Melatonin in mice: rhythms, response to light, adrenergic stimulation, and metabolism

Transcription

1 Am J Physiol Regulatory Integrative Comp Physiol 282: R358 R365, 2002; /ajpregu Melatonin in mice: rhythms, response to light, adrenergic stimulation, and metabolism D. J. KENNAWAY, A. VOULTSIOS, T. J. VARCOE, AND R. W. MOYER Department of Obstetrics and Gynaecology, Adelaide University, Medical School, Adelaide, South Australia 5005 Received 28 June 2001; accepted in final form 14 September 2001 Kennaway, D. J., A. Voultsios, T. J. Varcoe, and R. W. Moyer. Melatonin in mice: rhythms, response to light, adrenergic stimulation, and metabolism. Am J Physiol Regulatory Integrative Comp Physiol 282: R358 R365, 2002; /ajpregu There has been relatively little research conducted on pineal melatonin production in laboratory mice, in part, due to the lack of appropriate assays. We studied the pineal and plasma rhythm, response to light, adrenergic stimulation, and metabolism of melatonin in CBA mice. With the use of a sensitive and specific melatonin RIA, melatonin was detected in the pineal glands at all times of the day 21 fmol/gland in CBA mice but not in C57Bl mice. Both plasma and pineal melatonin levels peaked 2 h before dawn in a 12:12-h light-dark photoperiod ( pm and 1, fmol/gland, respectively). A brief light pulse (200 lx/15 min), 2 h before lights on, suppressed both plasma and pineal melatonin to near basal levels within 30 min. Exposure to light pulses 4 h after lights off or 2 h before lights on resulted in delays and advances, respectively, in the early morning decline of plasma and pineal melatonin on the next cycle. Administration of the -adrenergic agonist isoproterenol (20 mg/kg) 2 and 4 h after lights on in the morning resulted in a fivefold increase in plasma and pineal melatonin 2.5 to 3 h after the first injection. In the mouse, unlike the rat, melatonin was shown to be metabolized almost exclusively to 6-glucuronylmelatonin rather than 6-sulphatoxymelatonin. These studies have shown that the appropriate methodological tools are now available for studying melatonin rhythms in mice. pineal gland; circadian rhythm; phase shift; 6-glucuronylmelatonin; 6-sulphatoxymelatonin DESPITE THEIR USEFULNESS IN many diverse areas of biomedical research, there has been surprisingly little systematic research conducted on the production of melatonin by the pineal gland in laboratory mice. The most likely reasons for this situation are the small amount of blood that can be obtained from mice, their small pineal glands, and insensitive melatonin assays. Nevertheless, it has always been presumed that mice, like all other mammals, would synthesize melatonin during darkness. This assumption was challenged in 1986 when it was shown that C57Bl mice had 10 pg melatonin per pineal gland at various times during the day and night (8). Subsequently, it was shown that only two of the commonly used mouse strains, C3H and CBA, had high-amplitude melatonin rhythms with peak pineal melatonin content occurring 2 h before lights on (13). Other commonly used mouse strains (BALB/c, DBA/2, 129/Sv, AKR/J, and CF#1) had undetectable melatonin ( 20 pg/gland) at 2, 6, and 10 h after darkness onset. Melatonin is produced in the pineal gland by N- acetylation of serotonin by N-acetyltransferase (NAT) followed by O-methylation by hydroxyindole-o-methyltransferase (HIOMT). NAT is the rate-limiting enzyme in the synthesis of melatonin and is under -adrenergic control from the superior cervical ganglion. When the possible causes of the impaired melatonin production in mice were examined, it was found that C57Bl, AKR/J, and BALB/c mice had neither NAT nor HIOMT activity in the pineal gland, whereas another strain (NZB/BLNJ) had NAT but no HIOMT activity (7). The lack of NAT activity in the pineal gland was later confirmed in BALB/c mice (34). In the case of C57Bl/6J mice, it was found that a point mutation in the NAT gene results in a truncated protein that has little or no enzyme activity (30), whereas different mutations are present in BALB/c and 129/Sv strains (30). To date, there have been no reports on the nature of the HIOMT mutations. The aim of the present study was to evaluate pineal melatonin synthesis, secretion, metabolism, and the response to light in CBA, C57Bl, and BALB/c mice using a variety of approaches. Of particular interest was the possibility of establishing a method to monitor melatonin metabolite excretion rhythms in a similar way to that conducted in laboratory rats (16). MATERIALS AND METHODS Adult male CBA, C57Bl, CBAxC57Bl (F1), and BALB/c mice were obtained from the University of Adelaide Central Animal house, and wild derived mice (32) were obtained from the Animal Resources Centre (Canning Vale, Western Australia) where they had been maintained since birth on a 12:12-h light-dark photoperiod. On arrival in our facility, they continued to be on the 12:12-h light-dark photoperiod Address for reprint requests and other correspondence: D. J. Kennaway, Dept. of Obstetrics and Gynaecology, Adelaide Univ., Medical School, Frome Road, Adelaide, South Australia 5005 ( david.kennaway@adelaide.edu.au). The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. R /02 $5.00 Copyright 2002 the American Physiological Society

2 MELATONIN IN MICE R359 (lights off 1900, on 0700) and were fed standard laboratory mouse food ad libitum. Pineal and plasma melatonin rhythm studies. Male CBA and C57Bl mice were killed by decapitation at 2-h intervals in dim red light, starting at 1900 and ending at Trunk blood was collected into heparinized tubes, and plasma was harvested by centrifugation and stored frozen before assay by RIA. Skull caps with attached pineal glands were cut away and frozen in 1 ml phosphate buffer containing 0.5% BSA (PBS/BSA). Before assay, pineals were removed together with the attached dura, homogenized in the buffer, and assayed for melatonin by RIA. Isoproterenol stimulation. Groups of five CBA mice were injected with isoproterenol (20 mg/kg sc; Sigma, St. Louis, MO) or saline, 2 h after lights on and again 2 h later and killed at various times after the last injection. Pineal glands and blood were collected as above. In a separate experiment, groups of 10 CBA and 10 C57Bl male mice were injected with isoproterenol (20 mg/kg) or saline (CBA only) at 0900 and 1100, killed at 1130, and pineals and blood were collected as above. Effect of light on pineal and plasma melatonin. Male CBA mice entrained to a 12:12-h light-dark photoperiod were killed at 0500, 0600, and 0700 (n 14, 10, and 10, respectively). An additional group of five animals was exposed to 200-lx light for 15 min commencing at 0500 and killed at Pineal glands and blood were collected as previously described. In two separate experiments, the effects of a 200 lx/15 min light pulse on the timing of the morning decline in pineal melatonin content and plasma melatonin concentration were determined. Male CBA mice entrained to a 12:12-h lightdark photoperiod were exposed to light pulses at either 2300 or 0500 or were left in darkness. The lights remained off for the remainder of the experiment, and groups of five animals were killed the next morning at 0500, 0600, and Pineal glands and blood were collected for melatonin analysis as above. Urine collections. Male CBA and C57Bl mice (n 5) were acclimated in metabolism cages for several days and fed a liquid diet (Osmolite HN, Abbot Laboratories) as previously reported for rats (16). Urine was collected at 2-h intervals from 1900 to 0900 for two nights and for one night of continuous light. The mice appeared to tolerate the liquid diet, and urine volumes in excess of 200 l/2 h were regularly obtained overnight, with the urine flow decreasing slightly before the expected time of lights on. To investigate the nature of the metabolism and possible metabolites of melatonin in the mouse, five CBA and five C57Bl mice were housed in metabolism cages, placed on a liquid diet, and urine was collected at 2-h intervals. At 1600 (3 h before expected darkness), mice were injected subcutaneously with 100 g melatonin, and the lights were kept illuminated. Samples (20 l) were assayed for 6-sulphatoxymelatonin and melatonin by RIA. Assays. Melatonin was assayed in plasma from individual mice ( l) by RIA using reagents obtained from Buhlmann Laboratories (Allschwil, Switzerland). This assay uses the G280 antibody developed in our laboratory (18). Samples diluted to 1 ml with PBS/BSA were added to C18 reversephase columns, and melatonin was eluted according to the instructions accompanying the assay kit. Standards were reconstituted in PBS/BSA and were also extracted to ensure that both standards and samples were treated identically. Sensitivity of the assay ranged from 21 to 43 pm depending on the volume of sample used. Pineal gland melatonin was assayed in 100 l of homogenate by direct RIA using the Buhlmann reagents. The sensitivity was 21 fmol/gland. Urinary melatonin was assayed by direct RIA of 20 l of urine. There has been no formal attempt to prove that the immunoreactivity detected in the mouse plasma is authentic melatonin due to the very low amounts of sample available and the low levels detected. Nevertheless, the G280 antibody used has been well characterized in terms of specificity with a large number of synthetic indoles, and it has low crossreactivity with all naturally occurring indoles (15, 18). The urinary metabolite 6-sulphatoxymelatonin was assayed in urine by direct RIA as previously reported for rats (1, 17) using reagents obtained from Stockgrand, Guildford, Surrey, UK. Chromatography. To investigate the metabolism of melatonin, CBA, BALB/c, and wild derived mice and a Wistar albino rat were injected intraperitoneally with 4 mg (mice) or 40 mg melatonin (rat), and urine was collected overnight. For chromatography, 5 l of undiluted urine were applied to a silica gel thin-layer plate along with standard solutions of melatonin, 6-hydroxymelatonin, 6-sulphatoxymelatonin, and 6-glucuronylmelatonin. The plates were developed in butanol: acetic acid:water (4:1:1), and indoles were detected by spraying with Ehrlich s reagent ( p-dimethylaminobenzaldehyde). Melatonin and 6-hydroxymelatonin were obtained from Sigma. 6-Sulphatoxymelatonin was synthesized using the method of Fellenberg et al. (9). 6-Glucuronylmelatonin was synthesized enzymatically using a procedure originally designed for steroid glucuronylation (36). Briefly, this involved incubating UDP-glucuronic acid (68 mg; Sigma), bovine liver UDP-glucuronyl transferase (100 mg; Sigma), and 6-hydroxymelatonin [25 mg (Sigma) in 0.5 ml of dimethylformamide] in 4.5 ml of incubation buffer (0.3 M Tris HCl buffer, 1 mm calcium chloride, ph 8) at 37 C for 16 h. The reaction was stopped by freezing the mixture. Thin-layer chromatography revealed two major indole-positive spots, one corresponding to the starting material and another consistent with 6-glucuronylmelatonin (22). Aliquots of the reaction mixture (0.5 ml) were applied to a C18 Sep-Pak cartridge (Waters), and the presumptive 6-glucuronylmelatonin was eluted with 40% methanol (11). The material has not been crystallized or subjected to any further chemical analysis at this stage. Statistics. The pineal and plasma melatonin rhythm data for CBA and C57Bl mice were analyzed by ANOVA, and differences between times at night were tested by post hoc Student-Newman-Keuls test. Significance was set at P All studies were conducted according to the guidelines set out by the National Health and Medical Research Council of Australia and were approved by the Animal Ethics Committee of Adelaide University. RESULTS Pineal and plasma melatonin rhythms. Melatonin was detected in the pineal glands of CBA mice at all times from 1900 to 0700, increasing from 0100, peaking at 0500, and decreasing eightfold by 0700 (Fig. 1). The changes in melatonin content with time were significant (ANOVA; F 5.6, P 0.001). Post hoc analysis indicated that the 0500 levels (1, fmol/ gland) were higher than all other times except By contrast, melatonin content of the C57Bl mice never exceeded 55 fmol/gland ( 12 pg/gland) at any time (Fig. 1). Nevertheless, ANOVA indicated that there was a significant effect of time (F 3.5, P 0.01) with the highest melatonin content ( fmol/gland)

3 R360 MELATONIN IN MICE Fig. 1. Pineal gland melatonin content and plasma melatonin concentration in CBA mice (A, C) and C57Bl mice (B, D) throughout the dark period of a 12: 12-h light-dark photoperiod. Data are means SE for 4 to 5 animals at each time point. One CBA mouse and 3 C57Bl mice were excluded from analysis for reasons outlined in the text. Pineal melatonin data are indicated with F and plasma melatonin with E. The relationship between pineal and plasma melatonin in these animals is shown in C and D. Note the scale difference for C57Bl pineal levels in B and D. Where no SE bar is shown, the line is obscured by the symbol. occurring at The lowest value ( fmol/ gland) occurred at 2100 [the values at 0500 were significantly higher than those at 2100 (P 0.04) and 2300 (P 0.047)]. Plasma melatonin levels in CBA mice reflected the patterns of pineal melatonin content with the highest plasma melatonin levels at 0300 and 0500, coinciding with the time of highest pineal content (Fig. 1). One plasma sample at 2100 returned an assay value of 200 pm that was more than six standard deviations outside the means of the 1900 and 2300 sample times. This animal had very low pineal melatonin content, and so data from this animal were excluded from further analysis. In C57Bl mice, plasma melatonin was 43 pm throughout the night (Fig. 1), although one sample at 0300 and two samples at 0500 returned assay results above 100 pm. These values were 6 to 12 standard deviations above the mean values for the strain. As in the case of the CBA plasma sample, unfortunately, there was insufficient sample available for a repeat analysis. The three C57Bl mice with elevated plasma melatonin had undetectable pineal melatonin content. Correlation analysis in the two strains indicated a significant relationship between pineal and plasma melatonin in CBA but not C57Bl mice. Isoproterenol. Administration of isoproterenol (20 mg/kg) at 0900 and again at 1100 resulted in a robust and prolonged increase in pineal melatonin content and plasma melatonin levels 30 and 60 min after the last injection in CBA mice (Fig. 2). Saline administration to CBA mice caused no changes in pineal or plasma melatonin levels (pineal melatonin fmol/gland; plasma melatonin 30 pm). When the response of CBA and C57Bl mice to isoproterenol was compared 30 min after the last injection, the pineal melatonin levels were and 21 fmol/gland, respectively; plasma melatonin values were 63 5 and 30 pm, respectively. No obvious acute ill effects were observed in the mice following isoproterenol administration. Fig. 2. Effect of isoproterenol administration (20 mg/kg) and saline vehicle on pineal and plasma melatonin in CBA mice. Data are means SE (n 5) for pineal melatonin (fmol/gland; isoproterenol, F; saline, ) and for plasma melatonin (pm; isoproterenol, E; saline, ). 2Time of administration of isoproterenol with 0 being 2 h after lights on. Where no SE bar is shown, the line is obscured by the symbol.

4 MELATONIN IN MICE R361 Light effects. Exposure to light 2 h before dawn (0500) resulted in the suppression of pineal melatonin content and plasma melatonin levels to basal levels within 45 min of the start of light exposure (Fig. 3). These levels were indistinguishable from the usual 0700 values. When mice were exposed to light at 2300 or 0500, kept in darkness, and then killed after 30 or 24 h, around the time of the expected melatonin peak (0500), there were significant effects on the timing of the morning decline. The 2300 pulse resulted in pineal melatonin content and plasma melatonin levels at 0600 that were significantly higher than in mice that had not been exposed to light (Fig. 4, A and B). By 0700, there was no difference between the two groups. Animals exposed to light at 0500 had undetectable plasma melatonin levels at 0500 and 0600 and very low pineal melatonin content at these times compared with the animals that remained in darkness throughout (Fig. 4, C and D). When the 6-sulphatoxymelatonin RIA was applied to mouse urine, it was immediately evident that the excretion rate was very low ( 2 pmol/h), demanding assay volumes up to 20 l and thus increasing the possibility of nonspecific immunoreactivity being detected. Analysis of the 2-h urine collections revealed a small but significant increase in immunoreactivity in urine collected between 0600 and 1000 in CBA mice (Fig. 5). This rise was not evident when the lights were left on throughout the night. By contrast, there was somewhat less immunoreactivity in the C57Bl mice and no evidence of a rise toward dawn. No attempt was made to confirm the identity of the immunoreactivity by chromatography because the levels were too low. Administration of 100 g of melatonin to mice resulted in an increase in 6-sulphatoxymelatonin-like immunoreactivity and melatonin immunoreactivity peaking at 4 h and decreasing slowly thereafter to be undetectable by 14 h postinjection (Fig. 6). When the total amounts of 6-sulphatoxymelatonin and melatonin recovered in urine were calculated, they represented 1.7 and 0.4% of the administered melatonin dose, respectively. The chromatographic identity of the immunoreactivity was not investigated. When large amounts of melatonin (4 mg/mouse and 40 mg/rat) were administered, urine analyzed by thinlayer chromatography, and plates sprayed with Ehrlich s reagent, a single strong indole-positive blue spot was observed in mouse urine and one major and two minor indole-positive blue spots were observed in rat urine (Fig. 7). The CBA, C57Bl, and wild derived mice had identical metabolite profiles. The indole in the mouse urine had an R f consistent with the synthetic 6-glucuronylmelatonin (R f 0.15). The position of the major indole metabolite in the rat urine was consistent with 6-sulphatoxymelatonin (R f 0.32), whereas the minor spots were consistent with 6-glucuronylmelatonin (R f 0.15) and had an unidentified compound slightly more polar than 6-glucuronylmelatonin (R f 0.18). Neither melatonin nor 6-hydroxymelatonin was evident in urine collected from mice and rats following melatonin administration. DISCUSSION Fig. 3. Pineal melatonin content and plasma melatonin concentration in CBA mice 2 h before lights on at 0700 and the response to a 15-min light pulse at The figure shows the results for untreated animals ( ; means SE; n 5 10). Data from 5 animals that were light pulsed for 15 min at 0500 and killed at 0530 are shown with F. Where no SE bar is shown, the line is obscured by the symbol. The CBA mouse strain has a high-amplitude pineal melatonin rhythm with a peak 2 h before lights on (13, 33). The decrease during the 2 h before dawn is profound, with near basal levels measured at zeitgeber time 0 (ZT0; ZT12 is defined as lights off), and it is highly repeatable, indicating that it is likely to be under clock control. This is to our knowledge the first report of plasma melatonin measurement in CBA mice, and we have shown that the pineal melatonin content is highly correlated with plasma melatonin levels. Masana et al. (25) reported pineal and plasma melatonin rhythms in the related C3H strain and found that the highest plasma levels coincided with the lowest pineal melatonin content 2 h after dark onset. This relationship was not observed in animals kept under constant darkness conditions. It remains to be confirmed whether this finding is a special feature of the C3H strain. We also confirmed the greatly impaired melatonin synthesis in the commonly used C57Bl

5 R362 MELATONIN IN MICE Fig. 4. Effect of a light pulse 4 h after lights off [zeitgeber time 16 (ZT16); A, B] or 2 h before dawn (ZT22; C, D) on pineal melatonin content (fmol/gland) and plasma melatonin concentration (pm) around the time of the expected melatonin peak the next subjective morning. Note that the mice were kept in darkness from the time of the pulses until they were killed. Data are means SE (n 5). The untreated animals are indicated with F and pulsed animals with. *Pulsed and unpulsed groups were significantly different (P 0.05) at that time of day. Where no SE bar is shown, the line is obscured by the symbol. Fig. 5. Urinary 6-sulphatoxymelatonin (amt.6s) excretion rate in CBA and C57Bl mice in a 12:12-h light-dark photoperiod and over 1 night of constant light. Data are means SE (n 5) with CBA mice shown as and C57Bl mice as. The dark period is indicated by black horizontal bars at top. strain. The slight, but statistically significant, elevation in pineal melatonin content (but not plasma melatonin) in C57Bl mice toward light onset is difficult to explain considering that the mutation in NAT responsible for the low melatonin production has been shown to be due to a truncated nonfunctional acetyltransferase (30). Little is known about the nature of the HIOMT defect in mouse strains and whether the mutant enzyme retains some enzyme activity. It is possible that small amounts of N-acetylserotonin produced by other transferases are methylated by residual pineal HIOMT enzyme. The rhythm could then come from a low-amplitude rhythm in pineal HIOMT activity similar to the rhythm reported in rats (29). It should be pointed out that very low levels of pineal melatonin synthesis have also been reported elsewhere in C57Bl mice (33, 35). Exposure to bright light at the time of maximal pineal melatonin content and plasma melatonin levels resulted in rapid suppression of melatonin production within 30 min. This is consistent with a previous report (12) that showed that pineal melatonin content in CBA mice decreased to basal levels after 5-min exposure to 100-lx light. The rapid decrease in plasma melatonin levels following light exposure as opposed to the gradual morning decline between 0500 and 0700 suggests that the decline may be a clock-controlled event. Although complete profiles were not obtained due to the large numbers of animals required, a light pulse in the early dark period clearly resulted in a small but significant delay of the morning decline in melatonin production and secretion, whereas a pulse 2 h before lights on resulted in what we interpret as an advance of more than 2 h. These results are consistent with light-induced phase shifts in activity rhythms induced by light pulses in this strain (Kennaway et al., unpublished results) and other strains (3 5), although the

6 MELATONIN IN MICE R363 Fig. 6. Excretion of immunoreactive melatonin and immunoreactive 6-sulphatoxymelatonin following the administration of melatonin at Data are means SE (pmol/2 h; n 5). The experiment was conducted in constant light. magnitude of the shifts is smaller. This may be due to the use of the Aschoff Type II protocol (2) rather than the application of pulses after a long period of constant darkness. The apparently large phase advance in the melatonin rhythm within one cycle following light exposure at 0500 was a little unexpected because in rats, it can take up to 4 days under similar circumstances for advances of the pineal NAT rhythm to become apparent (14). A possible explanation for this difference is the species difference in the endogenous period, which for rats free running in constant darkness is greater than 24 h and for mice less than 24 h [CBA mice in our laboratory have a free-running rhythm of wheel running of 23.4 h (Kennaway et al., unpublished results)]. Thus, in rats, morning light pulses before producing a net advance must first counteract the natural tendency to delay, whereas in mice, morning pulses assist the natural tendency for rhythms to advance. Recent in vitro studies by von Gall et al. (35) highlighted the similarities between rats and mice in the control systems responsible for the induction of NAT and synthesis of melatonin. We have shown that in vivo stimulation of the mouse pineal gland by the -adrenergic agonist isoproterenol during the light period increased pineal melatonin content and melatonin secretion into blood in CBA but not C57Bl mice. In preliminary studies, we observed only small transient increases in melatonin production following a single injection of isoproterenol. By contrast, two injections, 2 h apart in the morning, provided reliable induction of melatonin synthesis, which is reminiscent of the requirements of Syrian hamsters for prolonged adrenergic stimulation to increase melatonin synthesis (31). Interestingly, both mice and hamsters have maximum melatonin production late in the dark period. The failure to detect increased melatonin production or synthesis in C57Bl mice in response to isoproterenol administration supports the hypothesis that the small amounts of melatonin detected in the pineals of some animals at night may be synthesized from N-acetylserotonin produced by a transferase enzyme outside the pineal gland that is not induced by -adrenergic stimulation. Melatonin produced by the pineal gland of mammals appears in the blood and is metabolized by the liver to 6-hydroxymelatonin, conjugated, and excreted into the urine (20). The nature of the conjugation products of melatonin was reported over 40 yr ago in rats to be 6-sulphatoxymelatonin and 6-glucuronylmelatonin (20), and this was subsequently confirmed in humans (9, 22, 23, 37). With the development of a RIA for 6-sulphatoxymelatonin (1), it has been possible to monitor the excretion of 6-sulphatoxymelatonin into the urine of humans (10), rats (17), Djungarian hamsters (21), minks (26), and pigs (19). In all cases, the excretion rate of 6-sulphatoxymelatonin is rhythmic and can be used as an index of pineal gland rhythmicity (10). There is a clear advantage in such a procedure due to the noninvasive nature and simplicity of urine collection that reduces the number of animals needed for Fig. 7. Thin-layer chromatogram of urine from CBA, BALB/c, and wild mice and a rat following administration of 4 or 40 mg (rat) melatonin. The solvent mixture used to develop the chromatogram was butanol:acetic acid:water (4:1:1). The figure shows the Ehrlich s reagent-positive spots for the metabolites and standards.

7 R364 MELATONIN IN MICE experiments. It was thus expected that mice would produce and excrete 6-sulphatoxymelatonin and that RIA procedures might be used to follow pineal rhythms in this species. It should, however, be pointed out that while Kopin et al. (20) used both rats and mice in their influential study, the nature of the melatonin metabolites in their mouse studies was not reported. We have provided unequivocal evidence that mice do not excrete significant amounts of 6-sulphatoxymelatonin, presumably due to a lack of an appropriate sulphotransferase. Instead, melatonin is excreted as 6-glucuronylmelatonin with only very small amounts excreted unmetabolized. Our method of detection of metabolites involved both RIA and thin-layer chromatography. In the case of the RIA, the antibody has 0.5% cross-reactivity with 6-glucuronylmelatonin, and this may be the cause of the increase in 6-sulphatoxymelatonin immunoreactivity observed in CBA mice between 0600 and 1000 and following melatonin administration. In the case of the thin-layer chromatography identification of metabolites, because Ehrlich s reagent only detects intact indole rings, we can neither exclude nor quantitate other nonindolic melatonin metabolites in the mouse. It is also important to point out that the inability to produce 6-sulphatoxymelatonin is not confined to laboratory mouse strains because the wild mouse strain also excreted only 6-glucuronylmelatonin. The very low levels of unmetabolized melatonin excreted are in contradiction to a study by Perissin et al. (28), who not only reported melatonin in mouse urine but also found a more than fivefold increase in unmetabolized melatonin in urine at night. These researchers used a strain of mouse (BD2F1) that is a cross of two strains (DBA2 and C57Bl) both known to be melatonin deficient (13). A more recent study describing atypical plasma and pineal melatonin rhythms in this strain is remarkable for the extremely low levels of melatonin measured (24). We cannot provide an explanation for this discrepancy other than the possibility that the assay used was not specific for melatonin. Similar conclusions may be warranted in the case of the two recent publications that have challenged the idea that BALB/c, C57Bl, and AKR mice are melatonin deficient. In the first study, in each strain, melatonin was reported to be elevated for 15 min around ZT18;at all other times, the pineal melatonin content was very low (6). In the second study, melatonin was detected in BALB/c, C57Bl/6, and OF1 Swiss mice pineal glands, but the levels were very low and the pattern was very different from that observed for CBA and C3H mice (6). Perspectives The results of this study indicate that the CBA mouse strain has a robust melatonin rhythm that is under -adrenergic control and that melatonin production can be suppressed by brief light pulses. There is also compelling preliminary evidence that light pulses presented at different times of the night phase shift melatonin rhythms. The need to kill animals for pineals or blood may restrict the usefulness of mice in studies of melatonin rhythmicity, although the knowledge that 6-glucuronylmelatonin is the major urinary melatonin metabolite provides an incentive to develop assays for this compound. Apparently, all knockout and transgenic mouse lines are on genetic backgrounds that make the animals melatonin deficient (27). This is not acceptable for researchers studying circadian rhythms or reproduction. If strains of interest (i.e., knockout) are crossed with CBA mice, an isoproterenol stimulation test based on results presented in this study would provide a simple method of monitoring the inheritance of melatonin production capacity at least until the mutations in NAT and HIOMT are defined and PCR-based genotyping strategies are developed. The authors thank S. Palnok for assistance in the early stages of the study. This study was supported by a grant to D. J. Kennaway from the National Health and Medical Research Council of Australia. REFERENCES 1. Aldhous ME and Arendt J. Radioimmunoassay for 6-sulphatoxymelatonin in urine using an iodinated tracer. Ann Clin Biochem 25: , Aschoff J. Response curves in circadian periodicity. In: Circadian Clocks, edited by Aschoff J. Amsterdam: North Holland, 1965, p Benloucif S and Dubocovich ML. Melatonin and light induce phase shifts of circadian activity rhythms in the C3H/HeN mouse. J Biol Rhythms 11: , Bradbury MJ, Dement WC, and Edgar DM. Serotonin-containing fibers in the suprachiasmatic hypothalamus attenuate light-induced phase delays in mice. Brain Res 768: , Colwell CS and Foster RG. Photic regulation of Fos-like immunoreactivity in the suprachiasmatic nucleus of the mouse. J Comp Neurol 324: , Conti A and Maestroni GJ. HPLC validation of a circadian melatonin rhythm in the pineal gland of inbred mice. J Pineal Res 20: , Ebihara S, Hudson DJ, Marks T, and Menaker M. Pineal indole metabolism in the mouse. Brain Res 416: , Ebihara S, Marks T, Hudson DJ, and Menaker M. Genetic control of melatonin synthesis in the pineal gland of the mouse. Science 231: , Fellenberg AJ, Phillipou G, and Seamark RF. Specific quantitation of urinary 6-hydroxymelatonin sulphate by gas chromatography mass spectrometry. Biomed Mass Spectrom 7: 84 87, Fellenberg AJ, Phillipou G, and Seamark RF. Urinary 6-sulphatoxy melatonin excretion during the human menstural cycle. Clin Endocrinol (Oxf) 17: 71 75, Francis PL, Leone AM, Young IM, Stovell P, and Silman RE. Gas chromatographic-mass spectrometric assay for 6-hydroxymelatonin sulfate and 6-hydroxymelatonin glucuronide in urine. Clin Chem 33: , Goto M and Ebihara S. The influence of different light intensities on pineal melatonin content in the retinal degenerate C3H mouse and the normal CBA mouse. Neurosci Lett 108: , Goto M, Oshima I, Tomita T, and Ebihara S. Melatonin content of the pineal gland in different mouse strains. J Pineal Res 7: , Illnerova H, Vanecek J, and Hoffmann K. Different mechanisms of phase delays and phase advances of the circadian rhythm in rat pineal N-acetyltransferase activity. J Biol Rhythms 4: , 1989.

8 MELATONIN IN MICE R Kennaway DJ. Radioimmunoassay of 5-methoxy tryptophol in sheep plasma and pineal glands. Life Sci 32: , Kennaway DJ. Urinary 6-sulphatoxymelatonin excretory rhythms in laboratory rats: effects of photoperiod and light. Brain Res 603: , Kennaway DJ, Blake P, and Webb HA. A melatonin agonist and N-acetyl-N2-formyl-5-methoxykynurenamine accelerate the re-entrainment of the melatonin rhythm following a phase advance of the light-dark cycle. Brain Res 495: , Kennaway DJ, Gilmore TA, and Seamark RF. Effect of melatonin feeding on serum prolactin and gonadotropin levels and the onset of seasonal estrous cyclicity in sheep. Endocrinology 110: , Klante G, Brinschwitz T, Secci K, Wollnik F, and Steinlechner S. Creatinine is an appropriate reference for urinary sulphatoxymelatonin of laboratory animals and humans. J Pineal Res 23: , Kopin IJ, Pare CMB, Axelrod J, and Weissbach H. The fate of melatonin in animals. J Biol Chem 236: , Korenman EM, Watson BW, and Silman RE. The pattern of motor activity, light, and melatonin production in Syrian hamsters. J Pineal Res 4: , Leone AM, Francis PL, and Silman RE. The isolation, purification, and characterisation of the principal urinary metabolites of melatonin. J Pineal Res 4: , Leone RM and Silman RE. An investigation of demethylation in the metabolism of methoxytryptamine and methoxytryptophol. J Pineal Res 2: 87 94, Li XM, Liu XH, Filipski E, Metzger G, Delagrange P, Jeanniot JP, and Levi F. Relationship of atypical melatonin rhythm with two circadian clock outputs in B6D2F(1) mice. Am J Physiol Regulatory Integrative Comp Physiol 278: R924 R930, Masana MI, Benloucif S, and Dubocovich ML. Circadian rhythm of mt(1) melatonin receptor expression in the suprachiasmatic nucleus of the C3H/HeN mouse. J Pineal Res 28: , Maurel D, Mas N, Roch G, Boissin J, and Arendt J. Diurnal variations of urinary 6-sulphatoxymelatonin in male intact or ganglionectomized mink. J Pineal Res 13: , McWhir J, Schnieke AE, Ansell R, Wallace H, Colman A, Scott AR, and Kind AJ. Selective ablation of differentiated cells permits isolation of embryonic stem cell lines from murine embryos with a non-permissive genetic background. Nat Genet 14: , Perissin L, Zorzet S, Rapozzi V, and Giraldi T. A noninvasive simple method for measurement of urinary excretion of melatonin in undisturbed mice. J Pineal Res 15: , Ribelayga C, Garidou ML, Malan A, Gauer F, Calgari C, Pevet P, and Simonneaux V. Photoperiodic control of the rat pineal arylalkylamine-n-acetyltransferase and hydroxyindole- O-methyltransferase gene expression and its effect on melatonin synthesis. J Biol Rhythms 14: , Roseboom PH, Namboodiri MA, Zimonjic DB, Popescu NC, Rodriguez IR, Gastel JA, and Klein DC. Natural melatonin knockdown in C57BL/6J mice: rare mechanism truncates serotonin N-acetyltransferase. Mol Brain Res 63: , Santana C, Menendez-Pelaez A, Reiter RJ, and Guerrero JM. Treatment with forskolin for 8 hours during the day increases melatonin synthesis in the Syrian hamster pineal gland in organ culture: the long lag period is required for RNA synthesis. J Neurosci Res 25: , Urosevic N, Silvia OJ, Sangster MY, Mansfield JP, Hodgetts SI, and Shellam GR. Development and characterization of new flavivirus-resistant mouse strains bearing Flv(r)- like and Flv(mr) alleles from wild or wild-derived mice. J Gen Virol 80: , Vivien-Roels B, Malan A, Rettori MC, Delagrange P, Jeanniot JP, and Pevet P. Daily variations in pineal melatonin concentrations in inbred and outbred mice. J Biol Rhythms 13: , Vollrath L, Huesgen A, Manz B, and Pollow K. Day/night serotonin levels in the pineal gland of male BALB/c mice with melatonin deficiency. Acta Endocrinol 117: 93 98, Von Gall C, Lewy A, Schomerus C, Vivien-Roels B, Pevet P, Korf HW, and Stehle JH. Transcription factor dynamics and neuroendocrine signalling in the mouse pineal gland: a comparative analysis of melatonin-deficient C57BL mice and melatonin-proficient C3H mice. Eur J Neurosci 12: , Werschkun B, Wendt A, and Thiem J. Enzymic synthesis of -glucuronides of estradiol, ethynylestradiol and other phenolic substrates on a preparative scale employing UDP-glucuronyl transferase. J Chem Soc, Perkin Trans 1: , Young IM, Leone RM, Francis P, Stovell P, and Silman RE. Melatonin is metabolized to N-acetyl serotonin and 6-hydroxymelatonin in man. J Clin Endocrinol Metab 60: , 1985.