By Beth Azar
Monitor staff
The '90s could be hailed as the decade of the gene.

A steady stream of researchers has claimed to have found the genes responsible for everything from breast cancer to obesity. In the latest of such discoveries, researchers have developed a powerful genetic technique that allows psychologists to combine their behavioral expertise with molecular biology to search for genes that affect the behaviors associated with alcoholism.

Researchers have known for more than 100 years that alcoholism runs in families. Family, twin and adoption studies prove that genes, as well as environment, play a role in drinking problems. The premise is that genes-the body's underlying control manual-cause problems when they're either too active, inactive or act aberrantly. But few researchers believe they'll find a single alcoholism gene. More likely, they expect to find many genes that together predispose a person toward abusing or depending on alcohol.

Behavioral scientists have identified many traits that seem to affect a person's propensity to drink, but alcoholics may have countless combinations of these traits. Alcoholics tend to have different sensitivity to alcohol than nonalcoholics: They may develop tolerance for large amounts of alcohol and dependence on its effects, they may feel rewarded when they drink and suffer withdrawal when they don't. Temperament may also play a role in drinking; alcoholics tend to be more aggressive, hyperactive and prone to risky behaviors.

Any or all of these traits may be inherited, said Ellen Witt, PhD, program director for behavior in the National Institute on Alcohol Abuse and Alcoholism's Division of Basic Research. For any trait that is inherited, researchers search for genes that affect how that trait is expressed.

Animal models

Other trait models developed by researchers include high- and low-drinking strains; strains that sleep for a long time after drinking; strains that sleep for a short time after drinking; strains that develop severe alcohol withdrawal symptoms after chronic alcohol exposure stops; and strains that develop mild withdrawal symptoms.

That the models exist proves that these traits have some genetic component, said Witt. Although no animal models duplicate the complexity of human alcoholism, these individual trait models allow researchers to study pieces of the problem. Genetically, even the simple, easy-to-measure traits, such as sedation, are likely controlled by multiple genes. Therefore, trait models are ideal for gene searches because researchers can use individual strains of animals to search for genes for one trait at a time. A new genetic technique, called quantitative trait loci mapping, or QTL, allows researchers to identify genes that contribute to complex traits.

In the neighborhood

'It puts you in the neighborhood on the chromosome,' said psychologist Christopher Cunningham, PhD, a QTL researcher at Oregon Health Sciences University.

The technique usually identifies a few QTLs for any one trait, implying that more than one gene affects that trait. But because an initial QTL search is preliminary, each location found must be verified.

Mouse chromosomes are fairly well mapped so once a QTL is confirmed, researchers can look at a genetic map to see if any known genes exist at that location. If so, they can see if the gene's function could affect the behavior in question.

If no known genes are mapped at the QTL, researchers must randomly search the location for a gene, then study the gene's function.

QTL mapping has had a major impact on behavioral genetics, according to John Crabbe, PhD, a leading QTL alcohol researcher at Oregon Health Sciences University and the Veterans Administration Medical Center. He and John Belknap, PhD, also of OHSU, coordinate a QTL program project grant from the NIAAA. Their group of more than 12 researchers has already found QTLs for many alcoholic behaviors and is moving ahead rapidly.

'In the next year we'll have five times to 10 times more information than we currently do,' said Crabbe.

Once researchers find genes for alcohol-related behaviors in mice, they can then look for a similar gene in humans, said Robert Karp, PhD, program director for genetics in NIAAA's Division of Basic Research.

If a human gene exists, researchers can check if the gene differs in human alcoholics as compared to nonalcoholics.

As part of a study called the Cooperative Agreement on Genetics of Alcoholism, NIAAA is in the final stages of collecting genetic samples from a large group of families affected by alcoholism. This sample will be the ideal place to test genes found through QTL mapping in animals.

Success stories

At the definitive site-a region holding as many as 100 functional genes-two genes could be involved in alcohol withdrawal, said Crabbe.

The most plausible of the two is a gene called Gad-1. This gene encodes an enzyme that promotes the production of GABA, an inhibitory neurotransmitter that has been associated with alcohol withdrawal seizures, said Crabbe. A disruption in GABA production, caused by problems with the Gad-1 gene, could theoretically affect alcohol withdrawal. Another researcher on Crabbe's team, Kari Buck, PhD, is manipulating the Gad-1 gene to study this association.

Cunningham works with more complex behaviors, such as conditioned place preference, which predicts whether an animal likes alcohol. To measure conditioned place preference, researchers put a mouse in a cage with a characteristic floor-a grid-and inject it with alcohol until it associates the environment with alcohol. They then put the mouse in a cage with half the floor the same as the first cage and half with holes instead of a grid. If a mouse likes alcohol, it will spend most of its time in the alcohol-related side of the cage. Cunningham has identified potential QTLs for conditioned place preference on several different chromosomes and is seeking to verify them.

QTL researchers must study between 500 and 1,000 mice for each trait. At one phase of the process, they measure the trait of interest-withdrawal symptoms, alcohol consumption, sleep time-in each mouse to see how strong the trait is. The harder a behavior is to measure, the harder the study will be to complete, said Karp. As such, most of the QTL studies to date concentrate on alcohol sensitivity traits that tend to be easier to measure than temperament, he said.

NIAAA would like to see researchers begin genetic studies on even more complex behavioral models. Behavioral researchers have a wide array of rodent models for complex behaviors such as impulsiveness, novelty seeking, aggression and anxiety.

To encourage people who do interesting behavioral work to think about conducting a genetic analysis, NIAAA recently released a Request for Applications for QTL research. NIAAA hopes to foster collaborations between behavioral researchers and geneticists, said Karp.

'We're working at the level of behavior and then exploiting the techniques that molecular biologists have developed,' said Cunningham.

'This research ultimately has the potential to identify specific genes and makes it easier to get at underlying biological factors and sort out the nature-versus-nurture debate,' he said.