Crimson Iris
Back to Blog

The genetics of gluttony

5 min readBy Joe Pickrell

Summary: One in 400 people have a mutation that strongly increases risk of obesity. The subjective experience of having a mutation like this is poorly documented.

The rise of GLP-1 receptor agonists as treatments for obesity has driven home for many just how much of obesity is psychological rather than metabolic. There are now a number of forum threads and popular articles describing how these drugs quiet 'food noise', the low-level and sometimes intrusive thoughts about when and what to eat next.

Over one in 400 people have a genetic mutation that has a large effect on obesity. An interesting question is: are these mutations similarly psychological rather than (or in addition to) metabolic?

First, let's look at the set of loss-of-function variants in the UK Biobank that have the most significant effects on body mass index. The table below shows the top six genes, which have a cumulative frequency of about one in 400 people, and effect sizes (for a man of average height) that range from an additional 10 to 35 pounds in weight.

Loss-of-function variants with the most significant effects on BMI in the UK Biobank, with the frequency of heterozygous genotypes and effect sizes in standard deviations of BMI or of weight for a 5'9'' male
GeneFrequencyEffect Size (sd/lbs)
MC4R0.000420.55/22
KIAA11090.001470.24/10
SLTM0.000180.61/25
PTPRG0.000490.36/15
UBR30.000180.58/23
SLC12A50.000080.85/35

Two things stand out if you survey the literature on these variants:

1. Effectively nothing is known about five of the six genes. The discoveries that KIAA1109, SLTM, PTPRG, UBR3, and SLC12A5 influence obesity in the general population has only been made in the last few years from biobank studies, so literature is non-existent. There's some evidence that the effects of some of these genes depend on gender, but otherwise there's nothing like case studies or community forums to understand the subjective experience of having mutations like these.

2. MC4R effects are indeed mediated through behavior (at least partially). There is an extensive literature on MC4R, though a number of the case studies and behavioral papers focus on homozygous individuals rather than the heterozygous individuals like those in the UK Biobank. There are multiple references to children with MC4R stealing food and otherwise being insatiable (referred to as 'hyperphagia'), and self-reports of MC4R carriers in places like Reddit include people saying they are 'obsessed' with food. A nice visualization of the effects of hyperphagia are in the blow figure from Arnouk et al.

Hyperphagia illustration or data visualization

Visualization of questionnaire responses of parents of children with hyperphagia, include homozygous MC4R individuals. Presumably heterozygous individuals would be intermediate on this plot? From Arnouk et al.

In sum, there is a bit of evidence that genetic factors that increase risk of obesity generally act similarly on behavior (though in the reverse direction) like GLP-1 receptor agonists.