Photoperiodic Responses Differ among Inbred Strains of Golden Hamsters (Mesocricetus auratus)'

Transcription

1 BIOLOGY OF REPRODUCTION 49, (1993) Photoperiodic Responses Differ among Inbred Strains of Golden Hamsters (Mesocricetus auratus)' MARTHA HOTZ VITATERNA 2 and FRED W. TUREK Department of Neurobiology and Physiology, NSF Center for Biological Timing, Northwestern University Evanston, Illinois ABSTRACT Inbred strains of golden hamsters differ in both the free-running period of the circadian rhythm of locomotor activity in constant darkness, and in the phase angle of entrainment of activity to a 14L:10D cycle. To determine whether these differences in circadian entrainment affect photoperiodic time measurement, we measured the critical photoperiod for maintaining testicular function as well as the rate of response for four different inbred strains (MHA/SsLak, LSH/SsLak, BIO 1.5, and BIO 87.20) and an outbred stock (Lak:LVG(SYR)) of golden hamsters. Hamsters of each group were maintained for 12 wk under one of five different LD cycles. Animals of all groups maintained testis size in 14L:10OD and 12.5L:11.5D. Significant strain differences were observed in the critical photoperiod for maintaining testis size after 12 wk; the LSH/SsLak inbred strain showed complete testicular regression during exposure to 12L: 1 2D, while little change was observed in any of the other strains under this photoperiod. Some degree of testicular regression was observed in all groups exposed to 1 1.5L:12.5D, while complete regression was observed in all animals exposed to 6L:18D. The rate of testicular regression differed markedly between the different groups under both 6L:18D and 11.5L:12.5D. The differences in critical photoperiod and rate of testicular regression observed between the various strains could not be correlated with the known strain differences in entrainment or circadian period, indicating that genetic differences in photoperiodic response are not related to the genetic differences in circadian rhythmicity. Furthermore, these results demonstrate that for any given genotype of golden hamster, the rate of testicular regression during exposure to an inhibitory photoperiod varies as a function of the day length, and thus the apparent critical photoperiod may depend upon the length of exposure to a given day length. INTRODUCTION For a variety of animal species, reproduction is timed to an appropriate time of year by responses to the seasonal changes in day length. For many small mammalian species, such as the golden hamster, exposure to short days inhibits, while exposure to long days stimulates, neuroendocrinegonadal activity [1, 2]. Thus, the critical photoperiod is the shortest day length needed to sustain or stimulate reproductive function, i.e., the "shortest long day" [3]. In all mammals that have been examined, photoperiodic time measurement involves the use of a circadian clock that controls a circadian rhythm of responsiveness to light. In studies that have examined the photoperiodic response of animals maintained in a variety of experimental lighting schedules, a consistent finding has been that the light-dark cycle will be interpreted as a "long" day only if light coincides with the animal's "photosensitive phase" during the subjective night [3, 4]. The circadian clock involved in photoperiodic time measurement is thought to regulate most, if not all, circadian rhythms in hamsters, including the rhythm of locomotor activity. Small changes in the period of the entraining light-dark cycle (e.g., from 24.0 to h) or small chemically induced changes in the endogenous pe- Accepted April 13, Received April 13, 'This research was supported by an NSF Science and Technology Center Grant, by NIH grants R-01-HD and P-01-HD to F.W.T. and NIH NRSA 5-F31- MH09278 to M.H.V 2Correspondence: FAX: (708) riod of the clock are associated with dramatic changes in the phase relationship between the activity rhythm and the light-dark cycle, which in turn are associated with alterations in the photoperiodic response [5, 6]. The circadian rhythm of sensitivity to light that is involved in the photoperiodic response appears to be the rhythm of pineal melatonin release [7, 8]. Previous studies in our laboratory have demonstrated that genetic differences exist between inbred strains of golden hamster in their circadian rhythms of locomotor activity [9]. Clear differences among four inbred strains of hamsters have been found in the free-running periods of animals kept in constant darkness. The strain order from shortest to longest was BIO 1.5, MHA/SsLak and LSH/SsLak, BIO A similar strain ordering from early to late activity onset times was observed in the activity rhythms of animals entrained to a 14L:10D cycle, indicating that a common set of genetic differences between strains influences both period and phase angle of entrainment. To determine whether the genetic differences in period and phase angle of entrainment observed between strains might also result in genetic differences in their reproductive responses to day length, the reproductive responses to five different LD cycles were determined for four inbred strains and an outbred stock of golden hamsters. Because testis size is strongly correlated with reproductive function in hamsters [1] and testis size is amenable to repeated measurement without stressful, invasive procedures, we chose to follow changes in testis size over time as a measure of the photoperiodic response. 496

2 PHOTOPERIODIC RESPONSE OF INBRED HAMSTERS 497 MATERIALS AND METHODS Animals Male golden hamsters (Mesocricetus auratus) were received from breeders at 6-7 wk of age and initially grouphoused 5-6 per cage with same-strain individuals under a 14L:10D cycle. Animals were of four different inbred strains and one outbred stock (LAK/LVG(SYR)). The outbred animals and the LSH/SsLak and MHA/SsLak animals were received from Charles River (Wilmington, MA); the BIO 1.5 and BIO inbred strains originated from BioBreeders (Watertown, MA). All inbred strains used in this study have been maintained by more than 20 generations of full-sibling mating, and thus can be considered as fully homozygous genotypes [10]. Food and water were available ad libitum throughout the experiment, and cages were changed weekly. Photoperiods At wk of age, the hamsters were grouped 3-5 per cage with same-strain individuals and given individually identifying ear punches; 7-9 animals of each strain or stock were placed in each of the following LD cycles: 6:18, 11.5:12.5, 12:12, 12.5:11.5, and 14:10. The animals were then maintained under their respective photoperiods for 12 wk. The animals in the 14L:1OD groups were kept in the same room where all animals were initially housed, while the groups in 6L:18D were moved into a different light-controlled room. The groups in the other photoperiods were maintained in light-tight, ventilated boxes (internal dimensions: 56 cm x 44 cm x 44 cm), up to 6 cages per box. These boxes were individually equipped with a timer and a single fluorescent light fixture (General Electric F40CW) mounted on the top inside the box. The light intensities measured at the level of the cage floor with cage lid, food, and water bottle in place ranged from lux in the boxes, and from lux in the rooms. Testis Measurements The width of each animal's right testis was determined on the day of assignment to a photoperiod group or transfer from 14L:10D to another photoperiod (Week 0) and subsequently every 2 wk, until the animals had been exposed to their photoperiod for 12 wk. The hamsters were each injected i.m. with 0.1 ml Ketaset (100 mg/ml ketamine HCl, Aveco Co., Inc.), except for the BIO 1.5 animals, which were given 0.07 ml because of heightened sensitivity in this strain to ketamine. The right testis was then gently palpated and measured through the scrotal skin with a pair of digital calipers. Animals with testes too small to palpate were assigned a width of 4.5 mm since this was judged to be the minimum width that could be reliably measured. Data Analysis and Statistics The individual testis width measurements were not used as such for analysis because large differences were apparent between strains in both initial body weight and testis size, and thus the individual testis width measurements were not considered the best index of changes in reproductive state. Rather, each testis width measurement was expressed as a ratio of the initial testis width of the same individual hamster at wk of age under 14L:10D (i.e., width relative to Week 0). Ratios were subjected to a three-way (strain by time by photoperiod) Generalized Linear Models (GLM) analysis of variance with repeated measures using SAS/STAT software for the IBM PC [11]. Data from each photoperiod were then subjected to a two-way (strain by time) GLM analysis and Tukey's Studentized Range tests when appropriate for comparisons within a given week and photoperiod. A probability of p was considered significant. RESULTS The analysis revealed significant effects on testis size of strain, photoperiod, and time, as well as significant interactions of strain by time, strain by photoperiod, photoperiod by time, and strain by photoperiod by time. Within the 14L:10D groups, significant time effects and time-by-strain effects were present; however, no significant strain effect was present. This could be attributed to gradual growth in testis size over time under this photoperiod, with some strains growing more than others. Hamsters of all five groups maintained fully functional testis size under these conditions (Fig. la). Within the 12.5L:11.5D groups, significant time, strain, and time-by-strain effects were present. Animals of all five strains maintained fully functional testis size under these conditions (Fig. lb). However, the differences in size between strains became more pronounced over time. Within the 12L:12D groups, significant time, strain, and time-by-strain effects were present. Significant differences between strains were present at Weeks 10 and 12 (Table 1). This was because the LSH/SsLak strain had significantly smaller relative testis sizes than the other four groups at both Weeks 10 and 12, i.e., the LSH/SsLak hamsters underwent clear gonadal regression while the others did not (Fig. lc). Within the 11.5L:12.5D groups, significant time, strain, and time-by-strain effects were present. Significant differences in relative testis size between strains were present at Weeks 4, 6, 8, and 10 (Table 1), indicating considerable variability between strains in the rate of gonadal regression (Fig. Id). Much of the strain difference was due to the more rapid gonadal regression exhibited by the LSH/SsLak strain compared to other strains and to the failure to show full gonadal regression by some individuals of the MHA/SsLak strain. Within the 6L:18D groups, there were significant time, strain, and time-by-strain effects. Although all animals of all strains exhibited testicular regression, there was a large

3 498 HOTZ VITATERNA AND TUREK a) N (n C: 0 a) I 3. -c -4, a) I a. -. '.! LD 14: b , 0. 6 LD 12.5:1 15 e. ~LD 12.5:11.5~~1 1\ 0-- Lok:LVG(SYR) O--O BIO 1.5 A A MHA/SsLak 0-0 LSH/SsLak O--O BO I I I 9 - LD 6: r~ ~' Weeks of exposure to photoperiod amount of variability in the rate of regression between strains (Fig. 2e), with the LSH/SsLak strain showing the most rapid testicular regression. Significant strain differences in relative testis width were present at Weeks 2, 4, 6, and 8, although at Week 8, there were no significant pairwise differences (Table 1). In order to assess the critical photoperiod of each strain, the relative testis widths of animals after 12 wk of exposure to each photoperiod were compared (Fig. 2). Significant strain differences were found at this time only in the 12L:12D photoperiod: the LSH/SsLak hamsters had regressed testes while none of the hamsters of the other strains exhibited appreciable testicular regression. The difference in response to this single photoperiod indicates that the LSH/ SsLak strain has a different critical photoperiod than the other strains. DISCUSSION Clear differences in both rate of response and critical day length were found between strains, confirming that there is genetic variability in the photoperiodic response in the golden hamster. Within-species differences in reproductive responses to day length have been described in a variety of small mammals, including deer mice [12], white-footed mice [13], voles [14], and Djungarian hamsters [15]. Although many of these intraspecific differences in response to photoperiod have a demonstrable genetic basis [16-18], this is the first study to directly examine genetic influences on critical photoperiod in a mammalian species. However, the genetic differences in photoperiodic response demonstrated in the strain differences described here represent quantitative rather than the qualitative differences described in many other species. The strain differences in photoperiodic response do not seem to be attributable to genetic differences in circadian rhythmicity. The two strains with the shortest and longest free-running periods (BIO 1.5 and BIO 87.20, respectively) exhibited responses to photoperiod closest to those of the outbred stock. Hamsters of the LSH/SsLak strain exhibited phase angles of entrainment to 14L:10D and free-running periods in constant darkness indistinct from those of outbred Lak:LVG(SYR) animals. These results are in contrast to the findings in Djungarian hamsters, in which animals that are nonresponsive to photoperiod exhibited abnormal entrainment to a "short day" light cycle [15]. Similarly, while Eskes and Zucker [6] found that lengthening the period of the circadian rhythm of activity with deuterium oxide was as- FIG. 1. Changes in testis size over 12 wk of exposure to 5 different photoperiods. Testis widths are expressed relative to Week 0 measurements for each animal. Strain means + SEM of hamsters kept on the same photoperiod are shown in each panel. All hamsters were initially housed under 14L:10D, and at wk of age were either maintained under 14L:10D (la), or transferred to 12.5L:11.5D (lb), to 12L:12D (1c), to 11.5L:12.5D (ld), or to 6L:18D (le).

4 PHOTOPERIODIC RESPONSE OF INBRED HAMSTERS 499 TABLE 1. Strain differences by week: Tukey's groupings LD 12:12 Strain Week: Lak:LVG(SYR) A* A A A A A A BIO 1.5 A A A A A A A MHA/SsLak A A A A A A A LSH/SsLak A A A A A B B BIO A A A A A A A LD 11.5: 12.5 Strain Week: Lak:LVG(SYR) A A A AC A AB A BIO 1.5 A A AB B B B A MHA/SsLak A A A A AC A A LSH/SsLak A A C B B B A BIO A A BC C BC B A LD 6:18 Strain Week: Lak:LVG(SYR) A AB A AB A A A BIO 1.5 A A AB AC A A A MHA/SsLak A BC A B A A A LSH/SsLak A ABC B C A B B BIO A C AB ABC A A A *Strains with the same letter within week are not significantly different as determined by Tukey's Studentized Range test. sociated with an alteration in the critical photoperiod for maintaining testis size, no such relationship between freerunning period and the photoperiodic response was observed among the four inbred strains examined here. It may be that the altered reproductive responses to day length of the LSH/SsLak hamsters are more analogous to : *--0 Lak:LVGi O--O BIO 1.5 e A... A MHA/Ss o LSH/Ssl, c0 BIO > T i I Hours of light per day FIG. 2. Testis size of hamsters of 4 different inbred strains and 1 outbred stock after 12 wk of exposure to 5 different photoperiods. Testis widths are expressed relative to Week 0 measurements for each animal. the intraspecific differences observed in Peromyscus species than in Djungarian hamsters. Peromyscus, which are insensitive to short days, do not show any alterations in melatonin rhythms [19, 20] and are not responsive to exogenous melatonin administration [21,22]. In contrast, LSH/ SsLak animals show a more rapid reproductive regression in response to exogenous melatonin than do outbred animals [23]. Thus, it is possible that the LSH/SsLak animals represent the converse of nonresponsive Peromyscus: rather than an insensitivity to photoperiod due to a loss of melatonin sensitivity, LSH/SsLak animals may have an increased sensitivity to short photoperiods due to a lowered melatonin threshold. Alterations at other levels in the system, even possibly at the level of the testis, could also be involved. The present data do not address this issue; however, further examination of the LSH/SsLak strain might be informative in identifying factors controlling the rate of reproductive involution. We cannot rule out the possibility that altered melatonin secretion could be involved in the genetic differences in photoperiodic response. Lauber and Vriend [24] have found that nocturnal melatonin levels were considerably reduced in the albino inbred strain MHA/SsLak when compared with outbred Lak:LVG(SYR) hamsters. The MHA strain, interestingly, was among the slowest to respond to short photo-

5 50 HOTZ VITATERNA AND TUREK periods, and 2 (of 8) individuals of this strain still did not have regressed testes by Week 12 in 11.5L:12.5D in the present study. Lauber and Vriend attributed the reduced melatonin levels to the albinism of this strain. The BIO 1.5 strain is also an albino strain, and in the present study was among the more rapid to respond to a short photoperiod. Thus the reduced melatonin may not be related to albinism, and albinism per se does not appear to influence the rate of response to a short photoperiod. Alternatively, the BIO 1.5 strain may have other changes in the response of the neuroendocrine-gonadal axis to day length that can compensate for an albinism-induced reduction in melatonin. It is of interest that the rate of testicular regression appears to be a function of how near the day length is to the critical photoperiod. Testicular regression took longer in animals in 11.5L:12.5D when compared (within strain) with animals in 6L:18D, for all strains. The strain with the longest critical photoperiod for maintaining testicular size (i.e., LSH/ SsLak) exhibited the most rapid testicular regression in both photoperiods. Testicular regression in the LSH/SsLak strain was also slowest in 12L:12D compared to the regression observed during exposure to the two shorter photoperiods. This leads us to speculate that the apparent critical photoperiod may depend upon the length of exposure to that photoperiod. Certainly in the present study a different conclusion about the critical photoperiod would have been reached if Week 8 had been the endpoint rather than Week 12. Indeed, it is possible that with an additional 12 wk of exposure to 12L:12D, animals of other strains might have exhibited testicular regression. Given this demonstrated variability in rate of response, one may need to observe animals for several weeks of exposure before drawing conclusions concerning the effects of a given photoperiod. Our finding that LSH/SsLak hamsters show more rapid gonadal regression during exposure to short days when compared with outbred hamsters is consistent with the findings by others with both female [25] and male [26] LSH/ SsLak hamsters. However, the comparison with other inbred strains has not been made previously. Our results demonstrate that the more rapid response of LSH/SsLak hamsters is not simply a consequence of inbreeding and the resultant homozygous genotype (i.e., inbreeding depression), but instead is a characteristic of this strain's specific genotype. The critical photoperiod of outbred animals in our study differed from that determined in a previous study [4]. We employed hamsters of the same outbred stock (Lak:LVG(SYR)), yet found that a shorter day length was required to observe testicular regression. Similarly, Reiter [27] observed that the rate of reproductive atrophy appeared to be decreasing over the years. Since we have observed a correlation between rate of testicular regression and critical photoperiod, the slower response and longer critical photoperiod may represent the same phenomenon. It may be that over the last ten to twenty years there has been a decrease in the sensitivity to the inhibitory effects of short days on reproductive function due to the relaxed selective pressure for this trait with laboratory breeding, or, as Reiter has suggested [27], inbreeding or nutritional improvements could be responsible as well. Loss of photoperiodic response has been observed with laboratory breeding of another species: the reproductive response to short day length was absent in voles from an outbred population laboratorybred for 12 yr but not in animals descended from more recent wild-trapping [14]. However, a study comparing the outbred Lak:LVG(SYR) stock with hamsters derived from more recent wild-trapping failed to show any difference in rate of reproductive involution [28]. Since the founding population from which laboratory stocks and strains of hamsters descended consisted of three littermates [29], one might expect that there was very little genetic variability in the photoperiodic response for either inbreeding or selection to have an effect. However, the differences among inbred strains reported here demonstrate that there is indeed genetic variability in photoperiodic response among laboratory strains of hamsters. Whether this variability derives from the original allelic differences in the wild-caught animals or from mutations accumulated with laboratory breeding cannot be determined here. ACKNOWLEDGMENTS The authors thank Dr. John Kirby for his assistance with the statistical analysis and Ms. Susan Losee-Olson and Ms. Deborah Hinch for their technical assistance. REFERENCES 1. Gaston S, Menaker M. Photoperiodic control of hamster testis. Science 1967; 158: Turek FW, Campbell CS. Photoperiodic regulation of neuroendocrine-gonadal activity. Biol Reprod 1979; 20: Stetson MH, Elliott JA, Menaker M. Photoperiodic regulation of hamster testis: circadian sensitivity to the effects of light. Biol Reprod 1975; 13: Elliott JA. Circadian rhythms and photoperiodic time measurement in mammals. Fed Proc 1976; 12: Elliott JA, Goldman BD. Circadian rhythms, entrainment and photoperiodism in the Syrian hamster. In: Adler NT (ed.), Neuroendocrinology of Reproduction: Physiology and Behavior. New York: Plenum Publishers; 1981: Eskes GA, Zucker. Photoperiodic regulation of the hamster testis: dependence on circadian rhythms. Proc Natl Acad Sci USA 1978; 75: Bartness TJ, Goldman BD. Mammalian pineal melatonin: a clock for all seasons. Experientia 1989; 45: Karsch FJ, Bittman EL, Foster DL, Goodman RL, Legan SJ, Robinson JE. Neuroendocrine basis of seasonal reproduction. Recent Prog Horm Res 1984; 40: Hotz MM, Turek FW. Effects of light on the circadian clock in inbred strains of golden hamsters. Soc Neurosci Abstr 1989; 15(part 1):492 (abstract 201.5). 10. Falconer DS. Introduction to Quantitative Genetics. New York: Longman; SAS. SAS/STAT Release Cary, NC: Statistical Analysis Systems Institute Inc.; Desjardins C, Lopez MJ. Environmental cues evoke differential responses in pituitary-testicular function in deer mice. Endocrinology 1983; 112: Lynch GR, Gendler SL. Multiple responses to a short day photoperiod occur in the mouse Peromyscus leucopus. Oecologia 1980; 45: Nelson RJ. Photoperiodic regulation of reproductive development in male prairie voles: influence of laboratory breeding. Biol Reprod 1985; 33: Puchalski W, Lynch GR. Characterization of circadian function in Djungarian hamsters insensitive to short day photoperiod. J Comp Physiol A 1988; 162:

6 PHOTOPERIODIC RESPONSE OF INBRED HAMSTERS Desjardins C, Bronson FH, Blank JL. Genetic selection for reproductive photoresponsiveness in deer mice. Nature 1986; 322: Lynch GR, Lynch CB, Kliman RM. Genetic analysis of photoresponsiveness in the Djungarian hamster, Phodopus sungorus. J Comp Physiol 1989; 164: Wichman HA, Lynch CB. Genetic variation for seasonal adaptation in Peromyscus leucopus: nonreciprocal breakdown in a population cross. J Hered 1991; 82: Blank JL, Nelson RJ, Vaughan MK, Reiter RJ. Pineal melatonin content in photoperiodically responsive and non-responsive phenotypes of deer mice. Comp Biochem Physiol 1988; 91A: Lynch GR, Sullivan JK, Heath HW, Tamarkin L. Daily melatonin rhythms in photoperiod sensitive and insensitive white-footed mice (Peromyscus leucopus). In: The Pineal and Its Hormones. New York: Alan R Liss, Inc.; 1982: Heath HW, Lynch GR. Intraspecific differences for melatonin-induced reproductive regression and the seasonal molt in Peromyscus leucopus. Gen Comp Endocrinol 1982; 48: Blank JL, Freeman DA. Differential reproductive response to short photoperiod in deer mice: role of melatonin. J Comp Physiol A 1991; 169: Reiter RJ, Sabry I, Nordio M, Vaughan MK, Migliaccio S. Rate of reproductive involution following either exposure to short days or daily administration of melatonin is faster in inbred than in random-bred female Syrian hamsters. J Endocrinol 1989; 120: Lauber JK, Vriend J. Melatonin reduction by lithium and albinism in quail and hamsters. Gen Comp Endocrinol 1989; 76: Hauser UE, Benson B. Early short photoperiod effects in female LSH/SsLak Syrian hamsters. Endocrinology 1987; 120: Urbanski HF, Kim SO, Connolly ML. Influence of photoperiod and 6-methoxybenzoxazolinone on the reproductive axis of inbred LSH/SsLak male hamsters. J Reprod Fertil 1990; 90: Reiter RJ. The pineal and its hormones in the control of reproduction in mammals. Endocr Rev 1980; 1: Reiter RJ, Golovko V. Failure of inbreeding to influence pineal-induced gonadal regression in short day exposed hamsters. Arch Androl 1983; 10: Adler S. Origin of the golden hamster Cricetus auratus as a laboratory animal. Nature 1948; 162: