 |
Disruption of Growth Arrest and Apoptosis in Cancer
Summary: Wafik El-Deiry's research is aimed at understanding pathways of cell cycle control and cell death in normal and cancer cells. These efforts may lead to novel diagnostic and/or therapeutic strategies for cancer.
We are investigating the molecular basis of human tumorigenesis by
focusing on genetic and biochemical aberrations of cell cycle
checkpoint control and cell death pathways in human cancer cells. We
are studying the mechanism of programmed cell death (apoptosis) by p53
and the role of novel downstream targets we have identified as
determinants of chemo- and radiosensitivity. The most commonly mutated
gene in human cancer, p53 is altered in more than 50 percent of
all cancers. The p53 protein is involved in the cellular response to
stress, including exposure to DNA-damaging agents, hypoxia, nutrient
deprivation, and oncogene activation. Stabilization of p53 leads to
inhibition of cell growth, primarily because of cell cycle arrest or
apoptosis, although there is some evidence that p53 may be involved in
differentiation and senescence. Thus p53-dependent growth inhibition is
a major checkpoint in mammalian cells.
p53-Dependent Apoptosis and Chemosensitivity
The p53 protein is activated by a number of cellular stresses, leading
to cell cycle arrest and apoptosis. Deficient apoptosis is an important
mechanism of tumorigenesis and resistance of cancer to chemo- and
radiotherapy. We are investigating the effects of p53 and various
genotoxic stresses on the cleavage of caspases, enzymes that mediate
apoptosis. We cloned a unique TRAIL (TNF [tumor necrosis
factor]-related apoptosis-inducing ligand) receptor gene called
KILLER/DR5 as an up-regulated transcript when chemosensitive
cells were exposed to cytotoxic doses of adriamycin. To determine how
the KILLER/DR5 gene is regulated by p53- and non-p53-dependent
means, we cloned the human genomic KILLER/DR5 locus. We
identified three candidate high-affinity p53 DNA-binding sites, all of
which appear to bind to p53 in vitro. Binding-site 2, located in the
first intron of the human KILLER/DR5 gene, appears to be
required for p53-dependent regulation of KILLER/DR5 expression.
We have established that p53 regulates the extrinsic pathway of cell
death. With evidence for caspase 8 and 9 activation, and blockade of
p53-dependent death by cFLIP (cellular FLICE-inhibitory protein) and
Bcl-XL, it appears that p53 signals leading to cell death
are transduced by components of both the extrinsic and the intrinsic
cell death pathways.
Two recent developments in our lab are shaping our understanding of
the p53-dependent DNA damage response in vivo. The first is the
recognition that p53 activates its downstream target genes in a highly
regulated, selective, tissue-specific manner that likely correlates
with observed sensitivity to damage. Even within certain tissues there
is compartmentalization of target-gene induction. For example, the
p53–up-regulated mediator of apoptosis (PUMA) is induced strongly
in splenic white pulp, whereas the p53 target BH3-containing proteins
Noxa and Bid are induced in the splenic red pulp. We are interested in
the molecular basis for this regulation.
The second development is the recognition of a subset of
proapoptotic p53 target genes that lower the threshold for death in
response to DNA-damaging therapeutic agents. These genes (including
Apaf1, caspase 6, and Bid) and probably others
lower the threshold for death, providing a molecular mechanism for the
role of p53 in chemo- and radiosensitivity. The model asserts that
transcriptional up-regulation of these genes leads to a buildup of
their encoded proteins; then second signals generated by drug exposure
result in processing and "activation" of the death-inducing proteins,
leading to efficient cell death.
Each of the known "chemosensitivity" genes (Apaf1, caspase
6, and Bid) fits this paradigm. For example, p53-dependent
induction of Apaf1 is not by itself toxic to cells unless a second
signal leads to mitochondrial cytochrome c release, apoptosome
activation, and rapid death through downstream caspase activation.
Pro–caspase 6 protein is not toxic unless it is processed to the
active form, in which case it can cleave its specific substrates,
including the nuclear lamins, whose breakdown leads to the dissolution
of the nuclear envelope during apoptosis. Bid, which is strategically
located and allows "crosstalk" between the extrinsic and intrinsic
death pathways, can be induced by p53. Bid must also, however, be
cleaved to truncated Bid, which then becomes myristolylated prior to
insertion into the mitochondrial membrane, leading to subsequent events
in cell death signaling. The model proposed for the activity of
p53-activated chemosensitivity genes provides a molecular mechanism by
which the presence of wild-type p53 in cells leads to sensitization to
therapeutic agents, such as DNA-damaging chemo- and radiotherapy.
Studies in collaboration with scientists at Pfizer are yielding
insights into the mechanism of action of a unique drug (CP-31398) that
appears to stabilize a wild-type conformation in mutated p53 proteins
that are found commonly in human cancers. CP-31398 appears to act
through a mechanism different from the currently understood mechanism
in cells exposed to DNA-damaging agents. Following DNA damage, the ATM
and Chk2 kinases become activated, phosphorylate p53, and release it
from binding its negative regulator MDM2. Thus p53 is stabilized by
reduced ubiquitin-mediated proteolysis. In cells treated with CP-31398,
however, there is no increase in p53 phosphorylatrion at residues
phosphorylated by ATM or Chk2 (serines 15 or 20, respectively), p53
stabilization occurs in ATM-deficient cells, and p53:MDM2 binding is
not reduced. CP-31398 appears to lead to reduced cellular levels of
ubiquitinated p53, either through inhibition of p53 ubiquitination or
possibly through acceleration of deubiquitination of MDM2-ubiquitinated
p53. These results provide insight into drug development strategies for
tumors with MDM2 overexpression or ARF deletion, at a point of
intervention that occurs after MDM2 binding to p53.
The p53 target-gene promoters we have cloned are facilitating our
efforts to image gene expression in vivo in animals and tumors and to
evaluate the role of various DNA damage–inducible genes in
anticancer chemotherapy and radiation. This effort includes the use of
molecular beacons to image p53 target-gene activation in vivo. Our
studies are focused on the role of p53 target genes in toxicity to
tissue and sensitivity of engrafted tumors during therapy. The goal is
to understand the molecular basis for these responses. The development
of cell-based assays coupled with imaging tools provides a way to
identify small molecules with anticancer activity through modulation of
p53 signaling.
BRCA1-Dependent Regulation of p53 Selectivity and DNA
Repair
Evidence suggests that the familial breast cancer gene product BRCA1
may in part suppress cancers through the regulation of gene expression.
We are using cDNA arrays to identify genes whose expression may be
controlled by BRCA1. Our goal is to explain the coordinated role of
BRCA1 in transcription, DNA damage response, and repair pathways. We
find that BRCA1 can coactivate p53-dependent transcription and have
recently recognized that the effect of BRCA1 on p53 results in
selective up-regulation of growth arrest and DNA repair targets but not
apoptosis-inducing targets. These results provide a system for
elucidating the molecular basis of the selectivity of p53 for
target-gene activation. Thus it is now possible to determine whether
p53 localizes and binds to the genomic loci of proapoptotic target
genes in cells that express high levels of BRCA1 and whether blockade
of BRCA1 modulates such binding and DNA repair responses. Moreover, if
p53 can still bind to its proapoptotic target-gene loci when BRCA1 is
overexpressed, then it becomes possible to investigate whether this p53
is post-translationally modified to no longer activate transcription
from particular promoters or perhaps identify recruited deacetylases or
other p53-interacting proteins that may modulate its selectivity in
target-gene activation. Because it is highly likely that the phenotype
of growth arrest and repair versus cell death depends on the pattern of
p53-induced genes in the normal DNA damage response, it becomes
important to understand the molecular basis for its regulation.
A second area of focus that has emerged from the observation that
BRCA1 can up-regulate DNA repair genes, which can occur in part through
effects on p53, involves questions surrounding the role of
transcription in the BRCA1-dependent repair response. Other important
questions include the role of particular DNA repair genes in breast
cancer development, mammary epithelial and fibroblast cellular
transformation, and the repair response.
This work was supported in part by grants from the National
Institutes of Health.
Last updated June 09, 2004
|
|
|
|
