Skip to main content
NCBI home page
Endocrinology logoLink to Endocrinology
. 2009 Jul 9;150(10):4663–4671. doi: 10.1210/en.2009-0432

Estradiol Negative Feedback Regulation by Glutamatergic Afferents to A15 Dopaminergic Neurons: Variation with Season

Sushma R Singh 1, Stanley M Hileman 1, John M Connors 1, Christina J McManus 1, Lique M Coolen 1, Michael N Lehman 1, Robert L Goodman 1
PMCID: PMC2754677  PMID: 19589862

Abstract

It is now clear that seasonal breeding in ewes is due to an increase in response to estradiol (E2) negative feedback in the nonbreeding season (anestrus) that is mediated by the A15 group of dopaminergic (DA) neurons. Because A15 cells do not contain estrogen receptors, we have postulated the presence of estrogen-responsive afferents and recently reported evidence that input from neurons containing γ-aminobutyric acid (GABA) contribute to the control of A15 activity by E2. However, GABAergic afferents account for only a fraction of A15 synaptic input and do not appear to vary with season. We therefore investigated the possible role of stimulatory glutamatergic input to A15 neurons. In experiments 1 and 2, local administration into the A15 of either a N-methyl-d-aspartate (NMDA) receptor or a kainate/α-amino-3-hydroxyl-5-methyl-4-isoxazole-propionate (AMPA) receptor antagonist stimulated episodic LH secretion in a dose-dependent manner in ovary-intact anestrous ewes. In experiment 3, we examined the number of glutamatergic close contacts onto A15 neurons using dual immunocytochemistry in tissue from E2-treated ovariectomized anestrous and breeding season ewes. All A15 DA neurons were contacted by glutamatergic vesicles, and the number of close contacts was significantly higher in anestrus than the breeding season. Finally, using a triple-label immunocytochemistry procedure, we did not observe any colocalization of markers for GABA and glutamate in vesicles contacting A15 neurons. These results support the hypothesis that glutamatergic afferents actively stimulate A15 DA neurons in ovary-intact anestrous ewes and raise the possibility that alterations in this input may contribute to increased A15 neural activity during anestrus.

Glutamatergic afferents to A15 dopaminergic neurons mediate estradiol negative feedback in anestrus and decrease in breeding season ewes in parallel with inhibitory actions of estradiol.

Temperate-region sheep are seasonal breeders with the breeding season beginning in the early fall and ending late winter in response to changing photoperiod (1). Anestrus, the period of reproductive acyclicity, in ewes is characterized by diminished pulsatile GnRH and LH secretion (1,2,3) induced by both steroid-dependent and steroid-independent actions of inhibitory photoperiod (4). The former, which is evident as a marked increase in the responsiveness of the GnRH secretory system to estrogen negative feedback at this time of year (1,2,3,4,5), is primarily responsible for seasonal breeding in ewes (1,2). Studies using sheep and horses (6,7,8) have shown that inhibitory dopaminergic (DA) neurons are an integral part of the neural circuitry that mediates this seasonally enhanced sensitivity to estradiol (E2). In ewes, A15 DA neurons residing in the retrochiasmatic area (RCh) play an important role in this increased sensitivity to estrogen and are largely responsible for the slow LH pulse frequency in ovary-intact anestrous ewes (7,8,9). Estrogen receptors (ER), however, are not expressed in this neuronal population (10,11), indicating that steroid negative feedback is conveyed to these DA neurons via other E2-sensitive afferents during anestrus. We have recently reported evidence that neurons containing γ-aminobutyric acid (GABA) contribute to this afferent input (12). However, GABAergic boutons account for only approximately 25–30% of the synaptic close contacts onto A15 neurons (12,13), so other types of neural input may also mediate the stimulatory actions of estrogen on A15 neurons. The excitatory amino acid neurotransmitter glutamate is another likely candidate because many glutamate neurons contain ERα in the ewe (14).

Glutamate has been implicated as one of the primary regulators of GnRH secretion in several species (for review see Ref. 15). The reproductive neuroendocrine effects of glutamate are primarily exerted via ionotropic receptor subtypes: N-methyl-d-aspartate (NMDA), kainate, and α-amino-3-hydroxyl-5-methyl-4-isoxazole-propionate (AMPA) (15). NMDA, AMPA and kainate elicit GnRH release (15,16,17), and subsequently LH release (15,17,18,19,20,21,22,23) in a variety of species, including rats, sheep, and primates, whereas administration of NMDA and AMPA/kainate (non-NMDA) receptor antagonists inhibit gonadotropin release (15,17,24). The NMDA receptor subunit NR1 (25), AMPA receptor subunits GluR1-4 (26), and kainate receptor subunits KA2 and GluR5-GluR7 (17) are coexpressed in GnRH neurons. In rats, glutamate release within the preoptic area is elevated during the afternoon of the steroid-induced LH surge (15,27). In addition, changes in glutamatergic tone have been implicated in age-related changes in reproductive function that occur at puberty (15,23,28,29,30) and during reproductive senescence (16,25,31).

In the present study, we first investigated the possibility that stimulatory glutamatergic tone onto A15 DA neurons is necessary for the slow LH pulse frequency in ovary-intact anestrous ewes using local administration of an NMDA receptor antagonist and a non-NMDA ionotropic glutamate receptor antagonist into the RCh. Because these experiments indicated a role for gluamatergic afferents to A15 neurons in anestrus, we next determined whether this input varied seasonally by comparing the number of close contacts containing the vesicular glutamate transporter-2 (vGlut2), a marker for glutamatergic synapses (32), on A15 neurons in tissue collected from anestrous and breeding season ewes. Finally, because of evidence that glutamate and GABA are found in the same boutons innervating GnRH neurons in rats (33), we determined whether this also occurred in A15 afferents using triple-label immunofluorescence for vGlut2, vesicular GABA transporter (vGAT), and tyrosine hydroxylase (TH).

Materials and Methods

Animals

Adult mixed-breed blackface ewes were maintained in an open barn and moved indoors 3–7 d before surgeries. Ewes were fed a daily quota of silage and had free access to water. Lights were adjusted every 2 wk to mimic the duration of natural lighting. Anestrous experiments were carried out from May through July, and tissue from breeding-season ewes was collected between October and the end of January; the seasonal status of each ewe was determined based on estrous behavior and/or status of the ovaries at the time of ovariectomy. All procedures were approved by the West Virginia University Animal Care and Use Committee and followed National Institutes of Health guidelines for use of animals in research.

Surgeries

All surgical procedures were carried out as previously described (34,35) under sterile conditions using 3% halothane supplemented with nitrous oxide as anesthesia. Ovariectomies were performed via midventral laporatomy. For implanting chronic guide cannulae, the head of the ewe was positioned in a stereotaxic apparatus, the surface of the skull was exposed, and a small portion of bone was removed just rostral to bregma to expose the surface of the brain. Radioopaque dye, Omnipaque 350 (Iohexol, Winthrop, NY) was injected into the right lateral ventricle, and x-ray radiography used to visualize placement of the guide cannulae. Bilateral cannulae were placed 3.0 mm from midline at the posterior border of the optic chiasm and lowered to a position 2.0 mm dorsal to the floor of optic recess. Guide cannulae were cemented into place using dental acrylic and stainless steel screws, protected with a plastic cap, and occluded using stainless steel, 22-gauge obturators. The animals were treated with dexamethasone and penicillin, from 1 d before surgery to 3 d afterward, as previously described (34,35).

Experiments

Effects of local administration of glutamate receptor antagonists in anestrous ewes

Low dose.

Based on previous work in rats (15) and sheep (36), we chose doses of 5 μg/injection site for both the NMDA receptor antagonist, AP-5 [d(−)-2-amino-5-phosphonopentanoic acid; Sigma-Aldrich Co., St. Louis MO], and the kainate/AMPA (non-NMDA) receptor antagonist DNQX (6,7-dinitroquinoxaline-2,3-dione; Research Biochemicals International, Natick, MA) for our initial experiment. This dose of AP-5 was delivered in 0.3 μl sterile water, but due to solubility constraints, DNQX was delivered in 0.3 μl of 10% dimethylsulfoxide (DMSO) in water. The volume of the microinjection was selected to encompass the A15 without encroaching on the third ventricle (12).

In the initial replicate, we gave seven ovary-intact anestrous ewes bilateral microinjections of AP-5, DNQX, water, or 10% DMSO and collected blood samples by jugular venipuncture every 12 min from 36 min before to 4 h after microinjection. Drugs/vehicles were rapidly administered using a 1-μl Hamilton syringe with an attached needle that extended 2 mm beyond the guide tube. Blood was collected in heparinized tubes, and plasma stored at −20 C for LH analysis. The experimental protocol was repeated four times so that each ewe received all treatments with a 4- to 5-d gap between treatments; treatment order was randomized. Ewes were given gentamycin im at the end of each blood collection.

High dose.

The effects of bilateral microinjections of 25 μg AP-5/0.3 μl water or 0.3 μl water on LH secretion in ovary-intact ewes were tested in these same animals using an identical protocol. To test a higher dose of DNQX, we used a microimplant (12,34,35) to avoid using higher concentrations of DMSO. Crystalline DNQX was tamped into 22-gauge stainless steel tubes that were then lowered into the guide cannulae so they extended 2 mm into tissue and left in place for the 4-h sampling period; empty 22-gauge tubes were used as control. These microimplants contain about 250 μg drug, but not all of the drug was delivered into the ewe because crystalline material was still visible in the lumen of the tubing when the microimplants were removed at the end of the blood collection. As in the previous experiment, all ewes received all treatments in a randomized order and were given gentamycin prophylactically at the end of each treatment.

At the end of these experiments, the sheep were heparinized (two injections of 25,000 U heparin), killed with an overdose of pentobarbital, and their brains perfused with 6 liters 4% paraformaldehyde as previously described (12). Tissue blocks were dissected out, incubated overnight in fixative (4 C), and then infiltrated with 20% sucrose in 0.1 m phosphate buffer (pH 7.4). A series of coronal sections (50 μm thick) through the implantation site were stained with cresyl violet and examined histologically to determine cannula placement.

LH concentrations were measured in duplicate 100- or 200-μl aliquots of serum by RIA as previously described (37); assay sensitivity averaged 0.3 ng/ml (NIH S24), and inter- and intraassay coefficients of variation of a pool that produced approximately 60% displacement of iodinated LH were 8.2 and 14.2%, respectively.

Seasonal comparison of glutamatergic input to A15 DA neurons

To avoid the confounding effects of seasonal differences in circulating ovarian steroids, we used the well characterized (2,9,12,13) E2-treated ovariectomized (OVX+E) ewe model for this experiment. Ewes were ovariectomized during breeding (n = 5) and anestrous (n = 6) seasons, and 3-cm-long SILASTIC brand (Dow Corning Corp., Midland MI) implants containing crystalline E2 were inserted in the axillary region while ewes were under anesthesia. Two weeks later, ewes were killed and their heads perfused with 4% paraformaldehyde as described above. Frozen coronal sections (50 μm thick) were cut and stored at −20 C in cryopreservative until used for immunocytochemistry (ICC).

A series of free floating sections, 250 μm apart through the RCh were used for ICC. All incubations were done on a shaker at room temperature unless otherwise noted, and all rinses consisted of three consecutive 5-min incubations in 0.1 m PBS. Sections were washed overnight in PBS, rinsed, incubated in 4% donkey serum for 1 h to minimize nonspecific binding, and then rinsed and placed in 1:200 mouse anti-TH monoclonal antibody (Chemicon International Inc., Temecula, CA) overnight. Sections were next rinsed and incubated in Alexafluor 555 conjugated to donkey antimouse secondary antibody (Molecular Probes, Eugene OR) for 30 min. Sections were then rinsed and incubated in 8% normal goat serum for 1 h, followed by overnight incubation in 1:10,000 guinea pig anti-vGlut2 antibody (AB5907; Chemicon) at 4 C. This vGlut2 antibody has been used extensively in rodents (33,38,39,40); Western blot analysis using extracts of ovine hypothalamus and this antibody (1:32,000 dilution) resulted in a single band at the expected molecular mass of vGlut2 (52 kDa). The sections were then rinsed, incubated in Alexafluor 488 conjugated to goat anti-guinea pig antiserum (Molecular Probes), rinsed again, mounted in 50% glycerol in PBS, and coverslipped. Controls for both antisera included omission of one primary antiserum, which eliminated staining for the appropriate antigen without affecting staining of the other antigen.

Input to A15 DA and rostral A12 DA neurons, which were medial to posterior A15 cells in the same sections, were analyzed using a Carl Zeiss laser scanning microscope. Scans at each appropriate wavelength were done independently to avoid bleed-through between channels. Ten A15 and three to eight A12 DA neurons, with complete cell bodies visible, were randomly selected in sections from each ewe, and two sets of images were taken along the z-plane (z-stack) for each neuron at an interval of 1 μm. These z-stacks were then converted to a series of tif files, and each neuron was traced through the z-stacks and digitally recreated in three dimensions using Neurolucida 6.0 software (Microbrightfield Inc., Willistan, VT). While tracing the TH-immunoreactive (TH-ir) neuron each apparent vGlut-positive close contact was identified using an assigned marker, which was placed on the trace in that particular stack. The three-dimensional trace with markers was then rotated 360°, and only the markers that touched the neuron in all dimensions were counted as close contacts (13). All analyses were done blind to treatment group.

Triple immunolabeling for vGlut, vGAT, and TH

Additional free-floating sections from four of these ewes (two breeding-season and two anestrous) were labeled with TH and vGlut2 antibodies using the procedure described above. After the 1-h incubation in Alexafluor 488-conjugated goat anti-guinea pig antiserum, the sections were rinsed in PBS and incubated in 8% normal goat serum containing 3% BSA for 1 h, followed by overnight incubation in rabbit anti-vGAT1 antiserum (33) (1:15,000; Chemicon). The vGAT signal was then amplified using the tyramide signal amplification system (12,41). Sections were rinsed and incubated sequentially in 1) 1:200 biotinylated secondary antibody (Jackson ImmunoResearch Laboratories, West Grove, PA) for 1 h, 2) 1:200 Vectastain ABC-elite (Vector Laboratories Burlingame CA) for 1 h, 3) 1:250 tyramide signal amplification biotin system (PerkinElmer Life Sciences Boston, MA) in PBS containing 0.003% H202 for 10 min, and 4) streptavidin conjugated to Alexafluor 647 (Molecular Probes) for 30 min. Sections were rinsed, mounted, coverslipped, and analyzed for triple labeling using confocal microscopy as described above. Omission of one primary antiserum abolished signal for its antigen without affecting staining for the other two antigens. Close contacts containing vGAT or vGlut were then identified with the confocal microscope as described above.

Statistics

LH pulses were identified using previously described criteria (4): pulse amplitude was greater than assay sensitivity, the peak occurred within two samples of the previous nadir, and the peak was 2 sd above the previous and succeeding nadirs, based on assay variability. LH pulse frequency was analyzed by Wilcoxon-Mann-Whitney and Friedman’s two-way ANOVA because parametric statistics are not appropriate for this noncontinuous variable. Mean LH concentrations were analyzed by one-way ANOVA with repeated measures. For ICC data, the number of vGlut-ir close contacts on TH-positive neurons was averaged for each animal and differences between breeding-season and anestrous ewes compared using Student’s t test.

Results

Experiment 1: effect of glutamate receptor antagonists on LH secretion in anestrous ewes

Low dose

Six of the seven ewes received bilateral treatments that were within the A15 group (Fig. 1); the seventh animal, which was not included in the analysis, did not respond to any drug treatments. Four of these six ewes had more LH pulses after microinjection of 5 μg DNQX or AP-5 into the A15 compared with microinjection of vehicle (Fig. 1). However, neither mean LH pulse frequency (Fig. 2) nor mean plasma LH concentration (data not shown) was significantly different between vehicle and antagonist treatment groups.

Figure 1.

Figure 1

Open in a new tab

Effect of glutamate receptor antagonists on LH secretion in ovary-intact anestrous ewes. Left panels, Schematic coronal sections of the rostral (C), middle (B), and claudal (C) depicting bilateral sites of microinjections/microimplants in anestrous ewes. Solid circles, Correct placements; hatched circles, incorrect placement (too dorsal) in one ewe (B). fx, Fornix; OT, optic tract; PVN, paraventricular nucleus; V, third ventricle; VMH, ventromedial hypothalamic nucleus. Right panels, Individual pulse patterns in response to AMPA and NMDA receptor antagonists. Top, LH concentrations in a single ewe after microinjection (arrows) of either 5 μg (19.8 nmol) of the NMDA receptor antagonist AP-5, 5 μg (25.3 nmol) of the non-NMDA receptor antagonist DNQX (right panels), or their respective vehicles (left panels). Bottom, LH concentrations in a single ewe after microinjection of 25 μg (127 nmol) of the NMDA receptor antagonist AP-5 or insertion of a microimplant containing the non-NMDA receptor antagonist DNQX (right panels) and their respective control treatments (left panels). Solid circles identify the peak of LH pulses.

Figure 2.

Figure 2

Open in a new tab

Effects of bilateral microinjections or microimplants of low doses (top panel) and high doses (bottom panel) of glutamate receptor antagonists (black bars) or controls (gray bars) into the A15 of anestrous ewes on mean (± sem) LH pulse frequency.

High dose

Because microinjection of a low dose of either AP-5 or DNQX produced a modest, albeit statistically insignificant, increase in LH pulse frequency, we administered higher doses of the antagonists to determine whether that would elicit a more robust response. Microimplants of DNQX and microinjections of 25 μg AP-5 increased the number of LH pulses in all of the ewes above that seen with control treatments and significantly increased LH pulse frequency (Figs. 1 and 2). Treatment with either antagonist also elicited a significant increase in mean LH concentrations (water 1.1 ± 0.2 vs. AP-5 2.1 ± 0.2 ng/ml, P < 0.05; empty microimplant 0.9 ± 0.1 vs. DNQX 2.4 ± 0.3, P < 0.02) but did not significantly increase LH pulse amplitudes (water 1.7 ± 0.6 vs. AP-5 2.7 ± 0.8, P = 0.08; empty microimplant 1.8 ± 0.9 vs. DNQX 2.9 ± 0.7, P = 0.18).

vGlut2 close contacts on DA neurons in A15

Virtually all of the TH-ir neurons showed vGlut2-ir close contacts in these OVX+E ewes (Fig. 3). Total mean numbers of vGlut-positive close contacts on TH neurons were significantly higher (P < 0.05) in anestrous ewes compared with breeding-season ewes (Fig. 4). Although numbers of vGlut close contacts on TH-containing cell bodies were elevated during anestrus compared with breeding season (8.5 ± 2.3 vs. 4.0 ± 0.5), this did not achieve statistical significance (P = 0.06). In contrast, the numbers of vGlut close contacts on TH-positive dendrites were significantly higher in anestrous animals (P < 0.005). This effect was specific for A15 neurons because no seasonal difference was observed in glutamate input to DA neurons in the proximal A12 (total contacts 6.2 ± 0.7 in anestrus vs. 5.4 ± 0.6 in breeding season). Compared with A15 neurons (Fig. 4), fewer vGlut2-positive contacts were observed on cell bodies (anestrus, 3.8 ± 0.4; breeding season, 3.8 ± 0.5) and dendrites (anestrus, 2.4 ± 0.3; breeding season, 1.6 ± 0.5) of these A12 neurons.

Figure 3.

Figure 3

Open in a new tab

vGlut2-positive close contacts on A15 dopamine neurons in the RCh. A single, 1-μm-thick optical section along the z-plane depicting vGlut2-containing (green) close appositions on TH-positive (red) neurons. Arrows indicate some of the vGlut2-ir close contacts on dendrites of DA neurons. Scale bar, 20 μm.

Figure 4.

Figure 4

Open in a new tab

Seasonal changes in vGlut-positive close contacts on A15 DA neurons. Bars indicate mean numbers ± sem of vGlut-ir close contacts on TH-positive neurons (cell bodies, dendrites, and combined total) in the RCh area of hypothalami collected from OVX+E ewes during anestrus (n = 6) or the breeding season (n = 5). *, P < 0.05, breeding-season vs. anestrous ewes.

vGlut2/vGAT/TH triple immunolabeling

Close appositions containing vGAT were observed on virtually every TH-ir neuron in the A15 area. However, in the RCh, none of the close contacts coexpressed both vGAT1 and vGlut2 immunoreactivity; instead TH-positive cell bodes and dendrites were contacted by distinct vGAT- and vGlut-positive terminals (Fig. 5). As previously reported (12), there were no obvious seasonal differences in the number of GABAergic close contacts onto A15 neurons, but these data were not analyzed statistically because of the low number of animals (two per season).

Figure 5.

Figure 5

Open in a new tab

A single confocal slice along the z-plane (1 μm thickness) depicting vGAT1-containing (blue) and vGlut2-containing (green) terminals contacting TH-positive (red) neuron in the A15 area. Note that individual vGAT-containing (arrowheads) and vGlut-positive (arrows) vesicles make close contact on to TH neurons, but no colocalization of these markers in the same vesicle is observed. Scale bar, 20 μm.

Discussion

The results of the present study support the hypotheses that 1) glutamatergic afferents to A15 neurons mediate, at least in part, the E2 negative feedback-induced suppression of LH pulse frequency in anestrus and that 2) alterations in this input contribute to seasonal changes in responsiveness to E2 feedback. The ability of glutamate receptor antagonists administered locally into the RCh area of ovary-intact anestrous ewes to increase pulsatile LH release in a dose-dependent manner clearly demonstrates that glutamatergic tone in this region holds LH pulse frequency in check in these animals. In theory, this action of glutamate could reflect either the steroid-independent or steroid-dependent inhibition of LH secretion in anestrus (1,4). However, our previous work and that of others showed that A15 DA neurons in the RCh mediate the steroid-dependent, not the steroid-independent, actions of photoperiod (1,7,8,9,42). These A15 neurons are interposed between estrogen-sensitive afferents and GnRH neurons in the neuronal circuit mediating E2 negative feedback during seasonal anestrus (1,9,34,35,43). Thus, the simplest explanation for these and earlier observations is that estrogen increases glutamate release from afferents to this region, which in turn stimulates the activity of A15 neurons to inhibit GnRH and LH pulse frequency. Glutamate could be acting directly on A15 DA neurons or on short interneurons or glial cells in the RCh that then act on DA neurons. The fact that virtually all A15 DA neurons are contacted by vGlut2-containing boutons implies that the negative feedback effects of E2 on GnRH release are likely conveyed via direct synaptic input from glutamatergic neurons onto DA neurons. We thus propose that E2 suppresses GnRH pulse frequency in anestrus by stimulating release of glutamate from synapses onto inhibitory A15 DA neurons afferent to GnRH cells (Fig. 6, left panel).

Figure 6.

Figure 6

Open in a new tab

Proposed pharmacological model for seasonal changes in the neural circuit responsible for E2 negative feedback in the ewe. During anestrus, E2 stimulates glutamatergic and inhibits GABAergic input to A15 DA neurons, resulting in release of DA that inhibits GnRH pulse frequency. During the breeding season, glutamatergic input and the ability of E2 to inhibit GABAergic tone is decreased, resulting in continual inhibition of A15 DA neurons.

This hypothesis is consistent with the report that in ewes, hypothalamic glutamatergic neurons express ERα (14), the ER subtype that mediates E2 negative feedback in ewes (35), and respond to E2 treatment, as indicated by increased Fos expression (14). Moreover, the increased frequency of episodic LH secretion induced by the high dose of either glutamate receptor antagonist approached that seen in ovariectomized anestrous ewes sampled concurrently with this study (12), suggesting that glutamatergic input to the A15 may be the dominant system mediating the negative feedback actions of E2 in anestrus. It should be noted that we have recently reported data supporting a role for GABAergic input to the A15 in E2 negative feedback, but in that case, E2 appears to inhibit GABAergic tone (12). Thus, E2 secretion from the ovary during anestrus would result in maximal glutamate and minimal GABA release onto A15 neurons (Fig. 6) so that blocking the former with a receptor antagonist could be sufficient to disrupt estrogen negative feedback.

The ability of either antagonist to increase LH pulse frequency indicates that NMDA as well as non-NMDA ionotropic receptor subtypes are involved in mediating glutamate-induced stimulation of A15 DA neurons. Previous reports have also implicated multiple glutamate receptors in the control of GnRH and LH secretion. For example, agonists for all three ionotropic receptor subtypes (NMDA, AMPA, and kainite) stimulate GnRH and LH release (15,16,17,18,19,20,21,22,23). In rodents, glutamate and aspartate levels are elevated in the preoptic area during the afternoon of the proestrous LH surge (15,27), and injection of the NMDA antagonist MK801 and the non-NMDA antagonist DNQX inhibits the LH surge (15,24). Thus, all three receptor subtypes appear to mediate the positive effects of E2 on the GnRH surge. Interestingly, the GABAergic modulation of A15 neural activity appears to involve both GABAA and GABAB receptors (12).

Proposed mechanism of the seasonal increase in responsiveness to E2 negative feedback

Several lines of evidence indicate that the A15 neurons that mediate E2 negative feedback in anestrus are not functional during the breeding season (1,7,8,9). Although these changes in control of A15 neural activity appear to be important for seasonal variations in response to E2 negative feedback, the underlying neuroendocrine mechanisms remain unknown. Our observation that the mean number of vGlut2 close contacts on A15 neurons during anestrus was almost 30% higher than during the breeding season suggests that seasonal changes in this stimulatory input could account for seasonal differences in the effects of E2 on A15 neural activity (Fig. 6). These seasonal morphological changes may reflect anatomical differences in synaptic input because the total number of synapses on A15 neurons is higher in anestrus (13). Alternatively, the increase in vGlut-positive close contacts could reflect an increased functional activity of this input in light of evidence that the degree of vGlut expression correlates with quantal release of glutamate at the synapse (44,45). In either case, increased glutamatergic tone would increase the activity of A15 neurons in anestrus.

It should be noted that this seasonal change was observed in OVX+E ewes, not in ovary-intact animals. We chose this model because interpretation of seasonal data in ovary-intact ewes is complicated by changes in endogenous steroids. During the breeding season, ewes have either elevated progesterone or falling progesterone and increasing E2 depending on the stage of the cycle, whereas anestrous animals have undetectable progesterone and infrequent episodes of E2 secretion (1). There is no information on the effects of either progesterone or E2 on vGlut expression in sheep, but E2 does increase vGlut1 expression in rodent hippocampal neurons (46). Because we used a model in which E2 concentrations were held constant, we can conclude that the seasonal differences are not due to changes in this steroid. It will be important to confirm in future work that similar morphological changes occur in ovary-intact ewes, but this seems likely in light of the strong evidence that the neuroendocrine changes observed in OVX+E ewes are responsible for seasonal breeding in ovary-intact animals (1,2,7).

In contrast to the clear seasonal differences in glutamatergic input to A15 neurons, there does not appear to be any difference in GABA-containing close contacts onto A15 neurons between seasons (12). Thus, anatomical alterations in GABAergic afferents are unlikely to be important for seasonal breeding in the ewe. This does not preclude important functional roles for this inhibitory input. At a minimum, continuous GABAergic inhibition throughout the year may be critical for suppressing A15 activity in the breeding season when glutamatergic tone is decreased. In addition, there may be seasonal changes in the activity of the GABAergic input (Fig. 6). For example, if ERα expression in GABAergic afferents decreased during the breeding season, E2 would no longer be able to inhibit GABAergic tone, as it does in anestrus (12), so that A15 neurons would be inactive regardless of circulating E2 concentrations. Thus, we propose that during anestrus when A15 neurons are active, their afferent input is predominantly glutamatergic, whereas during the breeding season when they are inactive, inhibitory GABAergic input predominates.

One key question that remains to be resolved is the location of the glutamatergic and GABAergic afferents to the A15. The ventromedial preoptic area and RCh area are two likely candidates because ERα-containing neurons in each area project to the A15 (47), and E2 microimplants in either area inhibit LH pulse frequency via a DA pathway in anestrus (34,35,48). The arcuate nucleus is a third possibility because many glutamatergic neurons in this region contain ERα, and preliminary data indicated that this is also a site of E2 negative feedback in anestrous ewes (49) and food-restricted lambs (50). It should be noted that these two aminergic inputs most likely arise from separate perikarya because we did not observe any colocalization of vGAT and vGlut in synaptic terminals onto A15 neurons in anestrus or the breeding season.

Only a few previous studies have examined the role of glutamatergic neural systems in mediating seasonal changes in GnRH and LH release. In both rams (51) and ewes (52), the ability of NMDA to stimulate LH secretion is greater when endogenous LH concentrations are suppressed by inhibitory photoperiod. This difference could reflect seasonal changes in glutamatergic tone or an increase in releasable pools of GnRH when GnRH secretion is low (51,52). The report that NMDA stimulates LH in OVX+E ewes, but not in ovariectomized ewes, during anestrus (19) supports the latter explanation. In male hamsters, NMDA is more effective in stimulating LH secretion in both intact and castrated males exposed to inhibitory photoperiod (53,54,55,56). In contrast, no effects of photoperiod were observed on the ability of AMPA to stimulate LH release (55). These workers have thus postulated that an increase in glutamatergic tone acting via the NMDA receptor enhances GnRH secretion during stimulatory photoperiod in the hamster. However, the key experiment testing whether blockade of NMDA receptors prevents the stimulatory effects of photoperiod has yet to be done. It is also important to note that these studies on the role of glutamate in seasonal breeding, and more generally studies on the control of GnRH (15), are most likely examining direct stimulatory effects of glutamate on GnRH neurons. In contrast, this report supports an inhibitory role for glutamate because it appears to be mediating E2 negative feedback by stimulating the inhibitory A15 DA neurons in anestrus.

In summary, the data presented here support the hypothesis that in ovary-intact anestrous ewes, E2 stimulates glutamate release onto A15 DA neurons, which in turn inhibit GnRH pulse frequency. Moreover, the decrease in number of glutamatergic close contacts onto A15 neurons in the breeding season may well contribute to the inability of E2 to stimulate these neurons at this time of year and may represent a critical component of the neural circuitry underlying the seasonal variation in response to E2 negative feedback.

Acknowledgments

We thank Karie Hardy and Heather Clemmer at the West Virginia University Food Animal Research Facility for care of animals and Paul Harton for his technical assistance in sectioning tissue. We also thank Dr. Al Parlow and the National Hormone and Peptide Program for reagents used to measure LH and the West Virginia University Image Analysis Center for use of their confocal microscope.

Footnotes

This work was supported by National Institutes of Health Grant R01 HD017864.

Disclosure Summary: The authors have nothing to disclose.

First Published Online July 9, 2009

Abbreviations: AMPA, α-Amino-3-hydroxyl-5-methyl-4-isoxazole-propionate; AP-5, d(−)-2-amino-5-phosphonopentanoic acid; DA, dopaminergic; DMSO, dimethylsulfoxide; DNQX, 6,7-dinitroquinoxaline-2,3-dione; E2, estradiol; ER, estrogen receptor; GABA, γ-aminobutyric acid; ICC, immunocytochemistry; NMDA, N-methyl-d-aspartate; OVX+E, E2-treated ovariectomized; RCh, retrochiasmatic area; TH, tyrosine hydroxylase; TH-ir, TH-immunoreactive; vGAT, vesicular GABA transporter; vGlut2, vesicular glutamate transporter-2.

References

  1. Goodman RL, Inskeep EI 2006 Neuroendocrine control of the ovarian cycle of the sheep. In: Neill JD, ed. Knobil, Neill’s physiology of reproduction. 3rd ed. Vol 2. Amsterdam: Elsevier; 2389–2447 [Google Scholar]
  2. Karsch FJ, Bittman EL, Foster DL, Goodman RL, Legan SJ, Robinson JE 1984 Neuroendocrine basis of seasonal reproduction. Recent Prog Horm Res 40:185–232 [DOI] [PubMed] [Google Scholar]
  3. Barrell GK, Moenter SM, Caraty A, Karsch FJ 1992 Seasonal changes of gonadotropin-releasing hormone secretion in the ewe. Biol Reprod 46:1130–1135 [DOI] [PubMed] [Google Scholar]
  4. Goodman RL, Bittman EL, Foster DL, Karsch FJ 1982 Alterations in control of LH pulse frequency underlie the seasonal variation in estradiol negative feedback. Biol Reprod 27:580–589 [DOI] [PubMed] [Google Scholar]
  5. Karsch FJ, Dahl GE, Evans NP, Manning JM, Mayfield KP, Moenter SM, Foster DL 1993 Seasonal changes in GnRH secretion in ewe: alteration in response to the negative feedback action of estradiol. Biol Reprod 49:1377–1383 [DOI] [PubMed] [Google Scholar]
  6. Besognet B, Hansen BS, Daels PF 1997 Induction of reproductive function in anestrous mares using a dopamine antagonist. Theriogenology 47:467–480 [DOI] [PubMed] [Google Scholar]
  7. Thiéry JC, Gayrard V, Le Corre S, Viguié C, Martin GB, Chemineau P, Malpaux B 1995 Dopaminergic control of LH secretion by the A15 nucleus in anoestrous ewes. J Reprod Fertil(Suppl 49):285–296 [PubMed] [Google Scholar]
  8. Havern RL, Whisnant CS, Goodman RL 1994 Dopaminergic structures in the ovine hypothalamus mediating estradiol negative feedback in anestrous ewes. Endocrinology 134:1905–1914 [DOI] [PubMed] [Google Scholar]
  9. Lehman MN, Durham DM, Jansen HT, Adrian B, Goodman RL 1996 Dopamine A14/A15 neurons are activated during estradiol negative feedback in anestrous, but not breeding season, ewes. Endocrinology 137:4443–4450 [DOI] [PubMed] [Google Scholar]
  10. Lehman MN, Karsch FJ 1993 Do gonadotropin-releasing hormone, tyrosine hydroxylase, and β-endorphin-immunoreactive neurons contain estrogen receptors? A double-label immunocytochemical study in the Suffolk ewe. Endocrinology 133:887–895 [DOI] [PubMed] [Google Scholar]
  11. Skinner DC, Herbison AE 1997 Effects of photoperiod on estrogen receptor, tyrosine hydroxylase, neuropeptide Y, and β-endorphin immunoreactivity in the ewe hypothalamus. Endocrinology 138:2585–2595 [DOI] [PubMed] [Google Scholar]
  12. Bogusz AL, Hardy SL, Lehman MN, Connors JM, Hileman SM, Sliwowska JH, Billings HJ, McManus CJ, Valent M, Singh SR, Nestor CC, Coolen LM, Goodman RL 2008 Evidence that γ-amino butyric acid is part of the neural circuit mediating estradiol negative feedback in anestrous ewes. Endocrinology 149:2762–2772 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Adams VL, Goodman RL, Salm AK, Coolen LM, Karsch FJ, Lehman MN 2006 Morphological plasticity in the neural circuitry responsible for seasonal breeding in the ewe. Endocrinology 147:4843–4851 [DOI] [PubMed] [Google Scholar]
  14. Pompolo S, Pereira A, Scott CJ, Fujiyma F, Clarke IJ 2003 Evidence for estrogenic regulation of gonadotropin-releasing hormone neurons by glutamatergic neurons in the ewe brain: an immunohistochemical study using an antibody against vesicular glutamate transporter-2. J Comp Neurol 465:136–144 [DOI] [PubMed] [Google Scholar]
  15. Brann DW, Mahesh VB 1994 Excitatory amino acids: function and significance in reproduction and neuroendocrine regulation. Front Neuroendocrinol 15:3–49 [DOI] [PubMed] [Google Scholar]
  16. Zuo Z, Mahesh VB, Zamorano PL, Brann DW 1996 Decreased gonadotropin-releasing hormone neurosecretory response to glutamate agonists in middle-aged female rats on proestrus afternoon: a possible role in reproductive aging? Endocrinology 137:2234–2238 [DOI] [PubMed] [Google Scholar]
  17. Ping L, Mahesh VB, Bhat GK, Brann DW 1997 Regulation of gonadotropin-releasing hormone and luteinizing hormone secretion by AMPA receptors. Evidence for a physiological role of AMPA receptors in the steroid-induced luteinizing hormone surge. Neuroendocrinology 66:246–253 [DOI] [PubMed] [Google Scholar]
  18. Honaramooz A, Chandolia RK, Beard AP, Rawlings NC 1998 Excitatory amino acid regulation of gonadotropin secretion in prepubertal heifer calves. Biol Reprod 59:1124–1130 [DOI] [PubMed] [Google Scholar]
  19. Estienne MJ, Schillo KK, Hileman SM, Green MA, Hayes SH 1990 Effect of N-methyl-d,l-aspartate on luteinizing hormone secretion in ovariectomized ewes in the absence and presence of estradiol. Biol Reprod 42:126–130 [DOI] [PubMed] [Google Scholar]
  20. Ondo JG, Wheeler DD, Dom RM 1988 Hypothalamic site of action for N-methyl-d-aspartate (NMDA) on LH secretion. Life Sci 43:2283–2286 [DOI] [PubMed] [Google Scholar]
  21. Kumar V, Lincoln GA, Tortonese DJ 1993 Effects of excitatory amino acid receptor agonists and antagonists on the secretion of melatonin, luteinizing hormone and prolactin in the ram. J Neuroendocrinol 5:649–654 [DOI] [PubMed] [Google Scholar]
  22. Eyigor O, Jennes L 2000 Kainate receptor subunit-positive gonadotropin-releasing hormone neurons express c-Fos during the steroid-induced luteinizing hormone surge in the female rat. Endocrinology 141:779–786 [DOI] [PubMed] [Google Scholar]
  23. Plant TM, Gay VL, Marshall GR, Arslan M 1989 Puberty in monkeys is triggered by chemical stimulation of the hypothalamus. Proc Natl Acad Sci USA 86:2506–2510 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. López FJ, Donoso AO, Negro-Vilar A 1990 Endogenous excitatory amino acid neurotransmission regulates the estradiol-induced LH surge in ovariectomized rats. Endocrinology 126:1771–1773 [DOI] [PubMed] [Google Scholar]
  25. Ottem EN, Godwin JG, Petersen SL 2002 Glutamatergic signaling through the N-methyl d-aspartate receptor directly activates medial subpopulations of LHRH neurons, but does not appear to mediate the effects of estradiol on LHRH gene expression. Endocrinology 143:4837–4845 [DOI] [PubMed] [Google Scholar]
  26. Bailey JD, Centers A, Jennes L 2006 Expression of AMPA receptor subunits (GluR1–GluR4) in gonadotrophin-releasing hormone neurones of young and middle-aged persistently oestrous rats during the steroid-induced luteinising hormone surge. J Neuroendocrinol 18:1–12 [DOI] [PubMed] [Google Scholar]
  27. Jarry H, Hirsch B, Leonhardt S, Wuttke W 1992 Amino acid neurotransmitter release in the preoptic area of rats during the positive feedback actions of estradiol on LH release. Neuroendocrinology 56:133–140 [DOI] [PubMed] [Google Scholar]
  28. Terasawa E, Luchansky LL, Kasuya E, Nyberg CL 1999 An increase in glutamate release follows a decrease in γ-aminobutyric acid and the pubertal increase in luteinizing hormone releasing hormone release in female Rhesus monkeys. J Neuroendocrinol 11:275–282 [DOI] [PubMed] [Google Scholar]
  29. Wu FC, Howe DC, Naylor AM 1990 N-Methyl-d,l-aspartate receptor antagonism by d-2-amino-5-phosphonovaleric acid delays onset of puberty in the female rat. J Neuroendocrinol 2:627–231 [DOI] [PubMed] [Google Scholar]
  30. I'Anson H, Herbosa CG, Ebling FJ, Wood RI, Bucholtz DC, Mieher CD, Foster DL, Padmanabhan V 1993 Hypothalamic versus pituitary stimulation of luteinizing hormone secretion in the prepubertal female lamb. Neuroendocrinology 57:467–475 [DOI] [PubMed] [Google Scholar]
  31. Gore AC, Yeung G, Morrison JH, Oung T 2000 Neuroendocrine aging in the female rat: the changing relationship of hypothalamic gonadotropin-releasing hormone neurons and N-methyl-d-aspartate receptors. Endocrinology 141:4757–4767 [DOI] [PubMed] [Google Scholar]
  32. Herzog E, Bellenchi GC, Gras C, Bernard V, Ravassard P, Bedet C, Gasnier B, Giros B, El Mestikawy S 2001 The existence of a second vesicular glutamate transporter specifies subpopulations of glutamatergic neurons. J Neurosci 21:RC181 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Ottem EN, Godwin JG, Krishnan S, Petersen SL 2004 Dual phenotype GABA/glutamate neurons in adult preoptic area: sexual dimorphism and function. J Neurosci 24:8097–8105 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Anderson GM, Connors JM, Hardy SL, Valent M, Goodman RL 2001 Oestradiol microimplants in the ventromedial preoptic area inhibit secretion of luteinising hormone via dopamine neurones in anoestrous ewes. J Neuroendocrinol 13:1051–1058 [DOI] [PubMed] [Google Scholar]
  35. Hardy SL, Anderson GM, Valent M, Connors JM, Goodman RL 2003 Evidence that estrogen receptor α, but not β, mediates seasonal changes in the response of the ovine retrochiasmatic area to estradiol. Biol Reprod 68:846–852 [DOI] [PubMed] [Google Scholar]
  36. Hileman SM, Schillo KK, Estienne MJ 1992 Effects of intracerebroventricular administration of d,l-2-amino-5-phosphonovaleric acid, an N-methyl-d-aspartate receptor antagonist, on luteinizing hormone release in ovariectomized lambs. Biol Reprod 47:1168–1172 [DOI] [PubMed] [Google Scholar]
  37. Goodman RL, Coolen LM, Anderson GM, Hardy SL, Valent M, Connors JM, Fitzgerald ME, Lehman MN 2004 Evidence that dynorphin plays a major role in mediating progesterone negative feedback on gonadotropin-releasing hormone neurons in sheep. Endocrinology 145:2959–2967 [DOI] [PubMed] [Google Scholar]
  38. Lin W, McKinney K, Liu L, Lakhlani S, Jennes L 2003 Distribution of vesicular glutamate transporter-2 messenger ribonucleic acid and protein in septum-hypothalamus of the rat. Endocrinology 144:662–670 [DOI] [PubMed] [Google Scholar]
  39. Todd AJ, Hughes DI, Polgár E, Nagy GG, Mackie M, Ottersen OP, Maxwell DJ 2003 The expression of vesicular glutamate transporters VGLUT1 and VGLUT2 in neurochemically defined axonal populations in the rat spinal cord with emphasis on the dorsal horn. Eur J Neurosci 17:13–27 [DOI] [PubMed] [Google Scholar]
  40. Kiss J, Halász B, Csáki A, Liposits Z, Hrabovszky E 2007 Vesicular glutamate transporter 2 protein and mRNA containing neurons in the hypothalamic suprachiasmatic nucleus of the rat. Brain Res Bull 74:397–405 [DOI] [PubMed] [Google Scholar]
  41. Hunyady B, Krempels K, Harta G, Mezey E 1996 Immunohistochemical signal amplification by catalyzed reporter deposition and its application in double immunostaining. J Histochem Cytochem 4:1353–1362 [DOI] [PubMed] [Google Scholar]
  42. Meyer SL, Goodman RL 1986 Separate neural systems mediate the steroid-dependent and steroid-independent suppression of tonic luteinizing hormone secretion in the anestrous ewe. Biol Reprod 35:562–571 [DOI] [PubMed] [Google Scholar]
  43. Gayrard V, Malpaux B, Tillet Y, Thiéry JC 1994 Estradiol increases tyrosine hydroxylase activity of the A15 nucleus dopaminergic neurons during long days in the ewe. Biol Reprod 50:1168–1177 [DOI] [PubMed] [Google Scholar]
  44. Moutsimilli L, Farley S, Dumas S, El Mestikawy S, Giros B, Tzavara ET 2005 Selective cortical VGLUT1 increase as a marker for antidepressant activity. Neuropharmacology 49:890–900 [DOI] [PubMed] [Google Scholar]
  45. Daniels RW, Collins CA, Gelfand MV, Dant J, Brooks ES, Krantz DE, DiAntonio A 2004 Increased expression of the Drosophila vesicular glutamate transporter leads to excess glutamate release and a compensatory decrease in quantal content. J Neurosci 24:10466–10474 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Jelks KB, Wylie R, Floyd CL, McAllister AK, Wise P 2007 Estradiol targets synaptic proteins to induce glutamatergic synapse formation in cultured hippocampal neurons: critical role of estrogen receptor-α. J Neurosci 27:6903–6913 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Lehman MN, Coolen LM, Goodman RL, Viguié C, Billings HJ, Karsch FJ 2002 Seasonal plasticity in the brain: the use of large animal models for neuroanatomical research. Reprod Suppl 59:149–165 [PubMed] [Google Scholar]
  48. Gallegos-Sánchez J, Delaleu B, Caraty A, Malpaux B, Thiéry JC 1997 Estradiol acts locally within the retrochiasmatic area to inhibit pulsatile luteinizing hormone release in the female sheep during anestrus. Biol Reprod 56:1544–1549 [DOI] [PubMed] [Google Scholar]
  49. Billings HJ, Geer SN, Holaskova I, Goodman RL 2008 Estradiol in the arcuate nucleus inhibits LH pulse frequency and steroid receptor coactivator-1 colocalization with estrogen receptor-α. Biol Reprod 78(Suppl 1):520 (Abstract) [Google Scholar]
  50. McManus CJ, Goodman RL, Llanza NV, Valent M, Dobbins AB, Connors JM, Hileman SM 2005 Inhibition of luteinizing hormone secretion by localized administration of estrogen, but not dihydrotestosterone, is enhanced in the ventromedial hypothalamus during feed restriction in the young wether. Biol Reprod 73:781–789 [DOI] [PubMed] [Google Scholar]
  51. Lincoln GA, Wu FCW 1991 Luteinizing hormone responses to N-methyl-d,l-aspartate during a photoperiodically induced reproductive cycle in the ram. J Neuroendocrinol 3:309–317 [DOI] [PubMed] [Google Scholar]
  52. Viguie C, Caraty A, Locatelli A, Malpaux B 1995 Regulation of luteinizing hormone-releasing hormone (LHRH) secretion by melatonin in the ewe. II. Changes in N-methyl-d,l-aspartic acid-induced LHRH release during the stimulation of luteinizing hormone secretion by melatonin. Biol Reprod 52:1156–1161 [DOI] [PubMed] [Google Scholar]
  53. Urbanski HF 1990 A role for N-methyl-d-aspartate receptors in the control of seasonal breeding. Endocrinology 127:2223–2228 [DOI] [PubMed] [Google Scholar]
  54. Urbanski HF 1992 Photoperiodic modulation of luteinizing hormone secretion in orchidectomized Syrian hamsters and the influence of excitatory amino acids. Endocrinology 131:1665–1669 [DOI] [PubMed] [Google Scholar]
  55. Ebling FJ, Hui Y, Mirakhur A, Maywood ES, Hastings MH 1993 Photoperiod regulates the LH response to central glutamatergic stimulation in the male Syrian hamster. J Neuroendocrinol 5:608–618 [DOI] [PubMed] [Google Scholar]
  56. Ebling FJ, Mirakhur A, Maywood ES, Hastings MH 1994 Photoperiodically induced changes in glutamatergic stimulation of LH secretion in male Syrian hamsters: role of circulating testosterone and endogenous opioids. Gen Comp Endocrinol 96:50–62 [DOI] [PubMed] [Google Scholar]

Articles from Endocrinology are provided here courtesy of The Endocrine Society

RESOURCES