Skip to main content
PMC home page
Endocrinology logoLink to Endocrinology
. 2008 Jul 10;149(11):5500–5508. doi: 10.1210/en.2008-0453

Critical Roles for Fast Synaptic Transmission in Mediating Estradiol Negative and Positive Feedback in the Neural Control of Ovulation

Catherine A Christian 1, Suzanne M Moenter 1
PMCID: PMC2584596  PMID: 18617615

Abstract

A switch in the balance of estradiol feedback actions from negative to positive initiates the GnRH surge, triggering the LH surge that causes ovulation. Using an ovariectomized, estradiol-treated (OVX+E) mouse model that exhibits daily switches between negative in the morning and positive feedback in the evening, we investigated the roles of fast synaptic transmission in regulating GnRH neuron firing during negative and positive feedback. Targeted extracellular recordings were used to monitor activity of GnRH neurons from OVX+E and OVX mice in control solution or solution with antagonists to both ionotropic glutamate and γ-aminobutyric acid receptors (blockade). Blockade had no effect on activity of OVX cells. In contrast, in OVX+E cells in the morning, blockade increased activity compared with control cells, whereas in the evening, blockade decreased activity. In vivo barbiturate sedation of OVX+E mice that blocked LH surge induction prevented the in vitro evening changes in firing rate and response to blockade. These observations suggest at least partial inversion of the negative-to-positive switch in estradiol feedback action and indicate that changes in fast synaptic transmission to GnRH neurons and within the network of cells presynaptic to GnRH neurons are critical for mediating estradiol negative and positive feedback actions on GnRH neurons. Fast synaptic transmission may also affect GnRH neuron activity indirectly through altering release of excitatory and inhibitory neuromodulators onto GnRH neurons at specific times of day. Fast synaptic transmission is thus critical for proper generation and timing of the GnRH surge.

THE STEROID HORMONE estradiol has profound effects throughout the central nervous system (1). A principal role for estradiol is in the regulation of the neural signal for ovulation. At the end of the follicular phase of the reproductive cycle (proestrus in rodents), the balance of feedback actions of estradiol on the GnRH neural network switches from negative to positive, leading to an increase in GnRH neuron firing activity (2). This causes a surge of GnRH release (3,4,5,6) that signals a surge of LH from the pituitary, triggering ovulation. The mechanisms of the GnRH surge and the negative-to-positive switch in estradiol action are not well understood. In rodents the surge appears dependent on not only estradiol feedback but also a daily signal that times the GnRH/LH surge to a particular time of day (i.e. late afternoon in nocturnal species) (2,7,8). There may be some diurnal dependency in humans as the LH surge in women tends to begin in the early morning (9).

To investigate the mechanisms of estradiol negative (−FB) and positive (+FB) feedback, we developed a mouse model that exhibits daily switches between −FB and +FB after ovariectomy and constant replacement of physiological levels of estradiol (2). As circulating levels of estradiol are constant, the primary variable is the time of day at which experiments are performed, allowing examination of diurnal changes. In this model the firing activity of GnRH neurons from estradiol-treated mice is low during −FB, but increased during +FB; cells from animals not treated with estradiol show no diurnal changes in firing (2). Estradiol may act directly and/or transsynaptically to mediate these effects (10,11,12,13,14,15). Recent work has indicated a role for γ-aminobutyric acid (GABA), which can excite GnRH neurons via the GABAA receptor (16,17,18), in mediating estradiol feedback. Specifically, estradiol decreases GABA transmission to GnRH neurons during −FB but increases it during +FB (19). Preliminary data suggest similar changes in glutamate transmission mediated by ionotropic receptors on GnRH neurons (20). Based on these findings, we hypothesized that fast synaptic transmission mediated by GABA and glutamate, two major inputs to GnRH neurons, plays a role in changing GnRH neuron activity during −FB and +FB.

Here we used single-unit recordings of GnRH neuron firing to investigate the roles of fast synaptic transmission in these different estradiol feedback states. Our results indicate that disturbing fast synaptic transmission disrupts both −FB and +FB effects of estradiol on GnRH neuron firing activity, suggesting fast synaptic inputs are critical for GnRH surge regulation.

Materials and Methods

Animals

GnRH neurons were recorded from adult (2–4 months old) GnRH-green fluorescent protein transgenic mice (21). Mice were maintained in 14-h light, 10-h dark cycle (lights off 1630 h EST) with Harlan 2916 food and water available ad libitum. As described previously (2), mice were ovariectomized (OVX) and either simultaneously implanted with a 1.8-cm-long SILASTIC brand capsule (1.98 mm internal × 3.18 mm external diameter, no. 508–009; Dow Corning, Midland, MI) containing 0.625 μg estradiol (E) suspended in sesame oil (OVX+E, n = 45 mice) or not treated further (OVX, n = 21 mice). OVX+E treatment in this manner induces daily negative-to-positive switches in estradiol feedback, generating LH surges that peak around the time of lights off (2). Estradiol was solely administered in vivo and was not present in any recording solutions. A subset of OVX+E animals received a 60–65 mg/kg ip injection of Nembutal (Abbott, Abbott Park, IL) on the day of recording to temporarily block neural activity and prevent LH surge induction (22,23). Animals were typically sedated for 1–2 h. For LH measurements, trunk blood was collected after CO2 anesthesia. Blood was refrigerated overnight at 4 C and centrifuged for at least 30 min at room temperature (∼21–23 C). Serum was stored at −20 C until assay. Serum LH concentration was determined using a modified, supersensitive two-site mouse LH sandwich RIA (24). All procedures were approved by the University of Virginia Animal Care and Use Committee.

Brain slice preparation

Reagents were purchased from Sigma (St. Louis, MO). Sagittal brain slices (300 μm) were prepared 2–4 d after OVX surgery as previously described (2,25,26). Mice were euthanized at either 0900–0930 h (negative feedback, −FB) or 1430–1500 h (positive feedback, +FB). For recording, slices were placed in a chamber on the stage of a BX50WI upright fluorescent microscope (Opelco, Dulles, VA) and superfused at 5–6 ml/min with oxygenated normal saline (NS) maintained at 30–32 C.

Recordings

Targeted single-unit extracellular recordings (loose-patch) were used because this configuration allows recording with minimal impact on the behavior of the recorded cell (26). Recordings were performed between 1030 and 1400 h (−FB) or 1600 and 1930 h (+FB). Up to three cells per animal, two cells per treatment, and only one cell per slice were recorded. Recording pipettes (1–3 mΩ) were filled with normal HEPES-buffered solution (26). Seal resistances (5–50 mΩ) were monitored at least every 30 min during recording. Recordings were made in voltage-clamp mode with the pipette holding potential at 0 mV, and signals were filtered at 10 kHz. Cells were recorded for 30–60 min. If no firing was observed after 60 min, 15 mm KCl was added to the bath solution to depolarize cells and induce firing for confirmation of recording integrity and cell health. If no firing occurred with KCl treatment, the data were discarded.

For on-cell measurements of GnRH neuron responsiveness to local rapid GABA application, recording pipettes were filled with a solution containing (in millimoles) 125 K-gluconate, 20 KCl, 10 HEPES, 5 EGTA, 4 MgATP, 0.4 NaGTP, and 0.1 CaCl2, and a gigaohm seal was formed between the pipette and the GnRH neuron. A pressure pipette containing 1 mm GABA in HEPES-buffered solution was placed adjacent to the cell body, and GABA was applied via a 5-msec pulse of 5–10 lb/in.2 as described previously (16). Action current spikes were detected with the pipette holding potential at 0 mV and signals filtered at 10 kHz. Seal resistance was monitored every 60 sec and remained greater than 1 GΩ.

Drug treatments

Bath solution either contained NS alone or NS with the addition of D(-)2-amino-5-phosphonovaleric acid (APV; 20 μm) and 6-cyano-7-nitroquinoxaline (CNQX; 10–20 μm) to block ionotropic glutamate receptors and picrotoxinin (100 μm) to block GABAA receptors (blockade). Blockade-treated cells were preincubated in blockade solution for 15 min or longer before recording. Order of recording was randomized between control and blockade-treated cells so that times of recording were equivalent between treatment groups for each time of day. Recordings of GnRH neuron responsiveness to GABA application were performed in either NS alone or the presence of APV and CNQX.

Data analysis

For extracellular recordings, each time the recording trace crossed the threshold for event detection, 10 msec worth of data centered on the threshold crossing were recorded and stored to a digital file. Events were confirmed off line using custom programs in Igor Pro (26) and binned at 1-min intervals. Binned data were transferred to Excel (Microsoft, Redmond, WA) and InStat or Prism4 (GraphPad, San Diego, CA) for further analysis. Data were analyzed for mean firing rate, percentage of time spent in quiescence, and duration of quiescence. Mean firing rate (Hertz) was calculated by dividing the total number of events over the duration of recording. Quiescence was defined as 1-min bins containing one or fewer event. The percentage of bins that were quiescent and the longest duration of consecutive quiescent bins were calculated for each cell. Percentage of quiescence provides a measure of overall activity, whereas changes in the duration of quiescent time indicate changes in the pattern of firing. Firing data were log transformed and group means compared using two-way ANOVA followed by Bonferroni multiple comparisons tests. Comparison of firing in medial vs. lateral slices was done using two-tailed t tests. LH levels were compared using nonparametric Kruskal-Wallis followed by Dunn’s multiple comparisons tests. Data are presented as mean ± sem. Statistical significance was set at P < 0.05.

Results

Effects of anatomical location on GnRH neuron firing during estradiol +FB

GnRH neurons receive synaptic inputs that arise from several anatomically diverse areas (13,27,28). Furthermore, the scattered nature of GnRH neurons suggests that there may be subpopulations that function in specific ways (19,29,30). Previously, we examined changes in firing activity of GnRH neurons in coronal slices (2) and did not observe any differences in firing based on anatomical location. In more recent work, however, we showed that during +FB GABA transmission to GnRH neurons in sagittal slices is elevated in medial slices but lower in lateral slices (19). Here we investigated the effect of anatomical location on firing of GnRH neurons from OVX+E mice in sagittal slices during +FB. Preparation of 300-μm sagittal slices typically yields five slices containing GnRH neurons; the middle slice (midsagittal) and the first slice on either side (parasagittal) were designated as medial, and the other two slices as lateral. Cells recorded in medial slices (n = 14 cells from 13 mice) showed significantly higher firing activity than cells in lateral slices (n = 5 cells from five mice) as measured by increased mean firing rate and decreased percentage of time in quiescence and duration of quiescent time (P < 0.05 all parameters) (Fig. 1). This effect correlates well with the patterns of both cFos expression (31) and GABA transmission (19) during the surge. Furthermore, these data in sagittal slices support and extend our previous findings in coronal slices (2), indicating components of the surge mechanism persist in the brain slice preparation. As the cells that underlie surge generation appear to be primarily located in medial sagittal slices, medial slices were used for the remainder of the experiments.

Figure 1.

Figure 1

Open in a new tab

Effect of anatomical location on GnRH neuron firing activity during estradiol-positive feedback in sagittal slices. A and B, Representative recordings from GnRH neurons from OVX+E mice located in a medial (in this case parasagittal) slice (A) and a lateral slice (B). Vertical lines at the top of each graph illustrate time of individual action currents recorded. Firing rate is displayed at 30-sec intervals. C–E, Mean values for mean firing rate (C), percentage of time spent in quiescence (D), and longest duration of quiescence (E). *, P < 0.05. med, medial; lat, lateral.

Fast synaptic transmission blockade increases GnRH neuron firing at the normal time of estradiol −FB

To investigate the roles of fast synaptic transmission in regulating GnRH neuron firing activity, cells were recorded in either control normal saline (NS) or in NS with ionotropic glutamate and GABA receptor antagonists added (blockade) to block fast synaptic transmission throughout the brain slice network. Under normal conditions, excitatory and inhibitory transmission are maintained in a balanced state in the network. Blocking one but not the other can lead to network imbalances that are evident in experiments, such as these, in which the spontaneous firing rate of a neuron is monitored and no voltage control is imposed on the recorded cell by the amplifier. In particular, blocking GABAA receptors alone can disinhibit excitatory transmission, mainly mediated by ionotropic glutamate receptors. This is used as an in vitro model for epilepsy (32). In GnRH neurons, when ionotropic glutamate receptors are already blocked subsequent blockade of GABAA receptors decreases firing rate (17), whereas blockade of GABAA receptors alone increases firing, likely through disinhibition mediated via increased glutamatergic transmission (17,33). To avoid this problem here, both GABA and glutamate receptors were simultaneously blocked to preclude changes in firing activity attributable to network imbalances of excitatory and inhibitory neurotransmission. Because ionotropic receptors are blocked throughout the slice preparation, this technique cannot precisely specify locations of altered fast synaptic transmission within the network. It can, however, allow for assessment of roles of fast synaptic transmission in the network as a whole and, in combination with complementary studies examining changes in fast synaptic transmission directly to GnRH neurons (19,20), may provide at least preliminary information on direct vs. indirect actions of altered fast synaptic transmission.

During the time of −FB (1030–1400 h) in estradiol-treated animals in vivo, control cells in brain slices from OVX+E mice show reduced activity compared with cells from OVX mice (OVX+E, 0.04 ± 0.02 Hz, n = 9 cells from eight mice; OVX, 0.18 ± 0.06 Hz, n = 11 cells from eight mice; P < 0.01), confirming previous results (2). In OVX+E cells during −FB, blockade markedly increased firing activity relative to that in control solution (n = 8 cells from seven mice). Representative examples are shown in Fig. 2, A and B. Firing parameters from these recordings were quantified revealing that blockade increased mean firing rate and decreased percentage and duration of quiescent time (P < 0.01) (Fig. 2, C–E). Interestingly during −FB, the level of firing induced by blockade treatment of OVX+E cells surpassed even that in control OVX cells (see Fig. 6) (mean firing rate, percentage, and duration of quiescent time, P < 0.05).

Figure 2.

Figure 2

Open in a new tab

Blockade of ionotropic GABA and glutamate receptors increases firing activity in medial GnRH neurons from OVX+E mice recorded during negative feedback (morning). A and B, Representative recordings from GnRH neurons recorded in either control solution (A; con) or solution with receptor blockers added (B; blk). See Fig. 1 for details. C–E, Mean values for mean firing rate (C), percentage of time spent in quiescence (D), and longest duration of quiescence (E). *, P < 0.05; con, control; blk, blockade.

Figure 6.

Figure 6

Open in a new tab

Firing rate or pattern of GnRH neurons from OVX mice does not change in response to blockade of fast synaptic transmission at either time of day. A–D, Representative recordings in the morning (AM; corresponding to time of −FB in OVX+E animals) (A and B) or evening (corresponding to time of +FB in OVX+E animals) (C and D) from GnRH neurons recorded either in control solution (A and C) or in solution with fast synaptic blockers added (B and D). Note that no switches between −FB and +FB occur in OVX groups, but recording times were matched to OVX+E groups in which these states do occur; these terms are used for comparison. E and G, Mean values for mean firing rate (E), percentage of time spent in quiescence (F), and longest duration of quiescent time (G). Open bars indicate cells recorded in control solution, and filled bars indicate cells recorded in the presence of fast synaptic blockers (blockade).

Blockade decreases GnRH neuron firing during the normal time of estradiol +FB

As previously reported (2), during +FB control cells from OVX+E mice show increased activity, compared with control OVX cells (OVX+E, 0.85 ± 0.2 Hz, n = 14 cells from 13 mice; OVX, 0.3 ± 0.06 Hz, n = 13 cells from 10 mice; P < 0.05). In contrast, blockade-treated cells (n = 14 cells from 13 mice) from OVX+E mice showed decreased mean firing rate (P < 0.05) and increased percentage and duration of quiescent time (P < 0.05), compared with control cells from OVX+E mice (Fig. 3).

Figure 3.

Figure 3

Open in a new tab

Blockade of ionotropic GABA and glutamate receptors reduces activity in GnRH neurons from OVX+E mice recorded during positive feedback (evening). A and B, Representative recordings from GnRH neurons recorded either in control solution (A; con) or solution with fast synaptic blockers added (B; blk). See Fig. 1 for details. C–E, Mean values for mean firing rate (C), percentage of time spent in quiescence (D), and longest duration of quiescent time (E). *, P < 0.05.

Blockade reverses diurnal changes in GnRH neuron firing activity

Control OVX+E cells exhibited large diurnal changes in firing, with increased firing activity during +FB compared with −FB, as previously shown in coronal slices (2). Interestingly, blockade reversed some diurnal changes in firing observed in control conditions. These changes are easier to see by plotting −FB and +FB values for control and blockade cells side by side, as shown in Fig. 4. Specifically, during −FB blockade cells showed increased activity quantified as decreased percentage (P < 0.05) and duration of quiescent time (P < 0.01), compared with +FB. Although there is a trend toward decreased mean firing rate during +FB, compared with −FB this effect was not statistically significant (P > 0.05), suggesting other factors in addition to GABA/glutamate may be involved in driving increased GnRH neuron firing activity during +FB or suppressing activity during −FB.

Figure 4.

Figure 4

Open in a new tab

Reversal of diurnal changes in GnRH neuron activity with blockade of fast synaptic transmission. A–C, Mean values for mean firing rate (A), percentage of time spent in quiescence (B), and longest duration of quiescence (C). Open bars indicate cells recorded in control solution, and filled bars indicate cells recorded in the presence of fast synaptic blockers (blockade). Note that the data are replotted from Figs. 2 and 3 for ease of comparison. *, P < 0.05 vs. −FB in respective recording bath solution.

In vivo barbiturate sedation prevents both increased firing during +FB and the change in response to blockade at this time

If fast synaptic transmission is critical to the negative-to-positive switch in estradiol action, as indicated by the partial reversal of diurnal changes in blockade-treated cells (Fig. 4), a sedation treatment to block ovulation presumably at the central level (22,23) would likely prevent the diurnal changes in GnRH neuron firing and alter the response to in vitro blockade treatment. To test this possibility, we first established the time of sedation that effectively blocked the LH surge. Sedation for 1–2 h, beginning at 1100 h (5.5 h before lights out) on the day of recording was shown to block LH surge induction as measured in serum LH levels at 1600 h (sedated, n = 7 mice; control, n = 6 mice; P < 0.05 vs. control) (Fig. 5A); sedation at later times of day did not block the LH surge (1200 h, 1300 h, n = 3 mice each) (Fig. 5A), recapitulating studies in the rat, indicating that there is a critical window during which the LH surge and ovulation can be blocked (22,23) but after which the surge and ovulation become inevitable.

Figure 5.

Figure 5

Open in a new tab

In vivo barbiturate sedation that blocks LH surge induction in OVX+E mice blunts the diurnal increase in GnRH neuron firing activity and the diurnal switch in response to blockade treatment. A, Serum LH levels measured at 1600 h in animals that were injected with Nembutal at 1100 h, 1200 h, 1300 h, or not sedated (control). B, Representative recordings of cells from animals sedated at 1100 h recorded in control solution or in solution with fast synaptic blockers added (blockade). C–E, Mean values for mean firing rate (C), percentage of time spent in quiescence (D), and longest duration of quiescent time (E). Open bars indicate cells recorded in control solution, and filled bars indicate cells recorded in the presence of fast synaptic blockers (blockade). *, P < 0.05. con, Control; sed, sedated.

GnRH neurons from 1100 h-sedated animals recorded in control solution during the normal time of +FB (n = 8 cells from eight mice) showed decreased mean firing rate, compared with cells from unsedated control animals (P < 0.01) (Fig. 5, B and C). In contrast, blockade-treated cells from sedated animals (n = 7 cells from six mice) showed increased activity, compared with cells from sedated animals recorded in control solution (Fig. 5, B–E), reflected as increased mean firing rate (P < 0.001) and decreased quiescence (P < 0.05). Note the direction of response to blockade treatment in cells from sedated mice recorded during the normal time of +FB is similar to that observed during the time of −FB in cells from unsedated mice. Thus, a treatment that inhibits the diurnal switch in estradiol feedback action on LH also interferes with the diurnal changes in GnRH neuron firing and responsiveness to fast synaptic transmission blockade that are characteristic of +FB.

Estradiol is required for a response to fast synaptic transmission blockade

To examine whether estradiol is required for a response to fast synaptic transmission blockade, firing activity of GnRH neurons from OVX mice not treated with estradiol was recorded in control and blockade solutions. Blockade did not alter any measure of firing activity in OVX cells either during −FB (n = 11 cells from eight mice control; n = 6 cells from six mice blockade) or +FB (n = 13 cells from 10 mice control; n = 8 cells from eight mice blockade) (Fig. 6). Thus, estradiol is required for a response of GnRH neuron firing activity to blockade of fast synaptic transmission, just as it is required for the normal diurnal change in firing (2) and changes in fast synaptic transmission (19) to GnRH neurons in this model.

GnRH neurons can be excited by GABAA receptor activation during both −FB and +FB

It is possible that the diurnal shift in response to fast synaptic transmission blockade reflects a diurnally regulated change in the direction of response to GABAA receptor activation in GnRH neurons. Such shifts have been suggested to underlie circadian changes in firing in the suprachiasmatic nucleus (SCN) (34,35,36). Previous work indicated that GABAA receptor activation in GnRH neurons is excitatory throughout the circadian cycle (16) and on all days of the estrous cycle in the rat (18). These studies, however, were performed before the daily LH surge model used here was developed. To examine whether the response to GABAA receptor activation may vary with time of day in this model, we assessed the response of GnRH neurons from OVX+E mice to local, rapid GABA application during −FB (n = 7 cells from four mice) and +FB (n = 6 cells from two mice). The on-cell recording configuration was used to preserve the internal chloride milieu of the recorded GnRH neuron, and thus, the response to GABA. Rapid GABA application elicited action currents in GnRH neurons during both −FB and +FB (Fig. 7, A and B), suggesting that chloride homeostasis in GnRH neurons does not change sufficiently in a diurnal manner to alter the direction of response to GABAA receptor activation. Although GABA action is excitatory during both −FB and +FB, the number of action currents elicited by the local application of GABA tended to be less during −FB, perhaps indicating the magnitude of excitation may be reduced in −FB. Importantly, GABA application elicited firing in GnRH neurons in both the presence and absence of APV and CNQX (Fig. 7, B and C), indicating that excitation of GnRH neurons by GABA is not dependent on the absence of endogenous glutamatergic input.

Figure 7.

Figure 7

Open in a new tab

Activation of GABAA receptors excites GnRH neurons during both −FB and +FB, independent of ionotropic glutamate input. A and B, Representative recordings of the response of GnRH neurons from OVX+E mice during −FB (A) and +FB (B) to local rapid pressure application (arrow) of GABA (1 mm) recorded in the on-cell recording configuration in NS alone. C, Recording from the same cell shown in (B), made in the presence of APV and CNQX.

Discussion

The mechanisms of estradiol negative and positive actions on the GnRH network, and switches between the two, remain fundamental questions. Here we present evidence that fast synaptic transmission plays a significant role in mediating both negative and positive feedback effects of estradiol. Blockade of fast synaptic transmission increased GnRH neuron firing during the normal time of −FB, but decreased it during +FB, suggesting fast transmission mediated by GABA and glutamate is required for both generating and timing the different estradiol feedback actions. A caveat to working in a brain slice preparation is that normal network circuitry is inevitably disrupted. The maintenance of diurnal and estradiol-dependent changes in GnRH neuron firing and synaptic transmission that directly correlate with changing in vivo LH levels, however, suggests that at least parts of the network sufficient to regulate the normal changes in GnRH neuron activity characteristic of −FB and +FB are functionally maintained in the slice preparation. This is consistent with findings in other brain slice preparations in which circuitry has been examined (37,38).

Estradiol may influence synaptic transmission to GnRH neurons by acting on cells that are directly presynaptic to GnRH neurons (first order) and/or on cells that communicate indirectly to GnRH neurons through other cells (second order and above). Previous whole-animal studies measuring neurotransmitter levels in the preoptic area showed that GABA levels decline before and during the LH surge (39,40,41,42), and glutamate release in the preoptic area increases during the LH surge (43). Alterations in neurotransmitter concentrations, however, do not provide information regarding which neurons are affected by these changes. It is possible lower GABA levels are required to diminish inhibition on upstream inputs (which are presumably inhibited by GABAA receptor activation, in contrast to GnRH neurons), which thereby induce increases in GnRH neuron activity.

During −FB, fast synaptic transmission to GnRH neurons via first-order GABA/glutamate neurons is of low frequency and amplitude (19,20). This relative lack of excitatory transmission likely is one contributor to the low level of GnRH neuron activity during −FB. Both GABA and glutamate depolarize GnRH neurons during both negative and positive feedback. This makes the increase in activity when their receptors are blocked during −FB appear paradoxical. One possibility is that the low amplitude of GABAergic postsynaptic currents in GnRH neurons during −FB (19) could have an inhibitory effect. Specifically, subthreshold depolarizing GABA postsynaptic potentials may act in a shunting manner to decrease GnRH neuron excitability by increasing membrane conductance, thereby reducing the response to excitatory transmission or intrinsic conductances (44,45). Although such depolarizing inhibition may occur during −FB, the low frequency of transmission at this time (19) suggests this effect would be minor and likely does not explain the large increase in firing after simultaneous blockade of ionotropic GABA and glutamate receptors at this time. An alternative, speculative explanation is that during −FB estradiol acts on second-order cells to alter fast synaptic transmission to first-order presynaptic cells, which in turn lower GnRH neuron firing through mechanisms other than GABA/glutamate transmission, such as neuromodulators. Disruption of fast synaptic transmission by blockade would alter the activity of these neuromodulatory first-order presynaptic cells to decrease release of inhibitory and/or increase release of excitatory factors and thereby increase GnRH neuron activity.

In contrast to the low rate of GABA/glutamate transmission to GnRH neurons during −FB, during +FB, transmission directly to GnRH neurons is elevated (19,20). Under these conditions, the decreased firing caused by blockade suggests the estradiol-induced increase in GABA/glutamate transmission from first-order neurons directly to GnRH neurons is critical to increase firing at this time. Blockade treatment would also block fast synaptic transmission between second-order cells and first-order cells and thus may also alter release of neuromodulators from first-order cells onto GnRH neurons.

The areas in which estradiol acts to induce changes in fast synaptic transmission in association with −FB and +FB have yet to be determined. The SCN, the central circadian pacemaker (46), is one candidate. The SCN projects to GnRH neurons (27,28,47) and SCN lesions abolish the LH surge (48). Estradiol feedback and circadian signals may also converge downstream of the SCN, for example in the anteroventral periventricular area (AVPV) because AVPV cells projecting to GnRH neurons express estrogen receptor-α (49). The AVPV receives direct input from the SCN (50,51), and AVPV lesions also block the LH surge (52). The SCN and AVPV coexhibit increased cFos expression with GnRH neurons during the GnRH/LH surge (53,54,55). Both of these areas are GABAergic, and cells in the AVPV may also release glutamate (56).

The role of the circadian system in the control of ovulation in rodents has long been recognized (23,57,58,59), but the nature of the daily signal is not precisely known. The present work, in which contrasting responses to the same treatment were elicited at different times of day, potentially via different mechanisms (direct vs. network), suggests a working model in which there is a series of daily signals that coordinate release of inhibitory vs. excitatory factors (and perhaps GnRH neuron response to these factors as well). The balance of these actions determines whether GnRH neuron activity is low or high at a particular time of day. If these coordinated changes do not occur at the proper time of day, the surge and subsequent ovulation are blocked until the critical window is present again (23). Consistent with this, in vivo sedation that blocks LH surge induction also inhibited the increase in GnRH neuron firing. Sedation also changed the response to fast synaptic transmission blockade, suggesting estradiol-induced changes in neurobiological mechanisms were altered by sedation. Notably, cells from animals in which the sedation treatment either did not produce an hour of sedation or in which complete sedation was not achieved until after 1200 h (n = 2 mice, not shown) did not show this effect. This suggests the different firing properties and response to blockade between cells from control and sedated animals are not due to nonspecific neural effects of the sedation treatment but are specific to the time and duration of the sedation. These studies thus link together diurnal changes in firing, alterations in response to blockade, and LH surge induction and further support the hypothesis that increased GnRH neuron activity characteristic of the surge is driven at least in part by changes in fast synaptic transmission. Importantly, if these changes do not occur at the correct time of day, GnRH neuron activity and response to blockade are held back from reaching the +FB state.

The importance of generating a GnRH surge to passing on genetic information is sufficient that multiple diurnal signals may be used to generate the surge. Blockade of fast synaptic transmission reversed diurnal changes in firing, profoundly disrupting the timing of estradiol-induced changes in GnRH neuron activity. Increased firing after blockade during −FB suggests GnRH neuron responsiveness to excitatory factors is in place before excitatory factors are released (15). This does not exclude the possibility that blockade increases firing activity indirectly by removing inhibitory neuromodulation [such as gonadotropin-inhibitory hormone input (60)] of GnRH neurons. Of interest, the inversion of diurnal changes in mean firing rate was only a strong trend. This suggests other factors [e.g. kisspeptin (55), vasoactive intestinal polypeptide (61,62,63,64), or vasopressin (65,66,67)] may help regulate the diurnal increase in GnRH neuron firing activity. This multiplicity underlies the complexity of the surge mechanism, as illustrated in the model proposed here.

Previous work in our laboratory on d 5–9 after surgery largely during −FB showed that blockade eliminated the effects of estradiol on GnRH neuron firing, so that firing of OVX+E cells was no longer different from OVX cells (26). Here, in contrast, blockade during −FB increased firing of OVX+E cells to a level significantly higher than that of OVX cells recorded at the same time of day. Several changes in methodology may account for these differences. We used a model in which daily LH surges are generated in OVX+E mice on d 2–4 after surgery (2). The LH surge appears to dampen after d 5; replacement of the estradiol capsule at this time restores the surge (Christian, C. A., and S. M. Moenter, unpublished observations). The lack of a positive feedback response after d 5 may reduce the response to blockade. Furthermore, the previous work examined cells in coronal slices, whereas sagittal slices were used here. Slice orientation affects available GABA transmission to GnRH neurons (19) and may influence other forms of neurotransmission as well. These differences notwithstanding, altogether these findings indicate critical roles for estradiol-sensitive afferents in modulating GnRH neuron activity through changes in fast synaptic transmission.

In summary, the present studies indicate critical roles for fast synaptic transmission in mediating both negative and positive actions of estradiol on GnRH neuron firing rate and pattern. These inputs also appear to regulate the diurnal switch between negative and positive feedback and are thus essential for both the generation and proper timing of the GnRH surge.

Acknowledgments

We thank Debra Fisher for expert technical assistance and Justyna Pielecka-Fortuna, Alison Roland, and Pei-San Tsai for helpful editorial comments.

Footnotes

This work was supported by National Institute of Child Health and Human Development/National Institutes of Health Grant R01 HD41469 and National Institute of Neurological Disorders and Stroke National Research Service Award F31 NS53253 (to C.A.C.).

Disclosure Statement: C.A.C. and S.M.M. have nothing to disclose.

First Published Online July 10, 2008

Abbreviations: APV, D(-)2-amino-5-phosphonovaleric acid; AVPV, anteroventral periventricular area; CNQX, 6-cyano-7-nitroquinoxaline; E, estradiol; +FB, estradiol positive feedback; −FB, estradiol negative feedback; GABA, γ-aminobutyric acid; NS, normal saline; OVX, ovariectomized; SCN, suprachiasmatic nucleus.

References

  1. McEwen BS, Alves SE 1999 Estrogen actions in the central nervous system. Endocr Rev 20:279–307 [DOI] [PubMed] [Google Scholar]
  2. Christian CA, Mobley JL, Moenter SM 2005 Diurnal and estradiol-dependent changes in gonadotropin-releasing hormone neuron firing activity. Proc Natl Acad Sci USA 102:15682–15687 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Sarkar DK, Chiappa SA, Fink G, Sherwood NM 1976 Gonadotropin-releasing hormone surge in pro-oestrous rats. Nature 264:461–463 [DOI] [PubMed] [Google Scholar]
  4. Moenter SM, Caraty A, Locatelli A, Karsch FJ 1991 Pattern of GnRH secretion leading up to ovulation in the ewe: existence of a preovulatory GnRH surge. Endocrinology 129:1175–1182 [DOI] [PubMed] [Google Scholar]
  5. Pau KY, Berria M, Hess DL, Spies HG 1993 Preovulatory gonadotropin-releasing hormone surge in ovarian-intact rhesus macaques. Endocrinology 133:1650–1656 [DOI] [PubMed] [Google Scholar]
  6. Clarke IJ, Thomas GB, Yao B, Cummins JT 1987 GnRH secretion throughout the ovine estrous cycle. Neuroendocrinology 46:82–88 [DOI] [PubMed] [Google Scholar]
  7. Norman RL, Blake CA, Sawyer CH 1973 Estrogen-dependent 24-hour periodicity in pituitary LH release in the female hamster. Endocrinology 93:965–970 [DOI] [PubMed] [Google Scholar]
  8. Legan SJ, Karsch FJ 1975 A daily signal for the LH surge in the rat. Endocrinology 96:57–62 [DOI] [PubMed] [Google Scholar]
  9. Kerdelhue B, Brown S, Lenoir V, Queenan JT, Jones GS, Scholler R, Jones HW 2002 Timing of initiation of the preovulatory luteinizing hormone surge and its relationship with the circadian cortisol rhythm in the human. Neuroendocrinology 75:158–163 [DOI] [PubMed] [Google Scholar]
  10. Skynner MJ, Sim JA, Herbison AE 1999 Detection of estrogen receptor α and β messenger ribonucleic acids in adult gonadotropin-releasing hormone neurons. Endocrinology 140:5195–5201 [DOI] [PubMed] [Google Scholar]
  11. Hrabovszky E, Steinhauser A, Barabas K, Shughrue PJ, Petersen SL, Merchenthaler I, Liposits Z 2001 Estrogen receptor-beta immunoreactivity in luteinizing hormone-releasing hormone neurons of the rat brain. Endocrinology 142:3261–3264 [DOI] [PubMed] [Google Scholar]
  12. Couse JF, Korach KS 1999 Estrogen receptor null mice: what have we learned and where will they lead us? Endocr Rev 20:358–417 [DOI] [PubMed] [Google Scholar]
  13. Wintermantel TM, Campbell RE, Porteous R, Bock D, Grone H-J, Todman MG, Korach KS, Greiner E, Perez CA, Schutz G, Herbison AE 2006 Definition of estrogen receptor pathway critical for estrogen positive feedback to gonadotropin-releasing hormone neurons and fertility. Neuron 52:271–280 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. DeFazio RA, Moenter SM 2002 Estradiol feedback alters potassium currents and firing properties of gonadotropin-releasing hormone neurons. Mol Endocrinol 16:2255–2265 [DOI] [PubMed] [Google Scholar]
  15. Chu Z, Moenter SM 2006 Physiologic regulation of a tetrodotoxin-sensitive sodium influx that mediates a slow afterdepolarization potential in gonadotropin-releasing hormone neurons: possible implications for the central regulation of fertility. J Neurosci 26:11961–11973 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. DeFazio RA, Heger S, Ojeda SR, Moenter SM 2002 Activation of A-type γ-aminobutyric acid receptors excites gonadotropin-releasing hormone neurons. Mol Endocrinol 16:2872–2891 [DOI] [PubMed] [Google Scholar]
  17. Moenter SM, DeFazio RA 2005 Endogenous γ-aminobutyric acid can excite gonadotropin-releasing hormone neurons. Endocrinology 146:5374–5379 [DOI] [PubMed] [Google Scholar]
  18. Yin C, Tanaka N, Kato M, Sakuma Y 2006 GABA depolarizes GnRH neurons isolated from adult GnRH-EGFP transgenic rats. Neurosci Res 55S:S159 [Google Scholar]
  19. Christian CA, Moenter SM 2007 Estradiol induces diurnal shifts in GABA transmission to gonadotropin-releasing hormone neurons to provide a neural signal for ovulation. J Neurosci 27:1913–1921 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Christian CA, Moenter SM 2007 Diurnal changes in NMDA and AMPA/KA receptor-mediated glutamate transmission to gonadotropin-releasing hormone (GnRH) neurons are associated with the GnRH/LH surge. Proc 37th Annual Meeting of the Society for Neuroscience, San Diego, CA, 2007 (Abstract 518.4) [Google Scholar]
  21. Suter KJ, Song WJ, Sampson TL, Wuarin JP, Saunders JT, Dudek FE, Moenter SM 2000 Genetic targeting of green fluorescent protein to gonadotropin-releasing hormone neurons: characterization of whole-cell electrophysiological properties and morphology. Endocrinology 141:412–419 [DOI] [PubMed] [Google Scholar]
  22. Everett JW, Sawyer CH, Markee JE 1949 A neurogenic timing factor in control of the ovulatory discharge of luteinizing hormone in the cyclic rat. Endocrinology 44:234–250 [DOI] [PubMed] [Google Scholar]
  23. Everett JW, Sawyer CH 1950 A 24-hour periodicity in the “LH-release apparatus” of female rats, disclosed by barbiturate sedation. Endocrinology 47:198–218 [DOI] [PubMed] [Google Scholar]
  24. Haavisto AM, Pettersson K, Bergendahl M, Perheentupa A, Roser JF, Huhtaniemi I 1993 A supersensitive immunofluorometric assay for rat luteinizing hormone. Endocrinology 132:1687–1691 [DOI] [PubMed] [Google Scholar]
  25. Chu Z, Moenter SM 2005 Endogenous activation of metabotropic glutamate receptors modulates GABAergic transmission to gonadotropin-releasing hormone neurons and alters their firing rate: a possible local feedback circuit. J Neurosci 25:5740–5749 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Nunemaker CS, DeFazio RA, Moenter SM 2002 Estradiol-sensitive afferents modulate long-term episodic firing patterns of GnRH neurons. Endocrinology 143:2284–2292 [DOI] [PubMed] [Google Scholar]
  27. Boehm U, Zou Z, Buck LB 2005 Feedback loops link odor and pheromone signaling with reproduction. Cell 123:683–695 [DOI] [PubMed] [Google Scholar]
  28. Yoon H, Enquist LW, Dulac C 2005 Olfactory inputs to hypothalamic neurons controlling reproduction and fertility. Cell 123:669–682 [DOI] [PubMed] [Google Scholar]
  29. Rubin BS, Mitchell S, Lee CE, King JC 1995 Reconstructions of populations of luteinizing hormone releasing hormone neurons in young and middle-aged rats reveal progressive increases in subgroups expressing Fos protein on proestrus and age-related deficits. Endocrinology 136:3823–3830 [DOI] [PubMed] [Google Scholar]
  30. Sim JA, Skynner MJ, Herbison AE 2001 Heterogeneity in the basic membrane properties of postnatal gonadotropin-releasing hormone neurons in the mouse. J Neurosci 21:1067–1075 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Lee WS, Smith MS, Hoffman GE 1990 Luteinizing hormone-releasing hormone neurons express Fos protein during the proestrous surge of luteinizing hormone. Proc Natl Acad Sci USA 87:5163–5167 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Salazar P, Tapia R, Rogawski MA 2003 Effects of neurosteroids on epileptiform activity induced by picrotoxin and 4-aminopyridine in the rat hippocampal slice. Epilepsy Res 55:71–82 [DOI] [PubMed] [Google Scholar]
  33. Han SK, Todman MG, Herbison AE 2004 Endogenous GABA release inhibits the firing of adult gonadotropin-releasing hormone neurons. Endocrinology 145:495–499 [DOI] [PubMed] [Google Scholar]
  34. Choi HJ, Lee CJ, Schroeder A, Kim YS, Jung SH, Kim JS, Kim DY, Son EJ, Han HC, Hong SK, Colwell CS, Kim YI 2008 Excitatory actions of GABA in the suprachiasmatic nucleus. J Neurosci 28:5450–5459 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Wagner S, Castel M, Gainer H, Yarom Y 1997 GABA in the mammalian suprachiasmatic nucleus and its role in diurnal rhythmicity. Nature 387:598–603 [DOI] [PubMed] [Google Scholar]
  36. De Jeu M, Pennartz C 2002 Circadian modulation of GABA function in the rat suprachiasmatic nucleus: excitatory effects during the night phase. J Neurophysiol 87:834–844 [DOI] [PubMed] [Google Scholar]
  37. Smith BN, Dudek FE 1996 Amino acid-mediated regulation of spontaneous synaptic activity patterns in the rat basolateral amygdala. J Neurophysiol 76:1958–1967 [DOI] [PubMed] [Google Scholar]
  38. Rempe DA, Bertram EH, Williamson JM, Lothman EW 1997 Interneurons in area CA1 stratum radiatum and stratum oriens remain functionally connected to excitatory synaptic input in chronically epileptic animals. J Neurophysiol 78:1504–1515 [DOI] [PubMed] [Google Scholar]
  39. Jarry H, Perschl A, Wuttke W 1988 Further evidence that preoptic anterior hypothalamic GABAergic neurons are part of the GnRH pulse and surge generator. Acta Endocrinol 118:573–579 [DOI] [PubMed] [Google Scholar]
  40. Robinson J, Kendrick K, Lambart C 1991 Changes in the release of γ-aminobutyric acid and catecholamines in the preoptic/septal area prior to and during the preovulatory surge of luteinizing hormone in the ewe. J Neuroendocrinol 3:393–399 [DOI] [PubMed] [Google Scholar]
  41. Mitsushima D, Shwe TT, Funabashi T, Shinohara K, Kimura F 2002 GABA release in the medial preoptic area of cyclic female rats. Neuroscience 113:109–114 [DOI] [PubMed] [Google Scholar]
  42. Tin-Tin-Win-Shwe, Mitsushima D, Shinohara K, Kimura F 2004 Sexual dimorphism of GABA release in the medial preoptic area and luteinizing hormone release in gonadectomized estrogen-primed rats. Neuroscience 127:243–250 [DOI] [PubMed] [Google Scholar]
  43. Jarry H, Leonhardt S, Schwarze T, Wuttke W 1995 Preoptic rather than mediobasal hypothalamic amino acid neurotransmitter release regulates GnRH secretion during the estrogen-induced LH surge in the ovariectomized rat. Neuroendocrinology 62:479–486 [DOI] [PubMed] [Google Scholar]
  44. Koch C, Poggio T, Torre V 1983 Nonlinear interactions in a dendritic tree: localization, timing, and role in information processing. Proc Natl Acad Sci USA 80:2799–2802 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Staley KJ, Mody I 1992 Shunting of excitatory input to dentate gyrus granule cells by a depolarizing GABAA receptor-mediated postsynaptic conductance. J Neurophysiol 68:197–212 [DOI] [PubMed] [Google Scholar]
  46. Ralph MR, Foster RG, Davis FC, Menaker M 1990 Transplanted suprachiasmatic nucleus determines circadian period. Science 247:975–978 [DOI] [PubMed] [Google Scholar]
  47. van der Beek EM, Horvath TL, Wiegant VM, Van den Hurk R, Buijs RM 1997 Evidence for a direct neuronal pathway from the suprachiasmatic nucleus to the gonadotropin-releasing hormone system: combined tracing and light and electron microscopic immunocytochemical studies. J Comp Neurol 384:569–579 [DOI] [PubMed] [Google Scholar]
  48. Brown-Grant K, Raisman G 1977 Abnormalities in reproductive function associated with the destruction of the suprachiasmatic nuclei in female rats. Proc R Soc Lond B Biol Sci 198:279–296 [DOI] [PubMed] [Google Scholar]
  49. Simerly RB, Carr AM, Zee MC, Lorang D 1996 Ovarian steroid regulation of estrogen and progesterone receptor messenger ribonucleic acid in the anteroventral periventricular nucleus of the rat. J Neuroendocrinol 8:45–56 [DOI] [PubMed] [Google Scholar]
  50. de la Iglesia HO, Blaustein JD, Bittman EL 1995 The suprachiasmatic area in the female hamster projects to neurons containing estrogen receptors and GnRH. Neuroreport 6:1715–1722 [DOI] [PubMed] [Google Scholar]
  51. Watson RE, Langub MC, Engle MG, Maley BE 1995 Estrogen-receptive neurons in the anteroventral periventricular nucleus are synaptic targets of the suprachiasmatic nucleus to the gonadotropin-releasing hormone system: combined tracing and light and electron microscopic immunocytochemical studies. Brain Res 689:254–264 [DOI] [PubMed] [Google Scholar]
  52. Wiegand SJ, Terasawa E 1982 Discrete lesions reveal functional heterogeneity of suprachiasmatic structures in regulation of gonadotropin secretion in the female rat. Neuroendocrinology 34:395–404 [DOI] [PubMed] [Google Scholar]
  53. de la Iglesia HO, Meyer J, Schwartz WJ 2003 Lateralization of circadian pacemaker output: activation of left- and right-sided luteinizing hormone-releasing hormone neurons involves a neural rather than a humoral pathway. J Neurosci 23:7412–7414 [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Le WW, Berghorn KA, Rassnick S, Hoffman GE 1999 Periventricular preoptic area neurons coactivated with luteinizing hormone (LH)-releasing hormone (LHRH) neurons at the time of the LH surge are LHRH afferents. Endocrinology 140:510–519 [DOI] [PubMed] [Google Scholar]
  55. Smith JT, Popa SM, Clifton DK, Hoffman GE, Steiner RA 2006 Kiss1 neurons in the forebrain as central processors for generating the preovulatory luteinizing hormone surge. J Neurosci 26:6687–6694 [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. 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]
  57. Docke F, Dorner G 1965 The mechanism of the induction of ovulation by oestrogens. J Endocrinol 33:491–499 [DOI] [PubMed] [Google Scholar]
  58. Chappell PE 2005 Clocks and the black box: circadian influences on gonadotropin-releasing hormone secretion. J Neuroendocrinol 17:119–130 [DOI] [PubMed] [Google Scholar]
  59. de la Iglesia HO, Schwartz WJ 2006 Minireview: timely ovulation: circadian regulation of the female hypothalamo-pituitary-gonadal axis. Endocrinology 147:1148–1153 [DOI] [PubMed] [Google Scholar]
  60. Kriegsfeld LJ, Mei DF, Bentley GE, Ubuka T, Mason AO, Inoue K, Ukena K, Tsutsui K, Silver R 2006 Identification and characterization of a gonadotropin-inhibitory system in the brains of mammals. Proc Natl Acad Sci USA 103:2410–2415 [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Christian CA, Moenter SM 2008 Vasoactive intestinal polypeptide can excite gonadotropin-releasing hormone neurons in a manner dependent on estradiol and gated by time of day. Endocrinology 149:3130–3136 [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Gerhold LM, Rosewell KL, Wise PM 2005 Suppression of vasoactive intestinal polypeptide in the suprachiasmatic nucleus leads to aging-like alterations in cAMP rhythms and activation of gonadotropin-releasing hormone neurons. J Neurosci 25:62–67 [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Harney JP, Scarbrough K, Rosewell KL, Wise PM 1996 In vivo antisense antagonism of vasoactive intestinal peptide in the suprachiasmatic nuclei causes aging-like changes in the estradiol-induced luteinizing hormone and prolactin surges. Endocrinology 137:3696–3701 [DOI] [PubMed] [Google Scholar]
  64. van der Beek EM, Swarts HJM, Wiegant VM 1999 Central administration of antiserum to vasoactive intestinal peptide delays and reduces luteinizing hormone and prolactin surges in ovariectomized, estrogen-treated rats. Neuroendocrinology 69:227–237 [DOI] [PubMed] [Google Scholar]
  65. Palm IF, van der Beek EM, Wiegant VM, Buijs RM, Kalsbeek A 1999 Vasopressin induces a luteinizing hormone surge in ovariectomized, estradiol-treated rats with lesions of the suprachiasmatic nucleus. Neuroscience 93:659–666 [DOI] [PubMed] [Google Scholar]
  66. Palm IF, van der Beek EM, Wiegant VM, Buijs RM, Kalsbeek A 2001 The stimulatory effect of vasopressin on the luteinizing hormone surge in ovariectomized, estradiol-treated rats is time-dependent. Brain Res 901:109–116 [DOI] [PubMed] [Google Scholar]
  67. Miller BH, Olson SL, Levine JE, Turek FW, Horton TH, Takahashi JS 2006 Vasopressin regulation of the proestrous luteinizing hormone surge in wild-type and Clock mutant mice. Biol Reprod 75:778–784 [DOI] [PubMed] [Google Scholar]

Articles from Endocrinology are provided here courtesy of The Endocrine Society

RESOURCES