ÎçÒ¹¸£Àû1000¼¯ºÏ

Martin Raff has a theory: all living cells are genetically programmed
to kill themselves; it is only constant support from their neighbours that
keeps them alive. So convinced is Raff about the suicidal tendencies of
cells that he has offered £1000 to anyone who finds a cell capable
of living in isolation.

Nobody has yet come to Raff, who is professor of biology at University
College London, to claim the prize. Indeed, the more biologists study apoptosis
– the cellular death wish – the more examples they find and, paradoxically,
the more profound its implications appear to be for the health and wellbeing
of organisms.

Ten years ago, the hunt began for the genes that carry these vital suicide
instructions (see ‘Life and death in the condemned cell’, New Scientist,
25 January 1992). Now researchers are moving on to the next step: working
out why cells decide to act on these instructions, how genes involved in
cell suicide communicate with one another, and what the links are with diseases
like cancer. Already they are looking to the suicide genes, and the proteins
they encode, for therapies that might attack cancer cells. Apoptosis is
no longer just a buzzword in biology. It is becoming part of the vocabulary
of medicine.

Suicidal tendencies

This would have once come as a surprise to biologists. The notion that
certain cells can commit suicide emerged almost a century ago – a fetal
hand, for example, is sculpted by the death of cells that form webs between
the fingers. But biologists did not suspects that they would see programmed
cell death outside the fetus. It took a group of pathologists from the University
of Edinburgh to prove them wrong.

In 1972, Andrew Wyllie, John Kerr and the late Alastair Currie became
the first to describe cell suicide in the cells of a mature organism. To
name the phenomenon they coined the term ‘apoptosis’, from the Greek word
meaning the shedding of leaves in autumn. At the time their discovery, which
was made in a type of skin cancer called squamous cell carcinoma, was viewed
as an anomaly. Not until another decade had passed did researchers begin
to see apoptosis as ubiquitous, and it was a further five years before its
implication in various diseases – cancer in particular – led to a surge
of interest from molecular biologists and drugs companies.

We now know that cells can die by two pathways. Necrosis is the untimely
death caused by damaging external factors such as poisons, excessive heat
or lack of oxygen. The process is messy: membranes rupture and the contents
of cells spill out uncontrollably. In addition, some of the cell’s chemicals
may have damaging effects on the surrounding tissue.

Apoptosis proceeds in a much more orderly way. A cell triggered to commit
suicide first makes the enzymes it will need to destroy itself. Under the
influence of suicide genes, the cell body then shrinks to a characteristic
spherical shape, while the enzymes within repackage its contents, including
the genetic material, into digestible fragments that neighbouring cells
can consume. Within a few hours the cell starts breaking up. Each fragment
has chemical markers on its surface that attract phagocytic cells, the body’s
scavengers, which quickly remove all trace of it.

It is has become clear in recent years that cells that lose the ability
to commit suicide can start to run out of control, redirecting a destructive
urge into one for survival. Many cancer cells have mutations in one or more
of the genes that control apoptosis. These defects pose a double threat:
as well as preventing death, they can also confer tolerance to chemotherapy
and radiotherapy. The cells most resistant to these treatments survive,
and go on to produce particularly virulent growths. This problem is compounded
by the fact that most cancer therapies induce DNA damage, increasing the
likelihood of mutations in apoptosis genes.

The search is now on for cancer treatments that will pinpoint and then
reverse mutations to the genes controlling apoptosis. To find ways of coercing
cancer cells into killing themselves, researchers are probing the enzymes
and genes at the very heart of apoptosis. It is becoming clear that all
cells are balanced on a knife edge, with death on one side and growth on
the other. Cells react to normal day-to-day damage by activating genes with
one of two opposing aims. The first group, the death genes, put a brake
on cell growth, nudging the cell towards suicide in an attempt to prevent
replication of mutated DNA that might wreak havoc. Their action is balanced
by survivor genes, which aim to keep cells alive so that damage can be repaired.
But how does a cell decide whether damage is irreparable? Where does the
fulcrum lie in normal cells?

The issue is a contentious one. While it is clear that some types of
tissue are more susceptible to apoptosis than others, the level and type
of damage that triggers suicide in different tissues is far from being quantified.
Furthermore, as Raff points out, cell damage is not a prerequisite for apoptosis.
‘Cells are continuously proliferating, yet their numbers remain constant,’
he says. ‘All this normal cell death is caused by apoptosis.’

Few biologists would disagree, but Raff takes the notion one step further.
He argues that the normal state for a cell is death. Raff’s interest in
apoptosis developed in the early 1980s from studies of brain cells and how
they communicate. When he tried to grow individual cells in culture, they
all died within hours, from apoptosis. This response was not confined to
cells from the central nervous system, and he concluded that all cells are
stimulated by certain substances – in particular cytokines and growth factors
– produced by the cells around them. Without this stimulation, apoptosis
is triggered.

Raff’s theory – which he describes as ‘still very unpopular’ – is that
cells compete for limited amounts of cytokines and growth factors in their
environment. The result is that even healthy cells are sacrificed if their
neighbours compete more effectively. Raff recognises that this runs counter
to the accepted idea that cells collaborate for the good of the organism,
but he believes the advantages of this system over random survival of cells
are obvious. ‘People don’t like the idea of cells competing,’ he says, ‘because
that’s what cancer cells do and it’s dangerous.’

Factors of life

Raff’s radical ideas get some support from research by Gerard Evan of
the Imperial Cancer Research Fund in London. Evan has shown that a survival
gene called c-myc, which was identified in the early 1980s, acts as a kind
of switch for apoptosis. If the c-myc gene is turned on, it instructs the
cell to proliferate; if it is not expressed, the cell kills itself. Another
link with Raff’s work is the discovery that normal c-myc cannot function
without the serum in which cells sit. Cytokines within the serum are needed
to bring the gene to life, Evan has found.

In research reported two weeks ago, Evan’s team shows that c-myc activation
is under dual control from cytokines. One group turns on cell growth but
at the same time switches on an order to self-destruct. A second group –
in particular an insulin-like protein called IGF-1 – overrides the self-destruct
instruction. Evan believes that without c-myc cancers would be much more
common than they are. What makes it so difficult for a normal cell to become
cancerous is that death and growth are linked. But in almost all types
of cancer, the c-myc gene appears in a mutated form that cannot be switched
off. Evan now hopes that manipulation of IGF-1 could be one way of attacking
such tumours.

Important though c-myc is as a switch to apoptosis, it does not act
in isolation. Other genes seem to impose checks and restraints on the way
c-myc behaves. One such gene is MTS1, which was reported in Science this
April by Alexander Kamb and his team at Myriad Genetics, a biotechnology
company in Salt Lake City, Utah. The gene acts as a ‘gatekeeper’ for c-myc,
preventing its activation and so promoting apoptosis. MTS1 produces a protein
that blocks the action of an enzyme in the pathway that allows messages
from cell surface receptors to switch on c-myc. So c-myc stays switched
off, and apoptosis can take place. In more than half of all human cancers,
functional copies of MTS1 are missing, leaving c-myc unchecked and the pathway
to cell division open.

The good news is that the MTS1 protein is small, and so might be a suitable
candidate for a drug therapy. ‘In terms of therapeutic potential’, says
Kamb, ‘MTS1 may be the most important tumour-suppressor gene yet discovered.’

Molecule of the year

It may even upstage p53, another gene involved in decisions about cell
growth and suicide. Discovered 15 years ago by David Lane of the University
of Dundee, p53 has come to be known as the ‘guardian of the genome’ because
it swings into action when DNA is damaged. In essence, the gene controls
whether the damaged cell repairs itself or commits suicide. The process
is not a simple one, but it is already clear that when a cell suffers genetic
damage from ionising radiation or other causes, the p53 gene produces a
protein which latches onto other genes to control their activities. So crucial
is this thought to be that the p53 protein was awarded the title ‘1993 molecule
of the year’ by the journal Science.

The process first came to light last November, when teams from Johns
Hopkins University in Baltimore and from Baylor College of Medicine in Houston
reported the discovery of CIP1/WAF1, a growth-arresting gene which is switched
on by p53 protein (see New Scientist, Science, 22 January). The protein
pro-duced by this gene inhibits the activity of a kinase called Cdk2, which
is a key factor in initiating cell division.

The p53 gene has two ways in which it can help to check uncontrolled
growth of damaged cells: by putting the brakes on cell growth and division,
or by opening the way to apoptosis. In cancer cells examined to date, the
p53 gene is the one that is most often mutated. Unable to commit suicide
in response to DNA damage, these cells are free to proliferate out of control.
Such cancers are particularly difficult to treat, since mutant p53 confers
resistance to radiation and chemotherapy. The tumours containing aberrant
forms of the p53 protein are the most aggressive: they are particularly
likely to spread and to cause death quickly.

Many pharmaceuticals companies already have teams working on cancer
therapies that target p53. The most direct approach, gene therapy – introducing
a good copy of the gene into a tumour – is soon to be tested by Jack Roth
and his colleagues at the University of Texas, MD Anderson Cancer Center.
Later this year they hope to begin clinical trials to treat 14 people with
advanced lung cancer. In animal tests they have already achieved an 80 per
cent success rate by delivering the p53 gene inside a retrovirus.

In addition, several drugs companies are funding research into p53 vaccines,
reasoning that fragments of the mutated p53 protein are recognised by the
immune system’s T cells. The plan here is to inject synthesised peptides
which resemble these fragments, in the hope that this will boost the immune
response. Under normal circumstances, high levels of p53 protein appear
only when DNA has been damaged. In cells with mutated p53 genes, protein
levels remain high at all times, acting as a beacon to an enhanced immune
response.

The most radical therapy exploiting p53 has been pioneered by Lane and
his colleagues at Dundee. Their research is based on the observation that
around 70 per cent of p53 mutations result in proteins that differ from
normal p53 protein in just a single amino acid. This causes it to fold incorrectly,
in a form that is inactive. ‘We estimate that about half of those have the
potential to reconfigure or refold into the active form,’ says Lane. He
explains that by introducing drug molecules that bind to inactive forms
of p53, it might be possible to coax these into adopting the shape of the
active form. Now several drugs companies have adopted this approach. ‘It’s
a very bold venture,’ says Lane, ‘but I’m optimistic.’

When they are working properly, both MTS1 and p53 prevent damaged cells
from dividing by steering them in the direction of suicide. A third gene,
bcl-2, has the opposite effect: it keeps cells alive. The danger here is
that cells may become malignant if they have a mutant form of bcl-2 that
is unusually active. How bcl-2 works is only now becoming clear.

Last October, Stanley Korsmeyer, a molecular biologist at Washington
University in Saint Louis, reported that mice devoid of bcl-2 genes unexpectedly
turned grey if they survived to puberty, as though undergoing an accelerated
ageing process. It was already known that most of the proteins produced
by bcl-2 in normal cells are found on the membranes of mitochondria, the
organelles that consume oxygen to form the cellular fuel ATP. These two
clues led Korsmeyer to the idea that bcl-2 might have a role in protecting
cells against free radicals, which are formed as by-products of aerobic
respiration and are also implicated in ageing.

Free radical connection

Support for this idea has come from David Hockenbery, a former member
of Korsmeyer’s team, now at the Fred Hutchinson Cancer Research Center in
Seattle. His team discovered that apoptosis can be blocked almost as effectively
by antioxidants that mop up free radicals as it can be by bcl-2. The researchers
also found signs of lipid peroxidation – a specific type of damage caused
by free radicals – in cells undergoing the early stages of apoptosis. Hockenbery
concludes that cells use free radicals as a suicide weapon. ‘There are two
possibilities,’ he says. ‘Either they act as a signalling mechanism that
might activate downstream genes or enzymes leading to damage, or there are
crucial targets directly damaged by peroxidase stress.’ Bcl-2, he says,
cannot stop free radical formation but it can somehow prevent these molecules
causing devastation.

But a mystery remains: bcl-2 is more effective in some cells than in
others. Epithelial cells – those on the body’s surfaces – tend to have a
higher threshold for apoptosis than cells found deep within the body. In
evolutionary terms this makes sense because such layers need to be tougher
to cope with bombardment by damaging chemicals and radiation. But it creates
a problem for doctors trying to treat cancers. The most common cancers affect
epithelial tissues. Yet they are also the most difficult to treat, because
they are often more resistant to drugs and radiotherapy than the cells around
them.

With a better understanding of the action of bcl-2 there is hope of
overcoming this problem. Research in the past year reveals that the bcl-2
protein is just one member of a family of proteins that interact in apoptosis,
with a variety of consequences. Some combinations have even been shown to
promote cell suicide. Craig Thompson at the University of Chicago’s Howard
Hughes Institute, has discovered that one of these proteins, bcl-x, exists
in two forms. The first, found in long-lived cells such as neurons, reinforces
the effects of the bcl-2 protein. The second form, which is found in short-lived
cells such as immature immune cells, has the opposite effect: it renders
bcl-2 less effective in keeping cells alive. Another protein, bax, binds
to bcl-2 and so suppresses its action, but can also bind to itself to give
a molecule that accelerates apoptosis. More proteins continue to come to
light, and the next step for medical researchers is to manipulate them in
order to target cancer therapies at particular tissue types.

As research into the genetics of apoptosis continues it is revealing
just how sketchy our understanding of the process is. ‘We know virtually
nothing about the machinery by which a cell kills itself,’ Evan warns. ‘It
will be many years before we are able to induce the death programme,’ adds
Raff. But, like most people researching apoptosis, Raff is predominantly
optimistic. ‘All of this will eventually lead to useful drugs,’ he says.
‘It’s as good as any other way of going after cancer.

* * *

Genetic clocks

What makes a cancer cell immortal? A discovery earlier this year shows
that it may be nothing more than an enzyme. Calvin Harley and colleagues
at McMaster University in Hamilton, Ontario, identified telomerase as the
culprit. This enzyme is usually produced only by the cells that make eggs
and sperm. But Harley found that many cancer cells also produce it, allowing
them to escape the natural ageing process.

ÎçÒ¹¸£Àû1000¼¯ºÏy cells are mortal because they contain a genetic clock. At the
end of each chromosome is a chain of DNA made up of ‘beads’ of genetic material
called telomeres. A young cell that has formed in the embryo has over 1000
telomeres on each chromosome tip. But every time it clones itself, between
10 and 20 of them are lost. At first this is no problem, as telomeres do
not code for any proteins, but eventually the chain becomes so short that
the cell can no longer survive. It starts to accumulate damage, and eventually
dies.

Harley discovered that telomerase in malignant cells can repair telomere
chains damaged during cell division. This has led him to speculate that
cancers might be cured if the enzyme could be blocked. Harley argues that
most cancer cells are old; they have divided many times. So if a drug could
be found to block the enzyme, these cells might quickly die of old age.
Because normal cells do not produce telomerase, such a treatment would target
only cancer cells.