Phase control of ultradian feeding rhythms in the common vole (Microtus arvalis): the roles of light and the circadian system.

In their ultradian (2- to 3-hr) feeding rhythm, common voles show intraindividual synchrony from day to day, as well as interindividual synchrony between members of the population, even at remote distances. This study addresses the question of how resetting of the ultradian rhythm, a prerequisite for such synchronization, is achieved. Common voles were subjected to short light-dark cycles (1 hr darkness with light varying between 0.7 and 2.5 hr); to T cycles (long light-dark cycles in the circadian range--16 hr darkness and 3-13 hr light); to light pulses (15 min) during different circadian and ultradian phases; and to addition of D2O to the drinking water (25%). Short light-dark cycles and D2O were also applied to voles without circadian rhythmicity, after lesions of the suprachiasmatic nuclei. In these experiments, four hypotheses on synchronization of ultradian rhythmicity were tested: (I) synchronization by a direct response to light; (II) synchronization via the circadian system with multiple triggers, here called "cogs," each controlling a single ultradian feeding bout; and (III and IV) synchronization via the circadian system with a single "cog," which resets an ultradian oscillator and either (III) originates directly from the circadian pacemaker, or (IV) is mediated via the overt circadian activity rhythm. Short light-dark cycles failed to entrain ultradian rhythms, either in circadian-rhythmic or in non-circadian-rhythmic voles; light pulses did not cause phase shifts; and in extreme T cycles no stable phase relationship with light could be demonstrated. Thus, Hypothesis I was rejected. Changes in the circadian period (tau) were generated as aftereffects of light pulses, by entrainment in various T cycles, and by the addition of D2O to the drinking water. These changes in tau did not lead to parallel, let alone proportional, changes in the ultradian period. This excluded Hypothesis II. Both in T-cycle experiments and in the D2O experiments with circadian-rhythmic voles, the phase of ultradian feeding bouts was locked to the end of circadian activity rather than to the most prominent marker of the pacemaker, the onset of circadian activity. This was not expected under Hypothesis III, but was consistent with entrainment via activity (Hypothesis IV). On the basis of these experiments, we conclude that the most likely mechanism of ultradian entrainment is that of a light-insensitive ultradian oscillator, reset every dawn by the termination of the activity phase controlled by the circadian pacemaker, which is itself entrained by the light-dark cycle. Neither in circadian-rhythmic nor in non-circadian-rhythmic voles was the period of the feeding rhythm lengthened by administration of D2O. This insensitivity to deuterium is exceptional among biological rhythms.