By JOHN DINSMORE
"Could rheumatoid arthritis be one of the casualties of the battle
between the sexes? That's the
implication of an intriguing study reported recently by researchers
at Harvard Medical School, who
have created two strains of mice, both of which are genetically
identical. But while one can inherit
rheumatoid arthritis, the other group is free of it.
The key to this genomic mystery is a curious trick that some genes
can pull off called imprinting.
Imprinting is a way of bypassing the normal rules of inheritance
that may also be involved in other
disorders, including autism, schizophrenia and even cancer. But
let's begin at the beginning - with sex.
One of the basic ideas behind modern evolutionary genetics is that
males and females have different,
and sometimes conflicting, interests. It pays for males, for example,
to spread their DNA around, while
females benefit from husbanding their reproductive resources more
carefully.
This is where imprinting comes in. One of the surprising genetic
findings of recent years is that genes
can have different effects depending on whether they come from the
mother or the father. In classic
genetics, genes are sexless, they do the same thing regardless of
their origins, but researchers now
know of at least 40 genes that are "imprinted".
Paternally imprinted genes generally function only when they come
from sperm, while maternally
imprinted ones are turned off unless they originate from an egg.
The father's genes go for growth, while
the mother's ones try to keep these competitors under control.
A good example is what happens as soon as the fertilised egg arrives
in the womb. Immediately it starts
building a placenta, which hooks up to the mother's blood supply.
The genes that control this are
paternally imprinted. Their overriding aim is to draw as many nutrients
from the mother as possible.
However, they are matched by a set of maternally imprinted genes
that damp their voracious activity.
It is not in the mother's interest to have unlimited supplies going
to this one baby - she needs to
conserve her strength for others. The male genes, however, may never
inhabit that womb again. Animal
studies show how disastrous it is if either side wins this arms
race. Animals with the paternal genes
knocked out never get enough nourishment and those without the maternal
ones grow so large they
damage the mother.
However, research by Azim Surani at Cambridge suggests the role of
imprinted genes goes way
beyond a tug-of-war over the food supply. In experiments with mice
he has found that paternally and
maternally imprinted genes are responsible for building different
areas of the brain.
Contrary to the old stereotype about rational men and emotional women,
it looks as if it is paternally
imprinted genes that build the limbic system, the region that controls
our desires and emotions, while
the part that looks after judgment and intelligence uses only mum's.
No one has yet shown that the same goes for humans, but two rare
genetic diseases suggest that we
are indeed built along the same lines. If the masculine half of
a particular imprinted gene is missing, then
the maternal one controls development and the result is a child
suffering from Prader-Willi syndrome. It
will be mentally retarded, as well as placid, tending to over-eat
and later have little interest in sex - all
functions controlled by the limbic system.
However, when the female half of another imprinted gene is missing
the result is Angelman syndrome -
a male gene trying but failing to control cortical development and
so the child is retarded, has poor
control of movement and can't speak.
All of which brings us back to rheumatoid arthritis and the work
of Denise Faustman at Harvard. She
was originally working with a strain of mice that were susceptible
to diabetes - some, but not all, were
likely to get it. So she took the unaffected animals, mated them
and bred a strain that didn't develop
diabetes. To her surprise, however, another strain emerged that
got rheumatoid arthritis. "It's
remarkable," she says, "that the same genes can result in inherited
diabetes, arthritis or a healthy
animal."
Investigating further, she found that the difference lay in the imprinting
on some of the genes. She had
evidence that imprinting can affect cytokines, a group of molecules
that control the immune system.
Imprinting is a hot field of research that may hold clues to other
diseases, such as autism. Last autumn,
researchers at Duke University found both a paternally and a maternally
imprinted gene common in
children with autism, but as yet they don't know what the genes
do.
Another imprinting connection has been made with cancer. The link
comes from the first imprinted gene
to be discovered, a paternal one known as IGF2, which is, as you
might expect, a growth promoter.
Knock it out and rats are born 30 per cent smaller. It is balanced
by the maternally imprinted gene -
M6P - and both are involved in cancer. That is because when they
go wrong, you get uncontrolled cell
growth.
"Imprinted genes pose a danger to health because they have no backup,"
says Randy Jirtle, a
researcher at Duke University. "If they go wrong, that's it."
But there are many unexpected connections in the imprinting story
- IGF2, for instance, has also been
linked to having a high IQ. What's more, the gene illustrates how
sophisticated the imprinting arms race
can become. Recently scientists found that the ovary can disable
some paternally imprinted genes.
Imprinted genes have chemical markers on them, known as methyl groups,
but scientists at the Max
Planck Institute in Berlin found that eight hours after the sperm
carrying a modified IGF2 had entered
an egg, most of its methyl groups had been removed. Why the egg
targets only the IGF2 gene is not
clear.
Other results suggest imprinting may not just be about sexual advantage.
Last year Surani found that a
paternally imprinted gene called Peg3 controls the way female mice
look after their young. Without it
they make hopeless mothers. If the sexual conflict theory is true,
a father's genes would have to benefit
more than the mother's if their grandchildren are well cared for.
But since that's unlikely, imprinting
remains a mystery.
Imprinting may one day point to a new range of treatments but for
now it is largely a big and very
intriguing puzzle."
2. "Imprinted Expression of Small Nucleolar RNAs in Brain: Time for RNomics".
3. "Maternal Methylation Imprints on Human
Chromosome 15 are Established During or After Fertilization".