An evolutionary perspective on the genetics of anorexia nervosa

Abstract

Anorexia nervosa (AN) typically emerges around adolescence and predominantly affects females. Recent progress has been made in identifying biological correlates of AN, but more research is needed to pinpoint the specific mechanisms that lead to its development and maintenance. There is a known phenotypic link between AN, growth and sexual maturation, yet the genetic overlap between these phenotypes remains enigmatic. One may hypothesize that shared factors between AN, energy metabolism and reproductive functions may have been under recent evolutionary selection. Here, we characterize the genetic overlap between AN, BMI and age at menarche, and aimed to reveal recent evolutionary factors that may help explain the origin of AN. We obtained publicly available GWAS summary statistics of AN, BMI and age at menarche and studied the polygenic overlap between them. Next, we used Neandertal Selective Sweep scores to explore recent evolutionary selection. We found 22 loci overlapping between AN and BMI, and 9 loci between AN and age at menarche, with 7 of these not previously associated with AN. We found that loci associated with AN may have been under particular evolutionary dynamic. Chronobiology appeared relevant to the studied genetic overlaps and prone to recent evolutionary selection, offering a promising avenue for future research. Taken together, our findings contribute to the understanding of the genetic underpinning of AN. Ultimately, better knowledge of the biological origins of AN may help to target specific biological processes and facilitate early intervention in individuals who are most at risk.

Fig. 1. Anorexia nervosa genetically overlaps with body mass index and age at menarche.

Stratified quantile-quantile (QQ) plots displaying (A) increasing levels of SNP enrichment for AN conditioned on association p-values for BMI, and vice versa; (B) increasing levels of SNP enrichment for AN conditioned on association p-values for age at menarche, and vice versa.

Fig. 2. Identification of loci shared between anorexia nervosa and body mass index, and anorexia nervosa and age at menarche.

Manhattan plots presenting in color the significant loci identified as shared between (A) AN | BMI and; (B) AN | Menarche.

Fig. 3. Comparison to Neanderthal genome suggests recent evolutionary selection for SNPs associated with anorexia nervosa.

Quantile-Quantile (QQ) plot showing the enrichment of more negative Neanderthal Selective Sweep scores in SNPs associated with AN, pointing toward recent positive evolutionary selection.

Fig. 4. Compared to body mass index and age at menarche, SNPs associated with anorexia nervosa appear to be under stronger evolutionary selection.

Distribution of Neanderthal Selective Sweep scores for candidate SNPs associated with (A) AN, BMI and those shared between AN | BMI; (B) AN, age at menarche and those shared between AN | Menarche; (C) BMI, age at menarche, and those shared between BMI | Menarche. Negative Neanderthal Selective Sweep scores point toward recent positive evolutionary selection.

Fig. 5. Genes associated with anorexia nervosa, highlighted from an evolutionary perspective.

Genes mapped to the loci associated with AN (top panel), AN | BMI (middle panel) or AN | Menarche (bottom panel). The x-axis represents Neanderthal Selective Sweep scores, with more negative Neanderthal Selective Sweep scores pointing toward recent positive evolutionary selection. Neanderthal Selective Sweep score per gene was determined as an average of the SNPs from which the gene was mapped. Location on y-axis is random (jitter) and was used for legibility of otherwise overlapping text. Darker colours are used to highlight genes with the lowest mean NSS scores.