K_ir2.1 encodes the inward rectifier potassium channel in rat arterial smooth muscle cells

Abstract

1. The molecular nature of the strong inward rectifier K⁺ channel in vascular smooth muscle was explored by using isolated cell RT-PCR, cDNA cloning and expression techniques.

2. RT-PCR of RNA from single smooth muscle cells of rat cerebral (basilar), coronary and mesenteric arteries revealed transcripts for K_ir2.1. Transcripts for K_ir2.2 and K_ir2.3 were not found.

3. Quantitative PCR analysis revealed significant differences in transcript levels of K_ir2.1 between the different vascular preparations (n = 3; P < 0.05). A two-fold difference was detected between K_ir2.1 mRNA and β-actin mRNA in coronary arteries when compared with relative levels measured in mesenteric and basilar preparations.

4. K_ir2.1 was cloned from rat mesenteric vascular smooth muscle cells and expressed in Xenopus oocytes. Currents were strongly inwardly rectifying and selective for K⁺.

5. The effect of extracellular Ba²⁺, Ca²⁺, Mg²⁺ and Cs²⁺ ions on cloned K_ir2.1 channels expressed in Xenopus oocytes was examined. Ba²⁺ and Cs⁺ block were steeply voltage dependent, whereas block by external Ca²⁺ and Mg²⁺ exhibited little voltage dependence. The apparent half-block constants and voltage dependences for Ba²⁺, Cs⁺, Ca²⁺ and Mg²⁺ were very similar for inward rectifier K⁺ currents from native cells and cloned K_ir2.1 channels expressed in oocytes.

6. Molecular studies demonstrate that K_ir2.1 is the only member of the K_ir2 channel subfamily present in vascular arterial smooth muscle cells. Expression of cloned K_ir2.1 in Xenopus oocytes resulted in inward rectifier K⁺ currents that strongly resemble those that are observed in native vascular arterial smooth muscle cells. We conclude that K_ir2.1 encodes for inward rectifier K⁺ channels in arterial smooth muscle.

Small arteries are a major contributor to the control of systemic blood pressure and local blood flow. Metabolic demand in these small arteries is linked to blood flow in part through the release of vasodilating substances such as potassium ions. Extracellular potassium concentrations have been demonstrated to reach >10 mM during ischaemia in both the cerebral (Sieber et al. 1993) and coronary (Kleber, 1983; Weiss et al. 1989) circulation. Unlike many other types of smooth muscle, elevations of extracellular potassium in these arteries do not lead to vasoconstriction, but to vasodilatation and (ultimately) increased blood flow (Katz & Linder, 1938; Bonaccorsi et al. 1977). Elevations in extracellular potassium concentrations occur in the heart and brain under physiological as well as pathophysiological conditions. While evidence suggests that the overall health of the heart and brain is dependent upon the fine regulation of coronary and cerebral blood flow by substances such as potassium ions, the mechanism(s) by which these substances regulate arterial diameter remain to be fully characterized.

A number of different mechanisms have been suggested to be involved in the regulation of arterial diameter by low concentrations of extracellular K⁺ including Na⁺-K⁺-ATPase (Webb & Bohr, 1978; McCarron & Halpern, 1990) and the inwardly rectifying potassium (K_ir) conductance (Edwards et al. 1988; McCarron & Halpern, 1990; Knot et al. 1996). The first evidence for inward rectifier K⁺ channels in arteries was provided by measuring currents in intact voltage-clamped cerebral (Hirst et al. 1986; Edwards et al. 1988) and mesenteric (Edwards & Hirst, 1988) arteries (for review see Hirst & Edwards, 1989). Subsequently, inward rectifier K⁺ currents were identified in isolated smooth muscle cells from cerebral (Quayle et al. 1993) and coronary arteries (Robertson et al. 1996; Quayle et al. 1996). The K_ir channels identified in these isolated arterial smooth muscle cells possess the characteristics of the K_ir2 subfamily - strong inward rectification, a conductance dependent upon extracellular potassium concentration, and a voltage- and time-dependent gating process (Quayle et al. 1993, 1996). Inward rectifier K⁺ channels in isolated cerebral and coronary myocytes also exhibit a distinct quantitative pattern of block by external barium, caesium, calcium and magnesium ions (Quayle et al. 1993; Robertson et al. 1996) (Table 1). Three distinct isoforms of the K_ir2 channel subfamily have been identified in the rat brain: K_ir2.1 (Kubo et al. 1993), K_ir2.2 (Takahashi et al. 1994) and K_ir2.3 (Morishige et al. 1994). Recently, a fourth member (K_ir2.4) has been identified and differs significantly from the other three members with respect to barium block (Töpert et al. 1998). Members of the K_ir2 family also show strong inward rectification and possess consensus sites for phosphorylation by protein kinases A and C (Henry et al. 1996). However, the molecular form(s) of K_ir channel expressed in vascular smooth muscle cells has not been identified.

Table 1.

Comparison of the electrophysiological properties for K_ir2.1, 2.2 and 2.3 expressed in oocytes and native arterial smooth muscle inward rectifier K⁺ currents

	Vascular smooth muscle	Kir2.1	Kir2.2	Kir2.3
RT-PCR of isolated arterial smooth muscle cells	Does not apply	+	—	—
Ba2+ block	Kd(0) = 38.1 μM (Quayle et al. 1993); Kd(0) = 21.1 μM (Robertson et al. 1996)	Kd(0) = 41.7 μM, μ = 0.51; Kd(0) = 62 μM, μ = 0.54(Shieh et al. 1998)	K1/2˜5 μM at −50 mV (Takahashi et al. 1994)	K1/2˜10 μM at −60 mV (Morishige et al. 1994)
Cs+block	Low affinity block; no measurable block with 50 μM at −60 mV; Kd(0) = 19.0 mM (Robertson et al. 1996); 4% reduction with 100 μM at −60 mV (Quayle et al. 1993)	Low affinity block; 0 to < 10% inhibition by 50 μM at −60 mV; Kd(0) = 50.8 mM, μ = 1.95, K1/2 = 457 μM at −60 mV; Kd(0) = 54 mM, μ = 1.57(Abrams et al. 1996), K1/2 = 300 μM to 1 mM (Kubo et al. 1993)	High affinity block; maximal block with 50 μM at −60 mV; K1/2 < 5 μM (Takahashi et al. 1994)	High affinity block; maximal block with 50 μM at −60 mV (Morishige et al. 1994)
Ca2+ block	49.3% reduction with 10 mM Ca2+ at −60 mV (Robertson et al. 1996)	42.8% reduction with 10 mM Ca2+ at −60 mV	Not done	Not done
Mg2+ block	52.8% reduction with 10 mM Mg2+ at −60 mV (Robertson et al. 1996)	58.1% reduction with 10 mM Mg2+ at −60 mV	Not done	Not done
Inactivation at hyper-polarizing potentials	Minimal	Minimal	Pronounced inactivation (Takahashi et al. 1994)	Minimal

The purpose of the present study was to test the hypothesis that one or more members of the K_ir2 channel subfamily is expressed in vascular smooth muscle cells and that cloned members of this subfamily possess similar pharmacological and electrophysiological characteristics to inward rectifier K⁺ currents recorded in native cells. We found only one form (K_ir2.1) of the inward rectifier potassium channel in single smooth muscle cells from cerebral, coronary and mesenteric arteries, with the highest level being observed in coronary arteries. Cloning and expression of K_ir2.1 from arterial smooth muscle in Xenopus oocytes yielded K⁺ currents that exhibited strong inward rectification. Inward rectifier K⁺ currents in native arterial myocytes exhibited a very distinct pattern of blocking properties of barium, caesium, calcium and magnesium ions which reflect the nature and positions of ion binding sites in the channel (Quayle et al. 1993; Robertson et al. 1996). Comparison of the blocking properties of these ions on native inward rectifier K⁺ currents and cloned K_ir2.1 current reveals striking quantitative similarities, and is different to what has been previously reported for K_ir2.2, 2.3 and 2.4 channels (Koyama et al. 1994; Morishige et al. 1994; Takahashi et al. 1994; Töpert et al. 1998).

METHODS

Isolation and collection of vascular smooth muscle tissues and cells

Sprague-Dawley rats (12–14 weeks, approximately 300 g) were overdosed with pentobarbitone (150 mg kg⁻¹ administered by intraperitoneal injection) and the heart, brain and mesenteric artery were removed after exsanguination. All experiments were carried out in accordance with the guidelines of the animal care committees of the Universities of Nevada and Vermont. Samples were placed in ice-cold extracellular solution containing 55 mM NaCl, 80 mM sodium glutamate, 6 mM KCl, 10 mM Hepes, 2 mM MgCl₂, 10 mM glucose and 100 μM CaCl₂ (pH 7.3). Coronary arteries were isolated from the right side of the ventricular septum. After dissection of basilar, coronary and mesenteric arteries, connective tissue was removed. The endothelium was removed on a portion of the artery samples by inverting the artery and scraping the lumenal surface with a glass slide.

Vascular smooth muscle cells were isolated by a procedure previously described (Robertson et al. 1996). Briefly, intact arteries were placed into the above-mentioned extracellular solution which included 1 mg ml⁻¹ papain, 1 mg ml⁻¹ bovine serum albumin (BSA), and 1 mg ml⁻¹ dithioerythritol, and incubated for various time periods (20 min for coronary arteries; 30 min for basilar arteries; 35 min for mesenteric arteries) at 37°C. Samples were then transferred to a second vial containing saline with 1 mg ml⁻¹ collagenase (30 % collagenase H and 70 % collagenase F; Sigma) and 1 mg ml⁻¹ BSA and incubated for 10 min at 37°C. The tissue was washed 5 times with ice-cold extracellular solution to remove any enzymes. Arteries were then triturated with a fire-polished Pasteur pipette to yield single smooth muscle cells. Cells were transferred to an experimental chamber and allowed to adhere to a glass coverslip for 10 min prior to perfusion of the bath with sterile extracellular solution (to remove any cellular debris). Microelectrodes were pulled from borosilicate glass (Sutter Instruments, Novato, CA, USA) using a vertical pipette puller yielding 40–50 μm tip diameters. After positioning the tip of the electrode over individual single smooth muscle cells, negative pressure was applied and individual cells were collected; approximately 60 myocytes were collected for each RNA preparation. Only spindle-shaped cells (length 100–150 μm; diameter 5–10 μm) were selected to ensure that samples were not contaminated by other cell types. Cells were expelled from the pipette tip into an RNase-free microcentrifuge tube after which the electrode was rinsed twice with phosphate-buffered saline solution to ensure that all cells were recovered. Cells and tissue samples were flash-frozen in liquid nitrogen and stored at −80°C for future use.

RNA isolation and reverse transcription-polymerase chain reaction (RT-PCR)

Total RNA was prepared from rat vascular smooth muscle tissues, brain and cells isolated from basilar, coronary, and mesenteric arteries using the SNAP Total RNA Isolation kit (Invitrogen, Carlsbad, CA, USA) as per the manufacturer's instructions. Polyinosinic acid (an RNA carrier; 20 μg) was added to the vascular smooth muscle cell lysates. First strand cDNA was synthesized from the RNA preparations using the Superscript II RNase H Reverse Transcriptase kit (Gibco BRL); RNA (1 μg) was reversed transcribed using an oligo (dT)_12–18 primer (500 μg μl⁻¹). To perform PCR the following sets of primers were used: K_ir2.1 sense, nucleotides (nt) 864–883, and anti-sense, nt 1170–1189 (gene accession no. GA L48490); K_ir2.1 sense, nt 335–356, and anti-sense, nt 1613–1632 (gene accession no. GA X73052). K_ir2.2 sense, nt 419–438, and anti-sense, nt 1235–1254 (GA X78461); and K_ir2.3 sense, nt 407–426, and anti-sense, nt 1303–1322 (GA S71382); and β-actin sense, nt 2383–2402 and anti-sense, nt 3071–3091 (GA V01217). PCR primers for β-actin were used to assess the viability of RNA samples as well as to detect genomic DNA contamination whereby the primers were designed to span an intron in addition to two exons. Brain tissue-derived RNA is included as a species control for the K_ir specific primers as all three forms of K_ir are expressed in brain neuronal tissue. Complimentary DNA (20 % of the first strand reaction) was combined with sense and anti-sense primers (20 μM), 1 mM dNTPs, 40 mM Tris-HCl (pH 8.3), 100 mM KCl, 3 units of TAQ (Promega), 1 Ampliwax Gem 100 (Perkin Elmer, Applied Biosystems, Foster City, CA, USA), and RNase-free water to a final volume of 50 μl. PCR was performed in a COY II Thermal Cycler under the following conditions: 32 cycles at 94°C, 1 min; 57°C, 30 s; 72°C, 1 min; and then, incubated at 72°C, 10 min. A second round of PCR (5 μl of 1st PCR reaction in lieu of cDNA) was performed under the same conditions. PCR products were separated by 2 % agarose gel electrophoresis, directly ligated into pCR2.1 vector constructs (Invitrogen), and transformed using the TA Cloning Kit (Invitrogen). DNA sequences were determined using an automatic nucleotide sequencer (Model 310, Applied Biosystems, Foster City, CA, USA).

Quantitative PCR (Q-PCR)

Q-PCR was performed using the PCR MIMIC Construction Kit (Clontech, Palo Alto, CA, USA) which is based upon a competitive PCR approach - non-homologous engineered DNA standards (referred to as PCR MIMICs) compete with target DNA for the same gene-specific primers. PCR MIMICs were constructed for K_ir2.1 and β-actin and competitive PCR was carried out by titration of sample cDNA with known amounts of the desired non-homologous PCR MIMIC constructs; serial dilutions (tenfold and then, twofold) of these constructs were then added to PCR amplification reactions. Following PCR, products were separated by 2 % agarose gel electrophoresis and quantified using Molecular Analyst (Biorad, Hercules, CA, USA).

Data analysis and statistical treatments for molecular studies

Qualitative PCR was performed on 60-cell preparations isolated from basilar, coronary and mesenteric arteries from four different rats. For quantitative PCR studies, arteries from two to four different rats were pooled for each RNA isolation; at least four different RNA isolations were performed. The concentrations of the target DNA and β-actin were determined for each sample; the target DNA concentration was then normalized to β-actin expression. All results are expressed as means ±s.e.m. Data for quantitative PCR were analysed by one-way analysis of variance (ANOVA) and differences between tissues were illustrated by Newman-Keuls multiple comparison tests; P values < 0.05 were considered to be significant.

Cloning and in vitro transcription of vascular K_ir2.1

K_ir2.1 was cloned from rat mesenteric smooth muscle cells by first performing PCR in the presence of gene-specific K_ir2.1 sense (nt 335–356; GA X73052) and anti-sense (nt 1613–1632; GA X73052) primers. Full-length cDNA fragments were ligated into pCR2.1 vector constructs (Invitrogen) and transformed using the TA cloning kit (Invitrogen). Clones were then sequenced using an automated nucleotide sequencer (Applied Biosystems, Model 310). Capped RNA (cRNA) was in vitro transcribed from pCR2.1 plasmids containing full-length K_ir2.1 cDNA using the Ambion mMessage mMachine transcription kit (Ambion, Austin, TX, USA). Briefly, reactions contained 2–5 μg of linearized template DNA, 500 μM ribonucleotides, 1 × transcription buffer and 40 units of T7 RNA polymerase; the concentration of the final cRNA was adjusted to 1 ng μl⁻¹.

Xenopus oocyte preparation and K_ir2.1 RNA injection

Female Xenopus laevis frogs were anaesthetized by immersion in 0.2 % tricaine (pH 7.5) for 3–5 min. Stage V-VI oocytes were obtained by partial ovarectomy and washed in ND96 solution, comprising (mM): 96 NaCl, 2 KCl; 1.8 CaCl_2, 1 MgCl₂, 5 Hepes, pH 7.5. Oocytes were defolliculated by rotation in Ca²⁺-free ND96 containing 2 mg ml⁻¹ collagenase (Type 1A, Sigma) for 45 min at room temperature (22°C). Collagenase was removed and the follicular layer of individual oocytes was then carefully removed using fine-tipped forceps while the oocytes were immersed in ND96 solution. Oocytes were then stored at 16°C for 24 h in ND96 solution supplemented with 0.55 mg ml⁻¹ pyruvate and 50 μg ml⁻¹ gentamicin after which they were injected (Nanoject, Drummond, PA, USA) with water (control) or 40–50 ng of cRNA (final volume 50 nl). Oocytes were maintained following injection at 16°C in ND96 supplemented solution for 3–6 days prior to electrophysiological recording.

Double electrode voltage clamp

Experiments were performed at room temperature (22°C) using the two-electrode voltage clamp setup of Dr Alan Segal (Ion Channel Group, University of Vermont). Microelectrodes, pulled from borosilicate glass, were bevelled to ensure clean impalement (1–5 MΩ), and filled with 1 M KCl for recording. Currents were recorded using a two-electrode voltage clamp amplifier (Warner Instrument Corp.) and filtered at 400 Hz using a low-pass Bessel filter (Frequency Devices). Data were acquired and analysed using pCLAMP software (Axon Instruments,) on PC-compatible computers.

Membrane currents were recorded using either voltage ramps or voltage pulses. Voltage ramps of 480 ms duration from −100 to +60 mV (0.3 mV ms⁻¹) were performed by initially holding the membrane potential at 0 mV and stepping to −100 mV for 15 ms prior to the start of the ramp. During voltage pulses (1, 5 or 10 s duration), cells were held at −10 mV and stepped from +60 to −120 mV in −20 mV increments.

The standard bath solution contained the following (mM): 90 KCl, 5 Hepes, 0.1 CaCl₂, and 1 MgCl₂. KCl was replaced with an equimolar concentration of NaCl to test the effect of [K⁺]_o. MgCl₂ was not added to the bathing solution when testing inhibition by Ca²⁺ and Mg²⁺. Ca²⁺ (100 μM) was present in all experiments.

Currents other than those through inward rectifier K⁺ channels were measured by performing voltage protocols in the presence of high concentrations (10 times the half-block constant at 0 mV for inward rectifier currents) of Ba²⁺ (300 or 500 μM). Unless indicated, the Ba²⁺-insensitive current was subtracted from test currents to reveal the Ba²⁺-sensitive current (Robertson et al. 1996). Current at the end of voltage pulses, where inhibition was observed to have reached a steady level (10 s for Ba²⁺; 5 s for Ca²⁺ and Mg²⁺; and 1 s for Cs⁺), was measured to determine current levels and calculate fractional inhibition.

Patch clamp electrophysiology

The whole-cell configuration of the patch clamp technique (Hamill et al. 1981) was used on vascular smooth muscle cells isolated from rat coronary arteries. The standard pipette (internal) solution contained the following (mM): 107 KCl, 33 KOH, 10 EGTA, 10 Hepes, 1 MgCl₂, 1 CaCl₂ (∼20 nM free Ca²⁺), 3 Na₂ATP, 0.1 NaADP, and 0.2 Na₃GTP (pH 7.2). The bath solution contained 140 mM KCl, 1 mM MgCl₂, 10 mM Hepes and 100 μM CaCl₂ (pH 7.4). All experiments were performed at room temperature (22°C). Pipettes were pulled from borosilicate glass (Sutter Instruments) using a Narishige puller (PP-83) to a resistance of 3–5 MΩ. Tips were coated with sticky dental wax and fire polished (Narishige MF-83). Whole-cell currents were acquired using an Axopatch 200 amplifier (Axon Instruments) and an 8-pole Bessel filter. Currents were filtered at 1 kHz. Data were acquired and analysed using pCLAMP software (Axon Instruments). All results are expressed as means ±s.e.m. of n cells.

To eliminate other K⁺ currents, the membrane potential was measured generally negative to −20 mV (which would limit the activation of Ca²⁺-activated K⁺ (K_Ca) channels and voltage-activated K⁺ channels; Nelson & Quayle, 1995). In addition, the pipette solution contained low concentrations of Ca²⁺ (to minimize the activity of K_Ca channels) and 3 mM ATP (to inhibit ATP-sensitive K⁺ channel activity). To determine the Ba²⁺-sensitive current a protocol similar to that described above was used to subtract currents other than those through inward rectifier channels.

Data fitting

Fractional inhibition of inward currents by ionic block (Ba²⁺ and Cs⁺) were fitted by the following equation (eqn (1)) (e.g. for Ba²⁺):

(1)

where I_Ba is the K_ir current in Ba²⁺, I_con is the K_ir current in the control solution, [Ba²⁺] is the Ba²⁺ concentration, and K_d is the apparent dissociation constant for Ba²⁺.

To determine the voltage dependence of the dissociation constants the following equation (eqn (2)) was used to fit the data:

(2)

where K_d(0) is the dissociation constant at 0 mV, V is the voltage, z is the valency of the ion, F is Faradays constant, μ is the slope constant (indicating the sensitivity of the K_d to the applied voltage), R is the gas constant, and T is the temperature in Kelvin.

To determine the kinetics of Ba²⁺ block, the onset of Ba²⁺ block was fitted with a single exponential. The time constant (τ) is related to the blocking (k_b) and unblocking (k_−b) rate constants and the Ba²⁺ concentration by the following equation (eqn (3)):

(3)

RESULTS

Identification of the K_ir2 channel subfamily members

The molecular species of K_ir2 channel subfamily members expressed in vascular smooth muscle cells was determined by isolating total RNA from basilar, coronary and mesenteric smooth muscle cell preparations. Qualitative analysis of transcriptional expression in dispersed and selected smooth muscle cells minimizes the possibility of contaminating expression from non-smooth muscle cell types. First strand cDNA was synthesized and then amplified by PCR. As indicated in Fig. 1A, RT-PCR yielded visible amplified products of K_ir2.1 in mRNA isolated from basilar, coronary and mesenteric arterial smooth muscle cells. In contrast, transcripts for K_ir2.2 or 2.3 subfamily members were not found within these same cell preparations (Fig. 1B and C, respectively). Amplification products corresponding to K_ir2.2 and 2.3 were detected in mRNA isolated from rat cerebellum (neuronal) (lane 2, B and C, respectively), indicating that the primers were capable of amplifying the gene specific products in rat. Two amplifications were performed for detection of K_ir2.2 and 2.3 expression. Therefore, at this level of detection, these K_ir family members could not be detected. We further tested mRNA preparations derived from intact, denuded arteries to test whether the detection level of isolated cells was too low to detect transcriptional expression of these K_ir forms. Neither K_ir2.2 nor 2.3 expression could be detected from these preparations. The K_ir2.1 fragment sequences were determined to be 100 % homologous to rat tumour mast cell line K_ir2.1 mRNA (Wischmeyer et al. 1995).

Figure 1. RT-PCR detection of K_ir2.1 in anatomically different vascular smooth muscle cell preparations.

Amplified PCR products generated using gene-specific primers for K_ir2.1 (A), K_ir2.2 (B), or K_ir2.3 (C) were fractionated on 2 % agarose gels and size markers were used to indicate the size of the experimental fragments (lane 1). Transcriptional expression for K_ir2.1 was detected in basilar, coronary and mesenteric arteries, whereas no expression for K_ir2.2 and K_ir2.3 was observed. Rat cerebellum was used as a positive control for the K_ir2 primers.

While qualitative RT-PCR demonstrated the molecular species of K_ir2 channel subfamily members present in vascular smooth muscle cells, differences in expression levels of these transcripts in the different arterial preparations were detected by quantitative RT-PCR. Studies using quantitative competitive PCR were performed by including gene-specific primers for K_ir2.1 or β-actin in the PCR reaction. A constant volume of the target cDNA was used and varying concentrations of a specific non-homologous engineered DNA fragment was included in the reaction. A representative competitive PCR gel is shown in Fig. 2A. Actual concentrations of the K_ir2.1 cDNA and, in turn, mRNA in the various arteries are expressed relative to β-actin RNA concentration (Fig. 2B). Transcriptional levels of K_ir2.1 in coronary arteries (0.0146 ± 0.0032) were significantly greater than those detected in mesenteric (0.0071 ± 0.001) or basilar (0.0068 ± 0.001) preparations (n = 3, P < 0.05). These results indicate that arterial smooth muscle cells express K_ir2.1 transcripts.

Figure 2. Differential transcriptional expression of K_ir2.1 in rat arteries.

Competitive PCR products were resolved on 2 % ethidium bromide agarose gels. A representative gel of quantitative RT-PCR for K_ir2.1 in rat basilar, coronary and mesenteric arteries is shown in A; two-fold serial dilutions of mimic DNA were included in the PCR reactions while the target cDNA (K_ir2.1) concentration remained constant. The actual concentrations of target cDNA were calculated and expressed relative to β-actin cDNA concentration (B). Results are expressed as means ±s.e.m.; *significant difference when compared with transcript levels detected in coronary artery.

Characterization of cloned K_ir2.1 expressed in Xenopus oocytes and comparison with native K_ir channel currents

To test the hypothesis that K_ir2.1 encodes inward rectifier potassium channels in arterial smooth muscle, currents were recorded from cloned K_ir2.1 expressed in Xenopus oocytes and were compared with K_ir channels recorded from native vascular smooth muscle cells. Two-electrode voltage clamp was performed on Xenopus oocytes 3–6 days after injection with RNA encoding the K_ir2.1 channel protein. In a bathing solution containing 90 mM K⁺ (i.e. E_K = 0 mV), voltage steps of 10 s duration from +60 to −100 mV (from a holding potential of −10 mV) evoked a strong, inwardly rectifying current which was selective for K⁺ ions (Fig. 3A). These currents were not observed in control oocytes injected with 50 nl of H₂O (data not shown). Figure 3B demonstrates the Ba²⁺-sensitive (500 μM) currents observed in an isolated coronary (septal) artery smooth muscle cell in a bathing solution containing 140 mM K⁺ (i.e. E_K = 0 mV). This cell was held at a membrane potential of 0 mV with voltage pulses between −20 and −100 mV applied every 10 s in 20 mV increments.

Figure 3. Arterial smooth muscle inward rectifier currents.

A, inwardly rectifying potassium currents recorded from a Xenopus oocyte injected with RNA encoding a K_ir2.1 channel protein cloned from vascular smooth muscle. The oocyte was held at a membrane potential of −10 mV and voltage pulses (of 10 s duration), from +60 to −100 mV were applied every 4 s in 20 mV increments. The bathing solution contained 90 mM K⁺. B, Ba²⁺-sensitive (500 μM) currents in an isolated coronary (septal) artery smooth muscle cell in a bathing solution containing 140 mM K⁺ (i.e. E_K = 0 mV). The cell was held at a membrane potential of 0 mV with voltage pulses of between −20 and −100 mV applied every 10 s in 20 mV increments.

K⁺ selectivity of cloned K_ir2.1 expressed in Xenopus oocytes

We investigated the effect of changing external K⁺ concentrations (5.4, 30, 60 and 90 mM K⁺) on the properties of K_ir2.1 currents expressed in Xenopus oocytes (Fig. 4A). The slope conductance of the current (measured from the straight portion of the I-V relationship) increased with external K⁺ (Fig. 4A). A similar change in the slope conductance and reversal potential (E_rev) was also observed in Fig. 4B, which demonstrates the Ba²⁺-sensitive currents (average of five voltage ramps from −140 to +50 mV) observed in an isolated coronary (septal) artery smooth muscle cell when external (bathing) K⁺ was changed. The slope conductances and E_rev values for this cell were 0.77 and 1.99 nS, and −47.3 mV (E_K =−49.5 mV) and −3.4 mV (E_k= 0 mV) with 20 and 140 mM external K⁺, respectively. Slope conductances for Xenopus oocytes injected with K_ir2.1 cRNA were 57.2 ± 7.9, 97.6 ± 10.7, 115.9 ± 11.6 and 125.5 ± 10.4 μS with 5.4, 30, 60 and 90 mM K⁺, respectively (n = 8) (Fig. 5A). The slope conductance was equal to ([K⁺]_o)^0.28 (Fig. 5A). The E_rev became more positive with an increase in external K⁺, consistent with the potassium-selective nature of the expressed K_ir2.1 channels (Fig. 5B).

Figure 4. Inward rectification shifts with a change in external [K⁺].

A, current-voltage relationship from the same Xenopus oocyte in a bathing solution containing 90, 60, 30 or 5.4 mM K⁺. I-V relationships were obtained by performing voltage ramps initially by holding at a membrane potential of 0 mV, followed by a pulse to −100 mV for 15 ms before a voltage ramp of 480 ms duration from −100 to +60 mV (0.3 mV ms⁻¹). B, Ba²⁺-sensitive currents (average of 5 voltage ramps from −140 to +50 mV) observed in an isolated coronary (septal) artery smooth muscle cell with 20 and 140 mM external (bathing) K⁺. The slope conductance and E_rev for this cell were 0.77 and 1.99 nS, and −47.3 mV (E_K =−49.5 mV) and −3.4 mV (E_K = 0 mV) with 20 and 140 mM external K⁺, respectively.

Figure 5. Elevation of external [K⁺] increases the conductance of inward rectifier potassium currents and shifts the reversal potential.

A, the slope conductance of the inward rectifier current increased with external [K⁺] (n = 8 oocytes). Slope conductance (C) was fitted with (C)([K⁺]_o)ⁿ, giving C = 36.7μS and n = 0.28. B, changing the concentration of K⁺ in the bathing solution shifted the reversal potential (30 mM K⁺, n = 8 oocytes; 60 mM K⁺, n = 8 oocytes; 90 mM K⁺, n = 24 oocytes). The line represents the best fit to the relationship between external K⁺ and the reversal potential, with a slope corresponding to 53.1 mV shift in E_rev for a 10-fold change in [K⁺].

External Ba²⁺ inhibition of K_ir2.1 currents in Xenopus oocytes

An important feature of native K_ir channels is high affinity block by external barium ions. The effect of external Ba²⁺ concentrations on K_ir2.1 expressed in Xenopus oocytes was determined. Figure 6A illustrates currents evoked from a holding potential of −10 mV to −60 mV under control conditions and in the presence of 10 and 100 μM Ba²⁺ in the bathing solution. K_ir2.1 currents was inhibited by external Ba²⁺ in a concentration-dependent manner. The rate of onset and the degree of Ba²⁺ block increased with the concentration of Ba²⁺. Figure 6B and C shows the voltage- and concentration-dependent inhibition of K_ir current at the end of the 10 s voltage pulse when block had reached a steady-state level. We compared the Ba²⁺ sensitivity of K_ir2.1 with that of native inward rectifier currents from vascular tissues. The fractional inhibition of K_ir2.1 currents (expressed as the Ba²⁺-sensitive current measured at different voltages and concentrations of Ba²⁺ divided by the Ba²⁺-sensitive current in control) was then plotted against the concentration of Ba²⁺ (Fig. 6C; n = 5). The data presented in Fig. 6C were fitted by eqn (1) (refer to Methods) to determine the apparent dissociation constant (K_d) at each potential. The following K_d values were obtained: 21.4 μM at −20 mV, 7.2 μM at −40 mV, 3.5 μM at −60 mV, 1.6 μM at −80 mV, and 0.8 μM at −100 mV (Table 1). Figure 6D demonstrates that the K_d decreased e-fold per 25 mV hyperpolarization; these data were fitted with eqn (2) (refer to Methods) with a K_d(0) of 41.7 μM and a slope (μ) of 0.51.

Figure 6. Ba²⁺ inhibition of inward rectifier potassium currents in Xenopus oocytes injected with RNA encoding K_ir2.1 cloned from vascular smooth muscle cells.

Membrane current recorded from the same oocyte with voltage steps from a holding potential of −10 mV to −60 mV in the presence of 0 (control), 10 and 100 μM Ba²⁺. External K⁺ was 90 mM. Dotted line indicates zero current level. B, current-voltage relationship demonstrating the effect of applying 0.3, 1, 3, 10, 30 and 100 μM Ba²⁺ on membrane currents recorded at the end of 10 s voltage pulses (n = 5). Bathing K⁺ was 90 mM. C, relationship between external Ba²⁺ concentration and the fractional inhibition of inward current at −20, −40, −60, −80 and −100 mV. Data were fitted with eqn (1), to give K_d values of (μm): 21.4, 7.2, 3.5, 1.6 and 0.8 at −20, −40, −60, −80 and −100 mV, respectively. D, voltage dependence of the dissociation constants (K_d) from C. Data were fitted with eqn (2) with a K_d at 0 mV of 41.7 μM, and a slope (μ) of 0.51.

Immediately upon membrane hyperpolarization Ba²⁺ block increases at a rate which depends upon the voltage applied and the concentration of Ba²⁺ present. Assuming that the blocking reaction is a simple bimolecular reaction, the reduction in K_ir current by Ba²⁺ can be fitted by a single exponential decay function with time constant τ. At a given membrane potential, τ decreased with increasing Ba²⁺ concentration: e.g. with 10 μM Ba²⁺, at −20 mV, τ= 2814 ± 356 ms; at −40 mV, τ= 1647 ± 151 ms; at −60 mV, τ= 1159 ± 137 ms; at −80 mV, τ= 688 ± 91 ms; and at −100 mV, τ= 387 ± 55 ms (n = 5). Also, at a given membrane potential τ decreased with increasing Ba²⁺ concentration: e.g. at −60 mV, with 1 μM Ba²⁺, τ= 2810 ± 94 ms; with 3 μM Ba²⁺, τ= 2151 ± 149 ms; with 10 μM Ba²⁺, τ= 1159 ± 137 ms; with 30 μM Ba²⁺, τ= 496 ± 62 ms (n = 5). The time constant is related to the blocking (k_b) and unblocking (k_−b) rate constants, and the Ba²⁺ concentration as described in eqn (3) (refer to Methods). The reciprocal of τ plotted against [Ba²⁺] gives a straight line with slope equal to k_b, and intercept equal to k_−b (Fig. 7A). The k_b was an exponential function of membrane potential, indicating that k_b increases with membrane hyperpolarization, while k_−b changed relatively little (Fig. 7B). Fitting k_b values with an equation analogous to eqn (2) (refer to Methods) demonstrated a k_b at 0 mV of 7.4 M⁻¹ ms⁻¹, and a μ value of −0.43. k_−b changed only slightly over the membrane potential range, demonstrating a value of 0.26 s⁻¹ at 0 mV, and a μ value of −1.02, indicating that the majority of the voltage dependence of the K_d comes from the voltage dependence of the blocking rate constant, k_b. Since the K_d is a function of k_b and k_−b, such that K_d=k_−b/k_b, the rate constants allow an alternative estimate of the K_d at different membrane potentials. Using this procedure, the calculated K_d values were 16.4 μM at −20 mV, 10.3 μM at −40 mV, 4.9 μM at −60 mV, 1.9 μM at −80 mV, and 1.3 μM at −100 mV (see Table 1). Since k_−b=K_d× k_b, k_−b can also be determined using the K_d and the k_b calculated above, and is 0.31 s⁻¹ at 0 mV.

Figure 7. Concentration and voltage dependence of blocking kinetics of Ba²⁺ on inward rectifier K⁺ currents from vascular smooth muscle expressed in Xenopus oocytes.

A, the onset of Ba²⁺ block was fitted with a single exponential after which the mean reciprocal values (±s.e.m.) of the time constant (τ) of current decay was plotted against Ba²⁺ concentration (n = 5 oocytes). These data were fitted by eqn (3) to determine the k_b, the slope of the line, and k_−b, the intercept (see Table 1). B, voltage dependence of the blocking rate, k_b, and unblocking rate k_−b, obtained from the slope and intercept of lines fitted in A. Data were fitted with an equation equivalent to eqn (2) with k_b values of 7.4 m⁻¹ s⁻¹ at 0 mV and slope (μ) of −0.43, and k_−b values of 0.26 s⁻¹ at 0 mV and μ of −1.02.

External caesium inhibition of inward rectifier K⁺ currents expressed in Xenopus oocytes

Another important characteristic of native K_ir channels in arterial smooth muscle is the steep voltage dependence of block by caesium ions. To determine the effect of external Cs⁺ on K_ir2.1 channel activity, we examined the effect of different concentrations of Cs⁺ on currents evoked by 1 s voltage pulses to membrane potentials between +40 and −120 mV, in 90 mM [K⁺]_o. External Cs⁺ blocked K_ir2.1 currents expressed in Xenopus oocytes. Inhibition of the current increased with both Cs⁺ concentration and hyperpolarization (Fig. 8A-C). Note in Fig. 8A that the current recorded at −100 mV in the presence of 1 mM Cs⁺ is smaller than that observed at −10 mV (holding potential) in control, demonstrating the voltage dependence of block. Figure 8B demonstrates that as the membrane potential becomes more hyperpolarized the current inhibition with Cs⁺ increases, even at the same concentration (n = 8 oocytes). Data were fitted by using eqn (1) (refer to Methods). The K_d values for Cs⁺ at each membrane potential were calculated as follows: 2911.7 μM at −40 mV; 458.3 μM at −60 mV; 91.6 μM at −80 mV; 19.9 μM at −100 mV; and 6.5 μM at −120 mV (Fig. 8C). The K_d decreased e-fold per 13 mV (Fig. 8D), with an estimated K_d(0) of 50.8 mM, and a slope (μ) of 1.95. The expressed channels exhibited the characteristic steep voltage dependence of caesium ion block observed in inward rectifier channels from arterial smooth muscle.

Figure 8. External Cs⁺ block of inward rectifier potassium currents expressed in Xenopus oocytes by injection with RNA encoding K_ir2.1 cloned from vascular smooth muscle.

A, membrane current recorded from the same oocyte with voltage steps from a holding potential of −10 to −100 mV in the presence of 0 (control), 10, 100 and 1000 μM Cs⁺. Note that the current recorded at −100 mV in the presence of 1 mM Cs⁺ is smaller than that observed at −10 mV in control, demonstrating the voltage dependence of block. External K⁺ was 90 mM. B, current-voltage relationship demonstrating the voltage dependence of block of Cs⁺ at different membrane potentials (n = 8 oocytes). Currents were measured at the end of 1 s voltage pulses. C, relationship between Cs⁺ concentration and fractional inhibition of inward current at −40, −60, −80, −100 and −120 mV. Data were fitted using eqn (1) demonstrating K_d values of (μm): 2911.7, 458.3, 91.6, 19.9 and 6.5 at −40, −60, −80, −100 and −120 mV, respectively. D, voltage dependence of K_d. Data were fitted using eqn (2) with a K_d (0) of 50.8 mM and μ of 1.95.

External Ca²⁺ and Mg²⁺ inhibition of inward rectifier K⁺ currents expressed in Xenopus oocytes

External calcium and magnesium ions block K_ir channel currents in native coronary myocytes at physiological concentrations. We therefore examined the effects of external Ca²⁺ and Mg²⁺ on K_ir2.1 channel activity were determined by using 5 s voltage steps from a holding potential of −10 mV to −60 mV with 90 mM [K⁺]_o. The effects of adding 1, 5 and 10 mM Ca²⁺ or Mg²⁺ were compared with currents evoked in control (100 μM Ca²⁺ and nominal Mg²⁺ present). At −60 mV addition of 1, 5 or 10 mM Ca²⁺ reduced inward currents by 13.3, 31.9 and 42.8 % (n = 8), respectively (Fig. 9A and B); Fig. 9B also demonstrates that the Ca²⁺ block is relatively unaffected by voltage. K_ir2.1 currents were also inhibited in a concentration-dependent manner by Mg²⁺, with 1, 5 and 10 mM Mg²⁺ inhibiting the current by 18.1, 47.5 and 58.1 % (n = 8), respectively, at −60 mV. Inhibition of currents by Mg²⁺ was also independent of voltage (Fig. 9C and D).

Figure 9. External Ca²⁺ and Mg⁺ block of inward rectifier currents expressed in Xenopus oocytes by injection of RNA encoding K_ir2.1 cloned from vascular smooth muscle.

A, membrane current recorded from an oocyte in response to a 5 s voltage step from −10 to −60 mV in control (100 μM Ca²⁺) and after addition of 5 mM Ca²⁺. The bathing solution contained 90 mM K⁺ and nominal Mg²⁺. B, fractional inhibition of inward rectifier currents by addition of 1, 5 and 10 mM Ca²⁺ at different voltages (n = 8 oocytes). Inhibition of currents by Ca²⁺ is not voltage dependent. C, membrane current recorded from an oocyte in response to a 5 s voltage step from −10 to −60 mV in control (0 Mg²⁺) and in the presence of 5 mM Mg²⁺. The bathing solution contained 90 mM K⁺ and nominal Mg²⁺. D, fractional inhibition of inward rectifier currents by 1, 5 and 10 mM Mg²⁺ at different voltages (n = 8 oocytes). Inhibition of currents by Mg²⁺ is not voltage dependent.

DISCUSSION

Molecular identification of K_ir2 channel subfamily members

The current study indicates that transcripts for K_ir2.1 are expressed in vascular smooth muscle cells. While transcripts for K_ir2.1 were identified in the various arterial cell preparations, transcripts for the other members of the K_ir2 subfamily (K_ir2.2 and 2.3) were not. Transcripts for an inward rectifier channel, K_ir2.4, recently cloned from rat brain were not examined, since it is unlikely this molecular species is responsible for K⁺-induced dilatations of arteries due to the low sensitivity of the currents to Ba²⁺ ([K⁺]_i= 390 μM) (Töpert et al. 1998). There was significantly more K_ir2.1 transcript expressed in coronary arteries than either basilar or mesenteric arteries. If these differences were translated into significantly more current in coronary artery myocytes, this could lead an increased effect of K⁺-induced dilatations in these vessels. Inward rectifier K⁺ current density is higher in myocytes from smaller than larger diameter porcine coronary arteries (Quayle et al. 1996). Consistent with this finding, we show that myocytes from the smallest diameter artery (coronary) which we tested had the highest level of K_ir transcript. These coronary arteries also exhibit pronounced barium-sensitive K⁺-induced dilatations (Knot et al. 1996). However, smooth muscle cells of small (arcuate) arteries from the renal vasculature have a relatively low inward rectifier K⁺ current density and, consistent with this, do not exhibit Ba²⁺-sensitive K⁺ dilatations (Prior et al. 1998).

Characterization of cloned K_ir2.1 expressed in Xenopus oocytes

Following the molecular characterization studies, we determined whether the RNA encoding the K_ir2.1 channel protein cloned from isolated vascular smooth muscle cells demonstrated the same inwardly rectifying K⁺ currents as those observed in native vascular smooth muscle cells (Table 1). Previous studies have demonstrated that strong, inwardly rectifying K⁺ currents in native vascular smooth muscle cells are sensitive to extracellular K⁺ and low (μM) concentrations of Ba²⁺(Quayle et al. 1993; Robertson et al. 1996). It has previously been shown that a change in external K⁺ had a number of effects on K_ir currents: (1) inward rectification shifted with E_K; (2) E_rev for the current was a function of the E_K; and (3) inward conductance increased with external K⁺ concentration. Extracellular Ba²⁺ block has been characterized as being voltage and concentration dependent and due to the binding of Ba²⁺ ions within the membrane voltage field of the channel pore (Quayle et al. 1993; Robertson et al. 1996). Due to the relative specificity of Ba²⁺ for inward rectifier channels over other K⁺ channels in smooth muscle (such as K_ATP channels or K_V channels) (Nelson & Quayle, 1995), Ba²⁺ has become a useful probe for investigating the functional role of K_ir currents in the regulation of arterial (myogenic) tone (McCarron & Halpern, 1990; Knot et al. 1996; Prior et al. 1998; Johnson et al. 1998). Thus, if K_ir2.1 is the molecular structure responsible for generating the strong, inwardly rectifying current in native vascular smooth muscle cells it should possess the above-mentioned characteristics when expressed in Xenopus oocytes.

Our studies demonstrate that expression of arterial K_ir2.1 vascular in Xenopus oocytes does indeed result in a strong, inwardly rectifying K⁺ current which is inhibited by micromolar concentrations of extracellular Ba²⁺. The half-block constant for Ba²⁺ at −60 mV in our experiments was 3.5 μM which compares well with those of 2.2 or 2.1 μM reported previously for inward rectifier currents from cerebral and coronary myocytes, respectively (Quayle et al. 1993; Robertson et al. 1996). As expected for an ion binding within the membrane voltage field (Woodhull, 1973), the K_d for Ba²⁺ was an exponential function of the voltage (Fig. 6D). The K_d decreased e-fold per 25 mV hyperpolarization, which is similar to that reported in myocytes from rat cerebral (e-fold per 23 mV; Quayle et al. 1993) and coronary (e-fold per 25 mV; Robertson et al. 1996) arteries. The K_d(0) for Ba²⁺ was determined to be 41.7 μM, which again is similar to values previously reported in native cerebral (K_d(0) = 38.1 μM; Quayle et al. 1993) and coronary (K_d(0) = 21.1 μM; Robertson et al. 1996) smooth muscle cells, and similar to the K_d(0) reported for murine K_ir2.1 (K_d(0) = 62 μM; Shieh et al. 1998). The slope of Ba²⁺ inhibition in Xenopus oocytes expressing K_ir2.1 is identical (0.51) to that measured previously in native coronary myocytes (0.51; Robertson et al. 1996), and similar to that reported for murine K_ir2.1 (μ= 0.54; Shieh et al. 1998), consistent with Ba²⁺ binding halfway down the channel pore.

In addition to block by Ba²⁺, inward rectifier currents in vascular smooth muscle cells have also been reported to be sensitive to external Cs⁺, Ca²⁺ and Mg²⁺ concentrations (Quayle et al. 1993; Robertson et al. 1996). External Cs⁺ inhibition of inward rectifier K⁺ currents in native vascular smooth muscle cells has been determined to be voltage dependent whereas Ca²⁺ and Mg²⁺ block were relatively voltage independent. In fact, the effect of Ca²⁺ and Mg²⁺ on K_ir2.1 channel properties has also not been examined. External Cs⁺, Ca²⁺ and Mg²⁺ inhibited K_ir2.1 currents in Xenopus oocytes (Figs 8 and 9). Inhibition by Cs⁺ was highly sensitive to voltage whereby block increased e-fold per 13 mV of membrane hyperpolarization. Also, the slope of Cs⁺ block in Xenopus oocytes expressing K_ir2.1 was 1.95 (Fig. 8), similar to that previously described (μ= 1.57) by Abrams and colleagues (1996) for murine K_ir2.1. These findings are consistent with those previously reported by Robertson et al. (1996) in coronary myocytes (Cs⁺ block increased e-fold per 17 mV and the slope was 1.51). In contrast to Cs⁺ block, external Ca²⁺ and Mg²⁺ block of cloned K_ir2.1 currents were was relatively unaffected by voltage (Fig. 9). External Ca²⁺ or Mg²⁺ (5 mM) reduced K_ir2.1 currents in Xenopus oocytes by 32 or 47.5 % (at −60 mV), while similar conditions have been reported to inhibit 47 or 41 % of inward rectifier current in native coronary myocytes. The lack of voltage dependence of Ca²⁺ and Mg²⁺ block suggests that the Ca²⁺ and Mg²⁺ binding sites are superficial. These results suggest that the cloned K_ir2.1 possesses the same low-affinity blocking site(s) for Ca²⁺ and Mg²⁺ as the strong, inwardly rectifying K⁺ currents present in native vascular myocytes.

Electrophysiological evidence that K_ir2.1 (and not K_ir2.2 or 2.3) encodes for the inward rectifier potassium channel in vascular smooth muscle

RT-PCR on isolated smooth muscle cells provided molecular evidence for K_ir2.1 in arterial smooth muscle (Figs 1 and 2). Transcripts for the other known members of the K_ir2 family (2.2 and 2.3) were not present (Fig. 1; Table 1). Barium, caesium, calcium and magnesium block of inward rectifier potassium currents in native arterial myocytes and the expressed K_ir2.1 currents in oocytes were quantitatively very similar (Table 1), consistent with the idea that K_ir2.1 encodes the inward rectifier in arterial myocytes. In further support of this idea, caesium blocked expressed K_ir2.2 and K_ir2.3 currents with at least 100-fold higher affinity than it blocked K_ir2.1 currents or native arterial smooth muscle currents (Morishige et al. 1994; Takahashi et al. 1994). In addition K_ir2.2 currents exhibited pronounced inactivation at negative voltages, unlike native inward rectifier currents, and K_ir2.1 and K_ir2.3 currents (Table 1). Our study also provides the first measurements of external calcium and magnesium block of cloned K_ir2.1 channels (Fig. 9). Barium block appears to be similar for all three subtypes, although it has not been quantitatively examined for K_ir2.2 and 2.3 currents. Therefore, the electrophysiological profile (barium, caesium, magnesium and calcium block as well as inactivation) of inward rectifier potassium currents in arterial myocytes is consistent with the properties of K_ir2.1 channels and does not correspond to either K_ir2.2 or 2.3 currents (Table 1).

In conclusion, the molecular biological results coupled with the electrophysiological and pharmacological studies performed on native cells and oocytes expressing K_ir2.1 channels suggest that the molecular structure of the channel responsible for generating strong inwardly rectifying potassium currents in vascular smooth muscle cells consists of K_ir2.1 subunits. With the molecular target now identified, future studies will focus on the relationship of transmitter regulation on K_ir2.1 expressed in Xenopus oocytes and K_ir current in native cells.