Objective To conduct a genome-wide association study (GWAS) of anorexia nervosa and to calculate genetic correlations with a series of psychiatric, educational, and metabolic phenotypes. Method Following uniform quality control and imputation using the 1000 Genomes Project (phase 3) in 12 case-control cohorts comprising 3,495 anorexia nervosa cases and 10,982 controls, we performed standard association analysis followed by a meta-analysis across cohorts. Linkage disequilibrium score regression (LDSC) was used to calculate genome-wide common variant heritability [ hSNP2, partitioned heritability, and genetic correlations (rg)] between anorexia nervosa and other phenotypes. Results Results were obtained for 10,641,224 single nucleotide polymorphisms (SNPs) and insertion-deletion variants with minor allele frequency > 1% and imputation quality scores > 0.6. The hSNP2 of anorexia nervosa was 0.20 (SE=0.02), suggesting that a substantial fraction of the twin-based heritability arises from common genetic variation. We identified one genome-wide significant locus on chromosome 12 (rs4622308, p=4.3×10−9) in a region harboring a previously reported type 1 diabetes and autoimmune disorder locus. Significant positive genetic correlations were observed between anorexia nervosa and schizophrenia, neuroticism, educational attainment, and high density lipoprotein (HDL) cholesterol, and significant negative genetic correlations between anorexia nervosa and body mass index, insulin, glucose, and lipid phenotypes. Conclusions Anorexia nervosa is a complex heritable phenotype for which we have found the first genome-wide significant locus. Anorexia nervosa also has large and significant genetic correlations with both psychiatric phenotypes and metabolic traits. Our results encourage a reconceptualization of this frequently lethal disorder as one with both psychiatric and metabolic etiology.

The maintenance of normal body weight is disrupted in patients with anorexia nervosa (AN) for prolonged periods of time. Prior to the onset of AN, premorbid body mass index (BMI) spans the entire range from underweight to obese. After recovery, patients have reduced rates of overweight and obesity. As such, loci involved in body weight regulation may also be relevant for AN and vice versa. Our primary analysis comprised a cross-trait analysis of the 1000 single nucleotide polymorphisms (SNPs) with the lowest p-values in a genome-wide association meta-analysis (GWAMA) of AN (GCAN) for evidence of association in the largest published GWAMA for BMI (GIANT). Subsequently we performed sex-stratified analyses for these 1000 SNPs. Functional ex vivo studies on four genes ensued. Lastly, a look-up of GWAMA-derived BMI related loci was performed in the AN GWAMA. We detected significant associations (p-values < 5×10−5, Bonferroni corrected p < 0.05) for 9 SNP alleles at 3 independent loci. Interestingly, all AN susceptibility alleles were consistently associated with increased BMI. None of the genes (chr. 10: CTBP2, chr. 19: CCNE1, chr. 2: CARF and NBEAL1; the latter is a region with high linkage disequilibrium) nearest to these SNPs has previously been associated with AN or obesity. Sex-stratified analyses revealed that the strongest BMI signal originated predominantly from females (chr. 10 rs1561589; poverall: 2.47 × 10−06/pfemales: 3.45 × 10−07/pmales: 0.043). Functional ex vivo studies in mice revealed reduced hypothalamic expression of Ctbp2 and Nbeal1 after fasting. Hypothalamic expression of Ctbp2 was increased in diet induced obese (DIO) mice as compared to age-matched lean controls. We observed no evidence for associations for the look-up of BMI related loci in the AN GWAMA. A cross-trait analysis of AN and BMI loci revealed variants at three chromosomal loci with potential joint impact. The chromosome 10 locus is particularly promising given that the association with obesity was primarily driven by females. In addition, the detected altered hypothalamic expression patterns of Ctbp2 and Nbeal1 as a result of fasting and DIO implicate these genes in weight regulation.

Disturbances of serotonergic neurotransmission and temperamental vulnerability have both been implicated in the pathogenesis of anorexia nervosa (AN). We genotyped the –1438 A/G polymorphism in the 5-HT2A receptor gene and serotonin transporter linked-polymorphic region (5-HTTLPR) in 132 adolescent subjects with AN and in 93 healthy controls. Personality dimensions in AN patients were assessed with the Temperament and Character Inventory. In a case-control model, we tested the hypothesis that these genetic variants confer susceptibility to AN. We also analyzed whether two polymorphisms show association with temperamental and character traits. No significant difference was found in the 5-HTTLPR frequency between AN patients and controls; however, there was a statistical trend towards a higher frequency of the A allele of the –1438 A/G polymorphism in patients than in controls (64.9 vs. 56.7%, χ² test, p = 0.08). We also found a significant association between the A allele of this polymorphism and two temperamental traits. Patients homozygous for the A allele showed lower reward dependence than G/G homozygotes, and A/A homozygotes showed lower harm avoidance than heterozygotes. Low reward dependence and harm avoidance were more characteristic of the restrictive-type AN than of other subtypes of the disorder. No association of 5-HTTLPR with personality dimensions in AN patients was observed. Our results may suggest that the A allele of the –1438 A/G polymorphism confers some genetic risk for adolescent AN patients, especially in those with personality traits, which are typical of the restrictive-type AN.

Objectives: Several lines of evidence suggest that brain-derived neurotrophic factor (BDNF) plays an important role in weight regulation and eating behaviors as well as in the activity-dependent neuroplasticity underlying learning and memory behaviors involving the hippocampus. In anorexia nervosa (AN) patients, abnormal serum BDNF concentrations, cognitive impairments and specific personality traits have been traditionally observed. This study explores the levels of four serum neurotrophins [BDNF, neurotrophin 3 (NTF3), neurotrophin 4 (NTF4) and glial cell line-derived neurotrophic factor (GDNF)] with respect to their use as potential biomarkers for AN. This study also investigates any associations that might exist between serum neurotrophin levels and neurocognitive impairment or personality traits. Methods: Serum neurotrophin concentrations were measured in 60 AN patients (AN group) and 45 healthy controls (HC group). We correlated the serum levels of the four neurotrophins BDNF, NTF3, NTF4 and GDNF and the clinical type of anorexia. We also analyzed the relationship between serum neurotrophin levels and the Beck Depression Inventory, body mass index, executive functions by the Wisconsin Card Sorting test (WCST) and personality dimensions by the Temperament and Character Inventory (TCI) test. Results: Serum NTF4 concentrations were significantly lower when comparing all AN patients (34.7 ± 72.5 pg/ml) or restriction type AN patients (29.1 ± 62.5 pg/ml) with the HC group (58.4 ± 135.8 pg/ml; p = 0.004 and p = 0.005, respectively). A significant correlation (p < 0.005) between BDNF serum levels and patient personality dimensions as measured by the TCI test was observed. Furthermore, significant correlations were observed between NTF4 and GDNF serum levels and executive function as measured by the WCST. Conclusions: These data suggest that NTF4 might serve as a biomarker for AN. Furthermore, BDNF and GDNF serum levels appear to be associated with personality traits and executive function.

We conducted a genome-wide association study (GWAS) of anorexia nervosa (AN) using a stringently defined phenotype. Analysis of phenotypic variability led to the identification of a specific genetic risk factor that approached genome-wide significance (rs929626 in EBF1 (Early B-Cell Factor 1); P = 2.04 × 10−7; OR = 0.7; 95% confidence interval (CI) = 0.61–0.8) with independent replication (P = 0.04), suggesting a variant-mediated dysregulation of leptin signaling may play a role in AN. Multiple SNPs in LD with the variant support the nominal association. This demonstrates that although the clinical and etiologic heterogeneity of AN is universally recognized, further careful sub-typing of cases may provide more precise genomic signals. In this study, through a refinement of the phenotype spectrum of AN, we present a replicable GWAS signal that is nominally associated with AN, highlighting a potentially important candidate locus for further investigation.