Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Calcium-activated K⁺ channels exert far-reaching effects on cell excitability. They regulate the shape of the action potential and thus help control neuronal firing rate and pattern (Pennefather et al., 1985; Lancaster and Nicoll, 1987; Viana et al., 1993). These channels are also expressed in a variety of vascular and visceral smooth muscle cells, and they mediate the hyperpolarization and resultant relaxation elicited by mediators such as endothelium-derived hyperpolarizing factor (Murphy and Brayden, 1995). In addition, they regulate catecholamine secretion from adrenal chromaffin cells (Nagayama et al., 1998) and gonadotropin secretion from the anterior pituitary (Tse et al., 1995).

Calcium-activated K⁺ channels can be divided into two major categories: large-conductance (BK) channels and small-conductance (SK) channels (Sah, 1996). They differ not only in their single channel conductances, but also in their sensitivity to cytosolic calcium and to blockers such as tetraethylammonium (TEA) and charybdotoxin that antagonize BK-mediated currents, and apamin, which antagonizes SK-mediated currents (Köhler et al., 1996; Sah, 1996; Hirschberg et al., 1999). BK-associated currents are involved in membrane repolarization observed during the descending phase of the action potential and in the fast component of the rebound afterhyperpolarization (AHP) observed immediately following an action potential (Pennefather et al., 1985; Lancaster and Nicoll, 1987; Viana et al., 1993; Zhang and McBain, 1995). SK-associated currents, on the other hand, are responsible for the slower components of the AHP, and they underlie spike frequency adaptation (Pedarzani and Storm, 1993; Sah, 1996; Bond et al., 1999).

SK currents are derived from three channel subtypes, namely the SK1, SK2, and SK3 channel subtypes (Köhler et al., 1996). These subtypes can be distinguished from one another on the basis of their sensitivity to the bee venom apamin. SK1-mediated currents are refractory, whereas SK2- and SK3-mediated currents are sensitive to apamin (Köhler et al., 1996; Bond et al., 1999). SK1-mediated slow AHPs and their underlying currents have been extensively characterized in hippocampal pyramidal neurons (Lancaster and Nicoll, 1987; Pedarzani and Storm, 1993; Velumian and Carlen, 1999) and in neurons of the dorsal motor nucleus of the vagus (Sah, 1993). Apamin-sensitive AHPs and associated currents, on the other hand, have a more widespread distribution. Not only have they been observed in hippocampal interneurons (Zhang and McBain, 1995) and pyramidal neurons (Stocker et al., 1999), but they have also been reported in many other areas including, but not limited to, sympathetic ganglia (Pennefather et al., 1985; Dunn, 1994) and magnocellular neurosecretory cells of the hypothalamus (Bourque and Brown, 1987).

The preoptic area (POA) of the rostral hypothalamus is an important integrative brain region involved in the control of reproductive function (Fleming et al., 1994; Numan and Sheehan, 1997). The POA contains a large proportion of the centrally localized somata of gonadotropin-releasing hormone (GnRH) neurosecretory cells (Silverman et al., 1979). These cells drive the hypothalamic-pituitary-gonadal axis by releasing GnRH into the hypophysial portal vasculature to stimulate the secretion of gonadotropins [i.e., follicle-stimulating hormone, luteinizing hormone (LH)] from the anterior pituitary (Ferin et al., 1984). GnRH neurons are the focal point of gonadal steroid (i.e., estrogen) feedback on the reproductive axis (Ferin et al., 1984). GnRH neurons historically have been thought not to contain the classical estrogen receptor (Herbison, 1997a). However, several recent reports indicate that GnRH neurons express the transcript and protein for both the - and -isoforms of the estrogen receptor (Butler et al., 1999; Skynner et al., 1999; Hrabovszky et al., 2000), and estrogen directly inhibits these neurons via a membrane hyperpolarization (Kelly and Wagner, 1999). Other mechanisms of estrogen-induced feedback involve the modulation of both inhibitory [e.g., -aminobutyric acid (GABA)ergic] and stimulatory (e.g., noradrenergic) inputs to GnRH neurons (Condon et al., 1989; Herbison et al., 1991; Simonian et al., 1999). Monoamine neurotransmitters inhibit the slow AHP in hippocampal pyramidal neurons, thereby increasing cell excitability (Lancaster and Nicoll, 1987; Pedarzani and Storm, 1993; Pedarzani and Storm, 1996). The majority of POA neurons express an AHP (Wagner et al., 2000), and it is plausible that noradrenergic input to the POA increases neuronal excitability by decreasing the AHP in a manner dependent on the gonadal steroid milieu. While noradrenergic inputs terminate in close proximity to GnRH neurons, they also directly influence the neuronal activity of POA GABAergic neurons, and in doing so they can indirectly affect GnRH output (for review, see Herbison, 1997b).

In the present study we characterized the current underlying the AHP in POA neurons and tested the hypothesis that norepinephrine attenuates this current in a steroid-dependent fashion. To this end, sharp electrode and whole-cell patch recordings were performed in hypothalamic slices prepared from ovariectomized, estrogen-, or vehicle-treated female guinea pigs. Isoproterenol, methoxamine, timolol, and prazosin were used to evaluate the effects of - and, for comparison, ₁-adrenergic receptor activation and blockade (Emilien and Maloteaux, 1998; Docherty, 1998; Frishman and Kotob, 1999), respectively. Post hoc identification of neuronal phenotype was accomplished by combined histofluorescence and in situ hybridization for glutamic acid decarboxylase (GAD)₆₅. The results reveal that norepinephrine negatively modulates an apamin-sensitive SK current in POA neurons, including GABAergic neurons, via both ₁- and -adrenergic receptor-mediated mechanisms. Furthermore, estrogen rapidly augments the ₁-adrenergic receptor-mediated inhibition of the SK current. This novel finding suggests a mechanism by which noradrenergic input to the POA promotes estrogen-induced feedback of the reproductive axis.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals and Treatments. Female Topeka guinea pigs (470-660 g) were obtained from our institutional breeding facility and maintained under constant temperature (72.4 ± 0.1°F) and light (on from 6:30 AM-8:30 PM). Animals were housed individually, with food and water provided ad libitum. They were ovariectomized under ketamine/xylazine anesthesia (33 mg/kg and 6 mg/kg, respectively, s.c.) 5 to 9 days prior to experimentation and given either estradiol benzoate (EB; 25 µg, s.c.) or its sesame oil vehicle (0.1 ml, s.c.) 24 h prior to experimentation. Serum estrogen concentrations were determined by radioimmunoassay from trunk blood collected on the day of experimentation. This treatment regimen produced physiological levels of 17-estradiol (vehicle: <10 pg/ml; 24-h EB: 243 ± 26 pg/ml; n = 6-31). All animal procedures described in this study are in accordance with institutional guidelines based on National Institutes of Health standards.

Drugs. All drugs were purchased from Sigma Chemical Co. (St. Louis, MO) unless otherwise specified. EB was dissolved in sesame oil to a concentration 250 µg/ml. Tetrodotoxin (TTX) was dissolved in Milli-Q H₂O and further diluted with 0.1% acetic acid (final concentration, 1 mM; pH 4-5). TEA chloride was dissolved in Milli-Q H₂O to a stock concentration of 1 M. Cadmium chloride (CdCl₂) was dissolved in Milli-Q H₂O to a stock concentration of 0.1 M. The peptide apamin (C₇₉H₁₃₁N₃₁O₂₄S₄) was dissolved in 0.05 M acetic acid to a stock concentration of 0.5 mM. 1,1'-(1,10-Decanediyl)bis(4-amino-2-methylquinolinium) chloride (dequalinium dichloride; Tocris Cookson Inc., Ballwin, MO) was dissolved in 70% dimethyl sulfoxide to a stock concentration of 4.3 mM. 1-[3',4'-Dihydroxyphenyl]-2-isopropylaminoethanol hydrochloride (isoproterenol HCl) and (S)- 1-[(1,1-dimethylethyl)amino]-3-[[4-(4-morpholinyl)-1,2,5-thiadiazol-3-yl]oxy]-2-propanol [(S)-timolol maleate; Tocris Cookson Inc.] were dissolved in Milli-Q H₂O to a stock concentration of 10 mM. 2-Amino-1-(2,5-dimethoxyphenyl)-propan-1-ol hydrochloride (methoxamine HCl; Burroughs-Wellcome, Research Triangle Park, NC) was dissolved in Milli-Q H₂O to a stock concentration of 1 mM. 1-(4-Amino-6,7-dimethoxy-2-quinazolinyl)-4-(2-furanylcarbonyl)piperazine (prazosin hydrochloride; Pfizer, Groton, CT) was dissolved in 100% dimethyl sulfoxide to a stock concentration of 1 mM. 17-estradiol (E₂) was purchased from Steraloids (Wilton, NH), recrystallized to ensure purity, and dissolved in 100% ethanol to a stock concentration of 1 mM. Aliquots of the stock solutions were stored as appropriate until needed.

Tissue Preparation. On the day of experimentation, the animal was decapitated, its brain removed from the skull, and the hypothalamus dissected. The resultant hypothalamic block was mounted on a plastic cutting platform that was then secured in a vibratome well filled with ice-cold, oxygenated (95% O₂, 5% CO₂) artificial cerebrospinal fluid (aCSF, in mM: NaCl, 124; NaHCO₃, 26; dextrose, 10; HEPES, 10; KCl, 5; NaH₂PO₄, 2.6; MgSO₄, 2; CaCl₂, 1). Four coronal slices (350 µM) through the POA were then cut. The slices were transferred to a multi-well auxiliary chamber containing oxygenated aCSF and kept there until electrophysiological recording.

Electrophysiology. Sharp electrode recordings in current clamp and whole-cell patch recordings in voltage clamp were performed as previously described (Wagner et al., 2000). Briefly, slices were maintained in a chamber perfused with warmed (35°C), oxygenated aCSF containing the same constituents and respective concentrations, except for CaCl₂, which was raised to 2 mM. Artificial CSF and all drug solutions were perfused via a peristaltic pump at a rate of 1.5 ml/min. Drug solutions were prepared in 20-ml syringes by diluting the appropriate stock solution with aCSF, and the flow was controlled via a three-way stopcock. Microelectrodes (100-225 M) were assembled from borosilicate glass pipettes (Sutter Instrument Co., Novato, CA; 1.2-mm o.d.) pulled on a P-97 Flaming Brown puller (Sutter Instrument Co.) and filled with a 3% biocytin solution in 1.75 M KCl and 0.025 M Tris (pH 7.4).

Post hoc Identification of POA GABAergic Neurons. Sometimes following electrophysiological recording, slices were fixed in 4% paraformaldehyde in 0.03 M Sorensen's phosphate buffer (90-180 min, 4°C) and then soaked overnight in buffer containing 20% sucrose. All solutions were prepared with diethyl pyrocarbonate-treated Milli-Q H₂O and molecular grade reagents. Frozen slices were sectioned at 20 µm on a cryostat (Leitz model 1720 Digital Cryostat, Wetzlar, Germany), mounted on Superfrost-plus slides, and then washed (5 min) with 0.1 M phosphate buffer. Streptavidin-CY3 (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA), diluted with seaweed gelatin solution in the presence of a RNase inhibitor (RNAsin; 60 units/ml) and sodium heparin (1.25 mg/ml), was then applied (2 h). The reaction was terminated by washing with 0.1 M phosphate buffer. Biocytin-filled, GABAergic neurons were then identified by in situ hybridization using a guinea pig GAD₆₅ riboprobe as described previously (Wagner et al., 2001). Briefly, slides were postfixed in fresh 4% paraformaldehyde (40 min), rinsed with Sorensen's phosphate buffer, and treated with Proteinase-K (1.0 µg/ml, 2 min, 37°C). All sections were then treated (3 min) with 0.1 M triethanolamine, followed by 0.25% acetic anhydride in 0.1 M triethanolamine (10 min). Thereafter, the sections were rinsed in 2× SSC and hybridized (56-58°C, 18 h). Sections were rinsed in 2× SSC (30 min) on a shaker, reacted with RNase (20 µg/ml, 30 min, 37°C), and sequentially rinsed in 1×, 0.5×, and 0.25× SSC (55°C). Slides were finally washed (30 min, 65°C) in 0.1 × SSC containing 1.0 mM dithiothreitol. The sections were dehydrated in increasing concentrations of ethanol, and together with autoradiographic [¹⁴C]microscales (Amersham Pharmacia Biotech, Arlington Heights, IL) were exposed to Hyperfilm-_max X-ray film (PerkinElmer Life Science Products, Boston, MA) for 5 to 6 days at 4°C. Slides were then dipped in Kodak nuclear track -2 emulsion and exposed for up to 16 days at 4°C (Kodak Scientific Imaging Systems, Rochester, NY). Sections were evaluated and photographed under dark-field illumination using a Zeiss microscope configured with a dark light attachment (Foster, Inc.).

Statistical Analyses. Comparison between two groups were performed using either the Student's two-tailed t test, paired t test, or the Mann-Whitney U test. The homogeneity of variance was evaluated using Cochran's C test. Comparisons between two or more groups were performed using either a one-way or a multifactorial analysis of variance followed by the least significant difference test. Differences were considered statistically significant if the probability of error was less than 5%.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Figure 1A shows a typical example of spike frequency adaptation from a sharp electrode recording of a POA neuron. A 100-pA depolarizing current pulse, 1 s in duration, elicits an initial cluster of action potentials, with each one exhibiting an AHP at its tail end. Over time, the interspike interval between action potentials increases, thereby decreasing the firing rate during the latter phase of the pulse. An AHP is also observed on the off-step of the current pulse. In the presence of 1 µM TTX, spike generation is eliminated, but the AHP associated with the current pulse off-step is still present (Fig. 1B).

View larger version (12K): [in this window] [in a new window]   Fig. 1.   A, sharp electrode recording showing an example of spike frequency adaptation in a POA neuron from an ovariectomized, vehicle-treated guinea pig. Action potentials (not fully replicated by the chart recorder) were generated by a depolarizing, 100-pA current pulse (1 s in duration) delivered at rest (51 mV). AHPs were observed at the tail end of each action potential, and on the off-step of the current pulse. B, another POA neuron recorded at rest (55 mV) via sharp electrode in the presence of 1 µM TTX. Under these conditions, the same depolarizing current pulse does not elicit action potentials, but the AHP observed on the off-step of the pulse still remains.

We sought to characterize the current underlying the AHP in POA neurons from vehicle-treated animals using whole-cell patch recordings. Outward tail currents were evoked immediately following a 100-mV depolarizing voltage command (50-150 ms). At a holding potential of 50 mV, which closely approximates the mean resting potential of POA neurons (Wagner et al., 2000), these currents exhibited a latency-to-peak and a _inactivation of 26.3 ± 8.3 and 80.7 ± 20.6 ms, respectively (n = 47). While varying the EGTA concentration of the internal solution significantly altered the resting membrane potential (1 mM EGTA: 50.7 ± 2.1 mV, n = 27; 11 mM EGTA: 65.2 ± 2.5 mV, n = 12, p < 0.05), it did not affect peak tail current magnitude (1 mM EGTA: 36.7 ± 4.8 pA, n = 32; 11 mM EGTA: 43.5 ± 8.8 pA, n = 18). On the other hand, the nonselective Ca²⁺ channel blocker CdCl₂ (200 µM) completely abolished the tail current (Fig. 2). In addition, both apamin (300 nM-1 µM) and dequalinium (3 µM) produced an appreciable diminution of the peak current (Fig. 3). Collectively, these data indicate that functional SK2 and/or SK3 channel subtypes are expressed in POA neurons.

View larger version (8K): [in this window] [in a new window]   Fig. 2.   A, whole-cell patch recording from a POA neuron that illustrates the outward tail current underlying the AHP (IAHP) obtained prior to (a) and in the presence of (b) 200 µM CdCl2. B, a composite bar graph that illustrates the abolition of the IAHP in POA neurons by 200 µM CdCl2 (Cd2+). Columns represent means, and vertical bars represent 1 S.E.M. (n = 4) of the baseline IAHP (open column) and that attained during the bath application of 200 µM Cd2+ (filled column). *, values of IAHP obtained in the presence of Cd2+ that are significantly different (Mann-Whitney U test; p < 0.05) from the baseline control values.

View larger version (22K): [in this window] [in a new window]   Fig. 3.   Examples of the IAHP obtained in POA neurons prior to (a) and in the presence of (b) apamin (1 µM; A) and dequalinium (3 µM; B). The cell in A is a GAD-positive neuron. C, a composite bar graph that shows the decrement in the IAHP caused by apamin (300 nM and 1 µM) and dequalinium (3 µM). Columns represent means, and vertical bars represent 1 S.E.M. (n = 4-7) of the percentage of the baseline control IAHP observed during the bath application of 300 nM apamin (open column), 1 µM apamin (solid column), or 3 µM dequalinium (gray column). Baseline IAHP control values for 300 nM and 1 µM apamin and for 3 µM dequalinium were 52.0 ± 16.4, 26.2 ± 7.4, and 30.4 ± 8.0 pA, respectively. *, significant decrease of the IAHP (paired t test; p < 0.05).

We next examined the potential for noradrenergic modulation of the outward tail current in POA neurons. As shown in Fig. 4, the -adrenergic agonist isoproterenol (1 µM) decreased the peak current, which was restored to its original magnitude upon clearance of the drug from the slice. The inhibitory effect of isoproterenol was completely blocked by the -adrenergic receptor antagonist timolol (3 µM). Likewise, bath application of the ₁-adrenergic receptor agonist methoxamine (10 µM) produced a reversible decrease of the peak current, which was significantly attenuated by the ₁-adrenergic antagonist prazosin (10 µM; Fig. 5). The inhibitory effect of these adrenergic receptor agonists was also dose-dependent. The dose-response curves in Fig. 6 show that methoxamine decreased peak current amplitude with an estimated I_max of 46% and an IC₅₀ of 4.4 µM. Isoproterenol had a comparable I_max of 60% and an IC₅₀ of 132.6 nM (Fig. 7).

View larger version (15K): [in this window] [in a new window]   Fig. 4.   A, examples of the IAHP in a POA neuron obtained prior to (a), in the presence of (b), and following recovery from (c) 1 µM -adrenergic receptor agonist isoproterenol. B, examples of the IAHP in another POA neuron procured in the presence of 3 µM the -adrenergic receptor antagonist timolol (a), and in the combined presence of 3 µM timolol and 1 µM isoproterenol (b). C, a composite bar graph that illustrates the blockade by timolol of the isoproterenol-induced decrease in the IAHP. Columns represent means, and vertical bars represent 1 S.E.M. (n = 5) of the change in the IAHP caused by the bath application of 1 µM isoproterenol alone (open column) or in combination with 3 µM timolol (filled column). Baseline control IAHP values for 1 µM isoproterenol and isoproterenol plus 3 µM timolol were 48.4 ± 15.4 and 34.0 ± 14.2 pA, respectively. *, significant decrease of the IAHP (paired t test; p < 0.05).

View larger version (15K): [in this window] [in a new window]   Fig. 5.   A, examples of the IAHP from a POA neuron that were obtained prior to (a), in the presence of (b), and following recovery from (c) 10 µM 1-adrenergic receptor agonist methoxamine. B, examples of the IAHP from another POA neuron that were attained prior to (a) and in the combined presence of (b) 10 µM methoxamine and 10 µM 1-adrenergic receptor antagonist prazosin. C, a composite bar graph depicting the attenuation by prazosin of the methoxamine-induced decrease in the IAHP. Columns represent means, and vertical bars represent 1 S.E.M. (n = 4-5) of the change in the IAHP due to bath application of 10 µM methoxamine alone (open column) or in combination with 10 µM prazosin (filled column). Baseline control values for 10 µM methoxamine and methaxamine plus 10 µM prazosin were 53.0 ± 16.0 and 57.6 ± 11.6 pA, respectively. *, significant decrease of the IAHP (paired t test; p < 0.05).

View larger version (18K): [in this window] [in a new window]   Fig. 6.   Examples of the IAHP that were obtained before (a) and during (b) the bath application of 10 µM methoxamine in POA neurons from vehicle- (A) and EB-treated (25 µg, 24 h prior; B) animals. C, composite dose-response curves for methoxamine that were derived from POA neurons in vehicle- () and EB-treated () animals. The curves were fit via a logistic equation (see Materials and Methods) to the experimental data points. Symbols represent means, and vertical bars represent 2 S.E.M. (n = 3-8) of the percent decrease in the IAHP (normalized to the baseline control value) elicited by various concentrations of methoxamine (0.3-30 µM). *, significant main effect of EB on the methoxamine-induced inhibition of the IAHP (multifactorial analysis of variance/least significant difference; p < 0.05).

View larger version (18K): [in this window] [in a new window]   Fig. 7.   Examples of the IAHP procured before (a) and during (b) the bath application of 300 nM isoproterenol in GAD-positive POA neurons from vehicle- (A) and EB-treated (B) animals. C, composite dose-response curves for isoproterenol generated from POA neurons in vehicle- () and EB-treated () animals. The curves were fit by a logistic equation (see Materials and Methods) to the experimental data points. Symbols represent means, and vertical bars represent 2 S.E.M. (n = 3-5) of the percent decrease in the IAHP (normalized to the baseline control value) caused by various concentrations of isoproterenol (0.1-1 µM).

We then investigated the ability of estrogen to modulate the ₁- and -adrenergic receptor-mediated inhibition of the outward tail current. EB-treatment of ovariectomized animals (25 µg, 24 h prior) had no effect per se on the resting membrane potential, latency-to-peak, _inactivation, peak current, or AHP amplitude (data not shown). It did, however, markedly increase the responsiveness of POA neurons to methoxaminenearly doubling the estimated I_max (85%) and apparently reducing the estimated IC₅₀ by more than 50% (2.1 µM; Fig. 6). By contrast, EB affected neither the estimated I_max nor the IC₅₀ of the isoproterenol-induced decrease in peak current amplitude (Fig. 7). It is known that estrogen modifies synaptic transmission and thereby stimulus-secretion coupling not only through gene transcription, but also via much more rapid mechanisms (Kelly and Wagner, 1999). We endeavored to gain an appreciation for how rapid the onset of steroid action was. Therefore, we bath-applied E₂ (100 nM) for 15 to 20 min to slices from vehicle-treated animals just before testing with a submaximal concentration of methoxamine (3 µM). As shown in Fig. 8, short-term E₂ treatment, just like the longer-term EB-treatment paradigm, greatly augmented the methoxamine-induced inhibition of peak current amplitude.

View larger version (13K): [in this window] [in a new window]   Fig. 8.   A, IAHP values obtained in a POA neuron from a vehicle-treated animal prior to (a) and in the presence of (b) a submaximal concentration of methoxamine (3 µM). B, IAHP values procured prior to (a) and in the presence of (b) 3 µM methoxamine from a GAD-positive POA neuron (shown in Fig. 9, C and D) given a brief (20-min) exposure to E2 (100 nM). C, a composite bar graph that illustrates the potentiation of the methoxamine-induced decrease in the IAHP observed at 3 µM caused by the bath application of E2 (100 nM; 15-20 min). Columns represent means, and vertical bars represent 1 S.E.M. (n = 4-5) of the percent decrease in the IAHP due to the bath application of methoxamine (3 µM) in POA neurons with no intervening steroid treatment (open column), and in POA neurons given a short-term exposure to E2 (filled column). The baseline IAHP control values for 3 µM methoxamine, alone and with an intervening steroid exposure just prior to testing with the agonist, were 30.6 ± 5.8 and 55.8 ± 11.8 pA, respectively. *, decreases in the IAHP caused by 3 µM methoxamine in POA neurons with prior E2 exposure that were significantly greater (Student's t test; p < 0.05) than those without any steroid exposure.

We did a post hoc evaluation of the GABAergic phenotype using combined histofluorescence and in situ hybridization for GAD₆₅ on 16 of the 90 POA neurons included into the present study. Of these 16 neurons, 12 (75%) were GAD-positive, reflecting the predominance of GABAergic neurons in the POA (Herbison, 1997a). Examples of GAD-positive POA neurons are shown in Fig. 9. These GAD-positive POA neurons displayed a tail current that was subject to inhibition by apamin, isoproterenol, and methoxamine, the latter of which was also observed in GAD-negative POA neurons (not shown). In addition, the methoxamine-induced inhibition of peak current amplitude in GAD-positive cells was rapidly and markedly potentiated by estrogen (Fig. 8).

View larger version (30K): [in this window] [in a new window]   Fig. 9.   A, photomicrograph of the biocytin-streptavidin-CY3 fluorescent labeling of a POA neuron whose IAHP was inhibited by methoxamine. B, an overlay of the fluorescent labeling in A and the hybridization signal that clearly illustrates the double-labeling for GAD65. C, photomicrograph of the biocytin-streptavidin-CY3 fluorescent labeling of the POA neuron shown in Fig. 8B. D, an overlay of the fluorescent labeling in C and the hybridization signal illustrating the double-labeling for GAD65.

What effect does the adrenergic receptor-mediated inhibition of the outward tail current have on the firing rate of POA neurons? Fig. 10 shows segments of spontaneous firing in a sharp electrode recording from a POA neuron just prior to and in the presence of various concentrations of methoxamine. Under baseline conditions, this POA neuron fires action potentials at a rate of 3.1 Hz. Bath application of either 10 or 30 µM methoxamine increased the number of action potentials obtained over an equivalent period, resulting in an increase in neuronal firing rate. Taken together, these results demonstrate that POA neurons, including GABAergic neurons, express an apamin-sensitive, SK current that underlies the I_AHP. This current is inhibited by both ₁- and -adrenergic receptor agonists, resulting in increased neuronal excitability. Furthermore, estrogen selectively enhances the ₁-receptor-mediated inhibition of the I_AHP.

View larger version (28K): [in this window] [in a new window]   Fig. 10.   Excerpts of spontaneous firing from a sharp electrode recording of a POA neuron (resting membrane potential = 53 mV) obtained in the absence (A) and presence of 10 µM (B) and 30 µM (C) methoxamine. The segments shown in A, B, and C are identical in duration. D, bar graph exemplifying the stimulatory effect of methoxamine on the firing rate of this POA neuron. In each instance, the firing rate was calculated as time necessary to digitally capture 40 action potentials. Under baseline conditions, the 40 action potentials were collected over a period of 12.3 s, resulting in a baseline firing rate of 3.2 Hz. Bath application of 10 µM methoxamine nearly doubled the firing rate to 5.7 Hz. In the presence of 30 µM methoxamine, the firing rate was increased slightly further to 7.1 Hz.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In the present study, we have shown that POA GABAergic neurons exhibit an apamin-sensitive, SK-mediated I_AHP that regulates spike frequency and is negatively modulated upon ₁- and -adrenergic receptor activation. Moreover, the ₁-adrenergic receptor-mediated inhibition of the I_AHP is augmented by estrogen in a rapid and sustained manner. These conclusions are based on the observations that 1) POA GABAergic neurons exhibit an AHP and spike frequency adaptation in response to a depolarizing stimulus, 2) they express the corresponding I_AHP that is sensitive to antagonism by CdCl₂, apamin, and dequalinium, 3) this I_AHP is reduced by ₁- and -adrenergic receptor agonists, resulting in an increased neuronal firing rate, and 4) both short-term (15-20 min) E₂ treatment and longer-term (24 h prior) EB treatment elicit a pronounced augmentation of the ₁-adrenergic receptor-mediated inhibition of the I_AHP.

Consistent with what we have shown previously, POA neurons express an AHP (Wagner et al., 2000). In the present study, we found that the current underlying this AHP is insensitive to the EGTA concentration within the internal solution. This conflicts with a number of reports describing an inhibition of the I_AHP by internally applied Ca²⁺ chelators (Lancaster and Nicoll, 1987; Viana et al., 1993; Velumian and Carlen, 1999). However, the I_AHP in POA neurons was entirely eliminated by the bath application of CdCl₂, which indicates that this current is dependent on the initial influx of extracellular Ca²⁺. That EGTA was ineffective in blocking the I_AHP suggests a close proximity between the Ca²⁺ channels and the Ca²⁺-activated K⁺ channels on the somata of POA neurons. A similar juxtaposition of these channels has also been proposed to exist in hippocampal pyramidal neurons (Hirschberg et al., 1999).

Currents underlying the AHP like those extensively studied in native hippocampal pyramidal neurons and in the dorsal motor nucleus of the vagus exhibit a slow rise time and a _inactivation greater than 1 s (Lancaster and Nicoll, 1987; Sah, 1993; Sah, 1996). This type of I_AHP mediates the slow AHP responsible for spike frequency adaptation, is apamin-insensitive, and is believed to arise from the SK1 channel subtype (Pedarzani and Storm, 1993; Sah, 1996; Hirschberg et al., 1999). On the other hand, SK2 and SK3 channel subtypes mediate the apamin-sensitive I_AHP (Köhler et al., 1996; Bond et al., 1999), which underlies the medium AHP observed in several different brain regions (Viana et al., 1993; Sah, 1996; Velumian and Carlen, 1999) and which displays a comparatively faster kinetic profile (Pennefather et al., 1985; Sah, 1993; Viana et al., 1993; Zhang and McBain, 1995; Stocker et al., 1999). These two currents have been shown to coexist in hippocampal pyramidal neurons and in neurons of the dorsal motor nucleus of the vagus (Sah, 1993; Stocker et al., 1999). The I_AHP observed presently in POA neurons, including GABAergic neurons, is sensitive to antagonism by apamin and the nonpeptide blocker of the apamin-sensitive I_AHP, dequalinium (Dunn, 1994). Moreover, GnRH and also dopamine neurosecretory cells in the POA express the SK3 channel subtype as revealed by in situ hybridization (unpublished findings). On the other hand, we observed a residual, apamin-insensitive component of the I_AHP. This may reflect the small yet discernible level of expression of the SK1 subtype compared with the SK3 subtype in the POA as measured by ribonuclease protection assay (unpublished findings). These observations, coupled with the relatively transient latency-to-peak and _inactivation (26.3 and 80.7 ms, respectively) that we measured, indicate that the current underlying the AHP in POA neurons is composed primarily of a medium I_AHP. Moreover, it would appear that apamin-sensitive channel subtypes play a role in the spike frequency adaptation observed in POA neurons.

Noradrenergic input to hippocampal pyramidal neurons serves to inhibit the slow AHP and thus spike frequency adaptation, thereby promoting cell excitability (Lancaster and Nicoll, 1987; Pedarzani and Storm, 1993; Pedarzani and Storm, 1996). This inhibitory action of norepinephrine involves the stimulation of ₁-adrenergic receptors and the protein kinase A signal transduction pathway (Lancaster and Nicoll, 1987; Pedarzani and Storm, 1993). -Adrenergic receptors do not appear to be involved, although a synergistic interaction between the two receptor systems has been implicated (Pedarzani and Storm, 1996). Presently, we have found that activation of both - and ₁-adrenergic receptors results in a suppression of the medium I_AHP in POA neurons. This is the first description of an ₁-adrenergic receptor-mediated inhibition of an I_AHP, which ultimately leads to an increase in the firing rate of POA neurons. Likewise, ₁-adrenergic receptor agonists can induce phasic burst firing in neurons from the hypothalamic arcuate nucleus, including GnRH neurons (Condon et al., 1989). Our finding is also consistent with the observed inhibition of a linear, voltage-independent K⁺ conductance arising from activation of ₁-adrenergic receptors in midbrain dopamine neurons (Grenhoff et al., 1995).

In the present study, we have discovered that estrogen produces a rapid and prolonged increase in the sensitivity of POA neurons to the ₁-adrenergic receptor-mediated inhibition of the medium I_AHP. We have shown previously that estrogen can rapidly disrupt the coupling of GABA_B receptors to their effector K⁺ channels in neurons from the hypothalamic arcuate nucleus (Kelly and Wagner, 1999). Similarly, it is well established that estrogen can rapidly uncouple µ-opioid receptors from inwardly rectifying K⁺ channels in -endorphin neurons through a protein kinase A pathway (for review, see Kelly and Wagner, 1999). Through a G-protein (G_q/11) linkage, activation of ₁-adrenergic receptors results in the production by phospholipase C of inositol 1,4,5-trisphosphate and diacylglycerol, the latter of which then stimulates protein kinase C (Docherty, 1998; García-Sáinz et al., 2000). In addition, estrogen has been shown to up-regulate ₁-adrenergic receptors (Petitti et al., 1992) and to stimulate protein kinase C catalytic activity (Ansonoff and Etgen, 1998) in the POA. It is therefore possible that estrogen synergistically interacts with the ₁-adrenergic receptor and the protein kinase C pathway to rapidly potentiate the inhibition of the medium I_AHP in POA neurons. This may account for the observed increase in the efficacy of the response. It follows that estrogen would greatly augment the increase in firing rate produced by ₁-adrenergic receptor agonists, similar to the potentiation of phasic burst firing caused by methoxamine in hypothalamic arcuate neurons (Condon et al., 1989). Future studies will be conducted to determine whether this is the case.

Noradrenergic neurons emanating from the brainstem A2 cell group are estrogen-sensitive and provide a prominent projection to the POA (Simonian et al., 1999), which contains the majority of GnRH neurons (Silverman et al., 1979). In ovariectomized animal models, estrogen rapidly exerts negative feedback on the reproductive axis, manifest by a suppression of LH secretion. In the guinea pig, negative feedback induced by systemic injection of the steroid lasts up to 40 h, followed by positive feedback and the characteristic LH surge (Wagner et al., 2001). The former arises from a direct inhibition of GnRH neurons by estrogen (Kelly and Wagner, 1999). This is in conjunction with a steroid-induced increase in the release of inhibitory neurotransmitters such as GABA and -endorphin (Herbison et al., 1991; Kelly and Wagner, 1999). The augmented transmitter release is initiated by estrogen's uncoupling the µ-opioid receptor and the GABA_B receptor from their effector K⁺ channels in -endorphin neurons and in POA GABAergic neurons, respectively, and is maintained for at least 24 h (Kelly and Wagner, 1999; Wagner et al., 2001). Furthermore, noradrenergic neurons also synapse on GABAergic neurons in the POA (for review, see Herbison, 1997a,b), and presently we show that estrogen enhances the ₁-adrenergic receptor-mediated inhibition of the medium I_AHP in these GABAergic cells. This suggests that ascending noradrenergic input helps drive the GABAergic inhibition of GnRH neurons, and in doing so would participate in the coordinated, steroid-induced negative feedback of the reproductive axis.

These multiple mechanisms of estrogen-induced negative feedback combine to dramatically reduce impulse flow in GnRH neurons. It would appear that under conditions of negative feedback, this predominant inhibitory tone overwhelms the steroid's rapid potentiation of the ₁-adrenergic receptor-mediated reduction of the medium I_AHP in GnRH neurons. At the time of the LH surge, however, the inhibitory GABAergic tone is decreased and the release of the amine in the POA is increased (Herbison, 1997b; Wagner et al., 2001). Hence, around the time of the preovulatory LH surge, the conditions are favorable for a relatively unabated, noradrenergic stimulation of GnRH neurons via the ₁-adrenergic receptor-mediated inhibition of the medium I_AHP.

In conclusion, we have shown that stimulation of - and ₁-adrenergic receptors inhibits the medium I_AHP in POA neurons and increases their excitability. Moreover, the ₁-adrenergic receptor-mediated suppression of the medium I_AHP is positively modulated by estrogen. These results provide new insight into the mechanism by which noradrenergic input to POA neurons helps govern GnRH output over the course of the reproductive cycle.