480 (C) and LS174T (D) cell lines. Cell numbers were determined and normalized to control groups. Caspase-3 cleavage was determined through western blot analysis in IEC-18, SW480 and LS174T cells following 48 h 10 mM DCA and/or anoxic (<0.1% O2) treatment (E). (a p ≤ 0.05 vs. control, b p ≤ 0.05 vs. anoxia-treated)

58

Figure 2. PDK1 and 3 are highly expressed in human CRC cells. qRT-PCR analysis was performed to determine mRNA expression of PDK isoforms 1-4 in human CRC cell lines SW480 (A) and LS174T (B). Expression levels were normalized to β-Actin reference gene to determine relative mRNA expression, as shown. Protein levels of PDK isoforms in non-tumorigenic NCM460, IEC-18 and human CRC SW480 and LS174T cell lines were assessed via western blotting (C); α-tubulin was used as a loading control. Expected molecular weights are noted. (* p ≤ 0.05 vs. PDK2 and 4)

59

Figure 3. PDK1 and 3 levels are upregulated in response to treatment. qRT-PCR was used to assess PDK mRNA expression in human CRC cells SW480 (A) and LS174T (B) following 24 h 10 mM DCA exposure in normoxia and anoxia (< 0.1% O2). PDK mRNA expression levels were normalized to β-Actin reference gene, and then normalized to control samples to determine relative fold expression. Samples from 24 h treatments were subjected to western blot analysis to determine PDK1 and PDK3 protein expression. (a p ≤ 0.05 vs. control, b p ≤ 0.05 vs. DCA-treated, c p ≤ 0.05 vs. anoxia-treated)

60

Figure 4. Phosphorylation status of PDH at residues Ser293 and Ser300 varies in response to treatment. Western blot analysis was used to examine PDH phosphorylation following 24 h 10 mM DCA and/or anoxic (< 0.1% O2) treatment. Representative blots are shown (A) above graphical outputs for densitometric analysis of pPDH normalized to native PDH at residues Ser300 and Ser293 (B). (a p ≤ 0.05 vs. control, b p ≤ 0.05 vs. DCA-treated, c p ≤ 0.05 vs. anoxia-treated)

61

Figure 5. Mitochondrial response to DCA exposure is cell line-dependent. Mitochondrial activity (MA) of cells exposed to 10 mM DCA in normoxia and anoxia (< 0.1% O2) following 24 h was determined through mitochondrial staining using MitoTracker probes and subsequent flow cytometry analysis. Representative plots are shown in (A). Values displayed in each plot represent the % of cells with active mitochondria (i.e. in the upper right quadrant). Quantification of values from three biological replicates ± SEM is shown in (B). (a p ≤ 0.05 vs. control, b p ≤ 0.05 vs. DCA-treated)

62

Figure 6. DCA does not significantly alter metabolic status of human CRC cells. Conditioned media was collected from cells treated with 10 mM DCA under normoxia and anoxia (< 0.1% O2) following 24 h. Glucose consumption (A) and lactate production (B) was determined using commercially available kits. Graphs represent the average amount of glucose consumed and lactate produced per 104 cells from each cell line in various treatments from three biological replicates ± SEM. (a p ≤ 0.05 vs. control, b p ≤ 0.05 vs. DCA-treated)

63

CHAPTER III – METABOLIC DISRUPTION VIA 3-BROMOPYRUVATE IN HUMAN COLORECTAL CANCER CELLS

Introduction

Tumors upregulate key glycolytic enzymes to utilize and rely on the glycolytic

pathway for energy production regardless of oxygen tension in a phenomenon called the

Warburg effect (Koppenol et al, 2011; Warburg et al, 1927). This metabolic strategy

promotes tumorigenesis by re-routing glycolytic intermediates to fuel anabolic processes

to sustain rapid proliferation (Vander Heiden et al, 2009). Furthermore, highly glycolytic

tumors are associated with tumor aggressiveness (Simonnet et al, 2002).

The family of hexokinase enzymes function to catalyze the first irreversible step

of glycolysis in which glucose is phosphorylated to glucose-6-phosphate. While HK

isoforms are differentially expressed in various tissues, HKII is upregulated in many

tumors (Kwee et al, 2012; Palmieri et al, 2009; Wolf et al, 2011). Mito-HKII promotes

metabolic coupling of glycolysis and oxidative phosphorylation while also playing a

cytoprotective role by maintaining mitochondrial integrity (Majewski et al, 2004b;

Nakashima et al, 1986; Pastorino et al, 2002).

HKII expression is regulated by the AKT signaling pathway. AKT, or protein

kinase B, is a serine/threonine kinase that is often hyperactivated in human cancers. AKT

phosphorylates and stimulates the activity of target molecules involved in glucose

metabolism, cell proliferation, migration, and survival (Miyamoto et al, 2009a;

Miyamoto et al, 2009b; Scheid & Woodgett, 2001; Sussman et al, 2011; Thompson &

Thompson, 2004). AKT is able to phosphorylate HKII at residue Thr473 and increase

subsequent mitochondrial localization (Majewski et al, 2004a; Roberts et al, 2013).

64

Furthermore, the anti-apoptotic effects of AKT require mitochondrial-bound HKII

(Gottlob et al, 2001).

In Chapter II, I demonstrated that while DCA exposure in CRC cells led to

changes in metabolic signaling and variations in mitochondrial activity, 1) it did not

ultimately lead to significant changes in cellular metabolism and growth and 2) the doses

used were above physiological tolerance. While the effects of DCA at physiologically

doses have not shown promising results in pre-clinical studies, other compounds that

target metabolism in cancer cells in vitro have led to dramatic outcomes in preclinical and

human trials. 3-bromopyruvate (3BP) is a pyruvate analog with alkylating properties that

depletes cellular ATP levels and induces rapid cell death in tumor cells with limited

cytotoxic effects against normal cells. 3BP treatment led to eradication of tumors in all 19

rats tested without apparent cytotoxic effects, and the first human case report suggested

that 3BP was able to prolong survival in a cancer patient diagnosed with hepatocellular

carcinoma in 2012 (Ko et al, 2004; Ko et al, 2012). As previously mentioned, 3BP has

been reported to effectively kill other cancer cell types in vitro. 3BP is able to dissociate

and inhibit mitochondrial HKII function, thereby reducing ATP production and free

binding sites for pro-apoptotic molecules previously occupied by HKII to promote the

release of cytochrome c into the cytosol and induce eventual cell death (Chen et al,

2009b; Kim et al, 2008). However, the effects of 3BP on AKT signaling have not been

well studied.

In most solid tumors, imbalances between cancer cell proliferation and the

angiogenic response lead to some cancer cells being at a great distance from blood

vessels. This produces patchy regions of ischemic tumor microenvironment, due to a

65

reduction in both oxygen and nutrient availability, including glucose. This ischemia

ultimately leads to changes in metabolic pathways within cancer cells in response to a

reduction in glucose consumption that would otherwise be upregulated in well-perfused

regions of the solid tumor. Thus, it is important to study not only the impact of oxygen

tension on chemotherapeutic resistance, but the effects of hypoglycemia as well. Xu and

colleagues have shown that both HCT116 CRC cells and Raji lymphoma cells grown

under hypoxic conditions were more sensitive to 3BP than in normoxia (Xu et al, 2005),

but the effects of glucose exposure on 3BP responses are not well examined.

In this chapter, I examined the effects of 3BP on different human colorectal

cancer cell lines, with particular emphasis on examining the HKII/AKT signaling axis. I

also investigated whether glucose availability may play a role in targeting CRC cells with

3BP. I hypothesized that HKII expression will correlate with sensitivity to 3BP exposure

in human colorectal cancer cells and that a knockdown in its expression will decrease this

sensitivity. Furthermore, decreasing glucose availability in culturing conditions will lead

to an increase in 3BP resistance due to alterations in metabolic signaling pathways.




Cell culture and media-glucose reduction

Human colorectal cancer cell lines HCT116, CaCo2, SW480 and DLD-1 were

obtained from ATCC. Cells were maintained in high glucose DMEM supplemented with

66

10% FBS and incubated at 37oC in a humidified atmosphere containing 5% CO2 in room

air. Stock 1 M 3BP was prepared in H2O, filter-sterilized and aliquots were stored at -

80oC for future use.

Cells were routinely maintained under logarithmic growth by passaging them

every 3-4 days at a 1/8 dilution. At each passage, media-glucose concentrations were

reduced by 2.5 mM, by supplementing no glucose DMEM with 10% FBS and the desired

glucose concentration using a sterile stock of glucose. Cells were maintained at each

reduced media glucose concentration for at least 3 additional passages before further

experimentation.

Measurement of cell growth

Cell growth was approximated through crystal violet staining. Briefly, 5 x 103

cells were seeded into a 96-well plate and incubated overnight. Cells were then treated

with 1 – 100 µM 3BP for up to 72 h. Post-treatment, media was aspirated and 1% crystal

violet solution was added to each well and incubated for 10 min in room temperature.

Plates were then aspirated, rinsed and left to dry overnight. 10% acetic acid was used to

dissolve the crystals and the absorbance at 590 nm was recorded.

Protein isolation and western immunoblot analysis

Sodium orthovanadate (1 mM) was added 15 min prior to protein extraction.

Following treatment with 3BP or 50 µM etoposide, cells were lysed on ice with lysis

buffer supplemented with 2 µg/ml aprotinin, 1 mM PMSF and phosphatase inhibitor

cocktail II. Samples were centrifuged at 12,000 x g for 15 min at 4oC, and supernatant

67

was aliquoted and stored at -80oC for future use. Protein was quantified using the Bio-

Rad DC Protein Assay Kit. 30 µg of total protein was resolved in a 10% polyacrylamide

gel using SDS-PAGE then transferred to a PVDF membrane with wet transfer buffer at

100 V for 2 hr. Gels used to subsequently detect caspase-3 expression were transferred

onto PVDF membranes using a Trans-Blot SD Semi-Dry Transfer Cell apparatus (Bio-

Rad) with semi-dry transfer buffer at 16 V for 30 min. Following transfer, membranes

were blocked for 1 h at room temperature in 5% non-fat dry milk diluted in TBST

(blocking solution) followed by an overnight incubation in blocking solution with

primary antibodies. After washing, membranes were incubated for 1 h at room

temperature with the appropriate peroxidase-conjugated secondary antibody, washed, and

subjected to Luminata Forte chemiluminescent substrate. Membranes were imaged using

the ChemiDoc XRS+ system and densitometry was performed using Image Lab software.

Primary antibodies used included rabbit anti-pAKT Ser473 (1:1,000), anti-pAKT Thr308

(1:1,000), anti-pan-AKT (1:1,000), anti-caspase 3 (1:1,000), anti-HKII (1:5,000), anti-

pPDK1 (1:1,000), anti-PDK1 (1:1,000), and anti-PTEN (1:1,000), mouse anti-α-tubulin

(1:100,000), and rabbit or mouse HRP-labeled secondary antibodies (all 1:10,000).

Cell death assay

Cell death was determined through annexin V/PI staining using the Alexa Fluor®

488 annexin V/Dead Cell Apoptosis Kit following the manufacturer’s protocol. Briefly, 5

x 105 cells were seeded onto 6-well plates and incubated overnight. Following 48 h

treatment with 3BP, cells were trypsinized, washed, resuspended in annexin-binding

buffer containing AV and PI dyes and left to incubate for 15 min at room temperature,

68

protected from light. Cells were subsequently analyzed using a FACScanTM flow

cytometer. In total, 1 x 104 events were counted per sample. Fluorescence scatter plots

were gated and analyzed using the Cyflogic software (Terho & Korkeamaki, 2008) to

determine the percentage of viable and dead cells.

DNA fragmentation

Following 48 h treatment with 3BP or etoposide, cells were trypsinized, pelleted

and stored at -80oC. DNA extractions were carried out using the Omega Bio-Tek Tissue

DNA Kit according to the manufacturer’s protocol with the following changes: 1)

samples were left to precipitate in ethanol overnight, 2) samples were subsequently

centrifuged at 6,000 x g for 1 min, 3) DNA Wash Buffer step was carried out once and

centrifuged for 3 min at 20,000 x g for 3 min, 4) samples were eluted in 50 µl elution

buffer. DNA samples were run on either a 0.8% or 2% agarose gel at 150 V for 2 hr or

100 V for 1.5 h, respectively, and visualized using the ChemiDoc XRS+ system.

RNA interference

Knockdown experiments were performed by transient transfection of HKII

siRNA using INTERFERin according to the manufacturer’s protocol. Briefly, 5 x 104

cells were seeded into 12-well plates and incubated overnight. Cells were transfected

with 10 nM siRNA for 24 h in Opti-MEM and subsequently allowed to recover in high

glucose DMEM for the times indicated. The sense sequences of double-strand siRNA

used were previously published (Yuan et al, 2008):

siHKIIa: 5’-GGAUAAGCUACAAAUCAAA[dT][dT]-3’ siHKIIb: 5’-CGGGAAAGCAACUGUUUGA[dT][dT]-3’

69

3BP dose response following HKII knockdown was conducted as follows. 2 x 105

cells were seeded into 60 mm2 plates and incubated overnight. Cells were transfected

with 10 nM siRNA for 24 h in Opti-MEM and subsequently allowed to recover in high

glucose DMEM for 24 h. A mixture of both HKII siRNA sense sequences was used.

Following its recovery, a 4 h 3BP dose response was conducted.


Data were presented as means of independent measurements +/- SE. GraphPad

Prism v6.0 software was used to perform statistical analysis. One-way ANOVA followed

by Tukey’s post-hoc tests was performed to determine differences between means within

each cell line for flow cytometry and cell growth. Non-parametric Kriskal-Wallis

ANOVA followed by Dunn’s post-hoc tests was performed to determine differences

between means within each cell line for western blot densitometric analysis. At least

three biological replicates were used for each analysis unless otherwise specified, and

treatments were considered significantly different if a p-value ≤ 0.05 was achieved.

Results

HKII expression is correlated with 3BP sensitivity in human CRC cells

The protein expression of HKII by various human CRC cell lines was screened

via western blot analysis. All four cell lines showed HKII expression to varying degrees

(Figure 7A). HCT116 and CaCo2 cells showed moderate and low HKII expression,

respectively, while SW480 and DLD-1 cells showed high HKII expression.

Corresponding IC50 values were determined through 72 h crystal violet cell growth assays

70

(Figure 7B) and were as followed: HCT116 = 22.5 ± 0.7 µM, CaCo2 = 36.6 ± 2.1 µM,

SW480 = 16.9 ± 1.0 µM, DLD-1 = 16.9 ± 1.3 µM. 3BP sensitivity of HCT116, SW480,

and DLD-1 cells were statistically different from CaCo2, but not from each other (p <

0.05). Thus, high HKII-expressing SW480 and DLD-1 cells were significantly more

sensitive to the effects of 3BP than low HKII-expressing CaCo2 cells.

3BP-induced cell death is dose dependent

To determine the cytotoxic effects of 3BP on CRC cells, cell death was measured

using AV/PI staining and subsequent flow cytometry analysis. Following 48 h 3BP

treatment, cell death was observed in all cell lines examined (Figure 8). Cells staining

positive for AV was indicative of an apoptotic event while PI positivity correlated to a

necrotic event. Cells treated with 50 µM etoposide served as a positive control for

apoptosis and are shown in APPENDIX III. In all cell lines examined, both forms of cell

death were induced upon 3BP exposure. Limited cell death was observed with 5 µM 3BP

treatment. At higher doses (30 and 50 µM), the percentage cell death observed in each

cell line was associated with levels of HKII expression previously examined. In moderate

and low HKII expressing HCT116 and CaCo2 cell lines, significant cell death was only

observed following 50 µM 3BP treatment (56.8 and 26.9%, respectively). In high HKII

expressing SW480 and DLD-1 cell lines, significant cell death was observed at both 30

µM (31.1 and 78.1%, respectively) and 50 µM 3BP treatment (90.1 and 92.9%,

respectively) (Figure 8).

SW480 cells showed considerable caspase-3 cleavage even under control

conditions. HCT116 cells were most resistant to etoposide-induced caspase-3 cleavage

71

compared to CaCo2, SW480, and DLD-1 cells. Following 3BP exposure, no cleaved

caspase-3 was observed in HCT116 or CaCo2 cells while limited caspase-3 cleavage was

observed in SW480 and DLD-1 cells (Figure 9A). However, cells exposed to 50 µM 3BP

showed overall reduced cellular protein as evidenced by reduced full length caspase-3 but

no detectable cleaved product, and reduced α-tubulin expression. This observation,

suggestive of cellular necrosis, was verified using DNA fragmentation analysis, where 50

µM 3BP treatment on SW480 cells led to DNA smearing on the agarose gel, and no

obvious nucleosome laddering expected with apoptotic cell death (Figure 9B, C).

3BP induces rapid AKT phosphorylation at residue Thr308

Through a 24 h time course study of 3BP on AKT activation, maximum

phosphorylation of AKT at residue Thr308 was observed after 4 h 3BP treatment (Figure

10A). Dose-dependent changes in AKT phosphorylation following 3BP treatment were

observed at residue Thr308 with no changes at residue Ser473 (Figure 10B,C;

APPENDIX IV). Native AKT expression was not affected. The trends in AKT

phosphorylation at residue Thr308 matched the sensitivity to 3BP in different cell lines;

phosphorylation occurred at lower doses in cell lines highly expressing HKII (SW480

and DLD-1) (Figure 10C). In CaCo2 cells, AKT phosphorylation appeared to remain

constitutively active regardless of 3BP treatment (Figure 10B). Phosphorylation of AKT

at residue Thr308 occurs through PIP3-mediated recruitment of AKT and

phosphoinositide-dependent kinase-1 (PDK1) to the plasma membrane (Alessi et al,

1997). Similar to AKT phosphorylation, changes in the expression of these proteins

occurred in a dose-dependent manner. Changes in PTEN expression were limited to

72

SW480 and DLD-1 cells, where a decrease in its expression was observed at higher 3BP

doses (Figure 10C). At 3BP doses that induced cytotoxic effects, the molecular weight of

PDK1 shifted by approximately 3 kDa. In SW480 and DLD-1 cells, an increase in PDK1

expression at higher 3BP doses was observed.

Glucose availability affects 3BP sensitivity in human CRC cells

HCT116 and DLD-1 cells were adapted and maintained in DMEM with varying

glucose concentrations. Cells were maintained in DMEM containing 25, 5.5 and 1 mM

glucose for subsequent experimentation, denoted as 'high', 'normal' and 'low' glucose

conditions, respectively. Western blot analysis showed that HKII levels decreased with

lower media glucose levels in both cell lines examined (Figure 11A,B). Differences in

HKII expression between cells grown under high vs. normal glucose were more profound

in DLD-1 cells. Both CRC cell lines proliferated exponentially under high and normal

glucose concentrations, but growth was limited in cells cultured in low glucose over 72 h

(Figure 11C). Cells grown in normal glucose showed an increase in 3BP resistance vs.

cells grown in high glucose. Significant differences in cell growth were observed

between high and normal glucose for HCT116 cells treated with 30 and 40 µM 3BP and

DLD-1 cells treated with 20 and 30 µM 3BP (Figure 11D). This pattern is consistent

with the finding that DLD-1 cells are more sensitive to 3BP than HCT116 cells.

HKII knockdown does not lead to changes in 3BP sensitivity

To determine whether HKII was the reason for the differences observed in 3BP

sensitivity between different CRC cells, HKII expression was knocked down in HCT116

73

and DLD-1 cells using transient RNAi transfection. Effective knockdown of HKII

expression was observed in both cell lines (Figure 12A). After 24 h siRNA transfection,

treatment media was replaced with DMEM to allow subsequent assessment of 3BP

effects in standard treatment media. Protein levels remained knocked down for 48 h post-

transfection with a slight increase in HKII expression after 72 h of post-transfection

recovery (Figure 12B). Given this stability in HKII knockdown for at least 48 h post-

transfection, a 4 h 3BP dose-response was conducted on both HCT116 and DLD-1 cells

after the 24 h recovery period (i.e. samples were collected after 28 h knockdown). There

were no differences in 3BP sensitivity as revealed by AKT phosphorylation when

comparing scrambled controls to HKII knockdown cells in either HCT116 (Figure 12C)

or DLD-1 cells (Figure 12D). The dose-response patterns observed mimicked those

previously seen in parental HCT116 and DLD-1 cells (Figure 10B,C), suggesting that,

contrary to expectation, a reduction in HKII expression did not lower 3BP sensitivity in

the cell lines examined.

Discussion

While the promising anti-cancer effects of pyruvate/lactate analog 3-

bromopyruvate on cancer cells in vitro have been well documented, the molecular

mechanisms surrounding its potent effects remain understudied. In this chapter, I sought

to investigate the potential link between HKII expression and the sensitivity of human

colorectal cancer cells towards 3BP, and whether glucose availability may factor into

these cytotoxic effects. I found that although limiting glucose availability in the culturing

conditions led to a reduction in HKII expression and a decrease in 3BP sensitivity, the

74

expression of HKII alone cannot account for the differences in 3BP sensitivity observed

amongst the different CRC cells examined.

Due to its high affinity for glucose compared to other isoforms, HKII is

upregulated during tumorigenesis. This increases glucose consumption and alleviates the

metabolic requirements for glycolytic intermediates, thus supporting key anabolic

processes that sustain rapid cellular proliferation (Kwee et al, 2012; Palmieri et al, 2009;

Wolf et al, 2011). I examined the expression of HKII in 4 different CRC cell lines and

observed that the expression of HKII correlated with their reported aggressiveness in vivo

(Flatmark et al, 2004). Western blot results reflected the data generated from 3BP IC50

determination via crystal violet staining, where high HKII-expressing SW480 and DLD-1

cells were significantly more sensitivity to the cytotoxic effects of 3BP versus low HKII-

expressing CaCo2 cells. Like in Chapter II, crystal violet was used to monitor the effects

on cell growth as 3BP has the potential to interfere with the tetrazolium reduction assays

(i.e. MTT, WST-1) generally used to evaluate cytotoxicity (Ganapathy-Kanniappan et al,

2010).

I next sought to determine the mechanisms of 3BP-induced cytotoxicity. AV/PI

labeling was used to determine whether 3BP triggered cell death in the cells examined. I

observed that, much like the results from crystal violet staining, the amount of cell death

induced in CRC cells was correlative to HKII levels. In all cell lines treated, a significant

portion of cells following > 50 µM 3BP exposure appeared in the AV+/PI+ quadrant,

signifying either late apoptosis or necrosis (Kroemer et al, 1998). Analyzing AV and PI

staining individually did not provide further insight into the mechanisms of cell death

induced by 3BP (See APPENDIX V).

75

As AV/PI labeling was insufficient to fully elucidate the route of 3BP-induced

cell death, I performed additional assays including detection of caspase-3 cleavage and

DNA nucleosome fragmentation, both characteristics of apoptotic events (Zhivotovsky et

al, 2001). Activation of initiator caspases (caspase 8 and 9) by either ‘extrinsic’ or

‘intrinsic’ apoptotic signaling pathways, respectively, leads to the cleavage and activation

of executioner pro-caspase 3, an common event for both 'intrinsic' and 'extrinsic'

apoptotic pathways (McIlwain et al, 2013). In high HKII expression CRC cells, I found a

slight induction of caspase-3 cleavage following 30 µM 3BP treatment. This effect

however was lost with 50 µM 3BP treatment, in which total protein was also reduced,

presumably due to microtubule degradation during necrotic cell death. Calvino and

colleagues observed the same trend in myeloid leukemia cells: increase in cleaved

caspase-3 following 30 µM 3BP exposure which was lost at 60 µM (Calvino et al, 2014).

Programmed cell death is not solely dependent on the cleavage and activation of

caspases. BH3 domain-only proteins, such as tBID, BIM and BAD can kill cells

independently of caspase activation (Cheng et al, 2001). A common feature in all forms

of programmed cell death is DNA fragmentation (Zhivotovsky et al, 2001). However, I

was unable to visualize the presence of small DNA fragments (representing nucleosomes)

in 3BP-treated samples, suggesting minimal apoptotic induction following 30 µM 3BP

treatment. However, there was visible DNA smearing in 50 µM 3BP treated SW480 cells,

suggesting a necrotic event (Zhivotovsky et al, 2001). Combining the AV/PI labeling,

caspase-3 cleavage, and DNA fragmentation results is thus consistent with apoptosis

induction under lower doses of 3BP and necrosis at higher doses.

76

The specific dose at which 3BP was able to induce these effects was dependent on

the cell line. Opinions on the mechanisms of 3BP-induced cell death vary in the

literature. Necrosis was induced in both hepatocyte and cancer cell lines following 50 µM

3BP treatment (Ihrlund et al, 2008; Kim et al, 2008). Moreover, a 60% reduction in

intracellular ATP levels was observed in hepatoma cells, further suggesting that 3BP

induces ATP depletion-dependent necrosis at higher concentrations (Kim et al, 2008). In

myeloid leukemic cells, both apoptosis and necrosis were induced at 20-30 µM and a

pure necrotic response was seen at 60 µM (Calvino et al, 2014). Other studies report the

ability of 3BP to induce apoptosis at 100 µM (Gwak et al, 2005; Yu et al, 2012). The

variation observed in 3BP sensitivity and mechanisms of cell death induction may thus be

attributed to both the dose and the metabolic phenotype of the cell models analyzed.

AKT plays a role in the localization of HKII to the mitochondria (Majewski et al,

2004a). In addition to HKII phosphorylation, AKT is able to phosphorylate and inhibit

glycogen synthase kinase-3β (GSK3β), a constitutively active serine-threonine kinase

able to phosphorylate VDAC and prevent HKII binding. Thus, inhibition of GSK3β leads

to increased mitochondrial localization of HKII (Gall et al, 2011; Miura & Tanno, 2012;

Pastorino et al, 2005; Rasola et al, 2010). I examined the effects of 3BP on

phosphorylation of AKT and its upstream effectors. Initial studies using an approximate

IC50 dose for DLD-1 showed that the serine residue 473 remained constitutively

phosphorylated. However, maximum AKT phosphorylation was induced after 4 h of 3BP

exposure at residue Thr308. Furthermore, the effects of 3BP on AKT phosphorylation

occurred at residue Thr308 not residue Ser473 in all 4 CRC cell lines examined. Rapid

induction of AKT phosphorylation following 3BP treatment was previously shown at

77

residue Ser473 in myeloid leukemic cells (Calvino et al, 2014). These changes in AKT

phosphorylation correspond with expression of the upstream signaling molecule PTEN in

my CRC cells. Phosphorylation of PIP2 to activated PIP3 recruits AKT and PDK1 to the

plasma membrane. This in turn causes a major conformational change in AKT to allow

PDK1 to phosphorylate residue Thr308 in the activation loop of AKT (Alessi et al, 1996;

Alessi et al, 1997; Calleja et al, 2007). Maximum AKT activity is dependent on

phosphorylation of both Thr308 and Ser473 residues, the latter of which is

phosphorylated by mTORC2 (Alessi et al, 1996; Sarbassov et al, 2005). PTEN

dephosphorylates PIP3 and prevents AKT membrane recruitment and phosphorylation

(Stambolic et al, 1998).

Interestingly, I observed a shift in the molecular weight of PDK1 by

approximately 3 kDa at cytotoxic 3BP doses in cell lines. In SW480 and DLD-1 cells,

PDK1 expression increased in a dose-dependent manner. Scheid and colleagues have

previously shown that a shift in the molecular weight of PDK1 is a result of

hyperphosphorylation in response to survival signaling. Treatment of HEK 293 cells with

IGF-1 caused a decrease in PDK1 mobility, and co-treatment with calf alkaline

phosphatase reversed this effect, supporting a phosphorylation event (Scheid et al, 2005).

PDK1 has shown to be constitutively phosphorylated on at least five serine residues.

Residue Ser241 is located in the activation loops of the PDK1 kinase domain, and its

phosphorylation is required for PDK1 activity (Casamayor et al, 1999). Native PDK1

levels reflected the changes observed in PDK1 phosphorylation at residue Ser241. This

would suggest that there was an increase in PDK1 levels as well as an increase in

phosphorylation (or possibly other posttranslational modification) at various residues. In

78

agreement with this finding, Schied and colleagues also observed changes in PDK1

phosphorylation patterning changes in AKT phosphorylation at residue Thr308 (Scheid et

al, 2005).

As previously discussed, the in vitro conditions may have a profound effect on the

outcome when testing the efficacy of anti-cancer agents. The concentration of glucose

present in my culturing media (4.5 g/L) is nearly 5 fold greater than what is considered

physiological blood sugar levels. As HKII expression is a function of the rate of glucose

consumption, I wanted to determine whether glucose availability would alter HKII

expression and subsequently change 3BP sensitivity. HCT116 and DLD-1 cells

maintained in high, normal and low media glucose concentrations were subjected to 3BP

treatment. The normal glucose concentration (5.5 mM) used for these studies is

equivalent to that of normal human physiological steady-state conditions (Freckmann et

al, 2007). I observed a decrease in HKII expression with decreasing glucose

concentrations. This change also led to a change in 3BP sensitivity; cells treated in

normal glucose concentrations were most resistant to the effects of 3BP. It has previously

been shown that HCT116 cells induced to exhibit an increased glycolytic phenotype via

oligomycin treatment were also more sensitive to the effects of 3BP (Sanchez-Arago &

Cuezva, 2011).

To definitively confirm whether levels of HKII expression alone can explain the

differences observed in 3BP sensitivity, I elected to knock down HKII expression in

HCT116 and DLD-1 cells. I hypothesized that reduction in HKII expression would lead

to increased resistance against 3BP. Surprisingly, effective and sustained HKII

knockdown in both cell lines did not lead to any changes in AKT phosphorylation

79

induced by 3BP. Thus, although high HKII-expressing CRC cells were more sensitive to

the effects of 3BP versus low HKII-expressing CRC cells, this observation cannot

explain the differences in 3BP-sensitivity. 3BP affects various enzymes involved in both

glycolysis and mitochondrial respiration, including SDH, GADPH and PGK (Pereira da

Silva et al, 2009). It has previously been suggested that an ester derivative of 3BP, 3-

bromopyruvate propyl ester, exhibits a more potent inhibitory effect on GADPH than on

HKII in CRC cells HCT116 and HT29 (Tang et al, 2012). The fact that 3BP can affect

multiple glycolytic enzymes may explain the lack of change in 3BP sensitivity in HKII

knockdown cells. The differences in 3BP sensitivity when cells were treated under

different glucose concentrations may thus be explained by changes in the activity of these

other glycolytic enzymes.

Along with its ability to inhibit tumor growth as a single agent, 3BP has also been

shown to exert potentiating effects in combination with platinum drugs cisplatin and

oxaliplatin (Ihrlund et al, 2008) and with doxorubicin (Nakano et al, 2011). Wintzell and

colleagues showed that cisplatin-resistant ovarian carcinoma SKOV3 cells were more

sensitive to 3BP than parental cells, and in both cell lines, 3BP was able to potentiate the

effects of cisplatin in vitro (Wintzell et al, 2012). Similarly, Isayev and colleagues

showed that 3BP exposure reverted gemcitabine resistance in various pancreatic ductal

adenocarcinoma cell models and provided an additive effect in combination with

gemcitabine in vivo (Isayev et al, 2014). The mechanisms of 3BP potentiation are not

known, but can be speculated upon. The loss of intracellular ATP upon 3BP exposure

may affect a number of processes that are critical to chemoresistance, including efficient

DNA repair or the function of multi-drug resistance ATP-binding cassette transporters

80

(Nakano et al, 2011). It is also possible that 3BP-induced dissociation of HKII from the

mitochondria may lower mitochondrial integrity and sensitize cells to pro-apoptotic

factors binding more freely onto open sites on VDAC upon exposure to other anti-cancer

drugs.

In summary, 3BP is a promising anti-cancer agent capable of inducing rapid ATP

loss and cell death in numerous cancer types. Since HKII has been proposed to be one of

its principal targets, I hypothesized that HKII expression would correlate with 3BP

sensitivity in CRC cell lines, and reduction in HKII expression would reduce cell

sensitivity. I showed that high HKII-expressing CRC cells displayed increased 3BP

sensitivity, and that alteration of culture conditions via glucose availability affected HKII

expression. However, RNAi studies showed that HKII expression could not explain for

the variability in 3BP sensitivity between different cell lines. The mechanism by which

lowered glucose availability increases 3BP resistance is thus not known. Elucidating the

alterations in metabolic pathways that may lead to the observed resistance could provide

better understanding of the impact of the tumor microenvironment on colorectal cancer

cells, and clinical insight to the mechanisms of chemoresistance.

81

Figure 7. HKII expression is correlated to 3BP sensitivity. HKII protein expression in CRC HCT116, CaCo2, SW480, and DLD-1 cells was assessed via western blot (A). Crystal violet staining assay was used to determine 3BP IC50 values in HCT116, CaCo2, SW480, and DLD-1 cells (B). IC50 values shown represent the means of three biological replicates ± SEM.

82

Figure 8. 3BP-induced cell death is cell-line dependent. Following 48 h 3BP treatment, cell death was assessed through AV/PI labeling and subsequent flow cytometry analysis. Representative plots are shown in (A) and quantification of live vs. dead cells from three biological replicates ± SEM are shown in (B). Viable cells appear in the lower left quadrant. 3BP induced significant cell death in all four cell lines. (* p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001, **** p ≤ 0.0001)

83

Figure 9. Type of cell death induced by 3BP is dose-dependent. Caspase-3 cleavage and DNA fragmentation was assessed to further elucidate the mechanisms of 3BP-induced cell death. Caspase-3 cleavage was examined through western blot analysis (N=3) (A). An image of a longer exposed membrane to better detect cleaved caspase-3 is shown below each full caspase-3 blot. DNA fragmentation was examined through gel electrophoresis in SW480 cells following 48 h 3BP treatment using 0.8% (B) and 2% (C) gels (N=2). Etoposide (E; 50 µM) was used as a positive control for caspase-3 cleavage and DNA fragmentation; note nucleosome laddering in etoposide treated cells in (C) indicative of apoptosis.

84

Figure 10. 3BP treatment leads to rapid AKT phosphorylation at residue Thr308. Time course assessment of AKT phosphorylation in response to 3BP. Example from DLD-1 cells shown in (A). Western blot analysis of proteins involved in AKT phosphorylation following 4 h 3BP dose-response in HCT116, CaCo2, SW480, and DLD-1 cells (B,C).

85

Figure 11. Glucose availability alter 3BP sensitivity in CRC cells. HCT116 and DLD-1 cells were treated with 3BP under different media glucose concentrations. HKII expression was assessed via western blot analysis. Representative blots from three biological replicates (A) and graphical outputs for densitometric analysis of HKII expression normalized to α-tubulin from three biological replicates ± SEM are shown (B). Growth curves for CRC cells grown under different media glucose concentrations in the presence or absence of 3BP over 72 h (C). Dose-effects of 3BP on cell growth following 72 h treatment in CRC cells under different media glucose concentrations (D). (* p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001, **** p ≤ 0.0001 vs. control (0 µM 3BP))

86

Figure 12. HKII expression knockdown did not lead to changes in AKT phosphorylation following 3BP exposure. HKII expression in HCT116 and DLD-1 cells were knocked down using RNA interference. Both siRNA sequences used showed efficient knockdown of HKII expression following 24 h transfection (A). Representative blots from DLD-1 cells are shown. Recovery time of CRC cells following siRNA transfection (B). Expression of AKT and related proteins following 4 h 3BP dose response on HCT116 (C) and DLD-1 cells (D). siRNA experiments were performed twice with equivalent results.

87

CHAPTER IV – HEXOKINASE II AS A POTENTIAL PROGNOSTIC BIOMARKER FOR HUMAN COLORECTAL CANCER

Introduction

CRC is the second leading cause of cancer-related deaths and the third most

common cancer in Canada (Canadian Cancer Society). In 2015, it is estimated that CRC

will account for 17.6% of all new cancer cases and 25.3% of all cancer-related deaths

(Siegel et al, 2015). Decreasing mortality rates have been attributed to early screening,

reduced prevalence of risk factors, and/or improved treatments (Bosetti et al, 2011;

Edwards et al, 2010). However, outcome for patients with advanced and metastatic

disease remains poor.

In 2011, deregulating cellular energetics was identified as an emerging hallmark

of cancer (Hanahan & Weinberg, 2011). Many tumors upregulate key glycolytic enzymes

to utilize and rely on the glycolytic pathway for energy production regardless of oxygen

tension in a phenomenon called the Warburg effect (Warburg et al, 1927). Studies have

correlated the glycolytic state of cancer cells with tumor aggressiveness (Simonnet et al,

2002). Measurement of this unique metabolic change specific to cancer cells using FDG-

PET is widely used as a diagnostic tool for assessing various cancers including CRC

(Artiko et al, 2015; Kruse et al, 2013). Therefore, specific glycolytic enzymes may serve

as possible biomarkers of CRC aggressiveness, providing useful prognostic tools.

As discussed in Chapter III, upregulation of HKII is considered a consequence of

metabolic re-programming in cancer, and has been shown in breast carcinoma (Sato-

Tadano et al, 2013), laryngeal squamous cell carcinoma (Chen et al, 2014), glioblastoma

multiforme (Wolf et al, 2011), and gastric adenocarcinoma (Qiu et al, 2011). Its

88

assessment in CRC has been very limited to date; immunohistochemical analyses on 104

surgically resected CRC samples showed a positive correlation between HKII expression

with extensive tumor diameter, advanced tumor depth, presence of lymph node

metastasis, and shorter recurrence-free survival (Hamabe et al, 2014).

The rapid and disorganized proliferation associated with tumorigenesis leads to

ischemic regions within solid tumors (Raghunand et al, 2003). Cells within these regions

are exposed to poor blood flow leading to transient or chronic hypoxic and hypoglycemic

conditions. The impact of these microenvironments on the effectiveness of anti-cancer

therapies is not well understood. HKII expression is upregulated under hypoxic

conditions as a result of stabilized HIF1α transcription factor binding to the promoter

region of HKII, activating gene expression (Srinivas et al, 2001). Results from Chapter

III showed that limiting glucose availability in vitro led to a decrease in HKII expression.

The family of carbonic anhydrase (CA) enzymes catalyzes the hydration of carbon

dioxide to bicarbonate and proton (Potter & Harris, 2004). CAIX is a HIF-inducible gene

that has been proposed as an ischemic marker and is overexpressed colorectal cancer

(Loncaster et al, 2001; Niemela et al, 2007).

Although I demonstrated that HKII expression does not affect cellular sensitivity

to 3BP, there is still therapeutic potential in targeting HKII due to its importance in

aerobic glycolysis. Therefore, it is still worth examining the HKII expression in CRC

samples and determining its value as a prognostic marker and also as a chemotherapeutic

target. In this study, I evaluated the expression of HKII and CAIX in FFPE specimens of

colorectal cancer tumors and assessed its relationship to clinicopathologic or prognostic

relevance. Based on findings from other studies, I hypothesized that high HKII

89

expression in colorectal cancer tumor samples would be associated with shorter

progression-free status and worse outcome.




Samples

Sixty FFPE colorectal cancer samples used for this study were gathered by the

Ontario Tumor Bank, Ontario Institute for Cancer Research. Samples were collected

between 2005 and 2011. The patients’ medical records were reviewed and general

information, including sex, age range at diagnosis, histological grade, and TNM staging

(Edge & Compton, 2010), are provided in Table 3. Follow-up data were retrieved for

survival analysis. The last follow-up was May, 2013. Cases were anonymized and I was

blinded to tumor and patient details until completion of image capture and analysis.

Cell lines and immunoblot analysis

Human colorectal cancer cell lines HCT116 and DLD-1 were obtained from

ATCC and were maintained in high glucose DMEM supplemented with 10% FBS and

incubated at 37oC in a humidified atmosphere containing 5% CO2 in room air. Anoxic

conditions were achieved using a Modular Incubator Chamber modified to allow

continuous flushing of the chamber with a humidified mixture of 95% N2 and 5% CO2.

Cells were lysed on ice with lysis buffer supplemented with 2 µg/ml aprotinin, 1 mM

90

PMSF and phosphatase inhibitor cocktail II, following the manufacturer’s protocol.

Protein was quantified using the Bio-Rad DC Protein Assay Kit. 30 µg of total protein

was resolved in a 10% polyacrylamide gel using SDS-PAGE then transferred to a PVDF

membrane with wet transfer buffer at 100 V for 2 hr. Membranes were blocked for 1 h at

room temperature in 5% non-fat dry milk diluted in TBST followed by an overnight

incubation in blocking solution with primary rabbit anti-HKII (1:5,000), mouse anti-

CAIX (1:1,000) and mouse anti-α-tubulin (1:100,000). HRP-conjugated goat anti-rabbit

or mouse antibodies (1:10,000) were incubated for 1 h at room temperature. Luminata

Forte chemiluminescent substrate was used for visualization.

Immunofluorescent staining

FFPE histological sections were deparaffinized in xylene, rehydrated in

isopropanol, and washed with H2O. Antigen retrieval was performed by incubating

samples in 10 mM sodium citrate buffer (pH 6.0) at 95oC for 15 min and subsequently

left to cool at room temperature for 30 min. Sections were blocked with 1:1 solution of

5% goat serum in PBS + 0.1% Tween (PBST) and protein block solution for 1 hr at room

temperature. Sections were subsequently exposed to primary antibodies, rabbit mAb

HKII (1:400) and mouse mAb CAIX (1:400), diluted in PBST and incubated at 4oC

overnight in a humidified chamber. Sections were washed with PBST and then incubated

for 1 hr at room temperature with goat anti-rabbit Alexa 488-conjugated (1:300) and goat

anti-mouse Cy3-conjugated (1:200) antibodies diluted in PBST. Slides were

counterstained using DAPI and washed with PBS twice for 5 min, and coverslips were

91

mounted with Fluorescence Mounting Media. Slides were stored flat in the dark at 4oC

and imaged the following day.

Imaging and quantification of immunofluorescence

Slides stained with HKII and CAIX antibodies were photographed with a QICAM

camera (QImaging, BC, Canada) with a 40X objective at 1.4 megapixel resolution, using

excitation and barrier filter sets for fluorescein (green emission), rhodamine (red

emission) and DAPI (blue emission). Images were saved as TIFF files. In total, 5 images

were captured per slide, covering different regions of the sample in an unbiased,

systematic fashion. Brightness and contrast adjustments were applied using Photoshop

CS6 (Adobe) on all images; images from the same slides were adjusted using the same

parameters. Fiji (Schindelin et al, 2012) was used to process and analyze the images to

determine HKII and CAIX staining. For CAIX staining in all images, a minimum

threshold value of 20 was used in the software and the percent area stained was

subsequently calculated. The Mean and SD from the 5 images were calculated for each

sample and used for subsequent analysis. A variation of the semi-quantitative H-Score

immunohistochemistry evaluation method (Hatanaka et al, 2003) was used to assess

HKII staining, and given the term “F-Score”. The percent area of HKII staining was

determined at 2 minimum threshold values (25 and 85), and the areas were added

together to form the F-Score as denoted in the following equation:

F-Score = (% area stained at threshold=25) + (% area stained at threshold=85)

The Mean and SD from the 5 images per sample were calculated and used for subsequent

analysis.

92


GraphPad Prism v6.0 software was used for statistical analysis. Correlations

between staining parameters and patients’ characteristics were analyzed with the Fisher’s

exact test. Progression-free survival and overall survival were calculated per the Kaplan-

Maier method and were compared between two groups by using the logrank test. Patients

lost to follow-up, and those still progression-free and/or still alive at last follow-up were

right-censored for survival analysis. Results were considered significantly different when

a two-sided p value ≤ 0.05 was achieved.

Results

Patient characteristics

Clinicopathological characteristics of the study cohort are shown in Table II.

Briefly, there were slightly more men (N=34, 57%) than women in this study. The cohort

contained both low grade (I & II; N=30, 50%) and high grade (III & IV; N=28, 47%)

tumors; grading for two cases was not available. Sixty three percent of patients presented

with low TNM tumors (Stage 1, 2 & 3; N=38) while 22% presented with high TNM

tumors (Stage 4; N=13). TNM stage was not available for 15% (N=9) of the cases.

Immunoblot and immunofluorescent detection of HKII and CAIX

Human colorectal cancer cells showed an upregulation in both HKII and CAIX

expression following 24 h anoxic exposure in vitro (Figure 13). IF revealed

heterogeneous staining for both HKII and CAIX in FFPE sections of human CRC;

93

representative staining patterns are shown in Figure 14. Both markers were localized to

the cytoplasm in neoplastic cells. HKII staining was predominantly located in areas of

low CAIX staining, suggesting HKII expression is primarily present in non-ischemic

regions of the tumors.

Levels of CAIX and HKII staining

Computational analysis was used to examine both CAIX and HKII staining.

Representative analyses for areas staining low or high for CAIX are shown on the left

and right panels of Figure 15, respectively. The median area stained for CAIX was

15.5% with a minimum of 1.3% and maximum of 65.5%, and this median value was used

to dichotomize the cases into ‘low’ and ‘high’. For HKII staining the F-Score was

obtained by the summation of percent area HKII staining at two minimum thresholds, of

25 and 85, as represented in Figure 16. Examples of HKII staining and corresponding F-

Scores are shown in Figure 17. The median F-Score of the five images per sample was

24.7 with a minimum of 3.4 and maximum of 71.5. This median F-Score was used to

dichotomize the cases into ‘low’ and ‘high’.

Association of CAIX and HKII expression with patient characteristics

Age was significantly associated with 'high' CAIX expression (p=0.038). The

odds of patients 70 years of age or older staining high for CAIX expression was

approximately 3.5 times greater than patients younger than 70. HKII expression was not

statistically associated with sex, age at diagnosis, histological grade, TNM stage, nor

primary tumor size (Table III).

94

Association of CAIX and HKII expression with patient outcome

No significant differences in PFS were observed between patients with low CAIX

expression (median PFS, 2132 days) and those with high expression (median PFS, 548

days); p = 0.16) (Figure 18A). Likewise, no significant differences in OS were observed

between patients with low CAIX expression (median OS, undefined) and those with high

expression (median OS, 1011 days); p = 0.085) (Figure 18B), although both PFS and OS

trended towards better outcome with low CAIX. Kaplan-Meyer survival analysis showed

a significant association between HKII expression and median survival duration of CRC

patients (Figure 18C,D). Specifically, patients who scored high for HKII had

significantly longer progression-free survival (Undefined vs. 497 days; p=0.016) and

overall survival (Undefined vs. 1060 days; p=0.025) than patients who scored low for

HKII.

Association of stromal HKII staining with patient characteristics and outcome

Intense stromal HKII staining was present in at least 1 of 5 images in 30 of the 60

samples analyzed (shown by arrows in Figure 17E). This stratification was used to

dichotomize the cases into “stromal positive” and “stromal negative”. Stromal HKII

staining was not statistically associated with sex, age at diagnosis, histological grade,

TNM stage, nor primary tumor size (Table IV). Samples positive for stromal HKII

staining were associated with larger primary tumor sizes, although not significantly (p =

0.119). Kaplan-Meyer survival analysis showed that stromal HKII staining was

associated with median survival duration of CRC patients (Figure 19). Specifically,

95

patients whose tumors presented with stromal HKII staining had significantly shorter

progression-free survival (Figure 19A; 879 vs. 2132 days; p=0.049) and overall survival

(Figure 19B; 1011 vs. Undefined days; p=0.024) than patients whose tumors did not

present with stromal HKII staining.

Discussion

The results of this study revealed that low HKII scoring was associated with

shorter progression-free and overall survival in human CRC patients. During

hepatocellular carcinogenesis, the normal expression of HKIV in the liver is silenced

whereas HKII gene expression is induced, thereby promoting the glycolytic phenotype

(Mayer et al, 1997). Results from Chapter III suggested HKII is upregulated in more

aggressive CRC cell lines, but the impact of ischemia on cell signaling in the context of

solid tumors is poorly understood. In this study, I sought to quantify HKII expression in

FFPE samples of human colorectal cancer using immunofluorescence. This approach was

utilized to allow for co-localization between staining of HKII and of carbonic anhydrase

IX, a marker of tissue ischemia (Potter & Harris, 2004). My findings were subsequently

correlated to clinicopathological characteristics of the patients, and to patient outcome.

I initially wanted to verify the correlation between low oxygen conditions and the

induction of CAIX expression by exposing human CRC cells to anoxia in vitro. I found

that in HCT116 and DLD-1 cells, both HKII and CAIX expression was upregulated

under anoxic conditions (Figure 13). Upregulation of CAIX expression under low

oxygen conditions has been reported in breast and prostate cancer (Li et al, 2009; Wykoff

et al, 2001). Significant correlation between HKII and HIF1α, a marker of hypoxia, has

96

also been reported in hepatocellular and gastric carcinoma tissues (Qiu et al, 2011;

Yasuda et al, 2004). However, those studies did not examine HKII expression within

different tumor microenvironments via co-localization.

The abnormal vasculature that develops as a result of tumor progression leads to

ischemic areas that consist of both hypoxic and hypoglycemic conditions (Raghunand et

al, 2003). While the effects of hypoxia on tumor progression have been well documented,

the effects of hypoglycemia have been less studied. My results reveal that HKII staining

was present primarily in regions negative for CAIX staining, indicating that HKII is

predominantly expressed in non-ischemic regions of these CRC tumors. This suggests

that while hypoxia and hypoglycemia may have opposite effects on HKII expression, the

availability of glucose to cancer cells may ultimately determine the expression of this

enzyme. Previous work in our laboratory using preclinical models has established the

importance of hypoglycemia in CRC progression, including its ability to influence

mutation rate, DNA repair, and DNA methylation (Plumb et al, 2009; Shahrzad et al,

2005; Skowronski et al, 2014; Skowronski et al, 2010).

Since I found that HKII and CAIX expression were not co-localized, the

assessment of their expression patterns was performed separately. A number of studies

have shown that hypoxia-related overexpression of CAIX in the tumor is associated with

poor survival rates, including esophageal cancer (Birner et al, 2011), endometrioid

ovarian cancer (Choschzick et al, 2011), and oral cavity squamous cell carcinoma

(Brockton et al, 2012). CAIX is generally upregulated in the tumor compared to normal

intestinal epithelium (Kivela et al, 2001; Saarnio et al, 1998a; Saarnio et al, 1998b). In

the present study, CAIX expression was found to only be significantly associated with

97

age and none of the other clinicopathological characteristics. Increased CAIX expression

was associated with shorter progression free and shorter overall survival in our patient

set. However, these results were not statistically significant, perhaps reflective of the

small sample size.

Increased HKII expression has been observed via immunohistochemistry when

comparing hepatocellular carcinoma to non-dysplatic cirrhosis (Guzman et al, 2015).

HKII is a potential prognostic marker in breast carcinoma (Sato-Tadano et al, 2013),

laryngeal squamous cell carcinoma (Chen et al, 2014), glioblastoma multiforme (Wolf et

al, 2011), and gastric adenocarcinoma (Qiu et al, 2011). In the present study, I

hypothesized that high HKII expression would correlate with worse patient outcome (as

has been previously reported for CRC; Hamabe et al, 2014), due to an increased capacity

to increase glucose consumption and produce the necessary glycolytic intermediates to

sustain rapid proliferation. HKII expression was not associated with any of the

clinicopatholigical characteristics assessed. Of most interest, lower HKII expression was

found to be significantly associated with shorter progression-free status and worse overall

survival. While this is the first such observation in colorectal cancer, Lyshchik et al had

also observed this correlation in pancreatic cancer, where patients with high HKII

expression survived longer than those with low HKII expression (Lyshchik et al, 2007).

I was interested in optimizing a protocol that would allow for automated analysis

of immunofluorescence staining of any target through computational algorithms that can

be used as a rapid prognostic tool in the future. I therefore developed a variation of the H-

Score used for HKII staining, denoted as the F-Score, which took into account the 2

distinct staining intensities that were observed. This approach took into consideration

98

different expression levels of HKII throughout the tumor sections analyzed. As

mentioned, there are conflicting reports on the impact of HKII expression on cancer

progression. This, combined with the observation of intense HKII staining in the stroma

of 50% of the cases in my study, suggested that HKII may prove a more useful

prognostic marker using stromal vs. neoplastic cell staining as a focal point of analysis.

It has become increasingly apparent that the tumor microenvironment plays a

critical role in tumorigenesis. Stratification of colon cancer patients based on tumor

stroma revealed a correlation between patients with high stromal content and shorter

progression free and overall survival (Mesker et al, 2007). Stromal fibroblasts that

undergo myofibroblast differentiation lose caveolin 1 (Cav-1) expression which induces a

cancer-associated fibroblast phenotype, characterized by an increase in glycolysis

(Pavlides et al, 2009). This observation has been termed the ‘reverse Warburg effect’ in

which there is metabolic coupling between supporting glycolytic stromal cells and

oxidative tumor cells. Loss of stromal Cav-1 expression was associated with both shorter

disease-free and overall survival in CRC (Zhao et al, 2015). Cav-1 null stromal cells also

upregulate key glycolytic genes, including HKII (Pavlides et al, 2009).

My study showed intense HKII stromal staining in 30 of 60 cases. Based on the

trend between positive stromal HKII staining and increased tumor sizes, my results are

consistent with the possible role of the tumor stroma in feeding neoplastic cells and

supporting tumor growth. Further analysis revealed that patients with tumors presenting

stromal HKII staining had significantly shorter progression-free and overall survival than

patients with tumors lacking stromal HKII staining. This association of survival and

tumor aggressiveness (as revealed by outcome) with stromal HKII is novel for CRC and

99

indicates the possibility of the reverse Warburg effect occurring in some cases of

colorectal cancer. Stromal HKII may be a more informative prognostic marker than

overall HKII expression within the tumor. Further study with a larger number of patients

is needed to confirm these findings. Additional studies are needed to identify which

stromal cell type(s) are expressing HKII, and whether that expression is associated with

changes in Cav-1 as shown previously (Pavlides et al, 2009).

My study had several limitations. The retrospective nature of this study is a major

limitation, due to possible confounding factors and bias associated with case selection.

The samples for this study were gathered from different tumor bank sites around Ontario.

Thus, the fixation protocols may have been slightly altered leading to possible variability

in the staining outcome and the sample size is relatively small. Further studies using

tumor microarrays can be used to control staining protocols and increase sample size.

In conclusion, I have demonstrated that HKII is expressed in non-ischemic

regions of tumors and that lower expression by neoplastic cells is associated with worse

outcome in patients with colorectal cancer. HKII was also highly expressed in stromal

cells of some cases, where it was associated with worse outcome, suggesting an

important role for HKII in glycolytic stromal cells. Until recently, the Warburg effect

phenomenon was thought to occur in most cancer cells (Gatenby & Gillies, 2004). There

is now a growing appreciation for the complexity of tumor metabolism and the interplay

between the tumor and adjacent stromal tissue (Pavlides et al, 2009). Identifying

glycolytic enzymes that are upregulated in the stromal compartment, such as HKII, may

help to improve the efficacy of targeting tumor metabolism as therapeutic strategy for

human CRC. These findings may be widely applicable, as many different cancer subtypes

100

exhibit an increased glycolytic phenotype and in many cases a large proportion of the

tumor mass is stromal in nature.

101

Table II. Patient characteristics.

Variable N %

Sex

Male 34 57

Female 26 43

Age at diagnosis (years)

40-49 6 10

50-59 12 20

60-69 13 22

70-79 17 28

80-89 9 15

90-94 3 5

Histological Grade

I 5 8

II 25 42

III 19 32

IV 9 15

Not available 2 3

TNM Stage

1 2 3

2 16 27

3 20 33

4 13 22

Not available 9 15

102

Table III. Associations of CAIX and HKII staining with patient characteristics and tumor pathology.

Variable High CAIX (N=30) High HKII F-Score (N=30)

N n % P value n % P value

Sex

Male 34 16 47.1 0.795 17 50.0 1.000

Female 26 14 53.8 13 50.0


< 70 31 11 35.5 0.038* 14 45.2 0.797

≥ 70 29 19 65.5 16 55.2

Histological Grade

I/II 30 13 43.3 0.600 15 50.0 0.799

III/IV 28 15 53.6 13 46.4

TNM Stage

1-3 38 19 50.0 0.336 24 63.2 0.057

4 13 9 69.2 4 30.8

Primary tumor size (cm)

≤ 5 27 14 51.9 1.000 13 48.1 1.000

> 5 33 16 48.5 17 51.5

103

Table IV. Associations of stromal HKII staining with patient characteristics and tumor pathology.

Variable Presence of Stromal HKII (N=30)

N n % P value

Sex

Male 34 16 47.1 0.795

Female 26 14 53.8


< 70 31 13 41.9 0.302

≥ 70 29 17 58.6

Histological Grade

I/II 30 15 50.0 0.799

III/IV 28 15 53.6

TNM Stage

1-3 38 17 44.7 1.000

4 13 6 46.2

Primary tumor size (cm)

≤ 5 27 10 37.0 0.119

> 5 33 20 60.6

104

Figure 13. HKII and CAIX immunoblotting in colorectal cancer cell lines. HKII and CAIX expression was detected by western blot analysis. Two human colorectal cancer cells were exposed to either normoxia (C; 20% O2) or anoxia (A; < 0.1% O2) for 24 hours. In response to anoxia, both cell lines showed upregulation in HKII and CAIX expression.

105

Figure 14. HKII and CAIX immunostaining in colorectal cancer. Each row in descending order represents immunostaining of HKII (green), CAIX (red), DAPI nuclear counterstain (blue), and merged images. CAIX was used as a marker of ischemia. The left panel shows images from a non-ischemic region of a tumor sample, noted by the lack of CAIX staining. The right panel shows images from an ischemic region of a different tumor sample, noted by the presence of CAIX staining. Images were taken at 40X objective.

106

Figure 15. Evaluation of CAIX immunostaining. Images from CAIX staining, as shown on the top panel, were subjected to computational analysis. Images were taken at 40X objective. Bottom panel shows analysis of CAIX staining when the detection threshold was set at 20. Percent area stained (outlined in yellow) is calculated by the amount of staining set by the threshold value and numerically shown on the bottom right corner. Examples of low and high CAIX staining from two different tumors are shown on the left and right panels, respectively.

107

Figure 16. Evaluation of HKII immunostaining. Images from HKII staining from two different CRC specimens, as shown on the top panel, were subjected to computational analysis. Images were taken at 40X objective. Middle and bottom panels show analysis of HKII staining when the detection threshold was set at 25 and 85, respectively. Percent area stained (outlined in yellow) is calculated by the amount of staining set by the threshold values, and numerically displayed on the bottom right corner.

108

Figure 17. Immunofluorescent staining of HKII and corresponding F-Scores. Images were taken at 40X objective. The F-Score is calculated by summation of the two values generated by the percent area stained at the two thresholds. Images in the top panel with a F-Score of < 25 were considered to have a low F-Score. Images in the bottom panel with a F-Score of > 25 were considered to have a high F-Score. Arrows in E indicate stromal cell staining.

109

Figure 18. Progression-free survival and overall survival by CAIX and HKII staining. Kaplan-Meier curves of PFS (A, C) and OS (B, D) in patients with CRC based on CAIX (A, B) and HKII (C, D) staining. Ticks indicate censored subjects.

110

Figure 19. Progression-free survival and overall survival by the presence of stromal HKII staining. Kaplan-Meier curves of PFS (A) and OS (B) in patients with CRC based on the presence of stromal HKII staining. Ticks indicate censored subjects.

111

GENERAL DISCUSSION

Metabolic reprogramming is an important aspect of tumorigenesis that remains to

be fully understood. Initial observations on lactate buildup in tumors under normal

oxygen tension have been further characterized through alterations in metabolic signaling

pathways in cancer cells, a phenomenon known as the Warburg effect (Warburg et al,

1927). These alterations are a result of mutations in key glycolytic enzymes and are

aberrantly regulated by oncogenic signaling (Koppenol et al, 2011). The tumor

microenvironment is an important modulator of tumor behavior (Raghunand et al, 2003;

Subarsky & Hill, 2003). What remains poorly understood are the effects of the tumor

microenvironment on drug sensitivity, especially pertaining to treatments that affect

metabolic pathways. Previous data generated in our laboratory suggesting that DCA, a

potential anti-cancer agent, provided a protective effect to ischemic CRC cells. I therefore

sought to further understand this phenomenon and to determine the effects of compounds

targeting metabolic pathways in CRC.

Over the course of the 21st century, much research interest has focused on better

comprehending the metabolic pathways that permit rapid tumor growth. A major issue

with studies elucidating the mechanisms of cancer metabolism lies in the experimental

models used. It is well known and widely accepted that results collected through in vitro

experimentation are not necessarily reproducible in vivo, and that therapeutic success in

preclinical models and early phase studies rarely translates to advanced trials. One

possible reason for this disconnect is the influence of the natural microenvironment seen

in solid tumors. The effect of the tumor microenvironment on drug sensitivity is

dependent on both the therapeutic agent and the cells being studied (Strese et al, 2013).

112

This is one reason why targeting various aspects of cancer metabolism may show anti-

cancer potential in vitro but not show any benefit in vivo.

Our laboratory previously showed that DCA provided a cytoprotective effect to

human CRC cells under anoxic conditions. In contrast, Xu and colleagues found that

targeting lymphoma and CRC cells with another compound, 3BP, was more effective

under hypoxic conditions (Xu et al, 2005). I wanted to follow up our previous study and

determine the molecular mechanisms by which DCA-induced cytoprotection was

conferred. In Chapter II, I found differential regulation of PDH phosphorylation between

different human CRC cells led to differences in mitochondrial activity following DCA

exposure. However, these effects did not result in significant changes in cellular

metabolism nor growth. This suggests that DCA may only be beneficial in treating a

subset of tumor types based on their molecular profiles of different PDK isoform

expression.

As with hypoxia, glucose concentration varies in a gradient fashion resembling

what is observed with oxygen levels in the extravascular space of solid tumors

(Kallinowskil et al, 1985). Tumor hypoxia is a common occurrence, and since it is

primarily due to insufficient blood flow, reduced nutrient supply such as low glucose

availability is likely also associated with those conditions. As previously discussed,

aerobic cancer cells are able to utilize lactate as fuel and allow glucose to support

glycolytic cells at greater distances from supporting blood vessels (Sonveaux et al, 2008).

Therefore, a tumor with inadequate vascularization would experience both hypoxic and

hypoglycemic conditions, although glucose may be able to diffuse over greater distances

due to the ability of aerobic cells to preferentially metabolize lactate.

113

The bulk of research examining tumor ischemia has focused on hypoxia and the

effects of hypoglycemia on tumor progression are not well studied. This issue would be

particularly important when studying the effects of compounds that may alter cellular

metabolism, which is heavily reliant on glucose consumption. In particular, 3BP is a

pyruvate analog that inhibits HKII activity and effectively kills cancer cells (Ko et al,

2001). Therefore, the effects of this drug on cytotoxicity may be altered by glucose

availability. I found that human CRC cells were far more sensitive to the cytotoxic effects

of 3BP when cultured under high glucose conditions versus normal glucose conditions.

The IC50 value against 3BP in SKOV3 cells increases by over 3 fold when treated in

galactose medium vs. glucose medium, suggesting the effects of simply treatment

medium alone affects drug response in vitro (Wintzell et al, 2012).

Glucose-dependent alteration of 3BP-sensitivity led me to suspect that my

observations could be explained by the increase in HKII expression I saw under high

glucose conditions. However, HKII knockdown using RNA interference resulted in no

differences in 3BP sensitivity. Thus, I believe that the differences observed in 3BP

sensitivity in CRC cells under altered glucose availability conditions cannot be explained

by HKII expression alone. Since 3BP acts as an alkylating agent and affects other

glycolytic enzymes, it is possible that the main effects of 3BP in CRC are through other

metabolic pathways. Although HKII expression did not ultimately correlate with 3BP

sensitivity, it is still worthwhile to continue examining the therapeutic potential of 3BP

against colorectal cancer. The potent effects of 3BP on colorectal cancer have been

shown in a xenograft mouse model, where 3BP treatment over 6 days led to over 50%

tumor regression (Sanchez-Arago & Cuezva, 2011).

114

Under glucose limiting conditions, cancer cells including glioma (Beckner et al,

2005; Bouzier et al, 1998), cervical, breast carcinoma, hepatoma, and pancreatic (Plecita-

Hlavata et al, 2008; Rossignol et al, 2004; Smolkova et al, 2010) switch from aerobic

glycolysis to oxidative phosphorylation. Furthermore, removal of glucose, or inhibition

of glycolysis by iodoacetate in HeLa cells led to a rapid switch toward glutamine

utilization (Brand et al, 1994). These findings suggest the flexibility of cancer cells to

utilize different metabolic pathways and survive in various microenvironments.

The majority of metabolic enzymes implicated in tumorigenesis are not mutated

and furthermore are expressed in both transformed and normal cells (Koppenol et al,

2011). This leads to a considerable set of challenges regarding target specificity when

using metabolic modulating approaches. However, regardless of enzyme expression, the

differences observed between transformed and normal metabolism provides a window for

therapeutic intervention. Furthermore, the additional issue of the tumor

microenvironment and its impact on the outcome from therapeutic intervention further

adds to the complexity of cancer metabolism (Pavlides et al, 2009). Future work studying

the effects of DCA and 3BP in CRC cells would include evaluating the combination of

both hypoxic and hypoglycemic conditions to better recapitulate what may occur in the

tumor microenvironment. Several other environmental factors present in solid tumors

may be studied beyond oxygen and nutrient supply, including interaction with stromal

cells (such as vascular endothelial cells, infiltrating immune cells, and tumor fibroblasts)

and acidity (in part secondary to hypoxia, and metabolism). These factors can be studied

using models that better resemble the in vivo conditions by the use of spheroid or

115

organoid cultures (Gao et al, 2014; Santini & Rainaldi, 1999). Treating cells for

prolonged incubation times beyond confluency could mimic possible contact inhibition

effects and study confluence dependent resistance in vitro (Pelletier et al, 1990).

While the ability of cancer cells to increase their glucose consumption has been

confirmed through FDG-PET scanning in vivo, the shuttling of glucose for further

processing through either glycolysis or oxidative phosphorylation cannot be evaluated

using this technique. Several glioma cell lines are highly dependent on oxidative

phosphorylation for ATP production (Griguer et al, 2005). 50-80% of ATP production in

various cancer cell lines was generated through oxidative phosphorylation (Guppy et al,

2002; Kallinowski et al, 1989; Kallinowski et al, 1988; Zu & Guppy, 2004). With

increasing reports outlining the importance of oxidative phosphorylation in cancer

metabolism, the Warburg effect is currently being challenged by other theories.

Growing evidence suggests the tumor microenvironment plays a pivotal role in

tumor progression and metastasis. A hypothesis gaining traction is termed the ‘reverse

Warburg effect’ formulated by Dr. Michael Lisanti; it suggests that aerobic glycolysis

occurs in the fibroblastic tumor stromal compartment rather than in tumor cells (Pavlides

et al, 2009). These cancer cells induce the Warburg effect in neighboring stromal

fibroblasts, resulting in differentiation of cancer-associated fibroblasts (CAFs) to

myofibroblasts. These tumor-associated myofibroblasts express muscle-specific actin,

and show an increased ability to secrete and remodel the extracellular matrix (Ronnov-

Jessen et al, 1996; Shekhar et al, 2003). The glycolytic end-products, pyruvate and

lactate, are then secreted by CAFs and taken up by adjacent cancer cells to sustain their

proliferative capacity and continued use of oxidative metabolism for energy production.

116

In human breast cancer tissues lacking Cav-1 expression, 2 key glycolytic enzymes,

PKM2 and LDH, are prominently expressed in CAFs and not within adjacent cancer

cells, consistent with the reverse Warburg effect hypothesis (Pavlides et al, 2009).

The reverse Warburg effect has also been independently confirmed by Pol and

colleagues, where it was also observed that Cav-1 deficient murine fibroblasts had

dysfunctional mitochondria and that restoring its expression recovered mitochondrial

function (Bosch et al, 2011). CAFs have been shown to express MCT4 for lactate

secretion while adjacent breast cancer cells express MCT1 for lactate uptake (Whitaker-

Menezes et al, 2011).

Under monolayer culturing conditions, MCF7 breast cancer cells exhibit very low

mitochondrial mass, a conventional aerobic glycolysis phenotype. However, when these

cells were co-cultured with fibroblasts, it led to a significant increase in the mitochondrial

mass of the MCF7 cells. Lactate administration to MCF7 cells also led to an increase in

mitochondrial mass, suggesting lactate produced by the fibroblasts may promote

mitochondrial biogenesis and increase mitochondrial respiration (Martinez-Outschoorn et

al, 2010). In a breast cancer mouse model, oncogenically driven human organoids

develop into tumors only if co-injected with immortalized fibroblasts and not with normal

primary fibroblasts, indicative of stromal activation in breast cancer formation (Wu et al,

2009). These results suggest that cross talk between transformed tumor cells and the

adjacent stroma exists.

Metabolic cooperation between adjacent cell compartments is not unique to

tumors. During ovarian folliculogenesis, the oocyte controls cumulus granulosa cell

metabolism to compensate for its lack of critical metabolic functions, including

117

cholesterol synthesis and amino acid uptake (Su et al, 2009; Sugiura et al, 2007).

Glycolytic astrocytes generate lactate to support oxidative phosphorylation in adjacent

neurons (Magistretti & Pellerin, 1996; Tsacopoulos & Magistretti, 1996). Likewise,

glycolytic fast-twitch muscle fibers produce lactate that is secreted for slow-twitch fibers

to utilize as oxidative phosphorylation substrates (Connett et al, 1984; McCullagh et al,

1997).

In Chapter IV, I observed the presence of strong HKII staining in the stromal

compartment and weaker staining in the adjacent tumor epithelial cells. This would

suggest the potential presence of the reverse Warburg effect in some CRC tumors. Other

studies found that colon cancer patients with high tumor stromal content show

dramatically increased tumor progression and poor survival (Benita et al, 2009; Mesker

et al, 2007). The observed HKII staining pattern provides further evidence to suggest that

cell culture experiments in vitro might not reflect what occurs in an in vivo experimental

system, in which the tumor microenvironment would affect metabolic strategies and

cellular responses. Another interesting finding from Chapter IV was the negative

correlation between HKII expression and survival measures, a finding that has also been

reported in pancreatic cancer (Lyshchik et al, 2007). This could suggest that aggressive

tumors have adapted an ability to better utilize other carbon fuels for its metabolic

processes, thereby reducing expression of glucose-related processing molecules such as

HKII. Glucose and glutamine account for most carbon and nitrogen metabolism in

mammalian cells (Wellen et al, 2010). Using ovarian cancer cell lines, Yang and

colleagues discovered a correlation between glutamine dependence and invasiveness;

118

high-invasive ovarian cancer cells were markedly glutamine dependent versus low-

invasive ovarian cancer cells (Yang et al, 2014).

A chronic disease characterized by hyperglycemia is diabetes mellitus. Globally,

approximately 387 million people aged between 20 to 79 years have diabetes. Type 2

diabetes, characterized by insensitivity to insulin, accounts for 85-95% of prevalent cases

(International Diabetes Foundation, 2014). Metformin is the first-line therapy for type 2

diabetes (Alexander et al, 2008). This metabolic modulator decreases plasma glucose

concentrations by decreasing hepatic gluconeogenesis and increasing cellular glucose

uptake (Iozzo et al, 2003; Wollen & Bailey, 1988). This is accomplished through

inhibition of complex I of the electron transport chain, thereby increasing AMP:ATP

ratio and activating AMPK signaling (El-Mir et al, 2000).

In 2005, Evans and colleague found a 23% reduction in total cancer incidence in

diabetic patients treated with metformin (Evans et al, 2005). Several meta-analyses also

show a correlation between metformin use and lowered cancer incidence (Gandini et al,

2014). Metformin inhibits proliferation and induces apoptosis in cancer cells, presumably

due to the re-wiring of cancer metabolism to better utilize mitochondrial respiration and

increase ATP production via AMPK activation (Luo et al, 2012; Owen et al, 2000; Zhou

et al, 2001). Similarly, numerous epidemiological studies suggest an association between

type 2 diabetes and cancer, including increased risk and mortality rates (Joung et al,

2015).

However, increased focus on metformin use and cancer risk has led to conflicting

data arising from different epidemiological, human, and animal studies. The variation in

estimated risks may differ due to the characteristics of the study population, including

119

genetic susceptibility, life-style behavior, and environmental exposure (Habib & Rojna,

2013). In a systematic review and meta-analysis on studies controlling for body mass

index, prospective studies, and studies without time-related biases, Gandini and

colleagues found that the association between metformin use and decreased overall

cancer incidence was considerably smaller than observed without such adjustments (10-

18% vs. 31%) (Gandini et al, 2014). High glucose media concentrations reduced the

effectiveness of metformin on breast cancer cell proliferation (Wahdan-Alaswad et al,

2013). The relationship between diabetes and cancer remain unclear; the hyperglycemic

conditions associated with diabetes may facilitate neoplastic proliferation, which is

reduced upon metformin exposure. This cancer-diabetics association may also be indirect

and due to common risk factors such as obesity (Gallagher & LeRoith, 2015). Clinical

trials are needed to determine whether metformin can in fact be applicable to non-

diabetic populations. Results from pre-clinical and retrospective studies support testing

the use of metformin as an adjuvant in chemotherapy (Zhang et al, 2013).

An alternative to directly targeting glycolytic enzymes with metabolic modulators

is to exploit the regulation of these enzymes. A field of molecular biology with rapidly

growing interest is the role of small interfering RNAs. In particular, microRNAs are non-

coding 18-24 nucleotide-long RNAs that match and specifically bind onto 3’ untranslated

regions of target mRNA, resulting in altered gene expression of a suite of target genes

(Flynt & Lai, 2008; Vasudevan et al, 2007). Over the past few years, a large number of

miRNAs have been implicated in tumor metabolic regulation and the Warburg effect

(Bienertova-Vasku et al, 2013; Chen et al, 2012; Hatziapostolou et al, 2013). miRNAs

have been shown to regulate several glycolytic enzymes, including GLUTs, GAPDH,

120

LDHA and PKM2 (Chan et al, 2015). HKII expression in particular is regulated by

numerous miRNAs. miR-143 expression is downregulated in human glioblastoma, and its

overexpression leads to glycolytic inhibition via HKII downregulation (Zhao et al, 2013).

Downregulation of HKII via miR-143 and inhibition of glucose metabolism was also

shown in head and neck squamous cell carcinoma (Peschiaroli et al, 2013), breast cancer

(Jiang et al, 2012), lung cancer (Fang et al, 2012), and colon cancer (Gregersen et al,

2012). Jiang and colleagues showed that miR-55 promoted HKII transcription through

activation of signal transducer and activator of transcription 3 (STAT3) in hepatocellular

carcinoma cells (Jiang et al, 2014). miR-125b was also found to regulate HKII

expression, and this finding was also observed in lung cancer (Fang et al, 2012) and in

chronic lymphocytic leukemia (Tili et al, 2012). Kim and colleagues demonstrated that

miR-21 and -34 regulates HKII expression in bladder cancer and lung cancer,

respectively, while both miRNAs were capable of repressing HKII expression in colon

cancer (Kim et al, 2013). Through bioinformatics analysis, it was revealed that miR103/7

might have the ability to target PDK4 (Wilfred et al, 2007).

As some miRNAs have been implicated in tumorigenesis and thus preferentially

expressed in cancer and not normal cells, therapeutic strategies can be developed to target

these small RNA molecules. Studies show that miRNA deregulation leads to reduction in

cancer cell growth with no effects on normal cells, suggesting its therapeutic potential

(Drakaki et al, 2013; Hatziapostolou et al, 2013). Thus, either downregulating miRNAs

important to cancer malignancy or upregulating those that would otherwise limit cancer

progression may be a potential anti-cancer strategy. A limitation to this approach is our

current lack of depth of knowledge in this field. Since miRNAs have numerous targets

121

and thus functions in cell biology, it is important to first comprehend the ability of the

various miRNAs that play a role in cancer progression before contemplating its

therapeutic potential. Furthermore, the effectiveness of in vivo delivery of miRNAs/anti-

miRs is still a major hurdle in clinical studies (Park et al, 2011).

122

SUMMARY AND CONCLUSIONS

The research described in this thesis highlighted the effects of dichloroacetate and

3-bromopyruvate on human colorectal cancer cells under varying culture conditions in

attempts to mimic aspects of the tumor microenvironment. The goals were to determine

how different in vitro conditions might alter metabolic signaling and ultimately affect

drug sensitivity in human CRC cells. In addition, I assessed the expression of a key

glycolytic enzyme, hexokinase II, in colorectal cancer tumor samples to better understand

its regulation within the microenvironment in vivo.

I initially studied the molecular mechanisms surrounding DCA-induced

cytoprotection in CRC cells. PDKs 1 and 3 were determined to be the predominant

isoforms expressed by different CRC cell lines. This in part could account for the

resistance to DCA-induced apoptosis observed, as PDK2 is reported to be most sensitive

to the effects of DCA. I found changes in expression of PDK isoforms under anoxia that

led to slight differences in PDH phosphorylation at residue Ser293 and mitochondrial

activity. However, these molecular changes did not lead to significant changes in cell

metabolism or proliferation. I propose that DCA may only be beneficial in treating a

subset of tumor types based on their molecular profiles of different PDK isoforms.

The next compound I focused on was 3-bromopyruvate and its proposed target,

hexokinase II. I found that the variation in HKII expression between different CRC cell

lines correlated to their sensitivity towards 3BP. Exposure to 3BP at lower doses led to

induction of apoptosis while necrotic cell death was observed at higher doses. Studies

involving the manipulation of glucose availability in culture conditions showed that cells

were less sensitive to the effects of 3BP when cultured under normal glucose vs. high

123

glucose concentrations. However, HKII knockdown experiments revealed no differences

in 3B sensitivity, even with robust HKII knockdown. This suggests that 3BP sensitivity

could be correlated to the bioenergetic signature of cancer cells and that HKII expression

is a surrogate marker for this signature. Such a situation explains the correlation found

between HKII expression and 3BP sensitivity while manipulations to HKII expression

led to no changes in 3BP sensitivity.

Although other enzymes are targeted by 3BP in CRC cells, HKII still plays an

important role in aerobic glycolysis. Thus, I examined its expression in FFPE samples of

CRC tumors with known clinical outcomes to determine whether HKII expression is a

prognostic marker for patient outcome. HKII expression was detectable only in non-

ischemic regions of tumors. Surprisingly, I found that low HKII expression correlated

with both shorter disease free progression and with worse overall survival in patients.

Stromal cells sometimes exhibited high HKII expression, agreeing with a novel theory on

cancer metabolism termed the ‘reverse Warburg effect’, in which the glycolytic

phenotype is exhibited by the stromal cells within the tumor and not the malignant cells

themselves.

In conclusion, these studies have provided a better understanding of how the

tumor microenvironment plays a role in altering the sensitivity of cancer cells to various

metabolic modulators. Both oxygen and glucose deprivation can alter the effects of

compounds such as DCA and 3BP. An issue with in vitro experimentation involves the

lack of external cues otherwise provided by the microenvironment in in vivo settings. A

better understanding of how the tumor microenvironment may affect the susceptibility of

cancer cells to metabolic re-wiring will allow researchers to develop and identify more

124

effective anti-cancer compounds and will provide better insight into the possible

translational effects in vivo.

125

REFERENCES

Aebersold DM, Burri P, Beer KT, Laissue J, Djonov V, Greiner RH, Semenza GL (2001) Expression of hypoxia-inducible factor-1alpha: a novel predictive and prognostic parameter in the radiotherapy of oropharyngeal cancer. Cancer Res 61(7): 2911-6 Alessi DR, Andjelkovic M, Caudwell B, Cron P, Morrice N, Cohen P, Hemmings BA (1996) Mechanism of activation of protein kinase B by insulin and IGF-1. The EMBO journal 15(23): 6541-51 Alessi DR, James SR, Downes CP, Holmes AB, Gaffney PR, Reese CB, Cohen P (1997) Characterization of a 3-phosphoinositide-dependent protein kinase which phosphorylates and activates protein kinase Balpha. Curr Biol 7(4): 261-9 Alexander GC, Sehgal NL, Moloney RM, Stafford RS (2008) National trends in treatment of type 2 diabetes mellitus, 1994-2007. Archives of internal medicine 168(19): 2088-94 Ardehali H, Printz RL, Whitesell RR, May JM, Granner DK (1999) Functional interaction between the N- and C-terminal halves of human hexokinase II. J Biol Chem 274(23): 15986-9 Artiko V, Odalovic S, Sobic-Saranovic D, Petrovic M, Stojiljkovic M, Petrovic N, Kozarevic N, Grozdic-Milojevic I, Obradovic V (2015) Can (18)F-FDG PET/CT scan change treatment planning and be prognostic in recurrent colorectal carcinoma? A prospective and follow-up study. Hellenic journal of nuclear medicine 18(1): 35-41 Astuti D, Latif F, Dallol A, Dahia PL, Douglas F, George E, Skoldberg F, Husebye ES, Eng C, Maher ER (2001) Gene mutations in the succinate dehydrogenase subunit SDHB cause susceptibility to familial pheochromocytoma and to familial paraganglioma. American journal of human genetics 69(1): 49-54 Babu E, Ramachandran S, CoothanKandaswamy V, Elangovan S, Prasad PD, Ganapathy V, Thangaraju M (2011) Role of SLC5A8, a plasma membrane transporter and a tumor suppressor, in the antitumor activity of dichloroacetate. Oncogene 30(38): 4026-37 Baker JC, Yan X, Peng T, Kasten S, Roche TE (2000) Marked differences between two isoforms of human pyruvate dehydrogenase kinase. J Biol Chem 275(21): 15773-81 Baker JP, Rabin BR (1969) Effects of bromopyruvate on the control and catalytic properties of glutamate dehydrogenase. Eur J Biochem 11(1): 154-9 Bauer DE, Hatzivassiliou G, Zhao F, Andreadis C, Thompson CB (2005) ATP citrate lyase is an important component of cell growth and transformation. Oncogene 24(41): 6314-22

126

Baysal BE, Ferrell RE, Willett-Brozick JE, Lawrence EC, Myssiorek D, Bosch A, van der Mey A, Taschner PE, Rubinstein WS, Myers EN, Richard CW, 3rd, Cornelisse CJ, Devilee P, Devlin B (2000) Mutations in SDHD, a mitochondrial complex II gene, in hereditary paraganglioma. Science 287(5454): 848-51 Beckner ME, Gobbel GT, Abounader R, Burovic F, Agostino NR, Laterra J, Pollack IF (2005) Glycolytic glioma cells with active glycogen synthase are sensitive to PTEN and inhibitors of PI3K and gluconeogenesis. Laboratory investigation; a journal of technical methods and pathology 85(12): 1457-70 Benita Y, Kikuchi H, Smith AD, Zhang MQ, Chung DC, Xavier RJ (2009) An integrative genomics approach identifies Hypoxia Inducible Factor-1 (HIF-1)-target genes that form the core response to hypoxia. Nucleic Acids Res 37(14): 4587-602 Bensaad K, Tsuruta A, Selak MA, Vidal MN, Nakano K, Bartrons R, Gottlieb E, Vousden KH (2006) TIGAR, a p53-inducible regulator of glycolysis and apoptosis. Cell 126(1): 107-20 Berra E, Richard DE, Gothie E, Pouyssegur J (2001) HIF-1-dependent transcriptional activity is required for oxygen-mediated HIF-1alpha degradation. FEBS Lett 491(1-2): 85-90 Bersin RM, Stacpoole PW (1997) Dichloroacetate as metabolic therapy for myocardial ischemia and failure. American heart journal 134(5 Pt 1): 841-55 Berwick DC, Hers I, Heesom KJ, Moule SK, Tavare JM (2002) The identification of ATP-citrate lyase as a protein kinase B (Akt) substrate in primary adipocytes. J Biol Chem 277(37): 33895-900 Bienertova-Vasku J, Sana J, Slaby O (2013) The role of microRNAs in mitochondria in cancer. Cancer Lett 336(1): 1-7 Birner P, Jesch B, Friedrich J, Riegler M, Zacherl J, Hejna M, Wrba F, Schultheis A, Schoppmann SF (2011) Carbonic anhydrase IX overexpression is associated with diminished prognosis in esophageal cancer and correlates with Her-2 expression. Annals of surgical oncology 18(12): 3330-7 Birsoy K, Wang T, Possemato R, Yilmaz OH, Koch CE, Chen WW, Hutchins AW, Gultekin Y, Peterson TR, Carette JE, Brummelkamp TR, Clish CB, Sabatini DM (2013) MCT1-mediated transport of a toxic molecule is an effective strategy for targeting glycolytic tumors. Nature genetics 45(1): 104-8 Blagosklonny MV, An WG, Romanova LY, Trepel J, Fojo T, Neckers L (1998) p53 inhibits hypoxia-inducible factor-stimulated transcription. J Biol Chem 273(20): 11995-8

127

Blouin JM, Penot G, Collinet M, Nacfer M, Forest C, Laurent-Puig P, Coumoul X, Barouki R, Benelli C, Bortoli S (2011) Butyrate elicits a metabolic switch in human colon cancer cells by targeting the pyruvate dehydrogenase complex. Int J Cancer 128(11): 2591-601 Bonnet S, Archer SL, Allalunis-Turner J, Haromy A, Beaulieu C, Thompson R, Lee CT, Lopaschuk GD, Puttagunta L, Bonnet S, Harry G, Hashimoto K, Porter CJ, Andrade MA, Thebaud B, Michelakis ED (2007) A Mitochondria-K+ Channel Axis Is Suppressed in Cancer and Its Normalization Promotes Apoptosis and Inhibits Cancer Growth. Cancer Cell 11(1): 37-51 Bos R, van der Groep P, Greijer AE, Shvarts A, Meijer S, Pinedo HM, Semenza GL, van Diest PJ, van der Wall E (2003) Levels of hypoxia-inducible factor-1alpha independently predict prognosis in patients with lymph node negative breast carcinoma. Cancer 97(6): 1573-81 Bosch M, Mari M, Herms A, Fernandez A, Fajardo A, Kassan A, Giralt A, Colell A, Balgoma D, Barbero E, Gonzalez-Moreno E, Matias N, Tebar F, Balsinde J, Camps M, Enrich C, Gross SP, Garcia-Ruiz C, Perez-Navarro E, Fernandez-Checa JC, Pol A (2011) Caveolin-1 deficiency causes cholesterol-dependent mitochondrial dysfunction and apoptotic susceptibility. Curr Biol 21(8): 681-6 Bosetti C, Levi F, Rosato V, Bertuccio P, Lucchini F, Negri E, La Vecchia C (2011) Recent trends in colorectal cancer mortality in Europe. Int J Cancer 129(1): 180-91 Bouzier AK, Voisin P, Goodwin R, Canioni P, Merle M (1998) Glucose and lactate metabolism in C6 glioma cells: evidence for the preferential utilization of lactate for cell oxidative metabolism. Developmental neuroscience 20(4-5): 331-8 Bowker-Kinley MM, Davis WI, Wu P, Harris RA, Popov KM (1998) Evidence for existence of tissue-specific regulation of the mammalian pyruvate dehydrogenase complex. Biochem J 329 ( Pt 1): 191-6 Brand RM, Lyons RH, Midgley AR (1994) Understanding the dynamics of cellular responsiveness to modifications of metabolic substrates in perifusion. J Cell Physiol 160(1): 10-6 Brizel DM, Sibley GS, Prosnitz LR, Scher RL, Dewhirst MW (1997) Tumor hypoxia adversely affects the prognosis of carcinoma of the head and neck. Int J Radiat Oncol Biol Phys 38(2): 285-9 Brockton NT, Klimowicz AC, Bose P, Petrillo SK, Konno M, Rudmik L, Dean M, Nakoneshny SC, Matthews TW, Chandarana S, Lau HY, Magliocco AM, Dort JC (2012) High stromal carbonic anhydrase IX expression is associated with nodal metastasis and decreased survival in patients with surgically-treated oral cavity squamous cell carcinoma. Oral oncology 48(7): 615-22

128

Brown JM (1979) Evidence for acutely hypoxic cells in mouse tumours, and a possible mechanism of reoxygenation. The British journal of radiology 52(620): 650-6 Brown JM (1999) The hypoxic cell: a target for selective cancer therapy--eighteenth Bruce F. Cain Memorial Award lecture. Cancer Res 59(23): 5863-70 Bustin SA, Benes V, Garson JA, Hellemans J, Huggett J, Kubista M, Mueller R, Nolan T, Pfaffl MW, Shipley GL, Vandesompele J, Wittwer CT (2009) The MIQE guidelines: minimum information for publication of quantitative real-time PCR experiments. Clin Chem 55(4): 611-22 Calleja V, Alcor D, Laguerre M, Park J, Vojnovic B, Hemmings BA, Downward J, Parker PJ, Larijani B (2007) Intramolecular and intermolecular interactions of protein kinase B define its activation in vivo. PLoS Biol 5(4): e95 Calvino E, Estan MC, Sanchez-Martin C, Brea R, de Blas E, Boyano-Adanez Mdel C, Rial E, Aller P (2014) Regulation of death induction and chemosensitizing action of 3-bromopyruvate in myeloid leukemia cells: energy depletion, oxidative stress, and protein kinase activity modulation. J Pharmacol Exp Ther 348(2): 324-35 Cambier N, Chopra R, Strasser A, Metcalf D, Elefanty AG (1998) BCR-ABL activates pathways mediating cytokine independence and protection against apoptosis in murine hematopoietic cells in a dose-dependent manner. Oncogene 16(3): 335-48 Canadian Cancer Society Canadian Cancer Statistics 2014. Cao W, Yacoub S, Shiverick KT, Namiki K, Sakai Y, Porvasnik S, Urbanek C, Rosser CJ (2008) Dichloroacetate (DCA) sensitizes both wild-type and over expressing Bcl-2 prostate cancer cells in vitro to radiation. Prostate 68(11): 1223-31 Carmeliet P, Jain RK (2000) Angiogenesis in cancer and other diseases. Nature 407(6801): 249-57 Casamayor A, Morrice NA, Alessi DR (1999) Phosphorylation of Ser-241 is essential for the activity of 3-phosphoinositide-dependent protein kinase-1: identification of five sites of phosphorylation in vivo. Biochem J 342 ( Pt 2): 287-92 Chan B, Manley J, Lee J, Singh SR (2015) The emerging roles of microRNAs in cancer metabolism. Cancer Lett 356(2 Pt A): 301-8 Chang GG, Hsu RY (1973) The substrate analog bromopyruvate as a substrate, an inhibitor and an alkylating agent of malic enzyme of pigeon liver. Biochem Biophys Res Commun 55(3): 580-7

129

Chen B, Li H, Zeng X, Yang P, Liu X, Zhao X, Liang S (2012) Roles of microRNA on cancer cell metabolism. Journal of translational medicine 10: 228 Chen J, Zhang S, Li Y, Tang Z, Kong W (2014) Hexokinase 2 overexpression promotes the proliferation and survival of laryngeal squamous cell carcinoma. Tumour Biol 35(4): 3743-53 Chen Y, Cairns R, Papandreou I, Koong A, Denko NC (2009a) Oxygen consumption can regulate the growth of tumors, a new perspective on the Warburg effect. PLoS One 4(9): e7033 Chen Z, Zhang H, Lu W, Huang P (2009b) Role of mitochondria-associated hexokinase II in cancer cell death induced by 3-bromopyruvate. Biochim Biophys Acta 1787(5): 553-60 Cheng EH, Wei MC, Weiler S, Flavell RA, Mak TW, Lindsten T, Korsmeyer SJ (2001) BCL-2, BCL-X(L) sequester BH3 domain-only molecules preventing BAX- and BAK-mediated mitochondrial apoptosis. Mol Cell 8(3): 705-11 Choschzick M, Oosterwijk E, Muller V, Woelber L, Simon R, Moch H, Tennstedt P (2011) Overexpression of carbonic anhydrase IX (CAIX) is an independent unfavorable prognostic marker in endometrioid ovarian cancer. Virchows Archiv : an international journal of pathology 459(2): 193-200 Christofk HR, Vander Heiden MG, Harris MH, Ramanathan A, Gerszten RE, Wei R, Fleming MD, Schreiber SL, Cantley LC (2008a) The M2 splice isoform of pyruvate kinase is important for cancer metabolism and tumour growth. Nature 452(7184): 230-3 Christofk HR, Vander Heiden MG, Wu N, Asara JM, Cantley LC (2008b) Pyruvate kinase M2 is a phosphotyrosine-binding protein. Nature 452(7184): 181-6 Chu QS, Sangha R, Spratlin J, L JV, Mackey JR, McEwan AJ, Venner P, Michelakis ED (2015) A phase I open-labeled, single-arm, dose-escalation, study of dichloroacetate (DCA) in patients with advanced solid tumors. Invest New Drugs Coloff JL, Mason EF, Altman BJ, Gerriets VA, Liu T, Nichols AN, Zhao Y, Wofford JA, Jacobs SR, Ilkayeva O, Garrison SP, Zambetti GP, Rathmell JC (2011) Akt requires glucose metabolism to suppress puma expression and prevent apoptosis of leukemic T cells. J Biol Chem 286(7): 5921-33 Connett RJ, Gayeski TE, Honig CR (1984) Lactate accumulation in fully aerobic, working, dog gracilis muscle. Am J Physiol 246(1 Pt 2): H120-8 Conroy CW, Maren TH (1985) The determination of osteopetrotic phenotypes by selective inactivation of red cell carbonic anhydrase isoenzymes. Clinica chimica acta; international journal of clinical chemistry 152(3): 347-54

130

Cuezva JM, Krajewska M, de Heredia ML, Krajewski S, Santamaria G, Kim H, Zapata JM, Marusawa H, Chamorro M, Reed JC (2002) The bioenergetic signature of cancer: a marker of tumor progression. Cancer Res 62(22): 6674-81 Cuezva JM, Ortega AD, Willers I, Sanchez-Cenizo L, Aldea M, Sanchez-Arago M (2009) The tumor suppressor function of mitochondria: translation into the clinics. Biochim Biophys Acta 1792(12): 1145-58 Dang CV (1999) c-Myc target genes involved in cell growth, apoptosis, and metabolism. Mol Cell Biol 19(1): 1-11 Dang CV, Kim JW, Gao P, Yustein J (2008) The interplay between MYC and HIF in cancer. Nat Rev Cancer 8(1): 51-6 Dang CV, Le A, Gao P (2009) MYC-induced cancer cell energy metabolism and therapeutic opportunities. Clin Cancer Res 15(21): 6479-83 Dang L, White DW, Gross S, Bennett BD, Bittinger MA, Driggers EM, Fantin VR, Jang HG, Jin S, Keenan MC, Marks KM, Prins RM, Ward PS, Yen KE, Liau LM, Rabinowitz JD, Cantley LC, Thompson CB, Vander Heiden MG, Su SM (2010) Cancer-associated IDH1 mutations produce 2-hydroxyglutarate. Nature 465(7300): 966 Dasu A, Toma-Dasu I, Karlsson M (2003) Theoretical simulation of tumour oxygenation and results from acute and chronic hypoxia. Physics in medicine and biology 48(17): 2829-42 David CJ, Chen M, Assanah M, Canoll P, Manley JL (2010) HnRNP proteins controlled by c-Myc deregulate pyruvate kinase mRNA splicing in cancer. Nature 463(7279): 364-8 Davidescu M, Sciaccaluga M, Macchioni L, Angelini R, Lopalco P, Rambotti MG, Roberti R, Corcelli A, Castigli E, Corazzi L (2012) Bromopyruvate mediates autophagy and cardiolipin degradation to monolyso-cardiolipin in GL15 glioblastoma cells. J Bioenerg Biomembr 44(1): 51-60 DeBerardinis RJ, Mancuso A, Daikhin E, Nissim I, Yudkoff M, Wehrli S, Thompson CB (2007) Beyond aerobic glycolysis: transformed cells can engage in glutamine metabolism that exceeds the requirement for protein and nucleotide synthesis. Proc Natl Acad Sci U S A 104(49): 19345-50 Delaney LM, Ho N, Morrison J, Farias NR, Mosser DD, Coomber BL (2015) Dichloroacetate affects proliferation but not survival of human colorectal cancer cells. Apoptosis : an international journal on programmed cell death 20(1): 63-74

131

Deprez J, Vertommen D, Alessi DR, Hue L, Rider MH (1997) Phosphorylation and activation of heart 6-phosphofructo-2-kinase by protein kinase B and other protein kinases of the insulin signaling cascades. J Biol Chem 272(28): 17269-75 Dhar S, Lippard SJ (2009) Mitaplatin, a potent fusion of cisplatin and the orphan drug dichloroacetate. Proc Natl Acad Sci U S A 106(52): 22199-204 Drakaki A, Hatziapostolou M, Iliopoulos D (2013) Therapeutically targeting microRNAs in liver cancer. Current pharmaceutical design 19(7): 1180-91 Draoui N, Feron O (2011) Lactate shuttles at a glance: from physiological paradigms to anti-cancer treatments. Disease models & mechanisms 4(6): 727-32 Druker BJ, Guilhot F, O'Brien SG, Gathmann I, Kantarjian H, Gattermann N, Deininger MW, Silver RT, Goldman JM, Stone RM, Cervantes F, Hochhaus A, Powell BL, Gabrilove JL, Rousselot P, Reiffers J, Cornelissen JJ, Hughes T, Agis H, Fischer T, Verhoef G, Shepherd J, Saglio G, Gratwohl A, Nielsen JL, Radich JP, Simonsson B, Taylor K, Baccarani M, So C, Letvak L, Larson RA, Investigators I (2006) Five-year follow-up of patients receiving imatinib for chronic myeloid leukemia. N Engl J Med 355(23): 2408-17 Druker BJ, Talpaz M, Resta DJ, Peng B, Buchdunger E, Ford JM, Lydon NB, Kantarjian H, Capdeville R, Ohno-Jones S, Sawyers CL (2001) Efficacy and safety of a specific inhibitor of the BCR-ABL tyrosine kinase in chronic myeloid leukemia. N Engl J Med 344(14): 1031-7 Duan Y, Zhao X, Ren W, Wang X, Yu KF, Li D, Zhang X, Zhang Q (2013) Antitumor activity of dichloroacetate on C6 glioma cell: in vitro and in vivo evaluation. OncoTargets and therapy 6: 189-98 Dunbar EM, Coats BS, Shroads AL, Langaee T, Lew A, Forder JR, Shuster JJ, Wagner DA, Stacpoole PW (2013) Phase 1 trial of dichloroacetate (DCA) in adults with recurrent malignant brain tumors. Invest New Drugs Dunbar EM, Coats BS, Shroads AL, Langaee T, Lew A, Forder JR, Shuster JJ, Wagner DA, Stacpoole PW (2014) Phase 1 trial of dichloroacetate (DCA) in adults with recurrent malignant brain tumors. Invest New Drugs 32(3): 452-64 Edge SB, Compton CC (2010) The American Joint Committee on Cancer: the 7th edition of the AJCC cancer staging manual and the future of TNM. Annals of surgical oncology 17(6): 1471-4 Edwards BK, Ward E, Kohler BA, Eheman C, Zauber AG, Anderson RN, Jemal A, Schymura MJ, Lansdorp-Vogelaar I, Seeff LC, van Ballegooijen M, Goede SL, Ries LA (2010) Annual report to the nation on the status of cancer, 1975-2006, featuring

132

colorectal cancer trends and impact of interventions (risk factors, screening, and treatment) to reduce future rates. Cancer 116(3): 544-73 El Sayed SM, El-Magd RM, Shishido Y, Chung SP, Diem TH, Sakai T, Watanabe H, Kagami S, Fukui K (2012) 3-Bromopyruvate antagonizes effects of lactate and pyruvate, synergizes with citrate and exerts novel anti-glioma effects. J Bioenerg Biomembr 44(1): 61-79 El-Mir MY, Nogueira V, Fontaine E, Averet N, Rigoulet M, Leverve X (2000) Dimethylbiguanide inhibits cell respiration via an indirect effect targeted on the respiratory chain complex I. J Biol Chem 275(1): 223-8 Elf SE, Chen J (2014) Targeting glucose metabolism in patients with cancer. Cancer 120(6): 774-80 Elstrom RL, Bauer DE, Buzzai M, Karnauskas R, Harris MH, Plas DR, Zhuang H, Cinalli RM, Alavi A, Rudin CM, Thompson CB (2004) Akt stimulates aerobic glycolysis in cancer cells. Cancer Res 64(11): 3892-9 Evans JM, Donnelly LA, Emslie-Smith AM, Alessi DR, Morris AD (2005) Metformin and reduced risk of cancer in diabetic patients. Bmj 330(7503): 1304-5 Fan J, Shan C, Kang HB, Elf S, Xie J, Tucker M, Gu TL, Aguiar M, Lonning S, Chen H, Mohammadi M, Britton LM, Garcia BA, Aleckovic M, Kang Y, Kaluz S, Devi N, Van Meir EG, Hitosugi T, Seo JH, Lonial S, Gaddh M, Arellano M, Khoury HJ, Khuri FR, Boggon TJ, Kang S, Chen J (2014) Tyr phosphorylation of PDP1 toggles recruitment between ACAT1 and SIRT3 to regulate the pyruvate dehydrogenase complex. Mol Cell 53(4): 534-48 Fang R, Xiao T, Fang Z, Sun Y, Li F, Gao Y, Feng Y, Li L, Wang Y, Liu X, Chen H, Liu XY, Ji H (2012) MicroRNA-143 (miR-143) regulates cancer glycolysis via targeting hexokinase 2 gene. J Biol Chem 287(27): 23227-35 Fantin VR, St-Pierre J, Leder P (2006) Attenuation of LDH-A expression uncovers a link between glycolysis, mitochondrial physiology, and tumor maintenance. Cancer Cell 9(6): 425-34 Farber S, Diamond LK (1948) Temporary remissions in acute leukemia in children produced by folic acid antagonist, 4-aminopteroyl-glutamic acid. N Engl J Med 238(23): 787-93 Figueroa ME, Abdel-Wahab O, Lu C, Ward PS, Patel J, Shih A, Li Y, Bhagwat N, Vasanthakumar A, Fernandez HF, Tallman MS, Sun Z, Wolniak K, Peeters JK, Liu W, Choe SE, Fantin VR, Paietta E, Lowenberg B, Licht JD, Godley LA, Delwel R, Valk PJ, Thompson CB, Levine RL, Melnick A (2010) Leukemic IDH1 and IDH2 mutations

133

result in a hypermethylation phenotype, disrupt TET2 function, and impair hematopoietic differentiation. Cancer Cell 18(6): 553-67 Fischer K, Hoffmann P, Voelkl S, Meidenbauer N, Ammer J, Edinger M, Gottfried E, Schwarz S, Rothe G, Hoves S, Renner K, Timischl B, Mackensen A, Kunz-Schughart L, Andreesen R, Krause SW, Kreutz M (2007) Inhibitory effect of tumor cell-derived lactic acid on human T cells. Blood 109(9): 3812-9 Flatmark K, Maelandsmo GM, Martinsen M, Rasmussen H, Fodstad O (2004) Twelve colorectal cancer cell lines exhibit highly variable growth and metastatic capacities in an orthotopic model in nude mice. European journal of cancer 40(10): 1593-8 Flynt AS, Lai EC (2008) Biological principles of microRNA-mediated regulation: shared themes amid diversity. Nature reviews Genetics 9(11): 831-42 Folkman J (1971) Tumor angiogenesis: therapeutic implications. N Engl J Med 285(21): 1182-6 Fonda ML (1976) Bromopyruvate inactivation of glutamate apodecarboxylase. Kinetics and specificity. J Biol Chem 251(1): 229-35 Freckmann G, Hagenlocher S, Baumstark A, Jendrike N, Gillen RC, Rossner K, Haug C (2007) Continuous glucose profiles in healthy subjects under everyday life conditions and after different meals. Journal of diabetes science and technology 1(5): 695-703 Frisch SM, Ruoslahti E (1997) Integrins and anoikis. Curr Opin Cell Biol 9(5): 701-6 Gall JM, Wong V, Pimental DR, Havasi A, Wang Z, Pastorino JG, Bonegio RG, Schwartz JH, Borkan SC (2011) Hexokinase regulates Bax-mediated mitochondrial membrane injury following ischemic stress. Kidney Int 79(11): 1207-16 Gallagher EJ, LeRoith D (2015) Obesity and Diabetes: The Increased Risk of Cancer and Cancer-Related Mortality. Physiological reviews 95(3): 727-748 Ganapathy-Kanniappan S, Geschwind JF, Kunjithapatham R, Buijs M, Syed LH, Rao PP, Ota S, Vali M (2010) The pyruvic acid analog 3-bromopyruvate interferes with the tetrazolium reagent MTS in the evaluation of cytotoxicity. Assay and drug development technologies 8(2): 258-62 Gandini S, Puntoni M, Heckman-Stoddard BM, Dunn BK, Ford L, DeCensi A, Szabo E (2014) Metformin and cancer risk and mortality: a systematic review and meta-analysis taking into account biases and confounders. Cancer prevention research 7(9): 867-85 Gao D, Vela I, Sboner A, Iaquinta PJ, Karthaus WR, Gopalan A, Dowling C, Wanjala JN, Undvall EA, Arora VK, Wongvipat J, Kossai M, Ramazanoglu S, Barboza LP, Di W, Cao Z, Zhang QF, Sirota I, Ran L, MacDonald TY, Beltran H, Mosquera JM, Touijer

134

KA, Scardino PT, Laudone VP, Curtis KR, Rathkopf DE, Morris MJ, Danila DC, Slovin SF, Solomon SB, Eastham JA, Chi P, Carver B, Rubin MA, Scher HI, Clevers H, Sawyers CL, Chen Y (2014) Organoid cultures derived from patients with advanced prostate cancer. Cell 159(1): 176-87 Gao P, Tchernyshyov I, Chang TC, Lee YS, Kita K, Ochi T, Zeller KI, De Marzo AM, Van Eyk JE, Mendell JT, Dang CV (2009) c-Myc suppression of miR-23a/b enhances mitochondrial glutaminase expression and glutamine metabolism. Nature 458(7239): 762-5 Gatenby RA, Gillies RJ (2004) Why do cancers have high aerobic glycolysis? Nat Rev Cancer 4(11): 891-9 Gillies RJ, Gatenby RA (2007) Adaptive landscapes and emergent phenotypes: why do cancers have high glycolysis? J Bioenerg Biomembr 39(3): 251-7 Gillies RJ, Robey I, Gatenby RA (2008) Causes and consequences of increased glucose metabolism of cancers. J Nucl Med 49 Suppl 2: 24S-42S Gilmore AP (2005) Anoikis. Cell Death Differ 12 Suppl 2: 1473-7 Goethals L, Debucquoy A, Perneel C, Geboes K, Ectors N, De Schutter H, Penninckx F, McBride WH, Begg AC, Haustermans KM (2006) Hypoxia in human colorectal adenocarcinoma: comparison between extrinsic and potential intrinsic hypoxia markers. Int J Radiat Oncol Biol Phys 65(1): 246-54 Gogvadze V, Zhivotovsky B, Orrenius S (2010) The Warburg effect and mitochondrial stability in cancer cells. Mol Aspects Med 31(1): 60-74 Goldberg MS, Sharp PA (2012) Pyruvate kinase M2-specific siRNA induces apoptosis and tumor regression. The Journal of experimental medicine 209(2): 217-24 Gordan JD, Lal P, Dondeti VR, Letrero R, Parekh KN, Oquendo CE, Greenberg RA, Flaherty KT, Rathmell WK, Keith B, Simon MC, Nathanson KL (2008) HIF-alpha effects on c-Myc distinguish two subtypes of sporadic VHL-deficient clear cell renal carcinoma. Cancer Cell 14(6): 435-46 Gottlob K, Majewski N, Kennedy S, Kandel E, Robey RB, Hay N (2001) Inhibition of early apoptotic events by Akt/PKB is dependent on the first committed step of glycolysis and mitochondrial hexokinase. Genes Dev 15(11): 1406-18 Grassian AR, Metallo CM, Coloff JL, Stephanopoulos G, Brugge JS (2011) Erk regulation of pyruvate dehydrogenase flux through PDK4 modulates cell proliferation. Genes Dev 25(16): 1716-33

135

Gregersen LH, Jacobsen A, Frankel LB, Wen J, Krogh A, Lund AH (2012) MicroRNA-143 down-regulates Hexokinase 2 in colon cancer cells. BMC cancer 12: 232 Griffiths EA, Pritchard SA, Valentine HR, Whitchelo N, Bishop PW, Ebert MP, Price PM, Welch IM, West CM (2007) Hypoxia-inducible factor-1alpha expression in the gastric carcinogenesis sequence and its prognostic role in gastric and gastro-oesophageal adenocarcinomas. Br J Cancer 96(1): 95-103 Griguer CE, Oliva CR, Gillespie GY (2005) Glucose metabolism heterogeneity in human and mouse malignant glioma cell lines. Journal of neuro-oncology 74(2): 123-33 Gruber G, Greiner RH, Hlushchuk R, Aebersold DM, Altermatt HJ, Berclaz G, Djonov V (2004) Hypoxia-inducible factor 1 alpha in high-risk breast cancer: an independent prognostic parameter? Breast cancer research : BCR 6(3): R191-8 Guppy M, Leedman P, Zu X, Russell V (2002) Contribution by different fuels and metabolic pathways to the total ATP turnover of proliferating MCF-7 breast cancer cells. Biochem J 364(Pt 1): 309-15 Guzman G, Chennuri R, Chan A, Rea B, Quintana A, Patel R, Xu PZ, Xie H, Hay N (2015) Evidence for heightened hexokinase II immunoexpression in hepatocyte dysplasia and hepatocellular carcinoma. Digestive diseases and sciences 60(2): 420-6 Gwak GY, Yoon JH, Kim KM, Lee HS, Chung JW, Gores GJ (2005) Hypoxia stimulates proliferation of human hepatoma cells through the induction of hexokinase II expression. Journal of hepatology 42(3): 358-64 Habib SL, Rojna M (2013) Diabetes and risk of cancer. ISRN oncology 2013: 583786 Hagg M, Wennstrom S (2005) Activation of hypoxia-induced transcription in normoxia. Exp Cell Res 306(1): 180-91 Haklar G, Sayin-Ozveri E, Yuksel M, Aktan AO, Yalcin AS (2001) Different kinds of reactive oxygen and nitrogen species were detected in colon and breast tumors. Cancer Lett 165(2): 219-24 Halestrap AP (1975) The mitochondrial pyruvate carrier. Kinetics and specificity for substrates and inhibitors. Biochem J 148(1): 85-96 Hamabe A, Yamamoto H, Konno M, Uemura M, Nishimura J, Hata T, Takemasa I, Mizushima T, Nishida N, Kawamoto K, Koseki J, Doki Y, Mori M, Ishii H (2014) Combined evaluation of hexokinase 2 and phosphorylated pyruvate dehydrogenase-E1alpha in invasive front lesions of colorectal tumors predicts cancer metabolism and patient prognosis. Cancer science 105(9): 1100-8 Hanahan D, Weinberg RA (2000) The hallmarks of cancer. Cell 100(1): 57-70

136

Hanahan D, Weinberg RA (2011) Hallmarks of cancer: the next generation. Cell 144(5): 646-74 Hatanaka Y, Hashizume K, Nitta K, Kato T, Itoh I, Tani Y (2003) Cytometrical image analysis for immunohistochemical hormone receptor status in breast carcinomas. Pathology international 53(10): 693-9 Hatziapostolou M, Polytarchou C, Iliopoulos D (2013) miRNAs link metabolic reprogramming to oncogenesis. Trends in endocrinology and metabolism: TEM 24(7): 361-73 Hatzivassiliou G, Zhao F, Bauer DE, Andreadis C, Shaw AN, Dhanak D, Hingorani SR, Tuveson DA, Thompson CB (2005) ATP citrate lyase inhibition can suppress tumor cell growth. Cancer Cell 8(4): 311-21 Hitosugi T, Fan J, Chung TW, Lythgoe K, Wang X, Xie J, Ge Q, Gu TL, Polakiewicz RD, Roesel JL, Chen GZ, Boggon TJ, Lonial S, Fu H, Khuri FR, Kang S, Chen J (2011) Tyrosine phosphorylation of mitochondrial pyruvate dehydrogenase kinase 1 is important for cancer metabolism. Mol Cell 44(6): 864-77 Hockel M, Schlenger K, Aral B, Mitze M, Schaffer U, Vaupel P (1996a) Association between tumor hypoxia and malignant progression in advanced cancer of the uterine cervix. Cancer Res 56(19): 4509-15 Hockel M, Schlenger K, Mitze M, Schaffer U, Vaupel P (1996b) Hypoxia and Radiation Response in Human Tumors. Seminars in radiation oncology 6(1): 3-9 Holash J, Maisonpierre PC, Compton D, Boland P, Alexander CR, Zagzag D, Yancopoulos GD, Wiegand SJ (1999) Vessel cooption, regression, and growth in tumors mediated by angiopoietins and VEGF. Science 284(5422): 1994-8 Hu CJ, Wang LY, Chodosh LA, Keith B, Simon MC (2003) Differential roles of hypoxia-inducible factor 1alpha (HIF-1alpha) and HIF-2alpha in hypoxic gene regulation. Mol Cell Biol 23(24): 9361-74 Hu W, Zhang C, Wu R, Sun Y, Levine A, Feng Z (2010) Glutaminase 2, a novel p53 target gene regulating energy metabolism and antioxidant function. Proc Natl Acad Sci U S A 107(16): 7455-60 Hur H, Xuan Y, Kim YB, Lee G, Shim W, Yun J, Ham IH, Han SU (2013) Expression of pyruvate dehydrogenase kinase-1 in gastric cancer as a potential therapeutic target. Int J Oncol 42(1): 44-54

137

Icard P, Zhang XD, Lemoisson E, Louis MH, Allouche S, Lincet H, Poulain L (2012) Experimental results using 3-bromopyruvate in mesothelioma: in vitro and in vivo studies. J Bioenerg Biomembr 44(1): 81-90 Ihrlund LS, Hernlund E, Khan O, Shoshan MC (2008) 3-Bromopyruvate as inhibitor of tumour cell energy metabolism and chemopotentiator of platinum drugs. Molecular oncology 2(1): 94-101 Inoki K, Corradetti MN, Guan KL (2005) Dysregulation of the TSC-mTOR pathway in human disease. Nature genetics 37(1): 19-24 International Diabetes Foundation (2014) IDF Diabetes Atlas, 6th edn Iozzo P, Hallsten K, Oikonen V, Virtanen KA, Parkkola R, Kemppainen J, Solin O, Lonnqvist F, Ferrannini E, Knuuti J, Nuutila P (2003) Effects of metformin and rosiglitazone monotherapy on insulin-mediated hepatic glucose uptake and their relation to visceral fat in type 2 diabetes. Diabetes Care 26(7): 2069-74 Isaacs JS, Jung YJ, Mole DR, Lee S, Torres-Cabala C, Chung YL, Merino M, Trepel J, Zbar B, Toro J, Ratcliffe PJ, Linehan WM, Neckers L (2005) HIF overexpression correlates with biallelic loss of fumarate hydratase in renal cancer: novel role of fumarate in regulation of HIF stability. Cancer Cell 8(2): 143-53 Isayev O, Rausch V, Bauer N, Liu L, Fan P, Zhang Y, Gladkich J, Nwaeburu CC, Mattern J, Mollenhauer M, Ruckert F, Zach S, Haberkorn U, Gross W, Schonsiegel F, Bazhin AV, Herr I (2014) Inhibition of glucose turnover by 3-bromopyruvate counteracts pancreatic cancer stem cell features and sensitizes cells to gemcitabine. Oncotarget Iyer NV, Kotch LE, Agani F, Leung SW, Laughner E, Wenger RH, Gassmann M, Gearhart JD, Lawler AM, Yu AY, Semenza GL (1998) Cellular and developmental control of O2 homeostasis by hypoxia-inducible factor 1 alpha. Genes Dev 12(2): 149-62 Jiang BH, Semenza GL, Bauer C, Marti HH (1996) Hypoxia-inducible factor 1 levels vary exponentially over a physiologically relevant range of O2 tension. Am J Physiol 271(4 Pt 1): C1172-80 Jiang JX, Gao S, Pan YZ, Yu C, Sun CY (2014) Overexpression of microRNA-125b sensitizes human hepatocellular carcinoma cells to 5-fluorouracil through inhibition of glycolysis by targeting hexokinase II. Molecular medicine reports 10(2): 995-1002 Jiang S, Zhang LF, Zhang HW, Hu S, Lu MH, Liang S, Li B, Li Y, Li D, Wang ED, Liu MF (2012) A novel miR-155/miR-143 cascade controls glycolysis by regulating hexokinase 2 in breast cancer cells. The EMBO journal 31(8): 1985-98 Joensuu H, Roberts PJ, Sarlomo-Rikala M, Andersson LC, Tervahartiala P, Tuveson D, Silberman S, Capdeville R, Dimitrijevic S, Druker B, Demetri GD (2001) Effect of the

138

tyrosine kinase inhibitor STI571 in a patient with a metastatic gastrointestinal stromal tumor. N Engl J Med 344(14): 1052-6 Johnson I, Spence, M. (2010) Probes for Organelles. In Molecular Probes Handbook, A Guide to Fluorescent Probes and Labeling Technologies, 10th edn, Chapter 12. Joung KH, Jeong JW, Ku BJ (2015) The association between type 2 diabetes mellitus and women cancer: the epidemiological evidences and putative mechanisms. BioMed research international 2015: 920618 Kaelin WG, Jr. (2003) The von Hippel-Lindau gene, kidney cancer, and oxygen sensing. J Am Soc Nephrol 14(11): 2703-11 Kallinowski F, Schlenger KH, Runkel S, Kloes M, Stohrer M, Okunieff P, Vaupel P (1989) Blood flow, metabolism, cellular microenvironment, and growth rate of human tumor xenografts. Cancer Res 49(14): 3759-64 Kallinowski F, Vaupel P, Runkel S, Berg G, Fortmeyer HP, Baessler KH, Wagner K, Mueller-Klieser W, Walenta S (1988) Glucose uptake, lactate release, ketone body turnover, metabolic micromilieu, and pH distributions in human breast cancer xenografts in nude rats. Cancer Res 48(24 Pt 1): 7264-72 Kallinowskil F, Davel S, Vaupell P, Baessler KH, Wagner K (1985) Glucose, lactate, and ketone body utilization by human mammary carcinomas in vivo. Advances in experimental medicine and biology 191: 763-73 Kamarajugadda S, Cai Q, Chen H, Nayak S, Zhu J, He M, Jin Y, Zhang Y, Ai L, Martin SS, Tan M, Lu J (2013) Manganese superoxide dismutase promotes anoikis resistance and tumor metastasis. Cell death & disease 4: e504 Kamarajugadda S, Stemboroski L, Cai Q, Simpson NE, Nayak S, Tan M, Lu J (2012) Glucose oxidation modulates anoikis and tumor metastasis. Mol Cell Biol 32(10): 1893-907 Kandoth C, McLellan MD, Vandin F, Ye K, Niu B, Lu C, Xie M, Zhang Q, McMichael JF, Wyczalkowski MA, Leiserson MD, Miller CA, Welch JS, Walter MJ, Wendl MC, Ley TJ, Wilson RK, Raphael BJ, Ding L (2013) Mutational landscape and significance across 12 major cancer types. Nature 502(7471): 333-9 Kantarjian HM, Hochhaus A, Saglio G, De Souza C, Flinn IW, Stenke L, Goh YT, Rosti G, Nakamae H, Gallagher NJ, Hoenekopp A, Blakesley RE, Larson RA, Hughes TP (2011) Nilotinib versus imatinib for the treatment of patients with newly diagnosed chronic phase, Philadelphia chromosome-positive, chronic myeloid leukaemia: 24-month minimum follow-up of the phase 3 randomised ENESTnd trial. The Lancet Oncology 12(9): 841-51

139

Kaye SB (1998) New antimetabolites in cancer chemotherapy and their clinical impact. Br J Cancer 78 Suppl 3: 1-7 Kefas B, Comeau L, Erdle N, Montgomery E, Amos S, Purow B (2010) Pyruvate kinase M2 is a target of the tumor-suppressive microRNA-326 and regulates the survival of glioma cells. Neuro-oncology 12(11): 1102-12 Kikuchi H, Pino MS, Zeng M, Shirasawa S, Chung DC (2009) Oncogenic KRAS and BRAF differentially regulate hypoxia-inducible factor-1alpha and -2alpha in colon cancer. Cancer Res 69(21): 8499-506 Kim JS, Ahn KJ, Kim JA, Kim HM, Lee JD, Lee JM, Kim SJ, Park JH (2008) Role of reactive oxygen species-mediated mitochondrial dysregulation in 3-bromopyruvate induced cell death in hepatoma cells : ROS-mediated cell death by 3-BrPA. J Bioenerg Biomembr 40(6): 607-18 Kim JW, Gao P, Liu YC, Semenza GL, Dang CV (2007) Hypoxia-inducible factor 1 and dysregulated c-Myc cooperatively induce vascular endothelial growth factor and metabolic switches hexokinase 2 and pyruvate dehydrogenase kinase 1. Mol Cell Biol 27(21): 7381-93 Kim JW, Tchernyshyov I, Semenza GL, Dang CV (2006) HIF-1-mediated expression of pyruvate dehydrogenase kinase: a metabolic switch required for cellular adaptation to hypoxia. Cell Metab 3(3): 177-85 Kim NH, Cha YH, Kang SE, Lee Y, Lee I, Cha SY, Ryu JK, Na JM, Park C, Yoon HG, Park GJ, Yook JI, Kim HS (2013) p53 regulates nuclear GSK-3 levels through miR-34-mediated Axin2 suppression in colorectal cancer cells. Cell cycle 12(10): 1578-87 Kivela AJ, Saarnio J, Karttunen TJ, Kivela J, Parkkila AK, Pastorekova S, Pastorek J, Waheed A, Sly WS, Parkkila TS, Rajaniemi H (2001) Differential expression of cytoplasmic carbonic anhydrases, CA I and II, and membrane-associated isozymes, CA IX and XII, in normal mucosa of large intestine and in colorectal tumors. Digestive diseases and sciences 46(10): 2179-86 Kluza J, Corazao-Rozas P, Touil Y, Jendoubi M, Maire C, Guerreschi P, Jonneaux A, Ballot C, Balayssac S, Valable S, Corroyer-Dulmont A, Bernaudin M, Malet-Martino M, de Lassalle EM, Maboudou P, Formstecher P, Polakowska R, Mortier L, Marchetti P (2012) Inactivation of the HIF-1alpha/PDK3 signaling axis drives melanoma toward mitochondrial oxidative metabolism and potentiates the therapeutic activity of pro-oxidants. Cancer Res 72(19): 5035-47 Knoechel TR, Tucker AD, Robinson CM, Phillips C, Taylor W, Bungay PJ, Kasten SA, Roche TE, Brown DG (2006) Regulatory roles of the N-terminal domain based on crystal structures of human pyruvate dehydrogenase kinase 2 containing physiological and synthetic ligands. Biochemistry 45(2): 402-15

140

Ko YH, Pedersen PL, Geschwind JF (2001) Glucose catabolism in the rabbit VX2 tumor model for liver cancer: characterization and targeting hexokinase. Cancer Lett 173(1): 83-91 Ko YH, Smith BL, Wang Y, Pomper MG, Rini DA, Torbenson MS, Hullihen J, Pedersen PL (2004) Advanced cancers: eradication in all cases using 3-bromopyruvate therapy to deplete ATP. Biochem Biophys Res Commun 324(1): 269-75 Ko YH, Verhoeven HA, Lee MJ, Corbin DJ, Vogl TJ, Pedersen PL (2012) A translational study "case report" on the small molecule "energy blocker" 3-bromopyruvate (3BP) as a potent anticancer agent: from bench side to bedside. J Bioenerg Biomembr 44(1): 163-70 Kolobova E, Tuganova A, Boulatnikov I, Popov KM (2001) Regulation of pyruvate dehydrogenase activity through phosphorylation at multiple sites. Biochem J 358(Pt 1): 69-77 Koppenol WH, Bounds PL, Dang CV (2011) Otto Warburg's contributions to current concepts of cancer metabolism. Nat Rev Cancer 11(5): 325-37 Kroemer G, Dallaporta B, Resche-Rigon M (1998) The mitochondrial death/life regulator in apoptosis and necrosis. Annual review of physiology 60: 619-42 Kroemer G, Galluzzi L, Brenner C (2007) Mitochondrial membrane permeabilization in cell death. Physiological reviews 87(1): 99-163 Kruse V, Cocquyt V, Borms M, Maes A, Van de Wiele C (2013) Serum tumor markers and PET/CT imaging for tumor recurrence detection. Annals of nuclear medicine 27(2): 97-104 Kumar A, Kant S, Singh SM (2012) Novel molecular mechanisms of antitumor action of dichloroacetate against T cell lymphoma: Implication of altered glucose metabolism, pH homeostasis and cell survival regulation. Chemico-biological interactions 199(1): 29-37 Kumar K, Wigfield S, Gee HE, Devlin CM, Singleton D, Li JL, Buffa F, Huffman M, Sinn AL, Silver J, Turley H, Leek R, Harris AL, Ivan M (2013) Dichloroacetate reverses the hypoxic adaptation to bevacizumab and enhances its antitumor effects in mouse xenografts. J Mol Med (Berl) 91(6): 749-58 Kurtoglu M, Gao N, Shang J, Maher JC, Lehrman MA, Wangpaichitr M, Savaraj N, Lane AN, Lampidis TJ (2007) Under normoxia, 2-deoxy-D-glucose elicits cell death in select tumor types not by inhibition of glycolysis but by interfering with N-linked glycosylation. Mol Cancer Ther 6(11): 3049-58

141

Kwee SA, Hernandez B, Chan O, Wong L (2012) Choline kinase alpha and hexokinase-2 protein expression in hepatocellular carcinoma: association with survival. PLoS One 7(10): e46591 Kwon OH, Kang TW, Kim JH, Kim M, Noh SM, Song KS, Yoo HS, Kim WH, Xie Z, Pocalyko D, Kim SY, Kim YS (2012) Pyruvate kinase M2 promotes the growth of gastric cancer cells via regulation of Bcl-xL expression at transcriptional level. Biochem Biophys Res Commun 423(1): 38-44 Landriscina M, Maddalena F, Laudiero G, Esposito F (2009) Adaptation to oxidative stress, chemoresistance, and cell survival. Antioxidants & redox signaling 11(11): 2701-16 Laughner E, Taghavi P, Chiles K, Mahon PC, Semenza GL (2001) HER2 (neu) signaling increases the rate of hypoxia-inducible factor 1alpha (HIF-1alpha) synthesis: novel mechanism for HIF-1-mediated vascular endothelial growth factor expression. Mol Cell Biol 21(12): 3995-4004 Le A, Cooper CR, Gouw AM, Dinavahi R, Maitra A, Deck LM, Royer RE, Vander Jagt DL, Semenza GL, Dang CV (2010) Inhibition of lactate dehydrogenase A induces oxidative stress and inhibits tumor progression. Proc Natl Acad Sci U S A 107(5): 2037-42 Lea MA, Qureshi MS, Buxhoeveden M, Gengel N, Kleinschmit J, Desbordes C (2013) Regulation of the proliferation of colon cancer cells by compounds that affect glycolysis, including 3-bromopyruvate, 2-deoxyglucose and biguanides. Anticancer Res 33(2): 401-7 Leary SC, Sasarman F, Nishimura T, Shoubridge EA (2009) Human SCO2 is required for the synthesis of CO II and as a thiol-disulphide oxidoreductase for SCO1. Human molecular genetics 18(12): 2230-40 Lee JH, Kim EJ, Kim DK, Lee JM, Park SB, Lee IK, Harris RA, Lee MO, Choi HS (2012) Hypoxia induces PDK4 gene expression through induction of the orphan nuclear receptor ERRgamma. PLoS One 7(9): e46324 Leiblich A, Cross SS, Catto JW, Phillips JT, Leung HY, Hamdy FC, Rehman I (2006) Lactate dehydrogenase-B is silenced by promoter hypermethylation in human prostate cancer. Oncogene 25(20): 2953-60 Levine AJ, Puzio-Kuter AM (2010) The control of the metabolic switch in cancers by oncogenes and tumor suppressor genes. Science 330(6009): 1340-4 Li AE, Ito H, Rovira, II, Kim KS, Takeda K, Yu ZY, Ferrans VJ, Finkel T (1999) A role for reactive oxygen species in endothelial cell anoikis. Circ Res 85(4): 304-10

142

Li Y, Wang H, Oosterwijk E, Selman Y, Mira JC, Medrano T, Shiverick KT, Frost SC (2009) Antibody-specific detection of CAIX in breast and prostate cancers. Biochem Biophys Res Commun 386(3): 488-92 Liu L, Wang S, Zhang Q, Ding Y (2008) Identification of potential genes/proteins regulated by Tiam1 in colorectal cancer by microarray analysis and proteome analysis. Cell biology international 32(10): 1215-22 Llambi F, Moldoveanu T, Tait SW, Bouchier-Hayes L, Temirov J, McCormick LL, Dillon CP, Green DR (2011) A unified model of mammalian BCL-2 protein family interactions at the mitochondria. Mol Cell 44(4): 517-31 Loncaster JA, Harris AL, Davidson SE, Logue JP, Hunter RD, Wycoff CC, Pastorek J, Ratcliffe PJ, Stratford IJ, West CM (2001) Carbonic anhydrase (CA IX) expression, a potential new intrinsic marker of hypoxia: correlations with tumor oxygen measurements and prognosis in locally advanced carcinoma of the cervix. Cancer Res 61(17): 6394-9 Longley DB, Harkin DP, Johnston PG (2003) 5-fluorouracil: mechanisms of action and clinical strategies. Nat Rev Cancer 3(5): 330-8 Lopez-Lazaro M (2006) Hypoxia-inducible factor 1 as a possible target for cancer chemoprevention. Cancer Epidemiol Biomarkers Prev 15(12): 2332-5 Lopez-Rios F, Sanchez-Arago M, Garcia-Garcia E, Ortega AD, Berrendero JR, Pozo-Rodriguez F, Lopez-Encuentra A, Ballestin C, Cuezva JM (2007) Loss of the mitochondrial bioenergetic capacity underlies the glucose avidity of carcinomas. Cancer Res 67(19): 9013-7 Lora J, Alonso FJ, Segura JA, Lobo C, Marquez J, Mates JM (2004) Antisense glutaminase inhibition decreases glutathione antioxidant capacity and increases apoptosis in Ehrlich ascitic tumour cells. Eur J Biochem 271(21): 4298-306 Lorini M, Ciman M (1962) Hypoglycaemic action of diisopropyl-ammonium salts in experimental diabetes. Biochem Pharmacol 11: 823-7 Lowman XH, McDonnell MA, Kosloske A, Odumade OA, Jenness C, Karim CB, Jemmerson R, Kelekar A (2010) The proapoptotic function of Noxa in human leukemia cells is regulated by the kinase Cdk5 and by glucose. Mol Cell 40(5): 823-33 Lu C, Ward PS, Kapoor GS, Rohle D, Turcan S, Abdel-Wahab O, Edwards CR, Khanin R, Figueroa ME, Melnick A, Wellen KE, O'Rourke DM, Berger SL, Chan TA, Levine RL, Mellinghoff IK, Thompson CB (2012) IDH mutation impairs histone demethylation and results in a block to cell differentiation. Nature 483(7390): 474-8

143

Lu CW, Lin SC, Chien CW, Lee CT, Lin BW, Lee JC, Tsai SJ (2011) Overexpression of pyruvate dehydrogenase kinase 3 increases drug resistance and early recurrence in colon cancer. Am J Pathol 179(3): 1405-14 Lu H, Forbes RA, Verma A (2002) Hypoxia-inducible factor 1 activation by aerobic glycolysis implicates the Warburg effect in carcinogenesis. J Biol Chem 277(26): 23111-5 Luo Q, Hu D, Hu S, Yan M, Sun Z, Chen F (2012) In vitro and in vivo anti-tumor effect of metformin as a novel therapeutic agent in human oral squamous cell carcinoma. BMC cancer 12: 517 Lyng H, Brovig RS, Svendsrud DH, Holm R, Kaalhus O, Knutstad K, Oksefjell H, Sundfor K, Kristensen GB, Stokke T (2006) Gene expressions and copy numbers associated with metastatic phenotypes of uterine cervical cancer. BMC genomics 7: 268 Lyshchik A, Higashi T, Hara T, Nakamoto Y, Fujimoto K, Doi R, Imamura M, Saga T, Togashi K (2007) Expression of glucose transporter-1, hexokinase-II, proliferating cell nuclear antigen and survival of patients with pancreatic cancer. Cancer Invest 25(3): 154-62 Macheda ML, Rogers S, Best JD (2005) Molecular and cellular regulation of glucose transporter (GLUT) proteins in cancer. J Cell Physiol 202(3): 654-62 Madhok BM, Yeluri S, Perry SL, Hughes TA, Jayne DG (2010) Dichloroacetate induces apoptosis and cell-cycle arrest in colorectal cancer cells. Br J Cancer 102(12): 1746-52 Maekawa M, Taniguchi T, Ishikawa J, Sugimura H, Sugano K, Kanno T (2003) Promoter hypermethylation in cancer silences LDHB, eliminating lactate dehydrogenase isoenzymes 1-4. Clin Chem 49(9): 1518-20 Magistretti PJ, Pellerin L (1996) Cellular mechanisms of brain energy metabolism. Relevance to functional brain imaging and to neurodegenerative disorders. Ann N Y Acad Sci 777: 380-7 Majewski N, Nogueira V, Bhaskar P, Coy PE, Skeen JE, Gottlob K, Chandel NS, Thompson CB, Robey RB, Hay N (2004a) Hexokinase-mitochondria interaction mediated by Akt is required to inhibit apoptosis in the presence or absence of Bax and Bak. Mol Cell 16(5): 819-30 Majewski N, Nogueira V, Robey RB, Hay N (2004b) Akt inhibits apoptosis downstream of BID cleavage via a glucose-dependent mechanism involving mitochondrial hexokinases. Mol Cell Biol 24(2): 730-40 Mardis ER, Ding L, Dooling DJ, Larson DE, McLellan MD, Chen K, Koboldt DC, Fulton RS, Delehaunty KD, McGrath SD, Fulton LA, Locke DP, Magrini VJ, Abbott

144

RM, Vickery TL, Reed JS, Robinson JS, Wylie T, Smith SM, Carmichael L, Eldred JM, Harris CC, Walker J, Peck JB, Du F, Dukes AF, Sanderson GE, Brummett AM, Clark E, McMichael JF, Meyer RJ, Schindler JK, Pohl CS, Wallis JW, Shi X, Lin L, Schmidt H, Tang Y, Haipek C, Wiechert ME, Ivy JV, Kalicki J, Elliott G, Ries RE, Payton JE, Westervelt P, Tomasson MH, Watson MA, Baty J, Heath S, Shannon WD, Nagarajan R, Link DC, Walter MJ, Graubert TA, DiPersio JF, Wilson RK, Ley TJ (2009) Recurring mutations found by sequencing an acute myeloid leukemia genome. N Engl J Med 361(11): 1058-66 Marin-Valencia I, Yang C, Mashimo T, Cho S, Baek H, Yang XL, Rajagopalan KN, Maddie M, Vemireddy V, Zhao Z, Cai L, Good L, Tu BP, Hatanpaa KJ, Mickey BE, Mates JM, Pascual JM, Maher EA, Malloy CR, Deberardinis RJ, Bachoo RM (2012) Analysis of tumor metabolism reveals mitochondrial glucose oxidation in genetically diverse human glioblastomas in the mouse brain in vivo. Cell Metab 15(6): 827-37 Martinez-Outschoorn UE, Balliet RM, Rivadeneira DB, Chiavarina B, Pavlides S, Wang C, Whitaker-Menezes D, Daumer KM, Lin Z, Witkiewicz AK, Flomenberg N, Howell A, Pestell RG, Knudsen ES, Sotgia F, Lisanti MP (2010) Oxidative stress in cancer associated fibroblasts drives tumor-stroma co-evolution: A new paradigm for understanding tumor metabolism, the field effect and genomic instability in cancer cells. Cell cycle 9(16): 3256-76 Martinez-Zaguilan R, Seftor EA, Seftor RE, Chu YW, Gillies RJ, Hendrix MJ (1996) Acidic pH enhances the invasive behavior of human melanoma cells. Clinical & experimental metastasis 14(2): 176-86 Maschek G, Savaraj N, Priebe W, Braunschweiger P, Hamilton K, Tidmarsh GF, De Young LR, Lampidis TJ (2004) 2-deoxy-D-glucose increases the efficacy of adriamycin and paclitaxel in human osteosarcoma and non-small cell lung cancers in vivo. Cancer Res 64(1): 31-4 Mates JM, Segura JA, Campos-Sandoval JA, Lobo C, Alonso L, Alonso FJ, Marquez J (2009) Glutamine homeostasis and mitochondrial dynamics. Int J Biochem Cell Biol 41(10): 2051-61 Mathupala SP, Heese C, Pedersen PL (1997a) Glucose catabolism in cancer cells. The type II hexokinase promoter contains functionally active response elements for the tumor suppressor p53. J Biol Chem 272(36): 22776-80 Mathupala SP, Rempel A, Pedersen PL (1997b) Aberrant glycolytic metabolism of cancer cells: a remarkable coordination of genetic, transcriptional, post-translational, and mutational events that lead to a critical role for type II hexokinase. J Bioenerg Biomembr 29(4): 339-43 Matoba S, Kang JG, Patino WD, Wragg A, Boehm M, Gavrilova O, Hurley PJ, Bunz F, Hwang PM (2006) p53 regulates mitochondrial respiration. Science 312(5780): 1650-3

145

Matsuyama T, Nakanishi K, Hayashi T, Yoshizumi Y, Aiko S, Sugiura Y, Tanimoto T, Uenoyama M, Ozeki Y, Maehara T (2005) Expression of hypoxia-inducible factor-1alpha in esophageal squamous cell carcinoma. Cancer science 96(3): 176-82 Maxwell PH, Wiesener MS, Chang GW, Clifford SC, Vaux EC, Cockman ME, Wykoff CC, Pugh CW, Maher ER, Ratcliffe PJ (1999) The tumour suppressor protein VHL targets hypoxia-inducible factors for oxygen-dependent proteolysis. Nature 399(6733): 271-5 Mayer D, Klimek F, Rempel A, Bannasch P (1997) Hexokinase expression in liver preneoplasia and neoplasia. Biochem Soc Trans 25(1): 122-7 Mazurek S (2011) Pyruvate kinase type M2: a key regulator of the metabolic budget system in tumor cells. Int J Biochem Cell Biol 43(7): 969-80 Mazurek S, Boschek CB, Hugo F, Eigenbrodt E (2005) Pyruvate kinase type M2 and its role in tumor growth and spreading. Semin Cancer Biol 15(4): 300-8 McAllister A, Allison SP, Randle PJ (1973) Effects of dichloroacetate on the metabolism of glucose, pyruvate, acetate, 3-hydroxybutyrate and palmitate in rat diaphragm and heart muscle in vitro and on extraction of glucose, lactate, pyruvate and free fatty acids by dog heart in vivo. Biochem J 134(4): 1067-81 McCullagh KJ, Poole RC, Halestrap AP, Tipton KF, O'Brien M, Bonen A (1997) Chronic electrical stimulation increases MCT1 and lactate uptake in red and white skeletal muscle. Am J Physiol 273(2 Pt 1): E239-46 McIlwain DR, Berger T, Mak TW (2013) Caspase functions in cell death and disease. Cold Spring Harbor perspectives in biology 5(4): a008656 Mesker WE, Junggeburt JM, Szuhai K, de Heer P, Morreau H, Tanke HJ, Tollenaar RA (2007) The carcinoma-stromal ratio of colon carcinoma is an independent factor for survival compared to lymph node status and tumor stage. Cell Oncol 29(5): 387-98 Metallo CM, Gameiro PA, Bell EL, Mattaini KR, Yang J, Hiller K, Jewell CM, Johnson ZR, Irvine DJ, Guarente L, Kelleher JK, Vander Heiden MG, Iliopoulos O, Stephanopoulos G (2012) Reductive glutamine metabolism by IDH1 mediates lipogenesis under hypoxia. Nature 481(7381): 380-4 Meyer LM, Miller FR, Rowen MJ, Bock G, Rutzky J (1950) Treatment of acute leukemia with amethopterin (4-amino, 10-methyl pteroyl glutamic acid). Acta haematologica 4(3): 157-67 Michelakis ED, Sutendra G, Dromparis P, Webster L, Haromy A, Niven E, Maguire C, Gammer TL, Mackey JR, Fulton D, Abdulkarim B, McMurtry MS, Petruk KC (2010)

146

Metabolic modulation of glioblastoma with dichloroacetate. Sci Transl Med 2(31): 31ra34 Michelakis ED, Webster L, Mackey JR (2008) Dichloroacetate (DCA) as a potential metabolic-targeting therapy for cancer. Br J Cancer 99(7): 989-94 Miura T, Tanno M (2012) The mPTP and its regulatory proteins: final common targets of signalling pathways for protection against necrosis. Cardiovascular research 94(2): 181-9 Miyamoto S, Murphy AN, Brown JH (2009a) Akt mediated mitochondrial protection in the heart: metabolic and survival pathways to the rescue. J Bioenerg Biomembr 41(2): 169-80 Miyamoto S, Rubio M, Sussman MA (2009b) Nuclear and mitochondrial signalling Akts in cardiomyocytes. Cardiovascular research 82(2): 272-85 Nakano A, Miki H, Nakamura S, Harada T, Oda A, Amou H, Fujii S, Kagawa K, Takeuchi K, Ozaki S, Matsumoto T, Abe M (2012) Up-regulation of hexokinaseII in myeloma cells: targeting myeloma cells with 3-bromopyruvate. J Bioenerg Biomembr 44(1): 31-8 Nakano A, Tsuji D, Miki H, Cui Q, El Sayed SM, Ikegame A, Oda A, Amou H, Nakamura S, Harada T, Fujii S, Kagawa K, Takeuchi K, Sakai A, Ozaki S, Okano K, Nakamura T, Itoh K, Matsumoto T, Abe M (2011) Glycolysis inhibition inactivates ABC transporters to restore drug sensitivity in malignant cells. PLoS One 6(11): e27222 Nakashima RA, Mangan PS, Colombini M, Pedersen PL (1986) Hexokinase receptor complex in hepatoma mitochondria: evidence from N,N'-dicyclohexylcarbodiimide-labeling studies for the involvement of the pore-forming protein VDAC. Biochemistry 25(5): 1015-21 Newsholme EA, Crabtree B, Ardawi MS (1985) The role of high rates of glycolysis and glutamine utilization in rapidly dividing cells. Bioscience reports 5(5): 393-400 Niemann S, Muller U (2000) Mutations in SDHC cause autosomal dominant paraganglioma, type 3. Nature genetics 26(3): 268-70 Niemela AM, Hynninen P, Mecklin JP, Kuopio T, Kokko A, Aaltonen L, Parkkila AK, Pastorekova S, Pastorek J, Waheed A, Sly WS, Orntoft TF, Kruhoffer M, Haapasalo H, Parkkila S, Kivela AJ (2007) Carbonic anhydrase IX is highly expressed in hereditary nonpolyposis colorectal cancer. Cancer Epidemiol Biomarkers Prev 16(9): 1760-6 Nordsmark M, Bentzen SM, Rudat V, Brizel D, Lartigau E, Stadler P, Becker A, Adam M, Molls M, Dunst J, Terris DJ, Overgaard J (2005) Prognostic value of tumor oxygenation in 397 head and neck tumors after primary radiation therapy. An

147

international multi-center study. Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology 77(1): 18-24 Nordsmark M, Overgaard M, Overgaard J (1996) Pretreatment oxygenation predicts radiation response in advanced squamous cell carcinoma of the head and neck. Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology 41(1): 31-9 Ohashi T, Akazawa T, Aoki M, Kuze B, Mizuta K, Ito Y, Inoue N (2013) Dichloroacetate improves immune dysfunction caused by tumor-secreted lactic acid and increases antitumor immunoreactivity. Int J Cancer 133(5): 1107-18 Oronsky BT, Reid T, Knox SJ, Scicinski JJ (2012) The scarlet letter of alkylation: a mini review of selective alkylating agents. Translational oncology 5(4): 226-9 Orrenius S, Gogvadze V, Zhivotovsky B (2007) Mitochondrial oxidative stress: implications for cell death. Annual review of pharmacology and toxicology 47: 143-83 Oudard S, Boitier E, Miccoli L, Rousset S, Dutrillaux B, Poupon MF (1997) Gliomas are driven by glycolysis: putative roles of hexokinase, oxidative phosphorylation and mitochondrial ultrastructure. Anticancer Res 17(3C): 1903-11 Owen MR, Doran E, Halestrap AP (2000) Evidence that metformin exerts its anti-diabetic effects through inhibition of complex 1 of the mitochondrial respiratory chain. Biochem J 348 Pt 3: 607-14 Oyedotun KS, Lemire BD (2004) The quaternary structure of the Saccharomyces cerevisiae succinate dehydrogenase. Homology modeling, cofactor docking, and molecular dynamics simulation studies. J Biol Chem 279(10): 9424-31 Palmieri D, Fitzgerald D, Shreeve SM, Hua E, Bronder JL, Weil RJ, Davis S, Stark AM, Merino MJ, Kurek R, Mehdorn HM, Davis G, Steinberg SM, Meltzer PS, Aldape K, Steeg PS (2009) Analyses of resected human brain metastases of breast cancer reveal the association between up-regulation of hexokinase 2 and poor prognosis. Mol Cancer Res 7(9): 1438-45 Pani G, Colavitti R, Bedogni B, Fusco S, Ferraro D, Borrello S, Galeotti T (2004) Mitochondrial superoxide dismutase: a promising target for new anticancer therapies. Curr Med Chem 11(10): 1299-308 Papandreou I, Cairns RA, Fontana L, Lim AL, Denko NC (2006) HIF-1 mediates adaptation to hypoxia by actively downregulating mitochondrial oxygen consumption. Cell Metab 3(3): 187-97 Papetti M, Herman IM (2002) Mechanisms of normal and tumor-derived angiogenesis. Am J Physiol Cell Physiol 282(5): C947-70

148

Park J, Chen Y, Tishkoff DX, Peng C, Tan M, Dai L, Xie Z, Zhang Y, Zwaans BM, Skinner ME, Lombard DB, Zhao Y (2013) SIRT5-mediated lysine desuccinylation impacts diverse metabolic pathways. Mol Cell 50(6): 919-30 Park JK, Kogure T, Nuovo GJ, Jiang J, He L, Kim JH, Phelps MA, Papenfuss TL, Croce CM, Patel T, Schmittgen TD (2011) miR-221 silencing blocks hepatocellular carcinoma and promotes survival. Cancer Res 71(24): 7608-16 Parkins CS, Dennis MF, Stratford MR, Hill SA, Chaplin DJ (1995) Ischemia reperfusion injury in tumors: the role of oxygen radicals and nitric oxide. Cancer Res 55(24): 6026-9 Parsons DW, Jones S, Zhang X, Lin JC, Leary RJ, Angenendt P, Mankoo P, Carter H, Siu IM, Gallia GL, Olivi A, McLendon R, Rasheed BA, Keir S, Nikolskaya T, Nikolsky Y, Busam DA, Tekleab H, Diaz LA, Jr., Hartigan J, Smith DR, Strausberg RL, Marie SK, Shinjo SM, Yan H, Riggins GJ, Bigner DD, Karchin R, Papadopoulos N, Parmigiani G, Vogelstein B, Velculescu VE, Kinzler KW (2008) An integrated genomic analysis of human glioblastoma multiforme. Science 321(5897): 1807-12 Pastorino JG, Hoek JB, Shulga N (2005) Activation of glycogen synthase kinase 3beta disrupts the binding of hexokinase II to mitochondria by phosphorylating voltage-dependent anion channel and potentiates chemotherapy-induced cytotoxicity. Cancer Res 65(22): 10545-54 Pastorino JG, Shulga N, Hoek JB (2002) Mitochondrial binding of hexokinase II inhibits Bax-induced cytochrome c release and apoptosis. J Biol Chem 277(9): 7610-8 Pathak RK, Marrache S, Harn DA, Dhar S (2014) Mito-DCA: a mitochondria targeted molecular scaffold for efficacious delivery of metabolic modulator dichloroacetate. ACS chemical biology 9(5): 1178-87 Pavlides S, Whitaker-Menezes D, Castello-Cros R, Flomenberg N, Witkiewicz AK, Frank PG, Casimiro MC, Wang C, Fortina P, Addya S, Pestell RG, Martinez-Outschoorn UE, Sotgia F, Lisanti MP (2009) The reverse Warburg effect: Aerobic glycolysis in cancer associated fibroblasts and the tumor stroma. Cell cycle 8(23): 3984-4001 Pedersen PL (2007) Warburg, me and Hexokinase 2: Multiple discoveries of key molecular events underlying one of cancers' most common phenotypes, the "Warburg Effect", i.e., elevated glycolysis in the presence of oxygen. J Bioenerg Biomembr 39(3): 211-22 Pelicano H, Xu RH, Du M, Feng L, Sasaki R, Carew JS, Hu Y, Ramdas L, Hu L, Keating MJ, Zhang W, Plunkett W, Huang P (2006) Mitochondrial respiration defects in cancer cells cause activation of Akt survival pathway through a redox-mediated mechanism. J Cell Biol 175(6): 913-23

149

Pelletier H, Millot JM, Chauffert B, Manfait M, Genne P, Martin F (1990) Mechanisms of resistance of confluent human and rat colon cancer cells to anthracyclines: alteration of drug passive diffusion. Cancer Res 50(20): 6626-31 Pereira da Silva AP, El-Bacha T, Kyaw N, dos Santos RS, da-Silva WS, Almeida FC, Da Poian AT, Galina A (2009) Inhibition of energy-producing pathways of HepG2 cells by 3-bromopyruvate. Biochem J 417(3): 717-26 Peschiaroli A, Giacobbe A, Formosa A, Markert EK, Bongiorno-Borbone L, Levine AJ, Candi E, D'Alessandro A, Zolla L, Finazzi Agro A, Melino G (2013) miR-143 regulates hexokinase 2 expression in cancer cells. Oncogene 32(6): 797-802 Peters GJ, Backus HH, Freemantle S, van Triest B, Codacci-Pisanelli G, van der Wilt CL, Smid K, Lunec J, Calvert AH, Marsh S, McLeod HL, Bloemena E, Meijer S, Jansen G, van Groeningen CJ, Pinedo HM (2002) Induction of thymidylate synthase as a 5-fluorouracil resistance mechanism. Biochim Biophys Acta 1587(2-3): 194-205 Piao L, Fang YH, Cadete VJ, Wietholt C, Urboniene D, Toth PT, Marsboom G, Zhang HJ, Haber I, Rehman J, Lopaschuk GD, Archer SL (2010) The inhibition of pyruvate dehydrogenase kinase improves impaired cardiac function and electrical remodeling in two models of right ventricular hypertrophy: resuscitating the hibernating right ventricle. J Mol Med (Berl) 88(1): 47-60 Pinheiro C, Longatto-Filho A, Pereira SM, Etlinger D, Moreira MA, Jube LF, Queiroz GS, Schmitt F, Baltazar F (2009) Monocarboxylate transporters 1 and 4 are associated with CD147 in cervical carcinoma. Disease markers 26(3): 97-103 Pinheiro C, Reis RM, Ricardo S, Longatto-Filho A, Schmitt F, Baltazar F (2010) Expression of monocarboxylate transporters 1, 2, and 4 in human tumours and their association with CD147 and CD44. Journal of biomedicine & biotechnology 2010: 427694 Plas DR, Thompson CB (2005) Akt-dependent transformation: there is more to growth than just surviving. Oncogene 24(50): 7435-42 Plecita-Hlavata L, Lessard M, Santorova J, Bewersdorf J, Jezek P (2008) Mitochondrial oxidative phosphorylation and energetic status are reflected by morphology of mitochondrial network in INS-1E and HEP-G2 cells viewed by 4Pi microscopy. Biochim Biophys Acta 1777(7-8): 834-46 Plumb CL, Adamcic U, Shahrzad S, Minhas K, Adham SA, Coomber BL (2009) Modulation of the tumor suppressor protein alpha-catenin by ischemic microenvironment. Am J Pathol 175(4): 1662-74

150

Populo H, Caldas R, Lopes JM, Pardal J, Maximo V, Soares P (2015) Overexpression of pyruvate dehydrogenase kinase supports dichloroacetate as a candidate for cutaneous melanoma therapy. Expert Opin Ther Targets 19(6): 733-45 Potter C, Harris AL (2004) Hypoxia inducible carbonic anhydrase IX, marker of tumour hypoxia, survival pathway and therapy target. Cell cycle 3(2): 164-7 Pradelli LA, Beneteau M, Chauvin C, Jacquin MA, Marchetti S, Munoz-Pinedo C, Auberger P, Pende M, Ricci JE (2010) Glycolysis inhibition sensitizes tumor cells to death receptors-induced apoptosis by AMP kinase activation leading to Mcl-1 block in translation. Oncogene 29(11): 1641-52 Price GS, Page RL, Riviere JE, Cline JM, Thrall DE (1996) Pharmacokinetics and toxicity of oral and intravenous lonidamine in dogs. Cancer Chemother Pharmacol 38(2): 129-35 Purcell WT, Ettinger DS (2003) Novel antifolate drugs. Current oncology reports 5(2): 114-25 Qiu MZ, Han B, Luo HY, Zhou ZW, Wang ZQ, Wang FH, Li YH, Xu RH (2011) Expressions of hypoxia-inducible factor-1alpha and hexokinase-II in gastric adenocarcinoma: the impact on prognosis and correlation to clinicopathologic features. Tumour Biol 32(1): 159-66 Queiros O, Preto A, Pacheco A, Pinheiro C, Azevedo-Silva J, Moreira R, Pedro M, Ko YH, Pedersen PL, Baltazar F, Casal M (2012) Butyrate activates the monocarboxylate transporter MCT4 expression in breast cancer cells and enhances the antitumor activity of 3-bromopyruvate. J Bioenerg Biomembr 44(1): 141-53 Raghunand N, Gatenby RA, Gillies RJ (2003) Microenvironmental and cellular consequences of altered blood flow in tumours. The British journal of radiology 76 Spec No 1: S11-22 Ramanathan A, Wang C, Schreiber SL (2005) Perturbational profiling of a cell-line model of tumorigenesis by using metabolic measurements. Proc Natl Acad Sci U S A 102(17): 5992-7 Rardin MJ, Wiley SE, Naviaux RK, Murphy AN, Dixon JE (2009) Monitoring phosphorylation of the pyruvate dehydrogenase complex. Analytical biochemistry 389(2): 157-64 Rasola A, Sciacovelli M, Pantic B, Bernardi P (2010) Signal transduction to the permeability transition pore. FEBS Lett 584(10): 1989-96

151

Rempel A, Mathupala SP, Griffin CA, Hawkins AL, Pedersen PL (1996) Glucose catabolism in cancer cells: amplification of the gene encoding type II hexokinase. Cancer Res 56(11): 2468-71 Rigo P, Paulus P, Kaschten BJ, Hustinx R, Bury T, Jerusalem G, Benoit T, Foidart-Willems J (1996) Oncological applications of positron emission tomography with fluorine-18 fluorodeoxyglucose. European journal of nuclear medicine 23(12): 1641-74 Roberts DJ, Miyamoto S (2015) Hexokinase II integrates energy metabolism and cellular protection: Akting on mitochondria and TORCing to autophagy. Cell Death Differ 22(2): 248-57 Roberts DJ, Tan-Sah VP, Smith JM, Miyamoto S (2013) Akt phosphorylates HK-II at Thr-473 and increases mitochondrial HK-II association to protect cardiomyocytes. J Biol Chem 288(33): 23798-806 Robey RB, Hay N (2009) Is Akt the "Warburg kinase"?-Akt-energy metabolism interactions and oncogenesis. Semin Cancer Biol 19(1): 25-31 Ronnov-Jessen L, Petersen OW, Bissell MJ (1996) Cellular changes involved in conversion of normal to malignant breast: importance of the stromal reaction. Physiological reviews 76(1): 69-125 Ross CD, Gomaa MA, Gillies E, Juengel R, Medina JE (2000) Tumor grade, microvessel density, and activities of malate dehydrogenase, lactate dehydrogenase, and hexokinase in squamous cell carcinoma. Otolaryngology--head and neck surgery : official journal of American Academy of Otolaryngology-Head and Neck Surgery 122(2): 195-200 Rossignol R, Gilkerson R, Aggeler R, Yamagata K, Remington SJ, Capaldi RA (2004) Energy substrate modulates mitochondrial structure and oxidative capacity in cancer cells. Cancer Res 64(3): 985-93 Roy HK, Olusola BF, Clemens DL, Karolski WJ, Ratashak A, Lynch HT, Smyrk TC (2002) AKT proto-oncogene overexpression is an early event during sporadic colon carcinogenesis. Carcinogenesis 23(1): 201-5 Rozen S, Skaletsky HJ (1998) Primer3 Saarnio J, Parkkila S, Parkkila AK, Haukipuro K, Pastorekova S, Pastorek J, Kairaluoma MI, Karttunen TJ (1998a) Immunohistochemical study of colorectal tumors for expression of a novel transmembrane carbonic anhydrase, MN/CA IX, with potential value as a marker of cell proliferation. Am J Pathol 153(1): 279-85 Saarnio J, Parkkila S, Parkkila AK, Waheed A, Casey MC, Zhou XY, Pastorekova S, Pastorek J, Karttunen T, Haukipuro K, Kairaluoma MI, Sly WS (1998b) Immunohistochemistry of carbonic anhydrase isozyme IX (MN/CA IX) in human gut

152

reveals polarized expression in the epithelial cells with the highest proliferative capacity. The journal of histochemistry and cytochemistry : official journal of the Histochemistry Society 46(4): 497-504 Saed GM, Fletcher NM, Jiang ZL, Abu-Soud HM, Diamond MP (2011) Dichloroacetate induces apoptosis of epithelial ovarian cancer cells through a mechanism involving modulation of oxidative stress. Reprod Sci 18(12): 1253-61 Sale GJ, Randle PJ (1981) Analysis of site occupancies in [32P]phosphorylated pyruvate dehydrogenase complexes by aspartyl-prolyl cleavage of tryptic phosphopeptides. Eur J Biochem 120(3): 535-40 Sanborn BM, Felberg NT, Hollocher TC (1971) The inactivation of succinate dehydrogenase by bromopyruvate. Biochim Biophys Acta 227(2): 219-31 Sanchez WY, McGee SL, Connor T, Mottram B, Wilkinson A, Whitehead JP, Vuckovic S, Catley L (2013) Dichloroacetate inhibits aerobic glycolysis in multiple myeloma cells and increases sensitivity to bortezomib. Br J Cancer 108(8): 1624-33 Sanchez-Arago M, Chamorro M, Cuezva JM (2010) Selection of cancer cells with repressed mitochondria triggers colon cancer progression. Carcinogenesis 31(4): 567-76 Sanchez-Arago M, Cuezva JM (2011) The bioenergetic signature of isogenic colon cancer cells predicts the cell death response to treatment with 3-bromopyruvate, iodoacetate or 5-fluorouracil. Journal of translational medicine 9: 19 Santini MT, Rainaldi G (1999) Three-dimensional spheroid model in tumor biology. Pathobiology : journal of immunopathology, molecular and cellular biology 67(3): 148-57 Sarbassov DD, Guertin DA, Ali SM, Sabatini DM (2005) Phosphorylation and regulation of Akt/PKB by the rictor-mTOR complex. Science 307(5712): 1098-101 Sato-Tadano A, Suzuki T, Amari M, Takagi K, Miki Y, Tamaki K, Watanabe M, Ishida T, Sasano H, Ohuchi N (2013) Hexokinase II in breast carcinoma: a potent prognostic factor associated with hypoxia-inducible factor-1alpha and Ki-67. Cancer science 104(10): 1380-8 Schafer ZT, Grassian AR, Song L, Jiang Z, Gerhart-Hines Z, Irie HY, Gao S, Puigserver P, Brugge JS (2009) Antioxidant and oncogene rescue of metabolic defects caused by loss of matrix attachment. Nature 461(7260): 109-13 Scheid MP, Parsons M, Woodgett JR (2005) Phosphoinositide-dependent phosphorylation of PDK1 regulates nuclear translocation. Mol Cell Biol 25(6): 2347-63

153

Scheid MP, Woodgett JR (2001) PKB/AKT: functional insights from genetic models. Nat Rev Mol Cell Biol 2(10): 760-8 Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, Preibisch S, Rueden C, Saalfeld S, Schmid B, Tinevez JY, White DJ, Hartenstein V, Eliceiri K, Tomancak P, Cardona A (2012) Fiji: an open-source platform for biological-image analysis. Nature methods 9(7): 676-82 Scorrano L, Korsmeyer SJ (2003) Mechanisms of cytochrome c release by proapoptotic BCL-2 family members. Biochem Biophys Res Commun 304(3): 437-44 Seagroves TN, Ryan HE, Lu H, Wouters BG, Knapp M, Thibault P, Laderoute K, Johnson RS (2001) Transcription factor HIF-1 is a necessary mediator of the pasteur effect in mammalian cells. Mol Cell Biol 21(10): 3436-44 Sears R, Leone G, DeGregori J, Nevins JR (1999) Ras enhances Myc protein stability. Mol Cell 3(2): 169-79 Secomb TW, Hsu R, Dewhirst MW, Klitzman B, Gross JF (1993) Analysis of oxygen transport to tumor tissue by microvascular networks. Int J Radiat Oncol Biol Phys 25(3): 481-9 Seifert P, Baker LH, Reed ML, Vaitkevicius VK (1975) Comparison of continuously infused 5-fluorouracil with bolus injection in treatment of patients with colorectal adenocarcinoma. Cancer 36(1): 123-8 Selak MA, Armour SM, MacKenzie ED, Boulahbel H, Watson DG, Mansfield KD, Pan Y, Simon MC, Thompson CB, Gottlieb E (2005) Succinate links TCA cycle dysfunction to oncogenesis by inhibiting HIF-alpha prolyl hydroxylase. Cancer Cell 7(1): 77-85 Semenza GL (2003) Targeting HIF-1 for cancer therapy. Nat Rev Cancer 3(10): 721-32 Semenza GL (2006) Development of novel therapeutic strategies that target HIF-1. Expert Opin Ther Targets 10(2): 267-80 Semenza GL (2010) HIF-1: upstream and downstream of cancer metabolism. Curr Opin Genet Dev 20(1): 51-6 Semenza GL, Jiang BH, Leung SW, Passantino R, Concordet JP, Maire P, Giallongo A (1996) Hypoxia response elements in the aldolase A, enolase 1, and lactate dehydrogenase A gene promoters contain essential binding sites for hypoxia-inducible factor 1. J Biol Chem 271(51): 32529-37 Semenza GL, Wang GL (1992) A nuclear factor induced by hypoxia via de novo protein synthesis binds to the human erythropoietin gene enhancer at a site required for transcriptional activation. Mol Cell Biol 12(12): 5447-54

154

Shahrzad S, Lacombe K, Adamcic U, Minhas K, Coomber BL (2010) Sodium dichloroacetate (DCA) reduces apoptosis in colorectal tumor hypoxia. Cancer Lett 297(1): 75-83 Shahrzad S, Quayle L, Stone C, Plumb C, Shirasawa S, Rak JW, Coomber BL (2005) Ischemia-induced K-ras mutations in human colorectal cancer cells: role of microenvironmental regulation of MSH2 expression. Cancer Res 65(18): 8134-41 Shekhar MP, Pauley R, Heppner G (2003) Host microenvironment in breast cancer development: extracellular matrix-stromal cell contribution to neoplastic phenotype of epithelial cells in the breast. Breast cancer research : BCR 5(3): 130-5 Shen YC, Ou DL, Hsu C, Lin KL, Chang CY, Lin CY, Liu SH, Cheng AL (2013) Activating oxidative phosphorylation by a pyruvate dehydrogenase kinase inhibitor overcomes sorafenib resistance of hepatocellular carcinoma. Br J Cancer 108(1): 72-81 Shim H, Dolde C, Lewis BC, Wu CS, Dang G, Jungmann RA, Dalla-Favera R, Dang CV (1997) c-Myc transactivation of LDH-A: implications for tumor metabolism and growth. Proc Natl Acad Sci U S A 94(13): 6658-63 Siegel R, Desantis C, Jemal A (2014) Colorectal cancer statistics, 2014. CA: a cancer journal for clinicians 64(2): 104-17 Siegel RL, Miller KD, Jemal A (2015) Cancer statistics, 2015. CA: a cancer journal for clinicians 65(1): 5-29 Simonnet H, Alazard N, Pfeiffer K, Gallou C, Beroud C, Demont J, Bouvier R, Schagger H, Godinot C (2002) Low mitochondrial respiratory chain content correlates with tumor aggressiveness in renal cell carcinoma. Carcinogenesis 23(5): 759-68 Simons AL, Ahmad IM, Mattson DM, Dornfeld KJ, Spitz DR (2007) 2-Deoxy-D-glucose combined with cisplatin enhances cytotoxicity via metabolic oxidative stress in human head and neck cancer cells. Cancer Res 67(7): 3364-70 Skowronski K, Andrews J, Rodenhiser DI, Coomber BL (2014) Genome-wide analysis in human colorectal cancer cells reveals ischemia-mediated expression of motility genes via DNA hypomethylation. PLoS One 9(7): e103243 Skowronski K, Dubey S, Rodenhiser D, Coomber B (2010) Ischemia dysregulates DNA methyltransferases and p16INK4a methylation in human colorectal cancer cells. Epigenetics : official journal of the DNA Methylation Society 5(6): 547-56 Smolkova K, Bellance N, Scandurra F, Genot E, Gnaiger E, Plecita-Hlavata L, Jezek P, Rossignol R (2010) Mitochondrial bioenergetic adaptations of breast cancer cells to aglycemia and hypoxia. J Bioenerg Biomembr 42(1): 55-67

155

Sonveaux P, Vegran F, Schroeder T, Wergin MC, Verrax J, Rabbani ZN, De Saedeleer CJ, Kennedy KM, Diepart C, Jordan BF, Kelley MJ, Gallez B, Wahl ML, Feron O, Dewhirst MW (2008) Targeting lactate-fueled respiration selectively kills hypoxic tumor cells in mice. J Clin Invest 118(12): 3930-42 Sridharan M, Hubbard JM, Grothey A (2014) Colorectal cancer: how emerging molecular understanding affects treatment decisions. Oncology 28(2): 110-8 Srinivas V, Leshchinsky I, Sang N, King MP, Minchenko A, Caro J (2001) Oxygen sensing and HIF-1 activation does not require an active mitochondrial respiratory chain electron-transfer pathway. J Biol Chem 276(25): 21995-8 Stacpoole PW (1989) The pharmacology of dichloroacetate. Metabolism 38(11): 1124-44 Stacpoole PW, Felts JM (1970) Diisopropylammonium dichloroacetate (DIPA) and sodium dichloracetate (DCA): effect on glucose and fat metabolism in normal and diabetic tissue. Metabolism 19(1): 71-8 Stacpoole PW, Henderson GN, Yan Z, James MO (1998) Clinical pharmacology and toxicology of dichloroacetate. Environ Health Perspect 106 Suppl 4: 989-94 Stacpoole PW, Kerr DS, Barnes C, Bunch ST, Carney PR, Fennell EM, Felitsyn NM, Gilmore RL, Greer M, Henderson GN, Hutson AD, Neiberger RE, O'Brien RG, Perkins LA, Quisling RG, Shroads AL, Shuster JJ, Silverstein JH, Theriaque DW, Valenstein E (2006) Controlled clinical trial of dichloroacetate for treatment of congenital lactic acidosis in children. Pediatrics 117(5): 1519-31 Stacpoole PW, Kurtz TL, Han Z, Langaee T (2008) Role of dichloroacetate in the treatment of genetic mitochondrial diseases. Adv Drug Deliv Rev 60(13-14): 1478-87 Stacpoole PW, Lorenz AC, Thomas RG, Harman EM (1988) Dichloroacetate in the treatment of lactic acidosis. Annals of internal medicine 108(1): 58-63 Stacpoole PW, Nagaraja NV, Hutson AD (2003) Efficacy of dichloroacetate as a lactate-lowering drug. J Clin Pharmacol 43(7): 683-91 Stacpoole PW, Wright EC, Baumgartner TG, Bersin RM, Buchalter S, Curry SH, Duncan CA, Harman EM, Henderson GN, Jenkinson S, et al. (1992) A controlled clinical trial of dichloroacetate for treatment of lactic acidosis in adults. The Dichloroacetate-Lactic Acidosis Study Group. N Engl J Med 327(22): 1564-9 Stambolic V, Suzuki A, de la Pompa JL, Brothers GM, Mirtsos C, Sasaki T, Ruland J, Penninger JM, Siderovski DP, Mak TW (1998) Negative regulation of PKB/Akt-dependent cell survival by the tumor suppressor PTEN. Cell 95(1): 29-39

156

Stiehl DP, Wirthner R, Koditz J, Spielmann P, Camenisch G, Wenger RH (2006) Increased prolyl 4-hydroxylase domain proteins compensate for decreased oxygen levels. Evidence for an autoregulatory oxygen-sensing system. J Biol Chem 281(33): 23482-91 Stockwin LH, Yu SX, Borgel S, Hancock C, Wolfe TL, Phillips LR, Hollingshead MG, Newton DL (2010) Sodium dichloroacetate selectively targets cells with defects in the mitochondrial ETC. Int J Cancer 127(11): 2510-9 Strese S, Fryknas M, Larsson R, Gullbo J (2013) Effects of hypoxia on human cancer cell line chemosensitivity. BMC cancer 13: 331 Su YQ, Sugiura K, Eppig JJ (2009) Mouse oocyte control of granulosa cell development and function: paracrine regulation of cumulus cell metabolism. Seminars in reproductive medicine 27(1): 32-42 Subarsky P, Hill RP (2003) The hypoxic tumour microenvironment and metastatic progression. Clinical & experimental metastasis 20(3): 237-50 Sugiura K, Su YQ, Diaz FJ, Pangas SA, Sharma S, Wigglesworth K, O'Brien MJ, Matzuk MM, Shimasaki S, Eppig JJ (2007) Oocyte-derived BMP15 and FGFs cooperate to promote glycolysis in cumulus cells. Development 134(14): 2593-603 Sun RC, Fadia M, Dahlstrom JE, Parish CR, Board PG, Blackburn AC (2010) Reversal of the glycolytic phenotype by dichloroacetate inhibits metastatic breast cancer cell growth in vitro and in vivo. Breast Cancer Res Treat 120(1): 253-60 Sussman MA, Volkers M, Fischer K, Bailey B, Cottage CT, Din S, Gude N, Avitabile D, Alvarez R, Sundararaman B, Quijada P, Mason M, Konstandin MH, Malhowski A, Cheng Z, Khan M, McGregor M (2011) Myocardial AKT: the omnipresent nexus. Physiological reviews 91(3): 1023-70 Sutendra G, Dromparis P, Kinnaird A, Stenson TH, Haromy A, Parker JM, McMurtry MS, Michelakis ED (2013) Mitochondrial activation by inhibition of PDKII suppresses HIF1a signaling and angiogenesis in cancer. Oncogene 32(13): 1638-50 Swinson DE, Jones JL, Cox G, Richardson D, Harris AL, O'Byrne KJ (2004) Hypoxia-inducible factor-1 alpha in non small cell lung cancer: relation to growth factor, protease and apoptosis pathways. Int J Cancer 111(1): 43-50 Tanaka M, Nakamura F, Mizokawa S, Matsumura A, Matsumura K, Murata T, Shigematsu M, Kageyama K, Ochi H, Watanabe Y (2004) Role of lactate in the brain energy metabolism: revealed by Bioradiography. Neuroscience research 48(1): 13-20 Tang Z, Yuan S, Hu Y, Zhang H, Wu W, Zeng Z, Yang J, Yun J, Xu R, Huang P (2012) Over-expression of GAPDH in human colorectal carcinoma as a preferred target of 3-bromopyruvate propyl ester. J Bioenerg Biomembr 44(1): 117-25

157

Terho P, Korkeamaki M (2008) Cyflogic Thomlinson RH, Gray LH (1955) The histological structure of some human lung cancers and the possible implications for radiotherapy. Br J Cancer 9(4): 539-49 Thompson JE, Thompson CB (2004) Putting the rap on Akt. Journal of clinical oncology : official journal of the American Society of Clinical Oncology 22(20): 4217-26 Thorn CC, Freeman TC, Scott N, Guillou PJ, Jayne DG (2009) Laser microdissection expression profiling of marginal edges of colorectal tumours reveals evidence of increased lactate metabolism in the aggressive phenotype. Gut 58(3): 404-12 Tili E, Michaille JJ, Luo Z, Volinia S, Rassenti LZ, Kipps TJ, Croce CM (2012) The down-regulation of miR-125b in chronic lymphocytic leukemias leads to metabolic adaptation of cells to a transformed state. Blood 120(13): 2631-8 Tomlinson IP, Alam NA, Rowan AJ, Barclay E, Jaeger EE, Kelsell D, Leigh I, Gorman P, Lamlum H, Rahman S, Roylance RR, Olpin S, Bevan S, Barker K, Hearle N, Houlston RS, Kiuru M, Lehtonen R, Karhu A, Vilkki S, Laiho P, Eklund C, Vierimaa O, Aittomaki K, Hietala M, Sistonen P, Paetau A, Salovaara R, Herva R, Launonen V, Aaltonen LA, Multiple Leiomyoma C (2002) Germline mutations in FH predispose to dominantly inherited uterine fibroids, skin leiomyomata and papillary renal cell cancer. Nature genetics 30(4): 406-10 Torre LA, Bray F, Siegel RL, Ferlay J, Lortet-Tieulent J, Jemal A (2015) Global cancer statistics, 2012. CA: a cancer journal for clinicians 65(2): 87-108 Trachootham D, Alexandre J, Huang P (2009) Targeting cancer cells by ROS-mediated mechanisms: a radical therapeutic approach? Nature reviews Drug discovery 8(7): 579-91 Trastour C, Benizri E, Ettore F, Ramaioli A, Chamorey E, Pouyssegur J, Berra E (2007) HIF-1alpha and CA IX staining in invasive breast carcinomas: prognosis and treatment outcome. Int J Cancer 120(7): 1451-8 Tsacopoulos M, Magistretti PJ (1996) Metabolic coupling between glia and neurons. The Journal of neuroscience : the official journal of the Society for Neuroscience 16(3): 877-85 U.S. National Institutes of Health (2015) ClinicalTrials.gov Vander Heiden MG, Cantley LC, Thompson CB (2009) Understanding the Warburg effect: the metabolic requirements of cell proliferation. Science 324(5930): 1029-33

158

Vasudevan S, Tong Y, Steitz JA (2007) Switching from repression to activation: microRNAs can up-regulate translation. Science 318(5858): 1931-4 Vaupel P, Kallinowski F, Okunieff P (1989) Blood flow, oxygen and nutrient supply, and metabolic microenvironment of human tumors: a review. Cancer Res 49(23): 6449-65 Vleugel MM, Greijer AE, Shvarts A, van der Groep P, van Berkel M, Aarbodem Y, van Tinteren H, Harris AL, van Diest PJ, van der Wall E (2005) Differential prognostic impact of hypoxia induced and diffuse HIF-1alpha expression in invasive breast cancer. Journal of clinical pathology 58(2): 172-7 Vousden KH, Ryan KM (2009) p53 and metabolism. Nat Rev Cancer 9(10): 691-700 Wahdan-Alaswad R, Fan Z, Edgerton SM, Liu B, Deng XS, Arnadottir SS, Richer JK, Anderson SM, Thor AD (2013) Glucose promotes breast cancer aggression and reduces metformin efficacy. Cell cycle 12(24): 3759-69 Wambolt RB, Lopaschuk GD, Brownsey RW, Allard MF (2000) Dichloroacetate improves postischemic function of hypertrophied rat hearts. Journal of the American College of Cardiology 36(4): 1378-85 Warburg O, Wind F, Negelein E (1927) The Metabolism of Tumors in the Body. J Gen Physiol 8(6): 519-30 Ward PS, Patel J, Wise DR, Abdel-Wahab O, Bennett BD, Coller HA, Cross JR, Fantin VR, Hedvat CV, Perl AE, Rabinowitz JD, Carroll M, Su SM, Sharp KA, Levine RL, Thompson CB (2010) The common feature of leukemia-associated IDH1 and IDH2 mutations is a neomorphic enzyme activity converting alpha-ketoglutarate to 2-hydroxyglutarate. Cancer Cell 17(3): 225-34 Ward PS, Thompson CB (2012) Metabolic reprogramming: a cancer hallmark even warburg did not anticipate. Cancer Cell 21(3): 297-308 Wargovich TJ, MacDonald RG, Hill JA, Feldman RL, Stacpoole PW, Pepine CJ (1988) Myocardial metabolic and hemodynamic effects of dichloroacetate in coronary artery disease. Am J Cardiol 61(1): 65-70 Wartenberg M, Ling FC, Muschen M, Klein F, Acker H, Gassmann M, Petrat K, Putz V, Hescheler J, Sauer H (2003) Regulation of the multidrug resistance transporter P-glycoprotein in multicellular tumor spheroids by hypoxia-inducible factor (HIF-1) and reactive oxygen species. Faseb J 17(3): 503-5 Wellen KE, Lu C, Mancuso A, Lemons JM, Ryczko M, Dennis JW, Rabinowitz JD, Coller HA, Thompson CB (2010) The hexosamine biosynthetic pathway couples growth factor-induced glutamine uptake to glucose metabolism. Genes Dev 24(24): 2784-99

159

Whitaker-Menezes D, Martinez-Outschoorn UE, Lin Z, Ertel A, Flomenberg N, Witkiewicz AK, Birbe RC, Howell A, Pavlides S, Gandara R, Pestell RG, Sotgia F, Philp NJ, Lisanti MP (2011) Evidence for a stromal-epithelial "lactate shuttle" in human tumors: MCT4 is a marker of oxidative stress in cancer-associated fibroblasts. Cell cycle 10(11): 1772-83 Wigfield SM, Winter SC, Giatromanolaki A, Taylor J, Koukourakis ML, Harris AL (2008) PDK-1 regulates lactate production in hypoxia and is associated with poor prognosis in head and neck squamous cancer. Br J Cancer 98(12): 1975-84 Wilfred BR, Wang WX, Nelson PT (2007) Energizing miRNA research: a review of the role of miRNAs in lipid metabolism, with a prediction that miR-103/107 regulates human metabolic pathways. Mol Genet Metab 91(3): 209-17 Wilson JE (1995) Hexokinases. Reviews of physiology, biochemistry and pharmacology 126: 65-198 Wilson JE (1997) An introduction to the isoenzymes of mammalian hexokinase types I-III. Biochem Soc Trans 25(1): 103-7 Wilson JE (2003) Isozymes of mammalian hexokinase: structure, subcellular localization and metabolic function. The Journal of experimental biology 206(Pt 12): 2049-57 Wilson PM, Labonte MJ, Lenz HJ (2010) Molecular markers in the treatment of metastatic colorectal cancer. Cancer journal 16(3): 262-72 Wintzell M, Lofstedt L, Johansson J, Pedersen AB, Fuxe J, Shoshan M (2012) Repeated cisplatin treatment can lead to a multiresistant tumor cell population with stem cell features and sensitivity to 3-bromopyruvate. Cancer biology & therapy 13(14): 1454-62 Wise DR, DeBerardinis RJ, Mancuso A, Sayed N, Zhang XY, Pfeiffer HK, Nissim I, Daikhin E, Yudkoff M, McMahon SB, Thompson CB (2008) Myc regulates a transcriptional program that stimulates mitochondrial glutaminolysis and leads to glutamine addiction. Proc Natl Acad Sci U S A 105(48): 18782-7 Wolf A, Agnihotri S, Micallef J, Mukherjee J, Sabha N, Cairns R, Hawkins C, Guha A (2011) Hexokinase 2 is a key mediator of aerobic glycolysis and promotes tumor growth in human glioblastoma multiforme. The Journal of experimental medicine 208(2): 313-26 Wollen N, Bailey CJ (1988) Inhibition of hepatic gluconeogenesis by metformin. Synergism with insulin. Biochem Pharmacol 37(22): 4353-8 Wong JY, Huggins GS, Debidda M, Munshi NC, De Vivo I (2008) Dichloroacetate induces apoptosis in endometrial cancer cells. Gynecologic oncology 109(3): 394-402

160

Wong KK, Engelman JA, Cantley LC (2010) Targeting the PI3K signaling pathway in cancer. Curr Opin Genet Dev 20(1): 87-90 Wu M, Jung L, Cooper AB, Fleet C, Chen L, Breault L, Clark K, Cai Z, Vincent S, Bottega S, Shen Q, Richardson A, Bosenburg M, Naber SP, DePinho RA, Kuperwasser C, Robinson MO (2009) Dissecting genetic requirements of human breast tumorigenesis in a tissue transgenic model of human breast cancer in mice. Proc Natl Acad Sci U S A 106(17): 7022-7 Wykoff CC, Beasley N, Watson PH, Campo L, Chia SK, English R, Pastorek J, Sly WS, Ratcliffe P, Harris AL (2001) Expression of the hypoxia-inducible and tumor-associated carbonic anhydrases in ductal carcinoma in situ of the breast. Am J Pathol 158(3): 1011-9 Xu RH, Pelicano H, Zhou Y, Carew JS, Feng L, Bhalla KN, Keating MJ, Huang P (2005) Inhibition of glycolysis in cancer cells: a novel strategy to overcome drug resistance associated with mitochondrial respiratory defect and hypoxia. Cancer Res 65(2): 613-21 Yan H, Parsons DW, Jin G, McLendon R, Rasheed BA, Yuan W, Kos I, Batinic-Haberle I, Jones S, Riggins GJ, Friedman H, Friedman A, Reardon D, Herndon J, Kinzler KW, Velculescu VE, Vogelstein B, Bigner DD (2009) IDH1 and IDH2 mutations in gliomas. N Engl J Med 360(8): 765-73 Yang L, Moss T, Mangala LS, Marini J, Zhao H, Wahlig S, Armaiz-Pena G, Jiang D, Achreja A, Win J, Roopaimoole R, Rodriguez-Aguayo C, Mercado-Uribe I, Lopez-Berestein G, Liu J, Tsukamoto T, Sood AK, Ram PT, Nagrath D (2014) Metabolic shifts toward glutamine regulate tumor growth, invasion and bioenergetics in ovarian cancer. Molecular systems biology 10: 728 Yasuda S, Arii S, Mori A, Isobe N, Yang W, Oe H, Fujimoto A, Yonenaga Y, Sakashita H, Imamura M (2004) Hexokinase II and VEGF expression in liver tumors: correlation with hypoxia-inducible factor 1 alpha and its significance. Journal of hepatology 40(1): 117-23 Yeluri S, Madhok BM, Jayne DG (2012) New horizons in cancer therapy: manipulating tumour metabolism. Biomedical Research 23: 17-23 Youle RJ, Strasser A (2008) The BCL-2 protein family: opposing activities that mediate cell death. Nat Rev Mol Cell Biol 9(1): 47-59 Yu JL, Rak JW, Carmeliet P, Nagy A, Kerbel RS, Coomber BL (2001) Heterogeneous vascular dependence of tumor cell populations. Am J Pathol 158(4): 1325-34 Yu SJ, Yoon JH, Yang JI, Cho EJ, Kwak MS, Jang ES, Lee JH, Kim YJ, Lee HS, Kim CY (2012) Enhancement of hexokinase II inhibitor-induced apoptosis in hepatocellular carcinoma cells via augmenting ER stress and anti-angiogenesis by protein disulfide isomerase inhibition. J Bioenerg Biomembr 44(1): 101-15

161

Yuan S, Fu Y, Wang X, Shi H, Huang Y, Song X, Li L, Song N, Luo Y (2008) Voltage-dependent anion channel 1 is involved in endostatin-induced endothelial cell apoptosis. Faseb J 22(8): 2809-20 Yun J, Rago C, Cheong I, Pagliarini R, Angenendt P, Rajagopalan H, Schmidt K, Willson JK, Markowitz S, Zhou S, Diaz LA, Jr., Velculescu VE, Lengauer C, Kinzler KW, Vogelstein B, Papadopoulos N (2009) Glucose deprivation contributes to the development of KRAS pathway mutations in tumor cells. Science 325(5947): 1555-9 Zhang P, Li H, Tan X, Chen L, Wang S (2013) Association of metformin use with cancer incidence and mortality: a meta-analysis. Cancer epidemiology 37(3): 207-18 Zhao S, Lin Y, Xu W, Jiang W, Zha Z, Wang P, Yu W, Li Z, Gong L, Peng Y, Ding J, Lei Q, Guan KL, Xiong Y (2009) Glioma-derived mutations in IDH1 dominantly inhibit IDH1 catalytic activity and induce HIF-1alpha. Science 324(5924): 261-5 Zhao S, Liu H, Liu Y, Wu J, Wang C, Hou X, Chen X, Yang G, Zhao L, Che H, Bi Y, Wang H, Peng F, Ai J (2013) miR-143 inhibits glycolysis and depletes stemness of glioblastoma stem-like cells. Cancer Lett 333(2): 253-60 Zhao Z, Han FH, Yang SB, Hua LX, Wu JH, Zhan WH (2015) Loss of stromal caveolin-1 expression in colorectal cancer predicts poor survival. World J Gastroenterol 21(4): 1140-7 Zhivotovsky B, Samali A, Orrenius S (2001) Determination of apoptosis and necrosis. Current protocols in toxicology / editorial board, Mahin D Maines Chapter 2: Unit 2 2 Zhong D, Liu X, Schafer-Hales K, Marcus AI, Khuri FR, Sun SY, Zhou W (2008) 2-Deoxyglucose induces Akt phosphorylation via a mechanism independent of LKB1/AMP-activated protein kinase signaling activation or glycolysis inhibition. Mol Cancer Ther 7(4): 809-17 Zhou CF, Li XB, Sun H, Zhang B, Han YS, Jiang Y, Zhuang QL, Fang J, Wu GH (2012) Pyruvate kinase type M2 is upregulated in colorectal cancer and promotes proliferation and migration of colon cancer cells. IUBMB Life 64(9): 775-82 Zhou G, Myers R, Li Y, Chen Y, Shen X, Fenyk-Melody J, Wu M, Ventre J, Doebber T, Fujii N, Musi N, Hirshman MF, Goodyear LJ, Moller DE (2001) Role of AMP-activated protein kinase in mechanism of metformin action. J Clin Invest 108(8): 1167-74 Zu XL, Guppy M (2004) Cancer metabolism: facts, fantasy, and fiction. Biochem Biophys Res Commun 313(3): 459-65

162

Zundel W, Schindler C, Haas-Kogan D, Koong A, Kaper F, Chen E, Gottschalk AR, Ryan HE, Johnson RS, Jefferson AB, Stokoe D, Giaccia AJ (2000) Loss of PTEN facilitates HIF-1-mediated gene expression. Genes Dev 14(4): 391-6

163

APPENDIX I – CHEMICAL LIST AND SUPPLIERS

Chemical Supplier 3BP Sigma-Aldrich, St. Louis, MO Acetic acid, glacial Fisher Scientific, Nepean, ON Acrylamide/Bis solution, 40% Bio-Rad Laboratories, Hercules, CA Agarose Life Technologies, Carlsbad, CA α-Tubulin mouse mAb; T5168 Sigma-Aldrich, St. Louis, MO Amido Black Bio-Rad Laboratories, Hercules, CA Ammonium persulfate Sigma-Aldrich, St. Louis, MO Amplex Red Glucose Assay Life Technologies, Carlsbad, CA Alexa Fluor® 488 Annexin V/Dead Cell Life Technologies, Carlsbad, CA Apoptosis Kit Aprotinin Sigma-Aldrich, St. Louis, MO pan-AKT rabbit mAb; 4691 Cell Signaling Technology, Danvers, MA pAKT (S473) rabbit pAb; 9271 Cell Signaling Technology, Danvers, MA pAKT (T308) rabbit mAb; 4056 Cell Signaling Technology, Danvers, MA Aurum Total RNA Fatty and Fibrous Kit Bio-Rad Laboratories, Hercules, CA BSA AMRESCO, Solon, OH CAIX mouse mAb; ab107257 Abcam plc, Cambridge, UK Caspase-3 rabbit pAb; 9662S Cell Signaling Technology, Danvers, MA Cell lysis buffer, 10X Cell Signaling Technology, Danvers, MA Crystal violet Fisher Scientific, Nepean, ON DC protein assay Bio-Rad Laboratories, Hercules, CA DCA Sigma-Aldrich, St. Louis, MO DMEM Sigma-Aldrich, St. Louis, MO DNA Ladder, 50 Kb SibEnzyme, Russia Etoposide Sigma-Aldrich, St. Louis, MO FBS Life Technologies, Carlsbad, CA Fluorescence Mounting Media Dako, Denmark Gentamicin Sigma-Aldrich, St. Louis, MO Glycine Fisher Scientific, Nepean, ON Goat anti-mouse Cy3 Ab Life Technologies, Carlsbad, CA Goat anti-mouse HRP Ab; A9917 Sigma-Aldrich, St. Louis, MO Goat anti-rabbit Alexa 488 Ab Life Technologies, Carlsbad, CA Goat anti-rabbit HRP Ab; A0545 Sigma-Aldrich, St. Louis, MO Goat serum Sigma-Aldrich, St. Louis, MO HKII rabbit mAb; 2867 Cell Signaling Technology, Danvers, MA Hydrogen peroxide, 30% Fisher Scientific, Nepean, ON INTERFERin Polyplus-transfections, France iScript Reverse Transcription Supermix Bio-Rad Laboratories, Hercules, CA Kplus DNA Ladder RTU FroggaBio, Toronto, ON L-Lactate Assay Kit I Eton Bioscience, San Diego, CA Luminata Forte Millipore, Billerica, MA M3:D culture media INCELL, San Antonia, TX Methanol, 100% Fisher Scientific, Nepean, ON

164

MitoTracker Green FM Life Technologies, Carlsbad, CA MitoTracker Orange CM-H2TMRos Life Technologies, Carlsbad, CA Opti-MEM Life Technologies, Carlsbad, CA PBS Sigma-Aldrich, St. Louis, MO PDH mouse mAb; ab110334 Abcam plc, Cambridge, UK PDK1 rabbit pAb; 3062 Cell Signaling Technology, Danvers, MA PDK1 rabbit pAb; NB100-2384 Novus Biologicals, Littleton, CO PDK2 rabbit mAb; NBP1-44428 Novus Biologicals, Littleton, CO PDK3 rabbit pAb; NBP1-32581 Novus Biologicals, Littleton, CO PDK4 rabbit pAb; NBP1-07047 Novus Biologicals, Littleton, CO pPDH (S293) rabbit pAb; NB110-93479 Novus Biologicals, Littleton, CO pPDH (S300) rabbit pAb; AP1064 Millipore, Billerica, MA pPDK1 (S241) rabbit mAb; 3438 Cell Signaling Technology, Danvers, MA Phosphatase Inhibitor Cocktail 2 Sigma-Aldrich, St. Louis, MO PMSF Sigma-Aldrich, St. Louis, MO Protein block solution Dako, Denmark PTEN rabbit mAb; 9188 Cell Signaling Technology, Danvers, MA PVDF membrane Roche, Laval, QC ReBlot Plus (Mild) Millipore, Billerica, CA RedSafeTM Nucleic Acid Staining Solution FroggaBio, Toronto, ON RiboZol AMRESCO, Solon, OH Scrambled siRNA Sigma-Aldrich, St. Louis, MO SDS, 20% Fisher Scientific, Nepean, ON Sodium chloride Fisher Scientific, Nepean, ON Sodium orthovanadate Fisher Scientific, Nepean, ON Sodium pyruvate Sigma-Aldrich, St. Louis, MO SsoFast EvaGreen Supermix Bio-Rad Laboratories, Hercules, CA TAE Buffer, 50X Stock Eppendorf, Hamburg, Germany TEMED Sigma-Aldrich, St. Louis, MO Tissue DNA Kit Omega Bio-Tek, Norcross, GA Tris base, powder Fisher Scientific, Nepean, ON Tris-HCl buffer, 0.5 M, pH 6.8 Bio-Rad Laboratories, Hercules, CA Tris-HCl buffer, 1.5 M, pH 8.8 Bio-Rad Laboratories, Hercules, CA Trypan blue solution, 4% Sigma-Aldrich, St. Louis, MO Tween-20 Fisher Scientific, Nepean, ON Xylene Fisher Scientific, Nepean, ON

165

APPENDIX II – PREPARATION OF MATERIALS

Acrylamide SDS-PAGE Gels (2 x 1.5 mm) Resolving gels of various % acrylamide were prepared as follows:

Acrylamide 7.5% 10% 12% 15% Distilled water 11 ml 9.7 ml 8.7 ml 7.2 ml 40% acrylamide/bis solution 3.72 ml 5 ml 6 ml 7.5 ml 1.5 M Tris-HCl buffer, pH 8.8 5 ml 5 ml 5 ml 5 ml 10% SDS 200 µl 200 µl 200 µl 200 µl 10% APS* 240 µl 240 µl 240 µl 240 µl TEMED 10 µl 10 µl 10 µl 10 µl

*1 g APS dissolved in 10 ml distilled water Once resolving gel was polymerized, stacking gel was prepared as follows:

Distilled water 3.2 ml 40% acrylamide/bis solution 500 µl 0.5 M Tris-HCl buffer, pH 6.8 1.26 ml 10% SDS 50 µl 10% APS 50 µl TEMED 5 µl

Plastic combs (1.5 mm) were added to the gels following addition of stacking gel. Agarose Gels (1.5%) 1.5 g of agarose/100 ml of 1X TAE buffer was added to a flask. Solution was heated in a microwave until completely dissolved and allowed to slightly cool. Once cooled, 5 µl of SafeRed was added, and te prepared solution was poured into a casting apparatus. Amido Black Stain Amido black 10B 0.5 g Glacial acetic acid 50 ml Methanol 200 ml H2O 250 ml Mix all components and store at RT. Citrate Buffer Stock Solution A (0.1 M Citric acid): 1.92 g in 100 ml H2O Stock Solution B (0.1 M Sodium citrate dihydrate): 14.7 g in 500 ml H2O Mix 9 ml of Stock Solution A with 41 ml of Stock Solution B. Top up with 450 ml of H2O and adjust pH to 6.0 with HCl. Store at 4oC

166

Crystal Violet Stain Crystal violet 5 g Methanol 100 ml H2O 400 ml Mix all components and store at RT. Electrophoresis Buffer 10X Tris base 30.25 g Glycine 144.1 g 20% SDS 50 ml H2O 600 mL Mix all components and adjust pH to 8.3 with HCl. Adjust final volume to 1 L. Store at RT. PBS-T 10X PBS 100 ml H2O 900 ml Tween-20 1 ml Mix all components and store at RT. Protein Loading Buffer 8X 400 mM Tris-HCl (pH 6.8) 16% SDS 0.8% Bromophenol blue 40% Glycerol 0.4 M DTT Store at -20oC. Once thawed, store in the dark at RT. TBS 10X Tris base 24.2 g NaCl 80 g H2O 600 ml Mix all components and adjust pH to 7.6 with HCl. Adjust final volume to 1 L. Store at RT. TBS-T 10X TBS 100 ml H2O 900 ml Tween-20 1 ml Mix all components and store at RT. Towbin Solution Tris base 30.25 g Glycine 141.1 g Dissolve in H2O to a final volume of 1 L. Store at RT.

167

Transfer Buffer (Semi-dry) Tris base 5.82 g Glycine 2.93 g 10% SDS 3.75 ml Methanol 200 ml Dissolve in H2O to a final volume of 1 L. Store at RT. Transfer Buffer (Wet) H2O 770 ml Towbin solution 80 ml 20% SDS 200 µl Methanol 150 ml Mix all components and store at 4oC.

168

APPENDIX III – AV/PI STAINING OF ETOPOSIDE-TREATED SAMPLES

Ctrl 50 µM Etoposide

Annexin V

Propidium Iodide

HC

T116

C

aCo2

SW

480

DLD

-1

169

APPENDIX IV – DENSITOMETRY OF AKT PHOSPHORYLATION AT DIFFERENT RESIDUES FOLLOWING 4 H 3BP EXPOSURE

Western blot analysis was used to examine AKT phosphorylation following 4 h 3BP treatment. Graphical outputs for densitometric analysis from three biological replicates of pAKT normalized to native AKT at residues Thr308 and Ser473 are shown above. Representative blots are shown in Figure 10B and C. (* p ≤ 0.05 vs. control (0 µM 3BP))

170

APPENDIX V – FLOW CYTOMETRY ANALYSIS OF AV AND PI STAINING FOLLOWING 48 H 3BP EXPOSURE

Quantification of AV and PI staining following 48 h 3BP treatment from three biological replicates ± SEM. As dead cells stained mostly in the top right quadrant (AV+/PI+), the trends observed when assessing AV and PI staining separately are similar to its combined analysis (e.g. see Figure 8B). These results therefore do not clarify the mechanism of 3BP-induced cell death in CRC cells. (* p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001, **** p ≤ 0.0001)

HCT116 CaCo2 SW480 DLD-10

20

40

60

80

100

Cell Line

% A

V +

0 µM5 µM30 µM50 µM

****

*

**

****

[3BP]

HCT116 CaCo2 SW480 DLD-10

20

40

60

80

100

Cell Line

% P

I +

0 µM5 µM30 µM50 µM

****

*

*

**

**

[3BP]

Targeting energy metabolism in colorectal cancer

Documents

Transcript of Targeting energy metabolism in colorectal cancer