Cryptochrome (Cry) 1 and 2 are essential for circadian rhythm generation, not only in the suprachiasmatic nucleus, the site of the mammalian master circadian clock, but also in peripheral organs throughout the body. CRY is also known as a repressor of arylalkylamine-N-acetyltransferase (Aanat) transcription; therefore, Cry deficiency is expected to induce constantly high pineal melatonin content. Nevertheless, we previously found that the content was consistently low in melatonin-proficient Cry1 and Cry2 double-deficient mice (Cry1−/−/Cry2−/−) on C3H background. This study aims to clarify the mechanism underlying this discrepancy. In the Cry1−/−/Cry2−/− pineal, expression levels of Aanat and clock gene Per1 were consistently high with no circadian fluctuation on the first day in constant darkness, demonstrating that CRY acts in vivo as a repressor of the pineal circadian clock and AANAT. In contrast, the enzyme activity and protein levels of AANAT remained low throughout the day, supporting our previous observation of continuously low melatonin. Thus, effects of Cry deficiency on the responses of β-adrenergic receptors were examined in cultured pineal glands. Isoproterenol, a β-adrenergic stimulant, significantly increased melatonin content, although the increase was smaller in Cry1−/−/Cry2−/− than in WT mice, during both the day and night. However, the increase in cAMP in response to forskolin was similar in both genotypes, indicating that CRY deficiency does not affect the pathway downstream of the β-adrenergic receptor. These results suggest that a lack of circadian adrenergic input due to CRY deficiency decreases β-receptor activity and cAMP levels, resulting in consistently low AANAT levels despite abundant Aanat mRNA.
Pineal melatonin synthesis in mammals is regulated by the circadian pacemaker in the suprachiasmatic nucleus (SCN) of the hypothalamus. The circadian pacemaker is entrained to the environmental light–dark cycle by photic signals via the retinohypothalamic tract (Rusak & Zucker 1979) and controls melatonin synthesis in the pineal gland through the sympathetic nerve, which releases noradrenalin at night (Larsen et al. 1998). Elimination of neural input to the pineal gland demonstrated an importance of this neural pathway not only for the nocturnal melatonin synthesis (Deguchi & Axelrod 1972) but also for day/night changes in transcription of thousands of genes (Hartley et al. 2015). Activation of the β1-adrenergic receptor in pinealocytes is known to be a critical step to increase intracellular cAMP. Increasing intracellular cAMP leads to a rapid activation of arylalkylamine N-acetyltransferase (AANAT), the rate-limiting enzyme in melatonin synthesis (Sugden et al. 1985, Klein 2007) by inhibiting proteasomal proteolysis (Ganguly et al. 2001) as well as by activating transcription level (Ganguly et al. 2005).
The molecular machinery generating the mammalian intracellular circadian rhythm is a transcription/translation feedback loop (core loop) involving the clock genes Period (Per) 1, Per2, Cryptochrome (Cry) 1, Cry2, Clock and Bmal1 and their protein products (Reppert & Weaver 2002). The promoter region of the Aanat gene contains an E-box element that is activated by CLOCK and BMAL1 (Chen & Baler 2000). Therefore, it is likely that pineal melatonin synthesis is also under the control of the core loop. Mutation or loss of these canonical clock genes alters not only circadian behavioral rhythms but also pineal melatonin synthesis in melatonin-proficient laboratory mice (Kennaway et al. 2003, Christ et al. 2010, Yamanaka et al. 2010). In Cry1 and Cry2 double-deficient (Cry1−/−/Cry2−/−) mice, expression of the clock genes Per1 and Per2 in the SCN and peripheral tissues are elevated and lose circadian rhythmicity under a light–dark (LD) cycle or constant darkness (DD) (Okamura et al. 1999, Van der Horst et al. 1999). These results are consistent with the core molecular loop model, since a lack of CRY, a repressor of Per transcription, results in the elevation of transcription (Reppert & Weaver 2002). In addition, a circadian rhythm has been observed in Cry1 and Cry2 expression in the rat pineal gland under a normal LD cycle and constant darkness (Nakamura et al. 2001, Simonneaux et al. 2004), suggesting the functioning of pineal peripheral clock. We previously examined the circadian profile of pineal melatonin content in Cry1−/−/Cry2−/− mice on a melatonin-proficient C3H background and found that the content was constantly low, without nocturnal increase (Yamanaka et al. 2010). The mechanism underlying a lack of nocturnal melatonin synthesis in Cry1−/−/Cry2−/− mice remains unexplored. Furthermore, this result seems to contradict the core molecular loop model wherein Aanat mRNA expression and melatonin levels are expected to be high throughout the day. Thus, Cry1−/−/Cry2−/− mice provide an ideal model for identifying the site of circadian regulation by the master clock in the SCN and the mechanism of transcriptional regulation of Aanat in the peripheral clock in the pineal gland. In the current study, we aim to identify the site at which CRY is critically involved in the circadian melatonin synthesis pathway and to determine whether the core molecular loop regulates Aanat expression.
Materials and methods
Animals and housing
Cry1−/−/Cry2−/− mice of C3H background (Masuki et al. 2005, Tanida et al. 2007) were the kind gift of Dr. Todo. The mice were maintained at the Hokkaido University Graduate School of Medicine as described previously (Yamanaka et al. 2010). Adult Cry1−/−/Cry2−/− (n = 324) and WT (C3H) (n = 283) control mice of both sexes were used for the present experiments. Mice were born and reared in animal quarters, where environmental conditions were controlled (lights on from 0600 to 1800 h; light intensity at the bottom of cage, approximately 100 lux; humidity, 50–60%). Food and water were available ad libitum. The present study was performed in compliance with the rules and regulations established by the Animal Care and Use Committee of Hokkaido University with the permission of the Animal Research Committee of Hokkaido University (approval no.13-0064).
Sampling of pineal glands
Pineal glands were sampled from 3- to 6-month-old WT (n = 40) and Cry1−/−/Cry2−/− (n = 36) mice under either LD or the first cycle after releasing to DD. Time of lights on was defined as Zeitgeber time 0 (ZT0) and that of lights-off as ZT12. The mice were killed by cervical dislocation without anesthesia at four time points (ZT 6, 10, 18, and 22; n = 4–5 per group) under LD or DD, and the pineal gland was collected under a dissecting microscope. Collection of the pineal gland during the dark phase was performed after cervical dislocation and eye enucleation under dim red light (<0.1 lux, <30 s).
Pineal glands were quickly removed at the indicated time and placed in 200 μL of ice-cold assay buffer (0.1 M Tris–HCl, pH 7.2). The tissue was homogenized by repeated freezing and thawing in liquid nitrogen for three cycles. The pineal homogenates were centrifuged for 15 min at 21,500 g, and the supernatant was stored at −30°C. The melatonin content of each sample was determined by radioimmunoassay (RIA) as previously described (Yamanaka et al. 2010). Each sample was assayed using 200 μL of supernatant. The minimum detection level for the melatonin concentration was 1.56 pg/tube. Intra- and inter-assay coefficient of variances were 5.6 and 6.2%, respectively.
Quantitative real-time PCR analysis of pineal Aanat and Per1 mRNA
Period 1 (Per 1) and Aanat mRNAs in the pineal gland were measured by quantitative real-time reverse transcription-PCR (qRT-PCR) as previously reported (Nishide et al. 2012) with some modifications. The pineal glands of the WT (n = 30) and Cry1−/−/Cry2−/− (n = 30) mice were quickly put in ice-cold tissue storage regent (RNA later, R0901, Sigma-Aldrich) at six time points (ZT 2, 6, 10, 14, 18, and 22; n = 5 per group) under LD or DD and stored at −80°C until mRNA extraction. Total RNA was extracted using RNeasy Micro Kit (Qiagen), and the concentration was measured using a NanoDrop 2000 spectrophotometer (Thermo Scientific). cDNA was synthesized using SuperScriptIII First-Strand Synthesis SuperMix (Invitrogen) according to the manufacturer’s instruction. qRT-PCR was performed with DyNAmo HS SYBR Green qPCR Kit (New England BioLabs, Ipswich, MA, USA) and a real-time thermal cycler (ABI PRISM 7700, Applied Biosystems) using the following primers: mGAPDH forward: 5′TGCGACTTCAACAGCAACTC3′; mGAPDH reverse: 5′ATGTAGGCCATGAGGTCCAC3′; mAanat forward: 5′CTCCAGCCCATCCAACTAG3′; mAanat reverse: 5′TGCACAGTTCAGAAGGCAAG3′; mPer1 forward: 5′CGTCCTACCTCCTTTATCCAGA3′; mPer1 reverse: 5′TGTTTGCATCAGTGTCATCAGC3′.
Measurement of pineal AANAT enzyme activity
The AANAT enzyme activity was assayed according to the method reported by Deguchi & Axelrod (1972) with slight modification. Pineal glands of the WT (n = 30) and Cry1−/−/Cry2−/− (n = 31) were quickly removed at one of six times of the day in DD (ZT2, 6, 10, 14, 18, and 22; n = 5–6 per group). Each pineal gland was homogenized in chilled 0.05 M phosphate buffer (50 μL, pH. 6.5) and stored at −30°C until the assay. On the day of assay, 10 μL each of 0.01 M Tryptamine-HCL and 3.4 nM (1-14C) acetyl coenzyme A (PerkinElmer) was added to 50 μL of homogenate in a small glass tube. After incubating at 37°C for 10 min, the reaction was stopped by the addition of 0.5 mL of 0.5 M borate buffer (pH 10.0). The homogenate was transferred into a glass tube containing 6 mL of toluene-isoamyl alcohol (97:3) and stirred using a vortex mixer. After centrifugation at 3000 g for 10 min, 5 mL of the organic phase was transferred into a scintillation vial containing 5 mL of scintillation cocktail (Aquazol2, PerkinElmer), and radioactivity was measured.
Immunoblotting of pineal AANAT protein
The AANAT protein concentration was determined by immunoblotting using the method reported elsewhere (Nishide et al. 2012) with some modifications. Briefly, the pineal glands were collected at four time points (ZT6, 10, 18 and 22) in DD. The pineal glands were pooled (WT, n = 8; Cry1−/−/Cry2−/−, n = 12) in tissue protein extraction regent (T-PER, Thermo Scientific) and homogenized by sonication followed by freezing and thawing four times in liquid nitrogen. The homogenized samples were frozen in liquid nitrogen and stored at −80°C until assay. After centrifuging at 21,500 g for 15 min at 4°C, the supernatant protein concentration was determined using the BCA Protein Assay Kit (Thermo Fisher Scientific). The protein extracts (50 μg/lane) were separated using SDS- PAGE. A sample of rat pineal gland (2.5 μg/lane) collected at ZT22 (expected peak of pineal AANAT protein) was used as an inter-assay standard. The sample was electrophoresed in NuPAGE Novex 4–12% Bis-Tris gel (Invitrogen) according to the manufacturer’s protocol and transferred onto a Hybond-P PVDF Membrane (GE Healthcare). The membrane was incubated with a blocking reagent (Block Ace, DS Pharma, Osaka, Japan) overnight at 4°C, followed by incubation with goat anti-AANAT (P-20) antibody (Santa Cruz Biotechnology) diluted 1:500 with immunoreaction enhancer solution (Can Get Signals, Toyobo, Osaka, Japan) for 1 h at room temperature, and then with HRP-conjugated donkey anti-goat IgG (Promega) diluted 1:500,000 in Block Ace buffer (Pharma Biomedical, Tokyo, Japan) for 1 h at room temperature. Immunoreactive bands were detected using Super Signal, the enhanced chemiluminescent substrate for HRP (Thermo Fisher Scientific) and Hyperfilm ECL (GE Healthcare). The membrane was stripped using Restore Western Blot Stripping Buffer (Thermo Fisher Scientific) and reacted with mouse anti β-ACTIN IgG (Sigma-Aldrich) and subsequently with HRP-conjugated goat anti-mouse IgG (Thermo Fisher Scientific). The bands of interest were quantified using ImageJ software (National Institutes of Health). The band intensities were calculated after subtracting the lane background from the raw data. The immunoblotting was repeated three times.
Pineal gland organ culture and isoproterenol stimulation
Pineal glands were collected from mice housed in LD at ZT3 (3 h after lights on) (WT, n = 24; Cry1−/−/Cry2−/−, n = 23) mice and ZT15 (3 h after lights-off) (WT, n = 24; Cry1−/−/Cry2−/−, n = 20). Each pineal gland was placed on a culture membrane (Millicell PICM01250, Merck Millipore) placed into a well of a 24-well plate and incubated with 200 μL of Dulbecco’s Modified Eagle’s Medium (DMEM; GIBCO/Thermo Fischer Scientific) at 37°C in a humidified atmosphere with 5% CO2 for 1 h. After this pre-incubation, the culture medium was replaced with DMEM containing 200 μL of the β-adrenergic agonist isoproterenol (ISO; final concentration, 2 μM) or vehicle (water) at projected ZT4 and 16. The dose was reported to increase Aanat mRNA and melatonin levels in cultured pineal gland most efficiently (Barbosa et al. 2008, Wongchitrat et al. 2011). After incubation for 2 h, each pineal gland was rinsed three times with 0.1 M Tris–HCl buffer (pH 7.2) and homogenized in 200 μL of Tris–HCl buffer by four cycles of freezing in liquid nitrogen and thawing. Two hours of incubation time was selected according to our preliminary experiment and previous reports (Santana et al. 1988, Fukuhara et al. 2005, Wongchitrat et al. 2011). The pineal homogenates were centrifuged at 21,500 g for 15 min at 4°C and the supernatants were stored at −80°C until measurement. The melatonin content of each sample was determined by the RIA.
AANAT enzyme activity after stimulation with isoproterenol and norepinephrine
Pineal glands of the WT (n = 18; male n = 9, female n = 9) and Cry1−/−/Cry2−/− (n = 19; male n = 9, female n = 10) mice were collected from mice housed in LD at ZT3. After preincubating for 1 h with DMEM, they were transferred to fresh medium containing either 2 μM of ISO or 10 μM of norepinephrine (NE) and ascorbic acid (0.01 mg/mL) at projected ZT4. After the treatment for 2 h, the pineal glands were rinsed three times with 0.05 M of phosphate buffer (PB, pH6.5) and homogenized in 50 μL of PB by four cycles of freezing in liquid nitrogen and thawing. The pineal homogenates were stored at −80°C until measurement. The AANAT activity was determined by the RIA.
Cyclic AMP content in the pineal gland after treatment with forskolin
Pineal glands of the WT (n = 21; male n = 9, female n = 12) and Cry1−/−/Cry2−/− (n = 21; male n = 10, female n = 11) mice were sampled from mice kept in LD at ZT3 and individually incubated with 200 μL of DMEM at 37°C and 5% CO2 for 30 min. After this pre-incubation, 200 μL of forskolin (FSK, final concentration, 50 μM) or vehicle (DMEM) was added to the culture medium at projected ZT3.5 and incubated at 37°C and 5% CO2 for 1 h. The dose was reported to increase cAMP level and melatonin synthesis in cultured pineal gland (Santana et al. 1988, 2001). The cultured pineal glands were rinsed by transferring them three times to new culture wells containing 200 μL of cold 0.1 M Tris–HCl buffer (pH 7.2). The pineal gland was placed in a tube containing 200 μL of cold 0.1 M Tris–HCl buffer (pH 7.2) and homogenized by four cycles of freezing in liquid nitrogen and thawing. The pineal homogenates were stored at −80°C until assayed. The cAMP level in the pineal homogenate was measured using the cAMP-Glo Max assay kit (Promega) by monitoring bioluminescence with luminometer.
Data are reported as the mean ± s.e.m. A one-way ANOVA (main effect of time) followed by a post hoc Tukey–Kramer test was used to determine the time-dependent difference for each genotype. A two-way ANOVA (main effect of genotype and interaction between genotype and time) followed by a post hoc unpaired t-test was used to compare the data between the two genotypes. P < 0.05 was considered statistically significant.
Circadian rhythms in pineal melatonin content in Cry1−/−/Cry2−/− mice
The circadian profile of pineal melatonin content under LD and DD differed significantly between WT and Cry1−/−/Cry2−/− mice (P < 0.001), with a significant interaction between genotype and time (two-way ANOVA test) (Fig. 1). Under both LD and DD conditions, WT mice exhibited a robust circadian variation, with a significant difference found between time points (P < 0.001; main effect time, one-way ANOVA), with the peak level at ZT22 (178 ± 35 pg/pineal in LD, 238 ± 44 pg/pineal in DD, n = 5). In contrast, Cry1−/−/Cry2−/− mice exhibited no significant circadian variation in pineal melatonin content (P = 0.12 in LD; P = 0.23 in DD; main effect time, one-way ANOVA test, n = 4–5). Under LD, the pineal melatonin content was significantly lower in Cry1−/−/Cry2−/− mice than in WT mice in the light phase (P < 0.01 at ZT6; P < 0.05 at ZT10; two-way ANOVA test with post hoc unpaired t-test) as well as in the dark phase (P < 0.01 at ZT18 and ZT22). Under DD, the melatonin content was significantly lower in Cry1−/−/Cry2−/− mice than in WT mice at ZT18 (P < 0.05) and ZT22 (P < 0.001).
Circadian rhythms in Per 1 and Aanat mRNA expression in the pineal lacking CRYs
Aanat is reported to have an E-box enhancer in the first intron, the function of which in the circadian regulation of Aanat expression is still unknown (Baler et al. 1999, Klein 2007). To examine the functional role of this E-box in Aanat expression, we determined the circadian profile of Per1 and Aanat mRNA in Cry1−/−/Cry2−/− mice. In the WT mice, both Per1 (Fig. 2A) and Aanat (Fig. 2B) mRNA levels in the pineal gland showed a significant rhythmicity (Per1, P < 0.05; Aanat, P < 0.05; main effect time, one-way ANOVA test, n = 5) and were significantly elevated in the subjective night as compared with those in the mid-subjective day (P < 0.05, post-hoc Tukey–Kramer test). The peak of Per1 expression was detected at ZT14, while that of Aanat was between ZT18 and ZT22. In Cry1−/−/Cry2−/− mice, the mRNA levels of both genes were constant throughout the day, exhibiting no significant rhythmicity (P = 0.23; P = 0.81; main effect time one-way ANOVA, n = 5). Compared with WT, the Per1 mRNA levels in Cry1−/−/Cry2−/− mice were significantly higher throughout the day except at ZT14, when the level was lower than that of wild type (P < 0.01). Despite the findings of low pineal melatonin levels, the Aanat mRNA level in Cry1−/−/Cry2−/− mice was consistently high at the circadian peak level of WT mice with statistical significance during the mid-subjective day (ZT6 and 10) and early subjective night (ZT14) (P < 0.01).
Circadian rhythm in pineal AANAT enzyme activity and protein concentration
To identify at which stage of melatonin synthesis the dissociation takes place between the consistently high Aanat mRNA and low melatonin levels, we examined the circadian rhythms in the levels of AANAT protein and enzyme activity. In WT mice, the AANAT protein concentration in the pineal gland exhibited a circadian rhythm with a peak level at ZT22 (P < 0.001, one-way ANOVA test). Whereas in Cry1−/−/Cry2−/− mice, AANAT protein did not exhibit a significant rhythm across the four circadian phases examined (Fig. 3), instead remaining at intermediate levels of the circadian amplitude in WT throughout the day. To apply the same amount of total protein for electrophoresis, we pooled the extracts from 8 and 12 pineal glands from the WT and Cry1−/−/Cry2−/− mice, respectively. The results indicate that the total protein content is lower in Cry1−/−/Cry2−/− pineal glands. Similarly, the AANAT enzyme activity in WT mice showed a significant rhythmicity (P < 0.001; main effect time, one-way ANOVA) (Fig. 4), with the peak reached at the late subjective night (ZT22), increased from the low daytime level (P < 0.05, ZT2, ZT6, ZT10 vs ZT22; post-hoc Tukey–Kramer test). In contrast, the AANAT enzyme activity in Cry1−/−/Cry2−/− mice did not show significant rhythmicity across the circadian phases (P = 0.07). Circadian patterns differed significantly between the two genotypes (P < 0.001, two-way ANOVA), with a significant difference at ZT 22 (P = 0.002, post-hoc unpaired t-test).
Effects of β-adrenergic stimulation on melatonin synthesis in the pineal lacking CRYs
We then sought to determine the mechanism underlying the discrepancy between Aanat mRNA and protein/enzyme activity levels in Cry1−/−/Cry2−/− pineal glands. The increase in intracellular cAMP in response to β-adrenergic stimulation via neural inputs from the SCN is known to be critical for stabilizing the AANAT protein and increasing melatonin synthesis (Ganguly et al. 2001). Therefore, we asked whether pinealocytes lacking CRYs can respond to β-adrenergic stimulation to increase melatonin synthesis. To determine the day–night difference, pineal glands of WT and Cry1−/−/Cry2−/− mice were sampled at two time points, ZT3 and ZT15, and stimulated with isoproterenol (ISO, 2 μM) for 2 h after a pre-incubation of 1 h. In wild-type pineal glands, ISO treatment significantly increased melatonin content compared with those of vehicle-treated controls, both at ZT3 (vehicle vs ISO, 174 ± 18 pg/gland (n = 12) vs 544 ± 104 pg/gland (n = 12); P = 0.002, two-way ANOVA test with post hoc unpaired t-test) and ZT15 (vehicle vs ISO, 60 ± 14 pg/gland (n = 12) vs 393 ± 91 pg/gland (n = 11), P = 0.002). In the Cry1−/−/Cry2−/− pineal gland, the melatonin content was also significantly elevated by ISO treatment, both at ZT3 (vehicle vs ISO, 34 ± 4 pg/gland (n = 13) vs 124 ± 29 pg/gland (n = 10); P = 0.004) and at ZT15 (vehicle vs ISO, 32 ± 3 pg/gland (n = 10) vs 65 ± 14 pg/gland (n = 10); P = 0.045) compared to the respective vehicle control. However, the response to ISO was markedly lower in the Cry1−/−/Cry2−/− than that in the wild type, both at ZT3 and ZT15 (ZT3, P < 0.001; ZT15, P = 0.004) (Fig. 5A). Three-way ANOVA revealed a significant interaction between the genotype (wild-type and Cry1−/−/Cry2−/−) and the stimulation (vehicle and ISO) (P = 0.003), but no significant interaction the genotype and the stimulation time).
Effect of ISO and NE on the AANAT enzyme activity of the pineal lacking CRYs
To explore the mechanism of the different response to ISO in the wild-type and Cry1−/−/Cry2−/− mice, the pineal glands were sampled at ZT3, and the AANAT activity was measured 2 h after the stimulation with 2 μM of ISO or 10 μM of NE at projected ZT4. We examined only at ZT3 because no stimulation time effect was detected by ISO in either genotype. The AANAT activity after stimulation with ISO tended to increase the AANAT activity on average as compared with the vehicle control in WT and Cry1−/−/Cry2−/− mice, though it did not reach statistical significance (Fig. 5B). After the stimulation with NE, the significant increases of the AANAT activity were observed in both genotypes at similar extent (Fig. 5C). Two-way ANOVA test revealed significant effect of genotype (WT and Cry1−/−/Cry2−/−) (P = 0.011) and the stimulation (Vehicle vs NE) (P = 0.002), but no significant interaction between the genotype and stimulation.
Effect of FSK stimulation on the pineal cAMP level
To assess whether the decreased melatonin synthesis in the Cry1−/−/Cry2−/− mice is caused by a decrease in intracellular cAMP or defects in the downstreams, the effect of FSK (5 μM and 50 μM) on cAMP levels were examined in cultured pineal glands sampled at ZT3. In both genotypes, FSK stimulation significantly increased the cAMP level in the pineal gland as compared to vehicle-treated controls (vehicle, 5 μM FSK and 50 μM FSK, 205 ± 10 nM (n = 7), 441 ± 48 nM (n = 7, P < 0.01) and 540 ± 33 nM (n = 7, P < 0.01) for WT; 125 ± 20 nM (n = 7), 380 ± 45 nM (n = 7, P < 0.01) and 397 ± 66 nM (n = 7, P < 0.01) for Cry1−/−/Cry2−/−, respectively) (Fig. 6). The Cry1−/−/Cry2−/− pineal exhibited FSK-induced increase in cAMP by FSK to a similar extent as WT. Thus, no genotype difference was detected in the response to FSK. Two-way ANOVA revealed a significant effect of stimulation (vehicle and FSK) on the cAMP level in the pineal gland (P < 0.001), but there was no significant interaction between genotype (WT and Cry1−/−/Cry2−/−) and stimulation. While no significant dose–response was observed in the two doses examined, cAMP tended to increase with a higher dose in both genotypes.
Our findings demonstrate that a loss of circadian melatonin rhythm in Cry1−/−/Cry2−/− mice is mainly caused by the loss of rhythmic noradrenergic signals from the circadian pacemaker in the SCN and partly caused by a decreased response to β-adrenergic signals. The pineal gland of Cry1−/−/Cry2−/− mice showed consistently high Aanat mRNA expression and low AANAT protein and enzyme activity throughout the day. The discrepancy between these findings is explained by the lack of rhythmic β-adrenergic signals (Klein & Moore 1979, Klein et al. 1983a) by loss of neural input from the SCN, because cAMP-dependent phosphorylation of AANAT leads binding with 14-3-3 protein, which activates the AANAT activity (Ganguly et al. 2005) and protects from proteasomal degradation of AANAT (Ganguly et al. 2001). In addition, the consistently high Aanat mRNA levels observed in the Cry1−/−/Cry2−/− pineal glands indicate that the E-box enhancer plays a functional role in the regulation of Aanat expression.
Cry genes are core components of the autoregulatory transcription/translation feedback loop (core loop) in mammalian circadian clocks. Cry1−/−/Cry2−/− mice demonstrated circadian rhythm in behavior under LD but immediately became arrhythmic after the transfer to DD (Van der Horst et al. 1999). Furthermore, expression of the clock genes Per1 and Per2 and multiunit neural activity in the SCN did not show circadian rhythms under DD (Okamura et al. 1999, Vitaterna et al. 1999, Albus et al. 2002). Nevertheless, we recently observed that individual SCN cells of Cry1−/−/Cry2−/− mice exhibit circadian rhythms in clock gene expression and that the arrhythmicity in clock gene expression at the SCN tissue level in adults is caused by desynchronization of individual cellular rhythms in the course of postnatal development (Ono et al. 2013). However, in the neonatal period, circadian rhythms in single SCN cells synchronize mainly via neuropeptides released in the SCN to exhibit robust circadian rhythms at the tissue level (Ono et al. 2016). If the low level of melatonin production in Cry1−/−/Cry2−/− mice is due to the absence of melatonin synthesis via rhythmic neuronal signals from the SCN, the pineal glands of Cry1−/−/Cry2−/− mice should exhibit increased melatonin production in response to ISO. However, we observed here that ISO-induced melatonin production occurred in both pineal glands sampled during subjective day (ZT3) and night (ZT15), suggesting that the Cry1−/−/Cry2−/− pineal glands can synthesize melatonin. Since the WT pineal glands also responded to ISO at both time points, this response to noradrenergic stimulation appears to be ungated. The ungated response has also been reported in the Aanat and Per1 mRNA levels of cultured pineal gland after the treatment with ISO (Fukuhara et al. 2005, Wongchitrat et al. 2011). Taken together, these findings indicate that the loss of circadian rhythm of pineal melatonin expression in Cry1−/−/Cry2−/− mice is caused by a loss of rhythmic output signals from the SCN to the pineal gland.
In rodents, pineal AANAT enzyme activity dramatically increases at night (Deguchi & Axelrod 1972) in response to noradrenalin released under the control of the circadian clock in the SCN (Perreau-Lenz et al. 2003). Upon binding of noradrenalin to the pineal β-adrenergic receptor, intracellular cAMP content increases (Sugden et al. 1985, Ho et al. 1988, 1989), which leads to an activation of adenylate cyclase. This effect is enhanced by α1 receptors via Ca2+ and phosphatidyl inositol activation of protein kinase C (Klein et al. 1983b, Ho & Klein 1987, Sugden & Klein 1988, Ho et al. 1989), thereby activating Aanat transcription via cAMP-responsive elements in the promoter and 1st intron of Aanat (Humphries et al. 2007). Stimulation of α1- and β-adrenergic receptors results in a marked increase in both cAMP and cGMP accumulation in pinealocytes (Vanecek et al. 1986, Sugden & Klein 1987). The effects of cGMP analogs on pineal physiology are usually weaker than the actions of cAMP analogs; however, it remains to be studied whether or not the cGMP accumulation is involved in the melatonin synthesis of the pineal gland of Cry1−/−/Cry2−/− mice.
Another possible mechanism underlying the observed circadian rhythms in Aanat expression is the E-box–mediated increase in a clock-controlled gene in the pineal peripheral clock (Chen & Baler 2000). The Aanat gene also has a canonical E-box (CACGTG) and E-box–like (CACATG) sequences in the 1st intron (Humphries et al. 2007); however, the role of the peripheral clock in the pineal gland has still been unclear. Previously, transcription of retinal but not pineal Aanat was shown to be increased by CLOCK and BMAL1 (Chen & Baler 2000), and the intronic E-box was reported to be involved not in the circadian regulation of transcription but in the tissue-specific silencing of Aanat expression (Humphries et al. 2007). With respect to the rhythmic pattern of AANAT transcription, pinealocyte clock of zebrafish and chicken is reported to drive AANAT directly via E-box–mediated transcription increase (Chong et al. 2000, Appelbaum et al. 2004). In the present study, a lack of CRY, the negative element of the core molecular clock, resulted in a continual increase in Aanat expression, similar to Per1 expression, indicating for the first time that the intronic E-box of the mouse Aanat gene has a functional role in vivo. Although no reports are available examining double labeling of CRY and AANAT so far, a continual increase of AANAT mRNA in the present results suggests that CRYs are expressed in the same cell as AANAT.
In addition to promoting Aanat transcription, an increase in cAMP also has a significant effect on nocturnal increases in melatonin synthesis that occurs via post-translational modifications of AANAT (Klein 2007). The AANAT protein is phosphorylated in a cAMP-dependent manner. The phosphorylation of AANAT leads to the formation of its complex with the 14-3-3 protein (pAANAT/14-3-3 complex), which protects AANAT from degradation (Ganguly et al. 2001) and elevates its enzymatic activity (Ganguly et al. 2005). Nocturnal melatonin synthesis critically depends on intracellular cAMP. Thus, melatonin synthesis is strictly influenced by the intracellular cAMP level, which is under the regulation of neural inputs from the circadian pacemaker in the SCN. According to our proposed mechanism, the decreased melatonin synthesis in Cry1−/−/Cry2−/− mice might be caused by the decrease in the intracellular cAMP level at night caused by a lack of coherent rhythm output from the SCN.
Although CRY deficiency resulted in a lack of circadian output from the SCN, we found that the pineal gland of Cry1−/−/Cry2−/− mice was capable of producing melatonin in response to noradrenergic signals. The response to ISO was reduced in Cry1−/−/Cry2−/−as compared to that of WT mice. Nevertheless, AANAT activity of Cry1−/−/Cry2−/−pineal increased by NE stimulation to the similar extent to the WT, suggesting significant role of α1 adrenergic receptor in melatonin synthesis. The same amount of ISO significantly increased melatonin content but the increase in AANAT activity was not statistically significant. The reason of the discrepancy is not known, but the difference in assay system might be involved. The possible mechanisms underlying this observation include the desynchronization of cellular rhythms in the pineal caused by a lack of entraining signals from the SCN and, alternatively, a decreased number and/or sensitivity of β-receptors on the pinealocytes. Nevertheless, the pathways for circadian melatonin synthesis downstream of β-receptors are intact in Cry1−/−/Cry2−/− pineal glands, indicating that CRYs are not involved in intracellular melatonin synthesis pathways, since NE and FSK increased AANAT activity and cAMP levels, respectively, in cultured pineal glands of Cry1−/−/Cry2−/− similar to those in WT mice (Fig. 6). Together, these results indicate that the pineal gland of Cry1−/−/Cry2−/− mice can induce melatonin synthesis in response to increased cAMP levels following NE release from the sympathetic nerve terminals. Previously, Deguchi & Axelrod (1972) reported the super-sensitivity in vitro to ISO stimulation in the denervated pineal gland and pineals from rats exposed to continuous lighting. Whereas in the present study, AANAT activity in Cry1−/−/Cry2−/− pineal was continuously low throughout the day but increased by the stimulation of NE to the similar level of the WT. Therefore, upregulation of adrenergic receptors was prevented, somehow, in the pineals lacking CRYs. CRY deficiency may affect the availability of receptor site and/or number of α1- and β-adrenergic receptor.
The limitation of the present study is that we used conventional Cry-knockout mice. CRY deficiency may have long-term effects on the development of adrenergic innervation of the pineal gland, which needs to be studied using time-dependent Cry gene targeting in future. Furthermore, it has been reported that adrenergic stimulation turns on transcription of nearly 600 genes via the same signaling pathway that induces Aanat gene (Bailey et al. 2009, Ho & Chik 2010). Genome-scale analysis might reveal a novel role of CRY in the pineal physiology.
In conclusion, the present data demonstrate that continuously low pineal melatonin in Cry1−/−/Cry2−/− mice is associated with continuously high Aanat mRNA levels and low AANAT protein and enzyme activity. E-box elements of the Aanat gene are involved in transcriptional regulation in the pineal gland in vivo. Finally, Cry genes are a necessary requirement for generating the circadian rhythm in pineal melatonin synthesis. CRYs are critically involved in the intracellular cAMP increase in response to nocturnal noradrenalin stimulation, which is disrupted in the Cry1−/−/Cry2−/− SCN due to desynchronization of cellular rhythms (Ono et al. 2013). CRYs are not necessary for melatonin synthesis via intracellular pathways.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
This work was supported in part by the Creation of Innovation Centers for Advanced Interdisciplinary Research Areas Program, Ministry of Education, Culture, Sports, Science and Technology, Japan and the Japan Society for the Promotion of Science Grants-in-Aid for Scientific Research (KAKENHI) (no. 15H04679 and no. 16K01723).
Author contribution statement
Y Yamanaka and S Honma designed the experiments; Y Yamanaka, Y Yamada and S Honma performed the experiments; Y Yamanaka, Y Yamada and S Honma analyzed the data; and Y Yamanaka, K Honma and S Honma wrote the paper.
The authors appreciate the generous supply of CRY-deficient mice on a C3H background from Prof. Todo of Osaka University, and melatonin antibody from Prof. K Kawashima, Kyoritsu Pharmaceutical College, Tokyo, Japan.
AppelbaumLToyamaRDawidIBKleinDCBalerRGothilfY 2004 Zebrafish serotonin-N-acetyltransferase-2 gene regulation: pineal-restrictive downstream module contains a functional E-box and three photoreceptor conserved elements. Molecular Endocrinology 18 1210–1221. (https://doi.org/10.1210/me.2003-0439)
BaileyMJCoonSLCarterDAHumphriesAKimJSShiQGaildratPMorinFGangulySHogeneschJB 2009 Night/day changes in pineal expression of >600 genes: central role of adrenergic/cAMP signaling. Journal of Biological Chemistry 284 7606–7622. (https://doi.org/10.1074/jbc.M808394200)
GangulySGastelJAWellerJLSchwartzCJaffeHNamboodiriMACoonSLHickmanABRollagMObsilT 2001 Role of a pineal cAMP-operated arylalkylamine N-acetyltransferase/14-3-3-binding switch in melatonin synthesis. PNAS 98 8083–8088. (https://doi.org/10.1073/pnas.141118798)
HoAKKleinDC 1987Activation of alpha 1-adrenoceptors, protein kinase C, or treatment with intracellular free Ca2+ elevating agents increases pineal phospholipase A2 activity. Evidence that protein kinase C may participate in Ca2+-dependent alpha 1-adrenergic stimulation of pineal phospholipase A2 activity. Journal of Biological Chemistry 262 11764–11770.
HoAKChikCLKleinDC 1988 Permissive role of calcium in alpha 1-adrenergic stimulation of pineal phosphatidylinositol phosphodiesterase (phospholipase C) activity. Journal of Pineal Research 5 553–564. (https://doi.org/10.1111/j.1600-079X.1988.tb00798.x)
KennawayDJVoultsiosAVarcoeTJMoyerRW 2003 Melatonin and activity rhythm responses to light pulses in mice with the Clock mutation. American Journal of Physiology. Regulatory Integrative and Comparative Physiology 284 R1231–R1240. (https://doi.org/10.1152/ajpregu.00697.2002)
KleinDCSmootRWellerJLHigaSMarkeySPCreedGJJacobowitzDM 1983a Lesions of the paraventricular nucleus area of the hypothalamus disrupt the suprachiasmatic leads to spinal cord circuit in the melatonin rhythm generating system. Brain Research Bulletin 10 647–652. (https://doi.org/10.1016/0361-9230(83)90033-3)
NishideSYOnoDYamadaYHonmaSHonmaK 2012 De novo synthesis of PERIOD initiates circadian oscillation in cultured mouse suprachiasmatic nucleus after prolonged inhibition of protein synthesis by cycloheximide. European Journal of Neuroscience 35 291–299. (https://doi.org/10.1111/j.1460-9568.2011.07952.x)
Perreau-LenzSKalsbeekAGaridouMLWortelJvan der VlietJvan HeijningenCSimonneauxVPévetPBuijsRM 2003 Suprachiasmatic control of melatonin synthesis in rats: inhibitory and stimulatory mechanisms. European Journal of Neuroscience 17 221–228. (https://doi.org/10.1046/j.1460-9568.2003.02442.x)
SantanaCGuerreroJMReiterRJTroianiME 1988 The in vitro activation of cyclic AMP production by either forskolin or isoproterenol in the Syrian hamster pineal during the day is not accompanied by an increase in melatonin production. Biochemical and Biophysical Research Communications 157 930–936. (https://doi.org/10.1016/S0006-291X(88)80963-X)
VanecekJSugdenDWellerJLKleinDC. 1986 See-saw signal processing in pinealocytes involves reciprocal changes in the α1-adrenergic component of the cyclic GMP response and the β-adrenergic component of cyclic AMP response. Journal of Neurochemistry 47 678–686. (https://doi.org/10.1111/j.1471-4159.1986.tb00665.x)