Rhythmic Expression of Adenylyl Cyclase VI Contributes to the Differential Regulation of Serotonin N-Acetyltransferase by Bradykinin in Rat Pineal Glands*

The rhythmic nocturnal production of melatonin in pineal glands is controlled by the periodic release of norepinephrine from the superior cervical ganglion. Norepinephrine binds to the β-adrenergic receptor and stimulates an increase in intracellular cAMP levels, leading to the transcriptional activation of serotonin N-acetyltransferase, which in turn promotes melatonin production. In the present study, we report that bradykinin inhibits basal- and forskolin-stimulated adenylyl cyclase activity, norepinephrine-induced cAMP generation, and N-acetyltransferase expression in a calcium-dependent manner. These effects were blocked by pretreatment with U73122 (a selective phospholipase C inhibitor), and 1,2-bis(o-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid (a Ca2+ chelator), but not pertussis toxin. The calcium ionophore, ionomycin, inhibited isoproterenol-mediated cAMP generation, similar to bradykinin. Interestingly, the inhibitory effect of bradykinin was evident only during the daytime. At midday, bradykinin inhibited the cAMP level by ∼50% but markedly stimulated cAMP production (by ∼50%) at night. Northern blotting and immunoblotting data disclosed circadian expression of calcium-inhibitable adenylyl cyclase type 6. Expression of adenylyl cyclase type 6 was maximal at Zeitgeber Time (ZT) 01 and very low at ZT 13. Our results suggest that bradykinin-induced calcium inhibits melatonin synthesis through the mediation of adenylyl cyclase type 6 expression.

The mammalian pineal gland is a neuroendocrine transducer that rhythmically synthesizes and secretes melatonin, a regulator of circadian rhythm, sleep, mood, reproduction, and aging (1). Melatonin synthesis is modulated by the hypothalamic circadian clock in the suprachiasmatic nucleus as well as light signals through a multisynaptic neuronal pathway projecting from the retina to the suprachiasmatic nucleus of the anterior hypothalamus via the retinohypothalamic tract (2). A dark signal perceived by the retina triggers norepinephrine release from postganglionic neurons originating in the superior cervical ganglion regulated by suprachiasmatic nucleus activation (3,4). Norepinephrine interacts with the ␤-adrenergic receptor on pinealocytes, inducing an increase in pineal cAMP generation. This elevation of cAMP, in turn, stimulates the conversion of tryptophan to melatonin through triggering transcription and activity of serotonin N-acetyltrans-ferase (AANAT, 2 acetyl coenzyme A:arylalkylamine N-acetyltransferase; EC 2.3.1.87), the key enzyme catalyzing the important step of melatonin synthesis (5,6). This ␤-adrenergic receptor-mediated cAMP signaling is further potentiated by stimulation of ␣ 1 -adrenergic receptors through increment of cytosolic Ca 2ϩ and activation of protein kinase C (PKC) (7). Protein kinase C additionally stabilizes the AANAT protein via phosphorylation (8).
Previous studies have suggested that multiple receptors for amino acids, neuropeptides, and biogenic amines other than norepinephrine present on pinealocytes participate in the modulation of melatonin synthesis (9). However, to date, no studies have established the presence of BK receptors on pinealocytes.
BK, a nonapeptide hormone, is generated from the high molecular weight precursor, kininogen, by the proteolytic action of kallikrein and functions as a potent mediator of inflammation, pain, asthma, and hypertension (10,11). The kallikrein-kinin system has been identified in the mammalian nervous system (12,13). Interestingly, kallikrein mRNA and enzyme activity are higher in the rat pineal gland than other brain regions (14,15). Consistent with this finding, immunoelectron microscopy data by Kudo et al. (16) disclosed that kallikrein is localized in perivascular spaces in the rat pineal gland. However, the functional role of the kallikrein-kinin system in rat pineal glands is currently unclear.
A number of earlier investigations reported that BK inhibits ␤-adrenergic receptor-mediated cAMP accumulation in various cell types (17)(18)(19). Moreover, Ca 2ϩ -inhibitable adenylyl cyclase (AC) is involved in the BK-induced inhibition mechanism, because stimulation of the BK receptor leads to increased [Ca 2ϩ ] i . To date, at least nine isoforms of AC have been cloned and characterized in mammals (20). Among these, types V and VI constitute a subfamily defined as Ca 2ϩ -inhibitable.
Here, we report that the treatment of rat pinealocytes with BK results in [Ca 2ϩ ] i increase and diurnal inhibition of pineal AC activity. Inhibition of ␤-adrenergic receptor-mediated cAMP generation and AANAT expression by BK is mediated by direct suppression of Ca 2ϩ -inhibitable AC type VI (AC6). Additionally, the expression patterns of AC6 mRNA and protein follow a diurnal rhythm.

EXPERIMENTAL PROCEDURES
Materials-Fura-2 pentaacetoxymethyl ester (Fura-2/AM) and BAPTA/AM were obtained from Molecular Probes (Eugene, OR).  12) with lights off at 7:00 pm for at least 1 week before the day of the experiment.
Rat Pinealocyte Culture-Pinealocytes were isolated from the pineal glands of male Sprague-Dawley rats (150 g, Hyochang Science) at postnatal week 6. Male Sprague-Dawley rats were decapitated, and pineal glands were washed with ice-cold Locke's solution (154 mM NaCl, 5.6 mM KCl, 10 mM glucose, 2.2 mM CaCl 2 , 1.2 mM MgCl 2 and 5 mM HEPES, adjusted to pH 7.4). Glands were dissected into small sections and treated with 0.1% collagenase solution (2 mg/ml collagenase type V in Locke's solution, Invitrogen) at 37°C for 30 min with gentle shaking. This was followed by treatment with 0.025% trypsin solution (Invitrogen) at 37°C for 15 min, and centrifugation at 180 ϫ g for 5 min. Dispersed pinealocytes were washed three times with Dulbecco's modified Eagle's medium (DMEM, Invitrogen) supplemented with 10% fetal bovine serum and 1% (v/v) antibiotics comprising 5000 units/ml penicillin G (sodium) and 5000 g/ml streptomycin sulfate in 0.85% saline (Invitrogen), pH 7.4. Pinealocytes were placed in coverslips or culture dishes coated with poly-L-lysine, and maintained in the above medium in a humidified atmosphere of 5% CO 2 at 37°C.
Single-cell Ca 2ϩ Imaging-Intracellular Ca 2ϩ levels were measured by fluorescence ratio imaging of the Ca 2ϩ indicator dye, Fura-2. Briefly, pinealocyte-attached coverslips were transferred to a 35-mm culture dish containing prewarmed medium. After 10 min, the medium was replaced with 1.5 ml of the same culture medium containing 3 M Fura-2/AM dissolved in Me 2 SO, and incubated for 1 h. Coverslips were rinsed twice with 1 ml of medium, placed in a perfusion chamber, and perfused with imaging buffer (154 mM NaCl, 5.6 mM KCl, 5.0 mM HEPES, 1.2 mM MgCl 2 , 2.2 mM CaCl 2 , 10 mM glucose, 1 mg/ml bovine serum albumin, pH 7.4). Rat pinealocytes were imaged using the Image-Master system (Photon Technology International). The average Ca 2ϩ level in individual rat pinealocytes was determined from the ratio of fluorescence emissions obtained using two different excitation wavelengths (340 and 380 nm).
Measurement of IP 3 -The IP 3 concentration in cells was determined using a competition assay with [ 3 H]IP 3 , as described previously (21), with minor modifications. In brief, pinealocytes were stimulated as indicated, and the reaction was terminated by aspirating medium off the cells, followed by the addition of 5% (w/v) ice-cold trichloroacetic acid containing 4 mM EGTA. Samples were left on ice for 30 min to extract water-soluble inositol phosphates. Samples were transferred to Eppen-dorf tubes, and trichloroacetic acid was removed by extraction with diethyl ether four times. The final preparation was neutralized with 200 mM Trizma base, and the pH was adjusted to ϳ7.4. Assay buffer (50 mM Tris/HCl, pH 8.0, 4 mM EGTA, 4 mg/ml bovine serum albumin), [ 3 H]IP 3 (0.1 Ci/ml), and IP 3 -binding protein were added to cell extracts. The mixture was incubated on ice for 15 min and centrifuged at 10,000 ϫ g for 5 min. Scintillation mixture was added to the pellet to measure radioactivity. The IP 3 concentration was determined using a standard curve. IP 3 -binding protein was prepared from bovine adrenal cortex according to the method of Challiss et al. (22).
Quantitation of cAMP-The cAMP concentration in rat pineal glands was determined using the [ 3 H]cAMP competition assay for evaluating interactions with cAMP-binding protein (23). Rats were decapitated at midday (12:00), and glands were placed directly into ice-cold DMEM containing 10% fetal bovine serum and 1% penicillin/streptomycin. After the removal of extraneous tissue, glands were placed on nylon mesh that rested on DMEM medium, and incubated at 37°C in a humidified atmosphere of 5% CO 2 for 1 h before the experiment. Glands were stimulated with agonists for 3 min at 37°C and transferred to a 1.5-ml Eppendorf tube containing 200 l of absolute ethanol for termination of the reaction. This was followed by incubation for 2 h at Ϫ20°C to extract cAMP. Glands in ethanol were sonicated for 10 s and centrifuged with 2,500 ϫ g for 10 min at 4°C. The supernatant was evaporated to dryness in a SpeedVac (Savant instruments, Farmingdale, NY). Residues were dissolved in 0.2 ml of Tris-HCl (pH 7.5) and 4 mM EDTA. Sample solution (50 l) was employed in the cAMP assay. The assay is based on competition between [ 3 H]cAMP and unlabeled cAMP in the sample for crude cAMP-binding protein prepared from bovine adrenal cortex according to the method of Brown et al. (24). Free [ 3 H]cAMP was adsorbed using charcoal and removed by centrifugation, and bound [ 3 H]cAMP in the supernatant was measured by liquid scintillation. Each unknown sample was incubated with 50 l of [ 3 H]cAMP (5 Ci/ml) and 100 l of binding protein for 2 h at 4°C. Separation of protein-bound cAMP from unbound cAMP was achieved by adsorption of free cAMP onto charcoal (100 l), followed by centrifugation (12,000 ϫ g) at 4°C. Supernatant fractions (200 l) were placed in Eppendorf tubes containing 1.2 ml of scintillation mixture to measure radioactivity. The cAMP concentrations in samples were determined using a standard curve and expressed as picomoles/gland.
AANAT Activity Assay-AANAT activity was measured as described previously (25). Briefly, rats were decapitated at midday (12:00), and pineal glands were placed directly into ice-cold DMEM containing 10% fetal bovine serum and 1% penicillin/streptomycin. After the removal of extraneous tissue, pineal glands were placed on nylon mesh that rested on DMEM medium, and incubated at 37°C in a humidified atmosphere of 5% CO 2 for 1 h before the experiment. Following stimulation, rat pineal glands were individually disrupted by ultrasound in 100 l of ice-cold phosphate buffer (50 mM, pH 6.8). Debris was removed by centrifugation (15,000 ϫ g, 5 min at 4°C), and the supernatant was transferred to a new tube and stored at Ϫ80°C until use. An amount (8 l or indicated volumes) of the supernatant was incubated in the presence of 5 l of tryptamine-HCl (10 mM), 1 l of acetyl CoA (0.5 mM), and 1 l of [ 3 H]acetyl CoA (3.6 Ci/mmol, 250 Ci/ml). Phosphate buffer (50 mM, pH 6.8) was added to obtain a final volume of 20 l. Incubation was performed at 37°C for 30 min, and the was reaction terminated by dilution with an additional 180 l of ice-cold phosphate buffer (50 mM, pH 6.8). Econofluor was rapidly added to the whole diluted reaction mixture, and after incubation for 15 min, the amount of radiolabeled acetyltryptamine was determined in a liquid scintillation counter.
Immunoblotting-Proteins from pineal glands prepared at indicated times were separated by SDS-PAGE and blotted onto nitrocellulose membrane (0.45 mm, Bio-Rad), as described previously (28). Membranes were incubated with the AC5/6 antibody (Santa Cruz Biotechnology, Santa Cruz, CA) or AANAT antibody (serum 2500, 1:30,000 dilution, a kind gift from Dr. D. C. Klein, National Institutes of Health, Bethesda, MD). Signals were detected with the ECL detection system.
Statistical Analysis-All quantitative data are presented as means Ϯ S.E. of a minimum of three experiments. Comparisons between two groups were analyzed via t test, and values of p Ͻ 0.05 were considered to be significant.

RESULTS
Following perfusion of pinealocytes with 100 nM BK, [Ca 2ϩ ] i increased transiently in 90% of the cell population. Serial perfusion of 100 nM BK, 1 M 1,1-dimethyl-4-phenylpiperazinium iodide (DMPP), 1 M phenylephrine, and 70 mM KCl (High K ϩ ), which are known [Ca 2ϩ ] i -mobilizing agents in rat pinealocytes (29,30), led to transient elevation of [Ca 2ϩ ] i (Fig. 1A). These responses correlate with the typical characteristics of rat pinealocytes. BK triggered an increase in [Ca 2ϩ ] i in a concentration-dependent manner (Fig. 1B). The concentration dependence curve discloses maximal and half-maximal effective con-centration (EC 50 ) values for BK of 1 M and 50 nM, respectively. The B1 BK receptor-specific agonist, [des-Arg 10 ]kallidin (DAK), had little effect on [Ca 2ϩ ] i mobilization ( Fig. 2A). To further define the subtype of BK receptor involved, cells were perfused with specific BK antagonists. In view of the finding that IP 3 induces Ca 2ϩ mobilization from intracellular stores through specific receptors, we measured the IP 3 generated as a result of BK treatment. Fig. 2C shows that BK treatment stimulated IP 3 generation, whereas DAK, a B1 BK receptor-specific agonist, had little effect. Antagonists of BK receptors had similar effects on IP 3 generation, as evident from [Ca 2ϩ ] i mobilization data. HOE140, a B2 BK receptor-specific antagonist, blocked BK-induced IP 3 generation. In contrast, dHOE140, a B1 BK receptor-specific antagonist, did not affect BK-induced IP 3 generation. The results collectively indicate that phospholipase C (PLC)-linked BK receptors exist in rat pinealocytes, and the subtype of BK receptors expressed in rat pinealocytes is B2.
To elucidate the physiological roles of BK, we focused on its effect on ␤-adrenergic receptor signaling, the major pathway of melatonin synthesis. Previous studies have shown that [Ca 2ϩ ] i increase potentiates the effects of ␤-adrenergic stimulation in pineal glands (31,32). However, BK-mediated modulation of ␤-adrenergic receptor signaling was distinct from the well established concept of Ca 2ϩ signaling on melatonin synthesis. BK differentially modulates ␤-adrenergic signaling according to the circadian time. At ZT05, isoproterenol-induced cAMP genera- tion was inhibited by about 30% by BK (Fig. 3A). On the other hand, at ZT17, isoproterenol-induced cAMP generation was increased 1.8-fold in the presence of BK (Fig. 3B). Furthermore, forskolin-induced AC activity was differentially regulated by BK, dependent on the time of day. At ZT05, co-treatment with BK resulted in about 50% inhibition of forskolin-stimulated AC activity (Fig. 3A), whereas at ZT17, BK enhanced forskolin-stimulated AC activity by about 1.5-fold (Fig. 3B). Additionally, differential regulation of basal pineal AC activity by BK was observed. At ZT05, BK inhibited basal AC activity about 80%, whereas at ZT17, enzyme activity was stimulated ϳ20-fold (Fig. 3C).
These results indicate that BK diurnally inhibits ␤-adrenergic receptor signaling through suppression of AC activity.
Because Ca 2ϩ -mediated potentiation of ␤-adrenergic stimulation in pineal glands is a well known concept, we focused on the opposing inhibitory effect of BK on ␤-adrenergic signaling during the daytime. Co-treatment with BK resulted in concentration-dependent inhibition of isoproterenol-induced cAMP generation (Fig. 4A) and forskolin-stimulated AC activation (Fig. 4B) in rat pineal glands. Furthermore, concentration-dependent blockage of isoproterenol-induced AANAT activation (Fig. 4C) was also observed in rat pineal glands. All the above experiments were conducted with rat pineal glands excised at ZT05. The results indicate that BK acts as a negative regulator of melatonin synthesis at daytime through direct inhibition of AC.
We further investigated the mechanism of BK signaling that inhibits ␤-adrenergic receptor-mediated cAMP generation (TABLE ONE). To confirm the hypothesis that BK-evoked inhibition of ␤-adrenergic signaling is mediated through a guanine nucleotide-binding regulatory protein (G protein), such as G i/o , rat pineal glands were preincubated with 300 ng/ml PTX, which uncouples the G i/o protein from its receptor by catalyzing ADP-ribosylation of the ␣ i/o subunit, for 6 h. PTX had no effect on BK-specific inhibition of isoproterenol-mediated cAMP generation. Because the B2 BK receptor generally links to PLC through G q protein (33), we additionally evaluated whether PLC-mediated [Ca 2ϩ ] i rise contributes to the inhibitory effect of BK. U73122, a PLC inhibitor, completely reversed BK-induced suppression of isoproterenol-medi-  ated cAMP generation. Next, we examined the possibility that BK-specific inhibition of cAMP production in rat pineal glands was due to the downstream event of PLC activation. Intact rat pineal glands were preincubated with 20 M BAPTA/AM, a cell-permeable Ca 2ϩ chelator, to block the [Ca 2ϩ ] i increase. As shown in TABLE ONE, inhibition of cAMP generation by BK was completely reversed following pretreatment with BAPTA/AM. In addition, treatment of rat pineal glands with ionomycin (200 nM) led to blockage of isoproterenol-mediated cAMP generation. These results indicate that the inhibitory effect of BK on isoproterenol-induced cAMP generation in rat pineal glands is due to the direct effect of [Ca 2ϩ ] i increase on pineal AC.
To explain the differential modulation of ␤-adrenergic signaling by BK, which are dependent on the time of day, we hypothesize that "Ca 2ϩinhibitable" AC proteins are present in the pineal gland with expression patterns following circadian rhythms. To verify this theory, the expression of AC isoforms in rat pineal glands was examined by reverse transcription-PCR analysis. PCR primer sets specific for known rat AC genes were selected from the reported cDNA sequences (26,27). As shown in Fig. 5, amplified products of the expected sizes for all AC isoforms (except type V) were detected in rat pineal glands. These products were identified in at least three sets of experiments using different rat pineal glands. Moreover, nucleotide sequences of the amplified DNA products were identical to those of rat AC isoforms. Among the AC isoforms expressed in rat pineal glands, the Ca 2ϩ -inhibitable AC is type VI (AC6). Accordingly, AC6 expression was investigated during the day-night cycle in rat pineal glands by Northern blot and immunoblot analyses. AC6 mRNA and protein expression displayed significant circadian patterns of expression (Fig. 6). Specifically, levels of AC6 mRNA and protein started to increase at midnight (ZT17), and peaked at midday (ZT05). This expression pattern is complementary to that of FIGURE 5. Analysis of AC isoform expression in rat pineal glands. Reverse transcription-PCR data on transcripts for AC isoforms reveal that all known AC isoforms, except type V, are expressed in rat pineal glands. Primer pairs specific for AC isoforms I, II, III, IV, V, VI, VII, VIII, and IX were employed, as described under "Experimental Procedures." The template for PCR analysis was analyzed in the presence (ϩ) and absence (Ϫ) of reverse transcriptase. A, BK inhibits isoproterenol-induced cAMP generation in a concentration-dependent manner. Glands were co-treated with 10 M isoproterenol and various concentrations of BK for 3 min, and cAMP generation was measured using a competition assay, as described under "Experimental Procedures." Data are presented as means Ϯ S.E. values from three independent experiments. B, BK additionally inhibits forskolin-induced cAMP generation in a concentration-dependent manner. Glands were co-treated with 10 M forskolin and various concentrations of BK for 3 min, and cAMP generation was measured using a competition assay, as described under "Experimental Procedures." Data are presented as mean Ϯ S.E. values from three independent experiments. C, BK inhibits isoproterenol-induced AANAT activation in a concentration-dependent manner. Glands (three each) were co-incubated with 10 M isoproterenol and various concentrations of BK for 7 h. Following incubation, glands were collected, and AANAT activity was measured with the liquid biphasic diffusion assay, as described under "Experimental Procedures." Results were expressed as relative values Ϯ S.E. (three independent experiments). Isoproterenol (10 M)-induced AANAT activity in the absence of BK was taken as 100%.

Involvement of the ͓Ca 2؉ ͔ i in BK-specific inhibition of ␤-adrenergic signaling in rat pineal glands
Rat pineal glands were pretreated with the following compounds for the indicated time: 300 ng/ml PTX for 12 h, 10 M U73122 for 10 min, 20 M BAPTA/AM for 10 min, and 200 nM ionomycin for 10 min, respectively. Glands were stimulated with 10 M isoproterenol in the presence or absence (Control) of 1 M BK for 3 min. Following stimulation, glands were homogenized, and cAMP generation was assayed, as described under "Experimental Procedures." Results were expressed as relative values Ϯ S.E. (three independent experiments). Isoproterenol (10 M)-induced cAMP generation in the absence of BK was taken as 100%. AANAT mRNA and protein. In this case, the levels of AANAT mRNA and protein started to decrease at midnight (ZT17), and reached a minimum at midday (ZT05). Our results strongly support the theory that the differential modulation of BK on ␤-adrenergic signaling and pineal AC activity during the day and night is mainly due to diurnal expression of AC6.

DISCUSSION
The present study reveals that the B2 BK receptors are present and differentially modulate ␤-adrenergic signaling, the major pathway of melatonin synthesis, by direct suppression of Ca 2ϩ -inhibitable AC type VI (AC6) in rat pinealocytes, and that the differential modulation of ␤-adrenergic signaling by BK-evoked [Ca 2ϩ ] i increase is due to the diurnal expression of pineal AC6 mRNA and protein.
Recently, the presence of BK receptors in mammalian neuronal and neuroendocrine systems was demonstrated by immunohistochemical (34,35) and radioligand binding experiments (36). Although the physiological roles of BK are currently unclear, the presence of BK receptors indicates a role in the brain or neuroendocrine system. BK increases the permeability of the blood-brain barrier by activation of B2 receptors on brain endothelial cells (37). Moreover, BK stimulates the release of hormones and neurotransmitters, such as gonadotropin-releasing hormone and norepinephrine, through B2 receptor activation (38,39). Activation of B2 receptors triggers norepinephrine release from rat sympathetic neurons (40). Our present data show that stimulation of BK results in concentration-dependent inhibition of ␤-adrenergic recep-tor-mediated cAMP generation and AANAT activation in rat pineal glands during the daytime. Our results also indicate that inhibition of ␤-adrenergic receptor-induced cAMP generation by BK is mediated by the increase in [Ca 2ϩ ] i , which is inconsistent with the well established concept of intracellular free Ca 2ϩ on melatonin synthesis. Previous studies have shown that [Ca 2ϩ ] i increase potentiates the effects of ␤-adrenergic stimulation (30 -32). However, the effects of intracellular Ca 2ϩ mobilization in pineal glands at various circadian time points remain to be determined. Tzavara et al. (41) demonstrated inhibition of AC activity in rat pineal glands by calcium, both in the presence and absence of calmodulin at midday but marked activation by the calcium-calmodulin complex at night. In agreement with this finding, we showed that BKevoked [Ca 2ϩ ] i diurnally suppresses ␤-adrenergic signaling through inhibition of pineal AC activity in rat pineal glands. To elucidate the differential modulation mechanism of BK, we focused on the Ca 2ϩ -inhibitableAC proteins. To date, nine isoforms of AC have been cloned (20). Among these, types V and VI constitute a subfamily, which has the remarkable property of being inhibited by submicromolar Ca 2ϩ concentrations in addition to G␣ i -mediated processes (42). Several previous studies showed the involvement of Ca 2ϩ -inhibitable AC in the inhibitory mechanism of BK in various cell types. Stimulation of BK receptors markedly reduced cAMP production in C6 -2B rat glioma cells through the PTX-insensitive elevation of intracellular Ca 2ϩ concentration (17). In NCB-20 cells, BK stimulates Ca 2ϩ mobilization, leading to direct suppression of AC (18). These reports suggest the possibility of direct inhibition of ␤-adrenergic receptor-linked AC by BK-mediated [Ca 2ϩ ] i in rat pineal glands. Initially, we identified the AC isoforms expressed in rat pineal glands. Reverse transcription-PCR analyses revealed the expression of types I, II, III, IV, VI, VII, VIII, and IX AC proteins, of which type VI (AC6) is the Ca 2ϩ -inhibitable AC. We examined the time-dependent expression patterns of AC6 mRNA and protein. Northern blot and immunoblot analyses revealed circadian expression of AC6 mRNA and protein, with maximal levels at ZT01 and minimal levels at ZT13. Moreover, the expression profiles of AC6 and AANAT were complementary. Specifically, when the AANAT mRNA level started to decrease, AC6 transcript expression increased. These findings support the possible involvement of AC6 in diurnal inhibition of ␤-adrenergic receptor-mediated cAMP generation and AANAT activity by B2 bradykinin receptor signaling in rat pineal glands.
Previous studies have reported the diurnal expression of AC1 mRNA in rat pineal glands (41,43) and circadian patterns of AC1 mRNA and protein activity. AC1 mRNA levels are maximal during the daytime and minimal at night. These results imply that expression of AC1 and AC6 mRNA follow the same circadian pattern. Interestingly, unlike AC6, AC1 mRNA expression and protein activity are not in phase, but rather they are inversely related. Whereas the AC1 transcript level peaks during the day, activity is maximal at night. Chan et al. (44) explained the uncoupling of AC1 mRNA expression and protein activity by showing the presence of the cAMP-inhibitable element in the AC1 promoter region. Based on our results, in association with earlier data, we have formed an interesting hypothesis. AC1 and AC6 are part of the Ca 2ϩregulated AC subclass. One is Ca 2ϩ -stimulatable (AC1), whereas the other is Ca 2ϩ -inhibitable (AC6). The activities of these two ACs follow the circadian rhythm with inverse patterns in rat pineal glands. During the daytime, AC6 activity reaches a peak. Conversely, AC1 activity peaks at night. In view of these findings, we propose that [Ca 2ϩ ] i increase acts as a Janus-faced modulator of AC signaling in pineal glands. At night, the [Ca 2ϩ ] i increase potentiates ␤-adrenergic receptor-mediated cAMP signaling through activation of AC1 (41) and protein kinase C (45). During the daytime, ␤-adrenergic receptor-mediated FIGURE 6. Circadian variation of AC6 expression in the rat pineal gland. Rats were housed in a controlled lighting environment (LD12:12), with lights on at circadian time (CT) 07. The filled bar represents the times when lights were off. A, Northern blot and immunoblot analysis display the circadian expression of rat AC type 6 (rAC6). AANAT mRNA and protein levels are shown as a comparing control. B, time course of relative AC6 protein (open circle) and AANAT mRNA expression (filled circle). Band intensities of AANAT transcripts and rAC6 protein were normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA and protein to correct for variations in loading. AANAT/ glyceraldehyde-3-phosphate dehydrogenase ratio at ZT17 and rAC6/glyceraldehyde-3phosphate dehydrogenase ratio at ZT01 were adjusted to 100%. cAMP signaling is suppressed via direct inhibition of AC6 by [Ca 2ϩ ] i . In other words, differential circadian modulation of the pineal cAMP level by [Ca 2ϩ ] i is influenced by circadian changes in the AC1/AC6 ratio in rat pineal glands. Indeed, this hypothesis is based on speculation rather than fact, and more extensive studies are necessary for confirmation.
Until now, several mechanisms have been proposed to explain the termination of nocturnal melatonin synthesis. There is no doubt that termination of melatonin synthesis is mainly initiated by the termination of norepinephrine release (46). Other mechanisms have been proposed, such as rapid proteasomal proteolysis of AANAT (47) and inhibition of ␤-adrenergic signaling by metabotropic glutamate receptors through the inhibitory guanine nucleotide-binding protein (Gi) (48). In addition, we have shown that rhythmic degradation of AANAT mRNA is also essential to achieve circadian oscillation of melatonin synthesis (49). Here, we propose an additional termination mechanism of nocturnal melatonin synthesis by the inhibition of ␤-adrenergic signaling through blocking Ca 2ϩ -inhibitable AC6 with [Ca 2ϩ ] i -mobilizing agents, such as BK. We additionally provide evidence that [Ca 2ϩ ] i differentially modulates the pineal cAMP level according to circadian time, due to periodic changes in AC1/AC6 activity in rat pineal glands.