Articles in PresS. Am J Physiol Regul Integr Comp Physiol (March 18, 2009). doi:10.1152/ajpregu.90914.2008
Phenobarbital blockade of the preovulatory LH surge: Association with phaseadvanced circadian clock and altered suprachiasmatic nucleus Period1 gene
expression
Sandra J. Legan1, Kathleen M. Donoghue2, Kathleen M. Franklin2 and Marilyn J. Duncan2
Departments of Physiology1 and Anatomy and Neurobiology2,
University of Kentucky Medical Center, Lexington, KY 40536-0298
Running head: Does phenobarbital phase-advance the circadian clock?
Address correspondence to:
Dr. Sandra J. Legan
Department of Physiology
University of Kentucky
Lexington, KY 40536-0298
[email protected]
ph (859) 323-6277
fax (859) 323-1070
Copyright © 2009 by the American Physiological Society.
1
ABSTRACT
The suprachiasmatic nucleus (SCN) controls the timing of the preovulatory LH surge in
laboratory rodents. Barbiturate administration during a critical period on proestrus delays the
surge and prolongs the estrous cycle one day. Because a non-photic timing signal (zeitgeber)
during the critical period that phase advances activity rhythms can also induce the latter effect,
we hypothesized that barbiturates delay the LH surge by phase-advancing its circadian timing
signal beyond the critical period. In Experiment 1, locomotor rhythms and estrous cycles were
monitored in hamsters for 2-3 weeks pre- and post-injection of vehicle or phenobarbital and post
transfer to darkness at zeitgeber time (ZT) 6 on proestrus. Phenobarbital delayed estrous cycles
in 5 of 7 hamsters, which exhibited phase shifts that averaged two-fold greater than those
exhibited by vehicle controls or phenobarbital-injected hamsters with normal cycles. Experiment
2 used a similar protocol, but injections were at ZT 5, and blood samples for LH determination
were collected from 1200-1800 h on proestrus and the next day via jugular cannulae inserted the
day before proestrus. Phenobarbital delayed the LH surge one day in all 6 hamsters, but it
occurred at an earlier circadian time, supporting the above hypothesis. Experiment 3
investigated whether phenobarbital, like other non-photic zeitgebers, suppresses SCN Period1
and Period2 transcription. Two hours post-injection, phenobarbital decreased SCN expression
of only Period1 mRNA, as determined by in situ hybridization. These results suggest that
phenobarbital advances the SCN pacemaker governing activity rhythms and hormone release in
part by decreasing its Period1 gene expression.
KEYWORDS
critical period, activity rhythm, estrous cycle, barbiturate, Per1, Per2
2
INTRODUCTION
In laboratory rodents, the timing of the preovulatory LH surge is controlled in part by a circadian
neural signal that is entrained by the daily light:dark (L:D) cycle, or photoperiod. Under a
standard laboratory photoperiod, such as 12L:12D or 14L:10D, the preovulatory LH surge occurs
in regularly cycling rats and hamsters about once every 4 days (96 hours) at the same time of day
(4; 6; 8; 51). The daily occurrence of a neural signal for triggering the LH surge was established
over 50 years ago by the finding that daily administration of barbiturates within a specific time
period on the afternoon of proestrus for 1-3 days prevents ovulation or the LH surge for 1-3 days,
respectively, thereby delaying the estrous cycle up to 3 days (16). Based on this finding, the
“critical period” was defined as the time on proestrus when barbiturate administration blocks the
preovulatory LH surge and ovulation, and has been interpreted to indicate that interval during
which the neural trigger for the LH surge occurs. The critical period begins about 2 or 3 hours
before the first detectable increase in circulating LH levels in proestrous Syrian hamsters (55) or
rats (19), respectively, and lasts for about 2 hours in both species (17; 40).
A variety of evidence demonstrates the existence of circadian regulation of the LH surge. For
example, daily LH surges occur in anestrous (4; 7; 47) or untreated ovariectomized Syrian
hamsters (57) and in estradiol-treated ovariectomized rats (5; 11; 23; 28) and hamsters (39). In
addition, persistence of the LH surge with a free-running rhythm during exposure to constant
light [rats (32)] or constant darkness [hamsters (54)] provides evidence that an endogenous
circadian pacemaker controls the neural signal for the LH surge. Furthermore, lesions of the
suprachiasmatic nucleus (SCN), site of the master circadian pacemaker in mammals, block the
LH surge and estrous cycles (56; 61). Similar to the regulation of other circadian rhythms, the
circadian rhythm in the neural signal for the LH surge is entrained by the light-dark cycle.
3
During exposure to different photoperiods, LH surges occur at a consistent time relative to the
midpoint of the dark phase, similar to activity onset (31; 36; 37), and can be phase advanced or
delayed by corresponding shifts in the light-dark cycle (36).
It has long been known that whenever the LH surge is temporarily blocked or delayed, there is a
delay in the estrous cycle. Thus in addition to the barbiturate-induced delay of estrous cycles,
the delay in LH surges that is caused by exposure to constant light (32) and the delay in LH
surges occurring in middle-aged rats (62; 63) are also associated with lengthened estrous cycles.
More recently a phase advance induced by a non-photic stimulus was also shown to lengthen
estrous cycles (25). Thus, transferring hamsters to a clean home cage or placing them in a novel
running wheel near the beginning of the critical period on proestrous afternoon can delay estrous
cycles by one day, but only in those animals whose circadian activity rhythm is phase advanced
more than ~1.5 hours (25). In conjunction with the finding that pentobarbital administration to
some strains of male mice phase can advance the locomotor activity rhythm (14), the latter
finding suggests that phenobarbital-induced delay of the estrous cycle in female hamsters is also
mediated by a phase advance. Therefore, we hypothesized that administration of phenobarbital
during the critical period on proestrous afternoon delays the LH surge and the estrous cycle by
advancing the phase of the circadian pacemaker past the critical period.
With regard to the mechanism by which phenobarbital advances the circadian pacemaker, other
nonphotic stimuli (e.g., transfer to a novel wheel, dark exposure, or injection of triazolam or 8OH-DPAT) that advance the phase of the circadian pacemaker during the midsubjective day
appear to re-set the clock by decreasing SCN expression of Period1 (Per1) and Period2 (Per2)
mRNAs (13; 24; 30; 33). The finding that SCN administration of Per1 antisense
4
oligonucleotides induces nonphotic-like phase advances in the hamster locomotor activity
rhythm demonstrates the mechanistic importance of this decrease in Per1 expression (22). Based
on these findings in male rodents, we also tested the hypothesis that phenobarbital administration
during the critical period on proestrus suppresses Per1 mRNA expression in the female hamster
SCN. Finally, many extra-SCN brain regions exhibit circadian expression of Per1 and Per2 (3;
21; 65), but whether nonphotic phase-resetting signals affect expression of these genes in extraSCN regions has not been reported. Therefore, we investigated the effect of phenobarbital on
expression of Per1 and Per2 mRNAs in several other brain regions.
MATERIALS AND METHODS
All procedures were performed according to the AALAC Guide and were approved by the
University of Kentucky Institutional Animal Care and Use Committee (IACUC).
Animals: Adult female Syrian hamsters were obtained from Harlan (Indianapolis, IN) and were
maintained between 20 and 22° C under a 14L:10D photoperiod (lights on at 0600 h). Light
intensity ranged between 100 and 500 lux. The hamsters were housed individually in cages
equipped with running wheels that were electronically interfaced with a computer. Wheel
revolutions were recorded using hardware and software from the Chronobiology Kit (Stanford,
CA). The cages were enclosed in light-tight, sound-attenuated, ventilated compartments that
contained samples of soiled bedding from male hamsters. Food and water were available ad
libitum. Estrous cyclicity was determined by daily examination of vaginal discharge for
viscosity, color and odor. On the day of estrus, the vaginal discharge is stretchy, white and
pungent (42).
5
Experiment 1: Effect of phenobarbital administration during the critical period on estrous cycles
and phase of the locomotor activity rhythm.
To determine whether administration of phenobarbital during the critical period on proestrous
afternoon delays the estrous cycle by causing a phase advance in the circadian pacemaker,
estrous cycles and activity rhythms were monitored daily. Hamsters exhibiting at least two
consecutive estrous cycles were injected subcutaneously (s.c.) with vehicle (1% ethanol in 0.9%
NaCl) or phenobarbital (100 mg/kg in vehicle) on proestrus at 1400 h, 6 hours before lights off
[zeitgeber time (ZT) 6], which is near the onset of the critical period. ZT 12 is conventionally
defined as the time of lights off. This dose of phenobarbital induces drowsiness but not
unconsciousness, as assessed by responses to toe-pinches or observations of changes in the
animal’s position or location in her cage every 30-40 minutes for the next 3 hours. Immediately
after injection, all hamsters were exposed to constant darkness (DD), and wheel running rhythms
and vaginal discharge were recorded daily using dim red light (< 10 lux) for the next 2-3 weeks.
Experiment 2: Effect of phenobarbital administration on estrous cycles, phase of the locomotor
activity rhythm, and timing of the LH surge.
Based on the results of Experiment 1, we proceeded to determine whether administration of
phenobarbital during the critical period on proestrous afternoon delays the estrous cycle by
rapidly phase advancing the timing signal for the LH surge past the critical period, such that the
LH surge in DD occurs one day later but at an earlier time. Hamsters were treated similarly as in
Experiment 1, with the following two differences. On diestrus, hamsters that had exhibited at
least 3 consecutive estrous cycles were anesthetized with isoflurane and fitted with right atrial
cannulae for serial blood sampling. On proestrus (Day 1) vehicle (1% ethanol in sterile 0.9%
NaCl) or phenobarbital (100 mg/kg in vehicle) was injected s.c. one hour earlier than in
6
Experiment 1, i.e., at 1300 h, 7 hours before lights off (ZT 5). Immediately after the injections,
all hamsters were exposed to constant darkness (DD), and activity rhythms and appearance and
characteristics of vaginal discharge were recorded daily using dim red light for the next 2-3
weeks. Blood samples (0.2-0.3 ml) were obtained for determination of circulating LH
concentrations at 1200 h, and hourly from 1400 to 1800 h on the day of the expected LH surge,
and at 1400, 1600 and 1800 h on the other day. In DD conditions, samples were obtained using
dim red light. To avoid possible effects of handling on the phase of the circadian clock, in some
animals in each group, samples were collected via an extended cannula. After each blood
sample was collected, it was centrifuged in a microfuge at ~6000 rpm (~2900 x g) for 10
minutes, and the plasma was drawn off and stored at -20° C until the time of radioimmunoassay.
The red blood cells were reconstituted in a volume of sterile heparinized saline (10 IU/ml) equal
to 50% of the total sample volume and replaced after the next sample was withdrawn in order to
maintain the hematocrit within the normal range. Only sterile syringes, collection tubes and
pipet tips were used to collect samples and reconstitute the red blood cells.
Experiment 3: Effect of phenobarbital administration on Per1 and Per2 expression in select brain
areas.
After exhibiting at least two consecutive estrous cycles, either vehicle (1% ethanolic saline) or
phenobarbital (100 mg/kg in 10 ml vehicle) was injected subcutaneously on proestrus at 1300 h
(ZT 5), as in Experiment 2. The hamsters were decapitated 2 h later, based on previous findings
that other pharmacological agents, e.g., triazolam and 8-OH-DPAT, that act as nonphotic
resetting signals suppress SCN Per1 mRNA at 2 but not at 1 or 4 hours after treatment (24). The
brains were removed and frozen with powdered dry ice and stored at -80o C. Coronal sections
(20 µm thick) through the SCN were cut with a cryostat, mounted onto negatively-charged
7
slides, and stored at -80o C until processed by in situ hybridization. Per1 and Per2 mRNA
expressions were assessed in the SCN and in hippocampal, cortical, and thalamic regions, based
on previous reports that these brain regions exhibit circadian rhythms in clock gene expression
(1; 29; 60).
Circadian Rhythm Analysis: Phase shifts were calculated by the linear regression analysis
method using the computer program, ClockLab (Actimetrics, Willmette, IL). Regression lines
were fitted through the daily onsets of activity for seven days before and eight-nine days after
each treatment. A phase shift was defined as the difference between the onset of activity
predicted by the pre-treatment regression line and the actual onset shown by the post-treatment
regression line. Positive numbers represent phase advances while negative numbers represent
phase delays. Phase shift data were analyzed by t-test, or in the case of more than two
experimental groups, by ANOVA. When the main effects were significant as indicated by
P<0.05, Bonferroni’s Test was used for post hoc analyses.
In Experiment 1, periodogram analysis (Clocklab) was used to determine endogenous period
length of the circadian activity rhythms expressed in constant darkness after the injections. The
endogenous period length (mean ± S.E.M.) was virtually identical for the two treatment groups:
Veh, 23.7 ± 0.05 versus Pheno, 23.7 ± 0.08 h (P=0.68). In view of the absence of any effect of
phenobarbital on period length in Experiment 1, this parameter was not determined in
Experiment 2.
Radioimmunoassay: Plasma LH concentrations were determined in 2.5 to 75 μl aliquots of
plasma by means of an RIA described previously (27). The first antibody was CSU 120 (kindly
8
provided by Dr. Terry Nett, Colorado State University, Fort Collins, CO), used at a working
dilution of 1:10,000 in 1:100 normal rabbit serum (Millipore Corp., St. Charles, MO, formerly
Linco Research Inc.) in 0.05 M phosphate-buffered saline (PBS) containing 0.1% gelatin (gelPBS). The tubes were incubated at 4° C for 48 h following the addition of 100 μl first antibody
and after adding radiolabeled LH (~22-32,000 cpm/100 μl diluent, MP Biomedicals LLC, Santa
Ana, CA), and for 72 h after addition of 200 μl second antibody (anti-rabbit gamma globulin,
Milllipore Corp., St. Charles, MO, 1:50 in gel-PBS). Plasma LH concentrations are reported in
terms of ng RP-3 per ml (National Hormone and Peptide Program, NIDDK, and Dr. Parlow).
All samples were analyzed in 2 assays for which the sensitivity (100% - 2 SD of maximum
binding) averaged 0.17 ng/tube, and the interassay coefficient of variation was 8.8%.
LH Data Analysis
Mean peaks of LH surges, i.e. the maximum levels attained, and the times at which the peak
levels occurred were compared between treatment groups by Student’s t tests.
In situ Hybridization
In situ hybridizations for Per1 and Per2 mRNAs were conducted as described previously (13).
Riboprobes were produced using partial cDNA sequences of hamster Per1 (nucleotides 7261367) in pBluescript SK(+) (Stratagene, La Jolla, CA) and hamster Per2 (nucleotides 822-1601)
in pGEM-T Easy Vector (Promega, Madison, WI), which were generously provided by Drs. H.
Okimura (Kobe University, Kobe, Japan) and S. Shibata (Waseda University, Saitama, Japan),
respectively. Bacteria were transformed with these plasmids and vector DNA was purified for cRNA synthesis. [The characteristics of these clones and their utility for in situ hybridization
have been published previously (24)].
9
In situ hybridization was conducted using slide-mounted tissue sections that were pre-treated as
follows: equilibrated to room temperature, fixed for 15 min in 4% paraformaldehyde in 0.1 M
phosphate-buffered saline (pH 7.4, PBS), and washed 2 X 5 min each in PBS. The sections were
acetylated in triethanolamine (0.1 M, pH 8.0) and 0.25% acetic anhydride, dehydrated and
delipidated, and air dried. The sections were hybridized overnight at 55° C with a saturating
concentration (determined from preliminary studies) of 35S-labelled riboprobes that were diluted
in hybridization cocktail containing salmon sperm DNA (250 μg/ml), yeast total RNA (625
μg/ml), Tris-HCl (20 mM, pH 7.4), EDTA (1 mM), NaCl (300 mM), deionized formamide (50%
v/v), dextran sulfate (10% v/v), Denhardt’s solution (1x), dithiothreitol (100 mM), SDS
(0.1%w/v), and Na-thiosulfate (0.1% w/v). After 18 hours, sections were rinsed twice for 10 min
each in 2X SSC/10 mM DTT (1X SSC=0.25 M NaCl, 0.015 M sodium citrate, pH 7.2) at 22° C,
treated with RNAse (40 mg/ml) for 30 min at 37o C, washed for 15 min in 1X SSC and incubated
for 1 h at 63o C in 0.1X SSC. The sections were rinsed in 0.1X SSC at 22° C, quickly
dehydrated and air dried. All solutions, except Tris buffer, were prepared with diethyl
pyrocarbonate-treated water (0.1%) and autoclaved. Slides and radioactive standards (14Cmicroscales, Amersham Life Sciences) were exposed to X-ray film (Biomax MR, Kodak) for
times determined in preliminary experiments. Autoradiograms were analyzed by
microdensitometry, using an M4 image analysis system (Imaging Corp., St. Catherines, Ontario,
Canada), as previously described.
For Per1 and Per2 mRNAs, the effect of phenobarbital in each brain region was assessed by
Student’s t test.
10
RESULTS
Experiment 1: Effect of phenobarbital administration during the critical period on estrous cycles
and phase of the locomotor activity rhythm.
Vehicle administration did not alter estrous cyclicity in the 5 control hamsters (Fig. 1, top panel).
In contrast phenobarbital treatment at ZT 6 blocked the next expected estrous discharge in 4 of 6
hamsters (Fig. 1, middle and bottom panels; missing data for animal F14). Unexpectedly,
subsequent cycles in 5 of these 7 pentobarbital-treated hamsters were blocked for variable
periods of time (middle panel). However when estrous discharges reappeared, they were delayed
by one day more than a multiple of the expected 4 day length of a normal estrous cycle. In the
other 2 hamsters, phenobarbital did not delay the next estrous discharge (bottom panel).
Although the timing of ovulation was not determined in these studies, ovulation is dependent on
the LH surge and occurs on the morning after an LH surge, which usually but not always
coincides with an estrous vaginal discharge. Pentobarbital delay of LH surges also delays
ovulation until the day after the LH surge for up to 3 days (55). Therefore, it is highly likely that
ovulation was delayed until the day after the LH surge, whenever it next occurred. Actograms
from a representative animal in each group are depicted in Fig. 2. Delay of the estrous cycle in
response to phenobarbital treatment was associated with the magnitude of the resulting phase
advances in the circadian activity rhythms. In animals in which phenobarbital delayed the
estrous cycle, the average phase advance (2.9 ± 0.3 h) was approximately 2-fold greater than that
observed in either vehicle-treated hamsters (1.4 ± 0.3 h) or those in which phenobarbital had no
effect on estrous cycles (1.5 ± 0.1 hours) (Fig. 3). In general, phase advances ≥ 2.3 h were
associated with delays in the estrous cycle and phase advances < 2.3 h were not, suggesting that
the threshold phase advance for delaying estrous cycles is about 2.3 hours. Thus, one vehicle-
11
injected hamster exhibited a phase advance of 2.38 hours in the absence of any change in estrous
cyclicity, whereas one phenobarbital injected hamster in which estrous cycles were delayed had a
2.35 hour phase advance (Fig. 1).
Experiment 2. Effect of phenobarbital administration on estrous cycles, phase of the locomotor
activity rhythm, and timing of the LH surge.
Vehicle administration at ZT 5 had no effect on estrous cycles in 4 of 6 hamsters based on
recorded estrous discharges, the timing of subsequent estrous discharges relative to the expected
4-day intervals (Fig. 4, left panel), and on occurrence of the LH surge on proestrus (Day 1) (Fig.
6, left panel). In the remaining 2 hamsters (F75 and F87), there was a one day delay in estrous
cycles. In contrast, based on the same three criteria, phenobarbital administration at ZT 5
blocked the next expected estrous discharge in 6 of 6 hamsters (Fig. 4, right panel). In 5 of the 6
phenobarbital-treated hamsters, estrous cycles were blocked for one day more than a multiple of
the expected 4 day estrous cycle. In the sixth animal (F80) estrous discharges appear to have
been delayed 3 days more than a multiple of the 4 day estrous cycle. The phase shifts within
each group were not different whether animals were transferred to a small container for
collection of each blood sample or remained in their cages and were sampled remotely (P = 0.16,
vehicle and P = 0.31, phenobarbital), therefore data from all animals within a group were
combined. Just as in Experiment 1, the delay in estrous cycles after phenobarbital treatment was
associated with magnitude of the resulting phase shifts (Fig. 5). Generally, a phase advance of
about 2 hours or longer resulted in a one-day delay of the estrous cycle; a shorter phase advance
had no effect on estrous cyclicity. This relationship was observed in 5 of the 6 animals in both
groups. In addition, in all 8 animals in which estrous cycles were delayed by one day (2 vehicle-
12
treated and 6 phenobarbital-treated), the LH surge was also delayed by one day and occurred 1.6
hours earlier on the average than that in vehicle-treated animals, with no difference in magnitude
(Fig. 6).
Experiment 3: Effect of phenobarbital administration on Per1 and Per2 expression in select
brain areas.
Phenobarbital administration selectively suppressed SCN Per1 mRNA expression. Namely,
Per1 mRNA expression in the SCN was approximately 50% lower in phenobarbital- treated
hamsters than in vehicle-treated hamsters (P=0.01), while Per2 mRNA expression in the SCN
was not affected by phenobarbital (P=0.31) (Figs. 1, 2). In contrast to the SCN, phenobarbital
treatment attenuated Per2 mRNA but not Per1 mRNA in the cingulate cortex, but had no effect
on the expression of either Per1 or Per2 mRNA in the piriform cortex, CA3, or dentate gyrus
(Fig. 2). In the midline thalamus, Per1 mRNA expression was higher in phenobarbital-injected
hamsters versus vehicle-injected hamsters (P<0.005; Fig. 2), while Per2 mRNA was
undetectable in this region in both treatment groups.
DISCUSSION
These findings support the hypothesis that phenobarbital administration to female hamsters near
the beginning of the critical period causes a robust, rapid phase advance of the circadian
pacemaker, which entrains both the activity rhythm and the neural circadian signal for the LH
surge. Although several previous studies have investigated the effects of barbiturates on
circadian locomotor activity rhythms in male rodents, the effects were inconsistent and appear to
depend on species or strain. Thus, pentobarbital (50 mg/kg i.p.) induces circadian phase
advances during the mid-subjective day and small phase delays during the subjective night when
13
administered to male mice of the DBA/2 or SK strains, but not when given to C57 BL mice or
male rats or hamsters (14; 46). In contrast, the present results clearly show that in female
hamsters, phenobarbital (100 mg/kg s.c.) administered at ZT 5-6 induces a phase advance in both
circadian locomotor activity rhythms and the timing signal for the LH surge. Although vehicletreated hamsters also exhibited phase advances in their wheel-running rhythms, these phase
advances were about half as large as those seen in animals treated with phenobarbital, and the
small phase advances were not sufficient to delay the LH surge. These smaller phase advances
may have resulted at least in part from the transfer to constant darkness because exposure to a
dark “pulse” of at least 3 h duration, or to constant darkness, during the mid-subjective day
following exposure to constant light can induce a 2-3 h phase shift (2; 9). In addition, it is
possible that the volume of vehicle injected (10 ml), which was much larger than that typically
used in phase shifting studies, contributed to the phase shift magnitude. [This volume was
chosen to mimic the volume used previously to block the LH surge in hamsters (55).] The
repeated arousal for hourly blood sampling was most likely not a contributory factor because
there was no difference in the phase shifts between animals that were sampled and those that
were not (compare vehicle-treated animals in Figs. 3 and 5). Finally, it is unlikely that the 1%
ethanol in the vehicle contributed to the phase advance because it has been shown that ethanol
attenuates phase advances induced either by light or a nonphotic stimulus, triazolam (48).
It is interesting to note that in contrast to most other nonphotic stimuli that phase advance the
circadian pacemaker, such as dark pulses and triazolam (59), phenobarbital strongly suppresses
rather than stimulates activity and alertness. Physical activity appears to be necessary for the
phase shifting effect of some nonphotic stimuli (35; 38; 59), perhaps through increased arousal.
In support of this possibility, sleep deprivation is sufficient to induce a phase shift (35). The
14
present findings suggest that neither increased arousal state nor activity is required for a
phenobarbital-induced phase advance in female hamsters.
It has long been known that pentobarbital administration in rats can block the LH surge and
ovulation, thereby delaying the estrous cycle by one day. However, it is essential to recognize
that there are important functional differences between the two barbiturate anesthetics that block
LH surges and ovulation in rodents, which are revealed when comparing their actions between
rats and hamsters. Thus, although administration of either pentobarbital or phenobarbital can
block ovulation in rats (16), only phenobarbital blocks LH surges in hamsters (40). Both drugs
are completely effective only if administered at the beginning of the two-hour critical period. If
either drug is administered two hours later, neither of them blocks the LH surge. In both species,
the dose of phenobarbital that is sufficient to block the LH surge does not induce surgical
anesthesia, as does the surge-blocking dose of pentobarbital, indicating that these two drugs may
be acting by different mechanisms to block the LH surge (16; 40). In this regard, administration
of pentobarbital to proestrous hamsters at the beginning of the critical period delays the time
when the neural signal for the LH surge occurs, as indicated by a one-hour delay in the LH surge,
and a delay in the critical period, i.e. the time when phenobarbital administration can block the
LH surge (40). These findings suggest that in hamsters, pentobarbital may act by delaying the
critical period, whereas our findings suggest that phenobarbital acts via a phase advance in the
circadian clock.
It should also be noted that although phenobarbital administration advanced the timing of the LH
surge on the next day, it remains to be determined whether this phase shift is permanent (stable)
or only transient. The former possibility receives strong support from our finding that
15
phenobarbital administration induces a permanent phase shift in the circadian pacemaker
regulating the onset of activity and on previous findings that the onset of activity and timing of
the LH surge are strongly linked in hamsters (36; 37; 58). Addressing the issue of a permanent
phase shift in the timing of the LH surge would be more feasible in hamsters that are exhibiting
daily LH surges, such as anestrous hamsters exposed to short days (6; 7; 47) or ovariectomized
hamsters in the presence or absence of estradiol (39; 57), than in intact, long day-exposed
hamsters such as those used in the present study.
When the phenobarbital-induced phase advance is greater than about 2 hours, the next estrous
discharge is blocked. Based on previous findings that the duration of the critical period is about
2 hours, and on the finding herein that after phenobarbital, the LH surge occurs between 1 and 2
hours earlier one day later, these results suggest that phenobarbital delays the hamster estrous
cycle one day in part by advancing the phase of the circadian clock past the critical period for the
neural signal that initiates the LH surge. Although phenobarbital phase-advances the circadian
clock, it remains to be demonstrated that this is sufficient to block the LH surge, and whether
there are other non-circadian mechanisms whereby barbiturates block the LH surge. For
example in ewes, pentobarbital suppresses LH pulse frequency (20), which is a possible
mechanism for blockade of the LH surge by phenobarbital in rodents. In addition, in induced
ovulators like rabbits, in which the LH surge is not controlled by the circadian clock,
pentobarbital only reliably blocks the LH surge when administered such that it blocks the coitusinduced neural signal for LH release (34). Therefore in species which lack circadian control of
the preovulatory LH surge, barbiturate blockade of the LH surge must occur via a different
mechanism than in spontaneous ovulators (e.g., rodents). We are currently using a non-photic
16
zeitgeber to address whether a phase advance alone is sufficient to block the LH surge and
advance the circadian phase of its peak in hamsters.
The difference in effectiveness of phenobarbital to delay estrous cycles between Experiments 1
and 2 may be related to the time of phenobarbital administration in relation to the critical period.
In previous studies, phenobarbital administration to proestrous hamsters at ZT 5 blocked LH
surges in 100% of animals (55), therefore in Experiment 2 the time of injection was advanced
one hour to ZT 5 and succeeded in delaying estrous cycles in all animals. These results suggest
that the critical period in hamsters begins between ZT 5 and ZT 6, about 2 hours before onset of
the LH surge.
Our observation that phenobarbital administration blocks the vaginal estrous discharge, often for
a number of days, suggests that this drug may exert reproductive effects on other tissues (e.g.,
vaginal epithelium) besides the SCN circadian pacemaker. The estrous discharge reflects mucus
secretion by the vaginal epithelial cells (26), and although it is often used to monitor estrous
cyclicity, it is not always an accurate indicator of LH release or ovulation. It remains to be
determined whether phenobarbital chronically decreases vaginal mucus production, thereby
disrupting the estrous vaginal discharge selectively, while regular 4-day estrous cycles, including
LH surges and ovulation, continue to occur delayed by one day. Alternatively, it is possible that
the cessation of vaginal estrous discharges is caused by the disruption of normal circulating
levels of estrogen and/or progesterone occurring as a result of the phenobarbital administration.
In this case, estrous cycles would also be disrupted during the absence of vaginal estrous
discharges.
17
Phase resetting of the master circadian pacemaker is associated with changes in expression of the
core circadian clock genes, Per1 and Per2, as demonstrated by many studies in male rodents (13;
24; 30; 33). Thus in hamsters, induction of phase shifts by light at night is associated with
increased SCN expression of Per1 and Per2 (24). In contrast, induction of phase shifts during
the daytime by a variety of nonphotic stimuli, such as systemic injections of either triazolam or
8-OH-DPAT, is associated with decreased SCN expression of Per1 and Per2 (13; 24; 30; 33).
Similarly, our results show that phenobarbital administration to proestrous hamsters in the early
afternoon (ZT 5) decreased SCN expression of Per1 mRNA two hours later. Based on the
observation that acute reduction of SCN Per1 mRNA by local administration of antisense
oligonucleotides induces phase advances in the circadian locomotor activity rhythm in male
hamsters (22), our finding that phenobarbital decreases SCN Per1 mRNA elucidates the
mechanism by which phenobarbital advances the circadian rhythm of wheel running in
proestrous hamsters. This finding also supports the hypothesis that phenobarbital blockade of
the LH surge is mediated at least in part by an advance in the timing of the master circadian
pacemaker.
In contrast to its effect on SCN Per1 mRNA expression, phenobarbital administration to female
hamsters in the present study did not alter SCN Per2 mRNA expression at two hours postinjection. The differential effect of phenobarbital on these two Per genes at this timepoint
distinguishes it from most other nonphotic stimuli, which suppress expression of both Per1 and
Per2 when presented in the middle of the subjective day (13; 24; 30; 33). However, the present
finding parallels that of another study in which the GABAA receptor agonist, muscimol, was
microinjected at CT 6 into the SCN region of hamsters exposed to either 14 days or 42 hours of
constant darkness (15). This treatment suppressed SCN Per1 mRNA at either two or three hours
18
post-injection, but suppression of Per2 mRNA was only observed 3 hours, not 2 hours, postinjection in hamsters that were exposed to 42 h of darkness (15).
The differential suppressive effect of muscimol on Per1 versus Per2 mRNA at two hours postinjection might be related to the different temporal patterns of SCN expression of these genes.
The circadian rhythm of Per2 mRNA expression in the male hamster SCN, after exposure to
darkness for three days, peaks at CT 8, four hours later than that of SCN Per1 mRNA expression
(24). In male hamsters exposed to a light:dark cycle (14L:10D), SCN Per2 mRNA expression
exhibited a broad “peak” from ZT4-ZT15, in contrast to the sharp peak in SCN Per1 mRNA
expression at ZT 4 (30). Thus, it may be easier to induce suppression of Per1 mRNA than Per2
mRNA in the mid-afternoon (either CT or ZT 4-8) because expression of the former but not the
latter is normally decreasing during this interval. Also, muscimol might only induce a decrease
in SCN Per2 mRNA at 3 h but not 2 h post-injection (15) because the former timepoint but not
the latter occurs during the falling phase of Per2 mRNA expression during exposure to darkness
(24).
Muscimol shares a common neurochemical mechanism of action with phenobarbital; both drugs
have a high affinity for GABAA receptor complexes that form ligand-gated chloride channels
(12). Activation of GABAA receptors stimulates opening of chloride channels, allowing an
influx of chloride ions and thereby inducing hyperpolarization of the plasma membrane. Binding
of phenobarbital to the barbiturate binding sites on the GABAA receptor complex increases the
opening time of the channel as well as the affinity of the receptor for GABA binding (12; 50).
GABAA receptors are expressed in many brain regions, including the SCN (41; 49).
Furthermore, in male hamsters, activation of SCN GABAA receptors by microinjections of
19
muscimol during the midsubjective day not only reduces SCN Per1 and Per2 mRNA expression,
as discussed above, but also induces circadian phase advances (15; 52; 53). These findings are
very similar to the effects we observed following phenobarbital administration in female
hamsters and suggest that activation of SCN GABAA receptors may constitute a common
mechanism for induction of phase advances.
In the present study, phenobarbital’s effects on Per1 and Per2 expression differed in a brainregion-dependent manner. For example, although phenobarbital induced a large decrease in
Per1 mRNA in the SCN, no phenobarbital-induced decreases in Per1 mRNA were observed in
any other region examined, and surprisingly an increase was observed in the midline thalamus.
Further, although Per2 mRNA expression was unaffected by phenobarbital in most regions
investigated, it was suppressed in the cingulate cortex. These regional differences in the effects
of phenobarbital may be due to differences among brain regions in the temporal patterns of Per1
and Per2 expressions and/or in the time required for the drug to exert its effects. The former
possibility is support by a number of previous findings demonstrating that the temporal
expression and mechanisms of regulation of Per in the cortex differ from those in the SCN.
Thus, Per1 and Per2 gene expressions in the SCN both peak in the daytime (24; 30). In contrast
Per gene expression in the cerebral cortex peaks at night in both female rats (29) and male mice
(1; 60) exposed to a light:dark cycle. In the parietal cortex of female rats, Per1 and Per2
mRNAs exhibit 24-h rhythms but neither gene shows a 24-h expression rhythm in the cingulate
cortex (60). Cortical Per expression is also affected differently from the SCN by various stimuli
that alter the time of activity and arousal. For example, chronic treatment of rats with
methamphetamine reverses the day:night phases of the locomotor rhythm concomitant with a
large shift in the peak of Per1 mRNA in the cortex but not the SCN. In mice, restricted daytime
20
feeding selectively shifts the Per1 mRNA peak in the cortex but not the SCN (60). Also, sleep
deprivation of mice increases cortical but not SCN expression of Per1 and Per2 mRNAs (18;
64). In view of these observations, one caveat of our results is that only one injection time and
one time of brain collection post-injection were investigated.
In conclusion, the results herein confirm that administration of phenobarbital during the critical
period on proestrus delays the LH surge and the estrous cycle by one day, as shown previously.
In addition, they demonstrate for the first time that phenobarbital administration during the
critical period phase advances the wheel running activity rhythm as well as the time of the peak
of the LH surge that occurs on the following day. Because phenobarbital advances both of these
circadian rhythms, it appears to reset the circadian pacemaker in the SCN, thereby supporting the
hypothesis that phenobarbital delays the LH surge by advancing the circadian pacemaker past the
critical period. The foregoing results also show that phenobarbital administration to female
hamsters at ZT 5 on proestrus selectively suppresses Per1 mRNA but not Per2 mRNA in the
SCN at 2 h post-injection. This effect of phenobarbital resembles the effect of other nonphotic
phase advancing signals, including injection of muscimol, another drug that activates GABAA
receptors. In contrast to its effects in the SCN, very few effects of phenobarbital on either Per1
mRNA or Per2 mRNA expression were observed in other brain regions investigated, which
included the thalamus, cortex, and hippocampus.
PERSPECTIVES AND SIGNIFICANCE
The findings that phenobarbital administration on the afternoon of proestrus advances both the
circadian activity rhythm and the timing of the delayed LH surge, as well as suppresses SCN
21
expression of Per1, demonstrates that this drug acts as a nonphotic zeitgeber that resets the
master circadian pacemaker. These findings also demonstrate that female rodents exhibit a
similar molecular mechanism of nonphotic phase resetting previously established in males.
Although the depressant effects of barbiturates are known to occur widely throughout the central
nervous system, the effect of phenobarbital on Per1 expression is selective for the SCN, at least
at the timepoint (2 h post-injection at ZT 6) investigated, suggesting that the molecular circadian
oscillations expressed in various brain regions are differentially responsive to this phase
advancing, nonphotic zeitgeber. Elucidation of the role of the circadian phase advance in
mediating phenobarbital blockade of the LH surge will be important for understanding the
mechanisms involved in this phenomenon and perhaps also in the regulation of other circadian
endocrine rhythms.
The present studies add to a growing body of evidence revealing interactions between the
circadian timing system and pharmaceutical agents. Although circadian influences on the
efficacy, duration, and toxicity of barbiturates and other drugs are well known (10), the current
findings demonstrate that barbiturates can robustly reset the circadian pacemaker and thus
suggest that time of drug therapy, including the use of phenobarbital for epilepsy, may influence
many circadian rhythms and the processes that they modulate. Also, the demonstrated
association between alterations in circadian rhythms and temporary disruption of a reproductive
hormone rhythm may have clinical relevance and also provide some insight into the alterations in
endocrine rhythms characteristic of depression in pregnant, post-partum, and post-menopausal
women (43-45).
22
ACKNOWLEDGEMENTS
We thank Dr. Terry Nett for generously providing us with anti-ovine LH, the National Hormone
& Peptide Program (NHHP), NIDDK and Dr. Parlow for rLH-RP-3, the reference preparation,
and Ms. Xiao-Li Peng, Ms. Verda Davis and Mr. Christopher Reinhardt for their expert technical
assistance with conduct of the experiments.
23
GRANTS
This project was supported in part by NIH R01 NS055228 (to SJL and MJD), by a University of
Kentucky Medical Center Research Grant (to SJL) and by NIH R01 AG13418 (to MJD).
24
REFERENCES
1. Abe H, Honma S, Namihira M, Masubuchi S, Ikeda M, Ebihara S and Honma K.
Clock gene expressions in the suprachiasmatic nucleus and other areas of the brain
during rhythm splitting in CS mice. Brain Res Mol Brain Res 87: 92-99, 2001.
2. Albers HE. Response of hamster circadian system to transitions between light and
darkness. Am J Physiol 250: R708-R711, 1986.
3. Amir S, Lamont EW, Robinson B and Stewart J. A circadian rhythm in the expression
of PERIOD2 protein reveals a novel SCN-controlled oscillator in the oval nucleus of the
bed nucleus of the stria terminalis. J Neurosci 24: 781-790, 2004.
4. Bast JD and Greenwald GS. Serum profiles of follicle-stimulating hormone, luteinizing
hormone and prolactin during the estrous cycle of the hamster. Endocrinology 94: 12951299, 1974.
5. Bethea CL and Weiner RI. Inhibition of prolactin secretion delays extinction of
circadian LH surges in ovariectomized rats bearing estradiol implants. Proc Soc Exp
Biol Med 172: 65-69, 1983.
6. Bex FJ and Goldman BD. Serum gonadotropins and follicular development in the
Syrian hamster. Endocrinology 96: 928-933, 1975.
7. Bridges RS and Goldman BD. Diurnal rhythms in gonadotropins and progesterone in
lactating and photoperiod induced acyclic hamsters. Biol Reprod 13: 617-622, 1975.
8. Butcher RL, Collins WE and Fugo NW. Plasma concentration of LH, FSH, prolactin,
progesterone and estradiol-17β throughout the 4-day estrous cycle of the rat.
Endocrinology 94: 1704-1708, 1974.
9. Canal MM and Piggins HD. Resetting of the hamster circadian system by dark pulses.
Am J Physiol Regul Integr Comp Physiol 290: R785-R792, 2006.
10. Chassard D and Bruguerolle B. Chronobiology and anesthesia. Anesthesiology 100:
413-427, 2004.
11. Chazal G, Faudon M, Gogan F, Hery M, Kordon C and LaPlante E. Circadian
rhythm of luteinizing hormone secretion in the ovariectomized rat implanted with
oestradiol. J Endocrinol 75: 251-260, 1977.
25
12. Czapinski P, Blaszczyk B and Czuczwar SJ. Mechanisms of action of antiepileptic
drugs. Curr Top Med Chem 5: 3-14, 2005.
13. Duncan MJ, Franklin KM, Davis VA, Grossman GH, Knoch ME and Glass JD.
Short-term constant light potentiation of large-magnitude circadian phase shifts induced
by 8-OH-DPAT: effects on serotonin receptors and gene expression in the hamster
suprachiasmatic nucleus. Eur J Neurosci 5: 2306-2314, 2005.
14. Ebihara S and Hayakawa H. Pentobarbital-induced phase shifts of circadian rhythms of
locomotor activity are not mediated through stimulated activity in mice. Experientia 46:
1023-1026, 1990.
15. Ehlen JC, Novak CM, Karom MC, Gamble KL, Paul KN and Albers HE. GABAA
receptor activation suppresses Period 1 mRNA and Period 2 mRNA in the
suprachiasmatic nucleus during the mid-subjective day. Eur J Neurosci 23: 3328-3336,
2006.
16. Everett JW and Sawyer CH. A 24-hour periodicity in the "LH-release apparatus" of
female rats, disclosed by barbiturate sedation. Endocrinology 47: 198-218, 1950.
17. Everett JW and Tejasen T. Time factor in ovulation blockade in rats under differing
lighting conditions. Endocrinology 80: 790-792, 1967.
18. Franken P, Thomason R, Heller HC and O'Hara BF. A non-circadian role for clockgenes in sleep homeostasis: a strain comparison. BMC Neurosci 8: 87, 2007.
19. Gay VL, Midgley J and Niswender GD. Patterns of gonadotrophin secretion associated
with ovulation. Fed Proc 29: 1880-1887, 1970.
20. Goodman RL and Meyer SL. Effects of pentobarbital anesthesia on tonic luteinizing
hormone secretion in the ewe: evidence for active inhibition of luteinizing hormone in
anestrus. Biol Reprod 30: 374-381, 1984.
21. Granados-Fuentes D, Saxena MT, Prolo LM, Aton SJ and Herzog ED. Olfactory
bulb neurons express functional, entrainable circadian rhythms. Eur J Neurosci 19: 898906, 2004.
22. Hamada T, Antle MC and Silver R. The role of Period1 in non-photic resetting of the
hamster circadian pacemaker in the suprachiasmatic nucleus. Neurosci Lett 362: 87-90,
2004.
26
23. Henderson SR, Baker C and Fink G. Effect of oestradiol-17β exposure on the
spontaneous secretion of gonadotrophins in chronically gonadectomized rats. J
Endocrinol 73: 455-462, 1977.
24. Horikawa K, Yokota Si, Fuji K, Akiyama M, Moriya T, Okamura H and Shibata S.
Nonphotic entrainment by 5-HT1A/7 receptor agonists accompanied by reduced Per1
and Per2 mRNA levels in the suprachiasmatic nuclei. J Neurosci 20: 5867-5873, 2000.
25. Janik LY and Janik D. Nonphotic phase shifting in female Syrian hamsters: Interactions
with the estrous cycle. J Biol Rhythms 18: 307-317, 2003.
26. Joseph T, Daly M and Kramer B. Alterations in rat vaginal histology by exogenous
gonadotrophins. J Anat 193: 469-472, 1998.
27. Legan SJ and Callahan WH. Suppression of tonic luteinizing hormone secretion and
norepinephrine release near the GnRH neurons by estradiol in ovariectomized rats.
Neuroendocrinology 70: 237-245, 1999.
28. Legan SJ and Karsch FJ. A daily signal for the LH surge in the rat. Endocrinology 96:
57-62, 1975.
29. Masubuchi S, Honma S, Abe H, Ishizaki K, Namihira M, Ikeda M and Honma K.
Clock genes outside the suprachiasmatic nucleus involved in manifestation of locomotor
activity rhythm in rats. Eur J Neurosci 12: 4206-4214, 2000.
30. Maywood ES, Mrosovsky N, Field MD and Hastings MH. Rapid down-regulation of
mammalian period genes during behavioral resetting of the circadian clock. Proc Natl
Acad Sci USA 96: 15211-15216, 1999.
31. McCormack CE. Acute effects of altered photoperiods on the onset of ovulation in
gonadotropin-treated immature rats. Endocrinology 93: 403-410, 1973.
32. McCormack CE and Sridaran R. Timing of ovulation in rats during exposure to
continuous light: Evidence for a circadian rhythm of luteinizing hormone secretion. J
Endocrinol 76: 135-144, 1978.
33. Mendoza JY, Dardente H, Escobar C, Pevet P and Challet E. Dark pulse resetting of
the suprachiasmatic clock in Syrian hamsters: behavioral phase-shifts and clock gene
expression. Neuroscience 127: 529-537, 2004.
27
34. Mills T, Copland A and Osteen K. Factors affecting the postovulatory surge of FSH in
the rabbit. Biol Reprod 25: 530-535, 1981.
35. Mistlberger R, Belcourt J and Antle M. Circadian clock resetting by sleep deprivation
without exercise: dark pulses revisited. J Biol Rhythms 17: 227-237, 2002.
36. Moline ML and Albers HE. Response of circadian locomotor activity and the
proestrous luteinizing hormone surge to phase shifts of the light-dark cycle in the
hamster. Physiol Behav 43: 435-440, 1988.
37. Moline ML, Albers HE, Todd RB and Moore-Ede MC. Light-dark entrainment of
proestrous LH surges and circadian locomotor activity in female hamsters. Horm Behav
15: 451-458, 1981.
38. Mrosovsky N. Locomotor activity and non-photic influences on circadian clocks. Biol
Rev Camb Philos Soc 71: 343-372, 1996.
39. Norman RL, Blake CA and Sawyer CH. Estrogen-dependent twenty-four-hour
periodicity in pituitary LH release in the female hamster. Endocrinology 93: 965-970,
1973.
40. Norman RL, Blake CA and Sawyer CH. Delay of the proestrous ovulatory surge of LH
in the hamster by pentobarbital or ether. Endocrinology 91: 1025-1029, 1972.
41. O'Hara BF, Andretic R, Heller HC, Carter DB and Kildruff TS. GABAA, GABAC,
and NMDA receptor subunit expression in the suprachiasmatic nucleus and other brain
regions. Brain Res Mol Brain Res 28: 239-250, 1995.
42. Orsini MW. The external vaginal phenomena characterizing the stages of the estrous
cycle, pregnancy, pseudopregnancy, lactation, and the anestrous hamster, Mesocricetus
auratus waterhouse. Waterhouse Proc Anim Care Panel 11: 193-206, 1961.
43. Parry BL, Meliska CJ, Sorenson DL, Lopez AM, Martinez LF, Nowakowski S,
Elliott JA, Hauger RL and Kripke DF. Plasma melatonin circadian rhythm
disturbances during pregnancy and postpartum in depressed women and women with
personal or family histories of depression. Am J Psychiatry 165: 1551-1558, 2008.
44. Parry BL, Meliska CJ, Sorenson DL, Lopez AM, Martinez LF, Nowakowski S,
Hauger RL and Elliott JA. Increased melatonin and delayed offset in menopausal
depression: Role of years past menopause, follicle-stimulating hormone, sleep end time,
and body mass index. J Clin Endocrinol Metab 93: 54-60, 2008.
28
45. Parry BL and Newton RP. Chronobiological basis of female-specific mood disorders.
Neuropsychopharmacology 25: S102-S108, 2001.
46. Schwartz WJ. In vivo metabolic activity of hamster suprachiasmatic nuclei: use of
anesthesia. Am J Physiol Regul Integr Comp Physiol 252: R419-R422, 1987.
47. Seegal RF and Goldman BD. Effects of photoperiod on cyclicity and serum
gonadotropins in the Syrian hamster. Biol Reprod 12: 223-231, 1975.
48. Seggio JA, Logan RW and Rosenwasser AM. Chronic ethanol intake modulates photic
and non-photic circadian phase responses in the Syrian hamster. Pharmacol Biochem
Behav 87: 297-305, 2007.
49. Shimura M, Harata N, Tamai M and Akaike N. Allosteric modulation of GABAA
receptors in acutely dissociated neurons of the suprachiasmatic nucleus. Am J Physiol
270: C1726-C1734, 1996.
50. Sieghart W. GABAA receptors: ligand-gated Cl- ion channels modulated by multiple
drug-binding sites. Trends Pharmacol Sci 13: 446-450, 1992.
51. Smith MS, Freeman ME and Neill JD. The control of progesterone secretion during the
estrous cycle and early pseudopregnancy in the rat: Prolactin, gonadotropin and steroid
levels associated with rescue of the corpus luteum of pseudopregnancy. Endocrinology
96: 219-226, 1975.
52. Smith RD, Inouye S-I and Turek FW. Central administration of muscimol phase-shifts
the mammalian circadian clock. J Comp Physiol A Neuroethol Sens Neural Behav
Physiol 164: 805-814, 1989.
53. Smith RD, Turek FW and Slater NT. Bicuculline and picrotoxin block phase advances
induced by GABA agonists in the circadian rhythm of locomotor activity in the golden
hamster by a phaclofen-insensitive mechanism. Brain Res 530: 275-282, 1990.
54. Stetson MH and Anderson PJ. Circadian pacemaker times gonadotropin release in freerunning female hamsters. Am J Physiol 238: R23-R27, 1980.
55. Stetson MH and Watson-Whitmyre M. The neural clock regulating estrous cyclicity in
hamsters: Gonadotropin release following barbiturate blockade. Biol Reprod 16: 536542, 1977.
29
56. Stetson MH and Watson-Whitmyre M. Nucleus suprachiasmaticus: the biological
clock in the hamster? Science 191: 197-199, 1976.
57. Stetson MH, Watson-Whitmyre M and Matt KS. Cyclic gonadotropin release in the
presence and absence of estrogenic feedback in ovariectomized golden hamsters. Biol
Reprod 19: 40-50, 1978.
58. Swann JM and Turek FW. Multiple circadian oscillators regulate the timing of
behavioral and endocrine rhythms in female golden hamsters. Science 228: 898-900,
1985.
59. Van Reeth O and Turek FW. Stimulated activity mediates phase shifts in the hamster
circadian clock induced by dark pulses or benzodiazepines. Nature 339: 49-51, 1989.
60. Wakamatsu H, Yoshinobu Y, Aida R, Moriya T, Akiyama M and Shibata S.
Restricted-feeding-induced anticipatory activity rhythm is associated with a phase-shift
of the expression of mPer1 and mPer2 mRNA in the cerebral cortex and hippocampus
but not in the suprachiasmatic nucleus of mice. Eur J Neurosci 13: 1190-1196, 2001.
61. Wiegand SJ, Terasawa E, Bridson WE and Goy RW. Effects of discrete lesions of
preoptic and suprachiasmatic structures in the female rat. Alterations in the feedback
regulation of gonadotropin secretion. Neuroendocrinology 31: 147-157, 1980.
62. Wise PM. Alterations in proestrous LH, FSH, and prolactin surges in middle-aged rats.
Proc Soc Exp Biol Med 169: 348-354, 1982.
63. Wise PM. Estradiol-induced daily luteinizing hormone and prolactin surges in young and
middle-aged rats: correlations with age-related changes in pituitary responsiveness and
catecholamine turnover rates in microdissected brain areas. Endocrinology 115: 801809, 1984.
64. Wisor JP, O'Hara BF, Terao A, Selby CP, Kilduff TS, Sancar A, Edgar DM and
Franken P. A role for cryptochromes in sleep regulation. BMC Neurosci 3: 20, 2002.
65. Yoo SH, Yamazaki S, Lowrey PL, Shimomura K, Ko CH, Buhr ED, Siepka SM,
Hong HK, Oh WJ, Yoo OJ, Menaker M and Takahashi JS.
PERIOD2::LUCIFERASE real-time reporting of circadian dynamics reveals persistent
circadian oscillations in mouse peripheral tissues. Proc Natl Acad Sci USA 101: 53395346, 2004.
30
FIGURE LEGENDS
Fig. 1. Phenobarbital at ZT 6 blocks estrous cycles in most hamsters. Daily incidence of estrous
(E) or non-estrous (non-E) vaginal discharge is depicted during the 2 weeks before and after
injection (arrows) of vehicle (V, top panel) or phenobarbital (Ph, middle and bottom panels) at
ZT 6 on proestrous afternoon. Phenobarbital treatment blocked the next estrous discharge in
most animals (middle panel), but not in 2 of the hamsters (bottom panel). The expected days of
estrous discharge every 4th day after treatment are indicated by the vertical lines, either solid
(first estrus after treatment) or broken (subsequent cycles). Phase shifts (hours) in the circadian
activity rhythm of each animal are indicated in the upper right of each panel. The shaded area
represents the time during which animals were housed in DD.
Figure 2. Phenobarbital at ZT 6 induces large phase advances in most hamsters. Representative
actograms depicting wheel-running activity in 3 hamsters that were treated at ZT 6 on proestrus
with either vehicle (Veh, left panel), or phenobarbital (middle and right panels). Within each
panel, the rows depict wheel running activity as vertical black marks throughout consecutive 24
hour periods, with data from consecutive days arranged from top to bottom. Time of day in
hours is indicated on the horizontal bar above the actograms. The cross-hatched vertical bars to
the right of each actogram indicate exposure to 14L:10D and the solid vertical bars depict
exposure to DD. The asterisks indicate the time of injection. The phase (φ) shifts in hours are
indicated. Phenobarbital blocked estrous discharges in one of the hamsters (Phen block, middle
panel), but not in the other hamster (Phen No block, right panel).
Figure 3. Phase advances are about 2-fold greater in those hamsters in which phenobarbital at
ZT 6 delays estrous cycles. Mean (+ SE) phase shifts in vehicle- (Veh, n=5) or phenobarbital-
31
treated (Ph) hamsters are illustrated. Phenobarbital treatment blocked the next expected estrous
discharge in 5 of 7 hamsters (blk), in association with large phase shifts, but not in the remaining
2 (No blk), in which it elicited smaller phase shifts.
Figure 4. Phenobarbital at ZT 5 delays estrous discharges by one day more than a multiple of 4.
Daily incidence of estrous (E) or non-estrous (non-E) vaginal discharge in individual animals is
depicted during the 2 weeks before and 2-3 weeks after injection (arrows) of vehicle (Veh, left
panel) or phenobarbital (Ph, right panel) at ZT 5 on proestrous afternoon. The expected day of
estrous discharge immediately after treatment is indicated by the solid vertical line, and every 4th
day thereafter by the broken vertical lines. Phase shifts (hours) in the circadian activity rhythm
and time (hours) of the LH peak on the 2 days after injection in each animal are indicated in the 3
columns at the right of each panel. The LH peak times in parentheses on Day 2 in the vehicletreated animals are based on samples that were only obtained every 2 hours, whereas on Day 1,
samples were obtained hourly. The “+” sign indicates that an LH surge occurred, but the time of
the peak could not be determined due to missed samples. The shaded area represents the time
during which animals were housed in DD.
Figure 5. Phenobarbital at ZT 5 induces large (>2 hour) phase shifts. Left panel: Representative
actograms depicting wheel-running activity in 2 hamsters, one treated with vehicle (Veh), and
the other receiving phenobarbital (Phen) at ZT 5 on proestrus. The asterisks indicate the time of
injection. The phase (Φ) shifts in hours are indicated. For further details, see legend to Fig. 2.
Right panel: Mean (+ SE) phase shifts in groups treated with either vehicle (Veh, n=4) or
phenobarbital (Phen, n=6).
32
Figure 6. Phenobarbital at ZT 5 delays the day and phase advances the time of the LH surge.
Line drawing: Mean (+ SE) plasma LH concentrations from 1200 to 1800 h on the day of (Day
1, left panels) and the day after (Day 2, right panels) injection (Inj, arrows) of either vehicle
(Veh, upper panels) or phenobarbital (Phen, lower panels) on proestrus at ZT 5. Bar graphs:
Mean (+ SE) peak level of plasma LH concentrations (left panel) and time of the LH peak (right
panel) in hamsters treated with either vehicle (VEH) or phenobarbital (PHEN) at ZT 5 on
proestrous afternoon. Although all 6 phenobarbital-treated animals had an LH surge on Day 2,
results are only depicted for 5 of them due to missing samples in the sixth hamster (F78).
Figure 7. Pseudo-colored autoradiogams showing expression of Per1 mRNA and Per2 mRNA.
Darker, brighter colors (black and red) indicate higher levels of expression than lighter colors
(yellow, green, blue).
Figure 8. Effect of phenobarbital administration on Per1 mRNA and Per2 mRNA expressions.
Values shown are the mean and standard error. Thal, midline thalamus; Cg, cingulate cortex;
Pir, pyriform cortex; CA3, CA3 hippocampal subfield; DG, dentate gyrus, ND, not detectable.
*, P<0.05 versus vehicle in the same region; #, P<0.005 versus vehicle in the same region.
8
16
0
*
shift
0.92
Veh
8 8
16
0
*
shift
2.35
Phen
block
8 8
16
0
*
shift
1.56
Phen
No block
8
Phase Shift (h)
P<0.01
4
(5)
3
2
P<0.05
(5)
(2)
1
0
Veh
Ph
blk
Ph
No blk
Phase Shift (h)
P<0.05
4
3
2
(6)
(4)
1
0
Veh
Phen
Saline n=5 Day
1
Day 2
Veh
Inj
Sal Inj
20
Vehicle
n=4
10
0
40
Phenobarbital
n=5
30
Phen Inj
20
10
10
12
14
16
18
10
Hours x 100
12
14
16
18
Time of Peak
P<0.05
50
30
0
Peak [LH]
Plasma LH, ng/ml
Plasma LH ng/ml
40
Day of Expected
Estrous Discharge
(4)
40
17
(5)
30
18
16
20
15
10
14
0
Veh
Phen
Veh
Phen
13
Hours x 100
Day of Expected
Proestrus
VEH
CA3
DG
Per1
Thal
Pir
VEH
Per2
PHENO
Cg
SCN
PHENO