Antisense Oligodeoxynucleotides Directed Against Kv1.5 mRNA Specifically Inhibit Ultrarapid Delayed Rectifier K+ Current in Cultured Adult Human Atrial Myocytes
Abstract Several cloned K+ channel subunits are candidates to underlie macroscopic currents in the human heart, but direct evidence bearing on their role is lacking. The Kv1.5 K+ channel subunit has been suggested to play a potential role in human cardiac ultrarapid delayed rectifier (IKur) and transient outward (Ito) currents. To evaluate the role of proteins encoded by the Kv1.5 gene, we incubated cultured human atrial myocytes for 48 hours in medium containing antisense phosphorothioate oligodeoxynucleotides directed against octodecameric segments of the Kv1.5 mRNA coding sequence, the same concentration of homologous oligodeoxynucleotides with four mismatch mutations, or vehicle (control group). Cells exposed to antisense showed a highly significant (≈50%) reduction in IKur, whether measured by step current at the end of a 400-millisecond depolarizing pulse, tail current at −20 mV, or current sensitive to a concentration of 4-aminopyridine (50 μmol/L) that is highly selective for IKur, compared with control cells or cells exposed to mismatch oligodeoxynucleo-tides. In contrast, Ito was not different among the three experimental groups. When cultured human ventricular myocytes were exposed to Kv1.5 antisense oligodeoxynucleotides with the same controls, no changes occurred in either Ito or the sustained current at the end of a depolarizing pulse. We conclude that Kv1.5 channel subunits are essential to the expression of IKur and do not play a role in Ito in cultured human atrial myocytes. These studies provide the first direct evidence with an antisense approach for the equivalence between a macroscopic cardiac K+ current and a cloned K+ channel subunit and offer insights into the molecular electrophysiology of the human heart.
The widespread application of molecular biology techniques has resulted in the cloning of cDNAs coding for a wide variety of human cardiac K+ channel subunits.1 A number of these subunits form homomeric channels with properties that make them candidates to underlie native currents in human cardiac cells, but the definitive identification of the specific molecular basis for macroscopic currents remains a challenge.1 We have demonstrated the presence of a novel K+ current in human atrial myocytes that activates much more rapidly than IKr, and we have called this current the ultrarapid delayed rectifier, IKur.2 A variety of biophysical and pharmacological properties of IKur resemble those of currents carried by the expression of Kv1.5,2 3 4 5 6 and it has been suggested that IKur is the physiological manifestation of Kv1.5 expression.2 3 Mays et al7 showed evidence for the presence of Kv1.5 protein in human atrial and ventricular sarcolemma with the use of immunohistochemical techniques, but this approach does not allow for the assessment of whether the antigen detected corresponds to a functioning transmembrane ion channel. In fact, we found that IKur was absent in a series of human ventricular myocytes studied under conditions readily demonstrating the presence of IKur in atrial cells,8 despite the fact that Kv1.5 antigen is present in both human atrium and ventricle.7 Thus, there is a need for more direct evidence bearing on the relationship between Kv1.5 channel subunits and IKur in the human heart.
The molecular basis of Ito in the human heart is also an unresolved issue. Although an inactivating Shaker-type Kv1.4 K+ channel subunit with 4-AP sensitivity similar to that of human Ito was cloned from the human heart several years ago,9 its reactivation properties do not match those of human Ito.10 Among the possible explanations of this discrepancy are the existence of heterotetrameric channels involving Kv1.4 subunits and a slowly inactivating channel subunit like Kv1.511 or the introduction of rapid inactivation by a β-peptide12 13 14 15 to a slowly inactivating α-subunit like Kv1.5. Therefore, in addition to being a candidate to underlie IKur in the human heart, Kv1.5 may also be an important part of the channel underlying Ito.
Antisense oligodeoxynucleotides inhibit the production of proteins corresponding to the mRNA against which they are directed, acting in complex ways to affect mRNA expression and/or translation.16 They have recently been used to demonstrate the relationship between Kv1.5 channel subunits and K+ currents in pituitary cell lines.17 We have reported the characteristics of several ionic currents in adult human atrial myocytes maintained in primary culture for up to 7 days.18 Both IKur and Ito are readily recorded and have stable properties from the first through the seventh day in culture.18 The present study was designed to determine the effects of antisense oligodeoxynucleotides on the expression of IKur and Ito in cultured human atrial myocytes. To accomplish this, we compared cells incubated in the presence of either antisense octodecameric phosphoro-thioate oligodeoxynucleotides or control oligodeoxynucleo-tides with mismatch mutations (administered in a blinded fashion to avoid bias), along with cells maintained in control culture medium, to which only the vehicle for oligodeoxynucleotides was added.
Materials and Methods
Phosphorothioate oligodeoxynucleotides were synthesized on an Applied Biosystems 392 DNA synthesizer. After cleavage from the resin, the oligodeoxynucleotides were precipitated with ethanol and resuspended in water. Two adjoining antisense oligodeoxynucleotides were used in the experiments: 5′-GGGGCACCAGGGCGATCTCC-3′ (which will be referred to in the rest of the article as “antisense oligo A”) and 5′-CATGGCACCGCCGTTCTCC-3′ (which will be referred to as “antisense oligo B”). The two oligodeoxynucleotides were directed at nucleotides 3 to 22 and 24 to 42, respectively, of the human Kv1.5 coding sequence (nucleotide 1 corresponds to A of the initiating ATG codon). As a control, two other oligodeoxynucleotides were used: 5′-GGGACACCACGGCGAGCTAC-3′ and 5′-CATAGCACGGCTGTGCTCC-3′. Each of these is identical to one of the antisense oligodeoxynucleotides, with the exception of four mismatch mutations. These control oligodeoxynucleotides will be referred to as “mismatch oligo A” and “mismatch oligo B,” respectively.
Specimens of human right atrial appendage were obtained from the hearts of 9 patients (mean age, 61 years; range, 44 to 73 years) undergoing aortocoronary bypass surgery. The atria were all grossly normal and from patients without heart failure or atrial arrhythmias. The procedure for obtaining the tissue was approved by the Ethics Committee of the Montreal Heart Institute. The samples were quickly immersed in oxygenated (100% O2) nominally Ca2+-free Tyrode’s solution containing (mmol/L) NaCl 126.0, KCl 5.4, MgCl2 1.0, NaH2PO4 0.33, dextrose 10, and HEPES (Sigma Chemical Co) 10; pH was adjusted to 7.4 with NaOH. The tissue was immediately transported to the laboratory, where further processing was performed at 37°C. The specimens were diced into cubic chunks (1 mm3), and placed in a 25-mL flask containing 10 mL of the Ca2+-free Tyrode’s solution. The tissue was gently agitated by continuous bubbling with 100% O2 and stirring with a magnetic bar. After 5 minutes, the chunks were reincubated in a similar solution containing 390 U/mL collagenase (CLS II, Worthington Biochemical) and 4 U/mL protease (type XXIV, Sigma). The first supernatant was removed after 45 minutes, and the tissue was reincubated in a fresh enzyme-containing solution. Microscopic examination was performed every 15 minutes, and when the yield appeared to be maximal, the chunks were suspended in a storage solution containing (mmol/L) KCl 20, KH2PO4 10, glucose 10, glutamic acid 70, β-hydroxybutyric acid 10, taurine 10, EGTA 10, and albumin 1% (pH was adjusted to 7.4 with KOH) and gently pipetted. The remaining tissue chunks were then removed, the cell suspension was centrifuged at 250 rpm for 5 minutes, and the pellet was removed for culture. All procedures were performed with aseptic techniques.
Human ventricular myocytes were obtained from explanted right ventricles of two cardiac transplant recipients (aged 23 and 58 years) with the use of methods previously described in detail.8 In brief, a portion of the right ventricular free wall was perfused via a coronary artery, initially with Ca2+-free Tyrode’s solution (for 20 to 30 minutes) and then with a similar solution containing 200 to 300 U/mL of collagenase (CLS II, Worthington). The digested tissue was cut into small pieces (1 to 2 mm3) and placed in the storage solution described above, and the cells were obtained by gentle trituration with a Pasteur pipette. Both patients were suffering from idiopathic congestive cardiomyopathies, and the right ventricle was macroscopically normal in both cases. Microscopic examination showed interstitial fibrosis and mononuclear cell infiltrates in both.
Cell Culture and Exposure to Oligodeoxynucleotides
Cells were cultured in Petri dishes (35-mm, Nunc Co) containing medium 199 with 10% fetal bovine serum (both from GIBCO-BRL) supplemented with sodium penicillin G (1 U/mL) and streptomycin sulfate (1 μg/mL, GIBCO-BRL). Cells were allowed to adhere to the bottom of the Petri dish and maintained in the medium at 37°C in a humidified, 5% CO2-enriched atmosphere. Oligodeoxynucleotide treatment was started 24 hours after the onset of culture. Three groups of cultured cells were studied in all series of experiments: one (referred to in this article as the “control group”) was exposed to the vehicle for oligodeoxynucleotides (sterilized distilled water with the same volume as for oligodeoxynucleotide groups), a second group was exposed to mismatch oligodeoxynucleotides (mismatch oligo A and/or mismatch oligo B), and a third group was exposed to antisense (antisense oligo A and/or antisense oligo B) oligodeoxynucleotides. The oligodeoxynucleotides were prepared by one investigator (B.W.), who provided stocks that were coded so that the investigators performing culture, treatment, and electrophysiological study were unaware of treatment allocation until experiments and data analyses were completed. For each treatment, the growth medium was removed, and oligodeoxynucleotides or vehicle in medium without serum was added to the cells. After a 30-minute incubation at 37°C, heat-inactivated serum (final concentration, 10%) was returned to the medium. The same procedure was used for all groups and repeated every 12 hours for 48 hours. Patch-clamp experiments were performed after 48 hours of treatment.
Four series of experiments were performed. In the first series, cultured atrial cells from five hearts were exposed to 5 μmol/L antisense oligodeoxynucleotides (equal quantities of antisense oligo A and B), 5 μmol/L mismatch oligodeoxynucleotides (equal quantities of mismatch oligo A and B), or equal volumes of vehicle. In the second series, atrial cells from two hearts were exposed to the same solutions, but at a lower concentration (2 μmol/L). Because of possible nonspecific effects of oligodeoxynucleotides containing four successive guanine bases (as in antisense oligo A), a third series of experiments was performed in which atrial cells from two hearts were exposed to 5 μmol/L antisense oligo B, 5 μmol/L mismatch oligo B, or an equal quantity of vehicle. In the final series of experiments, cultured ventricular myocytes were exposed to 5 μmol/L antisense oligo A and B, 5 μmol/L mismatch oligo A and B, or equal volumes of vehicle. In each series of experiments, results were obtained with approximately equal numbers of cells from each heart under each condition studied. This was essential to avoid distortion of results by interpreparation variability in current amplitudes and to ensure that the results were representative of all hearts for each condition.
Solutions and Drugs
The extracellular solution for patch-clamp studies contained (mmol/L) choline chloride 126, KCl 5.4, MgCl2 1.0, CaCl2 1.0, HEPES 5, NaH2PO4 0.33, and dextrose 10. The pipette solution contained (mmol/L ) potassium aspartate 110, KCl 20, MgCl2 1.0, HEPES 5, EGTA 5, Mg-ATP 5, GTP 0.1, and Na2-phosphocreatine 5. The pH of external and internal solutions was adjusted to 7.3 with the use of CsOH. CdCl2 (200 μmol/L) was used to block Ca2+ current. 4-AP was prepared as a stock solution (1 mol/L) in distilled water, with pH adjusted to 7.3 with HCl. All chemicals and drugs were obtained from Sigma.
Data Acquisition and Analysis
The whole-cell patch-clamp technique was used to record ionic currents in the voltage-clamp mode. The 35-mm Petri dishes used for cell culture also served as baths for voltage-clamp experiments. All experiments were conducted at room temperature (20°C to 23°C). We used borosilicate glass electrodes (outer diameter, 1.0 mm) with tip resistances of 2.5 to 5 MΩ when filled. Currents were recorded with an Axopatch 1-D amplifier (Axon Instruments). Command pulses were generated by a 12-bit digital-to-analog converter controlled by pClamp software (Axon). Recordings were low pass–filtered at 1 kHz. Currents were digitized (model TM 125, Scientific Solutions) and stored on the hard disk of a personal computer.
Junction potentials were zeroed before formation of the membrane-pipette seal in 1 mmol/L Ca2+–containing Tyrode’s solution. Mean seal resistance averaged 19.3±2.8 GΩ (n=30). Several minutes after seal formation, the membrane was ruptured by gentle suction to establish the whole-cell configuration for voltage clamping. The series resistance was electrically compensated to minimize the duration of the capacitive surge on the current record. Series resistance along the clamp circuit was estimated by dividing the time constant obtained by fitting the decay of the capacitive transient by the calculated cell membrane capacitance (the time integral of the capacitive surge measured in response to 5-mV hyperpolarizing steps from a holding potential of −60 mV divided by the voltage drop). The decay of the capacitive surge was well fit by a single-exponential relation. Cell capacitance averaged 30.2±2.8 pF (n=30 ) and 28.7±2.6 pF (n=30) before and after compensation, respectively, in atrial myocytes and 93.6±7.2 pF (n=20) and 90.5±6.8 pF (n=20) in ventricular myocytes. The capacitive time constant averaged 162.8±20.3 and 68.9±7.3 microseconds before and after compensation, respectively, in atrial myocytes and 851±68 and 233±19 microseconds in ventricular myocytes. Series resistance averaged 5.6±0.8 and 2.5±0.3 MΩ before and after compensation, respectively, in atrial myocytes and 8.9±0.6 and 2.7±0.2 MΩ in ventricular myocytes. Leak currents were small, and no correction for leak currents was applied.
The amplitude of Ito was measured as the difference between the peak of Ito and the current level at the end of the pulse. Isus was measured as the amplitude of the current at the end of the test pulse relative to the zero current level. Tail currents were measured as the time-dependent deactivating component at −20 mV after an activating pulse to a more positive potential.
Comparisons among groups were performed by ANOVA with Scheffé’s contrasts. A two-tailed probability of 5% was taken to indicate statistical significance. Group data are presented as mean±SEM. Nonlinear curve fitting (Marquardt’s procedure) was performed using Clampfit in pClamp.
Studies of Atrial Myocytes Exposed to Pairs of Oligodeoxynucleotides
As we have previously reported,18 cells became round after 24 hours in culture, and no changes in cell morphology occurred thereafter. There were no differences in cell capacitance among the three study groups; eg, mean capacitance was 30.6±1.5 (n=55), 30.8±1.2 (n=55), and 31.6±1.6 (n=55) pF in control (vehicle), combined antisense oligo A and B (5 μmol/L), and combined mismatch oligo A and B (5 μmol/L) groups, respectively.
Fig 1A⇓ shows currents recorded from 39 randomly selected control cells (of a total of 55 studied) upon 400-millisecond depolarization from a holding potential of −80 to +50 mV. Cells showed a rapidly activating and inactivating Ito component of variable magnitude, with a residual component at the end of the pulse, which we have previously shown corresponds largely to IKur.2 18 Corresponding results obtained from a random sample of 39 cells (of a total of 55) incubated in combined antisense oligo A and B (5 μmol/L) are shown in Fig 1B⇓, and results from 39 randomly sampled cells exposed to combined mismatch oligo A and B (5 μmol/L) are shown in Fig 1C⇓. When currents from all 55 cells in each group were averaged, the results shown in Fig 1D⇓ were obtained. The mean current-time relations for control cells and those exposed to mismatch oligodeoxynucleotides could be superimposed, but the mean current recorded in cells exposed to antisense oligodeoxynucleotides showed a parallel displacement to smaller values. Fig 1E⇓ shows the current-time relation when mean currents in cells exposed to antisense oligodeoxynucleotides were subtracted from corresponding currents recorded in control cells. The subtracted current has the rapid activation kinetics and minimal inactivation over 400 milliseconds that is typical of IKur.2 Results obtained by subtracting mean currents in cells exposed to mismatch oligodeoxynucleotides from those in control cells are shown in Fig 1F⇓ and indicate that exposure to mismatch oligodeoxynucleotides did not significantly affect currents recorded with this protocol.
For each of the cells studied, we calculated the amplitude of Ito (difference between peak and end-pulse current upon depolarization from −80 to +50 mV) and the amplitude of sustained end-pulse current relative to the zero current level (Isus, an index of IKur2 ). The results of this analysis are shown in Fig 2⇓ (left panels). Antisense oligodeoxynucleotides (5 μmol/L) significantly reduced the amplitude of Isus (Fig 2A⇓) but had no effect on Ito (Fig 2B⇓). Mismatch oligodeoxynucleotides did not affect either Isus or IKur. The same approaches described above were used to study a lower concentration (2 μmol/L) of combined oligodeoxynucleotides. Results were obtained for 23 cells under each condition (Fig 2C⇓ and 2D⇓) and were qualitatively the same as those with the higher oligodeoxynucleotide concentration shown in Figs 1⇑, 2A⇓, and 2B⇓. These data indicate that antisense (but not mismatch) oligodeoxynucleotides significantly inhibited Isus without altering Ito.
Currents recorded in cultured human atrial cells upon depolarization from a holding potential of −80 mV typically contain both Ito and IKur. In order to evaluate directly the changes in IKur, the latter was isolated with the use of a holding potential of −50 mV and a depolarizing prepulse (100 milliseconds) to +40 mV at 10 milliseconds before a 180-millisecond test pulse, as previously reported.2 A 120-millisecond repolarization to −20 mV was applied after the test pulse in order to record IKur tail currents. As in the left panels of Fig 1⇑, the left panels of Fig 3⇓ show recordings from multiple cells in each group randomly selected from all 51 cells per group exposed to 5 μmol/L combined oligodeoxynucleotides studied with this protocol. Results from 18 control cells are shown in Fig 3A⇓. The rapid activation, lack of inactivation, and small tail currents typical of IKur are clear. Currents are substantially smaller in 19 cells exposed to antisense (Fig 3B⇓) and not obviously different from control in 18 cells exposed to mismatch oligodeoxynucleotides (Fig 3C⇓). Because of their relatively small amplitude, IKur tail currents are hard to observe in the left panels of Fig 3⇓. The right panels show tail currents from the same cells and protocol as in the left panels, but on an expanded current and time scale. Tail currents were recorded in all vehicle-exposed control cells (Fig 3D⇓), albeit with varying amplitude. Cells exposed to antisense oligodeoxynucleotides similarly showed tail currents (Fig 3E⇓), but of diminished size. Tail currents in cells exposed to mismatch oligodeoxynucleotides (Fig 3F⇓) resembled those in control cells exposed to vehicle (Fig 3D⇓).
Average current-time relations recorded with the use of the protocol in Fig 3⇑ in all 51 cells from each group are shown in Fig 4A⇓. Cells exposed to vehicle and mismatch oligodeoxynucleotides have similar current-time relations, but currents are substantially reduced in antisense-exposed cells. Antisense-sensitive mean currents obtained by digital subtraction are shown in Fig 4B⇓ and indicate that the current inhibited by antisense is rapidly activating, shows little inactivation, and has small tail currents upon repolarization. Fig 4C⇓ shows mean±SEM values of IKur step current in each group of cells exposed to 5 μmol/L combined oligodeoxynucleo-tides or vehicle. Antisense oligodeoxynucleotides significantly reduced IKur, whereas mismatch oligodeoxynucleotides had no effect. Mean IKur tail currents from all 51 cells in each group are shown on an expanded current and time scale in Fig 4D⇓ and show that antisense oligodeoxynucleotides inhibited tail current without altering its overall kinetics. This point is further made by the average current obtained by digitally subtracting mean tail currents from cells exposed to antisense from those incubated in vehicle (Fig 4E⇓). Overall mean±SEM tail current amplitudes from each group are shown in Fig 4F⇓ and show that IKur tail currents were significantly smaller in cells exposed to antisense than in the other two groups.
Fig 5⇓ illustrates the voltage dependence of Isus and Ito in the three groups of cells used to study the effects of combined 5 μmol/L oligodeoxynucleotides. Currents were elicited by 400-millisecond depolarizing pulses from −80 mV at 0.1 Hz. Isus, reflecting IKur, was significantly reduced in cells exposed to antisense at all voltages positive to 0 mV and was not significantly different at any voltage in cells exposed to vehicle compared with those exposed to mismatch oligodeoxynucleotides (Fig 5A⇓). In contrast, Ito was the same at all voltages in all groups (Fig 5B⇓).
As an additional test of the effects of antisense on IKur, we determined the density of 50 μmol/L 4-AP–sensitive current in five cells from each group used to study the effects of 5 μmol/L combined oligodeoxynucleotides. Cells were depolarized from a holding potential of −80 to +50 mV before and after superfusion of 4-AP, at a concentration (50 μmol/L) that we have previously shown specifically inhibits IKur in human atrial myocytes.2 18 Digital subtraction of recordings in the presence of 4-AP from those before 4-AP was used to obtain 4-AP–sensitive currents, which were then normalized to cell capacitance to obtain 4-AP–sensitive current density. The latter averaged 6.8±0.8 pA/pF in control vehicle-treated cells, 3.1±0.4 pA/pF in antisense-treated cells, and 6.5±0.7 pA/pF in cells exposed to mismatch oligodeoxynucleotides (P=NS for control versus mismatch oligodeoxynucleotides; P<.01 for antisense versus the other two groups).
Studies of Atrial Myocytes Exposed to Single Oligodeoxynucleotides
Because antisense oligo A contains four sequential guanine bases, a sequence that can have nonspecific effects, we incubated groups of cells with only antisense oligo B, its corresponding mismatch oligodeoxynucleotide (mismatch oligo B), or vehicle. Cells exposed to 5 μmol/L antisense oligo B (n=25) showed significant decreases in IKur when compared with the same number of cells exposed to mismatch oligo B or vehicle control cells. Fig 6A⇓ shows mean amplitudes of IKur tail current, measured with the use of the protocol illustrated in Fig 3⇑. Tail currents were ≈50% smaller in cells exposed to antisense compared with the other two groups. In contrast, Ito was not altered by exposure to antisense (Fig 6B⇓, n=25 cells per group).
Studies of Ventricular Myocytes
Fig 7A⇓ shows typical currents recorded upon depolarization to various voltages from a holding potential of −80 mV in a cultured human ventricular myocyte exposed to vehicle. A sizable Ito is observed, along with small sustained currents at the end of a depolarizing pulse. Exposure to 5 μmol/L combined antisense oligo A and B (Fig 7B⇓) or equivalent concentrations of mismatch oligo A and B (Fig 7C⇓) did not appreciably alter current recordings. Mean data for current densities in 13 cultured human ventricular myocytes exposed to antisense, 14 ventricular myocytes exposed to mismatch oligodeoxynucleotides, and 14 vehicle controls are shown in Fig 7D⇓ (for Isus) and Fig 7E⇓ (for Ito). In contrast to results in human atrial myocytes, antisense had no measurable effects on Isus density in ventricular cells.
In the present study, we have shown that antisense oligodeoxynucleotides directed against the Kv1.5 coding sequence specifically inhibit IKur in adult human atrial myocytes. To our knowledge, this is the first direct demonstration with the use of antisense techniques that a cloned K+ channel subunit underlies a native cardiac K+ current and is certainly the first such demonstration for a human cardiac K+ current.
Relevance to Understanding the Molecular Basis of Human Cardiac Electrophysiology
The first two time-dependent K+ channel subunits to be cloned from the human heart were initially designated HK1 and HK29 and correspond to Kv1.4 and Kv1.5 channels according to the present convention.19 More recently, clones designated HERG and minK or IsK have been discovered in human hearts.20 21 Functional similarities have been demonstrated between currents carried by channels encoded by human Kv1.4, Kv1.5, HERG, and IsK and cardiac Ito, IKur, IKr, and IKs, respectively.2 3 4 8 10 22 23 Despite these functional similarities, discrepancies remain, such as the different recovery time courses of Kv1.4 and human cardiac Ito10 24 and the apparent differences in inactivation properties between Kv1.5 and IKur.2 4 A variety of criteria have been suggested by Tamkun et al1 for identifying a native current definitively with a cloned channel subunit. These include (1) similar biophysical properties, (2) similar pharmacological responses, (3) immunohistochemical evidence for the presence in the membrane of protein encoded by the putative cDNA clone, (4) an ability to suppress channel function with isoform-specific antibodies to channel subunit protein, (5) affinity purification to confirm protein composition in terms of accessory subunits and heterotetramer formation, and (6) elimination of the function of the macroscopic current by deletion of the cloned channel gene in a transgenic model. With respect to the relationship between human IKur and Kv1.5, electrophysiological evidence to support criteria 1 and 2 has been obtained.2 3 4 5 6 25 A recent publication indicates the presence of Kv1.5-encoded protein in human cardiac cell membranes on the basis of immunohistochemical techniques, thus satisfying the third criterion.7 Physiological evidence of the type envisioned by criteria 4 and 6 has not been obtained: indeed, elimination of IKur by creating transgenic humans with a deleted Kv1.5 gene (criterion 6) would seem to be unfeasible. The data in the present study, however, speak to the objectives of criteria 4 and 6, by demonstrating that antisense to Kv1.5 inhibited strongly IKur expression in cultured human atrial myocytes. The lack of effect of Kv1.5 antisense on sustained current in ventricular myocytes is consistent with previous electrophysiological data suggesting that IKur is absent in the human ventricle.8 26 27
The molecular basis for Ito in human atrium is still uncertain. Although channels formed by Kv1.4 subunits have many properties like those of human cardiac Ito,10 their recovery from inactivation is two orders of magnitude slower than that of the native current.24 Therefore, it has been suggested that slowly inactivating channel subunits like Kv1.5 may be involved in carrying Ito, either via heterotetramer formation with inactivating channel subunits11 or by association with a β-peptide.12 13 14 15 The present work argues against an essential role for Kv1.5 channel subunits in human cardiac Ito and is consistent with recent findings that Ito in the human heart may be carried by Kv4.3 channel subunits.28
In a recent study, Yang et al29 showed that antisense oligodeoxynucleotides directed against minK mRNA inhibits IKr in AT-1 cells, without changing levels of minK mRNA. The authors hypothesized either coassembly of minK with another protein (such as HERG) or an indirect role (such as anchoring of the channel protein complex to the cell surface) to explain their results. Given the strong similarities between currents expressed by Kv1.5 and IKur, the present results are much more likely to be due to direct inhibition of production of the protein encoded by Kv1.5, which underlies the IKur channel.
It is possible that the expression of ion channels is altered in culture, and the extrapolation of results in a cultured system to ionic currents in vivo must be tempered in this light. The voltage and time dependence of IKur in culture are stable over time, as is its 4-AP sensitivity, and resemble those of freshly isolated cells.18 Our results can thus be related to the molecular basis for IKur in vivo with confidence. On the other hand, although Ito in cultured cells is stable over time, the voltage- and time-dependent inactivation of Ito in cultured human atrial myocytes is different from that in fresh cells.18
We did not observe full suppression of IKur, even after 48 hours of exposure to antisense. These findings resemble previous observations with Kv1.5 antisense in a pituitary cell line.17 Possible explanations include a long half-life of Kv1.5 mRNA and/or protein, cellular mechanisms that permit some gene expression even in the presence of antisense, and a role for channel subunits from another gene in macroscopic IKur.
It would have been highly desirable to document changes in mRNA and/or Kv1.5 protein expression in the atrial cells exposed to antisense oligodeoxynucleo-tides. Unfortunately, given the small size of atrial samples, the low yield of cells with the “chunk” dissociation technique, and cell loss during solution changes in culture, the number of cells available at the end of the antisense exposure protocol was too small (rarely >30 cells) to permit quantitative detection of mRNA or protein levels. Because the techniques available to us for quantification of Kv1.5 mRNA and protein concentrations require a much larger amount of tissue, we did not attempt to apply these techniques to the antisense-treated or control cells. The specificity of the response to antisense is supported by the lack of effect of mismatch oligodeoxynucleotides, the lack of antisense-induced change in Ito, and the lack of effect on currents in ventricular myocytes.
There was considerable intercell variability in current amplitudes and Ito kinetics (Fig 1⇑, left; Fig 3⇑, left). Because of this variability, we showed data for a large sample of individual cells as well as mean data. The underlying basis for this variable expression of K+ currents is an interesting issue that goes beyond the scope of the present study but must be considered in interpreting our observations.
Selected Abbreviations and Acronyms
|HERG||=||human ether-a-go-go–related gene|
|IKr, IKs||=||rapidly and slowly activating components of the delayed rectifier K+ current|
|IKur||=||ultrarapid delayed rectifier K+ current|
|Ito||=||transient outward current|
This study was supported by the Medical Research Council of Canada (Dr Nattel), the Quebec Heart Foundation (Dr Nattel), the American Heart Association, Northeast Ohio Affiliate (Dr Wible), and the Fonds de Recherche de l’Institut de Cardiologie de Montréal (FRICM, Dr Nattel). Dr Li is a research scholar of the Fonds de la recherche en santé du Québec. The authors thank Johanne Doucet for excellent technical help and Luce Bégin for secretarial assistance with the manuscript.
- Received May 28, 1996.
- Accepted December 26, 1996.
- © 1997 American Heart Association, Inc.
Tamkun MM, Bennett PB, Snyders DJ. Cloning and expression of human cardiac potassium channels. In: Zipes DP, Jalife J, eds. Cardiac Electrophysiology. From Cell to Bedside. Philadelphia, Pa: WB Saunders Co; 1995:21-31.
Wang Z, Fermini B, Nattel S. Sustained depolarization-induced outward current in human atrial myocytes: evidence for a novel delayed rectifier K+ current similar to Kv1.5 cloned channel currents. Circ Res. 1993;73:1061-1076.
Fedida D, Wible B, Wang Z, Fermini B, Faust F, Nattel S, Brown AM. Identity of a novel delayed rectifier current from human heart with a cloned K+ channel current. Circ Res. 1993;73:210-216.
Snyders DJ, Tamkun MM, Bennett PB. A rapidly activating and slowly inactivating potassium channel cloned from human heart. J Gen Physiol. 1993;101:513-543.
Wang Z, Fermini B, Nattel S. Effects of flecainide, quinidine and 4-aminopyridine on transient and sustained outward currents in human atrial myocytes. J Pharmacol Exp Ther. 1995;272:184-196.
Rampe D, Wang Z, Fermini B, Wible B, Dage RC, Nattel S. Voltage- and time-dependent block by perhexiline of K+ currents in human atrium and in cells expressing a Kv1.5 type cloned channel. J Pharmacol Exp Ther. 1995;274:444-449.
Mays DJ, Foose JM, Philipson LH, Tamkun MM. Localization of the Kv1.5 K+ channel protein in explanted cardiac tissue. J Clin Invest. 1995;96:282-292.
Li G-R, Feng J, Yue L, Carrier M, Nattel S. Evidence for two components of delayed rectifier K+ current in human ventricular myocytes. Circ Res. 1996;78:689-696.
Tamkun MM, Knoth KM, Walbridge JA, Kroemer H, Roden DM, Glover DM. Molecular cloning and characterization of two voltage-gated K+ channel cDNAs from human ventricle. FASEB J. 1991;5:331-337.
Po S, Snyders DJ, Baker R, Tamkun MM, Bennett PB. Functional expression of an inactivating potassium channel cloned from human heart. Circ Res. 1992;71:732-736.
Po S, Roberds S, Snyders DJ, Tamkun MM, Bennett PB. Heteromultimeric assembly of human potassium channels: molecular basis of a transient outward current? Circ Res. 1993;72:1326-1336.
Pongs O. Regulation of the activity of voltage-gated potassium channels by β subunits. Neurosciences. 1995;7:137-146.
England SK, Uebele VN, Shear H, Kodali J, Bennett PB, Tamkun MM. Characterization of a voltage-gated K+ channel β subunit expressed in human heart. Proc Natl Acad Sci U S A. 1995;92:6309-6313.
Morales MJ, Castellino RC, Crews AL, Rasmusson RL, Strauss HC. A novel β subunit increases rate of inactivation of specific voltage-gated potassium channel α subunits. J Biol Chem. 1995;270:6272-6277.
Stein CA, Chen Y-C. Antisense oligonucleotides as therapeutic agents: Is the bullet really magical? Science. 1993;261:1004-1012.
Chung S, Saal DB, Kaczmarek LK. Elimination of potassium channel expression by antisense oligonucleotides in a pituitary cell line. Proc Natl Acad Sci U S A. 1995;92:5955-5959.
Feng J, Li G-R, Fermini B, Nattel S. The properties of sodium and potassium currents of cultured adult human atrial myocytes. Am J Physiol. 1996;270:H1676-H1686.
Chandy KG. Simplified gene nomenclature. Nature. 1991;351:26. Letter.
Swanson R, Folander K, Antanavage J, Smith J. The IsK gene is expressed in human heart. Biophys J. 1991;59:452a. Abstract.
Freeman LC, Kass RS. Expression of a minimal K+ channel protein in mammalian cells and immunolocalization in guinea pig heart. Circ Res. 1993;73:968-973.
Fermini B, Wang Z, Duan D, Nattel S. Differences in rate dependence of transient outward current in rabbit and human atrium. Am J Physiol. 1992;263:H1747-H1754.
Snyders DJ, Knoth KM, Roberds SL, Tamkun MM. Time-, voltage-, and state-dependent block by quinidine of a cloned human cardiac potassium channel. Mol Pharmacol. 1991;41:322-330.
Konarzewska H, Peeters GA, Sanguinetti MC. Repolarizing K+ currents in nonfailing human hearts: similarities between right septal subendocardial and left subepicardial ventricular myocytes. Circulation. 1995;92:1179-1187.
Amos GJ, Wettwer E, Metzger F, Li Q, Himmel HM, Ravens U. Differences between outward currents of human atrial and subepicardial ventricular myocytes. J Physiol (Lond). 1996;4911:31-50.
Dixon JE, Shi W, Wang H-S, McDonald C, Yu H, Wymore RS, Cohen IS, McKinnon D. Role of the Kv4.3 K+ channel in ventricular muscle: a molecular correlate for the transient outward current. Circ Res. 1996;79:659-668.
Yang T, Kupershmidt S, Roden DM. Anti-minK antisense decreases the amplitude of the rapidly activating cardiac delayed rectifier K+ current. Circ Res. 1995;77:1246-1253.