Our previous studies showed that the eyes of Japanese quail contain a biological clock that drives a daily rhythm of melatonin synthesis. Furthermore, we hypothesized that these ocular clocks are pacemakers because eye removal abolishes freerunning rhythms in constant darkness (DD). If the eyes are indeed acting as pacemakers, we predicted that the two ocular pacemakers in an individual bird must remain in phase in DD and, furthermore, the two ocular pacemakers would rapidly regain coupling after being forced out of phase. These predictions were confirmed by demonstrating that1) the ocular melatonin rhythms of the two eyes maintained phase for at least 57 days in DD and 2) after ocular pacemakers were forced out of phase by alternately patching the eyes in constant light, two components of body temperature were observed that fused into a consolidated rhythm after 5–6 days in DD, showing pacemaker recoupling. The ability to maintain phase in DD and rapidly recouple after out-of-phase entrainment demonstrates that the eyes are strongly coupled pacemakers that work in synchrony to drive circadian rhythmicity in Japanese quail.
- body temperature
all organisms display daily (circadian) rhythms in a host of biochemical, physiological, and behavioral functions. Significantly, these rhythms persist when an organism is held in constant conditions demonstrating that they are driven by an internal circadian clock. The possession of an endogenous clock allows organisms to anticipate, rather than merely respond to, daily environmental cycles. Moreover, a clock provides internal temporal organization so that internal events occur at the appropriate time of day.
Numerous studies in mammals support the hypothesis that the suprachiasmatic nucleus (SCN) of the hypothalamus is the master circadian pacemaker; that is, it drives rhythms throughout the animal (see Ref. 16 for a review). Furthermore, the SCN receives photic input exclusively via the eyes. Evidence also supports a role for hypothalamic clocks in the circadian system of birds, although there is some controversy as to the precise location of these clocks. For example, lesions of the suprachiasmatic area of the hypothalamus of several avian species, including Japanese quail, abolish rhythmicity (8, 23, 24). Two nuclei, the medial hypothalamic nucleus (MHN, also termed the periventricular preoptic nucleus) and the lateral hypothalamic retinorecipient nucleus (LHRN, also termed the visual SCN), have been identified as possible homologs of the mammalian SCN on the basis of histochemical analyses and lesion experiments (4,18, 22). Recent studies have shown oscillations in clock gene expression in the MHN of several species, including pigeon, chicken, Japanese quail, and Java sparrow, and in the MHN and LHRN of house sparrows (3, 32). For the rest of the following discussion the term “SCN” will be used to identify the avian hypothalamic clock, recognizing that its precise location is still being resolved.
Although the SCN is an important component of the circadian systems of all vertebrates, nonmammalian vertebrates also possess circadian pacemakers located outside the hypothalamus. In birds, circadian pacemakers exist in the pineal organ and/or the eyes. These pacemakers are essential for maintaining rhythmicity in constant conditions; for example, pinealectomy has been shown to abolish rhythmicity in constant darkness (DD) in seven species of birds (30). In the Japanese quail, however, pinealectomy has little effect on rhythmicity in DD, but eye removal causes quail to become arrhythmic, suggesting that the eyes are the major pacemakers in this species (23,25). In the pigeon, both the pineal and eyes act as pacemakers because rhythmicity is disrupted only when both the eyes and pineal are removed (7, 9).
The fact that pinealectomy (or eye removal) abolishes rhythmicity in a variety of birds suggests that, in contrast to mammals, the hypothalamic pacemakers in birds are unable to maintain rhythmicity in the absence of periodic inputs from the pineal or eyes. Indeed, some investigators have proposed that the pineal and SCN of some birds are joined together in a neuroendocrine loop; that is, a daily melatonin output from the pineal is required to maintain rhythmicity in the SCN, and a daily neural output from the SCN is required to maintain rhythmicity in the pineal (5, 13, 14).
The Japanese quail is proving to be an excellent model for examining the nature of avian circadian organization. In this species the eyes appear to be the loci of major circadian pacemakers that are essential for maintaining rhythmicity in the SCN: eye removal consistently causes the body temperature rhythm to become arrhythmic in DD. It is likely that the SCN is responsible for the generation of body temperature rhythms, but the ocular pacemakers are needed to maintain SCN rhythmicity. However, only 25% of birds subjected to optic nerve section become arrhythmic, although rhythmicity in the remaining birds becomes less robust (25). These data strongly support the hypothesis that the eyes maintain rhythmicity in the SCN via both neural and hormonal outputs. The nature of the neural output is unknown, although it is tempting to speculate that it may involve the retinohypothalamic (RH) tract, which is present in birds and in mammals (6, 18). The hormonal output is likely melatonin because the eyes of quail synthesize and secrete melatonin rhythmically, and daily exogenous administration of melatonin can entrain the circadian system (15, 28). The eyes of quail entrained to a 12:12-h light-dark (LD) cycle contribute about one-third of the daily melatonin levels in the blood, and the pineal contributes the remaining two-thirds (27).
The pineal gland of the Japanese quail may not be the locus of an autonomous circadian clock because the pineal, in vitro, does not exhibit a persistent rhythm of melatonin release in DD (17). Pineal rhythmicity, however, is maintained in vivo by direct input from the daily LD cycle and by neural inputs from the SCN (4, 17). In contrast to a number of avian species, pinealectomy appears to have little effect on freerunning activity or body temperature rhythms in quail (25). This suggests that the daily neural and melatonin output from the eyes is sufficient to maintain robust rhythmicity in the SCN oscillators, even if the daily melatonin output from the pineal is eliminated.
The present investigation was undertaken to assess two predictions of our model of the Japanese quail circadian system. First, if the eyes are indeed the major circadian pacemakers in this system, the two ocular pacemakers in an individual bird must remain in phase with each other in prolonged DD. If the two ocular pacemakers do not maintain phase, it would be difficult to argue that they are responsible for driving the robust, persistent body temperature rhythm that is characteristic of quail freerunning in DD. Investigations were also undertaken to assess the strength of coupling between the two ocular clocks to lay the groundwork for studies into the coupling mechanism. Second, because most birds subjected to optic nerve section are rhythmic in DD, the model proposes that ocular synthesis and secretion of melatonin is capable of maintaining rhythmicity of the SCN. If this is true, pineal synthesis and secretion of melatonin must also be capable of maintaining rhythmicity in the SCN. This leads to the prediction that pineal melatonin synthesis and secretion in eyeless birds must be disrupted; otherwise, the pineal should be able to maintain rhythmicity in DD.
MATERIALS AND METHODS
Animals and Housing
Adult male and female Japanese quail (Coturnix japonica) were raised on a 14:10-h LD cycle until birds reached sexual maturity at 6 wk of age. Birds were then housed in a 12:12-h LD cycle until introduced into experiments at 8–20 wk of age. Because the 12:12-h LD cycle is less than the critical photoperiod for Japanese quail, birds were sexually regressed when placed in experiments. Animals were provided with food [North Carolina State University (NCSU) quail layer ration calculated to contain 16% protein and 2,550 kcal/kg] and water ad libitum. In experiments where body temperature was measured, animals were placed in Plexiglas cages (22 cm length × 20 cm width × 23 cm height) and isolated inside light-tight photoperiod boxes. Boxes were housed inside a room held in DD and maintained at 21 ± 3°C. Each photoperiod box contained a 4-watt fluorescent bulb that yielded an average intensity of 80 lx at the level of the bird.
In experiments 2, 5, and 6, groups of intact or blinded birds were housed in light-tight photoperiod boxes (122 cm length × 61 cm width × 41 cm height) that were located inside a room held in DD and maintained at 23 ± 2°C. Each large photoperiod box contained two 25-W incandescent light bulbs yielding an average intensity of 120 lx at the level of the birds.
LD cycles in all experiments were controlled by Tork Time Switches (model 8001). Continuous force-ventilation of all photoperiod boxes provided airflow and generated background noise that aided in the isolation between photoperiod boxes. All experiments were approved by the NCSU Animal Care and Use Committee and were conducted in accordance with the most current guiding principles for animal research (1).
In all surgical procedures, animals were anesthetized with an intramuscular injection of a mixture of ketamine hydrochloride (5–10 mg/kg body wt) and xylazine (2 mg/kg body wt). For measurement of body temperature, radiotransmitters were implanted in the right side of the peritoneal cavity of birds, below the last rib. Birds were blinded by the complete removal of both eyes. Briefly, eye removal was accomplished by dissecting the nictitating membrane away from the eye, using a needle and syringe to collapse the eye by removing the vitreous humor, and then cutting the ocular muscles and optic nerve. The eye was then removed, the orbital cavity was filled with gelfoam (Upjohn, Kalamazoo, MI), and the eye lids were sutured together. Blinded birds had no difficulty feeding and maintained normal body weights. Sham-blinded birds were subjected to a 10-mm incision through the skin at the top of the skull that was sutured immediately. Birds were administered a single daily injection of an analgesic (Banamine; 5 mg/kg body wt) for 3 days after all surgeries.
Body Temperature Measurement
Body temperature was continuously monitored via implanted transmitters (model VM-FH: Mini-Mitter, Sunriver, OR). A radio receiver (model RA 1010; Mini-Mitter) positioned directly beside each cage was connected to a computer dedicated to continuous data acquisition utilizing the Dataquest III Data Acquisition System (Data Sciences, St. Paul, MN). The transmitted frequency was proportional to body temperature with an accuracy of 0.1°C. The body temperature was measured at the end of each 10-min interval. Data analyses were performed using Circadia software (Behavioral Cybernetics, Cambridge, MA). The Circadia software was used to produce two types of graphical output: an actograph format in which a “pen deflection” was used to indicate each 10-min interval in which the body temperature exceeded the mean daily body temperature and a strip-chart format in which body temperature data were plotted against time. In the strip charts, the 10-min intervals were averaged into 30-min intervals.
Eye Patch Protocols
The behavior of the ocular pacemakers was tracked via the ocular melatonin rhythm or via the body temperature rhythm. Inexperiments 2 and 5 ocular melatonin levels were measured following a patching protocol that entrained the ocular pacemakers of individual quail out of phase, whereas, inexperiments 3 and 4 the body temperature rhythm was continuously monitored while the two ocular pacemakers in an individual quail were entrained out of phase (Expt. 3) or in phase (Expt. 4). Cone-shaped opaque patches (20 mm in diameter) were made from adhesive Band-Aids and painted with flat black acrylic paint. To increase adhesion to the bird, feathers were plucked from around the eyes at the time of patching, and rubber cement was used to hold eye patches in place. In the “alternate-patch (AP) protocol,” birds were held in constant light (LL) for 7 days, and, onday 8, an eye patch was placed over one eye for a 12-h period. After 12 h, the patch was removed and placed over the opposite eye for a 12-h period. This cycle of alternately patching the eyes lasted for 7 days, thereby exposing each eye to a 12:12-h LD cycle 180° out of phase with the opposite eye.
In the “in-phase (IP) patch protocol,” birds were held in LL for 7 days, and, on day 8, eye patches were placed over both eyes for a 12-h period. After 12 h, the patches were removed, and both eyes were exposed to light for a 12-h period. This cycle of patching the eyes in phase with one another lasted for 7 days. This protocol, therefore, exposed both eyes to the same 12:12-h LD cycle.
Subsequent to the patching protocols, the birds were placed in DD to assess the behavior of the ocular pacemakers as they either regained their normal phase relationship after the AP protocol or as they began freerunning in phase after the IP patching protocol. However, the disturbance caused by the twice daily placement and removal of the eye patches often obscured the body temperature rhythms. Therefore, in experiments in which body temperature was monitored, birds were exposed to a 1:11:1:11-h LDLD skeleton photoperiod (SP) for 14 days after the AP or IP patching protocol before being placed in DD. This LDLD cycle maintained the 12-h phase difference between the two ocular pacemakers engendered by the AP protocol, and it maintained the IP relationship between the two eyes after the IP patching protocol. Because the birds were not physically disturbed before being exposed to DD, the body temperature rhythms were more clearly discernable during the crucial period in which the ocular pacemakers were regaining their normal phase relationships. In all experiments in which SPs were used, the first 1-h light pulse was initiated immediately after removal of the last eye patch, and the last 1-h light pulse was given on the morning of day 14 of the 1:11:1:11-h LDLD cycle.
Blood melatonin levels.
Birds held in DD (Expt. 6) were removed from photoperiod boxes in the dark one time every 6 h on the day of blood sampling, and an opaque hood was placed over the bird's head. Blood samples (200–300 μl) were then taken under a dim light (∼3–5 lx) by pricking the brachial vein and collecting the blood in heparinized capillary tubes. The plasma was isolated from whole blood after centrifugation at 3,000 revolutions/min at 4°C and frozen at −80°C until RIA. Plasma melatonin levels were determined using a melatonin RIA kit from ALPCO Diagnostics (Windham, NH).
Ocular melatonin levels.
At the time of sampling, birds were transported to a nearby room in light-tight cages (Expts. 1, 2, and5). The birds were then decapitated within seconds of exposure to room light. The complete retina was removed from each eye, placed in a 1.5-ml microcentrifuge tube with 1 ml phosphate buffer solution, immediately frozen, and held at −80°C until RIA. Retinal melatonin levels were determined subsequently using a melatonin RIA kit from CIDTech Research (Cambridge, Ontario, CA).
Statistical tests were performed using the SAS and JMP statistical programs (SAS Institute, Cary, NC). For experiments 3 and 4 (see Fig. 6), the average phase was calculated by vector addition where the length of the average vector represents the scatter among phases. The Mardia-Watson-Wheeler test was used to determine whether distributions of freerunning phases between groups were significantly different (2). For experiment 6 (see Fig. 7), plasma hormone concentrations in the blinded group and sham-blinded group were analyzed for time, group (blinded group vs. sham-blinded group), and time by group effects using the repeated-measures ANOVA in the general linear models procedures of SAS (21). When group or time by group effects were found, differences between means of the two groups, within times, were assessed using ANOVA.
Experiment 1: Prolonged DD and ocular pacemaker coupling.
Adult female Japanese quail (n = 25) were implanted with radiotransmitters and held individually inside light-tight photoperiod boxes. Birds were exposed to an 8:16-h LD cycle (lights on at 0800) for 2 wk. On day 15, birds were placed in DD at 1600. On days 43–57 of DD, quail were killed atcircadian time (CT) 6 (midsubjective day) or CT 18 (midsubjective night), where CT 0marks the onset of the daily temperature rise above the mean daily temperature. Retinal tissues were collected, and ocular melatonin levels were determined via RIA.
Experiment 2: Effect of the AP protocol on the ocular melatonin rhythm.
Adult male Japanese quail were held in light-tight photoperiod boxes (n = 14/box) and exposed to a 12:12-h LD cycle (lights on at 0900) for 2 wk. On day 15, animals were placed in LL for 7 days. At 0900 on day 23, an AP protocol was initiated to entrain ocular pacemakers 180° out of phase. The patches were placed on the left eyes from 0900 to 2100 and on the right eyes from 2100 to 0900. On day 30, after 7 days of alternately patching the eyes, birds were placed in DD at 0900. Patches were not removed from eyes that were patched at the time of entry into DD. Animals were killed at 1500 (±1 h) on day 2(n = 14) and day 5 (n = 14) of DD. Ocular melatonin levels were determined via RIA.
Experiment 3: Effect of the APSP protocol on the body temperature rhythm.
Adult male Japanese quail (n = 12) were implanted with radiotransmitters and held individually in light-tight photoperiod boxes. Body temperature was monitored continuously while the birds were subjected to the AP protocol as described for experiment 2, exposed to 1:11:1:11-h LDLD cycles for 14 days to maintain the 180° phase relationship between the two ocular pacemakers, and then placed in DD. The transmitter batteries of five birds lasted long enough to expose the birds to an IP patching protocol as well (Expt. 4).
Experiment 4: Effect of the IPSP protocol on body temperature rhythm.
Adult male Japanese quail (n = 8, including 5 fromExpt. 3) were implanted with radiotransmitters and isolated in light-tight photoperiod boxes. Birds were exposed to a 12:12-h LD cycle (lights on at 0900) for 2 wk and, on day 15, placed in LL for 7 days. At 0900 on day 23, an IP patch protocol was employed that exposed both eyes to the same 12:12-h LD cycle. Onday 30, after 7 days of patching the eyes in phase with one another, the birds were placed on a 1:11:1:11-h LDLD SP for 14 days beginning at 0900 (IPSP protocol). On day 43 of the experiment, the birds were exposed to DD after the 1-h morning pulse. Ocular pacemaker phase was determined via body temperature.
Experiment 5: Effect of the APSP protocol on the ocular melatonin rhythm.
Adult male Japanese quail (n = 8) were held in a light-tight photoperiod box and exposed to a 12:12-h LD cycle (lights on at 0900) for 2 wk. On day 15, birds were placed in LL for 7 days. At 0900 on day 23, an AP protocol was initiated to entrain ocular pacemakers 180° out of phase. The patches were placed on the left eyes from 0900 to 2100 and on the right eyes from 2100 to 0900. On day 30, after 7 days of alternately patching the eyes, birds were exposed to 1:11:1:11-h LDLD light cycles for 14 days, which maintained the 180° phase relationship between the two ocular pacemakers. The birds were then exposed to DD and killed 18 h after the onset of the last 1-h light pulse (the projected midsubjective night of the right eyes). Melatonin levels in left and right eyes were subsequently measured.
Experiment 6. Effect of blinding on blood melatonin rhythms in DD.
Adult female Japanese quail were held inside light-tight photoperiod boxes, and blinding and sham surgeries were performed as described above. Blinded (n = 13) and sham-blinded (n = 14) birds were housed separately in two photoperiod boxes to ensure that sighted (sham-blinded group) birds did not injure blinded group birds. Birds were maintained on a 12:12-h LD cycle (lights on at 0900) for 14–16 days to allow complete recovery from surgery. After lights off on day 16, the birds were held in DD for 87 days. On day 87 of DD, blood samples were taken every 6 h from the brachial vein of each bird beginning at 1200 over the course of a 24-h sampling period. Plasma melatonin levels were determined via RIA.
Experiment 1: Ocular Pacemakers in Prolonged DD
Japanese quail exhibit a robust rhythm of body temperature that persists indefinitely in DD (25). Furthermore, the eyes of quail are the sites of ocular clocks that drive a rhythm of ocular melatonin synthesis: melatonin levels are high at night and lower during the day (26). If ocular clocks act as pacemakers driving rhythmicity throughout the bird, it would be predicted that the pacemakers in the left and right eyes of individual quail would remain coupled and maintain the same phase relationship with each other in constant conditions. Otherwise overt rhythms, such as body temperature, would be unable to maintain a coherent rhythm in DD. This prediction was tested by killing birds that had been exposed to DD for 43–57 days during either the midsubjective day or the midsubjective night and measuring the ocular melatonin levels. Because it was not possible to track the ocular melatonin rhythm continuously, the daily body temperature rhythm was used to assess the phase of the ocular melatonin rhythm. Melatonin levels were significantly higher (P ≤ 0.0001, paired t-test) in retinas collected in the midsubjective night [856 ± 20 (SE) pg/eye] compared with retinas collected in the midsubjective day (444 ± 36 pg/eye; Fig.1). In addition, ocular melatonin levels between the left and right eyes of individual birds were strongly correlated (r = 0.76). If the melatonin contents of left and right eyes were correlated perfectly regardless of the phase of the melatonin rhythm at which the birds were sampled, the regression line would show a slope of one. The slope of the regression line shown in Fig. 1 (slope = 0.8) is close to one.
Experiment 2: Effects of the AP Protocol on Ocular Melatonin Levels
Quail were subjected to an AP protocol for 7 days and then placed in DD. The birds were then killed either on day 2 or5 of DD, and the ocular melatonin levels in both eyes were measured subsequently. Figure 2 shows that ocular melatonin levels between the left and right eyes were significantly different when the birds were killed during the projected midnight of the left eye on day 2 of DD (P ≤ 0.001). By day 5 of DD, however, the interocular melatonin levels were not significantly different. The interocular differences in melatonin content on day 2 of the AP birds were significantly different from the day 5 AP birds (P ≤ 0.001, unpaired t-test) and from theexperiment 1 birds (P ≤ 0.001). There were no significant differences between the experiment 1 birds and the day 5 AP birds.
The data from experiment 2 show that the AP protocol entrained the left and right eyes out of phase, and the eyes were still significantly out of phase on day 2 of DD. By day 5 of DD, however, the eyes had regained their normal phase relationship: the average phase difference between the two eyes is minimal and is similar to that observed in normal birds in DD (Expt. 1) in which the eyes had not been previously driven out of phase.
Experiment 3: Effects of the APSP Protocol on Body Temperature
Figure 3 shows representative actograph and strip-chart records of a bird subjected to an AP protocol, followed by exposure to an SP regimen and then DD. Two small peaks in body temperature are seen during the first 1–2 days of DD in the actograph and strip chart. After 5–6 days in DD, however, a single robust daily rhythm in body temperature is observed (Fig.3 B). In addition, the length of time (in h) that the body temperature exceeds the mean daily temperature (α) shows a steady decline for the first 5–6 days in DD. This pattern is consistent with the behavior of two pacemakers that are 12 h out of phase during their first day of DD and regain their normal phase relationship over the next 5 days. The shortening in α demonstrates a mutual interaction between the two pacemakers: one phase advances and the other phase delays until they regain their normal phase relationship.
Two objective measures were used to assess the time until recoupling in the birds subjected to the APSP protocol. First, using the actograph records, the amount of time (in h) from the onset of the daily rise in body temperature to the end of the daily fall in body temperature (α) was measured for the first 10 days of DD (Table1). During the first 1–2 days of DD, when two daily peaks of body temperature could often be observed, α was measured from the onset of the first peak to the offset of the second peak. The α values showed a steady decrease from days 1 to 5 of DD, followed by a steady-state α, suggesting that it took an average of 5–6 days for the two ocular pacemakers to regain their normal phase relationship. The second objective measure used to calculate recoupling time involved a line drawn through the steady-state onsets of the daily rise in body temperature in DD and retrojected to the earliest day of exposure to DD (e.g., Fig. 3 A). The earliest day on which the retrojected line coincided with the onset of the rise of body temperature above the mean daily body temperature (day 5 of DD in Fig.3 A) was assumed to be the day on which the two ocular pacemakers regained their normal phase relationship. With the use of this criterion, the average number of days (±SE) to recoupling was 6.1 ± 0.4 (range 5–8, n = 9). It was not possible to calculate these two objective measures in two of the birds.
A line eye fitted through the onsets of the daily steady-state body temperature and retrojected to the first day of DD for the bird shown in Fig. 3 A falls at projected CT 8 (whereCT 0 is 12 h after the onset of the last 1-h light pulse). The retrojected phases of the 11 birds subjected to the APSP protocol are shown in Fig. 6; the average phase (±SE) is CT 6.5 ± 0.5 (r = 0.91). One bird was not included because it took significantly longer than normal to recouple (Fig.4). These data support the hypothesis that the two ocular pacemakers had an equal effect on one another; that is, upon release in DD, the two pacemakers mutually interacted via successive, and equal, phase advances and delays until they recoupled within 5–6 days. Consequently, the retrojected steady-state phase of the recoupled pacemakers falls midway between the phases the two pacemakers exhibited when they were 12 h apart.
The two ocular pacemakers of one bird out of all the birds subjected to the AP protocol (n = 12) took significantly longer to regain their normal phase relationship (Fig. 4). Clearly the coupling strength between these two pacemakers in this bird was significantly weaker than that exhibited by most birds. Although the underlying reason for the weaker coupling is unknown, the record of this bird is remarkable insofar as it clearly demonstrates that the AP protocol entrained the two ocular pacemakers 12 h out of phase, that is, this bird exhibits a “slow motion” display of the coupling interaction between the two ocular pacemakers. The initial 12-h phase difference between the two pacemakers gradually diminishes (∼26 min/day) until they regain their normal phase relationship on aboutday 28 of DD.
The average daily body temperatures of the birds subjected to the APSP protocol showed a decline of 0.18°C from day 1 of DD today 6 of DD (Table 1).
Experiment 4: Effects of the IPSP Protocol on Body Temperature
Figure 5 shows a representative example of an actograph and strip-chart record of a bird subjected to an IP patching protocol followed by exposure to the SP regimen and then to DD. Because the IP protocol entrained both eyes to the same phase, both ocular pacemakers presumably would choose the same SP as day and, furthermore, this “day” should correspond to the day phase produced by the IP protocol. This interpretation is supported by the fact that the retrojected line drawn through the steady-state body temperature rise onsets falls near the onset of the last dawn pulse. Figure6 plots the phases of the retrojected body temperature onsets of birds (n = 8) subjected to the IPSP protocol and shows that the phases cluster around the dawn pulse (0.5 ± 0.2 h after the onset of the last light pulse,r = 0.99).
The data, therefore, are fully compatible with the hypothesis that both ocular pacemakers were entrained to the same phase by the IPSP protocol (the day phase of the IP protocol) and began freerunning from that phase when placed in DD. Interestingly, the body temperature rhythm is less robust during the first 3–4 days of DD in all of birds subjected to the IPSP protocol (e.g., Fig. 5).
Experiment 5: Effects of the APSP Protocol on Ocular Melatonin Levels
The behavior of the body temperature rhythm in DD after the APSP protocol is fully compatible with the hypothesis that this protocol entrained the two ocular pacemakers 12 h out of phase. To confirm that the eyes were entrained out of phase, quail were subjected to the APSP protocol, placed in DD, and killed 18 h after the onset of the last 1-h light pulse (the projected midsubjective night of the right eyes). Ocular melatonin levels were determined subsequently. The average melatonin levels in the left and right eyes were 1,363 ± 176 and 1,974 ± 215 (SE) pg/eye, respectively, and were significantly different (P ≤ 0.001, pairedt-test). The mean interocular difference between the eyes was 611 ± 119 pg/eye and was significantly different compared with the mean interocular differences in control birds (Expt. 1) that did not receive an AP protocol (P ≤ 0.0001, unpaired t-test).
Experiment 6: Blood Melatonin Rhythms in DD
Figure 7 shows the plasma melatonin levels of blinded and sham-operated birds after the birds were held in DD for 87 days. A significant difference was found when comparing melatonin levels across time (P = 0.0011), and, furthermore, when analyzed between subject groups (treatments), there was a highly significant difference (P > F = 0.0001). Normally, both the eyes and the pineal synthesize and secrete melatonin in the blood, so it is not surprising that blinded birds had significantly lower mean melatonin levels, measured over 24 h, than sham-blinded birds (P ≤ 0.02,t-test). Blinded animals exhibited no detectable melatonin rhythm in DD, showing that blinding by complete eye removal disrupted the pineal's ability to synthesize and secrete melatonin rhythmically. One blinded bird was eliminated from analysis, since its peak blood melatonin levels were more than three SD from the group mean.
In mammals, the only photic input pathway to the circadian system is via the eyes, and the main circadian pacemaker is located in the SCN (16, 31). By contrast, birds possess at least three photic input pathways (the retina, the pineal organ, and extrapineal, extraretinal receptors in the brain), and circadian pacemakers located in the pineal and/or eyes drive circadian oscillators located in the SCN (29, 30). The SCN in most of the birds tested so far is less robust than the mammalian SCN insofar as the avian SCN cannot maintain activity or body temperature rhythms in constant conditions without daily input from the pineal or eyes.
Previously, we demonstrated that the eyes of Japanese quail are the loci of biological clocks that drive a daily rhythm of melatonin synthesis and secretion (26). Furthermore, we suggested that this clock is a circadian pacemaker, that is, it drives circadian rhythms outside of the eye itself because eye removal caused the activity and body temperature rhythms to decay into arrhythmicity in DD (25). Insofar as quail can maintain a robust circadian rhythm of body temperature in prolonged DD (25), the two putative ocular pacemakers in an individual bird must maintain the same phase; otherwise, a single circadian output could not be generated. We therefore sought to strengthen our hypothesis that the eyes are the circadian pacemakers in quail by determining if the ocular melatonin rhythm in the two eyes of individual quail maintain phase in prolonged DD. Melatonin levels in the eyes of birds killed in the middle of their subjective night were significantly higher than melatonin levels in the middle of their subjective day, and there was a strong correlation of melatonin levels between the eyes of individual birds (Fig. 1). These results showed that the eyes remained in phase in prolonged DD and are compatible with the hypothesis that the bilaterally distributed ocular clocks are pacemakers.
The strength of coupling between these bilaterally distributed pacemakers was also examined. In this discussion, the use of the term “coupling” is meant to describe the mechanism whereby the two eyes maintain a mutual phase relationship and is not meant to imply necessarily that the two eyes are directly communicating one with another. The coupling pathway, for example, may involve inputs from the central nervous system. Two approaches were used to determine the strength of coupling: either ocular melatonin levels or the body temperature rhythm was measured in birds subjected to an AP protocol. In experiment 2, an AP protocol was used to drive the ocular pacemakers of individual birds 12 h (180°) out of phase with each other, and the birds were then placed in DD to determine how rapidly the two eyes could regain the same phase relationship. Birds were then killed on day 2 or 5 of DD at the projected midday or midnight of the right and left eyes, respectively, and the ocular melatonin levels were measured subsequently (Fig. 2). Onday 2 of DD, the average ocular melatonin level in the subjective day eyes of individual birds was significantly lower than the average melatonin level in the subjective night eyes, but, byday 5, there was no significant difference between the average melatonin levels in the two eyes. Therefore, these data support the hypothesis that ocular pacemakers that are driven 180° apart can regain their normal phase relationship within 5 days.
The use of ocular melatonin rhythm to assess coupling strength had some significant disadvantages, for example, large numbers of birds had to be used, there was variability in melatonin levels between birds, and it was not possible to track the motion of the ocular pacemakers in individual birds. Consequently, we developed a second approach whereby each ocular pacemaker could be monitored continuously via body temperature. In these experiments, body temperature was monitored while the birds were subjected to the AP protocol for 7 days, placed in a 1:11:1:11-h LDLD lighting regimen for 14 days, and then exposed to DD. Each ocular pacemaker drove a separate peak in body temperature during the AP protocol that subsequently fused into a single body temperature rhythm in DD.
The LDLD lighting protocol was designed to maintain the 180° phase difference between the two ocular pacemakers that had been established by the AP protocol. A 1:11:1:11-h LDLD cycle would act as a “skeleton” of a 12-h photoperiod, and it is well known that circadian clocks can entrain to SPs. Furthermore, when confronted with SPs that are comprised of short (1-h) light pulses 12 h apart (1:11:1:11-h LDLD), an animal potentially could choose either of the 1-h light pulses as dawn of the 12-h SP; this has been termed the “bistability” phenomenon (10, 19). However, which SP is actually chosen is determined by the circadian time of the first light pulse the animal experiences (10). In the present experiment, each eye is 180° out of phase; consequently, when exposed to a 1:11:1:11-h LDLD cycle, the circadian clock in each eye experiences the first pulse at a circadian phase that is 180° out of phase with the other eye. Therefore, it would be predicted that the circadian pacemaker in one eye would choose the first light pulse as dawn and the pacemaker in the opposite eye would choose the second light pulse as dawn.
The use of the LDLD protocol before DD proved to be superior over going directly from the AP routine into DD because the twice daily disturbance caused by placing and removing the eye patches tended to disrupt the expression of the body temperature (unpublished data). These handling effects were assumed to represent effects on the expression of the body temperature rhythm but not on the circadian clocks driving the rhythm.
The expectation that the behavior of the ocular pacemakers could be assessed by observing the body temperature rhythm was confirmed. During exposure to a 1:11:1:11-h LDLD cycle after an AP protocol, a rise in body temperature is associated with both light pulses; presumably, each body temperature peak is being driven by a separate ocular pacemaker (Fig. 3). The bimodal appearance of the body temperature rhythm under a 1:11:1:11-h LDLD cycle, however, is not compelling evidence that the APSP protocol entrained the ocular pacemakers because a positive masking effect of light on body temperature would produce a similar pattern. However, the behavior of the body temperature rhythm(s) after introduction to DD is entirely consistent with the hypothesis that the APSP protocol entrained the two ocular pacemakers out of phase. Often, a bimodal body temperature pattern is seen during the first 1–2 days of DD after the APSP protocol. In addition, it is apparent that two components are interacting because there is a phase delay of the daily body temperature rise and a phase advance of the daily body temperature fall until a steady state is attained in ∼5–6 days (Fig. 3 and Table 1). Significantly, a retrojection of the phase of the daily rise in body temperature onsets of the steady-state freerun fell midway between the projected onsets of the two light pulses comprising the SP (Figs. 3 and 6). These results support the hypothesis that the two ocular pacemakers were held 12 h apart by the LDLD cycle and that the two pacemakers were equal in strength; each phase advanced or delayed by the same amount (∼1.2 h) each day until they regained their normal phase relationship.
During the first 1–6 days in DD, the two ocular pacemakers are presumably driving two subpopulations of SCN oscillators that, in turn, are driving two body temperature rhythms via their control over metabolism and thermal conductances. When the two pacemakers are out of phase, the heat-generating/conserving mechanisms driven by one pacemaker are being opposed by the other. Interestingly, mean daily body temperatures are significantly higher during the first 1–6 days of DD (Table 1), implying that the mechanisms promoting heat generation and/or heat conservation are more effective than those promoting a decrease in body temperature when the pacemakers are out of phase. When the two ocular pacemakers recouple in their normal phase relationship, all of the SCN oscillators are also driven to the same mutual phase relationship; consequently, the SCN-driven body temperature rhythm becomes unimodal and more robust.
Several other observations are also consistent with the conclusion that the AP protocol followed by a 1:11:1:11-h LD cycle successfully entrained the two ocular pacemakers 180° out of phase. 1) If both eyes were patched on the same phase, the pacemakers in both eyes chose the same SP as day, and, when placed in DD, the body temperature rhythm lacked the daily phase advances and delays characteristic of the AP birds (Fig. 5). Furthermore, this rhythm began freerunning from the day phase of the SP (Fig. 6). Somewhat surprisingly, however, the body temperature rhythm was markedly less robust during the first 3–4 days in DD. It is possible that, during this period, the ocular pacemakers were “recruiting” SCN oscillators that were initially out of phase with the ocular pacemakers. It is known that SCN oscillators can be directly “entrained” by light perceived by extraretinal photoreceptors (29) and the SP may have directly forced some SCN oscillators out of phase with the ocular pacemakers. 2) Ocular melatonin levels were measured in the eyes of birds that were subjected to an APSP protocol (Expt. 5), and the results showed that the melatonin levels in the left eyes were significantly different from the melatonin levels in the right eyes, that is, the patching protocol drove the two ocular pacemakers out of phase.3) In one bird, the two ocular pacemakers took ∼28 days to regain their normal phase relationship in contrast to the 5–6 days seen in all other birds (Fig. 4). In this case, it is abundantly clear that the two ocular pacemakers were indeed driven 12 h apart by the patching protocol.
The results of the AP experiments using either the ocular melatonin rhythm or the body temperature rhythm to determine the behavior of the ocular pacemakers are entirely consistent: both experiments demonstrated that the ocular pacemakers can be driven 180° apart, and, when released in DD, the pacemakers regain their normal phase relationship within 5–6 days. The rapidity with which phase is regained suggests that the two pacemakers are tightly coupled one to another. The mechanism by which the eyes communicate phase information to each other is currently unknown. It is possible that the eyes are communicating via the cyclic release of melatonin. The quail retina not only synthesizes and releases melatonin in the blood (27), but the avian retina also contains melatonin receptors (20). One could envision a mechanism whereby the retina not only releases melatonin but is also sensitive to the phase-shifting effects of melatonin. If the eyes are driven out of phase, daily melatonin release by the eyes (or pineal) may cause mutual phase advances or delays of the two ocular pacemakers until they regain their normal phase relationship. Alternatively, neural inputs (or outputs), or the SCN, may be involved in ocular pacemaker coupling. The eye possesses a single neural output pathway, the optic nerve (which includes the RH tract to the SCN), and a number of neural input pathways. Among the neural inputs to the eyes are autonomic inputs from the superior cervical ganglia and from the nucleus of Edinger-Westphal via the ciliary ganglia (4, 11, 12). Both the superior cervical ganglia and the nucleus of Edinger-Westphal receive input from the SCN in birds (4, 12).
Our current model of the circadian system of the Japanese quail postulates that the ocular pacemakers are communicating with the SCN via both neural and hormonal outputs because about three-quarters of quail subjected to optic nerve section remain rhythmic, although rhythmicity is less robust than normal birds, but all birds subjected to complete eye removal are arrhythmic in DD (25). Because optic nerve section does not prevent the eyes from synthesizing and secreting melatonin in the blood and because quail can be entrained by daily melatonin administration, the daily release of melatonin by the eyes may be the hormonal link between the eyes and the SCN (30). Normally, quail exhibit robust rhythms of melatonin synthesis and secretion by both the eyes and the pineal with higher melatonin levels at night: on a 12:12-h LD cycle, about two-thirds of the blood-borne melatonin is secreted by the pineal and one-third by the eyes (27). Melatonin is highly lipid soluble and is secreted as soon as it is synthesized, although some ocular melatonin may be metabolized within the eye and not secreted in the blood (27).
If our model is correct and a daily melatonin input is capable of keeping the oscillators in the SCN from damping out, or drifting out of phase, both ocular and pineal melatonin must be normally involved in maintaining circadian organization. The pineal of Japanese quail may not be the location of an autonomous clock, as it is in some other avian species, because melatonin rhythmicity does not persist in constant conditions in vitro, although rhythmicity can be driven directly by a 24-h LD cycle (17). However, pineal melatonin rhythmicity persists in constant conditions in vivo because pineal metabolism is driven by daily neural inputs from the SCN (5). Therefore, daily melatonin secretion from the pineal should be capable of maintaining rhythmicity of the SCN in DD. Why, then, are eyeless quail arrhythmic in DD, since they still have their pineal organs? We hypothesized that, in fact, the pineal organs of eyeless quail do not synthesize and secrete melatonin rhythmically in DD. In the absence of the ocular pacemakers, the SCN is unable to maintain rhythmicity; consequently, the daily neural output from the SCN to the pineal is disrupted, which, in turn, disrupts the pineal's ability to synthesize and secrete melatonin rhythmically. This hypothesis was tested by measuring daily blood melatonin levels in sham-operated quail and eyeless quail that had been held in DD for 87 days (Fig. 7). The results confirmed our hypothesis: sham-operated birds were still capable of showing a daily rhythm of melatonin in the blood, but birds in which the eyes were removed were not. The pineal organs of eyeless birds secreted significant amounts of melatonin in the blood, but this secretion did not exhibit circadian rhythmicity. These results also strengthen the hypothesis that the ocular clocks are the dominant pacemakers in the quail's circadian system: in the absence of the ocular pacemakers, neither pineal rhythmicity nor body temperature rhythmicity can persist in DD. In all likelihood, all overt circadian rhythmicity in DD is dependent on ocular pacemaker input to the central circadian system.
Calculations of the average melatonin levels in sham-operated and eyeless birds showed that the eyeless birds secreted about one-half of the amount of melatonin secreted by sham-operated (both eyes and pineal intact) birds, indicating that, in normal birds in DD, the eyes contribute one-half of the blood-borne melatonin and the pineal contributes the remaining one-half. When exposed to a 12:12-h LD cycle, the relative contributions of the pineal and the eyes to blood-borne melatonin are two-thirds and one-thirds, respectively (27).
Evidence that the eyes are the loci of master circadian pacemaker is compelling and includes the observations that: 1) the eyes contain a clock that drives the rhythm of ocular melatonin synthesis, and, presumably, this clock is the pacemaker that drives the other overt rhythms in the body, 2) eye removal abolishes rhythmicity in DD, including rhythms of body temperature, activity, and pineal melatonin, 3) the ocular pacemakers remain coupled in prolonged DD, 4) the ocular clocks can be entrained 180° out of phase, and each clock can be shown to drive a discrete component of the body temperature rhythm, and 5) out-of-phase clocks will rapidly recouple when exposed to constant conditions.
We thank Dr. Fred Davis for statistical assistance.
This work was supported by National Institute of Neurological and Communicative Disorders and Stroke Grant RO1 NS-20961 (to H. Underwood).
Address for reprint requests and other correspondence: C. T. Steele, NC State Univ., Dept. of Zoology, Box 7617, Raleigh, NC 27695-7617 (E-mail:).
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. Section 1734 solely to indicate this fact.
October 3, 2002;10.1152/ajpregu.00447.2002
- Copyright © 2003 the American Physiological Society