From the Department of Medicine (S.K., J. Duc, D.A., G.F.T.), Division of
Molecular and Cellular Cardiology, Johns Hopkins University, Baltimore,
Maryland; Department of Neurobiology and Behavior (J. Dixon, D.M.), State
University of New York at Stony Brook; Department of Medicine I (S.K., M.N.,
G.S.), Ludwig Maximilians University of Munich (Germany); and Department of
Medicine III (D.J.B.), University of Cologne (Germany).
Correspondence to Gordon Tomaselli, MD, 720 N Rutland Ave, Ross 844, Johns Hopkins University, Baltimore, MD 21205. E-mail gtomasel{at}welchlink.welch.jhu.edu
Methods and Results—We used ribonuclease protection assays and
whole-cell electrophysiological
recording to study changes in the level of Kv4.3 mRNA and
Ito in human tissues and isolated
ventricular myocytes, respectively. We found that the level
of Kv4.3 mRNA decreased by 30% in failing hearts compared with
nonfailing controls. Furthermore, this reduction correlated with the
reduction in peak Ito density measured in
ventricular myocytes isolated from adjacent regions of the
heart. There was no significant change in the steady-state level of any
other mRNA studied (HERG, Kv1.4, Kir2.1, Kvß1.3, and
the
Conclusions—These data provide further support for the hypothesis
that Kv4.3 encodes all or part of the native cardiac
Ito in humans and that part of the
downregulation of this current in heart failure may be
transcriptionally regulated.
Voltage-dependent K+ currents mediate
repolarization of cardiac myocytes and hence are critical determinants
of the action potential duration. Differential expression of a variety
of K+ currents may be important in defining
regional differences in the action potential profile in the
heart.10 11 12 Recently, the cellular and molecular
bases of action potential prolongation in failing human
myocardium were described. Terminal human heart failure is
associated with a functional downregulation of the
Ca2+-independent transient outward
K+ current (Ito) and
inward rectifier K+ current
(IK1).7 13 The
mechanism of the functional downregulation is unknown, but it may
involve altered transcription, translation, membrane trafficking,
subunit assembly, post-translational modification, degradation of
channel proteins, or a combination. We sought to determine the
mechanism of action potential prolongation in human heart failure by
quantifying the level of K+ channel gene
transcripts in normal and failing myocardium.
There are species-specific differences in voltage-dependent
K+ channel expression. For example, significant
levels of mRNA encoding the rapidly inactivating
K+ channels Kv1.4, Kv4.2, and Kv4.3 have been
observed in rat ventricle,14 whereas in canine
and human ventricle, there is no detectable Kv4.2, but mRNA encoding
Kv1.4, Kv1.5, Kv3.4, and Kv4.3 is present.15
Based on the biophysical properties and pharmacology of expressed Kv4.3
and its abundance in human ventricle, this gene has emerged as the
leading candidate for encoding Ito in both
dogs and humans.15 We measured
Ito in cells isolated from normal and
failing human myocardium and performed ribonuclease
protection assays (RPAs) on samples from the same hearts excised from
regions immediately adjacent to the sections used for cell isolation.
The level of Kv4.3 mRNA in ventricular
myocardium was compared with the
Ito density in isolated myocytes. To
determine whether Kv4.3 encodes all or part of
Ito in the human ventricle, we took
advantage of the known reduction in current density in myocytes
isolated from failing human ventricles compared with cells isolated
from nonfailing hearts. We measured a statistically significant
reduction in the level of Kv4.3 mRNA in failing ventricle compared with
nonfailing controls. The steady-state level of Kv4.3 mRNA correlated
with the density of Ito measured in cells
isolated from adjacent regions of the myocardium. These
data provide further support for the hypothesis that hKv4.3 encodes at
least part of the native Ito channel and
that the reduction in Ito in heart failure
is consistent with transcriptionally mediated regulation.
The standard nomenclature for K+ channel genes is
used throughout.16 For each control or target
transcript, probes were generated by PCR amplification of a region of
the cDNA as described in Table 1
Tissue and RNA Preparation
The tissue was quick-frozen in liquid nitrogen within 10 to 15 minutes
after tissue harvesting and stored at -80°C until further
processing. Total RNA was prepared either with TRIzol Reagent (GIBCO
BRL) according to the manufacturer's instructions or with
centrifugation through a CsCl
cushion.25 The integrity of all RNA samples was
confirmed by analysis on a denaturing agarose gel and
quantified by absorbance measurements at 260 nm.
Ribonuclease Protection Assay
A probe hybridizing to the I-II linker of the cardiac isoform of the
Na+ channel (hH1) was used as a myocyte-specific
control. The hH1 probe was normalized to the level of probe protecting
a segment of the 28S ribosomal RNA to correct for differences in RNA
loading. In general, the specific activity of the control probes was
Steady-state mRNA levels were quantified through exposure of the gels
on a storage phosphor screen and then scanning with a PhosphorImager
(Molecular Dynamics); quantification of the transcript levels was
performed with ImageQuant software (Molecular Dynamics). In cases in
which a doublet was observed, both bands were used in the
quantification. The level of target gene expression is given as either
the absolute density in arbitrary units or the relative density of the
protected fragment normalized to the density of the control protected
fragment to normalize for both RNA loading and the fraction of the
ventricular sample consisting of cardiac myocytes (hH1
probe). The averaged relative density of the channel transcripts from
failing samples was compared with nonfailing controls with the use of
ANOVA. The relative density of the Kv4.3 protected fragment is compared
to the maximal Ito density measured in
Isolation of Ventricular Myocytes and
Electrophysiology
The whole-cell configuration of the patch-clamp
technique27 was used to measure
Ito at room temperature (22°C to 23°C)
in cells isolated from 4 control and 6 failing ventricles as previously
described.28 Cell isolation was either not
performed or was inadequate for
electrophysiological recording in
the other 3 control and 11 failing hearts.
Ito was elicited by 500-ms voltage steps at
a frequency of 0.1 Hz as previously
described.7 8 28 The external solution for
recording Ito was (in mmol/L)
NaCl 138, KCl 4, CaCl2 2.0,
MgCl2 1, glucose 10,
NaH2PO4 0.33, HEPES 10,
CdCl2 0.3, and tetrodotoxin 0.003, pH 7.3. The
patch-clamp electrodes contained (in mmol/L) K-glutamate 120, KCl
10, MgCl2 2, HEPES 10, EGTA 5, and Mg-ATP 2, pH
7.2. Cell capacitance was calculated by integrating the area under an
uncompensated capacity transient elicited by a 20-mV depolarizing test
pulse from a holding potential of –80 mV. Series resistance was then
compensated as much as possible without ringing, typically 60% to
70%. Given the average series resistance of our electrodes, the
maximal uncompensated voltage error was <5 mV for the largest currents
studied. Currents were low-pass filtered at 2 kHz and digitized at 5 to
10 kHz with a Digidata 1200 A/D (Axon Instruments) interface for
off-line analysis.
The data were analyzed with the use of custom-written software.
Pooled data are presented as mean±SD. Statistical comparisons
were made with ANOVA, with P<0.05 considered significant.
Correlation of the action potential characteristics and current
densities to the normalized transcript levels was performed by linear
regression.
The demographics and drug regimens of the patients in the failing
cohort are summarized in Table 2
Expression of Kv4.3 mRNA in Heart Failure
Correlation Between mRNA Levels and Current Density
Expression of the hH1 Na Channel mRNA
Housekeeping genes such as glyceraldehyde-3-phosphate
dehydrogenase (GAPDH) or cyclophilin are typically used as
probes to normalize for the total amount of RNA. However, we observed
changes in the mRNA encoding both of these proteins in heart failure.
For example, there is a 17% decrease in the level of cyclophilin mRNA,
which does not reach statistical significance (Figure 1A
HERG mRNA in Heart Failure
There was greater intersample variability of HERG mRNA
expression in control samples compared with the variability of Kv4.3
and hH1 mRNA levels. The SD for the level of expression of Kv4.3 and
hH1 mRNAs in control hearts was 16% and 15% of the mean,
respectively. In contrast, the value for HERG mRNA was 28%.
We could not find any link between age or tissue handling that could
account for the variability in HERG expression in the
control samples. The control samples were all taken from the same
region of the left ventricle, excluding regional differences in
HERG expression as a cause of the variability.
Expression of Inward Rectifier Channel Genes in Heart
Failure
Expression of Other Ion Channel Genes in Heart Failure
The ß-subunits are known to modulate the function of
K+ channel
The L-type Ca2+ channel is active during the
plateau of the action potential. Alterations in the magnitude of this
current can change the plateau voltage and duration. The mRNA encoding
the
The correlation of the mRNA encoding Kv4.3 and the current density
suggests that changes in the message levels parallel reduction in the
functionally important, sarcolemmal component of the
Ito protein. In both
human7 13 and animal8
models of heart failure, there are no significant changes in
Ito kinetics or single-channel current
amplitude in cells isolated from the same region of normal and failing
ventricles; indeed, nonstationary fluctuation analysis was
consistent with a reduction in the number of functioning
channels in the cell membrane.8 The number of
functioning ion channels serves as a surrogate for the level of channel
protein. Therefore, it is reasonable to suggest that the reduction in
mRNA transcript is associated with a reduction in the level of
immunoreactive protein. However, it is possible that the level of
protein does not change despite a reduction in the steady-state level
of mRNA encoding Kv4.3 and reduction in Ito
density, and post-translational modifications occur that reduce channel
function without altering its biophysical properties. Alternatively,
changes in membrane trafficking may result in a reduction in the amount
of channel protein in the sarcolemma without a change in the total
cellular protein. It will be necessary to generate specific antibodies
to Kv4.3 and measure the protein levels directly to distinguish among
these possibilities. If Kv4.3 is a component of a
heteromultimeric channel that encodes
Ito, based on these data it appears to be
rate limiting in terms of production of functional
Ito channel.
The inward rectifier, IK1, is functionally
downregulated in human7 and
canine8 heart failure. In contrast to Kv4.3 and
similar to other K+ channel subunits measured by
RPA in this study, there is no change in the level of Kir2.1 mRNA in
failing ventricles compared with controls. It is possible that the
reduction in IK1 is mediated by a
post-translational change that eliminates channel function without
changing the steady-state level of RNA. It is also possible that the
current reduction is mediated post-transcriptionally and that protein
levels are indeed reduced in heart failure. Other Kir2 family members
have been cloned from human brain37 38 and murine
heart and brain,39 40 41 suggesting the changes in
the expression of these other genes could underlie the change in
IK1.
The delayed rectifier K channel is important in repolarization in heart
in a number of species. In humans, the HERG gene encodes the
rapid component of the inward rectifier
(IKr) and is mutated in the chromosome
7–linked form of the long QT syndrome, a congenital cardiac
arrhythmia characterized by disordered ventricular
repolarization.42 When the HERG gene
is expressed in oocytes or mammalian cells, its electrophysiology and
pharmacology resemble those of
IKr.42 43 44
IKr blockers prolong the human
ventricular action potential,45 46
and recently both the rapid component, IKr,
and the slow component, IKs, of the delayed
rectifier current were measured in human cardiac
myocytes.46 The mRNA level of HERG is
quite variable in human ventricle, but in transmural sections from
control and failing left ventricles, it does not change with heart
failure. The large variability might be the result of differences in
expression of the mRNA47 and current regionally,
so we cannot exclude the possibility that regional changes in
HERG mRNA expression occur with heart failure. Alternatively
spliced mRNA species of HERG and other mammalian homologs
have been recognized recently; the probe used in this study is common
to all of the HERG splice
variants.48 49
Further study of the apparent transcriptional regulation of Kv4.3 in
human heart failure will require study of the regulatory region of this
gene. Is the change in the transcript level an epiphenomenon due to the
altered neurohumoral environment in heart failure, or do the changes in
Kv4.3 transcript levels contribute to the development of heart failure?
Can the change in expression of this or other K channel genes be
altered pharmacologically: If so, will the change in expression alter
the incidence of sudden death due to ventricular
arrhythmias in heart failure?
Study Limitations
We did not evaluate changes in the level of the mRNA across the
ventricular wall and in different areas of the ventricles.
Regional changes in the K+ currents expressed are
likely to be important in the local control of ventricular
repolarization. However, a study of this sort is better suited to an
animal model in which the degree of heart failure, treatment regimen,
tissue harvesting, and cell isolation can be controlled.
Received February 10, 1998;
revision received May 22, 1998;
accepted June 3, 1998.
© 1998 American Heart Association, Inc.
Clinical Investigation and Reports
Molecular Basis of Transient Outward Potassium Current Downregulation in Human Heart Failure
A Decrease in Kv4.3 mRNA Correlates With a Reduction in Current Density
Abstract
Top
Abstract
Introduction
Methods
Results
Discussion
References
Background—Despite advances in
medical therapy, congestive heart failure remains a major cause of
death in the developed world. A disproportionate number of the deaths
of patients with heart failure are sudden and presumed to be
arrhythmic. Heart failure in humans and in animal models is associated
with prolongation of the action potential duration (APD), the result of
downregulation of K+ currents—prominently, the
Ca2+-independent transient outward current
(Ito). The mechanism for the reduction of
Ito in heart failure is unknown. The
K+ channel 
-subunit Kv4.3, a homolog of the
Drosophila Shal family, is most likely to encode all or
part of the native cardiac Ito in
humans. 1C subunit of the Ca2+ channel). mRNAs encoding
Kv1.2, Kv1.5, and Kv2.1 were found in low abundance in human
ventricle.
Key Words: currents • sodium • potassium • heart failure • repolarization
Introduction
Top
Abstract
Introduction
Methods
Results
Discussion
References
Congestive heart failure is a major cause of death
worldwide, with as many as 50% of affected patients dying
suddenly.1 Ventricular
tachyarrhythmias are a common cause of sudden death in
patients with heart failure2 ; however, the
underlying mechanism of these tachyarrhythmias is
poorly understood.3 Action potential prolongation
in the absence of other significant
electrophysiological changes is a hallmark
of failing ventricular
myocardium.4 5 6 7 8 Prolongation of the
action potential, particularly if it is heterogeneous, can
predispose to exaggerated dispersion of repolarization and nonexcitable
gap reentry.9 However, action potential
prolongation itself is arrhythmogenic; longer action potentials can be
associated with repolarization abnormalities such as
afterdepolarizations, which can predispose to triggered
arrhythmias.
Methods
Top
Abstract
Introduction
Methods
Results
Discussion
References
Preparation of cRNA Probes
The templates for preparing the human cRNA probes were generated
by subcloning small fragments of the cDNAs encoding ion channel and
control genes into pBluescript-SK, pBluescript II-SK (Stratagene),
pGEM7, or pSP73 (Promega). The cDNA fragments were isolated by
amplification of regions of the full-length cDNA clones using the
polymerase chain reaction (PCR), creating appropriate restriction sites
for subsequent subcloning. All constructs were confirmed by DNA
sequencing. Probe sequences were selected so that no probe had long
uninterrupted regions of identity with any mRNA other than the
transcript to be tested. There was no evidence for unwanted
cross-reaction between any probe and a nonspecific transcript.
Protected fragments of the anticipated size confirm specific
interaction of the cRNA probe and its target transcript. The hH1
template was designed to protect a fragment in the I-II linker of the
Na+ channel, which exists only in the cardiac
isoform; thus, this is a myocyte-specific cRNA probe. 
.
This table gives the probe, the reference, the numbers of the
nucleotide sequence that are protected, and the size of the
protected fragment. Each PCR primer pair includes a KpnI
site (forward) and a HindIII site (reverse) for cloning. The
riboprobe for Kv4.3 is as previously described.15
A commercially available cDNA probe with a protected size of 103 base
pairs (bp) was used for human cyclophilin (Ambion).
View this table:
[in a new window]
Table 1. RPA Probes
Human ventricular myocardium was
obtained from explanted failing hearts (failing; n=17) and donor hearts
unsuitable for transplantation (normal; n=7). Details of the
characteristics of the patients in the heart failure and control groups
are given in Table 2. Segments of
ventricular myocardium adjacent to the region
from which myocytes were isolated for
electrophysiological recording were
used for RNA isolation. The tissue was excised from the left
ventricular free wall between the left anterior descending
coronary artery (LAD) and the left circumflex artery. The
isolation procedure selects for cells from the midportion of the
ventricular wall. The tissue samples used for RNA isolation
were transventricular, including both the epicardium and
endocardium, and shared a border with the section of the heart used for
cell isolation. In the case of an anterior transmural myocardial
infarction, the left circumflex artery was perfused for cell isolation
and a segment of the lateral wall was used for RNA isolation.
View this table:
[in a new window]
Table 2. Patient Demographics and Medications at
Transplantation of Failing
Hearts
RPAs were performed as described
previously.14 15 26 All probes contained regions
of plasmid sequence at 1 or both ends of the transcript, permitting
easy distinction between any remaining undigested probe and the
shorter, specifically protected region of the probe. Ten micrograms of
yeast tRNA was used as a negative control to test for the presence of
probe self-protection. At least duplicate determinations were performed
on each ventricular sample, and intrasample variability was
<15% (in most cases, <10%). For each sample point, 10 µg of total
RNA was used in the assay. 
5-fold lower than that for the target channel probes. 
2
cells isolated from adjacent regions of ventricular
myocardium. Ito was
recorded in 2 or 3 cells in the normal hearts; in the failing
hearts, Ito was recorded in a median of
3 cells (range, 2 to 6).
Ventricular myocytes were isolated as previously
described with minor modifications.7 8 The
territory perfused by the LAD from the aortic root to the apex was
excised and perfused via the coronary artery with a
collagenase- and protease-containing Tyrode buffer. The
currents described were measured in cells isolated from the central
third of the myocardial wall, excluding endocardial and epicardial
layers, from the area of the left ventricular anterior free
wall between the LAD and the first diagonal branch. Only
Ca2+-tolerant cells with clear cross striations
and without spontaneous contraction or significant granulation (20%
to 40% of those isolated in both control and failing hearts) were
selected for the experiments.
Results
Top
Abstract
Introduction
Methods
Results
Discussion
References
RPAs were used to determine the levels of ion channel mRNAs in
normal and failing human myocardium. The levels of
K+ channel mRNA were correlated with currents
measured in myocytes isolated from adjacent segments of tissue. Samples
from 7 normal hearts and 17 failing hearts were examined. The average
age of the patients from whom the control hearts were obtained was
42±13 years, and that of the transplant recipients from whom the
failing hearts were harvested was 52±10 years. Two of the normal and 3
of the failing hearts were from female patients. The cause of heart
failure in the transplant recipients was dilated
cardiomyopathy in 10 of the patients (59%),
ischemic/postinfarction cardiomyopathy in 6
(35%), and postpartum cardiomyopathy in 1
(6%). 
. All of the patients were on standard
heart failure medical regimens at the time of transplantation,
including diuretics, ACE inhibitors, and digitalis;
4 of the patients were taking amiodarone, and no other
antiarrhythmic drugs were used in this cohort. The age, sex, and
reasons for not transplanting the normal hearts are given in Table 2.
We previously reported that the Kv4.3 channel is likely to
underlie a significant fraction, if not all, of the
Ca2+-independent
Ito.15 The level of
Kv4.3 mRNA in the ventricular myocardium of
normal and failing hearts was compared to further test the hypothesis
that Kv4.3 is a component of the native Ito
and to determine whether Ito downregulation
in heart failure is correlated with downregulation of Kv4.3 mRNA
levels. Figure 1A shows a
representative RPA for Kv4.3 mRNA in which samples from
normal and failing ventricular myocardium are
compared; lane P shows the size of the unprotected probe, and lane t
contains tRNA and no protection is observed (5 normal and 7 failing
samples are shown). There is sample-to-sample variability in both
normal and failing hearts: the range of transcript densities in
arbitrary units is 0.79 to 1.34 and 0.36 to 0.81 for normal and failing
samples, respectively. Repeated assays on the same samples reveal
little intrasample variability; the maximal difference on repeated
determinations was 17%, and in general, the densities of the protected
fragment for Kv4.3 varied by <8%. Overall, in these 12 samples, there
is a 34% reduction in the unnormalized Kv4.3 mRNA in failing
myocardium compared with that in the control hearts (Figure 1B, P<0.01).
View larger version (28K):
[in a new window]
Figure 1. Changes in Kv4.3 mRNA in heart failure. A,
Representative RPA with probes designed to protect a
337-bp fragment of human Kv4.3 and a 103-bp fragment of cyclophilin.
The first lane (P) contains the probes alone. The lane marked t
contains yeast tRNA. Shown is 10 µg of total RNA from 5 normal and 7
failing hearts. B, Bar plot of the unnormalized density of Kv4.3 mRNA
in 5 normal and 7 failing left ventricular samples; the
values are the average of duplicate determinations. The steady-state
level of mRNA encoding Kv4.3 is reduced by 34% in failing hearts
compared with controls (P=0.009) C,
Representative currents elicited by a series of
depolarizing voltage steps from –30 to +80 mV in 10-mV increments from
a holding potential of –60 mV. The normal traces are from the heart in
lane 4, and the failing traces are from the heart in lane 6 (A). D, Bar
plot of the current density measured at a test potential of +40 mV in
16 cells from the 5 normal ventricles and 25 cells from the 7 failing
ventricles in A. E, Correlation between Kv4.3 mRNA and
Ito density in ventricular
myocytes. In 10 samples, 6 from failing and 4 from normal ventricles,
RPAs were performed and currents were measured in at least 2 cells
isolated from an adjacent region of the myocardium. The
maximal Ito density is plotted against the
unnormalized level of Kv4.3 mRNA in arbitrary units determined on the
same RPA. F, A similar plot of the Kv1.4 mRNA density versus
Ito density reveals no correlation.
Myocytes were isolated from an adjacent region of the ventricle
for the 12 samples shown in Figure 1A; Ito
current density was recorded in at least 2 cells from 10 of these
hearts (normal sample No. 2 and heart failure No. 4 current records
were obtained from only 1 cell). Representative current
records elicited by a family of depolarizing voltage steps from
–30 to +80 mV in increments of 10 mV from a holding potential of –60
mV are shown in Figure 1C. The mean and SD values of the current
densities at +40 mV in 25 cells from 7 failing hearts and 16 cells from
5 control hearts are shown in Figure 1D (4.1±2.8 pA/pF for failing and
7.9±3 pA/pF for control, P=0.0035). The change in
Kv4.3 mRNA in failing myocardium is consistent in
both direction and magnitude with the change in native
Ito measured in ventricular
myocytes isolated from the same hearts. We previously demonstrated that
the reduction in Ito in heart failure is
due to a reduced number of functioning channels and therefore is a
reasonable index of the level of channel
protein.8 If Kv4.3 encodes the native channel, we
anticipate a correlation between the level of mRNA and the density of
Ito. Figure 1E illustrates that such a
correlation exists; this is a plot of the unnormalized mRNA density and
maximal Ito density at +40 mV (the circles
represent normal hearts, and the squares represent
failing ventricles). The plot reveals a significant correlation between
Kv4.3 mRNA and the current density (r=0.89,
P=0.0005). In contrast, the Ito
density did not correlate with the abundance of other mRNAs. Another
gene encoding an inactivating K channel, Kv1.4, showed no correlation
with the density of Ito (Figure 1E;
r=0.38, P=0.28).
It is possible that the change in Kv4.3 mRNA is due to a
generalized depression in mRNA levels in failing hearts or to a more
specific decline in the expression of genes encoding ion channels. To
test this possibility, we examined the level of mRNA encoding the hH1
Na+ channel in normal and failing hearts. We
designed a probe to the hH1 Na+ channel I-II
linker, which is unique to the cardiac isoform of this channel and
therefore myocyte specific. The level of hH1 mRNA does not change in
failing hearts (Figure 2). The level of
hH1 mRNA in failing samples is 96% of that of control hearts (Figure 2B, P>0.7). Similarly, when the steady-state level of hH1
mRNA is normalized to the density of 28S ribosomal RNA to control for
sample loading, there is no difference between control and failing
ventricles (data not shown). Thus, the reduction in Kv4.3 mRNA is not
the result of a generalized decrease in mRNA.
View larger version (28K):
[in a new window]
Figure 2. hH1 mRNA does not change in heart failure. A, A
myocyte-specific probe designed to the I-II linker of the cardiac
isoform of the Na+ channel is hybridized to 10 µg of
total RNA in 5 samples from normal and 4 samples from failing
ventricles. In the lane marked tRNA, the probe is hybridized to yeast
tRNA rather than myocardial RNA. B, Bar plot of the averaged density of
mRNA encoding hH1 reveals no significant difference in failing and
control hearts. C, Representative RPA exhibiting
reduction of the Kv4.3 protected fragment in failing samples compared
with normal controls. No significant change in hH1 density is observed.
D, Scatterplot of the density of Kv4.3 mRNA normalized to hH1 from 7
normal and 17 failing hearts. Each data point represents the
average of at least 2 determinations ( 
and 
, mean values for
normal and failing hearts, respectively; 1.2±0.16 control versus
0.84±0.16 failing, P=0.00005)., P>0.15). This trend in the cyclophilin mRNA level does
produce problems with respect to normalization of the channel
transcript levels to cyclophilin. The hH1 mRNA does not change with
heart failure (Figure 2A and 2B), which is consistent with the
lack of changes in Ina in human atrial
cells29 and canine ventricular cells
from failing hearts8 ; thus, hH1 may be a more
relevant control probe for K+ channel target
genes. We repeated the Kv4.3 RPA with hH1 as the normalizing control
probe in 7 control and 17 failing samples. The level of Kv4.3 mRNA
normalized to hH1 is reduced by 30% in failing cells compared with
nonfailing controls (Figure 2C and 2D, P=0.0005). Because
hH1 mRNA reflects the amount of RNA from cardiac myocytes, the
reduction of Kv4.3 mRNA is not simply the result of myocyte dropout;
instead, the level of Kv4.3 mRNA in ventricular myocytes
decreases. The intersample variability of Kv4.3 mRNA levels in the
heart failure cohort could not be ascribed to patient age, cause of
heart failure, or treatment with amiodarone. Three patients
with heart failure had Kv4.3 mRNA levels that overlapped with the range
of values in normal hearts: 2 of these patients had dilated
cardiomyopathy, and 1 had ischemic
cardiomyopathy and none were being treated with
amiodarone.
The ventricular action potential is prolonged in heart
failure; similarly, in the genetically determined cardiac
arrhythmia, the long QT syndrome (LQTS), repolarization is
prolonged. The chromosome 7–associated LQTS is caused by mutations in
the human ether a go-go–related gene
(HERG).30 To test the possible
involvement of the HERG gene product in the abnormal
repolarization associated with heart failure, we measured mRNA levels
of this gene in control and failing hearts (Figure 3). HERG RNA is abundantly
expressed in the human ventricle.30 31 There was
no significant change in HERG mRNA; the average value for
HERG mRNA expression in failing hearts was 85±18% of the
control value (Figure 3B, P=NS).
View larger version (21K):
[in a new window]
Figure 3. HERG mRNA exhibits significant
variability but no change in failing ventricular
myocardium. A, Representative RPA with 10
µg of control or failing ventricular
myocardium per lane. The 7 normal and 17 failing samples
exhibit large variability in the density of HERG mRNA.
B, The scatterplot of HERG mRNA density illustrates the
large range of values in both control and failing ventricles. The mRNA
density in normalized to the hH1 signal in control and failing hearts
is 1.05±0.43 and 0.9±0.2, respectively (P=0.28).
The inward rectifier K+ current
(IK1) is active in the terminal phases of
repolarization and is functionally downregulated in
human7 and canine8 heart
failure. The mRNA encoding IK1 (Kir2.1) is
abundant in human ventricle but does not change in failing hearts
compared with controls (Figure 4A and 4B). There is significant
sample-to-sample variability in the steady-state mRNA levels encoding
IK1; the SD for the level of expression
over all samples was 28%. The level of Kir2.1 mRNA in failing hearts
is not different from that in control ventricles (Figure 4B, P=0.56).
View larger version (38K):
[in a new window]
Figure 4. Kir2.1 mRNA does not change with heart failure. A,
Representative RPA of Kir2.1 (probe, 235 bp; protected
fragment, 194 bp) and hH1 in 5 control and 5 failing
ventricular samples. B, Scatterplot of the density of
Kir2.1 mRNA normalized to hH1 reveals no difference in control and
failing hearts. The normalized mRNA densities in arbitrary units are
0.88±0.16 and 0.94±0.26 for control and failing hearts, respectively
(P=0.56).
The reduction in Kv4.3 mRNA in failing hearts is in contrast to
other mRNAs examined (Figure 5). We
evaluated the transcript level of several other channel genes known to
be present in ventricular myocardium. There
was no change in the level of the Kv1 family gene Kv1.4, and we
observed very little Kv1.2 and Kv1.5 in human ventricle. The only
member of the Kv2 family of genes expressed in rat heart is
Kv2.1,14 and we detected no Kv2.1 in human
ventricle. This result is consistent with the absence of any
appreciable ultrarapid delayed rectifier K+
current in human ventricular myocytes. In other mammals,
Kv3 gene family members are present in low abundance in the
heart14 32 and are not likely to be relevant to
native ventricular
Ito.15
View larger version (58K):
[in a new window]
Figure 5. The level of mRNA encoding other voltage-dependent
channels is not different in normal and failing human
myocardium. A, Bar plots are shown for the level of mRNA
normalized to hH1, encoding Kv1.4, Kvß1.3, and the 1 subunit of
the cardiac Ca2+ channel. There is no significant
difference in the level of any of these transcripts in normal compared
with failing ventricles. B, Representative RPAs for
each of the transcripts shown in A.-subunits, in some cases changing
the phenotype of a channel from
noninactivating to
inactivating.33 The Kvß1.3 subunit is known to
be present in human heart34 but does not
change in heart failure (Figure 5). 1C subunit of Ca2+ channel is abundant in
human ventricle but does not change with heart failure (Figure 5).
Discussion
Top
Abstract
Introduction
Methods
Results
Discussion
References
One of the hallmarks of ventricular myocardial failure
is prolongation of the cellular action potential duration. In humans
and animal models, the action potential profile is changed; often, a
reduction in the phase 1 repolarizing notch is
observed.7 8 The prolongation and change in phase
1 are the result of decrease in the
Ca2+-independent Ito
density.7 8 Experiments in small mammals suggest
that the Shal subfamily K channel genes Kv4.2 and Kv4.3
encode the native cardiac
Ito.14 15 35 The K
channel, Kv4.3, is the leading candidate gene for
Ito in dogs and
humans.15 In further support of this hypothesis,
we observed a reduction in Kv4.3 mRNA in heart failure. Remarkably, the
reduction in Kv4.3 mRNA is similar to the observed reduction in
Ito density in cells isolated from failing
human ventricles.7 The level of Kv4.3 mRNA was
correlated with the Ito density of
ventricular myocytes isolated from segments of the heart
adjacent to the samples used for RPA. These data suggest that in the
human heart, Kv4.3 encodes all or part of the
Ca2+-independent Ito
and that downregulation of this current in heart failure is
consistent with altered transcription. However, changes in the
steady-state level of mRNA can occur by mechanisms other than altered
rates of transcription, and mRNA levels cannot be equated with the
level of protein.36
A significant limitation of any study in humans is the lack of
control of the conditions of the normal and especially the failing
ventricles. The anesthetic agents used at the time of cardiac
explantation were similar in patients with control and failing hearts,
but many other parameters were significantly different. The
cause of heart failure was variable, and the duration and severity
of myocardial dysfunction and the medical therapy at the time of
explantation of the failing hearts obviously were not controlled. Our
experiments were designed to correlate the level of functional protein
(ie, Ito density) in cells isolated from
these hearts with mRNA levels in immediately adjacent tissue from the
same hearts. A limitation of this design is the necessarily small
amount of tissue from which all of the measurements can be made; thus,
in this study the amount of immunoreactive hKv4.3 protein in the region
adjacent to the section of the ventricle used for cell isolation was
not determined. Despite these limitations, it is notable that we were
able to measure a statistically significant reduction in the mRNA level
encoding a K+ channel gene that is likely to play
a significant role in repolarization of the human heart. Furthermore,
the transcript level of Kv4.3 correlates well with the current density
of Ito in heart cells, consistent
with the hypothesis that this gene encodes a component of the native
channel in humans.
View this table:
[in a new window]
Table 3. Patient Demographics and Medications at
Transplantation of Control
Hearts
Acknowledgments
This work was supported by NIH grant HL-P50–52307 (SCOR in
Sudden Cardiac Death) and by Deutsche Forschungsgemeinschaft (S.K.). We
thank the following for generous gifts of the cDNA clones: Drs Mike
Tamkun (hKv1.4 and 1.5), Steve Ikeda and Rolf Joho (hKv2.1), Barbara
Wible (hKvß1.3), Barry Ganetsky and Jeff Warmke (HERG),
and Roland Kallen (hH1). We thank Colleen McCormack and Qing Xheng for
technical support and Dr Eduardo Marban for critical review of the
manuscript. We gratefully acknowledge the support of our colleagues,
the members of the heart failure and cardiac transplantation services
at the University of Munich and Johns Hopkins University.
References
Top
Abstract
Introduction
Methods
Results
Discussion
References
-subunits and voltage-gated K+ channel
currents in rat ventricular myocytes. J Gen
Physiol. 1996;108:405–419.
This article has been cited by other articles:
 |
|
 |
|
  C. Marionneau, S. Brunet, T. P. Flagg, T. K. Pilgram, S. Demolombe, and J. M. Nerbonne Distinct Cellular and Molecular Mechanisms Underlie Functional Remodeling of Repolarizing K+ Currents With Left Ventricular Hypertrophy Circ. Res., June 6, 2008; 102(11): 1406 - 1415. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Soh and S. A. N. Goldstein ISA Channel Complexes Include Four Subunits Each of DPP6 and Kv4.2 J. Biol. Chem., May 30, 2008; 283(22): 15072 - 15077. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Fischer, R. Dechend, A. Gapelyuk, E. Shagdarsuren, K. Gruner, A. Gruner, P. Gratze, F. Qadri, M. Wellner, A. Fiebeler, et al. Angiotensin II-induced sudden arrhythmic death and electrical remodeling Am J Physiol Heart Circ Physiol, August 1, 2007; 293(2): H1242 - H1253. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Fernandez-Velasco, G. Ruiz-Hurtado, O. Hurtado, M. A. Moro, and C. Delgado TNF-{alpha} downregulates transient outward potassium current in rat ventricular myocytes through iNOS overexpression and oxidant species generation Am J Physiol Heart Circ Physiol, July 1, 2007; 293(1): H238 - H245. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Nattel, A. Maguy, S. Le Bouter, and Y.-H. Yeh Arrhythmogenic Ion-Channel Remodeling in the Heart: Heart Failure, Myocardial Infarction, and Atrial Fibrillation Physiol Rev, April 1, 2007; 87(2): 425 - 456. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Teutsch, R. P. Kondo, D. A. Dederko, J. Chrast, K. R. Chien, and W. R. Giles Spatial distributions of Kv4 channels and KChip2 isoforms in the murine heart based on laser capture microdissection Cardiovasc Res, March 1, 2007; 73(4): 739 - 749. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Radicke, D. Cotella, E. M. Graf, U. Banse, N. Jost, A. Varro, G.-N. Tseng, U. Ravens, and E. Wettwer Functional modulation of the transient outward current Ito by KCNE {beta}-subunits and regional distribution in human non-failing and failing hearts Cardiovasc Res, September 1, 2006; 71(4): 695 - 703. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Hinterseer, M. Irlbeck, L. Ney, B.-M. Beckmann, A. Pfeufer, G. Steinbeck, and S. Kaab Acute respiratory distress syndrome with transiently impaired left ventricular function and Torsades de Pointes arrhythmia unmasking congenital long QT syndrome in a 25-yr-old woman Br. J. Anaesth., August 1, 2006; 97(2): 150 - 153. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Zhou, C. Ziegler, L. A. Birder, A. F.R. Stewart, and E. S. Levitan Angiotensin II and Stretch Activate NADPH Oxidase to Destabilize Cardiac Kv4.3 Channel mRNA Circ. Res., April 28, 2006; 98(8): 1040 - 1047. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. S. Barth and S. Kaab MAPK = Mitogen-Activated Protein KChIP2?: Unraveling Signaling Pathways Controlling Cardiac Ito Expression Circ. Res., February 17, 2006; 98(3): 301 - 302. [Full Text] [PDF]
|
|