Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Ionotropic glutamate receptors are cation-conducting channels that belong to a superfamily of ligand-gated ion channels. The three major families of ionotropic glutamate receptors have been denoted as the N-methyl-D-aspartate (NMDA), -amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA), and kainate (KA) receptor families. The KA family of glutamate receptors comprises the GluR5-7 and KA1-2 subunits (Hollmann and Heinemann 1994). Similar to GluR2 AMPA receptor subunits, RNA editing takes place in the M2 domain Q/R site of both GluR5 and GluR6 subunits (reviewed in Ozawa et al., 1998). GluR5-Q, GluR6-Q, and GluR6-R subunits can form homomeric channels, whereas GluR5-R subunits only form functional channels when coexpressed with KA1 or KA2 subunits (Herb et al., 1992). KA1 and KA2 subunits do not form homomeric channels but coexpress with GluR5-7 subunits forming receptors with new pharmacological and functional properties. For instance, GluR6 homomeric receptors are not activated by 1 mM AMPA, whereas it activates heteromeric receptors composed of GluR6 plus either KA1 or KA2 subunits (Herb et al., 1992; Sakimura et al., 1992). Although the subunit composition of native KA receptors in most neurons remains unknown, a number of biochemical and functional studies suggest that some KA receptors exist in vivo as heteromers (Sahara et al., 1997; reviewed in Ozawa et al., 1998).

Previous studies showed that KA receptors are acutely modulated by ethanol. Dildy-Mayfield and Harris (1995) demonstrated that concentrations of ethanol as low as 35 mM inhibited KA-gated currents in Xenopus oocytes expressing recombinant GluR6 receptors. We recently reported that recombinant GluR6 receptors expressed in human embryonic kidney (HEK) 293 cells are also inhibited by ethanol and that its mechanism of action does not involve protein phosphorylation (Valenzuela et al., 1998b). Moreover, we recently showed that pharmacologically isolated native KA receptors are acutely inhibited by ethanol in cerebellar granule neurons (Valenzuela et al., 1998a). However, ethanol's effects on KA receptors with different subunit compositions are unknown. In this article, we report the results of electrophysiological studies on the acute effects of ethanol on homomeric and heteromeric recombinant KA receptors composed of GluR5 or R6 subunits in the absence or presence of KA1 or KA2 subunits.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Cell Culture and DNA Transfections. HEK 293 cells were purchased from American Type Tissue Culture Collection (ATCC, Rockville, MD). Rat GluR5-Q, GluR5-R, GluR6-R, KA1, and KA2 receptor subunit cDNAs were subcloned in cytomegalovirus promoter-containing vectors and were generously provided by Drs. Yael Stern-Bach and Steve Heinemann (Salk Institute, La Jolla, CA). HEK 293 cells were maintained in minimum essential media (Hyclone, Logan, UT) plus 10% (v/v) fetal bovine serum (Gemini Bioproducts, Calabasas, CA), 2 mM glutamine, 100 U/ml penicillin, and 0.1 mg/ml streptomycin (all from Sigma Chemical Co., St. Louis, MO) at 37°C 5% CO₂ in a humidified atmosphere; these cells were transiently transfected by the calcium phosphate precipitation method (Chen and Okayama 1987).

Experiments with Xenopus Oocytes. The techniques for injection, culture, and two-voltage clamp electrophysiological recording from Xenopus oocytes used in this study are described in detail elsewhere (Dildy-Mayfield and Harris, 1995). Oocytes were injected with rat GluR5-Q or GluR6-R receptor cRNAs, which were transcribed in vitro with the mRNA capping kit from Stratagene (La Jolla, CA). Concanavalin A (2 µM) was preapplied for 1 min before KA application to prevent desensitization. Ethanol was coapplied with KA for 30 s and effects of drugs were calculated as the percentage of change from an average of control and washout responses.

Patch-Clamp Electrophysiological Recording. For electrophysiological recording, HEK 293 cells were plated on sterile 12-mm glass round coverslips coated with poly-L-lysine and transfected as described above. Cells were used for recording 48 to 72 h after transfection. Immediately before recording, coverslips were transferred to a perfusion chamber (Warner Instruments, Hampden, CT) and visualized under a Leitz Fluovert inverted microscope (Wetzlar, Germany) equipped with Hoffman modulation optics (Greenvale, NY). Currents were measured in the whole-cell patch clamp configuration. Membrane potential was clamped at 60 mV with an Axopatch 200 amplifier (Axon Instruments, Foster City, CA). Recording pipettes (borosilicate capillaries with filament, A. 1.5 mm, Sutter Instruments, Novato, CA) were prepared with a two-step puller (Narishige Instrument Co, Tokyo, Japan) and had resistances between 5 and 7 M. The external solution (all chemicals were from Sigma) contained: 130 mM NaCl, 5 mM KCl, 2 mM CaCl₂, 1 mM MgCl_2, 10 mM HEPES (pH 7.3), 11 mM glucose, and 1 µM concanavalin A. The internal solution contained: 130 mM KCl, 10 mM HEPES, 0.1 mM CaCl₂, 1 mM EGTA, 2 mM ATP, and 0.2 mM GTP (all chemicals were from Fluka, Milwaukee, WI). Drugs were applied with a fast-exchange flow-tube perfusion system that was driven by motor (Warner Instrument Co.) and controlled by a Master-8 stimulator (A.M.P.I, Jerusalem, Israel). Agonists were applied at 60-s intervals. Ethanol was coapplied with agonist and was also present in the buffer syringe. Data was acquired and analyzed with the Neuropro software package (RC Electronics, Santa Barbara, CA). Effects of drugs were calculated as the percentage of change from an average of control and washout responses.

Statistical Analysis. All statistical analysis and curve fitting was performed using GraphPad Prizm software (San Diego, CA). KA dose-response curves were fitted to a four-parameter logistic equation (sigmoid). Effects of ethanol were analyzed by one-sample Student's t test (against a theoretical mean of zero) and by one- or two-way ANOVA. Numbers given in parentheses (n =) refer to the number of individual cells or oocytes used in the statistical analysis. Data is presented as mean ± S.E.M. in all cases.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Effects of Ethanol on KA Receptors Expressed in Xenopus Oocytes. Application of KA in the presence of concanavalin A (2 µM) activated GluR5-Q receptors with an EC₅₀ of 27 µM (95% confidence interval 22-33 µM) and a Hill slope of 0.7 (95% confidence interval 0.6-0.8; n = 6-7; Fig. 1), and GluR6-R receptors with an EC₅₀ of 1.7 µM (95% confidence interval 0.9-3.2 µM) and a Hill slope of 0.8 (95% confidence interval 0.5-1; n = 7 - 8; Fig. 1). Ethanol (75 mM) inhibited GluR5-Q receptor-dependent KA currents by 40 ± 7, 28 ± 6, 31 ± 3, 24 ± 4, and 27 ± 3% at KA concentrations of 1 µM, 5 µM, 25 µM, 100 µM, and 5 mM, respectively (Fig. 1, inset). Ethanol (75 mM) inhibited GluR6-R receptor-dependent KA currents by 42 ± 6, 36 ± 4, 27 ± 4, 22 ± 3, and 19 ± 2% at KA concentrations of 0.5, 1, 3, 10, and 100 µM, respectively (Fig. 1, inset). It should be noted that ethanol did not change the KA EC₅₀ or Hill coefficient for these receptors. For GluR5-Q receptors, the EC₅₀ value in the presence of ethanol was 34 µM (95% confidence interval 20-59 µM) and the Hill coefficient was 0.7 (95% confidence interval 0.5-0.9; n = 6-7). For GluR6-R receptors, the EC₅₀ value in the presence of ethanol was 2.7 µM (95% confidence interval 1.5-5.3 µM) and the Hill coefficient was 0.8 (95% confidence interval 0.5-1.1; n = 8). The GluR5Q receptor E_max was decreased to 72% of control (95% confidence interval 63-82%) and the GluR6 receptor E_max was decreased to 80% of control (95% confidence intervals 68-98%).

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Effect of 75 mM ethanol on KA dose-response curve for homomeric GluR5-Q and GluR6-R receptors expressed in Xenopus oocytes. Currents were recorded in two-electrode voltage clamp configuration in the absence () and presence of ethanol (). Data were fitted to a four- parameter logistic equation (sigmoid) using GraphPad Prizm software. Each point represents average currents ± S.E.M. recorded from six to eight oocytes. Data were normalized with respect to maximal KA responses (5 mM for GluR5-Q and 100 µM for GluR6-R). Inset, ethanol-induced percentage of change of KA receptor currents as a function of the KA concentration used.

View larger version (22K): [in this window] [in a new window]   Fig. 2.   Comparison of ethanol's effect on GluR5-Q versus GluR6-R receptors expressed in Xenopus oocytes. Upper panel, representative tracings of effects of 50 mM ethanol. Lower panel, summary of effects of 25 to 100 mM ethanol. Ethanol was tested on receptors activated by concentrations at or below the KA EC50. Each point represents mean ± S.E.M. of four oocytes. Scale bar, 50 nA by 30 s.

View larger version (21K): [in this window] [in a new window]   Fig. 3.   Effects of ethanol on current/voltage relationships for GluR5-Q and GluR6-R receptors expressed in Xenopus oocytes. Currents were measured at indicated membrane holding potentials in the absence () and presence () of 75 mM ethanol. Each point represents mean ± S.E.M. of five to seven determinations. KA concentration was 5 µM. Currents were normalized with respect to responses obtained at 90 mV in the absence of ethanol. Inset, summary of percent change induced by ethanol at different holding membrane potentials.

Effects of Ethanol on Receptors Expressed in HEK 293 Cells. HEK 293 cells transfected with GluR6-R, GluR6-R + KA1, or GluR6-R + KA2 subunits were tested for sensitivity to 100 µM KA and 1 mM AMPA (Fig. 4). Homomeric GluR6 receptors were virtually insensitive to 1 mM AMPA. Conversely, 1 mM AMPA produced significant currents in cells transfected with GluR6-R plus either KA1 or KA2 (Fig. 4). It should be noted that all of the results reported in this article on heteromeric GluR6-R + KA1 or KA2 receptors were obtained from cells that significantly responded to 1 mM AMPA.

View larger version (30K): [in this window] [in a new window]   Fig. 4.   Representative tracings of currents gated by different GluR6-R subunit-containing KA receptors used in this study. Left, comparison of currents produced by 100 µM KA versus 1 mM AMPA. Right, effects of 25 to 100 mM ethanol on currents gated by KA ( EC50).

View larger version (29K): [in this window] [in a new window]   Fig. 5.   Studies with KA receptors containing GluR6-R subunits with and without KA1 or KA2 subunits. Upper, KA dose-response curves. Data were fitted to a four-parameter logistic equation (sigmoid) using GraphPad Prizm software. Each point represents average currents ± S.E.M. recorded from three to six cells. Data were normalized with respect to maximal KA responses (100 µM). Lower, acute effect of 25 to 100 mM ethanol on receptors activated by submaximal (EC50) KA concentrations. Each point represents mean ± S.E.M. of currents recorded from 6 to 14 cells. For results of statistical analyses, see text.

View larger version (21K): [in this window] [in a new window]   Fig. 6.   Studies with homomeric GluR5-Q KA receptors. Upper, domoate dose-response curves. Data were fitted to a four-parameter logistic equation (sigmoid) using GraphPad Prizm software. Each point represents average currents ± S.E.M. recorded from five to six cells. Data were normalized with respect to maximal domoate responses (25 µM). Middle and lower, sample tracings and summary of acute effect of 25 to 100 mM ethanol on submaximal (EC20-30) domoate currents. Each point in lower panel represents mean ± S.E.M. of currents recorded from eight cells. For results of statistical analyses, see text.

View larger version (28K): [in this window] [in a new window]   Fig. 7.   Studies with KA receptors containing GluR5-R subunits plus KA1 or KA2 subunits. Upper, sample tracings of the acute effect of 25 to 100 mM ethanol on submaximal (EC20-50) KA currents. Lower left, KA dose-response curves. Data were fitted to a four-parameter logistic equation (sigmoid) using GraphPad Prizm software. Each point represents average currents ± S.E.M. recorded from four to eight cells. Data were normalized with respect to maximal KA responses (100 µM). Lower right, summary of ethanol's effects on these receptors. Each point in lower panel represents mean ± S.E.M. of currents recorded from five to six cells. For results of statistical analyses, see text.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The results of the present study demonstrate that the function of recombinant KA receptors with different subunit compositions is acutely inhibited by ethanol. Previous studies on the actions of ethanol on recombinant KA receptors focused on homomeric GluR6 receptors expressed in Xenopus oocytes or HEK 293 cells, and these studies reported that 25 to 100 mM ethanol acutely inhibited the function of these receptors by 10 to 30% (Dildy-Mayfield and Harris 1995; Valenzuela et al., 1998b). We now report the effects of ethanol on homomeric KA receptors composed of GluR5-Q and heteromeric KA receptors composed of GluR6-R or GluR5-R plus KA1 or KA2 subunits. We found that ethanol inhibits, to a similar extent, KA receptors with all of these subunit compositions. We detected statistically significant differences in only two cases: GluR6 + KA2 receptors were inhibited significantly less than GluR6 and GluR6 + KA1 receptors, and GluR5-R + KA2 receptors were inhibited significantly more than GluR5-Q and GluR5-R + KA1 receptors. However, the differences between receptors with different subunit compositions were relatively small.

Present results with recombinant KA receptors are in agreement with a recent study from our laboratory with native KA receptors. We reported that acute exposure to 25 to 100 mM ethanol inhibited pharmacologically isolated KA receptor currents by 5 to 25% in cerebellar granule cells (Valenzuela et al., 1998a). In these cells, the majority of KA receptors were reported to contain either GluR5-R, GluR5-Q, or GluR6-Q subunits; GluR6-Q subunits appear to associate to KA2 subunits in these cells (Savidge et al., 1997; Pemberton et al., 1998). In cerebellar granule cells, we found that ethanol acutely inhibited NMDA receptors to a greater extent than either AMPA or KA receptors (Valenzuela et al., 1998a). Thus, NMDA receptors may be more important targets than KA receptors in the acute actions of pharmacologically relevant concentrations of ethanol. It should be emphasized, however, that the magnitude of ethanol's acute effects on KA receptor function might depend on the brain region studied. We recently found that KA receptor-mediated synaptic currents in the CA3 region of the hippocampus are inhibited by 11 to 50% in the presence of 20 to 80 mM ethanol (J. L. Weiner, T. V. Dunwiddie, and C. F. Valenzuela, unpublished observation). It would be interesting to study ethanol's actions on KA receptors expressed in neurons from central nervous system (CNS) regions other than the hippocampus and cerebellum and to establish the factors that determine this differential sensitivity of KA receptors to ethanol. This study and a previous one (Valenzuela et al., 1998b) suggest that neither subunit composition nor protein phosphorylation modulate acute ethanol's actions on recombinant KA receptors. Whether these factors, or other factors such as receptor clustering, localization and/or association with neuronal specific proteins, play a role in determining the sensitivity of native KA receptors to ethanol remains to be elucidated.

A challenging question for future study will be to locate the molecular site of action of ethanol on KA receptors and other ionotropic glutamate receptors. In the case of -aminobutyric acid_A (GABA_A) and glycine receptors, ethanol was reported to interact with a specific binding pocket located between transmembrane domains M2 and M3 (Mihic et al., 1997; Wick et al., 1998). Minami et al. (1998) found that in GluR6 receptors a specific amino acid, residue Gly-819 in M4, is important for the actions of the volatile anesthetics halothane, isoflurane, and enflurane. However, this residue does not appear to play a role in ethanol's actions on GluR6 KA receptors. Thus, more work will be required to determine the mechanism of action of ethanol on KA receptors at the molecular level.

Another question for future research will be to determine whether the acute effects of ethanol on KA receptors contribute significantly to the pathophysiology of acute intoxication. To answer this question, we need to learn more about the roles played by KA receptors in the CNS. KA receptors are found throughout many regions and they are particularly abundant in the granular cell layer of the cerebellum, CA3 region of the hippocampus, cerebral cortex, caudate/putamen, hypothalamus, reticulothalamic nucleus, and pontine nuclei (Petralia et al., 1994). Selective AMPA receptor antagonists such as LY300168 (GYKI-53655) were recently used to determine that presynaptic GluR5 KA receptors reduce the effectiveness of GABAergic synaptic inhibition in the rat hippocampus (Clarke et al., 1997; Rodriguez-Moreno et al., 1997). Moreover, KA receptors appear to be more complex than previously expected by acting not only as ionotropic receptors but also as metabotropic G protein-coupled receptors (Rodriguez-Moreno and Lerma, 1998). KA receptors mediate synaptic currents in neurons of the CA3 region of the hippocampus (Castillo et al., 1997; Vignes and Collingridge, 1997), and these currents are absent in GluR6 knockout mice (Mulle et al., 1998). KA receptor activation in interneurons increases tonic GABAergic inhibition of pyramidal cells (Cossart et al., 1998) and increases the frequency of spontaneous inhibitory postsynaptic currents (Frerking et al., 1998). Thus, decreases in synaptic KA receptor function in specific areas of the CNS may be important for acute ethanol intoxication. It is possible that relatively small effects of ethanol on KA receptors located in neuronal pathways where alcohol also affects the function of NMDA, GABA_A, voltage-gated ion channels, and/or metabotropic receptors, could contribute to the pathophysiology of acute ethanol intoxication. Clearly, more work will be required to test this hypothesis. For instance, it would be interesting to measure the acute effects of ethanol on the GluR6 knockout mice developed by Mulle et al. (1998).

In conclusion, we found that recombinant KA receptors composed of GluR5 or GluR6 subunits in the absence and presence of KA1 or KA2 subunits are all acutely inhibited by ethanol. Ethanol inhibited to a similar extent these recombinant KA receptors with different subunit compositions. It would be interesting to determine whether alcohol modulates in a similar manner native KA receptors with different subunit configurations in neurons of various CNS regions.