Recent data have suggested that the activity of various brain nuclei is modulated during sleep. In this context, we developed the idea that sleep may participate in adapting brain responsiveness to feeding, so as to tune the control of peripheral energy metabolism. In order to characterize the nature of a possible link between sleep and peripheral energy metabolism, we have investigated the relationship between sleep parameters [wakefulness (W), slow wave sleep (SWS), paradoxical sleep (PS)] and the intensity of peripheral lipid and protein deposition processes. To achieve this, by manipulating the amount and quality of food available to rats, we induced states of energy or protein depletion/repletion which would specifically affect lean or fat body mass, which was quantified by an analysis of body composition. In parallel, using a permanently implanted cortical electrode, we measured electroencephalogram signals (EEG) to quantify the time spent in W, SWS and PS. Analysis of EEG changes in relation to the changes induced in body composition, showed that (1) the amount of sleep (PS and SWS) followed the evolution of energy supply levels, and (2) the time spent in PS relative to SWS varied to a considerable degree (14–23.5%) and followed the same trend as the ratio of lean body mass to fat mass. These results suggest the possible existence of quantitative and qualitative interactions between sleep quality and the anabolic and catabolic processes of peripheral fat and protein deposition.