From Gut to Mind—Hormonal Satiety Signals and Anorexia Nervosa

The need to understand the regulation of appetite and energy balance has never been greater. The emergence of obesity as a global public health problem has imparted impetus to efforts directed at characterizing the complex neuroendocrine interactions that govern feeding. Similarly, eating disorders are responsible for very significant morbidity, mortality, and socioeconomic dysfunction. A recent review of outcomes in anorexia nervosa (AN) painted a gloomy picture, reporting a crude mortality rate of 5% and a recovery rate of less than 50% (1), figures that have changed little over the course of the last century. More effective therapeutic approaches are needed, and it is in this context that work to define better the pathogenesis of AN has been making progress. With some timely data on peptide YY (PYY) in AN, Misra et al. (2) supplement our existing knowledge of the changes in the gut hormone milieu that accompany changes in body weight.

PYY, so named because it has a tyrosine residue at either end, is a 36-amino-acid peptide released postprandially by the L-cells of the gastrointestinal tract into the circulation in proportion to the calories ingested (3). It is also present in the central nervous system (CNS), and a role for PYY in the regulation of food intake was initially predicated on the basis of observations that PYY injected directly into the CNS, either into the ventricular system or into the paraventricular nucleus of the hypothalamus, mediated an increase in food intake (4). A number of lines of evidence argued for a place for PYY, acting as a neurotransmitter, in the pathogenesis of eating disorders. In particular, similarities were noted between the hyperphagia induced by CNS injection of PYY and the pattern of ingestive behavior associated with bulimia nervosa, and several studies demonstrated differences in cerebrospinal fluid levels of PYY in abstaining bulimics when compared with controls, anorexics, or their own cerebrospinal fluid shortly after binge eating and vomiting (4).

Neuropeptide Y (NPY) is structurally related to PYY, and they bind to the same receptor family (5). NPY is present in the CNS in much larger quantities than PYY and, importantly, it is expressed in neurons of the hypothalamic arcuate nucleus (ARC) that project onto key appetite-regulatory areas, including the paraventricular nucleus (6). The orexigenic properties of NPY are well recognized, and the possibility therefore remains that the hyperphagia mediated by CNS-injected PYY is a pharmacological, rather than a physiological, effect, arising from the action of exogenous PYY at receptors that would physiologically be occupied by NPY.

In the circulation, the major form of PYY is the N-terminal-truncated PYY_3–36, which demonstrates relative specificity for the receptor subtype Y₂. This receptor is highly expressed on NPY neurons in the ARC, where it putatively acts as an inhibitory autoreceptor, providing a means for NPY to regulate its own release. PYY_3–36 injected systemically into a number of animal species, including humans, inhibits food intake, in contrast to the actions of CNS-injected peptide (7–11). This effect occurs at plasma levels of PYY_3–36 similar to those seen physiologically after a meal and is lost in Y₂ receptor knockout mice. Furthermore, peripherally injected PYY_3–36 induces expression of c-fos in the ARC, and injection directly into the ARC inhibits food intake, again contrasting with the effects of less-targeted intracerebroventricular administration (7). Notwithstanding a failure of some groups to reproduce the anorexigenic effect of peripherally injected PYY_3–36 in rodents (12), the weight of evidence now available strongly favors a satiety-promoting role for PYY.

The quest for novel therapies for diseases characterized by disordered energy homeostasis leads naturally to an examination of the possibility that abnormal gut hormone signaling may play a role in the pathogenesis of these conditions. Elevated levels of several gut hormones after gastric bypass surgery have been proposed as a possible mechanism by which postoperative reduction in appetite and loss of weight is achieved (13, 14). Similarly, PYY levels have been found to be lower in obese individuals than in lean controls but to rise after weight loss (8, 15, 16). The findings reported in this issue by Misra et al. (2) both of significantly elevated fasting plasma PYY levels in subjects with AN, and of a reciprocal change in PYY levels with weight change, are concordant with these previous observations.

The tantalizing thought that AN may arise out of a tendency toward elevated levels of gut satiety signals must be tempered, however, by a number of caveats. First of all, any model that defines the pathogenesis of AN in terms of dysregulation of the gut-brain axis must also explain the epidemiological characteristics of the condition in terms of sex, age, social class, and geographical distribution. Furthermore, levels of other gut peptides have been found to be altered in patients with AN, resulting in a mixture of both orexigenic and anorexigenic signals (17–19). The integration of these opposing drives into a net effect on food intake will be crucial to any understanding of the contribution of peripheral signals to the development of AN. The influence of gut hormones on food intake is a function not only of basal circulating peptide levels but also of the dynamic response to a meal. In this regard, the data presented here, as the authors acknowledge, is somewhat at odds with a previous, smaller study in which the secretion profile of PYY in response to a liquid mixed meal was found to be no different in subjects with AN compared with controls (20).

AN is a multisystem disorder, and Misra et al. (2) also report intriguing results demonstrating a correlation between log fasting PYY levels and a number of markers of bone turnover. This follows on from recent interest in the influence of Y₂ receptor activity on bone turnover (21). Again, although the existence of a correlation does not necessarily imply a causal relationship, if prevailing PYY activity does affect bone turnover, the implications would extend beyond AN. Any therapy for obesity, for instance, based on PYY or Y₂ receptor agonism would have to be carefully evaluated for its effects on bone density in a population already at increased risk of osteoporosis.

In conclusion, the data reported by Misra et al. (2) further illuminate our understanding of the interrelationship between body weight and circulating levels of PYY and are suggestive of a gut-bone interaction that introduces challenges for future therapeutic manipulation of the gut-brain axis. But will PYY antagonism be a useful treatment for AN? That depends on the nature of the association that has been revealed. The epidemiology of AN makes it unlikely that the primary cause of the disease is disordered gut hormone signaling. On the other hand, the elevation in PYY levels might result simply from an increase in speed of delivery of nutrients to the distal gut caused by chronic anxiety-related intestinal hurry—analogous to the discomfort many readers may have experienced when anticipating a stressful event. Against this explanation is the relative paucity of patterns of food intake characteristic of AN and also the observation that postmeal dynamic PYY levels in AN sufferers are similar to those in both normal weight and obese individuals (20).

Recent advances in knowledge of the gut-brain axis leave room for a third, more intriguing, explanation, namely that the psychopathology of AN results in direct up-regulation of neuroendocrine signaling from the distal gut. Further investigation of this possibility, perhaps using techniques including functional imaging and electrophysiological studies, may shed light on the mechanisms and pathways by which higher centers participate in the control of satiety. In the meantime, a realization that AN sufferers are exposed to abnormally elevated levels of a satiety-promoting gut hormone provides ample justification for an empirical trial of treatments aimed at ameliorating the imbalance between signal and body energy stores.

Acknowledgments

O.B.C. is supported by a Wellcome Trust Clinical Research Training Fellowship. B.C.T.F. is supported by a Medical Research Council Clinical Research Training Fellowship. S.R.B. is a director of, and has an equity interest in, Thiakis, a company with a commercial interest in PYY as a therapy for obesity.

Abbreviations:

 
• AN,

  Anorexia nervosa;

 
• ARC,

  arcuate nucleus;

 
• CNS,

  central nervous system;

 
• NPY,

  neuropeptide Y;

 
• PYY,

  peptide YY.

References