C h a p t e r

Characterization and pharmacological applications of a new technique

This chapter describes the characterization of a new technique to measure melatonin contents in the pineal gland of freely moving rats, by means of on-line microdialysis. Furthermore some exploratory pharmacological experiments are described, indicating the potential of this technique. Basal output during daytime was low, but well within the limits of the analytical capabilities. A marked increase was recorded during night-time. TTX was used to investigate the possible neuronal drive behind melatonin production. Such a drive appeared to be present during night, because melatonin levels decreased substantially during TTX perfusion. During daytime however, even high concentrations of TTX were not able to alter the melatonin production, indicating that no neuronal drive was respon sible for the low daytime output of melatonin. To exploit the high time resolution of the technique, the effects of short term changes in noradrenaline release (input) on melatonin secretion (output) were studied. A one minute light pulse was applied around midnight and resulted in an immediate decrease of both noradrenaline and melatonin. While noradrenaline returned to basal levels in 40 min, mela tonin did not reach the baseline within 2.5 h. This discrepancy in correlation between noradrenaline and melatonin indicates a rapid inac 3. Characterization and pharmacological applications of a new technique 73

tivation of N-acetyltransferase, followed by a slow reactivation, possibly by de novo synthesis of N-acetyltransferase. During daytime, a perfusion with 60 mM potassium for 30 min, resulted in a rapid and short stimulation of noradrenaline release, which was not followed by an increase in melatonin production. These studies show the possibility to study the coupling between noradrenergic innervation of the pineal gland and melatonin production with high resolution. Also, the technically complicated, but versatile dual probe approach was applied to pineal microdialysis. Implanting one probe in the dorsomedial hypothalamus (DMH) and the other in the pineal gland provided a unique situation in which the complex neuronal pathway from SCN to pineal could be studied. In this chapter we describe the effects of the GABA-agonist muscimol perfused in the DMH on both noradrenaline innervation and melatonin production in the pineal. Muscimol appeared to be very effective in inhibiting the night-time melatonin production, which was caused by a reduced noradrenaline release. From SCN-lesion studies, it appeared that the origin of this inhibitory input to the DMH is the SCN. Measuring the melatonin profile with microdialysis after lesioning the SCN did not only show an effect on rhythmicity. The profile was flat, with no circadian variation whatsoever, but the absolute level of melatonin was increased compared to normal daytime production. This presentation of a combination of various experiments shows the auxiliary possibilities of trans pineal microdialysis in pharmacological, biochemical, anatomical and chronobiological research.

Most data presented in this chapter are published in the following papers: Drijfhout WJ, Grol CJ and Westerink BHC (1993) Microdialysis of melatonin in the rat pineal gland: methodology and pharmacological applications. Neurochem. 61, 936-942. J. Drijfhout WJ, VanderLinde AG, DeVries JB, Grol CJ and Westerink BHC (1996) Microdialysis reveals dynamics of coupling between noradrenaline release and melatonin secretion in conscious rats. Neurosci. Lett. 202, 185-188. Kalsbeek A, Drijfhout WJ, Westerink BHC, VanHeerikhuize JJ, VanderWoude TP, VanderVliet J and Buijs RM (1996) GABA receptors in the region of the dorsomedial hypothalamus of rats are implicated in the control of melatonin and corticosterone release. Neuroendocrinology 63, 69-78.

74

3. Characterization and pharmacological applications of a new technique

3.1 Introduction
Over the last decades, extensive research has been performed on the pineal gland and its neuroendocrine hormone melatonin. Now that much is known about the biosynthesis and its regulation,348,396 the important role of melatonin as part of the biological clock has become evident.46,212,280 Thus far, however, most of these studies have been donein vitro with isolated glands, or in vivo by decapitation and dissecting the gland. The microdialysis technique has found a widespread use for in vivo monitoring of 377,406 neurotransmitter release in a variety of neuronal tissues. Dialysates are relatively clean and can easily be analyzed by HPLC, because the samples can be injected directly without clean-up procedure. The high sensitivity for electro-active compounds like catecholamines136 and the fact that the animals are conscious during the experiments contributed to its popularity. 19 Microdialysis of the pineal gland was first described by Azekawa et al. They used an I-shaped cannula to collect the samples. While melatonin was determined by HPLC and electrochemical detection,103,120 one of the problems they encountered was the level of sensitivity. Although their detection limit for melatonin was as low as 5pg/sample, some of the daytime samples could not be measured. However, for circadian studies daytime values are of crucial importance. The use of fluorescence detection applied in our studies dealt with that problem and allowed a variety of different kinds of microdia lysis experiments in the pineal gland. This chapter describes some of those experiments, starting with a characterization of this relatively new technique. Data on basal output, day/night variation, and the possible involvement of a neuronal drive in melatonin production, assessed by the use of TTX, provide a basic understanding of the kind of experiments described in this thesis. The role of the sympathetic nervous system in stimulating melatonin production is well known.348,280 However, the dynamics of the coupling between noradrenaline and melatonin in terms of its time dependency, is largely unknown. Short manipulations of the noradrenergic innervation may reveal thedynamics of this noradrenaline/melatonin coupling. It is well known that short light pulses during the dark period are very effective in suppressing pineal N-acetyltransferase activity35,133 and melatonin production.150 De397 pending on light intensity, pulses of several minutes to even milliseconds suffice to inhibit pineal metabolism. Furthermore, a short stimulation of sympathetic neurons can 408 be achieved by depolarization with potassium chloride. Also in the pineal gland this has been described as being useful to generate a short sympathetic stimulation during daytime.58 In chapter 4, the relationship between noradrenergic innervation and melatonin production is extensively studied, but in this chapter we focus on the dynamics of changes in sympathetic innervation on pineal melatonin productionin vivo, made possible by the high time resolution. The neural pathways involved in the regulation of the circadian rhythm in pineal activity are only partly known. Information from the SCN reaches the sympathetic system via a not well established pathway in which the dorsal hypothalamus takes a central posi tion.30,119,157 Recently, a GABA-containing projection from the SCN to the paraventricu 3. Characterization and pharmacological applications of a new technique 75

lar and dorsomedial hypothalamus was described (see Fig.1.15, page 37).38 In addition, the GABA-containing innervation of the dorsomedial and posterior hypothalamus has 208,209,321,400 been implicated in the control of sympathetic outflow. Accordingly, in this chapter, a study is described which determines whether the GABAergic projection from the SCN might be involved in depressing the activity of neurons in the hypothalamus which are associated with the control of the circadian melatonin rhythm, using the dual probe approach. These experiments are part of a large study addressing the role of GABA receptors in the dorsomedial hypothalamus (DMH) in the regulation of melatonin and corticosterone148 and was a collaboration with Dr. A. Kalsbeek, Netherlands Institute for Brain Research. 310 The activity of the SCN is in opposite phase with many other rhythms. The inhibitory 148 projection of the SCN to other areas as suggested, may well result in a stimulation of melatonin production after lesioning the SCN. Under such conditions, generally most 217 of the endogenous circadian rhythmicity is lost. The present study describes the effects of SCN-lesions as measured with microdialysis and was carried out in collaboration with Dr. G. Boer, Netherlands Institute for Brain Research and Prof. Dr. W. Rietveld, University of Leiden. The fact that both circadian information and data on absolute melatonin production are obtained simultaneously, is an advantage of this approach and another exploratory application of trans pineal microdialysis.

3.2 Experimental setup

Animals were used as described on page 58, except for the SCN-lesions, where male Wistar albino rats were obtained from the University of Leiden. Animals underwent surgery as described on page 59, one or two days before the experiments. Melatonin production was measured as described on page63, under basal conditions both in daytime and during light switch off, until several hours in the dark period. In a -6 second experiment, melatonin was measured during TTX perfusion (10 M, 1 h) both in the light and dark period. In the light period, melatonin was also measured following -5 perfusion with TTX in a concentration of 10 M (1 h). The dynamics of the coupling between noradrenergic innervation and resulting mela tonin production was done by measuring both melatonin (page63) and noradrenaline (off-line, page 67) in separate experiments. Their response to a 1 min light pulse (300 lux) at circadian time (CT) 18 and perfusion with potassium (60 mM, 30 min) during the light period was recorded.

76

3. Characterization and pharmacological applications of a new technique

SCN-lesion experiments SCN lesions were made under anaesthesia (Nembutal 60 mg/kg i.p.). A stainless steel electrode (diameter 0.4 mm) was inserted bilateral in the SCN with the aid of a David Kopf stereotaxic apparatus. The coordinates used for lesions were: 1.6 mm anterior from bregma, 0.5 mm lateral from midline, and 9.3 mm ventral below the dura, according to the atlas of Pellegrino. Lesions were made by leading a 1.5 mA anodal DC for 15 s through the electrode. To ensure that lesioned rats were completely arrhythmic, feeding and drinking behaviour was measured continuously for 4 weeks as described in detail by Rietveld et al.145 Ten arrhythmic rats were then transported by car from the Department of Physiology of the University of Leiden where the lesions were applied, to the Depart ment of Medicinal Chemistry of the University of Groningen. Upon arrival they were adapted to their new environment for one week before a microdialysis probe was implanted as described on page 59. One day after surgery, the melatonin rhythm was measured for 24 h. After the microdialysis experiments, animals were returned to Leiden for a further recording of drinking behaviour for two weeks. Finally the rats were anaesthetized with Nembutal (60 mg/kg i.p.) and perfused with 10% buffered formal dehyde. The brains were removed, kept in buffered formaldehyde for 1 week and then cut in 50 m frozen sections through the SCN area. The location and extent of the lesions was verified and defined on the basis of Kluver-Barrera myelin stained sections.

Figure 3.1 The effect of TTX on melatonin production. TTX was perfused in a -6 concentration of 10-6 M ( , n=4) during night-time and in concentrations of 10 M ( , n=4) and 10-5 M ( , n=3) during daytime. All perfusions lasted for 1 h and started at t = 0 min. Melatonin is expressed as percentage of mean day- or night-time levels, whichever is applicable and is presented as the mean S.E.M.

3. Characterization and pharmacological applications of a new technique

77

Figure 3.2 Day/night differences in the production of melatonin. Melatonin was sampled from the pineal gland from 1 h before until 5 h after lights off. Data are expressed as percentage of mean daytime level and presented as the mean S.E.M. (n=6).

Dual probe The dual probe experiments were carried out after implanting both a dialysis probe in the pineal (page 59) and in the dorsomedial hypothalamus (DMH) during one surgical procedure. The DMH probe was implanted by Dr. A. Kalsbeek, Netherlands Institute 148 for Brain Research, as described elsewhere. The U-shaped dialysis probe consisted of a dialysis fiber (molecular weight cut-off 6000 Da, 3 mm total length), glued into the ends of two parallel 25G stainless steel tubings. It was implanted just lateral to the DMH, at coordinates with flat skull of 2.8 mm caudal to bregma, 1.6 mm lateral to the midline and 8.0 mm below the brain surface. The loop of the probe was positioned in the rostro-caudal direction along the DMH. During the experiments, melatonin was measured as described before (page63). Noradrenaline was measured (on-line, page 66) in two animals that showed a marked effect on melatonin. The DMH probe was continuously perfused with Ringers solution at a flow rate of 3.0 l/min. Perfusion occurred by replacing Ringers solution with muscimol in a concentration of 10-5 M. Perfusion lasted for 40 min, although in one experiment two perfusions of 20 and 40 min respectively were carried out.

78

3. Characterization and pharmacological applications of a new technique

3.3 Results
Basal levels of melatonin The basal values varied among different animals, which indicated little differences in th e exact placement of the cannula. The average output of melatonin during daytime in the dialysates was 0.8 0.1 fmol/min (mean SEM; n=28). Generally, this implicated that typical daytime dialysates contained about 4 pg (16 fmol) in 20 min samples and 5 pg (24 fmol) in 30 min samples. TTX sensitivity of melatonin production Infusion of the sodium channel blocker TTX did not affect melatonin production during the light period. In Fig. 3.1 the results are shown of a 1 h perfusion with 10-6 M and 10-5 M TTX respectively. No significant change in melatonin output was obser ved. During -6 the dark period however, perfusion with 10 M TTX for one hour resulted in a marked decrease of melatonin levels to 28.5 3.2 % of basal levels (Fig. 3.1). Day/night differences in melatonin production In Fig. 3.2, the pattern of melatonin contents in the dialysates is shown during the last hour of the light period and the first five hours of the dark period. An increase in melatonin levels was observed, reaching a plateau value after four hours of about 1500 % compared to average daytime levels. In absolute amounts, this reflected an output of melatonin of about 12.3 1.9 fmol/min.

Figure 3.3 The effect of SCN lesions on melatonin production. Melatonin was measured for 24 h in constant darkness. Animals had their SCN lesioned and were arrhythmic in drinking behaviour. Melatonin is expressed as percentage of mean basal levels during the first two hours of the experiment and presented as the mean S.E.M. (n=6).

3. Characterization and pharmacological applications of a new technique

79

Effect of SCN-lesioning on melatonin production Fig. 3.3 shows the effect of lesioning the SCN on pineal melatonin production. The animals did not show any circadian variation in their melatonin production. However, the absolute amount of melatonin released was different. In three animals the output was much higher than under normal conditions. Their average output was 204.0 61.5 fmol/sample, which was in the range of normal night-time production. In the other three animals, basal output was 12.9 7.3 fmol/sample, which was comparable to normal daytime production. The average basal output of all six animals was 108.4 50.9 fmol/sample. It appeared to be difficult to correlate differences in basal output with the size and location of the lesion. In one animal, in which the absolute output was 27.1 fmol/sample, the SCN was not completely lesioned. From the drinking behaviour re cordings after the microdialysis experiments, it appeared that in this case some rhyth micity was restored. However, two other animals which had normal basal levels, did have complete lesions and did not show any rhythmicity in drinking behaviour whatso ever. The animals that showed high basal output of melatonin appeared to have also correct lesions and did not show rhythmicity in drinking behaviour after the microdialysis experiments. Effect of one minute of light on melatonin and noradrenaline release In Fig. 3.4, the effect of a one minute light pulse (300 lux) at circadian time (CT) 18 on both noradrenaline and melatonin are shown. Since CT 0 is the beginning of the light period, CT 18 is defined as the middle of the dark period. As is clear from the data, a maximal decrease of noradrenaline levels occurred within one sample (20 min), which was 24.5 6.2 % of basal night-time levels. This effect was significant for 40 min. At t = 80 min, the noradrenaline release had returned to baseline values. For melatonin, this effect lasted longer and the effect was even more pronounced. The initial decrease was similar to the decrease in noradrenaline, but lowest levels were reached at t= 60 min (6.6 2.7 %). Especially the recovery took longer than for noradrenaline. Significant lower levels were measured until t = 140 min, but basal levels of melatonin were not obtained within 2.5 h after the light pulse. Effect of potassium chloride on melatonin and noradrenaline release Stimulation of noradrenaline release with potassium in a concentration of 60 mM for a period of 30 min around CT 7, increased the low noradrenaline levels. As is shown in Fig. 3.5, the increase in noradrenaline release was rapid and significant to 286 60 % above baseline. At T = 60 min, levels had returned to basal values. Since basal pineal output of noradrenaline under daytime conditions was very low, cocaine, a re-uptake inhibitor of noradrenaline, was added to the perfusion medium in a concentration of 10-6 M. This resulted in an average basal output of 7.8 0.5 fmol/sample. As shown in chapter 4, adding cocaine up to a concentration of 10-5 M did not change the pharmacology of the system qualitatively. Unlike the light pulse experiment, melatonin produc tion did not follow the pattern of noradrenaline release. Throughout the experiment, no changes in melatonin production were obser ved.

80

3. Characterization and pharmacological applications of a new technique

Figure 3.4 The effect of a one minute light pulse on melatonin ( , n=5) and noradrenaline ( , n=4). Lights were turned on for 1 min at t= 0 min (CT18, midnight). Data are expressed as percentage of mean night-time levels and presented as the mean S.E.M. Asterisks (*) indicate statistical significance (p 0.05). <

Figure 3.5 The effect of potassium chlorideon melatonin ( , n=5) and noradrenaline ( , n=5). Potassium was perfused in a concentration of 60 mM for 0.5 h, starting at = t -6 0 min. Noradrenaline was determined in the presence of cocaine (10 M) throughout the experiment. Data are expressed as percentage of mean night-time levels and presented as the mean S.E.M. Asterisks (*) indicate statistical significance (p 0.05). <

3. Characterization and pharmacological applications of a new technique

81

Figure 3.6 The effect of muscimol in the DMH on pineal melatonin production. -5 Muscimol was perfused in the DMH in a concentration of 10 M for 20 and 40 min, starting at t = 19 h and t = 21 h respectively. At t = 24 h, lights went on. Melatonin is expressed as percentage of mean nightime level and presented as a single experiment.

Effect of muscimol in DMH on pineal melatonin and noradrenaline release -5 Perfusion with muscimol in a concentration of 10 M in the DMH caused an immediate and rapid decrease of melatonin levels (Fig. 3.6 and 3.7). A 40 min perfusion caused melatonin levels to decrease to approximately 40 % of the basal levels within 60 min. Recovery to basal levels commenced 100 min after the start of the muscimol perfusion, but was sometimes obscured by circadian decrease at the end of the night (Fig.3.6). Noradrenaline levels in the pineal showed decreases to about 30 % of basal levels, following muscimol perfusion (Fig 3.7).

82

3. Characterization and pharmacological applications of a new technique

3.4 Discussion
Basic considerations The first results indicated that it is well possible to monitor day- and night-time produc tion of melatonin directly in the pineal gland of freely moving animals. Therefore it may be a promising method for studying the neuroendocrine regulation of melatonin. Azek awa et al.19,20 were the first who used this technique to study the melatonin levels in the pineal gland. They used electrochemical detection in their studies, with a detection limit of 5 pg/sample, which made it impossible to see daytime melatonin levels in some animals. The fluorescence detection, used in our experiments, is more sensitive (detection limit of 1 pg/sample), specific for indolic compounds and very stable. Therefore, approxi mately 90% of the operated animals could be used in the experiments during daytime. The remaining 10% of the animals could not be used, due to incorrect placement of the cannula, or technical problems, such as obstruction of the cannula etc. The use of a trans pineal cannula prevented us from hitting the sinuses just above the pineal gland and caused only minor damage to brain tissue. The sharpened point of the tungsten wire cutting through the fleeces surrounding the gland, made it more easy to position the cannula exactly in the pineal. Rats recovered very fast from surgery, and could be used for experiments the subsequent day.

Figure 3.7 The effect of muscimol in the DMH on melatonin (left panel) and noradrenaline (right panel). Muscimol was perfused in the DMH in a concentration of 10-5 M for 40 min, starting at t= 0 min. Both melatonin and noradrenaline are expressed as percentage of mean night-time level and presented as the mean S.E.M. (n=5 for melatonin, n=2 for noradrenaline).

3. Characterization and pharmacological applications of a new technique

83

The percentage increase in melatonin levels during the night-time corresponded well 413 to that found in homogenated-tissue levels. Also the fact that it takes about three to four hours after the beginning of the dark period before melatonin levels reach their 242 maximum values, is in agreement with studies on post-mortem tissue. Compared to 19 the in vivo data of Azekawa et al., the day and night-time levels are in the same range. Hormone production is different from neurotransmitter release When studying the release of neurotransmitters such as catecholamines or acetylcholine, TTX is capable of inhibiting this neuronal release completely, by blocking sodium channels of the axons.236,407 Melatonin however, is not a neurotransmitter, but a hormone. Differences in its concentration represent differences in its biosynthesis, which can be stimulated and regulated by neuronal input. To demonstrate this difference between hormones and neurotransmitters, we investigated the effects of TTX. Even in high concentrations, TTX did not affect the biosynthesis of melatonin during daytime. From this we conclude that it is most likely that neuronal activity is absent during daytime. Melatonin levels represent some basal level ofN-acetyltransferase activity, which is not dependent on neuronal input. One could speculate that neuronal activity is present, consisting of both a stimulatory and inhibitory component. In this case TTX would block both the stimulating and inhibitory components at the same time, with no net results on melatonin levels. However there is no supporting evidence for this speculation. During the dark period, the situation is quite different. Now TTX does inhibit the melatonin release in an efficient way. Since under these circumstances all innervation is non-selectively blocked, it does not reveal the nature of the innervation responsible for the high level of melatonin production during night. On the other hand, it does prove that there is neuronal activity in the pineal during night and that this activity is necessary to have high levels of melatonin. SCN is negatively coupled to melatonin production In the regulation of a rhythmic melatonin release from the pineal, the SCN plays a crucial role. The mechanism by which SCN information is transferred to the pineal is still unsolved, but the involvement of other hypothalamic nuclei has been suggested. The fact that SCN activity is high during daytime and low during night-time, suggest an inverse relationship between SCN activity and melatonin production. Studies in SCN-lesioned monkeys,286 sheep365 and rats148 indicate that after SCN lesion, melatonin levels are elevated compared to normal daytime levels. The present data, in which the average melatonin output in SCN-lesioned animals was increased towards normal daytime levels, appear to confirm these findings and indicate that melatonin production is stimulated by an area upstream from the SCN, which is under rhythmic inhibition of the SCN. Ablation of the SCN therefore results in continuous melatonin release at a level above normal daytime levels. The anatomical localization of such a stimulatory nucleus is not known, but based on the large variation in melatonin levels obtained in this study and conflicting reports on this point, it may well be a nucleus close to the SCN, which can be easily damaged or removed when lesioning the SCN. The inhibitory GABA projection from the SCN to the DMH which is confirmed in the present dual-probe experiment in which muscimol in the DMH was able to suppress melatonin production in the pineal,

84

3. Characterization and pharmacological applications of a new technique

may be another indication of a continuous inhibitory projection from the SCN to target tissues. The DMH might be a candidate for the stimulatory nucleus involved in melatonin regulation. Dynamics of coupling between noradrenaline and melatonin From the literature, it is known thatN-acetyltransferase activity, which is high during the night, is decreased following treatment with-adrenergic blockers399 or light.131,150 This 72,389 decline is biphasic, with reportedhalf-lives of 3.8 and 70 min respectively. Consistent with this is the rapid decrease in melatonin release following a one minute light pulse recorded in this study. The exact overlap between noradrenaline and melatonin in this decrease emphasizes that noradrenaline release is an absolute prerequisite forN-acetyltransferase activity and melatonin production. The impairment in increase between noradrenaline and melatonin after the light pulse is more complicated. Vanecek and Illnerova reported that N-acetyltransferase activity started to increase within 5 min after stimulation with isoproterenol.390 In their experiments, isoproterenol stimulation oc curred already 7 min after the light pulse. The addition of protein synthesis inhibitors reduced basal levels of N-acetyltransferase activity and hindered its restoration after isoproterenol stimulation. Therefore it seems that N-acetyltransferase is continuously being synthesized and degraded. The longer N-acetyltransferase synthesis is inhibited, the less N-acetyltransferase is left. The present data show that noradrenaline release was inhibited for at least 40 min, leaving very little N-acetyltransferase left. N-acetyltransferase was newly synthesized in response to increased noradrenaline release. This caused the melatonin levels to take 2.5 h to return to basal levels. Also the effects of potassium chloride are consistent with this theory. The short-term increase in neurotransmitter release following 30 min perfusion with potassium chloride 58 is characteristic and reported in the pineal as well as other tissues.408 The elevated levels of noradrenaline will undoubtedly start to increase the synthesis ofN-acetyltransferase. However, since the noradrenaline levels remain elevated for only 30 min, the process of increasing N-acetyltransferase activity does not reach the level as to which it results in detectable increases in melatonin production. When stimulation of noradrenaline release stops, newly synthesized N-acetyltransferase will degrade rapidly, so neither during, nor after stimulation of noradrenaline release will the melatonin production increase. The possibility that melatonin production can not be elevated during daytime because of a direct inhibitory effect of light can be excluded. In the first experiment it was shown that the inhibitory effect of light was in fact mediated through inhibition of noradrenali ne release. Furthermore, long term stimulation of -adrenergic receptors actually does stimulate melatonin production during the light period. In chapter4 we show that perfusion with 10-6 M isoproterenol for 1.5 h resulted in marked increases in melatonin production.77 As would be expected from the above, RNA synthesis should play an important role in the induction of N-acetyltransferase. Literature data, although sometimes controver sial, indicate that this is indeed the case (for a review, see ref.353). Early studies have indicated that increased levels of cyclic AMP following-adrenergic stimulation increased 294 RNA synthesis, which could be blocked by actinomycin D, an effect that even showed 432 233 a circadian variation. Later, Morrissey and Lovenberg showed no changes in [3H]3. Characterization and pharmacological applications of a new technique 85

uridine incorporation into the RNA following isoproterenol stimulation. In addition, they reported a lack of effect of isoproterenol in changing the tissue level of ornithine decarboxylase.271 Controversially, an increase in melatonin production in Y79 human 142 retinoblastoma cells is described following RNA synthesis inhibitors. In conclusion, it was shown that noradrenaline release and melatonin production in the rat pineal gland are closely coupled. Decreases in noradrenaline release are immedi ately followed by decreases in melatonin production, whereas increased noradrenaline release is followed by increased melatonin only after a lag time. Short term increases in noradrenaline release will therefore not influence the melatonin production during daytime. Microdialysis by its nature is able to detect these time dependent changes with relatively high time resolution and is therefore suitable to study the relationship between innervation and output of peripheral tissues in conscious rats. Dual probe method opens even more possibilities In the present study a pronounced GABAergic inhibition of melatonin production was demonstrated at the level of the DMH. Infusion of muscimol in the DMH rapidly decreased melatonin, by reducing the noradrenergic input. Recently a GABA containing projection from the SCN to PVN and DMH was described (see Fig.1.15, page 37).38 Furthermore, elevated plasma melatonin concentrations in SCN-lesioned animals could 148 be inhibited by infusion of muscimol into the DMH. Taken these data together with the present data, apparently there is a GABA-containing projection from the SCN to the DMH that transmits an inhibitor y signal on melatonin secretion, mediated via the sympathetic innervation of the pineal gland. This tonic GABAergic inhibition of sympa thetic innervation at the level of the hypothalamus may not be restricted to melatonin production. Previous studies have shown similar effects on other components of sympa 208,209,321,400 thetic outflow as well, i.e. heart rate, blood pressure and plasma noradrenaline. Because SCN-lesions cause increased daytime heart rate and blood pressure meas ures,145,402 it may be that these components of the sympathetic system are under the inhibitory control of an SCN derived GABAergic input to the hypothalamus, similar to the pineal gland. The use of dual probe microdialysis has revealed important information on neuronal connections between different brain areas. Implanting probes in a combination of two 309 brain areas such as substantia nigra and striatum, septum and hippocampus228 are good examples. The combination DMH and pineal is yet another example of the additional value of this method to anatomical studies. Combinations of SCN/PVN and pineal would be a logical next step.

86

3. Characterization and pharmacological applications of a new technique