γ-Hydroxybutyrate inhibits excitatory postsynaptic potentials in rat hippocampal slices

Abstract

γ-Hydroxybutyrate (GHB) has been shown to mimic different central actions of ethanol, to suppress alcohol withdrawal syndrome, and to reduce alcohol consumption both in rats and in humans. The aim of the present study was to determine if GHB shared with alcohol the ability to inhibit glutamate action at both NMDA and AMPA/kainate receptors. The NMDA or the AMPA/kainate receptors-mediated postsynaptic potentials were evoked in CA1 pyramidal neurons by stimulation of Schaffer-collateral commissural fibers in the presence of CGP 35348, bicuculline to block the GABA_B and GABA_A receptors, and 10 μM 6,7-dinitroquinoxaline-2,3-dione (DNQX) or 30 μM dl-2-amino-5-phosphonovalerate (d-APV) to block AMPA/kainate or NMDA receptors, respectively. GHB (600 μM) produced a depression of both NMDA and AMPA/kainate receptors-mediated excitatory postsynaptic potentials with recovery on washout. The GHB receptors antagonist, NCS-382, at the concentration of 500 μM had no effect per se on these responses but prevented the depressant effect of GHB (600 μM) on the NMDA and AMPA/kainate-mediated responses. In the paired-pulse experiments, GHB (600 μM) depressed the amplitude of the first and the second evoked AMPA/kainate excitatory postsynaptic potentials, and significantly increased the paired-pulse facilitation (PPF). These results suggest that GHB inhibits excitatory synaptic transmission at Schaffer-collateral commissural–pyramidal neurons synapses by decreasing the probability of release of glutamate.

Introduction

γ-Hydroxybutyrate (GHB) is a naturally occurring metabolite of γ-aminobutyric acid (GABA) present in micromolar quantities in the central nervous system (Roth and Giarman, 1970; Maitre, 1997). The highest concentration of GHB is found in the synaptosomal fraction (Maitre et al., 1983b; Snead, 1987) and in vitro, GHB is released by neuronal depolarization in a Ca²⁺-dependent manner (Maitre and Mandel, 1982; Maitre et al., 1983a). A high-affinity, Na⁺-dependent uptake system for GHB has also been reported (Benavides et al., 1982a; Hechler et al., 1985; Vayer et al., 1987a). GHB high-affinity binding sites are present only in neurons, with a restricted specific distribution in the hippocampus, cortex and basal ganglia of the rat brain (Benavides et al., 1982b; Hechler et al., 1987; Hechler et al., 1991). Maximal high-affinity binding occurs in the CA1 field of the hippocampus where GHB induces increase of guanosine cyclic 3′,5′-phosphate (cGMP) levels (Vayer et al., 1987b; Vayer and Maitre, 1989; Hechler et al., 1992). This effect is blocked by the GHB receptors antagonist, NCS-382 (Maitre et al., 1990). The central GHB receptors are of G protein-linked receptor type, and it has been suggested that the G proteins implicated in GHB receptor coupling are of the G_i or G_o family (Ratomponirina et al., 1995).

Electrophysiological recordings of hippocampal or thalamocortical neurons in vitro have revealed a hyperpolarization of the membrane potential after local application of GHB (Olpe and Koella, 1979; Kozhechkin, 1980; Xie and Smart, 1992a; Williams et al., 1995). In hippocampal CA1 pyramidal neurons, a decrease in excitatory and inhibitory postsynaptic potentials was also described. These actions were blocked by GABA_B receptor antagonists, CGP 36742 or CGP 35348, suggesting that GHB can activate pre- and postsynaptic GABA_B receptors (Xie and Smart, 1992b). However, it has been shown that some neurophysiological and neuropharmacological effects of GHB are mediated by specific receptors. Indeed, NCS-382 has been reported to antagonize GHB-induced increase in guanosine cGMP levels in hippocampus and in vivo release of both dopamine and opioid-like substances (Vayer et al., 1987b). In the prefrontal cortex, two opposite effects on neuronal firing rate were observed with high and low doses of GHB where only the increase of firing rate of neurons at low doses was blocked by NCS-382 (Godbout et al., 1995).

GHB has been shown to reduce voluntary alcohol intake in alcohol-preferring rats (Fadda et al., 1988) and alcohol craving in humans (Gallimberti et al., 1992; Addolorato et al., 1998). In addition, GHB has been shown to suppress the severity of alcohol withdrawal symptoms in both ethanol-dependent rats (Fadda et al., 1989) and alcoholics (Gallimberti et al., 1989). The presence of symmetrical generalization between the discriminative stimulus effects of GHB and ethanol in rats (Colombo et al., 1995) could explain the pharmacotherapeutic use of GHB besides supporting the hypothesis that GHB may have ethanol-like action. Limited double-blind and more extended open studies indicate that GHB is highly effective in reducing craving, alcohol intake and relapses in alcoholic patients. GHB is self-administrated by rats, induces conditioned place preference, is voluntarily consumed by alcohol-preferring rats and is potentially abused by humans (Gallimberti et al., 1992; Addolorato et al., 1996). Although these studies reported no withdrawal symptoms when GHB was discontinued, it is worthy to mention that GHB has the potential to cause physical dependence (Galloway et al., 1997; Tunnicliff, 1997). Accumulating evidences suggest that neurophysiological and pathological effects of ethanol are mediated to considerable extent through the glutamatergic system (Tsai et al., 1995).

Ethanol inhibits both N-methyl-d-aspartate (NMDA) and α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA/kainate) types of glutamatergic synaptic transmission (Lovinger et al., 1989; Weight et al., 1991; Lovinger, 1993; Nie et al., 1994).

The aim of the present study was to determine whether GHB shares with alcohol the ability to inhibit both NMDA and AMPA/kainate-mediated excitatory postsynaptic potentials in the CA1 pyramidal cells in hippocampal slices.