QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services 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 Sinkkonen, S. T. Articles by Korpi, E. R. Search for Related Content PubMed PubMed Citation Articles by Sinkkonen, S. T. Articles by Korpi, E. R.

Receptor Subtype-Dependent Positive and Negative Modulation of GABA_A Receptor Function by Niflumic Acid, a Nonsteroidal Anti-Inflammatory Drug

Department of Pharmacology and Clinical Pharmacology, University of Turku, Turku, Finland (S.T.S, S.M., M.U.-O.); Turku Graduate School of Biomedical Sciences, University of Turku, Turku, Finland (S.T.S.); Clinical Research Group, Department of Psychiatry, University of Mainz, Mainz, Germany (H.L.); and Institute of Biomedicine, Pharmacology, University of Helsinki, Helsinki, Finland (S.T.S., T.M., E.R.K.)

Received January 14, 2003; accepted June 11, 2003.

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
In addition to blocking cyclooxygenases, members of the fenamate group of nonsteroidal anti-inflammatory drugs have been proposed to affect brain GABA_A receptors. Using quantitative autoradiography with GABA_A receptor-associated ionophore ligand [³⁵S]t-butylbicyclophosphorothionate (TBPS) on rat brain sections, one of the fenamates, niflumate, at micromolar concentration was found to potentiate GABA actions in most brain areas, whereas being in the cerebellar granule cell layer an efficient antagonist similar to furosemide. With recombinant GABA_A receptors expressed in Xenopus laevis oocytes, we found that niflumate potentiated 3 µM GABA responses up to 160% and shifted the GABA concentration-response curve to the left in 122 receptors, the predominant GABA_A receptor subtype in the brain. This effect needed the 2 subunit, because on 12 receptors, niflumate exhibited solely an antagonistic effect at high concentrations. The potentiation was not abolished by the specific benzodiazepine site antagonist flumazenil. Niflumate acted as a potent antagonist of 62 receptors (with or without 2 subunit) and of X22 receptors containing a chimeric 1 to 6 subunit, which suggests that niflumate antagonism is dependent on the same transmembrane domain 1- and 2-including fragment of the 6 subunit as furosemide antagonism. This antagonism was noncompetitive because the maximal GABA response, but not the potency, was reduced by niflumate. These data show receptor subtype-dependent positive and negative modulatory actions of niflumate on GABA_A receptors at clinically relevant concentrations, and they suggest the existence of a novel positive modulatory site on 122 receptors that is dependent on the 2 subunit but not associated with the benzodiazepine binding site.

The GABA_A receptor convulsant [³⁵S]t-butylbicyclophosphorothionate (TBPS) specifically binds to a picrotoxinin-sensitive site associated with the GABA_A receptor ionophore (Squires et al., 1983). It has been a useful neurochemical tool in GABA_A receptor research, because [³⁵S]TBPS binding can be studied on native receptor populations in brain sections and homogenates. Positive modulators, such as GABA and barbiturates, decrease [³⁵S]TBPS binding (Squires et al., 1983; Maksay and Ticku, 1985), whereas negative modulators, such as bicuculline, block this effect of GABA (Squires and Saederup, 1987). Furosemide, presently the most subtype-selective antagonist, noncompetitively increases the binding in the cerebellar granule cell layer and 6 subunit-containing recombinant receptors (Korpi et al., 1995). Modulation of [³⁵S]TBPS binding thus reveals also allosteric interactions within the receptor complex. Importantly, dissociation of [³⁵S]TBPS binding by GABA and other positive modulators has been shown to reflect receptor function itself as measured, for example, by ³⁶Cl^– flux assay (Im and Blakeman, 1991). Thus, [³⁵S]TBPS autoradiography on brain sections offers the advantage that pharmacological heterogeneity of the GABA_A receptor subtypes can be visualized in various brain regions and preliminarily correlated with function.

Niflumate (Fig. 1) belongs to the fenamate group of nonsteroidal anti-inflammatory drugs (NSAIDs). It is available for clinical use in several European countries. Its mechanism of action is believed to be based on inhibition of cycloxygenases (Cushman and Cheung, 1976) that results in antipyretic, analgesic, and anti-inflammatory effects (Vane and Botting, 1998). In addition to these effects on prostaglandin synthesis, it has been shown to interact with central GABA_A receptors in vitro. Niflumate decreases the enhancing effect of permeable anions on [³⁵S]TBPS binding (Evoniuk and Skolnick, 1988). It also decreases the binding of another convulsant, 4'-ethynyl-4-n[2,3-³H₂]propyl-bicycloorthobenzoate ([³H]EBOB), to the GABA_A receptor picrotoxinin-sensitive site on rat brain membranes and enhances the inhibitory effect of GABA on the binding (Maksay et al., 1998). Niflumate has GABA-enhancing actions on both cortical and cerebellar membranes, whereas furosemide (Fig. 1) antagonizes GABA in cerebellar membranes (Maksay et al., 1998). Niflumate inhibits GABA-induced currents in Xenopus laevis oocytes injected with rat brain total RNA (White and Aylwin, 1990), but it has been also reported to have bimodal effects on GABA-induced currents in X. laevis oocytes injected with rat cortical poly(A)⁺ RNA (Woodward et al., 1994), because 10 µM niflumate potentiates responses to 10 µM GABA and inhibits responses to maximal (3 mM) GABA with IC₅₀ value of 7 µM. As is evident from the above-mentioned findings, the mechanisms of positive and negative modulatory actions of niflumate on GABA_A receptor function remain to be clarified. Because NSAIDs permeate the blood-brain barrier (Bannwarth et al., 1989), there is a possibility that niflumate affects GABA_A receptors in vivo. In fact, another fenamate, mefenamic acid, has been shown to have severe central effects, including convulsions and coma in human overdose (Smolinske et al., 1990). We have now studied the mode of action of niflumate on GABA_A receptor subtypes by using 1) [³⁵S]TBPS autoradiography on rat brain sections to reveal heterogeneity in niflumate actions on native GABA_A receptors, and 2) two-electrode voltage-clamp recording of X. laevis oocytes expressing recombinant GABA_A receptor to assess the receptor subunit dependence of niflumate actions at functional level. We revealed subunit dependence for both positive and negative modulatory actions of niflumate.

View larger version (18K): [in this window] [in a new window]   Fig. 1. Structures of niflumic acid and furosemide.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
All experimental procedures were done with the permission from the Western Finland Provincial Government and the Institutional Animal Use and Care committee of the University of Turku (Turku, Finland).

Ligand Autoradiography. Male Wistar rats (n = 4; Central Animal Laboratory, University of Turku) were maintained in a stainless steel wire-mesh cage with pellet food (Special Diet Service; Witham, Essex, England) and tap water available ad libitum. Rats were decapitated at the age of 4 months, and whole brains were carefully dissected out, rinsed in ice-cold saline, and frozen on dry ice. The frozen brains were wrapped in plastic, and stored at –80°C. [³⁵S]TBPS autoradiography was modified from the standard assay (Sinkkonen et al., 2001). In brief, 14-µm horizontal rat brain sections were cut in a Microm HM 500 OM cryostat (Microm International GmbH, Walldorf, Germany). The sections were thaw-mounted onto gelatin-coated object glasses and stored frozen under desiccant at –20°C. For autoradiography, sections were preincubated in an ice-cold 50 mM Tris-HCl supplemented with 120 mM NaCl, pH 7.4, for 15 min. Final incubation in the preincubation buffer was performed with 3 nM (421 cpm/µl) [³⁵S]TBPS (PerkinElmer Life Sciences, Boston, MA) at room temperature (22°C) for 90 min. Nonspecific binding was determined with 100 µM picrotoxinin (Sigma-Aldrich, St. Louis, MO). Displacement of [³⁵S]TBPS binding was studied in the absence and presence of 3 µM or 1 mM GABA (Sigma-Aldrich). The effects of 10, 30, 100, 300, and 1000 µM niflumate (Sigma-Aldrich, dissolved in 0.1 N NaOH) and 100 µM furosemide (Sigma-Aldrich, dissolved in 0.1 N NaOH) on [³⁵S]TBPS binding were tested with or without 3 µM or 1 mM GABA. Niflumate or furosemide did not affect the pH of the incubation buffer. After incubation, the sections were washed three times for 30 min in ice-cold 10 mM Tris-HCl buffer, pH 7.4. Sections were then dipped into distilled water, air-dried at room temperature, and exposed with a plastic ¹⁴C standard to BioMax MR films (Eastman Kodak, Rochester, NY), for 3 days (basal [³⁵S]TBPS binding) or 2 weeks (other binding conditions). Regional labeling intensities of the sections were quantified from the films by using AIS image analysis devices and programs (Imaging Research, St. Catherine's, ON, Canada) the binding values given as radioactivity levels estimated for gray matter areas (in nanocuries per gram).

Recombinant Receptor Expression in X. laevis Oocytes. Capped cRNAs coding for rat GABA_A receptor subunits 1, 6, 2, and 2S (Lüddens et al., 1990; Shivers et al., 1989; Ymer et al., 1989), and 1-16 chimera [where amino acids of the 6 subunit including the first two transmembrane (TM) domains replace the corresponding amino acids in 1 subunit frame to gain furosemide sensitivity; Jackel et al., 1998], were transcribed in vitro from pRK5 plasmids using mMessage mMachine kit (Ambion, Austin, TX) according to manufacturer's instructions. Oocytes were dissected from adult X. laevis female frogs (Horst Kähler, Hamburg, Germany) anesthetized with 0.2% tricaine methanesulfonate (Sigma-Aldrich). Isolated oocytes were stored in normal frog Ringer (NFR): 115 mM NaCl, 2.5 mM KCl, 18 mM CaCl₂, and 10 mM HEPES, pH 7.5. Oocytes were then defolliculated manually and injected with 46 nl of a solution containing mixtures of subunit cRNAs (0.1–2.5 µg/µl) or pure H₂O with Drummond Nanoject injector (Drummond Scientific Co., Broomall, PA) via a glass micropipette with i.d. of 20 to 40 µm. The oocytes were incubated at 19°C in incubation solution [88 mM NaCl, 1 mM KCl, 2.4 mM NaHCO₃, 10 mM HEPES, 0.82 mM MgSO₄, 0.33 mM Ca(NO₃)₂, 0.91 mM CaCl₂, 0.5 mM theophylline, 2 mM sodium pyruvate, 10 U/ml penicillin and 10 µg/ml streptomycin, pH 7.5]. After injection (2 h–1 day) oocytes were digested for 30 min in Ca²⁺-free medium (82.5 mM NaCl, 2.5 mM KCl, 1 mM MgCl₂, 1 mM Na₂HPO₄, and 5 mM HEPES, pH 7.5) containing 0.3 U/ml collagenase type IA (Sigma-Aldrich). Thereafter, the oocytes were incubated in incubation solution until recordings.

Electrophysiological Recordings. For each experiment, oocytes from at least two different frogs were used. Electrophysiological recordings were made 1 to 3 days after cRNA injection. Oocytes were perfused with NFR ± drugs at a flow rate of 1 ml/min at room temperature (22°C) using Ismatec pump (Ismatec, Glattbrugg-Zürich, Switzerland) and 17 channel perfusion system with pinch valves. Drug combinations were mixed before experiments. Oocytes were impaled with two microelectrodes (1.0–2.5 M) filled with 3 M KCl plus 10 mM EGTA, and voltage clamped at –50 mV with Turbo TEC-05 two electrode voltage-clamp amplifier (NPI Electronic GmbH, Tamm, Germany). Experiments were controlled by EggWorks experimental control and data acquisition software program version 3.0.2 (NPI Electronic GmbH). GABA was dissolved in NFR. Niflumate and furosemide were dissolved in 0.1 M NaOH, stocks were diluted in NFR to a concentration of 10 mM, and pH was adjusted to 7.5. Drugs were applied for 10 s unless otherwise stated, and 180- to 600-s washout period was used, depending on drug concentrations. In the GABA concentration-response experiments, any given GABA concentration was first applied alone and thereafter in the presence of niflumate. All different GABA concentrations were tested in every oocyte.

Data Analysis and Statistics. Data analyses were performed using EggWorks Reader version 3.0.2 (NPI Electronic GmbH) and GraphPad Prism version 3.0 (GraphPad Software Inc., San Diego, CA) programs. For autoradiography, the specific [³⁵S]TBPS binding values were determined by subtracting the nonspecific binding values from the corresponding total binding values under each incubation condition. To assess the statistical significance of the niflumate effects on [³⁵S]TBPS binding, one-way analysis of variance (ANOVA) and Dunnett's post hoc test were used. Student's t test was used for 100 µM furosemide effect. For electrophysiological recordings, the amplitudes of peak currents induced by GABA + drug applications were determined from recorded traces, normalized to the corresponding GABA-induced peak currents estimated linearly between the GABA peak currents closest before and after the applications of GABA with the drugs, and presented as a percentage of the control GABA current. The peak currents induced by various GABA concentrations for each oocyte were normalized by setting the maximal GABA current without niflumate to 100%, and the GABA concentration-response curves were generated using nonlinear regression fit. The statistical significance of the niflumate modulation of the GABA response was assessed with one-way ANOVA and Dunnett's post hoc test. Furosemide and niflumate effects at 1,000 µM without additional GABA were assessed using Student's t test.

	Results

Top Abstract Materials and Methods Results Discussion References

 
Niflumate Effects on Native GABA_A Receptors in Brain Sections. [³⁵S]TBPS binding to picrotoxinin-sensitive convulsant site in GABA_A receptors was observed throughout the rat brain in ligand autoradiography (Fig. 2). Niflumate decreased the basal [³⁵S]TBPS binding in all brain regions at 300 and 1,000 µM concentrations, the latter leaving less than 10% of the basal binding (Fig. 2; Table 1). This was in accordance with previous results, because niflumate has been shown to decrease both [³⁵S]TBPS and [³H]EBOB bindings to rat brain homogenates without additional GABA (Evoniuk and Skolnick, 1988; Maksay et al., 1998). In contrast to niflumate, 100 µM furosemide increased the binding in the cerebellar granule cell layer but was inactive in the forebrain, in keeping with previous results (Korpi et al., 1995).

View larger version (115K): [in this window] [in a new window]   Fig. 2. Regional [35S]TBPS binding to rat brain sections and its modulation by niflumate (N, micromolar concentration) or furosemide (F, 100 µM) in the absence and presence of 3 µM and 1 mM GABA. Ctx, cortex; Gr, cerebellar granule cell layer; Hi, hippocampus; IC, inferior colliculus; Mol, cerebellar molecular layer; Str, striatum; Th, thalamus. Images in the first top row are from a film exposed for 3 days; others are from a film exposed for 2 weeks.

Adding 3 µM GABA clearly inhibited the basal [³⁵S]TBPS binding in all brain areas (Fig. 2; Table 2). Regional diversity of niflumate actions was revealed: whereas GABA-inhibited [³⁵S]TBPS binding was further decreased in a concentration-dependent manner in the forebrain, the cerebellar granule cell layer binding component was increased maximally to 200 ± 14% (mean ± S.E., n = 4) by 100 µM niflumate. This latter effect seemed to be biphasic, because 1,000 µM niflumate decreased the binding below control values. Furosemide at 100 µM similarly increased the GABA-inhibited [³⁵S]TBPS binding to maximally 332 ± 36% in the cerebellar granule cell layer. In a previous study, niflumate enhanced the inhibitory effect of 2 µM GABA on [³H]EBOB binding both in the cerebrocortical and cerebellar membranes, whereas furosemide was clearly antagonistic in the cerebellum (Maksay et al., 1998). In the present experiments using brain sections, we were able to differentiate the cerebellar effects of niflumate to an antagonism in the granule cell layer and to an agonism in the molecular layer; the net effect (i.e., when the cerebellum was analyzed as a whole) was negligible up to 300 µM (Table 2). Thus, it seems that in brain homognism by niflumate in the cerebellar granule cell layer seen now in autoradiography is masked by the agonism on the quantitatively larger receptor population of the molecular layer.

When 1 mM GABA was added to reveal the GABA-insensitive [³⁵S]TBPS binding component (Sinkkonen et al., 2001), [³⁵S]TBPS was displaced to background level in most brain regions, whereas many thalamic nuclei and cerebellar granule cell layer were still labeled (Fig. 2). Adding niflumate (300 µM) in this condition enhanced [³⁵S]TBPS binding in the granule cell layer maximally up to 414 ± 22% from the 1 mM GABA level. The thalamic component of the GABA-insensitive [³⁵S]TBPS binding was also increased about 50% by low niflumate concentrations, whereas it was decreased by 1,000 µM (Table 3). Furosemide efficiently blocked the 1 mM GABA action in the cerebellar granule cell layer and had a weak GABA antagonist effect also in the thalamus.

Positive and Negative Modulation of 122 and 622 GABA_A Receptor Function by Niflumate. To search for structural determinants for the regional heterogeneity of niflumate effects on GABA-modulated [³⁵S]TBPS binding, we applied two-electrode voltage clamp on X. laevis oocytes expressing recombinant GABA_A receptors. Because niflumate possesses some structural similarities to furosemide (Fig. 1), and inhibition of the GABA effect on [³⁵S]TBPS binding by these two compounds was regionally similar, to that by furosemide. Furosemide antagonizes GABA noncompetitively in recombinant receptors containing 6 and 2/3 subunits (Korpi et al., 1995), which are abundantly expressed in the cerebellar granule cell layer (Wisden et al., 1992). Most native GABA_A receptors contain 2 subunit, which holds true also in the cerebellar granule cell layer (Whiting et al., 1995). For these reasons, recombinant 622 receptors were chosen for antagonism studies. Because the GABA-potentiating action of niflumate was seen throughout the brain in the [³⁵S]TBPS autoradiography, subunit combination of 122, as the most abundant GABA_A receptor subtype in the native brain (Whiting et al., 1995), was selected for studies on positive modulation.

Different GABA concentrations were applied to oocytes to determine the GABA sensitivities of 122 and 622 receptors (Fig. 3b, inset). Nonlinear regression fit of normalized GABA currents yielded concentration-response curves with EC₅₀ values of 8.0 and 1.8 µM for 122 and 622 receptors, respectively. In autoradiography, niflumate and furosemide effects were tested in the presence of 3 µM GABA, which concentration induced about 30 and 60% of the maximal response in 122, and in 622 receptors, respectively. These were optimal responses for studies on positive and negative modulations, respectively. When niflumate was applied to oocytes expressing 122 receptors, it potentiated the 3 µM GABA-evoked currents in a concentration-dependent manner (EC₅₀ of 31 ± 3 µM, E_max of 162 ± 4% of the basal response), with potentiation of 61 ± 7% at the highest 1,000 µM concentration used (Fig. 3). When niflumate was applied alone, it elicited a small inward current at 1,000 µM concentration, lower concentrations having no effect (data not shown). Furosemide at 100 µM tended to enhance GABA-induced currents in 122 receptors, but the effect was not statistically significant. Niflumate antagonized the 3 µM GABA-evoked currents in oocytes expressing 622 receptors in a concentration-dependent manner (Fig. 3). Inhibition by 1000 µM niflumate was 74 ± 4% of the GABA-induced current. When 1,000 µM niflumate was applied alone, it elicited a small outward current, lower concentrations being ineffective (data not shown). Furosemide at 100 µM inhibited 3 µM GABA responses similarly to high niflumate concentrations. The traces shown in Fig. 3a suggest some "run-up" of the GABA response during experiments. This was variable between oocytes from different frogs and was not caused by repeated niflumate applications. For example, there was no run-up in the GABA concentration-response experiments (data not shown), when each GABA concentration was applied twice, first without and then with niflumate, in random order of GABA concentrations (see below).

View larger version (24K): [in this window] [in a new window]   Fig. 3. Modulation of GABA-induced currents by niflumate and furosemide in recombinant 122 and 622 GABAA receptors expressed in X. laevis oocytes. a, representative current traces for the receptor subtypes as affected by 3 µM GABA, various niflumate concentrations, and furosemide (F100, 100 µM). b, values are presented as mean ± S.E. (n = 20 and 7 for 122 and 622 combinations, respectively), the 3 µM GABA control response being set to 100%. F, 100 µM furosemide in the presence of 3 µM GABA; N, 1000 µM niflumate alone. Significance of the difference from the corresponding control value (one-way ANOVA followed by Dunnett's post hoc test for niflumate in the presence of GABA, Student's t test for furosemide or niflumate effects alone) as follows: *, p < 0.05; **, p < 0.01. Inset, the GABA concentration-response curves for both receptor subtypes. Dashed line indicates the efficacy of 3 µM GABA. Maximal current peaks elicited by GABA were 3.7 ± 0.6 (n = 14) and 2.3 ± 0.4 µA(n = 6) for 122 and 622 combinations, respectively.

To study the mechanisms of different niflumate effects in oocytes expressing 122 and 622 receptors, the GABA concentration-response curves were determined in the presence and absence of niflumate. Niflumate concentration of 100 µM was chosen for these experiments, because it resulted in clear positive and negative actions on 3 µM GABA response in 122 and 622 receptors, respectively (Fig. 3b). In 122 receptors, niflumate shifted the GABA concentration-response curve to the left (EC₅₀ of 12.5 and 6.3 µM in the absence or presence of niflumate, respectively), without notable change in the efficacy (maximal response 93 ± 5% of the response without niflumate; Fig. 4). In 622 receptors, niflumate reduced robustly the efficacy (E_max 58 ± 3% of the maximal GABA response without niflumate), whereas affinity to GABA was unaltered (EC₅₀ of 2.0 and 2.5 µM in the absence or presence of niflumate, respectively). These results suggest that niflumate acts as a positive allosteric modulator in 122 receptors, and as a noncompetitive negative modulator in 622 receptors.

View larger version (15K): [in this window] [in a new window]   Fig. 4. Effect of 100 µM niflumate on GABA concentration-response curves in recombinant 122 and 622 GABAA receptors expressed in X. laevis oocytes. The values are presented as mean ± S.E. (n = 7 for both receptor subtypes), the maximal GABA-induced currents without niflumate being set to 100%. They were 3.3 ± 0.3 and 2.4 ± 0.2 µA for 122 and 622 receptors, respectively.

To study whether niflumate effects on 122 and 622 receptors were reversible, we applied short niflumate pulses during long GABA applications. In both cases, niflumate elicited fast and reversible effects on GABA currents (Fig. 5).

View larger version (15K): [in this window] [in a new window]   Fig. 5. Reversibility of the niflumate effects on GABA-elicited currents in 122 and 622 GABAA receptors expressed in X. laevis oocytes. In the traces on the left, 3 µM GABA was applied for 90 s, and 100 µM niflumate for 10 s in the middle of GABA application, as indicated by horizontal bars. The traces on the right show the GABA-induced current without niflumate.

Niflumate has been shown to block Ca²⁺-activated Cl^– channels endogenously expressed in X. laevis oocytes (White and Aylwin, 1990). To assess the possible interference of this effect in our experiments, 1,000 µM niflumate was applied to oocytes injected with distilled H₂O. Only a minute outward current was detected (8.4 ± 3.5 nA, n = 8). This was negligible compared with niflumate effects on GABA-induced currents in oocytes expressing GABA_A receptors. Furthermore, our assay being sensitive to positive and negative modulations of niflumate should not involve this current at all, because both 62(2) and 112 pass anions and not allow Ca²⁺ to activate the additional current.

A Segment of 6 Subunit Is Sufficient to Induce Niflumate Antagonism in 122 Receptors. Furosemide antagonism on 6 receptors has been extensively studied using 1/6 subunit chimeras and point mutations. It was first shown with chimeric constructs that the main determinant for furosemide action is located in the N-terminal part of the TM1 of 6 subunit (Fisher et al., 1997; Jackel et al., 1998), where isoleucine at position 228 was later pin-pointed as the crucial amino acid (Thompson et al., 1999). Substitution of a 258-base pair fragment, including the TM1 and TM2 domains of the 1 subunit with that of 6 subunit gene (1-16 chimera), is thus enough to confer furosemide antagonism (Jackel et al., 1998). To study whether the same structural requirements apply for niflumate antagonism, we coexpressed chimeric 1-16 subunit together with wild-type 2 and 2 subunits in X. laevis oocytes. The GABA affinity of the chimeric receptor was between 122 and 622 receptors, EC₅₀ value being 5.1 µM (Fig. 6b, inset). When tested in the presence of 3 µM GABA, instead of potentiation seen in 122 receptors (Fig. 3), niflumate tended to decrease current amplitudes at all concentrations, but only at 1,000 µM the effect was statistically significant (Fig. 6). Furosemide at 100 µM antagonized GABA currents similarly to the highest niflumate concentration used (1000 µM). When applied alone, 1,000 µM niflumate elicited a small outward current.

View larger version (21K): [in this window] [in a new window]   Fig. 6. Modulation of GABA-induced currents by niflumate and furosemide in chimeric 1-1622 GABAA receptors expressed in X. laevis oocytes. a, representative current trace as affected by 3 µM GABA, various niflumate concentrations, and furosemide (F100, 100 µM). b, values are presented as mean ± S.E. (n = 7), the 3 µM GABA control response being set to 100%. F, 100 µM furosemide in the presence of 3 µM GABA; N, 1000 µM niflumate alone. Significance of the difference from the corresponding control value (one-way ANOVA followed by Dunnett's post hoc test for niflumate in the presence of GABA; Student's t test for furosemide or niflumate effects alone) as follows: *, p < 0.05; **, p < 0.01. Inset, the GABA concentration-response curve, dashed line indicates the efficacy of 3 µM GABA. Maximal current elicited by GABA was 2.8 ± 0.2 µA (n = 5).

Niflumate Potentiation Is Dependent on Receptor 2 Subunit. The 62/3 receptors are furosemide-sensitive irrespective of the presence of 2 subunit (Korpi et al., 1995; Korpi and Lüddens, 1997). To test the role of 2 subunit on niflumate actions, 12 and 62 receptors were expressed in X. laevis oocytes. Incorporation of 2 subunit in the receptor complex decreases GABA affinity, which was also seen in our GABA concentration-response studies, as EC₅₀ values for GABA were 1.7 and 0.46 µM for 12 and 62 receptors, respectively (Fig. 7b, inset). Because niflumate was hypothesized to have GABA-enhancing effects in 12 receptors, 1 µM GABA resulting in about 35% of maximal response was used. When niflumate was applied together with 1 µM GABA to 12 receptors, it was only active at the 1,000 µM concentration, which inhibited the GABA response by 29 ± 5% (Fig. 7). When applied alone, the same concentration of niflumate was inactive, as was furosemide. GABA at 3 µM induced about 90% of maximal response in 62 receptors and was suitable for antagonist studies. When niflumate was applied with 3 µM GABA on 62 receptors, concentration-dependent potent antagonism was evident (Fig. 7). Furosemide at 100 µM inhibited the GABA currents, and 1,000 µM niflumate alone elicited a small outward current.

View larger version (22K): [in this window] [in a new window]   Fig. 7. Modulation of GABA-induced currents by niflumate and furosemide in 12 and 62 recombinant GABAA receptors expressed in X. laevis oocytes. a, representative current traces for different receptor subtypes as affected by GABA, various niflumate concentrations, and furosemide (F100, 100 µM). b, values are presented as mean ± S.E. (n = 13 and 8 for 12 and 62 combinations, respectively), the GABA control response being set to 100%. F, 100 µM furosemide in the presence of GABA; N, 1000 µM niflumate alone. Significance of the difference from the corresponding control value (one-way ANOVA followed by Dunnett's post hoc test for niflumate in the presence of GABA; Student's t test for furosemide or niflumate effects alone) as follows: *, p < 0.05; **, p < 0.01; ***, p < 0.001. Inset, the GABA concentration-response curves for both receptor subtypes. Dashed line indicates the efficacy of GABA concentrations, which were used in the following experiments. Maximal current peaks elicited by GABA were 3.4 ± 0.4 (n = 9) and 1.1 ± 0.2 µA (n = 8) for 12 and 62 combinations, respectively.

Flumazenil Does Not Affect Niflumate Potentiation of 122 Receptors. Benzodiazepine potentiation of GABA_A receptors is dependent on the presence of a suitable subunit and the 2 subunit in the receptor complex (Pritchett et al., 1989). Because positive modulation by niflumate was present in 1 and 2 subunit-containing receptors, it was of interest to study, whether it would be mediated by the benzodiazepine binding site. We applied the benzodiazepine site positive modulator zolpidem at 1 µM concentration on 122 receptors together with 3 µM GABA. This resulted in 39 ± 6% potentiation of the GABA-induced current, which was totally blocked by 1 µM flumazenil, a benzodiazepine site antagonist (Fig. 8). When flumazenil was applied together with 100 µM niflumate and 3 µM GABA, it failed to affect the niflumate potentiation of the GABA-induced current (38 ± 4 versus 36 ± 6%).

View larger version (44K): [in this window] [in a new window]   Fig. 8. Potentiation of 3 µM GABA-induced currents by zolpidem (Zol, 1 µM) and niflumate (N, 100 µM), and their modulation by flumazenil (Flu, 1 µM) in 122 recombinant GABAA receptors expressed in X. laevis oocytes. Values are presented as mean ± S.E. (n = 8 and 13 for zolpidem and niflumate effects, respectively). ***, p < 0.001 for the statistical significance of the flumazenil effect on the zolpidem potentiation (paired t test).

	Discussion

Top Abstract Materials and Methods Results Discussion References

 
GABA_A receptor is well known for its many binding and effector sites for various chemical classes of ligands (Sieghart, 1995; Korpi et al., 2002). Many ligands may have more than one binding/effector site on the receptor complex, mediating, for example, GABA-potentiating and direct "GABA site-independent" agonistic effects on the GABA_A receptor. Good examples are intravenous anesthetics, such as propofol and pentobarbital (Inomata et al., 1988; Hales and Lambert, 1991; Korpi et al., 2002). Fewer compounds have been demonstrated to have distinct binding/effector sites mediating attenuation and potentiation of GABA responses. The -carboline negative allosteric modulator methyl-6,7-dimethoxy-4-ethyl--carboline-3-carboxylate has a potent inhibitory action via the 2 subunit-dependent benzodiazepine site and at higher concentrations a stimulatory "positive allosteric" effect via the coupling mechanism dependent on 2 and 3 subunits (Stevenson et al., 1995). The latter site is also involved in the antagonism of the GABA responses by furosemide (Thompson et al., 1999), this effect being also dependent on the 6 and 4 subunits (Korpi et al., 1995; Knoflach et al., 1996; Wafford et al., 1996). The present study reports on a fenamate NSAID, niflumate, that is a positive or negative modulator of the GABA responses, depending on the receptor subunit combination in both native and recombinant receptors.

The native receptors were studied using autoradiography of rat brain sections with picrotoxinin-sensitive [³⁵S]TBPS binding to the GABA_A receptor convulsant binding site. Without added GABA, niflumate decreased [³⁵S]TBPS binding in all brain areas (Fig. 2; Table 1), in keeping with previous binding studies (Evoniuk and Skolnick, 1988; Maksay et al., 1998). However, even without added GABA, brain sections containing endogenous GABA affecting [³⁵S]TBPS binding, which can be antagonized by GABA site antagonists (Korpi et al., 1992). It has been estimated that the endogenous GABA concentration reaches 1 µM in synaptosomal membranes (Im and Blakeman, 1991), and it is unlikely that the 15-min preincubation used here removes it completely. Thus, these niflumate effects might be interpreted as positive modulation of GABA effects rather than as independent "direct" action. This is in line with the negligible action of niflumate on recombinant receptors expressed in oocytes in the absence of exogenous GABA. These conclusions would imply that the niflumate potentiation of GABA action is a pure allosteric mechanism.

In search for structural correlates of the positive modulation by niflumate, we studied recombinant GABA_A receptors composed of 1, 2, and 2 subunits, the main receptor subtype in the brain (Whiting et al., 1995). The EC₅₀ for GABA 122 receptors was reduced by about 50%, indicating an allosteric potentiation by niflumate (Fig. 3). Furosemide did not consistently potentiate GABA, in keeping with Korpi et al. (1995). However, the 12 receptors were insensitive to niflumate up to 100 µM, and at 1,000 µM niflumate inhibited the GABA response by 30% in contrast to the potentiation of 122 receptors (Fig. 7). This suggests that niflumate potentiation is dependent on the presence of 2 subunit together with 1. The same subunit requirement is obligatory for the benzodiazepine binding site modulators, such as zolpidem (Pritchett et al., 1989). However, zolpidem and niflumate effects were differentially sensitive to the benzodiazepine binding site antagonist flumazenil in 122 receptors, because only the GABA potentiation by zolpidem was blocked by 1 µM flumazenil (Fig. 8), a concentration that saturates the benzodiazepine binding sites (K_D 1nMin 1x2 receptors; Pritchett and Seeburg, 1990). These data suggest that niflumate potentiation is mediated via a novel site of the GABA_A receptors dependent on , , and subunits.

Another fenamate group NSAID, mefenamic acid, was recently shown to potentiate GABA on 1 and 2/3 subunit-containing receptors, but to be inactive or inhibitory in 1 subunit-containing receptors (Halliwell et al., 1999). The GABA potentiating action was dependent on the asparagine residue of TM2 in 2 and 3 subunits (Halliwell et al., 1999), which is the critical residue for furosemide and methyl-6,7-dimethoxy-4-ethyl--carboline-3-carboxylate modulations as well (Stevenson et al., 1995; Thompson et al., 1999). Whether niflumate shows similar selectivity for the subunits is still open, but as its potentiating effect needs the 2 subunits, the mode of action is clearly different from that of the ligands solely needing subunits interacting with the 2 or 3 subunits and requiring specific feature in the TM2 domain.

Niflumate modulated the inhibition of [³⁵S]TBPS binding by 3 µM GABA depending on the brain region (Fig. 2; Table 2). Whereas positive modulation predominated in most brain areas, the negative modulation (antagonism) was evident in the cerebellar granule cell layer. When the cerebellum was analyzed as whole, the profound enhancement of [³⁵S]TBPS binding in the granule cell layer (GABA antagonism) was masked by reduced binding in the molecular layer (GABA potentiation), even with 3 µM GABA that inhibited the binding in the whole cerebellum already to 11% of basal binding. Our results on the whole cerebellum agree with the previous results on cerebellar membranes (Maksay et al., 1998), but they demonstrate the advantage gained by the spatial resolution when working with brain sections in comparison with brain homogenates.

GABA at 1 mM reveals an atypical GABA-insensitive [³⁵S]TBPS binding component in brain sections (Fig. 2), constituting a minor fraction of basal binding in selected brain regions (Sinkkonen et al., 2001). Niflumate greatly antagonized 1 mM GABA in the cerebellar granule cell layer and to a lesser degree in the thalamus (Table 3), whereas hardly altering anything in other regions up to 1,000 µM niflumate, which reduced the binding. Furosemide had similar effects, but it was more efficient in the cerebellar granule cell layer than thalamus. Furosemide antagonism in the cerebellar granule cell layer is related to the granule cell-restricted 6 subunit (Korpi et al., 1995). Using recombinant receptors, furosemide antagonism has also been observed in thalamus-enriched 4 subunit-containing receptors (Wisden et al., 1992; Knoflach et al., 1996; Wafford et al., 1996). A smaller furosemide antagonism was observed there now, although it was not detected in the thalamus previously, when a less sensitive [³⁵S]TBPS binding autoradiography was performed (Korpi and Lüddens, 1997), possibly because 4 subunit-containing receptors constitute only 20 to 27% of the thalamic receptors (Sur et al., 1999). These results suggest that niflumate antagonism is dependent on the presence of 6or 4 subunits.

Niflumate antagonized 3-µM GABA-elicited currents in a concentration-dependent manner in 62 and 622 combinations (Figs. 3 and 7). In 622 receptors, niflumate failed to alter the EC₅₀ value for GABA, but reduced the maximal GABA currents by about 40%. This is consistent with noncompetitive antagonism, which has been shown also for furosemide (Korpi et al., 1995). To study the site of niflumate antagonism more precisely, we coexpressed the 1-16 chimera, which has been shown to be furosemide-sensitive (Jackel et al., 1998), with 2 and 2 subunits. These receptors were also inhibited by niflumate (Fig. 6), supporting the idea that furosemide and niflumate share a binding/effector site on 6 subunit. Interestingly, the negative modulatory actions were not saturable, but fully blocked the GABA effects. Indeed, especially in 62 receptors high niflumate concentrations produced small outward, as opposed to GABA-induced inward, currents even in the absence of GABA. We believe this outward current is a small leakage current produced by spontaneously open channels, which has been suggested for homomeric 1 and 3 receptors (Sigel et al., 1989; Wooltorton et al., 1997). This property might have obscured the kinetics of drug action and prevented us from calculating IC₅₀ values. However, as can be seen from the present data of both ligand binding and electrophysiology experiments, niflumate concentrations needed for positive and negative effects were similar and already low micromolar ones were significantly effective.

Niflumate is clinically used in Europe. The usual clinical oral dosage of 250 mg of three times daily results in steadystate plasma concentrations of 20 to 70 µM (Houin et al., 1983). Although most of the drug is bound to plasma proteins, free niflumate concentrations in plasma reach the micromolar range. Because fenamates pass the blood-brain barrier efficiently (Bannwarth et al., 1989), niflumate can be estimated to reach micromolar concentrations in the brain. NSAID overdoses have been shown to result in severe central effects, such as convulsion and coma (Smolinske et al., 1990). Thus, some adverse effects of niflumate and other fenamates could be mediated by alterations of GABA_A receptor function.

In conclusion, niflumate has positive-negative modulatory profile on both native and recombinant GABA_A receptors depending on receptor subtype at concentrations that may be clinically or toxicologically relevant. Niflumate acts as a positive allosteric modulator on 122 and as a negative modulator on 62 and 622 (and 12) GABA_A receptors. The noncompetitive antagonistic action of niflumate is mediated by the same site as the furosemide action, whereas the site for the positive allosteric modulator action depends on the presence of the 2 subunit, but is different from the benzodiazepine binding site and remains to be characterized in detail. The present findings add to the diverse list of structural requirements of GABA_A receptor ligands and might offer a novel lead for subtype-selective drug development with a potential of having forebrain-preferring GABA potentiation without cerebellum-related motor impairment.

	Acknowledgements

 
We thank Eija Lehtovirta for technical assistance and Sirpa Lehti-Koivunen for maintenance of the frogs.

	Footnotes

 
This study was partially supported by funds from the Academy of Finland (to E.R.K.).

ABBREVIATIONS: TBPS, t-butylbicyclophosphorothionate; EBOB, 4'-ethynyl-4-n[2,3-H₂]propyl-bicycloorthobenzoate; NSAID, nonsteroidal anti-inflammatory drug; TM, transmembrane; NFR, normal frog Ringer; ANOVA, analysis of variance.

Address correspondence to: Dr. Esa R. Korpi, Institute of Biomedicine, Pharmacology, Biomedicum Helsinki, P.O. Box 63, University of Helsinki, Helsinki FIN-00014, Finland. E-mail: esa.korpi{at}helsinki.fiy

	References

Top Abstract Materials and Methods Results Discussion References

 
Bannwarth B, Netter P, Pourel J, Royer RJ, and Gaucher A (1989) Clinical pharmacokinetics of nonsteroidal anti-inflammatory drugs in the cerebrospinal fluid. Biomed Pharmacother 43: 121–126.[CrossRef][Medline]

Barnard EA, Skolnick P, Olsen RW, Mohler H, Sieghart W, Biggio G, Braestrup C, Bateson AN, and Langer SZ (1998) International Union of Pharmacology. XV. Subtypes of -aminobutyric acid_A receptors: classification on the basis of subunit structure and receptor function. Pharmacol Rev 50: 291–313.[Abstract/Free Full Text]

Bonnert TP, McKernan RM, Farrar S, le Bourdelles B, Heavens RP, Smith DW, Hewson L, Rigby MR, Sirinathsinghji DJ, Brown N, et al. (1999) , a novel -aminobutyric acid type A receptor subunit. Proc Natl Acad Sci USA 96: 9891–9896.[Abstract/Free Full Text]

Cushman DW and Cheung HS (1976) Effect of substrate concentration on inhibition of prostaglandin synthetase of bull seminal vesicles by anti-inflammatory drugs and fenamic acid analogs. Biochim Biophys Acta 424: 449–459.[Medline]

Evoniuk G and Skolnick P (1988) Picrate and niflumate block anion modulation of radioligand binding to the -aminobutyric acid/benzodiazepine receptor complex. Mol Pharmacol 34: 837–842.[Abstract]

Fisher JL, Zhang J, and Macdonald RL (1997) The role of 1 and 6 subtype amino-terminal domains in allosteric regulation of -aminobutyric acid_A receptors. Mol Pharmacol 52: 714–724.[Abstract/Free Full Text]

Hales TG and Lambert JJ (1991) The actions of propofol on inhibitory amino acid receptors of bovine adrenomedullary chromaffin cells and rodent central neurones. Br J Pharmacol 104: 619–628.[Medline]

Halliwell RF, Thomas P, Patten D, James CH, Martinez-Torres A, Miledi R, and Smart TG (1999) Subunit-selective modulation of GABA_A receptors by the nonsteroidal anti-inflammatory agent, mefenamic acid. Eur J Neurosci 11: 2897–2905.[CrossRef][Medline]

Houin G, Tremblay D, Bree F, Dufour A, Ledudal P, and Tillement JP (1983) The pharmacokinetics and availability of niflumic acid in humans. Int J Clin Pharmacol Ther Toxicol 21: 130–134.[Medline]

Im WB and Blakeman DP (1991) Correlation between -aminobutyric acid_A receptor ligand-induced changes in t-butylbicyclophosphoro[³⁵S]thionate binding and ³⁶Cl^– uptake in rat cerebrocortical membranes. Mol Pharmacol 39: 394–398.[Abstract]

Inomata N, Tokutomi N, Oyama Y, and Akaike N (1988) Intracellular picrotoxin blocks pentobarbital-gated Cl^– conductance. Neurosci Res 6: 72–75.[CrossRef][Medline]

Jackel C, Kleinz R, Mäkelä R, Hevers W, Jezequel S, Korpi ER, and Lüddens H (1998) The main determinant of furosemide inhibition on GABA_A receptors is located close to the first transmembrane domain. Eur J Pharmacol 357: 251–256.[CrossRef][Medline]

Knoflach F, Benke D, Wang Y, Scheurer L, Lüddens H, Hamilton BJ, Carter DB, Möhler H, and Benson JA (1996) Pharmacological modulation of the diazepam-insensitive recombinant -aminobutyric acid_A receptors 422 and 622. Mol Pharmacol 50: 1253–1261.[Abstract]

Korpi ER, Gründer G, and Lüddens H (2002) Drug interactions at GABA_A receptors. Prog Neurobiol 67: 113–159.[CrossRef][Medline]

Korpi ER, Kuner T, Seeburg PH, and Lüddens H (1995) Selective antagonist for the cerebellar granule cell-specific -aminobutyric acid type A receptor. Mol Pharmacol 47: 283–289.[Abstract]

Korpi ER and Lüddens H (1997) Furosemide interactions with brain GABA_A receptors. Br J Pharmacol 120: 741–748.[CrossRef][Medline]

Korpi ER, Lüddens H, and Seeburg PH (1992) GABA_A antagonists reveal binding sites for [³⁵S]TBPS in cerebellar granular cell layer. Eur J Pharmacol 211: 427–428.[CrossRef][Medline]

Lüddens H, Pritchett DB, Köhler M, Killisch I, Keinänen K, Monyer H, Sprengel R, and Seeburg PH (1990) Cerebellar GABA_A receptor selective for a behavioural alcohol antagonist. Nature (Lond) 346: 648–651.[CrossRef][Medline]

Maksay G and Ticku MK (1985) Dissociation of [³⁵S]t-butylbicyclophosphorothionate binding differentiates convulsant and depressant drugs that modulate GABAergic transmission. J Neurochem 44: 480–486.[CrossRef][Medline]

Maksay G, Korpi ER, and Uusi-Oukari M (1998) Bimodal action of furosemide on convulsant [³H]EBOB binding to cerebellar and cortical GABA_A receptors. Neurochem Int 33: 353–358.[CrossRef][Medline]

Pritchett DB and Seeburg PH (1990) -Aminobutyric acid_A receptor 5-subunit creates novel type II benzodiazepine receptor pharmacology. J Neurochem 54: 1802–1804.[Medline]

Pritchett DB, Sontheimer H, Shivers BD, Ymer S, Kettenmann H, Schofield PR, and Seeburg PH (1989) Importance of a novel GABA_A receptor subunit for benzodiazepine pharmacology. Nature (Lond) 338: 582–585.[CrossRef][Medline]

Shivers BD, Killisch I, Sprengel R, Sontheimer H, Köhler M, Schofield PR, and Seeburg PH (1989) Two novel GABA_A receptor subunits exist in distinct neuronal subpopulations. Neuron 3: 327–337.[CrossRef][Medline]

Sieghart W (1995) Structure and pharmacology of -aminobutyric acid_A receptor subtypes. Pharmacol Rev 47: 181–234.[Medline]

Sigel E, Baur R, Malherbe P, and Mohler H (1989) The rat 1-subunit of the GABA_A receptor forms a picrotoxin-sensitive anion channel open in the absence of GABA. FEBS Lett 257: 377–379.[CrossRef][Medline]

Sinkkonen ST, Hanna MC, Kirkness EF, and Korpi ER (2000) GABA_A receptor and subunits display unusual structural variation between species and are enriched in the rat locus ceruleus. J Neurosci 20: 3588–3595.[Abstract/Free Full Text]

Sinkkonen ST, Uusi-Oukari M, Tupala E, Sarkioja T, Tiihonen J, Panula P, Lüddens H, and Korpi ER (2001) Characterization of -aminobutyrate type A receptors with atypical coupling between agonist and convulsant binding sites in discrete brain regions. Brain Res Mol Brain Res 86: 168–178.[Medline]

Smolinske SC, Hall AH, Vandenberg SA, Spoerke DG, and McBride PV (1990) Toxic effects of nonsteroidal anti-inflammatory drugs in overdose. An overview of recent evidence on clinical effects and dose-response relationships. Drug Saf 5: 252–274.[Medline]

Squires RF, Casida JE, Richardson M, and Saederup E (1983) [³⁵S]t-Butylbicyclophosphorothionate binds with high affinity to brain-specific sites coupled to -aminobutyric acid-A and ion recognition sites. Mol Pharmacol 23: 326–336.[Abstract]

Squires RF and Saederup E (1987) GABA_A receptor blockers reverse the inhibitory effect of GABA on brain-specific [³⁵S]TBPS binding. Brain Res 414: 357–364.[CrossRef][Medline]

Stevenson A, Wingrove PB, Whiting PJ, and Wafford KA (1995) -Carboline -aminobutyric acid_A receptor inverse agonists modulate -aminobutyric acid via the loreclezole binding site as well as the benzodiazepine site. Mol Pharmacol 48: 965–969.[Abstract]

Sur C, Farrar SJ, Kerby J, Whiting PJ, Atack JR, and McKernan RM (1999) Preferential coassembly of 4 and subunits of the -aminobutyric acid_A receptor in rat thalamus. Mol Pharmacol 56: 110–115.[Abstract/Free Full Text]

Thompson SA, Arden SA, Marshall G, Wingrove PB, Whiting PJ, and Wafford KA (1999) Residues in transmembrane domains I and II determine gamma-aminobutyric acid type A receptor subtype-selective antagonism by furosemide. Mol Pharmacol 55: 993–999.[Abstract/Free Full Text]

Wafford KA, Thompson SA, Thomas D, Sikela J, Wilcox AS, and Whiting PJ (1996) Functional characterization of human -aminobutyric acid_A receptors containing the 4 subunit. Mol Pharmacol 50: 670–678.[Abstract]

Vane JR and Botting RM (1998) Mechanism of action of nonsteroidal anti-inflammatory drugs. Am J Med 104: 2S–8S.[Medline]

White MM and Aylwin M (1990) Niflumic and flufenamic acids are potent reversible blockers of Ca²⁺-activated Cl^– channels in Xenopus oocytes. Mol Pharmacol 37: 720–724.[Abstract]

Whiting PJ, McKernan RM, and Wafford KA (1995) Structure and pharmacology of vertebrate GABA_A receptor subtypes. Int Rev Neurobiol 38: 95–138.[Medline]

Wisden W, Laurie DJ, Monyer H, and Seeburg PH (1992) The distribution of 13 GABA_A receptor subunit mRNAs in the rat brain. I. Telencephalon, diencephalon, mesencephalon. J Neurosci 12: 1040–1062.[Abstract]

Woodward RM, Polenzani L, and Miledi R (1994) Effects of fenamates and other nonsteroidal anti-inflammatory drugs on rat brain GABA_A receptors expressed in Xenopus oocytes. J Pharmacol Exp Ther 268: 806–817.[Abstract/Free Full Text]

Wooltorton JR, Moss SJ, and Smart TG (1997) Pharmacological and physiological characterization of murine homomeric 3 GABA_A receptors. Eur J Neurosci 9: 2225–2235.[CrossRef][Medline]

Ymer S, Schofield PR, Draguhn A, Werner P, Köhler M, and Seeburg PH (1989) GABA_A receptor subunit heterogeneity: functional expression of cloned cDNAs. EMBO (Eur Mol Biol Organ) J 8: 1665–1670.[Medline]

This article has been cited by other articles:

				L. Cheng and M. C. Sanguinetti Niflumic Acid Alters Gating of HCN2 Pacemaker Channels by Interaction with the Outer Region of S4 Voltage Sensing Domains Mol. Pharmacol., May 1, 2009; 75(5): 1210 - 1221. [Abstract] [Full Text] [PDF]

				Y. Yuan and W. D. Atchison Methylmercury Induces a Spontaneous, Transient Slow Inward Chloride Current in Purkinje Cells of Rat Cerebellar Slices J. Pharmacol. Exp. Ther., May 1, 2005; 313(2): 751 - 764. [Abstract] [Full Text] [PDF]

				A. J. Smith, B. Oxley, S. Malpas, G. V. Pillai, and P. B. Simpson Compounds Exhibiting Selective Efficacy for Different {beta} Subunits of Human Recombinant {gamma}-Aminobutyric AcidA Receptors J. Pharmacol. Exp. Ther., November 1, 2004; 311(2): 601 - 609. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services 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 Sinkkonen, S. T. Articles by Korpi, E. R. Search for Related Content PubMed PubMed Citation Articles by Sinkkonen, S. T. Articles by Korpi, E. R.