NEW YORK (GenomeWeb) – By profiling a few thousand individuals at the extreme ends of the weight spectrum, a team from the Wellcome Sanger Institute, the Addenbrooke's Hospital Wellcome Trust-MRC Institute of Metabolic Science, and elsewhere has identified new and known body mass index-associated loci, including sites previously observed in cohorts exceeding 40,000 individuals.
The study, published in PLOS Genetics yesterday, "illustrates the value and increased power brought … by using clinically ascertained extremes to study complex traits and provides a valuable resource on which to study resistance to obesity in an increasingly obesogenic environment," co-corresponding authors Sadaf Farooqi, a metabolic researcher at Addenbrooke's Hospital, and Inês Barroso, a researcher affiliated with Addenbrooke's Hospital and the Sanger Institute, and their colleagues wrote.
The researchers did a genome-wide association study that included 1,471 healthy, thin individuals and 1,456 severely obese individuals with early onset obesity, narrowing in on 10 loci already linked to obesity, along with new BMI-associated loci or variants and variants found only in very large cohorts. By comparing the genetics of thin and control participants, meanwhile, they saw genetic influences at both ends of the BMI continuum.
"We show that persistent thinness and severe, early onset obesity are both heritable traits … that share a number of associated loci, and both are enriched for established BMI-associated loci," the authors wrote, though some loci showed stronger ties to high or low BMI.
In particular, the team noted that BMI-associated variants in and around the FTO gene had stronger associations with obesity, while variants near CADM2 were more strongly implicated in thinness. Still other BMI-associated loci seemed to have similar ties to both thinness and obesity.
"These data support expansion of genetic studies of persistent thinness as an approach to gain further insights into the biology underlying human energy homeostasis," they explained, "and as an alternative approach to uncovering anti-obesity targets for drug development."
For their GWAS, the researchers brought together exome array-based genotyping data for nearly 2,000 individuals with severe, childhood-onset obesity, 10,433 population controls, and more than 1,600 individuals classified as persistently thin, but healthy. Following quality control steps, they were left with genotyping data for 1,456 obese, 1,471 thin, and 6,460 control individuals.
Using these data, the team searched for variants associated with obesity, thinness, or BMI through separate comparisons of each individual at each weight extreme to population controls and of the obese and thin participants. For example, analyses focused on 97 loci linked to BMI in European individuals from a GIANT study in Nature in 2015 suggested that similar variants influence both thinness and obesity, albeit to different extents.
The researchers also searched for new obesity- and BMI-associated loci, uncovering a PKHD1 gene variant with apparent ties to BMI that was in linkage disequilibrium with a variant nominally linked to waist and hip circumference in a prior study. And they narrowed in on suspicious variants in and around genes such as FAM150B or PRDM6-CEP120 that were previously found in massive cohorts.
"Our ability to replicate loci found with much larger sample sizes demonstrates the value of clinical extremes," the authors wrote, "and suggest that characterization of the genetics of thinness may provide a more nuanced understanding of the genetic architecture of body weight regulation and may inform the identification of potential anti-obesity targets."