Ethanol Suppresses Fast Potentiation of Glycine Currents by Glutamate

Materials and Methods

Isolation of Neurons and Electrophysiological Recording.

The care and use of animals and the experimental protocol of this study were approved by the Institutional Animal Care and Use Committee of the University of Medicine and Dentistry of New Jersey (protocol number 00074). Sprague-Dawley rats (5- to 14-day-old) were decapitated as described earlier (Ye et al., 2001a). The brains were quickly excised, placed into ice-cold saline saturated with 95% O₂ and 5% CO₂, glued to the chilled stage of a Vibratome (Campden Instruments Ltd., Loughborough, Leicestershire, UK), and sliced to a thickness of 300 to 400 μm. Slices were transferred to the standard external solution containing 1 mg of pronase/6 ml and saturated with O₂ and incubated at 31°C for 20 min. After 20 min of additional incubation in 1 mg of thermolysin/6 ml, the VTA was identified medial to the accessory optic tract and lateral to the fasciculus retroflexus under a dissecting microscope. Micro-punches of the VTA were isolated and transferred to a 35-mm culture dish. Mild trituration through heat-polished pipettes of progressively smaller tip diameters dissociated single neurons. Within 20 min of trituration, isolated neurons attached to the bottom of the culture dish and were ready for electrophysiological experiments.

Under the light microscope, the cells acutely isolated from the VTA were of two types: bipolar and multipolar. The majority was bipolar, with one to three dendritic processes emerging from each end of the fusiform soma (20–40 μm in length and 15–25 μm in diameter). The multipolar neurons were larger, with a diameter of 35 to 60 μm, and had four to five major dendrites. Most of the cells were tyrosine hydroxylase-positive and therefore presumed to be dopaminergic. This is in good agreement with the recent report of Brodie et al. (1999). There were no appreciable differences in the response of these two groups of neurons to ethanol.

The saline in which the brain was dissected contained 128 mM NaCl, 5 mM KCl, 1.2 mM NaH₂PO₄, 26 mM NaHCO₃, 9 mM MgCl₂, 0.3 mM CaCl₂, and 2.5 mM glucose. The standard external solution contained 140 mM NaCl, 5 mM KCl, 1 mM MgCl₂, 2 mM CaCl₂, 10 mM glucose, and 10 mM HEPES. The pH was adjusted to 7.4 with Tris base and the osmolarity to 320 mM with sucrose. The patch pipette solution contained 120 mM CsCl, 21 mM tetraethylammonium chloride, 4 mM MgCl₂, 11 or 4 mM EGTA, 1 mM CaCl₂, 10 mM HEPES, and 2 mM Mg-ATP. The pH was adjusted to 7.2 with Tris base and the osmolarity to 280 mM with sucrose, or otherwise as indicated. In several experiments, instead of EGTA and CaCl₂, the pipette solution contained 1,2-bis(2-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid (BAPTA) (10 or 30 mM, as indicated). When filled with the above solutions, the patch electrodes had a resistance of 3 to 5 MΩ. Throughout the experiment, the bath was perfused with the standard external solution, at an ambient temperature of 20–23°C.

Whole-cell currents were recorded under voltage clamp with an Axopatch 200 B amplifier (Axon Instruments, Foster City, CA) interfaced to a Digidata 1320A data acquisition system (Axon Instruments) and directly digitized with pCLAMP 8 software for further off-line analysis. The junction potential between the patch pipette and the bath solutions was nulled just before forming the giga-seal. The liquid junction potential between the bath and the electrode was 3.3 mV, as calculated from the generalized Henderson equation using the Axoscope junction potential calculator (Barry, 1996). This value was corrected off-line when estimating the reversal potential ofI_Gly. In most experiments, the series resistance before compensation was 15 to 25 MΩ. Routinely, 80% of the series resistance was compensated; hence, there was a 3-mV error for 1 nA of current.

Chemical Applications.

Solutions of agonist, antagonists, and ethanol were prepared on the day of experimentation. Glycine, strychnine, glutamate, BAPTA, and EGTA were obtained from Sigma-Aldrich (St Louis, MO), and 1-[N,O-bis-(5-isoquinolinesulfonyl)-N-methyl-l-tyrosyl]-4-phenylpiperazine (KN-62) was obtained from Calbiochem (San Diego, CA). Ethanol (95%, prepared from grain), obtained from Pharmco Products Inc. (Brookfield, CT), was stored in glass bottles. Solutions were applied via a multibarreled pipette (as described previously: Ye et al., 2001a), the tip of which was usually placed 50 to 100 μm from a dissociated cell. While maintaining recording stability, this system allows complete exchange of solutions in its vicinity within 10 ms. A conditioning pulse of glutamate (1 mM for 1–3 s) was immediately followed by a brief pulse of glycine (30 μM, unless otherwise indicated). In some experiments, the extracellular Ca²⁺ concentration ([Ca²⁺]_o) was raised to 12 mM locally by superfusing the cell body only with a solution containing 12 mM Ca²⁺ (only these barrels and their respective reservoir syringes contained 12 mM Ca²⁺; the other barrels and syringes contained 2 mM Ca²⁺).

Data Analyses.

Whole-cell current decays were fitted by a Chebychev algorithm (pCLAMP). Concentration-response data were analyzed with a nonlinear curve-fitting program (Sigma Plot; Jandel Scientific, San Rafael, CA). Data were statistically compared using Student's t test at a significance level ofP < 0.05, unless otherwise indicated. For all experiments, average values are expressed as mean ± S.E.M., with the number of neurons indicated in brackets.

Results

Glutamate Potentiates I_Gly.

In agreement with our previous observations, most neonatal VTA neurons (82%) were sensitive to glycine. Glycine-induced current (I_Gly) was antagonized by 0.1 μM strychnine (Ye et al., 1998). Glutamate (0.1 and 1 mM) elicited inward currents in all VTA neurons tested. At a holding potential of –50 mV, larger peak currents were induced by 30 μM glycine (−520 ± 68 pA, n = 42) than by 1 mM glutamate (−338 ± 44 pA, n = 41). To examine the effect of glutamate onI_Gly, a pulse of glycine (10–1000 μM) was preceded by a brief conditioning pulse of glutamate (1–3 s). For this purpose, 1 mM glutamate was routinely used, because of its very predictable action and its previous use in comparable experiments (Fucile et al., 2000). However, substantial potentiation ofI_Gly could be obtained with 100 μM glutamate (see below).

The traces in Fig. 1A illustrate the effect of a glutamate prepulse (1 mM) on an immediately followingI_Gly, recorded from a neuron in standard external solution (2 mM Ca²⁺) and with 11 mM EGTA in the pipette. Because the glutamate-induced current returned to baseline immediately upon washout of glutamate, the potentiation of I_Gly cannot be the result of simple addition of the two currents, as illustrated in Fig.1B (and also Fig. 3A; see below). BecauseI_Gly rapidly returned to the control level after glutamate washout, the duration of glutamate's effect could not be measured accurately. Glutamate (1 mM) potentiatedI_Gly (induced by 30 μM glycine) in 85 of 131 (65%) of VTA neurons tested. The potentiation varied from 10 to >150% (Fig. 1C). I_Gly was increased from a control mean of 487 ± 50 to 643 ± 62 pA; that is, to 153 ± 8% of control (P < 0.001,n = 72, median 122%). To determine whether this was a true potentiation, we examined the reversal potential ofI_Gly. Plots of the current-voltage relationships of I_Gly were not changed after the glutamate pretreatment (Fig. 1, D–G), the reversal potential remaining close to the [Cl⁻] Nernst potential of 0 mV calculated for our solutions. This is in agreement with the previous report by Xu et al. (1999).

Involvement of Ca²⁺ in the Glutamate-Induced Potentiation of I_Gly.

According to recent reports, Ca²⁺ exerts a powerful and rapid modulation of glycine receptor/channels (Xu et al., 1999, 2000; Fucile et al., 2000). The following results suggest that Ca²⁺ also plays a role inI_Gly potentiation in VTA neurons.

Potentiation Was Greater when Extracellular Ca²⁺ Was Increased to 12 mM.

As shown in Fig.2, when [Ca²⁺]_o was raised locally to 12 mM, I_Glu increased by 65 ± 4% (P < 0.01, n = 4) and the magnitude of I_Gly potentiation by glutamate by 55 ± 10% of control tests of glutamate in 2 mM [Ca²⁺]_o (Fig. 2B,P < 0.01, n = 4). This effect of [Ca²⁺]_o was reversible: both I_Glu and the potentiation ofI_Gly returned to control values when 2 mM [Ca²⁺]_o was restored.

Reducing the Internal Concentration of EGTA Enhanced Both the Magnitude and Duration of Glutamate's Effect.

In recordings with pipettes containing 4 mM EGTA (instead of the usual 11 mM), the potentiation of peak I_Gly induced by a 3-s glutamate prepulse reached 405 ± 128% (n = 14; median 164%) of control. Note the unusually large effects of glutamate in traces A of Fig. 3, obtained with an electrode containing 4 mM EGTA; the potentiation persisted for more than 8 s (Fig. 3B). WhenI_Gly was recorded with such pipettes, even 100 μM glutamate induced a marked potentiation (146 ± 9%,n = 4; Fig. 3, C and D).

There Was Less Potentiation when Recording with Electrodes Containing 10 to 30 mM BAPTA.

In these recordings, the potentiation induced by glutamate decreased with time: 4 min after the start of whole-cell recording, I_Glywas potentiated to 153 ± 8% of control, but by 16 min, to only 126 ± 4% of control (paired t test, P= 0.029, n = 4; cf. traces in Fig. 3E and histograms in Fig. 3F). Presumably, this reflects the time required for BAPTA to equilibrate at the intracellular site of Ca²⁺action.

Although the glutamate-induced potentiation ofI_Gly thus seems to depend on an increase in intracellular free Ca²⁺, it was not affected by pretreating cells for 8 min with 5 μM KN-62, a selective calcium/calmodulin-dependent protein kinase II inhibitor: the large potentiations observed in the same five cells before and after applying KN-62 were to 372 ± 18 and 387 ± 21% of control, respectively (P = 0.67; Fig. 2, C and D).

Glutamate-Induced Potentiation Is Sensitive to Glycine Concentration.

Glutamate could augmentI_Gly either by increasing the number or conductance of functional glycine receptor channels or by modifying their sensitivity to glycine. To distinguish between these possibilities, we examined the effect of glutamate onI_Gly induced by 10 to 1000 μM glycine. The traces in Fig. 4 showI_Gly evoked by 30, 100, and 300 μM glycine, in the absence (Fig. 4A) and presence (Fig. 4B) of prepulses of glutamate (1 mM). Glutamate strongly potentiatedI_Gly induced by submaximal concentrations of glycine (30 and 100 μM, Fig. 4, a and b); but it had a weaker effect on I_Gly induced by supramaximal concentrations of glycine (≥300 μM; Fig. 4C). On average, 1 mM glutamate potentiated peakI_Gly elicited by 30, 100, 300, and 1000 μM glycine to 180 ± 21, 152 ± 22, 118 ± 22, and 121 ± 20%, respectively (n = 4, Fig. 4D)

Concentration-response data obtained in the absence and presence of conditioning pulses of glutamate (1 mM) are illustrated in Fig. 4C. The EC₅₀ and Hill coefficient were 87 μM and 1.96 in the absence of glutamate, and 47 μM and 1.83 after glutamate conditioning pulses. Thus glutamate reduced the EC₅₀ by nearly 50%. Similar observations on three other cells gave a mean reduction to 64 ± 3% (n = 4). The conditioning pulses of glutamate thus increased the apparent affinity of the glycine receptor for its agonist. In addition, maximal I_Gly was also larger after glutamate (121 ± 8%, n = 4), as found by Xu et al. (1999).

Glutamate Alters the Kinetics ofI_Gly.

Changes in either agonist affinity or channel opening efficacy can alter the EC₅₀ values of agonists (Colquhoun, 1998). Indeed, data from human embryonic kidney-AMPA cells transfected with αH1 demonstrate different kinetics ofI_Gly before and after a glutamate prepulse (Fucile et al., 2000). Therefore, we examinedI_Gly channel activation, deactivation, and desensitization, before and after glutamate conditioning pulses. To allow accurate measurement within the limits of the fast perfusion system (time constant of ∼10 ms), we applied glycine at a concentration of 30 μM (Fig. 5A). As previously observed (Ye et al., 2001a), both the onset and the decay ofI_Gly could be fitted by a single exponential function (Fig. 5, C and D). The activation time constant (τ_on) was significantly shortened by a glutamate prepulse, from 340 ± 14 ms to 183 ± 22 ms (pairedt test, P < 0.01, n = 8). In contrast, glutamate prolonged the deactivation time constant (τ_off), from 261 ± 19 ms to 350 ± 31 ms (Fig. 5B; paired t test, P < 0.01,n = 8). The slower decay indicates that glutamate increases the affinity of glycine for its receptor (Fucile et al., 2000).

Glutamate Accelerates Glycine Receptor Desensitization.

The potentiation of I_Gly by glutamate could result from a slower rate of receptor desensitization. To test for this possibility, we compared I_Glydesensitization in the absence and presence of glutamate prepulses. As shown in Fig. 5E, the current activated by a long pulse of glycine (30 μM) decayed more rapidly when applied after a brief pulse of glutamate. The ratio of the decay time constants (τ_Glu/τ_control) in Fig.5E was 0.56. For six neurons (Fig. 5F), 1 mM glutamate significantly shortened the time constant of desensitization from 6.9 ± 1.4 to 4.7 ± 0.8 s (paired t test, P < 0.05). Because glutamate enhanced the peak more than the steady-stateI_Gly, the ratio of steady-state to peak current amplitude declined from 0.92 ± 0.02 to 0.71 ± 0.04 (paired t test, P < 0.01,n = 17).

Ethanol Potentiates I_Gly But Inhibits Glutamate Current.

In agreement with our previous findings (Ye et al., 2001a), 0.1 to 100 mM ethanol enhancedI_Gly in 35% of VTA neurons from 5- to 14-day-old rats. This effect is illustrated in Fig.6A, whereI_Gly evoked by 30 μM glycine was potentiated by 0.1, 1, and 10 mM ethanol (Fig. 6, A-b–A-d). After washout of ethanol, I_Gly recovered to control amplitude (Fig. 6, A-e). For a series of neurons, 0.1, 1, 10, and 100 mM ethanol enhanced peak I_Glyto 116 ± 5% (n = 3), 135 ± 4% (n = 34), 127 ± 3% (n = 34), and 117 ± 6% (n = 4) of control, respectively. When a brief pulse of 10 mM ethanol was coapplied during a longer pulse of glycine, there was an immediate and rapidly reversible increase inI_Gly (Fig. 6B). In contrast to this potentiation of I_Gly, 10 mM ethanol depressed glutamate-evoked current to 74 ± 3% (n= 10) of control (Fig. 6C), in agreement with previous reports (Narahashi et al., 2001).

Ethanol Suppresses the Glutamate-Induced Potentiation ofI_Gly.

Records a and b of Fig. 6D illustrate the usual potentiation ofI_Gly by a conditioning prepulse of glutamate. When 10 mM ethanol was coapplied with glutamate and glycine (Fig. 6D-d), the potentiation of I_Glyby glutamate was significantly reduced: from 188 ± 72% of control in the absence of ethanol to 146 ± 26% in its presence (Fig. 6E; P < 0.05, n = 4).

In the experiments illustrated in Fig. 6D, ethanol was present when glutamate was applied. Therefore, the suppression of glutamate-induced potentiation of I_Gly could result from ethanol-induced reduction of Ca²⁺ entry via glutamate receptors. To assess this possibility, ethanol was coapplied with glycine immediately after the end of the glutamate prepulse (Fig.7A). Although either glutamate or ethanol alone increased I_Gly (Fig. 7, A-b and A-c), there was no further enhancement ofI_Gly by ethanol (Fig. 7A-d). The data from 13 cells studied with this protocol are summarized in Fig. 7B: there was a similar potentiation ofI_Gly by conditioning pulses of glutamate (as in Fig. 7A-b; 146 ± 12%), ethanol alone (Fig. 7A-c; 145 ± 11%), and ethanol applied after the glutamate conditioning pulse (Fig. 7A-d; 148 ± 16%). These findings suggest different mechanisms of potentiation by ethanol and by glutamate. As shown in Fig. 7A-b, glutamate enhanced mainly peakI_Gly and its rate of decay, whereas ethanol potentiated both peak and steady-stateI_Gly (Fig. 7A-c). When ethanol was coapplied with glycine after a conditioning pulse of glutamate, both peak and steady-state I_Gly were enhanced, as during ethanol treatment alone (Fig. 7A-d).

It is unlikely that maximal activation of glycine receptors explains these observations because the outcome was the same when ethanol was applied at a lower concentration (1 mM) (data not shown). Moreover, ethanol suppressed the potentiation by glutamate even in cells that showed no enhancement ofI_Gly by ethanol alone. This is illustrated in Fig. 7C, where 10 mM ethanol did not alterI_Gly (Fig. 7C-c) but suppressed its enhancement by glutamate (cf. Fig. 7, A-b and A-d). The results obtained from five cells are summarized in Fig. 7D, in whichI_Gly was not altered by ethanol alone (I_Gly was 103 ± 1% of control), although after glutamate,I_Gly was enhanced to 124 ± 4%. When glycine + ethanol were applied after the glutamate prepulse,I_Gly was 105 ± 5% of control, although peak glutamate currents were the same for the first and second pulse (274 ± 91 and 271 ± 97 pA; P > 0.1,n = 5). Thus, even when ethanol did not change the response to glycine, it still prevented the interaction between glutamate and glycine.

Discussion

Our previous research revealed glycine receptors in the majority of freshly dissociated VTA neurons (Ye et al., 1998), all of which also respond to glutamate (Wang and French, 1993; Wu and Johnson, 1996; Ye et al., 2001b). The present results show that in such VTA neurons, glutamate consistently enhances I_Gly, as previously observed in spinal neurons by Xu et al. (1999, 2000), and spinal and transfected cells by Fucile et al. (2000). Our principal new finding is that ethanol suppresses the glutamate-induced potentiation of I_Gly in VTA neurons. In keeping with the previous authors, our results point to the involvement of Ca²⁺, perhaps Ca²⁺ influx, in this phenomenon, although probably not calcium/calmodulin-dependent protein kinase II.

Comparison with Previous Reports of Interactions between Glutamate and Glycine.

The previous studies of glutamate-induced fast potentiation of I_Gly (Xu et al., 1999,2000; Fucile et al., 2000) attributed this effect to a rise in intracellular free Ca²⁺. Although outwardly similar, these reports differed in some important respects. The results of the perforated-patch recordings from freshly dissociated spinal neurons (Xu et al., 1999, 2000) led to the conclusion that the increase in I_Gly is not caused by a change in the affinity of glycine for its receptor and that the potentiation is mediated by activation of CAMKII. In contrast, conventional whole-cell recordings from transfected cells (Fucile et al., 2000) suggested potentiation was due to a higher affinity of glycine receptors and not mediated by known protein kinases. In our experiments, an increase in glycine receptor affinity was indicated by the slower deactivation ofI_Gly and the consistent reduction in EC₅₀. However, the glycine concentration-response plots also showed a clear increase in the maximalI_Gly. In VTA neurons, the potentiation thus appeared to be mediated by both mechanisms.

Is the Potentiating Action of Glutamate in VTA Neurons Mediated by Intracellular Ca²⁺?

Our findings that glutamate was more effective when [Ca²⁺]_o was raised to 12 mM and less effective when a stronger Ca²⁺ buffer, BAPTA or 11 mM EGTA (as compared with 4 mM EGTA) was present in the electrodes are consistent with some involvement of Ca²⁺. However, the relatively modest effects produced by these manipulations, especially when compared with those seen in the experiments of Xu et al. (1999,2000) and Fucile et al. (2000), seem more in keeping with a Ca²⁺-sensitive than a Ca²⁺-dependent process. Admittedly, by buffering slow changes in [Ca²⁺]_i, the routine presence of EGTA in the electrodes would tend to reduce Ca²⁺-mediated mechanisms. This may, at least in part, account for the relatively small and transient effects of glutamate observed in the current study as compared with the potentiation observed in the previous studies, where weaker or no buffers were used. Since even 30 mM BAPTA failed to abolish the action of glutamate, the potentiation of glycine receptors may occur at a site that is not entirely intracellular, or at least not easily accessible to intracellular chelators. Judging by the lack of effect of KN-62, CAMKII is probably not the agent of Ca²⁺-mediated modulation. Whether phosphorylation or dephosphorylation plays a significant role should be clarified by further tests of selective blockers.

Mechanisms of Ethanol's Actions.

Because ethanol alone inhibits glutamate-induced currents, it could exert its effect by a direct depression of glutamate receptor/channels and consequently a smaller rise in [Ca²⁺]_i(Gruol et al., 1997). This is unlikely because ethanol suppressed glutamate-induced potentiation when ethanol was applied after the end of the glutamate prepulse. Therefore, ethanol probably exerts its suppressant action at a site closer to the glycine receptors; for example, where Ca²⁺-sensitive phosphorylation occurs.

In VTA neurons, both glutamate prepulses and ethanol potentiated glycine responses. These effects showed several similarities. First, the potentiation of glycine receptors by either ethanol (Ye et al., 2001a) or glutamate decreases with higher glycine concentrations. Second, both agents lower the EC₅₀ of glycine. Third, they both accelerate glycine receptor desensitization. Thus, some occlusion between glutamate- and ethanol-induced potentiations is highly probable. However, because ethanol was effective in neurons that showed no potentiation of I_Gly by ethanol alone (Fig. 7 D), another mechanism is also likely involved.

Consequences of Ethanol's Inhibition of Glutamate-Induced Potentiation of Glycine Responses and Other Cellular Activities.

Because excitatory and inhibitory receptors coexist in many neurons, it is essential for us to understand how they interact. Modulation of glycine receptors by agents such as glutamate and ethanol is important because changes in the efficacy of glycinergic transmission have pathophysiological implications in nociception and motor behavior (Breitinger and Becker, 1998; Simpson and Huang, 1998). Moreover, Ca²⁺-dependent clustering of glycine receptors during synaptogenesis has been demonstrated (Kirsch and Betz, 1998). Being present in most VTA neurons, glycine receptors are likely to play an important role in modulating the excitability of VTA dopaminergic and nondopaminergic neurons and, consequently, the release of dopamine and other agents in the brain.

In conclusion, glutamate induced a fast potentiation ofI_Gly in VTA neurons of 5- to 14-day-old rats. This was enhanced when extracellular Ca²⁺ was raised from 2 to 12 mM and diminished by intraneuronal chelators, indicating the possible involvement of intracellular Ca²⁺. Glutamate increased both the apparent affinity of the glycine receptor for its agonist and the maximal response. Ethanol suppressed these effects of glutamate, even when ethanol was applied after a glutamate prepulse. Therefore, ethanol may act on a Ca²⁺-sensitive kinase or other pathways that regulate the function of glycine receptor channels.