|Home | About | Journals | Submit | Contact Us | Français|
A surge of gonadotropin-releasing hormone (GnRH) release from the brain triggers the luteinizing hormone (LH) surge that causes ovulation. The GnRH surge is initiated by a switch in estradiol action from negative to positive feedback. Estradiol signals critical for the surge are likely transmitted to GnRH neurons at least in part via estradiol-sensitive afferents. Using an ovariectomized estradiol-treated (OVX++E) mouse model that exhibits daily LH surges, we examined changes in glutamate transmission to GnRH neurons during negative feedback and positive feedback. Spontaneous glutamatergic excitatory postsynaptic currents (EPSCs) mediated by alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid/kainate receptors (AMPA/KA Rs) or N-methyl-d-aspartate receptors (NMDARs) were recorded in GnRH neurons from OVX++E and OVX mice. There were no diurnal changes in the percentage of GnRH neurons from OVX mice exhibiting EPSCs. In cells from OVX++E mice, the profile of AMPA/KA R-mediated and NMDAR-mediated EPSCs showed changes dependent on time of day. Comparison of AMPA/KA R-mediated EPSC frequency in OVX++E and OVX cells showed that estradiol suppressed transmission during negative feedback but had no effect during positive feedback. Tetrodotoxin treatment to block action potential firing did not affect AMPA/KA R-mediated EPSC frequency in OVX cells during negative feedback or in OVX++E cells during positive feedback, suggesting that estradiol-induced suppression of glutamate transmission may be primarily due to activity-independent changes. The diurnal removal of estradiol-induced suppression of AMPA/KA R-mediated glutamate transmission to GnRH neurons during positive feedback suggests that the primary role for estradiol-induced changes in glutamate transmission may be in mediating negative feedback.
The gonadotropin-releasing hormone (GnRH) neuronal network forms the final common pathway in the central regulation of fertility. Gonadal steroid feedback modulates GnRH neuron function and pituitary gland responsiveness to GnRH [1–4]. In females, the feedback action of ovarian estradiol switches from negative to positive at the end of the follicular phase of the reproductive cycle when estradiol levels peak (proestrus in rodents), leading to an increase in GnRH neuron firing activity  and a large surge of GnRH release [6–8]. The GnRH surge initiates a surge of luteinizing hormone (LH) release by the pituitary, which subsequently triggers ovulation. In rodents, the GnRH surge seems dependent not only on high estradiol levels but also on a daily neural signal, which times the GnRH surge to an appropriate time of day (i.e., late afternoon in nocturnal species) [5, 9, 10].
Estradiol may act directly and indirectly on GnRH neurons to mediate negative feedback and positive feedback effects [11–19]. Estradiol signals crucial for the control of ovulation are likely mediated at least in part via estradiol-sensitive afferents. In particular, GnRH neurons may not express estrogen receptor alpha (also known as ESR1), which is critical for negative feedback and positive feedback regulation of GnRH neurons [4, 13, 17, 20, 21]. In this regard, glutamatergic neurons in the hypothalamus express estrogen receptor alpha [22, 23], and GnRH neurons express the alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid/kainate (AMPA/KA) and N-methyl-d-aspartate (NMDA) subtypes of ionotropic glutamate receptors [24, 25]. Several lines of evidence indicate a role for glutamate, a major mediator of fast synaptic transmission, in surge regulation. Glutamate release in the preoptic area increases during the LH surge , and blockade of NMDA or AMPA/KA glutamate receptors blocks the LH surge [27–29]. A limitation of these approaches, however, is that measures of glutamate concentrations or administration of glutamate or agonists of its receptors do not provide information regarding the phenotype of the cells (i.e., GnRH neurons and/or other cells) affected by these changes in glutamate levels or treatments.
Synaptic activation of AMPA receptors in GnRH neurons has been observed , but endogenous activation of NMDA receptors (NMDARs) has not been reported. Herein, we investigate the relationship between glutamatergic transmission to GnRH neurons and estradiol negative feedback and positive feedback. We used a mouse model in which ovariectomy and treatment with an estradiol implant (OVX+E) producing a physiological level of steroid in the circulation induce daily LH surges for several days . We tested the hypotheses that glutamatergic transmission to GnRH neurons from OVX+E mice is suppressed relative to that in OVX mice during negative feedback and is increased relative to that in OVX mice during positive feedback.
Adult (2- to 4-mo-old) female transgenic mice expressing green fluorescent protein (GFP) in GnRH neurons , which allows for identification and targeted recording of GnRH neurons, were used for all experiments. Mice were housed under a 14L:10D photoperiod with lights off at 1630 h eastern standard time, with Harlan (Indianapolis, IN) 2916 chow and water available ad libitum. Two to four days before electrophysiology experiments, 37 mice were ovariectomized (OVX) under isoflurane (Burns Veterinary Supply, Westbury, NY) anesthesia and were implanted s.c. with a 1.8-cm-long Silastic capsule (1.98-mm internal × 3.18-mm external diameter, No. 508–009; Dow-Corning, Midland, MI) containing 0.625 μg of 17beta-estradiol suspended in sesame oil (OVX+E mice), and 32 mice were not treated further (OVX mice). As described previously , OVX+E treatment in this manner induces daily LH surges that peak around the time of lights off, and circulating estradiol levels (~30–35 pg/ml) are constant over successive days. It should be noted that no diurnal changes in GnRH neuron activity or LH release occur in OVX mice . For OVX mice, the times of day of the experiments were matched to times of negative feedback and positive feedback states in OVX+E mice for comparison. Long-acting postoperative local analgesia was provided by infiltration of each surgical site with 7 μl of 0.25% bupivacaine (Abbott Laboratories, North Chicago, IL). Estradiol was solely administered in vivo and was not present in any recording solutions. All procedures were approved by the University of Virginia Animal Care and Use Committee and were conducted in accord with the National Research Council's Guide for the Care and Use of Laboratory Animals.
All reagents were purchased from Sigma (St. Louis, MO) except where noted. Brain slices were prepared using slight modifications [5, 32] of previously described methods [12, 33]. Briefly, mice were euthanized by decapitation at 0900–0930 h (negative feedback groups) or at 1430–1500 h (positive feedback groups). The brain was rapidly removed and placed in ice-cold high-sucrose saline containing 250 mM sucrose, 3.5 mM KCl, 26 mM NaHCO3, 10 mM d-glucose, 1.3 mM Na2HPO4, 1.2 mM MgSO4, and 3.8 mM MgCl2. Sagittal (300 μm) slices were cut with a Vibratome 3000 (Ted Pella, Inc., Redding, CA). Slices were incubated 30 min at 30–32°C in 50% high-sucrose saline and 50% normal saline (NS) solution, with NS containing 135 mM NaCl, 3.5 mM KCl, 26 mM NaHCO3, 10 mM d-glucose, 1.25 mM Na2HPO4, 1.2 mM MgSO4, and 2.5 mM CaCl2 (pH 7.4). Slices were then transferred to 100% NS solution at room temperature (~21–23°C) for 0.5–2.5 h. For recording, slices were placed in a recording chamber on the stage of an Olympus Corp. BX50WI upright fluorescent microscope (Opelco, Dulles, VA) and were continuously superfused at 5–6 ml/min with oxygenated NS maintained at 30–32°C with an inline heating unit (Warner Instruments, Hamden, CT). Slices were stabilized in the recording chamber for at least 5 min before recording.
GnRH-green fluorescent protein neurons in the preoptic area and ventral hypothalamus were identified by brief illumination at 470 nm. Experiments were performed using an EPC 8 amplifier (HEKA Elektronik, Lambrecht/Pfalz, Germany) with the Pulse Control XOP (Instrutech, Port Washington, NY) running in Igor Pro software (Wavemetrics, Lake Oswego, OR) on a G4 Macintosh computer (Apple Inc., Cupertino, CA). Signals were digitized at 16-bit resolution through an ITC-18 acquisition interface (Instrutech). Recording pipettes (2–4 MΩ) were filled with a cesium gluconate-based pipette solution containing 125 mM d-gluconic acid, 80 mM CsCl, 10 mM Hepes, 5 mM ethyleneglycoltetracetic acid, and 0.1 mM CaCl2, with the addition of 4 mM MgATP and 0.4 nM NaATP before adjusting to pH 7.2 with CsOH. To isolate glutamatergic currents, the gamma aminobutyric acid (GABAA) receptor blocker picrotoxin (100 μM) was included in all recording solutions. In some recordings, tetrodotoxin (TTX) (0.5 μM; Calbiochem, La Jolla, CA) was added to block action potentials. To record excitatory postsynaptic currents (EPSCs) mediated by AMPA/KA receptors (AMPA/KA Rs), membrane potential was clamped at −70 mV; in a subset of experiments, the AMPA/KA R blocker 6-cyano-7-nitroquinoxaline (CNQX [10 μM]) was applied to confirm that these currents were mediated by AMPA/KA Rs. For currents mediated by NMDARs, membrane potential was clamped at 40 mV; these currents were blocked by the NMDAR antagonist d-2-amino-5-phosphonovaleric acid (APV [20 μM]). The depolarized holding potential removes the voltage-dependent pore block in the NMDAR by extracellular Mg2+  and thus reveals NMDAR-mediated EPSCs. Signals were low-pass filtered at 7 kHz with gain set at 10 mV/pA. Under these recording conditions, AMPA/KA EPSCs appear as inward currents, and NMDA EPSCs appear as outward currents. Liquid junction potential of 7 mV  was not corrected for.
Recordings were performed between 1100 and 1400 h (negative feedback) or between 1600 and 1900 h (positive feedback). An average of two cells were recorded per animal. Input resistance (Rin), series resistance (Rs), and membrane capacitance (Cm) at a holding potential of −60 mV were monitored every 120–180 sec as described previously [14, 36]. Only recordings with Rin >500 MΩ, Rs <20 MΩ, Cm >10 pF, and holding current at −60 mV between 0 and −100 pA were included in analyses. There were no differences among groups in any passive recording properties or Rs attributable to estradiol treatment or time of day of recording. The location of each GnRH neuron recorded was mapped on figures of sagittal sections obtained from a mouse brain atlas . Cell location, however, did not affect the data.
Multiple 120-sec recordings from each cell were stored as Event Tracker files using Pulse Control XOP and Igor Pro software. The mean event frequency (in hertz) was calculated from two to three 120-sec records for each cell to obtain the mean postsynaptic current frequency for each group. For examination of TTX effects, the last two of three 120-sec records obtained in the presence of TTX were used in analysis to allow for exchange of bath solutions. Stored traces were analyzed off-line using custom event detection software in Igor Pro  to identify EPSCs. The EPSCs were confirmed by eye, and detection errors were corrected manually. Recordings obtained at −70 and 40 mV were analyzed to determine event frequency and amplitude for AMPA/KA and NMDA EPSCs, respectively, for each cell. Averaged EPSC waveforms for each cell were generated after aligning events on the rising phase. Data were transferred to Excel (Microsoft, Redmond, WA) and to InStat or Prism4 (Graph Pad Software, San Diego, CA) for further analysis. Data were log transformed, and group means were compared using two-way ANOVA, followed by Bonferroni multiple comparisons tests. The EPSCs before and during TTX treatment were compared using two-tailed paired t-tests. Cumulative probability plots were created by using 100 randomly selected events per cell or all events if <100 were recorded. Probability distributions were compared using two-sample Kolmogorov-Smirnov goodness-of-fit tests (S-PLUS 8.0; Insightful Corp., Seattle, WA). When more than one cell from an animal was included in a given treatment group, within-animal variance was similar to between-animal variance; thus, cells are considered independent observations. Similarly, values from Days 2 to 4 after surgery were grouped together, as variance among days was similar to variance within values for each day for each treatment group. Data are presented as means ± SEM. Statistical significance was set at P < 0.05.
To test the hypothesis that glutamate transmission changes during the time of the GnRH surge in estradiol-treated mice, spontaneous glutamatergic postsynaptic currents (EPSCs) were recorded during negative (−FB) and positive feedback (+FB) from GFP-identified GnRH neurons in brain slices obtained from OVX+E (−FB: n = 16 cells from 11 mice; +FB: n = 25 cells from 12 mice) or OVX (−FB: n = 13 cells from 7 mice; +FB: n = 10 cells from 7 mice) animals. The EPSCs were recorded at membrane holding potentials of −70 and 40 mV; EPSCs at −70 mV showed fast kinetics and were blocked by the AMPA/KA R antagonist CNQX (10 μM), whereas currents at 40 mV showed slower kinetics and were blocked by the NMDAR antagonist APV (20 μM) (Fig. 1).
In cells from OVX and OVX+E mice, approximately 20%–35% of GnRH neurons did not exhibit EPSCs of either subtype, suggesting that many GnRH neurons (independent of estradiol treatment or time of day) do not receive fast glutamatergic transmission, at least as detected by recordings at the cell body under these conditions. With regard to the cells that showed EPSCs, cells from OVX mice showed no difference in the profile of current subtype expression with time of day (Fig. 2). In contrast, in GnRH neurons from OVX+E mice, some cells recorded during positive feedback (13%) showed only NMDAR EPSCs; no cells recorded during negative feedback showed NMDAR EPSCs in the absence of AMPA/KA R EPSCs. Because of the small percentage of cells that exhibited NMDAR currents (consistent with previous findings in which approximately 20% of GnRH neurons responded to NMDAR activation ), we focused on AMPA/KA R-mediated transmission for the remainder of the studies herein.
Examination of AMPA/KA R EPSC frequency in GnRH neurons from OVX mice exhibiting these currents at each time of day showed no diurnal change in frequency (0.29 ± 0.07 Hz in 31 cells from 20 mice [negative feedback] and 0.30 ± 0.10 Hz in 15 cells from 11 mice [positive feedback]), consistent with the lack of a diurnal change in GnRH neuron firing activity or LH release in similarly treated animals . Comparison of cells from OVX+E and OVX mice (F1,74 = 6.84, P = 0.01; two-way ANOVA) showed that estradiol decreased EPSC frequency during negative feedback (0.06 ± 0.02 Hz in 10 cells from eight mice [P < 0.05]), but there was no difference in EPSC frequency between cells from OVX and OVX+E mice during positive feedback (0.35 ± 0.09 Hz in 33 cells from 22 mice). In OVX+E cells, EPSC frequency was significantly increased during positive feedback compared with that during negative feedback (P < 0.01), and the interevent interval was decreased correspondingly (P < 0.01) (Fig. 3). The primary effect of estradiol on the frequency of glutamatergic transmission to GnRH neurons thus appears to be suppression during negative feedback.
Changes in EPSC frequency may be due to altered firing activity of presynaptic cells and/or activity-independent effects such as changes in synaptic connectivity or vesicle release probability [39, 40]. To begin to test the hypothesis that estradiol alters the firing activity of glutamatergic cells, EPSCs were recorded before and during in vitro addition of TTX (0.5 μM) to block action potential firing and thus activity-dependent presynaptic glutamate release. The EPSCs recorded in the presence of TTX are attributable to random action potential-independent vesicle release and are referred to as miniature EPSCs. Decreases in EPSC frequency are difficult to investigate, as TTX treatment would not reveal if diminished synaptic drive is due to decreased firing activity. We therefore took the approach of testing the effect of TTX on OVX cells recorded during the normal time of negative feedback. The rationale was that, if TTX decreased EPSC frequency in GnRH neurons from OVX mice, this would suggest that the basal level of glutamate release in the absence of estradiol is susceptible to action potential blockade; thus, decreased activity is a possible mechanism for estradiol-induced suppression of EPSC frequency during negative feedback. However, TTX did not affect EPSC frequency in OVX cells recorded during negative feedback (n = 9, 7.9% ± 14.4%; P > 0.6) (Fig. 4), suggesting that the estradiol-induced suppression of glutamate transmission during negative feedback may be primarily due to activity-independent changes.
To examine if changes in firing activity may drive the diurnal relief of estradiol suppression of glutamate transmission, we next tested the effect of TTX on EPSC frequency in OVX+E cells recorded during positive feedback. Similar to the lack of effect of TTX on OVX cells during negative feedback, there was no effect of TTX on EPSC frequency in OVX+E cells during positive feedback (n = 9, −13.2% ± 9.0%; P > 0.16) (Fig. 4). The diurnal change in EPSC frequency in GnRH neurons from OVX+E mice therefore appears to be due to activity-independent suppression of glutamate transmission during negative feedback.
To examine if the response of GnRH neurons to AMPA/KA R activation may change in a diurnal or estradiol-dependent manner, we examined changes in EPSC amplitude, a measure of conductance. In OVX+E cells, EPSC amplitude was decreased during negative feedback compared with that during positive feedback (P < 0.01) (Fig. 5). Comparison of OVX and OVX+E cells showed that estradiol decreased EPSC amplitude during negative feedback (P < 0.01), but there was no difference during positive feedback (P > 0.05). Within-cell comparisons before and during TTX treatment showed no effect of TTX on EPSC amplitude in OVX cells during negative feedback (−4.75% ± 3.85%, P > 0.16) or in OVX+E cells during positive feedback (−2.48% ± 2.59%, P > 0.45), suggesting that estradiol acts via activity-independent mechanisms to alter EPSC amplitude, similar to changes in EPSC frequency, and that this suppression is removed during positive feedback.
The neurobiological mechanisms underlying estradiol feedback to GnRH neurons are not fully elucidated, although a growing body of work [4, 13, 16, 19, 41] indicates that synaptic inputs to GnRH neurons have a critical role. Herein, we present evidence that glutamate transmission to GnRH neurons mediated by both AMPA/KA Rs and NMDARs exhibits shifts in association with the changes in GnRH neuron firing activity and pituitary hormone release characteristic of changing estradiol feedback states . The results herein also provide novel functional evidence for synaptic activation of NMDARs on GnRH neurons. Although normal network circuit behavior is disrupted by brain slice preparation, the ability to examine glutamate transmission directly at GnRH neurons outweighs this caveat, and the slice preparation has been used extensively to characterize synaptic transmission throughout the brain [39, 42, 43].
Together with anatomical studies in the literature [23, 44, 45], these data suggest that the GnRH neuron population may be subdivided into groups that may be defined by the type of fast synaptic transmission received and the plasticity of that transmission. Electrophysiological analyses of nucleated patches of GnRH neuronal membrane proximal to the soma revealed that most GnRH neurons respond to AMPA agonists but that only about 20% respond to NMDA agonists . In the present studies, we observed spontaneous NMDAR-mediated currents in a similar percentage of GnRH neurons, but a lower percentage of GnRH neurons exhibited spontaneous AMPA/KA R-mediated EPSCs compared with the percentage of patches that responded to AMPA agonists in the previous study . Indeed, approximately 20%–35% of GnRH neurons tested did not exhibit any EPSCs that were detectable by recordings at the cell body. Taken together, these data suggest that although the receptors are expressed (located proximal to the cell body and functional) they may be silent in some cases (i.e., lacking presynaptic activation). In contrast, essentially all GnRH neurons that we have recorded from [38, 46–48] receive detectable GABAergic transmission. Although controversial [49, 50], GABAA receptor activation has been reported to depolarize and excite even mature GnRH neurons [36, 38, 51, 52]. With regard to GABA, a functional plasticity subgroup exists; in the same animal model used in the present study, approximately one third of GnRH neurons were found to receive elevated GABAergic transmission during positive feedback .
The one fifth to one third of GnRH neurons in which no spontaneous EPSCs were observed at the soma may still be subjected to regulation via ionotropic glutamatergic receptors. One possibility is that in some cells the synaptic inputs are located far from the cell body on distal dendrites  and that the degree of signal decay diminishes reliable detection via somatic patch clamp [54, 55]. Such inputs could still affect GnRH neuron function, as recent evidence suggests that dendrites in GnRH neurons are active and can initiate action potential firing . At the other end of the cell, ionotropic glutamate receptor expression on GnRH neuron terminals in the median eminence  suggests the possibility of EPSCs in the terminals that are not propagated back to the cell body and thus are not recorded in this configuration. Depolarizing synaptic currents in presynaptic terminals have been shown to enhance neurotransmitter release ; thus, glutamatergic transmission at GnRH neuron terminals may alter GnRH release as well. Another possibility is that one subpopulation of GnRH neurons may receive elevated GABA transmission but no glutamate transmission during positive feedback, while another subpopulation is targeted for increases in both GABA and glutamate transmission. Recent modeling work suggests that synaptic GABA and AMPA inputs can interact to modulate, and in some cases amplify, action potential firing in GnRH neurons . In this regard, blocking of ionotropic GABA and glutamate receptors in brain slices disrupts the normal pattern of GnRH neuron firing during both negative feedback and positive feedback . The role of interactions between GABAergic and glutamatergic inputs in modulating GnRH neuron activity in different estradiol feedback states remains to be determined.
The population of cells that exhibited only NMDAR EPSCs during positive feedback is also notable in this regard, as NMDARs often function as coincidence detectors in forms of long-term potentiation, typically in cooperation with AMPA/KA Rs . The depolarizing action of GABA leads to the postulate that in at least some GnRH neurons NMDARs may act as coincidence detectors in cooperation with GABAA receptors. That is, GABAA receptor activation could depolarize the postsynaptic membrane potential to a level sufficient to remove the Mg2+ pore block in NMDARs and thus potentiate synaptic activation and drive increased GnRH neuron activity during positive feedback.
The estradiol-dependent decrease in EPSC frequency during negative feedback was accompanied by decreased EPSC amplitude, suggesting a reduction in ion conductance on AMPA/KA R activation in GnRH neurons. Altered postsynaptic current amplitude can be due to postsynaptic changes such as differences in the number of receptors or alterations in receptor phosphorylation or composition, as well as presynaptic effects such as decreased neurotransmitter vesicular content [61–63]. Anatomical findings have indicated that the expression of AMPA receptor subunits in GnRH neurons changes in a diurnal manner associated with the LH surge ; this may account at least in part for the changes in AMPA/KA R EPSC amplitude observed herein. Larger EPSCs may lead to greater depolarization in postsynaptic membrane potential. Thus, in combination with increased GABA postsynaptic current amplitude  and intrinsic effects of estradiol on GnRH neurons to hyperpolarize action potential threshold  and to raise cellular excitability , increased EPSC amplitude would enhance the probability for action potential firing in postsynaptic GnRH neurons. Decreased EPSC amplitude during negative feedback is thus consistent with lowered firing activity and reduced LH levels at this time .
In all groups tested, AMPA/KA R-mediated EPSCs were typically not observed in recordings obtained at a membrane holding potential of 40 mV, indicating that these currents may be rectified in GnRH neurons. The AMPA receptors that lack the GluR2 (also known as GRIA2) subunit are Ca2+ permeable and display inward rectification , and these receptors have been implicated in forms of synaptic strengthening [65, 66]. Immunocytochemical evidence indicates that GnRH neurons express GRIA2 , but the data presented herein provide some preliminary functional evidence that in at least some GnRH neurons (or under certain conditions) GRIA2-lacking AMPA receptors may be expressed.
The sources of glutamate input that may underlie the observed changes in glutamate transmission are unknown. Several areas in the hypothalamus (including the preoptic area and ventromedial, dorsomedial, supraoptic, paraventricular, and arcuate nuclei) express type 2 vesicular glutamate transporters (VGLUT2, official symbol SLC17A6), an anatomic marker for glutamatergic cells [23, 67]. Estrogen-responsive SLC17A6-positive cells are present in the anteroventral periventricular area (AVPV), ventromedial nucleus, and medial preoptic nucleus . The AVPV, a critical area with regard to estradiol regulation of GnRH [13, 23, 68–70], is the most likely candidate for mediating estradiol-induced and diurnal changes in glutamate transmission to GnRH neurons, as these cells synapse directly on GnRH neurons . Although the overall number of AVPV-derived synaptic contacts that contain SLC17A6 does not change with time of day, the number of dual-phenotype contacts that express both vesicular GABA transporters and SLC17A6 (as well as SLC17A6 immunoreactivity within these terminals) increases around the onset of the LH surge . Furthermore, the dual-phenotype terminals are only detectable with estradiol treatment. Herein, no difference in EPSC frequency was detected between cells from OVX vs. OVX+E animals during positive feedback; the functional role of these dual-phenotype terminals in driving changes in glutamate transmission to GnRH neurons remains to be determined. These anatomical changes, however, may reflect at least part of the activity-independent suppression in glutamate transmission during negative feedback, as well as its diurnal relief as part of the switch from negative to positive feedback. GnRH neurons themselves also express SLC17A6 , suggesting the possibility that GnRH neurons may form a short-loop feedback circuit via autaptic and/or GnRH neuron-GnRH neuron connections mediated at least in part by glutamate .
These studies were performed using an endocrine model in which LH surges are induced on successive days, a situation different from the normal mouse estrous cycle in which an LH surge only occurs on the day of proestrus. Although the circulating level of estradiol is physiological [5, 72], the constant treatment pattern is not. The advantage of a one-variable model, however, allows us to examine mechanism in a controlled manner. The daily-surge mouse model has been used to examine changes in GnRH neuron activity, intrinsic properties, neurotransmission, and neuromodulation [5, 15, 16, 20, 41, 73]. No differences in parameters have been observed among Days 2 through 4 after surgery, when the previous and present experiments were performed. The possibility remains, however, that surges on successive days may utilize slightly different mechanisms and that the mechanisms of surge induction in this model may differ from those used in the endogenous intact cycle.
The present studies indicate that estradiol alters glutamate transmission to GnRH neurons in a manner that correlates with the decreases in GnRH neuron firing activity and LH release observed during negative feedback. Glutamate, in addition to GABA and neuromodulatory factors, is thus poised to have an important role in the neural control of ovulation.
We thank Debra Fisher for expert technical assistance.
1Supported by National Institute of Child Health and Human Development/National Institutes of Health R01 HD41469 and National Institute of Neurological Disorders and Stroke National Research Service Award F31 NS53253 to C.A.C.