INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

POTASSIUM IONS ARE INVOLVED in many aspects of the function of the heart such as control of the resting membrane potential and repolarization of the action potential. TASK-1, [KCNK3; GenBank accession nos. AF006823 (human) and AF006824 (mouse)] is a member of the twin-pore domain K⁺ channel family, comprising four transmembrane segments and two pore-forming domains with intracellular termini. TASK-1 has the properties of a background conductance, instantaneous activation with voltage changes and a current-voltage relationship enabling the channel to function as an open K⁺-selective pore (5).

Several groups have established the presence of TASK-1 mRNA in the mammalian heart independent of species [human (14, 5), rat (10, 12), and mouse (5, 13)]. However, these studies contain conflicting evidence regarding TASK-1 mRNA expression levels within regions of the heart. TASK-1 mRNA distribution was first investigated in the mouse heart by Duprat et al. (5), who concluded from in situ hybridization that TASK-1 was present in atrial, but not ventricular, tissue. Kim et al. (9) reported the presence of TASK-1 mRNA in both rat atrial and ventricular myocytes, and further analysis showed that expression was at comparable levels (10). Conversely, Lopes et al. (13) reported TASK-1 mRNA in mice was predominantly located in ventricles, with much lower levels in the atrium. These variations of RNA expression levels may be due to the different probes used, or the limitations of sensitivity may differ in the experimental techniques.

This study investigated the localization of TASK-1 in the rat heart. The distribution of TASK-1 protein and mRNA in rat atria and ventricles was studied with Western blot, immunofluorescence, and RT-PCR. Additionally, colocalization of TASK-1 protein with the lectin wheat germ agglutinin (WGA) provides the first evidence for precise cellular localization of the TASK-1 K⁺ channel within rat atrial and ventricular myocytes.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Antibody specificity to TASK-1. The entire open reading frame of mouse TASK-1, encoding for 378 amino acids (Genbank accession no. AF006824) with the first amino acid E replaced by MKRQ, was subcloned into the plasmid pIRES-EYFP (Clontech, Oxford, UK). Transfection of the plasmid into a mammalian cell line of Chinese hamster ovary cells (CHO-K1) was performed with Lipofectamine 2000 reagent (Life Technologies, Paisley, UK). Cells were viewed 48 h after transfection. Cells were confirmed as being successfully transfected by expression of the marker yellow florescent protein (YFP) when viewed by confocal microscope (emission maximum 527 nm). Cells expressing YFP were further analyzed for TASK-1 cellular location by immunofluorescence (see Cellular location by immunofluorescence). Control CHO cells were also analyzed for TASK-1 protein: cells transfected with YFP only, cells transfected with no plasmid present, and cells not exposed to the transfection procedure.

Sample acquisition. Rats were humanely killed in accordance with Home Office guidelines, and the heart was removed, washed in PBS, and blotted dry. For immunofluorescence, single cardiac myocytes were isolated from the ventricles (7) and atria (16).

Analysis of protein expression. Dissected atria and ventricular tissue were snap-frozen, ground under liquid nitrogen, and homogenized in fresh buffer [in mM: 1 iodoacetamide, 1 benzathonium chloride, 5.7 PMSF, 10 EDTA, and 300 sucrose in 1% (wt/vol) SDS]. The homogenate was centrifuged at 10,000 rpm for 10 min at 4°C, the pellet was discarded, and the remainder was stored at 20°C before total protein determination by bicinchoninic acid assay (Sigma, Poole, UK). Samples (50 µg protein/lane) were separated by electrophoresis under reducing conditions by 10% SDS-PAGE, followed by transfer to a nitrocellulose membrane by the discontinuous blotting system (Pharmacia, Buckinghamshire, UK). The membrane was immersed in 5% (wt/vol) dried milk overnight at 4°C.

Cellular location by immunofluorescence. Single myocytes were plated into resin circles on polysine-coated slides and left to settle for 30 min. At room temperature, myocytes were fixed with 4% paraformaldehyde in PBS for 20 min, washed in PBS, and then subjected to 0.1% Triton X-100 for 20 min. Myocytes were stored in blocking solution [10% (vol/vol) serum in PBS] for 1 h. Anti-TASK-1 was applied to myocytes (1.2 µg/ml) in blocking solution and incubated overnight at 4°C, or, as a control, competitive inhibition of TASK-1 antibody was performed. Myocytes were washed, incubated for 1 h with the secondary antibody, swine anti-rabbit IgG conjugated to FITC (emission maximum 495 nm) or rhodamine (emission maximum 570 nm) (DAKO), washed again, and, to prevent photobleaching, mounted in Vectorshield (Vector, Burlingame, CA). All cellular myocyte membranes were stained by application of WGA conjugated to rhodamine (the lectin binds to N-acetylglucosamine within membranes; Vector) 2 h before mounting. Slides were stored in the dark at 4°C before examination by laser scanning confocal microscopy (Leica, Milton Keynes, UK). Images were taken at approximately midcell depth. Wavelengths were individually collected for each optical slice, and a colocalized image was produced by superimposing each wavelength. Single optical images were further analyzed by Scion image (Scion; National Institutes of Health, Bethesda, MD).

Detection of mRNA expression. Total RNA was extracted from dissected atria and ventricles by the Gen Elute extraction kit (Sigma) and eluted into water. Reverse transcription (RT) was performed on 1 µg of total RNA per sample according to the manufacturer's instructions (Life Technologies). RT was also performed in the absence of reverse transcriptase as a control for genomic DNA contamination. The RT material was subjected to PCR by using 10 pmol of each primer, 20 mM Tris, pH 8.8, 10 mM KCl, 10 mM (NH₄)₂SO₄, 2 mM MgSO₄, 0.1% Triton X-100, 100 µg/ml bovine serum albumin, and 200 nM dNTPs. During a "hot start," 2.5 units of Taq DNA polymerase was added, and 30 cycles of PCR were then performed (30 s at 94°C, 45 s at 60°C, 30 s at 72°C) in a Geneamp 2400 cycler (PE Biosystems, Foster City, CA). Two sets of specific primer pairs were used: rat TASK-1 5'-ACGATGAAGCGGCAGAATGTG-3' (sense) with 5'-ACGAAACCGATGAGCCATG-3' (antisense) and -actin 5'-TTGTAACCAACTGGGACGATATGGG-3' (sense) with 5'-GATCTTGATCTTCATGGTGCTAGG-3'(antisense). PCR products were analyzed by electrophoresis on a 1% agarose-Tris-disodium ethylenediamine gel, stained with ethidium bromide, and visualized by ultraviolet illumination.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Antibody specificity to TASK-1. CHO cells were confirmed as transfected with the plasmid-containing mouse TASK-1 by the expression of the marker YFP. Viewed by confocal microscope, the cells expressing YFP showed the marker to be located within the cytoplasm and increasing toward the outer cell membrane (Fig. 1A) compared with no expression in control cells (cells transfected with no plasmid present; Fig. 1B). On further analysis of the YFP-positive cells for TASK-1 protein expression, TASK-1 staining was observed at the outer cell membranes only (Fig. 1A). To demonstrate that both proteins were present but in different locations, YFP and TASK-1 images were overlaid with minimal colocalization (Fig. 1A). Therefore, these data confirmed that anti-TASK-1 antibody was specific to the TASK-1 protein.

View larger version (33K): [in this window] [in a new window]   Fig. 1.   Demonstration of antibody specificity to TASK-1. A: Chinese hamster ovary cells (CHO-K1) were transfected with a plasmid containing the marker yellow florescent protein (YFP) and mouse TASK-1. Transfected CHO cells expressing TASK-1 protein were stained with anti-TASK-1 (+), and transfection was confirmed by expression of the YFP marker. Points of TASK-1 and YFP colocalization are indicated in yellow (+/YFP). Bar, 10 µm. B: as control experiments CHO-K1 cells were subjected to the transfection procedure with no plasmid present, and these cells were then stained with anti-TASK-1 (+) and examined for YFP expression. Points of TASK-1 and YFP colocalization are indicated in yellow (+/YFP). Bar, 10 µm. C: TASK-1 protein expression was confirmed in rat atrial and ventricular tissue by Western blot (all bands detected are shown). The same samples exhibited no bands when the anti-TASK-1 antibody was competitively inhibited with its antigenic peptide.

Identification of TASK-1 protein in rat heart. Expression of TASK-1 protein in whole rat atria and ventricular tissue was confirmed by Western blot, which showed the presence of two bands per sample at the expected molecular weights (Fig. 1B). Bands in both samples were eliminated by competitive inhibition, and this also confirmed the specificity of the TASK-1 antibody (Fig. 1B).

Location of TASK-1 protein in rat heart. TASK-1 protein was stained by immunofluorescence in rat ventricular and atrial myocytes. Confocal microscopy showed that ventricular myocytes exhibited strong staining at the intercalated disks with transversely oriented striated fluorescence through the cell interior (Fig. 2A), whereas at higher resolution punctate spots were shown in an orderly manner, consistent with striations across the myocyte (Fig. 2B). In contrast, single rat atrial myocytes were intensely stained at the intercalated disks with a weak punctate, striated pattern (Fig. 2C). As a control, myocytes were subjected to competitive inhibition; under these conditions no staining occurred (Fig. 2, A and C). WGA stained all cellular membranes. The ventricular myocyte was stained at the outer membrane and across in a striated pattern as associated with the T tubule network (Fig. 2, A and B). The atrial myocyte was stained at the outer cell membrane and nuclear envelope (Fig. 2C).

View larger version (33K): [in this window] [in a new window]   Fig. 2.   Expression of TASK-1 protein in rat myocytes. Immunofluorescence images from single ventricular (A and B) and atrial (C) myocytes were stained with anti-TASK-1 (+), eliminated by competitive inhibition with the antigenic peptide (), and costained with wheat germ agglutinin (WGA). Colocalization of TASK-1 and WGA is indicated in yellow (+/WGA). B: high-resolution ventricular myocyte images. D: single ventricular myocytes treated with formamide to induce detubulation were observed labeled with TASK-1 (+) and WGA, and areas of colocalization are indicated in yellow (+/WGA). Bar, 15 µm.

Presence of TASK-1 mRNA in rat heart. Total RNA from atrial, ventricular, and whole heart tissue was subjected to RT-PCR with two TASK-1-specific primers that yielded a PCR product of the expected size of 500 bp (Fig. 3). The identity of this PCR product was confirmed as rat TASK-1 by DNA sequencing (data not shown). No PCR products were generated in the absence of reverse transcriptase, indicating that the PCR products were not the result of genomic DNA contamination (Fig. 3). Furthermore, TASK-1-specific primers were designed to span an intron to differentiate between mRNA and genomic-derived products. -Actin mRNA was detected in all tissues at 640 bp (Fig. 3).

View larger version (21K): [in this window] [in a new window]   Fig. 3.   Detection of TASK-1 mRNA in rat myocytes. RT-PCR experiments were performed in the presence (+) or absence () of reverse transcriptase on total RNA extracted from atrial, ventricular, and whole heart tissue. A single specific RT-PCR product of 500 bp for TASK-1 and 640 bp for endogenous -actin was found in each tissue sample.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Previously, TASK-1 was reported to be abundant in a broad spectrum of tissues irrespective of species (2). For example, TASK-1 mRNA was found to be highly expressed in the whole rodent heart compared with lower levels in rat lung and brain (9, 10, 13). No attention has been paid to the cellular location of the TASK-1 protein in the mammalian heart.

First, the specificity of the antibody to TASK-1 protein was confirmed by staining of CHO cells transfected with a plasmid expressing TASK-1 protein. Our Western blot data further demonstrated that the antibody was selective for TASK-1, by elimination of the positive bands with the competitive peptide, and that whole tissue from atria and ventricles of the rat was positive for expression of the TASK-1 protein. The anti-TASK-1 antibody used in this study has been demonstrated to be selective for TASK-1 by other authors; for example, Millar et al. (15) used anti-TASK-1 antibody to show TASK-1 protein within the plasma membrane of cerebellar granule neurons (13, 15).

By immunofluorescence, single ventricular myocytes were shown to have intense staining of TASK-1 protein at the intercalated disks, where extensive folding of the plasma membrane occurs, and a striated pattern across the cell that corresponded to the T tubule network. Removal of the T tubule network by treatment of the cell with formamide induced the disappearance of striated TASK-1 protein staining, but the intercalated disk remained stained. This confirmed the intracellular location of TASK-1. Atrial myocytes showed that TASK-1 was stained at the intercalated disks and displayed punctate spots across the cell. The exhibited punctate pattern is attributed to the underdeveloped T tubule network (6); hence, staining was weak compared with the ventricular striated pattern. For both ventricular and atrial myocytes, the location of TASK-1 protein within the rat myocyte membranes was supported by colocalization with WGA. Costaining of the T tubule network occurred only in ventricular cells (17). We have also demonstrated that TASK-1 mRNA is present in rat atrial and ventricular tissue.

Expression of TASK-1 in Xenopus oocytes results in a large outward, noninactivating K⁺ current, open at rest and at all membrane potentials (5, 12). This current lacks voltage- or time dependence and therefore results in TASK-1 being described as a "background" channel. Considering its pharmacology, TASK-1 is insensitive to the classic K⁺ channel blockers but inhibited by certain divalent ions such as zinc (12) and barium (1) and acutely sensitive to changes in extracellular pH. A background current known as I_kp has been described in guinea pig ventricular myocytes as having a role in determining action potential duration (1). The molecular correlate of this current has yet to be identified; a potential candidate is TASK-1. The differences between I_kp and TASK-1 currents are open probability and barium block, but these differences may be explained by species differences of homologous K⁺ channels (13).

TASK-1 in neurons has been demonstrated to be regulated by G protein-coupled receptors such as the muscarinic receptor (15). Hence, acetylcholine could act via receptors through this channel, altering the resting membrane potential and cell excitability. Therefore, the extensive distribution of TASK-1 expression in cardiac tissue may provide a potential future target of antiarrhythmic drugs, particularly those designed to prevent arrhythmias by prolonging the action potential duration.