91av

Picking the winner of a competition for the most useful biochemical
molecule is not easy. Five years ago, there was no doubt about it: proteins
and nucleic acids shared the prize. Proteins were the messengers and the
machine tools of the living cell, and nucleic acids formed the genes – the
blueprints for making proteins. But there is a third important class of
large, complex molecules that until recently went largely unnoticed. Now
some new discoveries about these molecules are forcing biochemists to revise
their opinions and biotechnologists to rewrite their rules.

The molecules in question are oligosaccharides – complex carbohydrates
mde up of chains of simple sugar units similar to glucose, called monosaccharides.
Oligosaccharides are produced by successive reactions between the hydroxyl
group of one monosaaccharide and the reducing group of another, with the
loss of water molecules. The bonds formed in this way are called glycosydic
bonds.

Oligosaccharides are often attached, in varying quantities, to other
molecules: proteins or lipids. In nature, oligosaccharides are added to
proteins in a process called glycosylation. The chains branch outwards,
in appearance rather like the flares of the solar corona around the Sun,
from the surfaces of many functional proteins – molecules such as hormones,
enzymes and receptors. Biochemists have known their basic structures for
several decades. But they regarded oligosaccharides, and indeed other carbohydrates,
as rather inert ‘filler’ molecules.

Then, about five years ago, Raymond Dwek, a biochemist working at the
University of Oxford, made some intriguing discoveries about the oligosacchrides
on one class of functional proteins, antibodies (see This Week, 1 August
1985). As a result, biochemists began to realise that the oligosaccharides
on proteins, as well as proteins themselves, depend on subtle differences
of structure to define their particular job in the body. This has immediate
implications for the biotechnology industry. In future biotechnologists
will have to define and control the sugar component, as well as the protein
component of therapeutic glycoproteins such as insulin: otherwise they may
not be able to predict for certain how these molecules will behave in the
body.

Dwek and his team have been studying oligosaccharides for 10 years.
They were the first to discover that oligosaccharides make up parts of the
signals by which proteins recognise both each other and receptor sites on
cell membranes. During the past decade, they have found out that the sugar
chains making up oligosaccharides are somehow deeply involved in reactions
between enzymes and substrates, between messenger molecules and their receptors,
and between antibodies and antigens. It seems that the way oligosaccharides
contribute to the shape of the reactive site in such molecules is often
more important than the precise sequence of the individual sugar molecules.
According to Dwek, oligosaccharides are ‘nature’s way of diversifying proteins’.
He even prefers to define life, not in terms of the accepted sequence –
DNA to RNA to protein – but as the variety made possibly by the diversities
of oligosaccharides on glycoproteins.

In 1985, Dwek discovered that a single antibody to a single antigen
might be one of over 900 types of glycoprotein, differing only in the type
of oligosaccharide on their surface. The challenge was to relate these differences
in structure to differences in their function. So Dwek looked at antibodies
in people affected by rheumatoid arthritis, where chemical differences from
the norm were known to exist. He found that the oligosaccharide chains on
some immunoglobulins – basic antibody molecules – were shorter than normal.
People with rheumatoid arthritis consistently had one sugar residue – galactose
– missing. Dwek and his team, with the American pharmaceuticals company
Monsanto, are now developing their discovery into a clinical test for rheumatoid
arthritis, which will be tried out in hospitals before the end of 1990.
If the trials are a success, it will be on sale as a diagnostic kit within
the next two years.

In the longer term, researchers, may be able to direct white blood cells
– lymphocytes – to produce antibodies with the correct sugar chains. For
example, they could insert genes for the appropriate enzymes into the stem
cells which produce the lymphocytes. Eliminating ‘foreign’ oligosaccharides
would switch off the auto-immune reaction these trigger. Some researchers
in pharmaceutical companies think such an auto-immune reaction leads to
joint damage and to the symptoms of rheumatoid arthritis.

Since 1983, Monsanto has provided 90 per cent of the funding for Dwek’s
laboratory – over 1 million Pounds each year – with Oxford University contributing
most of the other 10 per cent. But Monsanto’s main interest is in new and
improved medical drugs based on glycoproteins that have modified oligosaccharides.
It sub-licences patents it takes out on some of Dwek’s discoveries to another
company, Oxford Glycosystems, whose shareholders include Oxford University
(9 per cent), Dwek (9 per cent) and Monsanto (6 per cent).

Oxford Glycosystems’ main interest is to commercialise the techniques
being developed in Dwek’s laboratory to speed up and automate the sequencing
of oligosaccharides. This is many years behind the smoothly-automated technology
now used to sequence proteins and DNA. The teams now racing to catch up
also face far tougher problems than do sequences of proteins of DNA. One
is that oligosaccharide chains frequently branch. Conventional sequencing
machines, cannot cope with this because they lack the intelligence to do
more than work along a single, unbranched chain. Another is that successive
monosaccharides along a chain can be linked to each other not just in one
way, like bases or amino acids, but in 256 different ways. For example,
there may be more than one chain of sugars attached at one place to a single
protein molecule. Different chains may even be found on what have until
now been considered essentially identical protein molecules. Any one of
31 different monosaccharides may be found in a single oligosaccharide chain.

Today, Oxford Glycosystems is producing sugar chains to order for other
researchers to work on, and sequencing oligosaccharides sent for analysis
by other research groups. In 18 to 20 months it hopes to sell a machine
for analysing oligosaccharides, using selective enzymes to dissect the chains,
one monosaccharide at a time, with the aim of providing sequences and structures
in a completely automated way within hours.

In 1985, Steve Homans, one of Dwek’s group at Oxford, left to found
another laboratory in the biochemistry department at the University of Dundee
which specialises in the sequencing of oligosaccharides. Research at Dundee
is co-funded in a LINK programme by the Science and Engineering Research
Council, the Department of Trade and Industry, and four British-based comanies
– Glaxo, the Wellcome Foundation, Celltech and ICI – who provide half the
funding between them. Homans says that ‘the sequencing of oligosaccharides
presents such massive problems that it has distorted the thrust of research,
with scientists perhaps understandably often tending to avoid increasingly
important areas in which such sequencing appeared unavoidable, because of
lack of the equipment or the time and effort needed to undertake it.’ Homan’s
laboratory, which will offer further opportunities for industrial collaborations,
is putting much effort into developing automated sequencing techniques.
An important method is high resolution nuclear magnetic resonance, which
can identify an unambiguous sequence for an oligosaccharide chain detached
from its parent protein. Others are fast atom bombardment mass spectrometry
and gas chromatography mass spectrometry.

Since 1988 another group of researchers at Dundee, led by Mike Ferguson,
has been investigating ways of manipulating oligosaccharides with chemicals.
They hope to use the resulting compounds as medical drugs to attack parasites
or to make them vulnerable to the immune systems of their hosts. In this
way they hope to find new therapies for tropical parasitic diseases. Ferguson
is currently working with Steve Homans towards a drug to treat African sleeping
sickness. The parasite responsible for this disease, Trypanosomas brucei,
lives in the blood of its host, evading the host’s immune response by repeatedly
changing the antigens that form its outer coat. These antigens are called
variable surface glycoproteins (VSGs). They are so closely packed that ‘complement’
– the group of proteins that triggers the first steps in non-specific immune
response directed against any foreign organism – cannot penetrate the parasite’s
coat. Even if another arm of the immune system, the antigen-specific immune
response, succeeds in identifying, attacking and killing nearly all the
parasites in the bloodstream of an individual, a few parasites will escape
because they have just changed their coats and made different glycoproteins.
So the host’s antibodies have no effect on them. These survivors build up
a new population, which in turn will be mostly eliminated. But once again
a few will survive, and so on, so that the disease is cyclical.

‘The clink in the armour of this parasite,’ says Mike Ferguson, ‘could
be the fact that all the VSGs are attached to the membrane around the parasite
by a single glycolipid (an oligosaccharide with a lipid attached). The glycolipid
is a natural target for drugs designed to attack T brucei.’ When Ferguson
and his colleagues sequenced the oligosaccharides of the glycolopid they
discovered that some sequences of sugars are present in the parasite, but
not in its human host. They are now trying to work out how the parasites
make these specific glycolopids, so that they can devise enzyme inhibitors
to block the preocess without harming the human host.

One way to do this would be to distort the glycolipid that links the
VSGs and the parasite membrane. This would loosen the coat and allow complement
to penetrate to the membrane. There, it could trigger the non-specific immune
reaction. The researchers hope that this will prevent or delay the changeover
to a new coat long enough to allow antigen-specific immunity to destroy
all the parasites.

Ferguson’s team is also working on new strategies to attack other trypanosome
parasites. They are particularly interested in those that cause two diseases:
Leishamiasis, which is transmitted by sand flies in the Middle East and
Indian sub-continent, and Chagas’ disease in Central and Latin America,
which is transmitted by bugs and which causes fevers and sometimes death.
These parasites use the ability of oligosaccharides on proteins to recognise
receptors and through them enter macrophages, part of the human body’s immune
system. Here the parasites hide, rather audaciously, to evade detection
by that same immune system. When Ferguson and his team altered the surface
oligosaccharides of some trypanosomes they found that these parasites could
no longer enter the macrophages. Ferguson’s group is now designing inhibitors
of the enzymes glycosyl transferase which plays an essential role in producing
the oligosaccharides.

The design of such enzyme inhibitors is also one aim of the Link programme
at Dundee. There the inhibitors are being developed for another purpose,
however: to control the nature of the sugars added to proteins made in laboratory
cell cultures. There is growing concern about the role of oligosaccharides
in the ‘cloning’ of glycoproteins – the process of making proteins by inserting
human genes into laboratory cultures. These cultures are usually of a bacteria
such as E coli because they can be grown more rapidly and cheaply than animal
cells. But bacteria do not add oligosaccharides to proteins (a process called
glycosylation). Yeast cell cultures, which are also used for cloning proteins,
do, but these oligosaccharides are not necessarily the same as those that
would be added by human cells making the same protein.

Oligosaccharides are also being used in developing treatment for AIDS.
Particles of HIV 1 and 2 are covered in a cloud of oligosacchardies; half
the molecular weight of the glycoprotein coat of the AIDS virus is carbohydrate.
Somehow these oligosaccharides are involved in the virus’s ability to center
T-cells by reacting with the CD4 antigen on the cell surface. The enzymes
responsible for producing these oligosaccharides are cellular, not viral
enzymes. Irene Schultz of St Louis University and other researchers have
shown that analogues of animo sugars – compounds which mimic the ones cellular
enzymes normally use as building blocks – can inhibit the virus’s ability
to spread from cell to cell. Monsanto is now testing amino sugar analogues
on volunteers in the US who have AIDS. The results of earlier laboratory
tests were encouraging, but it remains to be seen whether there will be
any harmful side effects and if so whether these will be severe enough to
prevent the use of analogues as drugs to treat people with advanced AIDS.
if the treatment is successful similar techniques might be used to attack
other viruses where oligosaccharides are important in the mechanism of infection.

Laboratories are now in the thick of the battle to automate the sequencing
of oligosaccharides. Genetic engineers are beginning to grapple with the
problems of controlling the sugar sequences in the way that those of proteins
and DNA are controlled today. Gradually reasons are emerging why so many
glycoforms have evolved, although some of these may turn out to be random
and meaningless configurations. In a few years, when some biochemists know
how to produce glycoforms to order, others may also be ready to tailor them
to specific needs.

John Newell is editor of science, technology and medicine for the BBC’s
World Service.