- Email: [email protected]
Modulation of the GABAA Response in Rat Ventromedial Hypothalamic Neurons by Pregnanolone
Comp. Biochem. Physiol. Vol. 117A, No. 2, pp. 219–226, 1997 Copyright 1997 Elsevier Science Inc.
ISSN 0300-9629/97/$17.00 PII S0300-9629(96)00288-5
Modulation of the GABAA Response in Rat Ventromedial Hypothalamic Neurons by Pregnanolone Tian-Le Xu, Taiichiro Imanishi, and Norio Akaike Department of Physiology, Faculty of Medicine, Kyushu University, Fukuoka 812-82, Japan ABSTRACT. The effects of 5β,3α-pregnanolone (PGN) on γ-aminobutyric acid (GABA)-induced Cl2 current in acutely dissociated rat ventromedial hypothalamic neurons were investigated using the nystatin perforated patch recording mode under voltage-clamp conditions. The PGN at concentrations between 1027 and 1025 M evoked an inward current at a holding potential (VH) of 240 mV. The reversal potential of the PGN-induced current was close to the equilibrium potential of Cl2. PGN potentiated the GABA-induced Cl2 current in a concentration-dependent manner. The facilitatory effect was long lasting and disappeared slowly after being washed out. The effect of PGN on the GABA response was also affected by the pretreatment time of PGN. PGN shifted the concentration–response relationship for GABA to the left with a suppression of the maximal response to GABA, resulting from the rapid inactivation of the GABA response during PGN treatment. Facilitatory interactions were found to exist among GABA, pentobarbital, diazepam, and PGN. Three other PGN isomers were also effective in facilitating the GABA response. However, the isomers containing the 3α-hydroxy configurations potentiated the GABA response much more potently and only these isomers exhibited inward currents themselves. comp biochem physiol 117A;2:219–226, 1997. 1997 Elsevier Science Inc. KEY WORDS. GABAA response, nystatin perforated patch, neurosteroid, pregnanolone, diazepam, pentobarbital, rat, ventromedial hypothalamic neurons
INTRODUCTION Recent discoveries in the field of neuroactive steroids (neurosteroids) have led to a resurgence in the research efforts to better understand the influence of steroids on the membrane receptors of CNS neurons (7,12,26,28,29,35). Although the historical view of steroid action focuses on the genomic effects of hormones, it is now well known that many steroids can modulate the excitability of CNS neurons via their direct action on ligand-gated (10,26,44,45, 47) and voltage-dependent (14) ion channels. Furthermore, there is also evidence to support the idea of the local synthesis of some neurosteroids in the brain (12,21,23,26,29,35). The GABAA receptor–Cl2 channel complex is a major target for neurosteroid action (1,34,43), and several neurosteroids have been shown to be potent stereoselective allosteric modulators of the GABAA receptors. Thus, the 3-hydroxy ring-A reduced metabolites of progesterone and deoxycorticosterone, such as 5α,3α-pregnanolone, 5β,3αpregnanolone, and 5α,3α-THDOC, produce a potent enAddress reprint requests to: N. Akaike, Ph.D., Department of Physiology, Faculty of Medicine, Kyushu University, 812-82, Japan. Tel. 81-92-6411151, ext. 3332; Fax 81-92-633-6748; E-mail: [email protected]. kyushu-u.ac.jp *Present address: Department of Anatomy, The Fourth Military Medical University, Xi’an, Shannxi 710032, People’s Republic of China. Received 6 February 1996; accepted 1 August 1996.
hancement of GABAA receptor-mediated response in vitro (17,24,27,44), whereas they also induce behavioral sedation in vivo (8,9,24). At higher concentrations, neurosteroids themselves also exhibited a GABA-mimetic action (10,24). Barbiturates and benzodiazepines have long been shown to facilitate the GABAA response (1–4,18,48), and their pharmacological significance is quite clear. Considerable data have demonstrated that both barbiturates and benzodiazepines facilitate the affinity of GABA to GABAA receptors as the main mechanism of their potentiating effects (13,33). The presence of a steroid modulatory site on the GABAA receptor–Cl2 channel complexes might thus provide a novel locus for the potential pharmacological manipulation of GABAA receptor-mediated neurotransmission. In fact, the endogenous progesterone metabolites, 5α,3αpregnanolone (8,24) and 5β,3α-pregnanolone (24), have been shown to exhibit a potent anticonvulsant activity against seizures induced by GABAA receptor antagonists such as pentylenetetrazol, bicuculline, and picrotoxin. The anticonvulsant profile of these neurosteroids resembled that of the benzodiazepine (24). Although several studies have reported the actions of neurosteroids on the GABA-related responses (1,25,34,43,44), few systematic electrophysiological studies on both the facilitation of the GABAA response and GABA-mimetic action by neurosteroids have apparently been performed. Whether the activation of the puta-
T.-L. Xu et al.
220
tive steroid site(s) results in electrophysiological and pharmacological effects similar to those produced by benzodiazepines or barbiturates also still remains to be determined. On the other hand, an important central nervous site for steroid modulation of physiological functions such as food intake and reproduction is the ventromedial hypothalamus (VMH) (6). Therefore, in the present study, we investigated how 5β,3α-pregnanolone (PGN) modulates the GABAA receptor-mediated Cl2 response in acutely dissociated rat ventromedial hypothalamic (VMH) neurons using a nystatin perforated patch recording configuration. MATERIALS AND METHODS Preparation VMH neurons were acutely dissociated as described elsewhere (41). Briefly, 2-week-old Wistar rats were decapitated under pentobarbital-Na anesthesia. Slices at a thickness of 400 µm containing the VMH region thus were obtained using a vibratome tissue slicer (Dosaka, DSK-1000). The slices were preincubated in a well-oxygenated incubation solution (see below for the composition) for 50 min at room temperature (22–25°C). Thereafter, the slices were treated enzymatically in well-oxygenated incubation solution containing 1 mg/6 ml pronase for 20 min at 31°C and then were successively exposed to 1 mg/6 ml thermolysin for another 15 min at the same temperature. After the enzyme treatment, the slices were kept in an enzyme-free incubation solution for 1 hr. Next, a portion of the VMH region was micropunched out with an electrolytically polished injection needle and was transferred into a culture dish filled with standard external solution (see below). The neurons were mechanically dissociated with fire-polished Pasteur pipettes under the visual guidance of a phase contrast microscope (Nikon, TMS-1). Dissociated neurons adhered to the bottom of the dish within 20 min, thus allowing us to conduct the electrophysiological studies. Only neurons that retained their original morphological features, such as the dendritic processes, were used for the experiments. Solutions The ionic composition of the incubation solution was (mM): NaCl 124, NaHCO3 24, KCl 5, KH2PO4 1.2, CaCl2 2.4, MgSO4 1.3, and glucose 10, aerated with 5% CO2/95% O2 to a final pH of 7.4. The normal external standard solution was (mM): NaCl 150, KCl 5, CaCl2 2, MgCl2 1, HEPES 10, and glucose 10. The pH was adjusted to 7.4 with Trisbase. The patch pipette solution for nystatin perforated patch recording was (mM): KCl 150, HEPES 10. The pH was adjusted to 7.2 with Tris-base. A nystatin stock solution dissolved in acidified methanol at a concentration of 10 mg/ ml was prepared and stored at 220°C. The stock solution was added to the patch pipette solution in a final nystatin concentration of 400 µg/ml just before use. When the current–voltage (I-V) relationship for the PGN-induced cur-
rent was examined using ramp pulses, 1 µM tetrodotoxin (TTX) and 10 µM CdCl2 were added to the standard external solution and then 150 mM CsCl was used in the patch pipette solution instead of KCl. Whole-Cell Recording All electrical measurements were carried out in a wholecell mode using the nystatin perforated patch recording technique (2,19) at room temperature (22–25°). The patch pipettes were pulled from glass capillaries with an outer diameter of 1.5 mm (Narishige) on a two-stage puller (Narishige, PB-7). The resistance between the recording electrode filled with pipette solution and the reference electrode was 4–6 MΩ. The patch pipette was positioned on a neuron using a hydraulic micromanipulator (Narishige, MO-102). The current and voltage were measured with a patch clamp amplifier (Nihon Koden, CEZ-2300), filtered at 1 kHz (NF Electronic Instruments, FV-665), and then was monitored on both a storage oscilloscope (Iwatsu Electronic, 5100A) and a pen recorder (Nippondenki San-ei, Recti-Horiz8K21). The data were simultaneously stored on videotape after digitalization with a PCM processor (Nihon Koden, PCM 501 ESN). A function generator (NF Electronic Instruments, FG-121B) was used to deliver the ramp pulses for the current–voltage experiments. The membrane potential was maintained at 240 mV throughout the experiment, except for the ramp pulse studies. All measurements were started after the stabilization of either the PGN or GABA responses (15–25 min after cell attachment). Drugs Thermolysin, nystatin, 5β,3α-pregnanolone (PGN), 5α,3α-pregnanolone (5α,3α-PGN), 5β,3β-pregnanolone (5β,3β-PGN), and 5α,3β-pregnanolone (5α,3β-PGN) were purchased from Sigma. The other drugs included: pronase (Calbiochem), GABA (Kasei), diazepam (DZP) (Ishizu), pentobarbital-Na (PB) (Ishizu), and tetrodotoxin (TTX) (Sankyo). Stock solutions of neurosteroids and diazepam were prepared in dimethyl sulfoxide (DMSO) and diluted to final concentrations in the external standard solution. The final concentrations of DMSO were always less than 0.1%; they did not induce any ionic current and they also had no effect on the GABA or PGN responses at the concentrations used. The drug solutions were applied using the ‘‘Y-tube’’ method (32) throughout the study. This system allows for the complete exchange of external solution surrounding a neuron within 10–20 msec. RESULTS PGN-Induced Inward Currents PGN evoked inward currents in all VMH neurons tested at a holding potential (VH) of 240 mV. The currents became detectable at a concentration of 1027 M and thereafter in-
Pregnanolone and GABAA Response
221
of two curves obtained before (a) and during (b) PGN application gave a reversal potential of the PGN response of 22.0 mV, which was close enough to the theoretical equilibrium potential of Cl2 (ECl) calculated from the given external and internal Cl2 concentrations (161 and 150 mM, respectively) by the Nernst equation, thus indicating that the PGN-induced current passes through the Cl2 channels. Modulation of GABA Response by PGN The effects of PGN at various concentrations (1028 to 1025 M) on the 3 3 1026 M GABA-induced Cl2 current were quantitatively studied. The currents obtained at various concentrations of PGN, either alone or with 3 3 1026 M GABA, are illustrated in Fig. 2A. PGN enhanced the GABA response, which was apparent with a concentration of PGN as low as 3 3 1028 M, in which the PGN concentration elicited no response. It is also evident that the inactivation of the GABA response became faster as the PGN concentrations increased. The potentiation ratio of the GABA response by PGN was calculated by a formula indicated
FIG. 1. PGN-induced inward currents in dissociated VMH neurons. [A(a)] The inward currents induced by PGN of various concentrations at a holding potential (VH) of 240 mV. [A(b)] Concentration–response relationship for PGNinduced current. All responses were normalized to the peak current amplitude induced by 1025 M GABA alone. Each point shows the mean of six neurons and the vertical bars show 6 SEM. (B) A representative record to show the current–voltage relationship for PGN-induced current. Depolarized triangular voltage ramp pulses of 60 mV, followed by a mirror-image hyperpolarizing ramp, were applied before (a) and during (b) PGN application (see inset). MP represents the membrane potential (mV).
creased in a concentration-dependent manner. As shown in Fig. 1A(a), the currents induced by PGN did not show any inactivation at concentrations below 1026 M, even in the case when the application lasted up to 2 min (data not shown), although a slight inactivation could be observed from concentrations beyond 3 3 1026 M. Because of limitations in the PGN solubility, no concentrations over 1025 M could be evaluated. To identify the ion that is responsible for the PGN-induced current, the current–voltage (I-V) relationship for PGN was thus elucidated using ramp pulses in the external solution containing TTX, CdCl 2, and CsCl, and in the pipette solution containing CsCl (Fig. 1B). TTX and CdCl2 did not affect the PGN response. The intercept
FIG. 2. Facilitatory effect of PGN on GABA-induced Cl2
current. (A) Inward currents induced by PGN alone (a) or by simultaneous application of PGN and 3 3 1026 M GABA (b). All traces were obtained from the same neuron. (B) Concentration–response relationship for PGN-mediated augmentation of GABA response. Ordinate shows enhancement ratio of GABA response by PGN. Each point is the mean of six measurements. Vertical bars show 6 SEM.
T.-L. Xu et al.
222
FIG. 3. Recovery time courses of the potentiation of GABA
responses by PGN. (A) Typical examples of the potentiation of GABA response and the recovery after wash out. All data were form the same neuron. (B) Summary of the recovery time courses of 3 3 1026 M GABA responses in the presence of PGN of different concentrations. All responses were normalized to the peak amplitude induced by 3 3 1026 M GABA alone. Each point is the average of four to six measurements. Vertical bars show 6 SEM.
along the y-axis in Fig. 2B because PGN itself induced an inward current [see also Fig. 2A(a)]. The GABA response was augmented in a concentration-dependent manner by PGN. As shown in Fig. 3, the augmentatory action of PGN on the GABA response was reversible, although the residual facilitatory effect slowly disappeared after the removal of PGN. The recovery time courses of the PGN actions of different concentrations on the GABA response are illustrated in Fig. 3B. It is clearly shown that a longer recovery time was required when higher concentrations of PGN were applied. The facilitatory effect of PGN on the GABA response was also affected by the pretreatment time of PGN of different concentrations. In Fig. 4, the PGN of a low concentration (3 3 1028 M) required a longer pretreatment time to evoke the maximal potentiation. The PGN at higher concentrations induced a maximal potentiation even with only 10 sec pretreatment, but the potentiation ratio decreased with the prolongation of the pretreatment time. Thus, the facilitatory effect of PGN on the GABA response depends not only on the concentrations of each drug but also on the duration of PGN pretreatment. Figure 5 shows the concentration–response relationship for GABA with or without PGN. PGN shifted the concentration–response curve for GABA to the left in a concentration-dependent fashion (Fig. 5B). At the same time,
FIG. 4. Effect of various pretreatment time of PGN on the potentiation of GABA response. (A) The potentiation of GABA response by 3 3 1028 (a) and 3 3 1026 M (b) PGN applied with various pretreatment time. (B) Pretreatment effects of PGN at different concentrations on GABA response. All responses were normalized to the peak current induced by 3 3 1026 M GABA alone. Each point represents the average value from six neurons. The vertical bars are 6 SEM.
PGN also suppressed the maximal response to GABA. As shown in Fig. 5A, the inactivation of the current induced by GABA of the higher concentrations was accelerated more in the presence of PGN. Facilitatory Interactions Among GABA, PB, DZP, and PGN Considering the possible involvement of either the barbiturates or benzodiazepines binding site on the effect of PGN, the facilitatory interactions among GABA, PB, DZP, and PGN were examined. As shown in Fig. 6A, the simultaneous treatment of GABA or PB at concentrations eliciting slight responses and PGN exhibited a marked potentiation. On the other hand, DZP did not induce any currents itself, whereas it potentiated the PGN response to a lesser extent than that of GABA or PB. The inward current evoked by the coapplication of low concentrations of GABA, PB, and DZP could still be enhanced by adding PGN (Fig. 6B).
Pregnanolone and GABAA Response
223
FIG. 6. Facilitatory interactions among GABA, pentobarbital
(PB), diazepam (DZP), and PGN. (A) Currents were elicited by 3 3 1027 M GABA (a), 3 3 1025 M PB (b), or 3 3 1026 M DZP (c) alone or combined with 1027 M PGN. (B) Simultaneous treatment of 1027 M GABA, 1025 M PB, and 1026 M DZP in the presence or absence of 1027 M PGN. Similar results in both (A) and (B) were obtained from four to six neurons.
FIG. 5. Concentration–response relationship for GABA in
the presence of PGN. (A) Typical currents induced by GABA alone (a) or by the combination of GABA and 3 3 1027 M PGN (b). (B) Concentration–response relationships for GABA with or without PGN of different concentrations. All responses were normalized to the peak response induced by 1025 M GABA alone (*). Each point shows the mean of four to six neurons.
Enhancement of GABA Response by PGN Isomers We next investigated how the GABA response is differently modified by PGN isomers [i.e., PGN (5β,3α-PGN), 5α,3αPGN, 5β,3β-PGN, 5α,3β-PGN] (Fig. 7A). Because PGN facilitated the GABA response at lower concentrations and also induced the GABA-mimetic response at higher concentrations, two representative concentrations of each isomer were used for comparison with the PGN actions. As shown in Fig. 7B, all steroids tested were effective in facilitating the GABA response at both concentrations. However, the isomers containing a 3α-hydroxy configuration potentiated the GABA response much more potently and only
these isomers themselves exhibited a GABA-mimetic effect. DISCUSSION Neurosteroids have been shown to interact potently with the GABA receptor–Cl2 channel complex in brain slice preparations (43), cultured neurons from the hippocampus (24) and the spinal cord (44), and cultured chromaffin cells (10). The present study confirmed that PGN, one of the 3α-hydroxy ring-A-reduced metabolites of progesterone, potentiated the GABA-induced Cl2 currents and also exhibited GABA-mimetic current in acutely dissociated VMH neurons. The PGN-induced inward current was concentration dependent and showed little inactivation. The examination of the I-V relationship for PGN response revealed the PGN response to be carried by Cl2. A previous study reported that bicuculline, a GABAA receptor antagonist, blocked PGN response (10), which also supports the fact that Cl2 passes through the GABA receptor–Cl2 channel complex during PGN treatment. It is unlikely that the response is produced by the interaction of PGN and GABA
224
FIG. 7. Structure–activity relationship for PGN isomers to enhance GABA response. (A) Chemical structures of PGN isomers: a, down; b, up. (B) Typical examples showing effects of four isomers of PGN at 3 3 1028 M (a) or 1026 M (b) on 3 3 1026 M GABA response. Similar results were obtained from four neurons. [B(a)] and [B(b)] were from different neurons.
released from neighboring cells, because the neurons we used were freshly dissociated. Therefore, the present results reinforced the view that this steroid directly activates the GABAA receptor to open the associated Cl2 channel. The current amplitudes observed under our experimental conditions were apparently larger than those observed in the cultured hippocampal neurons (24), chromaffin cells (10), recombinant GABAA receptor subunits expressed in 293 cells (36), and isolated frog sensory neurons (1), in which the induction of PGN currents was observed only at steroid concentrations of 3 3 1027 or higher. Moreover, the long latency in the activation of the PGN current reported for cultured hippocampal neurons (24) was not observed in the present study. However, these findings may result from differences in the preparations and/or regional differences of CNS used (16,22). The facilitatory effect of PGN on the GABA-induced Cl2 current was similar to the previous results observed in both electrophysiological (1,10,44) and binding studies
T.-L. Xu et al.
(30,31). Kokate et al. reported that the coapplication of GABA and PGN induced a complex Cl2 current exhibiting multiple peaks, which may reflect the multiple binding sites of PGN on GABAA receptor (24). However, in the present study, such multiple peaks were not observed during PGNGABA coapplication or coapplication PGN and GABA following PGN pretreatment. In addition, we have also shown that the potentiation action of PGN was long lasting and was also affected by the pretreatment time of PGN. Similar phenomena have been found in the augmentation of the GABA response by DZP in frog sensory neurons (48). In the presence of PGN, the concentration–response curve for GABA shifted to the left concomitant with the suppression of the maximal response to GABA. As shown in Fig. 5A, the inactivation of the GABA-induced Cl2 current accelerated more in the presence of PGN. Therefore, the suppression of the maximal response might be due to the rapid inactivation of the GABA response, although the effect of neurosteroids on GABA receptor desensitization is still unclear. On the other hand, another possibility may also exist. The results from a single channel analysis showed that in the presence of highest concentration (1025 M) of steroids, the prolongation of the average open time of GABAinduced single Cl2 channel currents decreased, suggesting the concentration-dependent blockage of GABA receptor– Cl2 channel complexes by steroids (44). Neurosteroids are normally found in neuronal tissue at rather low concentrations, but their levels increase dramatically in response to stress and during certain stages of pregnancy and/or the menstrual cycle in females. The plasma levels of 3α-hydroxy metabolites of progesterone reach 30 ng/ml (approximately 1027 M) during the third trimester of pregnancy and thereafter remain at an elevated level for a few months (35). The levels in the rat brain of these 3αhydroxy steroids measured after swimming stress are approximately 3–10 ng/g tissue (1028 –3 3 1028 M) (38). Although it is now uncertain whether or not the GABAmimetic action of PGN is clinically relevant, its augmentatory action on the GABA response is pharmacologically and clinically even more essential, because this action was observed at concentrations 10 times less than that needed to induce GABA-mimetic action and therefore is within the range of clinically relevant concentrations (35). It is already known that specific binding sites exist for some barbiturates and benzodiazepines on GABA receptor– Cl2 channel complex that closely link each other to the GABA binding site. Our previous study demonstrated that both PB (1,3–5) and DZP (1,18,48) potentiate the GABA response. Therefore, it is possible that PGN potentiated the GABA response by interacting with either the barbiturates or the benzodiazepines binding site. In the present study, a marked facilitatory interaction was observed among GABA, PB, DZP, and PGN. Both PB and DZP elicited an enlarged inward current in the presence of PGN as well as in the case of GABA. Furthermore, the current induced by the coapplication of GABA, PB, and DZP was also potentiated
Pregnanolone and GABAA Response
with same procedure. Although not conclusive, such results suggested that the modulatory site of PGN on the GABA receptor is probably neither a barbiturate site nor a benzodiazepine site. Previous studies have also suggested that steroids modulate the GABA receptor–Cl2 channel complex at a site distinct from barbiturates and benzodiazepines (10,16,17,30,42). The structure–activity relationship obtained in the present study for the potentiation of the GABA-activated Cl2 current in the presence of either 3α- or 3β-hydroxy isomers was similar to that in the cultured rat hippocampal neurons by Kokate et al. (24), but was different from that reported by Harrison et al. (17), who showed that 3β-hydroxysteroids had little effect at the GABAA receptor. Because 5β,3βPGN has been proposed to be a specific antagonist of the neurosteroid facilitatory site on the GABAA receptor (37), it is possible that 3β-hydroxy steroids may also play a physiological role in the modulation of GABAA receptor, although whether or not 3β-pregnanolones are produced in the CNS remains uncertain. However, it is apparent from both our present findings and our previous studies (17,24) that isomers having 3α-hydroxy configuration potentiated the GABA response much more potently. In addition, the present study also revealed that only these 3α-isomers themselves produced an inward current at higher concentrations. In conclusion, the present study demonstrated that PGN has a dual effect on the GABAA receptor–Cl2 channel complexes, the potentiation of GABA-induced current and, at relatively higher concentrations, the direct induction of GABA-mimetic current. In addition, this endogenous 3α-hydroxy ring A-reduced steroid metabolite interacts with the GABAA receptors at a unique recognition site that is different from both the benzodiazepine and barbiturate sites and affects the Cl2 channel activity. GABAA receptors have been well established to modulate the anxiolytic, sedative–hypnotic, anticonvulsant, antinociceptive, and anesthetic effects of benzodiazepines and barbiturates (11,20,35). Some neurosteroids, like other GABA-potentiating drugs, also have anxiolytic (46), sedative–hypnotic (8), anticonvulsant (9,24), antinociceptive (15), and anesthetic activities (39,40,42). The low intrinsic toxicity and lack of hormonal properties of these compounds together with the findings obtained herein therefore suggest that it may in the future be possible to develop novel synthetic steroids as therapeutic agents. We wish to thank Dr. K. Morita and Prof. J. S. Li for their support and encouragement throughout the course of these experiments.
References 1. Akaike, N.; Harata, N. GABAA receptor–chloride channel complex. In: Narahashi, T. (ed). Ion Channels of Excitable Cells. San Diego, CA: Academic Press, Inc; 1994:148– 165. 2. Akaike, N.; Harata, N. Nystatin perforated patch recording
225
3.
4.
5. 6.
7. 8. 9. 10.
11.
12. 13. 14.
15. 16.
17.
18.
19. 20.
21.
and its applications to analyses of intracellular mechanisms. Jpn. J. Physiol. 44:433–473;1994. Akaike, N.; Hattori, K.; Inomata, N.; Oomura, Y. γ-Aminobutyric acid- and pentobarbitone-gated chloride currents in internally perfused frog sensory neurons. J. Physiol. 360:367– 386;1985. Akaike, N.; Oomura, Y. Interactions of γ-aminobutyric acid (GABA), pentobarbital, and homopantothenic acid (HOPA) on internally perfused frog sensory neurons. Cell. Mol. Neurobiol. 5:245–255;1985. Akaike, N.; Tokutomi, N.; Ikemoto, Y. Augmentation of GABA-induced current in frog sensory neurons by pentobarbital. Am. J. Physiol. 258:C452–C460;1990. Akesson, T.R.; Mantyh, P.W.; Mantyh, C.R.; Matt, D.W.; Micevych, P.E. Estrus cyclicity of 125I-cholecystokinin octapeptide binding in the ventromedial hypothalamic nucleus. Neuroendocrinology 45:257–262;1987. Baulieu, E.E.; Robel, P. Neurosteroids: A new brain function? J. Steroid Biochem. Mol. Biol. 37:395–403;1990. Belelli, D.; Bolger, M.B.; Gee, K.W. Anticonvulsant profile of the progesterone metabolite 5α-pregnan-3α-ol-20-one. Eur. J. Pharmacol. 166:325–329;1989. Belelli, D.; Lan, N.C.; Gee, W.K. Anticonvulsant steroids and the GABA/benzodiazepine receptor–chloride ionophore complex. Neurosci. Biobehav. Rev. 14:315–322;1990. Callachan, H.; Cottrell, G.A.; Hather, N.Y.; Lambert, J.J.; Nooney, J.M.; Peters, J.A. Modulation of the GABAA receptor by progesterone metabolites. Proc. R. Soc. Lond. B Biol. Sci. 231:359–369;1987. Cartmell, S.M.; Mitchell, D. Diazepam attenuates hyperexcitability and mechanical hypersensitivity of dorsal horn convergent neurons during reperfusion of rat’s tail following ischaemia. Brain Res. 659:82–90;1994. Costa, E.; Auta, J.; Guidotti, A.; Korneyev, A.; Romeo, E. The pharmacology of neurosteroidogenesis. J. Steroid Biochem. Mol. Biol. 49:385–389;1994. Costa, T.; Russell, L.; Pert, C.B.; Rodbard, D. Halide and γaminobutyric acid–induced enhancement of diazepam receptors in rat brain. Mol. Pharmacol. 20:470–476;1981. Ffrench-Mullen, J.M.H.; Danks, P.; Spence, K.T. Neurosteroids modulate calcium currents in hippocampal CA1 neurons via a pertussis toxin-sensitive G-protein-coupled mechanism. J. Neurosci. 14:1963–1977;1994. Frye, C.A.; Duncan, J.E. Progesterone metabolites, effective at the GABAA receptor complex, attenuate pain sensitivity in rats. Brain Res. 643:194–203;1994. Gee, K.W.; Tan, N.C. γ-Aminobutyric acid A receptor complexes in rat frontal cortex and spinal cord show differential responses to steroid modulation. Mol. Pharmacol. 40:995– 999;1991. Harrison, N.L.; Majewska, M.D.; Harrington, J.W.; Barker, J.L. Structure–activity relationships for steroid interactions with the γ-aminobutyric acid-A receptor complex. J. Pharmacol. Exp. Ther. 241:346 –353;1987. Hattori, K.; Oomura, Y.; Akaike, N. Diazepam activation on γ-aminobutyric acid-gated chloride currents in internally perfused frog sensory neurons. Cell. Mol. Neurobiol. 6:307–323; 1986. Horn, R.; Marty, A. Muscarinic activation of ionic currents measured by a new whole-cell recording method. J. Gen. Physiol. 92:145–159;1988. Howard, D.L.; Hammond, D.L. Effect of pentobarbital or diazepam on the antinociceptive actions of GABAA agonists in the spinal cord of the rat. Anesthesiology 75:A579; 1991. Hu, Z.Y.; Bourreau, E.; Jung-Testas, I.; Robet, P.; Baulieu, E.-E. Neurosteroids: oligodendrocyte mitochondria convert
T.-L. Xu et al.
226
22. 23.
24.
25. 26. 27.
28. 29. 30.
31. 32.
33. 34.
cholesterol to pregnenolone. Proc. Natl. Acad. Sci. USA 84: 8215–8219;1987. Jussofie, A. Brain region-specific effects of neuroactive steroids on the affinity and density of the GABA-binding site. Biol. Chem. Hoppe Seyler 374:265–270;1993. Karavolas, H.J.; Hodges, D.R. Neuroendocrine metabolism of progesterone and related progestins. In: Chadwick, D.; Widdows, K. (eds). Steroids and Neural Activity. Chichester: Wiley; 1990:22–56. Kokate, T.G.; Svensson, B.E.; Rogawski, M.A. Anticonvulsant activity of neurosteroids: Correlation with γ-aminobutyric acid-evoked chloride current potentiation. J. Pharmacol. Exp. Ther. 270:1223–1229;1994. Lambert, J.J.; Belelli, D.; Hill-Venning, C.; Peters, J.A. Neurosteroids and GABAA receptor function. Trends Pharmacol. Sci. 16:295–303;1995. Majewska, M.D. Neurosteroids: Endogenous bimodal modulation of the GABAA receptor. Mechanism of action and physiological significance. Prog. Neurobiol. 38:379–395;1992. Majewska, M.D.; Harrison, N.L.; Schwartz, R.D.; Barker, J.L.; Paul, S.M. Steroid hormone metabolites are barbiturate-like modulators of the GABA receptor. Science 232:1004 –1007; 1986. McEwen, B.S. Non-genomic and genomic effects of steroids on neural activity. Trends Pharmacol. Sci. 12:141–147;1991. Mellon, S.H. Neurosteroids: Biochemistry, modes of action, and clinical relevance. J. Clin. Endocrinol. Metab. 78:1003– 1008;1994. Morrow, A.L.; Pace, J.R.; Purdy, R.H.; Paul, S.M. Characterization of steroid interactions with γ-aminobutyric acid-gated chloride ion channels: Evidence for multiple steroid recognition sites. Mol. Pharmacol. 37:263–270;1990. Morrow, A.L.; Suzdak, P.D.; Paul, S.M. Steroid hormone metabolites potentiate GABA receptor-mediated chloride ion flux with nanomolar potency. Eur. J. Pharmacol. 142:483–485;1987. Murase, K.; Randic, M.; Shirasaki, T.; Nakagawa, T.; Akaike, N. Serotonin suppresses N-methyl-d-aspartate responses in acutely isolated spinal dorsal horn neurons of the rat. Brain Res. 525:84–91;1990. Olsen, R.W. Drug interactions as the GABA receptor ionophore complex. Annu. Rev. Pharmacol. Toxicol. 22:245–277; 1982. Orchinik, M.; Murray, T.F.; Moore, F.L. Steroid modulation of GABAA receptors in an amphibian brain. Brain Res. 646: 258–266;1994.
35. Paul, S.M.; Purdy, R.H. Neuroactive steroids. FASEB J. 6: 2311–2322;1992. 36. Piua, G.; Santi, M.; Vicini, S.; Pritchett, D.B.; Purdy, R.H.; Paul, S.M.; Seeburg, P.H.; Costa, E. Neurosteroids act on recombinant human GABAA receptors. Neuron 4:759–765; 1990. 37. Prince, R.J.; Simmonds, M.A. 5β-Pregnan-3β-ol-20-one, a specific antagonist at neurosteroid site of the GABAA receptor-complex. Neurosci. Lett. 135:273–275;1992. 38. Purdy, R.H.; Morrow, A.L.; Moore, P.H.J.; Paul, S.M. Stressinduced elevations of γ-aminobutyric acid type A receptoractive steroids in the rat brain. Proc. Natl. Acad. Sci. USA 88:4553–4557;1991. 39. Selye, H. The anesthetic effect of steroid hormones. Proc. Soc. Exp. Biol. Med. 46:116–121;1941. 40. Selye, H. The antagonism between anesthetic steroid hormones and pentamethylenetetrazol (metrazol). J. Clin. Endocrinol. 27:1051–1053;1942. 41. Tokutomi, N.; Agopyan, N.; Akaike, N. Penicillin-induced potentiation of glycine receptor-operated chloride current in ventro-meidal hypothalamic neurons. Br. J. Pharmacol. 106: 73–78;1992. 42. Turner, D.M.; Ransom, R.W.; Yang, J.S.-J.; Olsen, R.W. Steroid anesthetics and naturally occurring analogs modulate the γ-aminobutyric acid receptor complex at a site distinct from barbiturates. J. Pharmacol. Exp. Ther. 248:960–966; 1989. 43. Turner, J.P.; Simmonds, M.A. Modulation of the GABAA receptor complex by steroids in slices of rat cuneate nucleus. Br. J. Pharmacol. 96:409–417;1989. 44. Twyman, R.E.; Macdonald, R.L. Neurosteroid regulation of GABAA single-channel kinetic properties of mouse spinal cord neurons in culture. J. Physiol. 456:215–245;1992. 45. Valera, S.; Ballivet, M.; Bertrand, D. Progesterone modulates a neuronal nicotinic acetylcholine receptor. Proc. Natl. Acad. Sci. USA 89:9949–9953;1992. 46. Wieland, S.; Lan, N.C.; Mirasedeghi, S.; Gee, K.W. Anxiolytic activity of the progesterone metabolite 5α-pregnan-3αol-20-one. Brain Res. 565:263–268;1991. 47. Wu, F.S.; Gibbs, T.; Farb, H.D. Inverse modulation of γaminobutyric acid- and glycine-induced currents by progesterone. Mol. Pharmacol. 37:597–602;1990. 48. Yakushiji, T.; Tokutomi, N.; Akaike, N. Augmentation of GABA-induced chloride current in frog sensory neurons by diazepam. Neurosci. Res. 6:309–320;1989.
ISSN 0300-9629/97/$17.00 PII S0300-9629(96)00288-5
Modulation of the GABAA Response in Rat Ventromedial Hypothalamic Neurons by Pregnanolone Tian-Le Xu, Taiichiro Imanishi, and Norio Akaike Department of Physiology, Faculty of Medicine, Kyushu University, Fukuoka 812-82, Japan ABSTRACT. The effects of 5β,3α-pregnanolone (PGN) on γ-aminobutyric acid (GABA)-induced Cl2 current in acutely dissociated rat ventromedial hypothalamic neurons were investigated using the nystatin perforated patch recording mode under voltage-clamp conditions. The PGN at concentrations between 1027 and 1025 M evoked an inward current at a holding potential (VH) of 240 mV. The reversal potential of the PGN-induced current was close to the equilibrium potential of Cl2. PGN potentiated the GABA-induced Cl2 current in a concentration-dependent manner. The facilitatory effect was long lasting and disappeared slowly after being washed out. The effect of PGN on the GABA response was also affected by the pretreatment time of PGN. PGN shifted the concentration–response relationship for GABA to the left with a suppression of the maximal response to GABA, resulting from the rapid inactivation of the GABA response during PGN treatment. Facilitatory interactions were found to exist among GABA, pentobarbital, diazepam, and PGN. Three other PGN isomers were also effective in facilitating the GABA response. However, the isomers containing the 3α-hydroxy configurations potentiated the GABA response much more potently and only these isomers exhibited inward currents themselves. comp biochem physiol 117A;2:219–226, 1997. 1997 Elsevier Science Inc. KEY WORDS. GABAA response, nystatin perforated patch, neurosteroid, pregnanolone, diazepam, pentobarbital, rat, ventromedial hypothalamic neurons
INTRODUCTION Recent discoveries in the field of neuroactive steroids (neurosteroids) have led to a resurgence in the research efforts to better understand the influence of steroids on the membrane receptors of CNS neurons (7,12,26,28,29,35). Although the historical view of steroid action focuses on the genomic effects of hormones, it is now well known that many steroids can modulate the excitability of CNS neurons via their direct action on ligand-gated (10,26,44,45, 47) and voltage-dependent (14) ion channels. Furthermore, there is also evidence to support the idea of the local synthesis of some neurosteroids in the brain (12,21,23,26,29,35). The GABAA receptor–Cl2 channel complex is a major target for neurosteroid action (1,34,43), and several neurosteroids have been shown to be potent stereoselective allosteric modulators of the GABAA receptors. Thus, the 3-hydroxy ring-A reduced metabolites of progesterone and deoxycorticosterone, such as 5α,3α-pregnanolone, 5β,3αpregnanolone, and 5α,3α-THDOC, produce a potent enAddress reprint requests to: N. Akaike, Ph.D., Department of Physiology, Faculty of Medicine, Kyushu University, 812-82, Japan. Tel. 81-92-6411151, ext. 3332; Fax 81-92-633-6748; E-mail: [email protected]. kyushu-u.ac.jp *Present address: Department of Anatomy, The Fourth Military Medical University, Xi’an, Shannxi 710032, People’s Republic of China. Received 6 February 1996; accepted 1 August 1996.
hancement of GABAA receptor-mediated response in vitro (17,24,27,44), whereas they also induce behavioral sedation in vivo (8,9,24). At higher concentrations, neurosteroids themselves also exhibited a GABA-mimetic action (10,24). Barbiturates and benzodiazepines have long been shown to facilitate the GABAA response (1–4,18,48), and their pharmacological significance is quite clear. Considerable data have demonstrated that both barbiturates and benzodiazepines facilitate the affinity of GABA to GABAA receptors as the main mechanism of their potentiating effects (13,33). The presence of a steroid modulatory site on the GABAA receptor–Cl2 channel complexes might thus provide a novel locus for the potential pharmacological manipulation of GABAA receptor-mediated neurotransmission. In fact, the endogenous progesterone metabolites, 5α,3αpregnanolone (8,24) and 5β,3α-pregnanolone (24), have been shown to exhibit a potent anticonvulsant activity against seizures induced by GABAA receptor antagonists such as pentylenetetrazol, bicuculline, and picrotoxin. The anticonvulsant profile of these neurosteroids resembled that of the benzodiazepine (24). Although several studies have reported the actions of neurosteroids on the GABA-related responses (1,25,34,43,44), few systematic electrophysiological studies on both the facilitation of the GABAA response and GABA-mimetic action by neurosteroids have apparently been performed. Whether the activation of the puta-
T.-L. Xu et al.
220
tive steroid site(s) results in electrophysiological and pharmacological effects similar to those produced by benzodiazepines or barbiturates also still remains to be determined. On the other hand, an important central nervous site for steroid modulation of physiological functions such as food intake and reproduction is the ventromedial hypothalamus (VMH) (6). Therefore, in the present study, we investigated how 5β,3α-pregnanolone (PGN) modulates the GABAA receptor-mediated Cl2 response in acutely dissociated rat ventromedial hypothalamic (VMH) neurons using a nystatin perforated patch recording configuration. MATERIALS AND METHODS Preparation VMH neurons were acutely dissociated as described elsewhere (41). Briefly, 2-week-old Wistar rats were decapitated under pentobarbital-Na anesthesia. Slices at a thickness of 400 µm containing the VMH region thus were obtained using a vibratome tissue slicer (Dosaka, DSK-1000). The slices were preincubated in a well-oxygenated incubation solution (see below for the composition) for 50 min at room temperature (22–25°C). Thereafter, the slices were treated enzymatically in well-oxygenated incubation solution containing 1 mg/6 ml pronase for 20 min at 31°C and then were successively exposed to 1 mg/6 ml thermolysin for another 15 min at the same temperature. After the enzyme treatment, the slices were kept in an enzyme-free incubation solution for 1 hr. Next, a portion of the VMH region was micropunched out with an electrolytically polished injection needle and was transferred into a culture dish filled with standard external solution (see below). The neurons were mechanically dissociated with fire-polished Pasteur pipettes under the visual guidance of a phase contrast microscope (Nikon, TMS-1). Dissociated neurons adhered to the bottom of the dish within 20 min, thus allowing us to conduct the electrophysiological studies. Only neurons that retained their original morphological features, such as the dendritic processes, were used for the experiments. Solutions The ionic composition of the incubation solution was (mM): NaCl 124, NaHCO3 24, KCl 5, KH2PO4 1.2, CaCl2 2.4, MgSO4 1.3, and glucose 10, aerated with 5% CO2/95% O2 to a final pH of 7.4. The normal external standard solution was (mM): NaCl 150, KCl 5, CaCl2 2, MgCl2 1, HEPES 10, and glucose 10. The pH was adjusted to 7.4 with Trisbase. The patch pipette solution for nystatin perforated patch recording was (mM): KCl 150, HEPES 10. The pH was adjusted to 7.2 with Tris-base. A nystatin stock solution dissolved in acidified methanol at a concentration of 10 mg/ ml was prepared and stored at 220°C. The stock solution was added to the patch pipette solution in a final nystatin concentration of 400 µg/ml just before use. When the current–voltage (I-V) relationship for the PGN-induced cur-
rent was examined using ramp pulses, 1 µM tetrodotoxin (TTX) and 10 µM CdCl2 were added to the standard external solution and then 150 mM CsCl was used in the patch pipette solution instead of KCl. Whole-Cell Recording All electrical measurements were carried out in a wholecell mode using the nystatin perforated patch recording technique (2,19) at room temperature (22–25°). The patch pipettes were pulled from glass capillaries with an outer diameter of 1.5 mm (Narishige) on a two-stage puller (Narishige, PB-7). The resistance between the recording electrode filled with pipette solution and the reference electrode was 4–6 MΩ. The patch pipette was positioned on a neuron using a hydraulic micromanipulator (Narishige, MO-102). The current and voltage were measured with a patch clamp amplifier (Nihon Koden, CEZ-2300), filtered at 1 kHz (NF Electronic Instruments, FV-665), and then was monitored on both a storage oscilloscope (Iwatsu Electronic, 5100A) and a pen recorder (Nippondenki San-ei, Recti-Horiz8K21). The data were simultaneously stored on videotape after digitalization with a PCM processor (Nihon Koden, PCM 501 ESN). A function generator (NF Electronic Instruments, FG-121B) was used to deliver the ramp pulses for the current–voltage experiments. The membrane potential was maintained at 240 mV throughout the experiment, except for the ramp pulse studies. All measurements were started after the stabilization of either the PGN or GABA responses (15–25 min after cell attachment). Drugs Thermolysin, nystatin, 5β,3α-pregnanolone (PGN), 5α,3α-pregnanolone (5α,3α-PGN), 5β,3β-pregnanolone (5β,3β-PGN), and 5α,3β-pregnanolone (5α,3β-PGN) were purchased from Sigma. The other drugs included: pronase (Calbiochem), GABA (Kasei), diazepam (DZP) (Ishizu), pentobarbital-Na (PB) (Ishizu), and tetrodotoxin (TTX) (Sankyo). Stock solutions of neurosteroids and diazepam were prepared in dimethyl sulfoxide (DMSO) and diluted to final concentrations in the external standard solution. The final concentrations of DMSO were always less than 0.1%; they did not induce any ionic current and they also had no effect on the GABA or PGN responses at the concentrations used. The drug solutions were applied using the ‘‘Y-tube’’ method (32) throughout the study. This system allows for the complete exchange of external solution surrounding a neuron within 10–20 msec. RESULTS PGN-Induced Inward Currents PGN evoked inward currents in all VMH neurons tested at a holding potential (VH) of 240 mV. The currents became detectable at a concentration of 1027 M and thereafter in-
Pregnanolone and GABAA Response
221
of two curves obtained before (a) and during (b) PGN application gave a reversal potential of the PGN response of 22.0 mV, which was close enough to the theoretical equilibrium potential of Cl2 (ECl) calculated from the given external and internal Cl2 concentrations (161 and 150 mM, respectively) by the Nernst equation, thus indicating that the PGN-induced current passes through the Cl2 channels. Modulation of GABA Response by PGN The effects of PGN at various concentrations (1028 to 1025 M) on the 3 3 1026 M GABA-induced Cl2 current were quantitatively studied. The currents obtained at various concentrations of PGN, either alone or with 3 3 1026 M GABA, are illustrated in Fig. 2A. PGN enhanced the GABA response, which was apparent with a concentration of PGN as low as 3 3 1028 M, in which the PGN concentration elicited no response. It is also evident that the inactivation of the GABA response became faster as the PGN concentrations increased. The potentiation ratio of the GABA response by PGN was calculated by a formula indicated
FIG. 1. PGN-induced inward currents in dissociated VMH neurons. [A(a)] The inward currents induced by PGN of various concentrations at a holding potential (VH) of 240 mV. [A(b)] Concentration–response relationship for PGNinduced current. All responses were normalized to the peak current amplitude induced by 1025 M GABA alone. Each point shows the mean of six neurons and the vertical bars show 6 SEM. (B) A representative record to show the current–voltage relationship for PGN-induced current. Depolarized triangular voltage ramp pulses of 60 mV, followed by a mirror-image hyperpolarizing ramp, were applied before (a) and during (b) PGN application (see inset). MP represents the membrane potential (mV).
creased in a concentration-dependent manner. As shown in Fig. 1A(a), the currents induced by PGN did not show any inactivation at concentrations below 1026 M, even in the case when the application lasted up to 2 min (data not shown), although a slight inactivation could be observed from concentrations beyond 3 3 1026 M. Because of limitations in the PGN solubility, no concentrations over 1025 M could be evaluated. To identify the ion that is responsible for the PGN-induced current, the current–voltage (I-V) relationship for PGN was thus elucidated using ramp pulses in the external solution containing TTX, CdCl 2, and CsCl, and in the pipette solution containing CsCl (Fig. 1B). TTX and CdCl2 did not affect the PGN response. The intercept
FIG. 2. Facilitatory effect of PGN on GABA-induced Cl2
current. (A) Inward currents induced by PGN alone (a) or by simultaneous application of PGN and 3 3 1026 M GABA (b). All traces were obtained from the same neuron. (B) Concentration–response relationship for PGN-mediated augmentation of GABA response. Ordinate shows enhancement ratio of GABA response by PGN. Each point is the mean of six measurements. Vertical bars show 6 SEM.
T.-L. Xu et al.
222
FIG. 3. Recovery time courses of the potentiation of GABA
responses by PGN. (A) Typical examples of the potentiation of GABA response and the recovery after wash out. All data were form the same neuron. (B) Summary of the recovery time courses of 3 3 1026 M GABA responses in the presence of PGN of different concentrations. All responses were normalized to the peak amplitude induced by 3 3 1026 M GABA alone. Each point is the average of four to six measurements. Vertical bars show 6 SEM.
along the y-axis in Fig. 2B because PGN itself induced an inward current [see also Fig. 2A(a)]. The GABA response was augmented in a concentration-dependent manner by PGN. As shown in Fig. 3, the augmentatory action of PGN on the GABA response was reversible, although the residual facilitatory effect slowly disappeared after the removal of PGN. The recovery time courses of the PGN actions of different concentrations on the GABA response are illustrated in Fig. 3B. It is clearly shown that a longer recovery time was required when higher concentrations of PGN were applied. The facilitatory effect of PGN on the GABA response was also affected by the pretreatment time of PGN of different concentrations. In Fig. 4, the PGN of a low concentration (3 3 1028 M) required a longer pretreatment time to evoke the maximal potentiation. The PGN at higher concentrations induced a maximal potentiation even with only 10 sec pretreatment, but the potentiation ratio decreased with the prolongation of the pretreatment time. Thus, the facilitatory effect of PGN on the GABA response depends not only on the concentrations of each drug but also on the duration of PGN pretreatment. Figure 5 shows the concentration–response relationship for GABA with or without PGN. PGN shifted the concentration–response curve for GABA to the left in a concentration-dependent fashion (Fig. 5B). At the same time,
FIG. 4. Effect of various pretreatment time of PGN on the potentiation of GABA response. (A) The potentiation of GABA response by 3 3 1028 (a) and 3 3 1026 M (b) PGN applied with various pretreatment time. (B) Pretreatment effects of PGN at different concentrations on GABA response. All responses were normalized to the peak current induced by 3 3 1026 M GABA alone. Each point represents the average value from six neurons. The vertical bars are 6 SEM.
PGN also suppressed the maximal response to GABA. As shown in Fig. 5A, the inactivation of the current induced by GABA of the higher concentrations was accelerated more in the presence of PGN. Facilitatory Interactions Among GABA, PB, DZP, and PGN Considering the possible involvement of either the barbiturates or benzodiazepines binding site on the effect of PGN, the facilitatory interactions among GABA, PB, DZP, and PGN were examined. As shown in Fig. 6A, the simultaneous treatment of GABA or PB at concentrations eliciting slight responses and PGN exhibited a marked potentiation. On the other hand, DZP did not induce any currents itself, whereas it potentiated the PGN response to a lesser extent than that of GABA or PB. The inward current evoked by the coapplication of low concentrations of GABA, PB, and DZP could still be enhanced by adding PGN (Fig. 6B).
Pregnanolone and GABAA Response
223
FIG. 6. Facilitatory interactions among GABA, pentobarbital
(PB), diazepam (DZP), and PGN. (A) Currents were elicited by 3 3 1027 M GABA (a), 3 3 1025 M PB (b), or 3 3 1026 M DZP (c) alone or combined with 1027 M PGN. (B) Simultaneous treatment of 1027 M GABA, 1025 M PB, and 1026 M DZP in the presence or absence of 1027 M PGN. Similar results in both (A) and (B) were obtained from four to six neurons.
FIG. 5. Concentration–response relationship for GABA in
the presence of PGN. (A) Typical currents induced by GABA alone (a) or by the combination of GABA and 3 3 1027 M PGN (b). (B) Concentration–response relationships for GABA with or without PGN of different concentrations. All responses were normalized to the peak response induced by 1025 M GABA alone (*). Each point shows the mean of four to six neurons.
Enhancement of GABA Response by PGN Isomers We next investigated how the GABA response is differently modified by PGN isomers [i.e., PGN (5β,3α-PGN), 5α,3αPGN, 5β,3β-PGN, 5α,3β-PGN] (Fig. 7A). Because PGN facilitated the GABA response at lower concentrations and also induced the GABA-mimetic response at higher concentrations, two representative concentrations of each isomer were used for comparison with the PGN actions. As shown in Fig. 7B, all steroids tested were effective in facilitating the GABA response at both concentrations. However, the isomers containing a 3α-hydroxy configuration potentiated the GABA response much more potently and only
these isomers themselves exhibited a GABA-mimetic effect. DISCUSSION Neurosteroids have been shown to interact potently with the GABA receptor–Cl2 channel complex in brain slice preparations (43), cultured neurons from the hippocampus (24) and the spinal cord (44), and cultured chromaffin cells (10). The present study confirmed that PGN, one of the 3α-hydroxy ring-A-reduced metabolites of progesterone, potentiated the GABA-induced Cl2 currents and also exhibited GABA-mimetic current in acutely dissociated VMH neurons. The PGN-induced inward current was concentration dependent and showed little inactivation. The examination of the I-V relationship for PGN response revealed the PGN response to be carried by Cl2. A previous study reported that bicuculline, a GABAA receptor antagonist, blocked PGN response (10), which also supports the fact that Cl2 passes through the GABA receptor–Cl2 channel complex during PGN treatment. It is unlikely that the response is produced by the interaction of PGN and GABA
224
FIG. 7. Structure–activity relationship for PGN isomers to enhance GABA response. (A) Chemical structures of PGN isomers: a, down; b, up. (B) Typical examples showing effects of four isomers of PGN at 3 3 1028 M (a) or 1026 M (b) on 3 3 1026 M GABA response. Similar results were obtained from four neurons. [B(a)] and [B(b)] were from different neurons.
released from neighboring cells, because the neurons we used were freshly dissociated. Therefore, the present results reinforced the view that this steroid directly activates the GABAA receptor to open the associated Cl2 channel. The current amplitudes observed under our experimental conditions were apparently larger than those observed in the cultured hippocampal neurons (24), chromaffin cells (10), recombinant GABAA receptor subunits expressed in 293 cells (36), and isolated frog sensory neurons (1), in which the induction of PGN currents was observed only at steroid concentrations of 3 3 1027 or higher. Moreover, the long latency in the activation of the PGN current reported for cultured hippocampal neurons (24) was not observed in the present study. However, these findings may result from differences in the preparations and/or regional differences of CNS used (16,22). The facilitatory effect of PGN on the GABA-induced Cl2 current was similar to the previous results observed in both electrophysiological (1,10,44) and binding studies
T.-L. Xu et al.
(30,31). Kokate et al. reported that the coapplication of GABA and PGN induced a complex Cl2 current exhibiting multiple peaks, which may reflect the multiple binding sites of PGN on GABAA receptor (24). However, in the present study, such multiple peaks were not observed during PGNGABA coapplication or coapplication PGN and GABA following PGN pretreatment. In addition, we have also shown that the potentiation action of PGN was long lasting and was also affected by the pretreatment time of PGN. Similar phenomena have been found in the augmentation of the GABA response by DZP in frog sensory neurons (48). In the presence of PGN, the concentration–response curve for GABA shifted to the left concomitant with the suppression of the maximal response to GABA. As shown in Fig. 5A, the inactivation of the GABA-induced Cl2 current accelerated more in the presence of PGN. Therefore, the suppression of the maximal response might be due to the rapid inactivation of the GABA response, although the effect of neurosteroids on GABA receptor desensitization is still unclear. On the other hand, another possibility may also exist. The results from a single channel analysis showed that in the presence of highest concentration (1025 M) of steroids, the prolongation of the average open time of GABAinduced single Cl2 channel currents decreased, suggesting the concentration-dependent blockage of GABA receptor– Cl2 channel complexes by steroids (44). Neurosteroids are normally found in neuronal tissue at rather low concentrations, but their levels increase dramatically in response to stress and during certain stages of pregnancy and/or the menstrual cycle in females. The plasma levels of 3α-hydroxy metabolites of progesterone reach 30 ng/ml (approximately 1027 M) during the third trimester of pregnancy and thereafter remain at an elevated level for a few months (35). The levels in the rat brain of these 3αhydroxy steroids measured after swimming stress are approximately 3–10 ng/g tissue (1028 –3 3 1028 M) (38). Although it is now uncertain whether or not the GABAmimetic action of PGN is clinically relevant, its augmentatory action on the GABA response is pharmacologically and clinically even more essential, because this action was observed at concentrations 10 times less than that needed to induce GABA-mimetic action and therefore is within the range of clinically relevant concentrations (35). It is already known that specific binding sites exist for some barbiturates and benzodiazepines on GABA receptor– Cl2 channel complex that closely link each other to the GABA binding site. Our previous study demonstrated that both PB (1,3–5) and DZP (1,18,48) potentiate the GABA response. Therefore, it is possible that PGN potentiated the GABA response by interacting with either the barbiturates or the benzodiazepines binding site. In the present study, a marked facilitatory interaction was observed among GABA, PB, DZP, and PGN. Both PB and DZP elicited an enlarged inward current in the presence of PGN as well as in the case of GABA. Furthermore, the current induced by the coapplication of GABA, PB, and DZP was also potentiated
Pregnanolone and GABAA Response
with same procedure. Although not conclusive, such results suggested that the modulatory site of PGN on the GABA receptor is probably neither a barbiturate site nor a benzodiazepine site. Previous studies have also suggested that steroids modulate the GABA receptor–Cl2 channel complex at a site distinct from barbiturates and benzodiazepines (10,16,17,30,42). The structure–activity relationship obtained in the present study for the potentiation of the GABA-activated Cl2 current in the presence of either 3α- or 3β-hydroxy isomers was similar to that in the cultured rat hippocampal neurons by Kokate et al. (24), but was different from that reported by Harrison et al. (17), who showed that 3β-hydroxysteroids had little effect at the GABAA receptor. Because 5β,3βPGN has been proposed to be a specific antagonist of the neurosteroid facilitatory site on the GABAA receptor (37), it is possible that 3β-hydroxy steroids may also play a physiological role in the modulation of GABAA receptor, although whether or not 3β-pregnanolones are produced in the CNS remains uncertain. However, it is apparent from both our present findings and our previous studies (17,24) that isomers having 3α-hydroxy configuration potentiated the GABA response much more potently. In addition, the present study also revealed that only these 3α-isomers themselves produced an inward current at higher concentrations. In conclusion, the present study demonstrated that PGN has a dual effect on the GABAA receptor–Cl2 channel complexes, the potentiation of GABA-induced current and, at relatively higher concentrations, the direct induction of GABA-mimetic current. In addition, this endogenous 3α-hydroxy ring A-reduced steroid metabolite interacts with the GABAA receptors at a unique recognition site that is different from both the benzodiazepine and barbiturate sites and affects the Cl2 channel activity. GABAA receptors have been well established to modulate the anxiolytic, sedative–hypnotic, anticonvulsant, antinociceptive, and anesthetic effects of benzodiazepines and barbiturates (11,20,35). Some neurosteroids, like other GABA-potentiating drugs, also have anxiolytic (46), sedative–hypnotic (8), anticonvulsant (9,24), antinociceptive (15), and anesthetic activities (39,40,42). The low intrinsic toxicity and lack of hormonal properties of these compounds together with the findings obtained herein therefore suggest that it may in the future be possible to develop novel synthetic steroids as therapeutic agents. We wish to thank Dr. K. Morita and Prof. J. S. Li for their support and encouragement throughout the course of these experiments.
References 1. Akaike, N.; Harata, N. GABAA receptor–chloride channel complex. In: Narahashi, T. (ed). Ion Channels of Excitable Cells. San Diego, CA: Academic Press, Inc; 1994:148– 165. 2. Akaike, N.; Harata, N. Nystatin perforated patch recording
225
3.
4.
5. 6.
7. 8. 9. 10.
11.
12. 13. 14.
15. 16.
17.
18.
19. 20.
21.
and its applications to analyses of intracellular mechanisms. Jpn. J. Physiol. 44:433–473;1994. Akaike, N.; Hattori, K.; Inomata, N.; Oomura, Y. γ-Aminobutyric acid- and pentobarbitone-gated chloride currents in internally perfused frog sensory neurons. J. Physiol. 360:367– 386;1985. Akaike, N.; Oomura, Y. Interactions of γ-aminobutyric acid (GABA), pentobarbital, and homopantothenic acid (HOPA) on internally perfused frog sensory neurons. Cell. Mol. Neurobiol. 5:245–255;1985. Akaike, N.; Tokutomi, N.; Ikemoto, Y. Augmentation of GABA-induced current in frog sensory neurons by pentobarbital. Am. J. Physiol. 258:C452–C460;1990. Akesson, T.R.; Mantyh, P.W.; Mantyh, C.R.; Matt, D.W.; Micevych, P.E. Estrus cyclicity of 125I-cholecystokinin octapeptide binding in the ventromedial hypothalamic nucleus. Neuroendocrinology 45:257–262;1987. Baulieu, E.E.; Robel, P. Neurosteroids: A new brain function? J. Steroid Biochem. Mol. Biol. 37:395–403;1990. Belelli, D.; Bolger, M.B.; Gee, K.W. Anticonvulsant profile of the progesterone metabolite 5α-pregnan-3α-ol-20-one. Eur. J. Pharmacol. 166:325–329;1989. Belelli, D.; Lan, N.C.; Gee, W.K. Anticonvulsant steroids and the GABA/benzodiazepine receptor–chloride ionophore complex. Neurosci. Biobehav. Rev. 14:315–322;1990. Callachan, H.; Cottrell, G.A.; Hather, N.Y.; Lambert, J.J.; Nooney, J.M.; Peters, J.A. Modulation of the GABAA receptor by progesterone metabolites. Proc. R. Soc. Lond. B Biol. Sci. 231:359–369;1987. Cartmell, S.M.; Mitchell, D. Diazepam attenuates hyperexcitability and mechanical hypersensitivity of dorsal horn convergent neurons during reperfusion of rat’s tail following ischaemia. Brain Res. 659:82–90;1994. Costa, E.; Auta, J.; Guidotti, A.; Korneyev, A.; Romeo, E. The pharmacology of neurosteroidogenesis. J. Steroid Biochem. Mol. Biol. 49:385–389;1994. Costa, T.; Russell, L.; Pert, C.B.; Rodbard, D. Halide and γaminobutyric acid–induced enhancement of diazepam receptors in rat brain. Mol. Pharmacol. 20:470–476;1981. Ffrench-Mullen, J.M.H.; Danks, P.; Spence, K.T. Neurosteroids modulate calcium currents in hippocampal CA1 neurons via a pertussis toxin-sensitive G-protein-coupled mechanism. J. Neurosci. 14:1963–1977;1994. Frye, C.A.; Duncan, J.E. Progesterone metabolites, effective at the GABAA receptor complex, attenuate pain sensitivity in rats. Brain Res. 643:194–203;1994. Gee, K.W.; Tan, N.C. γ-Aminobutyric acid A receptor complexes in rat frontal cortex and spinal cord show differential responses to steroid modulation. Mol. Pharmacol. 40:995– 999;1991. Harrison, N.L.; Majewska, M.D.; Harrington, J.W.; Barker, J.L. Structure–activity relationships for steroid interactions with the γ-aminobutyric acid-A receptor complex. J. Pharmacol. Exp. Ther. 241:346 –353;1987. Hattori, K.; Oomura, Y.; Akaike, N. Diazepam activation on γ-aminobutyric acid-gated chloride currents in internally perfused frog sensory neurons. Cell. Mol. Neurobiol. 6:307–323; 1986. Horn, R.; Marty, A. Muscarinic activation of ionic currents measured by a new whole-cell recording method. J. Gen. Physiol. 92:145–159;1988. Howard, D.L.; Hammond, D.L. Effect of pentobarbital or diazepam on the antinociceptive actions of GABAA agonists in the spinal cord of the rat. Anesthesiology 75:A579; 1991. Hu, Z.Y.; Bourreau, E.; Jung-Testas, I.; Robet, P.; Baulieu, E.-E. Neurosteroids: oligodendrocyte mitochondria convert
T.-L. Xu et al.
226
22. 23.
24.
25. 26. 27.
28. 29. 30.
31. 32.
33. 34.
cholesterol to pregnenolone. Proc. Natl. Acad. Sci. USA 84: 8215–8219;1987. Jussofie, A. Brain region-specific effects of neuroactive steroids on the affinity and density of the GABA-binding site. Biol. Chem. Hoppe Seyler 374:265–270;1993. Karavolas, H.J.; Hodges, D.R. Neuroendocrine metabolism of progesterone and related progestins. In: Chadwick, D.; Widdows, K. (eds). Steroids and Neural Activity. Chichester: Wiley; 1990:22–56. Kokate, T.G.; Svensson, B.E.; Rogawski, M.A. Anticonvulsant activity of neurosteroids: Correlation with γ-aminobutyric acid-evoked chloride current potentiation. J. Pharmacol. Exp. Ther. 270:1223–1229;1994. Lambert, J.J.; Belelli, D.; Hill-Venning, C.; Peters, J.A. Neurosteroids and GABAA receptor function. Trends Pharmacol. Sci. 16:295–303;1995. Majewska, M.D. Neurosteroids: Endogenous bimodal modulation of the GABAA receptor. Mechanism of action and physiological significance. Prog. Neurobiol. 38:379–395;1992. Majewska, M.D.; Harrison, N.L.; Schwartz, R.D.; Barker, J.L.; Paul, S.M. Steroid hormone metabolites are barbiturate-like modulators of the GABA receptor. Science 232:1004 –1007; 1986. McEwen, B.S. Non-genomic and genomic effects of steroids on neural activity. Trends Pharmacol. Sci. 12:141–147;1991. Mellon, S.H. Neurosteroids: Biochemistry, modes of action, and clinical relevance. J. Clin. Endocrinol. Metab. 78:1003– 1008;1994. Morrow, A.L.; Pace, J.R.; Purdy, R.H.; Paul, S.M. Characterization of steroid interactions with γ-aminobutyric acid-gated chloride ion channels: Evidence for multiple steroid recognition sites. Mol. Pharmacol. 37:263–270;1990. Morrow, A.L.; Suzdak, P.D.; Paul, S.M. Steroid hormone metabolites potentiate GABA receptor-mediated chloride ion flux with nanomolar potency. Eur. J. Pharmacol. 142:483–485;1987. Murase, K.; Randic, M.; Shirasaki, T.; Nakagawa, T.; Akaike, N. Serotonin suppresses N-methyl-d-aspartate responses in acutely isolated spinal dorsal horn neurons of the rat. Brain Res. 525:84–91;1990. Olsen, R.W. Drug interactions as the GABA receptor ionophore complex. Annu. Rev. Pharmacol. Toxicol. 22:245–277; 1982. Orchinik, M.; Murray, T.F.; Moore, F.L. Steroid modulation of GABAA receptors in an amphibian brain. Brain Res. 646: 258–266;1994.
35. Paul, S.M.; Purdy, R.H. Neuroactive steroids. FASEB J. 6: 2311–2322;1992. 36. Piua, G.; Santi, M.; Vicini, S.; Pritchett, D.B.; Purdy, R.H.; Paul, S.M.; Seeburg, P.H.; Costa, E. Neurosteroids act on recombinant human GABAA receptors. Neuron 4:759–765; 1990. 37. Prince, R.J.; Simmonds, M.A. 5β-Pregnan-3β-ol-20-one, a specific antagonist at neurosteroid site of the GABAA receptor-complex. Neurosci. Lett. 135:273–275;1992. 38. Purdy, R.H.; Morrow, A.L.; Moore, P.H.J.; Paul, S.M. Stressinduced elevations of γ-aminobutyric acid type A receptoractive steroids in the rat brain. Proc. Natl. Acad. Sci. USA 88:4553–4557;1991. 39. Selye, H. The anesthetic effect of steroid hormones. Proc. Soc. Exp. Biol. Med. 46:116–121;1941. 40. Selye, H. The antagonism between anesthetic steroid hormones and pentamethylenetetrazol (metrazol). J. Clin. Endocrinol. 27:1051–1053;1942. 41. Tokutomi, N.; Agopyan, N.; Akaike, N. Penicillin-induced potentiation of glycine receptor-operated chloride current in ventro-meidal hypothalamic neurons. Br. J. Pharmacol. 106: 73–78;1992. 42. Turner, D.M.; Ransom, R.W.; Yang, J.S.-J.; Olsen, R.W. Steroid anesthetics and naturally occurring analogs modulate the γ-aminobutyric acid receptor complex at a site distinct from barbiturates. J. Pharmacol. Exp. Ther. 248:960–966; 1989. 43. Turner, J.P.; Simmonds, M.A. Modulation of the GABAA receptor complex by steroids in slices of rat cuneate nucleus. Br. J. Pharmacol. 96:409–417;1989. 44. Twyman, R.E.; Macdonald, R.L. Neurosteroid regulation of GABAA single-channel kinetic properties of mouse spinal cord neurons in culture. J. Physiol. 456:215–245;1992. 45. Valera, S.; Ballivet, M.; Bertrand, D. Progesterone modulates a neuronal nicotinic acetylcholine receptor. Proc. Natl. Acad. Sci. USA 89:9949–9953;1992. 46. Wieland, S.; Lan, N.C.; Mirasedeghi, S.; Gee, K.W. Anxiolytic activity of the progesterone metabolite 5α-pregnan-3αol-20-one. Brain Res. 565:263–268;1991. 47. Wu, F.S.; Gibbs, T.; Farb, H.D. Inverse modulation of γaminobutyric acid- and glycine-induced currents by progesterone. Mol. Pharmacol. 37:597–602;1990. 48. Yakushiji, T.; Tokutomi, N.; Akaike, N. Augmentation of GABA-induced chloride current in frog sensory neurons by diazepam. Neurosci. Res. 6:309–320;1989.