Genetic Aspects of Eating Disorders Research Paper

Although intuitively appealing, evidence for a central etiological role of sociocultural factors is hardly persuasive. Exposure to thin ideals and dieting are nearly universal in industrialized countries; however, only a relatively small number of young women actually succumb to clinically significant eating disorders. The recent surge in research regarding the contribution of genetic effects to liability to eating disorders offers a more powerful strategy for addressing the perennial question of differential susceptibility to the development of eating disorders in the presence of near universal exposure to a powerful societal risk factor.

This research paper considers how the application of recent advances in family and twin study methodology, as well as quantitative and molecular genetics, offer a firm new scientific footing for pursuing more refined searches for genetic and environmental factors that contribute to the etiology of eating disorders.

2. Components Of Genetic Epidemiology

There are three major research designs in genetic epidemiology that allow for the delineation and quantification of the relative contribution of genes and environment to complex behavioral traits. The first step is to determine whether the disorder aggregates in families. This hypothesis is addressed by the classic controlled family study which determines whether there is a statistically greater lifetime risk of eating disorders in blood relatives of individuals who have an eating disorder in comparison to relatives of individuals without eating disorders. If no differential risk is observed, then the probability of the disorder being genetically influenced is low. However, because genetic and environmental effects are confounded in the family study design, twin and adoption studies are needed to determine the extent to which the observed familial resemblance for illness is due to genetic or environmental effects. If genetic effects appear to be important in the transmission of the disorder, then linkage and association studies are next employed to determine the precise location, identity, and function of the genes that are implicated. Each of these approaches (with the exception of adoption studies) has been employed in the study of eating disorders.

2.1 Family Studies Of Eating Disorders

There are now a series of family studies addressing the fundamental question of ‘are eating disorders familial?’ Most (Gershon et al. 1983, Hudson et al. 1987, Kassett et al. 1989, Lilenfeld et al. 1998, Strober et al. 1990, 2000) but not all (Gershon et al. 1983, Halmi et al. 1991, Stern 1992) controlled family studies, have found a significantly greater lifetime prevalence of eating disorders among relatives of eating disordered individuals in comparison to relatives of controls. Intriguingly, several studies have found increased rates of both AN and BN (i.e., co-aggregation) in relatives of individuals with AN as well as individuals with BN, compared to rates among relatives of controls (Gershon et al. 1983, Hudson et al. 1987, Kassett et al. 1989, Strober et al. 1990, 2000), suggesting that AN and BN share transmissible risk factors. Moreover, relatives of individuals with AN and BN have also been found to have a significantly increased rate of subthreshold eating disorders compared to relatives of controls (Lilenfeld et al. 1998, Strober et al. 2000), suggesting that the eating disorders do not ‘breed true,’ but are expressed in families as a broad spectrum of eating-related pathology.

On balance, evidence from family study data suggests there is an elevation in the lifetime prevalence of eating disorders among the relatives of people with eating disorders. Moreover, suggestive evidence of the co-aggregation in families of AN, BN, and milder eating disturbances raises the provocative possibility of shared etiologic factors across these conditions.

2.2 Twin Studies

The goal of the classical twin study is to qualify and quantify similarities and differences between monozygotic (MZ) and dizygotic (DZ) twin pairs in order to identify and delineate genetic and environmental causes for a particular trait. Given that MZ twins, for most intents and purposes, share all of their genes, and DZ twins share, on average, half of their genes, any excess concordance in MZ twins over DZ twins suggests a genetic contribution to liability to the disorder. Conversely, any differences between MZ twins provides strong evidence for the role of environmental influences (Plomin et al. 1994, pp. 171–2), whereas differences between DZ twins can result from genetic and/or environmental effects. More complicated statistical modeling allows one to parse the variance in liability to ill ness into three sources: additive genetic effects (a²), shared environment al effects (c²), and unique environmental effects (e²). Additive genetic effects reflect the cumulative impact on a trait of many individual genes each of which has a relatively small individual effect on the behavioral phenotype. The presence of a² is inferred when the correlation between MZ twins is greater than the correlation between DZ twins. By contrast, common environmental effects reflect etiological influences to which both members of a twin pair are exposed, regardless of zygosity. Examples of such effects include the social class and religious preference of the family of origin. Unique environmental effects, on the other hand, result from etiological influences to which one member of a twin pair is exposed but not the other, and contribute to differences between members of a twin pair. Examples include one member of a twin pair being exposed to a traumatic experience not shared with the co-twin.

One of the key caveats to the twin study is the assumption of equal environments (EEA) which posits that MZ and DZ twins are equally correlated for their exposure to environmental influences that are of etiologic relevance to the trait under study (Plomin et al. 1994). If the EEA is incorrect, then the greater resemblance of MZ twins in comparison to DZ twins could actually be due to environmental factors. A violation of the EEA does not necessarily invalidate the results, but may influence the magnitude of the estimated genetic and environmental components. Studies of the EEA with regard to eating disorders suggest that this assumption has not been violated in twin studies (Bulik et al. 2000, Kendler et al. 1993, Klump et al. 2000, Sullivan et al. 1998).

2.3 Twin Studies Of Anorexia Nervosa

Beyond isolated case reports, the first systematic study of clinically ascertained twins with AN (Holland et al. 1984, 1988, Treasure and Holland 1989) found that the concordance for MZ twins was substantially greater than for DZ twins. Reanalyses of these data (assuming a population prevalence of AN of 0.75 percent) revealed evidence of familial aggregation with parameter estimates of 88 percent for a², 0 percent for c², and 12 percent for e². In short, the observed familial aggregation for AN appears to be influenced most strongly by additive genetic effects; however, the estimates were rather imprecise given the small sample size.

Population-based studies of anorexia nervosa are difficult to conduct given the relatively low prevalence of the disorder. There has been one published study that has been able to successfully apply twin modeling techniques. Wade et al. (2000) derived heritability estimates for AN in the context of studying the nature of the comorbid relationship between AN and major depression. The heritability of AN was estimated to be 58 percent, although the authors could not rule out a contribution of shared environment to the liability to AN. On balance, we can conclude from these twin and family studies (see Lilenfeld et al. 1997 for a review) that AN is familial. However, the definitive resolution of the independent contribution of genetic and shared environmental factors to the observed familiality of AN will require more ambitious collaborative efforts to obtain larger sample sizes with sufficient statistical power.

2.4 Twin Studies Of Bulimia Nervosa

Twin studies of BN have been more successful given the higher population prevalence of the disorder. Initial case series of twins with BN revealed consistently greater concordance for BN in MZ than DZ twin pairs (Fichter and Noegel 1990, Hsu et al. 1990, Treasure and Holland 1989). Pooling data from these case series for twin modeling and assuming a population prevalence of BN of 2.5 percent revealed evidence of familial aggregation with 47 percent of the variance accounted for by additive genetic effects, 30 percent by shared environmental effects, and 23 percent by unique environmental effects. However, the sample sizes were small and the estimates imprecise.

Population-based studies of BN have been conducted in the US (Bulik et al. 1998, Kendler et al. 1991) and Australia (Wade et al. in press). The studies that have estimated the heritability of BN based on a single occasion of measurement suggest a moderate contribution of additive genetic effects, a negligible contribution of shared environmental effects, and a more substantial contribution of unique environmental effects to liability to BN. Although a marked improvement over the clinical series, these studies still had limited statistical power given the relatively low population prevalence of BN and given that the reliability of the diagnosis of BN tends to be poor , which can lead to underestimation of both a² and c² (Bulik et al. 1998).

Two studies have augmented statistical power by incorporating more than one occasion of measurement into the twin model (Bulik et al. 1998, Wade et al. in press). This approach controls for unreliability of diagnosis, increases power to detect both a² and c², and therefore provides the most reliable information regarding the nature and magnitude of genetic and environmental contributions to BN. In short, these two studies reveal a markedly greater contribution of additive genetic effects to the liability to BN (59 and 83 percent, respectively), a negligible contribution of shared environment (0 percent in both studies), and a moderate contribution of unique environmental effects (41 and 17 percent). Although the parameter estimates for c² were 0, the confidence intervals did not completely rule out a contribution of shared environment. The results from these two studies confirm the central role of genes in the observed familiality of BN.

In summary, from twin and family studies, we can conclude that BN is familial and that there appears to be a moderate to substantial contribution made by genetic factors and unique environmental factors to liability to the disorder. The contribution of shared environment is less certain, but appears to be of lesser prominence than the effect of genes and unique environment. A reasonable next step for twin studies is to determine the precise nature of the unique environmental effects that increase risk for developing BN.

2.5 Molecular Genetic Studies

Several efforts have been made to identify the map position of individual genes conferring susceptibility to AN. To date, the focus has been on genes involved in serotonergic (5HT) neurotransmission. Collier et al. (1997) conducted an association study of 81 women with AN and 226 controls, finding a modest increase in the frequency of the 5HT2A -1438/A allele in AN compared to controls. These results have since been replicated by Sorbi et al. (1998) and by Enoch et al. (1998). Interestingly, in the Sorbi et al. study the increased frequency of this allelic variant was seen only in the subset of nonpurging restricting anorexics, whereas in the Enoch et al. study this genotype was not found to discriminate BN patients from controls, yet was also more common among patients with obsessive compulsive disease.

Two studies have failed to replicate the 5HT2A finding. Campbell et al. (1998) studied 157 anorexic patients and 150 controls, finding no association of this polymorphism with AN; however, stratification by the presence vs. absence of purging was not performed and the frequency of the 1438A genotype in their controls was higher than in other reported controls samples. Still, Hinney et al. (1998) studied 100 anorexics, with 100 underweight but nonanorexic controls, and 100 obese controls and found no intergroup differences in the frequency of this genotype. Ongoing international collaborative studies are continuing the search for susceptibility genes for both AN and BN (Kaye et al. 2000).

3. Summary

An increasing body of evidence supports the notion that AN and BN have a strong familial component, whereas twin studies suggest a substantial contribution of additive genetic effects to the observed familiality of both disorders. By contrast, the magnitude of the contribution of shared environmental effects remains unresolved. Moreover, both family and twin data suggest there is at least a moderate familial–genetic correlation between AN and BN, which may both reflect and explain the common presence of weight phobia in both disorders and the not infrequent development of binge eating during the natural course of AN.

That the contribution of additive genetic effects appears to be more prominent than shared environmental effects for both AN and BN is surprising given the historical focus on cultural and social factors in the study of eating disorders. There are several plausible explanations. First, environmental experiences that are shared in common by members of a family may have little bearing on risk to eating disorders. Thus, environmental factors that do influence risk may be those experiences that are unique to the individual, possibly acting to increase the likelihood of body dissatisfaction, or initiating weight control practices. Second, the statistical power of twin data to detect common environmental effects in the presence of genetic influences may be inadequate, although we note once again that the studies with greater statistical power support a greater contribution of genetic than shared environmental effects. Third, if familial factors exert effects in concert with an individual’s genetic propensities, the main effects of familial environment will be weak but effects operating in gene–environment interaction might be considerable. Finally, the effects of shared environment may be more robust in childhood, but diminish with age (Klump et al. 2000).

An obvious critical question that ensues from these findings is the nature of a purported inherited diathesis, as it is improbable that specific genes exist that code for body weight phobia or dietary restraint. Indeed, critical to future studies of etiology will be efforts at refining the phenotype or phenotype spectrum, identifying those components of the phenotype that are most heritable, and, through co-variation with symptoms of the disorder, are the most likely candidates for mediation of genetic risk. Incorporating a focus on certain quantitative temperamental or personality traits in future genetic studies of eating disorders may be of great promise in this regard. For example, restricting-type AN in particular is characterized by a remarkably consistent cluster of heritable behavioral traits, including emotional restraint, avoidance of novelty, anxious worry and self-doubt, compliancy, perfectionism, and perseverance in the face of nonreward (Strober 1995). The current focus on candidate genes regulating serotonergic neurotransmission as a risk factor for AN derives, intuitively and empirically, on the one hand from preclinical data linking serotonergic neurotransmission to restraint of reward motivation for exploring novel environments, inhibitory modulation of feeding and sexual behavior, and enhanced sensitivity of neurobehavioral systems to stimulus events (Kaye et al. 1993), and on the other to evidence adduced from clinical studies (Kaye et al. 1991) of impaired regulation of serotonergic activity in AN patients well after restoration to normal weight levels. Thus, a convergence of preclinical and clinical data implicate abnormalities of serotonergically mediated brain and behavioral systems in the pathogenesis of AN by conferring extreme propensities toward behavioral rigidity, obsessiveness, perfectionism, and constraint. However, the contribution to risk of abnormalities in other brain substrates cannot be discounted.

Although the presentation of BN is more variable than that of restricting-type AN, a role for heritable personality and behavioral traits as indices of a biologically mediated etiologic mechanism in the development of this disorder is also heuristically and theoretically defensible. In contrast to AN, women with BN more often display traits of thrill seeking and excitability, and are more dysphoric in response to rejection or nonreward (Strober 1995). These features may predispose to periodic lapses in control in eating and other realms under conditions of stress. In this same vein, possible heritable variations in the sensitivity of brain emotional systems to the reward and motivational properties of feeding behavior may constitute a diathesis for the development and reinforcement of dysregulated appetitive behavior at times of emotional despair or life stress. These models of ‘genetic’ risk are easily reconciled with prevailing psychological theories of BN which view binge eating as a compensatory modulation of dysphoric states linked to deeply entrenched negative self-schemas.

In closing, we believe that the search for susceptibility genes in eating disorders is an exciting and valid research agenda. It may well hold forth the promise of bringing resolution to the continuing uncertainty of whether AN and BN are truly separate entities, or have risk and vulnerability factors in common. Given the body of research reviewed in this research paper, future empirical investigation bridging the behavioral and biological sciences and exploiting the rapidly advancing science of molecular and family genetic epidemiology may yield keys to refining our knowledge of the etiology, nosology, and therapeutics of eating disorders.

Bibliography:

1. Bulik C M, Sullivan P F, Kendler K S 1998 Heritability of binge-eating and broadly defined bulimia nervosa. Biological Psychiatry 44: 1210–8
2. Bulik C M, Sullivan P F, Wade T D, Kendler K S 2000 Twin studies of eating disorders: A review. International Journal of Eating Disorders 27: 1–20
3. Campbell D A, Sundaramurthy D, Markham A F, Pieri L F 1998 Lack of association between 5-HT2A gene promoter polymorphism and susceptibility to anorexia nervosa. Lancet 351: 499
4. Collier D A, Arranz M J, Li T, Mupita D, Brown N, Treasure J 1997 Association between 5-HT2A gene promoter polymorphism and anorexia nervosa. Lancet 350: 412
5. Enoch M A, Kaye W H, Rotondo A, Greenberg B D, Murphy D L, Goldman D 1998 5-HT2A promoter polymorphism– 1438G A, anorexia nervosa, and obsessive-compulsive disorder. Lancet 351: 1785–6
6. Fichter M M, Noegel R 1990 Concordance for bulimia nervosa in twins. International Journal Eating Disorders 9: 255–63
7. Gershon E, Schreiber J, Hamovit J, Dibble E D, Kaye W H, Nurnberger J I, Andersen A, Ebert M 1983 Anorexia nervosa and major affective disorders associated in families: A preliminary report. In: Guze S B, Earls F G, Barrett J E (eds.) Childhood Psychopathology and Development. Raven Press, New York, pp. 279–84
8. Halmi K, Eckert E, Marchi P, Sampugnaro V, Apple R, Cohen J 1991 Comorbidity of psychiatric diagnoses in anorexia nervosa. Archives of General Psychiatry 48: 712–18
9. Hinney A, Ziegler A, Nothen M, Remschmidt H, Hebebrand J 1998 5-HT2a receptor gene polymorphisms, anorexia nervosa and obesity. Lancet 350: 1724–5
10. Holland A J, Hall A, Murray R, Russell G F M, Crisp A H 1984 Anorexia nervosa: A study of 34 twin pairs and one set of triplets. British Journal of Psychiatry 145: 414–19
11. Holland A, Sicotte N, Treasure J 1988 Anorexia nervosa: Evidence for a genetic basis. Journal of Psychosomatic Research 32: 561–71
12. Hsu G L K, Chesler B E, Santhouse R 1990 Bulimia Nervosa in eleven sets of twins: A clinical report. International Journal of Eating Disorders 9: 275–82
13. Hudson J I, Pope H G, Jonas J M, Yurgelun-Todd D, Frankenburg F R 1987 A controlled family history study of bulimia. Psychological Medicine 17: 883–90
14. Kassett J, Gershon E, Maxwell M, Guroff J J, Kazuba D M, Smith A L, Brandt H A, Jimerson D C 1989 Psychiatric disorders in the first-degree relatives of probands with bulimia nervosa. American Journal of Psychiatry 146: 1468–71
15. Kaye W H, Lilenfeld L, Berretini W, Strober M, Devlin B, Klump K, Goldman D, Bulik C M, Halmi K A, Fichter M, Kaplan A, Woodside B, Treasure J, Plotnikov K, Pollice C, Rao R, McConaha C 2000 A genome-wide search for susceptibility loci in anorexia nervosa: Methods and sample description. Biological Psychiatry 47: 794–803
16. Kaye W H, Weltzin T E, Hsu L K G 1993 Relationship between anorexia nervosa and obsessive-compulsive behaviors. Psychiatric Annals 23: 365–73
17. Kaye W H, Gwirtsman H E, George D T, Ebert M H 1991 Altered serotonin activity in anorexia nervosa after long-term weight restoration. Archives of General Psychiatry 48: 556–62
18. Kendler K S, MacLean C, Neale M C, Kessler R C, Heath A C, Eaves L J 1991 The genetic epidemiology of bulimia nervosa. American Journal of Psychiatry 148: 1627–37
19. Kendler K S, Neale M C, Kessler R C, Heath A C, Eaves L J 1993 A test of the equal environment assumption in twin studies of psychiatric illness. Behavioral Genetics 23: 21–7
20. Klump K L, Holly A, Iacono W, McGue M, Willson L 2000 Physical similarity and twin resemblance for eating attitudes and behaviors: A test of the equal environments assumption. Behavioral Genetics 30: 51–8
21. Klump K L, McGue M K, Iacono W G 2000 Age differences in genetic and environmental influences on eating attitudes and behaviors in adolescent female twins. Journal of Abnormal Psychology 109: 239–51
22. Lilenfeld L, Kaye W H, Greeno C, Merikangas K R, Plotnicov K, Pollice C, Rao R, Strober M, Bulik C M, Nagy L 1998 A controlled family study of restricting anorexia and bulimia nervosa: Comorbidity in probands and disorders in firstdegree relatives. Archives of General Psychiatry 55: 603–10
23. Lilenfeld L, Kaye W, Strober M 1997 Genetics and family studies of anorexia nervosa and bulimia nervosa. Balliere’s Clinical Psychiatry 3: 177–97
24. Plomin R, DeFries J C, McClearn G E, Rutter M 1994 Behavioral Genetics. Third Edition. W. H. Freeman & Co., New York Sorbi S, Nacmias B, Tedde A, Ricca V, Mezzani B, Rotella C 1998 5-HT2A promoter polymorphism in anorexia nervosa. Lancet 351: 1785
25. Stern S L, Dixon K N, Sansone R A, Lake M D, Nemzer E, Jones D 1992 Psychoactive substance use disorder in relatives of patients with anorexia nervosa. Comparative Psychiatry 33: 207–12
26. Strober M 1995 Family-genetic perspectives on anorexia nervosa and bulimia nervosa. In: Brownell K D, Fairburn C G (eds.) Eating Disorders and Obesity. Guilford Press, New York, pp. 212–18
27. Strober M, Freeman R, Lampert C, Diamond J, Kaye W 2000 A controlled family study of anorexia nervosa and bulimia nervosa. American Journal of Psychiatry 157: 393–401
28. Strober M, Lampert C, Morrell W, Burroughs J, Jacobs C 1990 A controlled family study of anorexia nervosa: Evidence of familial aggregation and lack of shared transmission with affective disorders. International Journal of Eating Disorders 9: 239–53
29. Sullivan P F, Bulik C M, Kendler K S 1998 The genetic epidemiology of binging and vomiting. British Journal of Psychiatry 173: 75–9
30. Treasure J, Holland A 1989 Genetic vulnerability to eating disorders: Evidence from twin and family studies. In: Remschmidt H, Schmidt M (eds.) Child and Youth Psychiatry: European Perspectives. Hogrefe & Huber, New York, pp. 59–68
31. Wade T D, Bulik C M, Kendler K S 2000 Anorexia nervosa and major depression: An examination of shared genetic and environmental risk factors. American Journal of Psychiatry 157: 469–71
32. Wade T D, Martin N, Neale M, Tiggemann M, Trealor S, Heath A, Bucholz K, Madden P in press The structure of genetic and environmental risk factors for three measures of disordered eating characteristic of bulimia nervosa. Psychological Medicine