Skip to main content
NCBI home page
The Journal of Physiology logoLink to The Journal of Physiology
. 1999 Feb 1;514(Pt 3):655–665. doi: 10.1111/j.1469-7793.1999.655ad.x

Wortmannin, an inhibitor of phosphatidylinositol kinases, blocks the MgATP-dependent recovery of Kir6.2/SUR2A channels

Lai-Hua Xie *, Makoto Takano *, Masafumi Kakei *, Midori Okamura *, Akinori Noma *
PMCID: PMC2269097  PMID: 9882737

Abstract

  1. In order to investigate the mechanism underlying MgATP-dependent recovery of ATP-sensitive potassium (KATP) channels, we expressed Kir6.2/SUR2A (inwardly rectifying K+ channel subunit/sulfonylurea receptor) or C-terminal-truncated Kir6.2 (Kir6.2ΔC26) in COS7 cells (Green monkey kidney cells), and carried out inside-out patch clamp experiments.

  2. After patch excision in ATP-free internal solution, the activity of Kir6.2/SUR2A channels could be maximally recovered by the application of 5 mM MgATP. Subsequent application of 100 μM Ca2+ induced a rapid decay of Kir6.2/SUR2A activity to 11·6 ± 1·1 % (mean ± s.e.m.) of the control level (Ca2+-induced run-down; n = 64).

  3. MgATP (5 mM) recovered 99·4 ± 4·2 % (n = 13) of the Ca2+-induced run-down. Protein kinase inhibitors such as W-7, H-7, H-8 and genistein did not inhibit this reaction. However, wortmannin, an inhibitor of phosphatidylinositol 3- and 4-kinases, blocked the MgATP-dependent recovery in a concentration-dependent manner; the magnitudes of recovery were 35·7 ± 7·2 % (10 μM) and 4·3 ± 2·5 % (100 μM) of the Ca2+-induced run-down.

  4. MgUDP (10 mM) reversed the Ca2+-induced run-down of Kir6.2/SUR2A channels by 60·4 ± 7·6 % (n = 5). Wortmannin failed to modify this reaction.

  5. Kir6.2ΔC26 channels, which opened in the absence of SUR2A, were less sensitive to Ca2+; Kir6.2ΔC26 channels were inactivated to 44·8 ± 4·4 % (n = 14) by 100 μM Ca2+. MgATP recovered the Ca2+-induced run-down of Kir6.2ΔC26 by 89·8 ± 7·7 % (n = 9), and 100 μM wortmannin inhibited this reaction (1·8 ± 2 %, n = 7).

  6. Application of 10 μM phosphatidylinositol-4,5-bisphosphate (PI-4,5-P2) recovered the activity of Kir6.2/SUR2A channels after Ca2+-induced run-down (104·3 ± 6·4 %, n = 10). Even after the MgATP-dependent recovery was blocked by 100 μM wortmannin, PI-4,5-P2 reactivated the channels (102·3 ± 8·6 %, n = 5). Similar results were obtained with Kir6.2ΔC26.

  7. These results suggest that the entity of MgATP-dependent recovery may be membrane lipid phosphorylation rather than protein phosphorylation, and that synthesis of PI-4,5-P2 or phosphatidylinositol-3,4,5-trisphosphate may upregulate Kir6·2 channels.

ATP-sensitive potassium (KATP) channels undergo ‘run-down’ after removal of intracellular ATP, but can be recovered by the application of MgATP in native cells (Findlay & Dunne, 1986; Ohno-Shosaku et al. 1987; Takano et al. 1990). KATP channels reconstituted by the co-expression of the inwardly rectifying K+ channel subunit (Kir6.2) and sulfonylurea receptor (SUR) genes or with a truncated form of Kir6.2 (Kir6.2ΔC26) gene alone also retained similar properties of run-down and MgATP-dependent recovery (Takano et al. 1996, 1998; Tucker et al. 1997; Okuyama et al. 1998). Despite intense studies, the mechanisms of MgATP-dependent recovery of KATP channel run-down have not been fully understood (Findlay, 1988; Furukawa et al. 1994, 1996; Hussain & Wareham, 1994). It has been speculated from the following observations that hydrolysis of ATP and phosphorylation are involved in the MgATP-dependent recovery. First, the recovery of KATP channel activity was not observed in the absence of Mg2+, or when ATP was replaced with a non-hydrolysable ATP analogue, 5′-adenylylimidodiphosphate (AMP-PNP). Secondly, MgATP-dependent recovery proceeded with a slow time course. However, Furukawa et al. (1994) suggested that the protein phosphorylation by serine/threonine protein kinases may not be involved in the MgATP-dependent recovery of cardiac KATP channels.

Another plausible mechanism underlying the MgATP-dependent recovery could be lipid phosphorylation. It was recently reported that phosphatidylinositol-4,5-bisphosphate (PI-4,5-P2) activated both the cardiac KATP channel and the reconstituted KATP channel (Hilgemann & Ball, 1996; Fan & Makielski, 1997). In the plasma membrane PI-4,5-P2 is produced by the consecutive phosphorylation of phosphatidylinositol (PI) and phosphatidylinositol-4-monophosphate (PI-4-P). Therefore, the synthesis of PI-4,5-P2 might be the entity of the MgATP-dependent recovery of KATP channels. In order to test this hypothesis, we examined the effects of a lipid kinase inhibitor, wortmannin, on the KATP channels reconstituted with Kir6.2 + SUR2A or C-terminus-truncated Kir6.2 (Kir6.2ΔC26) alone. Wortmannin is known to be a specific inhibitor of PI 3-kinase, but it has also recently been shown that wortmannin can also inhibit PI 4-kinase at higher concentrations (Nakanishi et al. 1995). In the present study, we will demonstrate that among a variety of kinase inhibitors so far examined, wortmannin, an inhibitor of membrane lipid kinase, successfully blocks MgATP-dependent recovery.

METHODS

Molecular biology

Kir6.2 cDNA (Takano et al. 1996), SUR2A cDNA (a gift from Professor S. Seino, Chiba University) and green fluorescent protein (GFP) cDNA (Moriyoshi et al. 1996) were subcloned into the pCI vector which possesses the CMV promoter/enhancer (Promega, Madison, WI, USA). Kir6.2ΔC26, a truncated form of Kir6.2 in which the last 26 amino acids of the C-terminus had been deleted, was made by introducing a stop codon at the appropriate residues by site-directed mutagenesis using PCR. Kir6.2ΔC26 was also subcloned into the pCI vector.

Transfection

COS7 cells (Green monkey kidney cells; Riken, Wako, Japan) were plated on coverslips in 35 mm culture dishes and cultured in Dulbecco's modified Eagle's medium supplemented with 10 % (v/v) fetal calf serum. Mixtures of the following amounts of vectors (μg per dish) were cotransfected into COS7 cells using Lipofectamine reagent and OPTI-MEM (Gibco): (1) 0.8 Kir6.2, 0.8 SUR2A and 0.4 GFP, (2) 1.6 Kir6.2ΔC26 and 0.4 GFP. The transfected cells could be identified with green fluorescence 24-48 h after the transfection.

Electrophysiology

A coverslip was transferred to a recording chamber and perfused with physiological saline containing (mM): 140 NaCl, 5.4 KCl, 1 Na2HPO4, 1.8 CaCl2, 0.5 MgCl2, 5 glucose and 5 Hepes (pH 7.4 with NaOH). Gigaohm seals were obtained in physiological saline using patch pipettes filled with a high-K+ external solution containing (mM): 140 KCl, 1.8 CaCl2, 0.5 MgCl2 and 5 Hepes (pH 7.4 with KOH). The membrane patch was then excised after the bath solution was switched to control internal solution containing (mM): 140 KCl, 2 MgCl2, 2 EGTA and 5 Hepes (pH 7.2 with KOH). The MgCl2 was omitted in the experiments analysing the activation effect of PI-4,5-P2. In the case of Ca2+-containing solution, EGTA was omitted from the control internal solution, and 100 μM or 1 mM CaCl2 was added. All experiments were carried out in the inside-out mode of the patch clamp technique (Hamill et al. 1981) using an Axopatch 200B amplifier (Axon Instruments). The electrode resistance was between 3 and 6 MΩ. ATP, UDP, H-7, H-8 and W-7 were dissolved in the control internal solution and pH was readjusted with KOH. Genistein and wortmannin were firstly dissolved in dimethyl sulfoxide (DMSO) to a concentration of 50 mM as a stock solution and were used at a concentration of 10-100 μM in the control internal solution. The final concentration of DMSO was ≤ 0.2 % (v/v). PI-4,5-P2 was dispersed in the control internal solution and was sonicated for > 20 min on ice in the dark. This procedure was essential to obtain consistent effects of PI-4,5-P2. When DMSO was included in the test solution containing PI-4,5-P2, the effect of PI-4,5-P2 was not stable for unknown reasons. Test solutions were applied either by bath perfusion or by a Y-tube apparatus which could exchange solutions within ∼50 ms (Ogata & Tatebayashi, 1991). All experiments were carried out at 22-25°C.

MgATP, Na2UDP, H-7 and genistein were purchased from Sigma; PI-4,5-P2 and H-8 were from Calbiochem. W-8 was from RBI and wortmannin was from Wako Pure-Chemical Industries Ltd (Osaka, Japan).

Data analysis

Current signals were stored on a DAT tape recorder (TEAC). For computer analysis, the signals were played back and digitized through a Digipack 1200A interface (Axon Instruments). Data analysis was carried out using commercial software (pCLAMP 6 and 7, Axon Instruments) on an IBM-PC compatible computer. The amplitude of the mean patch current (the integral of open channel current divided by the time for integration; MPC) was measured for 5-10 s. Statistical data are expressed as means ±s.e.m. Statistical difference was estimated by Student's paired or unpaired t test.

RESULTS

Evaluation of the magnitude of MgATP-dependent recovery of Kir6.2/SUR2A channels

We firstly confirmed that the ATP-sensitive K+ (KATP) channels reconstituted with Kir6.2 and SUR2A retained the properties of run-down and MgATP-dependent recovery as previously reported in KATP channels of the native tissues. When the patch membrane was excised into ATP-free control internal solution, an abrupt opening of KATP channels was observed in almost all GFP-positive cells. The inward single channel conductance was ∼66 pS. In ATP-free internal solution, the channel activity spontaneously decreased with various time courses and magnitudes (spontaneous run-down). Channel activity could be recovered by the application of 5 mM MgATP for ∼1 min (not shown). The channel activity remained stable after this procedure (a in Fig. 1A). We then applied 100 μM Ca2+ to the patch for 11.6 ± 0.5 s, and the channel activity declined quickly to 11.6 ± 1.1 % (n = 64) of the control level and did not recover after the wash-out of Ca2+ (Ca2+-induced run-down; b in Fig. 1A). The application of 5 mM MgATP completely inhibited the residual KATP channel activity. The subsequent removal of MgATP revealed the recovery of KATP channel activity (c in Fig. 1A).

Figure 1. Run-down and recovery of Kir6.2/SUR2A channels.

Figure 1

Open in a new tab

A, MgATP-dependent recovery. The current trace was recorded 1 min after the channel activity was maximally recovered by the application of 5 mM MgATP. At a, the mean patch current (MPC) was 544.7 pA. The patch was then exposed to 100 μM Ca2+ for 18 s as indicated by the bar above the trace. At b, MPC was decreased to 31.3 pA. The subsequent application of 5 mM MgATP completely inhibited the residual channels. Upon wash-out of MgATP, MPC was recovered to 568.3 pA at c. The magnitude of the recovery in this experiment was (568.3 - 31.3)/(544.7 - 31.3) × 100 = 104.4 %. Thereafter, the same protocol was repeated using 5 mM K2ATP. The magnitude of the recovery was 2.1 %. B, UDP-induced reactivation. At the beginning of the current trace, the maximal recovery was confirmed by the application of 5 mM MgATP. At d, MPC was 572.6 pA. The application of 100 μM Ca2+ for 8 s induced rapid run-down. MPC at e was 31.2 pA. The application of 10 mM UDP produced rapid and reversible activation of the channels. At f, MPC was 405.1 pA. The magnitude of the recovery by UDP was therefore 69.1 %. It should be noted that, after the channel activity had been maximally recovered by the application of 5 mM MgATP, 10 mM UDP did not produce further activation. In all of the current traces in the present paper, the transmembrane holding potential was -40 mV and the dotted lines indicate the closed channel levels.

In the present study, we evaluated the magnitude of the recovery of KATP channels with reference to the Ca2+-induced run-down; mean patch currents (MPCs) were measured at a, b and c, and the magnitude of recovery was calculated as (c - b)/(a - b) × 100 (%). For example, the magnitude of MgATP-dependent recovery in Fig. 1A was 104.4 %. This reaction completely disappeared either by omitting Mg2+ (K2ATP in Fig. 1A) or by replacing ATP with AMP-PNP (data not shown).

Run-down of KATP channels was also reversed by the application of Mg-nucleotide diphosphates (NDPs). The stimulatory effects of NDPs are mediated by the nucleotide binding domains (NBDs) of SUR (Gribble et al. 1997). As shown in Fig. 1B, UDP reactivated KATP channels which underwent Ca2+-induced run-down in a reversible manner. However, the magnitude of the UDP-induced reactivation was significantly smaller (60.4 ± 7.6 %) than that of the MgATP-dependent recovery (99.4 ± 4.2 %, P < 0.01).

Wortmannin does not block UDP-induced activation, but blocks MgATP-dependent recovery

In the experiments shown in Fig. 2, we examined the effect of wortmannin on both UDP-induced activation and MgATP-dependent recovery. Wortmannin (10 μM) had no direct effect on the activities of KATP channels. It was clear from Fig. 2A that wortmannin did not block UDP-induced activation of KATP channels. The magnitudes of UDP-induced activation in the presence and absence of wortmannin were 57.1 ± 8.2 and 60.4 ± 7.6 %, respectively (Fig. 2C, P > 0.05).

Figure 2. The effect of wortmannin on UDP-induced reactivation and MgATP-dependent recovery.

Figure 2

Open in a new tab

A, after Ca2+-induced run-down reached the steady state (from 1432.3 to 128.9 pA), the application of 10 mM UDP was repeated. Upon wash-out of UDP, a transient increase of the channel activity was observed, indicating that UDP partially inhibited the channel activity. At the end of UDP application, MPC was 866.1-899.0 pA before the exposure to 10 μM wortmannin (WMN). Under the influence of wortmannin, MPC at the end of UDP application was 839.6-848.0 pA. B, Ca2+-induced run-down was completely recovered by the application of 5 mM MgATP within 1 min (104.6 %). Under the influence of 10 μM WMN, the application of MgATP for a similar duration only produced a partial recovery (34.9 %). C, summary of the magnitudes of recovery. In the presence and absence of 10 μM WMN, the magnitude of UDP-induced reactivation was 57.1 ± 8.2 and 60.4 ± 7.6 %, respectively. Numbers in parentheses above the bars indicate the number of experiments. There was no significant difference between these two values (P > 0.05). The magnitude of MgATP-dependent recovery in the presence of 10 μM WMN was 35.7 ± 7.2 %, which was significantly smaller than the control (99.4 ± 4.2 %, **P < 0.01).

Figure 2B compares the MgATP-dependent recoveries of the KATP channel before and after the application of wortmannin. The application of MgATP was interrupted by the application of MgATP-free solution for short periods. The mean patch current in ATP-free solution gradually increased in parallel with the perfusion time of MgATP, demonstrating that a rather long time was needed to give maximum recovery. We therefore applied MgATP for at least 1 min in order to evaluate the magnitude of MgATP-dependent recovery, which was 99.4 ± 4.2 % (n = 13) in the control conditions. In contrast to the UDP-induced activation, the application of 10 μM wortmannin clearly blocked the MgATP-dependent recovery. Further prolongation of MgATP application did not increase the magnitude of recovery (data not shown). The inhibitory effect of wortmannin was irreversible. The magnitude of MgATP-dependent recovery was only 35.7 ± 7.2 % in the presence of 10 μM wortmannin (n = 10). As summarized in Fig. 2C, this value was significantly lower than that of the control group (P < 0.01).

Protein kinase inhibitors do not block MgATP-dependent recovery

Wortmannin inhibits calmodulin-myosin light chain kinase (MLCK) as well as PI 3-kinase and PI 4-kinase (Nakanishi et al. 1995). Therefore, we compared the inhibitory effects of wortmannin with those of other kinase inhibitors. In the experiments shown in Fig. 3A, we first produced the Ca2+-induced run-down of the KATP channel. The magnitudes of Ca2+-induced run-down were not significantly different between different sets of experiments (P > 0.05). We then applied 5 mM MgATP to the patch membrane in the presence of various kinase inhibitors, and compared the magnitudes of MgATP-dependent recovery with reference to the Ca2+-induced run-down. In the 100 μM wortmannin-treated group, MgATP-dependent recovery was clearly blocked even after prolonged exposure to MgATP (Fig. 3Aa). In contrast, 100 μM H-7, an inhibitor of protein kinase A (PKA) and protein kinase C (PKC), did not block the MgATP-dependent recovery (Fig. 3Ab). The same was the case for H-8, an inhibitor of protein kinase G (PKG) and PKA, and genistein, an inhibitor of tyrosine kinase (traces not shown). W-7, an inhibitor of calmodulin-dependent MLCK, directly inhibited KATP channels in a reversible manner. However, MgATP-dependent recovery examined in the residual KATP channels was not affected by W-7 (Fig. 3Ac).

Figure 3. Effects of protein kinase inhibitors and WMN on the MgATP-dependent recovery of Kir6.2/SUR2A channels.

Figure 3

Open in a new tab

A, after run-down of the channel activity was induced by 100 μM Ca2+, MgATP (5 mM) was applied to the intracellular side of the membrane in the presence of 100 μM wortmannin (a), 100 μM H-7 (b) and 10 μM W-7 (c). The bars above the current traces indicate the applications of Ca2+ and MgATP, and the bars under the current traces indicate the applications of wortmannin, H-7 and W-7. The magnitude of MgATP-dependent recovery was 14.0, 99.1 and 118.2 % in a, b and c, respectively. The application of 10 μM W-7 decreased the channel activity to 38.6 % in a reversible manner. B, summary of the magnitudes of MgATP-dependent recovery in the presence of 100 μM wortmannin, 100 μM H-7, 100 μM H-8, 100 μM genistein (Gen) and 10 μM W-7. The magnitudes of recovery are shown in the text. The numbers in parentheses above each bar indicate the number of experiments. WMN (100 μM) significantly decreased the magnitude of recovery (** P < 0.01). No significant difference was found in the other experiments with H-7, H-8, genistein and W-7 (P > 0.05).

Figure 3B summarizes the magnitudes of MgATP-dependent recovery examined in the presence of various kinase inhibitors. Clearly, wortmannin inhibited MgATP-dependent recovery in a concentration-dependent manner; magnitudes of recovery were 35.7 ± 7.2 % (n = 10) at 10 μM (Fig. 2C) and 4.3 ± 2.5 % (n = 8) at 100 μM (Fig. 3B), which were significantly lower than the control group (99.4 ± 4.2 %, n = 13, P < 0.01). In contrast, no significant differences were found between the control group and the groups treated with protein kinase inhibitors (H-7, H-8, genistein and W-7, P > 0.05). It was reported that the inhibitory effect of wortmannin was persistent and irreversible (Nakanishi et al. 1995). In accordance with this, pretreatment with wortmannin successfully inhibited the MgATP-dependent recovery as shown in Figs 4, 5 and 6.

Figure 4. Ca2+-induced run-down and MgATP-dependent recovery of Kir6.2ΔC26 channels.

Figure 4

Open in a new tab

A, the current trace was recorded 30 s after the channel activity was maximally recovered by the application of 5 mM MgATP. Ca2+ (100 μM) was applied for 10 s, and MPC was decreased from 29.9 pA to 13.0 pA. MgATP (5 mM) recovered the channel activity to an extent comparable to the initial level (MPC = 28.6 pA). B, the effect of 100 μM wortmannin pretreatment. Wortmannin was perfused for 1 min. Following application of 100 μM Ca2+ MPC decreased from 19.8 to 5.7 pA. No significant recovery was found after the application of MgATP; at the end of the current trace, MPC was 6.1 pA. C, magnitudes of Ca2 +-induced run-down of Kir6.2/SUR2A and Kir6.2ΔC26 alone. In Kir6.2/SUR2A, 100 μM and 1 mM Ca2+ were applied for 11.6 ± 0.5 s, and MPCs of the residual channel activity were 11.6 ± 1.1 and 4.5 ± 1.4 %, respectively. In Kir6.2ΔC26, prolonged application of Ca2+ (27.4 ± 2.1 s) produced significantly smaller inhibitory effects (**P < 0.01); 44.8 ± 4.4 % by 100 μM Ca2+ and 30.1 ± 5.9 % by 1 mM Ca2+, respectively. D, MgATP-dependent recovery. In the control condition, the recovery was 89.8 ± 7.7 % of Ca2+-induced run-down. Wortmannin pretreatment successfully inhibited the recovery (1.8 ± 2.0 %; **P < 0.01). Numbers in parentheses above each bar indicate the number of experiments.

Figure 5. Effect of PI-4,5-P2 on Kir6.2/SUR2A channels.

Figure 5

Open in a new tab

A, after the run-down induced by 100 μM Ca2+, PI-4,5-P2 (PIP2; 10 μM) was applied to the patch membrane for 30 s. Channel activity persisted even after wash-out of PI-4,5-P2. The magnitude of the recovery was 101.6 %. B, the patch membrane was pretreated with 100 μM wortmannin for 1 min before the Ca2+-induced run-down. The inhibition of MgATP-dependent recovery was confirmed by the application of 5 mM MgATP (7.3 % recovery). Even under these conditions, 10 μM PI-4,5-P2 completely restored the channel activity (110.9 % recovery). In both A and B, the dotted lines indicate the closed channel levels. The applications of MgATP, Ca2+, PI-4,5-P2 and wortmannin are indicated by bars above the current traces. C, percentages of recovery of KATP channels by MgATP and PI-4,5-P2 with or without the pretreatment of wortmannin. The values of recovery are shown in the text. The magnitude of MgATP-dependent recovery was significantly smaller in the wortmannin-pretreatment group (** P < 0.01). The effect of PI-4,5-P2 was not significantly altered by wortmannin pretreatment (P > 0.05). Numbers in parentheses above each bar indicate the number of experiments.

Figure 6. PI-4,5-P2 also recovered Kir6.2ΔC26 channels.

Figure 6

Open in a new tab

A, the top panel shows the current trace of Kir6.2ΔC26. The amplitude histograms a, b and c were constructed for 10 s at the corresponding times in the top panel. a, control; MPC, 1.76 pA. b, after the Ca2+-induced run-down; MPC, 0.46 pA. c, after the application of 10 μM PI-4,5-P2; MPC, 2.18 pA. The magnitude of recovery was 132.3 %. B, the effect of pretreatment with 100 μM wortmannin. C, summary of the magnitude of recovery caused by 10 μM PI-4,5-P2. No significant difference was found between the presence (108.1 ± 20.1 %) and absence of wortmannin (96.4 ± 19.1 %, P > 0.05). Numbers in parentheses above each bar indicate the number of experiments.

Wortmannin inhibits MgATP-dependent recovery of C-terminus-truncated Kir6.2 channel subunit

In the absence of SURs, Kir6.2 was not operative. SURs conferred the functional activity on Kir6.2 (Inagaki et al. 1995; Sakura et al. 1995). Among the members of the ATP-binding cassette superfamily, neither CFTR (cystic fibrosis transmembrane conductance regulator) nor MDR1 (multi-drug resistance gene) could activate Kir6.2 (Takano et al. 1996). However, when the last 26 or 36 amino acids of the C-terminus of Kir6.2 were deleted, Kir6.2ΔC26 or ΔC36 alone showed functional channel activity. The run-down and following MgATP-dependent recovery was observed in Kir6.2ΔC26 and Kir6.2ΔC36 alone, indicating that this mechanism was intrinsic to Kir6.2, rather than SURs (Tucker et al. 1997).

In Fig. 4, we examined the effect of wortmannin on the MgATP-dependent recovery of a truncated form of Kir6.2 (Kir6.2ΔC26). Most patches contained only four or five channels, suggesting that the level of functional expression of Kir6.2ΔC26 was lower than that of Kir6.2/SUR2A. Furthermore, the Ca2+-induced run-down was less significant and slower in development when compared with Kir6.2/SUR2A. Although we prolonged the duration of Ca2+ application (27.4 ± 2.1 s), 100 μM and 1 mM Ca2+ inactivated Kir6.2ΔC26 only to 44.8 ± 4.4 and 30.1 ± 5.9 % of the control level, respectively. These values are significantly larger (P < 0.01) than 11.6 ± 1.1 % at 100 μM and 4.5 ± 1.4 % at 1 mM, respectively, obtained by the short application (11.6 ± 0.5 s) of Ca2+ to Kir6.2/SUR2A.

Despite the above differences, MgATP recovered the channel activity of Kir6.2ΔC26 which underwent Ca2+-induced run-down by 89.8 ± 7.7 %. Pretreatment with 100 μM wortmannin completely blocked the MgATP-dependent recovery (1.8 ± 2.0 %) as shown in Fig. 4B and D.

PI-4,5-P2 activates KATP channels even after wortmannin treatment

Wortmannin inhibits PI 4-kinase as well as PI 3-kinase. The inhibition of PI 4-kinase should decrease the production of PI-4-P, and consequently PI-4,5-P2. Therefore, the blockade of MgATP-dependent recovery may be due to the inhibition of the production of the above compounds. If this is the case, the above products of membrane lipid phosphorylation should activate KATP channels even after the application of wortmannin.

In Fig. 5A, as expected, KATP channels which had previously undergone Ca2+-induced run-down, were reactivated by 10 μM PI-4,5-P2 almost completely to the initial level. The channel activity persisted after wash-out of PI-4,5-P2. In Fig. 5B, pretreatment with 100 μM wortmannin blocked the MgATP-dependent recovery, demonstrating that the effect of wortmannin was irreversible. Even under these conditions, 10 μM PI-4,5-P2 still showed a strong activating effect. Figure 5C summarizes the magnitude of the PI-4,5-P2 effect in the presence (102.3 ± 8.6 %, n = 5) and absence (104.3 ± 6.4 %, n = 10) of 100 μM wortmannin. No significant difference was found between these two values (P > 0.05).

Essentially the same results were obtained for KATP channels reconstituted with the truncated form of Kir6.2. Figure 6A illustrates the effect of PI-4,5-P2 on the Kir6.2ΔC26 channels. After run-down of the channel activity was caused by 1 mM Ca2+, 10 μM PI-4,5-P2 was applied. The channel activity increased gradually, and reached a steady level ∼1 min after the application of PI-4,5-P2. The amplitude histograms a, b and c shown in Fig. 6A were constructed from the segments indicated by a, b and c in the top panel. The amplitude histograms demonstrated that the unitary amplitude (∼3.1 pA) did not change throughout the experiments. These results revealed that PI-4,5-P2 increased the open probability but had no effect on the single channel conductance during the reactivation of KATP channels. Similar results were obtained after pretreatment of the patch membrane with 100 μM wortmannin (Fig. 6B). As summarized in Fig. 6C, PI-4,5-P2 reactivated the channel activity equally with (108.1 ± 20.1 %, n = 6) or without (96.4 ± 19.1 %, n = 4) the pretreatment of wortmannin. No significant difference was found between these values (P > 0.05).

DISCUSSION

The decline of KATP channel activity after excising the patch membrane from the cell has been termed ‘run-down’ and intense studies have been carried out on this subject. Since the non-hydrolysable analogue of ATP, AMP-PNP, had no effect on recovery and Mg2+ was necessary for the recovery, many researchers suggested that MgATP acted as a phosphoryl group donor in protein kinase-mediated phosphorylation reactions (Findlay & Dunne, 1986; Ohno-Shosaku et al. 1987; Takano et al. 1990). However, such protein kinases have not, as yet, been identified (Furukawa et al. 1994).

In the present study, we have observed for the first time that the MgATP-dependent recovery of KATP channels was blocked by wortmannin, an inhibitor of phosphoinositide kinases. Protein kinase inhibitors or calmodulin-MCLK inhibitor could not block MgATP-dependent recovery. These results suggested that MgATP might serve as a substrate for the PI kinases, not protein kinases. In accordance with this explanation, PI-4,5-P2, a product of the phosphatidylinositol phosphorylation pathways, recovered the activity of KATP channels after run-down, even after MgATP-dependent recovery had been blocked by wortmannin. PI-4,5-P2 also potentiated the activities of cloned Kir channels such as IRK1, ROMK1 and GIRK1 (Huang et al. 1998). Therefore, members of Kir channels may be generally upregulated by MgATP through the production of PI-4,5-P2.

Biochemical studies have revealed that wortmannin inhibits PI 4-kinase as well as PI 3-kinase at the IC50 range of 10−7-10−6 M (Nakanishi et al. 1995). In the present study, we used 10−5 and 10−4 M wortmannin, which blocked the MgATP-dependent recovery by 35.7 ± 7.2 and 4.3 ± 2.5 %, respectively. Since these concentrations are much higher than those used in the biochemical studies, we could not exclude the possibility that an unknown protein kinase could be blocked by wortmannin. However, it should be noted that discrepancies in effective concentrations used in the patch clamp experiments and biochemical experiments are also reported for many other compounds such as glibenclamide (French et al. 1990; Miller et al. 1991; Venkatesh et al. 1991).

In the cell membrane, PI 4-kinase phosphorylates PI to produce PI-4-P and then PIP-kinase phosphorylates PI-4-P to produce PI-4,5-P2. PI 3-kinase phosphorylates PI, PI-4-P and PI-4,5-P2, and produces PI-3-P (phosphatidylinositol-3-monophosphate), PI-3,4-P2 (phosphatidylinositol-3,4-bisphosphate) and PI-3,4,5-P3 (phosphatidylinositol-3,4,5-trisphosphate), respectively (Fig. 7). Fan & Makielski (1997) reported that anionic phospholipids such as PI-4-P and phosphatidylserine (PS) also activated KATP channels. Therefore, membrane phospholipids other than PI-4,5-P2 may also be critical requirements in the recovery of the KATP channel. The inhibition of PI 4-kinase should decrease the amount of PI-4-P, and consequently, the amount of PI-4,5-P2. The inhibition of PI 3-kinase should decrease PI-3-P, PI-3,4-P2 and PI-3,4,5-P3. Since PI-3-P, PI-3,4-P2 and PI-3,4,5-P3 are not commercially available, we could not examine their effects on the KATP channel. It was therefore difficult to conclude which substance is most critical for the MgATP-dependent recovery. For the same reason, it was also difficult to exclude the possibility that the wortmannin effect was due to the inhibition of PI 3-kinase rather than PI 4-kinase. In order to investigate this subject further, selective inhibitors of PIP-kinase or PI 4-kinase are required.

Figure 7. Schematic diagram of phosphatidylinositol metabolism.

Figure 7

Open in a new tab

PI, phosphatidylinositol. PI-3-P, phosphatidylinositol-3-monophosphate. PI-4-P, phosphatidylinositol-4-monophosphate. PI-3,4-P2, phosphatidylinositol-3,4-bisphosphate. PI-4,5-P2, phosphatidylinositol-4,5-bisphosphate. PIP3, phosphatidylinositol-3,4,5-trisphosphate. IP3, inositol-1,4,5-trisphosphate. DAG, diacylglycerol. PI3K, PI 3-kinase. PI4K, PI 4-kinase. PIPK, PIP kinase. PLC, phospholipase C. * Sites inhibited by wortmannin.

As a mechanism for the PI-4,5-P2 effect, Furukawa et al. (1994) suggested that the target of PI-4,5-P2 was the membrane-associated actin cytoskeletal network, which may interact with KATP channels. On the contrary, Fan & Makielski (1997) supposed an electrostatic interaction between the anionic head groups of PI-4,5-P2 and the positively charged amino acid residues of the C-terminus of Kir6.2. At present, the latter mechanism appears more likely because anionic phospholipids such as PIP and PS could also activate KATP channels, and polyvalent cations such as Mg2+ (Kozlowski & Ashford, 1990), La3+, Gd3+, neomycin and gentamicin (Fan & Makielski, 1997) induced run-down of KATP channels. In the present study, we exclusively utilized Ca2+-induced run-down to study the mechanism of the MgATP-dependent channel recovery. Although there has been no direct evidence yet, the Ca2+-induced run-down might be due to the hydrolysis of PI-4,5-P2 mediated by an endogenous Ca2+-dependent PLC (Hilgemann & Ball, 1996) and/or an interruption of the electrostatic interaction between the anionic head groups of PI-4,5-P2 and the positively charged amino acid residues of the C-terminus of Kir6.2. Tucker et al. (1997) suggested that both run-down and MgATP-dependent recovery are intrinsic to the Kir6.2 subunit. Our studies on Kir6.2ΔC26 channels confirmed this conclusion. Furthermore, it was revealed that PI-4,5-P2 also reactivates KATP channel activity due to the interaction with the Kir6.2 subunit. However, Kir6.2ΔC26 channels were less sensitive to Ca2+ than Kir6.2/SUR2A channels. Although the primary sites at which Ca2+ and PI-4,5-P2 interact to modulate channel activity seemed to be located on Kir6.2, this result suggests that functional coupling between SUR2A and Kir6.2 may also be interrupted.

PI-4,5-P2 is the product of phosphorylation of PI and PI-4-P. On the other hand, it can be hydrolysed by the receptor-mediated reaction of phospholipase C (PLC) to give inositol 1,4,5-trisphosphate (IP3) and diacylglycerol (DAG), which are important second messengers in the inositol lipid signal transduction mechanism. It has been reported that PI metabolism is altered during ischaemia, reperfusion (Schwertz & Halverson, 1992) and the stimulation of α1-adrenoceptors (Fedida et al. 1993), metabotropic glutamate receptors (Huber et al. 1998) and endothelin receptors (Van Heugten et al. 1993). Since the content of PI-4,5-P2 may also change during the phosphoinositide metabolism alteration, PI-4,5-P2 may play a linkage role between the inositol lipid signal transduction and KATP channel activity.

Acknowledgments

We are grateful to Professor M. Hirata for helpful discussion. We thank Professor S. Seino for providing us with the SUR2A clone and Dr K. Moriyoshi for the GFP clone. We also thank Mr M. Fukao and Ms K. Tsuji for technical support and Dr A. F. James for critical reading of the manuscript. Secretarial service by Ms K. Fujita is highly appreciated. This work was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science, Sports and Culture of Japan and by a Japan Heart Foundation & IBM Research Grant.

References

  1. Fan Z, Makielski JC. Anionic phospholipids activate ATP-sensitive potassium channels. Journal of Biological Chemistry. 1997;272:5388–5395. doi: 10.1074/jbc.272.9.5388. 10.1074/jbc.272.9.5388. [DOI] [PubMed] [Google Scholar]
  2. Fedida D, Braun AP, Giles WR. α1-Adrenoceptors in myocardium: functional aspects and transmembrane signaling mechanisms. Physiological Reviews. 1993;73:469–487. doi: 10.1152/physrev.1993.73.2.469. [DOI] [PubMed] [Google Scholar]
  3. Findlay I. Calcium-dependent inactivation of the ATP-sensitive K+ channel of rat ventricular myocytes. Biochimica et Biophysica Acta. 1988;943:297–304. doi: 10.1016/0005-2736(88)90561-5. [DOI] [PubMed] [Google Scholar]
  4. Findlay I, Dunne MJ. ATP maintains ATP-inhibited K+ channels in an operational state. Pflügers Archiv. 1986;407:238–240. doi: 10.1007/BF00580683. [DOI] [PubMed] [Google Scholar]
  5. French JF, Riera LC, Sarmiento JG. Identification of high and low (GTP-sensitive) affinity [3H]glibenclamide binding sites in cardiac ventricular membranes. Biochemical and Biophysical Research Communications. 1990;167:1400–1405. doi: 10.1016/0006-291x(90)90678-g. [DOI] [PubMed] [Google Scholar]
  6. Furukawa T, Virág L, Furukawa N, Sawanobori T, Hiraoka M. Mechanism for reactivation of the ATP-sensitive K+ channel by MgATP complexes in guinea-pig ventricular myocytes. The Journal of Physiology. 1994;479:95–107. doi: 10.1113/jphysiol.1994.sp020280. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Furukawa T, Yamane T, Terai T, Katayama Y, Hiraoka M. Functional linkage of the cardiac ATP-sensitive K+ channel to the actin cytoskeleton. Pflügers Archiv. 1996;431:504–512. doi: 10.1007/BF02191896. [DOI] [PubMed] [Google Scholar]
  8. Gribble FM, Tucker SJ, Ashcroft FM. The essential role of the Walker A motifs of SUR1 in K-ATP channel activation by Mg-ADP and diazoxide. The EMBO Journal. 1997;16:1145–1152. doi: 10.1093/emboj/16.6.1145. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Hamill OP, Marty A, Neher E, Sakmann B, Sigworth FJ. Improved patch-clamp techniques for high-resolution current recording from cells and cell-free membrane patches. Pflügers Archiv. 1981;391:85–100. doi: 10.1007/BF00656997. [DOI] [PubMed] [Google Scholar]
  10. Hilgemann DW, Ball R. Regulation of cardiac Na+,Ca2+ exchange and KATP potassium channels by PIP2. Science. 1996;273:956–959. doi: 10.1126/science.273.5277.956. [DOI] [PubMed] [Google Scholar]
  11. Huang CL, Feng S, Hilgemann DW. Direct activation of inward rectifier potassium channels by PIP2 and its stabilization by Gβγ. Nature. 1998;391:803–806. doi: 10.1038/35882. [DOI] [PubMed] [Google Scholar]
  12. Huber KM, Sawtell NB, Bear MF. Effects of the metabotropic glutamate receptor antagonist MCPG on phosphoinositide turnover and synaptic plasticity in visual cortex. Journal of Neuroscience. 1998;18:1–9. doi: 10.1523/JNEUROSCI.18-01-00001.1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Hussain M, Wareham AC. Rundown and reactivation of ATP-sensitive potassium channels (KATP) in mouse skeletal muscle. Journal of Membrane Biology. 1994;141:257–265. doi: 10.1007/BF00235135. [DOI] [PubMed] [Google Scholar]
  14. Inagaki N, Gonoi T, Clement JP, IV, Namba N, Inazawa J, Gonzalez G, Aguilar-Bryan L, Seino S, Bryan J. Reconstitution of IKATP: an inward rectifier subunit plus the sulfonylurea receptor. Science. 1995;270:1166–1170. doi: 10.1126/science.270.5239.1166. [DOI] [PubMed] [Google Scholar]
  15. Kozlowski RZ, Ashford MLJ. ATP-sensitive K+-channel run-down is Mg2+ dependent. Proceedings of the Royal Society B. 1990;240:397–410. doi: 10.1098/rspb.1990.0044. [DOI] [PubMed] [Google Scholar]
  16. Miller JA, Velayo NL, Dage RC, Rampe D. High affinity [3H]glibenclamide binding sites in rat neuronal and cardiac tissue: localization and developmental characteristics. Journal of Pharmacology and Experimental Therapeutics. 1991;256:358–364. [PubMed] [Google Scholar]
  17. Moriyoshi K, Richards LJ, Akazawa C, O'leary DDM, Nakanishi S. Labeling neural cells using adenoviral gene transfer of membrane-tagged GFP. Neuron. 1996;16:255–260. doi: 10.1016/s0896-6273(00)80044-6. [DOI] [PubMed] [Google Scholar]
  18. Nakanishi S, Yano H, Matsuda Y. Novel functions of phosphatidylinositol 3-kinase in terminally differentiated cells. Cellular Signalling. 1995;7:545–557. doi: 10.1016/0898-6568(95)00033-l. [DOI] [PubMed] [Google Scholar]
  19. Ogata N, Tatebayashi H. A simple and multi-purpose ‘concentration clamp’ method for rapid superfusion. Journal of Neuroscience Methods. 1991;39:175–183. doi: 10.1016/0165-0270(91)90083-c. [DOI] [PubMed] [Google Scholar]
  20. Ohno-Shosaku T, Zünkler BJ, Trube G. Dual effects of ATP on K+ currents of mouse pancreatic β-cells. Pflügers Archiv. 1987;408:133–138. doi: 10.1007/BF00581342. [DOI] [PubMed] [Google Scholar]
  21. Okuyama Y, Yamada M, Kondo C, Satoh E, Isomoto S, Shindo T, Horio Y, Kitakaze M, Hori M, Kurachi Y. The effects of nucleotides and potassium channel openers on the SUR/Kir6.2 complex K+ channel expressed in a mammalian cell line, HEK293T cells. Pflügers Archiv. 1988;435:595–603. doi: 10.1007/s004240050559. [DOI] [PubMed] [Google Scholar]
  22. Sakura H, Ämmälä C, Smith PA, Gribble FM, Ashcroft FM. Cloning and functional expression of the cDNA encoding a novel ATP-sensitive potassium channel subunit expressed in pancreatic beta-cells, brain, heart and skeletal muscle. FEBS Letters. 1995;377:338–344. doi: 10.1016/0014-5793(95)01369-5. [DOI] [PubMed] [Google Scholar]
  23. Schwertz DW, Halverson J. Changes in phosphoinositide-specific phospholipase C and phospholipase A2 activity in ischaemic and reperfused rat heart. Basic Research in Cardiology. 1992;87:113–127. doi: 10.1007/BF00801959. [DOI] [PubMed] [Google Scholar]
  24. Takano M, Ishii T, Xie L-H. Cloning and functional expression of rat brain KIR6.2 channel. Japanese The Journal of Physiology. 1996;46:491–495. doi: 10.2170/jjphysiol.46.491. [DOI] [PubMed] [Google Scholar]
  25. Takano M, Qin D, Noma A. ATP-dependent decay and recovery of K+ channels in guinea pig cardiac myocytes. American Journal of Physiology. 1990;258:H45–50. doi: 10.1152/ajpheart.1990.258.1.H45. [DOI] [PubMed] [Google Scholar]
  26. Takano M, Xie L-H, Otani H, Horie M. Cytoplasmic terminus domains of Kir6.x confer different nucleotide-dependent gating on the ATP-sensitive K+ channel. The Journal of Physiology. 1998;512:395–406. doi: 10.1111/j.1469-7793.1998.395be.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Tucker SJ, Gribble FM, Zhao C, Trapp S, Ashcroft FM. Truncation of Kir6.2 produces ATP-sensitive K+ channels in the absence of the sulphonylurea receptor. Nature. 1997;387:179–183. doi: 10.1038/387179a0. [DOI] [PubMed] [Google Scholar]
  28. Van Heugten HAA, Bezstarosti K, Dekkers DHW, Lamers JMJ. Homologous desensitization of the endothelin-1 receptor-mediated phosphoinositide response in cultured neonatal rat cardiomyocytes. Journal of Molecular and Cellular Cardiology. 1993;25:41–52. doi: 10.1006/jmcc.1993.1006. [DOI] [PubMed] [Google Scholar]
  29. Venkatesh N, Lamp ST, Weiss JN. Sulfonylureas, ATP-sensitive K+ channels, and cellular K+ loss during hypoxia, ischemia, and metabolic inhibition in mammalian ventricle. Circulation Research. 1991;69:623–637. doi: 10.1161/01.res.69.3.623. [DOI] [PubMed] [Google Scholar]

Articles from The Journal of Physiology are provided here courtesy of The Physiological Society

RESOURCES