The role of GABAA receptors in the neural systems of the medial preoptic area in the control of GnRH release during the luteal phase of the oestrous cycle in ewes

Recent studies from our laboratory suggested that functional interconnection of the GABAergic system with the GnRHergic, β-endorphinergic and catecholaminergic systems in the medial preoptic area (MPOA) plays an important role in the modulation of gonadoliberin (GnRH) release during the follicular phase of the oestrous cycle in ewes. Since the mode of action of γ-aminobutyric acid (GABA) on GnRH neurons is highly dependent on gonadal steroids, we extended our studies to the role of GABAA receptors in the MPOA on the activity of all of the above-mentioned systems during the luteal phase of the oestrous cycle. Stimulation of GABAA receptors by muscimol did not affect either GnRH/LH, β-endorphin release or catecholaminergic activity, expressed by extracellular concentrations of noradrenaline (NE), dopamine (DA) and their metabolites MHPG and DOPAC. Blockade of GABAA receptors decreased β-endorphin release, dopaminergic and noradrenergic activity but did not change GnRH release or LH secretion. It is suggested that progesterone secretion during the luteal phase may desensitize GABAA receptors to further stimulation.


INTRODUCTION
Neuroendocrine control of GnRH/LH release involves both excitatory and inhibitory components.Among these transsynaptic regulatory systems, γ-aminobutyric acid (GABA), a major inhibitory neurotransmitter in the hypothalamus (Decavel and van den Pol, 1990) and preoptic area (Robinson, 1995(Robinson, , 1997)), affects gonadoliberin (GnRH) release by two classes of membrane receptors: GABA A (Sieghart, 1995) and GABA B (Mott and Lewis, 1994).Numerous experiments indicate that GABA mediates both inhibition (Scott and Clarke, 1993;Leonhardt et al., 1995) as well as stimulation (Jarry et al., 1992;Leonhardt et al., 2000) of GnRH release by GABA A receptor mechanism(s).The dual inhibitory-stimulatory action of GABA on GnRH release is not clearly understood.GABA may inhibit or stimulate GnRH release in the hypothalamus in at least two ways: first, directly by a GABA A receptor mechanism on GnRH neurons (Petersen et al., 1993;Jung et al., 1998) and second, through GABA A receptor processes on inhibitory or stimulatory interneurons that impinge on GnRH cells (Horvath et al., 1992(Horvath et al., , 1993)).Indeed, our previous results strongly suggest that functional interconnection of GABAergic neurons with β-endorphinergic and dopaminergic systems in the ventromedial hypothalamus-nucleus infundibularis (VEN/NI) (Tomaszewska-Zaremba et al., 2003a) and in the preoptic area (MPOA) (Tomaszewska-Zaremba et al., 2003b) plays an important role in the control of GnRH release during the follicular phase of the oestrous cycle.Since GnRH neuronal activity and function of the β-endorphinergic (Domański et al., 1991;Conover et al., 1993), noradrenergic and dopaminergic (Chomicka et al., 1994) systems in the control of GnRH release depend on gonadal hormone concentrations and changes in various phases of the reproductive cycle, it is reasonable to suggest that the influence of GABA on these systems varies during the time course of the oestrous cycle.This assumption prompted us to investigate the effect of a GABA A receptor agonist, muscimol, and antagonist, bicuculline, on GnRH and β-endorphin (B-END) release, and on catecholaminergic activity in the MPOA of ewes during the luteal phase of the oestrous cycle.

Animals
The studies were performed on four-year-old Polish Merino ewes, well adapted to the experimental conditions.Using a stereotaxic procedure (Traczyk and Przekop, 1963;Welento et al., 1969), permanent guide cannulae were positioned on the skull at least three weeks prior to the perfusion procedure.The surgical opera-tion of ewes was carried out under Vetbutal (Biovet, Puławy, Poland) anaesthesia.The guide cannulae were directed towards the MPOA by a stereotaxic procedure and were secured to the skull with screws and dental cement.Throughout the study the animals were maintained indoors at a temperature of 12-15°C in individual pens and exposed to natural daylight.They always had visual contact with their neighbours, even during the sampling periods, to prevent the stress of social isolation.
Feed and water were available ad libitum.On the day of experiments two hours prior to perfusion, push-pull cannulae were introduced through the guide cannulae and placed in the proper position according to coordinates for the ovine hypothalamus.Each animal received two perfusions: first, a control perfusion with Ringer solution on the 7-th day of the oestrous cycle, and a second perfusion with muscimol (10 μg muscimol/ml Ringer solution) or bicuculline (30 μg bicuculline/ml Ringer solution) on the 7-th day of the next oestrous cycle.Each ewe served as a control.After the experiments the animals were euthanised with a barbiturate overdose.The procedures were done with the consent of the Local Ethical Committee at Warsaw University of Agriculture.All experiments were performed during September-November.

Perfusate collection
The perfusions were performed at a flow rate of 5 μl/min and perfusates were collected continuously for 5 h (9.00-14.00) in 30-min fractions into tubes containing 50 μl 0.1 mM ascorbic acid.The perfusates were kept in an ice bath during sampling and stored at -80 o C until assay.To determine the site of perfusion and to localize the place from which the perfusates were sampled, the brain of each animal was infused with 20 μl Prussian blue for 10 min.Then the brains were removed and sectioned sagitally under a stereoscopic binocular.Stained tissue was established in a spherical fraction about 2.0-2.5 mm around the tip of the cannula.The perfusates of animals with misplaced cannulae were excluded from analysis.

Analysis of β-endorphin in perfusates
Extraction of β-endorphin-like immunoreactivity compound(s) was performed by a method similar to the procedure described by Leshin and Malven (1984) and modified as described previously (Tomaszewska-Zaremba et al., 2003).β-endorphin antibodies were kindly provided by G. Martin from The University of West Australia, Needlands.The detection limit was 6 pg/50 μl.The recovery of β-endorphin-like immunoreactivity calculated on the basis of extraction using 125 Iβ-endorphin added to plasma samples was 85-87%.Intra-and interassay coefficients of variation were 11 and 16%, respectively.

Radioimmunoassay of GnRH
The RIA procedure for GnRH was similar to that previously reported (Domański et al., 1991).The reference GnRH was obtained from Sigma.GnRH was radioiodinated using the chloramine-T method.GnRH antibodies (kindly provided by J. Kosowicz, Medical School, Poznań, Poland) at a final dilution of 1:16000 were used for determination.The detection limit was 1.9 pg/100 μl.The intra-and interassay coefficients were 7 and 11%, respectively.

Radioimmunoassay of LH
Plasma LH concentration was assayed by a double-antibody RIA using antiovine -LH and antirabbit -gammaglobulin antisera and ovine LH standard (NIH-LH-SO18) according to Stupnicki and Madej (1976).The assay sensitivity was 0.06 ng/ml and intra-and interassay coefficients of variation were 4 and 10%, respectively.

Data analysis
Control levels of NE, DA, MHPG and DOPAC in the MPOA were evaluated in perfusates collected from Ringer-solution-treated animals.The effect of perfusion of muscimol or bicuculline into the MPOA on the extracellular concentration of these compound(s) was expressed as the percent change in the respective fraction pairs of muscimol or bicuculline to control animals, respectively.All results are presented as mean ± SEM.One-way ANOVA was used followed by Tukey's test to evaluate differences between means of concentrations in the respective pair fractions in perfusates from muscimol-or bicuculline-treated animals to controls.
Β-END concentrations are expressed as means ± SEM and were assessed by one-way ANOVA following by Tukey's test.GnRH concentrations are expressed as means ± SEM and were assessed by one-way ANOVA, differences between TOMASZEWSKA-ZAREMBA D. ET AL. muscimol or bicuculline and the control treatment in particular pair fractions were evaluated by Tukey's test.
Plasma LH concentrations are expressed as a mean ± SEM.The significance of differences between the results of control and muscimol or bicuculline treatment was assessed by ANOVA followed by the least significant differences (LSD) test (STATISTICA).The number of LH pulses and amplitude was determined by the PC-PULSAR computer program according to the method of Marriam and Wachter (1982) with G parameters: G1 = 3.98; G2 = 2.40; G3 = 1.68,G4 = 1.24 and G5 = 0.93.The frequency in controls and muscimol-or bicuculline-treated animals was defined as the number of identified pulses per collecting period and is expressed as a mean ± SEM.The amplitude was defined as the difference between peak and nadir values (Viguie et al., 1995).Differences in LH pulse frequency and amplitude in the vehicle and muscimol or bicuculline perfused groups were analysed by the Wilcoxon test.

Effects of muscimol and bicuculline perfusion in the MPOA on the extracellular concentration of GnRH in this structure and LH level in blood plasma
A schematic diagram showing the location of the perfusion sites in the MPOA is presented in Figure 1.The concentration of GnRH in the control perfusates was low and in some samples the value of this hormone was at, or near the assay sensitivity limit.Muscimol perfusion did change neither the GnRH concentration in perfusates (Figure 2a) nor LH pulse frequency, pulse amplitude and LH level in blood plasma (data not shown).In all animals after muscimol perfusion the oestrous cycle occurred at the physiological time.
Perfusion with bicuculline in the MPOA did not cause evident changes in the extracellular concentration of GnRH in all animals (Figure 2b).Bicuculline did not change either LH concentration in blood plasma or the pattern of this hormone's secretion, i.e. the pulse amplitude and frequency.

Effects of muscimol and bicuculline perfusion in the MPOA on the extracellular concentration of Β-END-LI in perfusates
During the luteal phase of the oestrous cycle in ewes, the extracellular concentration of Β-END-LI in this structure varied markedly.Analysis of the response magnitude to muscimol in individual ewes showed that only in one animal did the decline of the Β-END-LI concentration reach statistical significance as compared with controls.The mean concentration of Β-END-LI in perfusates in the muscimol-treated animals therefore did not differ significantly as compared with levels in these ewes during vehicle perfusion (Figure 3a).Analysis of the magnitude of the β-endorphin response to bicuculline treatment in individual ewes indicated that blockade of GABA A receptors failed to change this opioid concentration in 2 animals; in the remaining 4 ewes bicuculline significantly decreased Β-END-LI levels in perfusates.The overall mean concentration of Β-END-LI in perfusates collected during bicuculline treatment differed significantly from controls (Figure 3b).

Effect of muscimol and bicuculline perfusion in the MPOA on the extracellular concentration of NE, DA, MHPG, DOPAC
In the control treatment the concentrations of NE and DA in perfusates ranged from 6 to 42 pg/50 μl and from 15 to 104 pg/50 μl, respectively, depending on the individual ewe.The concentrations of MHPG in perfusates ranged from 27 to 276 pg/50 μl and DOPAC from 55 to 383 pg/50 μl in individual animals.Perfusion of muscimol did not cause a marked change in extracellular concentrations of catecholamines or their metabolites (Figure 4).
In all animals bicuculline treatment significantly depressed the extracellular concentrations of NE, DA and their metabolites, MHPG and DOPAC (Figure 4).

DISCUSSION
These data show that perfusion of muscimol into the MPOA of ewes during the luteal phase of the oestrous cycle did not significantly alter either GnRH or β-endorphin release or catecholaminergic activity in this structure.Lack of changes in the pattern of LH secretion strongly suggests that the GABA A receptor agonist in the MPOA did not alter GnRH release into the VEN/NI.This point of view is supported by our previous findings showing that muscimol perfusion in the VEN/NI apparently did not affect GnRH/LH secretion in ewes during the luteal phase of the oestrous cycle (Tomaszewska-Zaremba et al., 2002).Since blockade of GABA A receptors in the MPOA of ewes during the luteal phase of the oestrous cycle suppressed both β-endorphin release and catecholaminergic system activity, the failure of muscimol to alter activity of these systems is not likely a reflection of either a minor or no role for GABA A receptors in the MPOA in modulating GnRH, β-endorphin, NE and DA release.The dose of muscimol was chosen according to the literature (Scott and Clarke, 1993) and on the basis of our earlier works.Higher doses of muscimol caused toxic symptoms (convulsions, salivate).In this setting, one plausible explanation for lack of response of these systems to the GABA A receptor agonist would be that the progesterone-induced GABA A release during the luteal phase may desensitize GABA A receptors and make them resistant to further stimulation.Indeed, recent works have revealed a clear stimulatory effect of progesterone on GABA release in the medial preoptic area (Robinson and Kendrick, 1992;Robinson, 1995).In the light of these results it is reasonable to suggest that progesterone secreted during the luteal phase of the oestrous cycle may affect the dynamics of GnRH release through a influence on GABA release or indirectly by changes of β-endorphinergic and catecholaminergic system activities.In such multifactoral actions of GABA on gonadotropin secretion it is not possible to distinguish in detail the action of the GABAergic system in the control of GnRH release.This is particularly so because the applied method did not enable analysis of the direct influence of muscimol on the activity of GnRH neurons.
Blockade of GABA A receptors may also affect in different ways the mechanism(s) controlling GnRH release.The presented results indicate that bicuculline reduced the tone of β-endorphinergic activity but did not increase GnRH release; such event is not strange because it is well documented that the inhibitory effects of β-endorphin on GnRH release are only marginally apparent in the luteal phase of the oestrous cycle in ewes (Conover et al., 1993).The inhibitory effect of β-endorphin on GnRH/LH release (Curlevis et al., 1991;Domański et al., 1991;Conover et al., 1993) as well as the stimulatory influence of naloxone (Conover et al., 1993) on GnRH secretion in the VEN/NI in ewes increases distinctly throughout the follicular phase reaching the highest value just prior to the preovulatory surge of gonadotropin.However, at the MPOA of ovariectomized-progesterone treated ewes the decrease of opioidergic activity by naltrexon suppressed the stimulatory influence of progesterone on GABA release and enhanced LH secretion (Tartonese, 1999).Our present findings together with those of previous reports support the notion that the specific actions of β-endorphin on GnRH/LH secretion is dependent upon the ovarian hormones concentration.
Temporal analysis of the relationship between bicuculline treatment and extracellular concentrations of DA and DOPAC indicated that the GABA A receptor antagonist decreases dopaminergic activity in the MPOA and that dopamine does not play an important role in this structure on GnRH release during the luteal phase of the oestrous cycle.The role of dopamine in the regulation of GnRH release is still not consistent to provide a clear understanding of its action in this process.
The increased dopaminergic activity in the VEN/NI during the luteal phase suggests an inhibitory effect of DA on GnRH release (Chomicka et al., 1994); the other results rise the possibility that the dopaminergic system in this structure is part of the neural pathways mediating the positive feedback effects of oestradiol on GnRH secretion (Andersen et al., 2001).On the other hand, in the MPOA the progesteroneinduced increase of GABA secretion did not change DA release, thus suggesting that the GABA-induced suppression of LH release cannot be attributed to increased dopaminergic system activity in this structure (Robinson et al., 1991).Similarly, lack of changes in GnRH/LH release with a concomitant decrease of noradrenergic system activity during the luteal phase under bicuculline treatment indicates that suppression of NE release has a rather minor role in GnRH release in the MPOA.The physiological significance of NE in the control of gonadotropin secretion is still not unequivocally established.Most of the results implicating noradrenaline in the regulation of GnRH release suggest the MPOA as the main site of the stimulatory influence of the noradrenergic system on gonadotropin secretion (Robinson et al 1991;Robinson and Kendrick, 1992); however, the mode of action of noradrenaline on GnRH release seems highly dependent on the influence of season and steroid hormone concentration (Scott et al., 1992;Goodman, 1996).

Figure 4 .
Figure 4. Extracellular concentrations of NE, DA, MHPG and DOPAC in MPOA of ewes during perfusion of muscimol and bicuculline.Data are expressed as a percent of the control values in the respective pair fractions (muscimol/bicuculline treatment to Ringer solution treatment), data presented are mean values ± SEM (n=6, *P≤0.05,**P≤0.01)