Cellular Respiration and Carcinogenesis

Cellular Respiration and Carcinogenesis Shireesh P. Apte Rangaprasad Sarangarajan • Editors Cellular Respiration and...

Author: Shireesh Apte | Rangaprasad Sarangarajan

175 downloads 939 Views 5MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

Cellular Respiration and Carcinogenesis

Shireesh P. Apte Rangaprasad Sarangarajan •

Editors

Cellular Respiration and Carcinogenesis

Foreword by Gregg L. Semenza

123

Editors Shireesh P. Apte Alcon Laboratories Fort Worth, TX USA [email protected]

ISBN: 978-1-934115-07-7 DOI 10.1007/978-1-59745-435-3

Rangaprasad Sarangarajan Massachusetts College of Pharmacy and Health Sciences Worcester, MA USA [email protected]

e-ISBN: 978-1-59745-435-3

Library of Congress Control Number: 2008937240 c Humana Press, a part of Springer Science+Business Media, LLC 2009 All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Humana Press, c/o Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. Printed on acid-free paper springer.com

Foreword

Cancer is remarkably heterogeneous when one considers pathogenesis of the disease in affected individuals with the same diagnosis (such as “renal carcinoma”) or even when cancer in a single individual is considered over time and space. This heterogeneity is the single greatest obstacle to effective cancer therapy. As a result of disease heterogeneity, there are not too many universal truths in cancer biology. One of the most consistent findings, though, is the dramatic increase in glucose uptake by metastatic cancer cells compared with that of their normal counterparts. This is such a reliable characteristic that it is used clinically to identify occult disease by positron emission tomography after injection of [18 F]fluoro-2-deoxy-D-glucose. Increased glucose uptake reflects an increased rate of glycolysis, with conversion of glucose to lactate and decreased conversion of pyruvate to acetyl coenzyme A, the substrate for mitochondrial oxidative phosphorylation. Because of the relative inefficiency of glycolysis compared with that of oxidative phosphorylation as a means of generating ATP, flux through the glycolytic pathway must increase dramatically to maintain cellular energetics. Because it is observed in more than 95% of advanced cancers, understanding the mechanisms and consequences of this dramatic reprogramming of glucose and energy metabolism in cancer cells is an important challenge for contemporary cancer biology. The 12 chapters of this book provide the reader with state-of-the-art summaries of recent exciting progress that has been made in the field of cancer metabolism. This story starts at the beginning of the 20th century with the work of Otto Warburg, whom Hans Krebs described as a “Cell Physiologist, Biochemist, and Eccentric.” Warburg invented a device, now known as the Warburg manometer, with which he demonstrated that tumor cells consume less O2 (and produce more lactate) than do normal cells under the same ambient O2 concentrations. A century later, the struggle to understand how and why metastatic cancer cells manifest the Warburg effect is still ongoing, and 12 rounds of this heavyweight fight await the reader beyond this brief introduction. Why do cancer cells manifest altered energy metabolism? To answer this question, it is necessary to acknowledge that mitochondria, in addition to serving as the major source of ATP production, provide two other critical cellular functions. One is that they also serve as the major source of reactive oxygen species (ROS), which are generated when electrons donated to the respiratory chain are not successfully v

vi

Foreword

transferred from complex IV (cytochrome c oxidase) to O2 to form water but instead react with O2 prematurely to form superoxide anion, which is then converted to hydrogen peroxide through the action of mitochondrial superoxide dismutase. ROS are used by normal cells as signaling molecules (e.g., during cell proliferation). However, if generated in excess (amplitude or duration or both), ROS result in the oxidation of lipids, proteins, and nucleic acids leading to cell dysfunction or death, which introduces the other major role of mitochondria as arbiters of cell survival, as apoptosis is triggered through changes in mitochondrial structure and function. Thus, it appears that reduced mitochondrial respiration, which often is accompanied by reduced mitochondrial mass, occurs in cancer cells as a result of selection against excess ROS production and/or apoptosis. How do cancer cells reprogram energy metabolism? To answer this question (and to more completely answer “why?” beyond the pat generalization provided above), it is necessary to return full circle to the issue of heterogeneity. Because mitochondria play critical roles in energy production, redox balance, proliferation, and survival, many mutations that are selected for in advanced cancers impact mitochondrial function in one or more ways, for although mitochondria control cell fate, the cell controls mitochondria by virtue of the fact that most of the components of mitochondria are the products of nuclear, not mitochondrial, genes. A danger of a reductionist approach to cancer biology that focuses on changes involving key oncogenes and tumor suppressor genes is the misconception that carcinogenesis can be attributed to one or a few mutations in any given cell. As in every other aspect of cancer biology, it is the net effect of all of the mutations in nuclear and mitochondrial DNA of the cancer cell that will determine its metabolic phenotype. Another danger of a solely mutation-based approach to cancer is that it ignores the effect of the tumor microenvironment. Like normal cells, cancer cells react to changes in the concentrations of O2 , glucose, hydrogen ions, and free radicals. In fact, these changes represent the most important selective force acting on the mutant gene pool of cancer cells. As a result, not all changes in cancer metabolism are hard-wired by genetic alteration but may instead represent adaptive responses to the microenvironment. In other words, the reprogramming of energy metabolism is unique to each cancer cell. The complex considerations that are covered so superficially in this brief introduction are discussed in detail in the 12 chapters that follow. There is tremendous interest (and hope) that because altered glucose and energy metabolism appears to be a virtually universal characteristic of metastatic cancer cells (i.e., the cells that kill cancer patients), understanding the mechanisms and consequences of metabolic reprogramming may lead to novel therapeutic approaches that will more effectively fight this formidable foe. Ring the bell for Round 1 . . . Baltimore, Maryland

Gregg L. Semenza, M.D., Ph.D.

Contents

Oxidative Phosphorylation and Cancer: The Ongoing Warburg Hypothesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Michael Ristow and José M. Cuezva

1

The Electron Transport Chain and Carcinogenesis . . . . . . . . . . . . . . . . . . . . . 19 Jean-Jacques Brière, Paule Bénit, and Pierre Rustin Respiratory Control of Redox Signaling and Cancer . . . . . . . . . . . . . . . . . . . 33 Pauline M. Carrico, Nadine Hempel, and J. Andrés Melendez Cellular Respiration and Dedifferentiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 Roberto Scatena, Patrizia Bottoni, and Bruno Giardina Cellular Adaptations to Oxidative Phosphorylation Defects in Cancer . . . . 55 Sarika Srivastava and Carlos T. Moraes Regulation of Glucose and Energy Metabolism in Cancer Cells by Hypoxia Inducible Factor 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 ´ Túlio César Ferreira and Elida Geralda Campos The Role of Glycolysis in Cellular Immortalization . . . . . . . . . . . . . . . . . . . . . 91 Hiroshi Kondoh Metabolic Modulation of Carcinogenesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 Shireesh P. Apte and Rangaprasad Sarangarajan Mitochondrial DNA Mutations in Tumors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 Anna Czarnecka and Ewa Bartnik Cellular Respiration and Tumor Suppressor Genes . . . . . . . . . . . . . . . . . . . . . 131 Luis F. Gonzalez-Cuyar, Fabio Tavora, Iusta Caminha, George Perry, Mark A. Smith, and Rudy J. Castellani

vii

viii

Contents

Uncoupling Cellular Respiration: A Link to Cancer Cell Metabolism and Immune Privilege . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 M. Karen Newell, Elizabeth M. Villalobos-Menuey, Marilyn Burnett, and Robert E. Camley How Cancer Cells Escape Death . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161 Erica Werner Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179

Contributors

Shireesh P. Apte Senior Scientist, Process Research and Support, Mail stop: R4-11 Alcon Laboratories, 6201 South Freeway, Fort Worth, Texas 76134, USA Ewa Bartnik Department of Genetics and Biotechnology, University of Warsaw; and Institute of Biochemistry and Biophysics, Polish Academy of Sciences Warsaw, Pawinskiogo 5a. 02106, Poland Paule Bénit Hôpital Robert Debré, INSERM U676, Paris, France Patrizia Bottoni Dipartimento di Medicina di Laboratorio, Università Cattolica, Rome, Italy Jean-Jacques Brière Hôpital Robert Debré, INSERM U676, Paris, France Marilyn Burnett CU Institute of Bioenergetics and Immunology, University of Colorado at Colorado Springs, Colorado Springs, Colorado Iusta Caminha Department of Pharmacology, University of Maryland, Baltimore, Maryland Robert E. Camley CU Institute of Bioenergetics and Immunology, University of Colorado at Colorado Springs, Colorado Springs, Colorado ´ Elida Geralda Campos Departamento de Biologia Celular, Universidade de Bras´ılia, Bras´ılia, DF, Brazil Pauline M. Carrico Ordway Research Institute, Albany, New York Rudy J. Castellani Department of Pathology, University of Maryland, Baltimore, Maryland ix

x

Contributors

José M. Cuezva Departamento de Biolog´ıa Molecular, Centro de Biolog´ıa Molecular “Severo Ochoa,” Centro de Investigación Biomédica en Red de Enfermedades Raras, Universidad, Autónoma de Madrid, Madrid, Spain Anna Czarnecka Department of Genetics and Biotechnology, University of Warsaw; Warsaw, Poland; and Postgraduate School of Molecular Medicine, Warsaw, Poland Túlio César Ferreira Departamento de Biologia Celular, Universidade de Brasélia, Brasélia, DF, Brazil Bruno Giardina Dipartimento di Medicina di Laboratorio, Università Cattolica, Rome, Italy Luis F. Gonzalez-Cuyar Department of Pathology, University of Maryland, Baltimore, Maryland Nadine Hempel The Center for Immunology & Microbial Disease, Albany Medical College, Albany, New York Hiroshi Kondoh Department of Geriatric Medicine, Graduate School of Medicine, Kyoto University, Kyoto, Japan J. Andrés Melendez The Center for Immunology & Microbial Disease, Albany Medical College, Albany, New York Carlos T. Moraes Department of Neurology and Department of Cell Biology and Anatomy, University of Miami Miller School of Medicine, Miami, Florida M. Karen Newell CU Institute of Bioenergetics and Immunology, University of Colorado at Colorado Springs, Colorado Springs, Colorado George Perry College of Sciences, University of Texas at San Antonio, San Antonio, Texas; and Department of Pathology, Case Western Reserve University, Cleveland, Ohio Michael Ristow Department of Human Nutrition, Institute of Nutrition, University of Jena, Jena, Germany Pierre Rustin Hôpital Robert Debré, Paris, France

Contributors

xi

Rangaprasad Sarangarajan Associate Professor of Pharmacology & Toxicology, School of Pharmacy (Worcester) Massachusetts College of Pharmacy and Health Sciences, 19 Foster Street, Worcester, MA 01608, Gregg L. Semenza Director, Vascular Program, Institute for Cell Engineering; Professor, Departments of Pediatrics, Medicine, Oncology, Radiation Oncology, and the McKusick-Nathans Institute of Genetic Medicine, The Johns Hopkins University School of Medicine, Broadway Research Building, Suite 671, 733 North Broadway, Baltimore, MD 21205 USA Roberto Scatena Dipartimento di Medicina di Laboratorio, Università Cattolica del Sacro Cuore, Rome, Italy Mark A. Smith Department of Pathology, Case Western Reserve University, Cleveland, Ohio Sarika Srivastava Department of Neurology, University of Miami Miller School of Medicine, Miami, Florida Fabio Tavora Department of Pathology, University of Maryland, Baltimore, Maryland Elizabeth M. Villalobos-Menuey CU Institute of Bioenergetics and Immunology, University of Colorado at Colorado Springs, Colorado Springs, Colorado Erica Werner Department of Cell Biology, School of Medicine, Emory University, Atlanta, Georgia

Oxidative Phosphorylation and Cancer: The Ongoing Warburg Hypothesis Michael Ristow and José M. Cuezva

Abstract More than eight decades ago, the German physiologist Otto Warburg observed that cancer cells in the presence of oxygen produced large amounts of lactate and proposed that impaired oxidative metabolism may cause cancer. This formulation, later known as the Warburg hypothesis, was investigated and debated for several decades. The development of molecular biology and the discovery of oncogenes and tumor suppressor genes in subsequent years shifted the general interest in the cancer field into directions other than metabolism and promoted the abandoning of the Warburg hypothesis or its consideration as an epiphenomenon of cell transformation. In recent years, a renaissance in the field of mitochondria has occurred in biological studies. This has happened mostly by the recognition of the key functional role that this organelle plays in the execution of cell death and, thus, for its participation in the development of a vast array of human pathologies. The cancer field was not indifferent to these changes, and the Warburg hypothesis was brought back to the scene with renewed strength. Specifically, recent findings suggest that cancer is associated with a decrease in the activity and expression of ␤-F1-ATPase, a key subunit of the mitochondrial ATP synthase. This alteration has been shown to limit oxidative phosphorylation and to trigger the induction of glycolysis to provide energy to the cell thus configuring the earlier Warburg observation in an additional hallmark of the cancer cell. Moreover, increased and decreased cellular mitochondrial activities are respectively associated with suppression and development of cancer. We suggest that reactivating mitochondrial metabolism by pharmacologic or dietary measures and/or tackling the deviant glycolysis in cancer cells may efficiently suppress malignant growth.

M. Ristow (B) Chair, Department of Human Nutrition, Institute of Nutrition, University of Jena, D-07743 Jena, Germany e-mail: [email protected] J.M. Cuezva (B) Departamento de Biolog´ıa Molecular, Centro de Biolog´ıa Molecular “Severo Ochoa”, CIBER de Entermedades Rasas, Universidad Autónoma de Madrid, 28049 Madrid, Spain. e-mail: [email protected] S.P. Apte, R. Sarangarajan (eds.), Cellular Respiration and Carcinogenesis, DOI 10.1007/978-1-59745-435-3 1, C Humana Press, a part of Springer Science+Business Media, LLC 2009

1

2

M. Ristow, J.M. Cuezva

Keywords Mitochondria · Oxidative phosphorylation · H+ -ATP synthase · Glycolysis · Cancer Nowadays, malignant disorders are a major cause of premature death. The incidence of cancers increases with age, whereas metabolic activity and resting energy expenditure resembling oxidative metabolism continuously decrease during aging of humans. Hence, an inverse relationship between mitochondrial activity and the promotion of tumor growth might possibly exist. To analyze whether this apparent link is supported by scientific evidence rather than simply being a coincidence, tremendous efforts have been undertaken in the past to elucidate this possible connection. This chapter aims to summarize published evidence on mechanistic interdependencies between cancer biology and the activity of mitochondria. Other recent reviews dealing with this subject are also available [1, 2]. In the past decades, glucose has become a major component of Western diets [3]. The most frequently used nutritive sugar, the disaccharide sucrose, contains equal amounts of the monosaccharides fructose and glucose. Glucose is actively transported into mammalian cells, and subsequently is (as well as fructose) biochemically converted into pyruvate (Fig. 1). Pyruvate either is converted into lactate, which prevents its further intracellular breakdown, or is completely oxidized to carbon dioxide and water by mitochondrial activities linked to the supply of molecular oxygen (Fig. 1). The breakdown of glucose into pyruvate is termed glycolysis. Glycolysis takes place in the cytoplasm of eukaryotic cells, and it can proceed both in the absence and presence of oxygen (Fig. 1). In order to carry on glycolysis in the former case, the carbon skeleton of pyruvate needs to be reduced to lactate to regenerate the pool of oxidized NAD+ (Fig. 1). Many prokaryotes and lower eukaryotes including Saccharomyces cerevisiae (yeast) survive in the absence of oxygen by obtaining energy from anaerobic glycolysis only [4], whereas higher eukaryotes, especially mammals, require a continuous supply of oxygen for obtaining biological energy. Glycolysis in absence or presence of oxygen provides comparably little energy equivalents per mole of glucose oxidized when compared with its terminal oxidation in the mitochondria (Fig. 1). Nevertheless, anaerobic glycolysis supplies the precursors and building blocks for the biosynthesis of cellular macromolecules, and, in yeast, as well as in certain mammalian cell types, is sufficient to maintain viability for longer periods of time. Because highly developed organisms tend to use nutritive energy equivalents more efficiently than do less developed organisms [5], mammals have significantly more efficient ways to generate energy equivalents. This is illustrated by the further oxidation of a metabolite of glycolysis (pyruvate) in mitochondria, which yields an approximately 18-fold amount of energy equivalents per mole of glucose oxidized when compared with anaerobic glycolysis (Fig. 1). Hence, the major advantage of glycolytic metabolism is to be independent from oxygen, and the major disadvantage of this process is a comparably low yield of energy equivalents (Fig. 1). For many decades, mitochondrial organelles were considered eminently important for the generation of biological energy and as the cellular site that allowed

Oxidative Phosphorylation and Cancer: The Ongoing Warburg Hypothesis HK Glucose

PFK1

PGI

G-6-P

F-6-P

3

ALD

F-1,6-P

TPI

G-3-P

G-3-P + DHAP

GAPDH ENO

PEP

PGM

2-PG

PGK

3-PG

1,3-BPG

PK

02

Pyruvate

NAD+

LDH

Lactate

Net ATP produced: 2 Glucose utilization: HIGH

AEROBIOSIS

AEROBIOSIS / ANAEROBIOSIS

NADH

pyruvate

CO2 + H2O Net ATP produced: 36 Glucose utilization: LOW

Fig. 1 Cellular ATP provision defines the rate of aerobic glycolysis. Abbreviations of glycolytic intermediates and enzymes of glycolysis are presented sequentially up to the pyruvate crossroad. The key glycolytic enzymes hexokinase (HK), phosphofructokinase 1 (PFK1), and pyruvate kinase (PK) are highlighted in bold. The reduction of pyruvate to lactate by lactate dehydrogenase (LDH) regenerates NAD+ and allows glycolysis to proceed with very low yield of energy and a high consumption of glucose. Pyruvate could be oxidized to CO2 and water only in the presence of oxygen by mitochondria. In this situation, the yield of energy obtained is high and consequently glucose consumption is low

the integration of the metabolism of macronutrients, specifically carbohydrates, fatty acids, and amino acids. Interconversion of carbohydrate derivatives (i.e., pyruvate) into fatty acids, amino acids into carbohydrate precursors, and others occur in the matrix of mitochondria, and these activities are integrated in the Krebs cycle. Concurrently, the reducing equivalents obtained from the terminal oxidation of the metabolic intermediates of the Krebs cycle in the form of NADH and FADH2 are shuttled into the complexes of the respiratory chain (Fig. 2). The respiratory chain is composed of proteins that are located in the inner mitochondrial membrane and have, as overall function, the reoxidation of NADH and FADH2 to further allow the oxidation of more metabolic substrates (Fig. 2). The ordered transfer of electrons along the carriers of the respiratory chain to molecular oxygen (Fig. 2), which is the final electron acceptor, is the process known as respiration. Electron transfer in complexes of the respiratory chain trigger the proton pumping activity of complexes CI, CIII, and CIV (Fig. 2), which extrudes protons from the matrix interior of mitochondria and finally results in the generation of the proton electrochemical gradient (⌬␮H+ ) across the inner membrane of the organelle, which is the intermediate used for the synthesis of ATP [6] (Fig. 2). The reentry of protons to the matrix interior through the proton channel of the H+ -ATP synthase (CV in Fig. 2) drives the

4


O2·

CYTOPLASM O.M.

H+

H+

Cyt c – CI e

Q

NAD+ FADH2

CIV

e–

I.M.

CIII

– CII e

NADH

ΔμH+

H+

e– FAD

O2 H2O

CV

ATP H+ MATRIX

O2· Fig. 2 Mitochondrial activities: respiration, oxidative phosphorylation, and the production of ROS. The oxidation of metabolic substrates promotes the reduction of redox coenzymes (NAD+ and FAD) that feed the electrons (e-) to the complexes of the respiratory chain (CI to CIV) placed in the inner mitochondrial membrane (I.M.). Electron flow down the respiratory chain to molecular oxygen generates the water of respiration. Electron transfer is coupled with proton pumping from the matrix interior to the intermembrane space at CI, CIII, and CIV to generate the proton electrochemical gradient (⌬␮H+ ) across the inner membrane. ⌬␮H+ is the driving force used by the H+ -ATP synthase (CV) for the synthesis of ATP. However, at complexes I and III, some of the electrons can combine with molecular oxygen to generate the toxic superoxide radical (dashed lines). Q, ubiquinone; Cyt c, cytochrome c; O.M., outer membrane

biosynthesis of adenosine triphosphate (ATP) from adenosine diphosphate (ADP) and inorganic phosphate (Pi) in the process known as oxidative phosphorylation (OXPHOS) (Fig. 2). Whereas the metabolism of nutrient intermediates and OXPHOS were thought to be the primary roles of mitochondria for many decades, recent findings indicate that mitochondria play additional essential cellular functions. The protist Giardia lamblia carries remnant organelles resembling mitochondria that, though they are unable to provide ATP, instead synthesize ironsulfur (Fe-S) clusters [7]. Fe-S clusters are essential parts of specific mitochondrial enzymes, namely aconitase in the Krebs cycle as well as complexes of the respiratory chain [8]. Briefly, Fe-S clusters are known to carry non-heme iron, which can both donate and accept electrons [9]. Moreover, mitochondria are central regulators and executors of cell death [10], and the acquisition of a resistant phenotype to apoptosis is a critical aspect in the biology of tumors and its therapy [11, 12]. It is now firmly established that upon the activation of cell death by various types of stimuli (genetic damage, ischemia, chemical and/or metabolic stresses), mitochondria release proteins (cyt c, AIF, Endo G, Smac/DIABLO, Omi/HtrA2) and/or other molecules such as the superoxide anion (O2 . ) that participate in signaling and

Oxidative Phosphorylation and Cancer: The Ongoing Warburg Hypothesis

5

executing cell demise. It should be noted that in most cells, reactive oxygen species (ROS) are produced by single electron transfer reactions to molecular oxygen during mitochondrial respiration (Fig. 2). In summary, mitochondria are responsible for the production of Fe-S clusters, the conversion of primary energy substrates into each other, the generation of large amounts of ATP, and the signaling and execution of cell death. The major advantage of oxidative metabolism is the comparably high energy yield, and the major disadvantage is its dependency on the availability of oxygen and the subsequent production of ROS that could damage essential cellular constituents. Unaware of the above-mentioned facts, the physiologist Otto Warburg (Fig. 3 and Fig. 4) proposed as early as 1924 that malignant growth might be caused by increased glycolysis accompanied by impaired oxygen consumption [13–16]. This hypothesis was based on his original observations that cancer tissues show a low respiratory rate compared with that of adjacent healthy (i.e., nonmalignant) tissues. After becoming the Nobel laureate for medicine and physiology in 1931 [17], his hypothesis on initiation of cancer obtained increasing attention and caused widespread debate [18–22]. Warburg died in 1970 while his hypothesis was still

Fig. 3 Otto Warburg (date unknown). (Photograph courtesy of the Archives of the Max Planck Society, Berlin, Germany.)

6


Fig. 4 Otto Warburg in his laboratory (date unknown). (Photograph courtesy of the Archives of the Max Planck Society, Berlin, Germany.)

unconfirmed, denied, and/or neglected [23, 24] mostly because of the abandonment of metabolic studies in favor of the development of molecular biology and the discovery of oncogenes, tumor suppressor genes, and other advances in cancer research. In any case, and after both OXPHOS and respiration had been attributed to mitochondria, electron microscopy–based techniques were developed to study the morphology of these organelles in tumor tissues. In summary, mitochondria from rapidly growing cancers show less cristae and are smaller than those from betterdifferentiated cancers and normal tissues [25, 26]. Besides these structural differences in the mitochondria of the cancer cell, it has been further emphasized that some cancer tissues also have a significant reduction in the number of mitochondria per cell [25–27]. The abnormal high-aerobic glycolysis of tumors discovered by Warburg was confirmed by most researchers studying cancer cells, and a significant number of potential mechanisms have been put forward to explain the metabolic changes specific to cancer cells. These can be separated into mechanisms that impinge on cellular glucose uptake and consumption by glycolysis (Fig. 1) and those affecting the mitochondrial metabolism of pyruvate (Fig. 1 and Fig. 2). When comparing energy metabolism of tumors and healthy tissue, profound changes have been observed in virtually all cases of cancer or tissue types. Increased glycolysis has been found in cancer cells at the genomic [28], proteomic [26, 29, 30–32], and metabolic levels:


7

increased glucose uptake [33, 34], as well as increased production of lactate as the end product of glycolysis [35, 36]. A modern noninvasive imaging technique called positron emission tomography (PET) is nowadays in clinical use for diagnosing and staging cancer patients [37] because it is remarkably sensitive to detect tumor tissue and its metastasis in situ based on the principles of the Warburg hypothesis [32]. By injecting the nonmetabolizable glucose analogue [18 F]fluoro-2-deoxy-d-glucose (FDG) as a tracer substance into the human body, tissues with high FDG uptake (high glucose avidity) can easily be identified from the surrounding normal tissue in images of PET scans [38]. Using this approach, we have recently showed that lung tumors with high glucose avidity have a profound alteration of the bioenergetic proteome [32]. The bioenergetic proteome (BEC index) provides a signature of the tumor that informs of the preponderance of the metabolic pathways relevant for cellular energy provision at the protein level [2, 26]. Remarkably, the analysis of the bioenergetic signature in tumor specimens and in cells in culture [32] revealed that the downregulation of the bioenergetic function of mitochondria (Fig. 1 and Fig. 2), specifically of oxidative phosphorylation, underlies the high rates of glucose uptake and consumption by aerobic glycolysis in the carcinomas (Fig. 1) [32]. These results have provided the first indications that integrate molecular and functional data to confirm Warburg’s initial observations [32]. In a recent review, Gatenby and Gillies put forward an evolutionary model to explain the apparently high glycolytic rates of cancer cells [39]. Under conditions of extensive hypoxia, which most likely occurs during hyperplastic growth, evolutionary selection favors cells with elevated glycolytic capacity [40]. Moreover, increased glycolysis causes increased lactate production and subsequent lactate-dependent acidification. Selection is in favor of those cells that exhibit increased resistance toward an unfavorable acidic environment. Therefore, a premalignant cell will be less dependent on oxygen and is still able to expand under conditions that would kill a nonmalignant cell. In support of Warburg’s hypothesis, this simple model provides an explanation on the benefit of cancer cells from an energetic trade-off (i.e., preference for nonmitochondrial energy supply). Numerous experimental findings further emphasize this assumption. Induction of glucose-specific transporters (GLUTs), and, most importantly GLUT1, was several times correlated with tumor aggressiveness [41–43]. Moreover, additional evidence links a number of established cancer genes to control of glycolytic flux. In this regard, it has been shown that the c-myc oncogene induces expression of GLUT1 and a number of additional genes involved in glycolytic flux, including glucose-6phosphate isomerase, phosphofructokinase, glyceraldehyde-3-phosphate dehydrogenase, phosphoglycerate kinase, lactate dehydrogenase, and others (Fig. 1) [44]. Of interest, c-Myc also acts to transactivate the lactate dehydrogenase A (LDHA) gene which is necessary for c-myc–mediated transformation in fibroblasts [35]. Accordingly, expression of other oncogenes (ras, src, PDGF) causes increased glucose uptake in malignant cells [34, 35, 45]. Additionally, and acting as c-myc itself, such cells were capable of coactivating LDHA expression via proposed binding to a carbohydrate responsive element found in the LDHA promoter region [46].

8


Another relevant mechanism to regulate glycolysis and metabolic adaptation to oxygen depletion is that mediated by the hypoxia inducible factor 1 alpha (HIF-1␣). Subsequent to its identification in 1992 [47], a significant body of evidence has emerged proposing that HIF-1␣ might be relevant for many of the effects as suggested in the initial Warburg hypothesis [48]. HIF-1␣ forms dimers with its partner HIF-1␤ to become an active transcription factor binding to promoter elements of several genes including those encoding for proteins of glucose transport [49] and activating the expression of several enzymes of the glycolytic pathway (Fig. 1) including LDHA [50, 51]. Furthermore, the active HIF-dimer promotes angiogenesis and erythropoiesis [47, 52]. HIF-1␣ is constitutively expressed in most tissues and degraded under conditions of normal oxygen pressure. In contrast, when oxygen partial pressure is decreased, degradation of HIF-1␣ is blocked causing increased transcriptional activity to regulate the respective target genes following changes in oxygen availability [53]. Accordingly, deregulated HIF-1␣ or sustained stability under conditions of normal oxygen supply has been linked to numerous types of cancer in past years [54]. Furthermore, the ubiquitin ligase responsible for HIF-1␣ inactivation, the von Hippel–Lindau tumor suppressor (VHL) protein, has been affiliated with the von Hippel–Lindau syndrome, which promotes neuronal and kidney cancers [55]. Moreover, downregulation of pyruvate dehydrogenase might lead to accumulation of pyruvate and increased lactate production, which contributes to stabilization of HIF-1␣ [56]. Hence, although hypoxia might play a key role in tumor development, it seems not to be unambiguously required for sustained increases in glycolysis. Putatively, this may promote a vicious cycle of enhanced glycolysis leading to increased accumulation of pyruvate, which subsequently may stimulate HIF-1␣ and lastly further increase the expression of glycolytic enzymes [57]. Another candidate pathway controlling the glycolysis-dependent phenotype of malignancies, the Akt/PBK (protein kinase B) pathway, has been proposed. Akt/PKB has long been known for its capability of inducing glucose incorporation and glycolytic flux independently of malignancy-promoting capacity [58]. In malignant cells, elevated glycolytic flux was linked to activation of Akt/PKB [59]. Accordingly, blastoma cells lacking constitutive activation of Akt/PKB did not display altered glycolytic flux, whereas expression of Akt/PBK promoted cell death under conditions of glucose deprivation. Moreover, it has recently been shown that the overexpression of Akt/PBK in noninvasive radial-growth melanoma cells triggers an increase in protein markers of glycolysis and in the glycolytic flux of the melanoma concurrently with its transformation to the aggressive vertical-growth invasive behavior [60]. Altogether, these results suggest the possibility that deregulated Akt/PKB signaling might contribute to the Warburg effect in cancer cells. Additionally, interactions of glycolytic enzymes with the well-characterized p53 tumor suppressor protein have been described, further promoting the relevance of cancer-related genes in causing a potential metabolic shift toward increased glycolytic rate. In this regard, a putative role for hexokinase has been proposed [61]. In hepatoma cells, hexokinase was closely associated with the mitochondria, which was not observed in normal (i.e., untransformed) liver cells [62]. A subsequent study revealed that mutations in the p53 tumor suppressor protein might activate


9

hexokinase II expression via binding to specific sites of the corresponding hexokinase promoter [63, 64]. In further support of these interesting observations, a protein named TP53-induced glycolysis and apoptosis regulator (TIGAR ), an isoform of phosphofructokinase that can be induced by p53, was identified in recent years [65]. TIGAR acts as an inhibitor of glycolysis by lowering levels of fructose-2,6bisphosphate. Accordingly, in cases of p53 mutations as found in many cancers, activity of TIGAR is downregulated promoting an increase in glycolysis. In summary, a number of cancer-related genes have been shown to promote alterations on the glycolytic phenotype that could explain that particular aspect of the Warburg effect. Aspects other than deregulated glycolysis underscore that oxidative metabolism itself is also impaired in malignant cells, thereby representing another key aspect of Warburg’s hypothesis. The phenomenon of impaired mitochondrial activity [13, 15, 25] appears to be a key feature of numerous cancer specimens [26, 29, 31, 32, 66, 67]. Accordingly, mutations of mitochondrial DNA have repeatedly been associated with induction of tumor growth on a hypothetical basis [67, 68] and later on a factual basis [69–73]. Complexes of the mitochondrial respiratory chain (Fig. 2) including cytochrome c oxidase (COX) exhibit diminished activity in cancers, mainly due to decreased expression of the subunits encoded by the mitochondrial DNA [74]. Likewise, mutations in nuclear genes that affect the bioenergetic function of mitochondria, and thus compromise the provision of metabolic energy, have also been shown to predispose to different types of inherited neoplasia syndromes [75–77]. Moreover, over the past years it has been shown that most prevalent human tumors fit the original Warburg hypothesis because they show a decrease in the expression of ␤-F1-ATPase, which is the catalytic subunit of the mitochondrial H+ ATP synthase and thus a bottleneck of oxidative phosphorylation (CV in Fig. 2), concurrently with an increased expression of several makers of the glycolytic pathway (26, 29, 31 and references therein). This proteomic feature, which was defined as “the bioenergetic signature” of cancer [26], illustrates both alterations in the bioenergetic proteome of the mitochondrion and on the overall mitochondrial activity of the cancer cell [2] and provides markers of the two mechanisms that could interfere with mitochondrial activity in the various types of cancer cells [2]. This is especially relevant because depending upon the tissue type considered, a general reduction in mitochondrial mass or a specific reduction on the expression of the ATP synthase is associated with progression of the neoplasia [2, 26]. Interestingly, these recent findings perfectly complement earlier findings on reduced ATP(synth)ase activity in rodent models of liver carcinogenesis [78]. Additionally, Racker and coworkers also emphasized the putative relevance of ATP(synth)ase in the control of tumor proliferation [79, 80]. As indicated previously, we have recently documented both in vivo using FDGPET imaging in lung cancer patients and in vitro that the rates of glucose capture by lung tumors and of glucose utilization by aerobic glycolysis in cancer cells depend on the expression level of the ␤-F1-ATPase protein and on the activity of oxidative phosphorylation [32]. These results provide strong support to the

10


original Warburg hypothesis and to the principle of metabolic regulation known as the Pasteur effect [81]. This principle of metabolism, which basically states that “the metabolic flux of glycolysis in eukaryotic cells depends on the energy provided in the form of ATP by mitochondrial activity,” is what was used by Warburg as the “first stone” for the building of his seminal and outstanding hypothesis in the early dates of biochemistry [2]. Remarkably, the bioenergetic signature has been shown to provide a relevant marker of disease progression in colon [26], lung [30, 32], and breast [31] cancer patients. Furthermore, it has been documented that the bioenergetic signature also provides a marker of the cellular response to chemotherapy in colon [82] and liver [83] cancer cells that might allow its further use as a predictive marker of therapeutic intervention and/or as a target of future cancer treatments. Because it appears that alterations in mitochondrial energetic physiology contribute to the Warburg effect, that is, an impaired oxidative phosphorylation in mitochondria promotes the upregulation of glycolysis (Fig. 1), at least two questions arise: (i) what are the mechanism(s) that impair the OXPHOS activity of mitochondria, and (ii) what is the advantage and mechanistic contribution of the downregulation of OXPHOS for the cancer cell? These will be partially addressed in the following. Supported by previous findings on p53 as mentioned before, it has been shown that this tumor suppressor protein may directly be involved in regulating mitochondrial oxygen consumption through regulation of COX activity [84]. Accordingly, cancer cell lines lacking p53 showed decreased rates of OXPHOS as reflected by impaired ATP synthesis and decreased respiratory rate, and maintenance of cellular ATP levels was dependent on a concurrent increase in glycolysis [85, 86]. This specific regulation was promoted by Synthesis of Cytochrome c Oxidase 2 SCO2, which has previously been shown to be involved in the assembly of complex IV of the respiratory chain [86]. Nevertheless, it remains to be shown whether this particular activation can be directly linked to mitochondrial activity. Therefore, this tumor suppressor, p53, not only regulates glycolysis as shown above but possibly also directly affects mitochondrial oxygen consumption. Not only p53 has been shown to influence both glycolysis and mitochondrial activity. Similar conjunct activities have been put forward for both HIF-1␣ and Akt/PKB signaling pathways. In this regard, HIF-1␣ has previously been demonstrated to activate pyruvate dehydrogenase kinase (PDK), which is known to inactivate pyruvate dehydrogenase, preventing the decarboxylation of pyruvate and hence its further oxidation in the mitochondria having a negative impact on mitochondrial oxygen consumption [87, 88]. Likewise, the above-mentioned tumor suppressor protein VHL not only ubiquitinates HIF-1␣ but also promotes mitochondrial biogenesis and activity. Accordingly, loss of VHL, as repeatedly described in various types of cancer, might possibly impair mitochondrial activity strongly promoting the Warburg effect [66]. Moreover, reduced OXPHOS capacity has previously been suggested to promote phosphorylation and hence activation of Akt [89], a well-defined oncogene, in a Phosphatase and Tensin Homolog PTEN-dependent manner [90]. An additional alternative to explain the “Warburg phenotype” of the cancer cell is the one we have recently put forward [32] after the characterization of the biogenesis and functional development of mitochondria during the cell cycle [91]. We have


11

suggested that the downregulation of oxidative phosphorylation and the concurrent upregulation of glycolysis is the phenotype that identifies the energetic metabolism of most proliferating cells. In other words, glycolysis is the metabolic pathway that supports proliferation and tumor growth. In this regard, it is well established that lymphocytes switch to glycolysis as the energy-producing pathway during proliferation [92], and progression through the S/G2/M phases of the cell cycle, the socalled reductive phases of the “metabolic cycle” in which most of the mitochondrial components are being synthesized [91], is supported by nonrespiratory modes of energy generation [93]. In agreement with these findings, it has recently been shown that cyclin D1, which is involved in the phosphorylation and inactivation of the retinoblastoma protein marking in this way the entry of cells into the S phase, represses mitochondrial function in vivo [94, 95]. We presume that the effects that oncogenes and tumor suppressors could have on mitochondrial activity might be superimposed to the natural metabolic phenotype of proliferating cells. Mechanistically, and by analogy with findings on the biogenesis of mitochondria in rat hepatomas [27], in fetal rat liver [96–98] and during the cell cycle [91], the repression of ␤-F1-ATPase expression in human tumors [26, 29–31] could be exerted by translation masking (silencing) of the ␤-F1-ATPase mRNA, a condition that might also affect other OXPHOS mRNAs of the cancer cell [2]. According to Warburg’s hypothesis, impaired mitochondrial function should be considered a key property of cancer cells (i.e., an additional hallmark of cancer to be added to the list) [99]. Strongly supporting this hypothesis, it was shown that induction of mitochondrial energy conversion by forced expression of the mitochondrial protein frataxin [100] leads to reduced growth of cancer cells [101]. Frataxin has been shown to be involved in Fe-S cluster biogenesis [102] and induces OXPHOS [100]. In this regard, and as a novel pathway connecting induction of mitochondrial metabolism and decreased growth rates, mitogen activated protein kinase p38 has been proposed [101, 103]. This protein p38 has repeatedly been observed to act as an efficient tumor suppressor protein [104, 105]. In full accordance with these observations, hepatocyte-specific inactivation of murine frataxin reduces the activity of Fe-S cluster–containing enzymes within the mitochondria, promoting hepatic adenoma formation in such mice [103]. Consistently, defective Krebs cycle activity has been observed in colorectal cancer samples including reduced activity of the Fe-S cluster–containing enzyme aconitase [106]. Taken together, these findings indicate that activation of mitochondrial energy metabolism has a tumor suppressor effect that opposes the Warburg phenotype and could be a therapeutic tool in cancer therapy. Indeed, the execution of cell death requires an efficient oxidative phosphorylation [107–110], the molecular components of the H+ -ATP synthase being specifically required [111–113]. In this regard, we have recently shown that the activity of the H+ -ATP synthase is necessary for the efficient execution of apoptosis in cells that depend on oxidative phosphorylation for the provision of metabolic energy, whereas it is dispensable in those cells that rely heavily on glycolysis [83]. The role of the H+ -ATP synthase in the execution of cell death is mediated by controlling the generation of ROS [83], which in turn promote a severe oxidative damage on mitochondrial proteins, favoring in this way the release of apoptogenic molecules from the organelle [83]. Consistently, highly glycolytic cells

12


with an arrested mitochondrial biogenesis and thus with no dependence on oxidative phosphorylation for energy provision do not produce ROS and are resistant to mitochondria-geared cell death stimuli [83]. Altogether, these results provide additional evidence linking metabolism to cell death [114–117] and support strongly that the acquisition of a Warburg phenotype is another strategy of the cancer cell in order to ensure its perpetuation. In recent years, AMP activated protein kinase (AMPK) has been established as a key regulator of cellular energy balance controlling a significant number of pathways that control cellular ATP content [118]. AMPK is activated under conditions of low energy supply, induced by stressors like reduced nutrient availability or hypoxia [118]. Activation of AMPK is initiated by the upstream tumor suppressor kinase, LKB1, which phosphorylates AMPK [119]. Of note, a mutation of LKB1 has been shown to cause Peutz-Jeghers syndrome, a disease that has been linked to increased incidence of cancers. Moreover, two other downstream targets of AMPK, the tumor suppressors TSC1 and 2, have been shown to be mutated in tuberous sclerosis [120]. One of the major outcomes of AMPK and/or activation of its downstream substrates is the induction of mitochondrial biogenesis and/or activity [121]. Not surprisingly, AMPK appears to be relevant in the control of protein synthesis and proliferation [122], and may activate p53 causing cell cycle arrest and/or cellular senescence in states of decreased glucose availability [123]. Given the connection of AMPK on the one hand and several tumor suppressors, including TSCs and LKB1 (see above), on the other hand, as well as the fact that AMPK itself is of eminent relevance to mitochondrial activity and energy balance, disruption of this pathway might be considered of importance in regard to the molecular causes of the Warburg effect in states of cancer and malignant growth. In summary, though the Warburg hypothesis was formulated more than 80 years ago, nowadays it still (or rather again) is a subject of extensive debate. The recent findings on the metabolic signature of tumors and malignant cells on the one hand and several molecular pathways linked to derangements of cellular energy flux in transformed state on the other hand strongly support the overdue importance of one of the most long-standing hypotheses in oncology—Warburg’s hypothesis. Acknowledgments We thank members of our laboratories for their contributions and encouraging discussions on the subject of this review. This review article was written while the research activity in the authors’ laboratory was supported by grants from the German government to M.R. and from the Ministerio de Educación y Ciencia (BFU2007-65253) and Fundación Mutua Madrileña to J.M.C.

References 1. Ristow M. Oxidative metabolism in cancer growth. Curr Opin Clin Nutr Metab Care 2006; 9:339–345. 2. Cuezva JM, Sanchez-Arago M, Sala S, Blanco-Rivero A, Ortega AD. A message emerging from development: the repression of mitochondrial beta-F1-ATPase expression in cancer. J Bioenerg Biomembr 2007; 39:259–265.


13

3. Lichtenstein AH, Appel LJ, Brands M, et al. Diet and lifestyle recommendations revision 2006: a scientific statement from the American Heart Association Nutrition Committee. Circulation 2006; 114:82–96. 4. Pasteur L. Influence de l oxygene sur le developpement de la levure et la fermentation alcoolique. Bulletin de la Societe Chimique de Paris 1861; 79–80. 5. Pfeiffer T, Schuster S, Bonhoeffer S. Cooperation and competition in the evolution of ATPproducing pathways. Science 2001; 292:504–507. 6. Boyer PD. The ATP synthase. A splendid molecular machine. Annu Rev Biochem 1997; 66:717–749. 7. Tovar J, Leon-Avila G, Sanchez LB, et al. Mitochondrial remnant organelles of Giardia function in iron-sulphur protein maturation. Nature 2003; 426:172–176. 8. Lill R, Mühlenhoff U. Iron-sulfur-protein biogenesis in eukaryotes. Trends Biochem Sci 2005; 30:133–141. 9. Rouault TA, Klausner RD. Iron-sulfur clusters as biosensors of oxidants and iron. Trends Biochem Sci 1996; 21:174–177. 10. Wang X. The expanding role of mitochondria in apoptosis. Genes Dev 2001; 15: 2922–2933. 11. Green DR, Reed JC. Mitochondria and apoptosis. Science 1998; 281:1309–1312. 12. Jaattela M. Multiple cell death pathways as regulators of tumour initiation and progression. Oncogene 2004; 23:2746–2756. 13. Warburg O. On the origin of cancer cells. Science 1956; 123:309–314. 14. Warburg O. The metabolism of tumours. London: Constable, 1930. 15. Warburg O. On respiratory impairment in cancer cells. Science 1956; 124:269–270. ¨ 16. Warburg O, Posener K, Negelein E. Uber den Stoffwechsel der Tumoren [On metabolism of tumors]. Biochemische Zeitschrift 1924; 152:319–344. 17. Anonymous. The Nobel prize in physiology or medicine 1931. Available at: http://nobelprizeorg/medicine/laureates/1931/. 18. Burk D. A colloquial consideration on the Pasteur and Neo-Pasteur effects. Cold Spring Harbor Symposia on Quantitative Biology 1939; 7:420–460. 19. Burk D, Schade AL. On respiratory impairment in cancer cells. Science 1956; 124: 270–272. 20. Burk D, Woods M. Newer aspects of glucose fermentation in cancer growth and control. Arch Geschwulstforsch 1967; 28:305–319. 21. Weinhouse S. On respiratory impairment in cancer cells. Science 1956; 124:267–269. 22. Zu XL, Guppy M. Cancer metabolism: facts, fantasy, and fiction. Biochem Biophys Res Commun 2004; 313:459–465. 23. Warburg O. Annual meeting of Nobelists at Lindau, Germany, 1966. English edition by D. Burk. Bethesda, MD: National Cancer Institute. Available at: http://www.hopeforcancer. com/OxyPlus.htm. 24. Krebs H. Otto Warburg: cell physiologist, biochemist and eccentric. Oxford, UK: Clarendon, 1981. 25. Pedersen PL. Tumor mitochondria and the bioenergetics of cancer cells. Prog Exp Tumor Res 1978; 22:190–274. 26. Cuezva JM, Krajewska M, de Heredia ML, et al. The bioenergetic signature of cancer: a marker of tumor progression. Cancer Res 2002; 62:6674–6681. 27. López de Heredia M, Izquierdo JM, Cuezva JM. A conserved mechanism for controlling the translation of beta-F1-ATPase mRNA between the fetal liver and cancer cells. J Biol Chem 2000; 275:7430–7437. 28. Altenberg B, Greulich KO. Genes of glycolysis are ubiquitously overexpressed in 24 cancer classes. Genomics 2004; 84:1014–1020. 29. Isidoro A, Martinez M, Fernandez PL, et al. Alteration of the bioenergetic phenotype of mitochondria is a hallmark of breast, gastric, lung and oesophageal cancer. Biochem J 2004; 378:17–20.

14


30. Cuezva JM, Chen G, Alonso AM, et al. The bioenergetic signature of lung adenocarcinomas is a molecular marker of cancer diagnosis and prognosis. Carcinogenesis 2004; 25: 1157–1163. 31. Isidoro A, Casado E, Redondo A, et al. Breast carcinomas fulfill the Warburg hypothesis and provide metabolic markers of cancer prognosis. Carcinogenesis 2005; 26:2095–2104. 32. López-R´ıos F, Sánchez-Aragó M, Garc´ıa-Garc´ıa E, et al. Loss of the mitochondrial bioenergetic capacity underlies the glucosa avidity of carcinomas. Cancer Res 2007; 67:9013–9017. 33. Birnbaum MJ, Haspel HC, Rosen OM. Transformation of rat fibroblasts by FSV rapidly increases glucose transporter gene transcription. Science 1987; 235:1495–1498. 34. Flier JS, Mueckler MM, Usher P, Lodish HF. Elevated levels of glucose transport and transporter messenger RNA are induced by ras or src oncogenes. Science 1987; 235:1492–1495. 35. Shim H, Dolde C, Lewis BC, et al. c-Myc transactivation of LDH-A: implications for tumor metabolism and growth. Proc Natl Acad Sci USA 1997; 94:6658–6663. 36. Ziegler A, von Kienlin M, Decorps M, Remy C. High glycolytic activity in rat glioma demonstrated in vivo by correlation peak 1H magnetic resonance imaging. Cancer Res 2001; 61:5595–5600. 37. Rigo P, Paulus P, Kaschten BJ, et al. Oncological applications of positron emission tomography with fluorine-18 fluorodeoxyglucose. Eur J Nucl Med 1996; 23:1641–1674. 38. Gambhir SS. Molecular imaging of cancer with positron emission tomography. Nat Rev Cancer 2002; 2:683–693. 39. Gatenby RA, Gillies RJ. Why do cancers have high aerobic glycolysis? Nat Rev Cancer 2004; 4:891–899. 40. Raghunand N, Gatenby RA, Gillies RJ. Microenvironmental and cellular consequences of altered blood flow in tumours. Br J Radiol 2003; 76(Spec No 1):S11–22. 41. Grover-McKay M, Walsh SA, Seftor EA, Thomas PA, Hendrix MJ. Role for glucose transporter 1 protein in human breast cancer. Pathol Oncol Res 1998; 4:115–120. 42. Sakashita M, Aoyama N, Minami R, et al. Glut1 expression in T1 and T2 stage colorectal carcinomas: its relationship to clinicopathological features. Eur J Cancer 2001; 37:204–209. 43. Younes M, Lechago LV, Lechago J. Overexpression of the human erythrocyte glucose transporter occurs as a late event in human colorectal carcinogenesis and is associated with an increased incidence of lymph node metastases. Clin Cancer Res 1996; 2:1151–1154. 44. Osthus RC, Shim H, Kim S, et al. Deregulation of glucose transporter 1 and glycolytic gene expression by c-Myc. J Biol Chem 2000; 275:21797–21800. 45. Govindarajan B, Shah A, Cohen C, et al. Malignant transformation of human cells by constitutive expression of platelet-derived growth factor-BB. J Biol Chem 2005; 280: 13936–13943. 46. Dang CV, Lewis BC, Dolde C, Dang G, Shim H. Oncogenes in tumor metabolism, tumorigenesis, and apoptosis. J Bioenerg Biomembr 1997; 29:345–354. 47. Semenza GL, Wang GL. 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 1992; 12:5447–5454. 48. Robey IF, Lien AD, Welsh SJ, Baggett BK, Gillies RJ. Hypoxia-inducible factor-1alpha and the glycolytic phenotype in tumors. Neoplasia 2005; 7:324–330. 49. Ebert BL, Firth JD, Ratcliffe PJ. Hypoxia and mitochondrial inhibitors regulate expression of glucose transporter-1 via distinct cis-acting sequences. J Biol Chem 1995; 270: 29083–29089. 50. Firth JD, Ebert BL, Ratcliffe PJ. Hypoxic regulation of lactate dehydrogenase A. Interaction between hypoxia-inducible factor 1 and cAMP response elements. J Biol Chem 1995; 270:21021–21027. 51. Semenza GL, Roth PH, Fang HM, Wang GL. Transcriptional regulation of genes encoding glycolytic enzymes by hypoxia-inducible factor 1. J Biol Chem 1994; 269:23757–23763. 52. Maxwell PH, Dachs GU, Gleadle JM, et al. Hypoxia-inducible factor-1 modulates gene expression in solid tumors and influences both angiogenesis and tumor growth. Proc Natl Acad Sci USA 1997; 94:8104–8109.


15

53. Semenza GL. HIF-1, O(2), and the 3 PHDs: how animal cells signal hypoxia to the nucleus. Cell 2001; 107:1–3. 54. Semenza GL. HIF-1 mediates the Warburg effect in clear cell renal carcinoma. J Bioenerg Biomembr 2007; 39:231–234. 55. McKusick VA, Kniffin CL, Tiller GE, et al. Online Mendelian Inheritance in Man: von Hippel-Lindau syndrome (OMIM 193300). Available at: http://www.ncbi.nlm.nih.gov/ entrez/dispomim.cgi?id=193300.> 56. Koukourakis MI, Giatromanolaki A, Sivridis E, Gatter KC, Harris AL. Pyruvate dehydrogenase and pyruvate dehydrogenase kinase expression in non small cell lung cancer and tumor-associated stroma. Neoplasia 2005; 7:1–6. 57. Lu H, Dalgard CL, Mohyeldin A, McFate T, Tait AS, Verma A. Reversible inactivation of HIF-1 prolyl hydroxylases allows cell metabolism to control basal HIF-1. J Biol Chem 2005; 280:41928–41939. 58. Rathmell JC, Fox CJ, Plas DR, Hammerman PS, Cinalli RM, Thompson CB. Akt-directed glucose metabolism can prevent Bax conformation change and promote growth factorindependent survival. Mol Cell Biol 2003; 23:7315–7328. 59. Elstrom RL, Bauer DE, Buzzai M, et al. Akt stimulates aerobic glycolysis in cancer cells. Cancer Res 2004; 64:3892–3899. 60. Govindarajan B, Sligh JE, Vincent BJ, et al. Overexpression of Akt converts radial growth melanoma to vertical growth melanoma. J Clin Invest 2007; 117:719–729. 61. Pedersen PL. 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 2007; 39:211–222. 62. Bustamante E, Morris HP, Pedersen PL. Energy metabolism of tumor cells. Requirement for a form of hexokinase with a propensity for mitochondrial binding. J Biol Chem 1981; 256:8699–8704. 63. Mathupala SP, Heese C, Pedersen PL. Glucose catabolism in cancer cells. The type II hexokinase promoter contains functionally active response elements for the tumor suppressor p53. J Biol Chem 1997; 272:22776–22780. 64. Smith TA, Sharma RI, Thompson AM, Paulin FE. Tumor 18F-FDG incorporation is enhanced by attenuation of P53 function in breast cancer cells in vitro. J Nucl Med 2006; 47:1525–1530. 65. Bensaad K, Tsuruta A, Selak MA, et al. TIGAR, a p53-inducible regulator of glycolysis and apoptosis. Cell 2006; 126:107–120. 66. Hervouet E, Demont J, Pecina P, et al. A new role for the von Hippel-Lindau tumor suppressor protein: stimulation of mitochondrial oxidative phosphorylation complex biogenesis. Carcinogenesis 2005; 26:531–539. 67. Wu M, Neilson A, Swift AL, et al. Multiparameter metabolic analysis reveals a close link between attenuated mitochondrial bioenergetic function and enhanced glycolysis dependency in human tumor cells. Am J Physiol Cell Physiol 2007; 292:C125–136. 68. Hoberman HD. Is there a role for mitochondrial genes in carcinogenesis? Cancer Res 1975; 35:3332–3335. 69. Baggetto LG. Role of mitochondria in carcinogenesis. Eur J Cancer 1992; 29A:156–159. 70. Carew JS, Huang P. Mitochondrial defects in cancer. Mol Cancer 2002; 1:9. 71. Petros JA, Baumann AK, Ruiz-Pesini E, et al. mtDNA mutations increase tumorigenicity in prostate cancer. Proc Natl Acad Sci USA 2005; 102:719–724. 72. Singh KK. Mitochondrial dysfunction is a common phenotype in aging and cancer. Ann N Y Acad Sci 2004; 1019:260–264. 73. Taylor RW, Turnbull DM. Mitochondrial DNA mutations in human disease. Nat Rev Genet 2005; 6:389–402. 74. Herrmann PC, Herrmann EC. Oxygen metabolism and a potential role for cytochrome c oxidase in the Warburg effect. J Bioenerg Biomembr 2007; 39:247–250. 75. Baysal BE, Ferrell RE, Willett-Brozick JE, et al. Mutations in SDHD, a mitochondrial complex II gene, in hereditary paraganglioma. Science 2000; 287:848–851.

16


76. Habano W, Sugai T, Nakamura S, Uesugi N, Higuchi T, Terashima M, Horiuchi S. Reduced expression and loss of heterozygosity of the SDHD gene in colorectal and gastric cancer. Oncol Rep 2003; 10:1375–1380. 77. Neumann HP, Pawlu C, Peczkowska M, et al. Distinct clinical features of paraganglioma syndromes associated with SDHB and SDHD gene mutations. JAMA 2004; 292:943–951. 78. Deml E, Oesterle D, Wolff T, Greim H. Age-, sex-, and strain-dependent differences in the induction of enzyme-altered islands in rat liver by diethylnitrosamine. J Cancer Res Clin Oncol 1981; 100:125–134. 79. Racker E. Bioenergetics and the problem of tumor growth. Am Sci 1972; 60:56–63. 80. Racker E, Spector M. Warburg effect revisited: merger of biochemistry and molecular biology. Science 1981; 213:303–307. 81. Lehninger AL. Biochemistry. New York: Worth Publishers, 1970. 82. Shin YK, Yoo BC, Chang HJ, et al. Down-regulation of mitochondrial F1F0-ATP synthase in human colon cancer cells with induced 5-fluorouracil resistance. Cancer Res 2005; 65: 3162–3170. 83. Santamaria G, Martinez-Diez M, Fabregat I, Cuezva JM. Efficient execution of cell death in non-glycolytic cells requires the generation of ROS controlled by the activity of mitochondrial H+-ATP synthase. Carcinogenesis 2006; 27:925–935. 84. Zhou S, Kachhap S, Singh KK. Mitochondrial impairment in p53-deficient human cancer cells. Mutagenesis 2003; 18:287–292. 85. Ma W, Sung HJ, Park JY, Matoba S, Hwang PM. A pivotal role for p53: balancing aerobic respiration and glycolysis. J Bioenerg Biomembr 2007; 39:243–246. 86. Matoba S, Kang JG, Patino WD, et al. p53 regulates mitochondrial respiration. Science 2006; 312:1650–1653. 87. Kim JW, Tchernyshyov I, Semenza GL, Dang CV. HIF-1-mediated expression of pyruvate dehydrogenase kinase: a metabolic switch required for cellular adaptation to hypoxia. Cell Metab 2006; 3:177–185. 88. Papandreou I, Cairns RA, Fontana L, Lim AL, Denko NC. HIF-1 mediates adaptation to hypoxia by actively downregulating mitochondrial oxygen consumption. Cell Metab 2006; 3:187–197. 89. Pelicano H, Xu RH, Du M, et al. Mitochondrial respiration defects in cancer cells cause activation of Akt survival pathway through a redox-mediated mechanism. J Cell Biol 2006; 175:913–923. 90. Stiles B, Groszer M, Wang S, Jiao J, Wu H. PTENless means more. Dev Biol 2004; 273: 175–184. 91. Martinez-Diez M, Santamaria G, Ortega AD, Cuezva JM. Biogenesis and dynamics of mitochondria during the cell cycle: significance of 3 UTRs. PLoS ONE 2006; 1:e107. 92. Wang T, Marquardt C, Foker J. Aerobic glycolysis during lymphocyte proliferation. Nature 1976; 261:702–705. 93. Chen Z, Odstrcil EA, Tu BP, McKnight SL. Restriction of DNA replication to the reductive phase of the cell cycle protects genome integrity. Science 2007; 316:1916–1919. 94. Sakamaki T, Casimiro MC, Ju X, et al. Cyclin d1 determines mitochondrial function in vivo. Mol Cell Biol 2006; 26:5449–5469. 95. Wang C, Li Z, Lu Y, et al. Cyclin D1 repression of nuclear respiratory factor 1 integrates nuclear DNA synthesis and mitochondrial function. Proc Natl Acad Sci USA 2006; 103:11567–11572. 96. Luis AM, Izquierdo JM, Ostronoff LK, Salinas M, Santarén JF, Cuezva JM. Translational regulation of mitochondrial differentiation in neonatal rat liver. Specific increase in the translational efficiency of the nuclear-encoded mitochondrial beta-F1-ATPase mRNA. J Biol Chem 1993; 268:1868–1875. 97. Izquierdo JM, Ricart J, Ostronoff LK, Egea G, Cuezva JM. Changing patterns of transcriptional and post-transcriptional control of beta-F1-ATPase gene expression during mitochondrial biogenesis in liver. J Biol Chem 1995; 270:10342–10350.


17

98. Izquierdo JM, Cuezva JM. Control of the translational efficiency of beta-F1-ATPase mRNA depends on the regulation of a protein that binds the 3 untranslated region of the mRNA. Mol Cell Biol 1997; 17:5255–5268. 99. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000; 100:57–70. 100. Ristow M, Pfister MF, Yee AJ, et al. Frataxin activates mitochondrial energy conversion and oxidative phosphorylation. Proc Natl Acad Sci USA 2000; 97:12239–12243. 101. Schulz TJ, Thierbach R, Voigt A, et al. Induction of oxidative metabolism by mitochondrial frataxin inhibits cancer growth: Otto Warburg revisited. J Biol Chem 2006; 281:977–981. 102. Mühlenhoff U, Richhardt N, Ristow M, Kispal G, Lill R. The yeast frataxin homolog Yfh1p plays a specific role in the maturation of cellular Fe/S proteins. Hum Mol Genet 2002; 11:2025–2036. 103. Thierbach R, Schulz TJ, Isken F, et al. Targeted disruption of hepatic frataxin expression causes impaired mitochondrial function, decreased life span, and tumor growth in mice. Hum Mol Genet 2005; 14:3857–3864. 104. Bulavin DV, Fornace AJ Jr. p38 MAP kinase s emerging role as a tumor suppressor. Adv Cancer Res 2004; 92:95–118. 105. Timofeev O, Lee TY, Bulavin DV. A subtle change in p38 MAPK activity is sufficient to suppress in vivo tumorigenesis. Cell Cycle 2005; 4:118120. 106. Bi X, Lin Q, Foo TW, et al. Proteomic analysis of colorectal cancer reveals alterations in metabolic pathways: mechanism of tumorigenesis. Mol Cell Proteomics 2006; 5:1119–1130. 107. Dey R, Moraes CT. Lack of oxidative phosphorylation and low mitochondrial membrane potential decrease susceptibility to apoptosis and do not modulate the protective effect of Bcl-x(L) in osteosarcoma cells. J Biol Chem 2000; 275:7087–7094. 108. Kim JY, Kim YH, Chang I, et al. Resistance of mitochondrial DNA-deficient cells to TRAIL: role of Bax in TRAIL-induced apoptosis. Oncogene 2002; 21:3139–3148. 109. Park SY, Chang I, Kim JY, et al. Resistance of mitochondrial DNA-depleted cells against cell death: role of mitochondrial superoxide dismutase. J Biol Chem 2004; 279: 7512–7520. 110. Tomiyama A, Serizawa S, Tachibana K, et al. Critical role for mitochondrial oxidative phosphorylation in the activation of tumor suppressors Bax and Bak. J Natl Cancer Inst 2006; 98:1462–1473. 111. Matsuyama S, Xu Q, Velours J, Reed JC. The mitochondrial F0F1-ATPase proton pump is required for function of the proapoptotic protein Bax in yeast and mammalian cells. Mol Cell 1998; 1:327–336. 112. Gross A, Pilcher K, Blachly-Dyson E, et al. Biochemical and genetic analysis of the mitochondrial response of yeast to BAX and BCL-X(L). Mol Cell Biol 2000; 20:3125–3136. 113. Harris MH, Vander Heiden MG, Kron SJ, Thompson CB. Role of oxidative phosphorylation in Bax toxicity. Mol Cell Biol 2000; 20:3590–3596. 114. Plas DR, Thompson CB. Cell metabolism in the regulation of programmed cell death. Trends Endocrinol Metab 2002; 13:75–78. 115. Danial NN, Gramm CF, Scorrano L, et al. BAD and glucokinase reside in a mitochondrial complex that integrates glycolysis and apoptosis. Nature 2003; 424:952–956. 116. Azoulay-Zohar H, Israelson A, Abu-Hamad S, Shoshan-Barmatz V. In self-defence: hexokinase promotes voltage-dependent anion channel closure and prevents mitochondriamediated apoptotic cell death. Biochem J 2004; 377:347–355. 117. Vahsen N, Cande C, Briere JJ, et al. AIF deficiency compromises oxidative phosphorylation. EMBO J 2004; 23:4679–4689. 118. Kahn BB, Alquier T, Carling D, Hardie DG. AMP-activated protein kinase: ancient energy gauge provides clues to modern understanding of metabolism. Cell Metab 2005; 1:15–25. 119. Alessi DR, Sakamoto K, Bayascas JR. LKB1-dependent signaling pathways. Annu Rev Biochem 2006; 75:137–163. 120. Inoki K, Zhu T, Guan KL. TSC2 mediates cellular energy response to control cell growth and survival. Cell 2003; 115:577–590.

18


121. Reznick RM, Shulman GI. The role of AMP-activated protein kinase in mitochondrial biogenesis. J Physiol 2006; 574:33–39. 122. Shaw RJ. Glucose metabolism and cancer. Curr Opin Cell Biol 2006; 18:598–608. 123. Jones RG, Plas DR, Kubek S, et al. AMP-activated protein kinase induces a p53-dependent metabolic checkpoint. Mol Cell 2005; 18:283–293.

The Electron Transport Chain and Carcinogenesis Jean-Jacques Brière, Paule Bénit and Pierre Rustin

Abstract Major metabolic changes that affect the balance between respiration and glycolysis occur during carcinogenesis. It is therefore not surprising that it has long been suggested that abnormal activity of the mitochondrial electron transfer chain could play a role in the underlying pathophysiologic process. However, it has only recently been demonstrated that the specific impairment of one component of the electron transfer chain, namely complex II, can indeed be the primary event triggering carcinogenesis. Unexpectedly, rather than superoxide overproduction, the organic acid imbalance resulting from the complex II blockade appears to be instrumental in the early step of tumorigenesis. Nevertheless, because an abnormal handling of oxygen is frequently suspected of being associated with electron transfer dysfunction, a renewed interest is observed for the putative instability of the mitochondrial genome often reported in tumor tissues. In this chapter, we review and discuss the evidence advocated for the implication of the electron transfer chain in carcinogenesis. Keywords Carcinogenesis · Mitochondria · Organic acids · Superoxides · Electron transfer

1 Introduction Mitochondria play an irreplaceable role in all aerobic organisms (i.e., the handling of oxygen by the electron transfer chain for the mutual benefit of the cell host and of these Proteobacteria-derived organelles) [1]. In addition, through a fascinating and complex network of interactions, mitochondria have emerged as central executioners, or possibly even decision makers, for cell death under a large number of conditions [2]. In keeping with this paradigm, a set of mitochondria-releasable

P. Rustin (B) INSERM U676, Hôpital Robert Debré, 75019 Paris, France e-mail: [email protected] S.P. Apte, R. Sarangarajan (eds.), Cellular Respiration and Carcinogenesis, DOI 10.1007/978-1-59745-435-3 2, C Humana Press, a part of Springer Science+Business Media, LLC 2009

19

20

J.-J. Briére et al.

factors have been shown to be bifunctional proteins [3–5]. Involved in electron transfer chain activity or maintenance, both cytochrome c and apoptosis inducing factor (AIF) readily trigger apoptosis when released in the cytosol upon opening of the elusive mitochondrial permeability transition pore (MPTP) [6]. Furthermore, a severe shortage of mitochondrial ATP production, which presumably occurs when the electron transfer chain is deficient, is supposedly sufficient to cause cell necrosis [7]. However, more recently, it was discovered that respiratory chain complex II deficiency can favor cell proliferation and tumor formation rather than cell death [8]. This has led to a renewed interest in the potential role of respiratory chain dysfunction in tumor promotion.

2 Mitochondria and Cancer: A Long History Otto Warburg performed the seminal observation that, even under aerobic conditions, cancer cells shift their metabolism from an oxidative one (through the mitochondrial respiratory chain) to a highly glycolytic metabolism, thereby producing elevated levels of lactate [9]. This paradoxical effect, wherein oxygen is not utilized even though it is available, was thereafter referred to as the Warburg effect.This phenomenon is at variance with the Pasteur effect according to which glycolysis is naturally favored by conditions that turn off the oxygen-dependent mitochondrial respiratory chain. As a result of the shift of metabolic flux from respiration to glycolysis, the former would become of secondary importance, and this logically suggested the idea that the fate of the mitochondria in tumor tissues would be somewhat irrelevant to tumor development. Accordingly, up to very recently, the word cancer was indeed largely absent from mitochondrial disease–devoted meetings and reviews. Even the reports of decreased respiratory chain activities or of quantitative changes in mitochondrial DNA (mtDNA)-encoded mRNA in tumors were insufficient to really draw the attention of the mitochondria-oriented scientific community. However, several features of the mitochondrial electron transfer chain that could play a role in triggering, or participating in, tumorigenesis deserve to be presented.

3 The Respiratory Chain: Biochemical and Genetic Features The respiratory chain (RC) is constituted by an association of supercomplexes whose spatial organization favors a rapid exchange of electrons. Such a mechanism may serve to avoid—or at least to limit—direct reactions between reduced RC carriers and molecular oxygen that can produce superoxides [10]. Two major entities have been described: the respirasome consisting of the association of RC complexes I/III/IV and presumably a subfraction of cytochrome c; and the ATPasome, consisting of complex V plus the adenylate and the phosphate carriers. The

The Electron Transport Chain and Carcinogenesis

21

respirasome channels the electrons from respiratory substrates to molecular oxygen to form water with a concomitant proton extrusion from the mitochondrial matrix space to the intermembrane space (Fig. 1). This creates an electrochemical gradient across the inner membrane that can be further used by the ATPasome to generate ATP from ADP and inorganic phosphate. Additional processes make use of the electrochemical gradient, ensuring the transport of various components through the membrane, including various cations (e.g., calcium), anions (e.g., organic acids, fatty acids), or proteinaceous material (e.g., mitochondrial protein precursors) [1]. Dissipation of the electrochemical gradient can also be an aim per se, allowing the uncoupling of electron flow from the phosphorylation process [11]. This process liberates heat and mechanically favors the oxidized state of the RC

Fig. 1 The dual genetic origin and the organization of the respiratory chain. (A) Respiratory chain components are encoded by both the nuclear and the mitochondrial (mt) genomes. (B) A schematic view of the organization of the respirasome and the ATPasome. CI–CV, the various complexes of the respiratory chain; cyt c, cytochrome c; IM, inner membrane; Q: ubiquinone pools

22


components, thus reducing the chance to generate superoxides. UnCoupling Proteins (UCP), a large family of proteins, mediate this process, with a high tissue specificity. The physiologic function of a number of these UCPs is still a subject of debate [12]. The respirasome interacts with several other dehydrogenases (e.g., complex II, the electron transfer flavoprotein [ETF], the glycerol-3-phosphate and the dihydroorotate dehydrogenases) through the kinetically compartmentalized quinone pool in the inner mitochondrial membrane (Fig. 1) [12]. Depending on their specific metabolic demand, tissues are variably endowed with these dehydrogenases [13]. Redox status of the quinone pool therefore depends on the respective activities of the reducing enzymes (the various dehydrogenases) and of the oxidizing part of the RC (i.e., the cytochromic segment). This redox status is also important because it helps to reconstitute the reducing power of lipophilic antioxidants, such as tocopherol (vitamin E) [14]. This complex machinery constituted of about 100 proteins is encoded by both the mitochondrial genome (mtDNA) and the nuclear genome (Fig. 1) [15]. The maternally inherited mtDNA ranges from a few hundred (lymphocytes) to more than 100,000 (oocytes) copies per cell [16]. Each circular molecule consists of 16,569 base pairs with 37 genes encoding 13 proteins (polypeptides), 22 transfer RNA (tRNAs), and 2 ribosomal RNAs (rRNAs) [15]. In one cell, the mtDNA copies can be identical (a situation referred as cellular homoplasmy) or not (a situation known as heteroplasmy). In the particular case of deleterious mutations harbored by the mtDNA, a threshold between wild type (wt) and mutant mtDNA species has to be reached before an RC phenotype can be observed [17]. Because of the random distribution of mitochondria during cell division, an individual can have variable levels of heteroplasmy from tissue to tissue. The involvement of a given tissue containing mutant mtDNA will therefore depend on the actual level of heteroplasmy, the threshold value required for RC impairment, and the susceptibility of the tissue to the many consequences of an RC dysfunction (e.g., ATP shortage, superoxide overproduction, metabolic blockade, etc.). The remaining components of the RC are encoded by genes present in the nuclear genome. According to the theory of the symbiotic origin of the mitochondria, these genes have supposedly left the mitochondria to be inserted in the nuclear chromosomes during evolution [18]. Noticeably, nonfunctional copies of genes encoding mitochondrial proteins are frequently encountered in the nuclear genome [19]. This also stands true for genes still harbored by the mtDNA, a feature that should be carefully recorded when studying these genes [20]. Accordingly, the authenticity of the increased levels of mutant mtDNA frequently reported in tumor tissues has been recently questioned [21]. Finally, if as mentioned earlier, the RC does much more than to only transfer electrons and synthesize ATP, its functioning also requires much more than the hundreds of genes that encode its constituents. Indeed about 1500 genes on the 30,000 genes of the human genome are believed to be necessary for the sustained building, function and maintenance of mitochondrial functions.


23

4 Mutations in Three Nuclear Genes Encoding Respiratory Chain Complex II Can Promote Tumor Formation Mutations in three of the four genes encoding complex II have been shown to cause tumor formation and cancer. Mutations of succinate dehydrogenase (SDH) are frequently encountered in patients with an apparently sporadic form of paragangliomas (PGLs) or pheochromocytomas (PHEOs) [22]. PGLs are neuroendocrine tumors that may secrete catecholamines [23]. They occur most frequently in the head (glomus tympanicum and jugulare), neck (carotid body and glomus vagale), adrenal medulla, and extra-adrenal sympathetic ganglia. PGLs are generally benign, highly vascularized tumors occurring close to the major blood vessels and cranial nerves [24]. About 10% of these tumors are malignant. PGLs are usually diagnosed on the basis of the presence of a mass in the neck, pulsatile tinnitus, or hearing loss. According to current nomenclature, the term pheochromocytomas should be restricted to secreting PGLs evolving from the adrenal medulla, whereas a functional PGL refers to an extra-adrenal secreting PGL, which was previously known as an ectopic PHEO. The tumors now described as carotid body PGL were formerly known as chemodectoma due to the chemoreceptor function of the carotid body. PGLs seem to be inherited in about 30% of cases. The hereditary form of PGL is usually characterized by an early onset and a more severe presentation than that of the sporadic form. These tumors often display bilateral and multiple locations and may be recurrent or malignant. The presence of one or several secreting tumors is not rare and contributes to the severity of inherited PGL [23]. Such PGLs constitute a genetically heterogeneous group of diseases as four different loci have been implicated: PGL1 on 11q23, PGL2 on 11q13, PGL3 on 1q21, and PGL4 on 1p36. In 2000, linkage analysis and positional cloning allowed Baysal et al. [8] to report the first deleterious mutations in the SDHD gene, corresponding with the PGL1 locus. The use of a candidate gene approach has led to the identification of mutations in SDHC (PGL3) and SDHB (PGL4) [24, 25]. Several mutations in these three genes were subsequently reported in patients with hereditary PGL and in apparently sporadic cases [26]. These observations fueled a vigorous debate on the mechanism linking SDH deficiency to tumor formation. Based on previous findings on the potential consequences of a RC blockade, it has been suggested that superoxide overproduction may be at the origin of increased cell proliferation [27]. Accordingly, the mev1 mutant of the worm Caenorhabditis elegans defective in the cytochrome b subunit of CII was found to have a reduced lifespan that was attributed to overproduction of superoxides [28]. In the minds of many investigators, the superoxide overproduction provided a simple explanation linking RC defects to neoplastic transformation. However, no hyperplasia or indications of abnormal cell proliferation were reported in the mev1 mutant at variance with some other C. elegans mutants for proteins that are prone to trigger tumorigenesis when mutated in the homologous protein in mammals (e.g., cul1 mutant) [29]. Soon after the discovery that SDH mutations can result in tumor formation, it was shown in a family harboring a SDHD mutation that these tumors were highly

24


vascularized and that this corresponded with hypoxia-inducible factor (HIF) stabilization and the activation of the hypoxia pathway [30]. HIFs are composed of two subunits: an ␣ subunit, which is regulated by oxygen, and a ␤-subunit (HIF-1␤, also called the aryl hydrocarbon receptor nuclear translocator, or ARNT), which is constitutively and ubiquitously expressed. Three ␣-subunits have been identified to date (HIF-1␣, HIF-2␣, and HIF-3␣), each encoded by a different gene. Under normoxic conditions, HIF-␣ is continuously ubiquitinated and subsequently degraded by the proteasome [31]. The process of ubiquitination is started by their recognition by the von Hippel–Lindau (VHL) protein, which in turn requires the hydroxylation of two proline residues on HIF-␣ [32]. The very first step of HIF-␣ degradation under normoxic conditions is thus dependent on this hydroxylation, which is catalyzed by HIF prolyl hydroxylases (PHDs). PHDs belong to the superfamily of Fe(II)-dependent oxygenases and require reduced iron as a cofactor, alpha-ketoglutarate (␣-KG) and oxygen as cosubstrates, with carbon dioxide and succinate being the products of the reaction [33]. Under hypoxic conditions, the absence of oxygen prohibits PHD activity, and HIF-␣ is thus stabilized, allowing for its nuclear translocation and the subsequent activation of its target genes. The involvement of HIFs has been observed in numerous types of tumors, playing an active role in the progression of neoplasia [34]. Besides oxygen, others factors, which intervene in the reaction catalyzed by the PHD enzyme, can also affect its activity. Consistent with this hypothesis, it was shown that elevated levels of succinate, the product of the PHDs reaction, can result in the blockade of its activity with the consequent stabilization of the HIF-1␣ protein (Fig. 2) [35, 36]. A significant accumulation of succinate was found in SDH-deficient cells and tumors. Moreover, the addition of ␣-ketoglutarate, the substrate of the PHD,

Fig. 2 The interaction between the mitochondrion and type II hexokinase. The channeling of mitochondrial ATP by type II hexokinase favors the production of glucose 6-phosphate (G 6-P) and ultimately glycolysis. Ant, adenylate carrier; HKII, type II hexokinase; I–V, the various complexes of the respiratory chain; Q, ubiquinone; VDAC, voltage-dependent anion channel


25

was shown to abolish the nuclear translocation of HIF-1␣ in the nucleus of SDHdefective cells [35]. This provided an alternative explanation linking SDH deficiency to tumor formation and an explanation of the restriction of the induction of tumors to RC complex II blockade. Finally, the respiration of SDH-defective cells was shown to be not significantly decreased indicating that the electron flow to oxygen and the phosphorylation process are not significantly affected [37]. Additional support in favor of this latter hypothesis came from the observation that fumarase defect can lead to leiomyomatosis and renal cell cancer (HLRCC) syndrome [38]. In this latter case, fumarate, accumulated because of fumarase inactivation, was found to act as a competitive inhibitor of the PHD, thus inducing the abnormal stabilization of HIF-1␣. Noticeably, other structurally related organic acids can also inhibit PHD [39]. Therefore, a tricarboxylic acid cycle (TCA) cycle blockade may result in the induction of angiogenesis and in the upregulation of glycolysis during tumorigenesis, being at the origin of the Warburg effect. These observations support the view that rather than a blockade of the electron flow (potentially associated with all subtypes of RC defects), the impairment of a particular segment of the Krebs cycle can be at the origin of tumor formation [40]. It would therefore be at least rather imprudent to invoke SDH mutations as a general proof that a RC defect results in tumor formation.

5 Does a Defective Respiratory Chain Trigger Cell Proliferation? To date, there is no conclusive evidence that a perturbation of the electron flow through the RC is sufficient to increase cell proliferation and to constitute the primary cause of tumor formation. On one hand, none of the nuclear genes encoding RC components or involved in the building or maintenance of the chain have been demonstrated to be tumor suppressors, with the exception of genes encoding RC complex II (or fumarase; see above) [41]. Noticeably, the accumulation of ␣-ketoglutarate is a feature frequently associated with RC dysfunction, which will not favor HIF-1␣ stabilization through PHD inhibition. On the other hand, RC-specific poisons are not known as carcinogenic (e.g., the complex I–specific inhibitor rotenone shown to possibly induce parkinsonism in mammals). Finally, patients harboring deleterious mutation in genes encoding RC components are not known to be particularly at risk for tumor formation. This is even true for mutations affecting RC proteins such as ATPase (complex V) components that result in high levels of superoxide production [42]. These observations lend less credibility to the idea that an impairment of the RC function could per se be at the origin of cell proliferation and tumor formation. Even more puzzling is the fact that in a subset of tumor cells (liver and pancreatic tumors), the enhanced glycolysis might well require an active production of ATP by the RC. The hexokinase specifically expressed in most tumor cells uses both ATP and glucose to produce ADP and glucose-6-phosphate. This tumor-specific hexokinase is characterized by its poor affinity (high Km , about 10 mM) for glucose, similar to the hexokinase IV (glucokinase) found in normal hepatocytes [43].

26


Fig. 3 Induction of the hypoxia path by succinate through HIF-1␣ stabilization. Upon succinate dehydrogenase (complex I) blockade, succinate accumulates and is exported to the cytosol. There, it inhibits the prolyl hydroxylase (PHD) triggering HIF-1␣ stabilization. The nuclear translocation of this latter factor induces an increased transcription of the hypoxia pathway components. VHL, von Hippel–Lindau; I–V, the various respiratory chain complexes


27

It possibly allows for an additional capacity for glucose uptake from plasma and increased glucose phosphorylation by displacement of the cell glucose equilibrium. Interestingly, whereas the isoform with a low affinity is expressed in most tumor cells, in the tumoral liver, similarly to tumoral pancreatic cells, it is largely replaced by a high-affinity form (hexokinase II; HKII). This latter form can readily bind voltage-dependent anion channel (VDAC) at the outer mitochondrial membrane and utilize the mitochondrial ATP to produce glucose-6-phosphate thus favoring an aerobic glycolysis [44] so long as the electron transfer chain is functional (Fig. 3). Any significant impairment of RC activity would then tend to decrease the ATP produced by the RC and oppose the upregulation of glycolysis. In addition, it has been suggested that, upon binding to VDAC, HKII would be protected from product inhibition because of a conformational change and in turn may reduce the chance of VDAC to open and to facilitate release of proapoptotic factors [43, 45]. In this case, not only would carcinogenesis not be triggered by mitochondrial dysfunction, but also it would require the preservation of the respiratory chain function. Finally, there is ample evidence that an impairment of RC function under most conditions leads to cell death, tissue necrosis, and the related wide range of clinical symptoms observed in patients harboring mutations in genes necessary for the synthesis of the electron transfer chain [46]. Potential overproduction of superoxides has been shown to be responsible for triggering apoptotic cell death, even if low amount of superoxides might be required for normal cell division, and can target various cell cycle checkpoints. It therefore appears that a defective RC is probably not per se a primary cause of tumor formation. On the contrary, it could in some cases even constitute an obstacle to this process!

6 Mitochondria, Superoxides, Hypoxia The possibility nevertheless exists that low oxygenation in tumors may secondarily affect mitochondrial functioning favoring the formation of superoxides and peroxides by the RC (Fig. 4). In principle, activated oxygen species may in turn upregulate both oncogene growth factors and their tyrosine kinase receptors thus driving cell transformation [47], simultaneously promoting HIF-1␣ stabilization by inhibiting prolyl hydroxylase. In any event, superoxide production that may promote tumorigenesis would have to be restricted to a very narrow window of concentration, large overproduction readily favoring cell death rather than proliferation. The suggested role of superoxides in triggering tumorigenesis has long been advocated in support for a therapeutic use of antioxidants [48]. Actually, a wide range of antioxidant molecules is available that can potentially interfere with cell free radicals, possibly released by mitochondria. However, contrasting results from in vitro and in vivo (intervention trials) experiments have raised some doubt about the ability of antioxidants to actually fight cancer by such a mechanism. Indeed, most antioxidant molecules also act as prooxidants; this possibly accounts

28


Fig. 4 Linking electron transfer chain to cell death and proliferation. (A) According to a first hypothesis, the level of mitochondrial superoxide production largely determines the fate of the cells. A low superoxide production favors cell division, whereas increasing this production triggers dysplastic lesion and tumor formation. An excessive production rather results in cell death. (B) According to this second scheme, the blockade of the prolyl hydroxylase by organic acids (succinate, fumarase) results in HIF-1␣ stabilization and the activation of the hypoxia path allowing for cell proliferation. Release of mitochondrial proapoptotic factors (AIF, cytochrome c) and overproduction of superoxide (blockade of the ATPase; complex V) result in cell death

for their potential antitumoral activity. For instance, an antioxidant molecule like melatonin would exercise its antiproliferative effect on the growth of rat pituitary prolactin-secreting tumor cells in vitro by damaging mitochondria rather than by quenching superoxides [49]. Thus even if an increased superoxide production is observed in a subset of cancer cells, more evidence is needed to establish that, as a general rule, these superoxides are instrumental in triggering tumorigenesis. Recent studies however have established that overexpression of thioredoxin, manganese-dependent superoxide dismutase (MnSOD), glutathione peroxidase, or catalase all tend to reduce cell proliferation in different animal or cell models


29

suggesting that handling of superoxides or their derivatives might be critical at some point in a subset of cases [50, 51] (see, however, Refs. 52, 53].

7 Targeting the Respiratory Chain to Kill Tumors Although it is not clear that a defective respiratory chain actually favors tumor formation, mitochondria might represent the Achilles’ heel of cancer cells (Fig. 5). Indeed, mitochondria house several proapoptotic factors that are simultaneously

Fig. 5 Mitochondria and apoptosis. Release of intermembrane space components (such as AIF or cytochrome c) and/or the loss of membrane potential (⌬⌿) under the effect of numerous stimuli trigger cell commitment to die through an apoptotic process. Loss of membrane potential and of various matrix cofactors can result from the opening of the mitochondrial permeability transition pore (PTP). The balance between the proapoptotic and antiapoptotic members of the Bcl-2 family controls the opening of the pore. Alternatively, the members of this family can form channels that may also allow for the release of proapoptotic components present in the intermembrane space. PTP would be a complex formed between the voltage-dependent anion channel (VDAC; outer membrane) and the adenylate carrier (ANT; inner membrane), associated with several additional proteins. AIF, apoptosis inducing factor; Br, benzodiazepine receptor; CI–CV, the various complexes of the respiratory chain; c, cytochrome c; IM, inner membrane; OM, outer membrane

30


components of (or closely associated with) the electron transfer chain, and targeting tumors with reagents that induce release of these components has became a fashionable idea [54]. Accordingly, cisplatin, one of the most important chemotherapeutic agents ever developed, has been shown to readily interact with mitochondria to trigger apoptosis [55]. Resveratrol, a natural polyphenolic antioxidant, has been reported to possess a cancer chemopreventive potential that has been attributed to its ability to trigger mitochondrial dysfunction and cell apoptosis [56]. Actually, an exploding list of promising therapeutic and preventative agents is being shown (or has been suggested) to actually act through a quite similar mechanism, with or without the involvement of superoxide production.

8 Conclusion Mutations in both nuclear and mitochondrial genes encoding respiratory chain components have been suggested to be causative for triggering cancer. However, despite the numerous mtDNA mutations reported in many tumors, only germ-line mutations in the complex II–encoding genes (SDHB,C,D) have been conclusively shown as the primary cause of tumor formation. Interestingly enough, no inherited mtDNA mutation has been observed to cause tumor to date, contrasting the high frequency of mutant mtDNA in various type of tumors (see, however, Ref. 21). Although these may have a role in the oncogenic process at some point, they may be a secondary event totally irrelevant for this process. As a general rule, it seems that targeting respiratory chain function, except for complex II activity, provides a mechanism to inhibit rather than to promote tumor formation.

References 1. Tzagoloff A. Mitochondria. New York: Plenum Press, 1982. 2. Green DR, Kroemer G. The pathophysiology of mitochondrial cell death. Science 2004; 305(5684):626–629. 3. Krippner A, Matsuno-Yagi A, Gottlieb RA, Babior BM. Loss of function of cytochrome c in Jurkat cells undergoing fas-mediated apoptosis. J Biol Chem 1996; 271(35):21629–21636. 4. Green DR, Reed JC. Mitochondria and apoptosis. Science 1998;281(5381):1309–1312. 5. Vahsen N, Cande C, Briere JJ, et al. AIF deficiency compromises oxidative phosphorylation. EMBO J 2004; 23(23):4679–4689. 6. Garrido C, Kroemer G. Life s smile, death s grin: vital functions of apoptosis-executing proteins. Curr Opin Cell Biol 2004; 16(6):639–646. 7. Golstein P, Kroemer G. Cell death by necrosis: towards a molecular definition. Trends Biochem Sci 2007; 32(1):37–43. 8. Baysal BE, Ferrell RE, Willett-Brozick JE, et al. Mutations in SDHD, a mitochondrial complex II gene, in hereditary paraganglioma. Science 2000; 287(5454):848–851. 9. Warburg O, Wind F, Negelein E. The metabolism of tumors in the body. J. General Physiol 1926; 8(6):519. 10. Schagger H, Pfeiffer K. Supercomplexes in the respiratory chains of yeast and mammalian mitochondria. EMBO J 2000; 19(8):1777–1783. 11. Pecqueur C, Couplan E, Bouillaud F, Ricquier D. Genetic and physiological analysis of the role of uncoupling proteins in human energy homeostasis. J Mol Med 2001; 79(1):48–56.


31

12. Gutman M. Electron flux through the mitochondrial ubiquinone. Biochim Biophys Acta 1980; 594(1):53–84. 13. Pette D, Klingenberg M, Buecher T. Comparable and specific proportions in the mitochondrial enzyme activity pattern. Biochem Biophys Res Commun 1962; 7:425–429. 14. Geromel V, Darin N, Chretien D, et al. Coenzyme Q(10) and idebenone in the therapy of respiratory chain diseases: rationale and comparative benefits. Mol Genet Metab 2002; 77(1–2):21–30. 15. Larsson NG, Clayton DA. Molecular genetic aspects of human mitochondrial disorders. Annu Rev Genet 1995; 29:151–178. 16. Chen X, Prosser R, Simonetti S, Sadlock J, Jagiello G, Schon EA. Rearranged mitochondrial genomes are present in human oocytes. Am J Hum Genet 1995; 57(2):239–247. 17. Bourgeron T, Chretien D, Rotig A, Munnich A, Rustin P. Fate and expression of the deleted mitochondrial DNA differ between human heteroplasmic skin fibroblast and Epstein-Barr virus-transformed lymphocyte cultures. J Biol Chem 1993; 268(26):19369–19376. 18. Rustin P, Jacobs HT, Dietrich A, Lightowlers RN, Tarassov I, Corral-Debrinski M. Targeting allotopic material to the mitochondrial compartment: new tools for better understanding mitochondrial physiology and prospect for therapy. Med Sci (Paris) 2007; 23(5):519–525. 19. Ricchetti M, Tekaia F, Dujon B. Continued colonization of the human genome by mitochondrial DNA. PLoS Biol 2004; 2(9):E273. 20. Parfait B, Rustin P, Munnich A, Rotig A. Co-amplification of nuclear pseudogenes and assessment of heteroplasmy of mitochondrial DNA mutations. Biochem Biophys Res Commun 1998; 247(1):57–59. 21. Salas A, Yao YG, Macaulay V, Vega A, Carracedo A, Bandelt HJ. A critical reassessment of the role of mitochondria in tumorigenesis. PLoS Med 2005; 2(11):e296. 22. Briere JJ, Favier J, Gimenez-Roqueplo AP, Rustin P. Tricarboxylic acid cycle dysfunction as a cause of human diseases and tumour formation. Am J Physiol Cell Physiol 2006; 291(6): 1114–1120. 23. Favier J, Briere JJ, Strompf L, et al. Hereditary paraganglioma/pheochromocytoma and inherited succinate dehydrogenase deficiency. Horm Res 2005; 63(4):171–179. 24. Astuti D, Latif F, Dallol A, et al. Gene mutations in the succinate dehydrogenase subunit SDHB cause susceptibility to familial pheochromocytoma and to familial paraganglioma. Am J Hum Genet 2001; 69(1):49–54. 25. Niemann S, Muller U. Mutations in SDHC cause autosomal dominant paraganglioma, type 3. Nat Genet 2000; 26(3):268–270. 26. Nakamura E, Kaelin WG Jr. Recent insights into the molecular pathogenesis of pheochromocytoma and paraganglioma. Endocr Pathol 2006; 17(2):97–106. 27. Rustin P. Mitochondria, from cell death to proliferation. Nat Genet 2002; 30(4):352–353. 28. Senoo-Matsuda N, Yasuda K, Tsuda M, et al. A defect in the cytochrome b large subunit in complex II causes both superoxide anion overproduction and abnormal energy metabolism in Caenorhabditis elegans. J Biol Chem 2001; 276(45):41553–41558. 29. Piva R, Liu J, Chiarle R, Podda A, Pagano M, Inghirami G. In vivo interference with Skp1 function leads to genetic instability and neoplastic transformation. Mol Cell Biol 2002; 22(23):8375–8387. 30. Gimenez-Roqueplo AP, Favier J, Rustin P, et al. The R22X mutation of the SDHD gene in hereditary paraganglioma abolishes the enzymatic activity of complex II in the mitochondrial respiratory chain and activates the hypoxia pathway. Am J Hum Genet 2001; 69(6):1186–1197. 31. Hickey MM, Simon MC. Regulation of angiogenesis by hypoxia and hypoxia-inducible factors. Curr Top Dev Biol 2006; 76:217–257. 32. Kaelin WG. Proline hydroxylation and gene expression. Annu Rev Biochem 2005;74: 115–128. 33. Lee JW, Bae SH, Jeong JW, Kim SH, Kim KW. Hypoxia-inducible factor (HIF-1)alpha: its protein stability and biological functions. Exp Mol Med 2004; 36(1):1–12. 34. Gordan JD, Simon MC. Hypoxia-inducible factors: central regulators of the tumor phenotype. Curr Opin Genet Dev 2007; 17(1):71–77.

32


35. Briere JJ, Favier J, Benit P, et al. Mitochondrial succinate is instrumental for HIF1alpha nuclear translocation in SDHA-mutant fibroblasts under normoxic conditions. Hum Mol Genet 2005; 14(21):3263–3269. 36. Selak MA, Armour SM, MacKenzie ED, et al. Succinate links TCA cycle dysfunction to oncogenesis by inhibiting HIF-alpha prolyl hydroxylase. Cancer Cell 2005; 7(1):77–85. 37. Bourgeron T, Rustin P, Chretien D, et al. Mutation of a nuclear succinate dehydrogenase gene results in mitochondrial respiratory chain deficiency. Nat Genet 1995; 11(2):144–1449. 38. Tomlinson IP, Alam NA, Rowan AJ, et al. Germline mutations in FH predispose to dominantly inherited uterine fibroids, skin leiomyomata and papillary renal cell cancer. Nat Genet 2002; 30(4):406–410. 39. MacKenzie ED, Selak MA, Tennant DA, et al. Cell-permeating alpha-ketoglutarate derivatives alleviate pseudohypoxia in succinate dehydrogenase-deficient cells. Mol Cell Biol 2007; 27(9):3282–3289. 40. Briere JJ, Favier J, El Ghouzzi V, et al. Succinate dehydrogenase deficiency in human. Cell Mol Life Sci 2005; 62(19–20):2317–2324. 41. Kroemer G. Mitochondria in cancer. Oncogene 2006; 25(34):4630–4632. 42. Geromel V, Kadhom N, Cebalos-Picot I, et al. Superoxide-induced massive apoptosis in cultured skin fibroblasts harboring the neurogenic ataxia retinitis pigmentosa (NARP) mutation in the ATPase-6 gene of the mitochondrial DNA. Hum Mol Genet 2001; 10(11):1221–1228. 43. Brandon M, Baldi P, Wallace DC. Mitochondrial mutations in cancer. Oncogene 2006; 25(34):4647–4662. 44. Bustamante E, Pedersen PL. High aerobic glycolysis of rat hepatoma cells in culture: role of mitochondrial hexokinase. Proc Natl Acad Sci USA 1977; 74(9):3735–3739. 45. Mathupala SP, Ko YH, Pedersen PL. Hexokinase II: cancer s double-edged sword acting as both facilitator and gatekeeper of malignancy when bound to mitochondria. Oncogene 2006; 25(34):4777–4786. 46. Munnich A, Rustin P. Clinical spectrum and diagnosis of mitochondrial disorders. Am J Med Genet 2001; 106(1):4–17. 47. Aslan M, Ozben T. Oxidants in receptor tyrosine kinase signal transduction pathways. Antioxid Redox Signal 2003; 5(6):781–788. 48. Nishikawa M, Hashida M. Inhibition of tumour metastasis by targeted delivery of antioxidant enzymes. Expert Opin Drug Deliv 2006; 3(3):355–369. 49. Yang QH, Xu JN, Xu RK, Pang SF. Antiproliferative effects of melatonin on the growth of rat pituitary prolactin-secreting tumor cells in vitro. J Pineal Res 2007; 42(2):172–179. 50. Weydert CJ, Waugh TA, Ritchie JM, et al. Overexpression of manganese or copper-zinc superoxide dismutase inhibits breast cancer growth. Free Radic Biol Med 2006; 41(2):226–237. 51. Li GX, Hirabayashi Y, Yoon BI, et al. Thioredoxin overexpression in mice, model of attenuation of oxidative stress, prevents benzene-induced hemato-lymphoid toxicity and thymic lymphoma. Exp Hematol 2006; 34(12):1687–1697. 52. Lu YP, Lou YR, Yen P, et al. Enhanced skin carcinogenesis in transgenic mice with high expression of glutathione peroxidase or both glutathione peroxidase and superoxide dismutase. Cancer Res 1997; 57(8):1468–1474. 53. Welsh SJ, Bellamy WT, Briehl MM, Powis G. The redox protein thioredoxin-1 (Trx-1) increases hypoxia-inducible factor 1alpha protein expression: Trx-1 overexpression results in increased vascular endothelial growth factor production and enhanced tumor angiogenesis. Cancer Res 2002; 62(17):5089–5095. 54. Galluzzi L, Larochette N, Zamzami N, Kroemer G. Mitochondria as therapeutic targets for cancer chemotherapy. Oncogene 2006; 25(34):4812–4830. 55. Cepeda V, Fuertes MA, Castilla J, Alonso C, Quevedo C, Perez JM. Biochemical mechanisms of cisplatin cytotoxicity. Anticancer Agents Med Chem 2007; 7(1):3–18. 56. Fulda S, Debatin KM. Resveratrol modulation of signal transduction in apoptosis and cell survival: a mini-review. Cancer Detect Prev 2006; 30(3):217–223.

Respiratory Control of Redox Signaling and Cancer Pauline M. Carrico, Nadine Hempel and J. Andrés Melendez

Abstract In the past several decades, elegant work has been performed in a variety of organisms suggesting that decreases in oxidant burden can prolong life span and decrease the severity of age-associated degenerative disorders (Finkel T, Holbrook NJ. Nature 2000; 408:239–247). Cancer is primarily an age-related disease and is associated with the altered production of oxidants (Cerutti PA. Science 1985; 227:375–381). Compounds that generate reactive oxygen species (ROS) promote skin tumors in mice. Treatment with antioxidants that terminate the chain reactions initiated by ROS antagonizes this process. Many tumor cell types have also been shown to have an increased oxidant radical load relative to normal tissue. Thus, tumor cells have adapted to cope with an increased oxidant burden compared with that of normal nonmalignant tissue and may even harness the increased oxidant load to modulate signaling pathways that are critical for their survival. With the tight connection between oxidant production and tumor promotion and growth, it is quite surprising that antioxidant-based therapeutics have not become pervasive in treatments for many cancers (for a concise review on this topic, see Seifried HE, et al. Cancer Research 2003; 63:4295–4298). Hydrogen peroxide (H2 O2 ) is the electron-neutral, 2e- reduction product of oxygen that is generated both chemically and enzymatically and can modulate the activity of a variety of signaling molecules. This chapter will summarize how the metabolic production of H2 O2 can impact distinct signaling pathways that are essential for successful tumorigenesis and metastatic progression and will identify redoxsensitive pathways that may be targets for antioxidant-based cancer therapy. Keywords Reactive oxygen species · Cancer · Antioxidants · Redox signaling · Hydrogen peroxide

J.A. Melendez (B) Center for Immunology and Microbial Disease MC151, Albany Medical College, Albany, NY 12208 e-mail: [email protected] S.P. Apte, R. Sarangarajan (eds.), Cellular Respiration and Carcinogenesis, DOI 10.1007/978-1-59745-435-3 3, C Humana Press, a part of Springer Science+Business Media, LLC 2009

33

34

P.M. Carrico et al.

1 ROS and Redox Signaling All aerobic organisms generate ROS as by-products of normal cellular metabolism. Although the concentrations of ROS are normally low but measurable, accumulation of these species can be toxic to the cells. To protect themselves from the deleterious effects of ROS, cells have evolved an impressive repertoire of enzymatic and nonenzymatic antioxidant defenses. In normal cells, a balance between the rates of ROS production and the scavenging capacity by antioxidants is required to maintain the cellular redox homeostasis. An imbalance resulting from either an increase in ROS concentration or a decrease in the antioxidant capacity leads to the modulation of redox-dependent signaling events. ROS production can arise from a variety of sources in nonphagocytic cells. Most ROS are produced as by-products of the electron transport chain of the mitochondria at the level of complex I (NADH/ubiquinone oxidoreductase) and complex III (ubiquinol/cytochrome c oxidoreductase). Outside the mitochondria, ROS can originate from enzymes such as NADPH oxidase, dual oxidases, and lipoxygenases. At the cell surface, ROS production can be initiated after a variety of stimuli, such as integrin signaling, growth factors, and cytokines, such as TNF-␣, which activate NADPH oxidase. ROS can also arise from arachidonic acid metabolism via 5-lipoxygenase. Both cytoplasmic and mitochondrial pathways create the highly unstable superoxide anion (O2 -. ), which is either chemically or enzymatically converted to H2 O2 . The enzyme responsible for this dismutation is superoxide dismutase, isoforms of which can be found in the cytoplasm (copper zinc/sod1) or the mitochondria (manganese /sod2). Conditions where mitochondrial-derived ROS are produced at a rate that exceeds their scavenging capacity are commonly associated with metabolic perturbations that impact signaling. Recent evidence also suggests that ROS produced as a result of normal metabolic flux affect signaling. For example, Nemoto et al. [1] demonstrated that the addition of extracellular pyruvate to HeLa cells resulted in a dramatic increase in the phosphorylation/activation of JNK1, a modest increase in ERK activity, and no effect on JNK2 or p38. This activation was accompanied by an increase in mitochondrial oxidants and was reversible by the addition of N-acetylcysteine (NAC), overexpression of glutathione S-transferase, or extracellular catalase. These data suggest that the addition of pyruvate results in an increase in H2 O2 levels and the subsequent phosphorylation of JNK1. Furthermore, the activation of JNK1 coincided with the inhibition of glycogen synthase kinase 3-␤ and glycogen synthase. This work provided convincing evidence that ROS produced as a result of normal cellular metabolism could act as a regulator of signal transduction. The principle mediator of ROS-dependent signaling is H2 O2 . H2 O2 generated in response to receptor stimulation has been shown to be an efficient signal transducing molecule by its ability to reversibly oxidize active site cysteines [2]. Many protein tyrosine phosphatases (PTPs) are particularly susceptible to H2 O2 -dependent inactivation because of the lowered pKa of the active site cysteine. The essential cysteine residue in the PTP signature active-site motif Cys-(X)5 -Arg exists as a thiolate anion (Cys-S- ), which, at neutral pH, is susceptible to nucleophilic attack by H2 O2 .

Respiratory Control of Redox Signaling and Cancer

35

Oxidation of the active-site cysteine generates a sulfenic derivative (Cys-SOH) leading to enzyme inactivation that can be reversed by cellular thiols. This mode of inactivation is similar to that described for both bacterial and eukaryotic peroxiredoxin family members that are responsible for detoxification of organic hydroperoxides, H2 O2 and ONOO- [3]. A unique feature of the eukaryotic peroxiredoxins (PRXs) is their sensitivity to inactivation by H2 O2 levels that exceed steady state. This sensitivity to inactivation has led to the “floodgate” hypothesis, whereby PRX inactivation can lead to a rapid rising burst of signaling H2 O2 , which may lead to PTP inactivation and subsequent enhancement of kinase signaling [4]. Several PTP family members have been shown to be oxidatively inactivated by physiologic stimuli in vivo including the dual lipid protein phosphatase and tensin homolog deleted on chromosome 10 (PTEN), the low-molecular-weight PTPs (LMW-PTPs), and the mitogen activated protein (MAP) kinase phosphatase family members [5]. Such redox-based signaling, arising from a chronic shift in ROS homeostasis that results from metabolic changes or altered antioxidant scavenging, can impact critical components of the malignant phenotype. Tumor cell proliferation, survival, vascular recruitment, and migration are components that drive the malignant phenotype. Below we discuss how alterations in metabolic ROS production can regulate these mechanisms.

2 Mitochondrial ROS and Cell Cycle ROS have been shown to be potent mitogens. Effective detoxification of H2 O2 by catalase inhibits proliferation in Her-2/neu–transformed Rat-1 fibroblasts. Similarly, addition of exogenous H2 O2 at low concentrations and its intracellular production by the noninflammatory NADPH oxidase family members (Nox1-5) is also mitogenic [6]. H2 O2 may directly induce cell proliferation, however, it also acts as a downstream effector of platelet-derived growth factor (PDGF)-induced mitogenic signaling. In Rat-1 cells treated with PDGF, the SH2 domain–containing PTP, SHP-2, can be oxidatively inactivated, resulting in the phosphorylation and activation of MAPK. Both the inactivation of SHP-2 and resulting activation of MAPK is abrogated by treatment of cells with the H2 O2 scavenger NAC, demonstrating a regulatory role of H2 O2 in cell proliferation through the reversible oxidation of PTPs [7]. In this context, antioxidant strategies can attenuate the proliferative capacity of numerous tumor cell lines [8]. H2 O2 detoxification by the use of adenoviral catalase can effectively inhibit DNA synthesis and decrease the proliferation of human aortic endothelial cells. Goswami and co-workers have demonstrated that intracellular H2 O2 is also required for G1-S transit in mouse embryonic fibroblasts. Thus, in contrast with the growth inhibition commonly associated with the bolus addition of H2 O2 , the intracellular production of submicromolar concentrations of H2 O2 is mitogenic and stimulates proliferation, and its effective removal can prevent cell cycle progression. It is likely that the intracellular redox state fluctuates as a result of alterations in production or removal of mitochondrial-derived ROS. Menon and

36

P.M. Carrico et al.

Goswami have proposed that periodic fluctuations in metabolic ROS production likely control cell proliferation via redox-sensitive cell cycle regulatory proteins [9]. The rapid turnover of antioxidants may give rise to acute shifts in cellular oxidant production that can control normal cell cycle progression. The elucidation of oxidant scavenging or oxidant generating enzymes whose levels fluctuate with respect to cell cycle would give credence to this hypothesis. During the process of carcinogenesis, a similar response may occur as a result of the loss of tumor suppressor function or oncogene activation. Thus, the enhanced proliferative capacity of tumor cells may be attributed to chronic shifts in respiratory ROS production due to perturbations in metabolic or antioxidant function.

3 Mitochondrial ROS and Cell Survival H2 O2 has been implicated as a strong inducer of apoptosis in various cell types. Apoptosis or programmed cell death refers to a series of tightly coordinated events designed to eliminate potentially dangerous cells and cells that have reached the end of their life cycle. Apoptosis allows an organism to tightly control cell number and tissue size, as well as to protect itself from cells that interfere with homeostasis. Apoptosis is a tightly regulated process whereby a set of cysteine proteases (caspases) are made active through a complex signaling cascade resulting in degradation of cellular nuclear DNA [10]. In contrast with necrosis, apoptosis destroys only the cell in question with no effect on surrounding cells. Apoptosis can be triggered by both extracellular stimuli as well as intracellular events that mark cellular dysfunction. The hallmarks of apoptosis include membrane blebbing, cytoskeletal disorganization, and loss of mitochondrial integrity and DNA fragmentation. The ultimate step in apoptosis is the activation of caspases, which act on a number of downstream cellular proteins inactivating them by catalytic cleavage at an Asp-Xxx sequence. The pro/inactive forms of these enzymes are activated via cleavage by other upstream caspases or via autoactivation. Caspases are involved in the degradation of important cellular proteins as well as activation of caspase-activated DNAses that cleave mitochondrial DNA, ultimately leading to cell death. Neoplastic cells are able to evade apoptosis and circumvent cell death, an essential survival step during tumorigenesis. TNF-␣ is produced by cells under stress and induces signaling via binding to its cell surface receptor. TNF-␣ along with cycloheximide (CHX) has been established as a potent inducer of apoptosis in various tumor cell lines. One of the primary mechanisms of TNF-␣/CHX–induced apoptosis involves activation of the receptorassociated death domains that recruit and activate caspase-8, leading to activation of downstream caspases and finally of the executioner caspase, caspase-3. In addition, CHX enhances the induction of apoptosis via TNF-␣ by inhibition of new protein synthesis. One of the proteins that was first discovered to be essential for protection from TNF-mediated cell death is the mitochondrial manganese containing superoxide dismutase (Sod2) [11]. Since this initial observation by Wong and


37

Goeddel, numerous reports have demonstrated a role for Sod2 in protection from many apoptotic stimuli. The antiapoptotic effects of Sod2 suggest that either the removal of O2 -. or the generation of H2 O2 via Sod2 may have an inhibitory effect on programmed cell death. The mitochondrial generation of H2 O2 has been shown to confer resistance to apoptotic stimuli. Procaspases contain an active site cysteine that is susceptible to oxidation thereby preventing their activation via proteolytic cleavage. This may explain the ability of oxidants in some studies to deter apoptotic cell death. Cederbaum and Bai have demonstrated that catalase overexpression in HepG2 hepatocarcinoma cells increases their susceptibility to TNF-␣–induced apoptosis. Fas-induced apoptotic cell death can be enhanced by overexpression of mitochondrial-targeted catalase. The antioxidant flavonoid and terpenoid compounds derived from the Ginkgo biloba plant have been shown to enhance the expression of proapoptotic molecules as a result of its ability to scavenge H2 O2 in distinct cancer cell types. Recent studies from this lab indicate that efficient and mitochondrial-targeted H2 O2 detoxification can reverse the antiapoptotic properties of Sod2 overexpression [12]. Another mechanism whereby H2 O2 may augment cell survival is via oxidative inactivation of the LMW-PTPs. LMW-PTP activity is dependent upon the tyrosine phosphorylation on Tyr131 and Tyr132; phosphorylation on Tyr131 results in a 25fold increase in enzymatic activity, and phosphorylation of Tyr132 serves as a scaffold for the SH2 domain of the Grb2 adaptor protein. In PDGF-treated NIH-3T3 or human prostatic carcinoma PC3 cells, inactivation of LMW-PTP by glucose oxidase (G/O)-induced oxidative stress inhibits LMW-PTP auto-dephosphorylation on Tyr131 and Tyr132, thus enhancing the recruitment of Grb2 to LMW-PTP and, subsequently, ERK activation and cell survival [13]. Thus, even though it has been observed that bolus additions of H2 O2 induce apoptosis, these studies have established that submicromolar concentrations of mitochondrial-derived H2 O2 may provide a selective inhibition of programmed cell death and that effective H2 O2 detoxification may also sensitize cancer cells to apoptotic stimuli.

4 Mitochondrial ROS and Angiogenesis H2 O2 has been shown to regulate proangiogenic responses in a variety of systems including human retinal pigment epithelial cells, cultured keratinocytes, and bovine pulmonary artery endothelial cells. Monte et al. have demonstrated that H2 O2 contributes to promoting the angiogenic activity of tumor-bearing lymphocytes and that this activity can be inhibited by coadministration of H2 O2 -detoxifying enzyme catalase but not superoxide dismutase. In vitro angiogenesis of bovine thoracic aorta can be induced with relatively low concentrations (1 ␮M) of H2 O2 and blocked by catalase. The discovery of a noninflammatory NADPH oxidase (Nox1-5) has provided an additional source for the production of O2 -. and H2 O2 in a variety of tumor cell

38

P.M. Carrico et al.

types [14]. The Nox1-dependent generation of H2 O2 is a potent trigger of angiogenesis, increasing the vascularity of tumors and inducing molecular markers of angiogenesis including vascular endothelial growth factor (VEGF), VEGF receptors, and matrix metalloproteinase (MMP) activity in cultured cells and in tumors. Coexpression of catalase blocks the increased activity of the angiogenic markers, indicating that H2 O2 signals part of the switch to the angiogenic phenotype. Furthermore, MMPs are important contributors to the angiogenic switch, and their expression is redox-responsive [15]. The transcription factor Ets-1 is also an important regulator of angiogenesis and is sensitive to H2 O2 production [15]. The role of Ets-1 family members in tumorigenesis is largely due to their ability to activate the transcription of a number of genes involved in matrix degradation, such as MMP-1, 2, 3, and 9 as well as urokinase plasminogen activator. These proteases contribute to tumor invasion and the early phases of blood vessel formation [16]. In cisplatin-resistant ovarian carcinoma variants, elevations in H2 O2 are attributable to alterations of mitochondrial function relative to parental cell lines leading to an increase in transcription of Ets-1. The increase in transcription enhances binding of Nrf2, a key transcription factor in the cellular response to oxidative stress, to an antioxidant response element (ARE) in the promoter of Ets-1. Ets-1 expression has been correlated with poor prognosis in breast and ovarian cancer. Furthermore, the development of cisplatin resistance is a major limitation of this chemotherapy regime, which is widely used in the treatment of testicular and ovarian cancer. These observations suggest that shifts in the steadystate production of H2 O2 in response to receptor engagement or altered mitochondrial function resulting from the development of cisplatin resistance can enhance the expression of the angiogenic transcription factor Ets-1. VEGF is a key mediator in regulating the vascular recruitment of tumor endothelial cells. The transcriptional activity of VEGF is directly responsive to H2 O2 and involves a GC-rich region of the promoter between −95 and −51. Detailed characterization of the promoter has identified two SP1 and SP3 sites at −73/−66 and −58/−52 that represent the core mechanism of oxidative stress–triggered VEGF transactivation. Connor et al. have also established that mitochondrial-derived H2 O2 resulting from the dismuting activity of Sod2 was responsible for enhanced VEGFdependent angiogenic activity. The mechanism involved the H2 O2 -dependent oxidative inactivation of the dual lipid-protein phosphatase PTEN, leading to increase in activity of the serine/threonine-specific protein kinase AKT and subsequent VEGF expression [17]. Hypoxia inducible factor 1␣ (HIF-1␣), the master regulator of cellular responses to hypoxia, responds to shifts in the metabolic production of ROS [18]. HIF-1␣ regulates the transcriptional activity of genes that participate in angiogenesis, iron metabolism, glucose metabolism, and cell proliferation/survival. The regulation of HIF-1␣ in response to hypoxia is controlled by its hydroxylation state. Hydroxylation is controlled by a family of prolyl hydroxylases (PHDs) that require molecular oxygen for full activity. In the presence of oxygen, PHD activity is maximal, leading to HIF-1␣ hydroxylation. Hydroxylated HIF-1␣ is then targeted for polyubiquitination leading to its degradation by the cellular proteasome complex. In the absence of


39

O2 , hydroxylation is inhibited, leading to HIF-1␣ stabilization and activation of its target genes. Studies have determined that efficient mitochondrial ROS scavenging can prevent the hypoxic induction of HIF-1␣. This ROS-dependent stabilization of HIF-1␣ involves mitochondrial complex III–dependent production of superoxide and subsequent inhibition of PHD activity. The mechanisms for how mitochondrial ROS regulate HIF-1␣ are not clear but may involve a shift in the oxidation state of PHD-bound iron. It also possible that the hypoxic production of ROS may lead to the oxidative inactivation of phosphatases, such as PTEN, that contribute to the maximal activation of VEGF. Thus, it appears that angiogenic signaling is particularly sensitive to alterations in the metabolic production of oxidants and that mitochondrial free radicals may play a role in preserving levels of their initial substrate, oxygen, via HIF-1␣.

5 Mitochondrial ROS and Invasion/Migration A hallmark of malignancy and a primary cause of morbidity and mortality in cancer patients is metastasis. Tumor metastasis is a multistep process initiated by migration and invasion of cells into the surrounding vasculature. This is followed by extravasation from the circulation and proliferation at a secondary site where the formation of a metastatic lesion occurs. A key process in the initial movement and subsequent invasion of neoplastic cells into the circulatory system is the remodeling of the extracellular matrix (ECM). Emerging evidence has implicated ROS and the activation of redox-sensitive signaling pathways in invasion and migration. Intrinsic antioxidant enzymes are vital to the regulation of oxidative stress within cells. In vitro studies have shown that a number of cancer cell lines contain elevated levels of Sod2 and decreased levels of catalase and that this change in steady-state levels of H2 O2 correlates with increased metastasis, proliferation, and resistance to apoptosis. An integral part of migration and adhesion is the dynamic rearrangement of focal adhesion complexes in the cell. The role of Rho GTP-binding proteins (Rho, Rac, and Cdc42) in regulating cytoskeletal dynamics by activation of Src and focal adhesion kinase (FAK) and initiating remodeling of integrins has been firmly established. In adherent cells, mature focal adhesions have been shown to be regulated by RhoA, whose activation results in actin stress fiber formation. Rac and Cdc42 are considered to be promigratory resulting in the formation of focal complexes of lamellipodia and filopodia, respectively. It has been shown that ROS modulate Rac and Rho signaling. Rac can specifically downregulate RhoA in a ROS-dependent manner through oxidative inactivation of LMW-PTP, which normally dephosphorylates and inactivates the Rho GTPase-activating protein, p190Rho-GAP. Activation of p190Rho-GAP results in the conversion of the GTP-bound to the GDP-bound form of Rho, and its inactivation contributes to a loss of adhesion and a transition to a migratory phenotype. While ROS appear to disrupt the formation of mature focal adhesions of adherent cells, there is evidence that suggests a role of ROS in the turnover of leading edge focal complexes observed in lamellipodia of migrating cells. Rac has been shown to activate NADPH oxidase, which is spatially localized

40

P.M. Carrico et al.

to membrane ruffles of the leading edge of migrating cells, and was shown to be associated with cell migration. In this context, oxidants can reversibly inactivate the protein tyrosine phosphatase PTP-PEST that normally dephosphorylates and inactivates FAK, leading to a more rapid turnover of active FAK at focal complexes. It is also possible that mitochondrial-derived ROS are involved in disrupting mature focal adhesions yet contribute to active remodeling of focal complexes of the lamellipodia required for migration. Lipid signaling molecules also play an important role in regulating the migratory phenotype. Studies in phagocytic cells and the amoeba Dictyostelium discoideum demonstrate that phosphoinositide (3,4,5) 3-phosphate (PtdIns[3,4,5]P3 ) accumulates at the front of chemotaxing cells. Concomitant with the distribution of PtdIns(3,4,5)P3 to the leading edge is a redistribution of PTEN to the remaining outside perimeter of the cell where it degrades PtdIns(3,4,5)P3 by removing the 3phosphate. Alterations in the steady-state levels of H2 O2 can inactivate the tumor suppressor PTEN leading to an augmentation of AKT signaling and redistribution of phosphoinositides at the plasma-lamellar surface [17]. It is intriguing to hypothesize that receptor engagement in response to chemotactic triggers may lead to the local oxidant-dependent inactivation of PTEN and rapid reproportioning of lipid signaling molecules that promote the migratory phenotype. An essential and rate-limiting step in metastasis is the remodeling and degradation of the ECM and basement membrane by proteolytic enzymes. In this model, cells must interact with surrounding stromal cells, leading to the loss of matrix function and resulting in a compromised matrix boundary. MMPs are major contributors of stromal degradation and are vital to the process of cellular invasion. MMP-1/interstitial collagenase has a specificity for type I collagen, the primary collagen of the interstitial matrix; though it has been shown to degrade collagen types II, III, VII, and X. Studies using antisense mRNA against MMP-1 in melanoma cells have shown a significant attenuation in the capacity of these cells to invade type I collagen and Matrigel in vitro. Moreover, the expression of MMP-1 has been shown to be a definitive prognostic marker for breast lesions that will develop into cancer suggesting that the expression and regulation of MMP-1 may be important in malignancy. Studies from this and other laboratories have demonstrated that the Sod2dependent production of H2 O2 leads to increased expression of MMP family members and that there is a strong correlation between this increase in MMP levels and enhanced metastasis. Interestingly, the Sod2-dependent increases in MMP expression can be reversed by the H2 O2 detoxifying enzyme, catalase, or glutathione peroxidase [15]. Thus, H2 O2 -dependent upregulation of MMPs may, in part, contribute to increased invasion and metastatic capacity of tumors displaying elevated Sod2 levels. Maximal mitochondrial H2 O2 -dependent transcription of MMP-1 transcription is achieved by a proximal activator protein 1 (AP-1) site at –1602 and a single nucleotide polymorphism (SNP) guanine insertion at –1607, which creates an Ets-1 binding site [15]. This redox-sensitive MMP-1 protein expression requires activation of both ERK1/2 and JNK pathways. JNK signaling is largely responsible for


41

the H2 O2 sensitivity of the MMP-1 promoter, whereas ERK1/2 contributes to both its basal and H2 O2 dependence [19]. The ability of a tumor cell to migrate and invade through the extracellular matrix is attributed to local shifts in signaling that regulate the distribution of phospholipid signaling molecules, rearrangement of focal adhesion complexes, and matrix degradation. All of these processes are redox-sensitive and are likely responsive to metabolic shifts in the steady-state production of H2 O2 . As many of the processes are highly compartmentalized, it is possible that vicinal mitochondria may participate in these migratory signaling events by augmenting their local production of H2 O2 . With the development of fluorescent reporter molecules that detect focal shifts in the production of many of the migratory signaling intermediates and new redox-sensing mitochondrial targeted GFP constructs, we will be able to determine if these events are juxtaposed.

6 Mitochondrial ROS and Cancer Stem Cells Recently, the hypothesis has emerged that conventional therapies may fail to completely eradicate cancer because cancer stem cells, which are thought to act as cancer-initiating cells, are insensitive to these treatments. Work in leukemia, breast, brain, and lung cancers have provided further understanding of the role cancer stem cells play in malignancies. Although cancer stem cells appear to have some of the phenotypic and functional properties of embryonic stem cells, such as the potential to give rise to a larger population of cells, survival of these two different cell types requires different pathways that are frequently at odds with one another. Recent research indicates that signaling by ROS may regulate growth and differentiation in both hematopoietic stem cells and cancer stem cells. PTEN, like p53 is one of the most common targets for mutation in sporadic human cancers. The oxidative inactivation of PTEN and subsequent activation of the AKT signaling pathway may lead to the transformation of adult hematopoietic stem cells (HSCs) into leukemias. Transgenic mice with conditional deletions of PTEN in adult HSCs develop leukemias within 4 to 6 weeks along with a depletion of normal HSCs. These defects are dependent upon the downstream effector mammalian target of rapamycin (mTOR), a serine/threonine kinase that regulates metabolism and that has also been shown to generate ROS [20]. The oxidative inactivation of PTEN and subsequent activation of the AKT signaling pathway leads to the phosphorylation of the class O of forkhead transcription factors (FoxO), resulting in the retention of FoxO in the cytoplasm through interactions with 14-3-3 proteins [21]. FoxOs, which have also been postulated to act as tumor suppressors, regulate the transcription of genes involved in cell cycle arrest, apoptosis, glucose metabolism, and stress response. In C. elegans, the FoxO ortholog DAF-16 has been shown to affect life span through resistance to oxidative stress and adaptation to food shortage. DAF-16 promotes resistance to oxidative stress by activating transcription of Sod2 and catalase. In mammals, the FoxO

42

P.M. Carrico et al.

subgroup consists of four members (FoxO1, FoxO3, FoxO4, and FoxO6), of which three (FoxO1, FoxO3, FoxO4) overlap in function and expression patterns. Individual deletion of FoxO proteins in mice does not result in a tumor-prone phenotype, however, mice containing conditional deletions of FoxO1, FoxO3, and FoxO4 display a cancer-prone condition characterized by thymic lymphomas and hemangiomas [22]. In addition, mice with conditional deletions of FoxO1, FoxO3, and FoxO4 exhibited premature aging of HSCs, which was shown to be at least partly due to an increase in ROS levels in the HSC compartment and attributable to a decrease in expression of antioxidant genes [23]. Treatment of these mice with NAC restored the HSC compartment size, reestablished stem cell cycling, and normalized HSC apoptosis. FOXO3a is also regulated via p66Shc, which belongs to a family of adaptors that function in signal transduction of mitogenic and apoptotic responses, and whose absence has been shown to prolong life span both in cell culture and mouse models. Although p66Shc is activated through tyrosine phosphorylation in response to extracellular signals such as EGF and insulin, phosphorylation of p66Shc at serine 36 in response to oxidative stress leads to the subsequent phosphorylation and sequestration of FOXO3a in the cytosol [24]. This may further increase levels of ROS as FOXO3a has been shown to activate a reporter under the control of the human catalase promoter [24]. In a hypoxic environment, p66shc functions to promote survival of stem cells and mammary cancer stem cells through the upregulation of Notch-3 and carbonic anhydrase IX (CA-IX) [25]. In contrast, as discussed above, inhibition of FOXO proteins in response to H2 O2 augments apoptosis of stem cells. Recently, p66shc was also shown to regulate mitochondrial metabolism [26]. In immortalized mouse embryonic fibroblasts (MEFs) lacking p66shc, oxygen consumption is reduced 30% to 50% with a coincident increase in glycolysis and decrease in NADH metabolism. In addition, Giorgio et al. [27] demonstrated that p66shc utilizes reducing equivalents of the mitochondrial electron transfer chain through the oxidation of cytochrome c to generate mitochondrial H2 O2, which functions as a signaling molecule for apoptosis.

7 Conclusion The role of ROS in the process of carcinogenesis has evolved from damaging lipid, proteins, and nucleic acids that in turn promote tumor development to regulators of signal transduction events that control key facets of the malignant phenotype. Many of the redox-sensitive signaling proteins drive the expression or activation of genes and proteins that regulate tumor cell proliferation, survival, vascular recruitment, and migration. The source of the ROS that regulate these signaling networks has in many cases been assigned to the ubiquitous family of noninflammatory respiratory burst oxidases. However, emerging data indicates that mitochondrial-derived ROS also play a prominent role in regulating the tumorigenic phenotype. A number of


43

signaling proteins like the dual lipid protein phosphatase PTEN and the transcription factor Ets-1 are sensitive to perturbations in mitochondrial oxidant production and are involved in regulating numerous aspects of the tumorigenic phenotype. H2 O2 dependent activation of Ets-1 leads to enhanced tumor angiogenesis, migration, and invasion. Oxidative inactivation of PTEN leads to an increase in active AKT thereby promoting tumor cell survival, proliferation, and angiogenesis. This implies that alterations in mitochondrial function that enhance respiratory ROS production can control signaling molecules that synergize and potentially drive the malignant phenotype. PTEN contains an active site cysteine that renders it redox-sensitive. Ets-1 contains several cysteines that are essential for its DNA binding, and it is likely that oxidation would lead to its inactivation. The fact that oxidants promote Ets-1 activity suggest some as yet undefined oxidant-sensitive signaling molecule may control the redox-responsiveness of this transcription factor independent of its DNA-binding activity. It is clear that many facets of the tumorigenic phenotype are ROS responsive. The angiogenic factor VEGF is controlled by oxidants through promoter activation, stabilization of transcription factors (HIF-1␣), and activation of upstream signals. Ets-1 activation also contributes to the early phases of blood vessel formation. Thus, angiogenic signals appear to be highly sensitive to shifts in oxidant production and may be highly responsive to shifts in respiratory oxidant production. Further characterization and identification of oxidant-responsive molecular targets that control tumor progression and eradication of cancer stem cells will undoubtedly lead to the development of novel antioxidant-based therapies for the treatment of malignancy.

References 1. Nemoto S, Takeda K, Yu ZX, Ferrans VJ, Finkel T. Role for mitochondrial oxidants as regulators of cellular metabolism. Mol Cell Biol 2000; 20(19):7311–7318. 2. Rhee SG, Chang TS, Bae YS, Lee SR, Kang SW. Cellular regulation by hydrogen peroxide. J Am Soc Nephrol 2003; 14(8 Suppl 3):S211–S215. 3. Wood ZA, Poole LB, Karplus PA. Peroxiredoxin evolution and the regulation of hydrogen peroxide signaling. Science 2003; 300(5619):650. 4. Poole LB. Bacterial defenses against oxidants: mechanistic features of cysteine-based peroxidases and their flavoprotein reductases. Arch Biochem Biophys 2005; 433(1):240–254. 5. Tonks NK. Protein tyrosine phosphatases: from genes, to function, to disease. Nat Rev Mol Cell Biol 2006; 7(11):833–846. 6. Arnold RS, Shi J, Murad E, et al. Hydrogen peroxide mediates the cell growth and transformation caused by the mitogenic oxidase Nox1. Proc Natl Acad Sci USA 2001; 98(10): 5550–5555. 7. Tonks NK. Protein tyrosine phosphatases: from genes, to function, to disease. Nat Rev Mol Cell Biol 2006; 7(11):833–846. 8. Floyd RA. Antioxidants, oxidative stress, and degenerative neurological disorders. Proc Soc Exp Biol Med 1999; 222(3):236–245. 9. Menon SG, Goswami PC. A redox cycle within the cell cycle: ring in the old with the new. Oncogene 2007; 26(8):1101–1109. 10. Adams JM. Ways of dying: multiple pathways to apoptosis 2. Genes Dev 2003; 17(20): 2481–2495.

44

P.M. Carrico et al.

11. Wong GHW, Goeddel DV. Induction of manganous superoxide dismutase by tumor necrosis factor: possible protective mechanism. Science 1988; 242:941–944. 12. Dasgupta J, Subbaram S, Connor KM, et al. Manganese superoxide dismutase protects from TNF-alpha-induced apoptosis by increasing the steady-state production of H2 O2 . Antioxid Redox Signal 2006; 8(7–8):1295–1305. 13. Chiarugi P, Buricchi F. Protein tyrosine phosphorylation and reversible oxidation: two crosstalking posttranslation modifications. Antioxid Redox Signal 2007; 9(1):1–24. 14. Lambeth JD, Cheng G, Arnold RS, Edens WA. Novel homologs of gp91phox. Trends Biochem Sci 2000; 25(10):459–461. 15. Nelson KK, Melendez JA. Mitochondrial redox control of matrix metalloproteinases. Free Radic Biol Med 2004; 37(6):768–784. 16. Lincoln DW, Bove K. The transcription factor Ets-1 in breast cancer. Front Biosci 2005; 10:506–511. 17. Connor KM, Subbaram S, Regan KJ, et al. Mitochondrial H2 O2 regulates the angiogenic phenotype via PTEN oxidation. J Biol Chem 2005; 280(17):16916–16924. 18. Brunelle JK, Bell EL, Quesada NM, et al. Oxygen sensing requires mitochondrial ROS but not oxidative phosphorylation. Cell Metab 2005; 1(6):409–414. 19. Nelson KK, Ranganathan AC, Mansouri J, et al. Elevated sod2 activity augments matrix metalloproteinase expression: evidence for the involvement of endogenous hydrogen peroxide in regulating metastasis. Clin Cancer Res 2003; 9(1):424–432. 20. Kim JH, Chu SC, Gramlich JL, et al. Activation of the PI3K/mTOR pathway by BCRABL contributes to increased production of reactive oxygen species. Blood 2005; 105(4): 1717–1723. 21. Brunet A, Bonni A, Zigmond MJ, et al. Akt promotes cell survival by phosphorylating and inhibiting a Forkhead transcription factor. Cell 1999; 96(6):857–868. 22. Paik JH, Kollipara R, Chu G, et al. FoxOs are lineage-restricted redundant tumor suppressors and regulate endothelial cell homeostasis. Cell 2007; 128(2):309–323. 23. Tothova Z, Kollipara R, Huntly BJ, et al. FoxOs are critical mediators of hematopoietic stem cell resistance to physiologic oxidative stress. Cell 2007; 128(2):325–339. 24. Nemoto S, Finkel T. Redox regulation of forkhead proteins through a p66shc-dependent signaling pathway. Science 2002; 295(5564):2450–2452. 25. Sansone P, Storci G, Giovannini C, et al. p66Shc/Notch-3 interplay controls self-renewal and hypoxia survival in human stem/progenitor cells of the mammary gland expanded in vitro as mammospheres. Stem Cells 2007; 25(3):807–815. 26. Nemoto S, Combs CA, French S, et al. The mammalian longevity-associated gene product p66shc regulates mitochondrial metabolism. J Biol Chem 2006; 281(15):10555–10560. 27. Giorgio M, Migliaccio E, Orsini F, et al. Electron transfer between cytochrome c and p66Shc generates reactive oxygen species that trigger mitochondrial apoptosis. Cell 2005; 122(2):221–233.

Cellular Respiration and Dedifferentiation Roberto Scatena, Patrizia Bottoni, and Bruno Giardina

Abstract Mitochondria are far more than the “powerhouse” of the cell as they have classically been described. They are also the location where various catabolic and anabolic processes, calcium fluxes, reactive oxygen and nitrogen species, and numerous signal transduction pathways interact to maintain cell homeostasis and to modulate cellular responses to diverse stimuli. Because of the mitochondrion’s vital roles in cell physiology, this organelle squarely falls within the ambit of cancer pathophysiology. Indeed cancer-associated alterations in cell energy metabolism related to mitochondrial dysfunction have been recognized for many years in the socalled Warburg effect. Furthermore, these organelles appear to play a fundamental role in determining whether a cell will undergo cell death by apoptosis or necrosis; and key metabolic enzymes of the tricarboxylic acid (TCA) cycle, the fundamental oxidative pathway of the cell, may act as oncosuppressors. Recent findings indicate that modulation of mitochondrial activities in general, and of its electron respiratory chain in particular, could induce differentiation and/or death of cancer cells. Thus it may be possible to induce anticancer activities by deliberate modulation of cellular respiration. Elucidation of the role of mitochondria in cancer cell dedifferentiation processes may provide a new definition of the dedifferentiation/differentiation of cancer cells as well as important information about the pathophysiology of so-called stem cells, with important diagnostic, prognostic, and therapeutic consequences. Keywords Cancer cell differentiation · Oxidative phosphorylation · NADH dehydrogenase · Heat shock proteins · Reactive oxygen species · Warburg effect · Nitric oxide · Oncogenes · Oncosuppressors · Biomarkers

R. Scatena (B) Dipartimento di Medicina di Laboratorio, Università Cattolica del Sacro Cuore, 0168 Rome, Italy e-mail: [email protected] S.P. Apte, R. Sarangarajan (eds.), Cellular Respiration and Carcinogenesis, DOI 10.1007/978-1-59745-435-3 4, C Humana Press, a part of Springer Science+Business Media, LLC 2009

45

46

R. Scatena et al.

1 Introduction Cancer can be defined as a genetic disease characterized by dysregulation of various cellular pathways that orchestrate cell proliferation, differentiation, and death [1]. A cell becomes cancerous when a series of oncogenes and/or oncosuppressors become dysfunctional and hence induce a neoplastic phenotype in that cell. In this dramatic clonal disorder, the cell acquires high proliferation kinetics and loses cell-cell contact inhibition. These properties are progressively achieved by a series of genetic “hits,” including mutations and or other epigenetic modifications [1]. The final result is often defined as dedifferentiation of a neoplastic cell that morphologically becomes anaplastic. The grade of dedifferentiation (anaplasia) differs among cancer cell specimens and is often correlated with the malignancy of the neoplasia. In the dramatic perturbation of cell homeostasis that accompanies carcinogenesis, mitochondria are involved in various relevant, interconnected, but also indirect pathophysiologic roles. In this context, the significance of mitochondrial respiration has received inadequate attention. In spite of an abundance of data that link mitochondria to cancer, there have been few studies that have investigated the interrelationships between cellular respiration and cellular dedifferentiation. Here we briefly summarize some molecular mechanisms, which will be extensively considered in other chapters, that link cancer cell dedifferentiation to mitochondrial oxidative metabolism to provide an introduction to concepts that facilitate the understanding of these aspects of oncology:

1. Peculiar modification of cell metabolism (Warburg effect). Cancer cells generally produce ATP mainly through so-called aerobic glycolysis, a metabolic shift characterized by high glucose uptake and increased production of lactate. Originally, this altered metabolic state was considered to be the result of adaptation (via HIF and/or AKT) to the new potentially anoxic environment of a neoplastic lesion [2]. However it has become evident that the aerobic glycolysis of cancer cells is an epiphenomenon that results from a more complex metabolic rearrangement in which not only glycolysis but also Krebs cycle, beta-oxidation, and anabolic metabolism in general are altered to support the new primary function of the cell (e.g., uncontrolled proliferation) by providing not only energy but also the building blocks for the synthesis of nucleic acids, peptides, and lipids [3–5]. 2. Distinct properties of cancer cell mitochondria. Emerging data suggest that mitochondria have an active pathogenic role in the induction and maintenance of the Warburg effect. This role is emphasized by some typical metabolic peculiarities of mitochondria that enable cancer cells to be functionally distinguished from normal cells. Cancer cells exhibit reduced mitochondrial respiration (both in terms of ATP turnover and proton leakage), increased NADH/NAD+ ratios, and a pronounced increase in membrane potential [6]. Other morphologic (number, size, shape) and structural differences (phospholipids, cholesterol, and

Cellular Respiration and Dedifferentiation

3.

4.

5.

6.

47

membrane proteins) have been described, but their pathogenic roles are not understood [7]. Mitochondrial DNA (mtDNA) mutations. Mutations of mtDNA have been linked to some tumors. In particular, intragenic deletions, missense and chainterminating point mutations, and alterations of homopolymeric sequences have been described in various sporadic tumors. Intriguingly, mutations have been described in the hypervariable regions of the mitochondrial D-loop, the section of DNA that controls mtDNA transcription and replication. Mutations have also been described in all tRNAs, in rRNAs, and in all 13 of the mtDNA-encoded subunits of the electron respiratory chain [8]. However, Salas et al. [9] recently suggested that most of the reported mtDNA mutations in tumors may be due to laboratory-induced artifacts. Others contend that mtDNA mutations play a real pathogenetic role in cancer [10]. Hence there is ongoing debate about the role of mtDNA mutations and the consequent respiratory dysfunction in cancer cells. Nuclear DNA mutations and mitochondrial proteins. Succinate dehydrogenase (SDH) and fumarate hydratase (FH), mitochondrial enzymes of the tricarboxylic acid (TCA) cycle, can act as classic tumor suppressors, and mutations of the genes that encode them have been associated with various benign and malignant tumors. These observations have drawn attention to the potential pathogenetic role of the mitochondrion, especially of mitochondrial oxidative metabolism, in cancer [11, 12]. Given the role of SDH (complex II) in the electron respiratory chain, such findings suggest that there may be critical interrelationships between mitochondrial energy metabolism and timely surveillance of the differentiation/dedifferentiation state of cells. Apoptosis. Mitochondria have a well-defined role in caspase-mediated induction of the intrinsic pathway of apoptosis initiated by the release of cytochrome c and second mitochondrial activator of caspases (SMAC) from the mitochondrial intermembrane space in response to a variety of noxious stimuli, including DNA damage, loss of adherence to the extracellular matrix (ECM), oncogeneinduced proliferation, and growth factor deprivation. Perturbation in the release of apoptosis-inducing proteins regulated by proapoptotic and antiapoptotic members of the Bcl-2 family (e.g., Bcl-2, Bcl-XL, and Mcl-1) seems to influence not only etiopathogenesis but also cancer prognosis and therapeutic efficacy. Accordingly, derangements of the permeability transition pore (PTP) and electrochemical gradient, which strictly depend on mitochondrial respiration, have been implicated in these phenomena [13]. Free radicals (reactive oxygen species [ROS] and reactive nitrogen species [RNS]). Mitochondria are the main source of free radicals in cells. Thus mitochondrial dysfunction, especially at the level of the electron respiratory chain, can result in increased production of ROS and RNS. ROS are considered classic carcinogens because they can directly induce mutagenesis and thereby promote tumorigenesis and cancer progression [14]. However, the role of ROS and RNS in cancer is complex. In fact, the growth-promoting effects of ROS have been also related to activation of redox-responsive signaling cascades (MAPK, PKC,

48

R. Scatena et al.

AP-1, NF-␬B). Paradoxically, ROS can also act as tumor suppressors, activating via a p16ink4a -Rb pathway an irreversible cellular process of senescence that establishes stable G1 phase cell-cycle arrest [15]. 7. Heat shock proteins (HSPs). HSPs represent a ubiquitous and evolutionarily conserved family of proteins (classified according to their molecular weight) that were originally discovered because they can be induced by heat stress. Their production confers a peculiar cellular resistance to further stresses (heat or chemical). Initially, these proteins were considered to be chaperones—molecular guides acting to ensure proper protein folding. Other functions have since been discovered, including roles in gene expression regulation, DNA replication, signal transduction, cell differentiation, metastasis, apoptosis, senescence, immortalization, and, intriguingly, multidrug resistance development. However, the precise pathophysiologic roles of HSPs in general, and of mitochondrial Heat Shock Proteins (mitHSPs) in particular, are not well understood. Studies employing several methodological approaches (immunochemical, immunohistochemical, proteomic) have shown that HSP expression gradually increases from normal to dysplastic and then to neoplastic tissues, such that HSP overexpression correlates with cancer evolution. Hence some of these proteins (i.e., HSP60, HSP10, HSP70, PHB1, and PHB2) have been implicated in cancer diagnosis, prognosis, and therapy [16]. Overexpression of HSP70 (also known as 75-kDa glucose regulated protein, HSP70-9, HSP70-9B, mortalin 2) may be particularly important in cancer cell proliferation and dedifferentiation; its role may go beyond generic cytoprotection with respect to apoptosis and involve a more specific role in the pathogenesis of the peculiar cellular metabolic shift that characterizes the cancer cell phenotype [17]. 8. Hexokinase II. The cancer cell glycolytic phenotype is supported by overexpression of the hexokinase II isoenzyme, which is linked to the outer mitochondrial membrane via the porin-like protein voltage-dependent anion channel (PPVAC). Hexokinase II-PPVAC binding may have significant functional implications in cancer because it facilitates glucose phosphorylation and thus glucose utilization for energetic and biosynthetic purposes [18]. 9. Calcium. Because of the pivotal roles mediated by calcium ions in signal transduction pathways, calcium has been considered to be a potential physiopathogenetic marker in several chronic and acute cellular diseases. However, research findings have not confirmed the originally grand expectations for calcium in disease processes, at least in oncology. When Scorrano et al. [19] examined the interrelationships between calcium, mitochondrial respiration, and dedifferentiation, they found only an indirect role of Bax/Bak via regulation of calcium release from the endoplasmic reticulum. Interestingly, however, chemotherapeutic-resistant tumor cells sequester cytosolic calcium (Cacyt ) more efficiently than do normal cells; they show reduced release of calcium ions from intracellular storage sites upon apoptosis induction [20]. It has been proposed that transition to more invasive cellular behavior may be related to impaired release of calcium from the endoplasmic reticulum of various kinds of cancer cells. On the contrary, Amuthan et al. [21] showed in tumorigenic


49

human pulmonary carcinoma A549 cells that mitochondrial stress induced by damaging mtDNA (with ethidium bromide treatment) activates a mitochondria to nucleus signaling pathway mediated by increased Cacyt concentration and that activation of this pathway induces cell dedifferentiation associated with phenotypic changes and cell invasion.

2 Cellular Respiration and Dedifferentiation Although the connection between mitochondrial respiration and cancer cell dedifferentiation (i.e., resistance to apoptosis, cell proliferation, metabolic shift, cell invasion, genetic instability) is fundamental to cancer pathogenesis, the precise molecular interrelationships and regulatory processes are far from clarified. This lack of understanding is related, at least in part, to the traditional view that the socalled Warburg effect was essentially a reflection of mitochondria being damaged and dysfunctional organelles in cancer cells. Recent studies are reexamining mitochondrial function, especially cellular respiration, in cancer. However, a consensus among the findings has yet to be achieved [6, 21–25]. Herein some important findings linking cellular respiration to cell dedifferentiation in cancer are provided. It is worth noting that this putative link may have crucial pharmacologic and clinical implications. The work of Amuthan et al. [21] is particularly relevant in this regard because the findings provide mechanistic insight into how cancer progression and tumor invasion may be directly related to chemical damage of mtDNA that results in mitochondrial respiratory chain dysfunction, loss of mitochondrial membrane potential, and reduction of ATP synthesis. This mitochondrial derangement also causes calcium release into the cytoplasm, which activates a cell dedifferentiation program. Amuthan et al. [21] showed in C2C12 myoblasts and in human pulmonary carcinoma A549 cells that thus-induced dedifferentiation programs can transform cells into a more aggressive phenotype with the following properties: • induction of hexokinase, the enzyme that primes glucose by consuming ATP; • induction of phosphoenolpyruvate carboxykinase (PEPCK), which regulates gluconeogenesis, glyceroneogenesis, and many other anaplerotic reactions; • induction of tumor invasion markers (i.e., cathepsin L, TGF-␤1, ERK1, ERK2, calcineurin) and antiapoptotic proteins (Bcl2 and Bcl-XL ); • reduction of levels of the proapoptotic proteins Bid and Bax; • reduction of cell growth and occurrence of morphologic changes. In summary, the findings of Amuthan et al. [21] demonstrate that, at least in this set of experimental conditions, a generic mitochondrial stress caused by mtDNA depletion deranges cellular respiration and alters patterns of nuclear gene expression and thereby induces tumorigenic properties.

50

R. Scatena et al.

In general, it is well-known that mtDNA depletion can also render cells more sensitive to apoptotic stimuli. Amuthan et al. [21] explain that the outcome of the stress (cell dedifferentiation vs. apoptosis) could be related to the relative resistance to apoptosis and/or to the capacity of the adopted cellular models to buffer efficiently the secondary metabolic stresses. Furthermore, evidence obtained from this combination of methodological approaches demonstrated the existence of mitochondriato-nucleus stress signaling, through which alterations in cellular respiration have the ability to change nuclear gene expression and selectively promote the growth of more aggressive clones. These results compel us to consider other research demonstrating that mitochondria can mediate the proliferative or inhibitory actions of different physiologic or pharmacologic molecules that influence cellular respiration. Interestingly, it has been reported recently that some “nongenomic” effects of estrogens, that is; effects mediated by mitochondria rather than estrogen receptors, seem to modulate the expression of nuclear cell cycle genes and human breast tumor growth. Estrogens cause increases in the transcription of various complex subunits of the electron respiratory chain in mitochondria. For example, estradiol treatment (0.5 nM for 6 days) is associated with a 16-fold increase in cytochrome oxidase II (COII) mRNA in rat pituitary tumor cells. Similarly, ethinyl estradiol treatment (0.5 to 10 ␮M for 40 hours) caused a two- to threefold increase in cytochrome oxidase I (COI), COII, and NADH dehydrogenase subunit I. Estrogens seem to negatively affect mitochondria at the protein level as well. Several studies have shown that estrogens can inhibit mitochondrial respiratory complexes I, II, III, IV, and V. This perturbation of the electron respiratory chain can induce mitochondrial ROS production, which could promote the generation and progression of some cancers via oxidative stress and/or ROS-related signaling pathways [23]. Given the suggested role of estrogens in some cancers (i.e., breast cancer) and the discussed pathogenetic mechanisms at the basis of oncopromotion and oncoprogression activity, further evidence should be gathered to determine the relevance of these findings. The ability of lipophilic molecules to derange cellular respiration and alter cell differentiation/dedifferentiation processes is a neglected pathophysiologic facet of the mitochondrial respiratory chain. On the basis of the cytopathology of liver disease associated with use of fibrates (lipid-reducing drugs) and some experimental data indicating that fibrate binding to human hemoglobin can functionally modify the hemoprotein, we postulated that fibrates, and their thiazolidinedione derivatives, could interact with many hydrophobic protein domains and especially with complex I subunits of the mitochondrial electron respiratory chain. We found that this molecular interaction is rapid (30, MW 850 kDa), which because of these features may act as the main lipophilic molecular sink in mitochondria [26]. The rapid iatrogenic respiratory derangement caused by introduction of these lipophilic molecules can induce metabolic and ROS-mediated stress that provokes a differentiation process, also mitochondrial HSP-mediated, in human tumor cell lines. In particular, the derangement of mitochondrial complex I may induce ROS production and may impair NADH oxidation resulting in an energetic shortage in cancer cells. Importantly, considering that complex I is only partially deranged, the oxidative and energetic stress are in general not lethal. In this condition, cell growth is hampered and phenotypic differentiated features may reappear. This implies that there is modification of the actual differentiation/dedifferentiation definition and thus suggests that a redefinition of the role of some oncogenes and oncosuppressors is appropriate. Data confirming that fibrates and thiazolidinediones inhibit, via nongenomic activity, complex I–impairing NADH oxidation at the mitochondrial level can explain some of the pharmacologic activities of this class of drugs (i.e., antihyperlipidemia, insulin sensitization). The findings also may account for some adverse side effects of fibrates and thiazolidinediones, including myocardiotoxicity, hepatotoxicity, and, most relevant to this discussion, peroxisome proliferation, carcinogenicity, and cancer differentiating properties, considered typical peroxisome proliferator activated receptor (PPAR)-␣–mediated effects in rodents [27]. The differentiating activity of fibrates and thiazolidinediones on human tumor cell lines has been shown to be tightly correlated with the level of mitochondrial complex I derangement. This differentiating activity is independent of the expression level of PPARs, as well as the embryonic origin of the tissue. Stress-related differentiation in human tumor cell lines, including some human leukemia cell lines, is a well-known occurrence [28]. Fibrate/thiazolidinedione-induced differentiation is unusual; it is linked to the metabolic status of differentiated cells that shift toward a more glycolytic metabolism (paradoxical Warburg effect) and thus showing that original neoplastic phenotype yet used mitochondria. Moreover, metabolic parameters recorded during drug-induced cell differentiation (residual activity of the mitochondrial electron respiratory chain components glycerol 3-phosphate dehydrogenase and succinate dehydrogenase; levels of glycolytic and nonglycolytic acetate, pyruvate, and alanine) suggest that the so-called Warburg effect is really an epiphenomenon of a more complex metabolic rearrangement of the neoplastic cell. This cancer-associated metabolic rearrangement is induced by the cancer’s rigid genetic selection program, characterized by a high level of cell proliferation and diffusion, which is normally present only during embryonic and fetal development. Our understanding of the functional connections between alterations in cell metabolism and mitochondrial physiology in cancer is under profound revision. And all mitochondrial physiology is strongly affected by electron respiratory chain activities. The recent results of Fantin et al. [6] are consistent with the view that tumor

52

R. Scatena et al.

progression is strongly associated with the so-called Warburg effect and mitochondrial metabolism. These authors showed that attenuation of lactate dehydrogenase A (LDH-A) expression by LDH-A short hairpin RNAs in tumor cell lines reactivates mitochondrial respiration, in terms of proton leakage and ATP turnover, and restores the normally depolarized state of mitochondrial membranes. In this situation, cancer cells can no longer oxidize NADH via LDH but must rely on the mitochondrial respiratory chain. Interestingly, this abruptly induced change in metabolism inhibited cell growth. The authors concluded that these results demonstrate a fundamental role of aerobic glycolysis in general and LDH in particular in tumor maintenance and progression and further show that cancer-associated slowdown of mitochondrial respiration is not due to structural damage of the organelles but rather results from altered metabolic signals. NADH/NAD+ ratio and modulation of related mitochondrial shuttle systems are likely involved in these signals. The fundamental role of cellular respiration in cancer cell dedifferentiation is further emphasized by a recent study that examined energy metabolism in breast cancer brain metastasis, which undoubtedly represents a clear in vivo model of bioenergetic adaptation of dedifferentiated cells [29]. Interestingly, these metastatic cells showed increased expression of enzymes involved in glycolysis, in accordance with the Warburg effect, but also showed increased expression of enzymes involved in the TCA cycle, pentose phosphate pathways, beta-oxidation, and oxidative phosphorylation. Intriguingly, the authors also reported induction of enzymatic pathways related to glutathione synthesis, which together with activation of the pentose phosphate pathway could indicate the presence of antioxidant defense mechanisms to minimize production of ROS derived from enhanced oxidative metabolism [29].

3 Conclusion Emerging evidence, despite conflicts likely due to experimental differences, is shedding light on the interrelationships between cellular respiration and cancer cell dedifferentiation, in particular: • Confirms a more active role of mitochondrial respiration to maintain the cellular differentiated phenotype and at the same time underline the pathogenetic significance of perturbations of mitochondrial electron respiratory chain in promotion and progression of cancer. • Stresses the importance of the shift of all cellular metabolism to ensure the new energetic and structural needs of neoplastic cell. • Justifies some debated metabolic and functional evidence peculiar to cancer cell (i.e., induction of hexokinase II and its binding to ANT). • Enhances the role of NADH and NADH/NAD+ ratio in modulation of a signal transduction pathway that controls differentiation/dedifferentiation status. • Emphasizes the interplay of complex I and complex II to adapt cell metabolism to different cellular physiopathologic conditions.


53

• Last but not least, addresses the new research, which explores the interrelationships between cellular respiration and cell dedifferentiation, toward the regulatory function of complex I, complex II, and above all pushes to investigate deeper, in terms of cancer cell metabolism adaptions, the other pathways of electron transport (i.e., alpha glycerophosphate and fatty acyl CoA).

In conclusion, it is clear that there exists a significant link between cellular respiration and cell differentiation/dedifferentiation. Elucidating the molecular mechanisms at the basis of this link may have critical clinical implications in terms of cancer diagnosis, prognosis, and, most importantly, therapy. Disentangling the relationship between cancer cell oxidative metabolism and cellular differentiation state is fundamental not only for oncology but also for properly understanding and using all potential clinical applications related to the physiopathology of stem cells.

References 1. Bale AE, Brown SJ. Etiology of cancer: cancer genetics. In: De Vita VT, Hellman S, Rosenbreg SA, eds. Cancer—Principles & practice of oncology, 6th ed. Philadelphia: Lippincot Williams & Wilkins, 2001:170–183. 2. Gatenby RA, Gillies RJ. Why do cancers have high aerobic glycolysis? Nat Rev Cancer 2004; 4:891–899. 3. Moreno-Sanchez R, Rodriguez-Enriquez S, Marin-Hernandez A, Saavedra E. Energy metabolism in tumor cells. FEBS J 2007; 274:1393–1418. 4. Kim JW, Dang CV. Cancer’s molecular sweet tooth and the Warburg effect. Cancer Res 2007; 66:8927–8930. 5. Garber K. Energy deregulation: licensing tumor to grow. Science 2006; 312:1158–1159. 6. Fantin VR, St-Pierre J, Leder P. Attenuation of LDH-A expression uncovers a link between glycolysis, mitochondrial physiology, and tumor maintenance. Cancer Cell. 2006; 9:425–434. 7. Modica-Napolitano JS, Singh KK. Mitochondrial dysfunction in cancer. Mitochondrion 2004; 4:755–762. 8. Brandon M, Baldi P, Wallace DC. Mitochondrial mutations in cancer. Oncogene 2006 7; 25:4647–4662. 9. Salas A, Yao YG, Macaulay V, Vega A, Carracedo A, Bandelt HJ. A critical reassessment of the role of mitochondria in tumorigenesis. PLoS Med 2005; 2(11):e296. 10. Herrnstadt C, Preston G, Howell N. Errors, phantoms and otherwise, in human mtDNA sequences. Am J Hum Genet 2003; 72:1585–1586. 11. King A, Selak MA, Gottlieb E. Succinate dehydrogenase and fumarate hydratase: linking mitochondrial dysfunction and cancer. Oncogene 2006; 25:4675–4682. 12. Gottlieb E, Tomlinson IP. Mitochondrial tumour suppressors: a genetic and biochemical update. Nat Rev Cancer 2005; 5:857–866. 13. Cheng WC, Berman SB, Ivanovska I, et al. Mitochondrial factors with dual roles in death and survival. Oncogene 2006; 25:4697–4705. 14. Droge W. Free radicals in the physiological control of cell function. Physiol Rev 2002; 82: 47–95. 15. Takahashi A, Ohtani N, Yamakoshi K, et al. Mitogenic signalling and the p16INK4a-Rb pathway cooperate to enforce irreversible cellular senescence. Nat Cell Biol 2006; 8:1291–1297. 16. Calderwood SK, Khaleque MA, Sawyer DB, Ciocca DR. Heat shock proteins in cancer: chaperones of tumorigenesis. Trends Biochem Sci 2006; 31:164–172.

54

R. Scatena et al.

17. Czarnecka AM, Campanella C, Zummo G, Cappello F. Mitochondrial chaperones in cancer: from molecular biology to clinical diagnostics. Cancer Biol Ther 2006; 5:714–720 18. Mathupala SP, Ko YH, Pedersen PL. Hexokinase II: cancer s double-edged sword acting as both facilitator and gatekeeper of malignancy when bound to mitochondria. Oncogene 2006; 25(34):4777–4786. 19. Scorrano L, Oakes SA, Opferman JT, et al BAX and BAK regulation of endoplasmic reticulum Ca2+: a control point for apoptosis. Science 2003; 300:135–139. 20. Chandra D, Liu JW, Tang DG. Early mitochondrial activation and cytochrome c up-regulation during apoptosis. J Biol Chem 2002; 277:50842–50854. 21. Amuthan G, Biswas G, Ananadatheerthavarada HK, Vijayasarathy C, Shephard HM, Avadhani NG. Mitochondrial stress-induced calcium signaling, phenotypic changes and invasive behavior in human lung carcinoma A549 cells. Oncogene 2002; 21:7839–7849. 22. Scatena R, Bottoni P, Vincenzoni F, et al. Bezafibrate induces a mitochondrial derangement in human cell lines: a PPAR-independent mechanism for a peroxisome proliferator. Chem Res Toxicol 2003; 16:1440–1447. 23. Felty Q, Roy D. Estrogen, mitochondria, and growth of cancer and non-cancer cells. J Carcinog 2005; 4:1–18 24. Bottoni P, Giardina B, Martorana GE, et al. A two-dimensional electrophoresis preliminary approach to human hepatocarcinoma differentiation induced by PPAR-agonists. J Cell Mol Med 2005; 9:462–467. 25. Scatena R, Nocca G, Sole PD, et al. Bezafibrate as differentiating factor of human myeloid leukemia cells. Cell Death Differ 1999; 6:781–787. 26. Scatena R, Bottoni P, Martorana GE, et al. Mitochondria, ciglitazone and liver: a neglected interaction in biochemical pharmacology. Eur J Pharmacol 2007; 567:50–58. 27. Michalik L, Desvergne B, Wahli W. Peroxisome proliferator-activated receptor and cancers: complex stories. Nat Rev Cancer 2004; 4:61–70. 28. Richards FM, Watson A, Hickman JA. Investigation of the effects of heat shock and agents which induce a heat shock response on the induction of differentiation of HL-60 cells. Cancer Res 1988; 48:6715–20. 29. Chen EI, Hewel J, Krueger JS, et al. Adaptation of energy metabolism in breast cancer brain metastases. Cancer Res 2007; 67:1472–1486.

Cellular Adaptations to Oxidative Phosphorylation Defects in Cancer Sarika Srivastava and Carlos T. Moraes

Abstract Mitochondrial DNA (mtDNA) somatic mutations or mutations in nuclear genes encoding mitochondrial proteins important for the assembly, activity, or maintenance of the individual oxidative phosphorylation (OXPHOS) complexes have been observed in tumors. Although the functional consequence of such mutations is unclear at the moment, retrograde signaling in response to OXPHOS defects can activate various nuclear genes and signaling pathways that alter mitochondrial function, tumor invasion, metastasis, redox-sensitive pathways, programmed cell death pathways, calcium signaling pathways, and cellular pathways leading to global changes in cellular morphology and architecture. In addition, we have found that some cancer cell lines harboring deleterious mtDNA mutations upregulate the expression of members of the peroxisome-proliferator activated ␥ coactivator 1 family of coactivators, probably to sustain the necessary ATP production for cell proliferation. In this chapter, we describe such cellular adaptations and changes in response to OXPHOS defects that are associated with a variety of cancer cell types. Keywords mtDNA · Retrograde signaling · OXPHOS · Cancer · Mitochondria

1 Mammalian Oxidative Phosphorylation Mammalian mitochondrial oxidative phosphorylation (OXPHOS) occurs in five multimeric enzyme complexes (complexes I, II, III, IV, and V) using two electron transport carriers (ubiquinone or coenzyme Q10 and cytochrome c). Electrons from reduced equivalents (NADH and FADH2 ) are transported along these complexes via ubiquinone and cytochrome c to molecular oxygen, producing water. At the same time, protons are pumped across the mitochondrial inner membrane (from the matrix to intermembrane space) by enzyme complexes I, III, and IV thereby generating a C.T. Moraes (B) University of Miami Miller School of Medicine, Miami, FL 33136 e-mail: [email protected] S.P. Apte, R. Sarangarajan (eds.), Cellular Respiration and Carcinogenesis, DOI 10.1007/978-1-59745-435-3 5, C Humana Press, a part of Springer Science+Business Media, LLC 2009

55

56

S. Srivastava, C.T. Moraes Intermembrane space H+

H+

H+

eO2 e-

e-

e-

e-

H2 O

e-

Cyt c

CoQ10

ATP

ADP

H+ Complex I

Complex II

Complex III

Complex IV

Complex V

Matrix

Fig. 1 Schematic representation of mammalian mitochondrial OXPHOS. Mammalian electron transport chain (ETC) consists of four multimeric complexes (complexes I, II, III, and IV). Electrons from the reduced equivalents, NADH and FADH2 , enter the ETC and reduce complex I and complex II, respectively. Coenzyme Q10 or ubiquinone is an electron carrier in the inner mitochondrial membrane that accepts an electron from either complex I or complex II and donates it to complex III. Cytochrome c, another electron carrier in the mitochondrial intermembrane space, accepts an electron from complex III and donates it to complex IV. Complex IV donates electrons to molecular O2 , which results in the formation of H2 O. During the electron flow, complexes I, III, and IV pump the protons from the matrix toward the intermembrane space thereby generating an electrochemical gradient across the mitochondrial inner membrane. The energy in the electrochemical gradient is harnessed by complex V to generate ATP from ADP (oxidative phosphorylation)

proton gradient or membrane potential (⌬⌿m). The energy in this gradient is harnessed by complex V (ATP synthase) to synthesize ATP from ADP, a phenomenon termed OXPHOS, during which protons flow back from the mitochondrial intermembrane space to the mitochondrial matrix (Fig. 1). Mitochondria are also important for cytosolic Ca+2 buffering and are the predominant source of ROS production in most cell types. Superoxide anions are generated at the level of complex I and complex III in the inner mitochondrial membrane.

2 OXPHOS Defects in Cancer OXPHOS defects in cancer cells were first reported more than 50 years ago by Warburg who stated that “cancer cells are impaired in respiratory chain function and are very glycolytic” [1]. Studies on various types of cancer cells have found

Cellular Adaptations to OXPHOS Defects in Cancer

57

this hypothesis to be correct in several but not all tumor cells. Several independent groups have also reported that many cancer cells possess a full complement of respiratory chain enzymes and can couple electron transport chain to ATP production [2]. Isolated tumor mitochondria have also been found to respire on a variety of substrates. Specific changes that occur in the energy metabolism of tumor cells have not yet been established. In the past 10 years, studies on mitochondrial DNA (mtDNA) mutations that cause OXPHOS defects and their potential role in cancer have regained momentum. Mitochondrial DNA mutations have been increasingly identified in various types of cancer cells [3, 4]. Polyak et al. were first to report a detailed analysis of somatic mtDNA mutations in human colorectal cancer cells by sequencing their complete mitochondrial genome [3] (somatic mtDNA mutations are those present in the tumor tissue but absent from the adjacent normal tissue). After this initial finding, mtDNA from various tumor cell types was sequenced by different groups, and mtDNA mutations have now been reported in esophageal, ovarian, thyroid, head, neck, lung, bladder, and renal cancer cells [4–6]. The majority of the mtDNA mutations reported in different tumor cell types are transitions (G-to-A or T-to-C), a feature that is characteristic of reactive oxygen species (ROS)-derived mutations. Further, most of the somatic mtDNA mutations have been reported to be homoplasmic suggesting their dominance at intra- and intercellular levels.

2.1 Functional Significance of OXPHOS Defects in Cancer Although in the past decade, the list of mtDNA mutations affecting OXPHOS in cancer cells has increased steadily, the functional relevance of these mutations in tumor promotion or formation process are yet obscure. Because the majority of the somatic mtDNA mutations found in tumor cells have been reported to be homoplasmic, one of the existing hypotheses is that the homoplasmic levels of these mutations were probably achieved through some kind of replication or growth advantage that the mutant mtDNA harbored over the wild type in the tumor progenitor cells. It has also been suggested that mtDNA mutations may cause a moderate increase in ROS production that in turn stimulates cell growth [3, 4]. Low levels of ROS have been shown to stimulate mitosis in various cell types [7, 8]. In contrast with this hypothesis, Coller and co-workershave demonstrated that there is a sufficient opportunity for a tumor progenitor cell to achieve homoplasmy through unbiased mtDNA replication and segregation during cell division [9]. Coller et al.constructed a theoretical model based on computer simulation studies and showed that there is a good correlation between the reported frequency of homoplasmic mutations in tumor cells and the predicted frequency of homoplasmy in tumor progenitor cells that can be attained in the absence of any selection [9]. The authors suggested that mtDNA mutations such as silent substitutions of amino acid coding regions or alterations in the length of polynucleotide tracts that are polymorphic in healthy individuals are the ones that are unlikely to confer any selective advantage to the mtDNA or the

58

S. Srivastava, C.T. Moraes

host cell. However, the authors did not exclude the possibility that certain mtDNA point mutations that cause functional alterations (e.g., mutations affecting the conserved amino acid regions in the coding region of a protein) might be segregated in a nonrandom manner and may provide a selective growth advantage to the host cell [9]. mtDNA mutations leading to OXPHOS dysfunction have also been shown to increase tumorigenicity in certain tumor cell types. Mutations in the mtDNA COXI gene were shown to increase tumorigenicity in prostate cancer patients [10]. Petros et al. reported that ∼11% to 12% of prostate cancer patients harbor pathogenic COXI mtDNA mutations that alter the conserved amino acid regions in the protein and suggested that these mutations may play a role in the etiology of prostate cancer [10]. mtDNA ATP6 gene mutation (T8993G) has also been shown to increase tumorigenicity by two independent groups. The first study was from Petros and coworkers who showed that PC3 prostate cancer cybrid cells harboring T8993G ATP6 mutant mtDNA when introduced in nude mice formed tumors that were ∼7 times larger than cells harboring the wild-type mtDNA suggesting a role of mtDNA mutation in tumor promotion process [10]. They also showed that tumor cells harboring ATP6 mutant mtDNA generated significantly higher levels of ROS than did those harboring wild-type mtDNA, further suggesting that high levels of ROS promoted tumor cell growth. The second study was from Shidara and co-workers. They constructed HeLa transmitochondrial cybrids harboring homoplasmic T8993G ATP6 mtDNA mutation and showed that these cybrids grew much faster in culture as well as in nude mice forming tumors that were larger in size compared with the cybrids harboring the wild-type mtDNA suggesting that the mtDNA ATP6 mutation in these cells conferred a selective growth advantage [11]. Further, transfection of a wild-type nuclear version of the mitochondrial ATP6 gene in these HeLa cybrids harboring homoplasmic ATP6 mutant mtDNA followed by their transplantation in nude mice was shown to slow down the tumor cell growth. Conversely, the expression of a nuclear version of the mutant ATP6 gene in the wild-type cybrids was found to accelerate tumor cell growth thereby suggesting that the growth advantage in these tumor cells depended on the mitochondrial ATP6 function [11]. OXPHOS dysfunction caused by partial mtDNA depletion (genetic stress) or treatment with mitochondrial-specific inhibitors (metabolic stress) has been shown to induce cell invasion and tumor progression in otherwise noninvasive cells [12, 13]. A correlation between OXPHOS impairment and tumor aggression has also been reported in renal carcinomas [14]. Simonnet et al. studied the mitochondrial respiratory chain enzyme content in three different renal tumors and found that the OXPHOS impairment increased from the less aggressive to the most aggressive types of renal carcinomas [14]. They also found that renal oncocytomas (benign tumors) are exclusively deficient in mitochondrial complex I activity, whereas the activity of all other OXPHOS complexes are increased in these tumors [15]. These tumors further showed dense mitochondrial proliferation suggesting an increase in mitochondrial biogenesis as an attempt to compensate for the potential loss of OXPHOS function. Complex I deficiency was also found in normal cells adjacent to the tumor tissue leading the authors to suggest that the complex I deficiency could be


59

an early event in the formation of renal oncocytomas [15]. Similarly, OXPHOS dysfunction and mitochondrial proliferation have also been reported in thyroid oncocytomas [16]. The current studies on potential effects of mtDNA mutations and OXPHOS dysfunction in cancer cells therefore suggest their role in tumor promotion and/or progression. It is believed that nuclear DNA mutations and not mtDNA mutations control the tumor initiation process. Akimoto and co-workers have tested this hypothesis [17]. They showed that cells carrying nuclear DNA from tumor cells and mtDNA from normal cells form tumors in nude mice whereas those carrying nuclear DNA from normal cells and mtDNA from tumor cells do not form tumors [17]. However, mutations in the mitochondrial enzymes that are encoded by the nuclear genome have been associated with certain types of cancer. Germ-line mutations in the tricarboxylic acid (TCA) cycle enzyme fumarate hydratase have been shown to be associated with uterine fibroids, skin leiomyomata, and papillary renal cell cancer [18], and mutations in the succinate dehydrogenase (SDH) subunits B, C, and D have been shown to be associated with hereditary paragangliomas and pheochromocytomas [19–21]. The potential role of these mutations in the tumor formation process is yet unclear.

3 Cellular Adaptations to OXPHOS Defects in Cancer The cross-talk between mitochondria and nucleus plays an important role in maintaining mitochondrial function and integrity. Mitochondrial dysfunction affects the mitochondrial-nuclear cross-talk leading to altered signaling cascades (Fig. 2). The nuclear-mitochondrial stress signaling, also called retrograde signaling, is a phenomenon that involves changes in the nuclear gene expression in response to

OXPHOS regulatory or structural genes

Altered Gene Expression

Fig. 2 Mitochondrial-nuclear intergenomic signaling in response to OXPHOS defects. Mitochondrial OXPHOS dysfunction in cancer cells can activate a retrograde signaling cascade. Altered retrograde signaling can affect the expression of various nuclear genes and cellular pathways that allow tumors cells to adapt to the environment and/or promote tumor cell growth or progression

Calcium signaling pathway Redox signaling / antioxidant defense pathway Cellular morphology and architecture Tumor invasion or metastatic genes

Nucleus

Retrograde Signaling

OXPHOS Dysfunction

60


OXPHOS dysfunction. These nuclear gene expression changes allow the tumor cells to adapt to the environment that may promote tumor cell growth and/or progression. Studies have shown that a variety of nuclear genes are altered in response to OXPHOS defects in different tumor cell types and have implicated their role in tumorigenesis.

3.1 Retrograde Signaling in Cancer Cells Retrograde signaling has been implicated as an important mechanism in carcinogenesis although the precise mechanism(s) of this pathway is yet elusive in mammalian systems. Studies have shown that retrograde signaling in response to mitochondrial dysfunction can alter the expression of nuclear genes involved in energy metabolism, tumor invasion and metastasis, calcium signaling, redox signaling, antiapoptotic genes, as well as genes involved in regulating cellular morphology and architecture. 3.1.1 Altered Expression of OXPHOS Regulatory Genes and Subunits The mitochondrial OXPHOS complexes (I, III, IV, and V) comprise subunits encoded by both mtDNA and nuclear DNA. The expression of nuclear encoded OXPHOS subunits is under a tight regulation by regulatory genes, namely transcription factors and transcriptional coactivators. Nuclear respiratory factors 1 and 2 (NRF-1 and NRF-2) are the nuclear transcription factors, and the members of the peroxisome-proliferator activated ␥ coactivator 1 (PGC-1) gene family are the nuclear transcriptional coactivators that directly or indirectly modulate the expression of nuclear encoded OXPHOS genes, respectively. Besides regulating the expression of nuclear encoded OXPHOS genes, the NRFs also exert a regulatory control over the expression of mitochondrial transcription and replication machinery components, the protein import machinery components, and heme biosynthesis. A change in the expression levels or activity of NRFs and/or PGC-1 family of transcriptional coactivators would therefore influence the expression of OXPHOS subunits and mitochondrial biogenesis. Altered expression of OXPHOS subunits and transcriptional regulatory genes has been reported in different tumor cell types [22, 23]. We studied a human colorectal tumor cell line (termed V425) harboring a homoplasmic nonsense mutation in the mitochondrial COXI gene. Despite the mtDNA mutation load, these cells were found to maintain a very high rate of mitochondrial respiration (∼75% of control cells) and a low level of COX activity (∼10% of control cells). Interestingly, osteosarcoma cybrids harboring the V425 mutant mtDNA showed a significant decline in respiration and COX activity [22]. Previous studies have shown that ∼10% COX activity cannot sustain a functional respiratory chain [24]. We found that the steady-state transcript levels of a transcriptional coactivator PGC-1␣ and its homolog PGC-1␤ were highly upregulated in V425 cells. In parallel, we also observed an increase in the steady-state levels of several mitochondrial proteins (e.g., SDH, COXIV, and cytochrome c) in these


61

cells. Further, the overexpression of PGC-1␣ in osteosarcoma cybrids harboring the V425 mutant mtDNA showed a significant improvement in mitochondrial respiration and an increase in the steady-state levels of several mitochondrial proteins suggesting that PGC-1␣ upregulation can improve mitochondrial respiration in a colorectal cancer cell line harboring COX deficiency [22]. It has also been shown that PGC-1␣ stimulates mitochondrial biogenesis and respiration by activating the expression of NRFs, augmenting their transcriptional activity or activating both the expression and transcriptional activity [25]. Altered expression of PGC-1 related coactivator (PRC), NRF-1, and mitochondrial transcription factor A (TFAM) have also been observed in thyroid oncocytomas harboring an OXPHOS defect [23]. Savagner et al. found that dense mitochondrial proliferation in 90% of the tested oncocytic tumors were associated with overexpression of the PRC compared with the controls suggesting that increased mitochondrial biogenesis might be a feedback mechanism to compensate for the presence of OXPHOS deficit in these tumors. They also found that the increase in PRC levels was associated with a 5-fold increase in NRF-1 transcripts, 10-fold increase in TFAM transcripts, and a 3-fold increase in COX activity further suggesting that the overexpression of PRC pathway is likely responsible for the increased mitochondrial proliferation in thyroid oncocytomas [23]. In contrast with the above studies, low levels of expression of peroxisomeproliferator activated receptor gamma (PPAR-␥) and PGC-1 have also been reported in breast cancer tissues [26]. Jiang et al. showed that the human metastatic breast cancer tissues have low levels of PPAR-␥ and PGC-1 transcripts compared with those of the normal tissues. Further, lower levels of these molecules were found to be associated with poor clinical outcomes in breast cancer patients [26]. Interestingly, agonists of PPAR-␥ have also been shown to inhibit growth and proliferation of cancer cells by inducing apoptosis [27]. 3.1.2 Activation of Genes Involved in Tumor Invasion and Progression OXPHOS dysfunction has been shown to modulate tumor invasive properties by transcriptional upregulation of genes coding for specific members of matrix remodeling pathway and activation of tumor-specific marker genes. van Waveren et al. showed that OXPHOS dysfunction can alter the expression of nuclear genes coding for factors involved in extracellular matrix remodeling in human osteosarcoma cells [28]. These genes included members of the matrix metalloproteinases (MMPs) and tissue inhibitors of metalloproteinases (TIMP) family, urokinase plasminogen activator and its inhibitor, plasminogen-activator inhibitor 1 (PAI-1), and CTGF and CYR61 (members of the cysteine-rich 61, connective tissue growth factor and nephroblastoma-overexpressed [CCN] gene family of growth regulators). Changes at the protein level for some of these factors have also been observed, though at a lesser magnitude. The study also showed that osteosarcoma cells harboring mtDNA mutations were associated with an increased matrigel invasion [28]. Activation of tumor-specific marker genes in response to an OXPHOS dysfunction was observed by Amuthan et al. [12, 13]. They found that partial mtDNA

62


depletion (genetic stress) or treatment of cells with mitochondrial-specific inhibitors (metabolic stress) activated Ca2+ -dependent kinases (PKC, ERK1, ERK2, and calcineurin) and signaling pathways that in turn activated the expression of nuclear genes involved in tumor invasion and progression. Among these genes, the extracellular matrix protease cathepsin L, transforming growth factor ␤ (TGF-␤), and mouse melanoma antigen (MMA) were found to be upregulated in both genetically and metabolically stressed rhabdomyosarcoma and pulmonary carcinoma cells. Using in vitro Matrigel invasion assays and in vivo rat tracheal xenotransplants in Scid mice, they further showed that these genetically and metabolically stressed cells were ∼4to 6-fold more invasive than were their respective controls thereby implicating a role of OXPHOS dysfunction in inducing genes involved in tumor invasion and progression. The invasive behavior of mtDNA-depleted rhabdomyosarcoma and pulmonary carcinoma cells was reverted by the restoration of the mtDNA content [12, 13]. These findings clearly established a role of mitochondrial retrograde signaling in inducing phenotypic changes and tumor progression in osteosarcoma, rhabdomyosarcoma, and human pulmonary carcinoma cells. OXPHOS dysfunction was also shown to be associated with tumor aggressiveness in renal carcinoma cells [14]. Retrograde signaling in breast cancer has been reported by Delsite et al. They performed a comparative microarray analysis of the nuclear gene expression changes in a human breast cancer cell line (MDA-435) and its ␳o derivative devoid of mtDNA. They found that the expression of several nuclear genes involved in cell signaling, cell growth, energy metabolism, cell architecture, cell differentiation, and apoptosis were altered in the ␳o derivative of the breast cancer cell line [29]. Functional characterization of these genes is further required to gain a clear insight to the potential role of retrograde signaling in tumorigenesis. 3.1.3 Activation of Calcium Signaling Pathway Changes in intracellular calcium (from the approximate 100 nM at rest) mediate a number of processes that affect tumorigenesis, including angiogenesis, motility, cell cycle, differentiation, transcription, telomerase activity, and apoptosis [30]. Mitochondria take up Ca2+ in a process dependent on a ⌬⌿m [31]. Ca2+ uptake by mitochondria controls organelle metabolic activity, as pyruvate, ␣-ketoglutarate, and isocitrate dehydrogenases are activated by Ca2+ . In addition, some metabolite transporters have been shown to be regulated by Ca2+ as well and to enhance aerobic metabolism. Mitochondria, by buffering local [Ca2+ ] (generated by Ca2+ channels on the plasma membrane or the ER/SR), can also modulate intracellular Ca2+ [31]. Therefore, Ca2+ can regulate cell growth and survival by modulating OXPHOS function. Likewise, a defective OXPHOS impairs Ca2+ buffering. As described above, impaired Ca2+ buffering capacity by defective mitochondria can activate proteins involved in tumor invasion and progression, including PKC, ERK1, ERK2, and calcineurin. OXPHOS dysfunction has been shown to activate calcium-dependent signaling events in several tumor cell types. Avadhani and co-workers investigated the nuclear gene expression changes in C2C12 rhabdomyosarcoma and human A549 pulmonary carcinoma cells in response to OXPHOS dysfunction and found that


63

calcium-dependent signaling events were activated in these cells [12, 13, 32]. OXPHOS defects associated with decreased mitochondrial membrane potential (⌬⌿m) and ATP synthesis caused a sustained increase in intracellular calcium [Ca2+ ]i levels in these cells. They found that the expression of ryanodine receptor-1 (RyR-1) and ryanodine recptor-2 (RyR-2) genes was upregulated in C2C12 rhabdomyosarcoma and A549 pulmonary carcinoma cells, respectively. Increased RyR gene expression correlated with high levels of Ca2+ release from ER to the cytosol in response to caffeine stimulation (RyR agonist). Avadhani and co-workers suggested that the defect in ⌬⌿m was associated with the observed increase in [Ca2+ ]i levels as mitochondria are key players in sequestering cytosolic Ca2+ and a defect in ⌬⌿m would affect mitochondrial Ca2+ uptake ability. Restoration of the ⌬⌿m in both C2C12 rhabdomyosarcoma and A549 pulmonary carcinoma cells was found to revert the ryanodine receptor gene expression and [Ca2+ ]i levels to near normal levels suggesting a direct link between OXPHOS dysfunction and elevated [Ca2+ ]i in these cancer cells. Ca2+ responsive factors such as calcineurin, calcineurindependent NFATc (cytosolic counterpart of activated T-cell–specific nuclear factor), and JNK (c-Jun N-terminal kinase)-dependent ATF2 (activated transcription factor 2) were also found to be elevated in both C2C12 rhabdomyosarcoma and pulmonary carcinoma cells. Further, Ca2+ -dependent PKC (protein kinase C) and MAPK (mitogen activated protein kinase) genes were activated in C2C12 rhabdomyosarcoma and lung carcinoma cells, respectively. Another study by Avadhani and co-workers has shown that calcineurin that is activated in response to elevated cytosolic Ca2+ inactivates I-␬B␤ thereby leading to an activation of the NF-␬B/Rel family of transcription factors in the OXPHOS-deficient C2C12 rhabdomyosarcoma cells [33]. They suggested that the NF-␬B/Rel family of transcription factors might be involved in the activation of nuclear genes in response to OXPHOS dysfunction in these cells [33]. Rise in cytosolic Ca2+ in response to OXPHOS dysfunction has also been observed in rat pheochromocytoma PC12 cells. Luo et al. showed that the treatment of PC12 cells with mitochondrial uncoupler FCCP releases Ca2+ from internal stores leading to a rise in cytosolic Ca2+ , which in turn activates MAPK also suggesting a potential link between OXPHOS dysfunction and Ca2+ signaling in these cells [34]. 3.1.4 Activation of Redox Signaling Pathway Changes in nuclear gene expression in response to oxidative stress have been observed in a variety of tumor cells. mtDNA mutations or OXPHOS dysfunction predispose mitochondria to the generation of ROS leading to oxidative stress. Cellular oxidative stress has been implicated in initiation, promotion, and progression of carcinogenesis [35]. Low levels of ROS generation have been shown to be mitogenic in a variety of human and mouse cell types [36, 37]. For example, under cell culture conditions, 10 nM to 1 ␮M concentrations of both O2 •- and H2 O2 have been shown to stimulate the growth of hamster and rat fibroblasts [37]. Low levels of ROS were also shown to stimulate the growth of mouse epidermal cells, human fibroblasts, and human astrocytoma cells [38, 39]. The increased tumorigenesis of

64


NARP cells with a pathogenic mtDNA mutation described above [10] could also be related to increased ROS, as the mutation in ATPase 6 was reported to significantly increase ROS production [40]. Further, many chemical carcinogens initiate carcinogenesis by stimulating ROS production, which in turn activates the expression of early growth related genes such as c-fos, c-myc, and c-jun [41]. Although the molecular mechanism(s) of ROS-mediated cell growth are yet unclear, cellular ROS have been shown to activate the expression of redox-sensitive nuclear genes and signaling pathways. It has been shown that ROS activate the expression of nuclear factor kappa B (NF-␬B) and activator protein (AP-1) transcription factors [42, 43]. Both NF-␬B and AP-1 are redox-sensitive transcription factors. NF-␬B in its active form consists of two subunits, p50 and p65. In the absence of a stimulus, it is present in the cytoplasm in association with its inhibitory subunit I-␬B. In response to a stimulus that leads to the I-␬B subunit phosphorylation, NF-␬B dissociates from the I-␬B inhibitory complex and translocates to the nucleus where it activates the expression of specific target genes [44]. AP-1 is a heterodimer of transcription factors composed of Jun, Fos, or activating transcription factor (ATF) subunits and binds the AP-1 sites in the promoter region to activate the target gene expression [45]. Constitutive activation of NF-␬B and Ap-1 genes in response to oxidative stress has been observed in a variety of cancer cell types. Shi et al. demonstrated that OH radicals activated the expression of NF-␬B in Jurkat cells, macrophages, and mouse epidermal cells, and antioxidants that scavenge OH radicals inhibited its activation [46]. High levels of ROS were also shown to cause a constitutive elevation of NF-␬B and AP-1 transcription factors in transformed mouse keratinocyte cells [7]. Constitutive elevation of AP-1 activity has also been shown to be associated with the conversion of benign papillomas to malignant carcinomas in mouse epidermal cells suggesting that the target genes induced by activated AP-1 were involved in cell growth and metastasis [47]. Inhibition of Ap-1 activity by a dominant negative c-Jun mutant was shown to suppress the tumorigenic phenotype of the malignant mouse epidermal cells thereby blocking the tumor formation in nude mice [48]. Constitutive activation of NF-␬B has also been shown to induce genes involved in tumor progression and metastasis [49]. Further, treatment with an antioxidant, N-acetylcysteine (NAC), has been shown to inhibit the elevated expression of NF␬B and AP-1 levels in tumor cells implicating the role of ROS in the activation of redox-sensitive signaling pathways [7]. The activities of NF-␬B and AP-1 transcription factors are modulated by phosphorylation. Although the precise signaling events that activate the phosphorylation of these transcription factors in response to increase in cellular ROS levels are yet not fully understood, protein kinases are central in their activation. It has been observed that MAPK pathways are involved in signaling the activation of both NF␬B and Ap-1 transcription factors. There are three subtypes of MAPKs; the extracellular signal regulated kinases (ERKs), the c-jun N-terminal kinases (JNKs), and the p38 MAPKS. Gupta et al. showed that the NF-␬B activation in malignantly progressed mouse keratinocytes was associated with an increase in ERK-1/2 and p38 MAPK activities and that NAC treatment rapidly abolished the increase in MAPK activities [7]. Direct activation of I-␬B by MEKK1 (MAPK/ERK kinase kinase-1)


65

has also been demonstrated in vitro [50]. Activation of AP-1 by MAPK in response to various stimuli has been demonstrated [51]. There is also evidence that antioxidants can attenuate MAPK activation suggesting that MAPK signaling cascades are activated in response to cellular oxidative stress conditions [7]. 3.1.5 Altered Antioxidant Defense Pathway The antioxidant defense enzymes have been found to be altered in several human tumors. Superoxide dismutases (SODs) are the first line of cellular defense against increase in intracellular ROS. There are two forms of intracellular SODs: a manganese-containing SOD (MnSOD) that localizes to the mitochondria and a copper-zinc–containing SOD that localizes to the cytosol. Most human tumor cell types have been found to be deficient in their antioxidant defense function. Antioxidant enzymes including catalase and glutathione peroxidase have also been shown to be lowered in cancer cells [52, 53] suggesting that tumor cells have a deficient antioxidant function. Studies have shown that antioxidant treatment prevents the malignant transformation of cancer cells. Increased expression of MnSOD has been shown to suppress cancer phenotypes in several murine and human cancer cells. St Clair et al. showed that overexpression of human MnSOD in mouse cells significantly reduced the frequency of radiation-induced neoplastic transformation implicating a direct link between mitochondrial antioxidants and neoplastic transformation [54]. Zhao et al. demonstrated that in response to tumor promoters (phorbol esters), transgenic mice expressing human MnSOD in the skin showed a significant reduction in papilloma formation compared with that of the non-transgenic controls [55]. They further found that the decreased papilloma formation was associated with a delay and reduction in AP-1 transcription factor binding activity suggesting that the MnSOD overexpression suppressed the tumor formation and Ap-1 activation in these transgenic animals [55]. Overexpression of human MnSOD has also been shown to suppress the malignant phenotype of human pancreatic and breast cancer cells. Li et al. demonstrated that MnSOD-overexpressing MCF-7 breast cancer cells showed a marked inhibition in growth rate in vitro under culture conditions and in vivo upon inoculation in nude mice when compared with the wild-type MCF-7 cells [56]. In another study, they demonstrated that the transcriptional and DNA binding ability of NF-␬B and AP-1 were reduced by ∼50% in MnSOD-overexpressing MCF-7 cells [57]. They further showed that the expression of NF-␬B and AP-1 responsive genes was downregulated in these cells when compared with that of the wild-type MCF-7 cells suggesting that the tumor-suppressive effects of MnSOD were associated with the inhibition of NF-␬B and Ap-1 activities that in turn downregulated the genes involved in malignant progression [57]. 3.1.6 Altered Mitochondrial Morphology, Cell Surface, and Architecture Mitochondrial fission and fusion are the central events that regulate mitochondrial morphology. Changes in mitochondrial morphology appear to be a key event during retrograde signaling. Studies have implicated a role of mitochondrial morphology in

66


modulating respiratory activity. For example, downregulation of mitofusin 2 (Mfn 2) decreases mitochondrial respiration and increases the cellular glucose uptake levels for glycolysis. Further, ectopic upregulation of Mfn 2 increases the expression of OXPHOS subunits, glucose oxidation, and ⌬⌿m [58]. Optic Atrophy 1 (Opa 1) is an important player in the formation of mitochondrial cristae. Downregulation of Opa 1 has been shown to induce the loss of mitochondrial respiration and increased mitochondrial fragmentation [59]. Changes in mitochondrial morphology and ultrastructure were reported in different tumor cell types over several decades. Tumor cell mitochondria were found to contain abnormal size, shape (dumbbell or cup), cristae organization, or inclusions. For example, rapidly growing hepatomas were found to contain small-size mitochondria with fewer cristae, whereas slowly growing hepatomas were found to contain large-size mitochondria with densely packed cristae [60]. Changes in the mitochondrial morphology in tumor cells could therefore affect their respiratory activity. Conversely, it is also possible that in response to altered respiratory activity or OXPHOS dysfunction, the retrograde signaling events in tumor cells affect the expression of nuclear genes involved in the regulation of mitochondrial morphology and ultrastructure. Changes in cell morphology and architecture are a characteristic feature of many cancer cells. These changes occur in cell surface charge, glycoprotein composition, membrane organization, and/or the ability of cells to agglutinate on lectin. The initial evidence of a potential role of OXPHOS dysfunction in inducing cell surface changes was obtained in yeast. In 1980, Evans and colleagues showed that mitochondrial petite mutations in yeast that cause respiratory deficiency lead to drastic changes in cell surface charge, cell wall organization, and lectin agglutinability in contrast with the wild-type yeast cells suggesting that the OXPHOS deficiency in yeast altered the retrograde signaling pathway leading to gene expression changes and hence altered cell surface properties [61]. Preliminary evidence in literature has also suggested a link between OXPHOS dysfunction and cell surface properties in mammalian cells. For example, Soslau et al. showed that baby hamster kidney cells when treated with ethidium bromide (Etbr) show an altered glycoprotein composition in the plasma membrane [62]. Further, they also showed that the glycopeptide elution profile of these Etbr-treated cells was similar to that of the cells transformed with Rous sarcoma virus. Low concentrations of Etbr are known to intercalate to mtDNA and abrogate mtDNA replication. Exponentially growing cells when treated with Etbr (25 ng/mL to 2 ␮g/mL) undergo mtDNA loss leading to either partial or complete loss of mtDNA molecules, generating an OXPHOS defect. These findings therefore suggest that altered retrograde signaling events during OXPHOS dysfunction could activate nuclear genes or pathways that cause global changes in cell surface properties. 3.1.7 Activation of Antiapoptotic Genes Activation of antiapoptotic genes is a common occurrence in most human cancers. However, retrograde signaling pathway has also been shown to activate


67

antiapoptotic genes (e.g., Bcl-xL and Bcl-2) in certain cancer cell types [12, 63]. Increased expression of the antiapoptotic genes may also contribute to metabolic homeostasis. It has been shown that members of the Bcl-2 antiapoptotic family of proteins regulate ATP/ADP exchange across the mitochondrial membranes and can prevent the loss of mitochondrial respiration during apoptosis [64]. Manfredi and colleagues showed that Bcl-2/Bcl-xL can improve OXPHOS function in cells harboring pathogenic mtDNA mutations. The effect of Bcl-2 overexpression in mutated cells was found to be independent from apoptosis and was suggested to modulate the adenine nucleotide exchange between mitochondria and cytosol. This study therefore provided evidence that when OXPHOS function is reduced, Bcl-2 expression can improve it by regulating the levels of adenine nucleotides inside the mitochondria [64].

3.2 Mechanism of Retrograde Signaling The mechanism of retrograde signaling was first identified in yeast and it is still poorly understood in other systems. Three retrograde regulatory genes (RTG1, RTG2, and RTG3) play a central role in retrograde signaling in yeast [65] (Fig. 3). Rtg1p and Rtg3p are transcription factors, whereas Rtg2p is mitochondrial function sensor. During normal mitochondrial function, Rtg1p and Rtg3p form heterodimers

P Nucleus

Rtg3p

Rtg1p

Target gene

GTCAC

R box

P Rtg3p

P P P Rtg3p P

Rtg1p

Rtg2p Rtg1p

Sensor

Mitochondrial Dysfunction

Cytoplasm

Fig. 3 Schematic model of retrograde signaling in yeast. RTG genes play a central role in retrograde signaling. Rtg1p and Rtg3p are sequestered in cytoplasm during normal mitochondrial function. Rtg3p is highly phosphorylated during this state. In response to mitochondrial dysfunction (sensed by Rtg2p), the Rtg3p undergoes partial dephosphorylation. Both Rtg3p and Rtg1p translocate to the nucleus and activate target gene expression leading to changes in nuclear genes and/or signaling pathways that in turn can alter cellular function and homeostasis

68


and are sequestered in the cytoplasm. Rtg3p is phosphorylated at multiple sites during this state. In response to mitochondrial dysfunction (sensed by Rtg2p), a retrograde signaling pathway gets activated. The Rtg3p undergoes partial dephosphorylation and translocates to the nucleus. Rtg1p also follows and translocates to the nucleus. Both Rtg1p and Rtg3p activate transcription at the target genes. In the absence of Rtg2p, the Rtg1p/Rtg3p complex remains in the cytoplasm during active retrograde signaling pathway implicating its role as mitochondrial function sensor and in signal relay to the Rtg1p/Rtg3p complex [65]. The precise mechanism of retrograde signaling in mammalian systems is yet not known. The mammalian homologs of RTG genes have yet not been identified. However, human MYC protein has been found to share a significant homology with yeast Rtg3p, and it is involved in the regulation of many cellular functions such as regulation of glycolysis, cellular stress response cell cycle progression, and apoptosis [66, 67].

4 Conclusion Mitochondrial dysfunction has been reported in a variety of human cancers. Mitochondrial-nuclear intergenomic signaling is an important phenomenon in regulating mitochondrial and cellular function and homeostasis. Although the role of mtDNA mutations or OXPHOS defects in cancer progression is poorly understood, a considerable body of evidence has now shown that mitochondrial function has an important role in different aspects of tumorigenesis. Processes such as cell survival and cell invasion can be directly influenced by OXPHOS function. Less clear is the association between cell cycle progression and an OXPHOS defect. However, specific examples, such as SDH mutations in paragangliomas and pheochromocytomas, provide proof of principle that OXPHOS defects can somehow stimulate cell division. It appears likely that some properties related to OXPHOS dysfunction can have an advantageous consequence for tumor development. However, this feature could impair cell performance at a different level. Therefore, tumor cells with an OXPHOS defect, in some cases, have compensatory mechanisms, such as an increase in mitochondrial biogenesis via overexpression of PGC-1 family members. Studies on the interplay between these metabolic and signaling pathways will help us better understand the basic mechanisms of tumor progression.

Note A recent study by Ishikawa et al. (Ishikawa K, Takenaga K, Akimoto M, et al. ROS-generating mitochondrial DNA mutations can regulate tumor cell metastasis.


69

Science. 2008;320(5876):661–664) showed that specific mtDNA changes in complex I subunit can increase the metastatic properties of tumor cells.

References 1. Warburg O. On the origin of cancer cells. Science 1956; 123(3191):309–314. 2. Pedersen PL, Greenawalt JW, Chan TL, Morris HP. A comparison of some ultrastructural and biochemical properties of mitochondria from Morris hepatomas 9618A, 7800, and 3924A. Cancer Res 1970; 30(11):2620–2626. 3. Polyak K, Li Y, Zhu H, et al. Somatic mutations of the mitochondrial genome in human colorectal tumours. Nat Genet 1998; 20(3):291–293. 4. Fliss MS, Usadel H, Caballero OL, et al. Facile detection of mitochondrial DNA mutations in tumors and bodily fluids. Science 2000; 287(5460):2017–2019. 5. Liu VW, Shi HH, Cheung AN, et al. High incidence of somatic mitochondrial DNA mutations in human ovarian carcinomas. Cancer Res 2001; 61(16):5998–6001. 6. Nagy A, Wilhelm M, Sukosd F, Ljungberg B, Kovacs G. Somatic mitochondrial DNA mutations in human chromophobe renal cell carcinomas. Genes Chromosomes Cancer 2002; 35(3):256–260. 7. Gupta A, Rosenberger SF, Bowden GT. Increased ROS levels contribute to elevated transcription factor and MAP kinase activities in malignantly progressed mouse keratinocyte cell lines. Carcinogenesis 1999; 20(11):2063–2073. 8. Irani K, Xia Y, Zweier JL, et al. Mitogenic signaling mediated by oxidants in Ras-transformed fibroblasts. Science 1997; 275(5306):1649–1652. 9. Coller HA, Khrapko K, Bodyak ND, Nekhaeva E, Herrero-Jimenez P, Thilly WG. High frequency of homoplasmic mitochondrial DNA mutations in human tumors can be explained without selection. Nat Genet 2001; 28(2):147–150. 10. Petros JA, Baumann AK, Ruiz-Pesini E, et al. mtDNA mutations increase tumorigenicity in prostate cancer. Proc Natl Acad Sci USA 2005; 102(3):719–724. 11. Shidara Y, Yamagata K, Kanamori T, et al. Positive contribution of pathogenic mutations in the mitochondrial genome to the promotion of cancer by prevention from apoptosis. Cancer Res 2005; 65(5):1655–1663. 12. Amuthan G, Biswas G, Ananadatheerthavarada HK, Vijayasarathy C, Shephard HM, Avadhani NG. Mitochondrial stress-induced calcium signaling, phenotypic changes and invasive behavior in human lung carcinoma A549 cells. Oncogene 2002; 21(51):7839–7849. 13. Amuthan G, Biswas G, Zhang SY, Klein-Szanto A, Vijayasarathy C, Avadhani NG. Mitochondria-to-nucleus stress signaling induces phenotypic changes, tumor progression and cell invasion. EMBO J 2001; 20(8):1910–1920. 14. Simonnet H, Alazard N, Pfeiffer K, et al. Low mitochondrial respiratory chain content correlates with tumor aggressiveness in renal cell carcinoma. Carcinogenesis 2002; 23(5): 759–768. 15. Simonnet H, Demont J, Pfeiffer K, et al. Mitochondrial complex I is deficient in renal oncocytomas. Carcinogenesis 2003; 24(9):1461–1466. 16. Savagner F, Franc B, Guyetant S, Rodien P, Reynier P, Malthiery Y. Defective mitochondrial ATP synthesis in oxyphilic thyroid tumors. J Clin Endocrinol Metab 2001; 86(10):4920–4925. 17. Akimoto M, Niikura M, Ichikawa M, et al. Nuclear DNA but not mtDNA controls tumor phenotypes in mouse cells. Biochem Biophys Res Commun 2005; 327(4):1028–1035. 18. Tomlinson IP, Alam NA, Rowan AJ, et al. Germline mutations in FH predispose to dominantly inherited uterine fibroids, skin leiomyomata and papillary renal cell cancer. Nat Genet 2002; 30(4):406–410. 19. Astuti D, Latif F, Dallol A, et al. Gene mutations in the succinate dehydrogenase subunit SDHB cause susceptibility to familial pheochromocytoma and to familial paraganglioma. Am J Hum Genet 2001; 69(1):49–54.

70


20. Niemann S, Muller U. Mutations in SDHC cause autosomal dominant paraganglioma, type 3. Nat Genet 2000; 26(3):268–270. 21. Baysal BE, Ferrell RE, Willett-Brozick JE, et al. Mutations in SDHD, a mitochondrial complex II gene, in hereditary paraganglioma. Science 2000; 287(5454):848–851. 22. Srivastava S, Barrett JN, Moraes CT. PGC-1alpha/beta upregulation is associated with improved oxidative phosphorylation in cells harboring nonsense mtDNA mutations. Hum Mol Genet 2007; 16(8):993–1005. 23. Savagner F, Mirebeau D, Jacques C, et al. PGC-1-related coactivator and targets are upregulated in thyroid oncocytoma. Biochem Biophys Res Commun 2003; 310(3):779–784. 24. Villani G, Greco M, Papa S, Attardi G. Low reserve of cytochrome c oxidase capacity in vivo in the respiratory chain of a variety of human cell types. J Biol Chem 1998; 273(48): 31829–31836. 25. Wu Z, Puigserver P, Andersson U, et al. Mechanisms controlling mitochondrial biogenesis and respiration through the thermogenic coactivator PGC-1. Cell 1999; 98(1):115–124. 26. Jiang WG, Douglas-Jones A, Mansel RE. Expression of peroxisome-proliferator activated receptor-gamma (PPARgamma) and the PPARgamma co-activator, PGC-1, in human breast cancer correlates with clinical outcomes. Int J Cancer 2003; 106(5):752–757. 27. Ohta K, Endo T, Haraguchi K, Hershman JM, Onaya T. Ligands for peroxisome proliferatoractivated receptor gamma inhibit growth and induce apoptosis of human papillary thyroid carcinoma cells. J Clin Endocrinol Metab 2001; 86(5):2170–2177. 28. van Waveren C, Sun Y, Cheung HS, Moraes CT. Oxidative phosphorylation dysfunction modulates expression of extracellular matrix–remodeling genes and invasion. Carcinogenesis 2006; 27(3):409–418. 29. Delsite R, Kachhap S, Anbazhagan R, Gabrielson E, Singh KK. Nuclear genes involved in mitochondria-to-nucleus communication in breast cancer cells. Mol Cancer 2002; 1(1):6. 30. Monteith GR, McAndrew D, Faddy HM, Roberts-Thomson SJ. Calcium and cancer: targeting Ca2+ transport. Nat Rev Cancer 2007; 7(7):519–530. 31. Giacomello M, Drago I, Pizzo P, Pozzan T. Mitochondrial Ca2+ as a key regulator of cell life and death. Cell Death Differ 2007; 14(7):1267–1274. 32. Biswas G, Adebanjo OA, Freedman BD, et al. Retrograde Ca2+ signaling in C2C12 skeletal myocytes in response to mitochondrial genetic and metabolic stress: a novel mode of interorganelle crosstalk. EMBO J 1999; 18(3):522–533. 33. Biswas G, Anandatheerthavarada HK, Zaidi M, Avadhani NG. Mitochondria to nucleus stress signaling: a distinctive mechanism of NFkappaB/Rel activation through calcineurin-mediated inactivation of IkappaBbeta. J Cell Biol 2003; 161(3):507–519. 34. Luo Y, Bond JD, Ingram VM. Compromised mitochondrial function leads to increased cytosolic calcium and to activation of MAP kinases. Proc Natl Acad Sci USA 1997; 94(18): 9705–9710. 35. Cerutti PA. Prooxidant states and tumor promotion. Science 1985; 227(4685):375–381. 36. Burdon RH, Rice-Evans C. Free radicals and the regulation of mammalian cell proliferation. Free Radic Res Commun 1989; 6(6):345–358. 37. Burdon RH, Gill V, Rice-Evans C. Oxidative stress and tumour cell proliferation. Free Radic Res Commun 1990; 11(1–3):65–76. 38. Arora-Kuruganti P, Lucchesi PA, Wurster RD. Proliferation of cultured human astrocytoma cells in response to an oxidant and antioxidant. J Neurooncol 1999; 44(3):213–221. 39. Murrell GA, Francis MJ, Bromley L. Modulation of fibroblast proliferation by oxygen free radicals. Biochem J 1990; 265(3):659–665. 40. Mattiazzi M, Vijayvergiya C, Gajewski CD, et al. The mtDNA T8993G (NARP) mutation results in an impairment of oxidative phosphorylation that can be improved by antioxidants. Hum Mol Genet 2004; 13(8):869–879. 41. Amstad P, Crawford D, Muehlematter D, Zbinden I, Larsson R, Cerutti P. Oxidants stress induces the proto-oncogenes, C-fos and C-myc in mouse epidermal cells. Bull Cancer 1990; 77(5):501–502.


71

42. Abate C, Patel L, Rauscher FJ 3rd, Curran T. Redox regulation of fos and jun DNA-binding activity in vitro. Science 1990; 249(4973):1157–1161. 43. Flohe L, Brigelius-Flohe R, Saliou C, Traber MG, Packer L. Redox regulation of NF-kappa B activation. Free Radic Biol Med 1997; 22(6):1115–1126. 44. Baeuerle PA. The inducible transcription activator NF-kappa B: regulation by distinct protein subunits. Biochim Biophys Acta 1991; 1072(1):63–80. 45. Karin M, Liu Z, Zandi E. AP-1 function and regulation. Curr Opin Cell Biol 1997; 9(2): 240–246. 46. Shi X, Dong Z, Huang C, et al. The role of hydroxyl radical as a messenger in the activation of nuclear transcription factor NF-kappaB. Mol Cell Biochem 1999; 194(1–2):63–70. 47. Domann FE Jr, Levy JP, Finch JS, Bowden GT. Constitutive AP-1 DNA binding and transactivating ability of malignant but not benign mouse epidermal cells. Mol Carcinog 1994; 9(2):61–66. 48. Domann FE, Levy JP, Birrer MJ, Bowden GT. Stable expression of a c-JUN deletion mutant in two malignant mouse epidermal cell lines blocks tumor formation in nude mice. Cell Growth Differ 1994a; 5(1):9–16. 49. Nakshatri H, Bhat-Nakshatri P, Martin DA, Goulet RJ Jr, Sledge GW Jr. Constitutive activation of NF-kappaB during progression of breast cancer to hormone-independent growth. Mol Cell Biol 1997; 17(7):3629–3639. 50. Lee FS, Hagler J, Chen ZJ, Maniatis T. Activation of the IkappaB alpha kinase complex by MEKK1, a kinase of the JNK pathway. Cell 1997; 88(2):213–222. 51. Karin M. The regulation of AP-1 activity by mitogen-activated protein kinases. J Biol Chem 1995; 270(28):16483–16486. 52. Gupta A, Butts B, Kwei KA, et al. Attenuation of catalase activity in the malignant phenotype plays a functional role in an in vitro model for tumor progression. Cancer Lett 2001; 173(2):115–125. 53. Oberley LW, Oberley TD. Role of antioxidant enzymes in cell immortalization and transformation. Mol Cell Biochem 1988; 84(2):147–153. 54. St Clair DK, Wan XS, Oberley TD, Muse KE, St. Clair WH. Suppression of radiation-induced neoplastic transformation by overexpression of mitochondrial superoxide dismutase. Mol Carcinog 1992; 6(4):238–242. 55. Zhao Y, Xue Y, Oberley TD, et al. Overexpression of manganese superoxide dismutase suppresses tumor formation by modulation of activator protein-1 signaling in a multistage skin carcinogenesis model. Cancer Res 2001; 61(16):6082–6088. 56. Li JJ, Oberley LW, St. Clair DK, Ridnour LA, Oberley TD. Phenotypic changes induced in human breast cancer cells by overexpression of manganese-containing superoxide dismutase. Oncogene 1995; 10(10):1989–2000. 57. Li JJ, Oberley LW, Fan M, Colburn NH. Inhibition of AP-1 and NF-kappaB by manganesecontaining superoxide dismutase in human breast cancer cells. FASEB J 1998; 12(15): 1713–1723. 58. Pich S, Bach D, Briones P, et al. The Charcot-Marie-Tooth type 2A gene product, Mfn2, up-regulates fuel oxidation through expression of OXPHOS system. Hum Mol Genet 2005; 14(11):1405–1415. 59. Chen H, Chomyn A, Chan DC. Disruption of fusion results in mitochondrial heterogeneity and dysfunction. J Biol Chem 2005; 280(28):26185–26192. 60. Hruban Z, Swift H, Rechcigl M Jr. Fine structure of transplantable hepatomas of the rat. J Natl Cancer Inst 1965; 35(3):459–495. 61. Evans IH, Diala ES, Earl A, Wilkie D. Mitochondrial control of cell surface characteristics in Saccharomyces cerevisiae. Biochim Biophys Acta 1980; 602(1):201–206. 62. Soslau G, Fuhrer JP, Nass MM, Warren L. The effect of ethidium bromide on the membrane glycopeptides in control and virus-transformed cells. J Biol Chem 1974; 249(10):3014–3020. 63. Dey R, Moraes CT. Lack of oxidative phosphorylation and low mitochondrial membrane potential decrease susceptibility to apoptosis and do not modulate the protective effect of Bcl-x(L) in osteosarcoma cells. J Biol Chem 2000; 275(10):7087–7094.

72


64. Manfredi G, Kwong JQ, Oca-Cossio JA, et al. BCL-2 improves oxidative phosphorylation and modulates adenine nucleotide translocation in mitochondria of cells harboring mutant mtDNA. J Biol Chem 2003; 278(8):5639–5645. 65. Sekito T, Thornton J, Butow RA. Mitochondria-to-nuclear signaling is regulated by the subcellular localization of the transcription factors Rtg1p and Rtg3p. Mol Biol Cell 2000; 11(6):2103–2115. 66. Miceli MV, Jazwinski SM. Common and cell type-specific responses of human cells to mitochondrial dysfunction. Exp Cell Res 2005; 302(2):270–280. 67. Epstein CB, Waddle JA, Hale WT, et al. Genome-wide responses to mitochondrial dysfunction. Mol Biol Cell 2001; 12(2):297–308.

Regulation of Glucose and Energy Metabolism in Cancer Cells by Hypoxia Inducible Factor 1 ´ ´ César Ferreira and Elida Tulio Geralda Campos

Abstract A crucial aspect of carcinogenesis is the adaptation of cells to the changing environmental conditions of the tumor mass. To grow and proliferate, the cells need nutrients that have to be delivered by surrounding vessels. In addition, they have to cope with a gradient of oxygen tension as the tumor mass grows. Hypoxia inducible factor 1 (HIF-1) mediates important responses to physiologic and pathologic processes that allow angiogenesis, erythropoiesis, cell proliferation and survival, and metabolic changes to take place. It is a heterodimeric transcription factor regulated by a complex network. Among the major metabolic changes evident in cancer cells is the activation of the glycolytic pathway, even in the presence of oxygen. HIF-1 regulates glucose and energy metabolism by directly activating the gene expression of glycolytic enzymes, regulatory enzymes, and glucose transporters. HIF-1␣ (the regulated subunit of the HIF-1 complex) is regulated mainly at the posttranscriptional level through mechanisms that block its proteasomal degradation depending on the oxygen concentration. HIF-1 is overexpressed in many tumors, and evidence is mounting that HIF-1 regulation of glucose and energy metabolism is important in carcinogenesis and tumor progression. Keywords HIF-1 · HIF-1␣ · Glycolysis · von Hippel–Lindau tumor suppressor protein (pVHL) · Lactate · Glucose transporter · Regulation of gene expression · Hypoxia response element (HRE) · Prolyl hydroxylases

1 Oxygen Homeostasis Cells respond to extracellular and intracellular stimuli to maintain homeostasis, and hypoxia is one of the most fundamental environmental stimuli. Oxygen homeostasis is crucial to maintain the approximately 1014 cells in the adult human body with

´ E.G. Campos (B) Departamento de Biologia Celular, Universidade de Bras´ılia, CEP 70910-900, Bras´ılia, DF, Brazil e-mail: [email protected] S.P. Apte, R. Sarangarajan (eds.), Cellular Respiration and Carcinogenesis, DOI 10.1007/978-1-59745-435-3 6, C Humana Press, a part of Springer Science+Business Media, LLC 2009

73

74

´ T.C. Ferreira, E.G. Campos

an adequate supply of O2 . The maintenance of the O2 levels within strict limits is essential throughout life but is particularly important during periods of rapid cellular proliferation, such as in normal growth or during the development of neoplasias. In normal tissues, the average oxygen tension is 40 to 50 mm Hg, whereas in tumors these levels are 5 to 10 mm Hg. Control of O2 levels in the tissues of animals occurs through the combination of gene regulation and biochemical and physiologic mechanisms. Low levels of oxygen in the environment lead cells and tissues to adapt to this adverse condition. Significant changes in gene expression and activation of signaling pathways occurs in response to hypoxia, a characteristic condition in solid tumors, resulting in an elevated transcription of angiogenic, hematopoietic, and some metabolic genes. Hypoxia inducible factor 1 (HIF-1) was discovered in 1992 by Gregg L. Semenza and Guang L. Wang at the Johns Hopkins University School of Medicine. It is currently known as the central regulator of the response to low levels of oxygen. HIF-1 is a heterodimeric transcription factor and regulates genes whose products participate in glycolysis, glucose transport, angiogenesis, cell proliferation and survival, pH regulation, and cell migration.

2 HIF-1 Structure HIF-1␣ protein is one of the subunits that form HIF-1 and belongs to the basic helix-loop-helix (bHLH)-PAS family of transcription factors. PAS is an acronym that refers to the Drosophila proteins Period (or PER), ARNT, and Single-minded or (SIM), the first proteins in which this motif was identified. HIF-1␣ has the bHLHPAS domain, which is in the N-terminal region of the protein and has two repeats (A and B) in its structure. This subunit binds to the second subunit that forms HIF-1, the aryl hydrocarbon receptor nuclear translocator (ARNT), also referred as HIF-1␤. ARNT is a nuclear protein constitutively expressed that also participates in the transcriptional response to xenobiotics when interacting with the aryl hydrocarbon receptor (AHR). As with HIF-1␣, the ␤ subunit also belongs to the bHLH-PAS family of transcription factors. HIF-1␣ and HIF-1␤ share some amino acid sequence similarities. The bHLH-PAS domains in both subunits are crucial for the heterodimerization and binding of these proteins to DNA. HIF-1␣ is an 826-amino-acid protein that has two transactivation domains (TADs), one in the N-terminal region (TAD-N) that overlaps with the oxygen degradation dependent (ODD) domain and another in the C-terminal region (TAD-C), the latter also found in the HIF-1␤ (Fig. 1). The TAD domains are localized at amino acids 531 to 575 (TAD-N) and 786 to 826 (TAD-C) of the HIF-1␣. The TAD-C domain of HIF-1␣ interacts with the transcriptional activators CREB binding protein (CBP) and p300. Maintenance of cysteine 800 in the TAD-C in a reduced state is essential for p300/CBP binding. Therefore, HIF-1␣ redox state is involved in its activation. HIF-1␣ and p300 interaction requires a leucine-rich hydrophobic interface that is regulated by the reversible change between hydrophobic and hydrophilic Cys800 of HIF-1␣. The reduced state of this cysteine is maintained by redox factor-1

Regulation of Glucose and Energy Metabolism in Cancer Cells by HIF-1

75

Fig. 1 HIF-1 proteins ␣ and ␤ have similar structural domains. HIF-1␣ (upper drawing) is a 826-amino-acid protein with a basic helix-loop-helix-PAS (bHLH-PAS) domain in the N-terminal and repeats A and B. There are two transactivation domains (TADs), one located in the N-terminal region (TAD-C), which overlaps with the oxygen degradation dependent (ODD) domain, and other in the C-terminal region (TAD-C). Both TADs are connected by an inhibitory domain (ID). HIF1␤ (lower drawing) is a 789-amino-acid protein, which also belongs to the bHLH-PAS family of transcription factors. As with HIF-1␣, HIF-1␤ also has the bHLH-PAS domain in the N-terminal and repeats A and B. However, it has only one TAD, which is located in the C-terminal of the protein (see Color Insert)

[1] and mediated by thioredoxin (TRX). Therefore, compounds that inhibit thioredoxin cause inhibition of HIF-1␣–mediated transactivation of HIF-1 target genes. Both TAD-C and TAD-N domains are connected by a inhibitory domain (ID). The ODD domain is a proline-serine-threonine–rich domain involved in the regulation of protein stability as a function of O2 concentration. In HIF-1␣, the main target amino acids for hydroxylation are localized in this domain (described below). HIF1␣ protein contains two nuclear localization signals (NLS), one in the N-terminal (NLS-N) and the other in the C-terminal, which encompass the amino acids 17 to 23 and 718 to 721, respectively [1].

3 HIF-1␣ Regulation HIF-1␣ regulation is complex and is governed by posttranslational mechanisms (Fig. 2). Protein stability and activity are regulated by covalent modifications such as hydroxylation, acetylation, S-nitrosation, and phosphorylation. Among them, hydroxylation and phosphorylation are the most studied. In normoxia conditions, hydroxylated HIF-1␣ is recognized by the von Hippel–Lindau tumor suppressor protein (pVHL), which is part of a polyubiquitination complex. HIF-1␣ is hydroxylated by a HIF prolyl hydroxylase (PHD). pVHL recognizes the hydroxylated proline residues (Pro402 and Pro564) located in the human HIF-1␣ ODD domain. PHD is a dioxygenase that uses O2 and 2-oxoglutarate (2-OG) as substrates, the latter generated by the tricarboxylic acid (TCA) cycle or by the transamination of amino acids. This enzyme transfers the first oxygen atom to the proline residue, and the second oxygen atom reacts with the 2-OD producing succinate. In addition to oxygen and 2-OG, the PHD proteins require Fe2+ and ascorbate as cofactors [2]. The first 2-OG dioxygenase to be identified was procollagen prolyl hydroxylase, which participates in collagen synthesis and catalyzes trans-4-hydroxylation reactions. There are three PHD isoenzymes: 1, 2, and 3. PHDs hydroxylate specific proline residues inside a highly conserved domain whose sequence is LXXLAP

76


Fig. 2 Mechanisms of HIF-1␣ regulation. Under normoxia conditions, HIF-1␣ is hydroxylated by prolyl hydroxylase (PHD), an enzyme that belongs to the family of dioxygenases and requires O2 and 2-oxoglutarate (2-OG) as substrates. This enzyme also uses iron as cosubstrate, which is maintained in the reduced form (Fe+2 ) by ascorbic acid. Iron can be scavenged by treatment of cells in normoxia with desferrioxamine (DFO), mimicking a hypoxia condition. Nickel and cobalt also promote HIF-1␣ stabilization in normoxia condition by competing with iron. The pVHL recognizes the hydroxylated (OH) proline residues 402 and 564. pVHL is a member of E3-ubiquitinligase complex formed by elongin C (EloC), elongin B (EloB), and cullin 2 (CUL2) and RBX1.


77

(where X indicates any amino acid; P indicates the hydroxylated proline; L, leucine and A, alanine). Concerning the cellular localization, PHD1 is found exclusively in the nucleus, PHD2 in the cytoplasm, and PHD3 is found in both cytoplasm and nucleus, with predominance in the nucleus. Selective blockage of HIF-1␣ expression using RNA interference technology leads to a complete loss of PHD2 and PHD3 gene expression under hypoxia, whereas PHD1 mRNA levels remain unchanged [2]. It seems that hypoxic induction of PHD2 and PHD3 is critically dependent on HIF-1␣. This negative feedback mechanism probably limits HIF-1␣ accumulation in hypoxia and leads to accelerated degradation on reoxygenation after long period of hypoxia. RNA interference showed the participation of PHD2 on HIF-1␣ hydroxylation. Indeed, the specific silencing of PHD2 stabilized the HIF1␣ in normoxia conditions in several tested human cells. However, the silencing of PHD1 and PHD3 has no effect in HIF-1␣ stabilization in both normoxia and reoxgenation of cells briefly exposed to hypoxia. Therefore, it seems that PHD2 is the critical oxygen sensor that maintains HIF-1␣ in a low level in normoxic conditions [3]. Under normoxia, HIF-1␣ activity is also regulated by hydroxylation of asparagine 803 in the TAD-C domain, catalyzed by an enzyme known as factor inhibiting HIF-1 (FIH), asparaginyl hydroxylase. FIH also belongs to the superfamily of 2-OD–dependent dioxygenases whose activity requires O2 as substrate. The hydroxylation of Asn803 does not lead to HIF-1␣ degradation. Therefore, this residue is not directly involved in the O2 -sensing mechanism. Hydroxylation of Asn803 abolishes the interaction between HIF-1␣ and its transcriptional coactivators CBP and p300. CBP and p300 are coactivators that participate in the activities of many different transcription factors. The CBP/p300 proteins have a key role in coordinating and integrating multiple signal-dependent events. They make possible the adequate expression levels of genes in response to diverse physiologic stimuli and influences; for example, proliferation, differentiation, and apoptosis.

Fig. 2 (Continued) This complex is responsible for HIF-1␣ polyubiquitination prior to degradation by the proteasome. Factor inhibiting HIF-1 (FIH-1) hydroxylates the asparagine 803 (N803) residue of HIF-1␣ preventing interaction with the CBP/p300. ARD1 acetylates the lysine 532 residue of HIF-1␣ leading to its degradation. Hypoxia, oncogenes, and growth factors can activate the phosphatidylinositol 3-kinase (PI3K) pathway, which has AKT as a downstream target. AKT can also be activated by phosphatidylinositol-dependent protein kinase 1 (PDK-1). PI3K/AKT activation causes HIF-1␣ overexpression mediated by mTOR (mammalian target of rapamycin). MEK (MAPK kinase) participates in the phosphorylation pathway activating the extracellular signal–regulated kinases (ERK1/2), which phosphorylate HIF-1␣. Phosphorylated HIF-1␣ is stabilized and activated, enters into the nucleus, forms the heterodimer with HIF-1␤, is coactivated by CBP/p300 complex, and induces the transcription of more than 100 target genes. A cysteine residue is kept in the reduced state by redox factor-1 (REF-1) mediated by thioredoxin (TRX), which participate in HIF-1␣ stabilization. Under hypoxia, nitric oxide (NO) disfavors HIF-1␣ accumulation (activates PHDs), and under normoxia condition, NO “mimics” hypoxia conditions (inhibits PHDs) and leads to HIF-1␣ accumulation . Some evidence points to the role of reactive oxygen species (ROS) as messengers that regulate HIF-1 activity (see Color Insert)

78


pVHL is a component of the E3 ubiquitin-protein ligase complex that drives the proteolysis in the proteasome, preventing the transcriptional activation of its target genes. E3 ubiquitin ligases are a family of proteins that participate in the degradation and activity of many cellular proteins and function together with ubiquitinactivating enzyme E1 and ubiquitin-conjugating enzyme E2. In the hydroxylated HIF-1␣ degradation pathway, E3 ubiquitin-protein ligase complex is formed by pVHL, elongin C, elongin B, cullin 2, and RBX1 (Fig. 2). The E3 complex is capable of functioning with E1 ubiquitin-activation and E2 ubiquitin-conjugating enzymes to mediate the ubiquitination of HIF-1␣ leading to its rapid proteosomal degradation. HIF-1␣ has a half-life of approximately 5 minutes. However, when HIF-1␣ escapes the degradation pathway, its detection by Western blot experiments occurs within minutes. Indeed, HIF-1␣ is detectable in the nucleus of cells after less than 2 minutes of exposure to anoxia or hypoxia. In this case, HIF-1␣ escapes ubiquitination and proteosomal degradation, is stabilized by posttranslational modifications, and enters into the nucleus. A comprehensive study about HIF1␣ turnover (analyzed by electrophoretic mobility shift assay; EMSA) showed that its maximal protein expression occurs in cells exposed for 1 hour in anoxia/hypoxia. These expression levels are maintained after 4 hours at low levels of oxygen exposure. However, after reoxygenation of cells, a decrease in the HIF-1 DNA binding activity is observed after 2 minutes. In addition, the nuclear HIF-1␣ protein was reduced within 4 to 8 minutes, down to a level below the detection limit within 32 minutes [4]. Under hypoxia conditions, HIF-1␣ is no longer degraded; it is phosphorylated in the cytoplasm and activates its target genes inside the nucleus. Once the target genes are expressed, a physiologic response to low oxygen concentration is achieved. Little is known about the fine-tuning of the posttranslational modifications and their role in HIF-1␣ activity. The MAPK (mitogen-activated protein kinase) and phosphatidylinositol-3-kinase (PI3K) pathways are certainly involved in the regulation of HIF-1␣. ERK1/2 (also called p44/42) activation, which is a result of activation of the upstream proteins Ras/Raf-1/MEK-1/ERK1/2, leads to HIF1␣ phosphorylation. Although MAPK phosphorylates HIF-1␣ in vitro, studies on mutations in the phosphorylation sites in HIF-1␣ protein show that its transactivation capacity is not affected or directly mediated by phosphorylation. Regulation of HIF-1␣ by MAPK is still an open issue [5]. Loss of tumor suppressor activity may also contribute to HIF-1␣ overexpression. For example, interaction of the p53 tumor suppressor gene with HIF-1␣ negatively regulates HIF-1␣ protein levels. Therefore, HIF-1␣ overexpression correlates with expression of mutant p53 in human cancer. AKT (protein kinase B), a serine/threonine kinase, is commonly activated in cancer cells. This protein functions downstream to PIP3 (phosphatidylinositol 3,4,5-triphosphate) in the PI3K phosphorylation pathway (Fig. 2). PI3K/AKT signaling pathway regulates vascular endothelial growth factor (VEGF) expression through HIF-1␣ in several types of transformed cells, including ovarian cancer and human gastric adenocarcinoma. Cancer cells have constitutive AKT


79

activity through amplification of PI3K signaling pathway or deletion of phosphatase and tensin homolog deleted on chromosome 10 (PTEN), a PI3K antagonist. AKT induction is an important event in tumorigenesis as this pathway leads to a higher level of glycolysis in a dose-dependent manner that correlates with a more aggressive malignancy in vivo. Apparently, AKT has an effect more pronounced on glucose metabolism than it effects on proliferation and survival of these transformed cells. This phenomenon was observed in established human glioblastoma cells, which showed a switch to aerobic glycolysis and glucose dependence [6]. HIF-1␣ is also modified in the lysine residue 532 localized in the ODD domain of HIF-1␣ by an acetyl transferase called arrest defective-1 (ARD1). ARD1 was originally identified in the yeast Saccharomyces cerevisiae and received this name due to the phenotype of mutant yeast defective in the mitotic cell cycle. A human variant of ARD1 is able to partially decrease HIF-1␣ accumulation in hypoxia [7]. Another HIF-1␣ posttranslational modification is S-nitrosation. Nitric oxide (NO) promotes HIF-1␣ stabilization, DNA binding, and transactivation of target genes under normoxic conditions. This NO effect on HIF-1␣ activity was verified in several cell lines, including swine tubular cells (LLC-PK1), human glioblastoma and hepatoma, as well as endothelial cells from bovine pulmonary artery. These cells, when treated with different NO donors, such as S-nitroglutathione, (Z)-1[2-(2-aminoethyl)-N-(2-ammonioethyl)amino]diazen-1-ium-1,2-diolate (NOC-18), 1-hydroxy-2-oxo-3-(3-aminopropyl)-3-isopropyl-1-triazene (NOC-5), and S-nitroso -N-acetyl-D,L-penicillamine (SNAP), accumulate HIF-1␣ protein [8]. Therefore, under hypoxia, NO disfavors HIF-1␣ accumulation, whereas under normoxic conditions, NO “mimics” hypoxia conditions and leads to HIF-1␣ accumulation.

4 HIF-1 Isoforms There are two other members of the bHLH-PAS superfamily that share structural similarity to HIF-1␣: HIF-2␣, also named endothelial PAS-domain protein 1 (EPAS1), and HIF-3␣. The roles of the 2␣ and 3␣ subunits are poorly understood. Similarly to HIF-1␣, HIF-2␣ is also regulated by hydroxylation of the conserved proline residue, which causes its degradation under normoxic conditions via ubiquitin-E3-ligase. HIF-2␣ has 48% amino acid sequence identity with HIF-1␣. HIF-2␣ and HIF-3␣ also have the common ability to heterodimerize to HIF-1␤ and bind to the hypoxia response element (HRE) motif in the regulatory regions of target genes. Additionally, an alternative splice variant of HIF-3␣ exists, a protein that lacks the transactivation domain. This splice variant was termed inhibitory PAS (IPAS) protein, and it was identified as the dominant-negative regulator of HIF-1. IPAS interacts with the amino-terminal region of HIF-1␣ and prevents its DNA binding. The effect of the different members of the HIF-1 family on gene expression can vary depending on the cell type. For example, HIF-1␣ and HIF-2␣ are strongly expressed in the kidney, vascular cells, bone marrow, macrophages, lungs,

80


endothelium, and carotid body, whereas HIF-3␣ is mainly expressed in Purkinje cells of the cerebellum and mouse cornea epithelial cells. HIF-2␣ has more restrictive distribution inside tissues than does HIF-1 [9].

5 HIF-1␣ Induction by Nonhypoxic Stimuli In addition to hypoxia, HIF-1 is also responsive to several nonhypoxic stimuli, including insulin, platelet-derived growth factor (PDGF), transforming growth factor ␤ (TGF-␤), insulin-like growth factor (IGF-1), thrombin angiotensin II, cytokines, and carbachol (activates the muscarinic acetylcholine receptors). Whereas growth factor stimulation of HIF-1␣ is cell type specific, hypoxia increases HIF-1␣ expression in all cell types. In addition, HIF-1 is activated by some metals such as cobalt, chromium, nickel, arsenic, and iron. Desferrioxamine also activates HIF-1 by mimicking hypoxic conditions. Some evidences point to the role of reactive oxygen species (ROS) as messengers that regulate HIF-1 activity [10]. The oxidative status of the iron ion bound to the PHDs might also regulate HIF-1␣ activity, as these enzymes are Fe2+ -dependent. ROS generated by mitochondria and released into the cytoplasm of cells exposed to hypoxia promote oxidation of Fe2+ to Fe3+ . A decrease in the Fe2+ availability would cause reduction in the PHD activity. This mechanism allows HIF-1 accumulation and stabilization [11].

6 HIF-1 in Cancer Once stabilized and activated, inside the nucleus, HIF-1 recognizes and binds to hypoxia response elements (HREs) present in the regulatory regions of its target genes. The consensus HRE has the sequence 5 -BRCGTGVBBB-3 , where the RCGTG core is the HIF-1 binding site (HBS). Microarray experiments suggest that more than 200 HIF-1 target genes might exist. However, not all of them are likely to be directly regulated by an HRE in their regulatory regions [10]. More than 100 HIF1 target genes are known, including the genes that encode erythropoietin, VEGF, glucose transporters, and glycolytic enzymes. Erythropoietin (EPO) and VEGF play crucial roles in adaptive responses to systemic and local hypoxia. Both are implicated in tumor progression. EPO stimulates erythropoiesis, which increases O2 delivery, whereas VEGF stimulates angiogenesis, which increases delivery of both O2 and energy substrates such as glucose. Rigorous analysis has confirmed the role of HIF-1 in EPO and VEGF transcriptional activation, and consequently in carcinogenesis. HIF-1␣ is overexpressed in many human cancers [12]. Immunohistochemical staining analysis of surgical biopsy specimens show overexpression of HIF-1␣ protein in colon, ovary, lung, prostate, skin, stomach, brain, and breast cancers. Significant associations between HIF-1␣ overexpression and patient mortality have been shown in cancers of the breast, brain, cervix, oropharynx, esophagus, ovary, and


81

uterus. Premalignancy and neomalignant lesions as well as cancer metastases of some cancer types also show increased HIF-1␣ protein. For example, HIF-1␣ is overexpressed in the majority of breast and colon cancer metastases. By contrast, association between HIF-1␣ overexpression and decreased mortality were reported for patients with head and neck cancer and non–small cell lung cancer. Thus, the effect of HIF-1␣ overexpression seems to be dependent on the cancer type. HIF-1␣ can have antiapoptotic or proapoptotic effects on tumor cells [9]. Mutation in the VHL gene results in the so called von Hippel–Lindau syndrome, characterized by the formation of tumors and fluid-filled sacs (cysts) in many different parts of the body. Clear cell renal carcinoma (CCRC) cells lacking pVHL tumor suppressor constitutively express HIF-1␣ and HIF-1 target genes under nonhypoxic conditions [13]. CCRCs present all the main characteristics supposed to be maintained by proteins encoded by genes regulated by HIF-1: increased glycolysis, decreased respiration, increased glucose uptake and lactate production. pVHL lossof-function also occurs in hemangioblastoma. Therefore, at least in these two types of tumors, deregulation of HIF-1␣ regulatory pathway plays a causal role.

7 HIF-1, Cancer, and the Glycolytic Pathway The way cells change their metabolism under hypoxia shares many similarities to the way tumors cells change their metabolism to survive and proliferate. Indeed, many of the known oncogenic signaling pathways overlap with hypoxia-induced pathways. Cancer cells have special metabolic needs for their proliferation and survival. They need to cope with different levels of oxygen and adjust their metabolism to guarantee cell proliferation and minimize apoptosis. The vast majority of human and animal tumors display a high rate of glycolysis, increased glucose uptake, increased lactate production, and decreased respiration under aerobic conditions (a phenomenon known as the Warburg effect) [14]. Aerobic glycolysis provides cancer cells (both nonmetastatic and metastastic) a high competitive capacity over normal cells. This adaptive response can be driven by hypoxia, as well as by mutations that inactivate tumor suppressor genes or activate oncogenes. Hypoxia is a potent inducer of gene expression, especially of genes involved in glycolysis for maintaining cellular energy. The higher glycolytic rate in tumor cells compared with that of nontumorigenic cells can be attained by the increased transcription, followed by translation, of glycolytic genes and glucose transporters associated or not with inhibition of oxidative phosphorylation. Hypoxia has HIF-1 as the key regulator. Evidence is surmounting that HIF-1 regulation of glycolysis is important in carcinogenesis. However, there is much controversy as to whether HIF-1– induced glycolysis is a reactive process or a “cause” of cancer. A definitive solution to this key question is complicated by the fact that hypoxia and HIF-1 are not exclusive regulators of the glycolytic enzymes, and HIF-1 activation is not associated only with hypoxia (oncogenes and growth factors can also activate this pathway). In addition, high pyruvate concentrations result in HIF-1␣ stabilization indepen-

82


dently of hypoxia [15].Therefore, although hypoxic HIF-1 activation alone does not explain aerobic glycolysis, other factors may provide feedback to enhance normoxic HIF-1 induction. The structure, regulation, and action of HIF-1 was discussed in detail above and shows a very complex network (Fig. 2). Regardless of which pathway leads to HIF-1 activation, its effect on glucose metabolism is due to the upregulation of specific target genes. Regulation of metabolism by HIF-1␣ was evidenced through investigation of its function in mouse embryonic stem (ES) cells in which the Hif1␣ gene was inactivated by homologous recombination. Analysis of Hif1a+/+ , Hif1a+/- , and Hif1a-/ES cells revealed that HIF-1␣ is required for the induction of several genes whose products are involved in glucose metabolism; more specifically, aldolase A and C, enolase 1, hexokinase I and II, phosphofructokinase L, phosphoglycerate kinase I, pyruvate kinase M, glyceraldehyde-3-phosphate dehydrogenase, lactate dehydrogenase A, GLUT1 and GLUT3 [16]. HRE has been characterized in the aldolase A, enolase 1, and lactate dehydrogenase A human promoters and also in the aldolase C mouse promoter. Complete identification of the key gene products specifying HIFdependent tumor metabolism is crucial for therapeutic targeting of HIF-1 signaling in cancer. In the presence of oxygen, glucose is converted to pyruvate by the glycolytic enzymes. Pyruvate is transported to the mitochondrial matrix, where it is converted to acetyl-CoA and metabolized in the TCA cycle. Under hypoxia, pyruvate is converted to lactate through lactate dehydrogenase (LDH) A in the cytoplasm, which is a target for HIF-1 regulation. In cancer cells, aerobic glycolysis and lactate production is favored to the detriment of oxidative phosphorylation. To understand the role of HIF-1 in the regulation of glucose and energy metabolism in cancer cells, we reviewed the literature in search of experimental data demonstrating the direct involvement of HIF-1 in the induction of such genes in these cells. In some cases, the link between HIF-1 and upregulation of a gene in cancer cells was based on the presence of HRE motifs in their regulatory regions. In the next sections, we describe the link found between HIF-1 and the proteins involved in glucose metabolism in cancer cells.

8 Hexokinase Hexokinases (HKs) catalyze the first step in the glycolytic pathway, conversion of glucose to glucose-6-phosphate (G6P). Human cells express four isozymes of hexokinases: types I, II, III, and IV (glucokinase). The key hexokinase isoenzyme involved in cancer promotion is the type II (HKII), the major isozyme overexpressed in tumors exhibiting the high glycolytic phenotype. HKII has been extensively studied by Dr. Peter Pedersen from Johns Hopkins University School of Medicine and his co-workers [17]. They found that HKII binds to transmembrane channels formed by the porin-like protein voltage-dependent anion channel (VDAC) located within the outer mitochondrial membrane. This interaction reduces the enzyme’s sensitivity


83

to product inhibition by G6P and allows the enzyme to have direct and rapid access to mitochondrially generated ATP. It also protects HKII against proteolytic degradation. These combined properties, together with the high content of the enzyme in highly malignant tumors (>100-fold increase), result in the rapid production of G6P. This key metabolic intermediate-precursor serves as a major carbon source for most biosynthetic pathways that are essential for the growth and rapid proliferation of tumors. It also serves as the initial substrate for glycolysis and ATP generation. Enhanced transcription of HKII occurs in malignant tumors, implicating the HKII promoter activation and upregulation during tumorigenesis. Functional HREs (HIF-1 responsive) are located in the distal region of the HKII promoter. HKII can also be transcriptionally activated by mutant p53 or through demethylation of its promoter. Regulation of the gene encoding hepatoma HKII occurs through activation of its promoter that contains in its distal region two overlapping E-box/HIF-1 sequences separated by approximately 50 base pairs. These sequences are known to be associated in other gene promoters with responsiveness to glucose (E-box) or hypoxic conditions (HIF-1). The response of the HKII gene promoter to both glucose and hypoxic conditions is not confined to its distal region but involves significantly the proximal region as well. Elements in this region remain to be identified and studied in detail.

9 Glucose Phosphate Isomerase Glucose phosphate isomerase (GPI) is a housekeeping cytosolic enzyme of glucose metabolism that plays a key role in both glycolysis and gluconeogenesis pathways, catalyzing the reversible isomerization of glucose-6-phosphate and fructose6-phosphate. Upon secretion, GPI also acts as a cytokine also referred to as autocrine mobility factor (AMF) or NLK [18]. AMF was originally isolated as a cytokine autocrine factor that stimulates cell mobility and NLK as a lymphokine and neurotrophic factor for spinal and sensory neurons. GPI/AMF/NLK is associated with the development of tumors, and an increased amount of this protein is present in the urine and serum of patients with various malignancies. Therefore, in addition to its role in the glycolytic pathway, the induction of this gene under hypoxia may play a role in stimulating cell mobility, invasion, and metastasis in tumor cells. In human pancreatic cancer PC-3 cells, GPI mRNA is weakly expressed in cells cultured in normoxic conditions, but its synthesis is markedly increased in hypoxic conditions (1% oxygen plus CoCl2 ). YC-1, 3-[5 -hydroxymethyl-2 furyl]-1-benzyl indazole, a guanylate cyclase activator, reduces the mRNA levels of GPI in PC-3 cells in a dose-dependent manner under hypoxic conditions. Guanylate cyclase activation can increase production of NO, which inhibits the accumulation and activity of HIF-1 in hypoxia. Therefore, GPI transcriptional activation is intimately linked to HIF-1 activity.

84


10 Phosphofructokinase Phosphofructokinase 1 (PFK-1) catalyses the rate-limiting phosphorylation of fructose-6-phosphate to fructose-1,6-biphosphate, which is an energy-consuming step in glycolysis. The fructose-2,6-biphosphate is the most potent allosteric activator of glycolysis and exerts control over the rate of glucose utilization. It activates PFK-1 and inhibits the gluconeogenic enzyme fructose-1,6-biphosphatase. Because of the antagonistic effects in these enzymes, the product of the reaction catalyzed by 6-phosphofructo-2-kinase/fructose-2,6-biphosphatase (PFKBF) plays an essential role in the opposing glycolytic and gluconeogenic pathways. In human cells, there are at least four different genes encoding PFKFBs (PFKBP-1, PFKBP-2, PFKBP-3, and PFKBP-4). They all respond to hypoxia in vivo, and this response is diverse in different organs. Tissue-specific isoforms of PFKBP are not completely exclusive, and several tissues express more than one isoform. Among the PFKFB isoforms of mammalian origin, PFKFB3 has the highest kinase:phosphatase activity (K/B) and is the most highly expressed isozyme in transformed cells. The PFKFB-4 isozyme lacks a serine phosphorylation residue that is critical for the downregulation of its kinase activity. Therefore, one can conclude that it may have a high K/B ratio, and it should greatly promote glycolysis under conditions of limited oxygen supply. The use of dimethyloxalylglycine (a specific competitive inhibitor of PHDs) demonstrated the role of HIF-1␣ in hypoxic induction of the genes coding for PFKFB3 and PFKFB-4 isoforms (in human hepatoma Hep-3B and HepG2 cell lines, human prostate cancer cell, PC-3 and HeLa cells). A splice variant of PFKFB-4 (PFKFB-4 s) is also induced by hypoxia in DB-1 melanoma, besides the full-length PFKFB-4. Hypoxic induction of the transcription of these genes is mediated by HREs located in their promoter regions. In addition, PFKFB3 and PFKFB4 isoforms are overexpressed in solid malignant tumors from the breast and colon, overexpression of the PFKFB-4 protein being higher compared with that of PFKBF-3 [19].

11 Aldolase Aldolase (ALD) catalyzes the reversible aldol cleavage of fructose-6-phosphate into two trioses, glyceraldehyde-3-phosphate and dihydroxyacetone. It has a role in the glycolytic, gluconeogenic, and fructose metabolic pathways. Three tissue-specific forms exist in higher vertebrates: aldolase A (predominately in muscle and red blood cells), aldolase B (predominately in liver, kidney, and intestine), and aldolase C (predominately in brain). The direct association of aldolase gene expression and HIF-1 in tumors was observed in a study with a mouse hepatoma cell line, Hepa1c1c7, and its mutant derivative, ARNT-deficient c4T. The absence of the ARNT protein completely abolished the oxygen regulation of the aldolase A and phosphoglycerate kinase 1 (PGK1) gene expression, indicating that these genes share a common pathway of response to hypoxia involving activation of HIF-1. Among the three


85

isoenzymes, aldolases A and C seem to have evolved to perform the glycolytic reaction, fructose-1,6-phosphate cleavage, more efficiently than does aldolase B. Aldolase B seems to be more involved with the gluconeogenic pathway [20].

12 Glyceraldehyde Phosphate Dehydrogenase Glyceraldehyde phosphate dehydrogenase (GAPDH) is a key enzyme in glycolysis and also a multifunctional protein. It converts glyceraldehyde-3-phosphate into 1,3bisphosphoglycerate. Its gene is constitutively expressed in many tissues. Variations in GAPDH expression levels occurs in tissues at different developmental stages and have been observed in cells treated with a variety of agents such as mitogens, dexamethasone, calcium, insulin, and in cells under hypoxia conditions. Overexpression of GAPDH in tumor cells as a consequence of the development of a hypoxic cellular microenvironment is cell type specific. It has been observed in human prostate adenocarcinoma cells (cell line LNCap), human spontaneous cervical cancer cells (SiHA), but not in human malignant glioma cells (cell lines U373-MG, GaMG, and U251) ([21], [22]). It is not clear if the reason for the upregulation of GAPDH in tumor cells is only for increasing glycolysis. The role of GAPDH as a NADH generator and as a mediator of cell death has been highlighted in several studies, which adds to the complexity of its upregulation in tumor cells.

13 Phosphoglycerate Kinase Phosphoglycerate kinase (PGK) catalyzes the transfer of the phosphoryl group from the acyl phosphate of 1,3-bisphosphoglycerate to ADP. ATP and 3phosphoglycerate are the products. Human PGK1 gene harbors HREs in the 5 flanking and 5 UT regions, and mouse PGK1 harbors three HREs in the 5 flanking sequence [10]. The HREs of the PGK promoter mediate the transcriptional response to hypoxia. Indeed, the induction of PGK in different solid tumor cell lines in hypoxic conditions is under the control of HIF-1␣. Expression of PGK1 increases in mouse hepatoma Hepa1c1c7 cell lines exposed to hypoxia or to treatment with desferrioxamine [20].

14 Enolase Enolase (ENO) is also upregulated in many cancers, including breast, liver, colon, and lung. Direct evidence of HIF-1 upregulation of the enolase gene was obtained using YC-1, a drug that targets HIF-1. Levels of aldolase and enolase mRNAs decrease in YC-1–treated Hep3B hepatoma cells. The inhibitory action of YC-1 on these glycolytic genes probably inhibits cell survival under hypoxia and may promote cell death in hypoxic areas [23].

86


15 Pyruvate Kinase Pyruvate kinase (PK) is responsible for net ATP production within the glycolytic pathway. It is consistently altered during tumorigenesis. Four isoenzymes of pyruvate kinase have been identified: type L (present in tissues with gluconeogenesis such as liver and kidney); type R (expressed in erythrocytes); type M1 (muscle and brain); and type M2 (expressed in lung tissues and all cells with high rates of nucleic acid synthesis, such as embryonic cells, adult stem cells, and especially tumor cells). Investigation of the mRNA levels of HIF-1␣, PGK1 and muscle-type pyruvate kinase 2 (PKM2) were performed in various hepatoma cell lines as well as in mouse and human tumor specimens and the corresponding normal tissues. In five of eight human colorectal cancers investigated, PGK1 and PKM2 mRNA levels were increased in comparison with the corresponding normal tissues, whereas HIF1-␣ was not significantly changed [24]. Though probable, these studies do not conclusively demonstrate that the increase in PKM2 gene expression is, in fact, mediated by HIF-1–induced transcriptional activation. However, HIF-1␣ plays a critical role in PK expression during hypoxic challenge in glial cells.

16 Pyruvate Dehydrogenase The pyruvate dehydrogenase (PDH) complex catalyzes the conversion of pyruvate to acetyl-coenzyme A, which enters the TCA cycle, producing NADH and carbon dioxide. The PDH complex is composed of three subunits, E1 (pyruvate decarboxylase), E2 (dihydrolipoamide acetyltransferase), and E3 (dihydrolipoamide dehydrogenase), and is under the control of pyruvate dehydrogenase kinases (PDKs) 1 to 4. PDKs phosphorylate the E1 subunit of PDH causing its inhibition. The activity of PDK is regulated by the concentration of the metabolic products of pyruvate (NADH and acetyl-coA). HIF-1 upregulates the expression of PDK1 [25]. Upregulation of the expression of PDK1 by HIF-1 illustrates how it favors a metabolic switch toward glycolysis to the detriment of oxidative phosphorylation. Hypoxic cells compensate the inefficient glycolytic energy output by increasing glucose uptake, overexpressing and increasing the activities of the glycolytic enzymes.

17 Lactate Dehydrogenase and Lactate Carrier MCT4 Lactate dehydrogenase (LDH) converts pyruvate into lactate. LDH-A converts pyruvate to lactate during glycolysis under anaerobic conditions in normal cells. The other isoenzyme, LDH-B, kinetically favors the conversion of lactate to pyruvate and is found at high levels in aerobic tissues such as heart. Elevated LDH-A expression contributes to the ability of tumor cells to undergo aerobic glycolysis. Once produced, lactic acid exits cancer cells through the MCT4 lactate carrier that is also


87

induced in these cells. Under hypoxia, both pyruvate transformation into lactate with concomitant regeneration of NAD+ and lactate exit from the cell are stimulated because the genes encoding the LDH-A isoform and the lactate carrier MCT4 are HIF-1 targets. The c-myc gene is frequently activated in human cancers. This transcription factor forms a heterodimer with another protein (MAX) and binds to a 5 -CACGTG-3 consensus sequence in its target genes inducing gene expression. In addition to HIF-1, the LDH-A gene is also a c-MYC target. Therefore, the molecular basis for altered tumor glycolysis and lactate production cannot be solely explained by HIF-1 activation of specific genes [26]. In addition to its role in lactate production, LDH-A may also be involved in the regulation of gene expression and/or DNA replication. Therefore, its modulation by HIF-1 and c-MYC could be linked to effects other than to increased lactate production.

18 Glucose Transporters and HIF-1 Glucose enters animal cells through facilitated diffusion via glucose transporters (GLUT1 to GLUT5) and is subsequently phosphorylated by hexokinase. GLUT1 and GLUT3, present in nearly all mammalian cells, are responsible for basal glucose uptake and continually transport glucose into cells at an essentially constant rate. They have a high affinity for glucose, allowing transport of glucose at a high rate under normal physiologic conditions. Cancer cells overexpress both GLUT1 and GLUT3 to increase glucose uptake. Upregulation of glucose transporters has been observed during carcinogenesis in esophageal, gastric, breast, and colon cancers. A major technique used to measure glucose intake in tumor cells is positron emission tomography (PET) with [18 F]fluoro-2-deoxy-D-glucose (FDG). PET is a noninvasive imaging technique used in the clinic to detect malignant tumors, which measures the glucose metabolism in vivo. FDG is a fluorinated glucose analogue and accumulates in tumor cells in proportion to the rate of glucose metabolism. A high uptake of FDG is associated with a higher GLUT1 expression. The enhanced uptake of glucose often reflects tumor aggressiveness in patients with different types of tumors. The glucose transfer from the bloodstream to the inside of cells mediated by GLUT1 plays a central role in the development and malignant behavior of cancer cells. HIF-1 binding to the promoter of glucose transporters was seen in human HepG2 cells (GLUT1 and GLUT3) and mouse hepatoma cells exposed to hypoxia (GLUT1). Studies of GLUT mRNA expression in hypoxic cells have shown that the level of GLUT1 mRNA in human placenta choriocarcinoma cell line BeWo (JCRB911) increases 1.4-fold in the presence of 5% O2 and by 2.2-fold in the presence of 250 ␮M CoCl2 compared with that under 20% O2 . In the rat choriocarcinoma Rcho-1 cell line, the level of GLUT1 mRNA increases 1.8-fold in the presence of 5% O2 and by 3.2-fold in the presence of 250 μ M CoCl2 compared with that under 20% O2 . The GLUT1 protein levels also increase under

88


hypoxia and CoCl2 treatment, and these results suggest that GLUT mRNA and protein levels were directly mediated through HIF-1 induction [27].

19 HIF-1␣ Upregulation by TCA Intermediates Is Linked to Carcinogenesis Succinate dehydrogenase (SDH) is a TCA enzyme, which is also part of the mitochondrial respiratory chain. Succinate dehydrogenase deficiency causes either severe encephalomyopathy or tumor formation [28]. Germ-line heterozygote mutations in the SDH subunits cause inherited pheochromocytomas (adrenal gland tumors) and paragangliomas. The effects of SDH deficiency are postulated to be associated with the disturbance of two signaling pathways: one involving the superoxide radical and another involving HIF-1. SDH defect can increase superoxide radical production by the mitochondrial respiratory chain, which is able to trigger both cell death and proliferation. On the other hand, when succinate accumulates inside the mitochondria, it is transported into the cytosol where it can inhibit prolyl hydroxylases (PHD1 to PHD3), the enzymes responsible for HIF-1 destabilization. Upon PHD inhibition, HIF-1␣ is stabilized under normoxic conditions. Heterodimerization of stabilized HIF-1␣ and HIF-1␤ forms active HIF-1, which induces transcription of nuclear genes involved in tumor progression. Mutations in another TCA enzyme, fumarate hydratase (FH), cause cutaneous and uterine leiomyomas, as well as renal cell carcinomas. Mitochondrial proteins may also play a role in the development of sporadic kidney tumor oncocytoma. High concentrations of cytosolic fumarate may inhibit prolyl hydroxylases and have the same consequences as excess succinate on HIF-1 destabilization. Defective SDH and FH activity lead, respectively, to the accumulation of succinate and fumarate. These TCA intermediates, in addition to oxaloacetate, are capable of inhibiting the PHD activity and subsequently inhibiting HIF-1␣ degradation [29].

20 HIF-1 Activity as a Therapeutic Target The prognostic significance of HIF-1␣ expression in cancer cells, its central role in the adaptive response to hypoxia, and its participation in the Warburg effect make it a potential target for anticancer drug development. It is a reasonable speculation that inhibition of HIF-1 activity may be a useful cancer therapy. A number of anticancer agents that may target HIF-1 activity have been described (some cited in this text) and are treated more fully by Patiar and Harris [30]. The challenge remains to prove that the antitumorigenic effect of these drugs is due to HIF-1␣ inhibition. Selectivity may be resolved by the fact that tumor cells are more hypoxic than are normal cells and have specific intrinsic markers of HIF-1␣ activation. Further increase in this difference is possible through the use of angiogenesis inhibitors (e.g., thrombospondin-1) in association with HIF-1␣ inhibition.


89

21 Conclusion HIF-1␣ can be activated by hypoxic and nonhypoxic stimuli, including growth factors, oncogenes, metals, and ROS. Metabolism of glucose and oxidative phosphorylation are directly affected by the HIF-1 signaling pathway. A change in ATP production pathway from oxidative phosphorylation to aerobic glycolysis is a hallmark of cancer cells. There is an evident overlap between HIF-1 function and cancer cell metabolism. Cancer cells have special needs for rapid proliferation and survival. Some of these needs are fulfilled by the HIF-1 signaling cascade. In addition, HIF-1 is located at the center of the major pathways that define cell destiny toward adaptation to fast growth or apoptosis. The available data show that cancer cell survival is a much more pronounced HIF-1 effect than is cell death. Therefore, HIF-1 will continue to be explored as a potential target for cancer therapy.

References 1. Semenza GL. Expression of hypoxia-inducible factor 1: mechanisms and consequences. Biochem Pharmacol 2000;59: 47–53. 2. Metzen E, Berchner-Pfannschmidt U, Stengel P, et al. Intracellular localisation of human HIF-1␣ hydroxylases: implications for oxygen sensing. J Cell Sci 2003; 116:1319–1326. 3. Berra E, Benizri E, Ginouvès A, Volmat V, Roux D, Pouysségur J. HIF prolyl-hydroxylase 2 is the key oxygen sensor setting low steady-state levels of HIF-1␣ in normoxia. EMBO J 2003; 22:4082–4090. 4. Jewell UR, Kvietikova I, Scheid A, Bauer C, Wenger RH, Gassmann M. Induction of HIF–1␣ in response to hypoxia is instantaneous. FASEB J 2001; 15:1312–1314. 5. Höpfl G, Ogunshola O, Gassmann M. HIFs and tumors – causes and consequences. Am J Physiol Regul Integr Comp Physiol 2004; 286:R608–R623. 6. Elstrom RL, Bauer DE, Buzzai M, et al. AKT stimulates aerobic glycolysis in cancer cells. Cancer Res 2004; 64:3892–3899. 7. Kim SH, Park JA, Kim JH, et al. Characterization of ARD1 variants in mammalian cells. Biochem Biophys Res Commun 2006; 340:422–427. 8. Brüne B, Zhou J. The role of nitric oxide (NO) is stability regulation of hypoxia inducible factor-1␣ (HIF-1␣). Current Med Chem 2003; 10:845–855. 9. Lee J-W, Bae S-H, Jeong J-W, Kim S-H, Kim K-W. Hypoxia-inducible factor (HIF-1) ␣: its protein stabilization and biological functions. Exp Mol Med 2004; 36:1–12. 10. Wenger RH, Stiehl DP, Camenisch G. Integration of oxygen signalling at the consensus HRE. Sci Stke 2005;re12. 11. Pouyssegur J, Mechta-Grigoriou F. Redox regulation of the hypoxia-inducible factor. Biol Chem 2006; 387:1337–1346. 12. Zhong H, De Marzo AM, Laughner E, et al. Overexpression of hypoxia-inducible factor 1alpha in common human cancers and their metastases. Cancer Res 1999; 59:5830–5835. 13. Semenza GL. HIF-1 mediates the Warburg effect in clear cell renal carcinoma. J Bioenerg Biomembr 2007; 39:231–234. 14. Warburg O. On the origin of cancer cells. Science 1956; 123:309–314. 15. Lu H, Forbes RA, Verma A. Hypoxia-inducible factor 1 activation by aerobic glycolysis implicates the Warburg effect in carcinogenesis. J Biol Chem 2002; 277:23111–23115. 16. Iyer NV, Kotch LE, Agani F, et al. Cellular and developmental control of O2 homeostasis by hypoxia-inducible factor 1␣. Genes Dev 1998; 12:149–162.

90


17. Mathupala SP, Ko YH, Pedersen PL. Hexokinase II: cancer’s double-edged sword acting as both facilitator and gatekeeper of malignancy when bound to mitochondria. Oncogene 2006; 25:4777–4786. 18. Funasaka T, Yanagawa T, Hogan V, Raz A. Regulation of phosphoglucose isomerase/autocrine motility factor expression by hypoxia. FASEB J 2005; 19:1422–1430. 19. Minchenko OH, Ochiai A, Opentanova IL, et al. Overexpression of 6-phosphofructo2-kinase/fructose-2,6-bisphosphatase-4 in the human breast and colon malignant tumors. Biochimie 2005; 87:1005–1010. 20. Salceda S, Beck I, Caro J. Absolute requirement of aryl hydrocarbon receptor nuclear translocator protein for gene activation by hypoxia. Arch Biochem Biophys 1996; 334:389–394. 21. Lu S, Gu X, Hoestje S, Epner DE. Identification of an additional hypoxia responsive element in the glyceraldehyde-3-phosphate dehydrogenase gene promoter. Biochim Biophys Acta 2002; 1574:152–156. 22. Said HM, Hagemann C, Stojic J, et al. GAPDH is not regulated in human glioblastoma under hypoxic conditions. BMC Mol Biol 2007; 8:55. 23. Yeo EJ, Chun YS, Cho YS, et al. YC-1: a potential anticancer drug targeting hypoxiainducible factor 1. J Natl Cancer Inst 2003; 95:516–525. 24. Kress S, Stein A, Maurer P, et al. Expression of hypoxia-inducible genes in tumor cells. J Cancer Res Clin Oncol 1998; 124:315–320. 25. Kim JW, Tchernyshyov I, Semenza GL, Dang CV. HIF-1-mediated expression of pyruvate dehydrogenase kinase: a metabolic switch required for cellular adaptation to hypoxia. Cell Metab 2006; 3:177–185. 26. Shim H, Dolde C, Lewis BC, et al. c-Myc transactivation of LDH-A: implications for tumor metabolism and growth. Proc Natl Acad Sci USA 1997; 94:6658–6663. 27. Hayashi M, Sakata M, Takeda T, et al. Induction of glucose transporter 1 expression through hypoxia-inducible factor 1 under hypoxic conditions in trophoblast-derived cells. J Endocrinol 2004; 183:145–154. 28. Selak MA, Armour SM, MacKenzie ED, et al. Succinate links TCA cycle dysfunction to oncogenesis by inhibiting HIF-alpha prolyl hydroxylase. Cancer Cell 2005; 7:77–85. 29. Pollard PJ, Briere JJ, Alam NA, et al. Accumulation of Krebs cycle intermediates and overexpression of HIF1alpha in tumours which result from germline FH and SDH mutations. Hum Mol Genet 2005; 14:2231–2239. 30. Patiar S, Harris AL. Role of hypoxia-inducible factor-1alpha as a cancer therapy target. Endocr Relat Cancer 2006; 13(Suppl 1):S61–S75.

The Role of Glycolysis in Cellular Immortalization Hiroshi Kondoh

Abstract The clinical significance of the Warburg effect is well established, although it is not clear why and when cancer cells start to display a high glycolytic rate in vivo. The discovery of hypoxia-inducible transcriptional factor as a critical regulatory factor for glycolytic genes along with the recent advances in senescence biology imply that the metabolic shift to enhanced glycolysis would be observed in multiple stages during tumorigenesis in vivo. Increased glycolysis would be required in the early step of immortalization to avoid oxidative stress–mediated senescence, followed by the adaptation to hypoxic conditions through increased enhancement of glycolysis during transformation. Discovery of regulatory mechanisms for such a metabolic shift could provide important information for the future development of cancer diagnosis and anticancer therapy. Keywords Phosphoglycerate mutase · Glycolysis · Immortalization · p53 · Warburg effect

1 In Which Stage of Carcinogenesis Is the Warburg Effect Required? 1.1 Immortalization and Transformation During Multistep Carcinogenesis It is well-known that tumorigenesis is a multistep process that involves a series of genetic and epigenetic alterations [1]. Chemically induced skin tumor is the most established animal model of multistep tumorigenesis. The sequential application of carcinogen can effectively induce tumors in mice. The application of 7, H. Kondoh (B) Department of Geriatric Medicine, Graduate School of Medicine, Kyoto University, 54 Kawahara-cho, Shogoin, Sakyo-ku, Kyoto, 606-8507, Japan e-mail: [email protected] S.P. Apte, R. Sarangarajan (eds.), Cellular Respiration and Carcinogenesis, DOI 10.1007/978-1-59745-435-3 7, C Humana Press, a part of Springer Science+Business Media, LLC 2009

91

92

H. Kondoh

12-dimethyl-benzanthracene (DMBA), followed by treatment with phorbol esters (e.g., TPA), can result in skin tumors. In these experimental models, each step is called initiation, promotion, and progression, respectively, supporting the notion that cancer progression includes multistep events. From the pathologic and clinical viewpoint, the process of tumorigenesis in vivo can be divided into three or more stages: immortalization, transformation, and metastasis [2]. Among other factors, the acquisition of unrestricted proliferative potential, called cellular immortalization, would be involved in the very early stage in vivo. During immortalization, several biological events are supposed to be required; bypassing senescence, evasion of apoptosis and antigrowth signals, growth factor independence, and so on. Then, to grow more aggressively in vivo, these immortalized cells should be transformed to be anchorage-independent, resistant to contact inhibition, proangiogenic, and so forth. In the end stage, they easily detach from each other, more easily attach to and degrade matrix components, and invade and migrate to other tissues (metastasis). All these properties are also distinct hallmarks of cancerous cells compared with their normal counterpart [3]. Another candidate for inclusion in this list would be enhanced glycolysis, noted by Otto Warburg more than seven decades ago. He first reported that cancerous tissues or cells display increased glycolysis by an unknown mechanism. A high glycolytic rate, even under high oxygen conditions, is referred to as the Warburg effect. This property is well used in clinical practice for the detection of metastatic tumor mass by positron-emission scanning of [18 F]fluoro-2-deoxy-Dglucose. Thus there is no dispute about its clinical significance, whereas there has been a controversy as to exactly when cancer cells become highly glycolytic in vivo. In other words, one big emerging question is the following: In which stage of multistep oncogenesis would the enhancement of glycolysis be involved?

1.2 Enhanced Glycolysis Is Required During Transformation to Adapt to Hypoxic Conditions It was widely assumed that cancer cells maintain upregulated glycolytic metabolism to adapt to the hypoxic conditions in vivo, as solid aggressive tumors outgrow the blood supply of the feeding vasculatures [4]. Alternatively, it has been proposed that the increased glucose flux might improve efficiency of glucose utilization in a microenvironment in which glucose is limited. In such a context, the glycolytic response represents a successful metabolic adaptation of cancer cells in vivo. The concomitant induction of angiogenesis and enhancement of glycolysis with cell proliferation is mediated partly by activating the hypoxia-inducible transcriptional factor (HIF-1) [5]. Hypoxia increases HIF-1␣ levels in most cell types, and HIF-1 mediates adaptive responses to changes in tissue oxygenation. Thus, HIF-1 can directly upregulate expression of a set of genes involved in both local and global responses to hypoxia, including angiogenesis, erythropoiesis, respiration, and most of the glycolytic enzymes: Hexokinase 1 (HK1), Hexokinase 2 (HK2), Autocrine

The Warburg Effect and Immortalization

93

Motility Factor/Glucose-6-Phosphate Isomerase (AMF/GPI), Enolase 1 (ENO1), glucose transporter 1 (GLUT1), glyceraldehyde-3-P dehydrogenase (GAPDH), lactate dehydrogenase (LDH), 6-phosphofructo-2-kinase/fructose-2,6-bisphosphatase 3 (PFKBF3), phosphofructokinase liver form (PFKL), phosphoglycerate kinase 1 (PGK1), pyruvate kinase muscle form (PKM), triose phosphate isomerase (TPI). Interestingly, both HIF-1 and glycolytic enzymes are overexpressed in many tumors and cancer cells. Moreover, ectopic expression of glycolytic enzyme LDH can transform culture cells, whereas the inactivation of LDH ablates the transformation ability in cancer cells [6]. Altogether, these data support a functional link between enhanced glycolysis and cellular adaptation during tumor formation and expansion. In conclusion, the transformation during multistep oncogenesis would require the enhancement of glycolysis (Fig. 1).

1.3 Oncogene and Glycolysis: ras and c-Myc However, the Warburg effect cannot be explained solely by cellular adaptation to hypoxic conditions for the following reasons. 1. The cellular adaptation model does not explain the constitutive metabolic change that maintains high glycolytic rates in cultured cancer cells even under 20% oxygen in vitro [7].

A

B

c-Myc p53 KO

Immortalization Immortalization

Glycolysis

Hypoxia

Transformation

HIF-1

Hypoxia

Transformation

Glycolysis

Metastasis

HIF-1 Glycolysis

Metastasis

Fig. 1 Two models of glycolysis involvement during multistep tumorigenesis.(A) Enhanced glycolysis during transformation. In the growing solid tumors, the core of tumors would suffer hypoxia as they outgrow the feeding capacity of the neovasculature. To adapt to hypoxic conditions, HIF-1 is activated; which in turn enhances glycolysis. (B) Enhanced glycolysis during immortalization and transformation. Other than the model in (A), glycolysis would be involved also in the earlier step of immortalization. Immortalizing genes, induction of c-Myc, or knock-down of p53 (p53KO) would enhance glycolysis in cancerous cells

94

H. Kondoh

2. Ectopic expression of HIF-1 in culture cells induces cell cycle arrest, which argues against the proposition that glycolysis in cancer cells is regulated only by HIF-1 or hypoxia. 3. Indeed, emerging evidence indicates that glycolysis is regulated in a cellular context–dependent manner and by multiple genetic factors: oncogenes (ras, c-Myc), tumor suppressor genes (p53), signaling kinases (AMP kinase, Akt kinase, Pak1 kinase), and so forth. These facts suggest that there would be possible other mechanisms to increase glycolysis during multistep oncogenesis other than adaptation to hypoxia. Interestingly, several groups reported that c-Myc plays a critical role in cellular immortalization of human primary epithelial cells and fibroblasts via telomerase activation [8], and c-Myc could also enhance glycolysis through transcriptional upregulation of several glycolytic enzymes: HK, PFK, TPI, GAPDH, ENO, and LDH [9]. Taken together, it is possible that glycolysis could be enhanced in the earlier step of tumorigenesis before cellular transformation occurs.

2 Enhanced Glycolysis Is Also Required in Immortalization 2.1 Senescence Is a Barrier for Tumorigenesis Historically, senescence and aging research came into their own around the 1960 s. Hayflick and Moorhead reported replicative senescence in tissue cultured primary cells in 1967 [10], and Harman proposed a radical theory of aging in 1956 [11]. Recently, there has been a resurgence in senescence biology, as cancer cells are known to be immortal both in vivo and in vitro. Most somatic cells have a limited replicative capacity under standard tissue culture conditions and suffer a permanent cell cycle arrest, called replicative senescence [12]. Replicative senescence is induced by telomere erosion upon reaching replicative exhaustion, which can be bypassed by the ectopic expression of telomerase in human fibroblasts. It is well established that senescence can also be induced in a telomere-independent manner, called stress-induced senescence (SIS) [13]. Cells suffering either replicative or stress-induced senescence (SIS) are phenotypically similar: they adopt an enlarged and flattened appearance; deposit increased amounts of extracellular matrix; express elevated levels of inhibitor of plasminogen activator type 1 (PAI-1); develop lipofuscin granules, single prominent nuclei, senescenceassociated heterochromatin, senescence-associated ␤-galactosidase activity; and show negligible DNA synthesis (Fig. 2). Interestingly, although mouse telomeres are much longer than their human counterparts (60 kb vs. 12 kb), mouse embryonic fibroblasts (MEFs) stop growing within a shorter time when compared with human primary fibroblasts [14]. Apart from telomere structure, there are several other factors influencing cellular life span, as some cell types reach a proliferative barrier long before telomere erosion becomes critical. It is now recognized that cells can arrest virtually at any age, in


95 Premature senescence

Oxidative, or Culture stress

glycolysis declined Replicative senescence Telomeric erosion

Immortal

Primary cells

Radical Scavanger

reduced ROS

Enhanced glycolysis

Fig. 2 Glycolysis and cellular senescence. Senescence can be induced in telomere-dependent or -independent manner. In both cases, glycolysis declines. Enhanced glycolysis can immortalize primary cells with significant reduction of oxidative damage

a telomere-independent manner. For example, oncogenic stimuli can induce premature senescence, called oncogene-induced senescence. Thus, cellular senescence also constitutes a potent anticancer mechanism [15].

2.2 Oxidative Stress and Senescence Recent studies suggest that oxidative damage and the accumulation of reactive oxygen species (ROS) are closely involved in senescence. ROS, such as superoxide anion and hydroxyl radical, are produced during cellular metabolism mainly in the mitochondria. ROS are also produced in response to different environmental stimuli such as UV, IR, chemicals, hyperoxia, or hydrogen peroxide treatment. Abnormal ROS accumulation and its effects on intracellular macromolecules (oxidation of lipid, protein, and DNA) induces cumulative damage at the cell, tissue, and organism level. Mild oxidative stress (e.g., treatment with low concentrations of hydrogen peroxide) is enough to induce senescence in primary cells. Interestingly, premature senescence induced by culture stress or oncogene-induced stress is associated with oxidative damage in cells [16]. Increased ROS accumulation is also observed during replicative senescence. The replicative potential of both murine and human fibroblasts is significantly enhanced under low oxygen, associated with less oxidative damage inflicted than under

96

H. Kondoh

normoxia (O2 20%) [17]. Immortalized cells suffer less oxidative damage than do primary fibroblasts when cultured under 20% O2 . Moreover, immortalized cells are more resistant to the deleterious effects of hydrogen peroxide than are primary cells. Thus, ability to resist oxidative stress could be a clue for explaining the immortality of cancer cells. Several radical scavengers can protect cells against oxidative stress. The superoxide dismutase enzyme (SOD) converts superoxide anions into hydrogen peroxide, and hydrogen peroxide can be detoxified by catalase. Consequently, these antioxidant enzymes can impact the proliferation of both primary and immortal cells as they can counteract ROS effects. The ability of SOD to bypass senescence has been well studied and established in various cells or organisms. Increased expression of SOD can extend the life span of primary fibroblasts [18]. Conversely, knockdown of SOD using small interfering RNA (siRNA) induces premature senescence accompanied by p53 activation. Transgenic fly overexpressing SOD [19] or the detoxifying enzyme catalase [20] presents an extended organism life span. Although it is clearly established that these antioxidant scavengers are essential for proliferation of immortal cells, little is known to date on the specific regulation operating in cancer cells.

2.3 Senescence-Bypassing Effect of Enhanced Glycolysis We recently found that glycolytic enzymes can modulate the cellular life span of MEFs [21]. In a senescence-bypassing screening in MEFs using a retroviral cDNA library, we isolated the glycolytic enzyme phosphoglycerate mutase (PGM). PGM converts 3-phosphoglycerate to 2-phosphoglycerate during glycolysis. Analysis of the impact of other glycolytic enzymes on senescence in MEFs showed that glucophosphate isomerase (GPI) could also drive immortalization of MEFs. Ectopic expression of PGM or GPI increases glycolytic flux, decreases the oxidative damage that MEFs are exposed to, and extends the life span of primary MEFs. Conversely, knockdown of PGM or GPI via specific siRNA induces premature senescence. Moreover, we and others found that the glycolytic flux declines during senescence both in murine and human fibroblasts [22]. How can an increase in glycolysis immortalize primary cells? From the data presented above, it appears that enhanced glycolysis can protect cells from oxidative stress and as a consequence avoid senescence [21]. MEFs immortalized by PGM or GPI suffer less oxidative damage than do control cells as estimated by cytosolic ROS staining, or quantification of 8-hydroxydeoxyguanosine (8-OHdG), a hallmark of oxidative DNA damage lesions. Moreover, mouse embryonic stem (ES) cells also present a surprisingly high glycolytic rate [23]. Thus similarly to cancer cells, ES cells display the Warburg effect. ES cells are an exception that can bypass culture stress–induced senescence among other primary cells. We hypothesized that reversible modifications such as a specific factor playing a role in ES cell immortality enhances the glycolytic rate resulting in an increased protection from oxidative stress. This metabolic protection might contribute to preserve the genome integrity of ES cells thereby avoiding genetic


97

alterations and allowing them to maintain their self-renewal capacity. Also, these metabolic rates can be reversed, which would explain why differentiated cells do not present this enhanced metabolic level.

2.4 Glycolysis as Radical Scavenger How can enhanced glycolysis protect cells from oxidative damage? One clue is the common metabolic feature shared between several immortalized cells: cancer cells, immortalized primary cells, and ES cells [22]. All these cells display enhanced glycolysis with decreased mitochondrial respiration. Thus, increased glycolysis during immortalization could be accompanied by the downregulation of mitochondrial function by unknown mechanism(s) and would result in less intrinsic ROS production, which will be discussed later. A second possible mechanism is that increased glycolysis could imitate or activate radical scavengers. Interestingly, the antioxidant function of some scavengers (such as reduced gluthation (GSH) or Thioredoxin (TRX)) is closely coupled to the NADPH/NADP balance. Most of the NADPH/NADP is produced through the pentose phosphate pathway (PPP), a branching metabolic pathway derived from the glycolytic pathway. Enhanced glycolysis might activate PPP and increase the level of NADPH as by-products (Fig. 2). Although this remains a hypothesis, several reports support it. The impact of glucose-6-phosphate dehydrogenase (G6PD) activity on cell proliferation is well established [24]. G6PD catalyzes the rate-limiting step in the pentose phosphate PPP, which is responsible for the recycling of NADPH and maintenance of the redox balance as described above. G6PD-deficient human fibroblasts have a reduced life span that is attributable to oxidative stress and can be corrected by the ectopic expression of this enzyme [25]. Both G6PD activity and the NADPH pool decline during continued culture passage, presumably as a consequence of the accumulation of oxidative damage. Importantly, ES cells ablated from G6PD expression are extremely sensitive to oxidative damage, showing massive apoptosis at low concentration of oxidants that are nonlethal for wild-type ES cells. It would, therefore, be worth exploring in the future whether enhanced glycolysis can then promote increased NADPH production via the PPP thereby exerting its antisenescence function.

3 Opposing Effect of Glycolysis on Life Span Observed in Different Organisms 3.1 Discovery of Sir2 as Longevity Gene in Yeast Any discussion about longevity and glycolysis merits evaluation of the results observed in yeast. The hypothesis that enhanced glycolysis can promote proliferation would not be applied in every species. It seems quite opposite in the case

98

H. Kondoh

of budding yeast. In yeast, Sir2 gene was isolated as a longevity gene, which can be activated in a calorie-restricted condition [26]. Calorie restriction is also wellknown to extend the life span of mouse and yeast. In a calorie-restricted condition, glycolysis declines in yeast, accompanied by an elevation of the respiratory rate. Mutations in the glycolytic pathway have also been reported to extend life span in yeast, suggesting that decreased glycolysis has a causal effect on extension of yeast life span. Sir2 works as an NAD-dependent histone deacetylase. In yeast, restricting food availability affects two pathways that activate Sir2 enzymatic activity [27]. First, calorie restriction turns on a gene called PNC1, which produces an enzyme that rids cells of nicotinamide, a small molecule similar to vitamin B3 that normally represses Sir2. Second, activated respiration during calorie restriction is a mode of energy production that creates NAD as a by-product and lowers NADH. In consequence, NAD-dependent Sir2 activity is induced by calorie restriction. Thus, Sir2 is a key molecule linking calorie restriction and longevity in yeast.

3.2 Calorie Restriction and Sir2 It seems that glycolysis has an opposing effect on yeast life span, in sharp contrast with the Warburg effect observed in cancerous or immortalized mammalian cells. To date, it is quite difficult to understand these differences in glycolytic profile between long-lived yeast in calorie restriction and cancer cells in tissue culture, but there are some possible explanations. First, the mammalian homolog of Sir2 gene, called Sirt1, has less impact on aging of mammalian primary cells in tissue culture. Sirt1 can directly bind and modify several DNA binding proteins: histone, p53, Peroxisome Proliferator-Activated Receptor-␥ Coactivator-1 (PGC-1), and possibly other unknown transcriptional factors. At least p53 is not conserved in yeast. Mammalian Sirt1 might target more unknown different factors, which would mask its senescence-bypassing effect observed in yeast. Second, aging of budding yeast is quite different from senescence observed in primary culture cells. The mother yeast cells suffer the accumulation of budding scars on their surface during the division process. In parallel, aging yeast cells show ballooning nucleolus, which is caused by accumulation of circular ribosomal DNAs [28]. These aging phenotypes are quite specific to yeast, which might explain the difference of aging mechanisms between yeast and mammalian culture cells. Third, Sir2 could affect mammalian organismal aging rather than senescence of mammalian culture cells. Resveratrol is a chemical reagent that activates Sirt1 and is enriched in red wine. Resveratrol improves health and survival of mice on a highcalorie diet. In this sense, calorie restriction–induced longevity observed in yeast should be more relevant to mammalian organismal longevity in calorie-restricted condition rather than the Warburg effect observed in cancer cells.


A

99

B Calorie restriction

Glycolysis declined

Calorie restriction

Sir2 activated

Others?

Respiration increased

Sir2 activated

Longevity

Glycolysis?

Longevity

Fig. 3 Impact of calorie restriction and Sir2 on yeast longevity and mammalian organismal aging. (A) In budding yeast, life span is extended in calorie-restricted condition. In such a case, glycolysis declined associated with enhanced respiration. Enhanced respiration will promote NAD production, leading to activation of Sir2. (B) In mouse model, calorie restriction can delay organismal aging. In contrast, high-calorie diet can shorten the life span, which would be restored by treatment with resveratrol, activating chemicals for Sir2. The significance of glycolysis in this situation is still not clear

Collectively, Sir2 is a highly conserved protein and serves as a key to modulate metabolic switching. Sir2 would work as a longevity gene in calorie-limited conditions, but its effect on replicative and stress-induced senescence remains to be clarified (Fig. 3). Thus, yeast might not be a good model to explore and explain the Warburg effect that occurs primarily in cancerous and immortalized mammalian cells.

4 Novel Regulatory Mechanism of Glycolysis 4.1 p53 and Glycolysis Recent advances in understanding the regulatory mechanism of glycolysis also supports our hypothesis that the Warburg effect is a very early event in tumorigenesis. Inactivation of tumor suppressor gene p53 increases glycolytic flux in vitro andin vivo. Ablation of p53 significantly downregulates mitochondrial respiration in vitro and in vivo. p53 is the most frequently mutated gene in various types of cancer and functions as a transcriptional factor to induce cell cycle arrest, apoptosis, and so forth. Inactivation of p53 immortalizes primary cells in vitro, implicating that p53 can also affect the senescence process. p53 knockout mice are cancer prone, possibly due to this senescence effect. Partial activation of p53 induces premature organismal aging

100

H. Kondoh

Extra single copy of p53 Longevity with less tumor events

Activated allele of p53 Premature ageing p53 KO Short lifespan due to cancer prone

Old wild-type Cancer prone

Fig. 4 p53 and organismal life span. p53 null mice are cancer prone, suffering short life span. p53 activating allele knock-in mice display premature aging phenotype. In contrast, mice knocked-in for an extra copy of p53 acquire organismal longevity due to less oxidative damage and fewer tumor events KO-knock out

in vivo[29] (Fig. 4). One possible interpretation would be that one of the major targets of p53 involved in senescence induction impacts metabolic regulation, which renders cells sensitive to oxidative stress. TP53-induced glycolysis and apoptosis regulator (TIGAR) may be such a target through which p53 can modulate oxidative stress in vivo. Recent works suggest that mice knocked-in for an extra copy of p53 acquire longevity associated with a resistance to oxidative damage [30]. It is possible that the metabolic shift caused by p53 might modulate cellular senescence and organismal aging through reduction of oxidative damage in vivo and in vitro. p53 regulates both glycolysis and mitochondrial respiration through different targets. p53 can target glycolytic enzymes (HK and PGM) and modulate the overall flux of glycolysis in primary cells. On the other hand, p53 knockdown can decrease mitochondrial respiration [23]. This is mainly due to inhibition of cytochrome c oxidase 2 (SCO2) [31], a direct target of p53. SCO2 is critical in regulating the cytochrome c oxidase complex, the major site of oxygen utilization. In this way, p53 could modulate both glycolysis and respiration in a concerted manner and would affect cellular life span. This fact suggests that stable alterations at the genetic or epigenetic levels, including inactivation of p53, provide a possible explanation for the enhanced glycolysis that cancer cells present in vitro.

4.2 PGM and Its Regulation Since Dr. Warburg’s discovery, research on glycolysis has been mainly performed by using established cancer and muscle cell lines. But these cells are already immortal with several genetic alterations and presumably not suitable to dissect the impact of immortalizing genes, such as c-Myc or p53 knockdown, on the glycolytic pathway. For example, phosphofructokinase (PFK) is described as the rate-limiting enzyme for glycolysis, while its overexpression has no effect on glycolytic flux


101

in immortalized and nonimmortalized cells. Ectopic expression of other enzymes, PGM or GPI, can increase glycolytic flux drastically in primary cells, indicating that regulation of glycolysis pathway is largely cellular context–dependent. For example, PGM protein levels are tightly regulated by the ubiquitination pathway in primary cells, while they are not so regulated in immortalized cells. Chemical screening of a breast cancer drug identified the PGM inhibitor as the most potent one, suggesting again that PGM can be a key step in the glycolytic pathway in a tissue-specific or cellular context–dependent manner. Interestingly, HIF-1 is a key transcriptional factor that upregulates many glycolytic enzyme genes under hypoxic conditions, but PGM is the only exception. The regulatory mechanism for PGM remains to be clarified.

5 Conclusion Glycolysis serves not only as an energy source for cells but also is essential in the steps of immortalization and transformation during multistep tumorigenesis, as its enhancement can render cells resistant both to oxidative stress and hypoxic conditions, respectively. Cancerous cells show a concerted metabolic shift, including enhanced glycolysis with reduced mitochondrial respiration by an as yet poorly characterized mechanism. The regulation of glycolysis relevant to the senescence process is probably more complicated than we have expected. Future work to clarify its molecular mechanism should be employed in various cell types, including primary cells and ES cells—not just in cancer cells. p53 might be a key molecule as it plays a significant role both in organismal and cellular aging. These findings will also contribute to improve and identify new anticancer therapies in the future.

References 1. Wu X, Pandolfi PP. Mouse models for multistep tumorigenesis. Trends Cell Biol 2001; 11: S2–S9. 2. Braithwaite KL, Rabbitts PH. Multi-step evolution of lung cancer. Semin Cancer Biol 1999; 9:255–265. 3. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000; 100:57–70. 4. Dang CV, Semenza GL. Oncogenic alterations of metabolism. Trends Biochem Sci 1999; 24:68–72. 5. Semenza GL. Hypoxia-inducible factor 1: master regulator of O2 homeostasis. Curr Opin Genet Dev 1998; 8:588–594. 6. Shim H, Dolde C, Lewis BC, et al. c-Myc transactivation of LDH-A: implications for tumor metabolism and growth. Proc Nat Acad Sci USA 1997; 94:6658–6663. 7. Gatenby RA, Gillies RJ. Why do cancers have high aerobic glycolysis? Nat Rev Cancer 2004; 4:891–899. 8. Wang J, Xie LY, Allan S, Beach D, Hannon GJ. Myc activates telomerase. Genes Dev 1998; 12:1769–1774. 9. Kim JW, Zeller KI, Wang Y, et al. Evaluation of myc E-box phylogenetic footprints in glycolytic genes by chromatin immunoprecipitation assays. Mol Cell Biol 2004; 24:5923–5936.

102

H. Kondoh

10. Hayflick L, Moorhead PS. The serial cultivation of human diploid cell strains. Exp Cell Res 1961; 25:585–621. 11. Harman D. Aging: a theory based on free radical and radiation chemistry. J Gerontol 1956; 11:298–300. 12. Wright WE, Shay JW. Historical claims and current interpretations of replicative aging. Nat Biotechnol 2002; 20:682–688. 13. Sherr CJ, DePinho RA. Cellular senescence: mitotic clock or culture shock? Cell 2000; 102:407–410. 14. Serrano M, Blasco MA. Putting the stress on senescence. Curr Opin Cell Biol 2001; 13: 748–753. 15. Campisi J. Cellular senescence as a tumor-suppressor mechanism. Trends Cell Biol 2001; 11:S27–S31. 16. Serrano M, Lin AW, McCurrach ME, Beach D, Lowe SW. Oncogenic ras provokes premature cell senescence associated with accumulation of p53 and p16INK4a. Cell 1997; 88:593–602. 17. Itahana K, Zou Y, Itahana Y, et al. Control of the replicative life span of human fibroblasts by p16 and the polycomb protein Bmi-1. Mol Cell Biol 2003; 23:389–401. 18. Serra V, von Zglinicki T, Lorenz M, Saretzki G. Extracellular superoxide dismutase is a major antioxidant in human fibroblasts and slows telomere shortening. J Biol Chem 2003; 278: 6824–6830. 19. Parkes TL, Elia AJ, Dickinson D, Hilliker AJ, Phillips JP, Boulianne GL. Extension of Drosophila lifespan by overexpression of human SOD1 in motorneurons. Nat Genet 1998; 19:171–174. 20. Orr WC, Sohal RS. Extension of life-span by overexpression of superoxide dismutase and catalase in Drosophila melanogaster. Science 1994; 263:1128–1130. 21. Kondoh H, Lleonart ME, Gil J, et al. Glycolytic enzymes can modulate cellular life span. Cancer Res 2005; 65:177–185. 22. Kondoh H, Lleonart ME, Gil J, Beach D, Peters G. Glycolysis and cellular immortalization. Drug Discov Today 2005; 2:263–267. 23. Kondoh H, Lleonart ME, Nakashima Y, et al. A high glycolytic flux supports the proliferative potential of murine embryonic stem cells. Antioxid Redox Signal 2007; 9:293–299. 24. Tian WN, Braunstein LD, Apse K, et al. Importance of glucose-6-phosphate dehydrogenase activity in cell death. Am J Physiol 1999; 276:C1121–C1131. 25. Ho HY, Cheng ML, Lu FJ, et al. Enhanced oxidative stress and accelerated cellular senescence in glucose-6-phosphate dehydrogenase (G6PD)-deficient human fibroblasts. Free Radic Biol Med 2000; 29:156–169. 26. Sinclair DA, Lin SJ, Guarente L. Life-span extension in yeast. Science 2006; 312:195–197; author reply 195–197. 27. Sinclair DA, Guarente L. Unlocking the secrets of longevity genes. Sci Am 2006; 294:48–51, 54–57. 28. Sinclair DA, Guarente L. Extrachromosomal rDNA circles – a cause of aging in yeast. Cell 1997; 91:1033–1042. 29. Tyner SD, Venkatachalam S, Choi J, et al. p53 mutant mice that display early ageingassociated phenotypes. Nature 2002; 415:45–53. 30. Matheu A, Maraver A, Klatt P, et al. Delayed ageing through damage protection by the Arf/p53 pathway. Nature 2007; 448:375–379. 31. Matoba S, Kang JG, Patino WD, et al. p53 regulates mitochondrial respiration. Science 2006; 312:1650–1653.

Metabolic Modulation of Carcinogenesis Shireesh P. Apte and Rangaprasad Sarangarajan

Abstract Emerging evidence inextricably links cellular pathways that involve glucose metabolism (via oxidative phosphorylation and glycolysis), oxygen sensing, apoptosis, cell cycle regulation, and immune recognition and response. This accounts for the epidemiologic and epigenetic effects of calorie intake and quality, obesity, diabetes, immunosuppression, chronic infection or inflammation, aging, and viral diseases on carcinogenic rates and trends. The mitochondrion is a critical organelle that coordinates and integrates the above-mentioned signaling pathways so as to ultimately affect cell proliferation or cell death. The differential regulation of these pathways not only depends on the varying levels of stress induced by the above-mentioned epigenetic factors but also on acquired genetic mutations, thereby attempting the prediction of cell survival or death, an exceedingly difficult proposition. Concepts that have not received much attention in the literature— such as the role of the exogenous cytochrome C/NADH pathway in apoptosis and oxidative phosphorylation; the switch to glycolysis at lower pyruvate thresholds in cancer cells and the paradox that this glycolytic switch may be necessary to provide ATP and matrix alkalinization necessary for apoptosis to occur; reconciling the observations that caloric restriction causes a generalized decrease in apoptosis yet incidences of cancer are significantly decreased in calorie-restricted animals; the unsaturation patterns of the fatty acids comprising the mitochondria-specific phospholipids; and the variation of the intermediary metabolism of tumors—with time will be addressed from the viewpoint of altered cellular respiration. The implications of attenuation of cellular respiration on the mitochondrial adhesion and localization of key immunoregulatory and glycolytic-apoptotic enzymatic or protein complexes will be discussed. Keywords Glycolysis · Oxidative phosphorylation · Apoptosis · Carcinogenesis · Cell cycle · Hypoxia · Warburg · Calorie restriction · Cytochrome c

S.P. Apte (B) Alcon Laboratories, 6201 South Freeway, Forth Worth, Texas 76134, USA e-mail: [email protected] S.P. Apte, R. Sarangarajan (eds.), Cellular Respiration and Carcinogenesis, DOI 10.1007/978-1-59745-435-3 8, C Humana Press, a part of Springer Science+Business Media, LLC 2009

103

104

S.P. Apte, R. Sarangarajan

Oxidative phosphorylation (OXPHOS), the last of the major steps in aerobic respiration and energy generation, occurs in the mitochondrion, an intracellular organelle that also controls and modulates apoptosis. A dysfunctional OXPHOS or an inadequate oxygen supply can therefore modulate apoptotic function. It can also increase the relative or absolute contribution of an oxygen-independent pathway of respiration and energy generation—glycolysis—to meet cellular energy requirements. This adaptive response temporally progresses from being proapoptotic to becoming antiapoptotic in nature. The proapoptotic effect may predominate in the earlier time period/stage of adaptation if tumor suppressors and apoptotic machinery are functional. The antiapoptotic effect may predominate in the latter times/stages of adaptation if either the tumor suppressors and apoptotic machinery is dysfunctional or is made so by this adaptive response itself.

1 A Decreased OXPHOS Phenotype Attenuates Apoptosis A decreased content of OXPHOS complexes II, III, and IV, reduced level of ATPsynthase, and suppression of NDUFA1, an accessory component of OXPHOS, are all associated with the development of tumors, increased tumor aggressiveness, and antiapoptotic effects. Deficiencies in mtETC, ATP synthesis machinery (F0 F1 ATPase), or ANT abrogate the apoptotic effects of Bax. OXPHOS dysfunction is associated with a proinvasive and pro-proliferative modeling of the ECM by altering transcriptional regulation of genes coding for members of MMP/TIMP, UPA, and CCN in human osteosarcoma cells. An impaired mtETC may increase cellular NADH levels because electrons can no longer be transferred as efficiently to oxygen and NADH cannot be oxidized back to NAD+ . Increased levels of NADH induce protection from apoptosis and decreased cell adhesion by promoting the binding of the tumor suppressor E1A to the (NADH activated) corepressor CtBP. Progression toward a more tumorigenic state is associated with increased and decreased sensitivity to 2-deoxyglucose and oligomycin, respectively, consistent with an increase and decrease in glycolysis and OXPHOS, respectively [1]. Dysfunction or mutations in tumor suppressors may cause attenuation in OXPHOS. A decrease in COX activity due to p53 attenuation can lead to a diminished OXPHOS phenotype [2]. In hepatomas caused by mutated p53, the mitochondrial porin bound hexokinase II is increased [3]. This enzyme isoform binds to the mitochondrial membrane through interaction with VDAC and inhibits apoptosis by suppressing the release of intramembrane space proteins in part by preventing the opening of the PTP [4]. In CCRC, VHL deficiency is one of the factors responsible for downregulation of the biogenesis of OXPHOS complexes, as well as for upregulation of HIF-1␣. AIF deficiency has been shown to produce a reduced OXPHOS phenotype.

Metabolic Modulation of Carcinogenesis

105

2 Reactive Oxygen Species and Carcinogenesis A general picture that emerges from the literature is that cellular ROS can modulate cell-signaling pathways, carcinogenesis, or cell death depending upon their cellular levels. Whether a particular cellular level of ROS triggers proliferation or apoptosis also seems to depend on the magnitude of the cellular stress—which in turn depends on substrate and oxygen supply, the existence of mutations in key tumor suppressor genes, OXPHOS function, and the local cellular environment including the levels of antioxidant enzymes such as SOD and catalase. Mitochondria can generate ROS not only accidentally but also via a regulated enzymatic mechanism, thereby attesting to the importance of ROS in apoptosis. This mechanism involves the electron transfer between the proapoptotic signal p66Shc and cyt-C that generates ROS, which in turn triggers apoptosis by opening the PTP and subsequent caspase activation. Interestingly, the redox reaction between p66Shc and cyt-C is favored by an excess of reduced cyt-C; as may occur under hypoxia when COX activity is decreased. Proapoptotic molecules such as Bax and tBid cause an increase of free intermembrane cyt-C, which is then fully reduced by the rotenone-insensitive NADH-cyt b5 reductase (see Section 8). Therefore, apoptotic induction in attenuated OXPHOS or hypoxic cells routes through the exogenous NADH/cyt-C electron transport pathway. Mitogenic signals during low-stress conditions activate E2F—which in turn decreases ROS—so as to induce proliferation. In conditions of high cellular stress, however, tumor suppressor genes such as p53 and p16INK4a are activated, leading to inhibition of E2F. Consequently, ROS levels accumulate to senescence-promoting levels. High levels of ROS also induce a PKC-mediated block of cytokinesis. Under conditions where either OXPHOS or ROS scavenging is dysfunctional, increased ROS levels may inactivate certain tumor suppressors such as PTEN leading to increased activation of Akt and its downstream targets, HIF-1␣ and VEGF. The activation of the Akt oncogene has been demonstrated to be sufficient to stimulate the switch from OXPHOS to aerobic glycolysis, in part by increasing mitochondria-associated hexokinase activity. Akt is a negative regulator of AMPK activity and hence can overcome the tumor-suppressor activity of LKB1. ROS mediated impaired OXPHOS induced PTEN oxidation may also suppress p53 activity by allowing nuclear translocation of Mdm2.

3 The Hexosamine Biosynthetic Pathway Adaptive glycolysis triggered by impaired OXPHOS, increased ROS generation, hypoxia, or deregulation of nutrient uptake pathways may siphon more glucose into the hexosamine biosynthetic pathway (HBP). Apart from causing insulin resistance, an increased flux through the HBP may cause increased O-GlcNAc glycosylation

106


and inactivation of the proapoptotic protein Bnip3 [5], thereby conferring a survival advantage on glycolytic cells [6]. Furthermore, increased posttranslational modification of transcriptional factors such as Sp1—mediated by O-GlcNAc glycosylation—may result in downregulation of OXPHOS gene expression and decreased mitochondrial function by impairing the activity or expression of known regulators of OXPHOS transcription PGC-1␣ and NRF [7]. This phenomenon may further exacerbate the shift from OXPHOS to glycolysis.

4 Diversion of Pyruvate from Mitochondria The prevalence of the Crabtree effect (a preference to ferment glucose into ethanol) in the bakers’ yeast, Saccharomyces cerevisiae, has been attributed in part to differences in enzymes at the branch point between respiration (PDH) and fermentation (PDC). The concentration of pyruvate that gives half maximal activity (Km ) of PDH is lower than that for PDC. The latter enzyme exhibits cooperativity with substrate concentration and also has a higher maximal activity. The net result is to make metabolic flux through PDH favored at low pyruvate concentrations and flux through PDC favored at higher ones. It is instructive to compare aerobic metabolism in this organism to that exhibited by tumor cells. A decreased NAD:NADH ratio, which occurs in part as a consequence of impaired mtETC and which is characteristic of highly glycolytic malignant cells, downregulates the PDH complex [8] so that entry of pyruvate into the TCA cycle is impaired. The unusual metabolite, acetoin, unique to tumors and formed by the nonoxidative decarboxylation of pyruvate with acetic acid, is a competitive inhibitor of intramitochondrial tumor PDH. Km values for pyruvate for the PDH of tumor cell lines were significantly lower (up to threefold) than were those for the PDH of non-transformed cells [9]. Broadly interpreted, the concentration of pyruvate below which metabolic flux is favored through OXPHOS may be lower in tumor cells, thereby contributing toward an oxygen supply–independent “pyruvate siphoning” into glycolysis. This interpretation also implies that changes in metabolic enzyme activity/polymorphism may be sufficient to initiate the “switch to glycolysis” independent of the cellular oxygen concentration. Nonhypoxic activation of HIF-1 through loss of VHL or activation of Akt or other signal transduction pathways could increase PDK1 and trigger the Warburg effect. HIF-1–mediated increased PDK1 activity has been shown to inhibit PDH, causing the shunting of pyruvate from the mitochondria and subsequently attenuating mitochondrial respiration and ROS production in MEFs. The influx of respiratory substrate into mitochondria has been shown to be an important determinant of cell sensitivity to oxidant-induced apoptosis. Reduced OXPHOS may hence induce low sensitivity to oxidative stress.


107

5 Organelle and Cell Membrane Perturbations Complexation of the mitochondria-specific phospholipid cardiolipin (CL) with cytC enables the latter to act as a CL-oxygenase, which is activated during apoptosis and selectively oxidizes CL. Oxidized CL is required for the release of proapoptotic factors from the mitochondria into the cytosol. Progression of the apoptotic program is hence dependent on the susceptibility of CL to oxidation; which in turn is dependent on its polyunsaturation pattern [10]. Consistent with this observation, enrichment of CL with the highly unsaturated fatty acid docosahexaenoic acid conferred HL-60 cells with increased sensitivity to apoptosis induced by staurosporine [11]. Hypoxia may be partly the effect of the incorporation of adulterated, nonoxygenating, or oxidized polyunsaturated fatty acids into the phospholipids of cellular and mitochondrial membranes [12]. Such incorporation is hypothesized to cause changes in membrane permeability that impair oxygen transmission into the cell. Trans fats, partially oxidized PUFA entities, and inappropriate omega 6:omega 3 ratios are all potential sources of unsaturated fatty acids that can disrupt the normal membrane structure and oxygen diffusivity. Significant differences have been observed between the phospholipid fatty acid profiles of the human colorectal mucosa among the control, adenoma, and carcinoma groups [13]. Acetoin has been shown to have a potentiating effect on citrate formation, which in turn is responsible for cytoplasmic cholesterol synthesis from the truncated tumoral Krebs cycle [14]. Deregulation of cholesterol synthesis has been observed in various cancer cell lines leading to its accumulation in every type of cell membrane. Cholesterol enrichment in membranes induces changes in biophysical properties, such as increased rigidification and decreased passive proton permeability. The accumulation of cholesterol, in combination with altered ratios of omega 6:omega 3 fatty acids, partially oxidized PUFA entities, and trans-esters of fatty acids have been hypothesized to impair oxygen transmission into the cell.

6 How Calorie Restriction Inhibits Carcinogenesis Nowhere is the link between metabolism and carcinogenesis more profoundly demonstrated than in studies of caloric restriction. CR has unequivocally been associated with increased life span and decreased carcinogenesis. In contrast, metabolic disorders such as obesity [15] and diabetes [16] are known risk factors for the development of cancer. CR has been shown to promote mitochondrial biogenesis and increase bioenergetic efficiency by modulating eNOS, PGC-1␣, and SIRT1 expression [17]. IGF1 inhibition by ER downregulates the longevity inhibitory gene product p66Shc and upregulates the induction of NRF1 and NRF2␤␥ mRNA. The expression of mammalian SIRT1 is induced in CR rats as well as in human cells that are treated with serum from such animals. SIRT1 deacetylates the DNA repair factor ku70, causing it to sequester the proapoptotic factor Bax away

108


from mitochondria, thereby inhibiting stress-induced apoptotic cell death. CR rats showed less apoptosis in skeletal muscle compared with their age-matched cohorts. CR produces a reduction in body temperature of rodents and nonhuman primates. This has been suggested to occur by an increase in the coupling of oxidative phosphorylation to ATP synthesis. Energy-deprived HepG2 hepatoma cells showed increased mitochondrial biogenesis and OXPHOS. OXPHOS efficiency, as indicated by a higher ADP:O ratio, lower H2 O2 production, and lower mtDNA content was found to correlate with Drosophila simulans survival [18]. Nutrient-induced downregulation of OXPHOS genes in skeletal muscle has been suggested to represent a thrifty response to nutrient excess, which attempts to limit the compensatory increase in energy expenditure normally designed to counterbalance the increase in energy intake. If caloric restriction causes a generalized decrease in apoptosis, then it should be conducive to carcinogenesis. However, incidences of cancer are significantly decreased in calorie-restricted animals. This apparent paradox can be explained by invoking the hypothesis that growth advantage for mutated cells (along with the probability of monoclonality and malignancy) is considerably increased (in part due to decoupling and consequent deregulation of nutrient intake pathways independent of the extracellular matrix) if the normal multiclonal environment is damaged by a pathologic proapoptotic process that spares the apoptosis-resistant clones. If the multiclonal environment is more stress resistant and robust—as would be expected to be the case with CR—the selective growth advantage of mutated cells is diminished. As an example, senescence creates a local tissue environment that promotes the growth of initiated or preneoplastic epithelial cells both in culture and in vivo [19]. To illustrate this effect further, CR results in a negative regulation of p53mediated transcriptional activation due to its deacetylation by SIRT1. A slight reduction in p53 activity permits the expression of several genes involved in cell-cycle arrest (p21WAF1/CIP1 ), DNA repair (p53R2), and those involved in protection against oxidative stress (TIGAR, sestrins, GPX1), thereby allowing cell survival. In contrast; a proapoptotic process mediated by elevated levels of p53 leads to the induction of pro-oxidant genes (PIG3, praline oxidase), the repression of transcription of antioxidant genes (PGM, NQO1), and results in elevated intracellular ROS levels and cell death or senescence in nonmutated cells or tumorigenesis in mutated or apoptosis-resistant clones. This differential regulation of target genes in response to varying levels of stress seems to be generally applicable also to ROS and Ca2+ in the context of cell survival or death. Apoptosis in the earlier stage of carcinogenesis may be necessary to increase the selective pressure on the initiated cells to favor the emergence of proliferative, apoptosis-resistant cells. For example, it has been hypothesized that Bax toxicity may be a result of a Bax-induced impairment in the ability to perform OXPHOS. Consistent with this hypothesis, it has been reported that Bax-induced lethality is diminished under conditions that favor fermentation and that the cells that recover from growth arrest after Bax expression frequently exhibit a permanent loss of respiration competence [20]. Deregulated activity of bcl-2 family members enhances


109

tumor formation when overexpressed in the upper layers of the epidermis but retards tumor formation when overexpression is restricted to the basal layer. Furthermore, apoptosis-induced selection pressure in the early stage of carcinogenesis is supported by the finding that coexpression of two proapoptotic genes (transforming growth factor ␤1 and the hepatitis B virus X) potentiates c-Myc– induced hepatocarcinogenesis. Clonal expansion of potentially malignant p53-deficient cell variants was suppressed in Bcl-2 overexpressing tumors, presumably due to the lack of selective advantages for p53-deficient cells inside a population of cells resistant to apoptosis. There is an age-related increase in the proportion of liver mitochondria with fragile outer membranes as evidenced by an increased release of adenylate kinase in hypotonic medium. Peroxidized cardiolipid in the mitochondrial membrane lowers the threshold of Ca2+ for PTP induction and cyt-C release. Aging mice exhibit enhanced PTP activation in lymphocytes, brain, and liver. These age-related, chronic, mitochondrial degenerative, apoptosis-favoring phenomena further increase the probability of growth advantage of a subset of apoptosisresistant mutant clones. That may be why pathologic proapoptotic processes such as paroxysmal nocturnal hemoglobinuria, myelodysplastic syndromes, chronic myeloid leukemia, secondary acute leukemias, and immunosuppression-related non–Hodgkin’s lymphomas may be interpreted as “opportunistic” clonal and malignant diseases [21]. CR may cause a generalized increase in efficiency of OXPHOS because it causes a decreased respiratory substrate flux through the HBP and decreased levels of UDPGlcNAc, the main substrate for protein glycosylation (see Section 3).

7 Chronic Infection and Carcinogenesis Hypoxia is a characteristic feature of the tissue microenvironment during bacterial infection. HIF-1␣ has been shown to be induced by bacterial infection, even under normoxia, and regulates the production of key immune effector molecules, including granule proteases, antimicrobial peptides, nitric oxide, and TNF-␣. Mice lacking HIF-1␣ in their myeloid cell lineage showed decreased bactericidal activity and failed to restrict systemic spread of infection from an initial tissue focus. Conversely, activation of the HIF-1␣ pathway through deletion of VHL tumor-suppressor protein or pharmacologic inducers supported myeloid cell production of defense factors and improved bactericidal capacity. Therefore, any chronic infectious condition, whether or not triggered by pathogens, may have the ability to upregulate HIF in the vicinity of the affected cells. Among agents that have been reported to cause uncoupling of OXPHOS to ATP synthesis or a decrease in OXPHOS are Helicobacter pylori and hepatitis B and the Epstein-Barr viruses; these are also selectogens for non–Hodgkin s lymphoma, liver cancer, and Burkitt s lymphoma, respectively. Epstein-Barr virus has also been shown to induce synthesis of HIF-1␣, and the protein is also upregulated in

110


macrophages at normoxia after exposure to bacteria. Uncoupling protein 2 (UCP2) expression is increased in most human colon cancers, and the level of expression correlates with the degree of neoplastic changes. Bacterial lipopolysaccharide has been shown to increase the expression of mitochondrial UCP2 in several tissues. Differential compartmentalization of signaling molecules in cells is emerging as an important mechanism for regulating the specificity of signal transduction pathways. Altered stress tolerance (and therefore altered heat shock protein levels) in cells with dysfunctional OXPHOS can affect protein folding and thereby disrupt their membrane localization or rate of ubiquitination [22]. Candidates subject to such a pathologic response include the antiviral protein MAVS, whose disruption of mitochondrial localization renders it unable to activate NF-␬B or IRF3 to induce type-I interferons. The resultant attenuation of innate immunity can sensitize the cell to carcinogenesis [23]. Increased MHC class I expression serves to alert the immune system to cells with mitochondrial mutations [24]. Mice deficient in MHC I accumulate mitochondrial DNA deletions in various tissues. Therefore, immunocompromised individuals may not be effectively able to eliminate cells containing mitochondrial DNA mutations. Indeed, NHL has been reported to occur more frequently in AIDS and in posttransplant immunocompromised patients.

8 Electron Transport Modulation It has been shown that the desorbed intermembrane fraction of endogenous cyt-C can function as an electron shuttle between the outer and inner membranes of intact mitochondria. The oxidation of extramitochondrial NADH can proceed through the rotenone-insensitive respiratory chain of the outer membrane, leading to cytb5 reduction. From cyt-b5 , electrons are channeled via the intermembrane cyt-C to COX. In physiologic conditions, cyt-C is constitutively released outside the mitochondria to activate the exogenous NADH/cyt-C electron transport pathway. This pathway, at least part of which routes through the VDAC, may attempt to maintain the membrane electrochemical potential that is necessary for ATP synthesis in the event of dysfunction of the respiratory chain upstream of complex IV [25]. It has been demonstrated that electrons from the external Nde1p have the right of way over those coming from either Gut2p or Ndip in Saccharomyces cerevisiae. Might a differential regulation for electron transfer in mammalian cells exist, such that under normal conditions, electrons originating from the citric acid cycle are preferentially used over those originating from the exogenous NADH/cyt-C pathway? Consistent with this proposition is the observation from control analysis that NADH oxidation has negligible control over the carbon flux through OXPHOS. Under these circumstances, blockage of the citric acid cycle or a general reduction of mitochondrial OXPHOS could shift the bulk of the burden of electron transfer to the exogenous NADH/cyt-C pathway. This may necessitate a greater concentration of extramitochondrial cyt-C to maintain mitochondrial ATP levels, which, coincidently, may be enough to activate procaspase 9 and cause apoptosis. This may also


111

explain why cyt-C has been chosen by biological evolution to be a mediator of both respiration and apoptosis and why gene promoters regulating the cell cycle contain HRE elements (see Section 11).

9 The “Switch to Glycolysis” Is Necessary to Induce Apoptosis Lymphocytes, enterocytes, and fetal tissues that exhibit rapid growth are poorly oxidative, whereas highly oxidative tissues such as kidney cortex or brain are normally quiescent. It has long been noted that tumor cells display elevated rates of glycolysis, despite oxygen availability. Crabtree observed that even though most tumor cells are capable of oxidative phosphorylation, extracellular glucose represses oxidative phosphorylation in tumor cells via an unknown mechanism. Self-sufficiency in growth signals and the ability to take up nutrients independently of the extracellular environment has been postulated to be a primary acquired capability of cancer. It has been demonstrated that eosinophils die by apoptosis even in the absence of a functioning OXPHOS. They do so by importing and hydrolyzing glycolytic ATP into mitochondria to maintain a transmembrane potential. This may be accomplished by operating the bidirectional Fo F1 -ATPase proton pump in reverse. Stimuli triggering apoptosis have been shown to be partly attributable to the reverse operation of this H+ pump, which causes an increase in mitochondrial matrix pH—an early event associated with mitochondrial dependent apoptosis. Several reports in the literature suggest that the increase in ATP levels in cells en route to apoptosis derives from both OXPHOS and glycolysis. It has also been shown that apoptosis can occur in cells that lack functional mitochondria and that ATP synthesized via glycolysis plays a critical role in preventing the collapse of the mitochondrial membrane potential thereby allowing the cell to die by apoptosis. Further support for this hypothesis comes from the observation that staurosporineinduced apoptotic ATP elevation in caspase-competent HeLa cells occurs via Ca2+ dependent activation of glycolytic ATP synthesis, with little or no contribution from mitochondrial OXPHOS. TNF-induced hepatic apoptosis could be selectively blocked by ketohexose-mediated ATP depletion. Because apoptosis is generally recognized as an ATP-dependent process, glycolysis may provide adequate ATP to activate caspases. Procaspase 9 may be activated by an increase in cytosolic cyt-C levels, which in turn are increased due to age-related or pathogenic attenuation of mitochondrial OXPHOS (discussed in Section 8). Cytosolic acidification caused by the glycolytic product, lactic acid, has been shown to be required for induction of caspase activation and apoptosis; it also decreases the ability of hexokinase to bind to the outer mitochondrial membrane. Increased susceptibility to stress-induced apoptosis in c-myc transformed cells has been linked to the overexpression of the LDH gene. This leads to the intriguing—as yet experimentally unproven—proposition that transient aerobic glycolysis may precede apoptosis in OXPHOS-deficient nontransformed cells and may even be necessary for such cells to undergo programmed cell death.

112


On the other hand, in transformed or preneoplastic cells, where caspasedependent apoptotic pathways are blocked or rendered nonfunctional by other non– OXPHOS-related mutations (such as PTEN, P53, TSC2, VHL, etc.), the shift to glycolysis to supply adequate ATP to fuel caspase-dependent apoptotic pathways may, ironically, give these cells a selective growth advantage over non-transformed neighboring cells. Because glucose phosphorylation, the first committed step of glycolysis, is sufficient to promote cell survival by Akt, it seems possible that even a transient shift to glycolysis in the presence of non–OXPHOS-related mutations may promote cell immortalization.

10 Regulation of the Cell Cycle and the Apoptotic Pathway Is Sensitive to Glucose Concentration Increased rates of glycolysis in cultured proliferating human lymphocytes and rat thymocytes have been reported to peak in concert with DNA synthesis. The glycolytic enzymes GAPD and LDH have been found in transcriptional coactivator complexes with the OCT-S transcription factor, which stimulates the expression of histone H2A as cells enter the S-phase. Similar to these glycolytic enzymes that display nonglycolytic functions among their regulatory DNA sequences, inhibition of PFK-2, coded by the pfkfb3 gene, produces proapoptotic effects in HeLa cells. The pfkfb3 gene promoter contains HIF-1 binding sites. The cyclin D1 promotor contains two consensus HREs. Cyclin D1 has been shown to be overexpressed in VHL-deficient RCC cell lines. A cyclin D1–dependent kinase was found to phosphorylate and inactivate NRF1 with a consequent decrease in OXPHOS. The proapoptotic protein, BAD, is required to assemble a mitochondrially located complex that also contains hexokinase IV. Furthermore, the phosphorylation status of BAD regulates hexokinase IV activity. The GlcRE within the PK promoter was found to be sufficient to confer the modulation by HIF-1 of the glucose-dependent induction of the PK gene expression. HIF-1 also activates telomerase via transcriptional activation of the human telomerase reverse transcriptase gene. These results suggest that regulatory points exist in the cell cycle, oxygen sensing and apoptotic pathways that are sensitive to glucose metabolism thereby providing direct biochemical and genetic links between oxygen availability, metabolism, and carcinogenesis.

11 Temporal and Spatial Effects in Tumorigenesis In VHL-deficient RCC cell lines, the resulting stabilization of HIF-1␣ negatively regulates mitochondrial biogenesis by inhibiting the transcription of the gene encoding the coactivator factor PGC-1␤. This is achieved by the downregulation of c-myc. The concomitant stabilization of HIF-1␤, on the other hand, increases c-myc activ-


113

ity and promotes cell-cycle progression in part through increased c-myc promoter binding. Even though such opposing transcription factors (HIF-1␣ and HIF-1␤) may coexist during hypoxia, the differential activation both temporally and spatially of their downstream gene targets may still drive tumor growth instead of a net “cancellation” of their effects [26]. The variation of the intermediary metabolism of tumors with time has been proposed with respect to the cellular energy sensor, AMPK. This kinase facilitates glucose and fatty acid metabolism. However, it also switches off anabolic processes, including proliferation through mTOR via TSC2. Active AMPK, therefore, has properties that facilitate advanced tumorigenesis yet reduce the selective advantage of cancer cells. The solution to this paradox is that some advanced cancers often retain a sufficient complement of defective downstream tumor suppressors, which allow a degree of AMPK activation while mitigating its growth-limiting effect. For example, some cancers harbor TSC2 mutations, allowing AMPK activation without mTOR suppression. On the other hand, for premalignant cells that are not limited by substrate supply and that retain intact cellular senescence machinery such as p53, TSC2, and mTOR, active AMPK is a disadvantage because it slows cellular proliferation. The inhibition of AMPK at this stage in tumorigenesis may be achieved, in part, by increasing mitochondrial ATP production. Indeed, mitochondrial biogenesis has been shown to increase to its greatest extent at an intermediate degree of transformation rather than at the fully transformed state.

12 The Warburg Model The sequence of events arising from the defective functioning of the electron transfer chain in mitochondria may thus be projected to occur thus: Because the electron transfer chain uses reducing equivalents in the form of NADH as a source of electrons, a defect in this chain (due to multiple reasons) leads to an intramitochondrial elevation of (unused) NADH levels. This leads to retrograde product inhibition of the citric acid cycle, which in turns leads to a similar retrograde inhibition of the NADH shuttle that transfers reducing equivalents from the cytosol to the mitochondrion [8, 27]. The increase in mitochondrial NADH levels is thus spread to the cytoplasm. The increased NADH level in the cytosol facilitates the conversion of pyruvate into lactate via aerobic glycolysis, thereby diverting the pool of aerobic pyruvate away from the mitochondria to compensate for the lack of electron transport and oxidative phosphorylation. At the same time, this increased NADH/NAD+ ratio also acts to limit respiratory flux through glycolysis and promotes senescence; both effects are due to inhibition of key glycolytic enzymes, especially GAPD. The forced conversion of pyruvate to lactate via aerobic glycolysis [1] uses NADH thereby serving as a mechanism to decrease NADH-mediated Akt activation [28] and [2] generates NAD+ , which can in turn activate PARP to force the cell to undergo necrosis [29] partly by a mechanism that induces poly(ADP)ribosylation and inactivation of GAPD [30]. An increased concentration of lactate acidifies the

114


cytosol, which is conducive to activation of caspases and the accumulation of the proapoptotic protein BNIP3 [31]. Cytosolic acidification also decreases the ability of hexokinase to bind to the outer mitochondrial membrane. 2-Methyl glyoxal, the final product of aerobic glycolysis, acts to inhibit further increase in glycolysis by a retrograde product inhibition. Lastly, the aerobic route of glycolysis makes the cell more susceptible to death by glucose deprivation and death-receptor–mediated apoptosis [32] (Table 1). The increased cytosolic NADH levels are also postulated to increase the burden of electron transfer to the exogenous NADH/cyt-C electron transfer pathway. This necessitates an increase in the concentration of extra-mitochondrial cyt-C, which coincidently may be enough to trigger activation of APAF and the activation of procaspase 9—ultimately activating apoptosis. Because the mitochondria cannot produce ATP from the oxidation of glucose via the electron transfer chain, it attempts to bring in glycolytic ATP into the mitochondria via the ANT/VDAC channels. Because the import of phosphate into mitochon-

Table 1 Apoptotic Pathways Activated by Glycolysis Attribute Increasing flux through glycolysis

Effect NADH / NAD+

GAPD

ratio Exogenous NADH/cyt-C electron pathway ROS

PARP,

NADH

Akt

NAD+

PARP

Lactate Hexokinase binding to mitochondrial membrane

pH, Caspases

Increased apoptosis, necrosis

GAPD

BNIP3,

PTP flux Reverse operation of ATPase

ATP

Glycolysis, LDH

Mitochondrial pH Disproportionate decrease in flux through glycolysis than through OXPHOS Susceptibility to stress, glucose deprivation and death receptor

Increasing metabolic flux through glycolysis activates proapoptotic pathways and mechanisms. These may induce death in cells whose OXPHOS is impaired but which possess functional tumor suppressors and apoptotic machinery. Metabolic deviation may therefore act as a “self-regulating mechanism” in such cells to prevent carcinogenesis. On the other hand, OXPHOS dysfunction in association with other mutations may tip the balance toward proliferation, thereby causing carcinogenesis.


115

dria is blocked, the mitochondrial transmembrane potential can only be maintained by operating the FoF1-ATPase pump in reverse, by hydrolyzing the mitochondrial ATP pool to pump protons out into the intermembrane space. This results in the alkalinization of the mitochondrial matrix, which, in tandem with the acidification of the cytosol, is generally recognized as the first step to initiate apoptosis. Control analysis studies show that ATP consumption exercises significantly more control over the flux through glycolysis than it does through OXPHOS [33]; consequently, an overall decreased cellular ATP level should disproportionately decrease carbon flux through glycolysis than it does through OXPHOS.

13 Conclusion Although aberrant functioning of the mitochondrial OXPHOS respiratory apparatus is well documented in the literature beginning with the observations by Warburg, emerging evidence suggests that the respiratory contribution from OXPHOS and from the cell surface may not be negligible, when compared with that from glycolysis, in proliferating transformed cells. These observations, when taken together with evidence suggesting that glycolysis may be a cell-death mediator of last resort [34]—capable of activating cell death pathways either on its own or in conjunction with OXPHOS—suggest that deficient OXPHOS may be sufficient but not necessary to initiate malignancy, and the “switch to glycolysis” may be coincidental to but not causative of this transformation. In fact, it may be speculated that cells with adequately functioning tumor suppressors and apoptotic machinery may be able to derive the majority of their cellular energy requirements by “switching” back and forth between OXPHOS and glycolysis numerous times over their life span depending on substrate or oxygen availability or as a response to moderate oxidative stress without undergoing transformation. In pancreatic beta cells, the ATP derived from mitochondrial pyruvate metabolism does not significantly contribute to the glucose-stimulated insulin secretion; instead, the glycolytic metabolism of glucose is critically involved in this signaling process [35]. If the cellular energy sensor AMPK in tumor cells were altered similarly, its “setpoint” would be decreased significantly, allowing the cancer cell to maintain respiratory flux through anabolic metabolic processes even in the presence of significantly reduced cellular ATP (due to impaired OXPHOS). More studies such as the one cited in reference [36] are needed that specifically induce a selective block of aerobic glycolysis to ascertain its effect on the initiation and maintenance of the malignant phenotype. The contribution of pathways other than OXPHOS and glycolysis, such as the electron transport pathway activated by exogenous cyt-C and the trans plasma membrane electron transport pathway, should be investigated in much more detail with regard to substrate utilization, oxygen consumption, and energy generation. Experiments should be designed to supply cell cultures not only with just glucose or glutamine but also with the whole complement of other substrate fuels at physiologic concentrations. Metabolic alterations

116


should be examined in the absence of growth factor stimulation because of its facilitating effects on the metabolic shift toward aerobic glycolysis [36]. Such metabolic alterations can be quantitatively measured through control analysis, and this technique could be expanded to include hypoxic and substrate-enhancing or -limiting conditions as well. Levels of the whole complement of glycolytic enzymes should be measured in order to be able to make an accurate assessment of deviations of metabolic flux in non-transformed as well as in transformed cells. Effects should be compared between primary, immortalized, and neoplastic cells to dissect cellspecific effects.

14 Abbreviations AIF, apoptosis inducible factor; Akt, v-akt murine thymoma viral oncogene; ALDH2, aldehyde dehydrogenase allele 2; AMPK, AMP activated protein kinase; ANT, adenine nucleotide translocator; BCC, basal cell carcinoma; Bnip3, Bcl2/adenovirus E1B 19 kDa interacting protein 3; CCN, connective tissue growth factor and nephroblastoma–overexpressed gene; CCRC, clear cell renal carcinoma; cGDPH, cytosolic glycerol-3-phosphate dehydrogenase; CHO, Chinese hamster ovary; COX, cytochrome c oxidase; CR, calorie restriction; CREB, cAMP responsive element binding; CtBP, C-terminal binding protein; cyt-C, cytochrome C; DM2, type 2 diabetes mellitus; E1A, adenovirus 5 early region 1A oncoprotein; ECM, extracellular matrix; eNOS, endothelial nitric oxide synthase; ES, embryonic stem; FCCP, carbonylcyanide-p-trifluoromethoxyphenylhydrazone; FH, fumarate hydratase; GAPD, glyceraldehyde-3-phosphate dehydrogenase; GlcRE, glucose responsive element; GPI, glucose phosphatase isomerase; GPX1, glutathione peroxidase 1; Gut2p, glycerol 3-phosphate dehydrogenase; HBP, hexosamine biosynthetic pathway; HIF, hypoxia inducible factor; HRE, hypoxia regulating elements; IAP-2, inhibitor of apoptosis protein 2; IGF-1, insulin growth factor; IRF3, interferon regulating factor 3; LDH, lactate dehydrogenase; LKB1, serine threonine protein kinase 11; MAVS, mitochondrial antiviral signaling protein; Mdm2, mouse double minute 2; MDR, multidrug resistance; MEF, mouse embryo fibroblasts; MELAS, mitochondrial encephalomyopathy, lactic acidosis and strokelike episodes; MMP, matrix metalloproteinase; mtDNA, mitochondrial DNA; mtETC, mitochondrial electron transport chain; mTOR, mammalian target of rapamycin; NAD+ /NADH, nicotinamide adenine dinucleotide (oxidized and reduced forms, respectively); Nde1p and Nde2p, external NADH dehydrogenases; NDUFA1, NADH dehydrogenase (ubiquinone) alpha subcomplex 1; NF-␬B, nuclear factor kappa B; NHL, non–Hodgkin’s lymphoma; NQO1, NADPH dehydrogenase quinone 1; NRF, nuclear respiratory factor; NSCLC, non–small cell lung cancer; O-GlcNAc, O-linked acetylglucosamine; OXPHOS, oxidative phosphorylation; p21Waf1/Cip1 , cyclin dependent kinase inhibitor 1A; PDC, pyruvate decarboxylase; PDH, prolyl hydroxylase; PDH, pyruvate dehydrogenase; PDK1, pyruvate dehydrogenase kinase 1; PFK-2, 6-phosphofructo-2 kinase/fructose-2,6-bisphosphatase; PGC, peroxisome prolifer-


117

ator activator protein–␥–coactivator-1␣; PGM, phosphoglycerate mutase; PIG3, p53 inducible gene 3; PK, pyruvate kinase; PNP, purine nucleoside phosphorylase; PPAR, peroxisome proliferator activated receptor; PTEN, phosphatase and tensin homolog; PTP, permeability transition pore; PUFA, polyunsaturated fatty acids; RCC, renal cell carcinoma; ROS, reactive oxygen species; SCC, squamous cell carcinoma; SDH, succinyl dehydrogenase; SIRT1, silent information regulator 2-homolog 1; SOD, superoxide dismutase; TCA, tricarboxylic acid; TIGAR, TP53 induced glycolysis and apoptosis regulator; TIMP, tissue inhibitor of matrix metalloproteinase; TNF, tumor necrosis factor; TSC2, tuberous sclerosis complex; UCP2, uncoupling protein 2; UDP-GlcNAc, uridine diphosphate-N-acetylglucosamine; UPA, urokinase plasminogen activator; VDAC, voltage-dependent anion channel; VHL, von Hippel–Lindau; VLB, vinblastine.

References 1. Ramanathan A, Wang C, Schreiber SL. Perturbational profiling of a cell line model of tumorigenesis by using metabolic measurements. Proc Natl Acad Sci USA 2005; 102(17): 5992–5997. 2. Poyton RO. Assembling a time bomb – cytochrome c oxidase and disease. Nat Genet 1998; 20:316–317. 3. Pedersen P. Tumor mitochondria and the bioenergetics of cancer cells. Prog Exp Tumor Res 1978; 22:190–274. 4. Pastorino JG, Hock JB. Hexokinase II: the integration of energy metabolism and control of apoptosis. Curr Med Chem 2003; 10(16):1535–1551. 5. Manka D, Millhorn DE. A potential molecular link between aerobic glycolysis and cancer. Cell Cycle 2006; 5(4):343–344. 6. Lee H, Paik SG. Regulation of BNIP3 in normal and cancer cells. Mol Cells 2006; 21(1):1–6. 7. Scarpulla RC. Transcriptional activators and coactivators in the nuclear control of mitochondrial function in mammalian cells. Gene 2002; 286:81–89. 8. Matthew CK, Van Holde KE, Ahern KG. Biochemistry, 3rd ed. Boston: Addison-Wesley, 2000. 9. Baggetto LG, Lehninger AL. Isolated tumoral pyruvate dehydrogenase can synthesize acetoin which inhibits pyruvate oxidation as well as other aldehydes. Biochem Biophys Res Commun 1987; 145(1):153–159. 10. Kagan VE, Tyurin VA, Jiang J, et al. Cytochrome C acts as a cardiolipin oxygenase required for release of proapoptotic factors. Nat Chem Biol 2005; 1:223–232. 11. Kagan VE, Borisenko GG, Tyurina YY, et al. Oxidative lipidomics of apoptosis: redox catalytic interactions of cytochrome C with cardiolipin and phosphatidylserine. Free Radic Biol Med 2004; 37(12):1963–1985. 12. Peskin BS, Carter MJ. Chronic cellular hypoxia as the prime cause of cancer: what is the de-oxygenating role of adulterated and improper ratios of polyunsaturated fatty acids when incorporated into cell membranes? Med Hypoth 2008; 70:298–304 13. Shim YJ, Choi KY, Lee WC, Kim MK, Lee SY, Lee-Kim YC. Phospholipid fatty acid patterns in the mucosa of human colorectal adenomas and carcinomas. Nutr Res 2005; 25: 261–269. 14. Baggetto LG, Testa-Parussini R. Role of acetoin on the regulation of intermediate metabolism of Ehrlich Ascites tumor mitochondria: its contribution to membrane cholesterol enrichment modifying passive proton permeability. Arch Biochem Biophys 1990; 283(2):241–248. 15. Bianchini F, Kaaks R, Vainio H. Overweight, obesity and cancer risk. Lancet Oncol 2002; 3(9):565–574.

118


16. Coughlin SS, Calle EE, Teras LR, Petrelli J, Thun MJ. Diabetes mellitus as a predictor of cancer mortality in a large cohort of US adults. Am J Epidemiol 2004; 159:1160–1167. 17. Nisoli E, Tonello C, Cardile A, Cozzi V, Bracale R, Tedesco L. Calorie restriction promotes mitochondrial biogenesis by inducing the expression of eNOS. Science 2005; 310(5746): 314–317. 18. Melvin RG, Ballard JWO, William O. Intraspecific variation in survival and mitochondrial oxidative phosphorylation in wild-caught Drosophila simulans. Aging Cell. 2006; 5(3): 225–233. 19. Campisi J. Senescent cells, tumor suppression and organismal aging: good citizens, bad neighbors. Cell 2005; 120:513–522. 20. Harris MH, Vander Heiden MG, Kron SJ, Thompson CB. Role of oxidative phosphorylation in Bax toxicity. Mol Cell Biol 2000; 20(10):3590–3596. 21. Cucuianu A. Cell Darwinism, apoptosis, free radicals and haematological malignancies. Med Hypotheses 2001; 56(1):52–57. 22. Li XD, Sun L, Seth RB, Pineda G, Chen ZJ. Hepatitis C virus protease NS3/4A cleaves mitochondrial antiviral signaling protein off the mitochondria to evade innate immunity. Proc Natl Acad Sci USA 2005; 102(49):17717–17722. 23. Dunn GP, Old LJ, Schreiber RD. The immunobiology of cancer: immunosurveillance and immunoediting. Immunity 2004; 21:137–148. 24. Gu Y, Wang C, Roifman CM, Cohen A. Role of MHC class I immune surveillance of mitochondrial DNA integrity. J Immunol 2003; 170:3603–3607. 25. Piana GL, Marzulli D, Gorgoglione V, Lofromento NE. Porin and cytochrome oxidase containing contact sites involved in the oxidation of cytosolic NADH. Arch Biochem Biophys 2005; 436:91–100. 26. Gordan JD, Thompson CB, Simon MC. HIF and c-myc: sibling rivals for control of cancer cell metabolism and proliferation. Cancer Cell 2007; 12(2):108–113. 27. Nyengaard JR, Ido Y, Kilo C, Williamson JR. Interactions between hyperglycemia and hypoxia. Diabetes 2004; 53:2931–2938. 28. Pelicano H, Xu RH, Du M, et al. Mitochondrial respiration defects in cancer cells cause activation of Akt survival pathway through a redox–mediated mechanism. J Cell Biol 2006; 176(6):913–923. 29. Zong WX, Ditsworth D, Bauer DE, Wang ZQ, Thompson CB. Alkylating DNA damage stimulates a regulated form of necrotic cell death. Genes Dev 2004; 18:1272–1282. 30. Du X, Matsumura T, Edelstein D, et al. Inhibition of GAPDH activity by poly (ADP-ribose) polymerase activates three major pathways of hyperglycemic damage in endothelial cells. J Clin Invest 2003; 112: 1049–1057. 31. Kubasiak LA, Hernandez OM, Bishopric NH, Webster KA. Hypoxia and acidosis activate cardiac myocyte death through the Bcl-2 family protein BNIP3. Proc Natl Acad Sci USA 2002; 99(20):12825–12830. 32. Pinedo CM, Ruiz CR, Almodovar CR, Palacios C, Rivas AL. Inhibition of glucose metabolism sensitizes tumor cells to death receptor triggered apoptosis through enhancement of death inducing signaling complex formation and apical procaspase 8 processing. J Biol Chem 2003; 278(15):12759–12768. 33. Ainscow EK, Brand MD. Internal regulation of ATP turnover, glycolysis and oxidative phosphorylation in rat hepatocytes. Eur J Biochem 1999; 266:737–749. 34. Gatenby RA, Gillies RJ. Why do cancers have high aerobic glycolysis? Cancer 2004; 4: 891–899. 35. Mertz RJ, Worley JF, Spencer B, Johnson JH, Dukes ID. Activation of stimulur-secretion coupling in pancreatic beta cells by specific products of glucose metabolism. J Biol Chem 1996; 271(9):4838–4845. 36. Ben-Shlomo I, Kol S, Roeder LM, et al. Interleukin (IL)-1beta increases glucose uptake and induces glycolysis in aerobically cultured rat ovarian cells: evidence that IL-1beta may mediate the gonadotropin induced midcycle metabolic shift. Endocrinol 1997; 138(7):2680–2688.

Mitochondrial DNA Mutations in Tumors Anna Czarnecka and Ewa Bartnik

Abstract Mitochondria are subcellular organelles with the most well-known and best-characterized function of adenosine triphosphate (ATP) production through oxidative phosphorylation (OXPHOS). Mitochondria play an important role in apoptosis, a fundamental biological process by which cells die in a programmed manner and are the strategic point in the cell death cascade. Alterations in respiratory activity and mitochondrial DNA (mtDNA) abnormalities appear to be a general feature of malignant cells. The presence of mtDNA mutations has been reported in most types of cancer. However, the functional consequences and clinical relevance of these mutations are not clear. Keywords Mitochondria · Mutations · Cancer · Tumor · DNA

1 Introduction Mitochondria are eukaryotic organelles involved in many metabolic pathways and have a principal function of generating most of the cellular ATP through oxidative phosphorylation (OXPHOS). Mitochondria are semiautonomous organelles that not only perform essential functions in cellular metabolism but also play an important role in the regulation of cell death, signaling, and free radical generation [1]. Mitochondria possess a double-membrane structure and their own genome along with their own transcription, translation, and protein assembly machinery [2]. Nevertheless, mitochondrial functions involve an interplay between the mitochondrial and nuclear genomes [3]. Mitochondria are called “cellular powerhouses”; the primary function of these organelles is to produce ATP in the process of oxidative phosphorylation. At the same time, many housekeeping metabolic reactions take place in this E. Bartnik (B) Department of Genetics and Biotechnology, Faculty of Biology, University of Warsaw, and Institute of Biochemistry and Biophysics, Polish Academy of sciences, Pawinskiego 5a, 02106 Warsaw, Poland e-mail: [email protected] S.P. Apte, R. Sarangarajan (eds.), Cellular Respiration and Carcinogenesis, DOI 10.1007/978-1-59745-435-3 9, C Humana Press, a part of Springer Science+Business Media, LLC 2009

119

120

A. Czarnecka, E. Bartnik

cell compartment including production of intermediates of carbohydrate, nucleotide, fatty acid, and amino acid metabolism. Mitochondria are indispensable for the cell because of their key role in homeostasis, which is probably the reason for their maintenance within cells lacking mtDNA (called ␳o ) [4]. Processes believed to be affected by mitochondrial function include climatic adaptation, aging, longevity, degenerative disease, and cancer [3]. Numerous diseases are known to be caused by mutations in the mtDNA. A breakthrough in mitochondrial pathophysiology occurred in 1988 when Wallace et al. [5] found the first point mutation in the mtDNA of patients with Leber’s hereditary optic neuropathy. To date, the etiology and pathophysiology of mitochondrial diseases has remained perplexing. The essential role of mitochondrial oxidative phosphorylation in cellular energy production, the generation of reactive oxygen species, and the initiation of apoptosis has suggested a number of novel mechanisms for mitochondrial pathology. Mutations in mtDNA appear to be the primary factor responsible for mitochondria-related diseases and syndromes. The number of known point mutations, deletions, and duplications of mtDNA is more than several hundred and new mutations are annotated every month, and the number of mtDNA polymorphisms has exceeded 1000 [6]. The recent resurgence of interest in the study of mitochondria has been fueled in large part by the recognition that genetic and/or metabolic alterations in this organelle are causative or contributing factors in a variety of human diseases including cancer. The role for mitochondria in tumorigenesis was hypothesized when tumor cells were found to have an impaired respiratory system and high glycolytic activity [7]. Recent findings elucidating the role of mitochondria in apoptosis and the high incidence of mtDNA mutations in cancer samples further support this hypothesis [3]. Defects in mitochondrial function have long been suspected of contributing to the development and progression of cancer. In the early 1920s, Warburg pioneered the research on mitochondrial respiration alterations in the context of cancer and proposed a mechanism to explain how they evolve during cell transformation. He hypothesized that a key event in carcinogenesis involved the development of an injury to the respiratory machinery, resulting in compensatory increases in glycolytic ATP production. Eventually, malignant cells would satisfy their energy needs by producing a large portion of their ATP through glycolysis rather than through oxidative phosphorylation [7]. Because of the inherent inefficiency of glycolytic ATP generation, cancer cells represent a unique metabolic state and require high consumption of glucose to fulfill cellular energy requirements [4]. Hypoxia itself has been studied as a carcinogenic factor. It has been proved that aerobic glycolysis is highly active in malignant cancers. In benign tumors, the rate of aerobic glycolysis is only one third of the malignant cancer aerobic glycolytic activity. High aerobic glycolysis activity must correlate with glucose absorption by the cell. In vivo measurements have been performed, and high glucose uptake by cancerous tissue in comparison with normal tissues was reported for gliomas, meningiomas, and sarcomas [7]. To date, mitochondrial failure has been reported at all levels of their structure and function, including abnormal ultrastructure and metabolism deregulation. First

Mitochondrial DNA Mutations in Tumors

121

of all, aerobic energy metabolism alterations have been assigned to disruption of the nuclear encoded enzymes fumarate hydratase and succinate dehydrogenase expression (complex II of the respiratory chain). It is worth noting that mutations in these two genes can also behave as classic tumor suppressors linking mitochondrial physiology with cancer [8]. In addition, subunits of complexes of the respiratory chain are also encoded by the mitochondrial genome (mtDNA) [3]. Accordingly, mtDNA alterations are frequently observed in tumor cells [4, 9]. Mutations in the mtDNA may result in altered structure of mitochondrial proteins and thus disrupt OXPHOS, which might shift metabolism toward anaerobic respiration. However, evidence for direct linkage of respiratory deficiency in a specific tumor type with a specific mtDNA mutation is still missing [8].

2 The Mitochondrial Genome In 1981, the complete sequence of human mtDNA was determined by Sanger and co-workers [10]. All genes have been identified and characterized [3]. Human mtDNA is a double-stranded, closed-circular molecule of approximately 16.6 kb, which corresponds with a molecular weight of about 10 million Da, and in most cells it represents only about 0.5% to 1% of total DNA content [11]. The normal mtDNA state is thought to be a supercoiled structure, and it is poorly associated with proteins in comparison with nuclear DNA. The two strands of mtDNA can be distinguished due to their different G content and can be separated by density in denaturing gradients into a heavy or H-strand and a light or L-strand [2]. Each mitochondrion contains 2 to 10 copies of its genome, and a mammalian cell typically contains 200 to 2000 mitochondria. As a consequence of mtDNA multiplication in the cell, mitochondrial genomes can tolerate very high levels (up to 90%) of damaged DNA through complementation by the remaining wild type [2]. Mitochondrial DNA encodes a small, but essential number of polypeptides of the OXPHOS system. The coding sequences for two ribosomal RNAs (rRNAs), 22 transfer RNAs (tRNAs) and 13 polypeptides, which are subunits of four OXPHOS complexes, are contiguous and lack introns (see Fig. 1 and Table 1). The tRNAs are regularly interspersed between the rRNA and protein-coding genes, playing a crucial role in RNA maturation from the polycistronic transcripts. A single major 1.1-kblong noncoding region, called the displacement loop (D-loop), contains the main regulatory sequences for transcription and replication initiation [2]. The D-loop is a triple-stranded structure in which a nascent H-strand DNA segment of 500 to 700 nucleotides remains annealed to the parental L-strand. This is the region that is most variable in sequence. However, the whole mtDNA is much more variable in sequence than nuclear genome (nDNA), as it is more vulnerable to mutations due to the lack of histone protection, limited repair capacity, and close proximity to the electron transport chain, which constantly generates superoxide radicals. A higher rate of mutation is accompanied by higher ratio of nonsynonymous to synonymous substitution in mtDNA than in nuclear genes [12]. As mtDNA lacks introns, most mutations occur in the coding sequences and are thus likely to be of biological

122


Fig. 1 Human mitochondrial DNA molecule. ATP6/8, ATP synthase F0 subunit 6/8; COX I, cytochrome c oxidase subunit I; COX III, cytochrome c oxidase subunit III; CYT B, cytochrome b; HSP, heavy-strand promoter; LSP, light-strand promoter; ND1, NADH dehydrogenase subunit 1; ND2, NADH dehydrogenase subunit 2; ND3, NADH dehydrogenase subunit 3; ND4L, NADH dehydrogenase subunit 4L; ND4, NADH dehydrogenase subunit 4; ND5, NADH dehydrogenase subunit 5; ND6, NADH dehydrogenase subunit 6; OH, H-strand origin; OL, L-strand origin; A-W, tRNAs.

consequence. The mitochondrial genome has other characteristics that distinguish mtDNA from nDNA. These include not only a higher rate of mutation but also the use of a divergent genetic code and transmission by maternal inheritance [3]. It has been proposed that dysregulation of mtDNA genes may contribute to bioenergetic imbalances as it is known that some rapidly growing tumors display alterations of oxidative metabolism and high rates of aerobic glycolysis. Further experiments have shown that mutations in D-loop, OXPHOS genes, and mitochondrial tRNA are characteristic for cancer cells. Moreover, mutations and deletions in mtDNA that have been observed in various solid tumors and hematologic malignancies are associated with abnormal expression of mtDNA-encoded proteins [11]. Mutations in a single mtDNA genome are silent, but as soon as the proportion of mutant mtDNA exceeds a critical threshold concentration, a defect of OXPHOS, integrity of inner mitochondrial membrane, electron leakage, and increased reactive oxygen species (ROS) production result [3]. Clonal expansion of a single somatic mtDNA mutation has substantial implications for a cell, as mtDNA mutations are at the basis of a number of human pathologies, not only cancer. This state of art is a derivative of the fact that a combination of different respiratory chain complexes (I,


123

Table 1 Mitochondrial Genes Encoded in the Mitochondrial Genome (mtDNA) and Nuclear Genome (nDNA) Mitochondrial genes Mitochondrial complex

Genes encoded in mtDNA

Complex I NADH 7 dehydrogenase Complex II FAD 0 dehydrogenase Complex III cytochrome 1 b-c1 Complex IV 3 cytochrome oxidase Complex V ATP 2 synthase Total 13

mtDNA genes

Total number of mtDNA and nDNA genes

ND1, ND2, ND3, ND4, ND4L, ND5, ND6 —

43

Cytb

11

COX1, COX2, COX3

13

ATP6, ATP8

16

4

87

III, IV, and V) defects might have substantial physiologic effects, for example may lead to a dysregulation of oxidative phosphorylation, which in turn allows robust production of the carcinogenic ROS. Growing molecular evidence suggests that cancer cells exhibit increased intrinsic oxidative stress, due in part to oncogenic stimulation, increased metabolic activity, and mitochondrial malfunction, which turns on the vicious circle of oncogenesis [13]. Given the critical role of mitochondria in apoptosis, it is conceivable that mutations in mtDNA in cancer cells could also significantly affect the cellular apoptotic response to anticancer agents and promote multidrug resistance [14]. At the same time, other studies have shown that mouse cells carrying nuclei from tumors expressed tumorigenicity, whereas cell with noncancerous nuclei and mitochondria from cancer cells did not [15]. Thus at least for some cancer cells, the mitochondria do not seem to have a profound effect on the cancerous state of the cell. On the other hand, this does not exclude their role in tumor growth and progression, and this has been experimentally demonstrated [16, 17]. Interest in potential therapeutic agents that could act on the mitochondria in tumor cells has fueled an explosion in the research on mitochondria and mtDNA in recent years, and although the role of mitochondria in tumorigenesis remains unclear, numerous publications emphasize the impact of mtDNA mutations in the process of carcinogenesis [3, 4, 9]. Reports on mtDNA mutations in cancer published to date present an unclear view of overwhelming mutation/polymorphism data with doubtful clinical correlations. Moreover, much of the literature demonstrating mutations—whether heteroplasmic or homoplasmic—has been burdened with technical and conceptual errors that many times reduce the finding of mtDNA mutations in tumors to the level of technical artifacts [18, 19]. Thus far, a limited number of mtDNA mutations are known that can be classified as pathologic markers. These include a 4977-bp deletion in Hürthle cell thyroid carcinoma;

124


mutations of complex I and IV in thyroid carcinoma arising from A cells; and a 264-bp deletion (3323–3588) in renal cell carcinoma. Further analysis revealed that accumulation of mtDNA mutations is correlated with tumor aggressiveness and multidrug resistance or with poor prognosis and acute symptoms and signs [1, 4, 11]. The increasing ease with which the mitochondrial genome can be analyzed has helped to recognize mtDNA mutations as a frequent event in the carcinogenic process. However, it is difficult to estimate the true prevalence of mtDNA-related cancer owing to many clinical presentations and the involvement of numerous causative mutations.

3 Mitochondrial Genome Instability in Cancer Mitochondrial genome instability (mtGI) in cancer can be defined as the occurrence of mutations in tumor cells that are not found in the normal cells of the same individual [20]. The mtDNA mutation rate is 100-fold higher than in the nuclear genome [4, 11, 20], and several reasons are invoked to explain this fact, which include (a) mtDNA is less protected by proteins, (b) it is physically associated with the mitochondrial inner membrane where damaging ROS are generated, and (c) it appears to have less efficient repair mechanisms than does the nucleus, as subunit ␣ of mitochondrial polymerase ␥ has only limited 3 –5 exonuclease activity [4, 20]. Also, malfunction of mismatch repair (MMR) and slipped strand mispairing (SSM) should be considered as mtGI factors [20]. Mitochondrial DNA mutations arise not only as a consequence of carcinogenetic stress but also as a result of polymerase gamma errors (T to C or G to A transitions) and are accumulated in daughter cells as consequence of genetic drift [21]. One of the intrinsic factors responsible for mtDNA mutation generation are ROS produced by the OXPHOS system located in the inner membrane. Because the mitochondrial respiratory chain is a major source of ROS generation in the cells and the naked mtDNA molecule is in close proximity to the source of ROS, the vulnerability of the mtDNA to ROS-mediated damage appears to be a mechanism to amplify ROS in already stressed cancer cells. Damaged, misfolded, or unfolded OXPHOS proteins in turn promote even more pronounced ROS production, which drives a vicious circle of cell transformation [3, 4, 11, 20]. The proposal that mtDNA mutations and respiratory dysfunction may be linked directly to carcinogenesis via apoptotic or ROS-mediated pathways is challenging but urgently needs experimental proof in cancer cells, preferably in specimens from cancer patients. To date, ROS overproduction has been demonstrated only in cells harboring mtDNA mutations responsible for mitochondrial diseases, the most common being NARP (neurogenic muscle weakness, ataxia and retinitis pigmentosa), MELAS (mitochondrial encephalomyopathy lactic acidosis, and strokelike episodes), MERRF (myoclonic epilepsy and ragged-red fibers), LHON (Leber hereditary optic neuropathy), and KSS (KearnsSayre syndrome) [22]. Moreover, it is worth emphasizing that some somatic mutations found in cancer cells are sometimes identical to those found in mitochondrial disease [23–25], which


125

offers the possibility of extrapolating mitochondrial disease defects into cancer cells. Nevertheless, further in vivo experiments are indispensable to validate the proposed hypothesis.

4 Homoplasmy and Heteroplasmy in Cancer Cells Cells are polyploid with respect to mtDNA: most mammalian cells contain hundreds of mitochondria and subsequently mtDNA copies. If in a given individual, all mtDNA copies are identical, a condition known as homoplasmy arises, but if mutations occur, are amplified and coexist with wild-type mtDNA, this is designated heteroplasmy. At cell division, mitochondria and their genomes are randomly distributed to daughter cells, and hence, starting from a heteroplasmic situation, homoplasmy and different levels of heteroplasmy may be found in daughter cell lineages [2, 20, 26]. In cancer cells, mutated mtDNA can be found as homo- or heteroplasmic [4], but many reports show that mtDNA mutations in cancer cells are homoplasmic. This raises the question of mechanisms responsible for normal to mutated mtDNA shift in the process of carcinogenesis [20, 26]. At least two scenarios are probable that drive this change. First of all, mutant mitochondria that lose a certain mtDNA fragment by deletion can be considered to replicate more rapidly than normal ones, resulting in an advantage in intracellular mitochondrial competition (replicative advantage). If the competition is intense (high rate of proliferation), heteroplasmic cells possessing both types of mitochondria give rise to homoplasmic daughter cells including mutant mitochondria only, with high probability. According to mathematical models, the rate of switching from wild type to mutated mitochondrial DNA is affected by factors including the intensity of intracellular competition and the effective population size [27]. Researchers in several laboratories who reported a high frequency of homoplasmic mitochondrial DNA mutations in human tumors proposed not only replicative advantage for mutated mtDNA copies [28] but also an advantage in growth and/or tumorigenic properties for a cell containing certain mtDNA mutations [24]. Polyak et al. [28] have proposed that mitochondrial genomes replicate more rapidly to balance OXPHOS failure and reduced ATP production. On the other hand, it cannot be excluded that homoplasmy arises entirely by chance in tumor progenitor cells, without any physiologic advantage or tumorigenic requirement. Through extensive computer modeling, it has been demonstrated that there is sufficient opportunity for a tumor progenitor cell to achieve homoplasmy through unbiased mtDNA replication and sorting during cell division [29]. Computer modeling has shown that random drift is also a sufficient mechanism to explain the homoplasmic nature of cancer cells [4]. Relaxed replication phenomenon analysis in normal cells has proved that replicative or metabolic advantage is not indispensable for homoplasmy to arise [4, 24, 26]. Although a role for other mechanisms is not excluded, random processes are sufficient to explain the incidence of homoplasmic mtDNA mutations in human tumors. Depending on how the mutant copies are distributed between the stem and transition cells, the mutants are depleted or enriched, and the number of mutants in the lineage fluctuates as a random walk.

126


On average, only approximately 70 generations are required for a mutation that is destined to become homoplasmic. The number of 70 generations is small compared with the number of cell divisions that a tumor progenitor cell is expected to undergo [28]. To summarize, whereas it cannot be excluded that selection occurs in a subset of tumors, there is no need to invoke a physiologic advantage or a role for mitochondrial mutations in tumorigenesis to explain the existence of homoplasmy or its observed frequency [29]. However, a mathematical model cannot prove or disprove that a phenomenon is occurring in growing cells, as it is based on assumptions and does not constitute experimental proof.

5 Conclusion Since the initial publications by Warburg [7] more than 50 years ago, a number of cancer-related mitochondrial defects have been identified and described in the literature. Growing evidence suggests that cancer cells exhibit increased intrinsic mitochondrial stress, due in part to oncogenic stimulation, increased metabolic activity, and mitochondria-nucleus signaling malfunction [30]. Despite the increased identification of signatures of mtDNA damage in transformed cells, the phenotypic effects of these genetic changes remain to be established. While there are many reports of these phenomena, the mechanisms responsible for the initiation and evolution of mtDNA mutations and their roles in the development of cancer, drug resistance, and disease progression still remain to be elucidated [9]. Research into the identification of altered expression patterns of mitochondrial proteins in cancer cells has been made possible by the relatively recent development of mitochondrial functional proteomics. The potential of this field may be realized in the identification of new markers and risk assessment as well as therapeutic targets [14, 30]. Despite the difficulties with mitochondrial proteomics, it is likely that the combination of the mitochondrial genetic and the proteomic approaches will provide an effective weapon in the fight against cancer. It is hoped that this strategy will provide specific genetic markers and protein profiles that will provide early detection, risk assessment, and new targets for treatment. The recently initiated generation of mouse models for mtDNA-linked diseases may help to solve some of those open questions. There are more than 400 publications in Medline reporting mitochondrial DNA mutations in cancer. Recently, a whole issue of Oncogene was devoted to this problem. What, if any, is the take-home message about the role of mtDNA mutations in cancer? First of all, the literature is not entirely reliable. There are reports of data analysis that essentially denies any role for mtDNA mutations in tumors [18]. There are problems with some of the mtDNA mutation papers, one of them being that mtDNA haplogroups differ by specific polymorphisms throughout the mitochondrial genome, and are the results of mutations in distinct maternal lineages, as mitochondria are inherited exclusively in the maternal line [3, 31]. Thus the comparison of any mitochondrial sequence with the Cambridge reference sequence (CRS) is


127

simply a way of determining the differences between one sequence and another— and the results may have nothing to do with tumors and all to do with the fact that the sample and the CRS may belong to different haplogroups of mtDNA [6]. Moreover, in most cases the analyzed groups of samples are small, for all sorts of reasons, and conclusions cannot be drawn as to the “most common mitochondrial mutation” in a particular type of tumor on the basis of 10 or even 20 or 30 samples as the population of analyzed samples is not significant for statistical analysis. Thirdly, data get misquoted; for example, a paper quotes another one out of context as to the fact that a certain mitochondrial variant predisposes to a certain cancer even though the original authors did not claim this and had no reason to do so. And finally, the easily accessible mitochondrial databases are rarely used to see if something that is suspected of being important in cancer is not simply a common polymorphism found in many populations. The use of mtDBase [32] and MITOMAP [6] should be obligatory for any papers dealing with mtDNA variants. During the past 2 years, there have been increasing numbers of complete genome sequences compared with appropriate control tissues and not with the CRS. Brandon et al. [33] have performed a comprehensive analysis of the literature and showed that most of the encountered mutations are population polymorphisms, and they have put forth the attractive hypothesis that what happens in tumors is a phenomenon of selection that also took place during human adaptation to various environments [3]—though here it is the tumor cells and not the human beings who have adapted to special circumstances. Thus most mutations in tumors would not be dramatic changes profoundly affecting respiration, but small changes facilitating growth under the conditions of relative hypoxia and crowding [33]. Some mutations—fairly rare—are mutations that profoundly affect cellular respiration; not many of these have been observed. Petros et al. [16] and Shidara et al. [17] have shown that tumors containing mitochondrial DNA with the pathogenic NARP mutation at position A8993G in ATP6 grow faster and are more resistant to apoptosis than are tumors with the same type of cells but with wild-type mtDNA. This was accepted as evidence that the severe mitochondrial mutations in tumors may play a role of facilitating growth and decreasing susceptibility to apoptosis under the exacting conditions of tumor growth. However, Salas et al. [18] have suggested that as this particular mutation has not been found in tumors, this experiment does not prove anything. On the other hand, the data presented there are quite convincing as the growth/apoptosis perturbations can be abrogated by expression of an unmutated gene (expressed from a gene introduced into the nucleus, as it is not yet possible to introduce genes into human mitochondria) [17]. Brandon et al. [33] have suggested that such severe mutations arise in early stages of tumor development and are successively lost. Recently, Gasparre et al. [34] have shown that in oncocytic tumor, such mutations (defined as disruptive; i.e., nonsense or frameshift) occur in about one quarter of the tumors, and they all occur in complex I genes. However— and this would be in agreement with Brandon et al. [33]—cell cultures grown from these tumors lose the severe mutations [34]. Zhou et al. [35] have demonstrated for the first time that a ND2 (again complex I) mutation found in a tumor (human squamous cancer of the head and neck)

128


can increase growth of cells in culture, thus showing that growth can depend on the expression of a mutated mitochondrial gene. In this case, they also used a gene introduced into the nucleus. Moreover, they showed that ROS production was increased [35]. Thus there certainly are mutations in mtDNA in cancer cells and they do affect the growth properties of these cells. It should be borne in mind, however, that the mutations that are observed in most tumors are likely to be ones that have only slight adaptive effects on their growth, and the dramatic mutations are rare. We certainly have spent a lot of time looking for them and essentially found only one [23]. Another problem important both for scientists and for epidemiologists is the question whether some of us are more predisposed to certain types of tumors. Mitochondrial DNAs form haplogroups, and the differences between them believed until recently to be completely neutral and insignificant may contribute to the frequency of some diseases. Again there is a problem with the data. Statistics is a doubleedged sword; if the sample sizes are too small, many associations are uncertain. Samuels et al. [31] have recently shown that large cohorts are required for studies of associations of diseases with mitochondrial haplogroups. Studies with 500 controls and 500 cases would not always reliably show a relative risk of about 1.75 for a common haplogroup; larger sizes would be required for less common ones [31]. Rarely are the analyzed groups even one tenth of this size. Thus far, two studies of cancer—haplogroup associations seem quite robust—a study by Booker et al. [36] showing increased relative risk for haplogroup U for prostate cancer and renal carcinoma (though this has again been criticized by Salas et al. [18]), and a study of the G10398A polymorphism in ND3 in haplogroup N has shown that it is associated with an increased breast cancer risk [37]. Increased relative risks for certain haplogroups and certain types of tumors might be of interest to health care providers as predictors of cancer development. What could the rationale be for this? Wallace [3] has postulated that the differences in mitochondrial DNA sequence transferred to the function of the electron transport chain in mitochondria could give important differences in, for example, ROS production, which could affect how susceptible the cells would be to damage. The main problem with this hypothesis is that no differences have thus far been shown between various normal types of cells in ROS production for humans, though a number of pathogenic mutations have been shown to affect ROS production. Recently, Moreno-Loshuertos et al. [38] have shown differences between various mouse mitochondrial haplotypes both in respiration and in free radical production, though this has aroused some discussion [39]. As the techniques of mutation analysis improve and more data become available, the role of mitochondrial mutations and haplogroup associations should be elucidated in the near future. However, they are unlikely to be specific markers for any type of tumor, and there will probably not be very strong associations between haplogroups and tumor occurrence. Finally, the importance of the highest standards in mtDNA analysis to produce reliable results in the many investigations in carcinogenesis should be emphasized [18, 19].


129

Acknowledgment This work was partly supported by an intramural grant from The Faculty of Biology, University of Warsaw and grant NN 401 2327 33 from the Ministry of Science and Higher Education.

References 1. Amuthan G, Biswas G, Zhang SY, Klein-Szanto A, Vijayasarathy C, Avadhani NG. Mitochondria-to-nucleus stress signaling induces phenotypic changes, tumor progression and cell invasion. EMBO J 2000; 20:1910–1912. 2. Fernandez-Silva P, Enriquez JA, Montoya J. Replication and transcription of mammalian mitochondrial DNA. Exp Physiol 2003; 88:41–56. 3. Wallace DC. A mitochondrial paradigm of metabolic and degenerative diseases, aging, and cancer: a dawn for evolutionary medicine. Annu Rev Genet 2005; 39:359–407. 4. Carew JS, Huang P. Mitochondrial defects in cancer. Mol Cancer 2002; 1:1–12. 5. Wallace DC, Singh G, Lott MT, et al. Mitochondrial DNA mutation associated with Leber s hereditary optic neuropathy. Science 1988; 242:1427–1430. 6. Ruiz-Pesini E, Lott MT, et al. An enhanced MITOMAP with a global mtDNA mutational phylogeny. Nucleic Acids Res 2007; 35:D823–D828. 7. Warburg O. On the origin of cancer cells. Science 1956; 123:309–314. 8. Gottlieb E, Tomlinson JP. Mitochondrial tumor suppressors: a genetic and biochemical update. Nat Rev Cancer 2005; 5:857–866. 9. Czarnecka AM, Golik P, Bartnik E. Mitochondrial DNA mutations in human neoplasia. J Appl Genet 2006; 47:67–78. 10. Anderson S, Bankier AT, Barrell BG, et al. Sequence and organization of the human mitochondrial genome. Nature 1981; 290:457–465. 11. Penta JS, Johnson FM, Wachsman JT, Copeland WC. Mitochondrial DNA in human malignancy. Mutat Res 2001; 488:119–133. 12. Lynch M. Mutation accumulation in transfer RNAs: molecular evidence for Muller’s ratchet in mitochondrial genomes. Mol Biol Evol 1996; 13:209–220. 13. Birch-Machin M. Using mitochondrial DNA as a biosensor of early cancer development. Br J Cancer 2005; 93:271–272. 14. Don AS, Hogg PJ. Mitochondria as cancer drug targets. Trends Mol Med 2004; 10:372–378. 15. Akimoto M, Niikura M, Ichikawa M, et al. Nuclear DNA but not mtDNA controls tumor phenotypes in mouse cells. Biochem Biophys Res Commun 2005; 327:1028–1035. 16. Petros JA, Baumann AK, Ruiz-Pesini E, et al. MtDNA mutations increase tumorigenicity in prostate cancer. Proc Natl Acad Sci USA 2005; 102:719–724. 17. Shidara Y, Yamagata K, Kanamori T, et al. Positive contribution of pathogenic mutations in the mitochondrial genome to the promotion of cancer by prevention from apoptosis. Cancer Res 2005; 65:1655–1663. 18. Salas A, Yao YG, Macaulay V, Vega A, Carracedo A, Bandelt HJ. A critical reassessment of the role of mitochondria in tumorigenesis. PLoS Med 2005 2:e296. 19. Zanssen S, Schon EA. Mitochondrial DNA mutations in cancer. PLoS Med 2005; 2: 1082–1084. 20. Bianchi NO, Bianchi MS, Richard SM. Mitochondrial genome instability in human cancer. Rev Mut Res 2001; 488:9–23. 21. Brown DT, Samuels DC, Michael EM, Turnbull DM, Chinnery PF. Random genetic drift determines the level of mutant mtDNA in human primary oocytes. Am J Hum Genet 2001; 68:533–536. 22. Kirkinezos IG, Moraes CT. Reactive oxygen species and mitochondrial diseases. Semin Cell Dev Biol 2001; 12:449–457. 23. Lorenc A, Bryk J, Golik P,et al. Homoplasmic MELAS A3243G mtDNA mutation in colon cancer sample. Mitochondrion 2003; 3:119–124.

130


24. Chinnery PF, Samuels DC, Elson J, Turnbull DM. Accumulation of mitochondrial DNA mutations in ageing, cancer, and mitochondrial disease: is there a common mechanism? Lancet 2002; 360:1323–1325. 25. Meierhofer D, Mayr JA, Fink K, Schmeller N, Kofler B, Sperl W. Mitochondrial DNA mutations in renal cell carcinomas revealed no general impact on energy metabolism. Br J Cancer 2006; 94:268–274. 26. Augenlicht LH, Heerdt BG. Mitochondria: integrators in tumorigenesis? Nat Genet 2001; 28:104–105. 27. Yamauchi A. Rate of gene transfer from mitochondria to nucleus: effects of cytoplasmic inheritance system and intensity of intracellular competition. Genetics 2005; 171:1387–1396. 28. Polyak K, Li Y, Zhu H, et al. Somatic mutations of the mitochondrial genome in human colorectal tumors. Nat Genet 1998; 20:291–293. 29. Coller HA, Khrapko K, Bodyak ND, Nekhaeva E, Herrero-Jimenez P, Thilly WG. High frequency of homoplasmic mitochondrial DNA mutations in human tumors can be explained without selection. Nat Genet 2001; 28:147–150. 30. Pelicano H, Carney D, Huang P. ROS stress in cancer cells and therapeutic implications. Drug Resist Updat 2004; 7:97–110. 31. Samuels DC, Carothers AD, Horton R, Chinnery PF. The power to detect disease associations with mitochondrial DNA haplogroups. Am J Hum Genet 2006; 78:713–720. 32. Ingman M, Gyllensten U. mtDB: human mitochondrial genome database, a resource for population genetics and medical sciences. Nucl Acids Res 2006; 34:D749–D751. 33. Brandon M, Baldi P, Wallace DC. Mitochondrial mutations in cancer. Oncogene 2006; 25:4647–4662. 34. Gasparre G, Porcelli AM, Bonora E, et al. Disruptive mitochondrial DNA mutations in complex I subunits are markers of oncocytic phenotype in thyroid tumors. Proc Natl Acad Sci USA 2007; 104:9001–9006. 35. Zhou S, Kachhap S, Sun W, et al. Frequency and phenotypic implications of mitochondrial DNA mutations in human squamous cell cancers of the head and neck. Proc Natl Acad Sci USA 2007; 104:7540–7546. 36. Booker LM, Habermacher GM, Jessie BC, et al. North American white mitochondrial haplogroups in prostate and renal cancer. J Urol 2006; 175:468–472; discussion 472–473. 37. Darvishi K, Sharma S, Bhat AK, Rai E, Bamezai RN. Mitochondrial DNA G10398A polymorphism imparts maternal Haplogroup N a risk for breast and esophageal cancer. Cancer Lett 2007; 249:249–255. 38. Moreno-Loshuertos R, Acin-Perez R, Fernandez-Silva P, et al. Differences in reactive oxygen species production explain the phenotypes associated with common mouse mitochondrial DNA variants. Nat Genet 2006; 38:1261–1268. 39. Battersby BJ, Shoubridge E. Reactive oxygen species and the segregation of mtDNA sequence variants. Nat Genet 2007;39:571–572.

Cellular Respiration and Tumor Suppressor Genes Luis F. Gonzalez-Cuyar, Fabio Tavora, Iusta Caminha, George Perry, Mark A. Smith, and Rudy J. Castellani

Abstract More than 70 years have passed since Dr. Otto Warburg first documented that cancer cells relied primarily on aerobic glycolysis. However, it was not until the late-1980s and 1990s when cellular respiration, cellular oxygen sensors, and hypoxia were convincingly related to tumorigenesis and tumor progression. With the discovery of hypoxia inducible factor (HIF-1) and its target genes, the relationship that was once weak has become very strong. The expression of the HIF-1 molecule has been proved to occur by hypoxic stress and confers an adaptation advantage, upregulating genes involved in angiogenesis and glycolysis among others. It has been reported that HIF-1 is under the control of tumor suppressor genes such as the von Hippel–Lindau (pVHL) protein, which, under normoxia, ubiquitinates HIF-1 for proteosomal degradation. Additionally, the tumor suppressor gene p53 has been reported to be stabilized by HIF-1 and inhibits Mdm2-dependent degradation. Recently, it was reported that synthesis of cytochrome c oxidase, a regulator of the cytochrome c oxidase pathway, is a p53 target. Moreover, interactions of pVHL and p53 stabilize p53 and permits p53-mediated apoptosis. The aim of this chapter is to outline how the cellular microenvironment affects cellular respiration and perhaps induces tumorigenesis, as well as to discuss ways in which tumor suppressor genes are also involved. Keywords Cellular respiration · Hypoxia inducible factor · p53 · Tumor suppressor · Von Hippel–Lindau

1 Introduction For more than 70 years, the notion that cancer cells rely preferentially on glycolysis, a phenomenon identified as the Warburg effect [1], has been in existence, but only over the past 15 or so years have major advances been made in elucidating R.J. Castellani (B) Department of Pathology, School of Medicine, University of Maryland, 22 South Greene Street, Baltimore, MD, 21201 e-mail: [email protected] S.P. Apte, R. Sarangarajan (eds.), Cellular Respiration and Carcinogenesis, DOI 10.1007/978-1-59745-435-3 10, C Humana Press, a part of Springer Science+Business Media, LLC 2009

131

132

L.F. Gonzalez-Cuyar et al.

the role of oxygen and cellular respiration in neoplasia. A number of specific and well-characterized genes are now known to be involved, directly and indirectly, as well as a number of different well-characterized molecular cascades. Prominent among these is hypoxia inducible transcription factor (HIF-1), which upregulates the transcription of several target genes involved in cellular adaptation to the tumoral microenvironment (i.e., hypoglycemia, acidosis, and low oxygen tensions) and angiogenesis. Additionally, HIF-1 is under the posttranscriptional control of pVHL that ubiquitinates HIF-1 and tags it for proteosomal degradation. Two tumor suppressor genes have been identified in connection with changes in the respiratory metabolism in cancer: TP53 and the VHL tumor suppressor gene. TP53 is the most common mutated gene in human tumors. Whereas the majority of the studies on p53 in past decades have dealt with its role in cell division and death/apoptosis, some evidence now points toward its involvement also with cellular metabolism. Matoba et al. [2] recently found that the gene that encodes synthesis of cytochrome c oxidase 2 (SCO2) is upregulated in a p53-dependent manner. SCO2 may be a transcriptional target for p53 [3], and this relation may explain how inactivation of p53, as seen with somatic and familial mutations, may promote oncogenesis by decreasing cellular dependence on oxygen [2]. In fact, other studies have linked p53 in the control of cell glycolysis by influence of downstream molecules. For example, Kondoh et al. [4] showed that phosphoglycerate mutase, another glycolytic enzyme, is also controlled by p53 by posttranscriptional activity. Activation of the Akt kinase pathway is influenced by p53 and may also play a role in switching the cell metabolism from oxidative phosphorylation to glycolysis [5]. The pVHL protein is a positive regulator of p53, and germ-line or somatic mutations on the VHL gene are related to tumors such as renal cell carcinomas, hemangioblastomas, and pheochromocytomas. It has been postulated that VHL protein interacts with HIF-1 tagging it for proteosomal degradation [6]. In the absence of oxygen under normal circumstances or when pVHL is mutated, the HIF-1 transcription factor is inactivated with direct action upon molecules such as vascular endothelial growth factor (VEGF), culminating in angiogenesis and, thus, cancer progression [7]. In this chapter, we will discuss the role of tumor suppressor genes in the metabolism of cells, more specifically in the various components of cellular respiration, interactions with HIF-1, and how interactions of the gene products will promote oncogenesis, cell survival, and cancer progression through upregulation of transcription of genes involved in angiogenesis, resistance to apoptosis, and metastasis.

2 Otto Warburg and Cellular Respiration Otto Heinrich Warburg (1883–1970), a German biochemist, made major contributions to the study of respiration and neoplasia. He hypothesized early on that as cancer cells have an abnormal growth rate, the means by which the required energy is

Respiration and Tumor Suppressor Genes

133

obtained may be central to the pathogenesis of neoplasia [1, 8]. Results from earlier works demonstrating that rates of respiration of the sea urchin egg after fertilization increase up to sixfold motivated him to study cancer cells in a similar manner [1, 8]. As he measured the rates of respiration in transplantable cancer cells from the Flexner-Jobling rat carcinoma, he concluded that there was no increase in oxygen consumption and lactate was being produced by glycolysis, even in the presence of oxygen [1, 8]. Warburg concluded that contrary to cells in normal tissues, cancer cells could not suppress glycolysis in the presence of oxygen. Interestingly, he also described aerobic glycolysis in tissues exposed to ethylcarbilamine. Warburg further demonstrated that although there was no significant increment in levels of cellular respiration or ATP production of cancer cells, anaerobic glycolysis provided a major energy source. He hypothesized that when normal cells become cancerous that metabolic activity shifts in favor of glycolysis. This switch in his opinion was secondary to a “dedifferentiation” or “reorientation of gene expression” that diverted the baseline energy production to growth. After a paper by Goldblatt and Cameron [9] citing induction of fibrosarcoma from fibroblasts exposed to intermittent hypoxia, he demonstrated that embryonic cells exposed to hypoxia reverted to glycolysis and that normal metabolism, as in cancer, was irreversible, even in the presence of oxygen [1, 8]. Additionally he pointed out that subsequent animal inoculation led to the development of cancers. In 1956, Warburg described that irreversible damage to the cellular respiration can be achieved by low oxygen tensions, respiratory poisons such as arsenic, hydrogen sulfide, and urethane, and x-rays [1, 8]. In an excerpt from a 1967 paper entitled “The prime cause and prevention of cancer,” he simplified his conclusions: Cancer, above all other diseases, has countless secondary causes. Almost anything can cause cancer. But, even for cancer, there is only one prime cause. Summarized in a few words, the prime cause of cancer is the replacement of the respiration of oxygen in normal body cells by a fermentation of sugar.

Although his work was not studied for close to 70 years, Warburg’s conclusions on the rates of respiration are still valid with respect to tumorigenesis research and cancer therapeutics.

3 Molecular Homeostasis and Genetic Instability: Genetic Instability and Tumorigenesis There is a consensus pertaining to the biological definition of life, which is summarized in seven categories: homeostasis, organization, metabolism, growth, adaptation, response to stimuli, and reproduction. Mammals, as the highest life form, have an exquisite intrinsic balance of all of these functions at the molecular level [10–12]. Hanahan and Weinberg [13] described six essential characteristics dictating tumorigenesis and progression, which are somewhat analogous to the above:

134


self-sufficiency of growth signals, insensitivity to growth-inhibitory signals, evasion of programmed cell death, limitless replicative potential, sustained angiogenesis, and tissue invasion and metastases. Huang et al. [14] proposed that tumor development is composed of expansion and progression preceded by genetic adaptation (glycolysis, angiogenesis) and genetic alteration (malignant selection of characteristics), respectively. Decades of scientific research in the area of tumorigenesis and neoplasia have yielded an insurmountable amount of information as to how all of these essential characteristics of life are deranged in neoplasia [14–16]. Current thought suggests that cancer is a genetic disease that originates as a stepwise process in which normal cells go through several premalignant phenotypes by acquiring cumulative genetic alterations each conferring a more malignant phenotype leading eventually to progression and metastases [13]. Alterations in the genetic composition of cells either by nucleotide mutations or chromosomal aberrations are regarded as the initiator of neoplasia, while accumulation of these lead to progression, invasion, and metastases [14, 17]. The process by which clones are selected thus follows selection for mutations that confer a survival advantage in a hostile tumoral microenvironment [14, 17]. A central question with respect to oncogenesis is that of etiology. Although this remains an open question, current thought generally suggests that adverse conditions in the tumor microenvironment contribute to genetic instability and induction of mutagenesis by direct DNA damage and damage to DNA repair systems [18, 19]. According to some, intermittent hypoxia/reperfusion leads to generation of reactive oxygen species (ROS), leading in turn to the formation of 8-hydroxyguanine and thymine glycols, which lead to transversions [18–20]. Apart from such oxidative damage, aberrant DNA synthesis as seen by amplification as well as increased incidence of point mutations contribute to further progression [18, 19].

4 Hypoxia Hypoxia is among the most studied stressors in solid tumors, the central theme being the shift from aerobic respiration to glycolysis due to limited availability of oxygen and nutrients. Hypoxia occurs in tumor tissue that is between 100 and 200 ␮m from a functional blood supply [17, 21, 22]. Normal tissue displays an oxygen diffusion gradient of 400 ␮m from a blood supply, but significant hypoxia is seen in tumor cells even directly adjacent to a capillary with a mean oxygen tension of 2%, and cells located 200 ␮m from capillaries showed 0.02% [22, 23]. In addition, studies have suggested that hypoxia is an independent prognostic indicator. Reynolds et al. [20] demonstrated a fivefold increase in mutations in both tumors and cells cultured in hypoxia. Hypoxia has also been implicated as a stimulus for p53 activation and accumulation [24]. Hypoxia seems to not only induce the genetic instability that accounts for cumulative genetic aberration but also selects for malignant clones. It has been suggested that hypoxia-induced apoptosis can be mediated


135

by HIF-1 upregulation of p53-induced apoptosis. Therefore, to an extent, HIF-1 selects for tumor clones that have lost p53 and thus are less sensitive to hypoxia and hypoglycemia.

5 HRE and HIF-1 In 1992, Semenza and Wang [12] described a 50-nt region located 116-nt 3 of the polyadenylation site at the flanking sequence of the human erythropoietin (EPO) gene that was able to dramatically increase transcription in the presence of low oxygen tensions. The region known as the hypoxia responsive element (HRE) increased the transcription of EPO up to 7-fold in cells cultured at 2% O2 and up to 20-fold in cells cultured at 1% O2 [12]. Also it was described that mutations within the 4–12 nt and 19–23 nt would not produce the desired increase in transcription, thus it was hypothesized that these sites were crucial for binding of transcription factors [12]. Concordantly with manipulation of these mutation sites, researchers were able to discover HIF-1, which binds to the HRE at the 1–18 nt region, thus increasing the transcription of the desired gene under hypoxic but not normoxic conditions [12]. Because maximal levels of EPO mRNA after the induction by hypoxia takes about 2 to 4 hours, this suggested that maximization of gene products involving HRE in the face of hypoxia required de novo synthesis of proteins. Further molecular characterization of HIF-1 revealed that it is a heterodimeric transcription factor composed of two subunits, HIF-1␣ and HIF-1␤, which is also known as the aryl hydrocarbon nuclear translocator (ARNT) [11, 12, 25–28]. These proteins have a basic helix-loop-helix (bHLH) configuration with a Per, Ahr/ARNT domain (PAS) [11, 12, 25–28]. Whereas the former facilitates the nuclear dimerization of the two subunits, the latter is required for DNA binding [25, 26, 29, 30]. Thus far, three isoforms have been described for each subunit, HIF-1, 2, 3␣, and ARNT1, ARNT2, ARNT3, respectively [28, 29, 31–33]. It is important to note that the ARNT is constitutively transcribed, whereas the transcription of the ␣ subunit is stimulated by low oxygen tensions as well as several oncogenes [29, 31]. When dimerized, the subunits bind to the cis-acting HRE in the RCGTG conserved motif mediated by Ser22, Ala25, Arg30 [28, 29, 31–33]. Subsequently, increased transcription of the gene products involved in cellular adaptations to hypoxia such as VEGF and EPO are attained [12, 28, 29, 31–35]. The N-terminal region of the HIF-1␣ molecule is necessary for dimerization, whereas the C-terminus is required for nuclear localization, transactivation, and stabilization of the protein. The molecule also contains an oxygen dependent degradation (ODD) domain in the region of residues 401–603, which in addition contains two proline residues, 401 and 564, which are essential for degradation under normoxic conditions [31]. There are two areas of nuclear localization signals (NLS), one on the N-terminal (residues 17–33) and one on the C-terminal (residues 718–721) [31, 36]. Additionally, HIF-1␣ contains two transactivation domains

136


(TAD) at the C-terminal 531–575 and 786–826 [31, 33]. The area between these two nucleotides contains a domain that inhibits transactivation. Approximately 53% of malignant tumors express HIF-1 by immunohistochemistry including colon, breast, gliomas, prostate, renal cell carcinomas, and glioblastomas when compared with the respective normal tissue. Low-grade gliomas have little or no staining with HIF-1. Glioblastomas show intense staining, particularly in the pseudopalisading cells and in the neighboring areas of necrosis. Apparently, malignant gliomas take advantage of the HIF-1 activation by upregulation of platelet derived growth factor (PDGF) and VEGF, which appear directly proportional. To date, approximately 30 target genes that are transactivated by HIF-1 have been identified as downstream targets of HIF-1 including genes involved in glycolysis, glucose transport, angiogenesis, vascular remodeling and tone, erythropoiesis, and iron metabolism [10, 15].

6 Oxygen Sensors Under normoxic conditions, both subunits of HIF-1, HIF-1␣ and HIF-1␤, are transcribed into primary transcripts and exported to the cytoplasm where they are translated. Whereas constitutively expressed HIF-1␤ is stable under both normoxic and hypoxic conditions, HIF-1␣ is unstable under normoxic conditions. In the presence of oxygen, it is hydroxylated at two proline residues located on amino acid 401 and amino acid 564, and then tagged by pVHL, forming part of the ubiquinone-ligase complex [25, 29, 37–40], rendering it susceptible to proteosomal degradation. It is important to note that these proline hydroxylases that account for the protein’s posttranscriptional modification are oxygen dependent, therefore under hypoxic conditions the proline residues will not be hydroxylated so the protein will be translocated to the nucleus where it would dimerize with its counterpart [25, 29, 37–40]. After dimerization, the molecule interacts with several HREs located upstream of gene products involved in cellular adaptations to hypoxia, acidosis, and hypoglycemia. It has also been suggested that under hypoxic conditions, the pVHL–HIF-1 complex still forms but with ineffective degradation of HIF-1␣ [41]. Also, in the presence of iron or desferrioxamine, the pVHL–HIF-1 complex is not formed, thus demonstrating another way of HIF-1 stabilization [42]. However, it has been reported that this type of induction can be reversed by cyclohexamide, showing the need of de novo protein synthesis [43]. Additionally, in wild-type pVHL cell lines, investigators described that the HIF-1 DNA binding doublet contains pVHL, thus pointing out that there is pVHL–HIF-1 complex formation under normoxia [41].

7 Tumor Microenvironment The tumor microenvironment is fundamentally hostile for normal cellular metabolism, being characterized by hypoxia, hypoglycemia, and acidosis, thereby enhancing the selection process for malignant clones. In this regard, it is not


137

surprising that hypoxia has been established as an independent adverse prognostic factor for a variety of tumor types. Moreover, aberrant tumor vasculature may contribute to the low oxygen tension, and this lack of adequate perfusion, coupled with cellular overpopulation, leads to decreased cellular “resources” and increased need for the cells to adapt. As mentioned above, the hostile microenvironment at the epicenter of the tumor affects a shift from oxidative metabolism to glycolytic pathways. HREs have been identified in a variety of genes involved in this shift. These include EPO and VEGF, as well as several of the enzymes required for glycolysis such as glyceraldehyde phosphate dehydrogenase, enolase, phosphofructokinase, phosphoglycerate kinase, and pyruvate kinase, as well as other genes such as transferrin, transferrin receptor, IGF binding proteins, lactate dehydrogenase, and glucose transporters (GLT-1 and GLT-3). Some authors have hypothesized that products of glycolytic metabolism, such as pyruvate, can further stabilize the HIF-1␣ molecule by substituting 2-oxoglutarate from the proline hydroxylase and thus inhibit the hydroxylation of HIF-1␣ at the ODD domain [38]. Both lactate and carbonic anhydrases 9 and 12 may further foster an acidic environment contributing to tumorigenesis. These carbonic anhydrases appear to be also responsive to HIF-1–HRE induction, as they are not normally expressed in normoxic cells. Moreover, tumors that have a faulty glycolytic pathway, secondary to increased mutations, still have an acidic environment [44], which further suggests the role of these anhydrases in promoting a tumor microenvironment.

8 Cellular Respiration and Tumor Suppressor Genes 8.1 p53 p53 is a tumor suppressor gene located on 17p13.1 and encodes a 393-amino-acid DNA- binding protein that is mutated in more than 50% of human tumors and that is responsible for inhibiting uncontrolled cellular proliferation and invasion [2, 45]. p53 induces cell cycle arrest and apoptosis by enhanced transcription of target genes involved in every step of the apoptosis signaling pathway in response to DNA damage by various stimuli [3, 45–49]. p53 is activated by hypoxia, DNA damage, expression of certain oncogenes, and cytotoxic stimuli and, as with all tumor suppressor genes, inactivation is accomplished by loss in both alleles, usually by missense mutations [50, 51]. In the absence of cellular stress, the p53 gene product is ubiquitinated and tagged for proteosomal degradation by the ubiquitin ligase Mdm2 [24, 45]. Graber et al. [52] also described that the tumor expressing wild-type p53 underwent apoptosis in hypoxic areas whereas the mutant p53 cells would show resistance to hypoxia-induced apoptosis, giving the p53-deficient cells a selective advantage over their wild-type p53 counterparts [24, 52]. Under hypoxia and addition of HIF-1 mimetics such as CoCl2 and DFX [53], it has been demonstrated that HIF-1 and p53 associate leading to p53 stabilization [24, 54] Moreover,

138


it was reported by Carmeliet et al. [55] that accumulation of HIF-1 was induced by anoxia and hypoglycemia only in the presence of HIF-1. However, others have reported that accumulation of p53 is induced by the usual HIF-1 stimuli, even in cells that are HIF-1 deficient [24, 56]. p53 is the most common mutated gene in cancer cells, and it has been demonstrated that it can also affect metabolism [2, 3]. p53 has been demonstrated to have a role in cellular respiration, and it has a direct impact in cellular respiration by contributing to the assembly of cytochrome c oxidase complex [2]. In mice, loss of p53 resulted in decreased oxygen consumption, increased lactic acid production, but comparable ATP production [2]. However, p53-dependent mice had a marked decreased in stamina, which suggests that p53 plays a crucial role in adequate ATP production in sustained physical exercise [2,3]. SCO2 is a transcription target of p53, and p53-deficient cells have decreased levels of SCO2 and no aerobic respiration. After reintroduction of SCO2 in p53 null cells, cells are then driven into glycolysis due to increased ATP requirements. Matoba et al. [2] described SCO2 as the first direct target that modulates energy metabolism [2, 3]. SCO2 is required for adequate assembly of a critical component of the respiratory chain cytochrome c oxidase. Thus, the loss of p53 and the subsequent decrease in SCO2 might be the switch that leads cells to use a glycolytic pathway therefore giving the cell the capability to adapt to increased energy requirements. p53-SCO2 connects tumor suppressors to cellular respiration and to metabolism, which correlates with the previous views on how cellular respiration and tumorigenesis intersect. Mutations in p53 render cells less sensitive to apoptosis induced by hypoxic states, and tumors that lack p53 appear to be more resistant to radiation and chemotherapy [49]. Approximately 90% of mutations in p53 are point mutations of residues affecting the DNA binding domain [49, 57–59]. Additionally, p53 posttranscriptionally controls the level of phosphoglycerate mutase (PGM) [4]. Mechanistically, there are many schools of thought. For example, some find that hypoxia induces the accumulation of p53 not the downstream upregulation of its target genes [60]. On the other hand, others suggest that, under prolonged hypoxia, HIF-1 and p53 form a complex with Mdm2 therefore inhibiting HIF-1 downstream gene activation [61].

8.2 von Hippel–Lindau Syndrome The von Hippel–Lindau syndrome takes its name from German ophthalmologist Eugen von Hippel and Swedish neuropathologist Arvid Lindau who, in the early 19th century, described a syndrome of angiomas in the retina and the cerebellum, respectively. The VHL gene is located at 3p16-15 and is ubiquitously expressed. It consists of three exons that are transcribed into two transcripts that in turn are translated into three proteins [62]. The first transcript contains all three exons and


139

is translated into two proteins, VHL30 and VHL19 containing 213 and 160 amino acids, respectively, both of which are active in tumor suppression [63]. The second transcript contains exons 1 and 3 and is hypothesized to have decreased tumor suppressor activity [64–66]. The gene product consists of a 30-kDa protein that has an ␣ subunit, comprising residues 155–213, and a ␤ subunit that includes the first 154 residues [67, 68]. VHL exon 2 encodes for the ␤ domain, and exon 3 encodes for the ␣ domain [69] The ␤ domain is necessary for conjugation with HIF-1 and fibronectin and apparently mediates nuclear cytoplasmic shuttling of pVHL independent of both O2 and HIF-1. The ␣ domain is encoded by exon 3 and is required for the formation of the BC–cullin-2 complex, E3 ligase activity. Several mutations of the VHL gene account for three main phenotypes of this familial syndrome. The most common phenotype includes cerebellar hemangioblastomas and clear cell renal cell carcinoma with or without adrenal pheochromocytomas. It is often considered an autosomal dominant disease because although both VHL copies need to be mutated according to the Knudson two-hit hypothesis, with one inherited mutant copy of VHL, the second mutation is almost guaranteed [64, 70–72] Previously, we mentioned that pVHL interacts with HIF-1␣ under normoxia after hydroxylation of two proline residues of the latter by oxygen-dependent proline hydroxylases, which targets the protein for proteosomal degradation [16, 29, 71, 73, 74]. Elongin C recruits an E3 ubiquitin ligase complex that in turn consists of an Elongin B, CUL2, and RBX1. Additionally, an E2 ubiquitin-conjugating enzyme attaches to this complex and, later, mediates the ubiquitination of HIF-1. The ␤ subunit will recognize HIF-1-OH. Under hypoxic conditions, the hydroxylation of HIF-1 and pVHL-ligase complex recognition does not take place, and this leads to dimerization and nuclear translocation of both HIF-1 subunits with subsequent binding to HREs and transcriptional upregulation of target genes. Taking into consideration that HIF-1 mediates angiogenesis by transcriptional upregulation of vascular endothelial growth factor receptor (VGFR) among others and that pVHL is necessary for HIF-1 degradation, it is no surprise that two carcinomas with known VHL mutations, such as CCRCC and cerebellar hemangioblastomas, have such a prominent and characteristic vascular network [41, 72]. Several studies point to the fact that pVHL inactivation has only been linked to renal cell carcinoma (RCC) and hemangioblastoma (HB). In these tumors, the cellular composition varies from VHL–/– to VHL+/− whereas endothelial cells are VHL+/− and VHL−/− [72]. Immunocytochemistry shows cellular and vascular proliferation seen by the autocrine-acting TGF-␣ and paracrine-acting VEGF, PDGF, respectively, both downstream genes upregulated by HIF-1 [72]. Furthermore, it has been suggested that the pVHL has more targets, for example the pheochromocytoma-only variant (2C), which might retain its ability to ubiquitinate HIF-1 [37, 67, 68, 71, 72, 75, 76]. Not surprisingly, tumors notorious for their prominent vasculature are part of the VHL syndrome, perhaps due to ineffective ubiquitination of HIF-1␣ leading to upregulation of VEGF and other angiogneic gene products.

140


pVHL has also been reported to control the extracellular matrix (ECM) by controlling fibronectin deposition [77, 78]. In pVHL-deficient cells, the extracellular deposition seen possibly contributes to increased migration and perhaps increases their metastatic potential [78]. Fibronectin 1 expression is not affected by hypoxia, suggesting a HFI-1–independent mode of regulation by pVHL [62]. Thus, there are tumor suppressor pathways of pVHL that are independent of HIF-1 [62, 79, 80]. Additionally, it has been described that pVHL also controls ECM degradation regulating metalloproteinases 2 and 9, their inhibitors, and the urokinase type plasminogen activator system [62, 81–83]. Other genes regulated by the HIF-1/pVHL complex include aminopetidase A, collagen type IV, alfa1, cyclin G2, DEC1/Straqw, endothelin 1, low-density lipoprotein receptor-related protein-1, and MIC/CD99 [79, 80]. In fact, the total number of genes that pVHL targets, either directly or independently of HIF-1, is likely to be of the order of hundreds [80]. Mutations in VHL will stabilize HIF-1 and contribute to tumorigenesis and progression [69]. Independently of HIF-1, through inadequate deposition of ECM proteins, mutations in VHL might select for tumor clones with a phenotype that can more readily invade and metastasize.

8.3 Interactions Between p53 and VHL Apart from their independent contributions to tumor suppression, it has recently been described that the gene products of both genes can also work with each other for their common goal. In 2006, Roe and colleagues [84] showed that p53 binds to pVHL at the latter’s ␣ domain and thus competes for this site with Elongin C. As we mentioned before, the domain of pVHL is the one responsible for the recognition and subsequent proteosomal degradation of HIF-1␣. The pVHL-p53 complex leads inhibition of the Mdm2-mediated ubiquitination of p53, consequently leading to p53 stabilization and upregulated downstream transcription of proapoptotic p21 and Bax promoters [62]. Additionally, it has been shown that it increases the acetylation and association with p300 under genotoxic stress and thus p53 transcriptional activity and p53 mediated cell cycle arrest [62, 84, 85].

9 Conclusion Conditions in the cellular microenvironment, such as hypoxia, not only lead to genetic instability but also lead to the activation of molecular mechanisms determined to confer a survival advantage by phenotypic adaptation. Mechanisms such as the HIF-1 cascade permit the cells to attain these capabilities by activating downstream genes. Limited data exists on how the tumor suppressor genes potentially affect cellular respiration. This aspect perhaps will be a potential target for cancer therapy in the future.


141

References 1. Warburg O. On the origin of cancer cells. Science 1956; 123:309–314. 2. Matoba S, Kang JG, Patino WD, et al. p53 regulates mitochondrial respiration. Science 2006; 312:1650–1653. 3. Kruse JP, Gu W. p53 aerobics: the major tumor suppressor fuels your workout. Cell Metab 2006; 4:1–3. 4. Kondoh H, Lleonart ME, Bernard D, Gil J. Protection from oxidative stress by enhanced glycolysis: a possible mechanism of cellular immortalization. Histol Histopathol 2007; 22: 85–90. 5. Plas DR, Thompson CB. Akt-dependent transformation: there is more to growth than just surviving. Oncogene 2005; 24:7435–7442. 6. Mahon PC, Hirota K, Semenza GL. FIH-1: a novel protein that interacts with HIF-1alpha and VHL to mediate repression of HIF-1 transcriptional activity. Genes Dev 2001; 15:2675–2686. 7. Semenza GL. Targeting HIF-1 for cancer therapy. Nat Rev Cancer 2003; 3:721–732. 8. Krebs H. Otto Warburg: Cell physiologist, biochemist and eccentric. Oxford: Oxford University Press, 1981. 9. Goldblatt H, Cameron G. Induced malignancy in cells from rat myocardium subjected to intermittent anaerobiosis during long propagation in vitro. J Exp Med 1953; 97:525–552. 10. Semenza GL. Hypoxia, clonal selection, and the role of HIF-1 in tumor progression. Crit Rev Biochem Mol Biol 2000; 35:71–103. 11. Semenza GL, Roth PH, Fang HM, Wang GL. Transcriptional regulation of genes encoding glycolytic enzymes by hypoxia-inducible factor 1. J Biol Chem 1994; 269:23757–23763. 12. Semenza GL, Wang GL. 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 1992; 12:5447–5454. 13. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000; 100:57–70. 14. Huang LE, Bindra RS, Glazer PM, Harris AL. Hypoxia-induced genetic instability—a calculated mechanism underlying tumor progression. J Mol Med 2007; 85:139–148. 15. Semenza GL. Regulation of physiological responses to continuous and intermittent hypoxia by hypoxia-inducible factor 1. Exp Physiol 2006; 91:803–806. 16. Semenza GL. Hypoxia and cancer. Cancer Metastasis Rev 2007; 26(2):223–234. 17. Dang CV, Semenza GL. Oncogenic alterations of metabolism. Trends Biochem Sci 1999; 24:68–72. 18. Bindra RS, Crosby ME, Glazer PM. Regulation of DNA repair in hypoxic cancer cells. Cancer Metastasis Rev 2007; 26:249–260. 19. Bindra RS, Glazer PM. Genetic instability and the tumor microenvironment: towards the concept of microenvironment-induced mutagenesis. Mutat Res 2005; 569:75–85. 20. Reynolds TY, Rockwell S, Glazer PM. Genetic instability induced by the tumor microenvironment. Cancer Res 1996; 56:5754–5757. 21. Helmlinger G, Yuan F, Dellian M, Jain RK. Interstitial pH and pO2 gradients in solid tumors in vivo: high-resolution measurements reveal a lack of correlation. Nat Med 1997; 3:177–182. 22. Folkman J. Incipient angiogenesis. J Natl Cancer Inst 2000; 92:94–95. 23. Denko NC, Fontana LA, Hudson KM, et al. Investigating hypoxic tumor physiology through gene expression patterns. Oncogene 2003; 22:5907–5914. 24. Pan Y, Oprysko PR, Asham AM, Koch CJ, Simon MC. p53 cannot be induced by hypoxia alone but responds to the hypoxic microenvironment. Oncogene 2004; 23:4975–4983. 25. Wang GL, Jiang BH, Rue EA, Semenza GL. Hypoxia-inducible factor 1 is a basic-helixloop-helix-PAS heterodimer regulated by cellular O2 tension. Proc Natl Acad Sci USA 1995; 92:5510–5514. 26. Wang GL, Semenza GL. General involvement of hypoxia-inducible factor 1 in transcriptional response to hypoxia. Proc Natl Acad Sci USA 1993; 90:4304–4308.

142


27. Wang GL, Semenza GL. Purification and characterization of hypoxia-inducible factor 1. J Biol Chem 1995; 270:1230–1237. 28. Michel G, Minet E, Ernest I, et al. A model for the complex between the hypoxia-inducible factor-1 (HIF-1) and its consensus DNA sequence. J Biomol Struct Dyn 2000; 18:169–179. 29. Acker T, Plate KH. A role for hypoxia and hypoxia-inducible transcription factors in tumor physiology. J Mol Med 2002; 80:562–575. 30. Semenza GL, Nejfelt MK, Chi SM, Antonarakis SE. Hypoxia-inducible nuclear factors bind to an enhancer element located 3 to the human erythropoietin gene. Proc Natl Acad Sci USA 1991; 88:5680–5684. 31. Shi YH, Fang WG. Hypoxia-inducible factor-1 in tumour angiogenesis. World J Gastroenterol 2004; 10:1082–1087. 32. Jiang BH, Rue E, Wang GL, Roe R, Semenza GL. Dimerization, DNA binding, and transactivation properties of hypoxia-inducible factor 1. J Biol Chem 1996; 271:17771–17778. 33. Jiang BH, Zheng JZ, Leung SW, Roe R, Semenza GL. Transactivation and inhibitory domains of hypoxia-inducible factor 1 alpha: modulation of transcriptional activity by oxygen tension. J Biol Chem 1997; 272:19253–19260. 34. Kourembanas S, Hannan RL, Faller DV. Oxygen tension regulates the expression of the platelet-derived growth factor-B chain gene in human endothelial cells. J Clin Invest 1990; 86:670–674. 35. Kourembanas S, Marsden PA, McQuillan LP, Faller DV. Hypoxia induces endothelin gene expression and secretion in cultured human endothelium. J Clin Invest 1991; 88:1054–1057. 36. Vandromme M, Gauthier-Rouviere C, Lamb N, Fernandez A. Regulation of transcription factor localization: fine-tuning of gene expression. Trends Biochem Sci 1996; 21:59–64. 37. Clifford SC, Astuti D, Hooper L, Maxwell PH, Ratcliffe PJ, Maher ER. The pVHL-associated SCF ubiquitin ligase complex: molecular genetic analysis of elongin B and C, Rbx1 and HIF1alpha in renal cell carcinoma. Oncogene 2001; 20:5067–5074. 38. Lu H, Forbes RA, Verma A. Hypoxia-inducible factor 1 activation by aerobic glycolysis implicates the Warburg effect in carcinogenesis. J Biol Chem 2002; 277:23111–23115. 39. Mabjeesh NJ, Amir S. Hypoxia-inducible factor (HIF) in human tumorigenesis. Histol Histopathol 2007; 22:559–572. 40. Semenza GL. HIF-1 mediates the Warburg effect in clear cell renal carcinoma. J Bioenerg Biomembr 2007; 39(3):231–234. 41. Maxwell PH, Wiesener MS, Chang GW, et al. The tumour suppressor protein VHL targets hypoxia-inducible factors for oxygen-dependent proteolysis. Nature 1999; 399:271–275. 42. Goldberg MA, Dunning SP, Bunn HF. Regulation of the erythropoietin gene: evidence that the oxygen sensor is a heme protein. Science 1988; 242:1412–1415. 43. Wang GL, Semenza GL. Desferrioxamine induces erythropoietin gene expression and hypoxia-inducible factor 1 DNA-binding activity: implications for models of hypoxia signal transduction. Blood 1993; 82:3610–3615. 44. Wykoff CC, Beasley NJ, Watson PH, et al. Hypoxia-inducible expression of tumor-associated carbonic anhydrases. Cancer Res 2000; 60:7075–7083. 45. Levine AJ. p53, the cellular gatekeeper for growth and division. Cell 1997; 88:323–331. 46. Freedman DA, Epstein CB, Roth JC, Levine AJ. A genetic approach to mapping the p53 binding site in the MDM2 protein. Mol Med 1997; 3:248–259. 47. Wu L, Levine AJ. Differential regulation of the p21/WAF-1 and mdm2 genes after high-dose UV irradiation: p53-dependent and p53-independent regulation of the mdm2 gene. Mol Med 1997; 3:441–451. 48. Hollstein M, Rice K, Greenblatt MS, et al. Database of p53 gene somatic mutations in human tumors and cell lines. Nucleic Acids Res 1994; 22:3551–3555. 49. Michalak E, Villunger A, Erlacher M, Strasser A. Death squads enlisted by the tumour suppressor p53. Biochem Biophys Res Commun 2005; 331:786–798. 50. Kinzler KW, Vogelstein B. Life (and death) in a malignant tumour. Nature 1996; 379:19–20.


143

51. Vogelstein B, Kinzler KW. Cancer genes and the pathways they control. Nat Med 2004; 10:789–799. 52. Graeber TG, Osmanian C, Jacks T, et al. Hypoxia-mediated selection of cells with diminished apoptotic potential in solid tumours. Nature 1996; 379:88–91. 53. Garber K. Energy boost: the Warburg effect returns in a new theory of cancer. J Natl Cancer Inst 2004; 96:1805–1806. 54. Ravi R, Mookerjee B, Bhujwalla ZM, et al. Regulation of tumor angiogenesis by p53-induced degradation of hypoxia-inducible factor 1alpha. Genes Dev 2000; 14:34–44. 55. Carmeliet P, Dor Y, Herbert JM, et al. Role of HIF-1alpha in hypoxia-mediated apoptosis, cell proliferation and tumour angiogenesis. Nature 1998; 394:485–490. 56. Wenger RH, Camenisch G, Desbaillets I, Chilov D, Gassmann M. Up-regulation of hypoxiainducible factor-1alpha is not sufficient for hypoxic/anoxic p53 induction. Cancer Res 1998; 58:5678–5680. 57. Vousden KH. Activation of the p53 tumor suppressor protein. Biochim Biophys Acta 2002; 1602:47–59. 58. Ryan KM, Vousden KH. Cancer: pinning a change on p53. Nature 2002; 419:795–797. 59. Vousden KH, Lu X. Live or let die: the cell s response to p53. Nat Rev Cancer 2002; 2: 594–604. 60. Koumenis C, Alarcon R, Hammond E, et al. Regulation of p53 by hypoxia: dissociation of transcriptional repression and apoptosis from p53-dependent transactivation. Mol Cell Biol 2001; 21:1297–1310. 61. Schmid T, Zhou J, Kohl R, Brune B. p300 relieves p53-evoked transcriptional repression of hypoxia-inducible factor-1 (HIF-1). Biochem J 2004; 380:289–295. 62. Ohh M. Ubiquitin pathway in VHL cancer syndrome. Neoplasia 2006; 8:623–629. 63. Latif F, Tory K, Gnarra J, et al. Identification of the von Hippel-Lindau disease tumor suppressor gene. Science 1993; 260:1317–1320. 64. Gnarra JR, Tory K, Weng Y, et al. Mutations of the VHL tumour suppressor gene in renal carcinoma. Nat Genet 1994; 7:85–90. 65. Herman JG, Latif F, Weng Y, et al. Silencing of the VHL tumor-suppressor gene by DNA methylation in renal carcinoma. Proc Natl Acad Sci USA 1994; 91:9700–9704. 66. Sekido Y, Bader S, Latif F, et al. Molecular analysis of the von Hippel-Lindau disease tumor suppressor gene in human lung cancer cell lines. Oncogene 1994; 9:1599–1604. 67. Kaelin WG Jr. The von Hippel-Lindau tumor suppressor gene and kidney cancer. Clin Cancer Res 2004; 10:6290S–6295S. 68. Kim WY, Kaelin WG. Role of VHL gene mutation in human cancer. J Clin Oncol 2004; 22:4991–5004. 69. Bonicalzi ME, Groulx I, de Paulsen N, Lee S. Role of exon 2-encoded beta-domain of the von Hippel-Lindau tumor suppressor protein. J Biol Chem 2001; 276:1407–1416. 70. Knudson AG Jr. Mutation and cancer: statistical study of retinoblastoma. Proc Natl Acad Sci USA 1971; 68:820–823. 71. Semenza GL. VHL and p53: tumor suppressors team up to prevent cancer. Mol Cell 2006; 22:437–439. 72. Kondo K, Klco J, Nakamura E, Lechpammer M, Kaelin WG Jr. Inhibition of HIF is necessary for tumor suppression by the von Hippel-Lindau protein. Cancer Cell 2002; 1:237–246. 73. Semenza GL. HIF-1: using two hands to flip the angiogenic switch. Cancer Metastasis Rev 2000; 19:59–65. 74. Semenza GL. Oxygen-dependent regulation of mitochondrial respiration by hypoxiainducible factor 1. Biochem J 2007; 405:1–9. 75. Clifford SC, Cockman ME, Smallwood AC, et al. Contrasting effects on HIF-1alpha regulation by disease-causing pVHL mutations correlate with patterns of tumourigenesis in von Hippel-Lindau disease. Hum Mol Genet 2001; 10:1029–1038. 76. Hoffman MA, Ohh M, Yang H, Klco JM, Ivan M, Kaelin WG Jr. von Hippel-Lindau protein mutants linked to type 2C VHL disease preserve the ability to downregulate HIF. Hum Mol Genet 2001; 10:1019–1027.

144


77. Bluyssen HA, Lolkema MP, van Beest M, et al. Fibronectin is a hypoxia-independent target of the tumor suppressor VHL. FEBS Lett 2004; 556:137–142. 78. Ohh M, Yauch RL, Lonergan KM, et al. The von Hippel-Lindau tumor suppressor protein is required for proper assembly of an extracellular fibronectin matrix. Mol Cell 1998; 1: 959–968. 79. Xu Q, Briggs J, Park S, et al. Targeting Stat3 blocks both HIF-1 and VEGF expression induced by multiple oncogenic growth signaling pathways. Oncogene 2005; 24:5552–5560. 80. Wykoff CC, Pugh CW, Maxwell PH, Harris AL, Ratcliffe PJ. Identification of novel hypoxia dependent and independent target genes of the von Hippel-Lindau (VHL) tumour suppressor by mRNA differential expression profiling. Oncogene 2000; 19:6297–6305. 81. Merzak A, Koocheckpour S, Pilkington GJ. CD44 mediates human glioma cell adhesion and invasion in vitro. Cancer Res 1994; 54:3988–3992. 82. Los M, Zeamari S, Foekens JA, Gebbink MF, Voest EE. Regulation of the urokinase-type plasminogen activator system by the von Hippel-Lindau tumor suppressor gene. Cancer Res 1999; 59:4440–4445. 83. Zatyka M, Morrissey C, Kuzmin I, et al. Genetic and functional analysis of the von HippelLindau (VHL) tumour suppressor gene promoter. J Med Genet 2002; 39:463–472. 84. Roe JS, Kim H, Lee SM, Kim ST, Cho EJ, Youn HD. p53 stabilization and transactivation by a von Hippel-Lindau protein. Mol Cell 2006; 22:395–405. 85. Sansone P, Storci G, Pandolfi S, Montanaro L, Chieco P, Bonafe M. The p53 codon 72 proline allele is endowed with enhanced cell-death inducing potential in cancer cells exposed to hypoxia. Br J Cancer 2007; 96:1302–1308.

Uncoupling Cellular Respiration: A Link to Cancer Cell Metabolism and Immune Privilege M. Karen Newell, Elizabeth M. Villalobos-Menuey, Marilyn Burnett, and Robert E. Camley

Abstract Epidemiologic evidence strongly correlates suppression of maturation of the immune system to incidences of cancer. The effectiveness of chemotherapeutic agents significantly depends upon the ability of the cancer cells to express (costimulatory) molecules on their surface that can be recognized and engaged by their surrogate ligands of the immune system, thereby resulting in immune-directed cell killing. Alterations in the metabolic strategies of cancer cells (such as: the rate of glucose utilization, glucose or fatty acid oxidation, the magnitude of the membrane potential and the pH gradient across the mitochondrial membrane, and/or changes in uncoupling protein levels) can modulate the expression of some cell surface “death receptors”; thereby conferring resistance to certain cell death-inducing stimuli. Keywords Cancer cell metabolism · Fas · Drug resistance · Uncoupling protein · Metabolic strategy · Metabolic modulators · Mitochondria · Cellular respiration · Immune privilege

1 History Cancer cells exhibit distinct metabolic pathways to meet their energy demands. Otto Warburg postulated that there must be some sort of deficit or abnormality in what has long been described as the most energy-efficient strategy that a cell can use— mitochondrial respiration. The purpose of this chapter is to propose an alternate explanation and to present a hypothesis that can be rigorously tested.

M.K. Newell (B) CU Institute of Bioenergetics and Immunology, University of Colorado at Colorado Springs, Colorado Springs, CO 80918 e-mail: [email protected] e-mail: [email protected] S.P. Apte, R. Sarangarajan (eds.), Cellular Respiration and Carcinogenesis, DOI 10.1007/978-1-59745-435-3 11, C Humana Press, a part of Springer Science+Business Media, LLC 2009

145

146

M.K. Newell et al.

2 An Overview of Cellular Metabolism Every cell in the body uses carbohydrates, protein, or fat in different proportions to ensure that the cell has sufficient energy to perform its normal function. The cell’s choice of fuel (i.e., the cell’s metabolic strategy) changes depending on its activation or differentiation state as well as its environment. For example, a cell that is dividing has different energy demands than does one that is nondividing and, thus, may employ an alternative metabolic strategy. Another example would be the change in strategy for a cell that has been damaged by infection or stress. The majority of nondividing cells in the body use carbons derived from carbohydrate or glucose to efficiently produce the cell’s ultimate energy unit of exchange, adenosine triphosphate (ATP). The process begins with the binding of glucose to the glucose receptor/transporter and transport into the cytosol, where the molecule becomes phosphorylated by hexokinase. Once phosphorylated, the molecule continues to be processed into two molecules of pyruvate. Pyruvate then enters the Krebs cycle, a process that occurs in the mitochondria where ultimately the production of NADH and FADH2 provide reducing equivalents that flow through the four complexes of the electron transport chain, eventually creating a proton gradient between the inner mitochondrial membrane and the matrix. In concert, the electrochemical gradient drives the synthesis of adenosine diphosphate (ADP) to ATP as a result of the electrochemical gradient across the inner mitochondrial membrane.

3 Tumor-Specific Metabolic Pathways Rapidly dividing tumor cells display a characteristic metabolic strategy that distinguishes the tumor cell from its normal, nondividing counterpart cell of the tissues. The key difference is that the rate of glycolysis is much greater than the rate of glucose breakdown in a normal cell under resting conditions. Glycolysis is an oxygen-independent process of splitting the six-carbon glucose molecules and thereby providing energy in the form of ATP to the cell. Aerobic cells use the glycolysis pathway as the initial pathway that ultimately involves oxygen consumption and complete oxidation of the carbons derived from carbohydrates. From a series of 10 reactions, the first five of which are energy “investments” and the second five of which are energy generating, the final product is pyruvate [1]. The fate of the pyruvate molecules produced by glycolysis largely depends on the oxidative state of the cell in question. Glycolysis needs a pool of cytosolic NAD+ , derived from the oxidation of NADH, to proceed. In the presence of oxygen (aerobic glycolysis), the NADH is oxidized in the mitochondria. If the rate of glycolysis in the cell is faster than the rate at which the NADH can be oxidized by electron transport in the mitochondria, the excess NADH must then be used to drive the reduction of another substrate to maintain cellular redox balance. Pyruvate itself is such a substrate, and by the activity of the enzyme lactate dehydrogenase, pyruvate is reduced to lactic acid [1].

Uncoupling Cellular Respiration

147

Glycolysis, whether aerobic or anaerobic, releases only a very small percent of the stored energy in the glucose molecule. The evolution of aerobic metabolism made possible a more energy-efficient use of the stored energy in glucose and in other carbon-containing fuels. Pasteur was the first to observe that when cells metabolizing glucose anaerobically were exposed to oxygen, the rate of glycolysis slowed dramatically (a phenomenon known as the Pasteur effect). This effect likely occurs because it is far more energy efficient to completely oxidize glucose than via just the initial glycolytic steps alone. Much of the control over these processes is determined, not surprisingly, by the energy needs of the cell [1].

4 Mitochondrial Metabolism Pyruvate derived from carbohydrate is one of the three suppliers of acetyl-CoA for oxidation in the Krebs cycle. The conversion of pyruvate to acetyl-CoA is catalyzed by the enzyme pyruvate dehydrogenase. The glycolytic product, pyruvate, is transported from the cytosol to the mitochondria. Once there, the Ca2+ -dependent enzyme pyruvate dehydrogenase catalyzes its production to acetyl-CoA and in that manner feeds the glucose-derived carbons to the Krebs cycle ultimately providing the reducing equivalents necessary for electron transport, oxidative phosphorylation, and the very efficient enzymatic production of ATP. The activity of pyruvate dehydrogenase requires a high mitochondrial membrane potential across the inner mitochondrial membrane [1]. When pyruvate supplies are high or when there is insufficient Ca2+ to activate pyruvate dehydrogenase, the source of the acetyl-CoA switches from glucosederived pyruvate to the acetyl-CoA produced by fatty acid oxidation. The source of fatty acid can be endogenous or exogenous, composed of short-, medium-, or longchain fatty acids. When long-chain fatty acids are used, the process that provides acetyl-CoA is beta-oxidation. A key enzyme in the transport of fatty acids is carnitine palmitoyl transferase (either I or II depending on the tissue of origin). Tumor cells in general and drug-resistant tumor cells in particular, as well as other cells that remain “immune privileged,” have a flexible capacity for the ready switch to fatty acid oxidation as a result of mitochondrial expression of key players, including mitochondrial uncoupling proteins that uncouple glycolysis from oxidative phosphorylation. This process results in a lower mitochondrial membrane potential, decreased sensitivity to free radical damage, and insensitivity to apoptosis inducing stimuli, such as chemotherapy or radiation [1].

5 The Immune Response in Cancer Our work has addressed the possibility that the immune system monitors the metabolic state and degree of damage that characterizes all cells. We have found that, though perhaps preferentially destroying those in an excessively damaged state (i.e., stressed/damaged cells or infected cells), nonetheless, tumor cells often survive this surveillance by changing their metabolic strategy to one that is protective [24].

148

M.K. Newell et al.

When tumor cells are treated with traditional chemotherapeutic drugs or radiation, the damage from these treatments causes cell surface changes allowing the cell to be recognized once again by the immune system. Drug-resistant cell variants display a characteristic metabolic state that masks them from the immune system (even after treatment with chemotherapeutic drugs or radiation) and that confers upon the tumor cell profound protection from further insults [2]. We have focused on understanding the characteristic metabolic states that protect abnormally proliferating cells. Cancer cells, particularly multidrug-resistant variants, have a strategy for survival that involves selective fatty acid oxidation in the mitochondria. We have identified several drug candidates that block the activity of these proteins. Preliminary studies indicate that these compounds are capable of interfering with the metabolic strategy of most tumor cells and have striking therapeutic activity in tumor-bearing mice when used alone or in conjunction with standard chemotherapy. We are currently investigating several novel drugs designed to target metabolic strategies used by tumor cells, in particular multidrug-resistant tumors, and other diseases characterized by abnormally proliferating cells. Attempts to trigger an effective immune response against growing cancer cells have been a priority in a large body of research. This is not surprising given the results that tumor allografts are routinely rejected by an effective immune response, the observation that children and the elderly (considered somewhat immunocompromised) have increased frequencies of tumors relative to the general population, and the increased incidence of tumors in immunocompromised hosts. The problems have been to understand enough about how the immune system detects abnormal proliferation and, likely more importantly, how an immune response can distinguish self from non-self, leading to the question of whether tumor cells are foreign or self. Clearly, in many instances there are associations with viral infections, although in other cases there are no clear links between a pathogen and tumorigenesis.

6 Fundamental Questions Every year at least 6.2 million people die worldwide from cancer [3]. The development of drug resistance attenuates the effectiveness of chemotherapy in many cases. In this chapter, we explore whether chemotherapy can be enhanced by increasing the response of the immune system to the tumor cells and by making drug-resistant tumor cells more visible to the immune system. Specifically, we ask the following questions: 1. Do effective chemotherapeutics sensitize tumors to immune destruction by increasing cell surface expression of costimulatory molecules? 2. Can this effect be enhanced by other treatments? 3. Is drug resistance, in part, due to differences in metabolic behavior between drugsensitive and drug-resistant cells? Using this idea, can we create new methods for treating drug-resistant tumors?


149

7 The Immune System and Cancer Extraordinary advances in our understanding of the immune system have been made in the past 100 years, especially since the discovery of the T cell and the B cell [4–6]. Na¨ıve T cells require two signals for activation. These are recognition of antigens in major histocompatibility complex (MHC)-encoded molecules [4] and a costimulation signal [7–11] provided by the B7/CD28 family members or other costimulatory molecules such as Fas (CD95) [12]. Previously activated T cells can be reactivated by costimulation alone [13, 14]. In the absence of activation, T cells disregard the tissue. If a T cell is activated, the consequences can be (1) destruction of the damaged cells or (2) repair of damaged cells by promoting regeneration either directly or indirectly. There is substantial evidence that the immune system plays an extensive role in suppressing cancer. However, it is also clear that the ability of the immune system to control cancer is not perfect. Researchers have tried to stimulate the immune system as a therapeutic strategy against cancer for many years [15, 16]. These attempts have generally been ineffective, although there have been some recent successes [17]. There are many reasons for this variability. These include (1) the inability to activate T cells that can destroy the tumor due to the absence of signal one (recognition of the appropriate tumor antigen); (2) the presence of a signal two that results in the production of the wrong cytokines by T cells, which may lead to the growth of tumors, or (3) the failure of activated T cells to kill cancerous cells [18]. In contrast with therapies that concentrate primarily on activating the immune system, we will explore the idea that effective chemotherapeutics work in concert with the immune system and derive some of their effectiveness from the immune response. This means that by tuning the immune response, we may be able to increase effectiveness in cancer therapy. Cellular metabolism may play a dominant role in modulating the intercellular communication between a cancer cell and a T cell and thereby plays a dominant role in T-cell activation. In particular, we have observed a significant correlation between cellular metabolism and the expression of cell-surface Fas, one of the key costimulatory molecules. Fas is expressed on most rapidly dividing and self-renewing cells, including a wide variety of tumor cells [19]. Whereas Fas is widely known as a “death receptor” and can induce programmed cell death when engaged by Fas ligand (FasL, CD95L) or anti-Fas antibodies [20, 21], paradoxically, Fas engagement can also result in accelerated proliferation of T lymphocytes and some tumor cells [22, 23]. Cell surface Fas expression is upregulated in response to increasing glucose concentrations in vitro in transformed cell lines and in freshly isolated cells from a variety of tissue origins [19]. Obviously, glucose is a key source of fuel, and the availability of glucose drives metabolic activity. A further example of the connection between cellular metabolism and Fas is found in the work of Bhushan et al. [2] who showed that drug-resistant tumor cells fail to express cell surface Fas and the work of Harper et al. [24] who showed that drug-resistant tumor cells also demonstrated a unique metabolic strategy, quite different from drug-sensitive tumor cells.

150

M.K. Newell et al.

8 Drug Treatment and Fas-Induced Tumor Cell Death Anticancer agents may work to promote the death of tumor cells in multiple ways. First, these agents may work by direct cytolysis requiring active participation of the tumor cell in the death process [25]. Second, chemotherapeutics may promote the ability of the tumor cell to be recognized by cells of the immune system and to be killed by immune-directed cell death [26]. Third, these agents may work to “rewire” the death-inducing receptor/ligand pairs, which include Fas and FasL [23, 25, 26]. The second and third possibilities are not mutually exclusive and may work in concert to result in tumor cell death (see Table 1). Drug resistance is the leading cause of death in cancer [27]. Mechanisms that have been suggested to account for drug resistance include overexpression of a multidrug-resistance transporter (pgp -1) [27], failure to express death-inducing receptors [27, 28], and our recent description of a metabolic strategy that may provide protection from a variety of stresses [24]. Our hypothesis is that mitochondrial metabolism in tumor cells significantly influences susceptibility or resistance to Fas-dependent/anticancer agent–induced tumor cell death. Our preliminary work provided the framework for the hypothesis that metabolic strategy confers resistance or susceptibility to apoptotic death of tumor cells. We suggest that treatment of some cancer cells by chemotherapeutics does indeed produce substantially enhanced costimulatory signals and that the treatment of cancer cells by chemotherapeutics and immune system involvement results in substantially higher tumor cell death rates in the measured cases. Figure 1 shows that treatment of L1210 tumor cells (a mouse leukemic cell line [29]) cultured for 24 hours with methotrexate induces significant increases in cell surface Fas. The lower concentration of 10–8 M is physiologically relevant to chemotherapy because this is the upper limit of drug serum levels in humans. Our results here are consistent with some of our previous published work where we showed that a different costimulatory signal (B7.1) also increased in response to methotrexate [2]. Figure 2 shows that different chemotherapeutic agents induce an increase in Fas, but that different drugs can produce increases of varying amounts. This leads to the tantalizing possibility that drug effectiveness could be correlated with levels of Fas

Table 1 Percent Death of L1210 cells with or without a 96-Hour Incubation with 10–8 M Methotrexate and with or without an Exposure to Anti-Fas Antibodies∗ Percent death L1210 cells (%) Anti-Fas exposure

No methotrexate

With methotrexate

No anti-Fas Anti-Fas coated plates

1 7

4.72 79.98

∗ The results indicate that methotrexate sensitize leukemic cells to Fas-induced death.


151

Fig. 1 Expression of cell surface Fas as a function of treatment with a typical oncolytic agent. There is a significant increase in cell surface Fas as a result of drug treatment. The level of cell surface Fas was determined using fluorochromeconjugated anti-Fas antibodies and flow cytometry. The Fas levels are measured relative to a control for staining artifacts

expression and that this could be used to predict effective targeting of drugs via Fas-induced destruction. We have concentrated on Fas as an important example of the costimulatory signal; however, there are a number of other cell surface molecules that also play a role in T-cell activation. In Fig. 3 , we show that B7.2 levels in HL60 (human leukemic cells) also increase after treatment with 10–8 M Adriamycin. HL60 is a human cell

Fig. 2 Flow cytometry data showing increasing levels of cell surface Fas in L1210 cells as a result of treatment with different chemotherapeutic agents. The horizontal axis shows intensity of the fluorescence, and the vertical axis shows the number of cells at each intensity. Peaks that appear farther to the right have higher levels of Fas per cell

152

M.K. Newell et al.

Fig. 3 Expression of the cell surface costimulatory molecule B7.2 as a function of treatment with a typical oncolytic agent. The level of cell surface B7.2 was determined using fluorochrome-conjugated anti-B7.2 antibodies and flow cytometry. The B7.2 levels are measured relative to a control for staining artifacts [31]

line, and the drug is different than that in previous figures. This implies that the increase in the costimulatory signal as a result of drug treatment may be a general phenomenon. The data above strongly support the connection between the efficacy of drug treatment and an increase in costimulatory immune signals. However, it is important to know if this increase actually results in an increased immune-mediated death rate for the tumor cells. In contrast with treatment with anti-Fas or methotrexate alone, anti-Fas produces a striking increase in percent death for L1210 cells that have been cultured in methotrexate. Therefore, drug treatments result in increased expression of costimulatory signals and this, in turn, plays an important role in successful tumor cell treatment. Our second question concerned whether the treatment of tumor cells by chemotherapeutics supplemented with agents that promote reactive oxygen intermediates result in increased susceptibility to immune-mediated cell death. As we saw from the data above and from the work of others [28], chemotherapeutics cause an increase in the expression of costimulatory molecules. There are other treatments that can also produce a similar increase. For example, Fig. 4 shows the effects of culturing two tumor cell lines in subcytoxic concentrations of H2 O2 . The H2 O2 treatment results in a threefold increase in the costimulatory molecule B7.2. The increase from treatment with H2 O2 is also significantly larger that the increase resulting from treatment with Adriamycin alone (Fig. 3). Experiments with other cell types show that the expression of other costimulatory molecules can also be substantially changed by the H2 O2 treatment. For example, Fig. 5 shows that treatment of JB6 cells (a mouse fibroblast tumor cell line) with H2 O2 causes up to a fourfold increase in cell surface Fas expression. Clearly what


153

Fig. 4 Relative B7.2 levels for two different tumor cell lines with and without addition of subcytotoxic levels of H2 O2 . Obviously, the addition of H2 O2 increases the expression of cell surface costimulatory molecules significantly

is needed is to look at the effectiveness of combined approaches, for example, using both chemotherapeutics and H2 O2 . Our third question dealt with whether drug resistance can be overcome by treatments that promote immune involvement in the destruction of tumor cells and involved the hypothesis that uncoupling proteins regulate drug resistance by changing the metabolic state of a cell.

Fig. 5 Relative cell surface Fas expression in JB6 cells as a function of time after exposure to subcytotoxic levels of H2 O2 . We see, again, that the levels of costimulatory molecules increase substantially after exposure to H2 O2

154

M.K. Newell et al.

Fig. 6 A scatter diagram for drug-sensitive (left panel) and drug-resistant (right panel) tumor cells. The horizontal axis is a measure of the change in pH across the inner mitochondrial membrane. The vertical axis measures mitochondrial membrane potential. The results show that mitochondrial membrane potential and pH difference is lower for the majority of drug-resistant cells. The data are obtained from flow cytometry using a JC-1 dye

First we show that drug-sensitive cancer cells and drug-resistant tumor cells have very different metabolic properties. As is well-known, cell metabolism is largely governed by the action of the mitochondria. This, in turn, depends on some key parameters that include the potential difference across the inner mitochondrial membrane and the change in acidity across the same membrane. Figure 6 shows a scatter diagram for drug-sensitive and drug-resistant tumor cells indicating the range of both of these parameters. The abscissa is a measure of the change in pH across the inner mitochondrial membrane. The ordinate measures mitochondrial membrane potential. We see that the drug-resistant cells, in general, have a lower membrane potential, and the pH gradient is also significantly reduced for these cells. These data have been substantiated and extended to other cell lines (including HL60 and HL60/MDR) by other measurements [24]. This provides strong evidence that drug resistance is correlated with altered metabolic behavior. A second example of how drug-resistant cells exhibit a different metabolic behavior compared with drug-sensitive cells is provided by the rate of mitochondrial oleate consumption in these two different types of cells [24]. Figure 7 shows that the drug-resistant cells typically have higher levels of oleate consumption, and as glucose becomes limited, the rate of consumption increases dramatically in the drug-resistant cells only. This argues that the ability to ”burn fat” is substantially different between the two types of cells.


155

Fig. 7 Rates of fatty acid (oleate) oxidation for drug-sensitive and drug-resistant cells. The oxidation rates for drug-resistant cells are generally higher than those for drug-sensitive cells, and this becomes more pronounced at low glucose levels [24]

An important question is whether this metabolic difference leads to differences in the way tumor cells present themselves to the immune system. In Fig. 8 , we present evidence that drug-sensitive tumor cells (those in the left panels) typically display relatively high levels of cell surface Fas, one of the requirements for Fas-induced death [2, 30]. In contrast, the drug-resistant cells (seen in the right panels) show virtually no cell surface Fas. These data suggest that the differences in metabolic behavior are correlated with expression of cell surface Fas. Clearly, it is not known if this is a cause-and-effect behavior or if it is mere correlation. Nonetheless, the possibility of a causal relationship between cellular metabolism and the expression of cell surface Fas may offer an intriguing tool for making tumor cells visible to the immune system. Treatment of drug-resistant tumor cells with chemotherapeutic agents does not increase expression of cell surface Fas or other costimulatory molecules. This behavior is illustrated in Fig. 9 below. This is in stark contrast with the behavior of drug-sensitive tumor cells (see Fig. 2, Fig. 3, and Fig. 4), which do show substantial increases in Fas with chemotherapeutic treatments. The results above suggest that one possible reason that chemotherapy is ineffective for drug-resistant cells is that the costimulatory signal is absent. This naturally leads to the question of whether costimulation, particularly in drug-resistant cells, could be increased by other treatments. In Fig. 10 , we show a summary of experiments on three different (relatively) drug-resistant human cell lines. We see

156

M.K. Newell et al.

Fig. 8 Levels of cell surface Fas expression on drug-sensitive (left panels) and drug-resistant (right panels) cells as measured by flow cytometry. The horizontal axis shows intensity of the fluorescence, and the vertical axis shows the number of cells at each level of fluorescence intensity. Peaks that appear farther to the right have higher levels of Fas per cell. Note that the drug-sensitive cells have significantly higher levels of cell surface Fas

that treatment with subcytotoxic H2 O2 increases B7.2 expression in all cases and that it is increased by a factor of 2 to 3 in some cases. This increase is, in fact, very similar to that for the drug-sensitive cells (see Fig. 4 for example). We have suggested that differences in metabolic behavior correlate with differences in sensitivity to drugs [24]. One possible mechanism that would account for this could be the activity of the uncoupling protein UCP2. These proteins have been shown to produce a number of metabolic changes that are consistent with the metabolic behavior of drug-resistant cells. We measured the presence of UCP2 in mitochondria of drug-resistant and drug-sensitive cells by isolating mitochondria and using antibodies to UCP2 in Western blot analysis (Fig. 11 ). Significantly higher levels of UCP2 were detected consistently in the resistant cell lines (lanes 4 and 5). These results provide the rationale for an investigation of the role of uncoupling proteins as a part of the mechanism for drug resistance. If this mechanism were confirmed, this would create a new approach for treatments of drug-resistant cancers.

Uncoupling Cellular Respiration Fig. 9 Levels of cell surface Fas expression on drug-sensitive L1210 (upper panel) and drug-resistant L1210/DDP (lower panel) cells as measured by flow cytometry. Cells were cultured in the presence of chemotherapeutic agents overnight. The drug-sensitive cells show increases in Fas as a result of the drug. In contrast, the drug-resistant cells are unchanged with treatment. The horizontal axis shows intensity of the fluorescence, and the vertical axis shows the number of cells at each intensity. Peaks that appear farther to the right have higher levels of Fas per cell. Note that the drug-sensitive cells have significantly higher levels of cell surface Fas

Fig. 10 Relative expression of the costimulatory molecule B7.2 on three different drug-resistant human cell lines with and without treatment with subcytotoxic amounts of H2 O2 . In contrast with treatments with chemotherapeutics alone, we see here a substantial increase in the costimulatory signal, which is achieved by adding the H2 O2

157

158

M.K. Newell et al.

Fig. 11 Western blot analysis for UCP2 of mitochondrial extracts from drug-sensitive L1210 (lanes 2 and 3) and drug-resistant L1210/DDP (lanes 4 and 5) tumor cells. UCP2 levels are substantially larger in the drug-resistant cells [24]

9 Therapeutic Possibilities As an example of using a metabolic approach to provide therapeutic benefit, we will discuss the performance of one compound, dichloroacetate, in detail. There are two overall goals: first, to slow or inhibit high-rate glycolysis, and the second, to block fatty acid oxidation at several key enzymatic steps. We and others have identified 2-deoxyglucose (2DG) as a key regulator of highrate glycolysis. However, our work has demonstrated that for most rapidly dividing tumor cells, treatment with 2DG alone does not cause tumor cell death. In fact, in terms of percent viability, treatment with 2DG results in improved percent viability for the vast majority of tumors. The mechanism appears to be that the compound results in growth arrest of the cells. In combination with traditional chemotherapeutic agents, the addition of 2DG increases the oncolytic activity of the therapeutic. Though beneficial, this approach leaves open the likely possibility that treatment with 2DG and chemotherapeutic agents such as Taxol (paclitaxel) or Adriamycin will eventually result in drug resistance, the leading cause of death due to cancer. Targeting metabolism as a chemotherapeutic approach is a major focus of our work, and we focus on the inhibition of tumor cell–specific fatty acid oxidation. Toward that end, we have used the several fatty acid oxidation (FAO) inhibitors. We have extensive data showing the effectiveness of this approach in inducing the apoptotic death of a wide variety of tumor types. We note the listing of dichloroacetate (DCA) in each of these two categories of metabolic modifications: as a glycolytic inhibitor as well as a fatty acid oxidation inhibitor. With metabolic strategies, one must consider that cells have feedback strategies that communicate when energy demands are met or when energy demands are unmet. The cell will respond accordingly with either catabolic or anabolic pathways to provide optimization of energy expenditure. Because DCA is similar in structure to pyruvate, the molecule can act as a negative regulator of glycolytic processes. In addition, the structure of DCA is known to fit the negative regulatory binding pocket of pyruvate dehydrogenase kinase, an enzyme that increases under conditions of starvation promoting fatty acid oxidation when glucose or carbohydrate


159

stores are limited. Thus, DCA will act as a negative regulator of fatty acid oxidation, a key strategy of drug-resistant tumor cells. This selective interference causes cells that are normally selectively nonapoptotic to now die from apoptotic cell death. While theoretically promising, the effects of DCA when used in combination with current “standard of care” chemotherapeutics need to be rigorously tested for potentially harmful or toxic side effects. In summary, the technique of metabolic disruption shows promise for creating therapeutic strategies that may help combat cancer or other diseases. These strategies have the potential of changing the way the immune system interacts with tumors and other metabolically altered tissues.

References 1. Voet D, Voet J. Chapter 22: Electron Transport and Oxidative Phosphorylation. Biochemistry John Wiley & Sons, Inc. (2004). 2. Bhushan A, et al. Drug resistance results in alterations in expression of immune recognition molecules and failure to express Fas (CD95). Immunol Cell Biol1998; 76:350–356. 3. Brundtland GH. The Global Burden of Cancer In: Stewart BW, Kleihues P, eds. World cancer report. Lyon: International Agency for Research on Cancer, World Health Organization, 2003:352. 4. Marrack P, Kappler J. The T cell receptor. Science 1987; 238:1073–1079. 5. Bretscher PA, Cohn M. A theory of self-nonself discrimination. Science 1970; 169: 1042–1049. 6. Linsley PS, Ledbetter JA. The role of the CD28 receptor during T cell responses to antigen. Ann Rev Immunol 1993; 11:191–212. 7. Linsley PS, et al. Human B7-1 (CD80) and B7-2 (CD86) bind with similar avidities but distinct kinetics to CD28 and CTLA4. Immunity 1994; 1:793–801. 8. June CH, et al. The B7 and CD28 receptor families. Immunol Today 1994; 15:321–330. 9. Kuchroo VK, et al. B7-1 and B7-2 costimulatory molecules activate differentially the Th1/Th2 developmental pathways: application to autoimmune disease therapy. Cell 1995; 80: 707–718. 10. Lanier LL, et al. CD80(B7) and CD86(B70) provide similiar costimulatory signals for T cell proliferation, cytokine production, and generation of CTL. J Immunol 1995; 154: 97–105. 11. Alderson MR, et al. Fas transduces activation signals in normal human T lymphocytes. J Exp Med 1993; 178:2231–2235. 12. Nagata S, Human autoimmune lymphoproliferative syndrome, a defect in the apoptosisinducing Fas receptor: a lesson from the mouse model. J Hum Genet 1998; 43(1):2–8. 13. Desbarats J, et al. Dichotomy between na¨ıve and memory CD4+ T cell responses to Fas (CD95) engagement. Proc Natl Acad Sci USA 1999; 96:8104–8109. 14. Desbarats J, Newell MK. Fas engagement accelerates liver regeneration after partial hepatectomy. Nat Med 2000; 6(8):920–923. 15. Mavligit GM, et al. Cell-mediated immunity to human solid tumors: in vitro detection by lymphocyte blastogenic responses to cell-associated and solubilized tumor antigens. Natl Cancer Inst Monogr 1973; 37:167–176. 16. Whelan M, et al. Cancer immunotherapy: an embarrassment of riches? Drug Discov Today 2003; 8(6):253–258. 17. Martindale D. T cell triumph: immunotherapy may have finally turned a corner. Sci Am 2003; 288(2):18–19.

160

M.K. Newell et al.

18. Tsuruo T, et al. Molecular targeting therapy of cancer: drug resistance, apoptosis and survival signal. Cancer Sci 2003; 94(1):15–21. 19. Newell MK, et al. Does the oxidative/glycolytic ratio determine proliferation or death in immune recognition? Ann NY Acad Sci 1999; 887:77–82. 20. Nagata S, Golstein P. The Fas death factor. Science 1995; 267:1449–1456. 21. Nagata S. Apoptosis by death factor. Cell 1997; 88:355–365. 22. Schneider P, Tschopp J. Apoptosis induced by death receptors. Pharm Acta Helv 2000; 74(2–3):281–286. 23. Kataoka T, et al. Expression level of c-FLIP versus Fas determines susceptibility to Fas ligandinduced cell death in murine thymoma EL-4 cells. Exp Cell Res 2002; 273(2):256–264. 24. Harper M-E, et al. Characterization of a novel metabolic strategy used by drug-resistant tumor cells. FASEB J 2002;16(12):1550–1557. 25. Sinkovics JG, Horvath JC. Virological and immunological connotations of apoptotic and antiapoptotic forces in neoplasia. Int J Oncol 2001; 19(3):473–488. 26. Green DR, Evan GI. A matter of life and death. Cancer Cell 2002; 1(1):19–30. 27. Tolomeo M, Simoni D. Drug resistance and apoptosis in cancer treatment: development of new apoptosis-inducing agents active in drug resistant malignancies. Curr Med Chem AntiCancer Agents 2002; 2(3):387–401. 28. Landowski TH, et al. Myeloma cells selected for resistance to CD95-mediated apoptosis are not cross-resistant to cytotoxic drugs: evidence for independent mechanisms of caspase activation. Blood 1999; 94(1):265–274. 29. Venditti JM, Sheldon DR, Goldin A. Evaluation of antileukemic agents employing advanced leukemia L1210 in mice. VII. Cancer Res 1964; 24(1 Pt 1):145–210. 30. Villalobos-Menuey EM. Metabolic regulation of the cellular distribution and function of Fas (CD95). Master’s thesis, University of Colorado, Colorado Springs, 2001:1–83. 31. Newell MK, Melamede RJ, Villalobos-Menuey E, Swartzendruber D, Trauger R, Camley RE, Crisp W. The effects of chemotherapeutics on cellular metabolism and consequent immune recognition. J Immune Based Ther Vaccines 2004; 2(1):3.

How Cancer Cells Escape Death Erica Werner

Abstract Programmed cell death, or apoptosis, is a cell-autonomous mechanism to restrict deviation from normal cellular function. Central to this mechanism are Bcl-2 proteins, which monitor normal cellular function to define apoptotic fate by surveying, integrating, and decoding multiple survival and proapoptotic signals from the intra and extra cellular environment. Escape from these restrictive mechanisms is essential to accommodate the cellular changes driving tumor establishment and development. To do so, many of the factors promoting tumorigenesis, such as genetic lesions, hypoxia, p53 mutations and oncogene activation, gain control over the main apoptotic pathway to promote cell survival. For this reason, the apoptotic pathway is the focus of remarkable interest for targeted cancer therapy development. However, as apoptosis evasion is a tumor promoting factor itself, the results of first therapeutic targeting attempts reveal that these targets pose drawbacks similar to the exhibited by classic oncogenic targets, including the rapid development of resistance through high mutation rate, intra and inter tumor heterogeneity and narrow effective range. In addition to tumor promoting pathways, apoptosis evasion is fostered by factors secondary to transformation and tumor development. These factors, such as changes in the interaction with the extracellular environment and metabolism, have received less attention because of their underestimated role in transformation but might prove to be fruitful avenues for intervention as they are common to many tissues and affect apoptotic fate under the influence of reduced oncogenic pressure. The significance of these factors with respect to apoptosis evasion and their potential impact for therapy development are discussed. Keywords Glycolysis

Apoptosis · Bcl-2 · Mitochondria · Prosurvival · Environment ·

E. Werner (B) Department of Cell Biology, School of Medicine, Emory University, Atlanta, GA 30322 e-mail: [email protected] S.P. Apte, R. Sarangarajan (eds.), Cellular Respiration and Carcinogenesis, DOI 10.1007/978-1-59745-435-3 12, C Humana Press, a part of Springer Science+Business Media, LLC 2009

161

162

E. Werner

1 Introduction Death is universal and certain. At the cellular level, this fate is controlled by apoptosis, a genetic program present in every cell type of organisms, including cancer cells. Evasion of this program is deemed key to carcinogenesis and tumor progression. Thus, extraordinary efforts have been committed to understand how cancer cells evade the apoptotic pathway and thereby influence disease prognosis and targeted therapy design. Although past and current studies have consistently revealed the significant role of apoptosis evasion in the process of carcinogenesis, correlations of disease progression and prognosis with the expression of apoptosis-related markers have been inconsistent, and therapies targeting effectors of the apoptotic pathway have been of limited efficacy [1]. This suggests that additional cell autonomous or extracellular mechanisms are likely contributing to apoptotic outcome. This will be the major focus in this chapter. The apoptotic pathway is a surveillance mechanism monitoring cell activity and imposing strict limits to permissible cellular function. These limits are constantly challenged during cancer initiation and progression. Early deregulated proliferation demands multifaceted changes encompassing intracellular signaling pathways, metabolism, nutrient and energy requirements, and tissue architecture, which trigger intracellular and extracellular proapoptotic signals in normal cells. At later stages of tumor development, cancer expansion and metastasis adds a further cohort of apoptosis-inducing signals, generated in opposition to changes in the function of multiple cell types and altered environments. In contrast with mutations in oncogene and tumor suppressor genes, evasion of the apoptotic pathway does not initiate and drive carcinogenesis per se, but is instrumental for tumor cells to successfully thrive despite harboring alterations that would not be tolerated in normal circumstances. The mechanisms driving tumor progression are integrated with apoptosis resistance acquisition, at least at three levels, so as to ensure successful apoptosis evasion: Somatic mutations, promoter methylation, chromosomal translocations and deletions associated with the inherent genetic instability of cancer cells alter the expression of oncogenes as well as those of key apoptotic players. Even though this has been the most explored mechanism to explain the increased apoptosis resistance of tumor cells [1], correlations in most cases are difficult to generalize and in many cases are nonconfirmatory. One interpretation of this outcome could be that activity rather than expression levels are consequential for carcinogenesis, as the apoptotic players are mainly regulated through posttranslational mechanisms, such as phosphorylation or ubiquitin-targeted proteasome-dependent degradation. Another reason may be associated with the difficulty to evaluate whether the identified changes in expression are indeed a causal factor in disease progression or are merely coincidental. A second factor promoting apoptosis resistance is an overall shift in the nature of signals surveyed by the apoptotic machinery, reinforcing prosurvival signals while interfering with proapoptotic signals (see Fig. 1). This altered input is the direct result of two concurrent processes; one leading to independence from an

How Cancer Cells Escape Death

163

Fig. 1 Apoptosis input is balanced by prosurvival (gray) and proapoptotic (black) signals. These signals control the activity (crossed arrows represent inhibitory signals, and pointed arrows represent activation; dashed lines represent loss of that particular connection) of antiapoptotic Bcl-2 members (round boxes) and proapoptotic Bcl-2 family members (square boxes). (A) In normal cells, the extracellular environment through growth factors, cell adhesion, and cytoskeleton organization restrains activity of proapoptotic Bcl-2 family members. P53 responds to endogenous stress providing proapoptotic signals. (B) During carcinogenesis, control of the apoptotic pathway is taken over by oncogene-activated kinases, prosurvival cell-surface interactions, glycolysis activity, and loss of p53 function

164

E. Werner

environment rich in proapoptotic signals and the other altering the intracellular signal transduction network through oncogene activity and tumor suppressor silencing. The marked influence of these factors on apoptotic fate becomes apparent as apoptosis sensitivity can be restored by normalizing cancer cell communication with the extracellular environment or by neutralizing oncogene activity (see Section 3). A third, far less studied factor affecting apoptotic fate is the characteristic altered metabolism associated with transformed and tumor cells. Mitochondria are a central piece in the apoptotic pathway, but are a control center for normal and tumor metabolism as well. The importance of metabolism in carcinogenesis is gaining recognition and promises a source of potentially viable therapeutic approaches, as many of the involved pathways control enzymatic activities suitable for pharmacologic inhibition are linked to apoptosis. Because of space constrains, in this chapter I will only outline the essential players and concepts necessary to provide a conceptual framework illustrated by a few examples and will cite only select key reviews to refer the reader to further details in each section.

2 The Bare Pathway to Death The essential apoptotic pathway responds to specific triggers, such as proapoptotic extracellular ligands, cellular stressors, or deprivation of a survival signal. These elicitors activate an intracellular proteases cascade in charge of cleaving multiple targets to cause cell death. The resulting cell debris, organelle and nucleus fragments are packed into apoptotic bodies to be cleanly disposed by professional phagocytes and/or neighboring cells thereby averting inflammatory responses [2]. At the center of the apoptotic pathway is a regulatory step dictated by the proteinprotein interaction of the Bcl-2 family members, controlled by proapoptotic and prosurvival signals generated in the extracellular environment (extrinsic pathway) and inside the cell (intrinsic pathway). If the activity balance of Bcl-2 proteins is inclined toward cell death, proapoptotic members redistribute to mitochondria and cause outer membrane permeability, with the consequent release of proteins contained in the mitochondrial intermembrane space [3]. The proteins released into the cytosol activate caspases and other regulators/effectors of cell breakdown.

2.1 The Family of Bcl-2 Proteins: Initiators of the Intrinsic Pathway Cellular stress signals such as growth factor deprivation, DNA damage, oncogene activation, or hypoxia trigger apoptosis by regulating the activity of members of the Bcl-2 protein family. Each member shares a protein-protein interaction domain with homology to Bcl-2 (BH domains) and is grouped according to its role in apoptosis: multidomain antiapoptotic (Bcl-2, Bcl-xL, Bcl-w, Mcl-1, and Bfl-1/A1), multidomain proapoptotic (Bax and Bak), and BH3-only proapoptotic (Bid, Bim, Bad, Bik,


165

Noxa, Puma, Bmf, and HRK). Tissue-specific expression and posttranslational regulation contribute to establish an activity balance of these proteins, which mediates stimuli-selective and tissue-specific induction of apoptosis. The signals that influence apoptosis converge to control the activity of proapoptotic Bax and Bak, as genetic deletion of both proteins renders cells unable to undergo apoptosis via the intrinsic pathway [4]. When the balance favors activity of proapoptotic members, Bax translocates from the cytosol to the mitochondrial outer membrane, where it polymerizes with mitochondria-localized Bax to form a pore. Bcl-2 family members modulate this step by two alternative but nonexclusive mechanisms. Some signals induce apoptosis by activating proapoptotic BH3-only proteins, which bind to and neutralize selectively antiapoptotic Bcl-2 family members, releasing Bak/Bax to oligomerize. For example, proapoptotic Bad mediates induction of apoptosis by this mechanism [5]. Other proapoptotic BH3-only proteins such as Bid and Bim compete with antiapoptotic members but also have the potential to interact directly with Bax/Bak for activation. In both scenarios, the emerging consensus is that several of the antiapoptotic factors need to be simultaneously neutralized by proapoptotic factors for apoptosis to proceed, requiring the cooperation of multiple members of the family acting as sensitizers or de-repressors. Thus the current view is that because proapoptotic BH3-only members present a selective pattern of interaction with antiapoptotic Bcl-2 family members, the right combination of antiapoptotic and proapoptotic members has to be engaged to either halt or unleash apoptosis. These interactions between Bcl-2 family members are further regulated by phosphorylation status and/or targeted degradation (see Fig. 1 and below and Ref. 3).

2.2 Mitochondrial Outer Membrane Permeability Once Bak/Bax become activated, they increase mitochondrial outer membrane permeability to release proteins contained in the intermembrane space. This event causes mitochondrial membrane potential dissipation and matrix swelling, compromising mitochondrial function and structure. This is the mechanistically least understood step in the pathway, as the exact identity of the pore releasing the proteins is still a matter of debate [6]. A candidate participant and modulator is the permeability transition pore (PTP), a multiprotein complex spanning the outer and inner mitochondrial membrane formed by voltage-dependent anion channel (VDAC), adenine nucleotide translocase, benzodiazepine receptor, cyclophilin D, hexokinase, and other proteins. This pore mediates the exchange of ATP/ADP and other solutes in the normal state. Its participation in apoptosis is supported by the effects of pharmacologic inhibitors and activators for the different components and thus constitutes an attractive target for therapy bypassing Bcl-2–dependent regulation [7]. However, these inhibitors do not abolish all pathways conducive to apoptosis, and apoptosis still occurs in models with genetic deficiencies of these proteins, suggesting that this complex regulates other pores releasing the apoptotic effectors or that—in addition to the PTP—there are multiple mechanisms for proapoptotic effectors release. Regardless of the mechanism, however, once the mitochondrial membrane becomes

166

E. Werner

permeable, the cell is committed to die even if downstream caspase activity is abrogated with pharmacologic inhibitors [8]. Proteins residing in the intermembrane space such as cytochrome c, Smac/ DIABLO, apoptosis-inducing factor (AIF), Endo G, and Omi/HtrA2 are released simultaneously and shortly after Bax/Bak translocation. Free in the cytoplasm, cytochrome c binds Apaf-1, causing a conformational change facilitating ATP/dATP exchange and oligomerization followed by recruitment of caspase 9. This multiprotein complex, the apoptosome, is formed to activate the initiator caspase 9, which cleaves and activates the downstream effector caspases 3 and 7. Smac/DIABLO and Omi/HtrA2 are proteases that inactivate the endogenous cytosolic inhibitors IAPs (inhibitors of apoptosis) to facilitate the activity of the newly formed apoptosome [9]. AIF as well as Endo G translocate to the nucleus and participate in chromatin condensation and lysis [2]. The extent of their contribution to the decision to undergo apoptosis is still unclear, but they are significant contributors to biochemical and morphologic manifestations of apoptosis. AIF is a flavoprotein participating in normal mitochondrial bioenergetic and redox metabolism, thus, its release contributes to mitochondrial dysfunction and to apoptosis irreversibility [10].

2.3 Caspases Are the Executors of Apoptosis Active cell dismantling is brought about by proteases termed caspases, which use a nucleophilic cysteine in the active site to cleave immediately distal to an aspartatecontaining motif present in specific substrates. Among more than 14 members of this family in mammals, only 7 play known roles in apoptosis, while many participate in physiologic functions unrelated to apoptosis. All caspases exist as zymogens in the cytosol and are grouped according to their mechanism of activation. The initiator or apical caspases (caspase 2, 8, 9, and 10) are autocatalytically activated by proximity after associating through their N-terminal domain to a multiprotein activating complex incrementing their catalytic activity by several orders of magnitude. Effector caspases (3, 6, and 7) are activated by initiator caspases–mediated cleavage in a 20- to 30-amino-acid linker sequence [11]. A further level of regulation comes from the activity of cytosolic IAPs, which neutralize caspase activity. Discovered for their ability to suppress apoptosis in baculovirus-infected cells, the IAP proteins are regulated by ubiquitylationmediated proteasome degradation [12].

2.4 Changes in Apoptotic Players Associated with Cancer Unlike oncogenic mutations, changes in the apoptotic pathway are generally insufficient to initiate the carcinogenesis process but play a significant role in tumor promotion. Overexpression of antiapoptotic Bcl-2 or Bcl-xL is observed in multiple cancers and accounts for acquired chemoresistance [3]. Bcl-2 was first identified as an oncogene at a chromosomal breakpoint of t(14:18) in human follicular B-cell lymphoma. Unexpectedly, this oncogene localized to the mitochondrion and did not


167

induce cell proliferation. Accordingly, when Bcl-2 is overexpressed as a transgene in murine models of carcinogenesis, it is not sufficient to elicit oncogenesis but promotes clonogenic survival and growth of the primary tumor when expressed together with an archetypal oncogene. Bcl-2 coexpression with Myc drives tumorigenesis in lymphocytes and in mammary gland. Further studies conditioning Bcl-2 expression in a double transgenic mouse together with Myc shows that switching Bcl-2 expression off does not alter the onset of Myc-induced leukemia but results in tumor remission and prolonged mice survival. These experiments highlight the tumor-promoting function of antiapoptotic factors such as Bcl-2 while simultaneously revealing the persistence and continuous pressure of proapoptotic signals acting on a tumor. Contrary to expectations, deletion of apoptosome components does not contribute to oncogenic transformation. This is evidenced by experiments where the selective deficiency of Apaf-1 or caspase 9 does not enhance lymphomagenesis nor contributes to fibroblast transformation [13]. Analysis of tumor samples reveal that alterations in molecules regulating the apoptosome function are more frequent [9]. From these studies stands out a correlation between the increased expression in tumors of caspases inhibitors (IAPs), especially survivin, with poor disease prognosis. Enhanced survival capabilities become significant at advanced stages of tumor progression during invasion and metastasis. In fact, isolated prostate tumor cells circulating in the bloodstream in animals bearing an orthotopic xenograft of prostate carcinoma cells showed increased resistance to anoikis, a special type of cell death (see Section 3.1) due to induction of the caspase inhibitors XIAP, cIAP2, and survivin, correlating with enhanced metastatic potential [14]. Survivin is a member of the AIF protein family that interferes with caspase activity and is one of the most commonly expressed proteins in transformed cell lines and human tumors, while absent in normal tissues [15]. Moreover, it is a poor prognosis factor correlating with aggressiveness and invasive behavior. Survivin is a target of p53 transcriptional repression (see Section 4.1), therefore it is found upregulated in tumors bearing p53 mutations. Targeting survivin for therapeutic purposes, however, is problematic due to the essential role it plays in mitosis.

3 Deceiving the Surveillance Mechanism by Upregulating Survival Signals The activity of Bcl-2 family members is regulated by information collected from the intracellular and extracellular milieu, both of which undergo significant alterations during carcinogenesis.

3.1 Evading Restrictions Imposed by Cell-Cell Interactions and Tissue Architecture Multicellular organisms have a number of signaling mechanisms geared toward ensuring systematic cell proliferation and differentiation during development and limited clonal expansion of somatic cells during growth and cell renewal in the

168

E. Werner

adult organism. Furthermore, proper tissue organization is maintained by the spatially restricted availability of proliferation and prosurvival signals from the basement membrane and by repressive proapoptotic signals from neighboring cells. Repressive signals from adjacent cells actively promote apoptosis directing the selective death of cell populations to organize organs and tissues during development, as is the case for eliminating abnormal, misplaced, or harmful cells during lymphocyte T and B development or for neuronal pruning during development. In the adult state, controlled apoptosis actively maintains tissue structures. For example, in vivo and in vitro experiments show that apoptosis actively maintains the lumen in the mammary gland acini through a signal that promotes proapoptotic Bim activity. A lumen is maintained even when disorganization is experimentally introduced by forced expression of proliferation inducing proteins leading to abnormal cell multiplication in the acini. Lost of structure by lumen repopulation can occur only when survival signals are delivered either by the coexpression of antiapoptotic proteins or by an oncogene conferring both prosurvival and proliferative signals [16]. Thus, these experiments illustrate the existence of geometric cues built in the organization of tissues, which need to be surmounted very early in tumor development. Cell surface molecules involved in adhesion communicate these controlling signals, acting upon later stages in tumor development as well. In multistage cancer models such as chemically induced squamous cell carcinoma, expression of survivin as a transgene in skin opposes apoptotic signaling from surrounding cells, thereby significantly reducing the spontaneous regression rate of benign lesions [17]. The intercellular adhesion molecule E-cadherin mediates tissue architecture restrictive signals that are active during cancer initiation as well as during metastasis, as downregulation of E-cadherin expression correlates with skin carcinoma progression. Accordingly, restitution of E-cadherin expression in transformed keratinocytes reduces in vivo tumor formation potential [18]. E-cadherin silencing in mammary epithelial cells, accelerates invasive disease and metastasis formation in transgenic mice models, forming tumors resembling invasive lobular carcinoma [19]. An additional source of restrictive signals is the obligatory epithelial cell contact with the basement membrane, which delivers localized proliferation and prosurvival signals, actively opposing apoptosis [20]. When epithelial cells detach and lose contact, they undergo a special form of cell death designated anoikis. This mechanism maintains digestive epithelia homeostasis by inducing death and shedding of enterocytes at the intestinal villi tip. In contrast, it is not observed in highly motile cells such as fibroblasts, where interference with integrin-matrix interactions triggers instead a tissue repair and remodeling phenotype characterized by the activation of NF-␬B and matrix metalloproteinase release [21]. Cell detachment causes integrin disengagement from the extracellular matrix, focal adhesion complex disassembly, and cytoskeleton disorganization, triggering multiple proapoptotic signals. Interference with ␤1 integrin binding to the extracellular matrix using function blocking antibodies is sufficient to induce keratinocyte apoptosis characterized by caspase 8 activation and proapoptotic Bid translocation to mitochondria [22]. Disassembly of focal adhesion complexes disrupts


169

prosurvival signaling mediated by FAK, SRC, and PI3K-AKT kinases, all recruited and activated by organized focal adhesion complexes in substrate-attached cells. Interference with FAK activity induces rapid Bax translocation to mitochondria in mammary gland epithelial cells [23]. The disorganization of microfilament and microtubule cytoskeleton during cell detachment also releases proapoptotic BH3only proteins Bmf and Bim that neutralize Bcl-2 antiapoptotic activity. Proapoptotic Bmf is indirectly associated to the actin cytoskeleton by binding to dynein light chain 2 complexed to myosin V. Proapoptotic Bim associates indirectly to tubulin through dynein light chain 1 sequestered by associated dynein motor complex (see Ref. 24 and references therein). Resistance to anoikis is a key factor to cell survival during invasion, metastasis, and growth in new environments at distal places from the primary tumor. Consequently with their prominent role in anoikis prevention, suppression of FAK activity diminishes resistance to anoikis as well as metastatic potential [25]. An additional strategy used by tumor cells to escape apoptosis is to alter the cell surface repertoire of receptors. The apoptotic balance is modulated either by silencing cell surface molecules mediating proapoptotic signals in non-transformed cells and/or by inducing new molecules promoting cell survival. The receptor for netrin1, NHC5 or “deleted in colon cancer,” induces apoptosis when unengaged, thus its expression is reduced in most colorectal cancer types and many other tumor types [26]. Because survival signaling is integrin-selective, squamous cell skin carcinomas switch the expression of integrin subunit, repressing the anoikis-promoting alphav beta5 and inducing alphav beta6, which renders these cells protected by increasing AKT activation [27]. However, in leukemias the significance of cell adhesion and the environment in the control of apoptosis is indisputably most evident: These normally nonadherent cells exhibit a much higher sensitivity to apoptosis when tested separated from their environment in vitro than when tested in vivo [28]. In fact, ␤1 integrin–dependent adhesion of leukemia cells in vitro is sufficient to induce proapoptotic Bim destabilization and confer resistance to apoptosis. A role for the environment is further supported by experiments showing that multiple myeloma cells become significantly more resistant to chemotherapeutic drugs when adhering to bone marrow stroma [29].

3.2 Autonomous Survival Signals Acquired Through Oncogene Activation and Anti-oncogene Silencing The remarkable effects of oncogene-targeted therapeutic agents used in mouse tumor models reveal the dominant influence of an activated oncogene on tumor phenotype and survival. This has led to the concept of “oncogene addiction” [30]. Most oncogene-activated and tumor suppressor–silenced pathways drive proliferation and promote cell survival by increasing the activity of ERK, PI3K, and AKT kinases [31]. The activation of these kinases downstream of Src, Abl, and EGFR increases prosurvival signaling [30]. Mutations in the effectors of these pathways either increase their catalytic activity or silence the activity of negative regulators.

170

E. Werner

These kinases have a profound effect on the function of the apoptotic cascade by regulating the expression, function, and stability of Bcl2 family members through transcriptional and posttranslational mechanisms. ERK induces the expression of antiapoptotic proteins Bcl-2, Bcl-xL, and Mcl-1, while MAPK-dependent phosphorylation stabilizes Bcl-2 and causes proapoptotic Bim degradation [32]. AKT is the active kinase most frequently found in tumors and mediates prosurvival signals from many oncogenes and growth factors [33]. AKT exerts a dominant prosurvival activity through multiple mechanisms. AKT phosphorylates the BH3 domain of proapoptotic Bad, interfering with apoptotic function and promoting sequestration by cytosolic 14-3-3 proteins [34]. In addition, AKT regulates glucose metabolism thereby amplifying prosurvival signals (see Section 4.2).

4 Survival Effects of Metabolic Changes in Cancer Cells Unrestricted proliferation amid accumulated mutations drive the accrual of profound alterations in cell metabolism and function, in addition to signal transduction. Furthermore, metabolic changes are advanced by stress that follows local hypoxia and metabolite accumulation imposed by the substantial and rapid increment in cell mass. There is mounting evidence to suggest that these abnormalities, which would be a source of stress and proapoptotic signals in normal cells, are not only opposed by tumors but also are key in fostering the acquisition of survival capabilities and are a determinant factor in tumor promotion and expansion [35].

4.1 p53 When normal cells experience alterations in the cell cycle and metabolism, such as reactive oxygen species, hypoxia, double-strand DNA breaks, and abnormal proliferation rate, p53 tumor suppressor is activated to induce growth arrest and/or apoptosis to ensure DNA repair in order to proceed with cell replication. P53 induces apoptosis by multiple mechanisms. It promotes apoptosis through the transcriptional induction of the proapoptotic BH3-only proteins Bad, Bax, Puma, and Noxa. In a second mechanism, p53 activates Bax directly, releasing proapoptotic BH3only proteins sequestered by Bcl-X and promoting Bak oligomerization by displacing bound antiapoptotic Mcl-1. P53 activation can be detected in pre-neoplastic lesions, including colon adenomas, breast carcinomas in situ, and lung hyperplasias in response to replicative stress resulting from deregulated cell proliferation and to double-strand DNA breaks, but preceding any signs of genomic instability or genetic alterations in the p53 pathway [36]. This observation indicates that p53 is activated very early in carcinogenesis by oncogenic stress and acts as an apoptosis-promoter and a barrier for tumor development. This barrier, however, is early averted by inactivating or attenuating p53, as mutations can be detected in early dysplastic lesions of non–small cell lung carcinomas as well as in most human cancers, thereby driving tumor development [37]. Experiments in models


171

of myc-driven lymphomas show that the apoptosis promoting activity is sufficient for effective p53-dependent tumor suppression, while growth arrest function is dispensable [38]. Restitution of p53 expression in advanced tumors causes significant tumor regression by inducing apoptosis in lymphomas and senescence in sarcomas and liver carcinomas [39]. Thus, these experiments underscore the significant role of p53 in suppressing tumor development through the control of the apoptotic pathway.

4.2 Glycolysis A switch to glycolysis as an alternative or additional source for ATP is essential to the cancer phenotype. In normal cells, p53 imposes a limitation on the glycolytic rate by a two-pronged mechanism [40]. P53 induces the expression of a cytochrome oxidase regulatory subunit, synthesis of cytochrome C oxidase, increasing mitochondrial respiration, and induces the expression of Tigar, which controls fructose-2,6-biphosphate intracellular levels, inhibiting glycolysis at the level of 6-phosphofructo-1-kinase activity. During transformation, glycolysis increases by several mechanisms in addition to loss of p53 control. The levels of fructose-2,6biphosphate are regulated by several kinases, including AKT, which phosphorylates and activates 6-phosphofructo-2 kinase to increase glycolysis. Local hypoxic conditions, as well as alterations in metabolism resulting from accrued mutations and from oncogene activation, such as Bcr-Abl, Myc, and K-Ras, can trigger a switch to glycolysis [41]. Glucose utilization is a survival factor per se, conferring resistance to apoptosis proportional to the glycolytic rate [42]. Glycolysis inhibition in Myc-transformed cells precipitates apoptosis, whereas normal cells respond with cell cycle arrest [43]. Although the mechanism is not entirely clear yet, and glycolysis could modulate several steps in the apoptotic pathway, most of the evidence points to an alteration of mitochondrial function downstream of the Bcl-2 regulatory step. Tumor cells increase their glycolytic rate by promoting hexokinase localization to the mitochondrial outer membrane and binding to VDAC, where hexokinase uses ATP produced by oxidative phosphorylation to catalyze the first step of glycolysis [41]. Reconstitution of hexokinase binding to VDAC in liposomes shows that this event also promotes pore closure, thus this event could regulate apoptosis by modulating the mitochondrial outer membrane permeability. Consistent with this possibility, forced disruption of hexokinase association to mitochondria causes cytochrome c release and apoptosis by a mechanism independent of Bax/Bak and Bcl-2. Conversely, forced expression of mitochondria-binding hexokinase neutralizes the ability of activated proapoptotic Bid and Bax to induce apoptosis. AKT prosurvival activities include the upregulation of glucose transporters and maintenance of hexokinase binding and activity at mitochondria [41]. The resulting upregulation of glucose metabolism accounts for most of AKT antiapoptotic function, because this activity can be completely abolished by a hexokinase inhibitor [44].

172

E. Werner

The progressive metabolic dependence of tumors on glycolysis for ATP generation is currently being explored for therapy purposes, revealing glycolysis as an influential control point for apoptosis. The use of glycolysis inhibitors in tumor cell lines induces apoptosis even in multidrug-resistant cells and are most effective in cells with respiratory deficiencies or in those exposed to hypoxia [41, 45].

4.3 Hypoxia Hypoxia can impose stress on cancer cells due to their reduced access to nutrients and oxygen. However, it can also provide a selective pressure in tumors for the expansion of variants that have lost their apoptotic potential. Cellular oxygen tension is sensed by prolyl and asparaginyl hydroxylases, which at normal oxygen levels target hypoxia inducible factor (HIF) for von Hippel–Lindau dependent degradation. When oxygen tension decreases, HIF accumulates and is able to function as a transcription factor, inducing the expression of genes involved in both proapoptotic and antiapoptotic pathways, resulting in the removal of cells that have retained their apoptotic potential and proliferation of cells that have lost their apoptotic potential [46]. HIF-␣ induces cell death through a p53-dependent mechanism and by the direct transcriptional induction of proapoptotic BH3-only proteins Noxa, Bnip3, and Nix [47]. Bnip3 mediates a distinctive form of apoptosis, which is cytochrome c, AIF, and caspase independent, but involves mitochondrial dysfunction, a mechanism compatible with mitochondrial permeability transition and resembling necrosis in some aspects [48]. Consistent with this finding, high expression levels of the proapoptotic factors Bnip3 as well as Nix are found in necrotic areas of tumors and correlate with poor prognosis in cervical cancer, non–small cell lung carcinomas, and pancreatic cancer. Tumor hypoxia is an additional selective pressure for p53 inactivation and other changes leading to apoptosis resistant cells, including Bnip3 silencing. Pancreatic adenocarcinomas switch off the expression of Bnip3 late in progression by promoter hypermethylation. Additionally, hypoxia is a selective pressure at later stages of tumorigenesis by promoting apoptosis resistance through antiapoptotic Mcl-1 expression and thereby promoting the selection of highly metastatic Lewis lung carcinoma cells [49].

4.4 Tumor Acidosis The consequences of increased glucose utilization are lactate production and local acidification, therefore this will be an attribute of all glycolytic tumors. Relatively little is known about the molecular effects of environmental acidosis on tumor physiology. Some evidence suggests that local acidosis could be a barrier to tumor growth in early carcinogenesis. In vitro studies reveal that acidosis can cause growth arrest and can alter sensitivity to chemotherapeutic drugs [50]. Environmental acidosis induces p53-dependent cell death in colon adenocarcinoma cells in vitro. Other


173

evidence suggests that acidosis exacerbates hypoxia-mediated proapoptotic signals. This has been observed in normal cardiomyocytes under ischemia, where hypoxia induces the expression of proapoptotic Bnip3 while acidosis is necessary to accumulate Bnip3 protein in the mitochondria and activate cell death [51]. At later stages of tumor development though, acidosis follows the pattern of all the other factors reviewed here thus far, correlating with tumor aggressiveness and the acquisition of proinvasive properties during, for example, the transition from colon adenomas to invasive cancer and from carcinoma in situ to invasive breast cancer [52]. Human melanoma cells cultured at low pH exhibit increased invasion capabilities in vitro and in vivo. Thus, it is foreseeable that acidosis could be a factor affecting tumor progression. It remains uncertain whether acidosis alters the apoptotic pathway in tumor cells as a factor independent of glycolysis.

4.5 Mitochondria The pathways leading to cell death or survival converge at the mitochondria for integration, interpretation, and amplification. Even though mitochondria are at the center of apoptosis, very little is understood on how mitochondrial function affects apoptosis and vice versa. Most tumors harbor some level of mutations in mitochondrial DNA, many of them affecting oxidative phosphorylation [53]. Nevertheless, the role of mitochondrial respiration on apoptosis remains controversial. Some studies indicate that proper Bak-Bax function is dependent on a functional respiratory chain. Bax function is completely impaired in yeast mutants in any component of the respiratory chain. These studies contrast with other studies showing that ␳ho0 cells maintain the capacity to undergo apoptosis in vitro and in vivo, as well as sensitivity to doxorubicin and growth factor withdrawal [54]. The impact of mitochondrial respiration on apoptosis could be indirect. Acute mitochondrial respiration deficiency induced by culturing cells in EtBr or treatment with respiratory chain inhibitors is sufficient to increase the glycolytic rate, to facilitate cell growth in hypoxic conditions, and to reduce susceptibility to several chemotherapeutic agents. This phenotype induced by reducing mitochondrial respiration correlates with increased AKT activity and can be reversed with inhibitors for this kinase. Thus, this work provides an alternative interpretation to experiments examining respiratory chain function requirement in apoptosis, in addition to suggest that resistance could emerge from a switch to glycolysis rather than from a decrease in respiratory chain function per se [55]. Apoptosis can affect mitochondrial function as well. Many experiments hint to a role for apoptotic players in the control of mitochondrial function. In addition to cytochrome c participation in the respiratory chain, Bcl-x regulates outer membrane permeability to anions and inhibits apoptosis by facilitating ADP transport required to maintain membrane potential. AIF has NADH-dependent oxidase activity, which is required not only for complex I activity but also for anchorage-independent growth and tumor formation. Other mitochondrial changes remain poorly studied, although they may be of potential therapeutic interest. Mitochondria from

174

E. Werner

transformed cells and tumors exhibit a higher membrane potential than do normal cells, correlating with growth rate. Mitochondria isolated from tumors have a higher relative intrinsic resistance to mitochondrial outer membrane permeability. These properties have been exploited to concentrate toxic molecules with a delocalized positive charge in mitochondria and promote apoptosis and mitochondrial dysfunction in transformed cell lines and tumors [56].

5 Concluding Remarks: Implications for Therapy The wide recognition of the important role played by apoptosis evasion in carcinogenesis has fueled the development of therapies targeted at overcoming this resistance and thereby increasing cell susceptibility to chemotherapy [57]. After promising results in preclinical studies, approaches targeting the apoptotic pathway are currently being tested in clinical studies. The results observed illustrate the advantages and problems associated with these strategies. Initial studies with G3139/Oblimersen, an antisense oligonucleotide directed at interfering with antiapoptotic Bcl-2 expression, demonstrated a complete lack of effect as a single therapeutic agent. In combination with fludarabine and cyclophosphamide, which affect DNA metabolism, however, it caused a limited increase in drug sensitivity in relapsed chronic lymphocytic leukemia and advanced melanoma patients. Because drugs directed at proteins involved in apoptosis target a ubiquitous pathway, they have toxic effects in normal cells, causing neutropenia and thrombocytopenia, as well as disrupting angiogenesis by altering endothelial cell function [58]. The limited efficacy of Bcl-2 targeting could be explained by several factors. One possibility is that the results may not correlate with delivery and expression knock-down efficiency. Another significant factor conferring resistance could be the presence of other functionally compensating antiapoptotic family members. Thus, rather than disrupting the function of a single member of the family, interfering with BH-3 domain activity widens the inhibitory range of the potential drug. The drug ABT-737 is a potent BH3 peptidomimetic, antagonizing the activity of the antiapoptotic factors Bcl-2, Bcl-xL, and Bcl-w. However, this drug has proved ineffective when Bcl-2 is phosphorylated or when antiapoptotic Mcl-1 is present [59]. An alternative strategy is to bypass Bcl-2–dependent regulatory mechanism and target mitochondria directly to trigger apoptosis and release intermembrane effectors. Using this approach, the use of a Smac/Diablo mimetic to activate caspases directly has been successful in preclinical studies, overcoming resistance induced by increased Bcl-2 expression, autocrine growth factors, and adhesion in multiple myeloma [60]. Combination therapies would benefit altogether by considering environmental and bioenergetic factors affecting the apoptotic pathway, offering additional targets and aiding in proapoptotic therapy selectivity. In the case of ABT-737, therapeutic efficacy could be improved by using glucose metabolism inhibitors, which would cause antiapoptotic Mcl-1 destabilization, as turnover and activity is controlled by glucose-dependent inhibition of GSK-3 activity [41]. Furthermore, the tumor


175

characteristic alterations in metabolism could impart selectivity to apoptosistargeted therapy. Inhibition of glycolysis causes massive apoptosis accentuated in tumors with mitochondrial respiration deficiency and in hypoxic cells [45]. Reduction of mitochondrial respiration with As2 O3 increases drug-sensitivity as well. Furthermore, local acidosis could be exploited to activate chemotherapeutic drugs locally [50]. Thalidomide, a drug known for immunomodulatory effects, disrupts tumor homeostasis and interaction with the environment. It successfully induced apoptosis in refractory multiple myeloma cells and decreased the resistance to chemotherapeutic drugs in vitro, preventing adhesion to bone marrow cells [61]. In prostate cancer, this drug restores interactions with the environment by increasing E-selectin expression [62]. Given these results, thalidomide is currently being tested in phase II trials for refractory multiple myeloma and solid tumors. Therapies aimed at modifying the cellular metabolism in order to increase susceptibility to apoptosis address tumor heterogeneity, an additional issue arising in clinical circumstances and poorly reproduced by mouse models. This is a significant source of variability, accounting in large part for the disappointing inconsistency observed in the results obtained from preclinical and clinical studies, as well as the difficulties in correlating disease prognosis with apoptotic marker expression in tumor samples. Tumor models of clonal origin, where oncogenesis is driven by either the expression of a single oncogene in transgenic mice or by the graft of a cell line originally isolated as a clone from a tumor, are ideal models to dissect the role of particular components of a pathway, regulating factors in specific stages of disease, for evaluation of possible therapeutic targets. However, tumor samples from patients show high heterogeneity in expression levels of apoptosis markers in different areas of the same tumor, coexisting with areas of high apoptotic index. Single cell profiling reveals that heterogeneity determines differential response to cytokines and drugs and has been proposed to be a key factor for promoting disease progression in esophageal cancer. Thus, because drugs modulating cell metabolism do not target specific oncogenic pathways or mutations, they are expected to be more effective in overcoming tumor heterogeneity as well as in targeting a wider range of tumors. Acknowledgments This work was supported by an award from the American Heart Association and by grant number K22CA127136 from the National Cancer Institute.

References 1. Igney FH, Krammer PH. Death and anti-death: tumour resistance to apoptosis. Nat Rev Cancer 2002; 2:277–288. 2. Danial NN, Korsmeyer SJ. Cell death: critical control points. Cell 2004; 116:205–219. 3. Adams JM, Cory S. The Bcl-2 apoptotic switch in cancer development and therapy. Oncogene 2007; 26:1324–1337. 4. Zong WX, Lindsten T, Ross AJ, MacGregor GR, Thompson CB. BH3-only proteins that bind pro-survival Bcl-2 family members fail to induce apoptosis in the absence of Bax and Bak. Genes Dev 2001; 15:1481–1486.

176

E. Werner

5. Letai A, Bassik MC, Walensky LD, Sorcinelli MD, Weiler S, Korsmeyer SJ. Distinct BH3 domains either sensitize or activate mitochondrial apoptosis, serving as prototype cancer therapeutics. Cancer Cell 2002; 2:183–192. 6. Brenner C, Grimm S. The permeability transition pore complex in cancer cell death. Oncogene 2006; 25:4744–4756. 7. Bouchier-Hayes L, Lartigue L, Newmeyer DD. Mitochondria: pharmacological manipulation of cell death. J Clin Invest 2005; 115:2640–2647. 8. Chipuk JE, Green DR. Do inducers of apoptosis trigger caspase-independent cell death? Nat Rev Mol Cell Biol 2005; 6:268–275. 9. Schafer ZT, Kornbluth S. The apoptosome: physiological, developmental, and pathological modes of regulation. Dev Cell 2006; 10:549–561. 10. Vahsen N, Cande C, Briere JJ, et al. AIF deficiency compromises oxidative phosphorylation. EMBO J 2004; 23:4679–4689. 11. Boatright KM, Salvesen GS. Mechanisms of caspase activation. Curr Opin Cell Biol 2003; 15:725–731. 12. Riedl SJ, Shi Y. Molecular mechanisms of caspase regulation during apoptosis. Nat Rev Mol Cell Biol 2004; 5:897–907. 13. Scott CL, Schuler M, Marsden VS, et al. Apaf-1 and caspase-9 do not act as tumor suppressors in myc-induced lymphomagenesis or mouse embryo fibroblast transformation. J Cell Biol 2004; 164:89–96. 14. Berezovskaya O, Schimmer AD, Glinskii AB, et al. Increased expression of apoptosis inhibitor protein XIAP contributes to anoikis resistance of circulating human prostate cancer metastasis precursor cells. Cancer Res 2005; 65:2378–2386. 15. Ambrosini G, Adida C, Altieri DC. A novel anti-apoptosis gene, survivin, expressed in cancer and lymphoma. Nat Med 1997; 3:917–921. 16. Debnath J, Mills KR, Collins NL, Reginato MJ, Muthuswamy SK, Brugge JS. The role of apoptosis in creating and maintaining luminal space within normal and oncogene-expressing mammary acini. Cell 2002; 111:29–40. 17. Allen SM, Florell SR, Hanks AN, et al. Survivin expression in mouse skin prevents papilloma regression and promotes chemical-induced tumor progression. Cancer Res 2003; 63:567–572. 18. Navarro P, Gomez M, Pizarro A, Gamallo C, Quintanilla M, Cano A. A role for the E-cadherin cell-cell adhesion molecule during tumor progression of mouse epidermal carcinogenesis. J Cell Biol 1991; 115:517–533. 19. Derksen PW, Liu X, Saridin F, et al. Somatic inactivation of E-cadherin and p53 in mice leads to metastatic lobular mammary carcinoma through induction of anoikis resistance and angiogenesis. Cancer Cell 2006; 10:437–449. 20. Boudreau N, Sympson CJ, Werb Z, Bissell MJ. Suppression of ICE and apoptosis in mammary epithelial cells by extracellular matrix. Science 1995; 267:891–893. 21. Kheradmand F, Werner E, Tremble P, Symons M, Werb Z. Role of Rac1 and oxygen radicals in collagenase-1 expression induced by cell shape change. Science 1998; 280:898–902. 22. Marconi A, Atzei P, Panza C, et al. FLICE/caspase-8 activation triggers anoikis induced by beta1-integrin blockade in human keratinocytes. J Cell Sci 2004; 117:5815–5823. 23. Gilmore AP, Metcalfe AD, Romer LH, Streuli CH. Integrin-mediated survival signals regulate the apoptotic function of Bax through its conformation and subcellular localization. J Cell Biol 2000; 149:431–446. 24. Puthalakath H, Villunger A, O Reilly LA, et al. Bmf: a proapoptotic BH3-only protein regulated by interaction with the myosin V actin motor complex, activated by anoikis. Science 2001; 293:1829–1832. 25. Duxbury MS, Ito H, Zinner MJ, Ashley SW, Whang EE. Focal adhesion kinase gene silencing promotes anoikis and suppresses metastasis of human pancreatic adenocarcinoma cells. Surgery 2004; 135:555–562. 26. Mazelin L, Bernet A, Bonod-Bidaud C, et al. Netrin-1 controls colorectal tumorigenesis by regulating apoptosis. Nature 2004; 431:80–184.


177

27. Janes SM, Watt FM. Switch from alphavbeta5 to alphavbeta6 integrin expression protects squamous cell carcinomas from anoikis. J Cell Biol 2004; 166:419–431. 28. Munk Pedersen I, Reed J. Microenvironmental interactions and survival of CLL B-cells. Leuk Lymphoma 2004; 45:2365–2372. 29. Nefedova Y, Landowski TH, Dalton WS. Bone marrow stromal-derived soluble factors and direct cell contact contribute to de novo drug resistance of myeloma cells by distinct mechanisms. Leukemia 2003; 17:1175–1182. 30. Sharma SV, Gajowniczek P, Way IP, et al. A common signaling cascade may underlie “addiction” to the Src, BCR-ABL, and EGF receptor oncogenes. Cancer Cell 2006; 10:425–435. 31. Vogelstein B, Kinzler KW. Cancer genes and the pathways they control. Nat Med 2004; 10:789–799. 32. Breitschopf K, Haendeler J, Malchow P, Zeiher AM, Dimmeler S. Posttranslational modification of Bcl-2 facilitates its proteasome-dependent degradation: molecular characterization of the involved signaling pathway. Mol Cell Biol 2000; 20:1886–1896. 33. McCormick F. Cancer: survival pathways meet their end. Nature 2004; 428:267–269. 34. Zhou XM, Liu Y, Payne G, Lutz RJ, Chittenden T. Growth factors inactivate the cell death promoter BAD by phosphorylation of its BH3 domain on Ser155. J Biol Chem 2000; 275: 25046–25051. 35. Hammerman PS, Fox CJ, Thompson CB. Beginnings of a signal-transduction pathway for bioenergetic control of cell survival. Trends Biochem Sci 2004; 29:586–592. 36. Bartkova J, Horejsi Z, Koed K, et al. DNA damage response as a candidate anti-cancer barrier in early human tumorigenesis. Nature 2005; 434:864–870. 37. Levine AJ. p53, the cellular gatekeeper for growth and division. Cell 1997; 88:323–331. 38. Schmitt CA, Fridman JS, Yang M, Baranov E, Hoffman RM, Lowe SW. Dissecting p53 tumor suppressor functions in vivo. Cancer Cell 2002; 1:289–298. 39. Xue W, Zender L, Miething C, et al. Senescence and tumour clearance is triggered by p53 restoration in murine liver carcinomas. Nature 2007; 445:656–660. 40. Green DR, Chipuk JE. p53 and metabolism: inside the TIGAR. Cell 2006; 126:30–32. 41. Pelicano H, Martin DS, Xu RH, Huang P. Glycolysis inhibition for anticancer treatment. Oncogene 2006; 25:4633–4646. 42. Vander Heiden MG, Plas DR, Rathmell JC, Fox CJ, Harris MH, Thompson CB. Growth factors can influence cell growth and survival through effects on glucose metabolism. Mol Cell Biol 2001; 21:5899–5912. 43. Shim H, Chun YS, Lewis BC, Dang CV. A unique glucose-dependent apoptotic pathway induced by c-Myc. Proc Natl Acad Sci USA 1998; 95:1511–1516. 44. Gottlob K, Majewski N, Kennedy S, Kandel E, Robey RB, Hay N. Inhibition of early apoptotic events by Akt/PKB is dependent on the first committed step of glycolysis and mitochondrial hexokinase. Genes Dev 2001; 15:1406–1418. 45. Xu RH, Pelicano H, Zhou Y, et al. Inhibition of glycolysis in cancer cells: a novel strategy to overcome drug resistance associated with mitochondrial respiratory defect and hypoxia. Cancer Res 2005; 65:613–621. 46. Pouyssegur J, Dayan F, Mazure NM. Hypoxia signalling in cancer and approaches to enforce tumour regression. Nature 2006; 441:437–443. 47. Kim JY, Ahn HJ, Ryu JH, Suk K, Park JH. BH3-only protein Noxa is a mediator of hypoxic cell death induced by hypoxia-inducible factor 1alpha. J Exp Med 2004; 199:113–124. 48. Vande Velde C, Cizeau J, Dubik D, et al. BNIP3 and genetic control of necrosis-like cell death through the mitochondrial permeability transition pore. Mol Cell Biol 2000; 20: 5454–5468. 49. Koshikawa N, Maejima C, Miyazaki K, Nakagawara A, Takenaga K. Hypoxia selects for highmetastatic Lewis lung carcinoma cells overexpressing Mcl-1 and exhibiting reduced apoptotic potential in solid tumors. Oncogene 2006; 25:917–928. 50. Reichert M, Steinbach JP, Supra P, Weller M. Modulation of growth and radiochemosensitivity of human malignant glioma cells by acidosis. Cancer 2002; 95:1113–1119.

178

E. Werner

51. Kubasiak LA, Hernandez OM, Bishopric NH, Webster KA. Hypoxia and acidosis activate cardiac myocyte death through the Bcl-2 family protein BNIP3. Proc Natl Acad Sci USA 2002; 99:12825–12830. 52. Abbey CK, Borowsky AD, McGoldrick ET, et al. In vivo positron-emission tomography imaging of progression and transformation in a mouse model of mammary neoplasia. Proc Natl Acad Sci USA 2004; 101:11438–11443. 53. Carew JS, Huang P. Mitochondrial defects in cancer. Mol. Cancer 2002; 1:9–21. 54. Jacobson MD, Burne JF, King MP, Miyashita T, Reed JC, Raff MC. Bcl-2 blocks apoptosis in cells lacking mitochondrial DNA. Nature 1993; 361:365–369. 55. Vander Heiden MG, Chandel NS, Li XX, Schumacker PT, Colombini M, Thompson CB. Outer mitochondrial membrane permeability can regulate coupled respiration and cell survival. Proc Natl Acad Sci USA 2000; 97:4666–4671. 56. Fantin VR, Berardi MJ, Scorrano L, Korsmeyer SJ, Leder P. A novel mitochondriotoxic small molecule that selectively inhibits tumor cell growth. Cancer Cell 2002; 2:29–42. 57. Reed JC. Drug insight: cancer therapy strategies based on restoration of endogenous cell death mechanisms. Nat Clin Pract Oncol 2006; 3:388–398. 58. Zeitlin BD, Joo E, Dong Z, et al. Antiangiogenic effect of TW37, a small-molecule inhibitor of Bcl-2. Cancer Res 2006; 66:8698–8706. 59. Konopleva M, Contractor R, Tsao T, et al. Mechanisms of apoptosis sensitivity and resistance to the BH3 mimetic ABT-737 in acute myeloid leukemia. Cancer Cell 2006; 10:375–388. 60. Chauhan D, Neri P, Velankar M, et al. Targeting mitochondrial factor Smac/DIABLO as therapy for multiple myeloma (MM). Blood 2007; 109:1220–1227. 61. Corso A, Ferretti E, Lunghi M, et al. Zoledronic acid down-regulates adhesion molecules of bone marrow stromal cells in multiple myeloma: a possible mechanism for its antitumor effect. Cancer 2005; 104:118–125. 62. Efstathiou E, Troncoso P, Wen S, et al. Initial modulation of the tumor microenvironment accounts for thalidomide activity in prostate cancer. Clin Cancer Res 2007; 13:1224–1231.

Index

A Abate, C., 64 Abbey, C. K., 173 Abu-Hamad, S., 12 Acetoin, 106, 107 Acin-Perez, R., 128 Acker, T., 135 Aconitase in Krebs cycle, 4 Adams, J. M., 36, 164, 166 Adebanjo, O. A., 63 Adida, C., 167 Agani, F., 82 Ahern, K. G., 106, 113 Ahn, H. J., 172 Ainscow, E. K., 115 Akimoto, M., 59, 123 Akt/PBK (protein kinase B) pathway, 8 Alam, N. A., 25, 59 Alarcon, R., 138 Alazard, N., 58, 62 Alderson, M. R., 149 Aldolase (ALD), 84, 85 Alessi, D. R., 12 Allan, S., 94 Allen, S. M., 168 Almodovar, C. R., 114 Alonso, A. M., 6, 10 Alonso, C., 30 Alquier, T., 12 Altenberg, B., 6 Altieri, D. C., 167 Ambrosini, G., 167 Amir, S., 136 Amstad, P., 64 Amuthan, G., 48–50, 58, 61–63, 67, 119 Ananadatheerthavarada, H. K., 48–50, 58, 61–63, 67 Anbazhagan, R., 62

Anderson, S., 121 Andersson, U., 61 Angiogenesis, 37–39 Antiapoptotic genes, 66–67 Antioxidant antioxidant response element (ARE), 38 defense pathway, 65 enzyme levels, 105 Aoyama, N., 7 Apoptosis, 36, 47, 162–163 Bcl-2 family proteins, 164–165 caspases and, 166 Appel, L. J., 2 Apse, K., 97 Armour, S. M., 24 Arnold, R. S., 35, 38 Arora-Kuruganti, P., 63 Aryl hydrocarbon nuclear translocator (ARNT), 135 Asham, A. M., 134 Ashley, S. W., 169 Aslan, M., 27 Astuti, D., 23, 59, 136 Attardi, G., 60 Atzei, P., 168 Augenlicht, L. H., 126 Avadhani, N. G., 48–50, 58, 61–63, 67, 119 Azoulay-Zohar, H., 12 B Babior, B. M., 20 Bach, D., 66 Bader, S., 139 Bae, S. H., 24, 80, 81 Bae, Y. S., 34 Baeuerle, P. A., 64 Baggett, B. K., 8 Baggetto, L. G., 9, 106, 107

179

180 Baldi, P., 25, 27, 47, 127 Bale, A. E., 46 Ballard, J. W. O., 108 Bamezai, R. N., 128 Bandelt, H. J., 22, 47, 126–128 Bankier, A. T., 121 Baranov, E., 171 Barrell, B. G., 121 Barrett, J. N., 60, 61 Bartkova, J., 170 Bartnik, E., 121, 123, 126 Bassik, M. C., 165 Battersby, B. J., 128 Bauer, C., 78 Bauer, D. E., 8, 79, 113 Baumann, A. K., 9, 58, 64, 127 Bayascas, J. R., 12 Baysal, B. E., 9, 20, 23, 59 Beach, D., 94, 95 Beasley, N. J., 137 Bellamy, W. T., 29 Bell, E. L., 38 Benit, P., 24, 25 Benizri, E., 77 Bensaad, K., 9 Ben-Shlomo, I., 115, 116 Berardi, M. J., 174 Berchner-Pfannschmidt, U., 75, 77 Berezovskaya, O., 167 Berman, S. B., 47 Bernard, D., 132 Bernet, A., 169 Berra, E., 77 Bhat, A. K., 128 Bhat-Nakshatri, P., 64 Bhujwalla, Z. M., 137 Bhushan, A., 148 Bianchi, M. S., 124, 125 Bianchini, F., 107 Bianchi, N. O., 124, 125 Bindra, R. S., 134 Birch-Machin, M., 123 Birnbaum, M. J., 7 Birrer, M. J., 64 Bishopric, N. H., 114, 173 Bissell, M. J., 168 Biswas, G., 48–50, 58, 61–63, 67, 119 Bi, X., 11 Blachly-Dyson, E., 11 Blanco-Rivero, A., 2, 9–11 Blasco, M. A., 94 Bluyssen, H. A., 140 Boatright, K. M., 166

Index Bodyak, N. D., 57, 58, 125, 126 Bond, J. D., 63 Bonhoeffer, S., 2, 5, 9 Bonicalzi, M. E., 139 Bonni, A., 41 Bonod-Bidaud, C., 169 Bonora, E., 127 Booker, L. M., 128 Borisenko,G. G., 107 Borowsky, A. D., 173 Bottoni, P., 50, 51 Bouchier-Hayes, L., 165 Boudreau, N., 168 Bouillaud, F., 21 Boulianne, G. L., 96 Bourgeron, T., 22, 25 Bove, K., 38 Bowden, G. T., 57, 64, 65 Boyer, P. D., 3 Bracale, R., 107 Braithwaite, K. L., 92 Brand, M. D., 115 Brandon, M., 25, 27, 47, 127 Brands, M., 2 Braunstein, L. D., 97 Breitschopf, K., 170 Brenner, C., 165 Bretscher, P. A., 149 Briehl, M. M., 29 Briere, J. J., 12, 20, 23–25, 166 Brigelius-Flohe, R., 64 Briggs, J., 140 Briones, P., 66 Bromley, L., 63 Brown, D. T., 124 Brown, S. J., 46 Brundtland, G. H., 148 Brune, B., 138 Brüne, B., 79 Brunelle, J. K., 38 Brunet, A., 41 Bryk, J., 128 Buecher, T., 22 Bulavin, D. V., 11 Bunn, H. F., 136 Burdon, R. H., 63 Buricchi, F., 37 Burk, D., 5 Burne, J. F., 173 Bustamante, E., 8, 27 Butow, R. A., 67, 68 Butts, B., 65 Buzzai, M., 8, 79

Index C Caballero, O. L., 57 Caenorhabditis elegans and mev1 mutant, 23 Cairns, R. A., 10 Calcium signaling pathway, 62, 63 Calderwood, S. K., 48 Calle, E. E., 107 Camenisch, G., 138 Cameron, G., 133 Camley, R. E., 152 Campanella, C., 48 Campisi, J., 95, 108 Cancer, 46 apoptotic pathway in, 166–167 cells, 145 Achilles’ heel of, 29 AKT activation in, 78–79 bioenergetic signature of, 9 HIF-1α expression in, 88 Notch-3 and carbonic anhydrase IX (CA-IX), 42 PTENoxidative inactivation and AKT signaling pathway, 41 retrograde signaling, 60–67 Warburg effect in, 8, 46 Warburg phenotype of, 10–11 COXI mtDNA mutations, 58 drug treatment and Fas-induced tumor cell death, 150–157 fumarate hydratase (FH) mutations, 88 G3139/Oblimersen studies with, 174 HKII transcription, 83 homoplasmy and heteroplasmy, 125 immune response in, 147–148 system and, 149 metabolic changes in cell, survival effects glycolysis, 171–172 hypoxia, 172 mitochondria and, 173–174 p53 tumor suppressor, 170–171 tumor acidosis, 172–173 mitochondrial genome instability (mtGI) in, 124–125 OXPHOS defects in, 56 dysfunction and, 66 Smac/Diablo mimetic use in, 174–175 surveillance mechanism cell-cell interactions and tissue architecture, 167–169 oncogene activation and anti-oncogene silencing, 169–170

181 thalidomide drug for, 175 therapeutic possibilities, 158–159 Cande, C., 12, 20, 166 Cano, A., 168 Cappello, F., 48 Carcinogenesis aerobic respiration and energy generation, 104 calorie restriction inhibition energy expenditure and intake, 108 metabolic disorders, 107 UDPGlcNAc decreased levels, 109 chronic infection and signaling molecules, 110 diversion of pyruvate mitochondria respiration (PDH) and fermentation (PDC), 106 TCA cycle, 106 electron transport modulation NADH/cyt-C electron transport pathway, 110 glycolysis involvement in, 93 immortalization and transformation, 91–92 induced apoptosis, 111–112 metabolic modulation of, 103 organelle and cell membrane perturbations, 107 OXPHOS phenotype and apoptosis ATP synthase and suppression of, 104 reactive oxygen species and cell-signaling pathways, 105 mitogenic signals, 105 Warburg model, 113–115 Cardile, A., 107 Cardiolipin (CL), 107 Carew, J. S., 9, 120, 122–125, 173 Carling, D., 12 Carmeliet, P., 138 Carney, D., 126 Carothers, A. D., 126, 128 Carotid body PGL, 23 Carracedo, A., 22, 47, 126–128 Carter, M. J., 107 Casado, E., 9–11 Casimiro, M. C., 11 Caspases and cellular proteins, 36 Castilla, J., 30 C2C12 rhabdomyosarcoma, ryanodine receptor-1 (RyR-1) and-2 (RyR-2) genes, 63 Cellular respiration tumor suppressor genes p53, 137–138

182 Cellular respiration (cont.) von Hippel–Lindau syndrome, 138–140 and Warburg hypothesis, 132–133 Cepeda, V., 30 Cerutti, P., 64 Cerutti, P. A., 63 Chan, D. C., 66 Chandel, N. S., 173 Chandra, D., 48 Chang, G. W., 136 Chang, H. J., 10 Chang, I., 11 Chang, T. S., 34 Chan, T. L., 57 Chauhan, D., 174 Chemodectoma, see Carotid body PGL Chemotherapeutic-resistant tumor cells sequester cytosolic calcium (Cacyt ), 48 Chen, E. I., 52 Chen, G., 6, 10 Cheng, G., 38 Cheng, M. L., 97 Cheng, W. C., 47 Chen, H., 66 Chen, X., 22 Chen, Z., 11 Chen, Z. J., 65, 110 Cheung, A. N., 57 Cheung, H. S., 61 Chiarle, R., 23 Chiarugi, P., 37 Chieco, P., 140 Chilov, D., 138 Chinnery, P. F., 124–126, 128 Chipuk, J. E., 171 Chi, S. M., 135 Chittenden, T., 170 Cho, E. J., 140 Choi, J., 100 Choi, K. Y., 107 Chomyn, A., 66 Chretien, D., 22, 25 Chu, G., 42 Chun, Y. S., 171 Chu, S. C., 41 Cinalli, R. M., 8 Ciocca, D. R., 48 Cisplatin-resistant ovarian carcinoma, 38 Cizeau, J., 172 Clayton, D. A., 22 Clear cell renal carcinoma (CCRC) cells, 81 Clifford, S. C., 136, 139

Index Cockman, M. E., 139 Cohen, A., 110 Cohen, C., 7 Cohn, M., 149 Colburn, N. H., 65 Coller, H. A., 57, 58, 125, 126 Colombini, M., 173 Combs, C. A., 42 Connor, K. M., 37, 38, 40 Contractor, R., 174 Copeland, W. C., 121, 122, 124 Corral-Debrinski, M., 22 Corso, A., 175 Cory, S., 164, 166 Coughlin, S. S., 107 Couplan, E., 21 Cozzi, V., 107 Crawford, D., 64 Crisp, W., 152 Crosby, M. E., 134 Cucuianu, A., 109 Cuezva, J. M., 2, 6, 7, 9–12 Curran, T., 64 Cycloheximide (CHX), 36 Cytochrome c oxidase (COX), 9 Czarnecka, A. M., 48, 121, 123, 126 D Dachs, G. U., 8 Dalgard, C. L., 8 Dallol, A., 23, 59 Dalton, W. S., 169 Dang, C. V., 7, 10, 46, 92, 134, 171 Dang, G., 7 Danial, N. N., 12, 164, 166 Darin, N., 22 Darvishi, K., 128 Dasgupta, J., 37 Dayan, F., 172 Debatin, K. M., 30 Decorps, M., 7 de Heredia, M. L., 6, 7, 9–11 Dellian, M., 134 Delsite, R., 62 De, Marzo, A. M., 80 Deml, E., 9 Demont, J., 9, 10, 58, 59 Denko, N. C., 10, 134 de Paulsen, N., 139 De Pinho, R. A., 94 Derksen, P. W., 168 Desbaillets, I., 138 Desbarats, J., 149

Index Desvergne, B., 51 Dey, R., 11, 67 Diala, E. S., 66 Dickinson, D., 96 Dictyostelium discoideum and phosphoinositide (3, 4, 5) 3-phosphate (PtdIns[3, 4, 5]P3), 40 Dietrich, A., 22 Dimmeler, S., 170 Ditsworth, D., 113 Dolde, C., 7, 93 Domann, F. E., 64 Domann, F. E. Jr., 64 Don, A. S., 123 Dong, Z., 64, 174 Dor, Y., 138 Douglas-Jones, A., 61 Drago, I., 62 Droge, W., 47 Drosophila simulans, 108 Dubik, D., 172 Dujon, B., 22 Dukes, I. D., 115 Du, M., 10, 113 Dunn, G. P., 110 Dunning, S. P., 136 Du, X., 113 Duxbury, M. S., 169 E Earl, A., 66 Ebert, B. L., 8 Edelstein, D., 113 Edens, W. A., 38 Efstathiou, E., 175 Egea, G., 11 Electron transport chain (ETC) cell death and proliferation, 28 in mammalian cell, 56 modulation, 110–111 El Ghouzzi, V., 25 Elia, A. J., 96 Elson, J., 125 Elstrom, R. L., 8, 79 Endo, T., 61 Endothelial PAS-domain protein 1 (EPAS1), 79 Energy-deprived HepG2 hepatoma cells, mitochondrial biogenesis and OXPHOS, 108 Enolase (ENO), 85 Enriquez, J. A., 119, 121, 125 Epstein, C. B., 68, 137

183 Erlacher, M., 137 Ernest, I., 135 Erythropoietin (EPO) and VEGF, role in hypoxia, 80 E3 ubiquitin ligases and HIF-1α, 78 Eukaryotic peroxiredoxins (PRXs), 35 Evan, G. I., 150 Evans, I. H., 66 F Fabregat, I., 10–12 Faddy, H. M., 62 Faller, D. V., 135 Fang, H. M., 8, 133, 135, 136 Fang, W. G., 135 Fan, M., 65 Fantin, V. R., 46, 49, 51, 174 Favier, J., 23–25 Felty, Q., 50 Fernandez, A., 135 Fernandez, P. L., 6, 9, 11 Fernandez-Silva, P., 119, 121, 125, 128 Ferrans, V. J., 34 Ferrell, R. E., 9, 20, 23, 59 Ferretti, E., 175 Fibrate/thiazolidinedione-induced differentiation, 51 Finch, J. S., 64 Finkel, T., 34, 42 Fink, K., 124 Firth, J. D., 8 Flier, J. S., 7 Fliss, M. S., 57 Flohe, L., 64 Floodgate hypothesis, 35 Florell, S. R., 168 Floyd, R. A., 35 Foekens, J. A., 140 Foker, J., 11 Folkman, J., 134 Fontana, L., 10 Fontana, L. A., 134 Foo, T. W., 11 Forbes, R. A., 82, 136 Forkhead transcription factors (FoxO), 41–42 Fornace, A. J. Jr., 11 Fox, C. J., 8, 170, 171 Franc, B., 59 Francis, M. J., 63 Frataxin and Fe-S cluster biogenesis, 11 Freedman, B. D., 63 Freedman, D. A., 137 French, S., 42

184 Fridman, J. S., 171 Fuertes, M. A., 30 Fuhrer, J. P., 66 Fulda, S., 30 G Gabrielson, E., 62 Gajewski, C. D., 64 Gajowniczek, P., 169 Galluzzi, L., 30 Gamallo, C., 168 Gambhir, S. S., 7 GAPDH overexpression and human prostate adenocarcinoma cells (cell line LNCap), 85 Garber, K., 46, 137 Garća-Garća, E., 6, 7, 9, 10 Garrido, C., 20 Gasparre, G., 127 Gassmann, M., 78, 138 Gatenby, R. A., 7, 46, 50, 93, 115 Gatter, K. C., 8 Gauthier-Rouviere, C., 135 Gebbink, M. F., 140 Genes encoding complex II mutations, tumor and cancer formation, 23 Geromel, V., 22, 25 Giacomello, M., 62 Giardia lamblia, 4 Giardina, B., 50 Giatromanolaki, A., 8 Gil, J., 96, 97, 132 Gill, V., 63 Gillies, R. J., 7, 8, 46, 50, 93, 115 Gilmore, A. P., 169 Gimenez-Roqueplo, A. P., 23, 24 Ginkgo biloba, proapoptotic molecules expression, 37 Ginouvés, A., 77 Giorgio, M., 42 Giovannini, C., 42 Glazer, P. M., 134 Gleadle, J. M., 8 Glinskii, A. B., 167 Glucose-6-phosphate dehydrogenase (G6PD) activity on cell proliferation, 97 Glucose phosphate isomerase (GPI), 83 Glucose transporters (GLUT1 to GLUT5), 87 Glyceraldehyde phosphate dehydrogenase (GAPDH), 85 Glycolysis, 2 aldolase (ALD), 84–85 apoptotic pathways and, 114

Index ATP levels and OXPHOS, 111 cell cycle and apoptotic pathway, regulation, 112 and cellular senescence, 94–95 enolase (ENO), 85 glucose phosphate isomerase (GPI), 83 glyceraldehyde phosphate dehydrogenase (GAPDH), 85 glycolysis-dependent phenotype of malignancies, 8 hexokinases (HKs), 82–83 hypoxia inducible factor 1 alpha (HIF-1α), 8 induction of hypoxia path by succinate through stabilization of, 26 hypoxic conditions and, 92–93 lactate dehydrogenase and lactate carrier MCT4, 86–87 LMW-PTP activity and, 37 malignant growth and, 5 Nox1-dependent generation of H2 O2 , 38 2-OG dioxygenase, 75–76 oncogene and, 93–94 organisms life span and calorie restriction, 98–99 Sir2 as longevity gene in Yeast, 97–98 phosphofructokinase 1 (PFK-1), 84 phosphoglycerate kinase (PGK), 85 pyruvate dehydrogenase (PDH), 86 pyruvate kinase (PK), 86 as radical scavenger, 97 regulatory mechanism of p53 and, 99–100 and PGM, 100–101 Sp1 and Bnip3, 106 TIGAR acts as inhibitor of, 9 Glycolytic malignant cells, 106 Gnarra, J., 139 Gnarra, J. R., 139 Goeddel, D. V., 36 Goldberg, M. A., 136 Goldblatt, H., 133 Goldin, A., 150 Golik, P., 121, 123, 126, 128 Golstein, P., 20, 149 Gomez, M., 168 Gordan, J. D., 24, 113 Gorgoglione, V., 110 Goswami, P. C., 36 Gottlieb, E., 47, 121 Gottlieb, R. A., 20 Gottlob, K., 171 Goulet, R. J. Jr., 64

Index Govindarajan, B., 7, 8 Graeber, T. G., 137 Gramlich, J. L., 41 Gramm, C. F., 12 Greco, M., 60 Greenawalt, J. W., 57 Greenblatt, M. S., 137 Green, D. R., 4, 19, 20, 150, 171 Greim, H., 9 Greulich, K. O., 6 Grimm, S., 165 Gross, A., 11 Groszer, M., 10 Groulx, I., 139 Grover-McKay, M., 7 Guan, K. L., 12 Guarente, L., 98 Guppy, M., 5 Gupta, A., 57, 64, 65 Gutman, M., 22 Gu, W., 132 Gu, Y., 110 Guyetant, S., 59 Gyllensten, U., 127 H Habano, W., 9 Habermacher, G. M., 128 Haendeler, J., 170 Hagler, J., 65 Hale, W. T., 68 Hammerman, P. S., 8, 170 Hammond, E., 138 Hanahan, D., 11, 92, 134 Hanks, A. N., 168 Hannan, R. L., 135 Hannon, G. J., 94 Haraguchi, K., 61 Hardie, D. G., 12 Harman, D., 94 Harper, M-E., 149, 150, 154–158 Harris, A. L., 8, 134, 140 Harris, M. H., 11, 108, 171 Hashida, M., 27 Haspel, H. C., 7 Hayflick, L., 94 Hay, N., 171 Heat shock proteins (HSPs), 48 Heerdt, B. G., 126 Heese, C., 9 Helicobacter pylori, 109 Helmlinger, G., 134 Hendrix, M. J., 7

185 Hepatitis B and Epstein-Barr viruses, 109 Herbert, J. M., 138 Herman, J. G., 139 Hernandez, O. M., 114, 173 Herrero-Jimenez, P., 57, 58, 125, 126 Herrmann, E. C., 9 Herrmann, P. C., 9 Herrnstadt, C., 47 Hershman, J. M., 61 Hervouet, E., 9, 10 Hewel, J., 52 Hexokinase II, 48 Hexokinases (HKs), 82–83 Hexosamine biosynthetic pathway (HBP) nutrient uptake pathways, 105, 108 O-GlcNAc glycosylation, 105–106 Hickey, M. M., 24 Hickman, J. A., 51 Higuchi, T., 9 Hilliker, A. J., 96 Hirabayashi, Y., 29 Hirota, K., 132 Hoberman, H. D., 9 Hock, J. B., 104 Hoffman, M. A., 139 Hoffman, R. M., 171 Hogg, P. J., 123 Ho, H. Y., 97 Hollstein, M., 137 Hooper, L., 136 Höpfl, G., 78 Horejsi, Z., 170 Horiuchi, S., 9 Horton, R., 126, 128 Horvath, J. C., 150 Howell, N., 47 Hruban, Z., 66 HSP70 overexpression, 48 Huang, C., 64 Huang, L. E., 134 Huang, P., 9, 120, 122–126, 171, 173 Hudson, K. M., 134 Human osteosarcoma cells, 104 Human placenta choriocarcinoma cell line BeWo (JCRB911) and GLUT1 mRNA, 87 Human spontaneous cervical cancer cells (SiHA) and GAPDH overexpression, 85 Huntly, B. J., 42 Hwang, P. M., 10 Hypoxia, 38, 107, 109, 134–135 hypoxia response elements (HREs), 80

186 Hypoxia (cont.) hypoxic cells, studies of GLUT mRNA expression in, 87 Hypoxia inducible factor 1 α (HIF-1 α) in cancer, 80–81 and glycolytic pathway, 81–82 ectopic expression of, 94 glucose transporters and, 87–88 HRE and, 135–136 nonhypoxic stimuli, induction by, 80 regulation, 75 acetyl transferase, 79 MAPK and PI3K pathways, 78 polyubiquitination, 77–78 stabilization in normoxia condition, 76–77 structure of, 75 transactivation domains and, 74 TCA intermediates upregulation and carcinogenesis, 88 as therapeutic target, 88 I Ichikawa, M., 59, 123 Ido, Y., 113 Igney, F. H., 162 Immortalization, 92 enhanced glycolysis senescence and, 94–95 See also Carcinogenesis Inghirami, G., 23 Ingman, M., 127 Ingram, V. M., 63 Inoki, K., 12 Irani, K., 57 Iron sulfur (Fe-S) clusters, mitochondrial enzymes, 4 Isidoro, A., 6, 9–11 Isken, F., 11 Israelson, A., 12 Itahana, K., 96 Itahana, Y., 96 Ito, H., 169 Ivan, M., 139 Ivanovska, I., 47 Iyer, N. V., 82 Izquierdo, J. M., 6, 11 J Jaattela, M., 4 Jacks, T., 137 Jacobs, H. T., 22 Jacobson, M. D., 173 Jacques, C., 60, 61

Index Jagiello, G., 22 Jain, R. K., 134 Janes, S. M., 169 Jazwinski, S. M., 68 Jeong, J. W., 24, 80, 81 Jessie, B. C., 128 Jewell, U. R., 78 Jiang, B. H., 135 Jiang, J., 107 Jiang, W. G., 61 Jiao, J., 10 Johnson, F. M., 121, 122, 124 Johnson, J. H., 115 Jones, R. G., 12 Joo, E., 174 June, C. H., 149 Ju, X., 11 K Kaaks, R., 107 Kachhap, S., 10, 62, 127, 128 Kadhom, N., 25 Kaelin, W. G., 24, 139 Kaelin, W. G. Jr., 23, 139 Kagan, V. E., 107 Kahn, B. B., 12 Kanamori, T., 58, 123, 127 Kandel, E., 171 Kang, J. G., 10, 100, 132 Kang, S. W., 34 Kappler, J., 149 Karin, M., 64, 65 Karplus, P. A., 35 Kaschten, B. J., 7 Kataoka, T., 149 Kearns-Sayre syndrome (KSS), 124 Kennedy, S., 171 Khaleque, M. A., 48 Kheradmand, F., 168 Khrapko, K., 57, 58, 125, 126 Kilo, C., 113 Kim, H., 140 Kim, J. H., 41, 79 Kim, J. W., 10, 46, 94 Kim, J. Y., 11, 172 Kim, K. W., 24, 80, 81 Kim, M. K., 107 Kim, S., 7 Kim, S. H., 24, 79–81 Kim, S. T., 140 Kim, W. Y., 139 Kim, Y. H., 11 King, A., 47

Index King, M. P., 173 Kinzler, K. W., 137, 169 Kirkinezos, I. G., 124 Kispal, G., 11 Klatt, P., 100 Klausner, R. D., 4 Klco, J., 139 Klco, J. M., 139 Klein-Szanto, A., 58, 61, 62, 119 Klingenberg, M., 22 Kniffin, C. L., 8 Knudson, A. G. Jr., 139 Koch, C. J., 134 Koed, K., 170 Kofler, B., 124 Kohl, R., 138 Kollipara, R., 42 Kol, S., 115, 116 Kondoh, H., 96, 97, 100, 132 Kondo, K., 139 Konopleva, M., 174 Koocheckpour, S., 140 Korsmeyer, S. J., 164–166, 174 Koshikawa, N., 172 Kotch, L. E., 82 Koukourakis, M. I., 8 Koumenis, C., 138 Kourembanas, S., 135 Kovacs, G., 57 Ko, Y. H., 27, 48 Krajewska, M., 6, 7, 9–11 Krammer, P. H., 162 Krebs cycle, 3, 107 Krebs, H., 6, 133 Krippner, A., 20 Kroemer, G., 19, 20, 25, 30 Kron, S. J., 11, 108 Krueger, J. S., 52 Kruse, J. P., 132 Kubasiak, L. A., 114, 173 Kubek, S., 12 Kuchroo, V. K., 149 Kuzmin, I., 140 Kvietikova, I., 78 Kwei, K. A., 65 Kwong, J. Q., 67 L Lactate dehydrogenase A (LDH-A) expression, 52 and lactate carrier MCT4, 86, 87 Lambeth, J. D., 38 Lamb, N., 135

187 Landowski, T. H., 150, 169 Lanier, L. L., 149 Larochette, N., 30 Larsson, N. G., 22 Larsson, R., 64 Lartigue, L., 165 Latif, F., 23, 59, 139 Laughner, E., 80 Leber hereditary optic neuropathy (LHON), 124 Lechago, J., 7 Lechago, L. V., 7 Lechpammer, M., 139 Ledbetter, J. A., 149 Leder, P., 46, 49, 51, 174 Lee, F. S., 65 Lee, H., 106 Lee, J. W., 24, 80, 81 Lee, S., 139 Lee, S. M., 140 Lee, S. R., 34 Lee, S. Y., 107 Lee, T. Y., 11 Lee, W. C., 107 Lehninger, A. L., 10, 106 Leon-Avila, G., 4 Letai, A., 165 Leukemia, 109 Leung, S. W., 135 Levine, A. J., 137, 170 Levy, J. P., 64 Lewis, B. C., 7, 93, 171 Lichtenstein, A. H., 2 Lien, A. D., 8 Lightowlers, R. N., 22 Li, G. X., 29 Li, J. J., 65 Lill, R., 4, 11 Lim, A. L., 10 Lin, A. W., 95 Lin, Q., 11 Lin, S. J., 98 Lincoln, D. W., 38 Lindsten, T., 165 Linsley, P. S., 149 Lipid signaling molecules and migratory phenotype regulation, 40 Liu, J., 23 Liu, J. W., 48 Liu, V. W., 57 Liu, X., 168 Liu, Y., 170 Liu, Z., 64

188 Li, X. D., 110 Li, X. X., 173 Li, Y., 57, 125, 126 Li, Z., 11 Ljungberg, B., 57 Lleonart, M. E., 96, 97, 100, 132 Lodish, H. F., 7 Lofromento, N. E., 110 Lolkema, M. P., 140 Lonergan, K. M., 140 López de Heredia, M., 6, 11 López-Rós, F., 6, 7, 9, 10 Lorenc, A., 128 Lorenz, M., 96 Los, M., 140 Lott, M. T., 120, 127 Lou, Y. R., 29 Lowe, S. W., 95, 171 Lucchesi, P. A., 63 Lu, F. J., 97 Lu, H., 8, 82, 136 Luis, A. M., 11 Lunghi, M., 175 Luo, Y., 63 Lutz, R. J., 170 Lu, X., 138 Lu, Y., 11 Lu, Y. P., 29 Lynch, M., 121 M Mabjeesh, N. J., 136 Macaulay, V., 22, 47, 126–128 MacGregor, G. R., 165 MacKenzie, E. D., 24, 25 Maejima, C., 172 Maher, E. R., 136 Mahon, P. C., 132 Majewski, N., 171 Malchow, P., 170 Malthiery, Y., 59 Mammalian mitochondrial oxidative phosphorylation (OXPHOS), 4, 55 Bcl-2 expression and, 67 defects in cancer cellular adaptations and, 59–60 functional significance of, 57–59 dysfunction mtDNA mutations and, 58 tumor-specific marker genes, 61–62 gene expression and transcription, 106 mitochondrial-nuclear intergenomic signaling in, 59

Index phenotype and apoptosis, 104 representation of, 56 Manfredi, G., 67 Manganese-dependent superoxide dismutase (MnSOD) overexpression and cell proliferation, 28 Maniatis, T., 65 Manka, D., 106 Mansel, R. E., 61 Mansouri, J., 41 MAPK pathways, 64 Maraver, A., 100 Marconi, A., 168 Marin-Hernandez, A., 46 Marquardt, C., 11 Marrack, P., 149 Marsden, P. A., 135 Marsden, V. S., 167 Martin, D. A., 64 Martindale, D., 149 Martin, D. S., 171 Martinez-Diez, M., 10–12 Martinez, M., 6, 9, 11 Martorana, G. E., 50, 51 Marzulli, D., 110 Matheu, A., 100 Mathupala, S. P., 9, 27, 48 Matoba, S., 10, 100, 132 Matsumura, T., 113 Matsuno-Yagi, A., 20 Matsuyama, S., 11 Matthew, C. K., 106, 113 Mattiazzi, M., 64 Mavligit, G. M., 149 Ma, W., 10 Maxwell, P. H., 8, 136, 140 Mayr, J. A., 124 Mazelin, L., 169 Mazure, N. M., 172 McAndrew, D., 62 McCormick, F., 170 McCurrach, M. E., 95 McFate, T., 8 McGoldrick, E. T., 173 McKnight, S. L., 11 McKusick, V. A., 8 McQuillan, L. P., 135 Mechta-Grigoriou, F., 80 Meierhofer, D., 124 Melamede, R. J., 152 Melendez, J. A., 38, 40 Melvin, R. G., 108 Menon, S. G., 36

Index Mertz, R. J., 115 Merzak, A., 140 Metcalfe, A. D., 169 Metzen, E., 75, 77 Miceli, M. V., 68 Michael, E. M., 124 Michalak, E., 137 Michalik, L., 51 Michel, G., 135 Miething, C., 171 Migliaccio, E., 42 Millhorn, D. E., 106 Minami, R., 7 Minet, E., 135 Mirebeau, D., 60, 61 Mitochondria activities in, 4 and apoptosis, 29 cancer, 20 cell immune system, 110 cellular respiration and dedifferentiation, 49–52 DNA deletions and mutations, 110 genome, 121–124 H2 O2 -dependent transcription of MMP-1 transcription, 40–41 H2 O2 , generation of, 37 metabolism in, 147 mitochondrial-derived ROS, 34 angiogenesis, 37–39 cancer stem cells, 41–42 and cell cycle, 35–36 cell survival, 36–37 invasion/migration, 39–41 mitochondrial DNA (mtDNA) mutations, 47, 57–58 mitochondrial encephalomyopathy lactic acidosis, and strokelike episodes (MELAS), 124 mitochondrial HSP-mediated, 51 mitochondrial transcription factor A (TFAM), 61 optic atrophy 1 (Opa 1), 66 outer membrane permeability, 165–166 OXPHOS and respiration, 5–6, 10 peroxisome proliferator activated receptor (PPAR)-α–mediated effects in rodents, 51 p66shc regulation and, 42 respiration and ROS production, 106 specific phospholipid cardiolipin (CL) complexation, 107 superoxides role, 27

189 type II hexokinase, interaction between, 24 voltage-dependent anion channel (VDAC), 27 Mitofusin 2 (Mfn 2) downregulation, 66 Miyashita, T., 173 Miyazaki, K., 172 MnSOD-overexpressing MCF-7 cells, 65 Modica-Napolitano, J. S., 47 Mohyeldin, A., 8 Molecular homeostasis and genetic instability, 133–134 Montanaro, L., 140 Monteith, G. R., 62 Montoya, J., 119, 121, 125 Mookerjee, B., 137 Moorhead, P. S., 94 Moraes, C. T., 11, 60, 61, 67, 124 Moreno-Loshuertos, R., 128 Moreno-Sanchez, R., 46 Morris, H. P., 8, 57 Morrissey, C., 140 Mueckler, M. M., 7 Muehlematter, D., 64 Mühlenhoff, U., 4, 11 Muller, U., 23, 59 Munk Pedersen, I., 169 Munnich, A., 22, 27 Murad, E., 35 Murrell, G. A., 63 Muse, K. E., 65 Myoclonic epilepsy and ragged-red fibers (MERRF), 124 N NADPH oxidase family members (Nox1-5), 35 Nagata, S., 149 Nagy, A., 57 Nakagawara, A., 172 Nakamura, E., 23, 139 Nakamura, S., 9 Nakashima, Y., 96, 100 Nakshatri, H., 64 Nass, M. M., 66 Navarro, P., 168 Nefedova, Y., 169 Negelein, E., 5, 20 Neilson, A., 9 Nejfelt, M. K., 135 Nekhaeva, E., 57, 58, 125, 126 Nelson, K. K., 38, 40, 41 Nemoto, S., 34, 42 Neri, P., 174 Neumann, H. P., 9

190 Neurogenic muscle weakness, ataxia and retinitis pigmentosa (NARP), 124 Newell, M. K., 149, 152 Newmeyer, D. D., 165 Niemann, S., 23, 59 Niikura, M., 59, 123 Nishikawa, M., 27 Nisoli, E., 107 Nocca, G., 49, 50 Normoxia, 77 See also Hypoxia Nuclear DNA mutations and mitochondrial proteins, 47 Nuclear respiratory factors 1 and 2 (NRF-1 and NRF-2), 60 Nyengaard, J. R., 113 O Oakes, S. A., 48 Oberley, L. W., 65 Oberley, T. D., 65 Oca-Cossio, J. A., 67 Odstrcil, E. A., 11 Oesterle, D., 9 Ogunshola, O., 78 Ohh, M., 138–140 Ohta, K., 61 Ohtani, N., 48 Old, L. J., 110 Onaya, T., 61 Opferman, J. T., 48 Oprysko, P. R., 134 Orr, W. C., 96 Orsini, F., 42 Ortega, A. D., 2, 9–11 Osmanian, C., 137 Osthus, R. C., 7 Ostronoff, L. K., 11 Oxidative stress and senescence, 95–96 Oxygen homeostasis, 73 sensors, 136 Ozben, T., 27 P Packer, L., 64 Pagano, M., 23 Paik, J. H., 42 Paik, S. G., 106 Palacios, C., 114 Pandolfi, P. P., 91 Pandolfi, S., 140 Pang, S. F., 28 Pan, Y., 134

Index Panza, C., 168 Papandreou, I., 10 Papa, S., 60 Paragangliomas (PGLs), 23 Parfait, B., 22 Parkes, T. L., 96 Park, J. A., 79 Park, J. H., 172 Park, J. Y., 10 Park, S., 140 Park, S. Y., 11 Pasteur effect, 20 Pasteur, L., 2 Pastorino, J. G., 104 Patel, L., 64 Patino, W. D., 10, 100, 132 Paulin, F. E., 9 Paulus, P., 7 Pawlu, C., 9 Payne, G., 170 Pecina, P., 9, 10 Pecqueur, C., 21 Peczkowska, M., 9 Pedersen, P., 104 Pedersen, P. L., 6–9, 27, 48, 57 Pelicano, H., 10, 113, 126, 171, 172 Penta, J. S., 121, 122, 124 Perez, J. M., 30 Peroxisome-proliferator activated γ coactivator 1 (PGC-1) gene family, 60 Peskin, B. S., 107 Petrelli, J., 107 Petros, J. A., 9, 58, 64, 127 Pette, D., 22 Peutz-Jeghers syndrome, 12 Pfeiffer, K., 20, 58, 59, 62 Pfeiffer, T., 2, 5, 9 Pfister, M. F., 11 PGC-1 related coactivator (PRC), 61 Pheochromocytomas (PHEOs), 23 Phillips, J. P., 96 Phosphoenolpyruvate carboxykinase (PEPCK), 49 Phosphofructokinase 1 (PFK-1), 84 Phosphoglycerate kinase (PGK), 85 Piana, G. L., 110 Pich, S., 66 PI3K/AKT signaling pathway and VEGF expression, 78–79 Pilcher, K., 11 Pilkington, G. J., 140 Pineda, G., 110 Pinedo, C. M., 114

Index Piva, R., 23 Pizarro, A., 168 Pizzo, P., 62 Plas, D. R., 8, 12, 132, 171 Plasminogen-activator inhibitor 1 (PAI-1), 61 Plate, K. H., 135 Platelet-derived growth factor (PDGF)-induced mitogenic signaling, 35 Podda, A., 23 Polyak, K., 57, 125, 126 Poole, L. B., 35 Porcelli, A. M., 127 Porin-like protein voltage-dependent anion channel (PPVAC), 48 Posener, K., 5 Pouysségur, J., 77, 80, 172 Powis, G., 29 Poyton, R. O., 104, 113 Pozzan, T., 62 Preston, G., 47 Prosser, R., 22 p53 tumor suppressor protein, 8–9 p53 tumor suppressor gene and organismal life span, 99–100 Pugh, C. W., 140 Puigserver, P., 61 Pyruvate dehydrogenase (PDH), 86 Pyruvate kinase (PK), 86 Q Quesada, N. M., 38 Quevedo, C., 30 Quintanilla, M., 168 R Rabbitts, P. H., 92 Racker, E., 9 Raff, M. C., 173 Raghunand, N., 7 Rai, E., 128 Ramanathan, A., 104, 113 Ranganathan, A. C., 41 Ratcliffe, P. J., 8, 136, 140 Rathmell, J. C., 8, 171 Rauscher, F. J., 64 Ravi, R., 137 Reactive oxygen species (ROS), 33 and carcinogenesis, 105 dependent stabilization of HIF-1α, 39 and redox signaling, 34–35 related signaling pathways, 50 ROS-mediated cell growth, molecular mechanism(s), 64 signal transducing molecule, 34

191 Rechcigl, M. Jr., 66 Redondo A., 9–11 Redox-based signaling, 35 Redox signaling pathway, 63–65 Reed, J., 169 Reed, J. C., 4, 11, 20, 173, 174 Regan, K. J., 38, 40 Reichert, M., 172 Remy, C., 7 Respiration, 3 Respiratory chain (RC) biochemical and genetic features, 20 cell proliferation and, 25 genetic origin and organization, 21 redox status, 22 and tumors, 29–30 Retrograde signaling in cancer cell altered antioxidant defense pathway, 65 altered expression of OXPHOS regulatory genes and subunits, 60–61 altered mitochondrial morphology and, 65–66 antiapoptotic genes activation, 66–67 calcium signaling pathway activation, 62–63 genes activation involved in tumor invasion and progression, 61–62 mechanism of, 67–68 redox signaling pathway activation, 63–65 Reynier, P., 59 Reynolds, T. Y., 134 Reznick, R. M., 12 Rhee, S. G., 34 Rho GTPase-activating protein, 39 Ricchetti, M., 22 Rice-Evans, C., 63 Rice, K., 137 Richards, F. M., 51 Richard, S. M., 124, 125 Richhardt, N., 11 Ricquier, D., 21 Ridnour, L. A., 65 Riedl, S. J., 166 Rigo, P., 7 Ristow, M., 2, 11 Ritchie, J. M., 29 Rivas, A. L., 114 Roberts-Thomson, S. J., 62 Robey, I. F., 8 Robey, R. B., 171 Rockwell, S., 134 Rodien, P., 59 Rodriguez-Enriquez, S., 46

192 Roeder, L. M., 115, 116 Roe, J. S., 140 Roe, R., 135 Roifman, C. M., 110 Romer, L. H., 169 Rosenberger, S. F., 57, 64, 65 Rosen, O. M., 7 Ross, A. J., 165 Roth, J. C., 137 Roth, P. H., 8, 133, 135, 136 Rotig, A., 22 Rouault, T. A., 4 Roux, D., 77 Rowan, A. J., 25, 59 Roy, D., 50 Rue, E., 135 Rue, E. A., 135 Ruiz, C. R., 114 Ruiz-Pesini, E., 9, 58, 64, 120, 127 Rustin, P., 22–25, 27 Ryan, K. M., 138 Ryu, J. H., 172 S Saavedra, E., 46 Saccharomyces cerevisiae, 2, 106 arrest defective-1 (ARD1) in, 79 Crabtree effect, 106 Sadlock, J., 22 Sakamaki, T., 11 Sakamoto, K., 12 Sakashita, M., 7 Sala, S., 2, 9–11 Salas, A., 22, 47, 126–128 Salinas, M., 11 Saliou, C., 64 Salvesen, G. S., 166 Samuels, D. C., 124–126, 128 Sánchez-Aragó, M., 2, 6, 7, 9–11 Sanchez, L. B., 4 Sansone, P., 42, 140 Santamaria, G., 10–12 Santarén, J. F., 11 Saretzki, G., 96 Saridin, F., 168 Savagner, F., 59–61 Sawyer, D. B., 48 Scarpulla, R. C., 106 Scatena, R., 49–51 Schade, A. L., 5 Schagger, H., 20 Scheid, A., 78 Schimmer, A. D., 167

Index Schmeller, N., 124 Schmid, T., 138 Schmitt, C. A., 171 Schneider, P., 149 Schon, E. A., 22, 128 Schreiber, R. D., 110 Schreiber, S. L., 104, 113 Schuler, M., 167 Schulz, T. J., 11 Schumacker, P. T., 173 Schuster, S., 2, 5, 9 Scorrano, L., 12, 48, 174 Scott, C. L., 167 Seftor, E. A., 7 Sekido, Y., 139 Sekito, T., 67, 68 Selak, M. A., 9, 24, 25, 47 Semenza, G. L., 8, 10, 75, 81, 92, 132–136, 135, 136, 139 Senescence, 94 bypassing effect of enhanced glycolysis, 96–97 oxidative stress and, 95–96 Senoo-Matsuda, N., 23 Serizawa, S., 11 Serrano, M., 94, 95 Serra, V., 96 Seth, R. B., 110 Shah, A., 7 Sharma, R. I., 9 Sharma, S., 128 Sharma, S. V., 169 Shaw, R. J., 12 Shay, J. W., 94 Sheldon, D. R., 150 Shephard, H. M., 48–50, 58, 61–63, 67 Sherr, C. J., 94 Shidara, Y., 58, 123, 127 Shi, H. H., 57 Shi, J., 35 Shim, H., 7, 93, 171 Shim, Y. J., 107 Shin, Y. K., 10 Shi, X., 64 Shi, Y., 166 Shi, Y. H., 135 Shoshan-Barmatz, V., 12 Shoubridge, E., 128 Shulman, G. I., 12 Simonetti, S., 22 Simoni, D., 150 Simon, M. C., 24, 113, 134 Simonnet, H., 58, 59, 62

Index Sinclair, D. A., 98 Singh, G., 120 Singh, K. K., 9, 10, 47, 62 Sinkovics, J. G., 150 Sivridis, E., 8 Sledge, G. W. Jr., 64 Sligh, J. E., 8 Smallwood, A. C., 139 Smith, T. A., 9 Sohal, R. S., 96 Sole, P. D., 49, 50 Sorcinelli, M. D., 165 Soslau, G., 66 Spector, M., 9 Spencer, B., 115 Sperl, W., 124 Srivastava, S., 60, 61 Steinbach, J. P., 172 Stengel, P., 75, 77 Stiehl, D. P., 80, 85 Stiles, B., 10 Storci, G., 42, 140 Strasser, A., 137 Streuli, C. H., 169 Strompf, L., 23 Subbaram, S., 37, 38, 40 Succinate dehydrogenase (SDH) deficiency and cancer, 88 mutations, 23 Sugai, T., 9 Suk, K., 172 Sukosd, F., 57 Sung, H. J., 10 Sun, L., 110 Sun, W., 127, 128 Sun, Y., 61 Supra, P., 172 Swartzendruber, D., 152 Swift, A. L., 9 Swift, H., 66 Symons, M., 168 Sympson, C. J., 168 Synthesis of Cytochrome c Oxidase 2 (SCO2), 10 p 53 gene, 132 T Tachibana, K., 11 Tait, A. S., 8 Takahashi, A., 48 Takeda, K., 34 Takenaga, K., 172 Tang, D. G., 48

193 Tarassov, I., 22 Taylor, R. W., 9 Tchernyshyov, I., 10 Tedesco, L., 107 Tekaia, F., 22 Tennant, D. A., 25 Terashima, M., 9 Teras, L. R., 107 Testa-Parussini, R., 107 Thierbach, R., 11 Thilly, W. G., 57, 58, 125, 126 Thioredoxin (TRX) and cysteine, 74–75 overexpression and cell proliferation, 28 Thomas, P. A., 7 Thompson, A. M., 9 Thompson, C. B., 8, 11, 12, 108, 113, 132, 165, 170, 171, 173 Thornton, J., 67, 68 Thun, M. J., 107 Tian, W. N., 97 Tiller, G. E., 8 Timofeev, O., 11 TNF-α/CHX–induced apoptosis, 36 Tolomeo, M., 150 Tomiyama, A., 11 Tomlinson, I. P., 25, 47, 59 Tomlinson, J. P., 121 Tonello, C., 107 Tonks, N. K., 35 Tory, K., 139 Tothova, Z., 42 Tovar, J., 4 TP53-induced glycolysis and apoptosis regulator (TIGAR), 9, 100 Traber, M. G., 64 Transcription factor Ets-1, 38 Trauger, R., 152 Tremble, P., 168 Troncoso, P., 175 Tsao, T., 174 Tschopp, J., 149 Tsuda, M., 23 Tsuruo, T., 149 Tsuruta, A., 9 Tu, B. P., 11 Tumor cell mitochondria, 66 glucose-specific transporters (GLUTs) induction, 7 metastasis and, 39 microenvironment for, 136–137 PTEN suppressor, 40

194 Tumor (cont.) superoxides role in, 27 tumorigenesis temporal and spatial effects in, 112–113 tumor-specific metabolic pathways, 146–147 VEGF role, 38 Turnbull, D. M., 9, 124, 125 Tyner, S. D., 100 Tyurina, Y. Y., 107 Tyurin, V. A., 107 Tzagoloff, A., 19, 21 U Uesugi, N., 9 UnCoupling proteins (UCP), 21–22 uncoupling protein 2 (UCP2) expression in colon cancers, 110 Usadel, H., 57 Usher, P., 7 V Vahsen, N., 12, 20, 166 Vainio, H., 107 van Beest, M., 140 Vander Heiden, M. G., 11, 108, 171, 173 Vande Velde, C., 172 Vandromme. M., 135 van Holde, K. E., 106, 113 van Waveren, C., 61 Vega, A., 22, 47, 126–128 Velankar, M., 174 Velours, J., 11 Venditti, J. M., 150 Venkatachalam, S., 100 Verma, A., 8, 82, 136 Vijayasarathy, C., 48–50, 58, 61–63, 67, 119 Vijayvergiya, C., 64 Villalobos-Menuey, E., 152 Villalobos-Menuey, E. M., 155 Villani, G., 60 Villunger, A., 137 Vincent, B. J., 8 Vincenzoni, F., 50 Voest, E. E., 140 Voet, D., 146, 147 Voet, J., 146, 147 Vogelstein, B., 137, 169 Voigt, A., 11 Volmat, V., 77 von Hippel–Lindau syndrome, 8, 138–140 VHL gene, mutation in, 81 von Hippel–Lindau tumor suppressor protein (pVHL), 75

Index von Kienlin, M., 7 von Zglinicki, T., 96 Vousden, K. H., 138 W Wachsman, J. T., 121, 122, 124 Waddle, J. A., 68 Wahli, W., 51 Walensky, L. D., 165 Wallace, D. C., 25, 27, 47, 119–122, 120, 127, 128 Walsh, S. A., 7 Wan, X. S., 65 Wang, C., 11, 104, 110, 113 Wang, G. L., 8, 133, 135, 136 Wang, J., 94 Wang, S., 10 Wang, T., 11 Wang, X., 4 Wang, Y., 94 Wang, Z. Q., 113 Warburg, O., 5, 6, 20, 56, 81, 120, 126, 131–133 Warburg hypothesis hypoxia inducible factor 1 alpha (HIF-1α) and, 8 model, 113–115 principles of, 7 Warren, L., 66 Watson, A., 51 Watson, P. H., 137 Watt, F. M., 169 Waugh, T. A., 29 Way, I. P., 169 Webster, K. A., 114, 173 Weiler, S., 165 Weinberg, R. A., 11, 92, 134 Weinhouse, S., 5 Weller, M., 172 Welsh, S. J., 8, 29 Wenger, R. H., 78, 80, 85, 138 Weng, Y., 139 Wen, S., 175 Werb, Z., 168 Werner, E., 168 Weydert, C. J., 29 Whang, E. E., 169 Whelan, M., 149 Wiesener, M. S., 136 Wilhelm, M., 57 Wilkie, D., 66 Willett-Brozick, J. E., 9, 20, 23, 59 William, O., 108

Index Williamson, J. R., 113 Wind, F., 20 Wolff, T., 9 Wong, G. H. W., 36 Woods, M., 5 Wood, Z. A., 35 Worley, J. F., 115 Wright, W. E., 94 Wu, H., 10 Wu, L., 137 Wu, M., 9 Wu, X., 91 Wu, Z., 61 Wurster, R. D., 63 Wykoff, C. C., 137, 140 X Xia, Y., 57 Xie, L. Y., 94 Xue, W., 171 Xue, Y., 65 Xu, J. N., 28 Xu, Q., 11, 140 Xu, R. H., 10, 113, 171, 172 Xu, R. K., 28 Y Yamagata, K., 58, 123, 127 Yamakoshi, K., 48 Yamauchi, A., 125 Yang, H., 139 Yang, M., 171 Yang, Q. H., 28 Yao, Y. G., 22, 47, 126–128 Yasuda, K., 23 Yauch, R. L., 140 YC-1–treated Hep3B hepatoma cells, aldolase and enolase mRNAs, 85 Yeast, Sir2 as longevity gene in, 97–98 calorie restriction and, 98–99

195 Yee, A. J., 11 Yen, P., 29 Yoo, B. C., 10 Yoon, B. I., 29 Younes, M., 7 Youn, H. D., 140 Yuan, F., 134 Yu, Z. X., 34 Z Zaidi, M., 63 Zamzami, N., 30 Zandi, E., 64 Zanssen, S., 128 Zatyka, M., 140 Zbinden, I., 64 Zeamari, S., 140 Zeiher, A. M., 170 Zeitlin, B. D., 174 Zeller, K. I., 94 Zender, L., 171 Zhang, S. Y., 58, 61, 62, 119 Zhao, Y., 65 Zheng, J. Z., 135 Zhong, H., 80 Zhou, J., 79, 138 Zhou, S., 10, 127, 128 Zhou, X. M., 170 Zhou, Y., 172 Zhu, H., 57, 125, 126 Zhu, T., 12 Ziegler, A., 7 Zigmond, M. J., 41 Zinner, M. J., 169 Zong, W. X., 113, 165 Zou, Y., 96 Zummo, G., 48 Zu, X. L., 5 Zweier, J. L., 57

Cellular Respiration and Carcinogenesis

Liver Regeneration And Carcinogenesis: Molecular and Cellular Mechanisms

Hormonal Carcinogenesis IV

Chemical Carcinogenesis

Soil Respiration and the Environment

Soil Respiration and the Environment

Monitoring of Respiration and Circulation

Oxidation: The Cornerstone of Carcinogenesis: Oxidation and Tobacco Smoke Carcinogenesis. A Relationship Between Cause and Effect

Cellular

La respiration holotropique

Hepatobiliary and Pancreatic Carcinogenesis in the Hamster

Hepatobiliary and Pancreatic Carcinogenesis in the Hamster

Molecular and Cellular Signaling

Molecular and Cellular Biophysics

Molecular and Cellular Signaling

Molecular and Cellular Biophysics

Cellular automata and groups

Cellular and Molecular Neurobiology

Molecular and Cellular Signaling

Cellular and Molecular Neurobiology,

Fish Physiology: Nervous System, Circulation, and Respiration

Wireless and Cellular Telecommunications

Photosynthesis And Respiration (The Green World)

Oxidation: The Cornerstone of Carcinogenesis (Oxidation and tobacco smoke carcinogenesis. A relationship between cause and effect)

Photosystem I (Advances in Photosynthesis and Respiration)

Cellular Ceramics

Cellular convection

Cellular Domains

Molecular and Cellular MR Imaging

Cellulose. Molecular and Cellular Biology

Cellular Computing (Genomics and Bioinformatics)

Cellular Respiration and Carcinogenesis

Liver Regeneration And Carcinogenesis: Molecular and Cellular Mechanisms

Hormonal Carcinogenesis IV

Chemical Carcinogenesis

Soil Respiration and the Environment

Soil Respiration and the Environment

Monitoring of Respiration and Circulation

Oxidation: The Cornerstone of Carcinogenesis: Oxidation and Tobacco Smoke Carcinogenesis. A Relationship Between Cause and Effect

Cellular

La respiration holotropique

Hepatobiliary and Pancreatic Carcinogenesis in the Hamster

Hepatobiliary and Pancreatic Carcinogenesis in the Hamster

Molecular and Cellular Signaling

Molecular and Cellular Biophysics

Molecular and Cellular Signaling

Molecular and Cellular Biophysics

Cellular automata and groups

Cellular and Molecular Neurobiology

Molecular and Cellular Signaling

Cellular and Molecular Neurobiology,

Fish Physiology: Nervous System, Circulation, and Respiration

Wireless and Cellular Telecommunications

Photosynthesis And Respiration (The Green World)

Oxidation: The Cornerstone of Carcinogenesis (Oxidation and tobacco smoke carcinogenesis. A relationship between cause and effect)

Photosystem I (Advances in Photosynthesis and Respiration)

Cellular Ceramics

Cellular convection

Cellular Domains

Molecular and Cellular MR Imaging

Cellulose. Molecular and Cellular Biology

Cellular Computing (Genomics and Bioinformatics)

Recommend Documents