Gene variation linked to obesity risk

A common gene variation that may increase the risk of obesity by 22 percent might someday be used to find new drug targets and predict whether children will be fat, scientists said.

Investigators used a new technique to sift through about 100,000 gene variations and found one that proved to be a common predictor of obesity when tested in several independent studies, according to a study in the April 14 Science journal.

"We have found a very common determinant of obesity that's present in populations of different ethnicity — affects men and women equally and also children," said Alan Herbert, assistant professor in genetic engineering at Boston University and lead investigator, in a telephone interview today.

The study looked at small gene variations called single nucleotide polymorphisms, or SNPs, in parents to decide which might best explain traits in their children, Herbert said. The researchers said they expect that more SNPs may be found that interact and affect the same biochemical pathway — increasing the obesity risk or keeping people skinny — and pointing to drug targets for development.

The SNP that increased the risk of being fat by 22 percent "makes the difference between being overweight and obese," Herbert said.

The researchers started by studying the SNPS in people enrolled in the Framingham Heart Study, which began investigating cardiovascular disease in a large Massachusetts group in 1948. The scientists asked researchers in five other studies in the U.S., Poland and Germany to search for the same significant link to obesity.

"Everyone came back and said, 'Oh, we see the same thing,' " Herbert said. "This is amazing."

The one exception proved to be the Nurses Health Study, in which participants were a lot more lean than other people studied, he said.

Someday, people who have unsuccessfully dieted and exercised and are considering gastric bypass surgery may have drug treatment alternatives that target the chemical process that synthesizes fat, Herbert said.

The research was funded by the U.S. National Institutes of Health, the American Diabetes Association and the Whitaker Foundation, Herbert said.

"There was a lot of skepticism when we first started that we would find anything," Herbert said. "Now everyone is saying, 'What a great idea. Let's do the whole Framingham population, let's do 500,000 SNPs."'