Ebook The biology of cancer (2/E): Part 1

(BQ) Part 1 book The biology of cancer has contents: The biology and genetics of cells and organisms, the nature of cancer, cellular oncogenes, tumor viruses, tumor suppressor genes,... and other contents.

This page intentionally left blank to match pagination of print book the biology of CANCER SECOND EDITION Robert A Weinberg This page intentionally left blank to match pagination of print book the biology of CANCER SECOND EDITION Robert A Weinberg Garland Science Vice President: Denise Schanck Assistant Editor: Allie Bochicchio Production Editor and Layout: EJ Publishing Services Text Editor: Elizabeth Zayatz Copy Editor: Richard K Mickey Proofreader: Sally Huish Illustrator: Nigel Orme Designer: Matthew McClements, Blink Studio, Ltd Permissions Coordinator: Becky Hainz-Baxter Indexer: Bill Johncocks Director of Digital Publishing: Michael Morales Editorial Assistant: Lamia Harik © 2014 by Garland Science, Taylor & Francis Group, LLC This book contains information obtained from authentic and highly regarded sources Every effort has been made to trace copyright holders and to obtain their permission for the use of copyright material Reprinted material is quoted with permission, and sources are indicated A wide variety of references are listed Reasonable efforts have been made to publish reliable data and information, but the author and the publisher cannot assume responsibility for the validity of all materials or for the consequences of their use All rights reserved No part of this book covered by the copyright hereon may be reproduced or used in any format in any form or by any means— graphic, electronic, or mechanical, including photocopying, recording, taping, or information storage and retrieval systems—without permission of the publisher ISBNs: 978-0-8153-4219-9 (hardcover); 978-0-8153-4220-5 (softcover) Library of Congress Cataloging-in-Publication Data Weinberg, Robert A (Robert Allan), 1942The biology of cancer Second edition pages cm Includes bibliographical references ISBN 978-0-8153-4219-9 (hardback) ISBN 978-0-8153-4220-5 (pbk.) Cancer Molecular aspects Cancer Genetic aspects Cancer cells I Title RC268.4.W45 2014 616.99’4 dc23 2013012335 Published by Garland Science, Taylor & Francis Group, LLC, an informa business, 711 Third Avenue, New York, NY 10017, USA, and Park Square, Milton Park, Abingdon, OX14 4RN, UK Printed in the United States of America 15 14 13 12 11 10 Visit our website at http://www.garlandscience.com About the Author Robert A Weinberg is a founding member of the Whitehead Institute for Biomedical Research He is the Daniel K Ludwig Professor for Cancer Research and the American Cancer Society Research Professor at the Massachusetts Institute of Technology (MIT) Dr Weinberg is an internationally recognized authority on the genetic basis of human cancer and was awarded the U.S National Medal of Science in 1997 Front Cover A micrograph section of a human in situ ductal carcinoma with α-smooth muscle actin stained in pink, cytokeratins and in redorange, and cytokeratins and 18 in green (Courtesy of Werner Böcker and Igor B Buchwalow of the Institute for Hematopathology, Hamburg, Germany.) v Dedication I dedicate this second edition, as the first one, to my dear wife, Amy Shulman Weinberg, who endured long hours of inattention, hearing from me repeatedly the claim that the writing of this edition was almost complete, when in fact years of work lay ahead She deserved much better! With much love This page intentionally left blank to match pagination of print book vii Preface C ompared with other areas of biological research, the science of molecular oncology is a recent arrival; its beginning can be traced with some precision to a milestone discovery in 1975 In that year, the laboratory of Harold Varmus and J Michael Bishop in San Francisco, California demonstrated that normal cell genomes carry a gene—they called it a proto-oncogene—that has the potential, following alteration, to incite cancer Before that time, we knew essentially nothing about the molecular mechanisms underlying cancer formation; since that time an abundance of information has accumulated that now reveals in outline and fine detail how normal cells become transformed into tumor cells, and how these neoplastic cells collaborate to form life-threatening tumors The scientific literature on cancer pathogenesis has grown explosively and today encompasses millions of research publications So much information would seem to be a pure blessing After all, knowing more is always better than knowing less In truth, it represents an embarrassment of riches By now, we seem to know too much, making it difficult to conceptualize cancer research as a single coherent body of science rather than a patchwork quilt of discoveries that bear only a vague relationship with one another This book is written in a far more positive frame of mind, which holds that this patchwork quilt is indeed a manifestation of a body of science that has some simple, underlying principles that unify these diverse discoveries Cancer research is indeed a field with conceptual integrity, much like other areas of biomedical research and even sciences like physics and chemistry, and the bewildering diversity of the cancer research literature can indeed be understood through these underlying principles Prior to the pioneering findings of 1975, we knew almost nothing about the molecular and cellular mechanisms that create tumors There were some intriguing clues lying around: We knew that carcinogenic agents often, but not always, operate as mutagens; this suggested that mutant genes are involved in some fashion in programming the abnormal proliferation of cancer cells We knew that the development of cancer is often a long, protracted process And we knew that individual cancer cells extracted from tumors behave very differently than their counterparts in normal tissues Now, almost four decades later, we understand how mutant genes govern the diverse traits of cancer cells and how the traits of these individual cells determine the behavior of tumors Many of these advances can be traced to the stunning improvements in experimental tools The techniques of genetic analysis, which were quite primitive at the beginning of this period, have advanced to the stage where we can sequence entire tumor cell genomes in several days (This is in sharp contrast to the state of affairs in 1975, when the sequencing of oligonucleotides represented a formidable task!) Given the critical role of genotype in determining phenotype, we now understand, as least in outline, why cancer cells behave the way that they On the one hand, the molecular differences among individual cancers suggest hundreds of distinct types of human cancer On the other, molecular and biochemical analyses reveal that this bewildering diversity really manifests a small number of underlying common biochemical traits and molecular processes viii Preface Amusingly, much of this unification was preordained by decisions made 600 million years ago Once the laws and mechanisms of organismic development were established, they governed all that followed, including the behavior of both normal and neoplastic cells Modern cancer researchers continue to benefit from this rigid adherence to the fundamental, evolutionarily conserved rules of life As is evident repeatedly throughout this book, much of what we understand about cancer cells, and thus about the disease of cancer, has been learned by studying the cells of worms and fruit flies and frogs These laws and principles are invoked repeatedly to explain the complex behaviors of human tumors By providing context and perspective, they can be used to help us understand all types of human cancer While these basic principles are now in clear view, critical details continue to elude us This explains why modern cancer research is still in active ferment, and why new, fascinating discoveries are being reported every month While they create new perspectives, they not threaten the solidity of the enduring truths, which this book attempts to lay out These principles were already apparent seven years ago when the first edition of this book appeared and, reassuringly, their credibility has not been undermined by all that has followed In part, this book has been written as a recruiting pamphlet, as new generations of researchers are needed to move cancer research forward They are so important because the lessons about cancer’s origins, laid out extensively in this book, have not yet been successfully applied to make major inroads into the prevention and cure of this disease This represents the major frustration of contemporary cancer research: the lessons of disease causation have rarely been followed, as day follows night, by the development of definitive cures And yes, there are still major questions that remain murky and poorly resolved We still not understand how cancer cells create the metastases that are responsible for 90% of cancer-associated mortality We understand rather little of the role of the immune system in preventing cancer development And while we know much about the individual signaling molecules operating inside individual human cells, we lack a clear understanding of how the complex signaling circuitry formed by these molecules makes the life-and-death decisions that determine the fate of individual cells within our body Those decisions ultimately determine whether or not one of our cells begins the journey down the long road leading to cancerous proliferation and, finally, to a life-threatening tumor Contemporary cancer research has enriched numerous other areas of modern biomedical research Consequently, much of what you will learn from this book will be useful in understanding many aspects of immunology, neurobiology, developmental biology, and a dozen other biomedical research fields Enjoy the ride! Robert A Weinberg Cambridge, Massachusetts March 2013 Chapter 8: pRb and Control of the Cell Cycle Clock 316 Yet another line of evidence indicating a coupling between differentiation and the cell cycle machinery comes from research on the biochemistry of the Myc protein As mentioned earlier (see Section 8.9), the Myc protein shifts the balance between proliferation and post-mitotic differentiation in favor of proliferation In order to understand (A) (B) rhodopsin rods ONL OPL bipolar cells GFP DAPI INL IPL (C) p27Kip1 & Mxd negative relative cell number wild-type control control +AGN +AGN Gr-1 expression Numerous experiments have shown that differentiation can Figure 8.33 Differentiation of granulocytes, retinal cells, be prevented by forcing pRb phosphorylation and inactivation and muscle cells (A) The proper differentiation of retinal cells Conversely, signals that prevent pRb phosphorylation favor (see Figure 7.3A) depends on the presence of pRb During and often induce differentiation (C) The collaborative actions normal development, retinal progenitor cells move upward and of Mxd (the Myc antagonist; see Figure 8.27A) and p27Kip1 differentiate into various specialized cell types In the eyes of 2-week-old mice, rod cells that successfully differentiated express in promoting differentiation are revealed by the behavior of the rhodopsin photoreceptor protein (red), while those that failed promyelocytes whose differentiation into Gr-1 antigen–positive to so (dashed lines) are represented only by the DAPI stain that granulocytes can be induced by an agonist of the RXR nuclear marks their DNA (blue) The latter cells sit directly above GFPreceptor Here this agonist, termed AGN194024 (AGN), was used labeled bipolar cells that mark the locations where progenitor to induce differentiation of wild-type promyelocytes (left panel) cells had previously excised both copies of their Rb gene prior to or promyelocytes that were genetically deprived of the genes migrating upward (B) Myoblasts, the less differentiated precursors encoding Mxd and p27Kip1 (right panel) The display of the Gr-1 of muscle cells (myocytes), can be cultured in vitro and remain differentiation antigen was increased >20-fold in a majority of the in the undifferentiated state (upper panel) However, various treated wild-type cells, while being essentially unaffected (if not types of physiologic signals can induce them to differentiate into slightly reduced) in the doubly deficient cells (A, from myocytes, whereupon they fuse to form muscle fibers (red, lower S.L Donovan and M.A Dyer, Vision Res 44:3323–3333, 2004 TBoC2 b8.33,n8.105,106/8.33 panel) For example, removal of serum (and its associated growth B, from E.M Wilson et al., Mol Biol Cell 15:497–505, 2004 factors) from the culture medium leads to such differentiation C, from G.A McArthur et al., Mol Cell Biol 22:3014–3023, 2002.) Myc antagonizes differentiation Myc Max MyoD Id2 E12 MyoD /E47 MyoD Id2 muscle differentiation program no DNA binding, no muscle differentiation these actions, we need to know that many of the transcription factors controlling differentiation programs are, like Myc, members of the large family of bHLH transcription factors However, these bHLH proteins function very differently from Myc: they orchestrate complex, tissue-specific differentiation programs, while Myc acts in the TBoC2 b8.34/8.34 opposite direction to block differentiation and promote proliferation Differentiationassociated bHLH transcription factors have been most extensively studied in the embryonic development of several distinct cell lineages, including those leading to the formation of muscle, the nervous system, the pancreas, and the immune system For example, four distinct bHLH proteins—termed MyoD (see above), Myf5, myogenin, and MRF4—operate in myoblast-to-myocyte differentiation (see Figure 8.33) by controlling various phases of the muscle-specific differentiation program Myc works to increase production of the Id1 and Id2 proteins, and the latter act as antagonists of the bHLH transcription factors that program differentiation Id1 and Id2 are members of a group of related proteins (Id1 to Id4) that are also members of the bHLH family of transcription factors The Id proteins operate, however, as dominant-negative inhibitors of the other bHLH transcription factors (Figure 8.34) More specifically, the Ids can form heterodimers with bHLH transcription factors but lack the ability to bind to DNA (because they lack the basic domain involved in DNA recognition) This explains how the Id proteins can act as inhibitors of the bHLH differentiation-inducing transcription factors, thereby blocking differentiation Id proteins are present at high levels in many types of actively growing cells, and this in itself reduces the likelihood that these cells will differentiate For instance, by associating with MyoD, an Id protein can prevent MyoD from programming muscle differentiation in actively growing myoblasts During the normal course of differentiation, however, the levels of Id proteins sink to undetectably low levels, and MyoD, now free of interference by Ids, is able to dimerize with its bHLH partners (called E12 and E47) in order to activate the muscle-specific differentiation program (see Figure 8.34) Depressed synthesis of the Id1 protein is known to be required for cell cycle withdrawal and differentiation in numerous cell lineages: muscle, pancreas, mammary epithelial cells, myeloid cells, erythroid cells, myocardial cells, B cells, T cells, and osteoblasts As might be expected from the above, the Id proteins also have been associated with cancer pathogenesis In many types of normal cells, the molecules of Id2 are bound and sequestered by the far more abundant molecules of pRb However, in neuroblastomas, a relatively common pediatric tumor, Id2 is often overexpressed because its expression is driven by extra copies of the N-Myc protein, a myc cousin that acts on the same targets as myc in cells (see Section 4.5) Now the tables are turned: in neuroblastoma cells, Id2 accumulates to such high levels that it is in great (>10×) molar excess of pRb Consequently, pRb can no longer sequester and regulate these cells’ Id2 proteins Figure 8.34 Id transcription factors and inhibition of differentiation Lineage-specific bHLH transcription factors, such as MyoD (beige), form heterodimers with widely expressed E12 or E47 bHLH partners (blue) The resulting heterodimeric transcription factors orchestrate differentiation programs in a variety of tissues, including muscle (left) The formation of these functional heterodimeric transcription factor complexes can, however, be blocked by Id proteins (light gray), which form heterodimers with the lineage-specific bHLH proteins, thereby preventing association of the latter with E12 or E47 Because the Id proteins lack a DNA-binding domain, they act as natural “dominant-negative” inhibitors of the lineage-specific bHLH proteins Expression of certain Id proteins, such as the Id2 shown here, can be induced by Myc (or N-myc) acting with its partner Max (top right); this helps to explain the observed ability of the Myc oncoprotein to block differentiation of various cell types 317 318 Chapter 8: pRb and Control of the Cell Cycle Clock They are then free to block the actions of differentiation-inducing bHLH transcription factors More recently, an entire different dimension of regulation has been uncovered: in certain tumors, degradation of the usually highly labile Id proteins is blocked, leading to great increases in their concentration and, in turn, blocked differentiation This mechanism may underlie the relatively poorly differentiated state of many types of tumor cells (for example, see Supplementary Sidebar 8.4) A dramatic demonstration of the opposition between the Myc oncoprotein and cell differentiation has come from a mouse model of liver cancer pathogenesis, which depends on the targeted expression of a myc transgene in hepatocytes Large hepatocellular carcinomas form, and these regress when the myc transgene is shut down At the same time, many of the carcinoma cells, which previously lacked most of the traits of normal hepatocytes, rapidly differentiate into normal-appearing liver cells that assemble to reconstruct many of the histological features of the normal liver These various controls on cell differentiation, involving pRb, Myc, Ids, and other regulatory proteins, clearly have effects on the formation and development of various types of cancer, since tumors formed by more differentiated cells are usually less aggressive while those composed of poorly differentiated cells tend to be far more aggressive and carry a worse prognosis for the patient 8.12 Control of pRb function is perturbed in most if not all human cancers Deregulation of the pRb pathway yields an outcome that is an integral part of the cancer cell phenotype—unconstrained proliferation This explains why normal regulation of the R-point transition, as embodied in pRb phosphorylation, is likely to be disrupted in most if not all types of human tumor cells (Tables 8.3 and 8.4) These disruptive mechanisms are summarized in Figure 8.35 We are already familiar with the most direct mechanism for deregulating advance through the R point—inactivation of the Rb gene through mutation In some tumors, an equivalent outcome is achieved through methylation of the Rb gene promoter In others, pRb, though synthesized in normal amounts, may be functionally inactivated by viral oncoproteins, such as the HPV E7 protein, which prevent pRb from binding and regulating E2Fs Yet another strategy used by cancer cells to inactivate pRb function is indicated by the presence of very high levels of cyclin D1 in a variety of human tumor cells This is most widely documented in breast cancer cells, in which as many as half of the tumors PI3K Wnts Figure 8.35 Perturbation of the R-point transition in human tumors The decision to advance through the R-point transition (yellow, middle, bottom) can be perturbed in a variety of ways in human tumors Elements that favor advance through the R point are drawn in red, while those that undertake to block this advance are shown in blue Almost all human tumors show either a hyperactivity of one or more of the agents favoring this advance (red) or an inactivation of the agents blocking this advance (blue) Proteins whose expression levels or activities are not known to change during the process of transformation are shown in gray PTEN TGF-βR mitogens APC β-catenin Ras mitogens Smad3/4 NF-κB Raf FOXP3 USP1 Akt/PKB RTK Apc/Cdh1 Cks1 Skp2 Cul1 Myc Akt/PKB E7 p15INK4B RTK p27Kip1 Id2 D1 D–CDK4/6 cyclins pRb E–CDK2 p21Cip1 R-point transition Cdc25A Dub3 D2 Bcr-Abl Myc p16INK4A pRb function is perturbed in most cancers Figure 8.36 Amplification of the cyclin D1 gene The use of fluorescent in situ hybridization (FISH) makes it possible to detect the copy number of specific genes in histological sections Here is the result of using a probe for CCND1, the human gene encoding cyclin D1, to determine the copy number of this gene in the cells of a head-and-neck squamous cell carcinoma (HNSCC) Each bright spot represents a single copy of this gene Individual HNSCC nuclei are stained purple Hence, CCND1 is amplified to various extents here, being present in three to five copies per cell in this tumor More generally, CCND1 is found to be amplified in about one-third of HNSCC tumors, leading to corresponding increases in cyclin D1 expression and resulting loss of proper control of pRb phosphorylation Similar observations (not shown) can be made with human breast cancer cells (From K Freier et al., Cancer Res 61:1179–1182, 2003.) have been reported to show elevated levels of this protein In these and other carcinomas, the overexpression is sometimes achieved by increases in the copy number of the cyclin D1 genes (that is, gene amplification; Figure 8.36) More frequently, however, breast cancer cells acquire excessive cyclin D1 by altering the upstream signaling pathways (see Section 8.8) that are normally responsible for controlling expression of the cyclin D1 gene TBoC2 b8.36/8.36 A more devious ploy is frequently exploited by cancer cells to disable the pRb machinery: they shut down expression of their p16INK4A tumor suppressor protein Recall that the p16INK4A protein, like its p15INK4B cousin, inhibits the cyclin D–CDK4/6 responsible for initiating pRb phosphorylation In the absence of the p16INK4A protein, pRb phosphorylation operates without an important braking mechanism, resulting in excessive cyclin D–CDK4/6 kinase activity, deregulated pRb phosphorylation, and inappropriate inactivation of pRb (see Figure 8.35) Individuals suffering from one form of familial melanoma inherit defective versions of the p16INK4A gene It is unclear why loss of this particular CDK inhibitor, which seems to operate in all cell types throughout the body, should affect specifically the melanocytes in the skin that are the normal precursors of melanoma cells In sporadic (that is, nonfamilial) tumors of various sorts, cancer cells resort far more frequently to another strategy to shed p16INK4A function—they methylate the CpG sequences present in the promoter of the p16INK4A gene (see Section 7.8) Evidence of an even more cunning strategy for destabilizing this control circuit has been found in the genomes of a small number of both sporadic and familial melanomas In these cancers, point mutations in the CDK4 gene (the R24C mutation) create CDK4 molecules that are no longer susceptible to inhibition by the family of INK4 molecules (that is, p15, p16, p18, and p19) While these various CDK inhibitors may be perfectly intact and functional in such tumor cells, their normally responsive CDK4 target now eludes them Once again, this permits CDK4, together with its cyclin D partners, to drive the initial steps of pRb phosphorylation in a deregulated fashion [Since the R24C mutation creates a dominant allele of CDK4 (at the cellular level), only one of the two copies of the gene encoding this CDK needs to be mutated in order for a cancer cell to derive proliferative benefit Mice carrying one or two copies of this R24C allele develop a diverse variety of tumors, including those affecting mesenchymal, epithelial, and hematopoietic cell types.] The most critical CDK inhibitor involved in cancer pathogenesis may well be p27Kip1 As mentioned earlier, p27Kip1 is involved largely in inhibiting the activity of cyclin 319 320 Chapter 8: pRb and Control of the Cell Cycle Clock E–CDK2 complexes (for example, see Figure 8.15) As cells exit the cell cycle into the G0 quiescent state, p27Kip1 levels rise (Figure 8.37A) Conversely, as cells re-enter the cell cycle and advance through its G1 phase, the levels of p27Kip1 are reduced progressively throughout early and mid-G1 and then are made to fall precipitously during late Table 8.3 Molecular changes in human cancers leading to deregulation of the cell cycle clock Specific alteration Clinical result Alterations of pRb Inactivation of the Rb gene by mutation retinoblastoma, osteosarcoma, small-cell lung carcinoma Methylation of Rb gene promoter brain tumors, diverse others Sequestration of pRb by Id1, Id2 diverse carcinomas, neuroblastoma, melanoma Sequestration of pRb by the HPV E7 viral oncoprotein cervical carcinoma Alteration of cyclins Cyclin D1 overexpression through amplification of cyclin D1 gene breast carcinoma, leukemias Cyclin D1 overexpression caused by hyperactivity of cyclin D1 gene promoter driven by upstream mitogenic pathways diverse tumors Cyclin D1 overexpression due to reduced degradation of cyclin D1 because of depressed activity of GSK-3β diverse tumors Cyclin D3 overexpression caused by hyperactivity of cyclin D3 gene hematopoietic malignancies Cyclin E overexpression breast carcinoma Defective degradation of cyclin E protein due to loss of hCDC4 endometrial, breast, and ovarian carcinomas Alteration of cyclin-dependent kinases CDC25A overexpression breast cancers CDK4 structural mutation melanoma Alteration of CDK inhibitors Deletion of p15INK4B gene diverse tumors Deletion of p16INK4A gene diverse tumors Methylation of p16INK4A gene promoter melanoma, diverse tumors Decreased transcription of p27Kip1 gene because of action of Akt/PKB on Forkhead transcription factor diverse tumors Increased degradation of p27Kip1 protein due to Skp2 overexpression breast, colorectal, and lung carcinomas, and lymphomas Cytoplasmic localization of p27Kip1 protein due to Akt/PKB action breast, esophagus, colon, thyroid carcinomas Cytoplasmic localization of p21Cip1 protein due to Akt/PKB action diverse tumors Multiple concomitant alterations by Myc, N-myc, or L-myc Increased expression of Id1, Id2 leading to pRb sequestration diverse tumors Increased expression of cyclin D2 leading to pRb phosphorylation diverse tumors Increased expression of E2F1, E2F2, E2F3 leading to expression of cyclin E diverse tumors Increased expression of CDK4 leading to pRb phosphorylation diverse tumors Increased expression of Cul1 leading to p27Kip1 degradation diverse tumors Repression of p15INK4B and p21Cip1 expression allowing pRb phosphorylation diverse tumors pRb function is perturbed in most cancers Table 8.4 Alteration of the cell cycle clock in human tumors A plus sign indicates that this gene or gene product is altered in at least 10% of tumors analyzed Alteration of gene product can include abnormal absence or overexpression Alteration of gene can include mutation and promoter methylation More than one of the indicated alterations may be found in a given tumor Tumor type Gene product or gene Rb Cyclin E1 Glioblastoma + + Mammary carcinoma + + Lung carcinoma + + p16INK4A p27Kip1 CDK4/6 + + +/+ >80 + + + +/ >80 + + + +/ >90 Cyclin D1 a Pancreatic carcinoma % of tumors with or more changes + >80 Gastrointestinal carcinoma + + +b + + +/c >80 Endometrial carcinoma + + + + + +/ >80 Bladder carcinoma + + + + + + +d + +/ >90 Leukemia + Head and neck carcinomas + + + + +/ >90 + + +e +d + /+ >90 + + + + +/ >20 +/c >90 Lymphoma Melanoma + >70 Hepatoma + + + +d + Prostate carcinoma + + + + + Testis/ovary carcinomas + + +b + + Osteosarcoma + Other sarcomas + + + + + aCyclin D3 (not cyclin D1) is present and is up-regulated in some tumors D2 also is up-regulated in some tumors cCDK2 is also found to be up-regulated in some tumors dp15INK4B is also found to be absent in some tumors eCyclin D2 and D3 are also found up-regulated in some lymphomas Adapted from M Malumbres and M Barbacid, Nat Rev Cancer 1:222–231, 2001 bCyclin G1 phase by the actions of cyclin E–CDK2 complexes These low levels are not created by reduced transcription of the gene encoding p27Kip1 Instead, it seems p27Kip1 levels are reduced by the actions of the Skp2 protein, which acts together with Cul1 and several other proteins (see Figure 8.37B) to recognize p27Kip1 and ubiquitylate it, thereby tagging it for destruction in proteasomes Indeed, the declining levels of Skp2 explain the increase in p27Kip1 as cells enter into G0 (see Figure 8.37A and C) Interestingly, a very similarly structured complex (Supplementary Sidebar 8.5) is involved in the programmed degradation of cyclin E as cells pass through the G1/S transition (see Figure 8.10) The inverse relationship between p27Kip1 levels and those of Skp2 is especially apparent in various human cancers such as mammary and oral carcinomas, as well as lymphomas (see, for example, Figure 8.37D), with higher levels of Skp2 portending shorter patient survival Moreover, when the levels of p27Kip1 are measured in the cells of human esophageal, breast, colorectal, and lung carcinomas, poor patient survival is correlated with low levels of this CDK inhibitor In all of these tumors, Skp2 seems to act as a proliferation-promoting oncoprotein by forcing the degradation of the critical p27Kip1 CDK inhibitor One clue to the ultimate source of elevated Skp2 levels comes from research indicating that the Notch protein (see Section 6.12), which is hyperactive in many types of human cancer, increases transcription of the gene encoding Skp2 >70 +/ >90 +/ >80 /+ >90 321 Chapter 8: pRb and Control of the Cell Cycle Clock ubiquitin ligase p27Kip1 Cks1 (A) (B) C′ diploid fibroblasts 24 48 72 UbcH7 (E2) C′ hours w/o serum Skp2 p27Kip1 cyclin A substrate recognition F-box N′ C′ Skp1 Rbx1 Skp2 N′ C′ N′ Cul1 N′ Cul1 stalk (C) (D) differentiated mucosal cells Skp2 p27Kip1 % of p27kip1-positive cells 322 35 corr co-eff = 0.58 p = 0.001 30 25 20 15 10 0 10 15 20 25 30 35 % of Skp2-positive cells 40 45 Figure 8.37 Suppression of p27Kip1 levels by Skp2 Levels modifications of the Cul1 stock cause it to bend, allowing the E2 of p27Kip1are controlled by, among other factors, the levels of ubiquitin ligase to access its p27Kip1 substrate Myc fosters cell Skp2 the SCF cycle progression in part by inducing expression of Cul1, Csk1, complex, which constitutes an E3 ubiquitin ligase and Skp2, which proceed to ubiquitylate p27Kip1, thereby driving that ubiquitylates p27Kip1, thereby tagging it for degradation in proteasomes (A) As normal human cells enter into the G0, its degradation (The UbcH7 shown here is used to represent the structurally related E2 ubiquitin ligase that actually functions in this quiescent state, in this case because they are deprived of serum complex.) (C) As seen here, in the normal oral mucosa, when Skp2 mitogens, Skp2 levels fall, allowing the accumulation of p27Kip1; levels are low (left), levels of p27Kip1 are high (right) and help to the latter, in turn, is responsible for blocking the actions of CDK2 The loss of cyclin A demonstrates that cells have retreated TBoC2 b8.37/8.37 hold cells in a post-mitotic state; the opposite situation is observed from the active cell cycle This turquoise CDK is responsible for pushing in many cancers (not shown) The part of (B) (ubiquitin ligase) is (D) The inverse relationship between the cell through the late G1 phase of the cell cycle (B) After Skp2 levels and those of p27Kip1 is plotted here for a large number very poor resolution Is it possible to get a higher res version? - N p27Kip1 is phosphorylated by cyclin E–CDK2 (see Figure 8.25B), of oral epithelial dysplasias and carcinomas (Similar relationships, not shown, are seen in lymphomas and colorectal carcinomas.) it is recognized and bound by a large multiprotein complex that (A, from T Bashir et al., Nature 428:190–193, 2004 B, from ubiquitylates it, thereby targeting it for destruction in proteasomes B Schulman et al., Nature 408:381–386, 2000; N Zheng et al., The phosphorylated domain of p27Kip1 (dark green) associates Cell 102:533–539, 2000; N Zheng et al., Nature 416:703–709, with the Cks1 adaptor protein (purple) and the “palm” of Skp2 2002; and B Hao et al., Mol Cell 20:9–19, 2005 C and D, from (dark pink) This complex is connected to the ubiquitylating M Gstaiger, Proc Natl Acad Sci USA 98:5043–5048, 2001.) complex via the Cul1 stalk (green), which holds the enzyme (right) 100 angstroms away from its p27Kip1 substrate (left) Covalent We have read repeatedly about the hyperactive state of the Akt/PKB kinase in many human tumors, which is caused by a variety of molecular defects, including hyperactive growth factor receptors, loss of the PTEN tumor suppressor protein, and mutations in the PI3K gene It, too, has effects on the cell cycle clock in cancer cells, by reducing Synopsis and prospects the effective levels of important CDK inhibitors (see Figure 8.15) By phosphorylating p21Cip1 and p27Kip1, Akt/PKB ensures the cytoplasmic localization of these two critical antagonists of cell cycle advance, thereby marginalizing them At the same time, Akt/PKB can suppress expression of the gene encoding p27Kip1 by phosphorylating a transcription factor of the Forkhead family, which serves to further reduce the overall concentrations of p27Kip1 in the cell As if this were not enough, Akt/PKB ambushes p27Kip1 in a third way: by phosphorylating Skp2 (see Figure 8.37B), Akt/PKB activates the latter, enhancing its ability to bind p27Kip1, resulting thereafter in the ubiquitylation and degradation of p27Kip1 In Section 8.9, we learned of the various ways in which the Myc protein, operating in normal cells, extends its reach into the cell cycle clock and pulls many key regulatory levers The actions of the Myc oncoprotein are likely to be qualitatively similar The difference between the two is largely, and possibly entirely, a matter of protein level: in normal cells, levels of Myc protein are highly dependent upon extracellular mitogenic growth factors In many cancer cells, however, the Myc protein is produced constitutively, independent of mitogenic signals coming into the cell With this idea in mind, we can deduce that the many perturbations wrought by the Myc protein are magnified in the many types of human cancer cells that carry either myc oncogenes or the oncogenic versions of its cousins, N-myc and L-myc Among the many products induced by Myc, the Id proteins may well have the greatest physiologic importance On the one hand, they can act as dominant-negative inhibitors of bHLH transcription factors that program cell differentiation On the other, they can bind to pRb and seem to inhibit its functioning in a way that is reminiscent of the actions of viral oncoproteins such as large T, E7, and E1A In fact, ectopically expressed Id2 protein can replace the pRb-binding actions of SV40 large T antigen in experiments designed to reverse the quiescent growth state of cells This particular Id protein has been reported to be overproduced in diverse tumors including endometrial, head-and-neck, breast, pancreatic, esophageal, and cervical carcinomas, as well as melanomas and neuroblastomas Its expression is positively correlated with the degree of malignancy and invasiveness in many of these tumor types The Id proteins are synthesized at high rates and degraded rapidly, resulting in relatively short half-lives (less than 20 minutes for Id1, Id2, and Id3) and thus low steady-state concentrations in cells This suggests that their concentrations might be greatly increased in cancer cells by mechanisms that reduce their ubiquitylation and thus proteasome-mediated degradation; indeed precisely such mechanisms have been documented in a variety of cancer cells (see Supplementary Sidebar 8.4) The diverse genetic and biochemical strategies shown in Figure 8.35 are all focused on one common goal—that of overwhelming and deregulating pRb function, thereby destroying the tight control that it normally imposes on the R-point transition Darwinian selection, occurring in the microcosm of living tissues, favors the outgrowth of cells that, by hook or by crook, have succeeded in inactivating the critical pRb braking system and thus deregulating the R-point transition 8.13 Synopsis and prospects All of the physiologic signals and signaling pathways affecting cell proliferation must, sooner or later, be connected in some fashion to the operations of the cell cycle clock It represents the brain of the cell—the central signal processor that receives afferent signals from diverse sources, integrates them, and makes the final decisions concerning growth versus quiescence, and in the latter case, whether or not the exit from the active cell cycle will be reversible Connected with the latter decision are the mechanisms governing entrance into tissue-specific differentiation programs The core components of the cell cycle clock are already present in clearly recognizable form in single-cell eukaryotes, including the much-studied baker’s yeast, Saccharomyces cerevisiae Single-cell organisms such as this one respond to a far smaller range of external signals than metazoan cells residing within complex tissues These simple organisms lack the hundred and more distinct types of growth factor receptors that 323 324 Chapter 8: pRb and Control of the Cell Cycle Clock vertebrate cells display on their surfaces, as well as other receptors, such as integrins, that our cells use to sense and control attachment to the extracellular matrix Yeast cells also lack the growth-inhibitory receptors, such as the TGF-β receptors, that play such a critical role in the economy of mammalian tissues All this explains why the peripheral wiring that regulates the core cell cycle machinery of animal cells has been added relatively recently in the history of life on this planet— perhaps 600 million years ago when metazoa may first have appeared The need to respond to a wide variety of afferent signals explains why so many distinct layers of regulation have been imposed on the core machinery Without these additional regulators, notably the CDK inhibitors, the core machinery could not be made responsive to the diverse array of signals that impinge on individual metazoan cells and modulate their proliferation While these connections between the cell exterior and the cell cycle clock were being forged, other critical regulators became integrated into this complex circuitry Actually, the invention of the key governors of G1 progression—pRb and its two cousins, p107 and p130—seems to have occurred well before the rise of metazoa: a pRb–E2F signaling pathway involved in cell cycle control is already present in Chlamydomonas reinhardtii, a single-cell alga that is related to the ancestors of land plants (Indeed, manipulations of the orthologs of the mammalian p27Kip1 gene in soybeans and canola/rapeseed have led to significantly increased crop yields!) Similarly constructed pRb- and E2F-like proteins are present in worms and flies, indicating their presence in early metazoans pRb may not have been the first of these three pocket proteins to have evolved, but during the ascendance of mammals, it surpassed the others in its ability to govern the critical decision made at the R point These three proteins preside over various aspects of the growth-versus-quiescence decision of the cell and in this sense lie upstream of the cell cycle clock At the same time, by affecting gene transcription, these proteins create a coupling between the cell cycle clock and the downstream circuits that must be activated in order for cells to execute the complex biochemical changes that enable them to enter into S phase [An unresolved question is: since pRb controls cell proliferation in many cell types throughout the body, why does heterozygosity at the RB locus predispose humans specifically to retinal tumors and osteosarcomas (Supplementary Sidebar 8.6)?] Some of the neoplastic growth state can be explained by the workings of the cell cycle clock We can explain the deregulated proliferation of cancer cells in terms of the operations of pRb and the molecules that control its state of phosphorylation Without pRb at the helm, the requirement for the growth-promoting actions of oncoproteins such as Ras is greatly reduced, and cells advance through G1 without fulfilling many of the prerequisites that normally determine whether or not the R-point transition will proceed The coupling between proliferation and blocked differentiation, still incompletely understood, seems to be traceable to the operations of pRb and proteins such as Myc, which simultaneously drives the cell cycle clock forward through G1 and antagonizes some of the master regulators of differentiation programs Perhaps surprisingly, one important aspect of the proliferative control of cells operates independently of the cell cycle clock As we will learn in detail in Chapter 10, normal cells can replicate only a limited number of times before they lose the ability to proliferate further; cancer cells have an unlimited proliferative capacity—the phenotype of replicative immortality The molecular devices within cells that tally the number of replicative generations through which a cell lineage has passed are embedded in the chromosomal DNA, and these devices not seem to be controlled by the cyclin– CDK complexes regulating cell cycle advance Yet other aspects of the malignant growth program are not controlled by the cell cycle clock Cells respond to severe, essentially irreparable genomic damage by activating their cell suicide program—apoptosis; this response does not seem to be connected directly to the cell cycle machinery This program will be the subject of much of our discussions in the next chapter Synopsis and prospects Many other peculiarities of the cancer cell phenotype are largely the purview of cytoplasmic oncoproteins such as Ras Included here are phenotypes of cell motility, changes in cell shape, anchorage independence, alterations in energy metabolism, and invasiveness These behaviors are also not controlled by the cell cycle clock, which does its work in the nucleus In spite of significant alterations that the cell cycle clock suffers in cancer cells, we recognize that the effects of these changes are felt largely during the G1 phase of the cell cycle We can understand this by noting that the R-point transition occurring toward the end of G1 represents a critical decision point in the life of a cell; this decision must be deregulated if cancer cells are to gain proliferative advantage However, once a cell has moved past the R point and reached the G1/S transition, the remaining steps of the cell cycle proceed in an essentially automatic, pre-programmed fashion Accordingly, the S, G2, and M phases of cancer cells, which together represent an extraordinarily complex program of biochemical and cell-biological steps, closely resemble the comparable cell cycle phases of normal cells This helps to explain an often-noted aspect of cancer cells: their growth-and-division cycles are not necessarily shorter than those of many normal cells in the body Instead, cancer cells continue to enter into these cycles and thus continue to proliferate under conditions that would force normal cells to halt proliferation (see, for example, Figure 7.29) While the effects of deregulating the cell cycle clock are felt largely in late G1, there are more subtle changes that extend into later phases of the cell cycle For example, pRb–E2F1 complexes also seem to regulate the expression of genes that play key roles in organizing the complex steps of mitosis; the products of these genes contribute to chromosome condensation (during metaphase), centromere function, and general chromosomal stability These more recent discoveries now indicate that cells that have lost pRb function suffer from chromosome instability (CIN) in addition to deregulation of the R-point transition, suggesting that in pRb-negative cells, CIN conspires with deregulation of the G1/S transition to accelerate multi-step tumor progression In addition to the critical regulators of G1 progression and chromosomal stability, the only other components of the cell cycle clock that seem to be affected in the malignant growth state are the small cohort of proteins that serve as checkpoint controls during the G1/S transition, during S phase, and during M phase Their inactivation, which occurs in some cancers, is not directed toward the immediate goal of deregulating proliferation Rather, the inactivation of these checkpoint controls serves to destabilize the cellular genome, enabling incipient cancer cells to generate a wide variety of permutations of the normal human genome These cells and their descendants then test the resulting novel genetic configurations, searching for those that are particularly advantageous for neoplastic growth We will return to the theme of genomic destabilization in Chapter 12 Large gaps in this scenario remain to be filled in We still not really understand why pRb inactivation is so important for the creation of cancer cells while the inactivation of its two cousins, p107 and p130, seems rarely, if ever, to be a priority during the course of tumorigenesis We not understand how the proliferation-versus-differentiation decision is made in the great majority of the body’s cell types Relevant here may be observations indicating that pRb interacts with a number of other transcription factors in addition to the intensively studied E2Fs Some of these other transcription factors may well govern the expression of genes that contribute to certain differentiation programs And we still not understand how many oncoproteins that function as transcription factors (see Table 8.2) succeed in perturbing the workings of the cell cycle clock Moreover, many of the currently held preconceptions about the operations of the cell cycle clock, as described in this chapter, may one day require substantial revision For example, mutant mouse embryos that have been deprived of both copies of each of the three D-type cyclin genes (that is, with a D1–/– D2–/– D3–/– genotype) are able to pass through most stages of embryonic development, some dying as late as embryonic day 16 (out of the ~20 days of full-term gestation) How the cells of these embryos succeed in advancing through many cycles of growth and division without 325 Chapter 8: pRb and Control of the Cell Cycle Clock D-type cyclins driving their advance through the G1 phase of the cell cycle? Similarly, mouse embryos lacking both copies of each of the two cyclin E genes (that is, with an E1–/– E2–/– genotype) develop until mid-gestation, at which point they die because of placental defects; if these are circumvented, then the embryos can develop to term, whereupon they die The existing models provide no insight into how the cells of these embryos are able to complete the last steps of the G1 phase and initiate the S phase of the cell cycle These startling results might suggest the operations of a normally latent cell cycle clock, possibly inherited from our protozoan ancestors, that is able to assume control when the more modern, metazoan cell cycle machinery fails to its job The roles of the cell cycle clock are also expanding beyond the realm of cell cycle control, as mentioned earlier in the context of cyclin D1 and its interactions with transcription factors For example, the β cells of the pancreas are responsible for secreting insulin in response to elevation of circulating glucose level During this response, the insulin that is initially released acts in an autocrine manner on the β cells, resulting in cyclin D2 induction, phosphorylation of pRb by D2–CDK4 complexes, and resulting E2F1-dependent transcription of a gene (termed Kir6.2) that amplifies insulin secretion, greatly accelerating the insulin response and removal of excess glucose from the blood All this occurs in cells (that is, β cells) that are stably ensconced in the G0 phase of the cell cycle Examples like these indicate that the cell cycle apparatus, which was already highly developed in eukaryotic protozoa, has been adapted during metazoan evolution for various applications that are totally unrelated to control of the cell cycle While we have read about the details of cell cycle control and the various ways by which it is disrupted in many types of cancer cells, the ultimate motivation behind these discussions is a need to understand clinical disease: How these changes actually affect tumor progression and patient outcome? In fact, losses of pRb function can have particularly striking effects on the behavior of cancer cells and thus tumors For example, as indicated in Figure 8.38A, abnormally high levels of cyclin E in the cancer cells of breast carcinoma patients are strongly predictive of aggressive malignancy and poor patient outcome, while low levels indicate long-term, disease-free survival In this case, expression of the cyclin E mRNA may be elevated, and the degradative mechanisms that are normally responsible for reducing cyclin E levels are likely to (A) (B) stage III disease 1.0 proportion of patients who did not die of breast cancer 326 chromosomal DNA low total cyclin E levels 0.8 0.6 0.4 high total cyclin E levels 0.2 0.0 normal cells 10 centrosomes spindle fibers breast cancer cells with elevated cyclin E years after diagnosis Figure 8.38 Cyclin E and breast cancer progression (A) This Kaplan–Meier plot presents the clinical progression of disease in women with stage III breast cancer, that is, those having relatively large primary tumors and cancer cells in regional lymph nodes but lacking observable metastases at distant anatomical sites Plotted is the fraction of patients (ordinate) who had not died from their disease at the indicated times after initial diagnosis (abscissa) Total cyclin E includes both the high– and low–molecular-weight forms of this cyclin (B) During normal mitosis, bipolar spindles are observed (left) However, in aggressive breast cancers, low–molecular-weight forms of cyclin E accumulate in the cytoplasm of the associated carcinoma cells, where they drive the formation of multiple extra centrosomes and resulting multipolar spindles; eight are visible here (right) This leads, in turn, to karyotypic chaos and the acceleration of tumor progression (A, adapted from K Keyomarsi et al., N Engl J Med 347:1566–1575, 2002 B, from R Bagheri-Yarmand et al., Cancer Res 70:5074–5084, 2010.) Key concepts be compromised The resulting excessively high levels of cyclin E drive deregulated pRb phosphorylation and inactivation Moreover, lower–molecular-weight forms of cyclin E that are found in many aggressive tumors function abnormally to drive the formation of multiple centrosomes (see Figure 8.38B), which results in genetic instability and acceleration of tumor progression, a topic that we will pursue in depth in Chapter 12 Observations like these provide compelling indications that the cell cycle clock is, in its normal state, an important deterrent to cancer development and, when deranged by various lesions, a potent agent for promoting cancer progression Our discussions of pRb and cell cycle control would suggest the preeminent importance of this protein among all the many products of tumor suppressor genes In fact, pRb shares this position with a second protein, p53, which plays an equally important role in normal cell physiology and in cancer development In essence, the pRb circuitry deals with the relations between the cell and the outside world The p53 circuitry has a very different function, since it monitors the internal well-being of the cell and permits cell proliferation and cell survival only if all the vital operating systems within the cell are functioning properly As we will see in the next chapter, the inactivation of this p53 signaling pathway is as important to developing cancer cells as the deregulation of the controls governing pRb and the R-point transition Key concepts • The cell cycle is a precisely programmed series of events that enables a cell to duplicate its contents and to divide into two daughter cells This series of events is controlled by the machinery that is often termed the cell cycle clock • Specific steps of the cell cycle are controlled by changing the levels and availability of cyclins Cyclins function by activating the catalytic function of their partners—the cyclin-dependent kinases (CDKs), a family of serine/threonine protein kinases Additional control of the cell cycle is provided by CDK inhibitors, which antagonize the activities of cyclin-CDK complexes • While the levels of the D cyclins are controlled primarily through extracellular signals, the remaining cyclins operate on a preordained schedule, once the decision to advance into the late G1 phase has been made; the gradual accumulation of these other cyclins followed by their rapid destruction ensures that the cell cycle clock can move in only one direction • Checkpoint controls operating throughout the cell cycle ensure that a new step in cell cycle progression is not undertaken before the preceding step is properly completed and that cell cycle progression cannot proceed if a cell’s genome is damaged Many types of cancer cells have inactivated one or more of these checkpoint controls, thus helping themselves to accumulate the mutant genes and altered karyotypes that propel their neoplastic growth • The critical decisions concerning growth versus quiescence and entrance into post-mitotic differentiated states are made in the G1 phase of the cell cycle In normal cells, the decision to grow and replicate requires signals from the external environment (hence the dependence of D-cyclin levels on mitogenic signals) Thereafter, advance through the other phases of the cell cycle is relatively independent of external signals • The restriction point (R point) represents a point at which the cell commits itself, essentially irrevocably, to complete the remainder of the cell cycle or, alternatively, to remain in G1 and possibly retreat from the active cell cycle into the G0, quiescent state Deregulation of the R-point decision-making machinery accompanies the formation of most types of cancer cells, since it leads to unconstrained cell proliferation • Decisions concerning growth versus quiescence are governed by the state of phosphorylation of the three pocket proteins—pRb, p107, and p130—which in turn is controlled by the D cyclins and cyclin E pRb, the retinoblastoma protein, controls passage through the R point Hypophosphorylated pRb blocks passage through the R point, while hyperphosphorylated pRb permits this passage 327 328 Chapter 8: pRb and Control of the Cell Cycle Clock • pRb’s phosphorylation is carefully controlled Growth factors induce expression of D-type cyclins, which initiate pRb phosphorylation by hypophosphorylating it The resulting hypophosphorylation of pRb leaves it in a growth-inhibitory state, which appears to be a prerequisite for its hyperphosphorylation (and consequent inactivation) by cyclin E-CDK2 complexes • pRb acts by binding or releasing E2F transcription factors associated with promoters of genes that usher the cell from late G1 into S phase Hypophosphorylated pRb binds E2Fs, while hyperphosphorylated pRb releases them When viral oncoproteins are present, they mimic pRb hyperphosphorylation by preventing pRB from binding E2Fs • In cancer cells, a number of alternative mechanisms operate to ensure that cell proliferation is not constrained by pRb Many of these cause pRb hyperphosphorylation and resulting functional inactivation • pRb function can be lost in a variety of ways, including excessive mitogenic signals (since these lead to elevated levels of D cyclins); mutation of the Rb gene; binding of pRb by a viral oncoprotein (for example, HPV E7); and the actions of cellular oncoproteins (for example, Myc) that deregulate pRb phosphorylation or directly affect pRb activity • The control of cell differentiation is coupled to the regulation of cell cycle progression Hypophosphorylated pRB is needed to halt proliferation of cells and to facilitate their differentiation Conversely, other regulatory proteins, such as Myc and the Ids, work to inhibit cell differentiation • In most types of cancer, differentiation is partially or completely blocked In general, the more differentiated the cells are that form a tumor, the less aggressive is the disease of cancer Thought questions Why is pRb function compromised in human tumors through mutations of its encoding gene while the genes encoding its two cousins, p107 and p130, have virtually never been found to suffer mutations in the genomes of cancer cells? How cells ensure that the transcription-activating functions of the E2F transcription factors are limited to a narrow window of time in the cell cycle? What might occur if E2F function were allowed to continue throughout the cell cycle? Why have DNA tumor viruses evolved the ability to inactivate pRb function? In what ways does the Myc oncoprotein deregulate cell proliferation and differentiation? How might loss of a CDK inhibitor’s function affect the control of cell cycle advance? Why is it important that the cell cycle clock never runs backwards? What molecular mechanisms operate to ensure that once the decision to advance through the restriction point has been made, this leads to an essentially irreversible commitment to complete the remaining phases of the cell cycle through M phase? What kinds of experiments suggest that the cell cycle clock may be organized differently in certain cell types or stages of embryonic development? How are the decisions of cell growth versus quiescence coupled mechanistically with the decisions governing cell differentiation? Why must these two processes be tightly coupled? Additional reading Additional reading Adhikary S & Eilers M (2005) Transcriptional regulation and transformation by Myc proteins Nat Rev Mol Cell Biol 6, 635–645 Barbacid M, Ortega S, Sotillo R et al (2005) Cell cycle and cancer: genetic analysis of the role of cyclin-dependent kinases Cold Spring Harbor Symp Quant Biol 70, 233–240 Bashir T & Pagano M (2003) Aberrant ubiquitin-mediated proteolysis of cell cycle regulatory proteins and oncogenesis Adv Cancer Res 65, 101–144 Blais A & Dynlacht BD (2004) Hitting their targets: an emerging picture of E2F and cell cycle control Curr Opin Genet Dev 14, 527–532 Bucher N & Britten CD (2008) G2 checkpoint abrogation and checkpoint kinase-1 targeting in the treatment of cancer Br J Cancer 98, 523–528 Chen HZ, Tsai S-Y & Leone G (2009) Emerging roles of E2Fs in cancer: an exit from cell cycle control Nat Rev Cancer 9, 785–797 Classon M & Dyson N (2001) p107 and p130: versatile proteins with interesting pockets Exp Cell Res 264, 135–147 Classon M & Harlow E (2002) The retinoblastoma tumour suppressor in development and cancer Nat Rev Cancer 2, 910–917 Coletta RD, Jedlicka P, Gutierrez-Hartmann A & Ford HL (2004) Transcriptional control of the cell cycle in mammary gland development and tumorigenesis J Mammary Gland Biol Neoplasia 9, 39–53 Eilers M & Eisenman R (2008) Myc’s broad reach Genes Dev 22, 2755–2766 Ewen ME & Lamb J (2004) The activities of cyclin D1 that drive tumorigenesis Trends Mol Med 10, 158–162 Foster DA, Yellen P, Xu L & Saqcena M (2011) Regulation of G1 cell cycle progression: distinguishing the restriction point from a nutrientsensing cell growth checkpoint Genes Cancer 1, 1124–1131 Frescas D & Pagano M (2009) Deregulated proteolysis by the F-box proteins SKP2 and β-TrCP: tipping the scales of cancer Nat Rev Cancer 8, 438–449 Gil J & Peters G (2006) Regulation of the INK4b–ARF–INK4a tumour suppressor locus: all for one or one for all Nat Rev Mol Cell Biol 7, 667–677 Hochegger H, Takeda S & Hunt T (2008) Cyclin-dependent kinases and cell-cycle transitions: does one fit all? Nat Rev Mol Cell Biol 9, 910–916 Iaquinta PJ & Lees JA (2007) Life and death decisions by the E2F transcription factors Curr Opin Cell Biol 19, 649–657 Iavarone A & Lasorella A (2006) ID proteins in cancer and in neurobiology Trends Mol Med 12, 588–594 Kim WY & Sharpless NE (2006) The regulation of INK4/ARF in cancer and aging Cell 127, 265–275 Lammens T, Li J, Leone G & De Veylder L (2009) Atypical E2Fs: new players in the E2F transcription factor family Trends Cell Biol 19, 111–118 Lindquist A, Rodríguez-Bravo V & Medema RH (2009) The decision to enter mitosis: feedback and redundancy in the mitotic entry network J Cell Biol 185, 193–202 Liu H, Dibling B, Spike B et al (2004) New roles for the RB tumor suppressor protein Curr Opin Genet Dev 14, 55–64 Macaluso M, Montanari M & Giordana A (2006) Rb family proteins as modulators of gene expression and new aspects regarding the interaction with chromatin remodeling enzymes Oncogene 25, 5263–5267 Malumbres M & Barbacid M (2001) To cycle or not to cycle: a critical decision in cancer Nat Rev Cancer 1, 222–231 Massagué J (2004) G1 cell-cycle control and cancer Nature 432, 298–307 Massari ME & Murre C (2000) Helix-loop-helix proteins: regulators of transcription in eucaryotic organisms Mol Cell Biol 20, 429–440 Morgan DO (2007) The Cell Cycle: Principles of Control Sunderland, MA: Sinauer Associates Perk J, Iavarone A & Benezra R (2005) Id family of helix-loop-helix proteins in cancer Nat Rev Cancer 5, 603–614 Reed S (2003) Ratchets and clocks: the cell cycle, ubiquitylation and protein turnover Nat Rev Mol Cell Biol 4, 855–864 Sherr CJ (2001) The INK4a/ARF network in tumour suppression Nat Rev Mol Cell Biol 2, 731–737 Sherr CJ & McCormick F (2002) The RB and p53 pathways in cancer Cancer Cell 2, 103–112 Sherr CJ & Roberts JM (2004) Living with or without cyclins and cyclin-dependent kinases Genes Dev 18, 2699–2711 Siegel PM & Massagué J (2003) Cytostatic and apoptotic actions of TGF-β in homeostasis and cancer Nat Rev Cancer 3, 807–820 Varlakhanova NV & Knoepfler PS (2009) Acting locally and globally: Myc’s ever-expanding roles on chromatin Cancer Res 69, 7487–7490 Whitfield ML, George LK, Grant GD & Perou CM (2006) Common markers of proliferation Nat Rev Cancer 6, 99–106 Yamasaki L & Pagano M (2004) Cell cycle, proteolysis and cancer Curr Opin Cell Biol 16, 623–628 329 This page intentionally left blank to match pagination of print book ... 11 .1 11. 2 11 .3 11 .4 11 .5 11 .6 11 .7 11 .8 11 .9 11 .10 11 .11 11 .12 11 .13 11 .14 439 Most human cancers develop over many decades of 440 time Histopathology provides evidence of multi-step tumor 442... Factors, Receptors, and Cancer 5 .1 5.2 5.3 5.4 5.5 5.6 5.7 5.8 5.9 10 3 10 4 10 5 10 8 11 3 11 7 12 4 12 7 12 8 13 0 13 0 13 1 Normal metazoan cells control each other’s lives 13 3 The Src protein functions... ways The NF1 protein acts as a negative regulator of Ras signaling 17 5 17 7 18 0 18 2 18 8 18 9 19 3 2 01 202 204 206 209 212 216 217 227 228 228 2 31 232 234 235 238 2 41 243 248 249 254 255 7 .11 7 .12