Rhythms of Barbiturate-Induced Sleep Time in Deermice Entrained to Non-Twenty-Four Hour Photocycles

Transcription

1 Physiology & Behavior, Vol. 31, pp Pergamon Press Ltd., Printed in the U.S.A. Rhythms of Barbiturate-Induced Sleep Time in Deermice Entrained to Non-Twenty-Four Hour Photocycles NANCY G. FORGER AND RANDY J. NELSON Department of Psychology and Group in Endocrinology University of California, Berkeley, CA Received 27 December 1982 FORGER, N. G. AND R. J. NELSON. Rhythms of barbiturate-induced sleep time in deermice entrained to non-twentyfour hour photocycles. PHYSIOL BEHAV 31(3) , The present experiment examined whether entrainment to twenty-four hour days is advantageous for physiological and behavioral adaptations to a pharmacological challenge. Adult, male deermice (Peromyscus maniculatus) were injected with sodium pentobarbital at 8.5, 14.5, 20.5, 2.5 and again at 8.5 hr after light onset while entrained to a 24 hr day (light/dark (LD) 14:10). Time to recover from anesthesia was recorded. Subsequently, the dark phase of the LD cycle was reduced min every 10 days until a 23 hr photocycle was achieved, (LD 14:9; T=23 hr). Injections of sodium pentobarbital were again administered at 8.5, 14.5, 20.5 and 2.5 hr after light onset and sleep times recorded. This procedure was repeated at T=21 hr (LD 14:7) and T= 18 hr (LD 14:4). A variation in sleep time was observed during entrainment to all T-cycles except T= 18 hr. Sleep time after drug administration was generally longer during the light than during the dark phase of each photocycle. Recovery from anesthesia was not compromised in mice entrained to photocycles with periods of 23 or 21 hr; however, entrainment to very short days disrupted the normal pattern of response to barbiturate administration. Entrainment to a range of daylengths that deviate appreciably from 24 hr is consistent with normal physiological function. Circadian rhythm Entrainment T-Cycle Peromyscus maniculatus Sodium pentobarbital anesthesia CIRCADIAN rhythmicity is a common feature of mammalian physiological organization; twenty-four hour biological rhythms presumably reflect evolutionary adaptation to the daily solar cycle. Although entrainment of circadian rhythms to the external environment is presumed to confer adaptive advantages to organisms (e.g., [1]), there have been few direct tests of this proposition. Normal physiological responses may be contingent on entrainment of endogenous rhythms to a twenty-four hour zeitgeber. The ability of squirrel monkeys to defend core body temperature against a mild cold challenge was compromised when their circadian rhythms were not entrained by external zeitgebers [8,9]. Desynchronization of internal phase relationships occurs in "free-running" squirrel monkeys and in humans exposed to constant light (LL) [3, 8, 9, 16]. Thermoregulatory deficits were eliminated in squirrel monkeys maintained in LL when food was presented on a 24 hr cycle. The feeding schedule apparently synchronized a variety of internal processes that otherwise free-ran and dissociated in LL. Entrainment to a 24 hr day thus appears superior to the free-running condition for effective thermoregulation. The ability to recover from sodium pentobarbitai treatment also appears compromised in the free-running condition. The average duration of anesthesia was prolonged in rats exposed to LL compared to animals entrained to a 12 hr light: 12 hr dark (LD 12:12) cycle [15]. It remains to be determined whether entrainment per se or entrainment to a specific 24 hr zeitgeber is the critical element which restores efficient physiological operations. Efficient physiological coordination most likely is dependent upon optimal phase relationships among various endogenous processes achieved during entrainment to a 24 hr day. For example, longevity of blowflies entrained to non-24 hr photocycles was shorter than flies exposed to normal day lengths [14]. Entrainment to non-24 hr zeitgebers may compromise various homeostatic mechanisms and impair adaptation to environmental perturbations. Thus, it is not necessarily the free-running (non-entrained) condition that is disadvantageous, but any significant deviation from entrainment to 24 hr days. This question was addressed by investigating responsiveness of deermice to pharmacological challenges during entrainment to non-24 hr photocycles. A circadian variation in susceptibility to sodium pentobarbital has been documented for several rodent species entrained to T=24 hr (T=sum of light plus dark periods) [6, 12, 15]. In the present study, deermice were injected with sodium pentobarbital at four different times of the light-dark (LD) cycle during entrainment to various T-cycles (T=24, 23, 21 and 18 hr). If the period of the entraining photocycle has major consequences on physiological systems, then the normal pattern of anesthesia-induced sleeptime may be perturbed during entrainment to non-24 hr days. Copyright 1983 Pergamon Press Ltd /83/

2 380 FORGER AND NELSON METHOD Deermice (Peromyscus maniculatus bairdii) were trapped near Brookings, SD and shipped to Berkeley, CA. Mice used in the study were produced by matings of an F-1 with an F-2 or between two F-2 laboratory-born animals. Nine male deermice were housed individually in cages equipped with running wheels in a cabinet illuminated for 14 hr daily (LD 14:10; lights on at 8.00 hr PST). Purina Mouse Chow and water were available ad lib. Each revolution of the running wheel deflected a pen on chart paper moving at a rate of 45.7 cm per 24 hr. Chart paper was cut into 24 hr segments and assembled with the record for each day pasted beneath the one from the previous day. Charts were labelled and photographically reduced. After three weeks mice were injected IP with sodium pentobarbital (dose=0.056 mg Nembutal per gram of body weight) at 8.5, 14.5, 20.5, 2.5 and again at 8.5 hr after light onset. Injections during the dark phase of the illumination cycle were accomplished under dim red light. Thirty hours separated successive injections. Sleep time was determined by placing injected mice on their backs in individual containers two minutes after injection; the time required for mice to right themselves twice was recorded to the nearest minute. In a few instances when the first and second righting responses were separated by more than 10 rain, the times of the second and third righting responses were recorded. At the end of the first sequence of drug injections the LD cycle was reduced in decrements of min per ten days until a 23 hr photocycle was achieved (LD 14:9; T=23 hr). After two weeks at T=23 hr mice were injected with sodium pentobarbital, as described above, at 8.5, 14.5, 20.5 and 2.5 hr after light onset and sleep times were recorded. This procedure was repeated at T=21 hr (LD 14:7) and T= 18 hr (LD 14:4). At T= 18 hr only three injections were given because 20.5 and 2.5 hr after light onset occurs at the same time. After the third injection at T= 18 hr mice were placed in constant darkness (DD) for three weeks. Daily locomotor activity onset was defined as the time when five minutes or more of sustained wheel running occurred, followed within 30 minutes by additonal wheel running ~>three hours duration. The period (r) of activity onset was calculated by fitting a straight line by eye to seven or more days of wheel running data. Mice were considered entrained to the photoperiod if the absolute value of the period of the animal's activity minus the period of the LD cycle was less than or equal to 10 min (T-T<~10 rain). Two-tailed t-tests for dependent samples were used to analyze differences between mean sleep times. RESULTS All animals entrained to the 24 hr photocycle. The period of the wheel running rhythm for individual mice is shown for T=23, 21, and 18 hr in Table 1. Eight of 9 mice entrained to the T=23 hr photocycle and 6 of the 9 mice met the criterion for entrainment at T=21 hr. Seven of the 8 animals for which data were available entrained to the T=18 hr photocycle. The wheel running record of an animal that entrained to all photocycles is shown in Fig. 1. As seen in this figure, pentobarbital anesthesia did not seem to phase shift or otherwise perturb the animals' activity rhythms. The phase angle of running onset with respect to lights-on did not differ significantly over the four T-cycles. A comparable pattern of results was obtained whether data were restricted to animals entrained to the photocycle at TABLE 1 THE PERIOD OF WHEEL RUNNING ACTIVITY FOR MICE EXPOSED TO NON-24 HR PHOTOCYCLES Animal No. T=23.04 T=21.05 T= " 22.94* * * " *Not entrained; r TrI0 min.?insuffficient data due to equipment malfunction. time of injection, or included all available data. Data from all animals are included in the following analyses. During entrainment to T=24 hr a striking diel variation in sleep time was observed (Fig. 2A). Sleep time was maximal when animals were injected 8.5 hr after light onset (28.2±4 min at L+8.5 hr, and minimal (12.3±1.8 min) for injections at L+14.5 hr (p<0.02). Sleep times at L+20.5 and L+2.5 hr also were longer than those recorded at L hr (p<0.03 in both cases). Mean sleep time after a fifth injection (at L+8.5 hr) was significantly longer than the mean sleep time at L hr, but not different from sleep times following the initial L+8.5 hr injection. Tolerance to sodium pentobarbital apparently did not develop over the first four injections. The rhythm in sleep time persisted while animals were maintained in the T=23 hr photocycle (Fig. 2B). Longest mean sleep time again occurred in the middle of the light phase (L+8.5 hr) and mice recovered from the barbiturate most quickly at the beginning of the dark phase (L hr, p<0.01). Furthermore, mean sleep times were increased at L+2.5 hr relative to those recorded at L hr (p<0.01). The magnitude and pattern of sleep durations were similar during exposure to T=24 and T--23 hr photocycles. The basic pattern of responsiveness to sodium pentobarbital was retained at T=21 hr (Fig. 2C). However, sleep duration at L+8.5 hr was not significantly different from the minimum sleep time at L hr (p =0.09). Sleep durations at T=21 hr did not differ significantly from those observed in the T=24 hr condition for any of the injection times. At T=18 hr the longest mean sleep duration (40.2_+6.9 min) again was observed in the middle of the light phase (L+8.5 hr), and was marginally different from mean sleep time observed at L+14.5 hr (p=0.051). However, sleep times at T= 18 hr generally were longer than those observed at T=24, 23, or 21 hr. Recovery from sodium pentobarbital injected at L hr was more rapid in the T=24 hr than in the T=I8 hr photocycle (12.3±1.8 min vs. 28.8±5.3 min; p<0.01). DISCUSSION A variation in susceptibility to barbiturates was preserved in mice entrained to non-24 hr days. After sodium pentobar-

3 BARBITURATE SENSITIVITY RHYTHMS 381 4"218 HOURS 24 FIG. I. Continuous record of wheel running activity of deermouse No.218 exposed to decreasing T-cycle lengths. Numbers on the left indicate time in days; numbers on th~ right indicate T-cycle duration at time of injection. Mice were exposed to constant darkness (DD) at day 223. Time scale (abscissa) runs from left to right with successive days' activity shown beneath each other. The record has been double plotted to facilitate visualization.

4 382 FORGER AND NELSON c ILl I-" a. I/J c 3 W a. 42- A. T--24h 2s ' 84, 0,, ~ i " S ,6 C. T=21h 42- s4-2e- 18- o ee B. T=23h s 42- ~1 D. T= 18 h 34-2e-,, I 0 m, S 2.S S 2.5 TIME OF INJECTION (Hrs. after light onset) FIG. 2. Mean sleep times (_+standard error) after sodium pentobarbital injection in deermice maintained in different T-cycles. Sleep durations at L hr were compared with those at all other injection times: *p<0.05, **p<0.01, tp= bitai administration, sleep times during the dark phase were shorter than those during the light phase. Entrainment to photocycles deviating by as many as 6 hours from the natural 24 hr day damped, but did not totally disrupt this basic pattern. Successful entrainment to the 18 hr T-cycle by 7 of 8 mice is in itself of interest; the shortest entraining photocycle previously reported forperomyscus was T=23 hr [13]. However, these animals were entrained to very short light pulses (~<1 hr) so the data are not directly comparable. Animals were entrained, rather than passively dirven by the short photocycles; locomotor activity onsets in DD began in phase with the activity onsets of the previous LD cycle. Peromyscus manicu/atus previously entrained only to 24 hr photocycles demonstrate free-running periods of about 23.4 hr when placed in DD [13]. In this study average r for the first several days in DD was 22.6 hr; such a strong after-effect on ~" suggests entrainment to the T-cycle. The pattern of barbiturate-induced sleep durations reported here was in accord with reports on nocturnal rodents entrained to 24 hr photocyles. Sprague-Dawley rats displayed highest survival rates to potentially lethal doses of sodium pentobarbital when injections occurred late in the light phase or early in the dark phase [12]. Similarly, shortest sleep times occurred during the dark phase for several strains ofmus musculus [5] and for Peromyscus manicu/atus rufinus [6]. At T=24, 23, 21 and 18 hr we also recorded shortest sleep times during the dark (active) phase. Body temperature is lowest during the day in nocturnal rodents and hypothermia occurs during barbiturate anesthesia [10,11]. Reduced daytime body temperatures might account for increased sleep times during the light phase ]5]. For example, mice injected in warm ambient temperatures (36 C) or when body temperature was high recovered more quickly than cold animals from sodium pentobarbital administration [5,7]. However, longest barbiturate-induced sleep times occur in rats injected early during the dark phase, when body temperatures are presumably high [15]. The reason for the discrepancy is not clear. Body temperature may be only one of many factors contributing to anesthesia duration. Repeated injection of the same group of animals presents a potential problem of interpretation in the present study and in several of the investigations cited above. Here, injections within a given sequence were separated by at least 24 hr and approximately 7 weeks separated trials at the different T-cycles. Development of tolerance to sodium pentobarbital was not apparent within or between sequences. Rather, sleep times recorded during exposure to T= 18 hr (the final injection sequence) were generally longer than those recorded during earlier sequences. The design of this study leaves open the question of whether this represents an effect of entrainment to very short days, increasing age, or increased sensitivity to sodium pentobarbital. The pattern of sleep times observed at T=21 hr was very similar to that seen at T=24 and T=23 hr. However, because of large variations, sleep times at L+8.5 hr and L+14.5 hr were not significantly different in T=21 hr. During entrainment to T=I8 hr, sleep times at L+8.5 hr were marginally different from those recorded at L+14.5 hr, but the difference at these two times was smaller than the corresponding difference at T=24, 23, or 21 hr. This attenuation of the amplitude of the rhythm is due to the relatively long sleep times recorded at L hr in the T+ 18 hr condition. While entrainment to 23 and 21 hr photocycles preserves the basic variation in barbiturate-induced sleep time, a damping of the rhythm is seen during exposure to 18 hr days. Mus musculus maintained for several weeks in constant light did not display a circadian variability in sleep time after sodium pentobarbitai injection [5]. Damping of the circadian fluctuation was also reported for rats blinded or exposed to LL [15]. However, these animals were injected at fixed clock times without reference to the phase of each animal's freerunning rhythm. The observed damping may be explained by the desynchronization of circadian periods that is likely to occur within a group of free-running animals. In addition to a damping of the rhythm, blind rats and rats exposed to LL also demonstrated longer overall sleep times after a standard pentobarbital injection, compared to animals entrained to a 12:12 photocycle. In the free-running conditions, rats were more susceptible to the anesthetic. Thermoregulation was impaired in squirrel monkeys exposed to LL and normal functioning was restored by providing an external zeitgeber [8,9]. Physiological adaptation in general may be disrupted in the absence of entrainment, presumably because normal phase relations between internal rhythms are abolished. In blowflies, longevity can be influenced by entrainment to T-cycles. Flies entrained to photocycles shorter than 24 hr as well as flies kept in LL died sooner than those exposed to normal day lengths. This led to the hypothesis that 24 hr is the optimal zeitgeber period [14]. In the present study, entrainment to photocycles >21 hr was consistent with a circadian rhythm in induced sleep time. Attenuation of the normal sleep-duration rhythm at T = 18 hr suggests that there is a limit to the capability of entraining

5 BARBITURATE SENSITIVITY RHYTHMS 383 photocycles to sustain "circadian" responses to sodium pentobarbital. Entrainment to some non-24 hr cycles was sufficient to maintain normal responsiveness; synchronization to 24 hr days offered no obvious advantage over synchronization to 23 or 21 hr days. A range of phase relations among relevant physiological rhythms may be sufficient to mediate circadian rhythms in responsiveness to drugs and other stimuli; photocycles quite deviant from 24 hr appear to be adequate entraining cues in this respect. ACKNOWLEDGEMENTS This research was supported by N.I.H. Grant HD-02982; N.G.F. was supported by N.S.F. pre-doctoral fellowship ( ) and R.J.N. was supported by a N.I.M.H. pre-doctoral traineeship (T-32-MH ). We thank Darlene Frost and Thomas Shinder for technical assistance, John Schutz for photography and Laurie Mason for typing the manuscript. We are especially grateful to Irving Zucker for guidance and helpful comments throughout this study. REFERENCES 1. Aschoff, J. Survival value of diurnal rhythms. Syrup Zool Soc London 13: , Aschoff, J. Circadian rhythm of activity and of body temperature. In: Physiological and Behavioral Temperature Regulation, edited by J. D. Hardy, A. P. Gagge and J. A. J. Stolwijk. Springfield: Charles C. Thomas, 1970, pp Aschoff, J., O. Gerecke and R. Wever. Desynchronization of human circadian rhythms. Jpn J Physiol 17: , Daan, S. and C. S. Pittendrigh. A functional analysis of circadian pace-makers in nocturnal rodents. II. The variability of phase response curves. J Comp Physiol 106: , Davis, W. M. Day-night periodicity in pentobarbital response of mice and the influence of socio-psychological conditions. Experientia 18: , Emlen, S. T. and W. Kem. Activity rhythm in Peromyscus: Its influence on rates of recovery from Nembutal. Science 142: , Fuhrman, F. A. The effect of body temperature on the duration of barbiturate anesthesia in mice. Science 105: , Fuller, C. A., F. M. Sulzman and M. C. Moore-Ede. Effective thermoregulation in primates depends upon internal circadian synchronization. Comp Biochem Physiol 63: , Fuller, C. A., F. M. Sulzman and M. C. Moore-Ede. Circadian control of thermoregulation in the squirrel monkey, Saimiri sciureus. Am J Physiol 236: R153-R161, Halberg, F., H. A. Zander, M. W. Houglum and H. R. Nuhlemann. Daily variations in tissue mitoses, blood eosinophils and rectal temperatures of rats. Am J Physiol 177: , Maynert, E. W. In: Drill's Pharmacology in Medicine, 4th ed, edited by J. R. DiPalma. New York: Blakiston, 1971, pp Pauly, J. E. and L. E. Scheving. Temporal variations in the susceptibility of white rats to Pentobarbital Sodium and Tremorine, lnt J Neuropharmocol 3: , Pittendrigh, C. S. and S. Daan. A functional analysis of circadian pace-makers in nocturnal rodents. I. The stability and lability of spontaneous frequency. J Comp Physiol 106: , von Saint Paul, U. and J. Aschoff. Longevity among blowflies Phormia terranenovae R. D. kept in non-24-hour light-dark cycles. J Comp Physiol 127: , Scheving, L. E., D. F. Vedral and J. E. Pauly. A circadian susceptibility rhythm in rats to Pentobarbital Sodium. Ant Rec 160: , Wever, R. Mutual relations between different physiological functions in circadian rhythms in man. J lnterdiscipl Cycle Res 3: , 1972.