Camp. Biochem. Physiol. Vol. 104C, No. 2, pp. 221-224, 1993 Printed in Great Britain

0306~4492/93 $6.00 + 0.00 0 1993 Pergamon Press Ltd

DIURNAL VARIATION AND BINDING CHARACTERISTICS OF MELATONIN IN THE MOUSE

BRAIN AND GASTROINTESTINAL TISSUES

GEORGE A. BUBENIK,*~ LENNARD P. NILES,$ SHIU FUN PAN@ and PATRICIA J. PENTNEY*

*Department of Zoology, University of Guelph, Guelph, Ontario, NlG 2W1, Canada (Tel. (519)824-4120; Fax (519)767-1656); SDepartment of Biomedical Sciences, McMaster University, Hamilton, Ontario,

L8N 325, Canada and SDepartment of Physiology, Faculty of Medicine, University of Hong Kong, Hong Kong

(Received 7 July 1992; accepted for publication 19 August 1992)

Abstract-l. Compared to night time values, levels of melatonin (M), determined by specific RIA, were lower in the brain and the duodenum-jejunum segment of mice sacrificed in mid-photophase. In other parts of the gut no significant diurnal difference was observed.

2. Highest daytime levels were detected in the stomach (569pg/ml), lowest in the brain (62pg/ml). 3. Preliminary studies indicate the presence of specific binding sites for [‘Z51]iodomelatonin in the colon,

ileum, jejunum, stomach and brain. These sites, which exhibit an affinity of about 1 nM, may be involved in mediating the gastrointestinal and central effects of melatonin in this species.

INTRODUCITON

The first investigation of M in the tissues of the gastrointestinal tract (GIT) was performed by the Raikhlin group. They observed that the extract from human appendices bleached frog skin. The identifi- cation of the active compound as M was then determined by thin layer chromatography (Raikhlin et al., 1975). Later, this group localized M immuno- histologically in the enterochromaffin (EC) cells of the human appendix and the rabbit small intestine (Raikhlin and Kvetnoy, 1976). Similar immunohisto- logical techniques were used in subsequent years to localize M throughout the entire GIT system (Bubenik et al., 1977), as well as to investigate its ontogeny, circadian variation, and the effects of M treatment or pinealectomy (Bubenik, 1980; Holloway et al., 1980).

In their early publication Raikhlin and Kvetnoy (1976) speculated that M is synthesized in the serotonin-producing EC cells. The hypothesis that melatonin is produced in the gut tissues was later strengthened by the intestinal detection of hydroxyin- dole-O-methyl transferase (HIOMT), the melatonin synthesizing enzyme (Quay and Ma, 1976). Quantifi- cation of GIT melatonin by immunohistology has been attempted by Holloway and co-workers (1980). However, quantification of M in the GIT using a more precise method such as the radioimmunoassay (RIA) was performed only once (Vakkuri et al., 1985).

The determination of M tissue levels by RIA, detection of circadian variations and elucidation of binding sites, an integral part of investigations of other extrapineal production sites, such as the retina

TAuthor to whom correspondence should be addressed.

and the Harderian gland (Hoffman et al., 1985; Pang and Allen, 1986; Dubocovich, 1988) was, until very recently (Lee and Pang, 1992), not performed on GIT tissues. Recent data obtained in our laboratory which indicated a possible melatonin-serotonin interaction involved in GIT activity in the rat and the Swiss-Webster mouse (Bubenik, 1986; Bubenik and Dhanvantari, 1989), renewed our interest in this field and led to continuation of our earlier investigations (Bubenik et al., 1990, 1992).

In view of these preliminary findings, we have decided to investigate diurnal variations of M in the brain and the individual segments of the GIT, as well as to attempt to identify and characterize M binding sites in the brain and the GIT of the above-mentioned species.

MATERIALS ANDMETHODS

Diurnal variation of melatonin

Experimental animals. Two groups of adult Swiss-Webster Mice (six animals per group) were kept on 12/12 light/dark cycle and fed Purina rat chow ad libitum. One group of animals was sacrificed by decapitation in the middle of the photophase, the other in the middle of the scotophase. The brains, stomach and the gut tissues were quickly removed and immediately frozen in liquid nitrogen. Small intestine was divided into two equal parts labelled as “jejunum” (containing also duodenum) and “ileum”. The large intestine containing also caecum and rec- tum was labelled as “colon”. As no sexual differences in M levels were found in the preliminary experiment, mice of both sexes were used.

RIA determination of M. Brain, stomach and the gut samples were first weighed and then extracted as

221

222 G. A. BUBENSK et al.

described earlier (Pang et al., 1982). Meiatonin levels were measured by a specific RIA as previously de- tailed (Brown et al., 1985). The intra-assay and inter-assay coefficient of variations of our RIA were 6.2% (N = 5) and 16.5% (N = 7) respectively (Pang and Yip, 1988). The concentration of melatonin is expressed in pmol per gram of tissue.

Binding assays. Radioligand binding assays were carried out with 2-~~2sIJiodomelatonin ([‘*‘I]MEL), specific activity 15~-18~ Ci/mmol, which was syn- thesized as previously detailed (Niles et al., 1987). Melatonin was obtained from Sigma and all other chemicals and reagents were purchased from com- mercial sources.

Gastrointestinal or whole brain tissues were hom- ogenized by a Brinkmann Polytron (setting 5, 10 set) in about 10 vol. ice-cold 0.32 M sucrose and centri- fuged at 1000 g for 10 min. The pellet was discarded and the supernatant centrifuged at 27,SOOg for 20 min to yield a crude membrane fraction. Mem- branes were washed three times by resuspension in 20 vol. of 50 mM Tris-HCl buffer (pH 7.4 at 4°C) and centrifugation at 35,OOOg for 10 min. Membrane fractions were then suspended in buffer for binding assays.

In single point assays, membranes were incubated with about 0.2 nM of [iz51]MEL at 0°C for 60 min. Saturation binding in brain and colonic membranes utilized radioligand concentrations of about 0.03-2.6 nM and incubation was carried out at 0°C or 30°C for 60 or 30min respectively. Bound radioac- tivity was separated by vacuum filtration on GF/B filters. Non-specific binding was determined in the presence of 10 ,uM melatonin.

~t~t~~t~c~~ analysis. The means and standards errors were computed and the differences between

day and night samples were calculated using the Student’s f-test. The accepted level of significance was P < 0.05.

RESULTS

Diurnal variation of melatonin

No diurnal difference was detected in stomach, ileum and colon tissues (Fig. 1). Significantly lower (P -C 0.05) daytime levels were found in the brain and the jejunum. Highest daytime levels were detected in the stomach (569 pg/g tissue), with the lowest in the brain (62 pg/g tissue). No differences in night time levels were detected between tissues. Highest variabil- ity between individual levels was observed in the stomach and then in the colon.

As shown in Table 1, specific [‘?]MEL binding was detected in both daytime and night time gastrointestinal tissues except for the jejunum which had no daytime binding under the present binding conditions. In all the tissues examined, binding was higher at night with this difference being particularly noticeable in the stomach where there was about a five-fold difference between day and night time levels.

Saturation binding experiments revealed the presence of high-affinity sites in the brain and colon as shown in Figs 1 and 2 respectively. In the brain, these sites had a dissociation constant (kd) of about 1.0 nM and a density (B,,) of 2.1 fmol/mg protein. The corresponding values in the colon were k, = 0.8 nM and B,,, = 6.2 fmol/mg protein. As shown in Fig. 2, a significantly higher number of binding sites was detected in the colon at an incubation tem~rature of 30°C: kd = 0.9, B,,, = 12.4 fmol/mg protein.

600

3 .g 400 Y X

< If

200

0 BRAJN STOMACH JEJUNUM ILEUM CCLCN

Tissues Fig. I. Meiatonin levels in the brain and the gastrointestinal tract of mice sacrificed at midday and

midnight. * = significantly higher M levels (P < 0.05).

Diurnal variation and binding characteristics of melatonin 223

Table I. Specific binding of [‘2’l]melatonin in mouse gastrointestinal tissues

Region

Specific Binding fmol/mg protein

Daytime Ninht time

Stomach 0.25 + 0.06 1.28 + 0.50 Jeiunum ND 0.98 + 0.06 Ilium 0.46 + 0.21 0.59 E 0.07 Colon 0.80 k 0.1 I 1.83 kO.15

Means f SE for 34 determinations are presented. ND = not detectable. Mem- branes were incubated with about 0.2 nM [?]MEL at 0°C for 60 min.

DISCUSSION

The presence of HIOMT in the intestinal tissues (Quay and Ma, 1976) supports the hypothesis of Raikhlin and Kvetnoy (1976) that tissues of intestinal mucosa, particularly EC cells are producing M. The detection of M in the gut of newborn rats (Bubenik, 1980) indicates that intestinal M is probably not of pineal origin. As the activity of HIOMT was not detected in the rat pineal before the 7th day of postnatal life (Illnerova, 1972) the M in the GIT of newborn rats may be of maternal origin (Reppert and Klein, 1978) or may be synthesized in the gut. Further support for the GIT synthesis of M is the detection of M in the gut of pinealectomized rats (Bubenik, 1980). Conversely, a remarkable increase of im- munodetectable M in the GIT tract after exogenous M administration (Bubenik, 1980; Bubenik and Pang-unpublished) indicates that the GIT tissues have the capacity to pick up M from the circulation. As the night time M levels in the pineal and the blood are about IO-fold higher than the daytime levels (Klein and Weller, 1970) it is conceivable that the significantly higher M levels in the jejunum of mice sacrificed during the scotophase is not due to the night time increase of M production in the EC cells, but rather due to an increase in M uptake from the circulation. However, the 20-fold higher levels of M in the jejunum of pigeons sacrificed during the night

16 I I I I COLON

.

0 3 6 9 12 16

BOUND Fig. 2. Scatchard plots of [‘251]MEL binding in mouse colonic membranes. Assays were performed at 0°C (0) and 30°C (a). Means of triplicate determinations from one of

two separate experiments are shown.

2.0 I I I I I

BRAIN

\ 1 0.0 0.5 1.0 1.5 2.0 2.6 3.0

BOUND Fig. 3. Scatchard plot of [‘251]MEL binding in mouse brain membranes. Means of triplicate determinations from one of two separate experiments are presented. Binding was

performed at 0°C.

(Vakkuri, et al., 1985) indicate that at least part of the nocturnal increase of M may be due to its synthesis in the GIT tract. Our finding of more than five times higher night-time levels of M in our brain samples (containing the pineal gland) (Fig. 1) are in agreement with the well known fact of much higher nocturnal levels of M in the pineal gland (Klein and Weller, 1970). The consistently high M levels and the large individual variability which was detected in the stomach in this study as well as in the mice deprived of food (Bubenik et al., 1992) may be related to the individual variability of food intake.

In preliminary experiments, we have detected specific binding sites for [‘251]MEL in various gastro- intestinal tissues and in whole brain. The nanomolar affinity of these sites is similar to that found in hamster brain and splenic tissues (Pickering and Niles, 1990; Niles, 1989). While the pharmacological characteristics of these sites are not yet known, their presence in the gastrointestinal tract which does not secrete melatonin into the circulation, suggests a paracrine role for locally synthesized hormone.

As noted above, the levels of melatonin in the jejunum were significantly higher at night. Moreover [iZSI]MEL binding in this tissue was also higher at night and was, in fact, not detected in daytime samples. These findings are not consistent with the well-known down-regulation of binding sites associ- ated with elevated levels of an agonist (Sibley and Lefkowitz, 1985). However, it should be borne in mind that the single-point binding examined in the jejunum is insufficient to accurately assess diurnal differences. Clearly, saturation binding studies utilizing several points in the circadian cycle are necessary in order to clarify the relationship between diurnal melatonin concentrations and its binding sites in the gastrointestinal tract.

Recently we have found an increase of M in all GIT tissues of mice exposed to food deprivation and sacrificed at midday (Bubenik et al., 1992). In

224 G. A. BUBEN~K et al.

addition, M treatment prevented or alleviated of the rat by immunocytochemistry. J. histochem.

symptoms of experimentally-induced colitis, such Cytochem. 28, 255.

as the rectal bleeding and the damage to the Illnerova H. (1972) The effect of light on the development

mucosal epithelium (Pentney and Bubenik- of hydroxyindole-0-methyltransferase activity in the rat pineal gland. Acta ner. super. 14, 130.

unnublished observations). The discoverv of a diur- Klein D. C. and Weller J. L. (1970) Indole metabolism in 1 I d

nal variation of M as well as the detection of binding sites for this hormone in the GIT provides further support for a physiological role of M in digestive physiology.

REFERENCES

Res. 12, 313.

Brown G. M., Seggie J. and Grota L. J. (1985) Serum

Bubenik G. A. (1986) The effect of serotonin. N-Acetvlsero-

melatonin response to melatonin administration in the Syrian hamster. Neuroendocrinology 41, 31.

Bubenik G. A. (1980) Localization of melatonin in the digestive tract of the rat. Effect of maturation, diurnal variation, melatonin treatment of uinealectomy. Hormone

the pineal gland: a circadian rhythm in pineal serotonin N-acetyl transferase activity. Science 177, 532.

Niles L. P. (1989) High-affinity binding sites for melatonin in hamster spleen. Med. Sci. Res. 17, 179.

Niles L. P., Pickering D. S. and Sayer B. G. (1987) HPLC-purified 2-[‘251]iodome1atonin labels multiple sties in hamster brain. Biochem. Biophys. Res. Commun. 147, 949.

Pang S. F. and Allen A. E. (1986) Extra-pineal melatonin in the retina: its regulation and physiological function. Pineal Res. Reviews 4, 55.

J. Exp. Zool. 219, 217. Pane S. F. and Yin P. C. Y. 11988) Secretorv natterns of

Pang S. F., Tang P. L., Yu H. S. and Yip M. K. (1982) The levels of N-acetylserotonin and melatonin in the brain of male rat: Diurnal variations and effects of pinealectomy.

tonin melatonin in spontaneous contraction of isolated pkeal melatonin-in the rats. ‘Pharkacol. Bio%em. Behav. rat intestine. J. Pineal Res. 3, 41. 15, 179.

Bubenik G. A., Ball R. 0. and Pang S.F. (1990) The role Pickering D. S. and Niles L. P. (1990) Pharmacological of serotonin and melatonin in the digestive physiology. characterization of melatonin binding sites in Syrian

Bubenik G. A., Brown G. M. and Grbta L. J. (1977)

_. -.

Immunohistological localization of melatonin in the rat

Clin. Inv. Med. 13, A3.

digestive tract. Experientia 33, 662.

Bubenik G. A., Ball R. 0. and Pang S. F. (1992) The effect

Bubenik G. A. and Dhanvantari S. (1989) The influence of serotonin and melatonin on some parameters of gastroin- testinal activity. J. Pineal Res. 7, 33.

of food deprivation on brain and gastrointestinal tissue

Dubocovich M. L. (1988) Pharmacology and function of melatonin receptors. FASEB J. 2, 2765.

levels of tryptophan, serotonin, 5-hydroxyindoleacetic

Hoffman R. A., Johnson L. B. and Reiter R. J. (1985)

acid. and melatonin. J. Pineal. Res. 12. 7-16.

Harderian glands of golden hamsters: Temporal and sexual differences in immunoreactive melatonin. J. Pineal Res. 2, 161.

enterochromaffine cells. Acta histochem.’ 55, 19. Raikhlin N. T., Kvetnoy I. M. and Tolkachev V. N. (1975)

hamster hypothalamus. Eur. J. Pharmacol. 175, 71:

Melatonin may be synthesized in enterochromaffin cells. Nature 255, 344.

Quay W. B. and Ma Y. H. (1976) Demonstration of gastrointestinal hydroxyindole-O-methyltransferase. ICRS Med. Sci. 4, 563.

Raikhlin N. T. and Kvetnov I. M. (1976) Melatonin and

Reppert S. and Klein D. (1978) Transport of maternal ‘H-melatonin to suckling rats and fate of )H-melatonin in neonatal rats. Endocrinology 102, 582-588.

Sibley D. R. and Lefkowitz R. J. (1985) Molecular mechan- isms of receptor desensitization using the /I-adrenergic receptor-coupled adenylate cyclase system as a model. Nature 317, 124.

Vakkuri O., Rintamaki H. and Leppalnoto J. (1985) Pres- ence of immunoreactive melatonin in different tissues of the pigeon. Gen. Comp. Endocrinol. 58, 69-75.

Holloway W. R., Grota L. J. and Brown G. M. (1980) Determination of immunoreactive melatonin in the colon