Department of Neurobiology and Physiology, Northwestern University, Evanston, Illinois. A.M. Wolfe, F.W. Turek, and J.E.

BIOLOGY OF REPRODUCTION 53, 724-731 (1995) Blockade of Singular Follicle-Stimulating Hormone Secretion and Testicular Development in Photostimulated Djungarian Hamsters (Phodopus sungorus) by a Gonadotropin-Releasing Hormone Antagonist' A.M. Wolfe, F.W. Turek, and J.E. Levine 2 Department of Neurobiology and Physiology, Northwestern University, Evanston, Illinois ABSTRACT Photostimulated male Djungarian hamsters following placement in a long-day photoperiod exhibit a characteristic rise in serum FSH levels that occurs in the absence of a simultaneous rise in LH levels. It is not known whether this singular FSH secretion is dependent upon a differential responsiveness of the gonadotrophsto the pattern of pulsatile GnRH release or is instead driven by a GnRH-independent mechanism. We have assessed the GnRH dependence of this singular FSH secretion by testing the ability of a potent GnRH antagonist (GnRHa: WY-45760) to block FSH and testicular responses to photostimulation. Photoinhibited hamsters were transferred from a shortday (6L:18D) to a long-day photoperiod (16L:8D). Hamsters received two daily injections of a GnRH antagonist or vehicle (VEH). After 0 (short day), 3, 5, 10, 30, or 40 days the hamsters were killed; plasma was assayed for FSH, LH, and testosterone (T), and testes weights were recorded. Testes were sectioned and analyzed for tubular development. In VEH-treated animals, testicular weights increased after photostimulation, reaching mean values of 514 mg by 30 days. Treatment with GnRHa resulted in a significant (p < 0.01) attenuation of testicular growth after 30 days of photostimulation (mean testes weight = 110.1 mg). In VEH-treated hamsters there was a rapid increase in FSH levels after photostimulation that became significant by 5 days and peaked at 10 days. In the GnRHa-treated group, however, these FSH increments were completely blocked at 5 days and significantly reduced at 10 days compared to the values in the corresponding VEH-treated groups. Serum LH and T levels remained relatively low following photostimulation in both groups. These data demonstrate that, in Djungarian hamsters, the singular rise in FSH secretion observed after the transfer from short to long days is primarily dependent upon endogenous GnRH stimulation. This GnRH-dependent FSH rise is, in turn, necessary for normal testicular development following photostimulation. Thus alterations in FSH-releasing factor activity or gonadal hormone feedback need not be invoked to explain the differential effects of photostimulation on LH and FSH release during the early stages of photic-induced changes in neuroendocrine-gonadal activity. We hypothesize that photostimulation of hamsters induces a change in the pattern of endogenous GnRH neurosecretion, which in turn conveys FSH-specific signals for the secretion of the gonadotropin. INTRODUCTION The male Djungarian hamster (Phodopus sungorus) is a seasonal breeder that, when transferred from an inhibitory short-day photoperiod to a stimulatory long-day environment, exhibits rapid testicular growth and development. Photostimulation of the reproductive axis is characterized by a rapid increase in FSH secretion, which supports testicular growth [1], followed weeks later by increases in LH levels [2]. Changes in day length alter the duration of the nighttime release of pineal melatonin, and there is now substantial evidence to indicate that changes in the duration of melatonin release induce changes in neuroendocrine-gonadal activity [3]. Little is known, however, about the mechanisms that mediate the effects of melatonin on the hypothalamic-pituitary-gonadal axis. The secretions of the two gonadotropins are thought to be primarily governed by the same hypothalamic releasing factor, GnRH, and thus it is unclear how photostimulation Accepted May 8, 1995. Received December 9, 1994. 'This work was supported in part by N.I.H. grants P01 HD21921, R01 HD20677, R01 HD0988j5, R04 HD00879 (R.C.D.A. toj.e.l.), and P30 HD28048. Presented in part at the 27th annual meeting of the Society for the Study of Reproduction, Ann Arbor, MI; 1994 (abstract 219). 2Correspondence: Jon E. Levine, I)epartment of Neurobiology and Physiology, Northwestern University, 2153 North Campus Drive, Evanston, IL 60208-3520. FAX: (708) 491-5211. can induce a secretion of FSH in hamsters in the absence of any changes in LH secretion. One explanation for this phenomenon is that FSH can be additionally stimulated or disinhibited by some neural or gonadal, GnRH-independent, mechanism. For example, the neurosecretion of a distinct FSH-releasing factor (FSH-RF) that possesses little or no LH-releasing activity [4-7] may increase in response to a stimulatory photoperiod. Alternatively, photostimulation could induce acute changes in gonadal feedback influences, in turn allowing for the singular release of FSH. A drop in inhibin secretion, for example, has been implicated in the rapid differential release of FSH that is observed after ovariectomy [81 or during the secondary surge of FSH on the morning of estrus in female rats [9, 10]. An alternative hypothesis is that a GnRH-dependent mechanism may also mediate the singular FSH response to photostimulation. In virtually all species studied, GnRH release has been found to be pulsatile [11-14]; the frequency and amplitude of GnRH pulsatility is regulated under a variety of physiological circumstances and by a host of neuroendocrine regulators. Moreover, the results from GnRH infusion studies in monkeys [15], hamsters [16], sheep [17], and rats [18-20] have clearly demonstrated that the frequency of pulsatile release of GnRH can differentially affect the release and synthesis of the pituitary gonadotropins, with slower frequencies favoring FSH secretion and faster 724

FSH RELEASE IS DEPENDENT ON GnRH 725 frequencies favoring LH secretion. Thus, FSH and LH biosynthesis and secretion may be differentially "tuned" to optimal GnRH pulse frequencies, and the divergent gonadotropin secretion seen under some physiological circumstances may be directed through changes in the frequency of endogenous GnRH stimulation. In the present study, we sought to determine whether stimulation of the singular secretion of FSH during the early stages of photostimulation in Djungarian hamsters occurs through a GnRH-independent or a GnRH-dependent mechanism. To address this question, we assessed the ability of a potent GnRH antagonist (GnRHa; WY45760) to block photic-induced FSH secretion and testicular growth. Our results clearly demonstrate that the rapid, singular secretion of FSH that occurs following photostimulation in Djungarian hamsters is predominantly dependent upon endogenous GnRH neurosecretion. Animals MATERIAL AND METHODS All animal protocols were used with the approval of the Northwestern University Animal Care and Use Committee and in compliance with NIH guidelines for care and use of experimental animals. Djungarian hamsters (Phodopus sungorus) were born in the Northwestern University breeding colony and raised under long-day (16L:8D, lights-on at 0500 h) conditions until weaned at 18 days of age. The animals were derived from original stock provided by Dr. Bruce Goldman (Worcester Foundation for Experimental Biology, Shrewsbury, MA). The hamsters were fed Teklad (Madison, WI) hamster chow and tap water ad libitum. After weaning, animals were separated by sex, housed three to five per cage, and maintained in a short-day (SD) photoperiod (6L: 18D, lights-on at 0900 h) until selection for experiments. In preliminary tests of the effectiveness of GnRH antagonist treatment, retired female breeders were used; otherwise, all experimental animal subjects were juvenile males maintained in 6L:18D for 30 + 2 days before the beginning of the experiment. All male hamsters selected for the experiment exhibited no discernable testicular development as assayed by palpation. Preliminary Test of the Effectiveness of the GnRHa A preliminary study, using ovariectomized (OVX) female Djungarian hamsters, was performed to examine the effectiveness of the GnRHa in reducing FSH and/or LH levels. Adult female hamsters were bilaterally OVX under light pentobarbital anesthesia (60 mg/kg). At 2000 h on Day 7 after OVX, the animals received s.c. injections of either 25 jg of the potent GnRH antagonist, WY-45760 ([Ac-,(2)-D - Nal',4-F-D-Phe2,D-Trp3,D-Arg6]GnRH; Wyeth Laboratories, Philadelphia, PA; n = 5 animals) or an equal volume of sesame oil alone (n = 5 animals). This antagonist has previously been shown to completely abolish LH secretion in cycling female rats [21]. On the following morning at 0800 h the animals received a second injection of either GnRHa or oil. At 1500 h on the same day the animals were killed by exsanguination under light methoxyflurane anesthesia. The blood was collected by cardiac puncture and dispersed into microcentrifuge tubes containing 10 [l of 50 IU/ml heparin. Samples were centrifuged and plasma was collected and stored at - 70 0 C until RIA could be performed to measure levels of LH, FSH, and testosterone (T). Experimental Groups Hamsters were selected for experiments as described and were transferred from short- to long-day photoperiods. Animals received injections of GnRHa or vehicle (VEH) twice daily, at 2000 h and 0800 h. The animals were killed, at 1500 h as above, on Days 3, 5, 10, 30, and 40 (VEH group only) following the beginning of treatment. One group of animals (Day 0) were killed prior to the transfer to long days. Hamsters were anesthetized with methoxyflurane and killed by exsanguination. Blood was collected and assayed as described below. The testes were removed, weighed, and stored for histological analysis. Testis Histology Testes were hemisected and then fixed in Bouin's solution for 2-4 days. One hemitestis from each animal was serially dehydrated and embedded in paraffin. Serial sections, 10 gim, were cut on a rotary microtome and were stained in hematoxylin and eosin through use of standard procedures. Testis sections were examined under the light microscope for the presence of spermatocytes and spermatozoa and for the degree of tubular organization. RIA Immunoreactive LH and FSH levels were determined by RIA using material provided by the NIDDK (Baltimore, MD). The LH standard was LH-RP-3, and the antiserum was LH- S-11. The level of sensitivity of the assay was 143 pg/ml with an intraassay coefficient of variation (CV) of 12.2% and an interassay variation of 8.1%. The FSH standard was FSH RP- 2, and the antiserum was FSH-S-11. The level of sensitivity of the assay was 1.43 ng/ml with a CV of 13.0% and an interassay variation of 15.6%. The RIA for T was performed with a kit from ICN Biomedical (Costa Mesa, CA). The level of sensitivity of the assay was 100 pg/ml with a CV of 8.9% and an interassay variation of 15%. Data Analysis Testes weights, LH levels, and FSH levels were averaged within a group at each time point. For each measure within a treatment group, data were examined by ANOVA. Differ-

726 WOLFE ET AL. -3 1 5 4 0 Oil Antag Oil Antag 18 15 12 9 c ' N FIG. 1. Demonstration of the effectiveness of GnRHa in attenuating LH and FSH levels in OVX female Djungarian hamsters. The OVX hamsters received VEH or GnRHa (25 pg) at 2000 h and 0800 h, and blood was obtained at 1500 h. Values are mean + SEM. ** p < 0.01 vs. oil control, * p < 0.05 vs. oil control. ences among individual points were assessed by post hoc comparison with Tukey's protected t-test. Comparison between GnRHa and VEH groups was performed through use of two-way ANOVA for 3-day through 30-day groups, and differences were again assessed by post hoc comparison 2-8 6-0 - _ Day 0 f[ Oil Treatment d Antag. Treatment 1' \ d 61 ii 0 3 5 10 30 40 Days After Shift to Long Days FIG. 2. Effects of GnRHa on plasma FSH levels in male Djungarian hamsters at various times following photostimulation. Photostimulated hamsters were killed 0 (short day), 3, 5, 10, 30, and 40 days after placement in long days (16L:8D, lightson at 0500 h) at 1500 h. Animals received an i.p. injection of VEH or GnRHa (25 pg) each day at 2000 h and 0800 h. Values are mean + SEM. a = p < 0.05 different from same-day VEH; b = p < 0.05 different from Day 0; d = p < 0.01 different from Day 0. alid 3 0 with Tukey's protected t-test. Since T levels in the antagonist-treated group were all below the sensitivity of our assay, the nonparametric Kruskal-Wallis test was used to analyze T data. Differences were considered significant if p < 0.05. Preliminary Test of GnRHa RESULTS Plasma LH and FSH levels in VEH- and GnRHa-treated, OVX females are presented in Figure 1. Treatment with GnRHa (25 gg/injection twice daily) resulted in a near-total suppression of LH levels compared to those in VEH-treated hamsters (p < 0.01). Plasma FSH levels were also significantly reduced by treatment with GnRHa (p < 0.05) to approximately 60% of the values in the VEH-treated animals. Since these results closely paralleled those previously obtained in rats by means of GnRH antagonists [21, 22] or antiserum to GnRH [5], this dosage of GnRHa was judged to be appropriate for use in the subsequent photostimulation experiment. Effects of GnRHa in Photostimulated Male Hamsters Plasma FSH levels. Shown in Figure 2 are plasma FSH levels of photostimulated males treated with VEH or GnRHa. In VEH-treated animals, levels of FSH were increased significantly after 5 days of photostimulation (p < 0.05 vs. Day 0), peaking at Day 10 (p < 0.01 vs. Day 0). FSH levels had declined by 30 days and had returned to near baseline levels after 40 days. The magnitude and timing of the FSH response to photostimulation were in close agreement with previous results in juvenile Djungarian hamsters in this colony [1]. Serum FSH levels were lower, although not significantly, in the GnRHa-treated group after exposure to a long-day light cycle for 5 days. After 10 days of photostimulation, FSH levels were significantly lower in the GnRHa-treated animals than in the VEH-treated animals. FSH levels in the GnRHa-treated group were not significantly increased at 5 days when compared to values for the Day 0 group, but had significantly increased by 10 days (p < 0.01) and remained significantly higher than the Day 0 group values after 30 days of photostimulation (p < 0.01). Plasma LHand Tlevels. As shown in Figure 3a, plasma LH levels were generally close to the limit of sensitivity of our assay (143 pg/ml at dilution used) in both antagonisttreated and control animals at all time points measured. LH levels in VEH-treated or GnRHa-treated animals were not found to have significantly increased from the Day 0 group values at any time. LH levels in the VEH-treated animals at 30 days were, however, significantly higher than levels seen in GnRHa-treated animals (p < 0.05). The LH levels observed in this study were similar to those reported in a previous study [16], although in this experiment it was further

FSH RELEASE IS DEPENDENT ON GnRH 727 documented that LH levels do not exhibit an increase even after 40 days of photostimulation in juvenile hamsters. Plasma T levels also remained low in all groups (Fig. 3b), although levels of T were found to have increased slightly from Day 0 levels in the 30-day VEH-treated group. There was a significant difference between T levels in VEH-treated animals and GnRHa-treated animals only in the 30-day groups. In GnRHa-treated animals, T levels were not above the limit of sensitivity of our assay at any time point measured (0.1 ng/ml at dilution used). T levels in the VEH-treated animals peaked at just over 0.5 ng/ml after 30 days of photostimulation. These values are low when compared to levels observed in photostimulated adult hamsters (about 1.5 ng/ml [231). Paired testes weights and spermatogenesis. Figure 4 shows the paired testes weights of VEH- and GnRHa-treated animals. The VEH-treated animals had paired testes weights that increased throughout the time examined, reaching a mean value of over 700 mg after 40 days of photostimulation-weights comparable to those reported by others [23, 24]. After 10 days of photostimulation, testes weights were significantly larger than those of the Day 0 group (p < 0.05). Treatment with GnRHa reduced testicular growth in response to photostimulation, as the testes weights in the GnRHa-treated group were significantly attenuated after 30 days of photostimulation vs. values for the corresponding VEH-treated group (i.e., an 86% attenuation; p < 0.01). Histological examination of the testes showed a clear correlation between testis size and degree of testicular development in both the VEH and GnRHa experimental groups. In the Day 0 group, there was minimal luminal space in the tubule as well as an absence of organization of the germinal epithelium within the tubule (Fig. 5a). Spermatogonia were present, but spermatocytes were rarely if ever observed. Little sign of testicular development was observed in the VEH-treated animals after three long days, while a slight increase in the lumen size of the tubule, as well as the appearance of scattered spermatocytes within the tubules of some animals, was detected after five long days (data not shown). After 10 days of photostimulation, there was a pronounced increase in the degree of organization of the tubules, and spermatogonia and leptotene and pachytene spermatocytes were observed (Fig. 5c). VEHtreated animals exhibited nearly mature tubular organization with large lumen and mature spermatozoa after 30 long days (Fig. 5e). A further maturation of the testis as well as complete organization of the germinal epithelium was observed in VEH-treated animals after 40 days of photostimulation (Fig. 5f). During the first five long days, there was minimal development of the testes in the GnRHa-treated animals beyond that seen in the Day 0 animals. Thus, at five days it appeared that the GnRHa may have inhibited tubular organization and formation of spermatocytes (data not shown). By 10 E E "I bt 1.0 (a) I Day 0 Oil Treatment 0.8 - i Antag. Treatment 0.5 - I 0.3 - n ua.u 1.0 0.8-0.5 0.3 n, n CT II NI pi NH 0 3 5 10 30 Days After Shift to Long Days FIG. 3. Effects of GnRHa on plasma LH and T levels from male Djungarian hamsters at various times following photostimulation. a) LH levels in VEH groups did not increase significantly above Day 0 levels, and LH levels in GnRHa groups were consistently below the limit of detectability of the assay (0.143 ng/ml, represented by dashed line). b) T levels in GnRHa group never exceeded the limit of detectability of the assay (0.1 ng/ml at dilution used, represented by dashed line). Values are mean + SEM. a = p < 0.05 different from same-day VEH; b = p < 0.05 different from Day 0; c = p < 0.01 different from same-day VEH. o-u Q--.3 _ Day 0 o DUU ' 400 a) U.U 0 200 0 - - I.- h i i ---------- _.- (b).x ia lx b T ik.. _... &1 0 a 1- d" t E N mtlull [ Antag. Treatment u u 5 10 30 40 Days After Shift to Long Days FIG. 4. Effects of GnRHa on paired testes weights from Djungarian hamsters at various times following photostimulation. Values are mean + SEM. b = p < 0.05 different from Day 0; c = p < 0.01 different from same-day VEH; d = p < 0.01 different from Day 0. F; b d 1KSI K. b,c - I q 40 ~

728 WOLFE ET AL. FIG. 5. Representative photomicrographs of hematoxylin- and eosin-stained testis sections from GnRHa- and VEH-treated animals from selected groups. Shown are sections from (a) Day 0 group, (b) 10-day GnRHa-treated group, (c) 10-day VEH-treated group, (d) 30-day GnRHa-treated group, (e) 30-day VEH-treated group, and (f) 40-day VEH-treated group. All photomicrographs were taken at x 200. Scale bar = 0.1 mm for all sections.

FSH RELEASE IS DEPENDENT ON GnRH 729 days, however, the GnRHa-treated animals showed testicular development virtually indistinguishable from that seen in VEH-treated animals at the same time point (Fig. 5b). By 30 days there was a clear difference between the two groups. The 30-day GnRHa-treated group had a slightly greater degree of tubular organization and more spermatocytes (Fig. 5c) than the 10-day animals but had not nearly the testicular maturity of 30-day controls. No spermatid or spermatozoa were ever seen in the 30-day GnRHa-treated animals. DISCUSSION Secretion of the gonadotropins FSH and LH can be differentially regulated under a variety of physiological and experimental circumstances [20, 22, 25-27]. The singular secretion of FSH, for example, has been shown to occur during the secondary FSH surge in female rats [28] following ovariectomy [29], during puberty [30] and early pregnancy [29], in the presence of hyperprolactinemia [311, and, as confirmed in the present study, following exposure of photoinhibited hamsters to a long-day photoperiod. There are likely several different neural and endocrine mechanisms that may either stimulate or permit singular FSH secretion in these different situations; in general these mechanisms may be classified as either dependent on or independent of the neurosecretion of GnRH. In this study we have tested the hypothesis that the distinct secretion of FSH that occurs after photostimulation of Djungarian hamsters is dependent, at least in part, upon endogenous GnRH stimulation. Our results suggest that the photic-induced increment in FSH secretion is primarily dependent upon GnRH stimulation. Although our study demonstrates an important role for GnRH in photic-induced FSH secretion, the exact nature of this role can only be inferred from these data. It is possible, for instance, that the role of the decapeptide is a permissive one, in which sufficient GnRH stimulation need only occur, not necessarily alteration, to allow stimulation by a putative FSH-RF [7] or to allow alterations in gonadal hormone feedback [32, 33]. However, we regard the former possibility as unlikely since after three days of photostimulation, FSH levels in GnRHa-treated animals were not significantly different from those seen in VEH-treated animals, indicating that a basal, GnRH-independent, component of FSH release exists and that GnRHa attenuated only the subsequent increment in FSH secretion. This suggests that the increased FSH secretion seen in the VEH-treated animals at 5 and 10 days is driven by a dynamic, photic-induced change in GnRH release. We also regard it as unlikely that GnRH need only occur to permit altered gonadal feedback to stimulate FSH secretion, since in a related species, the golden hamster, inhibin levels do not drop significantly during photostimulated FSH secretion [34]. There is likewise little support for the hypothesis that selective gonadal feedback suppression of LH at the level of the gonadotroph permits photic-induced FSH secretion in the absence of LH secretion, since LH responses to GnRH have been shown to be unaltered by photostimulation [35] and because levels of T (the principal negative feedback regulator of LH) are extremely low throughout this entire period of singular FSH secretion. It is most likely the case, then, that photic-induced, singular, FSH secretion depends upon some dynamic change in the pattern of endogenous GnRH release, which in turn differentially evokes FSH but not LH secretion. Numerous studies in a variety of species support the hypothesis that secretion of the gonadotropins FSH and LH can be differentially regulated by alterations in the pattern of GnRH neurosecretion [361. Reducing the frequency of exogenously administered GnRH pulses caused LH levels to decline and FSH levels to increase in arcuate nucleuslesioned rhesus monkeys receiving infusions of pulsatile GnRH stimulation [15]. Similar results have been obtained in pituitary-disconnected, ovariectomized ewes [17]. In female rats both the secretion [20] and synthesis [181 of FSH and LH are differentially regulated via GnRH frequency modulation. Rat FSH-P subunit transcription is maximally stimulated at lower pulse frequencies, whereas LH-3 and -a subunit transcription are maximally stimulated at more rapid pulse frequencies [181. In a recent report [16] we demonstrated in Djungarian hamsters that FSH and LH secretion can be differentially stimulated by various GnRH frequencies; GnRH pulses delivered at a moderate frequency favor FSH secretion, while increases in the frequency preferentially stimulate LH secretion. Taken together with these previous findings, the present data indicate that GnRH pulse frequency modulation may mediate the singular release of FSH in photostimulated hamsters. Nevertheless, it remains to be shown that the frequency of endogenous GnRH pulses is indeed modulated by photoperiod in a manner predicted by the divergent FSH and LH release patterns and by the differential sensitivities of the gonadotropins to various GnRH pulse frequencies. It also remains to be determined how changes in GnRH pulse frequency may be registered and transduced by pituitary gonadotrophs to yield divergent secretion of the two hormones. In the adult male Djungarian hamster, photostimulation produces not only a rapid increase in FSH secretion but also a more slowly developing rise in LH levels that peaks at about 30 days after transfer to long days [2]. The occurrence of temporally dissociated FSH and LH secretion in response to photic stimulation either may be an epiphenomenal consequence of GnRH pulse generator activation or may bear physiological significance. Clearly, the neuroendocrine responses to photostimulation appear to be organized such that initial hormone secretions (FSH) convey specific signals for growth and development of the testes, while later signals (LH) lead to secretion of T and stimulation of sexual activity.

730 WOLFE ET AL. It is possible that the sequential secretions of FSH and LH in the adult of this species are stimulated independently of each other via GnRH pulse frequency modulation and that these temporally appropriate secretions are part of an integrated neuroendocrine response to photostimulation. No such increase in LH levels was observed even after 40 days of exposure to a long-day photoperiod in the juvenile hamsters studied in the present experiment. This represents the second report that in the photostimulated juvenile hamster, LH and T levels do not increase after photostimulation [161 as they do in the photostimulated adult [2]. The reason for this difference between the juvenile and adult responses to photic stimulation are not immediately apparent but are likely attributable to some immaturity in the neuroendocrine response system of the juvenile hamsters. It is possible, for example, that photostimulation of juvenile Djungarian hamsters provokes only a relatively moderate acceleration of GnRH pulse generator activity that is in turn sufficient to stimulate FSH secretion. Further study of GnRH release in juveniles and adult hamsters following photostimulation will be required to assess this possibility. Milette et al. [1] have previously provided evidence that the photic-induced growth of the testes in Djungarian hamsters is dependent upon the effects of FSH. The present study provides further evidence for this hypothesis, as GnRHa treatment was found to attenuate the increase of FSH levels seen in response to photostimulation and to prevent testicular development normally seen in photostimulated hamsters. The 30-day GnRHa-treated, photostimulated hamsters, for example, had an attenuated FSH response to long-day photostimulation and showed minimal growth of the testes, limited tubular development, and little evidence of spermiogenesis. Thus, the rapid growth of the testes in response to photoperiodic stimulation is dependent upon photic-induced FSH secretion and would seem to be, in turn, dependent upon GnRH neurosecretion. It is interesting to note, however, that robust testicular growth and development lag significantly after the peak of FSH secretion. It is possible that FSH actions become grossly evident only after sustained action of the hormone. It may also be the case that other photically stimulated factors, such as prolactin [37, 38], must be secreted in sufficient amounts to support or permit FSH actions. Although photic-induced FSH secretion was found to be predominantly dependent upon GnRH stimulation, a significant component of basal FSH secretion was resistant to the inhibitory effects of GnRHa. A significant component of FSH secretion is similarly resistant to GnRH receptor blockade in the rat and may therefore be GnRH-independent [21, 22]. It has been suggested that this putative GnRH-independent fraction of FSH secretion either is a constitutive process [36] or is dependent upon stimulation by a separate FSH-RF [7]. It is also possible, however, that some fraction of this GnRHa-resistant FSH secretion occurs as a result of incomplete blockade of GnRH receptors. For example, the low-level "breakthrough" stimulation by GnRH in the presence of GnRHa may more effectively support FSH vs. LH synthesis and/or secretion. Indeed, partial escape from the effects of a GnRHa are suggested in the present study, since some photic-induced FSH secretion, albeit delayed and attenuated, did occur in the photostimulated, GnRHa-treated groups. Thus, while we can conclude that much of the photic-induced, singular secretion of FSH depends upon GnRH stimulation, it cannot be determined from these experiments whether the GnRHa-insensitive component of FSH secretion is truly driven by GnRH-independent processes or whether it represents residual secretion resulting from our inability to achieve total and continuous blockade of GnRH receptors in these animals. In summary, we have demonstrated in Djungarian hamsters that the photic-induced increment in FSH secretion, as well as much of the testicular growth and development driven by this FSH release, is dependent upon endogenous GnRH neurosecretion. Although the mechanism by which GnRH may stimulate singular FSH secretion in response to photostimulation is unknown, it appears that some characteristic of the pulsatile GnRH release pattern is altered by photostimulation, which in turn provides a specific stimulus for the preferential release of FSH from the anterior pituitary gland. On the basis of work in hamsters and other species, it is hypothesized that GnRH pulse frequency may be one such critical feature of GnRH release that is modulated by photoperiod and that in turn directs these independent changes in gonadotropin secretion. ACKNOWLEDGMENTS We thank Dr. Virginia Carr for her invaluable technical advice, Dr. Erwin Goldberg for assistance with histological analysis, and the NIDDK for all LH and FSH RIA reagents. The technical assistance of Jeffrey Norgle, Yi-Rong Ge, and Brigitte Mann is also gratefully acknowledged. REFERENCES 1. Milette JJ, Schwartz NB, Turek FW. The importance of follicle-stimulating hormone in the initiation of testicular growth in photostimulated Djungarian hamsters. Endocrinology 1988; 122:1060-1066. 2. Simpson SM, Follett BK, Ellis DH. Modulation by photoperiod of gonadotrophin secretion in intact and castrated Djungarian hamsters. J Reprod Fertil 1982; 66:243-250. 3. Bartness TJ, Powers JB, Hastings MH, Bittman EL, Goldman BD. The timed infusion paradigm for melatonin delivery: what has it taught us about the melatonin signal, its reception, and the photoperiodic control of seasonal responses? J Pineal Res 1993; 15:161-190. 4. Culler MD, Negro-Vilar A. Evidence that pulsatile follicle-stimulating hormone secretion is independent of endogenous luteinizing hormone-releasing hormone. Endocrinology 1986; 118:609-612. 5. Culler MD, Negro-Vilar A. Pulsatile follicle-stimulatinghormone secretion is independent of luteinizing hormone-releasing hormone (LHRH): pulsatile replacement of LHRH bioactivity in LHRH-immunoneutralized rats. Endocrinology 1987; 120:2011-2021. 6. Kovacs M, Koppan M, Mez6 I, Teplan, Flerk6 B. Antiovulatory doses of antagonists of LH-RH inhibit LH and progesterone but not FSH and estradiol release. J Neuroendocrinol 1993; 5:603-608.

FSH RELEASE IS DEPENDENT ON GnRH 731 7. Lumpkin MD, MoltzJH, Yu WH, Samson WK, McCann SM. Purification of FSH-releasing factor: its dissimilarity from LHRH of mammalian, avian, and piscian origin. Brain Res Bull 1987; 18:175-178. 8. Marder ML, Channing CP, Schwartz NB. Suppression of serum follicle stimulating hormone in intact and acutely ovariectomized rats by porcine follicular fluid. Endocrinology 1977; 101:1939-1942. 9. Schwartz NB, Rivier C, Rivier J, Vale WW. Effect of gonadotropin-releasing hormone antagonists on serum follicle-stimulating hormone and luteinizing hormone under conditions of singular follicle-stimulatinghormone secretion. Biol Reprod 1985; 32:391-398. 10. Schwartz NB, Channing CP. Evidence for ovarian "inhibin": suppression of the secondary rise in serum follicle stimulating hormone levels in proestrous rats by injection of porcine follicular fluid. Proc Natl Acad Sci USA 1977; 74:5721-5724. 11. BourguignonJ-P, Gerard A, Debougnoux G, Rose J, Franchimont P. Pulsatile release of gonadotropin-releasing hormone (GnRH) from the rat hypothalamus in vitro calcium and glucose dependency and inhibition by superactive GnRH analogs. Endocrinology 1987; 121:993-999. 12. Levine JE, Ramirez VD. In vivo release of luteinizing hormone-releasing hormone estimated with push-pull cannulae from the mediobasal hypothalami of ovariectomized, steroid-primed rats. Endocrinology 1980; 107:1782-1790. 13. Carmel PW, Araki S, Ferin M. Pituitary stalk portal blood collection in rhesus monkeys: evidence for pulsatile release of gonadotropin-releasing hormone (GnRH). Endocrinology 1976; 99:243-248. 14. Dierschke DJ, Bhattacharya AN, Atkinson LE, Knobil E. Circhoral oscillations of plasma LH levels in the ovariectomized rhesus monkey. Endocrinology 1970; 87:850-853. 15. Wildt L, Hausler A, Marshall G, HutchisonJS, Plant TM, Belchetz PE, Knobil E. Frequency and amplitude of gonadotropin-releasing hormone stimulation and gonadotropin secretion in the rhesus monkey. Endocrinology 1981; 109:376-385. 16. Meredith JM, Turek FW, Levine JE. Differential regulation of LH and FSH by GnRH pulse frequency in the photoinhibited Djungarian hamster. Soc Neurosc 1992; 18:192 (abstract #90.5). 17. Clarke IJ, Cummins JT, FindlayJK, Burman KJ, Doughton BW. Effects on plasma luteinizing hormone and follicle-stimulating hormone of varying the frequency and amplitude of gonadotropin-releasing hormone pulses in ovariectomized ewes with hypothalamopituitary disconnection. Neuroendocrinology 1984; 39:214-221. 18. Haisenleder DJ, Dalkin AC, Ortolano GA, MarshallJC, Shupnik MA. A pulsatile gonadotropin-releasing hormone stimulus is required to increase transcription of the gonadotropin subunit genes: evidence for differential regulation of transcription by pulse frequency in vivo. Endocrinology 1991; 128:509-517. 19. Darkin AC, Haisenleder DJ. The frequency of gonadotropin-releasing-hormone stimulation differentially regulates gonadotropin subunit messenger ribonucleic acid expression. Endocrinology 1989; 125:917-924. 20. Kellom TA, O'ConnerJL. The induction of divergent gonadotropin secretion in vitro by variations in luteinizing hormone releasing hormone pulse regimen. Steroids 1991; 56:284-290. 21. Grady RR, Shin L, Charlesworth MC, Cohen-Becker IR, Smith M, Rivier C, RivierJ, Vale W, Schwartz NB. Differential suppression of follicle-stimulating hormone and luteinizing hormone secretion in vivo by a gonadotropin-releasing hormone antagonist. Neuroendocrinology 1985; 40:246-252. 22. Kartun K, Schwartz NB. Effects of a potent antagonist to gonadotropin-releasing hormone on male rats: luteinizing hormone is suppressed more than follicle-stimulating hormone. Biol Reprod 1987; 36:103-108. 23. Hoffmann K, Nieschlag E. Circadian rhythm of plasma testosterone in the male Djungarian hamster (Phodopus sungorus). Acta Endocrinol 1977; 86:193-199. 24. Milette JJ, Turek FW. Circadian and photoperiodic effects of brief light pulses in male Djungarian hamsters. Biol Reprod 1986; 35:327-335. 25. Condon TP, Sawyer CH, Whitmoyer DI. Episodic patterns of luteinizing hormone and follicle-stimulating hormone release: differential secretory dynamics and adrenergic control in ovariectomized rats. Endocrinology 1986; 118:2525-2533. 26. Batta SK, Zanisi M, Martini L. Role of prostaglandins in the control of gonadotropin secretion. Psychoneuroendocrinology 1975; 1:115-121. 27. Grady RR, Schwartz NB. Role of gonadal feedback in the differential regulation of LH and FSH in the rat. In: Franchiment P, Channing CP (eds.), Intragonadal Regulation of Reproduction. London: Academic Press; 1981: 377-392. 28. Schwartz NB, Talley WL. Effects of exogenous LH or FSH on endogenous FSH, progesterone and estradiol secretion. Biol Reprod 1978; 18:820-828. 29. Brown-Grant K, Greig F. A comparison of changes in the peripheral plasma concentrations of luteinizing hormone and follicle-stimulating hormone in the rat. J Endocrinol 1975; 65:389-397. 30. Chiappa SA, Fink G. Releasing factor and hormonal changes in the hypothalamic-pituitary-gonadotrophin and -adrenocorticotrophin systems before and after birth and puberty in male, female and androgenized rats. J Endocrinol 1977; 72:211-224. 31. Chandrashekar V, Bartke A, Sellers K. Prolactin modulates the gonadotropin response to the negative feedback effect of testosterone in immature male rats. Endocrinology 1987; 120:758-763. 32. Lambalk CB, van Rees GP, SchoemakerJ, De Koning J, van DietenJA. Acute desensitization of pituitary FSH response to LHRH in ovariectomized rats: further evidence that in the presence of ovarian proteins the LHRH-dependent, LH-like component of FSH release becomes apparent. J Endocrinol 1989; 123:221-226. 33. Brooks J, Struthers WJ, McNeilly AS. GnRH-dependent and -independent components of FSH secretion after acute treatment of anoestrous ewes with ovine follicular fluid and a GnRH antagonist. J Reprod Fertil 1993; 98:591-595. 34. Kirby JD, Jetton AE, Ackland JF, Turek FW, Schwartz NB. Changes in serum immunoreactive inhibin-a during photoperiod-induced testicular regression and recrudescence in the golden hamster. Biol Reprod 1993; 49:483-488. 35. Turek FW, Alvis JD, Menaker M. Pituitary responsiveness to LRF in castrated male hamsters exposed to different photoperiodic conditions. Neuroendocrinology 1977; 24:140-146. 36. Levine JE, Conaghan LA, Luderer U, Strobl FJ. Hypothalamic regulation of FSH secretion. In: Hunzicker-Dunn M, Schwartz NB (eds.), Follicle Stimulating Hormone: Regulation of Secretion and Molecular Mechanisms of Action. New York: Springer-Verlag; 1992: 66-82. 37. Yellon SM, Goldman BD. Effects of melatonin on long-day responses in short-day housed adult siberian hamsters. Am J Physiol 1988; 255(5 pt 2):R823-R830. 38. Yellon SM, Goldman BD. Photoperiod control of reproductive development in the male Djungarian hamster (Phodopus sungorus). Endocrinology 1984; 114:664-670.