| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Institut National de la Santé et de la Recherche Médicale, Unité Mixte de Recherche 788, Université Paris 11, 94276, Le Kremlin-Bicêtre, France
Address all correspondence and requests for reprints to: Dr. Martine. El-Etr, Institut National de la Santé et de la Recherche Médicale, Unité Mixte de Recherche 788, 80 rue du Général Leclerc, 94276, Le Kremlin-Bicêtre, France. E-mail: eletr{at}kb.inserm.fr.
| Abstract |
|---|
|
|
|---|
,1,
3,
4, and
2, corresponding to the NR1, NR2C, NR2D, and NR2B rat subunits) and increase their spontaneous release of GnRH when incubated in the presence of exogenous glutamate or NMDA. In addition, we found that these neurons are able to release and synthesize glutamate, as demonstrated by the presence of glutamate accumulated in the defined incubation medium of the neurons, during the experiment and the expression of mRNA coding for vesicular glutamate transporter 2, a specific marker of glutamatergic neurons. The potentiating effect of PREGS on the secretion of GnRH induced by glutamate is consistent with the role of the steroid as a positive allosteric modulator of NMDA receptors. Together these results point to a novel mechanism by which the neuroactive steroid PREGS may potentiate an autocrine excitatory loop in GnRH neurons. | Introduction |
|---|
|
|
|---|
-dihydrotestosterone (6), the progesterone metabolite allopregnanolone (7, 8), and dehydroepiandrosterone sulfate (DHEAS), which decreases the electrophysiological responsiveness of GnRH neurons to the activation of
-aminobutyric acid A (GABAA) receptors (9). Whether the neuroactive steroid pregnenolone sulfate (PREGS) can also modulate the release of GnRH is as yet unknown.
PREGS does not bind to any classical intracellular steroid receptor. Its effects on central nervous system excitability are rapid, mainly through modulation of different channel-gated neurotransmitter receptors; indeed, it negatively interacts with GABAA receptors (10), a common property of 3ß-hydroxysteroid sulfates (11), and counteracts glycine receptor function (12); it also inhibits D,L-
-amino-3-hydroxy-5-methyl-4-isoxazol propionic acid (AMPA) and kainate receptor activities (13), whereas it potentiates most N-methyl-D-aspartate (NMDA) receptor-mediated glutamate excitatory responses (14), depending on the subunit composition of the receptors (15). A recent study suggests that PREGS may interact with NMDA receptors at a hydrophobic binding pocket identified on the NR2B subunit (
2 in the mouse) (16).
Among the modulators of hypothalamic GnRH secretion, glutamate plays an important part, because it is involved in the mechanisms of the GnRH pulse generator (17, 18, 19, 20), the induction of puberty in males (21) and females (22), including monkeys (23), and the preovulatory GnRH surge (24). Moreover, GnRH responses to glutamate through NMDA receptor stimulation can be modulated according to the age, being significantly decreased in old rats (25, 26). Indeed, compromised reproductive physiology with aging may be related to changes in NMDA receptor subunit composition and stoichiometry in GnRH neurons, with an increase in NR2B subunit affecting functional channel characteristics (27).
Using GT1-7 cells, hypothalamic GnRH neurons of mice, immortalized by genetically targeted tumorigenesis (28), we observed that PREGS, in the absence of exogenous glutamate or NMDA, could stimulate the release of GnRH, and this effect involved the activation of NMDA receptors. GT1-7 cells were also found to express vesicular glutamate transporter 2 (VGLUT2) and to release glutamate into the incubation medium. Together, these results strongly suggest that PREGS may potentiate the autocrine stimulatory action of glutamate on GnRH neurons.
| Materials and Methods |
|---|
|
|
|---|
GnRH release
On the day of the experiment, confluent GT1-7 cells, seeded on six-well dishes (4.6 ± 0.7 x 106 cells/well) were washed three times for 5 min each time with 2 ml magnesium-free Lockes medium [154 mM NaCl, 5.6 mM KCl, 2.2 mM CaCl2, 6 mM NaHCO3, 10 mM glucose, and 2 mM HEPES (pH 7.4)]. They were then incubated at 37 C for 30 min in 1 ml Lockes medium supplemented with bacitracin (20 µM; Sigma-Aldrich Corp.) and glycine (10 µM; Merck & Co., Rahway, NJ) in the absence or presence of PREGS (5100 µM), glutamate (5 µM-1 mM), NMDA (5 µM to 1 mM,) and/or the GABAA receptor channel blocker picrotoxin (100 µM), the NMDA receptor antagonist D,L-2-amino-5-phosphono-valeric acid (AP5; 10 µM), NMDA receptor channel blockers MK 801 (5 µM) and Mg2+ (6 mM), or the kainate-AMPA receptor antagonist cyano-3,3-dihydro-7-nitroquinoxaline (CNQX; 10 µM; all from Sigma-Aldrich Corp.). Three wells were used per condition. At the end of the incubation period, media (1 ml) were stored at 20 C until radioimmunoassayed for GnRH.
GnRH RIA
As previously described (7), the concentration of GnRH released into the medium (1 ml/well) was measured by RIA in triplicate by a charcoal precipitation method using [125I]GnRH (2000 Ci/mol; 1 Ci = 37 GBq; Amersham Biosciences, Les Ulis, France), unlabeled GnRH (Fluka, Buchs, Switzerland; Sigma-Aldrich Corp.) as the reference standard, and the rabbit polyclonal antibody R 1245 (30), which is specific for the decapeptide (obtained from T. Nett, Colorado State University, Fort Collins, CO), Briefly, 100 µl antibody (1:76,800 final dilution) was added to each tube already containing 200 µl sample (or unlabeled GnRH for the standard curves) and 200 µl GnRH assay buffer (145 mM NaCl, 25 mM EDTA, 3.3 mM NaH2PO4, and 6.7 mM Na2HPO4) supplemented with merthiolate (10 mg/liter; Sigma-Aldrich Corp.) and gelatin (1 g/liter; pH 7.4). After 2 h at 4 C, 100 µl [125I]GnRH (12,000 cpm) was added to all tubes, and incubation was continued for an additional 24 h. Seven hundred and fifty microliters of a charcoal/dextran suspension (2.5 and 0.25 g/liter, respectively, in Ca2+- and Mg2+-free PBS) were then added. The tubes were centrifuged at 4000 x g for 10 min, and the supernatants were counted for radioactivity. All samples from an experiment were analyzed in the same assay. The lower limit of detection was 4 pg/ml.
Detection of NMDA receptor subunit mRNAs and VGLUT2 mRNA
The expression of mouse NMDA receptor subunits (
1 and
14, corresponding to the rat NR1 and NR2A-D subunits, respectively) was detected by RT-PCR. Total RNA was isolated from GT1-7 cells or from mouse brain regions using TRIzol reagent (Invitrogen Life Technologies, Inc.) and treated with deoxyribonuclease I-ribonuclease-free (Stratagene, La Jolla, CA) to remove potential contaminant DNA. RNAs were controlled on a 1.2% agarose gel stained with ethidium bromide. One microgram of total RNA was incubated for 5 min at 70 C with 100 U of the ribonuclease inhibitor HPRI (Amersham Biosciences) and random hexamer primers (New England Biolabs, Beverley, MA), then reverse transcribed for 90 min at 42 C using 200 U SuperScript II Moloney murine leukemia virus reverse transcriptase (Invitrogen Life Technologies, Inc.) in the presence of 200 µM of each nucleotide triphosphate (deoxy-NTPs; Invitrogen Life Technologies, Inc.) in a total volume of 26 µl. One height of the RT reaction (3 µl) was used as a template for amplification by PCR using a thermal cycler (Stratagene). The amplification mixture contained cDNA, 25 pmol of each specific forward and reverse primer, 200 µM deoxy-NTPs, 1.5 mM MgCl2, 5 µl buffer, and 1.25 U Hot Star Taq DNA polymerase (QIAGEN, Hilden, Germany) in a total volume of 50 µl. Primers were purchased from Sigma-Genosys Ltd. (Pampisford, UK). The nucleotide sequences for sense and antisense primers are shown in Table 1
. After an initial denaturation at 94 C for 2 min, each amplification cycle (30 cycles) consisted of denaturation at 94 C for 1 min; annealing at 66 C (
1), 48 C (
1), 50 C (
2), 68 C (
3), or 60 C (
4) for 1 min; and extension at 72 C for 1 min, with a final extension time of 10 min.
|
,1,
1,
2, and
4) or cerebellum (
3) as positive controls.
The expression of the glutamate transporter VGLUT2 mRNA was also studied by RT-PCR. Specific nucleotide sequences for sense and antisense primers are shown in Table 1
. Five micrograms of total RNA were used for the RT. PCR was performed (40 cycles) in the same way as described above, using a temperature of annealing of 50 C.
Glutamate and glycine release
When cells reached 90% confluence, C-DMEM was replaced by serum-free medium (OptiMEM; Invitrogen Life Technologies, Inc.) for 24 h. On the day of the experiment, GT1-7 cells were carefully washed three times (5 min) with Lockes medium, then incubated in Lockes medium with or without PREGS. This incubation medium was collected at 5 and 30 min and stored at 80 C. Glutamate and glycine concentrations were determined according to a modified version of the method described by Geddes and Wood (31), using a Waters HPLC-fluorometric detection system including a precolumn derivatization with o-phtaldialdehyde/mercaptoethanol reagent and a C18 (ODS2, 4.6 x 150 mm) Spherisorb column (Waters Corp., Grand Rapids, MI). A nonlinear gradient delivered through a Waters 600 pump was used to separate derivatives [solvent A, 0.1 M potassium acetate and 25% methanol (pH 5.5); solvent B, 0.05 M (13) potassium acetate and 60% methanol (pH 5.5)]. Samples (10 µl) were automatically injected (Waters 717 Plus autosampler) and analyzed using a Waters 474 detector. The limit of detection was 1 pmol/sample. Data were computed with Waters Millenium software (via a Waters bus SAT/IN module) running onto a personal computer system; compound identification and peak quantification were achieved by comparison with standard solutions.
Statistical analysis
Data were expressed as the mean ± SEM of three to five independent experiments, each performed in triplicate. They were analyzed by one-way ANOVA, followed by Fishers multiple comparison test at the 0.05 level of significance.
| Results |
|---|
|
|
|---|
|
-androstan-17-one sulfate) had no effect, but epiallopregnanolone sulfate (3ß-hydroxy-5
-pregnan-20-one sulfate) stimulated GnRH release by 2-fold when used at 10 µM (data not shown). All 3
-sulfated steroids tested, allopregnanolone sulfate (3
-hydroxy-5
-pregnan-20-one sulfate), pregnanolone sulfate (3
-hydroxy-5ß-pregnan-20-one sulfate), and androsterone sulfate (3
-hydroxy-5
-androstan-17-one sulfate), were inactive.
PREGS enhancement of GnRH release involves NMDA receptors, but not AMPA/kainate or GABAA receptors
When GT1-7 cells were incubated with PREGS (10 µM) in the presence of the GABAA receptor channel blocker picrotoxin (100 µM), GnRH release remained unchanged compared with that induced by PREGS alone (Fig. 2A
). Similarly, coincubation with the selective AMPA/kainate receptor antagonist CNQX (10 µM) did not modify the release of GnRH by PREGS (10 µM) (Fig. 2A
).
|
GT1-7 neurons express several NMDA receptor subunit mRNAs
Among the five subunits of the NMDA receptor identified in the mouse, we detected in GT1-7 neurons the expression of the obligatory
1 mRNA (NR1 in the rat), but no
1 mRNA (NR2A). We also found
3,
4, and to a lesser extent
2 subunit mRNAs, which, respectively, correspond to the rat, NR2C, NR2D, and NR2B subunits (Fig. 3
).
|
|
|
| Discussion |
|---|
|
|
|---|
The previous in vivo observations that NMDA elicited pituitary LH release in the prepubertal male monkey (32) and advanced the onset of puberty in the female rat and male monkey (33, 34, 35), whereas NMDA receptor antagonists suppressed the pulsatile secretion of LH (36), did not elucidate the precise site of action of NMDA: exclusively in the pituitary gland or also in the hypothalamus. In vitro experiments on rat hypothalamic explants demonstrated that glutamate or NMDA perifusion induced GnRH release (37, 38, 39), although the cellular target of excitatory amino acids remained unclear. Subsequently, the presence of NMDA receptors at the surface of GnRH neurons was demonstrated (34) and confirmed when transgenic mice expressing green fluorescent protein (GFP) under control of the GnRH promoter (GnRH-GFP mice) were generated and used to determine the electrophysiological properties of identified GnRH neurons in brain slice preparations (40, 41); it was thus clearly observed that functional glutamate receptor channels of the AMPA and NMDA subtypes were present in the membrane of GnRH neurons. Finally, Ottem et al. (42) recently demonstrated that the majority of GnRH neurons located in the medial preoptic area of the hypothalamus expressed the NMDA R1 gene.
In immortalized GT1 neurons, conflicting data have been reported. Mahesh et al. (43) did not find any electrophysiological or binding characteristics of NMDA receptors in GT1-7 cells despite the expression of NMDAR1 mRNA. In contrast, Mahachoklertwattana (44) reported that NMDA stimulated the release of GnRH from GT1-1 neurons, which expressed NMDAR1 receptor transcripts. Urbanski et al. (45) observed the presence of NMDAR1 mRNA in GT1-7 neurons, and Spergel et al. (46) showed that glutamate increased intracellular calcium and GnRH release from GT1-7 cells through NMDA receptors. In the present study we report a stimulating effect of glutamate and NMDA on GnRH release and observe that GT1-7 neurons express various NMDA receptor subunit mRNAs. Indeed, among the five subunits identified in the mouse (47), GT1-7 neurons express not only the obligatory
1 mRNA, but also
3,
4, and
2, but not
1 (NR1, NR2C, NR2D, NR2B, and NR2A, respectively, in the rat). Interestingly, a steroid modulatory domain has recently been identified on the NR2B subunit that is critical for PREGS sensitivity (16).
PREGS has been described as a positive allosteric modulator, not as an agonist of the NMDA receptor. However, in cultures of GT1-7 neurons, the effect of PREGS occurred in the absence of added NMDA or glutamate, suggesting the presence of endogenous excitatory amino acids. When measuring in the incubation medium, the amount of glutamate and glycine, a well-known coagonist of the NMDA receptor (48), their concentrations were found to increase between 5 and 30 min of incubation. The amount of secreted glutamate was too low to induce stimulation of GnRH release and probably did not participate in the basal release of GnRH, because the NMDA receptor antagonists AP5, MK 801, and Mg2+ did not lower the spontaneous liberation of GnRH. However, it was probably sufficient to allow positive modulation of the activity of NMDA receptors by PREGS, leading to stimulation of GnRH release.
The question remained whether the release of glutamate occurred after its uptake from OptiMEM medium, in which cells had been cultured 24 h before the experiments or whether GT1-7 cells could synthesize it. For a very long time, no reliable method for detecting glutamatergic neurons existed. This obstacle has recently been overcome with the characterization of VGLUTs, which are specific markers of glutamatergic neurons (49). Indeed, neurons that are able to synthesize glutamate must have the capacity to package and release it via specific VGLUT. Three types of VGLUTs have been described (50): although VGLUT3 is not detected and VGLUT1 is very scarce in the hypothalamus, VGLUT2 is expressed, especially in the arcuate nucleus and median eminence (51). Our data suggest that GT1-7 cells are likely to synthesize glutamate as they express VGLUT2 mRNA and release glutamate into the incubation medium. These data are in accordance with the recent observation by Hrabovsky et al. (52) of the expression of VGLUT-2 in GnRH neurons of the adult male rat.
Immortalized GT11 and GT1-7 neurons secrete GnRH, but have also been shown to secrete GABA (53), express GABAA receptors, and release GnRH under muscimol stimulation (Ref. 7 for GT11 cells and El-Etr, M., personal unpublished observations for GT1-7 cells), suggesting a possible autocrine influence of GABA on GnRH secretion. The cosecretion of GABA and glutamate by the same neurons has been previously described in the hypothalamus; indeed, some neurons present in the anteroventral periventricular nucleus, a region of the preoptic area involved in the regulation of GnRH neuron function, are able to secrete both amino acids, as demonstrated by their capacity to coexpress VGLUT2, glutamic acid decarboxylase, and vesicular GABA transporter (54). However, to our knowledge, this is the first observation reported of GABA (53) and glutamate (our present data) cosecretion by GnRH neurons.
GnRH neurons represent the final output pathway of a neuronal network that integrates multiple environmental and internal factors, such as steroid hormones, to control fertility in both sexes (55, 56). These regulations by steroids can be indirect, via astrocytes (3, 57) or surrounding interneurons releasing various neurotransmitters and peptides (2), which, in turn, modulate GnRH secretion. However, steroids can directly influence GnRH neurons in a genomic and even nongenomic way; indeed, 17ß-estradiol is able to induce a fast nongenomic stimulation of GnRH release through ERß, which might participate in the positive feedback of estradiol (4, 58) and provoke a rapid inhibition (or stimulation, depending on the dose of estradiol) of GnRH secretion, through Gi-coupled membrane ER
(59).
Besides classical steroid hormones, a few neuroactive steroids have been shown to rapidly modulate GnRH release; the first experiments performed on male rat hemihypothalami had shown an inhibiting effect of allopregnanolone on GnRH release, that involved GABAA receptors and was counteracted by PREGS, but not by DHEAS. In the same preparations, GABA suppressed GnRH release (60). However, those studies could not ascertain whether these effects occurred at the GnRH neuron itself and/or at the presynaptic cells. Recently, using hypothalamic slices from transgenic GnRH-GFP mice, adult GnRH neurons were shown to maintain high intracellular chloride levels such that the chloride reversal potential was depolarized relative to the threshold for action potential firing, and direct activation of GABAA receptors on these cells was excitatory regardless of the sex and hormonal state of the animals (61). In agreement with these observations, other studies reported that allopregnanolone enhanced the electrical responses to GABA of juvenile and adult female mouse GnRH neurons from hypothalamic brain slices (8) and increased the electrophysiological responsiveness to GABA of GnRH neurons from adult GnRH-GFP mice, whereas DHEAS reversed this effect (9). In both GT11 and GT1-7 cell cultures, allopregnanolone stimulates GnRH release through GABAA receptors, PREGS counteracts this effect, whereas PREGS alone is able to enhance GnRH release (7 ; and El-Etr, M., personal unpublished data). Overall, the local concentrations of magnesium and GABA might differentiate between the inhibitory actions of PREGS at the GABAA receptor and its stimulatory effects through NMDA receptors.
Our present results, which suggest a role of NMDA receptor modulation by PREGS in the control of GnRH release, are likely to be of physiological significance: 1) GnRH neurons indeed express NMDA receptors in vivo, and NMDA stimulates GnRH release by acting on the hypothalamus (see above); 2) PREGS is an important positive allosteric modulator of NMDA receptors (15, 16); 3) the positive modulation of NMDA receptors by PREGS has been shown to be involved in the regulation of important brain functions, including memory processes (62, 63). Whether PREGS is an endogenous neurosteroid in the rodent brain has recently been questioned (for a critical evaluation, see Ref. 64). Whatever the outcome of this controversial issue, however, it is well established that PREGS is present in human blood (64) and brain, including the hypothalamus (65).
PREGS stimulated the release of GnRH from GT1-7 neurons at concentrations ranging from 10100 µM. Whether these concentrations can be locally reached within the nervous system remains a matter of speculation and cannot be solved by currently available assay methods, even by the very sensitive analysis of steroids by mass spectrometry. However, it is known that in vitro experimental paradigms sometimes require high concentrations of steroids, and in particular of their sulfated forms. The modulation of NMDA receptors by PREGS using electrophysiological recordings has been extensively studied (15, 66). In their studies the positive allosteric modulation of NMDA receptors by PREGS in various neuronal preparations required very high micromolar concentrations of the steroid (10100 µM), consistent with our present findings. More recently, the same group identified a PREGS modulatory domain on the NMDA receptor, strongly suggesting that the steroid directly acts on the receptor (16).
In conclusion, in addition to the positive effect of PREGS on GnRH gene expression in rat GnRH neurons (67) the present data show that via NMDA receptors, the neuroactive steroid PREGS can rapidly enhance the release of GnRH by immortalized GT1-7 cells, through the potentiation of the autocrine actions of glutamate, and may thus contribute to a rapid regulation of GnRH release at the level of the GnRH neuron.
| Acknowledgments |
|---|
| Footnotes |
|---|
First Published Online March 2, 2006
Abbreviations: AMPA, D,L-
-Amino-3-hydroxy-5-methyl-4-isoxazol propionic acid; AP5, D,L-2-amino-5-phosphono-valeric acid; C-DMEM, conditioned-DMEM; CNQX, cyano-3,3-dihydro-7-nitroquinoxaline; DHEAS,dehydroepiandrosterone sulfate (3ß-hydroxy-5-androsten-17-one sulfate); GABA,
-aminobutyric acid; MK-801, dizocilpine maleate; NMDA, N-methyl-D-aspartate; GFP, green fluorescent protein; PREGS, pregnenolone sulfate (3ß-hydroxy-5-pregnen-20-one sulfate); VGLUT, vesicular glutamate transporter.
Received September 19, 2005.
Accepted for publication February 21, 2006.
| References |
|---|
|
|
|---|
-aminobutyric acid type receptor. Proc Natl Acad Sci USA 92:37693773
-aminobutyric acid postsynaptic currents in gonadotropin-releasing hormone neurons: possible mechanism for direct steroidal control. Endocrinology 144:43664375
-aminobutyric acid receptor complex in brain? Brain Res 404:355360[CrossRef][Medline]
-amino-butyric acid- and glycine-induced currents by progesterone. Mol Pharmacol 37:597602[Abstract]
-aminobutyric acid. J Neurochem 41:1624
-aminobutyric acid: evidence for an autocrine regulation. Eur J Neurosci 10:11451152[CrossRef][Medline]
-aminobutyric acid A receptor. J Endocrinol 158:121125[Abstract]
-aminobutyric acid receptors excites gonadotropin-releasing hormone neurons. Mol Endocrinol 16:28722891This article has been cited by other articles:
![]() |
B. M. Nathan, C. A. Hodges, P. J. Supelak, L. C. Burrage, J. H. Nadeau, and M. R. Palmert A Quantitative Trait Locus on Chromosome 6 Regulates the Onset of Puberty in Mice Endocrinology, November 1, 2006; 147(11): 5132 - 5138. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |