Fast inactivation of a brain K+ channel composed of Kv1.1 and Kvbeta 1.1 subunits modulated by G protein beta gamma subunits
1999; Springer Nature; Volume: 18; Issue: 5 Linguagem: Inglês
10.1093/emboj/18.5.1245
ISSN1460-2075
Autores Tópico(s)Neuroscience and Neuropharmacology Research
ResumoArticle1 March 1999free access Fast inactivation of a brain K+ channel composed of Kv1.1 and Kvβ1.1 subunits modulated by G protein βγ subunits Jie Jing Jie Jing Department of Physiology and Pharmacology, Sackler School of Medicine, Tel-Aviv University, 69978 Ramat Aviv, Israel Search for more papers by this author Dodo Chikvashvili Dodo Chikvashvili Department of Physiology and Pharmacology, Sackler School of Medicine, Tel-Aviv University, 69978 Ramat Aviv, Israel Search for more papers by this author Dafna Singer-Lahat Dafna Singer-Lahat Department of Physiology and Pharmacology, Sackler School of Medicine, Tel-Aviv University, 69978 Ramat Aviv, Israel Search for more papers by this author William B. Thornhill William B. Thornhill Department of Physiology and Biophysics, Mount Sinai School of Medicine, The Mount Sinai Hospital, New York, NY, 10029-6574 USA Search for more papers by this author Eitan Reuveny Eitan Reuveny Department of Biological Chemistry, Weizmann Institute of Science, Rehovot, Israel Search for more papers by this author Ilana Lotan Corresponding Author Ilana Lotan Department of Physiology and Pharmacology, Sackler School of Medicine, Tel-Aviv University, 69978 Ramat Aviv, Israel Search for more papers by this author Jie Jing Jie Jing Department of Physiology and Pharmacology, Sackler School of Medicine, Tel-Aviv University, 69978 Ramat Aviv, Israel Search for more papers by this author Dodo Chikvashvili Dodo Chikvashvili Department of Physiology and Pharmacology, Sackler School of Medicine, Tel-Aviv University, 69978 Ramat Aviv, Israel Search for more papers by this author Dafna Singer-Lahat Dafna Singer-Lahat Department of Physiology and Pharmacology, Sackler School of Medicine, Tel-Aviv University, 69978 Ramat Aviv, Israel Search for more papers by this author William B. Thornhill William B. Thornhill Department of Physiology and Biophysics, Mount Sinai School of Medicine, The Mount Sinai Hospital, New York, NY, 10029-6574 USA Search for more papers by this author Eitan Reuveny Eitan Reuveny Department of Biological Chemistry, Weizmann Institute of Science, Rehovot, Israel Search for more papers by this author Ilana Lotan Corresponding Author Ilana Lotan Department of Physiology and Pharmacology, Sackler School of Medicine, Tel-Aviv University, 69978 Ramat Aviv, Israel Search for more papers by this author Author Information Jie Jing1,‡, Dodo Chikvashvili1,‡, Dafna Singer-Lahat1, William B. Thornhill2, Eitan Reuveny3 and Ilana Lotan 1 1Department of Physiology and Pharmacology, Sackler School of Medicine, Tel-Aviv University, 69978 Ramat Aviv, Israel 2Department of Physiology and Biophysics, Mount Sinai School of Medicine, The Mount Sinai Hospital, New York, NY, 10029-6574 USA 3Department of Biological Chemistry, Weizmann Institute of Science, Rehovot, Israel ‡J.Jing and D.Chikvashvili contributed equally to this work *Corresponding author. E-mail: [email protected] The EMBO Journal (1999)18:1245-1256https://doi.org/10.1093/emboj/18.5.1245 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Modulation of A-type voltage-gated K+ channels can produce plastic changes in neuronal signaling. It was shown that the delayed-rectifier Kv1.1 channel can be converted to A-type upon association with Kvβ1.1 subunits; the conversion is only partial and is modulated by phosphorylation and microfilaments. Here we show that, in Xenopus oocytes, expression of Gβ1γ2 subunits concomitantly with the channel (composed of Kv1.1 and Kvβ1.1 subunits), but not after the channel's expression in the plasma membrane, increases the extent of conversion to A-type. Conversely, scavenging endogenous Gβγ by co-expression of the C-terminal fragment of the β-adrenergic receptor kinase reduces the extent of conversion to A-type. The effect of Gβγ co-expression is occluded by treatment with dihydrocytochalasin B, a microfilament-disrupting agent shown previously by us to enhance the extent of conversion to A-type, and by overexpression of Kvβ1.1. Gβ1γ2 subunits interact directly with GST fusion fragments of Kv1.1 and Kvβ1.1. Co-expression of Gβ1γ2 causes co-immunoprecipitation with Kv1.1 of more Kvβ1.1 subunits. Thus, we suggest that Gβ1γ2 directly affects the interaction between Kv1.1 and Kvβ1.1 during channel assembly which, in turn, disrupts the ability of the channel to interact with microfilaments, resulting in an increased extent of A-type conversion. Introduction Voltage-gated K+ (Kv) channels are involved in a host of cellular processes from setting the resting membrane potential and shaping action potential wave-form and frequency, to controlling synaptic strength (Rudy et al., 1988). Two major types of Kv channels have been described. The delayed-rectifier type inactivate slowly with a time course of up to several seconds (C-type inactivation) and are involved in repolarization of action potential and attenuation of cell excitability. The A-type channels inactivate rapidly with a time course of less than ∼100 ms (N-type inactivation) and are involved in regulation of firing patterns and the threshold for firing, and thus play a key role in encoding pre- and postsynaptic nervous signals (Connor and Stevens, 1970). It has been shown that delayed-rectifier channels can be converted to A-type channels upon association of pore-forming Shaker-related (Kv1; Stuhmer et al., 1989) α subunits with peripheral auxiliary β subunits (Kvβ1; Rettig et al., 1994). The rapid inactivation results from occlusion of the inner mouth of the channel pore with the N-terminal part of the Kvβ1 subunit (Rettig et al., 1994), in a mechanism termed 'ball-and-chain' inactivation (Armstrong and Bezanilla, 1977). However, we showed, by studying the heteromultimeric channels composed of Kv1.1 and Kvβ1.1 subunits, that conversion of the delayed-rectifier to A-type is not complete and is subject to modulations by cellular processes. The extent of conversion to A-type is dependent on interactions of the channel with microfilaments and on cellular mechanisms leading to phosphorylation (Levin et al., 1996) and dephosphorylation (Levy et al., 1998) of Ser446 at the C-terminus of Kv1.1. Part of the effect of phosphorylation is conveyed via regulation of the interaction of the Kv1.1's C-terminus with a postsynaptic density-95 (PSD-95)-like protein (Jing et al., 1997). In this study, we identify G protein βγ subunits as a cellular regulator of the channel's interaction with microfilaments and consequently as a regulator of A-type conversion. Such modulations of A-type activity that modify K+ channel gating properties can produce plastic changes in neuronal signaling and may be involved in neural mechanisms related to behavior and learning (Crow, 1988). The large GTP-binding proteins (G proteins), made up of Gαβγ heterotrimers, dissociate into Gα and Gβγ subunits, which interact separately with effector molecules. While Gα has been recognized for many years now as a signal-transducing molecule, Gβγ has been recognized only recently as a direct regulator of many target proteins (Clapham and Neer, 1997). Many ion channels are modulated by G proteins, most by activating intracellular cascades and some via direct interaction with the channel proteins (Wickman and Clapham, 1995). Gβγ has been demonstrated to inhibit some voltage-dependent Ca2+ channels via direct interactions (for a review see Dolphin, 1997) and to activate the G protein-activated K+ channels (GIRKs) by direct binding (for a review see Dascal, 1997). Very recently, it has been demonstrated that co-expression of Gβγ with rat brain type IIA voltage-gated Na+ channels induces a persistent current component with shifted voltage dependence of inactivation. Direct Gβγ interaction with the channel has been inferred in this case from results showing that peptides containing Gβγ-binding motifs from adenylyl cyclase 2 and the Na+ channel inhibited the Gβγ effect. Our present report is the first demonstration of a modulation of a voltage-gated K+ channel by direct interaction with Gβγ. Also, our results point to a novel mechanism of action by Gβγ. Immunocytochemical studies have shown that Kv1.1 and Kvβ1.1 proteins are widely distributed in adult rodent brain and occur in specific neuronal locations, in particular in juxtaparanodal regions of myelinated axons and in terminal fields (Sheng et al., 1993; Wang et al., 1993, 1994; Rhodes et al., 1995, 1996). This may indicate a role in axonal action potential propagation and repolarization of the membrane at the synaptic terminal at specific brain regions, e.g. cerebellar Purkinje cells where high immunoreactivity of both proteins have been detected (Veh et al., 1995; Rhodes et al., 1996). Modulation of A-type activity by mechanisms described in our previous publications and in this report may play a significant role in the regulation of neuronal excitability in these regions. Results Level of expression of G protein βγ subunits (Gβγ) affects the extent of inactivation of the Kv1.1–Kvβ1.1 current When Kv1.1 (α) is co-expressed with Kvβ1.1 (β) in oocytes injected with the corresponding mRNAs, a heteromultimeric Kv1.1–Kvβ1.1 channel is formed that conveys a rapidly inactivating current with a fast inactivating component (Ii) and a non-inactivating sustained component (Is). Is is defined as the current remaining at the end of a 120 ms depolarizing pulse to +50 mV (Levin et al., 1996; see graphs in Figure 1A). The extent of inactivation (the inactivating fraction) is defined as Ii/Ip (Ip = peak current) (Levin et al., 1996). Previously (Levin et al., 1996), we showed that the extent of inactivation of the Kv1.1–Kvβ1.1 current, and not the inactivation rate constant, increases up to saturation at Ii/Ip = 0.5–0.8 (depending on the batch of oocytes) as the injected mRNA ratio of Kvβ1.1 to Kv1.1 (β:α) is increased up to a ratio of ∼50:1 (depending on the batches of mRNAs and oocytes). In parallel, the amount of Kvβ1.1 protein co-precipitated with Kv1.1 protein, i.e. the number of Kv1.1 subunits that associate with Kvβ1.1 subunits, increases until saturation is reached at a β:α mRNA ratio of ∼50:1. Figure 1.Co-expression of Gβγ or C-βARK with Kv1.1 (α) and Kvβ1.1 (β) affects the extent of inactivation of the K+ currents in Xenopus oocytes. (A) Normalized current traces evoked by a depolarization to +50 mV recorded by the two-electrode voltage clamp technique from single oocytes injected with either Kvβ1.1 and Kv1.1 mRNAs alone at a Kvβ1.1:Kv1.1 (β:α) ratio of 30:1 or co-injected with either Gβ1γ2 or C-βARK mRNAs; τinacts were 4.98, 4.61 and 5.30, respectively. Ii, Is and Ip illustrate definitions in the text for the inactivating, non-inactivating and total current components of Kv1.1–Kv1.1 (αβ), respectively. The inset shows activation curves of currents elicited in oocytes injected with Kv1.1 and Kvβ1.1 alone or co-injected with Gβ1γ2 (four oocytes each) by 250 ms depolarizations to the indicated voltages, at 20 mV increments, every 20 s (to allow recovery from inactivation); the elicited currents were normalized to maximal current at +80 mV. (B) Normalized and averaged effects of Gβ1γ2 mRNA co-injection with saturating β:α mRNA ratios (50:1 to 200:1; left panel) and with non-saturating ratios (3:1 to 30:1; middle panel), and effect of C-βARK mRNA co-injection (right panel). *p <0.004; **p <0.0000001. Numbers above the bars indicate the number of batches of oocytes; numbers in parentheses indicate the number of oocytes. (C) Effect of Gβ1γ2 (left panel) or C-βARK (right panel) mRNAs co-injection with different β:α mRNA ratios, each assayed in a single batch of oocytes; shown are absolute values of inactivations. *p <0.02; NS, not significant. Download figure Download PowerPoint To examine the effect of G protein βγ subunits on the inactivation of the Kv1.1–Kvβ1.1 current, we co-expressed Gβ1γ2 subunits with the channel subunits and recorded currents in a two-electrode voltage clamp configuration (Figure 1; normalized values are used when average values are derived from several batches of oocytes because of variability among batches). In oocytes expressing the channel subunits at saturating ratios, Gβγ co-expression occasionally increased the extent of inactivation of the current that was assayed 2 days after mRNA injection. However, when averaged over many experiments, the effect was very small (Figure 1B). Reducing the β:α ratio below saturation (between 30:1 and 3:1) resulted in a marked and reproducible increase in the extent of inactivation in oocytes expressing Gβγ subunits (Figure 1A and B). A test performed in a single batch of oocytes showed that the lower the β:α ratio the larger the degree of increase in extent of inactivation (Figure 1C, left panel). Actually, at the saturating β:α ratio (50:1), the extent of inactivation of the channel was so high that Gβγ could not increase it further. The inactivation time constant (τinact) for channels in the presence and in the absence of Gβγ did not differ statistically and were determined in two batches of oocytes to be 4.96 ± 0.25 ms (n = 22) and 5.04 ± 0.09 ms (n = 25), respectively. Also, Gβγ expression did not affect the voltage dependence of activation of the channel (Figure 1A, inset). Gβγ co-expression with Kv1.1 subunits alone had no effect on the kinetics of the slowly inactivating (C-type inactivation) Kv1.1 current (not shown). Conversely, in oocytes co-expressing the C-terminal region of β-adrenergic receptor kinase (C-βARK1; in-frame with the first 15 N-terminal amino acids of src60), the extent of inactivation was reduced in 12 out of 18 batches of oocytes tested (Figure 1A and B). This fragment of C-βARK was shown previously to bind Gβγ subunits effectively (Koch et al., 1993) and to inhibit the activation of the G protein-coupled potassium channel (GIRK) in Xenopus oocytes (E.Reuveny, unpublished data). The presence of the N-terminal src60 amino acids was used to target the C-βARK1 fragment to the membrane (Dascal et al., 1995). The C-βARK effect was about the same for saturating and non-saturating β:α ratios (Figure 1C, right panel), suggesting a constitutive role for Gβγ in the inactivation and the existence of sufficient free endogenous Gβγ to sustain maximal inactivation attained at the saturating β:α ratio. In oocytes expressing C-βARK, the current amplitudes were ∼50% of those in oocytes not expressing C-βARK; the reason for this is not known. Interrelationships between the Gβγ effect, channel phosphorylation and microfilaments Previously we showed that the extent of inactivation of the Kv1.1–Kvβ1.1 channel can be increased by treatment of oocytes with dihydrocytochalasin B (DHCB), leading to microfilament depolymerization (Levin et al., 1996), presumably due to disruption of points of interaction between the channel and microfilaments. One point of interaction was found to be the very end of the C-terminus of the Kv1.1 subunit via a PSD-95-like protein (Jing et al., 1997). We also showed that phosphorylation of Kv1.1 at Ser446 is another factor that affects the extent of inactivation. The extent of inactivation of the wild-type channel which is substantially phosphorylated is larger than that of the unphosphorylated Kv1.1S446A–Kvβ1.1 channel (Levin et al., 1996); part of the effect of phosphorylation is due to its modulation of the C-terminal interaction of Kv1.1 with microfilaments (Jing et al., 1997). In this study, we set out to test the relationship of these inactivation-affecting factors to the Gβγ effect. To examine the relationship of the Gβγ effect to phosphorylation, we examined oocytes co-expressing Gβγ with the Kv1.1S446A–Kvβ1.1 channel and found that the increase in inactivation (Figure 2A, left panel) was similar to that of the wild-type channel (compare with Figure 1B). To test the relationship with the C-terminal-mediated interaction with microfilaments, Gβγ was co-expressed with channels of which the last amino acids of Kv1.1 that interact with the PSD-95-like proteins (Kim et al., 1995) were deleted (K490s.c; Jing et al., 1997). An experiment performed on a single batch of oocytes (Figure 2B) showed that this truncation increased the extent of inactivation, as shown before (Jing et al., 1997), and Gβγ co-expression further increased it but only up to about the same level as in the untruncated channel co-expressed with Gβγ. Figure 2A (right panel) shows normalized and averaged results of several such experiments. Thus, the Gβγ effect occluded the effect of disruption of the C-terminal interaction with microfilaments. To test the relationship between the Gβγ effect and the overall interaction of the channel with microfilaments, the effect of a 4 h treatment of oocytes with DHCB on channels co-expressed with Gβγ was examined and compared with that in the absence of Gβγ in two batches of oocytes (Figure 2C). DHCB treatment increased the inactivation up to about the same level as Gβγ expression. In oocytes expressing Gβγ, the DHCB treatment had a small effect, if any. Figure 2.Relationships of the Gβγ effect to the channel's phosphorylation and interactions with microfilaments. (A and B) Effects of Gβ1γ2 mRNA co-injection with Kvβ1.1 (β) and either Kv1.1S446A [α (S446A)] or Kv1.1S446A/K490s.c [α (S446A/K490s.c.)] mRNAs in non-saturating β:α mRNA ratios. The Gβγ effect was normalized and averaged over several batches of oocytes (A) or shown in a single batch injected with a β:α mRNA ratio of 7:1 (B). Numbers above the bars indicate the number of batches of oocytes; numbers in parentheses indicate the number of oocytes. (C) Effect of Gβ1γ2 mRNA co-injection with Kv1.1 (α) and Kvβ1.1 (β) mRNAs in non-saturating ratios in oocytes of a single batch that were not treated or treated with DHCB. **p <0.0004; *p <0.04; NS, not significant. Download figure Download PowerPoint Gβ interacts physically with the channel To probe for possible physical interaction between G protein βγ subunits and the channel Kv1.1 and Kvβ1.1 subunits, we have measured the in vitro binding of 35S-labeled Gβ1γ2 synthesized in reticulocyte lysate to GST fusion proteins corresponding to the major intracellular parts of Kv1.1 and a part of Kvβ1.1 subunits. The following GST fusion proteins were used: GSTαC, corresponding to amino acids 412–495, i.e. the whole C-terminus of Kv1.1; GSTαN1 and GSTαN2, corresponding to the whole of (amino acids 1–167) and most of the N-terminus (amino acids 14–162) of α, respectively; GSTαT1'B' corresponding to amino acids 72–143 of the N-terminus region that is involved in tetramerization of Kvα subunits (Li et al., 1992; Shen et al., 1993; Shen and Pfaffinger, 1995) and in Kvβ subunit binding (includes some residues essential for Kvβ binding; Sewing et al., 1996; Yu et al., 1996); and GSTβV, corresponding to amino acids 1–72 of Kvβ1.1, i.e. a domain that is variable among the different Kvβ proteins and is not involved in binding to Kvα. The scheme shown in Figure 3A illustrates the position of the different parts of Kv1.1. Figure 3B shows that Gβ bound to αN1, αN2 and βV but not to αC. Since 35S-labeled Gβ1, and not Gγ2, each synthesized alone in reticulocyte lysate, bound to the above proteins (not shown), we concluded that the Gβ1 interacts physically with the channel. Next, we tested the relationship between the binding of Gβ and Kvβ1.1 (35S-labeled and synthesized in reticulocyte lysate). Figure 3C shows that Kvβ1.1 bound to αN1 and αN2, as expected; its binding was more pronounced than that of Gβ. However, while Gβ bound to αNT1'B', Kvβ1.1 did not. In the presence of Gβγ, Kvβ1.1 binding to αN1 was reduced by ∼30% (not shown); this was probably not due to specific competitive binding of the two proteins to the N-terminus of Kv1.1, but rather to non-specific limitations of the experimental method, as Kvβ1.1 binding to αN1 was reduced to the same extent in the presence of PSD-95 protein synthesized in the lysate that binds to αC but not to αN1 (not shown). Figure 3.Interaction of in vitro synthesized 35S-labeled Gβ1γ2 and Kvβ1.1 with GST fusion proteins of Kv1.1 and Kvβ1.1 fragments. (A) Schematic presentation of the Kv1.1 subunit. (B) Left panel: interaction of Gβγ with the GST fusion proteins corresponding to parts of Kv1.1 and Kvβ1.1. Right panel: Coomassie Blue staining of the same gel. (C) Left and middle panels: comparison of the interaction of Kvβ1.1 and Gβ1γ2 with GST fusion proteins of Kv1.1 in a single experiment. Right panel: Coomassie Blue staining of one of the two gels that were run in parallel. The proteins were separated by SDS–12% PAGE and monitored using a PhosphorImager. αN1, αN2 and αT1'B' correspond to whole (amino acids 1–167), most of the length (amino acids 14–162) and tetramerization domain 'B' (amino acids 72–143) of the Kv1.1 N-terminus, respectively; αC corresponds to the whole C-terminus of Kv1.1 (amino acids 412–495); βV corresponds to the variable region of Kvβ1.1 (amino acids 1–72). Similar results to those shown in (B) and in (C) were obtained in four and two additional experiments, respectively. * Indicates the predicted position of the fusion protein. Numbers on the right refer to the mobility of pre-stained molecular weight standards. Download figure Download PowerPoint Co-expression of Gβγ increases the capacity of Kv1.1 to bind Kvβ1.1 subunits Next, we addressed the question of whether Gβγ co-expression affects the interaction between the channel Kv1.1 and Kvβ1.1 subunits. We performed two kinds of biochemical analyses of the channel proteins. In the first kind of experiment, Western blot analysis of Kvβ1.1 protein was done in oocytes tested electrophysiologically on the same day and displayed a pronounced Gβγ effect. For this purpose, we co-immunoprecipitated the channel subunits by antibody directed against Kv1.1, and they were subjected to SDS–PAGE, blotted onto membranes and probed with antibody directed against Kvβ1.1 subunit for co-precipitated Kvβ1.1 protein. In parallel, proteins of oocytes from the same batch were subjected directly to SDS–PAGE analysis without immunopurification, blotted and probed for total-Kvβ1.1 protein. The amount of Kvβ1.1 protein co-precipitated with Kv1.1 was quantified and normalized for total Kvβ1.1 protein. In four such experiments (see a representative experiment in Figure 4A), the normalized co-precipitated Kvβ1.1 protein was 2.0 ± 0.15 (p <0.001) fold larger in oocytes expressing Gβγ than in oocytes that did not express Gβγ, suggesting that co-expression of Gβγ increases the capacity of Kv1.1 subunits to bind Kvβ1.1 subunits in correlation with the increase in inactivation. Figure 4.Digitized PhosphorImager scans of SDS–PAGE analyses of the effect of Gβγ on the association of Kvβ1.1 (β) with Kv1.1 (α) proteins. (A) Immunoblot analysis of Kvβ1.1 in oocytes previously tested electrophysiologically. Proteins were precipitated from homogenates of oocytes either uninjected (c; lanes 1 and 4), injected with Kv1.1 and Kvβ1.1 mRNAs alone (lanes 2 and 5) or co-injected with Gβ1γ2 (lanes 3 and 6), with either an antibody against Kv1.1 followed by protein A–Sepharose (left panel) or with ethanol (right panel), electrophoresed on an 8% gel, transferred to PVDF transfer membranes and the resultant immunoblots probed with anti-Kvβ1.1 antibody to monitor the amounts of co-immunoprecipitated (C.I.P.; left panel) or total (right panel) Kvβ1.1. Signals were visualized using ECL. (B) Analysis of [35S]Met/Cys-labeled Kvβ1.1 and Kv1.1 proteins co-immunoprecipitated by either Kv1.1 (left panel) or Kvβ1.1 antibodies (right panel) from homogenates of oocytes injected with the same combinations as in (A) but with 10-fold larger Kv1.1 and Kvβ1.1 mRNA concentrations. The arrow indicating Gβ points to Gβ that was pulled down by protein A–Sepharose (see text). Numbers on the right refer to the mobility of pre-stained molecular weight standards. Download figure Download PowerPoint A second type of biochemical analysis was done in an effort to resolve by SDS–PAGE, in addition to Kvβ1.1, also the Kv1.1 protein. Both proteins were expressed from mRNA concentrations that were ∼10-fold larger than those in the electrophysiological experiments. Kv1.1 and Kvβ1.1 proteins were co-immunoprecipitated with either anti-Kv1.1 or anti-Kvβ1.1 antibodies from oocytes metabolically labeled with [35S]methionine/cysteine and were subjected to SDS–PAGE. The amount of Kvβ1.1 that was co-precipitated with Kv1.1 by the Kv1.1 antibody was quantified and normalized to the amount of precipitated Kv1.1 protein. In three such experiments (see a representative experiment in Figure 4B, left panel), the normalized co-precipitated Kvβ1.1 was 2.2 ± 0.09 (p <0.025) fold larger in oocytes expressing Gβγ than in oocytes that did not express Gβγ. The co-immunoprecipitation with anti- Kvβ1.1 antibody (Figure 4B, right panel) demonstrated that not all of Kvβ1.1 was bound to Kv1.1, as more Kvβ1.1 was precipitated with the Kvβ1.1 antibody than co-precipitated with the Kv1.1 antibody. Taken together, the two biochemical types of analyses point to the probability that Gβγ expression increases the capacity of Kv1.1 to bind Kvβ1.1. In calculating the amount of the Kvβ1.1 protein, we referred only to the upper band of ∼45 kDa, which corresponds to the predicted molecular weight; the nature of the ∼35 kDa band is unknown to us (Levin et al., 1996). The bands corresponding to Kv1.1 protein (∼57 kDa) in Figure 4B are not very prominent because the mRNA concentrations were minimal in order to be as close as possible to the physiological concentrations and at the same time to allow for reasonable resolution of the protein. Interestingly, when Kv1.1 and Kvβ1.1 concentrations were increased up to 30-fold of those in an electrophysiological experiment, Gβγ expression had little or no effect on the Kvβ1.1-binding capacity of Kv1.1, suggesting that in the presence of a large density of channel subunits, Gβγ loses its effect. The apparent Gβ co-precipitation with both antibodies (Figure 4B, lanes 3 and 6) is most probably an experimental artifact arising from the fact that this protein non-specifically binds to protein A–Sepharose (not shown). The above biochemical analyses were done with whole oocyte homogenates and thus correspond mainly to channel proteins located in the cytoplasm and internal organelles, including the Golgi apparatus and endoplasmic reticulum (ER). The plasma membrane content of the channel proteins is <1% of total oocyte content (Levin et al., 1995). Therefore, it seems that Gβγ increased the Kvβ1.1-binding capacity of Kv1.1 before reaching the plasma membrane. It has been shown in transfected mammalian cells that assembly of Shaker channel subunits takes place early in biosynthesis before reaching the plasma membrane (Shi et al., 1996; Nagaya and Papazian, 1997); this may be the case in oocytes. Consequently, one would predict that if there is a causative relationship between the effect of Gβγ on subunit assembly and its effect on inactivation of channels that are in the plasma membrane, the latter effect will be less prominent if Gβγ is not expressed together with the channel subunits, but later on, when much of the assembly of channels has already occurred. To address this prediction, we compared the effects of 2 days expression of Gβγ which was either co-injected with the channel subunits, as above, or injected 2 days following the injection of the channel subunits. Clearly, in two such experiments, the late expression of Gβγ did not result in increased inactivation; rather, the extent of inactivation decreased by a small but statistically significant extent (Figure 5A). At this stage, we do not understand the decrease in inactivation. To substantiate further the importance of the time at which the interaction between Gβ1γ2 and the channel occurs, we injected Gβ1γ2 1 day prior to the injection of the channel subunits, in order to allow already synthesized Gβ1γ2 to be present from the very beginning of channel synthesis and assembly. As expected, in four experiments, the early expression of Gβ1γ2 had a larger effect on the extent of inactivation than that of co-expressed Gβ1γ2 (Figure 5B). Figure 5.The effects of late and early expression of Gβγ. (A) Comparison between the effects of late expression and co-expression of Gβ1γ2 on the extent of inactivation of the K+ currents. Oocytes were injected on day 1 with either Kvβ1.1 (β) and Kv1.1 (α) alone or co-injected with Gβ1γ2 and assayed electrophysiologically on day 3 and later on day 5; on both days of assay, the co-expressed Gβ1γ2 increased the extent of inactivation. On day 3, some of the oocytes that were injected on day 1 with Kvβ1.1 and Kv1.1 alone were injected with Gβ1γ2 and assayed electrophysiologically on day 5; the late expression of Gβ1γ2 decreased the extent of inactivation (compared with oocytes injected on day 1 with Kvβ1.1 and Kv1.1 alone). (B) Comparison between the effects of early expression and co-expression of Gβ1γ2. Oocytes that were either not injected or injected with Gβ1γ2 on day 0 were injected on day 1 with Kvβ1.1 and Kv1.1 and assayed electrophysiologically on day 3. The early expression of Gβ1γ2 had a larger effect on the extent of inactivation. Numbers above the bars indicate number the of batches of oocytes; numbers in parentheses indicate the number of oocytes. *p <0.02. Download figure Download PowerPoint Discussion Gβγ levels regulate the extent of inactivation of the Kv1.1–Kvβ1.1 current Our results show that co-expression of Gβ1γ2 subunits with a rat brain voltage-gated K+ channel composed of Kv1.1 (α) and Kvβ1.1 (β) subunits r
Referência(s)