- Email: [email protected]
Estrogen modulation of K+ channel activity in hypothalamic neurons involved in the control of the reproductive axis
Steroids 67 (2002) 447– 456
Estrogen modulation of K⫹ channel activity in hypothalamic neurons involved in the control of the reproductive axis Martin J. Kelly*, Oline K. Rønnekleiv, Nurhadi Ibrahim, Andre H. Lagrange, Edward J. Wagner Department of Physiology & Pharmacology, Oregon Health & Science University, Portland, OR, USA
Abstract Here we report on the progress we have made in elucidating the mechanisms through which estrogen alters synaptic responses in hypothalamic neurons. We examined the modulation by estrogen of the coupling of various receptor systems to inwardly rectifying and small conductance, Ca2⫹-activated K⫹ (SK) channels. We used intracellular sharp-electrode and whole-cell recordings in hypothalamic slices from ovariectomized female guinea pigs. Estrogen rapidly uncouples -opioid receptors from G protein-gated inwardly rectifying K⫹ (GIRK) channels in -endorphin neurons, manifest by a reduction in the potency of -opioid receptor agonists to hyperpolarize these cells. This effect is blocked by inhibitors of protein kinase A and protein kinase C. Estrogen also uncouples ␥-aminobutyric acid (GABA)B receptors from the same population of GIRK channels coupled to -opioid receptors. At 24 h after steroid administration, the GABAB/GIRK channel uncoupling observed in GABAergic neurons of the preoptic area (POA) is associated with reduced agonist efficacy. Conversely, estrogen enhances the efficacy of ␣1-adrenergic receptor agonists to inhibit apamin-sensitive SK currents in these POA GABAergic neurons, and does so in both a rapid and sustained fashion. Finally, we observed a direct, steroid-induced hyperpolarization of both arcuate and POA neurons, among which gonadotropin-releasing hormone (GnRH) neurons are particularly sensitive. These findings indicate a richly complex yet coordinated steroid modulation of K⫹ channel activity that serves to control the excitability of hypothalamic neurons involved in regulating the reproductive axis. © 2002 Elsevier Science Inc. All rights reserved. Keywords: Estrogen; -Endorphin; GABA; Norepinephrine; K⫹ Channel; Protein kinase
1. Introduction It is becoming increasingly evident that the gonadal steroid hormone estrogen imparts a multifaceted influence over synaptic transmission in the mammalian central nervous system. Not only can estrogen alter synaptic responses via genomic mechanisms, but there is a wealth of information that indicates the steroid can also modulate cell-to-cell communication much more rapidly (for review see [1]). These synaptic alterations are brought about via changes in the cellular responsiveness to the activation of various receptor systems (both G proteincoupled and ionotropic) to their respective first messengers. For example, estrogen can modulate the cellular responsiveness to ionophoric glutamate (both N-methyl-D-aspartate (NMDA) and non-NMDA) receptor activation [2,3]. In addition, it can alter the linkage of G protein-coupled receptor systems such as opioid (both and ), ␥-aminobutyric acid * Corresponding author. Tel.: ⫹1-503-494-5833; fax: ⫹1-503-4944352. E-mail address: [email protected] (M.J. Kelly).
(GABA)B and dopamine D2 receptors to their respective effector systems [4 – 8]. Furthermore, it now appears that the steroid can function as a first messenger by activating an estrogen receptor that couples directly to K⫹ and Ca2⫹ channels by way of a pertussis toxin-sensitive G protein [9,10]. These fundamentally distinct signalling pathways give rise to a coordinated regulation by estrogen of complex physiological processes such as reproduction, fine motor control and memory. Given estrogen’s preeminent role in regulating reproductive physiology and behavior, we will focus on the estrogenic modulation of K⫹ channel activity as it pertains to the negative feedback control of the hypothalamic–pituitary– gonadal axis.
2. Estrogen modulates the linkage between G protein-coupled receptors and their effector potassium channels The principal action of estrogen is to regulate the output of gonadotropin-releasing hormone (GnRH) from the mediobasal hypothalamus. Although we found direct
0039-128X/02/$ – see front matter © 2002 Elsevier Science Inc. All rights reserved. PII: S 0 0 3 9 - 1 2 8 X ( 0 1 ) 0 0 1 8 1 - 7
448
M.J. Kelly et al. / Steroids 67 (2002) 447– 456
actions of estrogen to inhibit GnRH neuronal activity over 15 years ago [9,11], it has been only recently that estrogen receptors have been demonstrated in GnRH neurons [12–14]. Therefore, little attention has been paid to the direct effect of estrogen on GnRH neurons. Rather, estrogen responsiveness has been conferred to neurons that synapse on GnRH cells [15–19]. Indeed, hypothalamic -endorphin and GABAergic neurons, both of which provide a prominent synaptic input onto GnRH neurons, express estrogen receptors and concentrate radiolabeled estradiol [20 –22]. Opioid peptides and GABAergic ligands both serve to inhibit GnRH output [23,24] and thus luteinizing hormone (LH) release from the anterior pituitary [23,25–27]. While presynaptic interactions between opioid and GABAergic nerve terminals may help regulate this process [28,29], it is clear that both -opioid and GABAB receptor agonists affect a direct, postsynaptic inhibition of GnRH neurons [9]. Studies using the in vitro slice preparation have revealed that -opioid receptor-mediated inhibition of GnRH neurons arises from the activation of a member of the G protein-gated, inwardly rectifying K⫹ channel subfamily known as GIRK1– 4 or Kir3.1–3.4 [9,30,31]. This elicits a robust hyperpolarization in current clamp, or outward current in voltage clamp. In addition, -endorphin neurons are exquisitely responsive to -opioid receptor activation [32]. Transient estrogen exposure for no longer than 20 min results in a decreased potency of -opioid receptor agonists to hyperpolarize -endorphin neurons [7]. Similarly rapid effects of estrogen have been observed in the hippocampus, where the steroid potentiates kainate-induced currents in CA1 pyramidal cells [3]. The negative modulatory effect of estrogen (i.e. reduced -opioid receptor agonist potency in -endorphin neurons) persists for at least 24 h following systemic steroid administration [5]. What is the underlying cause of this estrogen-induced decrease in the responsiveness of -endorphin neurons to the -opioid receptor-mediated activation of an inwardly rectifying K⫹ channel? Studies that have examined -opioid receptor desensitization and opiate tolerance, both of which are associated with a refractoriness to -opioid receptor agonists, have implicated intracellular protein kinase pathways in these phenomena [33–35]. Indeed, we have shown previously that protein kinase A (PKA) inhibitors block the maintenance of cellular tolerance to -opioid receptor agonists following chronic morphine treatment that is observed in hypothalamic neurosecretory cells [36]. In addition, estrogen has been shown to rapidly stimulate PKA activity in peripheral (i.e. uterine) tissue, as well as to stimulate cyclic adenosine monophosphate responsive element binding protein (CREB) and c-fos expression [37– 41]. PKA activators such as Sp-cAMP mimic the effect of estrogen on -opioid receptor agonist potency (Fig. 1A). In the presence of non-selective protein kinase inhibitors (e.g. staurosporine) and selective PKA inhibitors such as Rp-cAMP and
Fig. 1. A, The selective PKA activator Sp-cAMP mimics the effect of estrogen on -opioid receptor agonist potency. DAMGO concentrationresponse curves were generated in this arcuate neuron after superfusion of the slice with 50 M Sp-cAMP (Œ, DAMGO EC50 ⫽ 104 nM) and then subsequently with 150 M Sp-cAMP (}, DAMGO EC50 ⫽ 175 nM). The dashed line with open circles represents a summary of control doseresponse curves generated prior to steroid administration. This effect was observed in 30% of arcuate neurons tested. B, Staurosporine (100 nM) and the more selective PKA inhibitors KT5720 (60 nM; n ⫽ 7) and Rp-cAMP (100 M; n ⫽ 8) blocked the effect of 17-estradiol (E2; 100 nM) on -opioid receptor agonist potency. These agents were superfused for 10 min before and during the 20 min steroid application. E2 was subsequently administered alone and found to reduce DAMGO potency (not shown), thereby confirming the estrogen sensitivity of these cells. **, P ⬍ 0.0001; *, P ⬍ 0.01 compared with controls. Reprinted from [59] with permission from the American Society for Pharmacology and Experimental Therapeutics.
KT5720, estrogen is without effect on potency of the -opioid receptor agonist DAMGO to hyperpolarize hypothalamic neurons (Fig. 1B). Similar findings were reported for estrogen-induced potentiation of kainate currents in hippocampal CA1 pyramidal cells [3]. This demonstrates that the modulation by estrogen of the -opioid receptor coupling to a GIRK channel is due to increased PKA activity. There is considerable evidence for crosstalk between various intracellular protein kinase pathways such as protein kinase C (PKC) and PKA in regulating receptor/K⫹ channel coupling. Sp-cAMP and PKC activators such as phorbol esters both mimic the effect of chronic morphine treatment in attenuating the potency of DAMGO to hyperpolarize
M.J. Kelly et al. / Steroids 67 (2002) 447– 456
449
Fig. 2. Estrogen attenuates the efficacy of GABAB receptor-mediated neurotransmission in the POA 24 h after its administration. a, Successively increasing doses of baclofen (1, 3 and 10 M) hyperpolarized this POA neuron from a vehicle-treated animal (resting Vm ⫽ ⫺45 mV) 10, 22.5 and 25.5 mV, respectively. The upward deflection represent the return of low-threshold spikes and/or action potentials (truncated) seen during the later stages of drug wash-out. b, Successively increasing doses of baclofen (3, 10, 30 and 100 M) hyperpolarized this POA neuron (resting Vm ⫽ ⫺50 mV) from an EB-treated (25 g; 24 h) animal by 1.5, 2.5, 3 and 4 mV, respectively. c, Composite dose-response curves from recordings of POA neurons obtained from vehicle- and EB-treated animals. Cells were perfused with successively higher concentrations of baclofen (1, 3, 10 and 30 M; 4 –7 min/dose; n ⫽ 2–10). Symbols represent means and vertical lines 1 S.E.M. of hyperpolarizations elicited by a given concentration of baclofen. The ⌬Vmax obtained via logistic fit for POA neurons from vehicle-treated animals was 13.5 mV, whereas that obtained for POA neurons from EB-treated animals was 7.5 mV. *, Hyperpolarizations obtained with 10 M and 30 M baclofen that are significantly different (multifactorial ANOVA/LSD; P ⬍ 0.05) than those obtained with 1 M or 3 M baclofen. #, Hyperpolarizations of POA neurons obtained from EB-treated animals are significantly lower (multifactorial ANOVA/LSD; P ⬍ 0.05) than those obtained from vehicle-treated animals at all doses tested. d, Composite bar graph illustrating the baclofen-induced ⌬g in POA neurons from vehicle- and EB-treated animals (n ⫽ 5– 8). Columns represent means and vertical lines 1 S.E.M. of the baclofen-induced ⌬g estimated by linear regression between ⫺60 & ⫺80 mV, and between ⫺100 & ⫺130 mV. *, Values of ⌬g obtained in POA neurons from EB-treated animals which are significantly different (multifactorial ANOVA/LSD; P ⬍ 0.05) than those obtained from vehicle-treated controls. Reprinted from [46] with permission. Copyright 2001 by The Society for Neuroscience.
hypothalamic neurons [36]. In addition, the protein kinase C inhibitor calphostin C blocks the estrogen-induced decrease in DAMGO potency [42]. However, PKA but not PKC inhibitors block the maintenance of cellular tolerance caused by chronic opiate exposure [36]. On the other hand, PKC activators potentiate -adrenergic receptor agonist stimulation of cyclic adenosine monophosphate (cAMP) accumulation in the POA [43]. These observations suggest that PKC and PKA regulatory pathways can co-exist in parallel and serial intracellular arrangements. As with -opioid receptor stimulation, the activation of GABAB receptors elicits a hyperpolarization or an outward
current, and hypothalamic GABAergic neurons in the arcuate nucleus and POA exhibit particularly robust responses [44 – 46]. These responses arise from the coupling of GABAB receptors to the same population of GIRK channels as -opioid receptors [44,47]. Short-term estrogen exposure also reduces the potency of the GABAB receptor agonist baclofen to hyperpolarize hypothalamic neurons [8]. In the arcuate nucleus, this reduction in agonist potency is maintained for at least 24 h after steroid administration [5]. In the POA however, the estrogen-induced uncoupling of the GABAB receptor with its effector K⫹ channel in GABAergic neurons is manifest by a reduction in agonist efficacy
450
M.J. Kelly et al. / Steroids 67 (2002) 447– 456
Fig. 3. A, IAHPs obtained in a POA neuron from a vehicle-treated animal prior to (a) and in the presence of (b) a sub-maximal concentration of the ␣1-adrenergic receptor agonist methoxamine (3 M). B, IAHPs procured prior to (a) and in the presence of (b) 3 M methoxamine from a GAD-positive POA neuron (shown in Fig. 5) 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 M, caused by the bath application of E2 (100 nM; 15–20 min). Columns represent means and vertical lines 1 S.E.M. (n ⫽ 4 –5) of the percent decrease in the IAHP due to the bath application of methoxamine (3 M) in POA neurons with no intervening steroid treatment (blank column), and in POA neurons given a short-term exposure to E2 (filled column). The baseline IAHP control values for 3 M 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 M 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. Reprinted from [50] with permission from the American Society for Pharmacology and Experimental Therapeutics.
(Fig. 2a– c). This finding is substantiated by the parallel decrease in the agonist-induced change in slope conductance (Fig. 2d), which suggests a somewhat different effect of estrogen on POA versus arcuate neurons. In addition to these inwardly rectifying K⫹ channels, hypothalamic neurons also express small-conductance, Ca2⫹-activated K⫹ (SK) channels that are sensitive to antagonism by the peptidergic, honeybee toxin apamin [48, 49]. These currents underlie the afterhyperpolarization (AHP) that is observed at the tail end of the action potential or a prolonged depolarizing stimulus [47,49]. In the POA,
we have found that the SK currents in GABAergic neurons are inhibited by - and ␣1-adrenergic receptor activation [50]. Of particular interest is that the ␣1-adrenergic receptor-mediated inhibition of the SK current in these GABAergic neurons is markedly potentiated by a transient exposure to estrogen (Fig. 3). This estrogen-induced enhancement of the coupling of ␣1-adrenergic receptors to their effector SK channels lasts for at least 24 h following systemic steroid administration (Fig. 4). An example of an estrogen-responsive GABAergic neuron identified via combined histofluorescence and in situ hybridization is shown in Fig. 5.
M.J. Kelly et al. / Steroids 67 (2002) 447– 456
451
Fig. 4. Examples of the IAHP that were obtained before (a) and during (b) the bath application of 10 M methoxamine in POA neurons from vehicle- (A) and EB-treated (25 g, 24 h prior; B) animals. C, Composite dose-response curves for methoxamine that were derived from POA neurons in vehicle- (E) and EB-treated (■) animals. The curves were fit via a logistic equation to the experimental data points. Symbols represent means and vertical lines 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 M). *, Denotes a significant main effect of EB on the methoxamine-induced inhibition of the IAHP (multifactorial ANOVA/LSD; P ⬍ 0.05). Reprinted from [50] with permission from the American Society for Pharmacology and Experimental Therapeutics.
3. Estrogen directly modulates ion channel activity through a G protein linkage It has been known for some time that iontophoretically applied estrogen to POA neurons alters their excitability primarily by affecting an inhibition of firing rate [51]. This inhibition is associated with a membrane hyperpolarization (Fig. 6) due to the opening of a K⫹ channel that has also been observed in the amygdala and arcuate nucleus [9,11,
52,53]. This latter region contains GnRH neurons which are particularly responsive to this estrogenic inhibition [9,11]. Estrogen also rapidly inhibits voltage-gated Ca2⫹ channels in medium spiny GABAergic neurons of the basal ganglia, an effect mediated via a pertussis toxinsensitive G protein [10]. Interestingly, the estrogen-induced augmentation of kainate currents in hippocampal CA1 pyramidal cells also involves an intervening G protein [3]. Thus, estrogen apparently is capable of relatively direct modulation of ion channel activity through a G
452
M.J. Kelly et al. / Steroids 67 (2002) 447– 456
Fig. 5. A, Photomicrograph of the biocytin–streptavidin–CY3 fluorescent labeling of the 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. 3B. D, An overlay of the fluorescent labeling in C, and the hybridization signal that clearly illustrates the double-labeling for GAD65. Reprinted from [50] with permission from the American Society for Pharmacology and Experimental Therapeutics.
protein, as well as more subtle modulation of G proteincoupled receptor/K⫹ channel coupling.
4. Significance As mentioned at the outset, estrogen primarily serves to regulate the output of GnRH from the mediobasal hypothalamus, and in doing so controls reproductive status through the hypothalamic–pituitary– gonadal axis. The work presented here sheds meaningful insight into the cellular mechanisms by which estrogen exerts negative feedback over the female reproductive cycle. First of all, estrogen negatively modulates the coupling of the -opioid and GABAB receptors to their effector K⫹ channel in -endorphin and GABAergic neurons, respectively. This uncoupling requires the activation of a hierarchal, intracellular phosphorylation cascade involving both PKC and PKA pathways. PKA also appears to be involved in the estrogen-induced augmentation of kainate-induced currents in hippocampal pyramidal cells [3]. PKA-dependent phosphorylation of the -opioid receptor hinders its association with Gi/o [54,55]. However, it remains to be seen if the -opioid receptor is indeed the
phosphorylation substrate that renders it less apt to couple to its GIRK channel. Alternative targets include Gi/o subunits, GIRK channels and protein regulators of G protein signalling. Future studies will endeavor to elucidate how the phosphorylation affects this uncoupling in -endorphin neurons. -Opioid and GABAB receptors serve as autoreceptors in their respective -endorphin and hypothalamic GABAergic neurons [32,45,46]. The fact that estrogen uncouples these autoreceptors from their GIRK channel implies that the steroid decreases the autocrine, somatodendritic inhibition of these cells, thereby increasing the release of these inhibitory neurotransmitters. Indeed, estrogen rapidly increases extracellular GABA concentrations in the POA as measured by push/pull perfusion and microdialysis [27,56]. Moreover, estrogen markedly potentiates the negative coupling between the ␣1-adrenergic receptor and its effector SK channel in POA GABAergic neurons, which serves to facilitate the excitation caused by ascending noradrenergic input onto these cells [22,50]. Coupled with the attenuated GABAB receptor-mediated autoinhibition of these GABAergic neurons, it stands to reason that estrogen would dramatically increase the firing rate of these neurons during
M.J. Kelly et al. / Steroids 67 (2002) 447– 456
453
Fig. 6. A, A current clamp recording of a POA neuron showing the reversible hyperpolarization (7 mV) elicited by 100 nM E2. The resting potential of this cell was ⫺62 mV. The downward deflections represent voltage responses to intermittent current pulses (not shown) used to monitor input resistance. The break in the trace represents the time necessary to complete a post-drug I/V relationship (⬇ 3 min). B, An I/V plot derived from the cell in A that illustrates the effect of bath applied E2. This neuron exhibited a reversal potential for the E2 response of ⫺76 mV, which is close to the equilibrium potential for K⫹, and a ⌬g of 5.68 nS measured between ⫺60 and ⫺80 mV. Reprinted from [60] with permission from Academic Press.
negative feedback. Given that both -endorphin and hypothalamic GABAergic neurons synapse onto GnRH neurons [57,58], the collective modulation by estrogen of K⫹ channel activity in these cells would greatly enhance the inhibitory tonus impinging on GnRH neurons. Finally, estrogen itself can hyperpolarize and thereby inhibit GnRH neurons by activating a K⫹ channel. While it has long been thought that GnRH neurons do not contain estrogen receptors [15–17], recent evidence indicates that these neurons express the  isoform of the estrogen receptor [12–14,19]. In addition, both the estrogen-induced augmentation of kainate currents in hippocampal CA1 pyramidal cells and the inhibition of Ca2⫹ currents in medium spiny GABAergic neurons appear to involve G proteins [3,10]. It is therefore conceivable that estrogen could interact with a membrane-associated receptor to induce a hyperpolarization of GnRH neurons by way of a G protein. Thus, there are at least three distinct mechanisms through which estrogen can exert negative feedback on GnRH output. The dynamics of
these cellular mechanisms are illustrated in schematic shown in Fig. 7.
Acknowledgments The experiments from the authors’ laboratory described in this review were supported by Public Health Service Grants NS 35944, NS 38809, DA 05158 and DA 00192 (Research Scientist Development Award to MJK). The authors would like to recognize Ms. Martha A. Bosch and Mr. Barry R. Naylor for their expert technical contributions to these studies.
References [1] Kelly MJ, Wagner EJ. Estrogen modulation of G-protein-coupled receptors. Trends Endocrinol Metab 1999;10:369 –74.
454
M.J. Kelly et al. / Steroids 67 (2002) 447– 456
Fig. 7. Schematic representation illustrating the central mechanisms of estrogen-induced negative feedback on the mammalian female reproductive axis. During negative feedback estrogen uncouples somatodendritic GABAB (GABAB-R) and -opioid (MOR) receptors from their effector K⫹ channels in GABAergic and -endorphin neurons [1]. [Note: Although the GABAB and -opioid receptors couple to the same population of GIRK’s in both GABAergic and -endorphin neurons, only the autoreceptors are illustrated.] This leads to a decreased autoinhibition and an increased inhibitory tone onto GnRH neurons. Additional negative feedback is provided by estrogen-sensitive, ascending noradrenergic input to GABAergic neurons, in which the steroid markedly potentiates the ␣1-adrenergic receptor (␣1R)-mediated inhibition of an apamin-sensitive SK current [2]. This results in a further increase of GABAergic neuronal excitability. Finally, estrogen can affect a direct inhibition of GnRH neurons via a G protein coupling to a K⫹ channel [3]. These three distinct mechanisms by which estrogen modulates K⫹ channel activity accounts for, at least in part, the negative feedback effect of estrogen on the hypothalamic– pituitary axis. [2] Wagner EJ, Moore KE, Lookingland KJ. Sexual differences in N-methyl-D-aspartate receptor-mediated regulation of tuberoinfundibular dopaminergic neurons in the rat. Brain Res 1993;611: 139 – 46. [3] Gu Q, Moss RL. 17-Estradiol potentiates kainate-induced currents via activation of the cAMP cascade. J Neurosci 1996;16:3620 –9. [4] Demotes-Mainard J, Arnauld E, Vincent JD. Estrogens modulate the responsiveness of in vivo recorded striatal neurons to iontophoretic application of dopamine in rats: Role of D1 and D2 receptor activation. J Neuroendocrinol 1990;2:825–32. [5] Kelly MJ, Loose MD, Rønnekleiv OK. Estrogen suppresses -opioidand GABAB-mediated hyperpolarization of hypothalamic arcuate neurons. J Neurosci 1992;12:2745–50. [6] Wagner EJ, Manzanares J, Moore KE, Lookingland KJ. Neurochemical evidence that estrogen-induced suppression of -opioid-receptormediated regulation of tuberoinfundibular dopaminergic neurons is prolactin-independent. Neuroendocrinology 1994;59:197–201. [7] Lagrange AH, Rønnekleiv OK, Kelly MJ. The potency of -opioid hyperpolarization of hypothalamic arcuate neurons is rapidly attenuated by 17-estradiol. J Neurosci 1994;14:6196 –204.
[8] Lagrange AH, Wagner EJ, Rønnekleiv OK, Kelly MJ. Estrogen rapidly attenuates a GABAB response in hypothalamic neurons. Neuroendocrinology 1996;64:114 –23. [9] Lagrange AH, Rønnekleiv OK, Kelly MJ. Estradiol-17, and -opioid peptides rapidly hyperpolarize GnRH neurons: A cellular mechanism of negative feedback. Endocrinology 1995;136:2341– 4. [10] Mermelstein PG, Becker JB, Surmeier DJ. Estradiol reduces calcium currents in rat neostriatal neurons via a membrane receptor. J Neurosci 1996;16:595– 604. [11] Kelly MJ, Rønnekleiv OK, Eskay RL. Identification of estrogenresponsive LHRH neurons in the guinea pig hypothalamus. Brain Res Bull 1984;12:399 – 407. [12] Skynner MJ, Sim JA, Herbison AE. Detection of estrogen receptor ␣ and  messenger ribonucleic acids in adult gonadotropin-releasing hormone neurons. Endocrinology 1999;140:5195–201. [13] Butler JA, Sjo¨ berg M, Coen CW. Evidence for oestrogen receptor ␣-immunoreactivity in gonadotropin-releasing hormone-expressing neurones. J Neuroendocrinol 1999;11:331–5. [14] Hrabovszky E, Shughrue PJ, Merchenthaler I , et al. Detection of estrogen receptor- messenger ribonucleic acid and 125I-estrogen
M.J. Kelly et al. / Steroids 67 (2002) 447– 456
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22] [23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31] [32]
binding sites in luteinizing hormone-releasing hormone neurons of the rat brain. Endocrinology 2000;141:3506 –9. Watson RE, Langub MC, Landis JW. Further evidence that most luteinizing hormone-releasing hormone neurons are not directly estrogen-responsive: simultaneous localization of luteinizing hormonereleasing hormone and estrogen receptor immunoreactivity in the guinea-pig brain. J Neuroendocrinol 1992;4:311–7. Herbison AE, Horvath TL, Naftolin F, Leranth C. Distribution of estrogen receptor-immunoreactive cells in monkey hypothalamus: relationship to neurones containing luteinizing hormone-releasing hormone and tyrosine hydroxylase. Neuroendocrinology 1995;61:1– 10. Sullivan KA, Witkin JW, Ferin M, Silverman A-J. Gonadotropinreleasing hormone neurons in the rhesus macaque are not immunoreactive for the estrogen receptor. Brain Res 1995;685:198 –200.. Simonian SX, Spratt DP, Herbison AE. Identification and characterization of estrogen receptor ␣-containing neurons projecting to the vicinity of the gonadotropin-releasing hormone perikarya in the rostral preoptic area of the rat. J Comp Neurol 1999;411:346 –58. Herbison AE, Skynner MJ, Sim JA. Lack of detection of estrogen receptor- ␣ transcripts in mouse gonadotropin-releasing hormone neurons. Endocrinology 2001;142:493. Morrell JI, McGinty JF, Pfaff DW. A subset of -endorphin- or dynorphin-containing neurons in the medial basal hypothalamus accumulates estradiol. Neuroendocrinology 1985;41:417–26. Leranth C, MacLusky NJ, Brown TJ, Redmond DE, Naftolin F. Transmitter content and afferent connections of estrogen-sensitive progestin receptor containing neurons in the primate hypothalamus. Neuroendocrinology 1992;55:667– 82. Herbison AE. Estrogen regulation of GABA transmission in rat preoptic area. Brain Res Bull 1997;44:321– 6. Ferin M, Van Vugt D, Wardlaw S. The hypothalamic control of the menstrual cycle and the role of endogenous opioid peptides. Recent Prog Horm Res 1984;40:441– 85. Mitsushima D, Marzban F, Luchansky LL, et al. Role of glutamic acid decarboxylase in the prepubertal inhibition of the luteinizing hormone releasing hormone release in female rhesus monkeys. J Neurosci 1996;16:2563–73. Akema T, Chiba A, Kimura F. On the relationship between noradrenergic stimulatory and GABAergic inhibitory systems in the control of luteinizing hormone secretion in female rats. Neuroendocrinology 1990;52:566 –72. Seltzer AM, Donoso AO. Restraining action of GABA on estradiolinduced LH surge in the rat: GABA activity in brain nuclei and effects of GABA mimetics in the medial preoptic nucleus. Neuroendocrinology 1992;55:28 –34. Jarry H, Leonhardt S, Wuttke W. The inhibitory effect of -endorphin on LH release in ovariectomized rats does not involve the preoptic GABAergic system. Exp Clin Endocrinol Diabetes 1995;103:317–23. Masotto C, Wisniewski G, Negro-Vilar A. Different ␥-aminobutyric acid receptor subtypes are involved in the regulation of opiate-dependent, and independent luteinizing hormone-releasing hormone secretion. Endocrinology 1989;125:548 –53. Brann DW, Zamorano PL, Putnam-Roberts CD, Mahesh VB. Gamma-aminobutyric acid-opioid interactions in the regulation of gonadotropin secretion in the immature female rat. Neuroendocrinology 1992;56:445–52. Loose MD, Kelly MJ. Opioids act as -receptors to hyperpolarize arcuate neurons via an inwardly rectifying potassium conductance. Brain Res 1990;513:15–23. Jan LY, Jan YN. Voltage-gated and inwardly rectifying potassium channels. J Physiol (Lond) 1997;505:267– 82. Kelly MJ, Loose MD, Ronnekleiv OK. Opioids hyperpolarize -endorphin neurons via -receptor activation of a potassium conductance. Neuroendocrinology 1990;52:268 –75.
455
[33] Narita M, Feng Y, Makimura M, Hoskins B, Ho IK. A protein kinase inhibitor, H-7, inhibits the development of tolerance to opioid antinociception. Eur J Pharmacol 1994;271:543–5. [34] Mestek A, Hurley JH, Bye LS , et al. The human opioid receptor: Modulation of functional desensitization by calcium/calmodulin-dependent protein kinase and protein kinase C. J Neurosci 1995;15: 2396 – 406. [35] Wang L, Medina VM, Rivera M, Gintzler AR. Relevance of phosphorylation state to opioid responsiveness in opiate naive and tolerant/dependent tissue. Brain Res 1996;723:61–9. [36] Wagner EJ, Rønnekleiv OK, Kelly MJ. Protein kinase A maintains cellular tolerance to -opioid receptor agonists in hypothalamic neurosecretory cells with chronic morphine treatment: Convergence upon a common pathway with estrogen in modulating -opioid receptor/ effector coupling. J Pharmacol Exp Ther 1998;285:1266 –73. [37] Aronica SM, Katzenellenbogen BS. Stimulation of estrogen receptor mediated transcription and alteration in the phosphorylation state of the rat uterine estrogen receptor by estrogen, cyclic adenosine monophosphate, and insulin-like growth factor-1. Mol Endocrinol 1993;7: 743–52. [38] Nakhla AM, Khan MS, Romas NP, Rosner W. Estradiol causes the rapid accumulation of cAMP in human prostate. Proc Natl Acad Sci 1994;91:5402–5. [39] Zhou Y, Watters JJ, Dorsa DM. Estrogen rapidly induces the phosphorylation of the cAMP response element binding protein in rat brain. Endocrinology 1996;137:2163– 6. [40] Gu G, Rojo AA, Zee MC, Yu J, Simerly RB. Hormonal regulation of CREB phosphorylation in the anteroventral periventricular nucleus. J Neurosci 1996;16:3035– 44. [41] Watters JJ, Campbell JS, Cunningham MJ, Krebs EG, Dorsa DM. Rapid membrane effects of steroids in neuroblastoma cells: effects of estrogen on mitogen activated protein kinase signalling cascade and c-fos immediate early gene transcription. Endocrinology 2001;138: 4030 –3. [42] Kelly MJ, Lagrange AH, Wagner EJ, Rønnekleiv OK. Rapid effects of estrogen to modulate G protein-coupled receptors via activation of protein kinase A and protein kinase C pathways. Steroids 1999;64: 64 –75. [43] Ansonoff MA, Etgen AM. Estradiol elevates protein kinase C catalytic activity in the preoptic area of female rats. Endocrinology 1998;139:3050 – 6. [44] Loose MD, Rønnekleiv OK, Kelly MJ. Neurons in the rat arcuate nucleus are hyperpolarized by GABA(B) and -opioid receptor agonists: Evidence for convergence at a ligand-gated potassium conductance. Neuroendocrinology 1991;54:537– 44. [45] Wagner EJ, Bosch MA, Kelly MJ, Rønnekleiv OK. A powerful GABAB receptor-mediated inhibition of GABAergic neurons in the arcuate nucleus. Neuroreport 1999;10:2681–7. [46] Wagner EJ, Rønnekleiv OK, Bosch MA, Kelly MJ. Estrogen biphasically modifies hypothalamic GABAergic neuronal function concomitantly with negative and positive control of luteinizing hormone release. J Neurosci 2001;21:2085–93. [47] Wagner EJ, Reyes-Vazquez C, Rønnekleiv OK, Kelly MJ. The role of intrinsic and agonist-activated conductances in determining the firing patterns of preoptic area neurons in the guinea pig. Brain Res 2000; 879:29 – 41. [48] Erickson KR, Rønnekleiv OK, Kelly MJ. Role of a T-type calcium current in supporting a depolarizing potential, damped oscillations, and phasic firing in vasopressinergic guinea pig supraoptic neurons. Neuroendocrinology 1993;57:789 – 800. [49] Kirkpatrick K, Bourque CW. Activity dependence and functional role of the apamin-sensitive K⫹ current in rat supraoptic neurones in vitro. J Physiol (Lond) 1996;494:389 –98. [50] Wagner EJ, Rønnekleiv OK, Kelly MJ. The noradrenergic inhibition of an apamin-sensitive, small conductance Ca2⫹-activated K⫹ channel in hypothalamic GABAergic neurons: pharmacology, estrogen
456
[51]
[52]
[53]
[54]
[55]
M.J. Kelly et al. / Steroids 67 (2002) 447– 456 sensitivity and relevance to the control of the reproductive axis. J Pharmacol Exp Ther 2001;299:21–30. Kelly MJ, Moss RL, Dudley CA. The effects of microelectrophorectically applied estrogen, cortisol and acetylcholine on medial preoptic-septal unit activity throughout the estrous cycle of the female rat. Exp Brain Res 1977;30:53– 64. Kelly MJ, Kuhnt U, Wuttke W. Hyperpolarization of hypothalamic parvocellular neurons by 17-estradiol and their identification through intracellular staining with procion yellow. Exp Brain Res 1980;40:440 –7. Nabekura J, Oomura Y, Minami T, Mizuno Y, Fukuda A. Mechanism of the rapid effect of 17-estradiol on medial amygdala neurons. Science 1986;233:226 – 8. Harada H, Ueda H, Wada Y, Katada T, Ui M, Satoh M. Phosphorylation of -opioid receptors - a putative mechanism of selective uncoupling of receptor - Gi interaction, measured with low-Km GTPase and nucleotide-sensitive agonist binding. Neurosci Lett 1989; 100:221– 6. Harada H, Ueda H, Katada T, Ui M, Satoh M. Phosphorylated -opioid receptor purified from rat brain lacks functional coupling
[56]
[57]
[58]
[59]
[60]
with Gi1, a GTP-binding protein in reconstituted lipid vesicles. Neurosci Lett 1990;113:47–9. Herbison AE, Heavens RP, Dye S, Dyer RG. Acute action of oestrogen on medial preoptic ␥-aminobutyric acid neurons: Correlation with oestrogen negative feedback on luteinizing hormone secretion. J Neuroendocrinol 1991;3:101– 6. Leranth C, MacLusky NJ, Sakamoto H, Shanabrough M, Naftolin F. Glutamic acid decarboxylase-containing axons synapse on LHRH neurons in the rat medial preoptic area. Neuroendocrinology 1985; 40:536 –9. Naftolin F, Leranth C, Horvath TL, Garcia-Segura LM. Potential neuronal mechanisms of estrogen actions in synaptogenesis and synaptic plasticity. Cell Mol Neurobiol 1996;16:213–23. Lagrange AH, Rønnekleiv OK, Kelly MJ. Modulation of G proteincoupled receptors by an estogen receptor that activates protein kinase A. Mol Pharmacol 1997;51:605–12. Kelly MJ, Rønnekleiv OK. Rapid membrane effects of estrogen in the CNS. In: Pfaff DW, editor. Hormones, Brain and Behavior. San Diego: Academic Press (in press).
Estrogen modulation of K⫹ channel activity in hypothalamic neurons involved in the control of the reproductive axis Martin J. Kelly*, Oline K. Rønnekleiv, Nurhadi Ibrahim, Andre H. Lagrange, Edward J. Wagner Department of Physiology & Pharmacology, Oregon Health & Science University, Portland, OR, USA
Abstract Here we report on the progress we have made in elucidating the mechanisms through which estrogen alters synaptic responses in hypothalamic neurons. We examined the modulation by estrogen of the coupling of various receptor systems to inwardly rectifying and small conductance, Ca2⫹-activated K⫹ (SK) channels. We used intracellular sharp-electrode and whole-cell recordings in hypothalamic slices from ovariectomized female guinea pigs. Estrogen rapidly uncouples -opioid receptors from G protein-gated inwardly rectifying K⫹ (GIRK) channels in -endorphin neurons, manifest by a reduction in the potency of -opioid receptor agonists to hyperpolarize these cells. This effect is blocked by inhibitors of protein kinase A and protein kinase C. Estrogen also uncouples ␥-aminobutyric acid (GABA)B receptors from the same population of GIRK channels coupled to -opioid receptors. At 24 h after steroid administration, the GABAB/GIRK channel uncoupling observed in GABAergic neurons of the preoptic area (POA) is associated with reduced agonist efficacy. Conversely, estrogen enhances the efficacy of ␣1-adrenergic receptor agonists to inhibit apamin-sensitive SK currents in these POA GABAergic neurons, and does so in both a rapid and sustained fashion. Finally, we observed a direct, steroid-induced hyperpolarization of both arcuate and POA neurons, among which gonadotropin-releasing hormone (GnRH) neurons are particularly sensitive. These findings indicate a richly complex yet coordinated steroid modulation of K⫹ channel activity that serves to control the excitability of hypothalamic neurons involved in regulating the reproductive axis. © 2002 Elsevier Science Inc. All rights reserved. Keywords: Estrogen; -Endorphin; GABA; Norepinephrine; K⫹ Channel; Protein kinase
1. Introduction It is becoming increasingly evident that the gonadal steroid hormone estrogen imparts a multifaceted influence over synaptic transmission in the mammalian central nervous system. Not only can estrogen alter synaptic responses via genomic mechanisms, but there is a wealth of information that indicates the steroid can also modulate cell-to-cell communication much more rapidly (for review see [1]). These synaptic alterations are brought about via changes in the cellular responsiveness to the activation of various receptor systems (both G proteincoupled and ionotropic) to their respective first messengers. For example, estrogen can modulate the cellular responsiveness to ionophoric glutamate (both N-methyl-D-aspartate (NMDA) and non-NMDA) receptor activation [2,3]. In addition, it can alter the linkage of G protein-coupled receptor systems such as opioid (both and ), ␥-aminobutyric acid * Corresponding author. Tel.: ⫹1-503-494-5833; fax: ⫹1-503-4944352. E-mail address: [email protected] (M.J. Kelly).
(GABA)B and dopamine D2 receptors to their respective effector systems [4 – 8]. Furthermore, it now appears that the steroid can function as a first messenger by activating an estrogen receptor that couples directly to K⫹ and Ca2⫹ channels by way of a pertussis toxin-sensitive G protein [9,10]. These fundamentally distinct signalling pathways give rise to a coordinated regulation by estrogen of complex physiological processes such as reproduction, fine motor control and memory. Given estrogen’s preeminent role in regulating reproductive physiology and behavior, we will focus on the estrogenic modulation of K⫹ channel activity as it pertains to the negative feedback control of the hypothalamic–pituitary– gonadal axis.
2. Estrogen modulates the linkage between G protein-coupled receptors and their effector potassium channels The principal action of estrogen is to regulate the output of gonadotropin-releasing hormone (GnRH) from the mediobasal hypothalamus. Although we found direct
0039-128X/02/$ – see front matter © 2002 Elsevier Science Inc. All rights reserved. PII: S 0 0 3 9 - 1 2 8 X ( 0 1 ) 0 0 1 8 1 - 7
448
M.J. Kelly et al. / Steroids 67 (2002) 447– 456
actions of estrogen to inhibit GnRH neuronal activity over 15 years ago [9,11], it has been only recently that estrogen receptors have been demonstrated in GnRH neurons [12–14]. Therefore, little attention has been paid to the direct effect of estrogen on GnRH neurons. Rather, estrogen responsiveness has been conferred to neurons that synapse on GnRH cells [15–19]. Indeed, hypothalamic -endorphin and GABAergic neurons, both of which provide a prominent synaptic input onto GnRH neurons, express estrogen receptors and concentrate radiolabeled estradiol [20 –22]. Opioid peptides and GABAergic ligands both serve to inhibit GnRH output [23,24] and thus luteinizing hormone (LH) release from the anterior pituitary [23,25–27]. While presynaptic interactions between opioid and GABAergic nerve terminals may help regulate this process [28,29], it is clear that both -opioid and GABAB receptor agonists affect a direct, postsynaptic inhibition of GnRH neurons [9]. Studies using the in vitro slice preparation have revealed that -opioid receptor-mediated inhibition of GnRH neurons arises from the activation of a member of the G protein-gated, inwardly rectifying K⫹ channel subfamily known as GIRK1– 4 or Kir3.1–3.4 [9,30,31]. This elicits a robust hyperpolarization in current clamp, or outward current in voltage clamp. In addition, -endorphin neurons are exquisitely responsive to -opioid receptor activation [32]. Transient estrogen exposure for no longer than 20 min results in a decreased potency of -opioid receptor agonists to hyperpolarize -endorphin neurons [7]. Similarly rapid effects of estrogen have been observed in the hippocampus, where the steroid potentiates kainate-induced currents in CA1 pyramidal cells [3]. The negative modulatory effect of estrogen (i.e. reduced -opioid receptor agonist potency in -endorphin neurons) persists for at least 24 h following systemic steroid administration [5]. What is the underlying cause of this estrogen-induced decrease in the responsiveness of -endorphin neurons to the -opioid receptor-mediated activation of an inwardly rectifying K⫹ channel? Studies that have examined -opioid receptor desensitization and opiate tolerance, both of which are associated with a refractoriness to -opioid receptor agonists, have implicated intracellular protein kinase pathways in these phenomena [33–35]. Indeed, we have shown previously that protein kinase A (PKA) inhibitors block the maintenance of cellular tolerance to -opioid receptor agonists following chronic morphine treatment that is observed in hypothalamic neurosecretory cells [36]. In addition, estrogen has been shown to rapidly stimulate PKA activity in peripheral (i.e. uterine) tissue, as well as to stimulate cyclic adenosine monophosphate responsive element binding protein (CREB) and c-fos expression [37– 41]. PKA activators such as Sp-cAMP mimic the effect of estrogen on -opioid receptor agonist potency (Fig. 1A). In the presence of non-selective protein kinase inhibitors (e.g. staurosporine) and selective PKA inhibitors such as Rp-cAMP and
Fig. 1. A, The selective PKA activator Sp-cAMP mimics the effect of estrogen on -opioid receptor agonist potency. DAMGO concentrationresponse curves were generated in this arcuate neuron after superfusion of the slice with 50 M Sp-cAMP (Œ, DAMGO EC50 ⫽ 104 nM) and then subsequently with 150 M Sp-cAMP (}, DAMGO EC50 ⫽ 175 nM). The dashed line with open circles represents a summary of control doseresponse curves generated prior to steroid administration. This effect was observed in 30% of arcuate neurons tested. B, Staurosporine (100 nM) and the more selective PKA inhibitors KT5720 (60 nM; n ⫽ 7) and Rp-cAMP (100 M; n ⫽ 8) blocked the effect of 17-estradiol (E2; 100 nM) on -opioid receptor agonist potency. These agents were superfused for 10 min before and during the 20 min steroid application. E2 was subsequently administered alone and found to reduce DAMGO potency (not shown), thereby confirming the estrogen sensitivity of these cells. **, P ⬍ 0.0001; *, P ⬍ 0.01 compared with controls. Reprinted from [59] with permission from the American Society for Pharmacology and Experimental Therapeutics.
KT5720, estrogen is without effect on potency of the -opioid receptor agonist DAMGO to hyperpolarize hypothalamic neurons (Fig. 1B). Similar findings were reported for estrogen-induced potentiation of kainate currents in hippocampal CA1 pyramidal cells [3]. This demonstrates that the modulation by estrogen of the -opioid receptor coupling to a GIRK channel is due to increased PKA activity. There is considerable evidence for crosstalk between various intracellular protein kinase pathways such as protein kinase C (PKC) and PKA in regulating receptor/K⫹ channel coupling. Sp-cAMP and PKC activators such as phorbol esters both mimic the effect of chronic morphine treatment in attenuating the potency of DAMGO to hyperpolarize
M.J. Kelly et al. / Steroids 67 (2002) 447– 456
449
Fig. 2. Estrogen attenuates the efficacy of GABAB receptor-mediated neurotransmission in the POA 24 h after its administration. a, Successively increasing doses of baclofen (1, 3 and 10 M) hyperpolarized this POA neuron from a vehicle-treated animal (resting Vm ⫽ ⫺45 mV) 10, 22.5 and 25.5 mV, respectively. The upward deflection represent the return of low-threshold spikes and/or action potentials (truncated) seen during the later stages of drug wash-out. b, Successively increasing doses of baclofen (3, 10, 30 and 100 M) hyperpolarized this POA neuron (resting Vm ⫽ ⫺50 mV) from an EB-treated (25 g; 24 h) animal by 1.5, 2.5, 3 and 4 mV, respectively. c, Composite dose-response curves from recordings of POA neurons obtained from vehicle- and EB-treated animals. Cells were perfused with successively higher concentrations of baclofen (1, 3, 10 and 30 M; 4 –7 min/dose; n ⫽ 2–10). Symbols represent means and vertical lines 1 S.E.M. of hyperpolarizations elicited by a given concentration of baclofen. The ⌬Vmax obtained via logistic fit for POA neurons from vehicle-treated animals was 13.5 mV, whereas that obtained for POA neurons from EB-treated animals was 7.5 mV. *, Hyperpolarizations obtained with 10 M and 30 M baclofen that are significantly different (multifactorial ANOVA/LSD; P ⬍ 0.05) than those obtained with 1 M or 3 M baclofen. #, Hyperpolarizations of POA neurons obtained from EB-treated animals are significantly lower (multifactorial ANOVA/LSD; P ⬍ 0.05) than those obtained from vehicle-treated animals at all doses tested. d, Composite bar graph illustrating the baclofen-induced ⌬g in POA neurons from vehicle- and EB-treated animals (n ⫽ 5– 8). Columns represent means and vertical lines 1 S.E.M. of the baclofen-induced ⌬g estimated by linear regression between ⫺60 & ⫺80 mV, and between ⫺100 & ⫺130 mV. *, Values of ⌬g obtained in POA neurons from EB-treated animals which are significantly different (multifactorial ANOVA/LSD; P ⬍ 0.05) than those obtained from vehicle-treated controls. Reprinted from [46] with permission. Copyright 2001 by The Society for Neuroscience.
hypothalamic neurons [36]. In addition, the protein kinase C inhibitor calphostin C blocks the estrogen-induced decrease in DAMGO potency [42]. However, PKA but not PKC inhibitors block the maintenance of cellular tolerance caused by chronic opiate exposure [36]. On the other hand, PKC activators potentiate -adrenergic receptor agonist stimulation of cyclic adenosine monophosphate (cAMP) accumulation in the POA [43]. These observations suggest that PKC and PKA regulatory pathways can co-exist in parallel and serial intracellular arrangements. As with -opioid receptor stimulation, the activation of GABAB receptors elicits a hyperpolarization or an outward
current, and hypothalamic GABAergic neurons in the arcuate nucleus and POA exhibit particularly robust responses [44 – 46]. These responses arise from the coupling of GABAB receptors to the same population of GIRK channels as -opioid receptors [44,47]. Short-term estrogen exposure also reduces the potency of the GABAB receptor agonist baclofen to hyperpolarize hypothalamic neurons [8]. In the arcuate nucleus, this reduction in agonist potency is maintained for at least 24 h after steroid administration [5]. In the POA however, the estrogen-induced uncoupling of the GABAB receptor with its effector K⫹ channel in GABAergic neurons is manifest by a reduction in agonist efficacy
450
M.J. Kelly et al. / Steroids 67 (2002) 447– 456
Fig. 3. A, IAHPs obtained in a POA neuron from a vehicle-treated animal prior to (a) and in the presence of (b) a sub-maximal concentration of the ␣1-adrenergic receptor agonist methoxamine (3 M). B, IAHPs procured prior to (a) and in the presence of (b) 3 M methoxamine from a GAD-positive POA neuron (shown in Fig. 5) 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 M, caused by the bath application of E2 (100 nM; 15–20 min). Columns represent means and vertical lines 1 S.E.M. (n ⫽ 4 –5) of the percent decrease in the IAHP due to the bath application of methoxamine (3 M) in POA neurons with no intervening steroid treatment (blank column), and in POA neurons given a short-term exposure to E2 (filled column). The baseline IAHP control values for 3 M 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 M 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. Reprinted from [50] with permission from the American Society for Pharmacology and Experimental Therapeutics.
(Fig. 2a– c). This finding is substantiated by the parallel decrease in the agonist-induced change in slope conductance (Fig. 2d), which suggests a somewhat different effect of estrogen on POA versus arcuate neurons. In addition to these inwardly rectifying K⫹ channels, hypothalamic neurons also express small-conductance, Ca2⫹-activated K⫹ (SK) channels that are sensitive to antagonism by the peptidergic, honeybee toxin apamin [48, 49]. These currents underlie the afterhyperpolarization (AHP) that is observed at the tail end of the action potential or a prolonged depolarizing stimulus [47,49]. In the POA,
we have found that the SK currents in GABAergic neurons are inhibited by - and ␣1-adrenergic receptor activation [50]. Of particular interest is that the ␣1-adrenergic receptor-mediated inhibition of the SK current in these GABAergic neurons is markedly potentiated by a transient exposure to estrogen (Fig. 3). This estrogen-induced enhancement of the coupling of ␣1-adrenergic receptors to their effector SK channels lasts for at least 24 h following systemic steroid administration (Fig. 4). An example of an estrogen-responsive GABAergic neuron identified via combined histofluorescence and in situ hybridization is shown in Fig. 5.
M.J. Kelly et al. / Steroids 67 (2002) 447– 456
451
Fig. 4. Examples of the IAHP that were obtained before (a) and during (b) the bath application of 10 M methoxamine in POA neurons from vehicle- (A) and EB-treated (25 g, 24 h prior; B) animals. C, Composite dose-response curves for methoxamine that were derived from POA neurons in vehicle- (E) and EB-treated (■) animals. The curves were fit via a logistic equation to the experimental data points. Symbols represent means and vertical lines 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 M). *, Denotes a significant main effect of EB on the methoxamine-induced inhibition of the IAHP (multifactorial ANOVA/LSD; P ⬍ 0.05). Reprinted from [50] with permission from the American Society for Pharmacology and Experimental Therapeutics.
3. Estrogen directly modulates ion channel activity through a G protein linkage It has been known for some time that iontophoretically applied estrogen to POA neurons alters their excitability primarily by affecting an inhibition of firing rate [51]. This inhibition is associated with a membrane hyperpolarization (Fig. 6) due to the opening of a K⫹ channel that has also been observed in the amygdala and arcuate nucleus [9,11,
52,53]. This latter region contains GnRH neurons which are particularly responsive to this estrogenic inhibition [9,11]. Estrogen also rapidly inhibits voltage-gated Ca2⫹ channels in medium spiny GABAergic neurons of the basal ganglia, an effect mediated via a pertussis toxinsensitive G protein [10]. Interestingly, the estrogen-induced augmentation of kainate currents in hippocampal CA1 pyramidal cells also involves an intervening G protein [3]. Thus, estrogen apparently is capable of relatively direct modulation of ion channel activity through a G
452
M.J. Kelly et al. / Steroids 67 (2002) 447– 456
Fig. 5. A, Photomicrograph of the biocytin–streptavidin–CY3 fluorescent labeling of the 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. 3B. D, An overlay of the fluorescent labeling in C, and the hybridization signal that clearly illustrates the double-labeling for GAD65. Reprinted from [50] with permission from the American Society for Pharmacology and Experimental Therapeutics.
protein, as well as more subtle modulation of G proteincoupled receptor/K⫹ channel coupling.
4. Significance As mentioned at the outset, estrogen primarily serves to regulate the output of GnRH from the mediobasal hypothalamus, and in doing so controls reproductive status through the hypothalamic–pituitary– gonadal axis. The work presented here sheds meaningful insight into the cellular mechanisms by which estrogen exerts negative feedback over the female reproductive cycle. First of all, estrogen negatively modulates the coupling of the -opioid and GABAB receptors to their effector K⫹ channel in -endorphin and GABAergic neurons, respectively. This uncoupling requires the activation of a hierarchal, intracellular phosphorylation cascade involving both PKC and PKA pathways. PKA also appears to be involved in the estrogen-induced augmentation of kainate-induced currents in hippocampal pyramidal cells [3]. PKA-dependent phosphorylation of the -opioid receptor hinders its association with Gi/o [54,55]. However, it remains to be seen if the -opioid receptor is indeed the
phosphorylation substrate that renders it less apt to couple to its GIRK channel. Alternative targets include Gi/o subunits, GIRK channels and protein regulators of G protein signalling. Future studies will endeavor to elucidate how the phosphorylation affects this uncoupling in -endorphin neurons. -Opioid and GABAB receptors serve as autoreceptors in their respective -endorphin and hypothalamic GABAergic neurons [32,45,46]. The fact that estrogen uncouples these autoreceptors from their GIRK channel implies that the steroid decreases the autocrine, somatodendritic inhibition of these cells, thereby increasing the release of these inhibitory neurotransmitters. Indeed, estrogen rapidly increases extracellular GABA concentrations in the POA as measured by push/pull perfusion and microdialysis [27,56]. Moreover, estrogen markedly potentiates the negative coupling between the ␣1-adrenergic receptor and its effector SK channel in POA GABAergic neurons, which serves to facilitate the excitation caused by ascending noradrenergic input onto these cells [22,50]. Coupled with the attenuated GABAB receptor-mediated autoinhibition of these GABAergic neurons, it stands to reason that estrogen would dramatically increase the firing rate of these neurons during
M.J. Kelly et al. / Steroids 67 (2002) 447– 456
453
Fig. 6. A, A current clamp recording of a POA neuron showing the reversible hyperpolarization (7 mV) elicited by 100 nM E2. The resting potential of this cell was ⫺62 mV. The downward deflections represent voltage responses to intermittent current pulses (not shown) used to monitor input resistance. The break in the trace represents the time necessary to complete a post-drug I/V relationship (⬇ 3 min). B, An I/V plot derived from the cell in A that illustrates the effect of bath applied E2. This neuron exhibited a reversal potential for the E2 response of ⫺76 mV, which is close to the equilibrium potential for K⫹, and a ⌬g of 5.68 nS measured between ⫺60 and ⫺80 mV. Reprinted from [60] with permission from Academic Press.
negative feedback. Given that both -endorphin and hypothalamic GABAergic neurons synapse onto GnRH neurons [57,58], the collective modulation by estrogen of K⫹ channel activity in these cells would greatly enhance the inhibitory tonus impinging on GnRH neurons. Finally, estrogen itself can hyperpolarize and thereby inhibit GnRH neurons by activating a K⫹ channel. While it has long been thought that GnRH neurons do not contain estrogen receptors [15–17], recent evidence indicates that these neurons express the  isoform of the estrogen receptor [12–14,19]. In addition, both the estrogen-induced augmentation of kainate currents in hippocampal CA1 pyramidal cells and the inhibition of Ca2⫹ currents in medium spiny GABAergic neurons appear to involve G proteins [3,10]. It is therefore conceivable that estrogen could interact with a membrane-associated receptor to induce a hyperpolarization of GnRH neurons by way of a G protein. Thus, there are at least three distinct mechanisms through which estrogen can exert negative feedback on GnRH output. The dynamics of
these cellular mechanisms are illustrated in schematic shown in Fig. 7.
Acknowledgments The experiments from the authors’ laboratory described in this review were supported by Public Health Service Grants NS 35944, NS 38809, DA 05158 and DA 00192 (Research Scientist Development Award to MJK). The authors would like to recognize Ms. Martha A. Bosch and Mr. Barry R. Naylor for their expert technical contributions to these studies.
References [1] Kelly MJ, Wagner EJ. Estrogen modulation of G-protein-coupled receptors. Trends Endocrinol Metab 1999;10:369 –74.
454
M.J. Kelly et al. / Steroids 67 (2002) 447– 456
Fig. 7. Schematic representation illustrating the central mechanisms of estrogen-induced negative feedback on the mammalian female reproductive axis. During negative feedback estrogen uncouples somatodendritic GABAB (GABAB-R) and -opioid (MOR) receptors from their effector K⫹ channels in GABAergic and -endorphin neurons [1]. [Note: Although the GABAB and -opioid receptors couple to the same population of GIRK’s in both GABAergic and -endorphin neurons, only the autoreceptors are illustrated.] This leads to a decreased autoinhibition and an increased inhibitory tone onto GnRH neurons. Additional negative feedback is provided by estrogen-sensitive, ascending noradrenergic input to GABAergic neurons, in which the steroid markedly potentiates the ␣1-adrenergic receptor (␣1R)-mediated inhibition of an apamin-sensitive SK current [2]. This results in a further increase of GABAergic neuronal excitability. Finally, estrogen can affect a direct inhibition of GnRH neurons via a G protein coupling to a K⫹ channel [3]. These three distinct mechanisms by which estrogen modulates K⫹ channel activity accounts for, at least in part, the negative feedback effect of estrogen on the hypothalamic– pituitary axis. [2] Wagner EJ, Moore KE, Lookingland KJ. Sexual differences in N-methyl-D-aspartate receptor-mediated regulation of tuberoinfundibular dopaminergic neurons in the rat. Brain Res 1993;611: 139 – 46. [3] Gu Q, Moss RL. 17-Estradiol potentiates kainate-induced currents via activation of the cAMP cascade. J Neurosci 1996;16:3620 –9. [4] Demotes-Mainard J, Arnauld E, Vincent JD. Estrogens modulate the responsiveness of in vivo recorded striatal neurons to iontophoretic application of dopamine in rats: Role of D1 and D2 receptor activation. J Neuroendocrinol 1990;2:825–32. [5] Kelly MJ, Loose MD, Rønnekleiv OK. Estrogen suppresses -opioidand GABAB-mediated hyperpolarization of hypothalamic arcuate neurons. J Neurosci 1992;12:2745–50. [6] Wagner EJ, Manzanares J, Moore KE, Lookingland KJ. Neurochemical evidence that estrogen-induced suppression of -opioid-receptormediated regulation of tuberoinfundibular dopaminergic neurons is prolactin-independent. Neuroendocrinology 1994;59:197–201. [7] Lagrange AH, Rønnekleiv OK, Kelly MJ. The potency of -opioid hyperpolarization of hypothalamic arcuate neurons is rapidly attenuated by 17-estradiol. J Neurosci 1994;14:6196 –204.
[8] Lagrange AH, Wagner EJ, Rønnekleiv OK, Kelly MJ. Estrogen rapidly attenuates a GABAB response in hypothalamic neurons. Neuroendocrinology 1996;64:114 –23. [9] Lagrange AH, Rønnekleiv OK, Kelly MJ. Estradiol-17, and -opioid peptides rapidly hyperpolarize GnRH neurons: A cellular mechanism of negative feedback. Endocrinology 1995;136:2341– 4. [10] Mermelstein PG, Becker JB, Surmeier DJ. Estradiol reduces calcium currents in rat neostriatal neurons via a membrane receptor. J Neurosci 1996;16:595– 604. [11] Kelly MJ, Rønnekleiv OK, Eskay RL. Identification of estrogenresponsive LHRH neurons in the guinea pig hypothalamus. Brain Res Bull 1984;12:399 – 407. [12] Skynner MJ, Sim JA, Herbison AE. Detection of estrogen receptor ␣ and  messenger ribonucleic acids in adult gonadotropin-releasing hormone neurons. Endocrinology 1999;140:5195–201. [13] Butler JA, Sjo¨ berg M, Coen CW. Evidence for oestrogen receptor ␣-immunoreactivity in gonadotropin-releasing hormone-expressing neurones. J Neuroendocrinol 1999;11:331–5. [14] Hrabovszky E, Shughrue PJ, Merchenthaler I , et al. Detection of estrogen receptor- messenger ribonucleic acid and 125I-estrogen
M.J. Kelly et al. / Steroids 67 (2002) 447– 456
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22] [23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31] [32]
binding sites in luteinizing hormone-releasing hormone neurons of the rat brain. Endocrinology 2000;141:3506 –9. Watson RE, Langub MC, Landis JW. Further evidence that most luteinizing hormone-releasing hormone neurons are not directly estrogen-responsive: simultaneous localization of luteinizing hormonereleasing hormone and estrogen receptor immunoreactivity in the guinea-pig brain. J Neuroendocrinol 1992;4:311–7. Herbison AE, Horvath TL, Naftolin F, Leranth C. Distribution of estrogen receptor-immunoreactive cells in monkey hypothalamus: relationship to neurones containing luteinizing hormone-releasing hormone and tyrosine hydroxylase. Neuroendocrinology 1995;61:1– 10. Sullivan KA, Witkin JW, Ferin M, Silverman A-J. Gonadotropinreleasing hormone neurons in the rhesus macaque are not immunoreactive for the estrogen receptor. Brain Res 1995;685:198 –200.. Simonian SX, Spratt DP, Herbison AE. Identification and characterization of estrogen receptor ␣-containing neurons projecting to the vicinity of the gonadotropin-releasing hormone perikarya in the rostral preoptic area of the rat. J Comp Neurol 1999;411:346 –58. Herbison AE, Skynner MJ, Sim JA. Lack of detection of estrogen receptor- ␣ transcripts in mouse gonadotropin-releasing hormone neurons. Endocrinology 2001;142:493. Morrell JI, McGinty JF, Pfaff DW. A subset of -endorphin- or dynorphin-containing neurons in the medial basal hypothalamus accumulates estradiol. Neuroendocrinology 1985;41:417–26. Leranth C, MacLusky NJ, Brown TJ, Redmond DE, Naftolin F. Transmitter content and afferent connections of estrogen-sensitive progestin receptor containing neurons in the primate hypothalamus. Neuroendocrinology 1992;55:667– 82. Herbison AE. Estrogen regulation of GABA transmission in rat preoptic area. Brain Res Bull 1997;44:321– 6. Ferin M, Van Vugt D, Wardlaw S. The hypothalamic control of the menstrual cycle and the role of endogenous opioid peptides. Recent Prog Horm Res 1984;40:441– 85. Mitsushima D, Marzban F, Luchansky LL, et al. Role of glutamic acid decarboxylase in the prepubertal inhibition of the luteinizing hormone releasing hormone release in female rhesus monkeys. J Neurosci 1996;16:2563–73. Akema T, Chiba A, Kimura F. On the relationship between noradrenergic stimulatory and GABAergic inhibitory systems in the control of luteinizing hormone secretion in female rats. Neuroendocrinology 1990;52:566 –72. Seltzer AM, Donoso AO. Restraining action of GABA on estradiolinduced LH surge in the rat: GABA activity in brain nuclei and effects of GABA mimetics in the medial preoptic nucleus. Neuroendocrinology 1992;55:28 –34. Jarry H, Leonhardt S, Wuttke W. The inhibitory effect of -endorphin on LH release in ovariectomized rats does not involve the preoptic GABAergic system. Exp Clin Endocrinol Diabetes 1995;103:317–23. Masotto C, Wisniewski G, Negro-Vilar A. Different ␥-aminobutyric acid receptor subtypes are involved in the regulation of opiate-dependent, and independent luteinizing hormone-releasing hormone secretion. Endocrinology 1989;125:548 –53. Brann DW, Zamorano PL, Putnam-Roberts CD, Mahesh VB. Gamma-aminobutyric acid-opioid interactions in the regulation of gonadotropin secretion in the immature female rat. Neuroendocrinology 1992;56:445–52. Loose MD, Kelly MJ. Opioids act as -receptors to hyperpolarize arcuate neurons via an inwardly rectifying potassium conductance. Brain Res 1990;513:15–23. Jan LY, Jan YN. Voltage-gated and inwardly rectifying potassium channels. J Physiol (Lond) 1997;505:267– 82. Kelly MJ, Loose MD, Ronnekleiv OK. Opioids hyperpolarize -endorphin neurons via -receptor activation of a potassium conductance. Neuroendocrinology 1990;52:268 –75.
455
[33] Narita M, Feng Y, Makimura M, Hoskins B, Ho IK. A protein kinase inhibitor, H-7, inhibits the development of tolerance to opioid antinociception. Eur J Pharmacol 1994;271:543–5. [34] Mestek A, Hurley JH, Bye LS , et al. The human opioid receptor: Modulation of functional desensitization by calcium/calmodulin-dependent protein kinase and protein kinase C. J Neurosci 1995;15: 2396 – 406. [35] Wang L, Medina VM, Rivera M, Gintzler AR. Relevance of phosphorylation state to opioid responsiveness in opiate naive and tolerant/dependent tissue. Brain Res 1996;723:61–9. [36] Wagner EJ, Rønnekleiv OK, Kelly MJ. Protein kinase A maintains cellular tolerance to -opioid receptor agonists in hypothalamic neurosecretory cells with chronic morphine treatment: Convergence upon a common pathway with estrogen in modulating -opioid receptor/ effector coupling. J Pharmacol Exp Ther 1998;285:1266 –73. [37] Aronica SM, Katzenellenbogen BS. Stimulation of estrogen receptor mediated transcription and alteration in the phosphorylation state of the rat uterine estrogen receptor by estrogen, cyclic adenosine monophosphate, and insulin-like growth factor-1. Mol Endocrinol 1993;7: 743–52. [38] Nakhla AM, Khan MS, Romas NP, Rosner W. Estradiol causes the rapid accumulation of cAMP in human prostate. Proc Natl Acad Sci 1994;91:5402–5. [39] Zhou Y, Watters JJ, Dorsa DM. Estrogen rapidly induces the phosphorylation of the cAMP response element binding protein in rat brain. Endocrinology 1996;137:2163– 6. [40] Gu G, Rojo AA, Zee MC, Yu J, Simerly RB. Hormonal regulation of CREB phosphorylation in the anteroventral periventricular nucleus. J Neurosci 1996;16:3035– 44. [41] Watters JJ, Campbell JS, Cunningham MJ, Krebs EG, Dorsa DM. Rapid membrane effects of steroids in neuroblastoma cells: effects of estrogen on mitogen activated protein kinase signalling cascade and c-fos immediate early gene transcription. Endocrinology 2001;138: 4030 –3. [42] Kelly MJ, Lagrange AH, Wagner EJ, Rønnekleiv OK. Rapid effects of estrogen to modulate G protein-coupled receptors via activation of protein kinase A and protein kinase C pathways. Steroids 1999;64: 64 –75. [43] Ansonoff MA, Etgen AM. Estradiol elevates protein kinase C catalytic activity in the preoptic area of female rats. Endocrinology 1998;139:3050 – 6. [44] Loose MD, Rønnekleiv OK, Kelly MJ. Neurons in the rat arcuate nucleus are hyperpolarized by GABA(B) and -opioid receptor agonists: Evidence for convergence at a ligand-gated potassium conductance. Neuroendocrinology 1991;54:537– 44. [45] Wagner EJ, Bosch MA, Kelly MJ, Rønnekleiv OK. A powerful GABAB receptor-mediated inhibition of GABAergic neurons in the arcuate nucleus. Neuroreport 1999;10:2681–7. [46] Wagner EJ, Rønnekleiv OK, Bosch MA, Kelly MJ. Estrogen biphasically modifies hypothalamic GABAergic neuronal function concomitantly with negative and positive control of luteinizing hormone release. J Neurosci 2001;21:2085–93. [47] Wagner EJ, Reyes-Vazquez C, Rønnekleiv OK, Kelly MJ. The role of intrinsic and agonist-activated conductances in determining the firing patterns of preoptic area neurons in the guinea pig. Brain Res 2000; 879:29 – 41. [48] Erickson KR, Rønnekleiv OK, Kelly MJ. Role of a T-type calcium current in supporting a depolarizing potential, damped oscillations, and phasic firing in vasopressinergic guinea pig supraoptic neurons. Neuroendocrinology 1993;57:789 – 800. [49] Kirkpatrick K, Bourque CW. Activity dependence and functional role of the apamin-sensitive K⫹ current in rat supraoptic neurones in vitro. J Physiol (Lond) 1996;494:389 –98. [50] Wagner EJ, Rønnekleiv OK, Kelly MJ. The noradrenergic inhibition of an apamin-sensitive, small conductance Ca2⫹-activated K⫹ channel in hypothalamic GABAergic neurons: pharmacology, estrogen
456
[51]
[52]
[53]
[54]
[55]
M.J. Kelly et al. / Steroids 67 (2002) 447– 456 sensitivity and relevance to the control of the reproductive axis. J Pharmacol Exp Ther 2001;299:21–30. Kelly MJ, Moss RL, Dudley CA. The effects of microelectrophorectically applied estrogen, cortisol and acetylcholine on medial preoptic-septal unit activity throughout the estrous cycle of the female rat. Exp Brain Res 1977;30:53– 64. Kelly MJ, Kuhnt U, Wuttke W. Hyperpolarization of hypothalamic parvocellular neurons by 17-estradiol and their identification through intracellular staining with procion yellow. Exp Brain Res 1980;40:440 –7. Nabekura J, Oomura Y, Minami T, Mizuno Y, Fukuda A. Mechanism of the rapid effect of 17-estradiol on medial amygdala neurons. Science 1986;233:226 – 8. Harada H, Ueda H, Wada Y, Katada T, Ui M, Satoh M. Phosphorylation of -opioid receptors - a putative mechanism of selective uncoupling of receptor - Gi interaction, measured with low-Km GTPase and nucleotide-sensitive agonist binding. Neurosci Lett 1989; 100:221– 6. Harada H, Ueda H, Katada T, Ui M, Satoh M. Phosphorylated -opioid receptor purified from rat brain lacks functional coupling
[56]
[57]
[58]
[59]
[60]
with Gi1, a GTP-binding protein in reconstituted lipid vesicles. Neurosci Lett 1990;113:47–9. Herbison AE, Heavens RP, Dye S, Dyer RG. Acute action of oestrogen on medial preoptic ␥-aminobutyric acid neurons: Correlation with oestrogen negative feedback on luteinizing hormone secretion. J Neuroendocrinol 1991;3:101– 6. Leranth C, MacLusky NJ, Sakamoto H, Shanabrough M, Naftolin F. Glutamic acid decarboxylase-containing axons synapse on LHRH neurons in the rat medial preoptic area. Neuroendocrinology 1985; 40:536 –9. Naftolin F, Leranth C, Horvath TL, Garcia-Segura LM. Potential neuronal mechanisms of estrogen actions in synaptogenesis and synaptic plasticity. Cell Mol Neurobiol 1996;16:213–23. Lagrange AH, Rønnekleiv OK, Kelly MJ. Modulation of G proteincoupled receptors by an estogen receptor that activates protein kinase A. Mol Pharmacol 1997;51:605–12. Kelly MJ, Rønnekleiv OK. Rapid membrane effects of estrogen in the CNS. In: Pfaff DW, editor. Hormones, Brain and Behavior. San Diego: Academic Press (in press).