Anorexia Nervosa, Susceptibility To

Description

Eating disorders are characterized by severe disturbances in eating behavior that typically have onset during late adolescence and early adulthood. Three major types are recognized: anorexia nervosa (AN), bulimia nervosa (BN; 607499), and eating disorder not otherwise specified (EDNOS). AN is characterized by obsessive fear of weight gain, severely restricted eating, and low body weight. In women, AN has the highest mortality among the psychiatric disorders (Sullivan, 1995). AN is divided into 2 clinical subtypes, restricting anorexia nervosa (RAN) and binge-eating/purging anorexia nervosa (BPAN). BN can occur at any body weight and is characterized by binge-eating and compensatory weight-loss behaviors. Family studies have indicated an increased prevalence of eating disorders in relatives of probands with AN (Lilenfeld et al., 1998), and twin studies (Holland et al., 1984; Wade et al., 2000) have estimated concordance rates for monozygotic twins with AN to be 52 to 56%, whereas concordance rates for dizygotic twins with AN have been estimated to be 5 to 11% (Grice et al., 2002).

Mapping

ANON1 on Chromosome 1p

Grice et al. (2002) performed a genomewide linkage analysis of 192 families with at least 1 affected relative pair with AN and related eating disorders, including BN. Only modest evidence for linkage was found, with the highest nonparametric linkage (NPL) score, 1.80, at marker D4S2367 on chromosome 4. Since the reduction of sample heterogeneity would increase power to detect linkage, the authors performed linkage analysis in a subset of 37 families in which at least 2 affected relatives had diagnoses of RAN, the clinical subtype characterized by severe limitation of food intake without binge-eating or purging behavior. When they limited the linkage analysis to this clinically homogeneous subgroup, the highest multipoint NPL score observed was 3.03 at marker D1S3721 on chromosome 1p. The genotyping of additional markers in this region led to a peak multipoint NPL score of 3.45, thereby providing suggestive evidence for the presence of an AN susceptibility locus, designated ANON, on 1p.

Heterogeneity

Devlin et al. (2002) measured psychiatric, personality, and temperament phenotypes of individuals diagnosed with eating disorders from 196 multiplex families and genotyped these individuals at 387 short tandem repeat markers distributed across the genome. A multipoint affected sib pair linkage analysis was employed which incorporated covariates. Two variables were identified (drive-for-thinness and obsessionality) which delimited populations among the affected sib pairs. By incorporating these covariates into the affected sib pair linkage analysis, several regions of suggestive linkage were identified: one close to genomewide significance on chromosome 1 (at 210 cM, D1S1660; lod = 3.46, P = 0.00003), another on chromosome 2 (at 114 cM, D2S1790; lod = 2.22, P = 0.00070) and a third region on chromosome 13 (at 26 cM, D13S894; lod = 2.50, P = 0.00035).

Bacanu et al. (2005) measured over 100 attributes thought to be related to liability to eating disorders in affected individuals of multiplex families from the AN cohort that had been previously studied by Grice et al. (2002) and Devlin et al. (2002) and the BN cohort that had been previously studied by Bulik et al. (2003). Six traits were selected for linkage analysis on the basis of expert evaluation and statistical analysis, including obsessionality, age at menarche, anxiety for quantitative trait locus (QTL) linkage analysis, lifetime minimum body mass index (BMI), concern over mistakes, and food-related obsessions. Using QTL linkage analysis, the BN cohort produced 4 suggestive signals: for age at menarche at 10p13 and for anxiety for QTL linkage analysis at 1q31.1, 4q35.2, and 8q13.1. Using covariate-based linkage analysis, the BN cohort showed the most significant and suggestive linkages: for minimum BMI, 1 significant linkage at 4q21.1 and 3 suggestive linkages at 3p23, 10p13, and 5p15.3; for concern over mistakes, 2 significant linkages at 16p13.3 and 14q21.1 and 3 suggestive linkages at 4p15.33, 8q11.23, and 10p11.21; and for food-related obsessions, 1 significant linkage at 14q21.1 and 5 suggestive linkages at 4p16.1, 10p13.1, 8q11.23, 16p13.3, and 18p11.31. For the AN cohort, results were less compelling. Using QTL linkage analysis, they found 2 suggestive linkages, for obsessionality at 6q21 and for anxiety for QTL linkage analysis at 9p21.3. Covariate-based linkage analysis of the AN cohort revealed 5 suggestive signals: for minimum BMI at 4q13.1, for concern over mistakes at 11p11.2 and 17q25.1, and for food-related obsessions at 17q25.1 and 15q26.2. There was minimum overlap between the 2 cohorts for substantial linkage signals.

Molecular Genetics

Associations Pending Confirmation

Urwin and Nunn (2005) hypothesized that genes encoding proteins which clear serotonin (5-HT) and norepinephrine from the synapse are prime candidates for affecting susceptibility to anorexia nervosa. Urwin and Nunn (2005) provided evidence that supports an interaction between the MAOA-uVNTR gene variant (309850.0002) and the 5-HTTLPR gene variant (182138.0001) in 114 anorexia nervosa nuclear families (patient with AN plus biologic parents). A 5-HTTLPR S/S genotype with no apparent individual effect on risk and known to be associated with anxiety was preferentially transmitted to children with AN and restricting anorexia nervosa (AN-R) alone, when the 'more active' MAOA-L gene variant was also transmitted. The increased risk of developing the disorder was up to 8 times greater than the risk imposed by the MAOA-L gene variant alone. Urwin and Nunn (2005) considered this an example of synergistic epistatic interaction.

Frisch et al. (2001) found an association between anorexia nervosa and the COMT val158 allele (V158M; 116790.0001) in a family-based study of 51 Israeli-Jewish AN trios. Gabrovsek et al. (2004) could not replicate this finding in a combined sample of 372 European AN families, suggesting that the findings of Frisch et al. (2001) were specific to a particular population and that val158 is in linkage disequilibrium with other molecular variations in the COMT gene, or its vicinity, which were the direct cause of genetic susceptibility to anorexia nervosa. Michaelovsky et al. (2005) studied 85 Israeli-Jewish AN trios, including the original sample of Frisch et al. (2001), comprising 66 anorexia nervosa restricting and 19 binge-eating/purging patients. They performed a family-based transmission disequilibrium test (TDT) for 7 SNPs in the COMT-ARVCF (602269) region including the V158M polymorphism. TDT analysis of 5-SNP haplotypes in the AN-R group revealed overall statistically significant transmission disequilibrium for 'haplotype B' (COMT 186C, 408G, 472G [val158] and ARVCF 659C[pro220] and 524T[val175]) (P less than 0.001), while 'haplotype A' (COMT 186T, 408C, 472A[met158] and ARVCF 659T[leu220] and 524C[ala175]) was preferentially not transmitted (P = 0.01). Haplotype B was associated with increased risk (RR of 3.38), while haplotype A exhibited a protective effect (RR of 0.40) for AN-R. Preferential transmission of the risk alleles and haplotypes from parents was mostly contributed by fathers.

Monteleone and Maj (2008) reviewed the genetics of eating disorders, including anorexia nervosa, and stated that there was no convincing evidence for association of candidate genes with eating disorders; they noted that the heterogeneity of eating disorder phenotypes was most likely responsible for the contradictory and inconclusive results.

For discussion of an association between susceptibility to anorexia nervosa (formerly designated ANON2) and polymorphism in the BDNF gene, see 113505.0002.