domingo, 7 de octubre de 2007

Molecular Biology Intelligence Unit

The ultimate goal of cancer research is the development of effective anticancer
therapy. During the last several decades, the discovery of oncogenes, tumor
suppressors, growth factors, signal transduction pathways has dramatically
escalated our understanding of cancer cell biology and mechanisms of cell
transformation.1-3 Hundreds of cellular proteins and pathways have been logically
considered as molecular targets in a mechanism-based approaches of anticancer drug
development.4-6
Yet, the progress in cancer treatment has not paralleled these dramatic achievements
in basic research. Certainly, a delay must exist between identification of molecular
targets and their clinical applications. However in many other fields of medicine,
effective drugs had been found prior to identification of their molecular mechanisms,
such as aspirin, anti-malaria drugs and antibiotics. Vaccination against viruses
such as smallpox had been developed almost two centuries before the immune system
and viruses were described. The most relevant parallel to anticancer drug
development is the discovery of antibiotics. Penicillin had revolutionized the
treatment of bacterial diseases long before its molecular target was identified. The
bacterial wall, a structure that does not exist in human cells, allows penicillin to kill a
bacteria without affecting a human cell, thus exercising dramatic selectivity. In this
light, the absence of the magic bullet against cancer is consistent with the lack of a
cancer-selective target. We have learned that there are very few molecules in cancer
cells that are dispensable or absent in normal cells. One of these few, Bcr-Abl, is a
selective target in Bcr-Abl-positive leukemia, even though molecular therapeutics that
inactivate Bcr-Abl have additional targets 7 Mutant p53 is another example of a
potential cancer-selective target.8, 9 Although telomerase exists in normal cells, its
functional significance in cancer cells allows us to consider this enzyme as a
reasonably-selective target. 10 However, these and other examples do not alter the general
conclusion: proto-oncogenes and signal transduction molecules are required for
proliferation and survival of normal cells and therefore most mechanism-based therapeutics
(e.g., inhibitors of kinases) will be also toxic for certain normal cells. The
absence of cancer-selective targets is the most important problem of the anticancer
drug-screen, because compounds toxic to cancer cells also kill normal cells, therefore
side-effects are inevitable. Besides, natural compounds are synthesized by microorganisms,
plants and animals in order to kill other organisms. They are not intended to
discriminate between normal and cancer cells and cannot selectively kill cancer cells.
In light of the low probability of finding a “magic bullet”, it is not surprising that
alternative approaches emerge. These range from the targeting of endothelial cells to
protection of normal cells, from a selective delivery of drugs using tissue-specific
markers to exploiting hypoxia and drug-resistance.
PREFACE
Loss of cell cycle checkpoints is the most universal alteration in human cancer.
11-13 Furthermore, as emphasized by Hanahan and Weinberg, the large and
diverse collection of cancer-associated genes can be tied to the operations of a small
group of regulatory circuits. 2 In other words, although numerous genetic alterations
may cause loss of normal checkpoints, common strategies might be developed against
a wide variety of cancers. As suggested by Paul Nurse, this would present a more
promising approach than unspecific attempts to block cell cycle progression, which
are less likely to distinguish between cancerous and normal cells.14 Aiming at defective
cell cycle checkpoints is different from targeting cancer-specific molecules. In the checkpoint
approach, it is not necessary to target cancer-promoting or key-functional molecules
(e.g., CDK), nor the molecule which is altered in cancer (mutated, overexpressed,
etc). A target may lie upstream of the affected function or may belong to parallel
pathways. Although the same molecule will be targeted in both cancer and normal
cells, the functional outcome can be different in cells with defective checkpoints. For
example, loss of the G1 checkpoint is common in cancer cells with mutant p53. In
response to DNA damage, such cancer cells are arrested in G2. The arrest at G2/M is
dramatically sensitive to even one double strand break because failure to arrest would
lead to the irreversible loss of chromosome fragments.15 Since G2 arrest in cells lacking
p53 depends on the Chk1 kinase, inhibition of this kinase results in abrogation of
the G2 checkpoint exclusively in cancer cells lacking p53.16-18 Following treatment
with DNA damaging drugs, mitotic progression of cancer cells will result in selective
killing of cells with defective checkpoints.19, 20
Even currently used chemotherapy, such as DNA-damaging and microtubule active
drugs, is effective in the treatment of some malignancies, especially of
apoptosis-prone leukemia and lymphomas, and some solid tumors such as testicular
cancer. Of course, as expected, the toxicity to normal cells limits effectiveness of chemotherapy
in many cases. More intriguingly however is the question of why these drugs
are useful and in some cases may cure the disease. Although most of these drugs target
nonselective and even nonmechanism-based targets, such as DNA, topoisomerases,
or tubulin, their ultimate effects converge on targeting checkpoints. These drugs indirectly
target checkpoints.
Modulation of cell cycle checkpoints may result in treatment regimens with improved
therapeutic indices by exploiting the disruption of checkpoints in tumor cells.21,
22 Loss of the G2/M delay might be more consequential to a cell carrying a defect in a
G1/S checkpoint than to an otherwise wild-type cell.15 Pharmacological abrogation of
the G2 checkpoint can increase sensitivity to chemotherapy in G1-checkpoint-deficient
cells, whereas cells with normal checkpoints may take refuge in G1. Furthermore, loss
of checkpoints could be used for selective protection of normal cells.23-27 Recently it has
been shown that inhibitors of CDK can prevent chemotherapy-induced hair loss in rats.28
Exploiting defective checkpoints is only in its infancy of development. However, as is
often in the history of medicine, unintentional exploitation of checkpoint loss in cancer
might be responsible for the effectiveness of standard therapies. The link between checkpoint
control and apoptosis also tempts novel therapeutic approaches.29-31 Rational design
based on the understanding of cell cycle control coupled with utilizing novel
mechanism-based therapeutics for manipulating the cell cycle will bring anticancer chemotherapy
to a new level.32
The Book Overview
This book “Cell Cycle Checkpoints and Cancer” addresses mechanisms of normal
and cancer cell cycling, checkpoint control, the link of mitogenic signaling and cell
cycle machinery. Considerable attention is devoted to the analysis of checkpoint mechanisms
from yeast to man allowing us to understand the logic of the cell cycle. Applications
to current and future anticancer therapies is discussed throughout the book and
especially in last Chapters.
Mitogenic signaling is normally initiated on the cellular membrane by mitogens,
growth factors and cytokines.33 Not surprisingly, autocrine production of mitogens is
common in malignant transformation. Autocrine and paracrine growth factor synthesis
contribute to angiogenic and metastatic properties of transformed cells. In the
following Chapter, James McCubrey et al review the mechanisms of autocrine production
of cytokines and growth factors. The cytokine model illustrates deregulation
of autocrine cytokine expression on several levels with potential therapeutic approaches.
In additional Chapter, McCubrey et al discuss signal transduction from cytokine
receptors to cell cycle machinery via Ras/Raf-1/MEK/ERK, PI3K/Akt, Jak-STAT and
other pathways. The Chapter spotlights links between mitogenic signaling and apoptotic
machinery and mechanisms that allow cancer cell to evade apoptosis.
In normal cells, growth factors are necessary to initiate and maintain the transition
through G1 phase leading to S phase. The point at G1 at which commitment occurs
and a cell no longer requires growth factors to complete the cell cycle has been termed
the restriction point by Arthur Pardee in 1974. This discovery shaped the main direction
of the research in cell cycle regulation culminating in the discovery of cyclins and
cyclin-dependent kinases. It is important that following growth-regulating stimuli, both
inhibitors and stimulators of CDKs are simultaneously induced. The choice between
proliferation and growth arrest is determined by the state of the restriction point. The
Chapter discusses that the restriction point could be considered as a prototype of cell
cycle checkpoints.
By arresting the cell cycle, activation of checkpoints presumably allows cells to
repair DNA. In “DNA damage, cell cycle control, and cancer” Jens Oliver Funk et al
describes series of events that is triggered in cells upon DNA damage as well as a
framework for the understanding of the functions of the core components involved in the
cell cycle response to DNA damage.
Cell cycle checkpoints are not restricted to DNA damage.34 As discussed by Dunkan
Clarke et al, checkpoints are mechanisms that establish dependence relationships between
biochemically unrelated cellular processes. For example, the S-phase checkpoint
ensures that genome duplication is completed before cell division. The
topoisomerase II-dependent checkpoint ensures that the topology of the newly replicated
DNA has been correctly organized before cells begin mitosis. Distinct checkpoints
monitor mitotic spindle assembly, preventing the onset of chromosome segregation
until all the chromosomes are correctly aligned, and prevent exit from mitosis until
anaphase chromosome segregation has been completed.
The p53 tumor suppressor play a key role in checkpoint control in mammalian
cells. Levels of p53 are regulated by the Mdm-2-dependent protein degradation.35
p53 can induce growth arrest and/or apoptosis. Intriguingly, p53-mediated
apoptosis involves both transcription-dependent and independent mechanisms.36
In this book, R. Vogt Sionov, I. L. Hayon and Ygal Haupt discuss mechanisms of
p53 induction and its effect on cell cycle checkpoints.
As emphasized, p21 is an important regulator of cell cycle checkpoints. The identification
of p21 (also named WAF1 by Wafik S.El-Deiry) as a p53-inducible protein
had culminated the search for a mediator of the p53 tumor suppressor by Bert Vogelstein
and his colleagues.37 Later, Wafik S. El-Deiry and coauthors have demonstrated that
p21 is transactivated by another tumor suppressor, BRCA1.38 In famalies that
inherit breast and ovarian cancer, BRCA1 mutations account for close to 100% of
resultant cancers. As discussed in this book by Timothy MacLachlan and Wafik
El-Deiry, among other qualities of BRCA1, it is influenced by and affects directly the
position of the cell cycle and the transition from phase to phase in the cell cycle intimately
involves BRCA1. Yet, many functions of BRCA1 are not clear. The authors
summarize recent advances leading to new hypothesis.
Discovered in 1994, BCRA1 is not the last tumor suppressor identified to date.
The logic of the discovery of tumor suppressors is illustrated in the Chapter by Carlo
M. Croce and coauthors. The novel tumor suppressor FHIT, fragile histidine triad
protein, is normally expressed in epithelial tissues and is inactivated in most common cancers
including lung and breast cancer. It is inactivated in more than 50% of these
tumors. FHIT is the most common genetic alteration in human cancer.
Amato J. Giaccia and his colleagues discuss hypoxia and the cell cycle. When tumors
are more than 150 μm or approximately ten cells in diameter, they exceed their
ability to obtain sufficient oxygen by diffusion alone; and hypoxia develops. As hypoxia
plays important roles in both tumor response to therapy and malignant progression,
it is essential to understand how hypoxia affects cell cycle and molecular mechanisms
involved in this process. The Chapter provides insights in the cell cycle control
by hypoxia.
Recent studies spotlight the importance of G2 checkpoint.39 Stewart and Pietenpol
discuss DNA-damage induced G2 checkpoint signaling pathways. The Chapter discuss
mechanism of G2 checkpoint activation and G2 checkpoint maintenance. The authors
analyzed how knowledge of these signaling pathways may lead to more efficient
use of current anticancer therapies and the development of novel agents.
As emphasized by Rosandra Kaplan and David E. Fisher in “p53, apoptosis, and
cancer therapy”, the challenge in cancer therapy focuses fundamentally on the paucity
of therapeutic exploitable differences between cancer cells and normal cells. The actions
of p53 likely mediates the successful treatment responses in those few tumors in
which chemotherapy produces durable cures.
p53 and p21 act as positive regulators of accelerated senescence in tumor cells, but
they are not absolutely required for this response.40 By contrasting the functions of
p53 as a positive regulator of apoptosis and as a negative regulator of mitotic catastrophe
with secondary cell death, Igor B. Roninson et al explain conflicting and paradoxical
results in the literature. In their provocative Chapter, the authors raised a prospect
that induction of program of accelerated senescence in tumor cells may be a feasible and
biologically justified approach to cancer therapy and that the induction of permanent
cytostatic arrest could be the primary mode of treatment response in certain clinical cases.
Geoffrey I. Shapiro reviews preclinical and clinical development of small molecule
inhibitors of cyclin-dependent kinases. As more potent and selective CDK inhibitors
are now eagerly anticipated, it is important to review the preclinical and clinical
results with the agents presently under development. According to Dr. Shapiro,
as novel CDK inhibitors are developed, with improved potency and selectivity, it
will be critical to determine whether they induce cytotoxicity, or whether they
are primarily cytostatic, and to continually evaluate the selectivity of CDK inhibition
for transformed cell types.
In the final Chapter “Cell Cycle Molecular Targets and Drug Discovery” John K.
Buolamwini focuses on potential molecular targets in cell cycle regulatory pathways
and their exploitation for small molecule drug design and discovery. The inhibition of
kinase catalytic activity has been successfully achieved with small molecules that have
advanced into clinical trials for cancer therapy. More potential anticancer molecular
targets are emerging including critical oncogenic kinases and regulatory proteins identified
in the progression through mitosis. These include aurora kinases, polo-like kinases,
and the anti-apoptotic protein survivin.
References
1. Hunter T. Oncoprotein networks. Cell 1997; 88:333-346.
2. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000; 100:57-70.
3. Sherr CJ. The Pezcoller lecture: Cancer cell cycle revisited. Cancer Res. 2000; 60:3689-3695.
4. Kaelin WGJ. Choosing anticancer drug targets in the postgenomic era. J Clin Invest 1999;
104:1503-1506.
5. Shapiro GI, Harper JW. Anticancer drug targets: cell cycle and checkpoint control. J Clin Invest
1999; 104:1645-1653.
6. Gibbs JB. Mechanism-based target identification and drug discovery in cancer. Science 2000;
287:1969-1973.
7. Druker BJ, Lydon NB. Lessons learned from the development of an Abl tyrosine inhibitor for
chronic myelogenous leukemia. J Clin Invest 2000; 105:3-7.
8. Blagosklonny MV. p53 from complexity to simplicity: mutant p53 stabilization,
gain-of-function, and dominant-negative effect. FASEB J 2000; 14:1901-1907.
9. Sigal A, Rotter V. Oncogenic mutations of the p53 tumor suppressor: the demons of the
guardian of the genome. Cancer Res. 2000; 60:6788-6793.
10. Hahn WC, Stewart SA, Brooks MW et al. Inhibition of telomerase limits the growth of human
cancer cells. Nat Med 1999; 5:1164-1170.
11. Pardee AB. A restriction point for control of normal animal cell proliferation. Proc Natl Acad Sci
USA 1974; 71:1286-1290.
12. Hartwell LH, Kastan MB. Cell cycle control and cancer. Science 1994; 266:1821-1828.
13. Zhou B-BS, Elledge SJ. The DNA damage response: putting checkpoints in perspective.
Nature 2000; 408:433-439.
14. Nurse P. A long twentieth century of the cell cucle and beyond. Cell 2000; 100:71-78.
15. Paulovich AG, Toczyski DP, Hartwell LH. When checkpoints fail. Cell 1997; 88:315-321.
16. Graves PR, Yu L, Schwarz JK et al. The Chk1 protein kinase and the Cdc25C regulatory pathways
are targets of the anticancer agent UCN-01. J Biol Chem 2000; 275:5600-5605.
17. Jackson JR, Gilmartin A, Imburgia C et al. An indolocarbazole inhibitors of human checpoint
kinase (Chk1) abrogates cell cycle arrest caused by DNA damage. Cancer Res 2000; 60:566-572.
18. Monks A, Harris ED, Vaigro-Wolff A et al. UCN-01 enhances the in vitro toxicity of clinical
agents in human tumor cell lines. Invest. New Drugs 2000; 18:95-107.
19. Wahl AF, Donaldson KL, Fairchild C et al. Loss of normal p53 function confers sensitization to
Taxol by increasing G2/M arrest and apoptosis. Nature Med 1996; 2:72-79.
20. Bunz F, Dutriaux A, Lengauer C et al. Requirement for p53 and p21 to sustain G2 arrest after
DNA damage. Science 1998; 282:1497-1501.
21. Fisher DE. Apoptosis in cancer therapy: crossing the threshold. Cell 1994; 78:539-542.
22. Kastan MB, Canman CE, Leonard CJ. P53, cell cycle control and apoptosis: Implications for
cancer. Cancer Metastasis Reviews 1995; 14:3-15.
23. Pardee AB, James LJ. Selective killing of transformed baby hamster kidney (BHK) cells. Proc Natl
Acad Sci USA 1975; 72:4994-4998.
24. Darzynkiewicz Z. Apoptosis in antitumor strategies: modulation of cell-cycle or differentiation. J
Cell Biochem 1995; 58:151-159.
25. Blagosklonny MV, Robey R, Bates S, Fojo T. Pretreatment with DNA-damaging agents permits
selective killing of checkpoint-deficient cells by microtubule-active drugs. J Clin Invest 2000;
105:533-539.
26. Blagosklonny MV, Bishop PC, Robey R et al. Loss of cell cycle control allows selective microtubule-
active drug-induced Bcl-2 phosphorylation and cytotoxicity in autonomous cancer cells. Cancer
Res 2000; 60:3425-2428.
27. Chen X, Lowe M, Herliczek T et al. Protection of Normal Proliferating Cells Against Chemotherapy
by Staurosporine-Mediated, Selective, and Reversible G(1) Arrest J Natl Cancer Inst 2000;
92:1999-2008.
28. Davis ST, Benson BG, Bramson HN et al. Prevention of chemotherapy-induced alopecia in rats by
CDK inhibitors. Science 2001; 291:134-137.
29. Reed JC. Dysregulation of apoptosis in cancer. J. Clin. Oncol. 1999; 17:2941-2954.
30. Sellers WR, Fisher DE. Apoptosis and cancer drug targeting. J Clin Invest 1999; 104:1655-1661.
31. Lowe SW, Lin AW. Apoptosis in cancer. Carcinogenesis 2000; 21:485-495.
32. Lees JA, Weinberg RA. Tossing monkey wrenches into the clock: new ways of treating cancer. Proc
Natl Acad Sci USA 1999; 96:4221-4223.
33. McCubrey JA, May WS, Duronio V, Mufson A. Serine/Threonine phosphorylation in cytokine
signal transduction. Leukemia 2000; 14:1060-1079.
34. Clarke DJ, Gimenez-Abian JF. Checkpoints controlling mitosis. Bioessays 2000; 22:351-363.
35. Haupt Y, Maya R, Kazaz A, Oren M. Mdm2 promotes the rapid degradation of p53. Nature 1997;
387:296-299.
36. Haupt Y, Rowan S, Shaulian E et al. p53 mediated apoptosis in HeLa cells: transcription dependent
and independent mechanisms. Leukemia 1997; 11:337-339.
37. El-Deiry WS, Tokino T, Veculescu VE et al. WAF1, a potential mediator of p53 tumor suppression.
Cell 1993; 75:817-825.
38. Somasundaram K, Zhang H, Zeng YX et al. Arrest of the cell cycle by the tumour-suppressor
BRCA1 requires the CDK-inhibitor p21WAF1/CIP1. Nature 1997; 389:187-190.
39. Chan TA, Hwang PM, Hermeking H et al. Cooperative effects of genes controlling the G2/M
checkpoint. Genes Dev 2000; 14:1584-1588.
40. Chang BD, Broude EV, Fang J et al. p21Waf1/Cip1/Sdi1-induced growth arrest is associated with
depletion of mitosis-control proteins and leads to abnormal mitosis and endoreduplication in
recovering cells. Oncogene 2000; 19:2165-2170.
CHAPTER 1
Cell Cycle Checkpoints and Cancer, edited by Mikhail V. Blagosklonny. ©2001 Eurekah.com.
Autocrine Transformation: Cytokine Model
James A. McCubrey, Xiao-Yang Wang, Paul A. Algate, William L. Blalock
and Linda S. Steelman
Abstract
Autocrine growth factor secretion by cells is a frequent event involved in malignant transformation.
Constitutive growth factor gene expression can in turn result in the deregulation of
survival. Furthermore, autocrine and paracrine growth factor synthesis can also contribute to
the enhanced angiogenic and metatastatic properties of transformed cells converting them into
more malignant tumors. We will discuss three fundamental mechanisms which can result in
autocrine transformation; first, mutations of the cytokine or growth factor genes themselves,
second, the aberrant expression of upstream receptors, kinases, or downstream transcription
factors which can induce autocrine growth factor synthesis and third, retrovirally induced cytokine
gene expression. We will discuss possible therapeutic strategies designed to inhibit these
events. We will use as a model the interleukin-3 (IL-3) gene and discuss how the aberrant
regulation of this gene can result in the prevention of apoptosis and lead to autocrine transformation.
Cytokine Regulation of Growth
Cytokine usually refers to growth factors which often affect the hematopoietic system.
Some cytokines were initially called lymphokines because they were produced by lymphocytes
and often, but not always, functioned on lymphocytes. Even though some cytokines such as
IL-3 and granulocyte/macrophage colony stimulating factor (GM-CSF) have quite different
sounding names, they share many properties, are closely genetically linked, and were most
likely derived from a common ancestral gene which underwent tandem duplication. In this
Chapter we will use the term cytokine more frequently than growth factor. However, it should
be kept in mind that IL-3 and GM-CSF are often referred to interchangeably as lymphokines,
cytokines, and growth factors.
Cytokines can stimulate cell cycle progression, proliferation, and differentiation, as well
as, inhibit apoptosis of hematopoietic and other types of cells.1-3 Peripheral blood cells are
generated from self-renewable, pluripotential hematopoietic stem cells in the bone marrow.
Cytokines such as IL-3, GM-CSF, stem cell factor (SCF, a.k.a. steel factor, c-Kit-L, macrophage
growth factor), FL (a.k.a. Flt-3L, the ligand for the flt2/3 receptor), erythropoietin (EPO), and
others affect the growth and differentiation of these early hematopoietic precursor cells into
cells of the myeloid, lymphoid, and erythroid lineages.1-4 This Chapter will concentrate on the
regulation of IL-3 since much of the knowledge of how cytokines affect cell growth, signal
transduction, cell cycle progression, and apoptosis has been elucidated from research with IL-3
and IL-3-dependent cell lines. IL-3 was initially defined over 20 years ago by its ability to
induce the enzyme 20-α-hydroxysteroid dehydrogenase in cultures of splenic lymphocytes from
nude mice.5 However, it soon became apparent that IL-3 was being studied by a number of
2 Cell Cycle Checkpoints and Cancer
investigators under a variety of aliases. It was called persisting cell-stimulating factor (PSF),6
mast cell growth factor (MCGF),7 hematopoietic cell growth factor (HCGF),8
histamine-producing cell-stimulating factor,9 multi-colony stimulating factor (Multi-CSF),10
Thy-1-inducing factor,5 and burst promoting activity (BPA).11 All of these growth stimulatory
activities were subsequently identified as the same protein and renamed IL-3. It is apparent
that the many names by which this cytokine was known reflected its diverse biological
properties. There are over 7000 citations in the Medline® database which use IL-3 as a keyword.
Interestingly, IL-3 has remained one of the most intensively studied growth factors for
over 20 years. This may be due, in part, to its strong anti-apoptotic activities.
IL-3 acts on both myeloid and lymphoid lineages. In vivo administration of pharmacological
doses of recombinant IL-3 to mice resulted in the increased production of red blood
cells, leukocytes, and platelets.12 Moreover, over-expression of the IL-3 gene in hematopoietic
progenitors via retroviral transduction of bone marrow cells resulted in a noneoplastic,
myeloproliferative syndrome in vivo.13 Infection of primary hematopoietic cells with retroviruses
encoding IL-3 does not normally result in malignant transformation, as over-expression of a
second gene is often required which will synergize with IL-3 and lead to autonomous growth.
Experimentally, the hox2.4 gene product has been shown to synergize with IL-3 and result in
the transformation of certain hematopoeitic cells.14
In addition to stimulating proliferation and differentiation of hematopoietic cells, cytokines
such as IL-3 also promote cell survival. IL-3-dependent cells undergo apoptosis after withdrawal
of IL-3 for a prolonged period of time (12 to 48 hours, depending upon the cell type
and species of origin).15 However, addition of IL-3 to IL-3-deprived cells can prevent apoptosis
in a significant proportion of these cells.15 This anti-apoptotic function of IL-3 remains intensively
studied today. Investigation of the effects of IL-3 on apoptosis has contributed significantly
to the apoptosis/programmed cell death field. In fact, the initial clues to the function of
Bcl-2 came after the observation that over-expression of Bcl-2 prolonged the survival of
IL-3-dependent cells cultured in the absence of cytokines.16
Regulation of IL-3 Expression: TCR Ligation and Mitogen Induced IL-3
Expression
Most hematopoietic cells do not usually synthesize the 26-kDa IL-3 protein. In those cells
that do express the IL-3 gene, the gene is normally under stringent controls.17-43 In peripheral
blood, activated T cells, natural killer cells, mast cells and some megakaryocytic cells can synthesize
IL-3.17-19 For optimal IL-3 expression, T cells must be activated via the T-cell receptor
(TCR)/CD3 pathway or by agents that mimic this pathway, e.g., the combination of the phorbol
ester, phorbol 12-myristate 13-acetate (PMA) and calcium ionophores.17-19 When a T cell is
activated, aggregation of the TCR/CD3 complex occurs. Receptor aggregation is followed by a
complex series of biochemical events leading to the activation of protein kinase C (PKC) and a
rise in the intracellular concentration of Ca2+.1,20 The activated serine/threonine PKC kinase
can phosphorylate the repressor protein, inhibitor κB (I-κB), which is subsequently ubiquitinated
and degraded, thus permitting I-κB to disassociate from nuclear factor-κB (NF-κB).21 The
unmasked NF-κB nuclear localization signals present on NF-κB allow it to enters the nucleus
and transactivate cytokine gene expression.
In addition, there is a complex of proteins which also phosphorylates I-κB: the I-κB
kinases α and β (I-κKα and I-κKβ).22 I-κK phosphorylates I-κB on serine residues. The I-κB
kinases can be also activated through serine/threonine phosphorylation by the NF-κB inducing
kinase (NIK), the mitogen-activated protein kinase kinase kinase-1 (MEKK1), and Akt
(see below).23 In summary, there are multiple mechanisms to activate I-κK, which in turn
regulates I-κB and subsequently NF-κB. An illustration of the regulation of IL-3 gene expression
is presented in Figure 1. Similar mechanisms mediate the expression of IL-2, GM-CSF,
and other T-cell derived cytokines (Fig. 1).
Autocrine Trasformation: Cytokine Model 3
There are other transcription factors which modulate the expression of cytokine genes.
Increased levels of intracellular Ca2+ following TCR aggregation allows calmodulin to activate
calcineurin, a serine/threonine phosphatase.20 Activated calcineurin dephosphorylates the
cytoplasmic (c) form of the transcription factor, NF-ATc (nuclear factor of activated T
cells) enabling NF-AT to translocate to the nucleus (n). This results in the transactivation
of cytokine gene expression, including IL-3 and GM-CSF whose promoters contain NF-AT
binding sites.24-42
There are also additional kinase cascades which regulate cytokine gene expression. PKC
can also activate the Ras pathway by inactivating the GTPase activating protein (GAP), a
negative regulator of Ras.26,35-37 Ras is a member of a large multi-gene family, which encodes
small GTP-binding proteins that serve as molecular switches. Inactivation of GAP stimulates
Ras activity, which results in the enhancement of activator-protein-1 (AP-1) binding activity as
discussed below.26, 35-37 A diagram of where some of these transcription factors bind the IL-3
promoter region is presented in Figure 2. AP-1 can then stimulate cytokine gene expression,
including IL-3. Interestingly, the neurofibromatosis-1 (NF1) gene, a tumor suppressor
frequently lost in juvenile chronic myelogenous leukemia (CML), is functionally related to
GAP.44,45 NF1 likely serves to block Ras activation, thus its loss leads to constitutive Ras activation
and contributes to the generation of CML.
Regulation of IL-3 Expression: Transcription Factor Binding Sites
The cis-acting elements of the human IL-3 promoter include two activation regions separated
by an inhibitory region.24-43 These genetic elements lie within a region that extends to
Fig. 1 Activation of IL-3 and GM-CSF expression. The effects of diacylglycerol (DAG) and Ca2+ on PKC
activation and the subsequent activation of calmodulin, calcineurin, NF-κB and NF-AT. Moreover the
effects of activation of the Ras/PI3K anti-apoptotic cascade are indicated. NF-κB can also be activated by
NIK. I-κB can be targeted for degradation by serine/threonine phosphorylation mediated by PKC Akt,
I-κKα and I-κKβ. The activated transcription factors are indicated in clear ovals. Once activated, NF-κB
and NF-AT enter the nucleus and stimulate IL-3 and GM-CSF expression. The sites of inhibition by the
immunosuppressive drugs CsA, FK506 and DSG are also shown in black on this diagram.
4 Cell Cycle Checkpoints and Cancer
~300 bp upstream of the transcription start site (Fig. 2, Panel A). An inhibitory element,
nuclear inhibitory protein (NIP), has been described that binds to the IL-3 promoter, which
suppresses IL-3 transcription. This binding site for this transcription factor is located between
bp -271 to -250.24 Unfortunately no further information has been provided about this factor
and the role that deletion of this transcription factor binding site plays in leukemia.
Fig. 2 (See Figure legend on opposite page)
Autocrine Trasformation: Cytokine Model 5
The IL-3 promoter region contains sequence motifs common to many cytokine promoters,
including CK (cytokine)-1 and CK-2/GC elements. These sequences appear to be dispensable
for the activity of the IL-3 promoter.26-30 A CT/GC-rich region located between bp -76 to
-47 confers a basal transcriptional activity to the IL-3 promoter and responds to trans-activation
by the human T-cell leukemia virus type I-encoded Tax protein.31 At least four transcription
factors, (AML1, EGR1, EGR2 and DB1) have been shown to bind to this region.28-31,40-42
The binding of EGR1 and EGR2 to these sites increases IL-3 promoter activity when an appropriate
cell is activated.34 In contrast to EGR1 and EGR2, DB1 binds constitutively and
enhances the transcriptional activity of the IL-3 promoter when trans-activated by Tax.31 The
AML1 transcription factor also binds this region, but it has a higher affinity binding site in
the -175 to -135 region.40-42
Two regions of the IL-3 promoter play important positive regulatory roles in the response
of T cells to mitogens. One region is called Act-1 and is located between -175 to -135 bp.27, 28
The 5´ part of this region binds a mitogen-inducible, T-cell specific, octamer-1-associated
protein (Oct-1).27 Nuclear factor-IL3A (NF-IL3A) binds the middle region, while the 3´ Act-1
sequence contains a consensus-binding site for a cAMP responsive element binding protein
(CREB).30,38 The role of the Act-1 region is to coordinate the functions of several cis-acting
transcription factors, which leads to a maximal effect upon IL-3 transcription.27,28 The AML1
transcription factor also binds to this region.40-42 The other important transcriptional regulatory
region is located between bp -315 to -274 and contains AP-1 and Elf-1 binding sites.19,25 The
c-Jun and c-Fos heterodimer (AP-1) binds to the AP-1 site,19 while the Elf-1 site is bound by an
Ets-related transcription factor, Elf-1.25,35 Tissue specific expression of the IL-3 gene may result
from interactions between the Act-1 and this latter region.35-37
There are many kinases which regulate the activity of the transcription factors that bind
the IL-3 promoter region. Signal transduction cascades originating from extracellular signals
(including cell stress) often regulate the activities of these kinases (MEKK1, MKK4, JNK,
ERK, MKK3/6, p38MAPK, p90Rsk, and others) which in turn can regulate IL-3 expression.34, 45-47
In addition to the cis-acting elements 5´ to the IL-3 transcription start site, there is
another set of cis-acting elements found in the intergenic region between the IL-3 and GM-CSF
genes which are sensitive to the immunosuppressive drugs CsA and FK-506 (Fig. 2, Panel B).39
This region contains four NF-AT sites, which are bound by NF-AT transcription factors upon
mitogen activation.
Proto-oncogenes, which are sometimes mutated in human cancer, (e.g., c-Fos, c-Jun NF-κB,
EGR, and Ets-related transcription factors) can bind the IL-3 promoter region.45, 46 This suggests
that abnormal expression of these oncoproteins may result in autocrine transformation and
Fig. 2. (opposite page) Transcription regulation of IL-3 and GM-CSF. Panel A, The effects of activated Ras,
PKC, KSR and Src family kinases on the Raf/MEK/ERK signal transduction pathway and IL-3 expression.
Ras activation can be induced by external stimuli but inhibited by NF1 or GAP (black ovals). Active Ras
and downstream kinases are indicated in gray ovals. Ras can further transmit the signal to Raf, MEK, ERK
and p90Rsk which can result in the activation of the AP-1 and CREB transcription factors (shown in clear
ovals) which bind the promoter region of the IL-3 and GM-CSF genes. Activation of Ras can also result in
the activation of PI3K and the subsequent activation of PDK1, PDK2 and Akt which can phosphorylate
and activate IκK. Raf can also be activated by KSR and Src family tyrosine kinases. Other transcription
factors (e.g., c-Jun, Elf-1) are activated by other kinases such as JNK and p38 that in turn are activated by
MKK4, MEKK1 and MKK3/6 and SEK. The NF-κB and NF-AT as well as the Oct-1, AML1 and CREB
transcription proteins bind to the ACT-1 region. Possible control mechanisms for NF-κB, I-κB and NF-AT
(NIK, IκK and calcineurin) were presented in Figure 1. Also shown in this picture are the negative effects
of the NIP protein which binds the NIP region and suppresses transcription. In addition, there are CK1
and CK2 transcription factors, which bind to the CK1 and CK2 regions, as well as the Tax, EGR1, EGR2,
DB1 and AML1 transcription factors, which bind to the CT/GC rich region. Panel B, This panel depicts
the binding of NF-AT molecules to the intergenic region between the IL-3 and GM-CSF genes. The binding
of these proteins to the intergenic region influences the chromatin configuration of this gene cluster.
6 Cell Cycle Checkpoints and Cancer
lead to leukemia. In many transformed cells, the pathways controlling the activities of these
transcription factors are dysregulated.45,46 For example, constitutive activation of the Ras/Raf/
MEK/ERK (extracellular regulated kinase) cascade can alter the activity of
transcription factors, to induce autocrine growth factor synthesis.48,49
Genetic Influences on IL-3 Expression: DNA Methylation
Genomic DNA demethylation is also believed to influence the propagation of specific Tcell
cytokine profiles. The extent of methylation of certain cytokine genes, such as IL-3 and
interferon-γ (IFN-γ), may contribute to distinct patterns of cytokine gene expression in T-cell
clones. Demethylation of the IL-3 promoter was shown to be confined to specific CpG sites
within the same clones.50 This is a potential mechanism that could lead to the ability of certain
T-cell clones to express specific cytokines.
Therapeutic Approaches Based Upon Reducing Cytokine Gene Expression
We have discussed the mechanisms by which T-cell activation can result in the transcription
of the IL-3 gene. Now we will discuss therapies that exploit the inhibition of IL-3 transcription.
There may be therapeutic approaches to inhibit the activity of NF-κB, which will
decrease cytokine gene expression. 15-Deoxyspergualin (DSG), an immunosuppressive drug
which has been through Phase I/II clinical trials, inhibits the localization of heat shock protein
70 (Hsp 70) to the nucleus in response to heat stress, as well as the intranuclear activation of
NF-κB, through its interaction with Hsp70.51 Another approach to inhibiting NF−κB activation
involves introducing adenoviral vectors into leukemic cells which overexpress I-κB.52 This would
result in a decrease in NF-κB activity and cytokine gene synthesis. This gene therapeutic approach
may prove beneficial in the suppression of tumor growth. Decreasing the levels of
NF-ATc would suppress cytokine gene expression. The immunosuppressive drugs cyclosporin
A (CsA) and FK506 mediate their activity by inhibiting calcineurin activation, thereby
preventing the dephosphorylation of NF-ATc.39,53 Another approach is to treat leukemic
patients with immunosuppressive drugs. This approach would have to be carefully monitored
as it could render the patient susceptible to life-threatening microbiological infections.
Other targets to inhibit cytokine gene synthesis include the upstream signal transduction
cascades. Ras is frequently targeted by anti-neoplastic drugs including farnesyl transferase (FT)
inhibitors (see below). Addition of a farnesyl group is necessary for Ras localization to the
cytoplasmic membrane. Drugs, which block Ras farnesylation, are currently being developed
by pharmaceutical companies for therapeutic use (e.g., Janssen, Merck).54 Inhibiting Ras could
decrease cytokine gene expression.
Pharmacological companies have developed inhibitors to some of the kinases involved in
regulation of cytokine gene expression (e.g., SB203580 is a p38MAPK inhibitor developed by
the Smith Klein Beecham Company).55 Blocking p38MAPK activity would suppress some of the
transcription factors involved in cytokine synthesis. The critical question remains: How do we
target these inhibitors exclusively to malignant cells rather than normal cells? It may be possible
to control, either directly or indirectly to control the activities of these important regulatory
molecules in transformed cells.26,34,45,51,56
Regulation of IL-3 Expression: Post-Transcription Regulation
We have described how TCR ligation and mitogen stimulation can activate kinase pathways
resulting in the activation of transcription factors, which bind the IL-3 promoter region
and induce expression of the IL-3 gene. The next point of IL-3 regulation to be discussed is the
control of IL-3 synthesis due to post-transcriptional mechanisms. IL-3 mRNAs are very unstable
and decay within one-half to one hour after their synthesis.57-75 This property appears to
be critical for their normal function, since the degradation of IL-3 mRNA, as well as other
cytokine mRNAs, is stringently controlled.57-75 An AU-rich element (ARE) found within the
3' untranslated region (UTR) of the IL-3 and other cytokine mRNAs is involved in the regulation
of IL-3 mRNA stability (Fig. 3).58-63 These evolutionarily conserved ARE sequences
Autocrine Trasformation: Cytokine Model 7
serve to tightly regulate cytokine expression, a critical function due to the potent growth
stimulatory and anti-apoptotic effects of cytokines. A diagram of the post-transcriptional regulation
of IL-3 is presented in Figure 3.
Electrophoretic mobility shift assays (EMSAs) and UV-crosslinking experiments have identified
the proteins that bind to cytokine AREs (62, 64-71). These included proteins with apparent
molecular weights of 36-, 40-, 43-, 46-, 55-, 57-, 68- and 95-kDa. The adenine/uridine
binding protein (AUF1, also known as heterogeneous nuclear ribonuclear protein D [hnRNP
D]) was shown to bind to the IL-3 ARE, by an EMSA followed by immunoprecipitation of
IL-3 ARE binding proteins with a specific α-AUF1 antibody.62
All three isoforms of hnRNP D, which exhibit apparent molecular weights of 40-, 43-,
and 46-kDa, bind to the IL-3 ARE.62 hnRNP C also binds to the IL-3 ARE region.62 Calcium
ionophore treatment prevents/reverses binding of these proteins to the IL-3 ARE and results in
stabilized IL-3 mRNA62 (Fig. 3, Panel B). The affinities of the hnRNP D proteins for their
RNA substrates were shown to be negatively correlated with mRNA stability.69
Fig. 3. Post-transcriptional regulation of normal and mutated IL-3 expression. Panel A, The wild-type IL-3
ARE is shown which binds the indicated proteins. This IL-3 mRNA would be induced in T cells after TCR
ligation. The binding of these proteins results in mRNA with a short mRNA half-life. p36 and p95 are the
only proteins that were demonstrated by northwestern analysis to bind directly to the IL-3 gene.3,53,55 p95
is depicted as a larger sphere due to artistic constraints. The exact sequences where p36 and p95 bind the
IL-3 UTR are not known, nor is the stoichiometry of binding. Panel B, Calcium ionophores disrupt the
binding of RNA- binding proteins to the IL-3 ARE resulting in conditional stabilization of IL-3 mRNA.
This stabilized IL-3 mRNA would be detected after treatment of T-cell lines with calcium ionophores. Panel
C, The RNA-binding proteins are prevented from binding the truncated IL-3 gene in tumorigenic autocrine
transformed cells which contain an IAP provirus inserted into the IL-3 ARE.3,58-62 The prevention of
binding of these proteins results in the continuous stabilization of IL-3 mRNA. The sizes of the coding and
noncoding IL-3 sequences are not drawn to scale.
8 Cell Cycle Checkpoints and Cancer
Therapeutic Approaches Based Upon Decreasing Cytokine mRNA Stability
CsA and FK506 decrease IL-3 production by certain mast cells via mRNA destabilization
as well as affecting NF-ATc activation.74 These results suggest three possibilities: 1) CsA and
FK506 may have additional targets besides calcineurin which regulate mRNA stability, 2)
calcineurin may have other targets besides NF-AT which regulate mRNA stability or 3) NF-AT
may regulate the expression of additional genes besides cytokines, which regulate mRNA stability.
The immunosuppressive drug rapamycin, which has a different biochemical target
than CsA, also destabilizes IL-3 mRNA in certain autocrine transformed cells.75 Rapamycin
is primarily thought to affect p70 ribosomal S6 kinase (p70S6K) phosphorylation, which
subsequently modulates the efficiency of protein translation (see below). This is believed
to result from rapamycin inhibiting the mammalian target of rapamycin (mTor), which is
downstream of phosphatidylinositol-3 kinase (PI3K) but upstream of p70S6K. The mechanisms
by which the immunosuppressive drugs CsA, FK506, and rapamycin prevent the binding
of proteins to the IL-3 ARE are unknown. The drugs may alter the phosphorylation states
of ARE binding proteins, preventing them from interacting with the IL-3 ARE.
Chromosomal Translocations which may Inhibit IL-3 Expression
We have discussed how IL-3 mRNA is synthesized and regulated in normal cells. Now we
will discuss how IL-3 can be abnormally expressed in certain leukemias and lymphomas.
Chromosomal translocations have been linked to aberrant IL-3 expression. In certain human
B-cell lymphomas, chromosomal translocations between the immunoglobulin heavy chain (IgH)
locus on chromosome 14 and the IL-3 gene on chromosome 5 [t (5; 14)(q31; q32)] were
detected.76,77 The IgH enhancer, a strong tissue specific enhancer involved in many chromosomal
translocations in hematopoietic cells (e.g., Burkitt’s lymphoma, follicular B-cell lymphomas
involving Bcl-2) induces the transcriptional activation of the IL-3 gene. The role IL-3 plays in
the growth of B-cells remains controversial. IL-3-dependent pro-B cell lines have been
available since 1985. These cell lines offer support to show that IL-3 can play a role in the
growth of some early hematopoietic cells.4 It is also possible that in human B-cell lymphomas
containing a translocated IL-3 gene, IL-3 serves as a paracrine growth factor to support the
growth of neighboring cells. This, in turn, provides the necessary growth factors for the B-cell
lymphoma. For example, the IL-3 produced by the B-cell lymphoma may stimulate the expression
of: IL-4, IL-5, IL-6, IL-7, in either neighboring cells or the B-cell lymphoma, which in
turn supports the lymphoma growth. Such complicated cytokine circuitry is common, although
often we do not know which growth factor plays the critical role.
Abnormal IL-3 Expression: Inhibition Due to Chromosomal Translocations
The AML1 transcription factor is normally a transcriptional activator which binds the
promoter region of genes such as IL-3 and stimulates its expression.40-42 Some chromosomal
translocations may result in the creation of chimeric transcription factors which suppress IL-3
expression. The AML-ETO fusion protein that is generated by the t:(8; 21) chromosomal
translocation encodes a transcriptional repressor which has been shown to suppress IL-3 promoter
activity in in vitro promoter activity assays. Moreover, the t:(12: 21) translocation
encodes the chimeric TEL-AML protein which also represses transcription of the IL-3 and
other genes as measured by in vitro promoter activity assays. This chromosomal translocation
is the most commonly identified molecular abnormality in childhood acute lymphoblastic
leukemia (ALL). In freshly isolated human ALL cells which have the TEL-AML1 fusion protein,
no IL-3 was detected.40-42 Thus some chromosomal translocations appear to suppress
IL-3. The roles that these chimeric transcription factors play in leukemogenesis are being investigated
in “knock-in” mice.43 It may be that suppression of IL-3 synthesis is unrelated to the
leukemic properties of the cells and that the real targets of suppression by these chimeric
transcription factors are other genes involved in the induction of differentiation.
Autocrine Trasformation: Cytokine Model 9
Abnormal Cytokine Gene Expression Due to Retroviral Infection
Retroviruses, such as human T-cell leukemia virus-I (HTLV-I), encode sequences which
can regulate gene expression. The tax gene product of HTLV-I is a transactivator which can
induce the expression of many genes including: IL-2, IL-3, IL-15, GM-CSF, TNF, c-fos, c-jun,
and IL-2Rα chain.31 The tax protein can induce genes whose promoter regions contain CREB,
Ap-1, and serum responsive element (SRE) sequences. Although most studies on HTLV-I
infection of hematopoietic cells have focused primarily upon IL-2 and IL-15 expression, there
may be some cases where HTLV-I infection can result in abnormal autocrine IL-3 expression
in certain early hematopoietic cells which lead to autocrine transformation. Recent studies
have shown that both IL-2 and IL-15 expression are necessary for autocrine transformation; as
treatment of cells with antibodies to either cytokine by themselves did not fully inhibit the
growth of the HTLV-I-infected T cells.79 In contrast, when antibodies to both IL-2 and IL-15
were added, a greater level growth inhibition was observed.78,79
Autocrine Cytokine Gene Expression Due to Activated Raf and MEK1
Expression
We have observed that introduction of activated Raf and MEK1 genes into
cytokine-dependent cells resulted in autocrine transformation.80-89 Initially, we observed the
synthesis of GM-CSF, but not IL-3 transcripts, in cells which would grow in response to either
Raf or MEK1 expression. Moreover, the GM-CSF cytokine was detected in the supernatants,
which supported the proliferation of the parental cells. However, when we treated the
autocrine-transformed cells with neutralizing antibodies to GM-CSF, the highest level of growth
inhibition observed was approximately 50%.80,81 When we examined the expression of other
cytokines, we noticed that mRNAs encoding additional cytokines were detected in the Raf and
MEK1 transformed cells including IL-5, IL-6, IL-10, and IL-12. Some of these cytokine transcripts
(e.g., IL-5 & IL-6) were detected in uninfected cells and were observed to be regulated
in the cells by the addition of either GM-CSF or IL-3 to the growth medium. In contrast,
IL-10 and IL-12 were only detected in the cells which grew in response to activated Raf and
MEK1. The contribution of these additional cytokines to autocrine responsive growth is
being determined. Thus, the activation of downstream signal transduction cascades by Raf
and MEK1 resulted in the activation of cytokine genes that were not detected in the parental
cytokine-dependent cells.
Abnormal Regulation of IL-3 Expression: Biological Consequences of IL-3
ARE Disruption
We have characterized autocrine-transformed cells, which secrete IL-3 and have an
intracisternal type A particle (IAP) transposed into the IL-3 ARE.58,59,62 In these
autocrine-transformed FL-IL-3R cells, only two AUUUA motifs adjacent to the IL-3 gene remain
intact due to the IAP transposition in the parental FL5.12 cell line (Fig. 4, Panel A). IL-3 mRNA
isolated from these cells has a much longer half-life (T1/2 = 16 to 24 hours) than wild-type IL-3
mRNA (T1/2 = 0.5 to 1 hour). Moreover, the IL-3-secreting hematopoietic cells expressing the
mutated IL-3 mRNAs were tumorigenic upon injection into immunocompromised nude mice
(Fig. 4).57, 59-61
To determine the regions of the rearranged IL-3 gene that were responsible for the abrogation
of cytokine-dependency, chimeric IL-3 gene constructs containing portions of the wild-type
and IAP-disrupted IL-3 genes were made, transfected into IL-3-dependent parental FL5.12
cells, and examined for their ability to abrogate cytokine-dependency. The resulting
factor-independent cells were then examined for their tumorigenicity upon injection into immunocompromised
mice.59-61 Recombinant IL-3 constructs were also made which tested the
abilities of various IAP-LTRs and exogenous retroviral LTRs (e.g., Moloney-Murine Leukemia
Virus, Mo-MuLV) to affect IL-3 expression and factor-dependency. In Figure 4, Panel B, we
10 Cell Cycle Checkpoints and Cancer
have illustrated the recombinant IL-3 constructs and their abilities to abrogate the
cytokine-dependency of the parental FL5.12 cells.
Transfection of cells with a germline IL-3 gene did not result in the frequent isolation of
factor-independent cells. In those cells that were factor-independent, amplification of the
introduced GIL3 construct was detected.60,61 In contrast after transfection with the RIL3
construct factor-independent cells were detected. These transfected cells had not inherited a
Fig. 4. (see figure legend on opposite page)
Autocrine Trasformation: Cytokine Model 11
Fig. 4. (opposite page) Effects of LTRs and ARE deletions on IL-3 expression and tumorigenicity. Structures
of germline and rearranged IL-3 genes that are contained in the FL-IL3-R2 cell line and modified IL-3 genes.
Panel A. Maps of the germline (G) IL-3 locus present in FL5.12 cells and the rearranged (R)IL-3 locus
contained in FL-IL3-R cells. The black thick line is the germline IL-3 locus from start of transcription to
termination of transcription. The open thick line is the rearranged IL-3 gene from start to termination of
transcription. Boxes indicate the five IL-3 exons. Panel B. The germline and rearranged IL-3 genes as well
as various constructs containing deletions of the AUUUA regions as well as additions of different LTR and
other genetic sequences were inserted into the pSV2neo expression vector.59-61 The constructs were transfected
into IL-3 dependent FL5.12 cells, and their abilities to abrogate cytokine dependence were determined
and compared. Relevant restriction sites are indicated (E = EcoRI, B = Bam HI, N = NcoI, H = Hae
III, K = KstI). LS-IAP-LTR = Lymphocyte specific IAP-LTR (identical to the IAP-LTR contained in the rIL3
gene), Bacterial DNA = insertion of a 450 bp piece of Bacteriodes fragilis DNA. Mo-MuLV -LTR = Moloney
Murine Leukemia Virus LTR, PCR-DNA at -0.2 is the parent construct for the other LTR insertion
constructs which contain the different LTRs at -0.25. The 5´G + 3´R and 5´R + 3´G are chimeric IL-3
constructs which contain respective portions of the germline and rearranged IL-3 genes. Key to induction
of factor independence following transfection of IL-3 constructs: (–) = no or very low level of
factor-independence, ++ moderate level of factor-independence, +++ = higher level of factor-independence,
++++ = highest level of induction of factor-independence. Key to tumorgenicity: (–) no tumors or very few
(sporadic) tumors, (+) tumors in all mice examined.
high copy number of the rIL3 construct indicating that inheritance of a single rIL3 construct
was sufficient to abrogate cytokine-dependency.
The effects of the 5´ and 3´ regions of the germline and rearranged regions of the IL-3
genes were examined by creating chimeras of these genes by cleaving them in the middle with
the Bam HI (B) restriction endonuclease. This resulted in two constructs, 5´R + 3´G and 5´G
+ 3´R. The 5´R + 3´G construct, which contained the wild-type ARE sequence, did not readily
abrogate the cytokine-dependency of FL5.12 cells, whereas the construct (5´G + 3´R) which
contained the IAP-truncated ARE sequence did. These results indicated that the promoter
region of the RIL3 gene did not have any mutant elements (DNA sequences) in it which
resulted in abrogation of cytokine-dependency and the 3´R region of the RIL3 gene was responsible
for abrogation of cytokine-dependency. The DNA sequence of the RIL3 promoter
region was determined and confirmed that there were no differences in the promoter regions of
the GIL3 and RIL3 genes.
To determine whether deletion of the ARE region of the IL-3 gene was sufficient for
abrogation of cytokine-dependency, an IL-3 construct was made lacking the AUUUA region.
Transfection of cells with the GIL3 + ΔAUUUA construct did not result in the frequent isolation
of factor-independent cells. To determine if an LTR region was also required to abrogate
cytokine-dependency, an IAP-LTR was added to the gIL3 + ΔAUUUA construct. Transfection
of cells with this construct resulted in the isolation of factor-independent cells. These results
indicated that addition of the IAP-LTR was necessary for the transcription of the IL-3 gene.
Additional IL-3 constructs were made containing the various LTRs inserted in different
positions. The exogenous Moloney Murine Leukemia Virus (Mo-MuLV) LTR was more effective
in inducing the expression of the IL-3 gene than the endogenous IAP-LTR. As a control,
bacterial DNA was inserted where the various LTRs were positioned. Transfection of IL-3
dependent cells with this control construct did not result in the isolation of factor-independent
cells. LTR-CAT transient transfection experiments indicated that the LS-IAP-LTR contained
in the rIL3 gene was weaker than other LTRs and enhancer regions, thus IAP transpositions
involving this class of IAP-LTR may require additional mutagenic events to stimulate sufficient
gene transcription to induce malignant transformation.
The effects of the retroviral LTRs and the presence of the ARE on the levels of IL-3
expression in the transfected factor-independent cells are illustrated in Figure 5. This was determined
by purifying supernatants from the various cell lines and then titering them on the
factor-dependent parental cell line. The activity in the supernatants was determined to be IL-3
as treatment of the supernatants with an α-IL-3 Ab inhibited their abilities to stimulate
12 Cell Cycle Checkpoints and Cancer
[3H]-thymidine incorporation. In contrast, when the supernatants were incubated with an
α-GM-CSF Ab, there was no inhibition of [3H]-thymidine incorporation. Transfection of the
parental FL5.12 cells with a germline IL-3 construct ligated to a strong LTR (e.g., Mo-MuLV
LTR) led to the highest level of IL-3 secretion detected. In contrast transfection of FL5.12 cells
with a LTR with a low level of activity (e.g., IAP-LTR) led to a lower level of IL-3 synthesis.
Transfection of FL5.12 cells with a rearranged IL-3 gene which had a deletion of the mRNA
stability region and an IAP LTR resulted in an intermediate amount of IL-3 expression.
These studies indicated that the IAP transposition stabilized IL-3 mRNA. The remaining
two AUUUA motifs could not efficiently destabilize IL-3 mRNA, and hence, the transfected
cells were autocrine-transformed and tumorigenic. Site-directed mutagenesis studies indicated
that destabilization of IL-3 mRNA requires a clustering of either the three 5´ or the distal three
3’ AUUUA motifs present in the IL-3 ARE. However, the cluster of the three 3´ AUUUA
motifs was a stronger destabilizer.63
In order to determine how the IAP transposition altered the binding of proteins to the
IL-3 ARE, EMSAs were performed. Proteins were specifically bound to the wild-type IL-3
mRNA ARE region, whereas no protein binding was detected to the RNA which had only two
AUUUA motifs, nor to an artificial RNA probe which did not contain any AUUUA motifs
(Fig. 3, Panel C).62 Thus, certain IAP transpositions disrupt IL-3 AREs and prevent the binding
of proteins to this region. These mutations result in autocrine growth stimulation leading
to malignant transformation.
Fig. 5. Effects of LTRs on the levels of IL-3 secretion. The levels of IL-3 secreted in the various transfected
cell lines were determined by preparing supernatants from some factor-independent cells transfected with
some of the IL-3 constructs presented in Figure 4. The level of [3H]-thymidine incorporation is a measure
of DNA synthesis and a marker of proliferation. The WEHI-3B supernatant is a control since it is prepared
from the WEHI-3B cell line which produces a large amount of murine IL-3 and is a source of IL-3 for the
growth of murine IL-3 dependent cells. The gIL3 supernatant was prepared from a rare factor-independent
FL5.12 cell line transfected with the germline IL-3.59-61
Autocrine Trasformation: Cytokine Model 13
Conclusions
This Chapter has examined the mechanisms of regulation of IL-3 expression in normal
and autocrine transformed cells. We have also described therapeutic approaches which might
be effective in treating autocrine tumors. Clearly the aberrant expression of growth promoting
cytokines represents a significant challenge in cancer therapeutics because they can be activated
by diverse mechanisms.
Acknowledgments
We sincerely thank Ms. Catherine Spruill for the outstanding artwork. This work was
supported in part by grants (R01CA51025) from the NCI and the North Carolina Biotechnology
Center (9805-ARG-0006, 2000-ARG-0003) to JAM.
References
1. Blalock WL, Weinstein-Oppenheimer C, Chang F, et al. Signal transduction, cell cycle regulatory,
and anti-apoptotic pathways regulated by IL-3 in hematopoietic cells: Possible sites for intervention
with anti-neoplastic drugs. Leukemia 1999; 13:1109-1166.
2. Wang XY, McCubrey JA. Regulation of interleukin-3 expression in normal and autocrine
transformed hematopoietic cells. Int Onco 1997; 10:989-1001.
3. Wang XY, Steelman LS, McCubrey JA. Abnormal activation of cytokine gene expression by intracisternal
type A particle transposition: Effects of mutations which result in autocrine growth
stimulation and malignant transformation. Cytokines Cell Mol Therapy 1997; 3:3-19.
4. McKearn JP, McCubrey J, Fagg B. Enrichment of Hematopoietic precursor cells and cloning of
multipotential b-lymphocyte precursors. Proc Natl Acad Sci USA 1985; 82:7414-7418.
5. Ihle JN, Pepersack L, Rebar L. Regulation of T-cell differentiation: In vitro induction of
20α-hydroxysteroid dehydrogenase in splenic lymphocytes from athymic mice by a unique
lymphokine. J Immunol 1981; 126:2184-2189.
6. Schrader JW, Lewis SJ, Clark-Lewis I, et al. The persisting (P) cell: Histamine content, regulation
by a T cell-derived factor, origin from a bone marrow precursor, and relationship to mast cells.
Proc Natl Acad Sci USA 1981; 78:323-327.
7. Bazill GW, Haynes M, Garland J et al. Characterization and partial purification of a hemopoietic
cell growth factor in WEHI-3 cell conditioned medium. Biochem J 1983; 210:747-759.
8. Dy M, Lebel B, Kamoun P et al. Histamine production during the anti-allograft response.
Demonstration of a new lymphokine enhancing histamine synthesis. J Exp Med 1981; 153:293-309.
9. Schrader JW, Clark-Lewis I. A T cell-derived factor stimulating multipotential hemopoietic stem
cells: molecular weight and distinction from T cell growth factor and T cell-derived
granulocyte-macrophage colony-stimulating factor. J Immunol 1982; 129:30-35.
10. Ihle JN, Keller J, Oroszlan S et al. Biological properties of homogeneous interleukin-3. I.
Demonstration of WEHI-3 growth factor activity, mast cell growth factor activity, P cell-stimulating
factor activity, colony-stimulating factor activity, and histamine-producing cell-stimulating factor
activity. J Immunol 1983; 131:282-287.
11. Metcalf D, Begley CG, Johnson GR et al. Effects of purified bacterially synthesized murine
multi-CSF (IL-3) on hematopoiesis in normal adult mice. Blood 1986; 68:46-57.
12. Chang JM, Metcalf D, Lang RA et al. Nonneoplastic hematopoietic myeloproliferative syndrome
induced by dysregulated multi-CSF (IL-3) expression. Blood 1989; 73:1487-1497.
13. Cockayne DA, Bodine DM, Cline A et al. Transgenic mice expressing antisense interleukin-3 RNA
develop a B-cell lymphoproliferative syndrome or neurologic dysfunction. Blood 1994; 84:2699-2710.
14. Perkins A, Kongsuwan K, Visvader J et al. Homeobox gene expression plus autocrine growth factor
production elicits myeloid leukemia. Proc Natl Acad Sci USA 1993; 87:8398-8402.
15. Kinoshita T, Yokota T, Arai K-I et al. Suppression of apoptotic death in hematopoietic cells by
signalling through the IL-3/GM-CSF receptors. EMBO J 1995; 14:266-275.
16. Nunez G, London L, Hockenbery D et al. Deregulated Bcl-2 gene expression selectively prolongs
survival of growth factor-deprived hemopoietic cell lines. J Immunol 1990; 144:3602-3610.
17. Nimer S, Zhang J, Avraham H et al. Transcriptional regulation of interleukin-3 expression in
megakaryocytes. Blood 1996; 88:66-74.
18. Guba SC, Stella G, Turka LA et al. Regulation of interleukin-3 gene induction in normal human
T cells. J Clin Invest 1989; 84:1701-1706.
19. Park J-H, Kaushansky K, Levitt L. Transcriptional regulation of interleukin-3 (IL3) in primary human
T lymphocytes. Role of AP-1- and octamer-binding proteins in control of IL3 gene expression.
J Biol Chem 1993; 268:6299-6308.
14 Cell Cycle Checkpoints and Cancer
20. Clipstone NA, Crabtree GR. Identification of calcineurin as a key signalling enzyme in T-lymphocyte
activation. Nature 1992; 357:695-697.
21. Link E, Kerr LD, Schreck R et al. Purified I kappa B is inactivated upon dephosphorylation. J Biol
Chem 1992; 267:239-246.
22. Nakano H, Shindo M, Sakon S et al. Differential regulation of I-kB kinase a and b by two upstream
kinases NF-kB inducing kinase and mitogen-activated protein kinase/ERK kinase kinase-1. Proc
Natl Acad Sci USA 1998; 95:3537-3542.
23. Madrid LV, Wang CY, Guttridge DC. Akt Suppresses apoptosis by stimulating the transactivation
potential of the Rel A/p65 subunit of NF-kappaB. Mol Cell Biol 2000; 20:1626-1638.
24. Mathey-Prevot B, Andrews NC, Murphy HS et al. Positive and negative elements regulate human
interleukin-3 expression. Proc Natl Acad Sci USA 1990; 87:5046-5050.
25. John S, Marais R, Child R et al. Importance of low affinity Elf-1 sites in the regulation of
lymphoid-specific inducible gene expression. J Exp Med 1996; 183:743-750.
26. Arai N, Naito Y, Watanabe M et al. Activation of lymphokine genes in T cells: roles of cis-acting
DNA elements that respond to T-cell activation signals. Pharmacol Therapeut 1992; 55:303-318.
27. Davies K, TePas EC, Nathan DG et al. Interleukin-3 expre ssion by activated T cells involves an
inducible, T-cell-specific factor and an octamer binding protein. Blood 1993; 81:928-934.
28. Taylor DS, Laubach JP, Nathan DG et al. Cooperation between core binding factor and adjacent
promoter elements contributes to the tissue-specific expression of interleukin-3. J Biol Chem 1996;
271:14020-14027.
29. Nishida J, Yoshida M, Arai K-I et al. Definition of a GC-rich motif as regulatory sequence of the
human IL-3 gene: Coordinate regulation of the IL-3 gene by CLE2/GC box of the GM-CSF gene
in T-cell activation. Int Immunol 1991; 3:245-254.
30. Koyano-Nakagawa N, Nishida J, Baldwin D et al. Molecular cloning of a novel human cDNA encoding
a zinc finger protein that binds to the interleukin-3 promoter. Mol Cell Biol
1994; 14:5099-5107.
31. Himes SR, Katsikeros R, Shannon MF. Costimulation of cytokine gene expression in T cells by
the human T-cell leukemia/lymphotropic virus type 1 trans activator Tax. J Virol 1996; 70:4001-4008.
32. Gerondakis S, Strasser A, Metcalf D et al. Rel-deficient T-cells exhibit defects in production of
interleukin-3 and granulocyte-macrophage colony-stimulating factor. Proc Natl Acad Sci USA 1996;
93:3405-3409.
33. Ryan GR, Vadas MA, Shannon MF. T-cell functional regions of the human IL-3 proximal promoter.
Mol Reprod & Develop 1994; 39:200-207.
34. Park JH, Levitt L. Overexpression of the mitogen-activated protein kinase (ERK1) enhances T-cell
cytokine gene expression: Role of AP1, NF-AT, and NF-κB. Blood 1993; 82:2470-2477.
35. Gottschalk LR, Giannola DM, Emerson SG. Molecular regulation of the human IL-3 gene: inducible
T cell-restricted expression requires intact AP-1 and Elf-1 nuclear protein binding sites. J Exp Med
1993; 178:1681-1692.
36. Leiden JM, Wang C-Y, Petryniak B et al. A novel Ets-related transcription factor, Elf-1, binds to
human immunodeficiency virus type 2 regulatory elements that are required for inducible trans
activation in T cells. J Virol 1992; 66:5890-5897.
37. Cameron S, Taylor DS, TePas EC et al. Identification of a critical regulatory site in the human
interleukin-3 promoter by in vivo footprinting. Blood 1994; 83:2851-2859.
38. Zhang W, Zhang J, Kornuc M et al. Molecular cloning and characterization of NF-IL3A, a transcriptional
activator of the human interleukin-3 promoter. Mol Cell Biol 1995; 15:6055-6063.
39. Cockerill PN, Shannon MF, Bert AG et al. The granulocyte-macrophage colony-stimulating factor/
interleukin-3 locus is regulated by an inducible cyclosporin A-sensitive enhancer. Proc Natl Acad
Sci USA 1993; 90:2466-2470.
40. Uchida H, Downing JR, Miyazaki Y et al. Three distinct domains in TEL-AML1 are required for
transcriptional repression of the IL-3 promoter. Oncogene 1999; 18:1015-1022.
41. Mao S, Frank RC, Zhang J et al. Functional and physical interactions between AML1 proteins and
an ETS protein, MEF: implications for the pathogenesis of t(8;21)-positive leukemias. Mol Cell
Biol 1999;19:3635-3644.
42. Uchida H, Zhang J, Nimer SD. AML1A and AML1B can transactivate the human IL-3 promoter.
J Immunol 1997; 158:2251-2258.
43. Rhoades KL, Hetherington CJ, Harakawa N et al. Analysis of the role of AML1-ETO in leukemogenesis,
using an inducible transgenic mouse model. Blood 2000; 96:2108-2115.
44. Largaespada DA, Brannan CI, Jenkins NA et al. NF1 deficiency causes Ras-mediated granulocyte/
macrophage colony stimulating factor hypersensitivity and chronic myeloid leukemia. Nature Genetics 1996;
12:137-143.
Autocrine Trasformation: Cytokine Model 15
45. Dehbi M, Bedard PA. Regulation of gene expression in oncogenically transformed cells. Biochem
& Cell Biol 1992; 70:980-997.
46. van Dam H, Huguier S, Kooistra K et al. Autocrine growth and anchorage independence: Two
complementing Jun-controlled genetic program of cell transformation. Genes & Dev
1998; 12:1227-1239.
47. Ikushima S, Inukai T, Inaba T et al. Pivotal role for the NFIL3/E4BP3 transcription factor
interleukin-3-mediated survival of pro-B lymphocytes. Proc Natl Acad Sci USA 1997; 94:2609-2614.
48. Ueffing M, Lovric J, Philipp A et al. Protein kinase C-epsilon associates with the Raf-1 kinase and
induces the production of growth factors that stimulate Raf-1 activity. Oncogene 1997; 24:2921-2927.
49. Oldham SM, Clark GJ, Gangarosa LM et al. Activation of the Raf-1/MAP kinase cascade is not
sufficient for Ras transformation of RIE-1 epithelial cells. Proc Natl Acad Sci USA 1996; 93:6924-6928.
50. Fitzpatrick DR, Shirley KM, McDonald LE et al. Distinct methylation of the interferon gamma
(IFN-gamma) and interleukin-3 (IL-3) genes in newly activated primary CD8+ T lymphocytes:
Regional IFN-gamma promoter demethylation and mRNA expression are heritable in CD44(high)CD8+
T cells. J Exp Med 1998; 188:103-107.
51. Nadler SG, Eversol AC, Tepper MA et al. Elucidating the mechanism of action of the immunosuppressant
15-deoxyspergualin. Ther Drug Monitoring 1995; 17:700-703.
52. Foxwell B, Browne K, Bondeson J et al. Efficient adenoviral infection with I-κBα reveals that
macrophage tumor necrosis factor alpha production in rheumatoid arthritis is NF-κB dependent.
Proc Natl Acad Sci USA 1998; 95:8211-8215.
53. Sigal NH, Dumont FJ. Cyclosporin A, FK-506, and rapamycin: Pharmacologic probes of lymphocyte
signal transduction. Annu Rev Immunol 1992; 10:519-560.
54. M. Barinaga M. From bench top to bedside. Science 1997; 278:1036-1039.
55. McCubrey JA, Steelman LS, Moye PW, et al. Effects of deregulated Raf and MEK1 expression on
the cytokine-dependency of hematopoietic cells. Adv Enzyme Regl 2000; 40:305-337.
56. Beltman J, McCormick F, Cook SJ. The selective protein kinase C inhibitor, Ro-31-8220, inhibits
mitogen-activated protein kinase phosphatase-1 (MKP-1) expression, and activates Jun N-terminal
kinase. J Biol Chem 1996; 271:27018-27024.
57. McCubrey JA, Holland G, McKearn J et al. Abrogation of factor-dependence in two IL-3-dependent
cell lines can occur by two distinct mechanisms. Oncogene Res 1989; 4:97-109.
58. Algate PA, McCubrey JA. Autocrine transformation of hemopoietic cells resulting from cytokine
message stabilization after intracisternal A particle transposition. Oncogene 1993; 8:1221-1232.
59. Mayo MW, Wang X-Y, Algate PA et al. Synergy between AUUUA motif disruption and enhancer
insertion results in autocrine transformation of interleukin-3-dependent hematopoietic cells. Blood
1995; 86:3139-3150.
60. Wang X-Y, McCubrey JA. Malignant transformation induced by cytokine genes: A comparison of
the abilities of germline and mutated interleukin-3 genes to transform hematopoietic cells by transcriptional
and posttranscriptional mechanisms. Cell Growth & Differ 1996; 7:487-500.
61. Wang XY, McCubrey JA. Differential effects of retroviral long terminal repeats on interleukin-3
gene expression and autocrine transformation. Leukemia 1997; 11:1711-1725.
62. Wang X-Y, McCubrey JA. Characterization of proteins binding specifically to the interleukin 3
(IL-3) mRNA 3' untranslated region in IL-3-dependent and autocrine transformed cells. Leukemia
1998; 12:520-531.
63. Stoecklin G, Hahn S, Moroni C. Functional hierarchy of AUUUA motifs in mediating rapid
interleukin-3 mRNA decay. J Biol Chem 1994; 269:28591-28597.
64. Gillis P, Malter JS. The adenosine-uridine binding factor recognizes the AU-rich elements
of cytokine, lymphokine, and oncogene mRNAs. J Biol Chem 1991; 266:3172-3177.
65. Bohjanen PR, Petryniak B, June CH et al. An inducible cytoplasmic factor (AU-B) binds selectively
to AUUUA multimers in the 3' untranslated region of lymphokine mRNA. Mol Cell Biol
1991; 11:3288-3295.
66. Brewer G. An A+U-rich element RNA-binding factor regulates c-myc mRNA stability in vitro.
Mol Cell Biol 1991; 11:2460-2466.
67. Zhang W, Wagner BJ, Ehrenman K et al. Purification, characterization, and cDNA cloning of an
AU-rich element RNA-binding protein, AUF1. Mol Cell Biol 1993; 13:7652-7665.
68. Kiledjian M, DeMaria CT, Brewer G et al. Identification of AUF1 (Heterogenous Nuclear Ribonucleoprotein
D) as a component of the α-globin mRNA stability complex. Mol Cell Biol 1997;
17:4870-4876.
69. DeMaria CT, Brewer G. AUF1 binding affinity to A+U-rich elements correlates with rapid mRNA
degradation. J Biol Chem. 1996; 271:12179-12184.
70. Hamilton BJ, Nagy E, Malter JS et al. Association of heterogeneous nuclear ribonucleoprotein A1
and C proteins with reiterated AUUUA sequences. J Biol Chem 1993; 268:8881-8887.
16 Cell Cycle Checkpoints and Cancer
71. Henics T, Sanfridson A, Hamilton BJ et al. Enhanced stability of interleukin-2 mRNA in MLA
144 cells: Possible role of cytoplasmic AU-rich sequence-binding proteins. J Biol Chem 1994;
269:5377-5383.
72. Hirch HH, Nair APK, Backenstoss V et al. Interleukin-3 mRNA stabilization by a trans-acting
mechanism in autocrine tumors lacking interleukin-3 gene rearrangements. J Biol Chem 1995;
270:20629-20635.
73. Hirsch HH, Backenstoss V, Moroni C. Impaired interleukin-3 mRNA decay in autocrine mast cell
tumors after transient calcium ionophore stimulation. Growth factors 1996; 13:99-110.
74. Nair APK, Hahn S, Banholzer R et al. Cyclosporin A inhibits growth of autocrine tumor cell lines
by destabilizing interleukin-3 mRNA. Nature 1994; 369:239-242.
75. Banhozer R, Nair APK, Hirsch HH et al. Rapamycin destabilizes interleukin-3 mRNA in autocrine
tumor cells by a mechanism requiring an intact 3´ untranslated region. Mol Cell Biol 1997;
17:3254-3260.
76. Grimaldi JC, Meeker TC. The t(5;14) chromosomal translocation in a case of acute lymphocytic
leukemia joins the interleukin-3 gene to the immunoglobulin heavy chain gene. Blood 1989;
73:2081-2085.
77. Meeker TC, Hardy D, Willman C et al. Activation of the interleukin-3 gene by chromosome
translocation in acute lymphocytic leukemia with eosinophilia. Blood 1990; 76:285-289.
78. Azimi N, Brown K, Bamford RN et al. Human T cell lymphotropic virus type I Tax protein
trans-activates interleukin 15 gene transcription through an NF-κB site. Proc Natl Acad Sci USA
1998; 95:2452-2457.
79. Azimi N, Jacobson S, Leist T et al. Involvement of IL-15 in the pathogenesis of human T
lymphotropic virus type I-associated myelopathy/tropical spastic paraparesis: Implications for therapy
with a monoclonal antibody directed to the IL-2/15R beta receptor. J Immunol 1999; 163:4064-4072.
80. McCubrey JA, Steelman LS, Hoyle PA et al. Differential abilities of activated Raf oncoproteins to
abrogate cytokine-dependency, prevent apoptosis and induce autocrine growth factor synthesis in
human hematopoietic cells. Leukemia 1998; 12:1903-1929.
81. Hoyle PE, Moye PW, Steelman LS, et al. Differential abilities of the Raf family of protein kinases
to abrogate cytokine-dependency and prevent apoptosis in murine hematopoietic cells by a
MEK1-dependent mechanism. Leukemia 2000; 14:642-656.
CHAPTER 2
Cell Cycle Checkpoints and Cancer, edited by Mikhail V. Blagosklonny. ©2001 Eurekah.com.
Signal Transduction Pathways:
Cytokine Model
James A. McCubrey, William L. Blalock, Fumin Chang, Linda S. Steelman,
Steven C. Pohnert, Patrick M. Navolanic, John G. Shelton,
Paul E. Hoyle, Phillip W. Moye, Stephanie M. Oberhaus,
Martyn K. White, John T. Lee and Richard A. Franklin
Abstract
Growth factors (GF) initiate and maintain transition through G1 to S phase. GFdependence
ends with phosphorylation of Rb by cyclin-dependent kinases (CDKs),
enabling cells to pass through the restriction (R) point and to complete the remaining
phases of the cell cycle. Cyclin D-dependent kinase phosphorylates Rb leading to induction of
cyclin E which in turn activates CDK2 and collaborates with cyclin D-CDKs to complete Rb
phosphorylation. GF simultaneously induce cyclins and CDK inhibitors. Not only their ratio
but also cellular context determines response: proliferation vs arrest. R-point, a prototype of
cell cycle checkpoints, is usually lost in cancer. Loss of R-point can be exploited for selective
killing of cancer cells by cycle-dependent chemotherapy.
Cytokine-Induced Signal Transduction Resulting in Growth
and the Prevention of Apoptosis
In the previous Chapter, we discussed the mechanisms by which IL-3 is synthesized after
T-cell activation, mitogen stimulation, chromosomal translocations, and retroviral infection.
Next, it is logical to consider the effects of the synthesized IL-3 on signal transduction pathways
leading to growth and the prevention of apoptosis. The intracellular signal transducing
machinery induced by cytokines, such as IL-3 and granulocyte/macrophage colony stimulating
factor (GM-CSF), represents a promising area to exploit for the therapy of leukemia. The
ultimate goal of many of these studies described below is the development of specific compounds
or therapies, which will modulate key intermediates in signal transduction pathways.
An overview of some of the growth and anti-apoptotic pathways induced by IL-3 is presented
in Figure 1.
Neither the α nor the β chains of the specific receptors for IL-3 (IL-3R) has any obvious
homology to known signaling molecules, such as kinases, phosphatases, nucleotide binding
proteins, or src homology (SH)-containing proteins.1-36 However, the IL-3 βc-chain functions
in the activation of signal transduction pathways by recruiting the necessary kinases.10-36 An
immediate response of cells upon IL-3 activation is the tyrosine phosphorylation of Jak and
STAT proteins1,2,8,10-32 and the activation of Ras, Raf, MEK, and MAPK (mitogen-activated
protein kinase, ERK1/2). MAP kinase is a generic name referring to a group of three serinethreonine
MAP kinases (ERK, p38, and JNK).1,2,15-20 Subsequently, these signals are
18 Cell Cycle Checkpoints and Cancer
Fig. 1. Proliferative and apoptotic pathways regulated by IL-3. This diagram is an overview of the different
effects which IL-3 has on cell growth and the prevention of apoptosis. IL-3 can stimulate Jak kinases, which
activate gene expression through STAT proteins. Some of the genes that are induced by STAT stimulate
proliferation (e.g., cyclins) or prevent apoptosis (e.g., Bcl-XL), whereas others (e.g., Cis and Socs), serve to
inhibit the Jak/STAT signal transduction pathway or regulate cell cycle progression (e.g., p21CIP1). IL-3 can
also induce anti-apoptotic pathways by stimulating the Ras, or PI3K pathways, which can result in the
phosphorylation of the pro-apoptotic Bad, Gsk-3, FKHR and caspase 9 proteins. Also shown are the
negative effects of phosphatases, which can dampen IL-3 mediated signal transduction.
Signal Transduction Pathways: Cytokine Model 19
transduced to the nucleus resulting in the transcriptional induction of proto-oncogenes such as
c-myc and c-fos.21-29
Adaptor Proteins that Couple Receptors with Downstream Pathways
Upon IL-3 stimulation, the adapter molecule Shc is also rapidly phosphorylated and associates
with the phosphorylated βc chain.30,36-40 Shc contains two domains that are capable of
interacting with tyrosine-phosphorylated proteins: an N-terminal phosphotyrosine-binding
(PTB) domain and a C-terminal SH2 domain.40 The PTB domain is responsible for the physical
association of the Shc protein to the receptor βc chain.40
Phosphorylated Shc protein binds to another adapter protein, Grb2 (growth-factorreceptor-
bound protein-2), which in turn associates with the GTP exchange factor, mSos (mammalian
son of sevenless homologue), to activate Ras.38,39 The protein tyrosine kinases responsible
for the phosphorylation of Shc have been suggested, but not exclusively identified. The
kinase which phosphorylates Shc is proposed to be Jak2.35,38
IL-3 stimulation also results in tyrosine phosphorylation of an SH2-containing inositol
phosphatase (SHIP), which forms a complex with Shc, Grb2, and SOS and may act to regulate
this pathway.41 Phosphorylation of SHIP does not appear to be necessary for its IP3-phosphatase
activity; rather, it may be involved in the binding of proteins necessary for targeting
SHIP to its correct subcellular component where its catalytic activity is necessary. Indeed,
phosphorylated SHIP is found predominantely in the membrane fraction of cells.41
The Vav protein is yet another adaptor/signaling molecule activated by IL-3/GM-CSF
stimulation.42,45 Vav contains a single SH2 domain and two SH3 domains.44,45 The SH2
domain mediates the interaction of Vav with Jak2, which has been proposed to be responsible
for the phosphorylation and activation of Vav.43 Once phosphorylated, Vav can interact with
the Tec protein kinase through Tec’s SH2 domains. Tec can then bind phosphatidylinositol
3-kinase (PI3K) and initiate additional signal transduction cascades. In addition, PKC can also
activate Vav leading to Ras/Raf or potentially Ras/PI3K activation.45
The Jak-STAT Pathway
Upon binding of IL-3 to the IL-3 receptor, the IL-3 receptor α and β chains heterodimerize,
and the entire receptor oligomerizes with other IL-3 receptors.1,30-32 The association of Jak2
with the cytoplasmic membrane-proximal region of the βc chain allows for the subsequent
oligomerization, phosphorylation, and activation of Jak2 upon IL-3 receptor aggregation.32 A
diagram of IL-3 mediated signal transduction pathways is presented in Figure 2.
Jak2 is a member of a multi-gene family including Jak1, Jak2, Jak3, and Tyk2.14,32-34 A
unique characteristic of Jak family members is that they contain two tyrosine kinase domains:
a carboxy-terminal catalytic domain and an amino-terminal pseudo-kinase domain.32-34 Jak2
has been demonstrated to be the molecule responsible for some of the immediate responses of
IL-3 stimulation and is required for mitogenesis.14,32,46
Jak activity is necessary for STAT activation by non-tyrosine kinase receptors, such as the
IL-3 receptor.12-32,36,40,46-63 IL-3 receptor binding leads to the activation of Jak2 and the
recruitment and subsequent tyrosine phosphorylation of STAT5 (Y694 of STAT5a and Y699
of STAT5b).32 Tyrosine phosphorylation of STATs leads to their dimerization and activation.32
These STAT dimers then translocate to the nucleus where they act as transcription factors by
binding regulatory sites within the promoter region of immediate-early genes such as c-myc, β-
casein, Osm, and Bcl-XL,32,47 as well as feedback inhibitors of the JAK/STAT pathway (e.g.,
Cis).47 Although STAT activation requires tyrosine phosphorylation by Jak kinases, STAT
translocation to the nucleus is enhanced by threonine phosphorylation via Raf/MEK/ERK
activation.2,64-66 The consequences of activation of this pathway will be discussed later.
Constitutive activation of members of the Jak-STAT pathway has been associated with
the onset of HTLV-induced adult T-cell leukemia67 as well as v-Abl,68 BCR-ABL69 and
v-Src70-72 mediated transformation of various hematopoietic cells. Mutant Jak proteins have
20 Cell Cycle Checkpoints and Cancer
Fig. 2. IL-3 mediated signal transduction resulting in proliferation and the prevention of apoptosis. IL-3
mediates activation of the Jak/STAT and Ras/Raf/MEK/ERK signal transduction pathways. IL-3 can also
affect apoptosis by inducing the PI3K pathway which can be regulated by the PTEN tumor suppressor gene
which functions as a phosphatase. Also shown are the activation of PKC and kinase suppressor of Ras (KSR),
which can also activate the Raf pathway. Inactivated proteins are depicted in clear ovals whereas the activated
forms are depicted in grey ovals. Proteins which have a negative role on cell growth are indicated in black
ovals. ER = endoplasmic reticulum.
Signal Transduction Pathways: Cytokine Model 21
Fig. 3 Activation of Akt by IL-3. A) In cytokine-deprived cells, Akt is not localized to plasma membrane.
Also the phosphatase encoded by the PTEN tumor suppressor can result in the inactivation of PI3K. The
LY294002 drug inhibits the catalytic activity of the p110 kinase. B) When IL-3 binds the receptor, phosphorylation
of a tryosine residue on the IL-3β chain occurs creating a binding site for the PI3K p85
regulatory subunit. This results in the recruitment of p85 via an Sh2 domain. P85 in turn activates the PI3K
p110 catalytic site. p110 PI3K then phosphorylates certain membrane lipids which result in the activation
of PDK-1 and PDK-2 which occurs via their plextrin homology domains. C) PDK-1 and PDK-2 phosphorylate
Akt on two different serine/threonine residues which results in activation of Akt. D) Akt
can phosphorylate many downstream targets which result in their activation/inactivation and the
prevention of apoptosis.
22 Cell Cycle Checkpoints and Cancer
also been observed in certain patients with immunodeficiencies.73-79 Thus, modulation of Jak
and STAT activities may be a method of therapeutic intervention in HTLV-I-induced leukemias,
chronic myelogenous leukemia, immunodeficiency and other diseases which rely upon
Jak/STAT mediated signal transduction.
The PI3K/Akt Pathway
The stimulation of appropriate target cells by IL-3 also leads to the rapid activation of
PI3K.2,80 PI3K is a heterodimeric protein consisting of an 85 kDa regulatory subunit, which
contains SH2 and SH3 domains and a 110 kDa catalytic subunit.2,80-85 IL-3 stimulation leads
to the creation of a binding site for PI3K on the IL-3R. The SH2 domain of the p85 subunit
associates with this site on the receptor βc chain.51,83 The p85 subunit is then phosphorylated
which subsequently leads to the activation of the p110 catalytic subunit that in turn activates
the downstream targets p70 S6 kinase (p70S6K) and protein kinase B (PKB), also known as
Akt.86-90
The kinase which phosphorylates PI3K may be a member of the Src tyrosine kinase family,
which includes Fgr, Fyn, Hck, Lyn, Src, Syk, Tec, and Yes in hematopoietic cells.91-94 In
addition, Ras, as well as other Rac and Rho family proteins, can activate or enhance PI3K
activity.2,80-95
Activated PI3K phosphorylates certain membrane lipids which serve to activate the
phosphoinositide kinase dependent kinases (PDK-1 and PDK-2). PDK-1 and PDK-2 then
phosphorylate the Akt kinase (aka protein kinase B, PKB) on a serine and a threonine residue.2,
96 A model of IL-3 induced Akt activity and the subsequent effects on the prevention of apoptosis
is presented in Figure 3.
Activated Akt can further transduce the signal to other targets (e.g., glycogen synthase-3
[Gsk-3] and Tec family kinases) and mediate anti-apoptotic functions by phosphorylating the
pro-apoptotic Bad protein and the regulatory caspase, caspase 9 (See below).2,9,88,89,95-107 In
contrast to the inactivation of the previous molecules by Akt phosphorylation, Akt can also
phosphorylate I-κK, which phosphorylates I-κB, resulting in its ubiquitination and subsequent
degradation in the proteasome.101-108 Since I-κB is disassociated from NF-κB, NF-κB
can then enter the nucleus and transactivate gene expression. NF-κB can promote gene expression
that, under certain circumstances, stimulates growth as well as prevents apoptosis.101-108,181-187,121
Akt can also phosphorylate certain transcription factors such as the Forkhead family of transcription
factors (FKHR).97,98 Phosphorylation of the FKHR family of transcription factors prevents
their ability to transactivate the expression of certain pro-apoptotic genes including Fas.
The PI3K pathway can also result in the activation of ribosomal protein kinases. The
p70S6K is an S6 ribosomal protein kinase that phosphorylates S6 in vitro and enhances protein
translation of certain mRNAs.109 The inhibitors wortmannin and LY294002 suppress the
activity of PI3K and rapamycin can inhibit the activity of p70S6K (see Fig. 7). Alternatively,
p70S6K can be activated by PI3K-independent means as well.
The PI3K pathway is also regulated by phosphatases which serve to decrease the activity of
PI3K. The phosphatase PTEN (phosphatase and tensin homologue deleted on chromosome
ten, aka MMAC1 mutated in multiple advanced cancers) has been proposed as a tumor suppressor
gene. PTEN is a dual specificity lipid and protein phosphatase that can remove the
phosphates on PI3K-phosphorylated substrates. PTEN downregulates events catalyzed in response
to Shc, Ras and ERK activation.5
The Ras/Raf/MEK/ERK Signal Transduction Pathway
The Ras/Raf/MEK/ERK cascade is perhaps one of the best-studied signal transduction
pathways. It is centrally involved in the transmission of mitogenic and anti-apoptotic signals as it
couples information initiating from membrane receptors to transcription factors which control
gene expression. Many of the members of this pathway (e.g., Ras, Raf, MEK), as well as additional
downstream targets (e.g., c-Fos, c-Jun, and Ets) are proto-oncogenes. One important reason why
Signal Transduction Pathways: Cytokine Model 23
this pathway was one of the better studied cascades is that certain transforming retroviruses
contained activated oncogenes encoding viral homologues of some of these genes. In contrast,
only the Akt gene and the downstream NF-κB gene have been shown to have viral counterparts
in the PI3K/Akt cascade and no viral counterparts have been detected in the Jak/STAT
pathway. The Ras/Raf/MEK/ERK pathway is often aberrantly regulated in transformed cells.
Thus, elucidation of the regulation of this pathway may aid in the development of drugs which
will be useful in the treatment of various malignancies.
Ras is a small monomeric GTP-binding protein whose GTP-bound form can associate
with its downstream target, which in some cases is Raf.110-123 Because Ras is ubiquitously
expressed and often mutated in human cancer, Ras was one of the first oncogenes identified as
a potential chemotherapeutic target by pharmaceutical companies.117-122 For Ras to be functional,
it must be farnesylated by the enzyme farnesyltransferase (FT) which attaches a 15-
chain fatty acid to Ras. This modification allows Ras to be tethered to the plasma membrane.
There is a large family of Ras-related proteins, including Rho and Rac.117 The roles of these
Ras-related proteins in the growth and transformation of hematopoietic cells remains undefined,
but they may serve as potential targets for the FT inhibitors as well.
Ras frequently passes its mitogenic signal onto the Raf proteins, a family of three serine/
threonine kinases (Raf-1, A-Raf and B-Raf ) which contain binding sites for interaction with
Ras.110-116 Activated Ras will induce the translocation of Raf from the cytosol to the plasma
membrane.110-116 Thus, mutations which alter Ras activity may also perturb the actions of Raf
and the downstream cascade. Evidence suggests that this pathway is intimately associated with
the control of apoptotic machinery in myelo-monocytic cells.124
Activation of the Raf-1 pathway is essential for growth factor-induced proliferation during
hematopoiesis.125 The events that lead to activation of Raf-1 at the plasma membrane are
not fully understood. However Raf-1 activation often occurs in the presence of GTP-Ras.
Inactive Raf-1 proteins are present in the cytosol bound to 14-3-3 chaperonin proteins. The
14-3-3 proteins may bind to a cysteine-rich domain (CRD) present in Raf.126 Cytosolic Raf-1
is translocated to the plasma membrane through interactions with GTP-Ras. This occurs between
the Ras binding domain (RBD) on Raf-1 (aa 55 to 131) and the switch region of GTPRas.
116, 127 Once Raf-1 is localized to the plasma membrane, Ras can interact with the Raf-
CRD via the Ras switch-2 region.116,127 These interactions between Ras and the Raf-CRD
serve to displace the 14-3-3 proteins from Raf-1 and uncover its kinase domain, allowing the
phosphorylation of two regulatory tyrosine residues (Y340 and Y341 on Raf-1) by a Src-related
protein-tyrosine kinase.127-145
Displacement of the 14-3-3 proteins from the Raf-CRD also permits the dephosphorylation
of two regulatory serine residues on Raf-1 (S259 and S621). Once all of these changes in
phosphorylation have occurred, Raf-1 is fully activated. It has also been noted that partial
activation of Raf-1 can also be achieved through phosphorylation by other membrane-associated
kinases. There is some evidence for direct activation of Raf-1 by certain protein kinase C
(PKC) isotypes.208,214-217 The α, δ, and ε isoforms of PKC will lead to phosphorylation of Raf-
1; however, only PKC e may functionally activate Raf-1.130,137-139 Alternatively, different PKC
isoforms may stimulate autocrine growth factor loops that, in turn, activate Raf-1.146
There is evidence for crosstalk between the Jak/STAT and the Ras/Raf pathways. Activation
of Raf-1 by Jak is dependent upon recruitment of Raf-1 to the plasma membrane by Ras
and occurs by phosphorylation of Raf-1 at Y340 and Y341.67,71,130,140 Moreover, Ras may
activate both Raf and PI3K.
Certain Raf proteins appear to promote cell cycle arrest. High levels of B-Raf and Raf-1
induce p21Cip1 expression, which inhibit the kinase activities of CDK4/6 and CDK2, thereby
preventing cell cycle progression.147-150 p21Cip1 functions by binding CDK/cyclin And blocking
the phosphorylation of inhibitor pocket proteins, such as Rb. Many of the effects of the Raf
and downstream MEK1 proteins have been elucidated by conditionally-active DRaf:ER and
ΔMEK1:ER constructs.2-9,150-152 These constructs have been developed by Dr. Martin McMahon
24 Cell Cycle Checkpoints and Cancer
Fig. 4. Activation of the ΔRaf:ER and ΔMEK1:ER constructs. The ΔRaf:ER and ΔMEK1:ER retroviruses
have been used to evaluate the interactions between different signaling pathways in the abrogation of the
cytokine-dependency of hematopoietic cells. A) In the absence of either β-estradiol or 4-HT, the ΔRaf:ER,
and ΔMEK1:ER proteins molecules are believed to be complexed with heat shock proteins and present in
monomeric forms. B) Upon addition of β-estradiol or the estrogen receptor antagonist, 4-HT, dimerization
of the ΔRaf:ER or ΔMEK:ER constructs occurs as well as disassociation of the heat shock proteins. C) The
MP-1 scaffolding protein may interact with the dimerized ΔRaf:ER and ΔMEK1:ER constructs creating
a more efficient signalling complex within the cell. One means to activate Raf in cells is by cross-linking two
Raf proteins together. Thus the dimerization of the ΔRaf:ER molecules by either β-estradiol or 4HT may
resemble a natural mechanism by which Raf molecules are activated. Activated Raf and MEK are then
believed to phosphorylate their respective targets and induce gene expression in the nucleus.
Signal Transduction Pathways: Cytokine Model 25
Fig. 5. Effects of activated Raf and MEK1 on the cytokine-dependency of FDC-P1 and TF-1 cells. An outline
of the effects of activated Raf and MEK1 expression on the cytokine—dependency of the FDC—P1 and TF-
1 cell lines is presented in this figure. Factor-independent FDC-P1 and TF-1 cells can be isolated after infection
with retroviruses encoding activated forms of Raf (ΔA-Raf:ER, ΔB-Raf:ER or ΔRaf-1:ER) or MEK-1
(ΔMEK1:ER). This results in the activation of downstream ERK activity and the secretion of an autocrine
growth factor (GM-CSF) as well as other cytokines (e.g., IL-5, IL-6). In the presence of ΔRaf:ER or ΔMEK1:ER
and autocrine cytokines, a decrease in the mitochondrial membrane potential does not occur and neither
downstream activation of the caspases nor fragmentation of cellular DNA occurs. Active kinases, transcription
factors and anti-apoptotic molecules are shown in gray ovals, whereas pro-apoptotic molecules are shown in
black ovals. IL-3 and the ΔRaf-1:ER and ΔMEK1:ER are shown in clear ovals.
26 Cell Cycle Checkpoints and Cancer
(University of California San Francisco) and can be related by the addition of β-estradiol or the
estrogen receptor antagonist 4-hydroxy-tamoxifen (4HT). A model for the activation of the
ΔRaf:ER and ΔMEK1:ER constructs is presented in Figure 4.
We have observed that the introduction of activated Raf and MEK1 oncogenes into the
FDC-P1 and TF-1 hematopoietic cells resulted in the abrogation of the dependency upon
exogenous growth factors in some of the cells. There was a hierarchy in terms of the ability of
the different Raf genes to abrogate cytokine-dependency as the ΔA-Raf:ER was more efficient
than the ΔRaf-1:ER which in turn was more efficient than either ΔMEK1:ER or ΔBRaf:
ER.151,152 Thus the weakest Raf kinase, ΔA-Raf:ER was more efficient in abrogating the
cytokine-dependency of these hematopoietic cells. In contrast, ΔB-Raf:ER, the strongest Raf
isoform, was more efficient in relieving contact inhibition in fibroblast NIH-3T3 cells. The
hematopoietic cells which grew in response to Raf and MEK1 activation synthesized
sufficient GM-CSF to promote autocrine growth. A model for the effects of Raf and
MEK1 on the downstream signal transduction and apoptotic pathways in FDC-P1 and
TF-1 cells is presented in Figure 5.
It is conceivable that there are specific interactions between certain Raf and 14-3-3 family
members.153-163 These interactions may modulate the activity of Raf proteins and regulate
their ability to lead, either directly or indirectly to the phosphorylation of the pro-apoptotic
Bad protein. Phosphorylation of Bad, which leads to sequestering of Bad by 14-3-3, can also
inhibit apoptosis (see below). We have shown that the B-Raf oncoprotein was the least efficient
Raf oncoprotein in abrogating the cytokine-dependency of human hematopoietic cells.151,152
This may be a reflection of the enhanced capacity of the B-Raf oncoprotein to induce the
expression of cell cycle inhibitory proteins or prevent the phosphorylation of the Bad protein.
In summary, there may be a delicate balance between inducing cell growth and inducing
cell cycle arrest in hematopoietic cells. A Raf oncoprotein with high kinase activity may
actually be the least efficient protein in terms of abrogating cytokine dependency. These studies
indicate that results obtained with fibroblastic models may not always be relevant for other cell
systems (e.g., hematopoietic cells).
The Ras/Raf/MEK/ERK Pathway: Downstream Kinase Activation
Raf activates the dual specific serine/threonine and tyrosine kinase, MEK1, which, in
turn, activates the MAP kinases ERK1 and ERK2 (p42/p44). In cells expressing normal MEK1,
the kinase appears as a 45-kDa protein.164-173 The amino terminal end of the kinase has a
negative-regulatory domain, since deletion of these residues results in constitutive activation of
MEK1, while the catalytic activity is localized to the carboxyl terminus of the protein.174,178
Proline-rich sequences between kinase domains IX and X of MEK1 are required for Raf-1
binding and subsequent activation of MEK1.164
Raf-1-mediated activation of human MEK1 requires the phosphorylation of MEK1 serine
residues 218 and 222.165-168 Substitution of either of these residues with aspartic or glutamic
acid results in a 10- to 50-fold increase in MEK1 activity.177,178 When both serines are replaced
with aspartic acid or aspartic and glutamic acid (218 & 222), MEK1 activity was 400-to 6000-
fold greater.164-168,177,178 These substitutions are believed to confer to the MEK1 protein a
configuration that is constitutively active.167,168 When these MEK1 mutants were transfected
into NIH3T3 cells, constitutive activation of p42/p44 ERKs occurred as well as foci formation.
164-168 This suggests ERK activation through Raf requires MEK1. In support of this hypothesis,
PD98059, a MEK1 inhibitor developed by Parke-Davis, prevents ERK activation
mediated by activated Raf constructs.151,152,179,180
In addition to the three Raf kinases, several other kinases influence MEK activity. One
such kinase is the oocyte-expressed proto-oncogene, Mos.169,170 Constitutively active forms of
Mos (v-Mos) transform fibroblasts via a MEK1-ERK-dependent pathway implicating Mos as a
MEK kinase.170 Mos preferentially phosphorylates MEK1 on serine 222.169,170 In contrast, the
MEK kinase-1 (MEKK1), which is associated with stress-activated pathways (SAPK) and whose
Signal Transduction Pathways: Cytokine Model 27
Fig. 6. Synergy between Raf/MEK and PI3K/Akt in the abrogation of cytokine-dependency of FL5.12 cells.
The stimulation of the Raf/MEK/ERK and Akt/PI3K pathways can lead to autocrine transformation of
FL5.12 cells. Inactive kinases are depicted in black ovals. Activated kinases, phosphatases and transcription
factors are depicted in clear ovals. Induced expression of ΔRaf:ER and Akt:ER may lead to GM-CSF
expression transcription and autocrine transformation. The steps after MEK1 activation, which result in
GM-CSF transcription, are not known at the present time, although a plausible pathway is indicated.
Potential effects of Raf and Akt on p38MAPK activation are indicated in dotted lines. Also shown are
additional pathways regulated by Akt which may be activated (TCF Responsive gene transcription) or
inactivated (Fas).
28 Cell Cycle Checkpoints and Cancer
activation is associated with the induction of apoptosis, phosphorylates MEK1 on serine 218.179
Thus, MEK1 may represent a common intermediate in pathways that exert anti- and proapoptotic
effects.
Interactions Between the Raf/MEK/ERK
and the PI3K/Akt Pathways
Another potential means of activating MEK1 is via the PI3K pathway.171 The effector
directly leading to MEK1 activation in this pathway has not been determined but does not
appear to be Raf-1.171 Active PI3K leads to the induction of ERK1 and ERK2 and may be
responsible for the prolonged ERK activity observed in certain cytokine-stimulated cells.171
Other activators of Raf, such as PKC, can also result in MEK1 and ERK activation.172,173 We
have recently observed that activated PI3K or Akt expression will synergize with activated Raf/
MEK1 expression and result in the abrogation of the cytokine-dependency of the FL5.12 hematopoietic
cell line. In contrast, activated Raf, MEK1, PI3K, or Akt expression by themselves
was sufficient to abrogate the cytokine-dependency of the other cells. In these cells, higher
levels of activated ERK1,2, p38MAPK and JNK were detected than in their cytokine-dependent
counterparts. These results suggest that activation of both Raf/MEK1 and PI3K/Akt enhance
the expression of signal transduction pathways leading to uncontrolled growth and the prevention
of apoptosis. These cells also expressed autocrine growth factors (see below). A model for
the interactions between these two signaling pathways is presented in Figure 6.
The Ras/Raf/MEK/ERK Pathway: A Tether Enhancing Signal
Transduction
Recently, a MEK1 binding partner, MP-1, was identified and shown to enhance the enzymatic
activation of the MAP kinase cascade.181MP-1 is a nonenzymatic, scaffolding protein
that serves to anchor both the MEK and ERK proteins facilitating the activation of both proteins.
Raf has not been demonstrated to be present in this scaffolding complex. It was suggested
that MP1 functions as an adapter to enhance the efficiency of the Ras/Raf/MEK/ERK cascade.
A functionally similar protein (JIP-1) has also been identified, which binds another MAP
kinase family member, JNK.182 The JIP-1 protein also serves as a scaffolding protein to bind the
MLK and MKK7 proteins, upstream activators of JNK, in a complex with JNK (see below). Thus,
these signal transduction cascades also contain other matrix proteins that serve to form scaffolding
devices, which can enhance sequential activation of downstream substrates.
The Ras/Raf/MEK/ERK Pathway: Regulation of Downstream
Transcription Factors
Ultimately, the signals generated by the Raf/MEK/ERK pathway are transmitted to the
nucleus where they lead to activation of various transcription factors necessary for the regulation of
cell growth and differentiation.183-194 Raf-1 can activate the c-Jun transcription factor. 191 c-Jun
and c-Fos heterodimers bind AP-1 driven elements which are contained in promoter regions of
many cytokine and immediate-early genes. Raf activity is also linked to activation of another
AP-1-like site by phosphorylation of the TAR, ATF3, and c-Jun transcription factors.161,163,164
This protein complex binds a response element and leads to cell survival. The removal of
growth factors leads to formation of a JunD/ATF3 complex, which acts as a repressor of this
response.191,194
Activated ERK can enter the nucleus where it acts as a kinase and phosphorylates certain
key regulatory proteins. For example, activation of Elk-1 by ERK occurs in the nucleus where
active Elk-1 binds the serum response element (SRE) contained in the promoter regions of
certain cytokine and immediate-early genes (i.e., c-fos).185,186 In addition, ERK can also directly
activate p90Rsk, an S6 kinase family member. p90Rsk activates the cyclic adenosine
monophosphate response element (CRE) binding protein (CREB) which binds CRE response
Signal Transduction Pathways: Cytokine Model 29
elements in cytokine and immediate-early gene promoters (See Chapter 1).188,190,195 Thus,
induction of the Raf pathway leads to activation of at least three transcription factors that bind
elements contained in some cytokine gene promoter regions (CREB, c-Fos and c-Jun).
Induction of Autocrine Gene Expression by Altered Raf/MEK
and PI3K/Akt Expression
An area which has not been well investigated is the mechanism by which the Raf/MEK/
ERK signal transduction pathway stimulates cytokine gene expression. Is this through the activation
of transcription factors such as Elk-1 and CREB or by Raf activating the PI3K/Akt
pathway, which can in turn activate p70S6 kinase that leads to protein stabilization? We have
observed enhanced PI3K activity after activation of Raf in Raf-responsive hematopoietic cells
(See Figs. 5 and 6).9 In these cells, Raf also induced p70S6 kinase activity. An additional mechanism
by which Raf could enhance cytokine gene expression is by increased protein synthesis
due to elevated p70S6K activity. The effects of signal transduction pathways on protein translational
efficiency leading to increased levels of cytokine expression have not been well investigated.
Mutations of Ras/Raf/MEK/ERK Cascade which Result in Neoplasia
Ras is one of the most frequently mutated oncogenes in human cancer.110,111 There are
three Ras related genes: Ha-Ras, Ki-Ras and N-Ras.110, 111 Mutations in Ras are observed in 10
to 50% of patients with myelodysplastic syndrome and acute myelogenous leukemia.196-200
These lesions often result from point mutations in three conserved codons, 12 (Ha-Ras,
Ki-Ras), 13 (Ki-Ras), or 61 (Ha-Ras, N-Ras), which convert all three Ras proteins into
constitutively active proteins.196-200 Deregulated Ras often has downstream effects which results in
altered Raf activity.
Deregulated Raf expression has been observed in diverse neoplasias including hematopoietic,
breast, cervical, renal, laryngeal, hepatocellular, small cell lung carcinomas, and in lung biopsies
recovered from cigarette smokers.201-230 However, the role(s) of Raf in the initial transformation
events are not clear, since the biopsied cells were established tumors.219 Mechanisms responsible
for activated Raf expression include point mutations, deletions, gene rearrangements, and gene
amplifications.221-228
Constitutive activation of MEK1 has been associated with a variety of neoplasias including
hepatocellular carcinoma, renal cell carcinoma, breast cancer, squamous cell carcinoma, AIDSrelated
Kaposi’s sarcoma, acute myelogenous leukemia, and chronic myelogenous
leukemia.219,227-243 In addition, constitutive activation of other downstream members of this pathway
has been implicated in oncogenic processes such as invasion, metastases, angiogenesis, and
radioresistance.236-240 One factor involved in invasiveness of cancers is the urokinase-plasminogen
activator gene whose synthesis is increased following ERK activation.210,211
Regulatory Phosphatases of the Ras/Raf/MEK/ERK Pathway
MAP kinase phosphatase-1 (MKP-1) is a phosphatase that is activated by mitogenic signals
and calcium.242,243 This phosphatase serves to turn off activated ERK in a negative feedback manner
and inhibits ERK-stimulated DNA synthesis.242 Alterations in MKP-1 expression have been
observed in prostate, colon, and bladder cancer where MKP-1 is over-expressed in the early stages,
but progressively diminishes with higher histological grade and metastases.242,243 Moreover, MKP-
1 over-expression inhibits the differentiation of myoblasts.244 In v-Raf-transformed cells, MKP-1
expression appears to be suppressed.245 This suppression has been speculated to occur via feedback
regulation of MKP-1 by v-Raf.245 MKP-1 might be a target for gene therapy in aggressive hematopoietic
tumors, which have lost MKP-1 expression. The involvement of this and other phosphatases
in hematopoietic malignancies is an area of research in its infancy and requires
further investigation.
30 Cell Cycle Checkpoints and Cancer
Fig. 7. MAPK signaling pathways activated by stress. In addition to the Ras/Raf/MEK/ERK pathway, there
are other signal transduction pathways which may crosstalk (interact) with each other. These pathways can
be induced by many different stimuli that induce cell stress. The kinases, which are functionally similar, are
at the same horizontal positions in the pathway. Additional related molecules and other names are shown
on the left and right hand sides of the figure. Kinases are indicated in gray ovals and transcription factors
are indicated in clear ovals.
Signal Transduction Pathways: Cytokine Model 31
Alternative MAPK Pathways Activated by Stress
In addition to the Jak/STAT, PI3K/AKT, and Raf/MEK/ERK pathways, there are alternative
signal transduction cascades activated by Ras-dependent and independent mechanisms.246-253
A diagram illustrating these pathways is presented in Figure 7. These related pathways can
interact with the Raf/MEK/ERK pathway. Ras-dependent activation of the MEK1 kinase
(MEKK1), a serine/threonine kinase, is responsible for activating three MAPK pathways.247-249
MEKK1 phosphorylates and activates MEK1, MEK2,192,250 and the dual serine/threonine
and tyrosine stress/extracellular regulated kinases (SEKs).250-255 Activation of the SEKs (MKK4
and MKK3/MKK6) results in the activation of the SAPKs, JNK, and p38MAPK.250-254 JNK is
a 46-kDa MAP kinase responsible for activating the AP-1 transcription factor component c-
Jun by phosphorylating it on serine residues 63 and 73.258 p38MAPK is responsible for phosphorylating
and activating certain members of the AP-1 transcription factor family, a C/EBP
transcription factor family member (CHOP), and the MAPK-associated protein kinase (MAPKAP).
251,252,256 We have observed that both of these kinases are activated after IL-3 and hydrogen
peroxide treatment of certain hematopoietic cells (JTL, JAM and RAF manuscript in preparation).
Activation of MEKK1 preferentially leads to the activation of JNK > p38MAPK > ERK1
and ERK2.251 On the other hand, Raf-1 leads to the activation of ERK1 and ERK2 but does
not result in activated JNK or p38MAPK.175 Although pharmaceutical companies have produced
specific inhibitors to p38MAPK, direct inhibitors to JNK and ERK have not yet been
developed.65
Recent evidence suggests that a significant amount of crosstalk occurs among the PI3K,
MEKK/SEK/SAPK, and Raf/MEK/ERK pathways.109,257-267 PI3K has been implicated in
MEKK1 activation as well as MEK1/ERK activation,171,260,262 whereas oncogenic Raf-1 and
MEK1 have also been reported to activate p70S6K in a PI3K-independent way.261
Default Pathways which Dampen Signaling
Signal transduction pathways are also under the regulation of phosphatases which serve to
dampen cytokine stimulated signaling.268-279 p145 SHIP is purported to be an inhibitory growth
regulator which down-regulates PI3K and Ras signaling activities. SHIP contains a conserved
SH2 domain and an inositol polyphosphate-5-phosphatase domain.39,83,268-283 Targeted
disruption of SHIP has suggested an important role of this molecule in controlling cytokine
signaling. SHIP knockout mice have increased numbers of granulocyte-macrophage progenitors,
possibly as a consequence of hyper-responsiveness to stimulation with IL-3 and other cytokines.275
Thus, we see the consequences of a deregulated IL-3 signal transduction pathway on the growth of
hematopoietic cells. In addition to the association of the Shc protein with SHIP, IL-3 induces the
transient association of SHIP with another phosphatase, SHP-2.39,269-275 The intracellular levels
of this complex may influence whether a cell proliferates or undergoes apoptosis.
Binding sites are also present on the IL-3βc chain for SHP-1, a tyrosine phosphatase bearing
two SH2 domains.274-276 SHP-1 is also known as HCP, SH-PTP1, and PTP1C. SHP-1 associates
with the βc chain via its N-terminal SH2 domain.284 SHP-1 is preferentially
expressed in hematopoietic cells275 and negatively regulates the growth stimulation induced by
hematopoietic growth factors.284 SHP-1 activity is an essential component for controlling the
events of cell activation induced by hematopoietic growth factors as evidenced by a mutation at
the SHP-1 gene, which is responsible for the moth-eaten phenotype observed in mice.277 Motheaten
mice have severe immunodeficiency and autoimmune syndromes. SHP-1 negatively regulates
erythropoiesis based upon the observations that erythroid cells from these mice are hypersensitive
to Epo.275 Clearly, cytokines induce phosphatases which play critical roles in regulating
signal transduction. Thus, cytokines also induce inhibitory molecules that serve to dampen the
effects of cytokines. Through the studies of knockout or naturally occurring mutant mice, it is
becoming apparent what devastating effects dysregulation of these phosphatases can have on
the development of the hematopoietic system, as well as viability and health.
32 Cell Cycle Checkpoints and Cancer
Fig. 8. Signal transduction pathway inhibitors. Sites of intervention of pharmaceutically-derived chemical
drugs and naturally-derived proteins (in black). Deletion or mutation of the genes encoding the PTEN,
MKP-1 and SHP-1 proteins shown in black can contribute to hematological defects and in some cases
malignant transformation.
Signal Transduction Pathways: Cytokine Model 33
Jak/STAT Inhibitors
Recently a family of proteins which inhibits the activation and/or function of Jaks and
STATs has been described. This family is referred to as the Cis family of proteins and consists of
Cis, Soc1, Soc2, and Soc3, as well as other proteins.48,285-289 Some of these proteins inhibit
multiple Jak and STAT proteins, whereas others only suppress a specific Jak or STAT
protein.48,285-289 In addition, expression of Soc proteins appears to be tissue-specific. These
inhibitors can function either by binding the Jak protein and blocking its activity or by binding
the cytokine receptors to prevent Jak binding and subsequent activation of the signal transduction
cascade. Moreover, there are naturally occurring STAT isoforms, which can function as dominantnegative
mutants.296 Thus, it may be possible to target the Jak-STAT pathway in certain cells
by introducing vectors which over-express these natural inhibitory proteins.
Finally, certain drugs, such as the tyrphostins AG490, AG198, and AG6450, specifically
inhibit Jak kinases as well as the progression of lympoblastic leukemia.365,366 Thus, there exists
both natural and chemical means to inhibit Jak and STAT activities. A diagram of potential
regulatory sites of these inhibitors is presented in Figure 8.
PI3K/p70S6K Inhibitors
Wortmannin, LY294002 and rapamycin exert their effects at different points in the PI3K/
p70S6K pathway.263,264,292 Wortmannin and LY294002 inhibit PI3K, while the immunosuppressive
drug rapamycin inhibits p70S6K via an upstream target of rapamycin (mTOR) (Fig.
8). However, once cells have entered the cell cycle, inhibition of p70S6K does not stop proliferation.
259,267 Use of these inhibitors has documented the PI3K-independent activation of
p70S6K.266 Rapamycin has been recognized for years as a potentially useful immunosuppressive
drug.293 Thus, derivatives of rapamycin may be useful in the treatment of patients
with certain leukemias.
Ras/Raf/MEK/ERK Pathway Inhibitors
Companies, including Janssen Pharmaceutica, Schering Plough, Merck & Company, and
Parke-Davis either plan to perform clinical trials or have clinical trials in progress with FT
inhibitors.117,122,294 In addition, there are also mutations that eliminate Raf-1 kinase activity.
Site-directed mutations, which change the lysine in the ATP-binding site of the catalytic domain,
render the kinase inactive and can serve as dominant-negative mutations.295,296 Kinaseinactive
mutants of Raf-1 inhibit Ha-Ras-mediated cell transformation although Ha-Ras can,
in some cells, signal through other molecules that will allow growth. Use of dominant-negative
mutants of upstream activators of MEK1, as well as the MEK-specific inhibitor PD98059, blocked
MEK1 and ERK activation.179,180 The MEK1 inhibitor, in turn, blocked Ras- and Raf-mediated
transformation and cytokine-stimulated growth.180 PD98059 and related drugs that inhibit MEK1
activity may be useful in turning off this major pathway in rapidly proliferating malignant cells.
Inhibitors to MKP-1 (e.g., Ro-31-8220) or related phosphatases have been isolated, suggesting
that it may be possible to modulate the activity of these proteins in certain leukemias.297 At first
glance, it may appear that this type of inhibitor might enhance proliferation.
PKC Inhibitors
Several pharmaceutical companies have focused on modifying PKC activity as a potential
therapeutic treatment. Many PKC inhibitors (G0, GF) as well as activators, including Bryostatin,
are being used in clinical trials with patients who suffer from certain types of cancer.2,297-300 Prolonged
activation of PKC by inducers like Bryostatin result in PKC downregulation.2 Thus, activation
of PKC by such differentiation inducing compounds as Byrostatin may be a viable therapeutic
approach. An obvious problem with PKC inhibitors is that they may affect multiple signal
transduction pathways leading to growth.
34 Cell Cycle Checkpoints and Cancer
Cytokine Regulation of Cell Cycle Progression
We have addressed control of IL-3 expression and the effects of IL-3 on signal transduction
cascades. An IL-3-dependent cell in G0 (a quiescent resting stage theoretically out of the cell
cycle)/G1 (Gap1) receives positive (presence of IL-3) and negative (absence of IL-3) stimuli. These
signals are integrated to decide whether or not the cell should enter the cell cycle.23,26,301,302 The
decision whether to progress from G1 into S phase is a position in the cell cycle termed the
restriction (R) point.302 Coordination of cell cycle progression is achieved through the activity
of the cyclins and their associated cyclin-dependent kinases (CDKs), as well as interactions
with tumor suppressor genes (e.g., pRb, p53, p21Cip1) that regulate the activity of these
complexes.
Links Between the Ras/Raf/MEK/ERK Pathway and Cell Cycle Proteins
The Ras-activated signal transduction pathway provides a link between IL-3 and stimulation
of cell cycle machinery.303-309 Ras is required for cell cycle progression and activation of both
CDK2 and CDK4 before the G1/S transition. Ectopic overexpression of Ras or its downstream
molecules, such as ERK or Ets-2,305 can lead to the induction of cyclin D but has little or no
effect on cyclin E or A. 303,305-309 The regulation of cyclin D is thought to be the more critical
and limiting step because of its rapid degradation. Its expression is dependent on continued
growth factor stimulation until cells pass the G1 restriction point. The cell cycle will also be
discussed in more detail by Drs. Blagosklonny and Pardee.
Cytokine Regulation of Apoptosis and Cell Death
The final regulatory aspect of IL-3 to be discussed in this chapter is the role of IL-3 in the
modulation of apoptosis. Many cytokines such as IL-3 act to prevent apoptosis. Research with
IL-3-dependent cell lines has proven very rewarding in terms of our understanding of the
mechanisms of apoptosis since removal of IL-3 from the culture medium causes the cells to
undergo apoptosis 24 to 48 hrs later. Apoptosis can be suppressed in these cells for approximately
3 to 6 days by the introduction of constitutively expressed anti-apoptotic genes such as Bcl-2.
Apoptotic Mediators: The Caspases
The caspase family of proteases comprise the effector arm of the apoptotic pathway.310-312 At
least 15 caspases, have been identified thus far.313,314 These cysteine proteases cleave substrates
carboxy-terminal to an aspartate residue, the P1 site.315 However, caspases can be divided into
three subgroups based on differences in substrate preference dictated by the residues immediately
amino-terminal to the P1 site.316
Caspases are synthesized as inactive proenzymes consisting of an amino-terminal prodomain
and large and small subunits. Caspase prodomains contain protein-protein interaction modules,
which facilitate the association of multiple factors required for caspase activation in response to
apoptosis-inducing stimuli. Activation requires proteolytic removal of the prodomain and
formation of a heterodimer between the large and small subunits. Upon tetramer formation, the
caspase complex is activated.317,318
Some caspase prodomains contain a death effector domain (DED) through which they
bind to other DED domains contained on adapter proteins (e.g., FADD/MORT1). This
targets the caspase to ligand-activated death receptors at the cell membrane where caspase
activation occurs in the death-inducing signaling complex (DISC).319-324 Formation of DISCs
is triggered by binding of death ligands such as TNF-α, Fas, and TRAIL to their respective
death receptors. Other caspase prodomains contain a caspase recruitment domain (CARD)
through which they bind to the apoptotic protease activating factor-1 (Apaf-1) resulting in the
formation of a cytoplasmic caspase-activating complex, the apoptosome.325-328 In both cases,
multiple factors are required for caspase activation. Caspases are activated in a cascade in which
“upstream” caspases are activated by cleavage in the apoptosome or DISC. These caspases, in
Signal Transduction Pathways: Cytokine Model 35
turn, cleave and activate “downstream” caspases whose substrates are cellular components critical
to the life of the cell.
Death receptors contain cytoplasmic death domains (DD) which, when bound by their
respective death ligand, recruit adapter proteins containing DD and DED domains.329,330 The
DED domains of adapter and pro-caspase proteins interact leading to autoproteolytic activation
of the upstream caspases and priming of the caspase cascade. Apoptosome formation in some
cells can be triggered by the release of cytochrome c from mitochondria in response to apoptosisinducing
stimuli.325,331 Cytochrome c binds to Apaf-1, which can interact with an upstream
caspase, procaspase-9, and an apoptosis inhibitor, Bcl-XL.326 Binding of dATP to Apaf-1 in
some cells is required for cleavage of procaspase-9, while binding of Bcl-XL can inhibit this
cleavage.332 Once activated, caspase-9 then proteolytically activates downstream caspases, such
as caspase-3, which directly cleaves life-sustaining cellular proteins. As stated earlier,
phosphorylation of caspase 9 by Akt results in its inactivation.
A caspase-3-like protease is also responsible for activating the cytoplasmic endonuclease,
CAD (caspase-activated deoxyribonuclease) that generates the oligonucleosomal DNA fragments
indicative of apoptosis.333 CAD is rendered inactive by its association with two isoforms
of a chaperone-like protein, ICAD/DFF-45 (inhibitor of CAD/DNA fragmentation factor-
45), which inhibit CAD endonuclease activity by concealing the CAD nuclear localization
sequence.331,333 Induction of apoptosis results in the cleavage and inactivation of ICAD/DFF-
45, the activation and nuclear translocation of CAD, and fragmentation of DNA into
oligonucleosomes. Another endonuclease, cyclophilin C, may be involved in generating the
50- to 200-kbp-sized DNA fragments.334
Roles of Bcl-2 Family Members in Cytokine-Mediated Regulation
of Apoptosis
The Bcl-2 gene was identified at the chromosomal breakpoints of t(14;18)-bearing follicular
B cell lymphomas. Having been translocated to a location near the enhancer elements of the
immunoglobulin heavy chain locus, the Bcl-2 gene was transcriptionally enhanced and the 26-
kDa Bcl-2 protein was overexpressed, contributing to malignant transformation.335-337 Most
importantly for this chapter, Bcl-2 protects hematopoietic cell lines from apoptosis following
growth factor withdrawal.338 Bcl-2 is a member of a growing family of related proteins that
play pivotal roles in determining whether or not apoptosis will proceed to completion.
The Bcl-2-related proteins share some structural similarities but are divided into subgroups
based on their structural differences and pro- versus anti-apoptotic activities.339 Bcl-2 contains
four conserved motifs, the Bcl-2 homology (BH) domains BH1 to BH4, as well as a transmembrane
domain. The most closely related Bcl-2 proteins contain 2-4 of the BH domains (at least
BH1 and BH2) and are anti-apoptotic rather than pro-apoptotic. These include Bcl-2, Bcl-XL,
Bcl-W, Mcl-1, and A1. The BAX subfamily is comprised of three proteins, BAX, Bak, and Bok,
which are structurally similar to Bcl-2, but promote apoptosis rather than survival. The BH3
subfamily includes proteins that contain only the BH3 domain and also
promote apoptosis: Bik, Blk, Hrk, BimL, Bad, Bid, and BNIP3.340-343
The Bcl-2 proteins function in a concentration-dependent manner by forming homo- and
heterodimers with other family members via the BH domains.334 For example, when Bcl-2 is in
excess, Bcl-2/BAX heterodimers are formed and apoptosis is inhibited; when BAX predominates,
BAX homodimers are formed and the cells are susceptible to apoptosis. The formation of
dimers among family members appears to be somewhat discriminate since some members
form dimers with any of the other members while some exhibit more limited associations.343-353
Mitochondrial Regulated Apoptosis
Many of the Bcl-2 related proteins, including Bcl-2, Bcl-XL, and BAX are localized in the
outer mitochondrial membrane. Bcl-2 and related, pro-survival proteins can inhibit apoptosis
upstream of caspase activation, presumably by binding Apaf-1 during formation of the
36 Cell Cycle Checkpoints and Cancer
apoptosome, thus preventing caspase activation.326 However, these pro-survival proteins do
not inhibit apoptosis induced by death receptor activation.354-357 Apoptosis is associated with
distinct changes in the mitochondrion including a reduction of ψm and the release of
cytochrome c and apoptosis inducing factor (AIF).358 Bcl-2-related proteins are thought to be
involved in regulating these changes by forming pores (channels) in the membrane which allow
the passage of cytochrome c and AIF.358-360 Both pro-apoptotic BAX and pro-survival Bcl-2-
related proteins form pores in vitro.361-367 However, whereas Bcl-2 and Bcl-XL have been shown to
inhibit cytochrome c release, BAX stimulates the release of cytochrome c. 365,367 Furthermore, Bcl-
2 can inhibit the capacity of BAX to form pores.362-364 The functional significance of the
mitochondrial pores in the initiation or progression of apoptosis is still controversial. Bcl-2 and
Bcl-XL can bind cytochrome c directly, suggesting that Bcl-2/Bcl-XL inhibition of apoptosis is
accomplished by sequestering cytochrome c from apoptosomes and preventing caspase
activation.368, 369 Certain pro-apoptotic Bcl-2 family members become activated during the
apoptotic response by caspase cleavage.370,371 Bid is a pro-apoptotic family member, which is
cleaved from a 22 kDa inactive protein to a 15 kDa active protein by caspase 8. 368,371 Active
p15 Bid then translocates to the mitochondria, oligomerizes with the pro-apoptotic Bak molecule,
and results in the release of cytochrome c.370,371 The mechanism of Bcl-2-mediated
regulation of apoptosis is still unclear, but like the regulation of caspases, it is likely to be
complex with multiple checkpoints.
Interactions Between Cytokine Signaling Pathways and Apoptosis
Cytokines can prevent apoptosis via different mechanisms.124 The carboxy region of the
IL-3Rβc chain has been associated with the IL-3-mediated prevention of apoptosis. Stimulation of
appropriate target cells by IL-3 leads to phosphorylation of Bad and its sequesterization by
14-3-3 proteins.372-374 The unphosphorylated form of Bad normally forms heterodimers with
anti-apoptotic factors, such as Bcl-2 or Bcl-XL, to induce cell death.372,374 Phosphorylation of
Bad induced by IL-3 stimulation releases Bad from Bcl-2 and Bcl-XL. The kinase(s), which
directly phosphorylates Bad, is unknown; however, ERK, Akt, and PKA have been speculated
to be involved.105-107,372-374 Sequestering of Bad by chaperonin 14-3-3 proteins allows Bcl-2
and Bcl-XL to bind BAX, resulting in the prevention of apoptosis. However this is not the only
mechanism by which cytokines suppress apoptosis. Expression of DNp85, a dominant-negative
version of the regulatory PI3K subunit, in BAF/3 cells suppressed apoptosis, but resulted in
similar levels of Bad phosphorylation as observed in unstimulated cells even when these cells
were treated with IL-3.375 Thus, there are other kinases besides PI3K, which phosphorylate
Bad. Even so, these cells were still resistant to apoptosis indicating at least one additional mechanism
by which IL-3 prevents apoptosis. This secondary means may be a result of Bcl-2 or Bcl-
XL phosphorylation.
Phosphorylation of Bcl-2: Positive and Negative Effects
Chemotherapeutic drugs and cytokines induce phosphorylation of Bcl-2, which is associated
with pro-apoptotic and anti-apoptotic activities respectively.2,376-381 Some
evidence has shown that Bcl-2 phosphorylation by chemotherapeutic drugs, such as Taxol,
results in inactivation or cleavage by caspases generating a truncated Bcl-2 protein which has
thereby gained pro-apoptotic properties. Other evidence suggest phosphorylation of the S70
residue of Bcl-2 by the Ras/Raf/MAPK pathway is anti-apoptotic.380,381 This phosphorylation
is correlated with enhanced Bcl-2 anti-apoptotic activity and the subsequent association of
protein phosphatase 2A (PP2A) with Bcl-2. PP2A dephosphorylates S70 to return Bcl-2
anti-apoptotic activity to basal levels.380,381
Future Remarks
More recently studies of the ability of oncogenes to either abrogate the cytokine-dependency
of hematopoietic cells or to induce various signal transduction pathways have been performed
with activated cellular or viral oncogenes inserted into retroviral vectors containing
Signal Transduction Pathways: Cytokine Model 37
either the oncogene-estrogen receptor (ER) or the oncogene-androgen receptor
(AR).2,9,65,149-152,221-223 Studies performed in FDC-P1 and TF-1 cells with either ΔRaf:ER or
ΔMEK1:ER inducible constructs have demonstrated that the activated viral oncogene is necessary
for abrogation of cytokine-dependency since the cells reverted to cytokine-dependency
when the steroid hormone was removed.2,9,151,152 These conditional constructs enable
investigators to distinguish between biochemical effects, which are due to the activated oncogene,
and effects which result from the cytokine. Moreover, studies performed with FL5.12 cells
have indicated that both activation of Raf/MEK1 and PI3K/Akt pathways were required for
efficient abrogation of cytokine-dependency.9 Given that some activated oncogene constructs
contain the androgen receptor, it is possible to activate one oncogene in the absence of the
other oncogene by adding testosterone in the absence of exogenous estrogen. We have determined
that human TF-1 as well as murine FDC-P1 and FL5.12 cells all infected with ΔRaf:ER
oncogenes secrete GM-CSF at sufficient levels to promote autocrine growth. Moreover the
autocrine cytokine production and the activated oncogene expression were both required for
the proliferation, providing further evidence for the theory that oncogenes induce autocrine
cytokine gene expression which is required for transformation. With conditionally active
oncogenes it is possible to directly monitor the effects of certain oncogenes on signal transduction
cascades as well as other pathways, which may be induced by the oncogene and determine the
signal transduction pathways involved in inducing autocrine transformation and promoting
tumorigenicity. These oncogene:ER or oncogene:AR constructs will aid us in determining which
signal transduction, cell cycle regulatory and apoptotic pathways are induced by oncogenes.
Knowledge of such information may allow the rational design of drugs which will affect multiple
points in different signal transduction cascades.
Acknowledgments
We sincerely thank Ms. Catherine Spruill for the outstanding artwork. This work was
supported in part by grants (R01CA51025) from the NCI and the North Carolina Biotechnology
Center (9805-ARG-0006, 2000-ARG-0003) to JAM. RAF was supported in part by
grants from the American Cancer Society (IRG-97-149), American Heart Association (9930099N)
and the North Carolina Biotechnology Center (9705-ARG-0009). MKW was supported in part
by PHS Grant DK45718. We sincerely thank Drs. Julian Downward, Martin McMahon and
Richard Roth for providing the Raf, MEK, PI3K and Akt encoding retroviruses.
References
1. Blalock WL, Weinstein-Oppenheimer C, Chang F et al Signal transduction, cell cycle regulatory,
and anti-apoptotic pathways regulated by IL-3 in hematopoietic cells: Possible sites for intervention
with anti-neoplastic drugs. Leukemia 1999; 13:1109-1166.
2. McCubrey JA, May WS, Duronio V et al. Serine/threonine phosphorylation in cytokine signal
transduction. Leukemia 2000; 14:9-21.
3. Blalock WL, Moye PW, Chang F et al. Combined effects of aberrant MEK1 activity and BCL2
overexpression on relieving the cytokine-dependency of human and murine hematopoietic cells.
Leukemia 2000; 14:1080-1096.
4. Moye PW, Blalock WL, Hoyle PE, et al. Synergy between Raf and BCL2 in abrogating the
cytokine-dependency of hematopoietic cells. Leukemia 2000; 14:1060-1079.
5. Weinstein-Oppenheimer CR, Blalock WL, Steelman LS et al. Raf signal transduction cascade as a
target for chemotherapeutic intervention in growth factor-dependent tumors Pharmacology &
Theraeutics 2000; 88:229-279.
6. Weinstein-Oppenheimer C, Steelman LS, Algate PA et al. Effects of deregulated Raf activation on
integrin, cytokine-receptor expression and the induction of apoptosis in hematopoietic cells.
Leukemia 2000; 14:1921-1938.
7. Blalock WL, Pearce M, Steelman LS et al. A conditionally-active form of MEK1 results in autocrine
transformation of human and mouse hematopoeitic cells. Oncogene 2000; 19:526-536.
8. Franklin RA and McCubrey JA. Kinases: Positive and negative regulators of apoptosis. Leukemia
2000; 14:2019-2034.
9. McCubrey JA, Steelman LS, Blalock WL et al. Synergistic effects of PI3K/Akt on abrogation of
cytokine-dependency induced by oncogenic Raf. Adv Enzyme Regl. 2001; 41:289-323.
38 Cell Cycle Checkpoints and Cancer
10. Miyajima A, Mui AL-F, Ogorochi T et al. Receptors for granulocyte-macrophage colonystimulating
factor, interleukin-3, and interleukin-5. Blood 1993; 82:1960-1974.
11. Miyajima A, Kitamura T, Harada N et al. Cytokine receptors and signal transduction. Annu Rev
Immunol 1992; 10:295-331.
12. Morla AO, Schreurs J, Miyajima A et al. Hematopoietic growth factors activate the tyrosine phosphorylation
of distinct sets of proteins in interleukin-3-dependent murine cell lines. Mol Cell Biol
1988; 8:2214-2218.
13. Kanakura Y, Druker B, Cannistra SA et al. Signal transduction of the human granulocyte-macrophage
colony-stimulating factor and interleukin-3 receptors involves tyrosine phosphorylation of
a common set of cytoplasmic proteins. Blood 1990; 76:706-715.
14. Silvennoinen O, Witthuhn BA, Quelle FW et al. Structure of the murine Jak2 proteintyrosine
kinase and its role in interleukin-3 signal transduction. Proc Natl Acad Sci USA
1993; 90:8429-8433.
15. Carroll MP, Clark-Lewis I, Rapp UR et al. Interleukin-3 and granulocyte-macrophage colonystimulating
factor mediate rapid phosphorylation and activation of cytosolic c-Raf. J Biol Chem
1990; 265:19812-19817.
16. Terada K, Kaziro Y, Satoh T. Ras-dependent activation of c-Jun N-terminal kinase/stress-activated
protein kinase in response to interleukin-3 stimulation in hematopoietic BaF3 cells. J Biol Chem
1997; 272:4544-4548.
17. Kinoshita T, Shirouzu M, Kamiya A et al. Raf/MAPK and rapamycin-sensitive pathways mediate
the anti-apoptotic function of p21Ras in IL-3-dependent hematopoietic cells. Oncogene 1997;
15:619-627.
18. Duronio V, Welham MJ, Abraham S et al. p21Ras activation via hemopoietin receptors and c-kit
requires tyrosine kinase activity but not tyrosine phosphorylation of p21Ras GTPase-activating protein.
Proc Natl Acad Sci USA 1992; 89:1587-1591.
19. Rapp UR, Troppmair J, Carroll M et al. Role of Raf-1 protein kinase in IL-3 and GM-CSFmediated
signal transduction. CTMI 1990; 166:129-139.
20. Okuda K, Sanghera JS, Pekech SL et al. Granulocyte-macrophage colony-stimulating factor,
interleukin-3, and steel factor induce rapid tyrosine phosphorylation of p42 and p44 MAP kinase.
Blood 1992; 79:2880-2887.
21. McCubrey JA, Steelman LS, Risser RG et al. Structure and expression of the T cell receptor gamma
locus in pre-B and early hemopoietic cells. Eur J of Immunol 1989; 19:2303-2308.
22. McCubrey JA, Steelman LS, Sandlin G et al. Effects of phorbol esters on an interleukin-3-dependent
cell line. Blood 1990; 76:63-72.
23. McCubrey JA, Steelman LS, McKearn JP. Interleukin-3 and phorbol esters induce different patterns
of immediate-early gene expression in an interleukin-3-dependent cell line. Oncogene Res
1991; 6:1-12.
24. Ways DK, Qin W, Riddle RS et al. Differential effect of phorbol esters and IL-3 on protein kinase
C isoform content and kinase activity in the FDC-P1 cell line. Blood 1991; 78:2633-2641.
25. White MK, DeVente JE, Robbins PJ et al. Differential regulation of glucose transporter expression
in hematopoietic cells by oncogenic transformation and cytokine stimulation. Oncology Reports
1994; 1:17-26.
26. McCubrey JA, Algate PA, Mayo MW et al. Differential effects of tumor promoters and cytokines
on proto-oncogene expression in a hematopoietic cytokine-dependent cell line. Oncology Reports
1994; 1:285-300.
27. Mayo MW, Steelman LS, McCubrey JA. Phorbol esters support the proliferation of a hematopoietic
cell line by upregulating c-Jun expression. Oncogene 1994; 9:1999-2008.
28. White MK, McCubrey JA. Changes in glucose transport associated with malignant transformation.
Int J Oncol 1995; 7:701-712.
29. McCubrey JA, Steelman LS, Mayo MW et al. Growth promoting effects of insulin-like growth
factor-1 on hematopoietic cells: Overexpression of introduced IGF-1 receptor abrogates
IL-3-dependency of murine factor dependent cells by a ligand-dependent mechanism.
Blood 1991; 78:921-929.
30. Brown CB, Pihl CE, Murray EW. Oligomerization of the soluble granulocyte-macrophage
colony-stimulating factor receptor: Identification of the functional ligand-binding species.
Cytokine 1997; 9:219-225.
31. Lia F, Rajotte D, Clark SC et al. A dominant negative granulocyte-macrophage colony-stimulating
factor receptor alpha chain reveals the multimeric structure of the receptor complex. J
Biol Chem 1996; 271:28287-28293.
32. Ihle JN. Cytokine receptor signaling. Nature 1995; 377:591-594.
Signal Transduction Pathways: Cytokine Model 39
33. Wilks AF. Two putative protein-tyrosine kinases identified by application of the polymerase chain
reaction. Proc Natl Acad Sci USA 1989; 86:1603-1607.
34. Firmbach-Kraft I, Byers M, Shows T et al. tyk2, prototype of a novel class of non-receptor tyrosine
kinase genes. Oncogene 1990; 5:1329-1336.
35. Heim MH, Kerr IM, Stark GK et al. Contribution of Stat SH2 groups to specific interferon
signalling by the Jak-Stat pathway. Science 1995; 267:1347-1349.
36. Sato N, Sakamaki K, Terada N et al. Signal transduction by the high-affinity GM-CSF receptor:
Two distinct cytoplasmic regions of the common β subunit responsible for different signaling.
EMBO J 1993; 12:4181-4189.
37. Pelicci G, Lanfrancone L, Grignani F et al. A novel transforming protein (Shc) with an SH2
domain is implicated in mitogenic signal transduction. Cell 1992; 70:93-104.
38. Blaikie P, Immanuel D, Wu J et al. A region in Shc distinct from the SH2 domain can bind
tyrosine-phosphorylated growth factor receptors. J Biol Chem 1994; 269:32031-32034.
39. Kavanaugh WM, Williams LT. An alternative to SH2 domains for binding tyrosine-phosphorylated
proteins. Science 1994; 266:1862-1865.
40. Pratt JC, Weiss M, Sieff CA et al. Evidence for a physical association between the Shc-PTB domain
and the βc chain of the granulocyte-macrophage colony-stimulating factor receptor. J
Biol Chem 1996; 271:12137-12140
41. Lioubin MN, Algate PA, Tsai S et al. p150Ship, a signal transduction molecule with inositol
polyphosphate-5-phosphatase activity. Genes Dev 1996; 10:1084-1095.
42. Matsuguchi T, Inhorn RC, Carlesso N et al. Tyrosine phosphorylation of p95Vav in myeloid cells
is regulated by GM-CSF, IL-3 and steel factor and is constitutively increased by p210BCR/ABL.
EMBO J 1995; 14:257-265.
43. Gulbins E, Coggeshall KM, Baier G et al. Tyrosine kinase-stimulated guanine nucleotide exchange
activity of Vav in T-cell activation. Science 1993; 260:822-825.
44. Gulbins E, Coggeshall KM, Langlet C et al. Activation of Ras in vitro and in intact fibroblasts by
the Vav guanine nucleotide exchange protein. Mol Cell Biol 1994; 14:906-913.
45. Adams JM, Houston H, Allen J et al. The hematopoietically expressed vav proto-oncogene shares
homology with the dbl GDP-GTP exchange factor, the bcr gene and a yeast gene (CDC24) involved
in cytoskeletal organization. Oncogene 1992; 7:611-618.
46. Quelle FW, Sato N, Witthuhn BA et al. Jak2 associates with the βc chain of the receptor for
granulocyte-macrophage colony-stimulating factor, and its activation requires the membrane-proximal
region. Mol Cell Biol 1994; 14:4335-4341.
47. Jaster R, Zhu Y, Pless M et al. JAK2 is required for induction of the murine DUB-1 gene. Mol
Cell Biol 1997; 17: 3364-3372.
48. Yoshimura A. The CIS/JAB family: Novel negative regulators of JAK signaling pathways. Leukemia
1998; 12:1851-1857.
49. Tian SS, Tapley P, Sincich C et al. Multiple signaling pathways induced by granulocyte colonystimulating
factor involving activation of JAKs, STAT5 and/or STAT3 are required for regulation
of three distinct classes of immediate-early genes. Blood 1996; 88:4435-4444.
50. Rao P, Mufson RA. A membrane proximal domain of the human interleukin-3 receptor beta c
subunit that signals DNA synthesis in NIH 3T3 cells specifically binds a complex of Src and Janus
family kinases and phosphatidylinositol 3-kinase. J Biol Chem 1995; 270:6886-6893.
51. Jucker, M Feldman RA. Identification of a new adapter protein that may link the common β
subunit of the receptor for granulocyte/macrophage colony-stimulating factor, interleukin (IL)-3,
and IL-5 to phosphatidylinositol 3-kinase. J Biol Chem 1995; 270:27817-27822.
52. Azam M, Erdjument-Bromage H, Kreider BL et al. Interleukin-3 signals through multiple isoforms
of Stat5. EMBO J 1995; 14:1402-1411.
53. Yoshimura A, Ohkubo T, Kiguchi T et al. A novel cytokine-inducible gene CIS encodes an SH2-
containing protein that binds to tyrosine-phosphorylated interleukin 3 and erythropoietin receptors.
EMBO J 1995; 14:2816-2826.
54. Takahashi-Tezuka M, Hibi M, Fujitani Y et al. Tec tyrosine kinase links the cytokine receptors to
PI-3 kinase probably through JAK. Oncogene 1997; 14:2273-2278.
55. Matsuda T, Takahashi-Tezuka M, Fukada T et al. Association and activation of Btk and Tec
tyrosine kinases by gp130, a signal transducer of the interleukin-6 family of cytokines. Blood 1995;
85:627-633.
56. Tsubokawa M, Tohyama Y, Tohyama K et al. Interleukin-3 activates Syk in a human myeloblastic
leukemia cell line, AML 193. Euro J Biochem 1997; 249:792-796.
57. Stomski FC, Woodchock JM, Zacharakis B et al. Identification of a Cys motif in the common beta
chain of the interleukin-3, granulocyte-macrophage colony-stimulating factor, and interleukin-5
40 Cell Cycle Checkpoints and Cancer
receptor essential for disulfide-linked receptor hetreodimerization and activation of all three
receptors. J Biol Chem 1998; 273:1192-1199.
58. Stomski FC, Sun Q, Bagley CJ et al. Human interleukin-3 (IL-3) induces disulfide-linked IL-3
receptor alpha- and beta-chain heterodimerization, which is required for receptor activation but
not high-affinity binding. Mol Cell Biol 1996; 16:3035-3046.
59. Anderson SM, Jorgensen B. Activation of src-related tyrosine kinases by IL-3. J Immunol 1995;
155:1660-1670.
60. Watanabe S, Ishida S, Koike K et al. Characterization of cis-regulatory elements of the c-myc
promoter responding to human GM-CSF or mouse interleukin-3 in mouse pro B cell BA/F3 cells
expressing the human GM-CSF receptor. Mol Biol Cell 1995; 6:627-636.
61. Barry SC, Korpelainen E, Sun Q et al. Roles of the N and C terminal domains of the interleukin-
3 receptor alpha chain in receptor function. Blood 1997; 89:842-852.
62. Nicola NA, Smith A, Robb L et al. The structural basis of biological actions of the GM-CSF
receptor. The molecular basis of cellular defense mechanisms. Wiley, Chichester Ciba Foundation
Symposium 1997; 204:19-27.
63. Doyle SE, Gasson JC. Characterization of the role of the human granulocyte-macrophage colony-stimulating
factor receptor alpha subunit in the activation of JAK2 and STAT5. Blood 1998; 92:867-876.
64. Barinaga M. Barinaga M. From bench top to bedside. Science 1997; 278:1036-1039..
65. McCubrey JA, Steelman LS, Moye PW, et al. Effects of deregulated Raf and MEK1 expression on
the cytokine-dependency of hematopoietic cells. Adv Enzyme Regl 2000; 40:305-337.
66. David M, Petricoin E, Benjamin C et al. Requirement for MAP kinase (ERK2) activity in interferon
alpha- and interferon beta-stimulated gene expression through Stat proteins. Science 1995;
269:1721-1723.
67. Migone TS, Lin JX, Cereseto A et al. Constitutively activated Jak-Stat pathway in T cells transformed
with HTLV-1. Science 1995; 269:79-81.
68. Danial NN, Pernis A, Rothman PB. Jak-Stat signaling induced by the v-abl oncogene. Science
1995; 269:1875-1877.
69. Ilaria RL, Van Etten RA. p210 and p190 (BCR/ABL) induce the tyrosine phosphorylation and DNA
binding activity of multiple specific Stat family members. J Biol Chem 1996; 271:31704-31710.
70. Murakami Y, Nakano S, Niho Y et al. Constitutive activation of Jak-2 and Tyk-2 in a v-Srctransformed
human gallbladder adenocarcinoma cell line. J Cell Physiol. 1998; 175:220-228.
71. Campbell GS, Yu CL, Jove R et al. Constitutive activation of JAK1 in Src-transformed cells. J Biol
Chem 1997; 272:2591-2594.
72. O’Shea JJ, Notarangelo LD, Johnston JA et al. Advances in the understanding of cytokine signal
transduction: the roles of Jaks and STATs in immunoregulation and the pathogenesis of immunodeficiency.
J Clin Immunol 1997; 17:431-447.
73. Paul WE. Interleukin 4: signalling mechanism and control of T cell differentiation. Ciba Foundation
Symposium 1997; 204:208-216.
74. Chen M, Cheng A, Chen YQ et al. The amino terminus of JAK3 is necessary and sufficient for
binding to the common gamma chain and confers the ability to transmit interleukin 2-mediated
signals. Proc Natl Acad Sci USA 1997; 94:6910-6915.
75. Berridge MJ. Lymphocyte activation in health and disease. Crit Rev Immun 1997; 17:155-178.
76. Riedy MC, Dutra AS, Blake TB et al. Genomic sequence, organization, and chromosomal
localization of human JAK3. Genomics 1996; 37:57-61.
77. Izuhara K, Heike T, Otsuka T, Yamaoka K et al. Signal transduction pathway of interleukin-4 and
interleukin-13 in human B cells derived from X-linked severe combined immunodeficient patients.
J Biol Chem 1996; 271:619-622.
78. Leonard WJ. Dysfunctional cytokine receptor signaling in severe combined immunodeficiency. J
Invest Med 1996; 44:304-311.
79. Rodig SJ, Meraz Ma, White JM et al. Disruption of the Jak1 gene demonstrates obligatory and
nonredundant roles of the Jaks in cytokine-induced biologic responses. Cell 1998; 93:373-383.
80. Corey S, Eguinoa A, Puyana-Theall K et al. Granulocyte macrophage-colony stimulating factor
stimulates both association and activation of phosphoinositide 3OH-kinase and src-related tyrosine
kinase(s) in human myeloid derived cells. EMBO J 1993; 12:2681-2690.
81. Suzuki J, Kaziro Y, Koide H. An activated mutant of R-Ras inhibits cell death caused by cytokine
deprivation in BaF3 cells in the presence of IGF-1. Oncogene 1997; 15:1689-1697.
82. Ahmed N, Kansara M, Berridge MV. Acute regulation of glucose transport in a monocytemacrophage
cell line: Glut-3 affinity for glucose is enhanced during the respiratory burst. Biochem
J 1997; 327:369-375.
Signal Transduction Pathways: Cytokine Model 41
83. Craddock BL, Welham MJ. Interleukin-3 induces association of the protein-tyrosine phosphatase
SHP2 and phosphatidylinositol 3-kinase with a 100-kDa tyrosine-phosphorylated protein in
hematopoietic cells. J Biol Chem 1997; 272:29281-29289.
84. Anderson SM, Burton EA, Koch BL. Phosphorylation of Cbl following stimulation with interleukin-3 and
its association with Grb2, Fyn and phosphatidylinositol 3-kinase. J Biol Chem 1997; 272:739-745.
85. Minshall C, Arkins S, Freund GG et al. Requirement for phosphatidylinositol-3- kinase to protect
hemopoietic progenitors against apoptosis depends upon the extracellular survival factor. J Immunol
1996; 156:939-947.
86. Ahmed NN, Grimes HL, Bellacosa A et al. Transduction of interleukin-2 anti-apoptotic and proliferative
signals via Akt protein kinase. Proc Natl Acad Sci USA 1997; 94:3627-3632.
87. Srivastava RK, Srivastava AR, Korsmeyer SJ et al. Involvement of microtubules in the regulation of
Bcl2 phosphorylation and apoptosis through cyclic AMP-dependent protein kinase. Mol Cell Biol
1998; 18:3509-3517.
88. Scheid MP, Duronio V. Dissociation of cytokine-induced phosphorylation of Bad and activation
of PKB/akt: involvement of MEK upstream of Bad phosphorylation. Proc Natl Acad Sci USA
1998; 95:7439-7444.
89. del Peso L, Gonzalez-Garcia M, Page C et al. Interleukin-3-induced phosphorylation of BAD through
the protein kinase Akt. Science 1997; 278:687-689.
90. Vanhaesebroeck B, Leevers SJ, Panayotou G et al. Phosphoinositide 3-kinases: A conserved family
of signal transducers. Trends Biochem Sci 1997; 22:267-272.
91. Takahashi-Tezuka M, Hibi M, Fujitani Y et al. Tec tyrosine kinase links the cytokine receptors to
PI-3 kinase probably through JAK. Oncogene 1997; 14:2273-2278.
92. Matsuda T, Takahashi-Tezuka M, Fukada T et al. Association and activation of Btk and Tec
tyrosine kinases by gp130, a signal transducer of the interleukin-6 family of cytokines. Blood 1995;
85:627-633.
93. Tsubokawa M, Tohyama Y, Tohyama K et al. Interleukin-3 activates Syk in a human myeloblastic
leukemia cell line, AML 193. Euro J Biochem 1997; 249:792-796.
94. Anderson SM, Jorgensen B. Activation of src-related tyrosine kinases by IL-3. J Immunol 1995;
155:1660-1670.
95. Downward J. Ras signalling and apoptosis. Cur Opin Gen Dev 1988; 8:49-54.
96. Wick MJ, Dong LQ, Riojas RA et al. Mechanism of phosphorylation of protein kinase B/Akt by a
constitutively active 3-phosphoinositide-dependent protein kinase-1. J Biol Chem 2000;
275:40400-40406.
97. Nakae J, Park BC, Accili D. Insulin stimulates phosphorylation of the forkhead transcription factor
FKHR on serine 253 through a Wortmannin-sensitive pathway. J Biol Chem 1999;
274:15982-15985.
98. Brunet A, Bonni A, Zigmond MJ et al. Akt promotes cell survival by phosphorylating and inhibiting
a Forkhead transcription factor. Cell 1999; 96:857-868.
99. Cardone MH, Roy N, Stennicke HR, Salveson GS et al. Regulation of cell death protease caspase-
9 by phosphorylation. Science 1998; 282:1318-1321.
100. Kelley TW, Graham MM, Doseff AI et al. Macrophage colony-stimulating factor promotes cell
survival through Akt/protein kinase B. J Biol Chem 1999; 272:26393-26398.
101. Ozes ON, Mayo LD, Gustin JA et al. NF-kappaB activation by tumor necrosis factor requires the
Akt serine-threonine kinase. Nature 1999; 401:82-85.
102. Xie P, Browning DD, Hay N et al. Activation of NF-kappaB by bradykinin through a
Galpha(q-) and Gbeta gamma-dependent pathway involves phosphoinsitide 3-kinase and Akt.
J Biol Chem 2000; 275:24907-24914.
103. Romashkova JA, Makarov SS. NF-kappaB is a target of Akt in anti-apoptotic PDGF signaling.
Nature 1999; 401:86-90.
104. Yanagihara Y, Basaki Y, Ikizawa K et al. Possible role of nuclear factor-kappa B activity in germline
C epsilon transcription in a human Burkitt lymphoma B cell line. Cell Immunol 1997; 176:66-74.
105. Duronio V, Scheid MP, Ettinger S. Downstream signalling events regulated by phosphatidylinositol
3-kinase activity. Cell Signal 1998; 10:233-239.
106. Downward J. Mechanisms and consequences of activation of protein kinase B/Akt. Curr Opin Cell
Biol 1998; 10:262-267.
107. Songyang Z, Baltimore D, Cantley LC et al. Interleukin 3-dependent survival by the Akt protein
kinase. Proc Natl Acad Sci USA 1997; 94:11345-11350.
108. Madrid LV, Wang CY, Guttridge DC. Akt Suppresses apoptosis by stimulating the transactivation
potential of the Rel A/p65 subunit of NF-kappaB. Mol Cell Biol 2000; 20:1626-1638.
109. Pullen N, Thomas G. The modular phosphorylation and activation of P70S6K. FEBS Letters
1997; 410:78-82.
42 Cell Cycle Checkpoints and Cancer
110. Cuadrado A, Bruder JT, Heidaran MA et al. H-Ras and Raf-1 cooperate in transformation of
NIH3T3 fibroblasts. Oncogene 1993; 8:2443-2448.
111. Saez R, Chan AM-L, Miki T et al. Oncogeni activation of human R-Ras by point mutations
analogous to those of prototype H-Ras oncogenes. Oncogene 1994; 9:2977-2982.
112. Wittinghofer A, Nassar N. How Ras-related proteins talk to their effectors. Trends Biochem Sci
1996; 21:488-491.
113. Vojtek AB, Hollenberg SM, Cooper JA. Mammalian Ras interacts directly with the serine/threonine
kinase Raf. Cell 1993; 74:205-214.
114. Marshall M. Interactions between Ras and Raf: Key regulatory proteins in cellular transformation.
Mol Repro Dev 1995; 42:493-499.
115. Brtva TR, Drugan JK, Ghosh S et al. Two distinct Raf domains mediate interaction with Ras. J
Biol Chem 1995: 270:9809-9812.
116. Drugan JK, Khosravi-Far R, White MA et al. Ras interaction with two distinct binding domains
in Raf-1 may be required for Ras transformation. J Biol Chem 1996; 271:233-237.
117. Cox AD, Der CJ. Farnesyltransferase inhibitors and cancer treatment: targeting simply Ras? Biochem
et Biophysica Acta 1997; 1333:F51-71.
118. Hernandez-Alcoceba R, Saniger L, Campos J et al. Choline kinase inhibitors as a novel approach
for antiproliferative drug design. Oncogene 1997; 15:2289-2301.
119. Byk G, Lelievere Y, Duchesne M et al. Synthesis and conformational analysis of peptide inhibitors
of farnesyltransferase. Bioorganic & Medic Chem 1997; 5:115-124
120. Sepp-Lorenzino L, Ma Z, Rands E et al. A peptidomimetic inhibitor of farnesyl:Protein transferase
blocks the anchorage-dependent and -independent growth of human tumor cell lines. Cancer Res
1995; 55:5302-5309.
121. Heimbrook DC, Oliff A. Therapeutic intervention and signaling. Current Opin in Cell Biol 1998;
10:284-288.
122. Gibbs JB, Oliff A. The potential of farnesyltransferase inhibitors as cancer chemotherapeutics. Ann
Rev Pharmacol & Toxicol 1997; 37:143-166.
123. Khosravi-Far R, White MA, Westwick JK et al. Oncogenic Ras activation of Raf/mitogen-activated
protein kinase-independent pathways is sufficient to cause tumorigenic transformation. Mol Cell
Biol 1996; 16:3923-3933.
124. Kinoshita T, Yokota T, Arai K-I et al. Suppression of apoptotic death in hematopoitic cells by
signalling through the IL-3/GM-CSF receptors. EMBO J 1995; 14:266-275.
125. Muszynski KW, Ruscetti FW, Heidecker G et al. Raf-1 protein is required for growth factorinduced
proliferation of hematopoietic cells. J Exp Med 1995; 181:2189-2199.
126. Clark GJ, Drugan JK, Rossman KL et al. 14-3-3 ζ negatively regulates Raf-1 activity by interactions
with the Raf-1 cysteine-rich domain. J Biol Chem 1997; 272:20990-20993.
127. Hu CD, Kariya K, Tamada M et al. Cysteine-rich region of Raf-1 interacts with activator domain
of post-translationally modified Ha-Ras. J Biol Chem 1995; 270:30274-30277.
128. Fabian JR, Daar IO, Morrison DK. Critical tyrosine residues regulate the enzymatic and biological
activity of Raf-1 kinase. Mol Cell Biol 1993; 13:7170-7179.
129. Marais R, Light Y, Paterson H et al. Ras recruits Raf-1 to the plasma membrane for activation by
tyrosine phosphorylation. EMBO J 1995; 14:101-110.
130. Cai H, Smola U, Wixler V et al. Role of diacylglycerol-regulated protein kinase C isotypes in
growth factor activation of the Raf-1 protein kinase. Mol Cell Biol 1997; 17:732-741.
131. Xia M, Mukhopadhyay NK, Inhorn RC et al. The cytokine-activated tyrosine kinase JAK2 activates
Raf-1 in a p21Ras-dependent manner. Proc Natl Acad Sci USA 1996; 93:11681-11686.
132. Marshall CJ. Raf gets it together. Nature 1996; 383:127-1281.
133. Farrar MA, Alberola-Ila J, Perlmutter RM. Activation of the Raf-1 kinase cascade by coumermycin
induced dimerization. Nature 1996; 383:178-181.
134. Luo Z, Tzivion G, Belshaw PJ et al. Oligomerization activates c-Raf-1 through a Ras-dependent
mechanism. Nature 1996; 383:181-185.
135. Daum G, Eisenmann-Tappe I, Fries H-W et al. UR. The ins and outs of Raf kinases. Trends
Biochem Sci 1994; 19:474-479.
136. Yao B, Zhang Y, Delikat S et al. Phosphorylation of Raf by ceramide-activated protein kinase.
Nature 1995; 378:307-310.
137. Zhang Y, Yao B, Delikat S et al. Kinase suppressor of Ras is ceramide-activated protein kinase.
Cell 1997; 89:63-72.
138. Kolch W, Heidecker G, Kochs G et al. Protein kinase C activates Raf-1 by direct phosphorylation.
Nature 1993; 364:249-252.
Signal Transduction Pathways: Cytokine Model 43
139. Marais R, Light Y, Mason C et al. Requirement of Ras-GTP-Raf complexes for activation of Raf-
1 by protein kinase C. Science 1998; 280:109-112
140. Singh K, Balligand JL, Fischer TA et al. Regulation of cytokine-inducible nitric oxide synthase in
cardiac myocytes and microvascular endothelial cells. Role of extracellular signal-regulated kinases
1 and 2 (ERK1/ERK2) and STAT1 alpha. J Biol Chem 1996; 271:1111-1117.
141. Toigoe T, O’Connor R, Santoli D et al. Interleukin-3 regulates the activity of the LYN proteintyrosine
kinase in myeloid-committed leukemic cell lines. Blood 1992; 80:617-624.
142. Hanazono Y, Chiba S, Sasaki K et al. c-fps/fes protein-tyrosine kinase is implicated in a signaling
pathway triggered by granulocyte-macrophage colony-stimulating factor and interleukin-3. EMBO
J 1993; 12:1641-1646.
143. Yousefi S, Hoessli DC, Blaser K et al. Requirement of Lyn and Syk tyrosine kinases for the prevention
of apoptosis by cytokines in human eosinophils. J Exp Med 1996; 183:1407-1414.
144. Pazdrak K, Schreiber D, Forsythe P et al. The intracellular signal transduction mechanism of
interleukin 5 in eosinophils: the involvement of the lyn tyrosine kinase and the Ras-Raf-1-MEKmicrotubule-
associated protein kinase pathway. J Exp Med 1995; 181:1827-1834.
145. Chao TS, Abe M, Hershenson MB et al. Src tyrosine kinase mediates stimulation of Raf-1 and
mitogen-activated protein kinase by the tumor promoter thapsigargin. Mol Cell Biol 1997;
57:3168-3173.
146. Ueffing M, Lovric J, Philipp A et al. Protein kinase C-epsilon associates with the Raf-1 kinase and
induces the production of growth factors that stimulate Raf-1 activity. Oncogene 1997; 24:2921-
2927.
147. Ravi RK, Weber E, McMahon M et al. Activated Raf-1 causes cell cycle arrest in small cell lung
cancer cells. J Clin Invest 1998; 101:153-159.
148. McCubrey JA, Smith SR, Algate PA et al. Retroviral infection can abrogate the factor-dependency
of hematopoietic cells by autocrine and non-autocrine mechanisms depending on the presence of a
functional viral oncogene. Oncogene 1993; 8:2905-15.
149. Lloyd AC, Obermuller F, Staddon S et al. Cooperating oncogenes converge to regulate cyclin/
CDK complexes. Genes Devel 1997; 11:663-677.
150. Woods D, Parry D, Cherwinski H, Bosch E et al. Raf-induced proliferation or cell cycle arrest is
determined by the level of Raf activity with arrest mediated by p21Cip. Mol Cell Biol 1997;
17:5598-5611.
151. McCubrey JA, Steelman LS, Hoyle PA et al. Differential abilities of activated Raf oncoproteins to
abrogate cytokine-dependency, prevent apoptosis and induce autocrine growth factor synthesis in
human hematopoietic cells. Leukemia 1998; 12:1903-1929.
152. Hoyle PE, Moye PW, Steelman LS, et al. Differential abilities of the Raf family of protein kinases
to abrogate cytokine-dependency and prevent apoptosis in murine hematopoietic cells by a MEK1-
dependent mechanism. Leukemia 2000; 14:642-656.
153. Dubois T, Rommel C, Howell S et al. 14-3-3 is phosphorylated by casein kinase I on residue
233. Phosphorylation at this site in vivo regulates Raf/14-3-3 interaction. J Biol Chem
1997; 272:28882-28888.
154. Hsu SY, Kaipia A, Zhu L et al. Interference of BAD (Bcl-xL/Bcl-2 associated death promoter)-induced
apoptosis in mammalian cells by 14-3-3 isoforms and P11. Mol Endo 1997; 11:1858-1867.
155. Zhang L, Wang H, Liu D et al. Raf-1 kinase and exoenzyme S interact with 14-3-3zeta through a
common siteinvolving lysine 49. J Biol Chem 1997; 272:13717-13724.
156. Xing H, Kornfeld K, Muslin AJ. The protein kinase KSR interacts with 14-3-3 protein and Raf.
Current Biol 1997; 7:294-300.
157. Dubois T, Howell S, Amess B et al. Structure and sites of phosphorylation of 14-3-3 protein: role
in coordinating signal transduction pathways. J Prot Chem 1997; 16:513-522.
158. Yaffe MB, Rittinger K, Volina S et al. The structural basis for 14-3-3: Phosphopeptide binding
specificity. Cell 1997; 91:961-971.
159. Muslin AJ, Tanner JW, Allen PM et al. Interaction of 14-3-3 with signaling proteins is mediated
by the recognition of phosphoserine. Cell 1996; 84:889-897.
160. Rommel C, Radziwill G, Lovric J et al. Activated Ras displaces 14-3-3 protein from the amino
terminus of c-Raf-1. Oncogene 1996; 12:609-619.
161. Freed E, Symons M, Macdonald SG et al. Binding of 14-3-3 proteins to the protein kinase Raf
and effects on its activation. Science 1994; 265:1713-1716.
162. Fu H, Xia K, Pallas DC et al. Interaction of the protein kinase Raf-1 with 14-3-3 proteins. Science
1994; 266:126-129.
163. Thorson JA, Yu LWK, Hsu AL et al. 14-3-3 proteins are required for maintenance of Raf-1 phosphorylation
and kinase activity. Mol Cell Biol 1998; 18:5229-5238.
44 Cell Cycle Checkpoints and Cancer
164. Huang W, Kessler DS, Erikson RL. 1995. Biochemical and biological analysis of MEK1 phosphorylation
site mutants. Mol Biol Cell 1995; 15:237-245.
165. Brunet A, Pages G, Pouyssegur J. Constitutively active mutants of MAP kinase kinase (MEK1)
induce growth factor-relaxation and oncogenicity when expressed in fibroblasts. Oncogene 1994;
9:3379-3387.
166. Krebs EG, Sege , Seger D et al. Overexpression of mitogen-activated protein kinase kinase (MAPKK)
and its mutants in NIH 3T3 cells. Evidence that MAPKK involvement in cellular proliferation is
regulated by phosphorylation of serine residues in its kinase subdomains VII and III. J Biol Chem
1994; 269:25699-25709.
167. Alessandrini A, Greulich H, Huang W et al. MEK1 phosphorylation site mutants activate Raf-1 in
NIH 3T3 cells. J Biol Chem 1996; 271:31612-31618.
168. Bottorff D, Stang S, Agellon S et al. Ras signaling is abnormal in c-Raf1 MEK1 double mutant.
Mol Cell Biol 1995; 15:5113-5122.
169. Okazaki K, Sagata N. MAP kinase activation is essential for oncogenic transformation ofNIH3T3
cells by Mos. Oncogene 1995; 10:1149-1157.
170. Pham CD, Arlinghaus RB, Zheng CF et al. Characterization of MEK1 phosphorylation by the v-
Mos protein. Oncogene 1995; 10:1683-1688.
171. Kranitz LM, Burns LA, Sutor SL et al. Interleukin-2 triggers a novel phosphotidylinositol 3-kinase-
dependent MEK activation. Mol Cell Biol 1995; 15:3049-3057.
172. Berra E, Diaz-Meco MT, Lozano J et al. Evidence for a role of MEK and MAPK during signal
transduction by protein kinase C zeta. EMBO J 1995; 14:6157-6163.
173. Marquardt B, Frith D, Stabel S. Signalling from TPA to MAP kinase requires protein kinase C,
Raf and MEK: Reconstitution of the signalling pathway in vitro. Oncogene 1994; 9:3213-3218.
174. Reuter CW, Catling AD, Jelinek T et al. Biochemical analysis of MEK activation in NIH-3T3
fibroblasts. Identification of B-Raf and other activators. J Biol Chem 1995; 270:7644-7655.
175. Williams JG, Roberts TM. Signal transduction pathways involving the Raf proto-oncogene. Cancer
Metastasis Rev 1994; 13:105-116.
176. Jaiswal RK, Moodie SA, Wolfman A et al. The mitogen-activated protein kinase Cascade is activated
by B-Raf in response to nerve growth factor through interaction with p21Ras. Mol Cell Biol
1994; 14:6944-6953.
177. Mansour SJ, Matten WT, Hermann AS et al. Transformation of mammalian cells by constitutively
active MAP kinase kinase. Science 1994; 265:966-970.
178. Weber MJ, Catling AD, Schaeffer HJ et al. A proline-rich sequence unique to MEK1 and MEK2
is required for Raf binding and regulates MEK function. Mol Cell Biol 1995; 15:5214-5225.
179. Dudley DT, Pung L, Decker SJ et al. A synthetic inhibitor of the mitogen-activated protein kinase
Cascade. Proc Natl Acad Sci USA 1995; 92:7686-7689.
180. Alessi DR, Cuenda A, Cohen P et al. PD098059 is a specific inhibitor of the activation of mitogen-
activated protein kinase kinase in vitro and in vivo. J Biol Chem 1995; 270:27489-27494.
181. Schaeffer HJ, Catling AD, Eblen ST et al. MP1: A MEK binding partner that enhances enzymatic
activation of the MAP kinase cascade. Science 1998; 281:1668-1671.
182. Whitmarsh AJ, Cavanagh J, Tournier C et al. A mammalian scaffolding complex that selectively
mediates MAP kinase activation. Science 1998; 281:1671-1674.
183. Bruder JT, Heidecker G, Rapp UR. Serum-, TPA-, and Ras-induced expression from AP-1/Etsdriven
promoters requires Raf-1 kinase. Genes Develop 1992; 6:545-556.
184. Griffin JD, Ranbaldi A, Vellenga E et al. Secretion of interleukin-1 by acute myeloblastic leukemia
cells in vitro induces endothelial cells to secrete colony stimulating factors. Blood 1987; 70:218-1221.
185. Janknecht R, Ernst WH, Pingould V et al. Activation of TCF Elk-1 by MAP kinases. EMBO J
1993; 12:5097-5104.
186. Marais R, Wynne J, Treisman R. The SRF accessory protein Elk-1 contains a growth factor-regulated
transcriptional activation domain. Cell 1993; 73:381-393.
187. Miltenberger RJ, Farnham PJ, Smith DE et al. v-Raf activates transcription of growth-responsive
promoters via GC-rich sequences that bind the transcription factor Sp1. Cell Growth Different
1995; 549:549-556.
188. Xing J, Ginty DD, Greenberg ME. Coupling of the Ras-MAPK pathway to gene activation by
Rsk2, a growth factor regulated CREB kinase. Science 1996; 273:959-963.
189. Kanno T, Siebenlist U. Activation of nuclear factor-kB via T cell receptor requires a Raf kinase
and Ca2+ influx. J Immunol 1996; 157:5277-5283.
190. Fisher TL, Blenis J. Evidence for two catalytically active kinase domains in pp90rsk. Mol Cell Biol
1996; 16:1212-1219.
191. Nilsson M, Ford J, Bohm S et al. Characterization of a nuclear factor that binds juxtaposed with
ATF3/Jun on a composite response element specifically mediating induced transcription in response
to an epidermal growth factor/Ras/Raf signaling pathway. Cell Growth Diff. 1997; 8:913-920.
Signal Transduction Pathways: Cytokine Model 45
192. Xu S, Robbins D, Frost J et al. MEKK1 phosphorylates MEK1 and MEK2 but does not cause
activation of mitogen-activated protein kinase. Proc Natl Acad Sci USA 1995; 92:6808-6812.
193. Davis RJ, Derijard B, Raingeaud J et al. Independent human MAP kinase signal transduction
pathways defined by MEK and MKK isoforms. Science 1995; 267:682-685.
194. Davis RJ. Transcriptional regulation by MAP kinases. Mol Repro Devel 1995; 42:459-467.
195. Samuels ML, Weber MJ, Bishop JM et al. Conditional transformation of cells and rapid activation
of the mitogen-activated protein kinase Cascade by an estradiol-dependent human Raf-1 protein
kinase. Mol Cell Biol 1993; 13:6241-6252.
196. Neubauer A, Greenberg P, Negrin R et al. Mutations in the Ras protooncogenes in patients with
myelodysplastic syndromes. Leukemia 1994; 8:638-641.
197. Buschle M, Janssen JW, Drexler H et al. Evidence for a pluripotent stem cell origin of idiopathic
myelofibrosis: Clonal analysis of a case characterized by a N-Ras gene mutation. Leukemia 1988;
2:658-660.
198. Flotho C, Valcamonica S, Mach-Pascual S et al. Ras mutations and clonality analysis in children
with juvenile myelomonocytic leukemia (JMML). Leukemia 1999;13:32-37.
199. Kalra R, Paderanga DC, Olson K et al. Genetic analysis consistent with the hypothesis that NF1
limits myeloid cell growth through p21 Ras. Blood 1994; 84:3435-3439.
200. Cline MJ, Jat PS, Foti A. Molecular mechanisms in the evolution of chronic myelocytic l leukemia.
Leukemia Lymphoma 1992; 7:283-287.
201. Cutler RE Jr, Morrison DK. Mammalian Raf-1 is activated by mutations that restore Raf signaling
in Drosophila. EMBO J 1997; 16:1953-1960.
202. Storm SM, Rapp UR. Oncogene activation: c-Raf-1 gene mutations in experimental and naturally
occurring tumors. Toxicology Letters 1993; 67:201-210.
203. Miyamoto K, Matsuoka M, Tomita N et al. A B-cell line having chromosome 14 aberration at
break band q11 derived from an adult T-cell leukemia patient. Jpn J Cancer Res 1988; 79:12-16.
204. Ishikawa F, Takaku F, Nagao M et al. Rat c-Raf oncogene activation by a rearrangement that
produces a fused protein. Mol Cell Biol 1987; 7:1226-1232.
205. Miyamoto K, Tomita N, Ishii A et al. Specific abnormalities of chromosome 14 in patients with
acute type of adult T-cell leukemia/lymphoma. Int J Can 1987; 40:461-468.
206. Ingvarsson S, Asker C, Szpirer J et al. Rat c-Raf oncogene is located on chromosome 4 and may be
activated by sequences from chromosome 13. Somatic Cell Mol Gen. 1987; 14:401-405.
207. Chung GTY, Huang DP, Wai Lo K et al. Genetic lesion in the carcinogenesis of cervical cancer.
Anticaner Res 1992; 12:1485-1490.
208. Sithanandam G, Dean M, Brennscheidt U et al. Loss of heterozygosity at the c-Raf locus in small
cell lung carcinoma. Oncogene 1989; 4:451-455.
209. Rapp UR, Huleihel M, Pawson T et al. Role of Raf oncogenes during lung carcinogenesis. Lung
Cancer 1988; 4:162-167.
210. Reynolds SH, Anna CK, Brown KC et al. Activated protooncogenes in human lung tumors from
smokers. Proc Natl Acad Sci USA 1991; 88:1085-1089.
211. Hubuchi T, Kinoshita H, Yamada H et al. Oncogene amplification in urothelial cancers with p53
gene mutation or MDM2 amplification. JNCI 1994; 86:1331-1335.
212. Nickell-Brady C, Hahn FF, Finch GI et al. Analysis of K-Ras, p53 and c-Raf-1 mutations in
beryllium-induced rat lung tumors. Carcinogensis 1994; 15:257-262.
213. Cadrian U, You M, Goodrow T et al. Activation of protooncogenes in spontaneously occurring
non-liver tumors from C57BL/6 X C3H F1 mice. Cancer Res 1991; 51:1148-1153.
214. Jung M, Notarrio V, Dritschilo A. Mutations in the p53 gene in radiation-sensitive and resistant
human squamous carcinoma cells. Cancer Res 1992; 52:6390-6393.
215. Harris P, Morton CC, Guglielmi P et al. Mapping by chromosome sorting of several gene probes,
including c-myc, to the derivative chromosomes of a 3:8 translocation associated with familial
renal cancer. Cytometry 1986; 7:589-594.
216. Pfeifer AMA, Jones RT, Bowden PE et al. Human bronchial epithelial cells transformed by the c-
Raf-1 and c-myc protooncogenes induce multidifferentiated carcinomas in nude mice: A model for
lung carcinogenesis. Cancer Res 1991; 51:3793-3801.
217. Eggstein S, Manthey G, Hirsch T et al. Raf-1 kinase, epidermal growth factor receptors, and mutant
Ras proteins in colonic carcinomas. Digestive Diseases Sci 1996; 41:1069-1075.
218. Callans LS, Naama H, Khandelwal M et al. Raf-1 protein expression in human breast cancer cells.
Annals Sur Onco 1995; 2:38-42.
219. Okuda K, Matulonis U, Salgia R et al. Factor-independence of human myeloid leukemia cell lines
is associated with increased phosphorylation of the protooncogene Raf-1. Exp Hematol 1994;
22:1111-1117.
46 Cell Cycle Checkpoints and Cancer
220. Schmidt CA, Oettle H, Ludwig WD et al. Overexpression of the Raf-1 proto-oncogene in human
myeloid leukemia. Leukemia Res 1994; 18:409-413.
221. Pritchard CA, Samuels ML, Bosch E et al. Conditionally oncogenic forms of the A-Raf and B-Raf
protein kinases display different biological and biochemical properties in NIH-3T3 cells. Mol Cell
Biol 1995; 15:9430-9442.
222. Pritchard C, McMahon M. Raf revealed in life-or-death decisions. Nature Genetics 1997;
16:214-215.
223. Bosch E, Cherwinski H, Peterson D et al. Mutations of critical amino acids affect the biological
and biochemical properties of oncogenic A-Raf and Raf-1. Oncogene 1997; 11:1021-1034.
224. Lee JE, Beck TW, Wojnowski L et al. Regulation of A-Raf expression. Oncogene 1996;
12:1669-1677.
225. Stephens RM, Sithanandam G, Copelan TD et al. 95-kilodalton B-Raf serine/threonine kinase: Identification
of the protein and its major autophosphorylation site. Mol Cell Biol 1992; 12:3733-3742.
226. Ikawa S, Fukui M, Ueyama Y et al. B-Raf, a new member of the Raf family, is activated by DNA
rearrangement. Mol Cell Biol 1988; 8:2651-2654.
227. Fukui M, Yamamoto T, Kawai S et al. Molecular cloning and characterization of an activated
human c-Raf-1 gene. Mol Cell Biol 1987; 7:1776-1781.
228. Williams JG, Roberts TM. Signal transduction pathways involving the Raf proto-oncogene. Cancer
Metastasis Rev 1994; 13:105-116.
229. Thiagalingam A, De Bustros A, Borges M et al. RREB-1, a novel zinc finger protein, is involved in
the differentiation response to Ras in human medullary thyroid carcinomas. Mol Cell Biol 1996;
16:5335-5345.
230. Benn J, Su F, Doria M et al. Hepatitis B virus HBx protein induces transcription factor AP-1 by
activation of extracellular signal-regulated and c-Jun N-terminal mitogen-activated protein kinases.
J Virol 1996; 70:4978-4985.
231. Doria M, Klein N, Lucito R et al. The hepatitis B virus HBx protein is a dual specificity cytoplasmic
activator of Ras and nuclear activator of transcription factors. EMBO J 1995; 14:4747-4757.
232. Ito T, Sasaki Y, Wands JR. Overexpression of human insulin receptor substrate 1 induces cellular
transformation with activation of mitogen-activated protein kinases. Mol Cell Biol 1996;
16:943-941.
233. Oka H, Chatan Y, Hoshino R et al. Constitutive activation of mitogen-activated protein (MAP)
kinases in human renal cell carcinoma. Cancer Res 1995; 55:4182-4187.
234. Silvaraman VS, Wang H, Nuovo GJ et al. Hyperexpression of mitogen-activated protein kinase in
human breast cancer. J Clin Invest 1997; 99:1478-1483.
235. Patel BK, Ray S, Whiteside TL et al. Correlation of constitutive activation of Raf-1 with morphological
transformation and abrogation of tyrosine phosphorylation of distinct sets of proteins in
human squamous carcinoma cells. Mol Carcino 1991; 8:1-6.
236. Riva C, Lavieille JP, Reyt E et al. Differential c-myc, c-jun, c-Raf, and p53 expression in squamous
cell carcinoma of the head and neck: Implication in drug and radioresistance. Euro J Cancer
1995; 31B:384-391.
237. Faris M, Ensoli B, Stahl N et al. Differential activation of the extracellular signal-regulated kinase,
Jun kinase and Janus kinase-Stat pathways by oncostatin M and basic fibroblast growth factor in
AIDS-derived Kaposi’s sarcoma cells. AIDS 1996; 10:369-378.
238. Silberman S, Janulis M, Schultz RM. Characterization of downstream Ras signals that induce alternative
protease-dependent invasive phenotypes. J Biol Chem 1997; 272:5927-5935.
239. Canman CE, Gilmer TM, Coutts SB et al. Growth factor modulation of p53-mediated growth
arrest versus apoptosis. Genes Devel 1995; 9:600-611.
240. Pirollo KF, Hao Z, Rait A et al. Evidence supporting a signal transduction pathway leading to the
radiation-resistant phenotype in human tumor cells. Biochem Biophys Res Commun 1997;
230:196-201.
241. Simon C, Juarez J, Nicolson GL et al. Effect of PD 098059, a specific inhibitor of mitogenactivated
protein kinase kinase, on urokinase expression and in vitro invasion. Cancer Res 1996;
56:5369-5374.
242. Loda M, Capodieci P, Mishra R et al. Expression of mitogen-activated protein kinase phosphatase-
1 in the early phases of human epithelial carcinogenesis. Amer J Pathol 1996; 149:1553-1564.
243. Magi-Galluzzi C, Mishra R, Fiorentino M et al. Mitogen-activated protein kinase phosphatase 1 is
overexpressed in prostate cancers and is inversely related to apoptosis. Lab Invest 1997; 76:37-51.
244. Bennett AM, Tonks NK. Regulation of distinct stages of skeletal muscle differentiation by mitogen-
activated protein kinases. Science 1997; 278:1288-1291.
Signal Transduction Pathways: Cytokine Model 47
245. Krautwald S, Buscher D, Dent P et al. Suppression of growth factor-mediated MAP kinase activation
by v-Raf in macrophages: A putative role for the MKP-1 phosphatase. Oncogene 1995;
10:1187-1192.
246. Wittinghofer A, Nassar N. How Ras-related proteins talk to their effectors. Trends Biochem Sci
1996; 21:488-491.
247. Yan M, Dai T, Deak JC et al. Activation of stress-activated protein kinase by MEKK1 phosphorylation
of its activator SEK1. Nature 1994; 372:798-800.
248. Lin A, Minden A, Martinetto H et al. Identification of a dual specificity kinase that activates the
Jun kinases and p38-Mpk2. Science 1995; 268:286-290.
249. Karin M, Minden A, Lin A et al. Differential activation of ERK and JNK mitogen-activated protein
kinases by Raf-1 and MEKK. Science 1994; 266:1719-1723.
250. Johnson NL, Gardner AM, Diener KM et al. Signal transduction pathways regulated by mitogen-
activated/extracellular response kinase kinase kinase induce cell death. J Biol Chem 1996;
271:3229-3237.
251. Zanke BW, Rubie EA, Winnett E et al. Mammalian mitogen-activated protein kinase pathways are
regulated through formation of specific kinase-activator complexes. J Biol Chem 1996;
271:29876-29881.
252. Han J, Lee JD, Bibbs L et al. A MAP kinase targeted by endotoxin and hyperosmolarity in mammalian
cells. Science 1994; 265:808-811.
253. Meier R, Rous J, Cuenda A et al. Cellular stresses and cytokines activate multiple mitogenactivated-
protein kinase kinase homologues in PC12 and KB cells Euro J Biochem 1996;
236:796-805.
254. Hibi M, Lin A, Smeal T et al. Identification of an oncoprotein- and UV-responsive protein kinase
that binds and potentiates the c-Jun activation domain. Genes & Devel 1993; 7:2135-2148.
255. Lange-Carter CA, Johnson GL. Ras-dependent growth factor regulation of MEK kinase in PC12
cells. Science 1994; 265:1458-1461.
256. Wang XZ, Ron D. Stress-induced phosphorylation and activation of CHOP (GADD153) by p38
MAP kinase. Science 1996; 272:1347-1349.
257. McIlroy J, Chen D, Wjasow C et al. Specific activation of p85-p110 phosphatidylinositol 3´-
kinase stimulates DNA synthesis by Ras- and p70 S6 kinase-dependent pathways. Mol Cell Biol
1997; 17:248-255.
258. Rodriguez-Viciana P, Marte BM, Warne PH et al. Phosphatidylinositol 3´kinase: One of the effectors
of Ras.Phil Trans Roy Soc London Series B: Biological Sciences 1996; 351:225-231.
259. Edelmann HM, Kuhne C, Petritsch C et al. Cell cycle regulation of p70 S6 kinase and p42/p44
mitogen-activated protein kinases in Swiss mouse 3T3 fibroblasts. J Biol Chem 1996; 271:963-971.
260. Blenis J, Grammer T. Evidence for MEK-independent pathways regulating the prolonged activation
of the ERK-MAP kinases. Oncogene 1997; 14:1635-1642.
261. Lenormand P, McMahon M, Pouyssegur J. Oncogenic Raf-1 activates the p70 S6 kinase via a
mitogen-activated protein kinase-independent pathway. J Biol Chem 1996; 271:15762-15768.
262. Gelfand EW, Ishizuka T, Oshiba A et al. Aggregation of the FceRI on mast cells stimulates c-Jun
amino-terminal kinase activity. J Biol Chem 1996; 271:12762-12766.
263. Calvo V, Wood M, Gjertson C et al. Activation of 70-kDa S6 kinase, induced by the cytokines
interleukin-3 and erythropoietin and inhibited by rapamycin, is not an absolute requirement for
cell proliferation. Euro J Immunol 1994; 24:2664-2671.
264. Price DJ, Grove JR, Calvo V et al. Rapamycin-induced inhibition of the 70-kilodalton S6 protein
kinase. Science 1992; 257:973-977.
265. Sugiyama H, Papst P, Gelfand EW et al. p70 S6 kinase sensitivity to rapamycin is eliminated by
amino acid substitution of Thr229. J Immunol 1996; 157:656-660.
266. Thomas G, Hall MN. Tor signalling and control of cell growth. Cur Opin Cell Bio 1997;
9:782-787.
267. Terada N, Franklin RA, Lucas JJ et al. Failure of rapamycin to block proliferation once
resting cells have entered the cell cycle despite inactivation of p70 S6 kinase. J Biol Chem
1993; 268:12062-12068.
268. Liu L. Damen JE, Ware MD, Krystal G. Interleukin-3 induces the association of the inositol 5-
phosphatase SHIP with SHP2. J Biol Chem 1997; 272:10998-11001.
269. Damen JE, Liu L, Wakao H et al. The role of erythropoietin receptor tyrosine phosphorylation in
erythropoietin-induced proliferation. Leukemia 1995; 11 Suppl 3:423-425.
270. Giuriato S, Payrastre B, Drayer AL et al. Tyrosine phosphorylation and relocation of SHIP are
integrin-mediated in thrombin-stimulated human blood platelets. J Biol Chem 1997;
272:26857-26863.
48 Cell Cycle Checkpoints and Cancer
271. Lamkin TD, Walk SF, Liu L et al. SHC interaction with Src homology 2 domain containing inositol
phosphatase (SHIP) in vivo requires the SHC-phosphotyrosine binding domain and two specific
phosphotyrosines on SHIP. J Biol Chem 1997; 272:10396-10401.
272. Pei D, Wang J, Walsh CT. Differential functions of the two Src homology 2 domains in protein
tyrosine phosphatase SH-PTP1. Proc Natl Acad Sci USA 1996; 93:1141-1145.
273. D’Ambrosio D, Fong DC, Cambier JC. The SHIP phosphatase becomes associated with Fc
gammaRIIB1 and is tyrosine phosphorylated during ‘negative’ signaling. Immunol Lett 1996;
54:77-82.
274. Pei D, Lorenz U, Klingmuller U et al. Intramolecular regulation of protein phosphatase SH-PTP1:
A new function for Src homology 2 domains. Biochemistry 1994; 33:15483-15493.
275. Helgason CD, Damen JE, Rosten P et al. Targeted disruption of SHIP leads to hemopoietic perturbations,
lung pathology, and a shortened life span. Genes Develop 1998; 12:1610-1620.
276. Yi T, Cleveland JL, Ihle JN. Identification of novel protein tyrosine phosphatases of hematopoietic
cells by polymerase chain reaction amplification. Blood 1991; 78:2222-2228.
277. Tsui HW, Siminovitch KA, de Souza L et al. Motheaten and viable motheaten mice have mutations
in the haematopoietic cell phosphatase gene. Nature Gen 1993; 4:124-129.
278. Lydon NB, Mett H, Mueller M et al. A potent protein-tyrosine kinase inhibitor which selectively
blocks proliferation of epidermal growth factor receptor-expressing tumor cells in vitro and in vivo.
Int J Can 1998; 76:54-163.
279. Damen JE, Liu L, Rosten P et al. The 145-kDa protein induced to associate with Shc by multiple
cytokines is an inositol tetraphosphate and phosphatidylinositol 3,4,5-trisphosphate 5-phosphatase.
Proc Natl Acad Sci USA 1996; 93:1689-1693.
280. Ioubin MN, Algate PA, Tsai S et al. p150Ship, a signal transduction molecule with inositol
polyphosphate-5-phosphatase activity. Genes Dev 1996; 10:1084-1095.
281. Liu L. Damen JE, Ware MD et al. Interleukin-3 induces the association of the inositol 5-phosphatase
SHIP with SHP2. J Biol Chem 1997; 272:10998-11001.
282. Giuriato S, Payrastre B, Drayer AL et al. Tyrosine phosphorylation and relocation of SHIP are
integrin-mediated in thrombin-stimulated human blood platelets. J Biol Chem 1997;
272:26857-26863.
283. Lamkin TD, Walk SF, Liu L et al. Shc interaction with Src homology 2 domain containing inositol
phosphatase (SHIP) in vivo requires the Shc-phosphotyrosine binding domain and two specific
phosphotyrosines on SHIP. J Biol Chem 1997; 272:10396-10401.
284. Yi T, Mui AL, Krystal G et al. Hematopoietic cell phosphatase associates with the interleukin-3
(IL-3) receptor beta chain and down-regulates IL-3-induced tyrosine phosphorylation and mitogenesis.
Mol Cell Biol 1993; 13:7577-7586.
285. Matsumoto A, Masuhara M, Mitsui K et al. CIS, a cytokine inducible SH2 protein, is a target of
the JAK-STAT5 pathway and modulates STAT5 activation. Blood 1997; 89:3148-3154.
286. Endo TA, Masuhara M, Yokouchi M et al. A new protein containing an SH2 domain that inhibits
JAK kinases. Nature 1997; 387:921-924.
287. Starr R, Wilson TA, Viney EM et al. A family of cytokine-inducible inhibitors of signalling. Nature
1997; 387:917-921.
288. Masuhara M, Sakamoto H, Suzuki R et al. Cloning and characterization of novel CIS family
genes. Biochem Biophys Res Commun 1997; 239:439-446.
289. Sakamoto H, Yasukawa K, Masuhara M et al. A Janus kinase inhibitor, JAB, is an interferon-γ-
inducible gene and confers resistance to interferons. Blood 1998; 92:1668-1676.
290. Wang D, Stravopodis D, Teglund S et al. Naturally occurring dominant negative variants of STAT5.
Mol Cell Biol 1996; 16:6141-6148.
291. Nielsen M, Kaltoft K, Nordahl M et al. Constitutive activation of a slowly migrating isoform of
Stat3 in mycosis fungoides: tyrphostin AG490 inhibits STAT 3 activation and growth of mycosis
fungoides tumor cell lines. Proc Natl Acad Sci USA 1997; 94:6764-6749.
292. Hara K, Yonezawa K, Sakaue H et al. Normal activation of p70 S6 kinase by insulin in cells
overexpressing dominant negative 85kDa subunit of phosphoinositide 3-kinase. Biochem Biophys
Res Commun 1995; 208:735-741.
293. Sigal NH, Dumont FJ. Cyclosporin A, FK-506, and rapamycin: pharmacologic probes of lymphocyte
signal transduction. Annu Rev Immunol 1992; 10:519-560.
294. Roush W. On the biotech pharm, a race to harvest new cancer cures. Science 1997; 278:1039-1040.
295. Kolch W, Heidecker G, Lloyd P et al. Raf-1 protein kinase is required for growth of induced
NIH-3T3 cells. Nature 1991; 349:426-428.
296. Quresh SA, Joseph CK, Hendrickson M et al. A dominant negative Raf-1 mutant prevents v-Src
induced transformation. Biochem Biophys Res Commun 1993; 192:969-975.
Signal Transduction Pathways: Cytokine Model 49
297. Beltman J, McCormick F, Cook SJ. The selective protein kinase C inhibitor, Ro-31-8220, inhibits
mitogen-activated protein kinase phosphatase-1 (MKP-1) expression, and activates Jun N-terminal
kinase. J Biol Chem 1996; 271:27018-27024.
298. Caponigro F, French RC, Kaye SB. Protein kinase C, a worthwhile target for anticancer drugs.
Anti-Cancer Drugs 1997; 8:26-33.
299. Das M, Stenmark KR, Ruff LJ et al. Selected isozymes of PKC contribute to augmented growth of
fetal and neonatal bovine PA adventitial fibroblasts. Amer J Physiol 1997; 273: L1276-1284.
300. Harris TE, Persaud SJ, Jones PM. Atypical isoforms of PKC and insulin secretion from pancreatic
beta-cells: Evidence using Go 6976 and Ro 31-8220 as PKC inhibitors. Biochem Biophys Res
Commun 1996; 227:672-676.
301. London L, McKearn JP. Activation and growth of colony-stimulating factor-dependent cell lines is
cell cycle stage dependent. J Exp Med 1987; 166:1419-1435.
302. Pardee AB. A restriction point for control of normal animal cell proliferation. Proc Natl Acad Sci.
USA 1974; 71:1286-1290.
303. Lavoie JN, L’Allemain G, Brunet A et al. Cyclin D1 expression is regulated by the p42/p44MAPK
and negatively by the p38/HOGMAPK pathway. J Biol Chem 1996; 271:20608-20616.
304. Liu J, Chao J, Jiang M et al. Ras transformation results in an elevated level of cyclin D1 and
acceleration of G1 progression in NIH 3T3 cells. Mol Cell Biol 1995; 15:3654-3663.
305. Albanese C, Johnson J, Watanabe G et al. Transforming p21Ras mutants and c-Ets-2 activate the
cyclin D1 promoter through distinguishable regions. J Biol Chem 1995; 270:23589-23597.
306. Akatas H, Cai H, Cooper GM. Ras links growth factor signaling to the cell cycle machinery via
regulation of cyclin D1 and the CDK inhibitor p27kip1. Mol Cell Biol 1997; 17:3850-3857.
307. Filmus J, Robles AI, Shi W et al. Induction of cyclin D1 overexpression by activated Ras. Oncogene
1994; 9:3627-3633.
308. Winston J, Dong F, Pledger WJ. Differential modulation of G1 cyclins and the CDK inhibitor
p27Kip1 by platelet-derived growth factor and plasma factors in density-arrested fibroblasts. J Biol
Chem 1986; 271:11253-11260.
309. Winston JT, Coats SR, Wang Y et al. Regulation of the cell cycle machinery by oncogenic Ras.
Oncogene 1996; 12:127-134.
310. Cryns V, Yuan J. Proteases to die for. Genes and Development 1998; 12:1551-1570.
311. Cryns V, Yuan J. The Cutting Edge: Caspases in Epoptosis and Eisease. In When Cells Die: a
Comprehensive Evaluation of Apoptosis and Programmed Cell Death (ed. Z. Zakeri, J. Tilly, and
R. Lockshin) pp. 177-210. 1998; John Wiley & Sons, New York, NY.
312. Allen RT, Cluck MW, Agrawal DK. Mechanisms controlling cellular suicide: Role of Bcl-2 and
caspases. Cell & Mol Life Sci 1998; 54:427-445.
313. Alnemri ES, Livingston DJ, Nicholson DW et al. Human ICE/CED-3 protease nomenclature.
Cell 1996; 87:171.
314. Wang S, Miuyra M, Jung Y-k, Zhu H et al. Murine caspase-11, and ICE-interacting protease, is
essential for the activation of ICE. Cell 1998; 92:501-509.
315. Black RA, Kronheim SR, Merriam JE et al. A pre-aspartate-specific protease from human leukocytes
that cleaves pro-interleukin-1β. J Biol Chem 1989; 264:5323-5326.
316. Nicholson DW, Hornberry N. Caspases: Killer proteases. Trends Biochem Sci 1997; 22:299-306.
317. Walker NPC, Taladian RV, Brady KD et al. Crystal structure of the cysteine protease interleukin-
1β-converting enzyme, a (p20/p10)2 homodimer. Cell 1994; 78:343-352.
318. Wilson KP, Black J-AF, Thomson JA et al. Structure and mechanism of interleukin-1β converting
enzyme. Nature 1994; 370:270-275.
319. Boldin MP, Goncharov TM, Goltsev YV et al. Involvement of MACH, a novel MORT1/FADDinteracting
protease, in Fas/Apo-1- and TNF receptor-induced cell death. Cell 1996; 85:803-815.
320. Fernandes-Alnemri T, Armstrong RC, Krebs J et al. In vitro activation of CPP32 and Mch3 by
Mch4, a novel apoptotic cysteine protease containing two FADD-like domains. Proc Natl Acad Sci
USA 1996; 93:7464-7469.
321. Muzio M, Chinnaiyan AM, Kischkel FC et al. FLICE, a novel FADD-homologous ICE/CED-3-
like protease, is recruited to the CD95 (Fas/Apo-1) death-inducing signaling complex. Cell 1996;
85:817-827.
322. Vincenz C, Dixit VM. Fas-associated death domain protein interleukin-1β-converting enzyme 2
(FLICE2), an ICE/Ced-3 homologue, is proximally involved in CD95- and p55-mediated death
signaling. J Biol Chem 1997; 272:6578-6583.
323. Boldin MP, VARFolomeev EE, Pancer Z et al. A novel protein that interacts with the death
domain of Fas/Apo-1 contains a sequence motif related to the death domain. J Biol Chem
1995; 270:7795-7798.
50 Cell Cycle Checkpoints and Cancer
324. Chinnaiyan AM, O’Rourke K, Lane BR et al. Interaction of CED-4 with CED-3 and CED-9: a
molecular framework for cell death. Science 1997; 275:1122-1126.
325. Zou H, Henzel W, Liu X et al. Apaf-1, a human protein homologous to C. Elegans CED-4, participates
in cytochrome c-dependent activation of caspase-3. Cell 1997; 90:405-413.
326. Pan G, O’Rourke K, Dixit VM. Capase-9, Bcl-xL, and Apaf-1 form a ternary complex. J Biol Chem
1998; 273:5841-5845.
327. Hofmann K, Bucher P, Tschopp J. The CARD domain: A new apoptotic signalling motif. Trends
Biochem Sci 1997; 22:155-156.
328. Irmler M, Hofman D, Tschopp J. Direct physical interaction between the Caenorhabditis elegans
“death proteins” CED-3 and CED-4. FEBBS Lett 1997; 406:189-190.
329. Nagata S. Apoptosis by death factor. Cell 1997; 88:355-365.
330. Nagata S. Apoptosis: Telling cells their time is up. Current Biol 1996; 6:1241-1243.
331. Liu X, Zou H, Slaughter C et al. DFF, a heterodimeric protein that functions downstream of caspase-
3 to trigger DNA fragmentation during apoptosis. Cell 1997; 89:175-184.
332. Li P, Nijhawan D, Budihardjo I et al. Cytochrome c and dATP-dependent formation of Apaf-1/
caspase-9 complex initiates an apoptotic protease cascade. Cell 1997; 91:479-489.
333. Enari M, Sakahira H, Yokoyama H et al. A caspase-activated DNase that degrades DNA during
apoptosis, and its inhibitor ICAD. Nature 1998; 391:43-50.
334. Montague JW, Hughes Jr FM, Cidlowski JA. Native recombinant cyclophilins A, B, and C degrade
DNA independently of peptidylpropyl cis-trans-isomerase activity. J Biol Chem 1997; 272:6677-6684.
335. Erikson J, Nishikura K, ar- Rushdi A et al. Translocation of an immunoglobulin kappa locus to a
region 3´ of an unrearranged c-myc oncogene enhances c-myc transcription. Proc Natl Acad Sci USA
1983; 80:7581-7515.
336. Bakhshi A, Jensen JP, Goldman P et al. Cloning the chromosomal breakpoint of t(14;18) human
lymphomas: Clustering around JH on chromosome 14 and near a transcriptional unit on 18. Cell
1985; 41:899-906.
337. Cleary ML, Sklar J. Nucleotide sequence of a t(14;18) chromosomal breakpoint in follicular lymphoma
and demonstration of a breakpoint-cluster region near a transcriptionally active locus on chromosome
18. Proc Natl Acad Sci USA 1985; 82:7439-7443.
338. Nunez G, London L, Hockenbery D et al. Deregulated Bcl-2 gene expression selectively prolongs
survival of growth factor-deprived hemopoietic cell lines. J Immunol 1990; 144:3602-3610.
339. Adams JM, Cory S. The Bcl-2 protein family: Arbiters of cell survival. Science 1998; 281:1322-1326.
340. Reed JC, Zha H, Aime-Sempe C et al. Structure-function analysis of Bcl-2 family protein, regulators
of programmed cell death. Adv Exp Med Biol 1996; 406:99-112.
341. Chittenden T, Flemington C, Houghton AB et al. A conserved domain in Bak, distinct from BH1
and BH2, mediates cell death and protein binding functions. EMBO J 1995; 14:5589-5596.
342. Kelekar A, Thompson CB. Bcl-2 family proteins: the role of the BH3 domain in apoptosis. Trends
Cell Biol 1998; 8:324-330.
343. Wang K, Yin X-M, Chao DT et al. BID: A novel BH3 domain-only death agonist. Genes Dev
1996; 10:2859-2869.
344. Oltvai ZN, Milliman CL, Korsmeyer SJ. Bcl-2 heterodimerizes in vivo with a conserved homolo,
BAX, that accelerates programmed cell death. Cell 1993; 74:609-69.
345. Inohara N, Ding L, Chen S et al. Harakiri, a novel regulator of cell death, encodes a protein that
activates apoptosis and interacts selectively with survival-promoting proteins Bcl-2 and Bcl-X(L).
EMBOJ 199; 16:1686-1694.
346. O’Connor L, Strasser A, O’Reilly LA et al. Bim: A novel member of the Bcl-2 family that promotes
apoptosis. EMBO J 1998; 17:384-395.
347. Chinnaiyan AM, O’Rourke K, Tewari M et al. FADD, a novel death domain-containing protein,
interacts with the death domain of Fas and initiates aoptosis. Cell 1995; 81:505-512.
348. Chaudhary D, O’Rourke K, Chinnaiyan AM et al. The death inhibitory molecules CED-9 and CED-
4L use a common mechanism to inhibit the CED-3 death protease. J Biol Chem 1998;
273:17708-17712.
349. Spector MS, Desnoyers S, Hoeppner DJ et al. Interaction between the C. elegans cell-death regulators
CED-9 and CED-4. Nature 1997; 385:653.
350. Wu D, Wallen HD, Nunez G. Interaction and regulation of subcellular localization of CED-4 by
CED-9. Science 1997; 275:1126-1129.
351. James C, Gschmeissner S, Fraser A et al. CED-4 induces chromatin condensation in
Schizosaccharomyces pombe and is inhibited by direct physical association with CED-9. Curr Biol
1997; 7:246-252.
352. Seshagiri S, Miller LK. Caenorhabditis elegans CED-4 stimulates CED-3 processing and CED-3-
induced apoptosis. Curr Biol 1997; 7:455-460.
Signal Transduction Pathways: Cytokine Model 51
353. Huang DC, Adams JM, Cory S. The conserved N-terminal BH4 domain of Bcl-2 homologues is
essential for inhibition of apoptosis and interaction with CED-4. EMBO J 1998; 17:1029-1039.
354. Hu Y, Benedict MA, Wu D et al. Bcl-XL interacts with Apaf-1 and inhibits Apaf-1-dependent caspase-
9 activation. Proc Natl Acad Sci USA 1998; 95:4386-4391.
355. Strasser A, Harris AW, Huang DC et al. Bcl-2 and Fas/APO-1 regulate distinct pathways to lymphocyte
apoptosis. EMBO J 1995; 14:6136-6147.
356. Newton K, Harris AW, Bath ML et al. A dominant interfering mutant of FADD/MORT1 enhances
deletion of autoreactive thymocytes and inhibits proliferation of mature T lymphocytes. EMBO J
1998; 17:706-718.
357. Smith KGC, Strasser A, Vaux DL. CrmA expression in T lymphocytes of transgenic mice inhibits
CD95 (Fas/APO-)-transduced apoptosis, but does not cause lymphadenopathy or autoimmune disease.
EMBO J 1996; 15:5167-5176.
358. Petit PX, Susin S-A, Zamzami N et al. Mitochondria and programmed cell death: Back to the future.
FEBS Lett 1996; 396:7-13.
359. Zamzami N, Susin SA, Marchetti P et al. Mitochondrial control of nuclear apoptosis. J Exp Med
1996; 183:1533-1544.
360. Zoratti M, Szabo I. The mitochondrial permeability transition. Biochim iophys Acta 1995;
1241:139-176.
361. Green DR Reed C. Mitochondria and apoptosis. Science 1998; 281:1309-1312.
362. Minn AJ, Velez P, Schendel SL et al. Bcl-XL forms an ion channel in synthetic lipid membranes.
Natur 1997; 385:353-357.
363. Antonsson B, Conti F, Ciavatta A et al. Inhibition of BAX channel-forming activity by Bcl-2. Science
1997; 277:370-372.
364. Schendel SL, Xie Z, Montal MO etal. Channel formation by antiapoptotic protein Bcl-2. Proc Natl
Acad Sci USA 1997; 94:5113-5118.
365. Schlesinger PH, Gross A, Yin XM et al. Comparison of the ion channel characteristics of proapoptotic
BAX and antiapoptotic BCL-2. Proc Natl Acad Sci USA 1997; 94:11357-11362.
366. Lam M, Bhat MB, Nunez G et al. Regulation of Bcl-xl channel activity by calcium. J Biol Chem
1998. 273:17307-17310.
367. Deveraux QL, Takahashi R, Salvesen GS, Reed JC. X-linked IAP is a direct inhibitor of cell death
proteases. Nature 1997; 388:300-304.
368. Rosse T, Olivier R, Monney L et al. Bcl-2 prolongs cell survival after BAX-induced release of cytochrome
c. Nature 1998; 393:496-499.
369. Kharbanda S, Pandey P, Schofield L et al. Role for Bcl-xL as an inhibitor of cytosolic cytochrome c
accumulation in DNA-damage iduced apoptosis. Proc Natl Acad Sci USA 1997; 94:6939-6942.
370. Wei MC, Lindsten T, Mootha VK et al. tBID, a membrane-targeted death ligand, oligomerizes
withBAK to release cytochrome c. Genes & Development 2000; 14:2060-2071.
371. Gross A, Yin XM, Wang K et al. Caspase cleaved BID targets mitochondria and is required for
cytochrome c release, which Bcl-XL prevents this release but not tumor necrosis factor-R1/Fas death.
J Biol Chem 1999; 274:1156-1163.
372. Thompson CB, Gajewski TF. Apptosis meets signal transduction: elimination of a Bad influence.
Cell 1996; 87:589-592.
373. Wang HG, Rapp UR, Reed JC. Bcl-2 targets the protein kinase Raf-1 to mitochondria. Cell 1996;
87:629-638.
374. Zha J, Harada H, Yang E et al. Serine phosphorylation of death agonist Bad in response to survival
factor results in binding to 14-3-3 not Bcl-xL. Cell 1996; 87:619628.
375. Craddock BL, Orchiston EA, Hinton HJ et al. Dissociation of apoptosis from proliferation, protein
kinase B activation, and Bad phosphorylation in interleukin-3-mediated phosphoinositide 3-kinase
signaling. J Biol Chem 1999; 274:10633-10640.
376. Blagosklonny MV. Cell death beyond apoptosis.Leukemia 2000; 14:1502-1508.
377. Blagosklonny MV, Chuman Y, Bergan RC, Fojo T. Mitogen-activated protein kinase pathway is
dispensable for microtubule active drug-induced Raf-1/Bcl-2 phosphrylation and apoptosis in leukemia
cells. Leukemia 1999; 13:1028-1036.
378. Blagosklonny MV. A node between proliferation apoptosis and growth arrest. BioEssays 1999;
21:704-709.
379. Haldar S, Chintapilli J, Croce CM. Taxol induces Bcl-2 phosphorylation and death of prostate cancer
cells. Cancer Res 1996; 56:1253-1255.
380. Ito T, Deng X, Carr B et al. Bcl-2 phosphorylation required for anti-apoptosis function. J Biol Chem
1997; 272:11671-11673.
381. Deng X, Ito T, Carr B et al. Reversible phosphorylation of Bcl2 following interleukin-3 or bryostatin
1 is mediated by direct interaction with protein phosphatase 2A. J Biol Chem 1998; 273:34157-34163.
CHAPTER 3
Cell Cycle Checkpoints and Cancer, edited by Mikhail V. Blagosklonny. ©2001 Eurekah.com.
The Restriction Point of the Cell Cycle
Mikhail V. Blagosklonny and Arthur B. Pardee
Introduction: Mitogen-Dependent and -Independent Phases
of the Cell Cycle
Production of two cells from one requires duplication of all molecules and organelles that
compose each cell. DNA does not duplicate throughout the cycle but only during
several hours in S-phase (Fig. 1). During S phase, the DNAs of the 100,000 genes located
on 23 pairs (in humans) of chromosomes are replicated, each in a timely fashion. This requires
the ordered assembly of many proteins at the origins of DNA replication to form a competent,
prereplicative chromosomal state. All eukaryotes use similar proteins to license replication origins.1
After formation, competent origins are activated by cyclin-dependent kinases. Once started,
DNA replication must be finished. Therefore, extracellular signals such as growth factors (GF)
must not and do not control S-phase progression. Although S-phase is independent of growth
factors, massive DNA damage or deprivation of nucleotides forces a cell to be arrested in S-phase,
but such arrest is usually accompanied by cell death.
After successful completion of DNA synthesis, cells enter G2 phase in preparation for
mitosis. Protein kinases that are activated in G2 phase prevent rereplication of the DNA.
Quiescent mammalian cells are usually diploid, so after DNA duplication a cell must divide.
Therefore, control by growth factors is also unnecessary in the G2 phase. This is a time to
check the internal signaling with events like DNA damage.
At the onset of mitosis, cyclin B/cdc2 phosphorylates laminin, a nuclear membrane
protein, thus dissolving the nuclear envelope. In metaphase of mitosis, condensing of chromosomes
associates with forming of mitotic spindles. The endoplasmic reticulum and the Golgi
complex break down into small vesicles. During mitosis, chromosomes are condensed, proteins
hyperphosphorylated and transcription and biosynthesis are reduced or absent. Progression
of mitosis is determined by monitoring of the microtubule function to ensure chromosomal
fidelity. External regulation by growth factors would be detrimental for a cell, and
consequently, mitosis is a growth factor-independent phase of the cell cycle. In anaphase of
mitosis the two sets of chromosomes are pulled apart. During telophase and cytokinesis the
preceding fragmentation process of intercellular membranes is reversed. The nuclear envelope
reappears around the chromosomes, endoplasmic reticulum is rebuilt and cytoplasmic
microtubules reassemble. The transcription and biosynthetic functions of the cell are normalized
and intracellular membrane traffic is resumed.2 As cells exit mitosis, the cell cycle is reset,
allowing the establishment of a new, competent replication state in G1 phase (Fig. 1). Therefore,
as the logic dictates, G1 phase is the only part of the cell cycle that can and must be growth
factor-dependent.
The Restriction Point
The major regulatory events leading to proliferation occur in the G1-phase of the cell
cycle. In vivo as well as in cell culture, most quiescent cells have a G1 DNA content. The
The Restriction Point of the Cell Cycle 53
growth of normal cells in culture is regulated by complex interactions between growth factors,
cell density, and attachment to substrate.3 Growth factors are necessary to initiate and maintain
the transition through G1 phase leading to S phase. In normal cells, withdrawal of growth
factors prevents the onset of S phase. The point at G1 at which commitment occurs and the
cell no longer requires growth factors to complete the cell cycle has been termed the restriction
(R) point.4 The R point has been temporally mapped at 2-3 hours prior to the onset of DNA
synthesis.5 Once beyond the R point, or point of no return, cells are committed to DNA
synthesis and they no longer require the extracellular growth factors during the remainder of the
cell cycle. Transition of the restriction point was proposed to be determined by accumulation of a
labile protein.4
In Search of Mediators of the Restriction Point
In 1974, comparison of growth of cancer and normal cells revealed that restriction point
is lost in cancer.4,6 Identification of the R-point provided the difference between normal and
cancer cell cycles. But what is the biochemical nature of the point of no return? R-protein is a
functionally short-lived (labile) regulatory protein, whose synthesis is sensitive to growth factors,
and needs to accumulate to a critical amount before a cell can pass the R point and proceed
towards DNA synthesis.5 When this question had been asked, oncogenes and signal transduction
pathways were not known. Many of the later discovered proteins were evaluated for
meeting the criteria of the R-protein.
Search for proteins that are differentially expressed in the late G1 phase versus resting cells
identified numerous proteins.7 In order to effectively identify and isolate the late G1 expressing
genes, Liang and Pardee developed a method known as differential display.8 The differential
display detects mRNA species that are different between sets of mammalian cells, permitting
their recovery and cloning of the corresponding cDNAs. Differential display allows one to
separate and clone individual mRNAs by PCR. The key element is to use a set of oligonucleotide
primers, one being anchored to the polyadenylate tail of a subset of mRNAs, the other
Fig. 1. Mitogen-dependent and -independent phases of the cell cycle.
54 Cell Cyle Checkpoints and Cancer
being short and arbitrary in sequence so that it anneals at different positions relative to the first
primer. The mRNA subpopulations defined by these primer pairs were amplified after reverse
transcription and resolved on a DNA sequencing gel. When multiple primer sets were used,
reproducible patterns of amplified complementary DNA fragments were obtained that showed
strong dependence on sequence specificity of either primer. Many late-G1 expressing genes were
isolated by differential display.9 However, R protein is also accumulated by the mechanism of
protein stabilization, but protein stabilization cannot be detected by the differential display method.
R-protein is a labile protein which is induced, stabilized and accumulated in response to
growth factors leading to growth factor independence. The discovery of G1-phase cyclins (D
and E) was an important breakthrough. Cyclin D1, a labile nuclear protein, accumulates following
growth factor stimulation.10-15 It is commonly overexpressed in human cancer. Cyclin
E was discovered in the course of a screen for human complementary DNAs that rescue a
deficiency of G1 cyclin function in budding yeast. The amounts of both the cyclin E protein
and an associated protein kinase activity (cyclin E-CDK2 complexes) fluctuated periodically
through the human cell cycle; both were maximal in late G1 and early S phases.16 In 1993,
Pardee and his colleagues suggested that the molecular basis of the R protein could be production
of cyclin E17,18 because cyclins satisfy all of the criteria for the R protein, which includes a
late G1 phase increase, a delay of appearance after inhibition of protein synthesis in
nontransformed cells, and a faster recovery in transformed cells.
Cyclins: From Mitogen Signaling to the Restriction Point
The progression from one phase to another is driven by enzymes named cyclin-dependent
kinases (CDK). Their activators, the cyclins, however are unstable and are “cycling” during the
cell cycle.19 Thus, cyclins control the activities of cyclin-dependent protein kinases (CDKs)
and play a key role in cell cycle regulation. As cells proceed through the cycle, four major
cyclins are produced sequentially (D, E, A, and B), and they activate CDKs (Fig. 2). B-type
cyclins associate with p34cdc2 to trigger mitosis. Progression through S phase requires cyclin
A, presumably in association with p33CDK2. Cyclins D and E drive a cell into S-phase. The
three D-type cyclins (cyclin D1, D2, D3) are very similar but they share very little homology
with cyclin E. During cell cycle progression, D cyclins start accumulating at mid-G1, whereas
cyclin E appears later, just prior to the G1/S transition.
Mitogen-dependent progression through G1 phase is mediated by induction of the cyclin
D family. Growth factors regulate cyclin D1 by four simultaneous mechanisms (Fig. 3):
1. transcriptional induction,
2. stabilization of the cyclin D protein,
3. its translocation to the nucleus, and
4. assembly with their catalytic partners, CDK-4 and CDK-6.14,15
The promoters of D-type cyclins respond to a variety of mitogen-activated signals such as
Ras, β-catenin-Tcf/lef pathways.13,20 The transcriptional induction of cyclin D1 by growth
factors is dependent on the Ras/Raf-1/Mek/ ERK pathway.21-23 However, cyclin D1 protein is
rapidly degraded and therefore it has a short half-life. It is important to emphasize that not
only cyclin D and E but other proteins that regulate or affect the restriction point can be
increased by protein stabilization. Normally, wt p53 is rapidly degraded by the proteasome, but
it is accumulated after DNA damage, almost exclusively by protein stabilization. Similarly the
CDK inhibitors p27 and p21, as well as cyclin E and the E2F-1 transcription factor are degraded
by the proteasome.24-28 Cyclin D1 turnover is governed by ubiquitination and proteasomal
degradation, which are stimulated by cyclin D1 phosphorylation on threonine-286.29 This
phosphorylation can be inhibited by signaling through a pathway that sequentially involves
Ras, phosphatidylinositol-3-OH kinase (PI3K), and protein kinase B (Akt). Thus turnover of
cyclin D1, like its assembly, is mitogen dependent.30 Nuclear localization and assembly of
newly synthesized cyclin D1 with CDK-4 is also GF-dependent.31
The Restriction Point of the Cell Cycle 55
After active cyclin D-dependent kinase is assembled in the nucleus, it phosphorylates the
retinoblastoma (Rb) protein,32 preventing its binding to E2F thus activating E2F-1 mediated
transcription.33 The E2F transcription factor activates genes whose products are involved in
nucleotide metabolism and DNA synthesis.34 For understanding the restriction point, it is
important to emphasize that E2F transactivates cyclins E and A.35 Cyclin E enters into a complex
with CDK2 and collaborates with cyclin D-CDKs to complete Rb phosphorylation. Cyclin
E-CDK has broader specificity than cyclin D-CDK. For example, cyclin E-CDK phosphorylates
the CDK inhibitor, p27, causing its degradation.36 The activity of cyclin E-CDK is inhibited
by p21, p27 and p57.15 In contrast, these CDK inhibitors, at least at low concentrations,
coactivate cyclin D-CDKs (Fig. 4). Another class of CDK inhibitors, p15, p16, and p18, specifically
inhibits cyclin D-CDKs.
The Restriction Point: A Knot of Mitogen and Inhibitory Signaling
Growth factors activate their receptors and other tyrosine kinases, Ras, and mitogen activated
pathways culminating in transcriptional induction of numerous genes, including
protooncogenes. Similarly, genes coding many growth factors, receptors, receptor-associated
proteins and kinases are all protooncogenes.37,38
One of the greatest breakthrough in the understanding of the regulation of cell cycle is the
connection of mitogenic stimulation to the cell cycle machinery. Expression of cyclins D, their
nuclear location, stability, and their assembly with CDK-4 and CDK-6 into active kinase complexes
are regulated by growth factors.15 Therefore, cyclins D are growth factor sensors. In
turn, the ability of cyclin D-dependent kinases to trigger phosphorylation of Rb in the mid- to
late G1 phase of the cell cycle makes inactivation of the growth suppressive function of Rb to be a
mitogen-dependent step. Rb participates in controlling the G1/S-phase transition, by binding
E2F transcription factor family members (Fig. 5). Absence of functional Rb is sufficient for S-phase
entry under growth-limiting conditions.39 E2F-1 accumulation bypasses a G1 arrest resulting
from the inhibition of G1 cyclin-dependent kinase activity.40-42 GF-dependency ends with the
phosphorylation of Rb, enabling cells to pass through the restriction point at the end of mid-G1
phase and to commit to completing the remaining phases of the growth cycle.43, 44 Given their
Fig. 2. Cyclin-dependent kinases (CDKs) and cyclins.
56 Cell Cyle Checkpoints and Cancer
ability to inactivate Rb, do cyclins D meet the criteria of R-protein? Not completely. Their
ability to act as growth factor sensors depends not only on their rapid induction by mitogens
but also on their protein instability, which ensures their precipitous degradation in cells deprived
of growth factors. The fact that D-type cyclins are labile proteins guarantees that the
subunit pool shrinks rapidly when cells are deprived of mitogens. Induction of cyclin D1 is not
sufficient for the transition from quiescence through G1 into S phase.20
Cyclin E in complex with CDK2 is downstream from cyclin D-CDK4. Cyclin E/CDK2
completes Rb phosphorylation (Fig. 5). This shift from cyclin D-CDK-4 to cyclin E-CDK2
accounts for the loss of dependency from growth factor. Precisely, the restriction point lies
between cyclins D and E (Fig. 5). As emphasized by Kohn, the mammalian G1/S cell cycle
phase transition comprise a highly nonlinear network that produces seemingly paradoxical
results and makes intuitive interpretations unreliable.45 Numerous feedback loops lead to a
situation that downstream events lie upstream of themselves. For example, pRb2/p130 and p27
both are involved in a negative feedback regulatory loop with cyclin E.46 Also, c-myc
expression is downstream from p21, CDKs, and E2F,47,48 yet c-myc is also an upstream
regulator of p21 and CDKs.49
Similarly, cyclin E is located downstream from Rb and E2F-1 because cyclin E is
transactivated by E2F-1 (Fig. 5). Yet, it is cyclin E that inactivates Rb and releases E2F (Fig. 5).
Positive loops ensure irreversibility of commitments. Once expressed, cyclin E become independent
of downstream GF-dependent cyclin D1. Therefore, cyclin E is a better candidate for
the R-protein than cyclin D. Dependence on GF ends with the phosphorylation of Rb,
enabling the cells to pass through the restriction point and to commit to completing the remaining
phases of the growth cycle.43 Yet, both cyclin D and E, as well as cyclin A, phosphorylate
Rb. The restriction "protein" splits into at least two slices (Fig. 6).
Furthermore, E2F-1 partially expresses functional features of the R-protein. In fact, E2F
is sufficient to drive a cell into S-phase.41 Under enlargement, the restriction “point” looks like
a restriction “knot” (Fig. 5 and 6).
Fig. 3. Cyclins D are growth factor sensors. Expression of cyclins D, their nuclear location, stability, and their
assembly with CDK4 and CDK6 into active kinase complexes are regulated by growth factors.
The Restriction Point of the Cell Cycle 57
Growth Arrest versus Proliferation
An isolated activation of signaling, which is downstream of the restriction point, induces
S phase entry. However, apoptosis may occur.50, 51 Apoptosis is suppressed by GF-dependent
events upstream of the restriction point. On the other hand, the choice between proliferation
and growth arrest in response to growth factors and inhibitors is determined by the state of the
restriction knot.
It is well known that exogenous stimuli such as EGF, TGFβ, CSF-1, interleukins, phorbol
esters and endogenous transduction molecules such as Ras, PKC, Raf-1, ERK can induce either
growth arrest or cycle progression (Fig. 7).52-61 Yet, one consideration is so simple that it
is often overlooked: growth stimulation can be detected only when an agent is applied to a resting
cell, whereas to detect growth inhibition the agent must be applied to a proliferating cell.51
There is increasing evidence that mitogenic signaling induces cyclin D and p21 simultaneously.
51 Not only their ratio but a proliferative status of a cell may determine the response.
In a resting cell, one can expect to find low basal levels of cyclins D1 and E but high levels of
the CDK inhibitor p27. In a proliferating cell, one would expect to find high levels of cyclin
D1 and E with low levels of p27. In resting fibroblasts, low levels of Raf-1 activity induces
cyclin D1 and proliferation, whereas, in cycling fibroblasts, high levels of Raf-1 activity leads to
growth arrest due to p21 induction. 57-59 Treatment with phorbol ester increased both cyclin
D1 and p21.60, 62, 63 While the growth of malignant melanoma cells was inhibited by TPA, the
growth of normal melanocytes was stimulated. 62 Similarly, stimuli that are mitogenic for mature
(resting) T-cells induce arrest in proliferating tumor T-cells, inducing both p21 and cyclin
D. 63, 60 E2F-1 and H-Ras can activate the p21 promoter, and induction of the p21 promoter
by activated Ras is mediated at least in part by E2F-1.64
The same stimuli can induce different outcomes depending on the Rb status. Cytokine
response gene induced p21 that blocked either G1/S or G2/M transition, but did not prevent
G1/S transition in Rb-negative (Rb-) cells, causing endoreduplication.65 The same stimuli
caused proliferation in quiescent normal human T cells.65 Levels of p21 are also important. Thus,
in HCT116 cells, low levels of endogenous p21 caused G2 but did not cause G1 arrest following
DNA damage.66,67 However, much higher levels of exogenous p21 arrested these cells in G1 phase.67
In primary cells, Ras is initially mitogenic but eventually induces premature senescence. Constitutive
activation of MEK induces both p53 and p16 and is required for Ras-induced senescence
of normal human fibroblasts.68 Furthermore, activated MEK permanently arrests primary murine
Fig. 4. Effects of “CDK inhibitors”.
58 Cell Cyle Checkpoints and Cancer
fibroblasts but forces proliferation in cells lacking either p53 or p16. This may explain the
opposite response of normal and immortalized cells to constitutive activation of MEK.68
It is very important that not only inhibitors and stimulators of CDKs are induced simultaneously,
but some of these molecules may act as both an inhibitor and a stimulator (Fig. 7).
While inhibiting cyclin-E-CDKs, p21 and p27 (at least at low levels) activate cyclin D1-CDKs
complexes.69 For example, p21 permits growth factor-induced cell cycle progression of vascular
smooth muscle.70 PKC alpha controls glioma cell cycle progression via upregulation of p21,
which facilitates active cyclin-CDK complex formation.71 In contrast, in many cell lines, PKC
induces growth arrest due to p21 induction,48 but the sensitivity to p21 may be lost in very
aggressive cancers.72 Increased expression of ectopic cyclin E in a mouse mammary epithelial
cell line inhibits rather than stimulates growth, and this may be due to increased expression of
the inhibitor p27.73
In addition to p21, p16(INK4a) and p15(INK4b), which are pure inhibitors of cyclin D
CDK-4 and -6, are induced through the MAPK pathway (Fig. 7). This pathway is involved in
p21 and p15 induction by TGF-β.74 Like contact inhibition, TGF-β can arrest the cell cycle in
G1 by inducing p27.75 Oncogenic Ras inhibits growth of primary cells due to induction of
p16 and p19(ARF).68,76 p15 is induced by oncogenic Ras to an extent similar to that of p16,
and expression of both is associated with G1 arrest and senescence.77 Ras-dependent induction
of these two INK4 genes is mediated mainly by the Raf-MEK-ERK pathway. 77 In addition,
high expression levels of both wild-type and oncogenic Ras inhibit growth of K562 leukemia
cells, which are deficient for p53, p16, p15, and p19 (ARF) genes. H-Ras increases p21 and
causes growth arrest in these cells.78
From Restriction- to “Check”-Points
The restriction point could be considered as a prototype of cell cycle checkpoints. The
restriction point is related to mitogen deprivation. Checkpoints are mostly related to DNA
damage79 and mitotic progression. Following DNA damage, a cell undergoes growth arrest to
repair DNA. Damaged DNA can propagate during S-phase and mitosis. Therefore, arrest
occurs in G1 (before S), and in G2 (before mitosis). Arrest in G1 prevents aberrant replication
of damaged DNA and arrest in G2 allows cells to avoid segregation of defective chromosomes.
Some mechanisms of G1 and G2 checkpoints are similar, including p53-dependent p21
induction and Rb dephosphorylation.66,80 Other mechanisms are distinct because different
sets of cyclins and kinases are activated in G1 and G2 phases. Massive DNA damage or deprivation
of nucleotides forces a cell to be arrested in S-phase.81,82 In the G2 checkpoint, p21 and
the 14-3-3 protein play distinct but complementary roles and cooperate to achieve arrest
following DNA damage.83 p21 overexpression may result in senescence-like growth arrest and
abnormal mitosis.84,85
Fig. 5. The restriction knot.
The Restriction Point of the Cell Cycle 59
In order to ensure faithful chromosome segregation, cells have evolved mechanisms that
delay progress into and out of mitosis until certain events are completed.86 A spindle checkpoint
in mitosis ensures equal chromosome segregation. The spindle checkpoint inhibits cell-cycle
progression in response to a signal generated by mitotic spindle damage or by chromosomes
that have not attached to microtubules. This prevents chromosome dysfunction and exit from
mitosis until all of the kinetochores (and thus chromosomes) are attached to the spindle. A
canonical example is the mitotic arrest by microtubule-active drugs such as Paclitaxel.87
The Restriction Point and G1 Checkpoint
As discussed, the restriction “point” lies between the accumulation of cyclins D and E and
the activation of cyclin E-CDK2. What is the topological relationship of this GF-dependent
restriction point and DNA-damage-induced G1 checkpoint?
DNA damage targets CDKs by two mechanisms. First is rapid and transient disappearance
of cyclin D1. DNA damage-induced pathways target cyclin D and then cyclin E-CDK2.
Following DNA damage, cyclin D1 is rapidly degraded by the proteasome.88 Cyclin D1 degradation
initiates a specific release of p21 from CDK4 complexes, leading to CDK2 inactivation.
p21 may be redistributed without an increase in total amount. This is a p53-independent
process. (Note: the restriction point control is also p53-independent). Second is a p53-dependent
induction of p21 resulting in inhibition of cyclin E-CDK2. The escape of cells with nonfunctional
p53 from the initial G1 arrest probably stems from the fact that the reservoir of p21 held
by cyclin D1/CDK4 complex is quickly exhausted after DNA damage.88 Simultaneously, p53
protein is accumulated. This results in p21 induction and sustained growth arrest. Importantly,
pRB is a critical component of this DNA damage checkpoint. The checkpoint pathway uses the
normal cell cycle regulatory machinery to induce the accumulation of the growth suppressive
form of pRb.89
Not only is the DNA damage checkpoint located nearby the restriction point, but many
regulators cause growth arrest by affecting the restriction point. Progesterone, a steroid
hormone, regulates proliferation and differentiation in the mammary gland and uterus and
induces p27 and p18 to inhibit cyclin E-CDK2 and CDK4.90 Since similar models have been
developed for growth inhibition by TGF-β and during adipogenesis, interaction between the
Cip/Kip and INK4 families of inhibitors may be a common theme in physiological growth
arrest.90 The protein kinase inhibitor staurosporine stops progression of normal nontransformed
cells in the G1 phase between cyclin D and cyclin E activities.91 In contrast, the histone
deacetylase inhibitor FR901228 causes nonphysiological G1 arrest with downregulation of
cyclin D1 and upregulation of cyclin E in cancer cells lacking cycle control.92
Fig. 6. “R-factor”.
60 Cell Cyle Checkpoints and Cancer
The Restriction Point and Therapy
Selective killing of malignant cells without killing of normal cells is the ultimate goal of
cancer therapy. Loss of the restriction point predisposes selective killing of cancer cells by
chemotherapy.93 Autonomous growth of cancer cells lacking the restriction point can be
exploited to selectively arrest growth of normal cells.94 For example, malignant transformation
is associated with loss of dependency on epidermal growth factor (EGF).95 MCF-10A,
EGF-dependent immortalized breast cells, underwent G0/G1 growth arrest following EGF
withdrawal. In contrast, MCF-7 cancer cells were not affected by EGF withdrawal. By using
inhibitors of the EGF receptor kinase, this dependence on EGF can be exploited for protection
of MCF-10A cells from microtubule-active drugs.96 Microtubule-active drugs, such as Paclitaxel
(Taxol) and vinblastine, kill cells in a cycle-dependent manner, in mitosis. Paclitaxel causes
microtubule dysfunction leading to mitotic arrest of cells that entered mitosis.97,98 Mitotic
arrest is accompanied by phosphorylation of serine in multiple regulatory proteins, including
Bcl-2, and this precedes cell death.87 A cell arrested in interphase by a nontoxic agent cannot
enter mitosis and therefore cannot undergo mitotic arrest when exposed to Paclitaxel. Hence,
arresting normal cells in interphase prior to treatment with Paclitaxel could increase the selectivity
of microtubule-active drugs against cancer cells by minimizing undesired toxicity to
normal cells. Low doses of AG1478, an inhibitor of the EGF receptor kinase, arrested the
MCF-10 breast cell line but not the MCF-7 cancer cell line.96 In fact, AG1478 abrogated Paclitaxel
effects in MCF-10A cells but not in MCF-7 cells.96 Also, by arresting growth, CDK inhibitors,
such as p21 and p16, can protect cells from apoptosis caused by cell-cycle dependent
chemotherapy.67, 99, 100 Targeting pathways that are not essential for cancer cells which lack the
restriction point can selectively protect normal cells against chemotherapy.
Fig. 7. Proliferation vs growth arrest.
The Restriction Point of the Cell Cycle 61
References
1. Pasero P, Schwob E. Think global, act local-how to regulate S phase from individual replication
origins. Opin Genet Dev 2000; 10:178-186.
2. Thyberg J, Moskalewski S. Role of microtubules in the organization of the Golgi complex. Exp
Cell Res. 1999; 246:263-279.
3. Baserga R. Oncogenes and the strategy of growth factors. Cell. 1994; 79:927-930.
4. Pardee AB. A restriction point for control of normal animal cell proliferation. Proc Natl Acad Sci
USA 1974; 71:1286-1290.
5. Campisi J, Medrano EE, Morro G, Pardee AB. Restriction point control of cell growth by a labile
protein: Evidence for increased stability in transformed cells. Proc Natl Acad Sci USA. 1982;
79:436-440.
6. Pardee AB. G1 events and regulation of cell proliferation. Science 1989; 246:603-608.
7. Croy RG, Pardee AB. Enhanced synthesis and stabilization of Mr 68,000 protein in transformed
BALB/c-3T3 cells: Candidate for restriction point control of cell growth. Proc Natl Acad Sci USA
1983; 80:4699-4703.
8. Liang P, Pardee AB. Differential display of eukaryotic messenger-RNA by means of the polymerase
chain-reaction. Science 1992; 257:967-971.
9. Alpan RS, Sparvero S, Pardee AB. Identification of mRNAs differentially expressed in quiescence
or in late G1 phase of the cell cycle in human breast cancer cells by using the differential display
method. Mol Med 1996; 2:469-4.
10. Matsushime H, Roussel MF, Ashmun RA, Sherr C. Colony-stimulating factor 1 regulates novel
cyclins During the G1 phase of the cell cycle. Cell 1991; 65:701-713.
11. Sewing A, Burger C, Brusselbach S et al. Human cyclin D1 encodes a labile nuclear protein whose
synthesis is directly induced by growth factors and suppressed by cyclic AMP. J Cell Sci 1993; 104
(Pt2):545-555.
12. Roussel M, Theodoras A, Pagano M, Sherr C. Rescue of defective mitogenic signaling by D type
cyclins. Proc Natl Acad Sci USA 1995; 92:6837-6841.
13. Surmacz E, Reiss K, Sell C, Baserga R. Cyclin D1 messenger RNA is inducible by platelet derived
growth factor in cultured fibroblasts. Cancer Res 1992; 52:4522-4525.
14. Sherr CJ. Cancer Cell Cycles. Science 1999; 274:1672-1677.
15. Sherr CJ, Roberts JM. CDK inhibitors: Positive and negative regulators of G1-phase progression.
Genes Dev 1999; 13:1501-1512.
16. Dulic V, Lees E, Reed SI. Association of human cyclin-E with a periodic G(1)-S phase protein
kinase. Science 1992; 257:1958-1961.
17. Dou QP, Levin AH, Zhao SC, Pardee AB. Cyclin-E and cyclin-A as candidates for the restriction
point. Cancer Res 1993; 53:1493-1497.
18. Keyomarsi K, Pardee AB. Redundant cyclin overexpression and gene amplification in breast cancer
cells. Proc Natl Acad Sci USA 1993; 90:1112-1116.
19. Koepp DM, Harper JW, Elledge SJ. How the cyclin Became a cyclin: Regulated proteolysis in the
cell cycle. Cell 1999; 97:431-434.
20. Won KA, Xiong Y, Beach D, Gilman MZ. Growth-regulated expression of D-type cyclin genes in
human diploid fibroblasts. Proc Natl Acad Sci USA 1992; 89:9910-9914.
21. Filmus J, Robles AI, Shi W et al. Induction of cyclin D1 overexpression by activated Ras. Oncogene
1994; 9:3627-3633.
22. Aktas H, Cai H, Cooper GM. Ras links growth factor signaling to the cell cycle machinery via
regulation of cyclin D1 and the CDK inhibitor p27KIP1. Mol Cell Biol 1997; 17:3850-3857.
23. Winston JT, Coats SR, Wang YZ, Pledger WJ. Regulation of the cell cycle machinery by
oncogenic Ras. Oncogene 1996; 12:127-134.
24. Tam SW, Theodoras AM, Pagano M. Kip1 degradation via the ubiquitin-proteasome pathway.
Leukemia 1997; 11 Suppl 3:363-366.
25. Blagosklonny MV, Wu GS, Omura S, El-Deiry WS. Proteasome-dependent regulation of p21WAF1/
CIP1 expression. Biochem Biophys Res Comm 1996; 227:564-569.
26. Campanero MR, Flemington EK. Regulation of E2F through ubiquitin-proteasome-dependent degradation:
Stabilization by the pRb tumor suppressor protein. Proc Natl Acad Sci USA 1997;
94:2221-2226.
27. Winston JT, Chu C, Harper JW. Culprits in the degradation of cyclin E apprehended. Genes Dev
1999; 13:2751-2757.
28. Varshavsky A, Turner G, Du F, Xie Y. The ubiquitin system and the N-end rule pathway. Biol
Chem 2000; 39(81):779-789.
62 Cell Cyle Checkpoints and Cancer
29. Diehl JA, Zindy F, Sherr CJ. Inhibition of cyclin D1 phosphorylation on threonine-286 prevents
its rapid degradation via the ubiquitin-proteasome pathway. Genes Dev 1997; 11:957-972.
30. Diehl JA, Cheng M, Roussel MF, Sherr CJ. Glycogen synthase kinase-3beta regulates cyclin D1
proteolysis and subcellular localization. Genes Dev 1998; 12:3499-3511.
31. Cheng M, Sexl V, Sherr CJ, Roussel MF. Assembly of cyclin D-dependent kinase and titration of
p27Kip1 regulated by mitogen-activated protein kinase kinase (MEK1). Proc Natl Acad Sci USA
1998; 95:1091-1096.
32. Hinds PW, Mittnacht S, Dulic V et al. Regulation of retinoblastoma protein functions by ectopic
expression of human cyclins. Cell 1992; 70:993-1006.
33. Harbour JW, Dean DC. The Rb/E2F pathway: Expanding roles and emerging paradigms. Genes
Dev 2000; 14:2393-2409.
34. Nevins JR. Toward an understanding of the functional complexity of the E2F and retinoblastoma
families. Cell Growth Diff 1998; 9:585-593.
35. Bartek J, Bartkova J, Lukas J. The retinoblastoma protein pathway and the restriction point. Curr
Opin Cell Biol 1996; 8:805-814.
36. Sheaff R, Groudine M, Gordon M et al. Cyclin E-CDK2 is a regulator of p27Kip1. Genes Dev
1997; 11:1464-1478.
37. Hunter T. Oncoprotein networks. Cell 1997; 88:333-346.
38. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000; 100:57-70.
39. Almasan A, Yin Y, Kelly RE et al. Deficiency of retinoblastoma protein leads to inappropriate
S-phase entry, activation of E2F-responsive genes, and apoptosis. Proc Natl Acad Sci USA 1995;
92:5436-5440.
40. DeGregori J, Leone G, Ohtani K et al. E2F-1 accumulation bypasses a G1 arrest resulting from
the inhibition of G1 cyclin-dependent kinase activity. Genes Dev 1995; 9:2873-2887.
41. Kowalik TF, DeGregori J, Schwarz JK, Nevins JR. E2F1 overexpression in quiescent fibroblasts
leads to induction of cellular DNA synthesis and apoptosis. J Virol 1995; 69:2491 2500.
42. Qin XQ, Livingston DM, Kaelin WG, Adams PD. Deregulated transcription factor E2F-1 expression
leads to S-phase entry and p53-mediated apoptosis. Proc Natl Acad Sci USA 1994;
91:10918-10922.
43. Planas-Silva MD, Weinberg RA. The restriction point and control of cell proliferation. Cur Opin
Cell Biol 1997; 9:768-772.
44. Weinberg RA. The retinoblastoma protein and cell cycle control. Cell 1995; 81:323-330.
45. Kohn KW. Functional capabilities of molecular network components controlling the mammalian
G1/S cell cycle phase transition. Oncogene 1998; 16:1065-1075.
46. Howard CM, Claudio PP, De Luca A et al. Inducible pRb2/p130 expression and growth suppressive
mechanisms: Evidence of a pRb2/p130, p27Kip1, and cyclin E negative feedback regulatory
loop. Cancer Res 2000; 60:2737-2744.
47. Zou X, Rudchenko S, Wong K, Calame K. Induction of c-myc transcription by the v-Abl tyrosine
kinase requires Ras, Raf1, and cyclin-dependent kinases. Genes Dev 1997; 11:654-662.
48. Blagosklonny MV, Prabhu NS, El-Deiry WS. Defects in p21WAF1/CIP1, Rb, c-myc signaling in
phorbol ester-resistant cancer cells. Cancer Res 1997; 57:320-325.
49. Mitchell, El-Deiry WS. Overexpression of c-myc inhibits p21WAF1/CIP1 expression and induces
S-phase entry in 12-O-tetradecanoylphorbol-13-acetate (TPA)-sensitive human cancer cells. Cell
Growth Diff 1999; 10:223-230.
50. Evan G, Littlewood T. A matter of life and cell death. Science 1998; 281:1317-1322.
51. Blagosklonny MV. A node between proliferation, apoptosis, and growth arrest. Bioessays 1999;
21:704-709.
52. MacLeod CL, Luk A, Castagnola J et al. EGF induces cell cycle arrest of A431 human epidermoid
carcinoma cells. J Cell Physiol 1986; 127:175-182.
53. Cowley S, Peterson H, Kemp P, Marshall CJ. Activation of MAP kinase kinase is necessary and
sufficient for PC12 differentiation and for transformation of NIH 3T3 cells. Cell 1994; 77:841-852.
54. Liu Y, Martindale JL, Gorospe M, Holbrook NJ. Regulation of p21 WAF1/CIP1 expression through
mitogen-activated protein kinase signaling pathway. Cancer Res 1996; 56:31-35.
55. Marshall CJ. Specificity of receptor tyrosine kinase signaling-transient versus sustained extracellular
signal-regulated kinase activation. Cell 1995; 80:179-185.
56. Pumiglia KM, Decker SJ. Cell cycle arrest mediated by the MEK/mitogen-activated protein kinase
pathway. Proc Natl Acad Sci USA 1997; 94:448-452.
57. Lloyd AC, Obermuller F, Staddon S et al. Cooperating oncogenes converge to regulate cyclin/
CDK complexes. Genes Dev 1997; 11:663-677.
58. Sewing A, Wiseman B, Lloyd AC, Land H. High-intensity Raf signal causes cell cycle arrest mediated
by p21CIP1. Mol Cell Biol 1997; 17:5588-5597.
The Restriction Point of the Cell Cycle 63
59. Woods D, Parry D, Cherwinski H et al. Raf-induced proliferation or cell cycle arrest is determined
by the level of Raf activity with arrest mediated by p21CIP1. Mol Cell Biol 1997; 17:5598-5611.
60. Blagosklonny MV. The mitogen-activated protein kinase pathway mediates growth arrest or
E1A-dependent apoptosis in SKBr3 human breast cancer cells. Int J Cancer 1998; 78:511-517.
61. Lee AWM, Nambirajan S, Moffat JG. CSF-1 activates MAPK-dependent and p53 independent
pathways to induce growth arrest of hormone-dependent human breast cancer cells. Oncogene
1999; 18:7477-7494.
62. Coppock DL, Buffolino P, Kopman C, et al. Inhibition of the melanoma cell cycle and regulation
at the G1/S transition by 12-O-tetradecanoylphorbol-13-acetate (TPA) by modulation of CDK2
activity. Exp Cell Res 1995; 221:92-102.
63. Desrivieres S, Volarevic S, Mercep L, Ferrari S. Evidence for different mechanisms of growth inhibition
of T-cell lymphoma by phorbol esters and concanavalin A. J Biol Chem 1997; 272:2470-2476.
64. Gartel AL, Najmabadi F, Goufman E, Tyner AL. A role for E2F1 in Ras activation of p21(WAF1/
CIP1) transcription. Oncogene 2000; 19:961-964.
65. Fan W, Richter G, Cereseto A et al. Cytokine response gene 6 induces p21 and regulates both cell
growth and arrest. Oncogene 1999; 18:6573-6582.
66. Bunz F, Dutriaux A, Lengauer C et al. Requirement for p53 and p21 to sustain G2 arrest after
DNA damage. Science 1998; 282:1497-1501.
67. Blagosklonny MV, Robey R, Bates S, Fojo T. Pretreatment with DNA-damaging agents permits
selective killing of checkpoint-deficient cells by microtubule-active drugs. J Clin Invest 2000;
105:533-539.
68. Lin AW, Barradas M, Stone JC et al. Premature senescence involving p53 and p16 is activated in
response to constitutive MEK/MAPK mitogenic signaling. Genes Dev 1998; 12:3008 3019.
69. Cheng M, Olivier P, Diehl JA et al. The p21CIP1 and p27Kip1 CDK ‘inhibitors’ are essential
activators of cyclin D-dependent kinases in murine fibroblasts. EMBO J 1999; 18:1571-1583.
70. Weiss RH, Joo A, Randour C. 21(Waf1/Cip1) is an assembly factor required for platelet derived
growth factor-induced vascular smooth muscle cell proliferation. J Biol Chem 2000; 275:10285-10290.
71. Besson A, Yong VW. Involvement of p21(Waf1/Cip1) in protein kinase C alpha-induced cell cycle
progression. Mol Cell Biol 2000; 20:4580-4590.
72. Salnikow K, Costa M, Figg WD, Blagosklonny MV. Hyperinducibility of hypoxia responsive genes
without p53/p21-dependent checkpoint in aggressive prostate cancer. Cancer Res 2000;
60:5630-5634.
73. Sgambato A, Han EK, Zhou P et al. Overexpression of cyclin E in the HC11 mouse mammary
epithelial cell line is associated with growth inhibition and increased expression of p27(Kip1). Cancer
Res 1996; 56:1389-1399.
74. Hu PPC, Shen X, Huang D et al. The MEK pathway is required for stimulation of p21(WAF1/
CIP1) by transforming growth factor-beta. J Biol Chem 1999; 274:35381-35387.
75. Polyak K, Kato JY, Solomon MJ et al. p27Kip1, a cyclin-CDK inhibitor, links transforming growth
factor-beta and contact inhibition to cell cycle arrest. Genes Dev 1994; 8:9-22.
76. Sherr CJ. The Pezcoller lecture: Cancer Cell Cycle Revisited. Cancer Res 2000; 60:3689 3695.
77. Malumbres M, De Castro IP, Hernandez MI et al. Cellular response to oncogenic Ras involves
induction of the CDK4 and CDK6 inhibitor p15(INK4b). Mol Cell Biol 2000; 20:2915 2925.
78. Delgado MD, Vaque JP, Arozarena I et al. H-, K- and N-Ras inhibit myeloid leukemia cell proliferation
by a p21(WAF1)-dependent mechanism. Oncogene 2000; 19:783-790.
79. Zhou B-BS, Elledge SJ. The DNA damage response: putting checkpoints in perspective. Nature
2000; 408:433-439.
80. Flatt PM, Tang LJ, Scatena CD et al. p53 regulation of G2 checkpoint is retinoblastoma protein
dependent. Mol Cell Biol 2000; 20:4210-4223.
81. Wyllie FS, Haughton MF, Bond JA et al. S phase cell-cycle arrest following DNA damage is
independent of the p53/p21(WAF1) signalling pathway. Oncogene 1996; 12:1077-1082.
82. Agarwal ML, Agarwal A, Taylor WR et al. A p53-dependent S-phase checkpoint helps to protect
cells from DNA damage in response to starvation for pyrimidine nucleotides. Proc Natl Acad Sci
U S A 1998; 95:14775-14780.
83. Chan TA, Hwang PM, Hermeking H et al. Cooperative effects of genes controlling the G2/M
checkpoint. Genes Dev 2000; 14:1584-1588.
84. Chang B-D, Broude EV, Dokmanovic M et al. A senescence-like phenotype distinquishes tumor
cells that undergo terminal proliferation arrest after exposure to anticancer drugs. Cancer Res 1999;
59:3761-3767.
85. Chang BD, Broude EV, Fang J et al. p21WAF1/Cip1/Sdi1-induced growth arrest is associated
with depletion of mitosis-control proteins and leads to abnormal mitosis and endoreduplication in
recovering cells. Oncogene 2000; 19:2165-2170.
64 Cell Cyle Checkpoints and Cancer
86. Clarke DJ, Gimenez-Abian JF. Checkpoints controlling mitosis. Bioessays 2000; 22:351 363.
87. Blagosklonny MV, Fojo T. Molecular effects of Paclitaxel: Myths and reality. Int J Cancer 1999;
83:151-156.
88. Agami R, Bernards R. Distinct initiation and maintenance mechanisms cooperate to induce G1
cell cycle arrest in response to DNA damage. Cell 2000; 102:55-66.
89. Brugarolas J, Moberg K, Boyd SD et al. Inhibition of cyclin-dependent kinase 2 by p21 is necessary
for retinoblastoma protein-mediated G1 arrest after gamma-irradiation. Proc Natl Acad Sci
USA 1999; 96:1002-1007.
90. Swarbrick A, Lee CS, Sutherland RL, Musgrove EA. Cooperation of p27(Kip1) and p18(INK4c)
in progestin-mediated cell cycle arrest in T-47D breast cancer cells. Mol Cell Biol 2000;
20:2581-2591.
91. Gong J, Traganos F, Darzynkiewicz Z. Staurosporine blocks cell progression through G1 between
the cyclin D and cyclin E restriction points. Cancer Res 1994 Jun 15;54(12):3136-9 1994;
54:3136-3139.
92. Sandor V, Senderowicz A, Mertins S et al. P21-dependent G1 arrest with downregulation of cyclin
D1 and upregulation of cyclin E by the histone deacetylase inhibitor FR901228. Br J Cancer
2000; 83:817-825.
93. Pardee AB, James LJ. Selective killing of transformed baby hamster kidney (BHK) cells. Proc Natl
Acad Sci USA 1975; 72:4994-4998.
94. Darzynkiewicz Z. Apoptosis in antitumor strategies: modulation of cell-cycle or differentiation. J
Cell Biochem 1995; 58:151-159.
95. Cherington PV, Smith BL, Pardee AB. Loss of epidermal growth factor requirement and malignant
transformation. Proc Natl Acad Sci U S A 1979; 76:3937-3941.
96. Blagosklonny MV, Bishop PC, Robey R et al. Loss of cell cycle control allows selective
microtubule-active drug-induced Bcl-2 phosphorylation and cytotoxicity in autonomous cancer cells.
Cancer Res 2000; 60:3425-2428.
97. Horwitz SB. Mechanism of action of Taxol. Trends Pharmacol Sci 1992; 13:134-136.
98. Jordan MA, Wilson L. Microtubules and actin filaments: dynamic targets for cancer chemotherapy.
Curr Opin Cell Biol 1998; 10:123-130.
99. Li W, Fan J, Banerjee D, Bertino JR. Overexpression of p21 (waf1) decreases G2-M arrest and
apoptosis induced by Paclitaxel in human sarcoma cells lacking both p53 and functional Rb protein.
Mol Pharmacol 1999; 55:1988-1093.
100. Stone S, Dayananth P, Kamb A. Reversible, p16-mediated cell cycle arrest as protection from
chemotherapy. Cancer Res 1996; 56:3199-3202.
DNA Damage, Cell Cycle Control, and Cancer 65
CHAPTER 4
Cell Cycle Checkpoints and Cancer, edited by Mikhail V. Blagosklonny. ©2001 Eurekah.com.
DNA Damage, Cell Cycle Control and Cancer
Jens Oliver Funk, Temesgen Samuel and H. Oliver Weber
Abstract
Cell cycle checkpoints constitute a network of signal transduction mechanisms to monitor
DNA damage and regulate progression through the cell cycle. A series of events is
triggered in cells upon DNA damage. Here we describe a framework for the
understanding of the functions of the core components involved in the cell cycle response to
DNA damage and the relevance to the origin of cancer.
Introduction
Various mechanisms exist to maintain genetic stability of cells facing DNA damage. A
state of genetic instability may be seen at the chromosomal level, at the nucleotide level, or be
reflected by chromosomal translocations and gene amplifications.1 Decisions to enter S phase
and proliferate, arrest in the G0/G1 phase of the cell cycle or differentiate are based on multiple
internal and external environmental stimuli. These processes are tightly regulated and well
balanced. A hallmark of cancer cells is that the normal balance of these processes is perturbed,
that they are prone to uncontrolled proliferation, and that they may further display progressive
genetic instability.2
Multiple pathways are involved in the maintainance of genetic integrity, most of which
link to the cell cycle.3 The inactivation of these pathways as part of a multi-step process contributes
significantly to the origin of tumors. By arresting the cell cycle, checkpoints presumably
allow cells to repair DNA. Checkpoints can be seen as a network of surveillance systems,
i.e., signal transduction systems, that interrupt cell cycle progression, when damage to the
genome or failure of a previous activity in the cell cycle is detected.3,4 Some cell types may
primarily undergo apoptosis, avoiding the risk of generating genetically altered progeny.
In order to signal cell cycle arrest, checkpoint control pathways must sense DNA damage
and then transduce the signal. To delay cell cycle progression after DNA damage, these mechanisms
affect the activity of critical cell cycle regulators. Conceptually, checkpoint control pathways
consist of three elements: stimuli (i.e., different types of DNA damage), a signal transduction
machinery, and its targets (i.e., different basal cell cycle regulators). The DNA damage
checkpoint5,6 can be regarded as a coherent signal transduction system which allows cells to
transfer the information from a DNA lesion to the cell cycle. This signal affects at least three
stages in the cell cycle: the G1/S transition, S phase progression and the G2/M boundary.
Although there are many points of cell cycle arrest, the DNA damage surveillance system can
be regarded as one checkpoint because all forms of arrest are signaled by DNA damage, often
simultaneously, and many genes are involved in more than one stage of the arrest.
In this Chapter, we focus primarily on the main signal transduction pathways and targets and
describe recent concepts, as to how mammalian cells arrest the cell cycle after DNA damage, and
to what extent some of these scenarios might be involved in the generation of cancer.
66 Cell Cycle Checkpoints and Cancer
Origins of DNA Damage
DNA damage could broadly be understood as any major change in the basic structure and
function of the DNA double helix that needs repair. Although any of the polynucleotide structures
can be damaged, the genomic DNA is of particular interest due to heritability of such
changes.
DNA damage can originate from internal or external causes. Damage from internal sources
may arise from biochemical products and replication errors, which occur during cellular
metabolism and division. External causes include irradiation and mutagenic chemicals.
Importantly, different types of DNA damage can occur during different cell cycle phases, e.g.,
oxidative damage mostly accumulates in G1, incomplete replication or nucleotide
misincorporation occurs in S phase, and cells in mitosis may be prone to chromosome breakage
during chromatid segregation.
DNA Damage of Intrinsic Origin
Oxygen and its metabolic products, like reactive oxygen species, are one major source of
intrinsic DNA damage. The high reactivity of DNA as a biomolecule renders it particularly
susceptible to this process. A mammalian genome may undergo over 105 modifications a day,7
and it is estimated that mitochondrial DNA is oxidized at the frequency of 1 base per 100kb.8
Certain physiological processes may also constitute a form of DNA damage. Recombination
during VDJ immunoglobulin synthesis or meiosis, and the DNA replication process by itself
incorporate steps of DNA breakage, which can be considered as DNA damage.9
DNA Damage of External Origin
DNA damaging agents may interfere with DNA structure and function in various ways,
some of which are still incompletely understood. The type, extent, and outcome of these types
of damage depend on the dose and duration of the damaging stimuli, as well as the sensitivity
of a cell. For example irradiation of cells by ultraviolet (UV) or ionizing radiation (IR) may
cause cross-links, double strand breaks, or may just knock off sub-atomic particles to generate
radicals in the DNA structure. Isolated nucleotide crystals have been irradiated to induce free
radical formation on both base and sugar moieties,10 indicating that the damage caused by
radiation can affect both the bases and the backbone of the DNA. Moreover, clustered damage
during IR resulted from both low and high radiation doses,11 implying that the outcome depends
on the type of radiation rather than on its dose. Evidence for the variability of cellular
response to the type of irradiation stems mostly from the analysis of ataxia telangiectasia (AT)
cells which are exquisitely sensitive to IR but not UVR (see below).12,13
Chemicals may interfere with DNA integrity and replication fidelity by reacting with
atoms in the DNA molecule and by modifying nucleotides, thereby distorting the normal
base-pairing pattern. Similar to the damage by IR, the nature and hence the response to DNA
damage induced by chemical agents is variable. Multiple chemical agents that are known to be
mutagenic/carcinogenic or to damage DNA are used for the induction of (programmed) cell
death, e.g., in anticancer therapy. Given the variability of DNA damaging agents, mechanisms
of damage, and their cellular responses it is necessary to consider DNA damage not as a single
entity, but a whole spectrum of insults to a cell which in turn may react with a spectrum of
cellular responses.
Upstream DNA Damage Signaling
In response to DNA damage, especially IR and UVR, two phosphatidyl-inositol kinase
(PIK)-related proteins, ataxia telangiectasia mutated (ATM) and ATM-rad3-related (ATR),
play an important role. They are characterized by similar structure and have sequence homologies
to PI3K; however, they are not lipid but protein kinases. After DNA damage, they function
as the most proximal signal transducers. Although ATR appears to be particularly important
in the cellular response to UVR, it also participates in other DNA damage pathways, since
DNA Damage, Cell Cycle Control, and Cancer 67
cells expressing a dominant-negative ATR are sensitive to many forms of DNA damage.14 So
far, no human disease is known to involve exclusively ATR, and ATR-/- mice are not viable.15
Therefore less is known about ATR function, but available data collectively indicates that ATR
operates in pathways parallel to ATM. Here, we focus on signalling of ATM and its downstream
targets, CHK2 and p53 (Fig. 1).
ATM-Dependent Signaling Pathways
The ATM gene was found to be inactivated in the autosomal recessive disorder AT.16 AT
is characterized by diverse symptoms, such as cerebellar degeneration, oculocutaneous telangiectasia,
immunodeficiency, hypogonadism, chromosomal instability, premature aging, cancer
predisposition, extreme IR sensitivity, and cell cycle abnormalities.17,18 Therefore, ATM must
have multiple cellular functions. In the majority of AT patients ATM is inactivated by mutations.
The product of the ATM gene is a 370 kD phosphoprotein of variable cellular localization.
Most notably, ATM is involved in the regulation of cellular responses to IR, while it
appears to be less important for signaling after UVR, damage by alkylating agents, or intrinsic
damage. It appears that ultimately the vast majority of cellular responses to DNA double strand
breaks (of different origin, including IR, radiomimetic agents, or topoisomerase inhibitors)
depends on intact ATM. ATM is a serine/threonine protein kinase whose phosphorylation
targets include p53, MDM2, CHK2, NBS1, and BRCA1. The functional interaction with
p53 (see below) is intriguing due to p53’s well known and universal relevance in DNA damage
scenarios. The observation that some of these targets are still phosphorylated in IR-treated AT
cells, albeit with delayed kinetics, points to the existence of additional pathways responding to IR.
Importantly, features of ATM-/- mice closely resemble the human disease.19 The mice are
viable and show growth retardation and infertility. This further shows that ATM also functions
in development since double strand breaks may occur as a part of physiological processes.20
Cells derived from AT patients are defective in G1, S and G2 arrest following IR.21,22 The
ATM kinase activity increases after IR by mechanisms poorly understood. Probably double
strand breaks induced by IR damage either directly or indirectly signal to ATM. Extensive
studies have been carried out on the mechanisms by which ATM might transduce signals from
damaged DNA to other cellular partners.
The exact nature of the exquisite radiosensitivity of AT cells is debated. While there is no
solid evidence for a deficient double strand break repair in AT cells, there is evidence that
defects in cell cycle checkpoints are responsible for this radiosensitivity. However, several arguments
point to an opposite functional relationship:23,24 Radiosensitivity and cell cycle defects
are separable phenotypes, and there is no correlation between radiosensitivity and increased
apoptosis induced by IR in AT cells. Finally, even under experimental conditions with additional
time for repair of DNA damage, the survival of AT cells is not increased. Instead, it is
assumed that subtle alterations in double strand break rejoining activity and chromatin, or
other cellular activities of ATM, such as induction of NF-κB, could be involved in the
radiosensitivity.13,25
CHK2—The Next Line of Defense
CHK2 is a protein kinase which is activated, by post-translational modifications, after
DNA damage. CHK2 is phosphorylated in an ATM-dependent manner after IR and in an
ATM-independent manner after UVR or stalled replication (Fig. 1). Another homolog, CHK1,
is phosphorylated after DNA damage in an ATR-dependent manner, but its precise role in
mediating downstream signals remaines elusive. Therefore, here we will discuss only CHK2
activities and refer to these in the context of the G1/S and G2/M checkpoints (see below). The
best known targets identified in vitro are p53 and CDC25C; however, the in vivo relevance of
the mechanisms involved in the DNA damage reponse remains to be shown.
68 Cell Cycle Checkpoints and Cancer
p53—The Core of the DNA Damage Pathways
The p53 tumor suppressor protein is a transcription factor with “sensor functions”
integrating signals from the external and internal environment.26,27 Biochemically, p53 is a
sequence-specific transcriptional activator and a nonspecific transcriptional repressor. It has
been known for years that the half-life of the otherwise short-lived p53 is increased several fold
in cells treated with DNA-damaging reagents.28-30 This correlates well with cellular responses
such as cell cycle arrest or apoptosis, depending on the particular cell type and damaging
agent.26,31 However, it is still poorly understood why one cell type undergoes cell cycle arrest
whereas another cell type dies by apoptosis.
There are currently two models that try to explain these phenomena.32 According to one
model, activated p53 always sends the same signals of cell cycle arrest and apoptosis, but apoptosis
occurs only if other pathways, e.g., survival or oncogenic pathways, either augment or impede
apoptosis, possibly incorporating also p53-independent signals. The other model suggests that
the choice lies mainly within the activities of p53 itself, either determined by p53 protein
abundance-low protein level leading to cell cycle arrest, high protein level leading to apoptosisor
specific posttranslational modifications. These models are not mutually exclusive. In both
cases p53 is likely regulated by other cellular partners in a specific manner to tightly control its
possibly deadly effect on the cell.
Thus, p53 is a central DNA damage checkpoint gene and is clearly required for a complete
DNA damage response. However, cells with mutant p53 grow normally and p53-/- mice
are normal, indicating that p53 is not essential for cell cycle progression.33 Importantly, the
fidelity of mitotic chromosome transmission is reduced in p53 mutant cells. Furthermore, the
rate of gene amplification is several orders of magnitude higher as compared to p53 wild-type
Fig. 1. Simplified model of the core cell cycle pathways induced by DNA damage, such as UVR or IR.
DNA Damage, Cell Cycle Control, and Cancer 69
cells,34 and p53-/- mice exhibit an abnormally high frequency of spontaneous cancers. As with the
other checkpoint genes in this pathway, such as the downstream genes p21 and 14-3-3σ,35 this
clearly points to subtle defects accumulating in mutant cells, even in the
absence of external DNA damage, which likely contribute to genetic instability. If an additional
disturbance intervenes, even if not primarily inducing DNA damage, a defect phenotype may
then be obvious.
Regulatory Effects Converging on p53
Multiple, mainly posttranslational, mechanisms regulate p53 activity. The best characterized
mechanisms include the regulation of p53 protein half-life, phosphorylation, and acetylation
events, all of which contribute to influence p53 stability.36 The stability can also be regulated
by the interaction with cellular partners of p53.
The MDM2 oncoprotein is a crucial player in the regulation of p53 by targeting p53 for
degradation in the ubiquitin-dependent proteosome pathway.37-39 Central to this degradation
process is the direct binding of MDM2 to p53. In addition, MDM2 is also able to directly
inhibit the transactivational activity of p53.40-42 Since MDM2 is transactivated by p5339 this
constitutes a tightly regulated feedback loop between both partners which likely reflects a wellbalanced
equilibrium.
The importance of the MDM2/p53 interaction in the regulation of the p53 protein levels
and functions is also evident from the fact that different pathways involve MDM2 or the
MDM2/p53 complex. A product from the INK4a/ARF tumor suppressor gene locus, p14ARF,
is a cell cycle inhibitor43 whose functions are, at least in part, p53-dependent. Upon oncogenic
signals such as c-MYC and E1A,44,45 ARF binds to MDM2 and inhibits the ability of MDM2
to promote p53 degradation, thus increasing its half-life and transactivation of p53-responsible
genes.46-49 It is open as to what extent ARF might also contribute to DNA-damage responses.50
Recently, ARF was shown to be implicated in p53-dependent cell cycle arrest after IR exposure
and mictrotubular disruption, while it did not affect arrest after ribonucleotide depletion or
actinomycin D treatment.51
p53 is also regulated by its phosphorylation and acetylation. This includes the phosphorylation
of serine 6/9,52 serine 15,53 serine 20,54 serine 33,55 serine 37,56 serine 315,57 serine 378,58
serine 392,59 as well as acetylations at lysine 382 and 320.60 The appearance of so many regulatory
phosphorylation/acetylation sites on p53 suggest a network of different signals
integrating into p53.36 Indeed many kinases have been connected to specific phosphorylation
events on the p53 protein, including CKI, CAK, CDK2, CDK1, pCAF, PKC, and p300.
IR leads to the phosphorylation of serines 15/20, both of which have been shown to
negatively influence the binding of MDM2 to p53 leading to stabilization of p53.53,61,62 Serine
15 is phosphorylated following IR by ATM placing p53 in an IR-induced DNA damage pathway
downstream of ATM (Fig. 1). ATR is also able to phosphorylate serine 15. However, this
occurs following UVR and not IR exposure.14 In contrast, serine 20 is phosphorylated following
IR in an ATM-dependent manner, not directly by ATM but indirectly through the CHK2
kinase.63 ATM is able to phosphorylate CHK2, which in turn phosphorylates serine 20 on p53
leading to stabilization of p53.63-65 In addition, the activation of CHK2 by ATM-mediated
phosphorylation leads to phosphorylation of CDC25C (see below).
Many phosphorylation/acetylation events affect the carboxy-terminal end of p53 which is
implicated in the regulation of the DNA binding capacities of p53. Phosphorylations at
carboxyterminal amino acids, binding with a specific antibody directed against the carboxy
terminus, and acetylations at lysine 320/382 have been shown to lead to an increased specific
DNA-binding function of p53.60,66,67 It appears, that many phosphorylation/acetylation events
of p53 reported in the last years have a predominantly cooperating effect in regulating p53
stability and transcriptional activity.68,69
70 Cell Cycle Checkpoints and Cancer
The G1/S Checkpoint
At the G1/S boundary the prerequisites for continuation with DNA replication are checked.
DNA damage leads to arrest in G1 via p53-dependent transactivation of genes, primarily p21.
Importantly, even in the absence of extrinsic DNA damage, unregulated S phase entry can lead
to genetic instability or apoptosis. The exact nature of DNA damage associated with unregulated
progression into S phase is unknown. Examples include experiments with overexpressed
G1 cyclins.70-72 Such cells enter S phase prematurely, after a shortened G1, exhibit genetic
instability,73 and an enhanced dependence on checkpoint functions for survival.9 This might
be explained by DNA damage occurring before S phase that cannot be repaired during replication.
On the other hand, premature entry into S phase could be an additional cause of DNA
damage, e.g., either due to failure to complete replication before entry into mitosis, or due to
abnormal nucleotide pools as a result of shortened G1. The importance of maintaining sufficient
nucleotide pools before entering S phase is reflected in the fact that cells presumably have a
mechanism to monitor nucleotide pools and arrest in a p53-dependent manner if these pools
are insufficient.74
p21CIP1—A Two-Tailed Cell Cycle Regulator
Among the transcriptional targets of p53, the CDK inhibitor p21CIP1 plays a key role in
mediating G1 arrest. Cells lacking p21 have a defect in the DNA damage-induced G1/S
arrest.75,76 The inhibition of CDK activities by CDK inhibitors constitutes a powerful cell
cycle control mechanism and provides an important link to other signaling pathways during
proliferation, differentiation, and senescence.77 The CIP/KIP family members p21CIP1, p27KIP1
and p57KIP2 share broad specificity for binding to and inhibition of most CDK/cyclin complexes
through conserved motifs for CDK and cyclin Binding in the amino termini of the
inhibitors,78-80 but only p21 is directly involved in the DNA damage-induced cell cycle arrest.
The identification of similar cyclin Binding motifs in several unrelated proteins, which may
function as inhibitors, substrates, or associated regulators of CDK/cyclin complexes, stresses the
importance of these motives for various cell cycle-related activities.81 The CIP/KIP inhibitors can
also function as adaptors to promote CDK/cyclin complex assembly.82
p21 is distinguished from the other CDK inhibitors by several unique features (Fig. 2).
The carboxy terminus of p21 binds to proliferating cell nuclear antigen (PCNA) and thereby
inhibits PCNA-dependent DNA replication but not PCNA-dependent nucleotide excision
repair.83,84 A second cyclin-binding motif overlaps with the PCNA binding motif and can
inhibit CDK activity in vitro79,80,85 and induce G1 arrest in vivo,85 though its precise role in
modulating the cell cycle is unclear. This indicates that inhibition of PCNA is not the sole
activity of the p21 carboxy terminus that is required for the cell cycle arrest.86 Inactivation of
the p21 inhibitory activities could occur by several mechanisms,81 which might be involved in
tumorigenesis.87 The carboxy-terminal nuclear localization signal in p21 also serves to target
CDK/cyclin complexes to the nucleus. We have further proposed that proteins which bind to
the carboxy terminus of p21 could also target p21/CDK/cyclin complexes to other subcellular
locations or substrates.81
Given the critical role p21 plays in coordinating cell cycle and G1 transition, DNA
replication, terminal differentiation and senescence, it was surprising that its elimination did
not result in a tumorigenic phenotype in mice.88 Furthermore, mutations in p21 are rarely
found in human tumors. Perhaps the positive requirement for p21 in CDK/cyclin Assembly or
other cell cycle-related functions precludes mutations. Also, other proteins that modulate p21
activity may be mutated in cancer.
Recently, a role for p21 in the G2/M phase of the cell cycle was identified;89-92 however, the
mechanisms underlying this function remaine elusive. Though supraphysiological levels of p21
can inhibit the activity of CDK1/cyclin B1 complexes in vitro,85 it is doubtful that this is the
primary mechansim by which p21 regulates G2/M. Disruption of the p21/p53 pathway has been
associated with mitotic spindle pole defects and the appearance of multiple centrosomes,93,94 and
DNA Damage, Cell Cycle Control, and Cancer 71
p21-/- cells showed S/M uncoupling with subsequent polyploidy after treatment with DNA
damaging agents.95 In the absence of any direct mechanistic link to the mitotic spindle poles and
centrosome, recent evidence has suggested that these abnormalities may arise as a result of
alterations in CDK2/cyclin E activity at the time of centrosome duplication in S phase.96 A recent
model further illuminating the function of p21 in G2/M is discussed below.
The G2/M Checkpoint
The G2/M checkpoint pathways converge mainly on activity and intracellular localization
of the CDK1/cyclin B1 complex. Activation of this complex is essential for cells to enter mitosis.
Just prior to mitosis, CDK1/cyclin B1 complexes translocate to the nucleus and trigger the
initiation of mitotic changes like chromosome condensation and nuclear membrane breakdown.
Control of the Unperturbed G2/M Transition
The entry into mitosis is regulated by the spatial distribution of the CDK1 kinase complex:
during interphase, CDK1 is sequestered in the cytoplasm.97 Also, cyclin B1 is predominantly
cytoplasmic due to CRM-1-mediated export of cyclin B1 out of the nucleus.98,99 The cytosolic
retention of the complex is also complemented by the inactivating phosphorylation of CDK1
(see below).
The activation of the CDK1/cyclin B1 complex requires that the inhibitory phosphorylations
at threonine 14 and tyrosine 15 of CDK1 are removed by a phosphatase. The known
phosphatase for this activating dephosphorylation is CDC25C, while the inhibitory phosphorylation
is performed by the kinases Weel100 and myt1.101 Activation of CDK1 by dephosphorylation
is associated with entry of cells into mitosis and is accompanied by chromosome condensation
and nuclear membrane breakdown. Some reports indicate, however, that early nuclear
mitotic events take place before CDK1/cyclin B1 complexes are intranuclear.102
CARB (CIP1-associated regulator of cyclin B1), a protein functioning to regulate cyclin
B1 in a p21-dependent manner, has also been recently suggested to contribute to the cytosolic
Fig. 2. Schematic model of the p21 protein and its key functional activities. Apparently, p21 can localize
either to the nuclear or the centrosomal compartment.
72 Cell Cycle Checkpoints and Cancer
retention of cyclin B1 proximal to the centrosome.103 Cyclin B1 is subject to tight control of
subcellular localization.92,99,104,105 Sequestration of cyclin B1 to centrosomal CARB may form
a storage site for cyclin B1 from which it can be rapidly mobilized, or may protect cyclin B1
from proteolytic degradation. Consistent for a role for p21 in CARB regulation, the cyclin B1-
CARB association is potently inhibited by the carboxy terminus of p21, suggesting that free
p21 is able to dissociate this complex under certain conditions. Since CARB is found associated
with either p21 or cyclin B1 but not with both, it seems unlikely that the association of
CARB and cyclin B1 is mediated by p21.103 CARB sequesters cyclin B1 in the cytoplasm close
to the nuclear membrane, and thus may regulate the availability of cyclin B1 for nuclear import
(Fig. 2 and 3). A similar suggestion that cyclin B1 is retained in the cytoplasm by an active
mechanism has been proposed previously106 but a candidate molecule was not identified. Sequestration
by CARB complements mechanisms for the CDK1/cyclin B1 regulation. CARBdependent
regulation of cyclin B1 bioavailability may regulate the onset of prophase, since
availability of p21 for CARB/cyclin B1 complex disruption could only occur if p21 was not
required for nucleotide excision repair or affecting other growth arrest stimuli.
This suggests that the p21 carboxy terminus can act in a manner to promote release of
cyclin B1 from CARB. This function is distinct from the potent inhibitory activities of p21,
and one interpretation of the data to date would be that p21 actually promotes the progression
into M phase by increasing cyclin B1 bioavailability. As a consequence, the evolution of such
bifunctionality would confer no selective advantage to cells which completely eliminate p21
since under DNA damaged conditions they would be able to abrogate the G1/S damage checkpoint
but be unable to efficiently initiate mitosis. Such a possibility may explain the lack of
sporadic tumor formation in p21-/- mice and the very low frequency of p21 inactivations in
human cancers.107,108
Regulation of the CDC25C Phosphatase
Although the regulation of CDK1 by CDC25C phosphatase is well established,109-111 the
dynamics and regulation of CDC25C remain to be clearly shown. CDC25C can be phosphorylated
at two major sites, at serine 216 and at its uncharacterized amino-terminal end. Serine
216, which appears to be the major regulatory site, is phosphorylated by CHK1 and
CHK2.64,111,112 C-TAK1 is another kinase that has been shown to phosphorylate serine 216 in
vitro,113 although the relevance of this event in vivo is unknown. The amino terminus is phosphorylated
by CDK1114 and the polo-like kinase Plk.115 Whereas serine 216 phosphorylation
inhibits CDC25C phosphtase activity, aminoterminal phosphorylation enhances the activity.
Serine 216 of CDC25C is also a regulatory residue which upon phosphorylation dictates
the localization of the protein (Fig. 3). Phosphorylation of this residue, as it occurs in interphase
or after DNA damage, may regulate its subcellular localization by creating a 14-3-3
binding site.113,116 The binding to 14-3-3 proteins prevents the nuclear import of CDC25C117
and hence prevents the G2/M transition. On the other hand, in a feedback loop, CDK1
phosphorylates the activating amino-terminal region of CDC25C during mitotic progression,
thereby amplifying its own activation.118 The role of CDC25C phosphorylation by Plk
in vivo is not known.
DNA Damage and the G2/M Transition
CHK1, CHK2, and ultimately p53 are phosphorylated upon DNA damage. In addition
to phosphorylating p53, both CHK1 and CHK2 also directly phosphorylate CDC25C at
serine 216, leading to its inactivation. This inactivation could lead to a p53-independent G2
arrest by blocking CDK1 activation.119 Since serine 216-phosphorylated CDC25C binds to
14-3-3 proteins, remains cytoplasmic,116,117 and is probably unable to activate CDK1, the exit
from G2 and hence entry into mitosis is prevented after DNA damage (Fig. 3). This implies
that the DNA damage-induced G2 arrest is doubly secured through p53-dependent and
-independent mechanisms, by which CDK1 and CDC25C are sequestered by 14-3-3 proteins
and kept inactive due to sustained phosphorylation.97,116,117
DNA Damage, Cell Cycle Control, and Cancer 73
14-3-3σ is upregulated in a p53-dependent manner and is involved in the G2 checkpoint
arrest.120 Cells lacking 14-3-3σ are defective in G2 arrest and undergo mitotic catastrophe
upon DNA damage.97 Cells lacking both 14-3-3σ and p21 are even more sensitive to DNA
damage35 indicating the cooperative role of these inhibitors in the G2 arrest. However, some
evidences also indicate a positive or negative role of p21 in G2/M (see above). The nuclear
accumulation of p21 at the onset of mitosis,89 as well as the regulation of the CARB-cyclin B1
interaction by p21,103 hint at the positive regulation of p21 during G2/M. On the other hand
the observation that p21 inhibits phosphorylation of CDK1 at threonine161121 contributing
to the G2 arrest, indicates a negative role for p21 at G2/M. Further work is needed to clarify the
contribution of each of these mechanisms to G2 arrest. Finally, it is debated, to what extent the
upregulation of apoptotic regulators, such as BAX,122 may favor an apoptotic outcome, even in
cells overcoming G2 arrest.
Links to Cancer and Genetic Instability
Loss of cell cycle checkpoint control has emerged as a central cause of genetic instability.3,6
Consequently, chances that these unstable cells progress to cancer are increased. This notion
has several important implications:
1. Since checkpoints may determine the ultimate response (arrest vs. apoptosis), the integrity
of these checkpoints influences the susceptibility of cells to DNA damage. This is relevant
either to the cells’ fate after accumulation of undesired DNA damage or to the cells’ sensitivity
to desired damage during chemo-/radiotherapy.
2. Exploring the early checkpoint defects in cancerous or precancerous lesions may serve as a
prognostic or, in certain tissues, as an additional diagnostic marker.
Fig. 3. Simplified model of the G2/M transition and the key proteins involved in the regulation of the
CDK1/cyclin B1 complex.
74 Cell Cycle Checkpoints and Cancer
3. Furthermore, known defects of pivotal checkpoint genes may help to predict treatment
outcome or to design more specific therapeutic strategies. In addition, checkpoint components
which are defective in certain cancer cells may be targeted during therapy to enhance
the anti-tumor effect, e.g., by preventing arrest and/or by forcing cells into apoptosis. Work
is in progress to develop novel therapeutic strategies with an increased therapeutic index.
4. Moreover, strategies could be considered to restore missing or dysfunctional checkpoints in
order to provide additional time for DNA repair and delay the onset of cancer.
5. Finally, since some of the components that are involved in the DNA damage checkpoint are
also involved in other cellular regulatory activities, e.g., during senescence, differentiation,
or certain immunological responses, this could lead to cross-signalling into other pathways
and might permit new strategies to influence related cellular functions.
Acknowledgments
Research in J.O.F.’s lab is funded by grants from the Deutsche Forschungsgemeinschaft,
the Boehringer Ingelheim Fonds, and the ELAN-Program of the University of Erlangen-
Nuremberg. We thank Andreas Baur, Jason Eppel, Manfred Lutz, Andy McShea, Pia Rauch,
and Berlinda Verdoodt for comments and discussion, and Gerold Schuler and Alexander
Steinkasserer for support and encouragement.
References
1. Lengauer C, Kinzler KW, Vogelstein B. Genetic instabilities in human cancers. Nature 1998;
396:643-649.
2. Funk JO. Cancer cell cycle control. Anticancer Res 1999; 19:4772-4780.
3. Hartwell LH, Kastan MB. Cell cycle control and cancer. Science 1994; 266:1821-1828.
4. Hartwell LH, Weinert TA. Checkpoints: Controls that ensure the order of cell cycle events. Science
1989; 246:629-634.
5. Elledge SJ. Cell cycle checkpoints: preventing an identity crisis. Science 1996; 274:1664-1672.
6. Paulovich AG, Toczyski DP, Hartwell LH. When checkpoints fail. Cell 1997; 88:315-321.
7. Friedberg EC. Relationships between DNA repair and transcription. Annu Rev Biochem 1996;
65:15-42.
8. Rodriguez H, Akman SA, Holmquist GP et al. Mapping oxidative DNA damage using
ligation-mediated polymerase chain reaction technology. Methods 2000; 22:148-156.
9. Zhou BB, Elledge SJ. The DNA damage response: Putting checkpoints in perspective. Nature
2000; 408:433-439.
10. Hole EO, Sagstuen E, Nelson WH, Close DM. Free radical formation in X-irradiated crystals of
2'-deoxycytidine hydrochloride. Electron magnetic resonance studies at 10 K. Radiat Res 2000;
153:823-834.
11. Sutherland BM, Bennett PV, Sidorkina O, Laval J. Clustered damages and total lesions induced in
DNA by ionizing radiation: oxidized bases and strand breaks. Biochemistry 2000; 39:8026-8031.
12. Kastan MB. Signalling to p53: Where does it all start? Bioessays 1996; 18:617-619.
13. Rotman G, Shiloh Y. ATM: A mediator of multiple responses to genotoxic stress. Oncogene 1999;
18:6135-6144.
14. Tibbetts RS, Brumbaugh KM, Williams JM et al. A role for ATR in the DNA damage-induced
phosphorylation of p53. Genes Dev 1999; 13:152-157.
15. Brown EJ, Baltimore D. ATR disruption leads to chromosomal fragmentation and early embryonic
lethality. Genes Dev 2000; 14:397-402.
16. Savitsky K, Bar-Shira A, Gilad S et al. A single ataxia telangiectasia gene with a product similar to
PI-3 kinase. Science 1995; 268:1749-1753.
17. Kastan M. Ataxia-telangiectasia--broad implications for a rare disorder. N Engl J Med 1995;
333:662-663.
18. Lavin MF, Shiloh Y. The genetic defect in ataxia-telangiectasia. Annu Rev Immunol 1997;
15:177-202.
19. Barlow C, Hirotsune S, Paylor R et al. ATM-deficient mice: A paradigm of ataxia telangiectasia.
Cell 1996; 86:159-171.
20. Taylor AM. What has the cloning of the ATM gene told us about ataxia telangiectasia? Int J
Radiat Biol 1998; 73:365-371.
DNA Damage, Cell Cycle Control, and Cancer 75
21. Kastan MB, Zhan Q, el-Deiry WS et al. A mammalian cell cycle checkpoint pathway utilizing p53
and GADD45 is defective in ataxia-telangiectasia. Cell 1992; 71:587-597.
22. Paules RS, Levedakou EN, Wilson SJ et al. Defective G2 checkpoint function in cells from individuals
with familial cancer syndromes. Cancer Res 1995; 55:1763-1773.
23. Murnane JP, Schwartz JL. Cell checkpoint and radiosensitivity. Nature 1993; 365:22.
24. Murnane JP. Cell cycle regulation in response to DNA damage in mammalian cells: A historical
perspective [published erratum appears in Cancer Metastasis Rev 1995 Sep;14(3):253-4]. Cancer
Metastasis Rev 1995; 14:17-29.
25. Rotman G, Shiloh Y. ATM: from gene to function. Hum Mol Genet 1998; 7:1555-1563.
26. Levine AJ. p53, the cellular gatekeeper for growth and division. Cell 1997; 88:323-331.
27. Hansen R, Oren M. p53: From inductive signal to cellular effect. Curr Opin Genet Dev 1997;
7:46-51.
28. Maltzman W, Czyzyk L. UV irradiation stimulates levels of p53 cellular tumor antigen in
nontransformed mouse cells. Mol Cell Biol 1984; 4:1689-1694.
29. Price BD, Calderwood SK. Increased sequence-specific p53-DNA binding activity after DNA damage
is attenuated by phorbol esters. Oncogene 1993; 8:3055-3062.
30. Maki CG, Howley PM. Ubiquitination of p53 and p21 is differentially affected by ionizing and
UV radiation. Mol Cell Biol 1997; 17:355-363.
31. Chen X, Ko LJ, Jayaraman L, Prives C. p53 levels, functional domains, and DNA damage determine
the extent of the apoptotic response of tumor cells. Genes Dev 1996; 10:2438-2451.
32. Vousden KH. p53: Death star. Cell 2000; 103:691-694.
33. Jones SN, Sands AT, Hancock AR et al. The tumorigenic potential and cell growth characteristics
of p53-deficient cells are equivalent in the presence or absence of Mdm2. Proc Natl Acad Sci U S
A 1996; 93:14106-14111.
34. Livingstone LR, White A, Sprouse J et al. Altered cell cycle arrest and gene amplification potential
accompany loss of wild-type p53. Cell 1992; 70:923-935.
35. Chan TA, Hwang PM, Hermeking H, Kinzler KW, Vogelstein B. Cooperative effects of genes
controlling the G(2)/M checkpoint. Genes Dev 2000; 14:1584-1588.
36. Giaccia AJ, Kastan MB. The complexity of p53 modulation: Emerging patterns from divergent
signals. Genes Dev 1998; 12:2973-2983.
37. Haupt Y, Maya R, Kazaz A, Oren M. Mdm2 promotes the rapid degradation of p53. Nature
1997; 387:296-299.
38. Kubbutat MH, Jones SN, Vousden KH. Regulation of p53 stability by Mdm2. Nature 1997;
387:299-303.
39. Lane DP, Hall PA. MDM2—Arbiter of p53´2 destruction. Trends Biochem Sci 1997; 22:372-374.
40. Momand J, Zambetti GP, Olson DC, George D, Levine AJ. The mdm-2 oncogene product forms
a complex with the p53 protein and inhibits p53-mediated transactivation. Cell 1992; 69:1237-1245.
41. Oliner JD, Pietenpol JA, Thiagalingam S et al. Oncoprotein MDM2 conceals the activation
domain of tumour suppressor p53. Nature 1993; 362:857-860.
42. Thut CJ, Goodrich JA, Tjian R. Repression of p53-mediated transcription by MDM2: A dual
mechanism. Genes Dev 1997; 11:1974-1986.
43. Sherr CJ. Tumor surveillance via the ARF-p53 pathway. Genes Dev 1998; 12:2984-2991.
44. de Stanchina E, McCurrach ME, Zindy F et al. E1A signaling to p53 involves the p19(ARF)
tumor suppressor. Genes Dev 1998; 12:2434-2442.
45. Zindy F, Eischen CM, Randle DH et al. Myc signaling via the ARF tumor suppressor regulates
p53-dependent apoptosis and immortalization. Genes Dev 1998; 12:2424-2433.
46. Kamijo T, Weber JD, Zambetti G et al. Functional and physical interactions of the ARF tumor
suppressor with p53 and Mdm2. Proc Natl Acad Sci USA 1998; 95:8292-8297.
47. Pomerantz J, Schreiber-Agus N, Liégeois NJ et al. The Ink4a tumor suppressor gene product,
p19Arf, interacts with MDM2 and neutralizes MDM2‘s inhibition of p53. Cell 1998; 92:713-723.
48. Zhang Y, Xiong Y, Yarbrough WG. ARF promotes MDM2 degradation and stabilizes p53:
ARF-INK4a locus deletion impairs both the Rb and p53 tumor suppression pathways. Cell 1998;
92:725-734.
49. Stott FJ, Bates S, James MC et al. The alternative product from the human CDKN2A locus,
p14(ARF), participates in a regulatory feedback loop with p53 and MDM2. Embo J 1998;
17:5001-5014.
50. Kamijo T, Zindy F, Roussel MF et al. Tumor suppression at the mouse INK4a locus mediated by
the alternative reading frame product p19ARF. Cell 1997; 91:649-659.
51. Khan SH, Moritsugu J, Wahl GM. Differential requirement for p19ARF in the p53-dependent
arrest induced by DNA damage, microtubule disruption, and ribonucleotide depletion. Proc Natl
Acad Sci USA 2000; 97:3266-3271.
76 Cell Cycle Checkpoints and Cancer
52. Milne DM, Palmer RH, Campbell DG, Meek DW. Phosphorylation of the p53 tumour-suppressor
protein at three N-terminal sites by a novel casein kinase I-like enzyme. Oncogene 1992;
7:1361-1369.
53. Shieh SY, Ikeda M, Taya Y, Prives C. DNA damage-induced phosphorylation of p53 alleviates
inhibition by MDM2. Cell 1997; 91:325-334.
54. Shieh SY, Taya Y, Prives C. DNA damage-inducible phosphorylation of p53 at N-terminal sites
including a novel site, Ser20, requires tetramerization. Embo J 1999; 18:1815-1823.
55. Ko LJ, Shieh SY, Chen X et al. p53 is phosphorylated by CDK7-cyclin H in a p36MAT1-dependent
manner. Mol Cell Biol 1997; 17:7220-7229.
56. Lees-Miller SP, Chen YR, Anderson CW. Human cells contain a DNA-activated protein kinase
that phosphorylates simian virus 40 T antigen, mouse p53, and the human Ku autoantigen. Mol
Cell Biol 1990; 10:6472-6481.
57. Price BD, Hughes-Davies L, Park SJ. CDK2 kinase phosphorylates serine 315 of human p53 in
vitro. Oncogene 1995; 11:73-80.
58. Baudier J, Delphin C, Grunwald D, Khochbin S, Lawrence JJ. Characterization of the tumor
suppressor protein p53 as a protein kinase C substrate and a S100b-binding protein. Proc Natl
Acad Sci USA 1992; 89:11627-11631.
59. Hall SR, Campbell LE, Meek DW. Phosphorylation of p53 at the casein kinase II site selectively
regulates p53-dependent transcriptional repression but not transactivation. Nucleic Acids Res 1996;
24:1119-1126.
60. Sakaguchi K, Herrera JE, Saito S et al. DNA damage activates p53 through a phosphorylationacetylation
cascade. Genes Dev 1998; 12:2831-2841.
61. Banin S, Moyal L, Shieh S et al. Enhanced phosphorylation of p53 by ATM in response to DNA
damage. Science 1998; 281: 1674-1677.
62. Kapoor M, Hamm R, Yan W, Taya Y, Lozano G. Cooperative phosphorylation at multiple sites is
required to activate p53 in response to UV radiation. Oncogene 2000; 19:358-364.
63. Chehab NH, Malikzay A, Appel M, Halazonetis TD. Chk2/hCds1 functions as a DNA damage
checkpoint in G(1) by stabilizing p53. Genes Dev 2000; 14:278-288.
64. Matsuoka S, Huang M, Elledge SJ. Linkage of ATM to cell cycle regulation by the Chk2 protein
kinase. Science 1998; 282:1893-1897.
65. Shieh SY, Ahn J, Tamai K, Taya Y, Prives C. The human homologs of checkpoint kinases Chk1
and Cds1 (Chk2) phosphorylate p53 at multiple DNA damage-inducible sites [published erratum
appears in Genes Dev 2000 Mar 15;14(6):750]. Genes Dev 2000; 14:289-300.
66. Hupp TR, Meek DW, Midgley CA, Lane DP. Activation of the cryptic DNA binding function of
mutant forms of p53. Nucleic Acids Res 1993; 21:3167-3174.
67. Hupp TR, Lane DP. Two distinct signaling pathways activate the latent DNA binding function of
p53 in a casein kinase II-independent manner. J Biol Chem 1995; 270:18165-18174.
68. Sakaguchi K, Sakamoto H, Lewis MS et al. Phosphorylation of serine 392 stabilizes the tetramer
formation of tumor suppressor protein p53. Biochemistry 1997; 36:10117-10124.
69. Ashcroft M, Kubbutat MH, Vousden KH. Regulation of p53 function and stability by phosphorylation.
Mol Cell Biol 1999; 19:1751-1758.
70. Resnitzky D, Gossen M, Bujard H, Reed SI. Acceleration of the G1/S phase transition by expression
of cyclins D1 and E with an inducible system. Mol Cell Biol 1994; 14:1669-1679.
71. Ohtsubo M, Theodoras AM, Schumacher J, Roberts JM, Pagano M. Human cyclin E, a nuclear
protein essential for the G1 to S phase transition. Mol Cell Biol 1995; 15:2612-2624.
72. Reed SI. Cyclin E: In mid-cycle. Biochim Biophys Acta 1996; 1287:151-153.
73. Spruck CH, Won KA, Reed SI. Deregulated cyclin E induces chromosome instability. Nature
1999; 401:297-300.
74. Linke SP, Clarkin KC, Di Leonardo A, Tsou A, Wahl GM. A reversible, p53-dependent G0/G1
cell cycle arrest induced by ribonucleotide depletion in the absence of detectable DNA damage.
Genes Dev 1996; 10:934-947.
75. Waldman T, Kinzler KW, Vogelstein B. p21 is necessary for the p53-mediated G1 arrest in human
cancer cells. Cancer Res 1995; 55:5187-5190.
76. Brugarolas J, Chandrasekaran C, Gordon JI et al. Radiation-induced cell cycle arrest compromised
by p21 deficiency. Nature 1995; 377:552-557.
77. Sherr CJ, Roberts JM. Inhibitors of mammalian G1 cyclin-dependent kinases. Genes Dev 1995;
9:1149-1163.
78. Zhu L, Harlow E, Dynlacht BD. p107 uses a p21CIP1-related domain to bind cyclin/CDK2 and
regulate interactions with E2F. Genes Dev 1995; 9:1740-1752.
79. Chen J, Saha P, Kornbluth S, Dynlacht BD, Dutta A. Cyclin-binding motifs are essential for the
function of p21CIP1. Mol Cell Biol 1996; 16:4673-4682.
DNA Damage, Cell Cycle Control, and Cancer 77
80. Adams PD, Sellers WR, Sharma SK et al. Identification of a cyclin-CDK2 recognition motif present
in substrates and p21-like cyclin-dependent kinase inhibitors. Mol Cell Biol 1996; 16:6623-6633.
81. Funk JO, Galloway DA. Inhibiting CDK inhibitors: New lessons from DNA tumor viruses. Trends
Biochem Sci 1998; 23:337-341.
82. LaBaer J, Garrett MD, Stevenson LF et al. New functional activities for the p21 family of CDK
inhibitors. Genes Dev 1997; 11:847-862.
83. Waga S, Hannon GJ, Beach D, Stillman B. The p21 inhibitor of cyclin-dependent kinases controls
DNA replication by interaction with PCNA. Nature 1994; 369:574-578.
84. Li R, Waga S, Hannon GJ, Beach D, Stillman B. Differential effects by the p21 CDK inhibitor
on PCNA-dependent DNA replication and repair. Nature 1994; 371:534-537.
85. Ball KL, Lain S, Fåhraeus R, Smythe C, Lane DP. Cell-cycle arrest and inhibition of CDK4 activity
by small peptides based on the carboxy-terminal domain of p21WAF1. Curr Biol 1996; 7:71-80.
86. Funk JO, Waga S, Harry JB et al. Inhibition of CDK activity and PCNA-dependent DNA replication
by p21 is blocked by interaction with the HPV-16 E7 oncoprotein. Genes Dev 1997;
11:2090-2100.
87. Hermeking H, Funk JO, Reichert M, Ellwart JW, Eick D. Abrogation of p53-induced cell cycle
arrest by c-Myc: evidence for an inhibitor of p21WAF1/CIP1/SDI1. Oncogene 1995; 11:1409-1415.
88. Elledge SJ, Winston J, Harper JW. A question of balance: The role of cyclin-kinase inhibitors in
development and tumorigenesis. Trends Cell Biol 1996; 6:388-392.
89. Dulic V, Stein GH, Far DF, Reed SI. Nuclear accumulation of p21CIP1 at the onset of mitosis: a
role at the G2/M-phase transition. Mol Cell Biol 1998; 18:546-557.
90. Bunz F, Dutriaux A, Lengauer C et al. Requirement for p53 and p21 to sustain G2 arrest after
DNA damage. Science 1998; 282:1497-1501.
91. Niculescu AB, 3rd, Chen X, Smeets M et al. Effects of p21(Cip1/Waf1) at both the G1/S and the
G2/M cell cycle transitions: pRb is a critical determinant in blocking DNA replication and in
preventing endoreduplication [published erratum appears in Mol Cell Biol 1998 Mar;18(3):1763].
Mol Cell Biol 1998; 18:629-643.
92. Winters ZE, Ongkeko WM, Harris AL, Norbury CJ. p53 regulates Cdc2 independently of inhibitory
phosphorylation to reinforce radiation-induced G2 arrest in human cells. Oncogene 1998;
17:673-684.
93. Fukasawa K, Choi T, Kuriyama R, Rulong S, Vande Woude GF. Abnormal centrosome amplification
in the absence of p53. Science 1996; 271:1744-1747.
94. Mantel C, Braun SE, Reid S et al. p21(cip-1/waf-1) deficiency causes deformed nuclear architecture,
centriole overduplication, polyploidy, and relaxed microtubule damage checkpoints in human
hematopoietic cells. Blood 1999; 93:1390-1398.
95. Waldman T, Lengauer C, Kinzler KW, Vogelstein B. Uncoupling of S phase and mitosis induced
by anticancer agents in cells lacking p21 [see comments]. Nature 1996; 381:713-716.
96. Meraldi P, Lukas J, Fry AM, Bartek J, Nigg EA. Centrosome duplication in mammalian somatic
cells requires E2F and CDK2-cyclin A. Nat Ce Biol 1999; 1:88-93.
97. Chan TA, Hermeking H, Lengauer C, Kinzler KW, Vogelstein B. 14-3-3Sigma is required to
prevent mitotic catastrophe after DNA damage [see comments]. Nature 1999; 401: 616-620.
98. Hagting A, Karlsson C, Clute P, Jackman M, Pines J. MPF localization is controlled by nuclear
export. Embo J 1998; 17:4127-4138.
99. Toyoshima F, Moriguchi T, Wada A, Fukuda M, Nishida E. Nuclear export of cyclin B1 and its
possible role in the DNA damage-induced G2 checkpoint. Embo J 1998; 17:2728-2735.
100. Parker LL, Sylvestre PJ, Byrnes MJ, III, Liu F, Piwnica-Worms H. Identification of a 95-kDa
WEE1-like tyrosine kinase in HeLa cells. Proc Natl Acad Sci U S A 1995; 92:9638-9642.
101. Mueller PR, Coleman TR, Kumagai A, Dunphy WG. Myt1: A membrane-associated inhibitory
kinase that phosphorylates Cdc2 on both threonine-14 and tyrosine-15. Science 1995; 270:86-90.
102. Talizawa CG, Morgan DO. Control of mitosis by changes in the subcellular location of cyclin
B1-CDK1 and Cdc25C. Curr Op Cell Biol 2000; 12:658-665.
103. McShea A, Samuel T, Eppel JT, Galloway DA, Funk JO. Identification of CIP-1-associated regulator
of cyclin B (CARB), a novel p21-binding protein acting in the G2 phase of the cell cycle. J
Biol Chem 2000; 275:23181-23186.
104. Ohi R, Gould KL. Regulating the onset of mitosis. Curr Opin Cell Biol 1999; 11:267-273.
105. Clute P, Pines J. Temporal and spatial control of cyclin B1 destruction in metaphase. Nat Cell
Biol 1999; 1:82-87.
106. Pines J, Hunter T. The differential localization of human cyclins A and B is due to a cytoplasmic
retention signal in cyclin B. Embo J 1994; 13: 3772-3781.
107. Harper JW, Elledge SJ. CDK inhibitors in development and cancer. Curr Op Genet Dev 1996;
6:56-64.
78 Cell Cycle Checkpoints and Cancer
108. Sherr CJ. Cancer cell cycles. Science 1996; 274:1672-1677.
109. Strausfeld U, Labbe JC, Fesquet D et al. Dephosphorylation and activation of a p34cdc2/cyclin B
complex in vitro by human Cdc25 protein. Nature 1991; 351:242-245.
110. Atherton-Fessler S, Parker LL, Geahlen RL, Piwnica-Worms H. Mechanisms of p34cdc2 regulation.
Mol Cell Biol 1993; 13:1675-1685.
111. Furnari B, Blasina A, Boddy MN, McGowan CH, Russell P. Cdc25 inhibited in vivo and in vitro
by checkpoint kinases Cds1 and Chk1. Mol Biol Cell 1999; 10:833-845.
112. Brown AL, Lee CH, Schwarz JK et al. A human Cds1-related kinase that functions downstream of
ATM protein in the cellular response to DNA damage. Proc Natl Acad Sci USA 1999; 96:3745-3750.
113. Peng CY, Graves PR, Ogg S et al. C-TAK1 protein kinase phosphorylates human Cdc25C on
serine 216 and promotes 14-3-3 protein binding. Cell Growth Differ 1998; 9:197-208.
114. Hoffmann I, Clarke PR, Marcote MJ, Karsenti E, Draetta G. Phosphorylation and activation of
human Cdc25-C by cdc2--cyclin B and its involvement in the self-amplification of MPF at mitosis.
Embo J 1993; 12:53-63.
115. Kumagai A, Dunphy WG. Purification and molecular cloning of Plx1, a Cdc25-regulatory kinase
from Xenopus egg extracts. Science 1996; 273:1377-1380.
116. Peng CY, Graves PR, Thoma RS et al. Mitotic and G2 checkpoint control: regulation of 14-3-3
protein binding by phosphorylation of Cdc25C on serine-216. Science 1997; 277:1501-1505.
117. Kumagai A, Yakowec PS, Dunphy WG. 14-3-3 proteins act as negative regulators of the mitotic
inducer Cdc25 in Xenopus egg extracts. Mol Biol Cell 1998; 9:345-354.
118. Poon RY, Chau MS, Yamashita K, Hunter T. The role of Cdc2 feedback loop control in the
DNA damage checkpoint in mammalian cells. Cancer Res 1997; 57:5168-5178.
119. Herzinger T, Funk JO, Hillmer K et al. Ultraviolet B irradiation-induced G2 cell cycle arrest in
human keratinocytes by inhibitory phosphorylation of the cdc2 cell cycle kinase. Oncogene 1995;
11:2051-2056.
120. Hermeking H, Lengauer C, Polyak K et al. 14-3-3 sigma is a p53-regulated inhibitor of G2/M
progression. Mol Cell 1997; 1:3-11.
121. Smits VAJ, Klompmaker R, Vallenius T et al. p21 inhibits Thr161 phosphorylation of cdc2 to
enforce the G2 DNA damage checkpoint. J Biol Chem 2000; 275:30638-30643.
122. Polyak K, Xia Y, Zweier JL, Kinzler KW, Vogelstein B. A model for p53-induced apoptosis. Nature
1997; 389:300-305.
DNA Damage-Independent Checkpoints from Yeast to Man 79
CHAPTER 5
Cell Cycle Checkpoints and Cancer, edited by Mikhail V. Blagosklonny. ©2001 Eurekah.com.
DNA-Damage-Independent Checkpoints
from Yeast to Man
Duncan J. Clarke, Adrian P.L. Smith and Juan F. Giménez-Abián
Introduction
Checkpoints are mechanisms that establish dependence relationships between biochemically
unrelated cellular processes. The temporal order of many critical cell cycle events must be
strictly maintained to ensure cell survival and integrity. A simple example is that of genome
duplication which must be completed before cell division. The relationship between these
processes is controlled by the S-phase checkpoint. After S-phase, the Topoisomerase II-dependent
checkpoint ensures that the topology of the newly replicated DNA has been correctly
organized before cells begin mitosis. During mitosis itself, distinct checkpoints monitor mitotic
spindle assembly, preventing the onset of chromosome segregation until all the chromosomes
are correctly aligned on the mitotic spindle, and prevent exit from mitosis until anaphase
chromosome segregation has been completed. In this Chapter, we discuss these checkpoint
control systems (Fig. 1).
An elegant way to define checkpoint pathways has been by analysis of loss-of-function
mutants in the genetically manipulable yeast systems. For example, the S-phase checkpoint was
described in budding yeast by the isolation of mutants that initiate mitosis despite a replication
block enforced by the ribonucleotide reductase inhibitor hydroxyurea (HU). Other checkpoint
systems were originally described in mammalian cells. Indeed, the existence of checkpoint
controls had been inferred from mammalian cell-fusion studies earlier than the genetic
analyses performed in yeast (Fig. 2).1 More recently, mammalian checkpoints have been
investigated by demonstrating that some checkpoint arrests can be overridden by drugs such as
caffeine.2-9 Caffeine inhibits the kinase ATM, a key component of eukaryotic checkpoint pathways.
10-13 Although mammalian cells are less amenable to genetic studies than are yeast, they
have proven to be important for the study of checkpoint biology. Our descriptions of checkpoint
controls use a commonly adopted format which divides the pathways into (1) the sensor,
(2) the transducer and (3) the target. Checkpoints do not necessarily follow simple linear pathways,
however. Many of the sensor and transducer components are likely to be assembled into
large complexes. Still, this nomenclature allows for a framework to be drawn up, on which the
details can be built. The sensor components are those that monitor completion of the relevant
process, for example DNA replication. The transducer transmits the signal from the sensor to
the target of the checkpoint. It is the activity of the target that controls cell cycle progression.
Budding Yeast versus Higher Eukaryotes
Since checkpoint pathways in both budding yeast and higher eukaryotes will be discussed
in this Chapter, it is important to describe a fundamental difference in budding yeast cell cycle
organization that sets it apart from other species. In budding yeast, checkpoints promote the
activity of anaphase inhibitors, whereas in most eukaryotes, checkpoints inhibit the activity of
80 Cell Cycle Checkpoints and Cancer
the mitotic kinase (cyclin/CDK), required for passage through the G2/M transition. A need
for distinct modes of control is related to differences in the spindle assembly pathway. In many
eukaryotes, including mammals, the mitotic spindle does not assemble until mitosis. However,
Fig. 1. DNA-damage-independent checkpoints. A summary of the five checkpoints that are discussed in
detail (see magnifying glass). In mammals the G2/M transition is regulated by at least 2 non-DNA damage
checkpoint pathways (1 and 2). These prevent the initiation of mitosis until DNA replication is complete
(1, S-phase checkpoint), and until DNA is sufficiently decatenated (2, Topoisomerase II-dependent checkpoint).
The Chfr checkpoint (3) appears to restrict chromosome condensation when spindle assembly is
perturbed. Also within mitosis the spindle assembly checkpoint (4) delays the onset of anaphase until all
the chromosomes of the karyotype have been correctly arranged on the mitotic spindle. In budding yeast,
similar checkpoint controls inhibit the onset of anaphase rather than preventing passage beyond the G2/
M transition. Therefore the G2/M transition in mammals and the metaphase-anaphase transition in budding
yeast are somewhat analogous. Indeed, sensor and signaling components of these checkpoint pathways
are conserved. However, the checkpoint targets differ between mammals and budding yeast, and the
topoisomerase II-dependent checkpoint is not functional in budding yeast. Exit from mitosis (5) is also
under checkpoint control, to ensure that anaphase has been completed before cell division. Photomicrographs
of onion root meristematic cells depict the mitotic stages. The cartoons compare cell cycle stages in
budding yeast and mammals.
DNA Damage-Independent Checkpoints from Yeast to Man 81
budding yeast spindles assemble during S-phase; checkpoints must inhibit spindle elongation even
while DNA is being replicated. In addition, sister chromatid cohesion, established during DNA
replication, must be maintained until the onset of anaphase. In budding yeast, an inhibitor of
anaphase, Pds1, can perform both of these tasks.14-18 Before anaphase, Pds1 binds to protease
Esp1 and thereby inhibits the anaphase-promoting activity of Esp1.18 During an
unperturbed cell cycle, Pds1 becomes poly-ubiquitinated at the metaphase to anaphase transition
by a multi-subunit enzyme complex known as the APC (Anaphase Promoting Complex); the
modified forms are recognized and degraded by 26S proteasomes.16 Once released from Pds1,
Esp1 induces cleavage of Scc1, a cohesin required to maintain cohesion between sister chromatids.
18-21 Concurrently with loss of sister cohesion, Esp1 induces spindle elongation.22 Not
surprisingly, Pds1 is a major target of checkpoints controlling anaphase onset. Vertebrate
proteins named securins, that are at least partial functional homologues of Pds1, have been
identified,23 making the study of checkpoints in budding yeast highly relevant. Moreover,
budding yeast and higher eukaryotes employ a common strategy for controlling exit from
mitosis. In this case, regulation of cyclin/CDK activity appears to be a universally adopted
mode of control.
S-Phase Checkpoint
The S-phase checkpoint ensures that the onset of mitosis is dependent on the completion
of DNA replication.24-26 Since little is known about S-phase checkpoint control in mammals, the
components of this pathway in budding yeast will be described (Fig. 3). As mentioned above,
budding yeast cells initiate DNA replication and mitotic spindle formation at a common cell cycle
point, early in S-phase. To prevent the generation of aneuploid daughter cells that are inviable, it is
essential that the mitotic spindle does not elongate before DNA replication has been completed.
The order of these two processes is normally maintained by a timing mechanism rather than a
checkpoint control: DNA replication takes only 20-30 minutes and spindle formation takes
around 60 minutes; thus, spindle assembly is not completed before genome replication. This
example illustrates how the temporal order of two events can be maintained independently of
checkpoint controls, i.e., if the processes have a common starting point and each require a
differing minimum amount of time for their completion.
Fig. 2. Premature chromosome condensation (PCC). Interphase cells can be fused with mitotic cells by
Sendai virus treatment. This induces PCC, suggesting that interphase checkpoints might normally inhibit
the onset of chromosome condensation. Photomicrographs show fusions of S—phase/G2 with M—phase
Muntiacus muntjak cells.
82 Cell Cycle Checkpoints and Cancer
A checkpoint pathway does exist, however, to ensure that the dependence between spindle
elongation and DNA replication is always maintained. If DNA replication is inhibited with
HU,27 the cells arrest with fully assembled short G2 spindles. After removal of the HU, spindle
elongation is delayed until replication is complete. The S-phase checkpoint does not only control
the mitotic spindle, however. All eukaryotes establish sister cohesion during DNA replication, and
it must be maintained until the onset of anaphase.20,21 Maintenance of sister chromatid cohesion
is of great importance to mammals and yeast alike, and is a function of the S-phase checkpoint.
At least in yeast, cohesion is established at some loci very early in S-phase and must therefore be
maintained for the remainder of the S-phase period as well as during G2 and until the moment
of anaphase onset. The homologs of yeast S-phase checkpoint components are therefore likely
to be important regulators of mammalian sister chromatid cohesion.
To define the budding yeast S-phase checkpoint, loss-of-function mutations causing
sensitivity to HU were identified. The proteins encoded by these genes were determined to
have S-phase checkpoint functions by showing that the loss-of-function mutations allowed
entry into mitosis when DNA replication was blocked with HU. Thus the S-phase checkpoint
is defined as that which restrains entry into mitosis when replication is blocked. However,
kinetic analyses of various checkpoint mutants, grown in the presence of a concentration of
HU that allows replication to proceed, but more slowly than in an unperturbed cell cycle, have
revealed genetically distinct S-phase checkpoint systems (see below).
To monitor ongoing DNA replication, cells seem to have replication sensors that reside at
replication forks. In budding yeast, the putative sensor components include Pol2, Rfc5, Dpb11,
Drc1 and Sgs1 (Refs. 28-32). POL2 encodes the replicative DNA polymerase, Polε and Rfc5 is
a replication factor C subunit involved in recruiting polymerases to replication forks. Dpb11 is
Fig. 3. S-phase checkpoint. Linear pathways are drawn for simplicity though more complex interactions
between the checkpoint components are likely. In budding yeast, a signal generated by replication forks is
transmitted by kinases Mec1 and Rad53. Rad53 activation is dependent on the Mec1-Ddc2 complex.
Downstream of Rad53 and Mec1, sequential pathways operate: a Mec1-Rad53 pathway is required early
in S phase, and a Mec1-Pds1 pathway is required part way through S phase. The Mec1-Rad53 pathway also
induces a Dun1-dependent transcriptional response which protects cells from replicative stress. The
mammalian pathway drawn on the right is speculative, based mainly on HU-induced phosphorylation
events and HU-induced localization of proteins at replication foci.
DNA Damage-Independent Checkpoints from Yeast to Man 83
also required for DNA replication; it can bind to Polε and is thought to help recruit Polα-
primase complexes to ARS sequences at replication origins.33 Dpb11 also binds to Drc1, which
is itself essential for DNA replication.31 Together with Srs2, Sgs1 has a redundant but essential
role in DNA replication.34
Evidently, the sensor proteins also play important roles in DNA replication itself, and
with hindsight it might seem elementary that components of the replication fork machinery
are involved in generating the checkpoint signal. For each of these components, it was important
to know that their checkpoint functions could be distinguished from their roles in DNA
replication. This is the case because the S-phase checkpoint cannot be activated until DNA
replication has begun,35 a fact illustrated by the phenotype of cells carrying a heat-inducible
cdc45 degron mutant.36 Cdc45 binds to replication origins before S-phase in budding yeast
and is required for origin firing. The cdc45 degron mutant is rapidly degraded at the restrictive
temperature. When degradation was induced in G1 of the cell cycle, replication origins could
not fire and the cells entered mitosis without replicating any DNA. When the temperature
shift was performed within S-phase, DNA replication was immediately inhibited because Cdc45
is also needed for elongation of replication forks, but in this case mitotic progression was inhibited.
Thus, the S-phase checkpoint signal requires the presence of replication forks that have fired,
and Cdc45 is not a component of the checkpoint response.
Mec1 and Rad53 kinases are traditionally described as components of the signal transduction
element of the S-phase checkpoint.24,25,37 When replication is perturbed, these checkpoint
kinases are activated in a manner dependent on the sensor components. Exactly how Rad53
and Mec1 activation occurs is not known, but several physical interactions have been identified
that may represent key steps. Sgs1 was found to co-localize with Rad53 in discrete nuclear foci
during S-phase. Intriguingly, Sgs1 is reported to interact with the FHA domain of Rad53,38 the
domain required for the formation of the Rad53-Rad9 complex, required for DNA damage checkpoint
signaling.39 Rad53-Sgs1 association may have revealed an Sgs1-dependent loading of
Rad53 onto specific chromatin regions that might be involved in monitoring replication. That
Sgs1 is involved in S-phase checkpoint control,32 adds weight to this model. But, the S-phase
checkpoint defect of sgs1 mutants is rather weak, not nearly as substantial as other S-phase
checkpoint mutants. This suggests that Sgs1 has a redundant checkpoint function, perhaps
with another helicase such as Srs2.
Another activator of signal transduction may be Ddc2 (also known as Lcd1), a component
that physically associates with Mec1 and is phosphorylated by Mec1.40,41 Phosphorylated Ddc2
is present in unperturbed S-phase cells and in HU-treated cells, and Ddc2 is required for cell
cycle arrest in the presence of HU. Phosphorylation and activation of Rad53 in response to
replication arrest is Ddc2-dependent. Therefore, Ddc2 appears to mediate between Mec1 and
Rad53 in response to ongoing DNA replication and when fork progression is blocked. It
remains to be tested whether the association of Ddc2 with Mec1, or the phosphorylation of
Ddc2 by Mec1, depends on the sensor components, and how these events are regulated.
In response to replication inhibition, Mec1 and Rad53 enforce cell cycle arrest partly by
blocking Pds1 degradation. In addition, these kinases induce transcription of genes involved in
DNA repair and that help deal with the perturbed replication process.24 This safety system
most likely protects stalled replication forks, allowing them to be reinitiated when conditions
have improved. The transcription pathway depends on Rad53-dependent phosphorylation of
the kinase Dun1 (damage uninducible).24,42 Activation of Dun1, in response to DNA damage
or DNA replication blocks, induces transcription of genes that promote efficient DNA repair.
24 This transcription response is partially initiated by Crt1 hyperphosphorylation.42 Crt1
represses transcription of DNA damage-inducible genes by binding to their promoter regions;
binding is prevented by hyperphosphorylation. Activation of a transcription program clearly
has the potential to enforce a wide range of Mec1-and Rad53-dependent functions, and a
growing literature has made clear that Mec1 and Rad53 are involved in numerous cellular
processes. For example, Mec1 and Rad53 inhibit the firing of late replication origins during
early S-phase. Eukaryotic cells replicate their genomes by initiating DNA synthesis from mul84
Cell Cycle Checkpoints and Cancer
tiple replication origins. Some fire early in S-phase, others are initiated part way through Sphase.
When cells are arrested in early S-phase with HU, late firing origins are kept dormant by
a dominant process that requires the Rad53-Mec1 pathway.43 Mec1 and Rad53 are also involved
in the regulation of telomere length and in transcriptional silencing at telomeres.44-46
It is not clear, however, whether the Mec1/Rad53/Dun1-dependent transcriptional
response contributes to cell cycle arrest in the presence of HU, since dun1 null mutants are not
S-phase checkpoint defective. In agreement with this, the cell cycle checkpoint defects of mec1
and rad53 mutants are somewhat different. Both mutants elongate their mitotic spindles when
DNA replication is blocked with HU, so it seems that no checkpoint response remains in these
cells. However, rad53 mutants delay in anaphase, while mec1 mutants exit mitosis. Thus some
aspects of mitotic progression are inhibited in rad53 mutants. Light was shed on the basis of
this difference by analysis of pds1 mutants, revealing that there are several S-phase checkpoint
targets. Pds1 is not an essential target in early S-phase because pds1 mutants can inhibit spindle
elongation when replication is blocked with HU in early S. However, kinetic analyses determined
that, part-way through S-phase, a critical point is reached where Pds1 becomes essential:
pds1 mutants elongate spindles and lose sister chromatid cohesion when roughly 2/3 of the
genome has been replicated.47 These experiments were performed in the presence of a
concentration of HU that does not fully block replication, but instead, slows the rate of DNA
replication. In these experiments, mec1 or rad53 mutant cells began anaphase when very little
DNA had been replicated (as is the case when replication is blocked). Therefore, a Pds1-
independent system restrains spindle elongation in early S-phase, but later in S-phase, Pds1 is
required. A reasonable prediction is that Pds1 and Rad53 function downstream of Mec1 in the
context of S-phase checkpoint control, and that these pathways run in parallel, and are temporally
regulated; one necessary in early S-phase, the other part way through S-phase. Presumably,
Mec1 controls Pds1 stability in late S-phase. Several details remain unresolved, however. For
example, the fact that pds1 null mutants are able to restrain spindle elongation and prevent
premature loss of sister chromatid cohesion in early S-phase necessitates a novel Mec1/Rad53
target at that point in the cell cycle.
An explanation for the duality of S-phase checkpoint control in budding yeast is the
linkage of spindle elongation with regulation of sister cohesion. Sister cohesion is established
during DNA replication, and must be maintained until the onset of anaphase.20,21 Once cohesion
is established, part way through S-phase,48 checkpoint control of anaphase must coordinate release
of cohesion with spindle elongation. Early in S-phase, prior to replication of critical cohesion
regions and concomitant establishment of cohesion, spindle elongation might be regulated
independently of cohesion. Therefore, the switch in the mode of checkpoint control from the
Mec1-Rad53 pathway to the Mec1-Pds1 pathway may be controlled by the establishment of
sister chromatid cohesion.
It remains to be determined how Pds1 levels are controlled in late S-phase when DNA
replication is perturbed. Recent evidence has linked two yeast genes to regulation of Pds1 in
this context.49 Rad23 or Ddi1 overproduction was found to rescue the sensitivity of pds1 mutant
cells to HU. Rad23 is a nucleotide excision repair protein, but recent studies suggest a novel
role of Rad23 in ubiquitin-dependent proteolysis.50 Rad23 binds to mono- or
di-ubiquitinated proteins, but cannot bind when the ubiquitin chains have been elongated.
Crucially, Rad23 blocks extension of the ubiquitin chains. For most ubiquitinated proteins
that are targeted for degradation, ubiquitin chain elongation is critical for efficient recognition by
the 26S proteasome. Therefore, Rad23 might have an important function in preventing or delaying
the degradation of proteasome targets. Although this mechanism has not been tested directly
in the context of S-phase checkpoint control, overexpression of Rad23 is able to stabilize a
mutant pds1 protein, suggesting that Rad23 might play a role in S-phase checkpoint signaling.
49 A role of Rad23 in checkpoint signaling may be utilized by virally expressed proteins.
The HIV-1 encoded protein Vpr has been shown to bind to the C-terminal UBA (Ubiquitin
associated) domain of human Rad23 (HHR23A).51-53 This interaction is needed for one of the
DNA Damage-Independent Checkpoints from Yeast to Man 85
cellular functions of Vpr – the ability of Vpr to induce G2 cell cycle arrest which allows time for
viral replication. It seems plausible that Vpr mimics an endogenous cellular checkpoint response
that involves binding of the Rad23 UBA to an unknown protein, inducing G2 arrest.
Mammalian cells also need an S-phase checkpoint. The initiation of mitosis must be
prevented during S-phase and chromatid cohesion must be maintained. Mammalian Sgs1
homologs are clearly important for S-phase regulation.38 Sgs1 is a budding yeast member of the
Escherichia coli recQ helicase family, and sgs1 mutants are genomically unstable.54
Mammalian recQ helicase family members include WRN (mutated in Werner’s syndrome
patients)55 and BLM (mutated in Bloom’s syndrome patients).56 Bloom’s syndrome is
characterized by genomic instability and a high incidence of cancer, whereas Werner’s
syndrome causes premature ageing. The BLM protein was recently identified as a component of a
large complex that includes tumor suppressor proteins, DNA repair and checkpoint
proteins that localize to nuclear foci when cells are treated with HU.57 Indeed, cultured cells from
Bloom’s syndrome patients have S-phase defects, but it is not clear whether these abnormalities
include checkpoint abrogation. In general, there are mammalian homologs of all the budding
yeast checkpoint proteins, but their potential roles in S-phase checkpoint control have not
been thoroughly investigated.
The Mec1 homologs are ATM and ATR.58,59 ATM, the gene mutated in ataxia telangiectasia
patients who have an increased incidence of cancer, is a nuclear protein kinase. ATR
(ataxia telangiectasia and rad3 related) is also a protein kinase, and is structurally more
homologous to Mec1, than is ATM. Both ATM and ATR are clearly involved in DNA damage
checkpoint signaling,60,61 but do not seem to be required for preventing the onset of mitosis
during S-phase. Whether these proteins have roles in regulating cohesion has not been
addressed. In the context of the DNA damage checkpoint, the tumor suppressor proteins p53
and BRCA1 (breast cancer gene 1) seem to be targets of ATM/ATR, but again there is little
evidence for roles in S-phase.62-68 BRCA1 is, however, phosphorylated by ATR in response to
HU treatment.69
The mammalian Rad53 homolog, kinase Chk2 (Checkpoint kinase),70 and the
mammalian homolog of budding and fission yeast Chk1 (also named Chk1), are required for the
ATM-dependent DNA damage checkpoint.71,72 After γ-irradiation, mammalian Chk2
phosphorylation (on thr-68, within the serine/threonine cluster domain of Chk2) and activation
is ATM-dependent.73 Chk2 kinase also becomes phosphorylated and activated upon HU
treatment, but in an ATM-independent manner, and not on thr-68 (Refs.70,73,74) If the
HU-induced phosphorylation is relevant for S-phase checkpoint control, it might be that the
mammalian S-phase checkpoint operates by a kinase distinct from ATM.
Chk1 is not needed for S-phase checkpoint control in budding yeast, but does seem to be
required in some higher eukaryotes. Xenopus Chk1 is activated in post-MBT (mid-blastula
transition) embryonic cells treated with HU.75 Similarly, there is good evidence for a role of
Drosophila Chk1 (named Grapes) in coordinating embryonic DNA replication with the onset
of mitosis.76,77 In Xenopus egg extracts, immunodepletion of Chk1 impairs an ability to delay
cell cycle progression in response to replication blocks.78 Immunodepletion of ATR has the
same effect since Chk1 activity depends on phosphorylation by ATR when unreplicated DNA
is present.79 Human ATR has been implicated in Chk1 phosphorylation in response to HU treatment,
but it is not known if cells lacking Chk1 or ATR have defective S-phase checkpoint controls.
80 In Xenopus, Chk1 activation also depends on a protein named Claspin which has a close
human homolog. Xenopus egg extracts depleted of Claspin are S-phase checkpoint deficient.81
Both Chk1 and Chk2 can phosphorylate Cdc25C on ser-216 in humans and it is thought
that this phosphorylation prevents Cdc25C from activating the mitotic kinase, cyclinB1/
Cdc2.70,71 Cdc25C is a protein phosphatase that promotes entry into mitosis by dephosphorylating
Cdc2. This phosphorylated residue creates a binding site for a 14-3-3 protein, resulting
in Cdc25C inhibition.72 Interestingly, expression of a mutant Cdc25C that cannot be phosphorylated
on ser-216 induces mitosis in the presence of unreplicated DNA, suggesting that
this pathways may be important for S-phase checkpoint control.72
86 Cell Cycle Checkpoints and Cancer
Topoisomerase II-Dependent Checkpoint
Topoisomerase II (topo II) is required for chromosome condensation and segregation in
eukaryotes.3,82-87 Although these are mitotic processes, their successful completion depends
partly on topo II activity during DNA replication and in G2 phase (Fig. 4). Chromosome
replication creates two identical sister DNA molecules that are knotted together (catenated).
Topo II removes the catenations; in higher eukaryotes, the majority must be resolved before
entry into mitosis to allow accurate chromosome condensation.3,88 A G2 checkpoint ensures
that DNA catenations have been sufficiently resolved before mammalian cells enter mitosis.3
It had been known for some time that topo II inhibitors block or delay mammalian cell cycle
progression in G2, but the inhibitors used were also known to cause DNA damage; it was
assumed that the G2 arrest was due to activation of the DNA damage checkpoint. That cells
also need to monitor topo II activity in G2 was an attractive hypothesis, however, and the
characterization of novel topo II inhibitors that do not damage DNA allowed this theory to be
tested. A variety of assays that measured DNA damage were used to assess the effects of
bisdioxopiperazine topo II inhibitors on mammalian cells and on isolated DNA.
Bisdioxopiperazine were found not to induce DNA breaks in vivo or in vitro (Ref. 3 and refs.
therein), but did block mammalian cells in G2. In these studies, entry into mitosis was assessed
based on the onset of chromosome condensation. Yet, topo II is needed for chromosome condensation,
albeit a requirement at late steps in the process.3,88 It was therefore necessary to
prove that cells were physically capable of initiating chromosome condensation when topo II
activity was absent. Such evidence could not be sought by isolating loss-of-function budding or
fission yeast mutants, since the topo II-dependent checkpoint seems to be absent in yeast.85,89
Therefore the checkpoint was first described in mammals, by demonstrating checkpoint bypass
induced with caffeine or kinase and phosphatase inhibitors.3 Cells treated with ICRF-
193, the most potent of the bisdioxopiperazines, could be forced into mitosis with caffeine; the
cells began to condense chromosomes without delay. Fully condensed chromosomes were not
formed, consistent with the essential role of topo II late in the condensation process. Thus, the
topo II-dependent checkpoint prevents the onset of chromosome condensation, a process which
the cells can begin, but cannot complete in the absence of topo II activity. Although the
evidence for this checkpoint in mammals is substantial, its absence in yeast has prevented the
checkpoint components from being rapidly identified. Indeed, it is not known whether topo II
levels are sensed directly, or if physical structures within chromosomes, such as catenations
(Fig. 5), are monitored. Some data suggest the latter is more likely. Replicative catenations are
introduced between daughter DNA duplexes during S-phase. Disentangling daughter duplexes
is of crucial importance since they otherwise could not separate and segregate during mitosis.
Most replicative catenations are resolved in G2, when cellular topo II activity increases, but
since the decatenation reaction is reversible, topo II activity inevitably promotes some catenation.
This generates nonreplicative catenations, that can involve distant regions of chromatin, and
can join different chromosomes together.90,91 Cells inhibited for topo II activity late in G2 and
forced to enter mitosis with caffeine, have striking chromosome aberrations caused by persistent
nonreplicative catenations that join chromosomes together and create ο figures within
individual chromosomes (Figs. 4 and 5).92 Circumstantial evidence suggests that the
removal of nonreplicative catenations in G2, the process that promotes chromosome individualization,
may be monitored by the topo II-dependent checkpoint.92
Another question is what is the nature of the topo II-dependent checkpoint sensor? It might be
that sensors bind to sites of DNA catenation. Though purely speculative, there is a precedent
for such a proposal, that a signaling cascade might be activated by protein complexes at sites of
DNA crossover. In bacteria, stable maintenance of the natural multicopy plasmid CoIE1
requires a cer sequence element (Fig. 6). cer is necessary for recombination that converts unstable
plasmid multimers to monomers.93,94 The expression of Rcd, a transcript encoded from
within cer, is specifically expressed in cells containing multimers. Rcd1 enforces a cell cycle
checkpoint that inhibits cell division when multimers are present,95 thus allowing time for siteDNA
Damage-Independent Checkpoints from Yeast to Man 87
specific recombination to occur. An interesting observation is that the Rcd1 promoter resides
within the cer sequence, and that the topology of cer is likely to be altered in multimers that
assemble recombination complexes at the cer sites.96 This difference in topology might influence
Rcd1 transcription, providing an elegant mechanism to activate the checkpoint in the
presence of multimers. The parallel between this phenomenon in bacteria, and the topo
II-dependent checkpoint signal generated by persistent DNA catenations in mammalian cells
is remarkable.
The target of the topo II-dependent checkpoint is presumed to be mitotic cyclin/CDK
activity: recent work has shown that a topo II-dependent checkpoint exists in plant cells that
can be overridden by overexpressing a mitotic cyclin (JFGA, unpublished data). How the
mitotic kinase is regulated in response to topo II-dependent checkpoint activation is not known,
although some data give clues as to the upstream components of the pathway. The topo II
inhibitor genistein arrests mammalian cells in G2 by activating Chk2 kinase, which in turn
leads to the inhibition of Cdc25C-dependent tyr-15 dephosphorylation of CDK1.97 Activation
of Chk2 occurs very efficiently at genistein doses that inhibit topo II but cause minimal
DNA damage compared to other topo II inhibitors such as etoposide. In the case of genistein,
ATR might be the kinase upstream of Chk2, since caffeine overrides the G2 arrest whereas
Wartmannin does not. In contrast to ATM, which is inhibited by caffeine and Wartmannin,
ATR is only efficiently inhibited by caffeine.97 The topo II-dependent checkpoint and DNA
damage checkpoint might be regulated primarily by ATR and ATM respectively. Although
these checkpoints are distinct, the possibility remains that they are closely linked pathways.
One way to address this issue will be to test whether other components of the DNA damage
checkpoint, such as Chk1, p53 and 14-3-3, are activated in the context of ICRF-193-induced
G2 arrest.
Checkpoint Control in Prophase
A recent study identified a novel mammalian checkpoint protein Chfr (Checkpoint with
FHA and Ring finger) apparently acting to slow chromosomal condensation in prophase and
prometaphase when microtubule polymerization is perturbed.98 In a cohort of 8 human tumor
cell lines, three were identified that failed to express Chfr at the transcriptional level, although
this was not due to loss of both gene copies. Furthermore, a mutation was identified in a fourth
cell line leading to loss of Chfr function. In tumor cell lines lacking Chfr function, mitotic
chromosome condensation occurred at the same rate in the presence or absence of nocodazole
(or Taxol). In cells with functional Chfr, or cells lacking Chfr but transiently transfected with a
functional copy, chromosome condensation occurred at a reduced rate in nocodazole treated
cells, relative to cell cycle progression from G2 to metaphase based on the accumulation of
cyclin/CDK activity and prophase separation of centrosomes. Examination of nuclear morphology
and DNA content following 48 hours of microtubule depolymerizing treatment demonstrated
that Chfr defective cells undergo aberrant mitosis, implicating this checkpoint in
chromosome instability. While there is no definitive yeast homolog for Chfr, two S. cerevisiae
open reading frames and one S. pombe gene, defective in mitotic arrest (Dma1),99 appear closely
related. Clearly study of a larger cohort of tumor cell lines and further mechanistic studies need
to be performed to fully assess the import of this checkpoint in tumorigenesis. These may also
substantiate the authors claims that Chfr is inactivated more frequently than ‘all known spindle
checkpoint proteins combined’.98
Spindle Assembly Checkpoint
The spindle assembly checkpoint is an example of how a combination of yeast genetics
and cell biology in higher eukaryotes has rapidly expanded our knowledge of a biological system.
26,100-104 Eukaryotic cells arrest in metaphase when microtubule polymerization is
disturbed. Higher eukaryotic cells, with normal mitotic spindles, also arrest if chromosomes
fail to become bioriented on the spindle and have not congressed to the metaphase plate.105
88 Cell Cycle Checkpoints and Cancer
Other defects such as spindle pole body (SPB), kinetochore and centromere abnormalities also
activate the checkpoint,100 and it is known that incorrect spindle orientation (relative to the
cell axis) can delay the onset of anaphase.106 Therefore, the spindle assembly checkpoint monitors
the process of bipolar attachment of all the chromosomes to the mitotic spindle and ensures that
the spindle is correctly positioned.
Chromosomes become bioriented by amphitelic attachment of their kinetochores to spindle
microtubules.100,107 The process of chromosome capture by the spindle occurs more or less
randomly:107 within the same species and cell type, it is accomplished quickly in some cells,
but takes much longer in others.105 Therefore, in animals and in yeast, the checkpoint is needed
every cell cycle.108-110 However, biorientation of the chromosomes on the mitotic spindle forms
a stable structure,107,111 thus the correct alignment of chromosomes, creating the metaphase
plate, is favored. Once the last chromosome becomes bioriented, the spindle assembly checkpoint
signal diminishes and anaphase is initiated in a highly regimented manner.105 At least
one facet of the checkpoint signal emanates from kinetochores that have not attached to the
spindle.112 In higher eukaryotes, a phospho-epitope (recognized by the 3F3/2 antibody) is
present on unattached kinetochores (Fig. 7).107,113 Attachment of microtubules to a kinetochore
induces dephosphorylation of the 3F3/2 phospho-epitopes at that kinetochore. As chromosomes
Fig. 4. Chromosome dynamics mediated by Topoisomerase II (Topo II).Topo II decatenation reactions are
needed for various mitotic steps in mammalian cells. Before mitosis, chromosome individualization is
promoted by the resolution of nonreplicative catenations (left). This process may be monitored by the G2
topo II-dependent checkpoint since forcing G2 cells into mitosis in the absence of topo II activity produces
ο figures (see Checkpoint override). Topo II is also needed for sister chromatid resolution in prometaphase
(middle) and sister separation in anaphase (right). Whether checkpoints controls these processes has not
been fully investigated. The photomicrographs show Muntiacus muntjak chromosomes.
DNA Damage-Independent Checkpoints from Yeast to Man 89
attach to the spindle, the 3F3/2 epitopes are dephosphorylated, and the checkpoint becomes
inactive . The molecular basis of this phosphorylation is not understood, but mechanistically, it is
thought that checkpoint sensors, that are tension-sensitive complexes residing within the
kinetochores, control the kinetochore phosphorylation status.105,107,112,114 Elegant studies have
shown that tension exerted on kinetochores, applied by manipulating chromosomes with a
micro-needle, induces loss of the kinetochore 3F3/2 epitopes.115 Therefore, it appears as though
a lack of tension generates the checkpoint signal. The identity of the kinase which creates the
3F3/2 epitope is not known, but recent work indicates that it is an integral component of
kinetochores. Cells lysed in detergent do not contain kinetochores that are
reactive against the α-3F3/2 antibody, but the α-3F3/2 reactivity can be reinstated by the addition
of ATP.116,117 The activity of the kinase must be tightly associated with kinetochores.
Furthermore, the substrate and kinase are likely to be associated. In theory, this ‘in vitro’ system
could be used as a biochemical assay to identify the kinase.Genetic studies have revealed
components of the yeast spindle assembly checkpoint (Fig. 8). Several groups of checkpoint
proteins were identified in genetic screens designed to find mutants sensitive to microtubule
antagonists. These are Mad1, Mad2, Mad3 (Mitotic Arrest Defective),109 and Bub1, Bub2,
Bub3 (Budding Uninhibited by Benzimidazole).118 In addition, Mps1 is required.119 Many of
these proteins have homologs in higher eukaryotes (see Table 1). One of these proteins, Mad2,
was shown to bind selectively to phosphorylated kinetochores in vertebrate cells.117
Conversely, Mad2 binding was inhibited by kinetochore-microtubule attachment.120 Therefore,
phosphorylated components of attachment-sensitive or tension-sensitive complexes might be
recognized by Mad2. The current hypothesis is that Mad2 binding to 3F3/2 positive epitopes
leads to the formation of an active checkpoint complex. In this model, kinetochores are
Fig. 5. Location of putative topoisomerase II-dependent checkpoint sensors. The cartoons depict a model
of chromosome structure showing loops of chromatin attached to a chromosome core or scaffold, and the
location of nonreplicative and long-lived replicative catenations. (Checkpoint 1) Perturbed topo II activity
in G2 results in persistent nonreplicative catenations between chromatin loops, positioned at distant points
within the same metaphase chromosome but fortuitously closer during interphase (left). Nonreplicative
catenations produce a cytologically observable chromosome aberration named ο-figures. When these catenations
involve different chromosomes they promote interchromosomal recombination. ο-figures are
likely to generate tension at the base of the loops at sites of catenation. Such regions might have the potential
to produce a checkpoint response in G2. (Checkpoint 2) Long lived replicative catenations are present at
the bases of chromatin loops, in the regions where loops are attached to the closely opposed sister chromatid
cores, until prometaphase (right). These replicative catenations might help maintain sister chromatid
cohesion until prometaphase. Tension at sites of replicative catenation, created by ongoing chromosome
condensation, might generate a checkpoint response during mitosis.
90 Cell Cycle Checkpoints and Cancer
catalytic sites for formation of the checkpoint signaling element, namely the activated Mad2
complex.
The target of the activated checkpoint complex was revealed in key experiments demonstrating
that cell cycle arrest is brought about by inhibition of APC activity, which in turn
prevents Pds1 ubiquitination and subsequent degradation. Yeast Mad2 was shown to bind to
Cdc20, a component of the APC required for Pds1 degradation,121,122 and this binding can
inhibit ubiquitination of APC substrates.121 Overexpression of CDC20, or expression of a
cdc20 mutant that cannot bind to Mad2, bypasses the spindle assembly checkpoint arrest.122
In the “catalytic kinetochore” model, unattached kinetochores might form an active site at
which Mad2-Cdc20 complexes are assembled, then released, thereby excluding Cdc20 from APCs.
Alternatively, active Mad2 complexes might be released from kinetochores allowing them to
inhibit APCCdc20 in other cellular locations. The latter model is supported by measurements of
Mad2 localization dynamics in living cells; Mad2 is a transient component of
unattached kinetochores, having a t1/2 of roughly 25 seconds.123
But what is the nature and function of the active Mad2 complex? Studies in yeast have
shown that spindle defects activate kinase Mps1, resulting in Mad1 hyperphosphorylation
(perhaps directly by Mps1).119,124 Overexpression of Mps1 alone can activate the checkpoint
and this arrest is (at least partly) dependent on Mad1, Mad2, Mad3 and Bub1, Bub2, Bub3.
Mad1 phosphorylation also requires Bub1, Bub3 and Mad2.124,125 This modified form of
Mad1 is required to mediate metaphase arrest. Significantly, Mad1 has been shown to bind to
Mad2, and in this complex, Mad1 is a better substrate for Mps1 kinase than is unbound Mad1.
At least in Xenopus egg extracts, Mad1 is required for the association of Mad2 to kinetochores
that are not attached to the mitotic spindle.126 Together, the yeast genetic data and studies in
higher eukaryotes indicate that checkpoint activation relies on the recognition of unattached
kinetochores by Mad2, and the formation of an activated Mad1-Mad2 complex in which Mad1
is hyperphosphorylated. But how do the other checkpoint components fit into this scheme?
Somewhat parallel to the case of the Mad1-Mad2 complex, yeast Bub1 and Bub3 are tightly
associated.127 This is also the case in mammalian cells, and the Bub1 domain required for Bub3
binding is also needed for localization of Bub1 to kinetochores.128 The implication is that
Fig. 6. A model for the bacterial Rcd checkpoint. Rcd imposes a cell cycle arrest before cell division that
allows time for plasmid multimers to be recombined into stable monomers. Rcd is encoded from within
the cer element of the recombination site. Transcription of Rcd occurs only in multimers and is induced by
the topological environment created when the XerC/XerD/ArgR/PepA recombination complex forms. In
theory, such a system could be utilized in mammalian cells to monitor the catenation state of DNA.
DNA Damage-Independent Checkpoints from Yeast to Man 91
Bub3 drives localization of Bub1 to kinetochores, as is the case for Mad1 and Mad2. A recent
study of the budding yeast proteins sheds some light into how these complexes might be
related.108 Mad1 was shown to associate with Bub1 and Bub3 in unperturbed cell cycles and
the amount of the complex in cells was increased in response to spindle checkpoint activation.
Mad2 and Mps1 are required for the formation of the Bub1-Bub3-Mad1 complex in yeast.108
A Mad1 mutation that abolished Bub1-Bub3-Mad1 complex formation, also led to a defective
spindle checkpoint.
But how are the Mad1-Mad2 and Bub1-Bub3 complexes related? It may be the case that each
complex becomes localized to kinetochores under slightly different conditions, in order to broaden
the scope of defects that the checkpoint can detect. However, since deletion mutants of any one of
these components results in a fully defective checkpoint, it is hard to argue that the complexes
play entirely redundant roles. Instead, the different complexes might well detect different aberrations,
but still all be needed for generating the active checkpoint complex that inhibits
APC.Cdc20 Recent studies have allowed a working model to explain such an interconnection
between the Mad1-Mad2 and the Bub1-Bub3 complexes, and how APCCdc20 might be inhib-
Fig. 7 Tension-sensitive kinetochore phosphorylation. The kinetochores of prometaphase chromosomes (a-c
and the left cell in g) become phosphorylated forming a 3F3/2 antibody-reactive epitope (open circles). The
phospho-epitope is lost (filled circles) as chromosomes attach to the mitotic spindle forming the metaphase
plate (b-d). This inactivates the spindle assembly checkpoint and cells begin anaphase roughly 20 minutes
later (e-f and the right cell in g). Micrographs show onion root meristematic cells. The left cell is represented
schematically in c, in which one chromosome has not congressed (also see arrow in g).
92 Cell Cycle Checkpoints and Cancer
ited.108 Although Bub1-Bub3-Mad1 complexes exist in yeast, and this complex forms in a
manner dependent on Mad2 and Mps1 kinase activity, Mad2 is not found in this complex.108
Additionally, the Bub1-Bub3-Mad1 complex does not seem to be able to bind to Cdc20.108
This might suggest that an exchange mechanism is necessary to generate the active checkpoint
complex (Fig. 9). In such a model, Mad2 is displaced from Mad1-Mad2
complexes, induced by Mps1 kinase, simultaneously stimulating the formation of the Bub1-
Bub3-Mad1 complex on the one hand and an active Mad2 complex on the other.108,129 Mad1
phosphorylation, dependent on Mps1, Bub1, Bub3 and Mad2, may also be involved in this
exchange. The nature of the activated Mad2 complex in not known, but may include Mad3.129
In support of this model, animal homologs of Mad1, Mad2, Mad3, Bub1 and Bub3 are
found at the kinetochores of prophase and prometaphase (not yet bioriented) chromosomes.
Following congression to the metaphase plate, these proteins seem to dissociate.130 These local-
Fig. 8. Two branches of the spindle checkpoint. Mad2 and Bub2 are both activated by the kinase Mps1, but
induce arrest at different stages of mitosis. The Mad2 (spindle assembly) pathway inhibits the metaphase
to anaphase transition by preventing APCCdc20-dependent Pds1 ubiquitination while the Bub2 pathway
inhibits mitotic exit by preventing APCCdh1-dependent B-type cyclin degradation and by maintaining
CDK activity. Both branches are required for indefinite arrest although a significant delay in mitosis can be
invoked when either pathway is activated. Photomicrographs of onion root meristematic cells depict the
mitotic stages. See text for detailed descriptions.
DNA Damage-Independent Checkpoints from Yeast to Man 93
ization studies suggest that formation of an active checkpoint complex within kinetochores is
likely to be a conserved mechanism that activates the checkpoint pathway. But how is the
checkpoint signal mobilized? How does a single unattached kinetochore generate a signal that
inhibits anaphase spindle elongation and prevents loss of sister chromatid cohesion of all the
other chromosomes? One study has revealed important information that should help to resolve
this question. Rieder et al, examined the timing of anaphase onset in cells that contain two
functional and independent spindles.131 Such polykaryons are generated by cell fusion. When
two cells at different stages of the cell cycle are fused, cell cycle progression of their nuclei soon
becomes synchronized, allowing measurements of anaphase timing in independent spindles
that share a common cytoplasm. This analysis revealed that the inhibitor emanating from a
single unattached kinetochore is not freely diffusible, but rather is likely to be associated with
the spindle itself. Therefore activated complexes might track from kinetochores along spindles.
Checkpoint Control of Mitotic Exit
Many of the components of the mitotic exit machinery have been identified by the cell
cycle phenotype of budding yeast mutants which arrest as a large dumbbells with elongated
spindles (see Table 1 and 2). This phenotype is consistent with arrest at the anaphase/telophase
transition. Mutants are unable to pass this arrest and the proteins are therefore essential for exit
from mitosis. These genes appear to define a GTP-dependent kinase signaling cascade,
ultimately releasing a phosphatase that induces spindle disassembly, cytokinesis and mitotic
exit. Control of mitotic exit therefore resides in the inhibition of this essential pathway.
Exit from mitosis absolutely requires inhibition of B-type cyclin/CDK activity. Under
normal circumstances this is mediated by both inhibition of CDK activity and by degradation
of mitotic cyclins. Study of the S. cerevisiae spindle checkpoint has revealed that the ‘spindle
assembly’ checkpoint is branched, inhibiting both the transition from metaphase to anaphase
and mitotic exit (Fig. 8). The different functions of the two branches begs the question as to
whether it is erudite to continue calling both branches by the term ‘spindle assembly checkpoint’.
Others have begun to call the two branches the ‘spindle assembly’ and ‘spindle position’
checkpoints.132 For the purposes of this review we continue to use the term spindle assembly
checkpoint for the inhibition of the metaphase-anaphase transition and the generic term
‘mitotic exit control’ for the branch that regulates the activity of the B-type cyclin/CDK activity.
The mammalian machinery for mitotic exit control is currently being elucidated while
substantial inroads into understanding the mechanism has been achieved in S. cerevisiae. Here we
describe current knowledge in the S. cerevisiae checkpoint control of mitotic exit (Fig. 10). Comparison
to mammalian homologs and discussion of their possible clinical importance in
tumorigenesis is left for the next section.
Evidence that Bub2 operates in a separate checkpoint pathway to the Bub1, 3, Mad1-3
pathway (hereafter collectively referred to as the Mad2 pathway) came from studies of double
mutants treated with antitubulin drugs.133-138 Double mutant combinations that included bub2
failed to arrest in nocodazole whilst double mutant combinations that did not include Bub2
retained a mitotic delay. However, in mad2 cells treated with nocodazole the metaphase-anaphase
transition occurs with kinetics comparable to those of untreated cells while bub2 cells delay the
metaphase-anaphase transition. In addition, delay of the cell cycle in ctf13 mutants (limited for
a key kinetochore component) requires Bub1 and 3 and Mad 1, 2 and 3 but is independent of
Bub2.139 In cdc20 mutants, the mitotic arrest caused by maintained Pds1 levels is dependent on
Bub2 but independent of the Mad2 pathway genes.140 Inhibition of Dbf2 in late mitotic arrest
requires Bub2 but not the Mad2 pathway proteins.141 Together, these studies define distinct
pathways. The Mad2 pathway ultimately targets Pds1, preventing spindle elongation and loss
of sister chromatid cohesion at the metaphase to anaphase transition. The Bub2 pathway
inhibits mitotic cyclin/CDK activity, and thus prevents spindle disassembly and exit from mitosis.
Since the Bub2 pathway also monitors spindle integrity, and is triggered by microtubule
depolymerizing agents, there is a common upstream element, kinase Mps1. However, in
94 Cell Cycle Checkpoints and Cancer
Table 1. Mitotic checkpoint components
S. cerevisiae S. pombe M. musculus H. sapiens Chromosomal location
(H.s.)
BFA1 byr4+— — —
BUB1 bub1+ Bub1 BUB1 2q14
BUB2 cdc16+ Mm29982 * VRP* 2
BUB3 spac23h3.08c * Bub3 BUB3 10q26
CDC14 † spac782.09c * Mm.6355 */Mm28909 * CDC14A/CDC14B 1p21 / 9
CDC15 † cdc7+ Mess1 * STK4 */STK3 * 20q11.2-13.2 / ?
CDC20 † slp1+ — CDC20 9q13-q21
CDC5 † plo1+ Plk/Cnk/Stk18 * PLK/CNK/SNK/STK18 * 16 / 1 / 5/ 4q27-q28
CDH1 spbc1198.12 * Mm.2440 * hCDH1 19p13.3
DBF2 sid2+/spcc417.06 * Lats1 */Lats2 * LATS1 */LATS2 * ? / 13q11-q12
DBF20 sid2+/spcc417.06 * Lats1 */Lats2 * LATS1 */LATS2 * ? / 13q11-q12
ESP1 † cut1+ — ESP1 8
LTE1 efc25+ */ste6+ * Rasgrf1 * RASGRF1 * 15q24
MAD1 spbcd6.04 */spac26a3.10 * — MAD1L1 7p22
MAD2 mad2+ Mad2L1 MAD2L1/MAD2L2 4q27 / 1p36
MAD3 bub1+ */spcc895.02 * Bub1b hBUB1R1 15q14-21
MOB1 † mob1+ — FLJ10788 * —
MPS1 † mph1+ Ttk TTK 6q13-q21
NET1 spbc1711.05 * Mm.4480 * LAD1 * 1q25.1-q32.3
PDS1 cut2+ mSec PTTG 5q35.1
SIC1 — Nfatc3 * — —
TEM1 † spg1+ Rab5b * RAB36 * 22q11.22
BIM1 mal3+ Eb1 EB1 —
* denotes ortholog based solely upon sequence homology. † denotes essential gene in S. cerevisiae.
DNA Damage-Independent Checkpoints from Yeast to Man 95
contrast to the inhibition of anaphase onset via APCCdc20 regulation mediated by the Mad2
pathway, the Bub2 pathways appears to primarily act by inhibition of mitotic cyclin degradation
and maintenance of mitotic cyclin dependent kinase activity. This is achieved by suppression of
APCCdh1 and Sic1 activity which promote Clb1/Clb2 degradation and inhibit mitotic cyclin
dependent kinase activity respectively.
The mitotic exit branch of the checkpoint is essential if microtubule polymerization is
perturbed during anaphase i.e., after APCCdc20 dependent degradation of Pds1. Normal
progression of the cell cycle ensures that Cdc20 remains active and bound to the APC until
after the onset of anaphase when Cdh1 replaces Cdc20 as the APC specificity factor targeting
B-type cyclins for degradation. However, deletion of CDH1 is not sufficient to prevent mitotic
exit since inactivation of mitotic cyclin dependent kinase activity by Sic1 is sufficient to allow
exit from mitosis. The redundancy of APCCdh1 and Sic1 activity ensures that cells may exit
mitosis in the absence of checkpoint stimulation. In the presence of checkpoint stimulation the
activity of both Sic1 and APCCdh1 is inhibited by nucleolar sequestration of the phosphatase
Cdc14. Indeed, it appears that release of Cdc14 from the nucleolus is a key event in mitotic
exit. The mechanism of Cdc14-mediated exit from mitosis appears to be three-fold. First, by
dephosphorylating Cdh1 the APC targets the B-type cyclins for degradation. Ordinarily,
phosphorylation of Cdh1 by Cdc28 is inhibitory and thus is self protecting,142 but when Cdh1
Table 2. Function/localization of mitotic checkpoint components
Gene Function Localization
BFA1 ? dSPB
BUB1 S/T kinase nucleus/kinetochore
BUB2 GAP (GTPase activating protein) dSPB
BUB3 ? nucleus/kinetochore
CDC14 † Phosphoprotein phosphatase nucleolus/nucleus
CDC15 † S/T kinase dSPB/bud neck
CDC20 † APC specificity factor kinetochores/SPB/spindle
CDC5 † S/T kinase SPB
CDH1 APC specificity factor ?
DBF2 S/T kinase SPB/bud neck
DBF20 S/T kinase nucleus
ESP1 † Separin nucleus/spindle
LTE1 GEF (GDP/GTP exchange factor) bud
MAD1 ? nucleus
MAD2 Cdc20 inhibitor kinetochores/SPB
MAD3 ? nucleus/kinetochore
MOB1 † ? SPB
MPS1 † S/T kinase ?
NET1 sequesters Cdc14 nucleolus
PDS1 sequesters ESP1; securin nucleus/spindle
SIC1 B-type cyclin/CDK Inhibitor ?
TEM1 † GTP binding memebr of RAS superfamily dSPB
BIM1 ? ?
† denotes essential gene in S. cerevisiae. dSPB denotes daughter-bound spindle pole body.
S/T kinase denotes serine/threonine protein kinase.
96 Cell Cycle Checkpoints and Cancer
is dephosphorylated by Cdc14 this self-protection is removed. Second, by dephosphorylating
Swi5, activating the transcription of Sic1, and third by directly dephosphorylating Sic1 itself.
Thus Cdc14 both inhibits the activity of cyclin/CDK activity and induces the destruction of
the cyclin components.143
How does the ‘mitotic exit’ checkpoint inhibit the release of Cdc14 from the nucleolus?
Throughout most of the cell cycle Cdc14 is held inactive within the nucleolus in complex with
Net1/Cfi1 (Fig. 11)144,145 termed the RENT complex (regulator of nucleolar silencing and
telophase). This inactive localization appears to be dependent upon Tem1;145 a GTP binding
protein localized to the daughter-bound spindle pole body (SPB). Cdc14 may also play a structural
role in the nucleolus.146 A recent paper proposes a mechanism for monitoring the completion
of anaphase and presents a compelling model.147 When SPB-associated Tem1-GDP locates
to the bud at the end of anaphase it interacts with a cortical protein Lte1 which is a GDP/
GTP exchange factor (GEF). Tem1-GDP is activated by conversion to Tem1-GTP triggering
the release of Cdc14 via a kinase cascade termed the mitotic exit network (MEN see below).
Even when localized to the bud cortex, Tem1 activation could be inhibited by GAP (GTPase
activating protein) activity of Bub2, preventing exit from mitosis. There are several compelling
reasons for supposing this to be the checkpoint mechanism. Bub2 has considerable sequence homology
to cdc16+ in S. pombe which is known to form a two component GAP with byr4+. Together
they activate the GTPase encoded by spg1+, the S. cerevisiae homolog of Tem1. Furthermore,
deletion of the cerevisiae homolog of byr4+ (Bfa1) causes a phenotype similar to bub2, as
does overexpression of Tem1. Finally, localization of Bub2 to the SPB and preferentially to that
Fig. 9. Model for the formation of the activated spindle assembly checkpoint complex. Association of Mad1-
Mad2 and Bub1-Bub3 complexes with kinetochores is stimulated by the onset of spindle assembly and/or
spindle aberrations. Kinase Mps1 induces exchange between the complexes, forming an active Mad2
complex, perhaps containing Mad3. Exchange may be stimulated by Mad1 phosphorylation. Activated
Mad2, released from kinetochores, binds to Cdc20 to inactivate the APC.
DNA Damage-Independent Checkpoints from Yeast to Man 97
destined for the daughter cell,138,148,149 provides strong circumstantial evidence in support of
this model.
The mechanism by which activation of Tem1 leads to the release of Cdc14 from the
nucleolus via the MEN involves a number of key components including Cdc15, Cdc5, Dbf2,
Dbf20 and Mob1.150 Most appear essential for mitotic exit since single mutants are lethal,
while in the case of Dbf2 and Dbf20, it is only the double mutant that is synthetically lethal.151
Localization and phosphorylation of these proteins appear to be important factors during the
cell cycle and probably contribute to their function (Table 2). In particular, many of the com-
Fig. 10. Mitotic exit (Bub2) pathway: Many components of the MEN localize to the spindle pole bodies
either symmetrically or asymmetrically with preference for that destined for the daughter cell. Components
of the RENT complex localize mainly to the nucleolus. Release of Cdc14 from the nucleolus dephosphorylates
Sic1, Swi5 and Cdh1 both activating APC dependent B-type cyclin degradation and inhibiting Btype
cyclin-dependent kinase activity. Components underlined are essential for mitotic exit as mutants
arrest in telophase with a ‘dumbbell morphology’. The involvement of Cdc5 in MEN and interactions of
the Mob1/Dbf2/Dbf20 complex in MEN have yet to be defined. Heavy arrows represent activation with
demonstrated physical interaction.
98 Cell Cycle Checkpoints and Cancer
ponents localize to the nucleolus or are asymmetrically distributed between the SPBs,
being preferentially bound to that destined for the daughter cell. Cdc15 localizes to the SPB
during mitosis and relocates to the bud neck after telophase.152,153 Furthermore, Cdc15 phosphorylation
increases gradually during the cell cycle until it is rapidly dephosphorylated in late
mitosis.152,154 Like Cdc15, Dbf2 localizes to the SPB and moves to the bud neck in telophase.155
While asymmetric localization to the daughter-bound SPB is true for some components, others
such as Mob1 (or at least the S. pombe homolog), localize symmetrically to both spindle pole
bodies. Localization to the SPB and the relocation of these components to the bud neck in
telophase mimics the localization in S. pombe in which their homologs principally act by
regulating cytokinesis rather than mitotic exit per se.
While mechanistic details of the MEN/RENT complex are being reported, much work
has still to be completed before a clear picture can be drawn. Some details of physical
interactions and colocalization can at least allow some description of the cascade (Fig. 10). The
rapid dephosphorylation of Cdc15 in late mitosis appears to be mediated by Cdc14.153 However,
there also appears to be a role for Cdc15 as an activator of Cdc14, and it is thus both an
activator and substrate.154 Also there appears to be a role for Pds1 as an inhibitor of B-type
cyclin degradation independent of its role as a securin156 thus forming potential ‘crosstalk’
between the Mad2 and Bub2 checkpoint branches. What is less clear is the role of Cdc5 which
physically interacts with Dbf4 (part of the DNA replication machinery) but as yet has no
clearly defined role in the mitotic exit pathway. As Cdc5 is a target of DNA damage checkpoint
control, this component offers an attractive link between the damage checkpoint and mitotic
exit control. Direct interactions have been established for a number of MEN components
while the exact nature of many interactions remain unclear. For example, physical interaction
between Mps1 and Mob1 has been demonstrated. Furthermore, interactions between Mob1 and
Dbf2 and Dbf20 have been demonstrated. However, there has been no direct link established
between these components and the remaining MEN components to date. Similarly, the role of
Cdc5 in the MEN has yet to be elucidated. Nonetheless, we present an attempt to order the
Fig. 11. Model for colocalization of Tem1/Lte1 inducing mitotic exit in budding yeast. Mitotic exit is
triggered by Tem1-GTP that induces release of Cdc14 from the nucleolus. Inactive Tem1-GDP is bound
to the SPB destined for the daughter cell. Once the spindle has elongated into the bud, the guanine
nucleotide exchange factor Lte1 activates Tem1.
DNA Damage-Independent Checkpoints from Yeast to Man 99
events of mitotic exit regulation based upon physical interactions and localization in Figure 10.
In this scheme we present Cdc15 ‘upstream’ of Cdc14 as the former localizes to the SPB
coincident with other upstream elements.
Evidence suggests that Bub2, and its associated partner Bfa1, participate in an essential
checkpoint that is also activated by DNA damage.157,158 Thus the maintenance of B-type cyclin/
CDK activity by the Bub2 pathway may represent a universal mechanism that can respond to
stress at any stage of G2 and M-phase. Other genetic interactions of Bub2 include synthetic
lethality of a bub2 arc35-1 mutant159 and that Bub2 is essential for arrest in tub4-1 cells.160 Thus,
the Bub2 pathway seems to respond to defects in spindle orientation, spindle localization, spindle
damage and DNA damage.
Oncological Implications of Mitotic Checkpoint Homologs
The existence of numerous mitotic exit and spindle assembly checkpoint protein
homologs in S. pombe and higher eukaryotes suggests that similar mechanisms regulate
mitotic exit in all eukaryotes despite the fact that the asymmetric cytokinesis in S. cerevisiae
appears to have fundamentally different spatial and temporal strategies. As aberrant mitosis
frequently results in asymmetric distribution of the genetic material and aneuploid daughter cells,
dysfunctional regulation of these checkpoints has become an attractive hypothetical mechanism
for chromosomal instability in mammalian tumorigenesis. Indeed, some established
tumor cell lines and tumors appear to have dysfunctional checkpoint controls23,161,162 and
some of the checkpoint proteins appear to be targets of oncogenic viral proteins.163 However,
screening of aneuploid colorectal tumor panels for such mutations revealed only mutations in
the human hBUB1/hBUBR1 genes.164 A similar study of 31 aneuploid lung, and head and
neck, tumors showed no such mutations.165 One hBUB1 somatic mutation that led to an
amino acid substitution was found among 30 human primary lung cancer tumors.166 hBUB1
and hBUBR1 mutants have also been found in adult T-cell leukemia’s/lymphomas 167 and
some colorectal tumor cell lines.161 Perhaps significantly, one study has implicated Brca2, which
is responsible for a fraction of the inherited susceptibilities to breast cancer, in the spindle
assembly checkpoint.168 Brca2 was found to interact with hBubR1, and was phosphorylated
by hBubR1 in vitro, though no direct role in the checkpoint was demonstrated. Although
inactivation of Bub1 appears to confer chromosomal instability,161 more studies are required to determine
whether mutations in BUB1 and other mitotic checkpoint proteins represent
significant causative events or whether other checkpoints may account for aneuploid tumorigenesis.
In addition to the MAD2 and BUB2 pathway components, a number of other S. cerevisiae
genes appear to have mammalian homologs which have been implicated as either protooncogenes
or as tumor suppressor genes (see Table 1). Notably S. cerevisiae Pds1 may have two mammalian
homologs, at least one of which is associated with pituitary tumors.169-172 Most of the human
genes have been localized at least to the chromosome level and recent publication of the human
genome will therefore facilitate further study. Furthermore, many mouse homologs have been
identified and murine models of tumorigenesis may further elucidate the contribution of these
genes to tumorigenesis. At the present time only two knockout mouse models have been
reported, MAD2 173 and BUB3,174 both of which are early embryonic lethals. In the BUB3
mouse, from day 3.5 onwards, embryonic cells display mitotic aberrations such as micronuclei,
anaphase chromosome laggards and bridging. In the presence of microtubule antagonists, the
cells fail to arrest in metaphase.
References
1. Rao PN, Johnson RT. Mammalian cell fusion: Studies on the regulation of DNA synthesis and
mitosis. Nature 1970; 225:159-162.
2. Downes CS, Musk SR, Watson JV et al. Caffeine overcomes a restriction point associated with
DNA replication, but does not accelerate mitosis. J Cell Biol 1990; 110:1855-9.
3. Downes CS, Clarke DJ, Mullinger AM et al. A topoisomerase II-dependent G2 cycle checkpoint
in mammalian cells. Nature 1994; 372:467-70.
100 Cell Cycle Checkpoints and Cancer
4. Ghosh S, Schroeter D, Paweletz N. Okadaic acid overrides the S-phase check point and accelerates
progression of G2-phase to induce premature mitosis in HeLa cells. Exp Cell Res 1996; 227:165-9.
5. Musk SR, Downes CS, Johnson RT. Caffeine induces uncoordinated expression of cell cycle functions
after ultraviolet irradiation. Accelerated cycle transit, sister chromatid exchanges and premature
chromosome condensation in a transformed Indian muntjac cell line. J Cell Sci 1988; 591-9.
6. Schlegel R, Pardee AB. Caffeine-induced uncoupling of mitosis from the completion of DNA replication
in mammalian cells. Science 1986; 232:1264-1266.
7. Steinmann KE, Belinsky GS, Lee D et al. Chemically induced premature mitosis: differential
response in rodent and human cells and the relationship to cyclin B synthesis and p34cdc2/cyclin
B complex formation. Proc Natl Acad Sci USA 1991; 88:6843-7.
8. Tolmach LJ, Jones RW, Busse PM. The action of caffeine on X-irradiated HeLa cells. I. Delayed
inhibition of DNA synthesis. Radiat Res 1977; 71:653-65.
9. Tolmach LJ, Busse PM. The action of caffeine on X-irradiated HeLa cells. IV. Progression delays
and enhanced cell killing at high caffeine concentrations. Radiat Res 1980; 82:374-92.
10. Moser BA, Brondello JM, Baber FB et al. Mechanism of caffeine-induced checkpoint override in
fission yeast. Mol Cell Biol 2000; 20:4288-94.
11. Blasina A, Price BD, Turenne GA et al. Caffeine inhibits the checkpoint kinase ATM. Curr Biol
1999; 9:1135-8.
12. Sarkaria JN, Busby EC, Tibbetts RS et al. Inhibition of ATM and ATR kinase activities by the
radiosensitizing agent, caffeine. Cancer Res 1999; 59:4375-82.
13. Zhou BB, Chaturvedi P, Spring K et al. Caffeine abolishes the mammalian G(2)/M DNA damage checkpoint
by inhibiting ataxia-telangiectasia-mutated kinase activity. J Biol Chem 275:10342-8.
14. Yamamoto A, Guacci V, Koshland D. Pds1p, an inhibitor of anaphase in budding yeast, plays a
critical role in the APC and checkpoint pathway(s). J Cell Biol 1996; 133:99-110.
15. Yamamoto A, Guacci V, Koshland D. Pds1p is required for faithful execution of anaphase in the
yeast, Saccharomyces cerevisiae. J Cell Biol 1996; 133:85-97.
16. Cohen-Fix O, Peters JM, Kirschner MW et al. Anaphase initiation in Saccharomyces cerevisiae is
controlled by the APC-dependent degradation of the anaphase inhibitor Pds1p. Genes Dev 1996;
10:3081-93.
17. Cohen-Fix O, Koshland D. The anaphase inhibitor of Saccharomyces cerevisiae Pds1p is a target of
the DNA damage checkpoint pathway. Proc Natl Acad Sci USA 1997; 94:14361-6.
18. Ciosk R, Zachariae W, Michaelis C et al. An ESP1/PDS1 complex regulates loss of sister chromatid
cohesion at the metaphase to anaphase transition in yeast. Cell 1998; 93:1067-76.
19. Michaelis C, Ciosk R, Nasmyth K. Cohesins: chromosomal proteins that prevent premature separation
of sister chromatids. Cell 1997; 91:35-45.
20. Uhlmann F, Nasmyth K. Cohesion between sister chromatids must be established during DNA
replication. Curr Biol 1998; 8:1095-101.
21. Uhlmann F, Lottspeich F, Nasmyth K. Sister-chromatid separation at anaphase onset is promoted
by cleavage of the cohesin subunit Scc1. Nature 1999; 400:37-42.
22. Jensen S, Segal M, Clarke DJ et al. A novel role of the budding yeast separin Esp1 in anaphase
spindle elongation: Evidence that proper spindle association of Esp1 is regulated by Pds1. J Cell
Biol 2001; 152:27-40.
23. Zou H, McGarry TJ, Bernal T et al. Identification of a vertebrate sister-chromatid separation
inhibitor involved in transformation and tumorigenesis. Science 1999; 285:418-422.
24. Allen JB, Zhou Z, Siede W et al. The SAD1/RAD53 protein kinase controls multiple checkpoints
and DNA damage-induced transcription in yeast. Genes Dev 1994; 8:2401-15.
25. Weinert TA, Kiser GL, Hartwell LH. Mitotic checkpoint genes in budding yeast and the dependence
of mitosis on DNA replication and repair. Genes Dev 1994; 8:652-65.
26. Elledge SJ. Cell cycle checkpoints: Preventing an identity crisis. Science 1996; 274:1664-72.
27. Weinert T. DNA damage checkpoints update: Getting molecular. Curr Opin Genet Dev 1998; 8:185-93.
28. Navas TA, Zhou Z, Elledge SJ. DNA polymerase epsilon links the DNA replication machinery to
the S phase checkpoint. Cell 1995; 80:29-39.
29. Sugimoto K, Shimomura T, Hashimoto K et al. Rfc5, a small subunit of replication factor C
complex, couples DNA replication and mitosis in budding yeast. Proc Natl Acad Sci USA 1996;
93:7048-52.
30. Araki H, Leem SH, Phongdara A et al. Dpb11, which interacts with DNA polymerase II(epsilon)
in Saccharomyces cerevisiae, has a dual role in S-phase progression and at a cell cycle checkpoint.
Proc Natl Acad Sci USA 1995; 92:11791-5.
31. Wang H, Elledge SJ. DRC1, DNA replication and checkpoint protein 1, functions with DPB11 to
control DNA replication and the S-phase checkpoint in Saccharomyces cerevisiae. Proc Natl Acad
Sci U S A 1999; 96:3824-9.
DNA Damage-Independent Checkpoints from Yeast to Man 101
32. Frei C, Gasser SM. The yeast Sgs1p helicase acts upstream of Rad53p in the DNA replication
checkpoint and colocalizes with Rad53p in S-phase-specific foci. Genes Dev 2000; 14:81-96.
33. Masumoto H, Sugino A, Araki H. Dpb11 controls the association between DNA polymerases
alpha and epsilon and the autonomously replicating sequence region of budding yeast. Mol Cell
Biol 2000; 20:2809-17.
34. Lee SK, Johnson RE, Yu SL et al. Requirement of yeast SGS1 and SRS2 genes for replication and
transcription. Science 1999; 286:2339-42.
35. Michael WM, Ott R, Fanning E et al. Activation of the DNA replication checkpoint through
RNA synthesis by primase. Science 2000; 289:2133-2137.
36. Tercero JA, Labib K, Diffley JF. DNA synthesis at individual replication forks requires the essential
initiation factor Cdc45p. Embo J 2000; 19:2082-93.
37. Sanchez Y, Desany BA, Jones WJ et al. Regulation of RAD53 by the ATM-like kinases MEC1 and
TEL1 in yeast cell cycle checkpoint pathways. Science 1996; 271:357-60.
38. Frei C, Gasser SM. RecQ-like helicases: the DNA replication checkpoint connection. J Cell Sci
2000; 2641-6.
39. Sun Z, Hsiao J, Fay DS et al. Rad53 FHA domain associated with phosphorylated Rad9 in the
DNA damage checkpoint. Science 1998; 281:272-4.
40. Paciotti V, Clerici M, Lucchini G et al. The checkpoint protein Ddc2, functionally related to S.
pombe Rad26, interacts with Mec1 and is regulated by Mec1-dependent phosphorylation in budding
yeast. Genes Dev 2000; 14:2046-59.
41. Rouse J, Jackson SP. LCD1: An essential gene involved in checkpoint control and regulation of
the MEC1 signalling pathway in Saccharomyces cerevisiae. Embo J 2000; 19:5801-5812.
42. Huang M, Zhou Z, Elledge SJ. The DNA replication and damage checkpoint pathways induce
transcription by inhibition of the Crt1 repressor. Cell 1998; 94:595-605.
43. Santocanale C, Diffley JF. A Mec1- and Rad53-dependent checkpoint controls late-firing origins
of DNA replication. Nature 1998; 395:615-8.
44. Longhese MP, Paciotti V, Neecke H et al. Checkpoint proteins influence telomeric silencing and
length maintenance in budding yeast. Genetics 2000; 155:1577-91.
45. Craven RJ, Petes TD. Involvement of the checkpoint protein Mec1p in silencing of gene expression
at telomeres in Saccharomyces cerevisiae. Mol Cell Biol 2000; 20:2378-84.
46. Mills KD, Sinclair DA, Guarente L. MEC1-dependent redistribution of the Sir3 silencing protein
from telomeres to DNA double-strand breaks. Cell 1999; 97:609-20.
47. Clarke DJ, Segal M, Mondesert G et al. The Pds1 anaphase inhibitor and Mec1 kinase define
distinct checkpoints coupling S phase with mitosis in budding yeast. Curr Biol 1999; 9:365-8.
48. Blat Y, Kleckner N. Cohesins bind to preferential sites along yeast chromosome III, with differential
regulation along arms versus the centric region. Cell 1999; 98:249-59.
49. Clarke DJ, Mondesert G, Segal M et al. Dosage suppressors of pds1 implicate UBA domains in
checkpoint control. Mol Cell Biol 2001; 21:1997-2007.
50. Ortolan TG, Tongaonkar P, Lambertson D et al. The DNA repair protein Rad23 is a negative
regulator of multi-ubiquitin chain assembly. Nat Cell Biol 2000; 2:601-608.
51. Bartz SR, Rogel ME, Emerman M. Human immunodeficiency virus type 1 cell cycle control: Vpr
is cytostatic and mediates G2 accumulation by a mechanism which differs from DNA damage
checkpoint control. J Virol 1996; 70:2324-31.
52. Withers WE, Jowett JB, Stewart SA et al. Human immunodeficiency virus type 1 Vpr interacts
with HHR23A, a cellular protein implicated in nucleotide excision DNA repair. J Virol 1997;
71:9732-42.
53. Gragerov A, Kino T, Ilyina GG et al. HHR23A, the human homologue of the yeast repair protein
RAD23, interacts specifically with Vpr protein and prevents cell cycle arrest but not the transcriptional
effects of Vpr. Virology 1998; 245:323-30.
54. Watt PM, Louis EJ, Borts RH et al. Sgs1: a eukaryotic homolog of E. coli RecQ that interacts
with topoisomerase II in vivo and is required for faithful chromosome segregation. Cell 1995;
81:253-60.
55. Yu CE, Oshima J, Fu YH et al. Positional cloning of the Werner’s syndrome gene. Science 1996;
272:258-62.
56. Ellis NA, Groden J, Ye TZ et al. The Bloom’s syndrome gene product is homologous to RecQ
helicases. Cell 1995; 83:655-66.
57. Wang Y, Cortez D, Yazdi P et al. BASC, a super complex of BRCA1-associated proteins involved
in the recognition and repair of aberrant DNA structures. Genes Dev 2000; 14:927-39.
58. Savitsky K, Bar SA, Gilad S et al. A single ataxia telangiectasia gene with a product similar to PI-
3 kinase. Science 1995; 268:1749-53.
59. Bentley NJ, Holtzman DA, Flaggs G et al. The Schizosaccharomyces pombe rad3 checkpoint gene.
Embo J 1996; 15:6641-51.
102 Cell Cycle Checkpoints and Cancer
60. Kastan MB, Zhan Q, el Deiry WS et al. A mammalian cell cycle checkpoint pathway utilizing p53
and GADD45 is defective in ataxia-telangiectasia. Cell 1992; 71:587-97.
61. Wright JA, Keegan KS, Herendeen DR et al. Protein kinase mutants of human ATR increase
sensitivity to UV and ionizing radiation and abrogate cell cycle checkpoint control. Proc Natl Acad
Sci U S A 1998; 95:7445-50.
62. Tibbetts RS, Brumbaugh KM, Williams JM et al. A role for ATR in the DNA damage-induced
phosphorylation of p53. Genes Dev 1999; 13:152-7.
63. Banin S, Moyal L, Shieh S et al. Enhanced phosphorylation of p53 by ATM in response to DNA
damage. Science 1998; 281:1674-7.
64. Canman CE, Lim DS, Cimprich KA et al. Activation of the ATM kinase by ionizing radiation and
phosphorylation of p53. Science 1998; 281:1677-9.
65. Cortez D, Wang Y, Qin J et al. Requirement of ATM-dependent phosphorylation of BRCA1 in
the DNA damage response to double-strand breaks. Science 1999; 286:1162-1166.
66. Bunz F, Dutriaux A, Lengauer C et al. Requirement for p53 and p21 to sustain G2 arrest after
DNA damage. Science 1998; 282:1497-501.
67. Chan TA, Hermeking H, Lengauer C et al. 14-3-3Sigma is required to pevent mitotic catastrophe
after DNA damage. Nature 1999; 401:616-20.
68. Hermeking H, Lengauer C, Polyak K et al. 14-3-3 sigma is a p5-regulated inhibitor of G2/M
progression. Mol Cell 1997; 1:3-11.
69. Tibbetts RS, Cortez D, Brumbaugh KM et al. Functional interactions between BRCA1 and the
checkpoint kinase ATR during genotoxic stress. Genes Dev 2000; 14:2989-3002.
70. Matsuoka S, Huang M, Elledge SJ. Linkage of ATM to cell cycle regulation by the Chk2 protein
kinase. Science 1998; 282:1893-7.
71. Sanchez Y, Wong C, Thoma RS et al. Conservation of the Chk1 checkpoint pathway in mammals:
Linkage of DNA damage to CDK regulation through Cdc25. Science 1997; 277:1497-501.
72. Peng CY, Graves PR, Thoma RS et al. Mitotic and G2 checkpoint control: regulation of 14-3-3
protein binding by phosphorylation of Cdc25C on serine-216. Science 1997; 277:1501-5.
73. Matsuoka S, Rotmn G, Ogawa A et al. Ataxia telangiectasia-mutated phosphorylates Chk2 in vivo
and in vitro. Proc Natl Acad Sci USA 2000; 97:10389-10394.
74. Chaturvedi P, Eng WK, Zhu Y et al. Mammalian Chk2 is a downstream effector of the ATMdependent
DNA damage checkpoint pathway. Oncogene 1999; 18:4047-54.
75. Kappas NC, Savage P, Chen KC et al. Dissection of the XChk1 Signaling Pathway in Xenopus
laevis Embryos. Mol Biol Cell 2000; 11:3101-3108.
76. Sibon OC, Stevenson VA, Theurkauf WE. DNA-replication checkpoint control at the Drosophila
midblastula transition. Nature 1997; 388:93-7.
77. Yu KR, Saint RB, Sullivan W. The Grapes checkpoint coordinates nuclear envelope breakdown
and chromosomecondensation. Nat Cell Biol 2000; 2:609-615.
78. Kumagai A, Guo Z, Emami KH et al. The Xenopus Chk1 protein kinase mediates a caffeinesensitive
pathway of checkpoint control in cell-free extracts. J Cell Biol 1998; 142:1559-69.
79. Guo Z, Kumagai A, Wang SX et al. Requirement for atr in phosphorylation of Chk1 and cell
cycle regulation in response to DNA replication blocks and UV-damaged DNA in xenopus egg
extracts. Genes Dev 2000; 14:2745-56.
80. Liu Q, Guntuku S, Cui XS et al. Chk1 is an essential kinase that is regulated by Atr and required
for the G(2)/M DNA damage checkpoint. Genes Dev 2000; 14:1448-59.
81. Kumagai A, Dunphy WG. Claspin, a novel protein required for the activation of Chk1 during a
DNA replication checkpoint response in xenopus egg extracts. Mol Cell 2000; 6:839-849.
82. Adachi Y, Luke M, Laemmli UK. Chromosome assembly in vitro: Topoisomerase II is required for
condensation. Cell 1991; 64:137-48.
83. Hirano T, Mitchison TJ. Cell cycle control of higher-order chromatin assembly around naked
DNA in vitro. J Cell Biol 1991; 115:1479-89.
84. Hirano T, Mitchison TJ. Topoisomerase II does not play a scaffolding role in the organization of
mitotic chromosomes assembled in Xenopus egg extracts. J Cell Biol 1993; 120:601-12.
85. Uemura T, Ohkura H, Adachi Y et al. DNA topoisomerase II is required for condensation and
separation of mitotic chromosomes in S. pombe. Cell 1987; 50:917-25.
86. Wood ER, Earnshaw WC. Mitotic chromatin condensation in vitro using somatic cell extracts and
nuclei with variable levels of endogenous topoisomerase II. J Cell Biol 1990; 2839-50.
87. Rose D, Holm C. Meiosis-specific arrest revealed in DNA topoisomerase II mutants. Mol Cell
Biol 1993; 13:3445-55.
88. Giménez-Abián JF, Clarke DJ, Mullinger AM et al. A postprophase topoisomerase II-dependent
chromatid core separation step in the formation of metaphase chromosomes. J Cell Biol 1995;
131:7-17.
DNA Damage-Independent Checkpoints from Yeast to Man 103
89. Holm C, Goto T, Wang JC et al. DNA topoisomerase II is required at the time of mitosis in
yeast. Cell 1985; 41:553-563.
90. Koshland D, Strunnikov A. Mitotic chromosome condensation. Annu Rev Cell Dev Biol 1996;
12:305-33.
91. Dietzel S, Jauch A, Kienle D et al. Separate and variably shaped chromosome arm domains are
disclosed by chromosome arm painting in human cell nuclei. Chromosome Res 1998; 6:25-33.
92. Giménez-Abián JF, Clarke DJ, Devlin J et al. Premitotic chromosome individualization in mammalian
cells depends on topoisomerase II activity. Chromosoma 2000; 109:235-44.
93. Summers DK, Sherratt DJ. Multimerization of high copy number plasmids causes instability: CoIE1
encodes a determinant essential for plasmid monomerization and stability. Cell 1984; 36:1097-103.
94. Summers DK, Sherratt DJ. Resolution of ColE1 dimers requires a DNA sequence implicated in
the three-dimensional organization of the cer site. Embo J 1988; 7:851-8.
95. Patient ME, Summers DK. ColE1 multimer formation triggers inhibition of Escherichia coli cell
division. Mol Microbiol 1993; 9:1089-95.
96. Sharpe ME, Chatwin HM, Macpherson C et al. Analysis of the CoIE1 stability determinant Rcd.
Microbiology 1999; 2135-44.
97. Darbon JM, Penary M, Escalas N et al. Distinct Chk2 activation pathways are triggered by genistein
and DNA-damaging agents in human melanoma cells. J Biol Chem 2000; 275:15363-9.
98. Scolnick DM, Halazonetis TD. Chfr defines a mitotic stress checkpoint that delays entry into
metaphase. Nature 2000; 406:430-5.
99. Murone M, Simanis V. The fission yeast dma1 gene is a component of the spindle assembly checkpoint,
required to prevent septum formation and premature exit from mitosis if spindle function is
compromised. Embo J 1996; 15:6605-16.
100. Rudner AD, Murray AW. The spindle assembly checkpoint. Curr Opin Cell Biol 1996; 8:773-80.
101. Nishimoto T, Uzawa S, Schlegel R. Mitotic checkpoints. Curr Opin Cell Biol 1992; 4:174-9.
102. Straight AF. Cell cycle: Checkpoint proteins and kinetochores. Curr Biol 1997; 7:R613-6.
103. Straight AF, Murray AW. The spindle assembly checkpoint in budding yeast. Methods Enzymol
1997; 283:425-40.
104. Waters JC, Salmon E. Pathways of spindle assembly. Curr Opin Cell Biol 1997; 9:37-43.
105. Rieder CL, Schultz A, Cole R et al. Anaphase onset in vertebrate somatic cells is controlled by a checkpoint
that monitors sister kinetochore attachment to the spindle. J Cell Biol 1994; 127:1301-10.
106. O’Connell CB, Wang YL. Mammalian spindle orientation and position respond to changes in cell
shape in a dynein-dependent fashion. Mol Biol Cell 2000; 11:1765-74.
107. Nicklas RB. How cells get the right chromosomes. Science 1997; 275:632-7.
108. Brady DM, Hardwick KG. Complex formation between Mad1p, Bub1p and Bub3p is crucial for
spindle checkpoint function. Curr Biol 2000; 10:675-8.
109. Li R, Murray AW. Feedback control of mitosis in budding yeast. Cell 1991; 66:519-531.
110. Taylor SS, McKeon F. Kinetochore localization of murine Bub1 is required for normal mitotic
timing and checkpoint response to spindle damage. Cell 1997; 89:727-35.
111. Hyman AA, Karsenti E. Morphogenetic properties of microtubules and mitotic spindle assembly.
Cell 1996; 84:401-10.
112. Rieder CL, Cole RW, Khodjakov A et al. The checkpoint delaying anaphase in response to chromosome
monoorientation is mediated by an inhibitory signal produced by unattached kinetochores. J
Cell Biol 1995; 130:941-8.
113. Gorbsky GJ, Ricketts WA. Differential expression of a phosphoepitope at the kinetochores of moving
chromosomes. J Cell Biol 1993; 122:1311-21.
114. McIntosh JR. Structural and mechanical control of mitotic progression. Cold Spring Harb Symp
Quant Biol 1991; 56:613-9.
115. Nicklas RB, Ward SC, Gorbsky GJ. Kinetochore chemistry is sensitive to tension and may link
mitotic forces to a cell cycle checkpoint. J Cell Biol 1995; 130:929-39.
116. Campbell MS, Daum JR, Gersch MS et al. Kinetochore “memory” of spindle checkpoint signaling
in lysed mitotic cells. Cell Motil Cytoskeleton 2000; 46:146-56.
117. Waters JC, Chen RH, Murray AW et al. Mad2 binding by phosphorylated kinetochores links error
detection and checkpoint action in mitosis. Curr Biol 1999; 9:649-52.
118. Hoyt MA, Totis L, Roberts BT. S. cerevisiae genes required for cell cycle arrest in response to loss
of microtubule function. Cell 1991; 66:507-17.
119. Weiss E, Winey M. The Saccharomyces cerevisiae spindle pole body duplication gene MPS1 is part
of a mitotic checkpoint. J Cell Biol 1996; 132:111-23.
120. Waters JC, Chen RH, Murray AW et al. Localization of Mad2 to kinetochores depends on
microtubule attachment, not tension. J Cell Biol 1998; 141:1181-91.
104 Cell Cycle Checkpoints and Cancer
121. Li Y, Gorbea C, Mahaffey D et al. MAD2 associates with the cyclosome/anaphase-promoting complex
and inhibits its activity. Proc Natl Acad Sci USA 1997; 94:12431-6.
122. Hwang LH, Lau LF, Smith DL et al. Budding yeast Cdc20: a target of the spindle checkpoint.
Science 1998; 279:1041-4.
123. Howell BJ, Hoffman DB, Fang G et al. Visualization of Mad2 dynamics at kinetochores, along
spindle fibers, and at spindle poles in living cells. J Cell Biol 2000; 150:1233-1250.
124. Hardwick KG, Weiss E, Luca FC et al. Activation of the budding yeast spindle assembly checkpoint
without mitotic spindle disruption. Science 1996; 273:953-6.
125. Hardwick KG, Murray AW. Mad1p, a phosphoprotein component of the spindle assembly checkpoint
in budding yeast. J Cell Biol 1995; 131:709-20.
126. Chen RH, Shevchenko A, Mann M et al. Spindle checkpoint protein Xmad1 recruits Xmad2 to
unattached kinetochores. J Cell Biol 1998; 143:283-95.
127. Roberts BT, Farr KA, Hoyt MA. The Saccharomyces cerevisiae checkpoint gene BUB1 encodes a
novel protein kinase. Mol Cell Biol 1994; 14:8282-91.
128. Taylor SS, Ha E, McKeon F. The human homologue of Bub3 is required for kinetochore localization
of Bub1 and a Mad3/Bub1-related protein kinase. J Cell Biol 1998; 142:1-11.
129. Hardwick KG, Johnston RC, Smith DL et al. MAD3 encodes a novel component of the spindle
checkpoint which interacts with Bub3p, Cdc20p, and Mad2p. J Cell Biol 2000; 148:871-82.
130. Nasmyth K. Separating sister chromatids. Trends Biochem Sci 1999; 24:98-104.
131. Rieder CL, Khodjakov A, Paliulis LV et al. Mitosis in vertebrate somatic cells with two spindles:
implications for the metaphase/anaphase transition checkpoint and cleavage. Proc Natl Acad Sci
USA 1997; 94:5107-12.
132. Hoyt MA. Exit from mitosis: spindle pole power. Cell 2000; 102:267-70.
133. Li R. Bifurcation of the mitotic checkpoint pathway in budding yeast. Proc Natl Acad Sci USA
1999; 96:4989-94.
134. Gardner RD, Burke DJ. The spindle checkpoint: Two transitions, two pathways. Trends Cell Biol
2000; 10:154-8.
135. Burke DJ. Complexity in the spindle checkpoint. Curr Opin Genet Dev 2000; 10:26-31.
136. Taylor SS. Chromosome segregation: Dual control ensures fidelity. Curr Biol 1999; 9:R562-4.
137. Alexandru G, Zachariae W, Schleiffer A et al. Sister chromatid separation and chromosome reduplication
are regulated by different mechanisms in response to spindle damage. Embo J 1999;
18:2707-21.
138. Fraschini R, Formenti E, Lucchini G et al. Budding Yeast Bub2 Is Localized at Spindle Pole
Bodies and Activates the Mitotic Checkpoint via a Different Pathway from Mad2. J Cell Biol
1999; 145:979-991.
139. Wang Y, Burke DJ. Checkpoint genes required to delay cell division in response to nocodazole
respond to impaired kinetochore function in the yeast Saccharomyces cerevisiae. Mol Cell Biol 1995;
15:6838-44.
140. Tavormina PA, Burke DJ. Cell cycle arrest in cdc20 mutants of Saccharomyces cerevisiae is independent of
Ndc10p and kinetochore function but requires a subset of spindle checkpoint genes. Genetics 1998;
148:1701-13.
141. Fesquet D, Fitzpatrick PJ, Johnson AL et al. A Bub2p-dependent spindle checkpoint pathway regulates the
Dbf2p kinase in budding yeast. Embo J 1999; 18:2424-34.
142. Jaspersen SL, Charles JF, Morgan DO. Inhibitory phosphorylation of the APC regulator Hct1 is
controlled by the kinase Cdc28 and the phosphatase Cdc14. Curr Biol 1999; 9:227-36.
143. Visintin R, Craig K, Hwang ES et al. The phosphatase Cdc14 triggers mitotic exit by reversal of
CDK-dependent phosphorylation. Mol Cell 1998; 2:709-18.
144. Visintin R, Hwang ES, Amon A. Cfi1 prevents premature exit from mitosis by anchoring Cdc14
phosphatase in the nucleolus. Nature 1999; 398:818-23.
145. Shou W, Seol JH, Shevchenko A et al. Exit from mitosis is triggered by Tem1-dependent release
of the protein phosphatase Cdc14 from nucleolar RENT complex. Cell 1999; 97:233-44.
146. de Almeida A, Raccurt I, Peyrol S et al. The Saccharomyces cerevisiae Cdc14 phosphatase is implicated
in the structural organization of the nucleolus. Biol Cell 1999; 91:649-63.
147. Bardin AJ, Visintin R, Amon A. A mechanism for coupling exit from mitosis to partitioning of the
nucleus. Cell 2000; 102:21-31.
148. Pereira G, Hofken T, Grindlay J et al. The Bub2p spindle checkpoint links nuclear migration with
mitotic exit. Mol Cell 2000; 6:1-10.
149. Daum JR, Gomez ON, Winey M et al. The spindle checkpoint of saccharomyces cerevisiae
responds to separable microtubule-dependent events. Curr Biol 2000; 10:1375-8.
150. Jaspersen SL, Charles JF, Tinker KR et al. A late mitotic regulatory network controlling cyclin
destruction in Saccharomyces cerevisiae. Mol Biol Cell 1998; 9:2803-17.
DNA Damage-Independent Checkpoints from Yeast to Man 105
151. Toyn JH, Johnston LH. Spo12 is a limiting factor that interacts with the cell cycle protein kinases
Dbf2 and Dbf20, which are involved in mitotic chromatid disjunction. Genetics 1993; 135:963-71.
152. Cenamor R, Jimenez J, Cid VJ et al. The budding yeast Cdc15 localizes to the spindle pole body
in a cell-cycle-dependent manner. Mol Cell Biol Res Commun 1999; 2:178-84.
153. Xu S, Huang HK, Kaiser P et al. Phosphorylation and spindle pole body localization of the Cdc15p
mitotic regulatory protein kinase in budding yeast. Curr Biol 2000; 10:329-32.
154. Jaspersen SL, Morgan DO. Cdc14 activates cdc15 to promote mitotic exit in budding yeast. Curr
Biol 2000; 10:615-8.
155. Frenz LM, Lee SE, Fesquet D et al. The budding yeast Dbf2 protein kinase localises to the centrosome and
moves to the bud neck in late mitosis. J Cell Sci 2000; 3399-3408.
156. Cohen-Fix O, Koshland D. Pds1p of budding yeast has dual roles: inhibition of anaphase initiation
and regulation of mitotic exit. Genes Dev 1999; 13:1950-9.
157. Neff MW, Burke DJ. A delay in the Saccharomyces cerevisiae cell cycle that is induced by a dicentric
chromosome and dependent upon mitotic checkpoints. Mol Cell Biol 1992; 12:3857-64.
158. Wang Y, Hu F, Elledge SJ. The Bfa1/Bub2 GAP complex comprises a universal checkpoint
required to prevent mitotic exit. Curr Biol 2000; 10:1379-82.
159. Schaerer BC, Riezman H. Saccharomyces cerevisiae Arc35p works through two genetically separable
calmodulin functions to regulate the actin and tubulin cytoskeletons. J Cell Sci 2000; 521-32.
160. Spang A, Geissler S, Grein K et al. gamma-Tubulin-like Tub4p of Saccharomyces cerevisiae is
associated with the spindle pole body substructures that organize microtubules and is required
for mitotic spindle formation. J Cell Biol 1996; 134:429-41.
161. Cahill DP, Lengauer C, Yu J et al. Mutations of mitotic checkpoint genes in human cancers.
Nature 1998; 392:300-3.
162. Li Y, Benezra R. Identification of a human mitotic checkpoint gene: hsMAD2. Science 1996;
274:246-8.
163. Jin DY, Spencer F, Jeang KT. Human T-cell leukemia virus type 1 oncoprotein Tax targets the
human mitotic checkpoint protein MAD1. Cell 1998; 93:81-91.
164. Cahill DP, da CL, Carson WE et al. Characterization of MAD2B and other mitotic spindle
checkpoint genes. Genomics 1999; 58:181-7.
165. Yamaguchi K, Okami K, Hibi K et al. Mutation analysis of hBUB1 in aneuploid HNSCC and
lung cancer cell lines. Cancer Lett 1999; 139:183-7.
166. Gemma A, Seike M, Seike Y et al. Somatic mutation of the hBUB1 mitotic checkpoint gene in
primary lung cancer. Genes Chromosomes Cancer 2000; 29:213-218.
167. Ohshima K, Haraoka S, Yoshioka S et al. Mutation analysis of mitotic checkpoint genes (hBUB1
and hBUBR1) and microsatellite instability in adult T-cell leukemia/lymphoma. Cancer Lett 2000;
158:141-150.
168. Futamura M, Arakawa H, Matsuda K et al. Potential role of BRCA2 in a mitotic checkpoint after
phosphorylation by hBUBR1. Cancer Res 2000; 60:1531-5.
169. McCabe CJ, Gittoes NJ. PTTG-a new pituitary tumour transforming gene. J Endocrinol 1999;
162:163-6.
170. Zhang X, Horwitz GA, Heaney AP et al. Pituitary tumor transforming gene (PTTG) expression in
pituitary adenomas. J Clin Endocrinol Metab 1999; 84:761-7.
171. Zou H, McGarry TJ, Bernal T et al. Identification of a vertebrate sister-chromatid separation
inhibitor involved in transformation and tumorigenesis. Science 1999; 285:418-22.
172. Prezant TR, Kadioglu P, Melmed S. An intronless homolog of human proto-oncogene hPTTG is
expressed in pituitary tumors: evidence for hPTTG family. J Clin Endocrinol Metab 1999; 84:1149-
52.
173. Dobles M, Liberal V, Scott ML et al. Chromosome missegregation and apoptosis in mice lacking
the mitotic checkpoint protein Mad2. Cell 2000; 101:635-45.
174. Kalitsis P, Earle E, Fowler KJ et al. Bub3 gene disruption in mice reveals essential mitotic spindle
checkpoint function during early embryogenesis. Genes Dev 2000; 14:2277-2282.
106 Cell Cycle Checkpoints and Cancer
CHAPTER 6
Cell Cycle Checkpoints and Cancer, edited by Mikhail V. Blagosklonny. ©2001 Eurekah.com.
The Regulation of p53 Growth Suppression
Ronit Vogt Sionov, Igal Louria Hayon and Ygal Haupt
Abstract
The p53 tumor suppressor protein plays a pivotal role in the cellular response to stress. A
variety of stress signals trigger accumulation and activation of p53 to halt the cell cycle
and to prevent replication of damaged DNA. The p53 protein is required for a proper
G1 arrest, it is essential for maintaining the G2 arrest, and it contributes to the mitotic spindle
checkpoint. p53 exerts these actions by inducing multiple target genes. Under defined conditions,
p53 induces programmed cell death by mechanisms that are partially understood and
involve a combination of transcriptional dependent and –independent activities. The choice
between arrest and cell death depends on the final integration of antagonistic signals. These
include the type and intensity of the stress signal, the spectrum of the target genes induced, the
type of cell and its oncogenic status, and the presence of growth and survival factors. The stability
of the p53 protein and its activities are tightly regulated by many factors among which the Mdm2
proto-oncoprotein is the central player. Inhibitory effects of Mdm2 on p53 stability and activities
are modulated by multiple mechanisms including post-translational modifications of p53 and
Mdm2 and by other interacting proteins. Importantly, p53 is also regulated at the level
of its sub-cellular localization. Sequestration of p53 into the cytoplasm is sufficient for
its inhibition. In contrast, accumulation of p53 in the nucleus induces its transcriptional
activity. This activity can be further enhanced by specific post-translational modifications
and by recruitment of p53 into nuclear bodies. We discuss current views on the
regulation of p53 and its growth inhibitory activities.
Introduction
The p53 tumor suppressor protein plays a key role in the regulation of the cell cycle and
cell death. The p53 protein is also involved in cell differentiation, DNA repair, senescence and
angiogenesis.1-8 Wild type (wt) p53 and intact signaling pathways are essential for the prevention
of cancer, consistent with a high tumor incidence observed in p53 null mice9 and in p53-
heterozygous Li-Fraumeni patients.10 It is estimated that approximately one half of human
cancers contain a mutation in p53.11 It is predicted that in the majority of the remaining
tumors the p53 signaling pathway is inactivated by up-regulation of p53 inhibitors, such as Mdm2,
or by down-regulation of p53 cooperators, such as ARF.12,13 The increased predisposition to
tumor development in the absence of p53 is due to the accumulation of genetic alterations and
failure to eliminate these defective cells.14
Wt p53 is a labile protein with a short half-life. Accumulation and activation of the protein
can be triggered by a variety of stress signals including DNA damage, hypoxia, nucleotide
deprivation, viral infection, heat shock, and mitogenic or oncogenic activation.15-17 The specific
activity of p53 is further enhanced by post-translational modifications and by a variety of
positive and negative regulators.6,18,19 Activated p53 elicits cellular responses that ultimately
lead to growth arrest and/or programmed cell death (apoptosis).3,8 In this Chapter we will
The Regulation of p53 Growth Suppression 107
focus on the regulation of p53, the growth inhibitory activities of p53 and the cellular choice to
stall or die.
Regulation of p53
The p53 protein is subject to tight regulation at multiple levels. This is achieved by a
variety of positive and negative regulators, often creating feedback loops. Three major levels of
regulation are recognized: protein stability, protein activity, and subcellular distribution. New
information has broadened our understanding of how p53 is regulated under normal and stress
conditions, and the molecular mechanisms governing these regulatory processes are beginning
to emerge. It is beyond the scope of this review to discuss all aspects of p53 regulation. Instead,
we will focus on two major aspects: the regulation of p53 stability, with emphasis on the pivotal
role of Mdm2, and the regulation of p53 subcellular localization. Comprehensive reviews on
other aspects of p53 regulation, such as p53 post-translational modifications and their effects
on p53 activities have been recently published.6,18-20
Mdm-2-Negative Autoregulatory Feedback Loop
The major negative regulator of p53 is the Mdm2 proto-oncogene.12,21,22 Mdm2 is
transcriptionally induced by p53, thus p53 triggers its own destruction through a negative
feedback loop.12 This feedback loop leads to oscillations in the expression of both proteins
following DNA damage. These oscillations may enable a more effective execution of a reversible
p53 response.23 The importance of Mdm2 in this negative regulation is demonstrated by the
lethal effect during early stages of embryonic development of mdm2 knockout mice. The early
mortality of these mice is abolished by the simultaneous inactivation of p53.24,25 Overexpression
of Mdm2 in many cancers is often sufficient to inactivate p53 without further mutation.12
Similarly, mere loss of Mdm2 is sufficient to induce p53-mediated apoptosis in vivo.26 Therefore,
Mdm2 is critical for keeping p53 in check. The following sections discuss the regulation of p53
by Mdm2 and how this regulation is modulated.
Mdm2 Signals p53 for Proteasomal Degradation
The finding that Mdm2 promotes p53 degradation through the ubiquitin-proteasome
system27,28 was central to our understanding of how p53 stability is regulated. It also helped to
explain the elevated expression of p53 mutant proteins in cancer cells, which results from their
inability to induce mdm2 expression.29 Reduced affinity of p53 mutants for Mdm2 may
contribute to this effect.30 Mdm2 promotes p53 for degradation by acting as an E3 ubiquitin
ligase,31 even though it does not contain the HECT (Homologous to E6AP Carboxy Terminus)
domain. The RING-finger domain of Mdm2 is crucial for the E3 activity of Mdm2.32,33 The
oligomerization of p53 appears to be essential for the degradation of p53 by Mdm2.34,35
Removal of the last 30-40 C-terminal amino acids from p53 impairs its degradation by Mdm2,34
but has minimal effect on the extent of p53 ubiquitination.35 Within this region there are
several lysines that serve as potential ubiquitination sites. Substitution of the 6 lysines between
residues 370-386 markedly reduces the ubiquitination of p53 by Mdm2 and its susceptibility
to Mdm2-mediated degradation.36,37 In addition, two N-terminal regions have been recently
shown to affect the susceptibility of p53 for destabilization by Mdm2. The polyproline region
(residues 62-91) of p53 may provide protection for p53 from Mdm2. p53 lacking this region is
excessively sensitive to inhibition and destabilization by Mdm2.38 On the other hand, the
adjacent region (residues 92-112) appears to be required for the degradation of p53 by Mdm2.39
These findings suggest that additional proteins are probably involved in the ubiquitination of
p53 by Mdm2 in vivo. A possible candidate is JNK that has been implicated in the ubiquitination
and degradation of p53 in unstressed cells, and binds p53 between residues 97 and 116.40
Inhibition of p53 Activities by Mdm2
Mdm2 binds the transactivation domain of p53 (residues 18-28) in a region important
for the interaction of p53 with components of the transcription machinery, such as TBP and its
108 Cell Cycle Checkpoints and Cancer
associated factors (TAFs),12 and with its transcriptional coactivator p300.41 This observation
prompted the ‘masking model’, whereby the binding of Mdm2 to p53 conceals its transactivation
domain.42 It is difficult to distinguish between effects of Mdm2 on p53 stability from its effect
on p53 activity. Several studies have suggested that Mdm2 can inhibit p53-mediated apoptosis
independent of protein degradation.38,43,44 This has been demonstrated by flow cytometric
measurement of the inhibitory effect of Mdm2 on p53 apoptotic activity, without reduction in
p53 stability. This notion is supported by the fact that the Mdm2 analogue MdmX (Mdm4)
inhibits p53 transcriptional activation without promoting its degradation.45,46 Similarly, the
inhibition of p53-dependent transcription by Mdm2 may not require p53 degradation.47 In
apparent conflict, recent studies suggested that in the absence of p53 ubiquitination, Mdm2 is
unable to block p53 transcriptional activity,37 and the degradation of p53 is obligatory for
blocking p53 transrepression and apoptotic activities.47 These findings are difficult to reconcile,
especially because both p53 and Mdm2 were overexpressed and different ratios of these
proteins were used. This important question needs to be addressed at physiological levels of
p53 and Mdm2.
Modulation of the p53/Mdm2 Autoregulatory Loop
In order to exert its growth inhibitory activities, p53 degradation by Mdm2 should be
regulated. Modulations of the autoregulatory feedback loop involving Mdm2 and p53 have
been the subject of intensive study. In the following sections, we discuss the major mechanisms
governing this important regulation.
Post-translational Modifications
The Mdm2-p53 interaction is obligatory for p53 degradation.27,28 Preventing this
interaction is sufficient to protect p53 from degradation.48-51 Following various stress stimuli,
such as γIR and UV, this interaction is interrupted. How is this interaction regulated in
response to stress? Phosphorylation of p53 on serine 20 (Ser20), which resides within the Mdm2
binding domain, reduces the affinity of Mdm2 for p53,44 and enhances the stabilization of p53 in
response to γIR and UV light.52,53 Checkpoint kinase 2 (Chk2) was identified as the kinase
responsible for phosphorylation of Ser-20,54 without which the stabilization of p53 in response
to DNA damage is abrogated.55 Hence, a cascade of DNA damage signaling from ATM to
Chk2 to p53 phosphorylation leads to the accumulation of p53. In response to γIR, p53 is also
phosphorylated on Thr18 by a CKI-like kinase, a modification that reduces the binding affinity
between p53 and Mdm2.56,57 However, the physiological relevance of the latter modification is
yet to be explored.
In addition to these modifications, p53 is subjected to multiple phosphorylations at the
N- and C- termini (Table 1). While modifications such as phosphorylation of Ser-15 or
Ser-37,57,58 do not appear to regulate p53 stability, they play important regulatory roles in p53
binding to DNA and p53 transcriptional activity (Table 1; reviewed in6,18,19). Similarly, acetylation
of p53 on lysine residues within its C-terminus enhances its transcriptional activity
(Table 1). Finally, p53 undergoes sumoylation, a modification that mildly enhances its
activities without affecting p53 stability.59-61
The Mdm2 protein is also subjected to modifications that may affect its ability to regulate
p53. For example, the sumoylation of Mdm2, which is reduced in response to DNA damage,
enhances its ability to ubiquitinate p53.62 Further, Mdm2 undergoes rapid phosphorylation in
vivo by ATM in response to γIR or radiomimetic drugs. 63 The effect of this modification on the
auto-regulatory loop is yet unclear. In response to DNA damage, Mdm2 is also phosphorylated at
multiple sites by DNA-PK. This phosphorylation impairs p53/Mdm2 interaction,64 but its
relevance in vivo is currently unknown. The combined modification of Mdm2 and p53 may
have synergistic effects on the auto-regulatory loop in response to stress signals.
The Regulation of p53 Growth Suppression 109
Table 1. Posttranslational modifications of p53 and their effects on p53
Modifying enzymes Stress signal p53 modifications Effectson p53 References
A. Phosphorylation
ATM IR Ser-15 Enhanced SST 169,209,210
ATR IR, UV Ser-15, Ser-37 Enhanced SST 170
DNA-PK IR, UV Ser-15,Ser-37 Enhanced SST 168
Chk1 IR, UV Ser-20 Reduced Mdm2 binding 211
Chk2 IR, UV Ser-20 Reduced Mdm2 binding 54,55
JNK IR, UV Ser-33 Enhanced stability 212
CAK ND Ser-33, Ser-376, Ser-378 Enhanced SST 126,213
p38 UV Ser-15, Ser-33, Ser-46, Ser-392 Enhanced SST and stability 214-216
CKII UV Ser-392 Enhanced SST 217
PKR IFN, dsRNA Ser-392 Enhanced SST 218
CKI IR Thr-18 Reduced Mdm2 binding 49, 57
CyclinA-CDK2 ND Ser-315 Enhanced SST 219-220
PKC UV Ser-371, Ser-376, Ser-378 Enhanced SST 6
B. Dephosphorylation
Unknown IR Ser-376 Increase binding to 14-3-3 221
Cdc14 ND Ser-315 Unknown 222
C. Acetylation
p300/CBP IR, UV Lys-373, Lys-382 Enhanced SST and stability 166,223
PCAF UV Lys-320 Enhanced SST 166
D. Sumoylation
Unknown enzyme ND Lys-386 Enhanced SST 59-61
E. Ribosylation
PARP IR Poly(ADP)ribosylation Enhanced SST and stability 224-225
Abbreviation: SST – Sequence specific transactivation, IR – ionizing radiation, ND – not determined,
IFN – interferon, dsRNA – double stranded RNA
110 Cell Cycle Checkpoints and Cancer
Involvement of Interacting Proteins
The majority of the positive regulators of p53 can be classified into two general groups.
The first group consists of proteins that activate and stabilize p53 by neutralizing the inhibitory
effects of Mdm2 (e.g., ARF, c-Abl), and their contribution will be discussed below. The
second group comprises proteins that bind the C-terminus of p53 and activate it by relieving p53
from the inhibitory effect of this region on DNA binding or by stabilizing the p53 tetramer
(e.g., c-Abl, BRCA-1, 14-3-3-σ, Ref-1).
The p53-ARF axis is critical for eliminating potential tumor cells containing deregulated
oncogene expression.13,65 Either ARF or p53 is frequently inactivated during tumorigenesis.66
However, the tumor spectrum differs in mice lacking either ARF or p53. Whereas p53-null
mice develop predominantly T-cell lymphomas (~70%), ARF-null mice exhibit primarily poorly
differentiated sarcomas (~50%).9,67 This indicates that ARF is selectively activated by a subset
of stress signals that potentially activate p53. ARF is activated by deregulation of proto-oncogenes,
such as E2F-1, by mitogenic oncogenes such as c-myc or Ras and by the adenoviral E1A.13 In turn,
ARF triggers p53-dependent growth arrest in G1 and G2 phases. In the presence of
appropriate signals, ARF sensitizes cells to apoptosis in a p53-dependent manner.68 The activation
of p53 by ARF is auto-regulated by a feedback loop. p53 down-regulates the expression of ARF
by directly suppressing its promoter69 and by blocking E2F-1 activation which induces ARF
expression.70,71 ARF activates p53 by neutralizing Mdm2-mediated ubiquitination and
degradation of p53.71-73 ARF sequesters Mdm2 into the nucleolus, thereby preventing the nuclear
export of p53.74-76 Two regions within ARF, the N-terminus and an Arg-rich C-terminus,
contribute to the nucleolar import of Mdm2.77,78 It is proposed that the binding of ARF to
Mdm2 unmasks a cryptic nucleolar localization signal (NrLS; residues 466-473), which resides
within the RING finger of Mdm2. This sequence promotes shuttling of the ARF/Mdm2 complex
to the nucleolus.78,79
Another cooperator of p53 is c-Abl, a stress activated protein with multiple effects on cell
cycle regulation.80,81 Mice deficient for c-abl exhibit retarded growth,82,83 whereas mice doubly
deficient for p53 and c-abl are nonviable.84 Under nonstressed conditions, c-Abl is required for
the proliferation of cells that lack p53, but not normal cells.84 These effects of c-Abl are poorly
understood. On the other hand, in response to genotoxic stress c-Abl induces growth arrest or
apoptosis (reviewed in references 80,81,85). c-Abl enhances the transcriptional activity of p53,
and by binding to its C-terminus it stabilizes the specific interaction of p53 with DNA.86 We
have previously shown that cooperation between c-Abl with p53 is achieved by neutralizing the
inhibitory effects of Mdm2 on p53 activity and stability.87 These inhibitory activities of Mdm2 are
also neutralized by other proteins. The retinoblastoma protein stabilizes p53 and relieves its apoptotic
activity from inhibition by Mdm2.88 The catenin protein, which is often deregulated in colon
cancer, accumulates p53 and enhances its transcriptional activity. This presumably serves as a
safeguard to protect cells from deregulated oncogenic expression.89 While proteins such as ARF,
protect p53 in response to mitogenic and oncogenic signals, other proteins, such as c-Abl, stabilize
p53 in response to DNA damage.
Regulation of Intracellular Distribution of p53
The p53 protein shuttles between the cytoplasmic and the nuclear compartments in a cell
cycle-dependent fashion.98,99 The accumulation of p53 in the nucleus is crucial for its tumor
suppressive activity. Prevention of nuclear accumulation provides an efficient mechanism by
which tumor cells may continue to proliferate in the presence of wt p53. Indeed, cytoplasmic
sequestration of p53 has been commonly observed in certain tumors, such as neuroblastomas,
breast and colon cancer.90-92 In at least a subset of these tumors, Mdm2 is responsible for the
cytoplasmic accumulation of p53.93 Several viral proteins also influence p53 localization. The E6
protein promotes the nuclear export of p53 in HPV-infected cervical carcinomas,94 whereas the
adenoviral E1B 55kD protein and the hepatitis B virus HBx protein keep p53 in cytoplasmic
structures.95-97 In addition, defects in the import/export machinery of p53 may alter its distribuThe
Regulation of p53 Growth Suppression 111
tion. For example, truncation of importin-α was identified in breast cancer and this modification
blocks the nuclear import of p53.90 Therefore, abrogating the appropriate sub-cellular distribution
of p53 is an important mechanism for eliminating p53 functions in cancer development.
The regulatory mechanism governing p53 import/export has begun to emerge only
recently. The import of p53 into the nucleus is an active process involving the association of the
importin complex with p53 via the nuclear localization signals (NLS). The major NLSI (residues
316-325) mediates the interaction between p53 and importin- and it is essential for the
import.90 Two minor sites, NLSII (residues 369-375) and NLSIII (residues 379-384), facilitate
this import.100 In addition to these sites, two adjacent basic residues, K305 and R306, are
important for import, while the spacer between them and the NLSI (residues 326-355) is
important for the cytoplasmic sequestration of p53.101 Presumably this cytoplasmic region serves
as an anchor for attachment by cytoplasmic factors that remain to be identified. Alternatively,
protein binding to this region masks the NLSI.100 p53 may be sequestered in the cytoplasm by
associating with one or more of cytoplasmic proteins such as tubulin, hsc70, hsc84, F-actin
and vimentin.102 A recent study revealed that p53 is associated with microtubules and is
transported to the nucleus by dynein, a motor protein. This facilitates nuclear accumulation of
p53 in response to DNA damage.103
The nuclear export of p53 is essential for its proteasomal degradation.104,105 Blocking p53
nuclear export by leptomycin B is sufficient to accumulate p53 in its active form in the
nucleus.104,106 Leptomycin B inhibits the nuclear exporting protein CRM1, which binds the
nuclear export signals (NES) of proteins.107 Yet the regulation of p53 nuclear export is still
controversial. It is generally accepted that Mdm2 promotes the nuclear export of p53.
However, it is debatable whether the NES of Mdm2,105 the NES of p53,108,109 or both are
responsible for p53 export. Intriguingly, the NES of p53 resides within the p53 tetramerization
domain. This raises the attractive model that NES is masked as long as p53 remains tetrameric, but
becomes unmasked when p53 acquires dimeric or monomeric form.109 However,
oligomerization of p53 increases its affinity to Mdm2, and oligomerized p53 is ubiquitinated
more efficiently than monomeric or oligomerization-defective mutant.34,35,110 Nuclear
ubiquitination of p53 seems to be a prerequisite for its nuclear export because Mdm2 lacking the
RING finger harboring the E3 ligase activity is unable to promote p53 export.111,112 This suggests
that the ubiquitination of p53 exposes NES allowing p53 to interact with CRM1. Consequently,
p53 is exported to the cytoplasm (Fig. 1). This may be a simplistic model, in
particular because multiple lysines within the regulatory C-terminal region of p53 need to be
ubiquitinated by Mdm2 for degradation.36,37 Further, this model challenges the dogma that
the ubiquitination of p53 by Mdm2 is important primarily for signaling p53 degradation. It is
possible that the ubiquitination of p53 is primarily aimed at promoting export of p53 to the
cytoplasm. Whether p53 is stored in a cytoplasmic pool, processed by the 26S proteasome or
recycled back to the nucleus presumably depends on the regulation of p53 in the cytoplasm.
In the nucleus, p53 may be sequestered into small structures termed promyelocytic leukemia
protein-nuclear bodies (PML-NB)113,114 (also known as PML oncogenic domains [PODs] and
nuclear domain 10 [ND10]). These structures contain several proteins, including PML, Sp100,
Sumo-1, p300/CBP, and HMG1, and are thought to be involved in transcriptional regulation.115,116
Some forms of PML (e.g., PML3) bind and sequester p53 to the PML-NBs,
enhancing transcriptional activity of p53.113,114 This regulation is important for a full p53
response to stress. In response to γIR, transcriptional and DNA binding activities of p53 are
impaired in PMLγIR primary cells.114 This explains thetolerance of PML-/- mice to lethal doses
of γIR.117 Moreover, oncogenic Ras upregulates PML and promotes the formation of a trimeric
p53-PML-p300/CBP complex within PML-NB,118 enhancing acetylation of p53 on Lys-
382 by p300/CBP.118 Thus, PML may activate p53 by recruiting its coactivator p300. Since
overexpression of PML leads to apoptosis,116 this protein may play a central regulatory role in
p53 response to stress.
112 Cell Cycle Checkpoints and Cancer
p53-Mediated Growth Regulatory Functions
Once activated, p53 triggers either growth arrest or apoptosis. Factors that influence this
decision are discussed below. p53 is crucial for the induction of growth arrest by numerous
stress signals, but it is dispensable for cell cycling under normal conditions.14 The following
sections describe the mechanisms of p53-mediated growth arrest.
p53-Mediated Cell Cycle Arrest
Activation of p53 may lead to growth arrest at both G1 and G2 phases of the cell cycle.
Arrest in G1 prevents replication of damaged DNA, while arrest in G2 prevents improper
segregation of chromosomes. p53 may also arrest DNA replication in S phase, which is usually
masked by the prior G1 arrest.119 Moreover, p53 is involved in a mitotic spindle checkpoint
preventing endoreduplication of 4N cells.120 The ability of p53 to arrest cells at multiple
checkpoints is crucial for suppression of amplification of genetic alterations which otherwise
can lead to cancer. Inactivation of wt p53 results in a loss of the DNA damage-induced G1/S
Fig. 1. A model for the regulation of p53 sub-cellular distribution. p53 is imported into the nucleus by
interaction with microtubulin and dynein motor protein. This import is mediated through interaction of
p53 nuclear localization signal (NLS) with importin-α. Following certain signals, p53 is transported into
the PML-nuclear bodies (PML-NB) where it is further activated. Mdm2-mediated ubiquitination of p53
in the nucleus may expose the nuclear export signal (NES) of p53, thereby allowing its nuclear export via
interaction with CRM1. In the cytoplasm p53 is degraded by the 26S proteasome system. Whether all the
cytoplasmic p53 pool is being degraded, or some is recycled back into the nucleus, is not known. Activation
of ARF by oncogenic stimuli protects p53 from Mdm2 by recruiting Mdm2 into the nucleolus. p53 is also
protected from Mdm2 by c-Abl following DNA damage.
The Regulation of p53 Growth Suppression 113
checkpoint,121 impaired G2 arrest,122 and the appearance of aneuploid and polyploid cells.120
While the pathway leading to G1 arrest is well established, the pathway leading to G2 arrest is
beginning to emerge (discussed in this book by Stewart and Pietenpol).
p53-Mediated G1-Arrest
The p53-target gene, p21 (waf-1/cip-1), is the key player in G1 arrest. p21 inhibits different
complexes of cyclin/cyclin-dependent kinases (CDKs) (Cyclin D-CDK4/6 and Cyclin A,
E-CDK2) that sequentially phosphorylate the retinoblastoma (pRb) protein, and as a result
release the S phase-promoting E2F-1 transcription factor.123 Cells deficient in p21 show aberrant
G1 arrest following radiation.124 Unlike p53 null mice, those lacking p21 develop normally,
exhibit normal apoptotic response and are not susceptible to spontaneous malignancies.124 p53
also promotes G1 arrest by directly inhibiting the activity of the CDK-activating kinase (CAK)
complex CDK7/CyclinH1/Mat1, which activates cyclin A-CDK2 by phosphorylation.125 CAK
is also a component of the TFIIH transcription factor complex controlling the transcriptional
activity of RNA polymerase II. Hence, binding of p53 to CAK results in a strong reduction of
Fig. 2. Induction of G1 cell cycle arrest by p53. The activation of p21 is central for mediating G1 arrest by
inhibition of multiple cyclin/CDK complexes. The induction of PC3 and inhibition of CAK activity
contribute to this effect. As a result, pRb is not phosphorylated and it inactivates E2F-1 through recruitment
of histone deacetylase (HDAC1).
114 Cell Cycle Checkpoints and Cancer
CAK activity towards both CDK2 and the C-terminal repeat domain (CTD) of RNA
polymerase II.125,126 In addition, p53 may activate pRb via PC3 (TIS21,BTG2), a newly
identified p53-target gene. The PC3 gene product promotes accumulation of hypophosphorylated
pRb by reducing Cyclin D1 protein level and thereby inhibiting CDK4 activity.127 These pathways
are summarized in Fig. 2.
p53-Mediated G2-Arrest
Activation of p53 can promote and maintain G2 arrest. This depends on functional pRb128
and is mediated by several target genes (Fig. 3). Acting in concert, p53 and its target genes
efficiently inhibit the CyclinB1/Cdc2 activity, which is essential for cells to enter mitosis. p21
inhibits the activity of the CyclinB1/Cdc2 complex.128 GADD45 binds Cdc2 and disrupts its
ability to complex with Cyclin B.129,130 The importance of GADD45 is manifested in GADD45-/- mice
which exhibit both genetic instability, failure of G2 arrest and centrosome amplification.131
The 14-3-3-σ protein sequesters and inhibits the phosphorylated form of Cdc25C.132,133 The
Cdc25C phosphatase dephosphorylates and thereby activates the CyclinB/Cdc2 complex.134
Also, 14-3-3-σ sequesters Cdc2 in the cytoplasm, preventing it from translocating into the
nucleus in the late G2.135 These effects of p21, GADD45 and 14-3-3-σ are further compounded
by the transcriptional repression of cdc2 and cyclin B1 by p53.136-138 Although the
role of p53 in triggering G2 arrest is still unclear, its ability to induce the expression of p21 and
14-3-3-σ is essential for sustaining this arrest.122,135,139
Fig. 3. Contribution of p53 to G2 cell cycle arrest. p53 triggers several parallel pathways that block the
formation of the mitotic cyclinB/Cdc2 complex and inhibit its activity. The activation of Chk1 by ATM
is also important for this effect. Defects in one of these pathways cause premature entry into M phase. This
activity of p53 is essential for maintaining the G2 arrest.
The Regulation of p53 Growth Suppression 115
Mitotic Spindle Checkpoint
Loss of p53 function leads to genomic instability, abnormal centrosome duplication, and
formation of aneuploid and polyploid cells. This suggests that p53 plays a role in the control of
centrosome duplication and normal chromosomal segregation.120,140,141 The mitotic spindle
checkpoint proceeds normally in the absence of p53, but some responses to the mitotic spindle
damage requires p53.120,142 Induction of G1-like arrest prevents reentry into the S-phase, thereby
avoiding 4N cell formation. This process depends on p21 which inhibits CyclinE-CDK2,143 a
complex that drives initiation of centrosome duplication and when constitutively expressed it
uncouples the centrosome duplication from DNA replication.144,145 This is consistent with the
tendency of p21 deficient cells to undergo endoreduplication.146-148 The GADD45-/- cells also
show genomic instability, aneuploidy and centrosome amplification,131 suggesting a complementary
role for GADD45. This is consistent with the idea that proper G2 checkpoint is
required for preventing premature entry into mitosis.
A cooperative role may exist between p53 and its interacting protein BRCA-1 in regulating
chromosomal segregation. Both proteins associate with centrosomes in mitosis and bind to
γ-tubulin.103,120,149 BRCA-1 null cells resemble p53 null cells in genomic instability and
abnormal centrosome duplication.150,151 BRCA-1 deficient cells posses a defective G2/M checkpoint,
152 indicating the role of G2 checkpoint for proper chromosome segregation.
p53-Mediated Apoptosis
The apoptotic activity of p53 is crucial for eliminating defective and potentially carcinogenic
cells. Accumulating evidence demonstrates that p53 induces apoptosis by multiple
pathways.3,8,153 While growth arrest largely depends on p53 target genes, a full apoptotic response
requires both transcription and transcription-independent functions including repression
of survival-promoting genes. p53-mediated apoptosis involves generation of reactive oxygen
species (ROS),154,155 and depolarization of the mitochondrial electropotential gradient
(ΔΨm), thus releasing caspase-9 and Apaf-1 which trigger apoptosis through the activation of
the caspase cascade.156,157 The topic of p53-mediated apoptosis is covered elsewhere in this
book and has been reviewed previously.3,8,158
The Choice Between Growth Arrest and Apoptosis
As discussed above, p53 is activated by a complex array of signals and in turn induces
multiple biological responses, predominantly growth arrest and apoptosis. The choice between
these responses is influenced by a variety of extracellular and intracellular signals. The major
factors are outlined in the following sections.
The Type and Intensity of the Stress Signal
UV radiation and γIR cause different types of DNA damage, leading to different cellular
responses.6 For instance, the osteosarcoma U2OS cells undergo apoptosis in response to UV
but undergo growth arrest upon γIR.159 The accumulation of p53 in response to γIR is rapid
and transient, whereas in response to UV it is slow, intense and persists for long period.159-161
These differences can be explained, at least in part, by the different kinetics by which double
strand breaks (DSB) versus bulky adducts are being repaired. The nucleotide excision repair
(following UV) is slower than the DSB repair (following γIR).6 Moreover, UV and γIR trigger
different signal transduction pathways that ultimately lead to different pattern of post-translational
modifications of p53.6,18,19,162 This reflects differential activation of upstream kinases.
ATM is the critical upstream effector of p53 after γIR whereas p38 is essential for the UVinduced
p53 activation. ATR and DNA-PK phosphorylate p53 following both signals. Unlike
γIR, UV triggers significant phosphorylation of Ser-392163-165 and prolonged phosphorylation of
Ser-37.166 Acetylation of Lys-320 occurs early with UV but late with -IR.166 In contrast,
phosphorylation of Ser-15 and Ser-20 is faster after γIR than after UV.53,167-170 After UV
exposure, phosphorylation of Ser-46 appears to be a late event.167 Furthermore, high doses of
116 Cell Cycle Checkpoints and Cancer
UV are required to phosphorylate Ser-15 and 392, but low doses are sufficient for rapid
phosphorylation of Ser 9, 20 and 372.162 These differences are summarized in Table 1.
Contribution of all of these modifications ultimately affects the outcome of p53 response.
The Spectrum of Target Genes Induced by p53
A particular repertoire of target genes which are activated by p53 may determine the cell
fate. A different set of target genes are required for p53-mediated growth arrest and apoptosis.
The repertoire is affected by many factors including the pattern of post-transcriptional modification
of p53, type and strength of a stress signal, and expression of p53 regulatory proteins.
p53 binds to its responsive promoters with different affinity due to sequence heterogeneity
among the various responsive elements.171-173 Moreover, the sequence specific DNA binding
can be affected by changes in p53 conformation and by the conformation of cognate DNA
sequences.174,175 This is demonstrated by the identification of mutant p53 proteins, including
tumor-derived mutants, which manifest distinct promoter specificities. p53 mutants (e.g.,
143Ala, 120Arg, 175Pro, 181Leu, and 283His) that retain their ability to transactivate the
high affinity promoter of p21, but have lost the ability to transactivate the lower affinity promoter
of BAX, are of particular interest. As a result, these mutants induce growth arrest, but
they fail to promote apoptosis.8,29,176-178 The loss of binding to the BAX promoter can be
“corrected” in some of these mutants (e.g., 175Cys and 181Leu) by creating multiple binding
sites,172 suggesting that these mutations reduced the affinity for specific sites.
The promoter specificity of p53 may be further defined. The p53 mutant M246I retains
the ability to induce mdm2 but not PIG3,179 whereas p53 S121F mutant specifically activates
p53AIPI, but not p21, PIG3 or mdm2.167 The promoter specificity of p53S121F explains the
enhanced apoptotic activity of this p53 mutant.167,180 Similarly, variation in promoter-specificity
has been demonstrated in p53 mutant lacking the proline-rich region, which has impaired
apoptotic activity.179,181 The requirement for specific p53 conformation in the induction
of p53AIP1 is further supported by the observation that this pro-apoptotic gene is only
induced by p53 phosphorylated on Ser46.167 These data demonstrate that post-translational
modification of p53 can alter its promoter specificity. Finally, the promoter specificity may be
altered under specified cellular conditions. The Pw1/Peg3 gene is induced during p53/c-myc
or p53/E2F-1 mediated apoptosis, but neither during p53-mediated G1 growth arrest nor by
c-myc alone.182 Similarly, the TRAIL receptor KILLER/DR5 is induced during p53-dependent
apoptosis, but not during growth arrest.183 Mdm2 and p21 are induced by γIR but repressed
by UV.21,184
Cell Type-Dependence
Different cell types respond differently to the activation of p53 by DNA-damage. Whereas
T lymphocytes often undergo extensive apoptosis, fibroblasts undergo growth arrest.3,8 This
may reflect differential induction of relevant target genes. In lymphoid and myeloid cells but
not in fibroblasts, γIR induces BAX and rapid apoptosis.185
Thymus, spleen and intestine cells which undergo apoptosis in response to γIR express
similar levels of full-length (p90) and truncated (p76) Mdm2 186. However, testis, brain, heart
and kidney cells, which respond to γIR by growth arrest, express predominantly the p90 form.
The truncated form is induced by p53 in response to UV 187 and it antagonizes the inhibitory
activity of p90.186 Thus, the ratio of the two Mdm2 proteins may be decisive for the cellular
response to DNA damage.
UV-induced p53 response is enhanced in XP-A and Cockayne’s syndrome (CS-A; CS-B)
in which transcription-coupled repair is defective.188 Cells derived from these patients show
increased sensitivity to UV radiation with rapid and prolonged accumulation of p53.189,190
This may explain the increased susceptibility of these cells to p53-mediated apoptosis. Overall,
it is consistent with the observations that (i) extensive DNA damage promotes apoptosis rather
than growth arrest and (ii) high levels of p53 are required for apoptosis, whereas lower levels are
sufficient for growth arrest.191
The Regulation of p53 Growth Suppression 117
Deregulated Expression of Oncogenes
Activation of p53 in cells with intact stress signaling pathways, such as fibroblasts, causes
G1 arrest. On the other hand, cells with defective checkpoints (e.g., cells deficient in ATM or
pRb) tend to undergo apoptosis. Often in the latter cases E2F-1 is activated. Deregulated E2F-
1 cooperates with p53 to promote apoptosis192 through the induction of ARF13 and p73.193
Following p53 activation, fibroblasts overexpressing c-myc or E1A tend to undergo apoptosis,
whereas fibroblasts expressing Ras or ARF enter premature senescence, 68 reflecting differential
effects of these oncogenes on p53 activity. Both c-myc and E1A enhance p53 expression through
upregulation of ARF.13 In addition, E1A inhibits p53-induced mdm2 expression.194 On the
other hand, Ras causes upregulation of both ARF and Mdm2,13,195 which have antagonistic
effects on p53 expression levels. Consequently, unlike Myc and E1A, Ras does not induce p53
to the levels sufficient for triggering apoptosis. Similarly, loss of p21 or cleavage of the p21
protein renders cells prone to apoptosis in response to DNA damage.196-200 This effect is further
enhanced by the concurrent loss of 14-3-3-σ.139 Reconstitution of these cells with growth
arrest-promoting genes such as, p21, Rb, GADD45 and p53R2 can overcome apoptosis.200-203
p21 not only promotes growth arrest, but also antogonizes apoptosis by inhibiting the apoptosisregulating
kinase 1 (ASK1) and JNK1/SAPK.204 In overall, the status of the G1 and G2
checkpoints is critical for determining the cellular fate in response to stress.
Growth and Survival Factors
Induction of growth arrest rather than apoptosis is favored by the presence of survival factors,
such as cytokines and growth factors.8 Two major survival pathways have been implicated in
prevention of apoptosis. First, a cross talk exists between growth signaling pathways and members
of the Bcl-2 family. For instance, IL-3 dependent phosphorylation of Bad leads to its sequestration,
with subsequent release of free Bcl-2 and Bcl-XL.205 The second mechanism involves the
induction of mdm2 by growth factors, such as bFGF, IGF-1, and the thyroid hormone,206-208
and by the Ras pathway.195
Concluding Remarks
A big leap has been made towards a deeper understanding of how p53 is regulated at the
molecular level. The major mechanism for controlling the p53 response is at the level of protein
stability. The tight inhibition of p53 by Mdm2 has to be relieved in order for p53 to accumulate
and be activated. It is difficult to estimate the relative contribution of protein destabilization
versus inhibition of activities, for the overall negative effect of Mdm2 on p53. Future experiments
under physiological conditions are needed. Different pathways are triggered in response to
different stresses, and the various steps in these pathways are constantly being dissected. Often
multiple and parallel pathways are elicited in order to achieve a coordinated and rapid
response. This concept is well exemplified by the activation of ATM in response to DNA
damage, which not only phosphorylates p53 and Mdm2 directly, but also activates concurrently
several positive regulators of p53 including Chk2, c-Abl, NBS1 and BRCA1. This
amplifies the stress signal leading to p53 activation and ensures the transmission of the signal
even if one or more pathways are defective. The post-translational modifications of p53 modulate
the auto-regulatory feedback p53/Mdm2 loop, but also monitor the fine-tuning of p53 transcriptional
activities. These modifications, among many other factors, influence the cellular decision to
die or pause. In contrast to our increasing knowledge how p53 is triggered in response to stress,
we know very little about how the signaling for p53 activation is being turned off. For example,
how the completion of DNA repair signals for p53 deactivation? Does it involve specific
phosphatases? How they are regulated and what are their targets?
Relatively little progress has been made towards the understanding of how p53 kills cells.
No one major pathway has been defined, but rather combinations of many parallel pathways
are probably required to fulfill this task. The ability of p53 to kill defective cells with genomic
instability, or with aberrant oncogenic events, is probably the most critical function of p53 as a
118 Cell Cycle Checkpoints and Cancer
tumor suppressor. The role of p53 in the G1 checkpoint, and in the maintenance of the G2
checkpoint, minimizes the accumulation of genetic abnormalities.
Acknowledgment
We thank Sue Moody-Haupt for critical comments and to Yaara Zwang for her help with
the references. The work in the author’s lab is supported by the Research Career Development
Award from the Israel Cancer Research Fund and by the Center for Excellence Grant of the
Israel Science Foundation.
References
1. Agarwal ML, Taylor WR, Chernov MV et al. The p53 network. J Biol Chem 1998; 273:1-4.
2. Almog N Rotter V. An insight into the life of p53: A protein coping with many functions. Biochim
Biophys Acta 1998; 1378:R43-54.
3. Bates S, Vousden KH. Mechanisms of p53-mediated apoptosis. Cell Mol Life Sci 1999;
55:28-37.
4. Colman MS, Afshari CA, Barrett JC. Regulation of p53 stability and activity in response to genotoxic
stress. Mutat Res 2000; 462:179-188.
5. Lakin ND, Jackson SP. Regulation of p53 in response to DNA damage. Oncogene 1999;
18:7644-7655.
6. Ljungman M. Dial 9-1-1 for p53: Mechanisms of p53 activation by cellular stress. Neoplasia 2000;
2:208-225.
7. Oren M. Regulation of the p53 tumor suppressor protein. J Biol Chem 1999; 274:36031-36034.
8. Vogt Sionov R, Haupt Y. The cellular response to p53: The decision between life and death.
Oncogene 1999; 18:6145-6157.
9. Choi J, Donehower LA. p53 in embryonic development: Maintaining a fine balance. Cell Mol Life
Sci 1999; 55:38-47.
10. Akashi M, Koeffler HP. Li-Fraumeni syndrome and the role of the p53 tumor suppressor gene in
cancer susceptibility. Clin Obstet Gynecol 1998; 41:172-199.
11. Soussi T. The p53 tumor suppressor gene: From molecular biology to clinical investigation. Ann
N Y Acad Sci 2000; 910:121-137.
12. Momand J, Wu HH, Dasgupta G. MDM2-master regulator of the p53 tumor suppressor protein.
Gene 2000; 242:15-29.
13. Sherr CJ, Weber JD. The ARF/p53 pathway. Curr Opin Genet Dev 2000; 10:94-99.
14. Carr AM. Cell cycle. Piecing together the p53 puzzle. Science 2000; 287:1765-1766.
15. Appella E, Anderson CW. Signaling to p53: Breaking the posttranslational modification code. Pathol
Biol 2000; 48:227-245.
16. Ashcroft M, Kubbutat MH, Vousden KH. Regulation of p53 function and stability by phosphorylation.
Mol Cell Biol 1999; 19:1751-1758.
17. Giaccia AJ, Kastan MB. The complexity of p53 modulation: Emerging patterns from divergent
signals. Genes Dev 1998; 12:2973-2983.
18. Jimenez GS, Khan SH, Stommel JM, Wahl GM. p53 regulation by post-translational modification
and nuclear retention in response to diverse stresses. Oncogene 1999; 18:7656-7665.
19. Meek DW. Mechanisms of switching on p53: A role for covalent modification? Oncogene
1999; 18:7666-7675.
20. Prives C, Hall PA. The p53 pathway. J Pathol 1999; 187:112 -126.
21. Freedman DA, Wu L, Levine AJ. Functions of the MDM2 oncoprotein. Cell Mol Life Sci
1999; 55:96-107.
22. Juven Gershon T, Oren M. Mdm2: The ups and downs. Mol Med 1999; 5:71-83.
23. Bar-On RL, Maya R, Segel LA et al. Generation of oscillations by the p53-Mdm2 feedback loop:
A theoretical and experimental study. Proc Natl Acad Sci USA 2000; 97:11250-11255.
24. Jones SN, Roe AE, Donehower LA, Bradley A. Rescue of embryonic lethality in Mdm2-deficient
mice by absence of p53. Nature 1995; 378:206-208.
25. Montes de Oca Luna R, Wagner DS, Lozano G. Rescue of early embryonic lethality in
mdm2-deficient mice by deletion of p53. Nature 1995; 378:203-206.
26. de Rozieres S, Maya R, Oren M, Lozano G. The loss of mdm2 induces p53-mediated apoptosis.
Oncogene 2000; 19:1691-1697.
27. Haupt Y, Maya R, Kazaz A, Oren M. Mdm2 promotes the rapid degradation of p53. Nature
1997; 387:296-299.
28. Kubbutat MHG, Jones SN, Vousden KH. Regulation of p53 stability by Mdm2. Nature
1997; 387:299-303.
The Regulation of p53 Growth Suppression 119
29. Blagosklonny MV. p53 from complexity to simplicity: Mutant p53 stabilization, gain-of-function,
and dominant-negative effect. FASEB J 2000;14: 1901-1907.
30. Buschmann T, Minamoto T, Wagle N et al. Analysis of JNK, Mdm2 and p14ARF contribution to
the regulation of mutant p53 stability. J Mol Biol 2000; 295:1009-1021.
31. Honda R, Tanaka H, Yasuda H. Oncoprotein MDM2 is a ubiquitin ligase E3 for tumor suppressor
p53. FEBS Lett 1997; 420:25-27.
32. Fang S, Jensen JP, Ludwig RL et al. Mdm2 is a RING finger-dependent ubiquitin protein ligase
for itself and p53. J Biol Chem 2000; 275:8945-8951.
33. Honda R, Yasuda H. Activity of MDM2, a ubiquitin ligase, toward p53 or itself is dependent on
the RING finger domain of the ligase. Oncogene 2000; 19:1473-1476.
34. Kubbutat MHG, Ludwig RL, Ashcroft M, Vousden KH. Regulation of Mdm2-directed degradation
by the C-terminus of p53. Mol Cell Biol 1998; 18:5690-5698.
35. Maki CG. Oligomerization is required for p53 to be efficiently ubiquitinated by MDM2. J Biol
Chem 1999; 274:16531-16535.
36. Nakamura S, Roth JA, Mukhopadhyay T. Multiple lysine mutations in the C-terminal domain of
p53 interfere with Mdm2-dependent protein degradation and ubiquitination. Mol Cell Biol 2000;
20:9391-9398.
37. Rodriguez MS, Desterro JMP, Lain S et al. Multiple C-terminal lysine residues target p53 for
ubiquitin-proteasome-mediated degradation. Mol Cell Biol 2000; 20:8458-8467.
38. Berger M, Vogt Sionov R, Levine AJ, Haupt Y. A role for the polyproline domain of p53 in its
regulation by Mdm2. J Biol Chem 2001; 276: In Press.
39. Gu J, Chen D, Rosenblum J et al. Identification of a sequence element from p53 that signals for
Mdm2 -targeted degradation. Mol Cell Biol 2000; 20:1243-1253.
40. Fuchs SY, Adler V, Buschmann T et al. JNK targets p53 ubiquitination and degradation in
nonstressed cells. Genes Dev 1998; 12:2658-2663.
41. Wadgaonkar R, Collins T. Murine double minute (MDM2) blocks p53-coactivator interaction, a
new mechanism for inhibition of p53-dependent gene expression. J Biol Chem 1999;
274:13760-13767.
42. Oliner JD, Pietenpol JA, Thiagalingam S et al. Oncoprotein MDM2 conceals the activation domain
of tumour suppressor p53. Nature 1993; 362:857-860.
43. Haupt Y, Barak Y, Oren M. Cell type-specific inhibition of p53-mediated apoptosis by mdm2.
EMBO J 1996; 15:1596 -1606.
44. Unger T, Juven-Gershon T, Moallem E et al. Critical role for Ser20 of human p53 in the negative
regulation of p53 by Mdm2. EMBO J 1999; 18:1805-1814.
45. Jackson MW, Berberich SJ. MdmX protects p53 from Mdm2-mediated degradation. Mol Cell Biol
2000; 20:1001-1007.
46. Stad R, Ramos YF, Little N et al. Hdmx stabilizes Mdm2 and p53. J Biol Chem 2000;
275:28039-28044.
47. Yap DBS, Hsieh J-K, Lu X. Mdm2 inhibits the apoptotic function of p53 mainly by targeting it
for degradation. J Biol Chem. 2000; 275:37296-37303.
48. Blaydes JP, Wynford-Thomas D. The proliferation of normal human fibroblasts is dependent upon
negative regulation of p53 function by mdm2. Oncogene 1998; 16:3317-3322.
49. Böttger A, Böttger V, Sparks A et al. Design of a synthetic Mdm2-binding mini protein that
activates the p53 response in vivo. Curr Biol 1997; 7: 860 -869.
50. Chen L, Lu W, Agrawal S et al. Ubiquitous induction of p53 in tumor cells by antisense inhibition
of MDM2 expression. Mol Med 1999; 5: 21 -34.
51. Wasylyk C, Salvi R, Argentini M et al. p53 mediated death of cells overexpressing MDM2 by an
inhibitor of MDM2 interaction with p53. Oncogene 1999; 18:1921-1934.
52. Chehab NH, Malikzay A, Stavridi ES, Halazonetis TD. Phosphorylation of Ser-20 mediates
stabilization of human p53 in response to DNA damage. Proc Natl Acad Sci USA 1999;
96:13777-13782.
53. Shieh S-Y, Taya Y, Prives C. DNA damage-inducible phosphorylation of p53 at N-terminal sites
including a novel site, Ser20, requires tetramerization. EMBO J 1999; 18:1815-1823.
54. Chehab NH, Malikzay A, Appel M Halazonetis TD. Chk2/hCds1 functions as a DNA damage
checkpoint in G1 by stabilizing p53. Genes Dev 2000; 14:278-288.
55. Hirao A, Kong YY, Matsuoka S et al. DNA damage-induced activation of p53 by the checkpoint
kinase Chk2. Science 2000; 287:1824-1827.
56. Böttger V, Böttger A, Garcia Echeverria C et al. Comparative study of the p53-mdm2 and
p53-MDMX interfaces. Oncogene 1999; 18:189-199.
120 Cell Cycle Checkpoints and Cancer
57. Sakaguchi K, Saito S, Higashimoto Y et al. Damage-mediated phosphorylation of human p53
threonine 18 through a cascade mediated by a casein 1-like kinase. Effect on Mdm2 binding. J
Biol Chem 2000; 275:9278-9283.
58. Dumaz N, Meek DW. Serine15 phosphorylation stimulates p53 transactivation but does not directly
influence interaction with HDM2. EMBO J 1999; 18:7002-7010.
59. Gostissa M, Hengstermann A, Fogal V et al. Activation of p53 by conjugation to the ubiquitin-like
protein SUMO-1. EMBO J 1999; 18:6462-6471.
60. Müller S, Berger M, Lehembre F et al. c-Jun and p53 activity is modulated by SUMO-1 modification.
J Biol Chem 2000; 275:13321-13329.
61. Rodriguez MS, Desterro JM, Lain S et al. SUMO-1 modification activates the transcriptional response
of p53. EMBO J 1999; 18:6455-6461.
62. Buschmann T, Fuchs SY, Lee CG et al. SUMO-1 modification of Mdm2 prevents its
self-ubiquitination and increases Mdm2 ability to ubiquitinate p53. Cell 2000; 101:753-762.
63. Khosravi R, Maya R, Gottlieb T et al. Rapid ATM-dependent phosphorylation of MDM2 precedes
p53 accumulation in response to DNA damage. Proc Natl Acad Sci USA 1999;
96:14973-14977.
64. Mayo LD, Turchi JJ, Berberich SJ. Mdm-2 phosphorylation by DNA-dependent protein kinase
prevents interaction with p53. Cancer Res 1997; 57:5013-5016.
65. Artandi SE, DePinho RA. Mice without telomerase: what can they teach us about human cancer?
Nat Med 2000; 6:852-855.
66. Serrano M. The INK4α/ARF locus in murine tumorigenesis. Carcinogenesis 2000; 21:865-869.
67. Kamijo T, Bodner S, van de Kamp E et al. Tumor spectrum in ARF-deficient mice. Cancer Res
1999; 59:2217-2222.
68. Bringold F, Serrano M. Tumor suppressors and oncogenes in cellular senescence. Exp Gerontol
2000; 35:317-329.
69. Robertson KD, Jones PA. The human ARF cell cycle regulatory gene promoter is a CpG island
which can be silenced by DNA methylation and down-regulated by wild-type p53. Mol Cell Biol
1998; 18:6457-6473.
70. Bates S, Phillips AC, Clark PA et al. p14ARF links the tumour suppressors RB and p53. Nature
1998; 395:124-125.
71. Stott FJ, Bates S, James MC et al. The alternative product from the human CDKN2A locus,
p14ARF, participates in a regulatory feedback loop with p53 and MDM2. EMBO J 1998;17:5001-
5014.
72. Honda R, Yasuda H. Association of p19ARF with Mdm2 inhibits ubiquitin ligase activity of Mdm2
for tumor suppressor p53. EMBO J 1999; 18:22-27.
73. Pomerantz J, Schreiber Agus N, Liegeois NJ et al. The Ink4α tumor suppressor gene product,
p19Arf, interacts with MDM2 and neutralizes MDM2’s inhibition of p53. Cell 1998; 92:713-723.
74. Tao W, Levine AJ. p19ARF stabilizes p53 by blocking nucleo-cytoplasmic shuttling of Mdm2. Proc
Natl Acad Sci USA 1999; 96:6937-6941.
75. Weber JD, Taylor LJ, Roussel MF et al. Nucleolar Arf sequesters Mdm2 and activates p53. Nat
Cell Biol 1999; 1:20-26.
76. Zhang Y, Xiong Y. Mutations in human ARF exon 2 disrupt its nucleolar localization and impair
its ability to block nuclear export of MDM2 and p53. Mol Cell 1999; 3:579-591.
77. Rizos H, Darmanian AP, Mann GJ, Kefford RF. Two arginine rich domains in the p14ARF tumour
suppressor mediate nucleolar localization. Oncogene 2000; 19:2978-2985.
78. Weber JD, Kuo ML, Bothner B et al. Cooperative signals governing ARF-mdm2 interaction and
nucleolar localization of the complex. Mol Cell Biol 2000; 20:2517-2528.
79. Lohrum MA, Ashcroft M, Kubbutat MH, Vousden KH. Identification of a cryptic
nucleolar-localization signal in MDM2. Nat Cell Biol 2000; 2:179-181.
80. Shaul Y. c-Abl: activation and nuclear targets. Cell Death Differ 2000; 7:10-16.
81. Van Etten RA. Cycling, stressed-out and nervous: Cellular functions of c-Abl. Trends Cell Biol
1999; 9:179-186.
82. Schwartzberg PL, Stall AM, Hardin JD et al. Mice homozygous for the ablm1 mutation show poor
viability and depletion of selected B- and T-cell populations. Cell 1991; 65:1165-1175.
83. Tybulewicz VL, Crawford CE, Jackson PK et al. Neonatal lethality and lymphopenia in mice with
a homozygous disruption of the c-abl proto-oncogene. Cell 1991; 65:1153-1163.
84. Whang YE, Tran C, Henderson C et al. c-Abl is required for development and optimal cell proliferation
in the context of p53 deficiency. Proc Natl Acad Sci USA 2000; 97:5486-5491.
85. Kharbanda S, Yuan ZM, Weichselbaum R, Kufe D. Determination of cell fate by c-Abl activation
in the response to DNA damage. Oncogene 1998; 17:3309-3318.
The Regulation of p53 Growth Suppression 121
86. Nie Y, Li HH, Bula CM, Liu X. Stimulation of p53 DNA binding by c-Abl requires the p53
C-terminus and tetramerization. Mol Cell Biol 2000; 20:741-748.
87. Vogt Sionov R, Moallem E, Berger M et al. c-Abl neutralizes the inhibitory effect of Mdm2 on
p53. J Biol Chem 1999; 274:8371-8374.
88. Hsieh JK, Chan FS, O’Connor DJ et al. RB regulates the stability and the apoptotic function of
p53 via MDM2. Mol Cell 1999; 3:181-193.
89. Damalas A, Ben Ze’ev A, Simcha I et al. Excess beta-catenin promotes accumulation of transcriptionally
active p53. EMBO J 1999; 18:3054-3063.
90. Kim IS, Kim DH, Han SM et al. Truncated form of importin α identified in breast cancer cell
inhibits nuclear import of p53. J Biol Chem 2000; 275:23139-23145.
91. Moll UM, Ostermeyer AG, Haladay R et al. Cytoplasmic sequestration of wild-type p53 protein
impairs the G1 checkpoint after DNA damage. Mol Cell Biol 1996; 16:1126-1137.
92. Momand J, Jung D, Wilczynski S, Niland J. The MDM2 gene amplification database. Nucleic
Acids Res 1998; 26:3453-3459.
93. Lu W, Pochampally R, Chen L et al. Nuclear exclusion of p53 in a subset of tumors requires
MDM2 function. Oncogene 2000; 19:232-240.
94. Thomas M, Pim D, Banks L. The role of the E6-p53 interaction in the molecular pathogenesis of
HPV. Oncogene 1999; 18:7690-7700.
95. Elmore LW, Hancock AR, Chang SF et al. Hepatitis B virus X protein and p53 tumor suppressor
interactions in the modulation of apoptosis. Proc Natl Acad Sci USA 1997; 94:14707-14712.
96. Konig C, Roth J, Dobbelstein M. Adenovirus type 5 E4orf3 protein relieves p53 inhibition by
E1B-55-kilodalton protein. J Virol 1999; 73:2253-2262.
97. Wienzek S, Roth J, Dobbelstein M. E1B 55-kilodalton oncoproteins of adenovirus types 5 and 12
inactivate and relocalize p53, but not p51 or p73, and cooperate with E4orf6 proteins to destabilize
p53. J Virol 2000; 74:193-202.
98. Komarova EA, Zelnick CR, Chin D et al. Intracellular localization of p53 tumor suppressor protein
in gamma-irradiated cells is cell cycle regulated and determined by the nucleus. Cancer Res 1997;
57:5217-5220.
99. Shaulsky G, Ben Ze’ev A, Rotter V. Subcellular distribution of the p53 protein during the cell
cycle of Balb/c 3T3 cells. Oncogene 1990; 5:1707-1711.
100. Liang SH, Clarke MF. A bipartite nuclear localization signal is required for p53 nuclear import
regulated by a carboxyl-terminal domain. J Biol Chem 1999; 274:32699-32703.
101. Liang SH, Clarke MF. The nuclear import of p53 is determined by the presence of a basic domain
and its relative position to the nuclear localization signal. Oncogene 1999; 18:2163-2166.
102. Klotzsche O, Etzrodt D, Hohenberg H et al. Cytoplasmic retention of mutant tsp53 is dependent
on an intermediate filament protein (vimentin) scaffold. Oncogene 1998; 16:3423-3434.
103. Giannakakou P, Sackett DL, Ward Y et al. p53 is associated with cellular microtubules and is
transported to the nucleus by dynein. Nat Cell Biol 2000; 2:709-717.
104. Freedman DA, Levine AJ. Nuclear export is required for degradation of endogenous p53 by MDM2
and human papillomavirus E6. Mol Cell Biol 1998; 18:7288-7293.
105. Roth J, Dobbelstein M, Freedman DA et al. Nucleo-cytoplasmic shuttling of the hdm2 oncoprotein
regulates the levels of the p53 protein via a pathway used by the human immunodeficiency virus
rev protein. EMBO J 1998; 17:554-564.
106. Lain S, Midgley C, Sparks A et al. An inhibitor of nuclear export activates the p53 response and
induces the localization of HDM2 and p53 to U1A-positive nuclear bodies associated with the
PODs. Exp Cell Res 1999; 248:457-472.
107. Kudo N, Matsumori N, Taoka H et al. Leptomycin B inactivates CRM1/exportin 1 by covalent
modification at a cysteine residue in the central conserved region. Proc Natl Acad Sci USA 1999;
96:9112-9117.
108. Henderson BR Eleftheriou A. A comparison of the activity, sequence specificity, and CRM1-dependence
of different nuclear export signals. Exp Cell Res 2000; 256:213-224.
109. Stommel JM, Marchenko ND, Jimenez GS et al. A leucine-rich nuclear export signal in the p53
tetramerization domain: regulation of subcellular localization and p53 activity by NES masking.
EMBO J 1999; 18:1660-672.
110. Marston NJ, Jenkins JR, Vousden KH. Oligomerisation of full length p53 contributes to the interaction
with mdm2 but not HPV E6. Oncogene 1995; 10:1709-1715.
111. Boyd SC, Tsai KY, Jacks T. An intact Hdm2 RING-finger domain is required for nuclear exclusion
of p53. Nat Cell Biol 2000; 2:563-568.
112. Geyer RK, Yu ZK, Maki CG. The Mdm2 RING-finger domain is required to promote p53 nuclear
export. Nat Cell Biol 2000; 2:569-573.
122 Cell Cycle Checkpoints and Cancer
113. Fogal V, Gostissa M, Sandy P et al. Regulation of p53 activity in nuclear bodies by a specific PML
isoform. EMBO J 2000; 19:6185-6195.
114. Guo A, Salomoni P, Luo J et al. The function of PML in p53-dependent apoptosis. Nat Cell Biol
2000; 2:730-736.
115. Matera AG. Nuclear bodies: Multifaceted subdomains of the interchromatin space. Trends Cell
Biol 1999; 9:302-309.
116. Seeler JS, Dejean A. The PML nuclear bodies: actors or extras? Curr Opin Genet Dev 1999;
9:362-367.
117. Wang ZG, Ruggero D, Ronchetti S et al. PML is essential for multiple apoptotic pathways. Nat
Genet 1998; 20:266-272.
118. Pearson M, Carbone R, Sebastiani C et al. PML regulates p53 acetylation and premature senescence
induced by oncogenic Ras. Nature 2000; 406:207-210.
119. Agarwal ML, Agarwal A, Taylor WR et al. A p53-dependent S-phase checkpoint helps to protect
cells from DNA damage in response to starvation for pyrimidine nucleotides. Proc Natl Acad Sci
USA 1998; 95:14775-14780.
120. Meek DW. The role of p53 in the response to mitotic spindle damage. Pathol Biol 2000;
48:246-254.
121. Kastan MB, Zhan Q, el Deiry WS et al. A mammalian cell cycle checkpoint pathway utilizing p53
and GADD45 is defective in ataxia-telangiectasia. Cell 1992; 71:587-597.
122. Bunz F, Dutriaux A, Lengauer C et al. Requirement for p53 and p21 to sustain G2 arrest after
DNA damage. Science 1998; 282:1497-1501.
123. Brehm A, Miska EA, McCance DJ et al. Retinoblastoma protein recruits histone deacetylase to
repress transcription. Nature 1998; 391:597-601.
124. Deng C, Zhang P, Harper JW et al. Mice lacking p21CIP1/WAF1 undergo normal development, but
are defective in G1 checkpoint control. Cell 1995; 82:675-684.
125. Schneider E, Montenarh M, Wagner P. Regulation of CAK kinase activity by p53. Oncogene
1998; 17:2733-2741.
126. Ko LJ, Shieh SY, Chen X et al. p53 is phosphorylated by CDK7-cyclin H in a p36MAT1-dependent
manner. Mol Cell Biol 1997; 17:7220-7229.
127. Guardavaccaro D, Corrente G, Covone F et al. Arrest of G1-S progression by the p53-inducible gene
PC3 is Rb dependent and relies on the inhibition of cyclin D1 transcription. Mol Cell Biol 2000;
20:1797-1815.
128. Flatt PM, Tang LJ, Scatena CD et al. p53 regulation of G2 checkpoint is retinoblastoma protein
dependent. Mol Cell Biol 2000; 20:4210-4223.
129. Jin S, Antinore MJ, Lung FD et al. The GADD45 inhibition of Cdc2 kinase correlates with
GADD45 -mediated growth suppression. J Biol Chem 2000; 275:16602-16608.
130. Zhan Q, Antinore MJ, Wang XW et al. Association with Cdc2 and inhibition of Cdc2/Cyclin B1
kinase activity by the p53-regulated protein GADD45. Oncogene 1999; 18: 2892-2900.
131. Hollander MC, Sheikh MS, Bulavin DV et al. Genomic instability in GADD45a-deficient mice.
Nat Genet 1999; 23:176-184.
132. Hermeking H, Lengauer C, Polyak K et al. 14-3-3-σ is a p53-regulated inhibitor of G2/M progression.
Mol Cell 1997; 1:3-11.
133. Peng CY, Graves PR, Thomas RS et al. Mitotic and G2 checkpoint control: regulation of 14-3-3
protein binding by phosphorylation of Cdc25C on serine-216. Science 1997; 277:1501-1505.
134. Ferrell JE, Jr. How regulated protein translocation can produce switch-like responses. Trends Biochem
Sci. 1998; 23:461-465.
135. Chan TA, Hermeking H, Lengauer C et al. 14-3-3-σ is required to prevent mitotic catastrophe
after DNA damage. Nature 1999; 401:616-620.
136. Innocente SA, Abrahamson JL, Cogswell JP, Lee JM. p53 regulates a G2 checkpoint through cyclin
B1. Proc Natl Acad Sci USA 1999; 96:2147-2152.
137. Park M, Chae HD, Yun J et al. Constitutive activation of cyclin B1-associated cdc2 kinase overrides
p53-mediated G2-M arrest. Cancer Res. 2000; 60:542-545.
138. Taylor WR, DePrimo SE, Agarwal A et al. Mechanisms of G2 arrest in response to overexpression
of p53. Mol Biol. Cell 1999; 10:3607-3622.
139. Chan TA, Hwang PM, Hermeking H et al. Cooperative effects of genes controlling the G2/M
checkpoint. Genes Dev 2000; 14:1584-1588.
140. Tarapore P, Fukasawa K. p53 mutation and mitotic infidelity. Cancer Invest 2000; 18:148-155.
141. Yin XY, Grove L, Datta NS et al. c-myc overexpression and p53 loss cooperate to promote genomic
instability. Oncogene 1999; 18:1177-1184.
The Regulation of p53 Growth Suppression 123
142. Sablina AA, Agapova LS, Chumakov PM, Kopnin BP. p53 does not control the spindle assembly
cell cycle checkpoint but mediates G1 arrest in response to disruption of microtubule system. Cell
Biol Int 1999; 23:323-334.
143. Stewart ZA, Leach SD, Pietenpol JA. p21WAF1/Cip1 inhibition of cyclin E/CDK2 activity prevents
endoreduplication after mitotic spindle disruption. Mol Cell Biol 1999; 19:205-215.
144. Lacey KR, Jackson PK, Stearns T. Cyclin-dependent kinase control of centrosome duplication.
Proc Natl Acad Sci USA 1999; 96:2817-2822.
145. Mussman JG, Horn HF, Carroll PE et al. Synergistic induction of centrosome hyperamplification
by loss of p53 and cyclin E overexpression. Oncogene 2000; 19:1635-1646.
146. Khan SH Wahl GM. p53 and pRb prevent rereplication in response to microtubule inhibitors by
mediating a reversible G1 arrest. Cancer Res 1998; 58:396-401.
147. Lanni JS, Jacks T. Characterization of the p53-dependent postmitotic checkpoint following spindle
disruption. Mol Cell Biol 1998; 18:1055-1064.
148. Mantel C, Braun SE, Reid S et al. p21cip-1/waf-1 deficiency causes deformed nuclear architecture, centriole
overduplication, polyploidy, and relaxed microtubule damage checkpoints in human
hematopoietic cells. Blood 1999; 93:1390-1398.
149. Hsu LC, White RL. BRCA1 is associated with the centrosome during mitosis. Proc Natl Acad Sci
USA 1998; 95:12983-12988.
150. Deng CX, Brodie SG. Roles of BRCA1 and its interacting proteins. Bioessays 2000; 22:728-737.
151. Scully R, Livingston DM. In search of the tumour-suppressor functions of BRCA1 and BRCA2.
Nature 2000; 408:429-432.
152. Xu X, Weaver Z, Linke SP et al. Centrosome amplification and a defective G2-M cell cycle checkpoint
induce genetic instability in BRCA1 exon 11 isoform-deficient cells. Mol Cell 1999;
3:389-395.
153. Yap DB, Hsieh JK, Chan FS, Lu X. mdm2: A bridge over the two tumour suppressors, p53 and
Rb. Oncogene 1999; 18:7681-7689.
154. Johnson TM, Yu ZX, Ferrans VJ et al. Reactive oxygen species are downstream mediators of p53-
dependent apoptosis. Proc Natl Acad Sci USA 1996; 93:11848-11852.
155. Li PF, Dietz R, von Harsdorf R. p53 regulates mitochondrial membrane potential through reactive
oxygen species and induces cytochrome C-independent apoptosis blocked by Bcl-2. EMBO J 1999;
18:6027-6036.
156. Rich T, Allen RL, Wyllie AH. Defying death after DNA damage. Nature 2000; 407:777-783.
157. Soengas MS, Alarcon RM, Yoshida H et al. Apaf-1 and caspase-9 in p53-dependent apoptosis and
tumor inhibition. Science 1999; 284:156-159.
158. May P, May E. Twenty years of p53 research: Structural and functional aspects of the p53 protein.
Oncogene 1999; 18:7621-7636.
159. Allan LA, Fried M. p53-dependent apoptosis or growth arrest induced by different forms of
radiation in U2OS cells: p21WAF1/CIP1 repression in UV induced apoptosis. Oncogene
1999; 18:5403-5412.
160. Blaydes JP, Craig AL, Wallace M et al. Synergistic activation of p53-dependent transcription by
two cooperating damage recognition pathways. Oncogene 2000; 19:3829-3839.
161. Lu X, Lane DP. Differential induction of transcriptionally active p53 following UV or ionizing
radiation: defects in chromosome instability syndromes? Cell 1993; 75:765-778.
162. Buschmann T, Adler V, Matusevich E et al. p53 phosphorylation and association with murine
double minute 2, c-Jun NH2-terminal kinase, p14ARF, and p300/CBP during the cell cycle and
after exposure to ultraviolet irradiation. Cancer Res 2000; 60:896-900.
163. Blaydes JP, Hupp TR. DNA damage triggers DRB-resistant phosphorylation of human p53 at the
CK2 site. Oncogene 1998; 17:1045-1052.
164. Kapoor M, Lozano G. Functional activation of p53 via phosphorylation following DNA damage
by UV but not-radiation. Proc Natl Acad Sci USA 1998; 95:2834-2837.
165. Lu H, Taya Y, Ikeda M, Levine AJ. Ultraviolet radiation, but not-radiation or etoposide-induced
DNA damage, results in the phosphorylation of the murine p53 protein at serine-389. Proc Natl
Acad Sci USA 1998; 95:6399-6402.
166. Sakaguchi K, Herrera JE, Saito S et al. DNA damage activates p53 through a phosphorylationacetylation
cascade. Genes Dev 1998; 12:2831-2841.
167. Oda K, Arakawa H, Tanaka T et al. p53AIP1, a potential mediator of p53-dependent apoptosis,
and its regulation by Ser-46-phosphorylated p53. Cell 2000; 102:849-862.
168. Shieh SY, Ikeda M, Taya Y, Prives C. DNA damage-induced phosphorylation of p53 alleviates
inhibition by MDM2. Cell 1997; 91:325-334.
169. Siliciano JD, Canman CE, Taya Y et al. DNA damage induces phosphorylation of the amino
terminus of p53. Genes Dev 1997; 11:3471-3481.
124 Cell Cycle Checkpoints and Cancer
170. Tibbetts RS, Brumbaugh KM, Williams JM et al. A role for ATR in the DNA damage-induced
phosphorylation of p53. Genes Dev 1999; 13:152-157.
171. Albrechtsen N, Dornreiter I, Grosse F et al. Maintenance of genomic integrity by p53: Complementary
roles for activated and nonactivated p53. Oncogene 1999; 18:7706-7717.
172. Flaman JM, Robert V, Lenglet S et al. Identification of human p53 mutations with differential
effects on the BAX and p21 promoters using functional assays in yeast. Oncogene 1998;
16:1369-1372.
173. Thukral SK, Lu Y, Blain GC et al. Discrimination of DNA binding sites by mutant p53 proteins.
Mol Cell. Biol 1995; 15:5196-5202.
174. Kim E, Albrechtsen N, Deppert W. DNA-conformation is an important determinant of
sequence-specific DNA binding by tumor suppressor p53. Oncogene 1997; 15:857-869.
175. Thornborrow EC, Manfredi JJ. One mechanism for cell type -specific regulation of the BAX promoter
by the tumor suppressor p53 is dictated by the p53 response element. J Biol Chem 1999;
274:33747-33756.
176. Friedlander P, Haupt Y, Prives C, Oren M. A mutant p53 that discriminates between p53-responsive
genes cannot induce apoptosis. Mol Cell Biol 1996; 16:4961-4971.
177. Ludwig RL, Bates S, Vousden KH. Differential activation of target cellular promoters by p53
mutants with impaired apoptotic function. Mol Cell Biol 1996; 16:4952-4960.
178. Ryan KM, Vousden KH. Characterization of structural p53 mutants which show selective defects
in apoptosis but not cell cycle arrest. Mol Cell Biol 1998; 18:3692-3698.
179. Roth J, Koch P, Contente A, Dobbelstein M. Tumor-derived mutations within the DNA-binding
domain of p53 that phenotypically resemble the deletion of the proline-rich domain. Oncogene
2000; 19:1834-1842.
180. Saller E, Tom E, Brunori M et al. Increased apoptosis induction by 121F mutant p53. EMBO J
1999; 18:4424-4437.
181. Zhu J, Jiang J, Zhou W et al. Differential regulation of cellular target genes by p53 devoid of the
PXXP motifs with impaired apoptotic activity. Oncogene 1999; 18:2149-2155.
182. Relaix F, Wei X, Li W et al. Pw1/Peg3 is a potential cell death mediator and cooperates with
Siah1a in p53-mediated apoptosis. Proc Natl Acad Sci U S A 2000; 97:105-2110.
183. Wu GS, Burns TF, McDonald ER 3rd et al. Induction of the TRAIL receptor KILLER/DR5 in
p53-dependent apoptosis but notgrowth arrest. Oncogene 1999; 18:6411-6418.
184. Zhu Q, Wani MA, El Mahdy M et al. Modulation of transcriptional activity of p53 by ultraviolet
radiation: Linkae between p53 pathway and DNA repair through damage recognition. Mol Carcinog
2000; 28:215-224.
185. Amundson SA, Myers TG, Fornace AJ, Jr. Roles for p53 in growth arrest and apoptosis: Putting
on the brakes after genotoxic stress. Oncogene 1998; 17:3287-3299.
186. Perry ME, Mendrysa SM, Saucedo LJ et al. p76MDM2 inhibits the ability of p90MDM2 to desabilize
p53. J Biol Chem 2000; 275:5733-5738.
187. Saucedo LJ, Myers CD, Perry ME. Multiple murine double minute gene 2 (MDM2) proteins are
induced by ultraviolet light. J Biol Chem 1999; 274:8161-8168.
188. Ljungman M, Zhang F. Blockage of RNA polymerase as a possible trigger for U.V. light-induced
apoptosis. Oncogene 1996; 13:823-831.
189. Dumaz N, Duthu A, Ehrhart JC et al. Prolonged p53 protein accumulation in trichothiodystrophy
fibroblasts dependent on unrepaired pyrimdine dimers on the transcribed strands of cellular genes.
Mol Carcinog 1997; 20:340-347.
190. McKay BC, Ljungman M, Rainbow AJ. Persistent DNA damage induced by ultraviolet light inhibits
p21WAF1 and BAX expression: implications for DNA repair, UV sensitivity and the induction of
apoptosis. Oncogene 1998; 17:545-555.
191. Chen X, Ko LJ, Jayaraman L, Prives C. p53 levels, functional domains, and DNA damage determine
the extent of the apoptotic response o tumor cells. Genes Dev 1996; 10:2438-2451.
192. Wu X, Levine AJ. p53 and E2F-1 cooperate to mediate apoptosis. Proc Natl Acad Sci USA 1994;
91:3602-3606.
193. Irwin M, Marin MC, Phillips AC et al. Role for the p53 homologue p73 in E2F-1-induced
apoptosis. Nature 2000; 407:645-48.
194. Thomas A White E. Suppression of the p300-dependent mdm2 negative-feedback loop induces the
p53 apoptotic function. Genes Dev 1998; 121975-1985.
195. Ries S, Biederer C, Woods D et al. Opposing effects of Ras on p53: Transcriptional activation of
mdm2 and induction of p19ARF. Cell 2000; 103:321-330.
196. Allan LA, Duhig T, Read M, Fried M. The p21WAF1/CIP1 promoter is methylated in Rat-1 cells:
Stable restoration of p53-dependent p21WAF1/CIP1 expression after transfection of a genomic clone
containing the p21WAF1/CIP1 gene. Mol Cell Biol 2000; 20:1291-1298.
197. Gervais JL, Seth P, Zhang H. Cleavage of CDK inhibitor p21CIP1/Waf1 by caspases is an early event
during DNA damage-induced apoptosis. J Biol Chem 1998; 273:19207-19212.
The Regulation of p53 Growth Suppression 125
198. Gorospe M, Cirielli C, Wang X et al. p21WAF1/Cip1 protects against p53-mediated apoptosis of
human melanoma cells. Oncogene 197; 14:929-935.
199. Polyak K, Waldman T, He TC et al. Genetic determinants of p53-induced apoptosis and growth
arrest. Genes Dev 196; 10:1945-1952.
200. Waldman T, Zhang Y, Dillehay L et al. Cell-cycle arrest versus cell death in cancer therapy. Nat
Med 1997; 3:1034-036.
201. Bissonnette N, Hunting DJ. p21-induced cycle arrest in G1 protects cells from apoptosis induced
by U.V.-irradiation or RNA polymerase II blockage. Oncogene 1998; 16:3461-3469.
202. Haupt Y, Rowan S, Oren M. p53-mediated apoptosis in HeLa cells can be ovecome by excess
pRB. Oncogene 1995; 10:1563-1571.
203. Tanaka H, Arakawa H, Yamaguchi T et al. A ribonucleotide reductase gene involved in a
p53dependent cell-cycle checkpoint for DNA damage. Nature 2000; 404:42-49.
204. Asada M, Yamada T, Ichijo H et al. Apoptosis inhibitory activity of cytoplasmic p21CIP1/WAF1 in
monocytic differentiation. EMBO J 1999; 18:1223-1234.
205. Zha J, Harada H, Yang E et al. Serine phosphorylation of dath agonist BAD in response to survival
factor results in binding to 14-3-3 not BCL-XL. Cell 1996; 87:619-628.
206. Leri A, Liu Y, Claudio PP et al. Insulin-like growth factor-1 induces Mdm2 and down-regulates
p53, attenuating the myocyte renin-angiotensin system and stretch-mediated apoptosis. Am J Pathol
1999; 154:567-580.
207. Qi JS, Yuan Y, Desai Yajnik V, Samuels HH. Regulation of the mdm2 oncogene by thyroid
hormone receptor. Mol Cell Biol 1999; 19:864872.
208. Shaulian E, Resnitzky D, Shifman O et al. Induction of Mdm2 and enhancement of cell survival
by bFGF. Oncogene 1997; 15:2717-2725.
209. Banin S, Moyal L, Shieh S et al. Enhanced phosphorylation of p53 by ATM in response to DNA
damage. Science 1998; 281:1674-1677.
210. Canman CE, Lim DS, Cimprich KA et al. Activation of the ATM kinase by ionizing radiation and
phosphorylation of p53. Science 1998; 281:1677-1679.
211. Shieh SY, Ahn J, Tamai K et al. The human homologs of checkpoint kinases Chk1 and Cds1 (Chk2)
phosphorylate p53 at multiple DNA damage-inducible sites. Genes Dev 2000; 14:28-300.
212. Fuchs SY, Adler V, Pincus MR, Ronai Z. MEKK1/JNK signaling stabilizes and activates p53. Proc
Natl Acad Sci U S A 1998; 95:10541-10546.
213. Lu H, Fisher RP, Bailey P, Levine AJ. The CDK7-cycH-p36 complex of transcription factor IIH
phosphorylates p53, enhancing its sequence-specific DNA binding activity in vitro. Mol Cell Biol
1997; 17:5923-5934.
214. Bulavin DV, Saito S, Hollander MC et al. Phosphorylation of human p53 by p38 kinase
coordinates N-terminal phosphorylation and apoptosis in response to UV radiation. EMBO
J 1999; 18:6845-6854.
215. Sanchez Prieto R, Rojas JM, Taya Y, Gutkind JS. A role for the p38 mitogen-acitvated protein
kinase pathwayin the transcriptional activation of p53 on genotoxic stress by chemotherapeutic
agents. Cancer Res 2000; 60:2464-2472.
216. She QB, Chen N, Dong Z. ERKs and p38 kinase phosphorylate p53 protein at serine 15 in
response to UV radiation. J Biol Chem 2000; 275:20444-20449.
217. Sayed M, Kim SO, Salh BS et al. Stress-induced activation of protein kinase CK2 by direct interaction
with p38 mitogen-activated protein kinase. J Biol Chem 2000; 275:16569-16573.
218. Cuddihy AR, Wong AH, Tam NW et al. The double-stranded-RNA-activated protein kinase PKR
physically associates with the tumor suppressor p53 protein and phosphorylates p53 on serine 392
in vitro. Oncogene 1999; 18:2690 -2072.
219. Luciani MG, Hutchins JR, Zheleva D Hupp TR. The C-terminal regulatory domain of p53 contains
a functional docking site for cyclin A. J Mol Biol. 2000; 300:503-518.
220. Wang Y, Prives C. Increased and altered DNA binding of human p53 by S and G2/M but not G1
cyclin-dependent kinases. Nature 1995; 376:88-91.
221. Waterman MJ, Stavridi ES, Waterman JL, Halazonetis TD. ATM-dependent activation of
p53 involves dephosphorylation and association with 14-3-3 proteins. Nat Genet 1998;
19:175-178.
222. Li L, Ljungman M, Dixon JE. The human Cdc14 phosphatases interact with and dephosphorylate
the tumor suppressor protein p53. J Biol Chem 2000; 275:2410 -2414.
223. Gu W, Roeder RG. Activation of p53 sequence -specific DNA binding by acetylation of the p53
C-terminal domain. Cell 1997; 90:595 -606.
224. Kumari SR, Mendoza Alvarez H, Alvarez Gonzalez R. Functional interactions of p53 with
poly(ADP-ribose) polymerase (PARP) during apoptosis following DNA damage: Covalent
poly(ADP-ribosyl)ation of p53 by exogenous PARP and noncovalent binding of p53 to the M(r)
85,000 proteolytic fragment. Cancer Res 1998; 5:5075 -5078.
225. Wang X, Ohnishi K, Takahashi A, Ohnishi T. Poly(ADP-ribosyl)ation is required for p53-dependent
signal transduction induced by radiation. Oncogene 1998; 17:2819-2825.
126 Cell Cycle Checkpoints and Cancer
CHAPTER 7
Cell Cycle Checkpoints and Cancer, edited by Mikhail V. Blagosklonny. ©2001 Eurekah.com.
Functional Interactions Between BRCA1
and the Cell Cycle
Timothy K. MacLachlan and Wafik El-Deiry
Introduction
The onset of breast cancer in women is one of the most devastating diseases known
today, afflicting approximately one in nine women in Western countries.1 In families
that inherit breast and ovarian cancer, BRCA1 mutations account for close to 100% of
resultant cancers, and in pedigrees that solely inherit breast cancer, BRCA1 alterations are
present in nearly two-thirds of the families.2 These findings have led to the terminology of
BRCA1 as a true tumor suppressor. Its discovery in 1994 initially did not lead to any insights
into its functions, as the domains of the 220kDa protein were not exceptionally homologous to
any known proteins.3 However, research into the function of BRCA1 has yielded several theories
regarding its purpose in normal cells. Identification of interacting proteins, production of antibodies
against the protein, development of knockout and transgenic mouse models and
comparisons between BRCA1 wild type and mutant expressing cells has assisted in placing
functional characteristics with the BRCA1 protein. Among other qualities of BRCA1, it has
become clear that it is influenced by and affects directly the position of the cell cycle. From
phosphorylation to subcellular localization, protein-protein interaction to transcription
activation, it has become clear that BRCA1 activities are closely related with cell cycle events.
The transition from phase to phase in the mammalian cell cycle intimately involves the BRCA1
tumor suppressor.
BRCA1 Protein and mRNA during the Cell Cycle
Detection of BRCA1 transcripts initially did not identify a cell cycle component to its
regulation. However, a study looking specifically at the status of BRCA1 mRNA in G0 cells
found that the transcript was greatly reduced4. While expression was high in exponentially
growing cells, withdrawal of growth factor from human mammary epithelial cells resulted in a
disappearance of BRCA1 altogether. In addition, senescent cells also had dramatically reduced
BRCA1 transcript. The lack of BRCA1 transcript in non-dividing cells led to the notion that
this may be a component of the cell division machinery. At least in the case of senescence,
recent reports of the ability of p53 to repress the transcription of BRCA1 may underlie this
result5,6. Another contributor to the regulation of BRCA1 mRNA expression is the pRb-E2F
complex7. The promoter of BRCA1 contains several E2F binding sites and pRb is able to
repress transcription from the BRCA1 promoter in an E2F dependent fashion.
Development of antibodies against the full length BRCA1 protein further established its
role in the cell division process. Several papers described the shift of the BRCA1 protein band
to a higher mobility upon the onset of DNA replication, or if cells had been arrested in S phase
by such agents as hydroxyurea8-11. This shift was apparently due to phosphorylation, as
addition of phosphatase to extracts resulted in a collapse of the band to its normal state, and
Functional Interactions between BRCA1 and the Cell Cycle 127
phosphorylation occurred predominantly on serine (Fig. 1). The phosphorylation of BRCA1
continued throughout S and onto the G2/M phases, after which it was progressively dephosphorylated.
Blockage of cell cycle progression also resulted in phosphorylation, either at the
G1/S border or at G2/M by treatment with colchicine. Later, BRCA1 was found to posses a
cdk2 phosphorylation site at serine 1497.12 This site was found to be efficiently phosphorylated
in vitro by cdk2 complexed with either cyclin A or E. Kinases complexed with cyclin D
have also been shown to phosphorylate BRCA19 Therefore, at least one of the kinases that
force the hyperphosphorylation of BRCA1 at S phase is a major component of the cell cycle
machinery. To date, there exist several other kinases that are capable of BRCA1 phosphorylation
including casein kinase 2, DNA damage responsive kinases such as ATM, ATR, hCds1
and the AKT kinase as stimulated by heregulin.13-20 Whether these kinases also play a role in
cell cycle mediated alteration of BRCA1 remains to be tested. Also, it remains to be seen if cdk2
is capable of phosphorylating BRCA1 as late as G2/M phase—a phase that retains high levels
of phosphorylated BRCA1 yet has classically been thought to possess low levels of cdk2 kinase
activity. One report suggests that BRCA1 is actually predominantly tyrosine phosphorylated at
G2/M stages, so there may be other cell cycle regulated kinases that are able to affect BRCA1
phosphorylation status at different stages of the cycle.21 Thus far, the only concrete effect of
phosphorylation on BRCA1 to be shown has been a change in protein-protein interaction with
such regulators of its transcriptional activity as CtIP.22 There exists correlative evidence suggesting
that phosphorylation may affect subcellular localization and the conferring of sensitivity
to DNA damage,8,16,19 but no data thus far has placed a functional significance on cell
cycle-dependent phosphorylation of BRCA1.
Subcellular Localization
One of the first reports describing a cell cycle-dependent phosphorylation phenotype of
BRCA1 also noted a peculiar cellular distribution of BRCA1 (Fig. 2). In cells that were undergoing
DNA synthesis, BRCA1 protein appeared as foci when immunofluorescence staining
was performed.8 When S phase was interrupted by treatment with hydroxyurea or DNA dam-
Fig. 1. The subcellular localization of BRCA1 changes with respect to the cell cycle In G1 phase of the cell
cycle, BRCA1 is typically expressed in low amounts and what protein is Present, is distributed ubiquitously
throughout the nucleus (Black is the cytoplasm, dark green represents the nucleus and a relatively low level
of BRCA1 expression). As the cell progresses towards DNA replication, BRCA1 expression increases and
forms foci along replication forks in the genomic DNA (bright green dots represents high concentration of
BRCA1 protein isolated to regions of replication in the nucleus). Expression of BRCA1 remains high
through G2 phase And then relocalizes to the centrosomes after nuclear envelope breakdown (bright green
dots represent high concentration of BRCA1 protein at the centrosomes extending microtubules to the
condensed chromatin). During spermatogenesis and oogenesis, cells enter meiosis at which time BRCA1
relocalizes to the synaptonemal complexes (bright green dots represent high concentration of BRCA1
protein on the condensed chromatin.
128 Cell Cycle Checkpoints and Cancer
aging agents, these foci dispersed, indicating a correlation of the foci with ongoing DNA replication.
Proteins that are known to associate with BRCA1 such as BARD1 and Rad51 colocalized
with BRCA1 to these foci. Interestingly, during S phase these foci also colocalized with PCNA
positive replication structures, suggesting that BRCA1 containing complexes are an integral
part of the DNA replication machinery present at the replication fork. It may be that BRCA1
is essential for maintaining high fidelity replication in concert with its associated repair proteins
such as Rad51.
At another phase of the cell cycle however, BRCA1 possesses an entirely different localization.
In mitosis, the nuclear membrane breaks down, allowing the normally nuclear localized of
BRCA1 to distribute itself around the cell. One area of concentrated BRCA1 protein is the
centrosome.23 Immunofluorescence localizes BRCA1 to the polar ends of the cell during mitosis
and is able to be immunoprecipitated with antibodies against gamma tubulin, a centrosome
component. Interestingly, the gamma tubulin precipitated form of BRCA1 is hypophosphorylated.
This is at odds with previous data suggesting that phosphorylation of BRCA1 continues throughout
S and into G2/M. Perhaps there exist multiple populations of BRCA1 in the cell. It is of note
that microtubule destabilizing agents such as nocodazole induce a strong shift in phosphorylation
of BRCA1,10 raising the possibility that this phosphorylation induction may dissociate
Fig. 2. Accumulation and phosphorylation of BRCA1 during the cell cycle. In early G1 phase, BRCA1
protein is expressed at very low levels and is at best underphosphorylated. As the cell progresses towards the
DNA replication checkpoint, protein levels accumulate and begin to be phosphorylated. At the onset of S
phase, BRCA1 becomes increasingly phosphorylated by, among other possible kinases, cdk2. This level of
phosphorylation status remains high through G2 phase and then becomes progressively dephosphorylated
during mitosis.
Functional Interactions between BRCA1 and the Cell Cycle 129
BRCA1 from the microtubule organizing centers such as the centrosome. It will be interesting
to see if BRCA1 remains associated with the polar ends of the cell after treatment
with colchicine or nocodazole.
In meiotic cells, BRCA1 also has specific cellular localization. Along the chromosomes
during a period when synaptonemal complexes form, BRCA1 is isolated to regions that could
be undergoing recombination. 24This is consistent with the induction of BRCA1 that is seen
during this process. Interestingly, the DNA repair protein also colocalizes with BRCA1 during
this meiotic process. Perhaps the two proteins are intimately involved in the proper exchange of
genetic information between homologous chromosomes in spermatocytes or oocytes.
Activity at Cell Cycle Checkpoints
One aspect of subcellular localization of BRCA1 remains constant, regardless of cell cycle
phase. It is always involved in the protection of genomic DNA. Two major cell cycle checkpoints
that are required for genomic stability are DNA replication, where the lack of fidelity could result
in a mutation that may be deleterious to the daughter cell, and mitosis, where the separation of
sister chromatids must be performed carefully lest the distribution of unequal amounts of genetic
information be passed on to each cell at division. As it turns out, through knockout and
overexpression models, BRCA1 seems to be an integral component of both of these checkpoints.
The first attempt at eliminating BRCA1 from the genomes of mice resulted in very early
embryonic lethality.25 The complete lack of BRCA1 in developing mice forced death at embryonic
day.7-8 What was interesting about these embryos is that when checked for expression levels of
p21WAF1, the amounts were staggeringly high, indicative of the activation of the G1/S checkpoint.
This finding suggested that sufficient DNA damage had occurred in the absence of
BRCA1 enough to activate the G1/S checkpoint. BRCA1 itself may also be involved in activating this
checkpoint when it is present in cells. Overexpression of BRCA1 has been shown to cause cell
cycle arrest, but this effect requires the presence of p21WAF1 or pRb, both proteins that are
intimately involved in the G1/S checkpoint.26,27 Therefore, while it is clear that the G1/S
checkpoint is still intact in the absence of BRCA1, some events that are necessary to take place
(or to avoid) require a functional BRCA1 protein.
After the disastrous effect deletion of BRCA1 had on embryogenesis, a few studies sought
to find types of BRCA1 knockouts that would allow development of embryos, at least until
fibroblasts could be harvested. It has long been known that a truncated form of BRCA1, lacking
the coding sequence from exon 11, is expressed in cells, albeit to a much lower level than full
length and is mostly only expressed in developing embryos.28 One study developed a transgenic
mouse that only expressed this truncated form of BRCA1, called BRCA1Δ11. Mouse embryo
fibroblasts derived from these animals senesced much faster than wild-type cells and harbored
a plethora of genomic abnormalities. The chromosomal alterations seemed to be the result of
unequal recombination and breakage—two anomalies that predominantly occur during mitosis.
Indeed, these fibroblasts treated with DNA damaging agents failed to enact their G2/M checkpoint,
and progressed into mitosis as if there was no chromosomal damage. One possible factor
in the increase of chromosomal breakage in BRCA1Δ11 MEFs was the amplification of
centrosomes in mitotic cells. Pulling on chromatids in several directions as opposed to just to
polar ends of the dividing cell could certainly have an effect on the state of the chromosomes
after anaphase. As previously mentioned, the centrosomes are a prime location for BRCA1
during this process—perhaps the lack of full length protein at these complexes could account
for the amplification. The overall lack of G2/M checkpoint control pointed clearly at the
involvement of the exon 11 region of BRCA1 as an absolutely necessary factor in qualifying
cells for division. The overexpression of the C-terminus of BRCA1 in normal breast epithelial
cells has also been shown to adversely affect G2/M checkpoint control.29
From this data it is then clear that cells lacking a wild-type, DNA damage-responsive
BRCA1 proceed throughout the cell cycle past mitosis regardless of the anomalies that occur
during mitosis. These aberrations pile up and lead to a cell that is genetically incapable of
130 Cell Cycle Checkpoints and Cancer
dividing. In the case of the G1/S checkpoint, BRCA1 is apparently not required to halt the cell
cycle in the case of damage; however, it is necessary for proper repair of damage that may occur
during or prior to S phase. On the other hand, in human cancer, mutations in BRCA1 lead to
unrestricted cell growth, therefore, these transgenic and knockout models do not necessarily
reflect the true in vivo nature of cancerous BRCA1 protein. In breast cancer, BRCA1 is not
entirely deleted, usually mutations result in single amino acid changes in the N-terminal to middle
portions of the protein or in small truncations at the C-terminus. Such miniscule mutations may
permit semi-normal function of a particular cell, yet at the same time allow genetic alterations
here and there to slip by the damage sensing that is otherwise detected by BRCA1. The right
(or wrong, depending on your point of reference) mutations left undetected could cut the
brakes on the speed of cell growth.
Interactions with Cell Cycle Proteins
While being post-translationally modified by the cell cycle machinery and also controlling
critical steps of the checkpoint pathways, it seems logical that BRCA1 would associate with
proteins that are part of this process. Following the findings that BRCA1 is phosphorylated in
a cell cycle dependent manner, a number of reports detailed the interaction of BRCA1 with
cyclins, cyclin-dependent kinases, E2Fs and the Rb protein. Not surprisingly, BRCA1 binds
cyclin-dependent kinase 2, and interacts with it as an active kinase, as immunoprecipitation of
BRCA1 co-precipitates with kinase activity12. The activating cyclin in this complex bound to
BRCA1 appears to be cyclin A. Whether or not cyclin E, another cyclin Bound to cdk2 that allows
phosphorylation of BRCA1 in vitro, is also bound to BRCA1 in cells remains to be seen. A slew of
other cell cycle machinery proteins have also been described to bind BRCA1 including cdc2,
cyclins B and D, cdks 2 and 4 and E2F4, however the functional significance of these interactions
has yet to be determined.30
While BRCA1 is dependent, at least in some cell types, on the presence of pRb to arrest
cell growth,27 BRCA1 is also able to interact with pRb,31 BRCA1 was found to interact directly
with pRb as well as the pRb interacting proteins RbAP48/46. This binding also led to an
indirect association with histone deacetylase 1 (HDAC1), significant in that there has been
much speculation as to the involvement of BRCA1 in transcriptional control. This association
provides at least an indirect link to an enzymatic process that has been definitively shown to be
involved in transcription. The effect these interactions have on the progress of the cell cycle or with
the activity of pRb has not been found. It will be interesting to see if the phosphorylation of pRb
at the G1/S border, approximately the same temporal location as BRCA1 phosphorylation, has
any effect on the interaction with BRCA1.
A few other proteins that were not originally thought to play a role in cell cycle control
also appear to associate with BRCA1 in a cell cycle specific manner. BARD1, a protein that is
structurally similar to BRCA1, was one of the first proteins found to bind directly to BRCA1.32
The nuclear dot pattern that is characteristic of BRCA1 during S phase also is true for BARD1.
This foci formation of BARD1 only occurs during S phase, and the dots are overlapping with
BRCA1 foci, indicating that association of the two proteins may be cell cycle specific.24,33 The
overall expression of BARD1 however appears to be ubiquitous throughout the cell cycle,
therefore it is likely that post-translational modifications such as phosphorylation may be a
determining factor for this association. While clearly being a true binding partner to BRCA1 in
vivo, the function of BARD1 and the significance of BRCA1 interaction remains elusive. The
copurification of BARD1 with CstF-50, an mRNA stabilization factor, could be a lead on a
possible involvement in BRCA1 transcriptional control.34
The BRCA1 binding protein CtIP has been characterized as a protein that is able to
inhibit the transcriptional activation of promoters such as p21WAF1 by BRCA1.35-37 It is in a
complex with BRCA1 and BARD1, but in contrast to BARD1, CtIP is in fact expressed in a cell
cycle dependent manner, roughly mirroring the expression pattern of BRCA1.37 The interaction
of the two proteins is therefore cell cycle specific. While inhibiting BRCA1’s transcriptional
Functional Interactions between BRCA1 and the Cell Cycle 131
activity, the association has recently been shown to be broken by phosphorylation of BRCA1
by the gamma-irradiation responsive kinase ATM.22 The removal of CtIP allows BRCA1
transcription activation to proceed. As this activation of BRCA1 is true for ionizing radiation,
another study has found that the CtIP interaction is stable throughout several other DNA
damaging stimuli such as UV, adriamycin or hydrogen peroxide.37 Whether BRCA1 transcription
activity is affected by these other treatments is yet to be known.
Transcription of Cell Cycle Genes
It is well established that BRCA1 is likely involved in cell cycle checkpoint maintenance
and/or DNA damage sensing. However, another faction of BRCA1 research is the link of
BRCA1 to transcription of specific genes (Fig. 3). Early on, BRCA1 was found by biochemical
purification to be associated with the RNA polymerase holoenzyme and to induce transcription
from a synthetic promoter when tethered to it.38,39 Nevertheless, this data merely associated
BRCA with general transcription, not differential specific activation or repression. Indeed, no
evidence had been shown that BRCA1 possesses a promoter binding sequence nor had it been
able to bind DNA directly. However, much work since then has suggested that the expression
of BRCA1 forces changes in the expression patterns of certain genes that are not so coincidentally
linked to processes it has been proven to be involved in such as DNA damage response and cell
cycle progression. BRCA1 has been shown to be bound to such transcriptional regulators as p53,
c-Myc, the estrogen receptor and p300.40-43 Its effect on each of these proteins’s activity is what
one would expect from a tumor suppressor—co activation of p53, inhibition of c-Myc, etc.
Activation of the p21WAF1 gene expression was the first in a line of BRCA1 regulated
genes to be discovered.26 Overexpression of BRCA1 strongly activates the p21WAF1 promoter
and interestingly is not dependent on p53. Expression of BRCA1 from an exogenous promoter,
as previously mentioned, is able to cause growth arrest in most cell lines. This arrest in
G1 phase is dependent on the presence of p21WAF1, as expression of BRCA1 in isogenic cell lines
lack p21WAF1 are able to progress into DNA replication and eventually arrest in G2 phase.48
Two reports describing expression of BRCA1 in different cell lines identified the DNA
damage response gene GADD45 as a strong target of BRCA1 as well. In both, an induction of
GADD45 mRNA was clearly visible by northern blot analysis after BRCA1 transcript induction.
In one case, BRCA1 induction had an apoptotic effect in the U2OS cell line by use of a
tetracycline inducible system.44 As GADD45 had been previously shown to activate the
proapoptotic c-Jun N-terminal kinase a link was drawn between JNK and subsequent apoptosis,
by coexpression of a dominant negative mutant of JNK and a concomitant decrease in apoptosis
seen by BRCA1 expression. However, recent evidence has suggested that GADD45 in fact is
not involved in JNK activation and that GADD45 is not a component of apoptosis induction
in vivo.45-47 Therefore, whether BRCA1 causes apoptosis by activating GADD45 expression,
or if in fact this is a cell-type specific effect has yet to be determined.
In another case, an adenovirus expressing BRCA1 was infected into several different cell
lines with varying cell cycle changes, but no apoptosis.48 A common theme among all the cell
cycle changes in each cell type was an increase in the G2/M phase content. Interestingly,
GADD45 knockout mice possess a defective G2/M checkpoint, and the involvement of BRCA1
in the maintenance of genomic stability at mitosis has already been established.28,47 Therefore,
it is possible that BRCA1 may in part induce the GADD45 protein in the interest of activation
of the G2/M checkpoint.
Cyclin B1 is also an apparent target of BRCA1, however its expression levels are decreased
by the exogenous expression of BRCA1.48 Again, the links to the involvement in G2/M phase
control are evident. Upon the detection of anomalies in mitosis, the first step in allowing repair
to proceed is to halt progression of the cycle. The inactivation of cdc2 kinase by depletion of
the activating cyclin B1 is one way to accomplish this goal. In this study, a G2/M arrest induced
by BRCA1 overexpression was found to be abrogated by coexpression of exogenous
cyclin B1, indicating that one way by which BRCA1 is able to arrest the cell cycle at G2/M
132 Cell Cycle Checkpoints and Cancer
phase is by repression of cyclin B1. It will be interesting to see the possible involvement of the
histone deacetylase 1 enzyme bridged to BRCA1 by RbAP46/48 in this process.
The induction of the serum responsive, early response growth factor 1 (EGR1) has also
been found to be induced by BRCA1 in array screening for BRCA1 targets,44 EGR1 belongs to
a group of proteins that are involved in the progress through G1 phase of the cell cycle following
growth factor stimulation, which includes c-Myc, c-Jun and c-Fos. These other immediate
early genes however are not induced by BRCA1. The induction of EGR1 by BRCA1 may be
indirect, owing to its late upregulation and abrogation of the JNK pathway.
Finally, the physical link of BRCA1 to the regulatory regions of these genes that it is
proposed to affect has been lacking. However, a recent finding of a protein, termed ZBRK1,
has provided some insight into this necessary interaction,49 BRCA1 binds directly to ZBRK1,
which in turn binds to a consensus DNA sequence defined as GGGxxxCAGxxxTTT. Interestingly,
this sequence resides in many of the genes described in recent papers to be up or down
regulated as a result of BRCA1 expression. These include p21 (3 sites), GADD45, Gadd153,
Ki-67 and EGR1. Co expression of ZBRK1 and BRCA1 were found to actually repress the
GADD45 promoter, contrary to what one would expect given previous findings of activation
of GADD45 expression by BRCA1. Thus far, it is hypothesized that overexpression of BRCA1
may titrate ZBRK1 away from the promoter, allowing transcription to occur. In vivo, the
situations may be different, as they involve such other events as phosphorylation by a number
of kinases, binding to other repressors such as CtIP and cell cycle specific localization.
Conclusion
The links between BRCA1 and the cell cycle have been made clear with solid data published
over the last few years. Advances in the field have allowed visualization of BRCA1 localization
during the cell cycle, during mitosis, meiosis and DNA replication that provide insight
Fig. 3: Activation of genes by BRCA1 relevant to the cell cycle. BRCA1 has been shown to control the
expression levels of four genes—p21WAF1, GADD45, Cyclin B1 and EGR1—that have implications on
checkpoint activation. The p21WAF1 and EGR1 genes are involved in G1 phase of the cell cycle and may effect
the onset of a G1 block prior to DNA replication in the face of DNA damage. The induction of GADD45
and repression of Cyclin B1 may represent an attempt by BRCA1 to halt progression towards mitosis.
Functional Interactions between BRCA1 and the Cell Cycle 133
into how BRCA1 functions with respect to the cell cycle. Expression at both the mRNA and
protein levels have also been linked to this.
Further study into this association between BRCA1 and the cell cycle will allow delineation
such issues as whether BRCA1 is a regulator or is regulated by the progression of the cell
cycle, how phosphorylation affects the localization and function of BRCA1 during this progress
and if transcriptional control by BRCA1 is a requirement for the normal transition from phase
to phase of the mammalian cell cycle clock.
References
1. Alberg AJ, Helzlsouer KJ. Epidemiology, prevention and early detection of breast cancer. Curr
Opin Oncol 1997; 9:505-511.
2. Easton DF, Bishop DT, Ford D et al. Genetic linkage analysis in familial breast and ovarian
cancer: results from 214 families. The Breast Cancer Linkage Consortium. Am J Hum Genet 1993;
52:678-701.
3. Miki Y, Swensen J, Shattuck-Eidens D et al. A strong candidate for the breast and ovarian cancer
susceptibility gene BRCA1. Science 1994; 266:66–71.
4. Gudas JM Li T Nguyen H et al. Cell cycle regulation of BRCA1 messenger RNA in human breast
epithelial cells. Cell Growth Diff 1996; 7:717-23.
5. Arizti P,Fang L,Park I et al. Tumor suppressor p53 is required to modulate BRCA1 expression.
Mol Cell Biol 2000; 20:7450-7459.
6. MacLachlan TK, Dash BC, Dicker DT et al. Repression of BRCA1 through a feedback loop involving
p53. J Biol Chem 2000; 275: 31869-31875.
7. Wang A, Schneider-Broussard R, Kumar AP et al. Regulation of BRCA1 expression by the Rb-E2F
pathway. J Biol Chem 2000; 275:4532-6.
8. Scully R, Chen J, Ochs RL et al. Dynamic changes of BRCA1 subnuclear location and phosphorylation
state are initiated by DNA damage. Cell 1997; 90:425-35.
9. Chen Y, Farmer AA, Chen CF et al. BRCA1 is a 220-kDa nuclear phosphoprotein that is expressed
and phosphorylated in a cell cycle-dependent manner. Cancer Res 1996; 56(14):3168-72.
10. Ruffner H, Verma IM. BRCA1 is a cell cycle-regulated nuclear phosphoprotein. Proc Natl Acad
Sci USA 1997; 94:7138-43.
11. Thomas JE, Smith M, Tonkinson JL et al. Induction of phosphorylation on BRCA1 during the
cell cycle and after DNA damage. Cell Growth Diff 1997; 8:801-9.
12. Ruffner H, Jiang W, Craig AG et al. BRCA1 is phosphorylated at serine 1497 in vivo at a cyclindependent
kinase 2 phosphorylation site. Mol Cell Biol 1999;19:4843-54.
13. O’Brien KA, Lemke SJ, Cocke KS et al. Casein kinase 2 binds to and phosphorylates BRCA1.
Biochem Biophys Res Comm 1999; 260:658-64.
14. Gatei M, Scott SP, Filippovitch I et al. Role for ATM in DNA damage-induced phosphorylation
of BRCA1. Cancer Res 2000; 60:3299-304.
15. Kim ST, Lim DS, Canman CE et al. Substrate specificities and identification of putative substrates
of ATM kinase family members. J Biol Chem 1999; 274:37538-43.
16. Cortez D, Wang Y, Qin J et al. Requirement of ATM-dependent phosphorylation of BRCA1 in
the DNA damage response to double-strand breaks. Science 1999; 286:1162-6.
17. Chen J. Ataxia telangiectasia-related protein is involved in the phosphorylation of BRCA1 following
deoxyribonucleic acid damage. Cancer Res 2000; 60:5037-9.
18. Tibbetts RS, Cortez D, Brumbaugh KM et al. Functional interactions between BRCA1 and the
checkpoint kinase ATR during genotoxic stress. Genes Dev 2000; 14: 2989-3002.
19. Lee JS, Collins KM, Brown AL et al. hCds1-mediated phosphorylation of BRCA1 regulates the
DNA damage response. Nature 2000; 404:201-4.
20. Altiok S, Batt D, Altiok N et al. Heregulin induces phosphorylation of BRCA1 through
phosphatidylinositol 3-Kinase/AKT in breast cancer cells. J Biol Chem 1999; 274:32274-8.
21. Zhang HT, Zhang X, Zhao HZ et al. Relationship of p215BRCA1 to tyrosine kinase signaling
pathways and the cell cycle in normal and transformed cells. Oncogene 1997; 14:2863-9.
22. Li S, Ting NS, Zheng L et al. Functional link of BRCA1 and ataxia telangiectasia gene product in
DNA damage response. Nature 2000; 406:210-5.
23. Hsu LC, White RL. BRCA1 is associated with the centrosome during mitosis. Proc Natl Acad Sci
USA 1998; 95:12983-8.
24. Scully R, Chen J, Plug A et al. Association of BRCA1 with Rad51 in mitotic and meiotic cells.
Cell 1997; 88:265-75.
134 Cell Cycle Checkpoints and Cancer
25. Hakem R, de la Pompa JL, Sirard C et al. The tumor suppressor gene BRCA1 is required for
embryonic cellular proliferation in the mouse. Cell 1996; 85:1009-23.
26. Somasundaram K, Zhang H, Zeng YX et al. Arrest of the cell cycle by the tumour-suppressor
BRCA1 requires the CDK-inhibitor p21WAF1/CiP1. Nature 1997; 389:187-90.
27. Aprelikova ON, Fang BS, Meissner EG et al. BRCA1-associated growth arrest is RB-dependent.
Proc Natl Acad Sci USA 1999; 96:11866-71.
28. Xu X, Weaver Z, Linke SP et al. Centrosome amplification and a defective G2-M cell cycle checkpoint
induce genetic instability in BRCA1 exon 11 isoform-deficient cells. Mol Cell 1999; 3:389-95.
29. Larson JS, Tonkinson JL, Lai MT. A BRCA1 mutant alters G2–M cell cycle control in human
mammary epithelial cells. Cancer Res 1997; 57:3351–3355.
30. Wang H, Shao N, Ding QM et al. BRCA1 proteins are transported to the nucleus in the absence
of serum and splice variants BRCA1a, BRCA1b are tyrosine phosphoproteins that associate with
E2F, cyclins and cyclin dependent kinases. Oncogene 1997; 15:143-57.
31. Yarden RI, Brody LC. BRCA1 interacts with components of the histone deacetylase complex. Proc
Natl Acad Sci USA 1999; 96:4983-8.
32. Wu LC, Wang ZW, Tsan JT et al. Identification of a RING protein that can interact in vivo with
the BRCA1 gene product. Nat Genet 1999; 14:430-40.
33. Jin Y, Xu XL, Yang MC et al. Cell cycle-dependent colocalization of BARD1 and BRCA1 proteins
in discrete nuclear domains. Proc Natl Acad Sci USA 1997; 94:12075-80.
34. Kleiman FE, Manley JL. Functional interaction of BRCA1-associated BARD1 with polyadenylation
factor CstF-50. Science 1999; 285:1576-9.
35. Li S, Chen PL, Subramanian T et al. Binding of CtIP to the BRCT repeats of BRCA1 involved in the
transcription regulation of p21 is disrupted upon DNA damage. J Biol Chem 1999; 274:11334-8.
36. Wong AK, Ormonde PA, Pero R et al. Characterization of a carboxy-terminal BRCA1 interacting
protein. Oncogene 1998; 17:2279-85.
37. Yu X, Baer R. Nuclear localization and cell cycle-specific expression of CtIP, a protein that associates
with the BRCA1 tumor suppressor. J Biol Chem 2000; 275:18541-9.
38. Scully R, Anderson SF, Chao DM et al. BRCA1 is a component of the RNA polymerase II
holoenzyme. Proc Natl Acad Sci USA 1997; 94:5605-10.
39. Chapman MS, Verma IM. Transcriptional activation by BRCA1. Nature 1996; 382:678-9.
40. Somasundaram K, MacLachlan TK, Burns TF et al. BRCA1 signals ARF-dependent stabilization
and coactivation of p53. Oncogene 1999; 18:6605-14.
41. Zhang H, Somasundaram K, Peng Y et al. BRCA1 physically associates with p53 and stimulates its
transcriptional activity. Oncogene 1998; 16(13):1713-21.
42. Wang Q, Zhang H, Kajino K et al. BRCA1 binds c-Myc and inhibits its transcriptional and
transforming activity in cells. Oncogene 1998; 17:1939-48.
43. Fan S, Wang J, Yuan R et al. BRCA1 inhibition of estrogen receptor signaling in transfected cells.
Science 1999; 284:1354-6.
44. Harkin DP, Bean JM, Miklos D et al. Induction of GADD45 and JNK/SAPK-dependent apoptosis
following inducible expression of BRCA1. Cell 1999; 97:575-86.
45. Shaulian E, Karin M. Stress-induced JNK activation is independent of GADD45 induction. J Biol
Chem 1999; 274:29595-8.
46. Wang X, Gorospe M, Holbrook NJ. GADD45 is not required for activation of c-Jun N-terminal
kinase or p38 during acute stress. J Biol Chem 1999; 274:29599-602.
47. Hollander MC, Sheikh MS, Bulavin DV et al. Genomic instability in GADD45a-deficient mice.
Nat Genet 1999; 23:176-84.
48. MacLachlan TK, Somasundaram K, Sgagias M et al. BRCA1 effects on the cell cycle and the
DNA damage response are linked to altered gene expression. J Biol Chem 2000; 275:2777-85.
49. Zheng L, Pan H, Li S et al. Sequence-Specific Transcriptional Corepressor Function for BRCA1
through a novel zinc finger protein, ZBRK1. Mol Cell 2000; 6:757-764.
CHAPTER 8
The Role of FHIT in Carcinogenesis
Yuri Pekarsky, Kay Huebner and Carlo M. Croce
Abstract
Tumor suppressor genes and oncogenes can be identified by positions of chromosomal
translocations (in leukemia and lymphomas) and by detection of homozygous dele
tions and loss of heterozygosity (in solid tumors). Using these approaches, we
identified a specific locus from the short arm of chromosome 3, cloned, and characterized the
FHIT gene at 3p14.2 which is involved in chromosomal abnormalities in most common human
tumors. In lung cancer associated with smoking, inactivation of FHIT occurs very early in
tumor development. In other tumors, such as clear cell renal carcinoma and breast carcinoma,
this inactivation occurs in later stages of tumor progression. Thus, evaluation of FHIT
expression in premalignant lesions and tumors may be important in the diagnosis and
prognosis of human cancer
Chromosomal Changes in Cancer
Carcinogenesis is a complex process which includes consecutive genetic changes affecting
oncogenes and tumor suppressor genes. Most human malignancies show multiple genetic
abnormalities and are heterogenous even if they originate from a single cell. Therefore, different
cells of the same tumor may be affected differently by chemotherapy, gene therapy or
radiation therapy.
Chromosomal rearrangements are associated with human malignancies involve most
human chromosomes.1,2 Four major types of such cytogenetic rearrangements have been observed:
deletions, amplifications, translocations and inversions. Most human leukemias and lymphomas
show consistent chromosomal rearrangements, mostly translocations or inversions, resulting in
activation of oncogenes3,4 or in loss of function of tumor suppressor genes,5 causing malignant
transformation.
In solid tumors, chromosomal deletions cause inactivation of tumor suppressor genes.
This inactivation usually results in abnormal cell cycle control and/or increased cell survival
leading to malignant transformation. A prototypical example of such loss is observed in human
retinoblastomas where a deletion within the 13q14 region causes the inactivation of the retinoblastoma
tumor suppressor gene.1 Other such examples are deletions of the p53 gene and
Wilm’s tumor gene.6 In many of these cases, the first copy of the gene is deleted while the
second copy is mutated (mutations can be somatic or germline). These genetic changes result
in inactivation of tumor suppressors in malignant cells.7 Although the development of some solid
tumors is caused by the inactivation of known tumor suppressor genes, the initiating events in the
most common human cancers, including breast, prostate and lung cancer, are not known.
The identification and understanding of the earliest genetic changes that initiate transformation
in solid tumors and detection of these events in premalignant lesions may result in the
development of new therapeutic approaches (or novel drugs) to destroy premalignant cells
providing new opportunities for cancer prevention. In addition, we could use the proteins
Cell Cycle Checkpoints and Cancer, edited by Mikhail V. Blagosklonny. ©2001 Eurekah.com.
136 Cell Cycle Checkpoints and Cancer
encoded by the genes involved in these early steps of tumor development (or their biochemical
pathways) as targets for novel therapeutical agents.
FHIT Loci is the Target of Chromosomal Abnormalities at 3p14.2
Chromosomal deletions and loss of heterozygosity involving the short arm of chromosome 3
(3p) have been described as a frequent event in most common epithelial tumors such as breast,
lung and kidney cancers.8-12 Cytogenetic analysis of bronchial dysplastic lesions11 and lung
tumors10 has shown the frequent involvement of the short arm of chromosome 3 in these
malignancies. Therefore it seemed possible that rearrangements in gene(s) on the short arm of
chromosome 3p may occur in preneoplastic lesions (bronchial dysplasia) causing the development
of malignant tumors. Although four major regions of 3p (3p25, 3p21.3, 3p14.2 and 3p12) are
involved in allelic losses in cancer, the region 3p14.2 attracts most of the interest. In addition to
its involvement in cancer deletions, this region contains a familial kidney cancer-associated
3p14.2 translocation break t(3;8)(p14.2;q24),13 most common human fragile site, FRA3B,14
and papilloma virus integration sites15 (Fig. 1).
To identify gene(s) at 3p14.2 affected by these alterations, it was necessary to define
precisely the region involved in deletions. First, we investigated the loss of heterozygosity (LOH)
of gastrointestinal and kidney malignancies, evaluating the presence of markers on 3p in the
DNA of tumor tissues and their normal counterparts. The 3p14.2 region is often involved in
hemi-and homozygous deletions in human tumors, thus narrowing the region of interest to
several hundred kilobases of DNA.16 We also determined that this critical region overlapped
with the one involved in homozygous deletions in tumor-derived cell lines and primary tumors
and was located very close to the t(3;8)(p14.2;q24) chromosome translocation breakpoint in a
hereditary renal cell carcinoma.16 To identify gene(s) located in this narrowed region we
constructed a cosmid contig covering the region and used this contig in exon trapping
experiments. A single exon was identified then subjected to RACE (rapid amplification of cDNA
ends) analysis leading to the discovery of 1.1 kb ubiquitously expressed FHIT. This 1.1 kb mRNA
encoded a 17 kD FHIT protein containing 147 amino acids and showing ~70% similarity to a
diadenosine 5´,5"’-P1,P4-tetraphosphate (Ap4A) hydrolase, a member of a histidine triad protein
(HIT) family from the fission yeast Schizosaccaromyces pombe.16 This newly identified gene was
named FHIT because it encodes a HIT family protein containing the most common human
fragile site. Because of the similarity of FHIT and yeast (Ap4A) hydrolase, FHITt was tested for
this activity. These experiments demonstrated that FHIT possesses Ap3A (diadenosine 5´,5"’-
P(1),P(3)-triphosphate) hydrolytic activity: it cleaved Ap3A into ADP and AMP.17 Physical
mapping of 5´ and 3´ ends of FHIT mRNA revealed that FHIT is one of the largest human
genes, it spans a 1-2 mb fragment of genomic DNA. Further genomic analysis revealed that
FHIT contains 10 small exons, whereas exon 5 contains an initiating ATG codon (Fig.1). The
most common human fragile site, FRA3B, a papilloma virus integration site and a
translocation breakpoint of t(3;8)(p14.2;q24) associated with familial kidney cancer were mapped
within the FHIT gene.16 Some of the homozygous deletions within the FHIT locus observed
in tumor derived cell lines are shown on Figure 1. Siha, a cervical carcinoma-derived cell line,
has two overlapping allelic deletions, thus generating a small homozygous deletion in intron 4.
Therefore, Siha cells do not encode a full-length FHIT mRNA nor protein. Similarly, Kato III,
a gastric cancer cell line exhibits independent deletions in both alleles (Fig. 1), although these
cells retain a copy of each FHIT exon. Thus, Kato III and Siha can not express a full-length
FHIT mRNA or a FHIT protein.18 Both, AGS, a stomach carcinoma-derived cell line and
LS180, a colon carcinoma cell line exhibit three separate regions of homozygous deletions as
shown on Figure 1.
Taking together the observation that some tumors and tumor-derived cell lines contain
homozygous deletions in the FHIT gene and that translocation breakpoints of t(3;8)(p14.2;q24)
are associated with familial kidney cancer and LOH in most common human malignancies, we
conclude that FHIT indeed is a target of chromosomal abnormalities at 3p14.2 observed in
human cancer.
The Role of FHIT in Carcenogenesis 137
Inactivation of FHIT mRNA and Protein Expression in Cancer
Given that FHIT is involved in LOH and chromosomal deletions at 3p14.2, we carried
out detailed studies of the FHIT mRNA and protein expression in cervical, stomach, lung,
kidney, breast tumors, in both tumor-derived cell lines and primary tumors (Fig. 2). As determined
by immunohistochemistry, FHIT is abundantly expressed in epithelial tissues of all
human organs including lung, stomach and kidney.19-21 As expected, immunohistochemical
studies and immunoblot analysis of human malignancies demonstrated that tumors and cell
lines expressing altered FHIT transcripts, with genomic rearrangements of the FHIT locus,
usually do not express FHIT or express reduced levels of FHIT.18-20 We investigated a large collection
(474 cases) of stage 1 nonsmall cell lung cancers (NSCLC) by immunohistochemistry.21
FHIT was not expressed in 73% of these tumors (Fig. 2), indicating a very high frequency of
loss of FHIT expression in NSCLC (Table 1). A significant difference in the frequency of loss
of its expression in adenocarcinoma (59%) versus squamous cell carcinoma (87%) has been
found. Loss of FHIT expression was observed in 69% of large cell lung cancers. Interestingly,
loss of FHIT expression was less frequent in tumors of nonsmokers than smokers (Table 1).
We also examined levels of FHIT in precancerous lesions by immunohistochemistry. Loss
of FHIT was already apparent in bronchial dysplasia, a precancerous condition, indicating this
is an early event in the development of lung cancer (Table 1). Interestingly, loss of the FHIT
protein was more frequent and occurred earlier than alterations of the EGF receptor and more
common than p53 mutation in lung cancer and preneoplastic lesions.21 It has also been
reported that allele loss at the FHIT locus is more frequent in the bronchial epithelium and tumors
of smokers than nonsmokers and more frequent than loss at other tumor suppressor regions.22
Further studies have demonstrated that abnormalities in FHIT mRNA and protein in
the most common human malignancies (Fig. 2). When compared to normal levels in epithelium,
FHIT was absent or reduced in 67% of stomach tumors,18, 20 40-89% of kidney tumors,16,18,23
76% of cervical tumors,24 62% of pancreatic tumors25 and 30%-70% of breast carcinomas.26-28
Therefore, the FHIT gene is normally expressed almost exclusively in the epithelial tissues, an
origin of most common human neoplasias. It is inactivated in more than 50% of these tumors,
more frequently than any other known tumor suppressor gene.
Fig. 1. The map of FRA3B/FFHIT loci at 3p14.2. The FRA3B region is showed by the solid area at p14.2.
FHIT exons are numbered 1 through 10, coding exons are in black. Positions of viral integration sites and
t(3;8) translocation are marked with arrows. Gaps in the lines represent FHIT locus in the tumor cell lines
demonstrate deletions.
138 Cell Cycle Checkpoints and Cancer
Despite the involvement of the FHIT gene in most common human cancers, molecular
mechanisms of FHIT deletions remain unknown. To examine molecular basis of chromosomal
fragility at 3p14.2, we sequenced 870 kilobases of the FHIT/FRA3B locus, spanning intron 3
through intron 7 of FHIT.29,30 The sequence of the region surrounding exon 5, a center of the
fragile site (Fig. 1), revealed that it is poor in Alu and rich in LINE 1 repeat elements. Interestingly,
most of breakpoints that are involved in deletions in a variety of human tumor-derived
cell lines occur in long terminal repeats (LTR) or LINE 1 elements larger than 1 kb suggesting
these deletions may be caused by the homologous recombination between these elements.29
The Tumor Suppressor Activity of FHIT
Given that both FHIT alleles are frequently altered in human tumors and the translocation
associated with hereditary kidney cancer disrupts one FHIT allele,16 it is reasonable to
consider FHIT as a tumor suppressor gene.16
To demonstrate that FHIT is a bona fide tumor suppressor,31 the human FHIT cDNA
was transfected into four FHIT-negative tumor cell. Then FHIT-expressing cells were implanted
into nude mice. In these experiments, FHIT-expressing cells had lost their ability to form
tumors, suggesting the tumor suppressor activity of FHIT.31 Also, mutant FHIT protein with
reduced enzymatic activity suppressed tumorigenicity, suggesting that FHIT enzymatic activity
is not required for tumor suppression.31
Using adenovirus, FHIT cDNA was transduced into human lung cancer cells and head
and neck carcinoma cells lacking FHIT expression.32 In these cells but not in normal
human bronchial epithelial cells, re-expression of FHIT inhibited cell growth, induced apoptosis
and accumulation of cells in S phase.32 Less than 20% of Ad-FHIT-infected cancer cells
survived by day 7. Furthermore,in vivo tumorigenicity of H1299 human lung cancer cells was
eliminated by infection with Ad-FHIT but not with the empty adenoviral vector. The expression
of FHIT in H460, a FHIT-negative human lung cancer cell, inhibited cell growth and
induced p53-independent apoptosis, suggesting proapoptotic functions of FHIT. 33
The ultimate proof of a tumor suppressor activity is the development of neoplasias in
knockout mouse models. Although it is still early to discuss the phenotype of FHIT-/- mouse,
the encouraging results describing carcinogen induced tumors in FHIT+/- mice have recently
been reported.34 In this study FHIT +/+ and +/- mice were treated intragastrically with carcinogen
nitrosomethylbenzylamine and observed for 10 wk posttreatment. A total of 25% of
the +/+ mice developed adenoma or papilloma of the forestomach, whereas 100% of the +/-
mice developed multiple tumors that were a mixture of adenomas, squamous papillomas,
invasive carcinomas of the forestomach, as well as tumors of sebaceous glands. The visceral
and sebaceous tumors, which lacked FHIT protein, were similar to those of Muir-Torre familial
cancer syndrome.
Fig. 2. Expression of FHIT in most common human tumors.
The Role of FHIT in Carcenogenesis 139
Taken together, results in vitro and in vivo imply that the FHIT gene is a tumor
suppressor gene.
Toward FHIT Function
As discussed above, the FHIT protein in vitro possesses the Ap3A hydrolase activity17 and
this activity is not required for tumor suppression.31 Another study reported that mutant FHIT
which can not cleave Ap3A but suppresses tumorigenicity in vivo 31 still bind Ap3A.35 This
suggests that the Ap3A bound form of FHIT may be the active tumor suppressor.
To investigate whether the loss of the FHIT gene would cause any phenotype in Drosophila
melanogaster and C. elegans, we cloned the Drosophila and C. elegans FHIT genes.36
Although Drosophila and C. elegans FHIT mutants were not available, we discovered that
both proteins occur as a fusion proteins of ~450 amino acids in which a C-terminal FHIT
domain is fused with a novel 35 kD domain showing homology with yeast and bacterial
proteins of unknown function and with plant nitrilases.36 Therefore, the~450 amino acid
Drosophila and C. elegans FHIT proteins that contain an N-terminal nitrilase domain and a
C-terminal FHIT domain were designated as NitFHIT (Fig. 3). Like human FHIT, Drosophila
NitFHIT possesses Ap3A hydrolyzing activity.36 Thus, in Drosophila and C. elegans.
FHIT is a chimeric protein, possibly with a dual enzymatic function. We next isolated human
and murine NIT homologs. These separate genes, designated as NIT1, are located on human
chromosome 1q21 and mouse chromosome 1.
In several eukaryotic biosynthetic pathways, multiple steps are catalyzed by multi-enzymatic
proteins containing two or more functional domains. In prokaryotes, the same steps are often
carried out by a single enzyme with homology to individual domains of the corresponding
eukaryotic protein.37 For example, Gars, Gart and Airs are domains of the same protein in
Drosophila and mammals. These domains catalyze different steps in de novo synthesis of purines.
All three homologs (PurM, PurN and PurD) are separate proteins in bacteria. In yeast, Gars
and Airs homologs (Ade5 and Ade7) are domains of a bi-enzymatic protein and the Gart
homolog (Ade8) is a separate protein.37 If domains of a multi-enzymatic protein are expressed
as individual proteins in other organisms, these individual proteins participate in the same pathway.
38 This observation implies that FHIT and Nit1 may participate in the same molecular pathway
in mammalian cells. Further functional studies are required to determine these pathway(s).
Table 1. Loss of FHIT expression in stage 1 non small cell lung carcinomas and
preinvasive bronchial lesions
Tumor Type % FHIT negative Total Cases
Bronchial lesions: 0
Dysplasia 85% 20
Carcinoma in situ 100% 25
Lung carcinoma: 0
Adenocarcinoma 57% 196
Squamous cell carcinoma 87% 233
Smokers 75% 451
Nonsmokers 39% 23
140 Cell Cycle Checkpoints and Cancer
In a recent report,39 the interaction between either wild-type or mutant FHIT and tubulin
in vitrois described. Both wild-type and mutant forms of FHIT specifically bind to tubulin
without causing nucleation or formation of microtubules, promoting their assembly to a greater
extent than microtubule-associated proteins alone.39
In yeast two hybrid system, which identified proteins physically interacting with FHIT,
the human ubiquitin-conjugating enzyme 9 (hUBC9) specifically interacted with FHIT.40
The C-terminal region of FHIT is responsible for this interaction. Given that yeast UBC9 is
involved in the degradation of S- and M-phase cyclins, FHIT may be also involved in cell cycle
control through its interaction with hUBC9.40 Additional biochemical studies are necessary to
determine the exact role of FHIT in this molecular pathway.
Conclusions
To identify tumor suppressor genes and oncogenes involved in the development and/or
progression of human cancer it is critical to determine their chromosomal regions. In leukemias
and lymphomas, these regions can usually be determined by cytogenetic positions of
chromosomal translocations. This approach was effective in the discovery of critical cancer genes.3,4
In most solid tumors, chromosomal translocations are rare and different approaches such as detection
of homozygous deletions and loss of heterozygosity are necessary.
By using a combination of these approaches, we identified a specific locus from the short arm
of chromosome 3, cloned and characterized the FHIT gene at 3p14.2 involved in chromosomal
abnormalities in most common human tumors. The FHIT gene is the second largest known
human gene and contains a t(3;8) translocation in familial renal cell carcinoma and the FRA3B,
the most common human fragile site.
In lung cancer associated with smoking, inactivation of FHIT occurs very early in tumor
development.21 In other tumors, such as clear cell renal carcinoma and breast carcinoma, its
inactivation occurs in later stages of tumor progression.23,28 Thus, evaluation of FHIT expression
in premalignant lesions and tumors may be important for diagnosis and prognosis.
Re-expression of FHIT protein in human cancer cell lines causes apoptosis and suppresses
tumorigenicity in vivo.31-33 The study of the FHIT knockout mouse model proved that FHIT
is indeed a tumor suppressor gene. Therefore, FHIT may be a candidate gene for gene therapy.
Future goals are to find biological functions of FHIT that affect cell growth, differentiation
and cell death. This may lead to the development of novel approaches for diagnosis and treatment
of common human cancers.
References
1. Croce CM. Role of chromosome translocations in human neoplasia. Cell 1987; 49:155-156.
2. Soloman E, Borrow J, Goddart A. Chromosome aberrations and cancer. Science 1991;
254:1153-1160.
3. Ar-Rushdi A, Nishikura K, Erikson J et al. Differential expression of the translocated and of the
untranslocated c-myc oncogene in Burkitt lymphoma. Science 1983; 222:390-393.
4. Tsujimoto Y, Cossman J, Jaffe E et al. Involvement of the Bcl-2 gene in human follicular lymphoma.
Science 1985; 228:1440-1443.
5. Arakawa H, Nakamura T, Zhadanov AB et al. Identification and characterization of the ARP1 gene,
a target for the human acute leukemia ALL1 gene. Proc Natl Acad Sci USA 1998; 95:4573-4578.
6. Weinberg R. Tumor suppressor genes. Science 1991; 254:1138-1146.
7. Hinds PW, Weinberg RA. Tumor suppressor genes. Cur Opin Genet and Devel 1994; 4:135-141.
8. Druck T, Kastury K, Hadaczek P et al. Loss of heterozygosity at the familial RCC t(3;8) locus in
most clear cell renal carcinomas. Cancer Res 1995; 55:5348-5353.
9. Kastury K, Baffa R, Druck T et al. Potential gastrointestinal tumor suppressor locus at the 3p14.2
FRA3B site identified by homozygous deletions in tumor cell lines. Cancer Res 1996; 56:978-983.
10. Naylor SL, Johnson BE, Minna JD et al. Loss of heterozygosity of chromosome 3p markers in
small cell lung cancer. Nature 1987; 329:451-454.
11. Hibi K, Takahashi T, Yamakawa K, et al. Three distinct regions involved in 3p deletions in human
lung cancer. Oncogene 1992; 7:445-449.
The Role of FHIT in Carcenogenesis 141
12. Huebner K, Hadaczek P, Siprashvili Z, et al. The FHIT gene, a multiple tumor suppressor gene
encompassing the carcinogen sensitive chromosome fragile site, FRA3B. Biochem Biophys Acta
(Reviews on Cancer) 1997; 1332:M65-M70.
13. Glover TW, Coyle-Morris JF, Frederick PL et al. Translocation t(3;8) (p14.2;q24.1) in renal cell
carcinoma affects expression of the common fragile site at 3p14 (FRA3B) in lymphocytes. Cancer
Genet Cytogenet 1988; 31:69-73.
14. Glover TW, Stein CK. Chromosome breakage and recombination at fragile sites. Am J Hum Genet
1988; 43:265-273.
15. Rassool FV, McKeithan TW, Neilly ME et al. Preferential integration of marker DNA into the
chromosomal fragile site at 3p14.2: A novel approach to cloning fragile sites. Proc Natl Acad Sci
USA 1991; 88:6657-6661.
16. Ohta M, Inoue H, Cotticelli MG, et al. The human FHIT gene, spanning the chromosome 3p14.2
fragile site and renal cell carcinoma associated translocation breakpoint, is abnormal in digestive
tract cancers. Cell 1996; 84: 587-597.
17. Barnes LD, Garrison PN, Siprashvili Z et al. FHIT, a putative tumor suppressor in humans, is a
dinucleoside 5´, 5´”-p1, p3 triphosphate hydrolase. Biochemistry 1999; 35: 11529-11535.
18. Druck T, Hadaczek P, Fu T-B et al. Structure and expression of the human FHIT gene in normal
and tumor cells. Cancer Res 1997; 57:504-512.
19. Sozzi G, Tornielli S, Tagliabue E et al. Absence of FHIT protein in primary lung tumors and cell
lines with FHIT gene abnormalities. Cancer Res 1997; 57:5207-5212.
20. Baffa R, Veronese M L, Santoro R et al. Loss of FHIT expression in gastric carcinoma. Cancer Res
1998; 58:4708-4714.
21. Sozzi G, Pastorino U, Moiraghi L et al. Loss of FHIT function in lung cancer and preinvasive
bronchial lesions. Cancer Res 1998; 58:5032-5037.
22. Mao L, Lee JS, Kurie JM et al. Clonal genetic alterations in the lungs of current and former
smokers. J Natl Cancer Inst 1997; 89: 857-862.
23. Hadaczek P, Siprashvili Z, Markiewski M et al. Absence or reduction of FHIT expression in most
clear cell renal carcinomas. Cancer Res 1998; 58:2946-2951.
Fig. 3. Nitrilase and FHIT homologs are encoded as fusion proteins in Drosophila melanogaster and
Caenorhabditis elegans.
142 Cell Cycle Checkpoints and Cancer
24. Greenspan DL, Connolly DC, Wu R et al. Loss of FHIT expression in cervical carcinoma cell
lines and primary tumors. Cancer Res 1997; 57:4692-4698.
25. Simon B, Bartsch D, Barth P et al. Frequent abnormalities of the putative tumor suppressor gene
FHIT at 3p14.2 in pancreatic carcinoma cell lines. Cancer Res 1998; 58:1583-1587.
26. Negrini M, Monaco C, Vorechovsky I et al. The FHIT gene at 3p14.2 is abnormal in breast
carcinomas. Cancer Res 1996; 56:3173-3179.
27. Ingvarsson S, Agnarsson BA, Sigbjornsdottir BI et al. Reduced FHIT expression in sporadic and
BRCA2-linked breast carcinomas. Cancer Res 1999; 59:2682-2689.
28. Campiglio M, Pekarsky Y, Menard S et al. FHIT loss of function in human primary breast cancer
correlates with advanced stage of the disease. Cancer Res 1999; 59:3866-3869.
29. Inoue H, Ishii H, Alder H et al. Sequence of the FRA3B common fragile region: Implications for
the mechanisms of FHIT deletion. Proc Natl Acad Sci USA 1997; 94: 14584-14589.
30. Mimori K, Druck T, Inoue H et al. Cancer-specific chromosome alterations in the constitutive
fragile region FRA3B. Proc Natl Acad Sci USA 1999; 96:7456-7461.
31. Siprashvili Z, Sozzi G, Barnes LD et al. Replacement of FHIT in cancer cells suppresses tumorigenicity.
Proc Natl Acad Sci USA 1997; 94: 13771-13776.
32. Ji L, Fang B, Yen N et al. Induction of apoptosis and inhibition of tumorigenicity and tumor
growth by adenovirus vector-mediated fragile histidine triad (FHIT) gene overexpression. Cancer
Res 1999; 59:3333-3339.
33. Sard L, Accornero P, Tornielli S et al. The tumor-suppressor gene FHIT is involved in the regulation
of apoptosis and in cell cycle control. Proc Natl Acad Sci USA 1999; 96:8489-8492.
34. Fong LY, Fidanza V, Zanesi N et al. Muir-Torre-like syndrome in FHIT-deficient mice. Proc Natl
Acad Sci USA 2000; 97:4742-4247.
35. Pace HC, Garrison PN, Robinson AK et al. Genetic, biochemical and crystallographic characterization
of FHIT-substrate complexes as the active, signaling form of FHIT. Proc Natl Acad Sci USA
1998; 95: 5484-5489.
36. Pekarsky Y, Campiglio M, Siprashvili Z et al. Nitrilase and FHIT homologs are encoded as fusion
proteins in Drosophila melanogaster and Caenorhabditis elegans. Proc Natl Acad of Sci USA 1998;
95: 8744-8749.
37. Davidson JN, Peterson M. Origin of genes encoding multi-enzymatic proteins in eukaryotes. Trends
Genet 1997; 13:281-285.
38. Marcotte EM, Pellegrini M, Ng HL et al. Detecting protein function and protein-protein interactions
from genome sequences. Science 1999; 285:751-753.
39. Chaudhuri AR, Khan IA, Prasad V et al. The tumor suppressor protein FHIT. A novel interaction
with tubulin. J Biol Chem 1999; 274:24378-24382.
40. Shi Y, Zou M, Farid NR et al. Association of FHIT (fragile histidine triad), a candidate
tumour suppressor gene, with the ubiquitin-conjugating enzyme hUBC9. Biochem J 2000;
352:443-448.
CHAPTER 9
Cell Cycle Checkpoints and Cancer, edited by Mikhail V. Blagosklonny. ©2001 Eurekah.com.
Hypoxia and Cell Cycle
Rachel A. Freiberg, Susannah L. Green and Amato J. Giaccia
Introduction
Tumor initiation is dependent on several key changes in the requirements for cell growth.
Three of the most important features that distinguish transformed cells from
untransformed cells are the loss of senescence, anchorage independent growth, and loss
of contact inhibition. Cells that adopt a transformed phenotype based on these criteria typically
exhibit rapid growth as they have escaped regulation from both internal and external regulatory
signals. This uncontrolled cellular growth leads to an accumulation of cells, which are initially
genetically identical, but continued growth results in additional mutations which can lead to
the outgrowth of transformed cell variants that possess a survival advantage in the context of
the tumor microenvironment. When the tumor is less than 150 mM or approximately ten cells
in diameter, the supply of oxygen and nutrients through passive diffusion is sufficient to support
growth and metabolism of transformed cells. However, as tumor cells continue dividing
unchecked, they exceed their ability to obtain sufficient nutrients and oxygen by diffusion
alone from pre-existing blood vessels. As a result, regions of the tumor become hypoxic and
start secreting mitogens such as VEGF (vascular endothelial cell growth factor) which can
stimulate the migration of microvascular endothelial cells into the tumor region to form new
microvessels. However, these new microvessels are often disorganized and malformed, resulting
in inefficient oxygen and nutrient delivery. Ultimately, this process results in tumor containing
areas that are either poorly perfused or subjected to cycling hypoxia through the opening and
closing of blood vessels. Angiogenesis has now been recognized as a critical modulator of tumor
cell expansion and angiogenic activity has also been implicated in the development of
metastases.1,2 Transformed cells unable to tolerate the nutrient and oxygen poor microenvironment
become necrotic or apoptotic, and the cells that survive are selected to do so by their
ability to resist apoptosis and reduce their oxygen requirements by cycling slowly or switching
to glycolysis. In this way, hypoxia has been suggested to act as a selective pressure for the
expansion of variants resistant to its growth restrictive conditions (Fig. 1).
A variety of publications have supported the hypothesis that human tumors possess hypoxic
regions.3-6 In addition to the increased likelihood of metastases, hypoxic tumors are associated
with poor therapeutic response and poor prognosis for a patient regardless of the therapeutic
modality.4,6 For example, radiation therapy depends in part on the presence of molecular oxygen
to mediate DNA damage.7,8 Therefore the damage induced by radiation therapy in oxygendeficient
regions of a tumor is substantially reduced and can lead to a three-fold decrease in
killing in response to radiation treatment. In contrast to radiation therapy, chemotherapeutic
drugs are delivered systemically through the circulatory system and act best on rapidly dividing
cells and those which are in close proximity to blood vessels. Since hypoxic tumor cells are
poorly perfused and growth retarded, current chemotherapy protocols have exhibited limited
therapeutic efficacy against hypoxic tumor cells.9
144 Cell Cycle Checkpoints and Cancer
Finally, cells located in hypoxic regions of tumors that are growth arrested possess varying
abilities to re-enter the cell cycle upon reoxygenation, suggesting that growth arrest by hypoxia
is reversible under certain conditions. As hypoxia plays important roles in both tumor response
to therapy and malignant progression, it is essential to understand how hypoxia and reoxygenation
modify cell cycle function and to identify the molecular mechanisms involved in this process.
Although knowledge of the mechanisms by which the sensors of oxygen deprivation act to signal
to cell-cycle checkpoint pathways is still incomplete, more extensive information exists on the
downstream cell-cycle regulatory proteins that are modulated by a low oxygen environment.
Cell Cycle and Check Points
Untransformed cells progress through G1, S, G2 and M-phases of the cell cycle in a highly
regulated manner. Starting in G1, the cell undergoes a phase of growth in which it accumulates
the necessary nucleotides and proteins needed for replication of DNA in S-phase. Upon the
completion of DNA synthesis, cells enter the G2-phase during which the cell continues growing
and generating additional cellular components needed for cell division in M-phase or mitosis.
During mitosis chromosomes condense, the nuclear membrane breaks down, chromosomes pair
and align along the midline of the cell, and sister chromatids separate through the shortening
action of the tubulin filaments attached to both the kinetochores of the chromosomes and the
centrioles. Other cellular components such as the endoplasmic reticulum, Golgi apparatus and
mitochondria divide somewhat randomly between the two daughter cells. This stage completes
the cell cycle.
Cyclins and cyclin dependent kinases (CDKs) are the major effector proteins that are
responsible for the progression through the stages of G1, S, G2 and M phases. (Fig. 2) Cyclins,
as their name implies, are present at predictable and specific times in the cell cycle. CDKs are
enzymes which are inactive as monomers, but become active subsequent to associating with a
cyclin, being transported into the nucleus, and being phosphorylated by the CDK-activating
kinase (CAK).10 When the complex has carried out its function, it is transported from the
nucleus to the cytoplasm and the cyclin is degraded via ubiquitin-mediated proteolysis. Other
Fig. 1. Hypoxia as a selective pressure. Normal cells aquire mutations which allow some cells to escape cell
cycle contol. Continued hyperplastic growth leads to accumulation of cells, some of which are too distant
from blood vessels to obtain oxygen and nutrients through passive diffusion. Cells removed from proper
growth conditions may either arrest and undergo apoptosis or continue growing at a reduced rate despite
the lack of oxygen and other growth factors (reprinted with permission from Giaccia A. The influence of
tumor hypoxia on malignant progression. In: Vaupel P, Kelleher D, eds. page 117).
Hypoxia and Cell Cycle 145
important cell cycle regulators are the p53 and pRB tumor suppressor genes, which act through
the transcriptional stimulation or inhibition of positive and negative growth regulatory genes.11
During the gap phases G1 and G2 preceding S and M phases respectively, the cell prepares
itself by synthesizing many of the proteins needed to initiate and complete DNA synthesis and
mitosis. Progression from one phase to the next is often dependent on the satisfaction of certain
criteria; if the build-up or degradation of essential cell-cycle proteins does not proceed in a
tightly regulated manner, a cell with all of its checkpoint machinery intact will remove itself
from the cell cycle. Such cell cycle “checkpoints” are thought to have evolved to ensure proper
replication and transmission of genetic material. For example, checkpoint pathways are triggered
by nutrient deprivation, incomplete DNA replication, or improper chromosome segregation.
Elegant studies from yeast to mammalian cells have demonstrated that “checkpoints” in
the cell cycle occur in all phases, and typically act via the temporary inactivation of CDKs.
The role of checkpoint genes has been demonstrated fairly well in response to DNA damage.
Two critical regulators of checkpoint response are the ATM and ATR proteins. ATM is
triggered by DNA damage and phosphorylates p53 on serine 15. In contrast, ATR seems to be
activated by perturbations in DNA replication, but can also phosphoylate many of the same
substrates as ATM.12-14 After cells are exposed to DNA damaging agents, p53 can either initiate
cell cycle arrest through the transcriptional induction of the p21 cyclin kinase inhibitor or
induce apoptosis.15 In a low oxygen environment, checkpoint control is thought to play a
pivotal role in ensuring that cell cycle progression does not occur under growth restrictive
conditions until physiologic homeostasis (e.g., reoxygenation) is returned. Hypoxia, though a
nongenotoxic stress, can lead to p53 phosphorylation on serine 15 and protein accumulation.
This response is intact in cells deficient in ATM, suggesting that hypoxia does not utilize the
same signaling pathway as DNA damage (E. Hammond, M. Kastan, and A. Giaccia, unpublished
observations). Studies are currently ongoing to determine whether ATR plays a role in
p53 regulation under hypoxic conditions. Interestingly, in cells exposed to hypoxic conditions,
p53 is not necessary for, nor does it contribute to, cell cycle arrest.16
Fig. 2.Cyclin Expression. Cyclical expression of D, E, A, and B type cyclins allows precise control of the
initiation and duration of different phases of the cell cycle. Cyclin D-family members are key to the
progression through the G1 phase of the cell cycle by facilitating the dephosphorylation of pRb. Continued
depohsphorylation of pRb by Cyclin E is required for entry into S-phase of the cell cycle. Cyclin A and to
a greater extent Cyclin B are both required for entry into mitosis.
146 Cell Cycle Checkpoints and Cancer
Although DNA damage can induce cell-cycle arrest in G1, S and G2 phases of the cell
cycle through the activation of checkpoint pathways, it is unclear what the stimulus is for cell
cycle arrest observed in response to hypoxia. While reports have proposed both biochemical and
genetic mechanisms to explain cell-cycle arrest under hypoxic conditions, we still do not know
how and if these mechanisms can be generalized or if they are specific for individual cell lines.
Hypoxia-Induced Arrest
In vivo, at the tumor level, there is conflicting evidence as to whether or not hypoxic
regions of tumors exhibit a decrease in cell proliferation. In some cases, investigators have
found that hypoxic tumor regions have fewer actively cycling cells17 where others have found
that hypoxic tumors proliferate at an increased rate.18,19 Studies in which proliferation and
hypoxia were simultaneously assessed in tumor sections showed that proliferating cells were
proximal to blood vessels and hypoxic cells were found in regions that were distal to blood
vessels.17,20,21 Analysis of single cells from disaggregated tumors indicated that oxic regions
exhibit substantially greater proliferation than hypoxic regions as assessed by in vivo incorporation
of the thymidine nucleotide analog bromodeoxyuridine (BrdU).
Studies have also assessed the effect of hypoxic conditions on cellular proliferation rates in
vitro by analyzing cellular incorporation of BrdU or 3H-thymidine. These studies indicate that
untransformed or transformed cells cultured under hypoxic conditions contain a greatly
reduced number of BrdU positive S-phase cells or reduced amounts of 3H-thymidine incorporation
as compared to cells grown under oxic conditions.22-24 Investigators utilizing both types
of nucleotide incorporation have revealed that the stringency of hypoxia is an important determinant
of the kinetics of cell cycle arrest. For example, mild hypoxia (i.e., 2% oxygen) has been
found to enhance cell growth.25 Additionally, there is little difference in cell cycle profile for
some cells incubated in normoxic (21.0% oxygen) versus moderate hypoxic conditions (1.0%
or 0.1% oxygen).26 After 24 or 48 hours under moderate hypoxic conditions, immortalized
mouse embryo fibroblasts (MEFs) exhibit no significant cell cycle inhibition as measured by
BrdU or 3H-thymidine incorporation. However, stringent hypoxia (oxygen tensions below
0.01% O2) elicits a robust S-phase arrest response; under these conditions, DNA synthesis
ceases within 5 hours and recovers within a few hours after reoxygenation. While cell cycle
arrest induced by hypoxia is reversible, the extent of reversibility is reduced with longer or more
stringent hypoxic treatments and is variable among different cell types.27 In fact, most cells are
able to resume DNA synthesis within 10 minutes to 3 hours after reoxygenation.28
Until recently, little has been published on the molecular mechanisms that govern the
re-entry of hypoxic cells into the cell cycle. One report found that recovery of DNA synthesis
after removal from hypoxic conditions is blocked by cycloheximide addition in Ehrlich ascites
cells,29,30 suggesting that synthesis of a rapidly degraded protein is necessary to resume DNA
synthesis after arrest. In a separate study, re-entry into the cell cycle was associated with the
reappearance of hyperphosphorylated pRb in T-47D breast cancer cells.31 However, the
investigators could only speculate whether the relationship between pRb phosphorylation and
hypoxia-induced cell cycle arrest was causal or correlative.
Hypoxia-induced cell cycle arrest raises several important questions. What role if any does
arrest play in protecting cells from the potentially deleterious effects of a low oxygen
environment? Does hypoxia-induced arrest contribute to or protect the cell from genomic
instability? Do hypoxia-arrested cells accumulate even more damage as they escape apoptosis
and necrosis or sustain less damage because of the arrest? Reports in the literature have indicated
that hypoxia and reoxygenation induce gene amplification. While this has been demonstrated
in experimental systems, it is unknown whether the presence of hypoxia in human tumors is in
any way related to increased genomic instability or gene amplification.32,33 The implication of
these studies is that S-phase arrest under hypoxic conditions may be somehow involved in gene
amplification as well as in genomic instability. Additionally, hypoxia itself may not directly
induce DNA damage, but may permit the accumulation or manifestation of DNA damage or
Hypoxia and Cell Cycle 147
mutagenic events. Results recently reported by Denko et al16 indicate that hypoxia induces
cell-cycle arrest in human tumor cells through a different mechanism than DNA damage, and
that loss of the p21 cyclin-CDK inhibitor does not lead to increased genomic instability under
hypoxic conditions. The observation that loss of p21, which renders cells more sensitive to
ionizing radiation induced genomic instability, does not increase genomic instability under
hypoxic conditions is consistent with the finding in the same report that hypoxia induced G1/
S-phase arrest does not rely on the p53 tumor suppressor gene. Additional studies will be
needed to more rigorously address the role of hypoxia-induced arrest in cellular protection
against a low oxygen environment.
Mechanisms Underlying Cell Cycle Arrest by Hypoxia
Initial studies investigating the inhibition of cell proliferation seen in response to hypoxia
focused on a biochemical mechanism for cell cycle arrest. Numerous biochemical studies were
performed on Ehrlich ascites tumor cells, which are capable of proliferating in the peritoneal
cavity of mice, an environment which is largely devoid of oxygen. When exposed to hypoxic
conditions, Ehrlich ascites tumor cells that are in S-phase stop synthesizing DNA and cells in
G2 and M phase proceed into G1-phase and undergo a G1 arrest.34-36 It has been proposed that
these cells arrest in response to depletion of deoxynucleotides needed for the synthesis of DNA
during S-phase. There are two oxygen-dependent enzymes, ribonucleotide reductase and
dihydroorotate dehydrogenase, both of which are necessary for the completion of the
deoxynucleotide biosynthetic pathway. The active site of ribonucleotide reductase, which converts
ribonucleotides to deoxyribonucleotides, contains a tyrosyl radical that requires molecular oxygen
for its regeneration. Dihydroorotate dehydrogenase is responsible for the conversion of
dihydroorotate to orotate, an early step in the de novo synthesis of pyrimidines. The enzyme’s
activity is dependent on the mitochondrial respiratory chain and is thus linked to oxygen
availability. Oxygen deprivation, therefore, could lead to S-phase arrest by directly preventing the
synthesis of the building blocks needed for DNA replication. Further studies indicated that reduced
DNA synthesis correlated with the suppression of replicon initiation and that chain elongation
and maturation were both unaffected by hypoxia.30, 37 Arrest found under hypoxic conditions
could sometimes be observed at oxygen tensions as high as 2% and consistently at oxygen tensions
below 0.2%.30 However, it was found that the addition of exogenous deoxyribonucleosides (which
can cross the cell membrane and be phosphorylated via an oxygen-independent pathway) allowed
some cells to enter S-phase, but without normal progression of the cell cycle, strengthening the
idea that hypoxia induced arrest may simply not be due to lack of nucleotides.34,37-39 Taken
together, these studies suggested that the arrest mechanism under hypoxic conditions was not
totally biochemical in origin and may possess a genetic or epigenetic component.
In most cells that have been investigated, hypoxia causes a decrease in DNA synthesis,
which may be accompanied by an accumulation of cells in the G1-phase of the cell cycle.
However, Chinese hamster fibroblasts, V79-379A, arrest in all phases of the cell cycle and thus
its cell cycle profile under hypoxic conditions appears unchanged even when DNA synthesis
has ceased.22 CV-1P monkey kidney cells also arrest in all phases of the cell cycle.40 More
common is the cell-cycle profile displayed by many transformed and untransformed human
and rodent cells in which the S-phase arrest is accompanied by an accumulation of cells in
G1.27,41 Since G1-and S-phase arrest are more commonly found in cells exposed to hypoxic
conditions, current research has focused on the cell cycle components regulating the progression
of cells through the G1 and S-phases.
As discussed above, cyclins are responsible for progression through the cell cycle. In particular,
the G1-phase of the cell cycle is controlled by D-type cyclins. Expression of cyclin-D
family members D1, D2 and D3, can depend on the differentiation state of the cell as well as its
histological origin.42,43 Overexpression of Dcyclins causes the cell to progress through G1 at an
accelerated rate, and conversely, removal of these cyclins causes a G1 arrest.44,45 During early
and mid G1-phase, removal of growth stimulus can cause arrest prior to the initiation of DNA
148 Cell Cycle Checkpoints and Cancer
synthesis. However, in late G1 there is a point at which the cell is committed to progressing to
S-phase, past which the removal of growth stimulus or the binding of cyclins will have no
effect. This is known as the restriction point. A key component in the regulation of this restriction
point is the retinoblastoma protein (pRB). The sequential phosphorylation of pRB by different
cyclin-CDK complexes is essential for progression of the cell from G1 into S-phase. During the G1
phase, D cyclins associate with the cyclin dependent kinases CDK4 and CDK6 to form functional
units that are capable of phosphorylating pRB and do so in early G1. Further phosphorylation
of pRb is mediated by cyclin E/CDK2, which begins functioning during mid G1, and cyclin A/
CDK2, which acts at the G1/S-phase boundary and maintains pRb phosphorylation through S
and G2-phases.46 Krtolica et al have observed decreased CDK4 and CDK2 activity in CV-1P
monkey kidney cells treated with ~1% O2, and this kinase inhibition is associated with an
increase in p27Kip1 and a decrease in CDK4, cyclin D and cyclin E protein levels. They propose
that this decreased kinase activity, coupled with increased PP-1 phosphatase activity, leads to
dephosphorylation of pRb and inhibition of cell cycle progression.47 Similar results to those found
in CV-1P cells were obtained in ovarian carcinoma cells, suggesting a possible common pathway.48
Unphosphorylated pRb binds to and inhibits the activity of the E2F family of proteins,
which are transcriptional activators of genes required for entry into S-phase such as PCNA,
Cdc2, dihydrofolate reductase, cyclin E, thymidine kinase, DNA polymerase and histone H2A,
among others. Upon hyperphosphorylation of pRB, E2F is released from pRB and can function
as a transcriptional activator.49 Additional regulators of pRb include the CIP/KIP family
of cyclin-CDK inhibitors, including p21, p27 and p57, which block phosphorylation of pRB
through their interactions with CDK4, CDK6 and CDK2.50 Another family of CDK4/CDK6
specific inhibitors known as the INK4 family is comprised of p15, p16, p18 and p19.50 (Fig. 3)
In a recent study, immortalized mouse embryo fibroblasts derived from mice deficient in either
pRb, p130, p21CIP1, p27Kip1 or both p21CIP1 and p27Kip1, along with genetically matched wild
type cell lines, were exposed to an oxygen tension of 0.01%.26 Following BrdU incorporation
under hypoxia, each of the cell types was analyzed by two-dimensional FACS analysis. Within
a few hours, DNA synthesis ceased. These results indicate that pRB, p130, p21 and p27 are not
needed for cell cycle arrest induced by hypoxia and do not contribute to this arrest.26 Interestingly,
when the cells were reoxygenated after hypoxic treatment, the p21-/- cells re-entered the
cell cycle more rapidly than wild type cells, whereas there was no acceleration of the S-phase reentry
of the p27-/- cells as determined by both FACS analysis and 3H incorporation assays.
The double knockout p21-/-p27-/- cells resumed S-phase even more quickly than the p21-/-
alone did. Thus, while p21 and p27 are not necessary for arrest under hypoxia, they do play a
role in inhibiting cell cycle re-entry upon reoxygenation. In contrast to the studies by Green et
al studies by Gardner et al51 have resulted in the conclusion that p27 mediates hypoxia-induced
arrest. According to their data, cells deficient in p27 were observed to resist hypoxiainduced
arrest upon exposure to 0.1-0.5% oxygen for 32 hours. At present, the data published
by Green and Gardner seem to be in conflict and can only be explained by the different experimental
procedures and conditions used in each study. However, both groups agree that the
kinase activity of CDK2 in cells was found to decrease dramatically under hypoxic conditions
without a corresponding decrease in protein levels of CDK2 or a decrease in cyclin A or E
abundance. Cyclin A and E immunoprecipitates from hypoxic cells exhibited decreased activity,
but no change in CDK2 association.
There are several possible explanations for a decrease in cyclin Activity without a corresponding
decrease in protein levels. (Fig. 4) One possibility is that if cyclin Activating kinase
(CAK) fails to phosphorylate threonine 160 of CDK2, the cyclin/CDK complex will be inactive.
This modification of threonine 160 results in increased mobility on SDS-PAGE, which is
unaffected in both WT and p21-/-p27-/- cells exposed to hypoxic conditions. Thus, reduced
T160 phosphorylation could not explain the decrease in kinase activity observed under
hypoxia-treated cells. A second possibility is the binding of an inhibitory protein, such as the
CIP/KIP or INK4 family members mentioned previously as well as the pRb family members
Hypoxia and Cell Cycle 149
p107 and p130 to the cyclin/CDK complex. The experiments described above eliminated p21,
p27 and p130 as candidate effectors of arrest, as cells lacking these proteins arrested normally.
Though the INK4 proteins do not interact with CDK2, they have been proposed to displace
CIP/KIP proteins from CDK4/6 and thereby indirectly inactivate CDK2. Hypoxia, however,
had little effect on protein levels of p16,INK4a p15,INK4b p18,INK4c p19,INK4d p130 and p107 in
MEFs. A third possibility is that CDK2 can be inactivated by inhibitory phosphorylation on
threonine 14 and tyrosine 15. To address this latter possibility CDC25B, a protein phosphatase
known to dephosphorylate these residues in vitro was used to treat extracts from oxic and
hypoxic cells. Treatment of CDK2 immunoprecipitates from control and hypoxic cells with GST
tagged and purified CDC25B produced a significant activation of CDK2 complexes, particularly
in the hypoxia treated samples. CDK2 immunoprecipitates from confluent and serum starved
cells, by contrast, remained inactive upon CDC25B treatment. These results suggest that inhibitory
Fig. 3. Possible hypoxic effects on the progressive phosphorylation of pRb. INK family members negatively
regulate Cyclin D and its partners CDK4 and CDK6, inhibiting their kinase activity on the pRb/E2F
complex. Hypoxia does not induce the CyclinD/CDK2/4 inhibitors of the CIP/KIP family. The Cyclin E/
CDK2 complex has been found to be inhibited by hypoxia.
150 Cell Cycle Checkpoints and Cancer
phosphorylation plays an important role in CDK2 inactivation under hypoxic conditions. Previous
studies have indicated that Tyr15 is the more important of the two phosphorylation sites.52 To
directly analyze this residue, CDK2 was immunoprecipitated and subjected to SDS-PAGE and
probing with phosphospecific antibody recognizing Tyr15-phosphorylated cdc2 and CDK2. Results
revealed that phosphorylation was not increased by hypoxia, but was in fact mildly decreased.
Given the evidence that CDC25B can restore activity, this would imply that phosphorylation
of Thr14 could contribute to CDK2 inhibition by hypoxia. These results suggest that
the phosphorylation of Tyr15 alone cannot account for the hypoxia-induced inhibition of CDK2,
and further investigation to distinguish the difference between the two phosphorylation sites is
needed.
Hypoxia Mimetics and Cell Cycle Arrest
The arrest seen in response to hypoxic stress seems independent of other pathways, such as
those used by the cell in responses to ionizing radiation, ultraviolet light and aberrant growth
signaling. As mentioned previously, early studies focused on a biochemical mechanism for cell
cycle arrest, such as depletion of nucleotides or nucleotide precursors. Much of the current
research does not support this hypothesis. A comparison between cell cycle arrest seen in response
to hypoxia and that induced by hypoxia mimetic drugs addresses how cells sense
hypoxia and to what extent the hypoxic response overlaps with other stress responses. While
significant progress has been made towards answering this question with regards to other
elements of the hypoxic response (i.e., gene induction and apoptosis), in the context of cell
cycle arrest they remain largely unanswered.
The chemical hydroxyurea is a radical scavenger and is thought to inactivate ribonucleotide
reductase in a manner similar to hypoxia. Mouse fibroblasts exposed to this chemical displayed
a rapid cell cycle arrest in which there was negligible DNA synthesis 6 hours after treatment
started. The S-phase accumulation of cells persisted for 12-24 hours and then
declined to levels similar to the control sample. Kinase assays revealed that hydroxyurea did not
affect CDK2 activity, as had been observed previously.53 The distinct differences between the
Fig 4.Mechanisms of CDK2 regulation: modified from Morgan.57
Hypoxia and Cell Cycle 151
cellcycle arrest responses to hypoxia and hydroxyurea indicate that the hypoxic cell cycle arrest
cannot be due to ribonucleotide reductase inactivation alone.
Desferrioxamine (DFO) is an iron chelator that has been classified as chemical “hypoxia.”
Based on the observation that iron chelators or divalent metals can induce some aspects of the
hypoxic response, in particular activation of the transcription factor HIF-1,54 the hypoxic response
is thought to rely partially on oxygen sensing by a heme-containing protein. DFO and
cobalt chloride are commonly used in lieu of hypoxia due to the practical challenges posed by
achieving hypoxia in the lab. However, these chemicals do not necessarily cause a response
identical to that induced by true hypoxia. Cells exposed to DFO experience cell cycle arrest in
mid S-phase in less than six hours, similar to that induced by hypoxia. However, cell division
appears to continue as the G1 population increases significantly after 12 and 24 hours. Additionally,
DFO causes a decrease in cyclin-E and cyclin-A associated kinase activities, without
changing the protein levels of these two cyclins.55 These data indicate a possible common
pathway for hypoxia-and-DFO induced cell cycle arrest. Other pharmacologic and metabolic
agents have also been examined for cell cycle inhibitors that include an inhibitor of uridine
synthesis (N-phosphonacetyl-L-aspartate-PALA)), glucose deprivation, and an electron transport
inhibitor (antimycin). Each of these treatments or agents inhibit cell cycle progression, but none
were significantly similar to the hypoxic induced arrest. (Green et al unpublished observations)
Hypoxia-Induced Inhibition of CDK2 Activity and Resistance
to Chemotherapy
The proliferation status of tumor cells has little effect on killing by radiotherapy. In contrast,
many commonly used chemotherapeutic agents target rapidly proliferating cells. A recent study by
Davis et al56 has demonstrated that inhibition of CDK2 activity was sufficient to reduce the
severity of chemotherapy-induced alopecia in 33 to 50% of the animals treated. This study makes
a direct connection between the rate of proliferation of hair follicle cells and sensitivity to chemotherapy.
Thus, if proliferation is a critical determinant for chemotherapeutic killing, inhibition of
cell proliferation by hypoxia can explain the decreased efficacy of chemotherapeutic agents in
killing hypoxic tumor cells. Furthermore, as multiple studies have demonstrated that hypoxia
inhibits CDK2 activity, then restoration of CDK2 activity in hypoxic cells should increase the
efficacy of chemotherapy. This hypothesis was tested by transfecting a mutant version of CDK2
(CDK2-A14F15) resistant to phosphorylation into oxic and hypoxic cells. The transiently
transfected HA-tagged CDK2-A14F15 mutant was analyzed for kinase activity under various
conditions. Anti-HA immunoprecipitates from transfected cells displayed substantial Histone
H1 kinase activity whereas extracts from untransfected cells exhibited no HA-associated kinase
activity. However, CDK2-AF transfected cells exposed to hypoxia displayed a progressive
inhibition of kinase activity, reaching 60% inhibition by 12 hours. While this inhibition was
slower and less severe than that of endogenous CDK2, the data indicates that there is still some
other phosphorylation-independent means of CDK2 inhibition activated by hypoxia. If this
inhibition is stoichiometric (i.e., inhibitor binding), then one should be able to compete out
the inhibition. Therefore CDK2-AF was co-transfected with Cyclin E to generate more active
complexes and thereby drive the cells into cycle. As expected, cells transfected with both molecules
had a much higher kinase activity than CDK2 AF alone. But the activity was still inhibited by
hypoxia, suggesting that the reduction in activity is not a result of inhibitor binding. Therefore,
introduction of CDK2 or CDK2-A14F15 will not be sufficient to stimulate cells to proliferate
under hypoxic conditions, and whether such stimulation would render hypoxic cells
chemosensitive is still unknown.
These results underscore the importance of mechanistically understanding how hypoxia
inhibits CDK2 activity in order to increase the sensitivity of hypoxic cells to killing by chemotherapy.
The goal of future studies should be to determine how CDK2 is being inhibited under
hypoxic conditions, and whether restoration of CDK2 activity will stimulate hypoxic cells to
proliferate and become more chemosensitive.
152 Cell Cycle Checkpoints and Cancer
Acknowledgments
This work was supported by NIH grants PO1 CA67166 and CA64489 to A.J.G. R.A.F.
was supported by PHS Grant Number CA09302, awarded by the National Cancer Institute,
DHHS. S.L.G. was supported in part by a Lucille P. Markey fellowship to Molecular Mechanisms
of Disease.
References
1. Weidner N, Semple JP, Welch WR et al. Tumor angiogenesis and metastasis--correlation in invasive
breast carcinoma. N Engl J Med 1991; 324:1-8.
2. Weidner N, Carroll PR, Flax J et al. Tumor angiogenesis correlates with metastasis in invasive
prostate carcinoma. Am J Pathol 1993; 143:401-409.
3. Nordsmark M, Overgaard M, Overgaard J. Pretreatment oxygenation predicts radiation response
in advanced squamous cell carcinoma of the head and neck. Radiother Oncol 1996; 41:31-39.
4. Brizel DM, Scully SP, Harrelson JM et al. Tumor oxygenation predicts for the likelihood of distant
metastases in human soft tissue sarcoma. Cancer Res 1996; 56:941-943.
5. Brizel DM, Sibley GS, Prosnitz LR et al. Tumor hypoxia adversely affects the prognosis of carcinoma
of the head and neck. Int J Radiat Oncol Biol Phys 1997; 38:285-289.
6. Hockel M, Schlenger K, Aral B et al. Association between tumor hypoxia and malignant progression
in advanced cancer of the uterine cervix. Cancer Res 1996; 56:4509-4515.
7. Prise KM, Gillies NE, Michael BD. Evidence for a hypoxic fixation reaction leading to the induction
of ssb and dsb in irradiated DNA. Int J Radiat Biol 1998; 74:53-59.
8. Siemann DW, Keng PC. Characterization of radiation resistant hypoxic cell subpopulations in
KHT sarcomas. (II). Cell sorting. Br J Cancer 1988; 58:296-300.
9. Tomida A, Tsuruo T. Drug resistance mediated by cellular stress response to the microenvironment
of solid tumors. Anticancer Drug Des 1999; 14:169-177.
10. Sherr CJ, Roberts JM. CDK inhibitors: Positive and negative regulators of G1-phase progression.
Genes Dev 1999; 13:1501-1512.
11. Vogelstein B, Lane D, Levine AJ. Surfing the p53 network [news] [In Process Citation]. Nature
2000; 408:307-310.
12. Tibbetts RS, Brumbaugh KM, Williams JM et al. A role for ATR in the DNA damage-induced
phosphorylation of p53. Genes Dev 1999; 13:152-157.
13. Canman CE, Lim DS, Cimprich KA et al. Activation of the ATM kinase by ionizing radiation and
phosphorylation of p53. Science 1998; 281:1677-1679.
14. Banin S, Moyal L, Shieh S et al. Enhanced phosphorylation of p53 by ATM in response to DNA
damage. Science 1998; 281:1674-1677.
15. Zhou BB, Elledge SJ. The DNA damage response: putting checkpoints in perspective [In Process
Citation]. Nature 2000; 408:433-439.
16 .Denko NC, Green SL, Edwards D et al. p53 checkpoint-defective cells are sensitive to X rays, but
not hypoxia. Exp Cell Res 2000; 258:82-91.
17. Kennedy AS, Raleigh JA, Perez GM et al. Proliferation and hypoxia in human squamous cell
carcinoma of the cervix: First report of combined immunohistochemical assays. Int J Radiat Oncol
Biol Phys 1997; 37:897-905.
18. Nordsmark M, Hoyer M, Keller J et al. The relationship between tumor oxygenation and cell
proliferation in human soft tissue sarcomas. Int J Radiat Oncol Biol Phys 1996; 35:701-708.
19. Hansgen G, Hintner I, Krause V et al. [Intratumor pO2, S-phase fraction and p53 status in cervix
carcinomas]. Strahlenther Onkol 1997; 173:385-387.
20. Raleigh JA, Zeman EM, Calkins DP et al. Distribution of hypoxia and proliferation associated
markers in spontaneous canine tumors. Acta Oncol 1995; 34:345-349.
21. Zeman EM, Calkins DP, Cline JM et al. The relationship between proliferative and oxygenation
status in spontaneous canine tumors. Int J Radiat Oncol Biol Phys 1993; 27:891-898.
22. Shrieve DC, Begg AC. Cell cycle kinetics of aerated, hypoxic and re-aerated cells in vitro using
flow cytometric determination of cellular DNA and incorporated bromodeoxyuridine. Cell Tissue
Kinet 1985; 18:641-651.
23. Rice GC, Spiro IJ, Ling CC. Detection of S-phase overreplication following chronic hypoxia using
a monoclonal anti-BrdUrd. Int J Radiat Oncol Biol Phys 1985; 11:1817-1822.
24 .Webster L, Hodgkiss RJ, Wilson GD. Simultaneous triple staining for hypoxia, proliferation, and
DNA content in murine tumours. Cytometry 1995; 21:344-351.
25. Balin AK, Fisher AJCarter DM. Oxygen modulates growth of human cells at physiologic
partialpressures. J Exp Med 1984; 160:152-166.
Hypoxia and Cell Cycle 153
26. Green SL, Freiberg RA, Giaccia AJ. p21CIP1 and p27 Kip1 regulate cell cycle reentry after hypoxic
stress but are not necessary for hypoxia-induced arrest. Mol Cell Biol 2001; 21:1196-1206.
27. Amellem O, Sandvik JA, Stokke T et al. The retinoblastoma protein-associated cell cycle arrest in
S-phase under moderate hypoxia is disrupted in cells expressing HPV18 E7 oncoprotein. Br J
Cancer 1998; 77:862-872.
28. Wilson RE, Keng PC, Sutherland RM. Changes in growth characteristics and macromolecular synthesis
on recovery from severe hypoxia. Br J Cancer 1990; 61:14-21.
29. Gekeler V, Epple J, Kleymann G et al. Selective and synchronous activation of early-S-phase replicons
of Ehrlich ascites cells. Mol Cell Biol 1993; 13:5020-5033.
30. Riedinger HJ, Gekeler V, Probst H. Reversible shutdown of replicon initiation by transient
hypoxia in Ehrlich ascites cells. Dependence of initiation on short-lived protein. Eur J Biochem 1992;
210:389-398.
31. Amellem O, Stokke T, Sandvik JA et al. The retinoblastoma gene product is reversibly dephosphorylated
and bound in the nucleus in S-and G2-phases during hypoxic stress. Exp Cell Res 1996;
227:106-115.
32. Hill RP. Tumor progression: Potential role of unstable genomic changes. Cancer Metastasis Rev
1990; 9:137-147.
33. Russo CA, Weber TK, Volpe CM et al. An anoxia inducible endonuclease and enhanced DNA
breakage as contributors to genomic instability in cancer. Cancer Res 1995; 55:1122-1128.
34. Loffler M. Towards a further understanding of the growth-inhibiting action of oxygen deficiency.
Evaluation of the effect of antimycin on proliferating Ehrlich ascites tumour cells. Exp Cell Res
1985; 157:195-206.
35. Loffler M. Characterization of the deoxynucleoside-dependent reversal of hypoxia- induced inhibition
of cell cycle progression in Ehrlich ascites tumor cells. Eur J Cell Biol 1985; 39:198-204.
36. Probst H, Schiffer H, Gekeler V et al. Oxygen dependent regulation of DNA synthesis and growth
of Ehrlich ascites tumor cells in vitro and in vivo. Cancer Res 1988; 48:2053-2060.
37. Probst H, Schiffer H, Gekeler V et al. Oxygen dependent regulation of mammalian ribonucleotide
reductase in vivo and possible significance for replicon initiation. Biochem Biophys Res Commun
1989; 163:334-340.
38. Loffler M. A cytokinetic approach to determine the range of O2-dependence of pyrimidine
(deoxy)nucleotide biosynthesis relevant for cell proliferation. Cell Prolif 1992; 25:169-179.
39. Loffler M. Restimulation of cell cycle progression by hypoxic tumour cells with deoxynucleosides
requires ppm oxygen tension. Exp Cell Res 1987; 169:255-261.
40. Ludlow JW, Howell RL, Smith HC. Hypoxic stress induces reversible hypophosphorylation of
pRB and reduction in cyclin A abundance independent of cell cycle progression. Oncogene
1993; 8:331-339.
41. Graeber TG, Peterson JF, Tsai M et al. Hypoxia induces accumulation of p53 protein, but activation
of a G1- phase checkpoint by low-oxygen conditions is independent of p53 status. Mol Cell
Biol 1994; 14:6264-6277.
42. Doglioni C, Chiarelli C, Macri E et al. Cyclin D3 expression in normal, reactive and neoplastic
tissues. J Pathol 1998; 185:159-166.
43. Joyce NC, Meklir B, Joyce SJ et al. Cell cycle protein expression and proliferative status in human
corneal cells. Invest Ophthalmol Vis Sci 1996; 37:645-655.
44. Baldin V, Lukas J, Marcote MJ et al. Cyclin D1 is a nuclear protein required for cell cycle progression
in G1. Genes Dev 1993; 7:812-821.
45. Quelle DE, Ashmun RA, Shurtleff SA et al. Overexpression of mouse D-type cyclins accelerates
G1 phase in rodent fibroblasts. Genes Dev 1993; 7:1559-1571.
46. Mittnacht S. Control of pRB phosphorylation. Curr Opin Genet Dev 1998; 8:21-27.
47. Krtolica A, Krucher NA, Ludlow JW. Hypoxia-induced pRB hypophosphorylation results from
downregulation of CDK and upregulation of PP1 activities. Oncogene 1998; 17:2295-2304.
48. Krtolica A, Krucher NA, Ludlow JW. Molecular analysis of selected cell cycle regulatory proteins
during aerobic and hypoxic maintenance of human ovarian carcinoma cells. Br J Cancer 1999;
80:1875-1883.
49. Dyson N. The regulation of E2F by pRB-family proteins. Genes Dev 1998; 12:2245-2262.
50. Tamrakar S, Rubin E, Ludlow JW. Role of pRB dephosphorylation in cell cycle regulation. Front
Biosci 2000; 5:D121-137.
51. Gardner LB, Li Q, Park MS et al. Hypoxia inhibits G1/S transition through regulation of p27
expression. J Biol Chem 2000; 276:7919-7926.
52. Gu Y, Rosenblatt J, Morgan DO. Cell cycle regulation of CDK2 activity by phosphorylation of
Thr160 and Tyr15. Embo J 1992; 11:3995-4005.
154 Cell Cycle Checkpoints and Cancer
53. Matsumoto Y, Hayashi K, Nishida E. Cyclin-dependent kinase 2 (CDK2) is required for centrosome
duplication in mammalian cells. Curr Biol 1999; 9:429-432.
54. Wang GL, Semenza GL. Desferrioxamine induces erythropoietin gene expression and hypoxiainducible
factor 1 DNA-binding activity: Implications for models of hypoxia signal transduction.
Blood 1993; 82:3610-3615.
55. Kulp KS, Green SL, Vulliet PR. Iron deprivation inhibits cyclin-dependent kinase activity and
decreases cyclin D/CDK4 protein levels in asynchronous MDA-MB-453 human breast cancer cells.
Exp Cell Res 1996; 229:60-68.
56. Davis ST, Benson BG, Bramson HN et al. Prevention of chemotherapy-induced alopecia in rats by
CDK inhibitors [In Process Citation]. Science 2001; 291:134-137.
57. Morgan DO. Principles of CDK regulation. Nature 1995; 374:131-134.
CHAPTER 10
Cell Cycle Checkpoints and Cancer, edited by Mikhail V. Blagosklonny. ©2001 Eurekah.com.
G2 Checkpoint and Anticancer Therapy
Zoe A. Stewart and Jennifer A. Pietenpol
Abstract
Over the past two decades, the basic molecular events controlling eukaryotic G2 to Mphase
cell cycle transition have been deciphered. Studies in a variety of organisms
have identified an evolutionarily conserved system for controlling mitotic onset through
regulation of Cdc2 kinase activity. Recently, investigations have focused on how the signaling
pathways that mediate the G2 transition are regulated and modified after cellular stresses. In
response to DNA damage, eukaryotic cells activate biochemical pathways, called checkpoints,
to halt cell cycle progression and allow cellular damage to be repaired. Recent studies suggest
that the DNA damage-induced G2 checkpoint is comprised of an early activation stage as well
as a subsequent maintenance phase. In the absence of proper G2 checkpoint function, cells
proceed to mitosis with damaged DNA, resulting in either apoptosis or permanent alteration
of the genome that may contribute to tumorigenesis. The ability to manipulate G2 checkpoint
signaling also has important clinical implications, as abrogation of the G2 checkpoint in human
tumor cells can enhance cellular sensitivity to chemotherapeutic regimens that induce
DNA damage. This Chapter will focus on
(i) eukaryotic DNA damage-induced G2 checkpoint signaling pathways and
(ii) how knowledge of these signaling pathways may lead to more efficient use of current anticancer
therapies and the development of novel agents.
Introduction
The G2/M transition is regulated by the cyclin A/Cdc2 and cyclin B/Cdc2 kinase complexes.1 Like
other CDK molecules, Cdc2 function is tightly regulated by a complex system involving posttranslational
modifications and protein-protein interactions.2 Cdc2 activation requires association
with a cyclin partner, phosphorylation on Thr-161 by CAK ,3,4 and dephosphorylation on
Thr-14 and Tyr-15 by a Cdc25 phosphatase.5,6 Several substrate proteins of Cdc2 have been
identified, including the nuclear lamins A and B,7 histone H1,8 survivin,9 and microtubuleassociated
proteins such as MAP4,10 Eg5,11 and stathmin.12
As previously noted, Cdc2 can interact with both cyclin A and cyclin B family members to
regulate the G2/M transition. In mammalian cells, two cyclin A family members have been
identified, cyclin A1 and cyclin A2. Mammalian cyclin A1 expression is restricted to the testis,
brain, and hematopoietic cells.13,14 Mice deficient for cyclin A1 develop normally but males
are sterile due to defective spermatogenesis from impaired Cdc2 activation.15,16 Cyclin A1 is
frequently overexpressed in acute myelocytic leukemias, presumably through activation of the
cyclin A1 promoter by fusion proteins created through chromosomal translocation.13,17 Cyclin
A2 is ubiquitously expressed and participates in both the G1/S and G2/M transitions by interacting
with CDK2 and Cdc2.18,19 Homozygous deletion of cyclin A2 in mice results in
156 Cell Cycle Checkpoints and Cancer
embryonic lethality, demonstrating the importance of this cyclin for normal development.20 However,
further studies are necessary to define the role of cyclin A2 kinase activity in G2/M transition.
Similar to the cyclin A family, mammalian cells contain two cyclin B family members,
cyclin B121,22 and cyclin B2;23 however, chickens, frogs, flies, and nematode worms possess a
third, more distant relative, cyclin B3.24,25 Both mammalian cyclin B family members are
frequently co-expressed in proliferating cells, although tissue-specific temporal expression of
cyclin B1 and cyclin B2 has been observed in murine germ cells.23 Further, the two cyclin B
family members have distinct subcellular localizations, with cyclin B2 strictly associated with
intracellular membranes26 while cyclin B1 is found both on intracellular membranes and in the
cytoplasm.10,27 Cyclin B2-null mice develop normally and both males and females are fertile.28
In contrast, cyclin B1-deficient mice die in utero prior to embryonic day 10.28 In late G2 phase
of the cell cycle, cyclin B1 enters the nucleus, as nuclear localization of cyclin B1 is required
during the G2/M transition.27,29 Unlike cyclin B1, cyclin B2 does not relocate to the nucleus
during the G2/M transition.26 The nuclear localization of Xenopus cyclin B1 is mediated by
phosphorylation of residues in the cytoplasmic retention signal sequence of the protein.29 Several
studies indicate that cyclin B1 is continuously exported from the nucleus by interaction with
the nuclear exporter CRM1 during interphase, as the cytoplasmic retention signal of cyclin B1
contains a functional nuclear export signal.30-32 Since the role of cyclin B1 in the G2/M transition
has been the most extensively studied and the best defined of any eukaryotic Cdc2-associated
cyclin, the subsequent discussion of Cdc2 activity will focus on regulation of the cyclin B1/
Cdc2 complex.
During the G2-phase of the cell cycle, inactive cyclin B1/Cdc2 complexes accumulate in
mammalian cells due to inhibitory phosphorylation of Cdc2 on Thr-14 and Tyr-15 by the
Wee1 and Myt1 kinases (Fig. 1).33,34 Wee1 is a dual-specificity protein kinase that was initially
identified as a dose-dependent inhibitor of mitosis in S. pombe.35 When Cdc2 is complexed
with cyclin B1, recombinant Wee1 can phosphorylate Cdc2 on Tyr-15 and inhibit Cdc2 kinase
activity.33 Wee1 is hyperphosphorylated and degraded during mitosis, suggesting that
negative regulation of Wee1 is part of the mechanism that mediates Cdc2 activation during the
G2/M transition.36-38 In support of this, studies in fission yeast have shown that Nim1 negatively
regulates the S. pombe Wee1 by phosphorylation of its C-terminal catalytic region.39-41
Further, Xenopus Wee1 is inhibited by phosphorylation in its N-terminal noncatalytic region
by an unidentified protein kinase.42 However, the kinases responsible for the phosphorylation
and negative regulation of Wee1 in mammalian cells have not been determined.
The Wee1-related protein kinase, Myt1, is a cytoplasmic, membrane-bound kinase found
in Xenopus and mammalian cells that can phosphorylate Cdc2 on both Thr-14 and Tyr-15, but
has a strong preference for Thr-14.34,43,44 Human Myt1 localizes to the endoplasmic reticulum
and Golgi complex 45 and specifically phosphorylates and inactivates Cdc2 complexes to
inhibit G2/M progression.46 Myt1 interacts with Cdc2 complexes through its carboxy terminus
and overexpression of human Myt1 prevents entry into mitosis by sequestering cyclin B1/cdc2
complexes in the cytoplasm.47,48 Thus, Myt1 inhibits the G2/M transition by disrupting the
nuclear localization of cyclin B1/Cdc2 complexes, as well as by phosphorylating Cdc2 on Thr-
14 and Tyr-15. Another Wee1-related kinase, Mik1, has been identified in fission yeast. Mik1
acts redundantly with Wee1 in S. pombe to negatively regulate Cdc2 through phosphorylation
of Tyr-15.49 While a null allele of Mik1 has no discernible phenotype in fission yeast, a Mik1/
Wee1 double mutant undergoes mitotic lethality that is correlated with loss of tyrosine
phosphorylation on Cdc2.49
In eukaryotic cells, the Cdc25 family of dual-specificity phosphatases dephosphorylates
CDKs to mediate CDK activation and cell cycle progression. Mammalian cells contain three
isoforms of Cdc25 (A, B, and C) that share 40-50% amino acid identity and regulate distinct
cyclin/CDK complexes throughout the cell cycle.50,51 Cdc25A dephosphorylates cyclin E/CDK2
and cyclin A/CDK2 complexes to regulate the G1 and S-phase progression.52 Cdc25B activity
peaks during the G2-phase and ablation of Cdc25B activity prevents mitotic entry, suggesting
G2 Checkpoint and Anticancer Therapy 157
Cdc25B is a positive mitotic regulator.53-55 Further, Cdc25B dephosphorylates Cdc2 on Thr-14
and Tyr-15 in vitro6 and overexpression of Cdc25B causes cells to prematurely enter mitosis.56
Several studies implicate Cdc25A and Cdc25B as human oncogenes, as overexpression of these
proteins is observed in primary breast tumors as well as head and neck cancers.57,58
Cdc25C is required for entry into mitosis and is believed to be the major phosphatase that
dephosphorylates Thr-14 and Tyr-15 of Cdc2 in vivo.5 In cycling Xenopus egg extracts, the
activity of Cdc25 (biochemically equivalent to human Cdc25C) is low during S-phase and
increases in mitosis,51 and recombinant human Cdc25C dephosphorylates Cdc2 in vitro.50,59
Xenopus Cdc25 is extensively phosphorylated in its N-terminal noncatalytic region during mitosis
and active cyclin B/Cdc2 activates Cdc25 by phosphorylating the N-terminal region of Cdc25
in an autocatalytic loop.60,61 Thus only a small amount of active cyclin B/Cdc2 is required for
rapid activation of all the cyclin B/Cdc2 in the cell. In addition to cyclin B/Cdc2, other protein
kinases are likely to be important for Cdc25 activation; these kinases may mediate the initial
activation of cyclin B/Cdc2 during G2/M.62 For example, the polo-like kinase, Plx1, is associated
with Cdc25 in Xenopus egg extracts and phosphorylates Cdc25 on its amino terminus to stimulate
its activity.63 The mammalian Plx1 homolog, Plk1, also phosphorylates Cdc25C, and Plk1
activity is necessary for the functional maturation of centrosomes in late G2/early prophase
and for the establishment of a bipolar spindle.64,65 Plk1 may be a useful prognostic marker for
head and neck,66 esophageal,67 and non-small cell lung68 carcinomas, as overexpression in
these tumors is correlated with poorer prognosis.
G2 Checkpoint Activation
At key transitions during eukaryotic cell cycle progression, signaling pathways monitor
the successful completion of upstream events prior to proceeding to the next phase. These
regulatory pathways are commonly referred to as cell cycle checkpoints.69 Cells can arrest at
cell cycle checkpoints temporarily to allow for
(1) the repair of cellular damage,
(2) the dissipation of exogenous cellular stressors, or
(3) the accumulation of essential nutrients and growth factors.
Fig. 1. G2/M Transition. The activity of Cdc2 is tightly regulated by phosphorylation and protein interactions.
Cdc2 activation requires association with cyclin partners and phosphorylation by CAK. During Sand
G2-phases, cells accumulate cyclin B1/Cdc2 in an inactive form due to inhibitory phosphorylations
by Wee1 and Myt1 kinases. The conversion of Cdc2 from an inactive to active form is mediated by the
Cdc25C phosphatase. Mitotic events are stimulated by active cyclin B/Cdc2 complexes.
158 Cell Cycle Checkpoints and Cancer
Checkpoint signaling may also activate pathways leading to programmed cell death if
cellular damage cannot be properly repaired. Defects in cell cycle checkpoints can result in
gene mutations, chromosome damage, and aneuploidy, all of which result in permanent alteration
of the genome.70 The checkpoint mechanisms that normally regulate the fidelity of cell
cycle progression are frequently disrupted in tumor cells, verifying the importance of intact
checkpoint signaling pathways for maintenance of the genome.71
In addition to controlling normal mitotic entry, regulation of Cdc2 phosphorylation plays
a critical role in activating the G2 checkpoint after DNA damage. Genotoxic stress activates
cell cycle checkpoint pathways that initiate DNA repair and arrest at the G2/M transition to
prevent the propagation of DNA lesions during mitosis.72 Initial experiments in the fission
yeast S. pombe demonstrated that phosphorylation of Cdc2 on Tyr-15 is required for G2 arrest
after ionizing radiation treatment.73 Expression of a dominant Cdc2 mutant (Tyr-15 changed
to phenylalanine) completely abolishes the mitotic delay observed after ionizing radiation
exposure.73 The mechanism by which Tyr-15 phosphorylation is upregulated in fission yeast
after DNA damage involves both increased Wee1 protein levels and kinase activity, as well as
decreased Cdc25 (biochemical equivalent of human Cdc25C) levels and activity.73-75 The
importance of this two-step mechanism to modulate Tyr-15 phosphorylation of Cdc2 is
evidenced by the finding that while Wee1 inactivation in S. pombe increases sensitivity to both
ionizing76 and ultraviolet77 radiation, Wee1-deficient yeast maintain G2 checkpoint control.78
In contrast, inactivation of both Wee1- and Cdc25-dependent function completely abolishes
checkpoint control in addition to sensitizing cells to ionizing and ultraviolet (UV) radiation.75
Thus, upregulation of Cdc2 Tyr-15 phosphorylation after DNA damage requires both increased
phosphorylation of Tyr-15 and a reduced rate of Tyr-15 dephosphorylation.75 Recent studies
indicate that Mik1 may also function during the G2 DNA damage checkpoint in fission yeast, as
Mik1 protein levels are induced in response to ionizing radiation.79,80
Similar mechanisms appear to govern the DNA damage-induced G2 checkpoint arrest in
mammalian cells, as phosphorylation of Cdc2 on Thr-14 and Tyr-15 plays a critical role in the
DNA damage-induced G2 arrest (Fig. 2).81 Treatment of Chinese hamster ovary cells with
either etoposide or ionizing radiation results in rapid Cdc2 Tyr-15 phosphorylation and
inhibition of cyclin B/Cdc2 kinase activity.82,83 Increased Tyr-15 phosphorylation of Cdc2 is
observed in HL-60 human myeloid leukemia cells after exposure to ionizing radiation.84
Further, the overexpression of a non-phosphorylatable Cdc2 mutant (Thr-14 and Tyr-15 mutated
to alanine and phenylalanine, respectively), significantly reduces the G2 delay observed in
HeLa cells exposed to ionizing radiation.85 Numerous studies have also shown that after genotoxic
stress, the G2 checkpoint can be the predominant phase of cell cycle arrest in primary epithelial
cells, including keratinocytes86 as well as prostate,87 and bronchial epithelial cells.88
While the biochemical pathways involved in the DNA damage-induced G2 arrest in mammalian
cells entail signaling cascades that converge to inhibit Cdc2 activation72, specific pathways that
mediate G2 arrest after DNA damage in higher eukaryotes have only recently been elucidated.
In human hematopoietic cells, ionizing radiation activates the Lyn kinase, a member of the Src
family of protein tyrosine kinases.89 Lyn localizes to the nucleus and directly binds to Cdc2 in
hematopoietic cells exposed to ionizing radiation.90 Further, recombinant Lyn phosphorylates
Cdc2 on Tyr-15 in vitro, suggesting this kinase may be important in the G2 checkpoint
response in hematopoietic cells.90,91 In addition to upregulation of Cdc Tyr-15 phosphorylation,
decreased dephosphorylation of Cdc2 may also be important in G2 arrest in mammalian cells,
as impaired Cdc25C activation is observed in human lymphoma cells treated with nitrogen mustard,92 as
well as in HeLa cells treated with either ionizing93 or UV 94 radiation.
Recent studies have identified critical evolutionarily conserved G2 checkpoint pathways
mediated by members of the phosphoinositide-3 kinase (PI-3K) family in response to genotoxic
stress. After DNA damage, the PI-3K family members ATM (Tel1 in fission and budding
yeast) and ATR (Rad3 and Mec1 in fission and budding yeast, respectively), become activated
and initiate specific signal transduction pathways that regulate DNA repair and cell cycle arrest
G2 Checkpoint and Anticancer Therapy 159
in yeast, Xenopus, and mammalian cells (Fig. 2). ATM phosphorylates and activates the Chk2
kinase (Cds1 and Rad53 in fission and budding yeast, respectively) in cells exposed to ionizing
radiation.95-98 Similarly, ATR-dependent signaling mediates activation of the Chk1 kinase (Chk1
in fission and budding yeast) in cells treated with UV radiation.99,100 Activated Chk1 and
Chk2 can phosphorylate Cdc25C on Ser-216, generating a consensus binding site for 14-3-3
proteins.95,101-104 The binding of 14-3-3 proteins to Cdc25C results in the nuclear export of
Cdc25C, cytoplasmic sequestration of the phosphatase, and G2 arrest due to inhibition of
Cdc2 activity.105
Fig. 2. G2 Checkpoint Activation After Genotoxic Stress. In response to genotoxic stress, the ATM/ATR
signaling pathway is activated leading to phosphorylation and activation of Chk1 and Chk2 kinases and
subsequent phosphorylation of Cdc25C. This latter phosphorylation promotes the interaction of Cdc25C
with 14-3-3 adaptor proteins and inhibits the ability of Cdc25C to activate cyclin B1/Cdc2, resulting in
G2 cell cycle arrest.
160 Cell Cycle Checkpoints and Cancer
In addition to phosphorylating Cdc25 on Ser-216, activated Chk1 also phosphorylates
Wee1 in fission yeast, resulting in stabilization of the protein and thus prolonged Wee1 kinase
activity during G2 checkpoint responses.74 Further, Chk1 is essential for the DNA damageinduced
G2 checkpoint in S. pombe as Chk1-deficient cells have increased sensitivity to ionizing
radiation due to defective G2 arrest.106 Thus Chk1 integrates Rad3-dependent sigaling in
fission yeast to mediate the two-step mechanism to modulate Tyr-15 phosphorylation of Cdc2
through activation of the Wee1 kinase and inactivation the Cdc25 phosphatase.75
The importance of the PI-3K G2 checkpoint pathways is exemplified by the fact that
yeast and mammalian cells deficient in function of these family members have defective G2
checkpoint responses after DNA damage resulting in enhanced cellular sensitivity to genotoxic
stressors. Inactivation of Rad3107,108 and Mec1109 in yeast cells ablates G2 arrest after ionizing
radiation, while the expression of a kinase-defective ATR in human fibroblasts abrogates the
G2 checkpoint and sensitizes the cells to both ionizing and UV radiation.110 Further, homozygous
deletion of ATR in mice results in embryonic lethality, as cells die by chromosome
fragmentation and apoptosis.111,112
There are overlapping functions between the ATR and ATM classes of PI-3K kinases in
yeast. Deletion of the ATM homolog, Tel1, in fission113 or budding114 yeast has minimal effect
on DNA damage-induced checkpoint responses, as Tel1 functions to maintain telomere integrity
in yeast.115 In contrast, loss of ATM in mammalian cells eliminates G2 checkpoint function
and sensitizes cells to ionizing radiation,116,117 but does not alter checkpoint maintenance or
cellular sensitivity to UV radiation. ATM-null mice exhibit growth retardation, neurologic
dysfunction, infertility, defective T lymphocyte maturation, and sensitivity to ionizing
radiation.117,118 The majority of ATM-deficient animals develop malignant lymphomas by 4
months of age, while ATM-/- fibroblasts have abnormal radiation checkpoint function after
exposure to ionizing radiation.117,118 ATM function is defective in patients with ataxia telangiectasia,
a disorder in which patients have increased sensitivity to radiation119 and are highly prone to
the development of leukemias and lymphomas in childhood.120 Further, individuals harboring
a heterozygous germ line mutation of ATM have an elevated incidence of breast cancer.120,121
Chk1 and Chk2 proteins are also essential for maintenance of the DNA damage-induced
G2 checkpoint in yeast and mammalian cells. Loss of Chk1 in fission yeast,106 budding yeast,122
and mammalian cells.99,123 results in defective G2 checkpoint function after exposure to ionizing
radiation. Abrogation of Chk1 in mice results in early embryonic lethality.99,123 Similarly, Cds1124
and Rad53125,126 are essential for G2 checkpoint integrity after DNA damage in fission and
budding yeast, respectively. Mammalian Chk2-/- embryonic stem cells also fail to undergo G2
arrest after ionizing radiation treatment, while Chk2-/- thymocytes are resistant to DNA damageinduced
apoptosis.127 Of note, heterozygous germline mutations of Chk2 occur in a subset of
individuals with Li-Fraumeni syndrome, a highly penetrant familial cancer syndrome associated
with significantly increased rates of brain tumors, breast cancers, and sarcomas that is typically
associated with germline mutations in the tumor suppressor p53.128
While regulation of Cdc25C by PI-3K family members plays a critical role during G2
checkpoint responses, other kinases are capable of phosphorylating Cdc25C on Ser-216 in
response to DNA damage in mammalian cells. For example, C-TAK1 is a kinase that is
ubiquitously expressed in human cells and phosphorylates Cdc25C on Ser-216 to promote
14-3-3 protein binding.129 Similarly, Prk is a kinase expressed in human ovary, placenta, and
lung that also phosphorylates Cdc25C on Ser-216 in vitro.130-132 Prk mRNA expression is
downregulated in human lung tumors, suggesting that disruption of specific G2 checkpoint
pathways may contribute to tumorigenesis.130 However, a role for C-TAK1 and Prk-dependent
phosphorylation of Cdc25C after DNA damage has not been established.
Recently, the Plk1 kinase (Cdc5 in fission and budding yeast) was linked to G2 DNA
damage checkpoint signaling. Smits et al65 reported that Plk1 activity is inhibited in the G2
phase of the cell cycle in human tumor cells exposed to ionizing radiation, camptothecin, and
doxorubicin. The fission133,134 and budding135 yeast homologs of mammalian Plk, Cdc5, also
G2 Checkpoint and Anticancer Therapy 161
participate in G2 checkpoint signaling by preventing anaphase entry and mitotic exit after
DNA damage. Expression of a mutant Plk1 in which residues necessary for Plk1 activation are
altered, prevents Plk1 inactivation and leads to G2 override in cells treated with doxorubicin.65
Similarly, Plk1 activity is persistent during abrogation of the G2 checkpoint in tumor cells by
caffeine treatment.65 Several independent studies demonstrate that normal epithelial cells136
and fibroblasts64 undergo a G2 arrest in response to Plk inactivation in the absence of DNA
damage. In contrast, inhibition of Plk in human tumor cells under the same circumstances
results in mitotic catastrophe, although this latter event is independent of Plk1-mediated phosphorylation
and inactivation of Cdc25.64,136
In budding yeast, Xenopus, and mammalian cells, the G2 checkpoint response may also be
regulated by the Pin1 protein. Pin1 is an essential peptidyl-prolyl isomerase that inhibits entry
into mitosis and is also necessary for mitotic progression.137 Depletion of Pin1 from yeast or
mammalian cells induces mitotic arrest, while Pin1 overexpression in these cells results in a G2
arrest.137 In both Xenopus and human cells, Pin1 directly interacts with Wee1, Myt1, Cdc25C,
and Plk1 in a phosphorylation-dependent manner.138,139 The binding of Pin1 to Cdc25C
inhibits the phosphatase activity of the latter, thus accounting for the ability of Pin1 to block
mitotic entry.138 Further, Winkler et al140 determined that Pin1 is essential for DNA replication
checkpoint responses, as depletion of Pin1 from Xenopus extracts results in inappropriate
G2 progression and mitotic entry in the presence of aphidocolin. Depletion of Pin1 activity
from human tumor cells by multiple mechanisms, including overexpression of Pin1 antisense
mRNA, overexpression of a dominant-negative Pin1, and treatment of cells with a chemical
inhibitor of Pin1, juglone, confirms that Pin1 catalytic activity is essential for both tumor cell
survival and mitotic entry.141 However, further studies are necessary to determine if Pin1 function
is required for the DNA damage-induced G2 checkpoint.
Proteins that mediate direct repair or detection of DNA damage may also be essential for
proper G2 checkpoint activation. MLH1 is a necessary component of the DNA mismatch
repair machinery and is thought to participate in the G2 checkpoint in human tumor cells.
Ovarian tumor cells lacking MLH1 expression have defective G2 cell cycle arrest after cisplatin
and 6-thioguanine treatment.142 Further, MLH1-deficient human colon carcinoma cells also
have decreased survival and concomitant G2 checkpoint deficiency after ionizing radiation or
6-thioguanine exposure as compared to genetically matched cells in which mismatch repair
function has been restored.143 Similarly, MLH1-/- mouse embryo fibroblasts (MEFs) are sensitized
to ionizing radiation and 6-thioguanine and have impaired G2 arrest after exposure to
these agents as compared to wild-type (wt) MEFs.143 Interestingly, Brown et al142 observed loss
of MLH1 expression in 9 of 10 human ovarian cell lines after cisplatin treatment in vitro, as
well as the loss of MLH1 expression in 4 of 11 tumors biopsied during second look laparotomy
after chemotherapy. Since, MLH1-deficiency in human ovarian cells renders them resistant to
numerous chemotherapeutic agents, including cisplatin, doxorubicin, 6-thioguanine, and Nmethyl-
N-nitrosourea, these latter findings have important therapeutic implications.142 Taken
together, these studies suggest that MLH1-mediated regulation of the G2 checkpoint is indispensable
for proper DNA damage detection and repair, although the mechanism by which
MLH1 enforces G2 checkpoint integrity has not been elucidated.
Members of the homeobox family of proteins have also been implicated in regulation of
the G2 checkpoint. HSIX1 is a homeobox protein expressed during the S and G2 phases of the
cell cycle.144 Overexpression of HSIX1 in human breast cancer cells abrogates the G2 cell cycle
checkpoint response after ionizing radiation.144 Further, HSIX1 expression is absent or very
low in normal mammary tissue but is elevated in nearly half of primary breast cancers and 90%
of metastatic lesions.144 HSIX1 expression is also elevated in a variety of human cancer cell lines,
suggesting an important function for the protein in multiple tumor types.144 Further studies are
necessary to determine if HSIX1 function is mediated through regulation of the G2 checkpoint.
Finally, several members of the MAPK family are activated in response to ionizing radiation;
however, the biological relevance of these kinases to the G2 checkpoint is uncertain. Recently,
162 Cell Cycle Checkpoints and Cancer
the p38γ kinase was shown to have an essential function in the G2 checkpoint in human tumor
cells exposed to ionizing radiation.145 Activation of p38γ occurs in tumor cells treated with
cisplatin, etoposide, or ionizing radiation.145,146 Further, p38γ-dependent signaling is required for
DNA damage-induced G2 arrest, as disruption of p38γ-mediated signaling abrogates the G2
arrest and enhances the cytotoxicity observed in human tumor cells and human fibroblasts
treated with ionizing radiation.145 The DNA damage-mediated activation of p38γ is ATMdependent.
ATM-deficient cells fail to induce p38γ activity after ionizing radiation treatment;
however, the downstream targets of p38γ activity during G2 checkpoint signaling have not
been elucidated.145
G2 Checkpoint Maintenance
Numerous studies indicate that the G2 arrest response is comprised of an early or activation
stage as well as a subsequent maintenance phase, with p53 signaling implicated to play a role in
the latter.147-150 While the importance of p53-mediated signaling in G1/S checkpoint function is
well documented, the role of p53 signaling at the G2 checkpoint has only recently been well
defined. Early studies showing p53-deficient cells maintain the DNA damage-induced G2
arrest suggested that p53 does not function to regulate the G2 checkpoint.151,152 However,
expression of p53, in the absence of cellular stress, induces cell cycle arrest at both the G1 and
G2 checkpoints, suggesting p53 signaling modulates the G2 checkpoint response.153-155
Subsequent studies showed that p53 and p21WAF1/Cip1 (p21) are necessary to maintain a G2
arrest following DNA damage, since tumor cells lacking these proteins enter into mitosis with
accelerated kinetics.150,156
p53 utilizes multiple signaling pathways to modulate the G2 checkpoint (Fig. 3). One of
the initial components of p53-dependent G2 checkpoint maintenance is the transcriptional
upregulation of 14-3-3σ. 14-3-3sσ is induced in a p53-dependent manner by exposure to ionizing
radiation and doxorubicin in colorectal carcinoma cells.157 Further, overexpression of 14-3-3σ in
proliferating cells induces a G2 arrest.157 Deletion of both alleles of 14-3-3σ in colorectal
carcinoma cells results in abrogation of G2 arrest and premature mitotic entry after exposure to
ionizing radiation and doxorubicin.158 The p53-dependent increase in 14-3-3σ is thought to
modulate cyclin B1/Cdc2 signaling, as the binding of 14-3-3σ to Cdc2 results in cytoplasmic
sequestration of the kinase.158
In addition to transcriptional upregulation of 14-3-3σ, the mechanism of p53-dependent
G2 arrest involves inhibition of cyclin B1/Cdc2 activity by p21 and a subsequent reduction of
cyclin B1 and Cdc2 protein levels.156,159-161 Similar to its regulation of the cyclin E/CDK2
complex at the G1/S checkpoint, p21 binds to and inhibits the cyclin B1/Cdc2 complex in
vitro by blocking the activating phosphorylation of Cdc2 on Thr-161,162 although p21 has a
significantly lower affinity for the cyclin B1/Cdc2 complex as compared to the G1 phase kinase
complexes.163 Thus, DNA damage-induced G2 delay is regulated by modulation of both the
activating and inhibitory phosphorylations of Cdc2. The reduced expression of cyclin B1/
Cdc2 is mediated in part by p53-dependent transcriptional repression of the cyclin B1 and
Cdc2 promoters, although this transrepression is not due to direct interaction of p53 with
these promoters.156,164 The CCAAT-biding factor NF-Y was recently shown to mediate
transcriptional inhibition of cyclin B1 and Cdc2 during p53-dependent G2 arrest.165 Cdc2
transrepression may also result from the interaction of p130 and E2F4 with the Cdc2 promoter.166
The importance of p53-dependent regulation of Cdc2 activity is exemplified by the finding
that constitutive activation of cyclin B1/Cdc2 overrides p53-mediated G2 arrest.167
The p53-mediated decrease in cyclin B1 and Cdc2 transcription also requires the
retinoblastoma protein (pRB). Abrogation of pRB function in cells containing wt p53 blocks
the down regulation of cyclin B1 and Cdc2 expression and leads to an accelerated exit from G2
after genotoxic stress.156 Thus, similar to what occurs in cells that are p21 and p53 deficient,
pRB loss can uncouple S phase and mitosis after genotoxic stress in tumor cells. pRB is a
transcription repressor that, in its hypophosphorylated state, binds to the E2F-family (E2F) of
G2 Checkpoint and Anticancer Therapy 163
transcription factors and blocks E2F-dependent transcription of genes whose products are necessary
for S-phase entry, G2 progression, and M phase transition.168 Activated G1-phase cyclin/
CDK complexes phosphorylate pRB, resulting in the dissociation of E2F and pRB and cell
cycle progression after E2F-mediated transcription.
In addition to modulation of cyclin B1/Cdc2 activity by indirect transcriptional mechanisms,
p53 may also exert G2 checkpoint responses through transcriptional upregulation of
the downstream target genes, GADD45, and Reprimo. Increased expression of GADD45 in
primary fibroblasts results in a G2 arrest that can be abrogated by the overexpression of cyclin
B1 or Cdc25C.169 This GADD45-induced G2 arrest is p53-dependent since overexpression of
GADD45 in p53-deficient fibroblasts fails to mediate a G2 arrest.169 Also, GADD45 has been
shown to directly inhibit the cyclin B1/Cdc2 complex after UV radiation by binding to Cdc2.170
Of note, the GADD45-dependent G2 arrest is induced only after specific types of DNA damage.
Lymphocytes from GADD45 knockout mice failed to arrest after exposure to UV radiation,
but retained the G2 checkpoint initiated after exposure to ionizing radiation.169 Reprimo
was recently identified as a novel p53 downstream target that is induced in MEFs exposed to
ionizing radiation.171 Reprimo is a highly glycosylated protein that induces G2 arrest when
overexpressed in human tumor cells regardless of p53 status.171 Cdc2 and cyclin B1 protein
Fig. 3. G2 Checkpoint Maintenance After Genotoxic Stress. After DNA damage, activated ATM/ATR as
well as Chk1 and Chk2 can phosphorylate p53, resulting in stabilization and activation of the tumor
suppressor. p53-dependent signaling contributes to maintenance of the G2 cell cycle arrest by upregulating
the 14-3-3σ protein that binds to Cdc2 and sequesters the kinase in the cytoplasm. p53-dependent transcription
also elevates the CDK inhibitor p21, that binds to cyclin/CDK complexes to reduce phosphorylation
of pRB. Hypophosphorylated pRB remains bound to E2F, preventing E2F from mediating the
biosynthesis of cyclin B1 and Cdc2. p53 may also play a role in G2 checkpoint maintenance through
transcriptional upregulation of GADD45 and Reprimo. GADD45 can directly impede cyclin B1/Cdc2
activity after UV radiation by binding to Cdc2. Reprimo can inhibit Cdc2 activity and inhibit nuclear
translocation of cyclin B1.
164 Cell Cycle Checkpoints and Cancer
levels are unaffected by Reprimo overexpression, although Cdc2 activity and cyclin B1 nuclear
translocation are inhibited.171
p53-mediated transrepression of target genes may also contribute to its role in G2 arrest.
One critical cell cycle target of p53 transrepression is stathmin/oncoprotein 18.172,173 Stathmin
is frequently overexpressed in breast174,175 and ovarian176 cancers, as well as hematologic
malignancies,177,178 suggesting a critical function in cell cycle control. Stathmin is a microtubule-
associated phosphoprotein that functions in the regulation of microtubule dynamics
during mitosis.179 Inhibition of stathmin by mutation of sites required activating phosphorylations
by Cdc2 and MAPK prevents mitotic spindle formation.12,180-182 Overexpression of
stathmin can override p53-dependent G2 arrest in human tumor cells exposed to ionizing
radiation, indicating that regulation of stathmin expression is critical to p53-regulated G2
checkpoint responses.173
In addition to their previously described role in G2 checkpoint function, the ATM, ATR,
Chk1, and Chk2 kinases also directly phosphorylate the amino terminus of p53 after DNA
damage.183-186 ATM-and ATR-induced phosphorylation of human p53 on Ser-15 may be
important for p53 activation after genotoxic stress.183-187 Similarly, Chk1- and Chk2-mediated
phosphorylation of human p53 on Ser-20 may contribute to p53 activation after DNA
damage,127,188,189 since substitution of Ser-20 with alanine abrogates p53 stabilization after
exposure to either ionizing or UV radiation.190 Further, as previously described, cells deficient
in either ATM, ATR, Chk1, or Chk2 have defective G2 arrest after ionizing radiation, a
phenotype similar to that of p53-/- cells.150 However, p53 appears to function through the G1/
S checkpoint after activation by these kinases in response to DNA damage;189 although, a role
for these kinases in p53-dependent G2 checkpoint responses cannot be excluded.
The p53-mediated G2 checkpoint also modulates genomic integrity after MYC oncogene
overexpression. Felsher et al191 recently reported that overexpression of MYC protein in human
fibroblasts triggers aneuploidy through a mechanism involving p53-dependent G2 arrest. Loss
of the p53-mediated G2 checkpoint decreases the number of aneuploid cells after MYC
overexpression, as p53 inactivation reduces the population of G2-arrested cells with the potential
to become aneuploid.191 Thus, cells in the G2 or M phases of the cell cycle become aneuploid
regardless of p53 status in the presence of activated MYC; however, the loss of p53-dependent
cell cycle checkpoints and p53-mediated apoptosis enhances the ability of MYC-overexpressing
cells to progress through the cell cycle and become aneuploid.
Experiments analyzing viral proteins that inactivate p53 or pRB demonstrate the importance
of the p53-mediated, pRB-dependent G2 checkpoint after DNA damage. p53 function is
disrupted in cells by the ectopic expression of the human papillomavirus E6 protein. E6 binds
p53 and targets it for ubiquitin-mediated degradation, thus abrogating p53-dependent
signaling.192 Human fibroblasts and tumor cells expressing E6 have attenuated G2 checkpoint
function after ionizing radiation exposure148,156,193,194 or adriamycin,156,195 as a significantly
greater proportion of E6-containing cells enter mitosis as compared to control cells. Similarly,
loss of the G2 checkpoint is observed after exposure to ionizing radiation in fibroblasts196 and
tumor cells156 expressing the human papillomavirus E7 protein, as the E7 protein binds to and
inactivates pRB. The adenovirus E1A protein is also capable of disrupting the DNA damageinduced
G2 checkpoint.197,198 Mouse keratinocytes and human tumor cells expressing the
E1A protein are sensitized to treatment with cisplatin, adriamycin, and ionizing radiation due
to defective G2 arrest.199,200 However, it remains unclear if E1A-mediated disruption of the
G2 checkpoint is p53-dependent. Sanchez-Prieto et al199 failed to find a correlation between
the presence or absence of p53 in keratinocytes and the ability of E1A to impair G2 checkpoint
activation. In contrast, Bulavin et al197 found that in E1A-expressing rat fibroblasts exposed to
ionizing radiation, deregulation of p53-dependent signaling pathways contributed to defective
G2 delay. Regardless of its ability to modulate p53-dependent G2 checkpoint signaling,
E1A directly mediates transactivation of the Cdc2 promoter to facilitate G2 progression,
although the contribution of E1A-mediated events to abrogation of the G2 checkpoint
has not been elucidated.201
G2 Checkpoint and Anticancer Therapy 165
In some hematopoietic cell lines, p53 may accelerate the exit from G2 after DNA damage
and this accelerated mitotic entry contributes to p53-mediated apoptosis. In murine myeloid
leukemia cells bearing a temperature-sensitive p53 mutant, wt p53 positively modulates G2/M
progression after etoposide or ionizing radiation exposure.202,203 In these studies, the accelerated
G2/M progression induced by p53 was associated with enhanced cytotoxicity and apoptosis.
Similarly, myeloblast-enriched bone-marrow cells from p53+/+ mice have accelerated mitotic
entry after ionizing radiation as compared to cells from p53-/- mice.203 Thus, while p53 appears to
be required for DNA damage-induced G2 arrest in epithelial cells, the G2 checkpoint in
hematopoietic cells may be p53-independent.
Similar to hematopoietic cells, downregulation of Wee1 mRNA and protein expression
and loss of Cdc2 Tyr-15 phosphorylation is observed after activation of p53 in rat embryo
fibroblasts expressing the temperature-sensitive p53val135 mutant.204 Downregulation of Wee1
also occurs in thymus of p53+/+ but not p53-/- mice after exposure of the animals to ionizing
radiation.204 The p53-mediated reduction in Wee1 is likely due to a transrepression mechanism
similar to that discussed previously for cyclin B1 and Cdc2. Based upon the important
roles of Wee1 and Cdc2 in regulating the G2/M transition, this mechanism may represent one
biochemical pathway by which p53 modulates G2/M progression after DNA damage.
Modulation of the G2 Checkpoint—Therapeutic Implications
Since preclinical studies have shown that cells with defective checkpoint function are more
vulnerable to anticancer agents, it is hypothesized that the same will hold true in the clinical
setting. Indeed, numerous laboratories are now searching for compounds that interfere with
and/or override cell cycle checkpoints, in hope that such agents may be more effective in anticancer
therapy. A majority of tumor cells have defective G1 checkpoint function, making the
G2 checkpoint their “last line of defense” after exposure to DNA damaging agents. Thus, the
G2 checkpoint is a particularly attractive target for chemotherapeutic manipulation, since
ablation of G2 checkpoint function in tumor cells may result in enhanced susceptibility to
genotoxic anticancer drugs.
Chemical Approaches
There is strong evidence that abrogation of DNA damage-induced G2 arrest in human
cancer cell lines results in higher rates of apoptosis. Exposure of cells to ionizing radiation in
combination with caffeine or pentoxifylline, compounds which activate Cdc2 by activation of
Cdc25C phosphatase, results in G2 checkpoint override and increased rates of apoptosis.205-207
Caffeine disrupts the G2 checkpoint in p53-defective cells and results in radiosensitization of
tumor cells;205-207 however, the concentrations of caffeine required for abrogation of the G2
checkpoint in vitro are too cytotoxic for in vivo use. Nonetheless, elucidation of how caffeine
overrides the G2 checkpoint has provided important mechanistic insight to how the G2 checkpoint
can be modulated to enhance therapeutic efficacy. Caffeine inhibits the catalytic
activity of both ATM and ATR at drug concentrations similar to those that induce
radiosensitization.184,208 Treatment of tumor cells with caffeine blocks ATM-mediated
phosphorylation of Chk2 on Thr-68 after ionizing radiation.209 ATR phosphorylates Chk1 on
Ser-345 in human cells after ionizing and UV radiation exposure.99 In Xenopus, DNA damageinduced
phosphorylation of Chk1 is inhibited by caffeine,210 suggesting that caffeine will also
inhibit ATR-mediated phosphorylation of Chk1 in human cells. Caffeine also prevents the
ionizing and UV radiation-induced phosphorylation of p53 on Ser-15, presumably by disrupting
ATM and ATR function.208 However, since caffeine preferentially sensitizes p53-deficient cells
to DNA damage, the radiosensitizing effects of caffeine are most likely related to inhibition of
ATM- and ATR-mediated activation of Chk2 and Chk1, respectively.
Pentoxifylline is a methylxanthine derivative that enhances the sensitivity of a wide variety
of human tumor cells to DNA damaging agents.211-213 While the precise mechanism of action
of pentoxifylline has not been elucidated, several studies indicate that its ability to enhance
166 Cell Cycle Checkpoints and Cancer
chemosensitivity is a direct result of G2 checkpoint abrogation.211 Like caffeine, pentoxifylline
preferentially radiosensitizes p53-deficient cells.213,214 When used in combination with cisplatin,
thiotepa, carboplatin, or cyclophosphamide, pentoxifylline enhances the ability of these
compounds to inhibit the growth of murine mammary tumors in vivo.215,216 Pentoxifylline
also inhibits the growth of human bladder tumors217 and human lung tumors212 in mice
xenograft tumor models. These promising preclinical results have prompted the evaluation of
pentoxifylline in several clinical trials for efficacy against lung218 and cervical219 cancer in
combination therapy with genotoxic chemotherapeutics such as ionizing radiation and cisplatin.
Another related methylxanthine, lisofylline, also abrogates G2 checkpoint function and sensitizes
p53-deficient human tumor cells to ionizing radiation216,220 and cisplatin.221 Further, lisofylline
is more effective than pentoxifylline at enhancing the sensitivity of murine mammary tumor
cells to ionizing radiation.216,220
Staurosporine is a non-specific protein kinase inhibitor that can override DNA damageinduced
G2 delay in response to ionizing radiation.222 However, the cytotoxicity of staurosporine
has limited its potential clinical efficacy, leading to the development of staurosporine analogs
with improved specificity and reduced cytotoxicity.223 One such staurosporine derivative,
UCN-01, is also a potent abrogator of the G2 cell cycle checkpoint and increases the cytotoxic
effect of DNA-damaging agents in human tumor cells.224,225 UCN-01 significantly inhibits
the growth of a variety of human tumors in mice xenograft tumor models226,227 and is
currently in Phase I clinical trials showing promising results.228 Preclinical studies have provided
many mechanistic insights to UCN-01 activity. Treatment of tumor cells with UCN-01 results
in Wee1 inactivation and Cdc25C activation, although these are indirect effects of UCN-01
inhibition of upstream checkpoint kinases.229 UCN-01 specifically inhibits Chk1, as the related
Chk2 kinase and the upstream ATM kinase are refractory to inhibition by UCN-01.230
UCN-01 also inhibits C-TAK1 in vitro, although the contribution of this kinase to
phosphorylation of Cdc25C during G2 checkpoint activation has not been fully elucidated.231
Interestingly, UCN-01 selectively ablates the G2 checkpoint in cancer cells with defective p53
function.224 Since ablation of G2 checkpoint function by UCN-01 can occur in the absence of
p53, signaling to substrates other than p53 must be sufficient for G2 override.
Another approach to disrupt G2 checkpoint is the use of cell permeable peptides that can
block specific G2 signaling components. For example, Suganuma et al232 engineered short
peptides corresponding to amino acids 211-221 of human Cdc25C fused with the retroviral
TAT protein.232 The TAT protein allows these fusion proteins to permeabilize the cell membrane
and accumulate in excess of the endogenous Cdc25C, thus blocking Chk1 and Chk2 kinase
activity toward the endogenous Cdc25C. Human tumor cells treated with these Cdc25C
peptides are sensitized to DNA damage due to defective G2 checkpoint response, suggesting
that Chk1 and Chk2 are effective targets to mediate abrogation of the G2 checkpoint.232
Evidence supporting this latter strategy is provided by the recent report of SB-2180708, a
selective Chk1 inhibitor structurally related to staurosporine that disrupts the G2 checkpoint.233
In the presence of SB-2180708, HeLa cells exposed to ionizing radiation or topoisomerase I
inhibitors fail to undergo a G2 arrest.233 Further, inhibition of Chk1 activity in HeLa cells
enhanced the cytotoxic effects of genotoxic agents, thus supporting the validity of Chk1 as a
target for G2 checkpoint override.233
The p38γ cascade is another potential target for the development of radiosensitizing agents.
Activation of p38γ is required for G2 arrest in tumor cells and fibroblasts exposed to ionizing
radiation and disruption of p38γ-dependent signaling enhances the sensitivity of these cells to
ionizing radiation.145 Since p38γ is inactive under normal cellular growth conditions and
elimination of p38γ-dependent signaling does not alter normal cell cycle progression,145 the
molecules in this pathway are potential targets for the development of inhibitors that will
mediate increased sensitivity to radiation therapy. Thus, a further understanding of the mechanism
by which p38γ regulates G2 arrest may lead to the development of novel strategies for the
improvement of radiation therapy.
G2 Checkpoint and Anticancer Therapy 167
While the majority of anticancer drugs activate the G2 checkpoint after genotoxic stress
by directly targeting kinases involved in DNA damage signaling pathways, alternative compounds
that regulate the G2 checkpoint exist. For example, histone deacetylase inhibitors trigger
a G2 arrest in normal human cells; however, this G2 arrest fails to occur in a diverse range
of human tumor cells and they undergo mitotic catastrophe.234 These compounds block
histone deacetylase activity, increasing the acetylation state of the chromatin, altering chromatin
structure and regulation of gene expression. These inhibitors may represent a novel mechanism of
G2 override in tumor cells in the absence of concurrent DNA damage.235 Of note, histone
deacetylase inhibitors upregulate p21 to mediate cell cycle arrest at the G1/S and G2/M
transitions,236,237 as p21-/- colon carcinoma cells are resistant to these compounds.238 The anticancer
potential of histone deacetylase inhibitors has been demonstrated in both in vitro236,239
and in vivo240 model systems and several histone deacetylase inhibitors are currently being used
in clinical trials with promising early results.241
In addition to providing insight to mechanisms that result in enhanced clinical efficacy, a
further understanding of G2 checkpoint function may also lead to the identification of the
signaling pathways that mediate tumor cell chemoresistance. For example, overexpression of
the receptor tyrosine kinase, ERbB2 (HER2/neu), results in Paclitaxel resistance in breast
cancers.242 Yu et al242 demonstrated that ERbB2 overexpression in breast cancer cells results in
upregulation of p21, which binds cyclin B1/Cdc2 complexes and inhibits Paclitaxel-mediated
Cdc2 activation and mitotic entry.242 It is hypothesized that the ERbB2-mediated G2 arrest
inhibits the action of Paclitaxel, which requires cell transition into mitosis and/or cyclin B1/
Cdc2 activation.243 Based on this latter study, it was hypothesized that the Paclitaxel resistance
of ERbB2-overexpressing tumors could be eliminated by downregulation of ERb2 function. In
support of this hypothesis, Baselga et al244 recently demonstrated that combinatorial use of
Paclitaxel and anti-HER2 antibodies results in significant growth inhibition of HER2
overexpressing human breast cancer xenograft tumors as compared to treatment with either
agent alone.
Screens for New Compounds
In an effort to identify novel anticancer agents, including those which may abrogate G2
checkpoint function, the National Cancer Institute has utilized a panel of 60 human tumor
cell lines in a drug screen to identify and characterize compounds with anticancer activity.245
To date, 70,000 compounds have been tested in these cell lines and the results recorded in a
database.245,246 Of note, p53 gene mutations occur in a majority of the NCI drug screen cell
lines, as 39 of 58 cell lines analyzed contain a mutant p53 sequence and have defective biochemical
p53 activity.247 Thus, many of the cell lines analyzed in the drug screen are predicted
to have impaired G2 checkpoint maintenance following DNA damage. In support of this
hypothesis, cell lines containing mutant p53 exhibit less growth inhibition in this screen than
the wt p53 cell lines when treated with the majority of clinically used anticancer agents, including
DNA cross-linking agents, antimetabolites, and topoisomerase I and II inhibitors.247 This latter
result suggests that disruption of the G2 checkpoint may enhance the clinical efficacy of anticancer
reagents and exemplifies the need for the development of novel agents that might
induce G2 checkpoint override. Amundson et al recently evaluated the basal expression levels
of 10 transcripts from genes that participate in DNA damage signaling pathways in these same
NCI cell lines and correlated this data with the sensitivity of the cells to a panel of 122 standard
chemotherapy agents.246 Further, cDNA microarray analyses have been used to assess gene
expression profiles of these same 60 cancer cell lines in response to several standard
chemotherapeutic drugs.248 Grouping the cell lines by patterns of gene expression resulted in
different relationships than those obtained by clustering the cell lines as a function of their
chemosensitivity.248 Analyses such as these may provide valuable insight as to how the transcription
of specific genes relates to drug sensitivity. Similar strategies can be utilized in the future
to examine the patterns of gene expression after treatment of cells with novel anticancer agents.
168 Cell Cycle Checkpoints and Cancer
Investigators are also developing high-throughput screens to identify G2 checkpoint
inhibitors. Roberge et al249 used MCF-7 breast cancer cells that express a dominant-negative
mutant p53 in a high-throughput screen for compounds that override ionizing radiationinduced
G2 arrest and allow entry into mitosis. The loss of wt p53 function in these cells
eliminates G1 checkpoint function and causes the majority of cells to arrest at the G2 checkpoint
after exposure to ionizing radiation. To identify compounds that can ablate G2 checkpoint
function, the mutant MCF-7 cells were grown in 96-well plates, irradiated to induce G2
arrest, and then co-treated with nocodazole and various extracts from marine invertebrates.249
In the presence of a compound that could override the IR-induced G2 arrest, the cells were
trapped in mitosis by the presence of the microtubule inhibitor nocodazole. The plates were
then rapidly screened by use of an antibody that recognizes a phosphorylated form of nucleolin
present only in mitotic cells. The reported screening process of 1300 extracts was validated by
the isolation of staurosporine, a previously described G2 checkpoint inhibitor.249 The screen
identified one novel G2 inhibitor, isogranulatimide, a structurally unique compound.
Isogranulatimide shows only mild toxicity to cells when used alone; however, treatment of the
MCF-7 cells expressing mutant p53 with ionizing radiation and isogranulatimide results in
synergistic cytotoxicity.249 The use of this type of assay to identify G2 checkpoint inhibitors
should allow further isolation of novel compounds that override G2 arrest.
Genetic Approaches
Since the major signaling pathways and cell cycle checkpoints are conserved between yeast
and mammalian cells, yeast model systems may be manipulated to determine the molecular
mechanism of anticancer drugs and to identify novel molecular targets for rational drug design.
In contrast to mammalian cells, yeast offer the unique advantage of simple and rapid genetic
manipulations that, coupled with the availability of the S. cerevisiae genomic sequence and the
ongoing S. pombe genome sequencing project,250 makes this organism an attractive model in
which to evaluate chemotherapeutic agents. The Seattle Project encompassed multiple approaches
for the discovery of anticancer targets and drugs through use of yeast genomics.251 In one
approach, a panel of isogenic yeast strains, each having single or multiple mutations in pathways
involved in DNA repair, cell cycle checkpoint function, or cell cycle regulation were
generated and used to screen new and existing anticancer drugs.251 In a separate effort to
identify novel cellular pathways to target for anticancer drug discovery, Norman et al252
utilized budding yeast to genetically select peptide inhibitors. In this latter approach, peptides
were selected based on phenotypic analyses and genetic dissections of candidate target pathways
were performed to identify putative targets of the inhibitors.252
An additional example of the power of yeast genomics is the ability to perform cDNA
microarray analyses to determine how genome-wide expression can be modulated in cells
exposed to anticancer compounds.253,254 This methodology was recently exploited by Jelinsky
and Samson, as they used DNA chip technology (with the 6,200 S. cerevisiae genes
represented) to compare the transcription profiles of S. cerevisiae treated with an alkylating
agent, methyl methanesulfonate, to those of untreated cells.254 The potential for microarray
analyses to integrate existing regulatory networks is exemplified by the recent use of this technology
to link the DNA excision repair pathway of S. cerevisiae to proteasome-associated
control elements.255 Further, Hughes et al256 recently generated a database of expression profiles
corresponding to 300 diverse mutations and chemical treatments in S. cerevisiae and used
this database to identify novel genes required for various cellular functions as well as to identify
novel target genes of known compounds. Of note, subsequent findings indicate that many
yeast mutants exhibit chromosome-wide aneuploidy as compared to isogenic parental
wt strains.257 This observation has significant implications for interpreting whole-genome transcriptional
expression profile data, particularly data obtained from malignant or immortalized
cells that are known to be genetically unstable.257 Genetic approaches such as those described
G2 Checkpoint and Anticancer Therapy 169
here will further validate the signaling pathways used by current chemotherapies, identify compounds
with preferential lethality to cells with defective checkpoint function, and reveal additional
signaling pathways to target for new drug discovery.
Future Directions
A fundamental challenge in the development of anticancer agents is the identification of
molecular differences between cancer cells and normal cells that can be targeted for chemotherapeutic
intervention to preferentially eliminate cancer cells while minimizing the toxicity
to normal tissues. The drug discovery process for cancer will continue to be transformed by the
wealth of information generated by the genome projects across many organisms, both prokaryotic
and eukaryotic. As our understanding of cell cycle regulation and checkpoints increases so
will the number of signaling molecules and pathways that can be used as targets for rational
drug and therapy design. The hope is that from a detailed understanding of these processes,
more incisive, mechanism-based approaches to cancer treatment will evolve that exploit the
molecular defects in human tumors. To achieve this goal, we need to continue to develop
(1) technologies to precisely define the checkpoint defects in individual tumors,
(2) panels of anticancer agents that target cells with defined genetic alterations, and
(3) treatment regimens that are tailored to the resulting cell cycle phenotype.
Acknowledgments
This work was supported by the Susan G. Komen Breast Cancer Foundation Grant 99-
3038 (Z.A.S.), National Institutes of Health Institutional Training Grant GM07347 (Z.A.S.),
National Institutes of Health Grants CA70856 (J.A.P.) and U.S. Army Grant DAMD 17-99-
1-9422 (J.A.P.).
References
1. Nurse P. Universal control mechanism regulating onset of M-phase. Nature 1990;344:503-508.
2. Morgan DO. Principles of CDK regulation. Nature 1995; 374:131-134.
3. Russo AA, Jeffrey PD, Pavletich NP. Strucutral basis of cyclin-dependent kinase activation by
phosphorylation. Nature Struct Biol 1996; 3:696-700.
4. Draetta GF. Cell cycle: Will the real CDK-activating kinase please stand up. Curr Biol 1997; 7:50-
52.
5. Millar JBA, Blevitt J, Gerace L et al. p55Cdc25 is a nuclear protein required for the initiation of
mitosis in human cells. Proc Natl Acad Sci USA 1997; 88:10500-10504.
6. Honda R, Ohba Y, Nagata A et al. Dephosphorylation of human p34cdc2 kinase on both Thr-14
and Tyr-15 by human Cdc25B phosphatase. FEBS Lett 1993; 318:331-334.
7. Peter M, Nakagawa J, Doree M et al. In vitro disassembly of the nuclear lamina and M phasespecific
phosphorylation of lamins by cdc2 kinase. Cell 1990; 61:591-602.
8. Arion D, Meijer L, Brizuela L, Beach D. Cdc2 is a component of the M phase-specific histone H1
kinase: evidence for identity with MPF. Cell 1988; 55:371-378.
9. O’Connor DS, Grossman D, Plescia J et al. Regulation of apoptosis at cell division by p34cdc2
phosphorylation of survivin. Proc Natl Acad Sci USA 2000; 97:13103-13107.
10. Tombes RM, Peloquin JG, Borisy GG. Specific association of an M-phase kinase with isolated
mitotic spindles and identification of two of its substrates as MAP4 and MAP1B. Cell Regul 1991;
2(11):861-874.
11. Blangy A, Arnaud L, Nigg EA. Phosphorylation by p34cdc2 protein kinase regulates binding of the
kinesin-related motor HsEg5 to the dynactin subunit p150. J Biol Chem 1997; 272:19418-19424.
12. Marklund U, Brattsand G, Shingler V, Gullberg M. Serine 25 of oncoprotein 18 is a major cytosolic
target for the mitogen-activated protein kinase. J Biol Chem 1993; 268:15039-15047.
13. Yang R, Morosetti R, Koeffler HP. Characterization of a second human cyclin A that is highly
expressed in testis and in several leukemic cell lines. Cancer Res 1997; 57:913-920.
14. Yang R, Nakamaki T, Lubbert M et al Cyclin A1 expression in leukemia and normal hematopoietic
cells. Blood 1999; 93:2067-2074.
15. Liu D, Matzuk MM, Sung WK et al. Cyclin A1 expression is required for meiosis in the male
mouse. Nat Genet 1998; 20:377-380.
170 Cell Cycle Checkpoints and Cancer
16. Liu D, Liao C, Wolgemuth DJ. A role for cyclin A1 in the activation of MPF and G2-M transition
during meiosis of male germ cells in mice. Dev Biol 2000;224:388-400.
17. Muller C, Yang R, Park DJ et al. The aberrant fusion proteins PML-RARalpha and PLZF-RARalpha
contribute to the overexpression of cyclin A1 in acute promyelocytic leukemias. Blood 2000;
96:3894-3899.
18. Pines J, Hunter T. Human cyclin A is adenovirus E1A-associated protein p60 and behaves differently
from cyclin B. Nature 1990; 346:760-763.
19. Pagano M, Pepperkok R, Verde F et al. Cyclin A is required at two points in the human cell cycle.
EMBO J 1992; 11:961-971.
20. Murphy M, Stinnakre MG, Senamaud-Beaufort C et al. Delayed early embryonic lethality following
disruption of the murine cyclin A2 gene. Nat Genet 1997; 15:83-86.
21. Chapman DL, Wolgemuth DJ. Identification of a mouse B-type cyclin which exhibits developmentally
regulated expression in the germ line. Mol Reprod Dev 1992; 33:259-269.
22. Pines J, Hunter T. Isolation of a human cyclin cDNA: Evidence for cyclin mRNA and protein
regulation in the cell cycle and for interaction with p34 cdc2. Cell 1989; 58:833-846.
23. Chapman DL, Wolgemuth DJ. Isolation of the murine cyclin B2 cDNA and characterization of
the lineage and temporal specificity of expression of the B1 and B2 cyclins During oogenesis,
spermatogenesis and early embryogenesis. Development 1993; 118:229-240.
24. Kreutzer MA, Richards JP, De Silva-Udawatta MN et al. Caenorhabditis elegans cyclin A- and Btype
genes: A cyclin A multigene family, an ancestral cyclin B3 and differential germline expression.
J Cell Sci 1995; 108:2415-2424.
25. Gallant P, Nigg EA. Identification of a novel vertebrate cyclin: Cyclin B3 shares properties with
both A- and B-type cyclins. EMBO J 1994; 13:595-605.
26. Jackman M, Firth M, Pines J. Human cyclins B1 and B2 are localized to strikingly different structures: B1
to microtubules, B2 primarily to the Golgi apparatus. EMBO J 1995; 14:1646-1654.
27. Pines J, Hunter T. Human cyclins A and B1 are differentially located in the cell and undergo cell
cycle-dependent nuclear transport. J Cell Biol 1991; 115:1-17.
28. Brandeis M, Rosewell I, Carrington M et al. Cyclin B2-null mice develop normally and are fertile
whereas cyclin B1-null mice die in utero. Proc Natl Acad Sci USA 1998; 95:4344-4349.
29. Li J, Meyer AN, Donoghue DJ. Nuclear localization of cyclin B1 mediates its biological activity
and is regulated by phosphorylation. Proc Natl Acad Sci USA 1997; 94:502-507.
30. Hagting A, Karlsson C, Clute P et al. MPF localization is controlled by nuclear export. EMBO J
1998; 17:4127-4138.
31. Toyoshima F, Moriguchi T, Wada A et al. Nuclear export of cyclin B1 and its possible role in the
DNA damage-induced G2 checkpoint. EMBO J 1998; 17:2728-2735.
32. Yang J, Bardes ESG, Moore JD et al. Control of Cyclin B1 localization through regulated binding
of the nuclear export factor CRM1. Genes Dev 1998; 12:2131-2143.
33. Parker LL, Piwnica-Worms H. Inactivation of p34cdc2-cyclin B complex by the human WEE1
tyrosine kinase. Science 1992; 257:1955-1957.
34. Mueller PR, Coleman TR, Kumagai A, Dunphy A. Myt1: A membrane-associated inhibitory kinase
that phosphorylates cdc2 on both threonine-14 and tyrosine-15. Science 1995;270:86-93.
35. Nurse P, Thuriaux P. Regulatory genes controlling mitosis in the fission yeast Schizosaccharomyces
pombe. Genetics 1980; 96:627-637.
36. Watanabe N, Broome M, Hunter T. Regulation of the human WEE1Hu CDK tyrosine 15-kinase
during the cell cycle. EMBO J 1995; 14:1878-1891.
37. Parker LL, Sylvestre PJ, Byrnes MJ et al. Identification of a 95-kDa WEE1-like tyrosine kinase in
HeLa cells. Proc Natl Acad Sci USA 1995; 92:9638-9642.
38. McGowan CH, Russell P. Cell cycle regulation of human WEE1. EMBO J 1995;14:2166-2175.
39. Wu L, Russell P. Nim1 kinase promotes mitosis by inactivating Wee1 tyrosine kinase. Nature
1993; 363:738-741.
40. Parker LL, Walter SA, Young P, Piwnica-Worms H. Phosphorylation and inactivation of the mitotic
inhibitor Wee1 by the nim1/cdr1 kinase. Nature 1993; 363:736-738.
41. Coleman TR, Tang Z, Dunphy WG. Negative regulation of Wee1 protein kinase by direct action
of the nim1/cdr1 mitotic inducer. Cell 1993; 72:919-929.
42. Tang Z, Coleman TR, Dunphy WG. Two distinct mechanisms for negative regualtion of the
Wee1 protein kinase. EMBO J 1993; 12:3427-3436.
43. Atherton-Fessler S, Liu F, Gabrielli B et al. Cell cycle regulation of the p34cdc2 inhibitory kinases.
Mol Biol Cell 1994; 5:989-1001.
44. Kornbluth S, Sebastian B, Hunter T, Newport J. Membrane localization of the kinase which phosphorylates
p34cdc2 on threonine 14. Mol Biol Cell 1994; 5:273-282.
G2 Checkpoint and Anticancer Therapy 171
45. Liu F, Stanton JJ, Wu Z, Piwnica-Worms H. The human Myt1 kinase preferentially phosphorylates
Cdc2 on threonine 14 and localizes to the endoplasmic reticulum and golgi complex. Mol
Cell Biol 1997; 17:571-583.
46. Booher RN, Holman PS, Fattaey A. Human Myt1 is a cell cycle-regulated kinase that inhibits
Cdc2 but not CDK2 activity. J Biol Chem 1997; 272:22300-22306.
47. Wells NJ, Watanabe N, Tokusumi T et al. The C-terminal domain of the Cdc2 inhibitory kinase
Myt1 interacts with Cdc2 complexes and is required for inhibition of G2/M progression. J Cell Sci
1999; 112:3361-3371.
48. Liu F, Rothblum-Oviatt C, Ryan CE, Piwnica-Worms H. Overproduction of human Myt1 kinase
induces a G2 cell cycle delay by interfering with the intracellular trafficking of Cdc2-cyclin B1
complexes. Mol Cell Biol 1999; 19:5113-5123.
49. Lundgren K, Walworth N, Booher R, et al. Mik1 and wee1 cooperate in the inhibitory tyrosine
phosphorylation of cdc2. Science 1991; 270:86-90.
50. Strausfeld U, Labbe JC, Fesquet D et al. Dephosphorylation and activation of a p34cdc2/cyclin B
complex in vitro by human CDC25C protein. Nature 1991;351:242-245.
51. Galaktionov K, Beach D. Specific activation of Cdc25 tyrosine phosphatases by B-type cyclins:
Evidence for multiple roles of mitotic cyclins. Cell 1991; 67:1181-1194.
52. Blomberg I, Hoffmann I. Ectopic expression of Cdc25A accelerates the G1/S transition and leads to premature
activation of cyclin E- and cyclin A-dependent kinases. Mol Cell Biol 1999; 19:6183-6194.
53. Lammer C, Wagerer S, Saffrich R et al. The Cdc25B phosphatase is essential for the G2/M phase
transition in human cells. J Cell Sci 1998; 111:2445-2453.
54. Garner-Hamrick PA, Fisher C. Antisense phosphorothioate oligonucleotides specifically down-regulate
Cdc25B causing S-phase delay and persistent antiproliferative effects. Int J Cancer 1998; 76:720-728.
55. Nilsson I, Hoffmann I. Cell cycle regulation by the Cdc25 phosphatase family. Prog Cell Cycle
Res 2000; 4:107-114.
56. Karlsson C, Katich S, Hagting A et al. Cdc25B and Cdc25C differ markedly in their properties as
initiators of mitosis. J Cell Biol 1999; 146:573-583.
57. Galaktionov K, Lee AK, Eckstein J et al. CDC25 phosphatases as potential human oncogenes.
Science 1995; 269:1575-1577.
58. Gasparotto D, Maestro R, Piccinin S et al. Overexpression of CDC25A and CDC25B in head and
neck cancers. Cancer Res 1997; 57:2366-2368.
59. Kumagai A, Dunphy WG. The Cdc25 protein controls tyrosine dephosphorylation of the cdc2
protein in a cell-free system. Cell 1991; 64:903-914.
60. Hoffman I, Clarke PR, Marcote MJ et al. Phosphorylation and activation of human Cdc25C
by cdc2-cyclin B and its involvement in the self-amplification of MPF at mitosis. EMBO J
1993; 12:53-63.
61. Izumi T, Maller JL. Elimination of cdc2 phosphorylation sites in the Cdc25 phosphatase blocks
initiation of M-phase. Mol Biol Cell 1993; 4:1337-1350.
62. Izumi T, Maller JL. Phosphorylation and activation of the Xenopus Cdc25 phosphatase in the
absence of Cdc2 and CDK2 kinase activity. Mol Biol Cell 1995; 6:215-226.
63. Kumagai A, Dunphy WG. Purification and molecular cloning of Plx1, a Cdc25-regulatory kinase
from Xenopus egg extracts. Science 1996; 273:1377-1380.
64. Lane HA, Nigg EA. Antibody microinjection reveals an essential role for human Polo-like kinase 1
(Plk1) in the functional maturation of mitotic centrosomes. Nature Med 1996;2:630-631.
65. Smits VAJ, Klompmaker R, Arnaud L et al. Polo-like kinase-1 is a target of the DNA damage
checkpoint. Nat Cell Biol 2000; 2:672-676.
66. Knecht R, Elez R, Oechler M et al. Prognostic significance of polo-like kinase (PLK) expression in
squamous cell carcinomas of the head and neck. Cancer Res 1999;59:2794-2797.
67. Tokumitsu Y, Mori M, Tanaka S et al. Prognostic significance of polo-like kinase expression in
esophageal carcinoma. Int J Oncol 1999; 15:687-692.
68. Wolf G, Elez R, Doermer A et al. Prognostic significance of polo-like kinase (PLK) expression in
non-small cell lung cancer. Oncogene 1997; 14:543-549.
69. Hartwell LH, Weinert TA. Checkpoints: Controls that ensure the order of cell cycle events.
Science 1989; 246:629-634.
70. Paulovich AG, Toczyski DP, Hartwell LH. When checkpoints fail. Cell 1997; 88:315-321.
71. Stewart ZA, Pietenpol JA. Cell cycle checkpoints as therapeutic targets. J Mammary Gland Biol
Neoplasia 1999; 4:389-400.
72. Hwang A, Muschell RJ. Radiation and the G2 phase of the cell cycle. Radiat Res 1998;
150:S52-S59.
73. Rhind N, Furnari B, Russell P. Cdc2 tyrosine phosphorylation is required for the DNA damage
checkpoint in fission yeast. Genes Dev 1997; 11:504-511.
172 Cell Cycle Checkpoints and Cancer
74. O’Connell MJ, Raleigh JM, Verkade HM, Nurse P. Chk1 is a wee1 kinase in the G2 DNA damage
checkpoint inhibiting cdc2 by Y15 phosphorylation. EMBO J 1997; 16:545-554.
75. Raleigh JM, O’Connell MJ. The G2 DNA damage checkpoint targets both Wee1 and Cdc25. J
Cell Sci 2000; 113:1727-1736.
76. Rowley R, Hudson J, Young PG. The wee1 protein kinase is required for radiation-induced mitotic
delay. Nature 1992; 356:353-355.
77. al-Khodairy F, Carr AM. DNA repair mutants defining G2 checkpoint pathways in
Schizosaccharomyces pombe. EMBO J 1992;11:1343-1350.
78. Barbet NC, Carr AM. Fission yeast wee1 protein kinase is not required for DNA damage-dependent
mitotic arrest. Nature 1993; 364:824-827.
79. Christensen PU, Bentley NJ, Martinho RG et al. Mik1 levels accumulate in S-phase and may mediate
an intrinsic link between S phase and mitosis. Proc Natl Acad Sci USA 2000; 97:2579-2584.
80. Baber-Furnari BA, Rhind N, Boddy MN et al. Regulation of mitotic inhibitor Mik1 helps to
enforce the DNA damage checkpoint. Mol Biol Cell 2000; 11:1-11.
81. Blasina A, Paegle ES, McGowan CH. The role of inhibitory phosphorylation of CDC2 following
DNA replication block and radiation-induced damage in human cells. Mol Biol of the Cell 1997;
8:1013-1023.
82. Lock RB, Ross WE. Inhibition of p34cdc2 kinase activity by etoposide or irradiation as a mechanism
of G2 arrest in Chinese hamster ovary cells. Cancer Res 1990; 50:3761-3766.
83. Lock RB. Inhibition of p34cdc2 kinase activation, p34cdc2 tyrosine dephosphorylation, and mitotic
progression in Chinese hamster ovary cells exposed to etoposide. Cancer Res 1992; 52:1817-1822.
84. Kharbanda S, Saleem A, Datta R. Ionizing radiation induces rapid tyrosine phosphorylation of
p34cdc2. Cancer Res 1994; 54:1412-1414.
85. Jin P, Gu Y, Morgan DO. Role of inhibitory CDC2 phosphorylation in radiation-induced G2
arrest in human cells. J Cell Biol 1996; 134:963-970.
86. Herzinger T, Funk JO, Hillmer K et al. Ultraviolet B irradiation-induced G2 cell cycle arrest in
human keratinocytes by inhibitory phosphorylation of the cdc2 cell cycle kinase. Oncogene 1995;
11:2151-2156.
87. Girinsky T, Koumenis G, Graeber TG et al. Attenuated response of p53 and p21 in primary
cultures of human prostatic epithelial cells exposed to DNA-damaging agents. Cancer Res 1995;
55:3726-3731.
88. Gadbois DM, Lehnert BE. Cell cycle response to DNA damage differs in bronchial epithelial cells
and lung fibroblasts. Cancer Res 1997; 57:3174-3179.
89. Kharbanda S, Yuan ZM, Rubin E et al. Activation of Src-like p56/p53lyn tyrosine kinase by ionizing
radiation. J Biol Chem 1994; 269:20739-20743.
90. Kharbanda S, Saleem A, Yuan ZM et al. Nuclear signaling induced by ionizing radiation
involves colocalization of the activated p56/p53lyn tyrosine kinase with p34cdc21. Cancer Res
1996; 56:3617-3621.
91. Uckun FM, Tuel-Ahlgren L, Waddick KG et al. Physical and functional interactions between Lyn
and p34cdc2 kinases in irradiated human B-cell precursors. J Biol Chem 1996;271:6389-6397.
92. O’Connor PM, Ferris DK, Hoffman I et al. Role of the Cdc25C phosphatase in G2 arrest induced
by nitrogen mustard. Proc Natl Acad Sci USA 1994; 91:9480-9484.
93. Barth H, Hoffmann I, Kinzel V. Radiation with 1 Gy prevents the activation of the mitotic inducers
mitosis-promoting factor (MPF) and Cdc25-C in HeLa cells. Cancer Res 1996; 56:2268-2272.
94. Gabrielli BG, Clark JM, McCormack AK, Ellem KAO. Ultraviolet light-induced G2 phase cell cycle
checkpoint blocks Cdc25-dependent progression into mitosis. Oncogene 1997; 15:749-758.
95. Matsuoka S, Huang M, Elledge SJ. Linkage of ATM to cell cycle regulation by the Chk2 protein
kinase. Science 1998; 282:1893-1897.
96. Ahn JY, Schwarz JK, Piwnica-Worms H, Canman CE. Threonine 68 phosphorylation by ataxia
telangiectasia mutated is required for efficient activation of Chk2 in response to ionizing radiation.
Cancer Res 2000; 60:5934-5936.
97. Melchionna R, Chen XB, Blasina A, McGowan CH. Threonine 68 is required for radiation-induced
phosphorylation and activation of Cds1. Nat Cell Biol 2000; 2:762-765.
98. Matsuoka S, Rotman G, Ogawa A et al. Ataxia telangiectasia-mutated phosphorylates Chk2 in vivo
and in vitro. Proc Natl Acad Sci USA 2000; 97:10389-10394.
99. Liu QH, Guntuku S, Cui XS et al. Chk1 is an essential kinase that is regulated by ATR and
required for the G2/M DNA damage checkpoint. Genes Dev 2000; 14:1448-1459.
100. Guo Z, Kumagai A, Wang SX, Dunphy WG. Requirement for ATR in phosphorylation of Chk1
and cell cycle regulation in response to DNA replication blocks and UV-damaged DNA in xenopus
egg extracts. Genes Dev 2000; 14:2745-2756.
G2 Checkpoint and Anticancer Therapy 173
101. Sanchez Y, Wong S, Thoma RS et al. Conservation of the Chk1 checkpoint pathway in mammals:
Linkage of DNA damage to CDK regulation through Cdc25C. Science 1997; 277:1497-1501.
102. Furnari B, Rhind N, Russell P. Cdc25C mitotic inducer targeted by Chk1 DNA damage checkpoint
kinase. Science 1997; 277:1495-1497.
103. Peng C-Y, Graves PR, Thoma RS et al. Mitotic and G2 checkpoint control: Regulation of
14-3-3 protein binding by phosphorylation of Cdc25C on Serine-216. Science 1997;
277:1501-1505.
104. Peng CY, Graves PR, Thoma RS et al. Mitotic and G2 checkpoint control: Regulation of 14-3-3
protein binding by phosphorylation of Cdc25C on serine-216. Science 1997; 277:1501-1505.
105. Lopez-Girona A, Furnari B, Mondesert O, Russell P. Nuclear localization of Cdc25C is regulated
by DNA damage and a 14-3-3 protein. Nature 1999; 397:172-175.
106. Walworth N, Davey S, Beach D. Fisssion yeast Chk1 protein kinase links the rad checkpoint
pathway to cdc2. Nature 1993; 363:368.
107. Jimenez G, Yucel J, Rowley R, Subramani S. The rad3+ gene of Schizosaccharomyces pombe is
involved in multiple checkpoint functions and in DNA repair. Proc Natl Acad Sci USA 1992;
89:4952-4956.
108. Seaton BL, Yucel J, Sunnerhagen P, Subramani S. Isolation and characterization of the
Schizosaccharomyces pombe rad3 gene, involved in DNA damage and DNA synthesis checkpoints.
Gene 1992;119:83-89.
109. Weinert TA. Dual cell cycle checkpoints sensitive to chromosome replication and DNA damage in
the budding yeast Saccharomyces cerevisiae. Radiat Res 1992;132:141-143.
110. Wright JA, Keegan KS, Herendeen DR et al. Protein kinase mutants of human ATR increase
sensitivity to UV and ionizing radiation and abrogate cell cycle checkpoint control. Proc Natl Acad
Sci USA 1998; 95:7445-7450.
111. Brown EJ, Baltimore D. ATR disruption leads to chromosomal fragmentation and early embryonic
lethality. Genes Dev 2000; 14:397-402.
112. de Klein A, Muijtjens M, van Os R et al. Targeted disruption of the cell-cycle checkpoint gene
ATR leads to early embryonic lethality in mice. Curr Biol 2000; 10:479-482.
113. Naito T, Matsuura A, Ishikawa F. Circular chromosome formation in a fission yeast mutant defective
in two ATM homologues. Nat Genet 1998; 20:203-206.
114. Morrow DM, Morrow M, Tagle DA et al. TEL1, an S. cerevisiae homolog of the human gene
mutated in ataxia telangiectasia, is functionally related to the yeast checkpoint gene MEC1. Cell
1995; 82:831-840.
115. Matsuura A, Naito T, Ishikawa F. Genetic control of telomere integrity in Schizosaccharomyces
pombe: rad3(+) and tel1(+) are parts of two regulatory networks independent of the downstream
kinases Chk1(+) and cds(+). Genetics 1999; 152:1501-1512.
116. Scott D, Spreadborough AR, Roberts SA. Radiation-induced G2 delay and spontaneous chromosome
aberrations in ataxia-telangiectasia homozygotes and heterozygotes. Int J Radiat Biol 1994;
66:157-163.
117. Barlow C, Hirostune S, Paylor R et al. ATM-deficient mice: a paradigm of ataxia telangiectasia.
Cell 1996; 86:159-171.
118. Xu Y, Ashley T, Brainerd EE et al. Targeted disruption of ATM leads to growth retardation,
chromosomal fragmentation during meiosis, immune defects, and thymic lymphoma. Genes Dev
1996; 10:2411-2422.
119. Taylor AM, Harnden DG, Arlett CF et al. Ataxia telangiectasia: A human mutation with abnormal
radiation sensitivity. Nature 1975; 258:427-429.
120. Khanna KK. Cancer risk and the ATM gene: A continuing debate. J Natl Cancer Inst 2000;
92:795-802.
121. Lavin MF, Shiloh Y. Ataxia-telangiectasia: A multifaceted genetic disorder associated with defective
signal transduction. Curr Opin Immunol 1996; 8:459-464.
122. Liu Y, Vidanes G, Lin YC et al. Characterization of a Saccharomyces cerevisiae homologue of
Schizosaccharomyces pombe Chk1 involved in DNA-damage-induced M-phase arrest. Mol Gen Genet
2000; 262:1132-1146.
123. Takai H, Tominaga K, Motoyama N et al. Aberrant cell cycle checkpoint function and early embryonic
death in Chk1-/- mice. Genes Dev 2000; 14:1439-1447.
124. Rhind N, Russell P. The Schizosaccharomyces pombe S-phase checkpoint differentiates between different
types of DNA damage. Genetics 1998; 149:1729-1737.
125. Weinert TA, Kiser GL, Hartwell LH. Mitotic checkpoint genes in budding yeast and the dependence
of mitosis on DNA replication and repair. Genes Dev 1994; 8:652-665.
126. Allen JB, Zhou Z, Siede W et al. The SAD1/RAD53 protein kinase Controls multiple checkpoints
and DNA damage-induced transcription in yeast. Genes Dev 1994; 8:2401-2415.
174 Cell Cycle Checkpoints and Cancer
127. Hirao A, Kong YY, Matsuoka S et al. DNA damage-induced activation of p53 by the checkpoint
kinase Chk2. Science 2000;287:1824-1827.
128. Bell DW, Varley JM, Szydlo TE et al. Heterozygous germ line hCHK2 mutations in Li-Fraumeni
syndrome. Science 1999; 286:2528-2531.
129. Peng CY, Graves PR, Ogg S et al. C-TAK1 protein kinase phosphorylates human Cdc25C on
serine 216 and promotes 14-3-3 protein binding. Cell Growth Differ 1998; 9:197-208.
130. Li B, Ouyang B, Pan H et al. Prk, a cytokine-inducible human protein serine/threonine kinase
whose expression appears to be down-regulated in lung carcinomas. J Biol Chem 1996;
271:19402-19408.
131. Ouyang B, Pan H, Lu L et al. Human Prk is a conserved protein serine/threonine kinase involved
in regulating M phase functions. J Biol Chem 1997; 272:28646-28651.
132. Ouyang B, Li W, Pan H et al. The physical association and phosphorylation of Cdc25C protein
phosphatase by Prk. Oncogene 1999; 18:6029-6036.
133. Sanchez Y, Banchet J, Wang H et al. Control of the DNA damage checkpoint by Chk1 and rad53
protein kinases through discrinct mechanisms. Science 1999; 286: 1166-11671:
134. Bernstein HS, Couhglin SR. A mammalian homolog of the fission yeast Cdc5 regulates G2 progression
and mitotic entry. J Biol Chem 1998; 20:4666-4671.
135. Cheng L, Hunke L, Hardy CFJ. Cell cycle regulation of the Saccharomyces cerevisiae polo-like
kinase cdc5p. Mol Cell Biol 1998; 18:7360-7370.
136. Cogswell JP, Brown CE, Bisi JE, Neill SD. Dominant-negative polo-like kinase 1 induces mitotic
catastrophe independent of Cdc25C function. Cell Growth Differ 2000; 11:615-623.
137. Lu KP, Hanes SD, Hunter T. A human peptidyl-prolyl isomerase essential for regulation of mitosis.
Nature 1996; 380:544-547.
138. Shen M, Stukenberg PT, Kirschner MW, Lu KP. The essential mitotic peptidyl-prolyl isomerase
Pin1 binds to and regulates mitosis-specific phosphoproteins. Genes Dev 1998; 12:706-720.
139. Crenshaw DG, Yang J, Means AR, Kornbluth S. The mitotic peptidyl-prolyl isomerase, Pin1, interacts
with Cdc25 and Plx1. EMBO J 1998; 17:1315-1327.
140. Winkler KE, Swenson KI, Kornbluth S, Means AR. Requirement of the prolyl isomerase Pin1 for
the replication checkpoint. Science 2000; 287:1644-1647.
141. Rippmann JF, Hobbie S, Daiber C et al. Phosphorylation-dependent proline isomerization catalyzed
by Pin1 is essential for tumor cell survival and entry into mitosis. Cell Growth Differ 2000;
11:409-416.
142. Brown R, Hirst GL, Gallagher WM et al. hMLH1 expression and cellular responses of ovarian
tumour cells to treatment with cytotoxic anticancer agents. Oncogene 1997;15:45-52.
143. Davis TW, Wilson-Van Patten C, Meyers M et al. Defective expression of the DNA mismatch
repair protein, MLH1, alters G2-M cell cycle checkpoint arrest following ionizing radiation. Cancer
Res 1998; 58:767-778.
144. Ford HL, Kabingu EN, Bump EA et al. Abrogation of the G2 cell cycle checkpoint associated
with overexpression of HSIX1: A possible mechanism of breast carcinogenesis. Proc Natl Acad Sci
USA 1998; 95:12608-12613.
145. Wang X, McGowan CH, Zhao M et al. Involvement of the MKK-p38gamma cascade in gammaradiation-
induced cell cycle arrest. Mol Cell Biol 2000; 20:4543-4552.
146. Pillaire MJ, Nebreda AR, Darbon JM. Cisplatin and UV radiation induce activation of the stressactivated
protein kinase p38gamma in human melanoma cells. Biochem Biophys Res Commun
2000; 278:724-728.
147. Levedakou EN, Kaufmann WK, Alcorta DA et al. p21CIP1 is not required for the early G2 checkpoint
response to ionizing radiation. Cancer Res 1995; 55:2500-2502.
148. Kaufmann WK, Schwartz JL, Hurt JC et al. Inactivation of G2 checkpoint function and chromosomal
destabilization are linked in human fibroblasts expressing human papillomavirus type 16 E6.
Cell Growth Differ 1997; 8:1105-1114.
149. Paules RS, Levedakou EN,Wilson,S.J., Innes CL et al. Defective G2 checkpoint function in cells
from individuals with familial cancer syndromes. Cancer Res 1995; 55:1763-1773.
150. Bunz F, Dutriaux A, Lengauer C et al. Requirement for p53 and p21 to sustain G2 arrest after
DNA damage. Science 1998; 282:1497-1501.
151. Kastan MB, Onyekwere O, Sidransky D et al. Participation of p53 protein in the cellular response
to DNA damage. Cancer Res 1991; 51:6304-6311.
152. Little JB, Nagasawa H, Keng PC et al. Absence of radiation-induced G1 arrest in two closely related
human lymphoblast cell lines that differ in p53 status. J Biol Chem 1995; 270:11033-11036.
153. Vikhanskaya F, Erba E, D’Incalci M, Broggini M. Introduction of wild-type p53 in a human
ovarian cancer cell line not expressing endogenous p53. Nucleic Acids Res 1994; 22:1012-1017.
G2 Checkpoint and Anticancer Therapy 175
154. Stewart N, Hicks GG, Paraskevas F, Mowat M. Evidence for a second cell cycle block at G2/M by
p53. Oncogene 1995; 10:109-115.
155. Agarwal ML, Agarwal A, Taylor WR, Stark GR. p53 controls both the G2/M and the G1 cell cycle
checkpoints and mediates reversible growth arrest in human fibroblasts. Proc Natl Acad Sci USA
1995; 92:8493-8497.
156. Flatt PM, Tang LJ, Scatena CD, Szak ST, Pietenpol JA. p53 Regulation of G2 checkpoint is
retinoblastoma protein dependent. Mol Cell Biol 2000; 20:4210-4223.
157. Hermeking H, Lengauer C, Polyak K et al. 14-3-3s is a p53-regulated inhibitor of G2/M progression.
Mol Cell 1997; 1:3-11.
158. Chan TA, Hermeking H, Lengauer C et al. 14-3-3Sigma is required to prevent mitotic catastrophe
after DNA damage. Nature 1999; 401:616-620.
159. de Toledo S, Azzam E, Keng P et al. Regulation by ionizing radiation of CDC2, cyclin A, cyclin
B, thymidine kinase, topoisomerase IIa, and RAD51 expression in normal human diploid fibroblasts
is dependent on p53/p21. Cell Growth Differ 1998; 9:887-896.
160. Innocente SA, Abrahamson JLA, Cogswell JP, Lee JM. p53 regulates a G2 checkpoint through
cyclin B1. Proc Natl Acad Sci USA 1999; 96:2147-2152.
161. Badie C, Bourhis J, Sobczak-Thépot J et al. p53-dependent G2 arrest associated with a decrease in
cyclins A2 and B1 levels in a human carcinoma cell line. Br J Cancer 2000; 82:642-650.
162. Smits VAJ, Klompmaker R, Vallenius T et al. p21 inhibits Thr161 phosphorylation of cdc2 to
enforce the G2 DNA damage checkpoint. J Biol Chem 2000; 275:30638-30643.
163. Harper JW, Elledge SJ, Keyomarsi K et al. Inhibition of cyclin-dependent kinases by p21. Mol
Biol Cell 1995; 6:387-400.
164. Taylor WR, DePrimo SE, Agarwal A et al. Mechanisms of G2 arrest in response to overexpression
of p53. Mol Biol Cell 1999; 10:3607-3622.
165. Manni I, Mazzaro G, Gurtner A et al. NF-Y mediates the transcriptional inhibition of the cyclin
B1, cyclin B2, and CDC25C promoters upon induced G2 arrest. J Biol Chem 2000;
166. Taylor WR, Schonthal AH, Galante J, Stark GR. p130/E2F4 binds to and represses the cdc2
promoter in response to p53. J Biol Chem 2001; 276:1998-2006.
167. Park M, Chae HD, Yun J et al. Constitutive activation of cyclin B1-associated cdc2 kinase overrides
p53-mediated G2-M arrest. Cancer Res 2000; 60:542-545.
168. Helin K, Harlow E. The retinoblastoma protein as a transcriptional repressor. Trends Cell
Biol 1993;3:43-46.
169. Wang XW, Zhan QM, Coursen JD et al. GADD45 induction of a G2/M cell cycle checkpoint.
Proc Natl Acad Sci USA 1999; 96:3706-3711.
170. Zhan QM, Antinore MJ, Wang XW et al. Association with Cdc2 and inhibition of Cdc2/cyclin
B1 kinase activity by the p53-regulated protein Gadd45. Oncogene 1999; 18:2892-2900.
171. Ohki R, Nemoto J, Murasawa H et al. Reprimo, a new candidate mediator of the p53-mediated
cell cycle arrest at the G2 phase. J Biol Chem 2000; 275:22627-22630.
172. Ahn J, Murphy M, Kratowicz S et al. Down-regulation of the stathmin/Op18 and FKBP25 genes
following p53 induction. Oncogene 1999; 18:5954-5958.
173. Johnsen JI, Aurelio ON, Kwaja Z et al. p53-mediated negative regulation of stathmin/Op18 expression
is associated with G2/M cell-cycle arrest. Int J Cancer 2000;88:685-691.
174. Bieche I, Lachkar S, Becette V et al. Overexpression of the stathmin gene in a subset of human
breast cancer. Br J Cancer 1998; 78:701-709.
175. Curmi PA, Nogues C, Lachkar S et al. Overexpression of stathmin in breast carcinomas points out
to highly proliferative tumors. Br J Cancer 2000; 82:142-150.
176. Price DK, Ball JR, Bahrani-Mostafavi Z et al. The phosphoprotein Op18/stathmin is differentially
expressed in ovarian cancer. Cancer Invest 2000; 18:722-730.
177. Melhem RF, Zhu XX, Hailat N et al. Characterization of the gene for a proliferation-related phosphoprotein
(oncoprotein 18) expressed in high amounts in leukemia. J Biol Chem 1991;
266:17747-17753.
178. Roos G, Brattsand G, Landberg G et al. Expression of oncoprotein 18 in human leukemias and
lymphomas. Leukemia 1993; 7:1538-1546.
179. Marklund U, Larsson N, Gradin HM et al. Oncoprotein 18 is a phosphorylation-responsive regulator
of microtubule dynamics. EMBO J 1996; 5:5290-5298.
180. Gavet O, Ozon S, Manceau V et al. The stathmin phosphoprotein family: Intracellular localization
and effects on the microtubule network. J Cell Sci 1998; 111:3333-3346.
181. Larsson N, Marklund U, Gradin HM et al. Control of microtubule dynamics by oncoprotein 18:
Dissection of the regulatory role of multisite phosphorylation during mitosis. Mol Cell Biol 1997;
17:5530-5539.
176 Cell Cycle Checkpoints and Cancer
182. Larsson N, Melander H, Marklund U et al. G2/M transition requires multisite phosphoryation of
oncoprotein 18 by two distinct protein kinase systems. J Biol Chem 1995; 270:14175-14183.
183. Tibbetts RS, Brumbaugh KM, Williams JM et al. A role for ATR in the DNA damage-induced
phosphorylation of p53. Genes Dev 1999; 13:152-157.
184. Hall-Jackson CA, Cross DA, Morrice N, Smythe C. ATR is a caffeine-sensitive, DNA-activated
protein kinase with a substrate specificity distinct from DNA-PK. Oncogene 1999;18:6707-6713.
185. Canman CE, Lim DS, Cimprich KA et al. Activation of the ATM kinase by ionizing radiation and
phosphorylation of p53. Science 1998; 281:1677-1679.
186. Banin S, Moyal L, Shieh SY et al. Enhanced phosphorylation of p53 by ATM in response to DNA
damage. Science 1998; 281:1674-1677.
187. Canman CE, Lim DS. The role of ATM in DNA damage responses and cancer. Oncogene
1998; 17:3301-3308.
188. Shieh SY, Ahn J, Tamai K et al. The human homologs of checkpoint kinases Chk1 and
Cds1 (Chk2) phosphorylate p53 at multiple DNA damage-inducible sites. Genes Dev 2000;
14:289-300.
189. Chehab NH, Malikzay A, Appel M, Halazonetis TD. Chk2/hCds1 functions as a DNA damage
checkpoint in G1 by stabilizing p53. Genes Dev 2000; 14:278-288.
190. Chehab NH, Malikzay A, Stavridi ES, Halazonetis TD. Phosphorylation of Ser-20 mediates
stabilization of human p53 in response to DNA damage. Proc Natl Acad Sci USA 1999;
96:13777-13782.
191. Felsher DW, Zetterberg A, Zhu JY et al. Overexpression of MYC causes p53-dependent G2 arrest
of normal fibroblasts. Proc Natl Acad Sci USA 2000; 97:10544-10548.
192. Scheffner M, Werness BA, Huibregtse JM et al. The E6 oncoprotein encoded by human
papillomavirus types 16 and 18 promotes the degradation of p53. Cell 1990;63:1129-1136.
193. Thompson D, Belinsky G, Chang T et al. The human papillomavirus-16 E6 oncoprotein decreases
the vigilance of mitotic checkpoints. Oncogene 1997; 15:3025-3035.
194. Filatov L, Golubovskaya V, Hurt JC et al. Chromosomal instability is correlated with telomere
erosion and inactivation of G2 checkpoint function in human fibroblasts expressing human
papillomavirus type 16 E6 oncoprotein. Oncogene 1998; 16:1825-1838.
195. Passalaris TM, Benanti JA, Gewin L et al. The G2 checkpoint is maintained by redundant
pathways. Mol Cell Biol 1999; 19:5872-5881.
196. Song SY, Gulliver GA, Lambert PF. Human papillomavirus type 16 E6 and E7 oncogenes abrogate
radiation-induced DNA damage responses in vivo through p53-dependent and p53-independent pathways.
Proc Natl Acad Sci USA 1998; 95:2290-2295.
197. Bulavin DV, Tararova ND, Aksenov ND et al. Deregulation of p53/p21CIP1/Waf1 pathway contributes
to polyploidy and apoptosis of E1A+cHa-Ras transformed cells after gamma-irradiation. Oncogene
1999; 18:5611-5619.
198. Yageta M, Tsunoda H, Yamanaka T et al. The adenovirus E1A domains required for induction of
DNA rereplication in G2/M arrested cells coincide with those required for apoptosis. Oncogene
1999; 18:4767-4776.
199. Sanchez-Prieto R, Leonart M, Ramon y Cajal S. Lack of correlation between p53 protein level and
sensitivity of DNA-damaging agents in keratinocytes carrying E1A mutants. Oncogene 1995; 11:675-382.
200. Sanchez-Prieto R, Quintanilla M, Cano A et al. Carcinoma cell lines become sensitive to DNAdamaging
agents by the expression of the adenovirus E1A gene. Oncogene 1996; 13:1083-1092.
201. Tanimoto A, Chen H, Kao CY et al. Transactivation of the human cdc2 promoter by adenovirus
E1A in cycling cells is mediated by induction of a 110-kD CCAAT-box-binding factor. Oncogene
1998; 17:3103-3114.
202. Skladanowski A, Larsen AK. Expression of wild-type p53 increases etoposide cytotoxicity in M1
myeloid leukemia cells by facilitated G2 to M transition: implications for gene therapy. Cancer Res
1997; 57:818-823.
203. Guillouf C, Rosselli F, Krishnaraju K et al. p53 involvement in control of G2 exit of the cell cycle:
Role in DNA damage-induced apoptosis. Oncogene 1995; 10:2263-2270.
204. Leach SD, Scatena CD, Keefer CJ et al. Negative regulation of Wee1 expression and Cdc2
phosphorylation during p53-mediated growth arrest and apoptosis. Cancer Res 1998; 58:3231-3236.
205. Yao SL, Akhtar AJ, McKenna KA et al. Selective radiosensitization of p53-deficient cells by
caffeine-mediated activation of p34cdc2 kinase. Nature Med 1996; 2:1140-1143.
206. Powell SN, DeFrank JS, Connell P et al. Differential sensitivity of p53(-) and p53(+) cells to
caffeine-induced radiosensitization and override of G2 delay. Cancer Res 1995; 55:1643-1648.
207. Russell KJ, Wiens LW, Demers W et al. Abrogation of the G2 checkpoint results in differential
radiosensitization of G1 checkpoint-deficient and G1-checkpoint competent cells. Cancer Res 1995;
55:1639-1642.
G2 Checkpoint and Anticancer Therapy 177
208. Sarkaria JN, Busby EC, Tibbetts RS et al. Inhibition of ATM and ATR kinase activities by the
radiosensitizing agent, caffeine. Cancer Res 1999; 59:4375-4382.
209. Zhou BB, Chaturvedi P, Spring K et al. Caffeine abolishes the mammalian G2/M DNA damage checkpoint
by inhibiting ataxia-telangiectasia-mutated kinase activity. J Biol Chem 2000; 275:10342-10348.
210. Kumagai A, Guo ZJ, Emami KH et al. The Xenopus Chk1 protein kinase mediates a caffeinesensitive
pathway of checkpoint control in cell-free extracts. J Cell Biol 1998; 142:1559-1569.
211. Schiano MA, Sevin BU, Perras J et al. In vitro enhancement of cis-platinum antitumor activity by
caffeine and pentoxifylline in a human ovarian cell line. Gynecol Oncol 1991; 43:37-45.
212. Ohsaki Y, Ishida S, Fujikane T, Kikuchi K. Pentoxifylline potentiates the antitumor effect of cisplatin
and etoposide on human lung cancer cell lines. Oncology 1996; 53:327-333.
213. Theron T, Binder A, Verheye-Dua F, Bohm L. The role of G2-block abrogation, DNA doublestrand
break repair and apoptosis in the radiosensitization of melanoma and squamous cell carcinoma
cell lines by pentoxifylline. Int J Radiat Biol 2000; 76:1197-1208.
214. Fan S, Smith ML, Rivet DJ et al. Disruption of p53 function sensitizes breast cancer MCF-7 cells
to cisplatin and pentoxifylline. Cancer Res 1995; 55:1649-1654.
215. Teicher BA, Holden SA, Herman TS et al. Efficacy of pentoxifylline as a modulator of alkylating
agent activity in vitro and in vivo. Anticancer Res 1991; 11:1555-1560.
216. Wong JS, Ara G, Keyes SR et al. Lisofylline as a modifier of radiation therapy. Oncol Res 1996; 8:513-518.
217. Fingert HJ, Pu AT, Chen ZY et al. In vivo and in vitro enhanced antitumor effects by pentoxifylline
in human cancer cells treated with thiotepa. Cancer Res 1988; 48:4375-4381.
218. Kwon H, Kim S, Chung W et al. Effect of pentoxifylline on radiation response of non-small cell
lung cancer: a phase III randomized multicenter trial. Radiother Oncol 2000; 56:175-179.
219. Mannel RS, Blessing JA, Boike G. Cisplatin and pentoxifylline in advanced or recurrent
squamous cell carcinoma of the cervix: A phase II trial of the gynecologic oncology group.
Gynecol Oncol 2000; 79:64-66.
220. Russell KJ, Wiens LW, Demers GW et al. Preferential radiosensitization of G1 checkpoint-deficient
cells by methylxanthines. Int J Radiat Oncol Biol Phys 1996; 36:1099-1106.
221. Husain A, Rosales N, Schwartz GK, Spriggs DR. Lisofylline sensitizes p53 mutant human ovarian carcinoma
cells to the cytotoxic effects of cis-diamminedichloroplatinum (II). Gynecol Oncol 1998; 70:17-22.
222. Tam SW, Schlegel R. Staurosporine overrides checkpoints for mitotic onset in BHK cells. Cell
Growth Differ 1992; 3:811-817.
223. Courage C, Snowden R, Gescher A. Differential effects of staurosporine analogues on cell cycle,
growth, and viability in A549 cells. Br J Cancer 1996; 74:1199-1205.
224. Wang Q, Fan S, Eastman A et al. UCN-01: A potent abrogator of G2 checkpoint function in
cancer cells with disrupted p53. J Natl Cancer Inst 1996; 88:956-965.
225. Bunch RT, Eastman A. Enhancement of cisplatin-induced cytotoxicity by 7-hydroxystaurosporine
(UCN-01), a new G2-checkpoint inhibitor. Clin Cancer Res 1996; 2:791-797.
226. Akinga S, Gomi K, Morimoto M et al. Antitumor activity of UCN-01, a selective inhibitor of
protein kinase C, in murine and human tumor models. Cancer Res 1991; 51:4888-4892.
227. Akinga S, Nomura K, Gomi K, Okabe M. Enhancement of antitumor activity of mitomycin C in
vitro and in vivo by UCN-01, a selective inhibitor of protein kinase C. Cancer Chemother Pharmacol
1993; 32:183-189.
228. Senderowicz AM, Sausville EA. Preclinical and clinical development of cyclin-dependent kinase
modulators. J Natl Cancer Inst 2000; 92:376-387.
229. Yu L, Orlandi L, Wang P et al. UCN-01 abrogates G2 arrrest through a cdc2-dependent pathway
that is associated with inactivation of the Wee1Hu kinase and activation of the Cdc25C phosphatase.
J Biol Chem 1998; 273:33455-33464.
230. Graves PR, Yu LJ, Schwarz JK et al. The Chk1 protein kinase and the Cdc25C regulatory pathways
are targets of the anticancer agent UCN-01. J Biol Chem 2000; 275:5600-5605.
231. Busby EC, Leistritz DF, Abraham RT et al. The radiosensitizing agent 7-hydroxystaurosporine
(UCN-01) inhibits the DNA damage checkpoint kinase hChk1. Cancer Res 2000; 60:2108-2112.
232. Suganuma M, Kawabe T, Hori H et al. Sensitization of cancer cells to DNA damage-induced cell
death by specific cell cycle G2 checkpoint abrogation. Cancer Res 1999; 59:5887-5891.
233. Jackson JR, Gilmartin A, Imburgia C et al. An indolocarbazole inhibitor of human checkpoint
kinase (Chk1) abrogates cell cycle arrest caused by DNA damage. Cancer Res 2000; 60:566-572.
234. Ling Q, Burgess A, Fairlie DP et al. Histone deacetylase inhibitors trigger a G2 checkpoint in
normal cells that is defective in tumor cells. Mol Biol Cell 2000; 11:2069-2083.
235. Grunstein M. Histone acetylation in chromatin structure and transcription. Nature 1997; 389:349-352.
236. Archer SY, Meng SF, Shei A, Hodin RA. p21WAF1 is required for butyrate-mediated growth inhibition
of human colon cancer cells. Proc Natl Acad Sci USA 1998;95:6791-6796.
178 Cell Cycle Checkpoints and Cancer
237. Richon VM, Sandhoff TW, Rifkind RA, Marks PA. Histone deacetylase inhibitor selectively induces
p21WAF1 expression and gene-associated histone acetylation. Proc Natl Acad Sci USA
2000;97:10014-10019.
238. Kim YB, Ki SW, Yoshida M, Horinouchi S. Mechanism of cell cycle arrest caused by histone
dacetylase inhibitors in human carcinoma cells. J Antibiot 2000; 53:1191-1200.
239. Nakajima H, Kim YB, Terano H et al. FR901228, a potent antitumor antibiotic, is a novel histone
deacetylase inhibitor. Exp Cell Res 1998; 241:126-133.
240. Saito A, Yamashita T, Mariko Y et al. A synthetic inhibitor of histone deacetylase, MS-27-275,
with marked in vivo antitumor activity against human tumors. Proc Natl Acad Sci USA 1999;
96:4592-4597.
241. Warrell RP, He LZ, Richon V et al. Therapeutic targeting of transcription in acute promyelocytic
leukemia by use of an inhibitor of histone. J Natl Cancer Inst 1998; 90:1621-1625.
242. Yu D, Jing T, Liu B et al. Overexpression of ERbB2 blocks Taxol-induced apoptosis by upregulation
of p21CIP1, which inhibits p34.Cdc2 kinase Mol Cell 1998; 2:581-591.
243. Donaldson KL, Goolby G, Kiner PA, Wahl AF. Activation of p34cdc2 coincindent with Taxolinduced
apoptosis. Cell Growth Differ 1994; 5:1041-1050.
244. Baselga J, Norton L, Albanell J et al. Recombinant humanized anti-HER2 antibody (Herceptin)
enhances the antitumor activity of Paclitaxel and doxorubicin against HER2/neu overexpressing
human breast xenografts. Cancer Res 1998; 58:2825-2831.
245. Weinstein JN, Myers TG, O’Connor PM et al. An information-intensive approach to the molecular
pharmacology of cancer. Science 1997; 275:343-349.
246. Amundson SA, Myers TG, Scudiero D et al. An informatics approach identifying markers of
chemosensitivity in human cancer cell lines. Cancer Res 2000; 60:6101-6110.
247. O’Connor PM, Jackman J, Bae I et al. Characterization of the p53 tumor suppressor pathway in
cell lines of the National Cancer Institute anticancer drug screen and correlations with the growthinhibitory
potency of 123 anticancer agents. Cancer Res 1997; 57:4285-4300.
248. Scherf U, Ross DT, Waltham M et al. A gene expression database for the molecular pharmacology
of cancer. Nat Genet 2000; 24:236-244.
249. Roberge M, Berlinck RGS, Xu L et al. High-throughput assay for G2 checkpoint inhibitors and
identification of the structurally novel compound isogranulatimide. Cancer Res 1998; 58:5701-5706.
250. Perego P, Jimenez GS, Gatti L et al. Yeast mutants as a model system for identification of determinants
of chemosensitivity. Pharm Rev 2000; 52:477-491.
251. Hartwell LH, Szankasi P, Roberts CJ et al. Integrating genetic approaches into the discovery of
anticancer drugs. Science 1997; 278:1064-1068.
252. Norman TC, Smith DL, Sorger PK et al. Genetic selection of peptide inhibitors of biological
pathways. Science 1999; 285:591-595.
253. Spellman PT, Sherlock G, Zhang MQ et al. Comprehensive identification of cell cycle-regulated genes of
the yeast Saccharomyces cerevesiae by microarray hybridization. Mol Biol Cell 1998;9:3273-3297.
254. Jelinsky SA, Samson LD. Global response of Saccharomyces cerevisiae to an alkylating agent. Proc
Natl Acad Sci USA 1999; 96:1486-1491.
255. Jelinsky SA, Estep P, Church GM, Samson LD. Regulatory networks revealed by transcriptional
profiling of damaged Saccaromyces cerevisiae cells: rpn4 links base excision repair with proteasomes.
Mol Biol Cell 2000; 20:8157-8167.
256. Hughes TR, Marton MJ, Jones AR et al. Functional discovery via a compendium of expression
profiles. Cell 2000; 102:109-126.
257. Hughes TR, Roberts CJ, Dai H et al. Widespread aneuploidy revealed by DNA microarray expression
profiling. Nat Genet 2000; 25:33
P53, Apoptosis and Cancer Therapy 179
CHAPTER 11
Cell Cycle Checkpoints and Cancer, edited by Mikhail V. Blagosklonny. ©2001 Eurekah.com.
p53, Apoptosis and Cancer Therapy
Rosandra Kaplan and David E. Fisher
Abstract
Several decades of genetic and molecular study have revealed enormous insights into the
mechanistic underpinnings of cancer. From the identification of dominantly acting
oncogenes to the signaling pathways which modulate the cell cycle, our understanding of
the machinery of cell cycle progression as well as the regulatory circuits which control it have
never been so detailed. However the translation of these discoveries into improved therapeutic
approaches has been slow. The more recent appreciation of the pivotal role for cell survival
pathways in both the genesis of malignancies as well as their response to treatment has created
a burst of excitement for the prospect that the unraveling of these pathways stands to more
directly impact on therapeutic strategies for cancer. This chapter focuses on one major regulator
of cell survival in human cancer: the p53 tumor suppressor gene product. Abundant evidence
suggests that the action(s) of this one gene serves to prevent formation of many human
malignancies and likely mediates the successful treatment responses in those few tumors in
which current/traditional chemotherapy produces durable cures. Through the study of p53’s
central roles in the cancer cell, molecular oncology has become inextricably linked to the quest
for novel approaches to the therapy of cancer.
Introduction
The challenge in cancer therapy focuses fundamentally on the paucity of therapeutically
exploitable differences between cancer cells and normal cells. This small margin is the therapeutic
index, and it is the final endpoint for successful cancer therapy. Aside from the fact that cancer
cells are so similar to normal cells, data accumulated over the past decade have suggested that
most cancer cells are even more resistant than normal cells to a variety of death triggers.
Dismantling the machinery for cell death confers an obvious survival advantage, potentially
explaining a fundamental feature of neoplastic transformation. For this reason, cancer is
now recognized to be a disease involving both uncontrolled proliferation and a lack of
normal cell death.
There is a growing body of evidence that anticancer therapies kill not exclusively through
the toxic effects of disrupting cellular metabolism, but also by triggering pathways that lead to
the cell’s suicide, also known as apoptosis. In both animal models and human cancers, cells
with intact apoptotic death machinery are more susceptible to cancer therapies than those with
disrupted apoptosis pathways. p53, the gene most commonly mutated in human cancer, is a
major regulator of apoptosis. Among the few curable human cancers, even in advanced stages,
are pediATRic acute lymphoblastic leukemia, testicular cancer, and Wilms tumor, which all
show a relative paucity of p53 aberrations and remarkably intact apoptotic death machinery. In
these responsive cancers, apoptosis is likely a central mechanism in the induction of remission.
Apoptosis is genetically encoded, evolutionarily conserved, and is intrinsically linked to
mechanisms of cell growth and differentiation. The first genetic descriptions of apoptosis came
180 Cell Cycle Checkpoints and Cancer
from studies of death mutants in nematodes.1,2 In these organisms three genes were discovered
that are involved in cell survival and death decisions. One of these is homologous to mammalian
caspases—a family of cysteine proteases that are essential to apoptosis. Another gene is
homologous to mammalian Apaf-1, which participates in formation of a complex which relays
death signals to the caspases. The third major death gene is homologous to the Bcl-2 family,
members of which have either pro-apoptotic or anti-apoptotic activity. The identification of these
genes initiated a research burst which has shed enormous light on our understanding of mechanisms
of cell death both in normal development and under pathological conditions such as cancer.
Although apoptosis can involve a variety of diverse triggers and cellular intermediates,
there are certain common features as well (Fig. 1). Apoptosis involves an initiator trigger such
as DNA damage, osmotic stress or cues from the extracellular environment. Through either of
several pathways, signaling events emanating from the trigger eventually lead to activation of
upstream initiator caspases which, in turn, activate downstream, “executioner” caspases. These
downstream caspases, such as caspases 3 and 7 cleave numerous cellular targets, resulting in
profound changes in the protein, lipid, and DNA compartments. Regulation of apoptosis
occurs at numerous levels within a cell. Of particular importance is the modulation of initial
caspase activation by the triggering pathway. Many survival factors as well as death factors
likely operate at this level, as described below.
Virtually all apoptotic triggers appear to employ one of two main pathways (Figs. 2 and
3). One pathway initiates from death receptors which may lead quite directly to the activation
of caspases, or in certain cases may utilize mitochondria as intermediates. The other is a pathway
that appears to depend upon release of key mitochondrial factors in order to activate
cytosolic caspases in response to a large number of stress-like triggers.
Caspases exist in normal cells as inactive enzymes called pro-caspases, analogous to
zymogens involved in the regulation of blood clotting. When activated by removal of the
pro-domain, these proteases cleave proteins at aspartic acid residues contained within tetrapeptide
recognition motifs. There are two general types of caspases: initiators, also known as upstream
caspases, and downstream, or executioner, caspases. Caspases -8, -9, -10 are upstream caspases
and are activated via auto-proteolysis upon recruitment into complexes in which high local
concentrations are presumed to permit weak proteolytic activity of the pro-caspase to cleave a
neighboring pro-caspase, thereby setting off an amplifying cascade. Once these upstream caspases
are activated, they then activate the executioner caspases such as caspase -3, -6 and -7, which
cleave numerous proteins throughout the cell and lead to membrane changes and DNA cleavage
events which are pathognomonic of apoptosis.
There are important negative regulators of apoptosis that can act on the caspase cascade at
several points. These inhibitors of apoptosis (IAPs) include c-IAP-1, c-IAP-2, XIAP, and survivin,
and are thought to provide a safeguard mechanism which serves as an endogenous threshold
regulator to temper minimal activation of the cascade. The overexpression of IAPs typically
renders the cell resistant to a wide variety of apoptotic stimuli. It has been suggested that IAPs
act by directly interfering with the catalytic activity or activation of certain caspases.3-5 Cellular
levels of IAPs may thus determine the difference in sensitivities of cells to apoptosis-inducing
stimuli. The recent identification of Smac/Diablo as a mitochondrially released IAP antagonist
has given rise to the concept that regulation of IAPs may represent a general means of modulating
the apoptotic response.6,7
Fas is a death receptor which triggers an apoptotic pathway involved in creating tolerance
to self in B and T cells by inducing death of autoreactive clones. Fas ligand binds to the Fas
receptor causing oligomerization and conformational changes which lead to recruitment of the
Fas-associated death-domain-containing molecule, FADD (Fig. 2). This in turn leads to binding
of pro-caspase-8 oligomers to FADD and caspase-8 auto-activation8 presumably through
auto-cleavage due to the high local concentrations of both enzyme and substrate. Similar death
receptor events are thought to occur for the Tumor Necrosis Factor (TNF) pathway and other
related receptor family members.
P53, Apoptosis and Cancer Therapy 181
The mitochondrial pathway of apoptosis (Fig. 3) involves a variety of signaling events, not
entirely understood at present, which trigger release of various mitochondrial contents including
cytochrome C into the cytosol. Cytochrome C, once released from the mitochondria, acts
together with Apaf-1 and procaspase-9 to activate downstream executioner caspases-3, -6, and
-79. A major regulatory step in this pathway is release of cytochrome C10as well as Smac/
Diablo.6,7 Antiapoptotic members of the Bcl-2 family such as Bcl-2 or Bcl-xL when overexpressed
can prevent cytochrome C release.11,12 The mitochondrial pathway is also strongly regulated
by pro-apoptotic members of the Bcl-2 family, such as BAX and Bad. These proteins may
dimerize with the anti-apoptotic Bcl-2 family members as well as form channels in the
mitochondrial membrane or alter the activity of the existing channels allowing for cytochrome
C release.
There is immense interconnectedness among these apoptotic pathways. p53 has been reported
to induce apoptosis through several different points within these apoptotic pathways. p53
may transcriptionally activate genes that are essential to growth arrest and DNA damage such as
GADD45, PA26, IGFBP-3, SIAH-1, and 14-3-3. It can also activate the pro-apoptotic Bcl-2
family member BAX.13 In addition, separate experimental evidence has suggested that p53 might
mediate death without transcriptional activity.14-17 It may alter trafficking of Fas to the cell surface
with activation of the receptor leading to activation of the caspase cascade.18 p53 may also activate
XPB and XPD DNA helicases which may, through an incompletely understood pathway, trigger
caspase activation.19 The mechanistic connection between p53 and caspase activation has remained
incompletely understood, but remains of pivotal importance because of p53’s apparently major
role in modulating the decision to live or die in human cancer cells.
Fig. 1. Blueprint of the apoptosis pathway. A wide assortment of triggering events activate signaling cascades
which direct the modification of upstream (or initiator) caspases, especially caspases 8 and 9, from their
zymogens to the active proteases. In turn, these activated caspases cleave and activate downstream caspases
such as caspase 3 and 7, which cleave numerous cellular targets known as death substrates.
182 Cell Cycle Checkpoints and Cancer
p53’s Emergence as a Key Death Regulator
p53 was discovered in 1979 when it was found as a -53kd cellular protein associated with
the tumor (T) antigen present in SV40 transformed cells.20,21 The protein was found to vary
somewhat in size, but retain homology in hamster, monkey and human cells. Its presence in
murine embryonal cell lines as well as SV-40 infected or transformed cells led to the conclusion
that the protein is encoded by the host genome. It was believed that p53 played a role in
modulation of the transformed state.
In the 1980s p53 was thought to be a dominant nuclear oncogene.22-26 Evidence for
dominant transforming activities of p53 resulted from the fact that the cDNAs and p53
genomic clones used in previous studies possessed dominant negative mutations.27-29 p53 was
definitively demonstrated to be a tumor suppressor gene in 1989 as evidence of loss of wildtype
p53 expression in many tumor types became apparent.30-34 Wildtype p53 showed the ability to
produce a marked reduction in the number of transformed foci of rat embryo fibroblasts.35
The p53 gene was localized to chromosome 17 band p13.36 p53 germline mutations were
identified in the rare autosomal dominant cancer predisposition condition known as
Li- Fraumeni syndrome.37 Li-Fraumeni syndrome is characterized by early onset tumors
including breast, pancreatic, adrenocortical, and prostate carcinomas, soft tissue sarcomas, brain
tumors, osteosarcoma, and leukemia.
Over the past decade p53’s molecular functions began to be elucidated. p53 protein was
discovered to be a transcription factor that enhances transcription of many genes thought to
integrate the cellular responses to stress. Protein levels are increased in response to double
stranded breaks in DNA and the presence of DNA repair intermediates after ultraviolet radia-
Fig. 2. The death receptor pathway of apoptosis. Shown here is a schematic of the pathway initiated by the
Fas ligand (FasL). Similar events are thought to occur for the other members of the TNF receptor family
of death receptors. In this pathway FasL induces trimerization of the Fas receptor, thereby recruiting FADD
which binds pro-caspase 8. High local concentrations of pro-caspase 8 oligomers result in auto-proteolysis
and activation. Active caspase 8 may either directly activate pro-caspase 3 or alternatively cleave Bid,
resulting in stimulation of the mitochondrial apoptosis pathway.
P53, Apoptosis and Cancer Therapy 183
tion exposure or chemical damage to DNA.38-42 Increased p53 activity in damaged cells can
lead to either cell cycle arrest or apoptosis.
The original observation which linked p53 protein to induction of the apoptotic response
involved an elegant experiment using temperature-sensitive mutant p53. In this study, the
ts-p53 gene was expressed in myeloid leukemia cells. At the permissive temperature, p53 triggered
massive apoptosis in the tumor cell population.43 Subsequent studies demonstrated that
p53 wildtype mouse thymocytes undergo apoptosis in response to ionizing radiation whereas
thymocytes from p53 -/- mice do not undergo apoptosis when similarly irradiated.44,45 The
ability of p53 to trigger apoptosis in response to DNA damage was further highlighted by the
findings that multiple chemotherapeutic agents triggered p53-dependent apoptosis in murine
oncogene-transformed fibroblasts both in vitro and in a mouse sarcoma model.46,47
Clinical Aspects of p53
Functional loss of p53 appears to be a very frequent, and possibly essential, prerequisite
during tumorigenesis. p53 mutations are found in over 50% of all human cancers. There are
currently over 2,000 literature reports documenting p53 mutations in diverse human tumor
types48 with p53 mutation being the most frequent genetic event in human cancer demonstrated
to date.49 The most common mutations are missense point mutations located in four of
the five evolutionarily conserved domains between amino acids 120 and 300.50 p53 mutations
have been found in all major histologic groups, including cancers of the colon (60%), stomach
(60%), breast (20%), lung (70%), brain (40%), and esophagus (60%).51 Mutations have been
analyzed by several different methods including DNA analytical methods, immunohistochemical
analyses (which show accumulation of mutant p53), and functional analyses which can be used
to assess transactivating activity.52
There has been a significant literature addressing clinical correlations between p53 and
prognosis. Some studies have found that p53 mutations in tumor cells have been associated
with a worse prognosis.53-55 Those malignancies with the best clinical outcomes such as testis
cancer, pediATRic acute lymphoblastic leukemia, and Wilms tumor have a paucity of p53
mutations and are quite responsive to therapy.56-58 When these tumors relapse, ineffectiveness
of therapy has been correlated with the acquisition of p53 mutations.59,60
Complicating these analyses, there are malignancies with wildtype p53 that exhibit poor
prognosis. Initially this finding gave pause to the idea that p53 aberrations played a significant
role in therapeutic responsiveness. However these examples have proven insightful in suggesting
that p53 function, rather than gene mutation per se, is the key measure of its role in the
therapeutic response. For example in cervical and anogenital cancers, p53 has been found to be
genetically wildtype, but the protein is rapidly degraded via a mechanism triggered by the E6
protein encoded by human papillomovirus.61-63 Similarly, p53 is both functionally inhibited
and degraded by the action of the Mdm2 oncoprotein. Mdm2 amplification is seen in a variety
of poor prognosis tumors that retain wildtype p53.64-67 Multiple recent studies show that p14
ARF (the gene encoded by the alternative reading frame at the p16/Ink4a locus) regulates p53
interactions with Mdm2.68,69 The genetic locus for this protein is frequently mutated, deleted,
or hypermethylated in a wide variety of cancers. Another example in which wildtype p53 is
present but lacks functionality is seen in neuroblastoma and certain breast carcinomas in which
p53 protein is sequestered in the cytoplasm, preventing proper nuclear localization.70-72
p53 Actions: Arrest vs. Death
Far beyond its early suspected role in viral transformation, p53 has subsequently been
termed “the guardian of the genome” and “the gatekeeper for cellular growth and division”
because of its vital function in tumor suppression and growth regulation.60,73 Normally p53 is
present at low concentration with a short half-life. However, in response to multiple signals
including DNA damage, oncogene activation, hypoxia/acidosis, stimulation or deprivation by
cytokines, as well as depletion of nucleoside triphosphate pools, p53 protein levels surge in the
cell.41, 74-77 Upon activation, p53 can induce a variety of cellular responses, most notably cell
184 Cell Cycle Checkpoints and Cancer
cycle arrest or apoptosis. Additionally, p53 has been implicated in prevention of embryonic
malformations despite the fact that p53 null mice display a relatively normal pattern of development.
78,79 p53 is involved in cellular senescence and participates in the regulation of
centrosome number.80-82 The regulation of cell cycle arrest and apoptosis likely underlie p53’s
role in modulating cancer therapy and are explored here in further detail.
Cell Cycle Arrest
p53 intervenes at several points in the cell cycle. It is thought to do so by transcriptionally
activating genes that are key to the cellular replication machinery. p53’s sequence-specific DNA
binding domain is localized between amino acid residues 102 and 292 where hydrogen bonds
contact both the minor and major groove of the double helix.83 p53 can mediate both G1 and
G2/M cell cycle arrest in response to DNA damage. It likely performs these functions by activating
target genes that contain p53-dependent, cis-acting, DNA- responsive elements namely
p21/WAF1/CIP1, GADD45 and IGF-BP3.39,60,84
p21/WAF1/CIP1 is a potent mediator of the G1 cell cycle checkpoint because it is able to
associate with and inhibit G1 cyclin/cyclin-dependent kinase complexes, specifically cyclin
D1-CDK4, cyclin E-CDK2, cyclin A-CDK2, and cyclin A-cdc2.84-87 p21/WAF1/CIP1 has
been shown to bind directly to proliferating cell nuclear antigen (PCNA) and inhibit
PCNA-dependent DNA replication.88-90 Mice deficient in the p21 gene develop normally but
fibroblasts derived from these mice have a limited ability to arrest in G1 following DNA damage.
91 Nonetheless, p21/WAF1/CIP1 is not the only p53 mediated mechanism to arrest cells
in G1, since p21 null fibroblasts retain some of the p53-dependent G1 checkpoint activity.
GADD45 expression has been associated with both G1 and G2/M arrest.92,93 GADD45 can
stimulate excision repair and like p21WAF1/CIP1 also binds to PCNA and may play a backup
role for p21.94 IGF-BP3 gene codes for a protein that inhibits insulin-like growth factor, a mitogenic
factor. p53 has been shown to activate IGF-BP3 thereby limiting the mitogenic response.95
p53 may also interact via protein-protein interactions with other tumor related factors
including the Wilm’s tumor gene (WT1) and the Gas1 gene. WT1 gene when coexpressed
with p53 results in higher steady state levels of p53, enhanced transcriptional activity of p53,
and increased levels of p53 DNA sequence-specific binding.96 The Gas 1 gene encodes a membrane
protein associated with Go arrest which functions when p53 is present. Interestingly
p53’s transcriptional function appears dispensable for Gas 1 induction because p53 with a
mutation preventing p53 transcriptional activity can still induce Gas1.97
Chk2, a DNA damage-responsive kinase, can phosphorylate Cdc25C causing binding to
14-3-3 protein and inhibition of its ability to promote cell cycle progression via cdc2.98 In
addition, Chk2 has another important role in cell cycle regulation which is mediated by its
ability to phosphorylate p53 on serine 20. This phosphorylation prevents Mdm2 binding and
results in p53 stabilization.98 ATM (the gene defective in ataxia-telangectasia) can also phosphorylate
p53 on serine 15 following ionizing radiation. This phosphorylation is important for
activation of p53 as a transcription factor and may act synergistically with serine 20 phosphorylation.
99,100 ATM, Chk1 and Chk2 are all activated following DNA damage, though with
differential responsiveness depending on the agent inducing the DNA damage as well as the
particular cell type and the cell cycle stage during which the damage occurs.101 Chk2 is now
recognized as a tumor suppressor; germline mutations in Chk2 occur in Li-Fraumeni-like
syndrome patients who lack the more common germline p53 mutations.98
Apoptosis
p53’s role in modulating the apoptotic response likely plays an important role in p53’s
tumor suppressor activity as well as its ability to mediate therapy-directed tumor cell death.
Although its ability to trigger apoptosis following appropriate signals is profound, the mechanistic
pathway(s) connecting p53 to the apoptotic death machinery have remained complex
and elusive. In contrast to the cell cycle activity of p53 which is largely (though not entirely)
P53, Apoptosis and Cancer Therapy 185
attributable to the transcriptional activation of the p21 CDK-inhibitor, no single essential
mediator of p53’s apoptotic activity has emerged thus far. In fact, a variety of experimental
systems have suggested that p53 may fundamentally utilize at least two apoptotic pathways,
one involving transcription of specific target death genes, and another which is a transcription-independent
road to apoptotic death. Although the road to apoptosis may traverse distinct paths,
the common endpoint is activation of executioner caspases. The two pathways leading to such
caspase activation are the death receptor and the mitochondrial pathways (Figs. 2 and 3).
p53 Mediated Transcription Dependent Apoptosis
There has been a growing list of p53 target genes involved in apoptosis. These include
death-promoting genes such as BAX, CD95/Fas/Apo1, the PIG genes involved in redox regulation,
and the TNF-receptor family of genes containing death domains. BAX is the best known
pro-apoptotic p53 target and member of the bcl-2 family. Its promoter contains p53-binding
sites and is upregulated by p53 in response to DNA damage.13,102 BAX can function as an
effector of p53, allowing for chemotherapy-induced apoptosis and suppression of oncogenic
transformation.103 The precise mechanism by which BAX activation leads to apoptosis has
been somewhat unclear. BAX is found both in the cytosol and mitochondria, but its translocation
from cytosol to the mitochondria seems to be essential for BAX function.104 This translocation
of BAX may require the presence of a newly discovered mediator of apoptosis known as Peg3/
Pw1.104,105 Once localized to mitochondria BAX may participate in generation of membrane
channels leading to the release of cytochrome C to the cytosol where it binds with Apaf-1 and
activates procaspase 9.104,106 Thus BAX clearly links p53 with the apoptotic machinery of the
mitochondria. However, despite the importance of BAX as a p53 target gene, the observation
that BAX-deficient thymocytes retain intact radiation induced apoptosis107 suggests that BAX
cannot be the only mediator of p53-dependent apoptosis.
PIGs are p53-induced genes identified using SAGE analysis from colon carcinoma cells.108
Most were predicted to be involved in modulating the redox state of the cell. These genes
encode proteins which may directly or indirectly stimulate the production of reactive oxygen
species which signal mitochondrial release of cytochrome C or other caspase activators.
Another group of p53-targeted apoptosis modulators consists of members of the TNF
family of transmembrane death receptors. These include Fas/APO-1/CD95, TNFR1, DR-3/
APO-3/WSL-1/TRAMP, DR4/TRAIL-R1, and DR5/TRAIL-R2/KILLER. These receptors
share a conserved cysteine-rich repeat at their extracellular domains. Fas, TNFR1, DR3, DR5/
KILLER, and CAR1 carry a region of homology in the cytoplasmic tail called the death
domain, which is essential to its ability to transduce an apoptotic signal.109 The activating
ligands for the death receptors are structurally related and belong to the TNF gene superfamily.
Fas ligand (FasL) binds to Fas, Apo3 ligand (Apo3L) binds to DR3 and Apo2 ligand (Apo2L or
TRAIL) binds to DR5/Killler.110 When Fas binds to its ligand (FasL) it induces trimerization
of Fas and the cytoplasmic region of Fas, which contains the death domain, recruits FADD.
FADD also contains a death domain and it is the interaction between the Fas and FADD death
domains that propogate the death signal inside the cell.106 A single point mutation in the death
domain abrogates the apoptotic signal.111,112 Physical association of FADD with Fas leads to
oligomerization of pro-caspase 8 within the DISC complex and generation of active caspase 8
which propagates subsequent activation of the executioner caspases. DR5/KILLER has a similar
mechanism of action after binding by the ligand TRAIL (APO2L). Both TRAIL and KILLER/
DR5 when overexpressed can act independently to induce apoptosis. However there are decoy
receptors for TRAIL (DcR1 and DcR2) which can block TRAIL function.113 p53 may transcriptionally
activate Fas and FasL expression.114-116 It has also recently been found that KILLER/
DR5 is induced by DNA damage selectively in wild type p53 cells, and p53-dependent transcriptional
induction of KILLER/DR5 death receptor may be restricted to cells undergoing
apoptosis rather than growth arrest.117
IGF-BP3, in addition to its anti-mitotic function, can bind to IGF-1 and thereby antagonize
its anti-apoptotic influence. A number of cytokine growth factors display strong
186 Cell Cycle Checkpoints and Cancer
anti-apoptotic influences. p53 may act through induction of IGF-BP3 to promote apoptosis to
disrupt the anti-apoptotic signaling.95,118
There are several novel p53 target genes believed to be mediators of apoptosis about which
relatively little mechanism of action is known. PAG608 is a zinc finger protein that induces
apoptosis when introduced into human cancer cells.119 PERP is a recently identified p53 gene
target in the gas3 family which is specifically induced upon DNA damage during apoptosis
and is another candidate effector in the p53 transcription dependent apoptotic pathway120.
DRAL, a p53 responsive gene, has been implicated in the apoptotic pathway because its expression
efficiently triggers apoptosis in multiple cell lines of different origins.121 This protein
also contains a zinc finger domain similar to PAG608. Potential functions for this protein
include a scaffolding role or as a modulator of transcription.122,123 Noxa encodes a protein
containing the Bcl-2 homology 3 (BH3) motif seen in the Bcl-2 family of proteins. The promoter
region of the Noxa gene contains a p53 response element and increased expression of
Noxa mRNA was observed after cells were infected with adenovirus overexpressing p53.124
Noxa protein localizes to the mitochondrial membrane.
p53 can also transcriptionally repress certain genes, particularly in the context of promoters
lacking p53 consensus DNA binding elements. For example, bcl-2 expression may be
decreased in the presence of p53-mediated apoptosis.125 p53 also negatively regulates expression
of MAP4 a microtubule stabilizing protein.126
p53 Mediated, Transcription-Independent Apoptosis
A major puzzle has emerged from studies which suggested that p53 may, in certain contexts,
trigger apoptosis without the apparent need for its transcriptional activity.14,15,127,128 The original
data supporting this possibility arose from the observation that p53’s apoptotic activity was
resistant to actinomycin D or cycloheximide treatment. Additional evidence came from mutational
studies in which overexpression of transcriptionally incompetent p53 mutants produced
clear apoptotic activity.16 These observations as well as others which followed, effectively
uncoupled p53’s apoptotic activity from its transcriptional function.
Fig. 3. The mitochondrial apoptosis pathway. A wide variety of cellular stresses trigger signaling cascades
which produce mitochondrial release of cytochrome C and Smac/Diablo. These mitochondrial factors
then assist in the activation of pro-caspase 9 in the cytosol. Cytochrome C forms an activating complex
with Apaf-1 and pro-caspase 9, while Smac/Diably displaces the IAP apoptosis inhibitor
molecules from caspase 9.
P53, Apoptosis and Cancer Therapy 187
The proline rich region of p53 contains several SH3 binding PXXP motifs and displays
SH3 binding activity. This area may serve as a docking site for other SH3-containing proteins.
A human p53 mutant derived from familial breast cancer destroys one of these proline rich
binding motifs suggesting that the proline rich domain is important to p53’s apoptotic function.
129,130 A Li-Fraumeni syndrome family was found to harbor a p53 hinge domain mutation
which retained the ability to arrest the cell cycle but not trigger apoptosis when
overexpressed.131 Perhaps this hinge region serves as a regulatory domain for apoptosis. The
most common mutations seen in human cancers occur in the central DNA-binding domain of
p53 that lead to a loss in the DNA binding function of the p53 protein or a change in confirmation
of the protein.51,83 There is evidence that this central domain can bind 53BP2 protein
that interacts with bcl-2.132 Finally, a cell-free apoptosis system was devised in which p53
protein appears to modulate the activation of caspase 8 via as yet undefined intermediates.17, 133
Regulating p53 Activation in the Stress Response
Regulation of p53 occurs predominantly through control of its protein stability. Mdm2
directly binds residues within the N-terminal transactivation domain of p53 thereby repressing
transcriptional activation of p53 and targeting it for degradation.64,65,134 p53 transcriptionally
activates Mdm2 expression thus providing negative autoregulation. Mdm2 acts as ubiquitin
ligase for p53.135,136 The high levels of mutant p53 in many tumors is probably related to lack
of Mdm2-mediated p53 turnover.137 Other factors can act to degrade p53 such as papillomavirus
E6 and adenovirus E1B proteins.138,139 Onocogenes are potent stabilizers of p53, potentially
providing the organism with a mechanism to curb malignant transformation. Myc, Ras and
E1A are thought to stabilize p53 via E2F-mediate upregulation of p19ARF (P14ARF in
humans), which inhibits Mdm2 and therefore decreases p53 degradation.140-142
p53 activity is significantly affected by acetylation and phosphorylation. p53 is
phosphorylated on Ser 15 by the ATM protein kinase in response to DNA damage.99 This
phosphorylation leads to protein accumulation as well as the increased ability of p53 to
transactivate downstream target genes.100 Phosphorylation of p53 can also take place on Ser 20
by Chk2, which results in p53 stabilization.98 It appears that phosphorylation at Ser 15 acts
synergistically with Ser 20 phosphorylation.143 Acetylation of p53 at the C terminus promotes
DNA binding activity as well as proper oligomerization and subcellular localization.144,145 In
addition, phosphorylation at the N terminus mediates acetylation at the C terminus and can
enhance DNA binding.127,146
Cell Cycle Arrest vs Death
The decision of arrest vs. death is potentially of enormous clinical importance in cancer
therapy. Studies in fibroblasts have demonstrated that in the oncogene-transformed setting,
ionizing radiation triggers p53-dependent apoptosis whereas in primary, untransformed fibroblasts
ionizing radiation triggers p53-dependent cell cycle arrest.46,47 The relevance of these
behaviors to cancer therapy stems from both the selective sensitization towards death in the
oncogene-transformed setting, as well as the protective effects of cell cycle arrest on the ability
of normal cells to repair DNA damage. Thus an apparent therapeutic index is created in which
p53 simultaneously protects normal cells via its cell cycle activity while targeting tumor cells
for apoptotic death (Fig. 4).
How might p53 selectively trigger death vs. arrest? One hypothesis proposes that the
quantity of DNA damage and, by extension, the overall level of p53 protein induction may
determine the outcome. In this system, higher levels of p53 protein were associated with
apoptosis.147 The oncogene-specific death response in primary fibroblasts could also theoretically
arise via Ink4a/ARF effects. Since oncogenic upregulation of ARF is predicted to repress
Mdm2’s downregulation of p53, oncogene-transformed cells are predicted to contain higher
basal p53 levels. In certain settings evidence suggested that both cell cycle arrest genes and
apoptosis mediating genes are activated in cells undergoing arrest alone or apoptosis.108,142
188 Cell Cycle Checkpoints and Cancer
Such data could still be consistent with the above hypothesis, if the decision were also determined
by cellular context. In fact, whereas primary fibroblasts arrest following ionizing radiation,
certain primary cell types such as thymocytes undergo p53-dependent apoptosis instead.
Still, it is unclear whether p53 levels correlate in different cell systems with the death vs. arrest
decision. For these reasons it is also plausible that a cellular setpoint or apoptosis threshold
determines whether p53’s output will result in death.
Another important determinant of arrest versus death may be the presence or absence of
cytokines. For example in the hematopoietic Baf/3 cell system, irradiation in the presence of
IL-3 produces cell cycle arrest whereas irradiation in the absence of IL-3 produces rapid apoptosis.
In this system, p53-independent regulation of Gadd45 and p21/Waf1/Cip1 could play a role
in modulating the apoptosis vs. arrest threshold.148 Alternatively, it is plausible that activation
of the PI3 kinase-Akt survival pathway by growth factors or Ras-mediated signaling may “set”
the apoptotic threshold to respond differently to identical p53 dependent apoptosis triggers.
Finally, the KILLER/DR5 receptor is a TNF receptor family member whose expression is
upregulated by DNA damage in a p53 dependent fashion.117,149,150 Activation of this pathway
by the ligand TRAIL results in death receptor-mediated apoptosis. Interestingly, a decoy receptor
which binds the same ligand (TRAIL) is often expressed in non-tumor cells, thereby potentially
conferring tumor selectivity to TRAIL as a therapeutic. However recent evidence suggests
that p53 may upregulate expression of decoy receptors as well, thereby potentially blunting the
apoptotic response to TRAIL.151,152 This pathway, as well as the Fas/FasL pathway which is
also transcriptionally regulated by p53, may help explain p53’s apoptotic activity and may
simultaneously be exploitable for clinical benefit if tumor selectivity were consistently revealed.
Therapy
Numerous well-known variables determine therapeutic efficacy in cancer including drug
delivery, selectivity (therapeutic index), and mechanisms of intrinsic or acquired resistance. A
number of experimental therapeutic strategies for cancer have revolved around modulating
p53’s apoptotic activity presumably because of p53’s ability to selectively instruct tumor cells to
undergo apoptosis. While most of these approaches have theoretical pitfalls, the creative evolution
of this field lends optimism to the prospect that improved therapies may eventually emerge.
One treatment strategy is reintroduction of wildtype p53 itself into p53 deficient tumor
cells.153-155 Concerns of delivery, normal cellular toxicity, and overcoming dominant negative effects
of mutant p53 could limit the applicability of this approach, but potential benefits may be
seen in specific clinical settings. It is likely that p53 reintroduction would be especially useful in
tumors harboring dysregulated E2F-1 (the majority of human malignancies) since such cells should
be hypersensitive to p53-dependent apoptosis if the Ink4a/ARF pathway is intact.
Another strategy employs a C-terminal fragment of p53 peptide to reactivate endogenous
mutant p53.156 This approach is based upon the known inhibitory role of the endogenous
C-terminal region of p53. Displacement of that motif by exogenous peptide may permit
reactivation of the mutant protein, and since the mutant protein is often present at significantly
higher levels, the result could be tumor-selective apoptosis as described by Fine and
colleagues.157 In the long term, it is likely that small molecule drug discovery or peptido-mimetic
strategies will be necessary to translate this approach into an in vivo therapy.
There are certain tumors in which p53 is wildtype, but its function is apparently deficient.
For example in cervical carcinoma cells, disruption of E6 without compromising E7 can lead
to extensive apoptosis.158,159 Although the generality of these findings remain to be confirmed,
one group has found that certain neuroblastomas express genomic sequences from human
polyomavirus BK (BKV). Neuroblastomas typically lack p53 mutations, although p53 tends
to be cytoplasmically sequestered and therefore less than fully functional. In this setting,
treatment of neuroblastoma cells with BKV large T Ag antisense was reported to correct the
translocation of p53 to the nucleus with rescue of wildtype p53 function including apoptosis.160
Another approach focuses on altering p53 in those cancer cells with mutated p53.
Trans-splicing ribozymes were used that can simultaneously reduce mutant p53 expression and
P53, Apoptosis and Cancer Therapy 189
restore wildtype p53 activity in various human cancers. The ribozyme acts to repair mutant
p53 mRNAs with high fidelity and specificity. Results showed transactivation of p53-responsive
promoters and down-regulated expression of the multidrug resistant gene promoter.161 One
novel therapeutic approach involves exploiting the ATM pathway of p53 activation. Although
ATM deficiency significantly delays p53 activation following ionizing radiation,162 ATM deficiency
also confers profound radiation sensitivity to multiple organ systems in ATM knockout
mice and this radiosensitivity is largely p53-independent.163-165 Therefore ATM may serve as a
significant radiosensitizing drug target for p53-deficient cancers, although tumor selectivity is a
potential problem (which might be at least partially overcome through localized radiation treatment).
An alternate strategy for exploiting p53 in cancer therapy focuses on targeting p53 as a
radio-or chemo-protective treatment. Since most incurable human tumors are p53 deficient, it
was proposed that interfering with host p53 may diminish normal cellular toxicity during
chemotherapy, while having minimal effect on the tumor population (which is already p53
deficient). This strategy proved promising in recent studies,166 although it carries a theoretical
risk of facilitating oncogenic transformation, particularly during chemotherapy much of which
is mutagenic. However no increased tumor incidence was reported in the initial work, so transient
p53 inhibition may prove to be clinically useful.
Clearly much remains to be learned about the mechanisms which underlie p53’s tumor
selective activities. Specifically, it is still incompletely understood how the death vs. arrest decision
is made, and whether this pathway may be amenable to drug targeting for therapeutic
benefit in p53 deficient tumors. The explosion of information surrounding the numerous
pathways with which p53 interacts suggests that a clearer understanding of its activities may
Fig. 4. Role of p53 in the cellular stress response. In fibroblasts, primary cells undergo cell cycle arrest,
whereas oncogene-transformed cells undergo apoptosis in the setting of wildtype p53. Where p53 is deficient,
primary fibroblasts display aberrant G0/G1 cell cycle arrest, and oncogene transformed fibroblasts
are significantly resistant to the induction of apoptosis, thereby diminishing the potential therapeutic index
conferred by wildtype p53.
190 Cell Cycle Checkpoints and Cancer
not be far away. Moreover the frequency of its inactivation in human cancer and the potency of
its activities in experimental models suggest that the payoff may be well worth the wait.
References
1. Ellis HM, and Horvitz HR, Genetic control of programmed cell death in the nematode C. elegans;
Cell 1986; 44:817-29.
2. Hengartner MO and Horvitz, HR, C. elegans cell survival gene ced-9 encodes a functional
homolog of the mammalian proto-oncogene bcl-2. Cell 1994; 76:665-76.
3. Roy N, Deveraux QL, Takahashi R et al. The c-IAP-1 and c-IAP-2 proteins are direct inhibitors
of specific caspases. Embo J 1997; 16:6914-25.
4. Deveraux QL, Takahashi R Salvesen GS et al. X-linked IAP is a direct inhibitor of cell-death
proteases. Nature 1997; 388:300-4.
5. Takahashi R, Deveraux Q, Tamm I et al. A single BIR domain of XIAP sufficient for inhibiting
caspases. J Biol Chem 1998; 273:7787-90.
6. Du C, Fang, M Li, Y et al. Smac, a mitochondrial protein that promotes cytochrome c-dependent
caspase activation by eliminating IAP inhibition. Cell 2000; 102:33-42.
7. Verhagen AM, Ekert PG, Pakusch, M et al. Identification of DIABLO, a mammalian protein that
promotes apoptosis by binding to and antagonizing IAP proteins. Cell 2000; 102:43-53.
8. Muzio M, Stockwell BR, Stennicke, HR et al. An induced proximity model for caspase-8 activation.
J Biol Chem 1998; 273:2926-30.
9. Li P, Nijhawan, D Budihardjo, I et al. Cytochrome c and dATP-dependent formation of Apaf-1/
caspase-9 complex initiates an apoptotic protease cascade. Cell 1997; 91 479-89.
10. Garland JM, and Rudin C, Cytochrome c induces caspase-dependent apoptosis in intact hematopoietic
cells and overrides apoptosis suppression mediated by bcl-2, growth factor signaling,
MAP-kinase-kinase, and malignant change. Blood 1998; 92:1235-46.
11. Yang J, Liu X, Bhalla K et al. Prevention of apoptosis by Bcl-2: Release of cytochrome c from
mitochondria blocked [see comments]. Science 1997; 275:1129-32.
12. Scaffidi C, Fulda S, Srinivasan A et al. Two CD95 (APO-1/Fas) signaling pathways. Embo J 1998;
17:1675-87.
13. Miyashita T, and Reed JC, Tumor suppressor p53 is a direct transcriptional activator of the human
BAX gene. Cell 1995; 80:293-9.
14. Caelles C, Helmberg A, and Karin M, p53-dependent apoptosis in the absence of transcriptional
activation of p53-target genes [see comments]. Nature 1994; 370:220-3.
15. Wagner AJ, Kokontis JM, and Hay N, Myc-mediated apoptosis requires wild-type p53 in a manner
independent of cell cycle arrest and the ability of p53 to induce p21WAF1/cip1. Genes Dev 1994;
8:2817-30.
16. Haupt Y, Rowan S, Shaulian, E et al. Induction of apoptosis in HeLa cells by trans-activation-deficient
p53. Genes Dev 1995; 9:2170-83.
17. Ding HF, McGill G, Rowan S et al. Oncogene-dependent regulation of caspase activation by p53
protein in a cell-free system, J Biol Chem 1998; 273:28378-83.
18. Bennett M, Macdonald K, Chan SW et al., Cell surface trafficking of Fas: A rapid mechanism of
p53-mediated apoptosis [see comments]. Science 1998; 282:290-3.
19. Wang XW, Yeh H, Schaeffer L et al. p53 modulation of TFIIH-associated nucleotide excision
repair activity. Nat Genet 1995; 10:188-95.
20. Linzer DI, and Levine AJ, Characterization of a 54K dalton cellular SV40 tumor antigen present
in SV40-transformed cells and uninfected embryonal carcinoma cells. Cell 1979; 17:43-52.
21. Lane DP, and Crawford LV, T antigen is bound to a host protein in SV40-transformed cells.
Nature 1979; 278:261-3.
22. Eliyahu D, Raz A, Gruss, P et al. Participation of p53 cellular tumour antigen in transformation of
normal embryonic cells. Nature 1984; 312:646-9.
23. Parada LF, Land H, Weinberg, RA et al. Cooperation between gene encoding p53 tumour antigen
and Ras in cellular transformation. Nature 1984; 312:649-51.
24. Jenkins JR, Rudge K, and Currie GA, Cellular immortalization by a cDNA clone encoding the
transformation-associated phosphoprotein p53. Nature 1984; 312:651-4.
25. Wolf D, Harris N, and Rotter V, Reconstitution of p53 expression in a nonproducer
Ab-MuLV-transformed cell line by transfection of a functional p53 gene. Cell 1984; 38:119-26.
26. Eliyahu D, Michalovitz D, and Oren M, Overproduction of p53 antigen makes established cells
highly tumorigenic. Nature 1985; 316:158-60.
27. Finlay CA, Hinds,PW, and Levine AJ, The p53 proto-oncogene can act as a suppressor of transformation.
Cell 1989; 57:1083-93.
P53, Apoptosis and Cancer Therapy 191
28. Hinds P, Finlay C, and Levine AJ, Mutation is required to activate the p53 gene for cooperation
with the Ras oncogene and transformation. J Virol 1989; 63:739-46.
29. Eliyahu D, Goldfinger N, Pinhasi-Kimhi, O et al. Meth A fibrosarcoma cells express two transforming
mutant p53 species. Oncogene 1988; 3:313-21.
30. Masuda H, Miller C, Koeffler HP et al. Rearrangement of the p53 gene in human osteogenic
sarcomas. Proc Natl Acad Sci USA 1987; 84:7716-9.
31. Lubbert M, Miller CW, Crawford L et al. p53 in chronic myelogenous leukemia. Study of mechanisms
of differential expression. J Exp Med 1988; 167:873-86.
32. Vogelstein B, Fearon ER, Kern SE et al. Allelotype of colorectal carcinomas. Science 1989;
244:207-11.
33. Baker SJ, Fearon, ER Nigro, JM et al. Chromosome 17 deletions and p53 gene mutations in
colorectal carcinomas. Science 1989; 244:217-21.
34. Nigro JM, Baker SJ, Preisinger AC et al. Mutations in the p53 gene occur in diverse human
tumour types. Nature 1989; 342:705-8.
35. Eliyahu D, Michalovitz D, Eliyahu S et al. Wild-type p53 can inhibit oncogene-mediated focus
formation. Proc Natl Acad Sci U S A 1989; 86:8763-7.
36. Isobe M, Emanuel BS, Givol D et al. Localization of gene for human p53 tumour antigen to band
17p13. Nature 1986; 320:84-5.
37. Malkin D, Li FP, Strong LC et al. Germ line p53 mutations in a familial syndrome of breast
cancer, sarcomas, and other neoplasms [see comments]. Science 1990; 250:1233-8.
38. Jayaraman J, and Prives C, Activation of p53 sequence-specific DNA binding by short single strands
of DNA requires the p53 C-terminus. Cell 1995; 81:1021-9.
39. Ko LJ, and Prives C, p53: puzzle and paradigm. Genes Dev 1996; 10:1054-72.
40. Lee S, Elenbaas B, Levine A et al. p53 and its 14 kDa C-terminal domain recognize primary DNA
damage in the form of insertion/deletion mismatches. Cell 1995; 81:1013-20.
41. Linke SP, Clarkin KC, Di Leonardo A et al. A reversible, p53-dependent G0/G1 cell cycle arrest
induced by ribonucleotide depletion in the absence of detectable DNA damage, Genes Dev 1996;
10:934-47.
42. Lutzker SG, and Levine AJ, A functionally inactive p53 protein in teratocarcinoma cells is activated
by either DNA damage or cellular differentiation. Nat Med 1996; 2:804-10.
43. Yonish-Rouach E, Resnitzky D, Lotem J et al., Wild-type p53 induces apoptosis of myeloid
leukaemic cells that is inhibited by interleukin-6. Nature 1991; 352:345-7.
44. Clarke AR, Purdie CA, Harrison DJ et al. Thymocyte apoptosis induced by p53-dependent and
independent pathways [see comments]. Nature 1993; 362:849-52.
45. Lowe SW, Schmitt EM, Smith SW et al. p53 is required for radiation-induced apoptosis in mouse
thymocytes [see comments]. Nature 1993; 362:847-9.
46. Lowe SW, Ruley HE, Jacks T et al. p53-dependent apoptosis modulates the cytotoxicity of anticancer
agents. Cell 1993; 74:957-67.
47. Lowe SW, Bodis S, McClatchey, A et al. p53 status and the efficacy of cancer therapy in vivo.
Science 1994; 266:807-10.
48. Beroud C, Collod-Beroud G, Boileau C et al. UMD (Universal mutation database): A generic
software to build and analyze locus-specific databases. Hum Mutat 2000; 15:86-94.
49. Soussi T, Dehouche K, and Beroud C, p53 website and analysis of p53 gene mutations in human
cancer: forging a link between epidemiology and carcinogenesis. Hum Mutat 2000; 15:105-13.
50. Hollstein M, Rice K, Greenblatt MS et al., Database of p53 gene somatic mutations in human
tumors and cell lines, Nucleic Acids Res 1994; 22:3551-5.
51. Soussi T, The p53 tumor suppressor gene: from molecular biology to clinical investigation. Ann N
Y Acad Sci 2000; 910:121-37; discussion 137-9.
52. Soussi T, Legros Y, Lubin R et al. Multifactorial analysis of p53 alteration in human cancer: a
review. Int J Cancer 1994; 57:1-9.
53. Aas T, Borresen AL, Geisler S et al. Specific P53 mutations are associated with de novo resistance
to doxorubicin in breast cancer patients. Nat Med 1996; 2:811-4.
54. Hamelin R, Laurent-Puig P, Olschwang S et al., Association of p53 mutations with short survival
in colorectal cancer [see comments]. Gastroenterology 1994; 106:42-8.
55. Harris CC, and Hollstein M, Clinical implications of the p53 tumor-suppressor gene [see comments].
N Engl J Med 1993; 329:1318-27.
56. Fisher DE, Apoptosis in cancer therapy: crossing the threshold. Cell 1994; 78:539-42.
57. Heimdal K, Lothe RA, Lystad S et al. No germline TP53 mutations detected in familial and
bilateral testicular cancer, Genes Chromosomes Cancer 1993; 6:92-7.
58. Bardeesy N, Falkoff D, Petruzzi MJ et al. Anaplastic Wilms’ tumour, a subtype displaying poor
prognosis, harbours p53 gene mutations. Nat Genet 1994; 7: 91-7.
192 Cell Cycle Checkpoints and Cancer
59. Felix CA, Nau MM, Takahashi T et al., Hereditary and acquired p53 gene mutations in childhood
acute lymphoblastic leukemia. J Clin Invest 1992; 89:640-7.
60. Levine AJ, p53, the cellular gatekeeper for growth and division. Cell 1997; 88:323-31.
61. Scheffner M, Werness BA, Huibregtse JM et al. The E6 oncoprotein encoded by human
papillomavirus types 16 and 18 promotes the degradation of p53. Cell 1990; 63:1129-36.
62. Werness BA, Levine AJ, and Howley PM, Association of human papillomavirus types 16 and 18
E6 proteins with p53. Science 1990; 248:76-9.
63. Hoppe-Seyler F, and Butz K, Repression of endogenous p53 transactivation function in HeLa
cervical carcinoma cells by human papillomavirus type 16 E6, human mdm-2, and mutant p53. J
Virol 1993; 67:3111-7.
64. Kubbutat MH, Jones SN, and Vousden KH, Regulation of p53 stability by Mdm2. Nature 1997;
387:299-303.
65. Haupt Y, Maya R, Kazaz A et al. Mdm2 promotes the rapid degradation of p53. Nature 1997;
387:296-9.
66. Momand J, Zambetti GP, Olson DC et al. The mdm-2 oncogene product forms a complex with
the p53 protein and inhibits p53-mediated transactivation. Cell 1992; 69:1237-45.
67. Oliner JD, Pietenpol JA, Thiagalingam S et al. Oncoprotein MDM2 conceals the activation domain
of tumour suppressor p53. Nature 1993; 362:857-60.
68. Zhang Y, Xiong Y, and Yarbrough WG, ARF promotes MDM2 degradation and stabilizes
p53: ARF-INK4a locus deletion impairs both the Rb and p53 tumor suppression pathways.
Cell 1998; 92:725-34.
69. Kamijo T, Weber JD, Zambetti G et al. Functional and physical interactions of the ARF tumor
suppressor with p53 and Mdm2. Proc Natl Acad Sci USA 1998; 95:8292-7.
70. Moll UM, Riou G, and Levine AJ, Two distinct mechanisms alter p53 in breast cancer: Mutation
and nuclear exclusion. Proc Natl Acad Sci U S A 1992; 89:7262-6.
71. Takahashi K, Suzuki K, and Tsukatani Y, Induction of tyrosine phosphorylation and association of
beta-catenin with EGF receptor upon tryptic digestion of quiescent cells at confluence. Oncogene
1997; 15:71-8.
72. Vogan K, Bernstein M, Leclerc JM et al. Absence of p53 gene mutations in primary neuroblastomas,
Cancer Res 1993; 53:5269-73.
73. Lane DP, Cancer. p53, guardian of the genome [news; comment] [see comments]. Nature 1992;
358:15-6.
74. Kastan MB, Onyekwere O, Sidransky D et al. Participation of p53 protein in the cellular response
to DNA damage. Cancer Res 1991; 51:6304-11.
75. Graeber TG, Osmanian C, Jacks T et al. Hypoxia-mediated selection of cells with diminished
apoptotic potential in solid tumours [see comments]. Nature 1996; 379:88-91.
76. Schmaltz C, Hardenbergh PH, Wells A et al. Regulation of proliferation-survival decisions during
tumor cell hypoxia. Mol Cell Biol 1998; 18:2845-54.
77. Eizenberg O, Faber-Elman A, Gottlieb E et al. Direct involvement of p53 in programmed cell
death of oligodendrocytes. Embo J 1995; 14:1136-44.
78. Donehower LA, Harvey M, Slagle BL et al. Mice deficient for p53 are developmentally normal but
susceptible to spontaneous tumours. Nature 1992; 356:215-21.
79. Feinstein E, Gale RP, Reed J et al. Expression of the normal p53 gene induces differentiation of
K562 cells. Oncogene 1992; 7:1853-7.
80. Fukasawa K, Choi T, Kuriyama R et al. Abnormal centrosome amplification in the absence of
p53. Science 1996; 271:1744-7.
81. Atadja P, Wong H, Garkavtsev I et al. Increased activity of p53 in senescing fibroblasts. Proc Natl
Acad Sci USA 1995; 92:8348-52.
82. Bond J, Haughton M, Blaydes J et al. Evidence that transcriptional activation by p53 plays a
direct role in the induction of cellular senescence. Oncogene 1996; 13:2097-104.
83. Cho Y, Gorina S, Jeffrey PD et al. Crystal structure of a p53 tumor suppressor-DNA complex:
Understanding tumorigenic mutations [see comments], Science 1994; 265:346-55.
84. el-Deiry WS, Tokino T, Velculescu VE et al. WAF1, a potential mediator of p53 tumor suppression.
Cell 1993; 75:817-25.
85. el-Deiry WS, Harper JW, O’Connor PM et al. WAF1/CIP1 is induced in p53-mediated G1 arrest
and apoptosis. Cancer Res 1994; 54:1169-74.
86. Harper JW, Adami GR, Wei N et al. The p21 CDK-interacting protein Cip1 is a potent inhibitor
of G1 cyclin-dependent kinases. Cell 1993; 75:805-16.
87. Gu Y, Turck CW, and Morgan DO, Inhibition of CDK2 activity in vivo by an associated 20K
regulatory subunit. Nature 1993; 366:707-10.
P53, Apoptosis and Cancer Therapy 193
88. Flores-Rozas H, Kelman Z, Dean FB et al. CDK-interacting protein 1 directly binds with proliferating
cell nuclear antigen and inhibits DNA replication catalyzed by the DNA polymerase delta holoenzyme.
Proc Natl Acad Sci USA 1994; 91:8655-9.
89. Luo Y, Hurwitz J, and Massague J, Cell-cycle inhibition by independent CDK and PCNA binding
domains in p21CIP1. Nature 1995; 375:159-61.
90. Waga S, Hannon GJ, Beach D et al. The p21 inhibitor of cyclin-dependent kinases controls DNA
replication by interaction with PCNA [see comments]. Nature 1994; 369:574-8.
91. Deng C, Zhang P, Harper JW et al. Mice lacking p21CIP1/WAF1 undergo normal development.
but are defective in G1 checkpoint control. Cell 1995; 82:675-84.
92. Chin PL, Momand J, and Pfeifer GP, In vivo evidence for binding of p53 to consensus binding
sites in the p21 and GADD45 genes in response to ionizing radiation. Oncogene 1997; 15:87-99.
93. Wang XW, Zhan Q, Coursen JD et al. GADD45 induction of a G2/M cell cycle checkpoint. Proc
Natl Acad Sci USA 1999; 96:3706-11.
94. Smith ML, Chen IT, Zhan Q et al. Interaction of the p53-regulated protein Gadd45 with proliferating
cell nuclear antigen [see comments]. Science 1994; 266:1376-80.
95. Buckbinder L, Talbott R, Velasco-Miguel, S et al. Induction of the growth inhibitor IGF-binding
protein 3 by p53. Nature 1995; 377:646-9.
96. Maheswaran S, Englert C, Bennett, P et al. The WT1 gene product stabilizes p53 and inhibits
p53-mediated apoptosis. Genes Dev 1995; 9:2143-56.
97. Del Sal G, Ruaro EM, Utrera R et al. Gas1-induced growth suppression requires a transactivationindependent
p53 function. Mol Cell Biol 1995; 15:7152-60.
98. Hirao A, Kong YY, Matsuoka S et al. DNA damage-induced activation of p53 by the checkpoint
kinase Chk2 [see comments]. Science 2000; 287:1824-7.
99. Banin S, Moyal L, Shieh S et al. Enhanced phosphorylation of p53 by ATM in response to DNA
damage. Science 1998; 281:1674-7.
100. Canman CE, Lim DS, Cimprich KA et al. Activation of the ATM kinase by ionizing radiation and
phosphorylation of p53. Science 1998; 281:1677-9.
101. Carr AM, Cell cycle. Piecing together the p53 puzzle [comment]. Science 2000; 287:1765-6.
102. Zhan Q, Fan S, Bae I et al. Induction of BAX by genotoxic stress in human cells correlates with
normal p53 status and apoptosis [published erratum appears in Oncogene 1995 Mar 16;10(6):1259].
Oncogene 1994; 9:3743-51.
103. McCurrach ME, Connor TM, Knudson CM et al. BAX-deficiency promotes drug resistance and
oncogenic transformation by attenuating p53-dependent apoptosis. Proc Natl Acad Sci USA 1997;
94:2345-9.
104. Deng Y, and Wu X, Peg3/Pw1 promotes p53-mediated apoptosis by inducing BAX translocation
from cytosol to mitochondria [In Process Citation]. Proc Natl Acad Sci USA 2000; 97:12050-5.
105 Relaix F, Wei X, Li W et al. Pw1/Peg3 is a potential cell death mediator and cooperates with
Siah1a in p53-mediated apoptosis Proc Natl Acad Sci USA 2000; 97:2105-10.
106. Budihardjo I, Oliver H Lutter, M et al. Biochemical pathways of caspase activation during apoptosis.
Annu Rev Cell Dev Biol 1999; 15:269-90.
107. Knudson, CM, Tung, KS, Tourtellotte, WG et al., BAX-deficient mice with lymphoid hyperplasia
and male germ cell death, Science 1995; 270:96-9.
108. Polyak K, Xia Y, Zweier,JL et al. A model for p53-induced apoptosis [see comments]. Nature
1997; 389:300-5.
109. Nagata S, Apoptosis by death factor. Cell 1997; 88:355-65.
110. Ashkenazi A, and Dixit VM, Death receptors: signaling and modulation. Science 1998; 281:1305-8.
111. Boldin MP, VARFolomeev EE, Pancer, Z et al. A novel protein that interacts with the death
domain of Fas/APO1 contains a sequence motif related to the death domain. J Biol Chem
1995; 270:7795-8.
112. Chinnaiyan AM, O’Rourke K, Tewari M et al. FADD, a novel death domain-containing protein,
interacts with the death domain of Fas and initiates apoptosis. Cell 1995; 81:505-12.
113. Marsters SA, Sheridan JP, Pitti RM et al. A novel receptor for Apo2L/TRAIL contains a truncated
death domain. Curr Biol 1997; 7:1003-6.
114. Owen-Schaub LB, Zhang W, Cusack JC et al. Wild-type human p53 and a temperature-sensitive
mutant induce Fas/APO-1 expression. Mol CeLl Biol 1995; 15:3032-40.
115. Muller M, Strand S, Hug H et al. Drug-induced apoptosis in hepatoma cells is mediated by the
CD95 (APO- 1/Fas) receptor/ligand system and involves activation of wild-type p53. J Clin Invest
1997; 99:403-13.
116. Tamura T, Aoyama N, Saya H et al. Induction of Fas-mediated apoptosis in p53-transfected human
colon carcinoma cells, Oncogene 1995; 11:1939-46.
194 Cell Cycle Checkpoints and Cancer
117. Wu GS, Burns TF, McDonald ER, 3rd et al. Induction of the TRAIL receptor KILLER/DR5 in
p53-dependent apoptosis but not growth arrest. Oncogene 1999; 18:6411-8.
118. Rajah R, Valentinis B, and Cohen P, Insulin-like growth factor (IGF)-binding protein-3 induces
apoptosis nd mediates the effects of transforming growth factor-beta1 on programmed cell death
through a p53- and IGF-independent mechanism. J Biol Chem 1997; 272:12181-8.
119. Israeli D, Tessler E, Haupt Y et al. A novel p53-inducible gene, PAG608, encodes a nuclear zinc
finger protein whose overexpression promotes apoptosis. Embo J 1997; 16:4384-92.
120. Attardi LD, Reczek EE, Cosmas C et al. PERP, an apoptosis-associated target of p53, is a novel
member of the PMP-22/gas3 family [published erratum appears in Genes Dev 2000 Jul
15;14(14):1835]. Genes Dev 2000; 14:704-18.
121. Scholl FA, McLoughlin P, Ehler E et al. DRAL is a p53-responsive gene whose four and a half LIM
domain protein product induces apoptosis [In Process Citation]. J Cell Biol 2000; 151:495-506.
122. Chan KK, Tsui SK, Ngai SM et al. Protein-protein interaction of FHL2, a LIM domain protein
preferentially expressed in human heart, with hCDC47. J Cell Biochem 2000; 76:499-508.
123. Muller JM, Isele U, Metzger E et al. FHL2, a novel tissue-specific coactivator of the androgen
receptor. Embo J 2000; 19:359-69.
124. Oda E, Ohki R, Murasawa H et al. Noxa, a BH3-only member of the Bcl-2 family and candidate
mediator of p53-induced apoptosis. Science 2000; 288: 1053-8.
125. Miyashita T, Harigai M, Hanada M et al. Identification of a p53-dependent negative response
element in the bcl- 2 gene. Cancer Res 1994; 54: 3131-5.
126. Murphy M, Hinman A, and Levine AJ, Wild-type p53 negatively regulates the expression of a
microtubule- associated protein. Genes Dev 1996; 10:2971-80.
127. Chao C, Saito S, Kang J et al. p53 transcriptional activity is essential for p53-dependent apoptosis
following DNA damage [In Process Citation]. Embo J 2000; 19:4967-75.
128. Attardi LD, Lowe SW, Brugarolas J et al. Transcriptional activation by p53, but not induction of
the p21 gene, is essential for oncogene-mediated apoptosis. Embo J 1996; 15:3702-12.
129. Sun XF, Johannsson O, Hakansson S et al. A novel p53 germline alteration identified in a late
onset breast cancer kindred. Oncogene 1996; 13:407-11.
130. Walker KK, and Levine AJ, Identification of a novel p53 functional domain that is necessary for
efficient growth suppression. Proc Natl Acad Sci U S A 1996; 93:15335-40.
131. Aurelio ON, Cajot JF, Hua ML et al. Germ-line-derived hinge domain p53 mutants have lost
apoptotic but not cell cycle arrest functions. Cancer Res 1998; 58:2190-5.
132. Naumovski L, and Cleary ML, The p53-binding protein 53BP2 also interacts with Bc12 and
impedes cell cycle progression at G2/M. Mol Cell Biol 1996; 16:3884-92.
133. Ding HF, Lin, YL, McGill, G et al., Essential role for caspase-8 in transcription-independent
apoptosis triggered by p53, J Biol Chem 2000; 275: 38905-11.
134. Thut CJ, Goodrich JA, and Tjian R, Repression of p53-mediated transcription by MDM2: A dual
mechanism. Genes Dev 1997; 11:1974-86.
135. Honda R, Tanaka H, and Yasuda H, Oncoprotein MDM2 is a ubiquitin ligase E3 for tumor
suppressor p53. FEBS Lett 1997; 420:25-7.
136. Honda R, and Yasuda H, Association of p19(ARF) with Mdm2 inhibits ubiquitin ligase activity of
Mdm2 for tumor suppressor p53. Embo J 1999; 18: 22-7.
137. Ashcroft M, Kubbutat MH, and Vousden KH, Regulation of p53 function and stability by
phosphorylation. Mol Cell Biol 1999; 19:1751-8.
138. Kubbutat, MH, and Vousden, KH, New HPV E6 binding proteins: dangerous liaisons?, Trends
Microbiol 1998; 6: 173-5.
139. Querido E, Marcellus RC, Lai, A et al. Regulation of p53 levels by the E1B 55-kilodalton protein
and E4orf6 in adenovirus-infected cells. J Virol 1997; 71: 3788-98.
140. de Stanchina E, McCurrach ME, Zindy F et al. E1A signaling to p53 involves the p19(ARF)
tumor suppressor. Genes Dev 1998; 12: 2434-42.
141. Zindy F, Eischen CM, Randle DH et al. Myc signaling via the ARF tumor suppressor regulates
p53-dependent apoptosis and immortalization. Genes Dev 1998; 12:2424-33.
142. Bates S, Phillips AC, Clark PA et al. p14ARF links the tumour suppressors RB and p53 [letter].
Nature 1998; 395: 124-5.
143. Dumaz N, and Meek DW, Serine15 phosphorylation stimulates p53 transactivation but does not
directly influence interaction with HDM2. Embo J 1999; 18: 7002-10.
144. Hupp TR, Meek DW, Midgley CA et al. Regulation of the specific DNA binding function of
p53. Cell 1992; 71: 875-86.
145. Shaulsky G, Goldfinger N, Tosky MS et al. Nuclear localization is essential for the activity of p53
protein. Oncogene 1991; 6: 2055-65.
P53, Apoptosis and Cancer Therapy 195
146. Sakaguchi K, Herrera JE, Saito S, et al. DNA damage activates p53 through a phosphorylationacetylation
cascade. Genes Dev 1998; 12:2831-41.
147. Chen X, Ko LJ, Jayaraman L et al. p53 levels, functional domains, and DNA damage determine
the extent of the apoptotic response of tumor cells, Genes Dev 1996; 10:2438-51.
148. Canman CE, Gilmer TM, Coutts SB et al. Growth factor modulation of p53-mediated growth
arrest versus apoptosis. Genes Dev 1995; 9:600-11.
149. Wu GS, Burns TF, McDonald ER et al. KILLER/DR5 is a DNA damage-inducible p53-regulated
death receptor gene. Nat Genet 1997; 17:141-3.
150. Wu GS, Kim K, and el-Deiry WS, KILLER/DR5: A novel DNA-damage inducible death receptor
gene, links the p53-tumor suppressor to caspase activation and apoptotic death. Adv Exp Med Biol
2000; 465:143-51.
151. Sheikh MS, and Fornace AJ, Death and decoy receptors and p53-mediated apoptosis. Leukemia
2000; 14:1509-13.
152. Meng RD, McDonald ER, Sheikh MS et al. The TRAIL decoy receptor TRUNDD (DcR2,
TRAIL-R4) is induced by adenovirus-p53 overexpression and can delay TRAIL-, p53-, and KILLER/
DR5-dependent colon cancer apoptosis, Mol Ther 2000; 1:130-44.
153. Fujiwara T, Grimm EA, Mukhopadhyay T et al. Induction of chemosensitivity in human lung
cancer cells in vivo by adenovirus-mediated transfer of the wild-type p53 gene. Cancer Res 1994;
54:2287-91.
154. Spitz FR, Nguyen D, Skibber JM et al. Adenoviral-mediated wild-type p53 gene expression sensitizes
colorectal cancer cells to ionizing radiation. Clin Cancer Res 1996; 2:1665-71.
155. Blagosklonny MV, and el-Deiry WS, In vitro evaluation of a p53-expressing adenovirus as an
anti-cancer drug. Int J Cancer 1996; 67:386-92.
156. Selivanova G, Kawasaki T, Ryabchenko L et al. Reactivation of mutant p53: a new strategy for
cancer therapy. Semin Cancer Biol 1998; 8:369-78.
157. Kim AL, Raffo AJ, Brandt-Rauf PW et al. Conformational and molecular basis for induction of
apoptosis by a p53 C-terminal peptide in human cancer cells. J Biol Chem 1999; 274:34924-31.
158. Pan H, and Griep AE, Altered cell cycle regulation in the lens of HPV-16 E6 or E7 transgenic
mice: implications for tumor suppressor gene function in development. Genes Dev 1994; 8: 1285-99.
159. Howes KA, Ransom N, Papermaster DS et al. Apoptosis or retinoblastoma: alternative fates of
photoreceptors expressing the HPV-16 E7 gene in the presence or absence of p53 [published erratum
appears in Genes Dev 1994 Jul 15;8(14):1738]. Genes Dev 1994; 8:1300-10.
160. Jorgensen GE, Johnsen, JI Ponthan F et al. Human polyomavirus BK (BKV) and neuroblastoma:
Mechanisms of oncogenic action and possible strategy for novel treatment [In Process Citation].
Med PediATR Oncol 2000; 35: 593-6.
161. Watanabe T, and Sullenger BA, Induction of wild-type p53 activity in human cancer cells by
ribozymes that repair mutant p53 transcripts. Proc Natl Acad Sci USA 2000; 97:8490-4.
162. Kastan MB, Zhan Q, el-Deiry WS et al. A mammalian cell cycle checkpoint pathway utilizing p53
and GADD45 is defective in ataxia-telangiectasia. Cell 1992; 71:587-97.
163. Westphal CH, Rowan S, Schmaltz C et al. ATM and p53 cooperate in apoptosis and suppression
of tumorigenesis, but not in resistance to acute radiation toxicity. Nat Genet 1997; 16:397-401.
164. Westphal CH, Schmaltz C, Rowan S et al. Genetic interactions between ATM and p53 influence
cellular proliferation and irradiation-induced cell cycle checkpoints. Cancer Res 1997; 57:1664-7.
165. Westphal CH, Hoyes KP, Canman CE et al. Loss of ATM radiosensitizes multiple p53 null tissues,
Cancer Res 1998; 58:5637-9.
166. Komarov PG, Komarova EA, Kondratov, RV et al. A chemical inhibitor of p53 that protects mice
from the side effects of cancer therapy. Science 1999; 285:1733-1737.
196 Cell Cycle Checkpoints and Cancer
CHAPTER 12
Cell Cycle Checkpoints and Cancer, edited by Mikhail V. Blagosklonny. ©2001 Eurekah.com.
Non-Apoptotic Responses to Anticancer
Agents: Mitotic Catastrophe, Senescence
and the Role of p53 and p21
Igor B. Roninson, Bey-Dih Chang and Eugenia V. Broude
Abstract
Chemotherapeutic drugs and radiation inhibit tumor cell proliferation by inducing growth
arrest and cell death. The best-studied antiproliferative response to anticancer agents
is programmed cell death or apoptosis. Inhibition of apoptosis, however, has been
frequently found to have little or no effect on clonogenic survival of human tumor cells. This
apparent paradox is explained by an increase in the fraction of damaged cells that undergo
senescence-like irreversible growth arrest or die through mitotic catastrophe (aberrant mitosis
that leads to the formation of multiple micronuclei) rather than apoptosis. This review
discusses the phenotypic changes and molecular mechanisms of mitotic catastrophe and
damage-induced senescence, and the role of these processes in the treatment response. It also
describes the roles of p53 and its downstream effector p21 as negative regulators of mitotic
catastrophe, positive regulators of senescence, and mediators of the paracrine effects of cellular
damage. Mitotic catastrophe and senescence provide promising targets for augmentation and
modulation in cancer treatment.
Can Apoptosis Account for Tumor Cell Response
to Anticancer Agents?
Over the past decade, the major effort in the study of cellular responses to chemotherapy
and radiation has been concentrated on apoptosis, the general physiological program of cell
suicide, which is induced by different cytotoxic agents in normal and tumor cells.1,2 The
significance of apoptosis in the antiproliferative effects of anticancer drugs and radiation has
been investigated by modulating the expression of apoptosis-regulating genes, such as apoptosis
inhibitor BC-L2, or p53 that usually acts as a positive regulator of apoptosis. The role of
apoptosis as a determinant of treatment response in solid tumors has been suggested in part by
the works of Lowe et al.3,4 These studies demonstrated that the knockout of p53 rendered
transformed mouse fibroblast cell lines more resistant to drugs and radiation, in vitro and in
vivo, and that this resistance was associated with a decreased induction of apoptosis in p53-/-
cells. Decreased induction of apoptosis was also associated with increased radiation and drug
resistance in normal mouse tissues5 and in murine lymphomas,6 as well as in human leukemia
and lymphoma cells in vitro.1 In contrast, studies with cell lines derived from human solid
tumors failed to produce an unambiguous correlation between the p53 status, propensity to
Mitotic Catastrophe, Senescence 197
undergo apoptosis and treatment sensitivity (see ref. 7 for a recent discussion). The role of
apoptosis in determining the treatment outcome was also questioned by a series of studies that
examined the effects of BCL2 on the response of human solid tumor cell lines to aphidicolin,8
etoposide9 and ionizing radiation.10,11 In all of these studies, ectopic overexpression of BCL2
inhibited the induction of apoptosis, but it had little or no effect on the clonogenic survival
of treated cells. In particular, Wouters et al11 found that BCL2 expression in a derivative
of HCT116 colon carcinoma cells strongly inhibited radiation-induced apoptosis
in vitro and in vivo, but apoptosis-resistant xenograft tumors turned out to be more
sensitive rather than more resistant to radiation treatment.
The lack of correlation between apoptosis and treatment survival can be understood, however,
if one considers that apoptosis is not the only mechanism of terminal proliferative death
in tumor cells. Two other known mechanisms, which are the subject of the present Chapter, are
mitotic catastrophe and senescence-like terminal proliferation arrest. The role of these nonapoptotic
mechanisms in treatment response is illustrated by an example from our recent study.12
This work investigated how the MDR1 P-glycoprotein, which blocks apoptosis through an as
yet undefined mechanism distinct from its well-known function as a multidrug
transporter,13 affects cellular responses to ionizing radiation. Figure 1 shows the time course of
morphological changes induced by 9 Gy of radiation in HeLa-derived HtTA cells with tetracycline-
regulated MDR1 expression. When MDR1 expression is turned off, morphologically detectable
apoptosis is the most prominent response to radiation in this cell line. Upon MDR1
induction, apoptosis is greatly decreased (Fig. 1a), but this decrease is followed by a
compensating increase in the fractions of cells that undergo mitotic catastrophe (Fig. 1b) or
senescence (Fig. 1c). As a result, MDR1 induction has only a minimal effect on clonogenic
survival after radiation (Fig. 1d). Similar results were obtained through radiation dose
response analysis and with another cell line.12 Thus, although the damaged tumor cells may be
dying through apoptosis, a functional apoptotic program is not required for their proliferative
death, which in the absence of apoptosis occurs through the processes of senescence and
mitotic catastrophe.
Mitotic Catastrophe
Mitotic cell death, a.k.a. mitotic catastrophe, has been observed in cells treated with all
the major classes of anticancer drugs or radiation2,14 when such cells enter mitosis with unrepaired
DNA damage or in the presence of mitotic spindle-disrupting agents. The resulting mitosis is
aberrant and incomplete. Depending on the cell line and the nature of damage, the process of
mitosis may terminate at the stage of abnormal metaphase15 or at abnormal anaphase or late
prophase.16,17 This abortive mitosis does not produce proper chromosome segregation and cell
division but results in the formation of large non-viable cells with two or more micronuclei
that are completely or partially separated from each other (Fig. 2a). In the case of partial separation,
the cell displays a large multilobulated nucleus. Micronuclei arise through the
formation of multiple nuclear envelopes around chromatin clusters,18 with chromosomes
showing random distribution among the micronuclei.14,19 Micronucleated cells arising from
mitotic catastrophe can be easily distinguished from apoptotic cells by their morphology
(Fig. 2a). While apoptotic cells may also have fragmented nuclei, they are characterized by
shrunken cytoplasm and condensed chromatin, whereas micronucleated cells are large and
contain uncondensed chromosomes. Cells that undergo mitotic catastrophe do not usually
show DNA ladder formation or DNA breaks that are detectable by TUNEL staining in
apoptotic cells.
Mitotic catastrophe has been studied most extensively in the field of ionizing radiation,
where it has been characterized as the main form of radiation-induced cell death.2 Aberrant
mitosis and micronucleation have been observed in irradiated cells both in vitro and in vivo;20
in fact, micronuclei formation is a common indicator of radiation damage.21 Mitotic death was
also identified as a prominent response to different anticancer drugs, such as etoposide, Taxol,
198 Cell Cycle Checkpoints and Cancer
cisplatin or bleomycin.9,22,23 In our study,14 treatment of HT1080 fibrosarcoma cells with equitoxic
(ID85) doses of seven different chemotherapeutic drugs and ionizing radiation invariably induced
mitotic catastrophe in a high fraction (45-66%) of treated cells. Furthermore, moderate doses
of doxorubicin induced mitotic catastrophe in 11 other solid tumor cell lines, including those
that were resistant to apoptosis.14 Mitotic catastrophe and apoptosis often arise in the same
population of irradiated cells; as a general trend, lower doses of cytotoxins preferentially induce
mitotic catastrophe while higher doses induce apoptosis.22,23 The onset of mitotic catastrophe
may be followed in some cases by the activation of the apoptotic program.24,19,15 For this
reason, mitotic catastrophe is viewed sometimes as an early stage of apoptosis. The apoptotic
program, however, is not needed for the lethal effect of mitotic catastrophe. This was shown
not only in the above-described work of Ruth and Roninson12 (Fig. 1), but also in an earlier
study by Lock and Stribinskiene.9 When etoposide-induced apoptosis of HeLa cells was
suppressed by BCL2, drug-treated tumor cells died through mitotic catastrophe, and there was
no significant change in the clonogenic survival.9
The mechanisms that lead to mitotic abnormalities in cells that enter mitosis without
repairing the damage are still poorly understood. In yeast cells, mitotic catastrophe was shown
to result from premature cell entry into mitosis before the completion of DNA replication.25
The notion of mitotic catastrophe as premature mitosis has also been promulgated by the
analysis of human tumor cells treated with hyperthermia18 or ionizing radiation.16 In these
studies, mitotic catastrophe has been described as starting with uneven chromatin condensation
around nucleoli, which resembled premature chromosome condensation originally described
in the fusions of interphase cells (in S or G2) with mitotic cells.26 Based on this similarity
and on the observations of increased cellular levels of cyclin B1 and elevated Cdc2 kinase
activity at the onset of mitotic catastrophe, this process was proposed to represent premature
induction of mitosis in cells that are damaged in S or G2.16 Uneven chromosome condensation,
however, may result not only from PCC but also from localized defects in the assembly of
chromosome-condensing proteins (condensins and cohesins) in the damaged regions of DNA.
It is still unclear therefore if mitotic catastrophe in mammalian cells represents premature mitosis.
Another key feature of mitotic catastrophe is the amplification of centrosomes in cells
with fragmented or multilobulated nuclei.17,27 Centrosome amplification leads to multipolar
mitosis, which appears to be the most common type of mitotic abnormality in drug-treated
tumor cells.28 The role of the centrosome as a potentially critical target in mitotic catastrophe is
Fig. 1. Effects of MDR1 on radiation-induced apoptosis (A), mitotic catastrophe (B), senescence (C) and
clonogenic survival (D) (reproduced with permission from: Ruth, A., Roninson, I.B. Cancer Res. 2000;
60:2576-2579).
HtTA-MDR1 cells that express MDR1 from a tetracycline-inhibited promoter were irradiated with a single
dose or different doses of radiation. The fractions of cells that undergo apoptosis, mitotic catastrophe or
senescence were determined by morphological assays and changes in cell proliferation were analyzed by
clonogenic assays. Left to right: Time course of changes in the percentages of apoptotic, micronucleated and
SA-β-gal+ after irradiation with 9 Gy, and clonogenic survival of different doses of radiation. MDR1 was
turned off (•) or turned on (o).
Mitotic Catastrophe, Senescence 199
suggested by a study of Sibon et al.29 These authors found that treatment of Drosophila
embryos with several DNA-damaging agents resulted in a loss of γ-tubulin from centrosomes,
anastral spindle formation and failure of chromosome segregation. Since centrosome duplication
occurs prior to mitosis, this process, may represent an early event of mitotic catastrophe.
Mitotic catastrophe can be induced not only by DNA damage but also by genetic means,
through untimely activation of the cyclinB/Cdc2 kinase complex. Ectopic coexpression of cyclin
B1 and Cdc2, cyclin A and Cdc2, and cyclin B1 and Cdc25C (an enzyme that dephosphorylates
and activates Cdc2) were shown to induce premature mitosis and subsequent mitotic catastrophe
in mammalian cells.30 Mitotic catastrophe was also produced by perturbing the expression of
genes that control early stages of mitosis, such as RCC1 involved in chromosome condensation31
and centrosome-associated proteins NuMA (nuclear-mitotic apparatus)32 and survivin.33
Survivin is best known as an inhibitor of apoptosis, and the loss or displacement of survivin
from centrosomes may potentially account for the induction of apoptosis in cells undergoing
mitotic catastrophe. These observations outline a general picture of mitotic catastrophe as a
process that occurs either when a cell initiates mitosis prematurely, before all the required
mitosis-control mechanisms are in place, or when the proper course of mitosis is hindered by a
deficiency in some of these control mechanisms.
p53 as a Negative Regulator of Mitotic Catastrophe
The entry of damaged cells into mitosis is delayed by growth arrest at the G2 checkpoint,
and this checkpoint was shown to play a key role in preventing mitotic catastrophe. Agents that
abrogate the G2 checkpoint, such as caffeine, okadaic acid or staurosporine, promote mitotic
death in irradiated or drug-treated cells.34-36 Mitotic catastrophe is also enhanced by genetic
inactivation of proteins that control the G2/M checkpoint. As demonstrated by a number of
studies, p53 is an essential regulator of the G2/M checkpoint in human cells.17,37 The first p53
target gene that was shown to play a role in this checkpoint is p21WAF1/Cip1/Sdi1, a pleiotropic
inhibitor of different cyclin/CDK complexes and a regulator of several transcription factors
and cofactors.38 The role of p21 in G2 arrest was demonstrated by the observations that p21
knockout in human HCT116 colon carcinoma cells greatly diminishes damage-induced G2
arrest,17 that p21-/- mouse fibroblasts show more rapid entry into mitosis,39 and that ectopic
overexpression of p21 induces growth arrest in both G1 and G2.40 In addition to p21, another
p53-inducible protein, 14-3-3-σ, was shown to play a role in G2 arrest of damaged cells by
maintaining cytoplasmic sequestration of cyclin B1 and Cdc2.41 The knockout of p53, p21 or
14-3-3-σ in HCT116 cells drastically increased the induction of morphologically detectable
Fig. 2. Morphology of mitotic catastrophe (A) and senescence (B) (reproduced with permission from: Chang
BD et al. 1999; Oncogene 18:4808-4818 and Chang BD et al. 1999; Cancer Res 59:3761-3767). A.Cells with
multiple micronuclei (mn) and apoptotic cells (a) in doxorubicin-treated HCT116 colon carcinoma cells.
B.Expression of the SA-β-gal marker of senescence in HT1080 fibrosarcoma cells, untreated (left), treated
with 50 nM doxorubicin (middle), or following ectopic expression of p21 (right).5
200 Cell Cycle Checkpoints and Cancer
cell death by doxorubicin or radiation,17,19,41 including both mitotic catastrophe and apoptosis.42
In contrast, the induction of cell death by 5-fluorouracyl was decreased by p53 knockout in the
same cells.43 These opposite effects of p53 on apoptosis can be explained by contrasting the
function of p53 as a positive regulator of the primary (mitosis-independent) apoptotic response
with its role as a negative regulator of mitotic cell death and apoptosis that develops as
a consequence of mitotic catastrophe. These opposite roles of p53 in drug-and radiation-induced
cell death, together with its role in different forms of damage-induced growth arrest (see
below), are illustrated in Figure 3.
Induction of Senescence by DNA-Damaging Agents
Exposure of tumor cells to chemotherapeutic drugs or radiation induces not only mitotic
death and apoptosis but also cell cycle arrest, which is mediated in part by a direct effect of the
drugs (e.g., G2 arrest by topoisomerase II inhibitors) and in part by the activation of cellular
G1 and G2 checkpoints. Damage-induced growth arrest may be reversible, allowing the cells
to repair the damage and reenter the cycle, or irreversible, with the damaged cells unable to
divide but remaining physiologically functional. Studies on normal fibroblasts that underwent
DNA damage showed a striking phenotypic resemblance between irreversibly arrested cells and
cells that undergo replicative senescence, the normal physiological program of terminal growth
arrest.44,45 The phenotypic markers of senescence include enlarged and flattened morphology,
increased granularity, and expression of a relatively specific marker of senescence, senescenceassociated
β-galactosidase (SA-β-gal), which is detectable by X-gal staining at pH6.0 (ref. 46)
(Fig. 2b). The rapid senescence-like growth arrest which can be induced not only by DNAdamaging
agents but also by mutant RAS47 has been termed accelerated senescence. Accelerated
senescence, like apoptosis, was proposed to be a programmed protective response of the
organism to potentially carcinogenic damage.48 In contrast to replicative senescence, a slow
process that is mediated by telomere shortening,49,50 accelerated senescence does not appear to
involve changes in the telomere length,51-53 and it is not prevented by constitutive activation of
telomerase.54 However, the programs that execute terminal growth arrest in replicative and
accelerated senescence of normal cells appear to be identical. Both telomere shortening and
DNA damage activate p53, which then induces p21 that arrests the cell cycle in G1 and G2.
p21 activation in senescent cells (which can also be induced by p53-independent mechanisms)
is transient, and it is followed by stable activation of another CDK inhibitor, p16INK4A, which
maintains the growth arrest of senescent cells after the shutoff of p21.55,56 p53 and p16 (but
not p21) are two of the most commonly mutated tumor suppressor genes in human cancer,57
and it has long been thought that tumor cells, which are often deficient in one or both of these
genes, can no longer undergo senescence.
Over the past several years, it has become apparent, however, that immortal tumor-derived
cell lines, which tend to have short telomeres and to carry mutations that trigger accelerated
senescence in normal cells (such as oncogenic RAS mutations), can be forced into senescence.
This was first suggested by the reports that a senescent phenotype could be induced in tumor
cells by ectopic overexpression of tumor suppressor genes, such as p53, RB, p16 or p21.42,58-63
Perhaps more significantly, inhibition of papillomavirus oncoproteins E6 and E7 in cervical
carcinoma cell lines was recently shown to induce rapid senescence in almost 100% of tumor
cells.52 The latter result suggested that tumor cells may be “primed” to senesce once senescencerestraining
mechanisms (in this case E6 and E7) have been removed.52 These findings raised a
prospect that the enhancement of the extant program of accelerated senescence in tumor cells
may be a feasible and biologically justified approach to cancer therapy.
We and others have found that accelerated senescence can be readily induced in a sizable
fraction of cells in different tumor cell lines after treatment with different chemotherapeutic
drugs, ionizing radiation, or differentiating agents.14,53 A phenotype characterized by senescencespecific
morphological changes and SA-β-gal expression (Fig. 2b) was induced by cytotoxic
agents in a dose-dependent manner, along with mitotic catastrophe, and its induction was
Mitotic Catastrophe, Senescence 201
detectable even at the lowest drug doses.14 The senescent phenotype was also
efficiently induced in breast carcinoma cells that were treated with retinoids, in vitro or in vivo,
under the conditions of minimal cytotoxicity. Using a FACS-based procedure that separates
proliferating from growth-inhibited cells, we have found that SA-β-gal expression and the
senescent morphology distinguish drug-treated cells that are physically intact but are no longer
clonogenic from the cells that recover and form colonies after drug treatment.14 This result
indicates that detection of senescence markers in treated tumors may be useful as a diagnostic
procedure indicative of treatment response.
While the senescent phenotype is associated with a loss of clonogenicity, the mechanisms
of the clonogenic failure of senescent cells differ among tumor cell lines (see Fig. 3). In some
lines, such as HCT116, the senescent cells do not divide and maintain permanent cell cycle
arrest, despite the lack of p16.19,42 The nature of the genes that maintain permanent growth
arrest in such cells remains to be determined. In another class of tumor cell lines, typified by
HT1080 fibrosarcoma, most of the cells that develop the senescent phenotype eventually reenter
the cycle, but they stop dividing or die after 1-3 cell divisions.14 The proliferative failure in
these cells is accompanied by an increase in DNA ploidy14 and numerous mitotic abnormalities
(our unpublished data). These events closely parallel our observations in HT1080 cells where
growth arrest and the senescent phenotype were produced by expressing p21 from a regulated
promoter42 (see below).
The induction of the senescent phenotype by anticancer agents has been observed not
only in vitro but also in vivo, in xenograft tumors that were stained for SA-β-gal after treatment
with different drugs (ref. 14 and our unpublished data). While it is difficult to discern the
effects of accelerated senescence from the more rapid cytotoxic action of chemotherapy or
radiation in the clinics, some observations in radiation therapy suggest that the induction of
permanent cytostatic arrest could be the primary mode of treatment response in certain clinical
cases. In particular, complete regression of prostate cancers was reported in some patients to
take more than a year64 and of desmoid tumors up to two years65 after radiation treatment.
Stimulation of senescence in normal tissues may also represent an adverse clinical effect of
cancer therapy. For example, radiation-induced damage to normal tissues frequently develops late
after treatment, and such late effects are considered the dose-limiting factor in radiation therapy.66
Fig. 3. Cellular responses to DNA damage and the effects of p53. Thick arrows and boldface indicate the
events that are positively regulated by p53, and thin arrows and regular fonts indicate events that are
negatively regulated by p53.
202 Cell Cycle Checkpoints and Cancer
Role of p53 and p21 in Damage-Induced Senescence
and Abnormal Mitosis
As mentioned above, two of the key regulators of senescence in normal cells, p53 and p16,
are frequently inactivated in cancers. Moderate doses of doxorubicin, however, induced the
senescent phenotype in all tested cell lines with wild-type p53 and in one half of p53-deficient
lines; this response was also readily observed in p16-deficient tumor cells.14 To investigate the
role of p53 and p21 in the induction of senescence-like growth arrest, we have analyzed HT1080
fibrosarcoma cells where p53 function and p21 expression were blocked by a p53-derived
genetic suppressor element, and HCT116 colon carcinoma cells with homozygous knockout
of p53 or p21 (both HT1080 and HCT116 lines are p16-deficient and carry different RAS
mutations). In both cell lines, the inhibition or knockout of p53 or p21 strongly decreased but
did not completely abolish drug- or radiation-induced senescence. These results have indicated
that p53 and p21 act as positive regulators of accelerated senescence in tumor cells, but they are
not absolutely required for this response.42 The opposite roles of p21 as a positive regulator of
damage-induced senescence and a negative regulator of mitotic catastrophe help to explain the
paradoxical results of Waldman et al67 and Wouters et al,68 who found that the wild-type and
p21-/- HCT116 cell lines form a similar number of colonies after radiation or doxorubicin
treatment, despite the much higher degree of cell death observed after treatment in p21-/- cells.
The role of p53 as a positive regulator of both reversible and irreversible growth arrest (senescence)
is illustrated in Figure 3.
To investigate the role of p21 in drug-induced senescence, we have analyzed the effects of
p21 expression from a promoter inducible by a physiologically neutral β-galactoside IPTG in
HT1080 fibrosarcoma cells. p21 induction in these cells leads to growth arrest in both G1 and
G2, accompanied by markers of senescence (Fig. 2b). Concurrent inhibition of p53 in these
cells had no significant effect on the induction of the senescent phenotype by p21, indicating
that p21 function was sufficient for this phenotype. On the other hand, comparison of the
percentages of senescent cells and p21 protein levels between IPTG-induced and doxorubicintreated
cells showed that p21 induction was insufficient to account for the full extent of druginduced
senescent response. At the same level of senescence, IPTG-induced cells expressed
four times more p21 than did doxorubicin-treated cells, indicating that some other factors in
drug-treated cells cooperated with p21 in producing the senescent phenotype.42
The development of the senescent phenotype in HT1080 cells with IPTG-inducible p21
expression was accompanied by gradual loss of cellular clonogenic capacity upon release from
p21. This loss of clonogenicity correlated with the extent and duration of p21 induction, and
it was not due to a failure of the cells to reenter the cycle upon release from p21. When p21 was
induced by IPTG for three or more days and then turned off, all the cells reentered the cycle
but most of them either died (with features of mitotic catastrophe) or underwent terminal
arrest after a few cell divisions and failed to form colonies. This proliferative failure was associated
with the predominance of abnormal mitotic figures during the first cell division after release
from p21. We have found that this abnormal mitosis was associated with an imbalance in the
intracellular pools of mitosis-initiating proteins (such as cyclin B1, Cdc2, or polo-like kinase
(Plk1)) and mitosis-controlling proteins (such as spindle checkpoint control proteins Mad2 or
BubR1 or aurora kinase family of proteins that control centrosome function).
Transcription of the genes encoding such proteins was inhibited upon the induction of p21,
and the intracellular pools of these proteins decayed after two or more days of p21 induction.
Once p21 was turned off, these proteins were resynthesized asynchronously. As a result, cells
acquired sufficient amounts of mitosis-initiating proteins and entered mitosis with only a minimal
level of Mad2 and possibly other mitosis-control proteins, resulting in abnormal mitosis.69 It
remains to be determined whether the same or a similar mechanism is responsible for the abnormal
mitosis that occurs in drug-treated cells that reenter the cycle after prolonged growth arrest.
Mitotic Catastrophe, Senescence 203
Paracrine Activities of Senescent Cells: Implications for Treatment
Outcome and Side Effects of Cancer Therapy
Terminal proliferative arrest is not the only aspect of accelerated senescence that has
important clinical implications. Senescent cells, while not dividing, are not physiologically
neutral and produce secreted proteins with clinically important functions. Many far-ranging
paracrine activities have been associated with normal senescent cells.50,70,71 Some of the
proteins overexpressed in senescent cells are proteases and protease inhibitors, which have
tissue-reorganizing effects and may promote metastasis, whereas others are paracrine growth
factors that affect the growth of non-senescent tumor cells or de novo carcinogenesis. 50,70,71
The same tumor-promoting activities have also been associated with tumor-derived stromal
fibroblasts.72 The tumor-promoting effect of the stromal tissue was recently shown in a mouse
mammary carcinogenesis model to be induced by ionizing radiation.73 Induction of mitogenic
and angiogenic factors by ionizing radiation was also detected by gene expression profiling
studies.74
One of the strongest effects of ionizing radiation in mammary stromal fibroblasts75 and
many other cell types is the induction of p21. As discussed above, p21 induction is an essential
component of cell senescence. While the best-known function of p21 is the inhibition of cyclin/
cyclin-dependent kinase (CDK) complexes, this protein also binds and modulates the activity
of many transcription factors and cofactors.38 Through cDNA microarray analysis, we76 have
found that p21 induction in HT1080 fibrosarcoma cells upregulates the genes for many
secreted proteins, including senescence-associated extracellular matrix (ECM) proteins and
plaque-forming proteins that have been implicated in age-related amyloid diseases (e.g.,
Alzheimer’s amyloid precursor protein, serum amyloid A, tissue transglutaminase). p21 also
upregulates the expression of p66Shc, a mediator of oxidative damage that was shown to be a
negative determinant of lifespan in mice.77 Furthermore, p21 induces several antiapoptotic,
mitogenic and angiogenic secreted factors, and conditioned media from p21-induced cells
shows antiapoptotic and mitogenic activities.76
While DNA damage induces the expression of growth-promoting factors, this effect may
be counterbalanced by the concurrent induction of secreted growth inhibitors, such as TNFα,
serine protease inhibitors or IGF-binding protein 3.74 The induction of both growth-promoting
and growth-inhibiting factors in damaged cells was found to depend on the presence of
wild-type p53, and ectopic overexpression of p53 induces transcription of such genes.74 The
induction of growth-promoting factors by p53 is likely to be mediated at least in part through
p21. p21, however, was not found to upregulate any growth-inhibitory factors,76 suggesting
that the induction of secreted growth inhibitors is mediated by other downstream effectors of p53.
Conditioned media from irradiated or p53-induced cells was shown to be growth-inhibitory,74 in
contrast to the growth promoting effect of the media conditioned by p21-induced cells.76 It
remains to be determined whether secreted growth-inhibiting factors are expressed by senescent
cells along with the growth-stimulating factors. The diverse paracrine effects
associated with drug-induced senescence may have important consequences for the long-term
outcome and side effects of cancer therapy.
Mitotic Catastrophe and Senescence as Target Responses
in Cancer Treatment
The above described non-apoptotic responses to cancer therapeutic agents, i.e., mitotic
catastrophe and senescence, provide promising directions of research into improving the efficacy
of cancer treatment. Apoptosis is a powerful control mechanism that eliminates normal cells
with potential oncogenic changes, and therefore mutations that disable the program of apoptosis
are commonly selected in the course of carcinogenesis. These mutations include those that
inactivate p53, a positive regulator of the primary (mitosis-independent) apoptotic response.
In normal cells, however, the program of apoptosis is fully functional, and it accounts largely
for treatment-induced damage to normal tissues. Inhibition of p53 and p53-mediated apoptotic
204 Cell Cycle Checkpoints and Cancer
program has been shown to provide a striking increase in the survival of irradiated animals.5 In
contrast to apoptosis, mitotic catastrophe is potentiated by p53 mutations and by the weakening
of G2 and mitotic checkpoints, which are characteristic of many human tumors.79,80 Tumor
cells therefore are likely to be more susceptible to mitotic catastrophe than normal cells.
Another potential advantage of mitotic cell death lies in the observations that mitotic catastrophe
in vivo leads to necrosis, with its associated inflammatory effects.2 The inflammatory response
may conceivably increase the efficacy of treatment, in contrast to the non-inflammatory nature
of apoptotic cell death. This difference may potentially explain a paradoxical observation of
Wouters et al11 that the inhibition of apoptosis in a BCL2-transduced derivative of HCT116
cells has made this cell line more sensitive rather than more resistant to radiation in vivo.
Elucidation of the molecular mechanisms of mitotic catastrophe may help to exploit this preferred
mode of tumor cell death in a therapeutic setting.
While cell senescence, like apoptosis, represents a protective mechanism of eliminating
cells that have experienced potentially carcinogenic events,48 studies reviewed in the present
Chapter have demonstrated that many and perhaps most tumor cells have retained the ability
to undergo senescence. The program of senescence can be activated in tumor cells either by
inhibiting senescence-overriding proteins (such as E6 and E7 oncoproteins of papilloma virus),
52 or by treatment with chemotherapeutic drugs, radiation, or “differentiating” agents.14
The induction of tumor senescence is a significant antiproliferative effect of anticancer agents,
and its augmentation may increase the efficacy of cancer therapy. On the other hand, the
senescent phenotype of tumor or normal cells is accompanied by paracrine side effects that
may interfere with the success of treatment and increase its toxicity to normal tissues. For
example, secretion of antiapoptotic/mitogenic factors or proteases may promote tumor growth
and metastasis, and increased expression of amyloid proteins may contribute to the eventual
development of different amyloid-associated diseases. Elucidation of the molecular mechanisms
that are responsible for permanent growth arrest in drug-induced senescence and for the
induction of different factors secreted by senescent cells should make it possible to develop
strategies that would promote tumor cell senescence or inhibit its undesirable side effects.
References
1. Reed JC. Bcl-2 family proteins: regulators of apoptosis and chemoresistance in hematologic
malignancies. Semin Hematol 1997; 34:9-19.
2. Cohen-Jonathan E, Bernhard EJ, McKenna WG. How does radiation kill cells? Curr Opin Chem
Biol. 1999; 3(1):77-83.
3. Lowe SW, Ruley HE, Jacks T et al. p53-dependent apoptosis modulates the cytotoxicity of anticancer
agents. Cell 1993; 74:957-967.
4. Lowe SW, Bodis S, McClatchey A et al. p53 status and the efficacy of cancer therapy in vivo.
Science 1994; 266:807-810.
5. Komarov PG, Komarova EA, Kondratov RV et al. A chemical inhibitor of p53 that protects mice
from the side effects of cancer therapy. Science 1999; 285:1733-1737.
6. Schmitt CA, Rosenthal CT, Lowe SW. Genetic analysis of chemoresistance in primary murine
lymphomas. Nat Med 2000; 6:1029-1035.
7. Finkel E. Does cancer therapy trigger cell suicide? Science 1999; 286:2256-2258.
8. Yin DX, Schimke RT. BCL-2 expression delays drug-induced apoptosis but does not increase
clonogenic survival after drug treatment in HeLa cells. Cancer Res 1995; 55:4922-4928.
9. Lock RB, Stribinskiene L. Dual modes of death induced by etoposide in human epithelial tumor cells allow
Bcl-2 to inhibit apoptosis without affecting clonogenic survival. Cancer Res 1996; 56:4006-4012.
10. Kyprianou N, King ED, Bradbury D et al. Bcl-2 over-expression delays radiation-induced apoptosis without
affecting the clonogenic survival of human prostate cancer cells. Int J Cancer 1997; 70:341-348.
11. Wouters BG, Denko NC, Giaccia AJ et al. A p53 and apoptotic independent role for p21WAF1
in tumour response to radiation therapy. Oncogene 1999; 18:6540-6545.
12. Ruth AC, Roninson IB. Effects of the multidrug transporter P-glycoprotein on cellular responses
to ionizing radiation. Cancer Res 2000; 60:2576-2578.
13. Smyth MJ, KRasovskis E, Sutton VR et al. The drug efflux protein, P-glycoprotein, additionally
protects drug- resistant tumor cells from multiple forms of caspase-dependent apoptosis. Proc Natl
Acad Sci USA 1998; 95:7024-7029.
Mitotic Catastrophe, Senescence 205
14. Chang BD, Broude EV, Dokmanovic M et al. A senescence-like phenotype distinguishes tumor
cells that undergo terminal proliferation arrest after exposure to anticancer agents. Cancer Res 1999;
59:3761-3767.
15. Jordan MA, Wendell K, Gardiner S et al. Mitotic block induced in HeLa cells by low concentrations
of Paclitaxel (Taxol) results in abnormal mitotic exit and apoptotic cell death. Cancer Res
1996; 56:816-825.
16. Ianzini F, Mackey MA. Spontaneous premature chromosome condensation and mitotic catastrophe
following irradiation of HeLa S3 cells. Int J Radiat Biol 1997; 72:409-421.
17. Bunz F, Dutriaux A, Lengauer C et al. Requirement for p53 and p21 to sustain G2 arrest after
DNA damage. Science 1998; 282:1497-1501.
18. Swanson PE, Carroll SB, Zhang XF et al. Spontaneous premature chromosome condensation, micronucleus
formation, and non-apoptotic cell death in heated HeLa S3 cells. Ultrastructural observations.
Am J Pathol 1995; 146:963-971.
19. Waldman T, Lengauer C, Kinzler KW et al. Uncoupling of S phase and mitosis induced by anticancer
agents in cells lacking p21. Nature 1996; 381:713-716.
20. Falkvoll KH. The occurrence of apoptosis, abnormal mitoses, cells dying in mitosis and micronuclei
in a human melanoma xenograft exposed to single dose irradiation. Strahlenther
Onkol 1990; 166:487-492.
21. Muller WU, Nusse M, Miller BM et al. Micronuclei: A biological indicator of radiation damage.
Mutat Res 1996; 366:163-169.
22. Tounekti O, Pron G, Belehradek J, Jr. et al. Bleomycin, an apoptosis-mimetic drug that
induces two types of cell death depending on the number of molecules internalized. Cancer
Res 1993; 53:5462-5469.
23. Torres K, Horwitz SB. Mechanisms of Taxol-induced cell death are concentration dependent. Cancer
Res 1998; 58:3620-3626.
24. Demarcq C, Bunch RT, Creswell D et al. The role of cell cycle progression in cisplatin-induced
apoptosis in Chinese hamster ovary cells. Cell Growth Differ 1994; 5:983-993.
25. Russell P, Nurse P. Negative regulation of mitosis by wee1+, a gene encoding a protein kinase
homolog. Cell 1987; 49:559-567.
26. Johnson RT, Rao PN. Mammalian cell fusion: Induction of premature chromosome condensation
in interphase nuclei. Nature 1970; 226:717-722.
27. Sato N, Mizumoto K, Nakamura M et al. Radiation-induced centrosome overduplication and
multiple mitotic spindles in human tumor cells. Exp Cell Res 2000; 255:321-326.
28. Schimke RT, Kung A, Sherwood SS et al. Life, death and genomic change in perturbed cell cycles.
Philos Trans R Soc Lond B Biol Sci 1994; 345:311-317.
29. Sibon OC, Kelkar A, Lemstra W et al. DNA-replication/DNA-damage-dependent centrosome inactivation
in DROSOPHILA embryos. Nat Cell Biol 2000; 2:90-95.
30. Heald R, McLoughlin M, McKeon F. Human wee1 maintains mitotic timing by protecting the
nucleus from cytoplasmically activated Cdc2 kinase. Cell 1993; 74:463-474.
31. Ohtsubo M, Kai R, Furuno N et al. Isolation and characterization of the active cDNA of the
human cell cycle gene (RCC1) involved in the regulation of onset of chromosome condensation.
Genes Dev 1987; 1:585-593.
32. Compton DA, Luo C. Mutation of the predicted p34cdc2 phosphorylation sites in NuMA impair
the assembly of the mitotic spindle and block mitosis. J Cell Sci 1995; 108:621-633.
33. Li F, Ackermann EJ, Bennett CF et al. Pleiotropic cell-division defects and apoptosis induced by
interference with survivin function. Nat Cell Biol 1999; 1:461-466.
34. Yao SL, Akhtar AJ, McKenna KA et al. Selective radiosensitization of p53-deficient cells by caffeinemediated
activation of p34cdc2 kinase. Nat Med 1996; 2:1140-1143.
35. Yamashita K, Yasuda H, Pines J et al. Okadaic acid, a potent inhibitor of type 1 and type 2A
protein phosphatases, activates cdc2/H1 kinase and transiently induces a premature mitosis-like
state in BHK21 cells. EMBO J 1990; 9:4331-4338.
36. Yoshida M, Usui T, Tsujimura K et al. Biochemical differences between staurosporine-induced
apoptosis and premature mitosis. Exp Cell Res 1997; 232:225-239.
37. Agarwal ML, Agarwal A, Taylor WR et al. p53 controls both the G2/M and the G1 cell cycle
checkpoints and mediates reversible growth arrest in human fibroblasts. Proc Natl Acad Sci USA
1995; 92:8493-8497.
38. Dotto GP. p21(Waf1/Cip1): More than a break to the cell cycle? Biochim Biophys Acta 2000;
1471:M43-M56.
39. Dulic V, Stein GH, Far DF et al. Nuclear accumulation of p21CIP1 at the onset of mitosis: A
role at the G2/M-phase transition. Mol Cell Biol 1998; 18:546-557.
206 Cell Cycle Checkpoints and Cancer
40. Niculescu AB, III, Chen X, Smeets M et al. Effects of p21(Cip1/Waf1) at both the G1/S and the
G2/M cell cycle transitions: pRb is a critical determinant in blocking DNA replication and in
preventing endoreduplication Mol Cell Biol 1998; 18:629-643.
41. Chan TA, Hermeking H, Lengauer C et al. 14-3-3Sigma is required to prevent mitotic catastrophe
after DNA damage. Nature 1999; 401:616-620.
42. Chang BD, Xuan Y, Broude EV et al. Role of p53 and p21WAF1/cip1 in senescence-like terminal
proliferation arrest induced in human tumor cells by chemotherapeutic drugs. Oncogene
1999; 18:4808-4818.
43. Bunz F, Hwang PM, Torrance C et al. Disruption of p53 in human cancer cells alters the
responses to therapeutic agents. J Clin Invest 1999; 104:263-269.
44. Di Leonardo A, Linke SP, Clarkin K et al. DNA damage triggers a prolonged p53-dependent
G1 arrest and long-term induction of Cip1 in normal human fibroblasts. Genes Dev 1994;
8:2540-2551.
45. Robles SJ, Adami GR. Agents that cause DNA double strand breaks lead to p16INK4a enrichment
and the premature senescence of normal fibroblasts. Oncogene 1998; 16:1113-1123.
46. Dimri GP, Lee X, Basile G, et al. A biomarker that identifies senescent human cells in culture and
in aging skin in vivo. Proc Natl Acad Sci USA 1995; 92:9363-9367.
47. Serrano M, Lin AW, McCurrach ME et al. Oncogenic Ras provokes premature cell senescence
associated with accumulation of p53 and p16INK4a. Cell 1997; 88:593-602.
48. Weinberg RA. The cat and mouse games that genes, viruses, and cells play. Cell 1997; 88:573-575.
49. Shay JW. Telomerase in human development and cancer. J Cell Physiol 1997; 173:266-270.
50. Campisi J. Cancer, aging and cellular senescence. In Vivo 2000; 14:183-188.
51. Michishita E, Nakabayashi K, Ogino H et al. DNA topoisomerase inhibitors induce reversible
senescence in normal human fibroblasts. Biochem Biophys Res Commun 1998; 253:667-671.
52. Goodwin EC, Yang E, Lee CJ et al. Rapid induction of senescence in human cervical carcinoma
cells. Proc Natl Acad Sci USA 2000; 97:10978-10983.
53. Wang X, Wong SC, Pan J et al. Evidence of cisplatin-induced senescent-like growth arrest in
nasopharyngeal carcinoma cells. Cancer Res 1998; 58:5019-5022.
54. Morales CP, Holt SE, Ouellette M et al. Absence of cancer-associated changes in human fibroblasts
immortalized with telomerase. Nat Genet 1999; 21:115-118
55. Alcorta DA, Xiong Y, Phelps D et al. Involvement of the cyclin-dependent kinase inhibitor
p16 (INK4a) in replicative senescence of normal human fibroblasts. Proc Natl Acad Sci USA
1996; 93:13742-13747.
56. Stein GH, Drullinger LF, Soulard A et al. Differential roles for cyclin-dependent kinase inhibitors
p21 and p16 in the mechanisms of senescence and differentiation in human fibroblasts. Mol Cell
Biol 1999; 19:2109-2117.
57. Hall M, Peters G. Genetic alterations of cyclins, cyclin-dependent kinases, and CDK inhibitors in
human cancer. Adv Cancer Res 1996; 68:67-108.
58. Sugrue MM, Shin DY, Lee SW et al. Wild-type p53 triggers a rapid senescence program in human
tumor cells lacking functional p53. Proc Natl Acad Sci USA 1997; 94:9648-9653.
59. Xu HJ, Zhou Y, Ji W et al. Reexpression of the retinoblastoma protein in tumor cells induces
senescence and telomerase inhibition. Oncogene 1997; 15:2589-2596.
60. Uhrbom L, Nister M, Westermark B. Induction of senescence in human malignant glioma cells by
p16INK4A. Oncogene 1997; 15:505-514.
61. Vogt M, Haggblom C, Yeargin J et al.Independent induction of senescence by p16INK4a and
p21CIP1 in spontaneously immortalized human fibroblasts. Cell Growth Differ 1998; 9:139-146.
62. Fang L, Igarashi M, Leung et al. p21WAF1/Cip1/Sdi1 induces permanent growth arrest with markers of
replicative senescence in human tumor cells lacking functional p53. Oncogene 1999; 18:2789-2797.
63. Dai CY, Enders GH. p16 INK4a can initiate an autonomous senescence program. Oncogene
2000; 19:1613-1622.
64. Cox JD, Kline RW. Do prostatic biopsies 12 months or more after external irradiation for adenocarcinoma,
Stage III, predict long-term survival? Int J Radiat Oncol Biol Phys 1983; 9:299-303.
65. Bataini JP, Belloir C, Mazabraud A et al. Desmoid tumors in adults: the role of radiotherapy in
their management. Am J Surg 1988; 155:754-760.
66. Hellman S. Principles of cancer management: Radiation therapy. In: V.T.DeVita SH, S.A.Rosenberg,
editors. Cancer, Principles & Practice of Oncology. Philadelphia: Lippincott-Raven, 1997: 307-332.
67. Waldman T, Zhang Y, Dillehay L et al. Cell-cycle arrest versus cell death in cancer therapy. Nat
Med 1997; 3:1034-1036.
68. Wouters BG, Giaccia AJ, Denko NC et al. Loss of p21WAF1/Cip1 sensitizes tumors to radiation
by an apoptosis- independent mechanism. Cancer Res 1997; 57:4703-4706.
Mitotic Catastrophe, Senescence 207
69. Chang BD, Broude EV, Fang J et al. p21WAF1/Cip1/Sdi1-induced growth arrest is associated
with depletion of mitosis-control proteins and leads to abnormal mitosis and endoreduplication in
recovering cells. Oncogene 2000; 19:2165-2170.
70. Campisi J. The biology of replicative senescence. Eur J Cancer 1997; 33:703-709.
71. Campisi J. The role of cellular senescence in skin aging. J Investig Dermatol Symp Proc 1998; 3:1-5.
72. Olumi AF, Grossfeld GD, Hayward SW et al. Carcinoma-associated fibroblasts direct tumor
progression of initiated human prostatic epithelium. Cancer Res 1999; 59:5002-5011.
73. Barcellos-Hoff MH, Ravani SA. Irradiated mammary gland stroma promotes the expression of
tumorigenic potential by unirradiated epithelial cells. Cancer Res 2000; 60:1254-1260.
74. Komarova EA, Diatchenko L, Rokhlin OW et al. Stress-induced secretion of growth inhibitors: A
novel tumor suppressor function of p53. Oncogene 1998; 17:1089-1096.
75. Meyer KM, Hess SM, Tlsty TD et al. Human mammary epithelial cells exhibit a differential p53-mediated
response following exposure to ionizing radiation or UV light. Oncogene 1999; 18:5795-5805.
76. Chang BD, Watanabe K, Broude EV et al. Effects of p21WAF1/Cip1/Sdi1 on cellular gene expression:
implications for carcinogenesis, senescence, and age-related diseases. Proc Natl Acad Sci
USA 2000; 97:4291-4296.
77. Migliaccio E, Giorgio M, Mele S et al. The p66shc adaptor protein controls oxidative stress
response and life span in mammals. Nature 1999; 402:309-313.
78. Hallahan DE, Spriggs DR, Beckett MA et al. Increased tumor necrosis factor alpha mRNA after
cellular exposure to ionizing radiation. Proc Natl Acad Sci USA 1989; 86:10104-10107.
79. O’Connor PM. Mammalian G1 and G2 phase checkpoints. Cancer Surv 1997; 29:151-182.
80. Cahill DP, Lengauer C, Yu J et al. Mutations of mitotic checkpoint genes in human cancers.
Nature 1998; 392:300-303.
208 Cell Cycle Checkpoints and Cancer
CHAPTER 13
Cell Cycle Checkpoints and Cancer, edited by Mikhail V. Blagosklonny. ©2001 Eurekah.com.
Small Molecule Inhibitors
of Cyclin-Dependent Kinases
Geoffrey I. Shapiro
Introduction
Cyclin-dependent kinases (CDKs) are core components of the cell cycle machinery.Orderly
transition between cell cycle phases requires the scheduled activity of the
CDKs, governed in part by their associations with cyclins and CDK inhibitors, as well
as by their state of phosphorylation. In malignant cells, altered expression of CDKs and their
modulators, including overexpression of cyclins and loss of expression of CDK inhibitors,
results in deregulated CDK activity, providing a selective growth advantage. Because of their
critical role in cell cycle progression, as well as the association of their activities with the processes
of differentiation and apoptosis, the CDKs comprise an attractive set of targets for novel
antineoplastics.
Multiple strategies can be used to modulate cellular CDK activity, including the genetic
replacement of endogenous CDK inhibitors or the introduction of small peptides that reproduce
their effects. In addition, several known anticancer agents cause reduction in expression of
D-type cyclins, and other drugs under development reduce CDK stability or inactivate the
proteasome-mediated degradation of endogenous CDK inhibitors (reviewed in refs. 1-3). This
Chapter will focus on small molecules that directly modulate CDK activity.
The classes of small molecule CDK inhibitors are shown in Table 1. In addition to their
chemical categorization, they can be further classified based on their overall specificity for
CDKs and their ability to inhibit CDK4.4-6 Most of these drugs inhibit tumor growth in
preclinical models by inducing cell cycle arrest or apoptosis. Several have profound effects on
cellular transcription or have been shown to cause differentiation. To date, flavopiridol and
UCN-01 have advanced the furthest in clinical application, and both have demonstrated
significant synergy with standard chemotherapy drugs. Although not all of the biologic effects of
flavopiridol have been definitively linked to CDK modulation, it serves as a model for the potential
antitumor activity of these agents. In addition, preclinical data exist for new generations of purine
analogs as well as newer classes of CDK inhibitors, such as the paullones. As more potent and
selective CDK inhibitors are now eagerly anticipated, it is useful to review the preclinical and
clinical results with the agents presently under development.
Flavopiridol
Antiproliferative Mechanisms
Early Studies
Flavopiridol, also known as L86-8275 or HMR 1275, is the first CDK inhibitor to enter
clinical trials. The drug is a semi-synthetic flavonoid derived from rohitukine, an alkaloid
Small Molecule Inhibitors of Cyclin-Dependent Kinases 209
isolated from the stem bark of Dysoxylum binectariferum, a plant native to India. Flavopiridol
was initially selected for further study because it demonstrated modest inhibition of the epidermal
growth factor receptor (EGFR) tyrosine kinase (IC50 = 21 μM), relatively specific
compared to its inhibition of other kinases, such as protein kinase A (IC50 = 122 μM).7
However, early studies showed that flavopiridol inhibited the proliferation of exponentially
growing human breast and lung carcinoma cells at concentrations as low as 25-160 nM.8 In
addition, the mean IC50 across the 60-cell line NCI anticancer drug screen panel was 66 nM.
Furthermore, antiproliferative effects were independent of EGFR status, indicating alternative
cellular targets.
Cell Cycle Arrest
Flavopiridol was subsequently shown to block cell cycle progression. For example, when
MDA-MB-468 breast cancer cells are released from an aphidicolin-induced block at the G1/S
boundary in the presence of flavopiridol, they ultimately arrest in G2. When the same cells are
released from a nocodazole-induced block in the presence of flavopiridol, they complete mitosis
but arrest in G1,8 with an accumulation of the hypophosphorylated form of the retinoblastomasusceptibility
gene product, Rb.9 Consistent with G1 and G2 arrest, in studies using purified
CDKs, flavopiridol has been shown to directly inhibit the activities of CDK1 (cdc2), CDK2
and CDK4, with IC50 values in the 40-400 nM range.7,9,10 Recently, inhibition of CDK6
has also been demonstrated.11 Inhibition is competitively blocked by ATP, with a Ki
range of 41-65 nM. 9,10
The biochemical data are supported by structural studies, in which the deschloro derivative
of flavopiridol has been co-crystallized with CDK2. This work demonstrated that the
aromatic portion of flavopiridol binds to the ATP-binding pocket of CDK2 with good
complementarity, as the size of buried surfaces of the inhibitor and the target kinase are similar
upon complex formation.12 In addition, the phenyl group of flavopiridol allows contact points
with the enzyme not seen in the CDK2-ATP complex. The stretch of amino acids contacting
the phenyl ring of flavopiridol is well conserved among the CDKs, but is not present in kinases
less potently inhibited, explaining the relative specificity of flavopiridol for CDKs.12
Flavopiridol may also induce cell cycle arrest by other mechanisms. For example, in addition
to direct inhibition of CDKs 1, 2, 4 and 6, flavopiridol also inhibits CDK7, the CDKactivating
kinase, (IC50 ~ 300 nM),13 resulting in decreased phosphorylation of CDKs at
threonine 160/161, necessary for their activation.14 Finally, transcriptional repression of cyclin
D1 by flavopiridol may also contribute to G1 arrest. This has been demonstrated following
flavopiridol exposure in MCF-7 breast cancer cells, in which decreases in cyclin D1 mRNA
levels precede the loss of CDK4 activity. The depletion of cyclin D1 mRNA was associated
with a decline in cyclin D1 promoter activity, measured by a luciferase reporter assay.15
While the effects of flavopiridol on CDK4 activity are specific for cells retaining wild-type
Rb, inhibition of CDK2 by flavopiridol also allows G1 arrest in cells lacking Rb.8, 9 Cell cycle
arrest mediated by flavopiridol is also independent of p53, as it occurs in cell lines in which p53
is inactivated by mutation or deletion.8,9
Differentiation
In some instances, growth arrest via CDK inhibition has been associated with cellular
differentiation. For example, treatment of NCI-358 non-small cell lung cancer (NSCLC) cells
with flavopiridol resulted in growth arrest and the induction of mucinous differentiation. In
these experiments, the onset of differentiation coincided temporally with the loss of CDK2
activity. 16
Apoptosis
In addition to cell cycle arrest, flavopiridol can also induce p53-independent apoptotic
cell death, especially in cells of hematopoietic origin.17,18 Apoptosis has also been observed in
solid tumor cell lines, including those derived from head and neck cancers,19 as well as those
210 Cell Cycle Checkpoints and Cancer
derived from carcinomas of the esophagus, lung and breast.20-22 The mechanisms by which
flavopiridol mediates apoptosis have not yet been elucidated. Specifically, the contribution of
CDK inhibition to apoptotic cell death has not been defined. For example, in lung cancer cell
lines, CDK inhibition and cell cycle arrest frequently occur at concentrations of drug lower
than those required for apoptosis.21 In many solid tumor cell lines, prolonged exposure (i.e., 72
hours) to high concentrations of flavopiridol are also required to achieve apoptosis. In addition,
cytotoxicity has been shown to occur even in non-cycling cells, including starved or confluent
A549 NSCLC cells,21,23 resting lymphoid cells in treated mice,18 and human endothelial cells.24
These findings raise the possibility that effects of flavopiridol on additional cellular targets
may contribute to apoptotic cell death. Several lines of evidence indicate that flavopiridol most
likely interacts with DNA.25 For example, in a recent study, reverse-phase high-performance
liquid chromatography demonstrated that flavopiridol binds to genomic DNA to a similar
extent as ethidium bromide and Hoechst 33258. Nuclear magnetic resonance (NMR)
spectroscopy revealed that DNA caused extreme broadening of flavopiridol 1H NMR signals
that could be reversed by addition of ethidium bromide or by DNA melting, suggesting that
flavopiridol binds to and likely intercalates into duplex DNA. Equilibrium dialysis demonstrated
that the dissociation constant of the flavopiridol-DNA complex was in the same range
observed for binding of intercalators such as doxorubicin and pyrazoloacridine to DNA.
Futhermore, analysis of the cytotoxic effects of flavopiridol (i.e., LC50 data) in the NCI tumor
cell line panel using the COMPARE algorithm26 demonstrated that flavopiridol most closely
resembles cytotoxic antineoplastic intercalators.25 In addition, clinically achievable concentrations
of flavopiridol result in induction of p53 in A549 NSCLC cells.21,25 While this supports an
interaction of flavopiridol with DNA, p53 induction is not required for apoptosis,21 and it
remains unclear whether flavopiridol induces DNA damage. Flavopiridol does not inhibit
topoisomerases,17, 25 as has been shown for other flavones. In addition, the flavopiridol-DNA
interaction has only been demonstrated at very high flavopiridol concentrations (10-50 μM),
well beyond those required for antiproliferative effects.25 Nonetheless, the data collectively
suggest that DNA represents another cellular target that could be responsible for the cytotoxic
effects of flavopiridol, especially in non-cycling cells.
While flavopiridol-mediated apoptosis is clearly independent of p53, the role of Bcl-2 and
other anti-apoptotic proteins remains controversial, and may be cell type dependent. For
example, in some hematopoietic cell lines, down-regulation of anti-apoptotic proteins (including
Mcl-1, XIAP, Bag-1 and Bcl-2) has been observed.27, 28 However, in other cell lines, alterations
in Bcl-2 and BAX levels in response to flavopiridol do not occur.17,29 Furthermore, in a recent
study of HeLa cell and small cell lung cancer (SCLC) cell lines, flavopiridol-mediated apoptosis
was clearly independent of Bcl-2.30 In these cell lines, neither overexpression nor antisense
oligonucleotide-mediated down-regulation of Bcl-2 had any effect on flavopiridol-induced
apoptosis. Pathways involving activation of caspase 8, followed by activation of other caspases,
such as caspase 3, appeared to play a major role in the apoptotic response to flavopiridol. In
contrast, caspase 8-independent induction of mitochondrial cytochrome c release (modulated
by Bcl-2 family members) followed by caspase 9 activation did not contribute significantly.
Interestingly, in cell lines lacking caspase 8, flavopiridol triggered mitochondrial depolarization
in the absence of cytochrome c release, followed by caspase 3 activation and cell death. In
summary, flavopiridol is able to kill tumor cells that are normally resistant to chemotherapeutic
agents due to Bcl-2 overexpression or the absence of caspase 8.30
Transcriptional Effects
Two studies have demonstrated an effect of flavopiridol on cellular transcription. As
already indicated, human breast carcinoma cells treated with flavopiridol showed decreased
transcription of the gene encoding cyclin D1.15 In addition, high concentrations of flavopiridol
affected the level of 63 different mRNAs in Saccharomyces cerevisiae, including mRNA encoding
proteins regulating cell cycle progression and phosphate and cellular energy metabolism.31
Small Molecule Inhibitors of Cyclin-Dependent Kinases 211
Recently, it has been demonstrated that flavopiridol inhibits P-TEFβ, a protein kinase
Composed of CDK9 and a cyclin subunit derived from one of three different genes. P-TEFβ
controls the elongation phase of transcription by RNA polymerase II. The ability of flavopiridol
to inhibit P-TEFβ-mediated phosphorylation of the carboxyl-terminal domain of the large
subunit of RNA polymerase II is particularly potent with an IC50 ~ 6 nM, non-competitive
with ATP.32
P-TEFβ, composed of CDK9 and cyclin T1 is a required cellular cofactor for the human
immunodeficiency virus (HIV-1) transactivator, Tat. Consistent with its ability to inhibit PTEFβ,
flavopiridol blocked Tat transactivation of the viral promoter in vitro, and blocked
HIV-1 replication in both single-round and viral spread assays with an IC50 < 10 nM.32
Antiangiogenic and Antimetastatic Properties
Antiangiogenic and antimetastatic effects may also contribute to the antitumor activity of
flavopiridol. In vitro, incubation of human umbilical vein endothelial cells (HUVECs) with
flavopiridol results in apoptotic cell death, even in non-cycling populations.24 In vivo, flavopiridol
has been found to decrease blood vessel formation in a mouse Matrigel model of angiogenesis.
33 In addition, at low nM concentrations, flavopiridol has been shown to inhibit the induction
of vascular endothelial growth factor (VEGF) by hypoxia in human monocytes.34 Finally,
flavopiridol has been shown to inhibit the secretion of matrix metalloproteinases by breast
cancer cells, resulting in decreased invasiveness in a Matrigel cell invasion assay. Decreased
invasiveness was noted in parental cells, as well as those engineered to overexpress c-eRB-2.22
Other Enzyme Targets
Although flavopiridol is relatively specific for CDKs compared to other serine and threonine
kinases, recent data have also implicated other possible enzyme targets. For example,
immobilized flavopiridol binds not only to CDK4 in tumor cell extracts, but also to cytosolic
aldehyde dehydrogenase 1.35 In addition, flavopiridol inhibits glycogen phosphorylase, most
potently in the presence of glucose, suggesting that it may affect cancer cells by starving them
of glycolytic intermediates.36,37
Summary
Some of the biologic effects of flavopiridol in preclinical models are clearly explained by
its ability to inhibit CDKs, including cell cycle arrest and differentiation. In addition, CDK
inhibition probably contributes to apoptosis as well. However, the ability of flavopiridol to
cause apoptosis in non-cycling cells, interact with DNA, modulate transcription and P-TEFb
activity, affect endothelial cells in biological models and inhibit glycogen phosphorylase, all
suggest that antitumor effects seen in in vivo models are indeed due to the effect of the drug on
multiple cellular targets.
Preclinical Antitumor Activity and Pharmacology
Xenograft Models
Initial studies in tumor-bearing mice used both the sub-renal capsule assay as well as
subcutaneously implanted human tumor xenografts. Scheduling experiments demonstrated
that the drug was more active when administered frequently than intermittently (i.e., daily
versus weekly).7,38 Similar effects were seen with IV and oral dosing. When administered over
protracted periods, flavopiridol was cytostatic in most xenograft models. Flavopiridol significantly
inhibited growth under the renal capsule of human mammary, lung, ovarian and colon
carcinoma, as well as glioblastoma with T/C ratios from 31 to 77% (ratio of the median tumor
weight of treated animals over the median tumor weight of control animals). In subcutaneously
growing xenograft tumor models, it similarly inhibited growth of human lung, colon,
gastric and mammary carcinomas as well as melanoma and glioblastoma with T/C values ranging
from 29 to 86%.7 Some of the most pronounced effects were seen against prostate carci212
Cell Cycle Checkpoints and Cancer
noma xenografts. At the maximal tolerated dose of 10 mg/kg/day administered orally on days
1-4 and 7-11, flavopiridol caused tumor stasis lasting 4 weeks in one prostate xenograft model
and tumor regression of a second prostate xenograft.39
Preclinical Pharmacokinetics and Metabolism
Murine plasma concentration-time profiles for flavopiridol exhibited biexponential behavior
with mean α and β half-lives of 16.4 and 201.0 minutes, respectively. The mean totalbody
plasma clearance was 22.6 ml/min/kg, and the mean oral bioavailability is ~20% in mice
and rats and ~35% in dogs.7 The metabolism of flavopiridol has been determined using
isolated rat liver perfusion models. Flavopiridol undergoes glucuronidation in the liver.40
Monoglucuronides may be released into the systemic circulation but are predominantly
excreted into the biliary tract, resulting in the propensity of flavopiridol for enterohepatic
circulation. Preclinical pharmacologic and toxicologic evaluation identified dose-limiting
toxicities as reversible hematopoietic and gastrointestinal effects.7
Phase I Studies
72-Hour Continuous Infusion Schedule
The demonstration that preclinical antitumor activity was superior with frequent administration
over a prolonged period led to phase I trials in which flavopiridol was given as a 72-
hour continuous infusion every 2 weeks. In the NCI phase I trial, 76 patients were treated.
Dose-limiting toxicity was secretory diarrhea with a maximal tolerated dose of 50 mg/M2/ day .41
In the presence of antidiarrheal prophylaxis, including a combination of cholestyramine and
loperamide, patients tolerated higher doses, defining a second maximal tolerated dose of 78
mg/M2/day for 3 days. The dose-limiting toxicity observed at the higher dose level was reversible
hypotension and a pro-inflammatory syndrome, including fever, fatigue, local tumor pain
and increases in acute phase reactants that made it difficult to administer drug chronically.41
One patient with renal cell carcinoma achieved a partial response (shrinkage of >50% of
tumor masses) and three other patients had minor responses (shrinkage of <50%), including
patients with colon carcinoma, renal cell carcinoma and non-Hodgkin’s lymphoma. Fourteen
patients received flavopiridol for more than 6 months, including five who received drug for
more than 1 year and one patient who received drug for more than 2 years, indicating that
disease stabilization for protracted periods is possible.41
The mean plasma flavopiridol concentration was 271 nM at the 50 mg/M2/day x 3 dose
level and 344 nM at the 78 mg/M2/day x 3-dose level, within range for inhibition of CDKs in
preclinical in vitro systems. Total clearance ranged from 11.5-27.3 l/hr/M2 (range 1.3-29.1)
and apparent volume of distribution at steady state was 131.16 l/M2 (range 24.3-516.7). Pharmacokinetics
appeared linear with dose, although there was patient heterogeneity with respect
to concentration and total plasma clearance for a given dose.41
In a second phase I study using the same 72-hour continuous infusion schedule, performed
at the University of Wisconsin, secretory diarrhea defined a maximal tolerated dose of
40 mg/M2/ day x 3. Similar plasma concentrations of flavopiridol were observed.42 One patient
in this trial had gastric cancer metastatic to the liver and had failed to respond to 5-
fluorouracil. Following treatment with flavopiridol, this patient achieved a sustained complete
response with no evidence of disease more than 2 years after treatment was completed.
Phase II Studies
Overview
Phase II studies of infusional flavopiridol have been completed or are ongoing in a number of
disease types, including chronic lymphocytic leukemia, non-Hodgkin’s lymphoma, as well as prostate,
colon, gastric, renal and non-small cell lung cancer (NSCLC).43 In general, these trials employ
the schedule of 50 mg/M2/day as a 72-hour infusion every 14 days; in some trials, higher
Small Molecule Inhibitors of Cyclin-Dependent Kinases 213
doses have been permitted with anti-diarrheal prophylaxis in patients who have no toxicity. To
date, extensive data are available from trials of patients with renal cell carcinoma,44 gastric carcinoma45
and NSCLC.46 Several themes have emerged from the results of these three studies.
Toxicities
First, the toxicity profiles in the three trials were similar. The major toxicities reported
included fatigue/asthenia, diarrhea, and clotting events. Diarrhea was an expected toxicity,
based on the preclinical and phase I data. In the renal cell trial, diarrhea occurred in 27/35
(77%) of patients and was severe in 7/35 (20%). Similar toxicity occurred among 16 patients
enrolled in the gastric cancer trial. In the NSCLC trial, diarrhea occurred in 11/20 (55%) of
patients but no severe toxicity was reported.
In the renal cell cancer trial, the plasma concentrations of flavopiridol and flavopiridol
glucuronide were measured at 23, 47 and 71 hours during the flavopiridol infusion and
the metabolic ratios of flavopiridol glucuronide: flavopiridol were determined. At 71
hours, the metabolic ratios showed a bimodal distribution with an antimode of 1.2. Thirteen
patients experienced diarrhea and had lower metabolic ratios [0.72 (0.53-0.86)] at
71 hours than patients without diarrhea [2.25 (1.76-2.3); P = 0.002]. Eight of 11 extensive
glucuronidators (ratio > 1.2) did not develop diarrhea, whereas 10 of 11 poor
glucuronidators (ratio <1.2) developed diarrhea (P = 0.008). Thus, the glucuronidation
of flavopiridol is apparently polymorphic, suggesting a genetic etiology, and is inversely
associated with the risk of developing diarrhea.47
Fatigue and asthenia were common in the three trials as well, reported in 88% and 93% of
patients in the renal cell carcinoma and gastric carcinoma trials, respectively. Although this was
also noted in phase I trials, it seemed to be more distressing to the presumably better functioning
patients in these phase II studies. Half of the patients in the lung cancer trial experienced
fatigue (10/20), although it was only rated as severe in three patients. The symptom is difficult
to interpret in lung cancer patients, as fatigue occurs commonly in this population, and has
been reported as severe in significant percentages of patients in the best supportive care arm of
randomized trials.48 Therefore, it is difficult to definitively conclude that the fatigue observed
in the NSCLC trial was entirely drug-related.
A serious toxicity that was not appreciated in the phase I studies was the occurrence of
venous and arterial clotting events. In the renal cancer trial, six patients developed venous
thrombosis; two events were catheter related, two occurred in the lower extremities and two
patients suffered probably pulmonary emboli. In addition, in this trial, three patients experienced
arterial vascular adverse events, including myocardial infarction, transient ischemic
attack and transient vision loss. In the trials of gastric and lung cancer, only venous clotting
events were reported. Five patients in the gastric cancer study developed catheter-related
thrombosis. Venous thromboses occurred in 7/20 (35%) patients in the NSCLC trial. While
the majority of these events occurred at the catheter site, lower extremity deep venous thromboses
and probably pulmonary embolism also occurred.
The incidence of venous thromboses does not appear to be affected by the concentration
at which flavopiridol is administered. Although a high concentration of 2.5 mg/ml was used in
some patients in the gastric cancer trial, the renal cell cancer trial administered drug at a concentration
of 0.1-0.5 mg/ml and in the NSCLC trial, all patients received flavopiridol at a maximal
concentration of 0.2 mg/ml. Despite the presumed higher volume and flow rate of infusion
associated with the administration of lower drug concentrations, catheter-related thromboses
were common.
Although patients with metastatic gastric, renal or lung cancer have high incidences of
deep venous thrombosis, the frequency of clotting events was higher than expected compared
with other clinical trials in comparable cohorts.44,49 Nonetheless, whether flavopiridol truly
causes a clotting diathesis remains unclear. For example, in the phase I trial, only five instances
of catheter related thrombosis occurred among 76 patients, and spontaneous thrombosis in the
absence of foreign bodies was not observed, despite alterations in plasma fibrinogen levels.41
214 Cell Cycle Checkpoints and Cancer
Similarly, in a preliminary report of the phase II trial in patients with metastatic colorectal
cancer, only 1 of 14 patients developed a catheter-related subclavian thrombosis.50 Measurement
of multiple clotting events in future trials may help assess the etiology of these events.
Single Agent Activity of Flavopiridol
A second theme emerging is the limited single agent activity of flavopiridol in these
diseases, despite the promise of phase I results in patients with renal cell or gastric cancer.
Among the 35 minimally pre-treated renal cell cancer patients, two objective responses were
observed. In the trial designed for advanced gastric cancer 14/16 enrolled patients were evaluable
for response; only one minor response was achieved. Necrosis was noted in one selected posttreatment
tumor biopsy, although the patient had progression of disease shortly after. Among
the 20 patients in the NSCLC study, 18 patients were evaluable for response. Although three
patients had minor responses, again there were no major objective responses.
Despite the lack of responses, several patients in these trials did achieve stable disease. For
example, three patients with renal cell cancer remained on trial for >30 weeks, and one patient
reported remained on flavopiridol more than 1 year. Similarly, in the NSCLC trial, four
patients (including the three patients with minor responses) achieved stable disease for 4, 5, 10
and 16 months respectively. In addition, the median survival of 7.5 months achieved with
flavopiridol was similar to the range of 7.4-8.2 months recently reported in a randomized trial
of four chemotherapy regimens containing platinum analogs in combination with taxanes or
gemcitabine.46,51 These trials were not designed to evaluate cytostatic effects of flavopiridol,
and it is possible that the tumors of patients remaining on flavopiridol for long periods had
biologic characteristics predicting a more indolent growth rate. Nonetheless, the results of
phase I and II trials thus far suggest that protracted disease stability with acceptable toxicity can
be achieved in some patients.
Available pharmacokinetic data indicate that plasma levels of flavopiridol achieved in these
studies were in the range for CDK inhibition. In the renal cell trial, flavopiridol concentrations
were 389 nM (296-567 nM), 412 nM (297-566 nM) and 397 nM (196-553 nM) at 23, 47
and 71 hours during the infusion, respectively.47 Levels were somewhat lower in the NSCLC
trial, where mean steady state concentrations (Css), defined as the geometric mean of the daily
drug levels, were reported for the first two infusions. During the first infusion, the mean Css (+ SD)
was 200 + 90 nM; during the second infusion the mean was 225 + 124 nM. Preclinical work in
NSCLC cell lines had indicated that concentrations of drug of at least 300 nM were required
for cytotoxic effects, whether mediated by apoptosis or necrosis.21,23 In addition, in preclinical
models, cytotoxicity was far more likely when drug concentrations were >500 nM. Importantly,
only six patients in the NSCLC achieved a Css exceeding 300 nM during either infusion, and
no patient achieved a Css above 500 nM, consistent with the absence of major
responses. Interestingly, in this trial, there was a correlation between plasma flavopiridol level
and stable disease. Plasma concentrations of flavopiridol were determined in 9 of 10 patients
with stable disease at 8 weeks, including two with minor response. Among these nine patients,
seven achieved a Css that was one standard deviation or more above 150 nM (the minimum
concentration producing clear growth arrest in the majority of NSCLC cell lines) during either
the first or second infusion. In contrast, among the eight patients with disease progression at 8
weeks, only 2 achieved similar plasma concentrations. Among the two groups of patients, there
was a statistically significant difference in the highest Css of flavopiridol achieved.46
Correlative Studies
Another similarity among these trials has been the lack of utility of in vitro cell cycle
studies performed on peripheral blood mononuclear cells (PBMCs). In the renal cell trial,
PBMCs were collected at baseline and 71 hours into the infusion. It was postulated that cells
collected at the end of the infusion would be inhibited from entering the cell cycle when
stimulated with phytohemagglutinin A (PHA) and interleukin-2 (IL-2). However, there was
no difference in cell cycle parameters of PHA/IL-2-stimulated lymphocytes collected at baseline
Small Molecule Inhibitors of Cyclin-Dependent Kinases 215
or at the end of the flavopiridol infusion. The lack of observed effect was likely because cells
need to be continuously exposed to flavopiridol in order for cell cycle arrest to become manifest.
Thus, cells that are exposed to flavopiridol for 72 hours and then undergo exvivo growth
stimulation do not experience any permanent cell cycle perturbations. Similarly, in the gastric
cell trial, no evidence of cell cycle effects or apoptosis were seen in circulating lymphocytes
examined by TdT or 7-AAD labeling with flow cytometry; evidence of PARP cleavage and
caspase-3 activation was also absent.
While the concentrations of flavopiridol observed in these studies should be capable of
achieving CDK inhibition in cycling tumor cells, this has not yet been confirmed in clinical
tumor specimens. The importance of this is underscored by the recent reports that immobilized
flavopiridol can bind to cytosolic aldehyde dehydrogenase 135 and glycogen phosphorylase36 present
in various cell lines, which may reduce its ability to compete with cellular ATP for CDK binding.
Currently, phosphospecific Rb antibodies exist, directed against sites specifically phosphorylated
by either CDK2 or CDK4. CDK inhibition by flavopiridol should cause decreased phosphorylation
of Rb at these sites. Adaptation of these antibodies for immunohistochemical studies
should allow confirmation of CDK inhibition in selected biopsies in future studies. Similarly,
determination of cyclin D1 levels by immunohistochemistry may also confirm the biologic
activity of flavopiridol in primary tumor specimens.
Alternative Strategies for the Development of Flavopiridol
Bolus Administration
Preclinical Data
Recently, promising activity has been reported in human leukemia and lymphoma
xenografts, using bolus-dosing schedules. For example, after treatment with 7.5 mg/kg
flavopiridol bolus (either IV or intraperitoneal) on each of 5 consecutive days, 11/12 advanced
stage HL-60 xenografts underwent complete regressions, and animals remained disease-free
several months after one course of flavopiridol treatment.18 Similar results were seen in two
lymphoma xenograft models, with complete and major regressions noted.18 Evidence of apoptosis
was seen both in the xenografts and in normal hematopoietic organs. In contrast, continuous
infusion of flavopiridol for 3 days, which results in plasma levels >400 nM, resulted in only
modest antitumor effect in animals bearing HL-60 xenografts. In a similar study against a head
and neck squamous cell carcinoma xenograft, flavopiridol administered intraperitoneally at 5
mg/kg per day for 5 consecutive days induced a 60-70% reduction in tumor size, which was
sustained for 10 weeks. Apoptotic cell death and cyclin D1 depletion were observed in tissues
from xenografts treated with flavopiridol at peak plasma concentrations of 5-8 μM.19 In
summary, the best antitumor effects in xenografted animals have been observed after daily
bolus IV or IP administration of flavopiridol that results in peak plasma levels of approximately 7
μM, followed by a progressive decline to 100 nM in 8 hours. Achievement of relatively shortlived,
but repetitive high plasma levels of flavopiridol in the μM range appears to be an effective
way to produce the maximum antitumor effect with flavopiridol at least in hematopoietic
malignancy and head and neck cancer models.
Phase I Trial
Based on these results, the NCI initiated a phase I trial of a daily 1-hour bolus of flavopiridol
for 5 consecutive days every 3 weeks.52 Initially, 24 patients received doses ranging from 12-
52.5 mg/M2/day x 5. Dose-limiting toxicities of nausea, neutropenia and fatigue were reached
at the 52.5 dose level, resulting in a recommended phase II dose of 37.5 mg/M2/day x 5. The
Cmax of flavopiridol at this dose level was 1.4 + 0.1 μM. To attempt to achieve higher concentrations,
12 patients received 1-hour flavopiridol for 3 consecutive days every 3 weeks. Doselimiting
toxicity was reached at 62.5 mg/M2/day, including severe hepatotoxicity, resulting in
a recommended phase II dose of 50 mg/M2/day for 3 consecutive days. At this dose level, the
216 Cell Cycle Checkpoints and Cancer
median Cmax of flavopiridol reached 3.2 + 1.2 μM. The elimination half-life of flavopiridol was
shorter with IV bolus (5.5 + 5.2 hours) than with 72-hour infusion, perhaps because
enterohepatic circulation was not observed.53 Three patients on trial (with NSCLC, mantlecell
lymphoma and melanoma) have had stable disease >6 months.52 Phase II trials with this
schedule are currently under development.
Combinations with Standard Chemotherapy Agents
It is possible that CDK inhibitor drugs will be primarily cytostatic when used alone, especially
against solid tumors derived from cells not typically predisposed to apoptotic responses.
Although flavopiridol may not have impressive antitumor activity when used as a single agent,
it may have great utility when combined with standard chemotherapy. Several preclinical
studies have indicated that flavopiridol produces additive and synergistic cytotoxicity when
administered in combination with many chemotherapy agents used in standard practice.54,55
In most cases, this is a schedule-dependent phenomenon in which synergy is seen when standard
chemotherapy drugs are administered before flavopiridol.
Taxol
Several studies have explored the interaction of flavopiridol with Taxol. If flavopiridol is
used first, CDK1 (cdc2) inhibition and G2 arrest prevents entry into mitosis, inhibiting Taxolmediated
cell death, which depends at least in part on mitotic arrest. However, if flavopiridol is
used following Taxol, significant synergism is seen.54,56 The activation of caspases, specifically
caspase 3, is enhanced by the presence of flavopiridol in Taxol-treated cells.56 The events that
occur in cells treated sequentially with Taxol followed by either no further treatment or by
flavopiridol have been analyzed. When Taxol is used alone, a transient mitotic block is induced,
after which cells exit mitosis without undergoing cytokinesis. Although they retain 4N DNA
content, they express G1 markers, a stage referred to as pseudo-G1. Cell death most likely
occurs in relation to both the mitotic block and the exit from an abnormal mitosis. Addition of
flavopiridol after Taxol treatment appears to accelerate the exit from mitotic block, with a more
rapid decline of mitotic cell markers, such as MPM-2.56 It is possible that the more rapid
emergence from an abnormal mitosis, caused by the appropriately timed inhibition of CDK1
(cdc2) is associated with enhanced cell death. The data demonstrated synergism with a
concentration of flavopiridol of 300 nM administered for 24 hours. To date, the synergism of
Taxol and flavopiridol has only formally reported in cell lines expressing wild-type p53, including
A549 NSCLC cells, MKN-74 gastric cancer cells and MCF-7 breast cancer cells,54, 56 although
a recent report suggests synergism in Calu-1 cells, which lack p53.57 It will be of interest to
determine whether the same degree of synergism occurs in cells lacking p53, which are often
more sensitive to taxane therapy.
A phase I trial of sequential Taxol and flavopiridol has been completed. The recommended
phase II dose is 175 mg/M2 Taxol, administered over 3 hours on day 1, followed by a 24-hr
infusion of 80 mg/M2 flavopiridol, beginning on day 2.58 Promising clinical activity has been
reported, including responses in some patients with taxane refractory tumors. This has prompted
a phase II study of the combination in patients with esophageal cancer that is refractory to
prior Taxol. However, phase III trials will be required to assess the true impact of flavopiridol in
the combination.
Combinations with Agents Damaging DNA and Inhibiting DNA Synthesis
Cytotoxic synergy has also been observed when flavopiridol follows
1. S-phase specific agents, such as cytarabine, gemcitabine and 5-fluorouracil;54, 59, 60
2. topoisomeraseI inhibitors, such as topotecan and CPT-11, which also work primarily
during S phase;54,61
3. topoisomerase II inhibitors such as doxorubicin and etoposide, which work in G1 and G2,
as well as in S phase;54 and
Small Molecule Inhibitors of Cyclin-Dependent Kinases 217
4. mitomycin C, an alkylating agent.55
All of these agents can cause retardation of S phase progression when used alone in vitro at
non-cytotoxic doses, resulting in populations of cells effectively recruited to S phase. The known
activity of CDK2 during S phase suggests a possible mechanism for flavopiridol-mediated
apoptosis following S phase recruitment by standard chemotherapy drugs.
The initiation of S phase requires activation of the E2F-1 transcription factor, which
occurs at the G1/S transition following CDK-mediated phosphorylation of Rb. However, proper
S phase progression also requires the appropriately timed deactivation of E2F-1, which is in
part the result of cyclin A-CDK2-mediated phosphorylation of E2F-1 and its heterodimeric
partner, DP-1.62-65 Inhibition of CDK2 activity during S phase would be expected to result in
inappropriately persistent E2F-1 activity, which can drive cells down both p53-dependent and
independent apoptotic pathways (reviewed in ref. 66; Fig. 1). Recently, peptides that block the
interaction of cyclin A-CDK2 with substrates such as E2F-1 have been shown to induce S
phase arrest and abrupt apoptosis.67 Chemotherapy agents may effectively recruit cycling cells
to S phase by imposing an S phase delay. Flavopiridol-mediated cyclin A-CDK2 inhibition
during such a delay may account for enhanced cytotoxicity.
This mechanism suggests several testable hypotheses. For example, cells recruited to S
phase by other mechanisms should be sensitized to flavopiridol treatment. This has been shown
in cells released from a hydroxyurea-induced block at the G1/S boundary68 and in cells
engineered to express an antisense oligodeoxynucleotide that inhibits expression of p27Kip1 .69
Furthermore, in cells treated sequentially with chemotherapy followed by flavopiridol, cyclin
A-CDK2 inhibition should result in decreased phosphorylation of E2F-1/DP-1, and E2F-1
DNA-binding and transactivation activities should be increased. In this sense, flavopiridol
used during S phase should be an E2F-1-metic drug. In addition, apoptosis mediated by chemotherapy/
flavopiridol combinations may very well depend on the level of cellular E2F-1
activity.68 Cells with higher baseline levels of E2F-1 activity will also have higher E2F-1 activity
following CDK2 inhibition, capable of generating an apoptotic response that might not occur
in cells with lower baseline levels of E2F-1 activity. In this regard, tumor cells may be selectively
sensitive to chemotherapy/flavopiridol combinations because they have high levels of E2F-1 activity
compared to normal cells because of disruption of the Rb-axis. Indeed, peptide CDK2
antagonists have been shown to selectively kill tumor cells,67 and chemotherapy/flavopiridol
combinations have been shown to be selectively toxic to T-antigen transformed fibroblasts
compared to their normal counterparts.60
An NCI-sponsored phase I trial of gemcitabine, followed by flavopiridol, is currently
underway in patients with solid tumors. Gemcitabine is administered on day 1, and a 24-hour
infusion of flavopiridol is administered beginning on day 2. This trial should lay the groundwork
for phase III trials in which the true contribution of flavopiridol can be assessed.
Cisplatin
In the study of NSCLC cell lines, the combination of cisplatin and flavopiridol is also
synergistic,54,68 although synergism has not been detected in bladder carcinoma models.70 In
NSCLC cells, the most pronounced effects occur when cisplatin precedes flavopiridol. However, a
synergistic interaction has also been demonstrated when flavopiridol is administered first or
concomitantly with cisplatin.54 Recently, it has been reported that flavopiridol can enhance
cisplatin accumulation in an ovarian cancer cell line, providing a possible explanation for the
sequence-independent synergy.71 Currently, a phase I trial is in progress assessing doses of cisplatin
immediately followed by a 24-hour infusion of flavopiridol. In addition, cisplatin has recently
been added directly prior to flavopiridol in the Taxol/flavopiridol combination, in order to
exploit the synergy demonstrated in in vitro systems.
Delayed Administration of Antimetabolites
Use of a CDK inhibitor before or concomitant with most DNA-damaging agents would
be expected to cause G1 arrest and therefore to decrease the cytotoxicity of such drugs, and
218 Cell Cycle Checkpoints and Cancer
indeed, most sequence specific interactions require that chemotherapy be administered before
flavopiridol. However, a CDK inhibitor might also be used to synchronize cells at the G1/S
boundary. After a sufficient interval to allow CDK activity to resume and for cells to enter S
phase in a synchronized fashion, administration of an S phase-specific agent may be predicted
to cause greater cytotoxicity than when used on an asynchronous tumor population. In vitro,
following G1 arrest by flavopiridol, cytarabine or 5-FU was administered following a 48-72
hour delay, with markedly enhanced cytotoxicity.54 In the case of 5-fluorouracil, synergy is also
seen without a long delay between the two drugs (and is greater than the degree of synergy seen
when 5-fluorouracil is used first).54 It is possible that the G1/S block imposed by flavopiridol is
accompanied by a down-regulation of thymidylate synthetase levels.72 Following removal of
flavopiridol, subsequent cytotoxicity in response to 5-FU would be enhanced. Thus, the
interaction of flavopiridol with antimetabolites suggests additional strategies for clinical trial.
Prevention of Endoreduplication by Flavopiridol
In addition to the therapeutic strategies for the further development of flavopiridol, one
recent report suggests the possibility that CDK inhibition may be appropriate as
chemoprevention. Cells defective in checkpoint control tend to undergo endoreduplication in
response to damage to DNA or the spindle apparatus. In the case of premalignant cells, entrance
into S phase prior to completing the previous mitosis leads to aneuploidy and genomic instability,
resulting in eventual emergence of the malignant clone. Recently, it has been shown that following
microtubule inhibition, the endoreduplication (i.e., progression to 8N) of checkpoint deficient
cells lacking p53 or p21WAF1/Cip1 can be prevented by flavopiridol, which causes an arrest
of 4N cells in a pseudo-G1 state.73 While the current formulation and toxicities of flavopiridol
Fig. 1. CDK2 inhibition during S phase can sensitize cells to E2F-1-dependent apoptosis. Following the G1/
S transition, E2F-1 activity is derepressed and E2F-1 is released from Rb. E2F-1, bound to its heterodimeric
partner, DP-1, directs transcription of genes required for S phase. Importantly, however, this transcription
is activated only transiently. Orderly S phase progression requires the down regulation of E2F-1 activity,
accomplished in part by cyclin A-CDK2-mediated phosphorylation. E2F-1 forms stable complexes with
cyclin A-CDK2, which phosphorylates both E2F-1 and DP-1, resulting in inhibition of the transactivation
and DNA binding activities of E2F-1 (upper row). Inhibition of CDK2 by flavopiridol (or any other CDK2
inhibitor) during S phase would be expected to result in inappropriately persistent E2F-1 activity, which
is known to result in S phase delay and apoptosis (lower row).
Small Molecule Inhibitors of Cyclin-Dependent Kinases 219
do not make it suitable for use as a chemopreventive agent, oral preparations of newer CDK
inhibitors may be effective for chemoprevention in selected high-risk populations.
Resistance to Flavopiridol
If flavopiridol used alone is largely cytostatic, its use over long periods of time can be
anticipated in some patients. It is noteworthy that it has been difficult to generate resistant cell
lines thus far. The first resistant cell line recently reported was an ovarian carcinoma cell line
that developed spontaneous resistance to both flavopiridol and cisplatin upon repeated passage
in the absence of drug exposure. At least part of the resistance appeared attributable to reduced
intracellular accumulation of the two drugs.71 Although one study has demonstrated that
flavopiridol may interact with the multi-drug resistance efflux protein, MRP1,74 additional
studies indicate that expression of MDR1 or MRP1 does not affect sensitivity to flavopiridol.
For example, the resistant ovarian carcinoma cell line displayed no resistance to substrates of
the P-glycoprotein drug efflux pump such as Taxol, etoposide or doxorubicin. Similarly, a
colon carcinoma cell line 9-fold resistant to flavopiridol, established in vitro following exposure
to increasing concentrations of drug, demonstrated only low level cross-resistance to Taxol,
not attributable to elevated levels of P-glycoprotein.75 Importantly, cells overexpressing P-glycoprotein
do not have increased flavopiridol resistance.70,76,77
Recently, it has been shown that overexpression of the ATP-binding cassette half transporter,
ABCG2, confers resistance to flavopiridol.77,78 In addition, a 24-fold resistant MCF-7
breast carcinoma cell subline, maintained in 1 μM flavopiridol, overexpresses ABCG2 mRNA
and protein. These cells display cross-resistance to mitoxantrone and camptothecins, also handled
by this transporter.77 To date, no resistant cell lines based on alterations of CDKs or their
endogenous modulators have been reported.
The Paullones
Discovery
Several members of the paullone family were identified using the COMPARE algorithm
to detect compounds from the database tested in the NCI Human Tumor Cell Line Anti-
Cancer Drug Screen that had similar biochemical targets and cellular mechanism to
flavopiridol.26 Kenpaullone is a potent inhibitor of cyclin B-cdc2 (IC50 = 0.4 μM), cyclin
A-CDK2 (IC50 = 0.68 μM), cyclin E-CDK2 (IC50 = 7.5 μM) and CDK5/p35 (IC50 = .85
μM), and has much lower activity against cyclin D1/CDK4 (IC50 = > 100 μM) or other
kinases.79 Kenpaullone competitively inhibits the binding of ATP, and molecular modeling
has demonstrated that it binds to the ATP binding site with residue contacts similar to other
CDK2 inhibitors.
Cell Cycle Effects
The cell cycle effects of kenpaullone were characterized in the MCF10A breast epithelial
cell line. In exponentially growing cells, a small increase in S phase fraction is seen, perhaps
consistent with inhibition of cyclin A/CDK2. In subsequent experiments, cells were synchronized
in G0/G1 by serum starvation, and then stimulated to re-enter the cell cycle in the
presence of vehicle or kenpaullone at its approximate IC50 concentration of 30 μM. After 20
hrs, vehicle-treated cells entered S phase, while kenpaullone-treated cells were arrested at the
G1/S boundary, consistent with CDK2 inhibition.79
Kenpaullone Analogues and Other Paullones
Kenpaullone contains a 9-bromo substituent. Its replacement with either a 9-cyano or
9-nitro group results in significantly enhanced inhibition of CDK1 (cdc2).80 The same is true
of 2-substituted paullones.81 10-bromo-paullone (NSC 672234) inhibits various protein kinases
in addition to CDKs 1, 2 and 5, including several protein kinase C isoenzymes. Interestingly, it
is less selective between cyclin A/CDK2 and cyclin E/CDK2, and also causes down-regulation
220 Cell Cycle Checkpoints and Cancer
of CDK4 expression.82 G1 arrest can be detected following drug exposure in exponentially
growing cells or in cells synchronized in G0/G1 by serum starvation.79
Purine Derivatives
First Generation Compounds
Purine derivatives were among the first CDK inhibitors discovered. Initially, 6-
dimethylaminopurine was shown to inhibit mitosis in sea urchin embryos without inhibiting
protein synthesis.83,84 It was subsequently shown to inhibit CDK1 (cdc2) activity with an IC50
= 120 μM, but its specificity was poor.5 Further studies identified substituted purine derivatives,
including isopentenyladenine (IC50 = 55 μM) and olomoucine (IC50 = 7 μM) capable
of inhibiting cyclin B-CDK1 (cdc2) in an in vitro kinase assay.5,85,86 Roscovitine, a derivative
of olomoucine, is an even more potent CDK1 (cdc2) inhibitor (IC50 = 0.45 μM).5 Olomoucine
and roscovitine demonstrate good selectivity for CDKs.
Isopentyladenine, olomucine and roscovitine all inhibit CDK2 as well and have been
crystallized in complex with CDK2.87-89 All of these compounds bind to the ATP site in the
deep groove between the N- and C-terminal domains of CDK2, usually occupied by the adenine
ring of ATP. Analysis of the ATP-CDK2 structure indicates that the N6 amino group acts a
hydrogen- bond donor to the backbone carbonyl of Glu-81, and N1 accepts a hydrogen bond
from the backbone nitrogen of Leu-83. The presence of a hydrogen-bond acceptor for the α-
amine of Leu-83 is a common feature of all known CDK2-inhibitor structures. For example,
the purine rings of isopentenyladenine and olomoucine are oriented so that the N3 and N7
positions, respectively, accept a hydrogen bond from the backbone nitrogen of Leu-83. Their
respective N2 and N6 positions also form novel interactions with the peptide oxygen of Leu-83.
Alteration of these positions by methylation results in loss of CDK inhibitory activity, confirming
the importance of the interaction of CDK inhibitors with Leu-83.
In the isopentyladenine-CDK2 complex, the N9 acts as a hydrogen-bond donor to the
carbonyl oxygen group of Glu-81. In contrast, the orientation of the purine ring of olomoucine
and roscovitine within the ATP-binding site of CDK2 is rotated nearly 160° relative to that of
the adenosine ring of ATP, such that there is no equivalent interaction to that of ATP with the
backbone carbonyl of Glu-81. However, olomoucine and roscovitine interact with a region of
CDK2 outside of the ATP binding pocket; this region is also involved in complexes of flavopiridol
with CDK2 and is likely responsible for the relative selectivity of olomoucine, roscovitine and
flavopiridol for CDKs.12
Second Generation Compounds
The rotated orientation of the purine ring of olomoucine and roscovitine compared with
ATP suggested that new substituents at the 2, 6 and 9 positions of the purine ring might
enhance binding affinity, as well as selectivity. Compounds derived from combinatorial libraries
of 2, 6, 9-trisubstituted purines were screened in in vitro kinase assays, identifying several
novel CDK inhibitors. The most potent of these compounds, purvalanol B, has an IC50 against
cyclin A-CDK2 of 6 nM, corresponding to a 1000-fold increase over olomoucine.31 The overall
geometry of purvalanol B bound to CDK2 resembles that of the related CDK2-olomoucine
and CDK2-roscovitine complexes. However, purvalanol B is able to hydrogen bond to the backbone
oxygen of Glu-81 via the acidic C8 atoms of its purine ring. In addition, the increased affinity for
CDK2 results from improved hydrophobic contact sites within the active site and the inhibitor,
and steric constraints of the ring systems that limit the number of conformations of the inhibitor.
The combinatorial library strategy has generated a large number of additional potent
purine analog inhibitors directed against the ATP binding site of CDK2, including CVT-313
(IC50 = 0.5 μM).90 CVT-313 is relatively selective for CDK2; when added to CDK1 (cdc2) or
CDK4, an 8.5- and 430-fold higher concentration of drug is required for half-maximal inhibition
of enzyme activity. In addition, there is no effect on other, non-related ATP-dependent
serine/threonine kinases.
Small Molecule Inhibitors of Cyclin-Dependent Kinases 221
Cellular Effects
Consistent with inhibition of CDK2 and cdc2 (CDK1), olomoucine, roscovitine, CVT-
313 and purvalanol A (a more membrane-permeable relative of purvalanol B) cause cell cycle
arrest at the G1/S and G2/M boundaries in a variety of transformed and non-transformed cell
types.31, 90-93 In NSCLC and neuroblastoma cell lines, cell cycle arrest mediated by olomoucine
and roscovitine is also accompanied by apoptosis.94
In addition to cell cycle arrest, purine derivatives have recently been implicated in differentiation
processes. In one study, a library of 2, 6,9-substitute purine derivatives was screened
for the ability to promote the differentiation of a myeloid leukemic cell line in culture. From
this screen, aminopurvalanol (AP) was identified; in the presence of this compound, cells acquire
phenotypic characteristics of differentiated macrophages and became arrested in the cell
cycle with a 4N DNA content.95 AP also inhibited mitosis in Xenopus egg extracts and affinity
chromatography and biochemical reconstitution experiments with these extracts identified cyclin
B-cdc2 (CDK1) as a target of the compound. AP inhibited immunoprecipitates of both cdc2
and CDK2 in human leukemic cell extracts, although the data suggested that the compound
preferentially targets the G2/M-phase transition in vivo.95
UCN-01
Mechanisms of Action
Protein Kinase C Inhibition
UCN-01 (7-hydroxystaurosporine), a staurosporine analogue, is a relatively non-specific
CDK inhibitor, and also has activity against several protein kinase C (PKC) isoenzymes. 96 It is
most specific for the calcium-dependent protein kinase C (IC50 ~30 nM) and less potent against
calcium-independent isoenzymes (IC50 ~500 nM).97-99 In a study of A549 lung adenocarcinoma
cells with acquired resistance to UCN-01 (A549/UCN cells), derived following long-term culture
in medium containing UCN-01, expression of the PKC-α, -ε and -θ isoenzymes was drastically
reduced. On subsequent culture in drug-free medium, levels of these isoenzymes returned
to baseline, and drug sensitivity was restored.100
However, the antiproliferative activity of UCN-01 against multiple human tumor cell
lines is far greater than that of other equipotent specific protein kinase C inhibitors, suggesting
alternative cellular targets.101 In addition, cells resistant to other protein kinase C-selective
staurosporine analogues (CGP 41251 and Ro 31-8220) that also expressed markedly reduced
levels of several PKC isoenzymes retained sensitivity to UCN-01.100
G1 Arrest
The effect of UCN-01 on cell cycle progression was first studied in several breast carcinoma
cell lines. Following release from a nocodazole-induced mitotic block, cells were arrested in
G1, and entry into S phase was inhibited.102 Although progression through S phase was
retarded in a dose-dependent fashion following release from an aphidicolin-induced block at
the G1/ S boundary, entry into and completion of M phase were not affected by UCN-01. G1
arrest by UCN-01 was confirmed in A431 epidermoid carcinoma cells103 and in A549 NSCLC
cells104 by simultaneous analysis of DNA content and 5-bromo-2-deoxyuridine (BrdU)
incorporation. In both cell lines, there was accumulation of the hypophosphorylated form of
Rb. When UCN-01 was added to CDK2 immune precipitates, histone H1 and Rb kinase activities
were inhibited in vitro (IC50 = 530 and 640 nM, respectively), indicating that UCN-01
can directly inhibit CDK2 with moderate potency.105 In addition, CDK2 activities of A431
cells pretreated with UCN-01 for 24 hrs at 260 and 520 nM were markedly inhibited (with
IC50’s of far less than 260 nM). When the same cell lysates were analyzed by Western blotting
for CDK2, the active, phosphorylated form of the kinase was significantly reduced,105 suggesting
that UCN-01 may act on CDK-activating kinase. Furthermore, treatment with UCN-01
222 Cell Cycle Checkpoints and Cancer
induced expression of p21WAF1/Cip1 and its complex formation with CDK2. Induction of
p21WAF1/Cip1 occurred at the transcriptional level. Expression of p27Kip1 was also increased
following UCN-01 exposure. Therefore, both direct and indirect inhibition of CDK2 appears to
contribute to UCN-01-mediated G1 arrest.105
The importance of CDK2 inhibition in UCN-01-mediated G1 arrest was confirmed in a
recent analysis of the A549/UCN NSCLC subline, derived following long term exposure of
the parent cells to UCN-01. Under the treatment conditions used, parental cells were capable
of undergoing apoptosis; the subline was resistant to apoptosis and displayed enhanced G1
arrest. While the subline had altered expression of multiple cell cycle related proteins, it most
notably had significantly lower levels of CDK2 compared to parental cells. Enhanced G1 arrest
in response to UCN-01 was associated with a significantly greater degree of inhibition of CDK2
activity following treatment.106
G1 arrest mediated by UCN-01 is clearly independent of p53, as it has been demonstrated
in cell lines in which p53 has been mutated or deleted.102,103 Initially, it also appeared
that G1 arrest could occur in cells lacking Rb as well, such as the MDA-MB468 cells used in
the analysis of breast carcinoma cell lines.102 However, following the nocodazole-induced
mitotic block, inhibition of S phase entry in these cells was incomplete. In addition, it was
noted that the growth inhibitory IC50 for MDA-MB468 cells was significantly higher than
that for other breast cancer cell lines that retained wild-type Rb. More recent studies in both
breast and lung carcinoma lines indicate that Rb is an important determinant of UCN-01-
mediated G1 arrest.107,108 In the absence of Rb, G1 arrest is either absent or less pronounced,
and S phase arrest occurs. Furthermore, in isogenic models, Rb-expressing derivatives of Rbnull
and mutant bladder and prostate carcinoma cell lines were capable of G1 arrest following
UCN-01 treatment, while G1 arrest in the parent cells was markedly compromised.107
The Rb-dependence of the G1 arrest suggests that components of the CDK4/6 axis also
comprise critical targets of UCN-01. Direct inhibition of CDK4 has been described in in vitro
assays with purified CDK components, although the IC50 value was similar to that seen for
CDK2.104 In A431 cells, exposure to UCN-01 results in significant diminution of cyclin D1
levels.105 In breast epithelial cells, decreased CDK4 levels have been described, with redistribution
of p27Kip1 from CDK4 to CDK2, resulting in CDK2 inhibition as well.108 Such events
may dominate the response of certain cell types. They either do not occur or produce little
impact in cells lacking Rb.
Apoptosis
In some cell lines, including hematopoietic cell lines,101 a colon carcinoma cell line109 and
several breast cancer cell lines,110 UCN-01 was shown to induce apoptosis. Cell cycle analysis
of leukemic T cell lines treated with UCN-01 demonstrated concomitant loss of cells with G2/
M DNA content and an increase in the proportion of cells with either a G1/S or a sub-G1
DNA content.101 These events coincided with the induction of apoptosis. Importantly, apoptosis
correlated with decreased inhibitory tyrosine phosphorylation of CDK1 (cdc2) and CDK2, as
well as activation of the histone H1 kinase activity of CDK1 (cdc2) and to a lesser degree,
CDK2.101 Inappropriate activation of CDK1 (cdc2) has been associated with apoptosis in a
number of model systems, inducing premature chromosome condensation in interphase cell.111
These data raise the possibility that modulators of CDK1 (cdc2) phosphorylation, such as wee1
kinase or the Cdc25C phosphatase, are important targets in cells undergoing apoptosis in response
to UCN-01.
G2 Checkpoint Abrogation
The activation of CDK1 (cdc2) during the apoptotic response in hematopoietic cell lines
led to the hypothesis that UCN-01 could be used to prevent G2 arrest following DNA damage
in cell lines of diverse origin. During the cellular response to DNA damage, CDK1 (cdc2) is
maintained in an inactive state through phosphorylation at Thr-14 and Tyr-15. This results in
G2 checkpoint arrest, in part to allow time for repair so that damaged cells do not enter mitosis.
Small Molecule Inhibitors of Cyclin-Dependent Kinases 223
Following treatment with DNA damaging agents such as γ-irradiation or cisplatin, treatment
with UCN-01 prevents G2 arrest.112,113 The abrogation of the G2 checkpoint, resulting in the
mitotic entry of damaged cells, is frequently associated with increased cytotoxicity.
The events that mediate G2 checkpoint control in response to DNA damage are shown in
Figure 2. Following DNA damage, activation of the ATM and ATR phosphatidylinositol 3-
kinase-related kinase family members result in phosphorylation of the checkpoint kinases, Chk1
and Chk2. Chk1 phosphorylates Cdc25C at Ser-216, which promotes binding of Cdc25C to
14-3-3 proteins, resulting in its cytoplasmic sequestration, where it cannot dephosphorylate
and activate CDK1 (cdc2).114-117 As the inhibitory phosphorylation of CDK1 (cdc2) is maintained,
cells undergo G2 arrest. Recently, it has been demonstrated that UCN-01 causes loss of
Ser-216 phosphorylation and 14-3-3 binding to Cdc25C in DNA-damaged cells. In addition,
UCN-01 potently inhibits the ability of Chk1 to phosphorylate Cdc25C in vitro.118,119 These
results identify Chk1 (and the Cdc25C pathway) as potential targets of G2 checkpoint abrogation
by UCN-01. In contrast, Chk2 was refractory to inhibition by UCN-01 in vitro and was
Fig. 2. UCN-01 abrogates G2 checkpoint control following DNA damage by inhibiting Chk1. Following
DNA damage, activation of the ATM/ATR kinases occurs, resulting in phosphorylation and activation of
the checkpoint kinases, Chk1 and Chk2. Chk1 (and perhaps Chk2) phosphorylate Cdc25C on Ser-216,
resulting in its binding to 14-3-3 proteins and ultimate cytoplasmic sequestration. This prevents it from
removing inhibitory phosphates from CDK1 (cdc2); the latter remains inactive, preventing the entry of
damaged cells into mitosis. (The activities that Cdc25C cannot accomplish following DNA damage because
of its cytoplasmic sequestration are indicated with white arrows). UCN-01 causes loss of Ser-216 phosphorylation
and 14-3-3 binding of Cdc25C in DNA-damaged cells, allowing dephosphorylation and activation
of CDK1 (cdc2), abrogation of the G2 checkpoint, and entry into mitosis.
The ATM, Chk2 and Chk1 kinases also phosphorylate p53 following DNA damage, contributing
to its stabilization. Stabilization of p53 following DNA damage occurs even in the presence of UCN-01,
and can mediate G2 arrest via transcriptional induction of p21WAF1/Cip1 and 14-3-3σ, which contribute to
CDK1 (cdc2) inhibition. Therefore, in the presence of p53, UCN-01 does not prevent the arrest of cells
in G2 following DNA damage. In the absence of p53, the checkpoint is completely dependent on the
Cdc25C pathway, and is severely compromised by UCN-01-mediated Chk1 inhibition. Inappropriate
entry into mitosis following DNA damage is frequently lethal. Thus, UCN-01 should enhance the cytotoxicity
of DNA damaging agents selectively in cells deficient in p53.
224 Cell Cycle Checkpoints and Cancer
still phosphorylated in irradiated cells treated with UCN-01, excluding Chk2 and ATM as in
vivo targets of UCN-01.
It has been postulated that UCN-01 must affect other pathways in addition to the Cdc25CSer-
216 regulatory pathway, because the checkpoint abrogation observed after UCN-01
treatment is much more severe than that observed after expression of a 14-3-3-binding mutant
of Cdc25C.118 Premature mitotic entry following treatment with UCN-01 indicates that active
CDK1 (cdc2) is accumulating in the nucleus. Disruption of the Cdc25C pathway explains
activation of CDK1 (cdc2), but does not address its nuclear accumulation. Throughout
interphase, and during the cellular response to DNA damage, cyclin B-CDK1 (cdc2) complexes
shuttle between nucleus and cytoplasm. Nuclear export of cyclin B-CDK1 (cdc2) complexes
is regulated in part by phosphorylation of a nuclear export signal (NES) in cyclin B; therefore,
it has been proposed that kinases regulating cyclin B NES function could also be potential
targets inhibited by UCN-01.118
Abrogation of the G2 checkpoint by UCN-01 in DNA-damaged cells occurs selectively
in cells that lack p53. This was confirmed using parental and HPV16E6-expressing breast
carcinoma cells; absence of G2 arrest, sensitization to DNA damage and increased cytotoxicity
occurred primarily in E6 expressing cells.112 This is because p53-mediated events, including
the transcriptional induction of p21WAF1/Cip1 and 14-3-3σ results in CDK1 (cdc2) inhibition
and cytoplasmic sequestration,120,121 causing G2 arrest even with abrogation of the Chk1-
Cdc25C-cdc2 pathway in UCN-01 treated cells. Therefore, treatment with UCN-01 should
result in the selective enhancement of p53-deficient cells to DNA damaging treatments, an
attractive feature for its therapeutic use.
S Phase Checkpoint Abrogation
In addition to arrest at the G1/S or G2/M boundaries following DNA damage, S phase
retardation and arrest may occur, comprising an S phase checkpoint. Manifestation of the S
phase checkpoint most likely depends on several factors, including the cell type
undergoing damage and its expression of Rb,122 as well as on the type and dose of DNA
damage used. Two systems that have been studied include colon carcinoma cells, treated with
camptothecin (CPT-11), a topoisomerase I inhibitor,123 and CHO cells, treated with cisplatin.124
Both cell types undergo S phase delay and arrest in response to the respective treatments. In
both cases, UCN-01 has been found to abrogate the S phase arrest and enhance chemotherapymediated
cytotoxicity. In the case of camptothecin-treated colon carcinoma cells, this occurs in
concert with a modulation of cyclin A-CDK2 activity as well as activation of CDK1 (cdc2)
activity, an effect that is selective for p53-deficient cells. In cisplatin-treated CHO cells, UCN-
01 induces a redistribution of PCNA to the detergent-insoluble, DNA-bound fraction, which
presumably contributes to accelerated DNA synthesis. The UCN-01-mediated redistribution
of PCNA following cisplatin treatment may be more likely to occur in p53-deficient cells, in
which PCNA is not sequestered by high levels of p21WAF1/Cip1. Importantly, in both systems,
the effects were associated with an enhancement of chemotherapy-induced cytotoxicity at
concentrations of UCN-01 that alone were not cytotoxic and had no detectable effect on cell
cycle progression.
In the above in vitro models, cell death was presumably enhanced by the mitotic entry of
damaged cells, allowed by combined abrogation of the S and G2 checkpoints. In studies of
hematopoietic cells, UCN-01 has also resulted directly in the death of S phase arrested cells.
Recently, it has been shown that the PI 3-Kinase-AKT-BAD pathway is inhibited by UCN-01
in gemcitabine-treated human myeloid leukemia ML-1 cells. Although neither gemcitabine
nor UCN-01 alone had any effect of PI 3-Kinase activity, sequential gemcitabine-UCN-01
treatment resulted in significant diminution in this activity. As a consequence, the combination
treatment resulted in decreased phosphorylation of AKT as well as decreased phosphorylation
of BAD, permitting an apoptotic response.125
Small Molecule Inhibitors of Cyclin-Dependent Kinases 225
Effects on E2F-1 Protein Expression
Orderly progression through S phase requires appropriately timed diminution of E2F-1
activity. The diminution in E2F-1 activity occurs by phosphorylation mediated by cyclin
A-CDK2, which disrupts DNA binding and transactivation activities, followed by ubiquitinproteasome-
dependent degradation. It has recently been shown that UCN-01 enhances
degradation of E2F-1, which perhaps could contribute to abrogation of an S phase
checkpoint.72,126 One consequence of decreased E2F-1 activity is decreased thymidylate
synthetase gene expression, resulting in enhanced sensitivity to 5-fluorouracil.
Inhibition of DNA Repair Signaling
In addition to its effects as a checkpoint abrogator, UCN-01 may also directly inhibit
DNA repair. Interestingly, in studies of CHO cells, UCN-01 was capable of preventing cisplatin
induced G2 arrest in all CHO lines, regardless of their DNA repair status. However, UCN-01
enhanced cisplatin-mediated cytotoxicity only in repair proficient CHO/AA8 cells, and not in
repair-deficient CHO/UV41 cells (which are incapable of recovery from cisplatin-mediated
DNA damage).113 While it does not physically inhibit the repair apparatus, UCN-01 inhibits
repair signaling, resulting in the attenuation of the interaction of XPA and ERCC, two components
of the nucleotide excision repair (NER) pathway; the phosphorylation of an XPA-associated
protein is also inhibited in UCN-01 treated cells.127
Preclinical Antitumor Activity and Pharmacology
Administration of UCN-01 by an intravenous or intraperitoneal route resulted in
antitumor activity in several xenograft model systems, including breast carcinoma, renal
carcinoma, pancreatic carcinoma and leukemia.2,128 Both in vitro and in vivo, the antitumor
effects were greatest when UCN-01 was given over a protracted period, usually greater than 72
hours.2,102 Pharmacokinetics and toxicologic studies using several schedules were completed in
rats and dogs. Following a 72-hour continuous infusion of UCN-01 in beagle dogs, local
reactions at the site of injection and gastrointestinal toxic effects were dose limiting. A steady-state
plasma concentration of 330 nM was achieved. Pharmacokinetic parameters included a volume of
distribution of 6.09 L/kg, and a total clearance of 0.6 L/kg per hour with a β elimination
half-life of ~12 hours.129
Phase I Studies
UCN-01 as a Single Agent
The first phase I trial of UCN-01 utilized a 72-hour continuous infusion every 2 weeks.130
In this trial, an unexpectedly long half-life of approximately 4 weeks was observed, approximately
100 times longer than the half-life predicted from preclinical models. This was most
likely caused by the avid binding of UCN-01 to α1-acid glycoprotein.131 For this reason, the
protocol was amended to retreat every 4 weeks, and the continuous infusion duration was
reduced to 36 hours during the second and subsequent courses. In this trial, there was a relative
lack of myelotoxicity or gastrointestinal toxicity, which were also predicted based on experience
in animal models. Instead, prominent toxicities included nausea and vomiting, symptomatic
hyperglycemia associated with an insulin-resistance state, and pulmonary toxicity, characterized
by substantial hypoxia without obvious radiologic changes. Other toxicities included
myalgias, headaches and elevation of transaminases. Pharmacokinetic parameters included a
clearance of .014 L/hour and a terminal half-life of 574.4 (199-4099) hours. The concentration
of free UCN-01 was assessed in saliva. At the maximal tolerated dose of 42.5 mg/M2/day,
the median free concentration was 111 nM, which is associated with G2 checkpoint abrogation
in in vitro models.
One patient with refractory metastatic melanoma achieved a partial response that lasted 6
months. Several other patients with leiomyosarcoma, non-Hodgkin’s lymphoma and lung cancer
achieved disease stabilization for 6 months or greater.
226 Cell Cycle Checkpoints and Cancer
Because of the long half-life, two phase I trials have addressed administration of UCN-01
as a short infusion, given over either 1 hour or 3 hours. In the American trial, at doses up to 68
mg/M2 given over 1 hour, toxicity was mild and reversible, including fatigue, fever, hyperglycemia,
and elevated transaminases and creatinine.132 At higher doses given over 1 hour, doselimiting
hypotension occurred and accrual continues with a 3-hour infusion. Mean half-life
was 534 + 580 hours. In this study, no responses were seen among the first 15 patients, although
one patient with cervical cancer has remained on study for more than 1 year with stable
disease. Using a 3-hour infusion, the Japanese trial has reported on patients receiving up to
51.1 mg/M2 with only mild diarrhea, nausea and arrhythmia reported. Pharmacokinetic studies
have confirmed the low systemic clearance and distribution volume, as well as the extremely
long half-life.133
Combinations with Standard Chemotherapy Agents
Several phase I studies are planned or in progress in which UCN-01 is administered in
combination with standard chemotherapy drugs. For example, a trial in which UCN-01
follows fludarabine is in progress in patients with refractory low-grade lymphoid malignancies.
Trials in which UCN-01 follows cisplatin have been initiated in patients with solid tumors. In
in vitro models, sequential cisplatin-UCN-01 treatment results in decreased levels of cyclin A
and a phosphorylation pattern of CDK1 (cdc2) consistent with activated kinase. Serial biopsies
of tumors are planned in patients; immunostaining with cyclin A antibodies and appropriate
phosphospecific CDK1 (cdc2) antibodies should provide confirmation of UCN-01 mediated
abrogation of cisplatin-induced G2 arrest.134 Finally, a trial in which 5-fluorouracil and UCN-01
are combined sequentially is also in progress, with the hope that decreased thymidylate
synthetase levels mediated by UCN-01 will enhance 5-fluorouracil-induced cytotoxicity.
Novel Selective CDK Inhibitors
Inhibitors of CDK2
Several selective CDK2 inhibitors are currently under development. The most extensive
reporting to date has been on a series of compounds developed at Glaxo Wellcome. Several of
these compounds have demonstrated cell cycle arrest in normal cells, and selective killing of
transformed cells. For example, it has been possible to kill colon carcinoma cells under the
same conditions that only slow proliferation of human diploid fibroblasts.135 Thus, these compounds
appear to mimic the killing of transformed cells by cell-permeable CDK2 inhibitory
peptides, perhaps by a mechanism dependent on the high levels of E2F-1 activity in transformed
cells.67
Another series of substituted oxindole inhibitors of the ATP binding site of the cyclin ACDK2
have also been synthesized. Compound 4 is a potent and selective inhibitor of CDK2 (IC50
= 10 nM), and a weaker inhibitor of cyclin B-CDK1 (cdc2) and cyclin D1-CDK4 (IC50s = 110
and 130 nM, respectively), and showed an average IC50 of 2 μM against a panel of 12 other
protein kinases.136 This drug produced the expected reversible cell cycle arrest in human diploid
fibroblasts. In addition, when mink lung epithelial cells were exposed to compound 4 first,
followed by treatment with several standard cytotoxic chemotherapy agents, significant inhibition
of cytotoxicity was observed, as expected when a CDK inhibitor is used first in the sequence.
Compound 4 has been formulated for topical application, and has recently been shown to
confer protection from chemotherapy-induced alopecia in rodent models. In one set of
experiments, newborn rats received chemotherapy with or without topical compound 4. While
control animals lost all of their hair, -30-70% of those treated with compound 4 had at least
partial protection at the application site. These data suggest that CDK inhibition may not only
be useful as antitumor therapy, but may also be useful in the protection of normal cells.136
Small Molecule Inhibitors of Cyclin-Dependent Kinases 227
Table 1. Families of Small Molecule Inhibitors of Cyclin-dependent kinases
Family Agent IC50
(μM)
against Relative
cdc2 specificity CDK4
(CDK1) for CDKs inhibition Ref.
Purine
Derivatives Dimethylaminopurine 120 - - 83, 84
Isopentyladenine 55 - - 85
Olomoucine 7 + - 148
Roscovitine 0.45 + - 5, 88
CVT-313 4.2 + - 90
Purvalanol Derivatives 0.004 + - 31
Flavopiridols Flavopiridol 0.4 + + 10
Deschloroflavopiridol 12
Staurosporines Staurosporine 0.004 - - 85
UCN-01 0.03 - + 104
— Butyrolactone I 0.6 + - 149
— 9-hydroellipticine 1 ND ND 150
Polysulfates Suramin 4 - ND 151
— Toyocamycin 0.88 + ND 152
Paullones Kenpaullone 0.4 + - 79
10-bromopaullone 1.3 + ND 79
Indirubins Indirubin 10 + + 153
(Active constituent
of a Chinese herbal
medicine) Indirubin-3’-monoxime 0.18 + + 153
Chemical CDK inhibitors can be divided into nine families. However, not all are specific for CDKs.
Staurosporine, UCN-01, suramin, 6-methylaminopurine, and isopentenyladenine are relatively nonspecific
protein kinase inhibitors. In contrast, olomoucine, roscovitine, CVT-313, purvalanol derivatives,
flavopiridol, butryolactone I, paullones and indirubins are more selective for CDKs. Buryrolactone I,
olomoucine, roscovitine, CVT-313, purvalanol and the paullones are relatively selective for CDK1
(cdc2), CDK2 and CDK5, but are relatively inactive against CDK4 and CDK6. Flavopiridol, staurosporine,
UCN-01 and the indirubins can inhibit CDK4; flavopiridol has also been shown to inhibit CDK6, CDK7
and P-TEFb (containing CDK9). ND- not determined.
E7070
Another CDK2 inhibitor under investigation in Europe and Japan is E7070, a novel
chloroindolyl-sulphonamide, which demonstrated a cytotoxicity profile suggesting a unique
mechanism of action using the NCI COMPARE program.137 The primary mechanism of this
drug may be indirect inhibition of CDK2 activity, as it demonstrates affects on CDK2 phosphorylation
as well as on cyclin E transcription.138 Xenograft and orthotopic transplantation
studies have indicated suppression of tumor growth and reduction of tumor volume in both
colon and lung cancer models.139,140 Recently, the combined results of four phase I studies
utilizing different schedules were reported; in all schedules, dose-limiting toxicities were
hematologic. Partial and minor responses among 127 total patients have been observed in
228 Cell Cycle Checkpoints and Cancer
patients with breast, endometrial, renal and ovarian carcinoma.141 The drug displays a half-life
ranging 10-31 hours, with a large volume of distribution.142 Initiation of phase II studies is
anticipated. In human tumor xenograft models, E7070 appears synergistic with camptothecin
(CPT-11).143
Inhibitors of CDK4
The universal involvement of the Rb-axis in human tumors has also motivated the
development of compounds specific for the ATP-binding site of CDK4, with the hope of achieving
selectivity for transformed cells. Some concern for this strategy has been raised by the effects of
a dominant-negative CDK4 mutant on the cell cycle profile of transformed cells retaining wildtype
Rb. Expression of this mutant, which fails to bind ATP, in U2OS osteosarcoma cells, inhibits
cyclin D1-CDK4 activity but does not cause G1 arrest.144 Unlike ectopic expression of p16INK4A,
it does not cause redistribution of Cip/Kip proteins from CDK4 to CDK2, and hence does not
cause concomitant inhibition of CDK2. Nevertheless, several compounds, including
diaminothiazoles145 and [2,3-d] pyridopyrimidines,146,147 aimed at the ATP binding site of CDK4,
have been reported to result in Rb-dependent G1 arrest, and inhibit tumor cell proliferation in
vitro at submicromolar concentrations. In clonogenic survival assays, irreversible inhibition of
colony formation has been observed. Finally, delayed tumor growth and time to progression in
human colon tumor xenografts has been observed in vivo. Currently, these compounds are under
investigation for their ability to induce differentiation or apoptosis, or produce cytotoxic synergy
in a sequence-dependent fashion with standard chemotherapy agents.
Conclusion
The universal deregulation of CDKs in human cancer makes these enzymes extremely
attractive targets for antineoplastic drug development. The first two CDK modulators to reach
clinical trial, flavopiridol and UCN-01, may exert their effects on a number of biologic
pathways. Data thus far indicate acceptable toxicity profiles, and some patients have achieved
impressive disease stabilization. Nonetheless, the optimal schedules for administration of these
drugs remain to be determined. Although plasma concentrations achieved appear sufficient for
CDK modulation, confirmation of the intended targets has not yet been demonstrated in
tumor biopsies from treated patients.
As novel CDK inhibitors are developed, with improved potency and selectivity, it will be
critical to determine whether they induce cytotoxicity, or whether they are primarily cytostatic.
These agents have the potential for significant synergism with standard chemotherapy drugs,
where they may ultimately find their place in the anticancer armamentarium. Such combinations
may be very sequence-dependent; in most cases maximal effects can be expected if CDK inhibition
follows exposure to DNA or microtubule damage.
Finally, it will be important to continually evaluate the selectivity of CDK inhibition for
transformed cell types. In this regard, several strategies of CDK modulation, including increasing
E2F-1 activity during S phase via CDK2 inhibition, or abrogating G2 checkpoint control via
CDK1 (cdc2) activation, have demonstrated selectivity for transformed cells in vitro. In addition,
appropriate use of CDK inhibition may confer protection on normal cells. Thus, small
molecule CDK inhibitors have great promise for increasing both the efficacy and safety of current
cancer treatment.
References
1. Shapiro GI, Harper JW. Anticancer drug targets: Cell cycle and checkpoint control. J Clin Invest
1999; 104:1645-1653.
2. Senderowicz AM, Sausville EA. Preclinical and clinical development of cyclin-dependent kinase
modulators. J Natl Cancer Inst 2000; 92:376-387.
3. Mani S, Wang C, Wu K et al. Cyclin-dependent kinase inhibitors: Novel anticancer agents. Exp
Opin Invest Drugs 2000; 9:1849-1870.
4. Meijer L. Chemical inhibitors of cyclin-dependent kinases. Trends Cell Biol 1996; 6:393-397.
Small Molecule Inhibitors of Cyclin-Dependent Kinases 229
5. Meijer L, Kim SH. Chemical inhibitors of cyclin-dependent kinases. Methods Enzymol 1997;
283:113-128.
6. Meijer L, Leclerc S, Leost M. Properties and potential applications of chemical inhibitors of cyclindependent
kinases. Pharmacol Ther 1999; 82:279-284.
7. Sedlacek HH, Czech J, Naik R et al. Flavopiridol (L86 8275; NSC 649890), a new kinase inhibitor
for tumor therapy. Int J Oncol 1996; 9:1143-1168.
8. Kaur G, Stetler-Stevenson M, Sebers S et al. Growth inhibition with reversible cell cycle arrest of
carcinoma cells by flavone L86-8275. J Natl Cancer Inst 1992; 84:1736-1740.
9. Carlson BA, Dubay MM, Sausville EA et al. Flavopiridol induces G1 arrest with inhibition of
cyclin-dependent kinase (CDK)2 and CDK4 in human breast carcinoma cells. Cancer Res 1996;
56:2973-2978.
10. Losiewicz MD, Carlson BA, Kaur G et al. Potent inhibition of cdc2 kinase activity by the flavonoid
L86-8275. Biochem Biophys Res Commun 1994; 201:589-595.
11. Singh SS, Sausville EA, Senderowicz AM. Cyclin D1 and CDK6 are the targets for flavopiridolmediated
G1 block in MCF10A breast epithelial cell line. Proc Am Assoc Cancer Res 1999;
40:A184 (abstr).
12. De Azevedo WF, Mueller-Dieckmann HJM, Schulze-Gahmen U et al. Structural basis for specificity
and potency of a flavonoid inhibitor of human CDK2, a cell cycle kinase. Proc Natl Acad Sci
USA 1996; 93:2735-2740.
13. Carlson BA, Pearlstein RA, Naik RG et al. Inhibition of CDK2, CDK4 and CDK7 by flavopiridol
and structural analogs. Proc Am Assoc Cancer Res 1996; 37:A2897 (abstr).
14. Worland P, Kaur G, Stetler-Stevenson M et al. Alteration of the phosphorylation state of p34cdc2
kinase by the flavone L86-8275 in breast carcinoma cells. Biochem Pharmacol 1993; 46:1831-
1840.
15. Carlson B, Lahusen T, Singh S et al. Down-regulation of cyclin D1 by transcriptional repression in
MCF-7 human breast carcinoma cells induced by flavopiridol. Cancer Res 1999; 59:4634-4641.
16. Lee HR, Chang TH, Tebalt MJ et al. Induction of differentiation accompanies inhibition of CDK2
in a non-small cell lung cancer cell line. Int J Oncol 1999; 15:161-166.
17. Parker BW, Kaur G, Nieves-Neira W et al. Early induction of apoptosis in hematopoietic cell lines
after exposure to flavopiridol. Blood 1998; 91:458-465.
18. Arguello F, Alexander M, Sterry JA et al. Flavopiridol induces apoptosis of normal lymphoid cells,
causes immunosuppression and has potent antitumor activity in vivo against human leukemia and
lymphoma xenografts. Blood 1998; 91:2482-2490.
19. Patel V, Senderowicz AM, Pinto J, D. et al. Flavopiridol, a novel cyclin-dependent kinase inhibitor,
suppresses the growth of head and neck squamous cell carcinomas by inducing apoptosis. J Clin
Invest 1998; 102:1674-1681.
20. Schrump DS, Matthews W, Chen GA et al. Flavopiridol mediates cell cycle arrest and apoptosis in
esophageal cancer cells. Clin Cancer Res 1998; 4:2885-2890.
21. Shapiro GI, Koestner DA, Matranga CB et al. Flavopiridol induces cell cycle arrest and p53-
independent apoptosis in non-small cell lung cancer cell lines. Clin Cancer Res 1999; 5:2925-2938.
22. Li Y, Bhuiyan M, Alhasan S et al. Induction of apoptosis and inhibition of c-eRbB-2 in breast
cancer cells by flavopiridol. Clin Cancer Res 2000; 5:223-229.
23. Bible KC, Kaufmann SH. Flavopiridol: A cytotoxic flavone that induces cell death in noncycling
A549 human lung carcinoma cells. Cancer Res 1996; 56:4856-4861.
24. Brüsselbach S, Nettelbeck DM, Sedlacek HH et al. Cell cycle-independent induction of apoptosis
by the anti-tumor drug flavopiridol in endothelial cells. Int J Cancer 1998; 77:146-152.
25. Bible KC, Bible J, RH., Kottke TJ et al. Flavopiridol binds to duplex DNA. Cancer Res 2000;
60:2419-2428.
26. Paull KD, Shoemaker RH, Hodes L et al. Display and analysis of patterns of differential activity of
drugs against human tumor cell lines: Development of mean graph and COMPARE algorithm. J
Natl Cancer Inst 1989; 81:1088-1092.
27. Konig A, Schwarz GK, Mohammed RM et al. The novel cyclin-dependent kinase inhibitor
flavopiridol downregulates Bcl-2 and induces growth arrest and apoptosis in chronic B-cell leukemia
lines. Blood 1997; 90:4307-4312.
28. Kitada S, Zapata JM, Andreeff M et al. Protein kinase inhibitors flavopiridol and 7-hydroxystaursporine
down-regulate antiapoptosis proteins in B-cell chronic lymphocytic leukemia. Blood
2000; 15:393-397.
29. Byrd JC, Shinn C, Waselenko J et al. Flavopiridol induces apoptosis in chronic lymphocytic leukemia
cells via activation of caspase-3 without evidence of Bcl-2 modulation or dependence on
functional p53. Blood 1998; 10:3804-3816.
230 Cell Cycle Checkpoints and Cancer
30. Achenbach TV, Muller R, Slater EP. Bcl-2 independence of flavopiridol-induced apoptosis. J Biol
Chem 2000; 275:32089-32097.
31. Gray NS, Wodicka L, Thunnissen AMWH et al. Exploiting chemical libraries, structure, and
genomics in the search for kinase inhibitors. Science 1998; 281:533-538.
32. Chao SH, Fujinaga K, Marion JE et al. Flavopiridol inhibits P-TEFb and blocks HIV-1 replication.
J Biol Chem 2000; 275:28345-28348.
33. Kerr JS, Wexler RS, Mousa SA et al. Novel small molecule alpha v integrin antagonists: Comparative
anti-cancer efficacy with known angiogenesis inhibitors. Anticancer Res 1999; 19:959-968.
34. Melillo G, Sausville EA, Cloud K et al. Flavopiridol, a protein kinase inhibitor, down-regulates
hypoxic induction of vascular endothelial growth factor expression in human monocytes. Cancer
Res 1999; 59:5433-5437.
35. Schnier JB, Kaur G, Kaiser A et al. Identification of cytosolic aldehyde dehydrogenase 1 from nonsmall
cell lung carcinomas as a flavopiridol-binding protein. FEBS Lett 1999; 454.
36. Kaiser AU, Nishi K, Walsh DA et al. The cyclin-dependent kinase inhibitor flavopiridol inhibits
glycogen phosphorylase and affects glucose metabolism. Clin Cancer Res 1999; 5 (suppl):A121 (abstr).
37. Oikonomakos NG, Schnier JB, Zographos SE et al. Flavopiridol inhibits glycogen phosphorylase
by binding at the inhibitor site. J Biol Chem 2000; 275:34566-34573.
38. Czech J, Hoffmann D, Naik R et al. Antitumoral activity of flavone L86-8275. Int J Oncol 1995;
6:31-36.
39. Drees M, Dengler WA, Roth T et al. Flavopiridol (L86-8275): Selective antitumor activity in vitro
and activity in vivo in prostate carcinoma cells. Clin Cancer Res 1997; 3:273-279.
40. Jager W, Zembsch B, Wolschann P et al. Metabolism of the anticancer drug flavopiridol, a new
inhibitor of cyclin dependent kinases, in rat liver. Life Sci 1998; 62:1861-1873.
41. Senderowicz AM, Headlee D, Stinson SF et al. Phase I trial of continuous infusion flavopiridol, a
novel cyclin-dependent kinase inhibitor, in patients with refractory neoplasms. J Clin Oncol 1998;
16:2986-2999.
42. Thomas J, Cleary J, Tutsch K et al. Phase I clinical and pharmacokinetic trial of flavopiridol. Proc
Am Assoc Cancer Res 1997; 38:A1496 (abstr).
43. Wright J, Blatner GL, Cheson BD. Clinical trials of flavopiridol. Oncol 1998; 12:1018-1024.
44. Stadler WM, Vogelzang NJ, Amato R et al. Flavopiridol, a novel cyclin-dependent kinase inhibitor,
in metastatic renal cancer: a University of Chicago phase II consortium study. J Clin Oncol
2000; 18:371-375.
45. Schwartz GK, Ilson D, Saltz L et al. A phase II study of the cyclin-dependent kinase inhibitor
flavopiridol administered in patients with advanced gastric carcinoma. J Clin Oncol 2001;
19:1985-1992.
46. Shapiro GI, Supko JG, Paterson A et al. A phase II trial of the cyclin-dependent kinase inhibitor
flavopiridol in patients with previously untreated stage IV non-small cell lung cancer. Clin Cancer
Res 2001; 7:1590-1599.
47. Innocenti F, Stadler WM, Iyer L et al. Flavopiridol metabolism in cancer patients is associated
with the occurrence of diarrhea. Clin Cancer Res 2000; 6:3400-3405.
48. Shepherd FA, Dancey J, Ramlau R et al. Prospective randomized trial of docetaxel versus best
supportive care in patients with non-small cell lung cancer previously treated with platinum-based
chemotherapy. J Clin Oncol 2000; 18:2095-2103.
49. Lynch Jr TJ, Kalish L, Strauss G et al. Phase II study of topotecan in metastatic non-small cell
lung cancer. J Clin Oncol 1994; 12:347-352.
50. Bennett P, Mani S, O’Reilly S et al. Phase II trial of flavopiridol in metastatic colorectal cancer:
Preliminary results. Proc Am Soc Clin Oncol 1999; 18:A1065 (abstr).
51. Schiller JH, Harrington D, Sandler A et al. A randomized phase III trial of four chemotherapy
regimens in advanced non-small cell lung cancer (NSCLC). Proc Am Soc Clin Oncol 2000;
19:A2 (abstr).
52. Senderowicz AM, Messmann R, Arbuck S et al. A phase I trial of 1 hour infusion of flavopiridol,
a novel cyclin-dependent kinase inhibitor, in patients with advanced neoplasms. Proc Am Soc Clin
Oncol 2000; 19:A796 (abstr).
53. Zhai S, Figg WD, Headlee D et al. Pharmacokinetics (PK) of flavopiridol IV bolus in patients
with refractory neoplasms. Clin Cancer Res 2000; 6 (suppl):A314 (abstr).
54. Bible KC, Kaufmann SH. Cytotoxic synergy between flavopiridol and various antineoplastic agents:
The importance of sequence of administration. Cancer Res 1997; 57:3375-3380.
55. Schwartz GK, Farsi K, Maslak P et al. Potentiation of apoptosis by flavopiridol in Mitomycin-Ctreated
gastric and breast cancer cells. Clin Cancer Res 1997; 3:1467-1472.
Small Molecule Inhibitors of Cyclin-Dependent Kinases 231
56. Motwani M, Delohery TM, Schwartz GK. Sequential dependent enhancement of caspase activation
and apoptosis by flavopiridol on Paclitaxel-treated human gastric and breast cancer cells. Clin Cancer
Res 1999; 5:1876-1883.
57. Gumerlock PH, Mack PC, Lau AH et al. Aberrant nuclear mitosis associated with increased apoptosis
from Paclitaxel followed by flavopiridol in non-small cell lung carcinoma (NSCLC) cell lines. Lung
Cancer 2000; 29:A243.
58. Schwartz GK, Kaubisch A, Saltz L et al. Phase I trial of sequential Paclitaxel and cisplatin in
combination with the cyclin-dependent kinase inhibitor flavopiridol in patients with advanced solid
tumors. Clin Cancer Res 1999; 5 (suppl):A122 (abstr).
59. Jung CP, Motwani MV, Schwartz GK. Flavopiridol potentiates gemcitabine-induced apoptosis in
association with downregulation of ribonucleotide reductase. Proc Am Assoc Cancer Res 2000;
41:A209 (abstr).
60. Matranga CB, Shapiro GI. Selective sensitization of transformed cells to flavopiridol-induced
apopotosis during S phase. Proc Am Assoc Cancer Res 2000; 41:A210 (abstr).
61. Motwani MV, Schwartz GK. Flavopiridol potentiates the SN-38-induced apoptosis in association
with downregulation of cyclin dependent kinase inhibitor p21WAF1/Cip1 in HCT116 cells. Proc Am
Assoc Cancer Res 2000; 41:A208 (abstr).
62. Krek W, Ewen ME, Shirodkar S et al. Negative regulation of the growth-promoting transcription
factor E2F-1 by a stably bound cyclin A-dependent protein kinase. Cell 1994; 78:161-172.
63. Dynlacht BD, Flores O, Lees JA et al. Differential regulation of E2F transactivation by cyclin-
CDK2 complexes. Genes Devel 1994; 8:1772-1786.
64. Xu M, Sheppard KA, Peng CY et al. Cyclin A/CDK2 binds directly to E2F-1 and inhibits the
DNA-binding activity of E2F-1/DP-1 by phosphorylation. Mol Cell Biol 1994; 14:8420-8431.
65. Krek W, Xu G, Livingston DM. Cyclin A-kinase regulation of E2F-1 DNA binding function
underlies suppression of an S phase checkpoint. Cell 1995:1149-1158.
66. Rich T, Allen RL, Wyllie AH. Defying death after DNA damage. Nature 2000; 407:777-783.
67. Chen YNP, Sharma SK, Ramsey TM et al. Selective killing of transformed cells by cyclin/cyclindependent
kinase 2 antagonists. Proc Natl Acad Sci USA 1999; 96:4325-4329.
68. Shapiro GI. Potentiation of flavopiridol-mediated apoptosis in non-small cell lung cancer cell lines
following recruitment of cells to S phase. Proc Am Assoc Cancer Res 1999; 40:A2025 (abstr).
69. Achenbach TV, Muller R, Slater E. Synergistic antitumor effect of chemotherapy and antisensemediated
ablation of the cell cycle inhibitor p27Kip1. Clin Cancer Res 2000; 6:3006-3014.
70. Chien M, Astumian M, iebowitz D et al. In vitro evaluation of flavopiridol, a novel cell cycle
inhibitor, in bladder cancer. Cancer Chemother Pharmacol 1999; 44:81-87.
71. Bible KC, oerner SA, Kirkland K et al. Characterization of an ovarian carcinoma cell line resistant
to cisplatin and flavopiridol. Clin Cancer Res 2000; 6:661-670.
72. Hsueh CT, Kelsen D, Schwartz GK. UCN-01 suppresses thymidylate synthase gene expression and
enhances 5-fluorouracil-induced apoptosis in a sequence-dependent manner. Clin Cancer Res 1998;
4:2201-2206. 73. Motwani M, Li X, Schwartz GK. Flavopiridol, a cyclin-dependent
kinase inhibitor, prevents spindle inhibitor-induced endoredupliction in human cancer cells. Clin
Cancer Res 2000; 6:924-932.
74. Hooijberg JH, Broxterman HJ, Scheffer GL et al. Potent interaction of flavopiridol with MRP1.
Brit J Cancer 1999; 81269-276.
75. Smith V, Workman P, Kelland LR. Drug resistance to the cyclin-dependent kinase (CDK) inhibitor,
flavopiridol. Proc Am Asso Cancer Res 2000; 41:A207 (abstr).
76 Kelland LR. Flavopiridol, the first cyclin-dependent kinase inhibitor to enter the clinic: Current
status. Exp Opin Invest Drugs 2000; 9:2903-2911.
77. Robey RW, Medina-Perez WY, Nishiyama K et al. Overexpression of the ATP-binding cassette
half-transporter, ABCG2 (MXR/BCRP/ABCP1) in flavopiridol-resistant human breast caner cells.
Clin Cancer Res 2001; 7:145-152.
78. Schlegel S, Klimeck W, List AF. Breast cancer resistance protein (BCRP) associated drug export
and p53 inactivation impact flavopiridol antitumor activity. Proc Am Assoc Cancer Res 1999;
40:A4415 (abstr).
79. Zaharevitz DW, Gussio R, Leost M et al. Discovery and initial characterization of the paullones, a novel
class of small- molecule inhibitors of cyclin-dependent kinases. Cancer Res 1999; 59:2566-2569.
80. Schultz C, Link A, Leost M et al. Paullones, a series of cyclin-dependent kinase inhibitors:
Synthesis, evaluation of CDK1/cycli B inhibition, and in vitro antitumor activity. J Med Chem
1999; 42:2909-2919.
81. Kunick C, Schultz C, Lemcke T et al. 2-substituted paullones: CDK1/cyclin B-inhibiting property
and in vitro antipoliferative activity. Bio Med Chem Lett 2000; 10:567-569.
232 Cell Cycle Checkpoints and Cancer
82. Price T, Walton MI, Titley J et al. Studies on the mechanism of action of the cycli dependent
kinase inhibitors 9-nitro and 10-bromopuallone in human colon tumor cell lines. Clin Cancer Res
2000; 6:A319 (abstr).
83. Meijer L, Pondaven P. Cyclic activation of histone H1 kinase during sea urchin egg mitotic
divisions. Exp Cell Res 1988; 14:116-129.
84. Neant I, Guerrier P. 6-dimethylaminopurine blocks stARFish oocyte maturation by inhibiting a
relevant protein kinase activity. Exp Cell Res 1988; 176:68-79.
85. Rialet V, Meijer L. A new screening test for antimitotic compounds using the universal M phasespecific
protein kinase p34cdc2/cyclin B cdc13, affinity-immobilized on p13suc1-coated
microtitration plates. Anticancer Res 1991; 11:1581-1590.
86. Havlicek L, Hanus J, Vesely J et al. Cytokinin-derived cyclin-dependent kinase inhibitors: Synthesis
and cdc2 inhibitory activity of olomoucine and related compounds. J Med Chem 1997; 40:408-412.
87. Schultze-Gahmen U, Brandsen J, Jones HD et al. Multiple modes of ligand recognition: Crystal
structures of cyclin-dependent kinase 2 in complex with ATP and two inhibitors, olomoucine and
ispentenyladenine. Proteins 1995; 22:378-391.
88. De Azevedo WF, Leclerc S, Meijer L et al. Inhibition of cyclin-dependent kinases by purine
analogues: crystal structure of human CDK2 complexed with roscovitine. Eur J Biochem 1997;
243:518-526.
89. Noble MEM, Endicott JA. Chemical inhibitors of cyclin-dependent kinases; insights into design
from x-ray crystallographic studies. Pharmacol Ther 1999; 82:269-278.
90. Brooks EE, Gray NS, Joly A et al. CVT-313, a specific and potent inhibitor of CDK2 that prevents
neointimal proliferation. J Biol Chem 1997; 272:29207-29211.
91. Meijer L, Borgne A, Mulner O et al. Biochemical and cellular effects of roscovitine, a potent and
selective inhibitor of the cyclin-dependent kinases cdc2, CDK2 and CDK5. Eur J Biochem 1997;
243:527-536.
92. Buquet-Fagot C, Lallemand F, Montagne MN et al. Effects of olomoucine, a selective inhibitor of
cyclin-dependent kinases, on cell cycle progression in human cancer cell lines. Anticancer Drugs
1997; 8:623-631.
93. Alessi F, Quarta S, Savio M et al. The cyclin-dependent kinase inhibitors olomoucine and roscovitine
arrest human fibroblasts in G1 phase by specific inhibition of CDK2 activity. Exp Cell Res 1998;
245:8-18.
94. Schutte B, Nieland L, van Engeland M et al. The effect of the cyclin-dependent kinase inhibitor
olomoucine on cell cycle kinetics. Exp Cell Res 1997; 236:4-15.
95. Rosania GR, Merlie J, J., Gray N et al. A cyclin-dependent kinase inhibitor inducing cancer cell
differentiation: Biochemical identification using Xenopus egg extracts. Proc Natl Acad Sci USA
1999; 96:4797-4802.
96. Tamaoki T. Use and specificity of staurosporine, UCN-01, and calphostin C as protein kinase
inhibitors. Methods Enzymol 1991; 201:340-347.
97. Takahashi I, Kobayashi E, Asano K et al. UCN-01, a selective inhibitor of protein kinase C from
Streptomyces. J Antibiotic (Tokyo) 1987; 40:1782-1784.
98. Takahashi I, Saitoh Y, Yoshida M et al. UCN-01 and UCN-02, new selective inhibitors of protein
kinase C. II. Purification, physico-chemical properties, structural determination and biologic activities.
J Antibiot (Tokyo) 1989; 42:571-576.
99. Seynaeve CM, Kazanietz MG, Blumberg PM et al. Differential inhibition of protein kinase C
isozymes by UCN-01, a staurosporine analogue. Mol Pharmacol 1994; 45:1207-1214.
100. Courage C, Bradder SM, Jones T et al. Characterisation of novel human lung carcinoma cell lines
selected for resistance to anti-neoplastic analogues of staurosporine. Int J Cancer 1997; 73:763-768.
101. Wang Q, Worland PJ, Clark JL et al. Apoptosis in 7-hydroxystaurosporine-treated T lymphoblasts
correlates with activation of cyclin-dependent kinases 1 and 2. Cell Growth Differ 1995; 6:927-936.
102. Seynaeve CM, Stetler-Stevenson M, Sebers S et al. Cell cycle arrest and growth inhibition by the protein
kinase antagonist UCN-01 in human breast carcinoma cells. Cancer Res 1993; 53:2081-2086.
103. Akinaga S, Nomura K, Gomi K et al. Effect of UCN-01, a selective inhibitor of protein kinase C,
on the cell-cycle distribution of human epidermoid carcinoma, A431 cells. Cancer Chemother
Pharmacol 1993; 33:273-280.
104. Kawakami K, Futami H, Takahara J et al. UCN-01, 7 hydroxyl-staurospaurine, inhibits kinase
activity of cyclin-dependent kinases and reduces the phosphorylation of the retinoblastoma
susceptibility gene product in A549 human lung cancer cell line. Biochem Biophys Res Comm
1996; 219:778-783.
Small Molecule Inhibitors of Cyclin-Dependent Kinases 233
105. Akiyama T, Yoshida T, Tsujita T et al. G1 phase accumulation induced by UCN-01 is associated
with dephosphorylation of Rb and CDK2 proteins as well as induction of CDK inhibitor p21/
Cip1/WAF1/Sdi1 inp53-mutated human epidermoid carcinoma A431 cells. Cancer Res 1997;
57:1495-1501.
106. Sugiyama K, Akiyama T, Shimizu M et al. Decrease in susceptibility toward induction of apoptosis
and alteration in G1 checkpoint funtion as determinants of resistance of human lung cancer cells
against the antisignaling drug UCN-01 (7-hydroxystaurosporine). Cancer Res 1999; 59:4406-4412.
107. Mack PC, Gandara DR, Bowen C et al. RB status as a deerminant of response to UCN-01 in
non-small cell lung carcinoma. Clin Cancer Res 1999; 5:2596-2604.
108. Chen X, Lowe M, Keyomarsi K. UCN-01-mediated G1 arrest in normal but not tumor breast
cells is pRb-dependent and p53-independent. Oncogene 1999; 18:5691-5702.
109. Shao RG, Shimizu T, Pommier Y. 7-hydroxystaurosporine (UCN-01) induces apoptosis in human
colon carcinoma and leukemia cells independently of p53. Exp Cell Res 1997; 234:388-397.
110. Nieves-Neira W, Pmmier Y. Apoptotic response to camptothecin and 7-hydroxystaurosporine (UCN-
01) in the 8 human breast cancer cell lines of the NCI anticancer drug screen: Multifactorial
relationships with topoisomerase I, protein kinase C, Bcl-2, p53, MDM-2 and caspase pathways.
Int J Cancer 1999; 82:396-404.
111. Shi L, Nishioka WK, Th’ng J et al. Premature p34cdc2 activatio required for apoptosis. Science
1994; 263:1143-1145.
112. Wang Q, Fan S, Eastman A et al. UCN-01: A potent abrogator of G2 checkpoint function in
cancer cells with disrupted p53. J Natl Cancer Inst 1996; 88:956-965.
113. Bunch RT, Eastman A. Enhancement of cisplatin-induced cytotoxicity by 7-hydroxystarusporine
(UCN-01), a new G2 checkpoint inhibitor. Clin Cancer Res 1996; 2:791-797.
114. Weinert T. A DNA damage checkpoint meets the cell cycle engine. Science 1997; 277:1450-1451.
115. Furnari B, Rhind N, Russell P. Cdc25 mitotic inducer targeted by Chk1 DNA damage checkpoint
kinase. Science 1997; 277:1501-1505.
116. Sanchez Y, Wong C, Thoma S et al. Conservation of the Chk1 checkpoint pathway in mammals:
Linkage of DNA damage to CDK regulation through Cdc25. Science 1997; 277:1497-1501.
117. Peng CY, Graves PR, Thoma RS et al. Mitotic and G2 checkpoint control: regulation of 14-3-3
protein binding by phosphorylation of Cdc25C on serine-216. Science 1997; 277:1501-1505.
118. Graves PR, Yu L, Schwarz JK et al. The Chk1 protein kinase and the Cdc25C regulatory pathways
are targets of the anticancer agent UCN-01. J Biol Chem 2000; 275:5600-565.
119. Busby EC, Leistritz DF, Abraham RT et al. The radiosensitizing agent 7-hydroxystaurosporine
(UCN-01) inhibits the DNA damage checkpoint kinase hChk1. Cancer Res 2000; 60:2108-2112.
120. Bunz F, Dutriaux A, Lengauer C et al. Requirement for p53 and p21 to sustain G2 arrest after
DNA damage. Science 1998; 282:1497-1501.
121. Chan TA, Hermeking H, Lengauer C et al. 14-3-3σ is required to prevent mitotic catastrophe
after DNA damage. Nature 1999; 401:616-620.
122. Knudsen KE, Booth D, Naderi S et al. RB-dependent S-phase response to DNA damage. Mol Cell
Biol 2000; 20:7751-7763.
123. Shao RG, Cao CX, Shimizu T et al. Abrogation of an S-phase checkpoint and potentiation of
camptothecin cytotoxicity by 7-hydroxystaurosporine (UCN-01) in human cancer cell lines, possibly
influenced byp53 function. Cancer Res 1997; 57:4029-4035.
124. Bunch RT, Eastman A. 7-Hydroxystaurosporine (UCN-01) causes redistribution of proliferating
cell nuclear antigen and abrogates cisplatin-induced S phase arrest in Chinese hamster ovary cells.
Cell Growth Differ 1997; 8:779-788.
125. Shi Z, Plunkett W. Inhibition of the PI 3-kinase-Akt-Bad survival pathway by UCN-01 is associated
with abrogation of gemcitabine-induced S phase arrest and induction of apoptosis. Proc Am
Assoc Cancer Res 2000; 41:A1989 (abstr).
126. Hsueh C, Chiu c, Schwartz GK. UCN-01 supresses E2F-1 mediated by ubiquitin-proteasomedependent
degradation. Clin Cancer Res 2001; 7:669-674.
127. Jiang H, Yang LY. Cell cycle checkpoint abrogator UCN-01 inhibits DNA repair: association with
attenuation of the interaction of XPA and ERCC1 nucleotide excision repair proteins. Cancer Res
1999; 59:4529-4534.
128. Akinaga S, Gomi K, Morimoto M et al. Antitumor activity of UCN-01, a selective inhibitor of
protein kinase C, in murine and human tumor models. Cancer Res 1991; 51:4888-4892.
129. Kurata N, Kuwabara T, Tanii H et al. Pharmacokinetics and pharmacodynamics of a novel protein
kinase inhibitor, UCN-01. Cancer Chemother Pharmacol 1999; 44:12-18.
130. Sausville EA, Arbuck SG, Messmann R et al. Phase I trial of 72-hour continuous infusion UCN-01,
inhibitor, in patients with refractory neoplasms. J Clin Oncol 2001; 19:2379-2333
234 Cell Cycle Checkpoints and Cancer
131. Fuse E, Tanii H, Kurata N et al. Unpredicted clinical pharmacology of UCN-01 caused by
specific binding to human alpha1-acid glycoprotein. Cancer Res 1998; 58:3248-3253.
132. Dees EC, O’Reilly S, Figg WD et al. A phase I and pharmacologic study of UCN-01, a protein
kinase C inhibitor. Proc Am Soc Clin Oncol 2000; 19:A797 (abstr).
133. Tamura T, Sasaki Y, Minami H et al. Phase I study of UCN-01 by 3-hour infusion. Proc Am Soc
Clin Oncol 1999; 18:A611 (abstr).
134. Gumerlock PH, Mack PC, Lau AH et al. Increased activity of the cisplatin to UCN-01 treatment
is associated with M-phase and p53-independent induction of p27Kip1 in NSCLC cells. Lung Cancer
2000; 29:A39 (abstr).
135. Walker DH, Luzzio M, Veal J et al. The novel cyclin dependent kinase inhibitors, GW5181 and
GW9499 regulate cell cycle progession and induce tumor-selective cell death. Proc Am Assoc Cancer
Res 1999; 40:A4783 (abstr).
136. Davis ST, Benson BG, Bramson HN et al. Prevention of chemotherapy-induced alopecia in rats by
CDK inhibitors. Science 2001; 291:134-137.
137. Owa T, Yoshino H, Okauchi T et al. A novel antitumor agent ER-35744, tageting G1 phase. I.
Discovery and structure-activity relationships. Proc Am Assoc Cancer Res 1996; 37:A2664 (abstr).
138. Watanabe T, Sugi N, Ozawa Y et al. A novel antitumor agent ER-35744, tageting G1 phase. III.
Studies of mechanism of action. Proc Am Assoc Cancer Res 1996; 37:A2667 (abstr).
139. Sugi N, Ozawa Y, Watanabe T et al. A novel antitumor agent ER-35744, tageting G1 phase. II.
Antitumor activities in vitro and in vivo. Proc Am Assoc Cancer Res 1996; 37:A2668 (abstr).
140. Ozawa Y, Funahashi Y, Senba T et al. Antitumor effect of E7070(ER-35744) against orthotopically
transplanted mouse and human colorectal tumors. Proc Am Assoc Cancer Res 1997; 38:A3163 (abstr).
141. Raymond E, Fumoleau P, Roche H et al. Combined results of 4 phase I and pharmacokinetic
(PK) studies of E7070, a novel chloroindolyl-sulphonamide inhibiting the activation of CDK2 and
cyclin E. Clin Cancer Res 2000; 6 (suppl):A315 (abstr).
142. Raymond E, Fumoleau P, roche H et al. Pharmacokinetic study of E7070, a novel sulphonamide,
combining data from four phase I trials. Proc Am Assoc Cancer Res 1999; 40:A2545 (abstr).
143. Ozawa Y, Kai J, Kusano K et al. Synergistic effect of E7070 combined with CPT-11 in human
tumor xenograft models. Clin Cancer Res 2000; 6 (suppl):A502 (abstr).
144. Jiang H, Chou HS, Zhu L. Requirement of cyclinE-CDK2 inhibition in p16INK4A-mediated growth
suppression. Mol Cell Biol 1998; 18:5284-5290.
145. Lundgren K, Price SM, Escobar J et al. Diaminothiazoles: Potent, selective cyclin-dependent
kinase inhibitors with anti-tumor activity. Clin Cancer Res 1999; 5 (suppl):A125 (abstr).
1 46. Fry DW, Garrett MD. Inhibitors of cyclin-dependent kinases as therapeutic agents for the treatment
of cancer. Curr Opin Oncol Endocr Metab Invest Drugs 2000; 2:40-59.
147. Fry DW, Harvey PH et al. Cell cycle and biochemical effects of PD 0183812- a potent inhibitor
of cyclin- D-dependent kinase CDK4 and CDK6. J Biol Chem 2001; 276:16617-16623.
148 . Vesely J, Havlicek L, Strnad M et al. Inhibition of cyclin-dependent kinases by purine derivatives.
Eur J Biochem 1994; 224:771-786.
149 . Kitagawa M, Okabe T, Ogino H et al. Butyrolactone I, a selective inhibitor of CDK2 and cdc2
kinase. Oncogene 1993; 8:2425-2432.
150 . Ohashi M, Sugikawa E, Nakanishi N. Inhibition of p53 phosphorylation by 9-hydroxyellipticine: a
possible anticancer mechanism. Jpn J Cancer Res 1995; 86:819-827.
151. Bojanowski K, Nishio K, Fukuda M et al. Effect of suramin on p34cdc2 kinase in vitro and in
extracts from human H69 cells: Evidence for a double mechanism of action. Biochem Biophys Res
Commun 1994; 203:1574-1580.
152 . Park SG, Cheon JY, Lee YH et al. A specific inhibitor of cyclin-dependent protein kinases, cdc2
and CDK2. Mol Cells 1996; 6:679-683.
153 . Hoessel R, Leclerc S, Endicott JA et al. Indirubin, the active constitutent of a Chinese antileukaemia
medicine, inhibits cyclin-dependent knases. Nature Cell Biol 1999; 1:60.
CHAPTER 14
Cell Cycle Molecular Targets and Drug
Discovery
John K. Buolamwini
Abstract
There have emerged, within the aberrant cell cycle regulatory pathways frequently en
countered in cancer cells, several potential targets for novel anticancer drug
discovery.Cyclin-dependent kinases (CDKs) and their regulatory units, cyclins, play a
central role in cell cycle progression, and several have been shown to be viable anticancer targets.
The inhibition of kinase catalytic activity has been successfully achieved with small molecules
that have advanced into clinical trials for cancer therapy. As the molecular players in the
pathways and cascades involved in progression through the cell cycle are uncovered, more
potential anticancer molecular targets are emerging including critical oncogenic kinases and
regulatory proteins identified in the progression through mitosis. These include aurora kinases,
polo-like kinases, and the anti-apoptotic protein survivin.
Introduction
The cell cycle comprises a complex set of sequential, well coordinated specific events that
result in cell division,1-3 and is therefore central to proliferative diseases such as cancer.
Although the final stages of the cell cycle, mitosis, has been the target of conventional chemotherapy
with drugs like the vinca alkaloids and taxanes, this has been largely nonspecific and
cytotoxic. Recent advances in the molecular characterization of cell cycle regulatory pathways
have uncovered component oncogenic and tumor suppressor genes that are mutated and/or
aberrantly expressed in many human cancers and show potential as targets for novel cancer
therapies. This Chapter focuses on the emerging and potential molecular targets in cell cycle
regulatory pathways and their exploitation for small molecule drug design and discovery.
Cell cycle progression is generally divided into four phases: G1, S, G2 and M2 phases as
depicted in Figure 1. The progression of the cell division cycle depends on the catalytic activity
of cyclin-dependent kinases, a class of serine threonine kinases that require binding to regulatory
subunits known as cyclins for their activation. It is these cyclin-CDK complexes that
constitute the central players (“work horses”) that drive cell division cycle. Whereas the cellular
levels of CDKs remain relatively constant throughout the cell cycle, the levels of cyclins fluctuate
in a periodic and sequential manner, with particular cyclins expressed at specific stages of
the cell cycle. Progress has been made in exploiting the new cell cycle molecular targets, primarily
cyclin-dependent kinases for cancer chemotherapy. Clinical trials are on-going with two
small molecule drug candidates in this regard. The control of cell cycle progression is achieved
primarily by a sequential expression and degradation of cyclins, which bind to specific cyclindependent
kinases (CDKs) or groups thereof. CDK-cyclin Binding interactions that drive cell
cycle progression are as follows: CDK1 (or CDC2) binds cyclins A and B to modulate G2-M
Cell Cycle Checkpoints and Cancer, edited by Mikhail V. Blagosklonny. ©2001 Eurekah.com.
236 Cell Cycle Checkpoints and Cancer
transition, CDK2 binds to cyclins A, D, and E to drive the G1-S transition and S phase,
CDK4 and CDK6 bind cyclin D for progression through G1, whereas CDK7 binds cyclin H,
interestingly, for the activation of other cyclin-CDK complexes.4,5 Some CDKs may bind noncyclin
counterparts, such as the binding of CDK5 to p35, a protein that shares no sequence
homology to the cyclins. CDK7 has been found to be a component of the basal transcription
factor TFIIH as well. Interestingly, it has been shown that cyclin D binds to and activates the
estrogen receptor. Other non-cell cycle associations of CDKs and cyclins are the following:
CDK 8 binds to cyclin C, and CDK9 binds cyclin T in transcription regulation.
Events in Cell Cycle Progression
A simplified outline of the cell cycle progression is depicted in Figure 1. Quiescent cells in
(G0) enter the G1 phase to be prepared for the DNA synthesis as a result of mitogenic signals.6
At this initial stage, D cyclins are synthesized, which bind to and activate CDK4 and CDK6 to
begin the phosphorylation of the retinoblastoma gene product, pRb. Subsequently, cyclin E is
synthesized to activate CDK2 for further phosphorylation of pRb to move the cell cycle through
late G1 and carry it through the G1-S phase transition. Hyperphosphorylated pRb releases
transcription factor E2F and associated proteins for the transcription of genes necessary for cell
cycle progression. The cell then enters the S phase for DNA synthesis, and cyclin A is expressed
to activate its associated CDKs, CDK2 and CDK1 (Cdc2), to take the cell through the S and
G2 phases. Following the completion of the G2 phase, cyclin B-CDK1 complexes carry the
cell through mitosis (M phase) to complete the division cycle.7,8
In addition to the central players in cell cycle progression, there are complex auxiliary
networks of kinases, phosphatases and proteases as well as regulatory proteins5 that feed into
the mainstream to promote progression or to halt it (checkpoints). For example CDK activating
protein kinases (CAKs) such as cyclin H-CDK7 and protein phosphatases, such as the dual
specific Cdc25 protein phosphatase family, serve to activate CDKs and thereby promote cell
cycle progression.9-12 CAKs activate CDKs by phosphorylation of threonine residues (Thr160
Fig. 1. A simplified representation of the phases and molecular players in cell cycle progression.
Cell Cycle Molecular Targets and Drug Discovery 237
or 161 in CDK1 and CDK2, respectively) in the T-loop of CDKs, whereas Cdc25 dual-specific
phosphatase family activates CDKs by dephosphorylation of Thr14 and Tyr159.
Built-in surveillance mechanisms termed checkpoints exit in cell cycle transition phases to
ensure that the integrity of the genome is maintained, as well as its proper apportioning between
the two daughter cells. The G1 and G2 checkpoints monitor DNA status and prevent
inappropriate entry into the phase or mitosis, respectively.13,14 The spindle checkpoint ensures
correct chromatid alignment on the spindle prior to the beginning of anaphase.15,16 Cell cycle
checkpoint abrogation is an emerging therapeutic strategy for sensitizing cancer cells to DNA
damaging agents.17
Regulatory Pathways
To keep the wheels of cell division cycle spinning, CDKs must be activated. Several positive
and negative CDK regulatory proteins and enzymes have been identified, and the list is
sure to continue as much still remains to be known about the mechanisms that control the cell
cycle. Some of these regulatory mechanisms will be highlighted and potential cancer molecular
targets pointed out. Figure 2 shows a simple representation of the fate that may befall cyclin-
CDK complexes.7,18,19 Following the binding of an appropriate cyclin to its target CDK(s), the
complex is mostly activated through phosphorylation by a CDK activating kinase such as the
cyclin H-CDK7 complex.9 Two families of cell cycle regulatory proteins have been found to
exert their effects primarily by binding to and inactivating cyclin dependent kinases in the G1
phase of the cell cycle,20 and have therefore been termed cyclin-dependent kinase inhibitors
(CKIs). CKIs are classified into the Cip/Kip and Ink4 families.
The Cip/Kip family was the first to be recognized and currently consists of three proteins,
p21Cip/WAF1 (the first CKI,21-24 p27Kip125 and p57Kip2.21 In vivo, they act primarily on CDK2
complexes.20 It has been shown, however, that CDK4 complexes sequester p27Kip1. CKI p21WAF1/
CIP1 is induced by the p53 tumor suppressor gene to cause G1 arrest upon DNA damage.27,28
Four proteins have been identified in the Ink4 family, namely p16Ink4a,29 p15Ink4b,
p18Ink4c and p19Ink4d.20 They are basically inhibitors of cyclin D-CDK complexes and are
expressed in a cell-type specific manner. Ink4a has been classified as a tumor suppressor gene29,30
and has been found to be frequently mutated in melanomas.31 X-ray crystallographic studies
have indicated that CKIs such as p21WAF1/CIP1 inhibit CDKs by binding to the cyclin-CDK
complexes and using their 310 helix to block the ATP-binding site.32 Recent studies show an
increasingly more complex interplay among cell cycle regulating proteins than was thought
previously, especially the interaction between CDKs and CKIs.5 Another negative factor in the
CDK regulation is phosphorylation at the G2 checkpoint. Phosphorylation of amino acid
residues Thr 14 and Tyr 15 on CDK1 by Myt1 and Wee1 kinases18,33 keeps it inactive to
prevent mitosis until earlier cell cycle events are properly completed. This inactivation is
reversed by dephosphorylation of Thr 14 and Tyr 15 by the Cdc25 protein phosphatase family
to generate an active CDK.7 The activity of Cdc25 is kept in check through phosphorylation at
its Ser 216 residue by Ch1 kinase. This is thought to be the reason that G2 checkpoint
abrogation is achieved by inhibiting Chk1 kinase with the staurosporine derivative UCN-01
(1).5,27 The phosphorylation of Cdc25 allows binding to the 14-3-3 proteins, which sequester
Cdc25 and keep it away from cyclinB-CDK1.34 It has also recently been demonstrated that
14-3-3 sigma, a major p53-induced protein following irradiation, sequesters cyclin B1 and
cdc2/CDK1 and prevents them from entering the nucleus. These results may indicate a
mechanism for maintaining the G2 checkpoint and preventing mitotic death.35 Furthermore,
it has been shown that polo-like kinases are part of the team that activates Cdc25 phosphatase36
in a loop leading to cyclin B degradation by the proteasome complex that is also activated by
polo-like kinases.37 Activated G1 cyclin-CDK complexes target the retinoblastoma tumor
suppressor gene product, pRb and related “pocket proteins’ p103 (pRb2) and p1072,38-40 which
they phosphorylate and cause to dissociate from their complexes with E2F and related
transcription factors freeing them to induce the transcription of genes to drive the cell division
238 Cell Cycle Checkpoints and Cancer
cycle.41 It appears that pRb interacts also with histone deacetylases (HDAC1) to repress
transcription.42-44 It has recently been shown that p130 (pRb2), which complexes with E2F-4,
may cooperate with p27Kip1 in regulating cell proliferation in a negative feedback loop with
cyclin E.45 In a complex relationship, which is yet to be fully elucidated, p130 (pRb2) and
p107, both inhibit CDK2 kinase activity.
Although elucidation of the mechanisms and molecular players of M-phase progression
has lagged behind that of earlier events of the cell cycle, it is now progressing rapidly and
uncovering potential cancer intervention targets.15,16,33 At the onset of mitosis, cyclin B is
degraded by the proteasome. The mitotic spindle then begins to form with chromosome
condensation in prophase. Following this, chromosomes line up at the equatorial plate
attached to the mitotic spindle in metaphase. The sister chromatids then begin to migrate to
opposite poles of the parent cell in the metaphase to anaphase transition. The anaphasepromoting
complex/cyclosome (APC/C) plays a major role in this metaphase to anaphase
transition,33-46 which it does by ubiquitnating mainly the checkpoint protein Pds1, which
functions to prevent sister chromatids from separating. Polo-like kinases have been shown to
play an extensive role in mitosis, beginning from the onset of M phase, at spindle formation,
during the metaphase-anaphase transition (affecting the APC complex) and at late M phase.36
Mitotic progression from anaphase to telophase involves the disappearance of the mitotic spindle
and the formation of nucleoli and a new nuclear membrane. Cytokinesis ensues after that to
divide the parent cell into two daughter cells. The aurora/Ipl1p family of serine/threonine
kinases is expressed in a cell-cycle regulated fashion, and is a key promoter of chromosome
segregation and cytokinesis.47 Survivin, a newly identified inhibitor of apoptosis (IAP) protein,
48,49 is expressed in the G2/M phase in a cell cycle-regulated fashion and associates with
microtubules of the mitotic spindle at the beginning of mitosis.50 Survivin has also been shown
to control the mitotic spindle checkpoint,51 and was recently shown to have a similar cell-cycle
localization and gene knockout phenotype with the inner centromere protein INCENP.52
Fig. 2. Regulation of CDK-cyclin complexes. Arrow shows positive and hammer shows negative regulation,
respectively.
Cell Cycle Molecular Targets and Drug Discovery 239
Oncogenic Cell Cycle Targets
Table 1 gives a summary of cell cycle target aberrations in human tumors. Oncogenic
expression and/or mutation of cyclins and CDKs, expecially those involved in the G1 phase,
have been found frequently in many human cancers.1-3,7,53,54 Common cancers in which
oncogenic expression of cyclins and CDKs has been demonstrated include B- and T-cell
lymphomas, esophageal cancer, breast cancer, bladder cancer, small cell lung cancer, prostrate
cancer, colon cancer, glioblastoma, neuroblastoma, leukemias, retinoblastoma, melanoma and
adenomatous polyposis.25,55-57 The D and E families25,58 of G1 cyclins are the ones most
frequently associated with oncogenic expression.40,57 The overexpression of cyclin D1 causes
cell type dependent transformation,58-60 sometimes in cooperation with the myc oncogene.61,62
Overexpression or gene amplification of D cyclins is particularly prevalent in breast cancer,
with cyclin D1 overexpressed or amplified in more than 50% of breast cancers.41,63-65 Cyclin
D1 overexpression has also been suggested to play a role in drug resistance to antifolates in a
human fibrosarcoma cell line.66 Amplification of cyclin E is also rampant in breast cancers67,68
and shows a strong correlation with tumor aggressiveness.57 Cyclin E2 has recently been
identified and shown to be overexpressed in human cancers including breast, lung and cervical
cancers, and has been proposed to be a potential cancer-specific molecular target.69 Oncogenic
amplification and overexpression of CDKs have also been reported in gliomas and soft tissue
sarcomas,70 as well as other cancers.31,55,56 The E2F transcription factor gene family has also
been shown to be oncogenic.71
The oncogenic potential of Cdc25 proteins has also been demonstrated. Overexpression
of Cdc25B has been shown in 32% of human primary breast cancers.72 It is interesting that
specific Cdc25B protein phosphatase inhibitors discovered recently were shown to inhibit
tumor growth.73 Chk1 kinase, an important mediator of cell cycle arrest in G2 following DNA
damage, is becoming an important cancer therapeutic target.17,74 A 1.7 A resolution X-ray
crystal structure human Chk1 kinase domain in complex with an ATP analog has recently been
published, providing a template for structure-based drug design.75
New potential oncogenic targets are emerging from the end stage of the cell cycle, i.e.,
mitosis. Disruption of survivin-microtubule interactions results in loss of survivin’s anti-apoptosis
function and increased caspase-3 activity. The overexpression of survivin in cancer is thought
to overcome apoptosis and favor aberrant progression of transformed cells through mitosis.51
In one study, survivin messenger RNA was shown to be overexpressed in 85% of non-small cell
lung tumors.76 Polo-like kinases are also mitotic oncogenic targets found to be overexpressed in
non-small cell lung cancer. Aurora 2 kinase amplification has been shown in more that 50% of
colorectal cancers77 and in 12% of primary breast cancer samples.78 The overexpression of
polo-like kinase has been demonstrated in non-small cell lung cancer.79
Cell Cycle Molecular Target-Based Cancer Drug Discovery
Oncogenic G1 cyclin interactions have not been successfully disrupted, but down regulation
of these cyclins by antisense oligonucleotides to achieve an anticancer effect has been
demonstrated.80,81 Inhibition of the catalytic activity of CDKs has been a successful strategy
for the discovery of novel anticancer agents, some of which are under clinical development82,83
such as flavopiridol (2). This strategy has produced a large number of potent small molecule
CDK inhibitors,4,7,55,56,82,84-92 which will be highlighted in this section.
The CDK inhibitors discovered to date are all essentially ATP-binding competitors at the
kinase domain. The ATP competitive activity was an initial concern for selectivity, since there
is a plethora of kinases in the cell all using ATP. Surprisingly, however, it appears this fear was
unwarranted as excellent selectivity has been demonstrated not only in discrimination between
serine/threonine kinases on the one hand and tyrosine kinases on the other, but also among
very closely related serine/threonine kinases. This selectivity has been attributed to interaction
of inhibitors with regions in the vicinity of the binding site that do not interact with bound
ATP, and where the amino acid residues are not conserved among the CDKs.85 This finding
240 Cell Cycle Checkpoints and Cancer
has advanced the structure-based drug design efforts on CDK inhibitors. The structure-activity
relationships of CDK inhibitors have been reviewed recently.91 The many small molecule
CDK inhibitors discovered to date represent a chemically group of compounds including the
following major ones (structures shown in Fig. 3):
i) flavopiridol (2), which represents flavone class is the most extensively studied selective CDK
inhibitor, and now in clinical trials,
ii) purine derivatives introduced with olomoucine (3) and for which extensive structure-activity
relationship studies have been conducted using state-of-the art drug discovery technologies
including X-ray crystal structure-based computer-aided drug design, combinatorial
chemistry and high throughput screening, as well as functional genomics,93
iii) 7-hydroxy staurosporine derivative UCN-01 (compound 1), which also inhibits G2 check
point kinase Chk1, as well as protein kinase C, represents the bisindolylmaleimide class
and has entered clinical trials,
iv) the dihydroindolo[3,2-d][1]benzazepin-6(5H)-ones also known as paullones, represented
by kenpaullone, 4, are a more recent addition,83,94,95 as well as
Table 1. Oncogenic cell cycle molecular targets
Oncogenic Target Oncogenic Changes Major Tumors
Cyclin D1 Gene amplification
and/or overexpression breast cancer (40-80%)
Familial Polyposis (~70%)
B-Cell Lymphoma (50%)
NSC Lung Cancer (~50%)
Head and neck cancer (~35%)
Esophageal cancer (25-50%)
Bladder cancer (~25%)
Cyclin K (cyclin D-like) KSHV Infection Kaposi’s sarcoma
Cyclin E Amplification Breast cancer (30-80%)
Prostate cancer (~70%)
Ovarian cancer (~18%)
Gastric cancer
Cervical cancer
Cyclin E2 Overexpression Breast cancer
Small cell lung cancer
Cervical cancer
Cyclin B1 Overexpression Colorectal Cancer (~90%)
Cyclin A Stabilization Hepatocellular carcinomas
CDK2 Overexpression Colorectal cancer
CDK4 Amplification Sarcomas and gliomas
Point Mutation Familial melanoma
Cdc25 Overexpression Breast cancer (32%)
Polo-like kinases Overexpression Non-small cell lung cancer
Aurora 2 Amplification Colorectal cancer (>50%)
Breast cancer (12%)
Survivin Overexpression Non-small cell lung (85%)
Cell Cycle Molecular Targets and Drug Discovery 241
v) the indigoids derived from a Chinese antileukemia herbs,96 and represented by indirubin-
3´-monoxime (compound 5), and
vi) a new diaminothiazoles series represented by compound 6.97,98
The staurosporine class is the least selective, inhibiting protein kinase C and Chk1 kinase
activity as well.17,85 Flavopiridol is a relatively general CDK inhibitor that is active against
CDK1, CDK2 and CDK4, whereas the purine analogs olomoucine and congeners inhibit
CDK1 and CDK2 but are much less potent against CDK4.99 The paullone class of CDK
inhibitors, like flavopiridol, is selective towards CDK1 and CDK2, as well as CDK5.83 Additionally,
kenpaullone is relatively inactive against the protein kinase C (PKC) family, compared
to flavopiridol.95 The indigoid series have a more broad selectivity, inhibiting CDK1, CDK2,
CDK4, and CDK5.96 The most active compound in that report, indirubin-5-sulfonic acid
exhibited IC50 values of 0.055, 0.035, 0.15 and 0.3 μM against CDK1/cyclin B, CDK2/cyclin
A, CDK2/cyclin E, CDK4/cyclin D1 and CDK5-p35, respectively.
Cancer Drug Development of Small Molecule CDK Inhibitors
In in vitro studies flavopiridol inhibits CDK2 and CDK4 and causes cell cycle arrest in
G1 phase that is independent of p53 or Rb states,100 as well as G2 arrest.101 Flavopiridol has
been shown to inhibit human tumor xenografts of lung, colon, stomach, breast and brain
cancers. Flavopiridol has been in clinical development sponsored by the National Cancer Institute
for sometime now.101 Dose-limiting toxicities of flavopiridol did not include leukopenia,
anemia or thrombocytopenia,102 but instead have included diarrhea, hypotension and a proinflammatory
syndrome.82 Clinical response has been observed in some solid tumors and lymphomas.
82 UCN-01 causes cell growth arrest in G1 by a mechanism involving dephosphorylation
of the pRb and CDK2, as well as induction of p21WAF1/CIP1/SDI1.103 It has been
shown to potently inhibit the phosphorylation of Cdc25C by Chk1 in vitro, and to abrogate
the G2 checkpoint and sensitize cancer cells to DNA damaging agents.17 It is also currently
undergoing clinical trials.82,101 Dose-limiting toxicities include hyperglycemia, lactic acidosis,
nausea and vomiting, as well as pulmonary toxicity. Among the purine derivatives, olomoucine
inhibits the proliferation of human breast, gastric and pancreatic cancers, and lymphoma.105
Lymphoma cells were shown to be arrested in both G1 and G2 phases by this agent related to
its inhibition of cyclin E/CDK2 and cyclin B/CDK1 complexes.104 Other cellular targets of
the purine class of CDK inhibitors are nucleoside transporters, which they inhibit at micromolar
concentrations.105-107 Olomoucine and roscovitine are said to be possible clinical trials candidates.
108 The paulones, which in addition to inhibiting CDKs have also recently been shown
to inhibit glycogen synthase kinase-3 beta,109 cause cell cycle arrest in the G1 phase and have
shown anticancer activity in the NCI tumor cell line screen.94,95 Induribine-3´-monoxime,
one of the recently identified indigoid compounds, has been shown to inhibit the proliferation
of cancer cells including MCF-7 human breast cancer and human K562 leukemia cells causing
cell cycle arrest in at G2/M.96 Much less is still known about the diaminothiazoles, but they
show anti-tumor activity at submicromolar concentrations and cause cell cylce arrest in both in
G1 and G2 phases, which is associated with the phosphorylation status of pRb.110
Other Targets
The recent discovery of Cdc25 specific small molecule inhibitors73 opens another window
of opportunity for exploiting cell cycle targets in cancer therapy. Fy21-aa09 (7), one of the
Cdc25B inhibitors in the above series, exhibited a Ki of 7.6μM and caused growth arrest of
tsFT210 cells at G2/M phase of the cell cycle. Chk1 kinase inhibition is also an interesting area
to enhance the chemosensitivity of tumors by abrogating the G2 check point. Proteasome
inhibitors have been reported that may be useful for inhibiting CKI or cyclin degradation to
halt cell cycle progression in cancer.111 Other potentially important points of intervention in
cell cycle pathways yet to be exploited are CAK or Cdc25 activation of cyclin-CDK complexes
242 Cell Cycle Checkpoints and Cancer
and the catalytic activity of proteins such as aurora 2 and polo-like kinases.7,33 Polo kinases
contain a distinct region of homology in the C-terminal non-catalytic domain, termed the
polo-box, which is thought to be critical for their oncogenic properties. The development of
chemical compounds that will interact with and disrupt the polo-box of polo-like kinases has
also been envisioned as a potential strategy to for cancer therapy.112,113
Acknowledgments
The manuscript was supported in part by a grant from the National Cancer Institute, No.
CA 80730.
References
1. Schafer A. The cell cycle: A Review. Vet Pathol 1998;35:461-478.
2. Lundberg A S, Weinberg RA. Control of the cell cycle and apoptosis. Eur J Cancer 1999; 35:531-539.
3. Sherr CJ. Cancer cell cycles. Science 1996; 274:1672-1677.
4. Coleman KG, Lyssikatos JP, Yang BV. Chemical Inhibitors of cyclin-dependent kinases. Ann Rep
Med Chem 1997; 32:171-179
5. Sherr C J, Roberts JM. CDK inhibitors: Positive and negative regulators of G1-Phase progression.
Genes Dev 1999; 13:1501-1512
6. Hunter T. Oncoprotein networks. Cell 1997; 88:573-582
Fig. 3. Chemical structures of representative cell cycle target inhibitors
Cell Cycle Molecular Targets and Drug Discovery 243
7. Meijer L. Chemical inhibitors of cyclin-dependent kinases. In Progress in Cell Cycle Research,
Meijer L, Guidet S, Tung HYL, Eds; Plenum Press: New York, 1995; Vol. 1, pp. 351-363.
8. Sherr CJ. Mammalian G1 cyclins. Cell 1993; 73:1059-1065.
9. Solomon MJ, Lee T, Kirschner MW. Role of phosphorylation in p34cdc2 activation: Identification
of an activating kinase. Mol Biol Cell 1992; 3:13-27.
10. Draetta G, Eckstein J. Cdc25 protein phosphatases in cell proliferation. Biochim Biophys Acta
1997; 1332:M53-M63.
11. Morgan D. Cyclin-dependent kinases: engines, clocks and microprocessors. Annu. Rev Cell Dev
Biol 1997; 13:261-291.
12. Jinno S, Suto K, Nagata A et al. Cdc25A is a novel phosphatase functioning early in the cell cycle.
EMBO J 1994; 13:1549-1556.
13. Weinert T. DNA damage checkpoints update: getting molecular. Curr Opin Genet Dev 1998; 8:
185-193
14. O’Connell MJ, Walworth NC. The G2-phase DNA damage checkpoint. Trends Cell Biol 2000;
10:296-303.
15. Gardner RD, Burke DJ. The spindle checkpoint: Two transitions, two pathways. Trends Cell Biol
2000; 10:154-158.
16. Cerutti L, Simanis V. Controlling the end of the cell cycle. Curr Opin Genet Dev 2000; 10:65-69
17. Yu L, Graves P, Tempcyzk A et al. Check point kinase Chk1 is a target of the Anticancer Agent
UCN-01. Proc Am Assoc Cancer Res 1999; 40:Absr 2019.
18. Morgan DO. Principles of CDK regulation. Nature 1995; 374:131-134.
19. Coleman, TR, Dunphy WG. Cdc2 regulatory factors. Curr Opin Cell Biol 1994; 6:877-882.
20. Vidal A, Koff A. Cell-cycle inhibitors: Three families united by a common cause. Gene 2000; 247:1-15.
21. Xiong Y, Zhang H, Beach D. D-type cyclins associated with multiple protein kinases and the
DNA replication and repair factor PCNA. Cell 1992; 71:505-514.
22. Xiong Y, Hannon GJ, Zhang H et al. p21 is a universal inhibitor of cyclin kinases. Nature
1993; 366:701-704.
23. El-Deiry WS, Tokino T, Velculescu VE et al. WAF1, a aotential mediator of p53 tumor suppression.
Cell 1993; 75:817-825.
24. Harper JW, Adami GR, Wei N et al. The p21 CDK-interaction protein, Cip1 is a potent inhibitor
of G1 cyclin-dependent kinases. Cell 1993; 75:805-816.
25. Hunter T, Pines J. Cyclins and cancer II: cyclin D and CDK inhibitors come of age. Cell
1994; 79:573-582.
26. Lee MH, Renisdottir I, Massague J. Cloning of p57KIP2, a cyclin-dependent kinase inhibitor with
unique domain structure and tissue distribution . Genes Dev 1995; 9:639-649.
27. Bates S, Vousden KH. p53 in signaling checkpoint arrest or apoptosis. Curr Opin Genet Dev
1996; 6:12-19.
28. O’Connor PM, Kohn KW. A fundamental role for cell cycle regulation in the chemosensitivity of
cancer cells?. Sem Cancer Biol 1992; 3:409-416.
29. Guan K-L, Jenkins CW, Li Y et al. Growth suppression by p18, q p16INK4/MST1- and p14INK4B/
MST2-related CDK6 inhibitor, correlates with wild-type pRb function. Genes Dev 1994; 8:2939-2952.
30. Nobori TK, Mlura K, Wu DJ et al. Deletion of cyclin-dependent kinase 4 inhibitor gene in multiple
human cancers Nature 1994; 368:753-756.
31. Kamb A. Cyclin-dependent kinase inhibitors and human cancer. Curr Top Microbiol Immunol
1998; 227:139-148.
32. Pines J. Cyclin-dependent kinases: The age of crystals. Biochim Biophys Acta 1997; 1332:M39-M42.
33. Kraker AJ, Booher RN. New targets in cell cycle regulation. Ann Rep Med Chem 1999; 34:247-256.
34. Sanchez Y, Wong C, Thoma RS et al. Conservation of the Chk1 checkpoint pathway in mammals:
Linkage of DNA damage to CDK regulation through Cdc25 Science 1997; 277:1497-501.
35. Chan TA; Hermeking H; Lengauer C et al. 14-3-3 Sigma is required to prevent mitotic catastrophe
after DNA damage. Nature 1999; 401:616-620.
36. Glover DM., Hagan IM, and Tavares Á.A.M. Polo-like kinases: A team that plays throughout
mitosis. Genes Dev 1998; 12:3777-3787.
37. Feng Y, Longo DL, and Ferris D K. Polo-like kinase interacts with proteasomes and regulates their
activity. Cell Growth Different 2001; 12:29-37.
38. Weinberg RA. The retinoblastoma protein and cell cycle control. Cell 1995; 81:323-330
39. Lees EM, Harlow E. Cancer and the cell cycle. In Cell Cycle Control; Hutchison C, Glover DM,
Eds.; IRL Press: New York, 1995; pp.228-263.
40. Draetta G, Pagano M. Cell cycle control and cancer. Ann Rep Med Chem 1996; 31: 241-248.
41. Bernard R. E2F: a nodal point in cell cycle regulation. Biochim Biophys Acta 1997; 1333:M33-M40.
244 Cell Cycle Checkpoints and Cancer
42. Brem, A., Miska, E. A., McCance, D. J et al. Retinoblastoma protein Recruits Histone Deacetylase
to Repress Transcription. Nature 1998; 391:597-601.
43. Magnaghi-Jaulin L, Groisman R, Naguibneva I et al. Retinoblastoma protein represses transcription
by recruiting a histone deacetylase. Nature 1998; 391:601-605.
44. Luo RX, Postigio AA, Dean DC. pRb interacts with histone deacetylase to repress transcription.
Cell 1998; 92: 463-473.
45. Howard CM, Claudio PP, De Luca A et al. Inducible pRb2/p130 Expression and growth suppressive
mechanisms: Evidence of a pRb2/p130, p27KIP1, and cyclin E negative feedback regulatory loop.
Cancer Res 2000; 60:2737-2744.
46. Tyers M, Jorgensen P. Proteolysis and the cell cycle: With this RING I do thee destroy. Curr
Opin Genet Dev 2000; 10:54-64.
47. Bischoff JR and Plowman GD. The Aurora/Ipl1p kinase family: Regulators of chromosome
segregation and cytokinesis. trend cell biol 1999; 9:454-459.
48. Ambrosini G, Adida C, Altieri DC. A novel anti-apoptosis gene, survivin, expressed in cancer and
lymphoma. Nat Med 1997; 3:917-921.
49. Ambrosini G, Adida C, Sirugo G, Altieri DC. Induction of apoptosis and inhibition of cell proliferation
by survivin gene targeting. J Biol Chem 1998; 273:11177-82.
50. Guo M, Hay BA. Cell proliferation and apoptosis. Curr Opin Cell Biol 1999; 11:745-752.
51. Li F, Ambrosini G, Chu EY, Plescia J et al. Control of apoptosis and mitotic spindle checkpoint
by surviving. Nature 1998; 396:580-584.
52. Uren AG, Wong L, Pakusch M et al. Survivin and the inner centromere protein INCENP show
similar cell-cycle localization and gene knockout phenotype Curr Biol 2000; 10:1319-1328.
53. Hartwell L. H, Kastan MB. Cell cycle control and cancer. Science 1994; 266:1821-1828.
54. Dictor M, Ehinger M, Mertens F et al. Abnormal cell cycle regulation in malignancy. Am J Clin
Patho. 1999; 112 (Suppl. 1):S40-S52.
55. Webster KR. The therapeutic potential of targeting the cell cycle. Exp Opin Invest Drugs
1996; 5:1601-1615.
56. Imoto M. Molecular target therapy of cancer: a. Cell cycle. Kagaku Ryo no Ryoiki 1998; 14:13-19
57. Karp JE, Broder S. Molecular foundations of cancer: new targets for intervention. Nat Me. 1995;
1: 309-320.
58. Sherr CJ. G1 phase progression: Cycling on cue. Cell 1994; 79: 551-555
59. Filmus J, Robles AI, Shi W et al. Induction of cyclin D1 overexpression by activated Ras. Oncogene
1994; 9:3627-3633.
60. Daksis JI, Lu RY, Facchini LM et al. Myc induction of cyclin D1 overexpression in the absence of
de novo protein synthesis and links mitogen-stimulated signal transduction to the cell cycle.
Oncogene 1994; 9, 3635-3645.
61. Lovec H, Grzeschiczek A, Kowalski M-B, Moroy T. Cyclin D1/bcl-1 cooperates with myc genes in
the generation of B-cell lymphoma in transgenic mice. EMBO J. 1994; 13: 3487-3495.
62. Bodrug SE, Warner BJ, Bath ML et al. Cyclin D1 transgene impedes lymphocyte maturation and
collaborates in lymphomagenesis with the myc gene. EMBO J 1994; 13: 2124-2130.
63. Buckley, M F, Buckley MF, Sweeney KJ et al. Expression and amplification of cyclin genes in
human breast cancer. Oncogene 1993; 8: 2127-2133.
64. van Diest, PJ, Michalides RJ, Jannink L et al. Cyclin D1 expression in invasive breast cancer.
Correlations and prognostic value. Am J Pathol 1997; 150:705-711.
65. Collecchi P Passoni A Rocchetta M et al. Cyclin-D1 expression in node-positive (N+) and
node-negative (N-) infiltrating human mammary carcinomas. Int J Cancer 1999; 69:139-144.
66. Hochhauser D, Schnieders B, Ercikan-Abali E et al. Effect of cyclin D1 overexpression in a human
fibrosarcoma cell line. J Natl Cancer Inst 1996; 88:1269-1275.
67. Dutta A, Chandra R, Leiter L, Lester S. Cyclins as markers of tumor proliferation: immunocytochemical
studies in breast cancer. Proc. Natl. Acad. Sci. USA 1995; 92:5386-5390.
68. Keyomarsi K, Pardee AB. Redundant cyclin overexpression and gene amplification in breast cancer
cells. Proc.Natl.Acad. Sci. USA 1993; 90:1112-1116.
69. Gudas JM, Payton M, Thukral S et al. Cyclin E2, a novel G1 cyclin that binds CDK2 and is
aberrantly expressed in human cancers. Mol Cell Biol 1999; 19: 612-622.
70. Costello JF, Plass C, Arap W et al. Cyclin-dependent kinase 6 (CDK6) amplification in human
gliomas identified by using two-dimensional separation of genomic DNA. Cancer Res 1997; 57:1250-
1254 and references cited therein.
71. Brooks G and La Thangue, L B. The cell cycle and drug discover: the promise and the hope. Drug
Discovery Today 1999; 4:455-464.
72. Galaktionov K, Lee AK, Eckstein J et al. CDC25 phosphatases as potential human oncogenes.
Science 1995; 269:1575-1577.
Cell Cycle Molecular Targets and Drug Discovery 245
73. Ducruet AP, Rice RL, Tamura K et al. Identification of new Cdc25 dual specificity phosphatase
inhibitors in a targeted small molecule array Bioorg Med Chem 2000; 8:1451-1466.
74. Graves PR, Yu L, Schwarz JK et al. The Chk1 protein kinase and the Cdc25C regulatory pathways
are targets of the anticancer agent UCN-01. J Biol Chem 2000; 275:5600-5605.
75. Chen P, Luo C, Deng Y et al. The 1.7 A crystal structure of human cell cycle checkpoint kinase
Chk1: implications for Chk1 regulation. Cell 2000; 100:681-92.
76. Monzo M, Rosell R, Felip E et al. A novel anti-apoptosis gene: re-expression of survivin messenger
RNA as a prognosis marker in non-small-cell lung cancers. J Clin Oncol 1999; 17:2100-2104.
77. Bischoff JR, Anderson L, Zhu Y et al. A homologue of drosophila aurora kinase is oncogenic and
amplified in human colorectal cancers. EMBO J 1998;17:3052-65.
78. Zhou H, Kuang J, Zhong L et al. Tumour amplified kinase STK15/BTAK induces centrosome
amplification, aneuploidy and transformation. Nat Genet 1998; 20:189-93.
79. Wolf G, Elez R, Doermer A et al. Prognostic significance of polo-like kinase (PLK) expression in
non-small cell lung cancer. Oncogene 1997; 14:543-549.
80. Sauter ER, Nesbit M, Litwin S et al. Antisense cyclin D1 induces apoptosis and tumor shrinkage
in human squamous carcinomas. Cancer Res 1999; 59:4876-4881.
81. Mukhopadhyay T, Roth JA. Antisense regulation of oncogenes in cancer. Crit Rev Oncol. 1996;
7:151-190.
82. Seymore, L. Novel anti-cancer agents in development: exciting prospects and new challenges. Cancer
Treat Revs 1999; 25:301-312.
83. Sauseville, E. A.; Zaharevitz, D.; Gussio, R et al. Cyclin-dependent kinases: Initial approaches to
exploit a novel therapeutic target. Pharmacol Ther 1999; 82:285-292.
84. Lee JC, Adams JL. Inhibitors of serine/threonine kinases. Curr Opin Biotech 1995; 6:657-661.
85. Meijer L. Chemical inhibitors of cyclin-dependent kinases. Trends Cell Biol 1996; 6:393-397.
86. Meijer L, Kim S-H. Chemical inhibitors of cyclin-dependent kinases. Methods Enzymol 1997;
283:113-128.
87. Walker DH. Small molecule inhibitors of cyclin-dependent kinases: Molecular tools and potential
therapeutics. Curr Top Microbiol Immuno. 1998; 227:149-165.
88. Garrett MD, Fattaey A. CDK Inhibition and Cancer Therapy. Curr Opin Genet Dev 1999; 9:104-111.
89. Toledo LM, Lydon NB, Elbaum D. Structure-based design of ATP-site directed protein kinase
inhibitors. Curr Med Chem 1999; 6:775-805.
90. Gray N, Detivaud L, Doerig C, Meijer L. ATP-site directed inhibitors of cyclin-dependent kinases.
Curr Med Chem 1999; 6:859-875.
91. Buolamwini JK. Cell cycle molecular targets in novel anticancer drug discovery. Curr Pharm Design
2000; 6:379-392.
92. Crews CM, Mohan R. Small molecule inhibitors of the cell cycle. Curr. Opin. Chem. Biol. 2000; 4:47-53.
93. Gray NS, Wodika L, Thunnissen A-MWH et al. Exploiting chemical libraries, structure, and
genomics in the search for kinase inhibitors. Science 1998; 281:533-538.
94. Zaharevitz D, Gussio R, Leost M et al. Discovery and initial characterization of the paullones, a
novel class of small molecule inhibitors of cyclin-dependent kinases. Cancer Res 1999; 59:2566-2569.
95. Schultz C, Link A, Leost M et al. Paullones, a series cyclin-dependent kinase inhibitors: synthesis and
evaluation of CDK1/Cyclin B inhibition, and In Vitro Antitumor Activity. J Med Chem 1999;
42:2909-2919.
96. Hoessel R, Leclerc S, Endicott JA et al. Indirubin, the active constituent of a Chinese antileukemia
medicine, inhibits cyclin-dependent kinases. Nat Cell Biol 1999; 1:60-67
97. Li L Chong, W K M Duvadie, R K et al. Novel ATP-site cyclin-dependent kinase (CDK) inhibitors:
Selective inhibitors of CDK4/cyclin D. 218th ACS National Meeting, New Orleans, Aug. 22-26
1999, MEDI-215.
98. Duvadie RK, Chong WKM, Li L et al. Novel ATP-site cyclin-dependent kinase (CDK) inhibitors:
Selective CDK inhibitors. 218th Am. Chem. Soc. National Meeting, New Orleans, Aug. 22-26
1999; MEDI-214.
99. Vesely J, Havlicek L, Strand M et al. Inhibition of cyclin-dependent kinase by purine Analogues.
Eur J Biochem 1994; 224:771-786.
100. Carlson BA, Dubay, MM, Sausville EA et al. Flavopiridol induces G1 arrest with inhibition of
cyclin-dependent kinase (CDK) 2 and CDK4 in human breast carcinoma cells. Cancer Res 1996;
56:2973-2978.
101. Christain MC, Puda JM, Ho PTC et al. Promising new agents under development by the Division
of Cancer Treatment, Diagnosis, and Centers of the National Cancer Institute. Sem Oncol 1997;
24:219-140.
102. Schwartz GK. Protein kinase C inhibitors as inducers of apoptosis for cancer treatment. Exp Opin
Invest Drugs 1996; 5:1601-1615.
246 Cell Cycle Checkpoints and Cancer
103. Akiyama T, Yoshida T, Tsujita T et al. Phase Accumulation Induced by UCN-01 is associated
with dephosphorylation of Rb and CDK2 proteins as well as induction of CDK inhibitor p21CIP1/
WAF1/Sdi1. Cancer Res 1997; 57:1495-1501.
104. Fan S, Duba DE, O’Connor PM. Cellular effects of olomoucine in human lymphoma cells differing
in p53 function. Chemotherapy 1999; 45:437-445.
105. Legraverend M, Ludwig O, Bisagni E et al. Synthesis and In vitro evaluation of novel 2,6,9-trisubstituted
purines acting as cyclin-dependent kinase inhibitors. Bioorg Med Chem 1999; 7:1281-1293.
106. Buquet-Fagot C, Lallemand F, Montagne, M-N, Mester J. Effects of olomoucine, a selective inhibitor
of cyclin-dependent kinases, on cell cycle progression in human cancer cell lines. Anti-
Cancer Drugs 1997; 8:623-631.
107. Buolamwini, JK. Unpublished results.
108. Iseki H, Ko TC, Xue XY et al. Cyclin-dependent kinase inhibitors block proliferation of human
gastric cancer cells. Surgery 1997; 122:187-194.
109. Leost M, Schultz C, Link A et al. Paullones are potent inhibitors of glycogen synthase kinase-3beta
and cyclin-dependent kinase 5/p25. Eur J Biochem 2000; 267:5983-5994.
110. Lundgren K, Price SM, Escobar J et al. Diaminothiazoles: Potent, Selective Cyclin-dependent Kinase
inhibitors with anti-tumor efficacy. Proc. AACR-NCI-EORTC Int. Conf., Washington, DC., Dec.
16-19-1999, Abst #125.
111. Adams J, Palombella VJ, Sausville EA et al. Proteasome inhibitors: A novel class of potent and
effective antitumor agents. Cancer Res 1999; 59:2615-2622.
112. Song S, Grenfell TZ, GARField S et al. Essential function of the polo box of Cdc5 in subcellular
localization and induction of cytokinetic structures. Mol Cell Biol 2000; 20:286-298.
113. Lee KS, Grenfell TZ Yarm, FR, Erikson RL. Mutation of the polo-box disrupts localization and mitotic
functions of the mammalian polo kinase Plk. Proc Natl Acad Sci USA 1998; 95:9301-9306.

No hay comentarios: