| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Journal of Physiology (2001), 532.3, pp. 595-607
© Copyright 2001 The Physiological Society
 |
ABSTRACT |
|---|
 TopAbstract IntroductionMethodsResultsDiscussionReferences |
|---|
-Conotoxin GVIA (-CTx) and -Agatoxin TK (-Aga) both reduced evoked EPSC amplitude, while nicardipine and nickel had no effect. A combination of -CTx and -Aga completely abolished the evoked EPSCs.
-CTx but only partially in the presence of -Aga.
-CTx and -Aga.
-CTx or by OXT, thus further substantiating an action of both compounds on calcium channels.
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Dendrites are now known to play an active role not only in integrating afferent information and relaying it to the cell body, but in regulating their own afferent inputs. Neurons are capable of releasing neurotransmitters from the somatodendritic membrane (Isaacson & Strowbridge, 1998; Ludwig & Leng, 1998; Zilberter et al. 1999) as well as from axonal terminals; thus, attention has been directed at the elucidation of possible roles for these dendritically released transmitters in synaptic physiology. A particularly accessible nucleus for this study is the supraoptic nucleus (SON); here, dendrites of vasopressin- and oxytocin (OXT)-containing magnocellular neurons (MCNs) release peptides in response to physiologically relevant stimuli (Ludwig & Leng, 1998). Furthermore, interference with these dendritically released peptides affects the function of the nucleus (Neumann et al. 1996). In an in vitro voltage-clamp study, we determined that in the SON, dendritically released OXT acts as a retrograde transmitter to reduce glutamate release from the presynaptic terminals (Kombian et al. 1997). To further characterise this effect, it is important to determine the mechanism by which OXT acts to decrease excitatory transmission.
Alteration of neurotransmitter release can be achieved in many ways. Firstly, the activation of potassium currents in the presynaptic terminals can change action potential propagation and waveform. This in turn can affect the activity of voltage-dependent calcium channels, thus indirectly changing the calcium signal that triggers neurotransmitter release (Sabatini & Regehr, 1997). Secondly, calcium currents can be modulated directly at the terminal, which would alter the coupling between action potentials and transmitter release (Wu & Saggau, 1997). Thirdly, modulation can occur downstream of the calcium trigger by an action directed at the release machinery. Such modulation would be reflected not only in changes in the action potential-dependent release but also in the frequency of miniature events (Dittman & Regehr, 1996), which are extracellular calcium-independent spontaneous fusions of vesicles in the SON (Inenaga et al. 1998). OXT is known to have several distinct effects on neurons. In the spinal cord, OXT has been reported to modulate excitatory transmission by an extracellular calcium-dependent mechanism (Jo et al. 1998). OXT has also been found to increase intracellular calcium concentration, independent of calcium entry through calcium channels, by causing the release of calcium from internal stores (Lambert et al. 1994).
The entry of calcium into presynaptic terminals through voltage-gated calcium channels is a very important trigger for neurotransmitter release, and in many synapses these channels are a major target for neuromodulation (Dittman & Regehr, 1996; Qian et al. 1997; Isaacson, 1998). Calcium channels have been classified as high voltage-activated (HVA) channels, that is N-, P-, Q-, L- or R-type, and low voltage-activated (LVA) channels, namely T-type, based on their voltage dependence, kinetics and pharmacological properties. N- and P-type channels have been shown to mediate transmitter release in many mammalian central synapses (Dunlap et al. 1995). The involvement of Q-type (Wang et al. 1997; Bao et al. 1998), L-type (Bao et al. 1998) and R-type channels (Wu et al. 1998; Wang et al. 1999) have also been reported.
Our objective in this study was to determine if OXT affects a transmitter release pathway that is triggered by calcium influx through presynaptic voltage-dependent calcium channels. We therefore determined the types of voltage-dependent calcium channels involved in glutamate release in the SON using pharmacological blockers, and then asked if OXT decreases excitatory synaptic transmission by modulating any of these channels. Furthermore, analysis of miniature excitatory postsynaptic currents (mEPSCs) was performed to determine if OXT modulates the neurotransmitter release process downstream of calcium entry.
|
METHODS |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
All experiments were carried out in accordance with the guidelines established by the Canadian Council on Animal Care and were approved by the University of Calgary Animal Care Committee.
Adult male Sprague-Dawley rats were anaesthetised with halothane and quickly decapitated to obtain the brains. Coronal hypothalamic slices (400 µm) containing the SON were cut at 4 °C in a low Ca2+, low Na+-containing buffer solution of composition (mM): KCl, 2.5; NaH2PO4, 1.2; MgCl2, 4; CaCl2, 1; NaHCO3, 18; and sucrose, 250. Slices were then held at room temperature (22 °C) in artificial cerebrospinal fluid (ACSF) of the following composition (mM): NaCl, 126; KCl, 2.5; NaH2PO4, 1.2; MgCl2, 1.2; CaCl2, 2.4; NaHCO3, 18; and glucose, 11. A hemisected slice was then transferred into a recording chamber where it was submerged and perfused with ACSF at 27-29 °C, 2-3 ml min-1. All solutions were continuously bubbled with a mixture of O2 (95 %) and CO2 (5 %).
Whole cell nystatin-perforated patch recordings were made in the SON using an Axopatch-1D amplifier. Electrodes with tip resistance of 4-7 M
contained (mM): potassium acetate, 120; MgCl2, 5; EGTA, 10; and Hepes, 40. Nystatin was dissolved in DMSO with Pluronic-F127 and added to the internal solution to yield a final concentration of 450 µg ml-1. The pH of the solution was adjusted to between 7.2 and 7.4. MCNs were identified based on their characteristic membrane response to positive current injection (Tasker et al. 1991; Kombian et al. 1996). All experiments were done on MCNs voltage clamped at -80 mV. The holding current ranged from -10 to -100 pA amongst cells. Membrane currents were recorded without series resistance compensation, filtered at 1 kHz and digitised at 2-5 kHz. Twenty millivolt hyperpolarising pulses (75-100 ms duration) were applied regularly throughout each experiment, and the steady-state current and decay rate (
) of the capacitance transient were monitored as measures of input resistance and series/access resistance, respectively. The series/access resistance was 10-40 M. Data from cells that showed > 15 % change in these parameters were excluded from analysis. A bipolar tungsten stimulating electrode was placed in the hypothalamic region dorso-medial to the SON and synaptic responses were evoked using monophasic square voltage pulses (100-200 µs duration) at 30 s intervals. Excitatory postsynaptic currents (EPSCs) were pharmacologically isolated by blocking GABAA receptors with 50 µM picrotoxin. We have previously shown that the remaining postsynaptic currents were non-NMDA receptor mediated as they were blocked by 6-nitro-7-sulfamoylbenzo(f)quinoxaline-2,3-dione (NBQX), a selective antagonist for non-NMDA receptors (Kombian et al. 1996, 1997). All cells had graded evoked synaptic responses to increasing stimulation intensity and, unless otherwise indicated, an intensity giving 50-60 % of the maximum evoked EPSC was used to elicit evoked responses for all experiments. All data were acquired and analysed using pCLAMP (Clampex 7 and Clampfit 6.0.5, Axon Instruments) and Mini Analysis Program (Synaptosoft, Inc.). The current and voltage traces were also recorded on a Gould chart recorder.
Control data were collected for a minimum of 5 min prior to drug application. Baseline was calculated as the mean of the values obtained during the control period. All values are stated as means ± S.E.M. One-way ANOVA and Student's t tests (paired and unpaired) were used for comparison, as indicated. P < 0.05 was taken as significant.
All drugs were bath perfused at final concentrations as indicated, by diluting aliquots of stock in the ACSF immediately before use. Nicardipine stock was made in DMSO and kept in the dark due to its sensitivity to light. Drugs were all purchased from Sigma except oxytocin (Bachem, Torrance, CA, USA), NiCl2 (BDH Inc., Toronto, Canada) and Pluronic-F127 (BASF, Wyandotte, MI, USA).
|
RESULTS |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Calcium channel blockers and evoked EPSCs
The effects of relatively selective organic and inorganic calcium channel blockers on evoked EPSCs were examined in MCNs of the SON. -Agatoxin TK (-Aga), a P/Q-type channel blocker (Teramoto et al. 1995), reduced the evoked EPSCs by 44.9 ± 5.97 % at 200 nM (n = 9; Fig. 1A) with no change in holding current or input resistance of MCNs. Increasing its concentration to 500 nM or 1 µM did not produce a further decrease in evoked EPSC amplitude (39.7 ± 4.87 %; P > 0.05 compared with the 200 nM effect, paired t test; n = 4; Fig. 1B). This high dose of -Aga would block both P- and Q-type channels. However, P-type channels are known to have a higher sensitivity to -Aga, responding to much lower concentrations, such as 30 nM, than are required to block Q-type channels (Teramoto et al. 1995). When 30 nM -Aga was applied, the EPSC amplitude was inhibited by 27.6 ± 5.60 % (P < 0.05 compared with control, paired t test; n = 9; Fig. 1A, see also Fig. 3B). The existence of a component blocked by low concentrations of -Aga and another one by high concentrations of -Aga may be an indication that both P- and Q-type channels participate in excitatory synaptic transmission in the SON. The residual EPSC, despite its resistance to P/Q selective toxin blockade, was still non-NMDA dependent as it could be completely blocked by 1 µM NBQX (Fig. 1A). To determine if P/Q-type channels participate in synaptic transmission only at a threshold synaptic strength, we tested the ability of -Aga (200 nM) to inhibit the evoked EPSCs at different stimulus intensities. As shown in Fig. 1C, -Aga reduced the evoked EPSC with a similar potency over a wide range of stimulus intensities (P > 0.05, one-way ANOVA; n = 4), indicating similar recruitment of -Aga-sensitive channels throughout the entire range of synaptic strength.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 1. A, time-effect plot of a representative cell showing that
| ||
Blockade of N-type channels by bath application of -conotoxin GVIA (-CTx; 100 nM) (Boland et al. 1994) caused a reduction in the evoked EPSC by 59.3 ± 5.13 % (n = 12; Fig. 2A) without an alteration in the holding current or input resistance of MCNs. This effect of -CTx could not be washed out or reversed for the duration of our experiments. The -CTx-induced reduction of the evoked EPSC was maximal at 100 nM, as a higher concentration of the toxin (1 µM) caused no additional reduction (53.0 ± 6.97 %; P > 0.05 compared with 100 nM effect, paired t test; n = 5; Fig. 2B). -CTx, like -Aga, was effective in reducing the evoked EPSC over a wide range of stimulus intensities to a similar degree (P > 0.05, one-way ANOVA; n = 4; Fig. 2C).
 |
View larger version [in this window] [in a new window] |
|
|
Figure 2. A, time-effect plot of a representative cell showing that
| ||
As P/Q- and N-type blockers each suppressed a fraction of the evoked response, we tested if combined application of both blockers would produce an additive effect. In the presence of maximal concentration of both blockers (100 nM -CTx and 200 nM -Aga), the evoked EPSC was suppressed by 94.8 ± 2.22 % (P < 0.05; n = 6; Fig. 3). This effect was slightly less, by 9.4 %, than the sum of individual application of -CTx (59.3 %) and -Aga (44.9 %). While this may indicate that a fraction of EPSC depends on both types of calcium channels, the difference is not large enough to differentiate it from the synaptic variability within and among the recorded cells. The suppression of EPSC by a combination of P/Q- and N-type blockers was comparable to the blockade of calcium influx through high voltage-activated (HVA) calcium channels by 50 µM cadmium (100.1 ± 2.56 %, n = 4; P > 0.05 vs. -CTx + -Aga, unpaired t test) or elimination of extracellular calcium (96.0 ± 1.84 %, n = 3; P > 0.05 vs. -CTx + -Aga, unpaired t test). The contributions of other types of voltage-dependent calcium channels to the evoked EPSC were also tested in other cells using nicardipine (10 µM), an L-type channel blocker, and nickel (50 µM), which blocks T- and R-type channels (Wang et al. 1999). Although these concentrations have been shown to be effective in blocking L- (Harayama et al. 1998) and T-type currents (Fisher & Bourque, 1995) in SON MCNs, neither of these compounds affected the evoked EPSCs in these cells (1.39 ± 3.37 % and 1.76 ± 8.80 %, respectively; P > 0.05, paired t test; n = 5 and 7; Fig. 3B). Furthermore, the concentration of nickel we used can be expected to partially block R-type channels (Wang et al. 1999). These data indicate that N- and P/Q-type channels, but not L-, T- or R-type channels, are involved in evoked excitatory transmission under our stimulation and recording conditions in the SON.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 3. Evoked EPSCs are abolished by a combination of A, time-effect plot showing that cumulative application of
| ||
Calcium current blockade: site of action
The site of action of the calcium channel blockers to reduce EPSCs is likely to be at the presynaptic terminal, but the toxins could also have effects on local interneurons, which in turn may alter the postsynaptic cell properties or kinetics of glutamate receptors on the recorded neuron. Furthermore, even under voltage clamp, a possibility remains that distally located calcium channels may be activated during the excitatory synaptic transmission, resulting in an enhancement in the basal level of AMPA receptor-mediated current through calcium-dependent phosphorylation (Yakel et al. 1995). Since N- and P/Q-type calcium channels are expressed in MCNs (Fisher & Bourque, 1995; Foehring & Armstrong, 1996), blockade of these channels at distal dendrites by -Aga and -CTx could lead to a decrease in synaptic transmission. To differentiate among these possibilities and to determine the locus of the action of -Aga and -CTx to reduce synaptic transmission, we performed a battery of tests to differentiate between possible pre- and postsynaptic actions of these toxins. Our first approach was to look for changes in the paired-pulse ratio, a common index of change in release probability in the presynaptic terminal (Zucker, 1989). When a pair of stimuli was applied at an inter-stimulus interval of 50 ms to evoke two consecutive EPSCs, at the time of the peak reduction of the EPSCs induced by the channel blockers, the paired-pulse ratio (EPSC2/EPSC1) increased by 40.8 ± 11.3 % of control with -CTx (1.59 ± 0.31 vs. 2.23 ± 0.49; P < 0.05, paired t test; n = 7; Fig. 4A), and 34.6 ± 8.61 % of control for -Aga (1.27 ± 0.14 vs. 1.69 ± 0.19; P < 0.05, paired t test; n = 7; Fig. 4A). Next we analysed the rise times and decay constants () of evoked EPSCs in control and at the peak of synaptic depression to see if any alteration had occurred in kinetics or desensitisation of non-NMDA receptors. The rise times of evoked EPSCs (measured from baseline to peak) were not significantly altered by these compounds (3.05 ± 0.27 ms in control; n = 11, vs. 2.82 ± 0.41 ms for -CTx; n = 6, and 3.28 ± 0.37 ms for -Aga; n = 5; P > 0.05, paired t test). The decay was monoexponential and the decay time constant () as measured by the time from the peak to 37 % in control was 10.6 ± 0.87 ms (n = 11) as compared with 9.05 ± 0.90 for -CTx (n = 6) and 12.7 ± 1.11 ms (n = 5) for -Aga (P > 0.05 compared with control, paired t test). Thus scaled evoked EPSCs in control and calcium channel treatment were superimposable (Fig. 4B). Finally, neither compounds affected the steady-state AMPA current induced by brief bath applications of AMPA (-2.64 ± 5.18 % change for -CTx and 0.06 ± 4.89 % in the presence of both compounds; P > 0.05, one-way ANOVA; n = 4; Fig. 4C). The above data indicate that the synaptic depression caused by -CTx and -Aga does not involve an alteration in the properties of the recorded postsynaptic cells and support a presynaptic action of these toxins to reduce EPSCs.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 4. Presynaptic calcium channels mediate A1, averaged (6 traces) sample synaptic responses (from 2 separate cells) elicited by a pair of stimuli (50 ms interval) showing that
| ||
Calcium channel blockers interfere with oxytocin effect
To examine if OXT interacts with any or all of the N- and P/Q-type channels to decrease excitatory synaptic transmission, we performed occlusion experiments. The inhibitory effect on evoked EPSC of more than 20 % was observed in 14 out of 21 cells tested, and only those were used for further analysis. The input resistance did not change with the application of OXT (893.5 ± 169.1 M vs. 901.9 ± 163.5 M; P > 0.05, paired t test; n = 5). First we determined that repeated application of OXT caused a consistent reduction in the evoked EPSC; the first application induced 32.5 ± 8.02 % suppression and the second application 28.4 ± 6.50 % (n = 4; P > 0.05, paired t test). This depressant effect of OXT was of similar magnitude upon EPSCs of widely varying sizes, from < 50 pA to > 150 pA. In 10 other cells, either -CTx or -Aga was then applied and at the peak of their effects, the same concentration of OXT was re-applied (2 µM, see sequence in Fig. 5). In the presence of 100 nM -CTx, OXT caused a reduction of only 5.10 ± 3.47 % compared with the control OXT effect of 37.8 ± 5.67 % (n = 4; P < 0.05, paired t test; Fig. 5A). On the other hand, in the presence of 200 nM -Aga, the EPSC inhibitory effect of OXT was 16.9 ± 7.02 % compared with the control OXT effect of 35.3 ± 4.65 % (n = 6; P < 0.05, paired t test; Fig. 5B). These results indicate that OXT interacts with -CTx-, and to lesser extent -Aga-sensitive loci to decrease excitatory synaptic transmission in this nucleus.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 5. Exogenous oxytocin (OXT)-induced synaptic depression is completely blocked by A, time-effect plot of a representative cell showing that OXT (2 µM) causes a reversible depression of evoked EPSCs. In this same cell, following the recovery of the OXT effect, pretreatment with
| ||
It has been shown that the neuropeptides OXT and vasopressin are tonically released within the nucleus, although they are efficiently degraded by aminopeptidases (Burbach & Lebouille, 1983). Thus, an aminopeptidase inhibitor, amastatin, has been used in the in vitro slice preparation to unmask an effect of endogenous peptide in this nucleus (Kombian et al. 1997). In this study, we asked if these same calcium channel blockers affected the synaptic inhibitory action of the endogenous neuropeptides. Bath application of amastatin (10 µM) repeatedly reduced the evoked EPSC amplitude; the first application induced 26.1 ± 5.69 % reduction and the second application 27.3 ± 3.93 % (n = 4; P > 0.05, paired t test). This is an effect that we have previously shown to be blocked by an antagonist of vasopressin/OXT receptors, confirming the presence of biologically active neuropeptides in the extracellular fluid (Kombian et al. 1997). In 100 nM -CTx, amastatin was without effect on the evoked EPSC (-1.48 ± 3.85 % depression, P < 0.05 compared with the control amastatin effect, 24.3 ± 2.46 %, paired t test; n = 4; Fig. 6A and C). In the presence of -Aga (200 nM), one out of four cells still responded fully to amastatin while in the other three cells -Aga occluded the amastatin effect by 70-90 %. Overall, amastatin reduced the evoked EPSCs by 10.7 ± 4.62 % in -Aga (P < 0.05 compared with the control amastatin effect, 27.1 ± 3.61; paired t test; n = 4; Fig. 6B and C). The above results suggest that both exogenous and endogenous oxytocin reduces transmitter release triggered by calcium influx through mainly N-type and also partly through P/Q-type calcium channels in the SON.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 6. Endogenous neuropeptide-induced synaptic depression is completely blocked by A, time-effect plot of a representative cell showing that amastatin (AMAS; 10 µM) causes a reversible depression of evoked EPSCs. In this same cell, following the recovery of the amastatin effect, pretreatment with
| ||
Effect of oxytocin on miniature EPSCs
Another form of neurotransmitter release consists of action potential-independent events called miniature events. The glutamate-mediated miniature events, mEPSCs, in the SON have been shown to be independent of extracellular calcium both in our laboratory (M. Hirasawa, unpublished observation) and elsewhere (Inenaga et al. 1998). Therefore, if the effect of OXT on evoked glutamate release is calcium dependent, then it should not influence calcium-independent mEPSCs. We (Kombian et al. 2000a) and others (Kabashima et al. 1997) have previously shown that spontaneous EPSCs in the SON are all tetrodotoxin-insensitive mEPSCs, probably due to the slice preparation, where the somata of the presynaptic neurons are dissected. Another possibility is that the recording temperature (27-29 °C) or ACSF we use may somehow quieten down the activity of any interneurons existing in the slice. This enables us to study the action of the neuropeptide and the calcium channel blockers on mEPSCs in cells that are confirmed to have an apparent response to these compounds on evoked EPSCs. We first verified that mEPSCs were resistant to -CTx and as expected, the calcium channel blocker had no effect on either the frequency (2.36 ± 0.86 Hz vs. 2.97 ± 0.96 Hz; n = 6; P > 0.05, paired t test) or mean amplitude of mEPSCs (9.89 ± 0.78 pA vs. 10.6 ± 1.35 pA; n = 6; P > 0.05, paired t test; Fig. 7A and B). Similarly, in four other cells OXT (2 µM) did not change the frequency (3.46 ± 1.58 vs. 3.00 ± 1.38 Hz; P > 0.05, paired t test) or mean amplitude (12.0 ± 1.30 vs. 12.1 ± 1.79 pA; P > 0.05, paired t test) of mEPSCs (Fig. 7A and B). Between 21 and 703 events recorded in each condition in each cell were used for this analysis. This finding is consistent with an action of OXT on calcium influx into the terminal, but not on the process involved in calcium-independent release of glutamate.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 7. Miniature EPSCs are resistant to A, upper and lower panels show evoked and miniature EPSCs, respectively, recorded from representative cells, in control and in the presence of
| ||
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The present study demonstrates that N- and P/Q-type calcium channels are involved in glutamate release from presynaptic terminals in the SON. Furthermore, OXT preferentially modulates the N-type calcium channel-dependent pathway to cause a decrease in the evoked EPSC in this nucleus. This finding is supported by the observation that OXT does not affect the spontaneous release of glutamate, which is independent of calcium influx through voltage-gated calcium channels.
N- and P/Q-type calcium channels mediate the evoked EPSC in the SON
To enable us to examine the possible interaction between OXT receptor and presynaptic calcium current, it is necessary first to identify the types of calcium channels underlying excitatory transmitter release in the SON. We have accomplished this by using relatively selective calcium channel blockers to interfere with the excitatory synaptic transmission. First of all, a low concentration of cadmium, which would block the HVA calcium channels, abolished the evoked EPSCs. In contrast, nickel, which blocks LVA channels, namely T-type, was without effect. Thus, the candidates underlying the excitatory synaptic transmission are the HVA channels: N-, P/Q-, L- and R-type. Nickel also partially blocks R-type channels at the dose we used, thus R-type channel contribution would be minimal. Nicardipine had no effect on evoked EPSCs, indicating little contribution of L-type channels. In contrast, -CTx and -Aga, N- and P/Q-type calcium channel blockers, respectively, significantly reduced the evoked EPSCs. The locus of their action is most likely at the presynaptic excitatory terminals. This is because no change was observed in any of the parameters that indicate postsynaptic effect, i.e. the postsynaptic passive membrane properties, and the kinetics of AMPA receptor-mediated responses. Nonetheless, we cannot decisively reject the possibility that these blockers act at upstream, axo-axonal synapses to indirectly modulate glutamatergic transmission, by suppressing release of known inhibitory transmitters in the SON (Kombian et al. 1996; Richard & Bourque, 1996; Oliet & Poulain, 1999). However, such an effect should increase the evoked EPSC amplitude and decrease the paired-pulse ratio, effects opposite to those found in the present study.
In many central synapses different calcium channels co-operatively mediate synaptic transmission (Regehr & Mintz, 1994; Wu et al. 1998). The combination of calcium channels in the same presynaptic terminal is thought to provide neurons with a high degree of flexibility in the regulation and modulation of neurotransmitter release (Fossier et al. 1999). In the SON, N- and P/Q-type channels appear to contribute substantially to synaptic transmission. Furthermore, since both channel blockers exhibit blockade over a wide range of stimulus intensity, both channel types may be recruited concurrently and be involved in basal, as well as enhanced transmission. Because it has been suggested that the sensitivity of P- and Q-type calcium channels to -Aga differs (Teramoto et al. 1995), we tentatively suggest that both P- and Q-type calcium channels are present in these terminals and mediate glutamate release in the SON. We acknowledge, however, that we cannot differentiate complete block of P-type and partial block of Q-type current by the low concentration of -Aga, because in the presynaptic terminals, we cannot measure the calcium currents directly.
An important question arising from the above finding is, do these channels colocalise on the same terminal, acting synergistically to mediate transmitter release? This question cannot be adequately answered until we get a better understanding of these synapses and the relationship between calcium influx and transmitter release. From a mathematical standpoint, however, we can make certain inferences or predictions by calculating the proportion of each channel's contribution to transmitter release using the fraction of EPSC sensitive to the selective channel blocker. A power relationship between calcium influx and transmitter release has been shown in well-characterised synapses, with a power coefficient of 2-4 (Mintz et al. 1995; Reid et al. 1998). Because both -Aga and -CTx were effective in suppressing EPSCs,
Release 
(IAga + ICTx + Iother)n,
where IAga is -Aga-sensitive current, ICTx is -CTx-sensitive current, Iother is the current insensitive to those two toxins. The relative amplitude of EPSCs remaining after application of toxins -Aga, -CTx or both was 0.55, 0.41 and 0.05, respectively. If we assume that total calcium influx IAga + ICTx + Iother = 1, and n = 3, the relative contributions of each current can be calculated as follows: 0.55 = (1 - IAga)3, IAga = 0.18. Similarly, 0.41 = (1 - ICTx)3, ICTx = 0.26. Iother would be mediated by R-type channels, which are insensitive to -CTx and -Aga and, as described above, insensitive to nicardipine (Wu et al. 1998). This fraction of current can be calculated as Iother = 1 - (0.18 + 0.26) = 0.56. Then the estimated remaining EPSC after combined application of -Aga and -CTx would be 0.563 = 0.176. However, our experimental value, 0.05, is less than a third of this estimated value. This calculation indicates that if all N-, P/Q- and R-type channels contribute to transmitter release, they cannot be homogeneously distributed on the same terminal. While the above calculations depend upon certain assumptions, it should be pointed out that similar conclusions are obtained with power coefficients ranging from 2 to 4, the value obtained in other synapses examined (Mintz et al. 1995; Reid et al. 1998).
Alternatively, if each terminal has different types of dominant calcium channels, blockade of transmitter release by maximally effective concentrations of toxins would be all or none in each terminal. Then the effect of toxins would be additive when applied together, as follows:
Release (IAga)n + (ICTx)n + (Iother)n.
In the present study, -CTx and -Aga showed an additive effect, supporting this hypothesis. Furthermore, complete blockade of EPSCs by -CTx and -Aga suggests that toxin-insensitive currents, namely R-type, have little contribution. Although a rather high variation in the data precludes us from making a conclusive remark from this calculation, it further supports our result showing lack of effect of nickel on evoked EPSCs. Thus we suggest that the glutamate release in the SON is mediated by N- and P/Q-type channels, located on different terminals. Although a possibility remains that two distinct populations of MCNs, namely OXT and vasopressin neurons, receive a different profile of synaptic inputs, our data showing that both -CTx and -Aga had a partial inhibitory effect in all the cells tested suggests both populations of MCNs receive both types of input. Identification of cell types being recorded will be needed to confirm this possibility.
Similar heterogeneous distribution of calcium channels amongst different presynaptic terminals has been observed in other brain regions (Reid et al. 1998). It has been reported that in the hippocampus the terminals of interneurons impinging on pyramidal neurons possess either N- or P/Q-type channels only, depending on their origin (Poncer et al. 1997). In the neurohypophysis, Q-type channels control vasopressin secretion from magnocellular terminals (Wang et al. 1997) while R-type channels mediate OXT release (Wang et al. 1999). Having a unique distribution of calcium channels on the terminals may make it possible for neuromodulators to gate the afferent input in a highly specific manner.
Oxytocin depresses the evoked EPSCs by modulating mainly N-type calcium channels
We reported earlier that OXT, released from the dendrites of MCNs within the SON, could retrogradely target presynaptic excitatory terminals via OXT receptors to decrease synaptic transmission (Kombian et al. 1997). In the present study we showed that virtually all of this OXT effect, both exogenous and endogenous (i.e. brought about by local aminopeptidase inhibition), is occluded in the presence of -CTx, whereas a much lesser occlusion was seen in the presence of a maximal concentration of -Aga. Thus, OXT seems to affect N- or P/Q-type calcium channels to different degrees. This suggests that a presynaptic waveform change from a potassium current activation is unlikely, because such a change would most likely affect all types of calcium channels similarly (Sabatini & Regehr, 1997). The subtle difference between the effect of OXT and amastatin we observed could be due to the fact that amastatin would also increase the level of endogenous vasopressin. This could produce a complex effect because vasopressin is also known to modulate evoked EPSCs (Kombian et al. 2000b). Another caveat in this result with amastatin comes from the possibility that calcium channel blockers may act on the release process of neuropeptides at the postsynaptic dendrites. It is not possible to distinguish between these possibilities and further investigation will be required to directly analyse the subtypes of calcium channels involved in dendritic release.
One likely possibility is that OXT receptors interact with the calcium channels to alter channel kinetics leading to a decrease in calcium influx. Consistent with the critical roles of N- and P/Q-type channels in the neurotransmitter release (Wu & Saggau, 1997), these channels have been shown to be highly regulated by neuromodulator receptors coupled to G-proteins and/or intracellular protein phosphatases (Dolphin, 1996) by shifting the voltage dependence of the calcium channels or moving the channel conductance to a different state (Bean, 1989). Our present finding that OXT has a greater effect on -CTx over -Aga-sensitive currents is in keeping with the reports showing that N-type currents are subject to stronger inhibition by G-proteins than are P/Q-type currents (Currie & Fox, 1997). It is plausible to speculate that G-protein mediates this effect of OXT, since most OXT effects are thought to be G-protein mediated (Thibonnier et al. 1998). It is obvious from our data that OXT action upon N- and P/Q-type channels is not absolute, given that substantially more current is blocked by the respective toxins than is reduced by OXT. This reinforces the idea that OXT action at this synapse is a modulatory one, rather than a complete blockade of excitation.
OXT has been reported to promote extracellular calcium-independent calcium release from the internal stores in MCNs (Lambert et al. 1994). In spinal cord neurons, OXT was found to produce calcium-dependent facilitation of excitatory synaptic transmission (Jo et al. 1998). Our finding of an inhibitory action of OXT is in contrast to these reports of excitatory effects. It is possible that these glutamate terminals possess OXT receptors with different pharmacological profiles. However, even pharmacologically identical receptors can exert their actions via distinct mechanisms. For example, baclofen-sensitive receptors are known to be coupled to various ionic channels and release machinery (Dittman & Regehr, 1996; Shen & Slaughter, 1999), and have excitatory as well as inhibitory effects on distinct calcium channel subtypes (Shen & Slaughter, 1999) and synaptic efficacy (Brenowitz et al. 1998).
OXT has no effect on miniature EPSCs
OXT may also modulate transmitter release processes downstream of calcium entry, especially if different calcium channels employ different mechanisms to trigger transmitter release. This type of modulation could be expected to alter the probability of calcium-independent spontaneous release (reflected in the frequency of mEPSCs), in parallel with action potential-dependent release (Dittman & Regehr, 1996). Our observation that OXT did not cause detectable changes in mEPSC frequency or amplitude indicates that OXT does not target the release machinery. The present study therefore adds to the growing body of evidence showing that evoked and spontaneous transmitter release can be differentially modulated (Bao et al. 1998; Kombian et al. 2000a).
This selective effect of OXT on evoked EPSCs over mEPSCs raises questions as to its functional consequence. The difference between the two forms of transmission is that while one is a random release of transmitters, the other is timed by invasion of action potentials into the presynaptic terminals. This discrepancy might result in different impacts on the synaptic transmission, since some synaptic plasticity is strictly dependent on the timing of synaptic activity (Markram et al. 1997). Miniature events are now thought to convey functional information: indeed, recent observations indicate that spontaneous transmitter release may have important physiological implications in morphological (McKinney et al. 1999) or functional synaptic plasticity. We have previously reported a unique form of synaptic plasticity in the SON, i.e. a massive increase in mEPSCs that alters the postsynaptic activity (Kombian et al. 2000a), induced by the same stimuli that cause dendritic OXT release (and inhibition of the evoked EPSC) (Kombian et al. 1997). Such alteration of the ratio of spontaneous to evoked release might bias the cell to respond only to certain afferent information. mEPSCs may prime the ion channels on the postsynaptic dendrites through their activation, and significantly influence their activity kinetics and dendritic integration during the subsequent synaptic transmission. History of change in membrane potential and intradendritic calcium concentration has been shown to influence the synaptic integration and plasticity (Spruston et al. 1995). The selective modulatory effect of OXT on action potential-dependent transmitter release may therefore be one of the mechanisms by which the system maintains highly controlled synaptic transmission. Further studies in SON cells under different functional states (e.g. lactation) may provide more information in this regard.
|
REFERENCES |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
| BAO J., LI, J. J. & PERL, E. R. (1998). Differences in Ca2+ channels governing generation of miniature and evoked excitatory synaptic currents in spinal laminae I and II. Journal of Neuroscience 18, 8740-8750 | [Abstract/Full Text] |
| BEAN B. P. (1989). Neurotransmitter inhibition of neuronal calcium currents by changes in channel voltage dependence. Nature 340, 153-156 | [Medline] |
| BOLAND L. M., MORRILL, J. A. & BEAN, B. P. (1994). -Conotoxin block of N-type calcium channels in frog and rat sympathetic neurons. Journal of Neuroscience 14, 5011-5027 |
[Abstract] |
| BRENOWITZ S., DAVID, J. & TRUSSELL, L. (1998). Enhancement of synaptic efficacy by presynaptic GABA(B) receptors. Neuron 20, 135-141 | [Medline] |
| BURBACH J. P. & LEBOUILLE, J. L. (1983). Proteolytic conversion of arginine-vasopressin and oxytocin by brain synaptic membranes. Characterization of formed peptides and mechanisms of proteolysis. Journal of Biological Chemistry 258, 1487-1494 | [Abstract] |
| CURRIE K. P. & FOX, A. P. (1997). Comparison of N- and P/Q-type voltage-gated calcium channel current inhibition. Journal of Neuroscience 17, 4570-4579 | [Abstract/Full Text] |
| DITTMAN J. S. & REGEHR, W. G. (1996). Contributions of calcium-dependent and calcium-independent mechanisms to presynaptic inhibition at a cerebellar synapse. Journal of Neuroscience 16, 1623-1633 | [Abstract] |
| DOLPHIN A. C. (1996). Facilitation of Ca2+ current in excitable cells. Trends in Neurosciences 19, 35-43 | [Medline] |
| DUNLAP K., LUEBKE, J. I. & TURNER, T. J. (1995). Exocytotic Ca2+ channels in mammalian central neurons. Trends in Neurosciences 18, 89-98 | [Medline] |
| FISHER T. E. & BOURQUE, C. W. (1995). Voltage-gated calcium currents in the magnocellular neurosecretory cells of the rat supraoptic nucleus. Journal of Physiology 486, 571-580 | [Abstract] |
| FOEHRING R. C. & ARMSTRONG, W. E. (1996). Pharmacological dissection of high-voltage-activated Ca2+ current types in acutely dissociated rat supraoptic magnocellular neurons. Journal of Neurophysiology 76, 977-983 | [Medline] |
| FOSSIER P., TAUC, L. & BAUX, G. (1999). Calcium transients and neurotransmitter release at an identified synapse. Trends in Neurosciences 22, 161-166 | [Medline] |
| HARAYAMA N., SHIBUYA, I., TANAKA, K., KABASHIMA, N., UETA, Y. & YAMASHITA, H. (1998). Inhibition of N- and P/Q-type calcium channels by postsynaptic GABAB receptor activation in rat supraoptic neurones. Journal of Physiology 509, 371-383 | [Abstract/Full Text] |
| INENAGA K., HONDA, E., HIRAKAWA, T., NAKAMURA, S. & YAMASHITA, H. (1998). Glutamatergic synaptic inputs to mouse supraoptic neurons in calcium-free medium in vitro. Journal of Neuroendocrinology 10, 1-7 | [Medline] |
| ISAACSON J. S. (1998). GABAB receptor-mediated modulation of presynaptic currents and excitatory transmission at a fast central synapse. Journal of Neurophysiology 80, 1571-1576 | [Abstract/Full Text] |
| ISAACSON J. S. & STROWBRIDGE, B. W. (1998). Olfactory reciprocal synapses: dendritic signaling in the CNS. Neuron 20, 749-761 | [Medline] |
| JO Y. H., STOECKEL, M. E., FREUND-MERCIER, M. J. & SCHLICHTER, R. (1998). Oxytocin modulates glutamatergic synaptic transmission between cultured neonatal spinal cord dorsal horn neurons. Journal of Neuroscience 18, 2377-2386 | [Abstract/Full Text] |
| KABASHIMA N., SHIBUYA, I., IBRAHIM, N., UETA, Y. & YAMASHITA, H. (1997). Inhibition of spontaneous EPSCs and IPSCs by presynaptic GABAB receptors on rat supraoptic magnocellular neurons. Journal of Physiology 504, 113-126. | [Abstract] |
| KOMBIAN S. B., HIRASAWA, M., MOUGINOT, D., CHEN, X. & PITTMAN, Q. J. (2000a). Short-term potentiation of miniature excitatory synaptic currents causes excitation of supraoptic neurons. Journal of Neurophysiology 83, 2542-2553 | [Abstract/Full Text] |
| KOMBIAN S. B., MOUGINOT, D., HIRASAWA, M. & PITTMAN, Q. J. (2000b). Vasopressin preferentially depresses excitatory over inhibitory synaptic transmission in the rat supraoptic nucleus in vitro. Journal of Neuroendocrinology 12, 361-367 | [Medline] |
| KOMBIAN S. B., MOUGINOT, D. & PITTMAN, Q. J. (1997). Dendritically released peptides act as retrograde modulators of afferent excitation in the supraoptic nucleus in vitro. Neuron 19, 903-912 | [Medline] |
| KOMBIAN S. B., ZIDICHOUSKI, J. A. & PITTMAN, Q. J. (1996). GABAB receptors presynaptically modulate excitatory synaptic transmission in the rat supraoptic nucleus in vitro. Journal of Neurophysiology 76, 1166-1179 | [Medline] |
| LAMBERT R. C., DAYANITHI, G., MOOS, F. C. & RICHARD, P. (1994). A rise in the intracellular Ca2+ concentration of isolated rat supraoptic cells in response to oxytocin. Journal of Physiology 478, 275-287 | [Abstract] |
| LUDWIG M. & LENG, G. (1998). Intrahypothalamic vasopressin release. An inhibitor of systemic vasopressin secretion? Advance in Experimental Medical Biology 449, 163-173. | |
| MCKINNEY R. A., CAPOGNA, M., DURR, R., GAHWILER, B. H. & THOMPSON, S. M. (1999). Miniature synaptic events maintain dendritic spines via AMPA receptor activation. Nature Neuroscience 2, 44-49 | [Medline] |
| MARKRAM H., LUBKE, J., FROTSCHER, M. & SAKMANN, B. (1997). Regulation of synaptic efficacy by coincidence of postsynaptic APs and EPSPs. Science 275, 213-215 | [Abstract/Full Text] |
| MINTZ I. M., SABATINI, B. L. & REGEHR, W. G. (1995). Calcium control of transmitter release at a cerebellar synapse. Neuron 15, 675-688 | [Medline] |
| NEUMANN I., DOUGLAS, A. J., PITTMAN, Q. J., RUSSELL, J. A. & LANDGRAF, R. (1996). Oxytocin released within the supraoptic nucleus of the rat brain by positive feedback action is involved in parturition-related events. Journal of Neuroendocrinology 8, 227-233 | [Medline] |
| OLIET S. H. & POULAIN, D. A. (1999). Adenosine-induced presynaptic inhibition of IPSCs and EPSCs in rat hypothalamic supraoptic nucleus neurones. Journal of Physiology 520, 815-825 | [Abstract/Full Text] |
| PONCER J. C., MCKINNEY, R. A., GAHWILER, B. H. & THOMPSON, S. M. (1997). Either N- or P-type calcium channels mediate GABA release at distinct hippocampal inhibitory synapses. Neuron 18, 463-472 | [Medline] |
| QIAN J., COLMERS, W. F. & SAGGAU, P. (1997). Inhibition of synaptic transmission by neuropeptide Y in rat hippocampal area CA1: modulation of presynaptic Ca2+ entry. Journal of Neuroscience 17, 8169-8177 | [Abstract/Full Text] |
| REGEHR W. G. & MINTZ, I. M. (1994). Participation of multiple calcium channel types in transmission at single climbing fiber to Purkinje cell synapses. Neuron 12, 605-613 | [Medline] |
| REID C. A., BEKKERS, J. M. & CLEMENTS, J. D. (1998). N- and P/Q-type Ca2+ channels mediate transmitter release with a similar cooperativity at rat hippocampal autapses. Journal of Neuroscience 18, 2849-2855 | [Abstract/Full Text] |
| RICHARD D. & BOURQUE, C. W. (1996). Atrial natriuretic peptide modulates synaptic transmission from osmoreceptor afferents to the supraoptic nucleus. Journal of Neuroscience 16, 7526-7532 | [Abstract/Full Text] |
| SABATINI B. L. & REGEHR, W. G. (1997). Control of neurotransmitter release by presynaptic waveform at the granule cell to Purkinje cell synapse. Journal of Neuroscience 17, 3425-3435 | [Abstract/Full Text] |
| SHEN W. & SLAUGHTER, M. M. (1999). Metabotropic GABA receptors facilitate L-type and inhibit N-type calcium channels in single salamander retinal neurons. Journal of Physiology 516, 711-718 | [Abstract/Full Text] |
| SPRUSTON N., SCHILLER, Y., STUART, G. & SAKMANN, B. (1995). Activity-dependent action potential invasion and calcium influx into hippocampal CA1 dendrites. Science 268, 297-300 | [Medline] |
| TASKER J. G., HOFFMAN, N. W. & DUDEK, F. E. (1991). Comparison of three intracellular markers for combined electrophysiological, morphological and immunohistochemical analyses. Journal of Neuroscience Methods 38, 129-143 | [Medline] |
| TERAMOTO T., NIIDOME, T., MIYAGAWA, T., NISHIZAWA, Y., KATAYAMA, K. & SAWADA, K. (1995). Two types of calcium channels sensitive to -agatoxin-TK in cultured rat hippocampal neurones. NeuroReport. 6, 1684-1688 |
[Medline] |
| THIBONNIER M., BERTI-MATTERA, L. N., DULIN, N., CONARTY, D. M. & MATTERA, R. (1998). Signal transduction pathways of the human V1-vascular, V2-renal, V3-pituitary vasopressin and oxytocin receptors. Progress in Brain Research 119, 147-161 | [Medline] |
| WANG G., DAYANITHI, G., KIM, S., HOM, D., NADASDI, L., KRISTIPATI, R., RAMACHANDRAN, J., STUENKEL, E. L., NORDMANN, J. J., NEWCOMB, R. & LEMOS, J. R. (1997). Role of Q-type Ca2+ channels in vasopressin secretion from neurohypophysial terminals of the rat. Journal of Physiology 502, 351-363 | [Abstract] |
| WANG G., DAYANITHI, G., NEWCOMB, R. & LEMOS, J. R. (1999). An R-type Ca2+ current in neurohypophysial terminals preferentially regulates oxytocin secretion. Journal of Neuroscience 19, 9235-9241 | [Abstract/Full Text] |
| WU L. G., BORST, J. G. & SAKMANN, B. (1998). R-type Ca2+ currents evoke transmitter release at a rat central synapse. Proceedings of the National Academy of Sciences of the USA 95, 4720-4725 | [Abstract/Full Text] |
| WU L. G. & SAGGAU, P. (1997). Presynaptic inhibition of elicited neurotransmitter release. Trends in Neurosciences 20, 204-212 | [Medline] |
| YAKEL J. L., VISSAVAJJHALA, P., DERKACH, V. A., BRICKEY, D. A. & SODERLING, T. R. (1995). Identification of a Ca2+/calmodulin-dependent protein kinase II regulatory phosphorylation site in non-N-methyl-D-aspartate glutamate receptors. Proceedings of the National Academy of Sciences of the USA 92, 1376-1380 | [Abstract] |
| ZILBERTER Y., KAISER, K. M. & SAKMANN, B. (1999). Dendritic GABA release depresses excitatory transmission between layer 2/3 pyramidal and bitufted neurons in rat neocortex. Neuron 24, 979-988 | [Medline] |
| ZUCKER R. S. (1989). Short-term synaptic plasticity. Annual Reviews in Neuroscience 12, 13-31 | [Medline] |
Acknowledgements
We thank Drs T. Fisher, K. Lukowiak, B. A. MacVicar, W. Wildering and G. W. Zamponi for suggestions and critical comments on the manuscript. The work was supported by MRC (Canada) grants to Q.J.P. and Kuwait University grant no. FPT 116 to S.B.K. M.H. is supported by the Heart and Stroke Foundation of Canada and CIHR. Q.J.P. is an MRC and AHFMR Senior Scientist.
Corresponding author
M. Hirasawa: Department of Physiology and Biophysics, University of Calgary, 3330 Hospital Drive NW, Calgary, AB, Canada, T2N 4N1.
Email: hirasawa{at}ucalgary.ca
This article has been cited by other articles:
|
|
 |
|
  M. E. Quinlan, C. O. Alberto, and M. Hirasawa Short-term potentiation of mEPSCs requires N-, P/Q- and L-type Ca2+ channels and mitochondria in the supraoptic nucleus J. Physiol., July 1, 2008; 586(13): 3147 - 3161. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.-F. Wang and G. I. Hatton Dominant Role of {beta}{gamma} Subunits of G-Proteins in Oxytocin-Evoked Burst Firing J. Neurosci., February 21, 2007; 27(8): 1902 - 1912. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. G. Kozoriz, J. B. Kuzmiski, M. Hirasawa, and Q. J. Pittman Galanin Modulates Neuronal and Synaptic Properties in the Rat Supraoptic Nucleus in a Use and State Dependent Manner J Neurophysiol, July 1, 2006; 96(1): 154 - 164. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. A. Anderson, K. A. Feathergill, D. P. Waller, and L. J. D. Zaneveld SAMMA Induces Premature Human Acrosomal Loss by Ca2+ Signaling Dysregulation J Androl, July 1, 2006; 27(4): 568 - 577. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y.-F. Wang and G. I. Hatton Mechanisms Underlying Oxytocin-Induced Excitation of Supraoptic Neurons: Prostaglandin Mediation of Actin Polymerization J Neurophysiol, June 1, 2006; 95(6): 3933 - 3947. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Hirasawa, Y. Schwab, S. Natah, C. J. Hillard, K. Mackie, K. A. Sharkey, and Q. J. Pittman Dendritically released transmitters cooperate via autocrine and retrograde actions to inhibit afferent excitation in rat brain J. Physiol., September 1, 2004; 559(2): 611 - 624. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. L. Soldo, D. R. Giovannucci, E. L. Stuenkel, and H. C. Moises Ca2+ and frequency dependence of exocytosis in isolated somata of magnocellular supraoptic neurones of the rat hypothalamus J. Physiol., March 15, 2004; 555(3): 699 - 711. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Hirasawa, D. Mouginot, M. G. Kozoriz, S. B. Kombian, and Q. J. Pittman Vasopressin Differentially Modulates Non-NMDA Receptors in Vasopressin and Oxytocin Neurons in the Supraoptic Nucleus J. Neurosci., May 15, 2003; 23(10): 4270 - 4277. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. P. J. de Kock, K. D. B. Wierda, L. W. J. Bosman, R. Min, J.-J. Koksma, H. D. Mansvelder, M. Verhage, and A. B. Brussaard Somatodendritic Secretion in Oxytocin Neurons Is Upregulated during the Female Reproductive Cycle J. Neurosci., April 1, 2003; 23(7): 2726 - 2734. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||
