| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1International Cooperative Research Project/Solution Oriented Research for Science and Technology Cell Mechanosensing, Japan Science and Technology Agency, Nagoya, Japan; 2Department of Physiology, Nanjing Medical University, Nanjing, China; 3Department of Physiology, Nagoya University Graduate School of Medicine, Nagoya; and 4Department of Molecular Physiology, National Institute for Physiological Sciences, Okazaki, Japan
Submitted 4 August 2004; accepted in final form 9 June 2005
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
7nAChR antagonist -BGT or MLA prevented the SIGD increase by PREGS. Furthermore DMXB, a selective 7nAChR agonist, mimicked the PREGS effect on SIGD and antagonized the effect of PREGS. The presynaptic effect of PREGS was partially attenuated by the L-type Ca2+ channel (VGCC) blocker nifedipine. Based on these findings, we proposed a novel mechanism underlying the facilitated synaptic transmission by PREGS: this neurosteroid sensitizes presynaptic 7nAChR that is followed by an activation of L-type VGCC to increase the presynaptic glutamate release. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
). Some experimental evidence have shown that hippocampal pyramidal neurons and granule neurons of adult male rats are equipped with complete machinery for the synthesis of PREGS (Kimoto et al. 2001; Shibuya et al. 2003). This brain neurosteroid is supposed to play an important role in memory and cognitive performances. This idea is supported by the memory-enhancing effect of PREGS on several learning tasks in rodents (Barnes 1988; Darnaudery et al. 2000; Vallee et al. 2001). Deficient cognitive performance in aged rats has been suggested to be related with low levels of PREGS in the hippocampus (Flood et al. 1995), and memory impairment can be corrected by intrahippocampal injection of PREGS (Maurice et al. 1998; Vallee et al. 1997).
Given the diverse potential roles of PREGS in regulating behavioral and pathological states, it is important to have a full understanding of PREGS actions at the molecular level. The effects of PREGS on neuronal excitability and synaptic plasticity have been observed by a number of in vivo and in vitro studies (Partridge and Valenzuela 2001, 2002). It has been postulated that some of the neurobehavioral effects of PREGS are mediated by the modulation of postsynaptic channel receptors (Meyer JH et al. 1999; Rupprecht and Holsboer 1999). However, a series of recent studies suggest a novel role of PREGS at the presynaptic level. For example, PREGS enhances dopamine (Barrot et al. 1999) and acetylcholine releases (Mayo et al. 1993) with the improvement of spatial recognition in aged rats (Darnaudery et al. 2000, 2002). PREGS is also known to enhance paired-pulse facilitation in the rat hippocampal CA1 (Partridge et al. 2001; Thomas et al. 2005) and to increase the occurrence frequency of miniature excitatory postsynaptic currents (mEPSCs) in hippocampal neurons (Mameli et al. 2005; Meyer DA et al. 2002). However, the interpretation of these results may not be easy. Even the sites of PREGS action, pre- and/or postsynaptic neurons, have not been clear. One of the major reasons is that it is technically difficult to record simultaneously the pre- and postsynaptic responses. It is particularly difficult to measure presynaptic activity and glutamate release at synaptic clefts in real time.
It is well recognized that astrocytes play a major role in uptaking glutamate at synapses by glutamate transporters. Recently, an indirect method for real-time monitoring of the synaptic glutamate release in hippocampal slices has been reported (Bergles and Jahr 1997), by which the relative amount of the glutamate released from Schaffer collateral terminals is detected by recording synaptically activated glutamate transporter currents in astrocytes located in the stratum rediatum. Because the glutamate transporter in astrocytes is electrogenic (Brew and Attwell 1987), measurable currents can be induced when the transporter activity is increased by released glutamate (Mennerick and Zorumski 1994). Diamond et al. (1998) found that blockers of the glutamate transporters suppressed synaptically activated inward currents from the cell body of hippocampal astrocytes. Thus it has been believed that the activity of glutamate transporters represents a parallel change in the probability of presynaptic glutamate release. Because of the spherical expansion of astrocyte processes and the random distribution of astrocytes, transporter currents generated in astrocytes at various loci would be canceled out and could not be detected as an extracellular field potential. There, an optical recording technique in combination with voltage-sensitive dyes, which can simultaneously measure multiple sites neural activities (Iijima et al. 1996), may be more suitable to analyze changes in synaptic transmissions.
The optical signal reflects intracellular potential changes (Barish et al. 1996), giving more precise information on the alteration of neuronal activities. More recently Kojima et al. (1999) and Kawamura et al. (2004) developed a novel optical method by using the voltage-sensitive dye RH155, to monitor the synaptically induced glial depolarization (SIGD) caused by the glutamate uptake in astrocytes. A major basis of this method is that the observed SIGD could be completely eliminated by knockout of glutamate transporter 1 (GLT-1) that is specifically expressed in astrocytes (Diamond et al. 1998) or by specific GLT-1 glutamate transporter blockers.
In the present paper, based on experimental work using the optical recording technique, we report our finding that the glutamate release from perforant path (PP) axonal terminals, a major cortical input to the hippocampus, is enhanced by extracellularly applied PREGS. In addition, our results revealed that the enhanced glutamate release is caused by a positive modulation of presynaptic 7nAChRs by PREGS.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
All experiments were carried out in accordance with institutional guidelines. Male Wister rats aged 40–50 days were decapitated under deep anesthesia with ethyl ether. We confirmed that the rats were sufficiently anesthetized by pinching their skins. The brains were taken quickly after removing the skulls and then placed in ice-cold artificial cerebrospinal fluid (ACSF) for about 10 min. ACSF contained (in mM): NaCl 128, KCl 1.7, KH2PO4 1.24, MgSO4 1.3, CaCl2 2.4, NaHCO3 26, and glucose 10. ACSF was oxygenated with a gas mixture of 95% O2-5% CO2 and the pH was adjusted to 7.4. Horizontal slices (350 µm) were cut using a vibratome (Microslicer DTK 1500, Dousaka EM, Kyoto, Japan). The slices were obtained only from the middle part of the hippocampus and placed in a rest chamber containing freshly oxygenated ACSF at room temperature for>1 h to recover the slices from deterioration. All the slices used in this study were stained with a voltage-sensitive dye (RH482 0.1 mg/ml or RH155 0.2 mg/ml, Nippon Kanko Shikiso Kenkyujo, Okayama, Japan) for 15 min, and were kept in a chamber containing oxygenated ACSF for at least 30 min for recovery. The optical absorbency of the dye varies according to the membrane potential of cells (Grinvald et al. 1982). Konnerth and coworkers (1987) suggested that different dyes may preferentially partition into neuronal and nonneuronal membranes, and that it is thus possible to separate signals in neurons and glial cells. RH482 dye mainly stains neuronal cells in contrast to RH155, which is reported to stain glial cells preferentially over neuronal cells (Konnerth et al. 1987). For recordings, the slices were transferred to a recording chamber placed on the stage of an Olympus inverted microscope (IMT-2, Nikon, Tokyo, Japan), and was perfused continuously with ACSF at a flow rate of about 1–2 ml/min. The experiments were performed at room temperature (27 ± 2°C). Ca2+-free solution was prepared by replacing CaCl2 with the same concentration of MgCl2 (final Mg2+ concentration of 3.25 mM).
Chemicals
-Bungarotoxin (-BGT), methyllycaconitine (MLA), dihydro-
-erythroidine hydrobromide (DHE), 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX), 5-amino-D-phosphonovaleric acid (D-AP5), and haloperidol were purchased from Sigma Research Biochemicals (Natick, MA). 
-CgTX GVIA, -AgaTX IVA, Ni2+, dihydropyridines (nifedipine), dihydrokainate (DHK), tetrodotoxin (TTX), pregnenolone sulfate (3-hydroxy-5-pregnen-20-one sulfate; PREGS), and bicuculline were purchased from Sigma (Milan, Italy). BD1047 dihydrobromide was purchased from Tocris. 3-(4)-Dimethylaminocinnamylidine anabaseline (DMXB; code name GTS-21) was provided by Taiho Pharmaceuticals (Tokushima, Japan). Other chemicals of special grade were obtained from Wako chemical (Osaka, Japan). PREGS was dissolved in dimethyl sulfoxide (DMSO) and diluted to the superfusing solution at a final concentration of 0.1% DMSO. Treatment of slices with 0.1% DMSO alone had no effect on the basal levels of synaptic transmission including amplitude and duration of excitatory postsynaptic potential (EPSP) and SIGD.
Electrical stimulation
Orthodromic stimuli were delivered using an electrically polished bipolar tungsten electrodes or a glass electrode filled with 0.9% NaCl (5 M
). A stimulating electrode was placed on the perforant path in the molecular layer of the dentate gyrus (DG). Constant-current pulses (100 µs, 0.06 Hz) were supplied by a stimulator (SEN-3301, Nihon Kohden, Tokyo, Japan). The intensity of test stimuli was adjusted to evoke about 50% of the maximum EPSP responses.
Field potential recordings
In some experiments, extracellular field potential recordings from the stratum molecule were simultaneously performed with optical recordings to ensure that the optical response was consistent with the electric response. After a slice was transferred to a submerged recording chamber, two hydraulic MMW204 Narishige micromanipulators (MMW204, Narishige, Tokyo, Japan) mounted on the microscopic stage were used to place a bipolar tungsten electrode in the molecular layer to stimulate the perforant path. Field excitatory postsynaptic potentials (f-EPSPs) were recorded from the molecular layer with a 4- to 5-M resistance glass microelectrode filled with 0.9% NaCl and connected to a neutralized, high-input–impedance preamplifier.
Optical recordings
Light from a tungsten-halogen lamp (type JC-24v/200W, Kondo Philips, Tokyo, Japan) was collimated and rendered quasimonochromatically with a heat filter (Olympus Optical, Tokyo, Japan) and an interference filter having the transmission maximum at 700 ± 10 nm (Olympus Optical). Changes in light absorption associated with membrane potential changes were measured with a high-speed optical recording system (HR Deltaron-1700 Fujix; Fuji Photo Film, Tokyo, Japan), which consists of an area sensor with 128 x 128 photodiodes and a data-processing unit. Each photodiode receives light from a 25 x 25-µm sample area, thus creating a 3.3 x 3.3-mm entire receptive field with a 4x objective lens. Changes in light absorption were detected by metal oxide (MOS) image sensors with an aid of a 690-nm interference filter (bandwidth of ±30 nm). To minimize bleaching of the dye molecules and photodynamic damages to slices, a heat-absorption filter was placed on the slices, and the light path to the slice was opened only long enough to measure optical responses (1–2 s). In each trial, a background image recorded for 16 ms before electrical stimulation was stored as a reference image.
A series of optical images after stimulations were recorded at 0.6 ms/frame and digitized into 8-bit signals from which the reference image was subtracted, and digitized into 8-bit signals. The digitized signals were then amplified 400x. Sixteen trials were averaged to improve the signal-to-noise ratio. The level of neural activities was indicated with pseudocolor (red indicates a high level, green a medium level, and blue a low level). Because maximum signal levels were<0.1%, we mostly used the color table: red, >0.055%; green, 0.026–0.054%; blue<0.025%. To make more quantitative analyses of the amplitude–time relationship (time course) of optical signals, data from each pixel were stored and retrieved, and the amplitude (indicated as a percentage change in optical absorbency) was plotted as a function of time.
Data analysis
Data were retrieved and processed with Micrcal Origin 6.1 (Northampton, MA): 1) To evaluate the effects of PREGS on presynaptic glutamate release, the area of SIGD (SIGD area; area between the basal line and the curve of SIGD) from 0 to 80 ms was measured and used as an index of glutamate transporter activity. 2) The paired-pulse facilitation (PPF) of EPSP and SIGD was calculated with the formula: paired-pulse ratio (PPR) = 100 x (S2EPSP – S1EPSP)/S1EPSP or 100 x (S2SIGD – S1SIGD)/S1SIGD. Changes in PPREPSP were calculated as the percentage of the absolute value of the baseline ratio. For PPRSIGD measurements, taken at 25- to 75-ms interpulse intervals (IPIs), the decay phase of the S1SIGD was fitted with a single exponential and the predicted values at 25–75 ms were subtracted from the S2SIGD to adjust the baseline of S2SIGD. 3) To evaluate the effects of inhibitors on the PREGS-induced SIGD increase, SIGD area in the presence of inhibitor(s) was subtracted from that after an application of PREGS. The obtained difference was normalized by the amount of SIGDarea increase in the presence of PREGS alone.
The group data were expressed as means ± SE, where n represents the number of hippocampal slices examined. Statistical analyses were performed using STATA 7.0 software (Stata, College Station, TX). In the analysis of PPREPSP and PPRSIGD before and after PREGS application, statistical differences were determined with Student’s comparison t-test. Statistical differences among values for individual groups were determined by ANOVA, followed by the Bonferroni post hoc test when F ratios were significant (*P<0.05). The signal intensity was repeatedly measured at the same area (pixel) of the tissue in ACSF (control), during an application of drug and after washout.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Optical signals in the DG region of hippocampal slices stained with RH482 were recorded using a procedure as described in METHODS. Figure 1A shows a time series of control images of signal propagation evoked by a stimulation of the perforant path. A weak excitatory signal was detected around the stimulated locus 1.2 ms after stimulus delivery, followed by a bidirectional signal spread along the molecular layer, then the signal invaded into the granule cell layer at 6.0 ms poststimulus. Subsequently, the excitatory signal expanded to cover the entire DG at 7.2 ms, around which time it reached a maximum amplitude, after which the signal began to decline and gradually disappeared by 12 ms. The effects of 50 µM PREGS on the signal propagation in the same slice are shown in Fig. 1B, in which evoked signals increase in their intensity and covering area as well as in duration. Essentially the same results were obtained from 26 out of 30 slices. At the end of the recording, CNQX (10 µM) was applied to confirm that the signals of the optical propagation were mediated mainly by -amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor (data not shown).
|
|
Effects of PREGS on presynaptic properties
The augmentation of EPSP by PREGS must originate from an enhanced glutamate release and/or a sensitization of postsynaptic glutamate receptors. First, we examined a putative contribution of PREGS to presynaptic events by measuring the dependency of the EPSP amplitude on extracellular Ca2+ concentration ([Ca2+]o). The PREGS-induced increase of EPSP was found to be strongly dependent on [Ca2+]o with an approximate 1.8 mM of EC50, as illustrated in Fig. 3A. This result strongly suggests that PREGS modulates presynaptic glutamate release because it is widely accepted that the changes in [Ca2+]o modify the probability of neurotransmitter release (P
) from presynaptic terminals (Katz and Miledi 1968).
|
). In this study, however, perfusion with Mg2+-free external solution containing 50 µM AP5, a selective NMDA antagonist, did not influence the effect of PREGS on EPSP (see the left traces in Fig. 3B). As shown in Fig. 3B, PREGS (50 µM) significantly increased the amplitude of EPSP by 131.22 ± 5.44% and 129.35 ± 7.41% in the absence and presence of AP5 (50 µM), respectively. A similar result (124.53 ± 8.76% increase) was obtained in slices treated with 10 µM MK-801, an open-channel blocker of NMDA receptor, indicating that the PREGS-induced increase of EPSP may not be attributed to the sensitization of the NMDA receptor. We further examined the effect of PREGS on PPF of EPSP elicited by two successive stimulation pulses, which has been used as an index of presynaptic facilitation. PPREPSP was measured with various IPIs of 25, 50, 75, and 100 ms before and after a PREGS application. We observed that the PREGS (50 µM) application significantly decreased the PPREPSP at 25- to 50-ms IPIs (Fig. 3C). In particular there was a marked decline of PPREPSP (from 22.49 ± 3.56 to 1.014 ± 3.86%, P<0.001) with increasing S1EPSP amplitude (132.88 ± 5.52%) at 50-ms IPI. To avoid the synaptic vesicle depletion presumably caused by PREGS, the stimulus current after the application of PREGS was adjusted to produce an equivalent size of EPSP similar to that before PREGS application (see the S1EPSP in Fig. 3D, inset). As shown in Fig. 3D, the application of PREGS (50 µM) increased PPREPSP (from 21.63 ± 2.26 to 29.76 ± 3.41%, P<0.05) in response to the paired stimuli of 50-ms IPI. These observations suggested that PREGS at micromolar concentrations could enhance the glutamate release at perforant path–granular cell synapses.
Effects of PREGS on glutamate release
To obtain strong evidence for PREGS-enhanced glutamate release, we tried to optically monitor the synaptic activation of the glutamate transporters in hippocampal slices stained with RH155. The optical signals of this dye evoked at the molecular layer consisted of two components, a spikelike signal and a slow depolarizing component (Fig. 4A, first trace), similar to that in slices stained with RH482. However, a considerable difference was that a sizable delayed depolarizing response remained in the presence of CNQX (10 µM)/AP5 (50 µM) (Fig. 4A, second trace). The delayed component was perfectly abolished by 1 mM DHK, a specific blocker for glial glutamate transporter GLT-1 (Fig. 4A, third trace), indicating that the delayed depolarizing response reflects the SIGD.
|
; Kojima et al. 1999) showed that SIGD could be recorded in each region of rat hippocampus. To test whether the SIGD is sensitive to the changes in the glutamate release from perforant path terminals, two standard manipulations that alter P were used. The first one is the increasing glutamate release by two successive stimulation pulses, which is thought to result from a residual elevation of intraterminal Ca2+ (Zucker 1999); the other is the decreasing glutamate release elicited by lowering [Ca2+]o from 2.4 to 1.0 mM. As expected, PPRSIGD could be evoked by a 50-ms IPI (Fig. 4B). In addition SIGD was significantly attenuated in 1.0 mM [Ca2+]o (Fig. 4C) and abolished in Ca2+-free solution (Fig. 4A, fourth trace). These results provide useful evidence that the size of SIGD is a reliable reporter of changes in the amount of synaptically released glutamates. We found that SIGD with respect to its amplitude and duration significantly increased within 10 min after the application of PREGS (50 µM). DHK was applied after exposure to 50 µM PREGS to further confirm that PREGS-induced increase of SIGD was a result of increased activation of the glutamate transporter. Indeed, both of the SIGD parameters increased by PREGS and basal SIGD was abolished by the addition of 1 mM DHK. In the presence of PREGS (50 µM), SIGDarea (see METHODS) increased up to approximately 1.3 times of the basal SIGDarea (P<0.001), whereas another neurosteroid pregnenolone (PREG) had no detectable effect on the size of SIGDarea (Fig. 5B). Additionally, we found that the blockade of the glutamate transporter with DHK (1 mM) before PREGS (50 µM) application completely inhibited the SIGD increase by PREGS (Fig. 5C). The kinetics of PREGS inactivation is not well known because PREGS is a lipid soluble compound and may take a relatively long time to be cleared off from the tissue. To address this issue, we measured the time course of PREGS effects on SIGD for 60 min after a brief exposure to PREGS. As seen in Fig. 5D, a 10-min application of 50 µM PREGS induced a SIGD increase with a rapid onset (<5 min) and a peak reaching within 10 min (open circle). After washing 10–15 min, the PREGS-induced increase in SIGD almost disappeared and returned to the baseline level, indicating that the effect of PREGS on the presynaptic release is reversible. In the absence of PREGS the size of SIGD was nearly constant for a period>60 min (Fig. 5D).
|
Involvement of presynaptic ionotropic receptors in PREGS-augmented glutamate release
PREGS is reported to enhance spontaneous the glutamate release in hippocampal neurons by a mechanism involving a 
1 receptor (Meyer DA et al. 2002). However, we found that haloperidol (0.5 µM), a 1 receptor antagonist, had little effect on basal SIGD (Fig. 6H) and the PPRSIGD (Fig. 6I) and it failed to influence the SIGD increase by PREGS (Fig. 6, B and J). Application of BD1047, another specific inhibitor of the 1 receptor, also did not inhibit the SIGD increase by PREGS (Fig. 6, C and J). Thus, it is unlikely that the 1 receptor is involved in the mechanism of the SIGD increase by PREGS observed in this study.
|
-aminobutyric acid type A (GABAA) receptor (Teschemacher et al. 1997). One simple interpretation is that PREGS enhances the glutamate release through decreasing GABAergic inhibitory transmission. We found that bath perfusion of 10 µM bicuculline dramatically increased the basal SIGD (Fig. 6H) and decreased the PPRSIGD (Fig. 6I), indicating the augmentation of presynaptic glutamate release. However, the effect of bicuculline (10 µM) on the SIGD increase by PREGS was limited (Fig. 6, D and J).
Alternatively, highly Ca2+ permeable 7nAChR located in the presynaptic terminals may contribute to the augmentation of the glutamate release (McGehee et al. 1995; Wonnacott 1997) that is supposed to be important for learning and memory processes. We noted that -BGT (0.1 µM) and MLA (0.2 µM) slightly decreased the size of basal SIGD (Fig. 6H), but they almost did not alter PPRSIGD (Fig. 6I). Interesting was that -BGT and MLA at identical concentrations significantly prevented the SIGD increase by PREGS (Fig. 6, E, F, and J). In contrast, extracellular treatment of 42nAChR antagonist DHE (10 µM) did not affect basal SIGD (Fig. 6H), the PPRSIGD (Fig. 6I), and the SIGD increase by PREGS (Fig. 6, G and J).
In summary, among the receptors putatively located in the presynaptic terminals, only 7nAChR seems to be responsible for the augmentation of glutamate release induced by PREGS, suggesting that the presynaptic target of PREGS is 7nAChR. As a result, we found that the application of 10 µM DMXB, a selective 7nAChR agonist (De Fiebre et al. 1995; Kem 2000), could mimic the effects of PREGS on the glutamate release to cause a robust increase in basal SIGD (Fig. 7A) in 10 of 12 slices. This effect was completely inhibited by the 7nAChR antagonist MLA (0.2 µM; Fig. 7B), but not by the 42nAChR antagonist DHE (10 µM; Fig. 7C), which indicates that the effect of DMXB on the glutamate release is mediated by 7nAChR (Fig. 7D). On the other hand, DMBX significantly decreased PPRSIGD, suggesting a further enhancing effect of DMXB on glutamate release (P<0.001; Fig. 7E). As illustrated in Fig. 7F, the pretreatment of 10 µM DMXB, although it caused a significant increase of SIGD (Fig. 7A), significantly inhibited the further increase in SIGD by PREGS (P<0.001). These results strongly support our hypothesis that PREGS enhances the glutamate release by a positive modulation of 7nAChR function.
|
We found that the enhancing effect of PREGS disappeared when the PREGS application was carried out in a very low [Ca2+]o solution, indicating that its effect depends on extracellular Ca2+ influx. One explanation, suggested by the high Ca2+ permeability of 7nAChR, is that 7nAChR activation might lead to sufficient Ca2+ influx by the 7nAChR–ligand Ca2+ channel itself. An alternative possibility is that 7nAChR activation results in Na+ influx and consequent depolarization, sufficient to activate local voltage-gated Ca2+ channels (VGCCs). The second possibility was verified by using blockers selective to VGCCs that have been classified into four types based on their gating kinetics and pharmacological properties: N-, P/Q-, T-, and L-types. Several experimental evidences suggest that fast transmitter release in the hippocampus is mediated by N- and P/Q-type Ca2+ channels, which are selectively sensitive to -CgTX GVIA and -AgaTX IVA, respectively (Luebke et al. 1993; Wheeler et al. 1994; Wu and Barish 1994). Normal transmission evoked by a test pulse at CA3–CA1 synapse is known to be insensitive to the inhibition of T- and L-type Ca2+ channels by Ni2+ and nifedipine, respectively (Barish et al. 1996).
Similar results were obtained in our study at perforant path–granular cell synapses (Fig. 8F). SIGD was substantially suppressed by -AgaTX IVA (0.1 µM, n = 11, P<0.001) or -CgTX GVIA (0.1 µM, n = 9, P<0.001), whereas it was insensitive to either Ni2+ (100 µM, n = 8) or nifedipine (50 µM, n = 7). However, this pharmacological pattern of basal SIGD to the VGCC blockers was dramatically altered when these blockers were applied in the presence of PREGS, as illustrated in Fig. 8, B–E. The SIGD increase by PREGS was insensitive to -CgTX GVIA (Fig. 8B), -AgaTX IVA (Fig. 8C), and Ni2+ (Fig. 8D), but significantly attenuated after the addition of nifedipine (
50% inhibition; P<0.001; Fig. 8, E and G). Further testing was done to show whether the 7nAChR-induced increase in the glutamate release is accompanied by L-type VGCC. As expected, the pretreatment of nifedipine (50 µM) significantly attenuated the SIGD increase by DMXB (Fig. 8H). Consistent with the result on the basal SIGD in Fig. 8F, application of nifedipine (50 µM) did not cause any alteration in PPRSIGD (P>0.05; Fig. 8I) elicited by paired stimuli at 50-ms IPI, indicating that nifedipine had no effect on the glutamate release in the absence of PREGS.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
In this study, we investigated the acute effects of PREGS on the synaptic transmission in the rat hippocampal DG using an optical recording technique that enables us to measure both the transmitter release and the EPSP. We found that the EPSP in granule cells was acutely and significantly enhanced on application of PREGS. We concluded that the increase in EPSP by PREGS was caused predominantly by enhanced probability of the glutamate release based on the following observations. First, PREGS rapidly increased SIGD, which represents the elevation of glutamate levels in the synaptic cleft. Second, PREGS decreased both PPREPSP and PPRSIGD, thus implying the probability of glutamate release. Third, the change in [Ca2+]o clearly altered the magnitude of the PREGS effects, indicating that the effect of PREGS required Ca2+ influx. Finally, the effects of PREGS depended on the activation of 7nAChR located at the presynaptic terminal, but not the postsynaptic NMDA receptors.
Generally, presynaptic short-term plasticity is the result of the interplay between facilitation, which depends on residual Ca2+, and depression, which depends largely on a decrease in the available readily releasable pool of vesicles (Dittman et al. 2000). We consider that in our case either decrease or increase in the PREGS-induced PPREPSP and PPRSIGD simply reflects the enhancing effect of PREGS on the probability of glutamate release. Recently Thomas et al. (2005) reported that PREGS enhances the PPF of excitatory postsynaptic currents (EPSCs) without increasing S1 amplitude at low concentration (300 nM), but PPF enhancement was not observed at concentrations>3 µM, in perforant path–granule cell synapses.
SIGD as a reporter of glutamate release
Glutamate uptake by high-affinity transporters is electrogenic, a consequence of the translocation of net positive charges during each transport cycle, which makes it possible to electrophysiologically monitor the transport activities (Schwartz and Tachibane 1990; Wyllie et al. 1991). Unlike postsynaptic ionotropic receptors facing presynaptic release sites, the transporter may be spread over the glial membranes surrounding synaptic clefts (Chaudry et al. 1995). Although the transporter will experience a lower concentration of glutamate as a consequence of their geometric distribution compared to that of synaptic clefts, it would work as an efficient scavenger of released glutamate because of its higher affinity to glutamate than AMPA receptors (Bergles et al. 1997). Munir et al. (2000) reported that neuronal cells exposed to 100 µM glutamate for 30 min upregulated expression of transporters. If such an upregulation occurs in the slice preparations treated with PREGS, the changes in SIGD would be contaminated by the changes in the number of transporters. However, reversibility and rapid onset of the PREGS effect on SIGD as shown in Fig. 5D suggest that the action of PREGS on SIGD is attributable to a nongenomic mechanism. Furthermore, the reversible enhancing effect of PREGS on the spontaneous glutamate release (Meyer DA et al. 2002) is consistent with our finding, and thus supports the opinion that SIGD can be used to evaluate the probability of the presynaptic glutamate release.
Molecular target of PREGS
The hippocampus is a center for learning and memory and receives cholinergic innervations mainly from the medial septum and the diagonal band (Woolf 1991). A lesion study demonstrates that the cholinergic basal forebrain neurons that project to the hippocampus play an important role in learning and memory (Dunnett et al. 1991). Experimental evidence have shown that 7nAChR localizes abundantly in the hippocampal DG (Seguela et al. 1993) and is expressed at nearly all the presynaptic terminals of rat hippocampus (Alkondon et al. 1996; Zhang et al. 1996). Because of its exceptionally high Ca2+ permeability (Castro and Albuquerque 1995), it has been believed for some time that 7nAChR regulates neurotransmitter releases. Indeed many experimental results support this idea. For example, acetylcholine enhances glutamate release by acting on presynaptic 7nAChR (Gray et al. 1996). Similarly, our results also revealed that the specific agonist and antagonists of 7nAChR could regulate the probability of the glutamate release in perforant path–granular synapses. This gives prominent evidence indicating an electric stimulation (test pulse) of the perforant path evokes an activation of 7nAChR, presumably at the presynaptic terminals, and that 7nAChR contributes to the synaptic transmission to some degree.
It has been well established that PREGS is a negative modulator of GABAA receptor (Akk et al. 2001) and a positive modulator of NMDA receptor (Mukai et al. 2000). Although there is no direct evidence, the following findings suggest that PREGS could upregulate the function of 7nAChR to enhance glutamate release. One of these findings is that the enhancing effect of PREGS was abolished by the selective blockers of 7nAChR, -BGT and MAL, and was mimicked by the 7nAChR agonist DMXB. Another more important finding is that, as shown in our results, the 7nAChR agonist partially attenuated the effect of PREGS, suggesting that the 7nAChR agonist and PREGS share the same molecular target, 7nAChR.
Ke and Lukas (1996) and Uki et al. (1999) proposed another mechanism for steroid action on nAChR. They considered that several steroids exert their inhibitory effects on nicotinic receptor function by a perturbation of membrane lipid structures. However, the effects of PREGS observed in this study are unlikely to be nonspecific ones caused by the membrane solvation of steroids or membrane disordering by steroids, because another neurosteroid, i.e., pregnenolone (PREG), had no detectable effect on the presynaptic glutamate release (Fig. 5B). Meyer JH et al. (1999) reported that DHEAS had no presynaptic action because this neurosteroid did not produce any alteration of PPF at either 20- or 200-ms interpulse intervals.
Coupling mechanism between 7nAChR and L-type VGCC
The increase in the glutamate release is principally linked to increases in the level of presynaptic Ca2+. We found in this study that the blockers for 7nAChR nearly completely depressed the SIGD increase by PREGS, whereas the L-type VGCC blocker nifedipine exerted partial inhibition. In addition other VGCC blockers had no effect, suggesting that another Ca2+ influx route other than VGCCs is involved. The most probable candidate molecule may be 7nAChR itself because of its high Ca2+ permeability. It is conceivable that Ca2+ influx across 7nAChR in combination with L-type VGCC may directly contribute to the enhancement of the glutamate release by PREGS. If this is true, the question is how 7nAChR and L-type VGCC are coupled. Because the 7nAChR blockade resulted in a complete inhibition of the PREGS effect, whereas the L-type VGCC blockade resulted in a partial inhibition, 7nAChR activation must be an upstream event.
There are at least two possible mechanisms for the coupling of these ion channels. A simpler and straightforward one may be an electrical coupling. Activation of positively modulated 7nAChR by PREGS would lead to a larger depolarization of the terminal membrane, which somehow selectively activates L-type VGCC, probably attributable to a peculiar time course and amplitude of the depolarization. It is suggested that opening of the 7nAChR channel can lead to terminal depolarization, which in turn induces activation of L-type VGCC and subsequent amplification of Ca2+ transients (Rathouz and Berg 1994; Vijayaraghavan et al. 1992). However, as a direct measurement of the putative depolarization induced by 7nAChR is practically difficult, the electrical coupling mechanism is only a hypothesis. Another possible mechanism may be a chemical coupling, which can be realized by increasing the Ca2+ influx through the modulated 7nAChR.
Several studies have demonstrated that Ca2+/CaM-dependent protein kinase II (CaMK) can induce a longer opening of the L-type Ca2+ channel by CaMK-mediated phosphorylation of the channel protein (Dzhura et al. 2000; Wu et al. 2003). There is another but rather trivial possibility for the apparent coupling of L-type Ca2+ channel with 7nAChR. Donnelly-Roberts et al. (1995) reported that some L-type Ca2+ channel blockers could directly interact with nAChR ionophore to inhibit the activation of nAChRs. However, another report argued that the 7nAChR activity was scarcely affected by most of the tested blockers for VGCCs, even though 34nAChR looked sensitive to all of them (Horrero et al. 1999). Furthermore, as shown in our results, nifedipine did not affect the basal SIGD at all. First, the L-type Ca2+ channel alone is not involved in a normal transmission evoked by a test pulse. Second, nifedipine does not act on 7nAChR because 7nAChR is involved in the basal SIGD, which has been ascertained by the reduction of basal SIGD with -BGT or MLA application as mentioned in RESULTS. Thus it is unlikely that nifedipine directly inhibits 7nAChR.
Finally, there is a possibility that PREGS altered the spatial distribution of the L-type Ca2+ channel. It has been suggested that properties of neurotransmitter releases relate to the spatial distribution of Ca2+ and thus those of responsible Ca2+ channels in the presynaptic membrane (Miller 1987). Sustained depolarizing stimuli such as elevated external K+ can elicit nifedipine-sensitive transmitter releases in the preparation of which nerve-driven releases are nifedipine insensitive (Feuerstein et al. 1991; Momiyama and Takahashi 1994). Although the nifedipine-sensitive Ca2+ channels are localized at the portions remote from N/Q-type, and the other type of channels located at the transmitter release sites, they can be recruited to the release sites by long-lasting terminal action potentials (Wheeler et al. 1996). At the present stage, although we cannot determine which mechanism(s) is responsible for the apparent coupling between 7nAChR and L-type Ca2+ channels, we tentatively prefer the simple electrical coupling one.
Recently, nifedipine (Hirasawa et al. 2003) was reported to behave as a secretagogue to increase spontaneous transmitter releases in central synapses by acting on the release process independent of the L-type Ca2+ channel. Nifedipine increases not only the frequency of mEPSPs but also their mean amplitude, which are abolished by the AMPA receptor antagonist CNQX (Hirasawa et al. 2003), implying that nifedipine somehow modulates the activation of AMPA receptor. This may explain why nifedipine did not affect basal SIGD in the presence of CNQX/AP5. Although the mechanism responsible for the nifedipine inhibition of the increase in SIGD by PREGS is not known at present, our results imply that nifedipine acts not only on the postsynaptic site but also on the L-type Ca2+ channel in the presynaptic terminals.
ffrench-Mullen et al. (1994) reported that endogenous brain steroids inhibited postsynaptic Ca2+ currents in dissociated hippocampal CAI pyramidal neurons by a pertussis toxin–sensitive G-protein–coupled mechanism. PREG, a precursor of PREGS, was found to nonselectively inhibit N- and L-type Ca2+ currents, although the authors provided no evidence for the direct action of PREGS on the L-type VGCC. In the present study, PREG was not found to affect the glutamate release, whereas PREGS had substantial effects on it. Because our experiments have been done under the complete blockade of EPSP by CNQX/AP5, the effects of PREGS on postsynaptic calcium currents remain to be solved.
Physiological and pathological correlates
Tissue levels of PREGS were measured and estimated to be in the nanomolar range (Kimoto et al. 2001). However, our results showed that relatively high concentrations of PREGS were required for the enhancement of glutamate release, which appears to be consistent with the previously published results showing that over 10 µM concentrations of PREGS enhanced PPF and glutamate release (Meyer et al. 2002; Partridge and Valenzuela 2001). NMDA receptor is significantly modulated by micromolar concentrations of PREGS (Gibbs et al. 1999). Recently, Mameli et al. (2005) found that a PREGS-like neurosteroid released from depolarized postsynaptic CA1 neurons has a similar effect to that observed with an application of 17 µM PREGS. Several studies have demonstrated that a closely related sulfated neurosferoid might actually be the endogenous counterpart of PREGS (Higashi et al. 2003a,b; Liu et al. 2003). These findings suggest a possibility that the micromolar range of exogenously applied PREGS can mimic the action of the endogenous equivalent of PREGS working under physiological conditions.
Reliable estimation of PREGS concentration in the synaptic cleft seems to be a very difficult problem. The effective concentration of PREGS at the synaptic cleft should be considerably higher than that obtained in tissue measurements. Even in brain tissues, PREGS concentration varies from 8 ng to 80 g (Vallee et al. 1997) and depends on the activity of sulfatase (Caldeira et al. 2004). Moreover, synthesis and release of PREGS are dependent on postsynaptic Ca2+ influx by NMDA receptor (Shibuya et al. 2003) in an activity-dependent manner (Mameli et al. 2005). The level of neurosteroids is age dependent (Caldeira et al. 2004), being lowered in aged animals (Morley et al. 1997). Because of these complexities, further efforts are necessary to determine the concentrations and the roles of the intrinsic PREGS in physiologically active brains.
Pregnenolone sulfate has been shown to act dose-dependently against -amyloid peptide–induced deficits in learning and memory (Maurice et al. 1998). 7nAChR has been reported to be a primary target of -amyloid (Wang et al. 2000) and 7nAChR agonists can improve memory-related behaviors in nonhuman primates (Bjugstad et al. 1996; Briggs et al. 1997), nucleus-basalis lesioned rats (Meyer et al. 1997), and aged rats (Arendash et al. 1995). Based on our findings in the present study we speculate that PREGS may have a therapeutic value in the treatment of age-related neurodegenerative diseases and cognitive deficits in Alzheimer’s disease.
In conclusion, measurements of synaptic activities at the rat hippocampal DG using voltage-sensitive dyes, by which we can separately evaluate transmitter release and EPSP, clearly revealed that the PREGS-induced EPSP increase was mediated by enhanced transmitter release. The effect of PREGS seems to be primarily attributable to its positive modulation of 7nAChR followed by L-type Ca2+ channel activation.
|
GRANTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
Address: M. Sokabe, Department of Physiology, Nagoya University Graduate School of Medicine, 65 Tsurumai-cho, Showa-ku, Nagoya 466-8560, Japan (E-mail: msokabe{at}med.nagoya-u.ac.jp)
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
subunit. J Physiol 532: 673–684, 2001.Alkondon M, Pereira EFR, and Albuquerque EX. Characterization of nicotinic acetylcholine receptors in CA1 neurons of rat hippocampal slices. Soc Neurosci Abstr 22: 1267–12, 1996.
Arendash M, Sengstock GJ, Sanberg PR, and Kem WR. Improved learning and memory in aged rats with chronic administration of the nicotinic receptor agonist GTS-21. Brain Res 674: 252–259, 1995.[CrossRef][ISI][Medline]
Barish ME, Ichikawa M, Tominaga T, Matsumoto G, and Iijima T. Enhanced fast synaptic transmission and a delayed depolarization induced by transient potassium current blockade in rat hippcampal slice as studied by optical recordings. J Neurosci 16: 5672–5687, 1996.
Barnes CA. Spatial learning and memory processes: the search for their neurobiological mechanisms in the rat. Trends Neurosci 11: 163–169, 1988.[CrossRef][ISI][Medline]
Barrot M, Valle M, Gingras MA, Le Moal M, Mayo W, and Piazza PV. The neurosteroid pregnenolone sulphate increases dopamine release and the dopaminergic response to morphine in the rat nucleus accumbens. Eur J Neurosci 11: 3757–3760, 1999.[CrossRef][ISI][Medline]
Baulieu EE. Neurosteroids: of the nervous system, by the nervous system, for the nervous system. Rec Prog Horm Res 52: 1–32, 1997.[ISI][Medline]
Bergles DE and Jahr CE. Synaptic activation of glutamate transporters in hippocampal astrocytes. Neuron 19: 1297–1308, 1997.[CrossRef][ISI][Medline]
Bjugstad KB, Mahnir VM, Kem WR, and Arendash GW. Long-term treatment with GTS-21 or nicotine enhances water maze performance in aged rats without affecting the density of nicotinic receptor subtypes in neocortex. Drug Dev Res 39: 19–28, 1996.[CrossRef]
Brew H and Attwell D. Electrogenic glutamate uptake is a major current in the membrane of axolotl glial cells. Nature 327: 707–709, 1987.[CrossRef][Medline]
Briggs CA, Anderson DJ, Brioni JD, Buccafusco JJ, Buchley MJ, Campbell JE, Decker MW, Donnelly-Roberts D, Elliot RL, Gopalakrishnan M, Holladay M, Holladay W, Hui Y, Jackson W, Kim DJB, Marsh KC, O’Neill AO, Pendergast MA, Ryther KB, Sullivan JP, and Arneric SP. Functional characterization of the novel nicotinic receptor ligand GTS-21 in vitro and in vivo. Pharmacol Biochem Behav 57: 231–241, 1997.[CrossRef][ISI][Medline]
Caldeira JC, Wu Y, Mameli M, Purdy RH, Li PK, Akwa Y, Savage DD, Engen JR, and Valenzuela CF. Fetal alcohol exposure alters neurosteroid levels in the developing rat brain. J Neurochem 90: 1530–1539, 2004.[CrossRef][ISI]
Castro NG and Albuquerque EX. The -bungarotoxin-sensitive hippocampal nicotinic acetylcholine receptor has a high calcium permeability. Biophys J 68: 516–524, 1995.
Chaudry FA, Lehre K, van Lookeren Campagne M, Ottersen OP, Danbolt NC, and Storm-Mathisen J. Glutamate transporters in glial plasma membranes: highly differentiated localizations revealed by quantitative ultrastructural immunocytochemistry. Neuron 15: 711–720, 1995.[CrossRef][ISI][Medline]
Darnaudery M, Koehl M, Piazza PV, Le Moal M, and Mayo W. Pregnenolone sulfate increases hippocampal acetylcholine release and spatial recognition. Brain Res 852: 173–179, 2000.[CrossRef][ISI][Medline]
Darnaudery M, Pallares M, Piazza PV, Moal ML, and Mayo W. The neurosteroid pregnenolone sulfate infused into the medial septum nucleus increases hippocampal acetylcholine and spatial memory in rats. Brain Res 951: 237–242, 2002.[CrossRef][ISI][Medline]
De Fiebre CM, Meyer EM, Henry JC, Muraskin SI, Kem WR, and Papke RL. Characterization of a series of anabaseine-derived compounds reveals that the 3-(4)-dimethylaminocinnamylidine derivative is a selective agonist at neuronal nicotinic 7/125--bungarotoxin receptor subtypes. Mol Pharmacol 47: 164–171, 1995.[Abstract]
Diamond JS, Bergles DE, and Jahr CE. Glutamate release monitored with astrocytes transporter currents during LTP. Neuron 21: 425–433, 1998.[CrossRef][ISI][Medline]
Dittman JS, Kreitzer AC, and Regehr WG. Interplay between facilitation, depression, and residual calcium at three presynaptic terminals. J Neurosci 20: 1374–1385, 2000.
Donnelly-Roberts DL, Arneric SP, and Sullivan JP. Functional modulation of human "ganglionic-like" neuronal nicotinic acetylcholine receptors (nAChR) by L-type calcium channel antagonists. Biochem Biophys Res Commun 213: 657–662, 1995.[CrossRef][ISI][Medline]
Dunnett SB, Everitt BJ, and Robbins TW. The basal forebrain–cortical cholinergic system: interpreting the functional consequences of excitotoxic lesions. Trends Neurosci 14: 494–501, 1991.[CrossRef][ISI][Medline]
Dzhura I, Wut Y, Colbrant RJ, Balser JR, and Anderson ME. Calmodulin kinase determines calcium-dependent facilitation of L-type calcium channels. Nat Cell Biol 2: 173–177, 2000.[CrossRef][ISI][Medline]
) [*P<0.001 by two-way ANOVA, F(1,51) = 14.82, n = 8]. E: summary of the effects of PREGS on PPRSIGD. A series of paired stimuli (25-, 50-, and 75-ms IPIs) was applied before and in 10 min after application of PREGS (50 µM). PPRSIGD decreased significantly (paired t-test, n = 5, *P<0.001) at 50- and 75-ms IPIs after application of PREGS. Sample traces show the changes in S1SIGD and S2SIGD after application of PREGS at 25- and 50-ms IPIs.
