| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Department of Neurobiology and Physiology, Center for Reproductive Science, Northwestern University,Evanston, Illinois 60208
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Despite pronounced photoperiodic differences in circulating gonadotropin levels, consistent differences in the GnRH neuronal system have not been observed. That is, in golden and Siberian hamsters, GnRH content, GnRH-immunoreactive (-ir) cell number, or prepro-GnRH mRNA levels are sometimes enhanced on long days and other times on short days, and often no differences are detected (e.g., [2–14]). These results have led most investigators to conclude that photoperiodic differences in reproduction likely reflect changes in GnRH secretion rather than synthesis. Recently, however, we observed a transient increase in prepro-GnRH mRNA levels during the transition from short to long days in male Siberian hamsters [6]. The number of cells with detectable levels of prepro-GnRH mRNA, as measured by in situ hybridization, was significantly but transiently increased on the morning of the second day after transfer. Thus, a change in GnRH synthesis may be important during transitions between reproductive conditions, but not necessary to maintain differences in reproduction in animals held chronically in different photoperiods.
Within 3–5 days after transfer from short to long days, Siberian hamsters exhibit a significant GnRH-dependent increase in serum FSH levels [4, 15–17]. Therefore, increases in GnRH synthesis may occur in anticipation of, in response to, or in synchrony with changes in GnRH release shortly after transfer to long days. In the present study, we first determined whether or not GnRH protein content in brain homogenates increases following long-day transfer with a time course similar to that described previously for prepro-GnRH mRNA levels [6]. Second, to determine whether long day-induced increases in prepro-GnRH mRNA and protein levels reflect a transient increase in the number of neurons producing GnRH or an increase in the amount of GnRH produced by an otherwise stable population of GnRH neurons, we used immunocytochemistry to identify and quantify the number of GnRH neurons on the morning of the second long day.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Siberian hamsters (Phodopus sungorus) were born and raised in long days (16L:8D; lights-on 0500 h CST) in an animal facility at Northwestern University (Evanston, IL). Animals were weaned at 18 days of age and thereafter group-housed with same-sex siblings. Only male hamsters were used. Food and water were provided ad libitum.
Procedures
Photoperiodic treatments proceeded as described previously [6]. Briefly, after weaning at Day 18, hamsters were transferred to short days (6L:18D; lights-on 0900 h CST). Control animals were maintained on long days. After 4 wk on short days, half of the short-day animals were transferred back to long days (16L:8D) by advancing light onset by 4 h and delaying lights-off by 6 h. Two hours after lights were illuminated on the morning of the second long day, hamsters were killed with CO2. Age-matched short-day and long-day controls were killed 2 h after light onset on their respective photoperiods. In experiment 1, brains were removed immediately and frozen on dry ice. In experiment 2, animals were perfused by hand with 10 ml of 0.9% NaCl followed by 10 ml of cold 4% buffered paraformaldehyde (pH 8.5). Brains were removed, postfixed overnight at 4°C, and then cryoprotected in 30% sucrose in 0.1 M PBS (pH 7.2) prior to being frozen on dry ice.
Experiment 1: GnRH Content
GnRH protein content was measured in the preoptic area-mediobasal hypothalamus (POA-MBH) by RIA. Frozen brains were dissected on dry ice by making lateral cuts along the hypothalamic sulci, 2 mm anterior to the optic chiasm, 1 mm anterior to the mammillary bodies, and 2 mm dorsal to the ventral surface of the brain. Each tissue sample was homogenized in 1 ml of 0.02 N HCl/80% ethanol with a Polytron homogenizer (Brinkman, Westbury, NY) and then centrifuged at 2800 rpm for 25 min. The supernatant was transferred to a new tube and vacuum desiccated overnight in a speed vac. The pellet was resuspended in 1 ml of 0.01 M PBS/0.1% gelatin. GnRH levels were measured in triplicate in an RIA validated for use in Siberian hamsters using the EL-14 primary antibody [6]. The assay was sensitive to 0.2 pg/tube, and the intraassay coefficient of variation was less than 15%. Data were normalized for differences in the mass of the tissue dissection and are reported as picograms of GnRH per milligram tissue wet weight.
Experiment 2: GnRH Immunocytochemistry
GnRH-containing neurons were identified using immunocytochemistry. Perfused frozen brains were sectioned coronally on a cryostat (-20°C). Fifty-micrometer sections from anterior to the joining of the corpus callosum through the median eminence were saved into 0.1 M PBS (pH 7.2). Sections were then consecutively incubated in 0.5% H2O2, 10% normal goat serum in 0.3% PBST (0.3% Triton X-100 in 0.1 M PBS), and the LR-1 GnRH antibody (kindly provided by Dr. R. Benoit, The Montreal General Hospital, Montreal, PQ, Canada) diluted to 1:40 000 in 0.3% PBST. Sections were washed twice in PBS for 10 min each after tissue sectioning and between the H2O2 and normal goat serum steps. After 48 h in the GnRH antibody at 4°C, sections were washed in 0.1% PBST, incubated for 1 h in goat anti-rabbit IgG (1:250 dilution; Vector Labs., Burlingame, CA), washed again in 0.1% PBST, and incubated for 1 h in 1:250 avidin-biotin-horseradish peroxidase. After washes in PBS, the sections were incubated for 4 min in 330 µg/ml 3,3'-diaminobenzidine and 0.004% H2O2. Sections were mounted on gelatin-coated slides. GnRH-ir cells were examined by brightfield microscopy (x200). We counted and determined the phenotype (unipolar or bipolar) (see [18]) and localization of all GnRH-ir neurons from 200 µm anterior to the joining of the corpus callosum through the full extent of the median eminence. In addition, we performed a more extensive analysis on the GnRH-ir cells within the diagonal band of Broca/medial preoptic area (DBB/MPOA). These regions have been shown previously to contain the largest number of GnRH-ir cells and contain the population of cells showing plasticity in this species (e.g., [3, 18, 19]).
To permit comparison of our results with those of previous studies, in a final experiment we performed the immunocytochemical procedure on an additional 4 brains from 7- to 11-wk-old male hamsters housed on long days from birth. Brain collection and tissue processing proceeded as described above except that alternate sections were incubated in the primary antibody at a concentration of either 1:20 000 or 1:40 000. Total GnRH-ir neuron number was quantified for each dilution and then multiplied by 2 to derive an estimate of total cells per brain. We had determined previously that in brains stained completely with the 1:40 000 dilution, summing GnRH-ir cells from every other section and multiplying by 2 yields an estimate of total cell number equivalent to summing cells from all sections (data not shown).
All measures were made by an observer blind to the condition of the animals or the dilution of the antibody used.
Statistical Analyses
Comparisons between the three treatment groups were made with one-way ANOVAs. Post hoc comparisons were made with Fisher's Least Significant Difference procedure. Differences in GnRH-ir cell number estimates between the two dilutions of the primary antibody were assessed with a paired t-test. Statistical significance was assessed relative to p < 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
GnRH protein content in tissue blocks containing the POA-MBH was significantly elevated in male hamsters transferred from short to long days relative to both the long- and short-day control groups (F(2,15) = 4.5, p < 0.03; see Fig. 1). The latter two groups did not differ. Body mass and paired testes mass were significantly greater in long-day controls than in the short-day controls or in animals transferred to long days (both F > 7.4 and p < 0.01; see Table 1). The latter two groups did not differ from one another.
|
|
Experiment 2: GnRH Immunocytochemistry
GnRH-ir cell number from anterior to the joining of the corpus callosum through the median eminence was affected significantly by photoperiod (F(2, 23) = 5.9, p < 0.01; Fig. 2). Hamsters that were transferred from short to long days and the short-day controls had equivalent numbers of cells; however, both of these groups had significantly more GnRH-ir neurons than long-day controls. As described previously, more than half of the cells were located within the DBB/MPOA (e.g., [18]). We examined more thoroughly the effect of photoperiod on the number of unipolar and bipolar cells within these regions using a two-way, repeated measures ANOVA. Overall, the long-day controls had fewer GnRH-ir cells within the DBB/MPOA than the other two groups, which did not differ from one another in this respect (F(2, 23) = 6.2, p < 0.01; Fig. 3). In all three groups, there were more bipolar than unipolar neurons in the DBB/MPOA (F(1, 23) = 19.4, p < 0.001; Fig. 3), but the interaction between cell type and photoperiodic condition was not significant. Finally, as expected, long-day controls weighed more and had larger testes than either of the other two groups (both F > 24.1 and p < 0.0001; Table 1).
|
|
The total number of GnRH-ir cells measured with the two different dilutions of the primary antibody did not differ significantly (t(3) = 0.8, p > 0.5). When multiplied by 2, the estimates of total GnRH-ir cell number per brain for the 1:20 000 (mean = 244) and 1:40 000 (mean = 226) dilutions were similar to values for the long-day control group above (mean = 229).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Given the temporal relationship between changes in GnRH content and increases in pituitary gonadotropin release after long-day transfer, the rapid increase in GnRH synthesis may occur in anticipation of augmented release shortly thereafter. That is, while GnRH content is increased on the morning of the second long day, GnRH-dependent increases in serum FSH levels are observed only after 3–5 long days [4, 15–17]. Other data indicate that GnRH secretion may be altered within the first long day. We, and others, have observed significant increases in pituitary FSH content after 1 long day [16, 20]. Therefore, the observed increase in GnRH synthesis may compensate for increases in GnRH secretion that occur within the first long day. Support for the latter hypothesis comes from work in golden hamsters, in which GnRH content is decreased and serum FSH increased after 1 long day [11]. However, to determine unequivocally whether changes in GnRH content reflect an anticipatory, compensatory, or perhaps synchronous mechanism, it will be necessary to measure GnRH secretion directly from the pituitary-portal vessels during the transition from short to long days. Unfortunately, the small size of Siberian hamsters has made it difficult to apply the push-pull perfusion or microdialysis approaches that have been used successfully in larger animals (e.g., [21–24]).
While the increased GnRH content observed after 1 day does not represent an increase in the number of GnRH-ir neurons, we did observe a greater number of GnRH-ir neurons in short-day than in long-day hamsters. This result is consistent with reports in other photoperiodic rodents, but differs from previous findings for Siberian hamsters. In deer mice (Peromyscus maniculatus), white-footed mice (P. leucopus), and prairie voles (Microtus ochrogaster), the number of detectable GnRH-ir neurons is increased in short-day relative to long-day animals [25–27]. In golden hamsters, photoperiodic condition does not affect GnRH-ir cell number (cf. [7, 9]), but short-day animals have larger GnRH-ir cell bodies than long-day animals [13]. These results suggest that decreased GnRH secretion on short days results in an accumulation of the peptide within cell bodies that may increase the intensity of staining and therefore facilitate detection with immunocytochemical techniques (e.g., [7, 9, 13, 25–27]).
While our immunocytochemistry results are consistent with this hypothesis, we do not detect an overall difference in GnRH content in the brains of long- and short-day animals. In two previous studies, however, we did observe increased GnRH content in the MBH of long-day relative to short-day adult males [4, 10]. The tissue dissections in these studies, however, were posterior to the optic chiasm and therefore likely did not include the majority of GnRH cell bodies. The dissections in the present study included the GnRH cell bodies anterior to the optic chiasm in addition to the peptide stored in the GnRH terminals of the median eminence. The results of all these studies have led us to postulate that the photoperiodic differences in GnRH-ir cell number may reflect a photoperiodic difference in the distribution of GnRH within the brain rather than a difference in the releasable pool of GnRH. Because of increased GnRH secretion in long days, GnRH produced within cell bodies may be rapidly transported to the terminals in the median eminence to facilitate release. This would lead to an increase in GnRH content within the MBH in long-day relative to short-day hamsters [4, 10]. This would also effectively decrease GnRH content within cell bodies and thereby decrease their detection by immunocytochemistry (present study). Future studies examining GnRH content from microdissected brain regions (e.g., MBH vs. DBB/MPOA) will better address this hypothesis. Interestingly, GnRH content in the MBH of both white-footed and deer mice is elevated in short days compared to long days (e.g., [28, 29]). Therefore, alternative hypotheses may require exploration.
The reduction in GnRH-ir cell number in long days is surprising in light of previous reports in Siberian hamsters. Photoperiodic effects on cell number have been observed during puberty, but not in older animals. During development, GnRH-ir cell number increases between 15 and 25 days of age [30]. This increase can be delayed by exposure to short days or treatment with melatonin [19]. In hamsters maintained from birth on long or short days, there are no differences in GnRH-ir cell number at 40, 60, or 90 days of age ([14]; see also [2, 3]). Thus, by 40 days of age, photoperiodic differences in GnRH-ir cell number no longer persist in Siberian hamsters. Given these results, we did not anticipate the photoperiodic difference in GnRH-ir cell number we observed in the 48- to 50-day-old animals used in our study. The numbers of GnRH-ir cells in the short-day controls and animals transferred to long days are comparable to numbers reported previously (300–350 total per brain) (e.g., [18, 19]); but we detected significantly fewer cells in long-day animals (~230 per brain), particularly within the DBB/MPOA. There are at least two explanations for the differences in results.
First, while the previous and present studies used the LR-1 GnRH antibody, we used a 2-fold lower concentration of the primary antibody (1:20 000 vs. 1:40 000). As discussed above, there may be a long day-associated decrease in GnRH within cell bodies. If this is the case, then the lower titer of antibody we used may have failed to detect cells containing small amounts of GnRH. To address this issue, we collected additional brains from long-day control animals and processed alternate sections from each brain with the 1:20 000 or 1:40 000 dilution of the primary antibody. We then counted the cells for each dilution and multiplied by 2 to generate an estimate of total cell number per brain. The two dilutions indicated similar cell number, and both numbers were similar to the number observed for long-day animals in our initial experiment (i.e., ~220–240 cells per brain). Therefore, we do not believe that the differences in results between the various studies can be accounted for by a difference in antibody concentration. In addition, the similarity between cell number estimates for our short-day group and those of previous results also indicates that differences in assay conditions are likely not involved.
Second, in Siberian hamsters, photoperiodic differences in circulating LH are highly variable between and within studies (e.g., [16, 17, 31]). In the previous study examining photoperiodic differences in GnRH-ir cell number, no differences in circulating LH were observed and only small differences in FSH were detected [14]. Therefore, it is possible that the animals in different photoperiodic conditions did not have marked differences in GnRH secretion. If GnRH-ir cell number reflects GnRH secretory activity, then this may account for the lack of photoperiodic differences in GnRH-ir cell number reported previously [14]. We did not measure LH levels in the present study, but it is possible that LH levels were increased in the long-day animals and hence we detected a smaller number of GnRH-ir neurons. Future studies are needed to establish a relationship, if any, between circulating LH levels and the number of detectable GnRH-ir cells in Siberian hamsters. It should be noted, however, that a direct relationship between circulating LH levels and GnRH-ir cell number is not always apparent in all species (e.g., [26]).
In summary, during the transition from short to long days, male Siberian hamsters show a significant increase in brain content of GnRH protein on the morning of the second long day. There is no change in GnRH-ir cell number at this time. Therefore, the increase in protein synthesis appears to occur within a stable population of GnRH-producing cells. The function of this increase in GnRH production is unknown, but it may occur in anticipation of or compensation for increased GnRH secretion occurring in the first few days following long-day transfer. Thus, increased GnRH production may be an essential part of the photoperiodic mechanism driving transitions in reproductive physiology. In contrast to previous reports in Siberian hamsters, we also observed a marked reduction in GnRH-ir cell number in long-day compared to short-day animals. As argued for other species, this difference may reflect augmented GnRH secretion on long days. Additional studies are needed to clarify the disparity in results between this and previous studies in Siberian hamsters.
As we have discussed previously [32], our ability to describe the cascade of physiological, cellular, and molecular events mediating photoperiod-induced changes in reproductive function has been impeded by the relatively slow response of various reproductive parameters to changes in day length. For example, changes in circulating gonadotropin levels and gonadal state require days to weeks to be manifested. In the model system described in this and previous studies (e.g., [6, 20]), we observe changes in the GnRH neuronal system and in pituitary FSH content within 26 h of the transfer from short to long days. These rapid responses define a narrow temporal window in which to focus investigation into the necessary and sufficient mechanisms for photoperiodic changes in reproduction. We strongly believe that the application of molecular biological approaches to this model system will provide the means to identify the genes and their protein products underlying the photoperiodic response.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: Daniel J. Bernard, Department of Neurobiology and Physiology, Center for Reproductive Science, Northwestern University, 2153 N. Campus Drive, Evanston, IL 60208. FAX: 847 467 4065; dbernard{at}nwu.edu
Accepted: September 8, 1998.
Received: May 18, 1998.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. J. Paul, J. H. Park, T. H. Horton, M. I. Alvarez, M. K. Burke, N. J. Place, and I. Zucker Photoperiodic Regulation of Compensatory Testicular Hypertrophy in Hamsters Biol Reprod, August 1, 2006; 75(2): 261 - 269. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Avigdor, S. D. Sullivan, and P. D. Heideman Response to selection for photoperiod responsiveness on the density and location of mature GnRH-releasing neurons Am J Physiol Regulatory Integrative Comp Physiol, May 1, 2005; 288(5): R1226 - R1236. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Barrett, A. W. Ross, A. Balik, P. A. Littlewood, J. G. Mercer, K. M. Moar, T. Sallmen, J. Kaslin, P. Panula, S. Schuhler, et al. Photoperiodic Regulation of Histamine H3 Receptor and VGF Messenger Ribonucleic Acid in the Arcuate Nucleus of the Siberian Hamster Endocrinology, April 1, 2005; 146(4): 1930 - 1939. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Anand, S. Losee-Olson, F. W. Turek, and T. H. Horton Differential Regulation of Luteinizing Hormone and Follicle-Stimulating Hormone in Male Siberian Hamsters by Exposure to Females and Photoperiod Endocrinology, June 1, 2002; 143(6): 2178 - 2188. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. R. Pak, G. R. Lynch, and P.-S. Tsai Testosterone and Estrogen Act via Different Pathways to Inhibit Puberty in the Male Siberian Hamster (Phodopus sungorus) Endocrinology, August 1, 2001; 142(8): 3309 - 3316. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
LD) had significantly more GnRH in the POA-MBH than either the long (LD)- or short-day (SD) controls, which did not differ from one another. * p < 0.03.
