Modulation of the voltage-dependent sodium channel by agents affecting G-proteins: a study in Xenopus oocytes injected with brain RNA

The effects of agents known to affect G-proteins on voltage-dependent, tetrodoxin-sensitive Na⁺ channels were studied in Xenopus oocytes injected with rat brain RNA, using two-electrode voltage-clamp technique. The non-hydrolysable analogue of GTP, GTP-γ-S, Known to activate G-proteins, inhibited the Na⁺ current (I_Na). The decrease in the amplitude of I_Na was not accompanied by changes in activation or inactivation characteristics of the channel. The non-hydrolysable analogue of GDP, GDP-β-S, had no effect on I_Na. The responses to γ-aminobutyric acid and kainate in the same oocytes were also attenuated by GTP-γ-S. Pertussis toxin, which inactivates some G-proteins by catalyzing thier ADP-ribosylation, enhanced I_Na, but did not prevent the inhibition of I_Na by GTP-γ-S. We conclude that the Na⁺ channel, and possibly the GABA and kainate receptors and/or channels, are coupled to a G-protein. The activation of the G-protein modulates the channels either directly, or via activation of biochemical cascade possibly involving production of second messengers and channel phosphorylation.     

Cytoskeleton mediates inhibition of the fast Na+ current in respiratory brainstem neurons during hypoxia

Whole-cell Na+ currents (INa) were recorded in inspiratory neurons in a medullary slice preparation from neonatal mouse that contains the functional respiratory network. Hypoxia and metabolic poisoning with KCN rapidly inhibited INa by reducing the number of Na+ channels available for opening during depolarization. Application of agents specific for G-proteins, protein kinase C and A, intracellular Ca2+ and pH did not prevent the hypoxic inhibition of INa. The effects of hypo-osmolarity and hypoxia were additive, whereas hyperosmolarity partially prevented a subsequent hypoxic inhibition of INa. Cytochalasin B and colchicine decreased, and taxol or phalloidin increased INa and reduced its hypoxic inhibition. We conclude that cytoskeleton rearrangements during hypoxia are responsible for suppression of a fast INa in brainstem respiratory neurons, which could be mediated by the uncoupling of channel inactivation gates from cytoskeletal elements.     

AMPA receptor activation induces association of G-beta protein with the alpha subunit of the sodium channel in neurons

Glutamatergic transmission is mediated by ionotropic receptors that directly gate cationic channels and metabotropic receptors that are coupled to second messenger generating systems and to ionic channels via heterotrimeric guanine-nucleotide binding- (G) proteins. This distinction cannot be made for the ionotropic receptor subclass activated by alpha-amino-3-hydroxy-5-methylisoxazole-4-propionic acid (AMPA), which has been shown to be physically associated with the alpha-subunit of Gi1 protein and activates this G-protein. Here, we report that, in addition to a Ca2+ influx, AMPA induces the mobilization of Ca2+ from the mitochondrial pool by reversing the mitochondrial Na+/Ca2+ exchanger in mouse neurons in primary culture. Both processes required the activation of tetrodotoxin-sensitive Na+ channels. AMPA receptor activation modified the gating properties of the Na+ channel, independently of the AMPA current, suggesting a G-protein-mediated process. Indeed, co-immunoprecipitation experiments indicated that AMPA receptor activation induced the association of Gbeta with the alpha-subunit of the Na+ channel. These results suggest that, in addition to its ionic channel function, the AMPA receptor is coupled to Na+ channels through G-proteins and that this novel metabotropic function is involved in the control of neuronal excitability.     

Dopamine modulates voltage-activated potassium currents in zebrafish retinal on bipolar cells

We report a study of the characterization of voltage-activated potassium (K+) currents in retinal ON bipolar cells in zebrafish. At single-channels levels, the open probability of the K+ channels increased when the membrane potential was increased. The maximal open proportion was 0.76+/-0.05 under our testing conditions. In whole-cell recordings, the K+ current displayed two exponential components with the activation time constants of 11-22 msec (tau1) and 0.8-4 msec (tau2). Dopamine modulated the K+ current. Dopamine reduced the time constant tau2 when the membrane potential was depolarized to high voltages. A decrease in K+ current was seen when dopamine D1 receptors were selectively activated by SKF38393 or when the D1 receptor-coupled G-proteins were activated by GTP-gamma-S. The activation of adenylate cyclase by forskolin or the increase of intracellular cAMP concentrations by 8-Br-cAMP or Sp-cAMPS also resulted in a decrease in K+ current. Together, the data suggest that dopamine modulates the K+ current via D1 receptor-coupled G-protein pathways.     

Inhibition of calcium channels in rat CA3 pyramidal neurons by a metabotropic glutamate receptor.

L-Glutamate rapidly and reversibly suppressed Ca channel current in freshly dissociated pyramidal neurons from the CA3 region of the rat hippocampus. L-Glutamate inhibition of Ca channel current could be distinguished from activation of background conductance by appropriate ionic conditions and by distinct pharmacological profiles. Ca channel inhibition by glutamate was mimicked by quisqualate, ibotenate, racem?ct-ACPD and 1S,3R-ACPD but not by kainate, AMPA, L-aspartate, NMDA, L-2-amino-4-phosphonobutyric acid, or 1R,3S-ACPD; 6-cyano-7-nitroquinoxaline-2,3-dione did not inhibit the response. All agonists inhibited a similar fraction of high-voltage-activated Ca channel current, typically approximately 30%. Concentration-response relations for the agonists were consistent with mediation by a metabotropic glutamate receptor. The stereospecific agonist 1S,3R-ACPD was especially useful since it did not activate background conductances. The fraction of Ca channel current sensitive to 1S,3R-ACPD was partially blocked by omega-conotoxin GVIA but was not sensitive to dihydropyridine antagonists or agonists. The suppression of Ca channels by 1S,3R-ACPD became irreversible when cells were dialyzed with GTP-gamma-S. 1S,3R-ACPD suppressed Ca channel currents in outside-out membrane patches but not in cell-attached patches when applied outside the patch. These results suggest that metabotropic glutamate receptors suppress the activity of N-type Ca channels in CA3 neurons by a mechanism involving G-proteins but not readily diffusible second messengers.     

Glial cells of the oligodendrocyte lineage express proton-activated Na+ channels

Neurons and oligodendrocytes, but not type I astrocytes and Schwann cells, generate large Na+ currents in response to a step increase of [H+]. Proton-activated Na+ channels are the first cationic channels expressed in neuronal precursor cells from the mammalian brain. Glial precursor cells cultured from mouse brain are also capable of generating Na+ currents in response to step acidification (INa(H]. With further development along the oligodendrocyte lineage, this property is retained, whereas voltage-activated Na+ and K+ currents disappear. Comparing INa(H) of oligodendrocytes with INa(H) of their precursor cells did not reveal a difference in current amplitude, suggesting a higher density of INa(H) channels on the (smaller) precursor cells. The properties of INa(H) in glial precursor cells and oligodendrocytes are similar to those of neurons, with respect to activation conditions, time course, and the effect of extracellular Ca2+ concentrations. The results are consistent with previous observations which showed that oligodendrocytes partially preserve their chemically activated, but completely lose their voltage-activated, ion channels.     

Protein kinase C modulates neurotransmitter responses in Xenopus oocytes injected with rat brain RNA

Oocytes of the frog Xenopus laevis express various exogenous neurotransmitter receptors and ion channels when injected with RNA from excitable tissues. The oocytes serve as a convenient model system in which modulation of neurotransmitter responses can be studied. We examined the effects of activators and an inhibitor of protein kinase C (PKC) on responses to serotonin (5-HT), acetylcholine (ACh), kainate, and gamma-aminobutyric acid (GABA) in oocytes injected with RNA from rat brain. The PKC activators beta-phorbol esters 4 beta-phorbol-12-myristate-13-acetate (PMA) and 4 beta-phorbol-12,13-dibutyrate (PDBu), as well as the synthetic diacylglycerol, 1-oleyl-2-acetylglycerol (OAG), significantly inhibited the responses to 5-HT and ACh (both known to be mediated by mobilization of intracellular Ca2+); the first (transient) phase of these responses was affected stronger than the second, slow phase. PKC activators also reduced the response to GABA. The effect of PDBu on the response to kainate was dual; either inhibition or potentiation were observed at different concentrations of PDBu. The inactive analogue of PMA, the alpha-PMA, was without effect on the responses to 5-HT and GABA. The PKC inhibitor 1,5-isoquinolinesulfonyl-2-methylpiperazine (H7) suppressed the inhibitory effect of PDBu on 5-HT response. Amiloride, a blocker of the Na+/H+ exchange (which is known to be activated by PKC in some tissues), did not suppress the effects of PDBu. We concluded that activation of PKC down-regulates the responses to 5-HT, ACh and GABA, and has a dual effect on response to kainate. Possible mechanisms of these effects are discussed.     

Noise of the slowly inactivating Na current in suprachiasmatic nucleus neurons

Slow depolarization induces a riluzole-sensitive persistent Na current (INa,P) in several types of neurons and a pharmacologically similar slowly inactivating Na current (INa,S) in rat suprachiasmatic nucleus neurons. In isolated suprachiasmatic nucleus neurons, INa,S fluctuations were analyzed to characterize the Na channel responsible for INa,S. The single-channel current near resting potential was -0.53+/-0.04 pA in an external solution containing 151 mM Na+ and 5 mM Na+ in the patch pipette. Tetrodotoxin completely blocked INa,S and single-channel current noise; 25 microM riluzole also suppressed INa,S and current noise by about 80% without a significant effect on mean single-channel current. The data suggest that single-channel noise is due to the opening of channels mediating INa,S with a conductance of about 3.4 pS.     

Inhibition of delayed rectifier K+ conductance in cultured rat cerebellar granule neurons by activation of calcium-permeable AMPA receptors

Activation of AMPA (alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid) receptors in cerebellar granule cells during perforated-patch whole-cell recordings activated an inward current at negative voltages which was followed, after a delay, by the inhibition of an outward potassium current at voltages positive to -20 mV. The activated inward current was inwardly rectifying suggesting that the AMPA receptors were Ca2+-permeable. This was confirmed by direct measurements of intracellular calcium where Ca2+ rises were seen following AMPA receptor activation in Na+-free external solution. Ca2+ rises were equally large in the presence of 100 microM Cd2+ to block voltage-gated Ca2+ channels. Specific voltage-protocols, allowing selective activation of the delayed rectifier potassium current (KV) and the transient A current (KA), showed that kainate inhibited KV, but not to any great extent KA. The inhibition of KV was blocked by the AMPA receptor antagonist CNQX (6-cyano-7-nitroquinoxaline-2,3-dione) and was no longer observed when the KV current was abolished with high concentrations of Ba2+. The responses to kainate were not altered by pre-treating the cells with pertussis toxin, suggesting that the AMPA receptor stimulation of the G-protein Gi cannot account for the effects observed. Replacing extracellular Na+ with choline did not alter the inhibition of KV by kainate, however, removing extracellular Ca2+ reduced the kainate response. The inhibition of KV by kainate was unaffected by the presence of 100 microM Cd2+. The guanylyl cyclase inhibitor, ODQ (1H-[1,2,4]oxadiazolo[4,3-a]quinoxalin-1-one), did not alter kainate inhibition of KV. It is concluded that ion influx (particularly Ca2+ ions) through AMPA receptor channels following receptor activation leads to an inhibition of KV currents in cerebellar granule neurons.     

Effect of tunicamycin on the expression of functional brain neurotransmitter receptors and voltage-operated channels in Xenopus oocytes

The role of N-glycosylation on the expression of functional brain neurotransmitter receptors and voltage-operated channels was studied by injecting Xenopus oocytes with mRNA from rat brain or chick optic lobe, and culturing them in the presence or absence of tunicamycin, an inhibitor of asparagine linked glycosylation. Electrophysiological recordings were then made to assess the amounts of functional receptors and channels present in the oocyte membrane. The appearance of gamma-aminobutyric acid (GABA) receptors and voltage-activated Na+ channels was profoundly reduced. In contrast, the functional expression of kainate receptors, and voltage-activated K+ and Ca2+ channels was much less affected. Thus, it seems that kainate receptors, and K+ and Ca2+ channels can be expressed and function normally without being glycosylated. On the other hand, GABA receptors and Na+ channels may need to be N-glycosylated in order to function properly, or to ensure their correct insertion into the membrane.     

Messenger RNAs coding for receptors and channels in the cerebral cortex of adult and aged rats.

Poly(A)+ mRNAs from the cerebral cortex of aged (24 months) and young adult (3 months) rats were isolated and injected into Xenopus oocytes to express functional neurotransmitter receptors and voltage-operated channels. Electrophysiological recordings of induced membrane currents were used as a measure of the relative amounts of mRNA encoding different receptors and channels, and to study their functional properties. There were no large differences apparent between mRNAs from aged and adult rats, in marked contrast to the dramatic (1000-fold) changes in mRNA expression that occur during embryonic and postnatal development. The membrane currents induced by glutamate or acetylcholine (ACh) application were roughly one third smaller in oocytes injected with mRNA from aged cerebral cortex than in oocytes injected with mRNA from adult cerebral cortex, whereas currents induced by gamma-aminobutyric acid (GABA), kainate or serotonin (5-HT) application, and by activation of voltage-operated Na+ and Ca2+ channels were not significantly different. We did not observe any age-related differences in the properties of the receptors and channels studied.     

Effect of lithium on rod photoreceptor rhodopsin-coupled G-protein (transducin).

1. Lithium was known to inhibit both adrenergic and cholinergic agonist-induced activation of G-proteins in cerebral cortex. 2. Doly et al (1989) observed that lithium reduced the b-wave of the electroretinogram and suggested that the effect was due to inhibition of G-protein in photoreceptor cells. 3. This study was undertaken to test this hypothesis directly on photoreceptor cell membranes. Rod disk membranes containing the visual transduction machinery were isolated in the dark and the effect of lithium was tested on (a) activation by bright light of G-protein-mediated cyclic GMP hydrolysis, (b) light sensitivity of the activation, (c) the lifetime of the light-activated receptor, and (d) light activation of GTP-gamma-S binding to the membranes. 4. None of these processes were affected by lithium. It is therefore concluded that the effects of lithium on the b-wave of the electroretinogram should be due to influences on G-proteins in other parts of the retina.     

Glutamate opens Na+/K+ channels in cultured astrocytes

Glial cells from different brain regions and species are depolarized by the neurotransmitter glutamate. The depolarization or, if voltage-clamped at the resting membrane potential, the inward current induced by glutamate could be due either to activation of receptor-coupled ion channels or electrogenic uptake of the transmitter. In the present study we applied the patch-clamp technique in the whole-cell recording mode to analyze glutamate-induced currents in cultured astrocytes from rat cerebral hemispheres. At the resting membrane potential, glutamate induced an inward current ranging from 40 to 300 pA. This current decreased in size with depolarization and reversed at about 0 mV. The resulting current-to-voltage curve was linear and depended strongly on the transmembrane Na+ but not on the Ca++ or Cl- gradient. In the presence of glutamate, current noise increased at potentials positive or negative from the reversal potential indicating that ionic channels are activated by glutamate. Both kainate and quisqualate mimicked the effect of glutamate. We conclude that glutamate opens a Na+/K+ channel in cultured astrocytes because of activation of a receptor which shares many properties with the neuronal kainate/quisqualate receptor.     

Resolution of two [(35)S]GTP-gamma-S binding sites and their response to chronic morphine treatment: a binding surface analysis

The mechanisms by which prolonged exposure to morphine leads to tolerance are not fully understood. We investigated the effects of etorphine (ET) on [(35)S]guanosine 5'-(-thio)-triphosphate ([(35)S]GTP-gamma-S) binding in brains of rats made tolerant to morphine via the implantation of morphine (or placebo) pellets. Binding surface analysis was used to characterize the interactions of ET, Gpp(Np)H and GTP-gamma-S with sites labeled by [(35)S]GTP-gamma-S. Data sets were fitted to one- and two-site binding models using the nonlinear least squares curve fitting program MLAB-PC (Civilized Software, Bethesda, MD, USA). Two binding sites were readily resolved. Chronic morphine significantly increased the B(max) and K(d) of the high affinity binding site. ET stimulated [(35)S]GTP-gamma-S binding in placebo membranes via an increase in the B(max) of the high affinity binding site. In contrast, ET stimulated [(35)S]GTP-gamma-S in chronic morphine membranes via a large decrease in the K(d) of the high affinity site. These results suggest that chronic morphine treatment alters the mechanism by which ET stimulates [(35)S]GTP-gamma-S binding to G-proteins. Since proper G-protein/receptor coupling increases [(35)S]GTP-gamma-S binding via an increase in B(max) values, these results suggest that opioid receptors in chronic morphine membranes are not normally coupled to G-proteins. These findings corroborate earlier studies that reported changes in G-protein function in morphine tolerant animals.     

Differential state-dependent modification of inactivation-deficient Nav1.6 sodium channels by the pyrethroid insecticides S-bioallethrin, tefluthrin and deltamethrin

Pyrethroid insecticides disrupt nerve function by modifying the gating kinetics of transitions between the conducting and nonconducting states of voltage-gated sodium channels. Pyrethroids modify rat Na(v)1.6+β1+β2 channels expressed in Xenopus oocytes in both the resting state and in one or more states that require channel activation by repeated depolarization. The state dependence of modification depends on the pyrethroid examined: deltamethrin modification requires repeated channel activation, tefluthrin modification is significantly enhanced by repeated channel activation, and S-bioallethrin modification is unaffected by repeated activation. Use-dependent modification by deltamethrin and tefluthrin implies that these compounds bind preferentially to open channels. We constructed the rat Na(v)1.6Q3 cDNA, which contained the IFM/QQQ mutation in the inactivation gate domain that prevents fast inactivation and results in a persistently open channel. We expressed Na(v)1.6Q3+β1+β2 sodium channels in Xenopus oocytes and assessed the modification of open channels by pyrethroids by determining the effect of depolarizing pulse length on the normalized conductance of the pyrethroid-induced sodium tail current. Deltamethrin caused little modification of Na(v)1.6Q3 following short (10ms) depolarizations, but prolonged depolarizations (up to 150ms) caused a progressive increase in channel modification measured as an increase in the conductance of the pyrethroid-induced sodium tail current. Modification by tefluthrin was clearly detectable following short depolarizations and was increased by long depolarizations. By contrast modification by S-bioallethrin following short depolarizations was not altered by prolonged depolarization. These studies provide direct evidence for the preferential binding of deltamethrin and tefluthrin (but not S-bioallethrin) to Na(v)1.6Q3 channels in the open state and imply that the pyrethroid receptor of resting and open channels occupies different conformations that exhibit distinct structure-activity relationships.     

Protein kinase A reduces voltage-dependent Na+ current in Xenopus oocytes.

The voltage-dependent Na+ channel of the brain is a good substrate for phosphorylation by the cAMP-dependent protein kinase (protein kinase A, or PKA), but the physiological effects of PKA on Na+ channels are poorly documented. We studied modulation by PKA of voltage-dependent Na+ channels expressed in Xenopus oocytes injected with RNA coding for the alpha-subunit of the channel protein (rat brain type IIA and its variant VA200), using the two electrode voltage-clamp technique. Intracellularly injected cAMP or catalytic subunit of PKA, or extracellularly applied forskolin, inhibited the Na+ current by 20-30%. The effect of cAMP was attenuated by prior injection of PKA inhibitors. Injection of small doses of protein phosphatase 2A increased the Na+ current by 10%, whereas larger doses of protein phosphatase 1 and alkaline phosphatase were without effect. The inhibition by PKA showed little voltage dependence, being only slightly stronger at holding potentials at which the availability of the channels was reduced. The voltage dependence of activation and inactivation processes was not altered by cAMP. Similar effects were exerted by forskolin and cAMP on the Na+ channels expressed after the injection of heterologous (total) RNA from rat brain. Thus, PKA modulates the Na+ channel by a mechanism that does not involve major changes in the voltage dependency of the current and is exerted on the channel-forming alpha-subunit.     

Excitation of locus coeruleus neurons by vasoactive intestinal peptide: evidence for a G-protein-mediated inward current

Vasoactive intestinal polypeptide (VIP) caused a reversible increase in the firing rate of locus coeruleus (LC) neurons. Voltage-clamp at −60 mV revealed that VIP induced an inward current, associated with a small increase in conductance. The inward current persisted in the presence of Co²⁺ (to block Ca²⁺ channels) or tetrodotoxin (to block fast voltage-dependent Na⁺ channels). Substitution (80%) of Na⁺ with choline or Tris reduced the VIP-elicited inward current by approximately 75%. Changing external K⁺ concentrations did not alter the effect of VIP. The inward current induced by VIP became irreversible after the intracellular administration of GTPγS, a hydrolysis-resistant analog of GTP which can cause a prolonged activation of G-proteins. The intracellular application of GDPβS, which can interfere with G-protein activation, attenuated the effect of VIP. Pertussis toxin, an inactivator of certain G-proteins, did not block the effect of VIP. We conclude that VIP directly excites LC neurons by inducing a largely Na-dependent inward current. As this effect became irreversible in the presence of intracellular GTPγS, was attenuated by GDPβS, and was not eliminated by pertussis toxin, mediation through a pertussis toxin-insensitive G-protein is suggested.     

Fast Inactivation of Ca<Subscript>V</Subscript>2.2 Channels Is Prevented by the Gβ<Subscript>1</Subscript> Subunit in Rat Sympathetic Neurons

Voltage-dependent regulation of Ca_V2.2 channels by G-proteins is performed by the β (Gβ) subunit. Most studies of regulation by G-proteins have focused on channel activation; however, little is known regarding channel inactivation. This study investigated inactivation of Ca_V2.2 channels in superior cervical ganglion neurons that overexpressed Gβ subunits. Ca_V2.2 currents were recorded by whole-cell patch clamping configuration. We found that the Gβ₁ subunit reduced inactivation, while Gβ₅ subunit did not alter at all inactivation kinetics compared to control recordings. Ca_V2.2 current decay in control neurons consisted of both fast and slow inactivation; however, Gβ₁-overexpressing neurons displayed only the slow inactivation. Fast inactivation was restored by a strong depolarization of Gβ₁-overexpressing neurons, therefore, through a voltage-dependent mechanism. The Gβ₁ subunit shifted the voltage dependence of inactivation to more positive voltages and reduced the fraction of Ca_V2.2 channels resting in the inactivated state. These results support that the Gβ₁ subunit inhibits the fast inactivation of Ca_V2.2 channels in SCG neurons. They explain the long-observed sustained Ca²⁺ current under G-protein modulation.     

A tetrodotoxin- and Mn2(+)-insensitive Na+ current in Duchenne muscular dystrophy

Muscle myotube cultures were obtained from normal and Duchenne muscular dystrophy (DMD) biopsies by using an explant technique. The current-voltage (I/V) curve of the whole sodium (Na+) current (INa) in normal myotubes was similar to that obtained from DMD myotubes. However, the inactivation curve of the whole INa was different in normal myotubes when compared to that obtained from DMD myotubes. Addition of 10(-4) M tetrodotoxin (TTX, a fast INa blocker) decreased the whole INa in both preparations. The inorganic calcium (Ca2+) blocker manganese (Mn2+) completely blocked the remaining TTX-resistant INa of normal myotubes and decreased this current in DMD myotubes leaving behind a TTX- and Mn2(+)-insensitive INa that was insensitive to the Ca2+ blocker desmetoxyverapamil ((-)D888). The slow inward barium current (IBa) of both normal and DMD myotubes was blocked by Mn2+ and (-)D888. However the kinetics of the slow channel in normal myotubes was different from that of DMD myotubes. This study demonstrates the presence of a TTX- and Mn2(+)-insensitive INa in DMD myotubes. This channel may contribute to the increase of intracellular Na+ [( Na]i) in DMD and allow Ca2+ to enter the cells through the Na(+)-Ca2+ exchanger, thus contributing to calcium loading.     

A role for extracellular Na+ in the channel gating of native and recombinant kainate receptors

Ionotropic glutamate receptors of the kainate and AMPA subtypes share a number of structural features, both topographical and in terms of stoichiometry. In addition, AMPA and kainate receptors share similar pharmacological and biophysical properties in that they are activated by common agonists and display rapid activation and desensitization characteristics. However, we show here that in contrast to AMPA receptor-mediated responses (native or recombinant GluR3 receptor), the response of native and recombinant (GluR6) kainate receptors to glutamate was drastically reduced in the absence of extracellular Na+ (i.e., when replaced by Cs+). Removal of Na+ increases the rate of desensitization, indicating that external Na+ modulates channel gating. Whereas the size of the substituting cation is important in mimicking the action of Na+ (Li+>K+>Cs+), modulation was voltage independent. These results indicate the existence of different gating mechanisms for AMPA and kainate receptors. By using chimeric AMPA-kainate receptors derived from GluR3 and GluR6, we have identified a key residue in the S2 segment of GluR6 (M770) that is largely responsible for the sensitivity of the receptor to external Na+. Thus, these results show the existence of a specific kainate receptor gating mechanism that requires external Na+ to be operative.