| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Laboratoire de Neurophysiologie Cellulaire et Intégrée, UMR 7519 CNRS/ULP, 21 rue R. Descartes, 67084 Strasbourg Cedex, France
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
(Received 25 May 2004;
accepted after revision 22 June 2004;
first published online 2 July 2004)
Corresponding author J. Trouslard: Laboratoire de Neurophysiologie Cellulaire et Intégrée, UMR 7519 CNRS/ULP, 21 rue R. Descartes, 67084 Strasbourg Cedex, France. Email: trouslard{at}neurochem.u-strasbg.fr
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Insitu hybridization studies have shown that GlyT2 is predominantly expressed in spinal cord, brainstem and cerebellum where it is associated with strychnine-sensitive glycinergic receptors and thus is considered a reliable marker of glycinergic neurones (Adams et al. 1995; Luque et al. 1995; Zafra et al. 1995). GlyT2 does not appear to play an important role in clearing glycine at glycinergic synapses since its inhibition did not affect significantly the decay kinetics of glycinergic IPSCs in several studies (Titmus et al. 1996; Singer et al. 1998; Oku et al. 1999). GlyT2 is found in glycinergic boutons extrasynaptically outside the active zones (Spike et al. 1997), a location that may explain this lack of effect. Recently, hypoglossal motoneurones from GlyT2-deficient mice were shown to display glycinergic miniature IPSCs with a marked reduction in amplitude as compared to those recorded from wild-type mice, suggesting an essential role of GlyT2 for taking up glycine into glycinergic terminals (Gomeza et al. 2003b).
GlyT1 is distributed more widely in the CNS (as it is located mostly on glial processes), is not restricted to glycinergic terminals (Zafra et al. 1997), and is found in areas devoid of strychnine-sensitive receptors (Smith et al. 1992; Zafra et al. 1995). An essential role of GlyT1 is to lower extracellular glycine concentration at glycinergic synapses. This was clearly shown in GlyT1-deficient mice, where the decay time constant of glycinergic mIPSCs recorded in hypoglossal motoneurones was longer than that in wild-type mice (Gomeza et al. 2003a). An additional role of GlyT1 could be to modulate NMDA-mediated glutamatergic neurotransmission. This could be achieved at a postsynaptic level by controlling the level of glycine that the NMDA glutamate receptors are exposed to (Supplisson & Bergman, 1997; Bergeron et al. 1998), as well as at a presynaptic level where presynaptic glycine receptors have been shown to enhance glutamate release (Turecek & Trussell, 2001).
We explored the role of GlyT1 and GlyT2 in lamina X of the spinal cord, where we have demonstrated the existence of functional glycinergic (Bradaïa & Trouslard, 2002b) and glutamatergic synaptic transmission (Bordey et al. 1996). The neurones receiving these synaptic inputs are localized around the central canal (Bordey et al. 1996; Bradaïa & Trouslard, 2002a,b) and participate in the control of sympathetic outflow (Hosoya et al. 1994) and visceronociception (Nahin et al. 1983). In addition, lamina X displays one of the highest densities of GlyT1 and GlyT2 in the spinal cord (Zafra et al. 1995). We used the pharmacological agents ORG 24598 and ORG 25543 in order to block glial and neuronal GlyT, respectively, and examined the consequences of these blockades on glycinergic IPCSs and glutamatergic EPSCs. ORG 24598 is a highly selective inhibitor of GlyT1 (pIC50= 6.9) with negligible activity at GlyT2 (pIC50 < 3) or GABA transporter (pIC50 < 4) (Brown et al. 2001). In addition, binding experiments using [3H]strychnine and [3H]MDL 105 519 in the rat spinal cord and brain membranes revealed that ORG 24598 had negligible affinity (pKi < 4) for the strychnine-sensitive glycine receptor and for the glycine site of the NMDA receptor (Brown et al. 2001). The interaction of ORG 24598 with the glutamate site of the NMDA receptor has not been investigated. ORG 25543 is a highly selective inhibitor of GlyT2 (pIC50= 7.8) with negligible activity at GlyT2 (pIC50 < 4) and negligible affinity for the strychnine-sensitive glycine receptor (Caulfield et al. 2001).
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Spinal cord slices were prepared as previously described (Bradaïa & Trouslard, 2002a,b). Briefly (P12–P16) Wistar rats (Depré SARL, France) were decapitated under deep ether anaesthesia. All experiments were conducted in conformity with the rules set by the EC Council Directive (86/89/EEC) and the French Department of Agriculture (licence no. 67-182 to J. T.) The thoracolumbar section of the spinal cord was removed and placed in a high-osmolarity, calcium-free, sucrose-based solution bubbled with 95% O2–5% CO2 and maintained at 3°C. This solution contained (mM): sucrose (248); NaH2PO4 (1.25); KCl (2); NaHCO3 (26); glucose (11.1); NaCl (100); MgSO4 (2), kynurenic acid (1); pH 7.3. Transverse slices with a thickness of 300 µm were prepared using a vibratome (Pelco). Slices were transferred to a storage chamber containing an artificial cerebrospinal fluid (aCSF) containing (mM): NaCl (126); NaH2PO4 (1.25); KCl (2.5); CaCl2 (2); MgCl2 (2); glucose (11.1); NaHCO3 (26); kynurenic acid (2); pH 7.3. After recovery of at least 1 h at room temperature (20°C), the slice was transferred to the recording chamber.
Recordings
The slices were placed on an upright microscope (Zeiss Axioscop) mounted on a Gibraltar X–Y table. Slices were observed through a 40 x water immersion objective using a infrared-sensitive camera (Till Photonics). Cells were chosen visually in the laterodorsal part of lamina X (Bradaïa & Trouslard, 2002a,b).
Patch clamp recordings were performed in the whole cell configuration. Patch pipettes were filled with a standard internal solution containing (mM): KCl (130); NaCl (10); Hepes (10); MgCl2 (1.2); EGTA 5; CaCl2 (2.5); biocytin (0.01); pH 7.2. Pipettes had resistances of 2.5–4 M
. Voltage clamp was achieved using a Axopatch 200B amplifier (Axon Instruments, USA). Currents were stored on a DAT recorder at 10 kHz (Biologic DTR-1201) and digitized using a data acquisition board (Digidata 1200) operated by pCLAMP 6.0 software (Axon Instruments).
Focal bipolar stimulation was achieved using an electrode pulled from a theta tube and filled with the external solution. The electrode had a low resistance (500 k to 1 M) and was positioned in any direction 100 µm away from the cell under recording. Stimuli (100 µs, 20–50 V) were generated by a custom-made isolated stimulation unit.
Solutions
The slice was continuously superfused at a rate of 3–4 ml min–1 with oxygenated aCSF at 35°C. The temperature of the recording chamber was continuously monitored and controlled by a custom-made temperature controller. To study glycinergic neurotransmission, kynurenic acid (2 mM), CNQX (10 µM) and bicuculline (10 µM) were added to the bath in order to suppress the glutamatergic and the GABAergic transmission, respectively. When miniature glycinergic IPSCs were recorded, TTX was added at a final concentration of 0.5 µM. When miniature GABAergic IPSCs were recorded, aCSF was supplemented with kynurenic acid (2 mM), CNQX (10 µM), strychnine (1 µM) and TTX (0.5 µM). To record electrically evoked glutamatergic EPSCs, strychnine (1 µM) and bicuculline (10 µM) were added to a Mg-free aCSF. This Mg-free aCSF was obtained by omitting MgCl2 (2 mM) from the external solution without compensation for the loss of osmolarity or of the amount of divalent ions. The bath solution was also supplemented with either 10 µM CNQX or 10 µM APV to isolate NMDA- or non-NMDA-mediated EPSCs, respectively. ORG 24598 and ORG 25543 were bath applied. There was a typical delay of 3–4 min between the start of application and the arrival of the drug on the recorded cell. This minimal delay was estimated in some cells by monitoring the time necessary for bath applied TTX to affect voltage gated sodium current.
Data analysis
The detection and analysis of miniature and spontaneous synaptic currents were achieved by MiniAnalysis software (version 5.6.3, Synaptosoft, USA) (see details in Bradaïa & Trouslard, 2002b). Analysis of electrically evoked currents was made using Clampfit software. Complex events, i.e. multiple peak events, were discarded. The majority of deactivation phases of evoked, spontaneous and miniature IPSCs decay phases were best fitted by a single exponential function: y=Aexp(–t/
) + baseline. The quality of the fit was determined by visual inspection. In a small minority of mIPSCs (1%), the decay kinetics were multiphasic and were not included in the analysis. Significant differences between two distributions of mIPSC amplitudes were determined using the Kolmogorov-Smirnov test with a P value < 0.01 indicating significance.
Data were compared statistically with Student's paired t test and significance was assessed at P < 0.05. Results are expressed as means ±S.E.M.
Drugs
Strychnine, 5,7-dichlorokynurenic acid (DCKA), D-serine and L-serine were obtained from Sigma. Kynurenic acid, bicuculline, 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) and D-2-amino-5-phosphonovalerate (APV) were from Tocris. Tetrodotoxin (TTX) was from Latoxan, France. The drugs ORG 24598 and ORG 25543 were a generous gift from Organon Laboratories Ltd, UK. The structure of ORG 24598 is given in Brown et al. (2001). The structure of ORG 25543 (4-benzyloxy-3,5-dimethoxy-N-[(1-dimethylaminocyclopentyl) methyl]benzamide) is given in Caulfield et al. (2001) and corresponds to compound no. 16 found in that publication.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
General effects of the GlyT blockers. Cells were recorded in the presence of bicuculline, CNQX and kynurenic acid (Jahr & Jessell, 1985) in order to record glycinergic synaptic current in isolation. Initially, we tested ORG 24598 and ORG 25543 at a concentration of 10 µM and at room temperature (20°C). At a holding potential of –60 mV, ORG 24598 and ORG 25543 induced a slow developing inward current of a few picoamps (–20 to –50 pA) and an increase in membrane current noise. These effects were detected after 20–30 min of application and were in general maximal after 40–60 min of application. Increasing the temperature of the bath from 20°C to 35°C reduced considerably the time necessary to observe the onset of the effect and the magnitude of the maximal effects of GlyT blockers. At 35°C, a clear effect of both compounds on membrane noise and current baseline level was typically observed after 5 min of bath application and was maximal after 7–10 min of application. Since increasing the bath temperature greatly facilitated the observation of the effect of GlyT blockers, all the subsequent experiments were performed at 35°C.
At 35°C and a holding potential of –60 mV, ORG 24598 induced an inward current and increased the membrane noise level in all the cells tested (Fig. 1A). Similar effects were observed with ORG 25543 but only in 50% of the cells tested (Fig. 2A). The amplitude of the inward current induced by either of the blockers varied and was between –20 and –50 pA. Due to the increase in membrane noise and the presence of synaptic events, it was difficult to determine accurately the amplitude of the inward current induced by ORG compounds. The inward and fast deflections of the current trace represent glycinergic synaptic miniature IPSCs because the recordings were performed in the presence of TTX. These mIPSCs were blocked by 1 µM strychnine (see Figs 1A and 2A). In addition to the suppression of mIPSCs, strychnine also suppressed the inward current induced by ORG compounds (n= 3) as well as the associated increase in membrane current noise (Figs 1A and 2A).
|
|
20 min in the absence of ORG compounds. One of the blockers was then applied and we routinely waited for 10 min until the baseline current was stabilized. Events were then collected during 20 min. Decay time constant, frequency and amplitude distribution of synaptic events were determined and compared to those obtained under control conditions. Effect of ORG 24598, a GlyT 1 blocker, on glycinergic transmission. ORG 24598 increased significantly (P < 0.05) the decay time constant of mIPSCs from a control value of 4.5 ± 1.4 ms to 10.5 ± 3.2 ms (n= 7 cells) (Fig. 1C). ORG 24598 did not affect significantly (P > 0.05) the amplitude distribution (Fig. 1B) or the frequency of mIPSCs (Fig. 1D). ORG 24598 was also tested on the basic properties of glycinergic sIPSCs and eIPSCs recorded in the absence of TTX. The results are summarized in Fig. 3 and indicate that ORG 24598 also increased the decay time constant of eIPSCs (Fig. 3A, upper traces) and sIPSCs (not shown) without affecting significantly their mean amplitudes or their mean frequencies in the case of sIPSCs (P > 0.05).
|
–54 ± 17 pA (n= 3 cells) (Fig. 4B). Interestingly, this reduction in the amplitude of eIPSCs over a long period of time was not observed with the GlyT1 blocker ORG 24598. Indeed, in the same recording conditions, the mean amplitude of eIPSCs recorded in three other cells was –132 ± 27 pA under control conditions and was –125 ± 7 pA after 50 min of application of ORG 24598 (Fig. 4A).
|
In a previous study we reported that glycinergic mIPSCs and eIPSCs recorded from young rat (P2–P7) at room temperature were described by a monoexponential function with decay time constants of 9–10 ms and 11–13 ms, respectively (Bradaïa & Trouslard, 2002b). In the present study, mIPSCs and eIPSCs decayed also monoexponentially but with a faster time constant of 5 ms. There are two main differences between these studies: temperature of the bath solution and the age of the animal used. It has been reported that glycinergic receptors are subject to developmental maturation which results in a decrease in the time constant of the mIPSCs (Takahashi et al. 1992; Krupp et al. 1994; Gao et al. 1998). Although such a phenomenon cannot be excluded in our preparation, it must be emphasized that the difference in bath temperature (20°C versus 35°C) can totally account for the observed differences in IPSC decay kinetics. Indeed at the recording temperature of 20°C the decay time constant of mIPSCs recorded was 9.7 ± 4.5 ms (n= 3 cells), a value close to those found in younger rats. Increasing temperature from 20°C to 35°C decreased the decay time constant of mIPSCs to 5.0 ± 1.0 ms. Similarly, in the absence of TTX, the decay time constant of sIPSCs decreased from 9.7 ± 3.4 to 5.2 ± 0.4 ms (n= 3 cells) and that of eIPSCs decreased from 10.6 ± 1.3 to 5.4 ± 0.4 ms
(n= 7 cells) when the temperature was raised from 20°C to 35°C (Fig. 5). At a temperature of 35°C, bath application of ORG 25543 decreased progressively the decay time constant to a mean value of 13.0 ± 2.0 ms (n= 7 cells) thereby reproducing the effect of temperature on the decay kinetics of eIPSCs.
|
Since the activity of GlyT depends on extracellular Na+, we have also blocked the activity of the GlyT by replacing extracellular Na+ with NMDG. The NMDG-based solution was only applied for 5 min because longer application caused very large fluctuations in the holding current with a deterioration of the seal and a final loss of the cell. After 5 min of application, we recorded IPSCs and compared their properties with those of IPSCs under control conditions recorded prior to the application of NMDG-based solution. As shown in Fig. 6, NMDG treatment increased the decay time constant of sIPSCs and mIPSCs but did not significantly affect their amplitudes or frequency of occurrence.
|
We hypothesized that the strychnine-sensitive slow inward current and the increase in membrane current noise induced by GlyT1 and GlyT2 blockers could be due to the accmulation of glycine in the extracellular space and subsequent tonic activation of strychnine-sensitive glycine receptors. Then increasing the extracellular concentration of glycine by any other means should, at last in part, mimic the effects of the ORG compounds. We have taken the simplest approach to this question by bath applying a steady-state increasing concentration of glycine (1–5 µM) and analysing the consequences of this situation for the properties of mIPSCs, eIPSCs and sIPSCs. Application of a low concentration of glycine (1–3 µM) induced an increase in membrane noise and a slow inward current of similar amplitude to those produced by ORG compounds (not shown). At the higher concentration of 5 µM, glycine induced increases in current noise and postsynaptic currents which had amplitudes of the order of –100 pA, making the analysis of sIPSCs unreliable. The similarity of the effects induced by superfusion of a steady concentration of glycine and that produced by GlyT blockade indicated that the blockade of GlyT probably increased extracellular glycine concentration in the micromolar range (less than 5 µM). As shown in Fig. 7, bath application of glycine induced, in a concentration-dependent fashion, an increase in the values of the decay time constants of eIPSCs, sIPSCs and mIPSCs. Although a tendency to increase the decay time constants was detected with glycine of 1 or 3 µM, a statistically significant difference (P < 0.05) was observed only at a concentration of 5 µM.
|
In order to assess the specificity of ORG compounds with respect to glycine transmission, we have tested the effects of GlyT blockers on the properties of GABAergic mIPSCs (Figs 8 and 9). In six cells in which ORG 24598 was tested, the decay time of mIPSCs under control conditions was monoexponential with a mean time constant of 17.6 ± 0.9 ms and the mean frequency of GABAergic mIPSCs was 0.148 ± 0.07 Hz. These values were unaffected by the superfusion of 10 µM of ORG 24598. In the presence of ORG 24598 (Fig. 8), the mean decay time of mIPSCs was 17.6 ± 0.8 ms and the frequency was 0.16 ± 0.08 Hz. The cmulative distribution of the amplitudes of mIPSCs was not affected by ORG 24598 (10 µM). Taken together, these results indicate that ORG 24598 does not modify GABAergic mIPSCs.
|
|
From this set of experiments, we conclude that the effects of GlyT blockers were specific as the GABAergic transmission was not affected by GlyT blockers.
Effect of GlyT blockers on glutamatergic transmission
Lamina X neurones express functional NMDA receptors (Bordey et al. 1996). We hypothesized that the blockade of GlyT by ORG compounds may increase extracellular glycine and may subsequently enhance the NMDA response. This may indeed occur if the glycine site is not saturated under the normal recording conditions. To evaluate the degree of saturation of the glycine site at NMDA receptors we isolated the NMDA component of eEPSCs and tested increasing concentrations of exogenously applied glycine. All subsequent experiments were thus performed in Mg-free aCSF and in presence of bicuculline, strychnine and CNQX. Under these conditions, spontaneous or miniature synaptic events were rarely observed (1 event every 3–5 min) and no analysis of these events could therefore be performed. By contrast, electrically evoked EPSCs were recorded in 70% of the cells from which we recorded. At –60 mV, electrically evoked EPSCs had a mean peak amplitude of –48.2 ± 5.6 pA (n= 45 cells), a time to peak of 6 ms, and a monoexponential decay phase with a mean time constant of 37.5 ± 2.1 ms (n= 45). These EPSCs were identified as being mediated by NMDA receptors because (a) eEPSCs were blocked by 30 µM APV (96.3 ± 2.7% inhibition, n= 11 cells), an antagonist of NMDA receptor; (b) the I–V relationship showed the typical Mg-dependent rectification at negative potential when Mg was present in the bath (not shown); (c) glycine (1–5 µM) increased the amplitude of eEPSCs (Fig. 10A–B), and D-Serine (10 µM) increased the amplitude of eEPSCs by a factor of 1.6 ± 0.28 (n= 11) whereas L-serine (10 µM) had no effect on the amplitude of eEPSCs (n= 3) (Fig. 10D); and (d)5, 7-dichlorokynurenic acid (DCKA), a competitive antagonist of glycine site at the NMDA receptor, was found to abolish both eEPSCs (n= 6) under control conditions as well as during bath applications of glycine (n= 3, Fig. 10A) or D-serine (n= 11, not shown).
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
ORG compounds induced an inward current of a few picoamps in amplitude and an increase in membrane noise. It is unlikely that these blockers had a direct action on the recorded cell because GlyT2 is located on glycinergic terminals and GlyT1 is located almost exclusively on glial cells. There are several possibilities which would explain the induction of an inward membrane current and an increase in current noise. First, a blocker which serves also as a substrate may cause the release of either the transported neurotransmitter, e.g. glycine via a mechanism of heteroexchange (Volterra et al. 1996), or a different neurotransmitter, such as GABA or glutamate, present in the terminals of the neurones on which the glycine transporter is located (Raiteri et al. 2001, 2003). The released glycine, GABA or glutamate might in turn activate postsynaptic receptors present on the recorded cells. These two possibilities are unlikely because ORG compounds used in this study are probably not substrates for the glycine transporters. This was clearly shown for ORG 24598 (Walker et al. 1999) and ORG 26176, another GlyT2 blocker with a structure similar to that of ORG 245543. ORG 26176 corresponds to compound no. 1 in Caulfield et al. 2001) (H. Sundaram, personal communication). As expected for non-substrate blockers, application of ORG 24598 and ORG 26176 on oocytes expressing GlyT1 and Glyt2 did not induce membrane currents (S. Supplisson, personal communication; see also Fig. 4 in Roux & Supplisson, 2000). Similarly, no inward current induced by ORG 24598 and ORG 245543 could be recorded in our preparation in the presence of strychnine and in the absence of bicuculline, a situation which would have allowed either the detection of a current associated with the transport of GlyT blockers if these compounds were to be transported or the action of GABA on GABAA receptors. A second possibility is that the inward current and the membrane noise induced by ORG compounds were due to the tonic activation of GlyR by glycine, whose concentration in the external medium was increased in response to the blockade of GlyT. This simple hypothesis is consolidated by the fact that both inward current and the noise were blocked by strychnine. Such a strychnine-sensitive inward current and increase in noise were also recorded in brainstem motoneurones from GLYT1 knock-out mice and have been attributed to extracellular accmulation of glycine which activated tonically glycine receptors (Gomeza et al. 2003a).
The blockade of GlyT1 induced a response in all cells tested whilst GlyT2 blockade induced a response in only 50% of the cells. This observation suggests a fundamental role of GlyT1 in the control of extracellular glycine levels in both glycinergic and glutamatergic synapses. The situation seems slightly more complicated for GlyT2 as a heterogeneity in the distribution of GlyT2 has been reported. First, some glycinergic neurones may express a large number of GlyT2 immunoreactive afferents whereas other neurones do not. Second, GlyT2 is developmentally regulated. It seems to be present in the growth cone prior to the formation of synapses; then labelling changes from a diffuse pattern to a punctuate appearance after the establishment of functional glycinergic synapses where GlyT2 are absent from the active zone (Poyatos et al. 1997; Spike et al. 1997; Friauf et al. 1999). Due to this extrasynaptic distribution in mature synapses, we hypothesize that GlyT2 may play a less important role in the clearing of glycine from the synaptic cleft and a lack of effect of GLT2 blocker is expected in these cells. A more detailed study of the sensitivity to GlyT2 blockers during development is needed to confirm this supposition but, arguing in favour of this hypothesis, we have noticed that, in younger animals (day 5–7), where immature synapses are expected to be more abundant than in this study, all cells tested were sensitive to GlyT2 blockers (results not shown).
Role of neuronal and glial glycine transporters in the properties of glycinergic IPSCs
There is a consensus on the fact that the decay time constant of glycinergic IPSCs is mainly governed by the kinetic properties of glycine receptors and, in particular, by their mean open time, because glycine diffuses so rapidly out of the synaptic cleft that it cannot rebind to the receptor after it has dissociated from it (Clements, 1996; Titmus et al. 1996). According to this schema, the decay time constants are not affected by the blockade of GlyT in several preparations (Singer et al. 1998; Oku et al. 1999). Our results show a clear impact of GlyT blockers or of lowering external Na+ concentration on IPSC decay time constants. The exact mechanism by which the blockade of GlyT affected the kinetics of IPSCs is unknown, and no information on the single-channel properties of glycine receptors in lamina X neurones is yet available. It is, however, known that, in the case of GABAergic synaptic transmission, the blockade of GABA transporters leads to an enhancement of the decay time constant of GABAA-mediated IPSCs. This enhancement might be due to several phenomena including reassociation of GABA to its receptors after unbinding, desensitization of
GABAA receptors (Jones & Westbrook, 1996), or spillover to extrasynaptic GABAA receptors with different kinetic properties (Chery & de Koninck, 1999). It is tempting to speculate that similar mechanisms are involved in the case of glycinergic transmission in lamina X but it is unlikely that a spillover of glycine onto extrasynaptic receptors might have occurred. Indeed, in this case, we would expect that GlyT1 and GlyT2 blockers would induce the appearance of additional synaptic events with a different decay time constant and we would have noticed the appearance of synaptic events with a biexponential decay time course. This was, however, not the case.
The main difference between the two compounds used to block GlyTs was the inhibitory effect of neuronal GlyT on the frequency of mIPSCs and sIPSCs. Mice deficient in GlyT2 exhibit also a reduction in the frequency in mIPSCs as compared to wild-type mice whereas no difference in the frequency was noted in GlytT1-deficient mice (Gomeza et al. 2003a,b). This indicates that the neuronal GlyT2 blocker interferes with the presynaptic compartment. GlyT2 interacts with syntaxin1A (Geerlings et al. 2001), a protein involved in the exocytosis process, and might affect that process in some way. No effect of GlyT2 blocker on the frequency of GABAergic mIPSCs or on the amplitude of non-NMDA-mediated eEPSCs was observed. This suggests that the GlyT2 blocker did not interfere by itself with the exocytosis process. A more likely explanation of the effect of GlyT2 blocker on the frequency of mIPSCs was that blocking the neuronal
GlyT affected the cytosolic glycine concentration in the terminal thus leading to a progressive suppression of the refilling of synaptic vesicles with glycine. As synaptic events generated at terminals sensitive to the GlyT2 blocker became progressively smaller, they may be difficult to detect thus leading to a reduction in IPSC frequency. This fits rather well with the fact that we also observed a slow but significant reduction of the amplitude of glycinergic, electrically evoked IPSCs when treating the slice with the GlyT2 blocker but not with the GlyT1 inhibitor (Fig. 4). Lamina X neurones display GABAergic as well as glycinergic mIPSCs, and in most cases we have isolated glycinergic transmission by adding bicuculline to the external solution in order to block GABAA receptors. This situation precludes the detection of an eventual GABA–glycine cotransmission. Cotransmission of glycine and GABA seems now to be the rule rather than an exception and it is highly probable that lamina X also exhibits GABA–glycine cotransmission. A true cotransmission is suggested by the fact that vesicular transporter of inhibitory amino acids (VIAAT) does not discriminate between glycine and GABA (Dumoulin et al. 1999). Therefore the ratio of GABA/glycine in the vesicles is likely to be determined in part by the ratio of GABA and glycine in the cytoplasm. As a consequence, regulation in the expression or activity of neuronal GlyT will determine the cytoplasmic glycine content and might also determine in part the degree of GABA–glycine cotransmission. Little is known concerning the regulation of GlyT, and the physiological context in which such a modulation of the neuronal GlyT might occur remains to be identified.
Potentiation of NMDA-mediated EPSCs by GlyT blockers
NMDA-evoked EPSCs are enhanced by micromolar concentrations of D-serine or glycine whereas DCKA blocked NMDA-eEPSCs. This indicates that under normal condition the glycine site of the NMDA receptor (glyB site) is probably not saturated by glycine or D-serine as also reported for example in CA1 hippocampal cells (Martina et al. 2003). NMDA receptors are composed of NR1 and NR2 subunits. Although NR1 subunits possess the glyB site for the binding of glycine, the affinity of the receptor for glycine is mainly determined by the NR2 subunit. The lowest affinity (micromolar) has been attributed to the presence of NR2D while NR2A–C confer affinities in the submicromolar range. The composition of the NMDA receptors in lamina X is unknown but NR1 and NR2D are abundantly expressed subunits in lamina X (Tolle et al. 1993). Thus if the NMDA receptors are of the NR1/NR2D type, one might expect that they would be saturated by glycine, whose concentration in the extracellular medium of the rat spinal cord was found to be in the micromolar range (Whitehead et al. 2001). However, we doubt that synaptic NMDA receptors responsible of the eEPSCs recorded in our study were of the NR1/NR2D type because (a) these receptors have been shown to generate EPSCs with a very slow decaying phase lasting several seconds and such eEPSCs have never been recorded during these experiments, and (b) NR1/NR2D are mainly located extrasynaptically (Misra et al. 2000a,b). Whatever the subunit composition of NMDA receptors expressed in lamina X neurones might be, a submicromolar concentration of glycine would be expected to saturate the glyB site of these receptors. This is obviously not the case. In line with this observation, we showed that GlyT blocker enhanced markedly the amplitude of NMDA eEPSCs leading to the view that GlyT reduced the level of extracellular glycine close to synaptic NMDA receptors to a non-saturating level. Supplisson & Bergman (1997) have shown that GlyT could indeed lower the effective concentration of glycine close to the NMDA receptor by a factor of 10 as compared to the concentration of glycine in the superfusion medium.
We observed a large increase in the amplitude of the NMDA-eEPSCs when GlyT were blocked. Similar enhancement has been observed in hippocampal (Bergeron et al. 1998) and prefrontal cortex slices (Chen et al. 2003) and has been also observed only after the blockade of glial GlyT. Recently, it has been shown that glycine might spill over and reach glycinergic excitatory receptors present on glutamatergic terminals (Turecek & Trussell, 2001). This would also lead to a facilitation of glutamatergic transmission In this study, as GlyT did not affect the non-NMDA component of glutamatergic EPSCs and therefore did not affect the release of glutamate, this rules out a possible spillover of glycine at putative presynaptic glycine receptors on glutamatergic terminals. In the superficial dorsal horn, synaptically released glycine reached NMDA receptors and this spillover plays an important role in nociception (Ahmadi et al. 2003). The spillover of glycine and the potentiation of the NMDA response are markedly reduced by the activation of presynaptic receptors to the neuropeptide nociceptin, which inhibits the release of glycine. In lamina X, we have shown that terminal nicotinic receptors are present on glycinergic terminals and their activation, in contrast to nociceptin, enhances the release of glycine (Bradaïa & Trouslard, 2002b). If the spillover phenomenon exists in lamina X as in more superficial layers of the spinal cord, one would expect that nicotine receptor activation stimulates the synaptic NMDA receptor depending on the density, localization and activity of GlyT (especially GlyT2). This has to be proven, but nevertheless this points to the pivotal and crucial role of GlyT in the fine tuning of the synaptic NMDA receptor-mediated synaptic transmission in areas of the spinal cord involved in pain processing.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Ahmadi S, Muth-Selbach U, Lauterbach A, Lipfert P, Neuhuber WL & Zeilhofer HU (2003). Facilitation of spinal NMDA receptor currents by spillover of synaptically released glycine. Science 300, 2094–2097.
Bergeron R, Meyer TM, Coyle JT & Greene RW (1998). Modulation of N-methyl-D-aspartate receptor function by glycine transport. Proc Natl Acad Sci U S A 95, 15730–15734.
Bordey A, Feltz P & Trouslard J (1996). Nicotinic actions on neurones of the central autonomic area in rat spinal cord slices. J Physiol 497, 175–187.[Medline]
Bradaïa A & Trouslard J (2002a). Fast synaptic transmission mediated by alpha-bungarotoxin-sensitive nicotinic acetylcholine receptors in lamina X neurones of neonatal rat spinal cord. J Physiol 544, 727–739.
Bradaïa A & Trouslard J (2002b). Nicotinic receptors regulate the release of glycine onto lamina X neurones of the rat spinal cord. Neuropharmacology 43, 1044–1054.[CrossRef][Medline]
Brown A, Carlyle I, Clark J, Hamilton W, Gibson S, McGarry G, McEachen S, Rae D, Thorn S & Walker G (2001). Discovery and SAR of org 24598 – a selective glycine uptake inhibitor. Bioorg Med Chem Lett 11, 2007–2009.[CrossRef][Medline]
Caulfield WL, Collie IT, Dickins RS, Epemolu O, McGuire R, Hill DR, McVey G, Morphy JR, Rankovic Z & Sundaram H (2001). The first potent and selective inhibitors of the glycine transporter type 2. J Med Chem 44, 2679–2682.[CrossRef][Medline]
Chen L, Muhlhauser M & Yang CR (2003). Glycine tranporter-1 blockade potentiates NMDA-mediated responses in rat prefrontal cortical neurons in vitro and in vivo. J Neurophysiol 89, 691–703.
Chery N & Koninck Y (1999). Junctional versus extrajunctional glycine and GABAA receptor-mediated IPSCs in identified lamina I neurons of the adult rat spinal cord. J Neurosci 19, 7342–7355.
Clements JD (1996). Transmitter timecourse in the synaptic cleft: its role in central synaptic function. Trends Neurosci 19, 163–171.[CrossRef][Medline]
Dumoulin A, Rostaing P, Bedet C, Levi S, Isambert MF, Henry JP, Triller A & Gasnier B (1999). Presence of the vesicular inhibitory amino acid transporter in GABAergic and glycinergic synaptic terminal boutons. J Cell Sci 112, 811–823.[Abstract]
Friauf E, Aragon C, Lohrke S, Westenfelder B & Zafra F (1999). Developmental expression of the glycine transporter GLYT2 in the auditory system of rats suggests involvement in synapse maturation. J Comp Neurol 412, 17–37.[CrossRef][Medline]
Gadea A & Lopez-Colome AM (2001). Glial transporters for glutamate, glycine, and GABA III. Glycine transporters. J Neurosci Res 64, 218–222.[CrossRef][Medline]
Gao BX, Cheng G & Ziskind-Conhaim L (1998). Development of spontaneous synaptic transmission in the rat spinal cord. J Neurophysiol 79, 2277–2287.
Geerlings A, Nunez E, Lopez-Corcuera B & Aragon C (2001). Calcium- and syntaxin 1-mediated trafficking of the neuronal glycine transporter GLYT2. J Biol Chem 276, 17584–17590.
Gomeza J, Hulsmann S, Ohno K, Eulenburg V, Szoke K, Richter D & Betz H (2003a). Inactivation of the glycine transporter 1 gene discloses vital role of glial glycine uptake in glycinergic inhibition. Neuron 40, 785–796.[CrossRef][Medline]
Gomeza J, Ohno K, Hulsmann S, Armsen W, Eulenburg V, Richter DW, Laube B & Betz H (2003b). Deletion of the mouse glycine transporter 2 results in a hyperekplexia phenotype and postnatal lethality. Neuron 40, 797–806.[CrossRef][Medline]
Guastella J, Brecha N, Weigmann C, Lester HA & Davidson N (1992). Cloning, expression, and localization of a rat brain high-affinity glycine transporter. Proc Natl Acad Sci U S A 89, 7189–7193.
Hosoya Y, Nadelhaft I, Wang D & Kohno K (1994). Thoracolumbar sympathetic preganglionic neurons in the dorsal commissural nucleus of the male rat: an immunohistochemical study using retrograde labeling of cholera toxin subunit B. Exp Brain Res 98, 21–30.[Medline]
Jahr CE & Jessell TM (1985). Synaptic transmission between dorsal root ganglion and dorsal horn neurons in culture: antagonism of monosynaptic excitatory postsynaptic potentials and glutamate excitation by kynurenate. J Neurosci 5, 2281–2289.[Abstract]
Johnson JW & Ascher P (1987). Glycine potentiates the NMDA response in cultured mouse brain neurons. Nature 325, 529–531.[CrossRef][Medline]
Jones MV & Westbrook GL (1996). The impact of receptor desensitization on fast synaptic transmission. Trends Neurosci 19, 96–101.[CrossRef][Medline]
Kim KM, Kingsmore SF, Han H, Yang-Feng TL, Godinot N, Seldin MF, Caron MG & Giros B (1994). Cloning of the human glycine transporter type 1: molecular and pharmacological characterization of novel isoform variants and chromosomal localization of the gene in the human and mouse genomes. Mol Pharmacol 45, 608–617.[Abstract]
Krupp J, Larmet Y & Feltz P (1994). Postnatal change of glycinergic IPSC decay in sympathetic preganglionic neurons. Neuroreport 5, 2437–2440.[Medline]
Legendre P (2001). The glycinergic inhibitory synapse. Cell Mol Life Sci 58, 760–793.[CrossRef][Medline]
Liu QR, Nelson H, Mandiyan S, Lopez-Corcuera B & Nelson N (1992). Cloning and expression of a glycine transporter from mouse brain. FEBS Lett 305, 110–114.[CrossRef][Medline]
Luque JM, Nelson N & Richards JG (1995). Cellular expression of glycine transporter 2 messenger RNA exclusively in rat hindbrain and spinal cord. Neuroscience 64, 525–535.[CrossRef][Medline]
Martina M, Krasteniakov NV & Bergeron R (2003). D-Serine differently modulates NMDA receptor function in rat CA1 hippocampal pyramidal cells and interneurons. J Physiol 548, 411–423.
Misra C, Brickley SG, Farrant M & Cull-Candy SG (2000a). Identification of subunits contributing to synaptic and extrasynaptic NMDA receptors in Golgi cells of the rat cerebellum. J Physiol 524, 147–162.
Misra C, Brickley SG, Wyllie DJ & Cull-Candy SG (2000b). Slow deactivation kinetics of NMDA receptors containing NR1 and NR2D subunits in rat cerebellar Purkinje cells. J Physiol 525, 299–305.
Nahin RL, Madsen AM & Giesler GJ Jr (1983). Anatomical and physiological studies of the gray matter surrounding the spinal cord central canal. J Comp Neurol 220, 321–335.[CrossRef][Medline]
Oku Y, Hulsmann S, Zhang W & Richter DW (1999). Modulation of glycinergic synaptic current kinetics by octanol in mouse hypoglossal motoneurons. Pflugers Arch 438, 656–664.[CrossRef][Medline]
Poyatos I, Ponce J, Aragon C, Gimenez C & Zafra F (1997). The glycine transporter GLYT2 is a reliable marker for glycine immunoreactive neurons. Brain Res Mol Brain Res 49, 63–70.[Medline]
Raiteri L, Paolucci E, Prisco S, Raiteri M & Bonanno G (2003). Activation of a glycine transporter on spinal cord neurons causes enhanced glutamate release in a mouse model of amyotrophic lateral sclerosis. Br J Pharmacol 138, 1021–1025.[CrossRef][Medline]
Raiteri L, Raiteri M & Bonanno G (2001). Glycine is taken up through GLYT1 and GLYT2 transporters into mouse spinal cord axon terminals and causes vesicular and carrier-mediated release of its proposed co-transmitter GABA. J Neurochem 76, 1823–1832.[CrossRef][Medline]
Roux MJ & Supplisson S (2000). Neuronal and glial glycine transporters have different stoichiometries. Neuron 25, 373–383.[CrossRef][Medline]
