North Carolina
IF AESOP had written a fable about voles, he might have spun a tale around
the virtuous prairie vole and the shameless montane vole. Prairie voles, which
inhabit the Midwestern grasslands of the US, are devoted to their mates and
offspring. They rarely shack up with new partners, even when one dies. Couples
often raise multiple litters together and allow most of their grown youngsters
to stay on, rather than kicking them out of the burrow.
Contrast that with their close Rocky Mountain cousins, the montane voles, who
form few if any social bonds. Montane voles are promiscuous and heartless. They
live in solitary burrows and abandon their young shortly after birth.
Advertisement
What Aesop couldn’t have known is that the two species’ dramatically
different family values may be due to the way nature designed a single gene that
encodes a hormone receptor in their brains, according to psychiatrist and
neuroscientist Thomas Insel at Emory University in Atlanta.
The voles are not the only recent casualties of such biological determinism.
This August, geneticists announced the discovery of the first normal, healthy
gene that is known to have a big impact on the behaviour of a single species.
Fruit flies with one version of that gene tend to wander far from home, those
with another version tend to stay put.
Could variations in one healthy gene be all it takes to create stark
differences in how people behave too? The very suggestion conjures up images of
a future in which engaged couples use a simple genetic test to find out how
faithful their partners will be, where lawyers fight custody battles by arguing
about which parent has the better child-rearing genes, where employers pass over
job applicants whose genes suggest they are too shy or too lazy.
Fairy tales
If those scenarios send shivers down your spine you can take solace from the
fact that, despite work like Insel’s, there is little consensus among
behavioural geneticists about how much influence a single gene can have on a
healthy person’s personality. But they all agree that the relationship between
genes and behaviour is far from fairy-tale simple, and that practically everyone
who’s not a behavioural geneticist has got hold of the wrong end of the stick:
there will not be a single gene for shyness, another for good parenting skills,
and another for risk-taking. The reality is that multiple genes underlie every
physiological process in the body and must underlie every animal and human
behaviour as well. Add the environment’s giant influence to this mass of genes,
and to some geneticists this suggests that predicting a healthy person’s
personality from their genes will remain the province of science fiction.
“It may be so complex, with so many nonlinear gene-gene and gene-environment
interactions, that we will not be able to assign precise contributions to
particular genes,” says Steven Hyman, a psychiatrist and molecular
neurobiologist, and the director of the National Institute of Mental Health near
Washington DC.
But others are more bullish. They are confident that nothing in behavioural
genetics is so complicated that it cannot be prised open by new tools being
developed. Some even suspect that, as in voles and fruit flies, certain human
behaviours may be programmed by “major-effect” genes—single genes that
have a disproportionate impact.
For many behaviours there will be dozens or hundreds of genes and all of them
may exert small effects, says Dean Hamer, who studies genetics and behaviour at
the National Cancer Institute near Washington DC. But, he adds, “some behaviours
will have major-effect genes”.
Even so, discovering those major-effect genes in humans is going to be as
tricky as listening to a symphony and trying to hear the contribution of each
instrument—with a thundering rainstorm going on outside.
That rainstorm is, of course, the effect of the environment on behaviour.
Environment includes everything from exposure to sex hormones in the womb to
whether you live a life of coddled privilege or abject poverty. But those who
would take comfort from the idea that the environment is likely to overwhelm any
genetic component of a human behaviour best consider a recent study led by
Robert Plomin, a behavioural geneticist at the Institute of Psychiatry in
London. Plomin’s team compared mental abilities in elderly identical twins (who
share all their genes) with those of nonidentical twins (who share on average 50
per cent of their genes). Even though we clock up more unique experiences as we
age, and one might expect that we increasingly become a product of our
environment, the study found that the contribution genes make to mental
abilities remains remarkably constant throughout adulthood at around 62 per cent
(Â鶹´«Ă˝, 14 June 1997, p 16).
The bottom line of this study, and a whole mass of twin studies on
personality traits ranging from shyness to cheerfulness, is that genes are
profound shapers of behaviour, says Plomin. But Plomin, unlike Hamer, doubts the
existence of major-effect genes in humans. “If there were `big effect’ genes,
[geneticists] would have found them,” he says.
He also points out that in some ways it’s misleading to put percentages on
the relative contributions made by genes and the environment, because the two
are totally entwined, each influencing the other. Take a crude example: a person
born with a “music” gene will seek out a musical environment. That environment,
in turn, will switch on certain genes to create links between certain brain
cells. Over time, these links will make the person even more “musical”.
Clearly, even when genes play a strong role in a behaviour, working out their
individual contribution is a huge challenge. Unlike such heritable traits as eye
colour, behaviour is hard to define. How, for instance, would you define
friendliness or adaptability to change? Behaviour is also notoriously difficult
to quantify. Most behavioural traits vary only by degree—people may be
very shy, very gregarious, or something in-between—and those variations
can be nearly impossible to measure even with the most sophisticated
psychological tests.
In the past, behavioural geneticists have tried to get around these problems
by looking for genes linked to clear-cut behavioural abnormalities that run in
families. They monitor the inheritance of genetic markers—easily
identifiable stretches of DNA—in family members who have a particular
trait, for example depression, and those without. If the markers are inherited
only by the depressed members of the family, this shows that they lie close to
the gene or genes responsible for the depression. Once the stretch of the genome
has been narrowed down to a manageable size—an area that may still contain
several hundred genes—the gene hunters resort to techniques such as
transcript mapping and exhaustive DNA sequencing to pull out the actual gene. It
is “pedigree” analyses like these that have linked defective genes to extreme
violence in one Dutch family (Â鶹´«Ă˝, 30 October 1993, p6) some
types of mental retardation, and constant overeating.
Dirty spark plugs
The trouble is, these studies find damaged genes that cause distinctly
abnormal behaviour. No one has ever shown that healthy versions of such genes
are responsible for normal variations in behaviour, although that is just what
geneticists once speculated would be the case. This is rather like thinking that
because your car won’t start with a dirty spark plug, the spark plug alone
determines how well the car runs.
Instead, researchers who want to track down the genes that play a role in
normal behaviour are developing statistical tools that make it possible to look
beyond families stricken with behavioural disorders to large numbers of
unrelated people with a huge variety of genes and personalities. And with
hundreds of research teams now involved in the chase, it’s not surprising that
every few weeks a new study is published reporting a possible link between a
gene and a normal human behaviour: novelty-seeking, happiness, gayness and
tendencies towards anxiety, to mention but a few.
But as with any new gizmo, the wrinkles in the techniques have yet to be
smoothed out, and every few weeks one of the supposed links falls apart on
closer scrutiny. “The half-life of some of these stories is about six months,”
says Jeff Hall, a neurogeneticist at Brandeis University in Waltham,
Massachusetts, who studies the behavioural genetics of fruit flies.
So what makes Insel think his research on voles has put him on the trail of a
major-effect gene for a complex human behaviour like social bonding?
Because Insel and his colleagues work on animals, they could take a more
direct route to finding their gene. They compared the brains of the female
prairie voles and female montane voles, and found that the receptors for a
hormone called oxytocin, which is produced in large amounts after giving birth,
were located in completely different areas of the brain in the two species. This
suggests that the pattern of oxytocin receptors in the brain plays an important
role in female bonding with her offspring and her partner. Insel and his
colleagues ruled out the possibility that the behavioural differences were due
to levels of the hormone itself.
“You can give a montane vole all the oxytocin you want and [she] won’t become
monogamous,” says Insel. “That’s because [her] receptors are in an area of the
brain devoted to scratching and not pair bonding.”
Cheating loner
Next, the team took their search down to the level of the gene for the
oxytocin receptor. When they compared the oxytocin receptor gene from the
prairie and montane voles, there were slight differences in one part called the
promoter. The promoter serves as the dial that turns the gene’s activity up or
down, and so helps determine which cells will have the receptors. Insel now
thinks that this is why the female prairie vole is faithful to her family and
the female montane vole is a cheating loner. With different promoters for the
oxytocin receptor genes, they are turned on in different parts of the brain in
the two species, and that leads to dramatically different social behaviour.
Insel’s group conducted parallel studies on a gene for the receptor for
vasopressin—a hormone that plays a role in the male vole’s bonding with
mates and offspring—and obtained similar results.
What does this mean for humans? Like all mammals, we have genes for oxytocin
and vasopressin hormone receptors in our brains. But in the vole studies Insel
looked at differences between two species, which over aeons evolved different
versions of the genes and different behaviours. Humans, on the other hand, are a
single species made up of individuals with bonding behaviours that range from
lifetimes of monogamy to rampant polygamy.
What Insel’s work does do is give clues about the kind of genes that could
have a major impact on human behaviour. These genes won’t be “master genes” that
direct the activity of myriad less important genes; one thing is
certain—there is no strict gene hierarchy. Instead, large numbers of genes
work in concert to create behaviours rather like the instruments in that
symphony orchestra. Even so, some instruments, such as the lead violin, are more
important than others—say, a fifth chair flute. Genes for hormone
receptors are natural lead violins because they modulate the action of hormones,
many of which are known to influence the brain, behaviour, and other genes.
Insel says that if the oxytocin and vasopressin receptor genes do play an
important role in human bonding behaviour, there should be substantial variation
in the genes from individual to individual. No one has looked for this yet, but
a closely related gene—one for the vasopressin 2 hormone receptor—is
known to vary widely, suggesting that this class of genes is easily mutated into
new versions.
Under wandering stars
Other candidates for lead violin status are genes that have an impact on the
brain either by affecting how it is wired up as the fetus develops, or how brain
cells communicate with one another in the adult animal. For example, the
foraginggene that dictates how far a fruit fly strays from home, encodes an
enzyme that helps to relay messages between brain cells.
A “rover” fruit fly has a version of the gene that makes for slightly higher
levels of the enzyme in the brain than the version in a “sitter” fruit fly. What
makes this discovery different from all the other genes that alter fly behaviour
is that this one is not an artificially created mutation. Instead, subtle and
naturally occurring differences in one gene are enough to make a difference in
normal fruit fly behaviour.
That isn’t to say that the influence of the environment is trivial. “If you
starve a rover, it will appear more sitter-like,” says Marla Sokolowski at York
University in North York, Ontario, who led the team that discovered the gene,
“and if the food quality is poor, a sitter will start to roam.”
Humans have a gene for the same enzyme, and it may have similar functions in
our brain cells. The gene may even play a role in normal human eating behaviour,
says Sokolowski. But she cautions that drawing parallels between fruit flies and
humans is a shaky business because humans have far more complicated brains.
Human brains also contain roughly 30 000 active genes—far more than in the
fruit fly—making it even more difficult to find any major-effect behaviour
gene. “This is very tough to do in flies, and I can just imagine how hard it
will be in humans,” says Sokolowski. Her group discovered the strong behavioural
role of the fruit fly foraging gene only after 20 years of research,
having investigated seven other behaviours without finding a single major-effect
gene.
But there are others who think they are closing in on major-effect genes for
human behaviour. David Skuse, a child psychiatrist at the Institute of Child
Health in London, reported in Nature this summer (vol 387, p 705) that
he has evidence for the existence on the X chromosome of a single gene, or
possibly a small cluster of genes, that plays a major role in normal social
skills, such as awareness of other people’s feelings and the ability to chat and
make friends.
Skuse got his first hint that the gene existed from girls with Turner’s
syndrome, who have only one X chromosome. Girls who inherit their mother’s X
chromosome lack social skills, whereas girls who inherit their father’s have
normal skills. Next, Skuse showed that normal girls without the syndrome, all of
whom have a copy of their father’s X chromosome, generally have rather better
social skills than boys, who only get an X from their mother.
Unlike disease linkage studies, Skuse’s investigations show that the
completely routine absence or presence of an active gene, caused by a mechanism
called imprinting, underpins two very different—but entirely
normal—personality types. Skuse speculates that the gene or genes may
alter the wiring or levels of chemicals in the brain’s prefrontal cortex, where
tasks involving memory, reasoning and judgment are mediated.
“I suggest that girls are genetically preprogrammed to learn almost by
instinct to interpret social cues,” he says. “Boys on the other hand do not have
this advantage and have to work harder to get to the same point.”
Vole makeover
This has yet to be proven. But even if human geneticists never manage to pin
down a major-effect behaviour gene, it may one day be possible to predict a fair
bit about someone’s personality from clusters of “small-effect” genes, which are
becoming easier and easier to find. Geneticists have already spotted genes that
account for tiny variations in normal human behaviour.
For instance, Hamer has discovered that variations in a gene for a protein
that affects the amount of serotonin in the brain accounts for just 4 per cent
of the difference in the amount of anxiety normal people experience.
“The technology is getting more powerful,” he says. “Resolution will continue
to increase to the point that we will detect genes of such small effect that
they will no longer be interesting.”
There are several scientific developments spurring this progress. First,
there is the sheer number of new gene sequences being churned out by the Human
Genome Project. Combine that with an increasingly sophisticated understanding of
what genes with a particular sequence may or may not do, and researchers have
plenty of candidate genes on which to focus. Then there are “personality DNA
databases”. All around the world, behavioural geneticists are asking volunteers
to submit blood or cheek swabs for DNA testing, and having them fill out
psychological questionnaires that gauge personality.
Finally, there are improved computer programs and statistical techniques that
allow geneticists to sift through vast amounts of data, picking out the subtle
interactions between genes, the environment and personality traits. With these
tools, geneticists are able to rely more on “association studies” to track the
relationship between candidate genes and a particular trait across whole
populations. This is a far more powerful way of pinning down behaviour genes
than any that have gone before.
Plomin suggests that researchers may one day be able to announce that, say,
10 or 20 genes account for half of the normal variation seen in a particular
human behaviour. “But that,” he says, “is a long way down the line.”
In the meantime, animal studies continue to give researchers new insights
into the murky relationship between genes and behaviour. Insel, for his part,
wants to prove conclusively that the oxytocin and vasopressin receptor genes can
explain the differences in bonding behaviour in prairie and montane voles. He
plans to create transgenic voles in which the genes for the receptors have been
switched between the two species. If he can make a montane vole faithful and
family-oriented, and a prairie vole promiscuous, he will surely have revealed
the awesome power of some genes over behaviour.
- Further reading:
Natural behavior polymorphism due to a cGMP-dependent protein kinase of Drosophila
by K. A. Osborne and others, Science, vol 277, p 834 (1997) - A neurobiological basis of social attachment
by T. R. Insel, American Journal of Psychiatry, vol 154, p 726 (1997)