Chronobiology International, 2014; 31(5): 637–644 ! Informa Healthcare USA, Inc. ISSN: 0742-0528 print / 1525-6073 online DOI: 10.3109/07420528.2014.885528
ORIGINAL ARTICLE
Estrogen receptor 1 modulates circadian rhythms in adult female mice Margaret S. Blattner1,2 and Megan M. Mahoney2,3 1
College of Medicine, 2Neuroscience Program and 3Department of Comparative Biosciences, University of Illinois at Urbana Champaign, Urbana, IL, USA
Estradiol influences the level and distribution of daily activity, the duration of the free-running period, and the behavioral phase response to light pulses. However, the mechanisms by which estradiol regulates daily and circadian rhythms are not fully understood. We tested the hypothesis that estrogens modulate daily activity patterns via both classical and ‘‘non-classical’’ actions at the estrogen receptor subtype 1 (ESR1). We used female transgenic mice with mutations in their estrogen response pathways; ESR1 knock-out (ERKO) mice and ‘‘non-classical’’ estrogen receptor knock-in (NERKI) mice. NERKI mice have an ESR1 receptor with a mutation in the estrogen-response-element binding domain, allowing only actions via ‘‘non-classical’’ genomic and second messenger pathways. Ovariectomized female NERKI, ERKO, and wildtype (WT) mice were given a subcutaneous capsule with low- or high-dose estradiol and compared with counterparts with no hormone replacement. We measured wheel-running activity in a light:dark cycle and constant darkness, and the behavioral phase response to light pulses given at different points during the subjective day and night. Estradiol increased average daily wheel-running, consolidated activity to the dark phase, and shortened the endogenous period in WT, but not NERKI and ERKO mice. The timing of activity onset during entrainment was advanced in all estradiol-treated animals regardless of genotype suggesting an ESR1-independent mechanism. We propose that estradiol modifies period, activity level, and distribution of activity via classical actions of ESR1 whereas an ESR1 independent mechanism regulates the phase of rhythms. Abbreviations: ERKO, estrogen receptor knock-out; NERKI, nontraditional estrogen receptor knock-in; ESR1, estrogen receptor 1 (ERa); ESR2, estrogen receptor 2 (ERb)
Keywords: Estradiol, ERKO, ERa, locomotor activity, tau, phase response, phase angle
INTRODUCTION Developmental changes, including exposure to steroid hormones, also modulate the expression of circadian activity in adulthood as well as alter the responsiveness of the timekeeping system to circulating hormones. In adult hamsters, treatment with estradiol shortens the free running period in females but not males (Zucker et al., 1980). Adult female rats perinatally exposed to androgens have changes in tau in response to estradiol treatment which mimic those of males (Albers, 1981). Adult aromatase knock-out mice that cannot produce endogenous estradiol display decreased daily activity level, increased daytime activity, and alterations in their timing (phase angle) of activity onset relative to wildtype counterparts, suggesting that the absence of estradiol during development modifies the expression of daily activity rhythms throughout life (Brockman et al., 2011). Traditionally, estrogens mediate their actions through two distinct receptor subtypes: estrogen
20 14
In females, estrogens influence the amount and distribution of daily activity; however, the mechanisms underlying these effects remain largely unknown. The stage of the reproductive cycle influences activity patterns and daily rhythms. Specifically, gonadally intact female rats, hamsters, and degus have the greatest level of activity during proestrus and estrus, when circulating estrogen levels are highest (Mahoney et al., 2011; Morin et al., 1977b; Wollnik & Turek, 1988). Further, in these species the period of the endogenous free-running rhythm (tau) in constant conditions is shortest during estrus (Albers et al., 1981; Fitzgerald & Zucker, 1976; Labyak & Lee, 1995). Ovariectomy of female rodents causes a decrease of daily wheel-running activity and a lengthening of tau, and estradiol replacement increases activity levels and shortens tau (Albers, 1981; Gentry & Wade, 1976; Labyak & Lee, 1995; Morin et al., 1977b).
Submitted September 5, 2013, Returned for revision January 15, 2014, Accepted January 17, 2014
Correspondence: Megan M. Mahoney, Department of Comparative Biosciences, 3639 VMSB, MC-002, 2001 South Lincoln Avenue, Urbana, IL 61802, USA. Tel: +1 217 333 7578. Fax: +1 217 244 1652. E-mail: [email protected]
637
638
M. S. Blattner & M. M. Mahoney
receptor subtype 1 (estrogen receptor subtype 1 (ESR1), or estrogen receptor a) and estrogen receptor subtype 2 (ESR2, or estrogen receptor b). In ‘‘classical’’ nuclear receptor-mediated transcriptional activation, estrogens bind to receptors and the ligand–receptor complex then forms dimers. These dimers bind to the estrogen response element (ERE) on regulatory regions of target genes to alter gene expression. There are also more recently characterized ‘‘non-classical’’ mechanisms of estrogen signaling including ligand-independent estrogen receptor activation that occurs via second messenger pathways and estrogen receptor activation of gene transcription that occurs independently from ERE-dependent pathways (Hall et al., 2001; Kousteni et al., 2001; McDevitt et al., 2008). Previous data from this and other labs provide strong evidence that estrogens acting at ESR1 increase wheel-running activity. Further, classical and non-classical actions of ESR1 regulate the distribution of activity across the LD cycle and the response to light (Blattner & Mahoney, 2013; Ogawa et al., 2003). However, it is unknown if these modulatory effects of estrogens occur developmentally, in adulthood, or as some combination of these factors. Here we test the hypothesis that estradiol signaling at ESR1 mediates its effects on the expression of circadian rhythms and daily activity through both classical EREmediated and ‘‘non-classical’’ pathways. We used two strains of mice with transgenic modifications of their ability to respond to estrogens at ESR1. ESR1 knock-out (ERKO) mice are unable to respond to estradiol via this receptor. ‘‘Non-classical’’ estrogen receptor knock-in (NERKI) mice retain the ability to bind estrogens at ESR1; however, a mutation in the ERE-binding domain prevents ERE-mediated transcription via this response pathway. The hormone–receptor dimers can still act at non-ERE-dependent transcription factors. Both mouse strains retain the ability to respond to estrogens at ESR2. To isolate the organizational and activational effects of estradiol modification of circadian rhythms, we used gonadectomized female NERKI, ERKO, and WT mice with and without estradiol replacement. We characterized their daily wheel-running behaviors in 12:12 lightdark (LD) entrainment, constant dark conditions (DD), and determined the phase response to light pulses administered throughout the subjective day and night.
METHODS Animal breeding and care Adult female NERKI, ERKO, and WT animals were used for these experiments (total n ¼ 78). The sample sizes for each variable are indicated in the figure legends. ERKO mice lack the ESR1 gene (ESR1/) (Dupont et al., 2000) (received from Dr. Pierre Chambon, IGBMC, Illkirch CEDEX, France). NERKI mice (ESR1/AA) have one knock-in copy of a mutated form of ESR1 (AA) (Jakacka et al., 2002) (received from Dr. J. Larry Jameson; University of Pennsylvania, Philadelphia, PA).
In our breeding colony established at the University of Illinois, both NERKI and ERKO lines have been bred onto a congenic C57BL/6J background for the current experiments. Because NERKI (ESR1+/AA) heterozygous females are infertile, heterozygous ERKO (ESR1+/) females are used to produce both strains of mice. They are bred to ESR1+/ males to generate ERKO (ESR1/) and WT offspring and bred to ESR1+/AA males to generate NERKI (ESR1/AA) and WT offspring. The NERKI genotype is thus bred on an ERKO background; these animals are null for normal ESR1 and only express the non-ERE-binding knock-in ESR1 mutation. WT animals used as the controls in these experiments come from both the NERKI and ERKO breeding pairs. Food and water were given ad libitum. Breeding animals and pre-weaned animals (521 d) were maintained on Teklad 8626 rodent diet (Harlan Laboratories, Indianapolis, IN). Prior to activity monitoring, adult animals were given Teklad 2016 diet, which contains low soy estrogens (isoflavones) in the range of non-detectable to 20 mg/kg, and remained on this diet for the remainder of the studies. All animal procedures were approved by the Institutional Animal Care and Use Committee of the University of Illinois and were conducted in accordance with the NIH Guide for the Care and Use of Laboratory Animals. Research conducted here was also in accordance with the ethical standards of the journal (Portaluppi et al., 2008).
Gonadectomy and estradiol replacement Adult females (4 2 months old) were gonadectomized via bilateral incisions on the dorsolateral surface while under ketamine/xylazine anesthesia. Incisions were closed with clips and animals were given buprenorphine for post-operative analgesia. Animals were allowed to recover for at least 7 d before data were collected. At the time of surgery, animals were given one of the following treatments: no hormone replacement (ovx), one silastic capsule containing estradiol benzoate in sesame oil (low E), or three capsules containing estradiol benzoate (high E). Silastic capsules (1.02 mm inner diameter, 2.16 mm outer diameter, 10 mm functional length; Dow Corning, Midland, MI) contained 1 mg/mL estradiol benzoate and were sealed with silastic adhesive on both ends. In preliminary work, we found no difference in total activity, LD ratio, phase angle, and tau in DD between mice with a vehicle capsule compared with no capsule. The estradiol dose was selected based on previous reports of the dose of estradiol sufficient to elicit activity changes (Ogawa et al., 2003). Further, this dose was chosen because in trials in our lab it elicited an increase in wheel running revolutions in gonadectomized female WT mice. Serum concentrations of estradiol (E) were determined for a subset of mice 10 d following capsule implantation and were less than the detectable level of the assay (59 pg/mL, DRG International, Mountainside, NJ) even for the high E replacement group. Chronobiology International
ESR1 modifies circadian rhythms in females At the end of the experiments, animals were sacrificed and we examined the size and color of the estrogen sensitive tissues and the status of the capsule in treated animals. In all cases, a volume of oil was present within the capsules. Further, females treated with high E exhibited hypertrophied uteri at the end of the experiments compared with the relatively small and pale colored uteri of control animals.
Determination of daily activity patterns and circadian variables Mice were maintained in 12 h light:12 h dark (LD) cycles unless otherwise indicated, and the light intensity in the cages ranged from 220 to 360 lux (average 290 lux). Mice were individually housed in cages (28 16 12 cm) equipped with a metal wheel affixed to the top of the cage. Wheel-running activity was recorded and visualized using VitalView and ActiView (MiniMitter, Bend, OR). Wheel revolutions were registered by a magnetic switch and recorded in 10-min bins of activity. The following variables were quantified: average daily wheel revolutions, ratio of activity in light compared to dark (LD ratio), phase angle of activity onset relative to lightsoff, free-running period (tau), and length of active phase (alpha) in constant dark (DD) conditions. For all parameters in 12:12 LD (activity distribution, average daily activity, LD ratio, and phase angle), measures were calculated from at least five consecutive days. For each variable, we used a sample of animals from the total number (n ¼ 78), and there was no specific timeline of experiments that was standard for all animals. Animals were given 10 d between experiments to minimize any effects of previous conditions on the data collected. The average daily wheel revolutions for each animal were calculated and were then used to calculate the group averages for each genotype and sex. LD ratios were averaged for 5 d for each individual animal, and then these values were averaged to determine the LD ratio for each group. The phase angle of activity onset is measured relative to the time of lights-off. The onset of the daily activity bout was defined as the first of three consecutive 10-min bins of activity that was not separated by more than two 10-min bins of inactivity before the next recorded active bout. The free-running period was recorded in DD conditions for 10 consecutive days. Data from the first 3 d in constant conditions were disregarded to avoid transition effects. The tau value was obtained from a chi-square periodogram analysis of wheel-running activity using ActiView. The duration of the period of sustained activity during constant conditions (alpha) was calculated from 5 d of consecutive activity and averaged for each animal. Activity duration was measured from the onset of the daily activity bout to its cessation. Onset of activity was defined as above for phase angle. Activity cessation was defined as the last active bin in a period of consecutive activity of least 40 min that preceded at least 2 h with no sustained activity. !
Informa Healthcare USA, Inc.
639
Determination of phase response to light-pulse The shift in the timing of the activity onset that results from a light-pulse given at a particular time of the subjective day or night is called a phase response. A modified Aschoff Type II procedure was used which enables us to pulse animals with light before their freerunning rhythms drift significantly after entrainment to the light cycle (Mrosovsky, 1996). Animals were housed in a 12:12 LD cycle for at least 5 d prior to light pulse (range 220–360 lux, average 290 lux). They were then placed in DD for at least 24 h prior to a 1-h light-pulse at different times during the second day in DD which correspond with the former Zeitgeber times (ZT) 4, 16, and 22 (ZT 0 ¼ lights-on, ZT 12 ¼ lights-off). Thus animals were pulsed following 28 (ZT 16), 32 (ZT 22), or 40 (ZT 4) hours in DD. Following the light pulse, animals were returned to constant darkness for 7 d. The phase response, measured in minutes, was calculated as the difference between the pre-pulse and post-pulse regression lines on the day following the light pulse. The onset of the daily activity bout was defined as the first of three consecutive 10-min bins of activity that were not separated by more than two 10-min bins of inactivity before the next recorded active bout. The prepulse regression line was determined by the onset of activity for 3 d in LD prior to the light-pulse. The postpulse regression line was determined from the onset of activity for 7 d in constant darkness, eliminating the day immediately following the light pulse to avoid transition effects. Regression lines were determined by eye by two investigators that were blind to the genotype and treatment of the animals. The calculated phase shifts did not differ between the two investigators. Statistical analysis Average daily activity, LD ratio, time of activity onset, and tau and alpha in constant conditions were analyzed with two-factor ANOVAs using genotype and treatment as independent factors (SPSS v.11, IBM, Armonk, NY). A priori comparisons (Tukey HSD) were performed to determine differences within a treatment condition or within a genotype. If classical ESR1 actions modulate the behavioral outcomes then we predicted WT animals would differ from both NERKI and ERKO animals. Whereas if estradiol acted via non-genomic pathways to modify activity patterns then we predicted that NERKI animals would resemble controls and differ from ERKO animals. Within a genotype, we compared groups to determine if there was an effect of number of estradiol capsules. We predicted that high E animals would differ from non-treated animals. Phase shifts (min) were analyzed with two-factor ANOVA with treatment and genotype as independent factors for each time point separately (ZT 4, 16, and 22). A priori comparisons (Tukey HSD) were then performed for each genotype to determine effects of hormone capsule number. Finally, comparisons were made between gonadectomized females to determine the
M. S. Blattner & M. M. Mahoney
organizational effects of estrogen signaling on response to light. Differences were considered significant when p50.05.
RESULTS Average daily activity and activity distribution during LD entrainment We measured average daily wheel-running activity of female WT, NERKI, and ERKO mice with and without E replacement. Average activity levels were significantly affected by genotype [F(2,66) ¼ 7.816, p ¼ 0.001] and E treatment [F(2,66) ¼ 6.285, p ¼ 0.003] (Figure 1). There was no significant interaction between genotype and treatment. Post hoc analyses were used to identify differences between groups. Ovariectomized animals without hormone replacement had no significant differences in activity among genotype groups. Following high E treatment, however, WT females had significantly more activity than ERKO mice (10 526 ± 2488 revolutions relative to three 605 ± 963 revolutions, p ¼ 0.018).
12000
*
#
total acttivity (wheel revoluons))
10000
Average wheel-running activity of NERKI females with high-dose E replacement was intermediate to WT and ERKO mice and was not significantly different from either group. Among animals with low-dose estradiol replacement, WT females had increased activity relative to NERKI (p50.05) and ERKO females (p50.05), however, this low dose was not sufficient to increase activity relative to no hormone replacement in any genotype group. Within WT females, high E caused a significant increase in activity relative to low E and no hormone replacement (p50.05 for both comparisons). In contrast, neither dose of estradiol treatment caused a significant increase in wheel-running activity in NERKI or ERKO females. The amount of activity during the light relative to the dark phase is described by the LD ratio. An LD ratio close to zero reflects an animal that is strictly nocturnal. We found significant effects of genotype [F(2,66) ¼ 6.428, p ¼ 0.003] and treatment [F(2,66) ¼ 5.01, p ¼ 0.009] on LD ratio (Figure 2). Within WT females, mice with high E were significantly more nocturnal than the no treatment group (p ¼ 0.021), and females with low-dose E replacement had an LD ratio that was intermediate and not significantly different than either group. In contrast, among NERKI and ERKO females, E replacement did not significantly consolidate activity to the dark phase of the LD cycle relative to no hormone replacement. When female mice with high E replacement were compared,
0.20
8000
*
0.18
6000
a
ovx
0.16
low E
0.14
high E
b
4000
b
d
b
0.12
LD rao
640
ovx 0.10
low E
0.08
high E c
0.06
2000
a 0.04 0 04 0.02 0 00 0.00
0
WT
NERKI
ERKO
FIGURE 1. Average daily wheel-running activity (±SEM) for WT, NERKI, and ERKO ovariectomized females with no hormone replacement (ovx: white bar) and low-dose (grey bar) and highdose (black bar) estradiol replacement groups. High-dose estradiol replacement significantly increased average activity in WT females relative to low and no replacement (*p50.05). With high-dose estradiol replacement, WT females had significantly greater activity than ERKO females (#p50.05). Among females with low-dose estradiol replacement, WT mice have greater activity than NERKI and ERKO females (a relative to b, p50.05). Sample sizes for the groups are as follows: WT (n ¼ 6–9/group), NERKI (n ¼ 8–10/ group), and ERKO (n ¼ 8/group).
WT
NERKI
ERKO
FIGURE 2. LD ratio (±SEM) for WT, NERKI, and ERKO ovariectomized females with no hormone replacement (ovx: white bar) and low-dose (grey bar) and high-dose (black bar) estradiol replacement. * indicates a significant decrease in LD ratio between WT females with high-dose estradiol and no hormone treatment (p50.05). Among females with high-dose estradiol replacement, WT animals have decreased LD ratio relative to ERKO animals (a relative to b, p50.05). Among animals with low-dose estradiol replacement, NERKI females have decreased LD ratio relative to ERKO females (c relative to d, p50.05). Sample sizes for the groups are as follows: WT (n ¼ 7–9/group), NERKI (n ¼ 8–9/group), and ERKO (n ¼ 8/group). Chronobiology International
ESR1 modifies circadian rhythms in females 24.2
* *
*
WT
641
*
24.1 24
*
NERKI
23.9
*
tau (h hours)
23.8
ovx *
ERKO
*
low E high g E
0
10
20
30
40
phase angle (min aer lights-off)
Free-running period and active period during constant darkness We found a significant effect of treatment on tau in constant darkness [F(2,66) ¼ 5.659, p ¼ 0.005] using a two-factor ANOVA with genotype and treatment as independent factors (Figure 4). Among WT animals, high-dose E significantly shortened tau relative to no hormone replacement (23.6 ± .11 h and 23.95 ± .03 h, respectively, p ¼ 0.014). There were no significant genotype differences within any of the treatment conditions or an interaction between genotype and treatment. Neither ERKO nor NERKI females responded to estradiol with a shortening of tau. We found a significant effect of genotype on alpha, the duration of the active period in constant darkness [F(2,65) ¼ 6.34, p ¼ 0.003] (Figure 5), but no significant !
Informa Healthcare USA, Inc.
23.6
high E
23.4 23.3 23.2 23.1
WT
NERKI
ERKO
FIGURE 4. Free-running period (tau) in constant darkness (hours) (±SEM) for WT, NERKI, and ERKO ovariectomized females with no hormone replacement (ovx: white bars) and low-dose (grey bar) and high-dose (black bar) estradiol replacement. WT females with high-dose estradiol replacement have a shortened period relative to no treatment and low-dose estradiol replacement (*p50.05 for both comparisons). Sample sizes for the groups are as follows: WT (n ¼ 8–9/group), NERKI (n ¼ 8–10/group), and ERKO (n ¼ 7–8/ group).
1000
*
900
* a
800 700
alpha (min n)
WT mice had significantly more night time activity than ERKO mice (p ¼ 0.014), whereas NERKI mice with high E treatment had an intermediate value and did not differ from either WT or ERKO females. With low-dose E replacement, there was greater consolidation of activity in NERKI females than in ERKO females given the same treatment (p ¼ 0.041). The onset of wheel-running activity relative to the time of lights-off is the phase angle and a positive phase angle reflects an animal that began activity after lightsoff. There was a significant effect of hormone treatment on phase angle [F(2,65) ¼ 14.481, p50.001] (Figure 3). There was no significant effect of genotype or interaction between genotype and treatment. Regardless of genotype, all females treated with estradiol (low or high) had a significantly advanced activity onset relative to mice with no hormone replacement (p50.05 in all groups).
ovx low E
23.5
50
FIGURE 3. Phase angle of activity onset relative to lights-off (min after lights-off) (±SEM) for WT, NERKI, and ERKO ovariectomized females with no hormone replacement (ovx: white bar) and lowdose (grey bar) and high-dose dose (black bar) estradiol replacement. Both low- and high-dose estradiol replacement advances the phase angle of activity onset relative to no hormone replacement among all genotypes (* indicates p50.05 for all comparisons). Sample sizes for the groups are as follows: WT (n ¼ 8–10/group), NERKI (n ¼ 7–9/group), and ERKO (n ¼ 7–8/group).
23.7
*
d
c b
600
ovx 500
low E high E
400 300 200 100 0
WT
NERKI
ERKO
FIGURE 5. Duration of the active period (alpha) (±SEM) in constant darkness (min) for WT, NERKI, and ERKO ovariectomized females with no hormone replacement (ovx: white bar) and lowdose (grey bar) and high-dose (black bar) estradiol replacement. Within genotype, high-dose estradiol decreases alpha in both NERKI and ERKO females relative to low-dose replacement (*p50.05). In ERKO females only, high-dose estradiol decreases alpha relative to no hormone replacement (*p50.05). With highdose estradiol replacement, WT females have greater alpha than NERKI females (a relative to b, p50.05). Female NERKI mice with no hormone replacement have decreased alpha relative to ERKO mice (c relative to d, p50.05). Sample sizes for the groups are as follows: WT (n ¼ 7–9/group), NERKI (n ¼ 8–9/group), and ERKO (n ¼ 7–8/group).
M. S. Blattner & M. M. Mahoney
WT
NERKI 125
* phase respo onse (min)
phase respo onse (min)
125 75
ovx low E high E
25 -25 -75 -125 125
a
-175
75 25 -25 -75
b
-125 125 -175
ZT4
ZT16
ZT22
ERKO 125 phase respo onse (min)
642
75 25 -25 -75 125 -125 -175
ZT4
ZT16
ZT22
ZT4
ZT16
ZT22
FIGURE 6. Behavioral phase response to light pulse (min) given during the subjective day (ZT 4, ZT 16, and ZT 22) for WT, NERKI, and ERKO females with no (white bars), low-dose (grey bars), and high-dose (black bars) estradiol replacement. WT mice given high-dose estradiol had significantly greater phase response to light pulse at ZT 4 than WT mice with low and no hormone replacement (*p50.05 for both comparisons). Among gonadectomized females, WT females differed from NERKI females following a light pulse given at ZT 16 (a relative to b, p ¼ 0.022). Sample sizes for the groups are as follows: WT (n ¼ 7–9/group), NERKI (n ¼ 6–10/group), and ERKO (n ¼ 7–9/ group).
effect of treatment or interaction between genotype and treatment. In gonadectomized animals with no hormone replacement, NERKI females had a shorter alpha than ERKO (p ¼ 0.035) females. WT females had a longer active period than NERKI animals with high E replacement (p ¼ 0.013). Within a genotype, replacement of low- or high-dose estradiol did not significantly alter alpha in WT and NERKI mice relative to no hormone replacement. Among ERKO mice, high E replacement lowered alpha relative to low and no replacement (p50.05 for both comparisons).
Phase response to light pulse All groups tested displayed a typical phase response with a phase advance following a light pulse given late in the subjective night (ZT 22) and a phase delay in response to light pulse given early in the subjective night (ZT 16; Figure 6). We examined the magnitude of the phase shifts at each time point using ANOVA with genotype and treatment as independent factors. At ZT 4, there was a significant interaction between genotype and hormone treatment [F(4,57) ¼ 2.87, p ¼ 0.031] but no effect of either factor alone. Post hoc comparisons of this time point indicate that WT females given high E had a larger phase response than either low E (p ¼ 0.013) or control animals (p ¼ 0.014). There was no significant effect of hormone treatment, genotype, or interaction of these variables at either ZT 16 or 22. When only gonadectomized females were compared, there was a significant effect of genotype [F(2,21) ¼ 4.35, p ¼ 0.026) on the phase shift following a light pulse given at ZT 16; WT females differed from NERKI females (p ¼ 0.022). We did not find a significant difference between genotypes at ZT 22 or ZT 4, although there was a trend towards a significance at ZT 4 (p ¼ 0.07) (Figure 6).
DISCUSSION In these studies, we found strong evidence that estradiol acts at ESR1 via ligand-dependent transcription to modulate endogenous period, activity amplitude, and the distribution of activity across the day. Estrogen treatment significantly shortened tau, increased locomotor activity, and consolidated this activity to the night in WT animals but not NERKI or ERKO mice which had altered or deleted ESR1 (Figures 1, 2, and 4). These data also reveal an ESR1-independent mechanism that is responsible for regulating the timing (phase) of activity in the LD environment; estradiol treatment phase advanced activity onset in all animals regardless of genotype (Figure 3). Finally, non-classical ESR1 signaling pathways may contribute to the amplitude of activity (Figure 1). Circadian rhythms are regulated by a master oscillator located in the suprachiasmatic nucleus (SCN). Circadian functionality including the endogenous period and photic entrainment mechanisms are regulated by the SCN and can be modified by circulating estrogens. Both ESR1 and ESR2 are expressed to varying degrees in the rat, mouse, and human SCN (Kruijver et al., 2003; Mitra et al., 2003; Vida et al., 2008). Estrogens may mediate their effects on circadian rhythms by acting directly on cells within the SCN; although these steroid receptors are present in relatively low numbers. It is also possible that estrogens modulate circadian functions, in part, by acting at targets which project to the SCN or by modifying downstream clock-controlled outputs. For example, the medial preoptic area of the hypothalamus both projects to and receives projections from the SCN. This region has a large number of ESR1 containing cells and when estradiol is implanted in this region wheel running activity is increased (de la Iglesia et al., 1995, 1999; Fahrbach et al., 1985; King, 1979). Further, wheel running and overall amount of daily Chronobiology International
ESR1 modifies circadian rhythms in females activity act as non-photic cues which feedback onto the timekeeping mechanism (Mrosovsky, 1999); rodents with a running wheel have a shorter tau than those without a wheel. Thus, the effects of estradiol on activity levels and tau observed here should be interpreted in light of the differences in locomotor activity observed between groups. The phase angle of activity onset has been characterized in different species across the estrous cycle and is typically advanced on days of high circulating estrogens (Albers et al., 1981; Fitzgerald & Zucker, 1976; Labyak & Lee, 1995; Morin et al., 1977a; Wollnik & Turek, 1988). In the current study, all animals regardless of the genotype had a phase advance in the timing of their activity in response to estradiol. These results indicate that estradiol is acting via ESR1-independent pathways and ESR2 signaling is likely one mechanism. It is intriguing to speculate that estradiol is acting on ESR2 containing cells located in the SCN to modulate circadian and daily activity patterns. In the mouse, SCN ESR2 is expressed in relatively higher numbers compared with ESR1 (Vida et al., 2008). Future experiments can tease apart the contributions of the two classical estrogen receptors ESR1 and ESR2 via the use of double knockout mice or treatment with receptor-specific ligands. Interestingly, while the timing of activity is not regulated by ESR1, the coupling of the activity to the LD cycle does appear to be regulated by ESR1. Gonadectomized animals have a larger ratio of activity in the light compared with the dark and treatment with estradiol consolidates that activity to the night, but only in WT animals. One interpretation is that estradiol is increasing the robustness of the coupling of the endogenous timekeeping system to the LD cycle, perhaps by altering photoresponsiveness as has been shown for androgens (Butler et al., 2012). Alternatively, estradiol may increase overall activity levels but the effect is seen robustly in the dark phase of the LD cycle due to the masking effects of light. It is likely that each of these mechanisms modulates the LD ration but the degree to which they contribute remains to be determined. It should be noted that the doses of estradiol that were sufficient to increase wheel-running and alter patterns of daily activity in WT females in the present study were lower than the doses normally used to induce female sexual behaviors. Serum concentrations of estradiol were determined for a subset of mice and were less than the detectable level of the assay (59 pg/mL) even with the high E replacement group. Nevertheless, we find stimulatory effects of estradiol on wheel-running in WT females in this study. Our finding is consistent with Ogawa et al. (2003) that also found significant effects of estradiol on wheel running activity amount despite serum hormone levels that were below threshold for the assay. Interestingly, the dose of estradiol needed to shorten tau and increase wheel running was greater than that needed to alter the phase of the activity rhythms. It remains to be elucidated whether there is a dose!
Informa Healthcare USA, Inc.
643
dependent effect of estradiol on these rhythmic functions. In male mice, there is a dose-dependent effect of androgen on the precision of the clock as well as activity duration and pattern (Butler et al., 2012). This has not been fully investigated with respect to estrogens. Our doses of estradiol are likely low relative to the physiological concentrations of sex hormone in intact females. However, manipulating the low doses revealed differences in the responsiveness of rhythmic functions to circulating hormones. That a low dose of estradiol is sufficient to alter phase angle, but not total daily wheelrunning, suggests increased sensitivity of this parameter to hormonal regulation relative to gross activity measures. Physiologically relevant estradiol concentrations may eliminate the differential responsiveness of activity parameters described here. However, it remains possible that other rhythmic factors such as light responsiveness and SCN cell firing rates or the expression of steroid receptors in the SCN have a dose-dependent response to sex hormones (Brockman et al., 2011; Kuljis et al., 2013). Identifying differences in hormone responsiveness of rhythmic functions may shed light on the degree to which endocrine signaling couples activity rhythms to the external light environment. Further, understanding the nuances of estrogen regulation of biological rhythms will provide insight into mechanisms underlying sex differences in sleep and circadian disorders. We hypothesized that developmental changes in exposure to estradiol would result in differences in response to light cues or daily activity patterns. For example, in aromatase knock-out mice, which do not produce endogenous estradiol, there are significant differences in the behavioral phase shift following light pulses in gonadectomized WT compared to knock-out females (Brockman et al., 2011). Further, there are sex differences between gonadally intact male and female hamsters with respect to the photic phase response curve (Davis et al., 1983). In the current studies, ovariectomized WT, ERKO, and NERKI without estradiol replacement did not differ with respect to activity levels, LD ratio, alpha, or in phase-dependent behavioral shifts in response to light pulses (Figure 6). Estradiol may not have a significant organizational role in the development of biological timekeeping or photic responsiveness, or perhaps, there are compensatory developmental mechanisms that attenuate the effect of altered ESR1 signaling. However, our study only examined behavioral outputs regulated by the circadian system and thus we cannot rule out that steroid hormone exposure during critical periods of the development can permanently modify the adult timekeeping system at other levels. Overall, our data support the well-known activational role of sex steroid hormones on activity level and free running period in females and extend this work by identifying molecular mechanisms underlying these effects. Our data strongly suggest that ESR1 acting via
644
M. S. Blattner & M. M. Mahoney
classic ligand-dependent transcriptional pathways regulates period, average daily level of activity, and distribution of activity across the LD cycle whereas non-ESR1 mechanisms modulate the phase of activity rhythms. Mice with a selective inability to respond to estradiol at ESR1 provide the opportunity to study how hormone actions at this receptor influence circadian rhythms and daily activity via both activational and organizational mechanisms. Together, these data provide new insight as to how reproductive physiology interacts with the timekeeping system to regulate behavioral patterns.
ACKNOWLEDGMENTS The authors thank James Allen, Aaron Fairbanks, Christopher Johnson, Kyle Klein, Athanasios Kondilos, Steven Lord, Arif Molla, and Lauran Wirfs for technical assistance with animal care, surgical procedures, and genotyping. We also thank the lab of Jodi Flaws for assistance with the estradiol assay. Finally, we thank an anonymous reviewer for comments on an earlier draft of this manuscript.
DECLARATION OF INTEREST The authors have no declarations of interest to make. Research conducted here was funded by the University of Illinois Campus Research Board and the Department of Comparative Biosciences.
REFERENCES Albers HE. (1981). Gonadal hormones organize and modulate the circadian system of the rat. Am J Physiol (Regul Integr Comp Physiol). 241:R62–6. Albers HE, Gerall AA, Axelson JF. (1981). Effect of reproductive state on circadian periodicity in the rat. Physiol Behav. 26:21–5. Blattner MS, Mahoney MM. (2013). Photic phase response curve in two strains of mice with impaired responsiveness to estrogens. J Biol Rhythms. 28:291–300. Brockman R, Bunick D, Mahoney MM. (2011). Estradiol deficiency during development modulates the expression of circadian and daily rhythms in male and female aromatase knockout mice. Horm Behav. 60:439–47. Butler MP, Karatsoreos IN, LeSauter J, Silver R. (2012). Dosedependent effects of androgens on the circadian timing system and its response to light. Endocrinology. 153:2344–52. Davis FC, Darrow JM, Menaker M. (1983). Sex differences in the circadian control of hamster wheel-running activity. Am J Physiol (Regul Integr Comp Physiol). 244:R93–105. de la Iglesia HO, Blaustein JD, Bittman EL. (1995). The suprachiasmatic area in the female hamster projects to neurons containing estrogen receptors and GnRH. NeuroReport. 6: 1715–22. de la Iglesia HO, Blaustein JD, Bittman EL. (1999). Oestrogen receptor-a-immunoreactive neurones project to the suprachiasmatic nucleus of the female Syrian hamster. J Neuroendocrinol. 11:481–90. Dupont S, Krust A, Gansmuller A, et al. (2000). Effect of single and compound knockouts of estrogen receptors alpha (ERalpha) and beta (ERbeta) on mouse reproductive phenotypes. Development. 127:4277–91.
Fahrbach SE, Meisel RL, Pfaff DW. (1985). Preoptic implants of estradiol increase wheel running but not the open field activity of female rats. Physiol Behav. 35:985–92. Fitzgerald K, Zucker I. (1976). Circadian organization of the estrous cycle of the golden hamster. Proc Natl Acad Sci USA. 73:2923–7. Gentry RT, Wade GN. (1976). Sex differences in sensitivity of food intake, body weight, and running-wheel activity to ovarian steroids in rats. J Comp Physiol Psychol. 90:747–54. Hall JM, Couse JF, Korach KS. (2001). The multifaceted mechanisms of estradiol and estrogen receptor signaling. J Biol Chem. 276:36869–72. Jakacka M, Ito M, Martinson F, et al. (2002). An estrogen receptor (ER)a deoxyribonucleic acid-binding domain knock-in mutation provides evidence for nonclassical ER pathway signaling in vivo. Mol Endocrinol. 16:2188–201. King JM. (1979). Effects of lesions of the amygdala, preoptic area, and hypothalamus on estradiol-induced activity in the female rat. J Comp Physiol Psychol. 93:360–7. Kousteni S, Bellido T, Plotkin LI, et al. (2001). Nongenotropic, sexnonspecific signaling through the estrogen or androgen receptors: Dissociation from transcriptional activity. Cell. 104:719–30. Kuljis DA, Loh DH, Truong DT, et al. (2013). Gonadal- and sexchromosome dependent sex differences in the circadian system. Endocrinology. 154:1501–12. Kruijver FP, Balesar R, Espila AM, et al. (2003). Estrogen-receptorbeta distribution in the human hypothalamus: Similarities and differences with ERalpha distribution. J Comp Neurol. 466: 251–77. Labyak SE, Lee TM. (1995). Estrus- and steroid-induced changes in circadian rhythms in a diurnal rodent, Octodon degus. Physiol Behav. 58:573–85. Mahoney MM, Rossi BV, Hagenauer MH, Lee TM. (2011). Characterization of the estrous cycle in Octodon degus. Biol Reprod. 84:664–71. McDevitt MA, Glidewell-Kenney C, Jimenez MA, et al. (2008). New insights into the classical and non-classical actions of estrogen: Evidence from estrogen receptor knock-out and knock-in mice. Mol Cell Endocrinol. 290:24–30. Mitra SW, Hoskin E, Yudkovitz J, et al. (2003). Immunolocalization of estrogen receptor beta in the mouse brain: Comparison with estrogen receptor alpha. Endocrinology. 144:2055–67. Morin LP, Fitzgerald KM, Rusak B, Zucker I. (1977a). Circadian organization and neural mediation of hamster reproductive rhythms. Psychoneuroendocrinology. 2:73–98. Morin LP, Fitzgerald KM, Zucker I. (1977b). Estradiol shortens the period of hamster circadian rhythms. Science. 196:305–7. Mrosovsky N. (1996). Methods of measuring phase shifts: Why I continue to use an Aschoff type II procedure despite the skepticism of referees. Chronobiol Int. 13:387–92. Mrosovsky N. (1999). Further experiments on the relationship between the period of circadian rhythms and locomotor activity levels in hamsters. Physiol Behav. 66:797–801. Ogawa S, Chan J, Gustafsson JA, et al. (2003). Estrogen increases locomotor activity in mice through estrogen receptor alpha: Specificity for the type of activity. Endocrinology. 144:230–9. Portaluppi F, Touitou Y, Smolensky MH. (2008). Ethical and methodological standards for laboratory and medical biological rhythm research. Chronobiol Int. 25:999–1016. Vida B, Hrabovszky E, Kalamatianos T, et al. (2008). Oestrogen receptor alpha and beta immunoreactive cells in the suprachiasmatic nucleus of mice: Distribution, sex differences and regulation by gonadal hormones. J Neuroendocrinol. 20:1270–7. Wollnik F, Turek FW. (1988). Estrous correlated modulations of circadian and ultradian wheel-running activity rhythms in LEW/Ztm rats. Physiol Behav. 43:389–96. Zucker I, Fitzgerald KM, Morin LP. (1980). Sex differentiation of the circadian system in the golden hamster. Am J Physiol (Regul Integr Comp Physiol). 238:R97–101.
Chronobiology International
Copyright of Chronobiology International: The Journal of Biological & Medical Rhythm Research is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
ORIGINAL ARTICLE
Estrogen receptor 1 modulates circadian rhythms in adult female mice Margaret S. Blattner1,2 and Megan M. Mahoney2,3 1
College of Medicine, 2Neuroscience Program and 3Department of Comparative Biosciences, University of Illinois at Urbana Champaign, Urbana, IL, USA
Estradiol influences the level and distribution of daily activity, the duration of the free-running period, and the behavioral phase response to light pulses. However, the mechanisms by which estradiol regulates daily and circadian rhythms are not fully understood. We tested the hypothesis that estrogens modulate daily activity patterns via both classical and ‘‘non-classical’’ actions at the estrogen receptor subtype 1 (ESR1). We used female transgenic mice with mutations in their estrogen response pathways; ESR1 knock-out (ERKO) mice and ‘‘non-classical’’ estrogen receptor knock-in (NERKI) mice. NERKI mice have an ESR1 receptor with a mutation in the estrogen-response-element binding domain, allowing only actions via ‘‘non-classical’’ genomic and second messenger pathways. Ovariectomized female NERKI, ERKO, and wildtype (WT) mice were given a subcutaneous capsule with low- or high-dose estradiol and compared with counterparts with no hormone replacement. We measured wheel-running activity in a light:dark cycle and constant darkness, and the behavioral phase response to light pulses given at different points during the subjective day and night. Estradiol increased average daily wheel-running, consolidated activity to the dark phase, and shortened the endogenous period in WT, but not NERKI and ERKO mice. The timing of activity onset during entrainment was advanced in all estradiol-treated animals regardless of genotype suggesting an ESR1-independent mechanism. We propose that estradiol modifies period, activity level, and distribution of activity via classical actions of ESR1 whereas an ESR1 independent mechanism regulates the phase of rhythms. Abbreviations: ERKO, estrogen receptor knock-out; NERKI, nontraditional estrogen receptor knock-in; ESR1, estrogen receptor 1 (ERa); ESR2, estrogen receptor 2 (ERb)
Keywords: Estradiol, ERKO, ERa, locomotor activity, tau, phase response, phase angle
INTRODUCTION Developmental changes, including exposure to steroid hormones, also modulate the expression of circadian activity in adulthood as well as alter the responsiveness of the timekeeping system to circulating hormones. In adult hamsters, treatment with estradiol shortens the free running period in females but not males (Zucker et al., 1980). Adult female rats perinatally exposed to androgens have changes in tau in response to estradiol treatment which mimic those of males (Albers, 1981). Adult aromatase knock-out mice that cannot produce endogenous estradiol display decreased daily activity level, increased daytime activity, and alterations in their timing (phase angle) of activity onset relative to wildtype counterparts, suggesting that the absence of estradiol during development modifies the expression of daily activity rhythms throughout life (Brockman et al., 2011). Traditionally, estrogens mediate their actions through two distinct receptor subtypes: estrogen
20 14
In females, estrogens influence the amount and distribution of daily activity; however, the mechanisms underlying these effects remain largely unknown. The stage of the reproductive cycle influences activity patterns and daily rhythms. Specifically, gonadally intact female rats, hamsters, and degus have the greatest level of activity during proestrus and estrus, when circulating estrogen levels are highest (Mahoney et al., 2011; Morin et al., 1977b; Wollnik & Turek, 1988). Further, in these species the period of the endogenous free-running rhythm (tau) in constant conditions is shortest during estrus (Albers et al., 1981; Fitzgerald & Zucker, 1976; Labyak & Lee, 1995). Ovariectomy of female rodents causes a decrease of daily wheel-running activity and a lengthening of tau, and estradiol replacement increases activity levels and shortens tau (Albers, 1981; Gentry & Wade, 1976; Labyak & Lee, 1995; Morin et al., 1977b).
Submitted September 5, 2013, Returned for revision January 15, 2014, Accepted January 17, 2014
Correspondence: Megan M. Mahoney, Department of Comparative Biosciences, 3639 VMSB, MC-002, 2001 South Lincoln Avenue, Urbana, IL 61802, USA. Tel: +1 217 333 7578. Fax: +1 217 244 1652. E-mail: [email protected]
637
638
M. S. Blattner & M. M. Mahoney
receptor subtype 1 (estrogen receptor subtype 1 (ESR1), or estrogen receptor a) and estrogen receptor subtype 2 (ESR2, or estrogen receptor b). In ‘‘classical’’ nuclear receptor-mediated transcriptional activation, estrogens bind to receptors and the ligand–receptor complex then forms dimers. These dimers bind to the estrogen response element (ERE) on regulatory regions of target genes to alter gene expression. There are also more recently characterized ‘‘non-classical’’ mechanisms of estrogen signaling including ligand-independent estrogen receptor activation that occurs via second messenger pathways and estrogen receptor activation of gene transcription that occurs independently from ERE-dependent pathways (Hall et al., 2001; Kousteni et al., 2001; McDevitt et al., 2008). Previous data from this and other labs provide strong evidence that estrogens acting at ESR1 increase wheel-running activity. Further, classical and non-classical actions of ESR1 regulate the distribution of activity across the LD cycle and the response to light (Blattner & Mahoney, 2013; Ogawa et al., 2003). However, it is unknown if these modulatory effects of estrogens occur developmentally, in adulthood, or as some combination of these factors. Here we test the hypothesis that estradiol signaling at ESR1 mediates its effects on the expression of circadian rhythms and daily activity through both classical EREmediated and ‘‘non-classical’’ pathways. We used two strains of mice with transgenic modifications of their ability to respond to estrogens at ESR1. ESR1 knock-out (ERKO) mice are unable to respond to estradiol via this receptor. ‘‘Non-classical’’ estrogen receptor knock-in (NERKI) mice retain the ability to bind estrogens at ESR1; however, a mutation in the ERE-binding domain prevents ERE-mediated transcription via this response pathway. The hormone–receptor dimers can still act at non-ERE-dependent transcription factors. Both mouse strains retain the ability to respond to estrogens at ESR2. To isolate the organizational and activational effects of estradiol modification of circadian rhythms, we used gonadectomized female NERKI, ERKO, and WT mice with and without estradiol replacement. We characterized their daily wheel-running behaviors in 12:12 lightdark (LD) entrainment, constant dark conditions (DD), and determined the phase response to light pulses administered throughout the subjective day and night.
METHODS Animal breeding and care Adult female NERKI, ERKO, and WT animals were used for these experiments (total n ¼ 78). The sample sizes for each variable are indicated in the figure legends. ERKO mice lack the ESR1 gene (ESR1/) (Dupont et al., 2000) (received from Dr. Pierre Chambon, IGBMC, Illkirch CEDEX, France). NERKI mice (ESR1/AA) have one knock-in copy of a mutated form of ESR1 (AA) (Jakacka et al., 2002) (received from Dr. J. Larry Jameson; University of Pennsylvania, Philadelphia, PA).
In our breeding colony established at the University of Illinois, both NERKI and ERKO lines have been bred onto a congenic C57BL/6J background for the current experiments. Because NERKI (ESR1+/AA) heterozygous females are infertile, heterozygous ERKO (ESR1+/) females are used to produce both strains of mice. They are bred to ESR1+/ males to generate ERKO (ESR1/) and WT offspring and bred to ESR1+/AA males to generate NERKI (ESR1/AA) and WT offspring. The NERKI genotype is thus bred on an ERKO background; these animals are null for normal ESR1 and only express the non-ERE-binding knock-in ESR1 mutation. WT animals used as the controls in these experiments come from both the NERKI and ERKO breeding pairs. Food and water were given ad libitum. Breeding animals and pre-weaned animals (521 d) were maintained on Teklad 8626 rodent diet (Harlan Laboratories, Indianapolis, IN). Prior to activity monitoring, adult animals were given Teklad 2016 diet, which contains low soy estrogens (isoflavones) in the range of non-detectable to 20 mg/kg, and remained on this diet for the remainder of the studies. All animal procedures were approved by the Institutional Animal Care and Use Committee of the University of Illinois and were conducted in accordance with the NIH Guide for the Care and Use of Laboratory Animals. Research conducted here was also in accordance with the ethical standards of the journal (Portaluppi et al., 2008).
Gonadectomy and estradiol replacement Adult females (4 2 months old) were gonadectomized via bilateral incisions on the dorsolateral surface while under ketamine/xylazine anesthesia. Incisions were closed with clips and animals were given buprenorphine for post-operative analgesia. Animals were allowed to recover for at least 7 d before data were collected. At the time of surgery, animals were given one of the following treatments: no hormone replacement (ovx), one silastic capsule containing estradiol benzoate in sesame oil (low E), or three capsules containing estradiol benzoate (high E). Silastic capsules (1.02 mm inner diameter, 2.16 mm outer diameter, 10 mm functional length; Dow Corning, Midland, MI) contained 1 mg/mL estradiol benzoate and were sealed with silastic adhesive on both ends. In preliminary work, we found no difference in total activity, LD ratio, phase angle, and tau in DD between mice with a vehicle capsule compared with no capsule. The estradiol dose was selected based on previous reports of the dose of estradiol sufficient to elicit activity changes (Ogawa et al., 2003). Further, this dose was chosen because in trials in our lab it elicited an increase in wheel running revolutions in gonadectomized female WT mice. Serum concentrations of estradiol (E) were determined for a subset of mice 10 d following capsule implantation and were less than the detectable level of the assay (59 pg/mL, DRG International, Mountainside, NJ) even for the high E replacement group. Chronobiology International
ESR1 modifies circadian rhythms in females At the end of the experiments, animals were sacrificed and we examined the size and color of the estrogen sensitive tissues and the status of the capsule in treated animals. In all cases, a volume of oil was present within the capsules. Further, females treated with high E exhibited hypertrophied uteri at the end of the experiments compared with the relatively small and pale colored uteri of control animals.
Determination of daily activity patterns and circadian variables Mice were maintained in 12 h light:12 h dark (LD) cycles unless otherwise indicated, and the light intensity in the cages ranged from 220 to 360 lux (average 290 lux). Mice were individually housed in cages (28 16 12 cm) equipped with a metal wheel affixed to the top of the cage. Wheel-running activity was recorded and visualized using VitalView and ActiView (MiniMitter, Bend, OR). Wheel revolutions were registered by a magnetic switch and recorded in 10-min bins of activity. The following variables were quantified: average daily wheel revolutions, ratio of activity in light compared to dark (LD ratio), phase angle of activity onset relative to lightsoff, free-running period (tau), and length of active phase (alpha) in constant dark (DD) conditions. For all parameters in 12:12 LD (activity distribution, average daily activity, LD ratio, and phase angle), measures were calculated from at least five consecutive days. For each variable, we used a sample of animals from the total number (n ¼ 78), and there was no specific timeline of experiments that was standard for all animals. Animals were given 10 d between experiments to minimize any effects of previous conditions on the data collected. The average daily wheel revolutions for each animal were calculated and were then used to calculate the group averages for each genotype and sex. LD ratios were averaged for 5 d for each individual animal, and then these values were averaged to determine the LD ratio for each group. The phase angle of activity onset is measured relative to the time of lights-off. The onset of the daily activity bout was defined as the first of three consecutive 10-min bins of activity that was not separated by more than two 10-min bins of inactivity before the next recorded active bout. The free-running period was recorded in DD conditions for 10 consecutive days. Data from the first 3 d in constant conditions were disregarded to avoid transition effects. The tau value was obtained from a chi-square periodogram analysis of wheel-running activity using ActiView. The duration of the period of sustained activity during constant conditions (alpha) was calculated from 5 d of consecutive activity and averaged for each animal. Activity duration was measured from the onset of the daily activity bout to its cessation. Onset of activity was defined as above for phase angle. Activity cessation was defined as the last active bin in a period of consecutive activity of least 40 min that preceded at least 2 h with no sustained activity. !
Informa Healthcare USA, Inc.
639
Determination of phase response to light-pulse The shift in the timing of the activity onset that results from a light-pulse given at a particular time of the subjective day or night is called a phase response. A modified Aschoff Type II procedure was used which enables us to pulse animals with light before their freerunning rhythms drift significantly after entrainment to the light cycle (Mrosovsky, 1996). Animals were housed in a 12:12 LD cycle for at least 5 d prior to light pulse (range 220–360 lux, average 290 lux). They were then placed in DD for at least 24 h prior to a 1-h light-pulse at different times during the second day in DD which correspond with the former Zeitgeber times (ZT) 4, 16, and 22 (ZT 0 ¼ lights-on, ZT 12 ¼ lights-off). Thus animals were pulsed following 28 (ZT 16), 32 (ZT 22), or 40 (ZT 4) hours in DD. Following the light pulse, animals were returned to constant darkness for 7 d. The phase response, measured in minutes, was calculated as the difference between the pre-pulse and post-pulse regression lines on the day following the light pulse. The onset of the daily activity bout was defined as the first of three consecutive 10-min bins of activity that were not separated by more than two 10-min bins of inactivity before the next recorded active bout. The prepulse regression line was determined by the onset of activity for 3 d in LD prior to the light-pulse. The postpulse regression line was determined from the onset of activity for 7 d in constant darkness, eliminating the day immediately following the light pulse to avoid transition effects. Regression lines were determined by eye by two investigators that were blind to the genotype and treatment of the animals. The calculated phase shifts did not differ between the two investigators. Statistical analysis Average daily activity, LD ratio, time of activity onset, and tau and alpha in constant conditions were analyzed with two-factor ANOVAs using genotype and treatment as independent factors (SPSS v.11, IBM, Armonk, NY). A priori comparisons (Tukey HSD) were performed to determine differences within a treatment condition or within a genotype. If classical ESR1 actions modulate the behavioral outcomes then we predicted WT animals would differ from both NERKI and ERKO animals. Whereas if estradiol acted via non-genomic pathways to modify activity patterns then we predicted that NERKI animals would resemble controls and differ from ERKO animals. Within a genotype, we compared groups to determine if there was an effect of number of estradiol capsules. We predicted that high E animals would differ from non-treated animals. Phase shifts (min) were analyzed with two-factor ANOVA with treatment and genotype as independent factors for each time point separately (ZT 4, 16, and 22). A priori comparisons (Tukey HSD) were then performed for each genotype to determine effects of hormone capsule number. Finally, comparisons were made between gonadectomized females to determine the
M. S. Blattner & M. M. Mahoney
organizational effects of estrogen signaling on response to light. Differences were considered significant when p50.05.
RESULTS Average daily activity and activity distribution during LD entrainment We measured average daily wheel-running activity of female WT, NERKI, and ERKO mice with and without E replacement. Average activity levels were significantly affected by genotype [F(2,66) ¼ 7.816, p ¼ 0.001] and E treatment [F(2,66) ¼ 6.285, p ¼ 0.003] (Figure 1). There was no significant interaction between genotype and treatment. Post hoc analyses were used to identify differences between groups. Ovariectomized animals without hormone replacement had no significant differences in activity among genotype groups. Following high E treatment, however, WT females had significantly more activity than ERKO mice (10 526 ± 2488 revolutions relative to three 605 ± 963 revolutions, p ¼ 0.018).
12000
*
#
total acttivity (wheel revoluons))
10000
Average wheel-running activity of NERKI females with high-dose E replacement was intermediate to WT and ERKO mice and was not significantly different from either group. Among animals with low-dose estradiol replacement, WT females had increased activity relative to NERKI (p50.05) and ERKO females (p50.05), however, this low dose was not sufficient to increase activity relative to no hormone replacement in any genotype group. Within WT females, high E caused a significant increase in activity relative to low E and no hormone replacement (p50.05 for both comparisons). In contrast, neither dose of estradiol treatment caused a significant increase in wheel-running activity in NERKI or ERKO females. The amount of activity during the light relative to the dark phase is described by the LD ratio. An LD ratio close to zero reflects an animal that is strictly nocturnal. We found significant effects of genotype [F(2,66) ¼ 6.428, p ¼ 0.003] and treatment [F(2,66) ¼ 5.01, p ¼ 0.009] on LD ratio (Figure 2). Within WT females, mice with high E were significantly more nocturnal than the no treatment group (p ¼ 0.021), and females with low-dose E replacement had an LD ratio that was intermediate and not significantly different than either group. In contrast, among NERKI and ERKO females, E replacement did not significantly consolidate activity to the dark phase of the LD cycle relative to no hormone replacement. When female mice with high E replacement were compared,
0.20
8000
*
0.18
6000
a
ovx
0.16
low E
0.14
high E
b
4000
b
d
b
0.12
LD rao
640
ovx 0.10
low E
0.08
high E c
0.06
2000
a 0.04 0 04 0.02 0 00 0.00
0
WT
NERKI
ERKO
FIGURE 1. Average daily wheel-running activity (±SEM) for WT, NERKI, and ERKO ovariectomized females with no hormone replacement (ovx: white bar) and low-dose (grey bar) and highdose (black bar) estradiol replacement groups. High-dose estradiol replacement significantly increased average activity in WT females relative to low and no replacement (*p50.05). With high-dose estradiol replacement, WT females had significantly greater activity than ERKO females (#p50.05). Among females with low-dose estradiol replacement, WT mice have greater activity than NERKI and ERKO females (a relative to b, p50.05). Sample sizes for the groups are as follows: WT (n ¼ 6–9/group), NERKI (n ¼ 8–10/ group), and ERKO (n ¼ 8/group).
WT
NERKI
ERKO
FIGURE 2. LD ratio (±SEM) for WT, NERKI, and ERKO ovariectomized females with no hormone replacement (ovx: white bar) and low-dose (grey bar) and high-dose (black bar) estradiol replacement. * indicates a significant decrease in LD ratio between WT females with high-dose estradiol and no hormone treatment (p50.05). Among females with high-dose estradiol replacement, WT animals have decreased LD ratio relative to ERKO animals (a relative to b, p50.05). Among animals with low-dose estradiol replacement, NERKI females have decreased LD ratio relative to ERKO females (c relative to d, p50.05). Sample sizes for the groups are as follows: WT (n ¼ 7–9/group), NERKI (n ¼ 8–9/group), and ERKO (n ¼ 8/group). Chronobiology International
ESR1 modifies circadian rhythms in females 24.2
* *
*
WT
641
*
24.1 24
*
NERKI
23.9
*
tau (h hours)
23.8
ovx *
ERKO
*
low E high g E
0
10
20
30
40
phase angle (min aer lights-off)
Free-running period and active period during constant darkness We found a significant effect of treatment on tau in constant darkness [F(2,66) ¼ 5.659, p ¼ 0.005] using a two-factor ANOVA with genotype and treatment as independent factors (Figure 4). Among WT animals, high-dose E significantly shortened tau relative to no hormone replacement (23.6 ± .11 h and 23.95 ± .03 h, respectively, p ¼ 0.014). There were no significant genotype differences within any of the treatment conditions or an interaction between genotype and treatment. Neither ERKO nor NERKI females responded to estradiol with a shortening of tau. We found a significant effect of genotype on alpha, the duration of the active period in constant darkness [F(2,65) ¼ 6.34, p ¼ 0.003] (Figure 5), but no significant !
Informa Healthcare USA, Inc.
23.6
high E
23.4 23.3 23.2 23.1
WT
NERKI
ERKO
FIGURE 4. Free-running period (tau) in constant darkness (hours) (±SEM) for WT, NERKI, and ERKO ovariectomized females with no hormone replacement (ovx: white bars) and low-dose (grey bar) and high-dose (black bar) estradiol replacement. WT females with high-dose estradiol replacement have a shortened period relative to no treatment and low-dose estradiol replacement (*p50.05 for both comparisons). Sample sizes for the groups are as follows: WT (n ¼ 8–9/group), NERKI (n ¼ 8–10/group), and ERKO (n ¼ 7–8/ group).
1000
*
900
* a
800 700
alpha (min n)
WT mice had significantly more night time activity than ERKO mice (p ¼ 0.014), whereas NERKI mice with high E treatment had an intermediate value and did not differ from either WT or ERKO females. With low-dose E replacement, there was greater consolidation of activity in NERKI females than in ERKO females given the same treatment (p ¼ 0.041). The onset of wheel-running activity relative to the time of lights-off is the phase angle and a positive phase angle reflects an animal that began activity after lightsoff. There was a significant effect of hormone treatment on phase angle [F(2,65) ¼ 14.481, p50.001] (Figure 3). There was no significant effect of genotype or interaction between genotype and treatment. Regardless of genotype, all females treated with estradiol (low or high) had a significantly advanced activity onset relative to mice with no hormone replacement (p50.05 in all groups).
ovx low E
23.5
50
FIGURE 3. Phase angle of activity onset relative to lights-off (min after lights-off) (±SEM) for WT, NERKI, and ERKO ovariectomized females with no hormone replacement (ovx: white bar) and lowdose (grey bar) and high-dose dose (black bar) estradiol replacement. Both low- and high-dose estradiol replacement advances the phase angle of activity onset relative to no hormone replacement among all genotypes (* indicates p50.05 for all comparisons). Sample sizes for the groups are as follows: WT (n ¼ 8–10/group), NERKI (n ¼ 7–9/group), and ERKO (n ¼ 7–8/group).
23.7
*
d
c b
600
ovx 500
low E high E
400 300 200 100 0
WT
NERKI
ERKO
FIGURE 5. Duration of the active period (alpha) (±SEM) in constant darkness (min) for WT, NERKI, and ERKO ovariectomized females with no hormone replacement (ovx: white bar) and lowdose (grey bar) and high-dose (black bar) estradiol replacement. Within genotype, high-dose estradiol decreases alpha in both NERKI and ERKO females relative to low-dose replacement (*p50.05). In ERKO females only, high-dose estradiol decreases alpha relative to no hormone replacement (*p50.05). With highdose estradiol replacement, WT females have greater alpha than NERKI females (a relative to b, p50.05). Female NERKI mice with no hormone replacement have decreased alpha relative to ERKO mice (c relative to d, p50.05). Sample sizes for the groups are as follows: WT (n ¼ 7–9/group), NERKI (n ¼ 8–9/group), and ERKO (n ¼ 7–8/group).
M. S. Blattner & M. M. Mahoney
WT
NERKI 125
* phase respo onse (min)
phase respo onse (min)
125 75
ovx low E high E
25 -25 -75 -125 125
a
-175
75 25 -25 -75
b
-125 125 -175
ZT4
ZT16
ZT22
ERKO 125 phase respo onse (min)
642
75 25 -25 -75 125 -125 -175
ZT4
ZT16
ZT22
ZT4
ZT16
ZT22
FIGURE 6. Behavioral phase response to light pulse (min) given during the subjective day (ZT 4, ZT 16, and ZT 22) for WT, NERKI, and ERKO females with no (white bars), low-dose (grey bars), and high-dose (black bars) estradiol replacement. WT mice given high-dose estradiol had significantly greater phase response to light pulse at ZT 4 than WT mice with low and no hormone replacement (*p50.05 for both comparisons). Among gonadectomized females, WT females differed from NERKI females following a light pulse given at ZT 16 (a relative to b, p ¼ 0.022). Sample sizes for the groups are as follows: WT (n ¼ 7–9/group), NERKI (n ¼ 6–10/group), and ERKO (n ¼ 7–9/ group).
effect of treatment or interaction between genotype and treatment. In gonadectomized animals with no hormone replacement, NERKI females had a shorter alpha than ERKO (p ¼ 0.035) females. WT females had a longer active period than NERKI animals with high E replacement (p ¼ 0.013). Within a genotype, replacement of low- or high-dose estradiol did not significantly alter alpha in WT and NERKI mice relative to no hormone replacement. Among ERKO mice, high E replacement lowered alpha relative to low and no replacement (p50.05 for both comparisons).
Phase response to light pulse All groups tested displayed a typical phase response with a phase advance following a light pulse given late in the subjective night (ZT 22) and a phase delay in response to light pulse given early in the subjective night (ZT 16; Figure 6). We examined the magnitude of the phase shifts at each time point using ANOVA with genotype and treatment as independent factors. At ZT 4, there was a significant interaction between genotype and hormone treatment [F(4,57) ¼ 2.87, p ¼ 0.031] but no effect of either factor alone. Post hoc comparisons of this time point indicate that WT females given high E had a larger phase response than either low E (p ¼ 0.013) or control animals (p ¼ 0.014). There was no significant effect of hormone treatment, genotype, or interaction of these variables at either ZT 16 or 22. When only gonadectomized females were compared, there was a significant effect of genotype [F(2,21) ¼ 4.35, p ¼ 0.026) on the phase shift following a light pulse given at ZT 16; WT females differed from NERKI females (p ¼ 0.022). We did not find a significant difference between genotypes at ZT 22 or ZT 4, although there was a trend towards a significance at ZT 4 (p ¼ 0.07) (Figure 6).
DISCUSSION In these studies, we found strong evidence that estradiol acts at ESR1 via ligand-dependent transcription to modulate endogenous period, activity amplitude, and the distribution of activity across the day. Estrogen treatment significantly shortened tau, increased locomotor activity, and consolidated this activity to the night in WT animals but not NERKI or ERKO mice which had altered or deleted ESR1 (Figures 1, 2, and 4). These data also reveal an ESR1-independent mechanism that is responsible for regulating the timing (phase) of activity in the LD environment; estradiol treatment phase advanced activity onset in all animals regardless of genotype (Figure 3). Finally, non-classical ESR1 signaling pathways may contribute to the amplitude of activity (Figure 1). Circadian rhythms are regulated by a master oscillator located in the suprachiasmatic nucleus (SCN). Circadian functionality including the endogenous period and photic entrainment mechanisms are regulated by the SCN and can be modified by circulating estrogens. Both ESR1 and ESR2 are expressed to varying degrees in the rat, mouse, and human SCN (Kruijver et al., 2003; Mitra et al., 2003; Vida et al., 2008). Estrogens may mediate their effects on circadian rhythms by acting directly on cells within the SCN; although these steroid receptors are present in relatively low numbers. It is also possible that estrogens modulate circadian functions, in part, by acting at targets which project to the SCN or by modifying downstream clock-controlled outputs. For example, the medial preoptic area of the hypothalamus both projects to and receives projections from the SCN. This region has a large number of ESR1 containing cells and when estradiol is implanted in this region wheel running activity is increased (de la Iglesia et al., 1995, 1999; Fahrbach et al., 1985; King, 1979). Further, wheel running and overall amount of daily Chronobiology International
ESR1 modifies circadian rhythms in females activity act as non-photic cues which feedback onto the timekeeping mechanism (Mrosovsky, 1999); rodents with a running wheel have a shorter tau than those without a wheel. Thus, the effects of estradiol on activity levels and tau observed here should be interpreted in light of the differences in locomotor activity observed between groups. The phase angle of activity onset has been characterized in different species across the estrous cycle and is typically advanced on days of high circulating estrogens (Albers et al., 1981; Fitzgerald & Zucker, 1976; Labyak & Lee, 1995; Morin et al., 1977a; Wollnik & Turek, 1988). In the current study, all animals regardless of the genotype had a phase advance in the timing of their activity in response to estradiol. These results indicate that estradiol is acting via ESR1-independent pathways and ESR2 signaling is likely one mechanism. It is intriguing to speculate that estradiol is acting on ESR2 containing cells located in the SCN to modulate circadian and daily activity patterns. In the mouse, SCN ESR2 is expressed in relatively higher numbers compared with ESR1 (Vida et al., 2008). Future experiments can tease apart the contributions of the two classical estrogen receptors ESR1 and ESR2 via the use of double knockout mice or treatment with receptor-specific ligands. Interestingly, while the timing of activity is not regulated by ESR1, the coupling of the activity to the LD cycle does appear to be regulated by ESR1. Gonadectomized animals have a larger ratio of activity in the light compared with the dark and treatment with estradiol consolidates that activity to the night, but only in WT animals. One interpretation is that estradiol is increasing the robustness of the coupling of the endogenous timekeeping system to the LD cycle, perhaps by altering photoresponsiveness as has been shown for androgens (Butler et al., 2012). Alternatively, estradiol may increase overall activity levels but the effect is seen robustly in the dark phase of the LD cycle due to the masking effects of light. It is likely that each of these mechanisms modulates the LD ration but the degree to which they contribute remains to be determined. It should be noted that the doses of estradiol that were sufficient to increase wheel-running and alter patterns of daily activity in WT females in the present study were lower than the doses normally used to induce female sexual behaviors. Serum concentrations of estradiol were determined for a subset of mice and were less than the detectable level of the assay (59 pg/mL) even with the high E replacement group. Nevertheless, we find stimulatory effects of estradiol on wheel-running in WT females in this study. Our finding is consistent with Ogawa et al. (2003) that also found significant effects of estradiol on wheel running activity amount despite serum hormone levels that were below threshold for the assay. Interestingly, the dose of estradiol needed to shorten tau and increase wheel running was greater than that needed to alter the phase of the activity rhythms. It remains to be elucidated whether there is a dose!
Informa Healthcare USA, Inc.
643
dependent effect of estradiol on these rhythmic functions. In male mice, there is a dose-dependent effect of androgen on the precision of the clock as well as activity duration and pattern (Butler et al., 2012). This has not been fully investigated with respect to estrogens. Our doses of estradiol are likely low relative to the physiological concentrations of sex hormone in intact females. However, manipulating the low doses revealed differences in the responsiveness of rhythmic functions to circulating hormones. That a low dose of estradiol is sufficient to alter phase angle, but not total daily wheelrunning, suggests increased sensitivity of this parameter to hormonal regulation relative to gross activity measures. Physiologically relevant estradiol concentrations may eliminate the differential responsiveness of activity parameters described here. However, it remains possible that other rhythmic factors such as light responsiveness and SCN cell firing rates or the expression of steroid receptors in the SCN have a dose-dependent response to sex hormones (Brockman et al., 2011; Kuljis et al., 2013). Identifying differences in hormone responsiveness of rhythmic functions may shed light on the degree to which endocrine signaling couples activity rhythms to the external light environment. Further, understanding the nuances of estrogen regulation of biological rhythms will provide insight into mechanisms underlying sex differences in sleep and circadian disorders. We hypothesized that developmental changes in exposure to estradiol would result in differences in response to light cues or daily activity patterns. For example, in aromatase knock-out mice, which do not produce endogenous estradiol, there are significant differences in the behavioral phase shift following light pulses in gonadectomized WT compared to knock-out females (Brockman et al., 2011). Further, there are sex differences between gonadally intact male and female hamsters with respect to the photic phase response curve (Davis et al., 1983). In the current studies, ovariectomized WT, ERKO, and NERKI without estradiol replacement did not differ with respect to activity levels, LD ratio, alpha, or in phase-dependent behavioral shifts in response to light pulses (Figure 6). Estradiol may not have a significant organizational role in the development of biological timekeeping or photic responsiveness, or perhaps, there are compensatory developmental mechanisms that attenuate the effect of altered ESR1 signaling. However, our study only examined behavioral outputs regulated by the circadian system and thus we cannot rule out that steroid hormone exposure during critical periods of the development can permanently modify the adult timekeeping system at other levels. Overall, our data support the well-known activational role of sex steroid hormones on activity level and free running period in females and extend this work by identifying molecular mechanisms underlying these effects. Our data strongly suggest that ESR1 acting via
644
M. S. Blattner & M. M. Mahoney
classic ligand-dependent transcriptional pathways regulates period, average daily level of activity, and distribution of activity across the LD cycle whereas non-ESR1 mechanisms modulate the phase of activity rhythms. Mice with a selective inability to respond to estradiol at ESR1 provide the opportunity to study how hormone actions at this receptor influence circadian rhythms and daily activity via both activational and organizational mechanisms. Together, these data provide new insight as to how reproductive physiology interacts with the timekeeping system to regulate behavioral patterns.
ACKNOWLEDGMENTS The authors thank James Allen, Aaron Fairbanks, Christopher Johnson, Kyle Klein, Athanasios Kondilos, Steven Lord, Arif Molla, and Lauran Wirfs for technical assistance with animal care, surgical procedures, and genotyping. We also thank the lab of Jodi Flaws for assistance with the estradiol assay. Finally, we thank an anonymous reviewer for comments on an earlier draft of this manuscript.
DECLARATION OF INTEREST The authors have no declarations of interest to make. Research conducted here was funded by the University of Illinois Campus Research Board and the Department of Comparative Biosciences.
REFERENCES Albers HE. (1981). Gonadal hormones organize and modulate the circadian system of the rat. Am J Physiol (Regul Integr Comp Physiol). 241:R62–6. Albers HE, Gerall AA, Axelson JF. (1981). Effect of reproductive state on circadian periodicity in the rat. Physiol Behav. 26:21–5. Blattner MS, Mahoney MM. (2013). Photic phase response curve in two strains of mice with impaired responsiveness to estrogens. J Biol Rhythms. 28:291–300. Brockman R, Bunick D, Mahoney MM. (2011). Estradiol deficiency during development modulates the expression of circadian and daily rhythms in male and female aromatase knockout mice. Horm Behav. 60:439–47. Butler MP, Karatsoreos IN, LeSauter J, Silver R. (2012). Dosedependent effects of androgens on the circadian timing system and its response to light. Endocrinology. 153:2344–52. Davis FC, Darrow JM, Menaker M. (1983). Sex differences in the circadian control of hamster wheel-running activity. Am J Physiol (Regul Integr Comp Physiol). 244:R93–105. de la Iglesia HO, Blaustein JD, Bittman EL. (1995). The suprachiasmatic area in the female hamster projects to neurons containing estrogen receptors and GnRH. NeuroReport. 6: 1715–22. de la Iglesia HO, Blaustein JD, Bittman EL. (1999). Oestrogen receptor-a-immunoreactive neurones project to the suprachiasmatic nucleus of the female Syrian hamster. J Neuroendocrinol. 11:481–90. Dupont S, Krust A, Gansmuller A, et al. (2000). Effect of single and compound knockouts of estrogen receptors alpha (ERalpha) and beta (ERbeta) on mouse reproductive phenotypes. Development. 127:4277–91.
Fahrbach SE, Meisel RL, Pfaff DW. (1985). Preoptic implants of estradiol increase wheel running but not the open field activity of female rats. Physiol Behav. 35:985–92. Fitzgerald K, Zucker I. (1976). Circadian organization of the estrous cycle of the golden hamster. Proc Natl Acad Sci USA. 73:2923–7. Gentry RT, Wade GN. (1976). Sex differences in sensitivity of food intake, body weight, and running-wheel activity to ovarian steroids in rats. J Comp Physiol Psychol. 90:747–54. Hall JM, Couse JF, Korach KS. (2001). The multifaceted mechanisms of estradiol and estrogen receptor signaling. J Biol Chem. 276:36869–72. Jakacka M, Ito M, Martinson F, et al. (2002). An estrogen receptor (ER)a deoxyribonucleic acid-binding domain knock-in mutation provides evidence for nonclassical ER pathway signaling in vivo. Mol Endocrinol. 16:2188–201. King JM. (1979). Effects of lesions of the amygdala, preoptic area, and hypothalamus on estradiol-induced activity in the female rat. J Comp Physiol Psychol. 93:360–7. Kousteni S, Bellido T, Plotkin LI, et al. (2001). Nongenotropic, sexnonspecific signaling through the estrogen or androgen receptors: Dissociation from transcriptional activity. Cell. 104:719–30. Kuljis DA, Loh DH, Truong DT, et al. (2013). Gonadal- and sexchromosome dependent sex differences in the circadian system. Endocrinology. 154:1501–12. Kruijver FP, Balesar R, Espila AM, et al. (2003). Estrogen-receptorbeta distribution in the human hypothalamus: Similarities and differences with ERalpha distribution. J Comp Neurol. 466: 251–77. Labyak SE, Lee TM. (1995). Estrus- and steroid-induced changes in circadian rhythms in a diurnal rodent, Octodon degus. Physiol Behav. 58:573–85. Mahoney MM, Rossi BV, Hagenauer MH, Lee TM. (2011). Characterization of the estrous cycle in Octodon degus. Biol Reprod. 84:664–71. McDevitt MA, Glidewell-Kenney C, Jimenez MA, et al. (2008). New insights into the classical and non-classical actions of estrogen: Evidence from estrogen receptor knock-out and knock-in mice. Mol Cell Endocrinol. 290:24–30. Mitra SW, Hoskin E, Yudkovitz J, et al. (2003). Immunolocalization of estrogen receptor beta in the mouse brain: Comparison with estrogen receptor alpha. Endocrinology. 144:2055–67. Morin LP, Fitzgerald KM, Rusak B, Zucker I. (1977a). Circadian organization and neural mediation of hamster reproductive rhythms. Psychoneuroendocrinology. 2:73–98. Morin LP, Fitzgerald KM, Zucker I. (1977b). Estradiol shortens the period of hamster circadian rhythms. Science. 196:305–7. Mrosovsky N. (1996). Methods of measuring phase shifts: Why I continue to use an Aschoff type II procedure despite the skepticism of referees. Chronobiol Int. 13:387–92. Mrosovsky N. (1999). Further experiments on the relationship between the period of circadian rhythms and locomotor activity levels in hamsters. Physiol Behav. 66:797–801. Ogawa S, Chan J, Gustafsson JA, et al. (2003). Estrogen increases locomotor activity in mice through estrogen receptor alpha: Specificity for the type of activity. Endocrinology. 144:230–9. Portaluppi F, Touitou Y, Smolensky MH. (2008). Ethical and methodological standards for laboratory and medical biological rhythm research. Chronobiol Int. 25:999–1016. Vida B, Hrabovszky E, Kalamatianos T, et al. (2008). Oestrogen receptor alpha and beta immunoreactive cells in the suprachiasmatic nucleus of mice: Distribution, sex differences and regulation by gonadal hormones. J Neuroendocrinol. 20:1270–7. Wollnik F, Turek FW. (1988). Estrous correlated modulations of circadian and ultradian wheel-running activity rhythms in LEW/Ztm rats. Physiol Behav. 43:389–96. Zucker I, Fitzgerald KM, Morin LP. (1980). Sex differentiation of the circadian system in the golden hamster. Am J Physiol (Regul Integr Comp Physiol). 238:R97–101.
Chronobiology International
Copyright of Chronobiology International: The Journal of Biological & Medical Rhythm Research is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
