Skinny Genes: Human Warming and the Reverse Butterfly Effect

Obesity in humans is associated with a region around FTO, a "fat mass and obesity-associated” gene on chromosome 16. Although this is the strongest genetic association, it accounts for only a 1-2% difference in Body Mass Index (BMI) and the mechanism was unknown. Here, the investigators looked at regulator proteins binding to variants DNA sequences within the FTO region, particularly those binding to sites with single nucleotide polymorphisms (SNPs), in 100 healthy Europeans -- 52 subjects were homozygous for 3 risk-variant SNPs (both alleles, all 3 loci) and the remaining 48 were homozygous for the non-risk variants.

They found that the change of a T-to-C at one SNP within a risk allele of FTO prevented the binding of the repressor protein ARID5B. For want of this binding site, the repressor is lost. For want of this repressor, the expression of 2 linked genes doubles: IRX3, located about half a million base pairs (~0.5 Mbp) away, and IRX5, ~1 Mbp away. IRX3 repression made mice thinner by “increased energy dissipation without a change in physical activity or appetite”, i.e. not changing eating or exercise but rather elevating ‘metabolism’. The doubled expression of IRX3 led to a 5-fold reduction in mitochondrial thermogenesis and a 7-fold difference in brown/white adipose tissue development.  (Brown fat is brown because it holds more mitochondria, little furnaces that burn fat and produce heat.)

The risk alleles are causative, and mediate through IRX3 and IRX5, because mimicking the repression of IRX3 or IRX5 in 8 carriers of the risk alleles, but not 10 carriers of the non-risk allele increased stimulated metabolism (Fig 3D, shown, left panel). And overexpression of IRX3 or IRX5 reduced stimulated metabolism in non-risk allele carriers (because their endogenous genes are repressed) but not in risk allele carriers (Fig 3D, right panel).

Figure 3D. Oxygen consumption rate (OCR), basal and stimulated, in cells with risk or non-risk alleles.

The take-home messages are that the causative SNP can be kilobases away from the genes that mediate the effect (not a huge surprise) and that small risks might develop from relatively big differences in particular developmental and cell biological pathways (reverse butterfly effect: not small-change-to-big-effect but big effect leads to small change (1-2% BMI)).

N Engl J Med. 2015 Sep 3;373(10):895-907. “FTO Obesity Variant Circuitry and Adipocyte Browning in Humans.” Claussnitzer M, Dankel SN, Kim KH, Quon G, Meuleman W, Haugen C, Glunk V, Sousa IS, Beaudry JL, Puviindran V, Abdennur NA, Liu J, Svensson PA, Hsu YH, Drucker DJ, Mellgren G, Hui CC, Hauner H, Kellis M