Posted on 07/08/2010 at 08:34:38 PM by Student BloggerBy: Matt T.
A question I hear frequently is, "How much of obesity is due to genes?"
The two most common, and seemingly disparate answers I hear are:
1) Heritability of obesity is about 50-70%, meaning 50-70% of the variation in weight in the population can be attributed to variation in genes (1).
2) The increase in obesity since 1980 cannot be explained by changes in the genome
What most people really want to know when they ask the question is how much of my body weight is determined by my genes. Unfortunately, it is impossible to parse an individual's weight so neatly into 'genetic' and 'environmental' compartments. Instead, a population is studied, and the question that is answered is, "how much of the variation in weight in this whole population is explained by variation in genes?" This is where ‘heritability' comes from.
In order to estimate heritability, populations are studied where the genetic variability is known. For example, the correlation of weight between identical twins gives us the heritability, because the variation in genes between twins is zero, so variability between twins must be attributable to the environment.
Notice, however, that the estimation of heritability does not account for the gene-environment interactions. This is where, in my opinion, heritability estimates are most limited.
Consider a simple thought experiment. We take a pair of identical twins, Seth and Hansel. We drop Seth off in the desert and tell him to walk home, and we drop Hansel into a candy house and tell him to eat his way out. I think few would argue that Seth will get home several pounds lighter, while Hansel will be plumper for the experience.
If we're right, what this tells us is that in certain environments, any genetic influences on our weight are swallowed up by a strong environmental influence, and this seems to agree with common experience.
Now let's suppose we take a hundred twin pairs and split them up, placing them into specific environments. If the twins are divided into very similar environments, then we would expect total variability to be small, and the estimate of heritability will be high. Conversely, if we divide the twins into a wide variety of environments from the desert to the candy house and everything in between, we would expect variability among twins' body weights to be high, according to these environments. The heritability estimate in this case will be low, not because the influence of genes has changed but because the range of environments has changed.
Now we're beginning to see how we can reconcile the two answers to the gene question. Barry Popkin has argued persuasively that, "the world is fat" (2), or more specifically, that the global environment since 1980 has become uniformly characterized by high energy intakes and low energy outputs. If he is correct and the global environment is fairly uniform, high heritability estimates are exactly what we would expect. This does not necessarily mean that genes are more important than the environment in determining body weight, but may simply indicate that our environment has become so overwhelmingly amenable to obesity that genetic influences are free to express themselves, so to speak.
Interpreting heritability is complex, but it's important to get it right. I know people who have given up, saying, "I can't fight my genetic predisposition." Unfortunately, due to genetic influences, it can be harder - sometimes much harder - for some individuals than others. Perhaps the most important message we can take away from the strong genetic influence on obesity is that negative stereotyping of overweight individuals is inappropriate. However, high heritability does not mean that changing the environment can't override genetic influences. The changes have to be big enough to override our biological predisposition to conserve energy, and they have to be permanent.