Circadian mechanisms of multiple meal anticipation

Circadian food anticipatory activity (FAA) has been modeled as the output of a dual food-entrainable oscillator (FEO) system, based on reports that rats can anticipate two but not three daily meals, and two meals recurring at different circadian intervals. Using operant responding to measure FAA, we found that rats can anticipate at least four daily meals, that the anticipation rhythm persists during food deprivation, and that the master light-entrainable circadian pacemaker in the suprachiasmatic nucleus (SCN) is not required. Intact and SCN-lesioned rats could also anticipate 2 daily meals recurring at 24-hour and 26-hour intervals. Critically the two bouts of FAA persisted with ~24- and ~26-hour periodicities during food deprivation tests. These findings are consistent with a multiple circadian FEO model of food anticipation. A circadian FEO model is challenged by limited evidence that rats can anticipate meals scheduled at long (e.g., 16-hour) but non-circadian intervals, suggesting a 'universal timing system' that is not constrained to circadian intervals. At many of these intervals, however, meals occur at a few circadian phases every two or three cycles. Apparent non-circadian anticipation could therefore be based on entrainment of multiple circadian FEOs. We found that rats can anticipate meals recurring at 16-hour, and possibly 18-hour intervals, but do so by becoming active at each of three or four possible daily mealtimes, respectively, rather than every 16 or 18 hours. Anticipation of circadian feeding intervals does not require the SCN, but anticipation of food on fixed-interval reinforcement schedules in the seconds-to-minutes range has been reported to be disrupted in mice housed in constant light (LL), which damps SCN rhythms. To determine whether LL-induced circadian desynchrony has a similar effect in rats, we tested 15- and 30-second interval timing in rats housed in LL. We found no evidence for a similar disruption of short interval timing. These results confirm and extend the evidence that timing of short (seconds to minutes) and circadian intervals rely on formally and neuroanatomically distinct mechanisms, and that timing of intermediate (e.g., 4-, 6-, 12-, 16-, 18-hour) and circadian intervals relies on a system of circadian oscillators entrainable by food.