Mercury interaction with the GABA_A receptor modulates the benzodiazepine binding site in primary cultures of mouse cerebellar granule cells

Abstract

Mercury compounds are neurotoxic compounds with a great specificity for cerebellar granule cells. The interaction of mercury compounds with proteins in the central nervous system may underlie some of their effects on neurotransmission. In this work we study the interaction of mercuric chloride (HgCl₂) and methylmercury (MeHg) with the GABA_A receptor in primary cultures of cerebellar granule cells. Both compounds increased, dose dependently, the binding of $\text{[}{}^3\text{H}\text{]}$flunitrazepam to the benzodiazepine recognition site. EC₅₀ values for this effect were 3.56 and 15.24 μM for HgCl₂ and MeHg, respectively, after 30 min exposure of intact cultured cerebellar granule cells. The increase of $\text{[}{}^3\text{H}\text{]}$flunitrazepam binding by mercury compounds was completely inhibited by the GABA_A receptor antagonists bicuculline and picrotoxinin, and by the organochlorine pesticide α-endosulfan. It was also partially inhibited by the anion transporter blocker DIDS, however this effect could be due to a possible chelation of mercury by DIDS. Intracellular events, like intracellular calcium, kinase activation/inactivation or antioxidant conditions did not affect $\text{[}{}^3\text{H}\text{]}$flunitrazepam binding or its increase induced by mercury compounds. The sulfhydryl alkylating agent N-ethylmaleimide mimicked the effect of mercury compounds on $\text{[}{}^3\text{H}\text{]}$flunitrazepam binding suggesting a common mechanism. We conclude that mercury compounds interact with the GABA_A receptor by the way of alkylation of SH groups of cysteinyl residues found in GABA_A receptor subunit sequences.

Introduction

Mercury exists in several chemical forms, including elemental mercury (Hg⁰), inorganic mercury (mercuric chloride, HgCl₂) and organic mercury (methylmercury, CH₃HgCl, MeHg). All forms of mercury have toxic effects in a number of organs; in the brain, methylmercury (MeHg) was the causative of the neurological Minamata disease and the clinical manifestations of the Iraqi outbreak (Goyer, 1996). Magnetic resonance examination of patients suffering from Minamata disease revealed brain cortical and cerebellar atrophy with diffuse loss of cerebellar granule cells, while Purkinje cells and brain stem neurones were intact (Korogi et al., 1994). Rats experimentally exposed to MeHg showed both clinical signs of neurologic dysfunction characterized by ataxic behaviour and degeneration of cerebellar granule cells with preservation of Purkinje cells (Nagashima, 1997). In vitro exposure of primary cultures of rat cerebellar granule cells to chronic low MeHg concentrations results in apoptosis (Castoldi et al., 2000).

According to the Mercury Research Strategy for the period 2001–2005 of the US Environmental Protection Agency (EPA), the route targeted for human health effects and exposure involves fish consumption where mercury is: released to the air, transported and deposited on land and water, converted to methylmercury in water bodies, consumed by fish, and then accumulated in mammals, including humans, that eat fish. The EPA's Reference Dose of 0.1 micrograms per kilogram of body weight per day is a scientifically justifiable level for protecting human health from the adverse effects of methylmercury (EPA, 2000).

Although MeHg is the most important mercuric compound in terms of environmental exposure, metallic mercury is the most common form to which workers are exposed. Mercury concentration in the pituitary gland was found to be 40 times higher in a Swedish dental staff population with respect to controls (Nylander and Weiner, 1989). Elemental mercury and MeHg undergo biotransformation in tissues, being accumulated in the body to the more reactive mercuric inorganic compounds (HgCl₂) (Smith et al., 1994). Thus, MeHg and HgCl₂ represent the two main chemical forms of mercury compounds for the effects of mercury in the central nervous system.

Because of the high bonding affinity between mercury and sulphur, there is a focused interest in the interactions that occur between mercury compounds and the thiol group(s) of proteins, peptides and amino acids. Interaction of mercury compounds with proteins in the central nervous system may be the basis for some of the reported effects of mercury compounds on neurotransmission. In addition to the effect of mercury compounds on amino acid and neurotransmitter transport (Estevez et al., 1998, Aschner et al., 2000, Faro et al., 2000, Gassó et al., 2000, Maekawa et al., 2000), actions on receptors for specific neurotransmitters have also been reported. Muscarinic and nicotinic receptors (Bondy and Agrawal, 1980, Castoldi et al., 1996), NMDA receptors (Rajanna et al., 1997) and dopamine receptors (Bondy and Agrawal, 1980, Scheuhammer and Cherian, 1985) have been described as being inhibited by micromolar concentrations of mercury compounds. Mercury compounds also modify GABA_A receptor-mediated inhibitory responses. First, Arakawa et al. (1991) reported a stimulation of GABA-induced current in dorsal root ganglion neurones by mercuric chloride but not by methylmercury, this effect being mediated by protein kinases and protein Gs (Huang and Narahashi, 1997a, Huang and Narahashi, 1997b). Yuan and Atchison (1997) reported that MeHg blocks both inhibitory postsynaptic potentials and currents as well as responses evoked by the GABA_A receptor agonist muscimol in hippocampal CA1 slices. In addition, Allan and Baier (1992) found that mercury compounds inhibited GABA-induced Cl⁻ flux in cortical microsacs. Komulainen et al. (1995) reported that MeHg marginally increased $\text{[}{}^3\text{H}\text{]}$flunitrazepam binding in the absence of GABA in rat brain membranes. Thus, a clear picture of the interaction of mercury compounds with the GABA_A receptor is not fully addressed in the literature.

The GABA_A receptor/Cl⁻ ionophore complex is an oligomeric protein that has separate but allosterically interacting binding sites for the endogenous neurotransmitter GABA, for benzodiazepines and for picrotoxinin-like convulsants. A wide spectrum of drugs, toxic agents and metals modify the GABA_A receptor function by directly interacting with these binding sites or with others not yet well described present in this receptor complex (Macdonald and Olsen, 1994, Sieghart, 1995, Mehta and Ticku, 1999). Primary cultures of cerebellar granule cells constitutively express functional GABA_A receptors. We have recently shown that primary cultures of cerebellar granule cells are a good in vitro model to study allosteric interactions between GABA, benzodiazepine and picrotoxinin recognition sites (Vale et al., 1997) and receptor-mediated neurotoxicity (Pomés et al., 1993, Vale et al., 1998).

The aim of the present study was to determine whether mercury compounds specifically interact with the GABA_A receptor. We used intact cerebellar granule cells as an in vitro model of the neuronal type selectively injured by mercury. We describe here that inorganic (HgCl₂) and organic (MeHg) mercury compounds increased $\text{[}{}^3\text{H}\text{]}$flunitrazepam binding at the benzodiazepine recognition site in primary cultures of mouse cerebellar granule cells, this effect being GABA_A receptor and chloride channel mediated. Because of the importance of redox status for the function of ionotropic receptors such as the NMDA, GABA_A and glycine receptors (Gozlan and Ben-Ari, 1996), we also studied whether the effect of mercury compounds could be antagonized or mimicked by reducing, antioxidant or sulfhydryl alkylating agents. Furthermore, we studied cellular GABA_A modulatory systems known to be related to mercuric disturbance as kinase activities, microtubule network integrity and intracellular calcium homeostasis.