Science: Advanced smelling tests: Ruth McKernan follows her nose to discover a fragrant world of complex sensory experiences

Why is it that when Jilly Goolden, of BBC 2's Food and Drink programme, inhales deeply over a glass of wine she can identify so many smells - 'barrels full of blackcurrants, oak, lemon and tar with a hint of gooseberries'? Her ability to detect multiple components in a complex mixture demonstrates nasal acuity on a par with that of a professional perfume designer. When I do the same thing I smell - well, wine, really. Some kinds may be more acidic or fruity than others, but that is the extent of my discriminatory powers.

Scientists agree that there is considerable genetic variation in our ability to smell. Recent research suggests, however, that with training it may be possible to improve our sense of smell.

According to Randall Reed, professor of molecular biology and genetics at Johns Hopkins University in the United States, once an odorous chemical is inhaled, smelling it requires three steps. 'First, the chemical binds to a receptor molecule on target olfactory cells in the nose. Second, the signal is amplified inside the cell. Each single odorant molecule generates the production of tens of thousands of intracellular signals. And third, these signals trigger electrical changes in the properties of cells which stimulate nerve impulses to the brain,' he says.

This mechanism conforms to a basic pattern underlying many of our senses. On a molecular level we smell in much the same way as we see. According to Professor Reed, the structure of the receptor molecules and the signalling systems that respond to light and smell are similar.

In the eye there are three types of receptor which are stimulated by red, green or blue light. For smell, there is much more scope for subtlety. Two years ago Linda Buck and Richard Axel from Columbia University in New York discovered a huge family of receptors for odorants. The most recent estimate is that there are between 500 and 800 receptor molecules, each of which is activated by chemicals of a specific structure. While we recognise colour as a combination of three elements, we recognise an odour as a combination of hundreds.

Many of the things we smell are composed of complex mixtures of chemicals. Take perfume, for example. The recently launched DNA scent has a unique floral fragrance. The helix-shaped bottle contains 193 ingredients which combine to make the scent. For the average human nose, detecting 193 chemicals is no problem. According to Professor Reed we can probably smell up to 10,000 different molecules. Odorant molecules are generally small, volatile and lacking electrical charge so they can evaporate, disperse in air and be carried easily into the nose.

We cannot all detect the same odours. The smell of freesias is a classic example: an estimated 10 per cent of the British population is unable to detect this flower's scent. Some people are also unable to smell a breakdown product of green vegetables. Juvenal Urbino, hero of the novel Love in the Time of Cholera by Gabriel Garcia Marquez, appreciated the 'immediate pleasure of smelling a secret garden in his urine that had been purified by lukewarm asparagus'. This delight is not available to all.

Charles Wysocki, a neuroscientist at the Monell Chemical Senses Center in Philadelphia, has investigated why some people are insensitive to a particular chemical. 'We know from studies on twins that it (the ability to smell) has a large genetic component,' Dr Wysocki says. Identical twins have the same sense of smell. 'The difference between smellers and non-smellers derives from the nasal epithelium,' he says. When the ability to smell a specific substance is missing so is the response in the brain. The fault, Dr Wysocki says, can be traced back to the receptor. 'If you can't smell a substance, the most likely explanation is that the receptor for it is not expressed.' We may have up to 800 receptors for different chemicals but we do not all have the same repertoire.

Professor D B Gower, from the department of clinical biochemistry at the London Hospital Medical College, has been studying the role of one group of compounds, 16-androstenes, for 20 years. One of these compounds, 5-androstenone, present in human under-arm sweat and also responsible for the smell of cooking pork, cannot be detected by about 50 per cent of adult males and 10 per cent of adult females. 'It is fascinating that some people smell it. Interestingly, though, we have had a couple of people over the years who, when they first came to the lab, couldn't smell it. By the time they'd finished their PhDs, they could,' he says.

Scientists at the Monell Chemical Senses Center experienced the same phenomenon. Researchers, including Dr Wysocki, who were unable to detect the odour at the start of the project could discern 5-androstenone after working with it for a while. This led them to conduct a controlled study. When a group of people sniffed a bottle of the compound for three minutes three times a day, half of them could smell it by the end of the six-week trial. It was clearly possible to induce 5-androstenone sensitivity in people by repeated exposure.

Dr Wysocki's team has carried out similar experiments on two breeds of mice with differential sensitivity to the unpleasant smell and now have more insight into how this happens. When unresponsive mice were exposed to 5- androstenone intermittently each day for a few weeks new electrical responses were produced in the olfactory cells of the nose. The effect was long-lasting and the ability to smell the chemical persisted for several weeks. 'We speculate that new receptors can be generated in response to environmental stimuli,' Dr Wysocki says.

Unlike sight or hearing, which decline with age, smell, the first sense to develop, is flexible and subject to adjustment throughout life. Pregnant women, for example, are more sensitive to eugenol (a constituent of clove oil) and less sensitive to 5-androstenone. Specific anosmia (the inability to smell) could in some cases be reversed by repetitive stimulation.

The latest research offers the hope that we might be able to improve our sense of smell. Perhaps I should try to expand my repertoire of nasal receptors by regularly opening a bottle of claret and breathing deeply. It seems to work for Jilly Goolden.

(Photograph omitted)