Daily variation of body temperature, locomotor activity and maximum nonshivering thermogenesis in two species of small rodents

Abstract

Many mammals are known to have clear circadian rhythms of body temperature (T_b). Large parts of the rhythms correspond to the oscillation of locomotor activity, but this activity does not explain the entire variation of T_b. We tested to what extent daily variation of nonshivering thermogenesis (NST), the most important source of heat in small mammals, may elucidate T_b oscillations in rodents. We studied daily variation of body temperature, locomotor activity, the rate of NST (MMR_NST; measured as a maximum response to noradrenaline (NA)) and resting metabolic rates in thermoneutral ambient temperature (RMR_TNZ) in two rodent species: yellow-necked mouse Apodemus flavicollis and root vole Microtus oeconomus. Both species were nocturnally active in our study. Mice had a clear circadian rhythm of T_b with minimum of 36.6±0.2°C at 8:00 and maximum of 38.1±0.3°C at 23:00, corresponding to the variation in RMR_TNZ. Their MMR_NST also exhibited significant daily variation, with the greatest response to NA between 16:00 and 19:00, when T_b rising and activity was still low before the onset of the dark phase. High level of MMR_NST at this time corresponded with elevated resting metabolic rates below thermoneutrality, but not RMR_TNZ. The smallest response to NA was recorded at the time of low T_b. In voles the lack of circadian T_b variation (average of 38.6±0.1°C) was associated with no variation in MMR_NST or RMR_TNZ, and contrasted with a clear rhythm of locomotor activity. Our results indicate that NST plays more important role than activity in shaping circadian rhythm of T_b.

Introduction

In the majority of mammals body temperature (T_b) is not regulated at a constant level over the course of a day, but rather demonstrates a reproducible oscillation (Aschoff (1970), Aschoff (1982); Refinetti, 1999a). The oscillations of T_b may be characterized by different periods and amplitudes. The circadian rhythm has a period of about 24 h (Aschoff, 1967; Fuller et al., 1979) and ultradian rhythm of approximately 2–6 h (Refinetti and Menaker, 1992). Several factors have been shown to play a role in regulation of the circadian rhythm of body temperature. One often-discussed factor is the influence of locomotor activity on the existence and amplitude of body temperature changes. Changes of activity in the circadian rhythm are correlated with body temperature rhythms (Refinetti, 1994; Brown and Refinetti, 1996; Decoursey et al., 1998; Weinert and Waterhouse, 1998; Benstaali et al., 2001). For example, the activity level explains 70% of T_b variation in golden hamsters Mesocricetus auratus (Refinetti, 1994), and the alternation of sleep and waking accounts for 84% of T_b variation in rats (Franken et al., 1992). Most of the authors who studied the correlation between body temperature and activity suggested that activity affected, but did not determine the rhythm of T_b (see Refinetti and Menaker, 1992 for review). Correlation between activity and body temperature in the laboratory mouse was strong only during the light phase and dark–light transition, whereas during the dark phase and light–dark transition the correlation was nonsignificant in many night-active animals (Weinert and Waterhouse, 1998). Moreover, Decoursey et al. (1998) found that the rise in T_b preceded the increase of activity level by about 1 h in golden hamsters and by 20 min in eastern chipmunks Tamias striatus. This suggests that the heat production related to the locomotor activity cannot account for the initial phase of the rise of T_b.

In small mammals, endogenous heat production is mainly achieved through nonshivering thermogenesis (NST), a heat production mechanism liberating chemical energy due to processes that do not involve muscular contractions (Sellers et al., 1954; Heldmaier, 1971; Jansky, 1973). Thus the heat production under conditions of basal metabolism, i.e. at rest in the postabsorptive state and at the ambient temperature of the thermoneutral zone (TNZ) is mostly nonshivering thermogenesis (called obligatory NST). Ambient temperatures below the TNZ elicit the additional nonshivering heat production (regulatory NST) that enables maintaining stable T_b. The capacity of an animal to generate heat by NST can be assessed by injection of noradrenaline (NA) (Hsieh and Carlson, 1957).

If the rise of T_b in circadian rhythm results from an elevated rate of NST, then it is reasonable to predict that NST, measured as metabolic response to noradrenaline (MMR_NST), would also show daily variation with maximum response during the rising period of T_b. Daily variation of NST has been reported by few authors (e.g. Haim et al., 1995; Haim and Zisapel, 1999; Jefimow et al., 2000) but we are not aware of any studies comparing daily T_b variations, activity rhythms and NST oscillations. The aim of this study was therefore to test a hypothesis that rodents exhibiting circadian rhythm of T_b also change their response to NA during the day, and that these changes are more important than locomotor activity in shaping daily variation of T_b.

Existence of daily rhythm of body temperature may result from changing rate of heat production or heat dissipation, or a combination of both (Rubal et al., 1992). Assuming that daily rhythm of T_b is determined primarily by the rhythm of regulatory NST, one can predict that the amplitude of MMR_NST variation should correlate with the amplitude of T_b variation. Therefore, it would be of interest to measure MMR_NST in: (1) animals with clear circadian T_b rhythm during different periods of the day, when T_b is low, increasing, and at the high level and, (2) animals without circadian T_b rhythm. We measured daily changes of body temperature, the respective maximum MMR_NST, resting rates of metabolism, and locomotor activity in yellow-necked mice A. flavicollis and root voles M. oeconomus. Mice species are known to have clear-cut circadian T_b rhythm with high amplitude of T_b variation (Rubal et al., 1992; Haim et al., 1995; Weinert and Waterhouse, 1999). On the other hand, vole species are known to have ultradian rhythm of T_b with no- or low-circadian rhythmicity. The amplitude of ultradian T_b oscillation in voles is lower than circadian T_b variation in mice (Moshkin et al., 2001; A. Myrcha and J. Żukowski, pers. comm.).