What's your poison?

Why some people become addicted to alcohol when others don't

There are also around 12 million adult smokers in the UK, with smoking-related diseases killing 120,000 people a year. One question that interests psychologists is this: why is it that some people can take or leave alcohol or cigarettes, while others become addicted? If scientists can unravel this puzzle, it could provide new ways of weaning people off drink and tobacco.

New research by scientists at the universities of Bristol and Oxford has added an important piece to the complicated jigsaw of addiction, which is made up of subtle interactions between molecular, genetic, social and environmental factors. The researchers are focusing on the neurotransmitter dopamine, a chemical that is released in the brain in response to activities that we associate with pleasure. These activities can include anything from eating when we are hungry, or drinking when we are thirsty, to having sex.

The release of dopamine is the brain's way of rewarding us for carrying out functions that are essential to our survival, and motivating us to carry on doing these things. However, some chemicals, such as nicotine, alcohol and cocaine, can also elicit a dopamine rush - and paradoxically these substances are more likely to harm us than help us to survive.

"With food and sex and other natural stimulants, the brain rewards you with a modest release of dopamine, so that you learn that these behaviours are worth persisting with," says Dr Marcus Munafo, an experimental psychologist at Bristol. "With nicotine and alcohol you get a much more powerful release."

But why doesn't everyone who tries alcohol or cigarettes become addicted? One strand of research that scientists around the world have been following is based on a natural variation in people's dopamine systems. When dopamine is released in the brain, it exerts its effect by docking on to specific protein molecules in nerve cells called dopamine receptors. Several years ago scientists identified the genes that encode the dopamine receptors, of which there are five types. Studies showed that one class of dopamine receptor, called D2, was the one involved in the effects of nicotine and alcohol.

Importantly, it turned out that there are a small number of variants of the D2 gene that occur throughout the population. Some variants, including one called Taq1A, result in a relatively lower density and distribution of the D2 receptor, while others result in a relatively higher density and distribution.

"In the population as a whole, about 30 per cent of people will have the lower-density version of Taq1A, and 70 per cent the higher-density version," Munafo says. Around 15 years ago a study in the US examined the relationship between alcoholism and the D2 gene. The researchers found that people with the version of the gene that resulted in a lower density of receptors were more likely to become alcoholic than those with the higher density.

"One hypothesis was that the lower number of receptors meant that people get less gratification from drinking a given amount of alcohol, so to compensate they drink more," Munafo says. "This can lead to alcoholism."

But the new work by UK researchers has confused this picture. Munafo and his colleagues were given access to two health surveys, which included data about people's lifestyles - including their smoking and drinking habits - and, crucially, had blood samples for each respondent.

Using modern techniques for genetic analysis, the researchers were able quickly to ascertain which type of D2 gene variant each respondent possessed, and to see if this was reflected in the individual's drinking habits.

"In our sample of just under 1,000 people, around 35 per cent had the D2 gene type that results in a lower density of receptors, while 65 per cent had a higher density version," says Munafo. "What we found was that, on average, people with the lower-density variant drank two units of alcohol a week less than those with the higher-density variant. In other words, if you are within the group with the lower density of D2 receptors, you are likely to drink less than the group with higher density. However, this same group is more likely to produce alcoholics."

The reasons for this apparent contradiction are not clear. "We can speculate that, in general, people who have fewer dopamine receptors of the D2 type will obtain less pleasure from consuming alcohol than those with more receptors," Munafo says. "The physiological reward they obtain is not as pronounced, so they may feel disinclined to drink more - they can take it or leave it. However, if they are relying on alcohol to help them cope with stressful events in their lives, then people with fewer receptors might need to consume more alcohol to achieve the desired effect, and prolonged, increased consumption could lead to alcoholism. This could account for both groups - moderate social drinkers and alcoholics - having the gene variant that leads to fewer receptors. At the moment, though, this is just speculation."

The finding opens up the possibility for new research to examine if there is a correlation between an individual's propensity to experience stress, the person's dopamine D2 system and the likelihood that if the person does drink, he or she is more likely to drink heavily.

As well as investigating the relationship between genetics and behaviour, Munafo and his colleagues are trying to discover if particular dopamine gene variants can predict how an individual is likely to respond to different smoking cessation programmes. "Given that some people do become dependent on drink or tobacco and they want to quit, can genetics tell us which type of treatment is likely to be of most benefit?"

In one study, the researchers followed people who were using nicotine patches to help them quit smoking. "We found that people with a reduced density of D2 receptors benefited more from patches than those with a greater density, and that the effect was especially strong in women," Munafo says.

It is conceivable that in the future doctors could carry out simple genetic tests on people who want to give up smoking, and combine this information with social and personal data on the individual to prescribe the most efficient regime for quitting.

Munafo accepts that such research has important ethical and social implications. "People do sometimes become concerned that insurance companies, for example, might use such data to help in deciding who should or should not be offered cover. The fact is that insurance companies have agreed not to use genetic-test data, and in any event this work is at an extremely early stage. We are not talking about well-defined diseases. We are dealing with much more subtle effects, which are not going to force anyone to behave in one way or another. Insurance companies have much more robust data on which to base their assessments.

"Nevertheless, the ethical and social issues do need to be worked through. Do you tell someone that their genetic profile indicates that they are more likely to become an alcoholic? Do they want to know? If you go to your GP for advice on quitting smoking, and the doctor prescribes a treatment on the basis of a DNA test, and you subsequently fail, will this undermine your confidence?

"We must proceed at a measured pace rather than trying to push ahead regardless. But people do get addicted to drink and tobacco, and if they want to avoid this or if they want to stop smoking and drinking, and if we can better understand some of the underlying factors that might help them to achieve their aims and improve the quality of their lives, then I think this can be a good thing."