Download Changes in Ion Channel Gene Expression Underlying Heart Failure

Changes in Ion Channel Gene Expression Underlying Heart
Failure-Induced Sinoatrial Node Dysfunction
Joseph Yanni, MD, PhD*; James O. Tellez, PhD*; Michal Ma˛czewski, MD, PhD*;
Urszula Mackiewicz, PhD; Andrzej Beresewicz, MD, PhD; Rudi Billeter, PhD;
Halina Dobrzynski, PhD†; M.R. Boyett, PhD†
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 12, 2017
Background—Heart failure (HF) causes a decline in the function of the pacemaker of the heart—the sinoatrial node (SAN).
The aim of the study was to investigate HF-induced changes in the expression of the ion channels and related proteins
underlying the pacemaker activity of the SAN.
Methods and Results—HF was induced in rats by the ligation of the proximal left coronary artery. HF animals showed an
increase in the left ventricular (LV) diastolic pressure (317%) and a decrease in the LV systolic pressure (19%)
compared with sham-operated animals. They also showed SAN dysfunction wherein the intrinsic heart rate was reduced
(16%) and the corrected SAN recovery time was increased (56%). Quantitative polymerase chain reaction was used to
measure gene expression. Of the 91 genes studied during HF, 58% changed in the SAN, although only 1% changed in
the atrial muscle. For example, there was an increase in the expression of ERG, KvLQT1, Kir2.4, TASK1, TWIK1,
TWIK2, calsequestrin 2, and the A1 adenosine receptor in the SAN that could explain the slowing of the intrinsic heart rate.
In addition, there was an increase in Na⫹-H⫹ exchanger, and this could be the stimulus for the remodeling of the SAN.
Conclusions—SAN dysfunction is associated with HF and is the result of an extensive remodeling of ion channels; gap junction
channels; Ca2⫹-, Na⫹-, and H⫹-handling proteins; and receptors in the SAN. (Circ Heart Fail. 2011;4:496-508.)
Key Words: sinoatrial node 䡲 heart failure 䡲 ion channels
T
activity in HF,10 which also could be true in the case of
HF-induced sick sinus syndrome.
he prevalence of heart failure (HF) is increasing in
modern industrial nations, and it is estimated that ⬎20
million people have this condition worldwide.1 Sudden cardiac death accounts for ⬇50% of deaths in patients with HF,2
and bradyarrhythmias account for ⬇42% of the sudden
deaths in the hospital.3 It is now known that HF causes
dysfunction of the pacemaker of the heart—the sinoatrial
node (SAN) (ie, sick sinus syndrome)4—as well as dysfunction of the atrioventricular node.5 In human, dog, and rabbit,
there is a decrease in the intrinsic heart rate (heart rate in
absence of autonomic influence) during HF.6 – 8 Sanders et al6
demonstrated that in patients with congestive HF as well as a
decrease in the intrinsic heart rate, there is an increase in the
corrected SAN recovery time (SNRTc), a caudal shift of
the leading pacemaker site, and abnormal propagation of the
action potential from the SAN (ie, an increase of the SAN
conduction time). Changes in the sensitivity of the SAN to
acetylcholine and vagal nerve stimulation have been observed
in HF in rabbit and dog.8,9 In other cardiac tissue types, it has
been shown that a remodeling of ionic currents (and underlying ion channels) is responsible for the changes in electric
Clinical Perspective on p 508
In the SAN, various inward ionic currents are responsible
for the pacemaker potential and, therefore, pacemaker activity: the funny current (If), the Na⫹ current (INa), L- and T-type
Ca2⫹ currents (ICa,L and ICa,T), and inward Na⫹-Ca2⫹ exchange current (INaCa) activated by diastolic release of Ca2⫹
from sarcoplasmic reticulum (SR).11 The delayed rectifier K⫹
currents, of which there are at least 3 types (ultrarapid [IK,ur],
rapid [IK,r], and slow [IK,s]), are activated during the action
potential, and after the action potential, they deactivate during
the pacemaker potential; the deactivation of the outward
currents allows the pacemaker potential to develop (K⫹ decay
hypothesis).12 Another important factor responsible for pacemaking is the lack of inward rectifier K⫹ current (IK,1) in the
SAN; in the working myocardium, IK,1 is responsible for the
stable resting potential, and its absence in the SAN facilitates
pacemaking.13 There is some evidence of a remodeling of
these ionic currents (and underlying ion channels) in HF-
Received May 23, 2010; accepted March 25, 2011.
From the University of Manchester, Manchester, UK (J.Y., J.O.T., H.D., M.R.B.); Medical Center of Postgraduate Education, Warsaw, Poland (U.M.,
A.B.); and University of Nottingham, Nottingham, UK (R.B.).
*Drs Yanni, Tellez, and Ma˛czewski are joint first authors.
†Drs Dobrzynski and Boyett are joint senior authors.
The online-only Data Supplement is available at http://circheartfailure.ahajournals.org/cgi/content/full/CIRCHEARTFAILURE.110.957647/DC1.
Correspondence to M.R. Boyett, PhD, Cardiovascular Medicine, University of Manchester, Core Technology Facility, 46 Grafton St, Manchester M13
9NT, UK. E-mail [email protected]
© 2011 American Heart Association, Inc.
Circ Heart Fail is available at http://circheartfailure.ahajournals.org
496
DOI: 10.1161/CIRCHEARTFAILURE.110.957647
Yanni et al
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 12, 2017
induced sick sinus syndrome: Block of If by zatebradine
causes a smaller decrease in heart rate in rabbit in HF,14 and
the decrease in SAN pacemaking in this model has been
attributed to a decrease in If.15 In the dog, a decrease in
hyperpolarization-activated cyclic nucleotide-gated (HCN) 2
and HCN4 mRNA and protein (ion channels responsible for
If) by up to ⬇80% is observed in the SAN in HF.7 It is not
known whether HF affects other ion channels in the SAN. In
various models of cardiac hypertrophy and HF, the Na⫹-H⫹
exchanger (NHE1) has been shown to play a surprising
central role: NHE1 is upregulated, perhaps as a response to
overload and stretch of the myocardium, and this may be
responsible for the remodeling of the myocardium because
inhibition of NHE1 (eg, by cariporide) results in prevention
or regression of hypertrophy.16
The aim of the present study was to investigate changes in
the expression of genes that potentially could be responsible
for HF-induced sick sinus syndrome. Genes for ion channels;
connexins (responsible for electric coupling between myocytes); Ca2⫹-, Na⫹- and H⫹-handling proteins (including
NHE1); and receptors were investigated. Widespread changes
were observed.
Methods
HF was induced in 10 male Sprague-Dawley rats (aged 12 weeks) by
ligation of the proximal left coronary artery as described previously.17 The ligation resulted in a large infarct of the left ventricle
(LV). Ten sham-operated rats were used as controls. Twelve weeks
postoperatively, cardiac function was assessed in the anesthetized
animal using echocardiography, ECG recordings, and catheterization. The animals were then euthanized, and SAN function was
characterized in the isolated Langendorff-perfused heart using ECGlike recordings. Gene expression was characterized using quantitative polymerase chain reaction (qPCR).
Functional Measurements
Rats were anesthetized with ketamine HCl 100 mg/kg body weight
IP and xylazine 5 mg/kg body weight IP. They underwent echocardiography (MyLab25; Esaote, Italy) with a 13-MHz linear array
transducer. Briefly, left ventricular (LV) end-diastolic and end-systolic areas were measured with a planimeter from the parasternal
long-axis view. LV ejection fraction was calculated as (LV diastolic
area⫺LV systolic area)/LV diastolic area, as reported previously.17
Then limb ECG electrodes were attached, and surface ECG recordings were obtained. In vivo normal heart rate was derived from at
least 60 consecutive heart beats, and atrioventricular conduction time
was measured as the PQ interval. A micromanometer-tipped catheter
(Millar Instruments; Houston, TX) was advanced through the right
carotid artery into the LV for recording of LV pressures and peak
rate of rise and decline of LV pressure (dP/dtmax and dP/dtmin,
respectively). After heparinization (1000 IU/kg body weight IP) and
pentobarbital anesthesia (50 mg/kg body weight IP), the heart was
excised. Lung tissue and excess connective tissue were removed and
the heart weighed. The heart was then Langendorff perfused with
Krebs-Henseleit buffer containing NaCl, 100 mmol/L; KCl,
4 mmol/L; MgSO4, 1.2 mmol/L; KH2PO4, 1.2 mmol/L; CaCl2,
1.8 mmol/L; NaHCO3, 25 mmol/L; and glucose, 10 mmol/L. The
solution was bubbled with 95% O2 and 5% CO2 to give a pH of 7.4.
Recording hook electrodes were inserted into the LV and right atrial
appendage. Bipolar pacing hook electrodes were inserted into the
right atrial appendage. The heart was allowed 10 minutes to stabilize.
In vitro intrinsic heart rate was determined from at least 60
consecutive heart beats. SNRT and SNRTc were determined during
atrial pacing at a frequency of 2 Hz higher than the intrinsic heart rate
for 10 s duration. The SNRT was measured as the time from the last
paced stimulus to the onset of the first spontaneous P wave. The
HF and Ion Channels in the Sinoatrial Node
497
Table 1. Functional Characterization of Sham-Operated and
HF Animals
Characteristic
Sham (n⫽10)
Body weight, g
528⫾10.5
Heart weight, g
1.8⫾0.1
2.7⫾0.2*
0.003⫾0.0001
0.005⫾0.0004*
Heart weight/body weight ratio
HF (n⫽10)
532⫾21
LV end diastolic area, mm2
78⫾2.3
LV end systolic area, mm2
36⫾1.5
95⫾8.2*
LV ejection fraction
0.5⫾0.007
0.2⫾0.02*
LV diastolic pressure, mm Hg
4.8⫾0.1
20⫾0.4*
LV systolic pressure, mm Hg
129⫾2.8
104⫾3.8*
LV developed pressure, mm Hg
124⫾2.8
84⫾5.5*
LV dP/dtmax, mm Hg/s
6026⫾150
3783⫾160*
LV dP/dtmin, mm Hg/s
4333⫾186
2560⫾207*
62⫾1.5
74⫾1.7*
In vivo PQ interval, ms
118⫾8.2*
See Methods section for details. HF indicates heart failure; LV, left ventricle.
*Significantly different (unpaired t test) at P⬍0.05 from sham-operated rats.
measurement was repeated 3 times, and the reported SNRT is a mean
from 3 measurements. To control for differences in intrinsic heart
rate, SNRT was normalized to the resting heart rate by subtracting
the SAN cycle length from the SNRT (SNRTc⫽SNRT⫺SAN cycle
length). A 60-s period was allowed to elapse between each successive pacing. Functional characteristics of all animals are shown in
Table 1. Results are presented as mean⫾SEM. Differences were
evaluated by Student t test and considered significant at P⬍0.05.
RNA Isolation
After functional measurements, intact SAN preparations were dissected from the hearts of sham-operated and myocardial infarction
rats as previously described.18 Tissue samples (⬇1⫻1 mm) were
taken from the center of the SAN as well as from the right atrial free
wall. The samples were taken approximately at the expected level of
the leading pacemaker site in the SAN (at the level of the main
branch from the crista terminalis). The samples were immersed in a
drop of freezing medium (OCT, BDH) and frozen in liquid N2.The
tissue subsequently was cut into 20-␮m sections on a cryostat for
RNA isolation. Total RNA isolation was performed on these samples
with Qiagen RNeasy Micro spin columns. Using random hexamer
priming to obtain cDNA, 100 ng of total RNA was reversed
transcribed with superscript III reverse transcriptase (Invitrogen).
Quantitative Polymerase Chain Reaction
The qPCR reaction was performed using an Applied Biosystems
7900HT instrument with low-density Taqman array cards (onlineonly Data Supplement Table 1). Additional transcripts were quantified using Power SYBR green master mix. Primer assays were
purchased from Qiagen, or individual primers were used (onlineonly Data Supplement Table 2); 28S was used as a housekeeping
transcript. A common calibrator sample comprising a combination of
different RNAs that represented all tissue types used was used with
each run. Expression levels were calculated using ⌬Ct and ⌬⌬Ct
methods (see online-only Data Supplement for details)18; the results
were qualitatively similar with the different methods. Data obtained
with the ⌬Ct method (implemented using StatMiner version 4.1
software [Integromics]) are shown in the Figures. In the case of the
⌬Ct method, the abundance of the transcript of interest is normalized
to the abundance of a housekeeping transcript to correct for variations in input RNA. The ⌬Ct method allows the abundance of
different transcripts to be roughly compared as well as allows the
abundance of transcript of interest in different tissues and different
treatment groups to be compared. However, the efficiency of the
reverse transcription (described previously) for different transcripts
can perhaps vary up to 10 times; therefore, only ⱖ10 times
498
Circ Heart Fail
July 2011
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 12, 2017
Figure 1. Characterization of the heart failure (HF) model. A, Echocardiography of the left side of heart in sham-operated and HF animals during diastole and systole. B, Right side of heart from sham-operated and HF animals; right atrium is outlined by dashed line. C,
Relative abundance of NHE1 and Tbx3 transcripts in AM and SAN in sham-operated and HF animals. Data are presented as
mean⫾SEM (n⫽7). AM indicates atrial muscle; LA, left atrium; LV, left ventricle; NHE1, Na⫹-H⫹ exchanger; SAN, sinoatrial node; Tbx3,
T-box transcription factor. *Significantly different (false discovery rate) from corresponding data from sham-operated animals.
differences in the abundance of different cDNAs (corresponding to
different transcripts) were interpreted as differences in the corresponding transcripts. Changes in mRNA with HF were evaluated
with the false discovery rate, which has become a standard for
multiple tests correction in microarray data analysis (see online-only
Data Supplement for details).
Hierarchical Clustering and
Multidimensional Scaling
Data obtained with the ⌬⌬Ct method (implemented using methods
described by Tellez et al18) were used. In the case of the ⌬⌬Ct
method, first the abundance of a transcript of interest is normalized
to the abundance of a housekeeping transcript (yielding a ratio of
abundance of transcript of interest to abundance of housekeeping
transcript) to correct for variations in input RNA; this ratio for the
sample then is normalized to the same ratio for the calibrator to
correct for variations between runs. The ⌬⌬Ct values for each
transcript and sample were transformed to log2 and then centered for
each individual sample and each transcript, resulting in data for each
transcript having a mean of 0 and an SD of 1 and thus allowing for
comparison of the data sets. For hierarchical clustering and multidimensional scaling, J-Express version 7 was used. Hierarchical
clustering was carried out using a selected group of 28 transcripts
with Euclidian distances and average linkage. Multidimensional
scaling was carried out with data from all transcripts with Euclidian
distances.
Results
systolic pressure (19%), LV developed pressure (32%), maximum rate of rise of LV pressure (37%), maximum rate of fall
of LV pressure (41%), and LV ejection fraction (60%) (Table
1). The hearts from the HF animals were hypertrophied in that
compared with the hearts from sham-operated animals, they
were heavier, and the heart-to-body weight ratio was greater
(Table 1). As expected, echocardiography showed an enlargement of the LV and left atrium in HF animals (Figure 1A);
this is confirmed by the measurements of LV end-diastolic
and end-systolic areas shown in Table 1. The right atrium also
was enlarged as shown in Figure 1B. We propose that this is
the cause of the remodeling of the SAN as explained in the
Discussion section.
The HF animals showed SAN dysfunction. Although in the
anesthetized animals there was no difference in the heart rate
between the HF and sham-operated animals (Figure 2A and
2B), in the isolated Langendorff-perfused hearts from the HF
animals the intrinsic heart rate was significantly slower
(Figure 2A and 2C), and the SNRTc was significantly greater
(Figure 2D and 2E) than in hearts from sham-operated
animals. In the HF animals, there also was evidence of
dysfunction of the atrioventricular node; for example, the PQ
interval was prolonged in both the anesthetized animals and
in the isolated Langendorff-perfused hearts (Figure 2B, Table 1).
Dilatation of Right Atrium and Nodal Dysfunction
After Infarction in the LV
qPCR Measurements
The myocardial infarction rats showed characteristic signs of
HF and are referred to as HF animals. Compared with the
sham-operated animals, the HF animals showed an increase
in the LV diastolic pressure of 317% and decreases in the LV
The relative abundance of 91 transcripts important for electric
activity (transcripts for ion channels; gap junction channels;
Ca2⫹-, Na⫹-, and H⫹-handling proteins; and receptors) was
measured in tissue samples taken from the SAN and atrial
Yanni et al
HF and Ion Channels in the Sinoatrial Node
499
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 12, 2017
Figure 2. HF-dependent changes in SAN function. A, Mean⫾SEM (n⫽10) normal heart rate (in vivo) and intrinsic heart rate (in vitro) in
sham-operated and HF animals. B, ECG recordings from sham-operated and HF animals (anesthetized). C, ECG-like recordings from
Langendorff-perfused hearts from sham-operated and HF animals. Measurement of BCL and PQ interval are shown. D, Mean⫾SEM
(n⫽10) SNRTc in sham-operated and HF animals. E, ECG-like recordings from Langendorff-perfused hearts from sham-operated and
HF animals showing recording of SNRT. Final paced beats (10-s pacing) are shown (stimuli indicated by Œ). SNRT was time to first
SAN beat after train. BCL indicates basic cycle length; SNRT, SAN recovery time; SNRTc, corrected SAN recovery time. Other abbreviations as in Figure 1. *Significantly different from corresponding data from sham-operated animals.
muscle from HF and sham-operated animals. Of the genes
studied, during HF, 53 changed significantly in the SAN, but
only 1 changed significantly in the atrial muscle (Table 2).
The qPCR results are discussed in detail later. Many differences were observed in the abundance of transcripts between
the SAN and atrial muscle, but generally, these are not commented on because similar data are discussed elsewhere.18
HF-Induced Changes in Two Key Regulatory
Genes in SAN
Figure 1C shows that during HF, there was a substantial
increase in NHE1 in the SAN, but, surprisingly, not in the
atrial muscle. T-box transcription factor Tbx3 plays a key role
in the regulation of gene expression in the SAN. As expected,
it was highly expressed in the SAN but not in the atrial
muscle (Figure 1C). During HF, there was a substantial
increase in Tbx3 in the SAN (Figure 1C).
Upregulation of Naⴙ and Ca2ⴙ Channel Subunits
Expression of the cardiac Na⫹ channel Nav1.5, which is
largely responsible for INa, was not affected by HF (Figure
3A). However, neuronal Na⫹ channels are known to be
expressed in cardiac myocytes,19 and the expression of the
neuronal Na⫹ channels Nav1.1 and Nav1.3 was approximately doubled in the SAN (but not in atrial muscle) during
HF (Figure 3A), which also is the case for the Na⫹ channel
accessory subunits Nav␤1 to Nav␤3 (Figure 3A). The Ca2⫹
channels Cav1.2 and Cav3.1 are largely responsible for ICa,L
and ICa,T, respectively, and expression of both was significantly higher in the SAN (but not in atrial muscle) during HF
(Figure 3B). The expression of the Ca2⫹ channel accessory
subunits Cav␤2, Cav␤3, Cav␣2␦1, Cav␣2␦2, and Cav␥7 was
significantly higher in the SAN (but not in atrial muscle)
during HF (Figure 3B).
Is Upregulation of Kir2.4, TASK1, and TWIK2
Responsible for Decrease of Intrinsic Heart Rate
During HF?
Kv1.4, Kv4.2, Kv4.3, KChIP2, and MIRP3 are K⫹ channel
subunits responsible for the transient outward K⫹ current
(Ito); Kv1.5, ERG, and KvLQT1 are delayed rectifier K⫹
channels responsible for IK,ur, IK,r, and IK,s, respectively; and
Kv1.2 and Kv2.1 are 2 other delayed rectifier K⫹ channels
important in the rat.20,21 Some of these K⫹ channel subunits
(Kv1.4, Kv2.1, ERG, KvLQT1, KCHiP2, and MIRP3) were
more highly expressed in the SAN (but not in atrial muscle)
during HF (Figure 4A). The following 4 K⫹ channels active
in diastole and, therefore, capable of slowing pacemaking
were upregulated in the SAN (but not in atrial muscle) during
HF: Kir2.4 (Kir2 channels are responsible for IK,1) and the
twin pore K⫹ channels TASK1, TWIK1, and TWIK2 (Figure
4). The abundance of the small-conductance Ca2⫹-activated
K⫹ channels SK1 and SK3 was greatly increased in the SAN
during HF (Figure 4B). In contrast, the abundance of another
small-conductance Ca2⫹-activated K⫹ channel, SK2, was
decreased in the atrial muscle during HF (Figure 4B).
500
Circ Heart Fail
July 2011
Table 2. Summary of Transcripts Judged To Change
Significantly (Based on the FDR) During HF in the SAN
Table 2.
Transcripts That Change With HF in SAN
Receptors
Continued
Transcripts That Change With HF in SAN
Adjusted P (FDR)
Inward current carrying channel subunits
Adjusted P (FDR)
1 ␣1A adrenergic receptor
0.122
1 HCN2*
0.066
1 ␣1B adrenergic receptor*
0.097
1 HCN4*
0.069
1 ␤1 adrenergic receptor*
0.083
1 Nav1.1
0.069
1 ␤2 adrenergic receptor*
0.088
1 Nav1.3*
0.066
1 Nav␤1*
0.069
1 Nav␤2*
0.087
1 Nav␤3*
0.078
1 Cav1.2*
0.088
1 Cav3.1*
0.066
1 Cav␤2
0.119
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 12, 2017
1 Cav␤3*
0.103
1 Cav␣2␦1*
0.083
1 Cav␣2␦2*
0.088
1 Cav␥7*
0.097
Outward current carrying channel subunits
1 Kv1.4*
0.088
1 Kv2.1
0.087
1 ERG
0.179
1 KvLQT1
0.16
1 Kir2.4*
0.066
1 Kir6.2
0.103
1 TASK1
0.083
1 TWIK1
0.164
1 TWIK2*
0.066
1 SK1
0.119
1 SK3*
0.069
1 KChIP2*
0.088
1 MIRP3
0.066
Ca2⫹-handling proteins
1 TRPC3*
0.083
1 TRPC4
0.145
1 RYR2
0.12
1 RYR3*
0.069
1 Calsequestrin 2
0.134
1 PMCA1*
0.088
1 Type 1 IP3 receptor*
0.088
1 Type 2 IP3 receptor
0.165
1 Type 3 IP3 receptor
0.192
Connexins
1 Cx30.2*
0.182
1 Cx45
0.104
Pumps and exchangers
1 ␣3 Na⫹-K⫹ pump*
⫹
⫹
0.088
2 ␤1 Na -K pump
0.066
1 AE1
0.103
1 AE2*
0.088
1 NBC1*
0.179
1 NHE1*
0.088
(Continued)
1 A1 adenosine receptor*
0.088
1 P2X4
0.069
Other
1 Tbx3*
0.092
1 ␣MHC
0.138
1 ␤MHC*
0.066
Arrows indicate direction of change. In the atrial muscle, FDR showed a
significant change in SK2 (adjusted P⫽0.184), and the t test showed
significant differences in Nav␤3 and the ␣3 isoform of the Na⫹-K⫹ pump.
AE indicates Cl⫺-HCO3⫺ exchanger; FDR, false discovery rate; HCN,
hyperpolarization-activated cyclic nucleotide-gated channels; HF, heart
failure; IP3, inositol 1,4,5-trisphosphate; MHC, myosin heavy chain; NBC1,
Na⫹-HCO3⫺ cotransporter 1; NHE1, Na⫹-H⫹ exchanger; PMCA1, plasma
membrane Ca2⫹ pump isoform 1; RYR, ryanodine receptor; SAN, sinoatrial
node; Tbx3, T-box transcription factor; TRPC, transient receptor potential
channels.
*Transcripts judged to change significantly based on an unpaired t test.
Unexpected Changes in HCN Channels, Transient
Receptor Potential Channels, and Connexins
HCN channels are responsible for If and, unexpectedly,
HCN2 and HCN4 expression was significantly higher in
the SAN (but not in atrial muscle) during HF (Figure 5A).
Transient receptor potential channels (TRPC) have been
proposed to be store (ie, sarcoplasmic reticulum [SR])
operated Ca2⫹ channels involved in SAN pacemaking,22
and again unexpectedly, TRPC3 and TRPC4 expression
was significantly higher in the SAN during HF; TRPC4
was not detected in the atrial muscle (Figure 5B). The
connexin Cx30.2 is primarily expressed in the SAN, and its
abundance was increased ⬇2-fold in the SAN during HF
(Figure 5C), and whether this can explain the increase in
the SAN conduction time in HF is considered in the
Discussion section. The abundance of Cx45 was also
significantly increased in the SAN during HF (Figure 5C).
Changes in Ca2ⴙ-, Naⴙ-, and Hⴙ-Handling
Proteins in the SAN During HF
Spontaneous diastolic release of intracellular Ca2⫹ from the
SR has been implicated in the pacemaker activity of the
SAN.11 The expression of the following 5 SR Ca2⫹ release
channels was significantly increased in the SAN (but not in
atrial muscle) during HF: the ryanodine receptors RYR2 and
RYR3 and types 1 to 3 inositol 1,4,5-trisphosphate (IP3)
receptors (Figure 6). Two other Ca2⫹-handling proteins,
calsequestrin 2 (SR Ca2⫹-binding protein) and PMCA1
(plasma membrane Ca2⫹ ATPase 1), also were increased in
the SAN (but not in atrial muscle) during HF (Figure 6).
Regulation of intracellular Ca2⫹ is tied to the regulation of
intracellular Na⫹ and H⫹. Of the 3 Na⫹-K⫹ pump isoforms
(␣1 to ␣3) responsible for the regulation of intracellular Na⫹,
Yanni et al
HF and Ion Channels in the Sinoatrial Node
501
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 12, 2017
Figure 3. Relative abundance of Na⫹ (A)
and Ca2⫹ (B) channel transcripts in AM
and SAN in sham-operated and HF animals. Relatively poorly expressing transcripts are shown on the left, and more
highly expressed transcripts are shown
on the right; dotted line separates
the 2 groups. Data are presented as
mean⫾SEM (n⫽7). Abbreviations as in
Figure 1. *Significantly different (false discovery rate) from corresponding data
from sham-operated animals.
the expression of the ␣3 isoform was significantly greater in
the SAN than in the atrial muscle (the other 2 isoforms were
more uniformly distributed in SAN and atrial muscle) (Figure
7A). The expression of the ␣3 isoform was significantly
increased in the SAN during HF, whereas the expression of
the Na⫹-K⫹ pump ␤1 subunit was depressed (Figure 7A).
The expression of intracellular H⫹-handling proteins also was
altered during HF: The expression of AE1 and AE2 (Cl⫺HCO3⫺ exchangers), NBC1 (Na⫹-HCO3⫺ cotransporter 1),
and MCT1 (monocarboxylate transporter 1) was significantly
increased in the SAN (but not in atrial muscle) during HF
(Figure 7A). An increase in NHE1 (Figure 1C) has already
been commented on.
Upregulation of Receptors in SAN During HF
There was an increase in expression of the ␣1A, ␣1B, ␤1, and
␤2 adrenergic receptors in the SAN (but not in atrial muscle)
during HF (Figure 7B). The expression of the A1 adenosine
receptor also was significantly increased in the SAN during
HF (Figure 7B).
Upregulation of a Hypertrophy Marker
and Summary
An increase in the expression of ␤-myosin heavy chain
(MHC) occurs as part of a hypertrophy program.23 Perhaps as
expected, the expression of ␤MHC was significantly higher
in the SAN (but, surprisingly, not in atrial muscle) during HF
(Figure 7C). The expression of ␣MHC also was significantly
higher in the SAN (but not in atrial muscle) during HF
(Figure 7C). However, the increase in expression of ␤MHC
was greater than that of ␣MHC.
Hierarchical clustering was carried out on a subset of the
transcripts as shown in Table 2 (see the online-only Data
Supplement for details), which grouped the SAN samples
from the HF animals separately from the SAN samples from
the sham-operated animals (Figure 8A). The atrial muscle
samples were separated from the SAN samples, but there was
no systematic separation of the HF and sham-operated atrial
muscle samples on the basis of these transcripts. Multidimensional scaling (Figure 8B) was carried out with the data for all
transcripts (online-only Data Supplement). Figure 8B shows
each sample in a 2D matrix separated from the others by a
502
Circ Heart Fail
July 2011
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 12, 2017
Figure 4. A and B, Relative abundance of
K⫹ channel transcripts in AM and SAN in
sham-operated and HF animals. See Figure 3 legend for key. Abbreviations as in
Figure 1.
distance proportional to the aggregate of all transcripts. The
atrial muscle samples, which are grouped closely together,
are separate from the SAN samples, which are less closely
grouped (ie, there was more variation between SAN samples)
(Figure 8B). The overall aggregate of all transcripts separates
the SAN samples, but not the atrial muscle samples, from the
HF and sham-operated animals. In summary, the hierarchical
clustering and the multidimensional scaling shows that ion
channel expression is distinct between (1) the SAN and atrial
muscle and (2) the SAN (but not atrial muscle) from HF and
sham-operated animals.
Discussion
To our knowledge, the present study shows for the first time
that HF results in SAN dysfunction and widespread remodeling
(at the mRNA level) of ion channels and related proteins (Figure
8, Table 2).
Stimulus for SAN Remodeling in HF
It is not obvious why myocardial infarction in the LV should
cause dysfunction of the SAN located on the right side of the
heart. One possible explanation is that the changes in the
SAN result from neurohormonal changes known to occur in
HF. Another possibility is that the reduced contractility of the
LV and the resulting decrease in the LV ejection fraction
(Table 1) causes (1) fluid retention and (2) a build-up of blood
in the left atrium; the pulmonary vasculature; the right side of
the heart; and, finally, the right atrium (Figure 1). As a result,
the right atrium is volume overloaded and stretched, and this
could initiate changes in gene expression. There was a
substantial increase in NHE1 in the SAN during HF (Table
2). An increase in NHE1 activity (including upregulation of
NHE1) has been observed in many models of HF.16 Activation of NHE1 is known to occur in response to stretch.16 It is
believed that NHE1 triggers intracellular signaling pathways
to result in hypertrophy and remodeling.16 It is interesting to
speculate that the HF-induced nodal dysfunction will be
cariporide sensitive. Could NHE1 be acting on Tbx3? Tbx3
has been shown to play a key role in the development of the
SAN, controlling the expression of many genes that are key to
the pacemaker function of the SAN.24 For example, ectopic
expression of Tbx3 in the atrial muscle causes changes in the
expression of ⬇500 genes and the expression of an SAN
phenotype.25 Tbx3 expression was substantially greater in the
Yanni et al
HF and Ion Channels in the Sinoatrial Node
503
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 12, 2017
Figure 5. Relative abundance of HCN (A),
TRPC (B), and connexin (C) transcripts in
AM and SAN in sham-operated and HF
animals. See Figure 3 legend for key.
HCN indicates hyperpolarization-activated
cyclic nucleotide-gated channels; TRPC,
transient receptor potential channels.
Other abbreviations as in Figure 1.
504
Circ Heart Fail
July 2011
Figure 6. Relative abundance of Ca2⫹handling transcripts in AM and SAN in
sham-operated and HF animals. See Figure 3 legend for key. IP3 indicates inositol
1,4,5-trisphosphate; NCX1, Na⫹-Ca2⫹
exchanger; PMCA1, plasma membrane
Ca2⫹ pump isoform 1; RYR3, ryanodine
receptor 3; SERCA2a, sarcoplasmic reticulum Ca2⫹ ATPase. Other abbreviations
as Figure 1.
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 12, 2017
SAN during HF (Table 2). Many of the genes Tbx3 has been
shown to upregulate25 were upregulated in the SAN during
HF, as follows: Cav3.1, Cav␤2, Cav␣2␦1, Cav␣2␦2, ERG,
HCN2, HCN4, Cx30.2, Cx45, and the type 1 IP3 receptor
(Table 2).
It is interesting and unexpected that there was no upregulation of NHE1 in the atrial muscle despite the dilatation of
the right atrium (Figure 1). This finding could explain the
relative lack of remodeling of ion channels and so forth in the
atrial muscle, whereas 58% of transcripts studied changed in
the SAN and only 1% changed in the atrial muscle (Table 2).
Explanation of HF-Induced Decrease of Intrinsic
Heart Rate
In the present study (Figure 2), as in previous studies of
humans and animal models,6 – 8 HF was accompanied by
dysfunction of the SAN. For example, HF was accompanied
by a decrease in the intrinsic heart rate (Figure 2). This must
be the result of a decrease in inward current or increase in
outward current during the pacemaker potential (ie, diastole).
The HCN channels (primarily HCN4) are responsible for If,
which is a key inward current during the pacemaker potential.
Previously, it has been reported that with other animal models
of HF, both the density of If and the expression of HCN4
decline in the SAN in response to HF.7,15 Therefore, it was
surprising that in the present study, an increase in HCN2 and
HCN4 expression in response to HF was seen (Table 2). We
suggest that this is an HF model difference, and in this case,
the increase in HCN expression may be a compensatory
response to the changes in K⫹ channels (discussed later). The
increase in HCN4 expression may be driven by the increase
in Tbx3 expression, as discussed previously, and for the
control and HF atrial and SAN samples, there is a significant
correlation between HCN4 expression and Tbx3 expression
(R2⫽0.72, P⬍0.0001).
Changes in Na⫹ channel subunits (responsible for inward
INa) cannot explain the decrease in intrinsic heart rate. During
HF, there was an increase in Nav1.1, Nav1.3, Nav␤2, and
Nav␤3 in the SAN (Table 2), and yet, block of neuronal ion
channels (eg, Nav1.3)26 and knockout of Nav␤227 have been
shown to decrease the heart rate. Again, it is difficult to
explain the decrease in intrinsic heart rate in terms of Ca2⫹
channels because expression of Cav1.2 and Cav3.1, which are
responsible for ICa,L and ICa,T (known to contribute to
pacemaking),11 was increased (Table 2). There was an increase in the expression of various Ca2⫹ channel accessory
subunits (Cav␣2␦1, Cav␣2␦2, Cav␤2, Cav␤3, and Cav␥7) in
the SAN during HF (Table 2); however, this is expected to
lead to an increase in ICa because they have been shown to
increase Ca2⫹ channel incorporation into the cell membrane.28 Changes in TRPC (perhaps responsible for inward
store-operated Ca2⫹ current involved in SAN pacemaking22)
also cannot explain the decrease in intrinsic heart rate because
expression of TRPC3 and TRPC4 was increased in the SAN
during HF (Table 2).
However, an increase in the expression of K⫹ channels
(ERG, KvLQT1, Kir2.4, TASK1, TWIK1, and TWIK2) (Table 2) responsible for outward currents during diastole could
be responsible for the decrease in the intrinsic heart rate. IK,r
and IK,s (for which ERG and KvLQT1, respectively, are
responsible) are activated during the action potential, and
they deactivate during diastole, allowing the development
of the pacemaker potential.12 An increase in IK,r and IK,s,
therefore, may slow pacemaking. However, IK,s has been
reported to decrease in the rabbit during HF.15 Kir2.4 (in part
responsible for IK,1), TASK1, TWIK1, and TWIK2 are
expected to carry outward current during diastole, and their
increased expression in the SAN during HF (Table 2) is
expected to slow pacemaking; the increase in TASK1 may be
particularly important because TASK1 was ⬎10 times more
abundant than the other channels.
Diastolic Ca2⫹ release from the SR, by activating inward
INaCa, contributes to pacemaking.29 There was an increase in
the expression of various SR Ca2⫹ release channels (RYR2,
Yanni et al
HF and Ion Channels in the Sinoatrial Node
505
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 12, 2017
Figure 7. Relative abundance of transcripts for pumps, exchangers, and
cotransporters (A); receptors (B); and miscellaneous proteins (C) in AM and SAN in
sham-operated and HF animals. See Figure 3 legend for key. AE indicates Cl⫺HCO3⫺ exchanger; ANP, atrial natriuretic
peptide; BNP, brain natriuretic peptide;
NBC, Na⫹-HCO3⫺ cotransporter; MCT1,
monocarboxylate transporter 1; MHC,
myosin heavy chain. Other abbreviations
as in Figure 1.
506
Circ Heart Fail
July 2011
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 12, 2017
Figure 8. A, Hierarchical cluster analysis of selected gene expression in AM and SAN. Each row represents a gene, and each column
represents a tissue sample. Red indicates values above average for a given transcript, white indicates average values, and blue indicates values below average (after normalization of log2 expression values). Trees above the block diagram represent correlation of
samples, and trees on the side represent correlation of transcripts. Length of each twig indicates closeness of correlation
(short⫽close). B, MDS. MDS indicates multidimensional scaling; RA, right atrial muscle. Other abbreviations as in Figures 1, 5, and 6.
RYR3, and types 1 to 3 IP3 receptors) in the SAN during HF
(Table 2). Expression of the type 1 IP3 receptor also has been
shown to increase in the ventricles during HF.30 However,
intuitively, the increase in expression of these SR Ca2⫹
release channels is expected to increase the intrinsic heart rate
rather than to decrease it. Nevertheless, the functional role of
RYR3 in cardiac tissue has yet to be studied, and the greater
abundance of RYR3 in the SAN compared to the atrial
muscle (also seen in mouse and rabbit31,32) and its further
increase in the SAN during HF (Table 2) suggest that RYR3
could play an important role in SAN function. Calsequestrin
2 is a high-capacity Ca2⫹-binding protein that is found in the
lumen of the SR. Overexpression of calsequestrin 2 leads to
a decrease in intracellular Ca2⫹ and is associated with
hypertrophy.33 It is possible that the increase in the expression
of calsequestrin 2 in the SAN during HF may slow pacemak-
Yanni et al
ing as a result of the inhibition of Ca2⫹ release. Finally,
expression of the A1 adenosine receptor was substantially
increased in the SAN during HF (Table 2), which is consistent with studies on the atrioventricular node from a rabbit HF
model where an increase in sensitivity to adenosine was
reported.5 By activating the A1 receptor, adenosine is known
to cause bradycardia in humans,34 and overexpression of the
A1 receptor in mice also causes a reduction in the intrinsic
heart rate.35 Therefore, it is possible that the increased
expression of the A1 adenosine receptor in the SAN could
contribute to the decline in the intrinsic heart rate. In
summary, we suggest that the decrease in the intrinsic heart
rate in this model of HF may involve an upregulation of ERG,
KvLQT1, Kir2.4, TASK1, TWIK1, TWIK2, calsequestrin 2,
and the A1 adenosine receptor in the SAN.
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 12, 2017
HF-Induced Changes in Action
Potential Duration?
It is well-known that HF causes an increase in action potential
duration in the atria and ventricles.36,37 If such an increase
occurred in the SAN, it could contribute to the slowing of
pacemaking (although such a change in action potential duration
does not occur in the SAN in a rabbit model of HF at least15).
The increase in Na⫹ and Ca2⫹ channel subunits in the SAN
discussed previously could increase action potential duration,
but the increase in ERG and KvLQT1 as well as the increase in
the Ito channel subunits (Kv1.4, KChIP2, and MIRP3) and the
Ca2⫹-activated K⫹ channels SK1 and SK3 (known to contribute
to action potential repolarization38,39) in the SAN during HF
(Table 2) are expected to decrease action potential duration.
Connexins
It has been shown clinically that HF is associated with an
increase in the SAN conduction time.6 SAN conduction time
depends on electric coupling between myocytes provided by
gap junctions made up of connexins. There was an increase in
the expression of Cx30.2 and Cx45 in the SAN during HF
(Table 2). Cx30.2 protein has been shown to be highly
abundant in the SAN,40 which suggests that Cx30.2 plays an
important role in the SAN. The conductance of both Cx30.2
(⬇9 ps) and Cx45 (20 to 40 ps) is low.40,41 The increase in
expression of Cx30.2 and Cx45 by increasing the complement of connexins could increase electric coupling (and,
therefore, accelerate action potential conduction). However,
Cx30.2 is able to form small conductance heteromeric gap
junctions with other connexins (which alone would produce
large conductance gap junctions); therefore, the increase in
expression of Cx30.2 could decrease electric coupling and
slow action potential conduction. The conduction velocity
also depends on other factors, such as fibrosis.
Exchangers and Transporters Involved in
Regulation of Intracellular Naⴙ and pH
Regulation of intracellular ions in the SAN is likely to be
different during HF because in the SAN during HF the expression of the ␣3 Na⫹-K⫹ pump isoform, AE1, AE2, NBC1,
MCT1, and NHE1 was increased, and the expression of the
Na⫹-K⫹ pump ␤1 subunit was decreased (Table 2). Changes in
Ca2⫹-handling proteins already have been discussed. The
HF and Ion Channels in the Sinoatrial Node
507
changes in the regulation of intracellular Na⫹ and H⫹ could be
important for pacemaking because intracellular Na⫹ and H⫹ are
known to be linked with intracellular Ca2⫹, which is involved
with pacemaking as already discussed. In addition, pH is known
to have a profound effect on pacemaking.42 However, there may
not be a change in intracellular pH because the opposite effects of
the increases in expression of NHE1 and AE1 and AE2 may
balance one another; this occurs in other systems in hypertrophy.16
Receptors
The expression of the ␣1A, ␣1B, ␤1, and ␤2 adrenergic receptors
was increased in the SAN during HF (Table 2). This is expected
to increase the sensitivity of the SAN to sympathetic stimulation
and could explain how the HF animals were able to maintain the
same heart rate as the sham-operated animals even though the
intrinsic heart rate of the HF animals was slower (Figure 2). In
contrast, the expression of the M2 muscarinic receptor was not
altered (Figure 7). In contrast to these findings, in a rabbit model
of HF, there was an increase in acetylcholine sensitivity and no
change in norepinephrine sensitivity.8
Conclusions
We have identified many genes that are altered in the SAN
during HF. These include an increase in the expression of ERG,
KvLQT1, Kir2.4, TASK1, TWIK1, TWIK2, calsequestrin 2, and
the A1 adenosine receptor, which could be responsible for the
slowing of the intrinsic heart rate during HF, and an increase in
the expression of Cx30.2, which perhaps is responsible for the
increase in the SAN conduction time. We suggest that these
changes are brought about by a stretch-mediated increase in
NHE1. The identification of the mechanisms that lead to SAN
dysfunction during HF will aid the development of moreeffective clinical treatments, such as gene therapy, in the future.
Sources of Funding
This work was funded by a program grant from the British Heart
Foundation (RG/06/005).
Disclosures
None.
References
1. Cleland JG, Khand A, Clark A. The heart failure epidemic: exactly how
big is it? Eur Heart J. 2001;22:623– 626.
2. Uretsky BF, Sheahan RG. Primary prevention of sudden cardiac death in
heart failure: will the solution be shocking? J Am Coll Cardiol. 1997;30:
1589 –1597.
3. Stevenson WG, Stevenson LW, Middlekauff HR, Saxon LA. Sudden
death prevention in patients with advanced ventricular dysfunction.
Circulation. 1993;88:2953–2961.
4. Dobrzynski H, Boyett MR, Anderson RH. New insights into pacemaker
activity: promoting understanding of sick sinus syndrome. Circulation.
2007;115:1921–1932.
5. Muir-Nisbet A. The Electrophysiology of the Atrioventricular Node in
Normal and Failing Rabbit Hearts [dissertation]. Glasgow, UK: University of Glasgow; 2008.
6. Sanders P, Kistler PM, Morton JB, Spence SJ, Kalman JM. Remodeling
of sinus node function in patients with congestive heart failure: reduction
in sinus node reserve. Circulation. 2004;110:897–903.
7. Zicha S, Fernandez-Velasco M, Lonardo G, L’Heureux N, Nattel S. Sinus
node dysfunction and hyperpolarization-activated (HCN) channel subunit
remodeling in a canine heart failure model. Cardiovasc Res. 2005;66:
472– 481.
8. Opthof T, Coronel R, Rademaker HM, Vermeulen JT, Wilms-Schopman
FJ, Janse MJ. Changes in sinus node function in a rabbit model of heart
508
9.
10.
11.
12.
13.
14.
15.
16.
17.
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 12, 2017
18.
19.
20.
21.
22.
23.
24.
25.
26.
Circ Heart Fail
July 2011
failure with ventricular arrhythmias and sudden death. Circulation. 2000;
101:2975–2980.
White CW. Abnormalities in baroreflex control of heart rate in canine
heart failure. Am J Physiol. 1981;240:H793–H799.
Nattel S, Maguy A, Le Bouter S, Yeh YH. Arrhythmogenic ion-channel
remodeling in the heart: heart failure, myocardial infarction, and atrial
fibrillation. Physiol Rev. 2007;87:425– 456.
Mangoni ME, Nargeot J. Genesis and regulation of the heart automaticity.
Physiol Rev. 2008;88:919 –982.
Irisawa H, Brown HF, Giles W. Cardiac pacemaking in the sino-atrial
node. Physiol Rev. 1993;73:197–227.
Miake J, Marban E, Nuss HB. Biological pacemaker created by gene
transfer. Nature. 2002;419:132–133.
Ryu KH, Tanaka N, Ross J Jr. Effects of a sinus node inhibitor on the
normal and failing rabbit heart. Basic Res Cardiol. 1996;91:131–139.
Verkerk AO, Wilders R, Coronel R, Ravesloot JH, Verheijck EE. Ionic
remodeling of sinoatrial node cells by heart failure. Circulation. 2003;
108:760 –766.
Cingolani HE, Ennis IL. Sodium-hydrogen exchanger, cardiac overload,
and myocardial hypertrophy. Circulation. 2007;115:1090 –1100.
Maczewski M, Mackiewicz U. Effect of metoprolol and ivabradine on left
ventricular remodelling and Ca2⫹ handling in the post-infarction rat heart.
Cardiovasc Res. 2008;79:42–51.
Tellez JO, Dobrzynski H, Yanni J, Billeter R, Boyett MR. Effects of
aging on gene expression in the rat sinoatrial node. J Mol Cell Cardiol.
2006;40:982–982.
Maier SKG, Westenbroek RE, Yamanushi TT, Dobrzynski H, Boyett
MR, Catterall WA, Scheuer T. An unexpected requirement for brain-type
sodium channels for control of heart rate in the mouse sinoatrial node.
Proc Natl Acad Sci U S A. 2003;100:3507–3512.
Bou-Abboud E, Nerbonne JM. Molecular correlates of the calciumindependent, depolarization-activated K⫹ currents in rat atrial myocytes.
J Physiol. 1999;517:407– 420.
O’Connell KM, Whitesell JD, Tamkun MM. Localization and mobility of
the delayed-rectifer K⫹ channel Kv2.1 in adult cardiomyocytes. Am J
Physiol. 2008;294:H229 –H237.
Ju YK, Chu Y, Chaulet H, Lai D, Gervasio OL, Graham RM, Cannell
MB, Allen DG. Store-operated Ca2⫹ influx and expression of TRPC
genes in mouse sinoatrial node. Circ Res. 2007;100:1605–1614.
Gupta MP. Factors controlling cardiac myosin-isoform shift during hypertrophy and heart failure. J Mol Cell Cardiol. 2007;43:388 – 403.
Hoogaars WM, Engel A, Brons JF, Verkerk AO, de Lange FJ, Wong LY,
Bakker ML, Clout DE, Wakker V, Barnett P, Ravesloot JH, Moorman
AF, Verheijck EE, Christoffels VM. Tbx3 controls the sinoatrial node
gene program and imposes pacemaker function on the atria. Genes Dev.
2007;21:1098 –1112.
Hoogaars WMH. The Role of Tbx3 in the Formation of the Cardiac
Conduction System [dissertation]. Amsterdam, The Netherlands: University of Amsterdam; 2007.
Du Y, Huang X, Wang T, Han K, Zhang J, Xi Y, Wu G, Ma A.
Downregulation of neuronal sodium channel subunits Nav1.1 and Nav1.6
in the sinoatrial node from volume-overloaded heart failure rat. Pflugers
Arch. 2007;454:451– 459.
27. Maier S, Westenbroek RE, Chen C, Marble DR, Feigl EO, Isom LL,
Catterall WA, Scheuer T. The ␤2-subunit of voltage-gated sodium
channels is required for maintenance of regular sinus rhythm in the heart.
Biophys J. 2002;82:609a.
28. Walker D, De Waard M. Subunit interaction sites in voltage-dependent Ca2⫹
channels: role in channel function. Trends Neurosci. 1998;21:148–154.
29. Lancaster MK, Jones SA, Harrison SM, Boyett MR. Intracellular Ca2⫹
and pacemaking within the rabbit sinoatrial node: heterogeneity of role
and control. J Physiol. 2004;556:481– 494.
30. Go LO, Moschella MC, Watras J, Handa KK, Fyfe BS, Marks AR.
Differential regulation of two types of intracellular calcium release
channels during end-stage heart failure. J Clin Invest. 1995;95:888 – 894.
31. Masumiya H, Yamamoto H, Hemberger M, Tanaka H, Shigenobu K,
Chen SR, Furukawa T. The mouse sino-atrial node expresses both the
type 2 and type 3 Ca2⫹ release channels/ryanodine receptors. FEBS Lett.
2003;553:141–144.
32. Tellez JO, Dobrzynski H, Greener ID, Graham GM, Laing E, Honjo H,
Hubbard SJ, Boyett MR, Billeter R. Differential expression of ion channel
transcripts in atrial muscle and sinoatrial node in rabbit. Circ Res. 2006;
99:1384 –1393.
33. Sato Y, Ferguson DG, Sako H, Dorn GW, II, Kadambi VJ, Yatani A, Hoit
BD, Walsh RA, Kranias EG. Cardiac-specific overexpression of mouse
cardiac calsequestrin is associated with depressed cardiovascular function
and hypertrophy in transgenic mice. J Biol Chem. 1998;273:28470–28477.
34. DiMarco JP, Sellers TD, Berne RM, West GA, Belardinelli L. Adenosine:
electrophysiologic effects and therapeutic use for terminating paroxysmal
supraventricular tachycardia. Circulation. 1983;68:1254 –1263.
35. Matherne GP, Linden J, Byford AM, Gauthier NS, Headrick JP.
Transgenic A1 adenosine receptor overexpression increases myocardial
resistance to ischemia. Proc Natl Acad Sci U S A. 1997;94:6541– 6546.
36. Li D, Melnyk P, Feng J, Wang Z, Petrecca K, Shrier A, Nattel S. Effects
of experimental heart failure on atrial cellular and ionic electrophysiology. Circulation. 2000;101:2631–2638.
37. Janse MJ. Electrophysiological changes in heart failure and their relationship to arrhythmogenesis. Cardiovasc Res. 2004;61:208 –217.
38. Xu Y, Tuteja D, Zhang Z, Xu D, Zhang Y, Rodriguez J, Nie L, Tuxson HR,
Young JN, Glatter KA, Vazquez AE, Yamoah EN, Chiamvimonvat N.
Molecular identification and functional roles of a Ca2⫹-activated K⫹ channel
in human and mouse hearts. J Biol Chem. 2003;278:49085–49094.
39. Tuteja D, Xu D, Timofeyev V, Lu L, Sharma D, Zhang Z, Xu Y, Nie L,
Vazquez AE, Young JN, Glatter KA, Chiamvimonvat N. Differential
expression of small-conductance Ca2⫹-activated K⫹ channels SK1, SK2,
and SK3 in mouse atrial and ventricular myocytes. Am J Physiol Heart
Circ Physiol. 2005;289:H2714 –H2723.
40. Kreuzberg MM, Sohl G, Kim JS, Verselis VK, Willecke K, Bukauskas
FF. Functional properties of mouse connexin30.2 expressed in the conduction system of the heart. Circ Res. 2005;96:1169 –1177.
41. Boyett MR, Inada S, Yoo S, Li J, Liu J, Tellez JO, Greener ID, Honjo H,
Billeter R, Lei M, Zhang H, Efimov IR, Dobrzynski H. Connexins in the
sinoatrial and atrioventricular nodes. Adv Cardiol. 2006;42:175–197.
42. Aberra A, Komukai K, Howarth FC, Orchard CH. The effect of acidosis
on the ECG of the rat heart. Exp Physiol. 2001;86:27–31.
CLINICAL PERSPECTIVE
It is estimated that ⬎20 million people have heart failure (HF) worldwide, and bradyarrhythmias account for about half
of the sudden deaths of patients with HF. Consistent with the high incidence of bradyarrhythmic deaths, HF is known to
cause dysfunction of the cardiac conduction system, including the sinoatrial node (SAN). The dysfunction of the SAN is
likely to be the result of a remodeling of the ion channels and related proteins responsible for the pacemaker activity of
the SAN, and the aim of the study was to investigate this. HF was induced in rats by the ligation of the proximal left
coronary artery. In the HF animals, there was an increase in the left ventricular diastolic pressure and a decrease in the left
ventricular systolic pressure and SAN dysfunction, the intrinsic heart rate was reduced, and the corrected SAN recovery
time was increased. Quantitative polymerase chain reaction was used to measure gene expression in the SAN and
surrounding atrial muscle. There was a widespread remodeling of ion channels, gap junction channels, calcium-, sodium-,
and proton-handling proteins, and receptors in the SAN. The decrease of the intrinsic heart rate can be explained by an
upregulation of various potassium channels. Curiously, the atrial muscle was much less sensitive to HF. Of the 91 genes
studied, 41% changed in the SAN, but only 7% changed in the atrial muscle. The elucidation of the mechanisms responsible
for SAN dysfunction in HF opens the way to the development of new treatments.
Changes in Ion Channel Gene Expression Underlying Heart Failure-Induced Sinoatrial
Node Dysfunction
Joseph Yanni, James O. Tellez, Michal Maczewski, Urszula Mackiewicz, Andrzej Beresewicz,
Rudi Billeter, Halina Dobrzynski and M.R. Boyett
Downloaded from http://circheartfailure.ahajournals.org/ by guest on May 12, 2017
Circ Heart Fail. 2011;4:496-508; originally published online May 12, 2011;
doi: 10.1161/CIRCHEARTFAILURE.110.957647
Circulation: Heart Failure is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX
75231
Copyright © 2011 American Heart Association, Inc. All rights reserved.
Print ISSN: 1941-3289. Online ISSN: 1941-3297
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circheartfailure.ahajournals.org/content/4/4/496
Data Supplement (unedited) at:
http://circheartfailure.ahajournals.org/content/suppl/2011/05/12/CIRCHEARTFAILURE.110.957647.DC1
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Circulation: Heart Failure can be obtained via RightsLink, a service of the Copyright Clearance Center, not
the Editorial Office. Once the online version of the published article for which permission is being requested is
located, click Request Permissions in the middle column of the Web page under Services. Further information
about this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation: Heart Failure is online at:
http://circheartfailure.ahajournals.org//subscriptions/
SUPPLEMENTAL MATERIAL
Changes in Ion Channel Gene Expression Underlying
Heart Failure-induced Sinoatrial Node Dysfunction
J. Yanni, J.O. Tellez, M. Mączewski, U. Mackiewicz, A. Beresewicz, R. Billeter, H. Dobrzynski,
M.R. Boyett
Statistical analysis of qPCR data
The qPCR data were analysed in three ways. First, they were analysed using the ΔCt and ΔΔCt methods
implemented using the methods described by Tellez et al.;1 this included outlier exclusion. Differences in
expression between sham-operated and heart failure (HF) groups were tested using the unpaired t test (with
no correction for multiple tests). Secondly, the data were analysed using the ΔCt method implemented using
StatMiner Version 4.1 software (Integromics); this included outlier exclusion using an algorithm within
StatMiner. The data were analysed using StatMiner, because this allowed statistical analysis using the FDR
(see below) as well as the unpaired t test. The data obtained using StatMiner are shown in Figs. 1 and 3-7
and Table 2. However, the data obtained with all three methods were qualitatively similar.
Multiple comparisons in large bodies of data, such as microarray data sets, pose major statistical
challenges.2, 3 Typically, a correction (e.g. the Bonferonni correction) is made when making multiple tests to
avoid false-positive results (type I error).2, 3 Unfortunately, reducing the type I error increases the chance of
false-negative results (type II error).2, 3 In relation to large data sets, it has been argued by Rothman3 that a
policy of not making corrections for multiple tests is preferable. Rothman3 states that “scientists should not
be so reluctant to explore leads that may turn out to be wrong that they penalize themselves by missing
possibly important findings”. This approach is justified in the present study, because each comparison is
independent of the others (the conclusions of the present study do not depend on a group of differences being
significant). In the absence of a correction for multiple comparisons, a single difference is significant at the
5% (P<0.05) level (if the conclusions of the present study depended on a group of differences being
significant, a correction for multiple comparisons would have to be made, and the group of differences
would be significant at the 5% level).
Although multiple test corrections of the classical type are not commonly used in the analysis of
microarray data, a new type of multiple tests correction, the false discovery rate (FDR), has become a
standard for multiple tests correction in microarray data analysis.2 The theory underlying the FDR is beyond
the scope of this discussion. In the present study, we followed the example of Myers et al.4 and used the
Benjamini-Hochberg FDR as well as the unpaired t test (with no correction for multiple tests) as
implemented by StatMiner. Table 2 reports the results of this analysis – for each transcript, the FDRcorrected P value is reported. Different investigators use a FDR-corrected P value of less than 0.05, 0.1 or
even 0.2 to indicate significance. In Table 2, transcripts for which the FDR-corrected P value<0.2 are shown.
An advantage of the FDR is that it gives an estimate of how many of the reported positives (i.e., the passed
tests) are false-positives on average (i.e. the false discovery rate). A FDR-corrected P value of 0.1 and 0.2
corresponds to a false discovery rate (FDR) of 10% and 20%, respectively. For example, Table 2 shows that
in the sinoatrial node (SAN): the FDR-corrected P value<0.1 in 35 cases – of these 10%, i.e. 4, on average
will be false positives; and the FDR-corrected P value<0.2 in 53 cases – of these 20%, i.e. 11, on average
will be false positives. In Figs. 1 and 3-7, transcripts for which the FDR-corrected P value<0.2 are flagged by
an asterisk. The unpaired t test flagged 38 transcripts as showing significant changes (P<0.05). In Table 2, an
asterisk indicates a significant change as determined by the unpaired t test. 36 of the transcripts flagged by
the unpaired t test as showing significant changes were also flagged by the FDR as showing changes (Table
2).
Hierarchical clustering and multidimensional scaling
For hierarchical clustering and multidimensional scaling, J-Express Version 7 was used. Analysis was
carried out on CT calculations. Outliers were not excluded. Two class unpaired statistical analysis of
microarrays (SAM) was carried out; 1000 iterations were carried out for each comparison (SAN and atrial
muscle). A FDR of 0 was used to identify transcripts that were different between HF and sham-operated
animals. This showed that 28 transcripts were differently expressed in the SAN from sham-operated and HF
animals. With one exception (Kv1.5), the transcripts were a subset of those in Table 2. Hierarchical
clustering was carried out on this group of 28 transcripts with Euclidian distances and average linkage.
Multidimensional scaling (MDS) with Euclidian distances was carried out with data from all transcripts; 500
iterations were run.
References
1.
Tellez JO, Dobrzynski H, Yanni J, Billeter R, Boyett MR. Effects of aging on gene expression in the
rat sinoatrial node. Journal of Molecular and Cellular Cardiology. 2006;40:982-982.
2.
Katagiri F. Multiple tests correction, false discovery rate and q value. Physiology News. 2009;77:3538.
3.
Rothman KJ. No adjustments are needed for multiple comparisons. Epidemiology. 1990;1:43-46.
4.
Myers SA, Eriksson N, Burow R, Wang SC, Muscat GE. Beta-adrenergic signaling regulates NR4A
nuclear receptor and metabolic gene expression in multiple tissues. Mol Cell Endocrinol.
2009;309:101-108.
Table S1. Details of assays used on Taqman low density array cards.
Gene
Symbol
Adora1
Adra1a
Adra1b
Adra1d
Adrb1
Adrb2
Chrm2
Atp1a1
Atp1a2
Atp1a3
Atp1b1
Atp2a2
Atp2b1
Cacna1c
Cacna1d
Cacna1g
Cacna2d
1
Cacna2d
2
Cacna2d
3
Cacnb1
Cacnb2
Cacnb3
Cacng4
Cacng7
Casq2
Clcn2
Gja5
Gja1
Hcn1
Hcn2
calcium channel, voltage-dependent, alpha2/delta subunit 1
Accession
Number
NM_017155.2
NM_017191.2
NM_016991.2
NM_024483.1
NM_012701.1
NM_012492.2
NM_031016.1
NM_012504.1
NM_012505.1
NM_012506.1
NM_013113.2
NM_001110139.
2
NM_053311.1
NM_012517.2
NM_017298.1
NM_031601.3
NM_001110847.
1
calcium channel, voltage-dependent, alpha 2/delta subunit 2
NM_175592.2
Rn00457825_m1
calcium channel, voltage-dependent, alpha 2/delta 3 subunit
calcium channel, voltage-dependent, beta 1 subunit
calcium channel, voltage-dependent, beta 2 subunit
calcium channel, voltage-dependent, beta 3 subunit
calcium channel, voltage-dependent, gamma subunit 4
calcium channel, voltage-dependent, gamma subunit 7
calsequestrin 2
chloride channel 2
gap junction membrane channel protein alpha 5 (Cx40)
gap junction membrane channel protein alpha 1 (Cx43)
hyperpolarization-activated, cyclic nucleotide-gated potassium channel 1
hyperpolarization activated cyclic nucleotide-gated potassium channel 2
NM_175595.2
NM_017346.1
NM_053851.1
NM_012828.2
NM_080692.1
NM_080695.1
NM_017131.2
NM_017137.1
NM_019280.1
NM_012567.2
NM_053375.1
NM_053684.1
Rn00598241_m1
Rn00569267_m1
Rn00587789_m1
Rn00432233_m1
Rn00589903_m1
Rn00519216_m1
Rn00567508_m1
Rn00567553_m1
Rn00570632_m1
Rn01433957_m1
Rn00584498_m1
Rn01408575_gH
Gene Name
adenosine A1 receptor
adrenergic receptor, alpha 1a
adrenergic receptor, alpha 1b
adrenergic receptor, alpha 1d
adrenergic receptor, beta 1
adrenergic receptor, beta 2
cholinergic receptor, muscarinic 2
ATPase, Na+/K+ transporting, alpha 1
ATPase, Na+/K+ transporting, alpha 2
ATPase, Na+/K+ transporting, alpha 3
ATPase, Na+/K+ transporting, beta 1
ATPase, Ca++ transporting, cardiac muscle
ATPase, Ca++ transporting, plasma membrane 1
calcium channel, voltage-dependent, L type, alpha 1C subunit
calcium channel, voltage-dependent, L type, alpha 1D subunit
calcium channel, voltage-dependent, T type, alpha 1G subunit
Assay ID
Rn00567668_m1
Rn00567876_m1
Rn01471343_m1
Rn00577931_m1
Rn00824536_s1
Rn00560650_s1
Rn02532311_s1
Rn00560766_m1
Rn00560789_m1
Rn00560813_m1
Rn00565405_m1
Rn00568762_m1
Rn00584038_m1
Rn00709287_m1
Rn00568820_m1
Rn00581051_m1
Rn00563853_m1
Table S1 continued.
Gene
Symbol
Hcn3
Hcn4
Itpr1
Itpr2
Itpr3
Ryr2
Ryr3
Kcna2
Kcna4
Kcna5
Kcnab1
Kcnb1
Kcnd2
Kcnd3
Kcne3
Kcne4
Kcnh2
Kcnj2
Kcnj12
Kcnj4
Kcnj14
Kcnj3
Kcnj5
Kcnj8
Kcnj11
Gene Name
hyperpolarization-activated cyclic nucleotide-gated potassium channel 3
hyperpolarization-activated, cyclic nucleotide-gated 4
inositol 1,4,5-triphosphate receptor 1
inositol 1,4,5-triphosphate receptor 2
inositol 1, 4, 5-triphosphate receptor 3
ryanodine receptor 2 (cardiac)
ryanodine receptor 3
potassium voltage-gated channel, shaker-related subfamily, member 2
(Kv1.2)
potassium voltage-gated channel, shaker-related subfamily, member 4
(Kv1.4)
potassium voltage-gated channel, shaker-related subfamily, member 5
(Kv1.5)
potassium voltage-gated channel, shaker-related subfamily, beta member
1
potassium voltage gated channel, Shab-related subfamily, member 1
(Kv2.1)
potassium voltage gated channel, Shal-related family, member 2 (Kv4.2)
potassium voltage gated channel, Shal-related family, member 3 (Kv4.3)
potassium voltage-gated channel, Isk-related subfamily, member 3
(MiRP2)
minK-related peptide 3 (MiRP3)
potassium voltage-gated channel, subfamily H (eag-related), member 2
(ERG1)
potassium inwardly-rectifying channel, subfamily J, member 2 (Kir2.1)
potassium inwardly-rectifying channel, subfamily J, member 12 (Kir2.2)
potassium inwardly-rectifying channel, subfamily J, member 4 (Kir2.3)
potassium inwardly-rectifying channel, subfamily J, member 14 (Kir2.4)
potassium inwardly-rectifying channel, subfamily J, member 3 (Kir3.1)
potassium inwardly-rectifying channel, subfamily J, member 5 (Kir3.4)
potassium inwardly-rectifying channel, subfamily J, member 8 (Kir6.1)
potassium inwardly rectifying channel, subfamily J, member 11 (Kir6.2)
Accession
Number
NM_053685.1
NM_021658.1
NM_001007235.
1
NM_031046.3
NM_013138.1
AF112257.1
XM_342491.3
Assay ID
Rn00586666_m1
Rn00572232_m1
Rn01425738_m1
Rn00579067_m1
Rn00565664_m1
Rn01470303_m1
Rn01328415_g1
NM_012970.3
Rn00564239_m1
NM_012971.2
Rn02532059_s1
NM_012972.1
Rn00564245_s1
NM_017303.2
Rn00568877_m1
NM_013186.1
NM_031730.2
NM_031739.2
Rn00755102_m1
Rn00581941_m1
Rn00709608_m1
AJ271742.1
NM_212526.1
Rn00573728_m1
Rn01769979_s1
NM_053949.1
NM_017296.1
NM_053981.2
NM_053870.2
NM_170718.1
NM_031610.3
NM_017297.2
NM_017099.3
NM_031358.3
Rn00588515_m1
Rn00568808_s1
Rn02533449_s1
Rn01502359_m1
Rn00821873_m1
Rn00434617_m1
Rn01789221_mH
Rn00567317_m1
Rn01764077_s1
Table S1 continued.
Gene
Symbol
Abcc9
Kcnk2
Kcnk3
Kcnk6
Kcnk1
Kcnn1
Kcnn2
Kcnn3
Kcnq1
Pias3
Kcnip2
Scn1a
Scn3a
Scn4a
Scn5a
Scn1b
Scn2b
Scn3b
Scn4b
Slc16a1
Slc4a1
Slc4a2
Slc4a3
Slc4a4
Slc4a7
Slc8a1
Slc9a1
Trpc1
Trpc2
Gene Name
ATP-binding cassette, sub-family C (CFTR/MRP), member 9 (SUR2)
potassium channel, subfamily K, member 2 (Trek-1)
potassium channel, subfamily K, member 3 (TASK-1)
potassium channel, subfamily K, member 6 (TWIK-2)
potassium channel, subfamily K, member 1 (TWIK-1)
potassium intermediate/small conductance calcium-activated channel,
subfamily N, member 1
potassium intermediate/small conductance calcium-activated channel,
subfamily N, member 2
potassium intermediate/small conductance calcium-activated channel,
subfamily N, member 3
potassium voltage-gated channel, subfamily Q, member 1 (KvLQT1)
protein inhibitor of activated STAT 3 (KChAP)
Kv channel-interacting protein 2 (KChIP2)
sodium channel, voltage-gated, type 1, alpha polypeptide
sodium channel, voltage-gated, type III, alpha polypeptide
sodium channel, voltage-gated, type IV, alpha polypeptide
sodium channel, voltage-gated, type V, alpha polypeptide
sodium channel, voltage-gated, type I, beta polypeptide
sodium channel, voltage-gated, type II, beta polypeptide
sodium channel, voltage-gated, type III, beta
sodium channel, voltage-gated, type IV, beta
solute carrier family 16, member 1 (MCT1)
solute carrier family 4, member 1 (AE1)
solute carrier family 4, member 2 (AE2)
solute carrier family 4, member 3 (AE3)
solute carrier family 4, member 4 (NBC1)
solute carrier family 4, sodium bicarbonate cotransporter, member 7
(NBC3)
solute carrier family 8 (sodium/calcium exchanger), member 1 (NCX1)
solute carrier family 9, member 1 (NHE1)
transient receptor potential cation channel, subfamily C, member 1
transient receptor potential cation channel, subfamily C, member 2
Accession
Number
NM_013040.2
NM_172041.2
NM_033376.1
NM_053806.2
NM_021688.3
Assay ID
Rn00564842_m1
Rn00597042_m1
Rn00583727_m1
Rn00821542_g1
Rn00572452_m1
NM_019313.1
Rn00570904_m1
NM_019314.1
Rn00570910_m1
NM_019315.2
NM_032073.1
NM_031784.2
NM_020095.2
NM_030875.1
NM_013119.1
NM_013178.1
NM_013125.1
NM_017288.1
NM_012877.1
NM_139097.3
NM_001008880.
1
NM_012716.2
NM_012651.2
NM_017048.1
NM_017049.1
NM_053424.1
Rn00570912_m1
Rn00583376_m1
Rn00582371_m1
Rn01411450_g1
Rn00578439_m1
Rn00565438_m1
Rn00565973_m1
Rn00565502_m1
Rn00441210_m1
Rn00563554_m1
Rn00594710_m1
NM_058211.1
NM_019268.2
NM_012652.1
NM_053558.1
AF136401.1
Rn00589539_m1
Rn00570527_m1
Rn00561924_m1
Rn00585625_m1
Rn00575304_m1
Rn01418016_m1
Rn00562332_m1
Rn00561909_m1
Rn00566910_m1
Rn00436642_m1
Rn00584747_m1
Table S1 continued.
Gene
Symbol
Trpc3
Trpc4
Trpc6
Trpc7
P2rx4
Myh6
Myh7
Tbx3
Hprt
Nppa
Nppb
Gene Name
transient receptor potential cation channel, subfamily C, member 3
transient receptor potential cation channel, subfamily C, member 4
transient receptor potential cation channel, subfamily C, member 6
transient receptor potential cation channel, subfamily C, member 7
purinergic receptor P2X, ligand-gated ion channel 4
myosin heavy chain, polypeptide 6
myosin, heavy polypeptide 7, cardiac muscle, beta
T-box 3
hypoxanthine guanine phosphoribosyl transferase
natriuretic peptide precursor type A
natriuretic peptide precursor type B
Accession
Number
NM_021771.1
NM_080396.1
NM_053559.1
XM_225159.4
NM_031594.1
NM_017239.1
NM_017240.1
NM_181638.1
NM_012583.2
NM_012612.2
NM_031545.1
Assay ID
Rn00572928_m1
Rn00584835_m1
Rn00585635_m1
Rn01448763_m1
Rn00580949_m1
Rn00568304_m1
Rn00568328_m1
Rn00710902_m1
Rn01527840_m1
Rn00561661_m1
Rn00580641_m1
Table S2. Additional SYBR green assays used.
Target
28S
Accession
number
AF460236
Cx30.2
AY863055
Cx45
AF536559
Phospholamban
NM_022707
Primer sequence 5′- 3′
Qiagen
Assay No.
-
GTTGTTGCCATGGTAATCCTGCTCAGTACG
TCTGACTTAGAGGCGTTCAGTCATAATCCC
CGAGGACGAGCAGGAGGA
GGAACAGCCAGAAGCGGTAG
AATAAAGAGCAGAGCCAACCAAAA
GCCCACCTCAAACACAGTCC
QT01290058