Retinal melatonin is metabolized within the eye of Xenopus laevis

Transcription

1 Proc. Nati. Acad. Sci. USA Vol. 86, pp , February 1989 Neurobiology Retinal melatonin is metabolized within the eye of enopus laevis (eyecup/aryl acylamidase/monoamine oxidase/methoxyindoles/circadian rhythm) GRGORY M. CAILL AND JOSP C. BSARS Department of Anatomy and Cell Biology, mory University School of Medicine, Atlanta, GA 3322 Communicated by John. Dowling, November 18, 1988 ABRSTRACT Retinal synthesis of melatonin, a potent modulator of rhythmic retinal processes, is elevated at night as a result of regulation by a circadian clock. Despite high nocturnal synthetic capacity, both melatonin content and release are low in the retina of the frog enopus laevis. We report here that cultured eyecups from enopus have the capacity for rapid metabolic breakdown of melatonin. Pharmacological analysis indicates that the initial step in this degradation pathway is deacetylation of melatonin by the enzyme aryl acylamidase (aryl-acylamide amidohydrolase, C ). This produces 5-methoxytryptamine, which is then deaminated by monoamine oxidase [amine:oxygen oxidoreductase (deaminating) (flavin-containing), C ], producing 5-methoxyindoleacetic acid and 5-methoxytryptophol. Inhibition of aryl acylamidase with eserine dramatically increases the release of endogenous melatonin by eyecups cultured at night, indicating that this pathway is the normal fate of retinal melatonin. Metabolism within the eye suggests a local neuromodulatory role for retinal melatonin, in contrast to the hormonal role of pineal melatonin. Melatonin (N-acetyl-5-methoxytryptamine) appears to play a major role in circadian regulation of retinal physiology (1). It mimics darkness in its effects on rod photoreceptor disc shedding (2), cone myoid length (3), movement of pigment granules in the pigment epithelium (4-6), and retinal dopamine release (7). Melatonin synthesis in the retina, as in the pineal gland, is elevated at night as a result of regulation by light and a circadian oscillator (1, 8-13). Melatonin may therefore be a circadian signal for night in the retina. Pineal melatonin content and release into the circulation appear to reflect directly its synthesis (8, 9). owever, evidence from several species suggests that other unidentified mechanism(s) may be involved in the regulation ofretinal melatonin (14, ). For example, in chicken retina, the activity of serotonin N-acetyltransferase, the penultimate enzyme in the synthesis of melatonin, and melatonin content are both high at night, but very little retinal melatonin enters the bloodstream (14). Despite high nocturnal serotonin N- acetyltransferase activity in the retina of enopus (12, 13), we found in preliminary experiments that retinal melatonin content and release of melatonin by cultured eyecups were both surprisingly low (<1 pg per retina and <1 pg/hr, respectively). One possible explanation for these results is that retinal melatonin is degraded within the eye. We have used a cultured enopus eyecup preparation to examine this possibility. MATRIALS AND MTODS yecup Culture. Postmetamorphic juvenile enopus laevis were maintained in a cycle of 12 hr light/12 hr dark for at least 4 weeks before use in experiments. yecups, including the The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C solely to indicate this fact. retina, pigment epithelium, choroid, and sclera, were prepared and cultured individually in 1 ml of defined culture medium as described (16). For experiments in which the metabolism of exogenous melatonin was measured, eyecups were prepared in the light just before the time of dark onset and cultured for 3 hr in the dark in medium that included either 8 gci of [methoxy-3]melatonin (DuPont/NN) per ml (1 Ci = 37 GBq) (1 nm) or 1 JLM melatonin (Sigma). For studies of endogenous melatonin release, eyecups were prepared and cultured for 4 hr at different times of day in the light or darkness. Dissections during the dark period were performed under infrared light with the aid of infrared viewers (FJW Industries, Mount Prospect, IL). ith.performance Liquid Chromatography (PLC). Tissue extracts were made by sonicating the combined retina, pigment epithelium, and choroid from each eyecup in 1.l of methanol. A sample ofthis suspension was reserved for determination of protein content by the method of Lowry et al. (17). After centrifugation, 1 pl of supernatant was added to 2 pul of.5 M ammonium acetate (p 4.25). Samples (1 pl) of culture medium and tissue extracts were analyzed by reverse-phase PLC. A Rainin Microsorb 5-gm C18 column ( x 4.6 mm) was used. The standard mobile phase was 33% methanol in.5 M ammonium acetate (p 4.25) with a flow rate of 1 ml/min. To identify unknown peaks, standard compounds were added to the samples and chromatographic conditions were altered by varying the methanol concentration in the range of 2-4%o, and by substitution of.1 M sodium acetate for ammonium acetate. Fluorescence was measured with a Laboratory Data Control Milton Roy (Riviera Beach, FL) Fluoromonitor III with a g lamp, 245-nm excitation filter, and 3- to 4-nm emission filter. For radiotracer experiments, fractions of.3 ml were collected and 3 activity was determined by liquid scintillation spectrometry. Counts per minute (cpm) are not corrected for counting efficiency, which was essentially constant. Melatonin RIA. The melatonin content of unextracted samples (5-2,p) of culture medium was determined by the RIA method of Rollag and Nisswender (18) as modified by Takahashi et al. (19). The 1251I-labeled melatonin analogue for the assay was obtained from azelton Biotechnologies (Vienna, VA). Melatonin concentrations >5 pg/ml were quantifiable in this assay. Validation of the RIA. Inhibition curves for various quantities of melatonin and pooled medium from nighttime eserine-treated eyecups were fitted to a 4-parameter logistic equation by a nonlinear least-squares curve fitting routine (2). The curves were parallel, with slopes of and , respectively. Melatonin (2.5, 5, 1, 25, 5, and 1 pg) added to samples of culture medium from eserine-treated day and night and control night eyecups was quantitatively recovered in the assay. The slopes of the recovery curves were 1.3, 1.7, and.92, respectively. RSULTS To determine whether ocular tissue has the capacity for metabolism of melatonin, we cultured enopus eyecups in 198

2 defined medium containing 1 nm [methoxy-3]melatonin. After a 3-hr incubation, samples of the culture medium and extracts of the combined retina, pigment epithelium, and choroid were separated by reverse-phase PLC. Analysis of fractions revealed the production of three radioactive metabolites from the added melatonin (Fig. 1 A-C). In this experiment, the amount of [3]melatonin in the medium was reduced 18%, from 687, cpm per 1Al in control medium to 56, + 47,5 cpm per 1 A.l (mean + SD; n = 4) during a 3-hr incubation. This reduction was entirely accounted for by the production of the three metabolites. The tissue concentration of [3]melatonin at the end of this experiment was.24 ±.84 pmol per mg of protein, and metabolites were produced at an average rate of 16.9 ± 4.9 pmol per hr per mg of protein (means ± SD). Thus, under these conditions, the entire tissue content of [3]melatonin was turned over in 1 min. The PLC retention times of melatonin and its metabolites are reduced slightly by the tritiated methoxy group, prevent- A. Tissue extract C4 5 _ C) B. Culture medium 61 C#, 4- ' 2- I- co C. CO, x I. co, Neurobiology: Cahill and Besharse V 5 1 Retention time, mi v Mel D. Culture medium. 1- nifi a; ) Co ) Standards Retention time, min FIG. 1. PLC analysis of melatonin metabolites produced by cultured enopus eyecups. (A and B) When eyecups are incubated in medium containing [methoxy-3]melatonin (Mel), three radiolabeled metabolites (arrowheads) are found in 1-1,u samples of tissue extract (A) and culture medium (B). (C) Control medium incubated without eyecups. (D) When eyecups are incubated with 1,uM unlabeled melatonin, metabolites (arrowheads) are detectable in the culture medium by fluorescence. () Chromatogram of synthetic indole standards (1 ng). The metabolites cochromatograph with synthetic 5-methoxytryptamine (peak 4), 5-methoxyindoleacetic acid (peak 5), and 5-methoxytryptophol (peak 6). Other standard peaks shown are serotonin (peak 1), N-acetylserotonin (peak 2), 6-hydroxymelatonin (peak 3), and melatonin (peak 7). Proc. Natl. Acad. Sci. USA 86 (1989) 199 ing direct identification of the radiolabeled metabolites by comparison with unlabeled standards. Therefore, eyecups were incubated in medium containing 1 pum unlabeled melatonin, and the culture medium was analyzed by PLC with fluorescence detection. Three fluorescent compounds, similar to the radiolabeled metabolites in relative amounts and chromatographic behavior, were released by the eyecups into the culture medium (Fig. 1D). These compounds, with retention times of 8.2, 1.9, and 12.1 min under standard chromatographic conditions, were identified as 5-methoxytryptamine, 5-methoxyindoleacetic acid, and 5-methoxytryptophol, respectively. Under several different chromatographic conditions, these compounds were indistinguishable from standards. When eyecups were incubated without added melatonin, 5-methoxyindoleacetic acid was detectable at low levels by PLC with fluorescence detection in culture medium; endogenous melatonin and the other metabolites were detectable only after extraction and concentration (data not shown). A pathway that produces these metabolites from melatonin has previously been described as a minor route of melatonin metabolism in the liver (21). This pathway (Fig. 2) involves deacetylation of melatonin by aryl acylamidase (arylacylamide amidohydrolase, C ) to produce 5- methoxytryptamine. Deamination of 5-methoxytryptamine by monoamine oxidase [amine:oxygen oxidoreductase (deaminating) (flavin-containing), C ] produces 5- methoxyindoleacetaldehyde, which is further oxidized to 5-methoxyindoleacetic acid or reduced to 5-methoxytryptophol. We used inhibitors of monoamine oxidase and aryl acylamidase to test the hypothesis that this is the pathway for melatonin metabolism in the eye. When pargyline, a monoamine oxidase inhibitor, was added to the culture medium at a concentration of 1,uM, the production of 5-methoxyindoleacetic acid and 5-methoxytryptophol from [3]melatonin was blocked, and 5-methoxytryptamine accumulated in the tissue (Fig. 3A). This indicates that the initial step in the breakdown of melatonin is deacetylation to 5-methoxytryptamine, which is then deaminated by monoamine oxidase. Alcohol dehydrogenase C3 t3. C3. C2C2NCOC3 Melatonin Aryl acylamidase I C3 > C2C2N2 5-Methoxytryptamine Monoamine oxidase 4 5-Methoxytryptophol C3 -. C2CO 5-Methoxyindoleacetaldehyde C2C2 Aldehyde dehydrogenase C3 N 3 C2COO N 5-Methoxyindoleacetic acid FIG. 2. Proposed enzymatic pathway for ocular metabolism of melatonin. Aryl acylamidase and monoamine oxidase are inhibited by eserine and pargyline, respectively (21, 22).

3 11 Neurobiology: Cahill and Besharse Proc. NatL Acad. Sci. USA 86 (1989) 1.5 r m Control M Pargyline ** co x 1._ 1 'a 4". A.. l..5 a If F * t 5MT 5MIAA 5MToI Total metabolites ['1A. :ND 5MT 5MIAA 5MTol Mel issue content 5V C. 4 I = Control M serine * 1 co 3 I Irl x._=, Ca 2-3 cl. om i - ** 75 c.6, co 5 25 B ^ T 5MT 5MIAA 5MToI Total metabolites 5MT 5MIAA 5MToI Mel Tissue content FIG. 3. Pharmacological analysis of melatonin metabolism in cultured eyecups. (A) ffects of 1,LM pargyline on metabolism of [3]melatonin. (Left) Total metabolites produced. (Right) Tissue content at the end of a 3-hr incubation. The production of 5- methoxyindoleacetic acid (5MIAA) and 5-methoxytryptophol (5MTol) was blocked by pargyline, and there was an increased accumulation of 5-methoxytryptamine (5MT) in the tissue. (B) ffects of 1,um eserine salicylate on metabolism of [3]melatonin. serine inhibited the production of all melatonin metabolites and resulted in accumulation of melatonin (Mel) in the tissue. Values are means ± SM from four individually cultured eyecups per group; asterisks indicate significant differences from controls as follows: *, P <.5; **, P <.1; one-tailed t test. ND, not detectable. Aryl acylamidase, a deacetylase that appears to be related to acetylcholinesterase, can be inhibited by eserine (22). serine (1 AM) inhibited the production of all metabolites from [3]melatonin and caused increased accumulation of melatonin in the tissue (Fig. 3B), consistent with a primary role for aryl acylamidase in the ocular metabolism of melatonin. Lower concentrations of eserine (1 and 1,uM) were ineffective. To determine whether this metabolism is the normal fate of melatonin synthesized in the retina, we used eserine to block aryl acylamidase in eyecups cultured at different times of day and night in the absence ofadded melatonin. The endogenous melatonin released into the medium by individual eyecups was measured by radioimmunoassay (Fig. 4). At night, when serotonin N-acetyltransferase activity is high, eyecups U treated with eserine released -7 times as much melatonin as controls. In all daytime groups, melatonin levels were near or below the level of accurate quantification by the RIA. Because eserine is expected to increase retinal acetylcholine levels by inhibiting acetylcholinesterase, other eyecups were cultured with carbachol, an agonist of acetylcholine receptors. Carbachol had no effect, indicating that the increased melatonin release was not the result of increased retinal acetylcholine levels. DISCUSSION We have shown that cultured enopus eyecups have the capacity for rapid metabolism of melatonin. The results of pharmacological experiments support the hypothesis that the

4 Neurobiology: Cahill and Besharse,>, ) Io-Control l Carbachol T 4' Tine of day, hr FIG. 4. ffects of eserine on the release of melatonin by cultured eyecups at different times of day. yecups were incubated in the light during the day (open bars) and in the dark at night (hatched bars). Times and 12 are the times of lights on and off, respectively. The culture medium contained no drug (control; Top), 1,uM eserine salicylate (Middle), or 1,uM carbachol (Bottom). Values are means ± SM (n = 5) of total radioimmunoassayable melatonin in 1 ml of culture medium after 4-hr incubations. In the presence of eserine, nighttime melatonin release by eyecups is increased with respect to control nighttime samples and to all daytime samples (analysis of variance, Scheffe's test: P <.1). In untreated controls, significant melatonin release is detectable only at night. The acetylcholine agonist carbachol has no effect on melatonin release when compared to untreated controls. Proc. Natl. Acad. Sci. USA 86 (1989) 111 first step in this breakdown pathway is deacetylation by aryl acylamidase. Several forms of aryl acylamidase have been described that differ in tissue localization, substrate specificity, and sensitivity to inhibitors (22, 23). The concentration of eserine required to significantly inhibit the metabolism of melatonin (1,uM) was relatively high. This may indicate that the ocular aryl acylamidase is one of the forms less sensitive to eserine or that access of the drug to the enzyme was limited in the eyecup. When eyecups were cultured with eserine at a concentration sufficient to inhibit the deacetylation of melatonin, the release of melatonin into culture medium was increased dramatically, but eserine had no effect during the day. The resulting diurnal variation in melatonin release is similar to that reported for the melatonin synthetic enzyme serotonin N-acetyltransferase in enopus retina (12, 13). These results suggest that retinal melatonin production is elevated at night, but release of melatonin from the eyecup is prevented by local metabolism. A similar mechanism may provide an explanation for the relatively small contribution of retinal melatonin to circulating levels in avian species (14, ) and the low melatonin content in the retina of mammals (24, 25). Ocular metabolism suggests a local role for retinal melatonin. This pathway may be a mechanism for regulation of local concentrations of melatonin, which is a potent modulator of cellular processes in photoreceptors (2, 3), the retinal pigment epithelium (4-6), and dopaminergic amacrine cells (7). An alternative function for this pathway may be in the synthesis of other biologically active methoxyindoles. 5- Methoxytryptophol has effects on disc shedding similar to those of melatonin (2), and 5-methoxytryptamine has been shown to affect serotonergic neurotransmission in the central nervous system (26). Synthesis of these compounds in the retina has been demonstrated previously, but it was assumed that they were made by methylation of serotonin (5- hydroxytryptamine) and deaminated serotonin metabolites through the action of hydroxyindole-o-methyltransferase (27). Our results show that they can also be synthesized through the degradation of melatonin. Previous attempts to demonstrate deacetylation of melatonin by aryl acylamidase in the brain and pineal gland have been unsuccessful (21, 23). In fact, no endogenous substrate has been found for the brain and pineal forms of this enzyme. The majority of circulating melatonin is cleared by conversion to 6-hydroxymelatonin in the liver; only 1-2% is deacetylated by liver aryl acylamidase (21, 28, 29). Thus, the eye may be unique in its use of this pathway as a major route of melatonin metabolism. We thank G. Nisswender for supplying the melatonin antibody. This work was supported by National ye Institute Grant Y2414 to J.C.B. and Fellowship Y5964 to G.M.C. 1. Besharse, J. C., Iuvone, P. M. & Pierce, M.. (1988) Prog. Retinal Res. 7, Besharse, J. C. & Dunis, D. A. (1983) Science 219, Pierce, M.. & Besharse, J. C. (1985) J. Gen. Physiol. 86, Kraus-Rupert, R. & Lembeck, F. (1965) Pflugers Arch. ur. J. Physiol. 284, Cheze, G. & Ali, M. A. (1976) Can. J. Zool. 54, Pang, S. F. & Yew, D. I. (1979) xperientia 35, Dubocovich, M. L. (1983) Nature (London) 24, Reppert, S. M. & Klein, D. C. (198) in The ndocrine Functions ofthe Brain, ed. Motta, M. (Raven, New York), pp Klein, D. C., Auerbach, D. A., Namboodiri, A. A. & Wheler, G.. T. (1981) in The Pineal Gland VI. Anatomy and Biochemistry, ed. Reiter, R. (CRC, Boca Raton, FL), pp Binkley, S., ryshchyshyn, M. & Reilly, K. (1979) Nature (London) 281, amm,.. & Menaker, M. (198) Proc. Natl. Acad. Sci. USA 77, Besharse, J. C. & Iuvone, P. M. (1983) Nature (London) 35, Iuvone, P. M. & Besharse, J. C. (1983) Brain Res. 273, Reppert, S. M. & Sagar, S. M. (1983) Invest. Ophthalmol. Visual Sci. 24, Underwood,., Binkley, S., Siopes, T. & Mosher, K. (1984) Gen. Comp. ndocrinol. 56, Besharse, J. C. & Dunis, D. A. (1983) xp. ye Res. 36, Lowry,.., Rosebrough, N. J., Farr, A. L. & Randall, R. J. (1951) J. Biol. Chem. 193, Rollag, M. D. & Nisswender, G. D. (1976) ndocrinology 98, Takahashi, J. S., amm,.. & Menaker, M. (198) Proc. Natl. Acad. Sci. USA 77, DeLean, A., Munson, P. J. & Rodbard, D. (1978) Am. J. Physiol. 235, Rogawski, M. A., Roth, R.. & Aghajanian, G. K. (1979) J. Neurochem. 32, Oommen, A. & Balassubramanian, A. S. (1978) Biochem. Pharmacol. 27, su, L. L. (1982) Int. J. Biochem. 14, Reiter, R. J., Richardson, B. A., Matthews, S. A., Lane, S. J. & Ferguson, B. N. (1983) Life Sci. 32, Dubocovich, M. L., Lucas, R. C. & Takahashi, J. S. (1985) Brain Res. 335,

5 112 Neurobiology: Cahill and Besharse 26. DeMontigny, C. & Aghajanian, G. K. (1977) Neuropharmacology 16, Balemans, M. G. M., Pevet, P., van Benthem, J., aldar-misra, C., Smith, I. & endriks,. (1983) J. Neural Trans. 56, Proc. Natl. Acad. Sci. USA 86 (1989) 28. Kopin, I. J., Pare, C. M. B., Axelrod, J. & Weissbach,. (1961) J. Biol. Chem. 236, Kveder, S. & McIsaac, W. M. (1961) J. Biol. Chem. 236,