Two-pore domain K+ channels are an emerging family of K+ channels that may contribute to setting membrane potential in both electrically excitable and non-excitable cells and, as such, influence cellular function. The human uteroplacental unit contains both excitable (e.g. myometrial) and non-excitable cells, whose function depends upon the activity of K+ channels. We have therefore investigated the expression of two members of this family, TWIK (two-pore domain weak inward rectifying K+ channel)-related acid-sensitive K+ channel (TASK) and TWIK-related K+ channel (TREK) in human myometrium. Using RT-PCR the mRNA expression of TASK and TREK isoforms was examined in myometrial tissue from pregnant women. mRNAs encoding TASK1, 4 and 5 and TREK1 were detected whereas weak or no signals were observed for TASK2, TASK3 and TREK2. Western blotting for TASK1 gave two bands of approximately 44 and 65 kDa, whereas TREK1 gave bands of approximately 59 and 90 kDa in myometrium from pregnant women. TASK1 and TREK1 immunofluorescence was prominent in intracellular and plasmalemmal locations within myometrial cells. Therefore, we conclude that the human myometrium is a site of expression for the two-pore domain K+ channel proteins TASK1 and TREK1.
Two-pore domain K+ (K2P) channels are an emerging family of K+ channels. They are different from other classic K+ channels in that they possess two pore-forming domains and exhibit different pharmacological properties (Patel & Honoré 2001). A key biophysical feature of these channels is their relative voltage-insensitivity and, as such, they are thought to contribute to resting membrane potential and have been termed leak or background channels. However, it is becoming more evident that these channels have diverse mechanisms of modulation. Factors influencing these channels include extracellular pH (Duprat et al. 1997), membrane stretch, arachidonic acid (Bang et al. 2000, Miller et al. 2003) and oxygen tension (hypoxia; Lewis et al. 2001, Miller et al. 2003).
Since the cloning of the first member of the K2P channel family, TWIK1 (two-pore domain weak inward rectifying K+ channel) (Lesage et al. 1996), numerous other members have been described including TASK (TWIK-related acid-sensitive K+ channels)1–5 and TREK (TWIK-related K+ channels)1 and 2. TASK channels have been localised to both electrically excitable and non-excitable tissues (e.g. brain, pancreas, lung and kidney; reviewed in O’Connell et al. 2002). TREK channels have a predominantly neural location but their mRNA has also been found in small intestine, kidney and pancreas (Medhurst et al. 2001). Functionally, TASK channels are exquisitely sensitive to perturbations in extracellular pH: acidosis inhibits TASK currents whereas alkalosis increases the current (Duprat et al. 1997). TREK channels, on the other hand, are modulated by membrane stretch and arachidonic acid (Bang et al. 2000, Miller et al. 2003). One common regulatory mechanism, described for TASK1, TASK3 and TREK1 is their inhibition by hypoxia (Hartness et al. 2001, Lewis et al. 2001, Miller et al. 2003).
The K+ channel superfamily contains over 100 different subunits/proteins (Coetzee et al. 1999) yet current understanding of uterine K+ channel expression is very limited. Large-conductance Ca2+ -activated K+ channel mRNA, protein and function has been documented (Khan et al. 2001). Voltage-gated K+ channel currents have been studied, although there is scant evidence regarding molecular identities (Knock et al. 1999). Alternatively, KATP mRNA expression has been studied during gestation (Curley et al. 2002) but little functional evidence documented. Clearly, there is a need for a more complete analysis of K+ channel expression in uterine tissue.
To this end, we decided as an initial step to study the expression of K2P channels in pregnant human myometrium. This would give important information on the expression of these channels and, if present, offer insights into their possible roles in determining uterine excitability. Although there are likely to be many K+ channel subtypes responsive to pH, the K2P channels, as mentioned above, do have a uniqueness as a K+ channel sub-family in their reported sensitivities to extracellular pH, experimental hypoxia and stretch – all stimuli that influence animal and human myometrial contractility (Taggart et al. 1997, Bugg et al. 2002a, Monir-Bishty et al. 2003). For both these reasons, we focused our attention on examining the expression and localisation of TASK and TREK in human myometrium.
Materials and Methods
Myometrium was taken from women undergoing elective Caesarean section at term after informed written consent and following local ethical committee approval. Expression studies were performed on myometrial biopsies taken from the mid-upper anterior lip. Samples were snap-frozen in liquid nitrogen and stored at −80 °C until required.
Total RNA was isolated from homogenised myometrium using Trizol reagent (Invitrogen). RNA was reverse transcribed using Moloney murine leukaemia virus reverse transcriptase (RT) according to the manufacturer’s instructions (Invitrogen) in a Perkin-Elmer Cetus thermal cycler (Perkin-Elmer, Beaconsfield, South Bucks, UK). An initial RT-PCR, using β-actin, in which the RT was omitted was performed to ensure that all RNA samples were not contaminated with genomic DNA. Subsequent RT-PCR, using standard techniques, with a hot start was employed using primer pairs already utilised in other studies or designed by us: TASK1, annealing temperature (Ta) 59 °C (Hamilton et al. 2000); TASK2, Ta 62 °C (Reyes et al. 1998); TASK3 5′-cggatcaaggggaagtacaa-3′ (forward) with 5′-tgcccataacc-tatggtggt-3′ (reverse), Ta 60 °C; TASK4 5′-gcatctg-gaaccttctgctc-3′ (forward) with 5′-atagctgcacccagcctcta-3′ (reverse), Ta 62 °C; TASK5, Ta 62 °C (Kim & Gnatenco 2001); TREK1 and 2, Ta 55 and 56 °C respectively (Lesage et al. 2000). BLAST searches were performed to ensure primers had no homology with any other known gene products. The number of cycles was 35 for each primer pair. An appropriate positive (human kidney and brain; Becton Dickinson Bioscience, Oxford, UK) and negative control (water replacing the template) were used. β-Actin was routinely amplified from all samples, confirming sample integrity and amplification capacity. PCR products were purified and subsequently sequenced using in-house facilities to confirm identity. The sequenced products had 98% (TASK1, 2, 3), 99% (TASK5) and 100% (TASK4 and TREK1) homology with the known published sequences when compared using BLAST (not shown).
Myometrial tissues, different from those used for RNA extraction, were homogenised on ice in homogenisation buffer (in mol/l): 0.01 Hepes, 0.001 EDTA, 0.25 sucrose (pH 7.4) with an anti-protease inhibitor cocktail (4-(2-aminoethyl)benzenesulphonyl fluoride, pepstatin A, E-64, bestatin, leupeptin and aprotinin; Sigma-Aldrich). Rat brain was used as a positive control for TASK and TREK expression. Barrier bred male Wistar rats where housed at BSU, University of Manchester and Schedule I killed (Stunning followed by cervical dislocation) according to UK Home Office guidelines. We routinely used the post-nuclear supernatant, obtained after centrifugation at 4000 g for 10 min, for our blotting experiments.
In a subset of experiments, we examined the sub-cellular distribution of TREK1 protein. After homogenisation of myometrial tissue, an initial centrifugation at 1000 g for 10 min was performed to yield a pellet predominately containing nuclear protein. The resulting supernatant was further centrifuged at 100 000 g for 1 h, after which a crude membrane pellet and supernatant with cytosolic protein were obtained. Both the nuclear and membrane pellets were resuspended in homogenisation buffer. All sample protein concentrations were determined using a commercial protein assay kit (Bio-Rad). Samples were stored at −80 °C.
Protein from myometrium (elective Caesarean section, 50 or 100 μg as indicated) and rat brain (50 μg) was mixed with a reducing, non-boiling loading buffer containing 8 mol/l urea, 5% SDS (w/v), 0.04% bromophenol blue (v/v) and 0.455 mol/l dithiothreitol, in 0.05 mol/l Tris–HCl (pH 6.9). Proteins were electrophorectically separated in 10% polyacrylamide gels and transferred to PVDF membranes. The membranes were blocked for 1 h using blocking buffer (3% dried milk powder (w/v) in 0.05% Tween 20 (v/v), Tris–buffered saline (TBS (in mol/l): 0.015 Tris, 0.150 NaCl; pH 8.0)). The membrane was probed for 2 h at room temperature with anti-TASK1 or anti-TREK1 antibody (Alomone Labs, Jerusalem, Israel) at a 1:100 or 1:200 dilutions respectively in blocking buffer. TASK1 rabbit polyclonal antibody was raised against the peptide (C-terminus) corresponding to residues 252–269 of the human TASK1 channel. TREK1 rabbit polyclonal antibody was raised against residues 8–28 of human TREK1 channel. Three washes in TBS/0.05% Tween 20 were followed by incubation with a horseradish peroxidase-conjugated goat anti-rabbit secondary antibody at a 1:2000 dilution (DAKO, Ely, Cambs, UK) for 1 h. After washing in TBS, membranes were developed by enhanced chemiluminescence (Amersham Pharmacia Biotech). Appropriate negative and pre-absorption controls were also performed. These controls consisted of omission of the primary antibody and pre-absorption of the primary antibody for 1 h at ambient room temperature with a 3-fold excess of the epitope to which it was raised.
Frozen myometrial tissue sections (4 μm thick, sampled randomly from three separate biopsies) were air-dried and fixed with 2% paraformaldehyde (v/v), 4% sucrose (w/v) in PBS (in mol/l: 0.137 NaCl, 0.0027 KCl, 0.0081 Na2HPO4, 0.0019 NaH2PO4; pH 7.4) for 10 min, then permeabilised with 0.3% Triton X-100 (v/v; Sigma) in PBS for 10 min. Non-specific binding sites were blocked with 4% fetal calf serum (v/v; FCS), 0.1% Tween 20 (v/v) in PBS for 30 min. Anti-TASK1 or anti-TREK1 was applied at a 1:100 dilution for 1 h at room temperature, then washed in 4% FCS in PBS and incubated further for 30 min at room temperature with a 1:50 dilution of the secondary antibody (DAKO), which was FITC-labelled anti-rabbit IgG. The sections were washed with PBS and mounted with Vectashield (Vector Labs, Peterborough, Cambs, UK) containing propidium iodide. Sections were then viewed using confocal laser-scanning microscopy (Bio-Rad Radiance 2000). The localisation of TASK1 and TREK1 within myometrial tissue was determined using an × 40 oil immersion objective. In negative control and pre-absorption experiments, an equivalent volume of PBS replaced the primary antibody or the primary antibody was pre-absorbed for 1 h at ambient room temperature with a 3-fold excess of the epitope to which it was raised.
Figure 1A demonstrates that the amplification products from control tissues (human kidney or human brain) using gene-specific primers for TASK1-5 or TREK1-2 appear as single amplicons at the predicted molecular size, validating the primers used in this study. These primers were then used to screen myometrial biopsies taken from four individual women (Fig. 1B). Detectable signals, which co-migrated with those in the positive controls, were observed for TASK1, TASK4, TASK5 and TREK1, weaker and more variable signals were observed for TASK2 and TASK3. TREK2 mRNA was undetected.
The presence of mRNA, however, does not necessarily mean that translation into protein occurs. To examine this further, the expression of TASK1 and TREK1 at the protein level was studied, selected on the basis that TASK1 is the main K2P channel reported to be present by both Western blotting and immunolocalisation in other tissues (see Discussion) and that the message for TREK1 was qualitatively similar to TASK1. Using anti-TASK1, a single band of approximately 55 kDa was observed in rat brain (Fig. 2A). However, in myometrium from individual pregnant women we observed two bands of ~ 44 kDa and 65 kDa (Fig. 2A). Using anti-TREK1, two bands of ~ 60 and 90 kDa were observed (Fig. 2D) in myometrium and a single band of ~ 60 kDa in rat brain. In negative control and competition experiments, when either the primary antibody was omitted or pre-absorbed with the epitope against which it was raised, no signals were observed (Fig. 2B and E respectively). Comparable signals were obtained for β-actin in all samples confirming equal protein loading between samples (Fig. 2C and F). In a separate series of experiments, we examined the sub-cellular distribution of TREK1 protein (Fig. 2G). As can be seen, TREK1 was localised to all three cellular fractions, with a more predominate signal observed in the cytosol.
We decided to examine sites of possible expression using immunostaining, the results of which are shown in Fig. 3. Figure 3A shows the pattern of TASK1 staining observed in pregnant myometrium. TASK1 was observed throughout plasmalemmal and intracellular locations. TREK1 also exhibited staining consistent with an intra-cellular location, as evidenced by a near-nuclear distribution (Fig. 3E). In similar negative control and competition experiments to those described for Western blotting, when the primary antibody was omitted (Fig. 3D, TASK1 shown) or pre-absorbed with a 3-fold excess of the epitope (Fig. 3F, TREK1 shown), no staining was observed.
We have investigated the expression of TASK1-5 and TREK1 and TREK2 within human myometrium. mRNA for TASK1, 4 and 5 and TREK1 were observed. At the level of protein, TASK1 was detected in myometrium, as was TREK1. Immunocytochemistry revealed a relatively uniform plasmalemmal and intracellular distribution of TASK1 and TREK1 in myometrial cells.
Using RT-PCR and Northern blotting, both TASK1 and TASK2 have been shown to be relatively weakly expressed in whole mouse and human uterus (Reyes et al. 1998, Medhurst et al. 2001), as has TASK3 (Medhurst et al. 2001). The screening of myometrial strips (Fig. 1B) gave us consistently detectable signals for TASK1, 4 and 5, with mRNA for the latter two channels being identified in the human myometrium for the first time. TREK1 mRNA has been shown to be expressed in human uterus (Reyes et al. 1998, Meadows et al. 2000) and our data suggest TREK1 expression is at the level of the myometrial cells and we document, for the first time, the lack of a detected signal for TREK2 in human myometrium, suggesting that myometrium is unlikely to be a significant site of expression of TREK2.
Western blotting for TASK1 revealed the presence of two bands, one of 65 kDa and another of 44 kDa in myometrium (Fig. 2A). Other studies in various tissues using this commercial antibody, whose epitope is conserved between human, rat and mouse, have documented bands of differing sizes. Lopes et al.(2000) found a single band of approximately 55–60 kDa in mouse heart. In rat heart, bands of 65 and 130 kDa have been observed (Jones et al. 2002), as has a band of 45 kDa (Rajan et al. 2002). In rat brain, TASK1 has been reported to be 50 kDa (Rajan et al. 2002). Clearly variability exists in the reported molecular size of TASK1. A core molecular mass for rat TASK1 has been suggested to be 45 kDa (Leonoudakis et al. 1998).
There are also limited Western blotting and immunostaining data for TREK1. Using an antibody corresponding to residues 5–21 of human TREK1, Hervieu et al.(2001) demonstrated the presence of bands that were 56 kDa and, more predominately, 112 kDa in size. In cells overexpressing TREK1, Maingret et al.(2000) used a polyclonal antibody directed against residues 1–44 and 71–114 to visualise a band of 84 kDa under non-reducing conditions that subsequently migrated to an apparent molecular mass of 45 kDa under reducing conditions (β-mercaptoethanol). However, Western blotting of murine and rat brain revealed a single band of ~ 92 kDa. This antibody recognised only a single band of ~ 45 kDa when subsequently used to blot rat heart microsomes (Terrenoire et al. 2001). A core size for TREK1 has been estimated to be 44 kDa (Hervieu et al. 2001). The bands we document here are ~ 60 and 90 kDa in pregnant human myometrium and ~ 60 kDa in rat brain. These bands are comparable to those already reported in other tissues. Although we detected messages for TASK4 and 5 in human myometrium, there are as yet no antibodies available to TASK4 or TASK5, therefore we could not test for protein expression.
The origins of different sized molecular mass bands for TASK1 and TREK1 are unknown, but may reflect different states of post-translational modification, e.g. there are putative glycosylation sites on TASK1 (Duprat et al. 1997, Leonoudakis et al. 1998) and TREK1 (Fink et al. 1996, Meadows et al. 2000). The 44 kDa band in myometrium could represent an unprocessed form of core TASK1 protein with the 65 kDa band reflecting a fully post-translationally modified protein (Fig. 2A). A similar situation may exist for TREK1. However, the higher molecular mass band might represent some form of dimerisation of TASK or TREK channels. Collectively, these observations imply differential TASK1 and TREK1 processing in different tissues, both within the same species and between different species.
To date there are no reports on immunostaining of TASK1 in the human myometrium. Our novel results (Fig. 3A) show TASK1 in myometrial cells with both a plasmalemmal and predominantly intracellular non-nuclear location. Staining for TREK1 (Fig. 3E) also exhibited plasmalemmal and intracellular staining, the latter evidenced as a prominent near-nuclear location. Taken together, the immunofluorescence and Western blotting data on the sub-cellular localisation of TREK1 (Fig. 2G) question a role for TREK in the determination of membrane potential at late term in myometrium.
The data we describe within the current study, therefore, provide a platform on which to base future functional studies that will address the role of TASK and TREK in the control of uterine excitability and contractility. This presents a considerable challenge in freshly isolated, fully differentiated human myometrial cells due to the absence of specific pharmacological inhibitors of these channel subtypes. The putative TASK1 channel blocker anandamide has been reported to inhibit K2P-like currents in other smooth muscle cells by ~ 25% (Gurney et al. 2003).
As TASK1, at least in other tissues, is a functional K+ channel, a plasma membrane localisation in myometrial cells may have an impact upon cellular function. Hypoxia or metabolic inhibition has been shown to reduce human myometrial contractility (Bugg et al. 2002b, Monir-Bishty et al. 2003). As myometrial contractility is in part dependent upon electrical excitability and membrane potential, the presence of ion channels that are sensitive to hypoxia and pH may influence contraction. As TASK1 (Lewis et al. 2001) and TASK3 (Hartness et al. 2001) are sensitive to hypoxia and pH, this makes them attractive candidates for roles in the control of myometrial function. A role for TREK1 in myometrial contractility is, at this moment, harder to define. Our data suggest a predominantly non-plasmalemmal location for TREK1 protein in term tissue, thus making it difficult to predict an impact upon myometrial contractility. However, the intracellular localisation might represent a functional down-regulation. As the myometrium becomes distended during pregnancy, the presence of a stretch-modulated K+ channel such as TREK1 (Bang et al. 2000, Miller et al. 2003) may contribute to the mechanisms by which the cell membrane remains hyperpolarised, dampening tissue excitability. Functional down-regulation of stretch-sensitive K+ channels by intra-cellular redistribution may result in the setting of a membrane potential that favours contractile activity.
Received 28 July 2004 First decision 21 September 2004 Revised manuscript received 17 December 2004 Accepted 14 January 2005
The authors would like to acknowledge Maureen O’Hara for technical assistance, Dr Peter March for assistance with confocal microscopy and the efforts of midwives and clinicians in recruiting patients. This work was supported by Tommy’s, the baby charity, and the Wellcome Trust (Grant no. 060997/Z/00). The authors declare that there is no conflict of interest that would affect the impartiality of this scientific work.
Bang HKim Y & Kim D2000 TREK-2 a new member of the mechanosensitive tandem-pore K+ channel family. Journal of Biological Chemistry27517412–17419.
Bugg GBaker PN & Taggart MJ2002a Altered length-dependent contractile activation of isolated human myometria with gestation. Journal of the Society for Gynecologic Investigation9205A.
Bugg GBaker PN & Taggart MJ2002b Influence of hypoxia on myometrial contractility from pregnant and non-pregnant women. Journal of the Society for Gynecologic Investigation9302A.
Coetzee WAAmarillo YChiu JChow ALau DMcCormack Tet al.1999 Molecular diversity of K+ channels. Annals of the New York Academy of Sciences868233–285.
Curley MCairns MTFriel AMMcMeel OMMorrison JJ & Smith TJ2002 Expression of mRNA transcripts for ATP-sensitive potassium channels in human myometrium. Molecular Human Reproduction8941–945.
Duprat FLesage FFink MReyes RHeurteaux C & Lazdunski M1997 TASK a human background K+ channel to sense external pH variations near physiological pH. EMBO Journal165464–5471.
Fink MDuprat FLesage FReyes RRomey GHeuteaux Cet al.1996 Cloning functional expression and brain localization of a novel unconventional outward rectifier K+ channel. EMBO Journal166854–6862.
Gurney AMOsipenko ONMacMillan DMcFarlane KMTate RJ & Kempsill FE2003 Two-pore domain K channel TASK-1 in pulmonary artery smooth muscle cells. Circulation Research93957–964.
Hamilton KClarson LHGlazier JDSibley CP & Greenwood SL2000 Expression of potassium channel mRNA in human placental trophoblast and microvascular endothelial cells. Journal of the Society for Gynecologic Investigation7342A.
Hartness MELewis ASearle GJO’Kelly IPeers C & Kemp PJ2001 Combined antisense and pharmacological approaches implicate hTASK as an airway O2 sensing K+ channel. Journal of Biological Chemistry27626499–26508.
Hervieu GJCluderay EGray CWGreen PJRanson JLRandall ADet al.2001 Distribution and expression of TREK-1 a two-pore-domain potassium channel in the adult rat CNS. Neuroscience103899–919.
Jones SAMorton MJHunter M & Boyett MR2002 Expression of TASK-1 a pH-sensitive twin-pore domain K+ channel in rat myocytes. American Journal of Physiology283H181–H185.
Knock GASmirnov SV & Aaronson PI1999 Voltage-gated K+ currents in freshly isolated myocytes of the pregnant human myometrium. Journal of Physiology518769–781.
Leonoudakis DGray ATWinegar BDKindler CHHarada MTaylor DMet al.1998 An open rectifier potassium channel with two pore domains in tandem cloned from rat cerebellum. Journal of Neuroscience18868–877.
Lesage FGuillemare EFink MDuprat FLazdunski MRomey Get al.1996 TWIK-1 a ubiquitous human weakly inward rectifying K+ channel with a novel structure. EMBO Journal151004–1011.
Lesage FTerrenoire CRomey G & Lazdunski M2000 Human TREK2 a 2P domain mechano-sensitive K+ channel with multiple regulations by polyunsaturated fatty acids lysophospholipids and Gs Gi and Gq protein-coupled receptors. Journal of Biological Chemistry27528398–28405.
Lewis AHartness MEChapman CGFearon IMMeadows HJPeers Cet al.2001 Recombinant hTASK1 is an O2-sensitive K+ channel. Biochemical and Biophysical Research Communications2851290–1294.
Lopes CMBGallagher PGBuck MEButler MH & Goldstein SA2000 Proton block and voltage gating are potassium-dependent in the cardiac leak channel Kcnk3. Journal of Biological Chemistry27516969–16978.
Maingret FLauritzen IPatel AJHeurteaux CReyes RLesage Fet al.2000 TREK-1 is a heat activated background K+ channel. EMBO Journal192483–2491.
Meadows HJBenham CDCairns WGloger IJennings CMedhurst ADet al.2000 Cloning localisation and functional expression of the human orthologue of the TREK-1 potassium channel. Pflügers Archiv439714–722.
Medhurst ADRennie GChapman CGMeadows HDuckworth MDKelsell REet al.2001 Distribution analysis of human two pore domain potassium channels in tissues of the central nervous system and periphery. Molecular Brain Research86101–114.
Miller PKemp PJLewis AChapman CGMeadows HJ & Peers C2003 Acute hypoxia occludes hTREK-1 modulation: re-evaluation of the potential role of tandem P domain K+ channels in central neuroprotection. Journal of Physiology54831–37.
Monir-Bishty EPierce SJKupittayanant SShmygol A & Wray S2003 The effects of metabolic inhibition on intracellular calcium and contractility of human myometrium. British Journal of Obstetrics and Gynaecology1101050–1056.
Rajan SPreisig-Muller RWischmeyer ENehring RHanley PJRenigunta Vet al.2002 Interaction with 14-13-3 proteins promotes functional expression of the potassium channels TASK-1 and TASK-3. Journal of Physiology54513–26.
Reyes RDuprat FLesage FFink MSalinas MRaman Net al.1998 Cloning and expression of a novel pH-sensitive two pore domain K+ channel from human kidney. Journal of Biological Chemistry27330863–30869.
Taggart MJSheader EAWalker SDNaderali EKMoore S & Wray S1997 External alkalinization decreases intracellular Ca2+ and spontaneous contractions in pregnant rat myometrium. American Journal of Obstetrics and Gynecology177959–963.
Terrenoire CLauritzen ILesage FRomey G & Lazdunski M2001 A TREK-1-like potassium channel in atrial cells inhibited by β-adrenergic stimulation and activated by volatile anesthetics. Circulation Research89336–342.