JOURNAL OF EXPERIMENTAL ZOOLOGY 296A:56–62 (2003) Development of Rhythmic Melatonin Secretion From the Pineal Gland of Embryonic Mummichog (Fundulus heteroclitus) DEBRA ROBERTS1, DARREN K. OKIMOTOn2, CARA PARSONS3, MARTIN STRAUME4, and MILTON H. STETSON3 1 Boston University School of Medicine, Boston, Massachusetts 2 American Samoa Community College, Pago Pago, American Samoa 3 Department of Biological Sciences, University of Delaware, Newark, Delaware 4 NSF Center for Biological Timing, University of Virginia, Charlottesville, Virginia ABSTRACT The pineal gland of vertebrates produces and secretes the hormone melatonin in response to changes in the light-dark cycle, with high production at night and low production during the day. Melatonin is thought to play an important role in synchronizing daily and/or seasonal physiological, behavioral, and developmental rhythms in vertebrates. In this study, the functional development of the pineal melatonin-generating system was examined in the mummichog, Fundulus heteroclitus, an euryhaline teleost. In this species, the pineal gland contains an endogenous oscillator, ultimately responsible for timing the melatonin rhythm. Oocytes from gravid females were collected and fertilized in vitro from sperm collected from mature males. Skull caps containing attached pineal glands were obtained from F. heteroclitus embryos at different embryonic stages and placed in static or perfusion culture under various photoperiodic regimes. Rhythmic melatonin secretion from pineal glands of embryonic F. heteroclitus embryos exposed to a 12L:12D cycle in static culture was observed at five days post-fertilization. The ontogeny of circadian-controlled melatonin production from F. heteroclitus pineal glands exposed to constant darkness for five days was also seen at day five post-fertilization. These data show that early development of the pineal melatonin-generating system in this teleost occurs prior to hatching. Pre-hatching development of the melatonin-generating system may confer some selective advantage in this species in its interactions with the environment. J. Exp. Zool. 296A:56–62, 2003. r 2003 Wiley-Liss, Inc. INTRODUCTION through which it reaches various target organs and tissues, including other regions of the brain The pineal gland of teleosts is located dorsal to (Takahashi et al., ’89; Underwood, ’89; Bolliet the forebrain just beneath the roof of the skull. et al., ’96b). The gland is attached to the diencephalon by a Melatonin is synthesized in pineal photorecep- long stalk and contains three major cell types: tor cells from serotonin (Takahashi et al., ’89; photoreceptor cells, ganglion cells, and interstitial Pickard and Tang, ’94). Serotonin is converted to cells, which have a supportive role (Pickard and N-acetylserotonin by the enzyme N-acetyltrans- Tang, ’94). Pineal photoreceptor cells, which are ferase (NAT), the rate-limiting step in the mela- similar in structure to those found in most tonin synthesis pathway. The daily changes in vertebrate retinas and in all vertebrates except melatonin secretion depend on the daily changes mammals, are fully developed, rudimentary in NAT activity (Falcon et al., ’89; Cahill, ’97). rod-like photoreceptor cells. In non-mammalian N-acetylserotonin is then methylated by the vertebrates, light-dark information is detected by enzyme hydroxyindole-O-methyltransferase to pineal photoreceptor cells through the transparent bones of the skull (Quay, ’79) and transduced into Grant sponsor: NSF; Grant number: IBN 93-18203. n a rhythmic endocrine signal in the form of Correspondence to: Darren K. Okimoto, American Samoa Com- munity College, P.O. Box 2609, Pago Pago, American Samoa 96799. the indoleamine, N-acetyl-5-methoxytryptamine E-mail: okimotod@hawaii.edu (melatonin). Melatonin is secreted from the highly Received 28 March 2002; Accepted 11 November 2002 Published online in Wiley InterScience (www.interscience.wiley. vascularized pineal gland into the circulation, com). DOI: 10.1002/jez.a.10241 r 2003 WILEY-LISS, INC.  PINEAL DEVELOPMENT IN THE MUMMICHOG 57 form melatonin. Little or no melatonin is synthe- receptor cells and the onset of circadian rhythmi- sized and released by the pineal gland during the city in the pineal gland of the mummichog, day or photophase, while much higher amounts Fundulus heteroclitus. We choose F. heteroclitus are released during the night or scotophase for this study because it is oviparous. Thus, F. (Meissl and Ekstrom, ’88; Zachmann et al., ’92). heteroclitus eggs were obtained from gravid Melatonin often induces the production or release females and fertilized in vitro, and subsequent of other hormones and is known to play an embryo development occurred under controlled especially vital role in regulating sex hormones conditions. In addition, previous experiments run and hence, reproductive development, function, in our laboratory have indicated that the pineal and behavior (Davis et al., ’86; Popek et al., ’92; gland of adult F. heteroclitus contains a circadian Bromage et al., ’95; Khan and Thomas, ’96). oscillator(s) that drives rhythmic melatonin secre- Most fish species examined to date contain an tion (Okimoto and Stetson, ’97). endogenous circadian oscillator(s) in the pineal gland (¼ clocked pineal) that regulates the daily MATERIALS AND METHODS production of melatonin (Falcon et al., ’89; Iigo All protocols in this study were approved by the et al., ’91; Bolliet et al., ’95, ’96a; Cahill, ’96; Lab Animal Care and Use Committee at the Okimoto and Stetson, ’97, ’99b). Other nonmam- University of Delaware and conducted in strict malian vertebrates that demonstrate clock-con- accordance with the NIH Guidelines for the Care trolled rhythmicity of melatonin release from and Use of Laboratory Animals. the pineal gland include the domestic chick (Takahashi et al., ’80), the zebra finch (Van’t Fish procurement Hof and Gwinner, ’96), and the lizard (Menaker and Wisner, ’83). Mature mummichogs were caught in minnow Two salmonids, the rainbow trout (Oncor- traps in Canary Creek at the University of hynchus mykiss) and masu salmon (Oncorhynchus Delaware Lewes Campus. They were brought to masou), however, are the only teleosts studied to the University of Delaware Newark Campus, date that have pineal glands that do not display a where eggs were collected from gravid females by clock-controlled rhythm in pineal melatonin pro- running a thumb along the abdomen of the fish duction (Gern and Greenhouse, ’88; Falcon et al., and allowing the eggs to extrude. Two mature ’92; Max and Menaker, ’92; Iigo et al., ’98). In males were then sacrificed via rapid decapitation. these fishes, in vitro melatonin production and The testes were removed, diced and mashed, and secretion remains elevated in constant darkness mixed with the ova in a 50-ml plastic vial. After 20 and is regulated by environmental light-dark min, 30 ml of a 30% (10 parts per thousand) sea cycles (Gern and Greenhouse, ’88; Falcon et al., water (SW) solution were added to this vial and ’92; Iigo et al., ’98). Two elasmobranchs, the the contents gently shaken. The eggs remained in smooth dogfish shark (Squalus acanthias) and this solution for two hr and then inspected for cell the scalloped hammerhead shark (Sphyrna division. Upon determination of cell division, lewini), also have pineal glands that demonstrate fertilized eggs were transferred to plastic petri constant, elevated melatonin release in constant dishes (50–75 per dish) lined with a saline-wetted darkness (Okimoto and Stetson, ’97). In addition, filter paper. The petri dishes were then placed in constant lighting (LL) conditions have been shown an incubation chamber (Billups-Rothenberg; Del to inhibit melatonin production in fish for the Mar, CA). The chamber was filled with 95% O2/5% duration of exposure (Bolliet et al., ’95; Okimoto CO2. The fertilized eggs were exposed to a and Stetson, ’99a). The chick pineal gland, how- 12L:12D cycle at 261C with lights off from 20:00 ever, will maintain rhythmic melatonin release in hr to 08:00 hr Eastern Standard Time. The LL, although the amplitude is somewhat attenu- dishes were wrapped with foil to produce dark ated (Csernus et al., ’98). conditions. Little is known about the functional develop- Pineal gland dissection ment of the pineal gland in vertebrates. The embryonic chick has been shown to entrain to a Surgical procedures were performed under a light-dark cycle on embryonic day 18, and the surgical dissecting scope. First, each embryo was zebrafish pineal gland responds to a light-dark removed from its egg, decapitated, and both eyes cycle by 43 hr post-fertilization. In this study we were completely removed. Then, the skull cap was examined the functional development of photo- separated from the rest of the cranium via two 58 D. ROBERTS ET AL. incisions from the corner of the eye posteriorly remained in constant darkness (DD) for four days. along the skull. The skull cap was cleaned of any In the second perfusion run, pineal glands were adhering brain tissue, but the pineal gland collected from embryos at 114 hr post-fertilization, remained attached to the skull cap. The skull cap transferred to seven perfusion chambers (n¼12 was then placed in sterile culture medium. The glands/chamber) and exposed to DD for five medium used was Eagle’s Minimum Essential days following lights off at 20:00 hr. Collected Medium (Sigma Chemical, M-4144) supplemented medium was stored at
201C until assayed for with sodium bicarbonate (2.2 gm/liter
1), L-gluta- melatonin. mine (0.292 gm/liter
1), penicillin-G (1000 units/ liter
1), streptomycin sulfate (0.1 gm/liter
1), and Radioimmunoassay amphotericin-B (2.5 mg/liter
1). Melatonin was assayed directly from the col- lected culture medium using the radioimmunoas- Pineal gland culture say method described by Nagy et al. (’96) with modifications described by Okimoto and Stetson Photoreceptor cell development was monitored (’99a). The only modifications made to this in static culture of embryonic pineal glands. procedure were the use of Scintisafe scintillant Embryos were tested beginning three days post- in place of the toluene-based scintillant and 7-ml fertilization through 11 days post-fertilization. scintillation vials instead of the 20-ml vials. Hatching occurred 10 days after fertilization. After dissection, six skull caps were placed in 1.0 Statistical analysis ml of culture medium in each of 4–6 wells of a 24- Analysis of static culture data was done using well culture plate. The plates were placed in an the SAS Statistical Institute software package incubation chamber, which was then filled with (Ver. 6.11). Data were analyzed by a two-way 95%O2/5%CO2. The chamber was incubated at GLM ANOVA for significant effects due to photo- 261C under a culture hood containing a 30-watt period and time. Comparisons were made using an fluorescent bulb. Glands were exposed to a LSMEANS test. Results were considered signifi- 12L:12D cycle with lights on from 08:00 hr to cant when Po0.05. Analysis of perfusion culture 20:00 hr. After each 12-hr period, glands were data for circadian rhythmicity was conducted with transferred to an adjacent well containing fresh the FFT-NLLS analysis software package devel- medium. The medium from which the glands were oped by Martin Straume (Straume et al., ’91; transferred was collected and frozen at
201C Plautz et al., ’97). The specified level of confidence until radioimmunoassay for melatonin. Dark con- probability used to estimate circadian periodicity ditions were created by wrapping the plates with was 95%. aluminum foil. Each set of glands harvested remained in culture for 48 hr. RESULTS Circadian development of the pineal gland was also examined in F. heteroclitus. A perfusion Photoreceptor development as indicated by the culture system was utilized to test for rhythmic production and secretion of higher melatonin melatonin production and release from embryonic levels in the dark period than in the preceding F. heteroclitus pineal glands. Twelve glands, from light period was initially seen in F. heteroclitus embryos at 114 hr (4.75 days) post-fertilization, pineal glands on the fifth day post-fertilization were obtained and placed in each of three (120–144 hr, Fig. 1). This rhythmic pattern of perfusion chambers. The chambers were kept in melatonin release in synchrony with the light- a water bath (261C) placed in a light-tight box dark cycle continued thereafter, throughout fitted with a 20-watt fluorescent bulb controlled by embryonic development (Fig. 1). an electronic programmable timer. Culture med- The circadian oscillator(s) in the F. heteroclitus ium was pumped via a peristaltic pump from pineal gland develops (matures) synchronously individual reservoirs through each chamber at a with photoreceptor development (120–144 hr). rate of 300 ml/hr-1. Samples were collected in glass This was demonstrated by rhythmic fluctuations tubes every 1.5 hr on a fraction collector. The in melatonin production and secretion from F. culture was started in the light at 15:30 hr with heteroclitus pineal glands exposed to DD (Fig. 2a lights off at 20:00 hr. The lights were programmed and 2b). Circadian rhythmicity was detected in to automatically turn on at 08:00 hr each morning seven of the ten groups of glands that were placed and off at 20:00 hr each night. The glands into perfusion culture. The period estimates of  PINEAL DEVELOPMENT IN THE MUMMICHOG 59 DISCUSSION This investigation has revealed from the time of fertilization, the developmental pattern of the pineal melatonin-generating system in the ovipar- ous mummichog. At about the midpoint of embryonic development (on day five post-fertiliza- tion), pineal photoreceptor cells in F. heteroclitus develop, as indicated by rhythmic melatonin production and secretion in vitro on this day and thereafter, prior to hatching. This finding is in agreement with results presented by Okimoto and Stetson (’97). While this investigation did not examine retinal photoreceptor development in F. heteroclitus, Hollyfield (’72), who studied the histogenesis of the Fundulus retina, reported that the first signs of retinal differentiation occurred at stage 28 (Armstrong and Child, ’65), approxi- mately 128 hr post-fertilization at 201C. Separate retinal cell-layers were clearly differentiated by stage 31 (168 hr at 201C), and outer segments on the photoreceptor cells could be seen clearly by stage 32 (192 hr at 201C). These data suggest that retinal photoreceptor cells may be functional by 192 hr post-fertilization. Our study was conducted at 261C; therefore retinal development would have occurred at a faster rate. In a recent study, Kazimi and Cahill (’99) found that the development of retinal photoreceptor cells takes place between 48 and 50 hr post-fertilization in embryonic zebra- fish. These investigators also demonstrated a light-dark rhythm in pineal melatonin production Fig. 1. Ontogeny of rhythmic melatonin secretion from between 31 and 43 hr post-fertilization, along with pineal glands of Fundulus heteroclitus embryos exposed to a photospecific-marker detection in the pinealocytes 12L:12D cycle in static culture. Six pineal glands collected from embryos ranging in age from 3–11 days (72–222 hr) post- at 24 hr post-fertilization. Zebrafish hatch on day fertilization were pooled per well and incubated in 1 ml of 3 post-fertilization when raised at 28.51C; there- culture medium at 261C (a-h). Each bar represents the mean fore zebrafish retinal and pineal photoreceptor 7 SE (N¼4–6). Samples were collected every 12-hr over a cells develop prior to hatching as those of 48-hr period. The age of the glands (in hr post-fertilization) F. heteroclitus. while in culture is depicted on the x-axis. The light-dark cycle is indicted by the black (dark period) and white bars (light It is interesting to note that photoreceptor cell period). *denotes Po0.05, **denotes Po0.01, ***denotes functionality appears to occur when approxi- Po0.001. mately 40% of the time has elapsed between fertilization and coordinated swimming and feed- ing are first observed in these fish species. In F. heteroclitus, photoreceptor cell development was these melatonin rhythms ranged from 22.64–27.9 observed on day five and coordinated swimming hr (Table 1). The free-running rhythm in Figure and feeding was observed on day 12 post-fertiliza- 2c represents the combined output of 24 glands tion (41.67%). In the zebrafish, photoreceptor cell (tissue chambers 1 and 3) in perfusion run 1, development was observed between 31–43 hr post- whereas the free-running rhythm in Figure 2d is fertilization (day 1.75), and coordinated swimming the aggregate profile of 60 glands (five tissue and feeding was observed between days four to five chambers) in run 2. In both cases (Figs. 2c and 2d), (38.9%). Studies run in our laboratory on the circadian rhythmicity was revealed by the FFT- cichlid, Oreochromis mossambicus, showed that NLLS analysis. photoreceptor cell development occurred on day 8 60 D. ROBERTS ET AL. Fig. 2. Circadian rhythmicity in melatonin running rhythm (closed circles, thin line) in (c) secretion from pineal glands of Fundulus het- represents the combined output from two tissue eroclitus embryos in perfusion culture. In the chambers (N¼24 glands) in run 1. In (d), the first perfusion run (a), pineal glands were free-running rhythm (closed circles, thin line) exposed to a 12L:12D cycle for one day followed represents the combined output from five tissue by constant darkness for four days, while in the chambers (N¼60 glands) in run 2. The culture second run (b), to constant darkness (DD) for time in hr under DD exposure is indicated on five days. The melatonin profiles (closed circles, the x-axis in (c) and (d). In (a-d), the dark, thin line) in (a) and (b) are typical examples that thicker line that overlays the DD portion of each represent the output from 12 glands that were profile is the theoretical best-fit line generated collected from embryos at 114 hr (4.75 days) by the FFT-NLLS analysis. The light-dark cycle post-fertilization, pooled in a single tissue is indicted by the black (dark period) and white chamber, and exposed to culture medium at a bars (light period). Estimates of the free- flow rate of 1 ml/hr-1 at 261C. The age of the running rhythm (Tau) are indicated in hr above gland (in hr post-fertilization) while in culture is each plot together with their associated 95% depicted on the x-axis. Samples were collected confidence limit estimates. every 1.5 hr over a five-day period. The free- post-fertilization, while coordinated swimming demonstrate a pineal melatonin rhythm that is and feeding did not take place until day 18–19 light responsive on day two post-hatch, and a post-fertilization (43.25%, unpublished results). retinal melatonin rhythm by day seven post-hatch While photoreceptor cell developmental studies (Van’t Hof and Gwinner, ’96). are decidedly lacking in fish, more research has Besides addressing the issue of functional been done with birds and mammals, most notably photoreceptor cell development, our study also the chick and the rat. Numerous studies have examined the development of the circadian oscil- indicated that the pineal gland of the chick embryo lator(s) in the pineal gland of embryonic F. is capable of entraining to a light-dark cycle by heteroclitus. In this study, we found it necessary embryonic day 18 (Zeman et al., ’92; Lamosova to pool twelve glands in a single tissue chamber in et al., ’95; Akasaka et al., ’95; Csernus et al., ’98; order to obtain sufficient quantities of secreted Mackova et al., ’98; Zeman et al., ’99). In the melatonin to assay. Thus, the melatonin profile neonate rat, retina-mediated light-dark entrain- obtained from each set of glands in perfusion ment begins as early as day six, and the process is culture represent the combined output or aggre- completed by day eight (Duncan et al., ’86). Zebra gate rhythm of that pool. One consequence of finches (Poephila guttata) have been shown to pooling is a greater potential for the introduction  PINEAL DEVELOPMENT IN THE MUMMICHOG 61 TABLE 1. Period estimates (Tau) from the FFT-NLLS analysis of possible sources that could contribute to the melatonin pro¢les in constant darkness (DD) from embryonic measured melatonin levels in the zebrafish. It is Fundulus heteroclitus pineal glands.Twelve glands were pooled in possible that the rhythm described is affected by each tissue chamber and maintained in perifusion culture at 261C. In retinal melatonin production, because it was perifusion run 1, gland were exposed to a 12L:12D cycle for one day noted earlier that retinal photoreceptor cells followed by DD for four days. In run 2, glands were exposed to DD for are developed by 48–50 hr post-fertilization, and ¢ve days. n.d. denotes that no circadian rythm was detected.The error estimates for Tau are the 95% con¢dence limit estimates there is no data supporting the notion that no melatonin is produced by the retina prior to this Perifusion Tau7error estimate time. Chamber run (hr) In conclusion, this study shows that light-dark entrainment of pineal melatonin production and 1 1 25.0671.69 2 1 n.d. secretion occurs early (e.g., prior to hatching) 3 1 24.9871.22 during embryonic development in the mummi- 1 2 22.6472.73 chog. This likely works to the advantage of this 2 2 27.9074.00 species, causing its activity rhythms to be set to 3 2 23.2770.40 the light-dark cycle; therefore, the young fish can 4 2 n.d. feed and be active at optimal times of day. Further 5 2 22.8971.60 studies need to be undertaken in F. heteroclitus to 6 2 23.3171.10 7 2 n.d. elucidate the functional development of the retinal melatonin-generating system. of noise in the time series data that could result from damping or desynchronization of circadian LITERATURE CITED oscillators among each set of glands in culture. Nevertheless, rhythmic melatonin production was Akasaka K, Nasu T, Katayama T, Murakami N. 1995. observed in the majority of pooled glands placed in Development of regulation of melatonin release in pineal cells in chick embryo. Brain Res 692:283–286. perfusion culture (Table 1). The inability of the Armstrong PB and Child JS. 1965. Stages in the normal FFT-NLLS analysis to detect a significant circa- development of Fundulus heteroclitus. Bio Bul 128:143–168. dian component in the other groups of glands Bolliet V, Falcon J, Ali MA. 1995. Regulation of the melatonin could be attributed to the pooling effect. secretion by light in the isolated pineal organ of the white In a previous study, using a static culture sucker (Catastomus commersoni). J Neuroendocrinol 7: 535–542. method, it was reported that F. heteroclitus did Bolliet V, Ali MA, Lapointe F-J, Falcon J. 1996a. Rhythmic not demonstrate a circadian rhythm of melatonin melatonin secretion in different teleost species: An in vitro production in DD at day eight or 10 post-fertiliza- study. J Comp Physiol B 165:677–683. tion (Okimoto and Stetson, ’97). The current Bolliet V, Begay V, Taragnat C, Ravault JP, Collin JP, Falcon study, using perfusion culture, determined that J. 1996b. Photoreceptor cells of the pike pineal organ as circadian oscillators. Eur J Neurosci 9:643–653. the circadian oscillator in the pineal gland of F. Bromage NR, Randall CF, Porter MJR, Davies B. 1995. How heteroclitus is functional at the time of photo- do photoperiod, the pineal gland, melatonin, and circannual receptor cell development, which is day five post- rhythms interact to co-ordinate seasonal reproduction in fertilization. The reason for the discrepancy may salmonid fish? In: Goetz F, Thomas, P, editors. Proceedings be due to the fact that the static culture technique of the Fifth International Symposium of Reproductive Physiology of Fish, The University of Texas at Austin, 2–8 requires disturbance of the gland at 12-hr inter- July 1995. Austin, TX: Fish Symposium 95. p 164–166. vals in order to collect the culture medium, while Cahill GM. 1996. Circadian regulation of melatonin produc- the perfusion technique allows the gland to sit tion in cultured zebrafish pineal and retina. Brain Res undisturbed in culture and be continuously ex- 708:177–181. posed to fresh medium. Cahill GM. 1997. Circadian melatonin rhythms in cultured zebrafish pineals are not effected by catecholamine receptor Rhythmic melatonin production in DD condi- agonists. Gen Comp Endocrinol 105:270–275. tions has been demonstrated in the sailfin molly Csernus V, Ghosh M, Mess B. 1998. Development and control (Poecilia velifera) as early as the day of birth of the circadian pacemaker for melatonin release in the (Okimoto and Stetson, ’99a) and in the zebrafish chicken pineal gland. Gen Comp Endocrinol 110:19–28 at 55 hr post-fertilization (Kazimi and Cahill, ’99). Davis KB, Goudie CA, Simco BA, Mac Gregor III R, Parker, NC. 1986. Environmental regulation and influence of the However, in the study using zebrafish, melatonin eyes and pineal gland on the gonadal cycle and spawning in levels were based on whole-embryo homogenates. channel catfish (Ictalurus punctatus). Physio Zool 59: Therefore, both the retina and pineal gland are 717–724. 62 D. ROBERTS ET AL. Duncan MJ, Banister MJ, Reppert SM. 1986. Developmental Okimoto D, Stetson M. 1997. Photoregulation of melatonin appearance of light-dark entrainment in the rat. Brain Res release in vitro from the pineal gland of cartilaginous and 369:326–330. teleostean fishes: A comparative approach. In: Webb S, Falcon J, Brun-Marmillon J, Claustrat B, Collin JP. 1989. Puig-Domingo A, Moller M, Pevet P, editors. Pineal Update. Regulation of melatonin secretion in a photoreceptive pineal Westbury, NY: PJD Publications LTD. p 121–126. organ: An in vitro study in the pike. J Neurosci 9:1943–1950. Okimoto D, Stetson M. 1999a. Properties of the melatonin- Falcon J, Thibault C, Begay V, Zachmann A, Collin JP. 1992. generating system of the sailfin molly, Poecilia velifera. Gen Regulation of the rhythmic melatonin secretion by fish Comp Endocrinol 114:293–303. pineal photoreceptor cell. In: Ali MA, editor. Rhythms in Okimoto D, Stetson M. 1999b. Presence of an intrapineal Fishes. New York, NY: Plenum Press. p 167–198. circadian oscillator in the teleostean family Poeciliidae. Gen Gern WA, Greenhouse SS. 1988. Examination of in vitro Comp Endocrinol 114:304–312. melatonin secretion from superfused (Salmo gairdneri) Pickard GE, Tang W-X. 1994. Pineal photoreceptors rhythmi- pineal organs maintained under diel illumination or cally secrete melatonin. Neurosci Lett 171:109–112. continuous darkness. Gen Comp Endocrinol 71:163–174. Plautz JD, Straume M, Stanewsky R, Jamison CF, Brandes C, Gern WA, Ralph CL. 1979. Melatonin synthesis by the retina. Dowse HB, Hall JC, Kay SA. 1997. Quantitative analysis Science 204:183–184. of Drosophila period gene transcription in living animals. Hollyfield JG. 1972. Histogenesis of the retina in the J Biol Rhythms 12:204–217. killifish, Fundulus heteroclitus. J Comp Neurol 144: Popek W, Bieniarz K, Epler P. 1992. Participation of the 373–380. pineal gland in sexual maturation of female rainbow Iigo M, Kezuka H, Aida K, Hanyu I. 1991. Circadian rhythms trout (Oncorhynchus mykiss Walbaum). J Pineal Res 13: of melatonin secretion from superfused goldfish (Carassius 97–100. auratus) pineal glands in vitro. Gen Comp Endocrinol Quay WB. 1979. The parietal eye-pineal complex. In: Gans C, 83:152–158. editor. Biology of the Reptilia Vol. 9. New York, NY: Iigo M, Kitamura S, Ikuta K, Sanchez-Vazquez FJ, Ohtani- Academic Press. p 245–406. Kaneko R, Hara M, Hirata K, Tabata M, Aida K. 1998. Straume M, Frasier-Cadoret SG, Johnson ML. 1991. Least Regulation by light and dark of melatonin secretion from squares analysis of fluorescence data. In: Lakowicz JR, the superfused masu salmon (Oncorhynchus masou) pineal editor. Topics in Fluorescence Spectroscopy, Volume 2: organ. Biol Rhythm Res 29:86–97. Principles. New York: Plenum Press. p 177–240. Kazimi N, Cahill G. 1999. Development of a circadian Takahashi JS, Hamm H, Menaker M. 1980. Circadian melatonin rhythm in embryonic zebrafish. Dev Brain Res rhythms of melatonin release from individual superfused 117:47–52. chicken pineals in vitro. Proc Nat Acad Sci USA. 77:2319– Khan IA, Thomas P. 1996. Melatonin influences gonadotropin 2322. II secretion in the Atlantic croaker (Micropogonias undula- Takahashi JS, Murakami N, Nikaido SS, Pratt BL, Robertson tus). Gen Comp Endocrinol 104:231–242. LM. 1989. The avian pineal, a vertebrate model system of Lamosova D, Zeman M, Mackova M, Gwinner E. 1995. the circadian system: Cellular regulation of circadian Development of rhythmic melatonin synthesis in cultured rhythms by light, second messengers, and macromolecular pineal glands and pineal cells isolated from chick embryo. synthesis. Recent Progr Horm Res 45:279–349. Experientia 51: 970–975. Underwood H. 1989. The pineal and melatonin: Regulators Mackova M, Lamosova D, Zeman M. 1998. Regulation of of circadian function in lower vertebrates. Experientia 45: rhythmic melatonin production in pineal cells of chick 914–921. embryo by cyclic AMP. Cell Mole Life Sciences 54:461–466. Van’t Hof TJ, Gwinner E. 1996. Development of post-hatching Meissl H, Ekstrom P. 1988. Photoreceptor responses to light rhythm in zebra finches (Poephila guttata). Experientia in the isolated pineal organ of the trout, Salmo gairdneri. 52:249–252. Neuroscience 25:1071–1076. Zachmann A, Knijff SCM, Ali MA, Anctil M. 1992. Effects of Max M, Menaker M. 1992. Regulation of melatonin production photoperiod and different intensities of light exposure on by light, darkness, and temperature in the trout pineal. melatonin levels in the blood, pineal organ, and retina of the J Comp Physiol A 170:479–489. brook trout (Salvelinus fontinalis Mitchill). Can J Zool Menaker M, Wisner S. 1983. Temperature-compensated 70:25–29. circadian clock in the pineal of Anolis. Proc Nat Acad Sci Zeman M, Gwinner E, Somogyiova E. 1992. Development of USA 80:6119–6121. melatonin rhythm in the pineal glands and eyes of chick Nagy T, Gower B, Stetson MH. 1996. Influence of photoperiod, embryo. Experientia 48:765–768. time, and sex on hormone concentrations in collared Zeman M, Gwinner E, Herichova I, Lamosova D, Kost’al L. lemmings (Dicrostonyx groenlandicus). Gen Comp Endocri- 1999. Perinatal development of circadian melatonin produc- nol 101: 53–62. tion in domestic chicks. J Pineal Res 26:28–34.