Tissue and Species Distribution of mRNA for the ikr-like K+ Channel, ERG
The human K+ channel gene, HERG, has been linked to the type 2 form of the autosomal dominant long-QT syndrome and has been suggested to encode the fast component of the delayed rectifier K+ current (IKr) found in heart. To date, the published electrophysiological and pharmacological data on the Xenopus-expressed HERG are very similar but are not identical to those of the endogenous IKr. In an effort to provide a different type of correlative data on the relationship between erg and IKr, cDNA fragments of erg homologues from guinea pig, rabbit, human, dog, and rat were cloned and used to test for the presence of erg mRNA in cardiac tissue. RNase protection assays reveal that erg message is found in the hearts of all five species and that it is expressed uniformly throughout the heart. The erg transcript is expressed at relatively high levels, being ≈50% more abundant than the most prevalent Kv-class K+ channel transcript in canine ventricle (Kv4.3). Erg transcripts were found to have a wide tissue distribution in rat and are abundant in the brain, retina, thymus, and adrenal gland and are also found in skeletal muscle, lung, and cornea. Since there were no published reports of an IKr-like current in the rat heart, electrophysiological studies were performed to test whether the significant level of erg message in rat heart was correlated with the presence of an IKr-like current in rat. In isolated rat ventricular myocytes, an E-4031–sensitive current was observed, which is consistent with the presence of IKr. These results strengthen the link between erg and the native IKr in heart and suggest that erg may play an important role in other noncardiac tissues.
The cardiac action potential is generated by a shifting balance of inward and outward currents. Whereas depolarization is caused by an influx of either Na+ or Ca2+ ions, depending on the tissue type, repolarization ensues when an activating efflux of K+ ions exceeds the inactivating inward currents. In most atrial muscle, a significant portion of this repolarization is produced by a rapidly activating and inactivating K+ current known as Ito.1 2 In the ventricular cardiac tissue types (endocardium, epicardium, and Purkinje) of some species, however, only phase 1 repolarization is due to a similar Ito, whereas final repolarization is dependent on the magnitude and kinetics of the delayed rectifier, IK.3 4 This delayed rectifier has been separated, on the basis of biophysical and pharmacological experiments, into both rapidly and slowly activating components, known as IKr and IKs, respectively.5 6 7
Until recently, no candidate gene existed for the rapidly activating component of the delayed rectifier. Two laboratories have now heterologously expressed HERG, the human homologue of the K+ channel gene erg,8 in Xenopus oocytes, and have shown that it has properties similar to those of IKr, suggesting that the erg gene encodes IKr.9 10 That this gene plays a role in cardiac repolarization seems probable, as defects in HERG have been previously linked to electrophysiological abnormalities associated with one form of the long-QT syndrome.11
In an attempt to further test the hypothesis that erg encodes IKr, we identified erg message in cardiac tissue from different species (guinea pig, rabbit, canine, human, and rat) as well as different cardiac tissue types in the same species. If erg does encode IKr then erg mRNA should be prevalent in all tissues that express significant IKr. Conversely, it would be expected that there should be minimal (if any) erg mRNA in species, or tissues, that have no detectable delayed rectifier.
Materials and Methods
Cloning of ERG Probes from Rabbit, Guinea Pig, Dog, Human, and Rat
It is possible to use highly conserved probes in Northern blot analyses across various species; this is not, however, the case with probes for RNase protection assays. Because of the intrinsic specificity of this assay system, it is necessary to obtain probes for each species to be examined. To obtain erg homologues from each of the five species in question, PCR amplification was used. Degenerate oligonucleotide primers were designed to correspond with amino acid sequences that are conserved across the entire known eag/erg superfamily.8 It was necessary to use regions conserved throughout the whole eag family, as there was only one published erg sequence.
The upstream primer was directed toward part of the S1 transmembrane segment encompassing the amino acid sequence FKAVWDW; the oligonucleotide sequence was 5′-TT(T/C) AA(A/G) (A/G)C(A/C/G/T) (A/G)(T/C)(A/C/G/T) TGG GA(T/C) TGG-3′. The downstream primer was directed to the S5 transmembrane region encompassing the amino acid sequence WKACIWY; the oligonucleotide sequence was 5′-(A/G)TA CCA (A/G/T)AT (A/G)CA (A/C/G/T)AG CCA (A/G)TG-3′.
Heart RNA was reverse-transcribed, and the subsequent cDNA was used as a template for the PCR reactions. The above two primers were used successfully for PCR amplification of erg homologues from each of the five species, using either TAQ polymerase in dimethyl sulfoxide containing buffer (guinea pig, dog, rabbit, and human, Perkin-Elmer Corp) or the Expand system (rat, Boehringer-Mannheim). The ≈480-bp PCR products were isolated by agarose gel electrophoresis, purified using the Glass-Max system (GIBCO BRL), then cloned into pBluescript (Stratagene), and transformed into Xl-1 blue Escherichia coli (Stratagene). Insert-bearing individual clones were identified, and the plasmid DNA was purified using the Wizard mini-prep system (Promega) and then sequenced with Sequenase (USB).
Preparation of RNA
Total RNA was prepared by quick freezing of the tissues isolated from adult animals in liquid N2, followed by homogenization in guanidinium thiocyanate. RNA was prepared essentially as described by Chirgwin et al12 and included a final step of pelleting the homogenate over a CsCl step gradient. Canine ventricular samples were dissected as a section of tissue across the width of the left ventricular free wall. Canine right atria were dissected as a section of tissue through the atrial wall. Free-running canine Purkinje fibers were dissected from both ventricles. Rabbit ventricle was dissected as a section of tissue through the left ventricle. Rabbit atrium was dissected as a section of tissue through the right atrium. Rabbit SA nodes were isolated from 10 rabbits. The rat and guinea pig tissues were homogenized after dissection and were not frozen. Rat and guinea pig ventricles were taken as left ventricular sections through the tissue and near the apex of the heart. Rat tissues (kidney, adrenal gland, thymus, cornea, lens, and retina) were isolated as whole organs and homogenized. Approximately one half of a gram of tissue was isolated from lung, liver, and small intestine for the homogenization. Both rat and guinea pig brains were whole hemispheric dissections. Human heart and brain RNA were mRNA preparations obtained commercially from Clontech. Poly-A+ RNA was isolated from adult animals using paramagnetic poly-dT beads (Dynal Inc); after isolation, canine Purkinje fibers were homogenized fresh in reagents supplied with the Dynal kit. Rabbit SA node tissue was homogenized frozen, and the poly-A+ RNA was extracted as described using the Dynal mRNA extraction kit.
RNase Protection Assays
RNA probes were prepared as described previously.13 In all experiments, a significant stretch of nonhybridizing sequence of ≈50 bp was included with the probe to allow easy distinction between the probe and the specific protected band. The specificity of the assay is such that there was no evidence for cross-reactivity between the probe and any other transcripts, with the possible exception of the guinea pig heart experiments. The guinea pig erg probe protects a predominant band in both brain and ventricle of the size expected; however, there are numerous smaller protected bands that appear primarily in the heart sample. This was observed with two different probes in two different RNA preparations (one total RNA and one poly-A+ ); it is unclear whether the probe is cross-reacting with other closely related genes or with alternative splices of the guinea pig erg gene.
RNase protection assays were performed essentially as described previously.13 For each experiment, 5 μg of total or 1 μg of poly-A+ RNA was used. Cyclophilin probes were included in the hybridization reaction to confirm that the sample was not lost during the course of the experiment. Rat and canine cyclophilin probes were PCR-cloned, using available sequence data to design primers for the PCR. The human cyclophilin probe was purchased from Ambion. In the guinea pig and rabbit experiments, the rat cyclophilin probe was used. The probe protected fragments smaller than the full-length band found in rat RNA. The same protected fragments appeared replicably, regardless of the tissue used. Five micrograms of yeast tRNA was used as a negative control for probe self-protection bands. The rat RNase protection assay figures are 2-day exposures; all of the other figures were generated from 1-day exposures.
For the comparison of erg and Kv4.3 mRNA levels in canine ventricle, specific signals were quantified directly from dried RNase protection gels using a PhosphorImager (Molecular Dynamics). In this experiment, three independent samples of RNA were used, and the erg and Kv4.3 samples were assayed separately so that no erg signal would contaminate the Kv4.3 protected band. Cloning and preparation of the Kv4.3 probe has been described previously.14
Myocyte Isolation Procedures
Adult rat ventricular myocytes were isolated according to techniques reported previously.15 Briefly, hearts were cannulated under sterile conditions and superfused with a Ca2+-free solution containing collagenase (CLS 2, Worthington) and hyaluronidase (type II, Sigma Chemical Co). Subsequently, the hearts were minced and digested with trypsin IX (Sigma) and deoxyribonuclease II (Worthington). Myocytes were filtered through nylon mesh, sedimented through Ca2+-containing Krebs-Henseleit bicarbonate buffer solution supplemented with albumin (bovine fraction V, Sigma), and stored in culture medium (medium 199) at room temperature. Myocytes were examined within 12 hours of harvest.
Myocytes were allowed to settle on the bottom of a Peltier-based temperature-controlled perfusion bath mounted on an inverted microscope and studied as described previously.16 Cells were superfused (0.8 mL/min) with a HEPES-buffered Tyrode's solution containing (mmol/L) NaCl 132, MgSO4 1.2, HEPES 20, glucose 11.1, KCl 4, and CaCl2 2 (pH 7.4 with HCl; bath temperature, 37°C).
Myocytes were accessed using whole-cell patch-clamp techniques and an Axopatch 200A amplifier (Axon Instruments). Patch pipettes were filled with a high-[K+] intracellular solution containing (mmol/L) potassium aspartate 125, KCl 20, EGTA 10, ATP (magnesium salt) 5, MgCl2 1, and HEPES-free acid 5, adjusted to 7.3 with 5N KOH. Junction potentials were compensated by offset adjustment of −10 mV immediately after cell break-in. Series resistance compensation was typically adjusted to values between 60% and 70%. Capacitance values of myocytes were determined from settings on the patch-clamp amplifier obtained after clamp tune-up at a holding potential of ≈−40 mV.
Extracellular solution contained nimodipine (4 to 5 μmol/L) to block L-type Ca2+ current; Na+ current was partially inactivated by holding potential (−75 mV) and further reduced by the use of slowly depolarizing ramp pulses to facilitate accommodation. E-4031 was added from a 5 mmol/L aqueous stock solution, for a final concentration of 5 μmol/L; this concentration has been shown to block nearly all of IKr in isolated ventricular myocytes from guinea pig (IC50, 0.4 μmol/L),6 dog,3 and neonatal mouse.17 Dofetilide was prepared from a 5 mmol/L stock solution dissolved in ethyl alcohol and used at a final concentration of 1 μmol/L. Dofetilide has been shown to selectively block IKr in rabbit myocytes (Kd, 3.9×10−9 mol/L).18
Preliminary experiments demonstrated that it was difficult to measure IKr in adult rat ventricular myocytes during depolarizing square test pulses because of the coincident activation of large rapidly activating transient and sustained outward currents and the lack of an agent to block these currents without blocking IK. Furthermore, the small amplitude of IKr tail currents upon repolarization to −40 mV suggested that IKr would be exceedingly small during strongly depolarizing test pulses, given the inwardly rectifying characteristics of this current in other species. Consequently, we used a three-phase clamp protocol to assess IKr consisting of (1) a slowly depolarizing ramp from a holding potential of −75 mV to +25 mV (duration of either 1 second or 600 milliseconds), followed by (2) a 200-millisecond “pedestal” step to 0 mV, followed by (3) a test pulse to −40 mV to elicit tail currents. The IKr was identified as a time-dependent drug-sensitive tail current recorded upon repolarization to −40 mV. The amplitude of IKr was assessed by measuring the amplitude of E-4031–sensitive or dofetilide-sensitive time-dependent current obtained by digital subtraction (ie, control minus drug). Recordings within a brief 5- to 6-minute period were compared to minimize possible nonspecific changes in membrane currents. Because of the small amplitude of the tail currents, averages from a minimum of three recordings in the absence and presence of drug were used for digital subtractions. The clamp protocols were repeated once every 10 seconds.
Cloning of RNase Protection Assay Probes
The RNase protection assay provides a means for determining the presence of mRNA transcripts that is both more specific and more sensitive than Northern blot analysis. To perform such experiments requires the use of a probe that is specific not only for the gene of interest but also for the species in question. It was necessary, therefore, to generate fragments of erg DNA from each of the five species examined in the present study. Degenerate oligonucleotide primers were designed that were conserved across the entire eag/erg superfamily and were used in the PCR, along with cDNA made from ventricular RNA, to generate ≈480-bp erg clones. Each of these clones was sequenced and compared with the published human sequence.8
An alignment of the sequences is shown in Fig 1⇓. These probes encompass most of the hydrophobic core of the channel from the beginning of the first transmembrane segment (S1) through the end of the fifth transmembrane region (S5). The canine and rabbit clones are 99% identical to the human sequence at the amino acid level; the rat and guinea pig clones are 96% identical to the human sequence. The main area of variability is in the S1-S2 extracellular loop; this same lack of conservation is found across the entire eag/erg superfamily at this position. The S4 voltage-sensing domain contains absolute identity across the various species.
There was no evidence for alternative splicing within the region encompassed by these clones. Amino- or carboxy-terminal splicing variants would not have been detected using these probes. No other eag-related genes were detected by PCR in the hearts of the species examined, even though these primers were able to amplify related cDNAs in other tissues.
Species Distribution of ERG mRNA in Heart
There have been published reports of IKr in the hearts of several species. Guinea pig, rabbit, dog, and human hearts have been shown to have significant E-4031–sensitive IK.3 7 19 20 To our knowledge, there is no previous report of this current in the rat. Since it has been proposed that the erg K+ channel gene encodes a subunit of the IKr channel, we examined the amount of erg mRNA in cardiac tissues from each of these species to determine whether message levels paralleled the presence of IKr.
Fig 2⇓ shows the results of RNase protection assays performed on RNA from the hearts of guinea pig, rabbit, dog, human, and rat, using the erg probes described above. The RNA used in all sample lanes is from the left ventricle, except for human RNA, which was whole-heart RNA (Clontech). A strong erg-protected signal is observed in the heart RNA from all of the species tested, including rat ventricle.
Distribution of ERG mRNA in Rabbit and Dog Hearts
IKr has been well characterized in rabbit SA node, atrium, and ventricle.19 21 Fig 3A⇓ shows that erg mRNA is approximately equally abundant in left ventricle, left atrium, and SA node of the rabbit heart.
IKr has also been well characterized in canine ventricle.3 4 RNase protection experiments were performed to determine whether erg mRNA was present in other regions of canine heart. Fig 3B⇑ reveals that there are similar levels of erg transcripts in the left ventricle, Purkinje fibers, and right ventricle. These results show that there is relatively uniform erg mRNA expression within the various specialized tissues of rabbit and dog hearts.
Tissue Distribution of ERG mRNA
In contrast to the characterizations of IKr in the hearts of several species, no reports could be found suggesting that an E-4031–sensitive IK exists in noncardiac tissues. A variety of tissues from rats was examined for the presence of erg mRNA. Fig 4⇓ indicates that the strongest signal (per microgram of RNA) is found in whole brain, but strong signals are also observed in RNA from left ventricle, adrenal gland, thymus, and retina; weaker erg-protected bands are seen in soleus, tibialis, lung, cornea, and lens. Although not visible in the autoradiographs shown, after a week-long exposure, a very faint erg-protected band is visible in kidney, liver, and small intestine.
Since rat brain had an abundance of erg mRNA, we examined two other species to determine the generality of this observation. Erg mRNA was abundantly expressed in human brain, similar to what was found in rat brain (Fig 5B⇓). Surprisingly, however, erg mRNA was significantly less abundant in guinea pig brain, suggesting marked species differences (Fig 5A⇓).
Comparison of ERG and Kv4.3 mRNA in Canine Heart
Kv4.3, which encodes the primary α subunit of Ito in canine and human heart, has previously been shown to be the most abundant Kv-class K+ channel transcript expressed in canine heart.14 In preliminary experiments, we observed that erg transcript levels in canine heart appeared to be quite high relative to Kv4.3. Using quantitative RNase protection experiments, we directly compared the levels of erg mRNA with Kv4.3 and found that erg mRNA levels are ≈1.5-fold more abundant than those of Kv4.3 in canine right ventricle (Fig 6⇓). The erg transcript appears to be the most prevalent K+ channel mRNA expressed in heart.
Delayed Rectifier in Rat Ventricle
The high level of erg message in rat heart was unexpected because IKr had not been previously reported in this species, although inferences have been made on the basis of the effect of class III antiarrhythmics on fetal rats.22 23 Biophysical and pharmacological experiments were carried out to determine whether a corresponding IKr-like current is present in rat cardiac myocytes.
Fig 7⇓ displays superimposed averaged current traces obtained from a typical adult rat ventricular myocyte in the absence (open circles) and presence (closed circles) of 5 μmol/l E-4031. E-4031 rapidly and consistently blocked the slowly decaying outward current elicited upon repolarization to −40 mV. Using this clamp paradigm, the average peak current density was 0.19±0.04 pA/pF (mean±SEM, n=9 myocytes, obtained from four hearts). The same protocol applied to two canine ventricular myocytes also revealed an E-4031–sensitive decaying tail current at −40 mV (authors' unpublished data, 1996), consistent with the previous demonstration of IKr in canine heart.3 4 E-4031 had no consistent effect on membrane currents during the depolarizing ramps (assessed at −20, 0, and +20 mV). Although averaged data showed a trend toward less outward current at the more positive ramp potentials, this did not achieve statistical significance (authors' unpublished data, 1996). Such an effect is consistent with the inwardly rectifying properties of IKr, which likely results from differences in the voltage- and time-dependent kinetics of activation versus fast inactivation.24
Dofetilide is another pharmacological agent that preferentially blocks IKr.18 To confirm the presence of IKr in rat myocytes, the effect of dofetilide was examined using the same protocol as in Fig 7⇑. The slow tail current component was abolished by 1 μmol/L dofetilide, consistent with the presence of IKr in rat myocytes; the amplitude of the dofetilide-sensitive tail current was 0.31±0.22 pA/pF (mean±SEM, n=6, obtained from two hearts).
IKr is also paradoxically reduced by lowering [K+]o.3 25 To further characterize the E-4031–sensitive current in rat myocytes as IKr, the effects of lowered [K+]o on the E-4031–sensitive current were examined. After initial recording in 4 mmol/L [K+]o, the solution was switched to one containing nominally 0 mmol/L [K+]o. This produced a reduction in the amplitude of the time-dependent tail current upon repolarization to −40 mV. The effects of E-4031 were then evaluated in the presence of 0 mmol/L [K+]o. In each of four myocytes tested, a time-dependent E-4031–sensitive current was negligible (0.026±0.005 pA/pF, mean±SEM), consistent with the near absence of IKr. under these recording conditions.
IKr has several unique properties, which include a sensitivity to E-4031, dependence on external K+, and inward rectification.7 None of the previously cloned voltage-gated(Kv), or inward rectifier (Kir)K+ channels strongly resembles IKr. When HERG was heterologously expressed in oocytes, it was clear that this channel was a candidate for IKr; it activated with increasing depolarization, showed marked voltage-dependent inward rectification, and had strong dependence on [K+]o. However, the activation kinetics of the expressed current were slower than the native IKr, and the sensitivity and characteristics of block by E-4031 differed as well from the cardiac myocyte IKr, causing some uncertainty.9 10 More recently, with modification of the protocols used on native myocytes, HERG expressed in oocytes has been shown to be blocked by the methanesulfonanilides initially used to characterize IKr, further supporting the hypothesis that IKr is encoded by erg.26 One purpose of the experiments described in the present study was to provide an alternative test of the hypothesis that erg encodes the native IKr in various cardiac tissues.
The abundant expression of erg message in the hearts of guinea pig, rabbit, dog, and human corresponds well with existing characterizations of the delayed rectifier.3 4 7 16 19 20 21 The presence of erg in all regions of the hearts examined is also consistent with reports of IKr in ventricle, atrium, and SA node in rabbit heart19 as well as ventricle and atrium in dog heart.3 4 Our results are also consistent with the finding that erg mRNA was detected in dispersed ventricle, SA node, and atrial myocytes isolated from ferret.27 The finding that erg mRNA is nearly as abundant in rat heart as in the other species examined was surprising in light of the fact that literature searches could identify no previous reports describing IK in rat heart, although there have been reports of class III antiarrhythmic effects in fetal rats.22 23 With the data from these RNase protection experiments in mind, we examined rat ventricular myocytes for the presence of delayed rectifier and found that the rat does indeed possess an IKr-like current, with properties similar to those described in other species.
Since IKr has been exclusively characterized as a cardiac channel, the original design of these experiments was to examine the erg mRNA content from noncardiac tissues in the expectation that there would be little expression outside the heart. Instead of limited expression, however, we found that erg mRNA is widely expressed in rat, and although it is not possible to directly compare RNA levels with functional currents, it would not be surprising to detect currents corresponding to erg mRNA in brain, adrenal glands, thymus, retina, and possibly skeletal muscle. The widespread distribution of erg gene expression suggests that pharmacological agents that suppress IKr in heart may produce unwanted effects in numerous other tissues and that careful consideration of these effects may be warranted when evaluating the pharmacological effects of these agents.
Northern blot analysis has shown that erg is expressed at low levels in the human brain.11 In contrast to this previous report, we have reproducibly detected an abundant erg message in the human brain, which is approximately equal in strength to that found in human heart. In every species we examined, erg is a ubiquitous heart channel; the presence of erg mRNA in the brain, however, was species variable. The guinea pig has considerably more erg mRNA in the ventricle than in the brain, in contrast to the rat, which, like the human, has high levels of erg mRNA levels in the brain.
Prior studies with ventricular myocytes have identified and characterized IKr as an inwardly rectifying K+ current blocked by the methanesulfonanilides E-4031 or dofetilide in ventricular cells from numerous species including dogs,3 16 guinea pigs,6 neonatal mice,17 28 and humans.20 The present study concludes that IKr is present in rat ventricular myocytes on the basis of (1) the presence of a slowly decaying outward tail current upon repolarization to −40 mV, (2) block of the tail current by E-4031 and dofetilide, (3) the reduction of tail current by superfusion with 0 mmol/L [K+]o, and (4) the lack of effect by E-4031 on tail currents in the presence of 0 mmol/L [K+]o. Together, these observations are consistent with the presence of a small IKr in adult rat ventricular myocytes sensitive to methanesulfonanilides. The small amplitude of IKr precluded more detailed studies of the characteristics of IKr in this species. Earlier reports of the absence of IKr in adult rat myocytes likely resulted from the small amplitude of this current relative to larger time-dependent currents. The small amplitude of IKr is also consistent with the lack of effect of dofetilide on the action potentials of rat papillary muscles.29
It has been shown that the Kv4.3 gene encodes the 4-aminopyridine–sensitive component of Ito in canine ventricle14 and that Kv4.3 transcripts are the most abundant Kv-class K+ transcripts expressed in heart. We have shown in the present study that the mRNA level for erg in canine ventricular myocardium actually exceeds that for Kv4.3 by ≈50%. It is difficult to meaningfully compare densities of different currents unless similar clamp protocols and experimental conditions are used. However, when identical disaggregation procedures and comparable internal and external solutions are used, it is evident that the density of Ito is much greater than that of IKr in canine ventricle. Specifically, the reported current densities for Ito in canine epicardium and midmyocardium are ≈30 pA/pF, declining to ≈5.6 pA/pF in canine endocardium.30 In contrast, the density of IKr in canine myocytes is only 0.22 pA/pF.3 Thus, despite the prevalence of erg mRNA over Kv4.3 mRNA, IKr current density is 25- to 136-fold smaller than that of Ito. This suggests that posttranscriptional events can contribute to determining the final amplitude of the erg currents in heart and points to limitations in comparing message levels with membrane currents in the absence of measurements of membrane protein levels. Although unlikely, it is possible that differences in the single-channel conductances of the Ito and IKr channels could account for some of the difference in current densities.
In conclusion, our experiments demonstrate that although erg is unlikely to encode a cardiac-specific channel, it is prevalent in all cardiac regions of all the species tested. The mechanism of posttranscriptional regulation of erg channel expression remains a fruitful area for further investigation, as does the identification of the role of the erg channel in the noncardiac tissues (brain, thymus, adrenal glands, and skeletal muscle) in which the erg message is expressed at significant levels.
Selected Abbreviations and Acronyms
|IK||=||delayed rectifier K+ current|
|IKr||=||rapidly activating component of IK|
|IKs||=||slowly activating component of IK|
|Ito||=||transient outward current|
|PCR||=||polymerase chain reaction|
This study was supported in part by National Institutes of Health grants HL-20558 and HL-28958 (Dr Cohen), NS-29755 (Dr McKinnon), and HL-49918 (Dr Gintant); by an American Heart Association Grant-in-Aid (Dr McKinnon); and by a SUNY, Stony Brook, pharmacology postdoctoral fellowship (T32-DK07521, Dr Wymore). Dofetilide was supplied courtesy of Pfizer Central Research. Sequences of the erg clones have been submitted to Genbank under the following accession numbers: U75210 (rat erg), U75211 (guinea pig erg), U75212 (rabbit erg), and U75213 (canine erg).
- Received September 16, 1996.
- Accepted November 27, 1996.
Gintant GA, Cohen IS, Datyner NB, Kline RP. Time dependent outward currents in the heart. In: Fozzard HA, Haber E, Katz AM, Morgan HE, eds. The Heart and Cardiovascular System. New York, NY: Raven Press Publishers; 1992.
Gintant GA. Regional Differences in Ik density in canine left ventricle: role of Ik,s in electrical heterogeneity. Am J Physiol. 1995;268:H604-H613.
Liu D-W, Antzelevitch C. Characteristics of the delayed rectifier current (IKr and IKs) in canine ventricular epicardial, midmyocardial, and endocardial myocytes: a weaker IKs contributes to the longer action potential of the M cell. Circ Res. 1995;76:351-365.
Sanguinetti MC, Jurkiewicz NH. Two components of cardiac delayed rectifier K+ current: differential sensitivity to block by class III antiarrhythmic agents. J Gen Physiol. 1990;96:195-215.
Sanguinetti MC, Jurkiewicz NH. Delayed rectifier outward K+ current: differential sensitivity to block by class III antiarrhythmic agents. Am J Physiol. 1991;260:H393-H399.
Warmke JW, Ganetzky B. A family of potassium channel genes related to eag in Drosophila and mammals. Proc Natl Acad Sci U S A. 1994;91:3428-3442.
Trudeau MC, Warmke JW, Ganetzky B, Robertson GA. HERG, a human inward rectifier in the voltage-gated potassium channel family. Science. 1995;269:92-95.
Chirgwin JM, Przybyla AE, MacDonald RJ, Rutter WJ. Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry. 1979;24:5294-5298.
Dixon JE, McKinnon D. Quantitative analysis of potassium channel mRNA expression in atrial and ventricular muscle of rats. Circ Res. 1994;75:252-260.
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.
Ellingsen O, Davidoff AJ, Prasad SK, Berger HJ, Springhorn JP, Marsh JD, Kelly RA, Smith TW. Adult rat ventricular myocytes cultured in defined medium: phenotype and electromechanical function. Am J Physiol. 1993;265:H747-H754.
Gintant GA. Two components of delayed rectifier current in canine atrium and ventricle: does IKs play a role in the reverse rate dependence of class III agents? Circ Res. 1996;78:26-37.
Carmeliet EE. Voltage- and time-dependent block of the delayed K+ current in cardiac myocytes by dofetilide. J Pharmacol Exp Ther. 1992;262:809-817.
Ito H, Ono K. A rapidly activating delayed rectifier K+ channel in rabbit sinoatrial node cells. Am J Physiol. 1995;269:H443-H452.
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.
Abrahamsson C, Palmer M, Ljung B, Duker G, Baarnhielm C, Carlsson L, Danielsson B. Induction of rhythm abnormalities in the fetal rat heart: a tentative mechanism for the embryotoxic effect of the class III antiarrhythmic agent almokalant. Cardiovasc Res. 1994;28:337-344.
Spector PS, Curran ME, Keating MT, Sanguinetti MC. Class III antiarrhythmic drugs block HERG, a human cardiac delayed rectifier K+ channel: open-channel block by methanesulfonanilides. Circ Res. 1996;78:499-503.
Brahmajothi MV, Morales MJ, Liu S, Rasmusson RL, Campbell DL, Strauss HC. In situ hybridization reveals extensive diversity of K+ channel mRNA in isolated ferret cardiac myocytes. Circ Res. 1996;78:1083-1089.
Wang L, Feng Z-P, Konodo CS, Sheldon RS, Duff HJ. Developmental changes in the delayed rectifier K+ channels in mouse heart. Circ Res. 1996;79:79-85.
Liu D-W, Gintant GA, Antzelevitch CA. Ionic bases for electrophysiological distinctions among epicardial, midmyocardial, and endocardial myocytes from the free wall of the canine left ventricle. Circ Res. 1993;72:671-687.