Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

K⁺ channels play an important role in modulating smooth muscle tone, such as in tracheal and bronchial smooth muscles, in airways (1). Previous studies using airway smooth muscles from various species have demonstrated voltage-dependent delayed rectifier K⁺ currents (K_V) (4), which contribute to the formation of membrane potential (3, 7). Therefore, the inhibition of K_V by 4-aminopyridine (4-AP), a selective antagonist, depolarizes the membrane (3, 7), resulting in increasing intracellular Ca²⁺ concentration ([Ca²⁺]_i) through the opening of the voltage-dependent Ca²⁺ channels. In addition, charybdotoxin-sensitive K⁺ channel, which is activated by an increase in [Ca²⁺]_i or -adrenergic stimulation, may also play a role in modulating smooth muscle tone in airway smooth muscle (5, 8). On the other hand, the inward rectifier K⁺ (Kir) channel, which is selectively inhibited by Ba²⁺, has been known to maintain membrane potential in cardiac myocytes and skeletal muscles (9), but several papers have reported the existence of Kir currents in vascular smooth muscle cells, such as small diameter coronary, cerebral, and mesenteric arteries or arterioles (10). The activation of Kir channels underlies the dilatation of these small arteries, which may be involved in autoregulation of local blood flow when the extracellular K⁺ concentration ([K⁺]_o) is raised under physiologic and pathologic conditions such as neurogenic stimulation or ischemia (18). In airways, it has been noticed that Ba²⁺ (1 mM) elicits large depolarization in canine small bronchi (21), proposing the existence of Kir channels, which has been recently reported in human small bronchioles (22). However, in tracheal smooth muscles or large diameter bronchus, the Kir channels have not been observed. These observations suggest that Kir channels may exist in the small diameter airways, possibly resulting in the regulation of local airway flow under physiologic and pathologic conditions.

Recent molecular studies have identified a superfamily of K⁺ channel genes and characterized specific gene subfamilies. Human airway myocytes have been reported to express mRNA K_V1.5, which may play an essential role in regulating airway tone (6). The Kir channels identified in the arterial smooth muscle cells (10) possess the characteristics of the Kir2 subfamily, where at least four distinct isoforms of the Kir2 channel family, Kir2.1 (23), Kir2.2 (24), Kir2.3 (25), and Kir2.4 (26), have been identified. In rat arterial smooth muscle cells (cerebral, coronary, and mesenteric arteries), RT-PCR of mRNA reveals the transcripts for Kir2.1, whereas transcripts for Kir2.2 and Kir2.3 are not found (13). In addition, recent targeted disruption of Kir2.1 gene, but not Kir2.2, reveals the essential role of Kir current in K⁺-mediated vasodilation of mice (27). These observations suggest that Kir2.1 gene encodes the inward rectifier K⁺ channel in arterial or arteriole vascular smooth muscle. On the other hand, in adult cardiac myocytes, Kir2.2 forms the native inward rectifier K⁺ channel (IK₁) (9), but the involvement of Kir2.1 gene has been reported in cardiac myocytes during pre- and postnatal development (28, 29). However, the molecular aspects of Kir channels in airway smooth muscle cells remain unknown.

The purpose of the present study was to clarify the properties of the inward rectifier K⁺ channel and then investigate which member of the Kir2 channel gene family forms the Kir channel in native human bronchial smooth muscle. Experiments using antisense oligonucleotides directed against Kir2.1 mRNA were done. Here, we provide direct evidence showing that the inward rectifier K⁺ current (Kir) is present in human bronchial smooth muscle cells (hBSMCs), and the Kir2.1 gene encodes the Kir channel protein in these cells.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Cell Preparation

Culture cells of human bronchial smooth muscles isolated from normal hBSMCs were purchased from the Clonetics Corporation (San Diego, CA) as reported previously (30). The cells were cultured in 25 cm² flasks, in culture medium for hBSMCs supplemented with 5% fetal calf serum, human epidermal growth factor (0.5 µg/ml), insulin (5 mg/ml), human fibroblast growth factor (1 µg/ml), gentamicin (50 µg/ml), and amphotericin B (0.05 µg/ ml) (SmGM-2 Buffer-Kit, Clonetics) in an atmosphere of 5% CO₂ and 95% air at 35°C. When the cells became confluent, they were subcultured in the same medium. At confluence, cells obtained from 25 cm² flasks were passaged using 0.25% trypsin in 0.02% EDTA. Medium was replaced twice weekly. Cells before confluence at passages 3-7 were detached from culture flasks with 0.25% trypsin in 0.02% EDTA, and used for later experiments. The cells were identified as smooth muscle cells, by which the expression of -actin was confirmed by immunostaining with biotin-conjugated antibody as shown previously (30). All experiments were performed at 35-37°C.

Solutions and Drugs

The composition of the control extracellular Tyrode solution was as follows (in mM): NaCl 136.5, KCl 5.4, CaCl₂ 1.8, MgCl₂ 0.53, glucose 5.5, and HEPES-NaOH buffer 5.5 (pH 7.4). When the external K⁺ concentration ([K⁺]_o) was changed, NaCl in the Tyrode solution was replaced with equimolar KCl. The patch pipette contained (in mM): KCl 140, EGTA 10, MgCl₂ 2, Na₂ATP 3, guanosine-5'-triphosphate (GTP, sodium salt; Sigma, St. Louis, MO) 0.1, and HEPES-KOH buffer 5 (pH 7.2). In experiments in which K⁺ currents were blocked, the patch pipette contained Cs⁺-internal solution (mM): CsCl 140, EGTA 10, MgCl₂ 2, Na₂ATP 3, GTP 0.1, and HEPES-CsOH 5 (pH 7.2). Nicardipine, 4-aminopyridine (4-AP), tetraethylammonium (TEA), charybdotoxin, and glibenclamide were purchased from Sigma. Nicardipine was dissolved into ethanol, and 10 mM stock solution was used.

Recording Technique and Data Analysis

Membrane currents and potentials were recorded with glass pipettes (outer diameter 1.5 mm; Kimble Glass Inc., Vineland, NJ) in the whole-cell clamp conditions (31, 32), using a patch-clamp amplifier (EPC-7; List Electronics, Darmstadt, Germany). Heat-polished patch pipettes, filled with the artificial internal solution (for composition, see above), had tip resistances of 3-6 M. Membrane currents and potentials were monitored with a high-gain storage oscilloscope (COS 5020-ST; Kikusui Electronics, Tokyo, Japan). At the start of each experiment, the series resistance was compensated. The data were stored on videotapes using a pulse-code-modulation (PCM) converter system (RP-880; NF Electronic Circuit Design, Tokyo, Japan). On playback, the data were reproduced, low-passed filtered at 2 kHz (3 dB) with a Bessel filter (FV-665, NF, 48 dB/ octave slope attenuation; Yokahama), sampled at 6 kHz, analyzed off-line on a computer using p-Clamp software (Axon Instruments, Foster City, CA). The capacitance of single hBSMCs and the input resistance of the cell/pipette assembly (the sum of the input resistance and the seal resistance) were measured under voltage-clamp conditions, where small hyperpolarizing pulses were applied from the membrane potential held at the resting membrane potential. 4-AP-sensitive current, which was elicited by depolarizing steps to +40 mV from a holding potential of 70 mV, was defined as the subtraction current from the current trace in control to the current trace in the presence of 10 mM 4-AP. Ba²⁺-sensitive current, which was elicited by hyperpolarizing potentials to 130 mV from a holding potential of 70 mV, were defined as the subtraction current from the current trace in control to the current trace in the presence of 5 mM Ba²⁺. To obtain the reversal potential of Kir, 4-AP (10 mM) was added to the bathing solution to block the delayed rectifier currents.

Data were expressed as the means ± S.D. Student's t test was used for statistical analysis and P < 0.05 was considered to be significant.

RNA Extraction and Reverse Transcriptase/Polymerase Chain Reaction

Total cellular RNA was extracted by using ISOGEN (Nippon Gene, Tokyo, Japan). For reverse transcriptase/polymerase chain reaction (RT-PCR), complementary DNA (cDNA) was synthesized from 1 µg of total RNA with reverse transcriptase with random primers (Toyobo, Osaka, Japan). The reaction mixture was then subjected to PCR amplification with specific forward and reverse oligonucleotide primers for 30 cycles consisting of heat denaturation (98°C), annealing (53°C), and extension (74°C). PCR products were size fractionated on 2% agarose gels, and visualized under UV light. Primers were chosen on the basis of the sequence of human Kir1.1 (Accession #U12541-5 and U65406, including all splicing variants), Kir2.1 (Accession #U12507), Kir2.2 (Accession #L36069), and Kir2.3 (Accession #U07364) (Table 1). Total RNA of human fetal brain (Toyobo) and human kidney (Sawady Technology, Tokyo, Japan) was used for a positive control.

Preparation of Antisense Constructs

Antisense oligomers (22 bp) were designed on the basis of the previously clarified sequences targeted to Kir2.1. The specificity of antisense oligomers was confirmed by comparison with all other sequences in GenBank with the use of the Basic Local Alignment Search Tool (BLAST). As a control, mismatch oligomers were prepared that were identical to each gene-specific antisense sequence with the exception of four mismatch mutations. The oligodeoxynucleotide sequences used are shown in Table 2. Phosphorothioate oligodeoxynucleotides were synthesized commercially by Amersham Pharmacia Biotech Inc. (Tokyo, Japan).

Exposure to Oligodeoxynucleotides

Oligodeoxynucleotide treatment was started 48 h after the onset of cell culture. Two groups of cultured cells were studied in all series of experiments. One group of cells was exposed to mismatch oligodeoxynucleotides, and a second group was exposed to antisense oligodeoxynucleotides. For each treatment, the growth medium was removed, and the cells were washed twice with serum-free medium and antibiotics. For experiments, antisense or mismatch oligodeoxynucleotides (0.25 µM per dish) were mixed with lipofectamine (Life Technologies Inc., Gaithersburg, MD) (30 µg per dish) and incubated at room temperature for 30 min before addition to the cultures. After 6 h of incubation at 37°C, the medium was removed and the cells were washed with fresh medium. The cells were then incubated in growth medium (including antibiotics and serum) for another 48 h before patch-clamp experiments and total RNA and protein extraction were performed.

Simultaneous Isolation of RNA and Protein Preparation

The simultaneous isolation of total RNA and protein fraction for RNase protection assay and Western blotting was performed using ISOGEN (Nippon Gene) according to manufacturer's protocol and briefly described below. hBSMCs were lysed directly in the culture dish by the addition of the reagent. The lysate was mixed with chloroform and centrifuged, which yielded the top aqueous phase, interphase, and the bottom organic phase. RNA was precipitated from the aqueous phase by addition of isopropanol, washed and dissolved in water. DNA was removed from interphase and organic phase. Proteins were precipitated from phenol-ethanol phase by addition of isopropanol. The protein pellet was washed and dissolved in 5% sodium dodecyl sulfate.

Preparation of RNA Probes and RNase Protection Assay

Kir probes were prepared from human fetal brain RNA by RT-PCR using the previously described RT-PCR primers. The length of the expected Kir2.1 fragment was 199 bp. A glyceraldehyde-3-phosphate dehydrogenase (GAPDH) probe was also prepared by RT-PCR using the following primers: upstream, CAATGAC CCCTTCATTGACC; downstream, TGGAAGATGGTGATGG GATT. This corresponded to an expected fragment of 135 bp. A quantity of 10-25 ng of PCR products were then ligated using the Lign'scribe kit (Ambion, Austin, TX) with T7 promoter and T4 DNA ligase. A second PCR was prepared with the T7 promoter-specific primer and the probe-specific sense primer. Antisense RNA probes were obtained with the DIG RNA Labeling Kit (Roche Diagnostics, Indianapolis, IN).

RNase protection assays were done with each 20 µg of total RNA isolated from hBSMCs using the RPA III RNase protection assay kit (Ambion). RNA was hybridized with 1,000 pg of DIG-UDP-labeled antisense RNA probe overnight at 42°C and digested with a mixture of RNase A and RNase T1 for 30 min at 37°C. The protected fragments were precipitated and separated on a denaturing 5% acrylamide/8 M urea gel and transferred to nylon membrane by capillary action. DIG-UDP-labeled RNA was detected by CDP-star (Tropix, Bedford, MA). Relevant RNA levels were calculated relative to those of GAPDH.

Western Blotting

For Western blotting, proteins were separated on a 12% polyacrylamide gel for 90 min at 25 mA. Proteins were then transferred to Hybond-P (Amersham Pharmacia Biotech, Piscataway, NJ) for 4 h at 60 V. Then, the membrane was blocked with 1% bovine serum albumin in Tris-buffered saline (137 mM NaCl and 25 mM Tris-HCl, pH7.4) with 0.1% Tween 20 (TBS-T) for 1 h. The membrane was then probed with anti-Kir2.1 polyclonal antibody diluted to 1:400 (Alomone Laboratory, Jerusalem, Israel) for 2 h at room temperature, and washed three times in TBS-T for 10 min each wash. The membrane was subsequently incubated with goat anti-rabbit IgG linked to peroxidase (Vector Labolatories, Burlingame, CA) diluted to 1:1,000 for 1 h at room temperature. After three additional washes, bound antibodies were detected by an ECL-plus (Amersham Pharmacia Biotech) and analyzed with an LAS-1000 image analyzer (Fuji-Film, Tokyo, Japan). The blot was stripped and reprobed with anti--actin monoclonal antibody (Sigma) to normalize lanes for the protein content.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Inward Rectifier K⁺ Current in hBSMCs

HBSMCs had Ca²⁺-activated K⁺ and Cl currents as described previously (30). In the present study, we investigated the characteristics and properties of Ca²⁺-independent K⁺ currents by using the whole-cell voltage clamp conditions. The patch pipette contained 10 mM EGTA and 3 mM ATP to block Ca²⁺-activated K⁺ and Cl currents and ATP-sensitive K⁺ channels. Figures 1 and 2 show typical K⁺ currents recorded from a same cell. Depolarizing voltage pulses from a holding potential of 70 mV induced outward current (Figure 1). The current rapidly activated, and then inactivated slowly, which was markedly inhibited by 4-AP (10 mM), a voltage-dependent K⁺ channel (K_V) blocker. Under the conditions with Cs⁺-internal solution, the current could not be observed (data not shown), suggesting that it consisted of K⁺ currents (K_V). The current- voltage (I-V) relationships at the initial peak and at the end of the pulse are illustrated in Figure 1B. The currents activated at potentials more positive than approximately 40 mV, and had delayed rectifying properties quite similar to the voltage-dependent K⁺ currents (K_V) previously described in human tracheobronchial smooth muscle cells (6). 4-AP (10 mM) markedly reduced the outward K⁺ currents at any command voltage steps (Figure 1B). Additionally, the inward current was elicited during the hyperpolarizing voltage pulses as shown in Figure 2. The I-V relationships were shown in Figure 2B. Voltage steps to various hyperpolarizing potentials from a holding potential of 70 mV induced a large inward current, but the depolarizing pulses to potentials between -60 mV and -30 mV elicited only small outward currents. Ba²⁺ (5 mM) blocked the inward current almost completely, and the remaining current might be only ascribed to the leakage current. Figure 2B illustrates the I-V relationships measured at the end of the pulse in the presence and absence of Ba²⁺. These I-V relations crossed the zero current level at 73 mV in this cell, and were linear for membrane potentials negative to about 90 mV. For membrane potentials positive to the reversal potential, however, the I-V relation was highly nonlinear and had inward rectifying properties for membrane potentials positive to 70 mV. When the external K⁺ concentration [K⁺]_o was increased from 5.4 to 20 and 50 mM (Figure 3) in the presence of 10 mM 4-AP, the I-V relations at the end of the pulse crossed zero current level at 83 mV in 5.4 mM [K⁺]_o, 48 mV in 20 mM [K⁺]_o, and 25 mV in 50 mM [K⁺]_o (Figure 3B), respectively. The slope conductance of the inward current region became much larger as [K⁺]_o increased; it was 16 nS at 5.4 mM [K⁺]_o, 25 nS at 20 mM [K⁺]_o, and 32 nS at 50 mM [K⁺]_o in this cell (Figure 3). Figure 4 shows the reversal potential of the current in the presence of 10 mM 4-AP, which demonstrated as Ba²⁺-sensitive current. The reversal potential was shifted when [K⁺]_o was changed. It shifted by 57 mV for a tenfold change in [K⁺]_o, indicating that the current consisted of K⁺ channels. The currents were also completely abolished under the conditions using Cs⁺-internal solution in the patch pipette. The inward current, which demonstrated as Ba²⁺-sensitive current, was detected at 42 of 45 cells tested. The current amplitude of the Ba²⁺-sensitive current elicited by hyperpolarizing pulses to -130 mV from a holding potential of -70 mV was -410 ± 220 pA (n = 42), and the current densities of the current were identified as 9.0 ± 4.3 pA/pF (n = 42).

View larger version (15K): [in this window] [in a new window]   Figure 1.   Voltage-dependent outward currents in human bronchial smooth muscle cells (hBSMCs). The cell was held at 70 mV, and various depolarizing pulses (620 ms in durartion) were applied. The bath contained nicardipine (1 µM) to block the voltage-dependent L-type Ca2+ channels, and the patch pipette contained K+-internal solution with 10 mM EGTA and 3 mM ATP to block Ca2+-dependent K+ and Cl currents and ATP-sensitive K+ channels. (A) Effects of 4-aminopyridine (4-AP) on the outward currents. The current traces in A are shown in control, and in the presence of 4-AP (10 mM). The vertical line represents zero current level. (B) The current-voltage relationships (I-V) measured at the peak and the end of pulse. The I-V relationships of the peak are shown in control (open circles) and in the presence of 4-AP (10 mM, closed circles). The I-V relationships measured at the end of pulses are illustrated in control (open square) and in the presence of 4-AP (10 mM, closed squares).

View larger version (19K): [in this window] [in a new window]   Figure 2.   Inward rectifier currents in hBSMCs. The data were obtained from a same cell shown in Figure 1. The cell was held at 70 mV, and command voltage pulses were applied to various membrane potentials. The current traces in A are shown in control, in the presence of Ba2+ (5 mM), and after washout of Ba2+. The zero current level is indicated by lines. (B) The I-V relationships measured at the end of the pulse, plotted in control (closed squares) and in the presence of Ba2+ (5 mM, closed circles). Note that both lines crossed the zero current level at approximately 73 mV, and the current had inwardly rectifying characteristics.

View larger version (27K): [in this window] [in a new window]   Figure 3.   Effects of changing the external concentration of K+ ions on the inward current. The cell was held at 70 mV and command voltage steps to various membrane potentials were applied. The current traces in Aa are shown in 5.4 mM [K+]o, 20 mM [K+]o, and 50 mM [K+]o in the absence of Ba2+. The current traces in Ab are indicated after the addition of Ba2+ (5 mM). (B) Effects of changing the external concentration of [K+]o on the I-V relationships at the end of pulses. The I-V relationships measured at the end of pulses were obtained by subtraction from control I-V relationships to that in the presence of Ba2+ (5 mM). Note that the activation voltages for inward rectification are changed when [K+]o was changed; when [K+]o was reduced, activation moved to more negative potentials. The slope conductance measured in the linear section of the I-V relationships in the potential range negative to the reversal potential was 16 nS at 5.4 mM [K+]o (triangles), 25 nS at 20 mM [K+]o (circles), and 32 nS at 50 mM [K+]o (squares), respectively.

View larger version (11K): [in this window] [in a new window]   Figure 4.   Effects of changes in [K+]o on the reversal potential for the inward rectifier current. The reversal potential was measured from I-V relationships measured at the end of pulses, which were obtained by subtraction from control I-V relationships to that in the presence of external Ba2+ (5 mM) as shown in Figure 4B, when the external K+ concentration ([K+]o) was changed. The mean and SD value are shown. The straight line, which is fitted to the data, has slopes of 57 mV/10-fold change in [K+]o. Data were obtained from six different cells.

Figure 5 shows the effects of various K⁺ channel blockers (charybdotoxin [CTX], glibenclamide, TEA, and Ba²⁺) on the inward current activated during the hyperpolarizing steps to -130 mV. Neither CTX, which blocks large-conductance Ca²⁺-activated K⁺ channels (Figure 5A), nor glibenclamide, a well-known ATP-sensitive K⁺ channel blocker, affected the K⁺ current (100 nM CTX by 4 ± 3% [n = 6] and 10 µM glibenclamide by 3 ± 2% [n = 5]). TEA at concentration of 2 mM, which preferentially inhibits large-conductance Ca²⁺-activated K⁺ channels, also did not inhibit it (2 mM TEA by 6 ± 4% [n = 5]), although high concentrations of TEA (20-140 mM) partly blocked it as shown in Figure 5C (140 mM TEA by 63 ± 12% [n = 5]). On the other hand, Ba²⁺ (5 mM) effectively inhibited it. Figure 6 illustrates the concentration-dependent effects of external Ba²⁺ on the inward K⁺ current. The cell was held at 70 mV and hyperpolarizing voltage pulses to 130 mV were applied to elicit the K⁺ current. In control, currents measured on hyperpolarization increased instantaneously and were maintained during the pulses. On the other hand, in the presence of Ba²⁺ (1 and 10 µM), the current declined from an initial value almost to a steady-state value before the end of the voltage pulse. Increasing Ba²⁺ concentration reduced the inward current in a concentration-dependent manner. Figure 6B shows the percent inhibition of the current level measured at the end of the pulse and the concentrations of external Ba²⁺. The half maximal inhibition (IC₅₀) was 1.3 µM in hBSMCs.

View larger version (26K): [in this window] [in a new window]   Figure 5.   Effects of various K+ channel blockers on the inward rectifier K+ current. The cells were held at 70 mV, and command voltage pulses to 130 mV were applied at 0.2 Hz. Effects of charybdotoxin (CTX, 100 nM; A), glibenclamide (10 µM; B), and tetraethylammium (TEA, 2-140 mM; C) on the inward rectifier current. The data present as a representative case obtained from five to six experiments in each case.

View larger version (14K): [in this window] [in a new window]   Figure 6.   Effects of external Ba2+ on the inward rectifier K+ current. The cells were held at 70 mV, and command voltage steps to 130 mV were applied. The current traces in A are shown in control and in the presence of Ba2+ (1, 10, and 100 µM, and 5 mM). (B) Concentration-dependent inhibition of the inward rectifier K+ current by Ba2+. The inhibitory effect of external Ba2+ on the current amplitude measured at the end of pulses is plotted against various concentrations of Ba2+. Data are shown as mean ± SD (n = 7).

Kir2.1 Encodes the Inward Rectifier K⁺ Channel in hBSMCs

The above results indicate that hBSMCs have Kir. Therefore, we further investigated the molecular aspects of Kir channel family members expressed in these cells. We examined the expression of Kir (1.1, 2.1, 2.2, and 2.3) mRNA. By RT-PCR, only expression of Kir2.1 mRNA was detected in these cells (Figure 7). The amplitude 199-bp Kir2.1 cDNA fragments were of predicted molecular size, identical to cDNA fragments amplified from reversely transcripted mRNA. In contrast, transcripts for Kir1.1, Kir2.2, and Kir2.3 subfamily members were not found within these same cells (Figure 7). Similar results were obtained from five different experiments, irrespective of the passage number (4 and 7). On the other hand, a positive control for Kir1.1, 2.1, 2.2, and 2.3 was observed in human fetal brain or human kidney (Figure 7).

View larger version (45K): [in this window] [in a new window]   Figure 7.   Expression of Kir2.1 mRNA in hBSMCs. Ethidium bromide-stained gel of the RT-PCR products for Kir1.1, Kir2.1, Kir2.2, and Kir2.3 mRNA. Note that Kir2.1, but not Kir1.1, Kir2.2, or Kir2.3, could be identified by RT-PCR. Kir2.1 N.C. = Kir2.1 negative control. As a positive control, Kir1.1 was identified in human kidney, and Kir2.1, 2.2, and 2.3 were observed in human fetal brain. X174/Hinc II digest was used for a marker.

The use of antisense oligonucleotides as specific gene inhibitors, in combination with measurements of biologic functions of specific proteins, offers a useful approach for evaluating the contributions of selected K⁺ channel isoforms to macroscopic K⁺ currents (28, 33). Therefore, inhibition of expression of Kir2.1, identified by RT-PCR, was attempted using antisense oligonucleotides directed to human Kir2.1. We first investigated the effects of antisense oligonucleotides on mRNA level and protein expression for Kir2.1 (Figure 8). RNase protection assay (Figure 8A) showed that in cells treated with antisense oligonucleotides, the mRNA level of Kir2.1 adjusted by the internal control (GAPDH mRNA) was only slightly decreased, compared with control cells treated with mismatch oligonucleotides (0.89 ± 0.04, n = 4). Figure 8B shows the effects of antisense oligonucleotides on the expression of Kir2.1 protein by using Western blotting. Antisense oligonucleotides significantly reduced the expression of Kir2.1 channel protein, as compared with the mismatch (0.19 ± 0.03, n = 4, P < 0.05).

View larger version (38K): [in this window] [in a new window]   Figure 8.   Effects of antisense oligonucleotides on mRNA and protein level of Kir2.1. The simultaneous isolation of total RNA and protein was done (see MATERIALS AND METHODS). (A) RNase protection assay for Kir2.1 mRNA. The mRNA level for Kir2.1 was compared with that for GAPDH in cells treated with antisense oligonucleotide for Kir2.1 and in cells treated with the mismatch. The data represent a typical recording obtained from four different experiments. (B) Western blotting analysis for the expression of Kir2.1 channel protein. The expression of Kir2.1 channel protein was compared with that of -actin in cells treated with antisense oligonucleotide for Kir2.1 and mismatch.

Figure 9 shows typical representative subtraction data obtained from a cell cultured in control medium with mismatch and a cell exposed for 6 h to the Kir2.1 antisense oligonucleotide. In a control cell, the voltage-dependent 4-AP-sensitive K⁺ currents (K_V) activated by depolarizing pulses and the Ba²⁺-sensitive K⁺ inward current activated by hyperpolarizing pulses (Figures 9Aa and 9Ba) were observed. On the other hand, in a cell treated with antisense oligonucleotide targeted to Kir2.1 mRNA, 4-AP-sensitive K⁺ current (Kv) activated by depolarizing pulses was similarly activated (Figure 9Ab), whereas Ba²⁺-sensitive current (Kir) activated by hyperpolarizing pulses was markedly reduced in a cell treated with antisense oligonucleotide (Figure 9Bb). Figure 10 summarizes the data using antisense experiments. The antisense oligonucleotides directed to human Kir2.1 significantly reduced the current density of the inward rectifying K⁺ current, which demonstrated as Ba²⁺-sensitive current, as compared with the mismatch-treated cells. On the other hand, it failed to affect the current density of the voltage-dependent K⁺ channels (K_V), which demonstrated as 4-AP-sensitive current.

View larger version (20K): [in this window] [in a new window]   Figure 9.   Suppression of the inward rectifier K+ current by exposure to antisense Kir2.1 oligonucleotide. Cells were held at 70 mV and command voltage steps to +30 mV, which activated the voltage-dependent K+ currents (KV), and to 130 mV, which activated the inward rectifier K+ current (Kir). The current traces (4-AP-sensitive and Ba2+-sensitive currents) are shown as the subtraction currents from the control trace to the current trace in the presence of 4-AP (10 mM) or Ba2+ (5 mM) (see MATERIALS AND METHODS). Examples of the currents elicited by these pulse protocols are shown in control cells after 6 h exposure to 0.25 µM mismatch Kir2.1 oligonucleotides (Aa and Ba), and a cell exposed to antisense Kir2.1 oligonucleotides (Ab and Bb). The data present a representative case in cells exposed to a mismatch Kir2.1 oligonucleotide, and cells exposed to an antisense Kir2.1 oligonucleotide. The data in Aa, Ba and Ab, Bb were obtained from a same cell, respectively. Note that in a cell treated with an antisense Kir2.1 oligonucleotide, the inward current reflected as the activation of Kir channel was markedly depressed.

View larger version (40K): [in this window] [in a new window]   Figure 10.   Suppression of the inward rectifier K+ current in hBSMCs by exposure to antisense Kir2.1 oligonucleotide. The cells were held at 70 mV, and command voltage pulses to +30 mV or 130 mV were applied to elicit the voltage-dependent K+ currents (KV), and the inward rectifier K+ current (Kir) as shown in Figure 9. The current densities of 4-AP (10 mM)-sensitive currents and Ba2+ (5 mM)-sensitive currents are plotted in cells exposed to mismatch Kir2.1 oligonucleotide and antisense Kir2.1 oligonucleotide. Mean ± SD value is indicated (n = 20). *P < 0.05 antisense versus mismatch. Shaded bars, mismatch; stippled bars, antisense.

Effects of Ba²⁺ on Membrane Potentials in hBSMCs

Figure 11 shows the effects of extracellular Ba²⁺ (5 mM) on membrane potentials. The application of Ba²⁺ (5 mM) markedly reduced the membrane potentials (from 45 ± 6 mV in control to 29 ± 8 mV in the presence of Ba²⁺ [n = 5]).

View larger version (7K): [in this window] [in a new window]   Figure 11.   Effects of Ba2+ ions (5 mM) on membrane potentials in hBSMCs. The patch pipette contained K+-internal solution. Results were representative of five similar experiments.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The major findings of the present study are: (i) the inward rectifier K⁺ (Kir) current as well as the voltage-dependent K⁺ currents (K_V) existed in hBSMCs; (ii) the Kir current was effectively inhibited by external Ba²⁺ with the IC₅₀ value of 1.3 µM; (iii) RT-PCR of mRNA showed transcripts for Kir2.1, but not Kir1.1, Kir2.2, or Kir2.3; (iv) treatment of cells with antisense oligonucleotides targeted to Kir2.1 resulted in a significant decrease in the current densities of Kir current and the expression of the Kir channel protein, whereas the current density of K_V did not change; and (v) Ba²⁺ markedly depolarized the membrane potential. These results demonstrate that the inward rectifier K⁺ (Kir) current is present in hBSMCs, and the Kir2.1 gene encodes the native Kir channel protein in these cells.

Our data showed that the inward rectifier K⁺ current as well as the voltage-dependent K⁺ currents existed in hBSMCs. The voltage-dependent K⁺ currents had outwardly rectifying properties, and were abolished by 4-aminopyridine (10 mM), similar to the previous studies using tracheal and bronchial smooth muscle cells in various kinds of species including human (3). hBSMCs express mRNA from several members of the K_V1 gene family (K_V1.1, K_V1.2, and K_V1.5), suggesting that these channel genes may play an essential role in forming the delayed outward K⁺ currents (K_V) in these cells (6). Furthermore, in the present study, the existence of Kir current was found in hBSMCs. There was no mention about the existence of Kir current in the previous paper (6), but the different properties among these papers might be due to the difference of the location of smooth muscles in the airway. It has been reported that Ba²⁺ (1 mM), which selectively inhibits Kir channel, depolarizes the membrane potential in canine small bronchi (3rd to 5th order) (21), proposing the existence of Kir channels in these cells. Recently, the inward rectifier K⁺ current has been reported in bronchial smooth muscle cells derived from human small intralobular bronchioles, but not in larger airway and trachea (22). Similarly, inward rectifier K⁺ current density is higher in myocytes from smaller than larger diameter porcine coronary arteries and rat coronary artery (13, 14). The Kir current expressed in hBSMCs had strong inward rectifier properties, which were inhibited by micromolar concentrations of external Ba²⁺. The IC₅₀ value of Ba²⁺ at 130 mV was 1.3 µM, which compares well with that of 2.2 or 2.1 µM reported previously for the inward rectifier K⁺ current found from cerebral and coronary myocytes, respectively (12, 17).

The present study also provides evidence showing that transcripts for 2.1 were expressed in hBSMCs, whereas transcripts for Kir1.1, Kir2.2, and Kir2.3 were not identified. Transcripts for Kir2.4, recently cloned from rat brain (26), were not investigated, due to the low sensitivity of the currents to external Ba²⁺ (IC₅₀ = 390 µM). These results were similar to those of the previous study of vascular smooth muscle cells showing that transcripts for Kir2.1, but not Kir2.2, and Kir2.3, are expressed in rat cerebral, coronary, and mesenteric artery (13). In addition, the targeted distruption of Kir2.1 and Kir2.2 gene shows that Kir2.1 gene expression, but not Kir2.2, which is considered to form native inward rectifier K⁺ current (IK₁) in cardiac myocytes (9), is required for Kir current in cerebral arteries (27). Thus, it is likely that Kir2.1 gene encodes the inward rectifier K⁺ current in hBSMCs as shown in arterial smooth muscle cells (13, 27). In fact, the use of antisense oligonucleotide technology provided additional molecular evidence that Kir2.1 was an essential component of the inward rectifier K⁺ current in hBSMCs. In these experiments, an antisense phosphorothioate oligonucleotide directed against Kir2.1 selectively reduced the inward rectifier K⁺ current, which was consistent with the analysis of the expression for Kir protein. On the other hand, it did not affect the voltage-dependent K⁺ currents (K_V), suggesting that the antisense oligonucleotide selectively inhibited Kir2.1 as previously shown in earlier papers using the antisense technology (28, 33). As shown in Figure 8, the decrease of Kir mRNA was relatively small in the antisense-treated cells, compared with the significant decrease of Kir protein, suggesting that the antisense oligonucleotide covering the translational initiation site used in the present study might act by inhibiting ribosomal translation of the target mRNA, rather than by inhibiting transcription from the gene (38).

Inward rectifying K⁺ channels have been thought to play an important role in maintaining membrane potential in both excitable and nonexcitable cells, including smooth muscle cells (10, 39). The membrane potential of hBSMCs was approximately 40-50 mV as reported previously (30), which is positive to the reversal potential of K⁺ channels, and therefore a physiologic role for Kir channel requires outward currents through the channel. In the present study, we could predict the outward currents of ~ 10-20 pA as shown in Figure 5. Considering the input resistance of the cells (~ 1 G), we could estimate the resulting effect on the cell membrane potential (10-20 mV), suggesting that the outward currents through the Kir channel have a physiologic role in regulating the membrane potential in bronchial smooth muscle cells. In fact, Ba²⁺ markedly reduced the membrane potential, as shown in Figure 11. In addition, if Kir channel is responsible, at least in part, for determining membrane potential in these cells, closure of the channel, for example by vasoconstrictors as reported in other cell types (42), would be an effective mechanism for causing depolarization, then causing voltage-dependent Ca²⁺ entry and constriction. Further studies are needed to clarify this possibility. The increased airway resistance seen during agonist challenge or during asthma attacks has been known to be due to contraction of small airways (43). Increases in [K⁺]_o occur in the heart and brain under the conditions such as hypoxia, ischemia, and neural activities (18). The rise in [K⁺]_o hyperpolarizes the membrane by activating Kir channels, resulting in vasodilation of small cerebral and coronary vessels, and thereby may cause a selective increase in the perfusion of metabolically active tissues. Similar mechanisms may be proposed for bronchial smooth muscle cells in the airways. Increases in [K⁺]_o under conditions such as ischemia may hyperpolarize the membrane by activating Kir channels, and thereby relax bronchial smooth muscle. Thus, K⁺-induced bronchodilation may be involved in the regulation of local airway flow, and control of ion channel expression may provide a means of regulating the responsiveness of bronchus of different diameter and therefore function.

In conclusion, the inward rectifier K⁺ current (Kir) is present in human bronchial smooth muscle cells, and the Kir2.1 gene encodes the Kir channel protein in these cells. Thus, it is likely that Kir channels play an important role in regulating local small bronchial tone under various pathophysiologic conditions.