|
The
Myc Oncogene and Cancer 
Our laboratory is interested in the molecular processes underlying carcinogenesis.
For some years, our principal focus has been the myc oncogenes, a family
of genes deregulated in the majority of human cancers. One member of the
myc gene family in particular, c-myc, is expressed in all normal proliferating
cells, and its experimental deregulation is sufficient to disrupt normal
growth control of many cell types. c-myc is therefore an oncogene because
it makes cells proliferate. Surprisingly, however, c-myc possesses another
unexpected attribute which greatly limits its carcinogenic potential --
it is also a potent activator of apoptosis (programmed cell death).
Thus, c-Myc drives two completely contradictory biological processes within
the cell: proliferation leading to cell gain and apoptosis leading to
cell loss. In fact, this contradictory duality of action is not restricted
to myc genes but is shared by all genetic lesions that drive cell proliferation.
From this we have argued that it is this obligatory coupling of cell proliferation
and cell death that is a fundamental mechanism which forestalls the inappropriate
and uncontrolled propagation of cells within tissues (Harrington et al.,
1994). When the proliferative machinery of the cells is engaged, so too
is a default suicide program which is lethal unless blocked by survival
signals. Because these necessary survival signals originate from neighbouring
cells, either as soluble factors or through specific cell-cell contacts,
propagation of any cell requires the connivance of both the cell concerned
and its local somatic environment. In effect, the autonomy of individual
somatic cells is restricted, thereby suppressing emergence of neoplastic
clones: surveillance against neoplastic transformation has been hardwired
into the proliferative machinery of every somatic cell (Evan and Littlewood,
1998).
Three key questions arise from these observations: First, how does the
c-myc protein, c-Myc, activate the cell suicide machinery at the molecular
level? Second, how is c-Myc-induced apoptosis suppressed in normal cells,
which proliferate yet stay alive? Third, what is the relevance of c-Myc-induced
apoptosis to the genesis and development of cancers in real tissues?
The Molecular Mechanism of Myc-Induced Apoptosis
Our earlier data indicated that expression of c-Myc renders cells of many
types sensitive to induction of cell death by a wide range of insults
and agents: for example, DNA damage, growth arrest, TNF/Fas signalling,
survival factor withdrawal, interferon, hypoxia and physical trauma (Evan
et al., 1992; Harrington et al., 1994; Hueber et al., 1997). This suggests
that c-Myc acts at some nodal point of intersection for these mechanistically
diverse insults -- presumably at the hub of cell death regulation.
Over the past few years, it has become clear that cytochrome C is an important
intracellular signal for cell death (Liu et al., 1996). Cytochrome C,
in its holo heme-bound form, is sequestered between the inner and outer
mitochrondrial membranes, where it fulfils its well-documented role in
electron shuttling. Recently it was shown that in response to many pro-apoptotic
triggers, holocytochrome C is rapidly released into the cytosol, where
it orchestrates assembly and activation of an "apoptosome" complex comprising
the adaptor protein Apaf-1 and the apical (initiator) Caspase 9 (Kharbanda
et al., 1997; Kluck et al., 1997; Li et al., 1997; Liu et al., 1996; Zou
et al., 1997). Caspases are the cysteine alanyl proteases that, acting
in proteolytic cascades, execute the process of apoptosis through cleavage
of critical cellular substrates (Thornberry and Lazebnik, 1998).
The "central" role of cytochrome C in apoptosis prompted us to ask whether
cytochrome C might be an apoptotic effector of c-Myc. We have shown that
c-Myc exerts its principal pro-apoptotic action through triggering release
of cytochrome C and that survival signals that suppress c-Myc-induced
apoptosis do so by inhibiting release of cytochrome C from mitochondria
(Juin et al., 1999). Perhaps most intriguingly, however, we have demonstrated
that cytosolic cytochrome C serves to prime cells to the same diverse
range of pro-apoptotic insults as c-Myc. Thus, artificially introducing
holocytochrome C into the cytosol by microinjection sensitises cells to
killing by Fas, TNF, DNA damage, hypoxia, and so on. C-Myc-induced apoptosis
is therefore a two-step process whereby the deregulated oncogene renders
the cell acutely sensitive, but does not commit the cell to apoptosis
(Juin et al., 1999) (Evan and Littlewood, 1998). An immediate conclusion
of this work is that two different strategies can be utilised by tumour
cells to evade apoptosis: one is to block release of cytochrome C from
mitochondria whilst the other is to uncouple released cytochrome C from
the downstream apoptotic machinery. Identifying which of these two mechanisms
has been acquired in specific tumour cells should aid development of strategies
to reinstate the cell suicide pathways that these malignant cells have
lost.
Signalling Pathways that Regulate Apoptosis
Evasion of apoptosis, at least to some degree, appears to be mandatory
in tumour cells. Identifying how tumours thwart cell death is, therefore,
a logical route forward for identifying new therapeutic targets for cancer
treatment.
We have therefore investigated the mechanism by which c-Myc triggers apoptosis
in cells as well as the mechanisms by which survival factors block it.
Recently, we made the unexpected discovery that c-Myc-induced apoptosis
in fibroblasts (our most typical model system) requires the interaction
on the cell surface between the Fas (Fas/Apo-1) receptor molecule and
its ligand FasL (Hueber et al., 1997). Fas is a member of a transmembrane
receptor family that directly activates the cell death machinery by recruiting
caspases the proteolytic effectors of the apoptotic programme (Boldin
et al., 1996; Muzio et al., 1996).
We have further mapped c-Myc action downstream of CD95 receptor activation:
thus, it appears that c-Myc acts to turn up the cell's sensitivity to
a pre-existing autocrine suicide signal that is always present in cells.
We now know that the sensitization afforded to Fas by c-Myc expression
is due to release of holocytochrome C into the cytosol (see above).
We have also extensively studied the signal transduction pathway by which
IGF-I suppresses apoptosis. They have shown that a survival signal travels
from the IGF-I receptor via Ras through to PI-3 kinase and the serine-threonine
kinase PKB/AKT (Kauffmann-Zeh et al., 1997). Paradoxically, however, Ras
is implicated not only in suppressing apoptosis but also in promoting
it. Andrea showed that this is because Ras does indeed exert these two
contradictory effects on cell viability. The choice of which cell fate
prevails -- death or proliferation -- depends upon the way in which downstream
Ras effector pathways intersect with other signalling pathways within
the cell. Importantly, in the absence of any countermanding signal the
net effect of Ras, like Myc, is to promote cell death. From this has emerged
our so-called "booby trap" model: we believe that inappropriate activation
of key regulators of cell proliferation, as happens following carcinogenic
mutations, triggers deletion of the affected cell, so suppressing the
incidence of cancer in somatic tissues (Evan and Littlewood, 1998).
To address the question of how c-Myc and survival factors interact with
the basal apoptotic machinery, we have been using the genetically tractable
fruit fly Drosophila melanogaster as a system with which to dissect out
the relationship between basal apoptotic machinery and the regulatory
pathways that control it. Some time ago, we identified a key effector
caspase, called drICE, that is required for apoptosis in fly cells (Fraser
and Evan, 1997; Fraser et al., 1997). Using a yeast two-hybrid assay to
identify candidate proteins that interact with drICE, we have now identified
a new Drosophila caspase called CARCASS/DRONC that associates with drICE.
CARCASS/DRONC has the extensive pro-domain characteristic of the "apical"
caspase - caspases that are activated directly by upstream regulatory
signals and, in so doing, cleave and initiate the downstream effector
caspase cascades that implement apoptosis.
The pro-domain of CARCASS/DRONC contains a CARD motif (Hofmann et al.,
1997) known to mediate a variety of interactions with caspase affector
and effector proteins. Intriguingly, although expression of full length
CARCASS/DRONC in yeast cells causes cell death through auto-activation
of the caspase, expression in fly or mammalian cells has no lethal effect.
In contrast, expression of a truncated CARCASS/DRONC lacking the CARD
pro-domain is lethal in animal cells. This suggests that both fly and
mammalian cells harbour an innate CARCASS/DRONC inhibitor which acts to
suppress CARCASS/DRONC activity through the caspase's CARD domain. In
fact a subsequent search for likely molecules that might act through the
CARCASS/DRONC CARD domain identified DIAP1, a cellular homologue of the
IAP protein originally identified as an inhibitor of apoptosis encoded
by the insect baculovirus (Clem and Miller, 1994; Hay et al., 1995). IAPs
have been conserved throughout metazoan evolution although their functions
remain unclear. Our data now suggest that the IAP proteins exert at least
a major part of their anti-apoptotic action through interaction with,
and inhibition of activation of, CARD domain caspases. Several of the
key apical caspases that initiate apoptosis in human cells possess CARD
domains, so it seems likely that similar interactions are involved in
regulating cell viability in man
.Role of c-Myc-Induced Apoptosis in Cancer in vivo
Our studies indicate that the availability of survival signals is a principal
determinant of the survival of proliferating and neoplastic cells. One
can therefore imagine two types of tissue. In the first, survival factors
are abundant and c-Myc-induced apoptosis is effectively suppressed. In
such a tissue, suppression of apoptosis would not be important in the
early genesis of cancers because it is already blocked. In contrast, other
types of tissue may, for reasons reflecting their function and architecture,
have limiting supplies of survival factors. In such a tissue, activation
of Myc would immediately trigger apoptosis in the affected cell and cancers
could never arise without early mutations that block cell death. Correcting
the anti-apoptotic lesion in such cancers would be an excellent way of
treating the disease. Where do each of these conditions apply in the body?
Put another way, in which tissue does Myc trigger proliferation and in
which cell death?
By definition, the complex web of survival signalling active within complex,
heterogeneous intact tissues cannot be replicated in tissue culture. Therefore,
to explore the relationship between c-Myc activation and cell death in
intact tissues we have developed a unique system in which a switchable
c-Myc protein is targeted to various mouse tissues. This switchable system
also has the advantage that any pathologies emerging from c-Myc activation
-- neoplasia or tissue degeneration -- are reversible. The switchable
c-Myc protein (MycERTM) we have designed is continuously dependent upon
the synthetic steroid 4-hydroxytamoxifen (4HT) for its activity.
Stella Pelengaris and Trevor Littlewood have generated transgenic mice
in which MycERTM has been targeted to suprabasal skin keratinocytes and
to the beta cells of pancreatic islets of Langerhans. Activation of c-Myc
in suprabasal skin leads to rapid onset of proliferation without any apparent
apoptosis, resulting in dramatic hyperplasia and, eventually, full-blown
papillomas showing arrested keratinocyte differentiation, areas of focal
dysplasia, and extensive neo-angiogenesis. Furthermore, withdrawal of
4HT inactivates c-Myc and within three weeks the papillomas, complete
with all their attendant neoplastic paraphernalia, completely regress.
For the first time this has allowed us to glimpse the kinetics and the
complexity of the neoplastic phenotype that can be generated through activation
of only a single oncogene as it both progresses and regresses. In addition,
our data indicate that the lack of apoptosis induced by Myc in skin is
not because Myc is unable to activate the death programme of skin cells
but because skin exemplifies a tissue with abundant survival signals in
which apoptosis is actively suppressed. Presumably skin, a shedding epithelium,
deals with potentially malignant cells by the simple expedient of discarding
them a strategy that enables preservation of the tissue's structural
integrity but which is, self-evidently, not applicable to non-epithelial
tissues.
Activation of c-Myc in pancreatic beta cells also induces dramatic cell
proliferation but, in contrast to skin, this is accompanied by massive
apoptosis. Clearly, in pancreas there is insufficient survival signalling
to support ectopically proliferating cells. The net consequence is rapid
involution of islet cells as cell death overwhelms cell gain: mice become
acutely diabetic within a week as their insulin-secreting cells die. As
predicted, co-expression of the anti-apoptotic oncogene bcl-2 in beta
cells forestalls Myc-induced apoptosis and permits rapid hyperplastic
expansion. This indicates that pancreatic islets are an example of a tissue
in which neoplasia could never gain a foothold without the very early
acquisition of a mutation that blocks apoptosis.
In T cells, the situation is different again. Bettina Rudolph has targeted
MycERTM to the T cell lineage via the lck promoter and shown that activation
of Myc has only a limited effect on these cells and their development.
It appears that T cells, a lineage at high risk of mutation because of
innate recombinases, are protected from the effects of a single oncogenic
lesion like c-Myc activation in a way not seen in skin, fibroblasts, or
beta cells.
It remains to be seen whether the different behaviour we have observed
in these three tissues can be extrapolated into general rules for differing
types of tissue such as liver and gut. We plan to address this fundamental
question in the future.
Selected Publications:
Evan, G., and Littlewood, T. (1998). A matter of life and cell death.
Science 281, 1317-22.
Fraser, A. G., James, C., Evan, G. I., and Hengartner, M. O. (1999). Caenorhabditi
elegans inhibitor of apoptosis protein (IAP) homologue BIR-1plays a conserved
role in cytokinesis. Curr Biol 9, 292-301. Hueber, A. O., and Evan, G.
I. (1998). Traps to catch unwary oncogenes. Trends Genet 14, 364-7.
Klefstrsem, J., Kovanen, P. E., Somersalo, K., Hueber, A. O., Littlewood,
T., Evan, G. I., Greenberg, A. H., Saksela, E., Timonen, T., and Alitalo,
K . (1999). c-Myc and E1A induced cellular sensitivity to activated NK
cells involves cytotoxic granules as death effectors. Oncogene 18, 2181-8.
McCarthy, N., and Evan, G. (1998). Methods for detecting and quantifying
apoptosis. Curr Top Dev Biol 36, 259-278.
Meier, P., and Evan, G. (1998). Dying like flies. Cell 95, 295-8. Rohn,
J. L., Hueber, A. O., McCarthy, N. J., Lyon, D., Navarro, P., Burgering,
B. M., and Evan, G. I. (1998). The opposing roles of the Akt and c-Myc
signalling pathways in survival from CD95-mediated apoptosis. Oncogene
17, 2811-8.
Shidrawi, R. G., Parnell, N. D., Ciclitira, P. J., Travers, P., Evan,
G., and Rosen-Bronson, S. (1998). Binding of gluten-derived peptides to
the HLA-DQ2 (alpha1*0501, beta1*0201) molecule, assessed in a cellular
assay. Clin Exp Immunol 111, 158-65.
Ulrich, E, Duwel, A., Kauffmann-Zeh, A., Gilbert, C., Lyon, D., Rudkin,
B., Evan, G., and Martin-Zanca, D. (1998). Specific TrkA survival signals
interfere with different apoptotic pathways. Oncogene 16, 825-832.
Vecino, E., Caminos, E., Becker, E., Rudkin, B. B., Evan, G. I., and Martin-Zanca,
D. (1998). Increased levels of TrkA in the regenerating retinal ganglion
cells of fish. Neuroreport 9, 3409-13.
Zs˙rnig, M., Hueber, A.-O., and Evan, G. (1998). p53-dependent impairment
of T-cell proliferation in FADD dominant-negative transgenic mice. Curr
Biol 8, 467-470.
Pelengaris, S., Littlewood, T., Khan, M., Elia, G., and Evan, G. (1999).
Reversible activation of c-Myc in skin: induction of a complex neoplastic
phenotype by a single oncogenic lesion. Mol Cell 3, 565-77.
Fraser, A. G., James, C., Evan, G. I., and Hengartner, M. O. (1999). Caenorhabditis
elegans inhibitor of apoptosis protein (IAP) homologue BIR-1 plays a conserved
role in cytokinesis. Curr Biol 9, 292-301.
Juin, P., Hueber, A. O., Littlewood, T., and Evan, G. (1999). c-Myc-induced
sensitization to apoptosis is mediated through cytochrome C release. Genes
Dev 13, 1367-81.
Contact Information:
Email: gevan@cc.ucsf.edu
Phone: 415/ 514-0438
Address: Box 0875, Room S 233
The University of California, San Francisco, CA 94143, (415) 476-9000
Copyright 2003, The Regents of the University of California.
|
