Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The cerebellum, an evolutionary conserved structure specific to vertebrates, is involved in the maintenance of balance and orientation, the refinement of motor action, motor memory storage, and possibly some aspects of cognition (Fiez, 1996). Several neurotransmitters, including acetylcholine, contribute to these functions. The predominant cholinergic system in the cerebellum is formed by the mossy fibers originated from the vestibular nuclei and projected to granule cells and unipolar brush cells (Jaarsma et al., 1997).

Nicotinic acetylcholine receptors (nAChRs) in cerebellum have been revealed by molecular biology, immunocytochemical, and binding studies. A differential subtype distribution was suggested by immunocytochemistry detection of 3, 4, and 2 proteins in granule cells, 4 and 2 in deep cerebellar nucleus, and 4, 7, and 2 in Purkinje cells (Wada et al., 1989; Nakayama et al., 1997). Binding to selective ligands indicates the presence of 42 (Flores et al., 1992) and 7 subunits (Didier et al., 1995). Recent electrophysiological study shows that nAChRs present in the soma of granule cells are predominantly the 42 subtype, whereas the 7 subtype is present preterminally and promotes the release of glutamate from mossy fibers (De Filippi et al., 2001). Receptors sensitive to dihydro--erythroidine were detected electrophysiologically in Purkinje cells from 5-day-old rats. However, they were barely detectable in older rats (more than 10 days old), indicating that nAChRs are regulated developmentally (Kawa, 2002).

The light/dark cycle modulates the function of many tissues including central and peripheral cholinergic synapses. Peripherally, we have shown that the number and the response of nAChRs located on sympathetic nerve terminals of rat vas deferens (Carneiro and Markus, 1990) present a circadian rhythm (Carneiro et al., 1991; Markus et al., 1996). In the central nervous system a daily variation in nicotinic function and number of binding sites was also observed. Nicotinic administration affected locomotor activity during the day but not during the night (Morley and Garner, 1990). The density of -[¹²⁵I]bungarotoxin binding sites in rat hypothalamus was lower at the end of a 12-h dark period when compared with the light period (Morley and Garner, 1990).

Environmental illumination changes are translated to internal body milieu by the nocturnal surge of the pineal gland hormone, melatonin, which is controlled by the suprachiasmatic nucleus of the hypothalamus and synchronized by the light/dark cycle to a 24-h period (Reiter, 1992). The gland is innervated primarily by the peripheral sympathetic tract (Kappers, 1979), which releases noradrenaline and ATP (Barbosa et al., 2000). Stimulation of ₁-adrenoceptors is a required step for induction of the rate-limiting enzyme, N-acetyltransferase, transcription (Klein et al., 1981). Therefore,  antagonists inhibit pineal melatonin production (Lipton et al., 1981).

Melatonin promotes different responses in the cerebellum or cerebellar cells. Ontoneurological examination of patients receiving melatonin confirms that this hormone plays a role in the sensorimotor control of balance (Fraschini et al., 1999). Besides being implicated in the control of equilibrium, melatonin was also shown to play a role in cell physiology. Melatonin concentration within the physiological range binds to calmodulin-inhibiting nitric-oxide synthase activation (Pozo et al., 1997). Antioxidative stress activity in cerebellar tissue and granular cerebellum cells (Mason et al., 1999) and inhibition of granule cells apoptosis (Persengiev, 2001) were demonstrated. Melatonin membrane receptors were identified in rat and human cerebellum by binding techniques (Laudon et al., 1988; Al-Ghoul et al., 1998); however, melatonin function on cerebellar neurons is still poorly understood.

Considering that 1) the main function of the cerebellum is to improve the accuracy of movements by comparing descending motor commands and sensitive vestibular input with motor action information, 2) the main cerebellar cholinergic input comes from vestibular nuclei (Jaarsma et al., 1997), and 3) melatonin interferes with the maintenance of proper balance (Fraschini et al., 1999), our aim was to show that nocturnal melatonin surge modulates cerebellar cholinergic response. The experimental model chosen was to measure the [³H]glutamate overflow from cerebellum slices, induced by stimulation of nAChRs. The following items were analyzed: 1) the subtype of nAChRs responsible for the effect, 2) the influence of the environmental light/dark cycle on the number and function of nAChRs, 3) the importance of nocturnal production of melatonin, and 4) the effect of melatonin reposition. In this study we show for the first time that [³H]glutamate released by stimulation of 7 nAChRs is different during the light and dark phases of the day and that the number of -[¹²⁵I]bungarotoxin binding sites is also under lighting control.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals and Tissue Preparation. Male Wistar rats (4 months old) were housed five per cage in an animal room lit and kept under a light/dark cycle of 12/12 h with water and food ad libitum. The animals were killed by decapitation 3 h before (light phase) or 6 h after (dark phase) lights off. These hours were chosen because in the vas deferens the lowest response to nAChRs is observed 3 h before lights off, whereas the maximum melatonin content occurs 6 h after lights off (Carneiro et al., 1991, 1993; Reiter, 1992). The cerebellum was dissected out, chopped at 300 µm thickness (McIwain Tissue Chopper), and immediately resuspended in a high osmolarity solution of the following composition: 124 mM NaCl, 3 mM KCl, 20 mM MgCl₂, 1.3 mM NaH₂PO₄, 0.5 mM CaCl₂, 26 mM NaHCO₃, 10 mM glucose, and 200 mM sucrose, gassed with 95% O₂/5% CO₂ (Richerson and Messer, 1995).

[³H]Glutamate Overflow. Tissue slices were first incubated with [³H]glutamate (8 nM; 51.9 Ci/mmol, 37°C) in the incubation solution for 5 min, and then washed for 2 min, centrifuged (2000g, 10 s), and resuspended in a second solution of the following composition: 118 mM NaCl, 4.8 mM KCl, 1.2 MgSO₄, 1.2 mM KH₂PO₄, 2.5 CaCl₂, 25 NaHCO₃, and 11 glucose, gassed with 95% O₂/5% CO₂ (Barnes et al., 1994). This solution was perfused for 10 min before beginning the collection of the samples, which were collected at 2-min intervals. After two baseline samples, a 2-min pulse of agonist (third sample) was given. Two sequential 2-min samples were collected before measuring the residual radioactivity of the slices. Therefore, four samples were collected (baselines 1 and 2, agonist present, third sample). The antagonists were incubated since slices were added to normal osmolar solution. Samples were transferred to 5-ml vials with scintillation liquid Ecolume. After 12 h, samples were counted for radioactivity in a Tricarb 2100TR Liquid Scintillation Counter (PerkinElmer Life Sciences, Boston, MA).

()-[³H]Nicotine and -[¹²⁵I]Bungarotoxin Binding Sites. Cerebellar membranes were prepared from animals killed 3 h before (light phase) and 6 h after (dark phase) lights off. The cerebella were washed and homogenized in Tris-sucrose buffer [10 mM Trizma (Tris base), 320 mM sucrose, pH 7.4; 4°C] plus 0.1 mM phenylmethylsulfonyl fluoride, centrifuged at 500g (10 min, 4°C). The supernatant was further centrifuged (40,000g, 10 min, 4°C), and the resulting pellet was resuspended in Tris-sucrose buffer (1 mg protein/ml). Protein concentration was estimated according to the method of Spector (1978), with bovine serum albumin taken as standard.

Drugs and Chemicals. -Bungarotoxin, choline chloride, dihydro--erythroidine, (+)-epibatidine, melatonin, methyllycaconitine citrate, ()-nicotine hydrogen tartrate, and DL-propranolol hydrochloride were purchased from Sigma/RBI (Natick, MA). [³H]Glutamate (specific activity 50-51.9 Ci/mmol) and -[¹²⁵I]bungarotoxin (specific activity 85 Ci/mmol) were purchased from PerkinElmer; ()-[N-methyl-³H]nicotine (specific activity 80 Ci/mmol) was purchased from Amersham Biosciences UK, Ltd. (Little Chalfont, Buckinghamshire, UK). Salts were purchased from Quimitra S/A (Rio de Janeiro, Brazil). Environmental safe liquid scintillation cocktail, Ecolume, was purchased from ICN Pharmaceuticals (Costa Mesa, CA).

Data Analysis. The percentage of tritium overflow in the presence of the agonists (third sample) normalized for basal response (second sample) was taken as agonist effect. We tested the correlation between the basal overflow and the overflow induced by 1 nM nicotine (MatLab Software, Natick, MA), the Pearson coefficient was 0.99 (p < 0.00001).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Effect of Nicotinic Agonists and Antagonists on the Release of [³H]Glutamate in Perfused Cerebellum Slices. Nicotine (10⁹-10⁶ M) and epibatidine (10¹⁴-10¹⁰ M) stimulate, in a concentration-dependent manner, the fractional release of [³H]glutamate from cerebellar slices (Fig. 1; animals killed at the light phase). The response to nicotine (log EC₅₀ value = 8.06 ± 0.58, n = 65, number of replicates for fitting the nonlinear sigmoidal curve) was 40,000-fold less potent than that for epibatidine (log EC₅₀ 13.29 ± 1.00, n = 43). Choline (3.10⁷ M)-induced [³H]glutamate release was blocked by methyllycaconitine (MLA, 10⁹ M). Epibatidine (10¹³ M)-induced effect was blocked by 10⁹ M MLA and 10⁹ M -bungarotoxin. However, dihydro--erythroidine (10⁶ M) was not able to block epibatidine (10¹³ M)-induced [³H]glutamate release (Fig. 2). Therefore, in our model, the cholinergic-evoked release of [³H]glutamate is mediated by stimulation of 7-nAChRs.

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Nicotine- and epibatidine-induced cerebellum slices with [3H]glutamate overflow (expressed as percentage over the basal release). Animals were killed 3 h before lights off. Data are shown as mean ± S.E.M. of the number of animals shown around the symbols.

View larger version (20K): [in this window] [in a new window]   Fig. 2.   Effect of MLA (109 M), -bungarotoxin (BTX, 109 M), and dihydro--erythroidine (DHE, 106 M) on choline (107 M)- and epibatidine (1013 M)-induced cerebellum slices with [3H]glutamate overflow. Animals were killed 3 h before lights off. Number of experiments is shown inside the bars. Experiments in the presence or absence of antagonists were run in parallel with slices from the same animal. The data are presented as the percentage of overflow in the presence or absence of the antagonists.

Cerebellar Nicotine Acetylcholine Receptor Binding Is Influenced by Environmental Lighting. The ()-[³H]nicotine and -[¹²⁵I]bungarotoxin binding to cerebellum membranes was saturable, reaching the maximum around 40 nM and 5 nM, respectively (Fig. 3). Specific binding represents about 90% of total binding for both ligands. K_d values, which were not significantly (p > 0.05) different between light (nicotine, 6.75 ± 1.9 nM; -bungarotoxin, 1.17 ± 0.23 nM, n = 4) and dark (nicotine, 8.58 ± 1.37 nM; -bungarotoxin, 0.97 ± 0.24 nM, n = 4) periods, are in agreement with values found in the literature for whole brain (Reavill et al., 1988). The effect of environmental lighting on the maximal number of receptors, however, was specific for each ligand. B_max values for ()-[³H]nicotine were not different between the two groups (light, 42.87 ± 3.08; dark, 48.38 ± 4.69 fmol/mg protein, n = 4). On the other hand, the maximal number of sites for -[¹²⁵I]bungarotoxin was lower in the dark phase (20.68 ± 1.52 fmol/mg protein) than in the light phase (32.25 ± 2.03 fmol/mg protein) (p < 0.05).

View larger version (15K): [in this window] [in a new window]   Fig. 3.   ()-[3H]Nicotine and -[125I]bungarotoxin binding saturation curves for membranes isolated from rat cerebellum. Nonspecific binding was based on nicotine (1 mM). Animals were killed 3 h before (open symbols) and 6 h after (closed symbols) lights off. Data represent the mean ± S.E.M. of three independent experiments.

Environmental Lighting and Melatonin Influence the Nicotine-Induced Release of [³H]Glutamate. [³H]Glutamate overflow induced by stimulation of nicotinic receptors was higher during the dark phase than during the light phase of the day (Fig. 4), when either nicotine or epibatidine was used as agonist of nAChRs. On the other hand, the release evoked by glutamate was not modified by the environmental illumination. The basal release of [³H]glutamate (before adding any agonist) was not dependent on environmental lighting. For example, in the experiments in which nicotine was used as agonist, the basal release at the light phase was 5.13 ± 1.08%/min (n = 5) and at the dark phase was 4.84 ± 1.03%/min (n = 5, p > 0.05).

View larger version (15K): [in this window] [in a new window]   Fig. 4.   Effect of light/dark cycle on nicotine (108 M)-, epibatidine (1013 M)-, and glutamate (107 M)-induced cerebellum slices with [3H]glutamate overflow. Data are shown as mean ± S.E.M. of the number of experiments shown inside the bars. Open and closed bars represent the data for rats killed during the light or the dark phase, respectively. , significantly different from data obtained during the light phase, p < 0.05. Number of experiments is shown in the graph.

8

View larger version (23K): [in this window] [in a new window]   Fig. 5.   Effect of endogenously produced and nocturnally administered melatonin on the nicotine (108 M)-induced cerebellum slices with [3H]glutamate overflow. LL, animals exposed to constant light for 48 h, and killed together with the other animals. In all other groups, animals were maintained in a 12/12-h light/dark cycle (LD), injected with propranolol or vehicle 1 h before lights off, and killed 6 h after lights off. In the group treated with propranolol and melatonin (9 mg/kg/day for 2 days), melatonin was injected during the dark period in the schedule shown under Materials and Methods. Data are shown as mean ± S.E.M. of the number of experiments shown in the graph. a, significantly different from animals maintained in LL, p < 0.05; b, significantly different from the LL group in the absence of propranolol, p < 0.05; c, significantly different from the group treated only with propranolol, 20 mg/kg, p < 0.05.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In the present study, our principal result shows that release of [³H]glutamate induced by stimulation of 7 nicotinic AChR and the B_max for -[¹²⁵I]bungarotoxin varies according to environmental illumination. Furthermore, using protocols that block nocturnal pineal melatonin production, and replacing the hormone, we demonstrate that the pineal hormone is responsible for the nocturnal increase in [³H]glutamate release.

Neuronal nAChRs are made of pentamers of different  (2-10) and  (2-4) subunits. The prevalent functional nAChRs in the mammalian brain are those composed of 7 or 42 subunits (Role and Berg, 1996), and can both mediate and modulate fast synaptic transmission in the brain. Activation of presynaptic or preterminal 7 or 42 nAChRs enhances the release of many neurotransmitters in diverse brain regions (Lena et al., 1993; Guo et al., 1998; De Filippi et al., 2001). 7-containing nAChRs mediate fast cholinergic synaptic transmission onto interneurons in the mammalian hippocampus, which indicates that postsynaptic and somatic nAChRs can have biological roles, as they have in the periphery (Frazier et al., 1998; Hefft et al., 1999).

In cerebellum slices of neonatal rats, 42- and 7-nAChRs are likely to be present in the somatic and presynaptic levels, respectively (De Filippi et al., 2001), and receptors sensitive to dihydro--erythroidine were shown to be present preterminally in Purkinje cells (Kawa, 2002). Using a neurochemical technique, we demonstrate that nicotine-induced [³H]glutamate release in adult cerebellar slices is mediated by 7-nAChRs, as was shown before for (4- to 15-day-old) rats using an electrophysiological technique, and neurochemically for adult rats (Renó and Markus, 2000). The three agonists tested, epibatidine, nicotine, and choline, were able to evoke the release of [³H]glutamate. Choline is a selective agonist for 7-nAChRs (Alkondon et al., 1997), and its effect was blocked by MLA, initially put out as a selective 7 antagonist (Alkondon et al., 1992), and recently proposed to block somatodendritic nAChRs [456(2)2], which release dopamine in the midbrain (Mogg et al., 2002). The response to epibatidine, which is able to stimulate either heteromeric or homomeric nAChRs, was blocked by MLA and -bungarotoxin and was not affected by dihydro--erythroidine, an antagonist that blocks heteromeric receptors (Wonnacott, 1997). Thus, our data clearly point to the conclusion that 7-nAChRs are responsible for evoking [³H]glutamate release from adult cerebellum slices, confirming previous observations (Renó and Markus, 2000; De Filippi et al., 2001).

We have previously shown that environmental illumination interferes with the number and response of peripheral 7-nicotinic receptors located in sympathetic nerve terminals (Markus et al., 1996; Zago and Markus, 1999). To evaluate whether these receptors, in the cerebellum, could be modified by environmental illumination, binding parameters (B_max and K_d) of cerebellum membranes from rats killed 3 h before, or 6 h after, lights off were analyzed. Two radioligands were used, ()-[³H]nicotine, which binds with higher affinity to heteromeric nAChRs (Flores et al., 1992), and -[¹²⁵I]bungarotoxin, which is selective for 7 binding sites (Seguela et al., 1993). ()-[³H]Nicotine binding site density in cerebellum membranes was higher than that of -[¹²⁵I]bungarotoxin binding sites. No difference between light and dark phase was observed for ()-[³H]nicotine B_max, whereas for -[¹²⁵I]bungarotoxin, a nocturnal reduction in B_max was observed. Therefore, the 7-nAChRs are down-regulated during the dark phase, when compared with the end of the light phase. This result was similar to that obtained in hypothalamus (Morley and Garner, 1990) but was different from the one obtained in peripheral tissue, where 7- nAChRs were shown to be up-regulated during the dark phase in preterminal noradrenergic neurons, which innervate the rat vas deferens (Markus et al., 1996; Zago and Markus, 1999). It is interesting to consider that even the clock genes present out-of-phase expression when different tissues are compared. Recently, a comparative analysis of circadian gene expression in vivo in mouse liver and heart showed that the distributions of circadian phases in the two tissues are markedly different (Storch et al., 2002). The divergence of sympathetic and mossy fiber nerve terminals in the melatonin regulation of the number of 7-nAChRs suggests a special role for circadian timing in each tissue.

The next step was to evaluate whether a functional response induced by stimulation of 7-nAChRs in cerebellum slices could depend on environment illumination. Cerebellum slices released more [³H]glutamate when 7-nAChRs were stimulated during the dark than during the light phase. An inverse correlation between functional response and number of binding sites was also observed previously in striatum, where an up-regulation resulted in a lower functional response (Wonnacott, 1997). However, in cortex a direct correlation between functional response and number of nAChRs was observed (Kawai and Berg, 2001). Accordingly, alternative explanations must be considered. A tentative consideration is that changes in number of 7-nAChRs expressed at presynaptic glutamatergic terminals are not revealed by binding studies made in whole cerebellum preparations. The 7-nAChRs expressed on Purkinje cells (Wada et al., 1989) and the soma of granule cells (De Filippi et al., 2001) possibly represent the majority of 7-nAChRs in cerebellum and at the same time may have overshadowed changes on presynaptic 7-nAchRs. Therefore, the ultimate relationship between lighting influence on functional response and number of nAChRs has yet to be determined.

The light/dark difference in nicotine-induced [³H]glutamate is a result of a nocturnal surge of melatonin, inasmuch as the inhibition of melatonin production (by maintaining the animal in constant light, or blocking the -adrenoceptor) impairs the nocturnal increase in nicotine-induced response. The conclusion that a nocturnal melatonin surge is essential for diurnal variation in nicotine-induced response in the cerebellum slices was reinforced by the nocturnal administration of melatonin to animals treated with propranolol, because this treatment increased the ability of nicotine in releasing [³H]glutamate.

Therefore, considering that the highest proportion of choline acetyltransferase-positive fibers is detected in the vestibulocerebellum, which is known to regulate equilibrium (Jaarsma et al., 1997), and melatonin was shown to control sensorimotor performance, related to postural balance (Fraschini et al., 1999), we may conclude that our data give the first neurochemical evidence that melatonin modulates cerebellar cholinergic input by interfering in nAChR receptor availability.