Photic phase response curve in Octodon degus: assessment as a function of activity phase preference

Martien J. H. Kas, Dale M. Edgar

Abstract

Light exposure during the early and late subjective night generally phase delays and advances circadian rhythms, respectively. However, this generality was recently questioned in a photic entrainment study in Octodon degus. Because degus can invert their activity phase preference from diurnal to nocturnal as a function of activity level, assessment of phase preference is critical for computations of phase reference [circadian time (CT) 0] toward the development of a photic phase response curve. After determining activity phase preference in a 24-h light-dark cycle (LD 12:12), degus were released in constant darkness. In this study, diurnal (n = 5) and nocturnal (n = 7) degus were randomly subjected to 1-h light pulses (30–35 lx) at many circadian phases (CT 1–6: n= 7; CT 7–12: n = 8; CT 13–18: n = 8; and CT 19–24: n = 7). The circadian phase of body temperature (Tb) onset was defined as CT 12 in nocturnal animals. In diurnal animals, CT 0 was determined as Tb onset + 1 h. Light phase delayed and advanced circadian rhythms when delivered during the early (CT 13–16) and late (CT 20–23) subjective night, respectively. No significant phase shifts were observed during the middle of the subjective day (CT 3–10). Thus, regardless of activity phase preference, photic entrainment of the circadian pacemaker inOctodon degus is similar to most other diurnal and nocturnal species, suggesting that entrainment mechanisms do not determine overt diurnal and nocturnal behavior.

  • circadian rhythm
  • body temperature
  • wheel running
  • phase shift
  • light
  • entrainment

the photic phase-shifting characteristics of mammalian circadian rhythms are thought to reflect the fundamental entrainment properties of the circadian pacemaker, located in the suprachiasmatic nucleus (SCN) of the hypothalamus (16, 23, 29). In most diurnal and nocturnal species (i.e., in unicellular alga, insects, rodents, and primates), light exposure during the early subjective night phase delays circadian rhythms and produces phase advances during the late subjective night (4, 7, 8, 10, 19, 20, 31), suggesting that the photic phase response curve (PRC) is highly conserved phylogenetically. However, a recent study in Octodon degus, a rodent species that exhibits multiple behavioral chronotypes (14), suggested that this species has an atypical photic PRC (15). Brief light pulses produced phase delays during the early subjective day and phase advances during the late subjective night. In two of the animals, brief light pulses produced phase advances at the end of the subjective night, but no phase delays at all (15). Based on the belief that degus are diurnal and have atypical PRCs, it was suggested that the entrainment mechanism within the mammalian circadian pacemaker is fundamentally different in diurnal and nocturnal species (15).

Although the fundamental shape of the photic PRC is similar in most species, environmental factors (e.g., intensity and duration of acute light exposure and background light intensity) can alter the shape of the photic PRC (1, 7, 21, 30). Furthermore, the magnitude of a photic phase shift can also vary due to interspecific differences, such as endogenous period length (τ) (3) or differences in photic sensitivity (22). Therefore, between-species comparisons of PRC shapes can be difficult to interpret. The fact that individual Octodon deguscan exhibit both diurnal and nocturnal phase preference (12) presents a unique opportunity to compare the shape of the PRC independent of environmental and interspecific factors.

Fundamental to the construction of any PRC is the accurate assessment of a standard circadian rhythm phase reference. Normally, the daily onset of activity behavior or a corresponding point in the body temperature (Tb) circadian rhythm is used as the phase reference when animals are housed under constant conditions. In diurnal species, this time of day is defined as circadian time zero (CT 0) and coincides with lights on during a 24-h light-dark cycle (beginning of the subjective day). In contrast, in nocturnal species, the beginning of the active phase is defined as CT 12 and starts at the beginning of the subjective night (onset of darkness in a 24-h light-dark cycle). BecauseOctodon degus can invert its activity phase preference from diurnal to nocturnal as a function spontaneous behavioral activity level (12), the definition of CT 0 is somewhat more complicated than in strictly diurnal or nocturnal species. Knowledge about the phase-preference status in degus when housed in constant conditions is in fact necessary to properly define CT 0 toward the construction of a photic PRC. In the present study, the photic entrainment properties of the circadian pacemaker were examined in Octodon degus as a function of each animal's diurnal or nocturnal activity phase preference using identical lightning conditions. When CT 0 was properly assigned with respect to phase preference, this species exhibited a prototypical photic PRC (phase delays and phase advances in the early and late subjective night, respectively).

METHODS

Animals and recording environment. Nine adult (age 27–33 mo) male Octodon degus were used in this study. Degus were individually housed in Nalgene cages (46-cm long × 24-cm wide × 20-cm deep) and maintained in a 24-h light-dark cycle (LD 12:12) within separate compartments of a ventilated, light-proof and sound-attenuating stainless steel recording chamber. Food and water were available ad libitum. Ambient temperature was 23 ± 1°C. Tb was detected using surgically implanted miniature biotelemetry transmitter (Barrows, Palo Alto, CA) and a calibrated telemetry receiver (Data Sciences, St. Paul, MN) located beneath the cage. A computer-based data collection system sampled Tb (°C) as discrete events every minute (5, 11). Data collection was continuous throughout the study.

Animal surgery. Degus were sedated with diazepam (1.6 mg/kg im) and anesthetized (2% isoflurane in medical grade oxygen), and a biotelemetry transmitter was surgically implanted in the peritoneal cavity. Antibiotics were administered postoperatively (gentamicin 20 mg/kg sc). Postoperative pain was managed with buprenorphine (0.03 mg/kg sc). Before beginning the studies, degus were given at least 3 wk to recover from surgery.

Light treatments. Animals were entrained to LD 12:12 for at least 3 wk [lights on: 30–35 lx (4-W fluorescent bulbs); darkness: <0.1 lx] to determine whether degus exhibited a diurnal or nocturnal phase preference (see Ref. 12). After stable circadian rhythm entrainment in LD 12:12 and determination of activity phase preference (diurnal or nocturnal), lighting was switched to constant darkness (DD) for 3 wk. Degus were then subjected to a 1-h light pulse (30–35 lx) randomly delivered at one of four different phases of the circadian day (CT 1–6, CT 7–12, CT 13–18, and CT 19–24). Additional light pulses were administered at ∼10-day intervals (range of 8–21 days) so that each animal was tested at multiple circadian phases. Animals were otherwise maintained in DD. Photic phase shifts were measured in three animals exhibiting a diurnal phase preference (housed without running wheels) and again when exhibiting a nocturnal phase preference [housed with running wheels (20-cm ID and 7-cm OD); see Ref. 12]. Photic phase shifts were also measured in six other animals that spontaneously exhibited a diurnal (n = 2) or a nocturnal (n = 4) phase preference without manipulation of the running wheel.

Analysis and statistics. Phase shifts and τ changes were manually calculated as a function of circadian phase of light exposure using digital raster plots of Tb. The density of the vertical tick marks in the raster plots was plotted proportional to Tb values exceeding a 72-h mean moved stepwise in daily increments through the time series. The beginning of each circadian Tb cycle (Tb onset) was identified when Tb exceeded the 72-h mean for two consecutive hours, followed by 4 h in which at least 3 h exceeded the 72-h mean. A line was manually drawn (eye fitted) through Tb onsets 7 days before and at least 7 days after each 1-h light pulse. Phase shifts were calculated from the difference between the fitted lines (in hours) on the day after the light pulse. Transient cycles after the light pulse were not included in the postpulse line fits. Period changes were calculated as the differences in τ before and after the light pulse. The circadian phase of Tb onset in animals with nocturnal phase preference was defined as CT 12. Because Octodon degus exhibited asymmetric Tb and locomotor activity wave forms [e.g., activity phase (α) > rest phase (ρ)], CT 0 in animals with a diurnal phase preference was defined as Tb onset + 1 h (on average, the α:ρ ratio was 13:11). This correction helped to assure that CT 12 and CT 0 were not artificially offset in the construction of the photic PRC.

The phase shifts and τ changes were plotted relative to the CT at which the light pulse was initiated. First, individual phase shifts and τ changes were plotted in a cluster plot. An eighth order regression line was fitted through all data points (from animals exhibiting diurnal and nocturnal phase preference) (SigmaPlot 2.01, Jandel Scientific, Corte Madera, CA). Second, the data were blocked into six 4-h bins (CT 1–4, CT 5–8, CT 9–12, CT 13–16, CT 17–20, and CT 21–24). When an animal received more than one light pulse in a particular bin, average phase shift and τ changes were computed. Time of day differences in phase shifts and τ changes across the 24-h circadian cycle were evaluated using one-way repeated-measures ANOVA (SigmaStat 1.0, Jandel Scientific). In the presence of a main effect, mean phase shifts and changes in τ for each bin were considered significant when they differed from zero (paired t-test). Mean phase shifts and τ changes were compared between diurnal and nocturnal degus after blocking the data into four 6-h bins (CT 1–6, CT 7–12, CT 13–18, and CT 19–24). Time of day variation in phase shifts and τ changes across the 24-h circadian day were tested between diurnal and nocturnal animals using a two-way ANOVA.

RESULTS

Phase shifts. On the basis of a total of 60 light pulses delivered across nine animals (CT 1–6: n = 7; CT 7–12: n = 8; CT 13–18: n = 8; and CT 19–24: n = 7), a significant interaction between time of day and phase shifts was observed (F = 15.4; P < 0.0001). Light pulses phase shifted the circadian Tb rhythm in degus at some, but not all, circadian phases (Fig.1). The magnitude of the phase delays at the beginning of the subjective night (bin CT 13–18) averaged −0.9 ± 0.2 h and were significantly different from zero (P < 0.001). When light was scheduled during the late subjective night (CT 19–24) and during the early subjective day (CT 1–6), degus exhibited significant phase advances (1.1 ± 0.2 and 0.6 ± 0.1 h, respectively; P < 0.003; Fig.2 C). No significant phase shifts were observed at other phases of the circadian day (see Fig.2 C). No significant differences were found in the magnitude and direction of the light-induced phase shifts when compared between degus exhibiting a diurnal or nocturnal phase preference (Fig. 2 B).

Fig. 1.

Representative light-induced phase shifts of body temperature (Tb) circadian rhythms in Octodon degus. One-hour light pulses (30–35 lx) delivered during early and late subjective night produced a phase delay (A) and phase advance (B), respectively. Time of light treatment is indicated by ○. These animals exhibited a nocturnal phase preference (based on entrained phase of circadian rhythm in LD 12:12); therefore Tb onset was defined as circadian time (CT) 12. Vertical ticks in raster plots indicate mean and above mean Tb.

Fig. 2.

Photic phase response curve (PRC) for Octodon degus. A polynomial regression was plotted through phase shift data for all 9 animals (A). Degus with diurnal and nocturnal phase preference are indicated by ○ and ●, respectively. PRC data averaged in 6-h bins (means ± SE) for degus exhibiting diurnal and nocturnal phase preference (B) revealed no significant difference between groups. All data averaged in 4-h bins revealed a prototypical photic PRC with significant phase delays (P < 0.001) and advances (P < 0.003) in early and late subjective night, respectively (C). Light intensity during 1-h light pulse was 30–35 lx.

Effects on τ. One-hour light pulses sufficient to produce significant phase shifts did not alter the subsequent circadian period. ANOVA revealed no significant interactions between time of day and changes in period (Fig. 3 C). There were also no significant differences in acute photic effects on τ between both diurnal and nocturnal degus (Fig. 3 B). A weak correlation between changes in τ and phase shifts was observed (r = 0.26; P = 0.04; n = 60).

Fig. 3.

Photic τ response curve for Octodon degus. τ Response curve was derived from same light treatments used to generate PRC in Fig. 2. A polynomial regression was plotted through τ response curve data for all 9 animals (A). Degus with diurnal and nocturnal phase preference are indicated by ○ and ●, respectively. τ Response curve data averaged in 6-h bins (means ± SE) for degus exhibiting diurnal and nocturnal phase preference (B) revealed no significant difference between groups. τ Response curve data averaged in 4-h bins (n = 9) did not resolve significant interaction with CT (C). Light intensity during 1-h light pulse was 30–35 lx.

DISCUSSION

This study shows that the photic PRC in degus is very similar to the prototypical photic PRC reported for many other diurnal and nocturnal species (4, 7, 8, 10, 19, 20, 31). In contrast to the atypical mammalian photic PRC previously reported for Octodon degus(15), our findings show that light induces phase delays at the beginning of the subjective night and produces phase advances at the end of the subjective night in all degus studied, regardless of whether they exhibit diurnal or nocturnal phase preference. No significant phase shifts were observed during the middle of the subjective day. Thus the photic responsiveness of the circadian pacemaker is independent from activity phase preference in this species.

The discrepancy between the photic PRCs generated for degus in this and in the previous report is uncertain. Because the dominant circadian activity phase in degus can occur during both the subjective day and night (12, 14), it is possible that CT 0 was not correctly determined in the previous study. For example, in the previous report (15), activity onset in a nocturnal animal may have been defined as CT 0 instead of CT 12. Indeed, an accurate assessment of phase preference is critical to the generation of the photic PRC in degus. When we generated a photic PRC for diurnal and nocturnal degus, assuming they were all diurnal, a PRC comparable to that previously reported for degus (15) was observed.

Although the general shape of the PRC is the same in most species, the amplitude of the PRC is affected by light intensity and duration (1, 7,21, 30). In the present study, degus were subjected to 1-h light pulses with a light intensity of 30–35 lx. At this light intensity, degus exhibited phase delays and phase advances of up to 2.1 and 2.6 h, respectively. This light intensity is also sufficient for stable entrainment of sleep-wakefulness, Tb, locomotor activity, and circadian rhythms in this species (11). In the previous report of atypical PRCs, degus were exposed to 250 lx of light for 20 min, resulting in two different types of PRCs (15), neither of which resembled classical type 1 or type 0 PRCs (32). Although the wave form of the PRC can somewhat broaden with increasing light intensity (see Ref. 17), the general shape of the PRC is normally conserved (e.g., the CT of peak delays and advances normally remains timed within the subjective night in most species). Therefore, it seems unlikely that the prototypical mammalian PRC for degus observed in the present study would be completely disrupted by increasing the light intensity from 30–35 to 250 lx.

Activity level can also influence the magnitude of photic phase shifts in mammals (24) via serotonergic modulation of glutamatergic terminals in the SCN (2, 18, 25, 28). Therefore, one could hypothesize that a change in activity phase preference could alter the shape of the photic PRC in degus. When exhibiting a diurnal phase preference, locomotor activity in degus occurs at times of day corresponding with the least sensitive portion of the photic PRC. When exhibiting a nocturnal phase preference, locomotor activity is occurring throughout the most sensitive portion of the photic PRC and therefore could potentially reduce the magnitude of the phase shifts. However, in degus exhibiting nocturnal phase preference, the masking effects of light inhibit locomotor activity (Kas and Edgar, unpublished observation). Therefore, the similarity in shape of the photic PRC in degus exhibiting diurnal and nocturnal phase preference may be due to counterbalancing effects of light-dark masking during the 1-h light pulse. Furthermore, a previous study has shown that light produces similar phase shifts in voles regardless of rest-activity state (6). Taken together, it seems unlikely that an atypical photic PRC results from an interaction between activity and light.

The photic PRC in degus observed in this study is remarkably similar to that of a wide variety of nocturnal and diurnal organisms, suggesting that photic entrainment mechanisms are conserved across species. Therefore, it seems unlikely that diurnal and nocturnal behavior is determined by entrainment mechanisms within the circadian pacemaker. Circadian rhythms of neuronal activity and glucose utilization are similar within the SCN of diurnal, nocturnal, and crepuscular species (9, 13, 26, 27). In contrast, circadian rhythms of neuronal activity in hypothalamic regions that are located just outside of the SCN are 180° out of phase in nocturnal and diurnal species (9, 13, 26). Thus coupling mechanisms between the pacemaker and central effectors that control organismal behavior are likely to determine nocturnal or diurnal behavior rather than photic entrainment mechanisms.

Acknowledgments

The authors thank Dr. Barbara Tate and Dr. Theresa Lee for providing the animals, Humberto Garcia and Ronny Tjon for technical assistance, and Dr. Serge Daan for insightful comments on the manuscript.

Footnotes

  • Address for reprint requests and other correspondence: D. M. Edgar, Sleep Research Center, Stanford Univ. School of Medicine, 701 Welch Rd. #2226, Palo Alto, CA 94304.

  • Research was supported by the Air Force Office of Scientific Research Program for Research Excellence and Transition (F49620–95–1-0388) and the National Institute on Aging (AG-11084).

  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

REFERENCES

View Abstract