This study was designed to identify the single-channel properties and molecular entity of ATP-sensitive K+ (KATP) channels in guinea pig gastric myocytes with patch-clamp recording and RT-PCR. Pinacidil and diazoxide activated KATP currents in a glibenclamide-sensitive manner. The open probability of channels was enhanced by the application of 10 μM pinacidil from 0.085 ± 0.04 to 0.20 ± 0.05 (n = 7) and was completely blocked by 10 μM glibenclamide. Single-channel conductance was 37.3 ± 2.5 pS (n = 4) between −80 and −20 mV in symmetrical K+ gradient conditions. In inside-out mode, KATP channels showed no spontaneous openings and were activated by the application of nucleotide diphosphates to the cytoplasmic side. These single-channel properties are similar to those of the nucleotide diphosphate-dependent K+ channels in vascular smooth muscle, which are composed of Kir6.1 and sulfonylurea receptor (SUR)2B. RT-PCR demonstrated the presence of Kir6.1, Kir6.2, and SUR2B in guinea pig stomach smooth muscle cells. These results suggest that KATP channels in smooth muscle cells of the guinea pig stomach are composed of Kir6.1 and SUR2B.
- adenosine 5′-triphosphate-sensitive potassium channels
- Kir6.1-sulfonurea receptor 2B
- smooth muscle cells
- guinea pig stomach
atp-sensitiveK+ (KATP) channels are inhibited by intracellular ATP, and so these channels play a role in linking cell metabolic state to membrane potential. KATP channels are associated with diverse cellular functions, such as shortening of action potential duration in cardiac myocytes (27), insulin release in pancreatic β-cells (4), regulation of excitability in skeletal muscle (31), and regulation of vascular smooth muscle contractility (32). KATP channels are heteromultimers composed of inwardly rectifying K+ channel subunits (Kir6.x) and sulfonylurea receptors (SURs) that associate in a 4:4 stoichiometry to form an octameric KATP channel. Various combinations of these two subunits convey the heterogeneity in channel properties observed in native cells such as Kir6.2-SUR1 in pancreatic β-cells, Kir6.2-SUR2A in cardiac and skeletal muscles, and Kir6.1-SUR2B or Kir6.2-SUR2B in smooth muscle (30).
KATP channels were identified in various smooth muscles such as rabbit portal vein (15), mouse colon (18), porcine coronary artery (24), rat mesenteric artery (39), porcine proximal urethra (34), human corporal smooth muscle (20), and cat trachea (36). The KATP channels in smooth muscle show different properties from other tissues, such as small conductance, different responses to K+ channel openers (KCOs) and sulfonylurea, and composition of subunits. Because the KATP channel is sensitive to intracellular ATP, it shows high activity in ATP-free conditions such as inside-out mode. In vascular smooth muscles, however, KATP channels show no activity when they are exposed to ATP-free conditions. These channels are activated by the application of nucleotide diphosphates (NDPs) intracellularly (3, 15, 16, 40). Because endogenous intracellular NDPs are the crucial regulators of these K+channels in smooth muscle, Beech et al. (3) designated these channels as nucleotide diphosphate-dependent K+(KNDP) channels. Yamada et al. (38) reported that the coexpressed SUR2B-Kir6.1 channel with unitary conductance of ∼33 pS did not spontaneously open in the absence of intracellular ATP in an inside-out mode and that NDPs such as UDP and GDP stimulated SUR2B-Kir6.1 channel activity. Thus they suggested that the SUR2B-Kir6.1 channel is not a classic KATP channel but rather resembles the KNDP channel in vascular smooth muscle (38). However, in other smooth muscles such as murine colon (18) and guinea pig urinary bladder (9), the KATP channel appears to be composed of Kir6.2 and SUR2B.
We observed the KATP channel current in guinea pig gastric myocytes, and this current was activated by KCOs and blocked by glibenclamide (13, 14). From the above results, in the present study, we tried to identify and characterize KATPchannels using single-channel recording and RT-PCR in guinea pig gastric myocytes. A preliminary report of the present study appeared in an abstract form (17).
MATERIALS AND METHODS
Guinea pigs of either sex weighing 300–350 g were exsanguinated after stunning. All experiments were performed in accordance with the guidelines for the animal care and use approved by the Seoul National University. The whole stomach was excised and placed in a bath containing oxygenated phosphate-buffered Tyrode solution at room temperature. The antral part of the stomach was cut, and the mucosal layer was separated from muscle layer. The circular muscle layer was dissected from the longitudinal muscle layer and cut into small segments in nominal Ca2+-free physiological salt solution (PSS) containing (in mM) 135 NaCl, 5 KCl, 1 MgCl2, 10 glucose, and 10 HEPES adjusted to pH 7.4 with NaOH (Ca2+-free PSS). These segments were incubated in a medium modified from Kraft-Brühe (KB) medium (12) for 30 min at 4°C. They were then incubated for 15–25 min at 37°C in digestion medium (Ca2+-free PSS) containing 0.1% collagenase (Wako), 0.05% dithioerythritol, 0.1% trypsin inhibitor, and 0.2% bovine serum albumin. After digestion, the supernatant was discarded, the softened muscle segments were transferred again into modified KB medium, and single cells were dispersed by gentle agitation with a wide-bore glass pipette. Isolated gastric myocytes were kept in modified KB medium at 4°C until use. All experiments were carried out within 12 h of harvesting cells and were performed at room temperature.
Voltage-clamp experiments using whole cell and single-channel recordings were performed on gastric smooth muscle cells. For the whole cell experiments, the external solution contained (in mM) 50 (or 80) NaCl, 90 (or 60) KCl, 10 HEPES, 10 glucose, 1 MgCl2, and 0.2 CaCl2, and pH was adjusted to 7.4 with NaOH. For the whole cell recordings, the pipette solution contained (in mM) 9 NaCl, 109 KCl, 10 HEPES, 10 EGTA, 1 GTP, 0.1 ATP, 1 MgCl2, and 1 CaCl2, and pH was adjusted to 7.2 with KOH (38 mM). Single-channel activities were recorded with the pipettes containing (in mM) 140 KCl, 10 HEPES, and 10 EGTA (pH 7.4 with KOH). After gigaseal formation, the extracellular solution was exchanged with a solution containing (in mM) 140 KCl, 10 HEPES, 1 MgCl2, 1 EGTA, and 10 glucose (pH 7.4 with KOH). The concentrations of free Ca2+ in the bath solutions of cell-attached and inside-out modes were adjusted to 10−6 and 10−9 M by adding appropriate amounts of CaCl2 calculated according to Fabiato (7). To minimize the activity of voltage-dependent K+ channels and Ca2+-activated K+channels, all currents were recorded at a holding potential of −60 mV and 100 nM charybdotoxin or 0.5 mM TEA was added to the pipette solution. Single-channel and whole cell currents were amplified through a patch-clamp amplifier (Axopatch-1D; Axon instruments) and were stored on a digital audio tape by a digital tape recorder (DTR-1204; Biologic). The single-channel data were filtered at 1 kHz with an eight-pole Bessel filter and were digitized at 5 kHz for analysis. Single-channel amplitudes and the open state probability ofN channels (NP o) were calculated from the all-point histograms. NP o values of the data recorded at a steady membrane potential were determined fromNP o = (Σtjj)/T, wheretj is the time spent with j = 1, 2,… N channels open and T is the duration of the measurement. Data are expressed as means ± SE.
Single cell collection.
Gastric smooth muscle cells, isolated by the procedure described inCell dissociation, were collected individually for multicell RT-PCR. Micropipettes were constructed from borosilicate glass (Clark Instruments) with 40- to 50-μm diameter tips. Cells were transferred to the stage of a phase-contrast microscope and allowed to stick lightly to the glass coverslip bottom of a small chamber for 10 min. The cells were then perfused with sterile normal Tyrode saline to remove cellular debris. Single smooth muscle cells were identified and collected by positioning the tip of the micropipette near the cell and applying light suction. Approximately 100–150 smooth muscle cells were collected from each dispersion. After collection, the cells were expelled from the pipette into an RNase-free microcentrifuge tube.
Distribution of Kir6x and SURs.
We used gastric antral tissue or isolated smooth muscle cells obtained as described in Single cell collection. Total RNA was prepared with SNAP total RNA isolation kits (Invitrogen, Carlsbad, CA), following the procedures of the manufacturer. Because ∼200 smooth muscle cells were used in each RNA isolation, 20 μg of polyinosinic acid (a carrier of RNA) were added to the lysates. First-strand cDNA was synthesized from the RNA preparations with a Superscript II RNase transcriptase kit (GIBCO-BRL, Gaithersburg, MD); RNA (1 pg) was reverse-transcribed by using random hexamers (50 μg/μl). To perform nested PCR, the following sets of primers were used: Kir6.1 forward nucleotides 137–154, 155–172 and reverse nucleotides 470–487, 506–521 (GenBank accession no. AF183918); Kir6.2 forward nucleotides 531–548, 549–566 and reverse nucleotides 834–851, 870–887 (AF183919); SUR1 forward nucleotides 1169–1186, 1205–1222 and reverse nucleotides 1508–1525, 1532–1549 (AF183921); SUR2A forward nucleotides 918–935, 954–971 and reverse nucleotides 1257–1274, 1281–1298 (AF183922); and SUR2B forward nucleotides 577–594, 613–620 and reverse nucleotides 916–933, 940–957 (AF183923) (29). PCR primers for Kir6.1 forward, nucleotides 782–799, 803–822, and reverse, 1475–1498, 1502–1521 (AF196330) were used to detect genomic DNA contamination, whereby the primers were designed to span an intron in addition to an exon. Complementary DNA (20% of the first-strand reaction) was combined with first sense and antisense primers (20 μM), 1 mM deoxynucleotide triphosphates, 60 mM Tris · HCl (pH 8.5), 15 mM (NH4)2SO4, 1.5 mM MgCl2, 2.5 units of Taq (Bioneer), and RNase-free water to a final volume of 50 μl. The reaction occurred in a Perkin-Elmer thermal cycler under the following conditions: an initial denaturation at 94°C for 4 min, followed by 40 cycles at 94°C for 30 s, 52°C for 30 s, 72°C for 1 min, with a final extension step at 72°C for 7 min. Five microliters of the first-round PCR product were then added to a new reaction mixture containing all of the components listed above except for second sense and antisense primers (20 μM), and 40 additional cycles of PCR were then performed. PCR products were separated by 2% agarose gel electrophoresis.
To record KATP channels in gastric antral myocytes, we first recorded KATP currents activated by KCOs such as pinacidil and diazoxide in conditions similar to those used in previous studies (13, 14). External K+ concentration was increased from 5 to 60 or 90 mM to increase the driving force for K+ at a holding potential of −60 mV. The intracellular ATP was lowered to 0.1 mM for steady-state activation of KATPchannels. Figure 1 shows whole cell current with a pipette solution containing 140 mM K+. After the whole cell configuration was established, inward steady-state currents were induced by application of high-K+ solution. In the experiment illustrated in Fig. 1 A, the first exposure of the cell to 10 μM pinacidil increased the current in the presence of external 90 and 60 mM K+, respectively. The application of 10 μM glibenclamide inhibited the pinacidil-activated currents. It should be noted that the glibenclamide-sensitive current was larger than pinacidil-activated current. This effect was reported in other KATP currents in smooth muscle and shows the possibility of contribution of KATP channel to resting membrane potential (18, 41). Figure 1 B shows the current-voltage (I-V) relationships of the glibenclamide-sensitive currents calculated from ramp pulses applied to whole cell current in Fig. 1 A. The reversal potential of the glibenclamide-sensitive currents with 90 and 60 mM K+ was −5.0 ± 0.3 (n = 6) and −14.8 ± 2.3 (n = 6) mV, respectively, and the estimated K+ equilibrium potential (E K) was −11 and −21 mV, respectively. The I-Vrelationship was linear between −100 and 40 mV and reversed nearE K. We also tested diazoxide on whole cell current in gastric smooth muscle cells. The application of 300 μM diazoxide activated KATP current in a glibenclamide-sensitive manner (Fig. 1 C).
Single-channel recordings of KATP channels in cell-attached mode.
To investigate the properties of the channels activated by KCOs, a single-channel experiment was performed in a symmetrical 140 mM K+ gradient using a cell-attached configuration. To minimize the activity of large-conductance Ca2+-activated K+ channels, most of the single-channel experiments were performed at a negative holding potential of −60 mV with 0.5 mM TEA or 100 nM charybdotoxin in the pipette solution. In cell-attached mode, spontaneous single-channel activities at −60 mV were observed in 29 of 88 pinacidil-activated patches (data not shown). The application of 10 μM pinacidil to the bath solution enhanced single-channel activities in a glibenclamide-sensitive manner (Fig.2 A). During the openings of the channel, it showed the fast transition between open and closed, and this bursting activity was described in other KATP channels (4, 27, 31). We observed single-channel activities withNP o of 0.085 ± 0.04 (n = 7) in control, and 10 μM pinacidil increasedNP o of these channels to 0.20 ± 0.05 (n = 7). Figure 2 B shows the all-point histogram of the single-channel openings obtained from channel activity in Fig. 2 A fitted with Gaussian curves to give a unitary current amplitude of −2.23 pA at a membrane potential of −60 mV. Figure 2 C shows the change of NP o by application of pinacidil and glibenclamide in Fig. 2 A. In the presence of 10 μM pinacidil, the I-V relationships for KATP channels were obtained by changing the holding membrane potentials from −80 mV to −20 mV by 20-mV increments in the cell-attached mode (Fig. 2D). The channel exhibited linearI-V relationships at negative holding potentials, and its conductance was 37.3 ± 2.5 pS (n = 4; Fig.2 E).
Disappearance and reactivation of channel activity in inside-out mode.
To suppress the opening of Ca2+-activated K+channels, we recorded the channel activity in low-Ca2+(pCa = 9) bath solution at negative membrane potential. We added 10 μM pinacidil to all bath solutions to enhance channel opening in the inside-out mode experiment. When a cell-attached patch showed openings of the KATP channel with 10 μM pinacidil present in the bath, patch excision resulted in an immediate disappearance of channel openings (Fig.3 A). Application of 1 mM ADP to the inner surface of the membrane patch restored the channel activity. NP o increased to 0.35 ± 0.06 (n = 4) in the presence of 1 mM ADP, and this enhanced channel activity was suppressed by 1 mM ATP in a reversible manner (Fig. 3 A). When Mg2+ was removed from the bath solution, the 1 mM ADP-induced openings of the KATP channel disappeared (Fig. 3 B). Figure4 showed the effect of other NDPs on the activity of the KATP channel. Application of 1 mM UDP or 1 mM GDP reactivated channel activity in the presence of pinacidil (Fig.4 A). Figure 4 B shows mean values of the enhancedNP o by 1 mM ADP, 1 mM GDP, and 1 mM UDP to 0.35 ± 0.06 (n = 4), 0.33 ± 0.06 (n = 4), 0.20 ± 0.06 (n = 3), respectively.
Molecular expression of KATP channels in guinea pig gastric myocytes.
To determine the presence of Kir subunits and SURs, RT-PCR was performed using Kir6.1, Kir6.2, SUR1, SUR2A, and SUR2B gene-specific primers in guinea pig gastric antrum and isolated circular myocytes. In tissue, RT-PCR detected transcripts for Kir6.1, Kir6.2, SUR1, and SUR2B, but the specific primers for amplification of SUR2A did not produce a cDNA fragment of 321 bp (Fig.5 A). This suggests that SUR2A is not present in guinea pig gastric antrum. Figure 5 B shows the presence of transcripts for Kir6.1, Kir6.2, and SUR2B but not SUR1 in smooth muscle cells. It is noteworthy that the cDNA fragment of SUR1 is present in tissue but absent in smooth muscle cells. The cDNA fragment of SUR1 seems to originate from neuronal cells of the guinea pig stomach (21). We also examined the contamination of genomic DNA by using specific primers for Kir6.1, which were generated to span an intron in addition to an exon. These primers showed no specific cDNA fragments (data not shown).
In this study, we investigated the properties and molecular composition of KATP channels in guinea pig gastric myocytes. KATP channels with conductance of 37 pS were activated by the application of pinacidil or diazoxide. Glibenclamide blocked the channel activity enhanced by pinacidil in cell-attached mode. However, the channel activity disappeared when the patch was excised and restored by the intracellular application of NDPs, an effect that was blocked by 1 mM ATP. The ADP-induced channel reactivation was Mg2+ dependent. The molecular composition assessed by RT-PCR analysis using subunit-specific primers revealed the presence of mRNA for inwardly rectifying K+ channels (Kir6.1 and 6.2) and SURs (SUR2B) in smooth muscle cells. The subunit profile, together with the electrophysiological properties, suggests that the KATP channel in smooth muscle cells of the guinea pig stomach is composed of Kir6.1 and SUR2B.
The properties of KATP channels in guinea pig gastric myocytes were similar to those of KATP channels in other smooth muscle. The single-KATP channel conductance of guinea pig gastric myocytes was recorded as 37 pS in cell-attached mode (Fig. 2). This is similar to other reported conductance values: 30 pS in porcine coronary artery (24), 43 pS in porcine proximal urethra (34), 40 pS in cat trachea (36), and 27 pS in murine colon (18). Zhang and Bolton (40) also reported that two types of KATPchannels with conductances of 50 and 22 pS were identified in rat portal vein. Although KATP channels show high activity when the patch is exposed to ATP-free solution in cardiac and pancreatic β-cells, the channel activities of KATP channels in smooth muscles such as rabbit portal vein (3), pig urethra (35),and murine colon (18) show rapid rundown or no spontaneous opening in inside-out mode. As shown in Fig.3 A, the channel activity was not observed when patches were excised in guinea pig gastric myocytes. We observed that the application of ADP, GDP, or UDP to the cytoplasmic side reactivated KATP channels in guinea pig gastric myocytes (Figs. 3 and4). Because the openings of KATP channels in vascular smooth muscle cells were regulated by intracellular NDPs but not by ATP, Beech et al. (3) suggested that endogenous intracellular NDPs are the crucial regulators of these K+channels in smooth muscles and they designated these channels as KNDP. The KNDP channel also has the characteristics of low conductance (∼30–40 pS) and no spontaneous opening in inside-out mode (3). For these reasons, the KATP channels in guinea pig gastric myocytes might be classified as KNDP.
Multiple channel types of KATP channels with defined physiological and pharmacological properties can be constituted by the heteromeric assembly of various Kir and SUR subunits. Yamada et al. (38) showed that the coexpression of Kir6.1 and SUR2B produces K+ channel currents with unitary conductance of ∼33 pS in the presence of pinacidil. In the inside-out mode, these channels did not open spontaneously in the absence of intracellular ATP. NDPs such as UDP and GDP stimulated SUR2B-Kir6.1 channel activity (38). The expressed SUR2B-Kir6.1 channel closely resembles the K+ channels in the vascular smooth muscle cells designated as KNDP channels, which are sensitive to sulfonylurea and KCO drugs but not to ATP (3, 15, 16, 40). Yamada et al. (38) suggested that the finding that KATP and KNDP channels show similar pharmacology but different ATP sensitivity may have occurred because KNDP and KATP channels share SUR but differ in Kir subtype. It was also reported that specific structural elements of Kir6.2 are involved in spontaneous opening in the absence of intracellular ATP. The cytosolic NH2-terminal domain of Kir6.2 was mandatory for spontaneous opening, although part of the COOH terminus was also involved (19). On the basis of the above results, the similar properties with KNDP channels imply that KATP channels in guinea pig gastric myocytes might be composed of Kir6.1 and SUR2B.
In Fig. 5, we observed transcripts for Kir6.1 and Kir6.2 in mRNA isolated from smooth muscle cells of the guinea pig stomach. It was reported that Kir6.1 is a subunit of the mitochondria KATPchannel (33). Kir6.2 has been thought to compose KATP channels with SUR2B in smooth muscles such as murine colon and guinea pig urinary bladder (9, 18). The conductances of Kir6.1 and Kir6.2 with SUR2A were 34.2 and 79.5 pS in symmetrical conditions, respectively (19). We could not find the channel with conductance of ∼70 pS. Although it is not clear whether Kir6.1 and Kir6.2 heteromultimerize with each other to compose KATP channels, it was reported that a dimer Kir6.1-Kir6.2 construct expressed with SUR2B had intermediate conductance between that of either Kir6.2 or Kir6.1 expressed with SUR2B and that this fact may contribute to the diversity of nucleotide-regulated K+channel currents seen in native tissues (2, 5). In heart, both Kir6.1 and Kir 6.2 mRNAs are abundantly expressed (11). However, the expressed KATP channels in heart are composed of Kir6.2 and SUR2A (28). It was reported that the gene expression of Kir6.1 and Kir6.2 is regulated differently in cardiac tissue (1, 22). Thus it is likely that the expression of inward-rectifier channels of the KATP channels in gastric myocytes would be regulated by an unknown mechanism different from that in cardiac myocytes.
In gastrointestinal smooth muscle, the physiological role of KATP channels is unclear. The changes in electrical and mechanical activities by metabolic inhibition would not involve KATP channels in guinea pig stomach smooth muscle (10, 25). However, the suppression of metabolism can activate KATP channels in vascular smooth muscle (3,6, 39). Franck et al. (8) reported that KATP channels might not be involved in nonadrenergic, noncholinergic neurostimulation such as ATP, nitric oxide (NO), and neurotensin. We observed spontaneous activity of KATPchannels and showed that the glibenclamide-sensitive current was larger than the pinacidil-activated current (Fig. 1 A). Koh et al. (18) offered the physiological interpretation that basally active KATP channels may contribute to the resting conductance of murine colonic myocytes because basal channel activity was observed at resting test potential range. It was reported that KATP channels show high activity and contribute to the control of the resting membrane potential in vascular smooth muscle cells (24, 37). The effects of several endogenous neurotransmitters, such as calcitonin gene-related peptide (CGRP; Ref. 26), angiotensin II (24), vasopressin (37), and NO (23) are mediated by KATP channels in vascular smooth muscle. Neuropeptide CGRP hyperpolarized the smooth muscle via activation of KATPchannels in guinea pig gallbladder (41). In guinea pig gastric myocytes, KATP channels were modulated by substance P and ACh (13, 14). KATP channels may play a role in regulation in membrane potential and/or in change of contractility mediated by neurotransmitters in guinea pig stomach.
From the above results, we identified single KATP channels in guinea pig gastric smooth muscle and demonstrated modulation by intracellular NDPs. We suggested that molecular entity of KATP channels in guinea pig gastric smooth muscle seems to be composed of Kir6.1-SUR2B.
We thank Dr. R. Preisig-Müller for providing the nucleotide sequences of gpKir6.1 and -6.2 and gpSUR1, -2A, and -2B genes before publication.
This study was supported by a grant from the Seoul National University College of Medicine Research Fund, by a grant of the Korea Health 21 R&D Project, Ministry of Health and Welfare (01-PJ1-PG3–21400–0024), and in part by year 2001 BK21 project for Medicine, Dentistry, and Pharmacy.
Address for reprint requests and other correspondence: K. W. Kim, Dept. of Physiology and Biophysics, Seoul National Univ. College of Medicine, 28 Yongon-Dong, Chongno-Gu, Seoul 110-799, Korea (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2002 the American Physiological Society