Tantalizing Neurobiological and Genetic Clues to the Development of Eating Disorders

Genetics, neurobiology, and environment each contribute to the development and perpetuation of eating disorders, according to four experts who participated in a special plenary session on neurobiology and genetics at the 2002 Academy for Eating Disorders meeting in Boston in April. As Moderator Dr. B. Timothy Walsh told the audience, “The nature versus nurture debate of 25 years ago has been succeeded by the clear recognition of an interaction between genes, biology, and the environment.”

The genetics of anorexia nervosa

Twin and family studies suggest substantial hereditability for anorexia nervosa, according to Wade Berrettini, MD, PhD. Dr. Berrettini also reported that investigators from the Price Foundation study of anorexia nervosa have identified a probable susceptibility gene for anorexia nervosa located at chromosome 1p34.

Noting that the general public believes that what is “genetic is written in stone,” Dr. Berrettini added that health professionals need to educate the public about the concept of a susceptibility gene. “A single susceptibility gene is neither necessary nor sufficient for the development of AN,” he said, and added that an affected person must inherit several susceptibility genes, each of which produces a small increase in risk for the disorder. Even then, the appropriate environmental influences must be also be present.

In family studies of AN, there is a substantial relative risk among first-degree relatives of AN probands, compared with first-degree relatives of control probands, Dr. Berrettini said. He also pointed to three characteristics of partially inherited susceptibility: First, partially inherited disorders tend to aggregate in families, and all types of behavior disorders show familial aggregation. A second characteristic is that the monozygotic (identical) twin concordance is substantially greater than the dizygotic (nonidentical) twin concordance in a host of behavioral disorders, including AN. This information can be used to estimate the fraction of the total risk for AN that can be attributed to genetic influences, he said. A third characteristic of partially inherited disorders comes from adoption studies showing that increased risk for the biological relatives of adoptees with the disorder. Any two pairs of siblings, by virtue of the fact that they have the same mother and father, share 50% of their DNA sequences, he said.

Biocultural influences on the development of eating disorders

Anthropologist Carol Worthman, PhD, outlined biocultural influences on the development of eating disorders. Although she doesn’t directly study eating disorders, she does study the interface between biology and culture, she said.

One of the measures of sociocultural change is age at menarche, and researchers have learned that humans are “very plastic” with respect with to growth and development,” she said. Changes in the timing at menarche can be affected by illness, nutrition, maternal background (maternal well-being in childhood and then in pregnancy), and cultural practices are now playing an enormous role. For example, among the Bundi tribe of the central highlands of New Guinea, the median age at menarche in the mid-1960s was 18 years. Tribal members who moved into more urban areas had more access to better nutrition, including high-protein foods. Over a short time, there was a decline of 0.8 years in menarche among the rural tribal members but a decline of more than 2 years among those in the urban setting.

Dr. Worthman noted that dehydroepiandrosterone sulfate (DHEAS), an adrenal androgen that shows a flex point at puberty and proceeds to climb linearly across the second decade and into the middle of the third decade, is a good marker of progression into puberty during a period when hormones are rising and falling. Urban children have higher levels of this hormone, presumably because they are at more advanced stages of development. The rural kids lag behind, she said, but the difference between the two groups diminishes with age. DHEAS peaks at age about age 25, so in many settings the rural children never catch up with the urban kids. This example of delayed development may be due to differences in the neuroendocrine system, and may have long-term consequences for neuroendocrine regulation. Typically, male and female Americans have very high levels of DHEAS. Lifetime patterns of steroid exposure can affect the amount of ovarian activity over the life span, Dr. Worthman said.

A study of depression in teens. Dr. Worthman also described some of the findings from the Great Smoky Mountain Study, which has been conducted over the last 8 years in 11 counties of western North Carolina. This population-based study includes about 1500 preteens and teens, 1100 of whom are population samples drawn from those counties and the other 450 of whom are a total sample of age-appropriate Cherokee. Researchers are trying to discover why girls develop depression later in life than boys, and have twice the rate of depression of boys by age 15. Boys are more likely than girls to be depressed at an early age, said Dr. Worthman. Another area of study is the significant increase in depression reported in girls as they become teens. She noted there is a linear relationship between estradiol levels and the prevalence of depression in girls.

The study has shown an environmentally interactive effect between depression and cortisol response. In low-risk environments, depression was associated with a lowered cortisol response. In high-risk environments, depression was associated with a high cortisol response.

“Because the ecologies in which kids grow up affect their gonadal function, on all kinds of levels, it is interesting to consider the relationship between gonadal steroids and the risk for mood disorders such as depression and the risk for mood disorders at adolescence in girls. Both estradiol and testosterone are implicated suggests that it may not just be the gonads but the adrenal glands as well,” said Dr. Worthman. She added, “I am more and more impressed by the lack of power of our models for dealing with mood and behavior disorders in boys. We have made enormous headway with respect to girls…but not boys. Our models for males are not yet as good as those for girls.” Males are also at risk for all sorts of risk-taking behaviors, including distortions in body image, she said.

The serotonin connection

“In light of recent findings, we have to find a place for serotonin mechanisms in a biopsychosocial framework on eating disorders development,” said Howard Steiger, PHD. He added that the real challenge is finding ways to specify how psychological factors, biological factors, and social imperatives favoring dieting and pursuit of thinness come together to explain why a person develops an eating disorder. Many useful clues come from comorbidity patterns, he said.

People with restricting patterns have a propensity to affective disturbances, anxiety disorders, and trait pathology characterized by perfectionism, preference for sameness, order, and control, and overly controlled, rigid, brittle behavior. In contrast, the person who binge-eats and purges is prone to anxiety and depression, but has a different spectrum of trait components, including impulsivity, mood lability, substance abuse, and dissociation. This is a far more dysregulated kind of profile, he said.

Gender-based effects of serotonin. Dr. Steiger reported that there is also evidence implicating serotonin in the development of eating disorders. Women have higher degrees of basal serotonin metabolism and apparently greater sensitivity to the serotonin system. He cited a recent study by Cowan showing that 3 weeks of modest dieting in women substantially alters serotonin levels. Thus, women are much more susceptible to the potential serotonergic implications of dieting than are men.

According to Dr. Steiger, in active cases of bulimia there is clear evidence of reduction of serotonin tone, blunted prolactin response to serotonin agonists, and reduced platelet binding of serotonin uptake inhibitors. In addition, dietary manipulations that reduce tryptophan also reduce availability of brain serotonin, and this exacerbates binge-eating among active bulimics. Active bulimia goes with reduced serotonin activity, and this is true during recovery as well. Kaye and colleagues have found various forms of normalization of sertonin activity in recovery, but evidence shows ongoing abnormality on the 5-HT(2A) receptor, and ongoing sensitivity to tryptophan depletion. There is also some evidence of persistence of serotonin anomaly in recovered bulimics.

In anorexia, one should see elevated serotonin tone, but in the active state there is a blunted prolactin response to serotonin agonists, decreased binding of serotonin uptake inhibitors, and reduced serotonin metabolism in bingeing-purging. Perhaps this is explained by phenomena during recovery where various serotonin indices normalize. If so, this would suggest that the reduction in serotonin activity seen in active anorexics has a lot to do with secondary effects of malnutrition.

Dr. Steiger noted that serotonin status can be related to psychopathological profiles among eating disordered patients. One subgroup seems to have relatively intact serotonin function, even though they have severe eating disorders; a second group has greater serotonergic dysregulation.

For patients with more intact serotonin systems, stabilizing eating and stopping dieting behaviors will improve their outcome. In the comorbid group, where developmental experiences have led to more fundamental dysregulation, pharmacologic and special psychotherapeutic adjuncts will need to be aimed at impulsivity and posttraumatic sequelae, Dr. Steiger said.

Genetic and environmental interactions: an animal model

Genetic and environmental influences team to produce modifications in behavior, according to Stephen J. Soumi, PhD. Traumatic experiences early in life may also play an important role, he said.

In his long-term studies with rhesus monkeys, Dr. Soumi found that from 5% to 10% of the monkeys seem to be unusually impulsive and inappropriately aggressive in response to mildly stressful conditions. This modification can take place not only at the level of behavior but or of physiology, or of neurotransmitter level or metabolites, but possibly even at the level of gene expression as well.

Genetic and environmental factors appear to interact throughout development to shape individual differences, he said. He added that the early environment is particularly important.

The Panel

Dr. Walsh is Ruane Professor of Pediatric Psychopharmacology at Columbia University and Director of the Eating Disorders Research Unit at The New York State Psychiatric Institute, New York, NY. Dr. Berrettini is Karl E. Rickels Professor of Psychiatry and Director of the Center for Neurobiology and Behavior at the University of Pennsylvania, Philadelphia. Dr. Worthman is Director of the Department of Anthropology, and Director of the Laboratory for Comparative Human Biology, both at Emory University, Atlanta. Dr. Steiger is Director of the Eating Disorders Program at Douglas Hospital, and professor of psychiatry at McGill University, Montreal. Dr. Soumi is Chief, Laboratory of Comparative Etiology, National Institutes of Child Health and Human Development, National Institutes of Health.