| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
NEUROENDOCRINOLOGY |
Department of Neurobiology and Physiology, Northwestern University, Evanston, Illinois 60208
Address all correspondence and requests for reprints to: Dr. Teresa H. Horton, Department of Neurobiology and Physiology, Northwestern University, 2-160 Hogan, 2153 North Campus Drive, Evanston, Illinois 60208. E-mail: . thorton{at}northwestern.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In the laboratory, seasonal changes in reproductive activity can be simulated by manipulation of photoperiod (7, 8, 9) and melatonin (10, 11). Siberian hamsters held in a short photoperiod for several weeks after weaning have negligible amounts of circulating gonadotropins and small testes (5, 6). After transfer from a short to a long photoperiod, FSH increases significantly within 3–5 d, peaks by the 10th day, and then decreases and plateaus to levels characteristic of a reproductively mature male (5, 6, 9). This increase in serum FSH can be blocked in part by injection of a GnRH antagonist; suggesting that secretion of FSH is partially dependent upon GnRH release (6). Testicular weight increases by the 10th day of photostimulation (5, 9) subsequent to FSH stimulation (12, 13). In contrast, despite the increase in FSH, serum LH levels either do not increase in response to photostimulation (5, 14) or increase much later than FSH (9). Correspondingly, serum T levels are low. Thus, photostimulation causes a rapid increase in serum FSH and testicular weight, but not in serum LH and T. These differences in the time course for release of FSH and LH are consistent with a model in which increasing day lengths trigger the release of FSH to promote the development of the testes, which requires several weeks to complete, in anticipation of improved environmental conditions. In contrast, the release of LH and T, which are required for the final stages of sperm maturation and regulate reproductive behaviors, may be delayed for several weeks or until additional proximate cues are present that more directly reflect environmental conditions required for successful reproduction. Given that both FSH and LH are required to induce full testicular maturation (12, 15) and the full complement of reproductive behaviors, we investigated whether an additional environmental stimulus is able to induce LH secretion in this species and whether the response to that stimulus is, in turn, dependent upon photoperiod.
In several species, serum LH levels in males and females can be influenced by exposure to a prospective mate. For example, serum LH levels of male laboratory mice can be increased by exposure to females (16). Similarly, the presence of an ewe facilitates an increase in plasma LH and redevelopment of testes in rams before the onset of breeding (17). An increase in serum LH and/or T in male Golden hamsters (Mesocricetus auratus) in response to female urine and vaginal secretions is well documented (16, 18, 19, 20). Hence, we investigated whether exposure to a female would serve as an additional environmental stimulus to induce LH secretion in males.
The first study reported here tested the hypothesis that presentation of a female can induce LH secretion in male Siberian hamsters and that long photoperiods provide a permissive condition enabling males to release LH upon exposure to a female. FSH levels were monitored to determine whether exposure to a female also triggered FSH release. Cortisol levels in male Siberian hamsters are known to decline upon establishment of a pair bond and to increase upon separation of the pair (21, 22). Cortisol is also elevated in Siberian hamsters housed in a short photoperiod compared with a long photoperiod (23). Because corticosteroids are known to influence the differential secretion of LH and FSH (24, 25), serum cortisol was measured when sufficient serum was available. Subsequent studies investigated whether the induction of LH in response to female exposure would be prevented by a short-day pattern of melatonin secretion and was dependent on GnRH. Exogenous melatonin was administered to long-day males for 7 wk to mimic short-day exposure; subsequently, males were presented with a female to determine whether the melatonin treatment would prevent female-induced LH secretion in males. In the last study a GnRH antagonist was administered before exposure of the male to a female to determine whether the secretion of LH upon female exposure was GnRH dependent.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Exp 1: regulation of FSH and LH by photoperiod and female exposure
Males were atrially catheterized and serially sampled to test whether the presentation of a female could induce LH secretion and also if long photoperiods provided a permissive condition enabling males to release LH upon female exposure.
Male hamsters were raised in LD until weaning (18 d of age) and were then divided into four groups (Fig. 1
and Table 1). One group remained in 16L:8D for 4–5.5 wk, i.e. until 7–8.5 wk of age (LD control group). The other three groups of animals were placed in short days (SD; 6L:18D; lights on at 0800 h CST) for 4 wk to inhibit reproductive development (5, 6). Of these three groups, one group of hamsters housed in SD remained in SD for an additional 1–2 wk to serve as a control group (SD control group). The second group of SD animals was reexposed to LD for 4 d (SD + 4 LD group), and the third group was reexposed to LD for 11 d (SD + 11 LD group). The timing of the transfer from SD to LD conditions was effected so that animals from the SD + 4 LD and SD + 11 LD groups were sampled on the same day along with representative hamsters from each control group. Hamsters used as control animals for chronic exposure to long and short days were sampled at 7–8.5 wk of age to cover the same span of ages as the animals that were transferred from SD back to LD. Subsequently, on the day of sampling (defined as the day of female exposure), half the males in each photoperiod treatment group were exposed to a female while the other half were not.
|
|
. For Exp 1, LH levels were measured from 35 µl of each sample in duplicate. The remaining sera from each animal were pooled to measure FSH and cortisol. Pooling was accomplished by combining four samples collected within a given hour (i.e. 0800, 0815, 0830, and 0845 h). The midpoint of sampling times for samples taken on the hour and 15, 30, and 45 min past the hour is 22.5 min past the hour. To simplify presentation of the data, the pooled sample for each animal is presented at the midpoint of the hour (i.e. samples taken at 0800, 0815, 0830, and 0845 h are presented as having been taken at 0830 h). FSH was measured in duplicate using 35 µl pooled serum. If sufficient serum remained (i.e. 25 µl), cortisol was measured. Because cortisol was measured from sera that remained after LH and FSH measurements, the sample sizes for cortisol measures are lower than those for LH and FSH. Siberian hamsters release cortisol, but not corticosterone, in response to stress (21). Cortisol levels were measured to determine whether there were any changes in circulating cortisol levels associated with the presence of a female or change in photoperiod that might influence the differential secretion of LH and FSH (21, 24, 25).
Exp 2: role of melatonin in the short photoperiod induced inhibition of LH release by males upon female exposure
This study tested the hypothesis that treatment with exogenous melatonin, in a manner known to induce testicular regression, would prevent female-induced LH secretion in males.
Male hamsters were raised in LD (16L:8D) until weaning (18 d of age). Beginning at weaning, they were injected once daily for 7 wk with either vehicle (0.1 ml ethanolic saline, 1:9) or melatonin (Sigma, St. Louis, MO; 5 µg in 0.1 ml ethanolic saline, 1:9) (29). Melatonin solution was prepared fresh daily by diluting 0.3 ml stock solution (0.1 g melatonin in 200 ml 95% ethanol) in 2.7 ml saline (29). The injections were administered 3–4 h before lights off (30), i.e. between 1500 and 1600 h. The dose and time were chosen because they are known to induce gonadal regression in this species (29, 30).
After 7 wk of injections, males were individually housed for 4 d before exposure to a female so as to remove any potential inhibitory effects of group housing. Injections continued each afternoon for the remainder of the experiment. On the day before female exposure, a baseline sample of blood (0.4 ml) was taken. On the day of female exposure, some males of the vehicle and melatonin injection groups were exposed to a female (n = 11 and 10), whereas the others were not (n = 8 and 5). The female remained in the male’s cage for 1.5 h. Blood samples were taken 1 and 1.5 h after introduction of the female into the male’s cage. Blood samples were collected by cardiac puncture after anesthetizing the males with halothane (Halocarbon Laboratories, River Edge, NJ; 0.1 ml). LH, FSH, and T levels were measured by RIA. Testes were removed and weighed.
Exp 3: role of GnRH in the induction of LH upon female exposure
To test the hypothesis that the induction of LH in response to female exposure was dependent on GnRH, a GnRH antagonist was administered to males 24 h before exposure to females.
Thirty-six male hamsters were raised in LD (16L:8D) and used at 3–4 months of age. They were individually housed for 7 d before exposure to a female. Four days before exposure to a female, baseline blood samples were taken by cardiac puncture. One day before exposure to a female, the males were injected with either vehicle (1:2, propylene glycol/water) or antide at one of three doses [Sigma; 1 (low dose), 4 (medium dose), or 8 (high dose) mg/kg BW, dissolved in vehicle] (31). Antide has been characterized as a GnRH antagonist with a long half-life. On the day of exposure to females, six males in each group were exposed to a female. Additional males treated with vehicle or the medium dose of antide were sampled, but not exposed to females (n = 6/group). Blood samples (0.4 ml) were taken 1 and 1.5 h after the onset of exposure to a female. Males that were and were not exposed to females were located in the same room in adjacent cages at the time blood samples were taken. Blood samples were collected by cardiac puncture after anesthetizing the males with halothane. LH, FSH, and T levels were measured by RIA. Animals were euthanized after the final blood sample, and testes were removed and weighed.
Atrial catheterization and blood sampling
In Exp 1 male hamsters were implanted with a catheter 4 d before the day of the experiment. The catheter was inserted through the jugular vein into the right atrium as described previously (32). The catheter was flushed daily with heparinized saline (0.3 ml, 7 U/ml) to maintain patency. On the day of the experiment, serial blood samples of 300 µl were withdrawn through the catheter every 15 min and replaced by an equivalent amount of donor blood [see Ref. 32 for preparation of donor blood]. Hematocrit levels were evaluated at the beginning, middle, and end of each bleed and were used as an index for compensation with donor blood. As shown previously (24), hematocrit levels averaged 43%; there was no significant change in hematocrit levels during the course of a sampling period.
Having established the time course for LH release after exposure to a female in Exp 1, blood samples were collected by cardiac puncture in Exp 2 and 3 to simplify the experimental protocols. Preliminary experiments were conducted to verify that the use of halothane as an anesthetic and cardiac puncture as the method for collecting blood did not alter the responses of males (data not shown).
Preparation of females
A separate female was placed in the cage of each male at the onset of the exposure period. That female remained with the male throughout the experiment. For Exp 1, 2- to 4-month-old LD females were individually housed for 4 d before presentation to a male so as to remove the potential inhibitory effects of group housing on their estrous cycles and possible effectiveness as a stimulus. However, we subsequently found that in this species the efficacy of a LD female as a stimulus to trigger LH release in male Siberian hamsters does not depend on her endocrine status (33). Exposure to a female, regardless of the stage of her estrous cycle, or even exposure to an ovariectomized female produces an increase in serum LH in male Siberian hamsters. Hence, in Exp 2 and 3, the females were not individually housed before exposure to a male.
Hormone measurements
Serum LH, FSH, cortisol, and T levels were measured by RIA. LH and FSH were measured using materials supplied by the NIDDK (Rockville, MD). All assays were performed at the Ligand Assay Core of Northwestern University. The LH and FSH assays have been validated previously for use in this species (5, 6). LH and FSH were measured in duplicate from 35-µl aliquots of serum. The LH standard used was rat LH RP-3, and the antibody used was rat LH-S-11. The intraassay coefficient of variation for LH was 10.9%, and the interassay coefficient of variation was 16.1%. The FSH standard used was rat FSH RP-3, and the antibody used was rat FSH S-11. The intraassay coefficient of variation for FSH was 9.5%, whereas the interassay coefficient of variation was 16.4%. Serial dilutions of male Siberian hamster sera were included with each LH and FSH assay, and their concentration curves were parallel to standard LH and FSH curves in the RIAs. Cortisol and T were measured in 25-µl aliquots of serum using RIA kits purchased from ICN Biomedicals, Inc. (Costa Mesa, CA). The T assay has been previously validated (5, 6). To validate the cortisol assay, serial dilutions of Siberian hamster serum were shown to yield displacement curves parallel to the standard included within the kit. The intraassay coefficient of variation for cortisol was 13.4%, whereas the intraassay coefficient of variation for T was 11.3%. The minimum detectable levels of the hormones were as follows: LH, 0.028 ng/ml; FSH, 1 ng/ml; cortisol, 1.5 ng/ml; and T, 0.02 ng/ml.
Statistical analyses
All statistical tests comparing serum hormone levels were conducted using NCSS 97 (Number Cruncher Statistical Systems, Kaysville, UT). Treatment effects were judged to be statistically significant if P < 0.05. Two-way ANOVA with repeated measures was used for comparisons of hormone levels between the groups. Post hoc comparisons were made with Duncan’s multiple comparison test.
For Exp 1, the 
2 test was used to compare the proportion of males in each group responding to a female, in addition to the use of ANOVA to compare serum hormone levels, as described above. Three criteria were used to classify a male as having released gonadotropins in response to the presentation of a female. First, the average and SD of the baseline concentration for each male were calculated from the three samples taken before time zero. A male was classified as having responded to the presence of a female if his serum hormone levels after presentation of a female exceeded his average baseline plus 3 SD. The second criterion compared the hormone levels of the males exposed to females to the average values for males not exposed to females at each sampling time. A male was classified as having responded to a female when hormone measurements exceeded the average plus 3 SD of the hormone levels for males that were not exposed to females. Thirdly, to be considered as a response to the presence of a female, the LH levels needed to be above the average baseline plus 3 SD for two consecutive time points. The half-life of LH is 42 min (34). Because LH measurements were made on samples collected every 15 min, LH levels needed to be elevated for at least 30 min to be classified as a significant response. FSH and cortisol were measured from pooled samples within a 1-h time interval; hence, the criterion of an elevation in two consecutive samples was not applied to those hormone measurements. Thus, all three criteria were required to be fulfilled to classify a positive LH response, whereas only the first two needed to be fulfilled for FSH and cortisol analyses. Using these criteria, a 2 analysis was performed to compare the proportions of males in each photoperiod group that showed an elevation in hormone levels on exposure to female.
Finally, three additional parameters were evaluated to determine whether the lengths of time hamsters were exposed to a long photoperiod altered the temporal dynamics or profile of hormone release. First, to evaluate whether the photoperiod treatment influenced the time course for initiation of the response, the time of onset of hormone increase for each male was calculated as the first time point at which hormone values exceeded 3 SD above baseline. Second, the duration of elevated LH secretion was calculated as the time between the first and last time points at which hormone values exceeded 3 SD above baseline. Finally, the maximal levels of LH secretion for each individual male during the period of elevated LH secretion were determined and averaged. Because the maximum level for each individual did not occur during the same 15-min sampling interval, the maximum levels were slightly higher than the average values reported for each group at specific sampling intervals in Fig. 2.
|
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
; P < 0.001).
LH levels.
Basal serum LH levels (defined as the average of the three samples taken immediately before introduction of the female into the male’s cage) were low (0.37 ± 0.08 ng/ml) and did not differ among the four groups before introduction of the female into the cage of the male (Fig. 2). The presence of a female in the male’s cage caused a significant increase in serum LH levels in the three groups of males subjected to LD (by ANOVA, female x time: F24,1212 = 11.01; P < 0.001).
Photoperiod significantly altered the ability of male Siberian hamsters to release LH upon exposure to a female (by ANOVA, photoperiod: F3,54 = 3.14; P < 0.05). No increase in LH was induced by the presence of the female in males that remained in SD, i.e. SD controls (Fig. 2). All three groups of males housed in LD, i.e. those continuously housed in LD (LD control males) and those that had been transferred from SD to LD for either 4 or 11 d (SD + 4 LD and SD + 11 LD males) released LH after exposure to a female. The average peak of the LH rise, however, was lower in males that had been exposed to LD for only 4 d (1.72 ng/ml at 1130 h) compared with males that had either been exposed to LD for 11 d (2.6 ng/ml at 1130 h) or been continuously exposed to LD (2.76 ng/ml at 1145 h; Fig. 2). Because the maximal value for all animals did not occur within the same sampling period, a separate analysis compared the maximal LH values for each animal. No significant difference in the maximal LH value was observed among the males in the three groups exposed to LD. No differences were observed among the three groups of males exposed to LD in either the time of onset of increase in LH, the time of the LH peak, or the duration of increase in LH. The initial increase in LH occurred within 15–45 min of female exposure in all three groups housed in LD.
The proportion of males that responded to a female was determined by categorizing males as having responded or not responded using the criteria described in Materials and Methods. Based on these criteria, 0% of SD control males (0 of 11), 57% of SD + 4 LD males (4 of 7), 83% of SD + 11 LD males (5 of 6), and 91% of LD control males (10 of 11) exhibited an increase in LH release after exposure to a female. This difference is significant (2 = 21.07; P < 0.001). To assess whether the proportion of males that released LH in response to the presence of a female increased with the length of exposure to LD, the analysis was redone, excluding the SD control group. Although there appeared to be a trend toward an increasing proportion of males releasing LH in response to the presence of a female with increasing time in LD, the differences in these proportions were not statistically significant (2 = 3.04; P = 0.39).
FSH levels.
FSH levels were significantly influenced by photoperiod (by ANOVA, photoperiod: F3,390 = 9.99; P < 0.001), but not by exposure to a female (by ANOVA, female, P = 1.0 Fig. 3). FSH levels of LD control males (5.22 ± 0.39 ng/ml) and SD + 11 LD males (6.45 ± 0.54 ng/ml) were significantly higher than those of SD control males (2.46 ± 0.38 ng/ml) and SD + 4 LD males (2.74 ± 0.41 ng/ml). None of the males exhibited an increase in FSH secretion after exposure to a female that met the criteria for female-induced gonadotropin secretion set forth in Materials and Methods. Linear regression analysis indicated that paired testes weights were positively and significantly correlated with serum FSH levels (r2 = 0.9; P < 0.05).
|
). Neither photoperiod nor the presence of a female exerted a significant main treatment effect on cortisol levels (by ANOVA, photoperiod, P = 0.12; female, P = 0.61). There was a significant interaction effect between photoperiod and time (by ANOVA, time x photoperiod, F15,284 = 1.74; P < 0.05) on serum cortisol. To elucidate this interaction, a one-way ANOVA was performed for the entire sampling period as well as for the final hour of the experiment, a time when cortisol levels had been stable for several hours. The final hour was evaluated to eliminate the possibility that the change in serum cortisol levels during the first 2 h of sampling may have obscured differences in cortisol due to photoperiod or the presence of a female. This analysis revealed a statistically significant difference in cortisol levels associated with photoperiod (by ANOVA, photoperiod, F3,47 = 2.93; P < 0.05), such that cortisol levels of SD males (47.6 ± 2.8) were significantly higher than those of LD males (35.0 ± 3.88). There was no change in serum cortisol in males upon exposure to a female either before or concurrent with the elevation of LH.
|
; by ANOVA, F1,33 = 40.97; P < 0.001) and regressed testes (Fig. 5B; by ANOVA, F1,34 = 202.41; P < 0.001). In this experiment, baseline LH (F1,33 = 4.92; P < 0.05), FSH (F1,33 = 13.43; P < 0.001), and T (F1,33 = 6.42; P < 0.05) levels were significantly reduced in melatonin-treated animals compared with vehicle-treated animals (Fig. 6). Other changes associated with exposure to short days (i.e. molt to white coat color and an increase in aggressive behavior toward the female) were observed, but not quantified. As expected, vehicle-treated males exhibited a significant increase in LH after exposure to a female (Fig. 6A). Prior treatment of the males with melatonin prevented the female-induced increase in circulating LH levels (Fig. 6A). Thus, there was a significant effect of melatonin on the ability of males to release LH when exposed to a female (by ANOVA, melatonin or vehicle treatment, F1,94 = 8.46; P < 0.05; female, F21,94 = 11.73; P < 0.05). Males not exposed to a female did not show an increase in LH regardless of melatonin treatment.
|
|
; by ANOVA, F1,94 = 24.23; P < 0.001). FSH levels in melatonin-injected males averaged 3.55 ± 0.32 ng/ml, whereas those in vehicle-injected males averaged 6.2 ± 0.29 ng/ml. FSH levels were not influenced by the presence of a female regardless of melatonin treatment. T levels were significantly affected by both melatonin treatment (by ANOVA, F1,92 = 61.1; P < 0.001) and female exposure (by ANOVA, F1,92 = 8.01; P < 0.05; Fig. 6C). Baseline T levels were lower in melatonin-treated animals. Exposure to a female resulted in a significant increase in T levels in vehicle-treated, but not melatonin-treated, males.
Exp 3
The GnRH antagonist, antide, blocked the female-induced increase in LH in males housed in LD in a dose-dependent manner (Fig. 7A; by ANOVA, F5,99 = 5.00; P < 0.05). Males in all groups had low baseline LH levels. LH levels were also low in males that were not exposed to females; there was no significant difference in LH levels before and after treatment with the medium dose of antide in males that were not exposed to females. Males that received vehicle injections and the low dose of antide showed significantly higher LH levels upon female exposure compared with their baseline LH levels. Males receiving medium and high doses of antide did not show a significant increase in LH levels over baseline upon female exposure. The high dose of antide appears to completely suppress female-induced LH secretion, whereas the low and medium doses of antide do not.
|
; by ANOVA). T levels were variable among groups before antide treatment or female exposure, thus making interpretation of antide effects on T difficult. T levels in males treated with the low dose of antide increased significantly from baseline after female exposure (Fig. 7C; by ANOVA, F2,95 = 3.39; P < 0.05). |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The results of Exp 1 demonstrate that secretion of FSH and LH in male Siberian hamsters is differentially regulated by two distinct environmental variables. Within a few days after transfer from a short to a long photoperiod, serum FSH concentrations increase selectively, without a concomitant increase in serum LH concentrations. During this time, presentation of a female can trigger a rapid and sustained increase in LH secretion, but not FSH secretion, in males. Female-induced LH secretion occurs only in males exposed to long photoperiods, but not in males exposed to short photoperiods. SD males regain the ability to release LH upon female exposure within 4 d of photostimulation. These data suggest that photoperiod modulates a sensory pathway, the neuroendocrine response to afferent signals from that pathway, or both to regulate LH release.
Siberian hamsters have very low levels of circulating gonadotropins when housed in SD. Previous studies indicate that after transfer from a short to a long photoperiod, GnRH gene expression and pituitary FSH levels increase within 26 h (35, 36). By 5 d of photostimulation, serum FSH levels are significantly elevated compared with SD control levels (5, 9). The rising levels of serum FSH peak between 10 and 15 d after the onset of photostimulation and decline to stable adult levels by 20–30 d of photostimulation (5, 9). This increase in FSH is the primary signal for stimulating testicular development in Siberian hamsters (13, 37). In the present study (Exp 1), we did not detect an increase in FSH by 4 d of photostimulation, but did see a significant increase in serum FSH by 11 d of photostimulation. Linear regression analysis revealed a positive correlation between FSH levels and paired testes weights within and between groups. The variations in FSH levels in response to 11 d of photostimulation in this study are consistent with those observed in previous studies (5, 9). The failure to see an increase in FSH within 4 d of photostimulation may be due to the outbreeding of our hamster colony or procedural differences, such as putting males into SD at weaning rather than at birth (9), keeping them in SD for only 28–35 d instead of up to 60 d (9), or maintaining them on a 6L:18D cycle instead of a 9L:15D (9) or 8L:16D (5) cycle.
In contrast to the relatively early rise in FSH levels, LH levels have been reported to remain low for at least 20 d, if not longer, after photostimulation (9, 36). The present data confirm and extend the observation that serum LH levels remain low in male Siberian hamsters for several days or weeks after photostimulation. Baseline concentrations of serum LH did not differ among males housed in the four photoperiod conditions in Exp 1 in the absence of exposure to a female (average for all four groups, 0.37 ± 0.08 ng/ml). We did detect a small, but significant, difference in baseline LH levels between vehicle- and melatonin-treated males in Exp 2. Exposure to LD or treatment with vehicle did, however, exert a permissive effect, enabling males to release LH after presentation of a female. Thus, the secretion of both FSH and LH is regulated by photoperiod and melatonin. FSH levels increase in direct response to an increase in photoperiod and do not increase in response to the presence of a female. In contrast, stimulation of LH release is sensitive to additional environmental cues, specifically the presence of a female, and the response to this cue can be modulated by photoperiod and melatonin.
In Exp 1, we have defined a time frame after transfer from a short to a long photoperiod over which an increase in LH secretion in the male in response to the presence of a female was seen. The time course for restoration of the response corresponds to the time course over which photoperiod alone has been shown to trigger changes in FSH secretion (5, 9), pituitary gonadotropin subunit mRNA (14), GnRH mRNA (35), and GnRH content and neuronal cell numbers (36). A significant increase in serum FSH levels was not detected on day 4 of Exp 1. When taken in total, the present data are consistent with those from previous studies in suggesting that a series of changes in the activity of the neuroendocrine system is induced rapidly after exposure to long days. The present data extend these changes to include the responsiveness to other environmental stimuli in addition to the direct control of gonadotropin secretion by photoperiod.
Exp 2 provides compelling evidence that the effects of photoperiod on the ability of males to release LH when presented with a female are influenced at least in part by the pineal hormone, melatonin. Exposure to SD results in a longer duration of melatonin secretion than exposure to LD (38). In Exp 2, the longer duration of melatonin signal was approximated by the daily injection of melatonin to LD males and was shown to prevent the release of LH in males when they were exposed to females. Although there appears to be a slight increase in LH levels in melatonin-injected males upon female exposure (Fig. 6A), this increase is not statistically significant and does not trigger a significant increase in T (Fig. 6C). Previous studies have shown that the pituitaries of male Golden hamsters housed in SD are capable of releasing large amounts of LH in response to the acute administration of GnRH (39, 40, 41, 42) or N-methyl-D,L-aspartate (NMDA) (42). NMDA is hypothesized to stimulate GnRH release in Siberian hamsters (43). Male Siberian hamsters housed in SD actually release larger amounts of LH in response to an equivalent dose of the glutamatergic agonist NMDA than males housed in LD (43). These observations suggest that the pituitaries of males housed in short photoperiods or treated with melatonin are capable of releasing large amounts of LH if an appropriate stimulus is received. It is possible that the failure to observe a large increase in LH in SD males after exposure to a female is a result of pituitary insensitivity to GnRH or insufficient reserves of LH; however, the existing data favor the alternative hypothesis that short photoperiods inhibit the release of GnRH by males in response to the presence of a female.
In Exp 2, FSH levels in melatonin-injected males are low, confirming that the longer duration of circulating melatonin inhibits the hypothalamic-pituitary-gonadal axis. The baseline levels of LH and T were significantly lower in melatonin-treated males compared with vehicle-treated males in Exp 2 in contrast to Exp 1, in which no photoperiod-dependent difference was observed in LH levels. It is possible that differences in the bleeding procedures used in Exp 1 and 2 might explain these different observations. Before the use of cardiac puncture, however, preliminary studies were conducted in LD males to verify that the anesthetic used (halothane) and the repeated use of cardiac puncture did not influence the ability of a male to release LH in the presence of a female. During these preliminary studies we did not detect a significant reduction or augmentation of LH release by males compared with the results of Exp 1. The variability in LH levels between experiments is consistent with observations from other laboratories (38) and emphasizes the role of the interaction between photoperiod and other environmental cues. We found no difference in LH between LD and SD males in Exp 1, but observed a difference between the melatonin-treated vs. vehicle-treated males in Exp 2. One possible explanation for this observation is that the melatonin treatment used in Exp 2 was more effective in suppressing LH release than the short photoperiod treatment used in Exp 1. The males in Exp 2 were treated with melatonin for 7 wk in contrast to the 4- to 5-wk exposure to short photoperiods in Exp 1. It is also possible that increased age or the length of time animals were exposed to LD exerted a permissive effect on the release of LH in males housed in LD; increasing LH release has been observed with increased exposure to LD (9). The animals in Exp 2 were sampled at approximately 10 wk of age; in contrast, animals in Exp 1 were sampled at 7–8 wk of age. Thus, the modest difference in LH levels between vehicle- and melatonin-injected males may reflect an increase in maturation of and LH secretion by the hypothalamic-pituitary-gonadal axis in vehicle-treated males housed in long photoperiod.
Because the presence of a female triggered the release of LH, but not FSH, we tested the hypothesis that GnRH was indeed required to stimulate the release of LH when males were exposed to a female (Exp 3). The GnRH antagonist, antide, was shown to block the increase in LH in the group receiving the high dose. In our hands, the baseline T data were highly variable and precluded us from drawing any conclusions about the effects of antide on T. There was no effect of antide treatment on either baseline FSH levels or FSH levels after exposure to a female. Thus, female-induced LH release from a male is GnRH dependent.
The results of these experiments do not indicate the mechanisms by which female presentation stimulates LH release in males. Subsequent experiments in this laboratory, however, indicate that a nonvolatile chemosignal from the female may trigger LH release in the male (33, 44). Exposures to short and long photoperiods alter the preference of female meadow voles for male vs. female odors (45) and alter Fos expression in the accessory olfactory bulbs of female meadow voles in response to males (46). Neural pathways involved in relaying chemosensory information to the hypothalamic-pituitary-gonadal axis have been identified in the Golden hamster. The vomeronasal organ and the accessory olfactory pathway convey chemosensory information to the medial nucleus of the amygdala, which, in turn, relays signals to the bed nucleus of stria terminalis and the medial preoptic nucleus (47, 48, 49, 50). This information may be conveyed from the medial preoptic nucleus to hypothalamic interneurons and GnRH neurons. If a similar pathway is involved in the Siberian hamster, then photoperiod may regulate the activity of this pathway or the response of the GnRH neurons to afferent signals from this pathway.
LH and T secretion as well as behavioral changes have been shown to be induced in male mice, sheep, and Golden hamsters by exposure to conspecific females (16, 17, 19). Social interactions can advance or delay the onset and end of reproductive activity in seasonally breeding animals (1, 51). The present results suggest that social stimuli may accelerate the onset of reproductive activity during the course of photostimulation in Siberian hamsters. Although the ability of members of the opposite sex to trigger LH secretion has been demonstrated in several species, there is no evidence that pheromones or social factors directly influence FSH secretion. Both LH and FSH are secreted as a consequence of stimulation of the anterior pituitary gland by hypothalamic GnRH. Our model is unique because FSH is selectively secreted on photostimulation and is not released in response to the presence of a female, whereas LH is selectively secreted upon female exposure. However, the ability of a male to release LH in response to a female is gated by photoperiod. The mechanisms for this dramatic separation of FSH and LH secretion in this species have not yet been determined.
It is known that the frequency and amplitude of GnRH stimulation can differentially regulate FSH and LH secretion in rodents, including the Siberian hamster (5, 52, 53, 54). Fast frequency GnRH pulses favor LH release, whereas slow frequency GnRH pulses favor FSH release. It is possible that changes in GnRH pulse frequency could prime the pituitary to produce increasing numbers of GnRH receptors, thus altering the response of the pituitary to GnRH. Although similar experiments have not been conducted in the Siberian hamster, treatment with GnRH does not induce the production of additional GnRH receptors on the pituitary of Golden hamsters (41). Future experiments should examine the roles of photoperiod and GnRH in the regulation of GnRH receptor number in the Siberian hamster and evaluate whether changes in GnRH receptor number or signaling could explain the observed differential patterns of FSH and LH secretion.
Several protein hormones, such as activin, follistatin, and inhibin, also contribute to the differential regulation of FSH and LH (55). Alternatively, the separate and distinct hypothalamic control of FSH and LH secretion may be mediated via two different hypothalamic releasing factors, namely, FSH-releasing factor and LH-releasing hormone. A recent study examined the potency of lamprey GnRH (L-GnRH III) as a selective FSH-releasing factor and suggests that this molecule may be conserved in mammals and may participate in the differential regulation of FSH and LH (56). The results of Exp 3 conclusively show that the release of LH, upon presentation of a female to a male, is GnRH dependent, but that FSH levels are not altered in response to treatment with a GnRH antagonist.
Glucocorticoids are also known to exert significant effects on the differential synthesis and secretion of LH and FSH. Exogenous cortisol attenuates the postgonadectomy increase in serum LH as well as suppresses exogenous GnRH-stimulated LH release in male and female rats. In contrast, cortisol increases serum FSH as well as pituitary FSH contents in laboratory rats (25, 57). Serum cortisol levels are reported to be higher in male Siberian hamsters housed in SD (average cortisol, 93.3 ng/ml) than in LD (average cortisol, 49.5 ng/ml) (23). The presence of a potential mate is known to reduce cortisol levels in male Siberian hamsters (21, 22). In view of this evidence, serum cortisol concentrations were measured to determine whether the bleeding procedure, photoperiod, or presence of a female induced a change in cortisol levels that could influence the secretion of FSH and LH. In our hands, there was no significant main effect of female or photoperiod on cortisol levels over the entire experimental period; however, there was a significant interaction between time and photoperiod, indicating that the change in cortisol level over time was dependent on photoperiod. At the final time point of the current experiment, cortisol levels in SD control males were significantly higher than cortisol levels in LD controls (Fig. 4, SD controls vs. LD controls). Our ability to identify significant treatment effects may have been limited by the sample sizes available for analysis of the cortisol data. Cortisol was measured only in those animals for which serum remained after assay of LH and FSH; thus, the sample sizes used in the analysis of the cortisol data are smaller than those used for the other analyses. The current results are consistent with previous observations that SD males have higher cortisol levels as well as more CRF mRNA than LD males (23). Also, higher cortisol levels during the first 2 h of sampling are explained by the stress associated with the sampling procedure and are expected in studies involving serial sampling. Note that cortisol levels in males in all photoperiod groups are normal (between 40 and 50 ng/ml) (21) by the third time point. Cortisol levels in males exposed to females compared with those in males not exposed to females are very similar in all photoperiod groups, indicating that cortisol levels were not affected by female exposure. Additional studies are required to evaluate whether the observed differences in serum cortisol levels contribute to the differential release of LH and FSH in this species.
To summarize, the secretion of the pituitary gonadotropins, FSH and LH, in male Siberian hamsters is differentially regulated by distinct environmental factors. Transfer of males from SD to LD elicits a rapid increase in FSH, but not LH, secretion. The release of LH can be induced shortly after the return to LD if the males are exposed to an additional environmental stimulus, which in this experiment was provided through exposure to a female. The ability of a male to release LH after female exposure is dependent on permissive photoperiod conditions and an LD pattern of melatonin secretion. The release of LH is also GnRH dependent. In conclusion, FSH and LH are differentially and selectively regulated by distinct environmental stimuli in male Siberian hamsters. These results are consistent with a model in which photoperiod acts as a predictive cue to trigger the release of FSH and the growth of the testes, which require several weeks to develop, in anticipation of improved environmental conditions. In contrast, secretion of LH and T may be withheld until much later or until additional environmental stimuli are encountered that more directly indicate the presence of proximate environmental conditions that contribute to successful reproduction. Keeping T levels low until the environment is conducive for mating has an adaptive significance of conserving metabolic energy and channeling it toward food acquisition and survival. This makes the Siberian hamster an excellent model to further examine the neuroendocrine mechanisms by which photoperiod and female exposure differentially regulate FSH and LH secretion as well as to elucidate the effects of photoperiod on the processing of social and chemosensory inputs to the GnRH neurons of the hypothalamus.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: CST, Central standard time; 16L:8D, 16 h of light, 8 h of darkness; LD, long days; SD, short days; NMDA, N-methyl-D,L-aspartate.
Received October 19, 2001.
Accepted for publication February 13, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, SD cortisol levels at the last time point were significantly different from those of the LD males.
