Animal Models of Eating Disorders
Jeanette E. Johansen and Martin Schalling
Eating disorders such as obesity, anorexia and bulimia are complex disorders displaying a variety of symptoms apart from an abnormal eating behaviour. Like many other motivated behaviours, feeding requires the integration of internal and external signals and it is not clear if the physiological correlates observed in these disorders are causes or effects of the altered eating behaviour. A good understanding of the physiology underlying feeding behaviour is therefore essential. This chapter deals with some of the many animal models that are being used to study feeding behaviour.
Early animal models of eating disorders include experimental studies of the effect of anorectic drugs on the amount of food consumed by rats. However, while successful in rats, pharmacological treatment of obesity is generally unimpressive in terms of weight loss in humans for a number of reasons.
That tumours in the region of the hypothalamus can cause obesity has been known for a long time. In 1940, Hetherington and Ranson confirmed the importance of the hypothalamus in the control of feeding and body weight. By performing electrolytic lesions in the hypothalamus of rats, they observed: 'A condition of marked adiposity characterized by as much as a doubling of body weight and a tremendous increase of extractable body lipids…' (Hetherington and Ranson, 1940). The damaged regions included the dorsomedial and ventromedial hypothalamic nuclei (DMH and VMH), the arcuate nucleus (Arc), the fornix, the lateral hypothalamic area (LHA) ventral to the fornix and possibly also the ventral premammillary nucleus. They also noted that lesions in the adjacent lateral hypothalamus could lead to decreased food intake. Anand and Brobeck pursued this observation and showed that bilateral electrolytic lesions of the LHA caused loss of feeding and even death by starvation (Anand and Brobeck, 1951). Thus, the concept arose of the LHA serving as a 'feeding centre' and the VMH as a 'satiety centre'—the dual centre model. This hypothesis has been widely questioned, and among the observations speaking against the dual centre model are findings that damage outside the hypothalamus can produce syndromes similar to those seen after lesions of the LHA or VMH. As cell-specific lesion methods emerged the focus was once again put on the VMH and LHA, and several studies showed that the LHA indeed could have a phagic function (Saper, 1985; Saper et al., 1986; Bittencourt, 1992). Today we know that hypothalamic cell populations and nuclei play important and specific roles in the regulation of food intake and other motivated behaviours.
Most of the population practices weight control, but in spite of that, weight seems to be stable in both lean and obese individuals. Dieting is usually not successful in the long run and most obese individuals eventually regain the lost weight (Wadden, 1993). The relative stability of weight in individuals indicates that there is a feedback loop controlling energy balance and maintaining constancy of total body energy stores. In 1953, Kennedy introduced his theory on a lipostatic mechanism that maintained energy homeostasis (Kennedy, 1953). He suggested that the size of the fat depots were sensed by a lipostat, which would regulate and adjust food intake and energy metabolism accordingly, to maintain body weight at a set point. He also proposed that an impaired lipostatic mechanism could lead to obesity. Further support for this hypothesis came from a study by Hervey (1958). He performed a series of parabiosis experiments, where the circulations of two animals are surgically joined, and showed that lesions in the VMH in one of the members in a parabiotic rat pair caused the lesioned rat to become obese as the unlesioned rat starved to death (Hervey, 1958). He suggested that the obese lesioned rat produced excessive amounts of a satiety factor that was transferred to the unlesioned rat, causing it to starve itself.
Over the past five years there has been a tremendous increase in the understanding of the genetic regulation of food intake and energy expenditure, using monogenic rodent models of obesity. Several genes have been cloned that, when mutated, cause obesity in the mouse and rat (Schalling et al., 1999; Barsh et al., 2000). These genes and their products have unravelled biochemical pathways involved in obesity. Some of these genes have been shown to be important for the regulation of food intake and/or metabolism also in humans. Crosses of mouse, rat, pig or chicken strains that are informative with regard to body mass or body fat have produced a number of quantitative trait loci (QTL) (Chagnon et al., 2000) that have opened the door to polygenenic approaches in the study of obesity in animals. As with the monogenic rodent models, these QTLs can be applied to a human genetic obesity map by identifying the syntenic chromosomal regions in the human.
There has been less of a focus on genetic models of anorexia. One reason might be that there are more genetic animal models that shift the regulation of food intake, satiety or metabolic turnover towards obesity than anorexia. A possible explanation for this could be that even a relatively mild anorectic phenotype could lead to malnutrition and/or death by starvation early enough in life to affect the number and viability of the offspring (see the anorexia mouse, the dopamine-deficient mouse and the HNF–3α deficient mouse below). A number of starvation, dehydration and chemically induced models have been developed and an example of each is discussed below. There are also numerous models of anorexia induced by either infection or cancer. Anorexia, or cachexia, is a frequent complication of malignant tumours and infectious and inflammatory diseases and is contributing significantly to the mortality of these disorders (Kotler et al., 1985; Tisdale, 1997; Larkin, 1998). This type of anorexia will not be dealt with in this chapter.
There are very few animal models of bulimia or binge eating. The models that exist are all based on cycles of food restriction