THE JOURNAL OF EXPERIMENTAL ZOOLOGY 275:67-70 (1996)

RAPID COMMUNICATION

Comparison of the Body Temperature Rhythms of Diurnal and Nocturnal Rodents

ROBERTO REFINETTI Department of Psychology, College of William and Mary, Williamsburg, Virginia 23187

ABSTRACT The daily rhythm of body temperature of six rodent species, three of them noctur- nal (Rattus noruegicus, Mesocricetus auratus, and Meriones unguiculatus) and three diurnal (Octodon degus, Spermophilus tridecemlineatus, and S. richardsoni), was studied by telemetry under a 14L: 10D light-dark cycle. Significant interspecies differences were found in the mean level (range: 36.0-37.4"C), amplitude (range: 2. lo4.3"C), robustness (range: 115-185 Q,), and acrophase (range: 8.5-16.0 hr) of the temperature rhythm. In addition, a positive correlation (r = 0.79) was found between body mass and robustness of the rhythm across species. However, the only consistent difference between the diurnal and nocturnal species was the acrophase, which occurred during the day in the diurnal species (range: 8.5-11.5 hr) and during the night in the nocturnal species (range: 13.5-16.0 hr). 0 1996 Wiley-Liss, Inc

Birds and mammals maintained in a constant- temperature environment exhibit a daily rhythm of body temperature with the approximate shape of a cosine wave and an amplitude of 1°-3"C (Refinetti and Menaker, '92b). Body temperature rhythms with unusual shape and exaggerated amplitude (up to 5OC) have been described in some homeothermic species (Binkley et al., '71; Lee et al., '90; Refinetti and Menaker, '92a), and much larger amplitudes and irregular shapes have been described in heterothermic animals that undergo daily torpor (Lyman, '82). However, very few at- tempts have been made at comparing the body temperature rhythms of different species studied under identical experimental conditions. This re- port compares the body temperature rhythms of three diurnal and three nocturnal rodent species studied with the same equipment and maintained under identical conditions of housing, ambient temperature, and illumination.

MATERIALS AND METHODS Four male adult specimens of each of six species

were used in this study. The three nocturnal spe- cies were Sprague-Dawley rats (Rattus norvegicus, 473-520 g), Syrian hamsters (Mesocricetus aura- tus, 134-182 g), and Mongolian gerbils (Meriones unguiculatus, 60-66 g). The three diurnal species were Chilean degus (Octodon degus, 212-267 g), thirteen-lined ground squirrels (Spermophilus tridecemlineatus, 148-192 g), and Richardson ground squirrels (Spermophilus richardsoni, 525-646 g). 0 1996 WILEY-LISS. INC.

All animals were housed in individual plastic cages (21 x 30 x 20 cm) lined with wood shavings under a 14L:lOD (800:O lux) light-dark cycle in- side a ventilated incubator maintained at 24°C. Prolab rodent pellets, Hartz gerbil food, and wa- ter were available ad libitum. Data collection was conducted during the summer and early fall.

Body temperature was measured by telemetry using a sensor-transmitter (model VM-FH, Mini- Mitter Co., Sunriver, OR) surgically implanted in the abdominal cavity under sodium pentobarbital anesthesia (40-80 mgkg) and a radio receiver (model RA-1010, Mini-Mitter Co.) attached to a computerized data acquisition system (Dataquest 111, Data Sciences Inc., St. Paul, MN). For each animal, body temperature was recorded every 6 min for 2 or more weeks and saved on disk for later analysis.

Analysis of the body temperature data consisted of calculations of the mean level, amplitude, ro- bustness, and acrophase (time at which peak tem- perature is reached) of the daily rhythm over blocks of 10 days. The mean level of a rhythm was calculated as the arithmetic mean of all read- ings during the 10-day period (2,400 data points), Amplitude was calculated as the difference be- tween the maximum and minimum values, the maximum value being defined as the highest read-

Received December 4, 1995; revision accepted February 26, 1996. Address reprint requests to Dr. R. Refinetti, Department o f Psy-

chology, College o f William and Mary, Williamsburg, VA 23187.

68

35

33 -

R. REFINETTI

--

..A I

I I

$ 35--

I "" I I

"D I I I

37

35

B 33 I I I t

39 1 I

37

35

33

39 i t i i

37

35

33 1 2 3 4 5

Days

Fig. 1. Five-day sections of the records of a representative animal of each of the six species studied ( A rat, B: golden hamster, C: gerbil, D: degu, E: 13-lined ground squirrel, F: Richardson ground squirrel). Data were collected and are plotted in 6-min intervals. Lights were on from 0000 to 1400 hr.

ing recorded at least ten times during the 10 days and the minimum value as the lowest reading re- corded at least ten times. The robustness of a rhythm was calculated ;IS the value of the Qp sta- tistic (for P = 24 hr) in the Sokolove-Bushel1 periodogram procedure (Sokolove and Bushell, '78), which has been shown to accurately reflect

the signal-to-noise ratio of the data (Refinetti, '93b). For calculation of the acrophase, the data were first smoothed by a 4-hr moving averages filter. Then, the acrophase for each animal was calculated as the mean of the times (in hours since lights-on) of the ten daily rises of body tempera- ture. Interspecies differences in each of the four

BODY TEMPERATURE RHYTHM OF RODENTS 69

0 0, 97.0

36.0

35.0

Mean Level

4.0 0, "T Amplitude

220 T Ro y t n e s s

8 160

130

100

I 18*o T Acrophase

U 16.0

8.0 el- +

e a -

tl

Fig. 2. Interspecies comparison of four parameters of the body temperature rhythm. In each graph, bars indicate the mean (T SE) of four animals.

. / 2oo T

lZ5V . 100 I

0 150 300 45 0 600 750 Body Mass (g)

Fig. 3. Robustness of the body temperature rhythm as a function of body mass. Each point corresponds to the mean values for each of the six species.

parameters were evaluated by analysis of vari- ance (ANOVA).

RESULTS Five-day sections of the records of representa-

tive animals of each of the six species are shown in Figure 1. Daily rhythmicity can be seen in all species, despite noticeable differences in mean level, amplitude, robustness, and acrophase.

The mean results for the six species are shown in Figure 2. Significant interspecies differences were found in the mean level of body tempera- ture [F(5,18) = 5.83, P < 0.011, the amplitude of the daily oscillation [F(5,18) = 7.10, P < 0.011, the robustness of the rhythm [F(5,18) = 3.62, P < 0.021, and the acrophase [F(5,18) = 12.58, P < 0.011. As expected, diurnal and nocturnal animals differed consistently in the acrophase of their body tem- perature rhythms. For the three diurnal species, the acrophase was approximately 10 hr after lights-on (or 4 hr before lights-off); for the three nocturnal species, the acrophase was approxi- mately 15 hr after lights-on (i.e., 1 hr after lights- off). No consistent difference between diurnal and nocturnal species was found in the other three parameters of the body temperature rhythm.

Because the body masses of the six species dif- fered by a full order of magnitude (from 60 g ger- bils to 640 g Richardson squirrels), the correlations between mean body mass and the mean value of each of the four parameters of the body tempera- ture rhythm were analyzed. As shown in Figure 3, robustness of the rhythm was strongly corre- lated with body mass (r = 0.79, which, for this small sample of six species, attains marginal sta- tistical significance; P = 0.06). The correlation be-

70 R. REPINETTI

tween body mass and each of the other three pa- rameters did not even approach significance (for mean level, r = -0.22; for amplitude, r = -0.13; for acrophase, r = -0.10).

DISCUSSION The results are consistent with previous obser-

vations of considerable interspecies variability in the parameters of the body temperature rhythm (Refinetti and Menaker, '92b). However, no evi- dence was found that diurnal and nocturnal spe- cies consistently differ in any parameter of the temperature rhythm except the acrophase. As ex- pected, the acrophase of the temperature rhythm occurred during the day in diurnal animals and during the night in nocturnal animals. These find- ings are consistent with the idea that, except for the difference in the posj tion of the activity phase in relation to the phase-response curve to light pulses, there is no significant difference in the cir- cadian physiology of diurnal and nocturnal ani- mals (Refinetti, '93a).

The finding of a positive correlation between body mass and the robustness of the body tem- perature rhythm was unexpected. A survey of the literature suggested a correlation between body mass and the amplitude of the temperature rhythm (Aschoff, ,821, but the analysis of robust- ness could not be previously performed because of the differences in experimental procedures among the various studies. In the present study, the different species were tested with the same equipment under identical environmental condi- tions, which allowed a reliable comparison of ro- bustness. The positive correlation between body mass and rhythm robushess indicates that larger

the amplitude of the daily oscillation. Whether the correlation found for the six rodent species inves- tigated in this study can be generalized to other mammalian species, or even to other rodent spe- cies, is an empirical question that must be ad- dressed by further investigation.

ACKNOWLEDGMENTS This work was supported by an NSF Career

Award to the author (IBN-9507452). Dr. Theresa Lee (University of Michigan) generously provided the Chilean degus.

LITERATURE CITED Aschoff, J. (1982) The circadian rhythm of body temperature

as a function of body size. In: A Companion to Animal Physi- ology. C.R. Taylor, K. Johansen, and L. Bolis, eds. Cam- bridge University Press, New York, pp. 173-188.

Binkley, S., E. Kluth, and M. Menaker (1971) Pineal func- tion in sparrows: Circadian rhythms and body temperature. Science, 174r311-314.

Lee, T.M., W.G. Holmes, and I. Zucker (1990) Temperature dependence of circadian rhythms in golden-mantled ground squirrels. J. Biol. Rhythms, 5:25-34.

Lyman, C.P. (1982) Who is who among the hibernators. In: Hibernation and Torpor in Mammals and Birds. C.P. Lyman, J.S. Willis, A. Malan, and L.C.H. Wang, eds. Academic Press, New York: pp. 12-36.

Refinetti, R. (1993a) A functional model of the mammalian circadian pacemaker. Int. J. Biomed. Comput., 32:45-60.

Refinetti, R. (1993b) Comparison of six methods for the de- termination of the period of circadian rhythms. Physiol. Be- hav., 54:869-875.

Refinetti, R., and M. Menaker (1992a) Body temperature rhythm of the tree shrew, !hpaia belangeri. J. Exp. Zool.,

Refinetti, R., and M. Menaker (1992b) The circadian rhythm of body temperature. Physiol. Behav., 51:613-637.

Sokolove, P.G., and W.N. Bushel1 (19783 The chi square periodogram: Its utility for analysis of circadian rhythms.

263A53-457.

animals have more robust rhythms, regardless of J. Theor. Biol., 72:131-160.