Melatonin synthesis in retina: circadian regulation of arylalkylamine N-acetyltransferase activity in cultured photoreceptor cells of embryonic chicken retina

Brain Research 973 (2003) 56–63 www.elsevier.com / locate / brainres

Research report

Melatonin synthesis in retina: circadian regulation of arylalkylamine N-acetyltransferase activity in cultured photoreceptor cells of embryonic chicken retina Tamara N. Ivanova a , P. Michael Iuvone a,b , * a

Department of Pharmacology, Emory University School of Medicine, 1510 Clifton Road, room 5107, Atlanta, GA 30322, USA b Department of Ophthalmology, Emory University School of Medicine, Atlanta, GA, USA Accepted 24 February 2003

Abstract The key regulatory enzyme in melatonin synthesis is arylalkylamine N-acetyltransferase (AANAT). In vivo, AANAT activity in chicken retinal photoreceptor cells exhibits a circadian rhythm that peaks at night. The purpose of the present study was to investigate the temporal development of light / dark and circadian oscillations of AANAT activity in cultured retinal cells prepared from 6- and 8-day-old chicken embryos (E6, E8, respectively). Photoreceptor cells prepared from E6 retinas and incubated under a 14-h light / 10-h dark (LD) cycle of illumination for 5–7 days displayed prominent daily fluctuations in AANAT activity on days 5 and 6 in vitro. However, when E6 cells, incubated for 5 days under LD, were transferred to continuous (24 h / day) darkness (DD) on day 6, no daily pattern of activity was observed. This result indicates that AANAT fluctuations were light-driven and not circadian at this stage. In contrast, cells prepared from E8 embryos and incubated under conditions identical to those for E6 cells displayed prominent rhythms of AANAT activity in both LD and DD, indicative of circadian control. To determine if circadian control of AANAT activity would develop in E6 cells incubated for a longer period of time to allow maturation, cells were incubated for 8 days in LD followed by 2 days in DD. AANAT activity in these cells was rhythmic in both LD and DD. In cells incubated in this manner, a 2-h light pulse in the middle of the subjective night suppressed AANAT activity, indicating that the enzyme activity in the cultured cells is acutely suppressed by light, as it is in vivo. These results indicate that the ability to express circadian regulation of AANAT activity is an intrinsic property of retinal cells that can develop in vitro. Development of light–dark regulation of AANAT activity appears to precede the circadian clock-control of enzyme activity.  2003 Elsevier Science B.V. All rights reserved. Theme: Sensory systems Topic: Retina and photoreceptors Keywords: Arylalkylamine N-acetyltransferase; Photoreceptors cell; Circadian rhythm; Melatonin; Retina

1. Introduction Circadian rhythms play a central role in the regulation of a variety of important aspects of photoreceptor metabolism and, particularly, in the biosynthesis of melatonin [5], which, in turn, influences the organization of visual processes. Serotonin N-acetyltransferase (arylalkylamine N-acetyltransferase, AANAT, EC 2.3.1.87) has been iden*Corresponding author. Department of Pharmacology, Emory University School of Medicine, 1510 Clifton Road, Room 5107, Atlanta, GA 30322, USA. Tel.: 11-404-727-5989; fax: 11-404-727-0365. E-mail address: [email protected] (P. Michael Iuvone).

tified as a key regulatory enzyme in the melatonin biosynthetic pathway [16]. In chickens, the enzyme is expressed primarily in the pineal gland and in the retina [4,5,13]. Retinal AANAT mRNA and, consequently, AANAT activity are expressed primarily in photoreceptor cells [5]. The levels of chicken retinal AANAT activity and melatonin biosynthesis exhibit circadian rhythms, peaking at night [9]. Additionally, AANAT activity in the chicken retina is also under control of light, which dramatically suppresses enzyme activity and promotes degradation of the enzyme protein [5,10,13]. The circadian rhythm of AANAT activity and melatonin biosynthesis may represent a direct output of a retinal circadian clock.

0006-8993 / 03 / $ – see front matter  2003 Elsevier Science B.V. All rights reserved. doi:10.1016 / S0006-8993(03)02540-X

T.N. Ivanova, P. Michael Iuvone / Brain Research 973 (2003) 56–63

However, Pierce [19] found that dispersed, cultured retinal cells from embryonic quail displayed light–dark differences in melatonin biosynthesis, but no circadian control, raising the possibility that clock control of melatonin synthesis may be indirect and not an intrinsic property of the photoreceptor cell. Alternatively, the circadian clock control of melatonin synthesis may develop later than the photic control mechanisms. The purpose of the present study was to investigate the temporal development of the diurnal and circadian oscillations of AANAT activity in cultured photoreceptor cells prepared from retinas of embryonic chickens. We found that photic control of AANAT activity precedes the circadian control and that circadian rhythmicity is indeed an intrinsic property of retinal cells.

57

mW/ cm 2 . The schedule (LD 14:10) was introduced on day 1 of incubation of eggs and continued through 5–10 DIV. Fig. 1 illustrates graphically the illumination and sampling schedules used in the experiments reported here. In some experiments illumination was switched from LD to constant (24 h / day) darkness (DD) (Fig. 1B–E) before expected onset of L at the start of DIV 6 or 9. In the experiment depicted in Fig. 1E, the light was switched on for 2 h during the subjective night of DIV 9. Illumination was switched from LD to constant light (LL) before expected onset of D on DIV 8 in the experiment illustrated in Fig. 1F; the light was switched off for 2 h before sampling one group of cells during the subjective night of DIV 9 (Fig. 1F). The cells were sampled on DIV 5–7 (Fig. 1A–C) or DIV 8–10 (Fig. 1D–F).

2.3. Assay of AANAT activity 2. Materials and methods

2.1. Cell preparation and culture Monolayer cultures of retinal cells were prepared from neural retinas of 6–8-day-old chicken embryos (E6, E8) by a modification of the method of Adler et al. [2]. Neural retinas were dispersed in 0.25% trypsin and cells (|3.43 10 6 cells) were plated on 35-mm Primaria culture plates (Becton Dickinson Labware, Franklin Lakes, NJ, USA) in medium 199 containing 20 mM HEPES, linoleic acid-BSA 110 mg / ml, 2 mM glutamine, penicillin G (100 U / ml) and 10% fetal bovine serum. Cells were maintained at 39.560.4 8C under a humidified atmosphere of 5% CO 2 in air. Days in vitro (DIV) are numbered successively from the day of dissection (DIV 0). On the first DIV, S-( pnitrobenzyl)-6-thioinosine (NBTI) was added in a final concentration of 5 mM. Medium was exchanged on DIV 4 with medium 199 supplemented with 1% fetal bovine serum, 1% horse serum, 5 mM NBTI, 5 nM insulin-like growth factor-1, 5 mM 9-cis-retinoic acid, HEPES, glutamine, linoleic acid-BSA, and penicillin at the concentrations described above. In the experiments depicted in Figs. 2 and 3, the retinoic acid was omitted from the culture medium in half of the groups in order to test the effect of the retinoid on AANAT activity. In the experiments shown in Figs. 5–7, the medium was replaced in DIV 4 and DIV 7.

2.2. Light cycles Fertilized eggs and cultured cells were exposed to a daily lighting regime of 14 h light (L) and 10 h dark (D), with light onset at Zeitgeber time (ZT) 0. Lighting was provided by an 8-W cool white fluorescent tube (General Electric Company, OH, USA); irradiance at the level of the egg racks and culture dishes was in the range of 30–50

Cells for AANAT measurements were collected in 150 ml of 0.25 M potassium phosphate (pH 6.5) containing 1.12 mM acetylcoenzyme A. Samples were frozen in dry ice and kept at 280 8C until analyzed. The AANAT enzyme activity was determined in cell homogenates by measuring the catalytic formation of N-acetyltryptamine from tryptamine and acetylcoenzyme A, as described by Thomas et al. [25]. The reaction product, Nacetyltryptamine, was quantified by HPLC with fluorescence detection. Enzyme activity was normalized to cellular protein, measured as described by Lowry et al. [18].

3. Results

3.1. Daily fluctuations of AANAT activity in embryonic chicken photoreceptor cells Retinal cells prepared from embryonic day 6 (E6) neural retinas and incubated under an LD 14:10 cycle, were sampled on DIV 5–7 at ZT 10 and at ZT 20 as shown in the diagram of experimental protocol (Fig. 1A). Fig. 2 illustrates typical daily fluctuations of AANAT activity in chicken photoreceptor cells. Enzyme activity was significantly higher at night than during the daytime on DIV 5 and 6 in cells incubated with and without 9-cis-retinoic acid (P,0.01). Inclusion of 9-cis-retinoic acid in the culture medium significantly increased AANAT activity on DIV 5 and 6 (P,0.05 at ZT 10; P,0.001 at ZT 20), but not on DIV 7. Night-time enzyme activity of retinoic acid-treated cells was lower on DIV 7 compared to DIV 5 (P50.016) and DIV 6 (P50.002). This apparent loss of responsiveness on the night of DIV 7 may reflect exhaustion of some media component, possibly retinoic acid, as a robust nocturnal increase of AANAT activity was observed on DIV 8 when the medium was replaced on DIV 7 (see Fig. 5).

58

T.N. Ivanova, P. Michael Iuvone / Brain Research 973 (2003) 56–63

Fig. 1. Schematic timelines of the experimental protocols. Diagrams A–F present graphically the duration of incubations in ovo and in vitro. Exposure to daily light / dark cycle (LD) in ovo is presented as alternating light gray and dark gray bars (L: 14 h, and D: 10 h). The embryonic day of retinal dissection and preparation of cell cultures was E6 (A, B, D, E, F) and E8 (C). LD cycles on days in vitro (DIV) are presented as white and black bars (L: 14 h, and D: 10 h). Continuous (24 h / day) darkness (DD) or light (LL) is indicated by the solid black or white bars, respectively. Exposure to light for 2 h during the subjective night on DIV 9 in DD, is illustrated by the thin white bars, while the 2 h of darkness during LL is illustrated by the thin black bar.

3.2. Development of circadian rhythmicity of AANAT activity in vitro Fig. 3 illustrates the activity of AANAT in E6 cells cultured for 5 days under LD followed by 2 days in constant darkness (see Fig. 1B for diagram of experimental protocol). Daily fluctuations of AANAT activity were evident in LD (DIV 5) with high values at night (ZT 10 vs. ZT 20, P,0.001), but in DD (DIV 6 and 7) the day / night rhythm of AANAT activity decayed completely. Inclusion of 9-cis-retinoic acid significantly increased AANAT activity on DIV 5 and 6 (P,0.001), but not on DIV 7. Retinoic acid had no demonstrable effect on the ability of the cells to sustain rhythmic expression of AANAT activity in DD. 9-cis-Retinoic acid was included in the medium for all subsequent experiments due to its ability to increase AANAT activity. To determine if circadian rhythmicity of AANAT activity was expressed in photoreceptor cells prepared

from older embryos, cells were prepared from E8 neural retinas and incubated as described above for E6 cells (see Fig. 1C for a diagram of experimental protocol). AANAT activity of E8 cultures (Fig. 4), which have a smaller percentage of total cells expressing the photoreceptor phenotype than the E6 cultures [1,14], was lower than that in E6 cells. However, in E8 cells, AANAT activity displayed prominent daily fluctuations in LD on DIV 5 (ZT 10 vs. ZT 20, P50.011) and in DD on DIV 6 (ZT 10 vs. ZT 20, P,0.001). Thus, circadian fluctuations of AANAT activity existed in E8 cells, but not in E6, with the same period of incubation in vitro. These observations suggest that the older developmental age of the E8 cultures (14–15 days of combined in ovo and in vitro development), compared with the E6 cultures, may promote the expression of circadian rhythmicity of AANAT activity or that some change occurs in the cells in ovo between E6 and E8 that is not replicated in vitro. To examine the former possibility, E6 photoreceptor

T.N. Ivanova, P. Michael Iuvone / Brain Research 973 (2003) 56–63

Fig. 2. Daily fluctuations of AANAT activity in E6 retinal cells cultured for 5–7 days. Photoreceptor cells, prepared from E6 retinas, were cultured under LD. On DIV 4, the initial culture medium was replaced with the low serum medium described in Materials and methods, with or without 5 mM 9-cis-retinoic acid (9cis RA). The cells were sampled on DIV 5–7 at ZT 10 and at ZT 20. White symbols represent AANAT activity at ZT 10 in light; black symbols represent ZT 20 in darkness. The horizontal white and black bars above the x-axis represent times of light and darkness, respectively. The numbers above the horizontal bars indicate the days in vitro (DIV). Data are presented as mean6S.E.M. N518–21 for all groups except on DIV 7, where n55–6. Activity of AANAT in these cells displayed daily fluctuations with elevations at night (ZT 10 vs. ZT 20 on DIV 5 and 6, P,0.01). A two-factor ANOVA indicated a significant effect of retinoic acid (F523.586, P,0.001); a significant effect of time (F517.775, P,0.001); and a significant interaction of retinoic acid and time (F52.692, P50.023).

cells were cultured for 8 days in LD and for the following 2 days in DD (See Fig. 1D for diagram of experimental protocol). Illumination was switched to constant darkness before the expected onset of light at the beginning of DIV 9, yielding a combined in ovo and in vitro age of 15–16 days at the time of sampling. Fig. 5 demonstrates that E6 photoreceptor cells, cultured in vitro for 8–10 days, exhibited clear circadian oscillations of AANAT activity under DD (ZT 10 vs. ZT 20 P,0.05 in all groups), similar to the rhythmicity recorded in E8 cells.

3.3. Influence of brief light exposure at night on the activity of AANAT in chicken photoreceptor cells In retinas of posthatch chickens, light exposure at night rapidly suppresses AANAT activity and melatonin levels [10]. To determine if E6 cells, cultured for 8–10 days as described above, develop a similar sensitivity to light exposure, cells were incubated under light for 2 h in the middle of the night on DIV 9 (see Fig. 1E for diagram of

59

Fig. 3. Absence of circadian fluctuations in AANAT activity in E6 retinal cells cultured for 5–7 days. Cells were prepared from E6 retinas and cultured under LD for 5 days. On DIV 4, the initial culture medium was replaced with the low serum medium described in Materials and methods, with or without 5 mM 9-cis-retinoic acid (9cis RA). Beginning on DIV 6, cells were incubated in constant darkness (DD) for 2 days. White symbols represent AANAT activity at ZT 10 in light; black symbols represent ZT 20 in darkness; gray symbols represent subjective day, ZT 10, in darkness. The horizontal white and black bars above the x-axis represent times of light and darkness, respectively. The numbers above the horizontal bars indicate the days in vitro (DIV). On DIV 5, AANAT activity at ZT 20 was significantly higher than at ZT 10 (P,0.001). In DD, rhythmic fluctuations of AANAT activity disappeared; AANAT activity at ZT 10 and at ZT 20 did not differ significantly for DIV 6 (P.0.05). A two-factor ANOVA indicated a significant effect of retinoic acid (F5185.07, P,0.001); a significant effect of time (F527.506, P,0.001); and a significant interaction of retinoic acid and time (F5 17.958, P,0.001). Data are presented as mean6S.E.M, n54–5 per group.

experimental protocol). Fig. 6 demonstrates the progressive increase of AANAT activity during DIV 9 under DD. Light exposure for 2 h at ZT 20 dramatically decreased AANAT activity to levels observed during the mid-day (ZT 22 in darkness vs. ZT 22 after 2 h light, P,0.001). Cells remaining in darkness, maintained high levels of AANAT activity, which continued to cycle during the next day (DIV 9, ZT 10 DD vs. ZT 20, P50.002; DIV 10, ZT 10 DD vs. ZT 20, P50.012).

3.4. Influence of continuous illumination on circadian oscillations of AANAT activity To determine if circadian rhythms of AANAT activity persist in constant light (24 h / day), E6 cells were incubated under LD for 8 days, followed by 2 days in LL (see Fig. 1F for a diagram of experimental protocol). Fig. 7

60

T.N. Ivanova, P. Michael Iuvone / Brain Research 973 (2003) 56–63

Fig. 4. Circadian fluctuations of AANAT activity in E8 retinal cells cultured for 5–7 days. Cells were prepared from E8 retinas and incubated for 5 days under LD. Illumination was switched from LD to DD before expected onset of light at the beginning of DIV 6. White symbols represent AANAT activity at ZT 10 in light; black symbols represent ZT 20 in darkness; gray symbols represent subjective day, ZT 10, in darkness. The horizontal white and black bars above the x-axis represent times of light and darkness, respectively. AANAT activity exhibited diurnal fluctuations on DIV 5 (P50.011) that persisted in DD on DIV 6 (P,0.001). Data are presented as mean6S.E.M, n54–6 per group.

demonstrates the suppressive effect of light on the activity of AANAT in embryonic photoreceptor cells cultured under constant illumination during DIV 9 and 10. Nevertheless, AANAT activity, depressed by constant light, still exhibited significant (P50.025) daily fluctuation during the first, but not the second, day in LL. Similar results were obtained with E8 cells incubated for 5 days in LD followed by 2 days in LL (data not shown). We also investigated the effect of 2 h of darkness on AANAT activity during the subjective night (ZT 20–22) of DIV 9 in cells incubated in LL (Fig. 7). Darkness had no significant effect on AANAT activity under these conditions (P.0.05). Thus, the suppressive effect of light is not readily reversible.

4. Discussion AANAT is a key regulatory enzyme in the melatonin biosynthetic pathway [17,16]. AANAT activity in retinal photoreceptor cells and pineal gland of chicken and many other vertebrate species exhibits circadian rhythmicity, with enzyme activity peaking at night [5,6,9]. Cultured photoreceptor cells prepared from embryonic chick retinas have been used extensively to elucidate regulatory mechanisms involved in retinal melatonin biosynthesis and

Fig. 5. Circadian fluctuations of AANAT activity in E6 retinal cells cultured for 8–10 days. Photoreceptor cells were prepared from E6 retinas and incubated for 8 days under LD. Illumination was switched from LD to DD before expected onset of light at the beginning of DIV 9. White symbols represent AANAT activity at ZT 10 in light; black symbols represent ZT 20 in darkness; gray symbols represent subjective day, ZT 10, in darkness. The horizontal white and black bars above the x-axis represent times of light and darkness, respectively. AANAT activity was significantly higher at night than during the day in LD on DIV 8 (P,0.001) and this fluctuation persisted in DD on DIV 9 (P50.011) and DIV 10 (P50.029). Data are presented as mean6S.E.M., n515–16 per group.

receptor-mediated actions [7,14,12], daily rhythms of photoreceptor retinomotor movements [22,24], and circadian regulation of iodopsin gene expression [20]. However, to date, the development of circadian oscillations of melatonin biosynthesis in cultured chicken photoreceptor cells has not been studied. The present study demonstrates that cultured photoreceptor cells prepared from neuronal retinas of chicken embryos express circadian rhythms of AANAT activity following entrainment to a light–dark cycle of illumination. Our results demonstrate that photoreceptor-enriched cell cultures prepared from E6 retinas or E8 retinas and incubated under LD, displayed prominent daily fluctuations in AANAT activity on days 5 and 6 in vitro. When entrained to LD in vitro for 5 days, E8 cells, but not E6 cells, exhibited circadian rhythms of AANAT activity on subsequent days in DD. Circadian fluctuation of AANAT activity in E6 cells appeared only after 8–9 days of incubation under LD in vitro. This study provides strong evidence that the circadian clock control of AANAT activity in cultured chicken photoreceptor cells develops at a later stage than the photic control mechanisms and that the circadian rhythmicity is indeed an intrinsic property of retinal cells. Moreover, photic and circadian control of

T.N. Ivanova, P. Michael Iuvone / Brain Research 973 (2003) 56–63

Fig. 6. Effect of light exposure at night on AANAT activity in E6 retinal cells cultured for 8–10 days. Photoreceptor cells were prepared from E6 retinas and incubated for 8 days under LD, followed by 2 days in DD. White circles represent AANAT activity at ZT 10 in light; black circles represent ZT 20 in darkness; gray symbols represent subjective day, ZT 10, in darkness; the white star represents samples exposed to light from ZT20-22. The cells displayed significantly lower AANAT activity during the daytime (ZT 10) than at night (ZT 20) in LD on DIV 8 (P,0.001). In DD on DIV 9, AANAT activity decreased during the subjective daytime (P,0.001) and then showed a progressive increase, peaking in the late night (ZT 10 vs. 22, P50.002). One group of cells was exposed to light for 2 h during the night of DIV 9 (ZT 20–22). Light dramatically suppressed AANAT activity compared to cells kept in darkness at the same time (ZT 22 DD vs. ZT 22 L, P,0.001). Data are presented as mean6S.E.M, n512–15 per group.

AANAT activity develops more rapidly in vitro than in ovo. In E6 cultures, daily fluctuations of enzyme activity in LD develop with a combined embryonic and in vitro age of #11 days. In contrast, daily fluctuations of AANAT activity in ovo were not observed until embryonic day 20 [11]. Similarly, circadian fluctuations are observable with a combined embryonic and in vitro age of 14–15 days, while circadian control of enzyme activity does not emerge in ovo until just prior to or after hatching [11]. The factors responsible for this accelerated development of photic and circadian control mechanisms in vitro are unknown. Photic control of retinal AANAT and melatonin production also precedes circadian control during development of other species in vivo. Daily fluctuations of AANAT mRNA are observed as early as postnatal day 2 in rat retina under LD, but circadian rhythms of the transcript in DD are not observed until after postnatal day 14 [21]. During ontogenesis of Xenopus laevis, melatonin release

61

Fig. 7. Effect of continuous illumination on circadian oscillations of AANAT activity in E6 retinal cells cultured for 8–10 days. Cells were prepared from E6 retinas and incubated for 8 days under LD followed by 2 days in constant light (LL). White circles represent AANAT activity in light at ZT 10; black circles represent ZT 20 in darkness; gray symbols represent subjective night, ZT 20, in light; the black star represents samples exposed to darkness from ZT20–22. AANAT activity was significantly higher at night in LD than during the daytime on DIV 8 (P,0.001). In LL, night time enzyme activity was noticeably depressed by constant light, but exhibited a weak rhythm on DIV 9 (ZT 10 vs. ZT 20, P50.025). On the second day of LL (DIV 10), no difference was observed between AANAT during the subjective day and subjective night (ZT 10 vs. ZT 20, P.0.05). Exposing cells to darkness for 2 h during the subjective night (ZT 20 –22) of DIV 9 had no effect on AANAT activity (ZT 20 light vs. ZT 22 dark, P.0.05). Data are presented as mean6S.E.M, n59–12 per group.

from eyes is weakly rhythmic in LD by stage 26 / 29, but no rhythm is observed in DD [8]; by stage 41, melatonin release from eyes is rhythmic with similar amplitudes in LD and DD. Quail retinal cells, prepared from E5 embryos and incubated for 4 days under LD, have a circadian clock that controls iodopsin gene expression on subsequent days in DD [19]. These cells also show day–night rhythms of melatonin release in LD, but the rhythms are not maintained in DD. This observation led to the suggestion by Pierce [18] that the circadian clock is not ‘hooked up’ to melatonin release or that different oscillators regulate melatonin release and iodopsin expression in cultured quail photoreceptors. Similarly, E6 chick retinal cells incubated for 5 days in LD display circadian control of iodopsin expression on subsequent days in DD [20]. Here we have shown that similar chick retinal cell cultures incubated for

62

T.N. Ivanova, P. Michael Iuvone / Brain Research 973 (2003) 56–63

5 days display day–night rhythms of AANAT in LD, but not in DD. Circadian control of AANAT activity in the cultured cells does not develop until the cultures are |3 days older. Thus, circadian control of iodopsin expression appears to develop earlier than clock control of AANAT activity and melatonin synthesis in chick retinal cells, as well as in quail retinal cells. The mechanisms underlying this differential circadian clock output are unknown, but may involve the development of specific clock output pathways. AANAT activity in the chicken retina is under control of light, which suppresses enzyme activity [5,9,10,13]. In retinas of posthatch chickens, light exposure at night rapidly suppresses AANAT activity and melatonin levels [9,10]. The present study demonstrates a similar effect in cultured photoreceptor cells. Brief (2 h) light exposure at night dramatically decreased AANAT activity in the cells incubated in DD. Furthermore, we found that constant illumination dramatically suppresses AANAT activity, as it does in posthatch chickens [5]. In cells incubated under LL, the lack of effect on AANAT activity of a brief period (2 h) of darkness during the subjective night indicates that the inhibition of AANAT activity by light is not readily reversible and that subsequent increases of enzyme activity require new enzyme synthesis. This conclusion is consistent with in vivo evidence that light stimulates the degradation of retinal AANAT by a proteasomal mechanism [13]. The present study also demonstrates that 9-cis-retinoic acid increases AANAT activity in cultured retinal cells. Retinoic acid promotes the differentiation and survival of photoreceptor cells in vitro [23,15], and this may account in part for the increase of AANAT activity. In addition, retinoic acid increases the expression in Y79 retinoblastoma cells of the mRNA and activity of hydroxindole-Omethyltransferase [3], the enzyme that converts the AANAT reaction product, N-acetylserotonin, to melatonin. Thus, retinoic acid may coordinately regulate both enzymes of the melatonin biosynthetic pathway. The effect of retinoic acid appeared to be transient; when added on DIV 4, it increased AANAT activity on DIV 5 and 6, but not on DIV 7. It seems likely that the loss of effect on DIV 7 is due to a time-dependent metabolism / isomerization of the retinoid. Hence, in experiments exceeding 6 DIV, cells should be re-fed often with fresh retinoic acid-containing medium. In conclusion, these studies demonstrate that chick retinal cell cultures develop photic and circadian control of AANAT activity that recapitulates development and control of the enzyme in vivo. However, development of these factors is accelerated in vitro. Thus, cultured chick retinal cells are a useful model to investigate the differential clock control of AANAT and other clock controlled genes, such as iodopsin, to probe the molecular mechanism of the circadian clock in photoreceptor cells, and to investigate

the signaling mechanisms whereby light entrains circadian oscillators and suppresses melatonin biosynthesis.

Acknowledgements We wish to express our thanks to Drs A.L. AlonsoGomez and R. Haque for pilot data relevant to this study. We also wish to thank A. Brown, Dr S.S. Chaurasia and J. Wessel for their assistance. This work was supported by grant EY04864 from the National Institute of Health. A preliminary report of some of these data was presented at the 2002 meeting of the Association for Research in Vision and Ophthalmology.

References [1] R. Adler, M. Hatlee, Plasticity and differentiation of embryonic retinal cells after terminalmitosis, Science 243 (1989) 391–393. [2] R. Adler, J.D. Lindsey, C.L. Elsner, Expression of cone-like properties by chick embryo retina cells in glial-free monolayer cultures, J. Cell Biol. 99 (1984) 1173–1178. [3] M. Bernard, D.C. Klein, Retinoic acid increases hydroxyindole-Omethyltransferase activity and mRNA in human Y-79 retinoblastoma cells, J. Neurochem. 67 (1996) 1032–1038. [4] M. Bernard, D.C. Klein, M. Zatz, Chick pineal clock regulates serotonin N-acetyltransferase mRNA rhythm in culture, Proc. Natl. Acad. Sci. USA 94 (1997) 304–309. [5] M. Bernard, P.M. Iuvone, V.M. Cassone, P.H. Roseboom, S.L. Coon, D.C. Klein, Avian melatonin synthesis: photic and circadian regulation of serotonin N-acetyltransferase mRNA in the chicken pineal gland and retina, J. Neurochem. 68 (1997) 213–224. [6] J.C. Besharse, P.M. Iuvone, M.E. Pierce, Regulation of rhythmic photoreceptor metabolism: a role for post-receptoral neurons., in: N. Osborne, G.J. Chader (Eds.), Progress in Retinal Research, Pergamon Press, Oxford, 1988, pp. 21–61. [7] J. Gan, A.L. Alonso-Gomez, G. Avendano, B. Johnson, P.M. Iuvone, Melatonin biosynthesis in photoreceptor-enriched chick retinal cell cultures: role of cyclic AMP in the K(1)-evoked, Ca(21)-dependent induction of serotonin N-acetyltransferase activity, Neurochem. Int. 27 (1995) 147–155. [8] C.B. Green, M.Y. Liang, B.M. Steenhard, J.C. Besharse, Ontogeny of circadian and light regulation of melatonin release in Xenopus laevis embryos., Dev. Brain Res. 117 (1) (1999) 109–116. [9] H.E. Hamm, M. Menaker, Retinal rhythms in chicks: circadian variation in melantonin and serotonin N-Acetyltransferase activity, Proc. Natl. Acad. Sci. USA 77 (1980) 4998–5002. [10] H.E. Hamm, J.S. Takahashi, M. Menaker, Light-induced decrease of serotonin N-acetyltransferase activity and melatonin in the chicken pineal gland and retina, Brain Res. 266 (1983) 287–293. [11] P.M. Iuvone, Development of melatonin synthesis in chicken retina: regulation of serotonin N-acetyltransferase activity by light, circadian oscillators, and cyclic AMP, J. Neurochem. 54 (1990) 1562– 1568. [12] P.M. Iuvone, J. Gan, Functional interaction of melatonin receptors and D1 dopamine receptors in cultured chick retinal neurons, J. Neurosci. 15 (1995) 2179–2185. [13] P.M. Iuvone, A.D. Brown, R. Haque, J. Weller, J.B. Zawilska, S.S. Chaurasia et al., Retinal melatonin production: role of proteasomal proteolysis in circadian and photic control of arylalkylamine Nacetyltransferase, Invest. Ophthalmol.Vis. Sci. 43 (2002) 564–572.

T.N. Ivanova, P. Michael Iuvone / Brain Research 973 (2003) 56–63 [14] P.M. Iuvone, G. Avendano, B.J. Butler, R. Adler, Cyclic AMPdependent induction of serotonin N-acetyltransferase activity in photoreceptor-enriched chick retinal cell cultures: characterization and inhibition by dopamine, J. Neurochem. 55 (1990) 673–682. [15] M.W. Kelley, J.K. Turner, T.A. Reh, Retinoic acid promotes differentiation of photoreceptors in vivo, Development 120 (1994) 2091–2102. [16] D.C. Klein, J.L. Weller, Indole metabolism in the pineal gland: a circadian rhythm in N-acetyltransferase, Science 169 (1970) 1093– 1095. [17] D.C. Klein, S.L. Coon, P.H. Roseboom, J.L. Weller, M. Bernard, J.A. Gastel et al., The melatonin rhythm-generating enzyme: molecular regulation of serotonin N-acetyltransferase in the pineal gland, Recent Prog. Horm. Res. 52 (1997) 307–357. [18] O.H. Lowry, N.J. Rosebrough, A.L. Farr, R.J. Randall, Protein measurement with the folin phenol reagent, J. Biol. Chem. 193 (1951) 265–275. [19] M.E. Pierce, Circadian organization in quail retina: differential regulation of melatonin synthesis and iodopsin gene expression in vitro, Vis. Neurosci. 16 (1999) 843–848. [20] M.E. Pierce, H. Sheshberadaran, Z. Zhang, L.E. Fox, M.L. Appleb-

[21]

[22]

[23]

[24]

[25]

63

ury, J.S. Takahashi, Circadian regulation of iodopsin gene expression in embryonic photoreceptors in retinal cell culture, Neuron 10 (1993) 579–584. K. Sakamoto, K. Oishi, N. Ishida, Ontogeny of circadian expression of serotonin N-acetyltransferase mRNA in the rat retina, Neurosci. Lett. 317 (1) (2002) 53–55. D.L. Stenkamp, R. Adler, Photoreceptor differentiation of isolated retinal precursor cells includes the capacity for photomechanical responses, Proc. Natl. Acad. Sci. USA 90 (1993) 1982–1986. D.L. Stenkamp, J.K. Gregory, R. Adler, Retinoid effects in purified cultures of chick embryo retina neurons and photoreceptors, Invest. Ophthalmol. Vis. Sci. 34 (1993) 2425–2436. D.L. Stenkamp, P.M. Iuvone, R. Adler, Photomechanical movements of cultured embryonic photoreceptors: regulation by exogenous neuromodulators and by a regulable source of endogenous dopamine, J. Neurosci. 14 (1994) 3083–3096. K.B. Thomas, J. Zawilska, P.M. Iuvone, Arylalkylamine (serotonin) N-acetyltransferase assay using high performance liquid chromatography with fluorescence or electrochemical detection of Nacetyltryptamine, Anal. Biochem. 184 (1990) 228–234.