|
|
||||||||
Articles |
a Departments of Obstetrics & Gynecology and Biology,
b Yale University School of Medicine, New Haven, Connecticut 06520-8063
c Reproductive Sciences Program and Departments of Obstetrics & Gynecology and Biology, University of Michigan, Ann Arbor, Michigan 48109-0404
| ABSTRACT |
|---|
|
|
|---|
| INTRODUCTION |
|---|
|
|
|---|
To understand the control of GnRH secretion in males and females, it is important to identify sex differences in the neural elements that determine GnRH release. The basic anatomy of the GnRH neurosecretory system is similar in male and female lambs. By mid-gestation, the number of GnRH neurons in fetal lambs of both sexes is comparable to that in adults [6]. Moreover, the distribution and gross morphology of GnRH neurons are not sexually dimorphic. Although sex differences in the anatomy of the GnRH neurons have not been observed at the light microscopic (LM) level, synaptic input to GnRH neurons may be sexually dimorphic. In this regard, GnRH neurons of female rats have been found to receive twice as many afferent synapses as those of males [7]. Sex differences in synapses on GnRH neurons may underlie sex differences in the control of GnRH secretion in males and females. The present study compared the number of synapses on GnRH neurons from male and female yearling lambs. To determine whether such sex differences are under the control of prenatal steroids, we also examined afferents to GnRH neurons in prenatally androgenized females.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Male, female, and androgenized female lambs (n = 5 each) were born in April at the Reproductive Sciences Program Sheep Research Facility in Ann Arbor, Michigan. Androgenized females were masculinized as previously described [8]. Briefly, female lambs were exposed to testosterone cypionate via weekly maternal injections (200 mg in 1.0 ml cottonseed oil; Sigma Chemical Co., St. Louis, MO) from 30 to 86 days of gestation (145 days is term). Treatment encompassed the entire critical period for sexual differentiation [8, 9], and the external genitalia of the androgenized females resembled those of normal male lambs. In addition, measurements of LH secretion showed that prenatal testosterone masculinized reproductive neuroendocrine function in the androgenized female lambs. Males and females that received no prenatal steroids served as controls. Lambs were group-housed outdoors in fence-line contact with adult rams and ewes. After weaning, all lambs were fed a commercial pelleted diet with vitamin supplements and alfalfa hay to ensure proper growth, and water was available at all times. All animals were cared for in accordance with the NIH Guide for the Care and Use of Laboratory Animals, and experimental protocols were approved by the University Committee on the Use and Care of Animals.
Gonadectomy and Steroid Replacement
To standardize the hormonal milieu and to provide a constant physiologic steroid feedback signal, the gonads were removed at 2 wk of age, and steroids were replaced with an estradiol implant as previously described [8]. The implants remained in place until perfusion. Estradiol was used in both male and female lambs because males respond similarly to estradiol and testosterone with regard to the regulation of LH pulse frequency [10] and because estradiol is considered to play an important role in inhibitory steroid feedback in the ram [11]. As determined from LH secretion in twice-weekly blood samples beginning at two weeks of age, all lambs reached sexual maturity by the autumn of their first year, some 6 mo before perfusion. In the estradiol-treated gonadectomized lamb, a sustained rise in circulating LH concentrations from infrequent blood samples reflects an increase in LH pulse frequency resulting from the pubertal reduction in sensitivity to inhibitory steroid feedback [5].
Perfusion and Sectioning
Lambs were killed in the spring at 1 yr of age (April 24), during the nonbreeding season. Perfusions followed the methods of Moenter et al. [12]. Briefly, lambs were decapitated under deep pentobarbital anesthesia, and the head was perfused via the carotid arteries with 3.5 L of 0.1 M sodium phosphate buffer (PB) containing 2% paraformaldehyde, 0.1% sodium nitrite for vasodilation, 1 U/ml heparin to minimize clotting, and 0.2% glutaraldehyde. Brains were removed and postfixed overnight in the perfusion fixative. A block of tissue containing the preoptic area and hypothalamus was dissected out, and 60-µm coronal sections were cut on a Vibratome (Technical Products International Inc., St. Louis, MO). Sections were stored in 6 serial vials containing PB with 0.01% sodium azide.
Immunocytochemistry
Free-floating sections were stained for GnRH using the LR-1 antiserum (1:40 000, gift of Dr. Robert Benoit, Montreal General Hospital, Montreal, PQ, Canada) as described previously [13]. This antiserum has been previously characterized for use in sheep [14]. Sections were incubated overnight at room temperature in primary antibody with 0.02% saponin and 4% normal donkey serum in PB (Jackson Immunoresearch Labs, West Chester, PA). Next, sections were exposed to biotinylated donkey anti-rabbit IgG (1:200, Jackson) and avidin-biotin-horseradish peroxidase (HRP, 1:100, Vectastain Elite Kit; Vector Labs, Burlingame, CA), each for 1 h at room temperature, with extensive rinsing between incubations. HRP was visualized with NiCl-enhanced 3,3' diaminobenzidine (DAB) to produce a dark blue-black reaction product over GnRH-immunoreactive neurons.
Embedding and Thin Sectioning
Individual GnRH neurons from the rostral portion of the medial preoptic area (MPOA) near the organum vasculosum of the lamina terminalis (OVLT, see Fig. 1A) were visualized under a dissecting microscope and removed from the section with a tissue punch (0.75-mm diameter). These punches were osmicated in 1% osmium tetroxide for 15 min, rinsed thoroughly in buffer, and dehydrated through graded alcohols. Punches were exposed to 1% uranyl acetate in 70% ethanol for 30 min, followed by 100% propylene oxide (2 times for 10 min each), and infiltrated with 50% Durcopan resin (Electron Microscopy Sciences [EMS], Fort Washington, PA) in propylene oxide overnight at 4°C. Finally, the punches were individually flat-embedded in Durcopan between liquid release-coated slides and Thermanox coverslips (EMS), and mounted onto Durcopan-filled gelatin capsules for thin sectioning. Thin sections (6070 nm) were cut on an LKB (Rockville, MD) ultramicrotome and collected on slotted Formvar (EMS)-coated copper grids. Thin sections were counterstained with lead citrate (0.25% for 3 min) before viewing on the electron microscope.
|
Electron Microscopy (EM)
Sections were examined with a Phillips CM10 transmission electron microscope (Eindhoven, The Netherlands). GnRH neurons were identified by dark, flocculent DAB material in the cell cytoplasm (Figs. 2 and 3). For each animal, at least 5 GnRH neurons (range 57) cut through the plane of the neuronal cell nucleus were photographed and examined at a final magnification of x33 000. Synapses were identified by the presence of a distinct synaptic bar and vesicles in the presynaptic terminal (see Fig. 4 for example of synapses on GnRH neurons). Because of dense staining in the GnRH neuron, bars in the postsynaptic terminal could not be clearly distinguished for all neurons. From each neuron, afferent synapses were counted by two observers blind to the treatment groups. In addition, the cell perimeter and the extent of glial ensheathment were measured using a cartographer's wheel. The number of synapses, expressed per 100 µm of plasma membrane, and the percentage of glial ensheathment on each neuron profile (mean perimeter 72.1 ± 3.5 µm; range 31177 µm) were determined. Values from the 5 neurons in each animal were averaged. Group means were compared by ANOVA with post-hoc comparisons by Scheffé's F-test. For all analyses, p < 0.05 was considered significant.
|
|
| RESULTS |
|---|
|
|
|---|
GnRH-immunoreactive neurons from male, female, and androgenized female lambs were present from the diagonal band of Broca through the mediobasal hypothalamus, as reported previously [6, 14, 15]. However, the majority of immunoreactive neurons were clustered in rostral MPOA at the level of the OVLT (Fig. 1A). Immunoreactive cells were predominantly multipolar, with branching dendrites (Fig. 1B). At the EM level, GnRH immunoreactive neurons were identified by the dark, flocculent DAB reaction. Although labeling was restricted to the cell cytoplasm in most GnRH neurons, there were several neurons with DAB staining in the cell nucleus as well (Figs. 2 and 3). As shown in Figure 2, GnRH neurons were often found in pairs or small groups, although contacts between GnRH neurons were difficult to distinguish. Occasionally, possible GnRH-GnRH contacts were observed (Fig. 4D), although the number of such contacts was insufficient to permit statistical comparison.
Synaptic input to GnRH neurons in yearling lambs was sparse, as described previously for the adult ewe [16, 17]. Figure 3 illustrates afferent synapses and glial ensheathment on a GnRH neuron from control male and female lambs. Synapses were found both on the cell soma and on proximal dendrites. Synapses on distal dendrites were not measured, since we were unable to observe uninterrupted dendrites more than 65 µm from the cell soma. Many synapses were found on small "spine-like" protuberances of the cellular membrane (Fig. 4B), as described previously [16]. In addition to direct synaptic input, there were many axon terminals in close apposition to the plasma membrane. However, much of the GnRH plasma membrane was covered by thin sheets of glia that were interposed between GnRH neurons and axon terminals (Fig. 4C).
|
Sex Differences in Synaptic Contacts on GnRH Neurons
Figure 5 presents a schematic outline of a representative GnRH neuron from each group. At the EM level, as at the LM level, there were no sex differences in the gross morphology of GnRH neurons. The average perimeter per neuron profile was not significantly different between males (80.6 ± 8.8 µm) and females (79.2 ± 8.9 µm). Although neurons from androgenized females tended to be smaller (53.4 ± 3.4 µm), the average perimeter was not significantly different in neurons from either males or control females (Table 1; p > 0.05). For all lambs, 76.5% of synapses were found on the cell soma (defined as the area within 10 µm of the nucleus, according to [18]) and 23.5% on proximal dendrites; 34.7% of all synapses were present on spines projecting from the cell soma or dendrites. However, there was a significant effect of sex on the number of synaptic contacts on GnRH neurons (Table 1). GnRH neurons from females received twice as many synapses (3.6 ± 0.7 per 100 µm plasma membrane) as those from males (1.6 ± 0.3, p < 0.05). Moreover, synaptic input to GnRH neurons in androgenized females (1.5 ± 0.5) was similar to that of males (p > 0.05). The increase in afferent input in female lambs was accompanied by reduced glial ensheathment of GnRH neurons in males. Whereas 69.3% of the neuron perimeter was covered by glia in male lambs, in females only 57.8% of the plasma membrane was enclosed by glia (p < 0.05). Glial ensheathment in androgenized females (62.4%) was intermediate between males and females.
|
|
| DISCUSSION |
|---|
|
|
|---|
GnRH secretion in the female lamb is more responsive to modulation by some sensory and endocrine cues. With regard to tonic secretion, the pubertal increase in pulsatile LH secretion in females (reflecting a decrease in responsiveness to inhibitory steroid feedback) is delayed until day lengths decrease during the autumn (see [5]). By contrast, males and androgenized females are relatively insensitive to photoperiod changes for timing of puberty [21]. For the preovulatory surge, estradiol induces a massive release of GnRH in female sheep [22], accompanied by Fos expression in GnRH neurons [12]. In males, the GnRH surge and associated Fos induction is neither present nor inducible [3, 13]. This sex difference is also established by gonadal steroids prenatally.
It is reasonable to expect that the neural input to GnRH neurons in females might reflect the additional physiologic regulation of GnRH release. Although GnRH neurons lack receptors for gonadal steroids [23, 24], they are a logical, if indirect, target for gonadal steroids to effect sexual differentiation of GnRH release. Nonetheless, the GnRH neurosecretory system of the mid-gestation ovine fetus is not sexually dimorphic at the LM level [6]. Although one study in adult Shiba goats reported that males have nearly twice as many GnRH neurons as females [25], studies in other species have found no evidence for sex differences in GnRH neurons at the LM level (rat [26, 27]; ferret [28]; monkey [29]).
It is more likely to find subtle remodeling of the GnRH neurosecretory system at the ultrastructural level. Synapses can be rapidly broken or restored in response to gonadal steroids. In this regard, estrogen suppresses synapses in the arcuate nucleus of the female rat within 24 h [30]. Sex differences in synapses have been described in several limbic structures [31], beginning with the pioneering work of Raisman and Field in MPOA [32] and Matsumoto and Arai in hypothalamus [33]. In particular, a study in rats reported that GnRH neurons of females have nearly twice as many synapses as those of males [7]. That study involved a detailed comparison of GnRH afferents from two male and two female rats. In the present study, GnRH afferents were sampled less intensively from a larger number of animals, with similar findings. It is important to note here that our findings apply only to synapses on the cell soma and proximal dendrites. We cannot make inferences about synaptic input to distal dendrites. Nonetheless, the results extend our understanding of the development of GnRH input by demonstrating that the prenatal steroid environment in sheep is responsible for the postnatal sex difference in the synaptology of GnRH neurons. Not surprisingly, glial ensheathment also decreases as synaptic input increases. Studies in monkeys [18, 34] have reported similar changes in glial apposition.
Other conditions that modify the synaptology of GnRH neurons include gonadectomy, aging, and photoperiodic transitions. However, it is difficult to generalize about the relationship between afferent input to GnRH neurons and GnRH release. In some instances, synapses increase along with GnRH release, as for long-term ovariectomy in rats [35], puberty in male rats [36], or seasonality in female sheep [17]. In other cases, synapses vary inversely with GnRH release, as with ovariectomy in monkeys [34], aging in male rats [37], puberty in male monkeys [38], or seasonality in male starlings [39]. Several studies found that no changes in synaptic input occur with aging in female rats [35], short-term (1-mo) gonadectomy in male or female rats [35, 40], or puberty in female monkeys [18]. Factors that may contribute to the diversity of synaptic changes include species differences in GnRH neurons and differences in local populations of GnRH neurons, as well as differences in sampling techniques. In this regard, GnRH neurons of rodents appear to have substantially fewer afferent synapses compared to those of monkeys and sheep, in addition to less complex morphology. The number of synapses per GnRH profile in the present study is similar to that described for the adult ewe [17].
Although our results suggest that the prenatal steroid environment is responsible for postnatal synapses on GnRH neurons, several questions remain. It is not known which steroids are responsible for sexual differentiation of GnRH synaptology and which neurotransmitters are involved. Presumably, exposure of the male brain to gonadal steroids in utero inhibits the development of neural elements responsible for complex control of GnRH release in females, a process termed defeminization [41]. Whether the additional synapses in females reflect increased inhibitory tone mediating photoperiodic inhibition of the GnRH system or whether the extra GnRH afferents in females stimulate the preovulatory GnRH surge is not known. Establishing the neurochemical identity of afferent synapses would help to resolve this question. In this regard, GnRH neurons in sheep appear to receive input from catecholamine neurons [42]. Studies in rats also demonstrate synaptic input to GnRH neurons from beta-endorphin, gamma aminobutyric acid, substance P, serotonin, somatostatin, and glutamatergic neurons (reviewed in [43]). Many of these neurotransmitter neurons have steroid receptors in both sheep [24, 44] and rats (see [43]). Additionally, we do not know the steroid signal responsible for defeminizing synaptic input to GnRH neurons. Whereas the expression of the LH surge in rats is determined by exposure to estradiol during a critical period for sexual differentiation (see [45]), recent work from our laboratory indicates that prenatal androgens may determine the postnatal control of tonic LH secretion [46]. The differential effect of androgens and estrogens on synaptic input to GnRH neurons remains to be resolved.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
2 Correspondence: Ruth I. Wood, Department of Cell&Neurobiology, USC School of Medicine, 1333 San Pablo St. BMT 401, Los Angeles, CA 90033. FAX: 323 442 3158. ![]()
Accepted: April 13, 1999.
Received: March 8, 1999.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
L. M. Jackson, K. M. Timmer, and D. L. Foster Organizational Actions of Postnatal Estradiol in Female Sheep Treated Prenatally with Testosterone: Programming of Prepubertal Neuroendocrine Function and the Onset of Puberty Endocrinology, May 1, 2009; 150(5): 2317 - 2324. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. M. Garcia-Segura, B. Lorenz, and L. L DonCarlos The role of glia in the hypothalamus: implications for gonadal steroid feedback and reproductive neuroendocrine output Reproduction, April 1, 2008; 135(4): 419 - 429. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Robinson Prenatal programming of the female reproductive neuroendocrine system by androgens. Reproduction, October 1, 2006; 132(4): 539 - 547. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. M. Foecking, M. Szabo, N. B. Schwartz, and J. E. Levine Neuroendocrine Consequences of Prenatal Androgen Exposure in the Female Rat: Absence of Luteinizing Hormone Surges, Suppression of Progesterone Receptor Gene Expression, and Acceleration of the Gonadotropin-Releasing Hormone Pulse Generator Biol Reprod, June 1, 2005; 72(6): 1475 - 1483. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. D. Sullivan and S. M. Moenter Prenatal androgens alter GABAergic drive to gonadotropin-releasing hormone neurons: Implications for a common fertility disorder PNAS, May 4, 2004; 101(18): 7129 - 7134. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. R. Thanky, R. Slater, and A. E. Herbison Sex Differences in Estrogen-Dependent Transcription of Gonadotropin-Releasing Hormone (GnRH) Gene Revealed in GnRH Transgenic Mice Endocrinology, August 1, 2003; 144(8): 3351 - 3358. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Gahr Male Japanese quails with female brains do not show male sexual behaviors PNAS, June 24, 2003; 100(13): 7959 - 7964. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. J. Scott, I. J. Clarke, and A. J. Tilbrook Neuronal Inputs from the Hypothalamus and Brain Stem to the Medial Preoptic Area of the Ram: Neurochemical Correlates and Comparison to the Ewe Biol Reprod, April 1, 2003; 68(4): 1119 - 1133. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |