Potassium currents are plastic entities that modify electrical activity of the heart in various physiological conditions including chronic thermal stress. We examined the molecular basis of the inward rectifier K+ current (IK1) in rainbow trout acclimated to cold (4°C, CA) and warm (18°C, WA) temperature. Inward rectifier K+ channel (Kir)2.1 and Kir2.2 transcripts were expressed in atrium and ventricle of the trout heart, Kir2.1 being the major component in both cardiac chambers. The relative expression of Kir2.2 was, however, higher (P < 0.05) in atrium than ventricle. The density of ventricular IK1 was ∼25% larger (P < 0.05) in WA than CA trout. Furthermore, the IK1 of the WA trout was 10 times more sensitive to Ba2+ (IC50 0.18 ± 0.42 μM) than the IK1 of the CA trout (1.17 ± 0.44 μM) (P < 0.05), and opening kinetics of single Kir2 channels was slower in WA than CA trout (P < 0.05). When expressed in COS-1 cells, the homomeric Kir2.2 channels demonstrated higher Ba2+ sensitivity (2.88 ± 0.42 μM) than Kir2.1 channels (24.99 ± 7.40 μM) (P < 0.05). In light of the different Ba2+ sensitivities of rainbow trout (om)Kir2.1 and omKir2.2 channels, it is concluded that warm acclimation increases either number or activity of the omKir2.2 channels in trout ventricular myocytes. The functional changes in IK1 are independent of omKir2 transcript levels, which remained unaltered by thermal acclimation. Collectively, these findings suggest that thermal acclimation modifies functional properties and subunit composition of the trout Kir2 channels, which may be needed for regulation of cardiac excitability at variable temperatures.
- inward rectifier potassium channels
- atrial myocytes
- ventricular myocytes
- thermal plasticity
strong inward rectifier potassium (Kir) channels conduct inward currents at membrane potentials negative to the K+ reversal potential but permit only limited K+ efflux at more positive voltages (18, 37) based on the voltage-dependent block of the channels by intracellular Mg2+ and polyamines (8, 9, 17). The small outward current is physiologically important, since it sets resting membrane potential (RMP), controls excitability, and participates in diverse body functions in various organs. In the heart, the inward rectifier current (IK1) clamps the RMP close to K+ equilibrium potential and contributes to the late phase 3 repolarization of the action potential (AP) and thereby participates in the regulation of AP duration (18).
On the basis of sequence homology, inward rectifier K+ channels have been classified into seven subfamilies, Kir1–Kir7 (5, 18). Inward rectifiers of the mammalian heart are homo- or heterotetrameric assemblies of Kir2.1–3 subunits (16, 26, 31, 36, 40) with substantial variation between species (4, 42). Further complexity is generated by chamber-related differences in the expression of Kir2 subunits (11, 14, 36). It is, however, incompletely understood to what extent species-specific and regional differences in the properties of cardiac IK1 are related to the relative expression of Kir2 subunits. For example, the atrioventricular differences in density and rectification of IK1 have been explained either by differences in free polyamine concentrations (39) or by variable expression of functionally different Kir2 subunits (4). Therefore, situations in which the cardiac IK1 is substantially modified within species by external or internal factors might shed new light on the relative importance of Kir subunits in the regulation of the cardiac IK1.
Body temperature of ectothermic animals can vary quite considerably, e.g., because of seasonal changes in ambient temperature that set special demands on ion channel function to maintain adequate excitability of cardiac myocytes in various temperatures. Indeed, prolonged exposure to low temperature induces compensatory shortening of AP that is associated with strong upregulation of the delayed-rectifier K+ current (IKr) (12, 34). Curiously, at the same time the density of the strong inward rectifier K+ current, the IK1, is reduced. The physiological importance of this cold-induced decrease is not completely understood, but it might increase cardiac excitability by reducing the demand for the depolarizing Na+ current. The objective of this study was to examine the molecular basis of the rainbow trout IK1 and to find out whether differences in molecular composition could explain temperature-induced changes and regional differences in IK1 density of the trout heart. To this end, we cloned Kir2.1 and Kir2.2 genes, determined their expression, and measured cardiac IK1 at the whole cell and single-channel levels from thermally acclimated trout.
MATERIALS AND METHODS
Rainbow trout (Oncorhynchus mykiss) were donated by a local fish farm (Kontiolahti, Finland). In the laboratory, the fish were kept in 500-liter stainless steel tanks with aerated groundwater continuously delivered to the tanks at ∼0.5 l/min and were acclimated to 4°C (cold acclimation, CA) or 18°C (warm acclimation, WA) for at least 3 wk before use in experiments. Fish were fed commercial trout fodder (Biomar, Brande, Denmark) to satiation three times a week. All experiments were conducted with consent of the local committee for animal experimentation and the Ministry of Agriculture and Forest Affairs (Finland).
Extraction of RNA and DNA.
Fish were stunned by a sharp blow to the head, and the spine was cut. Atrium, ventricle, brain, and pieces of gill, kidney, liver, and skeletal muscle were quickly removed and immediately frozen in liquid nitrogen. Total RNA was extracted with TRIzol reagent (Invitrogen, Carlsbad, CA) according to the manufacturer's instructions. DNA was extracted from liver by the method of Sambrook et al. (30). RNA and DNA were qualified by gel electrophoresis and quantified by UV spectrophotometry.
Cloning of Kir2.1 and Kir2.2 open reading frames from rainbow trout heart.
First-strand cDNA was prepared from total cardiac RNA and was treated with RQ1 RNase-free DNase (Promega, Madison, WI). Reverse transcription was performed with M-MuLV Reverse Transcriptase RNase H− (Finnzymes, Espoo, Finland) and an oligo(dT)18 primer. Partial cDNAs corresponding to rainbow trout (om)Kir2.1 and omKir2.2 were obtained by PCR using degenerative primers designed to the homologous regions of corresponding mammalian genes (Table 1). PCR was performed in a 25-μl reaction mixture containing 50 mM Tris·HCl, 1.5 mM MgCl2, 15 mM (NH4)2SO4, 0.1% Triton X-100, each dNTP at 200 μM, 1 U of DyNAzyme EXT (Finnzymes), 2 μl of cDNA, and 5 pmol of each primer. Amplification was performed under PCR conditions with initial denaturation at 94°C for 2 min followed by 1–4 cycles with low annealing temperature at 94°C for 30 s, 40°C for 30 s, and 72°C for 90 s, further followed by 31 cycles with higher annealing temperature at 55°C and final extension at 72°C for 5 min. PCR products were checked on a 0.8% agarose gel, and if no products were obtained 0.5 μl of the PCR product was reamplified under PCR conditions with initial denaturation at 96°C for 2 min followed by 35 cycles at 95°C for 30s, 53°C for 30 s, and 72°C for 90 s and final extension at 72°C for 5 min. New forward primers were designed on the basis of the sequences obtained and used to clone the rest of the open reading frames (ORFs) (Table 1). Oligo(dT) primer was used to clone the 3′-ends of the genes.
The Genome Walker Kit (Clontech, Palo Alto, CA) was used for the cloning of 5′-ends of the genes. To this end, high-molecular-mass genomic DNA was extracted from rainbow trout liver (30), and four genomic libraries were constructed by digesting genomic DNA with DraI, EcoRV, PvuII, and StuI, followed by ligation to the adaptor. Gene-specific primers and adaptor primers were used to amplify the 5′-ends of the genes. The first PCR was performed with GSP1 (gene-specific primer) and AP1 (adaptor primer) (Table 1) and a two-step PCR protocol recommended by the manufacturer: 6 cycles at 94°C for 25 s and 72°C for 3 min followed by 31 cycles at 94°C for 25 s and 67°C for 3 min and final extension at 72°C for 7 min. Of this PCR product, 0.5 μl was reamplified with GSP2 and AP2 (Table 1). The second PCR was identical to the first, except that the annealing temperature was 68°C for the first six cycles and 63°C for the rest. The sequences were confirmed by cloning the whole coding region of the Kir2.1 and Kir2.2 genes from genomic DNA. PCR was performed as described in the use of degenerative primers, except that 50 ng of DNA was used as a template.
After amplification, all PCR products were analyzed by gel electrophoresis, extracted from gel with the Qiaex II gel extraction kit (Qiagen, Valencia, CA), cloned into the pGEM-T Easy Vector (Promega), and sequenced by a ABI Prism 310 Genetic Analyzer (Applied Biosystems, Foster City, CA).
Ten micrograms of genomic DNA was digested to completion with EcoRI, HindIII, or NcoI. DNA fragments were size separated on a 1% agarose gel and blotted onto positively charged nylon membrane (Boehringer Mannheim, Mannheim, Germany). Probes were produced by PCR, amplifying nucleotides 1026–1284 and 1119–1341 from omKir2.1 and omKir2.2 plasmids, respectively. Probes were labeled with [α-32P]dCTP, using Ready-To-Go DNA Labeling Beads, and purified by ProbeQuant G-50 Micro Columns (all obtained from Amersham Biosciences). Nylon membranes were prehybridized 2 h at 42°C in 20 ml of prehybridization solution containing 50% formamide, 3× SSC (0.45 M NaCl, 0.045 M Na citrate), 2.5× Denhardt's solution (0.05% BSA, 0.05% polyvinylpyrrolidine, 0.05% Ficoll), 0.25% SDS, and 0.1 mg/ml herring sperm DNA, followed by hybridization overnight with the probe at the same conditions. Thereafter they were washed twice for 15 min at room temperature and twice at 37°C with 1× SSC-0.5% SDS solution, followed by a final wash at 65°C for at least 15 min with 0.1× SSC-1% SDS solution. Signals were detected by autoradiography.
Total RNA was extracted from heart, brain, gill, kidney, liver, and skeletal muscle of two WA trout for organ-specific gene expression and from atrium and ventricle of three WA and three CA trout for temperature acclimation-dependent gene expression. Each RNA sample (2 μg) was treated with RQ1 RNase-free DNase (Promega), and first-strand cDNA synthesis was performed with random hexamers and MMuLV RNase H+ (Finnzymes) at the following conditions: 25°C for 10 min, 37°C for 30 min, and 85°C for 5 min. A control run containing all other reaction components but RT enzyme was performed for every sample. Quantitative PCR (qPCR) was performed with the DyNAmo HS SYBR Green qPCR Kit (Finnzymes) and the Chromo4 Continuous Fluorescence Detector (MJ Research, Waltham, MA), using primer pairs listed in Table 2 (primers for DnaJA2 were a kind gift from Dr. Aleksei Krasnov, University of Kuopio, Finland). Polymerase was activated at 94°C for 15 min, and amplification was performed for 40 cycles at 94°C for 10 s, 57°C for 20 s, and 72°C for 30 s. After PCR, the specificity of the reaction was monitored by melting curve analysis. Two controls, one containing all the reaction components except the template and the other containing the product of cDNA synthesis performed without RT enzyme, were included in every experiment. DnaJA2 was selected as a reference gene. DnaJ is a small chaperonin, which belongs to the heat shock proteins, but unlike many commonly used reference genes, such as ribosomal proteins and β-actin, its expression remains quite constant in temperature acclimation (35). The suitability of DnaJA2 as an internal standard was further confirmed by a qPCR experiment using equal amounts of RNA extracted from the tissues of CA and WA trout as a template.
Modeling of trout Kirs.
Homology models of omKir2.1 and omKir2.2 channels were constructed with the DeepView 3.7 program (GlaxoSmithKline) and a SWISS-MODEL server. KirBac3.1 (PDB code 1XL4) and COOH terminus of mammalian Kir3.1 (PDB code 1N9P) (21) were used as templates for the transmembrane segments (G50-N185 or G55-N195) and the COOH terminus (N185-A358 or N195-A376) of omKir2.x, respectively. The fourfold rotational symmetry of the channels was adjusted manually by using KirBac3.1 and KcsA (PDB code 1BL8) (6) as templates.
Isolation of cardiac myocytes.
Atrial and ventricular myocytes were isolated by previously published methods (33). Briefly, fish were stunned with a quick blow to the head, the spine was cut, and the heart was excised. A metallic cannula was brought through the bulbus arteriosus into the ventricle, and the heart was retrogradely perfused with a nominally Ca2+-free, low-Na+ solution for 10 min and then with proteolytic enzyme solution for 15 min. Solutions were continuously gassed with 100% O2, and the enzyme solution was recycled with a peristaltic pump. The Ca2+-free saline contained (mM) 100 NaCl, 10 KCl, 1.2 KH2PO4, 4 MgSO4, 50 taurine, 20 glucose, and 10 HEPES adjusted to pH 6.9 with KOH at 20°C. For enzymatic digestion, 0.75 mg/ml collagenase (type IA, Sigma), 0.5 mg/ml trypsin (type IX, Sigma), and 0.5 mg/ml fatty acid-free bovine serum albumin (A6003, Sigma) were added to this solution. After enzymatic digestion, atrium and ventricle were chopped in small pieces in the Ca2+-free solution, and myocytes were released by agitating tissue pieces through the opening of a Pasteur pipette. Isolated myocytes were stored at 6°C and used within 8 h of the isolation.
Whole cell patch clamp.
Whole cell voltage-clamp recording of the IK1 was performed with an Axopatch 1-D (Axon Instruments) or an EPC-9 (HEKA Instruments) amplifier (22). Myocytes were superfused in a small recording chamber with a precooled external saline solution at the rate of 1.5–2.0 ml/min. Temperature was adjusted to 4°C and 18°C for CA and WA trout, respectively, with circulating water baths or a Peltier device (TC-10, Dagan) and was continuously monitored by a thermistor positioned closed to the myocyte. The external saline solution contained (mM) 150 NaCl, 5.4 KCl, 1.8 CaCl2, 1.2 MgCl2, 10 glucose, and 10 HEPES (pH adjusted to 7.7 with NaOH). Patch pipettes were pulled from borosilicate glass (Garner, Claremont, CA) and filled with K+-based electrode solution (mM: 140 KCl, 1 MgCl2, 5 EGTA, 4 MgATP, and 10 HEPES adjusted to pH 7.2 with KOH) giving a mean (±SE) pipette resistance of 2.81 ± 0.07 MΩ. After gaining access to the cell, pipette capacitance (7.55 ± 0.14 pF, n = 48) and series resistance (9.96 ± 0.48 MΩ, n = 48) were compensated. The inward rectifier current was determined as a Ba2+ (0.1 mM)-sensitive current in the presence of tetrodotoxin (0.5 μM, Tocris Cookson), nifedipine (10 μM, Sigma), glibenclamide (10 μM, Sigma), and E-4031 (1 μM, Alomone Labs) to block Na+, Ca2+, and ATP-sensitive K+ current and IKr, respectively. The IK1 was elicited every 5 s by repolarizing ramps or square wave pulses from a holding potential of −80 mV.
Ba2+ inhibition of the IK1 was determined in the presence of cumulatively added concentrations of BaCl2 (10−9–10−4 M). The cell was exposed to each Ba2+ concentration for 4 min. Dose-response curves were fitted with the Hill equation I = Imin + Imax [Ba]H/IC50H+ [Ba]H, where Imin is the minimum IK1 at the highest Ba2+ concentration, Imax the IK1 before Ba2+ addition, IC50 the drug concentration that causes half-maximal inhibition of the IK1, [Ba] the Ba2+ concentration, and H the Hill slope of the line.
Voltage dependence of inward rectification was determined as a portion of the Ba2+-sensitive current that was inhibited at depolarizing voltages relative to the unblocked current. The unblocked (nonrectifying) inward current was obtained from the current-voltage relationship between −120 mV and the reversal potential (EiK1) of the IK1 and extrapolated to the voltage area of inward rectification. Scattering data points around the EjK1 were omitted, and the current was fitted as the sum of two Boltzmann functions: IK1 = A[1 + exp(V − Vh1)/S1] + A2[1 + exp(V − Vh2)/S2], where A, Vh, and S are amplitudes, voltages, and slopes of half-block for steep and shallow components of the polyamine block, respectively.
Single-channel patch clamp.
Single-channel properties of the omKir2 channels were recorded in the cell-attached configuration with an EPC-9 amplifier and Pulse software and analyzed with TAC, TACFIT (Bruxton), and SigmaPlot 6.0 (SPSS) programs as described previously (23). Pipettes were pulled from thick-walled borosilicate glass (Garner, Claremont, CA) with a two-state puller (PP-83, Narishige, Tokyo, Japan), coated with Sylgard (WPI), and fire polished on a microforge (MF-83, Narishige). The mean resistance of the pipettes was 9.2 ± 0.6 MΩ when they were filled with K+-based solution (in mM: 134 KCl, 1.8 CaCl2, 2 MgCl2, 10 glucose, and 10 HEPES adjusted to pH 7.6 with KOH) ([K+] = 141 mM). Physiological saline solution was used as the bath solution. The activity of unwanted channels was abolished by including specific ion channel blockers in the pipette and Tyrode solutions (see Whole cell patch clamp). All single-channel recordings were sampled at 4 kHz and low-pass filtered at 2 kHz. Single-channel conductance was determined by applying 5-s square pulses from −200 to −20 mV in 20-mV increments every 10 s from the holding potential of −80 mV. Distributions of open and closed times were obtained from 20-s recordings at −100 mV. Open time and closed time analyses were performed on patches, which had only a single open current level. Open and closed times were detected with time course fitting, and probability density functions (pdf) were analyzed from idealized data with log-likelihood method on log (event times) (TACFIT).
Functional expression of omKir2 genes in COS-1 cells.
ORF sequences for putative ion channel-forming genes omKir2.1 and omKir2.2 were subcloned into the pcDNA3.1/Zeo(+) vector (Invitrogen) for expression in a COS-1 cell line. Correctness of the sequences was confirmed by DNA sequencing. COS-1 cells (ECACC) were cultured in Dulbecco's modified Eagle's medium (DMEM, Bio-Whittaker, Cambrex Verviers, Belgium) containing 10% fetal bovine serum (FBS; EuroClone, Milan, Italy) and 3,000 U/m of both penicillin and streptomycin (EuroClone). Cells were transiently cotransfected with pEGFP-N1 (Clontech) and either omKir2.1- or omKir2.2-pcDNA3.1/Zeo(+), using Effectene Transfection Reagent (Qiagen). Function of the cloned channels was tested with the whole cell patch-clamp method at room temperature (20°C) 48–64 h after transfection. The same external and internal solutions were used as in measurement of endogenous currents (see Whole cell patch clamp).
Differences between mean values from WA and CA trout and between atrial and ventricular IK1 were assessed by Student's t-test. If the data were not normally distributed, differences were tested with the nonparametric Mann-Whitney test. A P value of 0.05 was regarded as a limit of statistical significance.
Rainbow Trout Kir2.1 and Kir2.2 Genes
Coding regions of Kir2.1 (KCNJ2) and Kir2.2 (KCNJ12) genes were cloned from rainbow trout cDNA and genomic DNA. Genomic clones indicated that the coding regions of these genes were intronless. Two different alleles both for Kir2.1 and Kir2.2 were found and named omKir2.1a, omKir2.1b, omKir2.2a, and omKir2.2b. The nucleotide sequences for omKir2.1a and omKir2.1b ORF include 1,278 and 1,284 bp, respectively, omKir2.1a lacking nucleotides 93–98 of the omKir2.1b (Fig. 1A). The nucleotide sequences for omKir2.2a and omKir2.2b ORF include 1,341 bp coding 446 amino acid residues (Fig. 1B). Nucleotide sequences of the two alleles are 96.6% and 97.4% identical for omKir2.1 and omKir2.2, respectively, and show even higher amino acid identity. Southern blot hybridization of the genomic DNA with omKir2.1 or omKir2.2 probes revealed only one band for each gene, suggesting that the two alleles are located in equivalent loci of the genome (data not shown).
Amino acid alignments of omKir2.1 and omKir2.2 with mammalian Kir proteins reveal about 82% and 66% sequence similarity, respectively (Fig. 2). The lowest homology with mammalian Kirs is seen in the COOH terminus and in the M1-P loop (“the turret”), both in omKir2.1 and omKir2.2. Interestingly, the M1 transmembrane domain of omKir2.2 also differs greatly from the corresponding mammalian sequence. A structural model of omKir2.2 (Fig. 3) shows that the differences are located in the lipid bilayer-facing side of the membrane-spanning α-helix, where almost one-third (6) of the amino acids are leucines (Fig. 1). In the M2 transmembrane domain, the glycine G169 of the mammalian Kir2.2—the so-called glycine hinge—is replaced by serine (S173) in omKir2.2. The most prominent structural difference between omKir2.1 and omKir2.2 is the extracellular mouth formed by the M1-P loop, which in omKir2.1 is oriented straight upward and does not form any secondary structures, whereas in omKir2.2 it forms a short β-sheet and is turned toward the membrane, being possibly in contact with the lipid bilayer.
Functionally important sequences that are common to all Kir subfamilies, including the pore domain with K+ selectivity signature sequence (GYG), were identified in the rainbow trout Kir2s with multiple sequence alignments (Figs. 1 and 2). Amino acids, critical for polyamine, Mg2+, and phosphatidylinositol 4,5-bisphosphate (PIP2) binding were the same as in mammalian Kir2.1 and Kir2.2 channels. In contrast, differences existed in extracellular Ba2+ binding sites between fish and mammalian Kirs: in omKir2.1 glutamate E125 was replaced with asparagine, and in omKir2.2 threonine T141 was replaced with serine.
Expression of omKir2 mRNAs
The expression of omKir2.1 and omKir2.2 transcripts was determined from different organs by quantitative RT-PCR and normalized to the abundance of DnaJA2 mRNA. DnaJA2 expression was quite constant at both acclimation temperatures (data not shown) and therefore a suitable reference gene, as confirmed by qPCR using the same amount of RNA from the tissues of CA and WA trout as a template. Both omKir2.1 and omKir2.2 were expressed in all examined organs including heart, brain, gill, kidney, liver, and skeletal muscle (Fig. 4A). Of the two genes, omKir2.1 was more abundant in all other organs except the brain, where omKir2.2 was the dominant transcript. The expression of omKir2.2 in the brain was 10 times higher than in the atrium and much lower in ventricle, gill, skeletal muscle, and liver. The expression of omKir2.1 was highest in the kidney, being 7.1 times higher than in the atrium. In general, the total amount of omKir2 mRNAs in cells was very low in all tissues, e.g., in the atrium only ∼1% of the DnaJA2 expression level, which is not a very abundant protein either.
Since thermal acclimation changes IK1 of the trout heart and because there are striking differences in density and rectification of the IK1 between atrium and ventricle, Kir2 expression was studied in both cardiac chambers of WA and CA trout (Fig. 4B). Although omKir2.1 was the dominant isoform in both chambers, the atrium expressed relatively more omKir2.2 than the ventricle (P = 0.021), the transcripts of omKir2.1 being 2.7, 1.7, 7.3, and 6.1 times more abundant than omKir2.2 mRNA in CA atrium, WA atrium, CA ventricle, and WA ventricle, respectively. Quite surprisingly, no significant differences in omKir2 expression existed between CA and WA trout hearts.
Density and Inward Rectification of IK1
In accordance with previous findings (34) the peak outward and inward IK1 density of trout ventricular myocytes was 22–25% larger (P < 0.05) in WA than CA fish (Fig. 5A). The charge transfer of the outward current between EiK1 and 0 mV did not, however, differ between CA and WA trout (P = 0.12). There was a striking difference in current size between ventricular and atrial myocytes (Fig. 5A). The peak inward density and outward charge transfer of the atrial IK1 were only about 3% and 10%, respectively, of the values of the ventricular IK1.
The current-voltage relationships indicate that the ventricular IK1 rectifies completely at positive voltages, while inward rectification of the atrial IK1 is incomplete (Fig. 5A). Two Boltzmann functions were needed to fit the steady-state inward rectification of both atrial and ventricular IK1, which like the mammalian IK1 consists of a steep and a shallow component (Fig. 5B). The weaker inward rectification of atrial IK1 compared with ventricular IK1 is mainly due to the differences in the shallow component, which is larger and shallower in atrial than ventricular myocytes. Thermal acclimation did not affect inward rectification either in atrial or ventricular myocytes. Together, these findings indicate striking differences in IK1 between atrial and ventricular myocytes and suggest that warm acclimation slightly increases the number of active Kir channels in ventricular myocytes.
External K+ and Ba2+ Concentration Dependence of IK1
Different Kir2 channels vary in regard to their response to external K+ and Ba2+. To examine putative atrioventricular differences in Kir composition, K+ dependence, and Ba2+ sensitivity of the rainbow trout, cardiac IK1 were determined by whole cell patch clamp. Doubling of external K+ concentration ([K+]o) from 5.4 to 10.8 mM resulted in a Nernstian shift of the (EiK1) from about −80 to −60 mV (Table 3), and 1.64- to 1.74-fold and 1.70- to 2.43-fold increase of the inward IK1 in ventricular and atrial myocytes, respectively (Fig. 6). The effect of high [K+]o on the outward IK1 was, however, weak and opposite in the two cardiac chambers. In ventricular myocytes the outward IK1 was marginally (1.05- to 1.15-fold) increased, while in atrial myocytes it was slightly (0.79- to 0.93-fold) depressed. As an outcome of this difference, the current-voltage relationships of the outward IK1 crossed each other at the two [K+]o in ventricular but not in atrial myocytes. The ventricular crossover points differed between CA (−39.75 ± 1.77 mV) and WA (−45.88 ± 1.66 mV; P < 0.05) trout, but when the crossover points were expressed relative to the reversal potential of the IK1 at 5.4 mM [K]o, i.e., corrected for the temperature-dependent shift of the reversal potential, the difference disappeared (33.3 ± 1.5 and 32.5 ± 1.4 mV for CA and WA trout, respectively; P = 0.70). Together, these experiments indicate marked chamber-specific differences in K+ sensitivity in the trout cardiac IK1.
The trout atrial IK1 was so small that the effects of external Ba2+ concentration ([Ba2+]o) could be reliably measured only in ventricular myocytes. The [Ba2+] required for half-maximum block of the IK1 was almost an order of magnitude lower in WA (IC50 0.18 ± 0.13 μM) than in CA (1.17 ± 0.15 μM) (P < 0.05) trout ventricular myocytes (Fig. 7A). Moreover, the Hill slopes differed, being −0.69 ± 0.08 and −0.97 ± 0.08 (P < 0.05) in WA and CA trout, respectively. Differences in Ba2+ sensitivity of the native IK1 suggest temperature-related variation in omKir2 channel composition. To test this possibility omKir2.1 and omKir2.2 channels were expressed in COS-1 cells and their Ba2+ sensitivity was measured. Both genes formed K+-selective ion channels with clear inward rectifying properties (Fig. 7B). Current amplitude at −120 mV was 2.5 ± 0.5 (mean ± SE) and 2.2 ± 0.3 nA and reversal potential −70.2 ± 3.6 and −73.7 ± 2.0 mV for omKir2.1 (n = 7) and omKir2.2 (n = 10), respectively (P > 0.05). No inward rectifying K+ current was detected in nontransfected COS-1 cells (data not shown). Sensitivity of omKir2.1 to Ba2+ was almost an order of magnitude lower (IC50 24.99 ± 7.40 μM) than in omKir2.2 (IC50 2.88 ± 0.42 μM) (P < 0.05; Fig. 7B), but Hill slopes did not differ between omKir2.1 and omKir2.2 (1.15 ± 0.22 vs. 0.997 ± 0.064; P > 0.05).
Single-Channel Conductance of Native Kir Channels
To separate acute temperature effects from putative acclimation effects, single-channel currents were measured at the same experimental temperature (11°C) for both acclimation groups. Single-channel conductance was practically the same for CA (18.0 ± 1.3 pS, n = 24) and WA (18.0 ± 0.7 pS, n = 32) (P > 0.05) trout (Fig. 8). In contrast, there were clear temperature-related differences in gating kinetics. The time constant for opening was about four times slower in WA trout (23 ms) than in CA trout (6 ms) (P < 0.05), and three functions were needed to describe closing kinetics in WA trout (0.58, 5.37, and 11.3 ms), while two were sufficient in CA trout (0.96 and 6.04 ms) (Fig. 8).
Rainbow trout are moderately eurythermal fish that maintain activity throughout their thermal tolerance range (0–25°C). This requires thermal plasticity in various body functions including the cardiovascular system, i.e., the trout genome must be able to produce different phenotypes by differential gene expression according to thermal conditions of the environment (35). Shortening of the trout cardiac AP under chronic cold exposure (4°C) enables higher heart rates and larger cardiac output and is indicative for compensatory changes in cardiac ion channel function (1, 12). The inward rectifier, IK1, does not contribute to AP shortening because it is decreased by cold acclimation (Ref. 34; present study). IK1 has a stabilizing effect on membrane potential, and therefore the reduction of IK1 might increase excitability of trout ventricular myocytes in the cold. Here we show that cardiac IK1 of the rainbow trout is produced by omKir2.1 and omKir2.2 channels and that Ba2+ sensitivity and single-channel kinetics of the ventricular IK1 are modified by thermal acclimation. On the basis of the different Ba2+ sensitivities of homomeric omKir2.1 and omKir2.2 channels, it is concluded that warm acclimation increases either number or activity of omKir2.2 channels in trout ventricular myocytes. These functional changes are independent of omKir2 transcript levels, which remained unaltered by thermal acclimation.
Expression of omKir2.1 and omKir2.2
Kir2 inward rectifier K+ channels are involved in a number of cellular functions, and hence they are ubiquitously distributed in the animal body (29). In addition to their expression in heart, liver, and skeletal muscle, transcripts of omKir2.1 and omKir2.2 were especially abundant in brain, gill, and kidney. Trout brain was the only organ in which omKir2.2 was more abundant than omKir2.1, suggesting a significant role for omKir2.2 in trout brain function (27, 40). In the kidney, IK1 is involved in the regulation of renin secretion and consequently in blood pressure regulation and electrolyte homeostasis (25). Consistent with earlier findings from the rat kidney (15), omKir2.1 is also the main Kir2 transcript in the trout kidney. The relatively high expression of omKir2.1 in trout gill suggests its participation in ion regulation. Previously, Suzuki et al. (32) cloned a homolog of the mammalian Kir7.1 from the gills of the seawater-adapted eel. Thus the present study extends the expression of fish gill K+ channels to the Kir2 subfamily.
Sequence Structures of omKir2.1 and omKir2.2
Amino acid alignments show that omKir2.1 and omKir2.2 are relatively highly homologous to corresponding mammalian proteins. Even so, some interesting sequence differences exist, particularly in the transmembrane domains. For example, several amino acids in the M1 domain of the omKir2.2 are different from those of the mammalian Kir2.2. Growing evidence suggests that ion channels are regulated by their lipid environment (19, 28). Considering the large difference in body temperature between mammals and ectotherms, and associated differences in chemical composition of lipid membrane (3), it is not unexpected that the amino acids facing toward the lipid bilayer differ. Kir2 channels are almost exclusively located in the cholesterol-rich lipid rafts, and their activity is strongly modulated by membrane cholesterol (28). Because temperature acclimation has clear impact on cholesterol content of the lipid rafts in rainbow trout (41), the importance of membrane cholesterol on temperature-dependent regulation of the trout cardiac IK1 warrants further study.
In the mammalian Kir2.1, glutamates E224 and E299 of the inner vestibule are involved in polyamine binding and positioned so that the access of polyamines to aspartate D172, the blocking site in the transmembrane pore region, is enhanced (38). All three amino acids exist both in omKir2.1 (D168, E220, and E295) and omKir2.2 (D178, E230, and E305). Magnesium, the other important intracellular blocking agent of Kir2 channels, binds to the serine S165 in the mammalian Kir2.1 (10). This serine is also conserved in trout Kir2 channels: S161 in omKir2.1 and S171 in omKir2.2. Together these findings indicate that amino acids critical for inward rectification are highly conserved from fish to mammals. This structural similarity is reflected in inward rectification of the IK1, which is composed of steep and shallow components in both vertebrate classes (Ref. 24; this study).
In contrast, amino acid residues involved in Ba2+ binding are partially different in mammalian and fish Kirs. In the mammalian Kir2.1, glutamate E125 in the extracellular loop between M1 and the pore region interacts with Ba2+ and facilitates its access to the plugging site, threonine T141 in the pore (2). In omKir2.1, E125 is replaced by asparagine. The Kir2.2 subunit does not have the glutamate E125, only the threonine T141. In omKir2.2, T141 is replaced by serine. Analysis of cloned omKir2 channels showed that omKir2.2 is ∼10 times more sensitive to Ba2+ than omKir2.1. This agrees with previous studies that have established that homomeric Kir2.1 channels are 5- to 10-fold more resistant to blockade by Ba2+ than Kir2.2 channels (16, 26). Interestingly, IK1 of the WA trout heart was almost 10-fold more sensitive to Ba2+ than IK1 of the CA trout. This strongly suggests that warm acclimation increases either number or activity of the omKir2.2 subunits in ventricular myocytes of the trout heart. It is possible that in CA trout practically all Kir channels are composed of the omKir2.1 subunits and thus have low Ba2+ affinity, whereas in WA trout some portion of the channels might be homomeric Kir2.2 channels and therefore more sensitive to Ba2+. The low Hill slope value of the WA trout suggests that more than one type of Ba2+ binding site with different affinities to Ba2+ might exist in WA trout Kirs.
Chamber-Specific Differences of IK1
Contradiction between IK1 density and Kir2 expression is especially striking regionally, i.e., between atrium and ventricle of the trout heart. Despite ∼30 times larger IK1 in ventricular than atrial myocytes, the summed expression levels of omKir2.1 and omKir2.2 transcripts were not higher in the ventricle. The higher relative expression of omKir2.2 transcripts in the atrium cannot explain the large difference in IK1 density. There are several possible explanations for this contradiction. First, mRNA levels do not necessarily correlate with protein levels, as has been shown for canine Kir2.1 and Kir2.3 (36). Thus regional differences in the number of omKir2 proteins might exist. Even Kir2 protein expression does not always correlate well with IK1 density, suggesting that, in addition to protein density of Kir channels, subcellular distribution of channels or other factors is involved in producing the whole cell IK1 (20). Second, polyamines regulate the rectification of Kir channels and affect the amplitude of both inward and outward IK1. This raises the possibility that regional differences in IK1 might be explained by variation in free polyamine pool. Recently, Yan et al. (39) provided evidence that in the guinea pig lower spermine and spermidine concentrations of the ventricular tissue might explain the twice-larger IK1 of the ventricular myocytes compared with the atrial IK1. Atrioventricular differences in the density of IK1 are, however, much more extreme in trout than guinea pig, and therefore it is questionable whether differences in the amount of free polyamines alone could account for this large difference. Third, the membrane phospholipid PIP2 binds directly to Kir channels and activates them by stabilizing the open state (13). Therefore, one possible explanation for the atrioventricular difference in IK1 density may lie in the PIP2 regulation of Kir2 channels. All Kir2 channels are sensitive to PIP2, but their affinity to PIP2 differs. Kir2.1 has much higher affinity to PIP2 than Kir2.2 and Kir2.3 (7). Accordingly, the low IK1 density in atrial myocytes could result from a higher relative proportion of low-affinity Kir2.2 channels and/or low atrial PIP2 level. Fourth, trout heart might have other Kir2 channels in addition to Kir2.1 and Kir2.2, which could account for the regional differences in IK1. We were unable to detect Kir2.3 mRNA from the trout heart in the present study but cannot completely exclude the possibility that Kir2.3 is involved in the formation of the trout IK1.
Because species-specific antibodies were not available for omKir2.1 and omKir2.2 to determine omKir2 protein expression, we tried to resolve regional and temperature-related differences of the IK1 with functional studies. Because homomeric Kir2 channels differ in respect to external K+ dependence, single-channel conductance, and Ba2+ sensitivity, electrophysiological studies could reveal functional differences that may underlie Kir subunit variation of the IK1 between CA and WA trout. The drastic difference in Ba2+ sensitivity of the IK1 between CA and WA trout, as discussed above, is indicative of a higher omKir2.2 expression in WA ventricle. Temperature-related differences were also evident in single-channel properties of the trout ventricular IK1. The mammalian Kir channels differ in respect to their single-channel conductance, Kir2.2 having the largest and Kir2.3 the smallest conductance and Kir2.1 being intermediate between the two (16). The similarity of single-channel conductances of the native omKir2 channels in WA and CA trout does not provide any clues to the temperature-related changes in trout Kir channel composition. However, the kinetics of omKirs differed between WA and CA trout, the latter having faster gating. Whether this is due to higher omKir2.2 subunit expression, lipid-protein interaction, polyamine block, or regulatory mechanisms needs to be resolved by further experiments. Collectively, the electrophysiological studies suggest that functional heterogeneity of trout IK1 may originate from differential expression or assembly of omKir2.1 and omKir2.2 subunits in CA and WA trout.
A hallmark of the strong inward rectifier channels is the crossover of current-voltage curves on increase in extracellular K+. This crossover effect originates from the polyamine- and Mg2+-dependent negative slope conductance of the IK1 and an increase in the peak outward current in high-K+ solution. Atrial and ventricular IK1 of the trout heart had distinctly different properties in that the current-voltage curves of the atrial IK1, unlike those of the ventricular IK1, did not cross at different K+ concentrations. The absence of the crossover effect in atrial myocytes was caused by a decrease of the outward current in high-K+ solution. Previous studies on mammalian Kirs imply that the outward current magnitude increases in Kir2.1 but not in Kir2.3 channels (4). If this were the case, Kir2.3 might be involved in the formation of trout atrial IK1 and it would only be because of the methodological limitations that Kir2.3 was not found in the trout heart. However, recordings of the IK1 in cell-free outside-out patches indicate that in all Kir2 channels current-voltage relationships display the characteristic crossover effect (24). Accordingly, the absence of the crossover effect cannot be ascribed to any specific Kir2 channel but may be an outcome of an interaction of Kir2 channels with some intracellular regulatory factors. Taken together, the prominent differences in density and rectification of the trout IK1 between atrial and ventricular myocytes cannot be ascribed to different omKir2 transcripts but rather may depend on differences in polyamines and other intracellular regulators that interact with omKir2 subunits.
Kir2 transcript expression suggests that rainbow trout cardiac IK1 is composed of omKir2.1 and omKir2.2 subunits. The relative mRNA expression of the two subunits is not affected by temperature acclimation, while electrophysiological experiments indicate distinct functional properties of the IK1 in CA and WA trout, especially in regard to Ba2+ sensitivity and single-channel kinetics. Considering the different Ba2+ sensitivities of the homomeric omKir2 channels, functional variation of the native IK1 suggests that either activity or number of the omKir2.2 channels is increased by warm acclimation. Although the relative expression of the omKir2.2 was higher in atrium than ventricle, the atrioventricular differences in IK1 density cannot be explained by expression levels of omKir2.1 and omKir2.2 transcripts and may involve differences in polyamine-dependent regulation of atrial and ventricular Kirs or differential coassembly of the omKir2 subunits.
This work was financed by a research grant from the Academy of Finland to M. Vornanen (project no. 210400) and a personal grant of the Academy of Finland to V. Paajanen. M. Hassinen was supported by the Biological Interactions Graduate School.
We thank Aleksei Krasnov for the gift of the DnaJ primers and Holly Shiels for reading the manuscript and helpful suggestions. Kontiolahti Fish Farm is acknowledged for donating the trout. Anita Kervinen and Riitta Pietarinen are appreciated for technical assistance.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 the American Physiological Society