Researchers Identify Core Traits Strongly Linked To Anorexia And Bulimia
International team hones in on genes for two eating disorders that affect 11 million Americans
PITTSBURGH, September 9, 2005 — An international team of researchers led by investigators at the University of Pittsburgh School of Medicine and the University of North Carolina at Chapel Hill report that they have identified six core traits that appear to be linked to genes associated with two common eating disorders: anorexia and bulimia nervosa.
Their findings, which appear in two papers published in the Sept. 8 online edition of the American Journal of Medical Genetics Part B, bring researchers closer to being able to identify specific genes and may also have implications for genetic studies of other complex genetic disorders.
The team culled six core traits from a list of more than 100 behaviors and personality traits believed to contribute to development of anorexia and/or bulimia among a group of almost 400 individuals with eating disorders. The six traits are: obsessionality (a form of perfectionism), age at menarche (menstruation), anxiety, lifetime minimum Body Mass Index (BMI; a measure of body size based on height and weight), concern over mistakes and food-related obsessions.
The studies also found that minimum BMI, concern over mistakes, age at menarche and food-related obsessions appeared to be more closely linked to bulimia, whereas obsessionality and anxiety appeared to be more closely linked to anorexia, suggesting that, although closely related, the two conditions have some underlying differences.
“The research underscores how critically important genetics are in understanding the causes of eating disorders. We have been under the false impression for the past century that these are culturally caused disorders. This work puts us one step closer to identifying genetic risk factors that will, we hope, ultimately have implications for understanding cause, treatment and prevention,” said Cynthia M. Bulik, Ph.D., director of the Eating Disorders Program at UNC School of Medicine and the William and Jeanne Jordan Distinguished Professor of Eating Disorders in the department of psychiatry. Dr. Bulik also is professor of nutrition at the UNC School of Public Health.
Anorexia nervosa is a serious eating disorder that strikes women of all ages and also can affect males. It is characterized by the relentless pursuit of thinness and obsessive fears of being fat. The extreme weight loss caused by self-starvation and related medical complications that accompany it can result in death, giving it the highest mortality rate of any psychiatric disorder.
In contrast, bulimia nervosa is characterized by recurrent episodes of binge eating followed by self-induced vomiting, fasting, excessive exercise, or misuse of laxatives or other substances to prevent weight gain. Although people with bulimia usually maintain a normal weight, they, too, often have an obsessive fear of gaining weight, express a constant desire to lose weight and feel intensely dissatisfied with their bodies. In the United States, approximately 10 million females and 1 million males are afflicted by either anorexia or bulimia.
Historically, anorexia and bulimia have been considered closely related disorders or manifestations of the same disorder that is influenced primarily by social and cultural norms, such as society’s emphasis on thinness and being attractive. In recent years, however, research has increasingly pointed to substantial biological and genetic contributions as well.
Studies of families have shown that individuals with a mother or sister who has suffered from anorexia are 12 times more likely than people without a family history to develop the disorder and have a four-times greater risk for developing bulimia. Studies also have consistently linked anorexia and bulimia to a cluster of moderately heritable personality and behavioral traits.
However, identifying the genetic variations in the human genome that cause these traits has, until now, been a very complicated and cumbersome process, said Bernie Devlin, Ph.D., associate professor of psychiatry and human genetics at the University of Pittsburgh School of Medicine and a senior author of both studies.
“Scanning the human genome for the genetic contributions to eating disorders had presented a messy analytic problem because there were many, many traits—almost the same number of traits as families,” explained Dr. Devlin. “Yet, not all of these traits were closely related to inheritance of anorexia or bulimia. What we did new in these studies was to combine statistical analysis with expert opinion to come up with a core set of traits that were strongly associated with these two disorders to use in further analyses.”
Walter H. Kaye, M.D., professor of psychiatry at the University of Pittsburgh School of Medicine and head of this international group of clinicians, researchers and statisticians seeking to understand the genetic causes of eating disorders, agreed that this approach presents a major step forward. “This pioneering study has been possible because of the generous support from the Price Foundation, and more recently the National Institute of Mental Health. This support has allowed us to bring together the kinds of statistical and clinical expertise we needed to conduct these kinds of analyses. Not many places have this spectrum of resources.”
In the second of the two papers, the investigators conducted what is known as “linkage analyses” on these six traits to determine whether they could be used to pinpoint regions of the human genome that increase an individual’s risk for developing these two eating disorders. To their surprise, the analyses produced a pattern of significant linkage signals for bulimia but a less significant and somewhat different pattern of signals for anorexia.
Based on their findings, the investigators conclude that genes on chromosomes 10, 14 and 16 significantly affect risks for developing bulimia, but have a lesser impact on risks for anorexia. They found less significant, but still “suggestive” signals from other regions of the genome that were more closely linked to anorexia.
Interestingly, when they combined the data from anorexic and bulimic individuals and again conducted the linkage analyses, all of the significant genetic signals for bulimia were slightly or substantially lessened, while some of the suggestive signals for anorexia were amplified. Although it is possible that combining the data reduced the significant signals for bulimia because they were merely “false positives,” the investigators surmise instead that anorexia and bulimia differ at a fundamental biological and genetic level.
They note, for example, that the region on chromosome 10 that produced significant linkage with bulimia but not anorexia overlaps substantially with regions of chromosome 10 that previously have been linked with obesity. According to Dr. Bulik, this finding is consistent with other studies and observations.
“If you look at families with a high incidence of bulimia, you also find a high incidence of obesity,” she said. “However, if you look at families with a high incidence of anorexia, you don’t find a high incidence of obesity. So, although these are two genetically related disorders, there clearly appear to be some differences between them.”
Although the investigators caution that their findings require further replication in larger samples of individuals with anorexia and bulimia, they suggest that their methods have much broader applications. In particular, they believe that their approach can be applied and/or adapted to linkage studies of other complex conditions, such as diabetes and hypertension.
“Diabetes and hypertension are examples of complex, chronic conditions in which there are multiple sets of complications, or traits. Like anorexia and bulimia, identifying all of the potential genetic contributions to these conditions has been a monumental task,” said Dr. Devlin. “We believe that our approach—first doing an upfront analysis to identify a core set of traits before conducting genetic linkage analyses—will prove fruitful for such studies as well.”
In addition to Drs. Bulik, Devlin and Kaye, other authors on the two studies include Silviu-Alin Bacanu, Ph.D., Vibhor A. Sonpar and Weiting Xie, University of Pittsburgh, department of psychiatry; Kelly L. Klump, Ph.D., Michigan State University, department of psychology; Manfred M. Fichter, M.D., University of Munich (Germany), Hospital for Behavioral Medicine; Katherine A. Halmi, M.D., New York Presbyterian Hospital; Pamela Keel, Ph.D., University of Iowa, department of psychology; Allan S. Kaplan, M.D., and D. Blake Woodside, M.D., Toronto General Hospital, department of psychiatry; James E. Mitchell, M.D., Neuropsychiatric Institute, Fargo, ND; Alessandro Rotondo, M.D., University of Pisa, Italy, department of psychiatry, neurobiology, pharmacology and biotechnologies; Michael Strober, Ph.D., University of California at Los Angeles, department of psychiatry and behavioral sciences; Janet Treasure, FRCPsych, Institute of Psychiatry and South London and Maudsley National Health Service Trust, department of psychiatry; Andrew Bergen, Ph.D., National Institutes of Health, National Cancer Institute; and Wade H. Berrettini, M.D., Ph.D., University of Pennsylvania, Center for Neurobiology and Behavior.
The investigators are currently seeking families with at least two members who have or had anorexia nervosa, and who would be willing to participate in their ongoing study. Individuals who speak English (German for the Munich site) and who meet the study protocol are eligible for participation in this study. To learn more about this ongoing international study or how to participate, go to http://www.wpic.pitt.edu/research/angenetics.
These studies were supported by grants from the Price Foundation of Geneva, Switzerland, and the National Institute of Mental Health, National Institutes of Health.