INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

THE MEDIAL PREOPTIC AREA (MPOA), one of the hypothalamic regions, has an essential role in the regulation of various physiological functions, such as fluid volume (36), body temperature (14), and gonadal hormone release (20). The firing rates of hypothalamic neurons involved in such processes are sensitive to gonadal steroids (18, 23). In addition to genomic action of steroids as transcriptional regulators, they appear to have nongenomic activities such as rapid and direct interaction with membrane channels (23, 26, 31, 39). Progesterone and estrogen rapidly suppress luteinizing hormone (LH) release (33). One possible mechanism underlying such rapid action of steroids on neural function could be mediated by their metabolites. Neurosteroids, synthesized de novo in neurons (3) or glia (19) from steroids such as progesterone and androsterone, have been demonstrated to have nongenomic actions on various receptors (27, 30). The concentration of allopregnanolone (3-hydroxy-5-pregnan-20-one, 5-pregnan-3-ol-20-one, 3,5-tetrahydro-progesterone, the primary metabolite of progesterone) in the serum and brain changes in relation to naturally occurring hormonal dynamics such as the estrus cycle (27) and pregnancy (2), as well as during stress or after the intake of alcohol (40). Allopregnanolone also acts on the nervous system, where it has been shown to enhance Cl currents via -aminobutyric acid type A (GABA_A) receptors in the presynaptic nerve terminal (41).

MPOA neurons form local circuits in the preoptic area (10). Large numbers of MPOA neurons stain positive for glutamate decarboxylase, a GABA-synthesizing enzyme (5), and GABA is detectable at relatively high concentrations in the MPOA (22). Oscillations of GABA concentration in the MPOA produce synchronizing signals that may trigger release of gonadotropin-releasing hormone (GnRH) from the hypothalamus, resulting in constitutive LH secretion and pulsated release within the anterior pituitary (13). The effect of GABA on gonadotropic hormone release is mediated by the GABA_A receptor (20). Taken together, these facts suggest that GABAergic transmission plays an important role in the function of the MPOA.

To study the effects of neurosteroids on the activity of the GABAergic neural circuit in the MPOA, it will be necessary to better understand the effects of neurosteroids on the GABAergic neurotransmitter release from the presynaptic nerve terminal as well as their action on postsynaptic GABA_A receptors. Neurosteroid enhances the frequency of spontaneous GABA release in the MPOA (7). However, little known about the mechanisms of allopregnanolone action on neurotransmitter release from the presynaptic terminal in the central nervous system (CNS). The present study investigates the modulation of allopregnanolone on the release of GABA from the presynaptic nerve terminals in the MPOA by use of acutely dissociated MPOA neurons with still-attached afferent presynaptic terminals.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Preparation. Wistar rats (12-15 days old, either sex) were decapitated under pentobarbital sodium anesthesia (50 mg/kg ip). Brains were quickly removed and transversely sliced at a thickness of 350 µm with a microslicer (VT1000S; Leica, Solms, Germany). Before mechanical dissociation, slices were kept for 1 h at room temperature (22-25°C) in control incubation medium saturated with 95% O₂ and 5% CO₂. Pentobarbital is rapidly washed out of the pentobarbital-binding site of the receptor (42). Thus it is unlikely that the effect of pentobarbital used for anesthesia on chloride channels remained in recording. Immediately preceding dissociation, slices were transferred to a 35-mm culture dish (Primaria 3801, Becton Dickinson), and the region of the MPOA was identified under a binocular microscope (SMZ-1; Nikon). Details of the mechanical dissociation have been described previously (12). Briefly, mechanical dissociation was accomplished using a custom-built vibration device and a fire-polished glass pipette oscillating at a frequency of ~3-5 Hz (0.1-0.2 mm). The tip of the fire-polished glass pipette was gently placed on the surface of the MPOA region. The tip of the glass pipette, while lightly pressing on the tissue, was vibrated horizontally for ~2 min. Slices were removed from the dish while the remaining mechanically dissociated neurons were allowed to settle and adhere to the dish bottom for ~10 min. Neurons having undergone such dissociation retained their original morphological features, including short portions of the proximal dendrites.

Electrical measurements. All electrical measurements were performed using the nystatin-perforated patch recording method, which allows electrical access to the cytoplasm with limited intracellular dialysis (11, 25). All voltage clamp recordings were made at a holding potential (V_H) of 50 mV. Membrane voltage was controlled and currents recorded with the use of a patch-clamp amplifier (EPC-7; List). Patch pipettes were made from borosilicate capillary glass (1.5 mm OD, 0.9 mm ID; G-1.5; Narishige) in two stages on a vertical pipette puller (PB-7; Narishige). The resistance of the recording pipettes filled with internal solution measured between 3 and 6 M. Neurons were visualized under phase contrast on an inverted microscope (DMIRB; Leica). Current and voltage were continuously monitored on an oscilloscope (VC-6023; Hitachi) with a pen recorder (RECTI-HORIT-8K; San-ei) and recorded with a digital audiotape recorder (RD-120TE; TEAC). Membrane currents were filtered at 1 kHz (E-3201A Decade Filter; NF Electronic Instruments) digitized at 4 kHz and stored on a computer equipped with pCLAMP 8.0 software (Axon Instruments). All experiments were performed at room temperature (22-25°C).

Data analysis. Spontaneous miniature inhibitory postsynaptic currents (mIPSCs) were counted and analyzed using the MiniAnalysis program (Synaptosoft). Events were initially detected automatically by a preset amplitude threshold of 5 pA at a V_H of 50 mV and then visually accepted or rejected on the basis of rise and decay times. The amplitudes and inter-event intervals of large numbers of mIPSCs (>200) were examined by constructing cumulative probability distributions in each set of measurements. Numerical values are presented as means ± SE. Differences in amplitude and frequency were tested by Student's paired, two-tailed t-test using their absolute values. Values of P <0.05 were considered significant. Exponential fitting of the decay time of the synaptic currents was performed with the internal fitting routines of the MiniAnalysis program.

Solutions. The ionic composition of the slice incubation medium consisted of (in mM) 124 NaCl, 5 KCl, 1.2 KH₂PO₄, 24 NaHCO₃, 2.4 CaCl₂, 1.3 MgSO₄, and 10 glucose bubbled with 95% O₂-5% CO₂. The pH was ~7.45. The standard external recording solution consisted of (in mM) 150 NaCl, 5 KCl, 2 CaCl₂, 1 MgCl₂, 10 glucose, and 10 HEPES. Ca²⁺-free external solution consisted of (in mM) 150 NaCl, 5 KCl, 5 MgCl₂, 2 EGTA, 10 glucose, and 10 HEPES. External solutions were all adjusted to pH 7.4 with Tris base. For Na⁺-free extracellular solution, NaCl was replaced with Tris-Cl adjusted to pH 7.4. During mIPSC recordings, external solutions routinely contained 3 × 10⁷ M tetrodotoxin (TTX) to block voltage-dependent Na⁺ channels and 10⁵ M 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) and 10⁵ M DL-2-amino-5-phosphovaleric acid (APV) to block glutamatergic currents. The ionic composition of internal (patch pipette) solution for the nystatin-perforated patch recording was (in mM) 110 KCl, 40 K-methanesulfonate, and 10 HEPES, with pH adjusted to 7.2 with Tris base. Nystatin was initially dissolved in acidified methanol at 10 mg/ml. Just before use, this stock solution was diluted with the internal solution to a final concentration of 100 µg/ml.

Drugs. Drugs used in the present study were TTX, APV, bicuculline, CNQX, EGTA, bumetanide, allopregnanolone, and nystatin, all from Sigma (St. Louis, MO). Allopregnanolone, CNQX, bicuculline, and bumetanide were dissolved in dimethyl sulfoxide (DMSO) as a stock solution and then diluted with extracellular solution to the final concentration used. The final concentration of DMSO did not exceed 0.01%, which alone did not affect the frequency and amplitude of synaptic events. All solutions containing drugs were applied by the "Y-tube system" for rapid solution exchange within 20 ms (38).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

In the presence of 3 × 10⁷ M TTX, 10⁵ M CNQX, and 10⁵ M APV, spontaneous inward mIPSCs were observed in >90% (42/45) of the acutely dissociated MPOA neurons under voltage clamp whole cell recording at a V_H of 50 mV. The mIPSCs were completely and reversibly blocked by 10⁶ M bicuculline (Fig. 1A), suggesting that the mIPSCs observed were mediated by GABA_A receptor activation. External application of 10⁸ M allopregnanolone significantly increased mIPSC frequency to 158.9 ± 11.7% (n = 5) without affecting baseline current or mIPSC amplitude (116.2 ± 8.8% of control, n = 5; Fig. 1, Ba and C).

View larger version (28K): [in this window] [in a new window]   Fig. 1.   Enhancement of the frequency of -aminobutyric acid type A (GABAA)ergic miniature inhibitory postsynaptic currents (mIPSCs) by allopregnanolone. A: mIPSCs are completely blocked by 10-6 M bicuculline (bic) in the presence of 3 × 107 M tetrodotoxin, 105 M 6-cyano-7-nitroquinoxaline-2,3-dione and 105 M DL-2-amino-5-phosphovaleric acid. VH, holding potential. Ba: a typical trace of mIPSCs observed before, during, and after the exogenous application of 108 M allopregnanolone (Allo). The frequency of GABAAergic mIPSCs is reversibly increased by allopregnanolone. b: cumulative distribution of inter-event intervals (left) of allopregnanolone vs. control (K-S test), and amplitudes (right) of mIPSCs with and without 108 M allopregnanolone in the same neuron as shown in top trace (A). C: statistical summary of results from the 5 neurons responded to 108 M allopregnanolone. All frequencies and amplitudes are normalized to those of control mIPSCs. Error bar, ±SE. ** Statistically significant difference at P < 0.01.

Higher concentrations of allopregnanolone (10⁷ M) induced inward currents in the postsynaptic neurons (Fig. 2A, bottom trace), which could be blocked by 10⁵ M bicuculline (n = 3). The peak amplitude of allopregnanolone-induced postsynaptic currents increased in a concentration-dependent manner. Increase in mIPSC frequency by allopregnanolone was also concentration dependent, a significant increase of mIPSC frequency being observed at 10⁸ M. Thus the threshold concentration was between 10⁹ M and 10⁸ M (Fig. 2B). However, the mIPSC peak amplitude was not affected by allopregnanolone concentrations 10⁶ M compared with the control. These results suggest that 1) high concentrations (10⁷ M) of allopregnanolone induce bicuculline-sensitive currents in postsynaptic MPOA neurons, as has been demonstrated in the MPOA (7) and other brain areas (9, 21, 19); and 2) this neurosteroid enhanced spontaneous GABA release from nerve terminals attaching on MPOA neurons. To focus on elucidating the mechanism underlying allopregnanolone enhancement of GABA release, 10⁸ M allopregnanolone was employed in the following experiment, because allopregnanolone 10⁷ M directly induced the postsynaptic bicuculline-sensitive Cl currents.

View larger version (31K): [in this window] [in a new window]   Fig. 2.   Concentration-response relationship of allopregnanolone. A: typical traces of mIPSCs obtained from the same neuron in the absence and presence of different concentrations of allopregnanolone, as indicated above bars. B: summary of results of mIPSC frequency and amplitude from 5-6 neurons. All frequencies and amplitudes are normalized to those of control mIPSCs. Each point and error bars show the mean ± SE.

Because in general Ca²⁺ influx into the presynaptic nerve terminal triggers evoked and spontaneous neurotransmitter release from the presynaptic nerve terminal in the CNS (12) and peripheral nerve system (17), we wished to examine whether Ca²⁺ entry into the presynaptic terminal played a role during allopregnanolone enhancement of GABAergic mIPSC frequency. In Ca²⁺-free external solution with 2 mM EGTA, which decreases the mIPSC frequency to 46.8 ± 13.0% of control (n = 4, P < 005; Fig. 3, Aa and b and B), 10⁸ M allopregnanolone did not increase mIPSC frequency (93.6 ± 18.4% of control in Ca²⁺ free solution; n = 4, P > 0.1), which is compatible with the previous finding of an involvement of Ca²⁺ influx into synaptic terminal in enhancing spontaneous glutamate release by presynaptic GABA_A receptor activation in ventrohypothalamic neurons (12). Likewise, 10⁸ M allopregnanolone did not have any effect on the distribution and mean peak amplitude of the mIPSCs (Fig. 3Ac).

View larger version (23K): [in this window] [in a new window]   Fig. 3.   Effect of external Ca2+ to enhance mIPSC by allopregnanolone. Aa: traces of mIPSCs, as typically observed before, during, and after 108 M allopregnanolone application with and without Ca2+ in the external solution. Facilitation of the mIPSC frequency by allopregnanolone was blocked by removal of Ca2+ from external solution. Ab: time course of event frequency (same neuron as shown in a). No. of events in every 10-s period is summed and plotted. Ac: cumulative distribution of inter-event intervals (left) and amplitudes (right) in Ca2+-free external solution containing 2 mM EGTA. B: pooled data of 4 neurons, frequency (left) and amplitude (right), at 108 M allopregnanolone in the presence and absence of external Ca2+. Error bar, ±SE. * Statistically significant difference at P < 0.05.

Neurosteroids are known to enhance Cl conductance via GABA_A receptor at the presynaptic (43) and postsynaptic membrane (24). The effect of allopregnanolone on transmitter release might at least be dependent on Cl concentration in the presynaptic nerve terminal, since the extracellular Cl concentration was kept constant at 162 mM. Cation chloride cotransporters (CCCs) play principal roles in the regulation of intracellular Cl concentration in the neurons (15). Thus we investigated the effect of 5 × 10⁵ M bumetanide, a blocker of CCCs, on mIPSC frequency. Bumetanide itself did not affect the frequency or the mean peak amplitude of mIPSCs (112.2 ± 25.4 and 147.1 ± 44.3% of those without bumetanide, respectively; n = 4; Fig. 4A). Application of 10⁸ M allopregnanolone did not increase mIPSC frequency in the presence of bumetanide (97.7 ± 6.7% of control at 20 min after 5 × 10⁵ M bumetanide application; n = 4, P > 0.1; Fig. 4).

View larger version (27K): [in this window] [in a new window]   Fig. 4.   Effect of allopregnanolone in external 5 × 105 M bumetanide. Aa: representative recording of mIPSCs before, during, and after application of 108 M allopregnanolone in the absence and presence of 5 × 105 M bumetanide. Ab: Time course of event frequency in the top trace (a). No. of events in every 10-s period is summed and plotted. B: pooled data of 4 neurons, frequency (left) and amplitude (right), with and without bumetanide. All frequencies and amplitudes are normalized to those of control mIPSCs. Error bar, ±SE. * Statistically significant difference at P < 0.05.

Bumetanide-sensitive CCCs are known to include the K⁺-Cl cotransporter (KCC) and the Na⁺-K⁺-Cl cotransporter (NKCC). Because 5 × 10⁵ M bumetanide potently blocks NKCC function (6, 32, 41) but affects KCC less (16), the functional existence of NKCC and consequent high [Cl]_i in the presynaptic terminals might play a role in the enhancement of GABA release by allopregnanolone. NKCC accumulates Cl by using the Na⁺ gradient across the plasma membrane (32). Removal of extracellular Na⁺ immediately increased the mIPSC frequency (Fig. 5A). However, the previously observed increase of mIPSC frequency in the presence of 10⁸ M allopregnanolone did not persist in Na⁺-free solution (n = 4). In addition, repetitive application of allopregnanolone in Na⁺-free solution changed its action gradually into a decrease of the mIPSC frequency (Fig. 5B). Returning the extracellular Na⁺ concentration back to 150 mM immediately recovered mIPSC frequency to its original level (before removal of Na⁺), and the facilitatory effect of allopregnanolone on mIPSC frequency (Fig. 5) returned. No significant effect was observed on the average mIPSC peak amplitude during the Na⁺-free experiments (n = 4). Thus bumetanide-sensitive and Na⁺-dependent presynaptic mechanisms like NKCC appear essential for allopregnanolone-induced enhancement of GABA release.

View larger version (38K): [in this window] [in a new window]   Fig. 5.   Effect of allopregnanolone under the Na+-free external solution. A: continuous recording of mIPSCs affected by repetitive application of 108 M allopregnanolone in Na+-free and standard external solution. B: time course of number of events (same neuron as shown in A). Note that the allopregnanolone-induced facilitation of mIPSC frequency is gradually reduced in Na+-free external solution. In B, circles and diamonds indicate the numbers of events/10 s in the absence and presence of extracellular Na+, respectively. Open and closed symbols indicate those without and with allopregnanolone, respectively.

To further examine the action of allopregnanolone on GABAergic transmission at MPOA neurons, the kinetics of mIPSCs were analyzed with and without 10⁸ M allopregnanolone (V_H of 50 mV). The average decay time of 20 mIPSCs picked up at random was best fitted by the sum of two exponential components. Respective values for two time constants, _fast and _slow, were 18 ± 3 and 62 ± 12 ms for control (n = 6) and 30 ± 5 and 153 ± 28 ms in the presence of 10⁸ M allopregnanolone (n = 6; Fig. 6A). Significant prolongation of both values was observed in the presence of allopregnanolone, although the peak amplitudes of GABAergic mIPSCs were not affected (Fig. 6B)

View larger version (28K): [in this window] [in a new window]   Fig. 6.   Effect of allopregnanolone on decay time of mIPSCs. Aa: representative recordings of mIPSCs before, during, and after application of 108 M allopregnanolone. Ab: representative, scaled averaged mIPSC before (left, control) and during (middle) application of 108 M allopregnanolone from the same neuron as shown in a. Current decay is best fitted by 2 exponential components with time constants fast (f) and slow (s). B: average time constants (f and s) are calculated for each cell and averaged for recordings without (open bars) and with (filled bars) 108 M allopregnanolone. * Statistically significant difference at P < 0.01.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Neurosteroids have been known to change Cl conductance via actions on the GABA_A receptor (24) and bicuculline-sensitive Cl channels in the postsynaptic membrane (9, 19, 21). Consistent with these findings, the present study shows that allopregnanolone enhanced the decay of GABAergic mIPSCs without apparently affecting the peak current amplitudes (Fig. 2A), which appears similar to the action of allopregnanolone in the hippocampus (9) but not in the hypothalamus and pituitary intermediate lobe (28). Allopregnanolone does not potentiate the amplitude of GABA response elicited by high (saturated) concentration of GABA (28). Thus it is conceivable that the concentration of GABA reached in the synaptic cleft at individual release sites might be sufficient to saturate postsynaptic GABA_A receptor clusters in the MPOA in much the same way as has been demonstrated in the hippocampus (4). Regarding the effect of allopregnanolone on presynaptic function, allopregnanolone at physiological concentrations of 10-40 ng/ml (3 × 10⁸ ~ 10⁷ M) in blood and of 2-15 ng/g in the brain (34) has been shown to enhance GABA release from hypothalamic neuronal terminals (7, 28) and the still-attached afferent nerve terminals on acutely dissociated MPOA neurons (present study). Neurosteroids do activate GABA_A receptors of presynaptic nerve terminals in the pituitary gland (43). Combined together, these observations argue that neurosteroids appear to change Cl conductance of presynaptic terminals, which does seem to result in enhancement of transmitter releases.

An increasing Ca²⁺ concentration (8, 12, 17) may trigger neurotransmitter release. In the present study, allopregnanolone failed to increase the mIPSC frequency in Ca²⁺-free external solution. This result suggests that external Ca²⁺ is needed for allopregnanolone-facilitated release of GABA and appears to be in agreement with the reported triggering of GABA release from presynaptic nerve terminals in the MPOA by voltage-dependent calcium channels (VDCCs) in the active zone of synapses (8). Because of high Cl concentration at the presynaptic nerve terminal (37), activation of presynaptic Cl channels induces a Cl efflux, resulting in nerve terminal depolarization and Ca²⁺ entry into nerve terminal through the VDCCs (12).

In this study, bumetanide (Fig. 4) and removal of extracellular Na⁺ (Fig. 5) inhibited the enhancing effect of allopregnanolone on GABA release. Bumetanide is known to block CCCs such as KCC and NKCC. NKCC carries Cl into the cell, driven mainly by the inward Na⁺ force brought about by low intracellular and high extracellular Na⁺ concentrations (32). NKCC is responsible for keeping high intracellular Cl concentration of hippocampal presynaptic nerve terminals (12) and peripheral neurons (35). At the concentration of 5 × 10⁵ M employed in this study, bumetanide potently blocked NKCC (6, 32, 41) but affected KCC function less (16). Under blockade of NKCC, activation of presynaptic Cl conductance led to a gradual decrease of terminal [Cl]_i. Repetitive application of 10⁸ M allopregnanolone in the Na⁺-free condition gradually changed the action of allopregnanolone on mIPSC frequency from the facilitatory to the inhibitory (Fig. 5). The reversed effect of allopregnanolone on transmitter release in Na⁺-free extracellular solution is unclear. However, one possible explanation is that reversed NKCC function caused by a reversed Na⁺ driving force or Cl extrusion mechanism, such as KCC under the blockade of Cl intrusion by NKCC, might decrease the terminal [Cl]_i and cause terminal hyperpolarization.

At least two possible mechanisms of presynaptic action of allopregnanolone have been considered in the present study. 1) Allopregnanolone enhances GABA action on the presynaptic terminal. In our preparation, only presynaptic terminals were still attached to the MPOA neuron. Thus released GABA could act on presynaptic GABA_A receptors, which effect was enhanced by allopregnanolone. 2) Allopregnanolone acts as a direct effector on presynaptic channels (such as bicuculline-sensitive Cl channels) as reported at the postsynaptic membrane (7, 19, 28), this despite the fact that a low concentration of allopregnanolone (<10⁷M) did not activate Cl channels in the postsynaptic membrane (Fig. 2A). Here, the replacement with Na⁺-free extracellular solution immediately increased mIPSC frequency (Fig. 5). Removal of extracellular Na⁺ affected not only NKCC (15, 32, 41) but also other Na⁺-dependent mechanisms such as the Na⁺/Ca²⁺ exchanger, which also regulates transmitter release (29).

Large numbers of MPOA neurons are GABAergic (5) and form local circuits within this region (10). Progesterone is rapidly metabolized into neurosteroids in the brain (1). These facts considered, the modulation of neurosteroids such as allopregnanolone in this study on GABAergic transmission mediated by pre- and postsynaptic mechanisms is recognized to play a critical role in the GABAergic-related functions of the MPOA such as release of GnRH from the hypothalamus (20) and regulation of body temperature (14).