Society for Endocrinology - a world-leading authority on hormones

The Endocrinologist

Issue 132 Summer 2019

Endocrinologist > Summer 2019 > Features

Loading the dice: genetic factors influence BMI and obesity

Sadaf Farooqi | Features

On the face of it, the rising tide of obesity seems to have little to do with genetics. As a society, we eat more and move less than we need to. Weight gain and, in time, obesity (body mass index (BMI) >30kg/m2) follow. However, in an environment where highly palatable, calorie-dense food is cheap and easily available, many people maintain a normal body weight, and some are particularly thin. Why are people so different?

A commonly held view is that people who are obese choose to eat more and to exercise less. Obesity is seen as a direct consequence of a lack of willpower and self control: ‘It’s their fault!’ While some people do eat more and move less than others, genetic factors strongly influence these behaviours, and when faced with more calories than we need, genetic factors also influence the amount of weight we gain.


Studies of families, of identical and non-identical twins and of children who were adopted, all indicate that 40−70% of the variation in weight between people is heritable. In a landmark (indeed, heroic) Canadian study, 12 pairs of identical male twins lived under constant supervision for 100 days. They were given a diet that exceeded their energy requirements by 1000 calories a day. Not surprisingly, they all gained weight.1 The amount gained varied considerably across the group but was very similar between members of a twin pair. Similarly, twins given a reduced calorie diet lost very similar amounts of weight, but across the group there was a lot of variation.

Inherited factors influence how much food we eat, levels of fullness/satiety in response to a fixed meal, basal metabolic rate and the physiological response to a fixed amount of exercise. Lee Kaplan and colleagues showed that the response to bariatric surgery (Roux-en-Y bypass) is highly heritable.2


One way of assessing the contribution of many different genetic variants to a person’s BMI is to add up their effects to derive a risk score. BMI risk scores derived from hundreds of common variants found in genome-wide association studies are higher in obese people than in normal weight people, and very low in extremely thin people.3

Recently, Katherisan and colleagues tapped into genetic data from 0.5 million people involved in UK Biobank. By adding up the contribution of 2 million variants they derived a risk score that, for the first time, predicts obesity from age 12 years onwards.4 Their work, alongside that of others, now unequivocally shows that the dice are loaded against people who develop obesity, and in favour of others who carry variants that protect them from developing obesity5 and allow a subset of people to stay very thin.3


Some healthcare professionals (and indeed politicians) worry that knowing obesity is so strongly influenced by genes will give people permission to eat what they like ‘because it’s all genetic’. In practice, this seldom happens. Thinking about other clinical areas, if we recognise the real difficulties that people have with their weight, and why, then we may be able to have a more constructive relationship with patients with severe obesity, which can only be a good thing when managing chronic conditions.

Genetic testing (now available in the NHS) is particularly important in people who develop severe obesity at a young age (before the age of 5 years), who have a high chance of carrying penetrant variants which disrupt the hormone leptin and the hypothalamic neuropeptides and receptors involved in leptin signalling. One condition, albeit rare, responds dramatically to therapy, and there are other genetic obesity syndromes for which targeted clinical trials are ongoing ( Importantly, a genetic diagnosis challenges those who advocate removing children with severe obesity from the care of their own families.


There are likely to be hundreds, perhaps thousands, of genes that affect a person’s weight. Finding these genes remains a powerful tool for discovering the mechanisms that regulate weight, which can, in turn, highlight new approaches to therapy. One example of a potential target informed by genetic studies is MC4R (melanocortin 4 receptor), a G protein-coupled receptor which acts in the brain to suppress appetite. People who have variants that disrupt MC4R gain weight easily. Different variants in MC4R that increase its activity by signalling through the β-arrestin pathway are associated with protection from obesity and its metabolic complications.5 These studies suggest that drugs that mimic the protective variants could provide new, safer, weight-loss therapies.

So, finally, the evidence is piling up. When it comes to developing obesity, it is now very clear: the odds are stacked against some people and in favour of others.

Sadaf Farooqi, Wellcome Principal Research Fellow, Professor of Metabolism and Medicine, Wellcome−MRC Institute of Metabolic Science, University of Cambridge, and Honorary Consultant, Cambridge University Hospitals NHS Foundation Trust


  1. Bouchard C et al. 1990 New England Journal of Medicine 322 1477–1482.
  2. Hatoum IJ et al. 2011 Journal of Clinical Endocrinology & Metabolism 96 E1630−E1633.
  3. Riveros-McKay F et al. 2019 PLoS Genetics 15 e1007603.
  4. Khera AV et al. 2019 Cell 177 587−596.e9.
  5. Lotta LA et al. 2019 Cell 177 597−607.e9.

This Issue:

Summer 2019

Summer 2019