Djungarian hamsters bred at the authors' institute reveal two distinct circadian phenotypes, the wild-type (WT) and DAO type. The latter is characterized by a delayed activity-onset, probably due to a deficient mechanism for photic entrainment. Experiments with zeitgeber shifts have been performed to gain further insight into the mechanisms underlying this phenomenon. Advancing and delaying phase shifts were produced by a single lengthening or shortening of the dark (D) or light (L) time by 6 h. Motor activity was recorded by passive infrared motion detectors. All WT hamsters re-entrained following various zeitgeber shifts and nearly always in the same direction as the zeitgeber shift. On the other hand, a considerable proportion of the DAO animals failed to re-entrain and showed, instead, diurnal, arrhythmic, or free-running activity patterns. All but one of those hamsters that re-entrained did so by delaying their activity rhythm independently of the direction of the LD shift. Resynchronization occurred faster following a delayed than an advanced shift and also after changes of D rather than L. WT animals tended to re-entrain faster, particularly following a zeitgeber advance (where DAO hamsters re-entrained by an 18-h phase delay instead of a 6-h phase advance). However, the difference between phenotypes was statistically significant only with a shortening of L. To better understand re-entrainment behavior, Type VI phase-response curves (PRCs) were constructed. To do this, both WT and DAO animals were kept under LD conditions, and light pulses (15 min, 100 lux) were applied at different times of the dark span. In WT animals, activity-offset always showed phase advances, whereas activity-onset was phase delayed by light pulses applied during the first half of the dark time and not affected by light pulses applied during the second half. When the light pulse was given at the beginning of D, activity-onset responded more strongly, but light pulses given later in D produced significant changes only in activity-offset. In accord with the delayed activity-onset in DAO hamsters, no or only very weak phase-responses were observed when light pulses were given during the first hours of D. However, the second part of the PRCs was similar to that of WT hamsters, even though it was compressed to an interval of only a few hours and the shifts were smaller. Due to these differences, the first light-on or light-off following an LD shift fell into different phases of the PRC and thus caused different re-entrainment behavior. The results show that it is not only steady-state entrainment that is compromised in DAO hamsters but also their re-entrainment behavior following zeitgeber shifts.
