extrasynaptic localization of glycine receptors in the rat ... - UNIC, CNRS
Neuroscience 133 (2005) 175–183
EXTRASYNAPTIC LOCALIZATION OF GLYCINE RECEPTORS IN THE RAT SUPRAOPTIC NUCLEUS: FURTHER EVIDENCE FOR THEIR INVOLVEMENT IN GLIA-TO-NEURON COMMUNICATION C. DELEUZE,a1 G. ALONSO,a1 I. A. LEFEVRE,b A. DUVOID-GUILLOUa1 AND N. HUSSYa*
Glycine receptors (GlyRs) are well known for their participation in synaptic inhibition in the spinal cord and brain stem where glycine is the major inhibitory neurotransmitter. GlyRs are also present in numerous higher brain structures, such as hippocampus (Chattipakorn and McMahon, 2002), retina (Zhou, 2001), cerebellum (Kaneda et al., 1995), striatum (Sergeeva and Haas, 2001), amygdala (McCool and Botting, 2000), nucleus accumbens (Martin and Siggins, 2002; Jiang et al., 2004), cortex (Flint et al., 1998), substantia nigra (Mangin et al., 2002) or hypothalamus (Tokutomi et al., 1989; Hussy et al., 1997), but their involvement in synaptic transmission could be demonstrated only in retina and cerebellum (Dieudonne, 1995; Protti et al., 1997; Dumoulin et al., 2001). The role of the receptors in the other brain structures remains largely obscure. In fact, it has been proposed that these receptors could mediate a non-synaptic transmission, in hypothalamus (Hussy et al., 1997), cortical developing neurons (Flint et al., 1998), and hippocampus (Mori et al., 2002). A well-characterized system where such a non-synaptic activation of GlyRs occurs is the hypothalamo–neurohypophysial complex (see Hussy et al., 2000; Hussy, 2002). Neurons in the supraoptic nucleus (SON) express high levels of functional GlyRs on their soma (Hussy et al., 1997), as well as on their axon terminals in the neurohypophysis (Hussy et al., 2001). In vivo, blockade of these receptors by strychnine prevents part of the inhibition of the activity of vasopressin neurons induced by a decrease in plasma osmolarity (Hussy et al., 1997), directly implicating these GlyRs in the control of SON neuron electrical activity by peripheral osmotic stimuli. Strychnine also prevents the inhibitory action of hypoosmolarity on the release of vasopressin from isolated neurohypophysis, demonstrating the functional involvement of axonal GlyRs in the osmotic regulation of neurohormone secretion (Hussy et al., 2001). However, the involvement of a glycinergic inhibitory transmission in this system is doubtful. Only very few glycinergic fibers have been detected by immunohistochemistry in the rat SON (van den Pol and Gorcs, 1988; Rampon et al., 1996), and none in the neurohypophysis (Pow, 1993). Moreover, all miniature, spontaneous and evoked inhibitory postsynaptic potentials and currents in the SON seem to be GABAergic since they can all be blocked by GABAA receptor (GABAAR) antagonists (Randle and Renaud, 1987; Wuarin and Dudek, 1993; Kabashima et al., 1997; Mouginot et al., 1998). These results suggest the absence of glycinergic afferent input in the SON. To account for the osmodependent activation of GlyRs, it was
a
Biologie des Neurones Endocrines, CNRS-UMR5101 CCIPE, 141 rue de la Cardonille, 34094 Montpellier, Cedex 5, France b Sanofi-Aventis, 371 rue du Pr. J. Blayac, 34184 Montpellier, Cedex 04, France
Abstract—Neurons of the rat supraoptic nucleus (SON) express glycine receptors (GlyRs), which are implicated in the osmoregulation of neuronal activity. The endogenous agonist of the receptors has been postulated to be taurine, shown to be released from astrocytes. We here provide additional pieces of evidence supporting the absence of functional glycinergic synapses in the SON. First, we show that blockade of GlyRs with strychnine has no effect on either the amplitude or frequency of miniature inhibitory postsynaptic currents recorded in SON neurons, whereas they were all suppressed by the GABAA antagonist gabazine. Then, double immunostaining of sections with presynaptic markers and either GlyR or GABAA receptor (GABAAR) antibodies indicates that, in contrast with GABAARs, most GlyR membrane clusters are not localized facing presynaptic terminals, indicative of their extrasynaptic localization. Moreover, we found a striking anatomical association between SON GlyR clusters and glial fibrillary acidic protein (GFAP)-positive astroglial processes, which contain high levels of taurine. This type of correlation is specific to GlyRs, since GABAAR clusters show no association with GFAP-positive structures. These results substantiate and strengthen the concept of extrasynaptic GlyRs mediating a paracrine communication between astrocytes and neurons in the SON. © 2005 IBRO. Published by Elsevier Ltd. All rights reserved. Key words: synaptic currents, GABAA receptors, synaptic receptors, osmoregulation, neuroendocrine cells.
1
Present address: Department Neurology and Neurological Sciences, Stanford University Medical Center, Room M016, 300 Pasteur Drive, Stanford, CA 94305, USA (C. Deleuze); CNRS-UMR 5203, Institut de Génomique Fonctionnelle, 141 rue de la Cardonille, 34094 Montpellier Cedex 5, France (G. Alonso and A. Duvoid-Guillou). *Correspondence to: N. Hussy, Dynamique des Réseaux Neuronaux, INSERM U704 Domaine Universitaire, UFR Biologie, Bât B, 2280 rue de la Piscine, BP 53 38041 Grenoble Cedex 9, France. Tel: ⫹33-476635405; fax: ⫹33-4766-35415. E-mail address: [email protected] (N. Hussy). Abbreviations: aCSF, artificial cerebrospinal fluid; ATP, adenosine triphosphate; EGTA, ethylene glycol-bis-(2-aminoethylether)-N,N,N=,N=tetraacetic acid; EPSC, excitatory post-synaptic current; GABAAR, GABA receptor type A; GFAP, glial fibrillary acidic protein; GlyR, glycine receptor; IPSC, inhibitory post-synaptic current; mIPSC, miniature inhibitory post-synaptic current; PBS, phosphate-buffered saline; SNARE, soluble N-ethylmaleimide-sensitive factor attachment protein-receptor; SON, supraoptic nucleus; TTX, tetrodotoxin.
0306-4522/05$30.00⫹0.00 © 2005 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2005.01.060
175
176
C. Deleuze et al. / Neuroscience 133 (2005) 175–183
early proposed that taurine contained in astrocytes could be the natural agonist, acting via a non-synaptic mechanism (Hatton, 1990; Hussy et al., 1997). Indeed, a high level of taurine is found almost exclusively in astrocytes of the SON (Decavel and Hatton, 1995) and pituicytes of the neurohypophysis (Pow, 1993; Miyata et al., 1997; Pow et al., 2002). In both structures, glial taurine is released via volume-activated anion channels during cell swelling occurring in hypoosmotic conditions, with a very high sensitivity to even minute changes in osmotic pressure (Miyata et al., 1997; Deleuze et al., 1998; Brès et al., 2000). Moreover, taurine is an efficient agonist at both somatic and axonal GlyRs (Hussy et al., 1997, 2001). That taurine is the true endogenous agonist of the receptors is indicated by the observation that 1) there is no osmodependent release of the other GlyR agonists glycine and -alanine (Hussy et al., 1997), 2) the osmosensitivity of taurine release matches remarkably well that of GlyR activation in vivo (see Hussy et al., 2000), and 3) taurine depletion from isolated neurohypophysis prevents the osmodependent, strychnine-sensitive inhibition of vasopressin secretion (Hussy et al., 2001). Endogenous activation of GlyRs by a glial factor would imply an extrasynaptic localization of the receptors. We here confirm the complete absence of any glycinergic component of inhibitory transmission in the rat SON, and show by immunohistochemistry the lack of correlation of SON neuron GlyR clusters with presynaptic terminals. Furthermore we have found that these receptor clusters are specifically located in close apposition with astrocytic processes. These data constitute strong evidence for an extrasynaptic localization of GlyRs in SON neurons, clustered in membrane segments facing astrocytic elements, in a way compatible with their activation by a glial factor, reinforcing the idea that these ionotropic receptors serve to transmit a glia-to-neuron communication.
EXPERIMENTAL PROCEDURES Electrophysiological recordings Synaptic currents. Adult male Wistar rats (Iffa Credo, l’Arbresle, France) were decapitated and their brains removed and immersed for 1 min in a cold (4 °C) carbogenated artificial cerebrospinal fluid (aCSF) in which NaCl was replaced with sucrose (aCSF-sucrose, in mM: sucrose, 200; KCl, 2.3; KH2PO4, 1.2; CaCl2, 2.5; MgSO4, 1.2; NaHCO3, 26; glucose, 10; osmolarity 300 mOsmol l⫺1). A block containing the hypothalamus was isolated and 350 m thick coronal slices were cut at 4 °C using a vibratome (Campden Instruments, Leicester, UK). Slices containing the SON were then incubated for 1–2 h in a carbogenated aCSF (in mM: NaCl, 125; KCl, 2.3; KH2PO4, 1.2; CaCl2, 2.5; MgSO4, 1.2; NaHCO3, 26; glucose, 10; osmolarity 300 mOsmol l⫺1) to which was added 0.2 mM of the antioxidants thiourea and ascorbic acid. During the incubation period, the temperature was raised progressively up to 35 °C. Slices were transferred to the recording chamber, and perfused with aCSF. Test substances were applied in the perfusion solution. Recordings were performed under whole cell voltage-clamp using the patch-clamp technique. Electrodes had a resistance of 4 – 6 M⍀ when filled with the internal solution (in mM: KCH3SO3, 140; HEPES-Na, 10; CaCl2, 1; EGTA-Na, 5; ATP-Mg, 4; pH, 7.2;
osmolarity, 295 mOsmol l⫺1). The theoretical equilibrium potential for Cl⫺ ions (ECl) was ⫺105.5 mV. The ⫺10 mV junction potential, measured according to Neher (1992), was systematically corrected for. SON magnocellular neurons were identified by the presence of a transient outward rectifier K⫹ current and the absence of inward rectification (Armstrong et al., 1994). Spontaneous postsynaptic currents were recorded at holding potentials of ⫺40 to ⫺60 mV. In these conditions, excitatory and inhibitory post-synaptic currents (EPSCs and IPSCs) appeared as inward and outward currents, respectively. Currents were amplified by an Axopatch 200A (Axon Instruments, Union City, CA, USA) and stored on DAT (Biologic, Claix, France). They were then filtered at 1 kHz and digitized at 2.5 kHz using pClamp6 or Axotape software (Axon Instruments). Analysis of frequency and amplitude of synaptic currents was performed using Origin 4.1 software (Origin Lab, Northampton, MA, USA) with a subroutine written in the laboratory. Statistical significance was assessed by non-parametric Wilcoxon signed rank test. Tetrodotoxin (TTX), strychnine, forskolin, and gabazine (SR 95531) were from Sigma (St Quentin, Fallavier, France). Immunohistochemistry. After deep anesthesia with pentobarbital (300 mg/kg), animals were perfused through the ascending aorta with 100 ml phosphate-buffered saline (PBS), pH 7.4, followed by 500 ml of 4% paraformaldehyde, or 4% paraformaldehyde plus 0.5% glutaraldehyde in 0.1 M phosphate buffer, pH 7.4. The brain was removed and immersed in the same fixative for 8 h. Tissues were cut with a vibratome into 40-m-thick sections and rinsed in PBS. Sections were incubated for 48 h at 4 °C with antibodies against GlyRs (mAb4a, monoclonal mouse IgG, that recognizes all ␣ subunits; Alexis, Lausen, Switzerland; diluted 1:200), GABAARs (polyclonal rabbit IgG that recognizes the ␥2 subunit; Chemicon, Temecula, CA, USA; diluted 1:200), synaptobrevin (polyclonal rabbit IgG; kind gift of M. Seagar, UMR464 INSERM/Univ. de la Méditerranée, Marseille, France; diluted 1:2000), synaptophysin (monoclonal mouse IgG; Sigma; diluted 1:2000), and taurine (monoclonal mouse IgG; kindly provided by K. Magnusson, Colorado State University, Fort Collins, CO, USA; diluted 1:1000). After rinsing in PBS, sections were incubated for 4 h with secondary antibodies against rabbit or mouse IgG conjugated with Cy3 (Jackson Laboratories, Bar Harbor, ME, USA; diluted 1:2000) or with Alexa 488 (Molecular Probes, Eugene, OR, USA; diluted 1:2000). Primary and secondary antibodies were diluted in PBS containing 2% bovine serum albumin and 0.1% Triton X-100. After rinsing, sections were mounted in Mowiol (Calbiochem, San Diego, CA, USA) and observed under a Biorad MRC 1024 confocal laser scanning microscope (Carl Zeiss, Le Pecq, France) equipped with a krypton/argon mixed gas laser. Two laser lines emitting at 488 nm and 568 nm were used to excite the Alexa 488- and Cy3-conjugated secondary antibodies, respectively. The background noise of each confocal image was reduced by averaging five image inputs. Immunostained structures were studied on single confocal images of 1–2 m thick. Unaltered digitized images were transferred to a PC computer and Photoshop (Adobe, Paris, France) was used to prepare and print final figures. No fluorescent labeling was detected when the primary antibodies were omitted.
RESULTS High expression of GlyR clusters in the SON Electrophysiological measurements have revealed high levels of functional expression of GlyRs in all SON neurons (Hussy et al., 1997). Indeed, application of glycine onto acutely dissociated SON neurons recorded under wholecell configuration induced large inward currents that were mostly blocked after pre-incubation with 1 M strychnine
C. Deleuze et al. / Neuroscience 133 (2005) 175–183
177
Fig. 1. The SON exhibits high density of GlyRs as compared with other forebrain regions. The density of GlyR-immunostained clusters detected within the SON (A, B) appear lower than that detected within the dorsal spinal cord (E), but far higher than that found within the striatum (C) or the amygdala (D). Note that, although a high density of GlyR-immunolabeling is detected within the ventral portion of the nucleus that contains the dendritic processes (dz in A), a heavy staining is observed at the periphery of magnocellular neuron cell bodies throughout the whole nucleus (B). Scale bars⫽50 m in A; 15 m in E (applies for B–E).
(Hussy et al., 1997). This was confirmed by immunohistochemistry using the mAb4a antibody, which recognizes all ␣ subunits of GlyRs (Fig. 1). Labeling appeared punctuate, indicative of dense clusters dispersed throughout the whole SON, and with an especially dense labeling in the ventral portion of the nucleus (Fig. 1A), an area known as the dendritic zone of the SON which contains most of the magnocellular neuron dendrites. When observed at high magnification, such GlyR-immunostained clusters appeared to be distributed around most of the magnocellular cell bodies of the nucleus (Fig. 1B). When comparing the density of GlyR labeling in different brain regions, the SON appeared heavily labeled, areas like striatum (Fig. 1C) or amygdala (Fig. 1D) showing clear but scarcer clusters, a higher level of expression being evident only in the spinal cord (Fig. 1D).
Absence of glycinergic synaptic currents in the SON In the spinal cord, clusters of GlyRs are believed to represent synaptic receptors, aggregated in the post-synaptic density (Kneussel and Betz, 2000). Although earlier studies have shown that all evoked and miniature synaptic inhibitory currents and potentials in the SON are blocked by GABAAR antagonists (Wuarin and Dudek, 1993; Kabashima et al., 1997; Mouginot et al., 1998), the presence of glycinergic synapses has not been directly investigated, and the possibility remained that a minor glycinergic component of inhibitory transmission would have gone undetected after GABAAR blockade. Synaptic currents were therefore recorded from SON magnocellular neurons in slices, under whole-cell configuration. In our experimental conditions and at a holding
Abbreviations used in the figures dz dendritic zone OC optic chiasma Synb synaptobrevin
Synp synaptophysin
178
C. Deleuze et al. / Neuroscience 133 (2005) 175–183
Fig. 2. Strychnine does not inhibit mIPSCs. (A–C) Addition of strychnine (1 M) has no effect on either the amplitude or frequency of mIPSCs recorded in the presence of TTX, as shown on the recordings of synaptic activity (A) and on the cumulative histograms of amplitude and frequency of mIPSCs derived from these recordings (B). The average effect of strychnine is shown on the mean histograms in (C). In contrast, all mIPSCs (outward currents) are abolished by the application of 3 M gabazine (A). These traces also show the lack of effect of both antagonists on mEPSCs (inward currents). Holding potential was ⫺60 mV. (D–F) Application of forskolin dramatically increases the frequency of mIPSCs without changing their amplitude. Even in these conditions of enhanced activity, application of 1 M strychnine affects neither the amplitude nor the frequency synaptic currents. Holding potential was ⫺40 mV.
potential of ⫺60 mV, excitatory synaptic activity appeared as inward currents (EPSCs), and inhibitory synaptic activity appeared as outward currents (IPSCs). As shown earlier (Kabashima et al., 1997), these spontaneous synaptic currents are independent of cell electrical activity because neither their amplitude nor their frequency was affected by the application of 3.10⫺7 M TTX (data not shown). On average, the amplitude and frequency of spontaneous IPSCs in the presence of TTX represented 95⫾4% and 89⫾7% of control, respectively (n⫽8); these values were
not significantly different (P⬎0.05). Therefore, the vast majority of spontaneous inhibitory synaptic currents of SON neurons recorded in this preparation are miniature currents (mIPSCs) independent of neuronal electrical activity. The properties of mIPSCs were not affected by the application of 1 M strychnine (Fig. 2A–C). On average, the amplitude and frequency of mIPSCs in the presence of strychnine represented 94.3⫾2.7% and 99.3⫾6.1% of control, respectively (n⫽10, P⬎0.05). In contrast, all mIPSCs were blocked by the application of 3 M gabazine, a highly
C. Deleuze et al. / Neuroscience 133 (2005) 175–183
179
Fig. 3. GlyR clusters are not specifically associated with presynaptic terminals. Superimposed confocal images of sections double immunostained for either GlyRs and synaptobrevin (A–C), or GABAARs and synaptophysin (D–F). GlyR-immunostained clusters located at the periphery of the SON neuron cell bodies frequently appear at a distance from synaptobrevin-immunostained axon terminals, whereas a majority of the GABAARimmunostained clusters appears closely apposed to synaptophysin-immunostained terminals. The areas shown at higher magnification in (B, C) and (E, F), are delineated in (A) and (D) respectively. Scale bars⫽15 m in D (applies for A and D) and 7.5 m in F (applies for B, C, E, F).
specific inhibitor of GABAAR (n⫽7; Fig. 2A). However, because rare glycinergic events could still go undetected in these basal conditions, we checked the effect of strychnine in conditions of enhanced synaptic activity. Activation of adenylate cyclase is known to enhance the frequency of miniature synaptic currents by a presynaptic mechanism (Capogna et al., 1995; Chen and Regehr, 1997). Application of 10 M forskolin onto SON neurons did not modify the amplitude of mIPSCs (which represented 94.8⫾4.6% of control; n⫽6, P⬎0.05), but considerably increased their frequency (which reached on average 534⫾192% of control; n⫽6, P⬍0.05; Fig. 2D). Further addition of strychnine (1 M) did not significantly change the characteristics of mIPSCs, with mean amplitude and frequency representing 94.5⫾4.7% and 105.3⫾6.0% of those recorded in the presence of forskolin, respectively (n⫽6, P⬍0,05; Fig. 2D–F). These results strongly suggested that no functional glycin-
ergic synapse can be evidenced in the SON. The vast majority of inhibitory synapses appear to be GABAergic, involving postsynaptic GABAARs. Most GlyR clusters are not facing presynaptic terminals We then checked whether the absence of functional glycinergic synapses could be confirmed morphologically. We did co-immunolabeling of GlyRs and the presynaptic protein synaptobrevin, which belongs to the vesicular release soluble N-ethylmaleimide-sensitive factor attachment proteinreceptor (SNARE) complex. Synaptobrevin-like immunoreactivity appeared as a punctiform labeling spread over the nucleus (Fig. 3A). No clear association of GlyR clusters with synaptobrevin was apparent (Fig. 3A–C). In fact most GlyR clusters were located either in areas devoid of pre-
180
C. Deleuze et al. / Neuroscience 133 (2005) 175–183
synaptic labeling, or were not in close proximity of a presynaptic terminal. Only few receptor clusters were close enough to a labeled axon ending to suggest a postsynaptic localization. This observation strikingly contrasted with the frequent correspondence of GABAAR clusters with synaptophysin (another protein of the SNARE complex) -positive synaptic terminals (Fig. 3D, E). We could not perform double labeling of synaptobrevin and GABAAR because both antibodies were polyclonal rabbit antibodies, justifying the use of a monoclonal antibody directed against synaptophysin. However, synaptophysin and synaptobrevin co-localize in the same axon terminals (Becher et al., 1999), where they form a complex crucially involved in the regulation of vesicle fusion. These results indicate that GlyRs appear mostly extrasynaptic. Close association of GlyR clusters with glial elements This extrasynaptic localization of GlyRs in the SON is in agreement with their proposed activation by taurine. Taurine was initially shown by electron microscopy to be selectively concentrated in SON astrocytes (Decavel and Hatton, 1995) and has since been demonstrated to be specifically released from these glial cells (Deleuze et al., 1998; Hussy et al., 2000). We first confirmed the co-immunolabeling of taurine and glial fibrillary acidic protein (GFAP) under fluorescent microscopy (Fig. 4A–C). Most taurine-like immunoreactivity in the SON was strongly co-localized or closely associated with GFAP-immunolabeling. We next performed double immunohistochemical labeling of GlyRs and GFAP, and found a striking correlation between receptor clusters and glial processes (Fig. 4D–F). Most GlyR clusters were located along structures that surrounded neuronal cell bodies, and which were positively labeled by the GFAP antibody. Astrocytes in the SON are known to enwrap neuronal cell bodies and dendrites, and therefore any membrane receptor could appear closely apposed to a glial process. We checked for the specificity of the close association of SON GlyRs with glial structures by comparing it with the co-labeling of GABAARs and GFAP in the SON (Fig. 4G), and of GlyRs and GFAP in the spinal cord (Fig. 4H). Opposite to what we observed with SON GlyRs, neither GABAARs in the SON, nor GlyRs in the spinal cord, were facing GFAP-positive elements, affirming the unique targeting of GlyRs to the cellular population that releases its agonist. We made multiple attempts to directly show the presence of GlyRs facing astroglial membranes by electron microscopy, but in our hands, using either pre- or postembedding immunolabeling, most receptor labeling was lost. This is likely due to the glutaraldehyde fixation used for electron microscopy, which is deleterious for antigen recognition by mAb4a. A similar fixation procedure also eliminated the labeling of receptor clusters under fluorescence microscopy (data not shown).
DISCUSSION GlyRs in the SON have been proposed to be activated by endogenous taurine, which is released from astrocytes in an osmodependent manner (see Hussy et al., 2000; Hussy, 2002). We here show that no functional glycinergic synaptic currents can be detected in the SON and that most GlyR-like immunoreactivity is not found facing presynaptic terminals. Second, we found a striking and specific anatomical association between GlyR clusters and astrocytic processes, which contain the receptor agonist taurine. These data complete our previous physiological results on the functional activation of GlyRs by taurine released from astrocytes (Hussy et al., 1997, 2001; Deleuze et al., 1998), further arguing for the involvement of these ionotropic receptors in glia-to-neuron transmission. The expression of GlyRs is very high in the SON. GlyR-like immunostaining appears as punctuate labeling surrounding the cell bodies and dendrites of magnocellular neurons, strongly suggestive of receptor clusters on the membrane. Although such GlyR-immunostained clusters were detected in other forebrain regions such as the striatum or the amygdala, their density always appeared far lower than that detected within the SON. We examined the possible presence of glycinergic synapses in SON neurons by looking at miniature synaptic currents, considering that a glycinergic component of synaptic transmission should be detected when glycine is released by the spontaneous fusion of synaptic vesicles, as shown in the spinal cord and brain stem (Jonas et al., 1998). The lack of effect of strychnine on both mIPSC amplitude and frequency argues against a significant contribution of GlyRs to the inhibitory synaptic transmission in the SON, and suggests the absence of functional glycinergic synapses. These mIPSCs are all blocked by gabazine, a highly selective antagonist of GABAARs, confirming GABA as the principal, if not sole mediator of synaptic inhibition of magnocellular neurons (Randle and Renaud, 1987; Wuarin and Dudek, 1993; Kabashima et al., 1997; Mouginot et al., 1998). It should be noted however, that lack of strychnine-sensitive miniature currents by itself cannot be considered as conclusive evidence for the absence of glycinergic synapses: indeed, in cerebellar Golgi cells, it was recently shown that although spontaneous inhibitory synaptic activity is all GABAergic, a mixed glycine–GABA synaptic transmission can be evoked by stimulation with serotonin of a particular interneuron (Dumoulin et al., 2001). However, we found that strychnine had no effect even under conditions of strongly enhanced synaptic activity evoked by forskolin, suggesting that the contribution of glycinergic synapses to inhibitory transmission in the SON is at most very modest, unlikely to account for the high expression of the receptors in SON neurons. Moreover, these results agree with previous immunohistochemical studies that have failed to consistently detect glycinergic fibers within the SON (van den Pol and Gorcs, 1988; Rampon et al., 1996). Accordingly, we show that throughout the SON, most GlyR clusters are localized in areas not facing labeled presynaptic terminals, a situation that contrasts with that of GABAARs
C. Deleuze et al. / Neuroscience 133 (2005) 175–183
181
Fig. 4. Astrocytic processes contain taurine and are closely associated with GlyR clusters. (A–C) Confocal images of a section double immunostained for GFAP and taurine showing that intense taurine-immunostaining is associated with the GFAP-immunostained astrocytic processes extending throughout the SON (C corresponds to the superimposition of both immunostainings within the area delineated in A). (D–H) Superimposed confocal images of sections double immunostained for GFAP and either GlyRs (D–F, H) or GABARs (G) showing that 1) throughout the SON, GFAPimmunostained processes are closely associated with GlyR- (D–F) but not GABAAR- immunostained clusters (G), and 2) within the dorsal spinal cord, GFAP-immunostained processes are not specifically associated with GlyR-immunostained clusters (H). Scale bars⫽15 m.
detected in the same nucleus, which appear frequently apposed to such labeled presynaptic terminals. Instead, GlyR clusters are found very closely associated with GFAP-positive processes that run between and around
neuron somata and dendrites in the SON. Again, this relation to astrocytic elements is specific to GlyRs, since GABAAR clusters show no such association. Furthermore, such preferential anatomical relationship between
182
C. Deleuze et al. / Neuroscience 133 (2005) 175–183
GlyR clusters and astrocyte processes was not observed in the dorsal spinal cord. These results strongly argue for the extrasynaptic expression of GlyRs in magnocellular neurons. Together with our previous observations, namely the parallel activation of GlyRs and release of glial taurine in the SON and neurohypophysis by osmotic stimulation both in vivo and in vitro, the requirement for the presence of taurine in this activation, as well as the absence of glycine release in conditions of enhanced GlyR activity, (Hussy et al., 1997, 2001; Deleuze et al., 1998), the present results reinforce the notion of the particular role played by GlyRs in this structure, mediating a non-synaptic, paracrine transmission from glial cells to neurons. While other neuronal ionotropic receptors such as glutamate receptors have been shown to mediate gliato-neuron transmission in various brain areas (Araque et al., 2001; Haydon, 2001; Parri et al., 2001), such a role for GlyRs has only been shown in the hypothalamo– neurohypophysial system. Extrasynaptic activation of GlyRs by taurine has also been suggested in other parts of the brain, such as the developing cortex (Flint et al., 1998) or the hippocampus (Mori et al., 2002), but the origin of taurine was either proposed to be neuronal or not determined. Actually, a recent study on cultured hippocampal neurons has reported the synaptic localization of GlyR clusters (Brackmann et al., 2004), although the level of glial cells in the culture was not mentioned. Our results constitute a compelling argument for the activation of GlyRs by taurine released from glial cells. Moreover, taking advantage of the important knowledge of the physiology of the hypothalamo–neurohypophysial complex, we could clearly identify the role of GlyRs in the in vivo physiological osmoregulation of the activity of SON magnocellular neurons and of the release of vasopressin (Hussy et al., 2000; Hussy, 2002). The involvement of GlyRs in extrasynaptic, volume transmission in higher brain structures could well be more the rule than the exception, contrasting with the major function of GlyRs in fast synaptic transmission seen in spinal cord, brain stem, retina, and recently cerebellum. Actually, we know of little evidence for synaptic activation of these receptors in the forebrain, since search for glycinergic synaptic events have as yet been unsuccessful (this study, and Mangin et al., 2002; Mori et al., 2002). Interestingly, paracrine activation of GlyRs may also occur in spinal glial cells and neural stem cells, which express the receptors (Pastor et al., 1995; Nguyen et al., 2002), as well as outside the nervous system, with expression of GlyRs on spermatozoids (Sato et al., 2000), endothelial cells (Yamashina et al., 2001), hepatic cells (Qu et al., 2002), and various immune system-derived cells (Spittler et al., 1999; Wheeler et al., 2000; Froh et al., 2002). Acknowledgments—We are grateful to Pierre Fontanaud for writing the routine of analysis of synaptic currents, to Françoise Moos for critical reading of the manuscript, and to Mike Seagar and Kathy Magnusson for the generous gift of the synaptobrevin and taurine antibodies, respectively. Confocal microscopy was per-
formed using the facilities of the Centre Régional d’Imagerie Cellulaire (CRIC), Montpellier, France.
REFERENCES Araque A, Carmignoto G, Haydon PG (2001) Dynamic signaling between astrocytes and neurons. Annu Rev Physiol 63:795– 813. Armstrong WE, Smith BN, Tian M (1994) Electrophysiological characteristics of immunochemically identified rat oxytocin and vasopressin neurones in vitro. J Physiol 475:115–128. Becher A, Drenckhahn A, Pahner I, Margittai M, Jahn R, Ahnert-Hilger G (1999) The synaptophysin-synaptobrevin complex: a hallmark of synaptic vesicle maturation. J Neurosci 19:1922–1931. Brackmann M, Zhao C, Schmieden V, Braunewell KH (2004) Cellular and subcellular localization of the inhibitory glycine receptor in hippocampal neurons. Biochem Biophys Res Commun 324:1137–1142. Brès V, Hurbin A, Duvoid A, Orcel H, Moos FC, Rabie A, Hussy N (2000) Pharmacological characterization of volume-sensitive, taurine permeable anion channels in rat supraoptic glial cells. Br J Pharmacol 130:1976 –1982. Capogna M, Gahwiler BH, Thompson SM (1995) Presynaptic enhancement of inhibitory synaptic transmission by protein kinases A and C in the rat hippocampus in vitro. J Neurosci 15:1249 –1260. Chattipakorn SC, McMahon LL (2002) Pharmacological characterization of glycine-gated chloride currents recorded in rat hippocampal slices. J Neurophysiol 87:1515–1525. Chen C, Regehr WG (1997) The mechanism of cAMP-mediated enhancement at a cerebellar synapse. J Neurosci 17:8687– 8694. Decavel C, Hatton GI (1995) Taurine immunoreactivity in the rat supraoptic nucleus: prominent localization in glial cells. J Comp Neurol 354:13–26. Deleuze C, Duvoid A, Hussy N (1998) Properties and glial origin of osmotic-dependent release of taurine from the rat supraoptic nucleus. J Physiol 507(Pt 2):463– 471. Dieudonne S (1995) Glycinergic synaptic currents in Golgi cells of the rat cerebellum. Proc Natl Acad Sci USA 92:1441–1445. Dumoulin A, Triller A, Dieudonne S (2001) IPSC kinetics at identified GABAergic and mixed GABAergic and glycinergic synapses onto cerebellar Golgi cells. J Neurosci 21:6045– 6057. Flint AC, Liu X, Kriegstein AR (1998) Nonsynaptic glycine receptor activation during early neocortical development. Neuron 20:43–53. Froh M, Thurman RG, Wheeler MD (2002) Molecular evidence for a glycine-gated chloride channel in macrophages and leukocytes. Am J Physiol Gastrointest Liver Physiol 283:G856 –G863. Hatton GI (1990) Emerging concepts of structure-function dynamics in adult brain: the hypothalamo-neurohypophysial system. Prog Neurobiol 34:437–504. Haydon PG (2001) GLIA: listening and talking to the synapse. Nat Rev Neurosci 2:185–193. Hussy N (2002) Glial cells in the hypothalamo-neurohypophysial system: key elements of the regulation of neuronal electrical and secretory activity. Prog Brain Res 139:95–112. Hussy N, Brès V, Rochette M, Duvoid A, Alonso G, Dayanithi G, Moos FC (2001) Osmoregulation of vasopressin secretion via activation of neurohypophysial nerve terminals glycine receptors by glial taurine. J Neurosci 21:7110 –7116. Hussy N, Deleuze C, Desarmenien MG, Moos FC (2000) Osmotic regulation of neuronal activity: a new role for taurine and glial cells in a hypothalamic neuroendocrine structure. Prog Neurobiol 62:113–134. Hussy N, Deleuze C, Pantaloni A, Desarmenien MG, Moos F (1997) Agonist action of taurine on glycine receptors in rat supraoptic magnocellular neurones: possible role in osmoregulation. J Physiol 502(Pt 3):609 – 621. Jiang Z, Krnjevic K, Wang F, Ye JH (2004) Taurine activates strychninesensitive glycine receptors in neurons freshly isolated from nucleus accumbens of young rats. J Neurophysiol 91:248 –257. Jonas P, Bischofberger J, Sandkuhler J (1998) Corelease of two fast neurotransmitters at a central synapse. Science 281:419 – 424.
C. Deleuze et al. / Neuroscience 133 (2005) 175–183 Kabashima N, Shibuya I, Ibrahim N, Ueta Y, Yamashita H (1997) Inhibition of spontaneous EPSCs and IPSCs by presynaptic GABAB receptors on rat supraoptic magnocellular neurons. J Physiol 504(Pt 1):113–126. Kaneda M, Farrant M, Cull-Candy SG (1995) Whole-cell and singlechannel currents activated by GABA and glycine in granule cells of the rat cerebellum. J Physiol 485(Pt 2):419 – 435. Kneussel M, Betz H (2000) Receptors, gephyrin and gephyrinassociated proteins: novel insights into the assembly of inhibitory postsynaptic membrane specializations. J Physiol 525(Pt 1):1–9. Mangin JM, Guyon A, Eugene D, Paupardin-Tritsch D, Legendre P (2002) Functional glycine receptor maturation in the absence of glycinergic input in dopaminergic neurones of the rat substantia nigra. J Physiol 542:685– 697. Martin G, Siggins GR (2002) Electrophysiological evidence for expression of glycine receptors in freshly isolated neurons from nucleus accumbens. J Pharmacol Exp Ther 302:1135–1145. McCool BA, Botting SK (2000) Characterization of strychnine-sensitive glycine receptors in acutely isolated adult rat basolateral amygdala neurons. Brain Res 859:341–351. Miyata S, Matsushima O, Hatton GI (1997) Taurine in rat posterior pituitary: localization in astrocytes and selective release by hypoosmotic stimulation. J Comp Neurol 381:513–523. Mori M, Gahwiler BH, Gerber U (2002) Beta-alanine and taurine as endogenous agonists at glycine receptors in rat hippocampus in vitro. J Physiol 539:191–200. Mouginot D, Kombian SB, Pittman QJ (1998) Activation of presynaptic GABAB receptors inhibits evoked IPSCs in rat magnocellular neurons in vitro. J Neurophysiol 79:1508 –1517. Neher E (1992) Correction for liquid junction potentials in patch clamp experiments. Methods Enzymol 207:123–131. Nguyen L, Malgrange B, Belachew S, Rogister B, Rocher V, Moonen G, Rigo JM (2002) Functional glycine receptors are expressed by postnatal nestin-positive neural stem/progenitor cells. Eur J Neurosci 15:1299 –1305. Parri HR, Gould TM, Crunelli V (2001) Spontaneous astrocytic Ca2⫹ oscillations in situ drive NMDAR-mediated neuronal excitation. Nat Neurosci 4:803– 812. Pastor A, Chvatal A, Sykova E, Kettenmann H (1995) Glycine- and GABA-activated currents in identified glial cells of the developing rat spinal cord slice. Eur J Neurosci 7:1188 –1198. Pow DV (1993) Immunocytochemistry of amino-acids in the rodent pituitary using extremely specific, very high titre antisera. J Neuroendocrinol 5:349 –356.
183
Pow DV, Sullivan R, Reye P, Hermanussen S (2002) Localization of taurine transporters, taurine, and (3)H taurine accumulation in the rat retina, pituitary, and brain. Glia 37:153–168. Protti DA, Gerschenfeld HM, Llano I (1997) GABAergic and glycinergic IPSCs in ganglion cells of rat retinal slices. J Neurosci 17:6075–6085. Qu W, Ikejima K, Zhong Z, Waalkes MP, Thurman RG (2002) Glycine blocks the increase in intracellular free Ca2⫹ due to vasoactive mediators in hepatic parenchymal cells. Am J Physiol Gastrointest Liver Physiol 283:G1249 –G1256. Rampon C, Luppi PH, Fort P, Peyron C, Jouvet M (1996) Distribution of glycine-immunoreactive cell bodies and fibers in the rat brain. Neuroscience 75:737–755. Randle JC, Renaud LP (1987) Actions of gamma-aminobutyric acid on rat supraoptic nucleus neurosecretory neurones in vitro. J Physiol 387:629 – 647. Sato Y, Son JH, Meizel S (2000) The mouse sperm glycine receptor/ chloride channel: cellular localization and involvement in the acrosome reaction initiated by glycine. J Androl 21:99 –106. Sergeeva OA, Haas HL (2001) Expression and function of glycine receptors in striatal cholinergic interneurons from rat and mouse. Neuroscience 104:1043–1055. Spittler A, Reissner CM, Oehler R, Gornikiewicz A, Gruenberger T, Manhart N, Brodowicz T, Mittlboeck M, Boltz-Nitulescu G, Roth E (1999) Immunomodulatory effects of glycine on LPS-treated monocytes: reduced TNF-alpha production and accelerated IL-10 expression. FASEB J 13:563–571. Tokutomi N, Kaneda M, Akaike N (1989) What confers specificity on glycine for its receptor site? Br J Pharmacol 97:353–360. van den Pol AN, Gorcs T (1988) Glycine and glycine receptor immunoreactivity in brain and spinal cord. J Neurosci 8:472– 492. Wheeler M, Stachlewitz RF, Yamashina S, Ikejima K, Morrow AL, Thurman RG (2000) Glycine-gated chloride channels in neutrophils attenuate calcium influx and superoxide production. FASEB J 14:476 – 484. Wuarin JP, Dudek FE (1993) Patch-clamp analysis of spontaneous synaptic currents in supraoptic neuroendocrine cells of the rat hypothalamus. J Neurosci 13:2323–2331. Yamashina S, Konno A, Wheeler MD, Rusyn I, Rusyn EV, Cox AD, Thurman RG (2001) Endothelial cells contain a glycine-gated chloride channel. Nutr Cancer 40:197–204. Zhou ZJ (2001) A critical role of the strychnine-sensitive glycinergic system in spontaneous retinal waves of the developing rabbit. J Neurosci 21:5158 –5168.
(Accepted 31 January 2005) (Available online 19 April 2005)
EXTRASYNAPTIC LOCALIZATION OF GLYCINE RECEPTORS IN THE RAT SUPRAOPTIC NUCLEUS: FURTHER EVIDENCE FOR THEIR INVOLVEMENT IN GLIA-TO-NEURON COMMUNICATION C. DELEUZE,a1 G. ALONSO,a1 I. A. LEFEVRE,b A. DUVOID-GUILLOUa1 AND N. HUSSYa*
Glycine receptors (GlyRs) are well known for their participation in synaptic inhibition in the spinal cord and brain stem where glycine is the major inhibitory neurotransmitter. GlyRs are also present in numerous higher brain structures, such as hippocampus (Chattipakorn and McMahon, 2002), retina (Zhou, 2001), cerebellum (Kaneda et al., 1995), striatum (Sergeeva and Haas, 2001), amygdala (McCool and Botting, 2000), nucleus accumbens (Martin and Siggins, 2002; Jiang et al., 2004), cortex (Flint et al., 1998), substantia nigra (Mangin et al., 2002) or hypothalamus (Tokutomi et al., 1989; Hussy et al., 1997), but their involvement in synaptic transmission could be demonstrated only in retina and cerebellum (Dieudonne, 1995; Protti et al., 1997; Dumoulin et al., 2001). The role of the receptors in the other brain structures remains largely obscure. In fact, it has been proposed that these receptors could mediate a non-synaptic transmission, in hypothalamus (Hussy et al., 1997), cortical developing neurons (Flint et al., 1998), and hippocampus (Mori et al., 2002). A well-characterized system where such a non-synaptic activation of GlyRs occurs is the hypothalamo–neurohypophysial complex (see Hussy et al., 2000; Hussy, 2002). Neurons in the supraoptic nucleus (SON) express high levels of functional GlyRs on their soma (Hussy et al., 1997), as well as on their axon terminals in the neurohypophysis (Hussy et al., 2001). In vivo, blockade of these receptors by strychnine prevents part of the inhibition of the activity of vasopressin neurons induced by a decrease in plasma osmolarity (Hussy et al., 1997), directly implicating these GlyRs in the control of SON neuron electrical activity by peripheral osmotic stimuli. Strychnine also prevents the inhibitory action of hypoosmolarity on the release of vasopressin from isolated neurohypophysis, demonstrating the functional involvement of axonal GlyRs in the osmotic regulation of neurohormone secretion (Hussy et al., 2001). However, the involvement of a glycinergic inhibitory transmission in this system is doubtful. Only very few glycinergic fibers have been detected by immunohistochemistry in the rat SON (van den Pol and Gorcs, 1988; Rampon et al., 1996), and none in the neurohypophysis (Pow, 1993). Moreover, all miniature, spontaneous and evoked inhibitory postsynaptic potentials and currents in the SON seem to be GABAergic since they can all be blocked by GABAA receptor (GABAAR) antagonists (Randle and Renaud, 1987; Wuarin and Dudek, 1993; Kabashima et al., 1997; Mouginot et al., 1998). These results suggest the absence of glycinergic afferent input in the SON. To account for the osmodependent activation of GlyRs, it was
a
Biologie des Neurones Endocrines, CNRS-UMR5101 CCIPE, 141 rue de la Cardonille, 34094 Montpellier, Cedex 5, France b Sanofi-Aventis, 371 rue du Pr. J. Blayac, 34184 Montpellier, Cedex 04, France
Abstract—Neurons of the rat supraoptic nucleus (SON) express glycine receptors (GlyRs), which are implicated in the osmoregulation of neuronal activity. The endogenous agonist of the receptors has been postulated to be taurine, shown to be released from astrocytes. We here provide additional pieces of evidence supporting the absence of functional glycinergic synapses in the SON. First, we show that blockade of GlyRs with strychnine has no effect on either the amplitude or frequency of miniature inhibitory postsynaptic currents recorded in SON neurons, whereas they were all suppressed by the GABAA antagonist gabazine. Then, double immunostaining of sections with presynaptic markers and either GlyR or GABAA receptor (GABAAR) antibodies indicates that, in contrast with GABAARs, most GlyR membrane clusters are not localized facing presynaptic terminals, indicative of their extrasynaptic localization. Moreover, we found a striking anatomical association between SON GlyR clusters and glial fibrillary acidic protein (GFAP)-positive astroglial processes, which contain high levels of taurine. This type of correlation is specific to GlyRs, since GABAAR clusters show no association with GFAP-positive structures. These results substantiate and strengthen the concept of extrasynaptic GlyRs mediating a paracrine communication between astrocytes and neurons in the SON. © 2005 IBRO. Published by Elsevier Ltd. All rights reserved. Key words: synaptic currents, GABAA receptors, synaptic receptors, osmoregulation, neuroendocrine cells.
1
Present address: Department Neurology and Neurological Sciences, Stanford University Medical Center, Room M016, 300 Pasteur Drive, Stanford, CA 94305, USA (C. Deleuze); CNRS-UMR 5203, Institut de Génomique Fonctionnelle, 141 rue de la Cardonille, 34094 Montpellier Cedex 5, France (G. Alonso and A. Duvoid-Guillou). *Correspondence to: N. Hussy, Dynamique des Réseaux Neuronaux, INSERM U704 Domaine Universitaire, UFR Biologie, Bât B, 2280 rue de la Piscine, BP 53 38041 Grenoble Cedex 9, France. Tel: ⫹33-476635405; fax: ⫹33-4766-35415. E-mail address: [email protected] (N. Hussy). Abbreviations: aCSF, artificial cerebrospinal fluid; ATP, adenosine triphosphate; EGTA, ethylene glycol-bis-(2-aminoethylether)-N,N,N=,N=tetraacetic acid; EPSC, excitatory post-synaptic current; GABAAR, GABA receptor type A; GFAP, glial fibrillary acidic protein; GlyR, glycine receptor; IPSC, inhibitory post-synaptic current; mIPSC, miniature inhibitory post-synaptic current; PBS, phosphate-buffered saline; SNARE, soluble N-ethylmaleimide-sensitive factor attachment protein-receptor; SON, supraoptic nucleus; TTX, tetrodotoxin.
0306-4522/05$30.00⫹0.00 © 2005 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2005.01.060
175
176
C. Deleuze et al. / Neuroscience 133 (2005) 175–183
early proposed that taurine contained in astrocytes could be the natural agonist, acting via a non-synaptic mechanism (Hatton, 1990; Hussy et al., 1997). Indeed, a high level of taurine is found almost exclusively in astrocytes of the SON (Decavel and Hatton, 1995) and pituicytes of the neurohypophysis (Pow, 1993; Miyata et al., 1997; Pow et al., 2002). In both structures, glial taurine is released via volume-activated anion channels during cell swelling occurring in hypoosmotic conditions, with a very high sensitivity to even minute changes in osmotic pressure (Miyata et al., 1997; Deleuze et al., 1998; Brès et al., 2000). Moreover, taurine is an efficient agonist at both somatic and axonal GlyRs (Hussy et al., 1997, 2001). That taurine is the true endogenous agonist of the receptors is indicated by the observation that 1) there is no osmodependent release of the other GlyR agonists glycine and -alanine (Hussy et al., 1997), 2) the osmosensitivity of taurine release matches remarkably well that of GlyR activation in vivo (see Hussy et al., 2000), and 3) taurine depletion from isolated neurohypophysis prevents the osmodependent, strychnine-sensitive inhibition of vasopressin secretion (Hussy et al., 2001). Endogenous activation of GlyRs by a glial factor would imply an extrasynaptic localization of the receptors. We here confirm the complete absence of any glycinergic component of inhibitory transmission in the rat SON, and show by immunohistochemistry the lack of correlation of SON neuron GlyR clusters with presynaptic terminals. Furthermore we have found that these receptor clusters are specifically located in close apposition with astrocytic processes. These data constitute strong evidence for an extrasynaptic localization of GlyRs in SON neurons, clustered in membrane segments facing astrocytic elements, in a way compatible with their activation by a glial factor, reinforcing the idea that these ionotropic receptors serve to transmit a glia-to-neuron communication.
EXPERIMENTAL PROCEDURES Electrophysiological recordings Synaptic currents. Adult male Wistar rats (Iffa Credo, l’Arbresle, France) were decapitated and their brains removed and immersed for 1 min in a cold (4 °C) carbogenated artificial cerebrospinal fluid (aCSF) in which NaCl was replaced with sucrose (aCSF-sucrose, in mM: sucrose, 200; KCl, 2.3; KH2PO4, 1.2; CaCl2, 2.5; MgSO4, 1.2; NaHCO3, 26; glucose, 10; osmolarity 300 mOsmol l⫺1). A block containing the hypothalamus was isolated and 350 m thick coronal slices were cut at 4 °C using a vibratome (Campden Instruments, Leicester, UK). Slices containing the SON were then incubated for 1–2 h in a carbogenated aCSF (in mM: NaCl, 125; KCl, 2.3; KH2PO4, 1.2; CaCl2, 2.5; MgSO4, 1.2; NaHCO3, 26; glucose, 10; osmolarity 300 mOsmol l⫺1) to which was added 0.2 mM of the antioxidants thiourea and ascorbic acid. During the incubation period, the temperature was raised progressively up to 35 °C. Slices were transferred to the recording chamber, and perfused with aCSF. Test substances were applied in the perfusion solution. Recordings were performed under whole cell voltage-clamp using the patch-clamp technique. Electrodes had a resistance of 4 – 6 M⍀ when filled with the internal solution (in mM: KCH3SO3, 140; HEPES-Na, 10; CaCl2, 1; EGTA-Na, 5; ATP-Mg, 4; pH, 7.2;
osmolarity, 295 mOsmol l⫺1). The theoretical equilibrium potential for Cl⫺ ions (ECl) was ⫺105.5 mV. The ⫺10 mV junction potential, measured according to Neher (1992), was systematically corrected for. SON magnocellular neurons were identified by the presence of a transient outward rectifier K⫹ current and the absence of inward rectification (Armstrong et al., 1994). Spontaneous postsynaptic currents were recorded at holding potentials of ⫺40 to ⫺60 mV. In these conditions, excitatory and inhibitory post-synaptic currents (EPSCs and IPSCs) appeared as inward and outward currents, respectively. Currents were amplified by an Axopatch 200A (Axon Instruments, Union City, CA, USA) and stored on DAT (Biologic, Claix, France). They were then filtered at 1 kHz and digitized at 2.5 kHz using pClamp6 or Axotape software (Axon Instruments). Analysis of frequency and amplitude of synaptic currents was performed using Origin 4.1 software (Origin Lab, Northampton, MA, USA) with a subroutine written in the laboratory. Statistical significance was assessed by non-parametric Wilcoxon signed rank test. Tetrodotoxin (TTX), strychnine, forskolin, and gabazine (SR 95531) were from Sigma (St Quentin, Fallavier, France). Immunohistochemistry. After deep anesthesia with pentobarbital (300 mg/kg), animals were perfused through the ascending aorta with 100 ml phosphate-buffered saline (PBS), pH 7.4, followed by 500 ml of 4% paraformaldehyde, or 4% paraformaldehyde plus 0.5% glutaraldehyde in 0.1 M phosphate buffer, pH 7.4. The brain was removed and immersed in the same fixative for 8 h. Tissues were cut with a vibratome into 40-m-thick sections and rinsed in PBS. Sections were incubated for 48 h at 4 °C with antibodies against GlyRs (mAb4a, monoclonal mouse IgG, that recognizes all ␣ subunits; Alexis, Lausen, Switzerland; diluted 1:200), GABAARs (polyclonal rabbit IgG that recognizes the ␥2 subunit; Chemicon, Temecula, CA, USA; diluted 1:200), synaptobrevin (polyclonal rabbit IgG; kind gift of M. Seagar, UMR464 INSERM/Univ. de la Méditerranée, Marseille, France; diluted 1:2000), synaptophysin (monoclonal mouse IgG; Sigma; diluted 1:2000), and taurine (monoclonal mouse IgG; kindly provided by K. Magnusson, Colorado State University, Fort Collins, CO, USA; diluted 1:1000). After rinsing in PBS, sections were incubated for 4 h with secondary antibodies against rabbit or mouse IgG conjugated with Cy3 (Jackson Laboratories, Bar Harbor, ME, USA; diluted 1:2000) or with Alexa 488 (Molecular Probes, Eugene, OR, USA; diluted 1:2000). Primary and secondary antibodies were diluted in PBS containing 2% bovine serum albumin and 0.1% Triton X-100. After rinsing, sections were mounted in Mowiol (Calbiochem, San Diego, CA, USA) and observed under a Biorad MRC 1024 confocal laser scanning microscope (Carl Zeiss, Le Pecq, France) equipped with a krypton/argon mixed gas laser. Two laser lines emitting at 488 nm and 568 nm were used to excite the Alexa 488- and Cy3-conjugated secondary antibodies, respectively. The background noise of each confocal image was reduced by averaging five image inputs. Immunostained structures were studied on single confocal images of 1–2 m thick. Unaltered digitized images were transferred to a PC computer and Photoshop (Adobe, Paris, France) was used to prepare and print final figures. No fluorescent labeling was detected when the primary antibodies were omitted.
RESULTS High expression of GlyR clusters in the SON Electrophysiological measurements have revealed high levels of functional expression of GlyRs in all SON neurons (Hussy et al., 1997). Indeed, application of glycine onto acutely dissociated SON neurons recorded under wholecell configuration induced large inward currents that were mostly blocked after pre-incubation with 1 M strychnine
C. Deleuze et al. / Neuroscience 133 (2005) 175–183
177
Fig. 1. The SON exhibits high density of GlyRs as compared with other forebrain regions. The density of GlyR-immunostained clusters detected within the SON (A, B) appear lower than that detected within the dorsal spinal cord (E), but far higher than that found within the striatum (C) or the amygdala (D). Note that, although a high density of GlyR-immunolabeling is detected within the ventral portion of the nucleus that contains the dendritic processes (dz in A), a heavy staining is observed at the periphery of magnocellular neuron cell bodies throughout the whole nucleus (B). Scale bars⫽50 m in A; 15 m in E (applies for B–E).
(Hussy et al., 1997). This was confirmed by immunohistochemistry using the mAb4a antibody, which recognizes all ␣ subunits of GlyRs (Fig. 1). Labeling appeared punctuate, indicative of dense clusters dispersed throughout the whole SON, and with an especially dense labeling in the ventral portion of the nucleus (Fig. 1A), an area known as the dendritic zone of the SON which contains most of the magnocellular neuron dendrites. When observed at high magnification, such GlyR-immunostained clusters appeared to be distributed around most of the magnocellular cell bodies of the nucleus (Fig. 1B). When comparing the density of GlyR labeling in different brain regions, the SON appeared heavily labeled, areas like striatum (Fig. 1C) or amygdala (Fig. 1D) showing clear but scarcer clusters, a higher level of expression being evident only in the spinal cord (Fig. 1D).
Absence of glycinergic synaptic currents in the SON In the spinal cord, clusters of GlyRs are believed to represent synaptic receptors, aggregated in the post-synaptic density (Kneussel and Betz, 2000). Although earlier studies have shown that all evoked and miniature synaptic inhibitory currents and potentials in the SON are blocked by GABAAR antagonists (Wuarin and Dudek, 1993; Kabashima et al., 1997; Mouginot et al., 1998), the presence of glycinergic synapses has not been directly investigated, and the possibility remained that a minor glycinergic component of inhibitory transmission would have gone undetected after GABAAR blockade. Synaptic currents were therefore recorded from SON magnocellular neurons in slices, under whole-cell configuration. In our experimental conditions and at a holding
Abbreviations used in the figures dz dendritic zone OC optic chiasma Synb synaptobrevin
Synp synaptophysin
178
C. Deleuze et al. / Neuroscience 133 (2005) 175–183
Fig. 2. Strychnine does not inhibit mIPSCs. (A–C) Addition of strychnine (1 M) has no effect on either the amplitude or frequency of mIPSCs recorded in the presence of TTX, as shown on the recordings of synaptic activity (A) and on the cumulative histograms of amplitude and frequency of mIPSCs derived from these recordings (B). The average effect of strychnine is shown on the mean histograms in (C). In contrast, all mIPSCs (outward currents) are abolished by the application of 3 M gabazine (A). These traces also show the lack of effect of both antagonists on mEPSCs (inward currents). Holding potential was ⫺60 mV. (D–F) Application of forskolin dramatically increases the frequency of mIPSCs without changing their amplitude. Even in these conditions of enhanced activity, application of 1 M strychnine affects neither the amplitude nor the frequency synaptic currents. Holding potential was ⫺40 mV.
potential of ⫺60 mV, excitatory synaptic activity appeared as inward currents (EPSCs), and inhibitory synaptic activity appeared as outward currents (IPSCs). As shown earlier (Kabashima et al., 1997), these spontaneous synaptic currents are independent of cell electrical activity because neither their amplitude nor their frequency was affected by the application of 3.10⫺7 M TTX (data not shown). On average, the amplitude and frequency of spontaneous IPSCs in the presence of TTX represented 95⫾4% and 89⫾7% of control, respectively (n⫽8); these values were
not significantly different (P⬎0.05). Therefore, the vast majority of spontaneous inhibitory synaptic currents of SON neurons recorded in this preparation are miniature currents (mIPSCs) independent of neuronal electrical activity. The properties of mIPSCs were not affected by the application of 1 M strychnine (Fig. 2A–C). On average, the amplitude and frequency of mIPSCs in the presence of strychnine represented 94.3⫾2.7% and 99.3⫾6.1% of control, respectively (n⫽10, P⬎0.05). In contrast, all mIPSCs were blocked by the application of 3 M gabazine, a highly
C. Deleuze et al. / Neuroscience 133 (2005) 175–183
179
Fig. 3. GlyR clusters are not specifically associated with presynaptic terminals. Superimposed confocal images of sections double immunostained for either GlyRs and synaptobrevin (A–C), or GABAARs and synaptophysin (D–F). GlyR-immunostained clusters located at the periphery of the SON neuron cell bodies frequently appear at a distance from synaptobrevin-immunostained axon terminals, whereas a majority of the GABAARimmunostained clusters appears closely apposed to synaptophysin-immunostained terminals. The areas shown at higher magnification in (B, C) and (E, F), are delineated in (A) and (D) respectively. Scale bars⫽15 m in D (applies for A and D) and 7.5 m in F (applies for B, C, E, F).
specific inhibitor of GABAAR (n⫽7; Fig. 2A). However, because rare glycinergic events could still go undetected in these basal conditions, we checked the effect of strychnine in conditions of enhanced synaptic activity. Activation of adenylate cyclase is known to enhance the frequency of miniature synaptic currents by a presynaptic mechanism (Capogna et al., 1995; Chen and Regehr, 1997). Application of 10 M forskolin onto SON neurons did not modify the amplitude of mIPSCs (which represented 94.8⫾4.6% of control; n⫽6, P⬎0.05), but considerably increased their frequency (which reached on average 534⫾192% of control; n⫽6, P⬍0.05; Fig. 2D). Further addition of strychnine (1 M) did not significantly change the characteristics of mIPSCs, with mean amplitude and frequency representing 94.5⫾4.7% and 105.3⫾6.0% of those recorded in the presence of forskolin, respectively (n⫽6, P⬍0,05; Fig. 2D–F). These results strongly suggested that no functional glycin-
ergic synapse can be evidenced in the SON. The vast majority of inhibitory synapses appear to be GABAergic, involving postsynaptic GABAARs. Most GlyR clusters are not facing presynaptic terminals We then checked whether the absence of functional glycinergic synapses could be confirmed morphologically. We did co-immunolabeling of GlyRs and the presynaptic protein synaptobrevin, which belongs to the vesicular release soluble N-ethylmaleimide-sensitive factor attachment proteinreceptor (SNARE) complex. Synaptobrevin-like immunoreactivity appeared as a punctiform labeling spread over the nucleus (Fig. 3A). No clear association of GlyR clusters with synaptobrevin was apparent (Fig. 3A–C). In fact most GlyR clusters were located either in areas devoid of pre-
180
C. Deleuze et al. / Neuroscience 133 (2005) 175–183
synaptic labeling, or were not in close proximity of a presynaptic terminal. Only few receptor clusters were close enough to a labeled axon ending to suggest a postsynaptic localization. This observation strikingly contrasted with the frequent correspondence of GABAAR clusters with synaptophysin (another protein of the SNARE complex) -positive synaptic terminals (Fig. 3D, E). We could not perform double labeling of synaptobrevin and GABAAR because both antibodies were polyclonal rabbit antibodies, justifying the use of a monoclonal antibody directed against synaptophysin. However, synaptophysin and synaptobrevin co-localize in the same axon terminals (Becher et al., 1999), where they form a complex crucially involved in the regulation of vesicle fusion. These results indicate that GlyRs appear mostly extrasynaptic. Close association of GlyR clusters with glial elements This extrasynaptic localization of GlyRs in the SON is in agreement with their proposed activation by taurine. Taurine was initially shown by electron microscopy to be selectively concentrated in SON astrocytes (Decavel and Hatton, 1995) and has since been demonstrated to be specifically released from these glial cells (Deleuze et al., 1998; Hussy et al., 2000). We first confirmed the co-immunolabeling of taurine and glial fibrillary acidic protein (GFAP) under fluorescent microscopy (Fig. 4A–C). Most taurine-like immunoreactivity in the SON was strongly co-localized or closely associated with GFAP-immunolabeling. We next performed double immunohistochemical labeling of GlyRs and GFAP, and found a striking correlation between receptor clusters and glial processes (Fig. 4D–F). Most GlyR clusters were located along structures that surrounded neuronal cell bodies, and which were positively labeled by the GFAP antibody. Astrocytes in the SON are known to enwrap neuronal cell bodies and dendrites, and therefore any membrane receptor could appear closely apposed to a glial process. We checked for the specificity of the close association of SON GlyRs with glial structures by comparing it with the co-labeling of GABAARs and GFAP in the SON (Fig. 4G), and of GlyRs and GFAP in the spinal cord (Fig. 4H). Opposite to what we observed with SON GlyRs, neither GABAARs in the SON, nor GlyRs in the spinal cord, were facing GFAP-positive elements, affirming the unique targeting of GlyRs to the cellular population that releases its agonist. We made multiple attempts to directly show the presence of GlyRs facing astroglial membranes by electron microscopy, but in our hands, using either pre- or postembedding immunolabeling, most receptor labeling was lost. This is likely due to the glutaraldehyde fixation used for electron microscopy, which is deleterious for antigen recognition by mAb4a. A similar fixation procedure also eliminated the labeling of receptor clusters under fluorescence microscopy (data not shown).
DISCUSSION GlyRs in the SON have been proposed to be activated by endogenous taurine, which is released from astrocytes in an osmodependent manner (see Hussy et al., 2000; Hussy, 2002). We here show that no functional glycinergic synaptic currents can be detected in the SON and that most GlyR-like immunoreactivity is not found facing presynaptic terminals. Second, we found a striking and specific anatomical association between GlyR clusters and astrocytic processes, which contain the receptor agonist taurine. These data complete our previous physiological results on the functional activation of GlyRs by taurine released from astrocytes (Hussy et al., 1997, 2001; Deleuze et al., 1998), further arguing for the involvement of these ionotropic receptors in glia-to-neuron transmission. The expression of GlyRs is very high in the SON. GlyR-like immunostaining appears as punctuate labeling surrounding the cell bodies and dendrites of magnocellular neurons, strongly suggestive of receptor clusters on the membrane. Although such GlyR-immunostained clusters were detected in other forebrain regions such as the striatum or the amygdala, their density always appeared far lower than that detected within the SON. We examined the possible presence of glycinergic synapses in SON neurons by looking at miniature synaptic currents, considering that a glycinergic component of synaptic transmission should be detected when glycine is released by the spontaneous fusion of synaptic vesicles, as shown in the spinal cord and brain stem (Jonas et al., 1998). The lack of effect of strychnine on both mIPSC amplitude and frequency argues against a significant contribution of GlyRs to the inhibitory synaptic transmission in the SON, and suggests the absence of functional glycinergic synapses. These mIPSCs are all blocked by gabazine, a highly selective antagonist of GABAARs, confirming GABA as the principal, if not sole mediator of synaptic inhibition of magnocellular neurons (Randle and Renaud, 1987; Wuarin and Dudek, 1993; Kabashima et al., 1997; Mouginot et al., 1998). It should be noted however, that lack of strychnine-sensitive miniature currents by itself cannot be considered as conclusive evidence for the absence of glycinergic synapses: indeed, in cerebellar Golgi cells, it was recently shown that although spontaneous inhibitory synaptic activity is all GABAergic, a mixed glycine–GABA synaptic transmission can be evoked by stimulation with serotonin of a particular interneuron (Dumoulin et al., 2001). However, we found that strychnine had no effect even under conditions of strongly enhanced synaptic activity evoked by forskolin, suggesting that the contribution of glycinergic synapses to inhibitory transmission in the SON is at most very modest, unlikely to account for the high expression of the receptors in SON neurons. Moreover, these results agree with previous immunohistochemical studies that have failed to consistently detect glycinergic fibers within the SON (van den Pol and Gorcs, 1988; Rampon et al., 1996). Accordingly, we show that throughout the SON, most GlyR clusters are localized in areas not facing labeled presynaptic terminals, a situation that contrasts with that of GABAARs
C. Deleuze et al. / Neuroscience 133 (2005) 175–183
181
Fig. 4. Astrocytic processes contain taurine and are closely associated with GlyR clusters. (A–C) Confocal images of a section double immunostained for GFAP and taurine showing that intense taurine-immunostaining is associated with the GFAP-immunostained astrocytic processes extending throughout the SON (C corresponds to the superimposition of both immunostainings within the area delineated in A). (D–H) Superimposed confocal images of sections double immunostained for GFAP and either GlyRs (D–F, H) or GABARs (G) showing that 1) throughout the SON, GFAPimmunostained processes are closely associated with GlyR- (D–F) but not GABAAR- immunostained clusters (G), and 2) within the dorsal spinal cord, GFAP-immunostained processes are not specifically associated with GlyR-immunostained clusters (H). Scale bars⫽15 m.
detected in the same nucleus, which appear frequently apposed to such labeled presynaptic terminals. Instead, GlyR clusters are found very closely associated with GFAP-positive processes that run between and around
neuron somata and dendrites in the SON. Again, this relation to astrocytic elements is specific to GlyRs, since GABAAR clusters show no such association. Furthermore, such preferential anatomical relationship between
182
C. Deleuze et al. / Neuroscience 133 (2005) 175–183
GlyR clusters and astrocyte processes was not observed in the dorsal spinal cord. These results strongly argue for the extrasynaptic expression of GlyRs in magnocellular neurons. Together with our previous observations, namely the parallel activation of GlyRs and release of glial taurine in the SON and neurohypophysis by osmotic stimulation both in vivo and in vitro, the requirement for the presence of taurine in this activation, as well as the absence of glycine release in conditions of enhanced GlyR activity, (Hussy et al., 1997, 2001; Deleuze et al., 1998), the present results reinforce the notion of the particular role played by GlyRs in this structure, mediating a non-synaptic, paracrine transmission from glial cells to neurons. While other neuronal ionotropic receptors such as glutamate receptors have been shown to mediate gliato-neuron transmission in various brain areas (Araque et al., 2001; Haydon, 2001; Parri et al., 2001), such a role for GlyRs has only been shown in the hypothalamo– neurohypophysial system. Extrasynaptic activation of GlyRs by taurine has also been suggested in other parts of the brain, such as the developing cortex (Flint et al., 1998) or the hippocampus (Mori et al., 2002), but the origin of taurine was either proposed to be neuronal or not determined. Actually, a recent study on cultured hippocampal neurons has reported the synaptic localization of GlyR clusters (Brackmann et al., 2004), although the level of glial cells in the culture was not mentioned. Our results constitute a compelling argument for the activation of GlyRs by taurine released from glial cells. Moreover, taking advantage of the important knowledge of the physiology of the hypothalamo–neurohypophysial complex, we could clearly identify the role of GlyRs in the in vivo physiological osmoregulation of the activity of SON magnocellular neurons and of the release of vasopressin (Hussy et al., 2000; Hussy, 2002). The involvement of GlyRs in extrasynaptic, volume transmission in higher brain structures could well be more the rule than the exception, contrasting with the major function of GlyRs in fast synaptic transmission seen in spinal cord, brain stem, retina, and recently cerebellum. Actually, we know of little evidence for synaptic activation of these receptors in the forebrain, since search for glycinergic synaptic events have as yet been unsuccessful (this study, and Mangin et al., 2002; Mori et al., 2002). Interestingly, paracrine activation of GlyRs may also occur in spinal glial cells and neural stem cells, which express the receptors (Pastor et al., 1995; Nguyen et al., 2002), as well as outside the nervous system, with expression of GlyRs on spermatozoids (Sato et al., 2000), endothelial cells (Yamashina et al., 2001), hepatic cells (Qu et al., 2002), and various immune system-derived cells (Spittler et al., 1999; Wheeler et al., 2000; Froh et al., 2002). Acknowledgments—We are grateful to Pierre Fontanaud for writing the routine of analysis of synaptic currents, to Françoise Moos for critical reading of the manuscript, and to Mike Seagar and Kathy Magnusson for the generous gift of the synaptobrevin and taurine antibodies, respectively. Confocal microscopy was per-
formed using the facilities of the Centre Régional d’Imagerie Cellulaire (CRIC), Montpellier, France.
REFERENCES Araque A, Carmignoto G, Haydon PG (2001) Dynamic signaling between astrocytes and neurons. Annu Rev Physiol 63:795– 813. Armstrong WE, Smith BN, Tian M (1994) Electrophysiological characteristics of immunochemically identified rat oxytocin and vasopressin neurones in vitro. J Physiol 475:115–128. Becher A, Drenckhahn A, Pahner I, Margittai M, Jahn R, Ahnert-Hilger G (1999) The synaptophysin-synaptobrevin complex: a hallmark of synaptic vesicle maturation. J Neurosci 19:1922–1931. Brackmann M, Zhao C, Schmieden V, Braunewell KH (2004) Cellular and subcellular localization of the inhibitory glycine receptor in hippocampal neurons. Biochem Biophys Res Commun 324:1137–1142. Brès V, Hurbin A, Duvoid A, Orcel H, Moos FC, Rabie A, Hussy N (2000) Pharmacological characterization of volume-sensitive, taurine permeable anion channels in rat supraoptic glial cells. Br J Pharmacol 130:1976 –1982. Capogna M, Gahwiler BH, Thompson SM (1995) Presynaptic enhancement of inhibitory synaptic transmission by protein kinases A and C in the rat hippocampus in vitro. J Neurosci 15:1249 –1260. Chattipakorn SC, McMahon LL (2002) Pharmacological characterization of glycine-gated chloride currents recorded in rat hippocampal slices. J Neurophysiol 87:1515–1525. Chen C, Regehr WG (1997) The mechanism of cAMP-mediated enhancement at a cerebellar synapse. J Neurosci 17:8687– 8694. Decavel C, Hatton GI (1995) Taurine immunoreactivity in the rat supraoptic nucleus: prominent localization in glial cells. J Comp Neurol 354:13–26. Deleuze C, Duvoid A, Hussy N (1998) Properties and glial origin of osmotic-dependent release of taurine from the rat supraoptic nucleus. J Physiol 507(Pt 2):463– 471. Dieudonne S (1995) Glycinergic synaptic currents in Golgi cells of the rat cerebellum. Proc Natl Acad Sci USA 92:1441–1445. Dumoulin A, Triller A, Dieudonne S (2001) IPSC kinetics at identified GABAergic and mixed GABAergic and glycinergic synapses onto cerebellar Golgi cells. J Neurosci 21:6045– 6057. Flint AC, Liu X, Kriegstein AR (1998) Nonsynaptic glycine receptor activation during early neocortical development. Neuron 20:43–53. Froh M, Thurman RG, Wheeler MD (2002) Molecular evidence for a glycine-gated chloride channel in macrophages and leukocytes. Am J Physiol Gastrointest Liver Physiol 283:G856 –G863. Hatton GI (1990) Emerging concepts of structure-function dynamics in adult brain: the hypothalamo-neurohypophysial system. Prog Neurobiol 34:437–504. Haydon PG (2001) GLIA: listening and talking to the synapse. Nat Rev Neurosci 2:185–193. Hussy N (2002) Glial cells in the hypothalamo-neurohypophysial system: key elements of the regulation of neuronal electrical and secretory activity. Prog Brain Res 139:95–112. Hussy N, Brès V, Rochette M, Duvoid A, Alonso G, Dayanithi G, Moos FC (2001) Osmoregulation of vasopressin secretion via activation of neurohypophysial nerve terminals glycine receptors by glial taurine. J Neurosci 21:7110 –7116. Hussy N, Deleuze C, Desarmenien MG, Moos FC (2000) Osmotic regulation of neuronal activity: a new role for taurine and glial cells in a hypothalamic neuroendocrine structure. Prog Neurobiol 62:113–134. Hussy N, Deleuze C, Pantaloni A, Desarmenien MG, Moos F (1997) Agonist action of taurine on glycine receptors in rat supraoptic magnocellular neurones: possible role in osmoregulation. J Physiol 502(Pt 3):609 – 621. Jiang Z, Krnjevic K, Wang F, Ye JH (2004) Taurine activates strychninesensitive glycine receptors in neurons freshly isolated from nucleus accumbens of young rats. J Neurophysiol 91:248 –257. Jonas P, Bischofberger J, Sandkuhler J (1998) Corelease of two fast neurotransmitters at a central synapse. Science 281:419 – 424.
C. Deleuze et al. / Neuroscience 133 (2005) 175–183 Kabashima N, Shibuya I, Ibrahim N, Ueta Y, Yamashita H (1997) Inhibition of spontaneous EPSCs and IPSCs by presynaptic GABAB receptors on rat supraoptic magnocellular neurons. J Physiol 504(Pt 1):113–126. Kaneda M, Farrant M, Cull-Candy SG (1995) Whole-cell and singlechannel currents activated by GABA and glycine in granule cells of the rat cerebellum. J Physiol 485(Pt 2):419 – 435. Kneussel M, Betz H (2000) Receptors, gephyrin and gephyrinassociated proteins: novel insights into the assembly of inhibitory postsynaptic membrane specializations. J Physiol 525(Pt 1):1–9. Mangin JM, Guyon A, Eugene D, Paupardin-Tritsch D, Legendre P (2002) Functional glycine receptor maturation in the absence of glycinergic input in dopaminergic neurones of the rat substantia nigra. J Physiol 542:685– 697. Martin G, Siggins GR (2002) Electrophysiological evidence for expression of glycine receptors in freshly isolated neurons from nucleus accumbens. J Pharmacol Exp Ther 302:1135–1145. McCool BA, Botting SK (2000) Characterization of strychnine-sensitive glycine receptors in acutely isolated adult rat basolateral amygdala neurons. Brain Res 859:341–351. Miyata S, Matsushima O, Hatton GI (1997) Taurine in rat posterior pituitary: localization in astrocytes and selective release by hypoosmotic stimulation. J Comp Neurol 381:513–523. Mori M, Gahwiler BH, Gerber U (2002) Beta-alanine and taurine as endogenous agonists at glycine receptors in rat hippocampus in vitro. J Physiol 539:191–200. Mouginot D, Kombian SB, Pittman QJ (1998) Activation of presynaptic GABAB receptors inhibits evoked IPSCs in rat magnocellular neurons in vitro. J Neurophysiol 79:1508 –1517. Neher E (1992) Correction for liquid junction potentials in patch clamp experiments. Methods Enzymol 207:123–131. Nguyen L, Malgrange B, Belachew S, Rogister B, Rocher V, Moonen G, Rigo JM (2002) Functional glycine receptors are expressed by postnatal nestin-positive neural stem/progenitor cells. Eur J Neurosci 15:1299 –1305. Parri HR, Gould TM, Crunelli V (2001) Spontaneous astrocytic Ca2⫹ oscillations in situ drive NMDAR-mediated neuronal excitation. Nat Neurosci 4:803– 812. Pastor A, Chvatal A, Sykova E, Kettenmann H (1995) Glycine- and GABA-activated currents in identified glial cells of the developing rat spinal cord slice. Eur J Neurosci 7:1188 –1198. Pow DV (1993) Immunocytochemistry of amino-acids in the rodent pituitary using extremely specific, very high titre antisera. J Neuroendocrinol 5:349 –356.
183
Pow DV, Sullivan R, Reye P, Hermanussen S (2002) Localization of taurine transporters, taurine, and (3)H taurine accumulation in the rat retina, pituitary, and brain. Glia 37:153–168. Protti DA, Gerschenfeld HM, Llano I (1997) GABAergic and glycinergic IPSCs in ganglion cells of rat retinal slices. J Neurosci 17:6075–6085. Qu W, Ikejima K, Zhong Z, Waalkes MP, Thurman RG (2002) Glycine blocks the increase in intracellular free Ca2⫹ due to vasoactive mediators in hepatic parenchymal cells. Am J Physiol Gastrointest Liver Physiol 283:G1249 –G1256. Rampon C, Luppi PH, Fort P, Peyron C, Jouvet M (1996) Distribution of glycine-immunoreactive cell bodies and fibers in the rat brain. Neuroscience 75:737–755. Randle JC, Renaud LP (1987) Actions of gamma-aminobutyric acid on rat supraoptic nucleus neurosecretory neurones in vitro. J Physiol 387:629 – 647. Sato Y, Son JH, Meizel S (2000) The mouse sperm glycine receptor/ chloride channel: cellular localization and involvement in the acrosome reaction initiated by glycine. J Androl 21:99 –106. Sergeeva OA, Haas HL (2001) Expression and function of glycine receptors in striatal cholinergic interneurons from rat and mouse. Neuroscience 104:1043–1055. Spittler A, Reissner CM, Oehler R, Gornikiewicz A, Gruenberger T, Manhart N, Brodowicz T, Mittlboeck M, Boltz-Nitulescu G, Roth E (1999) Immunomodulatory effects of glycine on LPS-treated monocytes: reduced TNF-alpha production and accelerated IL-10 expression. FASEB J 13:563–571. Tokutomi N, Kaneda M, Akaike N (1989) What confers specificity on glycine for its receptor site? Br J Pharmacol 97:353–360. van den Pol AN, Gorcs T (1988) Glycine and glycine receptor immunoreactivity in brain and spinal cord. J Neurosci 8:472– 492. Wheeler M, Stachlewitz RF, Yamashina S, Ikejima K, Morrow AL, Thurman RG (2000) Glycine-gated chloride channels in neutrophils attenuate calcium influx and superoxide production. FASEB J 14:476 – 484. Wuarin JP, Dudek FE (1993) Patch-clamp analysis of spontaneous synaptic currents in supraoptic neuroendocrine cells of the rat hypothalamus. J Neurosci 13:2323–2331. Yamashina S, Konno A, Wheeler MD, Rusyn I, Rusyn EV, Cox AD, Thurman RG (2001) Endothelial cells contain a glycine-gated chloride channel. Nutr Cancer 40:197–204. Zhou ZJ (2001) A critical role of the strychnine-sensitive glycinergic system in spontaneous retinal waves of the developing rabbit. J Neurosci 21:5158 –5168.
(Accepted 31 January 2005) (Available online 19 April 2005)
