It has been known for many years that laboratory rats adjust food intake in response to changes in the energy content of their diet and thereby maintain energy intake at a constant level (Adolph, 1947). Control of energy intake has been observed in other species, including human subjects, and provides an important means for maintaining energy balance, especially in species with a varied food supply (Collier, 1986). This ability to maintain energy intake in the face of changes in dietary energy content implies the existence of a feedback signal that communicates changes in energy intake to mechanisms that control eating behaviour.Compensation for changes in the energy density of food appears to depend on the ability to catabolise the source of dietary energy (see Friedman, 1991). This suggests that the feedback signal controlling energy intake is not in the food, but rather is generated in the oxidation of metabolic fuels when the energy from food is finally realized (Friedman, 1991). Energy expenditure and the status of body energy stores also determine the number of kilojoules consumed. It is possible that these metabolic determinants influence energy intake via a signal derived from the breakdown of energy-yielding substrate (Friedman, 1991(Friedman, , 1995. In other words, control of energy intake, whether in response to changes in dietary energy content or in the service of body energy balance, is based on the ability to sense energy.In the present paper, I review evidence for a mechanism for the control of eating behaviour that is based on the detection of energy in its most fundamental biochemical form. The nature of the metabolic signal, the location of its sensor, and the means by which the signal is communicated to the brain are outlined. How such a mechanism might operate as an energy sensor to control energy intake in response to changes in dietary energy density, energy expenditure, and energy storage is then discussed.
METABOLIC CONTROL OF FOOD INTAKEWhat is the nature of the metabolic signal that controls food intake? The best-known theories focus on signals associated with glucose and fat metabolism (see Friedman & Stricker, 1976). According to 'glucostatic' and 'lipostatic' theories, intracellular glucose utilization or blood glucose level (Mayer, 1955;Campfield & Smith, 1990) provide one type of stimulus for eating behaviour, whereas body fat or fatty acid oxidation (Le Magnen et al. 1973;Harris & Martin, 1984) provide another. A different line of thought about the metabolic signal for food intake, which has evolved in parallel with the traditional glucostatic and lipostatic theories, vests control at a level of metabolism downstream from pathways involved in processing specific substrates. In this case, the signal is generated in the final oxidative pathways of metabolism where individual substrates are indistinguishable and where processes for the production of biochemically-useful energy predominate.Ugolev and his colleagues (Kassil et al. 1970) were the first to suggest that the control of f...