Genes associated with obesity shared between ancestors, researchers say

Obesity is a global epidemic that affects millions of people every day and is associated with comorbidities ranging from heart disease and type 2 diabetes to osteoarthritis and social stigma. Although lifestyle factors, such as diet and exercise, affect obesity, years of genetic research have identified about 20 genes that have a large effect on a person’s likelihood of developing the disease.

Now, a new study published in Natural communicationsby Penn State researchers, involving approximately 850,000 adults of six continental ancestries, identified 13 genes associated with obesity across ancestries. While eight of these genes had been discovered in previous studies, five were identified for the first time, having no previous links to obesity. Additionally, the team analyzed how these genes influence obesity-related comorbidities, such as type 2 diabetes and heart failure risk.

“Obesity affects millions of people, but most studies have focused on a few,” said Deepro Banerjee, a graduate student in Penn State’s bioinformatics and genomics program and first author of the study. “Previous studies relied primarily on populations of European ancestry, reflecting ancestral bias and missed opportunities to discover additional genes whose mutations might be more prevalent in other ancestry while remaining clinically relevant to Europeans.”

The findings provide insight into the genetic underpinnings of obesity worldwide, the researchers said, explaining that the information could help guide precision medicine efforts by revealing key genes that might be missed in single-population studies.

“Obesity is a complex trait that is influenced by many genetic and lifestyle factors,” said Santhosh Girirajan, T. Ming Chu Professor of Genomics and head of the department of biochemistry and molecular biology in the Penn State Eberly College of Science and author of the paper.

“Single population studies may cause us to miss important genes that are shared between populations, but may not reach statistical significance in any of them, even if they are clinically important in that population. New databases including greater representation of individuals with ancestry from around the world help mitigate this bias, but we still need more data from non-European populations.”

For the study, researchers used data from just over 450,000 adults from the UK Biobank – a biomedical database including genetic, physical and health data from mostly healthy people in the UK – and from almost 385,000 adults from the All of Us research program, a precision medicine initiative of the US National Institutes of Health with a more inclusive cohort that reflects America’s ancestral diversity. The six continental ancestries included were African, American, East Asian, European, Middle Eastern, and South Asian.

“Even with very large cohorts, rare and damaging variants can be difficult to find unless you look at diverse populations,” Banerjee said. “The UK Biobank is made up largely of Europeans, with only around 20,000 non-Europeans in our study sample. By combining the UK Biobank with All of Us, which contributed around 167,000 non-Europeans, we were able to measure the impact on body mass index (BMI), a measure of body fat percentage used as an indicator of obesity, of genes with rare variants and loss of function, independently in each of the six ancestral populations.

The researchers explained that they focused on rare loss-of-function variants because they are most likely to have significant effects on a disease. These variants disrupt the function of a gene and are often found at highly evolutionarily conserved sites in the genome. Their rarity reflects the fact that these harmful changes are generally not very common in the population.

The team combined the non-European populations and performed an association study of all protein-coding regions of the genome with BMI. They found 13 genes with a statistically significant association with BMI in the European group that were replicated in non-Europeans. Of these, eight had previously been associated with obesity, including well-known genes like MC4R and BSN.

Five genes (YLPM1, RIF1, GIGYF1, SLC5A3 and GRM7) had not been associated with obesity in previous studies of rare variants. The researchers found that these new genes conferred an approximately three-fold increased risk of severe obesity, a level of impact similar to that of MC4R and BSN. Like genes previously associated with obesity, the newly identified genes are expressed in the brain and adipose tissue (fat) and have been associated with characteristics of obesity such as increased body fat percentage.

“The new genes identified in our study highlight established and emerging pathways in obesity biology,” Banerjee said. “YLPM1, for example, is a little-studied transcription factor expressed in brain tissue, with links to mental disorders. It is a clear example of a gene whose lower prevalence in a population might have obscured it historically. In our cross-ancestry analysis, YLPM1 shows a remarkably consistent effect across ancestry, similar to MC4R.”

Researchers have also found that several of these genes contribute to other obesity-related conditions, including type 2 diabetes, hypertension and heart disease. Using a statistical method called mediation analysis, they showed different mechanisms by which the risk of comorbidity increases, helping to explain why obesity often leads to other serious health problems.

Mediation analysis helped the team determine whether these genes directly increase the risk of comorbid diseases, or indirectly by first increasing BMI, which in turn drives the risk of comorbidity. For example, the team found that genes such as BSN, GIGYF1 and SLTM increased the risk of type 2 diabetes through direct and indirect pathways, a phenomenon known as partial mediation. Although both effects were significant, the direct effect of these genes on disease risk was stronger than the indirect effect via BMI.

In a subset of individuals whose biobank entries included plasma proteomic data – a complete list of proteins present in their blood plasma – the team also identified changes in circulating proteins linked to the obesity genes they identified. These changes indicate potential drug targets and biomarkers that could guide future treatments and help track treatment response, the researchers said.

“Our results highlight the power and importance of crossover studies,” Girirajan said. “Some of the previously discovered obesity genes appear to only have a significant association with obesity in Europeans, which could limit their potential as therapeutic targets for a global population. We still found some of the most talked about obesity genes, like MC4R and BSN, but we also found several new genes with similar effect sizes, most with clear functional links to obesity.

“Our crossover approach helps us develop a more comprehensive view of the factors involved in obesity, which will hopefully help us develop effective therapies that can be applied through precision medicine.”