91av

What happens when a cell makes contact with a surface inside the body?
These sticky moments are much more important than scientists once thought.
But until recently their theories were based on observations of the behaviour
of cells on inorganic surfaces such as glass or plastic. Cells, in their
normal lives, never make such mundane contacts. Every surface they encounter
is tightly packed with proteins. Among these proteins are some known as
‘cell adhesion molecules’, or CAMs, which hold cells together like molecular
Velcro.

CAMs are found on the surface of almost every cell in our bodies. As
well as gluing our bodies together, they are key players in embryonic development
and the immune system. Should these molecules become too sticky or too slippery,
the effect on our bodies can be disastrous. Now companies have begun to
synthesise and manipulate CAMs, hoping to develop new treatments for everything
from cancer and heart disease to autoimmune diseases and viral infections.

Molecular biologists have identified several families of CAMs, and many
more probably remain to be discovered. The diverse roles that CAMs play
at every stage in our life is a clue to their importance. For example, cells
in the developing embryo need to know where they are, and to be able to
move to get to the ‘right spot’. Once they arrive, they need to be able
to anchor themselves. In each case, it is CAMs that do the trick.

In the adult body, too, CAMs have a role – in keeping cells firmly in
their place. For example, the ranks of epithelial cells that face the world
in the skin or the endothelial cells that line the guts and the arteries
are anchored to a layer of proteins, the basement membrane. Interactions
between CAMs on the cells and the basement membranes keep the ranks of stationary
cells in order, through the exchange of chemical signals as well as by adhesion.

Conversely, CAMs have a part to play in the mobility of the cells in
the immune system. All the white blood cells rely on CAMs to bind to, and
then pass through, the basement membranes so that they can leave the bloodstream
to reach injured or infected tissue. Once outside the blood vessel, their
CAMs make contact with connective tissue as they move through body tissues
before returning to the bloodstream.

Exactly what happens as CAMs make contact with a surface can only be
viewed in the artificial environment of the laboratory, and it may not be
the same as what happens deep in the body. But in the past few years, molecular
biology has revolutionised our understanding, and made it possible to unravel
some of the interactions that occur when cells touch.

ON THE SURFACE

In some ways, it seems that CAMs resemble hormones in the way they interact
with receptors on a surface. As in hormone-receptor interaction, only specific
peptides (fragments of proteins made up of sequences of several amino acids)
can interact. And the interaction at the surface seems to trigger the same
biochemical cascade: a flood of calcium ions into the cell and the phosphorylation
of molecules inside the cell. There are, however, big differences.

The CAM binds less tightly than a hormone does to a receptor, and the
CAM receptors are far more numerous than hormone receptors are. A cell may
have up to 500 000 CAMs on its surface, with perhaps 100 000 of them simultaneously
sticking to other membranes. But, then, the CAMs’ interactions must provide
a strong enough grip for a cell to be able to pull itself along and to anchor
itself once it has reached its destination. Viewed in the laboratory, a
cell settling on a surface makes several points of contact. At the leading
edge, it puts out clusters of CAMs, rather like tiny ‘feet’, which attach
themselves to the surface to anchor the cell. If it is on the move, the
cell disengages its feet at the rear. The cell then pulls itself forward
because the CAMs are linked to the long fibres of actin protein within the
cell that control the shape and movement of the cell.

SHORT BUT STICKY

One of best understood of the adhesive molecules is fibronectin. Martin
Humphries of the Department of Biochemistry at the University of Manchester
has found two tripeptides (each consisting of three amino acids) within
fibronectin that stick to corresponding peptides on the CAMs on cell surfaces.
The first sticky peptide, originally identified by Humphries and Erkki Ruoslahti
in the mid-1980s, is made up of arginine, glycine and aspartate and is known
as RGD. The second, identified last year, consists of leucine, aspartate
and valine and is known as LDV.

The proteins that RGD and LDV stick to on cell surfaces are members
of the integrin family of CAMs. But how can such short sequences of amino
acids encode specific signals such as ‘Stick here and don’t move’? The answer,
it now appears, is that it is not just the sequence of amino acids but the
exact arrangement of RGD and LDV (and maybe other peptides still to be discovered)
on the cell’s surface that provides the information. Humphries and his Manchester
colleague Andy Brass are working out the three-dimensional shapes of the
interacting molecules using computer graphics as well as molecular genetics.

‘We are now at the point where we know enough about RGD and LDV, and
about the integrin receptors involved in adhesion, to put the two sets
of molecules together and understand their interaction at the molecular
level,’ says Humphries. He is collaborating extensively with teams in Britain
and the US. Already several institutions and pharmaceuticals companies hold
patents in this area and hope to develop anticancer drugs or anti-inflammatory
drugs to treat rheumatoid arthritis.

CAMs are involved in cancer in more than one way. Some malignant cells
have lost the ability to respond to signals and fail to stick on the appropriate
membranes so that they end up where they do not belong. In Walter Bodmer’s
laboratory at the Imperial Cancer Research Fund, London, researchers have
recently pinpointed a gene which regulates cell adhesion, the CAR gene.
This gene makes malignant cells behave normally and attach to membranes
again. The CAR gene is not a gene for CAMs but for a regulatory molecule
that apparently encodes for the proteins that control adhesion. One day
it might be possible to treat cancer by somehow targeting a CAR gene into
cancer cells.

STOPPING THE SPREAD

Another approach could be to try to block the spread of cancer. Some
malignant cells remain responsive to RGD or LDV-type peptides, and the aim
is to ‘snare’ them as they wander around the bloodstream. For, once settled,
these malignant cells multiply and grow into tumours. Researchers are now
working to produce peptides that would mimic RGD and LDV but which would
be more stable so that they could persist in the bloodstream. Cancer cells
encountering such peptides would stick to them, ‘thinking’ they had found
a membrane to settle on. They would then remain in the circulation long
enough for the immune system to identify and destroy them.

A similar strategy could lead to new anti-arthritic drugs. In acute
inflammatory arthritis, the T cells of the immune system migrate from the
bloodstream to an inflamed joint, causing further swelling, pain and damage
to cartilage. Stable mimics of RGD, LDV and perhaps other tripeptides could
bind to T cells to make them ‘believe’ they had already crossed the membranes
to reach the site of inflammation.

Real hope of an effective treatment for multiple sclerosis rests on
yet another application of the same idea. T cells are normally excluded
from the central nervous system by the blood-brain barrier. In MS, the T
cells are able to attach to and pass through the membranes that form the
barrier, and then attack the myelin sheaths of neurons. Experimental autoimmune
encephalitis in mice, commonly used as a model of multiple sclerosis, has
been successfully treated using antibodies that bind to CAMs on T cells,
and so prevent the T cells from being able to pass through the blood-brain
barrier. Athena Neurosciences, a company working closely with scientists
at the University of California at Berkeley, is now engineering its mouse
antibodies so that they can be used to treat human patients without being
rejected. They hope to start clinical tests next year.

A team led by Mark Edwards of the Oxford company British Biotechnology
is specially interested in preventing damage to heart muscle after coronary
heart attacks. Mimics of the all-important RGD and LDV might be used to
latch onto the T cells that damage heart muscle after such attacks. Such
treatment could be given immediately after a coronary attack to limit the
damage.

At Guy’s Hospital in London, Robert Poston has found an excess of CAMs
in the walls of coronary arteries affected by atherosclerosis. The CAMs
attract white cells from the blood to settle on the arterial walls, where
they fill up with fat and contribute to the narrowing of the arteries. The
next step is to discover what causes the excess of CAMs, and then to try
to develop RGD and LDV mimics that block the settling process.

RGD, LDV and their opposite numbers, the integrin receptors, are not
the whole sticky story. Adhesion plays a specially important role in development,
especially in the nervous system. At Guy’s Hospital in London, Paul Doherty
is investigating this. His long-term aim is to find ways of promoting the
growth of axons, the long thin extensions of nerve cells, so as to repair
damage to the brain or spinal tissue. He has found that the growing tips
of axons carry neural cell adhesion molecules (NCAMs) which bind to CAMs
on other cells and that the binding stimulates axonal growth. But this happens
only during development of the central nervous system in the embryo, when
axons are growing towards target cells.

CHANGING WITH AGE

Doherty has shown that older neurons, which are no longer able to grow
new axons, still have NCAMs on the tips of their axons. But they are NCAMs
with two differences. The molecules have long extra side chains of the oligosaccharide
sialic acid, and an extra sequence of polypeptides. These two changes appear
to alter the functions of the NCAMs. In young neurons, NCAMS stimulate neuronal
growth, channelling it along paths signposted by other CAMs on other cells.
In mature, sedate neurons, the altered NCAMS prevent new growth but hold
existing circuitry together, thus ensuring the stability required for long-term
memory among other things.

Small chemical changes, then, can alter the function of NCAMs, causing
them to act as growth stimulants or as glue that ensures stability. During
the development of the central nervous system, NCAMs stimulate biochemical
pathways in the tips of the axons that cause an influx of calcium, which
in turn stimulates axon growth. If it becomes possible to bypass the stabilising
influence of NCAMs in older neurons, the way will be open to stimulate these
pathways directly so as to make nervous tissue regrow. A similar approach
might stimulate peripheral nerves, which do regenerate in adult life, making
them grow more rapidly and so speed the regrowth of severed nerves after
accidents. Several companies in the US are working in this area, with tests
on humans possible in three years’ time.

Many of the avenues for applications for CAMs that are now being explored
may turn out to be blind alleys, but sticky moments are so important to
the lives of cells that studying them seems bound to have a big impact one
way or another.

John Newell is editor of Science, Industry and Medicine for the BBC’s
World Service, and author of Playing God? (BBC/Broadside Books, 1991) which
looks at the implications of genetic engineering.