YOU might not have heard of John Pickett’s research, but you may soon be able
to smell it.
Pickett is working on two scents that some people find irresistible—one
smells similar to pine trees, the other like the herbal painkiller, oil of
wintergreen. His team at the IACR, one of Britain’s leading arable crop research
centres, based at Rothamsted in Hertfordshire, have discovered that these
chemicals are so repulsive to some insects that farmers could use them to
protect their crops against pests.
The idea of perfumes as insecticides may sound bizarre but it is simply
taking a leaf out of nature’s book. Pickett’s scents are produced by plants
themselves to help ward off unwanted pests. In fact, researchers are finding
that odours play a remarkably active role in plant defence. Many plants respond
to attack by sending out scented SOS signals that repel pests or summon help
from other insects that prey on the pests. Work on these smelly channels of
communication has even reawoken the controversial theory, put forward in the
early 1980s, that neighbouring plants warn each other of danger.
Advertisement
In commercial terms, the research has huge potential. The global market for
pest control chemicals is $7 billion, with conventional pesticides
accounting for more than 95 per cent of this. Until now, alternatives have been
thin on the ground. Pheromones—in particular the chemicals used by female
moths to attract males—have been sold for two decades. Some farmers use
them to disrupt moths’ mating behaviour and keep crops caterpillar-free. But
they are expensive because each sex pheromone is specific to an individual
species.
Today, however, scientists are devising protection regimes based on cocktails
of the odours used by plants, pests and their predators in nature. These can be
designed to combat more than one species of pest, and so have the potential to
be cheaper than pheromones.
The starting point of this research is to decode insects’ sense of smell. An
insect’s olfactory world has little in common with our own. For example, when
you smell a rose its scent flows over about 50 million odour-sensitive nerve
cells. Each responds to a broad set of volatile chemicals so that every odour
excites a huge number of nerves and they overlap so that different smells blend
seamlessly together. But Pickett’s team has found that an insect recognises a
rose by picking out discrete chemicals— the olfactory equivalent of
scanning a bar code. “We believe that the insect’s olfaction is based on the
responses of highly specific cells,” says Pickett.
Among the pests that Pickett’s group is working on are aphids which, like
moths, are a major threat to crops. The research has shown that not only do
aphids have very specific olfactory nerve cells, but they also have relatively
few of them. The odour-sensitive cells in each antenna number only in the
hundreds. It is this combination of precision and simplicity that may prove to
be the insects’ undoing. In theory, aphids should prove easy to fool using
synthetic versions of those molecules that catch the insects’ attention.
Of course, the real world is not so simple. Pickett and his colleagues have
found that aphid pheromones alone are not effective weapons. For a start, aphids
do not share the obsession for scents that makes moths an ideal target for
olfactory propaganda. Not only are they less sensitive to smells than moths,
they also have less control over where they fly. A strong breeze may be enough
to overrule an aphid’s choice of direction. And, because aphids reproduce
asexually for most of the year, sex pheromones are of little use in reducing
population growth.
One solution seems to be to use the odours as part of a lethal cocktail. For
example, if a predator attacks an aphid, the aphid produces an alarm pheromone
which makes neighbouring aphids scatter. Applying a synthetic version of this
chemical to crops does not disturb aphids for long enough to protect the plants.
But Pickett’s group has found that the chemical is a useful prod—driving
the insects into droplets of conventional insecticide or the spores of a
disease-causing fungus. Similarly, aphid sex pheromones are potent lures for
parasitic wasps, which use them to sniff out hosts in which to lay their eggs.
This may help to improve biological control, allowing the wasps to finish off
the aphids.
Most farmers are not prepared to wait for an infestation before they treat a
crop. Insecticides are often used to keep pests away in the first place. So
could the right smells also act as a deterrent? Gardeners who plant
strong-smelling “companion” plants such as herbs next to their vegetables hope
that the smell of the herbs will divert the pests. Research into companion
planting has produced mixed results, however, and even if it does work, the
method is not practical for farmers. They would need something much simpler and
less space-consuming.
Odour shield
In an attempt to create such an odour shield, Pickett and his colleagues are
making a list of aphids’ olfactory landmarks. To decode the aphid’s sense of
smell, they use microscopic electrodes to record the activity of individual
nerves in the insect’s antennae in response to odours. Aphids, it turns out,
detect odours not only from other aphids and from food plants but also from
plants they do not eat. “When we first entered the field, we encountered cells
which were clearly olfactory cells but we couldn’t find any stimulants for
them,” says Pickett. Now they have found that these cells respond to smells from
non-food plants, repelling the insects from them. This has reinforced the idea
that a vulnerable crop could be disguised as something that is not on an aphid’s
menu. For example, Pickett’s group, together with researchers from Imperial
College, London, found that odours from cabbage can repel bean aphids, Aphis
fabae, from broad beans.
In contrast, the cabbage aphid, Brevicoryne brassicae, is a brassica
specialist and is attracted to them. “The bean aphid and the cabbage aphid have
the same receptor system for detecting brassicas,” says Pickett, “but while the
cabbage aphid is attracted, the bean aphid is repelled.” Cabbages and their kin
protect themselves by accumulating unpalatable glucosinolates in their tissue,
some of which they break down into pungent chemicals called isothiocyanates.
Both species of aphid have receptor cells tuned to very specific
isothiocyanates.
Isothiocyanates smell strongly of mustard, so drenching the air above bean
fields with them is unlikely to receive a warm welcome. “The problem is that
although we eat them, they are in themselves quite unpleasant materials,” says
Pickett. In the past few years, however, the researchers have identified two
substances that could do the same job as isothiocyanates without leaving
farmers’ eyes smarting. Enter our two “scents”. One of them, myrtenal, is
characteristic of herbs such as mint, thyme, sage and rosemary and smells rather
like pine trees. The other, methyl salicylate, forms part of the odour of
several plants, including species in the rose and willow families. It is also a
major component of oil of wintergreen. Pickett’s group is now testing these
scents to see how well they can mask the smell of crops growing in a field.
Research on methyl salicylate is also helping to reveal how plants, pests,
and the predators of pests use scents to affect one another’s behaviour. When
they are under attack, plants activate defence genes to produce protective
chemicals called phenolics and terpenoids. The first group smells of coal tar,
the second, turpentine. Methyl salicylate is one of the phenolics.
Pickett’s team has found that methyl salicylate, which is made by beans,
repels bean aphids. The odour also repels aphids from cereal crops. And aphids
are not alone in responding to methyl salicylate—the scent can also help
to attract predators that eat pests. Marcel Dicke at the Agricultural University
in Wageningen, Netherlands, and his colleagues there and at the University of
Amsterdam have shown that when spider mites attack lima beans, the plants
respond by releasing a blend of volatile chemicals that includes methyl
salicylate. This mixture attracts predatory mites which eat the spider
mites.
Dicke and his team have now investigated the response of several plant
species to attacks from pests. There are two groups. The first includes
cabbages, potatoes and other plants that always contain high levels of defensive
chemicals. These chemicals are released in response to any damage, whether from
cutting or pest attack. Plants in the second group, including beans and
cucumbers, have a specific response to insect damage. “You cannot mimic the
effects of herbivores by mechanical damage alone,” says Dicke. “You have to add
insect saliva to the wound.”
This group responds to pests by releasing a blend of volatile chemicals which
can repel pests and, in many cases, attract their predators. Dicke and his
colleagues dubbed this the “SOS signal”. The odours involved vary by species of
plant and do not always include methyl salicylate. According to Dicke, the most
common feature in this chemical cry for help is the presence of two terpenoids,
known as C11 and C16.
Research in the US shows that plants may even time their SOS signal to
coincide with the activity of predators. Jim Tumlinson at the US government’s
Agricultural Research Service in Gainesville, Florida, and his colleagues have
found that damaged maize and cotton plants produce an SOS signal that attracts
parasitic wasps. The odour builds up over several hours, and once the full SOS
signal is operating, its strength fluctuates depending on the time of day. It
peaks in the early afternoon when the wasps are most active and disappears at
night when they are dormant.
Chemical cry for help
“It’s rather interesting that the plants’ release and the wasps’ hunting are
coordinated,” says Tumlinson. The researchers do not go so far as to say that
plants time their signal only for the wasps’ convenience. The system probably
evolved in response to other factors as well. The terpenoids C11 and C16, for
example, are also present in the scent of many night-opening flowers and so may
be important for attracting moths. If cotton and maize released these substances
after dark they may score an own goal by attracting more caterpillars.
Farmers should be exploiting the SOS signal, argues Tumlinson. But, as Dicke
points out, spraying these chemicals is not an option because a blanket covering
would confuse predators. The best weapon would be a plant that gives out a
strong SOS signal whenever it is attacked, and here there may be a problem.
Tumlinson thinks that crops selected for high yields and direct resistance to
pests and diseases could be losing their ability to call for help, just as
modern varieties of sweet peas and roses may have showy flowers but very little
scent. “There is a wild variety of cotton that releases eight to ten times as
much of these materials when it is attacked than any of the domestic varieties,”
points out Tumlinson. If he is correct, the answer may be genetic engineering
or conventional breeding to give varieties with strong SOS signals.
This may have an added benefit. Research on the SOS signal has led to renewed
claims that plants may communicate with one another. This idea dates back to
experiments in the 1980s on trees. Several researchers showed that damage by
pests to one tree sometimes leads to fewer pests on neighbouring trees. They
suggested that the damaged trees might be sending warnings to others nearby,
telling them to arm themselves against attack. Unfortunately, the resulting
storm of publicity over “talking trees” inhibited research for the rest of the
decade. Now, however, Dicke and his colleagues have shown that even if plants do
not “listen” to each other, they may at least borrow some of their neighbours’
armour.
The original “talking tree” research used trees growing in open conditions
where the results were hard to interpret. “There are so many variables in the
wild that you have to go to the more controlled conditions of the laboratory,”
says Dicke. His team uses wind tunnels to direct the flow of odour between
combinations of infested and healthy plants. They have found that healthy lima
bean plants become less attractive to spider mites and more attractive to
predatory mites after a spell downwind of plants infested by spider mites. The
healthy plants are either detecting the SOS signal from their infested
neighbours and switching on their own defences, or simply absorbing the volatile
chemicals released upwind. Either way, it is something Dicke wants to encourage.
“We need to breed for loudly crying plants,” he says, “rather than for deaf and
dumb plants.”
But there is an obstacle. Most farmers use large amounts of undiscriminating
insecticides, which kill not only pests but also their predators. “When you are
using high inputs, you’re on a treadmill and you can’t really just stop,” says
David Buffin of the Pesticides Trust in London, which campaigns for responsible
use of the chemicals. But if the price is right, the use of natural plant odours
may help farmers climb off this treadmill. Pest control based on odours such as
methyl salicylate could pay for itself by allowing a modest reduction in the use
of artificial insecticides. This in turn would give predator populations a
chance to build up.
Buffin points out that pressure for alternative methods of pest control will
continue to grow. Apart from public disquiet over pesticide use, pests
themselves are increasingly difficult to target. Some populations of the
Colorado potato beetle, for example, have become resistant to virtually all
insecticides. The doomsday case is that eventually pests will become resistant
to insecticides faster than chemicals companies can develop new active
ingredients. Already the balance is tipping in the wrong direction. “The number
of new active ingredients coming onto the market is decreasing,” says
Buffin.
A world in which farmers control pests with smells rather than insecticides
may sound like an environmental Utopia, but chemicals do not have to be toxic to
cause problems. Earlier this year, Bets Rasmussen at the Oregon Graduate
Institute of Science and Technology and her colleagues found that a major
component of the sex pheromones of many moths is also present in the sex
pheromone that female Indian elephants secrete in their urine. Obviously,
there’s no danger that male moths will be making advances to female elephants.
But farmers, beware: Rasmussen’s team has shown that a synthetic version of the
pheromone is enough to attract the attention of bull elephants.