Genetics Of Behaviors Associated With Eating Disorders

Eating disorders (ED), comprising anorexia, bulimia, and binge-eating, have become the third chronic disease among young females in Western societies. These disorders, which are characterized by irregular eating habits and severe distress or concern about body weight or shape, are closely related to poor quality of life, situations of obesity or malnutrition and high rates of psychosocial morbidity and premature mortality. These ED are complex, multifactorial diseases. The major susceptibility factors involved are sociocultural reasons, family environment, psychological traits and biological factors, mainly endocrine and genetic features. Yes, the existence of genetic predisposition in ED is widely accepted nowadays and it has been confirmed in twin studies designed to differentiate between the influence of genetics and that of the environment. In addition, many psychological traits that are often coupled with ED, have also been shown to display a certain degree of heritability.

The implications of genetics in ED and their psychopathological associated features are being heavily researched. In the last years, genome-wide association studies (GWAS) have flourished as designs capable of searching the whole genome for thousands of mutations. However, GWAS aimed to identify ED genetic determinants have not yielded conclusive results so far; there are some suggestive findings but few genes stand out across the different studies published.

The Link With Obesity

The main pathways involved in eating behavior are central neurotransmission, especially serotonin and dopamine; reward-related pathways, e.g. the cannabinoid endogenous system; and the central regulation of food intake. With regard to the latter, it is surprising that the majority of known genes involved in the central regulation of weight, such as those recently evidenced in obese individuals, have not yet been thoroughly explored in the ED setting. The existence of this gap in the research of the ED is especially surprising because one could easily make a compelling case for the involvement of obesity-related mutations in ED such as bulimia, binge-eating disorder or even the anorexia binge-purge subtype, where emotional and uncontrolled eating are also present.

Obesity is generally considered as a medical illness with metabolic and genetic origins, whereas ED has traditionally been regarded as Western culture-bound syndrome best treated by psychological interventions. However, many risk factors and personality dimensions (e.g. body dissatisfaction, low self-esteem, anxiety, depression, traits mediating substance abuse, dieting, binge-eating, a history of sexual and/or physical abuse, etc.) are shared between obese and ED patients. Even some common linkage peaks for obesity and bulimia have been found in certain chromosomes. Following this rationale, which argues for an intimate connection between ED and obesity, we can expect that alterations in genes involved in the regulation of eating behavior that may lead to obesity, could also help develop aberrant conduct patterns that favor the establishment of ED, should other susceptibility factors or personality dimensions were present. This has been our working hypothesis for a series of papers we have published. In them, we analyzed genes in the central nervous system whose genetic variability has been consistently associated with obesity are insufficiently tested in the field of ED. 

One such gene is the brain-derived neurotrophic factor (BDNF), which is essential for body weight control and energy homeostasis, as well as for the regulation of neurotransmitters systems that may be altered in psychiatric disorders. In fact, mutations in this gene have been related to hyperphagia, obesity, and several psychiatric illnesses. We hypothesized that changes in BDNF could affect personality features, which could, in turn, draw an already vulnerable patient (e.g. by familiar or environmental factors) to an ED or maybe aggravate associated psychopathological symptoms. We performed a combined analysis of several mutations in this gene and identified some combinations of these variants (haplotypes) which showed a profound effect on general psychopathological symptoms of both anorexia and bulimia patients. However, psychological traits more intimately related to ED were largely unaffected in our patients. This suggests that variability in the BDNF gene may contribute to symptoms that are commonly found in ED patients but that could be more indicative of comorbid disorders (depression, anxiety, etc.).

The expression of BDNF in the brain is controlled by the melanocortin 4 receptor (MC4R), which suggests that BDNF could be the effector through which MC4R controls energy balance. The importance of the latter in the regulation of weight is such that the presence of mutations in MC4R constitutes today the most common monogenic cause of human obesity. These mutations have also been associated with disinhibition and emotional eating. With this background, we carried out the sequencing of the MC4R gene to determine whether some of the mutations could also be present in binge-eating patients who were not obese. However, we found no evidence of such association in our population.

Another interesting gene in the central nervous system is the Neuronal Growth Regulator 1 (NEGR1), whose role is still unclear but that has been pinpointed in several studies as a gene linked to obesity and able to influence motivation and personality. We performed a combined association study of 21 relevant mutations in NEGR1 in relation to personality dimensions showed by ED patients. Interestingly, these analyses identified a central region of the gene whose variability strongly correlated with worse symptomatology of bulimia patients in several personality traits such as a drive for thinness, ineffectiveness and interoceptive awareness.

The last of the studies we have published on the influence of genetics in ED-related behavior is an analysis of the variability of two genes, TFAP2B and KCTD15, and the assessment of its combined influence on personality dimensions in anorexia and bulimia. These two genes interact in the brain to regulate feeding behavior and have also been related to obesity in the past. Epistasis analyses (the study of the combinations of mutations in different genes that work together) showed that there were indeed relevant interactions of some mutation pairs with the body mass index and also with the psychopathological symptoms showed by females with ED.

We believe these data are just the tip of the iceberg, as there are still many more central genes known to be related to obesity that need to be comprehensively tested in the ED setting. Some examples are the Fat Mass and Obesity Associated Gene (FTO), Src-Homology 2 Domain-Containing Protein B1 (SH2B1), Transmembrane Protein 18 (TMEM18), Glucosamine-6-phosphate Deaminase 2 (GNPDA2), Mitochondrial Carrier Homolog 2 (MTHC2) or the ETV5 Transcription Factor.

Conclusion And Future Perspectives

A common finding we have observed throughout our research is that mutations in obesity-related genes are more linked to pathological behaviors of patients with bulimia than those with anorexia. Indeed, we have seen that these obesity-related mutations are also related to impulsiveness, eating a great amount of food, increases in hunger score, snacking, etc., and these are of course behaviors that one could more easily relate to bulimia or binge-eating than anorexia. Paradoxically, however, the vast majority of available genetic association studies on the regulation of appetite and energy homeostasis have been carried out in anorexia patients. Therefore, there is, in general, a need for more studies in bulimia or binge-eating patients that can examine these central pathways. Future efforts should be put into collecting large samples from these diagnosis groups, as it is presently happening with anorexia thanks to the activities of different consortiums.

On the other hand, a shared problem of genetic studies on ED patients is that an increase in sample size is usually accompanied with a decrease of homogeneity; either from the point of view of diagnosis (there are significant differences, for instance, regarding the definition of binge-eating behavior across studies) or in the light of the many different tools utilized in the assessment of personality dimensions. The fact that these data are collected mostly through personal questionnaires makes difficult to compare information on patients that are interviewed by clinicians in different countries with different idiosyncrasies.

Another concern we should bear in mind is the extrapolations that are made from GWAS findings in ED patients. Even though GWAS are powerful tools that allow the study of a great amount of genes and mutations, it is imperative to move from GWAS signals to biological understanding.

In summary, we know that some of the mechanisms underlying the different behaviors displayed by patients with psychiatric disorders, including ED, are mediated by the contribution of genes of psychological traits. A better knowledge of these interactions, especially those of genes that are already known to be involved in obesity, will help elucidate the still obscure central pathways behind the development and course of an ED.

This article is a compilation of the articles below, with contribution from Guillermo Gervasini.