NEW YORK (GenomeWeb News) – Rare duplications and deletions contribute to childhood obesity, according to a study appearing online today in the American Journal of Human Genetics.
Researchers from the Children's Hospital of Philadelphia and elsewhere scanned the genomes of thousands of obese and lean American children of European descent, detecting 17 copy number variants in the obese group that weren't found in the genomes of lean children. Consistent with their role in pediatric obesity, almost half of these CNVs also turned up in obese but not lean African American children.
"This is another piece of the puzzle," co-senior author Struan Grant, a researcher affiliated with the University of Pennsylvania and Mount Sinai School of Medicine in New York and associate director of CHOP's Center for Applied Genomics, told GenomeWeb Daily News.
"This doesn't explain a huge amount of the missing heritability [for childhood obesity]," he added, "but I would argue that for the children that harbor these particular rare variants this is quite a sizable event for them."
Being obese increases an individual's risk of several serious health problems, including heart disease and type 2 diabetes. But although environmental factors have contributed to the rise in obesity that's been observed in many parts of the world, the researchers explained, genetic factors play an important role as well.
For example, past GWAS have identified more than a dozen loci associated with BMI. And a large meta-analysis published earlier this week uncovered 18 more loci missed by previous studies.
In an effort to find CNVs specifically involved in early onset obesity, Grant and his co-workers used the Illumina Infinium II HumanHap550K BeadChip to search for CNVs across the genomes of 1,080 obese European American children. They then compared these patterns with those found in the genomes of 2,500 lean children.
"Children have had a shorter period of environmental exposure," Grant explained. "So if they present with a certain condition, our notion is that it's going to have more of a genetic basis to it rather than the adult condition or counterpart, where we hypothesize there will be a stronger environmental component."
The obese children tested belonged to the top fifth percentile on the BMI spectrum while the lean children came below the 50th percentile.
In an effort to circumvent complications introduced by syndromes associated with the most severe cases of obesity, the team did not include children who were three or more standard deviations from mean BMI.
Based on genotype data at some 550,000 SNPs, the researchers identified nearly three- dozen apparent CNVs in the genomes of the obese children that were missing from the lean group. Of these, 17 obesity-related CNVs were missing from the Database of Genomic Variants and could be verified by quantitative PCR.
Most of the obesity-related CNVs fell in and around genes not previously tied to obesity, Grant explained. Even so, one obesity-related CNV did land near ARL15 — a gene associated with levels of the fat hormone adiponectin, as well as heart disease and type 2 diabetes risk through prior GWAS.
When the team looked for obesity-related CNVs in African American children, comparing 1,479 obese and 1,575 lean children, they again found that obese children carried CNVs not present in the lean group.
And almost half of the CNVs detected in the obese European American children — six of the deletions and two of the duplications — also turned up in the heavy group of African American children.
"Because many gene variants have different frequencies in different ethnic groups, detecting these same CNVs in both groups, exclusively in obese subjects, strengthens the probability that these CNVs play a genuine role in the development of obesity," co-senior author Hakon Hakonarson, director of the Center for Applied Genomics at CHOP, said in a statement.
"It was very compelling that almost 50 percent of the variants that we observed exclusive to European American cases were also exclusive to African American cases," Grant agreed.
Within the obese African American group, the team also found one CNV not detected in children of European descent: a deletion near the EPHA6-UNQ6114 locus.
Overall, such CNV findings should offer clues about the underlying biology of obesity, Grant explained, and may eventually help predict some individuals' susceptibility to becoming obese.
Along with translational studies aimed at unraveling the functional consequences of the obesity-related CNVs detected so far, Grant noted that more research is needed to identify additional rare and common genetic factors in obesity — and to determine whether there are genetic variants specific to childhood obesity.
He and his team are currently recruiting additional children and hope to get data on about twice as many European American and African American children as they assessed in the current study. They also plan to look at children from other ethnic backgrounds.
In addition to such array-based CNV studies, researchers at CHOP plan to use sequencing to track down rare variants involved in obesity and other conditions — a strategy that Hakonarson described in an interview with our sister publication In Sequence this week.
"With new advances in sequencing, one would expect further CNVs to be characterized in this condition so collectively they could build up to a meaningful contribution to the missing heritability for this trait," the team wrote.