- Email: [email protected]
The role of vasopressin in modulating circadian rhythm responses to phase shifts
Peptides, Vol 19, No. 7, pp. 1191–1208, 1998 Copyright © 1998 Elsevier Science Inc. Printed in the USA. All rights reserved 0196-9781/98 $19.00 1 .00
PII S0196-9781(98)00084-9
The Role of Vasopressin in Modulating Circadian Rhythm Responses to Phase Shifts HELEN M. MURPHY,*1 CYRILLA H. WIDEMAN* AND GEORGE R. NADZAM* *Departments of Psychology and Biology, John Carroll University, Cleveland, OH 44118 USA Received 24 December 1997; Accepted 3 April 1998 MURPHY, H. M., C. H. WIDEMAN AND G. R. NADZAM. The role of vasopressin in modulating circadian rhythm responses to phase shifts. PEPTIDES 19(7) 1191–1208, 1998.—Telemetered body temperature (BT), heart rate (HR), and motor activity (AC) data were collected in vasopressin-containing, Long–Evans (LE) and vasopressin-deficient, Brattleboro (DI) rats. In Experiment 1, the rats were initially exposed to a 12 h/12 h light/dark cycle under ad-libitum feeding and were then subjected to either a phase-advance or phase-delay shift of 6 h. After the phase-advance shift, neither strain adapted; however, after the phase-delay shift, both strains adapted rapidly. In Experiment 2, the animals were subjected to either a nocturnal or a diurnal restricted-feeding paradigm and were then exposed to either a phase-advance or phase-delay shift with synchronized feeding. In the nocturnal restricted-feeding paradigms, both strains rapidly adapted to both shifts. Concerning diurnal restrictedfeeding, DI animals readily entrained to the presentation of food in both shifts; whereas, LE animals exhibited a confused rhythmicity. In Experiment 3, animals were subjected to a phase-advance shift, while the time of feeding was held constant. Following the shift, LE animals responded to the onset of the dark at the new time; yet, were still influenced by the presentation of food. The DI animals maintained the preshift circadian pattern and continued to be dominated by the presentation of food. These experiments indicate that circadian rhythms of LE animals are dominated by the light entrainable oscillator (LEO) in ad-libitum feeding and by both the LEO and food entrainable oscillator (FEO) in restricted-feeding. On the other hand, the circadian rhythms of DI animals are dominated by the FEO unless food is provided ad-libitum. The demonstrated role of vasopressin in synchronizing circadian rhythms to the LEO may be of significance in understanding human circadian rhythm disturbances, such as jet lag. © 1998 Elsevier Science Inc. Telemetry Phase-advance Phase-delay Light entrainable oscillator (LEO) Food entrainable oscillator (FEO) Mutual reinforcement Ad-libitum feeding Jet lag Brattleboro rat
THE suprachiasmatic nucleus (SCN) of the hypothalamus has been shown to function as the “executive clock,” regulating circadian rhythms in mammals (29). This master circadian pacemaker is entrained by the light/dark cycle and is often referred to as the light entrainable oscillator (LEO). The pacemaker regulates circadian rhythms directly and also influences rhythms indirectly through the sleep–wake cycle (29). In animal studies, researchers frequently utilize the concept of the “rest-activity cycle” as a reflection of the sleep–wake cycle. Although the SCN is considered to be the master pacemaker, it has
Restricted-feeding
been demonstrated that some circadian rhythms persist after SCN ablation (21). The most extensively studied rhythm that persists after SCN destruction is that involving periodic food availability (4,5,26). Many animals exhibit food-anticipatory activity manifested in behaviors such as wheel running, unreinforced lever pressing, general cage activity, and drinking under daily schedules of restricted food availability (11). Periodic or restricted food availability entrains another oscillator known as the food entrainable oscillator (FEO) and many scientific studies indicate that the FEO is separate and
1 Requests for reprints should be addressed to Dr. Helen M. Murphy, Department of Psychology, John Carroll University, 20700 North Park Blvd., Cleveland, Ohio 44118. E-mail: [email protected]
1191
1192
distinct from the LEO (4,20,21,24,29). Assuming that the LEO and the FEO are separate entities and are entrained by different stimuli, it should be possible to couple or uncouple these two oscillators. In nocturnal animals, (e.g., rats) eating takes place primarily during the dark, even if food is available ad-libitum. Light/dark alternation is the most powerful synchronizer of circadian rhythms in ad-libitum feeding (18). Some circadian rhythms, however, undergo phase shifts when rats are presented with food during the light phase only and feeding is limited to 1– 4 h. For example, phase shifts are observed in activity (3,12,13,14,15), body temperature (7,8,14,15), and heart rate (15) rhythms under such a feeding regimen. A group of vasopressin-containing neurons has been found in the SCN (23), which function as output efferent neurons (32) and show a clearly-defined circadian rhythm in vasopressin levels (16). In an attempt to elucidate the functional role of vasopressin within the SCN, one study measured wheel running in the hamster following the microinjection of vasopressin into the SCN region. It was demonstrated that vasopressin did not affect the circadian timing system (1). Later publications, however, proposed that vasopressin may operate as a neurotransmitter within the SCN, either as a component of the circadian clock mechanism itself or as a part of the output pathway to other neural structures (19,32). Thus, the role of vasopressin in circadian rhythms is controversial. In order to gain insight into the effect of vasopressin on circadian rhythms, some researchers have utilized the Brattleboro rat (9,14,15,17). The Brattleboro rat lacks the ability to synthesize hypothalamic vasopressin (31), is a genetic mutant of the Long–Evans (LE) strain, exhibits hereditary diabetes insipidus, and is designated as a DI rat (30). Peterson et al. (17) demonstrated that the absence of vasopressin from the SCN did not change components of the endocrine (serotonin N-acetyltransferase levels in the pineal) and behavioral (wheel running and drinking) rhythms examined in female DI rats. These researchers concluded that, in the rat, the maintenance of circadian rhythmicity is not dependent upon the presence of vasopressin-containing neurons in the SCN. In another experiment, Groblewski et al. (9) examined circadian rhythms in cage activity and drinking behavior of female DI rats and proposed that the generation of circadian rhythms is not dependent upon central vasopressin. In the two studies just cited that found no differences between DI and LE animals, the number of hours in the light/dark cycle was manipulated and food and water were provided ad-libitum to the subjects. Another study examined 24 h pineal melatonin synthesis in male DI, LE, and Sprague–Dawley rats. Even though the results showed slight differences among these animals, the authors stated that their data “speaks against” an important role for vasopressin in the modification of circadian rhythms (22). These experimenters did make the
MURPHY ET AL.
point, however, that vasopressin may be involved in a more subtle manner in circadian rhythmicity which may not be readily apparent under standard lighting conditions. Contrary to the above findings, our laboratory has demonstrated the importance of vasopressin in circadian rhythms (14,15). In one of the studies, we examined the interaction of vasopressin and the photic oscillator in male DI and LE rats utilizing telemetry (15). In recent years, telemetry has been employed in circadian rhythm studies because it has been shown to be a very suitable technique for the continuous and undisturbed measurement of body temperature, heart rate, and motor activity (6,10,27,28). In our experiment, we demonstrated that with ad-libitum feeding and nocturnal restricted-feeding (two separated scheduled-feeding periods in the dark cycle), natural nocturnal cycles of body temperature, heart rate, and motor activity were maintained in both DI and LE animals. Marked changes, however, developed with diurnal restricted-feeding (two separated scheduled-feeding periods in the light cycle). With DI animals, the influence of the photic oscillator was lost and body temperature, heart rate, and motor activity shifted from nocturnal to diurnal patterns, reflecting the dominance of the food presentation in these vasopressindeficient animals. The LE animals, on the other hand, lost a well-defined circadian rhythmicity resulting from adherence to the photic oscillator, while at the same time being influenced by the presentation of food. We concluded that vasopressin has a significant interaction with the photic oscillator, which is obvious when the photic and nonphotic oscillators are uncoupled. In addition, we proposed that the strength of the photic oscillator was decreased or that the effect of this oscillator was masked in vasopressin-deficient rats compared to vasopressin-containing rats. In view of the results obtained in the above study, we asked the following questions: 1) Could vasopressin be significantly involved in circadian rhythm disturbances, particularly those concerned with phase shifts? (In circadian rhythm disturbances, natural cycles of rest-activity and feeding are disrupted, resulting in conflict between biological and environmental cues.) 2) Would the use of phaseadvance or phase-delay regimens and ad-libitum or restricted-feeding (nocturnal or diurnal) schedules reveal more information about how vasopressin interacts with the LEO and/or FEO in regulating rest-activity cycles? In the first experiment, we sought to determine whether or not a phase shift paradigm in which the dark cycle is either advanced or delayed would produce different effects under ad-libitum feeding. EXPERIMENT 1 Method The subjects utilized were 12 male homozygous DI rats and 12 male LE rats (Harlan–Sprague Dawley, Inc., Indianapolis, IN). Animals were 6 weeks of age and weighed between
VASOPRESSIN AND CIRCADIAN PHASE SHIFTS
100 –110 grams at the beginning of the experiment. Subjects were housed in individual translucent polypropylene cages with the room temperature maintained at 23 6 1°C and employed a 12 h/12 h light/dark cycle. Utilizing sterile techniques, all animals were anesthetized with sodium pentobarbital (4 mg or 5 mg/100 g body weight for DI or LE rats, respectively) and a biotelemetry transmitter (Model CTA-F40, Data Sciences, Inc., St. Paul, MN), capable of measuring body temperature (BT), electrocardiogram, and motor activity (AC), was implanted subcutaneously. Implant signals were transmitted to a personal computer that was equipped with receivers and a multiplexer (Model CTR-86 Receiver and Model MP-12 Multiplexer). The receivers converted the electrocardiogram waveform to heart rate (HR). Individual receivers were connected to a multiplexer where BT, HR, and AC information was sequentially passed to the data acquisition system. Data, which were collected every 5 min for the duration of the experiment, were stored on the hard disk of the personal computer for later analysis. Following the implantation of transmitters, the animals were subdivided into two groups of 6 DI and 6 LE rats each. Both groups were provided with food and water ad-libitum throughout the experiment. The animals had an habituation period of 14 days. Following the habituation period, 6 animals from each strain were subjected to a light period shortened by 6 h (phase-advance). The other 6 animals from each strain were subjected to a dark period lengthened by 6 h (phase-delay). The new light/dark cycle pattern was continued for 10 days. RESULTS Figure 1 summarizes the circadian rhythms of BT, HR, and AC for DI and LE animals under the phase-advance paradigm. The circadian rhythms were similar for DI and LE animals during both preshift and postshift conditions. During the preshift condition, all animals displayed a rhythm typical of nocturnal animals, namely, higher BT, HR, and AC during the dark period than during the light period. During the postshift period of the phase-advance paradigm, both strains of animals showed a strong adherence to the old light/dark cycle; the circadian rhythms became weaker; and the duration of AC became more persistent. The LE animals developed a modest peak with the onset of the new dark phase. Figure 2 summarizes the circadian rhythms of BT, HR, and AC for DI and LE animals under the phase-delay paradigm. The circadian rhythm was similar for DI and LE animals during both preshift and postshift conditions. During the preshift condition, all animals displayed a rhythm typical of nocturnal animals. During the phase-delay postshift period, even though both strains of animals synchronized to the new light/dark cycle, all animals initially showed anticipation of the old dark phase.
1193
Figure 3 displays actograms for AC for DI (A and C) and LE (B and D) animals under ad-libitum phase-advance (top) and phase-delay (bottom) paradigms. The figure shows that with phase-advance, both strains maintained a strong adherence to the old light/dark cycle; the LE animals developed a response to the new dark phase; and both strains developed more persistent AC. With phase-delay, both strains synchronized to the new light/dark cycle. Although the DI animals gradually adapted to the new schedule, the figure clearly shows that the LE animals adjusted immediately. In addition, all animals were instantly responsive to the care artifact in both phase-advance and phase-delay paradigms. EXPERIMENT 2 With the exception of entrainment of the photic oscillator to changing photic cues (phase shifts) by vasopressin-containing animals during ad-libitum feeding, one would conclude that vasopressin has little or no role to play in the generation or maintenance of circadian rhythms. In many aspects, the vasopressin-deficient animals resembled their vasopressincontaining counterparts when there was a 6-h phase-advance or phase-delay of the circadian cycle. However, from previous experiments, we have demonstrated that scheduled-feeding (FEO) presented during the light cycle, but not the dark cycle, has a significant role to play in modulating circadian rhythmicity in vasopressin-containing and vasopressin-deficient rats (14,15). As a result, we hypothesized that phase shift paradigms similar to the ones presented in Experiment 1 of this study would differentially effect vasopressin-containing and vasopressin-deficient animals if scheduled-feeding would be introduced. Thus, a second experiment was devised to delineate the differences between the two strains when the LEO and FEO were coupled or uncoupled and the phase shift paradigms of Experiment 1 were introduced. Method In this Experiment 24 DI rats and 24 LE rats were utilized. The same parameters for age, weight, and housing conditions employed in Experiment 1 were used in this experiment. The telemetry methodology was also the same as Experiment 1. The animals were subdivided into four groups of 6 DI and 6 LE rats each. Groups 1 and 2 were fed at the second hour of the dark cycle and groups 3 and 4 were fed at the second hour of the light cycle. The amount of food presented during each feeding for all groups was 10% of the body weight of the animal calculated from the previous day. (The 10% value was utilized because we had determined that animals could gain weight and survive throughout the experiment with such a feeding regimen.) Thus, all animals were food restricted and the feeding regimen was maintained throughout the experiment. Water was available adlibitum for all animals. Groups 1 and 3 were phase advanced
1194
MURPHY ET AL.
FIG. 1. Mean circadian rhythms of BT, HR, and AC for 6 DI and 6 LE rats under the phase-advance paradigm. The graphs show 2 days preshift, followed by the day of shift, and 5 days postshift under the ad-libitum feeding paradigm. The figure depicts a 24 point moving average of the 5 min data points. The darkened lines adjacent to the x-axis, represent the dark phase of the light/dark cycle. The short lines close to the x-axis, indicate the time of care and handling of animals. The long dark line above the care and handling lines, which extends for the duration of the experiment, indicates ad-libitum feeding.
VASOPRESSIN AND CIRCADIAN PHASE SHIFTS
1195
FIG. 2. Mean circadian rhythms of BT, HR, and AC for DI and LE rats under the phase-delay paradigm. All other graphic characteristics are the same as Figure 1.
1196
MURPHY ET AL.
FIG. 3. Mean circadian double-plotted actograms of AC for 6 DI (A) and 6 LE (B) rats subjected to ad-libitum feeding under phase-advance (top) and for 6 DI (C) and 6 LE (D) rats subjected to ad-libitum feeding under phase-delay (bottom) paradigms. The darkened lines adjacent to the x-axis (top and bottom) represent the dark phase of the light/dark cycle. The short lines close to the x-axis (top and bottom) indicate the time of care and handling of animals.
VASOPRESSIN AND CIRCADIAN PHASE SHIFTS
and groups 2 and 4 were phase delayed according to the paradigms utilized in Experiment 1. All animals had an habituation period of 14 days on their respective feeding regimens. Following the habituation period, the animals were subjected to their respective phase shifts. Feeding was shifted in synchrony with the light/dark cycle, that is, at the beginning of the second hour of the dark or light cycle. The new light/dark cycle pattern was continued for 10 days. Results Figure 4 summarizes the circadian rhythms of BT, HR, and AC for DI and LE animals under the nocturnal food-restricted, phase-advance paradigm. With the preshift condition, BT, HR, and AC were robust. The circadian rhythm patterns were those of nocturnal animals. During the postshift condition, the circadian rhythms remained strong and shifted immediately. Figure 5 summarizes the circadian rhythms of BT, HR, and AC for DI and LE animals under the nocturnal, foodrestricted phase-delay paradigm. With the preshift condition, BT, HR, and AC were robust. The circadian rhythm patterns were those of nocturnal animals. During the postshift condition, the circadian rhythms remained strong and shifted gradually for BT and AC and immediately for HR. In addition, even though both strains synchronized to the new light/dark cycle, all animals initially showed anticipation for the old dark phase and/or old time of food presentation. Figure 6 displays actograms for AC for DI (A and C) and LE (B and D) animals under nocturnal, food-restricted phase-advance (top) and phase-delay (bottom) paradigms. The figure shows that with phase-advance, the shift was immediate for both DI and LE animals as evidenced by increased AC at the onset of the new dark phase. For phase-delay, the shift was gradual for both DI and LE animals. Figure 7 depicts the circadian rhythms of BT, HR, and AC for DI and LE animals under the diurnal, food-restricted phase-advance paradigm. With the preshift condition, LE animals showed a confused rhythm characterized by peaks in both the dark and light phases in BT, HR, and AC. The DI animals had definite peaks in the light and troughs in the dark in BT and HR. The AC rhythm was somewhat confused in DI animals. During the postshift condition, the LE animals had a weak, confused rhythm in BT, HR, and AC; whereas, DI animals had a single, strong diurnal rhythm of BT, HR, and AC. Figure 8 gives the circadian rhythms of BT, HR, and AC for DI and LE animals under the diurnal, food-restricted phase-delay paradigm. With the preshift condition, LE animals showed a confused rhythm characterized by peaks in both the dark and light phases in BT, HR, and AC. The DI animals had a single, strong diurnal rhythm of BT, HR, and
1197
AC. During the postshift condition, the LE animals had weak rhythms in BT, HR, and AC; but, were synchronized with the strong diurnal rhythms of BT and AC of DI animals. Figure 9 displays actograms for AC for DI (A and C) and LE (B and D) animals under diurnal, food-restricted phaseadvance (top) and phase-delay (bottom) paradigms. The figure illustrates that with phase-advance, although LE animals showed peaks during both the dark and light phases, the shift of AC was immediate and coincided with the onset of the new dark period and the time of food presentation. On the other hand, the DI animals lost their nocturnal peaks and the shift of AC was immediate and coincided with the time of feeding. For phase-delay, although the LE animals shifted immediately to the presentation of food, strong peaks of AC were also immediately apparent with the onset of the dark period. The DI animals, however, lost their nocturnal peaks of AC and shifted their AC to coincide with the presentation of food. EXPERIMENT 3 In Experiment 2, the LEO and FEO were either tightly coupled or completely uncoupled and the photic and nonphotic zeitgebers were shifted in synchrony. Following the analysis of data obtained from the this experiment, the question arose: What if the two oscillators were weakly coupled and the photic zeitgeber was then phase advanced, while the nonphotic zeitgeber was held constant? The answers to this question would help to elucidate the relative dominance of the two oscillators in the two strains of animals, as well as provide insight into the phenomenon of coupling. Thus, a third experiment was conducted to provide such answers. Method In Experiment 3, 6 DI rats and 6 LE rats were utilized. The same parameters for age, weight, housing, and telemetry utilized in Experiments 1 and 2 were employed. During the habituation period of 14 days the animals were fed 10% of their body weight from the previous day at the third hour of the dark phase. Following the habituation period, the animals were subjected to a 6 h phase-advance shift. Feeding was not shifted. Thus, animals now received food at the beginning of the ninth hour of the dark phase. The new light/dark cycle pattern was continued for 10 days. Results Figure 10 summarizes the circadian BT, HR, and AC for DI and LE animals under nocturnal phase-advance with the time of food presentation remaining constant. The overall preshift circadian rhythms of BT, HR, and AC were similar for DI and LE animals in that they reflected patterns of nocturnal animals. The LE animals responded immediately to the onset of the dark through increased BT, HR, and AC;
1198
MURPHY ET AL.
FIG. 4. Mean circadian rhythms of BT, HR, and AC for 6 DI and 6 LE rats under the nocturnal food-restricted, phase-advance paradigm. The figure depicts a 24 point moving average of the 5 min data points. The darkened lines adjacent to the x-axis, represent the dark phase of the light/dark cycle. The short lines close to the x-axis, indicate the time of care and handling of animals. The short dark lines above the care and handling lines, indicate the time of feeding.
VASOPRESSIN AND CIRCADIAN PHASE SHIFTS
1199
FIG. 5. Mean circadian rhythms of BT, HR, and AC for DI and LE rats under the nocturnal food-restricted, phase-delay paradigm. All other graphic characteristics are the same as Figure 4.
1200
MURPHY ET AL.
FIG. 6. Mean circadian double-plotted actograms of AC for 6 DI (A) and 6 LE (B) rats subjected to nocturnal food-restricted feeding under phase-advance (top) and for 6 DI (C) and 6 LE (D) rats subjected to nocturnal food-restricted feeding under phase-delay (bottom) paradigms. The darkened lines adjacent to the x-axis (top and bottom) represent the dark phase of the light/dark cycle. The short lines close to the x-axis (top and bottom) indicate the time of feeding, as well as care and handling of animals.
VASOPRESSIN AND CIRCADIAN PHASE SHIFTS
1201
FIG. 7. Mean circadian rhythms of BT, HR, and AC for 6 DI and 6 LE rats under the diurnal food-restricted, phase-advance paradigm. All other graphic characteristics are the same as Figure 4.
1202
MURPHY ET AL.
FIG. 8. Mean circadian rhythms of BT, HR, and AC for DI and LE rats under the diurnal food-restricted, phase-delay paradigm. All other graphic characteristics are the same as Figure 4.
VASOPRESSIN AND CIRCADIAN PHASE SHIFTS
1203
FIG. 9. Mean circadian double-plotted actograms of AC for 6 DI (A) and 6 LE (B) rats subjected to diurnal food-restricted feeding under phase-advance (top) and for 6 DI (C) and 6 LE (D) rats subjected to diurnal food-restricted feeding under phase-delay (bottom) paradigms. All other graphic characteristics are the same as Figure 6.
1204
MURPHY ET AL.
FIG. 10. Mean circadian rhythms of BT, HR, and AC for 6 DI and 6 LE rats under the nocturnal food-restricted, phase-advance paradigm with the time of feeding held constant. All other graphic characteristics are the same as Figure 4.
VASOPRESSIN AND CIRCADIAN PHASE SHIFTS
1205
FIG. 11. Mean circadian double-plotted actograms of AC for 6 DI (A) and 6 LE (B) rats subjected to nocturnal food-restricted feeding under the phase-advance paradigm with the time of feeding held constant for preshift and postshift periods. All other graphic characteristics are the same as Figure 6.
but, the influence of food presentation was also apparent as manifested by maintenance of heightened physiological responses. On the other hand, the circadian rhythms of the DI animals were strongly controlled by presentation of food as observed by a sharp peak at the time of food presentation. By Day 1 of the postshift condition, the LE animals responded to the onset of dark at the new time; yet, still were influenced by the presentation of food. Despite the 6 h phase-advance, DI animals maintained the preshift circadian pattern and continued to be dominated by the presentation of food. Figure 11 displays actograms for AC for DI (A) and LE (B) animals under nocturnal, food-restricted phase-advance with the time of food presentation remaining constant. The figure shows that LE animals responded immediately to the new dark period as evidenced by increased levels of AC; while still maintaining peaks of AC coinciding with the presentation of food. The DI animals, however, lacked peaks of AC in the new dark phase until Day 6. They continued to show a peak of AC during feeding. DISCUSSION In Experiment 1, which utilized the ad-libitum feeding paradigm, the circadian rhythm of BT, HR, and AC in DI and LE animals were similar during the preshift condition.
During the phase-advance postshift condition, both strains maintained a strong adherence to the old light/dark cycle; whereas, during the phase-delay postshift condition, both strains readily adapted to the phase shift. With ad-libitum feeding, the LEO was dominant and the timing of the sleep–wake cycle (rest-activity cycle) was in the proper phase relationship with this master circadian pacemaker during the preshift condition. With the phase-advance shift, the light/dark cycle was manipulated and there was a dissociation between the LEO and the timing of rest. The rest (light) period was shortened by 6 h which led to a high degree of internal temporal disorganization within the animals. Both strains of rats displayed decreased BT, HR, and AC at the beginning of the new dark cycle and increased BT, HR, and AC at the beginning of the new light cycle. As a result, adherence to the old light/dark cycle led to a circadian rhythm that was maladapted to the new environmental cue. On the other hand, with the phase-delay shift, which involved manipulation of the light/dark cycle through a six hour extension of the dark phase, there was no dissociation between the LEO and the timing of rest. The dark period was merely lengthened by 6 h which was followed by a normal rest (light) period. This did not lead to a high degree of internal temporal disorganization within the animals. Both strains of rats displayed normal BT, HR, and AC
1206
at the beginning of the new dark and light cycles. As a result, adaptation to the light/dark cycle led to a circadian rhythm that was in synchrony with the new environmental cue. The phase-advance paradigm is indicative of an eastward jet lag phenomenon with the associated maladaptive responses. On the other hand, the phase-delay (westward) paradigm is without any indication of jet lag illness, but shows an anticipatory response to the dark phase. In Experiment 2, with the introduction of food restriction, another oscillator was unmasked (FEO). In nocturnal restricted-feeding when the LEO and FEO were coupled, no significant differences were noted between vasopressin-deficient and vasopressin-containing animals when they were maintained under a normal 12 h/12 h light/dark cycle or were phase advanced by 6 h. With nocturnal restrictedfeeding, both the LEO and FEO were functional and were coupled. Under the preshift condition, the timing of the rest-activity cycle was in the proper phase relationship with the coupled LEO and FEO. With the shift (6-h phaseadvance), both the LEO and FEO were shifted in synchrony. In this case, the light/dark cycle and rest-activity cycle rapidly realigned and synchronized, since the oscillators mutually reinforced one another. The reason that an adaptive circadian rhythmicity developed immediately, may be due to entrainment of the SCN pacemaker by food-restriction schedules (25). This entrainment of the LEO by the FEO occurs only when the periods of the pacemaker and the restriction schedule are very similar (2,25). In the present experiment, there was only a 1-h difference between the onset of the dark cycle and the presentation of food. Consequently, this could explain the differences obtained in the postshift condition of nocturnal restricted-feeding versus ad-libitum feeding of Experiment 1, where the FEO was nonfunctional since the nonphotic zeitgeber (restrictedfeeding) was absent and, thus, the oscillators were unable to reinforce one another. One way of assessing the relative dominance of the LEO versus the FEO and the role of vasopressin in influencing each oscillator is to uncouple the two oscillators and observe the ensuing rhythms. This was achieved by introducing a diurnal restricted-feeding paradigm. In this paradigm, significant differences were noted between DI and LE animals in BT, HR, and AC. In general, during the preshift condition, the DI animals had a reversal of peaks and troughs of these physiological variables from those observed in nocturnal restricted-feeding. The LE animals, on the other hand, had a confused rhythmicity with major nocturnal peaks evident in addition to the diurnal peak associated with food presentation. In the postshift phaseadvance and phase-delay conditions, the DI animals rapidly adapted to the phase shift and developed a definite circadian rhythmicity characterized by peaks associated with food presentation in the light and troughs in the dark. The LE animals, however, did not develop a definite rhythmicity
MURPHY ET AL.
(characterized by sustained peaks and troughs) but showed significant peaks which coincided with the onset of the new dark period, as well as the presentation of food. Under preshift and postshift conditions, the absence of vasopressin enabled the DI animals to adapt and develop a definite circadian rhythmicity governed by the FEO. In contrast to nocturnal-restricted feeding, where the LEO and FEO were coupled, diurnal restricted-feeding resulted in an uncoupling of the LEO and FEO. As a consequence of this uncoupling, the circadian rhythms of DI animals synchronized to the FEO and the response to the LEO was lost in the preshift condition. Because of this, the rest-activity cycle was governed by the FEO. On the other hand, the circadian rhythms of LE animals were influenced by both the LEO and FEO. Since the oscillators were uncoupled and since LE animals were strongly under the influence of the LEO (15), there was a dissociation between the master clock and the timing of rest due to the presentation of food when the animals would normally be sleeping. This led to a high degree of internal temporal disorganization, resulting in confused rhythmicity in BT, HR, and AC. In the postshift condition, the circadian rhythms of DI animals immediately shifted and synchronized to the presentation of food. Thus, the FEO, once again, governed the rest-activity cycle which was characterized by sustained peaks and troughs. In contrast, the LE animals, governed by the LEO and the FEO, exhibited a continuation of internal temporal disorganization resulting in a disruption of the rest-activity cycle. This disruption of the rest-activity cycle was manifested by confused rhythms of BT, HR, and AC which lacked sustained peaks and troughs that are characteristic of a normal restactivity cycle. In Experiment 3, where the LEO and FEO were weakly coupled in the preshift period (restricted feeding at the third hour of the dark cycle) differences were noted between DI and LE animals in that DI animals were almost exclusively attuned to the nonphotic zeitgeber; whereas, LE animals were influenced by both the photic and nonphotic zeitgebers. After the 6-h phase-advance shift with the feeding time being held constant (now nine hours into the dark cycle), DI animals continued to be dominated by the nonphotic zeitgeber and showed a circadian rhythm almost identical with that observed in the preshift condition. The LE animals, on the other hand, responded to the new photic cue while maintaining a response to the nonphotic zeitgeber. Although the restricted feeding at the third hour of the dark cycle began to uncouple the oscillators, the 6-h phase-advance of the light/dark cycle enhanced the separation of the zeitgebers; thus, further uncoupling the oscillators. Because of the strong influence of the LEO on LE animals, this uncoupling produced a disruption of the rest component of the restactivity cycle. There was now a dissociation between the LEO and the timing of rest in LE rats which led to a high degree of internal disorganization within the animals. On
VASOPRESSIN AND CIRCADIAN PHASE SHIFTS
1207
the other hand, because DI animals are dominated by the FEO, there was no disruption of the rest-activity cycle. In conclusion, the present studies help to unravel some of the mystery surrounding the role of vasopressin in circadian rhythms. Vasopressin-containing nocturnal animals are governed by the LEO when fed ad-libitum and strongly governed by the combined LEO and FEO during restricted feeding. The vasopressin-deficient animals, however, are governed by the FEO unless food is provided ad-libitum. Finally, the findings of the present study concerning the role of vasopressin in synchronizing various circadian rhythms of the organism to the LEO may be of importance in understanding human disturbances, such as jet lag. In this circadian rhythm disturbance there is a dissociation between the master pacemaker (LEO) and the timing of sleep. In jet
lag, crossing time zones can cause a person to fall asleep during the normally active period (day) and to wake during the rest period (night). As a result, a person’s circadian rhythms are disturbed and many days may pass before the body can readjust to the new time zone. This is particularly evident when flying eastward (phase-advance) as compared to flying westward (phase-delay). Experiment 3 of the present study provides an animal model for the disturbances encountered in eastward jet lag and evidence for the role of vasopressin in this circadian rhythm disorder. ACKNOWLEDGEMENTS This research was supported by a Focused Giving Grant from Johnson & Johnson to Helen M. Murphy and Cyrilla H. Wideman.
REFERENCES 1. Albers, H. E.; Ferris, C. F.; Leeman, S. E.; Goldman, B. D. Avian pancreatic polypeptide shifts hamster circadian rhythms when microinjected into the suprachiasmatic region. Science (Washington DC). 223:833– 835; 1984. 2. Aschoff, J.; von Goetz, C. Effects of feeding cycles on circadian rhythms in squirrel monkeys. J. Biol. Rhythms 1:267– 276; 1986. 3. Baker, R. A. A periodic feeding behavior in the albino rat. J. Comp. Physiol. Psychol. 46:422– 426; 1953. 4. Boulos, Z.; Rosenwasser, A. M.; Terman, M. Feeding schedules and the circadian organization of behavior in the rat. Behav. Brain Res. 1:39 – 65; 1980. 5. Clarke, J. D.; Coleman, G. J. Persistent meal-associated rhythms in SCN-lesioned rats. Physiol. Behav. 36:105–113; 1986. 6. Clement, J. G.; Mills, P.; Brockway, B. Use of telemetry to record body temperature and activity in mice. J. Pharmacol. Methods 21:129 –140; 1989. 7. Ehret, C. F.; Dobra, K. W. The oncogenic implications of chronobiotics in the synchronization of mammalian circadian rhythms: Barbiturates and methylated xanthines. In: Nieburgs, H. E., Ed. Proceedings of the Third International Symposium on the detection and prevention of cancer. New York: Marcel Dekker; 1977:1101–1114. 8. Ehret, C. F.; Groh, K. R.; Meinert, J. C. Circadian dyschronism and chronotypic ecophilia as factors in aging and longevity. In: Samis, H. V.; Capobianco, S., Eds. Aging and biological rhythms. New York: Plenum Publishing Corp.; 1978:185–213. 9. Groblewski, T. A.; Nunez, A. A.; Gold, R. M. Circadian rhythms in vasopressin deficient rats. Brain Res. Bull. 6:125– 130; 1981. 10. Kramer, K.; VanAcker, S. A. B. E.; Voss, H. P.; Grimbergen, J. A.; VanderVijgh, W. J. F.; Bast, A. Use of telemetry to record electrocardiogram and heart rate in freely moving mice. J. Pharm. Tox. Methods. 30:209 –215; 1993. 11. Mistleberger, R. E. Circadian food-anticipatory activity: formal models and physiological mechanisms. Neurosci. Biobehav. Rev. 18:171–195; 1994. 12. Mouret, J. R.; Bobiller, P. Diurnal rhythms of sleep in the rat: Augmentation of paradoxical sleep following alterations of the feeding schedule. Int. J. Neurosci. 2:265–270; 1971. 13. Mouret, J. R.; Doindet, J.; Chouvet, G. Circadian rhythms of
14. 15. 16.
17. 18. 19.
20. 21.
22. 23. 24. 25. 26.
sleep in the rat: Importance of the feeding schedule. Int. J. Chronobiol. 1:345; 1973. Murphy, H. M.; Wideman, C. H.; Nadzam, G. R. Vasopressin deficiency and circadian rhythms during food-restriction stress. Peptides. 14:1215–1220; 1993. Murphy, H. M.; Wideman, C. H.; Nadzam, G. R. The interaction of vasopressin and the photic oscillator in circadian rhythms. Peptides. 17:467– 475; 1996. Noto, T.; Hashimoto, H.; Doi, Y.; Nawajima, T.; Kato, N. Biorhythms of arginine-vasopressin in the paraventricular, supraoptic and suprachiasmatic nuclei of rats. Peptides. 4:875– 878; 1993. Peterson, G. M.; Watkins, W. B.; Moore, R. Y. The suprachiasmatic hypothalamic nuclei of the rat. Behav. Neural. Biol. 29:236 –245; 1980. Reinberg, A. Chronobiology and nutrition. In: Reinberg, A.; Smolensky, M. H., Eds. Biological rhythms and medicine. New York: Springer–Verlag; 1983:265–300. Reppert, S. M. Circadian rhythm of cerebrospinal fluid vasopressin: Characterization and physiology. In: Schrier, R. W., Ed. Vasopressin. New York: Raven Press; 1985: 455– 464. Rosenwasser, A. M.; Pelchat, R. J.; Adler, N. T. Memory for feeding time: Possible dependence on coupled circadian oscillators. Physiol. Behav. 32:25–30; 1984. Rusak, B. Biological rhythms: from physiology to behavior. In: Montplaisir, J.; Godbout, R., Eds. Sleep and biological rhythms. New York: Oxford University Press; 1990:11–24. Schro¨der, H.; Stehle, J.; Henschel, M. Twenty-four-hour pineal melatonin synthesis in the vasopressin-deficient Brattleboro rat. Brain Res. 459:328 –332; 1988. Sofroniew, M. V.; Weindl, A. Projections from the parvocellular vasopressin- and neurophysin-containing neurons of the suprachiasmatic nucleus. Am. J. Anat. 153:391– 430; 1978. Stephan, F. K. Limits of entrainment to periodic feeding in rats with suprachiasmatic lesions. J. Comp. Physiol. 143:401– 410; 1981. Stephan, F. K. The role of period and phase in interactions between feeding- and light-entrainable circadian rhythms. Physiol. Behav. 36:151–158; 1986. Stephan, F. K.; Swann, J. M.; Sisk, C. L. Anticipation of 24-hr feeding schedules in rats with lesions of the suprachiasmatic nucleus. Behav. Biol. 25:346 –363; 1979.
1208 27. Teelink, J. R.; Clozel, J. P. Hemodynamic variability and circadian rhythm in rats with heart failure: role of locomotor activity. Am. J. Physiol. 264:H2111–H2118; 1993. 28. Tornatzky, B. H. C.; Miczek, K. A. Long-term impairment of autonomic circadian rhythms after brief intermittent social stress. Physiol. Behav. 53:983–993; 1993. 29. Turek, F. W. Circadian rhythms. In: Bardin, C. W., Ed. Recent progress in hormone research, vol. 49. San Diego: Academic Press, Inc.; 1994:43–90. 30. Valtin, H.; Schroeder, H. A. Familial hypothalamic diabetes
MURPHY ET AL. insipidus in rats (Brattleboro strain). Am. J. Physiol. 206:425– 430; 1964. 31. Valtin, H.; Sawyer, W. H.; Sokol, H. W. Neurohypophysial principles in rats homozygous and heterozygous for hypothalamic diabetes insipidus (Brattleboro strain). Endocrinology. 77:701–706; 1965. 32. Watts, A. C.; Swanson, L. W. Efferent projections of the suprachiasmatic nucleus: II. Studies using retrograde transport of fluorescent dyes and simultaneous peptide immunoreactivity in the rat. J. Comp. Neurol. 258:230 –252; 1987.
PII S0196-9781(98)00084-9
The Role of Vasopressin in Modulating Circadian Rhythm Responses to Phase Shifts HELEN M. MURPHY,*1 CYRILLA H. WIDEMAN* AND GEORGE R. NADZAM* *Departments of Psychology and Biology, John Carroll University, Cleveland, OH 44118 USA Received 24 December 1997; Accepted 3 April 1998 MURPHY, H. M., C. H. WIDEMAN AND G. R. NADZAM. The role of vasopressin in modulating circadian rhythm responses to phase shifts. PEPTIDES 19(7) 1191–1208, 1998.—Telemetered body temperature (BT), heart rate (HR), and motor activity (AC) data were collected in vasopressin-containing, Long–Evans (LE) and vasopressin-deficient, Brattleboro (DI) rats. In Experiment 1, the rats were initially exposed to a 12 h/12 h light/dark cycle under ad-libitum feeding and were then subjected to either a phase-advance or phase-delay shift of 6 h. After the phase-advance shift, neither strain adapted; however, after the phase-delay shift, both strains adapted rapidly. In Experiment 2, the animals were subjected to either a nocturnal or a diurnal restricted-feeding paradigm and were then exposed to either a phase-advance or phase-delay shift with synchronized feeding. In the nocturnal restricted-feeding paradigms, both strains rapidly adapted to both shifts. Concerning diurnal restrictedfeeding, DI animals readily entrained to the presentation of food in both shifts; whereas, LE animals exhibited a confused rhythmicity. In Experiment 3, animals were subjected to a phase-advance shift, while the time of feeding was held constant. Following the shift, LE animals responded to the onset of the dark at the new time; yet, were still influenced by the presentation of food. The DI animals maintained the preshift circadian pattern and continued to be dominated by the presentation of food. These experiments indicate that circadian rhythms of LE animals are dominated by the light entrainable oscillator (LEO) in ad-libitum feeding and by both the LEO and food entrainable oscillator (FEO) in restricted-feeding. On the other hand, the circadian rhythms of DI animals are dominated by the FEO unless food is provided ad-libitum. The demonstrated role of vasopressin in synchronizing circadian rhythms to the LEO may be of significance in understanding human circadian rhythm disturbances, such as jet lag. © 1998 Elsevier Science Inc. Telemetry Phase-advance Phase-delay Light entrainable oscillator (LEO) Food entrainable oscillator (FEO) Mutual reinforcement Ad-libitum feeding Jet lag Brattleboro rat
THE suprachiasmatic nucleus (SCN) of the hypothalamus has been shown to function as the “executive clock,” regulating circadian rhythms in mammals (29). This master circadian pacemaker is entrained by the light/dark cycle and is often referred to as the light entrainable oscillator (LEO). The pacemaker regulates circadian rhythms directly and also influences rhythms indirectly through the sleep–wake cycle (29). In animal studies, researchers frequently utilize the concept of the “rest-activity cycle” as a reflection of the sleep–wake cycle. Although the SCN is considered to be the master pacemaker, it has
Restricted-feeding
been demonstrated that some circadian rhythms persist after SCN ablation (21). The most extensively studied rhythm that persists after SCN destruction is that involving periodic food availability (4,5,26). Many animals exhibit food-anticipatory activity manifested in behaviors such as wheel running, unreinforced lever pressing, general cage activity, and drinking under daily schedules of restricted food availability (11). Periodic or restricted food availability entrains another oscillator known as the food entrainable oscillator (FEO) and many scientific studies indicate that the FEO is separate and
1 Requests for reprints should be addressed to Dr. Helen M. Murphy, Department of Psychology, John Carroll University, 20700 North Park Blvd., Cleveland, Ohio 44118. E-mail: [email protected]
1191
1192
distinct from the LEO (4,20,21,24,29). Assuming that the LEO and the FEO are separate entities and are entrained by different stimuli, it should be possible to couple or uncouple these two oscillators. In nocturnal animals, (e.g., rats) eating takes place primarily during the dark, even if food is available ad-libitum. Light/dark alternation is the most powerful synchronizer of circadian rhythms in ad-libitum feeding (18). Some circadian rhythms, however, undergo phase shifts when rats are presented with food during the light phase only and feeding is limited to 1– 4 h. For example, phase shifts are observed in activity (3,12,13,14,15), body temperature (7,8,14,15), and heart rate (15) rhythms under such a feeding regimen. A group of vasopressin-containing neurons has been found in the SCN (23), which function as output efferent neurons (32) and show a clearly-defined circadian rhythm in vasopressin levels (16). In an attempt to elucidate the functional role of vasopressin within the SCN, one study measured wheel running in the hamster following the microinjection of vasopressin into the SCN region. It was demonstrated that vasopressin did not affect the circadian timing system (1). Later publications, however, proposed that vasopressin may operate as a neurotransmitter within the SCN, either as a component of the circadian clock mechanism itself or as a part of the output pathway to other neural structures (19,32). Thus, the role of vasopressin in circadian rhythms is controversial. In order to gain insight into the effect of vasopressin on circadian rhythms, some researchers have utilized the Brattleboro rat (9,14,15,17). The Brattleboro rat lacks the ability to synthesize hypothalamic vasopressin (31), is a genetic mutant of the Long–Evans (LE) strain, exhibits hereditary diabetes insipidus, and is designated as a DI rat (30). Peterson et al. (17) demonstrated that the absence of vasopressin from the SCN did not change components of the endocrine (serotonin N-acetyltransferase levels in the pineal) and behavioral (wheel running and drinking) rhythms examined in female DI rats. These researchers concluded that, in the rat, the maintenance of circadian rhythmicity is not dependent upon the presence of vasopressin-containing neurons in the SCN. In another experiment, Groblewski et al. (9) examined circadian rhythms in cage activity and drinking behavior of female DI rats and proposed that the generation of circadian rhythms is not dependent upon central vasopressin. In the two studies just cited that found no differences between DI and LE animals, the number of hours in the light/dark cycle was manipulated and food and water were provided ad-libitum to the subjects. Another study examined 24 h pineal melatonin synthesis in male DI, LE, and Sprague–Dawley rats. Even though the results showed slight differences among these animals, the authors stated that their data “speaks against” an important role for vasopressin in the modification of circadian rhythms (22). These experimenters did make the
MURPHY ET AL.
point, however, that vasopressin may be involved in a more subtle manner in circadian rhythmicity which may not be readily apparent under standard lighting conditions. Contrary to the above findings, our laboratory has demonstrated the importance of vasopressin in circadian rhythms (14,15). In one of the studies, we examined the interaction of vasopressin and the photic oscillator in male DI and LE rats utilizing telemetry (15). In recent years, telemetry has been employed in circadian rhythm studies because it has been shown to be a very suitable technique for the continuous and undisturbed measurement of body temperature, heart rate, and motor activity (6,10,27,28). In our experiment, we demonstrated that with ad-libitum feeding and nocturnal restricted-feeding (two separated scheduled-feeding periods in the dark cycle), natural nocturnal cycles of body temperature, heart rate, and motor activity were maintained in both DI and LE animals. Marked changes, however, developed with diurnal restricted-feeding (two separated scheduled-feeding periods in the light cycle). With DI animals, the influence of the photic oscillator was lost and body temperature, heart rate, and motor activity shifted from nocturnal to diurnal patterns, reflecting the dominance of the food presentation in these vasopressindeficient animals. The LE animals, on the other hand, lost a well-defined circadian rhythmicity resulting from adherence to the photic oscillator, while at the same time being influenced by the presentation of food. We concluded that vasopressin has a significant interaction with the photic oscillator, which is obvious when the photic and nonphotic oscillators are uncoupled. In addition, we proposed that the strength of the photic oscillator was decreased or that the effect of this oscillator was masked in vasopressin-deficient rats compared to vasopressin-containing rats. In view of the results obtained in the above study, we asked the following questions: 1) Could vasopressin be significantly involved in circadian rhythm disturbances, particularly those concerned with phase shifts? (In circadian rhythm disturbances, natural cycles of rest-activity and feeding are disrupted, resulting in conflict between biological and environmental cues.) 2) Would the use of phaseadvance or phase-delay regimens and ad-libitum or restricted-feeding (nocturnal or diurnal) schedules reveal more information about how vasopressin interacts with the LEO and/or FEO in regulating rest-activity cycles? In the first experiment, we sought to determine whether or not a phase shift paradigm in which the dark cycle is either advanced or delayed would produce different effects under ad-libitum feeding. EXPERIMENT 1 Method The subjects utilized were 12 male homozygous DI rats and 12 male LE rats (Harlan–Sprague Dawley, Inc., Indianapolis, IN). Animals were 6 weeks of age and weighed between
VASOPRESSIN AND CIRCADIAN PHASE SHIFTS
100 –110 grams at the beginning of the experiment. Subjects were housed in individual translucent polypropylene cages with the room temperature maintained at 23 6 1°C and employed a 12 h/12 h light/dark cycle. Utilizing sterile techniques, all animals were anesthetized with sodium pentobarbital (4 mg or 5 mg/100 g body weight for DI or LE rats, respectively) and a biotelemetry transmitter (Model CTA-F40, Data Sciences, Inc., St. Paul, MN), capable of measuring body temperature (BT), electrocardiogram, and motor activity (AC), was implanted subcutaneously. Implant signals were transmitted to a personal computer that was equipped with receivers and a multiplexer (Model CTR-86 Receiver and Model MP-12 Multiplexer). The receivers converted the electrocardiogram waveform to heart rate (HR). Individual receivers were connected to a multiplexer where BT, HR, and AC information was sequentially passed to the data acquisition system. Data, which were collected every 5 min for the duration of the experiment, were stored on the hard disk of the personal computer for later analysis. Following the implantation of transmitters, the animals were subdivided into two groups of 6 DI and 6 LE rats each. Both groups were provided with food and water ad-libitum throughout the experiment. The animals had an habituation period of 14 days. Following the habituation period, 6 animals from each strain were subjected to a light period shortened by 6 h (phase-advance). The other 6 animals from each strain were subjected to a dark period lengthened by 6 h (phase-delay). The new light/dark cycle pattern was continued for 10 days. RESULTS Figure 1 summarizes the circadian rhythms of BT, HR, and AC for DI and LE animals under the phase-advance paradigm. The circadian rhythms were similar for DI and LE animals during both preshift and postshift conditions. During the preshift condition, all animals displayed a rhythm typical of nocturnal animals, namely, higher BT, HR, and AC during the dark period than during the light period. During the postshift period of the phase-advance paradigm, both strains of animals showed a strong adherence to the old light/dark cycle; the circadian rhythms became weaker; and the duration of AC became more persistent. The LE animals developed a modest peak with the onset of the new dark phase. Figure 2 summarizes the circadian rhythms of BT, HR, and AC for DI and LE animals under the phase-delay paradigm. The circadian rhythm was similar for DI and LE animals during both preshift and postshift conditions. During the preshift condition, all animals displayed a rhythm typical of nocturnal animals. During the phase-delay postshift period, even though both strains of animals synchronized to the new light/dark cycle, all animals initially showed anticipation of the old dark phase.
1193
Figure 3 displays actograms for AC for DI (A and C) and LE (B and D) animals under ad-libitum phase-advance (top) and phase-delay (bottom) paradigms. The figure shows that with phase-advance, both strains maintained a strong adherence to the old light/dark cycle; the LE animals developed a response to the new dark phase; and both strains developed more persistent AC. With phase-delay, both strains synchronized to the new light/dark cycle. Although the DI animals gradually adapted to the new schedule, the figure clearly shows that the LE animals adjusted immediately. In addition, all animals were instantly responsive to the care artifact in both phase-advance and phase-delay paradigms. EXPERIMENT 2 With the exception of entrainment of the photic oscillator to changing photic cues (phase shifts) by vasopressin-containing animals during ad-libitum feeding, one would conclude that vasopressin has little or no role to play in the generation or maintenance of circadian rhythms. In many aspects, the vasopressin-deficient animals resembled their vasopressincontaining counterparts when there was a 6-h phase-advance or phase-delay of the circadian cycle. However, from previous experiments, we have demonstrated that scheduled-feeding (FEO) presented during the light cycle, but not the dark cycle, has a significant role to play in modulating circadian rhythmicity in vasopressin-containing and vasopressin-deficient rats (14,15). As a result, we hypothesized that phase shift paradigms similar to the ones presented in Experiment 1 of this study would differentially effect vasopressin-containing and vasopressin-deficient animals if scheduled-feeding would be introduced. Thus, a second experiment was devised to delineate the differences between the two strains when the LEO and FEO were coupled or uncoupled and the phase shift paradigms of Experiment 1 were introduced. Method In this Experiment 24 DI rats and 24 LE rats were utilized. The same parameters for age, weight, and housing conditions employed in Experiment 1 were used in this experiment. The telemetry methodology was also the same as Experiment 1. The animals were subdivided into four groups of 6 DI and 6 LE rats each. Groups 1 and 2 were fed at the second hour of the dark cycle and groups 3 and 4 were fed at the second hour of the light cycle. The amount of food presented during each feeding for all groups was 10% of the body weight of the animal calculated from the previous day. (The 10% value was utilized because we had determined that animals could gain weight and survive throughout the experiment with such a feeding regimen.) Thus, all animals were food restricted and the feeding regimen was maintained throughout the experiment. Water was available adlibitum for all animals. Groups 1 and 3 were phase advanced
1194
MURPHY ET AL.
FIG. 1. Mean circadian rhythms of BT, HR, and AC for 6 DI and 6 LE rats under the phase-advance paradigm. The graphs show 2 days preshift, followed by the day of shift, and 5 days postshift under the ad-libitum feeding paradigm. The figure depicts a 24 point moving average of the 5 min data points. The darkened lines adjacent to the x-axis, represent the dark phase of the light/dark cycle. The short lines close to the x-axis, indicate the time of care and handling of animals. The long dark line above the care and handling lines, which extends for the duration of the experiment, indicates ad-libitum feeding.
VASOPRESSIN AND CIRCADIAN PHASE SHIFTS
1195
FIG. 2. Mean circadian rhythms of BT, HR, and AC for DI and LE rats under the phase-delay paradigm. All other graphic characteristics are the same as Figure 1.
1196
MURPHY ET AL.
FIG. 3. Mean circadian double-plotted actograms of AC for 6 DI (A) and 6 LE (B) rats subjected to ad-libitum feeding under phase-advance (top) and for 6 DI (C) and 6 LE (D) rats subjected to ad-libitum feeding under phase-delay (bottom) paradigms. The darkened lines adjacent to the x-axis (top and bottom) represent the dark phase of the light/dark cycle. The short lines close to the x-axis (top and bottom) indicate the time of care and handling of animals.
VASOPRESSIN AND CIRCADIAN PHASE SHIFTS
and groups 2 and 4 were phase delayed according to the paradigms utilized in Experiment 1. All animals had an habituation period of 14 days on their respective feeding regimens. Following the habituation period, the animals were subjected to their respective phase shifts. Feeding was shifted in synchrony with the light/dark cycle, that is, at the beginning of the second hour of the dark or light cycle. The new light/dark cycle pattern was continued for 10 days. Results Figure 4 summarizes the circadian rhythms of BT, HR, and AC for DI and LE animals under the nocturnal food-restricted, phase-advance paradigm. With the preshift condition, BT, HR, and AC were robust. The circadian rhythm patterns were those of nocturnal animals. During the postshift condition, the circadian rhythms remained strong and shifted immediately. Figure 5 summarizes the circadian rhythms of BT, HR, and AC for DI and LE animals under the nocturnal, foodrestricted phase-delay paradigm. With the preshift condition, BT, HR, and AC were robust. The circadian rhythm patterns were those of nocturnal animals. During the postshift condition, the circadian rhythms remained strong and shifted gradually for BT and AC and immediately for HR. In addition, even though both strains synchronized to the new light/dark cycle, all animals initially showed anticipation for the old dark phase and/or old time of food presentation. Figure 6 displays actograms for AC for DI (A and C) and LE (B and D) animals under nocturnal, food-restricted phase-advance (top) and phase-delay (bottom) paradigms. The figure shows that with phase-advance, the shift was immediate for both DI and LE animals as evidenced by increased AC at the onset of the new dark phase. For phase-delay, the shift was gradual for both DI and LE animals. Figure 7 depicts the circadian rhythms of BT, HR, and AC for DI and LE animals under the diurnal, food-restricted phase-advance paradigm. With the preshift condition, LE animals showed a confused rhythm characterized by peaks in both the dark and light phases in BT, HR, and AC. The DI animals had definite peaks in the light and troughs in the dark in BT and HR. The AC rhythm was somewhat confused in DI animals. During the postshift condition, the LE animals had a weak, confused rhythm in BT, HR, and AC; whereas, DI animals had a single, strong diurnal rhythm of BT, HR, and AC. Figure 8 gives the circadian rhythms of BT, HR, and AC for DI and LE animals under the diurnal, food-restricted phase-delay paradigm. With the preshift condition, LE animals showed a confused rhythm characterized by peaks in both the dark and light phases in BT, HR, and AC. The DI animals had a single, strong diurnal rhythm of BT, HR, and
1197
AC. During the postshift condition, the LE animals had weak rhythms in BT, HR, and AC; but, were synchronized with the strong diurnal rhythms of BT and AC of DI animals. Figure 9 displays actograms for AC for DI (A and C) and LE (B and D) animals under diurnal, food-restricted phaseadvance (top) and phase-delay (bottom) paradigms. The figure illustrates that with phase-advance, although LE animals showed peaks during both the dark and light phases, the shift of AC was immediate and coincided with the onset of the new dark period and the time of food presentation. On the other hand, the DI animals lost their nocturnal peaks and the shift of AC was immediate and coincided with the time of feeding. For phase-delay, although the LE animals shifted immediately to the presentation of food, strong peaks of AC were also immediately apparent with the onset of the dark period. The DI animals, however, lost their nocturnal peaks of AC and shifted their AC to coincide with the presentation of food. EXPERIMENT 3 In Experiment 2, the LEO and FEO were either tightly coupled or completely uncoupled and the photic and nonphotic zeitgebers were shifted in synchrony. Following the analysis of data obtained from the this experiment, the question arose: What if the two oscillators were weakly coupled and the photic zeitgeber was then phase advanced, while the nonphotic zeitgeber was held constant? The answers to this question would help to elucidate the relative dominance of the two oscillators in the two strains of animals, as well as provide insight into the phenomenon of coupling. Thus, a third experiment was conducted to provide such answers. Method In Experiment 3, 6 DI rats and 6 LE rats were utilized. The same parameters for age, weight, housing, and telemetry utilized in Experiments 1 and 2 were employed. During the habituation period of 14 days the animals were fed 10% of their body weight from the previous day at the third hour of the dark phase. Following the habituation period, the animals were subjected to a 6 h phase-advance shift. Feeding was not shifted. Thus, animals now received food at the beginning of the ninth hour of the dark phase. The new light/dark cycle pattern was continued for 10 days. Results Figure 10 summarizes the circadian BT, HR, and AC for DI and LE animals under nocturnal phase-advance with the time of food presentation remaining constant. The overall preshift circadian rhythms of BT, HR, and AC were similar for DI and LE animals in that they reflected patterns of nocturnal animals. The LE animals responded immediately to the onset of the dark through increased BT, HR, and AC;
1198
MURPHY ET AL.
FIG. 4. Mean circadian rhythms of BT, HR, and AC for 6 DI and 6 LE rats under the nocturnal food-restricted, phase-advance paradigm. The figure depicts a 24 point moving average of the 5 min data points. The darkened lines adjacent to the x-axis, represent the dark phase of the light/dark cycle. The short lines close to the x-axis, indicate the time of care and handling of animals. The short dark lines above the care and handling lines, indicate the time of feeding.
VASOPRESSIN AND CIRCADIAN PHASE SHIFTS
1199
FIG. 5. Mean circadian rhythms of BT, HR, and AC for DI and LE rats under the nocturnal food-restricted, phase-delay paradigm. All other graphic characteristics are the same as Figure 4.
1200
MURPHY ET AL.
FIG. 6. Mean circadian double-plotted actograms of AC for 6 DI (A) and 6 LE (B) rats subjected to nocturnal food-restricted feeding under phase-advance (top) and for 6 DI (C) and 6 LE (D) rats subjected to nocturnal food-restricted feeding under phase-delay (bottom) paradigms. The darkened lines adjacent to the x-axis (top and bottom) represent the dark phase of the light/dark cycle. The short lines close to the x-axis (top and bottom) indicate the time of feeding, as well as care and handling of animals.
VASOPRESSIN AND CIRCADIAN PHASE SHIFTS
1201
FIG. 7. Mean circadian rhythms of BT, HR, and AC for 6 DI and 6 LE rats under the diurnal food-restricted, phase-advance paradigm. All other graphic characteristics are the same as Figure 4.
1202
MURPHY ET AL.
FIG. 8. Mean circadian rhythms of BT, HR, and AC for DI and LE rats under the diurnal food-restricted, phase-delay paradigm. All other graphic characteristics are the same as Figure 4.
VASOPRESSIN AND CIRCADIAN PHASE SHIFTS
1203
FIG. 9. Mean circadian double-plotted actograms of AC for 6 DI (A) and 6 LE (B) rats subjected to diurnal food-restricted feeding under phase-advance (top) and for 6 DI (C) and 6 LE (D) rats subjected to diurnal food-restricted feeding under phase-delay (bottom) paradigms. All other graphic characteristics are the same as Figure 6.
1204
MURPHY ET AL.
FIG. 10. Mean circadian rhythms of BT, HR, and AC for 6 DI and 6 LE rats under the nocturnal food-restricted, phase-advance paradigm with the time of feeding held constant. All other graphic characteristics are the same as Figure 4.
VASOPRESSIN AND CIRCADIAN PHASE SHIFTS
1205
FIG. 11. Mean circadian double-plotted actograms of AC for 6 DI (A) and 6 LE (B) rats subjected to nocturnal food-restricted feeding under the phase-advance paradigm with the time of feeding held constant for preshift and postshift periods. All other graphic characteristics are the same as Figure 6.
but, the influence of food presentation was also apparent as manifested by maintenance of heightened physiological responses. On the other hand, the circadian rhythms of the DI animals were strongly controlled by presentation of food as observed by a sharp peak at the time of food presentation. By Day 1 of the postshift condition, the LE animals responded to the onset of dark at the new time; yet, still were influenced by the presentation of food. Despite the 6 h phase-advance, DI animals maintained the preshift circadian pattern and continued to be dominated by the presentation of food. Figure 11 displays actograms for AC for DI (A) and LE (B) animals under nocturnal, food-restricted phase-advance with the time of food presentation remaining constant. The figure shows that LE animals responded immediately to the new dark period as evidenced by increased levels of AC; while still maintaining peaks of AC coinciding with the presentation of food. The DI animals, however, lacked peaks of AC in the new dark phase until Day 6. They continued to show a peak of AC during feeding. DISCUSSION In Experiment 1, which utilized the ad-libitum feeding paradigm, the circadian rhythm of BT, HR, and AC in DI and LE animals were similar during the preshift condition.
During the phase-advance postshift condition, both strains maintained a strong adherence to the old light/dark cycle; whereas, during the phase-delay postshift condition, both strains readily adapted to the phase shift. With ad-libitum feeding, the LEO was dominant and the timing of the sleep–wake cycle (rest-activity cycle) was in the proper phase relationship with this master circadian pacemaker during the preshift condition. With the phase-advance shift, the light/dark cycle was manipulated and there was a dissociation between the LEO and the timing of rest. The rest (light) period was shortened by 6 h which led to a high degree of internal temporal disorganization within the animals. Both strains of rats displayed decreased BT, HR, and AC at the beginning of the new dark cycle and increased BT, HR, and AC at the beginning of the new light cycle. As a result, adherence to the old light/dark cycle led to a circadian rhythm that was maladapted to the new environmental cue. On the other hand, with the phase-delay shift, which involved manipulation of the light/dark cycle through a six hour extension of the dark phase, there was no dissociation between the LEO and the timing of rest. The dark period was merely lengthened by 6 h which was followed by a normal rest (light) period. This did not lead to a high degree of internal temporal disorganization within the animals. Both strains of rats displayed normal BT, HR, and AC
1206
at the beginning of the new dark and light cycles. As a result, adaptation to the light/dark cycle led to a circadian rhythm that was in synchrony with the new environmental cue. The phase-advance paradigm is indicative of an eastward jet lag phenomenon with the associated maladaptive responses. On the other hand, the phase-delay (westward) paradigm is without any indication of jet lag illness, but shows an anticipatory response to the dark phase. In Experiment 2, with the introduction of food restriction, another oscillator was unmasked (FEO). In nocturnal restricted-feeding when the LEO and FEO were coupled, no significant differences were noted between vasopressin-deficient and vasopressin-containing animals when they were maintained under a normal 12 h/12 h light/dark cycle or were phase advanced by 6 h. With nocturnal restrictedfeeding, both the LEO and FEO were functional and were coupled. Under the preshift condition, the timing of the rest-activity cycle was in the proper phase relationship with the coupled LEO and FEO. With the shift (6-h phaseadvance), both the LEO and FEO were shifted in synchrony. In this case, the light/dark cycle and rest-activity cycle rapidly realigned and synchronized, since the oscillators mutually reinforced one another. The reason that an adaptive circadian rhythmicity developed immediately, may be due to entrainment of the SCN pacemaker by food-restriction schedules (25). This entrainment of the LEO by the FEO occurs only when the periods of the pacemaker and the restriction schedule are very similar (2,25). In the present experiment, there was only a 1-h difference between the onset of the dark cycle and the presentation of food. Consequently, this could explain the differences obtained in the postshift condition of nocturnal restricted-feeding versus ad-libitum feeding of Experiment 1, where the FEO was nonfunctional since the nonphotic zeitgeber (restrictedfeeding) was absent and, thus, the oscillators were unable to reinforce one another. One way of assessing the relative dominance of the LEO versus the FEO and the role of vasopressin in influencing each oscillator is to uncouple the two oscillators and observe the ensuing rhythms. This was achieved by introducing a diurnal restricted-feeding paradigm. In this paradigm, significant differences were noted between DI and LE animals in BT, HR, and AC. In general, during the preshift condition, the DI animals had a reversal of peaks and troughs of these physiological variables from those observed in nocturnal restricted-feeding. The LE animals, on the other hand, had a confused rhythmicity with major nocturnal peaks evident in addition to the diurnal peak associated with food presentation. In the postshift phaseadvance and phase-delay conditions, the DI animals rapidly adapted to the phase shift and developed a definite circadian rhythmicity characterized by peaks associated with food presentation in the light and troughs in the dark. The LE animals, however, did not develop a definite rhythmicity
MURPHY ET AL.
(characterized by sustained peaks and troughs) but showed significant peaks which coincided with the onset of the new dark period, as well as the presentation of food. Under preshift and postshift conditions, the absence of vasopressin enabled the DI animals to adapt and develop a definite circadian rhythmicity governed by the FEO. In contrast to nocturnal-restricted feeding, where the LEO and FEO were coupled, diurnal restricted-feeding resulted in an uncoupling of the LEO and FEO. As a consequence of this uncoupling, the circadian rhythms of DI animals synchronized to the FEO and the response to the LEO was lost in the preshift condition. Because of this, the rest-activity cycle was governed by the FEO. On the other hand, the circadian rhythms of LE animals were influenced by both the LEO and FEO. Since the oscillators were uncoupled and since LE animals were strongly under the influence of the LEO (15), there was a dissociation between the master clock and the timing of rest due to the presentation of food when the animals would normally be sleeping. This led to a high degree of internal temporal disorganization, resulting in confused rhythmicity in BT, HR, and AC. In the postshift condition, the circadian rhythms of DI animals immediately shifted and synchronized to the presentation of food. Thus, the FEO, once again, governed the rest-activity cycle which was characterized by sustained peaks and troughs. In contrast, the LE animals, governed by the LEO and the FEO, exhibited a continuation of internal temporal disorganization resulting in a disruption of the rest-activity cycle. This disruption of the rest-activity cycle was manifested by confused rhythms of BT, HR, and AC which lacked sustained peaks and troughs that are characteristic of a normal restactivity cycle. In Experiment 3, where the LEO and FEO were weakly coupled in the preshift period (restricted feeding at the third hour of the dark cycle) differences were noted between DI and LE animals in that DI animals were almost exclusively attuned to the nonphotic zeitgeber; whereas, LE animals were influenced by both the photic and nonphotic zeitgebers. After the 6-h phase-advance shift with the feeding time being held constant (now nine hours into the dark cycle), DI animals continued to be dominated by the nonphotic zeitgeber and showed a circadian rhythm almost identical with that observed in the preshift condition. The LE animals, on the other hand, responded to the new photic cue while maintaining a response to the nonphotic zeitgeber. Although the restricted feeding at the third hour of the dark cycle began to uncouple the oscillators, the 6-h phase-advance of the light/dark cycle enhanced the separation of the zeitgebers; thus, further uncoupling the oscillators. Because of the strong influence of the LEO on LE animals, this uncoupling produced a disruption of the rest component of the restactivity cycle. There was now a dissociation between the LEO and the timing of rest in LE rats which led to a high degree of internal disorganization within the animals. On
VASOPRESSIN AND CIRCADIAN PHASE SHIFTS
1207
the other hand, because DI animals are dominated by the FEO, there was no disruption of the rest-activity cycle. In conclusion, the present studies help to unravel some of the mystery surrounding the role of vasopressin in circadian rhythms. Vasopressin-containing nocturnal animals are governed by the LEO when fed ad-libitum and strongly governed by the combined LEO and FEO during restricted feeding. The vasopressin-deficient animals, however, are governed by the FEO unless food is provided ad-libitum. Finally, the findings of the present study concerning the role of vasopressin in synchronizing various circadian rhythms of the organism to the LEO may be of importance in understanding human disturbances, such as jet lag. In this circadian rhythm disturbance there is a dissociation between the master pacemaker (LEO) and the timing of sleep. In jet
lag, crossing time zones can cause a person to fall asleep during the normally active period (day) and to wake during the rest period (night). As a result, a person’s circadian rhythms are disturbed and many days may pass before the body can readjust to the new time zone. This is particularly evident when flying eastward (phase-advance) as compared to flying westward (phase-delay). Experiment 3 of the present study provides an animal model for the disturbances encountered in eastward jet lag and evidence for the role of vasopressin in this circadian rhythm disorder. ACKNOWLEDGEMENTS This research was supported by a Focused Giving Grant from Johnson & Johnson to Helen M. Murphy and Cyrilla H. Wideman.
REFERENCES 1. Albers, H. E.; Ferris, C. F.; Leeman, S. E.; Goldman, B. D. Avian pancreatic polypeptide shifts hamster circadian rhythms when microinjected into the suprachiasmatic region. Science (Washington DC). 223:833– 835; 1984. 2. Aschoff, J.; von Goetz, C. Effects of feeding cycles on circadian rhythms in squirrel monkeys. J. Biol. Rhythms 1:267– 276; 1986. 3. Baker, R. A. A periodic feeding behavior in the albino rat. J. Comp. Physiol. Psychol. 46:422– 426; 1953. 4. Boulos, Z.; Rosenwasser, A. M.; Terman, M. Feeding schedules and the circadian organization of behavior in the rat. Behav. Brain Res. 1:39 – 65; 1980. 5. Clarke, J. D.; Coleman, G. J. Persistent meal-associated rhythms in SCN-lesioned rats. Physiol. Behav. 36:105–113; 1986. 6. Clement, J. G.; Mills, P.; Brockway, B. Use of telemetry to record body temperature and activity in mice. J. Pharmacol. Methods 21:129 –140; 1989. 7. Ehret, C. F.; Dobra, K. W. The oncogenic implications of chronobiotics in the synchronization of mammalian circadian rhythms: Barbiturates and methylated xanthines. In: Nieburgs, H. E., Ed. Proceedings of the Third International Symposium on the detection and prevention of cancer. New York: Marcel Dekker; 1977:1101–1114. 8. Ehret, C. F.; Groh, K. R.; Meinert, J. C. Circadian dyschronism and chronotypic ecophilia as factors in aging and longevity. In: Samis, H. V.; Capobianco, S., Eds. Aging and biological rhythms. New York: Plenum Publishing Corp.; 1978:185–213. 9. Groblewski, T. A.; Nunez, A. A.; Gold, R. M. Circadian rhythms in vasopressin deficient rats. Brain Res. Bull. 6:125– 130; 1981. 10. Kramer, K.; VanAcker, S. A. B. E.; Voss, H. P.; Grimbergen, J. A.; VanderVijgh, W. J. F.; Bast, A. Use of telemetry to record electrocardiogram and heart rate in freely moving mice. J. Pharm. Tox. Methods. 30:209 –215; 1993. 11. Mistleberger, R. E. Circadian food-anticipatory activity: formal models and physiological mechanisms. Neurosci. Biobehav. Rev. 18:171–195; 1994. 12. Mouret, J. R.; Bobiller, P. Diurnal rhythms of sleep in the rat: Augmentation of paradoxical sleep following alterations of the feeding schedule. Int. J. Neurosci. 2:265–270; 1971. 13. Mouret, J. R.; Doindet, J.; Chouvet, G. Circadian rhythms of
14. 15. 16.
17. 18. 19.
20. 21.
22. 23. 24. 25. 26.
sleep in the rat: Importance of the feeding schedule. Int. J. Chronobiol. 1:345; 1973. Murphy, H. M.; Wideman, C. H.; Nadzam, G. R. Vasopressin deficiency and circadian rhythms during food-restriction stress. Peptides. 14:1215–1220; 1993. Murphy, H. M.; Wideman, C. H.; Nadzam, G. R. The interaction of vasopressin and the photic oscillator in circadian rhythms. Peptides. 17:467– 475; 1996. Noto, T.; Hashimoto, H.; Doi, Y.; Nawajima, T.; Kato, N. Biorhythms of arginine-vasopressin in the paraventricular, supraoptic and suprachiasmatic nuclei of rats. Peptides. 4:875– 878; 1993. Peterson, G. M.; Watkins, W. B.; Moore, R. Y. The suprachiasmatic hypothalamic nuclei of the rat. Behav. Neural. Biol. 29:236 –245; 1980. Reinberg, A. Chronobiology and nutrition. In: Reinberg, A.; Smolensky, M. H., Eds. Biological rhythms and medicine. New York: Springer–Verlag; 1983:265–300. Reppert, S. M. Circadian rhythm of cerebrospinal fluid vasopressin: Characterization and physiology. In: Schrier, R. W., Ed. Vasopressin. New York: Raven Press; 1985: 455– 464. Rosenwasser, A. M.; Pelchat, R. J.; Adler, N. T. Memory for feeding time: Possible dependence on coupled circadian oscillators. Physiol. Behav. 32:25–30; 1984. Rusak, B. Biological rhythms: from physiology to behavior. In: Montplaisir, J.; Godbout, R., Eds. Sleep and biological rhythms. New York: Oxford University Press; 1990:11–24. Schro¨der, H.; Stehle, J.; Henschel, M. Twenty-four-hour pineal melatonin synthesis in the vasopressin-deficient Brattleboro rat. Brain Res. 459:328 –332; 1988. Sofroniew, M. V.; Weindl, A. Projections from the parvocellular vasopressin- and neurophysin-containing neurons of the suprachiasmatic nucleus. Am. J. Anat. 153:391– 430; 1978. Stephan, F. K. Limits of entrainment to periodic feeding in rats with suprachiasmatic lesions. J. Comp. Physiol. 143:401– 410; 1981. Stephan, F. K. The role of period and phase in interactions between feeding- and light-entrainable circadian rhythms. Physiol. Behav. 36:151–158; 1986. Stephan, F. K.; Swann, J. M.; Sisk, C. L. Anticipation of 24-hr feeding schedules in rats with lesions of the suprachiasmatic nucleus. Behav. Biol. 25:346 –363; 1979.
1208 27. Teelink, J. R.; Clozel, J. P. Hemodynamic variability and circadian rhythm in rats with heart failure: role of locomotor activity. Am. J. Physiol. 264:H2111–H2118; 1993. 28. Tornatzky, B. H. C.; Miczek, K. A. Long-term impairment of autonomic circadian rhythms after brief intermittent social stress. Physiol. Behav. 53:983–993; 1993. 29. Turek, F. W. Circadian rhythms. In: Bardin, C. W., Ed. Recent progress in hormone research, vol. 49. San Diego: Academic Press, Inc.; 1994:43–90. 30. Valtin, H.; Schroeder, H. A. Familial hypothalamic diabetes
MURPHY ET AL. insipidus in rats (Brattleboro strain). Am. J. Physiol. 206:425– 430; 1964. 31. Valtin, H.; Sawyer, W. H.; Sokol, H. W. Neurohypophysial principles in rats homozygous and heterozygous for hypothalamic diabetes insipidus (Brattleboro strain). Endocrinology. 77:701–706; 1965. 32. Watts, A. C.; Swanson, L. W. Efferent projections of the suprachiasmatic nucleus: II. Studies using retrograde transport of fluorescent dyes and simultaneous peptide immunoreactivity in the rat. J. Comp. Neurol. 258:230 –252; 1987.