- Email: [email protected]
Docosahexaenoic Acid Reduces Both Cyclic Nucleotide and Melatonin Synthesis in Rat Pinealocytes
Docosahexaenoic Acid Reduces Both Cyclic Nucleotide and Melatonin Synthesis in Rat Pinealocytes Zaouali-Ajina Monia, Gharib Abdallah, Gharib Claude, and Nicole Sarda1 Laboratoire de Physiologie de l’Environnement, Faculte´ de Me´decine Grange-Blanche, 69373 Lyon Cedex 08, France The effect of docosahexaenoic acid (22:6n-3) on melatonin (MT) production in rat pinealocytes was measured. In pinealocytes, supplementation with 50 mM 22:6n-3 for 48 h decreased MT production after a norepinephrine (NE)-stimulation (1 and 10 mM). Using co-treatment with IBMX (1 mM), a cyclic nucleotide phosphodiesterase inhibitor known to increase NE-stimulated cAMP and prazosin (1 mM), an a1 antagonist, the results indicate that a) 22:6n-3 did not act on cyclic nucleotide phosphodiesterase activities to decrease both cAMP (264%) and cGMP (245%) levels and b) the inhibition of cGMP was higher (270%) in presence of prazosin. These results confirm that in rat pinealocytes the lowering effect of 22:6n-3 supplementation on cAMP is likely to be dependent on a1-adrenoceptor modulation whereas the decrease of cGMP seems to involve other intracellular mechanisms. The absence of 22:6n-3 effect on serotonin metabolites suggest that the lowering effect on MT may be also correlated with alterations of O-methylation step. Keywords: Rat; pineal; melatonin; cAMP; cGMP; Docosahexaenoic acid.
Introduction The pineal hormone melatonin (MT) plays a central role in controlling seasonal reproduction and other rhythmic events (1). In mammalian 1 Address correspondence to: N. Sarda, Laboratoire de Neuropharmacologie Mole´culaire, CNRS, UMR 5542, Faculte´ de Me´decine Lae¨nnec, rue Guillaume Paradin, 69007 Lyon, France; Phone: (33) 04 78 77 87 81; Fax: (33) 04 78 77 87 32.
Prostaglandins & other Lipid Mediators 55:291–300, 1998 © 1998 by Elsevier Science Inc. 655 Avenue of the Americas, New York, NY 10010
0090-6980/98/$19.00 PII S0090-6980(98)00028-8
Docosahexaenoic Acid, Nucleotides, and Melatonin: Monia et al. species, MT is synthesized nocturnally in the pinealocytes by N-acetylation and subsequent O-methylation of serotonin (5HT) after noradrenergic stimulation (2). The limiting factors for MT synthesis may be the substrate availability (i.e., 5HT), the cAMP level, the activity of the arylalkylamine-N-acetyltransferase (NAT), or the second enzyme (hydroxyindol-O-methyltransferase, HIOMT) involved in MT synthesis. In rat, norepinephrine (NE) increases cAMP and cGMP levels through an action involving both a1- and b-adrenoceptors and increases [Ca21]i and inositol phosphate turnover through a1-adrenergic mechanisms (3,4). The physiological role of cAMP is well known. cAMP participates in the stimulation of three enzymes, i.e., NAT, which is the rate-controlling enzyme in the pathway that converts 5HT to MT (2), tryptophan hydroxylase (5) and thyroxine type II 59 deiodinase (6). In contrast, the functional significance of cGMP remains still obscur. Recently, data have shown an involvement of NO in cGMP synthesis in response to NE in dissociated pinealocytes (7). At the same time, other studies have provided evidence for the role of polyunsaturated fatty acids (PUFAs) in pineal activity. Docosahexaenoic acid (22:6n-3) is among the major long-chain PUFAs of the n-3 family in the pineal of adult rat (8). It has been shown that dietary n-3 fatty acid deprivation effectively reduced the proportion of 22:6n-3 in the total pineal lipid fraction, as well as in the two major phospholipid classes, phosphatidylethanolamine (PE) and phosphatidylcholine (PC) (9). This deficiency led to a decrease of adenosine-stimulated cAMP and melatonin release from cultured rat pineal. In contrast, high levels of 22:6n-3 in the lipid fractions of pinealocyte membranes obtained by feeding adult rats with fish oil concentrates rich in n-3 PUFAs, have been reported to induce a marked increase of MT synthesis in response to NE stimulation (unpublished results). Recently, we showed that in rat pinealocytes, in vitro supplementation by 50 mM 22:6n-3 by inducing a marked increase in the 22:6n-3 levels in phospholipids (1240%) and in triacylglycerols (.1000%) decreased significantly the NE-stimulated cAMP production (10). The first aim of the present study was to examine whether under the same conditions, 22:6n-3 was able to affect the MT production. In addition, we also investigated whether the decreased cAMP production was associated with alteration in phosphodiesterase activities. In fact, it is known that in peripheral blood cells, 22:6n-3 enrichment of cellular phospholipids significantly increases both cAMPand cGMP-phosphodiesterase activities (11). For this reason, the effect of 22:6n-3 on cGMP accumulation was also included in this study. The present results show that 22:6n-3 supplementation is able to lower the MT production without modification of other indol metabolites after NE stimulation. The lowering effect of 22:6 n-3 on cAMP and cGMP accumulations is very likely to be independent on phosphodiesterase activities.
292
Prostaglandins & other Lipid Mediators 1998:55, April
Docosahexaenoic Acid, Nucleotides, and Melatonin: Monia et al. Materials and Methods Chemicals 22:6n-3 was obtained from Sigma-Chimie (L’Isle d’Abeau, France). Its concentration was adjusted after gas-liquid chromatography measurement before each experiment. Pineal cell culture reagents (medium 199HEPES, glutamine, gentamicin, hyaluronidase, DNase), free fatty acid bovine serum albumin (BSA), norepinephrine (NE), 3-isobutyl-1-methylxanthine (IBMX), and prazosin were purchased from Sigma-Chimie (L’Isle d’Abeau, France). Collagenase was obtained from Boehringer-Mannheim (Meylan, France), and horse and fetal calf sera from Flow Laboratories (Mc Lean, VA). [125I]-cAMP and [125I]-cGMP radioimmunoassay kits were purchased from Amersham France (Les Ulis, France). All solvents were of analytical grade and obtained from Merck (Darmstadt, Germany). Animals and Culture of Pinealocytes All experiments were performed in accordance to French (87/848/Ministe`re de l’Agriculture et de la Foreˆt) and European Economic Community (86/609/EEC) guidelines for care and use of laboratory animals and were approved by a regional ethical committee for animal use. Pineal glands were obtained from male Sprague-Dawley rats weighing around 200 g (IFFA CREDO, France). They were kept at least 1 week prior to experimentation under controlled temperature (22 6 1°C) and a 12:12 photoperiod (lights on at 0600), with food and water ad libitum. In all experiments, pineals were collected between 9:00 and 10:00 a.m. and immediately placed in medium 199-HEPES, which was gassed with 95% air, 5% CO2 at 37°C. Pinealocytes were prepared according to the method of Delton et al. (12). Briefly, pineals were dissociated by enzyme dispersion and pipette trituration in medium 199 containing collagenase (1 mg/mL), hyaluronidase (0.5 mg/mL), and DNase (10mg/mL) at 37°C. After the pineals were entirely dissociated, cells were resuspended in medium 199 containing 2 mM glutamicine and 1g/L free fatty acid BSAe (BSA) and plated at density of 2 3 105 cells/well. Cells were incubated at 37°C under a humidified 5% CO2 atmosphere. Docosahexaenoic Supplementation and Stimulation Experiments 22:6n-3 was used at a final concentration of 50 mM in medium after saponification by 5mL K2CO3 in 50 mL ethanol and 5O mL benzene as described by Delton et al. (12) and dilution in medium 199 containing 2.5% serum mixture, 12 mM free fatty acid BSA, 2 mM glutamine and 50 mg/mL gentamicin (ratio albumin to fatty acid 1/2). The cells were centrifuged (500 g, 10 min.) and resuspended either in control medium (without added fatty acid) or in supplemented medium (with added 22: 6n-3) and incubated for 48 h in a humidified atmosphere. The duration of incubation was chosen according to previous data (10). All drugs were
Prostaglandins & other Lipid Mediators 1998:55, April
293
Docosahexaenoic Acid, Nucleotides, and Melatonin: Monia et al. dissolved in water or dimethylsulfoxide and serial dilutions were made in water just before use. The highest concentration of dimethylsulfoxide in the medium (0.1% v/v) did not alter basal levels and had no effect on the magnitude of the adrenergic responses. Aliquots of cells (2 3 105 cells, 500 mL/tube) were exposed to various concentrations of NE. Prazosin (1 mM), an a1 antagonist, or IBMX (1 mM), a phosphodiesterase inhibitor, were preincubated 5 min. before adrenergic agonists. The quantities of cAMP and cGMP synthesized were measured 15 min. after 1 mM NEstimulation or 1 mM isoproterenol, except in time-course experiments, and 5 h after the NE-stimulation (from 0.1 to 10 mM) for melatonin or metabolites. Following treatments, cells were centrifuged (2 min., 3,000 rpm, 4°C), the medium was aspirated and the cell pellets immediately lysed by adding 200 mL perchloric acid 0.4 N and then immediately frozen at 220°C until assays of cAMP and cGMP contents. For the quantification of melatonin and its intermediary metabolites, the medium and cells were immediately frozen at 220°C. Assays of cAMP and cGMP cAMP and cGMP contents in pinealocytes was determined by radioimmunoassay following the non-acetylation (cAMP) or acetylation (cGMP) procedure. Melatonin Assay and Intermediary Metabolites Measurements Melatonin was extracted from 400 mL of cell suspension and quantified after appropriate dilutions, using the radioimmunoassay technique of Brun et al. (13). The simultaneous analysis of 5HT, N-ac-5HT, and 5HIAA was achieved from the remaining cell suspension (100 mL) using HPLC analysis according to the method of Gharib et al. (14) and quantified with external standards. Statistical Analyses Each experiment was performed in nine independent cultures. Data are presented as the mean 6 SEM of the indicated number of determinations as detailed in the legends and were analyzed using an analysis of variance, followed by a Mann-Whitney or Student’s t test.
Results cAMP and cGMP Accumulation In the control pinealocytes, the time course of NE (1 mM) stimulation indicated that the responses in cAMP and cGMP peaked at 5 min. and then declined gradually (Fig. 1A, B). After 1 h, cAMP was still threefold higher than its basal level, whereas cGMP returned to the baseline value in control cells but still remained higher in 22:6n-3 supplemented cells. The supplementation with 22:6n-3 (50 mM) for 48 h significantly reduced
294
Prostaglandins & other Lipid Mediators 1998:55, April
Docosahexaenoic Acid, Nucleotides, and Melatonin: Monia et al.
FIGURE 1. Effect of 22:6n-3 supplementation on time course NE stimulation of cAMP and c GMP accumulation. Pinealocytes (2 3 105 cells) were incubated in control medium (F) or in medium supplemented with 50 mM 22:6n-3 (E) for 48 h as described in Materials and Methods. Pinealocytes were then resuspended in control medium and stimulated with NE 1 mM for the indicated periods; cells were pelleted and cAMP (A) and cGMP (B) were determined by radioimmunoassays. Each value represents the mean 6 SEM of nine individual samples of cells. *(p # 0.05) significantly different from the corresponding control cells.
by 50% the amplitude of the peak of both nucleotides at 5 min. (Fig. 1A, B). Thereafter, the time course of the effects of NE on cAMP and cGMP in 22:6n-3 supplemented pinealocytes showed a slight but not significant decrease between 15 to 60 min. compared to control pinealocytes (Fig. 1A, B). In addition, cAMP and cGMP levels followed a slightly delayed pattern in 22:6n-3 supplemented cells, reaching peak 15–20 min. after NE stimulation and returning to control levels at 60 min. Incubation of control pinealocytes with a high concentration of phosphodiesterase inhibitor IBMX (1 mM) elevated NE-induced cAMP level in control cells (Table 1). The 22:6n-3 supplementation induced a marked decrease (263%) of cAMP. The a1-adrenoceptor antagonist prazosin (1 mM), used to block the a1-adrenoceptor component of NE, caused a significant decrease (273%) in NE-stimulated cAMP level in the control group. In the presence of IBMX, this effect did not become more pronounced in 22:6n-3 supplemented cells (Table 1). In presence of IBMX, the NE-stimulated cGMP in control cells was significantly increased (threefold). The 22:6n-3 supplementation caused a
Prostaglandins & other Lipid Mediators 1998:55, April
295
Docosahexaenoic Acid, Nucleotides, and Melatonin: Monia et al. TABLE 1. Effects of 22:6n-3 on cAMP and cGMP accumulation stimulated by 1 mM NE with and without prazosin (1 mM) in the presence of IBMX (1 mM). AMPc (fmol/200,000 cells)
GMPc (fmol/200,000 cells)
Treatment
Control
122:6
Control
122:6
Basal NE IBMX1NE IBMX1NE1Prazxosin
193 6 43 2524 6 449 11233 6 3143 3008 6 572
111 6 27 1164 6 167* 4058 6 1574* 1790 6 349
54 6 8 148 6 21 596 6 153 395 6 64
53 6 18 153 6 33 327 6 115* 104 6 27*
Pinealocytes (2 3 105 cells) were incubated in control medium or in medium supplemented with 50 mM 22:6n-3 for 48 h as described in Materials and Methods. Pinealocytes were stimulated with different combinations of drugs for 15 min. Each value represents mean 6 SEM for nine individual samples of cells. *(p # 0.05) significantly different from the corresponding control cells.
significant decrease (245%) in this response. This decrease was more pronounced (270%) after treatment by 1 mM prazosin (Table 1). With isoproterenol, 22:6n-3 did not alter the cyclic nucleotide accumulations (data not shown). Effect of 22:6n-3 Supplementation on Melatonin Production in NE Stimulation In the control pinealocytes, NE (0.1, 1, and 10 mM) increased MT production in a dose-dependent manner: the maximal reponse (fourfold increase compared with basal level) was achieved at 1 mM with EC50 approximately equals to 0.5 mM (Fig. 2). The 22– 6n-3 supplementation slightly decreased the basal level of MT secretion from pinealocytes but the statistical analysis could not demonstrate a significant inhibitory effect. The same observation was made with 0.1 mM NE. When the pinealocytes were stimulated with 1 or 10 mM NE, 22:6n-3 similarly decreased (250%) the MT secretion from pinealocytes (Fig. 2). NE also increased the N-ac5HT levels in a concentration-dependent manner in control pinealocytes but 22– 6n-3 did not modify significantly these levels. In addition, low level of MT in 22:6n-3 supplemented cells was not correlated with the metabolism of 5HT, the concentrations of 5HT and 5HIAA being not significantly modified as compared to control cells (data not shown).
Discussion The present study demonstrates for the first time that in vitro 22:6n-3 supplementation decreases MT synthesis after NE stimulation in rat pinealocytes. This lowering effect may be correlated with a decreased
296
Prostaglandins & other Lipid Mediators 1998:55, April
Docosahexaenoic Acid, Nucleotides, and Melatonin: Monia et al.
FIGURE 2. Effect of 22:6n-3 supplementation on NE stimulation of melatonin production. Pinealocytes (2 3 105 cells) were incubated in control medium or in medium supplemented with 50 mM 22:6n-3 for 48 h as described in Materials and Methods. Pinealocytes were then resuspended in control medium and stimulated with gradual concentrations of NE as indicated. Five hours after stimulation, melatonin accumulation was determined in cell suspension by radioimmunoassays. Each value represents the mean 6 SEM of nine individual samples of cells. *(p # 0.05); **(p , 0.01) significantly different from the corresponding control cells.
cAMP level. Our results also show a concomitant decrease of cGMP accumulation in the same conditions. Previously, we have reported that 22:6n-3 supplementation reduced the NE-stimulated cAMP response (10), this effect being obtained 15 min. after NE-stimulation. The time course studies confirm this lowering effect from 5 to 15 min. for cAMP and indicate a similar inhibitory effect for cGMP. Moreover, data obtained in presence of IBMX (1 mM) demonstrated that 22:6n-3 did not act on both cAMP- and cGMP-phosphodiesterase activities in pinealocytes. In contrast, Joulain et al. (11) reported that, in human blood mononuclear cells, 22:6n-3 significantly increased both cAMP-and cGMP-phosphodiesterase activities. Under the present experimental conditions, the lowering effect of 22:6n-3 on cyclic nucleotides in the pinealocytes involved other mechanisms. Usually, NAT activity correlates with the quantity of MT formed and the level of N-Ac-5 HT released into the medium is an indicator of MT production (14). In the 22:6n-3 supplemented cells, we did not observe any changes in the serotonin metabolism, 5HT and 5HIAA concentrations were not significantly different from control. This point explains partially why the amplitude of MT and cAMP inhibition are different and suggests that 22:6 n-3 supplementation may also affect the HIOMT activity. The lowering effect of 22:6n-3 supplementation in vitro on melatonin response was unexpected with regard to in vivo dietary experiments in
Prostaglandins & other Lipid Mediators 1998:55, April
297
Docosahexaenoic Acid, Nucleotides, and Melatonin: Monia et al. which feeding or deprivation 22:6n-3 respectively increases or decreases MT production. These apparent discrepancies suggest that the fatty acid regulation of endocrine activity is dependent on their respective concentration available to the cells and present in distinct compartments. In fact, Gazzah et al. (9) showed a preferential 22:6 n-3 incorporation in the sn 2-position of membrane phospholipids, especially PC and PE, whereas Delton-Vandenbroucke et al. (10), found that in vitro 22:6 n-3 supplementation, at concentration above 25 mM, resulted in a marked increase in the 22:6n-3 levels into phospholipids and triacylglycerols whose mass tended to be enhanced for these latters. In most cases PUFAs employed in vitro at mM concentrations are inhibitory in endocrine cells and the mechanisms by which they exert these actions are varied (15). Our data suggest that changes in the 22:6n-3 content of membrane phospholipids in the vicinity of adrenoceptors affect their coupling with the second messenger systems and are responsible for the observed results but the newly synthesized triacylglycerols cannot be excluded in the MT regulation via other intracellular mechanisms. It is well known that maximal stimulation of pineal cAMP accumulation by NE requires activation of both a1- and b-adrenergic receptors (3,16). In this study we found that: a) 22:6n-3 supplementation decreased significantly the melatonin response for NE concentrations up to 1 mM doses acting on both a1- and b-adrenergic receptors; b) 22:6n-3 supplementation did not modify cAMP accumulation after isoproterenol (1 mM), a dose acting preferentialy on b-adreno receptors (data not shown); and c) the co-treatment prazosin plus IBMX suppressed the inhibitory effect of 22:6n-3. These findings reinforce our hypothesis that in vitro 22:6n-3 supplementation acts on the a1-adrenoceptors sites (10). An additional argument was the uniform decrease of melatonin level (250%) obtained after the stimulation by NE (1 and 10 mM) in 22:6n-3 supplemented cells compared to control group. This effect is one specific for 22:6 n-3 as we found that: 1) 50 mM 22:6 n-3 does not affect pinealocyte viability (10,12); 2) 50 mM 20:5 n-3 supplementation increased cAMP (10); and 3) other essential fatty acids such as 50 mM oleic acid was ineffective on cAMP and MT production (unpublished results). As a final point, it is of interest to consider that the lowering effect of 22:6n-3 supplementation on cGMP accumulation is markedly increased by prazosin pre-treatment after NE-stimulation. Nonetheless, in this study we have not been able to explain such a target for 22:6n-3 in the modulation of cGMP. Based on current thinking about how cGMP production in the pineal gland is regulated (17), we suggest that 22:6 n-3 could act through other steps than a1 receptors. Several candidates may be involved in the regulation of cGMP such as [Ca21]i, Ca21 channels (18,19), and nitric oxide systems (7) (20). In vitro experiments should be useful in examining some of these questions further.
298
Prostaglandins & other Lipid Mediators 1998:55, April
Docosahexaenoic Acid, Nucleotides, and Melatonin: Monia et al. Acknowledgments This work was supported by INSERM, INRA AIP 95/P0009 and grants from B.Q.R Universite´ Claude Bernard Lyon. We thank Prof. M. Lagarde for helpful discussion.
References 1. Reiter, R.J. Pineal melatonin: Cell biology of its synthesis and its physiological interactions. Endocr. Rev. 12:151, 1991. 2. Klein, D.C., Weller, J.L. Rapid light-induced decrease in pineal serotonin N-acetyltransferase activity. Science 177:832, 1972. 3. Klein, D.C., Sugden, D., Weller, J.L. Postsynaptic a-adrenergic receptors potentiate the b-adrenergic stimulation of pineal serotonin N-acetyltransferase. Proc. Natl. Acad. Sci. USA 80:599, 1983. 4. Roseboom, P.H., Klein, D.C. Norepinephrine stimulation of pineal CREM phosphorylation: Primary role of b-adrenergic cyclic AMP mechanism. Mol. Pharmacol. 47:439, 1995. 5. Ehret, M., Pevet, P., Maitre, M. Tryptophan hydroxylase synthesis is induced by 39, 59 adenosine monophosphate during circadian rhythm in the rat pineal gland. J. Neurochem. 57:1516, 1991. 6. Murakami, M., Greer, M.A., Hyulstad, S., Greer, S.E., Tanaka, K. Evidence that rat pineal tyroxine 59deiodinase is primarily stimulated by beta and not alpha-adrenergic agonists and that its adrenergic stimulated and spontaneous rhythmic nocturnal rise require RNA and protein synthesis. Proc. Soc. Exp. Biol. Med. 190:190, 1989. 7. Spessert, R., Layes, E., Vollrath L. Adrenergic stimulation of cyclic GMP formation requires NO-dependent activation of cytosolic guanylate cyclase in rat pinealocytes. J. Neurochem. 61:138, 1993. 8. Sarda, N., Gharib, A., Croset, M., Moliere, P., Lagarde, M. Fatty acid composition of the rat pineal gland. Dietary modifications. Biochim. Biophys. Acta 1081:75, 1991. 9. Gazzah, N., Gharib, A., Delton, I., et al. Effect of an n-3 fatty acid-deficient diet on the adenosine-dependent melatonin release in cultured rat pineal. J. Neurochem. 61:1057, 1993. 10. Delton-Vandenbroucke, I., Sarda, N., Moliere, P., Lagarde, M., Gharib, A. Modulation of norepinephrine-stimulated cyclic AMP accumulation in rat pinealocytes by n-3 fatty acids. Eur. J. Pharmacol. 312:379, 1996. 11. Joulain, C., Guichardant, M., Lagarde, M., Prigent, A.F. Influence of polyunsaturated fatty acids on lipid metabolism in human blood mononuclear cells and early biochemical events associated with lymphocytes activation. J. Lipid Mediators Cell Signalling 11:63, 1995. 12. Delton, I., Gharib, A., Moliere, P., Lagarde, M., Sarda, N. Distribution and metabolism of arachidonic and docosahexaenoic acids in rat pineal cells. Effect of norepinephrine. Biochim. Biophys. Acta 1254:147, 1995. 13. Brun, J., Claustrat, B., Harthe, C., Vitte, P.A., Cohen, R., Chazot, G. Melatonin-RIA analytical and physiological criteria of validity. Adv. Biosci. 53:41, 1985. 14. Gharib, A., Reynaud, D., Sarda, N., Vivien-Roels, B., Pevet, P., Pacheco, H.
Prostaglandins & other Lipid Mediators 1998:55, April
299
Docosahexaenoic Acid, Nucleotides, and Melatonin: Monia et al.
15.
16.
17.
18.
19.
20.
Adenosine analogs elevate N-acetyl-serotonin and melatonin in rat pineal gland. Adv. Pineal Res. 5:47, 1989. Widmaier, E.P. In: Handbook of Essential Fatty Acid Biology. Biochemistry, Physiology and Behavioral Neurobiology. (S. Yehuda and D.I. Mostofsky, eds.) Humana Press, Totowa, New Jersey, 1997, p. 115. Vanecek, J., Sugden D., Weller, J., Klein, D.C. Atypical synergistic a1- and b-adrenergic regulation of adenosine 39,59-monophosphate and guanosine 39,59-monophosphate in rat pinealocytes. Endocrinology 116:2167, 1985. Breittmayer, J.P., Pelassy, C., Cousin, J.L., Bernard, A., Aussel, C. The inhibition by fatty acids of receptor-mediated calcium movements in Jurkat T-cells is due to increase calcium extrussion. J. Biol. Chem. 268:20812, 1993. Sugden, A.L., Sugden, D., Klein, D.C. Essential role of calcium influx in the adrenergic stimulation of cAMP and cGMP in rat pinealocytes. J. Biol. Chem. 261:11608, 1986. White, B.H., Klein, D.C. Stimulation of cyclic GMP accumulation by sodium nitroprusside is potentiated via a G(s) mechanism in intact pinealocytes. J. Neurochem. 64:711, 1995. Guerrero, J.M., Reiter, R.J., Poeggeler, B., Chen, L.D., Tan, D.X. Elevation of cyclic GMP levels in the rat pineal gland induced by nitric oxide. J. Pineal Res. 16:210, 1994.
Editor: Dr. B. Vargaftig
300
Recieved: 11–27-97
Accepted: 02– 02-98
Prostaglandins & other Lipid Mediators 1998:55, April
Introduction The pineal hormone melatonin (MT) plays a central role in controlling seasonal reproduction and other rhythmic events (1). In mammalian 1 Address correspondence to: N. Sarda, Laboratoire de Neuropharmacologie Mole´culaire, CNRS, UMR 5542, Faculte´ de Me´decine Lae¨nnec, rue Guillaume Paradin, 69007 Lyon, France; Phone: (33) 04 78 77 87 81; Fax: (33) 04 78 77 87 32.
Prostaglandins & other Lipid Mediators 55:291–300, 1998 © 1998 by Elsevier Science Inc. 655 Avenue of the Americas, New York, NY 10010
0090-6980/98/$19.00 PII S0090-6980(98)00028-8
Docosahexaenoic Acid, Nucleotides, and Melatonin: Monia et al. species, MT is synthesized nocturnally in the pinealocytes by N-acetylation and subsequent O-methylation of serotonin (5HT) after noradrenergic stimulation (2). The limiting factors for MT synthesis may be the substrate availability (i.e., 5HT), the cAMP level, the activity of the arylalkylamine-N-acetyltransferase (NAT), or the second enzyme (hydroxyindol-O-methyltransferase, HIOMT) involved in MT synthesis. In rat, norepinephrine (NE) increases cAMP and cGMP levels through an action involving both a1- and b-adrenoceptors and increases [Ca21]i and inositol phosphate turnover through a1-adrenergic mechanisms (3,4). The physiological role of cAMP is well known. cAMP participates in the stimulation of three enzymes, i.e., NAT, which is the rate-controlling enzyme in the pathway that converts 5HT to MT (2), tryptophan hydroxylase (5) and thyroxine type II 59 deiodinase (6). In contrast, the functional significance of cGMP remains still obscur. Recently, data have shown an involvement of NO in cGMP synthesis in response to NE in dissociated pinealocytes (7). At the same time, other studies have provided evidence for the role of polyunsaturated fatty acids (PUFAs) in pineal activity. Docosahexaenoic acid (22:6n-3) is among the major long-chain PUFAs of the n-3 family in the pineal of adult rat (8). It has been shown that dietary n-3 fatty acid deprivation effectively reduced the proportion of 22:6n-3 in the total pineal lipid fraction, as well as in the two major phospholipid classes, phosphatidylethanolamine (PE) and phosphatidylcholine (PC) (9). This deficiency led to a decrease of adenosine-stimulated cAMP and melatonin release from cultured rat pineal. In contrast, high levels of 22:6n-3 in the lipid fractions of pinealocyte membranes obtained by feeding adult rats with fish oil concentrates rich in n-3 PUFAs, have been reported to induce a marked increase of MT synthesis in response to NE stimulation (unpublished results). Recently, we showed that in rat pinealocytes, in vitro supplementation by 50 mM 22:6n-3 by inducing a marked increase in the 22:6n-3 levels in phospholipids (1240%) and in triacylglycerols (.1000%) decreased significantly the NE-stimulated cAMP production (10). The first aim of the present study was to examine whether under the same conditions, 22:6n-3 was able to affect the MT production. In addition, we also investigated whether the decreased cAMP production was associated with alteration in phosphodiesterase activities. In fact, it is known that in peripheral blood cells, 22:6n-3 enrichment of cellular phospholipids significantly increases both cAMPand cGMP-phosphodiesterase activities (11). For this reason, the effect of 22:6n-3 on cGMP accumulation was also included in this study. The present results show that 22:6n-3 supplementation is able to lower the MT production without modification of other indol metabolites after NE stimulation. The lowering effect of 22:6 n-3 on cAMP and cGMP accumulations is very likely to be independent on phosphodiesterase activities.
292
Prostaglandins & other Lipid Mediators 1998:55, April
Docosahexaenoic Acid, Nucleotides, and Melatonin: Monia et al. Materials and Methods Chemicals 22:6n-3 was obtained from Sigma-Chimie (L’Isle d’Abeau, France). Its concentration was adjusted after gas-liquid chromatography measurement before each experiment. Pineal cell culture reagents (medium 199HEPES, glutamine, gentamicin, hyaluronidase, DNase), free fatty acid bovine serum albumin (BSA), norepinephrine (NE), 3-isobutyl-1-methylxanthine (IBMX), and prazosin were purchased from Sigma-Chimie (L’Isle d’Abeau, France). Collagenase was obtained from Boehringer-Mannheim (Meylan, France), and horse and fetal calf sera from Flow Laboratories (Mc Lean, VA). [125I]-cAMP and [125I]-cGMP radioimmunoassay kits were purchased from Amersham France (Les Ulis, France). All solvents were of analytical grade and obtained from Merck (Darmstadt, Germany). Animals and Culture of Pinealocytes All experiments were performed in accordance to French (87/848/Ministe`re de l’Agriculture et de la Foreˆt) and European Economic Community (86/609/EEC) guidelines for care and use of laboratory animals and were approved by a regional ethical committee for animal use. Pineal glands were obtained from male Sprague-Dawley rats weighing around 200 g (IFFA CREDO, France). They were kept at least 1 week prior to experimentation under controlled temperature (22 6 1°C) and a 12:12 photoperiod (lights on at 0600), with food and water ad libitum. In all experiments, pineals were collected between 9:00 and 10:00 a.m. and immediately placed in medium 199-HEPES, which was gassed with 95% air, 5% CO2 at 37°C. Pinealocytes were prepared according to the method of Delton et al. (12). Briefly, pineals were dissociated by enzyme dispersion and pipette trituration in medium 199 containing collagenase (1 mg/mL), hyaluronidase (0.5 mg/mL), and DNase (10mg/mL) at 37°C. After the pineals were entirely dissociated, cells were resuspended in medium 199 containing 2 mM glutamicine and 1g/L free fatty acid BSAe (BSA) and plated at density of 2 3 105 cells/well. Cells were incubated at 37°C under a humidified 5% CO2 atmosphere. Docosahexaenoic Supplementation and Stimulation Experiments 22:6n-3 was used at a final concentration of 50 mM in medium after saponification by 5mL K2CO3 in 50 mL ethanol and 5O mL benzene as described by Delton et al. (12) and dilution in medium 199 containing 2.5% serum mixture, 12 mM free fatty acid BSA, 2 mM glutamine and 50 mg/mL gentamicin (ratio albumin to fatty acid 1/2). The cells were centrifuged (500 g, 10 min.) and resuspended either in control medium (without added fatty acid) or in supplemented medium (with added 22: 6n-3) and incubated for 48 h in a humidified atmosphere. The duration of incubation was chosen according to previous data (10). All drugs were
Prostaglandins & other Lipid Mediators 1998:55, April
293
Docosahexaenoic Acid, Nucleotides, and Melatonin: Monia et al. dissolved in water or dimethylsulfoxide and serial dilutions were made in water just before use. The highest concentration of dimethylsulfoxide in the medium (0.1% v/v) did not alter basal levels and had no effect on the magnitude of the adrenergic responses. Aliquots of cells (2 3 105 cells, 500 mL/tube) were exposed to various concentrations of NE. Prazosin (1 mM), an a1 antagonist, or IBMX (1 mM), a phosphodiesterase inhibitor, were preincubated 5 min. before adrenergic agonists. The quantities of cAMP and cGMP synthesized were measured 15 min. after 1 mM NEstimulation or 1 mM isoproterenol, except in time-course experiments, and 5 h after the NE-stimulation (from 0.1 to 10 mM) for melatonin or metabolites. Following treatments, cells were centrifuged (2 min., 3,000 rpm, 4°C), the medium was aspirated and the cell pellets immediately lysed by adding 200 mL perchloric acid 0.4 N and then immediately frozen at 220°C until assays of cAMP and cGMP contents. For the quantification of melatonin and its intermediary metabolites, the medium and cells were immediately frozen at 220°C. Assays of cAMP and cGMP cAMP and cGMP contents in pinealocytes was determined by radioimmunoassay following the non-acetylation (cAMP) or acetylation (cGMP) procedure. Melatonin Assay and Intermediary Metabolites Measurements Melatonin was extracted from 400 mL of cell suspension and quantified after appropriate dilutions, using the radioimmunoassay technique of Brun et al. (13). The simultaneous analysis of 5HT, N-ac-5HT, and 5HIAA was achieved from the remaining cell suspension (100 mL) using HPLC analysis according to the method of Gharib et al. (14) and quantified with external standards. Statistical Analyses Each experiment was performed in nine independent cultures. Data are presented as the mean 6 SEM of the indicated number of determinations as detailed in the legends and were analyzed using an analysis of variance, followed by a Mann-Whitney or Student’s t test.
Results cAMP and cGMP Accumulation In the control pinealocytes, the time course of NE (1 mM) stimulation indicated that the responses in cAMP and cGMP peaked at 5 min. and then declined gradually (Fig. 1A, B). After 1 h, cAMP was still threefold higher than its basal level, whereas cGMP returned to the baseline value in control cells but still remained higher in 22:6n-3 supplemented cells. The supplementation with 22:6n-3 (50 mM) for 48 h significantly reduced
294
Prostaglandins & other Lipid Mediators 1998:55, April
Docosahexaenoic Acid, Nucleotides, and Melatonin: Monia et al.
FIGURE 1. Effect of 22:6n-3 supplementation on time course NE stimulation of cAMP and c GMP accumulation. Pinealocytes (2 3 105 cells) were incubated in control medium (F) or in medium supplemented with 50 mM 22:6n-3 (E) for 48 h as described in Materials and Methods. Pinealocytes were then resuspended in control medium and stimulated with NE 1 mM for the indicated periods; cells were pelleted and cAMP (A) and cGMP (B) were determined by radioimmunoassays. Each value represents the mean 6 SEM of nine individual samples of cells. *(p # 0.05) significantly different from the corresponding control cells.
by 50% the amplitude of the peak of both nucleotides at 5 min. (Fig. 1A, B). Thereafter, the time course of the effects of NE on cAMP and cGMP in 22:6n-3 supplemented pinealocytes showed a slight but not significant decrease between 15 to 60 min. compared to control pinealocytes (Fig. 1A, B). In addition, cAMP and cGMP levels followed a slightly delayed pattern in 22:6n-3 supplemented cells, reaching peak 15–20 min. after NE stimulation and returning to control levels at 60 min. Incubation of control pinealocytes with a high concentration of phosphodiesterase inhibitor IBMX (1 mM) elevated NE-induced cAMP level in control cells (Table 1). The 22:6n-3 supplementation induced a marked decrease (263%) of cAMP. The a1-adrenoceptor antagonist prazosin (1 mM), used to block the a1-adrenoceptor component of NE, caused a significant decrease (273%) in NE-stimulated cAMP level in the control group. In the presence of IBMX, this effect did not become more pronounced in 22:6n-3 supplemented cells (Table 1). In presence of IBMX, the NE-stimulated cGMP in control cells was significantly increased (threefold). The 22:6n-3 supplementation caused a
Prostaglandins & other Lipid Mediators 1998:55, April
295
Docosahexaenoic Acid, Nucleotides, and Melatonin: Monia et al. TABLE 1. Effects of 22:6n-3 on cAMP and cGMP accumulation stimulated by 1 mM NE with and without prazosin (1 mM) in the presence of IBMX (1 mM). AMPc (fmol/200,000 cells)
GMPc (fmol/200,000 cells)
Treatment
Control
122:6
Control
122:6
Basal NE IBMX1NE IBMX1NE1Prazxosin
193 6 43 2524 6 449 11233 6 3143 3008 6 572
111 6 27 1164 6 167* 4058 6 1574* 1790 6 349
54 6 8 148 6 21 596 6 153 395 6 64
53 6 18 153 6 33 327 6 115* 104 6 27*
Pinealocytes (2 3 105 cells) were incubated in control medium or in medium supplemented with 50 mM 22:6n-3 for 48 h as described in Materials and Methods. Pinealocytes were stimulated with different combinations of drugs for 15 min. Each value represents mean 6 SEM for nine individual samples of cells. *(p # 0.05) significantly different from the corresponding control cells.
significant decrease (245%) in this response. This decrease was more pronounced (270%) after treatment by 1 mM prazosin (Table 1). With isoproterenol, 22:6n-3 did not alter the cyclic nucleotide accumulations (data not shown). Effect of 22:6n-3 Supplementation on Melatonin Production in NE Stimulation In the control pinealocytes, NE (0.1, 1, and 10 mM) increased MT production in a dose-dependent manner: the maximal reponse (fourfold increase compared with basal level) was achieved at 1 mM with EC50 approximately equals to 0.5 mM (Fig. 2). The 22– 6n-3 supplementation slightly decreased the basal level of MT secretion from pinealocytes but the statistical analysis could not demonstrate a significant inhibitory effect. The same observation was made with 0.1 mM NE. When the pinealocytes were stimulated with 1 or 10 mM NE, 22:6n-3 similarly decreased (250%) the MT secretion from pinealocytes (Fig. 2). NE also increased the N-ac5HT levels in a concentration-dependent manner in control pinealocytes but 22– 6n-3 did not modify significantly these levels. In addition, low level of MT in 22:6n-3 supplemented cells was not correlated with the metabolism of 5HT, the concentrations of 5HT and 5HIAA being not significantly modified as compared to control cells (data not shown).
Discussion The present study demonstrates for the first time that in vitro 22:6n-3 supplementation decreases MT synthesis after NE stimulation in rat pinealocytes. This lowering effect may be correlated with a decreased
296
Prostaglandins & other Lipid Mediators 1998:55, April
Docosahexaenoic Acid, Nucleotides, and Melatonin: Monia et al.
FIGURE 2. Effect of 22:6n-3 supplementation on NE stimulation of melatonin production. Pinealocytes (2 3 105 cells) were incubated in control medium or in medium supplemented with 50 mM 22:6n-3 for 48 h as described in Materials and Methods. Pinealocytes were then resuspended in control medium and stimulated with gradual concentrations of NE as indicated. Five hours after stimulation, melatonin accumulation was determined in cell suspension by radioimmunoassays. Each value represents the mean 6 SEM of nine individual samples of cells. *(p # 0.05); **(p , 0.01) significantly different from the corresponding control cells.
cAMP level. Our results also show a concomitant decrease of cGMP accumulation in the same conditions. Previously, we have reported that 22:6n-3 supplementation reduced the NE-stimulated cAMP response (10), this effect being obtained 15 min. after NE-stimulation. The time course studies confirm this lowering effect from 5 to 15 min. for cAMP and indicate a similar inhibitory effect for cGMP. Moreover, data obtained in presence of IBMX (1 mM) demonstrated that 22:6n-3 did not act on both cAMP- and cGMP-phosphodiesterase activities in pinealocytes. In contrast, Joulain et al. (11) reported that, in human blood mononuclear cells, 22:6n-3 significantly increased both cAMP-and cGMP-phosphodiesterase activities. Under the present experimental conditions, the lowering effect of 22:6n-3 on cyclic nucleotides in the pinealocytes involved other mechanisms. Usually, NAT activity correlates with the quantity of MT formed and the level of N-Ac-5 HT released into the medium is an indicator of MT production (14). In the 22:6n-3 supplemented cells, we did not observe any changes in the serotonin metabolism, 5HT and 5HIAA concentrations were not significantly different from control. This point explains partially why the amplitude of MT and cAMP inhibition are different and suggests that 22:6 n-3 supplementation may also affect the HIOMT activity. The lowering effect of 22:6n-3 supplementation in vitro on melatonin response was unexpected with regard to in vivo dietary experiments in
Prostaglandins & other Lipid Mediators 1998:55, April
297
Docosahexaenoic Acid, Nucleotides, and Melatonin: Monia et al. which feeding or deprivation 22:6n-3 respectively increases or decreases MT production. These apparent discrepancies suggest that the fatty acid regulation of endocrine activity is dependent on their respective concentration available to the cells and present in distinct compartments. In fact, Gazzah et al. (9) showed a preferential 22:6 n-3 incorporation in the sn 2-position of membrane phospholipids, especially PC and PE, whereas Delton-Vandenbroucke et al. (10), found that in vitro 22:6 n-3 supplementation, at concentration above 25 mM, resulted in a marked increase in the 22:6n-3 levels into phospholipids and triacylglycerols whose mass tended to be enhanced for these latters. In most cases PUFAs employed in vitro at mM concentrations are inhibitory in endocrine cells and the mechanisms by which they exert these actions are varied (15). Our data suggest that changes in the 22:6n-3 content of membrane phospholipids in the vicinity of adrenoceptors affect their coupling with the second messenger systems and are responsible for the observed results but the newly synthesized triacylglycerols cannot be excluded in the MT regulation via other intracellular mechanisms. It is well known that maximal stimulation of pineal cAMP accumulation by NE requires activation of both a1- and b-adrenergic receptors (3,16). In this study we found that: a) 22:6n-3 supplementation decreased significantly the melatonin response for NE concentrations up to 1 mM doses acting on both a1- and b-adrenergic receptors; b) 22:6n-3 supplementation did not modify cAMP accumulation after isoproterenol (1 mM), a dose acting preferentialy on b-adreno receptors (data not shown); and c) the co-treatment prazosin plus IBMX suppressed the inhibitory effect of 22:6n-3. These findings reinforce our hypothesis that in vitro 22:6n-3 supplementation acts on the a1-adrenoceptors sites (10). An additional argument was the uniform decrease of melatonin level (250%) obtained after the stimulation by NE (1 and 10 mM) in 22:6n-3 supplemented cells compared to control group. This effect is one specific for 22:6 n-3 as we found that: 1) 50 mM 22:6 n-3 does not affect pinealocyte viability (10,12); 2) 50 mM 20:5 n-3 supplementation increased cAMP (10); and 3) other essential fatty acids such as 50 mM oleic acid was ineffective on cAMP and MT production (unpublished results). As a final point, it is of interest to consider that the lowering effect of 22:6n-3 supplementation on cGMP accumulation is markedly increased by prazosin pre-treatment after NE-stimulation. Nonetheless, in this study we have not been able to explain such a target for 22:6n-3 in the modulation of cGMP. Based on current thinking about how cGMP production in the pineal gland is regulated (17), we suggest that 22:6 n-3 could act through other steps than a1 receptors. Several candidates may be involved in the regulation of cGMP such as [Ca21]i, Ca21 channels (18,19), and nitric oxide systems (7) (20). In vitro experiments should be useful in examining some of these questions further.
298
Prostaglandins & other Lipid Mediators 1998:55, April
Docosahexaenoic Acid, Nucleotides, and Melatonin: Monia et al. Acknowledgments This work was supported by INSERM, INRA AIP 95/P0009 and grants from B.Q.R Universite´ Claude Bernard Lyon. We thank Prof. M. Lagarde for helpful discussion.
References 1. Reiter, R.J. Pineal melatonin: Cell biology of its synthesis and its physiological interactions. Endocr. Rev. 12:151, 1991. 2. Klein, D.C., Weller, J.L. Rapid light-induced decrease in pineal serotonin N-acetyltransferase activity. Science 177:832, 1972. 3. Klein, D.C., Sugden, D., Weller, J.L. Postsynaptic a-adrenergic receptors potentiate the b-adrenergic stimulation of pineal serotonin N-acetyltransferase. Proc. Natl. Acad. Sci. USA 80:599, 1983. 4. Roseboom, P.H., Klein, D.C. Norepinephrine stimulation of pineal CREM phosphorylation: Primary role of b-adrenergic cyclic AMP mechanism. Mol. Pharmacol. 47:439, 1995. 5. Ehret, M., Pevet, P., Maitre, M. Tryptophan hydroxylase synthesis is induced by 39, 59 adenosine monophosphate during circadian rhythm in the rat pineal gland. J. Neurochem. 57:1516, 1991. 6. Murakami, M., Greer, M.A., Hyulstad, S., Greer, S.E., Tanaka, K. Evidence that rat pineal tyroxine 59deiodinase is primarily stimulated by beta and not alpha-adrenergic agonists and that its adrenergic stimulated and spontaneous rhythmic nocturnal rise require RNA and protein synthesis. Proc. Soc. Exp. Biol. Med. 190:190, 1989. 7. Spessert, R., Layes, E., Vollrath L. Adrenergic stimulation of cyclic GMP formation requires NO-dependent activation of cytosolic guanylate cyclase in rat pinealocytes. J. Neurochem. 61:138, 1993. 8. Sarda, N., Gharib, A., Croset, M., Moliere, P., Lagarde, M. Fatty acid composition of the rat pineal gland. Dietary modifications. Biochim. Biophys. Acta 1081:75, 1991. 9. Gazzah, N., Gharib, A., Delton, I., et al. Effect of an n-3 fatty acid-deficient diet on the adenosine-dependent melatonin release in cultured rat pineal. J. Neurochem. 61:1057, 1993. 10. Delton-Vandenbroucke, I., Sarda, N., Moliere, P., Lagarde, M., Gharib, A. Modulation of norepinephrine-stimulated cyclic AMP accumulation in rat pinealocytes by n-3 fatty acids. Eur. J. Pharmacol. 312:379, 1996. 11. Joulain, C., Guichardant, M., Lagarde, M., Prigent, A.F. Influence of polyunsaturated fatty acids on lipid metabolism in human blood mononuclear cells and early biochemical events associated with lymphocytes activation. J. Lipid Mediators Cell Signalling 11:63, 1995. 12. Delton, I., Gharib, A., Moliere, P., Lagarde, M., Sarda, N. Distribution and metabolism of arachidonic and docosahexaenoic acids in rat pineal cells. Effect of norepinephrine. Biochim. Biophys. Acta 1254:147, 1995. 13. Brun, J., Claustrat, B., Harthe, C., Vitte, P.A., Cohen, R., Chazot, G. Melatonin-RIA analytical and physiological criteria of validity. Adv. Biosci. 53:41, 1985. 14. Gharib, A., Reynaud, D., Sarda, N., Vivien-Roels, B., Pevet, P., Pacheco, H.
Prostaglandins & other Lipid Mediators 1998:55, April
299
Docosahexaenoic Acid, Nucleotides, and Melatonin: Monia et al.
15.
16.
17.
18.
19.
20.
Adenosine analogs elevate N-acetyl-serotonin and melatonin in rat pineal gland. Adv. Pineal Res. 5:47, 1989. Widmaier, E.P. In: Handbook of Essential Fatty Acid Biology. Biochemistry, Physiology and Behavioral Neurobiology. (S. Yehuda and D.I. Mostofsky, eds.) Humana Press, Totowa, New Jersey, 1997, p. 115. Vanecek, J., Sugden D., Weller, J., Klein, D.C. Atypical synergistic a1- and b-adrenergic regulation of adenosine 39,59-monophosphate and guanosine 39,59-monophosphate in rat pinealocytes. Endocrinology 116:2167, 1985. Breittmayer, J.P., Pelassy, C., Cousin, J.L., Bernard, A., Aussel, C. The inhibition by fatty acids of receptor-mediated calcium movements in Jurkat T-cells is due to increase calcium extrussion. J. Biol. Chem. 268:20812, 1993. Sugden, A.L., Sugden, D., Klein, D.C. Essential role of calcium influx in the adrenergic stimulation of cAMP and cGMP in rat pinealocytes. J. Biol. Chem. 261:11608, 1986. White, B.H., Klein, D.C. Stimulation of cyclic GMP accumulation by sodium nitroprusside is potentiated via a G(s) mechanism in intact pinealocytes. J. Neurochem. 64:711, 1995. Guerrero, J.M., Reiter, R.J., Poeggeler, B., Chen, L.D., Tan, D.X. Elevation of cyclic GMP levels in the rat pineal gland induced by nitric oxide. J. Pineal Res. 16:210, 1994.
Editor: Dr. B. Vargaftig
300
Recieved: 11–27-97
Accepted: 02– 02-98
Prostaglandins & other Lipid Mediators 1998:55, April