Brain Research Bulletin, Vol. 4, pp. 725-730. Printed in the U.S.A.

Effects of Melatonin on Spontaneous and Evoked Neuronal Activity in the Mesencephalic

Reticular Formation’

JORGE H. PAZ0

Laboratorio de Neurofisiologia, 2da. Catedra de Fisiologia Humana, Facultad de Medicina, Universidad de Buenos Aires, Paraguay 2155, 1121-Buenos Aires, Argentina.

Received 7 August 1979

PAZO, J. H. Effects of melatonin on spontaneous and evoked neuronal activity in the mesencephalic reticularformation. BRAIN RES. BULL. 4(6)72%730, 1979.-The acute effects of melatonin on the spontaneous activity of single cells in the mesencephalic reticular formation were studied in 40 mate rats unanesthetized and immobilized with Flaxedil. One hundred and ten neurons were explored. Only 64 modified their spontaneous activity after the intravenous administration of melatonin. This response consisted of an increase in neural firing (6 neurons), decrease (55 neurons) and biphasic response of decrease and increase (3 neurons). When the effect of melatonin on the evoked activity in the mesencephalic neurons by peripheral stimulation (sciatic and photic) was checked an increase of the number of neurons that showed inhibitory responses to photic stimulation was found. No changes in blood pressure and EEG were observed at the doses of melatonin used (200,400 and 600 &lOO g of body weight). However, with doses of 600 pg a tendency toward synchronization was seen in the EEG. The present observations indicate an inhibitory effect of melatonin on the spontaneous neuronal activity of the mesencephalic reticular formation. This effect may contribute to the changes in the sleep-wakefulness cycle and anticonvulsant action attributed to this hormone.

Melatonin effects MRF unit activity Melatonin and MRF activity

IT HAS been demostrated that melatonin, a hormone syn- thesized by the pineal gland produces some of its endocrine and non-endocrine effects by acting on the brain [S, 7, 12, 161. Changes in the EEG and convulsive threshold have been observed after melatonin administration in mammals and humans [3, 8, 121. Pinealectomy in rats and young chickens partially reverses such effects [ 14,151.

The sleep induced by melatonin was attributed to its ability to act on the hypothalamus. Crystal implants of this methoxyindole in cat hypothalamus is followed by sleep with the characteristic electroencephalographic pattern of slow- wave sleep [12]. On the other hand, the intraperitoneal injection of melatonin increases GABA and serotonin con- centration in the hypothalamus and midbrain. These sites show the highest uptake of the circulating tritiated melatonin 11, 2, 41.

The following study describes the effects of acute melatonin administration upon spontaneous and evoked unit activity in the mesencephalic reticular formation (MRF) of the rat.

METHOD

Experiments were performed on 40 male rats of the

Wistar strain weighing 245.80 +. 3 1.41 g. A tracheal cannula was inserted under ether anesthesia. The contralateral scia- tic nerve to the recording site, in the mesencephalic reticular formation, was exposed and fitted with a bipolar stimulating electrode.

Body temperature was maintained with a heat lamp and the heart rate was continuously monitored. A femoral vein was routinely cannulated for drug administration. In ten animals the blood pressure was recorded by means of a catheter placed in the femoral artery.

After completion of initial surgery the animals were placed in a stereotaxic instrument (David Kopf) and a craniotomy was performed. The exposed cortex was covered by agar solution in order to diminish pulsations. Wound edges and pressure points were infiltrated with xylocaine. Additional application of local anesthetic was carried out when necessary as judged by changes in heart rate and pupilhuy diameter. Gallamine triethiodide (Flaxedil) was administrated intravenously, the ether discontinued and artificial respiration started by means of a positive pressure respirator (Narco-Biosystem, V-1OOKG). Stroke rate and tidal volume were adjusted according to the Harvard venti- lation chart [9].

‘The author wishes to thank Mr. 0. Rivas for his technical assistance in the construction of the discriminator-integrator. This research was supported in part by a grant from the CONICET, Argentina.

Copyright a 1979 ANKHO International Inc.-0361-9230/79/060725-06$01.10/O

726 PAZ0

Local anesthesia was selected in preference to general anesthesia in order to assure the maximum sensitivity of the neuronal system under study.

Extracellular unit activity was recorded with stainless steel microelectrodes electrolytic~ly sharpened to a tip diameter of 1-3 wrn and mounted in a hydraulic microdrive. The depth of the microelectrode was determined on the scale of the micrometer. The microelectrode was capacity coupled to a low gain preamplifier (Ortec model 4661) and amplifier (Ortec, model 4660). The amplifier output was connected to an audiomonitor {Epic), displayed on an oscilloscope (Tek- tronix 565) and photographed by a Nihon-Kohden oscillo- scope camera. The amplified signal was put through an amplitude discriminator-integrator [ 161. The outputs from the disc~minator-integrator were sent to a polygraph (San-Ei-Biophysiograph, 110 system) for permanent records. The square pulse matched to each input signal from the discriminator was also observed on the second channel of the oscilloscope.This permits the superimposition of each action potential with its corresponding output pulse from the Schmidt-trigger to insure accurate recording.

In 20 rats the EEC was recorded from two galvanized screws cemented with dental acrylic over the frontal cortex and displayed on the polygraph and an oscilloscope.

Contralateral sciatic nerve was stimulated through an isolation unit (4V, 0.5 msec) by means of a Tektronix waveform and pulse generator. Photic stimuli were delivered from a Berger photostimulator. Ten single stimuli were applied every 2 set for each sensory modality and their effects on the unit firing were observed for 2000 msec after each stimuli.

One hour or more after the end of the surgical procedure, recording the unit activity in the MRF was started. Targets were between stereotaxic planes A: 1.020 to A: 0.350 accord- ing to the atlas of Kiinig and Klippel [ 1 If. After a period of spontaneous activity of 10 to 15 min, each of the two series of the sensory modalities were applied. This was followed by another spontaneous activity period of 10 min. Subsequently the melatonin was injected (IV). Following the injection, the spontaneous activity was again recorded for a period of 5 min and the two series of stimuli were then repeated. After this, unit discharges were monitored until they approached the pre-injection activity levels.

Firing rate for each unit was asseverated by direct mea- surement of the integrator output every five seconds. Spon- taneous activity after melatonin was expressed as percent of the mean pre-injection firing rate for 10 min before drug administration. A positive increase or decrease in firing rate following hormone administration was defined in each case as exceeding * 40% of control rate for a period of 3 min or more, since spontaneous pre-drug activity never exceeded this value.

The significance of the change in the firing rate caused by somatosensory stimulation was estimated by comparing, from the film, equal periods of spontaneous and evoked activity using the chi-square test of the following formula: xZ=(E-S)I(E+S)*p, where pKO.05 and E=evoked activity: S=spontaneous activity [13,17].

Melatonin (8 mg, Sigma) was dissolved in 0.2 ml of propylene glycol and made up to 1 ml with physiological saline. The final concentration was 400 pg in 50 ~1. The solution was prepared daily. Melatonin was injected intravenously at the doses of 200,400 and 600 pEL&I per 100 g of body weight. The control animals were given the same amount of melatonin diluent. Each animal received no more

TABLE 1 CHARACTERISTICS OF MELATONIN RESPONDING UNITS (n=W

Response No. of units 7c of all units

Increase 6 9.37

Decrease 55 89.93

Biphasic 3 4.68

than three melatonin injections at intervals of 60 min or longer.

Unit activity that failed to return to within r 40% of pre-injection levels was discarded.

At the end of the experiments a DC electrolytic lesion was made through the microelectrode in its lowest position. In some animals, to check distortions that appear during the histological procedure, two lesions were made along the electrode track: one at the end of the penetration and another at the first site where a cell activity was recorded. The brains were fixed in 10% neutralized formalin and the recording positions were verified histologically, on frozen sections of 50 Frn to 100 pm stained with gallocyanin.

RESULTS

The spontaneous unit discharges of 110 neurons were recorded from the ventral mesencephalic reticular forma- tion. Sixty four (58.18%) cells showed signi~~ant changes in the firing rate after melatonin administration, while 46(41.81%) neurons did not modify their activity. The melatonin responding neurons were classified as (a)those that showed increased or decreased frequencies of firing; and (b)those which had biphasic responses of decreases and increases. Table 1 summarizes the results.

Melatonin decreased the firing rate in 55 midbrain cells (Fig. 1) with a mean + SE latency of 3.5 r 0.95 min (n=55). Recovery from the melatonin influence was between 3 and 42 min (13.8 t 2.2 min; n=55). It was dose dependent, having a longer recovery with larger doses (Table 2).

Only six neurons showed sustained changes above their control firing rate after melatonin administration (Fig. 2). The time of onset of this effect was 4.5 t 2.47 min (n=6) and lasted 15.30 ? 5.48 min (n=6).

The biphasic responses (3 neurons) consisted of a de- crease followed by an increase in the neuronal discharges (Fig. 3).

When the whole population of neurons that respond to melatonin were analyzed according to their distribution of firing frequencies, the hormone produced a shift from rela- tively high frequencies during control periods to lower frequencies after melatonin. This effect is well observed ten minutes following hormone administration (Fig. 4).

No evident differences regarding the effect (increase. decrease or biphasic) were observed at doses of 200 and 400 ~g/lOO g of b.w of melatonin. Only when 600 CL&/l00 g of b.w was’used the neurons always responded with a decrease of the firing rate.

Control injection of melatonin diluent elicited responses in 10 out of 90 neurons investigated with the solvent. This effect consisted of a decrease in the spontaneous firing rate. Only 7 neurons of those 10 were inhibited by melatonin. The

MELATONIN AND NEURONAL ACTIVITY 721

Time after Melatonin in minutes

FIG. 1. Decrease of firing rate of two neurons in the MRF after melatonin at different doses. (A) with 200 &lOO g of b.w. and (B) with 600 pg/lOO g of b.w. The drug was administered at time zero. The

neuron activity is shown on a minute by minute basis.

TABLE 2

RECOVERY TIME OF 55 NEURONS IN THE MRF AFTER DIFFERENT DOSES OF MELATONIN IN PERCENT OF ALL UNITS IN EACH GROUP. THE

NUMBER IN PARENTHESES INDICATES THE NUDER OF UNITS

Doses l-10

200 c1g (15) 75%

400 Kg (30) 47% 600 cLg (lo) 0%

Recovery time in minutes >lo-20 >20-30 >40-50

IS% 10% 0% 47% 6% 0%

0% 0% 100%

Time after Melatonin in minutes

FIG. 2. Increase of firing rate of a single neuron in the MRF after melatonin (400 ~g/lOO g of b.w.) administered at time zero. The

neuron activity is shown on a minute by minute basis.

FIG. 3. Biphasic response of one neuron in the MRF after melatonin (400 ~g/lOO g of b.w.) administered at time zero. The neuron activity

is shown on a minute by minute basis.

A

Spikes/see

FIG. 4. ~js~~~~~iofl of resting firing rate of melatonin responding neurons (n-64) befofe and after drug administ~tj~n. (Al control. (B) 5 min after m&t&n, and (C) IO min after melatctnh

TABLE 3 PERCENTAGEANDNUMBEROFUNlTSNQTRESPONDlNGANDRESPQNDlNGSlGNJFICANTLYTOPERlPHERAL STIMULA~ONBEFOREANDAFTERMELA'TOFIIN.TNENUMBERINPAREhlTWESESINDfCATESTHENUMBEROF

UNITS

MetatORin R of all units %, of responsive units

Not respcmsive Responsive I m-ease Decrease

Sciatic (79)

Photic C30)

Before 41.77 (33) 58.23 (46) 80.43 (37) 19.57 f9)

Afier 31.65 (25) 68.35 {S4) 74.07 f40) 25.93 t 141

Befare 56.66 f 17) 43.34 (13) too iis 0 fQ& After 45.66 f 14) 53.34 (Id) 25 (4) 75 (121

other 3 increased their activity after the hormone. The inhibitory effects of diluent when given alone had shorter durations (5.33 Z+Z 1% min; n=10) as compared with the effect of melatonin (f3.80 rt 2.2 min; n=55).

The ~~~istrat~on of m&tonin (~~4~ ~g/ 100 g of b. wf was not associated with changes in rbe EEG patterns as com~~ed with contro1 periods. A tendency to synchroniza- tion was detectable only when larger doses were used (600 ~gilo0 g of b.w). On the other hand, this study confirms the finding of Barchas et at. [5] that the administration of melatonin did not change the blood pressure.

A total of 109 neurons were tested with peripheral stimu- lation. Forty six aut of 79 neurons responded significantly by increasing or decreasing their spontaneous activity after sciatic stimulation (single shock). However, these responses

did not show significant differences before and after melato- nin, although a tendency toward inhibitory respanses was observed foilawing mefatonin (Table 3 and Fig. 5).

In 30 neurons the effects of photic stimulation were recorded. Before me~atoaia, all the responsive units (n= t-7) s~~j~~aa~y increased their firing rate fo~~o~i~g photic stimuIatjo~. There was, however, a signi~Gaat increase in the number of units showing ~a~~b~to~ responses with a diminution of the neurons with facilitatory effects, after admin~strati~a of melatanin (Table 3 and Fig. 5).

DISCUSSION

The present experiments show that melatonin is capable of modifying the electrical activity of single ceils in the ventral mesencephalic reticular Formation of the rat. This effect was mainly a decrease in the spontaneous firing rate of most units recorded (Table I). This effect was independent

MELATONIN AND NEURONAL ACTIVITY 129

IQ Control of the anatomical localization of the neurons into the MRF

and it was of longer duration with the largest doses of melatonin (600 &lo0 g of b.w).

E ‘L 50 z ._ c c

1 Melatonin

1 , Sciatic , , Photic 1

On the other hand, it can be seen that administration of melatonin did not change the degree of responsiveness of the mesencephalic neurons to sensory stimulation. Nevertheless photic stimulation increases the percentage of cells showing inhibitory responses (Table 3 and Fig. 5). This difference in unit responses between photic and sciatic stimulation may reflect different synaptic properties of those sensory path- ways.

Since melatonin readily crosses the blood-brain barrier [10,18], an indirect effect on the mesencephalic neurons by acting on other neural structures connected with the mid- brain, are not ruled out. On the other hand it is not yet possible to say whether this pharmacological action on the reticular formation is due to the compound or to its metabo- lites, or by liberation of vasotocin in the cerebrospinal fluid P61.

The fact that no changes were observed in the EEG after melatonin in the rat is not in agreement with previous studies [3,8,12,15]. This may reflect differences in species, methodology or melatonin doses. In those studies ambula- tory animals (chickens, rabbits, mice, cats) and human beings were used and larger doses were administrated, in some case by different routes. Nevertheless a tendency toward a synchronized EEG was only seen when high doses of melatonin (600 /lg/lOO g of b.w.) were injected.

No sleep was observed after implants of crystalline melatonin in the mesencephalon of the cat [12]. This finding may be explained by the fact that the pool of neurons influenced by melatonin were not sufficient to induce sleep. The other possibility is that the inhibitory effect of melatonin upon cell activity in the MRF, as described in this paper, contributed to the changes in the sleep-wakefulness cycle but was not responsible for such changes.

100 1 FIG. 5. Distribution of responsive units to sensory stimulation before and after melatonin. Control sciatic: 46 neurons: melatonin sciatic: 54 neurons: control photic: 13 neurons: melatonin photic: 16

neurons.

REFERENCES

1. Ant6n-Tay, F. Pineal brain relationship. In The Pineal Gland, edited by G. E. W. Wolstonholme and J. Knight: Churchill Levingstone, London 1971, pp. 213-220.

2. Ant&-Tay, F., C. Chou, S. Anton, R. J. Wurtman. Brain serotonin concentration: elevation following intraperitoneal administration of melatonin. Science 162: 277-278, 1%8.

3. Anton-Tay, F., J. L. Diaz, A. Femandez-Guardiola. On the effect of melatonin upon human brain. Its possible therapeutic applications. Life Sci. 10: 841-850, 1971.

4. Anton-Tay, F., R. J. Wurtman. Regional uptake of 3H- melatonin from blood or cerebrospinal fluid by rat brain. Nature 221: 474-475, 1%9.

5. Barchas, J., F. DaCosta, S. Spector. Acute pharmacology of melatonin. Nature 214: 919920, 1967.

6. Bradley, W., C. Cowan, E. Glover and S. McCormick. Dis- crimunator and integrator instrument for on line frequency analysis of single unit discharges. Electroenceph. clin. Neurophysiol. 22: 177-179, 1967.

7. Fraschini, F., B. Mess and L. Martini. Pineal gland, melatonin and the control of luteinizing hormone secretion. Endocrinology 82: 919-924, 1%8.

8. Hishikawa, Y, H. Cramer, and H. Kuhlo. Natural and melatonin-induced sleep in young chickens. A behavioral and electrographic study. Exp. Brain Res. 7: 8&94, 1%9.

9. Kleiman, L. and E. P. Radford (Jr). Ventilation graph. Prepared for Harvard apparatus. Millis, Mass 02054,U.S.A.

730 PAZ0

10.

11.

12.

13.

14.

Kopin. I. J., C. M. B. Pare, J. Axelrod and H. Weissbach. The fate of melatonin in animals. J. Bid. Chc~m. 236: 3072-3075, 1961. Konig, J. F. R., R. A. Klippel. The Rut Bruin: A stereotaxic atlas of the forebrain and lower parts of the brain stem. The Williams and Wilkins Company, Baltimore, U.S.A. 1963. Marczynsky, T. J., N. Yamaguchi, G. M. Ling and L. Grod- zinska. Sleep induced by the administration of melatonin (5 methoxy-N-acetyltryptamine) to the hypothalamus in unre- strained cats. Experientin 20: 43.5-436, 1964. Nagler. J., N. Conforti and S. Feldman. Alterations produced by cortisol in the spontaneous activity and responsiveness to sensory stimuli of single cells in the tuberal hypothalamus of the rat. Nc,llrorndot.rinol(~~~ 12: 52-66, 1973. Nir, I., K. Behroozi, M. Assael, I. lvriani and F. G. Sulman. Changes in the electrical activity of the brain following pinealec- tomy. Nc,,rroPndoc,rinoloR?; 4: 122-127. 1969.

15.

16.

17.

18.

Pang, F. S., C. L. Ralph and J. A. Petrozza. Effects of melatonin administration and pinealectomy on the electroen- cephalogram of the chicken (Gallus domesticus) brain. L(’ Sc,i. 18: 961-966, 1976. Pave], S., D. Psatta and R. Goldstein. Slow-wave sleep induced in cats by extremely small amounts of synthetic and pineal vasotocin injected into the third ventricle of the brain. Broil7 Rrs. Bull. 2: 251-254, 1977. Takaori, S., M. Sasa and N. Fukuda. Responses of posterior hypothalamus neurons to electrical stimulation of the inferior alveolar nerve and distension of stomach with cold and warm water. Brain Rrs. II: 225-237. 1968. Wurtman, R. J., J. Axelrod and L. T. Potter. The uptake of “H-melatonin in endocrine and nervous tissues and the effects of constant light exposure. J. Phurrntrc. cxp, Tlrr~r. 143: 31&318. 1964.