Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Melatonin is an endogenous putative neuromodulator in the retina of various vertebrate species (Besharse and Dunis, 1983; Pang et al., 1991; Zawilska and Nowak, 1992; Faillace at al., 1996). As in the pineal gland (Klein, 1979), retinal melatonin content (Hamm and Menaker, 1980; Nowak et al., 1989) significantly changes during the 24-h cycle. In the hamster retina, melatonin levels were shown to be significantly higher during the night than during the day (Faillace et al., 1994). Because both the exposure to darkness in vitro and pinealectomy significantly increased retinal melatonin levels (Faillace et al., 1995), it seems likely that in the hamster, retinal melatonin is generated within the tissue itself, and it is mainly regulated by the photic information. On the other hand, it has been demonstrated that melatonin biosynthesis in the hamster retina is regulated by a local circadian oscillator (Tosini and Menaker, 1996). Recently, we have shown that -aminobutyric acid, acting through a -aminobutyric acid_A receptor, significantly increased melatonin content in light-exposed retinas (Jaliffa et al., 1999).

A large body of evidence suggests that dopamine is a key regulator of retinal melatonin biosynthesis in several species (Iuvone et al., 1990; Nowak et al., 1990; Zawilska, 1994). Dopamine completely inhibits the increase in N-acetyltransferase (NAT) activity during the night in the retina of Xenopus, chicken, rat, and rabbit (Iuvone, 1986; Nowak et al., 1989; Iuvone et al., 1990; Nguyen-Legros et al., 1996). In addition, dopamine reproduces the light-induced inhibition on NAT activity and light increases dopaminergic activity in the retina of several species, leading to the assumptions that retinal dopamine is involved in the action of light on the retinal melatonin biosynthesis (Iuvone et al., 1978; Godley and Wurtman, 1988; Boatright et al., 1989; Zawilska and Nowak, 1994).

Hitherto, the hypothesis of a regulatory role of dopamine on retinal melatonin has not been examined in the Golden hamster, despite the fact that this species is considered one of the best experimental models for the study of melatonin physiology. Therefore, we considered it worthwhile to assess the effect of dopamine on hamster retinal melatonin content. A characterization of dopamine receptor involved in its effect on melatonin content was also performed.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Reagents and Drugs. Dopamine, clozapine, 3,4-dihydroxyphenylacetic acid (DOPAC), 3,4-dihydroxybenzylamine (DHBA), melatonin, 3-isobutyl-1-methylxanthine (IBMX), cAMP, and TC 199 medium were purchased from Sigma Chemical Co. (St. Louis, MO). [³H]Melatonin was obtained from New England Nuclear Co. (Boston, MA). Quinpirole, SCH 23390, and spiperone were obtained from Research Biochemicals Inc. (Natick, MA). The rabbit anti-cAMP antiserum was kindly supplied by the National Institute of Diabetes and Digestive and Kidney Diseases.

Animals and Tissues. Male Golden hamsters (average weight, 120 ± 20 g), derived from a stock supplied by Charles River Breeding Laboratories (Wilmington, MA), were maintained, with standard food and water available ad libitum, under a 14-h light/10-h dark lighting schedule (lights on at 6:00 AM) for a minimum of 10 to 14 days before use. The animals were sacrificed by decapitation at 12:00 noon or midnight. The eyes were enucleated, vitreous was removed, and the retinas were dissected out and incubated as described below. In some experiments, the hamsters were kept under constant darkness for 48 h and sacrificed at subjective midday or midnight. In the case of dark-exposed hamsters, sacrifice and incubations were performed under dim red light.

Melatonin Assessment. The retinas were incubated at 37°C for 8 h in 500 µl of TC 199 medium, (adjusted to pH 7.4 with NaHCO₃ and bubbled with 5% CO₂, 95% O₂) with or without dopamine and dopaminergic agonists and/or antagonists. Dopamine or quinpirole was added 1 min after the antagonists. After removal of the medium, the retinas were homogenized, and melatonin was extracted with 5 ml of dichloromethane. The amount of melatonin in each sample was determined by radioimmunoassay (RIA), as previously described (Faillace et al., 1994). Briefly, suitable aliquots of the organic phase dried under vacuum were resuspended in 100 µl of buffer and mixed with [³H]melatonin (20,000-24,000 dpm; specific activity, 38.8 Ci/mmol) and 50 µl of a rabbit antiserum developed in our laboratory. After incubation for 2 h at 37°C, 250 µl of saturated (NH₄)₂SO₄ was added, and the samples were centrifuged at 5000g for 20 min at 4°C. The radioactivity was assessed in the pellet resuspended in water.

cAMP Measurement. The retinas were incubated for 30 min at 37°C in a buffer containing 140 mM NaCl, 5 mM KCl, 2.5 mM CaCl₂, 1 mM MgCl₂, 10 mM HEPES, 10 mM glucose, and 0.5 mM IBMX, adjusted to pH 7.4 with Tris base with or without dopaminergic agonists and/or antagonists. After removal of the medium, the retinas were homogenized in 1 ml of 0.5 mM IBMX and boiled for 2 min. The retinal content of cAMP was assessed as previously described (Faillace et al., 1994). Briefly, the homogenates were centrifuged at 5000g for 5 min at 4°C. cAMP content was measured in the supernatants by RIA after acetylation. For this purpose, aliquots of samples or standards were acetylated with acetic anhydride/triethylamine. The acetylated products were incubated with ¹²⁵I-cAMP (15,000-20,000 dpm; specific activity, 140 mCi/mmol) and a rabbit antiserum and incubated overnight at 4°C. After the addition of 2 ml of ethanol with 2% BSA, the antigen-antibody complexes were precipitated by centrifugation at 2000g for 30 min. The supernatants were separated by aspiration.

Assessment of Dopamine Turnover Rate. Steady-state levels of dopamine and DOPAC were analyzed by HPLC with electrochemical detection. Retinas were homogenized in 100 µl of ice-cold 0.1 M perchloric acid, 10 µM ascorbic acid, 0.1 mM EDTA, and 20 ng/ml DHBA (as internal standard) and centrifuged at 5000g for 5 min at 4°C. Aliquots (20-µl) of supernatant fraction was injected into a Beckman Ultrasphere-ODS reverse phase column (5-µm particle size, 25 × 0.46 cm; Beckman Instruments, San Ramon, CA). Dopamine and DOPAC were eluted at a flow rate of 1.5 ml/min with a mobile phase consisting of 100 mM phosphoric acid, 0.1 mM EDTA, 0.45 mM sodium octylsulfate, and 6% acetonitrile, adjusted to pH 2.6 to 2.7 with NaOH. The eluted products were quantified by amperometric detection at a glassy carbon, thin-layer electrode with an applied potential of 0.56 V versus an Ag/AgCl reference electrode. Dopamine and DOPAC were identified by relative retention times compared with those of extracted standards. Concentrations were determined by comparing peak areas of unknown samples with those of standards using a programmed integrator interfaced with the detector unit. Values were corrected for the recovery of the external standard. Dopaminergic turnover rate was estimated as the DOPAC/dopamine ratio. Protein content was determined by the method of Lowry et al. (1951).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

The effect of dopamine and dopaminergic antagonists on melatonin content in retinas excised at 12:00 noon is shown in Fig. 1. At the highest concentration tested (100 µM), dopamine significantly decreased melatonin levels (Fig. 1A). This effect was inhibited by spiperone and clozapine, selective D₂, and D₄/D₂ dopaminergic receptor antagonists, respectively, but not by SCH 23390 (a D₁ receptor antagonist), as shown in Fig. 1B. In addition, clozapine and spiperone per se (i.e., in the absence of added dopamine) significantly increased retinal melatonin content, whereas SCH 23390 was ineffective (Fig. 2). Quinpirole, a D₂ dopaminergic agonist, significantly decreased retinal melatonin levels, as shown in Fig. 3.

View larger version (35K): [in this window] [in a new window]   Fig. 1.   Effect of dopamine (DA) in the presence or absence of dopaminergic antagonists on melatonin content in hamster retinas excised at 12:00 noon and incubated under light. After sacrifice, the retinas were excised and incubated as described in Materials and Methods. A, dopamine significantly decreased melatonin levels only at the highest concentration tested (100 µM). B, effect of 100 µM dopamine on melatonin levels was significantly reduced in the presence of spiperone (10 µM) and clozapine (1 µM) but not of 10 µM SCH 23390. Data are mean ± S.E. values (n = 15 animals per group). **P < .01 by Dunnett's test.

View larger version (38K): [in this window] [in a new window]   Fig. 2.   Effect of dopaminergic antagonists per se on retinal melatonin content in hamster retinas excised at 12:00 noon and incubated under light. Both clozapine (1 µM) and spiperone (10 µM) significantly increased melatonin levels, whereas SCH 23390 (10 µM) was ineffective. Data are mean ± S.E. values (n = 15 animals per group). **P < .01 by Dunnett's test.

View larger version (49K): [in this window] [in a new window]   Fig. 3.   Effect of quinpirole on melatonin content in Golden hamster retina. Hamsters were sacrificed at 12:00 noon, and the retinas were excised and incubated under light for 8 h. Quinpirole significantly decreased melatonin content at all concentrations tested. Spiperone (10 µM) significantly inhibited the effect of quinpirole. Data indicate mean ± S.E. values (n = 15 animals per group). *P < .05, **P < .01 by Dunnett's test.

To further examine the action of dopamine on melatonin content, its effect on cAMP accumulation in the presence of IBMX was evaluated at 12:00 noon. Dopamine significantly increased cAMP accumulation with a threshold concentration of 10 µM, with its effect being reversed by SCH 23390 (Fig. 4A) but not by spiperone (data not shown). Quinpirole significantly decreased cAMP accumulation with a profile similar to that observed for the inhibition of melatonin content (Fig. 4B). The effect of quinpirole on melatonin content and cAMP accumulation was reversed by spiperone. Dopamine turnover rate was assessed at midday and midnight. DOPAC content was significantly higher at 12:00 noon than at midnight, whereas dopamine levels did not change between these daytime points. The dopamine turnover rate (assessed as the ratio DOPAC/dopamine) was significantly higher at 12:00 noon than at midnight (Table 1).

View larger version (38K): [in this window] [in a new window]   Fig. 4.   Effect of dopamine and quinpirole on cAMP accumulation in retinas of Golden hamster. Incubations were performed in the presence of IBMX (0.5 mM) as described in Materials and Methods. A, dopamine significantly increased cAMP accumulation, with a threshold concentration of 10 µM, and its effect was reversed by SCH 23390 (10 µM). B, quinpirole significantly decreased cAMP accumulation at all concentrations tested. Spiperone (10 µM) reversed the effect of quinpirole. Data indicate mean ± S.E. values (n = 15 animals per group). *P < .05, **P < .01 by Dunnett's test.

Table 2 shows, comparatively, the effects of dopamine and quinpirole on cAMP accumulation and melatonin content in retinas excised at 12:00 noon and midnight. Dopamine significantly increased cAMP accumulation during both day and night, whereas quinpirole significantly decreased cAMP accumulation in retinas excised at 12:00 noon but not at midnight. Dopamine and quinpirole significantly decreased melatonin content only at midday but were ineffective at midnight. Control levels of both cAMP and melatonin were significantly higher at midnight than at midday. Figure 5 shows the effect of dopamine and quinpirole on cAMP accumulation in retinas of hamsters kept under constant darkness and sacrificed at subjective midday and midnight. Dopamine significantly increased cAMP accumulation at both times, whereas quinpirole decreased it at subjective midday and was ineffective at subjective midnight. Dopamine turnover rate was significantly higher at subjective midday than at subjective midnight (Table 3).

View larger version (30K): [in this window] [in a new window]   Fig. 5.   Effect of dopamine and quinpirole on cAMP accumulation in retinas of hamsters kept under constant darkness and sacrificed at subjective midday or midnight. A, at subjective midday, dopamine increased, whereas quinpirole decreased the accumulation of cAMP in the presence of 0.5 mM IBMX. B, at subjective midnight, dopamine significantly increased, whereas quinpirole did not affect this parameter. Data indicate mean ± S.E. values (n = 15 animals per group). **P < .01 by Dunnett's test.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

These results show that dopamine, dose dependently and presumably through a dopaminergic D₂-like receptor, decreases hamster retinal melatonin levels in vitro. Our results agree with the extensively described effect of dopamine on NAT activity in both mammal and nonmammal retinas (Hamm and Menaker, 1980; Iuvone, 1986; Nowak et al., 1989; Zawilska, 1994; Nguyen-Legros et al., 1996). However, although the inhibitory effect of dopamine in those species was systematically evident only in dark-adapted retinas in vitro or during the night in vivo, in the Golden hamster, it decreased retinal melatonin content during the day but not during the night. There is no ready explanation for this discrepancy. It must be taken into account that most of the studies on the inhibitory effect of dopamine have focused primarily on NAT activity, whereas in the present work, its effect was assessed on melatonin itself. Although NAT activity is considered the key enzyme in the biosynthesis pathway of melatonin, the existence of other dopamine-triggered factor or factors that regulate melatonin synthesis, with a different time dependence and under the control of a cAMP-dependent mechanism (as discussed later), cannot be ruled out. Notwithstanding, because at night dopamine decreased melatonin synthesis and release in the Xenopus retina (Cahill and Besharse, 1991), it appears that species differences, rather than measurement of NAT versus melatonin, accounts for this discrepancy. Hence, it is possible that some features of retinal melatonin regulation in the hamster may be unique.

Because dopamine exerts its action by stimulating specific receptors localized to membranes of target cells, experiments were conducted to elucidate the dopamine receptor subtype involved in the regulation of retinal melatonin content. Both spiperone (a D₂ antagonist) and clozapine (a D₄/D₂ antagonist) blocked the effect of dopamine on melatonin levels, whereas SCH 23390 (a D₁ antagonist) was ineffective. In addition, the effect of dopamine was mimicked by quinpirole (a D₂ agonist). Furthermore, both clozapine and spiperone, but not SCH 23390, per se significantly increased retinal melatonin content, supporting the fact that melatonin content is regulated by endogenous dopamine in the hamster retina. In agreement with results obtained in other species, the effect of dopamine on retinal melatonin levels in the hamster seems to be mediated by a D₂-like dopaminergic receptor. cAMP is the classic second messenger involved in the induction of melatonin biosynthesis in both the pineal and the retina. Because dopamine increased and quinpirole decreased retinal cAMP accumulation, it seems likely that dopamine can either stimulate or inhibit the retinal cAMP generating system by acting on two pharmacologically different receptors. When both D₁ and D₂ receptors are activated by dopamine, a net increase in cAMP was observed, indicating the prevalence of D₁ activation over D₂ inhibition on adenylate cyclase in the hamster retina. Accordingly, dopamine increases cAMP levels in both chicken and rabbit retina (Dubocovich and Weiner, 1985; Zawilska et al., 1995), whereas activation of D₂ dopaminergic receptors decreases cAMP accumulation in hen and frog retina (Iuvone, 1986; Nowak et al., 1990). Quinpirole, on the other hand, had no effect on basal or forskolin-stimulated adenylate cyclase activity in the chicken (Zawilska et al., 1995). The coexistence of D₁ and D₂ receptors in retinas of several species is well documented, and it probably occurs in the Golden hamster as well. D₁ receptors were localized predominantly to horizontal and cone bipolar cells (Veruki and Wassle, 1996), whereas D₂-like receptors were found in photoreceptor inner segments, the outer nuclear layer, the outer plexiform layer, and the ganglion cell layer (Rohrer and Stell, 1995). Although the effect of dopamine may take place in other cellular types, a direct effect on photoreceptors is most likely, because these cells are the most abundant and they are a primary site for the biosynthesis of melatonin (Iuvone et al., 1991; Wiechmann, 1996). In this sense, the fact that dopamine increased cAMP levels while it decreased melatonin content suggests that a diminution of cAMP levels within photoreceptors may be masked by the increase induced in other cellular types.

There is evidence that dopaminergic activity in the retina of several species varies with changes in environmental light conditions (Iuvone et al., 1978; Zawilska and Nowak, 1994). Accordingly, dopaminergic turnover rate in the Golden hamster retina was significantly higher at midday than at midnight. As already discussed, dopamine and quinpirole decreased melatonin content at midday but not at midnight. Therefore, it is possible that the D₂-like dopaminergic response in hamster retina also changes along the 24-h cycle. On the other hand, although the dopamine-induced increase in cAMP accumulation was evident at midday as well as at midnight, the relative magnitude of this stimulatory effect was also higher at 12:00 noon than at midnight, suggesting that the D₁ dopaminergic effect also changes through the day-night cycle. How those dopaminergic receptors are modulated remains unknown. Taking into account the day-night variations in dopaminergic turnover rate, it can be hypothesized that dopaminergic receptors are up- and down-regulated by their own ligand. In this sense, it has been shown that prolonged exposure to light or darkness modifies retinal dopaminergic specific binding and reactivity, presumably through a down-regulation mechanism (Dubocovich et al., 1985; Zawilska et al., 1997). However, because exogenous dopamine was only effective at a daytime point at which dopaminergic turnover was higher and given the existence of a circadian oscillator that regulates melatonin biosynthesis, the occurrence of two independent circadian clock-controlled rhythms for dopaminergic turnover and dopaminergic response cannot be ruled out. In fact, the subjective midday-midnight differences in dopaminergic turnover rate and in dopaminergic responsiveness on cAMP accumulation persisted in hamsters kept under constant darkness, supporting their control by a circadian mechanism.

Several lines of evidence in other species demonstrate that dopamine is part of the pathway for light input to the retina. The present results (e.g., day-night variations in dopamine turnover rate and the highest sensitivity to dopamine during the day) additionally suggest that in the Golden hamster, dopamine is involved in the light message, particularly on the melatonin system.