As my Science colleague Jennifer Couzin reports below, many researchers are withholding judgement about the lastest claim for an obesity gene, even if it was published in Science...

Gene Variant May Boost Obesity Risk

By Jennifer Couzin
ScienceNOW Daily News
13 April 2006

Obesity has long been considered partly genetic, but if that's so, where are the genes? Now for the first time, scientists have unearthed a common genetic variant that appears to plump up the risk of obesity in several populations. But the finding hasn't held up in every group studied, and some obesity experts say they need more evidence to convince them that the variant exerts a noticeable pull over weight.

Several genes have been linked to obesity, but virtually all are extremely rare, found in just a handful of people worldwide (ScienceNOW, 13 August 2001). In their hunt for a more common genetic variant, Michael Christman and Alan Herbert of Boston University Medical School teamed up with nearly two dozen colleagues in the U.S. and Germany.

The scientists started with DNA from 923 participants of the Framingham Heart Study, which has followed residents of Framingham, Massachusetts, and their families for over 5 decades. Christman and Herbert's team sequenced tens of thousands of single nucleotide polymorphisms (SNPs) in each of the Framingham samples. SNPs are places where the genomes of people vary by a single base and are considered a useful guide for gene-hunters. After numerous analyses, the scientists found one SNP that appeared to boost the risk of obesity by about 30%, the group reports 14 April in Science.

Christman, Herbert, and their colleagues also tested the SNP's power to influence obesity in five other populations drawn from various studies. In every population but one, the scientists found that their SNP correlated with Body Mass Index and increased an individual's chance of being overweight.

"This is without a doubt the strongest evidence we've seen for a [common] variant associated with human obesity," says biostatistician David Allison, who studies obesity at the University of Alabama at Birmingham. He is concerned, however, about the population where the SNP had no obesity effects: the Nurses Health Study, from which the group tested the DNA of 2700 participants. "It makes one pause," he says, mainly because that group was so large. The Science authors suggest that failure to replicate the SNP effect in the Nurses Health Study may reflect BMI differences, because the nurses tended to be thinner than the general population.

Allison isn't alone in reserving judgment. Philippe Froguel, an obesity genetics expert at Imperial College London in the United Kingdom, and colleagues saw no link between the SNP and obesity in their study of 10,000 European individuals. "In our hands," he says of his still-unpublished work, "we found nothing." Herbert notes that the SNP may not predict obesity risk in all populations.