Interaction of Gq and Kir3, G Protein-Coupled Inwardly Rectifying Potassium Channels

【关键词】  Interaction

    Activation of substance P receptors, which are coupled to Gq, inhibits the Kir3.1/3.2 channels, resulting in neuronal excitation. We have shown previously that this channel inactivation is not caused by reduction of the phosphatidylinositol 4,5-bisphosphate level in membrane. Moreover, Gq immunoprecipitates with Kir3.2 (J Physiol 564:489-500, 2005), suggesting that Gq interacts with Kir3.2. Positive immunoprecipitation, however, does not necessarily indicate direct interaction between the two proteins. Here, the glutathione transferase pull-down assay was used to investigate interaction between Gq and the K+ channels. We found that Gq interacted with N termini of Kir3.1, Kir3.2, and Kir3.4. However, Gq did not interact with the C terminus of any Kir3 or with the C or N terminus of Kir2.1. TRPC6 is regulated by the signal initiated by Gq. Immunoprecipitation, however, showed that Gq did not interact with TRPC6. Thus, the interaction between Gq and the Kir3 N terminus is quite specific. This interaction occurred in the presence of GDP or GDP-AlF-4. The Gq binding could take place somewhere between residues 51 to 90 of Kir3.2; perhaps the segment between 81 to 90 residues is crucial. G, which is known to bind to N terminus of Kir3, did not compete with Gq for the binding, suggesting that these two binding regions are different. These findings agree with the hypothesis (J Physiol 564:489-500, 2005) that the signal to inactivate the Kir3 channel could be mainly transmitted directly from Gq to Kir3.

    Substance P (SP), an undecapeptide of the tachykinin family discovered by Chang and Leeman (1970), has been shown to excite various neurons. We have been investigating SP as a model excitatory transmitter. One of the mechanisms for neuronal excitation is inactivation of inward rectifier K+ channels (Stanfield et al., 1985; Velimirovic et al., 1995), including G protein-coupled inwardly rectifying K+ channels (GIRK, Kir3) (Dascal et al., 1993; Kubo et al., 1993).

    The mechanism of the Kir3 activation, which leads to neuronal inhibition, is well clarified. Logothetis et al. (1987) discovered that the Kir3 channel is activated by G subunits. Later, several groups of investigators (Huang et al., 1995; Inanobe et al., 1995; Krapivinsky et al., 1995b; Kunkel and Peralta, 1995) found that G binds directly to Kir3. Furthermore, Slesinger et al. (1995) determined that the activation signal is transmitted through this direct interaction of G and Kir3.

    Contrary to the situation of Kir3 activation, the signal transduction mechanism of Kir3 inactivation is poorly understood. It has been shown that Gq is involved in the signaling of Kir3 inactivation (Leaney et al., 2001; Lei et al., 2001; Koike-Tani et al., 2005). Beyond this, not much is clarified. A prevailing theory attributes the Kir3 inactivation to lowering of the PIP2 level in the membrane (Kobrinsky et al., 2000; Cho et al., 2001; Lei et al., 2001) or to the PKC-induced phosphorylation (Sharon et al., 1997; Mao et al., 2004). However, in a previous article (Koike-Tani et al., 2005), we presented evidence that the SP-induced inactivation of Kir3 channels cannot be explained by decline of the PIP2 level. Increasing the Kir3 affinity to PIP2 by mutation did not change the SP-induced Kir3 inactivation. Moreover, unlike the results using Xenopus laevis oocytes (Sharon et al., 1997; Mao et al., 2004), the role of PKC in the Kir3 inactivation is not considerable in mammalian heterologous systems (Lei et al., 2001; Koike-Tani et al., 2005).

    One possible mechanism of signal transduction from Gq to the Kir3 inactivation is the direct interaction between Gq and Kir3. Previously, Koike-Tani et al. (2005) have shown that Gq coprecipitates with Kir3.2, but not with Kir2.1 or Kir2.2. However, coimmunopreciptation cannot determine whether Gq and Kir3 interact directly without intermediate elements, because the lysates include various cell components. In the present experiments, we used glutathione transferase (GST) pull-down assays to observe direct interaction between Gq and Kir3 and to localize the sites of interaction.

    Culture and Transfection of HEK293A Cells. Human embryonic kidney 293A (HEK293A) cells were purchased from Qbiogene (Irvine, CA). They were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum (Invitrogen, Carlsbad, CA), penicillin (100 U/ml), and streptomycin (100 µg/ml) at 37°C with 10% CO2. Transfection was conducted with Effectene Transfection Reagent (QIAGEN, Valencia, CA). Cells on a 6-cm dish were transfected with the following cDNAs: 0.15 µg of Kir3.1, 0.15 µg of Kir3.2 (or Kir3.4), 0.2 µg of SP receptor, 0.3 µg of G1, 0.3 µg of G2 and 0.1 µg of green fluorescent protein (pEGFP-N1; Clontech, Mountain View, CA). Kir3.1, Kir3.2, and Kir3.4 cDNAs were from the rat, and G1 and G2 cDNAs were of bovine origin. SP receptor cDNA (Takeda et al., 1991) was human. The total amount of cDNAs was kept at 1.6 µg by adjusting the quantity of the empty vector. All cDNAs used were subcloned into pCMV5 (Andersson et al., 1989). One day after the transfection, cultures were replated on 3.5-cm dishes. Each of the dishes has a center well (1.2 cm in diameter) coated with rat-tail collagen (Roche Molecular Biochemicals, Indianapolis, IN). Electrophysiological experiments were performed 2 days after the re-plating.

    Electrophysiology. The whole-cell version of patch clamp was used. The external solution contained 141 mM NaCl, 10 mM KCl, 2.4 mM CaCl2, 1.3 mM MgCl2, 11 mM D-glucose, 5 mM HEPES-NaOH, and 0.0005 mM tetrodotoxin, pH 7.4. Patch pipette solution contained 141 mM potassium d-gluconate, 10 mM NaCl, 5 mM HEPES-KOH, 0.5 mM EGTA-KOH, 0.1 mM CaCl2, 4 mM MgCl2, 3 mM Na2ATP, and 0.2 mM GTP, pH 7.2. SP (0.5 µM; Peptide International, Louisville, KY) was applied through a sewer pipe system (ALA Scientific Instruments, Westbury, NY). Holding potential was -79 mV. Experiments were performed at room temperature. Statistical values are expressed as mean ± S.E.M.

    Protein Purification. To express GST-fusion proteins of the N and C termini of Kirs, each of the Kir cDNAs was constructed in pGEX-2T vector (GE Healthcare, Chalfont St. Giles, Buckinghamshire, UK). The Kir cDNAs used were from the rat, and GRK2 cDNA was of bovine origin. Escherichia coli strain JM109 or BL21 was transformed by pGEX-2T vector harboring cDNA interested. To enhance the solubility of some GST-fusion proteins, BL21(DE3) strain was cotransformed by pT-Trx (Yasukawa et al., 1995). Production of protein was induced by 0.1 to 0.3 mM isopropyl -D-thiogalactoside at A600 of 0.6, and then the bacteria were cultured for another 45 min to 16 h at 26-30°C. The bacterial pellet was suspended in 1/20 volume of lysis buffer with the following composition: 50 mM Tris-HCl pH 7.4, 1 mM EDTA, 20 mM 2-melcaptoethanol, 0.2 mg/ml lysozyme, and proteinase inhibitors (16 µg/ml phenylmethylsulfonyl fluoride, 16 µg/ml N-p-Tosyl-L-phenylalanine chloromethyl ketone, 16 µg/ml N-tosyl-L-lysine chloromethyl ketone, 3.2 µg/ml leupeptin, and 3.2 µg/ml lima bean trypsin inhibitor). The pellet was disrupted by sonication. After centrifugation at 100,000g for 30 min after incubation for 20 min at 4°C, the supernatant was loaded onto a glutathione Sepharose 4B (GE Healthcare) column. The column was washed with wash buffer N (50 mM Tris-HCl pH 7.4, 1 mM EDTA, 20 mM 2-melcaptoethanol, and 400 mM NaCl) and then with wash buffer G (50 mM Tris-HCl, pH 7.4, 1 mM EDTA, 20 mM 2-mercaptoethanol, and 10% glycerol). GST-fusion protein was eluted with elution buffer (50 mM Tris-HCl, pH 8.0, 1 mM EDTA, 10 mM 2-mercaptoethanol, 140 mM NaCl, 10% glycerol, and 5-100 mM glutathione). Peak fractions were exchanged with stock buffer (50 mM HEPES-NaOH, pH 7.5, 1 mM EDTA, 1 mM dithiothreitol, 150 mM NaCl, and 10% glycerol). The recombinant proteins of Gq, G1, and hexahistidine-tagged G2 expressed by the Sf9-bacurovirus system were purified as described before (Kozasa, 2004).

    GST Pull-Down Assays. To perform the pull-down assay with G12, 1 µg of GST-fusion protein (80-200 nM) and 40 nM G were incubated with glutathione Sepharose 4B at 4°C for 1 h in 200 µl of binding buffer A (PBS containing 2 mM EDTA, 5 mM 2-mercaptoethanol, 0.02% C12E10). After incubation, the resin was washed three times with 1 ml of binding buffer A, boiled with sample buffer, and then subjected to Western blotting. For the assay with Gq, 0.5 µg of GST-fusion proteins were incubated with 50 nM Gq in binding buffer B (PBS containing 10 mM MgCl2, 5 mM 2-mercaptoethanol, 0.02% C12E10, and 10 µM GDP with or without AlF-4 [30 µM AlCl3 and 5 mM NaF]). In competitive pull-down assays, binding buffer B with 1 mM MgCl2 was used.

    Precipitated proteins were boiled in a sample buffer (50 mM Tris-HCl at pH 6.8, 1% SDS, 2% 2-mercaptoethanol, 0.008% bromphenol blue, and 8% glycerol) and subjected to Western blotting using antibodies described below. The antibodies bound to proteins of interest on nitrocellulose membranes were visualized by the enhanced chemiluminescence detection system (Pierce, Rockford, IL) using anti-mouse or anti-rabbit Ig antibodies conjugated with horseradish peroxidase as secondary antibodies (GE Healthcare). Anti-G (T-20) and anti Gq/11 (C-19) antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA).

    Immunoprecipitation. HEK293A cells were transfected with plasmid cDNAs of Gq and TRPC6A or Myc-Kir3.2 by Trans IT-LT1 (Invitrogen). The amount of cDNA used for transfection was adjusted by adding pCMV5 to 20 µg total per 10-cm dish of the cells. After 48 h of the transfection, the cells were harvested from the dish and lysed in 550 µl of lysis buffer on ice for 20 min. The lysis buffer consisted of 20 mM HEPES-NaOH at pH 7.5, 150 mM NaCl, 5 mM MgCl2, 1 mM EDTA, 1 mM EGTA, 25 mM -glycerophosphate, 1 mM Na3VO4, 2 µg/ml aprotinin, 10 µg/ml leupeptin, 10 µg/ml pepstatin A, 0.5% Triton-X100, 10% glycerol, and 10 µM GDP either with or without AlF-4 (30 µM AlCl3 and 5 mM NaF). The samples were centrifuged at 15,000g for 20 min at 4°C. The supernatants were subjected to immunoprecipitation. The cell lysates (150 µl) were incubated at 4°C for 1 h with antibody (0.3-0.5 µg) and protein A-Sepharose (Amersham Biosciences). Precipitated immune complexes with protein A-Sepharose were washed three times with a wash buffer (20 mM HEPES-NaOH at pH 7.5, 150 mM NaCl, 5 mM MgCl2, 0.5% Triton-X100, and 10 µM GDP), either with or without AlF-4. Immunoprecipitated proteins were boiled in sample buffer and subjected to Western blotting. Mouse monoclonal antibody 9E10 against c-Myc epitope was purchased from Covance Research Products (Berkeley, CA). Anti-TRPC6 antibody was from Alomone Labs (Jerusalem, Israel).

    SP Inactivates the Kir3.1/3.4 Channels. Activation of the SP receptor inhibits the Kir3.1/3.2 channels with a half-time of 10 to 15 s (Koike-Tani et al., 2005). The Kir3.1/3.2 channels are abundant in brain neurons (Karschin et al., 1996; Liao et al., 1996; Kawano et al., 2004). In contrast, the Kir3.1/3.4 channels are predominantly expressed in cardiac myocytes and play a critical role in regulating heart rate (Krapivinsky et al., 1995a).

    We investigated whether the Kir3.1/3.4 channels are inhibited by SP with a time course similar to that for the Kir3.1/3.2 channels. The SP receptor, Kir3.1, Kir3.4 (or Kir3.2), G1, G2, and GFP were coexpressed in HEK293A cells. In those cells, because of the expression of G12, the Kir3 channels were active from the beginning.

    Application of SP to cells expressing Kir3.1/3.2 (Fig. 1A) as well as to cells expressing Kir3.1/3.4 (Fig. 1B) produced channel inactivation. The half-time (T0.5) of the inactivation was 16 s for Kir3.1/3.2 and 19 s for Kir3.1/3.4 channels (no significant difference). These results indicate that both Kir3.1/3.2 and Kir3.1/3.4 respond to the SP application in a similar manner (Fig. 1C).

    Fig. 1. Inactivation of Kir3.1/3.2 and Kir3.1/3.4 channels by SP. HEK293A cells were transfected with cDNAs for SP receptor, Kir3.1, Kir3.2 (or Kir3.4), G1, G2, and GFP. Application of SP to a cell expressing Kir3.1/3.2 (A) or to a cell expressing Kir3.1/3.4 (B) inactivated the Kir3 current. The whole-cell recordings were used in both A and B. Holding potential was -79 mV; over this steady holding potential, square-shape depolarizing pulses (20 mV/100 ms; upward swinging line), and square-shape hyperpolarizing pulses (50 mV/100 ms; downward swinging line) were periodically (once per 4 s) superimposed. In this way, changes of currents at -59, -79, and -129 mV could be followed by one experimental run. SP application produces a marked decrease in current amplitudes at -59 mV as well as at -129 mV, indicating that inward rectifying currents were reduced. In both A and B, the SP-induced inactivation recovered partially: for Kir3.1/3.2, the recovery (at 300 s) was 35 ± 14% (mean ± S.E.M., n = 6), ranging from 0 to 83%. For Kir3.1/3.4, recovery was 38 ± 13% (at 290 s), ranging from 12.5 to 89%. C, the speed of the SP-induced inactivation (k0.5) was plotted (k0.5 = 1/T0.5, in which T0.5 is half-time). Difference was not statistically significant.

    Gq Binds to the N terminus of Kir3. The N and C termini of Kir 2.1, Kir3.1, Kir3.2, and Kir3.4 were purified as GST-fusion proteins (Fig. 2D). Gq and G12 proteins were obtained using the Sf9-bacurovirus system as described previously (Kozasa, 2004). The pull-down assays showed that Gq coprecipitated with the N termini of Kir3.1, Kir3.2, and Kir3.4 but not with the C termini of any Kir3 channels tested (Fig. 2A). (Note: residues 2-96 of the N terminus would virtually correspond to the whole Kir3.2 N terminus. We therefore presented this domain as N-Kir3.2 in Figs. 2 and 5). The Gq binding to the N termini of Kir3.1, Kir3.2, or Kir3.4 occurred in the presence of GDP as well as in the presence of GDP-AlF-4.Gq did not bind to either the N or the C terminus of Kir2.1. The experiments in Fig. 2A were performed using Gq at 50 nM. We performed additional assays at lower Gq concentrations (2-20 nM) to examine the possible affinity difference in the presence or absence of AlF-4 toward the Kir3.2 N terminus. As shown in Fig. 2B, Gq, at these low concentrations, interacted with the Kir3.2 N terminus, and the interaction was still independent of the presence of AlF-4.

    Fig. 2. Gq binds to the N terminus, but not to the C terminus of Kir3s. A, representative data of GST pull-down assays between Gq and N- or C-terminal cytoplasmic domain of Kir2, Kir3, and GRK2-RH. Purified Gq protein (50 nM) was incubated with 0.5 µg of GST-fusion proteins (40-100 nM) in the presence of GDP or GDP-AlF-4. Both GDP and GDP-AlF-4 forms of Gq bound to the N-terminal domain but not to the C-terminal domains of Kir3.1, Kir3.2, and Kir3.4. However, Gq, in either form, did not bind to the N or C terminus of Kir2.1. B, pull-down assays were conducted at various low concentrations of Gq to examine the affinity of Gq to the N terminus of Kir3.2 in the presence or absence of AlF-4. In this experiment, binding buffer B contained MgCl2 at 1 mM instead of 10 mM. C, binding assays were performed between G12 and the C terminus, as well as between G12 and the N terminus of Kir3 and Kir2.1. Purified G12 protein (40 nM) was incubated with 1 µg of GST-fusion proteins of the N or C terminus of various Kirs. G12 was bound to the N- and C-terminal domains of Kir3, but not to those of Kir2.1. D, GST-fusion proteins used for the pull-down assays were visualized with the Coomassie Brilliant Blue (CBB) staining. GST-fusion proteins examined are the N terminus (residues 1-86) and the C terminus (179-428) of Kir2.1, the N terminus (1-86) and the C terminus (180-501) of Kir3.1, the N terminus (2-96) and the C terminus (191-414) of Kir3.2, and the N terminus (1-91) and the C terminus (185-419) of Kir3.4 and RH domain (1-178) of GRK2.

    Fig. 5. Gq and G12 do not compete with each other for the binding to the N terminus of Kir3.2. To examine possible competition between Gq and G12 for the binding site on the N terminus of Kir3.2, pull-down assays were performed at different concentrations of G12. In the presence of GDP-AlF-4, 0.5 µg of GST or GST-N-Kir3.2 (residues 2-96) was incubated with Gq and G12 at various concentrations.

    As a positive control, the RGS homology domain (residues 1-178) of G protein-coupled receptor kinase 2 (GRK2-RH) was used (Fig. 2, A and B). The binding between GRK2-RH and Gq occurred in the presence of AlF-4, in confirmation of Carman et al. (1999).

    We also performed pull-down assays of G12 with these GST-fusion proteins. G12 bound to both N and C termini of Kir3.1, 3.2, and 3.4. However, practically no binding was observed between G12 and the N or C terminus of Kir2.1 (Fig. 2C). These results agree with the data of previous publications (Huang et al., 1995; Inanobe et al., 1995; Krapivinsky et al., 1995b; Kunkel and Peralta, 1995), suggesting that our fusion proteins function as expected.

    Gq Does Not Interact with TRPC6A. The experiment of Fig. 2 shows that Gq interacts with the N terminus of Kir3s but not with Kir2. The Kir2 channel, being constitutively active and not regulated by Gq, served as a negative control. We, however, wished to perform another type of control experiment by using a channel that is regulated by Gq. For this purpose, we examined a transient receptor potential canonical type channel; this is a family of nonselective cation channels and is suggested to function downstream of Gq (Clapham et al., 2001).

    HEK293A cells were transfected with cDNAs for Gq and TRPC6A, and coimmunoprecipitation was carried out using anti-Gq/11 antibody. The results showed that TRPC6A did not coprecipitate with Gq in either GDP or GDP-AlF-4 containing solution, whereas Kir3.2 did coprecipitate with Gq (Fig. 3). We did not perform a further GST-pulldown assay between Gq and termini of the TRPC6 channel, because the absence of coprecipitation between Gq and TRPC6A clearly indicated the absence of interaction between the two types of proteins.

    Fig. 3. Gq does not interact with TRPC6A. HEK293A cells were transfected with Gq and TRPC6A (A) or Myc-Kir3.2 (B). Immunoprecipitation was performed using anti-Gq/11 antibody as described under Materials and Methods. Precipitated proteins were subjected to SDS-PAGE and immunoblotting with anti-Gq/11, anti-Myc, and anti-TRPC6 antibodies. A, TRPC6A was not coprecipitated with Gq either in the presence or absence of AlF-4. B, Myc-Kir3.2 was coprecipitated with Gq either in the presence or absence of AlF-4, in agreement with the result of Koike-Tani et al. (2005). Cell lysates were subjected to Western blotting to confirm protein expression (A and B, bottom blots).

    Gq Binding Region of Kir3. To locate the Gq binding region within the Kir3.2 N terminus, various GST-fusion proteins, each consisting of part of the Kir3.2 N terminus, were prepared (Fig. 4). Because Gq is likely to be anchored to the cell membrane through the palmitoyl moiety, we suspected that Gq might bind to part of the Kir3 N terminus that is close to the cell membrane. We, therefore, constructed GST-fusion proteins having the following Kir3.2 N terminus residues that either lack or retain membrane proximal sequences: 1 to 50, 1 to 60, 1 to 70, 1 to 80, 1 to 90, and 51 to 90 (Fig. 4C). Each of these partial domains of the Kir3.2 N terminus was tested for its capability to bind to Gq. As shown in Fig. 4A, GST constructs with the residues 1 to 90, 51 to 90, and 2 to 96 bound to Gq, but those with the residues 1 to 80, 1 to 70, 1 to 60, and 1 to 50 did not. The shortest domain that binds to Gq was the residues 51 to 90, whereas the residues 1 to 80 did not bind to Gq, suggesting that the residues 81 to 90 are critically involved in binding to Gq.

    Fig. 4. Mapping of the Gq binding site and the G12 binding site on the Kir3.2 N terminus. A, Gq binding to various parts of the Kir3.2 N terminus as well as to the whole N terminus of Kir2.1 and the N terminus of Kir3.1 were examined. Gq was incubated with GST-fusion proteins in the presence of GDP or GDP-AlF-4. B, G12 binding was examined on each of the seven different domains of the Kir3.2 N terminus as well as on the whole N terminus region of Kir2.1 and Kir3.1. C, GST-fusion proteins that were used for the pull-down assays were visualized with Coomassie Brilliant Blue (CBB) staining.

    We also investigated the G12 binding to these GST fusion constructs of the Kir3.2 N terminus (Fig, 4B). G12 bound to the 2 to 96 residues, whereas it did not bind to the 1-90 residues, suggesting that the residues 91 to 96 N-terminal domain may be critical for G12 binding.

    Gq Does Not Compete with G for Binding to the Kir3.2 N Terminus. The results of the previous section suggest that on the Kir3.2 N terminus, the domain where Gq attaches and the domain where G12 attaches are in close proximity. This could result in the situation, in which Gq and G12 compete with each other for the binding site. To examine this possibility, we performed pull-down assays between Gq (50 nM) and GST-Kir3.2 N terminus (residues 2-96) at various concentrations of G12. The tests were done in the presence of AlF-4 to keep Gq and G12 dissociated.

    As shown in Fig. 5, Gq and G12, when incubated together (Fig. 5, lanes 2 to 4 from the left), interacted with GST-N-Kir3.2, essentially to the same extent as in the case of Gq alone (lane 1) or G alone (lane 5). Indeed, G12, even at 20 times the concentration of Gq (1 µM; lane 4), did not interfere with the binding of Gq to the N terminus. These results suggest that each of Gq and G, in the presence of AlF-4, binds to the Kir3.2 N terminus separately, but not as a Gq heterotrimer. These data also suggest that Gq and G12 each binds to close but separate domains of the N terminus.

    The Role of Gq/Kir3 Interaction. The focus of this article is the interaction between Gq and Kir3s. In the beginning, we showed an example in which activation of SP receptor, coupled to Gq, inhibits Kir3 channels in a heterologously expressed system (Fig. 1). This type of channel inhibition through the Gq-Kir3 interaction may be in operation in many types of endogenous signaling. Activation of SP receptor is known to inhibit the endogenous Kir3 channels in locus ceruleus neurons (Velimirovic et al., 1995). Activation of the 1-adrenoceptors (Braun et al., 1992; Cho et al., 2001) and thyrotropin-releasing hormone receptor (Lei et al., 2001) inhibits Kir3 channels probably through Gq. Likewise, activation of orexin receptor (Hoang et al., 2003, 2004) or ghrelin receptor (Bajic et al., 2004) has been shown to inhibit Kir channels. All of these receptors are probably coupled to Gq (Offermanns, 2003). Thus, the central theme of this article would be the Gq-Kir3 interaction rather than the SP receptor and the Kir3.

    Gq Binds to Kir3. In the present experiments, we used the GST pull-down assay to study the interaction between Gq and Kir3. Our data show, for the first time, that Gq binds directly and specifically to the N terminus of Kir3s. The binding was direct without intervening proteins, because the binding was examined with the GST pull-down assay using purified proteins. The binding was specific in that Gq interacted with the N terminus of the Kir3s (Kir3.1, -3.2, and -3.4), whereas Gq did not interact with the C terminus of any of the Kir3s tested, and did not Gq interact with the N or C terminus of Kir2.1. Moreover, Gq, which interacts with Kir3, did not interact with TRPC6A, although the signal transduction cascades toward TRPC6 activation and toward Kir3 inhibition are both initiated by Gq.

    Our results differ from those of Clancy et al. (2005), who recently reported that Gq binds to the N terminus of both Kir2.1 and Kir3.2. Part of the discrepancy could have arisen from different conditions used. Clancy et al. (2005) used high concentrations of Gq (350 nM) and GST-fusion proteins (400 nM), whereas we used lower concentrations of these (50 nM for Gq and 70 nM for GST-fusion proteins). The concentration of Gq in our present experiments (50 nM) is similar to those used for functional examinations of Gq on phospholipase C (Hepler et al., 1996).

    The present pull-down assay indicated that Gq, both in GDP-bound form and in GDP-AlF-4-bound form, interacted with the N terminus of Kir3s. This type of interaction seems to be peculiar for a signal transduction involving a G protein. However, similar examples of signaling have been recently reported: G, independent of its conformation, can interact with regulator of G protein signaling (RGS) inside the signal complex consisting of the receptor, G protein and RGS (Benians et al., 2005, Abramov-Newerly et al., 2006). It is possible that inside these types of microspheres consisting of the receptor, Gq, and the channel, the concentration of Gq could become quite high, and only a high-concentration Gq could accomplish the channel inhibition. This type of mechanism could send information more accurately, because a low concentration of stray Gq molecules would become ineffectual. At present, this is a mere speculation.

    How then can this mode of signaling achieve an on-off switching mechanism? Switching can be accomplished by the fact that the GDP form of Gq has a high affinity to G, resulting in the formation of the heterotrimeric G protein, terminating the functional interaction between Gq and the K+ channel.

    Kir3 Domains That Interact with Gq and G. In addition to the above primary conclusion, the present experiments have shed light on the location of Gq binding versus the location of G binding on Kir3s. We showed that the domain encompassing the residues 51 to 90 of the Kir3.2 N terminus was capable of interacting with Gq, whereas the domain from the residues 1 to 80 was not. This suggests, although preliminarily, that the residues from 81 to 90 are critically involved in the interaction with Gq. Nevertheless, for now, it is safe to say that Gq can interact with the residues encompassing 51 to 90 of the Kir3.2. Our results also indicated that G12 binds to residues 2 to 96 but not to residues 1 to 90, of the Kir3.2 N terminus, suggesting that residues 91 to 96 are critical for G12 binding. Finally, the occupancy of Gq and that of G12 on the N terminus of Kir3.2 were independent of each other (Fig. 5), suggesting that the Gq-interacting surface and the G12-interacting surface have no overlap on Kir3.2.

    The N-terminal residues 51 to 96 of Kir3.2, where our results suggest that Gq and G12 attach, contain a domain designated as "slide helix" by Kuo et al. (2003). Based on the crystal structure of prokaryotic Kir (KirBac1.1), Kuo et al. (2003) propose that "slide helix" plays a role in channel gating: the negativity of the helix dipole, being located near the positive charge of the pore helices, may control the movement of the pore helices, thereby controlling the channel gating. Indeed, the attachment sites of Gq and G seem to be located at strategically crucial spots for the functional roles.

    Gq-Kir3 Interaction versus Gi-Kir3 Interaction. According to Dascal and his coworkers (Peleg et al., 2002; Ivanina et al., 2004), Gi, apart from its G sequestering effect, shows inhibitory effects on the basal Kir3 activity. Under ordinary conditions, this capability of Gi is overshadowed by the channel activating effect of G. On the other hand, Gq, having a strong suppressing effect on the channel, could suppress the activating effect of G. This situation could be related to the physiological data that SP effect to inhibit Kir3 channel overshadows the somatostatin effect to activate Kir3 (Velimirovic et al., 1995). In the present experiment, we observed that Gq binds to Kir3 channels only at the N terminus, not at the C terminus of all Kir3 channels tested (Kir3.1, Kir3.2, and Kir3.4). In contrast, Gi binds both N and C termini of Kir3 (Ivanina et al., 2004). This difference could suggest that the mode of Gq action might be basically different from that of Gi.

    Kir3.4 Behaves Similarly to Kir3.2. In the previous study (Koike-Tani et al., 2005), we investigated the properties of Kir3.1 and -3.2, as well as Kir2. In the present study, we have additionally observed the behavior of Kir3.4. Electrophysiological manifestation (Fig. 1) of Kir3.4 and the property of Gq binding were similar to that of Kir3.2. Thus, the cardiac type of GIRK (Kir3.1/3.4) does not seem to behave differently from the brain type (Kir3.1/3.2).

    Acknowledgements

    We thank David Saffen for supplying TRPC6 cDNA and James E. Krause for giving us the substance P receptor cDNA, Shihori Tanabe for instruction on the Sf9-bacurovirus system, Daisuke Urano for suggestions on the GST-fusion protein purification, and Pawinee Yongsatirachot for invaluable help.

    ABBREVIATIONS: SP, substance P; GIRK, G protein-coupled inwardly rectifying K+ channels; GST, glutathione transferase; HEK, human embryonic kidney; PIP2, phosphatidylinositol 4,5-bisphosphate; RGS, regulator of G protein signaling; GRK2-RH, RGS homology domain of G protein-coupled receptor kinase 2.

    1 Current affiliation: Department of Rehabilitation Medicine, Montefiore Medical Center, The University Hospital for the Albert Einstein College of Medicine, Bronx, New York.

【参考文献】
  Abramow-Newerly M, Roy AA, Nunn C, and Chidiac P (2006) RGS proteins have a signalling complex: interactions between RGS proteins and GPCRs, effectors, and auxiliary proteins. Cell Signal 18: 579-591.

Andersson S, Davis DL, Dahlback H, Jornvall H, and Russell DW (1989) Cloning, structure, and expression of the mitochondrial cytochrome P-450 sterol 26-hydroxylase, a bile acid biosynthetic enzyme. J Biol Chem 264: 8222-8229.[Abstract/Free Full Text]

Bajic D, Hoang QV, Nakajima S, and Nakajima Y (2004) Dissociated histaminergic neuron cultures from the tuberomammillary nucleus of rats: culture methods and ghrelin effects. J Neurosci Methods 132: 177-184.

Benians A, Nobles M, Hosny S, and Tinker A (2005) Regulators of G-protein signaling form a quaternary complex with the agonist, receptor, and G-protein. A novel explanation for the acceleration of signaling activation kinetics. J Biol Chem 280: 13383-13394.[Abstract/Free Full Text]

Braun AP, Fedida D, and Giles WR (1992) Activation of alpha 1-adrenoceptors modulates the inwardly rectifying potassium currents of mammalian atrial myocytes. Pflueg Arch Eur J Physiol 421: 431-439.

Carman CV, Parent JL, Day PW, Pronin AN, Sternweis PM, Wedegaertner PB, Gilman AG, Benovic JL, and Kozasa T (1999) Selective regulation of Gq/11 by an RGS domain in the G protein-coupled receptor kinase, GRK2. J Biol Chem 274: 34483-34492.[Abstract/Free Full Text]

Chang MM and Leeman SE (1970) Isolation of a sialogogic peptide from bovine hypothalamic tissue and its characterization as substance P. J Biol Chem 245: 4784-4790.[Abstract/Free Full Text]

Cho H, Nam GB, Lee SH, Earm YE, and Ho WK (2001) Phosphatidylinositol 4,5-bisphosphate is acting as a signal molecule in 1-adrenergic pathway via the modulation of acetylcholine-activated K+ channels in mouse atrial myocytes. J Biol Chem 276: 159-164.[Abstract/Free Full Text]

Clancy SM, Fowler CE, Finley M, Suen KF, Arrabit C, Berton F, Kozasa T, Casey PJ, and Slesinger PA (2005) Pertussis-toxin-sensitive Galpha subunits selectively bind to C-terminal domain of neuronal GIRK channels: evidence for a heterotrimeric G-protein-channel complex. Mol Cell Neurosci 28: 375-389.

Clapham DE, Runnels LW, and Strubing C (2001) The TRP ion channel family. Nat Rev Neurosci 2: 387-396.

Dascal N, Schreibmayer W, Lim NF, Wang W, Chavkin C, DiMagno L, Labarca C, Kieffer BL, Gaveriaux-Ruff C, Trollinger D, et al. (1993) Atrial G protein-activated K+ channel: expression cloning and molecular properties. Proc Natl Acad Sci USA 90: 10235-10239.[Abstract/Free Full Text]

Hepler JR, Biddlecome GH, Kleuss C, Camp LA, Hofmann SL, Ross EM, and Gilman AG (1996) Functional importance of the amino terminus of Gq. J Biol Chem 271: 496-504.[Abstract/Free Full Text]

Hoang QV, Bajic D, Yanagisawa M, Nakajima S, and Nakajima Y (2003) Effects of orexin (hypocretin) on GIRK channels. J Neurophysiol 90: 693-702.[Abstract/Free Full Text]

Hoang QV, Zhao P, Nakajima S, and Nakajima Y (2004) Orexin (hypocretin) effects on constitutively active inward rectifier K+ channels in cultured nucleus basalis neurons. J Neurophysiol 92: 3183-3191.[Abstract/Free Full Text]

Huang CL, Slesinger PA, Casey PJ, Jan YN, and Jan LY (1995) Evidence that direct binding of G beta gamma to the GIRK1 G protein-gated inwardly rectifying K+ channel is important for channel activation. Neuron 15: 1133-1143.

Inanobe A, Morishige KI, Takahashi N, Ito H, Yamada M, Takumi T, Nishina H, Takahashi K, Kanaho Y, Katada T, et al. (1995) G beta gamma directly binds to the carboxyl terminus of the G protein-gated muscarinic K+ channel, GIRK1. Biochem Biophys Res Commun 212: 1022-1028.

Ivanina T, Varon D, Peleg S, Rishal I, Porozov Y, Dessauer CW, Keren-Raifman T, and Dascal N (2004) Gi1 and Gi3 differentially interact with, and regulate, the G protein-activated K+ channel. J Biol Chem 279: 17260-17268.[Abstract/Free Full Text]

Karschin C, Dissmann E, Stuhmer W, and Karschin A (1996) IRK(1-3) and GIRK(1-4) inwardly rectifying K+ channel mRNAs are differentially expressed in the adult rat brain. J Neurosci 16: 3559-3570.[Abstract/Free Full Text]

Kawano T, Zhao P, Nakajima S, and Nakajima Y (2004) Single-cell RT-PCR analysis of GIRK channels expressed in rat locus coeruleus and nucleus basalis neurons. Neurosci Lett 358: 63-67.

Kobrinsky E, Mirshahi T, Zhang H, Jin T, and Logothetis DE (2000) Receptor-mediated hydrolysis of plasma membrane messenger PIP2 leads to K+-current desensitization. Nat Cell Biol 2: 507-514.

Koike-Tani M, Collins JM, Kawano T, Zhao P, Zhao Q, Kozasa T, Nakajima S, and Nakajima Y (2005) Signal transduction pathway for the substance P-induced inhibition of rat Kir3 (GIRK) channel. J Physiol 564: 489-500.[Abstract/Free Full Text]

Kozasa T (2004) Purification of G protein subunits from Sf9 insect cells using hexahistidine-tagged alpha and beta gamma subunits. Methods Mol Biol 237: 21-38.

Krapivinsky G, Gordon EA, Wickman K, Velimirovic B, Krapivinsky L, and Clapham DE (1995a) The G-protein-gated atrial K+ channel IKACh is a heteromultimer of two inwardly rectifying K+-channel proteins. Nature (Lond) 374: 135-141.

Krapivinsky G, Krapivinsky L, Wickman K, and Clapham DE (1995b) G binds directly to the G protein-gated K+ channel, IKACh. J Biol Chem 270: 29059-29062.[Abstract/Free Full Text]

Kubo Y, Reuveny E, Slesinger PA, Jan YN, and Jan LY (1993) Primary structure and functional expression of a rat G-protein-coupled muscarinic potassium channel. Nature (Lond) 364: 802-806.

Kunkel MT and Peralta EG (1995) Identification of domains conferring G protein regulation on inward rectifier potassium channels. Cell 83: 443-449.

Kuo A, Gulbis JM, Antcliff JF, Rahman T, Lowe ED, Zimmer J, Cuthbertson J, Ashcroft FM, Ezaki T, and Doyle DA (2003) Crystal structure of the potassium channel KirBac1.1 in the closed state. Science (Wash DC) 300: 1922-1926.[Abstract/Free Full Text]

Leaney JL, Dekker LV, and Tinker A (2001) Regulation of a G protein-gated inwardly rectifying K+ channel by a Ca2+-independent protein kinase C. J Physiol 534: 367-379.[Abstract/Free Full Text]

Lei Q, Talley EM, and Bayliss DA (2001) Receptor-mediated inhibition of G protein-coupled inwardly rectifying potassium channels involves Gq family subunits, phospholipase C, and a readily diffusible messenger. J Biol Chem 276: 16720-16730.[Abstract/Free Full Text]

Liao YJ, Jan YN, and Jan LY (1996) Heteromultimerization of G-protein-gated inwardly rectifying K+ channel proteins GIRK1 and GIRK2 and their altered expression in weaver brain. J Neurosci 16: 7137-7150.[Abstract/Free Full Text]

Logothetis DE, Kurachi Y, Galper J, Neer EJ, and Clapham DE (1987) The beta gamma subunits of GTP-binding proteins activate the muscarinic K+ channel in heart. Nature (Lond) 325: 321-326.

Mao J, Wang X, Chen F, Wang R, Rojas A, Shi Y, Piao H, and Jiang C (2004) Molecular basis for the inhibition of G protein-coupled inward rectifier K+ channels by protein kinase C. Proc Natl Acad Sci USA 101: 1087-1092.[Abstract/Free Full Text]

Offermanns S (2003) G-proteins as transducers in transmembrane signalling. Prog Biophys Mol Biol 83: 101-130.

Peleg S, Varon D, Ivanina T, Dessauer CW, and Dascal N (2002) Gi controls the gating of the G protein-activated K+ channel, GIRK. Neuron 33: 87-99.

Sharon D, Vorobiov D, and Dascal N (1997) Positive and negative coupling of the metabotropic glutamate receptors to a G protein-activated K+ channel, GIRK, in Xenopus oocytes. J Gen Physiol 109: 477-490.[Abstract/Free Full Text]

Slesinger PA, Reuveny E, Jan YN, and Jan LY (1995) Identification of structural elements involved in G protein gating of the GIRK1 potassium channel. Neuron 15: 1145-1156.

Stanfield PR, Nakajima Y, and Yamaguchi K (1985) Substance P raises neuronal membrane excitability by reducing inward rectification. Nature (Lond) 315: 498-501.

Takeda Y, Chou KB, Takeda J, Sachais BS, and Krause JE (1991) Molecular cloning, structural characterization and functional expression of the human substance P receptor. Biochem Biophys Res Commun 179: 1232-1240.

Velimirovic BM, Koyano K, Nakajima S, and Nakajima Y (1995) Opposing mechanisms of regulation of a G-protein-coupled inward rectifier K+ channel in rat brain neurons. Proc Natl Acad Sci USA 92: 1590-1594.[Abstract/Free Full Text]

Yasukawa T, Kanei-Ishii C, Maekawa T, Fujimoto J, Yamamoto T, and Ishii S (1995) Increase of solubility of foreign proteins in Escherichia coli by coproduction of the bacterial thioredoxin. J Biol Chem 270: 25328-25331.[Abstract/Free Full Text]

作者单位：Departments of Anatomy & Cell Biology (T. Ka., P.Z., C.V.F., Y.N.) and Pharmacology (T. Ko., S.N.), College of Medicine, University of Illinois at Chicago, Illinois