The Estrogen Receptor Modulator Tamoxifen Enhances Spontaneous Glycinergic Synaptic Inhibition of Hypoglossal Motoneurons
http://www.100md.com
《内分泌学杂志》
Institut National de la Santé et de la Recherche Médicale, Unité 488, Stérodes et Système Nerveux, 94276 Le Kremlin-Bicêtre Cedex, France
Abstract
Tamoxifen (Tam), a widely used anticancer agent, is now also used for healthy women with risk of breast cancer. Furthermore, it is the prototype of the selective estrogen receptor modulator family, with promise for neuroprotection. However, possible effects on neurotransmission have been little explored. Recently, Tam was shown to potentiate chloride responses to low concentrations of exogenous glycine in cultured spinal neurons from rat embryo. The present study investigates the possible modulation by Tam of the spontaneous synaptic glycinergic activity recorded from voltage-clamped hypoglossal motoneurons, using the whole-cell patch-clamp technique in brainstem slices from juvenile rat. Miniature currents were isolated with tetrodotoxin. Tam increased the mean amplitude of glycinergic miniature currents, by 68–79% at 2 μM (in nine of 10 cells) and by 47% at 0.5 μM (in four of nine cells). Furthermore, Tam markedly increased the frequency of glycinergic miniatures, by a factor reaching 15 in some neurons, even in the presence of the Ca2+ channel blocker Cd2+. Tam also increased the frequency of the total spontaneous glycinergic activity without tetrodotoxin. The increase in miniature amplitude is consistent with the increase in postsynaptic glycine receptor sensitivity previously reported. The increase in frequency indicates an additional presynaptic effect. Addition of exogenous glycine could also increase the frequency of glycinergic miniatures. Thus, one of the presynaptic effects of Tam might be potentiation of the basal activity of presynaptic glycine receptors facilitating glycine release. Possible risks related to modulation of glycinergic neurotransmission by Tam should be considered when recommending its use in healthy individuals.
Introduction
TAMOXIFEN (NOLVADEX; Tam) is one of the most commonly used anticancer agents. It is now used not only for treatment of breast cancer but has also emerged as a preventive agent even for healthy women with risk factors (1). In addition, new derivatives of Tam are under study for treatment of tumors that are resistant to presently available compounds (2, 3). Tam is also the prototype of a novel generation of molecules, the selective estrogen receptor modulators, under study not only for cancer treatment and prevention, but also as alternatives to traditional hormonal replacement therapy after menopause, as well as for neuroprotection. The effect of Tam and its derivatives in the nervous system are still little explored (for known effects of Tam on ionic channels see discussion, see also Ref.4). Tam not only crosses the brain-blood barrier, but its concentration has been reported to be much higher in the brain than in the serum (5). Its serum concentration in patients taking the usual therapeutic dose (20 mg/d) can reach 0.3 μM (see web site http://www.nolvadex.com). In clinical trials using higher concentrations of Tam, a clear neurological toxicity, including severe motor deficits, appeared to be dose limiting (6). Even at conventional therapeutic doses, Tam can induce some adverse neurophysiological effects, including hot flushes, nausea, increased pain, dizziness, fatigue, depression, and memory impairment (see web site http://www.biam2.org/www/Sub1716.html; see also Ref.7 and included references). However, this selective estrogen receptor modulator is also considered as a potentially useful neuroprotective agent (see, for example, Refs.8 and 9 for neuroprotection against glutamate toxicity in glial cells or cultured neurons; see Ref.10 for neuroprotection against kainic acid toxicity in the hippocampus; see Ref.11 for neuroprotection by high concentrations of Tam in a rat model of human stroke).
Strychnine-sensitive glycine receptors are very abundant in the nervous system, and glycinergic neurotransmission is particularly important in the spinal cord and brainstem, where it is one of the major forms of synaptic inhibition in the adult. Glycinergic neurotransmission is involved not only in motor control and pain sensation, but also in other functions, such as vision and audition (for review see Ref.12).
In a previous study performed on cultured spinal cord neurons dissociated from rat embryos (4), it was shown that at concentrations ranging from 0.2–5 μM, Tam potentiates responses to low concentrations of glycine, by decreasing the EC50 of glycine receptors. Glycine was applied exogenously, by fast perfusion of the whole neuron, and thus activated both synaptic and extrasynaptic glycine receptors. Responses to high concentrations of glycine were not potentiated and could even be reduced by Tam. Possible modulation of synaptic glycinergic currents was not investigated. The aim of the present study was to determine whether Tam also potentiates some form of glycinergic synaptic activity, in identified neurons expressing adult postsynaptic glycine receptors, with a preparation preserving glycinergic synapses. Rat brainstem slices containing the hypoglossal nucleus were selected for this study. The motoneurons of this nucleus are easily identified and their glycinergic responses have already been well studied: hypoglossal motoneurons already express mature glycine receptors in juvenile animals (from postnatal d 10) and receive an intense spontaneous glycinergic synaptic activity even after blockade of action potentials (13). From previous results (4), this miniature synaptic activity appeared to be the best synaptic candidate for potentiation by Tam.
Materials and Methods
Hypoglossal motoneurons were recorded in the whole-cell configuration of the patch-clamp technique on transverse brainstem slices (300 μm thick) from juvenile Sprague Dawley rats (11–14 d old) of both sexes (bred in laboratory animal house). After rapid decapitation (with a guillotine, as approved by the European Commission, directive 86/609/EEC), a block containing the cerebellum and brainstem was quickly removed, cleaned, and stuck on the vibratome platform. Slicing was performed at 4 C in a low Ca2+ extracellular solution, continuously bubbled with 95% O2/5% CO2, containing (in mM): 125 NaCl, 2.5 KCl, 1.25 NaH2PO4, 26 NaHCO3, 5 MgCl2, 0.1 CaCl2, and 0.4 sodium ascorbate. The slices were then incubated for half hour at 36 C in the extracellular solution containing (in mM): 125 NaCl, 2.5 KCl, 1.25 NaH2PO4, 26 NaHCO3, 1 MgCl2, 2 CaCl2, always bubbled with 95% O2/5% CO2. Equilibration at room temperature was then allowed for at least 1 h before recording. Hypoglossal motoneurons were easily identified from the characteristic location and aspect of the hypoglossal nucleus. The experiments were performed at room temperature, while continuously perfusing the bath with an oxygenated solution at a rate close to 3 ml/min. The control solution usually also contained 1 mM kynurenic acid and 5 μM 1(S), 9(R)-(-)-bicuculline methochloride (from Sigma, Schnelldorf, Germany, via Sigma, St. Quentin Fallavier, France) to block glutamatergic and -aminobutyric acid (GABA)ergic synaptic currents, as well as ethanol for comparison with the Tam-containing solution (see Results) and, in the case of miniature currents recordings, 0.5 μM tetrodotoxin (TTX with citrate, from Alomone Labs). Some experiments were performed by blocking glutamatergic synaptic responses with 10 μM 6-cyano-7-nitroquinoxaline-2,3-dione disodium salt (CNQX, from Sigma) and 25 μM D(-)-2-amino-5-phosphonopentanoic acid (D-AP-5, from Precision Biochemicals, Vancouver, British Columbia, Canada) instead of kynurenic acid. The stock solution of Tam citrate (from Tocris, Avonmouth, UK) was prepared each day in ethanol, usually at 10 mM (by sonication for 15 min); a tube containing only ethanol was handled in the same way for dilution in the control solution. A Teflon tap with four inputs (connected by Teflon tubing to glass syringes and to the recording chamber) allowed arrival of the selected solution to the chamber within 20–30 sec. However, because equilibration inside 300-μm slices may be much slower, the control solution was always applied for at least 6 min before the first control recording. All solutions were continuously bubbled with 95% O2/5% CO2. Because of its possible irreversible effects, once Tam had been applied to the bath, the slice was systematically replaced by a new one for the next experiment.
All of the effects reported in the present paper were studied by comparing the record obtained in the presence of the drug tested with the control record from the same neuron (because of the variability in synaptic activity from one motoneuron to the other, comparisons were not performed from data pooled from different neurons). Combined with the rather slow onset and reversibility of the effects of Tam, this method limited the number of neurons that were included in the analysis (recording conditions needed to be stable for a long duration). However, it allowed convincing comparisons in each neuron, independently of poorly controlled parameters.
Patch-clamp electrodes (1.8–2.5 M) were prepared from borosilicate glass capillaries (Hilgenberg 1406105) and were filled with (in mM): 150 CsCl (suprapur from Merck, Darmstadt, Germany), 10 NaCl, 1 MgCl2, 0.1 EGTA, 3 ATP-Mg, 0.1 GTP-Na, and 10 HEPES. The internal pH was adjusted to 7.2 using CsOH. For experiments performed in the absence of TTX, 10 mM N-(2,6-dimethylphenylcarbamoylmethyl) triethylammonium chloride (QX-314, from Alomone) was also added to the internal solution. Its final osmolarity was adjusted 0–10 mosmol below that of the extracellular solution, using cesium methane sulfonate or distilled water as necessary.
Hypoglossal motoneurons were voltage-clamped at –70 mV, using an Axopatch 200B amplifier, a Digidata 1322A interface, and pClamp8 software [all from Axon Instruments (Union City, CA, via DIPSI Industrie, Chatillon, France)]. The current was filtered at 1 kHz, sampled at 2 kHz, and stored on a computer. Access resistance was systematically checked before and after each experiment, and the experiments during which it increased above 15 M were eliminated. According to the neuron, the membrane resistance was between 50 and 200 M and did not change by more than 5–15% between the beginning and the end of the experiment. Recordings during which abrupt changes in membrane resistance spontaneously occurred were discarded. However, a slight and slow shift of the baseline current was often observed at –70 mV (slight increase in inward current often observed during the initial part of the experiment, sometimes followed by a reverse shift). These shifts appeared to be independent of exogenous extracellular applications (of Tam or glycine for example). Because of these slow spontaneous shifts, the present study cannot exclude some small and slowly developing effect of Tam on membrane resistance. Nevertheless, no rapid membrane resistance change was ever observed upon addition of Tam. Unless otherwise indicated, the records illustrated have not been corrected for baseline.
Synaptic events were automatically detected using MiniAnalysis software (Synaptosoft, Fort Lee, NJ) and were then checked visually. If spontaneous opening/closing of ionic channels (characterized by an abrupt offset and a reproducible amplitude above the baseline noise) were detected, the experiment was eliminated to avoid possible confusion with synaptic events. Several parameters, including peak amplitude and time to decay to 50% of the peak (T0.5), were measured for each detected event, and mean values were obtained from 2- to 4-min continuous recording. Cumulative histograms were built, in particular for peak amplitudes and inter-event intervals. Comparisons concern histograms obtained in different conditions from a given neuron, by including all the events detected in each condition in this neuron. In most cases, histograms were built from several hundreds of events, and in any case, from at least 100 events. The mean frequency was also estimated by dividing the total number of events by the recording duration.
Data are reported as mean ± SD, except for average plots of inter-event intervals (see Fig. 4) giving mean ± SEM. Final illustrations were prepared using Kaleidagraph (Synergy software, Reading, PA, via Logi Labo, Paris, France).
Results
Tam increases the amplitude and frequency of glycinergic miniature currents in hypoglossal motoneurons
Glycinergic miniature currents (action-potential-independent spontaneous synaptic events) were recorded in voltage-clamp at –70 mV in hypoglossal motoneurons, while perfusing the slice with kynurenic acid, bicuculline, and TTX. As expected, these events completely disappeared after addition of 1 μM strychnine (Fig. 1A). To investigate the effects of Tam, ethanol (the solvent) was also continuously applied during the entire recording (even in the control solution), at the same final dilution as in the Tam-containing solution (1/5000 unless otherwise noted, corresponding to 3.4 mM). The activity was successively recorded from each neuron in control (for 6–10 min), during Tam application and after washing. As illustrated in Fig. 1, B–E, addition of 2 μM Tam increased both the amplitude and the frequency of glycinergic miniature currents. In the experiment illustrated, the effects of Tam were strong enough to be detected by visual examination of the current traces illustrated at a 4-min time scale (Fig. 1B). These effects were reversed after several minutes washing with control solution. They are quantitatively demonstrated by comparison of histograms: Tam induced both a shift to the right of the cumulative histogram of peak amplitudes (Fig. 1D) and a shift to the left of the cumulative histogram of inter-event intervals (Fig. 1E). Note that because the control frequency was particularly low in this experiment, the control and wash recordings have been merged. Averages of miniature currents recorded in the absence or presence of Tam, illustrated on a 20-msec time scale (Fig. 1C), also show that Tam increased the mean amplitude of glycinergic miniature currents. In contrast, no significant effect of Tam on the decay of miniature currents was detected, neither by comparing normalized average traces (Fig. 1C, right traces), nor by comparing mean decay times or histograms of decay times (data not shown; see below for mean results). Similar results were obtained in the four experiments performed under identical conditions (in 3.4 mM ethanol, without cadmium). The mean percentage increase induced by 2 μM Tam in these conditions was 68 ± 29% (4).
In addition to these experiments, an initial set of experiments had been performed in the presence of a five times higher concentration of ethanol (17 mM), whereas Tam (2 μM) was only applied for 8 min (instead of 12–16 min). A reliable increase in the mean amplitude of glycinergic miniature currents was also observed in five of these 11 initial experiments. The mean percentage increase was 72 ± 31% (5). Interestingly, the motoneurons that did not show this effect were those in which the mean miniature amplitude was the highest in the control period. Thus, even though the conditions of these initial experiments were not ideal (higher concentration of ethanol, shorter duration of the application of Tam), their results reinforce those obtained with longer Tam applications and only 3.4 mM ethanol. A clear increase in miniature frequency was also observed in some cases, either during the perfusion with Tam (two of 11 cells) or a few minutes after the end of its addition, during the beginning of the washing period (three of 11 cells). The control frequency was too variable in four of 11 cells, and no significant effect of Tam on frequency was detected in the remaining two cells.
Experiments were also performed in the presence of 100 μM cadmium chloride (CdCl2), to reduce the calcium influx that might be activated by depolarization of presynaptic terminals. These experiments were performed in the presence of only 3.4 mM ethanol. As illustrated in Fig. 2, Tam could still increase both the amplitude and the frequency of glycinergic miniature currents in the presence of cadmium. In the experiment illustrated, the increase in amplitude was already clear 6–8 min after addition of Tam (Fig. 2B, average traces), but it was maximum only after the 14-min Tam application, during the initial washing period. This indicates that Tam slowly penetrated into the slice and was slowly washed out (see Discussion). The effect of Tam on amplitude was almost completely reversible after 24 min washing (Fig. 2C). In the same way, the increase in frequency was already detectable during Tam application (compare the control trace and the trace recorded at 12–14 min after addition of Tam in Fig. 2A) but developed further during the initial washing period. In this motoneuron, the effect on frequency was particularly slow and strong (increase by a factor of 3.4 12–14 min after Tam addition, maximum increase by a factor of 14.6) and was only partly reversed after 24–26 min washing (Fig. 2D). At 2 μM, Tam induced a reliable increase in miniature amplitude in five of six experiments performed in the presence of cadmium, and the mean percentage increase of miniature amplitude was 79 ± 35% (5). Interestingly, in the cell that did not respond to Tam by an increase in mean miniature amplitude, the mean control value was particularly high (125 pA). At 2 μM, Tam also induced a reliable frequency increase in five of six experiments performed in the presence of cadmium. In the remaining cell, the control frequency was very high and too variable (between 3.6 and 6.7) to allow detection of an effect of Tam. The maximum frequency increase was usually observed during Tam application, between 4 and 12 min after the beginning of perfusion with Tam.
Whatever the experimental conditions used, no effect of Tam (2 μM) was ever detected on the time course of decay of glycinergic miniature currents (as already indicated by the normalized average traces of Fig. 1B; comparison performed for each neuron). The average (±SD) of the mean T0.5 was 4.90 ± 0.70 msec in the controls and 4.89 ± 0.91 msec at the maximum Tam effect for the 10 experiments using 3.4 mM ethanol in which Tam increased the amplitude and/or frequency of glycinergic miniatures. The average of the mean T0.5 was 4.96 ± 0.74 msec in the controls and 4.76 ± 0.91 msec at the maximum Tam effect for the five experiments using 17 mM ethanol in which Tam increased the amplitude of glycinergic miniatures.
Another set of experiments was performed (in the presence of 3.4 mM ethanol) to test possible effects of a lower Tam concentration, 0.5 μM. In these experiments, Tam was added for 12–16 min. As illustrated in Fig. 3, even at 0.5 μM, Tam could increase the mean amplitude and the mean frequency of glycinergic miniature currents. However, each of these effects was only observed with four of nine cells tested. The percentage increase of mean miniature amplitude induced by 0.5 μM Tam was 47 ± 14% (4). In three of the cells that did not respond to Tam by an increase in mean miniature amplitude, the control value was particularly high (> 85 pA), as already noted above for experiments using 2 μM Tam and 17 mM ethanol. Again, no effect of Tam was detected on the time course of decay of miniature currents.
In conclusion, bath application of rather low concentrations of Tam (0.5 or 2 μM) can increase the mean amplitude and the mean frequency of glycinergic miniature currents, even in the presence of cadmium ions, within a time scale of a few minutes. Interestingly, the two effects were not always associated: in a few cases, only one of them was detected. Furthermore, the increase in frequency and the increase in amplitude did not always develop (or reverse) simultaneously (see for example Fig. 2).
Tam also increases the frequency of spontaneous glycinergic currents in the absence of TTX
The glycinergic spontaneous activity of hypoglossal motoneurons was also recorded in the absence of TTX, to detect both miniature synaptic events and action-potential driven spontaneous synaptic events. All of these experiments were performed without cadmium, in bicuculline and 3.4 mM ethanol, after blockade of glutamatergic activity either by kynurenic acid (in most experiments) or by CNQX (10 μM) and D-AP-5 (25 μM). The control spontaneous glycinergic activity was variable from one motoneuron to the other, as illustrated by the range of control mean frequency shown in Fig. 4F (horizontal axis). In most experiments (in nine of 11 in kynurenic acid and in two of three in CNQX and D-AP-5), Tam (2 μM) clearly increased the frequency of the total spontaneous glycinergic activity. In two of three remaining cases, the control frequency was not stable enough to reach a conclusion, whereas in the third, a strong increase in frequency was observed but only after Tam application, during the beginning of the washing period. Figure 4F gives the maximum mean frequency in Tam as a function of the mean control frequency for each successful experiment (without including the experiment described in Fig. 4E). One of the experiments of Fig. 4F is described in more detail in Fig. 4, A–D. The current traces recorded in the control period and with Tam are shown at two different time scales (Fig. 4, A and B), the average inter-event interval each minute is plotted (Fig. 4C), and cumulative histograms of spontaneous current amplitudes and inter-event intervals are also illustrated (Fig. 4D). Tam clearly shortened the mean inter-event interval and shifted the cumulative histogram of inter-event intervals to the left, demonstrating a strong increase in frequency. This effect was partly reversible during washing (Fig. 4D). In contrast, Tam only slightly affected the cumulative histogram of amplitudes. In the absence of TTX, Tam did not usually induce a significant increase of the mean amplitude of spontaneous glycinergic currents. Figure 4E illustrates an additional experiment, in which the effect of Tam on frequency was especially strong and rapid, but was mainly transient: the maximum effect would have been off the scale of Fig. 4F (the maximum frequency was higher than 30 at the beginning of the application of Tam). In this case, a strong increase in amplitude was also transiently induced by Tam, but this was unusual and is not shown. The mean frequency of spontaneous glycinergic synaptic currents increased by a factor of 3.08 ± 2.54 (10) when measured at the maximum of the effect of 2 μM Tam (over a period of at least 1 min, without including the experiment in Fig. 1E). In many experiments, the effect of Tam was partly transient, so that the increase of the mean frequency was only by a factor of 2.03 ± 1.22 (10) after more than 10 min Tam.
Are there presynaptic glycine receptors able to affect the frequency of glycinergic synaptic currents
It is known from studies performed on other preparations that glycine receptors can modulate transmitter release by acting presynaptically, in some cases via a direct effect on presynaptic terminals (see Discussion). It is also known that Tam can markedly potentiate the activity of glycine receptors when the agonist concentration is low (4). Thus, if some glycine receptors were expressed presynaptically, their activation by endogenous agonists (glycine and/or taurine) would be favored by Tam and might influence the frequency of glycinergic synaptic events. To test the possible role of presynaptic glycine receptors, the frequency of glycinergic miniatures was compared before and after addition of exogenous glycine. Detection of miniatures can be hampered by the noise increase resulting from activation of postsynaptic glycine receptors (very abundant on hypoglossal motoneurons). Thus, the concentration of glycine to be added had to be as low as possible. In the experiment illustrated in Fig. 5, A–C, addition of glycine (30 μM) immediately induced, as expected, a slight increase in noise between synaptic events (see current traces in Fig. 5B). After a few minutes, glycine also induced a strong increase in the frequency of miniatures: this is shown both by the current traces illustrated (Fig. 5, A and B), and by the cumulative histograms of inter-event intervals (Fig. 5C). Note the shift to the left induced by glycine and the almost complete reversibility after washing. Figure 5D illustrates another similar experiment performed by adding only 10 μM glycine. In four of seven experiments, addition of glycine (10, 20, or 30 μM) thus increased the frequency of glycinergic miniatures, by a factor ranging from 1.6–8.3. The data obtained did not show any reproducible effect on the mean miniature amplitude. The fact that glycine can increase the mean miniature frequency suggests that some glycinergic terminals express presynaptic glycine receptors that facilitate glycine release when activated.
Discussion
In a previous study performed on cultures of dissociated embryonic spinal neurons, Tam was shown to markedly increase the sensitivity of some glycine receptors to low concentrations of exogenous glycine, whereas it never increased responses to saturating glycine concentrations (4). Here, using brainstem slices that preserve native synaptic connections, it is shown that Tam increases the glycinergic spontaneous synaptic activity in motoneurons known to express adult glycine receptors (13). The observed increase in miniature synaptic activity results both from an increase in the mean amplitude of miniature currents and from an increase in their frequency.
The time scale of the development and reversibility of Tam effects on glycinergic synaptic currents was variable from one motoneuron to another. In a few cases, an effect was observed as soon as the Tam-containing solution reached the recording chamber (in less than 30 sec). However, several minutes were usually necessary to observe the maximum effects, and in some cases, the perfusion with Tam was not long enough and the maximum effects were observed during the beginning of the washing period. The reversibility was also rather slow and not always complete. Several factors can explain this variability and relative slowness. Tam must diffuse through the thickness of the slice to reach all the synapses contributing to the recorded current. Furthermore, this lipophilic compound is likely to accumulate progressively in membranes. Note that, even when Tam was applied to isolated cultured neurons by local fast perfusion, some preincubation was necessary for maximum potentiation; furthermore, complete reversibility was not always observed even after 15 min fast perfusion without Tam (4). Bath perfusion of synaptic contacts in 300 μm thick slices is obviously much more difficult than perfusion of an isolated soma via the strong and optimally directed fluid flow used in our previous study.
The increase in glycinergic miniature current amplitude by Tam is consistent with an increase in glycine sensitivity of postsynaptic receptors, similar to that previously observed in spinal neurons. It indicates that postsynaptic receptors are usually not saturated by the release of a single vesicle. The same conclusion had already been drawn from the potentiating effect of low concentrations of Zn2+ ions on glycinergic miniatures of zebrafish larvae neurons and embryonic rat spinal neurons (14, 15). However, these miniatures were also prolonged by Zn2+, whereas no prolongation was observed here upon addition of Tam. Note that Tam and Zn2+ are known to potentiate glycine responses by different mechanisms (4). The observation that Tam did not prolong glycinergic synaptic responses also contrasts with the prolongation observed upon activation of all G proteins in cultured embryonic spinal neurons (16). Note that this difference might involve differences in experimental conditions (glycine receptor isoforms expressed, recording temperature... ) and does not exclude possible activation of some G proteins by Tam. The effects of Tam also differ from those of volatile anesthetics, which did not affect the mean amplitude of miniature currents, but prolonged their duration in rat spinal motoneurons (17). In contrast, the effects of Tam are qualitatively similar to those of ethanol, which also increased both the amplitude and the frequency of miniature glycinergic currents without affecting their duration in juvenile rat hypoglossal motoneurons (18). The protocol used here (continuous application of the same concentration of ethanol during control, Tam and washing) guarantees that the effects observed upon addition of Tam are actually its own, rather than those of its solvent. Nevertheless, the fact that ethanol can already affect glycinergic miniatures at 10 mM (18) may partly explain why modulation by Tam was not systematically observed when 17 mM ethanol was continuously present. Ethanol shifts the dose-response curve of some glycine receptors toward lower glycine concentrations by increasing their affinity for glycine, as shown in hypoglossal motoneurons (19, 20, 21). The fact that Tam did not increase the mean miniature amplitude when it was already high in control, in particular in 17 mM ethanol, agrees with previous results obtained close to saturation of spinal receptors (4). Usually, Tam did not increase the mean amplitude of spontaneous synaptic responses recorded without TTX (including large amplitude events due to action-potential-dependent simultaneous release of several vesicles). This is consistent with the selective potentiation of responses to small amounts of released glycine.
By increasing the amplitude of the smallest synaptic events, Tam could bring more events above the detection threshold. This could contribute to the observed increase in frequency of miniature and spontaneous synaptic currents. However, this is unlikely to be the only explanation. The increase in miniature frequency was sometimes very marked. Furthermore, it was not always correlated with a change in amplitude. In addition, in the absence of TTX, the frequency increase was usually observed without any change in mean amplitude. Thus, Tam probably also acts at the presynaptic level. Possible mechanisms are discussed below.
The increase in frequency of glycinergic synaptic responses induced by Tam might involve an increase in the presynaptic Ca2+ concentration. However, Tam still increased the miniature frequency in the presence of the Ca2+ channel blocker Cd2+. Thus, the observed frequency increase did not require activation (by a putative Tam-induced depolarization) of the presynaptic Cd2+-sensitive voltage-gated Ca2+ channels usually involved in transmitter release. Note also that known effects of Tam on voltage-gated Ca2+ channels are inhibitory (22, 23), and that, at 1 μM, Tam prevented the intracellular Ca2+ increase induced by addition of external K+ in sensory neurons (Ref.24 ; see also25). Nevertheless, in some nonneuronal cells, Tam can increase intracellular Ca2+ (26, 27). Thus, it remains possible that Tam increases the presynaptic Ca2+ concentration, either by activating some Cd2+-insensitive Ca2+ permeable channels or by releasing intracellular Ca2+.
The present study shows that, at least in some cases, glycine itself can increase the frequency of miniature glycinergic synaptic currents. This suggests that Tam could increase the frequency of glycine release simply by potentiating the activity of some presynaptic glycine receptors (provided that an endogenous agonist is present at a sufficient concentration to allow activation of these receptors, at least in the presence of the potentiator). Note that such a scheme could explain why other potentiators of glycine receptors such as ethanol (18) also increase the frequency of glycinergic miniatures (additional mechanisms remain of course possible). An analogous scheme has already been proposed to explain the increase in frequency of GABAA spontaneous and miniature currents induced by allopregnanolone, a neurosteroid potentiating Cl– responses to low concentrations of GABA (Ref.28 ; for nerve terminals from 4-wk-old rats, see also Refs.29 and 30). It was already known that some glycinergic presynaptic receptors can facilitate transmitter release in a few other brain areas (31, 32, 33). However, these effects were no longer observed in TTX (33) or Cd2+ (31). For technical reasons, these studies, as well as the present one, were performed using young animals. It remains to demonstrate that similar effects persist in the adult, at least at some synapses. This is likely because, even in 4- to 6-wk-old animals, Cl– responses to glycine or GABA can be depolarizing in some interneurons, whereas they are hyperpolarizing in the principal cells (see Ref.34 for glycinergic interneurons in the dorsal cochlear nucleus; see also Refs.30 and 35). Note that, at the age used in the present study, postsynaptic glycine responses of hypoglossal motoneurons are already mature and hyperpolarizing (13) [in younger animals, they can be depolarizing but also inhibit the electrical activity (36)]. If the Cl– concentration is higher in the presynaptic terminal than in the postsynaptic motoneuron, the facilitatory effect of glycine on miniature frequency could be explained by a glycine-induced small presynaptic depolarization, for example activating Cd2+-insensitive voltage-gated Ca2+-channels and/or an intracellular Ca2+ release, possibly in the absence of Ca2+ influx (37).
An increase in cAMP might also be involved in presynaptic effects of Tam. Different types of membrane estrogen receptors have been described recently (for review see Ref.38). Interestingly, some of them are coupled to a Gs protein-stimulating adenylate cyclase and can be activated by Tam (39, 40). Furthermore, it is well known that transmitter release can be increased by cAMP, and modulation of glycine release by this pathway has been reported (Refs.41 and 42 ; see also Ref.43 for hypoglossal motoneurons).
The relief of some tonic inhibition of glycine release by some presynaptic metabotropic receptors could also partly explain the increase in glycine release induced by Tam. Indeed, activation of some membrane estrogen receptors has been reported to uncouple some metabotropic receptors from their associated G protein (44, 45, 46). Finally, Tam can inhibit some protein kinases C (PKC) (47). However, this property is unlikely to explain the observed increase in miniature frequency because it is PKC activation (not inhibition) that facilitates neurotransmitter release by increasing the Ca2+-sensitivity of vesicle fusion, a property likely to be general (see Ref.48 and included references). Tam can also enhance the membrane translocation of PKC in intact cells (49). This other type of effect might facilitate glycine release. Note also that contrasting modulations of glycine responses by PKC have been described in a variety of other neurons (12). Thus, a role for PKC in Tam-induced postsynaptic effects on hypoglossal motoneurons remains possible, even though it was excluded in spinal neurons (4).
The increase in frequency of the total spontaneous glycinergic synaptic activity induced by Tam was often partly transient. This could result from a progressive decrease of the number of releasable presynaptic vesicles after the very strong initial release triggered by Tam [note that, in the hippocampus, the number of releasable synaptic vesicles adjacent to the presynaptic membrane of GABAergic synapses, is decreased 24 h after injection of 17-estradiol benzoate (50)]. Possible effects of Tam on other types of ionic channels might also be inhibitory for action-potential evoked glycine release (22, 23, 51) for partial block of some voltage-gated Ca2+ or TTX-sensitive Na+ channels (see also Refs.52 and 53 for possible activation of some K+ channels).
Potentiation of glycinergic inhibition by Tam might be one of the early mechanisms contributing to its neuroprotective effects in various pathological situations (see Refs.10 and 11 and included references and Ref.9). Increasing tonic glycinergic activity, by increasing both the synaptic spontaneous glycinergic activity (present results) and possible nonsynaptic responses to the ambient concentration of glycinergic agonists, might oppose the excitotoxicity due to overactivation of glutamatergic receptors (see Ref.54 for neuroprotection by potentiation of GABA-induced Cl– responses). Note that a strong release of taurine (glycinergic agonist) is induced by various pathological conditions, including ischemia, and largely persists even in animals dialysed with high concentrations of Tam (55, 56). However, overactivation of ligand-gated Cl– channels can also be neurotoxic (see, for example, Ref.57 and included discussion). The neuroprotective effect of Tam also seems to involve a reduction of the release of excitatory amino acids during ischemia (11, 56).
In any case, under normal physiological conditions, the striking effects of Tam on glycinergic neurotransmission and more generally on the glycine receptors present in many brain areas (at some nonglycinergic presynaptic sites and at many nonsynaptic sites) are likely to induce some neurophysiological dysfunction, in particular in motor control. Possible risks related to the effects of Tam on glycinergic synapses and receptors should be taken into account when considering its administration to healthy individuals and patients. For example, Tam might increase the incidence of obstructive sleep apnea. This severe respiratory problem is indeed favored by ethanol, via a modulation of glycinergic neurotransmission in the hypoglossal nucleus (see Ref.21 and included references), which is very similar to the effects described here for Tam.
Acknowledgments
I am very grateful to Dr. Michael Schumacher for having allowed me to introduce patch-clamp stations in Unité 488 and for having encouraged and supported this work from its very beginning. I thank Alain Péchard (Mechanics workshop, Ecole Normale Supérieure, Paris) for his major contribution to the building of the experimental set-up. I also thank Boris Barbour, Mariano Casado, and Stephane Supplisson for their help and advice concerning the slice recording set-up and data processing. I am also grateful to Stephane Dieudonné and Marie-Pierre Morin-Surun for their advice and help in preparing the first brainstem slices. Finally, I thank Krzysztof Rajkowski for helpful comments on the manuscript.
Footnotes
Abbreviations: CNQX, 6-Cyano-7-nitroquinoxaline-2,3-dione disodium salt; D-AP-5, D(-)-2-amino-5-phosphonopentanoic acid; GABA, -aminobutyric acid; PKC, protein kinase C; Tam, tamoxifen; TTX, tetrodotoxin.
References
Schwartz J 2004 Tamoxifen: an emerging preventive. Front Biosci. 9:2827–2847
Sugimoto Y, Tsukahara S, Imai Y, Sugimoto Y, Ueda K, Tsuruo T 2003 Reversal of breast cancer resistance protein-mediated drug resistance by estrogen antagonists and agonists. Mol Cancer Ther. 2:105–112
Top S, Vessières A, Leclercq G, Quivy J, Tang J, Vaissermann J, Huché M, Jaouen G 2003 Synthesis, biochemical properties and molecular modelling studies of organometallic specific estrogen receptor modulators (SERMs), the ferrocifens and hydroxyferrocifens: evidence for an antiproliferative effect of hydroxyferrocifens on both hormone-dependent and hormone-independent breast cancer cell lines. Chem Eur J. 9:5223–5236
Chesnoy-Marchais D 2003 Potentiation of glycine responses by dideoxyforskolin and tamoxifen in rat spinal neurons. Eur J Neurosci. 17:681–691
Lien EA, Wester K, Lonning PE, Solheim E, Ueland PM 1991 Distribution of tamoxifen and metabolites into brain tissue and brain metastases in breast cancer patients. Br J Cancer. 63:641–645
Trump DL, Smith DC, Ellis PG, Rogers MP, Schold SC, Winer EP, Panella TJ, Jordan VC, Fine RL 1992 High-dose oral tamoxifen, a potential multidrug-resistance-reversal agent: phase I trial in combination with vinblastine. J Natl Cancer Inst. 84:1811–1816
Eberling JL, Wu C, Tong-Turnbeaugh R, Jagust WJ 2004 Estrogen- and tamoxifen-associated effects on brain structure and function. Neuroimage 21:364–371
Shy H, Malaiyandi L, Timiras PS 2000 Protective action of 17-estradiol and tamoxifen on glutamate toxicity in glial cells. Int J Dev Neurosci. 18:289–297
O’Neill K, Chen S, Diaz Brinton R 2004 Impact of the selective estrogen receptor modulator, tamoxifen, on neuronal outgrowth and survival following toxic insults associated with aging and Alzheimer’s disease. Exp Neurol. 188:268–278
Ciriza I, Carrero P, Azcoitia I, Lundeen SG, Garcia-Segura LM 2004 Selective estrogen receptor modulators protect hippocampal neurons from kainic acid excitotoxicity: differences with the effect of estradiol. J Neurobiol. 61:209–221
Kimelberg HK, Jin Y, Charniga C, Feustel PJ 2003 Neuroprotective activity of tamoxifen in permanent focal ischemia. J Neurosurg. 99:138–142
Lynch JW 2004 Molecular structure and function of the glycine receptor chloride channel. Physiol Rev. 84:1051–1095
Singer JH, Talley EM, Bayliss DA, Berger AJ 1998 Development of glycinergic synaptic transmission to rat brain stem motoneurons. J Neurophysiol. 80:2608–2620
Suwa H, Saint-Amant L, Triller A, Drapeau P, Legendre P 2001 High-affinity zinc potentiation of inhibitory postsynaptic glycinergic currents in the zebrafish hindbrain. J Neurophysiol. 85:912–925
Laube B 2002 Potentiation of inhibitory glycinergic neurotransmission by Zn2+: a synergistic interplay between presynaptic P2X2 and postsynaptic glycine receptors. Eur J Neurosci. 16:1025–1036
Yevenes GE, Peoples RW, Tapia JC, Parodi J, Soto X, Olate J, Aguayo LG 2003 Modulation of glycine-activated ion channel function by G-protein subunits. Nat Neurosci. 6:819–824
Cheng G, Kendig JJ 2002 Pre- and postsynaptic volatile anaesthetic actions on glycinergic transmission to spinal cord motor neurons. Br J Pharmacol. 136:673–684
Eggers ED, O’Brien JA, Berger AJ 2000 Developmental changes in the modulation of synaptic glycine receptors by ethanol. J Neurophysiol. 84:2409–2416
Aguayo LG, Tapia JC, Pancetti FC 1996 Potentiation of the glycine-activated Cl– current by ethanol in cultured mouse spinal neurons. J Pharmacol Exp Ther. 279:1116–1122
Ye JH, Tao L, Ren J, Schaefer R, Krnjevic K, Liu PL, Schiller DA, McArdle JJ 2001 Ethanol potentiation of glycine-induced responses in dissociated neurons of rat ventral tegmental area. J Pharmacol Exp Ther. 296:77–83
Eggers ED, Berger AJ 2004 Mechanisms for the modulation of native glycine receptor channels by ethanol. J Neurophysiol. 91:2685–2695
Sartor P, Vacher P, Mollard P, Dufy B 1988 Tamoxifen reduces calcium currents in a clonal pituitary cell line. Endocrinology 123:534–540
Song J, Standley PR, Zhang F, Joshi D, Gappy S, Sowers JR, Ram JL 1996 Tamoxifen (estrogen antagonist) inhibits voltage-gated calcium current and contractility in vascular smooth muscle from rats. J Pharmacol Exp Ther. 277:1444–1453
Chaban VV, Mayer EA, Ennes HS, Micevych PE 2003 Estradiol inhibits ATP-induced intracellular calcium concentration increase in dorsal root ganglia neurons. Neuroscience 118:941–948
Morley P, Whitfield JF 1994 Effect of tamoxifen on carbachol-triggered intracellular calcium responses in chicken granulosa cells. Cancer Res. 54:69–74
Somjen D, Kohen F, Lieberherr M 1997 Nongenomic effects of an anti-idiotypic antibody as an estrogen mimetic in female human and rat osteoblasts. J Cell Biochem. 65:53–66
Chang H-T, Huang J-K, Wang J-L, Cheng J-S, Lee K-C, Lo Y-K, Liu C-P, Chou K-J, Chen W-C, Su W, Law Y-P, Jan C-R 2002 Tamoxifen-induced increases in cytoplasmic free Ca2+ levels in human breast cancer cells. Breast Cancer Res Treat. 71:125–131
Uchida S, Noda E, Kakazu Y, Mizoguchi Y, Akaike N, Nabekura J 2002 Allopregnanolone enhancement of GABAergic transmission in rat medial preoptic area neurons. Am J Physiol Endocrinol Metab. 283:E1257–E1265
Haage D, Johansson S 1999 Neurosteroid modulation of synaptic and GABA-evoked currents in neurons from the rat medial preoptic nucleus. J Neurophysiol. 82:143–151
Haage D, Druzin M, Johansson S 2002 Allopregnanolone modulates spontaneous GABA release via presynaptic Cl– permeability in rat preoptic nerve terminals. Brain Res. 958:405–413
Turecek R, Trussell LO 2001 Presynaptic glycine receptors enhance transmitter release at a mammalian central synapse. Nature 411:587–590
Kawa K 2003 Glycine facilitates transmitter release at developing synapses: a patch clamp study from Purkinje neurons of the newborn rat. Brain Res Dev Brain Res. 144:57–71
Jeong H-J, Jang I-S, Moorhouse AJ, Akaike N 2003 Activation of presynaptic glycine receptors facilitates glycine release from presynaptic terminals synapsing onto rat spinal sacral dorsal commissural nucleus neurons. J Physiol. 550:373–383
Golding NL, Oertel D 1996 Context-dependent synaptic action of glycinergic and GABAergic inputs in the dorsal cochlear nucleus. J Neurosci. 16:2208–2219
Chavas J, Marty A 2003 Coexistence of excitatory and inhibitory GABA synapses in the cerebellar interneuron network. J Neurosci. 23:2019–2031
Marchetti C, Pagnotta S, Donato R, Nistri A 2002 Inhibition of spinal or hypoglossal motoneurons of the newborn rat by glycine or GABA. Eur J Neurosci. 15:975–983
De Crescenzo V, ZhuGe R, Velazquez-Marrero C, Lifshitz LM, Custer E, Carmichael J, Lai FA, Tuft RA, Fogarty KE, Lemos JR, Walsh Jr JV 2004 Ca2+ syntillas, miniature Ca2+ release events in terminals of hypothalamic neurons, are increased in frequency by depolarization in the absence of Ca2+ influx. J Neurosci. 24:1226–1235
Toran-Allerand CD 2004 Minireview: a plethora of estrogen receptors in the brain: where will it end Endocrinology 145:1069–1074
Razandi M, Pedram A, Greene GL, Levin ER 1999 Cell membrane and nuclear estrogen receptors (ERs) originate from a single transcript: studies of ER and ER expressed in Chinese hamster ovary cells. Mol Endocrinol. 13:307–319
Thomas P, Pang Y, Filardo EJ, Dong J 2005 Identity of an estrogen membrane receptor coupled to a G protein in human breast cancer cells. Endocrinology. 146:624–632
Katsurabayashi S, Kubota H, Wang ZM, Rhee JS, Akaike N 2001 cAMP-dependent presynaptic regulation of spontaneous glycinergic IPSCs in mechanically dissociated rat spinal cord neurons. J Neurophysiol. 85:332–340
Katsurabayashi S, Kubota H, Moorhouse AJ, Akaike N 2004 Differential modulation of evoked and spontaneous glycine release from rat spinal cord glycinergic terminals by the cyclic AMP/protein kinase A transduction cascade. J Neurochem. 91:657–666
Umemiya M, Berger AJ 1995 Presynaptic inhibition by serotonin of glycinergic inhibitory synaptic currents in the rat brain stem. J Neurophysiol. 73:1192–1201
Kelly MJ, Qiu J, Wagner EJ, Ronnekleiv OK 2002 Rapid effects of estrogen on G protein-coupled receptor activation of potassium channels in the central nervous system (CNS). J Steroid Biochem Mol Biol. 83:187–193
Qiu J, Bosch MA, Tobias SC, Grandy DK, Scanlan TS, Ronnekleiv OK, Kelly MJ 2003 Rapid signaling of estrogen in hypothalamic neurons involves a novel G-protein-coupled estrogen receptor that activates protein kinase C. J Neurosci. 23:9529–9540
Mize AL, Young LJ, Alper RH 2003 Uncoupling of 5-HT1A receptors in the brain by estrogens: regional variations in antagonism by ICI 182,780. Neuropharmacology 44:584–591
O’Brian CA, Liskamp RM, Solomon DH, Weinstein IB 1985 Inhibition of protein kinase C by tamoxifen. Cancer Res. 45:2462–2465
Lou X, Scheuss V, Schneggenburger R 2005 Allosteric modulation of the presynaptic Ca2+ sensor for vesicle fusion. Nature 435:497–501
Gundimeda U, Chen ZH, Gopalakrishna R 1996 Tamoxifen modulates protein kinase C via oxidative stress in estrogen receptor-negative breast cancer cells. J Biol Chem. 271:13504–13514
Ledoux VA, Woolley CS 2005 Evidence that disinhibition is associated with a decrease in number of vesicles available for release at inhibitory synapses. J Neurosci. 25:971–976
Hardy SP, deFelipe C, Valverde MA 1998 Inhibition of voltage-gated cationic channels in rat embryonic hypothalamic neurones and C1300 neuroblastoma cells by triphenylethylene antioestrogens. FEBS Lett. 434:236–240
Dick GM, Rossow CF, Smirnov S, Horowitz B, Sanders KM 2001 Tamoxifen activates smooth muscle BK channels through the regulatory 1 subunit. J Biol Chem. 276:34594–34599
Dick GM, Sanders KM 2001 (Xeno)estrogen sensitivity of smooth muscle BK channels conferred by the regulatory 1 subunit: a study of 1 knockout mice. J Biol Chem. 276:44835–44840
Green AR, Hainsworth AH, Jackson DM 2000 GABA potentiation: a logical pharmacological approach for the treatment of acute ischaemic stroke. Neuropharmacology 39:1483–1494
Saransaari P, Oja SS 2000 Taurine and neural cell damage. Amino Acids 19:509–526
Feustel PJ, Jin Y, Kimelberg HK 2004 Volume-regulated anion channels are the predominant contributors to release of excitatory amino acids in the ischemic cortical penumbra. Stroke 35:1164–1168
Allen NJ, Rossi DJ, Attwell D 2004 Sequential release of GABA by exocytosis and reversed uptake leads to neuronal swelling in simulated ischemia of hippocampal slices. J Neurosci. 24:3837–3849(D. Chesnoy-Marchais)
Abstract
Tamoxifen (Tam), a widely used anticancer agent, is now also used for healthy women with risk of breast cancer. Furthermore, it is the prototype of the selective estrogen receptor modulator family, with promise for neuroprotection. However, possible effects on neurotransmission have been little explored. Recently, Tam was shown to potentiate chloride responses to low concentrations of exogenous glycine in cultured spinal neurons from rat embryo. The present study investigates the possible modulation by Tam of the spontaneous synaptic glycinergic activity recorded from voltage-clamped hypoglossal motoneurons, using the whole-cell patch-clamp technique in brainstem slices from juvenile rat. Miniature currents were isolated with tetrodotoxin. Tam increased the mean amplitude of glycinergic miniature currents, by 68–79% at 2 μM (in nine of 10 cells) and by 47% at 0.5 μM (in four of nine cells). Furthermore, Tam markedly increased the frequency of glycinergic miniatures, by a factor reaching 15 in some neurons, even in the presence of the Ca2+ channel blocker Cd2+. Tam also increased the frequency of the total spontaneous glycinergic activity without tetrodotoxin. The increase in miniature amplitude is consistent with the increase in postsynaptic glycine receptor sensitivity previously reported. The increase in frequency indicates an additional presynaptic effect. Addition of exogenous glycine could also increase the frequency of glycinergic miniatures. Thus, one of the presynaptic effects of Tam might be potentiation of the basal activity of presynaptic glycine receptors facilitating glycine release. Possible risks related to modulation of glycinergic neurotransmission by Tam should be considered when recommending its use in healthy individuals.
Introduction
TAMOXIFEN (NOLVADEX; Tam) is one of the most commonly used anticancer agents. It is now used not only for treatment of breast cancer but has also emerged as a preventive agent even for healthy women with risk factors (1). In addition, new derivatives of Tam are under study for treatment of tumors that are resistant to presently available compounds (2, 3). Tam is also the prototype of a novel generation of molecules, the selective estrogen receptor modulators, under study not only for cancer treatment and prevention, but also as alternatives to traditional hormonal replacement therapy after menopause, as well as for neuroprotection. The effect of Tam and its derivatives in the nervous system are still little explored (for known effects of Tam on ionic channels see discussion, see also Ref.4). Tam not only crosses the brain-blood barrier, but its concentration has been reported to be much higher in the brain than in the serum (5). Its serum concentration in patients taking the usual therapeutic dose (20 mg/d) can reach 0.3 μM (see web site http://www.nolvadex.com). In clinical trials using higher concentrations of Tam, a clear neurological toxicity, including severe motor deficits, appeared to be dose limiting (6). Even at conventional therapeutic doses, Tam can induce some adverse neurophysiological effects, including hot flushes, nausea, increased pain, dizziness, fatigue, depression, and memory impairment (see web site http://www.biam2.org/www/Sub1716.html; see also Ref.7 and included references). However, this selective estrogen receptor modulator is also considered as a potentially useful neuroprotective agent (see, for example, Refs.8 and 9 for neuroprotection against glutamate toxicity in glial cells or cultured neurons; see Ref.10 for neuroprotection against kainic acid toxicity in the hippocampus; see Ref.11 for neuroprotection by high concentrations of Tam in a rat model of human stroke).
Strychnine-sensitive glycine receptors are very abundant in the nervous system, and glycinergic neurotransmission is particularly important in the spinal cord and brainstem, where it is one of the major forms of synaptic inhibition in the adult. Glycinergic neurotransmission is involved not only in motor control and pain sensation, but also in other functions, such as vision and audition (for review see Ref.12).
In a previous study performed on cultured spinal cord neurons dissociated from rat embryos (4), it was shown that at concentrations ranging from 0.2–5 μM, Tam potentiates responses to low concentrations of glycine, by decreasing the EC50 of glycine receptors. Glycine was applied exogenously, by fast perfusion of the whole neuron, and thus activated both synaptic and extrasynaptic glycine receptors. Responses to high concentrations of glycine were not potentiated and could even be reduced by Tam. Possible modulation of synaptic glycinergic currents was not investigated. The aim of the present study was to determine whether Tam also potentiates some form of glycinergic synaptic activity, in identified neurons expressing adult postsynaptic glycine receptors, with a preparation preserving glycinergic synapses. Rat brainstem slices containing the hypoglossal nucleus were selected for this study. The motoneurons of this nucleus are easily identified and their glycinergic responses have already been well studied: hypoglossal motoneurons already express mature glycine receptors in juvenile animals (from postnatal d 10) and receive an intense spontaneous glycinergic synaptic activity even after blockade of action potentials (13). From previous results (4), this miniature synaptic activity appeared to be the best synaptic candidate for potentiation by Tam.
Materials and Methods
Hypoglossal motoneurons were recorded in the whole-cell configuration of the patch-clamp technique on transverse brainstem slices (300 μm thick) from juvenile Sprague Dawley rats (11–14 d old) of both sexes (bred in laboratory animal house). After rapid decapitation (with a guillotine, as approved by the European Commission, directive 86/609/EEC), a block containing the cerebellum and brainstem was quickly removed, cleaned, and stuck on the vibratome platform. Slicing was performed at 4 C in a low Ca2+ extracellular solution, continuously bubbled with 95% O2/5% CO2, containing (in mM): 125 NaCl, 2.5 KCl, 1.25 NaH2PO4, 26 NaHCO3, 5 MgCl2, 0.1 CaCl2, and 0.4 sodium ascorbate. The slices were then incubated for half hour at 36 C in the extracellular solution containing (in mM): 125 NaCl, 2.5 KCl, 1.25 NaH2PO4, 26 NaHCO3, 1 MgCl2, 2 CaCl2, always bubbled with 95% O2/5% CO2. Equilibration at room temperature was then allowed for at least 1 h before recording. Hypoglossal motoneurons were easily identified from the characteristic location and aspect of the hypoglossal nucleus. The experiments were performed at room temperature, while continuously perfusing the bath with an oxygenated solution at a rate close to 3 ml/min. The control solution usually also contained 1 mM kynurenic acid and 5 μM 1(S), 9(R)-(-)-bicuculline methochloride (from Sigma, Schnelldorf, Germany, via Sigma, St. Quentin Fallavier, France) to block glutamatergic and -aminobutyric acid (GABA)ergic synaptic currents, as well as ethanol for comparison with the Tam-containing solution (see Results) and, in the case of miniature currents recordings, 0.5 μM tetrodotoxin (TTX with citrate, from Alomone Labs). Some experiments were performed by blocking glutamatergic synaptic responses with 10 μM 6-cyano-7-nitroquinoxaline-2,3-dione disodium salt (CNQX, from Sigma) and 25 μM D(-)-2-amino-5-phosphonopentanoic acid (D-AP-5, from Precision Biochemicals, Vancouver, British Columbia, Canada) instead of kynurenic acid. The stock solution of Tam citrate (from Tocris, Avonmouth, UK) was prepared each day in ethanol, usually at 10 mM (by sonication for 15 min); a tube containing only ethanol was handled in the same way for dilution in the control solution. A Teflon tap with four inputs (connected by Teflon tubing to glass syringes and to the recording chamber) allowed arrival of the selected solution to the chamber within 20–30 sec. However, because equilibration inside 300-μm slices may be much slower, the control solution was always applied for at least 6 min before the first control recording. All solutions were continuously bubbled with 95% O2/5% CO2. Because of its possible irreversible effects, once Tam had been applied to the bath, the slice was systematically replaced by a new one for the next experiment.
All of the effects reported in the present paper were studied by comparing the record obtained in the presence of the drug tested with the control record from the same neuron (because of the variability in synaptic activity from one motoneuron to the other, comparisons were not performed from data pooled from different neurons). Combined with the rather slow onset and reversibility of the effects of Tam, this method limited the number of neurons that were included in the analysis (recording conditions needed to be stable for a long duration). However, it allowed convincing comparisons in each neuron, independently of poorly controlled parameters.
Patch-clamp electrodes (1.8–2.5 M) were prepared from borosilicate glass capillaries (Hilgenberg 1406105) and were filled with (in mM): 150 CsCl (suprapur from Merck, Darmstadt, Germany), 10 NaCl, 1 MgCl2, 0.1 EGTA, 3 ATP-Mg, 0.1 GTP-Na, and 10 HEPES. The internal pH was adjusted to 7.2 using CsOH. For experiments performed in the absence of TTX, 10 mM N-(2,6-dimethylphenylcarbamoylmethyl) triethylammonium chloride (QX-314, from Alomone) was also added to the internal solution. Its final osmolarity was adjusted 0–10 mosmol below that of the extracellular solution, using cesium methane sulfonate or distilled water as necessary.
Hypoglossal motoneurons were voltage-clamped at –70 mV, using an Axopatch 200B amplifier, a Digidata 1322A interface, and pClamp8 software [all from Axon Instruments (Union City, CA, via DIPSI Industrie, Chatillon, France)]. The current was filtered at 1 kHz, sampled at 2 kHz, and stored on a computer. Access resistance was systematically checked before and after each experiment, and the experiments during which it increased above 15 M were eliminated. According to the neuron, the membrane resistance was between 50 and 200 M and did not change by more than 5–15% between the beginning and the end of the experiment. Recordings during which abrupt changes in membrane resistance spontaneously occurred were discarded. However, a slight and slow shift of the baseline current was often observed at –70 mV (slight increase in inward current often observed during the initial part of the experiment, sometimes followed by a reverse shift). These shifts appeared to be independent of exogenous extracellular applications (of Tam or glycine for example). Because of these slow spontaneous shifts, the present study cannot exclude some small and slowly developing effect of Tam on membrane resistance. Nevertheless, no rapid membrane resistance change was ever observed upon addition of Tam. Unless otherwise indicated, the records illustrated have not been corrected for baseline.
Synaptic events were automatically detected using MiniAnalysis software (Synaptosoft, Fort Lee, NJ) and were then checked visually. If spontaneous opening/closing of ionic channels (characterized by an abrupt offset and a reproducible amplitude above the baseline noise) were detected, the experiment was eliminated to avoid possible confusion with synaptic events. Several parameters, including peak amplitude and time to decay to 50% of the peak (T0.5), were measured for each detected event, and mean values were obtained from 2- to 4-min continuous recording. Cumulative histograms were built, in particular for peak amplitudes and inter-event intervals. Comparisons concern histograms obtained in different conditions from a given neuron, by including all the events detected in each condition in this neuron. In most cases, histograms were built from several hundreds of events, and in any case, from at least 100 events. The mean frequency was also estimated by dividing the total number of events by the recording duration.
Data are reported as mean ± SD, except for average plots of inter-event intervals (see Fig. 4) giving mean ± SEM. Final illustrations were prepared using Kaleidagraph (Synergy software, Reading, PA, via Logi Labo, Paris, France).
Results
Tam increases the amplitude and frequency of glycinergic miniature currents in hypoglossal motoneurons
Glycinergic miniature currents (action-potential-independent spontaneous synaptic events) were recorded in voltage-clamp at –70 mV in hypoglossal motoneurons, while perfusing the slice with kynurenic acid, bicuculline, and TTX. As expected, these events completely disappeared after addition of 1 μM strychnine (Fig. 1A). To investigate the effects of Tam, ethanol (the solvent) was also continuously applied during the entire recording (even in the control solution), at the same final dilution as in the Tam-containing solution (1/5000 unless otherwise noted, corresponding to 3.4 mM). The activity was successively recorded from each neuron in control (for 6–10 min), during Tam application and after washing. As illustrated in Fig. 1, B–E, addition of 2 μM Tam increased both the amplitude and the frequency of glycinergic miniature currents. In the experiment illustrated, the effects of Tam were strong enough to be detected by visual examination of the current traces illustrated at a 4-min time scale (Fig. 1B). These effects were reversed after several minutes washing with control solution. They are quantitatively demonstrated by comparison of histograms: Tam induced both a shift to the right of the cumulative histogram of peak amplitudes (Fig. 1D) and a shift to the left of the cumulative histogram of inter-event intervals (Fig. 1E). Note that because the control frequency was particularly low in this experiment, the control and wash recordings have been merged. Averages of miniature currents recorded in the absence or presence of Tam, illustrated on a 20-msec time scale (Fig. 1C), also show that Tam increased the mean amplitude of glycinergic miniature currents. In contrast, no significant effect of Tam on the decay of miniature currents was detected, neither by comparing normalized average traces (Fig. 1C, right traces), nor by comparing mean decay times or histograms of decay times (data not shown; see below for mean results). Similar results were obtained in the four experiments performed under identical conditions (in 3.4 mM ethanol, without cadmium). The mean percentage increase induced by 2 μM Tam in these conditions was 68 ± 29% (4).
In addition to these experiments, an initial set of experiments had been performed in the presence of a five times higher concentration of ethanol (17 mM), whereas Tam (2 μM) was only applied for 8 min (instead of 12–16 min). A reliable increase in the mean amplitude of glycinergic miniature currents was also observed in five of these 11 initial experiments. The mean percentage increase was 72 ± 31% (5). Interestingly, the motoneurons that did not show this effect were those in which the mean miniature amplitude was the highest in the control period. Thus, even though the conditions of these initial experiments were not ideal (higher concentration of ethanol, shorter duration of the application of Tam), their results reinforce those obtained with longer Tam applications and only 3.4 mM ethanol. A clear increase in miniature frequency was also observed in some cases, either during the perfusion with Tam (two of 11 cells) or a few minutes after the end of its addition, during the beginning of the washing period (three of 11 cells). The control frequency was too variable in four of 11 cells, and no significant effect of Tam on frequency was detected in the remaining two cells.
Experiments were also performed in the presence of 100 μM cadmium chloride (CdCl2), to reduce the calcium influx that might be activated by depolarization of presynaptic terminals. These experiments were performed in the presence of only 3.4 mM ethanol. As illustrated in Fig. 2, Tam could still increase both the amplitude and the frequency of glycinergic miniature currents in the presence of cadmium. In the experiment illustrated, the increase in amplitude was already clear 6–8 min after addition of Tam (Fig. 2B, average traces), but it was maximum only after the 14-min Tam application, during the initial washing period. This indicates that Tam slowly penetrated into the slice and was slowly washed out (see Discussion). The effect of Tam on amplitude was almost completely reversible after 24 min washing (Fig. 2C). In the same way, the increase in frequency was already detectable during Tam application (compare the control trace and the trace recorded at 12–14 min after addition of Tam in Fig. 2A) but developed further during the initial washing period. In this motoneuron, the effect on frequency was particularly slow and strong (increase by a factor of 3.4 12–14 min after Tam addition, maximum increase by a factor of 14.6) and was only partly reversed after 24–26 min washing (Fig. 2D). At 2 μM, Tam induced a reliable increase in miniature amplitude in five of six experiments performed in the presence of cadmium, and the mean percentage increase of miniature amplitude was 79 ± 35% (5). Interestingly, in the cell that did not respond to Tam by an increase in mean miniature amplitude, the mean control value was particularly high (125 pA). At 2 μM, Tam also induced a reliable frequency increase in five of six experiments performed in the presence of cadmium. In the remaining cell, the control frequency was very high and too variable (between 3.6 and 6.7) to allow detection of an effect of Tam. The maximum frequency increase was usually observed during Tam application, between 4 and 12 min after the beginning of perfusion with Tam.
Whatever the experimental conditions used, no effect of Tam (2 μM) was ever detected on the time course of decay of glycinergic miniature currents (as already indicated by the normalized average traces of Fig. 1B; comparison performed for each neuron). The average (±SD) of the mean T0.5 was 4.90 ± 0.70 msec in the controls and 4.89 ± 0.91 msec at the maximum Tam effect for the 10 experiments using 3.4 mM ethanol in which Tam increased the amplitude and/or frequency of glycinergic miniatures. The average of the mean T0.5 was 4.96 ± 0.74 msec in the controls and 4.76 ± 0.91 msec at the maximum Tam effect for the five experiments using 17 mM ethanol in which Tam increased the amplitude of glycinergic miniatures.
Another set of experiments was performed (in the presence of 3.4 mM ethanol) to test possible effects of a lower Tam concentration, 0.5 μM. In these experiments, Tam was added for 12–16 min. As illustrated in Fig. 3, even at 0.5 μM, Tam could increase the mean amplitude and the mean frequency of glycinergic miniature currents. However, each of these effects was only observed with four of nine cells tested. The percentage increase of mean miniature amplitude induced by 0.5 μM Tam was 47 ± 14% (4). In three of the cells that did not respond to Tam by an increase in mean miniature amplitude, the control value was particularly high (> 85 pA), as already noted above for experiments using 2 μM Tam and 17 mM ethanol. Again, no effect of Tam was detected on the time course of decay of miniature currents.
In conclusion, bath application of rather low concentrations of Tam (0.5 or 2 μM) can increase the mean amplitude and the mean frequency of glycinergic miniature currents, even in the presence of cadmium ions, within a time scale of a few minutes. Interestingly, the two effects were not always associated: in a few cases, only one of them was detected. Furthermore, the increase in frequency and the increase in amplitude did not always develop (or reverse) simultaneously (see for example Fig. 2).
Tam also increases the frequency of spontaneous glycinergic currents in the absence of TTX
The glycinergic spontaneous activity of hypoglossal motoneurons was also recorded in the absence of TTX, to detect both miniature synaptic events and action-potential driven spontaneous synaptic events. All of these experiments were performed without cadmium, in bicuculline and 3.4 mM ethanol, after blockade of glutamatergic activity either by kynurenic acid (in most experiments) or by CNQX (10 μM) and D-AP-5 (25 μM). The control spontaneous glycinergic activity was variable from one motoneuron to the other, as illustrated by the range of control mean frequency shown in Fig. 4F (horizontal axis). In most experiments (in nine of 11 in kynurenic acid and in two of three in CNQX and D-AP-5), Tam (2 μM) clearly increased the frequency of the total spontaneous glycinergic activity. In two of three remaining cases, the control frequency was not stable enough to reach a conclusion, whereas in the third, a strong increase in frequency was observed but only after Tam application, during the beginning of the washing period. Figure 4F gives the maximum mean frequency in Tam as a function of the mean control frequency for each successful experiment (without including the experiment described in Fig. 4E). One of the experiments of Fig. 4F is described in more detail in Fig. 4, A–D. The current traces recorded in the control period and with Tam are shown at two different time scales (Fig. 4, A and B), the average inter-event interval each minute is plotted (Fig. 4C), and cumulative histograms of spontaneous current amplitudes and inter-event intervals are also illustrated (Fig. 4D). Tam clearly shortened the mean inter-event interval and shifted the cumulative histogram of inter-event intervals to the left, demonstrating a strong increase in frequency. This effect was partly reversible during washing (Fig. 4D). In contrast, Tam only slightly affected the cumulative histogram of amplitudes. In the absence of TTX, Tam did not usually induce a significant increase of the mean amplitude of spontaneous glycinergic currents. Figure 4E illustrates an additional experiment, in which the effect of Tam on frequency was especially strong and rapid, but was mainly transient: the maximum effect would have been off the scale of Fig. 4F (the maximum frequency was higher than 30 at the beginning of the application of Tam). In this case, a strong increase in amplitude was also transiently induced by Tam, but this was unusual and is not shown. The mean frequency of spontaneous glycinergic synaptic currents increased by a factor of 3.08 ± 2.54 (10) when measured at the maximum of the effect of 2 μM Tam (over a period of at least 1 min, without including the experiment in Fig. 1E). In many experiments, the effect of Tam was partly transient, so that the increase of the mean frequency was only by a factor of 2.03 ± 1.22 (10) after more than 10 min Tam.
Are there presynaptic glycine receptors able to affect the frequency of glycinergic synaptic currents
It is known from studies performed on other preparations that glycine receptors can modulate transmitter release by acting presynaptically, in some cases via a direct effect on presynaptic terminals (see Discussion). It is also known that Tam can markedly potentiate the activity of glycine receptors when the agonist concentration is low (4). Thus, if some glycine receptors were expressed presynaptically, their activation by endogenous agonists (glycine and/or taurine) would be favored by Tam and might influence the frequency of glycinergic synaptic events. To test the possible role of presynaptic glycine receptors, the frequency of glycinergic miniatures was compared before and after addition of exogenous glycine. Detection of miniatures can be hampered by the noise increase resulting from activation of postsynaptic glycine receptors (very abundant on hypoglossal motoneurons). Thus, the concentration of glycine to be added had to be as low as possible. In the experiment illustrated in Fig. 5, A–C, addition of glycine (30 μM) immediately induced, as expected, a slight increase in noise between synaptic events (see current traces in Fig. 5B). After a few minutes, glycine also induced a strong increase in the frequency of miniatures: this is shown both by the current traces illustrated (Fig. 5, A and B), and by the cumulative histograms of inter-event intervals (Fig. 5C). Note the shift to the left induced by glycine and the almost complete reversibility after washing. Figure 5D illustrates another similar experiment performed by adding only 10 μM glycine. In four of seven experiments, addition of glycine (10, 20, or 30 μM) thus increased the frequency of glycinergic miniatures, by a factor ranging from 1.6–8.3. The data obtained did not show any reproducible effect on the mean miniature amplitude. The fact that glycine can increase the mean miniature frequency suggests that some glycinergic terminals express presynaptic glycine receptors that facilitate glycine release when activated.
Discussion
In a previous study performed on cultures of dissociated embryonic spinal neurons, Tam was shown to markedly increase the sensitivity of some glycine receptors to low concentrations of exogenous glycine, whereas it never increased responses to saturating glycine concentrations (4). Here, using brainstem slices that preserve native synaptic connections, it is shown that Tam increases the glycinergic spontaneous synaptic activity in motoneurons known to express adult glycine receptors (13). The observed increase in miniature synaptic activity results both from an increase in the mean amplitude of miniature currents and from an increase in their frequency.
The time scale of the development and reversibility of Tam effects on glycinergic synaptic currents was variable from one motoneuron to another. In a few cases, an effect was observed as soon as the Tam-containing solution reached the recording chamber (in less than 30 sec). However, several minutes were usually necessary to observe the maximum effects, and in some cases, the perfusion with Tam was not long enough and the maximum effects were observed during the beginning of the washing period. The reversibility was also rather slow and not always complete. Several factors can explain this variability and relative slowness. Tam must diffuse through the thickness of the slice to reach all the synapses contributing to the recorded current. Furthermore, this lipophilic compound is likely to accumulate progressively in membranes. Note that, even when Tam was applied to isolated cultured neurons by local fast perfusion, some preincubation was necessary for maximum potentiation; furthermore, complete reversibility was not always observed even after 15 min fast perfusion without Tam (4). Bath perfusion of synaptic contacts in 300 μm thick slices is obviously much more difficult than perfusion of an isolated soma via the strong and optimally directed fluid flow used in our previous study.
The increase in glycinergic miniature current amplitude by Tam is consistent with an increase in glycine sensitivity of postsynaptic receptors, similar to that previously observed in spinal neurons. It indicates that postsynaptic receptors are usually not saturated by the release of a single vesicle. The same conclusion had already been drawn from the potentiating effect of low concentrations of Zn2+ ions on glycinergic miniatures of zebrafish larvae neurons and embryonic rat spinal neurons (14, 15). However, these miniatures were also prolonged by Zn2+, whereas no prolongation was observed here upon addition of Tam. Note that Tam and Zn2+ are known to potentiate glycine responses by different mechanisms (4). The observation that Tam did not prolong glycinergic synaptic responses also contrasts with the prolongation observed upon activation of all G proteins in cultured embryonic spinal neurons (16). Note that this difference might involve differences in experimental conditions (glycine receptor isoforms expressed, recording temperature... ) and does not exclude possible activation of some G proteins by Tam. The effects of Tam also differ from those of volatile anesthetics, which did not affect the mean amplitude of miniature currents, but prolonged their duration in rat spinal motoneurons (17). In contrast, the effects of Tam are qualitatively similar to those of ethanol, which also increased both the amplitude and the frequency of miniature glycinergic currents without affecting their duration in juvenile rat hypoglossal motoneurons (18). The protocol used here (continuous application of the same concentration of ethanol during control, Tam and washing) guarantees that the effects observed upon addition of Tam are actually its own, rather than those of its solvent. Nevertheless, the fact that ethanol can already affect glycinergic miniatures at 10 mM (18) may partly explain why modulation by Tam was not systematically observed when 17 mM ethanol was continuously present. Ethanol shifts the dose-response curve of some glycine receptors toward lower glycine concentrations by increasing their affinity for glycine, as shown in hypoglossal motoneurons (19, 20, 21). The fact that Tam did not increase the mean miniature amplitude when it was already high in control, in particular in 17 mM ethanol, agrees with previous results obtained close to saturation of spinal receptors (4). Usually, Tam did not increase the mean amplitude of spontaneous synaptic responses recorded without TTX (including large amplitude events due to action-potential-dependent simultaneous release of several vesicles). This is consistent with the selective potentiation of responses to small amounts of released glycine.
By increasing the amplitude of the smallest synaptic events, Tam could bring more events above the detection threshold. This could contribute to the observed increase in frequency of miniature and spontaneous synaptic currents. However, this is unlikely to be the only explanation. The increase in miniature frequency was sometimes very marked. Furthermore, it was not always correlated with a change in amplitude. In addition, in the absence of TTX, the frequency increase was usually observed without any change in mean amplitude. Thus, Tam probably also acts at the presynaptic level. Possible mechanisms are discussed below.
The increase in frequency of glycinergic synaptic responses induced by Tam might involve an increase in the presynaptic Ca2+ concentration. However, Tam still increased the miniature frequency in the presence of the Ca2+ channel blocker Cd2+. Thus, the observed frequency increase did not require activation (by a putative Tam-induced depolarization) of the presynaptic Cd2+-sensitive voltage-gated Ca2+ channels usually involved in transmitter release. Note also that known effects of Tam on voltage-gated Ca2+ channels are inhibitory (22, 23), and that, at 1 μM, Tam prevented the intracellular Ca2+ increase induced by addition of external K+ in sensory neurons (Ref.24 ; see also25). Nevertheless, in some nonneuronal cells, Tam can increase intracellular Ca2+ (26, 27). Thus, it remains possible that Tam increases the presynaptic Ca2+ concentration, either by activating some Cd2+-insensitive Ca2+ permeable channels or by releasing intracellular Ca2+.
The present study shows that, at least in some cases, glycine itself can increase the frequency of miniature glycinergic synaptic currents. This suggests that Tam could increase the frequency of glycine release simply by potentiating the activity of some presynaptic glycine receptors (provided that an endogenous agonist is present at a sufficient concentration to allow activation of these receptors, at least in the presence of the potentiator). Note that such a scheme could explain why other potentiators of glycine receptors such as ethanol (18) also increase the frequency of glycinergic miniatures (additional mechanisms remain of course possible). An analogous scheme has already been proposed to explain the increase in frequency of GABAA spontaneous and miniature currents induced by allopregnanolone, a neurosteroid potentiating Cl– responses to low concentrations of GABA (Ref.28 ; for nerve terminals from 4-wk-old rats, see also Refs.29 and 30). It was already known that some glycinergic presynaptic receptors can facilitate transmitter release in a few other brain areas (31, 32, 33). However, these effects were no longer observed in TTX (33) or Cd2+ (31). For technical reasons, these studies, as well as the present one, were performed using young animals. It remains to demonstrate that similar effects persist in the adult, at least at some synapses. This is likely because, even in 4- to 6-wk-old animals, Cl– responses to glycine or GABA can be depolarizing in some interneurons, whereas they are hyperpolarizing in the principal cells (see Ref.34 for glycinergic interneurons in the dorsal cochlear nucleus; see also Refs.30 and 35). Note that, at the age used in the present study, postsynaptic glycine responses of hypoglossal motoneurons are already mature and hyperpolarizing (13) [in younger animals, they can be depolarizing but also inhibit the electrical activity (36)]. If the Cl– concentration is higher in the presynaptic terminal than in the postsynaptic motoneuron, the facilitatory effect of glycine on miniature frequency could be explained by a glycine-induced small presynaptic depolarization, for example activating Cd2+-insensitive voltage-gated Ca2+-channels and/or an intracellular Ca2+ release, possibly in the absence of Ca2+ influx (37).
An increase in cAMP might also be involved in presynaptic effects of Tam. Different types of membrane estrogen receptors have been described recently (for review see Ref.38). Interestingly, some of them are coupled to a Gs protein-stimulating adenylate cyclase and can be activated by Tam (39, 40). Furthermore, it is well known that transmitter release can be increased by cAMP, and modulation of glycine release by this pathway has been reported (Refs.41 and 42 ; see also Ref.43 for hypoglossal motoneurons).
The relief of some tonic inhibition of glycine release by some presynaptic metabotropic receptors could also partly explain the increase in glycine release induced by Tam. Indeed, activation of some membrane estrogen receptors has been reported to uncouple some metabotropic receptors from their associated G protein (44, 45, 46). Finally, Tam can inhibit some protein kinases C (PKC) (47). However, this property is unlikely to explain the observed increase in miniature frequency because it is PKC activation (not inhibition) that facilitates neurotransmitter release by increasing the Ca2+-sensitivity of vesicle fusion, a property likely to be general (see Ref.48 and included references). Tam can also enhance the membrane translocation of PKC in intact cells (49). This other type of effect might facilitate glycine release. Note also that contrasting modulations of glycine responses by PKC have been described in a variety of other neurons (12). Thus, a role for PKC in Tam-induced postsynaptic effects on hypoglossal motoneurons remains possible, even though it was excluded in spinal neurons (4).
The increase in frequency of the total spontaneous glycinergic synaptic activity induced by Tam was often partly transient. This could result from a progressive decrease of the number of releasable presynaptic vesicles after the very strong initial release triggered by Tam [note that, in the hippocampus, the number of releasable synaptic vesicles adjacent to the presynaptic membrane of GABAergic synapses, is decreased 24 h after injection of 17-estradiol benzoate (50)]. Possible effects of Tam on other types of ionic channels might also be inhibitory for action-potential evoked glycine release (22, 23, 51) for partial block of some voltage-gated Ca2+ or TTX-sensitive Na+ channels (see also Refs.52 and 53 for possible activation of some K+ channels).
Potentiation of glycinergic inhibition by Tam might be one of the early mechanisms contributing to its neuroprotective effects in various pathological situations (see Refs.10 and 11 and included references and Ref.9). Increasing tonic glycinergic activity, by increasing both the synaptic spontaneous glycinergic activity (present results) and possible nonsynaptic responses to the ambient concentration of glycinergic agonists, might oppose the excitotoxicity due to overactivation of glutamatergic receptors (see Ref.54 for neuroprotection by potentiation of GABA-induced Cl– responses). Note that a strong release of taurine (glycinergic agonist) is induced by various pathological conditions, including ischemia, and largely persists even in animals dialysed with high concentrations of Tam (55, 56). However, overactivation of ligand-gated Cl– channels can also be neurotoxic (see, for example, Ref.57 and included discussion). The neuroprotective effect of Tam also seems to involve a reduction of the release of excitatory amino acids during ischemia (11, 56).
In any case, under normal physiological conditions, the striking effects of Tam on glycinergic neurotransmission and more generally on the glycine receptors present in many brain areas (at some nonglycinergic presynaptic sites and at many nonsynaptic sites) are likely to induce some neurophysiological dysfunction, in particular in motor control. Possible risks related to the effects of Tam on glycinergic synapses and receptors should be taken into account when considering its administration to healthy individuals and patients. For example, Tam might increase the incidence of obstructive sleep apnea. This severe respiratory problem is indeed favored by ethanol, via a modulation of glycinergic neurotransmission in the hypoglossal nucleus (see Ref.21 and included references), which is very similar to the effects described here for Tam.
Acknowledgments
I am very grateful to Dr. Michael Schumacher for having allowed me to introduce patch-clamp stations in Unité 488 and for having encouraged and supported this work from its very beginning. I thank Alain Péchard (Mechanics workshop, Ecole Normale Supérieure, Paris) for his major contribution to the building of the experimental set-up. I also thank Boris Barbour, Mariano Casado, and Stephane Supplisson for their help and advice concerning the slice recording set-up and data processing. I am also grateful to Stephane Dieudonné and Marie-Pierre Morin-Surun for their advice and help in preparing the first brainstem slices. Finally, I thank Krzysztof Rajkowski for helpful comments on the manuscript.
Footnotes
Abbreviations: CNQX, 6-Cyano-7-nitroquinoxaline-2,3-dione disodium salt; D-AP-5, D(-)-2-amino-5-phosphonopentanoic acid; GABA, -aminobutyric acid; PKC, protein kinase C; Tam, tamoxifen; TTX, tetrodotoxin.
References
Schwartz J 2004 Tamoxifen: an emerging preventive. Front Biosci. 9:2827–2847
Sugimoto Y, Tsukahara S, Imai Y, Sugimoto Y, Ueda K, Tsuruo T 2003 Reversal of breast cancer resistance protein-mediated drug resistance by estrogen antagonists and agonists. Mol Cancer Ther. 2:105–112
Top S, Vessières A, Leclercq G, Quivy J, Tang J, Vaissermann J, Huché M, Jaouen G 2003 Synthesis, biochemical properties and molecular modelling studies of organometallic specific estrogen receptor modulators (SERMs), the ferrocifens and hydroxyferrocifens: evidence for an antiproliferative effect of hydroxyferrocifens on both hormone-dependent and hormone-independent breast cancer cell lines. Chem Eur J. 9:5223–5236
Chesnoy-Marchais D 2003 Potentiation of glycine responses by dideoxyforskolin and tamoxifen in rat spinal neurons. Eur J Neurosci. 17:681–691
Lien EA, Wester K, Lonning PE, Solheim E, Ueland PM 1991 Distribution of tamoxifen and metabolites into brain tissue and brain metastases in breast cancer patients. Br J Cancer. 63:641–645
Trump DL, Smith DC, Ellis PG, Rogers MP, Schold SC, Winer EP, Panella TJ, Jordan VC, Fine RL 1992 High-dose oral tamoxifen, a potential multidrug-resistance-reversal agent: phase I trial in combination with vinblastine. J Natl Cancer Inst. 84:1811–1816
Eberling JL, Wu C, Tong-Turnbeaugh R, Jagust WJ 2004 Estrogen- and tamoxifen-associated effects on brain structure and function. Neuroimage 21:364–371
Shy H, Malaiyandi L, Timiras PS 2000 Protective action of 17-estradiol and tamoxifen on glutamate toxicity in glial cells. Int J Dev Neurosci. 18:289–297
O’Neill K, Chen S, Diaz Brinton R 2004 Impact of the selective estrogen receptor modulator, tamoxifen, on neuronal outgrowth and survival following toxic insults associated with aging and Alzheimer’s disease. Exp Neurol. 188:268–278
Ciriza I, Carrero P, Azcoitia I, Lundeen SG, Garcia-Segura LM 2004 Selective estrogen receptor modulators protect hippocampal neurons from kainic acid excitotoxicity: differences with the effect of estradiol. J Neurobiol. 61:209–221
Kimelberg HK, Jin Y, Charniga C, Feustel PJ 2003 Neuroprotective activity of tamoxifen in permanent focal ischemia. J Neurosurg. 99:138–142
Lynch JW 2004 Molecular structure and function of the glycine receptor chloride channel. Physiol Rev. 84:1051–1095
Singer JH, Talley EM, Bayliss DA, Berger AJ 1998 Development of glycinergic synaptic transmission to rat brain stem motoneurons. J Neurophysiol. 80:2608–2620
Suwa H, Saint-Amant L, Triller A, Drapeau P, Legendre P 2001 High-affinity zinc potentiation of inhibitory postsynaptic glycinergic currents in the zebrafish hindbrain. J Neurophysiol. 85:912–925
Laube B 2002 Potentiation of inhibitory glycinergic neurotransmission by Zn2+: a synergistic interplay between presynaptic P2X2 and postsynaptic glycine receptors. Eur J Neurosci. 16:1025–1036
Yevenes GE, Peoples RW, Tapia JC, Parodi J, Soto X, Olate J, Aguayo LG 2003 Modulation of glycine-activated ion channel function by G-protein subunits. Nat Neurosci. 6:819–824
Cheng G, Kendig JJ 2002 Pre- and postsynaptic volatile anaesthetic actions on glycinergic transmission to spinal cord motor neurons. Br J Pharmacol. 136:673–684
Eggers ED, O’Brien JA, Berger AJ 2000 Developmental changes in the modulation of synaptic glycine receptors by ethanol. J Neurophysiol. 84:2409–2416
Aguayo LG, Tapia JC, Pancetti FC 1996 Potentiation of the glycine-activated Cl– current by ethanol in cultured mouse spinal neurons. J Pharmacol Exp Ther. 279:1116–1122
Ye JH, Tao L, Ren J, Schaefer R, Krnjevic K, Liu PL, Schiller DA, McArdle JJ 2001 Ethanol potentiation of glycine-induced responses in dissociated neurons of rat ventral tegmental area. J Pharmacol Exp Ther. 296:77–83
Eggers ED, Berger AJ 2004 Mechanisms for the modulation of native glycine receptor channels by ethanol. J Neurophysiol. 91:2685–2695
Sartor P, Vacher P, Mollard P, Dufy B 1988 Tamoxifen reduces calcium currents in a clonal pituitary cell line. Endocrinology 123:534–540
Song J, Standley PR, Zhang F, Joshi D, Gappy S, Sowers JR, Ram JL 1996 Tamoxifen (estrogen antagonist) inhibits voltage-gated calcium current and contractility in vascular smooth muscle from rats. J Pharmacol Exp Ther. 277:1444–1453
Chaban VV, Mayer EA, Ennes HS, Micevych PE 2003 Estradiol inhibits ATP-induced intracellular calcium concentration increase in dorsal root ganglia neurons. Neuroscience 118:941–948
Morley P, Whitfield JF 1994 Effect of tamoxifen on carbachol-triggered intracellular calcium responses in chicken granulosa cells. Cancer Res. 54:69–74
Somjen D, Kohen F, Lieberherr M 1997 Nongenomic effects of an anti-idiotypic antibody as an estrogen mimetic in female human and rat osteoblasts. J Cell Biochem. 65:53–66
Chang H-T, Huang J-K, Wang J-L, Cheng J-S, Lee K-C, Lo Y-K, Liu C-P, Chou K-J, Chen W-C, Su W, Law Y-P, Jan C-R 2002 Tamoxifen-induced increases in cytoplasmic free Ca2+ levels in human breast cancer cells. Breast Cancer Res Treat. 71:125–131
Uchida S, Noda E, Kakazu Y, Mizoguchi Y, Akaike N, Nabekura J 2002 Allopregnanolone enhancement of GABAergic transmission in rat medial preoptic area neurons. Am J Physiol Endocrinol Metab. 283:E1257–E1265
Haage D, Johansson S 1999 Neurosteroid modulation of synaptic and GABA-evoked currents in neurons from the rat medial preoptic nucleus. J Neurophysiol. 82:143–151
Haage D, Druzin M, Johansson S 2002 Allopregnanolone modulates spontaneous GABA release via presynaptic Cl– permeability in rat preoptic nerve terminals. Brain Res. 958:405–413
Turecek R, Trussell LO 2001 Presynaptic glycine receptors enhance transmitter release at a mammalian central synapse. Nature 411:587–590
Kawa K 2003 Glycine facilitates transmitter release at developing synapses: a patch clamp study from Purkinje neurons of the newborn rat. Brain Res Dev Brain Res. 144:57–71
Jeong H-J, Jang I-S, Moorhouse AJ, Akaike N 2003 Activation of presynaptic glycine receptors facilitates glycine release from presynaptic terminals synapsing onto rat spinal sacral dorsal commissural nucleus neurons. J Physiol. 550:373–383
Golding NL, Oertel D 1996 Context-dependent synaptic action of glycinergic and GABAergic inputs in the dorsal cochlear nucleus. J Neurosci. 16:2208–2219
Chavas J, Marty A 2003 Coexistence of excitatory and inhibitory GABA synapses in the cerebellar interneuron network. J Neurosci. 23:2019–2031
Marchetti C, Pagnotta S, Donato R, Nistri A 2002 Inhibition of spinal or hypoglossal motoneurons of the newborn rat by glycine or GABA. Eur J Neurosci. 15:975–983
De Crescenzo V, ZhuGe R, Velazquez-Marrero C, Lifshitz LM, Custer E, Carmichael J, Lai FA, Tuft RA, Fogarty KE, Lemos JR, Walsh Jr JV 2004 Ca2+ syntillas, miniature Ca2+ release events in terminals of hypothalamic neurons, are increased in frequency by depolarization in the absence of Ca2+ influx. J Neurosci. 24:1226–1235
Toran-Allerand CD 2004 Minireview: a plethora of estrogen receptors in the brain: where will it end Endocrinology 145:1069–1074
Razandi M, Pedram A, Greene GL, Levin ER 1999 Cell membrane and nuclear estrogen receptors (ERs) originate from a single transcript: studies of ER and ER expressed in Chinese hamster ovary cells. Mol Endocrinol. 13:307–319
Thomas P, Pang Y, Filardo EJ, Dong J 2005 Identity of an estrogen membrane receptor coupled to a G protein in human breast cancer cells. Endocrinology. 146:624–632
Katsurabayashi S, Kubota H, Wang ZM, Rhee JS, Akaike N 2001 cAMP-dependent presynaptic regulation of spontaneous glycinergic IPSCs in mechanically dissociated rat spinal cord neurons. J Neurophysiol. 85:332–340
Katsurabayashi S, Kubota H, Moorhouse AJ, Akaike N 2004 Differential modulation of evoked and spontaneous glycine release from rat spinal cord glycinergic terminals by the cyclic AMP/protein kinase A transduction cascade. J Neurochem. 91:657–666
Umemiya M, Berger AJ 1995 Presynaptic inhibition by serotonin of glycinergic inhibitory synaptic currents in the rat brain stem. J Neurophysiol. 73:1192–1201
Kelly MJ, Qiu J, Wagner EJ, Ronnekleiv OK 2002 Rapid effects of estrogen on G protein-coupled receptor activation of potassium channels in the central nervous system (CNS). J Steroid Biochem Mol Biol. 83:187–193
Qiu J, Bosch MA, Tobias SC, Grandy DK, Scanlan TS, Ronnekleiv OK, Kelly MJ 2003 Rapid signaling of estrogen in hypothalamic neurons involves a novel G-protein-coupled estrogen receptor that activates protein kinase C. J Neurosci. 23:9529–9540
Mize AL, Young LJ, Alper RH 2003 Uncoupling of 5-HT1A receptors in the brain by estrogens: regional variations in antagonism by ICI 182,780. Neuropharmacology 44:584–591
O’Brian CA, Liskamp RM, Solomon DH, Weinstein IB 1985 Inhibition of protein kinase C by tamoxifen. Cancer Res. 45:2462–2465
Lou X, Scheuss V, Schneggenburger R 2005 Allosteric modulation of the presynaptic Ca2+ sensor for vesicle fusion. Nature 435:497–501
Gundimeda U, Chen ZH, Gopalakrishna R 1996 Tamoxifen modulates protein kinase C via oxidative stress in estrogen receptor-negative breast cancer cells. J Biol Chem. 271:13504–13514
Ledoux VA, Woolley CS 2005 Evidence that disinhibition is associated with a decrease in number of vesicles available for release at inhibitory synapses. J Neurosci. 25:971–976
Hardy SP, deFelipe C, Valverde MA 1998 Inhibition of voltage-gated cationic channels in rat embryonic hypothalamic neurones and C1300 neuroblastoma cells by triphenylethylene antioestrogens. FEBS Lett. 434:236–240
Dick GM, Rossow CF, Smirnov S, Horowitz B, Sanders KM 2001 Tamoxifen activates smooth muscle BK channels through the regulatory 1 subunit. J Biol Chem. 276:34594–34599
Dick GM, Sanders KM 2001 (Xeno)estrogen sensitivity of smooth muscle BK channels conferred by the regulatory 1 subunit: a study of 1 knockout mice. J Biol Chem. 276:44835–44840
Green AR, Hainsworth AH, Jackson DM 2000 GABA potentiation: a logical pharmacological approach for the treatment of acute ischaemic stroke. Neuropharmacology 39:1483–1494
Saransaari P, Oja SS 2000 Taurine and neural cell damage. Amino Acids 19:509–526
Feustel PJ, Jin Y, Kimelberg HK 2004 Volume-regulated anion channels are the predominant contributors to release of excitatory amino acids in the ischemic cortical penumbra. Stroke 35:1164–1168
Allen NJ, Rossi DJ, Attwell D 2004 Sequential release of GABA by exocytosis and reversed uptake leads to neuronal swelling in simulated ischemia of hippocampal slices. J Neurosci. 24:3837–3849(D. Chesnoy-Marchais)