Time Course of Sensitivity of Golden Hamsters to Melatonin Injections Throughout the Day1

Transcription

1 BIOLOGY OF REPRODUCTION 29, (1983) Time Course of Sensitivity of Golden Hamsters to Melatonin Injections Throughout the Day1 MILTON H. STETSON2 and DORANNE E. TAY Physiology Section School of Life and Health Sciences University of Delaware Newark, Delaware ABSTRACT Twenty-four groups of adult male golden hamsters, housed on 14 h of light, 14L: 1OD, received daily injections of melatonin, each at a different hour of the day. Injections (15 ig melatonin/0. 1 ml ethanol:saline, 1:10, s.c.) persisted for 27 weeks. At the beginning of this investigation and every 2-4 weeks thereafter, laparotomies were performed on a few animals of each group and the testes measured to assess the efficacy of the melatonin injections in causing testicular regression. Two periods of melatonin sensitivity were identified. The first was brief, of 1-h duration, immediately prior to lights on ( h). The second period extended for 6 h throughout late afternoon and early evening ( h; lights out 2000 h). Regression was complete in all these groups except that injected at 1500 h in which the testes regressed to only 950 mg before recrudescence occurred. In all other affected groups, testes regressed to <400 mg. In those groups where melatonin did not cause full testicular regression, daily injections resulted in partial regression, usually to approximately 2000 mg prior to regrowth. INTRODUCTION Animals whose reproductive activities are firmly fixed to a particular season of the year require a system whereby environmental information can be received, integrated, and transduced into neural, neuroendocrine, and endocrine signals directly involved with controlling reproduction. The golden hamster, Mesocricetus auratus, is one of these animals, reproduction being normally confined to the spring and summer months (see Stetson and Tate-Ostroff, 1981, for review). This small rodent has been the subject of hundreds of investigations keyed to deciphering the means by which photoperiod, the environmental parameter read by this species, controls annual reproductive activity. A discovery of primary importance (Hoffman and Reiter, 1965a,b, 1966) implicated the pineal gland as a major component of the photoperiodic control system. Pinealectomy was shown to render the hamster incapable of responding to photo- Accepted April 21, Received March 17, Supported by NSF Research Grant PCM Reprint requests. period. Ablation and transplantation experiments, which had served endocrinologists so well since the turn of the century, were without success in pineal research (Reiter, 1967), suggesting that the gland a) did not function autonomously, and b) required a direct signal, endocrine or nervous, for normal activity. The latter proved to be the case. Postganglionic fibers from the superior cervical ganglia (SCG) were identified as the primary source of pineal innervation (Ariens Kappers, 1965). Removal of the SCG (Reiter and Hester, 1966; Reiter, 1967, 1968) or severing of its afferent or efferent projections was as equally effective as pinealectomy in preventing photoperiodically induced changes in gonada! activity in hamsters. The pineal was shown to produce the hydroxyindole melatonin (Wurtman et a!., 1963, 1964). We now recognize this substance as being the primary product of the mammalian pineal, produced during the hours of darkness, with production regulated by sympathetic input through the SCG (see e.g., Binkley et a!., 1973). Recognizing the importance of rhythmic (phasic) versus tonic melatonin production, Goldman and his colleagues embarked on a lengthy study of the effects of timed melatonin administration on reproductive activity in 432

2 DAILY SENSITIVITY TO MELATONIN 433 Chamber l0 II $2 Injection groups 1500, 160C $300, I $00,,, 0900, IOOC 0700, , ,0400 0l00,020C 2300, , ,2000 $700, 800 AA I I I I I I - I TIME FIG. 1. Photoperiods in the 12 chambers used in this study. White bars represent 14 h of light; stippled bars, 10 h darkness. Each chamber contained 2 groups of animals; one of which was injected at 1500 h and the other at 1600 h solar time (arrows and dotted lines). In this way, a group of animals was injected each hour of the 24-h day. hamsters. Their initial discovery was of primary importance; gonadal activity in golden hamsters can be interrupted by melatonin injections given only at specific times of the 24-h day (Tamarkin et a!., 1976, 1977a). In a follow-up report they showed that this effect occurred only in the presence of the pinea! (Tamarkin et a!., 1977b); in its absence hamsters failed to respond to a single, properly timed, daily injection of melatonin, but were shown to retain sensitivity to this hormone if it were administered in a thrice-daily injection paradigm, injections timed 3 h apart. Interestingly, the pinealectomized hamster appears to respond to melatonin at any time during the 24-h day, provided the 3 injections/6 h paradigm is maintained (Tamarkin et al., 1977b; Goldman et a!., 1979). We (Watson-Whitmyre and Stetson, 1983) have recently shown, however, that the pinealectomized hamster retains the ability to respond as an intact animal to melatonin injection in the evening (1900 h) but not morning (1200 h) provided the endogenous peak release of pinea! melatonin is simulated by a properly timed injection of exogenous hormone (0400 h; lights h). Thus the hamster s response to exogenous me!atonin is dependent on the time of administration of the hormone. A careful review of the literature reveals that in hamsters exposed to long days, only two periods of sensitivity to exogenous melatonin are recognized: around the time of lights out (a relatively broad period of 2-3 h, depending on the individual investigation) and at 0400 h on a photoperiod of 14L: 1OD (lights h; Tamarkin et al., 1977a). The purpose of this investigation is to refine our knowledge of the precise periods of sensitivity to exogenous melatonin in male hamsters held on a 14L:1OD photoperiod. This was accomplished by administering melatonin to 24

3 434 STETSON AND TAY 3500 a E I w U) IaJ C ) Ui TIME of INJECTION FIG. 2. Initial (dotted line), maximum (square) and minimum (circle) testes weight of each injection group throughout the 27 weeks of injection. Periods of melatonin sensitivity at h and h are clearly visible. Note that all groups displayed an initial decline in testes weight to approximately 2000 mg. Fifteen groups then recovered while the 9 indicated above continued to regress. Photoperiod is indicated by a solid line, darkness by stippled bars. Each point represents the mean of 3-8 individuals SEMs omitted for ease of data presentation. Those for affected groups provided in Table 1. Those of other groups were of similar magnitude (10-15% of the mean) as those depicted under the maximum heading in Table 1. groups of hamsters, 1 group each hour of the day. MATERIALS AND METHODS The adult male hamsters used in this investigation were raised from birth in our laboratory on a photoperiod of 14 h of light, 10 h of darkness (14L:1OD, lights on h) with free access to food and water. At 8 weeks of age the animals were divided at random into 24 groups of 16 males each. Two groups of males were transferred into each of 12 environmental chambers. The photoperiod in each chamber was 14L:1OD, but was programmed so that lights on in a chamber occurred 2 h later than in the preceding chamber (Fig. 1). That is, lights on in Chamber 1 at 0600 h, iii Chamber 2 at 0800 h, in Chamber 3 at l000h...inchamberl2ato4ooh(fig. 1).Theanimals were allowed 8 weeks to entrain to the particular 14L:1OD photocycle in their chamber. At this point, when the animals were 16 weeks of age, the experiment began. At 1500 or 1600 h EDT each day (Fig. 1) 1 group of animals in each chamber was injected subcutaneously with 15 tg melatonin (Sigma) in 0.1 ml ethanol:saline (1:10). Injections continued for 27 weeks. With this injection paradigm, I group of animals was injected each hour of the day (see Fig. 1). Every 2-4 weeks during the injection period a hemilaparotomy was performed on half of the animals of each group to assess the effect of daily melatonin injections on testicular size. The animals were anesthetized with ether. A amall lateral incision in the left or right scrotal sac exposed the corresponding testis whose length and width were then measured to the nearest 0.1 mm with a vernier calipers. The incision was closed with a steel suture and the animal returned to his cage. Attempts were made to measure a similar number of right and left testes in each group. Testicular dimensions were used to calculate testicular volume using the formula for the volume of a prolate,heroid (Kenagy, 1979; Watson-Whitmyre and Stetson, 1983). Testicular volume was transformed to paired testes weight using a regression formula based on volume-weight transformations of over a hundred testes (r=0.97). RESULTS In Fig. 2 we have plotted the initial, minimum, and maximum testes weight for each of the 24 injection groups during the 27 weeks of this study. Melatonin effectively caused a decline in testes weight in every group. For 15 of the 24 groups the decline in testes weight

4 DAILY SENSITIVITY TO MELATONIN 435 was of a minor nature, with minimal weight approximately 2000 mg (Fig. 2). In the remaining 11 groups, testicular regression was complete (10 groups), with minima! testes weight less than 400 mg, or nearly complete (1 group, 1500 h, minimal testes weight 950 mg). Maximum testes weight during the 27-week injection period exceeded 2500 mg in all but 1 group (2100 h, Fig. 2). Two periods of sensitivity to exogenous melatonin were identified in this study. The first is a brief period of 1-h duration at the end of the scotophase ( h, Fig. 2). Animals in the 0600 h group were always injected in the dark. This period of melatonin sensitivity was very well defined; in animals in the 0400 and 0700 h injection groups testicular regression was minimal, whereas it was complete in the 0500 and 0600 h groups (Fig. 2). The second period of sensitivity to exogenous melatonin extended from 1500 h to 2100 h (late afternoon-early evening, Fig. 2). Testicular regression was complete in all of these but the 1500 h group (Fig. 2). The 27-week injection paradigm employed in this study was of sufficient duration to allow us to examine the kinetics of melatonininduced testicular regression and testicular recrudescence. In Fig. 3 we have plotted the testicular maxima and minima throughout the study without regard to injection group. The testicular regression curve (weekly minimum) bears a marked similarity to the changes in testicular weight induced by exposure to short day lengths (Fig. 3). Full testicular regression persisted from Week 6 to Week 20, after which testicular recrudescence commenced (Fig. 3). We have included in Fig. 3 the complete testes weight data from those 2 groups which responded with full testicular regression most rapidly (1700 h injection group) and most slowly (0500 h injection group). A comparison of the 2 groups shows that full testicular regression was attained by the 1700 h group after only 6 weeks of injections, persisted to Week 12, at which point testicular recrudescence commenced, and was completed by the 24th week of injections (Fig. 3). In the slowest responding group (0500 h), testicular regression was not complete until Week 10 of injections (Fig. 3). Testes in these animals remained regressed until Week 20, at which point recrudescence commenced. At the termination of this study, after 27 weeks of daily melatonin injections, testicular recrudescence in the slow- E I w C/) Ui U) Ui 400C /. / *,,. #{149}.*s 0- - S., I WEEKS of INJECTION FIG. 3. Testes weight of the fastest (open squares; 1700 h) and slowest (open circles; 0500 h) responding injection groups as we!! as the groups with maximum (solid squares) and minimum (solid circles) testes weights each sample time throughout the 27 weeks of melatonin injections. Each symbol represents the mean testes weight of 3-8 individuals. SEMs omitted for ease of presentation (see legend of Fig. 2). est responding group was still in progress (Fig. 3). Thus, full regression and recrudescence were separated by a month or more between the 2 groups. Testes weight kinetics for all fully affected groups are presented in Table 1. Several salient facts should be noted. Regression and recrudescence proceeded more slowly in animals of the late evening injection groups (0500, 0600 h), and losses were greater, with fewer animals surviving the 27 weeks of this study. Animals in 3 of the 6 late afternoon and early evening injection groups completed a full cycle of testicular regression and recrudescence in 24 weeks (Table 1). The testes of animals in the remaining 3 groups had not fully recrudesced (i.e., testes weight <3000 mg) by Week 27 when the study was terminated. DISCUSSION Our results clearly define the times during a 14-h photoperiod when male golden hamsters are sensitive to exogenous me!atonin injections. These are from 5 h prior to 1 h after lights out, and for 1 h prior to lights on (Fig. 2). We must stress that these data were obtained from ham-

5 436 STETSON AND TAY sters housed on 14L:1OD only and that in animals held on other long days, the periods of sensitivity will probably vary in duration. A!- though Tamarkin et a!. (1977a) found male hamsters sensitive to melatonin injected at 0400 h, we found our hamsters exceptionally insensitive to the exogenous hormone at this time. In other respects our data are similar. A possible reason for this apparent difference in response may be differences in entrainment of the two populations of hamsters to the 14L: 1OD photoperiod used in each study. Although lights on and off were at identical hours in each study, slightly different entrainment patterns of the animals circadian system to the photoperiod could result in major differences in response to exogenous melatonin. The endogenous rhythm of pineal melatonin is timed by a circadian oscillator (Tamarkin et a!., 1980). The rhythm in our animals appears phase locked with the activity rhythm, peak pineal melatonin occurring 6-8 h after activity onset (Stetson et a!., unpublished; Rollag and Stetson, 1981). Similarly, we have shown that the hamster s period of sensitivity to exogenous melatonin is also timed by a circadian oscillator (Rollag and Stetson, 1982) and not to the onset or offset of the lights per Se. An interesting phenomenon was observed in every group of animals receiving melatonin. Periodic!aparotomies revealed that all animals, irrespective of injection group, responded transiently to melatonin with a decline in testicular size. In most animals this response was brief; the testes regressed to only twothirds of their normal weight before recrudescence ensued (Fig. 2). These partial responses may represent a transient effect of me!atonin directly on the testes, independent of the hypothalamo-hypophysial axis, or transient changes in pituitary gonadotropin release; full testicular regression induced by properly timed melatonin injections occurs because of a persistent reduction in gonadotropin release (Tate- Ostroff and Stetson, 1981). In those animals responding to melatonin with full testicular regression (Fig. 2), the rate of regression, the time full regression was maintained, and the rate of recrudescence varied from group to group (Table 1; Fig. 3). In general, those groups on the borders of the sensitivity windows (0500, 1500, 2100 h) responded more slowly to melatonin injections; full testicular regression was not achieved until the 10th week of injections (the 1500 h injection group never fully regressed; Fig. 2). Testicular recrudescence in these groups was incomplete at the 27th week of injections when the experiment was terminated (Fig. 3; Table 1). These variatons in the rate and duration of testicular regression and recrudescence are, we feel, directly attributable to the timing of melatonm administration in each group and not to individual variability among the animals. All animals in this investigation were from the same colony of selectively bred animals and of the same age (born within 4 days of one another). Instead, we feel that these slower responses are due to the fact that melatonin was administered to these groups at a time TABLE 1. Time course of testicular regression and recrudescence in groups which showed maximum responses to melatonin. Injec- Full Recrution regression Testes weights at descence Maximum testes Recrugroup period full regressiona period recrudescencea descence (h) (wks) (n) (wks) (n) complete?b ± ± 113 (4-8) ± 192(4) No ± ± 70(6-8) ± 154(3) No ± ± 284 (8) ± 153 (8) Yes ± 152 (6-8) ± 359(6) Yes ± ± 116 (6-7) ± 150(7) No ± 150 (6-8) ± 719 (6) Yes ± 42 (5-8) ± 206 (8) No ± ± 188 (6-8) ± 212 (8) No amean ± SEM in mg. Number of animals in group(s) in parentheses. bfuli testicular recrudescence defined as testes weight greater than 3000 mg.

6 DAILY SENSITIVITY TO MELATONIN 437 when the animals were only partially sensitive, i.e., at the extreme edges of the sensitivity periods, or windows. These windows, therefore, function to define discretely the times, during an 14L:1OD day, when golden hamsters will respond to exogenous melatonin injections. The slower responses may represent partial suppression of the reproductive control system by melatonin. They are not routinely observed in hamsters exposed to short day lengths. Here the response is more of an all or none (regression or no regression) nature (Elliott, 1976), rather than slower regression on longer short days, faster regression on shorter short days, as our melatonin data might infer. In the investigation reported here, not all early morning injections of melatonin effectively caused testicular regression (Fig. 2); injections given at 0500 or 0600 h were effective, injections at 0400 or 0700 h totally ineffective. Similarly, late afternoon injections ( h) caused testicular regression while injections on either side of this window did not. Many investigators have questioned the means by which melatonin injections given at the proper time cause testicular regression. None have answered the question, nor can we. Several hypotheses have been forwarded. One appears especially attractive in describing the response in Siberian hamsters (Pbodopus sungorus sungorus) and deals with the duration of pineal melatonin secretion on different photoperiods (Hoffmann, 1981; Goldman et a!., 1982). Briefly, the hypothesis states that short days (long nights) result in more melatonin released from the pineal over a longer period of time. The animal reads this as a short day and testicular regression ensues. In Phodopus, Goldman s investigations support this hypothesis completely (Goldman et a!., 1981; Yellon et a!., 1982). In this species, the pineal appears to secrete melatonin throughout the dark period, and animals on long days can be made to undergo testicular regression by infusion of melatonin for a prescribed duration at any time of the day. The criterion of importance appears to be the duration of melatonin exposure. It is probable, therefore, that a bolus injection of melatonin given to long-day Siberian hamsters a few hours before lights out will, in effect, extend the period of high serum melatonin levels to the point where the short day duration is reached, resulting in testicular regression. Indeed, daily injections administered in this way cause regression in Pbodopus (Stetson et a!., unpublished). Can the same hypothesis apply to the golden hamster? Data generated to date suggest not. Unlike Phodopus, exposure of golden hamsters to short days (long nights) has little effect on the magnitude and duration of pinea! melatonin release (Rollag et a!., 1980). Also, as we have shown here, exogenous melatonin administered in the dark, before and during endogenous release of melatonin ( h), is without effect on the testes (Fig. 2). Data from our 0500 and 0600 h injection groups may be interpreted as supporting this hypothesis by perhaps extending the period of high serum melatonin into the morning hours. However, injections given at 0700 h and after, which would have the same effect on serum melatonin, had a decidedly different effect on the testicular response (Fig. 2). Thus physically increasing the duration and level of melatonin to which golden hamsters are exposed cannot account for their gonadal response to melatonin injections. The question as to why (and how) exogenous melatonin, administered at a time of day when the hamster never experiences elevated melatonin levels (e.g., h on 14L:1OD, Fig. 2) results in gonada! regression, remains unresolved in Mesocricetus auratus. The solution may lie in the facts that pinea! physiology is under direct control of the animal s circadian system, that late afternoon sensitivity to exogenous melatonin is expressed only when the animal is exposed to a second pulse of melatonin at the proper time (Watson- Whitmyre and Stetson, 1983) and timed by a circadian oscillator (Rollag and Stetson, 1982). Melatonin injections may directly or indirectly affect circadian oscillators involved in timing any number of rhythms regulating the photoperiodic response in golden hamsters. REFERENCES Ariens Kappers, J. (1965). Survey of the innervation of the epiphysis cerebri and the accessory pineal organs of vertebrates. Prog. Brain Res. 10: Binkley, S. A, Klein, D. C. and Weller, J. L (1973). Dark induced increase in pineal serotonin N- acetyltransferase activity: A refractory period. Experientia 29: Efliott, J. A. (1976). Circadian rhythms and photoperiodic time measurement in mammals. Fed. Proc. 3 5: Goldman, B. D., Hall, V., Hollister, C., Roychoudhury, P., Tamarkin, L. and Westrom, W. (1979). Effects of melaronin on the reproductive system in intact and pinealectomized male hamsters

7 438 STETSON AND TAY malntained under various photoperiods. Endocrinology 104: Goldman, B. D., Hall, V., Hollister, C., Reppert, S., Roychoudhury, P., Yellon, S. and Taniarkin, L. (1981). Duirnal changes in pineal melatonin content in four rodent species: Relation to photoperiodism. Biol. Reprod. 24: Goldman, B. D., Carter, D. S., Hall, V. D., Roychoudhury, P. and Yellon, S. M. (1982). Physiology of pineal melatonin in three hamster species. In: Melatonin Rhythm Generating System (D. C. Klein, ed.). S. Karger, Basel, pp Hoffman, R. A. and Reiter, R. J. (1965a). Pineal gland: Influence on gonads of male hamsters. Science 148: Hoffman, R. A. and Reiter, R. J. (1965b). Influence of compensatory mechanisms and the pineal gland on dark-induced gonadal atrophy in male hamsters. Nature 203: Hoffman, R. A. and Reiter, R. J. (1966). Responses of some endocrine organs of female hamsters to pinealectomy and light. Life Sci. 5: Hoffmann, K. (1981). Photoperiodic function of the mammalian pineal organ. In: The Pineal Organ: Photobiology - Biochronometry- Endocrinology (A. Oksche and P. Pevet, eds.). E!sevierlNorth Holland Biomedical Press, Amsterdam, pp Kenagy, C. J. (1979). Rapid and repeatable surgical technique for measurement of testis size in small animals. J. Mammal. 60: Reiter, R. J. (1967). Effect of pineal grafts, pinealectomy and denervation of the pineal gland on the reproductive organs of male hamsters. Neuroendocrinology 2: Reiter, K. J. (1968). Morphological studies on the reproductive organs of blinded male hamsters and the effects of pinealectomy or superior cervical ganglionectomy. Anat. Rec. 160: Reiter, R. J. and Hester, K. J. (1966). Interrelationships of the pineal gland, the superior cervical ganglia and the photoperiod in the regulation of the endocrine systems of hamsters. Endocrinology 79: RoIlag, M. D. and Stetson, M. H. (1981). Ontogeny of the pineal melatonin rhythm in golden hamsters. Biol. Reprod. 24: Rollag, M. D. and Stetson, M. H. (1982). Melatonin injection into Syrian hamsters. Prog Clin. Biol. Res. 92: Rollag, M. D., Panke, E. S. and Reiter, R. J. (1980). Pineal me!atonin content in male hamsters throughout the seasonal reproductive cycle. Proc. Soc. Exp. BioL Med. 165: Stetson, M. H. and Tate-Ostroff, B. (1981). Hormonal regulation of the annual reproductive cycle of golden hamsters. Gen. Comp. Endocrino!. 45: Tamarkin, L., Westrom, W. C., Hamil!, A. I. and Goldman, B. D. (1976). Effect of melatonin on the reproductive systems of male and female Syrian hamsters: A diurnal rhythm in sensitivity to melatonin. Endocrinology 99: Tamarkin, L., Lefebre, N. C., Hollister, C. W. and Goldman, B. D. (1977a). Effect of melatonin administered during the night on reproductive function in the Syrian hamster. Endocrinology 101 : Tamarkin, L, Hol!ister, C. W., Lefebre, N. C. and Goldman, B. D. (1977b). Melatonin induction of gonadal quiescence in pinealectomized Syrian hamsters. Science 198: Taznarkin, L., Reppert, S. M., Klein, D. C., Pratt, B. and Goldman, B. D. (1980). Studies on the daily pattern of pineal melatonin in the Syrian hamster. Endocrinology 107: Tate-Ostroff, B. and Stetson, M. H. (1981). Melatonininduced suppression of gonadotropin titers in male golden hamsters: Effect on gonadal feedback mechanisms. Life Sci. 28: Watson-Whitmyre, M. and Stetson, M. H. (1983). Simulation of peak melatonin release restores sensitivity to evening me!atonin injections in pinealectomized hamsters. Endocrinology 112: Wurtman, R. J., Axelrod, J. and Phillips, L. S. (1963). Melatonin synthesis in the pineal gland: Control by light. Science 142: Wurtman, K. J., Axelrod, J. and Fischer, J. (1964). Melatonin synthesis in the pineal gland: Effect of light mediated by the sympathetic nervous system. Science 143: Yellon, S. M., Tamarkin, L., Pratt, B. L. and Goldman B. D. (1982). Pineal melatonin in the Djungarian hamster. Photoperiodic regulation of a circadian rhythm. Endocrinology 111: