The effect of food timing on torpor propensity and characteristics in laboratory mice during a common food restriction paradigm

Background Many behavioural, pharmacological, and metabolic studies in mice require fasting, yet the possibility of fasting-induced torpor affecting the data is rarely considered. Torpor is a state characterised by depressed metabolism and profound alterations to physiology and behaviour. In this study we aimed to determine how the effects of torpor on experimental outcomes could be mitigated. Methods To this end, timing and characteristics of fasting-induced torpor in response to feeding in the morning versus feeding in the night were compared using non-invasive monitoring of peripheral body temperature. Results Night-fed mice entered significantly more torpor bouts per day compared to morning-fed mice (Morning: 2.79±0.197 (mean ± SEM); Night: 4.79±0.533 (mean ± SEM); p=0.0125), but these bouts were shorter on average by ~1.5h. Latency to the first torpor bout following feeding tended to be shorter during night feeding (Morning: 9.57±0.8h (mean ± SEM); Night: 6.66±1.2h (mean ± SEM); p=0.0928). Moreover, torpor bouts typically occurred during the dark phase in the morning-fed group, whilst night feeding resulted in a shift of torpor occurrence to earlier in the day (Morning: 14.2±0.4 ZT h (mean ± SEM); Night: 12.2±0.9 ZT h (mean ± SEM); p=0.0933). There was a high degree of variation in torpor occurrence within and between animals in each group. Conclusions We recommend that feeding time is kept consistent between days and the same across animals to minimise variation in torpor occurrence. Moreover, the timing of food provision may be optimised to allow measurements to be taken during euthermia, to mitigate the effects of torpor on the variables investigated. Finally, we recommend that body temperature is monitored non-invasively to determine when torpor is occurring, and that testing, or sample collection is conducted when the torpor history is comparable between animals.