What Makes People Fat? View from the Chair


Department of Pharmacology, University of Alberta, 9-36 Medical Sciences Building, Edmonton, Alberta, Canada T6G 2H7. E-mail: william.colmers@ualberta.ca

Obesity is a persistent and increasingly widespread and severe disorder in the populations of affluent and, increasingly, emerging nations. The four papers accompanying this essay discuss different aspects of this, from the biology of hunter-gatherers to brain signals and two molecules, one of which definitely plays an essential role in energy balance (leptin) and one which has exciting prospects as an emerging player (fatty acid synthase inhibition).

In the first paper, Barry Levin outlines the essential physiological problem underlying Western-style, or diet-induced, obesity. In feral hunter-gatherer populations of humans, the ability to buffer energy stores, i.e., store excess fat in times of plenty and reduce energy expenditure in lean times is a physiological advantage that helps overcome the uneven availability of an adequate food supply. In modern advanced societies, there are no significant lean times. Nevertheless, the systems that evolved to buffer energy stores continue to operate, and the only physiological circumstance in which there is a relative paucity of food is when we try to lose weight by dieting. Dr. Levin argues that the “perfect survivor” traits work against us when highly palatable, energy dense foods are easily available. This is probably a genetic, although polygenic, trait and found in rodents as well as humans. To be able to bypass the consequences of our ancestral physiology will require a clearer understanding of the signals involved. Neurons in the brain respond to a plethora of signals related to energy balance, including glucose, peripheral hormones such as leptin, ghrelin, and insulin, and a large number of central nervous system signals. These regulate behavior not only related to ingestion but also arousal, foraging, reproduction, and other behaviors that determine the likelihood of finding food or whether to make the most of a positive energy balance state. However, despite this insight, one of the more enduring mysteries is the mechanism(s) by which the observed slow upward creep of body weight occurs with time when calorically dense foods are easily available.

The next paper, by Hans-Rudolf Berthoud, outlines a possible explanation for this upward movement in body weight. This author points out that tight homeostatic regulation of appetite is important; however, there exist non-homeostatic controls that, on balance, favor weight gain in a permissive environment. Thus, leptin is a good indicator of fat mass and, therefore, energy stores; its concordant action to regulate appetite in function of fat mass seems adaptive. However, the development of leptin resistance means that the signaling system is not meant to override other appetitive behaviors in the face of readily available, palatable, calorically dense food. In recent modeling work by de Castro cited by the author (1), cephalic components (reward centers) and intestinal components are shown to provide a significant, reward-driven, non-homeostatic component to the energy balance equation. The implication of this calculus is that a non-homeostatic, reward-based drive will add to the homeostatic components, resetting the physiological sensitivity of the homeostatic component to higher body weight levels. This makes some sense in that hierarchically, the accumbens, and cephalic reward centers evolved more recently that the homeostatic centers in the hypothalamus. It has other implications, namely that other, non-homeostatic forces may also prevail on this system to shift body weight in the downward direction. Thus, it would be very interesting to determine whether both sides of the energy balance equation are equally well defended in individuals with anorexia-cachexia, where the energy intake is already below the basal metabolic expenditure (2, 3). The author outlines the evidence for a direct connection between the accumbens and the hypothalamus, especially perifornical orexin neurons and arcuate neuropeptide Y neurons, and marked increases in food intake with manipulation of the accumbens opioid system. There are projections from different regions of hypothalamus to the accumbens and ventral tegmental area, particularly orexin inputs from lateral hypothalamus (4)(5), so there are reciprocal connections back to the accumbens, but it does seem that these evolutionarily more recent circuits may not be as tightly regulated as the intrahypothalamic circuits are.

Leptin and its receptors have been at the heart of the discussion on energy balance since their discovery in the mid-1990s. The one (of five structurally-related) receptor mediating the leptin signal (termed variously ObRb or LRb)1 is a Janus kinase/signal transducers and activators of transcription receptor, which can activate a number of signaling pathways simultaneously. Mutations in the LRb (most famously in the db/db mouse) result in gross obesity, type 2 diabetes, and a variety of other related disorders mimicking the metabolic syndrome in humans. The laboratory of Myers (6) has undertaken the considerable task of dissecting the signals emanating from this receptor on ligand activation. To identify neurons expressing the LRb, this laboratory has generated a mouse in which the bacterial DNA editing enzyme, cre-recombinase, is expressed in cells that express the LRb. By crossing these leprCre mice with other mice that express lox-P flanked reporter genes such as β-galactosidase, they were able to identify nerve cells in the brain naturally expressing the LRb. Numerous neurons in the arcuate, ventromedial, dorsomedial, and lateral hypothalamic nuclei express the LRb. These tools will permit the identification of neurochemical phenotypes of neurons that express LRb and provide a much clearer idea of the central pathways that respond to the leptin signal.

In the final paper in this section, the laboratory of Ronnett has taken a novel approach to treating the problem of overeating. The enzyme, fatty acid synthase (FAS), is responsible for the de novo, ATP-dependent synthesis of fatty acids. Inhibition of FAS (to reduce storage of fat by peripheral tissues) with an experimental compound called C75 was found to have a remarkable dose-limiting effect, that of weight loss. Much work has gone into clarification of the mechanism of action of this compound. One detailed study (7) has suggested that the action of C75 on arcuate nucleus neurons is non-specific. However, other studies discussed in detail here by the Ronnett group have suggested that the suppression of food intake and weight loss caused by C75 administration results from the inhibition of FAS in the brain. The controversy has not been entirely calmed by the availability of the prototypic FAS inhibitor, cerulein, which is structurally different from C75, but which is not suitable for the necessary in vivo studies. This approach, while certainly appealing, must await evidence that other FAS inhibitors, preferably ones with a different root structure, that have similar effects on appetite as is reported for C75.

In summary, obesity has many causes, some of which are arguably adaptive from an evolutionary perspective, but maladaptive in the present circumstance of unrestricted availability of highly palatable, energy dense foods. The signals that induce overeating can be seen as being beneficial for opportunistic foragers, but some can be seen to have evolved without a compensatory homeostatic component. One major homeostatic signal, leptin, has receptors expressed on a restricted subset of neurons in the hypothalamus, which may eventually predict the circuitry involved in homeostatic control of energy balance. Finally, a lipid synthetic enzyme, with a mundane task in the periphery, may lead an exciting double life in the brain as a major sensor of energy balance. Clearly, there is much to chew on here.


W.F.C. is a Medical Scientist of the Alberta Heritage Foundation for Medical Research. Work in his laboratory is supported by the Canadian Institutes for Health Research. I thank Dr. V Baracos for helpful suggestions on the manuscript.


  • 1

    Nonstandard abbreviations: LRb, leptin receptor b; FAS, fatty acid synthase.