- Email: [email protected]
Pineal perifusion with calcium channel blockers inhibits differently daytime and nighttime melatonin production in rat
Aokular and Zelluiar !ndocrindo9y
ELSEVIER
Molecular and Cellular Endocrinology lOl(1994) 189-196
Pineal perifusion with calcium channel blockers inhibits differently daytime and nighttime melatonin production in rat Zi-Yan Zhao, Yvan Touitou * Department of Biochemistry, Faculte’ de Mkdecine Pit&Salphri&e,
Boulevard de I’HGpital91, 75013 Paris, France
(Received 1 November 1993; accepted 20 December 1993)
Abstract In a previous study we have shown that the response of perifused pineal glands to calcium was different according to the circadian stage at which the glands were removed. This difference may be explained by circadian changes in calcium channel function. Therefore in the present study we documented the effects of calcium channel blockers in perifused rat pineal glands removed in the middle of the light and dark spans (7 and 19 HALO (hours after light onset), in a L/D 12: 12 regimen). Moreover, we have studied the effect of calcium channel blockers on adrenergically stimulated pineal glands removed 7 HALO. Inorganic (Co*+ and Cd*+) and organic (nifedipine and diltiazeml calcium channel blockers at 1O-4 mol/l all significantly reduced melatonin production and this inhibition was more effective with the glands removed 7 HALO. In a concentration of 10m5 mol/l, only Cd*+ and diltiazem reduced melatonin production significantly in pineal glands removed 7 HALO. Verapamil at 10m4 and 10e5 mol/l showed no significant effect on melatonin production in glands removed both during the light and dark spans. Mn*+ at 10e4 mol/l (but not at lop5 mol/l) appeared to stimulate melatonin production in glands removed both during the light and the dark (significant increase only with glands removed during the dark). Cobalt showed an immediate short inhibitory effect on both isoproterenol and norepinephrine-stimulated melatonin release, whereas nifedipine showed a significant inhibition only on isoproterenol-stimulated melatonin release. These results strongly suggest a circadian stage dependence of the pineal gland response to some calcium channel blockers and the involvement of calcium in the release of melatonin from pinealocytes. Key words:
Melatonin;
(Rat pineal gland); Calcium channel blocker; Circadian rhythm; Adrenergic
1. Introduction
Calcium ions play important roles in regulating a wide variety of cellular functions, among which the pineal gland function. In the pineal gland the sensitivity of P-receptors is dependent upon calcium ions (Wilkinson, 1978) and calcium influx is essential for norepinephrine to increase CAMP and cGMP (Sugden et al., 1986); besides, the activation of ar-adrenoceptors increases cytosolic calcium concentration (Sugden et al., 1987). In addition, calcium was shown to be necessary for the induction of serotonin N-acetyltransferase @NAT) activity (Zatz and Romero, 1978).
* Corresponding author. Tel.: (33 1) 40 77 96 63; Fax: (33 1) 40 77 96 65. Elsevier Science Ireland Ltd. SSDI 0303-7207(94)00013-Y
stimulation
In a previous study we have found that the response of perifused pineal glands to calcium was different according to the circadian stage at which the glands were removed, i.e., increased calcium concentration (within the physiological range) in the perifusion system stimulated melatonin production by pineal glands removed from rats killed in the middle of the dark period (19 h after light onset, HALO; L/D = 12: 12) but not by those removed in the middle of the light period (7 HALO) (Zhao and Touitou, 1993a). A possible explanation of this difference may be the circadian changes in calcium channel function. Therefore, we studied the effects of calcium channel blockers on melatonin production by perifused rat pineal glands removed in the middle of the light and dark spans (7 and 19 HALO) in order to document a possible circadian stage dependence of their effects which could support possible
190
Z.-Y. Zhao. Y. Touitou /Moleculur
changes in calcium channel function. These were examined on basal and on adrenergically lated pineal melatonin production.
and Cellular Endocrinology
effects stimu-
2. Materials and methods 2.1. Housirzg and ~~c~ro~~zation of rats Male albino Wistar rats (IFFA-CREDO, L’Arbresle, France), 4-5 weeks old (loo-120 g) on arrival at the laboratory, were housed in a chronobiologic animal facility (Enceinte Autonome d’Animalerie, Ref. E 110SP-6, ESI Flufrance, Arcueil, France) with food and water ad libitum. The rats were synchronised with a lighting regimen (L/D = 12: 12; the light intensity in the cages was 3.1-3.8 pW/cm’) for three weeks before each experiment. The chronobiologic facility was equipped with equispaced, sound-proof, temperaturecontrolled (21 f l.O’C) compartments provided with independent light-dark cycles. The 12 h of light span started in different compartments at 03 and 15 h (clock time), therefore the removal of pineal glands at 10 h resulted in glands obtained 7 and 19 HALO, i.e., in the middle of the light and dark spans, respectively. The rats were killed by decapitation at the age of 7-8 weeks (about 250 g) after 3 weeks’ synchronization in the chronobiologic facility. Decapitation of rats during the dark period (19 HALO) was carried out under a red light (0.4 pW/cm’ at the level of worktable) and each rat was exposed to the dim red light for less than 20 s to avoid alteration of melatonin production. The pineal glands were quickly removed and kept in an oxygenated Krebs-Ringer solution in an iced water bath until transferred into the perifusion chambers. 2.2. ~~emicats
IO1 (1994) 189-196
the temperature at 37 f 0.5”C, and a peristaltic pump (Ismatec IPN-12). The column and pistons delimited a 0.4 cm high space (perifusion chamber with a volume of 0.2 ml) which was filled with Krebs-Ringer solution (pH 7.4, gassed continuously with 95% O,-5% CO,). Pineal glands (two per chamber) were perifused with Krebs-Ringer solution at a constant flow rate of 0.2 ml/min. We had previously established (Zhao and Touitou, 1993b) that rat pineal glands in perifusion stabilize their melatonin production 3-4 h after the beginning of the perifusion, then melatonin production remains stable up to 8 h. Therefore in this study the effluent perifusate fractions were collected in glass tubes every 10 min from 3 h 30 min (210 min) to 8 h 30 min (5 IO min) of perifusion and stored at - 20°C until assayed for melatonin. After the first 4 fractions were collected (to calculate the basal level), the pineals were perifused in two sets of experiments: in the first set the calcium channel blockers were perifused for 30 min (250-280 min of the perifusion); in the second set the calcium channel blocker perifusion was preceded by an adrenergic stimulation by either isoproterenol or norepinephrine (adrenergic stimulation: 250-280 min; calcium channel blocker: 340-370 min). 2.4. Meltstonin radioimmunoassay Melatonin concentration in the collected fractions was measured directly by a modification of the RIA method of Fraser et al. (19831, using a 1251-melatonin tracer. The assay sensitivity was 5-10 pg/ml, the only significant (0.1%) cross-reactant was Ghydroxymelatonin (Ravault et al., 1989). The intra-assay and inter-assay coefficients of variation were 9% (N = 20). 2.5. Statistical analysis
Melatonin was purchased from Fluka (Buchs, 1251-melatonin from Dositek (Orsay, Switzerland), France), rabbit anti-melatonin serum and anti-immunoglobulin serum from INRA (Tours, France). The inorganic calcium channel blockers (cobalt chloride, cadmium chloride and manganese sulphate) were the generous gifts of the Pharmacy of PitiC-Salpetriere Hospital, Paris. Nifedipine was obtained from Bayer Pharma laboratory (Paris), diltiazem from SynthClabo (Bagneux, France) verapamil from Biosedra (Malakoff, France), (-Ksoproterenol from Sigma and (It)norepinephrine from Aldrich (Saint Quentin, France).
All the results obtained were expressed as means it S.E.M. To evaluate the effects of the calcium channel blockers, the melatonin production in any chamber was expressed as a percentage of the mean concentration (considered as a “basal level”) in the first 4 fractions (220, 230, 240 and 250 min) collected just before the introduction of drugs in the perifusion system. Student’s unpaired t-test was used to analyse the experimental data.
2.3. Experimental procedure
The melatonin production in perifused rat pineal glands reached a stable level after 3-4 h perifusion, and then remained fairly constant and roughly similar up to 8 h perifusion whatever the circadian stage (Zhao and Touitou, 1993). We therefore considered the mean
The perifusion system consisted of a plastic column (diameter 0.8 x 3.0 cm) closed with two pistons, a thermostatic bath (Techne TE-8J) which maintained
3. Resutts
Z.-Y. Zhao, Y. Touitou /Molecular
7 HALO
200
z a
191
and Cellular Endocrinology 101 (1994) 189-196
19 HALO
2OOf
1 ‘50-
IO-4moin
co*+
f-I
j B s _g 9 m B I
IOO-
&-m-
50-
1
04 210
z 270
o+ 210
270
.
’ 330
’ 390
’
*
330
390
450
5
450
P < 0.001
1 510
04 210
270
330
’
210
270
330
510
=
’
’
390
450
510
390
450
510
min
Fig. 1. Effect of cobalt on melatonin production in perifused rat pineal glands removed from rats killed 7 and 19 HALO (hours after light onset). Each point is the mean f SEM of data obtained in 3 chambers of perifusion (2 pineal glands per chamber).
The basal level of melatonin production in this study was 20.8 it 5.7 and 19.0 li: 5.9 pg melatonin/minjgland in perifused pineal glands removed 7 and 19 HALO, respectively (mean f SD of 56 and 54 chambers of perifusion, respectively). In another experiment, the perifused pineal glands were removed 7 HALO and isoproterenol or nore-
concentration of melatonin obtained between 210 and 250 min perifusion as a baseline. Two concentrations (10m5 and 10F4 mol/l) of calcium channel blockers (cobalt, cadmium, manganese, nifedipine, diltiazem, and verapamil) were introduced respectively in the perifusion chambers for 30 min perifusion.
5 5
7 HALO
200 I V4
I
d
1 5.
6 o”!
100
.E
0 -m
19 HALO
1 1 V4
molil Cd*+
f-+
molfiCd*+
#--I
i
50
5”
t
0-I 210 -!z 9, a
0 50
270
330
390
450
510
210
390
450
510
210
270
330
390
450
510
200 IV5
150I
mol/l Cd*+ f-I
7
210
270
330
I
*
I
I
I
270
330
390
450
510
min
Fig. 2. Effect of cadmium on melatonin production in perifused rat pineal glands removed from rats killed 7 and 19 HALO. Each point is the mean i SEM of data obtained in 4 to 5 chambers of perifusion (2 pineal glands per chamber).
Z.-Y Zhao, Y. Touitou /Molecular
192
and Cellular Endocrinology 101 (1994) 189-196
7 HALO
200
19 HALO
7
3
01
210
200
270
330
390
450
510
270
330
390
450
510
1
01 210
210
270
330
390
450
510
min
Fig. 3. Effect of manganese on melatonin production in perifused rat pineal glands removed from rats killed 7 and 19 HALO. Each point is the mean 5 SEM of data obtained in 4 chambers of perifusion (2 pineal glands per chamber).
pinephrine was introduced in the perifusion chambers for 30 min. One hour later after the melatonin production was stimulated, lop4 mol/l of cobalt or nifedipine
was perifused for 30 min which allowed the examination of the effects of both calcium channel blockers on adrenergically stimulated melatonin release.
7 HALO
19 HALO
1 Ow5mol/l Nifedipine r--+
0-i 210
,
,
270
330
. 390 min
.
,
450
510
J 210
I
270
I
I
330
390
.
450
‘
510
min
Fig. 4. Effect of nifedipine on melatonin production in perifused rat pineal gtands removed from rats killed 7 and 19 HALO. Each point is the mean i SEM of data obtained in 4 to 5 chambers of perifusion (2 pineal glands per chamber).
Z.-Y. Zhao, Y Touitou /Molecular 2001
193
and Cellular Endocrinology 101 (1994) 189-196 200
7 HALO
19 HALO 1
150-
1 V4 moi/l Diltiiem
50-
of
01
‘
210
270
330
390
450
510
200-I
B 1
P < 0.01
210 200
I
270
330
390
450
510
330
390
450
510
1
10m5 moVl Diltiarem
,5
t8 P 3 ae .!$
10
3 m
5
210
270
330
390
450
510
0-l
.
210
270
,
min
min
Fig. 5. Effect of diltiazem on melatonin production in perifused rat pineal glands removed from rats killed 7 and 19 HALO. Each point is the mean it SEM of data obtained in 4 to 6 chambers of perifusion (2 pineal glands per chamber).
3.2. Effects of inorganic calcium channel blockers on basal melaton~n ~ro~uct~n
(P < 0.01) more effective in pineal glands removed 7 HALO (80% inhibition) than in glands removed 19
A concentration of 10e4 mol/l of cobalt resulted in a significant (P < O.OOl>inhibitory effect on melatonin production (Fig. 1). This inhibition was significantly
HALO (57% inhibition). With a concentration of 10e5 mol/l cobalt had no significant effect on melatonin production in glands removed both 7 and 19 JX$LO. The effect of cadmium on melatonin production is
7 HALO
19 HALO
‘iii 5 I I .o Td 8 .z 9 -m $
210
5 5 7 I 0 ‘ii z .z s m 1 z
270
330
390
450
510
210
200
200
150
150
100
100
50
50
01 210
270
330
390 min
450
510
or 210
270
330
390
450
510
270
330
390
450
510
min Fig. 6. Effect of verapamil on melatonin production in perifused rat pineal glands removed from rats killed 7 and 19 HALO. Each point is the mean rfr:SEM of data obtained in 3 to 4 chambers of perifusion (2 pineal glands per chamber).
194
Z.-Y Zhao,
I’. Touitou /Molecular
and Cellular Endocrinology
shown in Fig. 2. Cadmium (lo-” mol/l) had a significant inhibitory effect (P < 0.001) on melatonin production in glands removed both 7 HALO (50% inhibition) and 19 HALO (40% inhibition). With a concentration of lo-” mol/l, cadmium showed a significant inhibitory effect on melatonin production only in glands removed 7 HALO (33% inhibition; P < 0.001). This inhibitory effect of cadmium persisted until the end of perifusion. Manganese showed no significant effect on melatonin production with a concentration of lo-’ mol/l (Fig. 3). In contrast to the other blockers, 10e4 mol/l of manganese resulted in a stimulatory effect on melatonin production both in pineal glands removed 7 and
101 (1994) 189-196
19 HALO, but this stimulation was significant only in glands removed 19 HALO (40% increase, P < 0.01). 3.2. Effects of organic calcium channel blockers on basal melatonin production Fig. 4 shows the effect of nifedipine on melatonin production in perifused rat pineal glands. low4 mol/l of nifedipine resulted in a significant inhibitory effect (P < 0.001) on melatonin production in glands removed both 7 and 19 HALO (60% and 55% inhibition, respectively). With a concentration of 10P5 mol/l, nifedipine failed to show any significant effect upon melatonin production.
6001
300 1
6001
210
270
330
390
450
510
270
330
390
450
510
04
210
I
210
270
330
390
I
1
I
270
330
390
450
510
0-l 210
450
510
min min Fig. 7. Effect of cobalt and nifedipine on adrenergically stimulated (isoproterenol or norepinephrine 10m4 mol/l) melatonin production perifused rat pineal glands removed from rats killed 7 HALO. Each point is the mean f SEM of data obtained in 5 chambers of perifusion pineal glands per chamber).
in (2
Z.-Y. Zhao, I’. Touitou /Molecular and Cellular Endocrinology 101 (1994) 189-196
Fig. 5 shows the effect of diltiazem. With a concentration of 10m4 mol/l, diltiazem induced a significant inhibition on melatonin production in glands removed both 7 and 19 HALO (26% and 23% inhibition, respectively). As with cadmium, low5 mol/l of diltiazem induced a significant inhibitory effect on melatonin production only in pineal glands removed 7 HALO (25% inhibition; P < 0.001). In contrast to other blockers, verapamil failed to show any significant effect on melatonin production with the two concentrations used (lo-’ and 10e4 mol/l) whatever the circadian stage of the removal of the pineal glands (Fig. 6). 3.3. Effects of cobalt and nifedipine on stimulated melatonin production Fig. 7 shows the effects of cobalt and nifedipine (10m4 mol/l) on adrenergically stimulated pineal melatonin production. Cobalt had a significant (P < 0.001) inhibitory effect on both isoproterenol and norepinephrine-stimulated melatonin production. The inhibitory effect was short: it occurred only at the time of drug perifusion and afterwards melatonin production returned immediately to its high stimulated level. Nifedipine showed a significant inhibitory (P < 0.001) effect on isoproterenol-stimulated pineal melatonin production but not on norepinephrine-stimulated melatonin release.
4. Discussion In a previous study we have shown that physiologic concentrations of calcium (5.2 mmol/l) and magnesium (1.34 mmol/l) for the rat had a stimulatory effect on melatonin production only in pineal glands removed in the middle of the dark span (19 HALO), but not in those removed in the middle of the light span (7 HALO) (Zhao and Touitou, 1993b). Our results strongly suggested that pineal glands respond differently to calcium and magnesium according to the circadian stage. This difference may be explained by circadian changes in calcium channel function. Therefore, in the present study, we have studied the effects of various calcium channel blockers on basal melatonin production by perifused rat pineal glands removed in the middle of the light and dark periods (7 and 19 HALO) or on adrenergically stimulated melatonin production by perifused rat pineal glands removed 7 HALO. With a concentration of 10e4 mol/l, most of the blockers studied (Co’+, Cd*+, nifedipine and diltiazem> showed an inhibitory effect on basal melatonin production and this inhibition was more effective in pineal glands removed in the middle of the light span
195
(7 HALO) than in those removed in the middle of the dark span (19 HALO), e.g., the inhibition of Co*’ reached 80% in pineal glands removed 7 HALO, and 57% in pineal glands removed 19 HALO (P < 0.01). With a concentration of lop5 mol/l, only Cd and diltiazem showed an inhibitory effect on glands removed during the light span (7 HALO). This effect was not significant in glands removed 19 HALO. From these results, it is evident that the rat pineal gland responds differently to calcium (Zhao and Touitou, 1993) and calcium channel blockers according to the circadian stage. It has been established that through voltage-dependent calcium channels (VDCCs), calcium regulates in vitro melatonin production in cultured chick pineal cells and that in vivo administration of calcium channel blockers inhibit SNAT activity in hen and rat pineal glands (Harrison and Zatz, 1989; Zawilska and Nowak, 1990, 1991). The “L-type” calcium channels appear to be involved in regulating melatonin synthesis in chick pineal cells (Harrison and Zatz, 1989). They were found to have the property of plasticity, i.e., their number in the brain has been shown to change during aging and in a number of pathological conditions (e.g., cardiomyopathy, and hypertension) and after chronic administration of drugs, such as opiates, phencyclidine, neuroleptics, and alcohol (Miller, 1987). Our results, showing a greater effect of calcium stimulation and a weaker effect of calcium channel blockers inhibition on melatonin production in pineal glands removed at night, might be explained by a circadian variation of calcium channels number, i.e., at night there would be more functioning channels than ‘during the day. Therefore the pineal gland would be more sensitive to calcium during dark than during the light and the calcium channel blockers would be less efficient during the dark, and vice versa. Conversely, there would be an opposite explanation, i.e., after the night peak of melatonin production, the calcium channels would be in a less sensitive situation, hence the calcium channel blockers would be less effective, and vice versa. Since calcium and magnesium have a direct stimulatory effect on SNAT by binding to the enzyme and enhancing its catalytic activity (Morton, 19891, another hypothesis would be that the binding affinity of the SNAT to ions might vary according to the circadian stage (i.e., weaker during the day) thus resulting in a reduced stimulating effect of calcium on melatonin production by pineal glands removed during the day. Further investigations are needed to validate these hypotheses. The results obtained with cadmium are different from those of other calcium channel blockers. Cd*+ is considered as an industrial toxic metallic ion, its toxicity could be blocked by other calcium channel blockers (Hinkle et al., 1987). The inhibitory effect of cadmium persisted till the end of perifusion showing its toxic
196
Z.-Y. Zhao, Y. Touitou /Molecular
and Cellular Endocrinology 101 (1994) 189-196
effect on calcium channels. Our results suggest that the toxicity of Cd2+ is both dose and circadian stage dependent. In contrast to other blockers, .Mn2+ had a stimulatory effect on melatonin production with a concentration of 10m4 mol/l and had no significant effect with a concentration of lop5 mol/l. Mn2+ was shown to have a significant inhibitory effect on melatonin production in cultured chick pineal cells (Zatz and Mullen, 1988; Harrison and Zatz, 19891, while in rat cortical membranes, Mn2+ provided a stimulation of [ 3H]nitrendipine binding, suggesting that Mn2+ presents calcium agonist as well as antagonist properties on calcium channels (Gould et al., 1982). In our experiment, Mn2+ in a concentration of 1O-4 mol/l showed a calcium agonist property on melatonin production in perifused rat pineal glands. Verapamil at concentrations of lop5 and lop4 mol/l showed no significant effect on melatonin production in glands removed both 7 and 19 HALO. Zawilska and Nowak (1991) reported that in vivo administration of verapamil (20 mg/kg, i.p.1 resulted in a marked suppression of the nighttime pineal SNAT activity in rat. In contrast to this effect in rat in vivo, verapamil failed to show any significant effect in perifused rat pineal glands whatever the circadian stage, no explanation can be given at the moment. The possibility exists that verapamil is converted into a more active metabolite in vivo. In our perifusion system, both pre- and postsynaptic action exist since the nerve terminals innervating the pineal gland are known to need more than 24 h to degenerate completely (Klein et al., 1983). The basal melatonin production level, however, was not the result of both pre- or’lpostsynaptic innervation, since the P-adrenergic antagonist, propranolol showed no effect on this basal melatonin production level (data will be published elsewhere). The adrenergic innervation of pineal gland plays an important role in the circadian rhythm of melatonin. We have therefore determined the effects of calcium channel blockers on adrenergitally stimulated melatonin release. The suppression of calcium influx effects by cobalt resulted in an immediate inhibition of melatonin release. It is very likely that melatonin synthesis was not impaired since its concentration in the perifusate recovered the increased release level due to isoproterenol stimulation as soon as cobalt was withdrawn. Nifedipine had a short significant inhibitory effect only on isoproterenol-stimulated melatonin release.
In conclusion, most of the calcium channel blockers studied had an inhibitory effect on melatonin production in perifused rat pineal glands, and this inhibition was more effective in pineal glands removed during the light phase. Cobalt had an immediate inhibitory effect on both isoproterenol and norepinephrine-stimulated pineal melatonin release limited to its time of presence in the perifusion system and nifedipine showed an inhibition only on isoproterenol-stimulated melatonin release. Our data suggest that calcium may be involved in the release of melatonin from the pinealocytes. This study is, to our knowledge, the first to show circadian stage differences in the effect of calcium channel blockers on melatonin production by the pineal gland.
5. Acknowledgements The technical assistance of Mrs Karine Lambert was greatly appreciated.We are indebted to Prof. E. Haus (Ramsay Medical Center, St. Paul, Minn. USA) for editorial assistance and critical review of the manuscript.
6. References Fraser, S., Cowen, P., Franklin, M., Franey, C. and Arendt, J. (1983) Clin. Chem. 29, 396-397. Gould, R.J., Murphy, K.M.M. and Synder, S.H. (1982) Proc. Natl. Acad. Sci. USA 79, 3656-3660. Harrison, N.L. and Zatz, M. (1989) J. Neurosci. 9, 2462-2467. Hinkle, P.M., Kinsella, P.A. and Osterhoudt, K.C. (1987) J. Biol. Chem. 262, 16333-16337. Klein, D.C., Sugden, D. and Weller, J.L. (1983) Proc. Nat]. Acad. Sci. USA 80, 599-603 Miller, R.J. (1987) Science 235, 46-52. Morton, D.J. (1989) J. Neural Transm. 75, 51-64. Ravault, J.P., Arendt, J., Tobler, I., Chesneau, D. and Maulin, 0. (1989) Chronobiol. Int. 6, 329-39. Sugden, A.L., Sugden, D. and Klein, DC. (1986) J. Biol. Chem. 261, 11608-11612. Sugden. A.L., Sugden, D. and Klein, D.C. (1987) J. Biol. Chem. 262, 741-745. Zatz, M. and Mullen, D.A. (1988) Brain Res. 463, 305-316. Zatz, M. and Romero, J.A. (1978) Biochem. Pharmacol. 27, 25492533. Zawilska, J.B. and Nowak, J.Z. (1990) Neurosci. Lett. 118, 17-20. Zawilska, J.B. and Nowak, J.Z. (1991) J. Neural Transm. 84, 171-182. Zhao, Z.-Y. and Touitou, Y. (1993a) J. Pineal Res. 14, 73-77. Zhao, Z.-Y. and Touitou, Y. (1993b) Acta Endocrinol. 129, 81-88.
ELSEVIER
Molecular and Cellular Endocrinology lOl(1994) 189-196
Pineal perifusion with calcium channel blockers inhibits differently daytime and nighttime melatonin production in rat Zi-Yan Zhao, Yvan Touitou * Department of Biochemistry, Faculte’ de Mkdecine Pit&Salphri&e,
Boulevard de I’HGpital91, 75013 Paris, France
(Received 1 November 1993; accepted 20 December 1993)
Abstract In a previous study we have shown that the response of perifused pineal glands to calcium was different according to the circadian stage at which the glands were removed. This difference may be explained by circadian changes in calcium channel function. Therefore in the present study we documented the effects of calcium channel blockers in perifused rat pineal glands removed in the middle of the light and dark spans (7 and 19 HALO (hours after light onset), in a L/D 12: 12 regimen). Moreover, we have studied the effect of calcium channel blockers on adrenergically stimulated pineal glands removed 7 HALO. Inorganic (Co*+ and Cd*+) and organic (nifedipine and diltiazeml calcium channel blockers at 1O-4 mol/l all significantly reduced melatonin production and this inhibition was more effective with the glands removed 7 HALO. In a concentration of 10m5 mol/l, only Cd*+ and diltiazem reduced melatonin production significantly in pineal glands removed 7 HALO. Verapamil at 10m4 and 10e5 mol/l showed no significant effect on melatonin production in glands removed both during the light and dark spans. Mn*+ at 10e4 mol/l (but not at lop5 mol/l) appeared to stimulate melatonin production in glands removed both during the light and the dark (significant increase only with glands removed during the dark). Cobalt showed an immediate short inhibitory effect on both isoproterenol and norepinephrine-stimulated melatonin release, whereas nifedipine showed a significant inhibition only on isoproterenol-stimulated melatonin release. These results strongly suggest a circadian stage dependence of the pineal gland response to some calcium channel blockers and the involvement of calcium in the release of melatonin from pinealocytes. Key words:
Melatonin;
(Rat pineal gland); Calcium channel blocker; Circadian rhythm; Adrenergic
1. Introduction
Calcium ions play important roles in regulating a wide variety of cellular functions, among which the pineal gland function. In the pineal gland the sensitivity of P-receptors is dependent upon calcium ions (Wilkinson, 1978) and calcium influx is essential for norepinephrine to increase CAMP and cGMP (Sugden et al., 1986); besides, the activation of ar-adrenoceptors increases cytosolic calcium concentration (Sugden et al., 1987). In addition, calcium was shown to be necessary for the induction of serotonin N-acetyltransferase @NAT) activity (Zatz and Romero, 1978).
* Corresponding author. Tel.: (33 1) 40 77 96 63; Fax: (33 1) 40 77 96 65. Elsevier Science Ireland Ltd. SSDI 0303-7207(94)00013-Y
stimulation
In a previous study we have found that the response of perifused pineal glands to calcium was different according to the circadian stage at which the glands were removed, i.e., increased calcium concentration (within the physiological range) in the perifusion system stimulated melatonin production by pineal glands removed from rats killed in the middle of the dark period (19 h after light onset, HALO; L/D = 12: 12) but not by those removed in the middle of the light period (7 HALO) (Zhao and Touitou, 1993a). A possible explanation of this difference may be the circadian changes in calcium channel function. Therefore, we studied the effects of calcium channel blockers on melatonin production by perifused rat pineal glands removed in the middle of the light and dark spans (7 and 19 HALO) in order to document a possible circadian stage dependence of their effects which could support possible
190
Z.-Y. Zhao. Y. Touitou /Moleculur
changes in calcium channel function. These were examined on basal and on adrenergically lated pineal melatonin production.
and Cellular Endocrinology
effects stimu-
2. Materials and methods 2.1. Housirzg and ~~c~ro~~zation of rats Male albino Wistar rats (IFFA-CREDO, L’Arbresle, France), 4-5 weeks old (loo-120 g) on arrival at the laboratory, were housed in a chronobiologic animal facility (Enceinte Autonome d’Animalerie, Ref. E 110SP-6, ESI Flufrance, Arcueil, France) with food and water ad libitum. The rats were synchronised with a lighting regimen (L/D = 12: 12; the light intensity in the cages was 3.1-3.8 pW/cm’) for three weeks before each experiment. The chronobiologic facility was equipped with equispaced, sound-proof, temperaturecontrolled (21 f l.O’C) compartments provided with independent light-dark cycles. The 12 h of light span started in different compartments at 03 and 15 h (clock time), therefore the removal of pineal glands at 10 h resulted in glands obtained 7 and 19 HALO, i.e., in the middle of the light and dark spans, respectively. The rats were killed by decapitation at the age of 7-8 weeks (about 250 g) after 3 weeks’ synchronization in the chronobiologic facility. Decapitation of rats during the dark period (19 HALO) was carried out under a red light (0.4 pW/cm’ at the level of worktable) and each rat was exposed to the dim red light for less than 20 s to avoid alteration of melatonin production. The pineal glands were quickly removed and kept in an oxygenated Krebs-Ringer solution in an iced water bath until transferred into the perifusion chambers. 2.2. ~~emicats
IO1 (1994) 189-196
the temperature at 37 f 0.5”C, and a peristaltic pump (Ismatec IPN-12). The column and pistons delimited a 0.4 cm high space (perifusion chamber with a volume of 0.2 ml) which was filled with Krebs-Ringer solution (pH 7.4, gassed continuously with 95% O,-5% CO,). Pineal glands (two per chamber) were perifused with Krebs-Ringer solution at a constant flow rate of 0.2 ml/min. We had previously established (Zhao and Touitou, 1993b) that rat pineal glands in perifusion stabilize their melatonin production 3-4 h after the beginning of the perifusion, then melatonin production remains stable up to 8 h. Therefore in this study the effluent perifusate fractions were collected in glass tubes every 10 min from 3 h 30 min (210 min) to 8 h 30 min (5 IO min) of perifusion and stored at - 20°C until assayed for melatonin. After the first 4 fractions were collected (to calculate the basal level), the pineals were perifused in two sets of experiments: in the first set the calcium channel blockers were perifused for 30 min (250-280 min of the perifusion); in the second set the calcium channel blocker perifusion was preceded by an adrenergic stimulation by either isoproterenol or norepinephrine (adrenergic stimulation: 250-280 min; calcium channel blocker: 340-370 min). 2.4. Meltstonin radioimmunoassay Melatonin concentration in the collected fractions was measured directly by a modification of the RIA method of Fraser et al. (19831, using a 1251-melatonin tracer. The assay sensitivity was 5-10 pg/ml, the only significant (0.1%) cross-reactant was Ghydroxymelatonin (Ravault et al., 1989). The intra-assay and inter-assay coefficients of variation were 9% (N = 20). 2.5. Statistical analysis
Melatonin was purchased from Fluka (Buchs, 1251-melatonin from Dositek (Orsay, Switzerland), France), rabbit anti-melatonin serum and anti-immunoglobulin serum from INRA (Tours, France). The inorganic calcium channel blockers (cobalt chloride, cadmium chloride and manganese sulphate) were the generous gifts of the Pharmacy of PitiC-Salpetriere Hospital, Paris. Nifedipine was obtained from Bayer Pharma laboratory (Paris), diltiazem from SynthClabo (Bagneux, France) verapamil from Biosedra (Malakoff, France), (-Ksoproterenol from Sigma and (It)norepinephrine from Aldrich (Saint Quentin, France).
All the results obtained were expressed as means it S.E.M. To evaluate the effects of the calcium channel blockers, the melatonin production in any chamber was expressed as a percentage of the mean concentration (considered as a “basal level”) in the first 4 fractions (220, 230, 240 and 250 min) collected just before the introduction of drugs in the perifusion system. Student’s unpaired t-test was used to analyse the experimental data.
2.3. Experimental procedure
The melatonin production in perifused rat pineal glands reached a stable level after 3-4 h perifusion, and then remained fairly constant and roughly similar up to 8 h perifusion whatever the circadian stage (Zhao and Touitou, 1993). We therefore considered the mean
The perifusion system consisted of a plastic column (diameter 0.8 x 3.0 cm) closed with two pistons, a thermostatic bath (Techne TE-8J) which maintained
3. Resutts
Z.-Y. Zhao, Y. Touitou /Molecular
7 HALO
200
z a
191
and Cellular Endocrinology 101 (1994) 189-196
19 HALO
2OOf
1 ‘50-
IO-4moin
co*+
f-I
j B s _g 9 m B I
IOO-
&-m-
50-
1
04 210
z 270
o+ 210
270
.
’ 330
’ 390
’
*
330
390
450
5
450
P < 0.001
1 510
04 210
270
330
’
210
270
330
510
=
’
’
390
450
510
390
450
510
min
Fig. 1. Effect of cobalt on melatonin production in perifused rat pineal glands removed from rats killed 7 and 19 HALO (hours after light onset). Each point is the mean f SEM of data obtained in 3 chambers of perifusion (2 pineal glands per chamber).
The basal level of melatonin production in this study was 20.8 it 5.7 and 19.0 li: 5.9 pg melatonin/minjgland in perifused pineal glands removed 7 and 19 HALO, respectively (mean f SD of 56 and 54 chambers of perifusion, respectively). In another experiment, the perifused pineal glands were removed 7 HALO and isoproterenol or nore-
concentration of melatonin obtained between 210 and 250 min perifusion as a baseline. Two concentrations (10m5 and 10F4 mol/l) of calcium channel blockers (cobalt, cadmium, manganese, nifedipine, diltiazem, and verapamil) were introduced respectively in the perifusion chambers for 30 min perifusion.
5 5
7 HALO
200 I V4
I
d
1 5.
6 o”!
100
.E
0 -m
19 HALO
1 1 V4
molil Cd*+
f-+
molfiCd*+
#--I
i
50
5”
t
0-I 210 -!z 9, a
0 50
270
330
390
450
510
210
390
450
510
210
270
330
390
450
510
200 IV5
150I
mol/l Cd*+ f-I
7
210
270
330
I
*
I
I
I
270
330
390
450
510
min
Fig. 2. Effect of cadmium on melatonin production in perifused rat pineal glands removed from rats killed 7 and 19 HALO. Each point is the mean i SEM of data obtained in 4 to 5 chambers of perifusion (2 pineal glands per chamber).
Z.-Y Zhao, Y. Touitou /Molecular
192
and Cellular Endocrinology 101 (1994) 189-196
7 HALO
200
19 HALO
7
3
01
210
200
270
330
390
450
510
270
330
390
450
510
1
01 210
210
270
330
390
450
510
min
Fig. 3. Effect of manganese on melatonin production in perifused rat pineal glands removed from rats killed 7 and 19 HALO. Each point is the mean 5 SEM of data obtained in 4 chambers of perifusion (2 pineal glands per chamber).
pinephrine was introduced in the perifusion chambers for 30 min. One hour later after the melatonin production was stimulated, lop4 mol/l of cobalt or nifedipine
was perifused for 30 min which allowed the examination of the effects of both calcium channel blockers on adrenergically stimulated melatonin release.
7 HALO
19 HALO
1 Ow5mol/l Nifedipine r--+
0-i 210
,
,
270
330
. 390 min
.
,
450
510
J 210
I
270
I
I
330
390
.
450
‘
510
min
Fig. 4. Effect of nifedipine on melatonin production in perifused rat pineal gtands removed from rats killed 7 and 19 HALO. Each point is the mean i SEM of data obtained in 4 to 5 chambers of perifusion (2 pineal glands per chamber).
Z.-Y. Zhao, Y Touitou /Molecular 2001
193
and Cellular Endocrinology 101 (1994) 189-196 200
7 HALO
19 HALO 1
150-
1 V4 moi/l Diltiiem
50-
of
01
‘
210
270
330
390
450
510
200-I
B 1
P < 0.01
210 200
I
270
330
390
450
510
330
390
450
510
1
10m5 moVl Diltiarem
,5
t8 P 3 ae .!$
10
3 m
5
210
270
330
390
450
510
0-l
.
210
270
,
min
min
Fig. 5. Effect of diltiazem on melatonin production in perifused rat pineal glands removed from rats killed 7 and 19 HALO. Each point is the mean it SEM of data obtained in 4 to 6 chambers of perifusion (2 pineal glands per chamber).
3.2. Effects of inorganic calcium channel blockers on basal melaton~n ~ro~uct~n
(P < 0.01) more effective in pineal glands removed 7 HALO (80% inhibition) than in glands removed 19
A concentration of 10e4 mol/l of cobalt resulted in a significant (P < O.OOl>inhibitory effect on melatonin production (Fig. 1). This inhibition was significantly
HALO (57% inhibition). With a concentration of 10e5 mol/l cobalt had no significant effect on melatonin production in glands removed both 7 and 19 JX$LO. The effect of cadmium on melatonin production is
7 HALO
19 HALO
‘iii 5 I I .o Td 8 .z 9 -m $
210
5 5 7 I 0 ‘ii z .z s m 1 z
270
330
390
450
510
210
200
200
150
150
100
100
50
50
01 210
270
330
390 min
450
510
or 210
270
330
390
450
510
270
330
390
450
510
min Fig. 6. Effect of verapamil on melatonin production in perifused rat pineal glands removed from rats killed 7 and 19 HALO. Each point is the mean rfr:SEM of data obtained in 3 to 4 chambers of perifusion (2 pineal glands per chamber).
194
Z.-Y Zhao,
I’. Touitou /Molecular
and Cellular Endocrinology
shown in Fig. 2. Cadmium (lo-” mol/l) had a significant inhibitory effect (P < 0.001) on melatonin production in glands removed both 7 HALO (50% inhibition) and 19 HALO (40% inhibition). With a concentration of lo-” mol/l, cadmium showed a significant inhibitory effect on melatonin production only in glands removed 7 HALO (33% inhibition; P < 0.001). This inhibitory effect of cadmium persisted until the end of perifusion. Manganese showed no significant effect on melatonin production with a concentration of lo-’ mol/l (Fig. 3). In contrast to the other blockers, 10e4 mol/l of manganese resulted in a stimulatory effect on melatonin production both in pineal glands removed 7 and
101 (1994) 189-196
19 HALO, but this stimulation was significant only in glands removed 19 HALO (40% increase, P < 0.01). 3.2. Effects of organic calcium channel blockers on basal melatonin production Fig. 4 shows the effect of nifedipine on melatonin production in perifused rat pineal glands. low4 mol/l of nifedipine resulted in a significant inhibitory effect (P < 0.001) on melatonin production in glands removed both 7 and 19 HALO (60% and 55% inhibition, respectively). With a concentration of 10P5 mol/l, nifedipine failed to show any significant effect upon melatonin production.
6001
300 1
6001
210
270
330
390
450
510
270
330
390
450
510
04
210
I
210
270
330
390
I
1
I
270
330
390
450
510
0-l 210
450
510
min min Fig. 7. Effect of cobalt and nifedipine on adrenergically stimulated (isoproterenol or norepinephrine 10m4 mol/l) melatonin production perifused rat pineal glands removed from rats killed 7 HALO. Each point is the mean f SEM of data obtained in 5 chambers of perifusion pineal glands per chamber).
in (2
Z.-Y. Zhao, I’. Touitou /Molecular and Cellular Endocrinology 101 (1994) 189-196
Fig. 5 shows the effect of diltiazem. With a concentration of 10m4 mol/l, diltiazem induced a significant inhibition on melatonin production in glands removed both 7 and 19 HALO (26% and 23% inhibition, respectively). As with cadmium, low5 mol/l of diltiazem induced a significant inhibitory effect on melatonin production only in pineal glands removed 7 HALO (25% inhibition; P < 0.001). In contrast to other blockers, verapamil failed to show any significant effect on melatonin production with the two concentrations used (lo-’ and 10e4 mol/l) whatever the circadian stage of the removal of the pineal glands (Fig. 6). 3.3. Effects of cobalt and nifedipine on stimulated melatonin production Fig. 7 shows the effects of cobalt and nifedipine (10m4 mol/l) on adrenergically stimulated pineal melatonin production. Cobalt had a significant (P < 0.001) inhibitory effect on both isoproterenol and norepinephrine-stimulated melatonin production. The inhibitory effect was short: it occurred only at the time of drug perifusion and afterwards melatonin production returned immediately to its high stimulated level. Nifedipine showed a significant inhibitory (P < 0.001) effect on isoproterenol-stimulated pineal melatonin production but not on norepinephrine-stimulated melatonin release.
4. Discussion In a previous study we have shown that physiologic concentrations of calcium (5.2 mmol/l) and magnesium (1.34 mmol/l) for the rat had a stimulatory effect on melatonin production only in pineal glands removed in the middle of the dark span (19 HALO), but not in those removed in the middle of the light span (7 HALO) (Zhao and Touitou, 1993b). Our results strongly suggested that pineal glands respond differently to calcium and magnesium according to the circadian stage. This difference may be explained by circadian changes in calcium channel function. Therefore, in the present study, we have studied the effects of various calcium channel blockers on basal melatonin production by perifused rat pineal glands removed in the middle of the light and dark periods (7 and 19 HALO) or on adrenergically stimulated melatonin production by perifused rat pineal glands removed 7 HALO. With a concentration of 10e4 mol/l, most of the blockers studied (Co’+, Cd*+, nifedipine and diltiazem> showed an inhibitory effect on basal melatonin production and this inhibition was more effective in pineal glands removed in the middle of the light span
195
(7 HALO) than in those removed in the middle of the dark span (19 HALO), e.g., the inhibition of Co*’ reached 80% in pineal glands removed 7 HALO, and 57% in pineal glands removed 19 HALO (P < 0.01). With a concentration of lop5 mol/l, only Cd and diltiazem showed an inhibitory effect on glands removed during the light span (7 HALO). This effect was not significant in glands removed 19 HALO. From these results, it is evident that the rat pineal gland responds differently to calcium (Zhao and Touitou, 1993) and calcium channel blockers according to the circadian stage. It has been established that through voltage-dependent calcium channels (VDCCs), calcium regulates in vitro melatonin production in cultured chick pineal cells and that in vivo administration of calcium channel blockers inhibit SNAT activity in hen and rat pineal glands (Harrison and Zatz, 1989; Zawilska and Nowak, 1990, 1991). The “L-type” calcium channels appear to be involved in regulating melatonin synthesis in chick pineal cells (Harrison and Zatz, 1989). They were found to have the property of plasticity, i.e., their number in the brain has been shown to change during aging and in a number of pathological conditions (e.g., cardiomyopathy, and hypertension) and after chronic administration of drugs, such as opiates, phencyclidine, neuroleptics, and alcohol (Miller, 1987). Our results, showing a greater effect of calcium stimulation and a weaker effect of calcium channel blockers inhibition on melatonin production in pineal glands removed at night, might be explained by a circadian variation of calcium channels number, i.e., at night there would be more functioning channels than ‘during the day. Therefore the pineal gland would be more sensitive to calcium during dark than during the light and the calcium channel blockers would be less efficient during the dark, and vice versa. Conversely, there would be an opposite explanation, i.e., after the night peak of melatonin production, the calcium channels would be in a less sensitive situation, hence the calcium channel blockers would be less effective, and vice versa. Since calcium and magnesium have a direct stimulatory effect on SNAT by binding to the enzyme and enhancing its catalytic activity (Morton, 19891, another hypothesis would be that the binding affinity of the SNAT to ions might vary according to the circadian stage (i.e., weaker during the day) thus resulting in a reduced stimulating effect of calcium on melatonin production by pineal glands removed during the day. Further investigations are needed to validate these hypotheses. The results obtained with cadmium are different from those of other calcium channel blockers. Cd*+ is considered as an industrial toxic metallic ion, its toxicity could be blocked by other calcium channel blockers (Hinkle et al., 1987). The inhibitory effect of cadmium persisted till the end of perifusion showing its toxic
196
Z.-Y. Zhao, Y. Touitou /Molecular
and Cellular Endocrinology 101 (1994) 189-196
effect on calcium channels. Our results suggest that the toxicity of Cd2+ is both dose and circadian stage dependent. In contrast to other blockers, .Mn2+ had a stimulatory effect on melatonin production with a concentration of 10m4 mol/l and had no significant effect with a concentration of lop5 mol/l. Mn2+ was shown to have a significant inhibitory effect on melatonin production in cultured chick pineal cells (Zatz and Mullen, 1988; Harrison and Zatz, 19891, while in rat cortical membranes, Mn2+ provided a stimulation of [ 3H]nitrendipine binding, suggesting that Mn2+ presents calcium agonist as well as antagonist properties on calcium channels (Gould et al., 1982). In our experiment, Mn2+ in a concentration of 1O-4 mol/l showed a calcium agonist property on melatonin production in perifused rat pineal glands. Verapamil at concentrations of lop5 and lop4 mol/l showed no significant effect on melatonin production in glands removed both 7 and 19 HALO. Zawilska and Nowak (1991) reported that in vivo administration of verapamil (20 mg/kg, i.p.1 resulted in a marked suppression of the nighttime pineal SNAT activity in rat. In contrast to this effect in rat in vivo, verapamil failed to show any significant effect in perifused rat pineal glands whatever the circadian stage, no explanation can be given at the moment. The possibility exists that verapamil is converted into a more active metabolite in vivo. In our perifusion system, both pre- and postsynaptic action exist since the nerve terminals innervating the pineal gland are known to need more than 24 h to degenerate completely (Klein et al., 1983). The basal melatonin production level, however, was not the result of both pre- or’lpostsynaptic innervation, since the P-adrenergic antagonist, propranolol showed no effect on this basal melatonin production level (data will be published elsewhere). The adrenergic innervation of pineal gland plays an important role in the circadian rhythm of melatonin. We have therefore determined the effects of calcium channel blockers on adrenergitally stimulated melatonin release. The suppression of calcium influx effects by cobalt resulted in an immediate inhibition of melatonin release. It is very likely that melatonin synthesis was not impaired since its concentration in the perifusate recovered the increased release level due to isoproterenol stimulation as soon as cobalt was withdrawn. Nifedipine had a short significant inhibitory effect only on isoproterenol-stimulated melatonin release.
In conclusion, most of the calcium channel blockers studied had an inhibitory effect on melatonin production in perifused rat pineal glands, and this inhibition was more effective in pineal glands removed during the light phase. Cobalt had an immediate inhibitory effect on both isoproterenol and norepinephrine-stimulated pineal melatonin release limited to its time of presence in the perifusion system and nifedipine showed an inhibition only on isoproterenol-stimulated melatonin release. Our data suggest that calcium may be involved in the release of melatonin from the pinealocytes. This study is, to our knowledge, the first to show circadian stage differences in the effect of calcium channel blockers on melatonin production by the pineal gland.
5. Acknowledgements The technical assistance of Mrs Karine Lambert was greatly appreciated.We are indebted to Prof. E. Haus (Ramsay Medical Center, St. Paul, Minn. USA) for editorial assistance and critical review of the manuscript.
6. References Fraser, S., Cowen, P., Franklin, M., Franey, C. and Arendt, J. (1983) Clin. Chem. 29, 396-397. Gould, R.J., Murphy, K.M.M. and Synder, S.H. (1982) Proc. Natl. Acad. Sci. USA 79, 3656-3660. Harrison, N.L. and Zatz, M. (1989) J. Neurosci. 9, 2462-2467. Hinkle, P.M., Kinsella, P.A. and Osterhoudt, K.C. (1987) J. Biol. Chem. 262, 16333-16337. Klein, D.C., Sugden, D. and Weller, J.L. (1983) Proc. Nat]. Acad. Sci. USA 80, 599-603 Miller, R.J. (1987) Science 235, 46-52. Morton, D.J. (1989) J. Neural Transm. 75, 51-64. Ravault, J.P., Arendt, J., Tobler, I., Chesneau, D. and Maulin, 0. (1989) Chronobiol. Int. 6, 329-39. Sugden, A.L., Sugden, D. and Klein, DC. (1986) J. Biol. Chem. 261, 11608-11612. Sugden. A.L., Sugden, D. and Klein, D.C. (1987) J. Biol. Chem. 262, 741-745. Zatz, M. and Mullen, D.A. (1988) Brain Res. 463, 305-316. Zatz, M. and Romero, J.A. (1978) Biochem. Pharmacol. 27, 25492533. Zawilska, J.B. and Nowak, J.Z. (1990) Neurosci. Lett. 118, 17-20. Zawilska, J.B. and Nowak, J.Z. (1991) J. Neural Transm. 84, 171-182. Zhao, Z.-Y. and Touitou, Y. (1993a) J. Pineal Res. 14, 73-77. Zhao, Z.-Y. and Touitou, Y. (1993b) Acta Endocrinol. 129, 81-88.