HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by De Saint Jan, D. Articles by Bregestovski, P. Search for Related Content PubMed PubMed Citation Articles by De Saint Jan, D. Articles by Bregestovski, P.

Journal of Physiology (2001), 535.3, pp. 741-755
© Copyright 2001 The Physiological Society

Activation of human 1 and 2 homomeric glycine receptors by taurine and GABA

Didier De Saint Jan, Brigitte David-Watine, Henri Korn and Piotr Bregestovski

	ABSTRACT

Top Abstract Introduction Methods Results Discussion Appendix References

1. Two ligand binding subunits, 1 and 2, of the human (H) glycine receptor (GlyR) are involved at inhibitory synapses in the adult and neonatal spinal cord, respectively. The ability of homomeric H1 and H2 GlyRs to be activated by glycine, taurine and GABA was studied in Xenopus oocytes or in the human embryonic kidney HEK-293 cell line.
2. In outside-out patches from HEK cells, glycine, taurine and GABA activated both GlyRs with the same main unitary conductance, i.e. 85 ± 3 pS (n = 6) for H1, and 95 ± 5 pS (n = 4) for H2.
3. The sensitivity of both H1 and H2 GlyRs to glycine was highly variable. In Xenopus oocytes the EC₅₀ for glycine (EC_50gly) was between 25 and 280 µM for H1 (n = 44) and between 46 and 541 µM for H2 (n = 52). For both receptors, the highest EC_50gly values were found on cells with low maximal glycine responses.
4. The actions of taurine and GABA were dependent on the EC_50gly: (i) their EC₅₀ values were linearly correlated to EC_50gly, with EC_50tau 10 EC_50gly and EC_50GABA 500-800 EC_50gly; (ii) they could act either as full or weak agonists depending on the EC_50gly.
5. The Hill coefficient (n_H) of glycine remained stable regardless of the EC_50gly whereas n_H for taurine decreased with increasing EC_50tau.
6. The degree of desensitization, evaluated by fast application of saturating concentrations of agonist on outside-out patches from Xenopus oocytes, was similar for glycine and taurine on both GlyRs and did not exceed 50 %.
7. Our data concerning the variations of EC_50gly and the subsequent behaviour of taurine and GABA could be qualitatively described by the simple del Castillo-Katz scheme, assuming that the agonist gating constant varies whereas the binding constants are stable. However, the stability of the Hill coefficient for glycine was not explained by this model, suggesting that other mechanisms are involved in the modulation of EC₅₀.

	INTRODUCTION

Top Abstract Introduction Methods Results Discussion Appendix References

In the mammalian central nervous system, inhibitory glycine receptors (GlyRs) are mainly expressed in the spinal cord and in the midbrain where they control motor and sensory pathways (Breitinger & Becker, 1998). They form chloride-selective ionic channels which are activated by glycine and, to a lesser extent, by -alanine, taurine and several other amino acids (Werman, 1972; Schmieden et al. 1995, 1999). Four subunits and one subunit have been cloned from mammals. It is generally believed that in adult, GlyRs are heteromers mainly composed of three 1 and two subunits, whereas fetal and neonatal receptors are homomeric 2 GlyRs (for reviews, see Rajendra et al. 1997; Betz et al. 1999), although strong functional evidence of the presence of synaptic homomeric GlyRs is still lacking (see Singer et al. 1998; Ali et al. 2000). The different GlyR subtypes exhibit different functional properties during ontogenesis (Takahashi et al. 1992; Singer et al. 1998; Ali et al. 2000).

We recently cloned an subunit from zebrafish GlyR (named Z1) which displays high sequence similarities to mammalian 1 subunits (David-Watine et al. 1999). Like all the subunits identified so far, Z1 is able to form a functional homomeric GlyR in Xenopus oocytes or in transiently transfected human cell lines. The functional properties of this GlyR are, however, surprisingly different from those composed of human subunits (David-Watine et al. 1999; Fucile et al. 1999). First, Z1 GlyRs are highly sensitive to taurine despite the presence of a valine at position 111, a residue that is thought to confer a low sensitivity to taurine on human GlyRs (Schmieden et al. 1992). Furthermore, Z1 GlyRs can be activated by GABA in the absence of mutations F159 and Y161 which are apparently necessary to transform GABA-insensitive human 1 GlyRs into GABA-sensitive GlyRs (Schmieden et al. 1993). To determine whether these discrepancies are related to species differences, we first re-examined the actions of taurine and GABA on homomeric H1 and H2 GlyRs.

We have also previously demonstrated that for Z1 GlyR the EC₅₀ for glycine (EC_50gly) and the relative maximum response of GABA (defined as the ratio I_maxGABA /I_maxgly) are closely correlated (Fucile et al. 1999). This implies that variations in EC_50gly alter the response to the other agonists dramatically. Although similar properties have never been established for the mammalian GlyRs, various data suggest that the ability of taurine and GABA to activate these GlyRs may also be correlated with the EC_50gly. Firstly, Taleb & Betz (1994) reported that when the EC_50gly of human H1 GlyRs is lowered at high receptor density in Xenopus oocytes, the sensitivity to taurine and to GABA increased. Secondly, the I_maxtau/I_maxgly ratio is always higher if the receptors are expressed in HEK cells, in which they display an EC_50gly of approximatively 30 µM (Rajendra et al. 1995; Lynch et al. 1997; Moorhouse et al. 1999), than in Xenopus oocytes, where the EC_50gly is usually above 200 µM (Schmieden et al. 1992, 1993, 1995, 1999). Thirdly, several mutations in the 1 subunit which increase the relative maximum response of taurine are accompanied by an elevation of the sensitivity of GlyR to glycine (Schmieden et al. 1999). Finally, H1 GlyRs become sensitive to GABA when their EC_50gly is decreased by the double mutation F159Y-Y161F (Schmieden et al. 1993). Thus, two other aims of our study were (i) to determine the relationships between the maximal responses to agonists (taurine or GABA) and the EC_50gly and (ii) to elucidate whether these relations are different for H1 and H2 GlyRs.

Preliminary results of this study have appeared in abstract form (De Saint Jan et al. 1999).

	METHODS

Top Abstract Introduction Methods Results Discussion Appendix References

Construction of pMT3 expression vectors for the human GlyR 1 and 2 sequences

The pBluescript SK-H1(Eco R1) and pST19(H2) vectors, provided by H. Betz (Grenningloh et al. 1990), were subcloned into the same vector (pMT3) and translational context as the Z1 subunit (David-Watine et al. 1999). The Eco R1 fragment of pBluescript SK-H1(EcoR1) was inserted into pMT3 and the 5' non-coding (NC) region of H1 was replaced by a Sma I-Bsp EI PCR fragment containing the 5' NC of Z1 and the H1 sequence up to the Bsp EI site, thus giving pMT3(Z1)H1. The human 2 sequence was amplified from pST19 (H2). The PCR product was then inserted between the Cla I and Eco R1 restriction sites of pMT3, giving pMT3(Z1)H2. PCR was carried out with the Pfu polymerase and Pfu buffer (Stratagene, France) for 25 cycles (95 °C, 1 min; 60 °C, 1 min; 72 °C, 4 min). All oligonucleotides were made and automatic sequencing was carried out by Eurogentec (Seraing, Belgium).

Extraction and injection of Xenopus oocytes

Experimental procedures and solutions used for the extraction and the injection of Xenopus oocytes were similar to those previously described (David-Watine et al. 1999; Fucile et al. 1999) and complied throughout with the guidelines of the French Animal Care Committee. Briefly, oocytes were surgically removed from adult female Xenopus laevis that had been anaesthetized by immersion in 0.03 % benzocaine (Sigma, USA) for 10 min. After the final collection the frogs were humanely killed. Oocytes were then isolated by incubation in a collagenase solution (20 mg ml^-1, 10 min, Boehringer Mannheim) and defolliculated manually using a pair of forceps in calcium-free Barth's solution (mM: 88 NaCl, 1 KCl, 0.82 MgSO₄, 0.33 Ca(NO₃)₂, 0.41 CaCl₂, 2.4 NaHCO₃ and 5 Tris-HCl; pH 7.4). Stage V-VI oocytes were injected at the centre of the animal pole (27 nl injection volume) with different concentrations of pMT3-H1 (1, 1.5, 2 and 10 pg nl^-1) or pMT3-H2 (2 and 10 pg nl^-1) to obtain various levels of expression of functional GlyR (Taleb & Betz, 1994; Fucile et al. 1999). Injected oocytes were kept at 18 °C for 3 days in Barth's solution containing antibiotics (penicillin and streptomycin, 10 µg ml^-1, Gibco, USA) and sodium pyruvate (1 mM, Gibco) and then for up to 10 days at 4 °C.

Two-electrode voltage-clamp recordings from Xenopus oocytes

Electrophysiological recordings were carried out 3-10 days after injection at room temperature (~25 °C). The oocytes were placed in a small volume chamber (~200 µl) and were continuously perfused with an external solution containing (mM): 96 NaCl, 2 KCl, 1.8 CaCl₂, 1 MgCl₂ and 5 Hepes-NaOH. Two-electrode voltage-clamp recordings were made using a GeneClamp amplifier (Axon Instruments, USA). The tips of the recording pipettes were plugged with a 3 M KCl solution containing 1 % agarose and the pipettes were then filled with 3 M KCl (Schreibmayer et al. 1994). The holding potential (V_h) was -60 or -70 mV. Drugs were dissolved in the external solution. The glycine concentration dependence of each cell was determined prior to the estimation of taurine and/or GABA dose-response relationships. The washout time between each application of agonist was 3-5 min. A saturating concentration of glycine was applied at the end of the taurine and GABA dose-response assessments. The maximum current obtained (I_maxgly) was then compared to those elicited by saturating concentrations of taurine (I_maxtau) and GABA (I_maxGABA). To avoid artefacts due to an increased osmolarity in the presence of high doses of GABA (up to 500 mM), similar concentrations of mannitol were applied. Oocytes that responded to mannitol were not included in the analysis.

Outside-out recordings from Xenopus oocytes

Patch-clamp recordings of outside-out patches from Xenopus oocytes were made at room temperature and at V_h = -20 mV after the removal of the vitelline membranes between 4 and 7 days after the injection. Glycine (1 mM), taurine (10 or 30 mM) and GABA (50 or 100 mM) were applied to the patch pipette using a fast perfusion system (SF 77A Perfusion Fast-step, Warner, USA) allowing a 10-90 % solution exchange time < 2 ms as measured by open electrode controls (1/10 NaCl). Agonists were diluted in an external solution containing (mM): 140 NaCl, 2.8 KCl, 2 CaCl₂, 2 MgCl₂ and 10 Hepes-NaOH (pH = 7.3). The patch pipettes were filled with a solution containing (mM): 140 CsCl, 2 Mg-ATP, 10 Hepes-CsOH and 10 BAPTA (pH = 7.3). Currents were recorded with an Axopatch-1A amplifier (Axon Instruments) and filtered at 2 kHz.

Single-channel recordings from HEK-293 cells

HEK-293 cells were transiently transfected with pMT3-H1 or pMT3-H2 (100 ng ml^-1) using an optimized calcium phosphate procedure. Co-transfection with Green Fluorescent Protein cDNA (GFP, 500 ng per dish) allowed identification of cells expressing GlyRs. Single-channel recordings were made in the outside-out configuration, at room temperature. External and internal solutions were the same as those used for outside-out recordings from Xenopus oocytes (with 5 mM of BAPTA in the internal solution). Single-channel currents were recorded with a EPC9 amplifier (HEKA Elektronics, Germany) and filtered at 1-3 kHz.

Data analysis

Whole-cell and single-channel data were stored and analysed on a PC computer using pCLAMP6 or pCLAMP8 software (Axon Instruments). For quantitative estimation of agonist actions, dose-response curves were fitted with the following equation:

(1)

where I is the current amplitude induced by the agonist at concentration [C], I_max is the maximum response of the cell, n_H is the Hill coefficient and EC₅₀ the concentration which induced 50 % of the maximal response.

	RESULTS

Top Abstract Introduction Methods Results Discussion Appendix References

Activation of H1 and H2 GlyRs by glycine

In this study we analysed the action of glycine, taurine and GABA on recombinant human homomeric H1 and H2 GlyRs. The EC₅₀ and Hill coefficient (n_H) of these agonists were estimated by fitting the concentration- response relationships with the Hill equation. The relative maximum responses of taurine and GABA were defined as the ratio I_maxtau/I_maxgly and I_maxGABA/I_maxgly, respectively.

It has been demonstrated that the injection of various amounts of GlyR cDNA in Xenopus oocytes results (i) in the modulation of I_maxgly and (ii) in changes of EC_50gly (Taleb & Betz, 1994; Fucile et al. 1999). We used a similar approach to analyse the properties of homomeric H1 and H2 GlyRs. It has also been shown that oocytes injected with large amounts of cDNA exhibit an agonist dose-response curve with two components (Taleb & Betz, 1994). No attempt was made here to analyse these complex behaviours and only five to nine doses of agonist were tested to estimate their EC₅₀ and I_max. We found that the sensitivity of both GlyR channels to glycine varied widely (Fig. 1A and B). The EC_50gly was between 25 and 280 µM for H1 (n = 44) and between 46 and 541 µM for H2 GlyRs (n = 52).

		View larger version [in this window] [in a new window]
Figure 1. Variability of the EC50 for glycine of human homomeric H1 and H2 GlyRs A, left: glycine-induced whole-cell currents from the most (upper traces) and the least (lower traces) sensitive oocytes expressing H1 GlyRs recorded in this study. Concentrations are indicated above each trace. A, right: concentration dependencies for the same cells fitted by the Hill equation. The lowest EC50gly of H1 was 25 µM, the highest 280 µM. B, similar presentation as in A for data recorded from oocytes expressing H2 GlyRs. The lowest and highest EC50gly were 46 and 541 µM, respectively.

The variability in sensitivity was particularly pronounced in oocytes injected with the largest amounts of cDNA (Fig. 2A). For example, following the injection of 270 pg of cDNA per oocyte, there was nearly 10-fold difference between the minimum and maximum EC_50gly for both subunits. Cells injected with the smallest amount of cDNA had statistically higher EC_50gly values. As previously reported (Taleb & Betz, 1994), maximal responses to glycine, although variable, were significantly higher for oocytes injected with the largest amounts of cDNA (Fig. 2B).

		View larger version [in this window] [in a new window]
Figure 2. Dependence of EC50gly and Imaxgly on the amount of cDNA injected into oocytes A, distribution of the EC50gly values (ordinates) for different amounts of injected cDNA (27, 40, 50 and 270 pg; abscissa). , mean EC50gly for each amount of cDNA. Note the nearly 10-fold variation in EC50gly for oocytes injected with 270 pg cDNA. B, distribution of the ratio Imaxgly/Vh (with Imaxgly in nA (200-10 000 nA) and with Vh = -60 or -70 mV) for various amounts of injected cDNA. , mean ratios.

The relationship between the maximal glycine response (relative to the holding potential i.e. I_maxgly/V_h) and the EC_50gly of each cell is shown in Fig. 3. Oocytes that gave a low response to glycine (I_maxgly/V_h < 80-90 nS) had high EC_50gly values (above 200 µM). For these cells the glycine sensitivity correlated with the I_max. For oocytes with I_maxgly/V_h > 80-90 nS, EC_50gly values were only weakly correlated with I_max values (Fig. 3).

		View larger version [in this window] [in a new window]
Figure 3. Relationship between the EC50gly for glycine and the maximum glycine response The EC50gly of each oocyte expressing either H1 GlyR () or H2 GlyR () was plotted against the maximum glycine response relative to the holding potential (i.e. Imaxgly/Vh expressed in nS).

Two main conclusions can be drawn from these results. First, a high variability of the EC₅₀ for glycine (more than 10-fold) is a common feature of human homomeric GlyRs, independent of the subunit composition. Second, the range of variation of the EC_50gly (see Fig. 1A and B) suggests that H1 are apparently more sensitive to glycine than H2 GlyRs.

Activation of human GlyR by taurine and by GABA

Cells that were responsive to glycine, expressing either H1 or H2 GlyRs, were also sensitive to taurine and GABA. To determine whether these agonists activated the same population of channels, we analysed current- voltage (I-V) relationships in outside-out patches excised from HEK-293 cells transiently transfected with plasmids encoding H1 or H2 subunits. Glycine (10 µM), taurine (25-50 µM) and GABA (1 mM) induced single-channel currents with several subconductance states, as previously reported for mammalian 1 and 2 GlyRs activated by glycine (Takahashi et al. 1992; Bormann et al. 1993). The most frequent openings, corresponding to the maximal current amplitudes, were used to determine I-V relationships. These revealed that the main conductance induced by the three agonists was similar at all recording potentials (Fig. 4A and B, right). The slope conductances, estimated at negative potentials, were smaller for H1 (85 ± 3 pS, n = 6) than for H2 (95 ± 5 pS, n = 4) subunits. On both GlyRs, the currents exhibited a small but significant rectification at positive potentials (Fig. 4A and B, right). In addition, ionic currents elicited by glycine, taurine and GABA applied to Xenopus oocytes expressing either H1 or H2 subunits were similarly antagonized by nanomolar concentrations of strychnine (not shown). These results strongly indicate that the three agonists activate the same channels.

		View larger version [in this window] [in a new window]
Figure 4. Single-channel currents evoked by glycine, taurine and GABA in outside-out patches excised from HEK 293 cells A, left: patch containing H1 GlyRs. Currents induced by the three agonists at the concentrations indicated above each trace recorded at Vh = -70 mV. Note that currents obtained with glycine, taurine and GABA had similar amplitudes. A, right: I-V relationship of single-channel currents induced on the same patch by the three agonists. Only the main conductance for each is presented. At hyperpolarizing potentials, the conductance slope was 86 pS. B, same presentation as in A but for H2 GlyRs. Typical traces (Vh = -60 mV) of agonist-induced currents are shown on the left. The I-V relationships from this patch (right) indicate that the three agonists activated a similar main conductance of 97 pS.

Analysis of the dose-response dependencies in Xenopus oocytes revealed that the actions of taurine and GABA were highly different depending on the EC_50gly. The I_max induced by glycine, taurine and GABA and the corresponding dose-response curves obtained from two typical oocytes expressing highly sensitive H1 and H2 GlyRs are illustrated in Fig. 5A and B. These two cells, with EC_50gly values of 58 µM and 97 µM, respectively, were fully activated by taurine (i.e. I_maxtau = 90 % I_maxgly for H1 and 86 % for H2). Thus, contrary to previous observations (Schmieden et al. 1992), taurine was equally efficient on both GlyRs. Moreover, millimolar concentrations of GABA induced currents of up to 70 % that of I_maxgly, indicating that this inhibitory neurotransmitter can effectively activate wild-type human homomeric GlyRs.

		View larger version [in this window] [in a new window]
Figure 5. Glycine, taurine and GABA responses of highly sensitive H1 and H2 GlyRs A and B, maximal currents induced by saturating concentrations of glycine, taurine and GABA and dose-response curves from two Xenopus oocytes expressing typical highly sensitive H1 (A) and H2 (B) subunits. EC50 values of the three tested agonists are indicated. The relative maximum responses of taurine and GABA, determined as Imaxagonist /Imaxgly from the illustrated traces, were 90 % and 69 % for the H1 GlyR and 86 % and 73 % for the H2 GlyR, respectively. Holding potentials were -70 mV.

Only oocytes expressing GlyRs with a low sensitivity to glycine displayed a pharmacological behaviour similar to that described by Schmieden et al. (1992, 1993). In such cases, taurine acted as a partial agonist and responses to GABA were very weak or even absent. This is illustrated in Fig. 6B: in an oocyte expressing H2 GlyRs with an EC_50gly = 171 µM, the values of I_maxtau and I_maxGABA were 34 % and 6 % of I_maxgly, respectively. The relative maximum response of these agonists was even lower in some cases. For example an oocyte expressing H1 GlyRs with a low sensitivity (EC_50gly = 278 µM) is shown in Fig. 6A: its I_maxtau and I_maxGABA were only 9 % and 4 % of I_maxgly, respectively. These different pharmacological behaviours were seen with both human GlyR subunits.

		View larger version [in this window] [in a new window]
Figure 6. Glycine, taurine and GABA responses of poorly sensitive oocytes expressing either H1 or H2 subunits A and B, presentation as in Fig. 5 of two oocytes expressing poorly sensitive H1 (A) or H2 (B) GlyRs. EC50 values are indicated above each dose-response curve, except for GABA (not determined). Vh = -60 mV. Imaxtau/Imaxgly = 9 % for H1 and 34 % for H2. ImaxGABA/Imaxgly = 4 % and 6 %, for H1 and H2, respectively.

To investigate further the action of taurine and GABA on human GlyRs, we analysed the relationships between the EC_50gly of these cells and their EC₅₀, Hill coefficient (n_H) or I_max values for taurine and GABA.

EC₅₀ values for taurine and GABA are closely related to EC_50gly

For oocytes expressing the H1 subunit, EC_50tau varied from 167 µM to 3.4 mM (n = 38) and EC_50GABA from 14.4 mM to ~160 mM (n = 31). In 48 oocytes expressing H2, EC_50tau ranged from 481 µM to 3.2 mM and EC_50GABA was between 64 and over 200 mM (n = 18). The EC_50tau and EC_50GABA were apparently linearly correlated with EC_50gly (Fig. 7A and B) and no striking differences in sensitivity of H1 and H2 GlyRs to taurine were observed (Fig. 7C).

		View larger version [in this window] [in a new window]
Figure 7. Relationship between the EC50 for glycine and for taurine and GABA of the H1 and H2 GlyRs A and B, linear relationships between the EC50 values for glycine and taurine (A) or GABA (B) for both H1 () and H2 () GlyRs. The data were approximated by the following equations: EC50tau = a + 10.3EC50gly, R = 0.94 (for H1) and EC50tau = a + 9.8EC50gly, R = 0.92 (for H2); EC50GABA = a + 483EC50gly, R = 0.95 (for H1) and EC50GABA = a + 765EC50gly, R = 0.64 (for H2). R, coefficient of correlation. C, relationships between the EC50 for glycine (abscissa) and for taurine and GABA (ordinate) for both H1 () and H2 (). For convenience of illustration, data are plotted on logarithmic scales.

For taurine, the slope coefficient estimated from the linear approximations indicates that the EC_50tau is about 10-fold higher than EC_50gly, regardless of the subunit composition (Fig. 7A). Therefore, this parameter cannot be used as a pharmacological criterion to distinguish H1 from H2 GlyRs.

The sensitivity to GABA was systematically lower for H2 than for H1 subunits (Fig. 7B). Equations from the linear fit, although less convincing than those for taurine, indicate that the EC_50GABA values are 480- and 765-fold higher than the EC_50gly for H1 and for H2 GlyRs, respectively (Fig. 7B).

The potency with which taurine and GABA activate the human GlyRs is related to the EC_50gly

To determine further whether taurine has different effects on H1 and on H2 GlyRs, the I_maxtau/I_maxgly ratios were plotted against EC_50gly. Figure 8A indicates that for both GlyR subunits the ratio I_maxtau /I_maxgly was closely related to EC_50gly in all oocytes tested. Regardless of the subunit composition, we found that (i) when the EC_50gly was lower than 80-90 µM, taurine acted as a full agonist (with I_maxtau reaching at least 80 % of I_maxgly), (ii) the relative maximum responses of taurine decreased as the EC_50gly increased from 80-90 µM to 250 µM, and (iii) for oocytes with an EC_50gly of over 250 µM, taurine acted as a weak partial agonist, producing currents that never exceeded 20 % of I_maxgly.

		View larger version [in this window] [in a new window]
Figure 8. Relative maximum responses of taurine and GABA depend on the EC50gly of the GlyRs The relative maximum responses of taurine (determined as Imaxtau/Imaxgly; A) and of GABA (ImaxGABA/Imaxgly; B) from oocytes expressing H1 or H2 GlyRs were plotted against the EC50gly. Note the similarities between the two GlyRs.

A similar relationship was obtained for GABA (Fig. 8B). Maximal currents induced by this amino acid reached 70-80 % of the maximal glycine currents only in oocytes expressing H1 or H2 GlyRs with an EC_50gly of below 100 µM. Furthermore, the relative maximal response of GABA decreased as the EC_50gly increased. Finally, in poorly sensitive oocytes (i.e. with EC_50gly of above 150 µM) this agonist induced either very small currents or no current, as previously reported (Schmieden et al. 1993; Taleb & Betz, 1994).

Relationships between the Hill coefficient (n_H) and the EC_50gly

The dose-response curves of glycine had Hill slopes of between 1.7 and 3 for H1 (mean 2.3 ± 0.3, n = 44) and between 1.4 and 3.5 for H2 (mean 2.45 ± 0.4, n = 52). As illustrated in Fig. 9A, no clear correlations between n_H and EC_50gly were revealed.

		View larger version [in this window] [in a new window]
Figure 9. Relationships between the EC50 for glycine and taurine and their Hill coefficient The Hill coefficient (nH) of glycine (A) and taurine (B) were plotted against the EC50 of these agonists for both H1 and H2 GlyRs. Note that only the nH for taurine decreased as the EC50 increased.

The n_H for taurine ranged from 1.2 to 3 for H1 (mean 2 ± 0.4, n = 38) and from 1.1 to 3.1 for H2 (mean 2.1 ± 0.5, n = 48). In contrast to n_Hgly, this parameter for taurine decreased when the EC_50tau increased (Fig. 9B) without any obvious difference between the two GlyR subtypes. It was only possible to obtain a good estimation of the n_HGABA on highly sensitive GlyRs for which the I_maxGABA could be determined. Thus, the relationship between n_HGABA and EC_50gly was not analysed.

Estimation of the desensitization induced by glycine, taurine and GABA on outside-out patches from Xenopus oocytes

The usual problem of currents recorded from Xenopus oocytes is that their desensitization cannot be assessed properly. It is thus possible that the partial agonist properties of taurine and GABA on highly sensitive GlyRs could not be resolved in whole oocytes. To evaluate if this parameter could alter our results, we estimated the degree of desensitization of currents induced by saturating concentrations of agonists on outside-out patches from oocytes with large I_maxgly expressing H1 or H2 GlyRs. Pulses of glycine (1 mM), taurine (10-30 mM) and GABA (50-100 mM) were applied using a fast perfusion system.

In the majority of patches containing H1 GlyRs, application of agonists induced currents with a rapid desensitization (Fig. 10A). The decay of the current following the peak was well fitted by the sum of two exponentials which displayed similar time constants for glycine (_fast = 10.6 ± 3.1 ms and _slow = 305 ± 212 ms, n = 13), taurine (_fast = 11.4 ± 3.8 ms and _slow = 222 ± 111 ms, n = 11), and GABA (_fast = 12.5 ± 8.4 ms and _slow = 245.6 ± 169 ms, n = 5). In eight patches tested with both glycine and taurine, the ratio I_tau/I_gly estimated at the peak or at the end of the pulse were similar (86.8 ± 4 % and 90.8 ± 6 %, respectively, P = 0.16, t test; Fig. 10C). These values were close to those obtained on whole oocytes with low EC_50gly. The ratio I_GABA/I_gly was statistically lower at the peak than at the steady state (57 ± 9 % and 89 ± 10 %, respectively, n = 6, P < 0.001, t test; Fig. 10C).

		View larger version [in this window] [in a new window]
Figure 10. Estimation of the desensitization from outside-out currents activated by glycine, taurine and GABA A and B, currents induced by high concentrations of glycine (1 mM), taurine (10 or 30 mM) and GABA (50 or 100 mM) applied on outside-out patches from Xenopus oocytes using a fast perfusion system. Vh = -20 mV. C, responses of taurine and GABA normalized to the Igly. For each patch, the ratio Iagonist/Igly was measured either at the peak () or at the end of the pulse (I500 ms, ). Results are given ± S.D. Asterisk indicates a statistical difference between the mean ratio at the peak and at 500 ms (P < 0.001, t test). D, degree of desensitization of the currents induced by the three agonists, given by the ratio of the amplitude at 500 ms (I500 ms) and the current at the peak (Ipeak). Results are given ± S.D. Only patches on which all three agonists could be tested were included in the analysis.

For H2 GlyRs, glycine and taurine responses exhibited a fast desensitization which was also well fitted by a double exponential with _fast = 20.3 ± 9.8 ms and _slow = 254 ± 131 ms for glycine responses (n = 6), and _fast = 15.8 ± 7.9 ms and _slow = 234 ± 153 ms for taurine (n = 4). In contrast, GABA-induced currents did not, or only weakly, desensitize (Fig. 10B). When the three agonists were tested on the same patch (n = 4) the I_tau/I_gly ratio was not significantly different when it was measured at the peak (82 ± 6 %) or at the end of the pulse (91 ± 6 %, P = 0.06, t test; Fig. 10C) and corresponded to the values obtained on highly sensitive oocytes. The ratio I_GABA/I_gly was smaller at the peak (40 ± 8 %) than at the end of the pulse (57.5 ± 15 %, n = 4) but there was no statistical differences between these values (P = 0.08, t test; Fig. 10C).

The extent of desensitization of glycine and taurine responses, estimated by the ratio of the current at the end of the 500 ms pulse to the current at the peak, was similar for both H1 GlyRs (I_{500 ms} /I_peak = 45 ± 16 % for glycine and I_{500 ms} /I_peak = 48 ± 19 % for taurine, n = 6) and H2 GlyRs (I_{500 ms}/I_peak = 60 ± 9 % for glycine and I_{500 ms} /I_peak = 66 ± 7 % for taurine, n = 4; Fig. 10D). The degree of desensitization of the GABA-induced currents was highly variable for H1 GlyRs (28.8 % < I_{500 ms} /I_peak < 100 %, mean 58 ± 23 %, n = 4) and negligible for H2 GlyRs (I_{500 ms} /I_peak = 85 ± 14 %, range 70-100 %, n = 4). No patches were obtained with low sensitive GlyRs which may display a less pronounced desensitization.

These results, obtained on presumably highly sensitive GlyRs, indicate that saturating concentrations of agonist applied on H1 or H2 GlyRs can evoke currents with a rapid desensitization which may not be seen on whole oocytes. As a consequence, it is likely that the I_max values were underestimated in our two-electrode voltage-clamp study. Figure 10C suggests that the I_maxGABA/I_maxgly ratio in particular should be subject to caution in the whole oocyte. However, these data demonstrate that desensitization does not change our main conclusions that the efficiency with which taurine and GABA activate H1 and H2 GlyRs depends on their EC_50gly and that taurine is similarly efficient as an agonist for both highly sensitive H1 and H2 GlyRs.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion Appendix References

The main goals of this study were to determine whether taurine can be used as a pharmacological descriminator between human 1 and 2 homomeric GlyRs and whether these receptors can be activated by GABA. In contrast to previous studies (Schmieden et al. 1992, 1993), we found that taurine is similarly effective for H1 and H2 GlyRs, and that these two receptors can also be activated by millimolar concentrations of GABA. We also demonstrated that the relative maximum responses of these two amino acids depend on the EC_50gly of the GlyRs which can vary from cell to cell by at least 10-fold, regardless of the subunit composition.

Actions of taurine and GABA on human 1 and 2 homomeric GlyRs

We have provided the first evidence that H2 and H1 GlyRs can be equally activated by taurine, an amino acid abundant in some parts of the developing brain, particularly the neocortex (Oja & Saransaari, 1996; Flint et al. 1998). These results are in apparent contradiction with those obtained by Schmieden et al. (1992) who reported that taurine was more potent on H1 than on H2 GlyRs. This observation can, however, be reconciled with our data by plotting the ratio I_maxtau/I_maxgly estimated in their study against the EC_50gly (see Fig. 11). The mean values obtained in their study are in fact similar to those obtained in our experiments.

		View larger version [in this window] [in a new window]
Figure 11. Comparison of the relative maximum responses of taurine on wild-type H1 GlyRs and on mutant H1 GlyRs The ratio Imaxtau /Imaxgly obtained in this study on H1 GlyRs () are compared with those observed by Lynch et al. (1997, ) on mutant and wild-type H1 GlyRs expressed in transfected HEK cells and with those from Schmieden et al. (1992, 1999; ) on wild-type and mutated H1 GlyRs expressed in Xenopus oocytes. Data from mutants I244A and W243A which exhibit an accelerated desensitization (Lynch et al. 1997) were not included in the graph.

Desensitization, which can affect the measurement of I_max in two-electrode voltage-clamped oocytes, may mask the partial agonist properties of taurine. However, using a fast perfusion system on outside-out patches, which allowed us to estimate properly the desensitization, we confirmed that taurine is a full agonist of highly sensitive H1 and H2 GlyRs.

It was also reported that H1 GlyRs are not (Schmieden et al. 1993) or only weakly (Taleb & Betz, 1994) activated by GABA, unless a double mutation F159Y,Y161F is introduced. It was therefore postulated that these residues are critical for glycine/GABA discrimination (see review by Betz et al. 1999). Our study demonstrated that wild-type human GlyRs can be efficiently activated by GABA in the absence of any mutations. Hence, we suggest that the appearance of responses to GABA following the double mutation F159Y,Y161F was due to the 10-fold decrease in the EC_50gly (from 260 µM to 22 µM in Schmieden et al. 1993) rather than to the specific elevation of the selectivity to GABA.

The agonist binding site of the GlyR

Based on the results from mutagenesis experiments, a model of an agonist binding site, composed of two subsites, has been proposed (Schmieden et al. 1992, 1993). It was suggested that the first subsite binds glycine with a high affinity and involves residues 159 and 161 and that the second, a low affinity subsite for -alanine and taurine involves residues I-111 and A-212 (Schmieden et al. 1992, 1993). As mentioned before, these amino acids were suggested to be key determinants for glycine/GABA discrimination and for the binding of taurine, respectively. Some studies have already suggested a revision of this model (Rajendra et al. 1995; Harvey et al. 2000). Furthermore, we found that the EC₅₀ values of glycine, taurine and GABA always varied in parallel (Fig. 7C) which suggests that these agonists interact with a common binding region on H1 and H2 subunits. Point mutations may alter this binding site or the gating properties of the GlyRs and non-selectively shift the EC₅₀ for all agonists. This hypothesis is supported by the fact that all the mutations known to affect the EC_50gly also induce changes in the relative maximum response of taurine. This is illustrated in Fig. 11 in which the EC_50gly and I_maxtau /I_maxgly of several mutants with amino acid substitution either in the NH₂ region (Schmieden et al. 1999), the intracellular M1-M2 loop or the extracellular M2-M3 loop (Lynch et al. 1997) are plotted. All of these values are close to those obtained in our experiments on wild-type H1 GlyRs, indicating that these mutations modify the binding or the efficacy of glycine and taurine to the same extent.

Self-inhibition of taurine

Taurine acts as a competitive antagonist of wild-type GlyRs displaying high EC_50gly values in Xenopus oocytes (Schmieden et al. 1995). To explain this property, Schmieden et al. (1995, 1999) speculated that -amino acids such as taurine, act on the GlyR in two conformations: the cis conformation mediates their agonistic action and the trans form mediates their antagonistic action. Based on this model, the authors suggested that variations in the I_maxtau /I_maxgly ratio reflect changes in the relative weight of the binding constant for inhibitors versus the binding constant for activators (Schmieden et al. 1999). Lynch et al. (1997) stated that it is unlikely that mutations in different parts of the GlyR (which all change the antagonistic action of taurine) would selectively affect a specific region involved in the binding site of -amino acids. Our results demonstrate that taurine is always a 10-fold weaker agonist than glycine and acts either as a partial or full agonist. Thus, as proposed by Rajendra et al. (1995), a more plausible explanation would be that taurine is only weakly capable of activating low affinity GlyRs, and, hence, acts as an antagonist by competing with glycine for the same binding site.

Variability of the EC_50gly of the GlyR

Our most striking result is the extreme diversity of the EC_50gly of human 1 and 2 homomeric GlyRs expressed in Xenopus oocytes. High variability of EC_50gly values may be a general property of the GlyRs conserved among different species. Indeed, a similar diversity of EC_50gly was recently observed for the zebrafish Z1 GlyR expressed either in Xenopus oocytes or in transiently transfected HEK-293 cells (Fucile et al. 1999). As illustrated in Table 1, recombinant GlyRs usually exhibit higher EC₅₀ values in Xenopus oocytes than in human cell lines. Furthermore, this analysis indicates that in addition to homomeric GlyRs, heteromeric / GlyRs (EC_50gly from 48 to 380 µM) also display a broad range of EC₅₀ values.

This variability could, at least in part, be due to some experimental limitations. Notably desensitization, which cannot be properly assessed in oocytes, may distort dose-response curves and lead to an underestimation of the EC₅₀. This factor is, however, unlikely to account entirely for the reported 10-fold variations of EC₅₀ values.

Taleb & Betz (1994) were the first to observe that the EC_50gly of GlyRs expressed in Xenopus oocytes is correlated with I_max values. To explain this correlation, they proposed that the sensitivity of the GlyR might be modulated via inter-receptor interactions occurring at high receptor density. An alternative explanation might be that the sensitivity of the GlyR to its agonists is modulated via its cytoplasmic domain by an intracellular mechanism. Indeed, we have recently demonstrated that the gating properties of neuronal or recombinant GlyRs are modulated by intracellular calcium acting through an unknown calcium-dependent cytoplasmic factor (Fucile et al. 2000). Variations in the intracellular calcium concentration or of the amount of this regulatory calcium binding protein may result in variations in EC_50gly. Further investigations are needed to unravel the mechanism(s) modulating the sensitivity of the GlyRs to its agonist in Xenopus oocytes.

Interpretation of our results using a simple kinetic model

It is tempting to use a kinetic model to interprete the variation of EC_50gly and the subsequent change of action of taurine and GABA observed in our experiments. However, although several models have been proposed for heteromeric GlyRs (Twyman & Macdonald, 1991; Legendre, 1998), there is no generally accepted model for homomeric GlyRs. Moreover, the result of any fit would be too speculative because the estimations of the EC₅₀, I_max and n_H from currents obtained on oocytes are susceptible to errors due to the invariable problem of desensitization. Thus, we only attempted to qualitatively interpret our results using the oversimplified del Castillo-Katz scheme (1957; see Appendix). This scheme, like other more realistic sequential (Twyman & Macdonald, 1991; Legendre, 1998; Lewis et al. 1998; Grewer, 1999) or allosteric models (see Galzi et al. 1996; Colquhoun, 1998), implies that the dose-response curve will have an EC₅₀ that depends on both binding (K_A) and gating (E) constants (Colquhoun, 1998).

For a first attempt at interpreting our data, we favoured a pure 'gating effect' (variation of E only, i.e. of the efficacy of the agonist to open the channel) as a plausible explanation for the observed variation of EC₅₀. Indeed, several arguments support this hypothesis. First, some startle disease mutations which are thought to alter the gating of the GlyRs (for example K276E; see Lewis et al. 1998), induce the same 'phenotype' as poorly sensitive wild-type GlyRs: they increase the EC_50gly and transform taurine from a full to a partial agonist (Lynch et al. 1997). Second, we have recently demonstrated that the EC_50gly of homomeric GlyRs is modulated by an intracellular diffusible factor which controls the gating properties of the channel (Fucile et al. 2000). Finally, the Appendix shows that the del Castillo-Katz scheme can theoretically predict most of our results (Figs 3, 7 and 8; see Fig. 12), assuming that only the gating constant E varies and with E_gly = 10E_tau and E_gly = 500E_GABA as suggested by Fig. 7.

		View larger version [in this window] [in a new window]
Figure 12. Predictions from the del Castillo-Katz scheme when the binding constant (KA) remains stable while the gating constant (E) varies A, dependence of Pmaxgly on c50gly. c50 is defined as the EC50 normalized to its binding constant and Pmax is the maximum fraction of open channel. Note the similar shape of this curve and the experimental results presented in Fig. 3. B, distribution of the c50 for taurine and GABA as a function of the c50 for glycine. Considering our result (Fig. 7), the efficacies of taurine and GABA were fixed here and in C to Etau = 0.1 Egly and EGABA = 0.002 Egly. Note the similarity with Fig. 7C. C, maximum fraction of open channel by taurine or by GABA relative to Pmaxgly decline with increasing c50gly. Note the similarities with Fig. 8.

However, this model cannot explain the different behaviours of the Hill coefficient for glycine and for taurine once the EC₅₀ increases. Indeed, any more realistic models, which assume the binding of at least two (Legendre, 1998) or three (Lewis et al. 1998) molecules of agonist to open the channel, predict that a decrease of E from a maximal E_gly between 10 and 60 (values estimated by Legendre, 1998; Lewis et al. 1998; Grewer, 1999) would be accompanied by a similar reduction of the Hill slope for both glycine and taurine (with E_gly = 10E_tau). This analysis thus indicates that a pure gating effect cannot account for all of our observations and suggests that another mechanism may be needed to induce the variation of EC_50gly.

	APPENDIX

Top Abstract Introduction Methods Results Discussion Appendix References

The general problem of affinity/efficacy relationships of receptor channels and the different models used to solve these questions were recently analysed by Colquhoun (1998). Here, we use the del Castillo-Katz scheme (1957) to clarify whether the general features of our results can be described by this simplest model, although it is clearly too oversimplified to describe the activation of the GlyR by the agonists. The scheme is

(2)

where A is the agonist, R the receptor, AR the inactive (closed) agonist-receptor complex and AR* the active (open) form of this complex. The binding constant is denoted by K_A and the gating constant by E. This mechanism implies that the fraction of channels open at equilibrium will be:

(3)

and thus predicts that the dose-response curve will have an EC₅₀ given by:

EC50 = KA/(1 + E).

(4)

If we define c_A as the agonist concentration normalized to its binding constant:

cA = [A]/KA,

(5)

the fraction of channels open at equilibrium will be:

(6)

and at high concentration of agonist, the maximum fraction of open channels is:

(7)

If we assume that K_A, although different for glycine, taurine and GABA, has similar values in every cell, then from eqn (4) the variation of EC₅₀ will only be due to cell-cell variability in E. We can now define c₅₀ as the EC₅₀ normalized with respect to its binding constant; from eqn (4) this is:

(8)

If we suggest that the binding constants (K_gly, K_tau and K_GABA) are stable and call the efficacies of the agonists E_gly, E_tau and E_GABA, the slope of the relationships between their EC₅₀ values (Fig. 7A and B) suggest that

Etau = 0.1Egly, (9)

(9)

and

EGABA = 0.002Egly.

(10)

From these simple equations in which E is the only variable, the general shape of the curve in Fig. 3 can be predicted by plotting, for a range of values of E_gly, P_max(E_gly) against c₅₀(E_gly). This is shown in Fig. 12A. Likewise the general form of the results in Fig. 7C can be predicted by plotting c₅₀(E_GABA) and c₅₀(E_tau) against c₅₀(E_gly) (Fig. 12B). Finally, the results in Fig. 8 can be mimicked by plotting P_max(E_GABA)/P_max(E_gly) and P_max(E_tau)/P_max(E_gly) against c₅₀(E_gly), as in Fig. 12C.

The del Castillo-Katz scheme is obviously oversimplified. It cannot explain, for instance, the behaviour of the Hill coefficient observed in our study. A more appropriate mechanism for homomeric GlyRs should involve five agonist molecules and several open states (Lewis et al. 1998; Fucile et al. 1999). However, the qualitative features of the curves shown in Fig. 12 should be similar for more complex, and physically realistic mechanisms.

	REFERENCES

Top Abstract Introduction Methods Results Discussion Appendix References

ALI D. W., DRAPEAU, P. & LEGENDRE, P. (2000). Development of spontaneous glycinergic currents in the Mauthner neuron of the zebrafish embryo. Journal of Neurophysiology 84, 1726-1736	[Abstract/Full Text]
BETZ H., KUHSE, J., SCHMIEDEN, V., LAUBE, B., KIRSCH, J. & HARVEY, R. J. (1999). Structure and functions of inhibitory and excitatory glycine receptors. Annals of the New York Academy of Sciences 30, 667-676
BORMANN J., RUNDSTRÖM, N., BETZ, H. & LANGOSCH, D. (1993). Residues within transmembrane segment M2 determine chloride conductance of glycine receptor homo- and hetero-oligomers. EMBO Journal 12, 3729-3737	[Abstract]
BREITINGER H. G. & BECKER, C. M. (1998). The inhibitory glycine receptor: prospects for a therapeutic orphan? Current Pharmaceutical Design 4, 315-334	[Medline]
DAVID-WATINE B., GOBLET, C., DE SAINT JAN, D., FUCILE, S., DEVIGNOT, V., BREGESTOVSKI, P. & KORN, H. (1999). Cloning, expression and electrophysiological characterization of glycine receptor alpha subunit from zebrafish. Neuroscience 90, 303-317	[Medline]
DEL CASTILLO J. & KATZ, B. (1957). Interaction at end-plate receptors between different choline derivatives. Proceedings of the Royal Society B 146, 369-381
DE SAINT JAN D., DAVID-WATINE, B., KORN, H. & BREGESTOVSKI, P. (1999). Activation of human homomeric 1 and 2 glycine receptors by taurine and GABA. Society for Neuroscience Abstracts 180, 14
FLINT A. C., LIU, X. & KRIEGSTEIN, A. R. (1998). Nonsynaptic glycine receptor activation during early neocortical development. Neuron 20, 43-53	[Medline]
FUCILE S., DE SAINT JAN, D., DAVID-WATINE, B., KORN, H. & BREGESTOVSKI, P. (1999). Comparison of glycine and GABA actions on the zebrafish homomeric glycine receptor. Journal of Physiology 517, 369-383	[Abstract/Full Text]
FUCILE S., DE SAINT JAN, D., PRADO DE CARVALHO, L. & BREGESTOVSKI, P. (2000). Fast potentiation of glycine receptor channels by intracellular calcium in neurons and transfected cells. Neuron 28, 571-583	[Medline]
GALZI J. L., EDELSTEIN, S. J. & CHANGEUX, J. P. (1996). The multiple phenotypes of allosteric receptor mutants. Proceedings of the National Academy of Sciences of the USA 93, 1853-1858	[Abstract]
GRENNINGLOH G., SCHMIEDEN, V., SCHOFIELD, P. R., SEEBURG. P. H., SIDDIQUE, T., MOHANDAS, T. K., BECKER, C. M. & BETZ, H. (1990). Alpha subunit variants of the human glycine receptor: primary structures, functional expression and chromosomal localization of the corresponding genes. EMBO Journal 9, 771-776	[Abstract]
GREWER C. (1999). Investigation of the alpha(1)-glycine receptor channel-opening kinetics in the submillisecond time domain. Biophysical Journal 77, 727-738	[Abstract/Full Text]
HANDFORD C. A., LYNCH, J. W., BAKER, E. G., WEBB, C., FORD, J. H., SUTHERLAND, G. R. & SCHOFIELD, P. R. (1996). The human glycine receptor subunit: primary structure, functional characterization and chromosomal localization of the human and murine genes. Molecular Brain Research 35, 211-219	[Medline]
HARVEY R. J., SCHMIEDEN, V., VON HOLST, A., LAUBE, B., ROHRER, H. & BETZ, H. (2000). Glycine receptors containing the alpha4 subunit in the embryonic sympathetic nervous system, spinal cord and male genital ridge. European Journal of Neuroscience 12, 994-1001	[Medline]
HOSIE A. M., AKAGI, H., ISHIDA, M. & SHINOZAKI, H. (1999). Actions of 3-[2'-phosphonomethyl [1,1'-biphenyl]-3-yl] alanine (PMBA) on cloned glycine receptors. British Journal of Pharmacology 126, 1230-1236	[Abstract/Full Text]
IMBODEN M., DE SAINT JAN, D., LEULIER, F., KORN, H., GOBLET, C. & BREGESTOVSKI, P. (2001). Isolation and characterization of a alpha 2-type zebrafish glycine receptor subunit. Neuroscience 103, 799-810	[Medline]
LANGOSCH D., LAUBE, B., RUNDSTRÖM, N., SCHMIEDEN, V., BORMANN, J. & BETZ, H. (1994). Decreased agonist affinity and chloride conductance of mutant glycine receptors associated with human hereditary hyperekplexia. EMBO Journal 13, 4223-4228	[Abstract]
LEGENDRE P. (1998). A reluctant gating mode of glycine receptor channels determines the time course of inhibitory miniature synaptic events in zebrafish hindbrain neurons. Journal of Neuroscience 18, 2856-2870	[Abstract/Full Text]
LEWIS T. M., SIVILOTTI, L. G., COLQUHOUN, D., GARDINER, R. M., SCHOEPFER, R. & REES, M. (1998). Properties of human glycine receptors containing the hyperekplexia mutation alpha1(K276E), expressed in Xenopus oocytes. Journal of Physiology 507, 25-40	[Abstract/Full Text]
LYNCH J. W., RAJENDRA, S., PIERCE, K. D., HANDFORD, C. A., BARRY, P. H. & SCHOFIELD, P. R. (1997). Identification of intracellular and extracellular domains mediating signal tranduction in the inhibitory glycine receptor channel. EMBO Journal 16, 110-120	[Abstract/Full Text]
MASCIA M. P., WICK, M. J., MARTINEZ, L. D. & HARRIS, R. A. (1998). Enhancement of glycine receptor function by ethanol: role of phosphorylation. British Journal of Pharmacology 125, 263-270	[Abstract]
MOORHOUSE A. J., JACQUES, P., BARRY, P. H. & SCHOFIELD, P. R. (1999). The startle disease mutation Q266H, in the second transmembrane domain of the human glycine receptor, impairs channel gating. Molecular Pharmacology 55, 386-395	[Abstract/Full Text]
OJA S. S. & SARANSAARI, P. (1996). Taurine as osmoregulator and neuromodulator in the brain. Metabolic Brain Disease 11, 153-164	[Medline]
PRIBILLA I., TAKAGI, T., LANGOSCH, D., BORMANN, J. & BETZ, H. (1992). The atypical M2 segment of the subunit confers picrotoxinin resistance to inhibitory glycine receptor channels. EMBO Journal 11, 4305-4311	[Abstract]
RAJENDRA S., LYNCH, J. W., PIERCE, K. D., FRENCH, C. R., BARRY, P. H. & SCHOFIELD, P. R. (1995). Mutation of an arginine residue in the human glycine receptor transforms -alanine and taurine from agonists into competitive antagonists. Neuron 14, 169-175	[Medline]
RAJENDRA S., LYNCH, J. W. & SCHOFIELD, P. R. (1997). The glycine receptor. Pharmacology and Therapeutics 73, 121-146	[Medline]
SCHMIEDEN V. & BETZ, H. (1995). Pharmacology of the inhibitory glycine receptor: agonist and antagonist action of amino acids and piperidine carboxylic acid compounds. Molecular Pharmacology 48, 919-927	[Abstract]
SCHMIEDEN V., KUHSE, J. & BETZ, H. (1992). Agonist pharmacology of neonatal and adult glycine receptor subunits: identification of amino acid residues involved in taurine activation. EMBO Journal. 11, 2025-2032	[Abstract]
SCHMIEDEN V., KUHSE, J. & BETZ, H. (1993). Mutation of glycine receptor subunit creates beta-alanine receptor responsive to GABA. Science 262, 256-258
SCHMIEDEN V., KUHSE, J. & BETZ, H. (1999). A novel domain of the inhibitory glycine receptor determining antagonist efficacies: further evidence for partial agonism resulting from self-inhibition. Molecular Pharmacology 56, 464-472	[Abstract/Full Text]
SCHREIBMAYER W., LESTER, H. A. & DASCAL, N. (1994). Voltage clamping of Xenopus laevis oocytes utilizing agarose-cushion electrodes. Pflügers Archiv 426, 453-458	[Medline]
SINGER J. H., TALLEY, E. M., BAYLISS, A. D. & BERGER, A. J. (1998). Development of glycinergic synaptic transmission to rat brain stem motoneurons. Journal of Neurophysiology 80, 2608-2620	[Abstract/Full Text]
TAKAHASHI T., MOMIYAMA, A., HIRAI, K., HISHINUMA, F. & AKAGI, H. (1992). Functional correlation of fetal and adult forms of glycine receptors with developmental changes in inhibitory synaptic receptor channels. Neuron 9, 1155-1161	[Medline]
TALEB O. & BETZ, H. (1994). Expression of the human glycine receptor alpha 1 subunit in Xenopus oocytes: apparent affinities of agonists increase at high receptor density. EMBO Journal 13, 1318-1324	[Abstract]
TWYMAN R. E. & MACDONALD, R. L. (1991). Kinetic properties of the glycine receptor main and sub-conductance states of mouse spinal cord neurones in culture. Journal of Physiology 435, 303-331	[Abstract]
WERMAN R. (1972). Amino acids as central neurotransmitters. Association for Research and Nervous Mental Disorder 50, 147-180
WICK M. J., BLECK, V., WHATLEY, V. J., BROZOWSKI, S. J., NIXON, K., CARDOSO, R. A. & VALENZUELA, C. F. (1999). Stable expression of human glycine 1 and 2 homomeric receptors in mouse L(tk-) cells. Journal of Neuroscience Methods 87, 97-103	[Medline]

Acknowledgements

We are grateful to Professor D. Colquhoun for critical and creative analysis of the manuscript and to D. Colquhoun and Dr L. Sivilotti for providing the material on which the Appendix is based. We also wish to thank Professor H. Betz for the kind gift of the pBluescriptSK-H1(EcoR1) and pST19(H2) vectors, Professor P. Ascher and Drs A. Devillers-Thiéry, L. Marubio, L. Prado de Carvalho and A. Triller for critical reading of manuscript, valuable suggestions and helpful discussions, and P. Caramelle for help with injection of oocytes. D.D.S.J. is supported by a fellowship from the Ministère de la Recherche and by the Fondation Pour La Recherche Médicale.

Corresponding author

P. Bregestovski: INSERM U 261, Institut Pasteur, 25, rue du Docteur Roux, 75724 Paris cedex 15, France.

Email: pbreges{at}pasteur.fr

This article has been cited by other articles:

				F. Wang, C. Xiao, and J. H. Ye Taurine activates excitatory non-synaptic glycine receptors on dopamine neurones in ventral tegmental area of young rats J. Physiol., June 1, 2005; 565(2): 503 - 516. [Abstract] [Full Text] [PDF]

				J. W. Lynch Molecular Structure and Function of the Glycine Receptor Chloride Channel Physiol Rev, October 1, 2004; 84(4): 1051 - 1095. [Abstract] [Full Text] [PDF]

				P. Castaldo, P. Stefanoni, F. Miceli, G. Coppola, E. M. del Giudice, G. Bellini, A. Pascotto, J. R. Trudell, N. L. Harrison, L. Annunziato, et al. A Novel Hyperekplexia-causing Mutation in the Pre-transmembrane Segment 1 of the Human Glycine Receptor {alpha}1 Subunit Reduces Membrane Expression and Impairs Gating by Agonists J. Biol. Chem., June 11, 2004; 279(24): 25598 - 25604. [Abstract] [Full Text] [PDF]

Z. Jiang, K. Krnjevic, F. Wang, and J. H. Ye Taurine Activates Strychnine-Sensitive Glycine Receptors in Neurons Freshly Isolated From Nucleus Accumbens of Young Rats J Neurophysiol, January 1, 2004; 91(1): 248 - 257.