university of groningen molecular adaptations in human
TRANSCRIPT
University of Groningen
Molecular adaptations in human atrial fibrillationBrundel, Bianca Johanna Josephina Maria
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.
Document VersionPublisher's PDF, also known as Version of record
Publication date:2000
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):Brundel, B. J. J. M. (2000). Molecular adaptations in human atrial fibrillation: mechanisms of proteinremodeling. s.n.
CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
The publication may also be distributed here under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license.More information can be found on the University of Groningen website: https://www.rug.nl/library/open-access/self-archiving-pure/taverne-amendment.
Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.
Download date: 03-04-2022
ISBN 90-367-1248-3
Nugi code 743
© Copyright 2000 Bianca J.J.M. Brundel
All rights are reserved. This publication is protected by copyright. No part of it
may be reproduced, stored in a retrieval system, or transmitted, in any form or by
any means — electronic, mechanical, photocopy, recording, or otherwise — with-
out the prior written permission of the author.
Lay-out: Peter van der Sijde, Groningen NL
Druk: Ponsen en Looijen bv, Wageningen NL
The studies described in this thesis were supported by grants 94.014 and 96.051
from the Netherlands Heart Foundation.
Publication of this thesis was financial supported by :
The Netherlands Heart Foundation
Groningen University Institute of Drug Exploration (GUIDE)
Faculteit Medische Wetenschappen RUG
Bristol-Myers Squibb BV
AstraZeneca BV
ASTA Medica BV
Novartis Pharma BV
Dr. Saal van Zwanenberg stichting
Cover:
Top of the picture shows the view from the ‘mountain’ of Kardinge
(Groningen). Bottom of the picture shows electron microscopic detail of a
human atrial myocyte (magnification x 4500).
Backside of the cover shows a picture of Captain Hook (Myrthe), Wendy (Jona)
and Peter Pan (Joachim).
Rijksuniversiteit Groningen
Molecular Adaptations in Human Atrial
Fibrillation:
Mechanisms of Protein Remodeling
Proefschrift
ter verkrijging van het doctoraat in
de Medische Wetenschappen
aan de Rijksuniversiteit Groningen
op gezag van de
Rector Magnificus, dr. D.F.J. Bosscher,
in het openbaar te verdedigen op
woensdag 29 november 2000
om 16.00 uur
door
Bianca Johanna Josephina Maria Brundel
geboren op 7 februari 1971
te Lichtenvoorde
Promotores:Prof. Dr. H.J.G.M. Crijns
Prof. Dr. W.H. van Gilst
Referenten:Dr. I.C. van Gelder
Dr. R.H. Henning
Table of contents
Chapter 1 Introduction 9
Part I Gene expression of proteins influencing calcium
homeostasis
Chapter 2 Gene expression of proteins influencing calcium homeostasis
in patients with persistent and paroxysmal atrial fibrillation
Cardiovascular Research 1999 (42) 443-454 21
Chapter 3 Alterations in gene expression of proteins involved in
calcium handling in patients with atrial fibrillation
Journal of Cardiovascular Electrophysiology
1999 (10) 552-560 39
Part II Gene expression of ion-channels
Chapter 4 Alterations in potassium channel gene expression in atria of
patients with persistent and paroxysmal atrial fibrillation
Differential regulation of protein and mRNA levels for
K+-channels
Journal of the American College of Cardiology, provisionally
accepted 55
Chapter 5 Ion channel remodeling is related to intra operative atrial
refractory periods in patients with paroxysmal and persistent
atrial fibrillation
Circulation, in press 69
Part III Gene expression of neurohormones
Chapter 6 Gene expression of the natriuretic peptide system in atrial tissue
of patients with paroxysmal and persistent atrial fibrillation
Journal Cardiovascular Electrophysiology 1999, (10) 827-835 85
Chapter 7 Endothelin-1 mRNA is upregulated in patients with persistent
atrial fibrillation with underlying valve disease
Submitted 99
Part IV Calpain activation, a new adaptive mechanism in AF
Chapter 8 Activation of proteolysis by calpain during paroxysmal and
persistent atrial fibrillation
Submitted 113
Chapter 9 Calpain activity is related to ion-channel, structural and electrical
remodeling in human paroxysmal and persistent atrial fibrillation
Submitted 121
Chapter 10 General discussion 137
Summary 153
Samenvatting 155
Dankwoord 157
9
Introduction
Introduction
Atrial Fibrillation, clinical aspects
Atrial fibrillation (AF) is currently the most common sustained clinical arrhythmia
and is responsible for a substantial proportion of hospital costs incurred in the treatment of
cardiac rhythm disorders.1 AF becomes increasingly common with age, having an incidence
averaging <0.5% in patients <40 years of age and reaching >5% in patients >65.2 Thus AF
is likely to become increasingly important with the ageing of the population. The arrhythmia
is defined by a very rapid atrial rate (generally >400/min in humans) along with irregular
atrial activation and lack of repetitive pattern of co-ordinated atrial activity. AF is associated
with a variety of complications, including thromboemboli resulting from coagulation in
the relatively static atrial blood pool, a loss of the fine adjustment of ventricular rate to the
body’s precise metabolic needs, potential impairment of cardiac function and subjective
symptoms like palpitations, dizziness, breathlessness and chest pain.
AF can occur in paroxysms of a duration shorter than 24 hours (but longer lasting
paroxysms are not unusual) with intermittent sinus rhythm. Paroxysmal AF either converts
spontaneously or is terminated with an intravenously administered antiarrhythmic drug.3,4
In contrast, during persistent AF, the arrhythmia is continuously present until the moment
of investigation, i.e. at least two consecutive electrocardiograms of AF more than 24 hours
apart and without intercurrent sinus rhythm. Persistent AF does not convert spontaneously.3,4
AF has the tendency to become more persistent over time. This is illustrated by the
fact that about 30% of patients with paroxysmal AF eventually will develop persistent
AF.5 Also pharmacological and electrical cardioversion and maintenance of sinus rhythm
thereafter become more difficult the longer the arrhythmia exists.6 The cumulative
percentage of patients who maintained sinus rhythm after serial cardioversion treatment
was not more than 42% after 1 year and 27% after 4 years.6 This relates to progression of
the underlying disease and possibly also to electrical remodeling of the atria.7
Mechanism: multiple-wavelet reentry
In 1959 Moe & Abildskov8 showed that AF could be produced by experimental
paradigms of both multiple circuit reentry and rapid activity and they suggested that either
type of mechanism might cause clinical AF. Moe put forward the ‘multiple wavelet
hypothesis’ of AF in 1962.9 This concept described the propagation of reentry waves as
involving multiple independent wavelets circulating around functionally refractory tissue.
The maintenance of AF then depends on the probability that electrical activity can be
sustained by a sufficient number of active wavelets at any time. Experimental support for
Moe’s ideas was obtained subsequently with the use of computerized mapping of AF
maintained in the presence of acetylcholine in dog hearts.10 It was demonstrated that during
10
Chapter 1
Figure 1
Schematic representation of inward and outward ionic currents involved in the ventricular action potential, resting
membrane potential and cytoplasmic Ca2+ transients.
The numbers 0 through 4 indicate the different phases of the action potential: 0, upstroke; 1, fast early repolarization
phase; 2, plateau phase; 3, repolarization phase; 4, resting membrane potential. Adapted from The Sicilian Gambit.47
INa
, sodium current encoded by the gene SNC5A; ICaT
, calcium current encoded by T-type Calcium channel; ICaL
,
calcium current L-type Calcium channel; ICl(Ca)
, calcium dependent transient outward current encoded by the
gene Kv4.2; INaCa
, sodium calcium exchanger; ITo
, transient outward current encoded by Kv4.3; IK1
, inward rectifier
K+ current encoded by Kir2.1; IKs
, slow delayed rectifier K+ current encoded by minK and KvLQT; IKr
, rapid
delayed rectifier K+ current encoded by HERG, IKACh
, acetylcholine dependent potassium current encoded by
Kir3.1 and Kir3.4; INa-K
, sodium potassium pump.
11
Introduction
AF, multiple independent wavelets activate the atria in a random reentrant way. Individual
wavelets could brake-up, fuse or collide with each other and wavelets would extinct when
they reached the border of the atria or met refractory tissue. From time to time a varying
number of wavelets was present in the atria and the duration of each individual wavelet
lasted only several hundreds of a millisecond. Further it was shown that the number of
wavelets that fit into the atria determined the perpetuation of AF. Below a critical number
of wavelets (between 3 and 6), there was a considerable chance for the wavelets to die out
all at the same time. When more than 6 independent wavelets were present, the arrhythmia
would not convert spontaneously anymore. The number of wavelets that fit into the atria
depends on the atrial refractory period, conduction velocity and atrial mass.8,9 During the
last decade, mapping studies in humans with AF have further confirmed the multiple wavelet
theory.11
Electrophysiology of the atria
The processes that signal the heart to contract (excitation-contraction coupling) begin
when an action potential depolarizes the plasma membrane surrounding the myocardial
cell. This electrical signal is generated by the passage of ions through ion channels in the
plasma membrane that changes the electrical potential of the interior of the cardiac cell
relative to the extracellular space. These ion fluxes include two major inward currents that
depolarize the heart. The phase 0 depolarization is initiated by a rapid inflow of sodium
ions through the voltage-gated Na-channels and later by calcium ions via the L-type calcium
channel (ICaL
) (Figure 1).12,13 A large transient outward current has been held responsible
for the phase 1 repolarization. This current is composed of a 4-aminopyridine-sensitive
component (ITo1
) and a Ca2+ activated and verapamil blocked Cl- current (ITo2
or ICl(Ca)
).14-16
During the plateau phase (phase 2), there is a delicate balance of inward and outward
currents. Inward currents can be carried through the Na+ channel and the Ca2+ channels.
During reverse-mode, the Na+/Ca2+ exchanger will transport three Na+ ions outside and
exchanged for one Ca2+ resulting in a net movement of charge across the plasma membrane
from inside to outside. The influx of calcium through the L-type calcium channels is
responsible for the plateau phase during repolarization and initiates calcium release from
the sarcoplasmic reticulum that binds to the contractile filaments of the myocardial cell,
causing contraction (see calcium homeostasis in normal cardiac cells).17 Outward currents
during the plateau are carried by a number of K+ channels (IKs
, IKr
, IKACh
) and the Na+-K+
pump. For the initiation of phase 3 the contributions of rapidly and slowly activating
delayed rectifier K+ currents are of key importance. Here the outward potassium current
IK1
is activated and the action potential returns to its transmembrane resting potential,
which remains at that level during phase 4 of the action potential until the cell becomes re-
activated.
12
Chapter 1
The duration of the repolarization phase is different between atrial and ventricular
myocardial cells. This phase is shorter in the atrium compared to ventricle.18 An other
difference is the distribution and magnitude of the ionic currents responsible for the rest-
ing membrane potential. The inward rectifier current, which is responsible for mainte-
nance of the resting membrane potential, IKs
is smaller in atrial cells compared with ven-
tricular cells.19,20 Secondly, the acetylcholine-dependent potassium current IKACh
is very
important in maintaining and hyperpolarizing the resting membrane potential in atrial
cells, while less active in ventricular cells.21 It is this current which is responsible for
hyperpolarization and shortening of the atrial action potential during parasympathic stimu-
lation. Vagal stimulation is indeed one of the oldest models to induce sustained atrial
fibrillation by applying vagal stimulation. The resulted increased parasympathetic tone
will shorten the atrial refractory period through opening of these acetylcholine-dependent
potassium channels. Furthermore, inhomogeneous distribution of vagal nerve endings will
increase the spatial dispersion in refractoriness.22 In this way, atrial fibrillation will persist
after induction with premature stimuli or atrial burst pacing as long as the parasympathetic
system is stimulated, either by vagal nerve stimulation or by acetylcholine application.
Figure 2
Schematic representation of the calcium triggered calcium release in the myocardial cell. Small amount of
calcium enters the cell via the Na+/Ca2+ exchanger, but mainly through voltage gated L-type calcium channels,
which open in response to the action potential. This causes a much larger amount of calcium to be released by
the ryanodine receptors (RyR) in the sarcoplasmic reticulum (SR). Calcium will bind to the troponin C
resulting in contraction. The calcium pump of the SR (SR Ca2+ ATPase) has a high affinity for calcium and
reduces cytosolic Ca2+ concentrations to levels low enough to dissociate this cation from its binding sites on
troponin C. The SR Ca2+ ATPase is inhibited by phospholamban (PLB). When PLB is phosphorylated this
inhibition is reversed.
13
Introduction
Calcium homeostasis in normal cardiac cells
When an action potential depolarizes the plasma membrane surrounding the
myocardial cell, the processes that signal the heart to contract begins. Calcium transients
underlying this excitation-contraction coupling in cardiac cells result mainly from calcium
release from the sarcoplasmic reticulum triggered by calcium entry during the action
potential. This process is called calcium-induced calcium release (Figure 2). Under normal
circumstances, calcium entry into cardiac myocytes is carried primarily via IcaL, whereas
additional fractions can enter via on reverse-mode Na+/Ca2+ exchange (one calcium ion is
transported inside the cell against three sodium ions outside) and ICaT
. All three pathways
are capable of triggering sarcoplasmic reticulum (SR) calcium release and contraction,
but the relative contribution and efficiency is largest for ICaL
.23,24 Upon depolarization,
influx of calcium through discrete clusters of L-type calcium channels in the plasma
membrane triggers the opening of ryanodine receptors in the sarcoplasmic reticulum
membrane, resulting in the major release of calcium into the cytoplasm which in turn
triggers myofibril contraction by binding of troponin C. Following contraction, the SR
calcium ATPase enzyme in the network of sarcoplasmic reticulum surrounding the
myofibrils, rapidly pumps the calcium back into the SR lumen, causing the myofibrils to
relax. The SR calcium ATPase is regulated by phospholamban, a small protein and when
phosphorylated it will activate SR calcium ATPase and the calcium content of the SR will
increase.25
Modern concepts of excitation-contraction coupling rely on a ‘local-control theory’ a
close association between L-type calcium channels and ryanodine receptors and
subsequently received experimental support documenting local functional interaction
between these channels.26 Changes in the close association between the L-type calcium
channels and ryanodine receptors result in a reduction of calcium release from the
sarcoplasmic reticulum and reduced contractility of the cardiac myocyte.26
The membrane machinery allows each myocyte to function as an autonomous
contractile unit. To produce a heart beat, the contractile capabilities of myocytes that
make up the heart have to work in a highly synchronous fashion. This requires both an
orderly spread of the wave of electrical activation and effective transmission of contractile
force from one cell to the next, throughout the heart.
Electrical remodeling of the atria
Over the past several years, AF-induced remodeling and its underlying mechanisms
have been studied in substantial detail. Wijffels and coworkers published the first study in
1995 which part of the underlying electrophysiological changes explaining the progressive
nature of AF was demonstrated.7 In healthy goats they showed that repetitive induction of
AF increased the duration of successive episodes of AF, until AF finally did not convert
14
Chapter 1
spontaneously any more. They discovered that the increased tendency of the atria to fibrillate
was paralleled by a progressive shortening of the atrial effective refractory period (AERP)
and loss of the physiological rate adaptation of the refractory period which they termed
atrial electrical remodeling.7 After cardioversion of atrial fibrillation that had been present
for two to four weeks, this so-called atrial electrical remodeling appeared to be completely
reversible within one week after restoration of sinus rhythm.
In 1995 Morillo et al.27 also showed that rapid atrial pacing (400 bpm) strongly
promotes the ability to maintain AF in dogs, with changes quite similar to those observed
by Wijffels et al.7 Later observations by Wijffels et al. suggested that acute volume loading,
opening of the ATP-dependent potassium channels, neurohumoral activation or an increase
in ANP were not responsible for the altered electrophysiological characteristics in this
experimental model. This supports the idea that AF-induced remodeling is primarily due
to the rapid atrial activation rates caused by AF.28 Consequentially, many investigators
have used rapid atrial pacing in experimental models to study the electrophysiological
changes caused by sustained atrial tachycardia, hoping to gain insights into the atrial
electrophysiological changes caused by AF in man.
Experimental and clinical studies have shown that sustained AF decreases the atrial
effective refractory period (AERP).7,29-32 AERP changes occur over a period of days to
weeks7,27,33,34, but AF can decrease AERP over a time interval as short as several minutes.32
Although the AERP reduction caused by AF favours arrhythmia maintenance, it cannot be
the only factor involved because AF-induced AERP alterations become maximal well before
AF-promoting effects stabilize.7,34 One of the AF-promoting effects is tachycardia induced
atrial conduction slowing.27,33,34 It has a slower time course than AERP changes, probably
due to structural changes and could account for at least a part of the continued development
of AF promotion after AERP changes have stabilized.
Tieleman and coworkers found that the AERP shortened with loss of the normal rate
adaptation in response to 24 hours of rapid atrial pacing in goats.35 Here the tendency of
the atria to fibrillate increased. With resumption of sinus rhythm after cessation of pacing,
the refractory period normalized over a period of slightly more than 24 hours. This electrical
remodeling could be modulated using several pharmacological agents. First, the calcium
channel blocker verapamil reduced atrial electrical remodeling; suggesting that tachycardia-
induced calcium overload might trigger the shortening of the refractory period.35 By contrast,
digoxin delayed the recovery from electrical remodeling of the atria.36 This could be due
to the effect of digoxin on calcium handling, preventing effective wash-out of calcium
after cessation of pacing.
Contractile dysfunction of the atrium
Besides the electrical remodeling observed in experimental studies, clinical studies
15
Introduction
demonstrated atrial contractile dysfunction after AF.37,38 Leistad and co-workers investigated
the contractile dysfunction in an experimental model for AF.39 They demonstrated that the
atrial contractile dysfunction after acute atrial fibrillation is reduced by the calcium channel
antagonist verapamil, which suggests that transsarcolemmal calcium influx contributed to
this dysfunction. The calcium agonist BAY K8644 increased postfibrillation atrial contractile
dysfunction. A remarkable finding was that atrial contractility increased in the first seconds
after atrial fibrillation before a longer period of reduced atrial contractility ensued. On the
basis of these observations they hypothesised that cytosolic calcium overload due to rapid
depolarization during the preceding fibrillation might be responsible for the atrial contractile
dysfunction. This hypothesis was also suggested by Shapiro et al. in 1988.40It should be
noted that in the pig model of Leistad and coworkers39 atrial contractile dysfunction was
observed after only five minutes of AF, indicating that contractile dysfunction, like electrical
remodeling in the goat7, starts early.39
Structural remodeling
Until now, research focused on electrical and contractile remodeling. AF is also
associated with structural changes.27,31,41 Ausma and coworkers described and quantified
the structural remodeling in atrial myocardium due to sustained atrial fibrillation in the
goat.41 They maintained atrial fibrillation in normal goats for a prolonged period of time.
After 9 to 23 weeks of sustained atrial fibrillation, several areas of the right and left atria
were examined by light and electron microscopy. They found that a substantial proportion
of the atrial myocytes (up to 92%) revealed marked changes in their cellular substructures,
such as loss of myofibrils, accumulation of glycogen, changes in mitochondrial shape and
size, fragmentation of sarcoplasmic reticulum and dispersion of nuclear chromatin. They
pointed out that these changes are the same as observed in chronically hibernating
myocardium. This is a condition that occurs in patients as a result of low flow ischemia
caused by stenosis of one or more coronary arteries.41 The clinical condition is defined as
the ability of the myocardium to adapt to chronic ischemia by down-regulating its contractile
function, thereby maintaining cell viability for a prolonged period of time. Furthermore, it
was found that these hibernating myocytes resembled a form of dedifferentiation as a
result of chronic atrial fibrillation.42 Some of the atrial cardiomyocytes acquired a
dedifferentiated phenotype, as deduced from the re-expression of α-smooth muscle actin,
the disappearance of cardiotin and the staining patterns of titin, which resembled those of
embryonic cardiomyocytes.
Considerations for the thesis
Thus, experimentally it has been shown that contractile dysfunction starts accutely
after AF onset and was reduced by verapamil but increased by BAY K8644.39 These results
16
Chapter 1
strongly indicate that changes in the calcium homeostasis triggered by tachycardia induced
intracellular calcium overload 43-45, play a pivotal role in the induction of atrial contractile
dysfunction.
Shortening of the atrial effective refractory period was an other important factor contributing
to the maintenance of AF 7,46. This shortening could also be mediated by verapamil and
gave a reduction of the tachycardia induced electrical remodeling of the atria.35 This finding
also suggests that electrical remodeling could be due to changes in the calcium homeostasis.
Aim of the thesis
The main goal was to study the molecular remodeling in human atrial fibrillation. We
focussed on gene expression of proteins which influence the calcium homeostasis and
action potential duration in human AF. The impact of modulating systems like the natriuretic
peptide system and the endothelin system were also studied.
For the purpose of the thesis, right and left atrial appendages were collected during
four years from different groups of patients undergoing cardiac surgery. The patients clinical
characteristics (underlying heart disease, type of AF, electrocardiograms, medication and
exercise tolerance) were assessed. The result was a unique human atrial appendage collection
of around 150 individual patients. Because of this amount of different atrial appendages it
was possible to match AF patients with control patients in sinus rhythm for age, sex,
underlying heart disease, left ventricular function and as far as possible for medication
use. In this way, dissecting the effect of AF was optimized.
During the progression of the study a discrepancy between alterations in mRNA and
protein expression of several ion-channels in patients with paroxysmal AF was observed.
No alterations in mRNA expression of ion-channels compared to important reductions on
the protein level were found in human paroxysmal AF. This new finding prompted us to
explore an adaptive mechanism unknown to occur in AF. We hypothesized reduction in
protein channels due to calcium activated neutral protease calpain. To test this the calpain
activity, protein expression and localization were determined in tissue of patients with
atrial fibrillation. Finally the role of calpain activity in ion-channel protein, structural and
electrophysiological remodeling was studied.
17
Introduction
References
1. Waktare JEP, Camm AJ. Acute treatment of atrial fibrillation: why and when to maintain sinus rhythm. J
Am Coll Cardiol 1998; 81:3C-15.
2. The National Heart Lung and Blood Institute Working Group on Atrial Fibrillation. Atrial fibrillation:
current understandings and research imperatives. J Am Coll Cardiol 1993; 22:1830-1834.
3. Gallagher MM, Camm AJ. Classification of atrial fibrillation. Pacing Clin Electrophysiol 1997; 20:1603-
1605.
4. Levy S, Breithardt G, Campbell RW, et al. Atrial fibrillation: current knowledge and recommendations for
management. The Working Group on Arrhythmias of the European Society of Cardiology. Eur Heart J
1998; 19:1294-1320.
5. Godtfredsen J. Etiology, course and prognosis. A follow-up study of 1212 cases. Copenhagen: University
of Copenhagen. Thesis 1975
6. Van Gelder IC, Crijns HJGM, Tieleman RG, et al. Value and limitation of electrical cardioversion in
patients with chronic atrial fibrillation - importance of arrhythmia risk factors and oral anticoagulation.
Arch Intern Med 1996; 156:2585-2592.
7. Wijffels MC, Kirchhof CJ, Dorland R, et al. Atrial fibrillation begets atrial fibrillation. A study in awake
chronically instrumented goats. Circulation 1995; 92:1954-1968.
8. Moe GK, Abildskov JA. Experimental and laboratory reports. Atrial fibrillation as a self-sustained arrhythmia
independent of focal discharge. Am Heart J 1959; 58:59-70.
9. Moe GK. On the multiple wavelet hypothesis of atrial fibrillation. Arch Int Pharmacodyn Ther 1962;
140:183-188.
10. Allessie MA, Lammers WJEP, Bonke FIM, et al. Experimental evaluation of Moe’s multiple wavelet
hypothesis of atrial fibrillation, in Zipes DP, Jalife J (eds). Cardiac Electrophysiology and Arrhythmias
1985; New York, Grune & Stratton:265-275.
11. Konings KT, Kirchhof CJ, Smeets J, et al. High-density mapping of electrically induced atrial fibrillation
in humans. Circulation 1994; 89:1665-1680.
12. Gettes LS, Reuter H. Slow recovery from inactivation of inward currents in mammalian myocardial fibres.
J Physiol (Lond) 1974; 240:703-724.
13. Weidman S. The effect of the cardiac membrane potential on the rapid availability of the sodium-carrying
system. J Physiol 1955; 127:213-224.
14. Coraboeuf E, Carmeliet E. Existence of two transient outward currents in sheep cardiac Purkinje fibers.
Pflugers Arch 1982; 1982:352-359.
15. Tseng GN, Hoffman BF. Two components of transient outward current in canine ventricular myocytes.
Circ Res 1989; 68:424-437.
16. Zygmunt AC, Gibbons WR. Calcium-activated chloride current in rabbit ventricular myocytes. Circ Res
1991; 68:424-437.
17. Fabiato A. Calcium-induced release of calcium from the cardiac sarcoplasmic reticulum. Am J Physiol
1983; 245:C1-14.
18. Whalley DW, Wendt DJ, Grant A. Basic concepts in cellular cardiac electrophysiology: Part I: Ion channels,
membrane currents, and the action potential. Pacing Clin Electrophysiol 1995; 18:1556-1574.
19. Giles WR, Imaizumi Y. Comparison of potassium currents in rabbit atrial and ventricular cells. J Physiol
(Lond) 1988; 405:123-145.
20. Heidbuchel H, Vereecke J, Carmeliet. Three different potassium channels in human atrium. Contribution
to the basal potassium conductance. Circ Res 1990; 66:1277-1286.
21. Yang ZK, Boyett MR, Janvier NC, et al. Regional differences in the negative inotropic effect of acetylcholine
within the canine ventricle. J Physiol 1996; 492:789-806.
22. Liu L, Nattel S. Differing sympathetic and vagal effects on atrial fibrillation in dogs: role of refractoriness
heterogeneity. Am J Physiol 1997; 273:H805-H816
23. Sipido KR, Carmeliet E, Van De Werf F. T-type Ca2+ current as trigger for Ca2+ release from the sarcoplasmic
reticulum in guinea-pig ventricular myocytes. J Physiol 1998; 508:439-451.
24. Sipido KR, Maes M, Van de Werf F. Low efficiency of Ca2+ entry through the Na+-Ca2+ exchanger as
trigger for Ca2+ release from the sarcoplasmic reticulum: a comparison between L-type Ca2+ current and
reverse mode Na+-Ca2+ exchange. Circ Res 1997; 80:1034-1044.
25. Sasaki T, Inui M, Kimura Y, et al. Molecular mechanism of regulation of Ca2+ pump ATPase by
phospholamban in cardiac sarcoplasmic reticulum. Effects of synthetic phospholamban peptides on Ca2+
pump ATPase. J Biol Chem 1992; 267:1674-1679.
18
Chapter 1
26. Gómez AM, Valdivia HH, Cheng H, et al. Defective excitation-contraction coupling in experimental
cardiac hypertrophy and heart failure. Science 1997; 276:800-805.
27. Morillo CA, Klein GJ, Jones D, et al. Chronic rapid atrial pacing. Structural, functional, and
electrophysiological characteristics of a new model of sustained atrial fibrillation. Circulation 1995; 91:1588-
1595.
28. Wijffels MC, Kirchhof CJ, Dorland R, et al. Electrical remodeling due to atrial firbrillation in chronically
instrumented conscious goats: role of neurohumoral changes, ischemia, atrial stretch, and high rate of
electrical activation. Circulation 1997; 96:3710-3720.
29. Franz MR, Karasik PL, Li C, et al. Electrical remodeling of the human atrium: similar effects in patients
with chronic atrial fibrillation and atrial flutter. J Am Coll Cardiol 1997; 30:1785-1792.
30. Yu WC, Chen SA, Lee SH, et al. Tachycardia-induced change of atrial refractory period in humans. Rate
dependency and effects of antiarrhythmic drugs. Circulation 1998; 97:2331-2337.
31. Goette A, Honeycutt C, Langberg JJ. Electrical remodeling in atrial fibrillation. Time course and mecha-
nisms. Circulation 1996; 94:2968-2974.
32. Daoud EG, Knight BP, Weiss R, et al. Effect of verapamil and procainamide on atrial fibrillation-induced
electrical remodeling in humans. Circulation 1997; 96:1542-1550.
33. Elvan A, Wylie K, Zipes DP. Pacing-induced chronic atrial fibrillation impaires sinus node function in
dogs: electrophysiological remodeling. Circulation 1996; 94:2953-2960.
34. Gaspo R, Bosch RF, Talajic M, et al. Functional mechanisms underlying tachycardia-induced sustained
atrial fibrillation in a chronic dog model. Circulation 1997; 96:4027-4035.
35. Tieleman RG, De Langen CDJ, Van Gelder IC, et al. Verapamil reduces tachycardia-induced electrical
remodeling of the atria. Circulation 1997; 95:1945-1953.
36. Tieleman RG, Blaauw Y, Van Gelder IC, et al. Digoxin delays the recovery from electrical remodeling of
the atria in the goat. Circulation 1999; 100:1836-1842.
37. Manning WJ, Silverman DI, Katz SE, et al. Impaired left atrial mechanical function after cardioversion:
relation to the duration of atrial fibrillation. J Am Coll Cardiol 1994; 23:1535-1540.
38. Daoud EG, Marcovitz P, Knight B, et al. Short-term effect of atrial fibrillation on atrial contractile func-
tion in humans. Circulation 1999; 99:3024-3027.
39. Leistad E, Aksnes G, Verburg E, et al. Atrial contractile dysfunction after short-term atrial fibrillation is
reduced by verapamil but increased by BAY K8644. Circulation 1996; 93:1747-1754.
40. Shapiro EP, Effron MB, Lima, et al. Transient atrial dysfunction after conversion of chronic atrial fibrilla-
tion to sinus rhythm. Am J Cardiol 1988; 62:1202-1207.
41. Ausma J, Wijffels M, Thone F, et al. Structural changes of atrial myocardium due to sustained atrial
fibrillation in the goat. Circulation 1997; 96:3157-3163.
42. Ausma J, Wijffels M, Van Eys G., et al. Dedifferentiation of atrial cardiomyocytes as a result of chronic
atrial fibrillation. Am J Pathol 1997; 151:985-997.
43. Lee HC, Clusin WT. Cytosolic calcium staircase in cultured myocardial cells. Circ Res 1987; 61:934-
939.
44. De Pauw M, Borgers M, Heyndrickx GR. Ultrastuctural calcium distribution in cardiac myocytes after
48h of rapid pacing in dogs. Circulation 1996; 94:I-604
45. Schouten VJA, Morad M. Regulation of Ca2+ current in frog ventricular myocytes by the holding poten-
tial, cAMP and frequency. Pflugers Arch 1989; 415:1-11.
46. Yue L, Feng J, Gaspo R, et al. Ionic remodeling underlying action potential changes in a canine model of
atrial fibrillation. Circ Res 1997; 81:512-525.
47. The Sicilian Gambit. a new approach to the classification of antiarrhythmic drugs based on their actions on
arrhythmogenic mechanisms. Task force of the Working Group on Arrhythmias of the European Society of
Cardiology. Circulation 1991; 84:1831-1851.
21
Gene expression of proteins influencing calcium homeostasis in patients
Chapter 2
Gene Expression of Proteins Influencing Calcium
Homeostasis in Patients with Persistent and Paroxysmal
Atrial Fibrillation
Bianca J. J. M. Brundela,b, Isabelle C. Van Geldera, Robert H. Henningb,
Anton E. Tuinenburga, Leo E. Deelmanb, Robert G. Tielemana,
Jan G. Grandjeanc, Wiek H. van Gilstb, Harry J. G. M. Crijnsa
From the departments of Cardiology (a), Clinical Pharmacology (b),
and Thoracic Surgery (c), Thoraxcenter, University Hospital Groningen,
Groningen, The Netherlands.
Cardiovascular Research 42 (1999) 443-454
Abstract
Objective: Persistent atrial fibrillation (AF) results in an impairment of atrial function.
In order to elucidate the mechanism behind this phenomenon, we investigated the gene
expression of proteins influencing the calcium handling. Methods: Right atrial appendages
were obtained from 8 patients with paroxysmal AF, 10 with persistent AF (> 8 months)
and 18 matched controls in sinus rhythm. All controls underwent coronary artery bypass
grafting whereas most AF patients underwent Cox’s MAZE surgery (n=12). All patients
had a normal left ventricular function. Total RNA was isolated and reversely transcribed
into cDNA. In a semi-quantitative polymerase chain reaction the cDNA of interest and of
glyceraldehyde-3-phosphate dehydrogenase were coamplified and separated by ethidium
bromide stained gel-electrophoresis. Slot blot analysis was performed to study protein
expression. Results: L-type calcium channel α1 and sarcoplasmic reticulum Ca2+-ATPase
mRNA (-57%, p=0.01 and -28%, p=0.04, respectively) and protein contents (-43%, p=0.02
and –28%, p=0.04, respectively) were reduced in patients with persistent AF compared to
the controls. mRNA contents of phospholamban, ryanodine receptor type 2 and sodium/
calcium exchanger were comparable. No changes were observed in patients with paroxysmal
AF. Conclusions: Alterations in gene expression of proteins involved in the calcium
homeostasis occur only in patients with long-term persistent AF. In the absence of underlying
heart disease, the changes are rather secundary than primary to AF.
22
Chapter 2
Introduction
Atrial fibrillation (AF) is the most common cardiac arrhythmia affecting millions of
people worldwide and its incidence increases with age [1]. Clinical observations showed
that immediately after restoration of sinus rhythm, atrial contractile function is severely
impaired or even absent [2, 3]. The contractile dysfuntion is reversible after restoration of
sinus rhythm, its time course being related to the previous duration of AF [2]. Restoration
does not seem to be complete in all patients presumably related to the extent of damage
occuring during AF.
In an experimental pig model atrial contractile dysfunction was observed after cessation
of pacing induced AF of a duration of only 1 to 30 minutes, indicating that contractile
remodeling, like electrical remodeling in the goat [4], is an early process [5]. There are
strong indications that abnormalities in the calcium handling, in response to tachycardia
induced intracellular calcium overload [6_8], play a pivotal role in the induction of atrial
contractile dysfunction. [5, 9_13]. The proteins and ion channels involved in the adaptation
processes during AF have not been clarified yet. Most likely, identification of the signaling
pathways and their target genes may lead to new therapeutic options for the treatment of
AF. Therefore, we investigated alterations in mRNA and protein expression of proteins
involved in the calcium handling of right atrial appendages (RAA) of patients with
paroxysmal and persistent AF undergoing cardiac surgery. To overcome the problem
whether changes were caused by AF itself, or by the concomitant underlying heart disease,
we selected AF patients with a normal left ventricular function and matched them for age,
sex and left ventricular function with patients in sinus rhythm who underwent coronary
artery bypass surgery.
Methods
Patients
The day before surgery, one investigator (AET) assessed the clinical characteristics
of the patients. The patients’ history and previous electrocardiograms were used to establish
type and duration of AF. In addition, the patients were asked for medication use and
exercise tolerance (according New York Heart Association classification). Echocardio-
graphy data were obtained within 3 months before surgery. RAAs were obtained from 8
patients with paroxysmal AF and from 10 with persistent AF without valvular heart disease
and a normal left ventricular function. The AF patients were matched for age, sex, left
ventricular function, and as far as possible for medication with 18 clinically stable patients
in sinus rhythm undergoing coronary artery bypass surgery. The Institutional Review Board
approved the study, and all patients gave written informed consent. Immediately after
excision, the RAAs were snap-frozen in liquid nitrogen and stored at -85 °C.
23
Gene expression of proteins influencing calcium homeostasis in patients
RNA isolation and cDNA synthesis
Total RNA was isolated from RAAs using the method of acid guanidinium thiocyanate/
-phenol/chloroform extraction followed by a RNeasy kit for RNA minipreps from tissues
(Qiagen). The amount of RNA was evaluated by absorption at 260 nm, using a GeneQuant
II (Pharmacia Biotechnology, The Netherlands). The ratio of absorption (260-280 nm) of
all preparations was between 1.8 and 2.0. First strand cDNA was synthesized by incubation
of 1 µg of total RNA, reverse transcription 10x buffer and 200 ng of random hexamers
with 200 units of Moloney Murine Leukemia Virus Reverse Transcriptase, 1mM of each
dNTP and 1 unit of RNase inhibitor (Promega, The Netherlands) in 20 µl. The synthesis
reaction lasted 10 minutes at 20 °C, 20 minutes at 42 °C, 5 minutes at 99 °C and 5 minutes
at 4 °C, respectively. All the products were checked on contaminating DNA (data not
shown).
Semi quantitative PCR analyses
Since a linear relationship between the amount of input template and amplification
product exists within the exponential range of amplification, a semi-quantitative polymerase
chain reaction (PCR) was developed [14]. The cDNA of interest and the cDNA of the
ubiquitously expressed housekeeping gene glyceraldehyde-3-phosphate dehydrogenase
(GAPDH) were coamplified in a single PCR reaction. Primers were designed for
Sarcoplasmic Reticulum Calcium ATPase (SR Ca2+-ATPase), Phospholamban (PLB), L-
type Calcium Channel α1 subunit (L-type Ca2+), Sodium/ Calcium exchanger (Na+/Ca 2+
exchanger), Ryanodine receptor type 2 (RyR2) and the housekeeping enzyme GAPDH
(Table 1). Eurogentec (Belgium) synthesized the oligonucleotides. For the semi-quantitative
PCR co-amplification of 1 µl of cDNA mixture, 0.5 unit of Taq polymerase (Eurogentec,
Belgium) was added to 17.5 nM dNTPs, 10x PCR buffer provided with Taq polymerase,
2.5 mM MgCl2, 40 pmol of sense and antisense primer for the gene of interest, 40 pmol of
sense and antisense GAPDH primer and water to bring the final volume to 50 µl. All
reaction mixtures were overlaid with 50 µl of mineral oil (Sigma, The Netherlands). After
3 min denaturation at 94 °C, n cycles (Table 1) of amplification were performed, each for
1 min at 94°C, 1 min at 56 °C and 1 min at 72 °C, using the thermocycler Perkin Elmer
480 (The Netherlands). After the last cycle the 72 °C elongation step was extended to 5
min. The PCR products were separated on 1-2.5% agarose gels by gel-electrophoresis and
stained with ethidium bromide. The densities of the PCR products were quantified by
densitometry (Aldus PhotoStyler 2.0, Grafic Workshop and ImageQuant Version 3.3).
During the PCR for L-type Ca2+ the three isoforms of this channel were amplified.
The primers were designed to amplify the IVS1 upto IVS5 region of the α1 subunit. The
PCR fragment contained the IVS3B adult isoform, the IVS3B deleted D1 isoform and the
IVS3A fetal isoform, which could be identified as products of 472 bp, 442 bp and 355 bp,
24
Chapter 2
respectively, after digestion with DdeI (Promega, The Netherlands).
The RyR2 PCR fragment was designed to contain the alternative splicing site, a 24
bp insert between residues 11145 and 11146 with a restriction site for BamH1. The RyR2
isoform with the 24 bp insert could be identified as products of 310 bp and 326 bp after
digestion with BamH1 (Promega, The Netherlands).
Determination of the absolute alterations of mRNA
To validate the semi quantitative PCRs the changes observed in ratios for L-type Ca2+
and SR Ca2+ ATPase were determined. Increasing amounts of gene of interest were added
to a fixed amount of GAPDH. Therefore, known amounts of GAPDH input template (range
10-2 to 10-4 ng, 410 bp) were added to a PCR sample mixture of 0.5 unit of Taq polymerase
(Eurogentec, Belgium),17.5 nM dNTPs, 10x PCR buffer, 2.5 mM MgCl2, 40 pmol of
sense and antisense GAPDH primer, 0.7 µl of cDNA mixture and water to bring the final
volume to 50 µl, and amplified for 25 cycles. Thereafter, a fixed amount of GAPDH input
template was used in a PCR with known amounts of the SR Ca2+-ATPase input template
(range 10-1.5 ng to 10-3.5 ng, 657 bp) or L-type Ca2+ input template (range 10-1.5 ng to 10-3.5
ng, 563 and 530 bp). The ratios SR Ca 2+ ATPase input or L-type Ca2+ input versus GAPDH
input were calculated.
Table 1. Sequence for the primers
Sequence cycles annealing
temp (0C)
Glyceraldehyde-3-phosphate dehydrogenase:
F 5'-CCC ATC ACC ATC TTC CAG GAG CG-3', - 56
R 5'-GGC AGG GAT GAT GTT CTG GAG AGC C-3'
Phospholamban:
F 5'-ATG GAG AAA GTC CAA TAC CTC ACT CGC-3', 25 56
R 5'-TCA GAG AAG CAT CAC GAT GAT ACA GAT CAG-3'
Sarcoplasmic reticulum calcium ATPase:
F 5’-TGT TCA TTC TGG ACA GAG TGG AAG G-3’ 25 56
R 5’-TTA ATA AAG TTG GCA GAG TCC TCA AGG-3’
L-type calcium channel :
F 5’-CTG GAC AAG AAC CAG CGA CAG TGC G-3’, 32 56
R 5’-ATC ACG ATC AGG AGG GCC ACA TAG GG-3’
Sodium-calcium exchanger:
F 5’- CTA CCA AGT CCT AAG TCA GCA GC3’, 27 56
R 5’-GAT CCG AGG CAA GCA AGT GTA GA-3’
Ryanodine receptor type 2:
F 5’-AAG GCA TCG GGC TGT CAA TCT-3’ 28 56
25
Gene expression of proteins influencing calcium homeostasis in patients
Protein Preparation and Slot-Blot Analysis
Frozen RAAs of 5 patients in sinus rhythm, 5 patients with paroxysmal and 5 with
persistent AF were homogenized in RIPA buffer (1% NP40, 0.5% sodium deoxycholate,
0.1% sodium dodecyl sulfate (SDS), 10 mM β mercapto-ethanol, 10mg/ml PMSF, 5µl/ml
aprotinin, 100 mM sodium orthovanadate, 5µl/ml benzamidine, 5µl/ml pepstatine A, 5 µl/
ml leupeptine in 1x PBS by use of an ultraturrax (Polytron, The Netherlands) with 10
seconds intervals. The homogenate was centrifuged at 14.000 rpm for 20 minutes at 4°C.
After centrifugation the supernatant was carefully removed and used for protein
concentration measurement. This was done according to the Bradford method (Sigma,
The Netherlands) with bovine albumin used as a standard. Samples of 10 µg protein were
denaturated by heating to 95°C before spotting on a TBS (10 mM Tris-HCl pH 8.0, 150
mM NaCl ) wetted nitrocellulose membrane (Bio-Rad, The Netherlands) by use of a slot
blot apparatus (Bio-Rad, The Netherlands). The membrane was washed twice with 200 µl
TBS buffer and the transfer was checked by staining the nitrocellulose membrane with
Ponceau S solution (Sigma, The Netherlands). Blocking was performed for 20 minutes in
blocking buffer (5% nonfat milk, TBS and 0.1% Tween 20). After three times washing for
5 minutes in TBS with 0.1% Tween 20 the membranes were incubated for 90 minutes with
primary antibodies: SR Ca 2+ ATPase 1:2500, PLB 1: 200 (gifts from dr. F. Wuytack
University Leuven, Belgium), GAPDH 1:5000 (Affinity Reagents, USA) or L-type Ca2+
anti α1-subunit (Alomone Labs, Israel). Immunodetection of the primary antibody was
performed, after three times washing for 5 minutes with TBS and 0.1% Tween 20, with
peroxidase conjugated secondary antibody anti-rabbit and anti-mouse IgG (Santa Cruz
Biotechnology, The Netherlands) for 60 minutes. The primary and secondary antibodies
were diluted in blocking buffer. The blot was washed two times for 5 minutes in TBS and
0.1% Tween 20 and one time in TBS also for 5 minutes. Consequently, the blot was
incubated with the ECL-detection reagent (Amersham, The Netherlands) for 1 minute,
and exposed to a X-OMAT x-ray film (Kodak, The Netherlands) for 15 to 90 seconds. The
band densities were evaluated by densitometric scanning using a Snap Scan 600 (Agfa,
The Netherlands). To test the linearity of the immunodetection system distinct amounts of
protein were analyzed. There was a linear relation between protein amounts spotted on the
membrane and the immunoreactive signals of L-type Ca2+, SR Ca 2+-ATPase, PLB and
GAPDH (data not shown).
Definitions
Persistent AF: continuous presence of AF until the moment of cardiac surgery, i.e. at
least two consecutive electrocardiograms of AF more than 24 hours apart, without
intercurrent SR. Persistent AF has a non-spontaneously converting character [15, 16].
Previously, this type of AF was classified as chronic AF.
26
Chapter 2
Paroxysmal AF: AF typically occurring in episodes of a duration shorther than 24
hours (but longer lasting paroxysms are not unusual) with intermittent sinus rhythm.
Paroxysmal AF is either converting spontaneously or is terminated with an intravenously
administered antiarrhythmic drug [15, 16]. It is non-controlled whether paroxysmal AF is
present at the moment of cardiac surgery.
Statistical Analysis
All PCRs and SDS-PAGEs were performed in duplo series. The mean values of the
ratios were used for statistical analysis. To compare the baseline characteristics between
groups for normally distributed variables, mean values and standard deviations are reported.
In case of skewed distribution of variables, the median values and ranges are given. Baseline
comparison between groups for normally distributed variables was performed by one-way
ANOVA for skewed distributed variables by the Wilcoxon two-sample test. The Chi-square
test with continuity correction or Fisher’s exact test was performed for group comparison
for categorical variables when appropriate. To determine which variables influenced mRNA
levels of proteins, univariate regression analyses were performed. Only variables with a p
value < 0.15 were selected for multiple regression analysis. To determine differences in
mRNA levels of these proteins between the four groups a Tukey correction for multiple
comparison was performed.
For determination of correlations the Spearman correlation test was used. The Mann-
Whitney U-test was performed for group to group comparisons of small numbers.
All p-values are two-sided, a p-value <0.05 was considered statistically significant.
SAS version 6.12 (Cary, NC) was used for all statistical evaluations.
Results
Patients
Included were 8 patients with paroxysmal and 10 patients with persistent AF. These
two groups were compared with two groups of control patients in sinus rhythm, who were
matched for sex, age and left ventricular function (Table 2). Six of the 8 patients with
paroxysmal AF suffered from intractable AF and were scheduled for Cox’s MAZE surgery.
The median duration of sinus rhythm before surgery was 1.5 days. The median frequency
of paroxysms was once a day (median duration of each paroxysm was 3 hours). Three
patients with paroxysmal AF had AF at the moment of surgery and harvesting of the RAA.
Control RAAs were obtained from clinically stable patients in sinus rhythm who were
scheduled for coronary artery bypass surgery. Although the AF groups and their controls
in sinus rhythm differed with respect to the underlying heart disease, all had a normal left
ventricular function and were in the functional class I or II for exercise tolerance. Also,
echocardiographic atrial and left ventricular dimensions were similar among groups.
27
Gene expression of proteins influencing calcium homeostasis in patients
Alterations in mRNA Levels in Paroxysmal and Persistent AF
Changes in mRNA levels of the gene of interest were determined by comparison of
gene-of-interest/GAPDH ratios between patients with persistent AF or with paroxysmal
AF, and their matched controls in sinus rhythm. The densities of the amplified GAPDH of
the 4 groups of patients were the same for all the genes investigated (data not shown).
Only patients with persistent AF showed a significant reduction of the cDNA ratios of L-
type Ca2+/GAPDH (-57%) and SR Ca2+-ATPase/GAPDH (-28%) (Figure 1A and 1B). The
cDNA ratios of Phospholamban/GAPDH , Na+/Ca2+ exchanger/GAPDH and RyR2/GAPDH
were unchanged compared to the controls in sinus rhythm (Figure 1C, 1D and 1E).
Table 2. Baseline characteristics of patients with paroxysmal AF, persistent AF and patients in sinus
rhythm at the moment of surgery.
PAF SR (PAF) CAF SR (CAF)
N 8 8 10 10
Male/ female (n) 6/2 6/2 6/4 6/4
Age 51 ±7 56 ± 11 63 ± 11 65 ± 17
Previous duration of AF (median, range (months) - - - 18 (8-64) -
Duration SR before surgery (median, range (days) 1.5 (0-30) - - -
Underlying heart disease (n)
Coronary artery disease 2* 8 4* 10
Hypertension 1 1 3 2
Lone AF 6* 0 5* 0
Surgical procedure
Coronary Artery Bypass Grafting 2* 8 4* 10
MAZE 6* 0 6* 0
New York Heart Association for exercise tolerance
Class I 7 5 6 5
Class II 1 3 4 5
Left atrial diameter (long axis, mm) 43 ± 7 41 ± 3 45 ± 7 44 ± 5
Left atrial diameter (apical, mm) 60 ± 6 64 ± 3 63 ± 4 64 ± 6
Right atrial diameter (apical, mm) 54 ± 9 54 ± 4 62 ± 7 57 ± 4
Left ventricular end-diastolic diameter (mm) 48 ± 4 49 ± 8 53 ± 3 53 ± 6
Left ventricular end-systolic diameter (mm) 35 ± 4 35 ± 7 33 ± 6 35 ± 4
Beta blockers 1* 5 3 6
Calcium antagonists 0 3 3 3
Digitalis 0 1 5 3
ACE inhibitors 0 1 4 2
* p-value < 0.05 compared to the control group
Values are presented as mean value ± SD. ACE indicates Angiotensin Converting Enzyme; CAF, chronic
persistent atrial fibrillation; PAF, paroxysmal atrial fibrillation; SR (CAF), matched controls in sinus rhythm of
patients with persistent AF; SR (PAF), matched controls in sinus rhythm of patients with paroxysmal AF.
28
Chapter 2
No changes were observed in patients suffering from paroxysmal AF. Table 3 shows that
the cDNA ratio L-type Ca2+/GAPDH in patients with persistent AF was neither influenced
Figure 1.
Individual cDNA ratios for L-type Ca2+/GAPDH (1A), SR Ca2+ ATPase/GAPDH (1B), PLB/GAPDH (1C), Na+/
Ca2+ exchanger/GAPDH (1D) and RyR2/GAPDH (1E). Mean values ar given for the 4 different groups ± SEM.
SR-CAF is the sinus rhythm control group for patients with chronic persistent AF, SR-PAF is the sinus rhythm
control group for patients with paroxysmal AF. All data are represented in density units/density units. Significance
was determined using the Tukey correction for multiple regression analysis for each mean of the four groups.
29
Gene expression of proteins influencing calcium homeostasis in patients
by the underlying heart disease nor by any (calcium handling influencing) drug. However,
patients in the sinus rhythm control group for paroxysmal AF who were treated with a beta
blocker (n=5) had an increased cDNA ratio for L-type Ca2+/GAPDH compared to those
who were not treated with the drug (n=3).
Validation of the Absolute mRNA Contents
To assess the amounts of L-type Ca2+ and SR Ca2+-ATPase in the cDNA mixtures in
different groups, a semi-quantitative PCR was developed with increasing amounts of input
cDNA of interest to a standard amount of GAPDH cDNA (0.018 ng).The ratios were
determined and plotted in a logarithmic way. This resulted in a straight line demonstrating
the validity of the method and enabling estimations of differences of L-type Ca2+ and SR
Ca2+ ATPase in the cDNA mixture (Figure 2A and 2B).
Figure 2.
Plot showing a significant correlation between the increasing doses of L-type Ca2+ (A) and SR Ca2+ ATPase (B)
input template and the ratios of L-type Ca2+/GAPDH and SR Ca2+ ATPase/GAPDH respectively. Correlation was
determined by the Spearman correlation test.
30
Chapter 2
Alterations in Proteins Levels in Paroxysmal and Persistent AF
Changes in protein levels were studied for the genes which showed alterations in
mRNA ratios (L-type Ca2+ and SR Ca2+-ATPase) and for phospholamban. Sufficient RAA
tissue to carry out protein isolations were available from 5 patients with paroxysmal AF, 5
with persistent AF and 5 patients in sinus rhythm. The protein levels of gene-of-interest/
GAPDH ratios were determined. Figure 3A shows the specificity of the antibodies used,
figure 3B the results of the slot-blot analysis. The protein ratio of L-type Ca2+/GAPDH
and SR Ca2+-ATPase/GAPDH were significantly reduced in patients with persistent AF
compared to the controls (-43%, p=0.02 and –28%, p=0.04, respectively, Figures 4A and
B). The PLB/GAPDH ratio showed a slight upregulation in patients with persistent AF
(Figure 4C). The GAPDH levels were comparable between the different groups (mean
values not shown). No alterations were found in the protein levels of patients with
paroxysmal AF (Figure 4). A positive correlation between the mRNA ratios and the protein
ratios of L-type Ca2+ and SR Ca2+-ATPase for patients with persistent AF and sinus rhythm
could be demonstrated (Figure 5).
Table 3. Effects of underlying heart disease and medication on the mRNA levels of l-type Ca2+ channel.
SR-CAF CAF SR-PAF PAF
CABG yes 2.4 (1.2–3.1, n=10) 1.6 (1.5–1.9, n=4) 2.6 (1.2–3.7, n=8) 2.6 (2.6-2.6, n=2)
no - 1.8 (0.7-2.4, n=6) - 2.2 (1.9-2.6,n=6)
MAZE yes - 1.8 (0.7-2.4, n=6) - 2.6 (2.6-2.6, n=6)
no 2.4 (1.2-3.1, n=10) 1.8 (1.5-2.2, n=4) 2.6 (1.2-3.7, n=8) 2.2(1.9-2.6, n=2)
Beta blockers yes 2.5 (1.9-3.1, n=6) 1.7 (1.5-1.9, n=3) 3.1 (2.5-3.7, n=5)* 2.6 (n=1)
no 2.2 (1.2-2.6, n=4) 1.8 (0.7-2.4, n=7) 1.8 (1.2-2.4, n=3)* 2.3 (1.9-2.6, n=7)
Calcium entry yes 2.7 (2.5-2.9, n=3) 1.8 (1.5-1.9, n=3) 2.4 (1.7-3.2, n=3) -
Blockers no 2.3 (1.2-3.1, n=7) 1.8 (0.7-2.4, n=7) 2.7 (1.2-3.7, n=5) 2.3 (1.9-2.6, n=8)
Digitalis yes 2.4 (1.9-2.7, n=3) 1.8 (1.5-2.2, n=5) 2.5 (n=1) -no 2.4 (1.2-3.1, n=7) 1.7 (0.7-2.4, n=5) 2.6 (1.2-3.7, n=7) 2.3 (1.9-2.6, n=8)
ACE-inhibitors yes 2.0 (1.9-2.1, n=2) 1.5 (0.7-2.2, n=4) 3.2 (n=1) -
no 2.5 (1.2-3.1,n=8) 1.9 (1.5-2.4, n=6) 2.5 (1.2-3.7, n=7) 2.3 (1.9-2.6, n=8)
* p value is 0.01 between SR-PAF with and without beta blocker (Mann-Whitney U-test)
The values are expressed as mean (range, number of patients). CAF, chronic persistent atrial fibrillation;
PAF, paroxysmal atrial fibrillation; SR (CAF), matched controls in sinus rhythm of patients with persistent
AF; SR (PAF), matched controls in sinus rhythm of patients with paroxysmal AF.
31
Gene expression of proteins influencing calcium homeostasis in patients
Figure 3.
A.Western blot analysis showing the specificity of the antibodies on human RAA. Anti-L-type calcium channel
α1 subunit (200 kD, lane 1), anti-SR Ca2+ ATPase (105 kD,lane 2), anti-phospholamban (6.5 kD, lane 3), GAPDH
(36 kD,lane 4) and the Marker (lane 5, Bio Rad, The Netherlands).
B. Slot-blot analysis of L-type Ca2+, SR Ca2+ ATPase, PLB and GAPDH of 5 control patients, 5 patients with
chronic persistent AF (CAF) and 5 patients with paroxysmal AF (PAF).
1 2 3 4 M
200 kD
116 kD97 kD
66 kD
45 kD
31 kD
22 kD14 kD
6.5 kD
A
GAPDH
PLB
SR Ca2+ ATPase
Control
CAF
PAF
Control
Control
CAF
CAF
PAF
PAF
PAF
Control
CAF
L-type Ca2+B
A
B
32
Chapter 2
Isoforms of L-type Calcium Channel
To investigate whether dedifferentiation occurred as an adaptation process in patients
with AF, the mRNA expression level of the fetal isoform compared to the adult isoforms
of L-type Ca2+ was determined. The amplified PCR fragment of L-type Ca2+ contained the
IVS3A (fetal form), the IVS3B (adult form) and the IVS3B deleted D1 form after digestion
with DdeI (Figure 6). No differences in percentages of the adult and fetal isoforms of L-
type Ca2+ were found in relation to the total quantity of the amplified L-type Ca2+ in the
different groups (Figure 7A). The ratio of the IVS3B form was significantly reduced in
patients with persistent AF (p=0.01, Figure 7B). No significant changes were seen in the
IVS3B deleted D1 form and the IVS3A form. Any significant alteration in the L-type Ca2+
ratios was observed in patients with paroxysmal AF (Figure 7B, 7C and 7D).
Figure 4.
Protein ratios of L-type Ca2+/GAPDH (A), SR Ca2+ ATPase/GAPDH (B) and PLB/GAPDH (C) of the individual
patients. SR is the sinus rhythm control group. All data are presented as density units/density units. Values are
mean ± SEM. Significant differences are indicated (Mann-Whitney U-test).
33
Gene expression of proteins influencing calcium homeostasis in patients
Figure 5.
Relationship between the mRNA and protein expression of L-type Ca2+(A), SR Ca2+ ATPase (B) and PLB (C).
(•) represents chronic persistent AF patients, (ο) represents SR patients. Correlation was determined by the
Spearman correlation test.
Discussion
Main Findings
We examined five different genes which play an important role in the calcium
homeostasis of the myocardial cell. By examining the mRNA and protein expression in
patients with paroxysmal AF and persistent AF, and for age, sex and cardiac function
matched controls in sinus rhythm we observed two important features. First, the mRNA
and protein contents of both SR Ca2+-ATPase and L-type Ca2+ were significantly reduced
in patients with persistent AF. Secondly, significant changes occurred only in patients with
34
Chapter 2
Figure 7.
Figure A shows the percentage IVS3B, IVS3B deletion and IVS3A in relation to the total L-type Ca2+ protein
mRNA expression. No differences in percentage were found between the three isoforms. The cDNA ratios of the
IVS3B/GAPDH was significantly lower in patients with persistent AF (B). No alterations were observed between
the groups of cDNA ratios of IVS3B deletion/GAPDH (C) and the fetal form IVS3A/GAPDH (D). SR-CAF are
the matched control patients in sinus rhythm for chronic persistent AF and SR-PAF for paroxysmal AF. All data
are presented as density units/density units. Values are mean ± SEM. Significant differences are indicated (Mann-
Whitney U-test).
Figure 6.
Typical example of an agarose gel. Here, the L-type Ca2+ isoforms (IVS3B, 472 bp; IVS3B deletion, 442 and
IVS3A, 355 bp) of 4 patients with chronic persistent AF and their matched controls in sinus rhythm and patients
with paroxysmal AF and their controls are shown.
IVS3BIVS3B del.
IVS3A
GAPDH
CAF CAF CAF CAF CAF PAF PAF
SR SR SR SR SR M SR SR
35
Gene expression of proteins influencing calcium homeostasis in patients
persistent AF but not in those with paroxysmal AF.
Alterations in Gene Expression of Ion Channels and Proteins Involved in the Calcium
Handling
The alterations observed in the present study of a reduction of the mRNA and protein
contents for the L-type Ca2+ without changes in L-type Ca2+ isoforms in patients with
persistent AF, fit in with the described changes in the above mentioned experimental models.
Clearly, a lower mRNA and protein expression of L-type Ca2+ may underlie reduced L-
type Ca2+ current densities [10, 11, 17]. Whereas in the experiments of Yue et al. significant
reduction of L-type Ca2+ current densities was observed within 6 weeks of rapid continuous
atrial pacing, we, however, observed significant alterations only in patients with persistent
longstanding (> 8 months) AF.
Data on alterations in the sarcoplasmic reticulum proteins influencing the calcium
handling in RAA tissue of patients with AF are lacking. The observed reduction of the SR
Ca2+-ATPase mRNA and protein expression in atrial tissue of patients with persistent AF
is comparable to data on left ventricular tissue of patients with severe heart failure due to
dilated and ischemic cardiomyopathy [18_23]. In our population both mRNA and protein
expression were reduced. In contrast, a discrepancy between adaptation of the SR Ca2+-
ATPase mRNA, protein and activity levels have been reported in ventricular tissue of
patients suffering from severe heart failure [19, 20, 23_25]. The mRNA and protein
expression of phospholamban, the regulatory protein of SR Ca2+-ATPase, was not
significantly different between the AF groups and their controls. A reduced gene expression
of SR Ca2+-ATPase and a not significantly changed expression of phospholamban yields a
reduced ratio of SR Ca2+-ATPase to phospholamban in patients with persistent AF. If we
assume that the stoichiometry of phospholamban to SR Ca2+-ATPase determines the level
of SR Ca2+-ATPase inhibition, this finding may indicate that in the basal low-phosphorylated
state, depression of SR calcium uptake is even more pronounced than would be expected
from the lower SR Ca2+-ATPase mRNA level in patients with persistent AF. This
interpretation would be consistent with functional abnormalities observed in the failing
human ventricular myocardium [22].
For the ryanodine receptor the amounts of a RyR2 region with or without the 24 bp
insert were examined. Alternative transcripts with or without this insertion might provide
a means for altering the binding affinity of this putative site for calcium [26]. No differences
were found in the expression of RyR2 with or without insert between the groups. This
suggests that persistent AF did not induce differences in the calcium binding affinity and
changes in fundamental nature of the RyR2. Furthermore, no changes in mRNA and protein
expression of RyR2 were observed in patients with AF.
36
Chapter 2
No Dedifferentiation of the L-type Calcium Channel During AF
Each IVS3 isoform is encoded by a separate but adjacent exon within a single genomic
clone and the various isoforms are generated by a developmentally regulated, mutually
exclusive exon splicing of the primary transcript [27, 28]. No reversion of the adult isoform
of L-type Ca2+ to its fetal isoform was observed in patients with AF. In contrast, however,
Gidh-Jain et al. demonstrated that in patients with left ventricular hypertrophy a significant
increase of the mRNA contents of the fetal isoform and reversion of fetal/ adult isoform
ratio to the fetal phenotype was observed in ventricular tissue [29]. In atrial goat tissue,
Ausma et al. showed that during pacing induced AF proteins which are present in embryonic/
fetal myocardial cells, e.g. α-smooth muscle actin, were reexpressed [30]. A reversion to
fetal isoforms of certain ion channels and proteins might be hypothesized to occur in
situations comparable to the embryonic situation, e.g. during higher heart rates as is the
case during AF. Therefore, a reexpression of fetal proteins might have occurred during AF.
No Changes in Gene Expression in Patients with Severe Paroxysmal AF
We observed significant alterations in expression of the investigated genes only in
patients with longstanding (> 8 months) persistent AF. No significant changes were observed
in those patients suffering from paroxysmal AF. Importantly, the included paroxysmal AF
patients had severe AF with daily episodes of AF. Moreover, 3 patients had AF at the
moment of surgery, i.e. at the moment of harvesting of the right atrial appendage. This
may suggest that episodes of sinus rhythm in between episodes of AF protect the myocardial
cell from alterations in gene expression.
Limitations of the Study
Drugs and differences in underlying diseases may influence gene expression of proteins
and ion channels influencing the calcium handling. In this study, to minimize the influence
of particular clinical parameters on gene expression, we included only patients with a
normal left ventricular function and, when possible, drugs were discontinued before surgery.
The present study does not clarify when the adaptive mechanisms of the atrial
myocardial cell, i.e. alterations in gene expression, start. Although our data suggest that
this is a late process, no patients with persistent AF with a duration between 1 day and 8
months were included .
No matched controlled analysis could be performed for determination of protein levels.
However, no significant changes in mRNA levels between the control groups were observed.
Therefore, in our opinion, a comparison between persistent AF, paroxysmal AF and sinus
rhythm patients seems to be justified.
37
Gene expression of proteins influencing calcium homeostasis in patients
Acknowledgments
Dr. Van Gelder was supported by Grant 94.014 of the Netherlands Heart Foundation, The
Hague, The Netherlands. The study was supported by Grant 96.051 of the Netherlands
Heart Foundation, The Hague, The Netherlands. We are indebted to Pieter J. De Kam for
statistical analysis.
References
1 Kannel WB, Abbott RD, Savage DD, McNamara PM. Epidemiologic features of chronic atrial fibrilla-
tion: the Framingham study. N Engl J Med 1982;306:1018-1022.
2 Manning WJ, Silverman DI, Katz SE, et al. Impaired left atrial mechanical function after cardioversion:
relation to the duration of atrial fibrillation. J Am Coll Cardiol 1994;23:1535-1540.
3 Van Gelder IC, Crijns HJ, Blanksma PK, et al. Time course of hemodynamic changes and improvement
of exercise tolerance after cardioversion of chronic atrial fibrillation unassociated with cardiac valve
disease. Am J Cardiol 1993;72:560-566.
4 Wijffels MC, Kirchhof CJ, Dorland R, Allessie MA. Atrial fibrillation begets atrial fibrillation. A study
in awake chronically instrumented goats. Circulation 1995;92:1954-1968.
5 Leistad E, Aksnes G, Verburg E, Christensen G. Atrial contractile dysfunction after short-term atrial
fibrillation is reduced by verapamil but increased by BAY K8644. Circulation 1996;93:1747-1754.
6 Lee HC, Clusin WT. Cytosolic calcium staircase in cultured myocardial cells. Circ Res 1987;61:934-
939.
7 De Pauw M, Borgers M, Heyndrickx GR. Ultrastuctural calcium distribution in cardiac myocytes after
48h of rapid pacing in dogs. Circulation 1996;94:I-604 (Abstract)
8 Schouten VJA, Morad M. Regulation of Ca2+ current in frog ventricular myocytes by the holding
potential, cAMP and frequency. Pflugers Arch 1989;415:1-11.
9 Ausma J, Wijffels M, Thone F, Wouters L, Allessie M, Borgers M. Structural changes of atrial myocar-
dium due to sustained atrial fibrillation in the goat. Circulation 1997;96:3157-3163.
10 Yue L, Feng J, Gaspo R, Li GR, Wang Z, Nattel S. Ionic remodeling underlying action potential changes
in a canine model of atrial fibrillation. Circ Res 1997;81:512-525.
11 Van Wagoner DR, Lamorgese M, Kirian P, Cheng Y, Efimov IR, Mazgalev TN. Calcium current density
is reduced in atrial myocytes isolated from patients in chronic atrial fibrillation. Circulation 1997;96:I-
180 (Abstract)
12 Daoud EG, Knight BP, Weiss R, et al. Effect of verapamil and procainamide on atrial fibrillation-in-
duced electrical remodeling in humans. Circulation 1997;96:1542-1550.
13 Sun H, Gaspo R, Leblanc N, Nattel S. Cellular mechanism of atrial contractile dysfuntion caused by
sustained atrial tachycardia. Circulation 1998;98:719-727.
14 Brundel BJ, Van Gelder IC, Henning RH, Tuinenburg AE, Van Gilst WH, Crijns HJ. Downregulation of
the mRNA expression of the acethylcholine-activated potassium channel and L-type calcium channel in
patients with chronic atrial fibrillation. Pacing Clin Electrophysiol 1997;20:II:1050 (Abstract)
15 Gallagher MM, Camm AJ. Classification of atrial fibrillation. Pacing Clin Electrophysiol 1997;20:1603-
1605.
16 Levy S, Breithardt G, Campbell RW, et al. Atrial fibrillation: current knowledge and recommendations
for management. The Working Group on Arrhythmias of the European Society of Cardiology. Eur
Heart J 1998;19:1294-1320.
17 Yue L, Wang Z, Gaspo R, Nattel S. The molecular mechanism of ionic remodeling of repolarization in a
dog model of atrial fibrillation. Circulation 1998;96:I-180 (Abstract)
18 Mercadier JJ, Lompre AM, Duc P, et al. Altered sarcoplasmic reticulum Ca2+-ATPase gene expression in
the human ventricle during end-stage heart failure. J Clin Invest 1990;85:305-309.
19 Hasenfuss G, Reinecke H, Studer R, et al. Relation between myocardial function and expression of
sarcoplasmic reticulum Ca2+-ATPase in failing and nonfailing human myocardium. Circ Res 1994;75:434-
442.
20 Linck B, Boknik P, Eschenhagen T, et al. Messenger RNA expression and immunological quantification
of phospholamban and SR-Ca2+-ATPase in failing and nonfailing human hearts. Cardiovasc Res
1996;31:625-632.
21 Studer R, Reinecke H, Bilger J, et al. Gene expression of the cardiac Na+-Ca2+ exchanger in end-stage
human heart failure. Circ Res 1994;75:443-453.
38
Chapter 2
22 Meyer M, Schillinger W, Pieske B, et al. Alterations of sarcoplasmic reticulum proteins in failing human
dilated cardiomyopathy. Circulation 1995;92:778-784.
23 Flesch M, Schwinger RH, Schnabel P, et al. Sarcoplasmic reticulum Ca2+ATPase and phospholamban
mRNA and protein levels in end-stage heart failure due to ischemic or dilated cardiomyopathy. J Mol
Med 1996;74:321-332.
24 Movsesian MA, Karimi M, Green K, Jones LR. Ca2+-transporting ATPase, phospholamban, and
calsequestrin levels in nonfailing and failing human myocardium. Circulation 1994;90:653-657.
25 Schwinger RH, Böhm M, Schmidt U, et al. Unchanged protein levels of SERCA II and phospholamban
but reduced Ca2+ uptake and Ca2+-ATPase activity of cardiac sarcoplasmic reticulum from dilated cardi-
omyopathy patients compared with patients with nonfailing hearts. Circulation 1995;92:3220-3228.
26 Tunwell RE, Wickenden C, Bertrand BM, et al. The human cardiac muscle ryanodine receptor-calcium
release channel: identification, primary structure and topological analysis. Biochem J 1996;318:477-
487.
27 Diebold RJ, Koch WJ, Ellinor PT, et al. Mutually exclusive exon splicing of the cardiac calcium channel
alpha 1 subunit gene generates developmentally regulated isoforms in the rat heart. Proc Natl Acad Sci
USA 1992;89:1497-1501.
28 Feron O, Octave JN, Christen MO, Godfraind T. Quantification of two splicing events in the L-type
calcium channel alpha-1 subunit of intestinal smooth muscle and other tissues. Eur.J.Biochem.
1994;222:195-202.
29 Gidh JM, Huang B, Jain P, Battula V, el Sherif N. Reemergence of the fetal pattern of L-type calcium
channel gene expression in non infarcted myocardium during left ventricular remodeling. Biochem
Biophys Res Commun 1995;216:892-897.
30 Ausma J, Wijffels M, Van Eys G., et al. Dedifferentiation of atrial cardiomyocytes as a result of chronic
atrial fibrillation. Am J Pathol 1997;151:985-997.
39
Alterations in gene expression of proteins
Chapter 3
Alterations in Gene Expression of Proteins Involved in the
Calcium Handling in Patients with Atrial Fibrillation
Isabelle C. Van Gelder, MD; Bianca J.J.M. Brundel, MSc; Robert H. Henning,
MD,#; Anton E. Tuinenburg, MD; Robert G. Tieleman, MD; Leo Deelman,
MSc#; Jan G. Grandjean, MD*Pieter Jan De Kam, MSc; Wiek H. Van Gilst,
PhD#; Harry J.G.M. Crijns, MD
From the Departments of Cardiology, Clinical Pharmacology (#) and Thoracic Surgery
(*), Thoraxcenter, University Hospital Groningen, Groningen, The Netherlands.
Journal of Cardiovascular Electrophysiology 10 (1999) 552 - 560
Abstract
Introduction: Atrial fibrillation (AF) leads to a loss of atrial contraction within hours
to days. During persistence of AF cellular dedifferentiation and hypertrophy occur,
eventually resulting in degenerative changes and cell death. Abnormalities in the calcium
handling in response to tachycardia induced intracellular calcium overload play a pivotal
role in these processes. Methods and Results: The purpose was to investigate the mRNA
expression of proteins and ion channels influencing the calcium handling in patients with
persistent AF. Right atrial appendages were obtained from 18 matched controls in sinus
rhythm (group 1) and 18 patients with persistent AF undergoing elective cardiac surgery.
Previous duration of AF was ≤ 6 months in 9 (group 2) and > 6 months in 9 patients (group
3). In a single semi-quantitative polymerase chain reaction the mRNA of interest and of
glyceraldehyde-3-phosphate dehydrogenase were coamplified and separated by gel-
electrophoresis. L-type calcium channel α1 subunit mRNA content was inversely related
to the duration of AF: -26% in group 2 compared to group 1 (p=0.2), and -49% in group 3
compared to group 1 (p=0.01). Inhibitory guanine nucleotide-binding protein iα2 mRNA
content was reduced in group 3 compared to group 1 (-30%, p=0.01). Sarcoplasmic
reticulum calcium ATPase, phospholamban and sodium-calcium exchanger mRNA contents
were not affected by AF. Conclusions: AF-induced alterations in mRNA contents of proteins
and ion channels
40
Chapter 3
involved in the calcium handling seem to occur in relation to the previous duration of AF.
In the present patient population these changes were significantly only if AF had lasted >
6 months.
Introduction
Atrial fibrillation (AF) is responsible for patient discomfort, thromboembolic
complications and heart failure.1-3 AF has the tendency to become permanent over time,4
which is illustrated by the fact that cardioversion to and maintenance of sinus rhythm
become increasingly difficult the longer the arrhythmia exists.5-7
AF leads to atrial contractile dysfunction, i.e. loss of atrial contraction, within hours
to days in both humans and experimental models.8-12 This process appears to be reversible,
its time course being related to the previous duration of AF.10-11,13,14 During persistence of
AF cellular dedifferentiation (resembling hibernation) and cellular hypertrophy may
occur,15,16 eventually resulting in degenerative changes like fibrosis and cell death.17-19 There
are strong indications that abnormalities in the calcium handling, in response to tachycardia
induced intracellular calcium overload,20-22 play a pivotal role in these adaptive
processes.10,12,15,23,24 The proteins and ion channels involved in the adaptation of the atria
during AF have not been clarified yet. Most likely, identification of the signaling pathways
and their target genes may lead to new therapeutic options for the treatment of AF. We
hypothesized that changes in mRNA expression of proteins and ion channels involved in
the calcium handling in the atria during AF would be comparable to those observed in the
ventricles during pacing-induced heart failure, and would be more pronounced the longer
AF had lasted. Therefore, it was our purpose to investigate the alterations in mRNA contents
of proteins and ion channels involved in the calcium handling in right atrial appendages of
patients undergoing cardiac surgery. A second aim was to determine whether these changes
were more pronounced in patients with AF of a duration > 6 months compared to those
with AF of a duration ≤ 6 months.
Methods
Patients
The day before surgery clinical characteristics of the patient were assessed by one
investigator (AET). Presence, type and duration of AF were assessed by patient’s complaints
and previous electrocardiograms. In addition, the patient was asked for medication use
and exercise tolerance (according to the New York Heart Association classification).
Echocardiography was performed within 3 months of date of surgery. Right atrial
appendages were obtained from 18 patients with persistent AF and from 18 controls in
sinus rhythm who were matched for age and sex, and as far as possible for underlying
disease and cardiac function. All patients were scheduled for elective cardiac surgery. The
41
Alterations in gene expression of proteins
study was approved by the Institutional Review Board, and written informed consent was
given by all patients. Immediately after excision the right atrial appendages were frozen in
liquid nitrogen and stored at -850C.
RNA Isolation and cDNA Synthesis
Total RNA was isolated from right atrial appendages using the method of acid
guanidinium thiocyanate/phenol/chloroform extraction followed by a RNeasy kit for RNA
minipreps for tissues (Qiagen, Hilden, Germany). The amount of RNA was evaluated by
absorption at 260 nm, using a GeneQuant II (Pharmacia LKB Biotechnology, The
Netherlands). The ratio of absorption (260-280 nm) of all preparations was between 1.8
and 2.0. First strand cDNA was synthesized by incubation of 1 µg of total RNA, reverse
transcription 10x buffer (Promega, The Netherlands), 200 ng of random hexamers (Promega,
The Netherlands) with 200 units of Moloney Murine Leukemia Virus Reverse Transcriptase
(Promega, The Netherlands), 1mM of each dinucleotidetriphosphate (dNTP) and 1 unit of
RNAse inhibitor (Promega, The Netherlands). The total volume was 20 µl. The synthesis
reaction lasted 10 minutes at 200C, 20 minutes at 420C, 5 minutes at 990C and 5 minutes at
40C, respectively. To assure that the amplification products did not arise from contaminating
DNA, a reverse transcriptase was performed without the addition of the enzyme. After
amplification of this product, no PCR products were detectable on an agarose gel implying
the absence of contaminating DNA in the RNA preparations.
Semi-Quantitative Polymerase Chain Reaction Analyses
Since a linear relationship between the amount of input template and amount of
amplification product exists within the exponential range of amplification, a semi-
quantitative polymerase chain reaction (PCR) was developed.25,26 To validate the results
obtained with this semi-quantitative PCR a competititive PCR method was performed. In
our experiments, the cDNA of interest and the cDNA of the ubiquitously expressed
housekeeping enzyme glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were
coamplified in a single PCR. Primers were designed for L type calcium channel α1 subunit
(L-type Ca2+), Sarcoplasmic Reticulum Calcium-ATPase 2A (SR Ca2+-ATPase),
phospholamban, sodium-calcium exchanger (Na+-Ca2+ exchanger), inhibitory guanine
nucleotide-binding protein iα2 (Giα
2), and the housekeeping gene GAPDH. The
oligonucleotides were synthesized by Eurogentec (Belgium). The sequence for each primer
is given in Table 1.
For the semi-quantitative PCR co-amplification of SR Ca2+-ATPase, phospholamban,
Na+-Ca2+ exchanger and Giα2 with GAPDH, 0.5 unit of Taq polymerase (Eurogentec,
Belgium) was added to 17.5 nM of each dNTP, 10x PCR buffer provided with Taq
polymerase, 2.5 mM MgCl2, 40 pmol of sense and antisense primer for the gene of interest,
42
Chapter 3
40 pmol of sense and antisense GAPDH primer, 1 µl of cDNA mixture and water to bring
the final volume to 50 µl. All reaction mixtures were overlaid with 50 µl of mineral oil
(Sigma). After 3 min denaturation at 940C, n cycles of amplification (Table 1) were
performed, each for 1 min at 940C , 1 min at the annealing temperature (Table 1) and 1 min
at 720C, using the thermocyclar Pharmacia LKB Gene. After the last cycle, the 720C
elongation step was extended to 5 min. The conditions for the semi-quantitative PCR
analysis of L-type Ca2+ with GAPDH were the same as above except that the MgCl2
concentration was reduced (2.0 mM). The PCR products were separated on a 1-2% agarose
gel by gel-electrophoresis and stained with ethidium bromide. The density of the PCR
products were quantified by densitometry (Aldus PhotoStyler 2.0, Graphic Workshop and
ImageQuant Version 3.3).
During the PCR for L-type Ca2+ the three isoforms of this channel were amplified.
The primers were designed to amplify the IVS1 up to IVS5 region of the α1 subunit. The
PCR fragment contained the IVS3B (adult form), the IVS3B deleted D1, and IVS3A (fetal
TABLE 1.Sequence for Primers
Protein sequence cycles annealing temp
(0C)
Glyceraldehyde-3-phosphate dehydrogenase:
5'-CCC ATC ACC ATC TTC CAG GAG CG-3' - 56
5'-GGC AGG GAT GAT GTT CTG GAG AGC C-3'
L type calcium channel α1 subunit :
5’-CTG GAC AAG AAC CAG CGA CAG TGC G-3’ 32 56
5’-ATC ACG ATC AGG AGG GCC ACA TAG GG-3’
Sarcoplasmic Reticulum Calcium-ATPase 2A:
5'-TGT TCA TTC TGG ACA GAG TGG AAG G-3' 25 56
5'-TTA ATA AAG TTG GCA GAG TCC TCA AGG-3'
Phospholamban:
5'-ATG GAG AAA GTC CAA TAC CTC ACT CGC-3' 25 56
5'-TCA GAG AAG CAT CAC GAT GAT ACA GAT CAG-3'
sodium-calcium exchanger:
5’- CTA CCA AGT CCT AAG TCA GCA GC3’ 27 56
5’-GAT CCG AGG CAA GCA AGT GTA GA-3’
Inhibitory guanine nucleotide-binding protein α2:
5’-AGG GAA GAG CAC CAT CGT CAA GCA G-3’ 25 56
5’-AGC ACC AAG TCA TAG GCG CTC AAG G-3’
43
Alterations in gene expression of proteins
form) isoforms. These products could be identified as products of 475 bp, 442 bp and 355
bp, respectively, after digestion with DdeI (Promega, The Netherlands).
Determination of the absolute alterations of mRNA
To validate the semi-quantitative PCR the changes observed in the ratio for L-type
Ca2+ were determined. Increasing amounts of gene of interest were added to a fixed amount
of GAPDH. Therefore, known amounts of GAPDH input template (range 10-2 to 10-4 ng,
410 bp) were added to a PCR sample mixture of 0.5 unit of Taq polymerase (Eurogentec,
Belgium),17.5 nM dNTPs, 10x PCR buffer, 2.0 mM MgCl2, 40 pmol of sense and antisense
GAPDH primer, 0,7 µl of cDNA mixture and water to bring the final volume to 50 µl, and
amplified for 25 cycles. Thereafter, a fixed amount of GAPDH input template was used in
a PCR with known amounts of L-type Ca2+ input template (range 10-2.5 ng to 10-3.5 ng, 563
and 530 bp). The ratio L-type Ca2+ input versus GAPDH input was calculated.
Statistical Analysis
A two-sided probability level < 0.05 was considered to indicate statistical significance.
Mean values and standard deviations or standard errors are reported for normally distributed
variables. In case of skewed distributed variables, the median values and ranges are given.
Categorical variables are presented by frequencies. For the comparison of clinical
characteristics between the persistent AF groups and the sinus rhythm group a student’s t
test was performed if variables were normally distributed. The chi-square test with continuity
correction or Fisher’s exact test were performed for categorical variables. The Wilcoxon
two-sample test was performed if variables were not normally distributed. To determine
parameters related to the mRNA levels of the investigated genes multivariate regression
analysis was performed. Only covariates with p values ≤ 0.15 in the univariate analysis
were entered in this model. The backward selection procedure was employed using the
clinical relevant first order interactions to derive a model with statistically significant
predictors. To determine differences in mRNA levels between the three groups (sinus
rhythm, AF ≤ 6 months and AF > 6 months), Tukey correction for multiple comparisons
was performed. Correlation between the increasing amounts of L-type Ca2+ input template
and the ratio L-type Ca2+/GAPDH was determined by the Spearman correlation test. The
analysis was performed by SAS statistical software (SAS, version 6.12, Cary, NC).
Results
Patients
Included were 18 patients with persistent AF and 18 patients in sinus rhythm at the
moment of surgery (Table 2). Age, sex distribution, exercise tolerance and underlying
heart disease were comparable between groups. However, more patients in sinus rhythm
44
Chapter 3
underwent coronary artery bypass grafting. Only two patients had lone AF and were
scheduled for Cox’s maze III surgery.27 Patients with severe left ventricular dysfunction
were excluded from this study.
Changes in mRNA transcription of different proteins and ion channels
Changes in transcription of the gene of interest were determined by comparison of
the gene of interest/ GAPDH ratio between patients with AF and patients with sinus rhythm
at the moment of surgery. The densities of amplified GAPDH did not differ between the
groups in the respective PCR reactions for the different genes (not shown).
The cDNA ratio of L-type Ca2+/ GAPDH was significantly lower in patients with
persistent AF, predominantly caused by the patients with AF of a duration > 6 months.
These patients had a significantly lower cDNA ratio L-type Ca2+/ GAPDH compared to
those in sinus rhythm (-49%, p<0.01). In contrast, patients with persistent AF ≤ 6 months
showed no significant reduction (-26%, p=0.20, Figure 1A). Patients with AF > 6 months
had a significantly lower content of the IVS3B form (475 bp), the IVS3B deleted D1 form
(442 bp) and the sum of the three isoforms. No significant changes were observed in the
fetal isoform IVS3A (335bp) of L-type Ca2+ (Figure 1B). Neither functional class for
exercise tolerance (Figure 1C), nor the use of calcium handling influencing drugs (Table
3), nor any other baseline characteristic, were related to the ratio cDNA L-type Ca2+/
GAPDH.
The cDNA ratios of SR Ca2+-ATPase/ GAPDH, phospholamban/ GAPDH, and Na+-
Ca2+ exchanger/ GAPDH were neither altered by the presence of persistent AF (Figure 2,
Panel A, B and C), nor by any other parameter listed in Table 2. The cDNA ratio of Giα2/
GAPDH was significantly lower in patients with persistent AF of a duration > 6 months (-
30%, p=0.01, Figure 2D). No other clinical parameters as presented in Table 2 influenced
the latter ratio.
Validation of the absolute mRNA contents
To assess the amount of L-type Ca2+ in the cDNA mixtures in different groups, a
semi-quantitative PCR was developed with increasing amounts of input cDNA of interest
to a standard amount of GAPDH cDNA (0.018 pg).The ratios were determined and plotted
in a logarithmic way. This resulted in a straight line demonstrating the validity of the
method and enabling estimations of differences of L-type Ca2+ in the cDNA mixture (Figure
3). The mean ratio cDNA for patients in sinus rhythm was 2.3, of patients with AF < 6
months duration 1.7 and of patients with AF > 6 months 1.2. (Figure 1A), which means an
amount of L-type Ca2+ of 2.1 pg, 0.85 pg and 0.41 pg, respectively (Figure 3).
45
Alterations in gene expression of proteins
Table 2. Baseline Characteristics of Patients with Persistent Atrial Fibrillation and Their Matched
Controls in Sinus Rhythm (at Surgery)
persistent atrial sinus rhythm
fibrillation
all≤ 6 months > 6 months p value
Duration of atrial fibrillation
(median months, range) 3 (0.1-6) 24 (12-240) -
Age (mean “ SD) 65 ± 10 60 ± 9 69 ± 9 63 ± 11 n.s.
Male/ female (n) 12 / 6 2 / 7 5 / 4 14 / 4 n.s.
Underlying heart disease (n)*
Coronary artery disease 4 0 4 10 0.10
previous myocardial infarction 4 2 2 5 n.s.
Mitral valve disease 4 4 0 3 n.s.
Mitral stenosis 0 0 0
Mitral regurgitation 4 4 0 2
Aortic valve disease 6 3 3 7 n.s.
Aortic stenosis 5 2 3 4
Aortic regurgitation 4 1 3 3
Hypertension 3 0 3 2 n.s.
“Lone” atrial fibrillation 2 1 1 - n.s.
New York Heart Association Class for Exercise tolerance (n) 0.23
Class I 5 2 3 5
Class II 3 1 2 8
Class III 10 6 4 5
Drugs at surgery (n)
Digitalis 14 7 7 1 0.01
Calcium entry blockers 5 1 4 7 n.s.
Beta-blockers 4 1 3 8 n.s.
Angiotensin converting enzyme inhibitors 8 5 3 6 n.s.
Acenocoumarol 8 4 4 0 0.01
Left atrial long axis view (mean±SD, mm) 51±11 50±14 51±5 47±7 n.s.
Left ventricular end diastolic diameter
(mean±SD, mm 53±10 53±9 53±11 49±6 n.s.
Left ventricular end systolic diameter
(mean±SD, mm) 36±14 37±3 35±10 35±3 n.s.
n.s.= not statistically significant. p value indicates comparison between all AF and sinus rhythm patients.
* > 1 underlying disease per patient scored.
Discussion
This study demonstrates selective changes in mRNA contents of proteins and ion
channels involved in the calcium handling in patients suffering from persistent AF. L-type
Ca2+ and Giα2 mRNA contents were reduced in patients with longstanding persistent AF,
whereas SR Ca2+-ATPase, phospholamban and Na+-Ca2+ mRNA contents were not affected
by AF. No significant alterations in the mRNA expression of proteins and ion channels
were observed in patients with AF of a duration ≤ 6 months.
46
Chapter 3
AF AF AF AF
SR SR SR SR
IVS3B 475 bp
IVS3B 442 bp
IVS3A 335 bp
GAPDH
Figure 1. A, The cDNA ratio L-type Ca2+ (CaLtype)/ GAPDH was significantly reduced (-49%, p=0.01) in
patients with AF of a duration > 6 months compared to the controls in sinus rhythm (SR). Mean values ± SEM.
B, Typical example of an agarose gel showing the results of a semi-quantitative PCR of L-type Ca2+ and GAPDH
of 4 patients with persistent AF at the moment of surgery and 4 controls in sinus rhythm. In the AF group there
was a lower expression the IVS3B form (475 bp), the IVS3B deleted D1 form (442 bp) and the sum of the three
isoforms of the L-type Ca2+. C, No influence of the functional class for exercise tolerance (NYHA) was demon-
strated on the ratio cDNA ratio L-type Ca2+/ GAPDH (mean ± SEM). All data are represented in density units/
density units.
A
B
C
47
Alterations in gene expression of proteins
Table 3.Influence of different drugs influencing the calcium handling on the cDNA ratio L-type Ca2+/
GAPDH (median value (range))
persistent atrial fibrillation sinus rhythm
≤ 6 months > 6 months
Digitalis yes (n=7) 2.2 (0.5-3.1) yes (n=7) 1.4 (0.2-3.2) yes (n=1) 4.1
no (n=2) 0.7 (0.6-0.8) no (n=2) 0.6 (0.3-0.9) no (n=17) 2.3 (0.5-3.3)
Calcium entry blockers yes (n=1) 3.1 yes (n=4) 1.2 (0.3-2.8) yes (n=7) 2.7 (2.4-3.1)
no (n=8) 1.6 (0.5-2.9) no (n=4) 1.3 (0.2-3.2) no (n=11) 2.2 (0.5-4.1)
Beta-blockers yes (n=1) 0.8 yes (n=3) 0.5 (0.2-0.9) yes (n=8) 2.7 (0.5-4.1)
no (n=8) 1.8 (0.5-3.1) no (n=6) 1.6 (0.3-3.2) no (n=10) 2.2 (1.3-3.3)
ACE inibitors yes (n=5) 1.8 (0.8-2.9) yes (n=3) 1.5 (0.9-2.8) yes (n=6) 2.2 (1.3-3.1)
no (n=4) 1.7 (0.5-3.1) no (n=6) 1.1 (0.2-3.2) no (n=12) 2.5 (0.5-4.1)
Figure 2. No differences in cDNA ratios SR Ca2+-ATPase/ GAPDH (panel A), phospholamban (PLB)/GAPDH
(Panel B), and Na+-Ca2+ exchanger/ GAPDH (Panel C) between patients with and without AF were observed.
The cDNA ratio Giα2/ GAPDH was reduced in patients with AF of a duration > 6 months compared to the
controls in sinus rhythm (- 30%, p=0.01, Panel D). All data are represented in density units/ density units and
mean ± SEM.
48
Chapter 3
Figure 3. Plot showing a significant correlation between the increasing amounts of L-type Ca2+ input template
and the ratio L-type Ca2+/GAPDH. The ratios are expressed in a logarithmic way. Each value represents the mean
of five measurements.
Changes in Gene Expression during AF
Several studies suggested that abnormalities in the calcium handling play a pivotal
role in the adaptive processes which occur during persistent AF. Experiments of pacing
induced AF,10,28,29 or rapid atrial pacing30 revealed that calcium lowering by verapamil
reduced both electrical and contractile dysfunction, while calcium loading or the Ca2+
agonist Bay 8644 had an opposite effect.10,28 Furthermore, depletion of contractile material
was demonstrated in atria of goats with AF induced for 9 to 23 weeks.15 Finally, calcium
lowering drugs, when administered in humans with persistent AF, reduced the recurrence
rate of AF after restoration of sinus rhythm.31 Only a few data are available on alterations
in gene expression or in ionic currents occurring in atrial tissue subjected to rapid atrial
rates due to AF or atrial pacing. Yue et al. demonstrated that 6 weeks of rapid atrial pacing
in dogs led to significant changes in ionic currents.23 The induction of atrial electrical
remodeling,32, i.e. a reduced atrial refractoriness together with loss of the physiological
rate adaptation to heart rate, was accompanied by a time-dependent decrease of L-type
Ca2+ current,23 transient outward current,23 and sodium current densities.33 Preliminary data
49
Alterations in gene expression of proteins
revealed that downregulation of the mRNA expression of the these ion channels was
responsible for the reduction of latter currents.34 The lower L-type Ca2+ current density has
been confirmed in humans with persistent AF24 and dilated atria.35,36 Protein expression of
the potassium channel Kv1.5, together with transient outward current and a sustained
outward potassium current densities were reduced in atrial myocardium of patients with
persistent AF.37 In contrast to the data of Yue et al.,23 the present study revealed significant
reductions only if AF had lasted > 6 months. This discrepancy between our data of patients
selected for cardiac surgery and data of the experimental canine model may be related to
(a) the difference in species (human versus dog); and (b) the presence of co-morbidity and
drug treatment in our patients versus the “clean” canine model.
Differences and Similarities between Changes in Atrial Gene Expression during AF
and Ventricular Gene Expression during Heart Failure.
Rapid ventricular pacing induces ventricular failure, and is frequently used as a model
to investigate alterations in gene expression during congestive heart failure in humans.38
Both AF and rapid atrial pacing induce atrial contractile dysfunction.10-14 We hypothesized
that during AF alterations in gene expression in atrial myocardium would be similar to
those demonstrated in ventricular myocardium during (pacing-induced) heart failure.39-44
However, in contrast to what was expected, the present study revealed important
differences between fibrillating atrial and failing ventricular myocardium. In the present
study, the expected reduction of mRNA contents of SR Ca2+-ATPase42,45-50 and
phospholamban,47-49 and upregulation of Na+-Ca2+ exchanger mRNA content46,51,52 could
not be demonstrated. Furthermore, in contrast to the increased level of Giα2 in failing
ventricular myocardium,40,48 a reduced mRNA content of Giα2 was observed in the right
atrial appendage. However, in agreement with most studies investigating the expression
or density of the L-type Ca2+ at the ventricular level, the present study showed a reduced
mRNA content of L-type Ca2+.41-45,52,53 Thus, the adaptation processes in the atria during
AF, as far as transcription is concerned, seem to be different from those occurring at the
ventricular level during heart failure. Furthermore, in contrast to the observed diffusealterations in the failing ventricular myocardium, changes in mRNA expression of proteins
and ion channels involved in the calcium handling occurred selectively in atrial myocardium
during AF. The mechanism behind these differences in genetic reprogramming, which
have to be confirmed in an animal model investigating gene expression in both the atrium
and ventricle, are unknown. However, (1) differences in distribution, number and type of
ion channels and proteins between atrial and ventricular myocardium,54 and (2) differences
in neurohumoral activation during AF and (pacing-induced) heart failure may have
influenced the latter.
50
Chapter 3
Limitations of the Study
The present study has several important limitations. First, only mRNA expression
was determined which neither reflects the protein expression nor its functional activity.
Furthermore, not a quantitative but a semi-quantitative PCR was used. Finally, drugs,
differences in hemodynamic function and underlying diseases may influence gene
expression of proteins and ion channels in humans. To minimize the influence of the above
mentioned clinical parameters on gene expression, no patients with severe left ventricular
dysfunction were included, the patients were matched, and we used multivariate analysis.
Using this statistical analysis, parameters were identified which independently influenced
alterations in the mRNA expression of the investigated proteins and ion channels. This
analysis revealed that in the present population the mRNA expressions were neither affected
by the underlying disease, nor by left atrial or left ventricular dimensions, nor by the
degree of heart failure, nor by any (calcium handling affecting) drug.
References
1. Kannel WB, Abbott RD, Savage DD, et al: Epidemiologic features of chronic atrial fibrillation. N Engl J
Med 1982;306:1018-1022.
2. Önundarson PT, Thorgeirsson G, Jonmundsson E, et al: Chronic atrial fibrillation-epidemiologic features
and 14 years follow-up: a case control study. Eur Heart J 1987;8:521-527.
3. Krahn AD, Manfreda J, Tate RB, et al: The natural history of atrial fibrillation: incidence, risk factors, and
prognosis in the Manitoba follow-up study. Am J Med 1995;98:476-484.
4. Godtfredsen J: Atrial fibrillation. Etiology, course and prognosis. A follow-up study of 1212 cases. Copen-
hagen: University of Copenhagen, 1975. Thesis
5. Bjerkelund C, Orning OM: An evaluation of DC shock treatment of atrial arrhythmias. Acta Med Scand
1968;184:481-491.
6. Van Gelder IC, Crijns HJGM, Tieleman RG, et al: Value and limitation of electrical cardioversion in
patients with chronic atrial fibrillation - importance of arrhythmia risk factors and oral anticoagulation.
Arch Int Med 1996;156:2585-2592.
7. Sopher SM, Camm AJ: Atrial fibrillation - maintenance of sinus rhythm versus rate control. Am J Cardiol
1996;77:24A-38A.
8. Graettinger JS, Carleton RA, Muenster JJ: Circulatory consequences of changes in cardiac rhythm pro-
duced in patients by transthoracic direct-current shock. J Clin Invest 1964;43:2290-2302.
9. Braunwald E: Symposium on cardiac arrhythmias: introduction with comments on the hemodynamic sig-
nificance of atrial systole. Am J Med 1964;37:665-669.
10. Leistad E, Aksnes G, Verburg E, et al: Atrial contractile dysfunction after short-term atrial fibrillation is
reduced by verapamil but increased by BAY K8644. Circulation 1996;93:1747-1754.
11. Manning WJ, Silverman DI, Katz SE, et al: Impaired left atrial mechanical function after cardioversion:
relation to duration of atrial fibrillation. J Am Coll Cardiol 1994;23:1535-1540.
12. Sun H, Gaspo R, Leblanc N, et al: Cellular mechanism of atrial contractile dysfunction caused by sus-
tained atrial tachycardia. Circulation 1998;98:719-727.
13. Shapiro EP, Effron MB, Lima S, et al: Transient atrial dysfunction after conversion of chronic atrial fibril-
lation to sinus rhythm. Am J Cardiol 1988;62:1202-1207.
14. Gallagher MM, Obel OA, Camm AJ: Tachycardia-induced atrial myopathy: an important mechanism in
the pathology of atrial fibrillation? J Cardiovasc Electrophysiol 1997;1065-1074.
15. Ausma J, Wijffels M, ThonJ F, et al: Structural changes of atrial myocardium due to sustained atrial
fibrillation in the goat. Circulation 1997;96:3157-3163.
16. Morillo CA, Klein GJ, Jones DL, et al: Chronic rapid atrial pacing: structural, functional and
electrophysiologic characteristics of a new model of sustained atrial fibrillation. Circulation 1995;91:1588-
1595.
51
Alterations in gene expression of proteins
17. Bailey GW, Braniff BA, Hancock W, et al: Relation of left atrial pathology to atrial fibrillation in mitral
valvular disease. Ann Int Med 1968;69:13-20.
18. Davies MJ, Pomerance A: Pathology of atrial fibrillation in man. Br Heart J 1972;34:520-525.
19. Mary-Rabine L, Pham TD, Hordof A, et al: The relationship of human atrial cellular electrophysiology to
clinical function and ultrastructure. Circ Res 1983;52:188-199.
20. Lee H-C, Clusin WT: Cytosolic calcium staircase in cultured myocardial cells. Circ Res 1987;61:934-939.
21. Schouten VJA, Morad M: Regulation of Ca2+ current in frog ventricular myocytes by the holding poten-
tial, c-AMP and frequency. Pflugers Arch 1989;415:1-11.
22. De Pauw M, Borgers M, Heyndrickx GR: Ultrastructural calcium distribution in cardiac myocytes after
48h of rapid pacing in dogs. Circulation 1996;94:I-604. (Abstract).
23. Yue L, Feng J, Gaspo R, et al: Ionic remodeling underlying action potential changes in a canine model of
atrial fibrillation. Circulation 1997;81:512-525.
24. Van Wagoner DR, Lamorgese M, Kirian P, et al: Calcium current density is reduced in atrial myocytes
isolated from patients in chronic atrial fibrillation. Circulation 1997;96:I-180. (Abstract).
25. Chelly J, Montarras D, Pinset C, et al: Quantitative estimation of minor mRNAs by cDNA polymerase
chain reaction - application to dystrophin mRNA in cultured myogenic and brain cells. Eur J Biochem
1990;187: 691-698.
26. Noonan KE, Beck C, Holzmayer TA, et al: Quantitative analysis of MDR1 (multi drug resistance) gene
expression in human tumors by polymerase chain reaction. Proc Natl Acad Sci USA 1990;87:7160-7164.
27. Cox JL, Jaquiss RD, Schuessler RB, et al: Modification of the maze procedure for atrial flutter and atrial
fibrillation. II. Surgical technique of the maze III procedure. J Thorac Cardiovasc Surg 1995;110:485-495.
28. Goette A, Honeycutt C, Langberg JJ: Electrical remodeling in atrial fibrillation. Time course and mecha-
nisms. Circulation 1996;94:2968-74.
29. Daoud EG, Knight BP, Weiss R, et al: Effect of verapamil and procainamide on atrial fibrillation - induced
electrical remodeling in humans. Circulation 1997;96:1542-1550.
30. Tieleman RG, De Langen CDJ, Van Gelder IC, et al: Verapamil reduces tachycardia induced electrical
remodeling of the atria. Circulation 1997;95:1945-53.
31. Tieleman RG, Van Gelder IC, Crijns HJGM, et al: Early recurrences of atrial fibrillation after electrical
cardioversion: a result of fibrillation-induced electrical remodeling of the atria? J Am Coll Cardiol
1998;31:167-173.
32. Wijffels MCEF, Kirchof CJHJ, Dorland R, et al: Atrial fibrillation begets atrial fibrillation. A study in
awake chronically instrumented goats. Circulation 1995;92:1954-1968.
33. Gaspo R, Bosch RF, Bou-Abboud E, et al: Tachycardia-induced changes in Na+ current in a chronic dog
model of atrial fibrillation. Circ Res 1997;81:1045-1052.
34. Yue L, Wang Z, Gaspo R, et al: The molecular mechanism of ionic remodeling of repolarization in a dog
model of atrial fibrillation. Circulation 1997;96:I-180. (Abstract).
35. Le Grand B, Hatem S, Deroubaix E, et al: Depressed transient outward and calcium currents in dilated
human atria. Cardiovasc Res 1994;28:548-556.
36. Ouadid H, Albat B, Nargeot J: Calcium currents in diseased human cardiac cells. J Cardiovasc Pharmacol
1995;25:282-91.
37. Van Wagoner DR, Pond AL, McCarthy PM, et al: Outward K+ current densities and Kv1.5 expression are
reduced in chronic human atrial fibrillation. Circ Res 1997;80:772-781.
38. Shinbane JS, Wood MA, Jensen N, et al: Tachycardia-induced cardiomyopathy: a review of animal mod-
els and clinical studies. J Am Coll Cardiol 1997;29:709-715.
39. Spinale FG, Tanaka R, Crawford FA, et al: Changes in myocardial blood flow during development of and
recovery from tachycardia-induced cardiomyopathy. Circulation 1992;85:717-729.
40. Vatner DE, Sato N, Galper JB, et al: Physiological and biochemical evidence for coordinate increases in
muscarinic receptors and Gi during pacing-induced heart failure. Circulation 1996;94:102-107.
41. Colston JT, Kumar P, Chambers JP, et al: Altered sarcolemmal calcium channel density and calcium-pump
ATPase activity in tachycardia heart failure. Cell Calcium 1994;16:349-356.
42. Cory CR, Shen H, O’Brien PJ: Compensatory asymmetry in down-regulation and inhibition of the myo-
cardial Ca2+ cycle in congestive heart failure produced in dogs by idiopathic dilated cardiomyopathy and
rapid ventricular pacing. J Mol Cell Cardiol 1994;26:173-184.
43. Hasenfuss G: Alterations of calcium-regulatory proteins in heart failure. Cardiovasc Res 1998;37:279-
289.
44. Mukherjee R, Hewett KW, Walker JD, et al: Changes in L-type calcium channel abundance and function
during transition to pacing-induced congestive heart failure. Cardiovasc Res 1998;37:432-444.
45. Takahashi T, Allen PD, Lacro RV, et al: Expression of dihydropyridine receptor (Ca2+ channel) and
calsequestrin genes in the myocardium of patients with end-stage heart failure. J Clin Invest 1992;90:927-
52
Chapter 3
935.
46. Studer R, Reinecke H, Bilger J, et al: Gene expression of the Na+-Ca2+ exchanger in end-stage human heart
failure. Circ Res 1994;75:443-453.
47. Schwinger RHG, Böhm M, Schmidt U, et al: Unchanged protein levels of SERCA II and phospholamban
but reduced Ca2+ uptake and Ca2+-ATPase activity of sarcoplasmic reticulum from dilated cardiomyopathy
patients compared with patients with nonfailing hearts. Circulation 1995;92:3220-3228.
48. Flesch M, Schwinger RHG, Schnabel P, et al: Sarcoplasmic reticulum calcium ATPase and phospholamban
mRNA and protein levels in end-stage heart failure due to ischemic or dilated cardiomyopathy. J Mol Med
1996;74:321-332.
49. Linck B, Boknik P, Eschenhagen T, et al: Messenger RNA expression and immunological quantification of
phospholamban and SR-Ca2+-ATPase in failing and nonfailing human hearts. Cardiovasc Res 1996;31:625-
632.
50. Movsesian MA, Schwinger RHG: Calcium sequestration by the sarcoplasmic reticulum in heart failure.
Cardiovasc Res 1998;37:352-259.
51. Flesch M, Schwinger RHG, Schiffer F, et al: Evidence for functional relevance of and enhanced expres-
sion of the Na+-Ca2+ exchanger in failing human myocardium. Circulation 1996;94:992-1002.
52. De Tombe PP: Altered contractile function in heart failure. Cardiovasc Res 1998;37:367-380.
53. Dixon IMC, Lee SL, Dhalla NS: Nitrendipine binding in congestive heart failure due to myocardial infarc-
tion. Circ Res 1990;66:782-788.
54. Schwinger RHG, Böhm M, Pieske B, et al: Different $-adrenoceptor-effector coupling in human ventricu-
lar and atrial myocardium. Eur J Clin Invest 1991;21:443-451.
55. Romanin C, Karlsson JO, Schindler H: Activity of L type Ca2+ channels is sensitive to cytoplasmic
calcium. Pflugers Arch 1992;421:516-518.
55
Alterations in potassium channel gene expression in atria
Chapter 4
Alterations in Potassium Channel Gene Expression in
Atria of Patients with Persistent and Paroxysmal Atrial
Fibrillation Differential regulation of protein and mRNA
levels for K+-channels
Bianca J. J. M. Brundel, MSc*,†; Isabelle C. Van Gelder, MD*; Robert H.
Henning, MD†; Anton E. Tuinenburg, MD*; Mirian Wietses†; Jan G. Grandjean,
MD‡; Arthur A. M. Wilde, MD§, Wiek H. Van Gilst, PhD†; Harry J. G. M.
Crijns, MD*.
Departments of Cardiology*, Clinical Pharmacology†, and Thoracic Surgery,
Thoraxcenter University Hospital Groningen and Department of Cardiology
University of Amsterdam and Utrecht§, The Netherlands.
Journal of the American College of Cardiology: Provisionally Accepted
AbstractObjectives: Our propose was to determine whether patients with persistent and
paroxysmal atrial fibrillation (AF) show alterations in potassium channel expression.
Background: Persistent AF is associated with a sustained shortening of the atrial action
potential duration and atrial refractory period. Underlying molecular changes have not
been studied in humans. We investigated whether a changed gene expression of specific
potassium channels is associated with these changes in patients with persistent and
paroxysmal AF.
Methods: Right atrial appendages were obtained from 8 patients with paroxysmal AF, 10
with persistent AF and 18 matched controls in sinus rhythm. All controls underwent CABG
whereas most AF patients underwent Cox’s MAZE surgery (n=12). All patients had normal
left ventricular function. mRNA levels were measured by semi-quantitative polymerase
chain reaction and protein content by Western blotting.
Results: mRNA levels of transient outward channel (Kv4.3), acetylcholine dependent
potassium channel (Kir3.4) and ATP dependent potassium channel (Kir6.2) were reduced
in patients with persistent AF (-35%, -47%, -36%, respectively, p<0.05), whereas only
Kv4.3 mRNA level was reduced in patients with paroxysmal AF (-29%, p=0.03). No
56
Chapter 4
changes were found for Kv1.5 and HERG mRNA levels in both groups. Protein levels of
Kv4.3, Kv1.5 and Kir3.1 were reduced both in patients with persistent (-39%, -84%, -
47%, respectively, p<0.05) and paroxysmal AF (-57%, -64%, -40%, respectively, p<0.05).
Conclusions: Persistent AF is accompanied by reductions in mRNA and protein levels of
several potassium channels. In patients with paroxysmal AF these reductions were observed
predominantly at the protein level and not at the mRNA level suggesting a post-
transcriptional regulation.
Abbreviations
AF atrial fibrillation
SR sinus rhythm
GAPDH glyceraldehyde-3-phosphate dehydrogenase
HERG gene encoding rapid component of the delayed recitifier IKr
Kir3.1 gene encoding part of the IKACH
, together with IKir3.4
Kir3.4 gene encoding part of the IKACh
, together with IKir3.1
Kir6.2 gene encoding part of the IKATP
Kv1.5 gene encoding ultra rapid component of the delayed rectifier IKur
Kv4.3 gene underlying calcium independent transient outward current ITo1
Introduction
Atrial fibrillation (AF) is a common cardiac arrhythmia affecting millions of people
worldwide (1). AF has the tendency to become more persistent and increasingly difficult
to treat over time. During recent years, experimental studies showed that shortening of the
atrial effective refractory period was one important factor contributing to the maintenance
of AF (2,3). This shortening has been confirmed in patients suffering from AF and atrial
flutter (4,5). Experimental and human data revealed that AF or tachycardia induced
shortening of atrial effective refractory period and action potential duration were associated
with a reduction of ICaL
, ITo1
and INa
currents due to reduced mRNA expression of these
channels (6-9). Previously we have demonstrated that mRNA and protein expression of
the L-type calcium channel in both patients with persistent AF with more severe underlying
heart disease (10) and the present patient population (11) were significantly reduced. No
alterations, however, were observed in either patients with paroxysmal or short term
persistent AF.
Theoretically, action potential duration and atrial effective refractory period can be
shortened by 1) an increase in K+ channel gene expression and activity, or 2) a decrease in
L-type Ca2+ channel (L-type Ca2+) gene expression and activity, or 3) a combination of
both. The present study was undertaken to evaluate the impact of persistent and paroxysmal
AF on gene expression of potassium channels in human right atrial appendages. Therefore,
57
Alterations in potassium channel gene expression in atria
the mRNA and protein expression of Kv4.3 (gene underlying the calcium independent
transient outward current ITo1
) (12), HERG (gene encoding the rapid component of the
delayed rectifier) (13), Kv1.5 (gene encoding the ultra rapid delayed rectifier, IKur
) (14,15),
Kir3.1/Kir3.4 (heterotetrameric complex of these two genes forms the acetylcholine
dependent K+ current, IKACh
) (16) and Kir6.2 (gene encoding the inward rectifier K+ current,
forming IKATP
with the sulfonylurea receptor) (17) were examined in patients with persistent
and paroxysmal AF undergoing cardiac surgery. Patients with lone AF or patients with AF
scheduled for coronary artery bypass grafting were matched with patients in sinus rhythm
without history of AF and undergoing coronary artery bypass grafting.
Materials and Methods
Patient selection and atrial tissue collecting
The day before surgery, one investigator (AET) assessed the clinical characteristics
of the patient. Patient’s history and previouw electrocardiograms were used to establish
type and duration of AF. In addition, medication use and exercise tolerance (according to
the NYHA classification) was determined. Echocardiography data were obtained within 3
months prior to surgery. Right atrial appendages were obtained from 10 patients with
persistent AF and from 8 patients with paroxysmal AF. All patients were euthyroid. The
AF patients were matched for age, sex and degree of heart failure with 18 clinically stable
patients in sinus rhythm undergoing CABG. The Institutional Review Board approved the
study and all patients gave written informed consent. Immediately after excision, the right
atrial appendages were snap-frozen in liquid nitrogen and stored at -85 °C.
RNA isolation and cDNA synthesis
Total RNA was isolated and processed as described previously (11). Briefly, first
strand cDNA was synthesized by incubation of 1 µg of total RNA in reverse transcription
10x buffer, 200 ng of random hexamers with 200 units of Moloney Murine Leukemia
Virus Reverse Transcriptase, 1mM of each dNTP and 1 unit of RNase inhibitor (Promega,
The Netherlands) in 20 µl. Synthesis reaction was performed for 10 minutes at 20 °C, 20
minutes at 42 °C, 5 minutes at 99 °C and 5 minutes at 4 °C. All the products were checked
for contaminating DNA.
Semi quantitative PCR analyses
We described and validated these methods before (11). In short, the cDNA of interest
and the cDNA of the ubiquitously expressed housekeeping gene glyceraldehyde-3-
phosphate dehydrogenase (GAPDH) were coamplified in a single PCR. Primers
(Eurogentec, Belgium) were designed for Kv4.3, HERG, Kv1.5, Kir3.4, Kir6.2 and the
housekeeping gene GAPDH (Table 1).
58
Chapter 4
The PCR products were separated on agarose gel by electrophoresis and stained with
ethidium bromide. The density of the PCR products was quantified by densitometry.
Linearity for the PCR was established by making a correlation between the number of
cycles and the density of gene of interest and GAPDH (data not shown).
Protein Preparation and Western Blotting
Frozen right atrial appendages of 5 patients in sinus rhythm, 5 patients with paroxysmal
AF and 5 patients with persistent AF were homogenized in RIPA buffer as previously
described (11). Protein concentration was determined according to the Bradford method
(Sigma, The Netherlands) with bovine albumin as a standard. Protein expression was
determined by Western-blot analysis and expressed as the ratio to levels of GAPDH.
Therefore, denatured protein (10 µg) was separated by SDS-PAGE, transferred to
nitrocellulose membranes (Stratagene, The Netherlands) and incubated with primary
antibodies against GAPDH (Affinity Reagents, USA), anti Kir3.1, anti Kv4.3 and anti
Kv1.5 (Alomone Labs, Israel). Anti-mouse IgG (Santa Cruz Biotechnology, The
Netherlands) was used as secondary antibody. Signals were detected by the ECL-detection
method (Amersham, The Netherlands) and quantified by densitometry. The specificity of
the band was tested by pre-incubation of the antibody with the antigen. The band densities
were evaluated by densitometric scanning using a Snap Scan 600 (Agfa, The Netherlands).
There was a linear relation between protein amounts on the membrane and immunoreactive
signals of Kir3.1, Kv4.3, Kv1.5 and GAPDH (data not shown).
Table 1. The sequence for the primers.
protein sequence
cycles annealing
temp (°C)
GAPDH: F 5'-CCC ATC ACC ATC TTC CAG GAG CG-3', var. var.
R 5'-GGC AGG GAT GAT GTT CTG GAG AGC C-3'
To1: F 5’-CAG CAA GTT CAC AAG CAT CC-3’ 31 52
Kv4.3 R 5’-AGC TGG CAG GTT AGA ATT GG-3’
Kr: F 5’-GTC AAT GCC AAC GAG GAG GT-3’, 27 58
HERG R 5’-CTG GTG GAA GCG GAT GAA CT-3’
Kur: F 5’-AAC GAG TCC CAG CGC CAG GT-3’ 32 64
Kv1.5 R 5’-AGG CGG ATG ACT CGG AGG AT-3’
KACh F 5’-CAC CCT GGT GGA CCT CAA GTG GCG C-3’ 30 56
Kir3.4 R 5’-AGC TCC GGG CTT GGC AGG TCA TGC-3’
KATP F 5’-CAC GCT GGT GGA CCT CAA GT-3’ 29 59
Kir6.2 R 5’-CAC GAT GAG GCT CAG GAT GG-3’
59
Alterations in potassium channel gene expression in atria
Definitions
Persistent AF: continuous presence of AF until the moment of cardiac surgery, i.e. at least
two consecutive electrocardiograms of AF more than 1 week apart, without intercurrent
sinus rhythm. Persistent AF has a non-spontaneously converting character. Previously,
this type of AF was classified as chronic AF (18).
Paroxysmal AF: AF typically occurring in episodes with a duration shorther than 24
hours (but longer lasting paroxysms are not unusual) with intermittent sinus rhythm.
Paroxysmal AF is either converting spontaneously or terminated with intravenously
administered antiarrhythmic drug. It is non-controlled whether paroxysmal AF is present
at the moment of cardiac surgery (18).
Statistical Analysis
All PCR and SDS-PAGE procedures were performed in duplicate series and mean
values were used for statistical analysis. For determination of correlation’s the Spearman
correlation test was used. One-way ANOVA was used for all group to group comparisons.
All p-values are two-sided, a p-value <0.05 was considered statistically significant. SAS
version 6.12 (Cary, NC) was used for all statistical evaluations.
Results
Patients
Included were 10 patients with persistent and 8 patients with paroxysmal AF. These
two groups were compared with two groups of controls in sinus rhythm, which were matched
for sex, age and left ventricular function (Table 2). Six of the 8 patients with paroxysmal
AF suffered from intractable paroxysmal AF without any underlying heart disease and
were scheduled for Cox’s MAZE surgery. The median duration of sinus rhythm before
surgery was 1.5 days. The median frequency of paroxysms was once a day with a median
duration of 3 hours. Three patients with paroxysmal AF were in AF at the moment of
surgery and harvesting of the right atrial appendage. Control right atrial appendages were
obtained from clinically stable patients in sinus rhythm who were scheduled for coronary
artery bypass surgery. Although the AF groups and their controls in sinus rhythm differed
with respect to the underlying heart disease, all had a normal left ventricular function and
were in the functional class I or II for exercise tolerance. Also, atrial and left ventricular
dimensions were similar among groups (data not shown).
Alterations in mRNA Levels in Persistent and Paroxysmal AF
Changes in transcription of the gene of interest were determined by comparison of
gene-of-interest/GAPDH ratios between patients with persistent AF or paroxysmal AF
60
Chapter 4
and their matched controls in sinus rhythm. No differences in GAPDH densities were
found between the groups (data not shown).
Patients with persistent AF showed significant reductions of mRNA contents for Kv4.3
(-35%, p=0.02), Kir3.4 (-47%, p=0.0003) and Kir6.2 (-36%, p=0.03) (Table 3). Patients
with paroxysmal AF showed only reduction of the Kv4.3 mRNA level (-29%, p=0.03,
Table 3). No differences in mRNA contents for Kv1.5 and HERG were found between
patients with persistent and paroxysmal AF compared to patients in sinus rhythm (Table
3). Although the group samples are small, the mRNA levels of Kv4.3, Kv1.5 and Kir3.4 in
both patients with persistent and paroxysmal AF seemed not to be influenced by any drug
(data not shown).
Table 2. Baseline characteristics of patients with paroxysmal AF, persistent AF and matched control patients
in sinus rhythm of both groups at the moment of surgery.
PAF SR (PAF) CAF SR (CAF)
8 8 10 10
M/F (n) 6/2 6/2 6/4 6/4
Age 51 ±7 56 ± 11 63 ±11 65 ± 17
Duration of AF: median, range (months) 18 (8 - 64)
Duration SR before surgery: median, range (days) 1.5 (0-30)
Underlying heart disease (n)
coronary artery disease 2* 8 3* 10
hypertension 1 1 3 2
lone AF 6* 0 5* 0
Surgical procedure
CABG 2* 8 4* 10
MAZE 6* 0 6* 0
New York Heart Association for exercise tolerance
Class I 7 5 6 5
Class II 1 3 4 5
Medication
Beta blockers 1* 5 3 6
Calcium antagonist 0 3 3 3
Digitalis 0 1 5 3
ACE inhibitor 0 1 4 2
* p-value < 0.05 compared to the control group
Values are presented as mean value ± SD or number of patients. ACE, inhibitor indicates Angiotensin
Converting Enzyme; AF, atrial fibrillation; CABG, Coronary Artery Bypass Grafting; CAF, chronic persistent
atrial fibrillation; M/F, male/female; PAF, paroxysmal atrial fibrillation; SR (CAF), matched controls in
sinus rhythm of patients with persistent AF; SR (PAF), matched controls in sinus rhythm of patients with
paroxysmal AF
61
Alterations in potassium channel gene expression in atria
Figure 1. The top of each panel shows a typical Western blot analysis of 10 µg of protein homogenates of 3
patients in sinus rhythm (SR), 3 patients with chronic persistent AF (CAF) and 3 patients with paroxysmal AF
(PAF). The immunoblot swere done for anti-Kv4.3 (A), anti-Kv1.5 (B), and anti-Kir3.1 (C) with GAPDH (37
kD) as an internal control.
All data are presented as density units/ density units. Values are mean ± SEM.
Alterations in Proteins Levels in Persistent and Paroxysmal AF
From the total patient group there were 5 patients with persistent AF, 5 with paroxysmal
AF and 5 patients in sinus rhythm with enough right atrial appendage tissue to isolate
proteins. Changes in protein expression were studied for Kv4.3, Kv1.5 and Kir3.1 in rela-
tion to protein levels of GAPDH. The protein expression of Kv1.5/GAPDH and Kir3.1/
62
Chapter 4
Table 3. Comparison of mRNA and protein expression for patients with persistent and paroxysmal AF with
their matched controls in sinus rhythm.
mRNA expression Correlation
mRNA and Protein
SR1 CAF SR2 PAF r
Kv4.3 1.12 ± 0.1 0.73 ± 0.09* 1.08 ± 0.11 0.77 ± 0.1* 0.75**
Kv1.5 1.35 ± 0.12 1.28 ± 0.11 1.37 ±0.14 1.56 ± 0.14 0.31
HERG 0.43 ± 0.04 0.42 ± 0.03 0.52 ± 0.04 0.62 ± 0.03 n.a.
Kir3.4/Kir3.1 1.9 ± 0.1 1.03 ± 0.1** 1.92 ± 0.11 1.59 ± 0.13 0.74**
Kir6.2 1.04 ± 0.07 0.67 ± 0.07* 1.1 ± 0.07 0.96 ± 0.07 n.a.
*, p<0.05, **, p<0.005
Figure 2. Correlation between the mean protein expression of Kv4.3 ( , r=0.67, p>0.05), Kv1.5 ( , r=0.97,
p=0.02) and Kir3.1 ( , r=0.56, p>0.05) in patients with paroxysmal AF and the duration of sinus rhythm before
surgery.
GAPDH was markedly reduced in patients with persistent AF compared to patients in
sinus rhythm (-84%, p=0.001 and -47%, p=0.002, respectively) and also in patients with
paroxysmal AF (-64%, p=0.005 and -40%, p=0.007, respectively, Figure 1B and C). Simi-
lar results were obtained for Kv4.3/GAPDH protein content, i.e. both a reduction in pa-
tients with persistent AF (-39%, p=0.04) and paroxysmal AF (-57%, p=0.001, Figure 1A).
A positive correlation between mRNA levels and protein levels of Kv4.3 and Kir3.1 for
patients with paroxysmal AF, persistent AF and sinus rhythm could be demonstrated, but
not for Kv1.5 (Table 3). Although the group samples are small the protein ratio of Kv4.3,
Kv1.5 and Kir3.1 seemed not to be influenced by any drug (data not shown).
63
Alterations in potassium channel gene expression in atria
Importantly, in patients with paroxysmal AF the mean protein expression of Kv1.5
appeared to be related to the duration of sinus rhythm after the last episode of AF. Patients
in AF at the moment of surgery showed the lowest protein expression, comparable to
patients with persistent AF. Patients in sinus rhythm at the moment of surgery showed the
highest protein expression (Figure 2).
Discussion
Both experimental (2,3,19,20) and human (4,5,21,22) AF is accompanied by shortening
of the action potential duration and effective refractory period. This shortening can be
mediated by either an increase in K+ channel gene products and/or activity, or a decrease
in L-type Ca2+ channel gene products and/or activity. Previously, we demonstrated a reduced
mRNA and protein expression of the L-type calcium channel in patients with longstanding
AF, but not in paroxysmal AF (10,11). The present study shows that in patients with long-
standing persistent AF, the mRNA and protein expression of almost all investigated
potassium channel genes were reduced. In paroxysmal AF patients, reduction in mRNA
levels was confined to Kv4.3, whereas the investigated protein levels (Kv4.3, Kv1.5 and
Kir3.1) were all importantly decreased. Finally, there was a significant positive correlation
between the duration of sinus rhythm after the last episode of paroxysmal AF and content
of protein expression of Kv1.5, suggesting a protective effect of high protein contents or
normalization of protein content after a longer duration of sinus rhythm.
Differences in mRNA and protein expression
We determined mRNA and protein levels of genes encoding a number of potassium
channels. Unfortunately, no antibodies against all the potassium channels have yet been
generated. Therefore, we could only study protein expression of Kv4.3, Kv1.5 and Kir
3.1. Nevertheless, this study reports profound changes in protein expression in both
persistent and paroxysmal AF. In contrast, reduction of mRNA contents seems almost an
exclusive feature for persistent AF. The observed reduction in Kv4.3 mRNA expression
(gene encoding the calcium independent transient outward current) in patients with both
persistent and paroxysmal AF is in accordance with experimental and human studies (3,8).
In dogs subjected to rapid atrial pacing (400 bpm) the transient outward current was reduced
by 70% after 6 weeks (3) with a concomitant reduction of mRNA and protein expression
(6).
In the heart, Kir3.1 and Kir3.4 gene products appear to be responsible for the
acetylcholine-activated K+ current (16), representing an important atrial inwardly rectifying
current. Activation of this channel, e.g. by vagal stimulation, shortens the action potential
duration and refractory period. The Kir3.4 gene was used for mRNA expression
determination and a reduction in Kir3.4 mRNA expression was found in patients with
64
Chapter 4
persistent AF. For Western blotting Kir3.1 was analyzed. A reduction in protein level was
observed in patients with both persistent and paroxysmal AF, which may occur to protect
the cell against further shortening of the action potential duration during AF. The
downregulation observed in our study is, however, in contrast to findings by others on the
electrophysiological level. In a comparable group of patients with persistent AF, an increase
in inward rectifying currents (IK1
and IKACh
) was measured in isolated myocardial cells
(23). This apparent inconsistency between protein level and current density can only be
explained by assuming a change in single channel properties in patients with persistent
AF, such as an increase of mean open-time, an increase in channel conductance or a change
in voltage dependency.
The reduction of Kir6.2 mRNA levels in patients with persistent AF may be related to
depletion of ATP by an increase in metabolic demand during AF. This depletion of ATP
could promote opening of Kir6.2 leading to enhanced repolarization (24) and subsequently
increased expression of this channel (25). When activation of Kir6.2 continues the myocyte
may eventually respond by reducing the gene expression of this channel. There is still
uncertainty whether atrial ischemia indeed plays a role in triggering electrical remodeling
by AF. First, in a canine model White et al. demonstrated that induced AF immediately
caused increase in coronary atrial perfusion and oxygen consumption of atrial myocardium,
but without induction of ischemia (26). On the other hand a progressive increase in metabolic
demand during persistent AF may lead to repeated episodes of atrial ischemia, contributing
to activation of the ATP dependent potassium channel. The latter is suggested by results of
Ausma et al. who demonstrated similarities between cellular structural changes induced
by AF and those seen in hibernating myocardium (27).
The observed reduction in protein expression of Kv1.5 in patients with persistent and
paroxysmal AF could be due to post-transcriptional changes, since at the mRNA level no
changes were found between the groups. The reduction in protein expression is in agreement
with the previous data in patients with persistent AF of Van Wagoner et al. (8). However,
in a canine model of the group of Nattel no changes could be found in the current density
of IKur
(3). It should be pointed out that the molecular species underlying canine I
Kur, probably
Kv3.1, is likely different than that underlying human atrial IKur
, Kv1.5 (28).
No changes in mRNA expression were found for the HERG gene, the gene encoding
the rapid component of the delayed rectifier. This is in accordance with data of Yue and
coworkers (3) and suggests that the HERG gene is less involved in repolarization at the
atrial level during AF.
Finally, we observed a positive correlation between the duration of sinus rhythm
before surgery and the protein levels of Kv1.5 in patients with paroxysmal AF; patients in
AF at the moment of surgery had lower protein levels compared with patients in sinus
rhythm. This finding may suggest that alterations in protein expression and possibly also
65
Alterations in potassium channel gene expression in atria
structural changes occur early (most paroxysms lasted shorter than 24 hours) and could be
reversible.
Underlying mechanisms
The observed reduction in gene expression of three potassium currents clearly can
not explain the observed shortening of effective refractory period and action potential
duration. One may hypothesize that reduction in potassium channels gene expression serves
as an adaptation mechanism to prolong the initially reduced atrial effective refractory
period and action potential duration.
The observed discrepancy between alterations in mRNA and protein expression in
patients with paroxysmal AF may suggest the influence of a different compensatory
mechanism. We hypothesize that reduction in protein channels occurs due to calcium
overload (20,29) and structural changes, including atrophy (27,30), in atrial tissue during
AF by an increased expression of proteolytic enzymes (31). An increased expression of
the proteolytic system is observed in heart tissue during atrophy, calcium overload and
stunning (32-36). Increased protein degradation in muscle atrophy and calcium overload
seemed predominantly induced by activation of a non-lysosomal ATP dependent proteolytic
process. Medina et al. showed that the ubiquitin-proteasome dependent pathway, a highly
conserved pathway consisting of ubiquitin, ubiquitin-conjugating enzymes, deubiquitinases
and proteasome, is activated in atrophying muscles of the heart during starvation (31).
Another common cytosolic proteinase regulating pathway in eukaryotes is the calcium-
dependent pathway, which consists of a diverse group of calcium-dependent cysteine
proteinases (calpains in vertebrate tissues) (37). The increase in cytosolic calcium (29,38)
during AF, could be an important activator of this calcium-dependent pathway by promoting
activation of neutral proteases such as calpains which, once accomplished, leads to
proteolysis of numerous cytoskeletal, membrane-associated and regulatory proteins (32-
35) leading to degeneration of the myocardial cell.
Limitations of the Study
Drugs and differences in underlying diseases may influence gene expression of ion
channels. In this study, to minimize the influence of particular clinical parameters on gene
expression, we included only patients with a normal left ventricular function and, when
possible, drugs were discontinued before surgery.
Because of the limited amount of tissue available, no matched controlled analysis could
be performed for determination of protein levels. However, no significant changes in mRNA
levels between the various control groups of patients in sinus rhythm were observed.
Therefore, in our opinion, a comparison between persistent AF, paroxysmal AF and sinus
rhythm patients seems to be justified.
66
Chapter 4
The paroxysmal AF patients, included in this study, represent patients who were
difficult to treat and underwent predominantly MAZE surgery. Furthermore, it should be
noted that in all groups the number of patients was small. Therefore, the present data can
not be extrapolated uncritically to all (paroxysmal) AF patients.
Acknowledgments
Dr. Van Gelder was supported by Grant 94.014 of the Netherlands Heart Foundation, The
Hague, The Netherlands. The study was supported by Grant 96.051 of The Netherlands
Heart Foundation, The Hague, The Netherlands.
References
1. Kannel WB, Abbott RD, Savage DD, McNamara PM. Epidemiologic features of chronic atrial fibrillation:
the Framingham study. N Engl J Med 1982;306:1018-22.
2. Wijffels MC, Kirchhof CJ, Dorland R, Allessie MA. Atrial fibrillation begets atrial fibrillation. A study in
awake chronically instrumented goats. Circulation 1995;92:1954-68.
3. Yue L, Feng J, Gaspo R, Li GR, Wang Z, Nattel S. Ionic remodeling underlying action potential changes in
a canine model of atrial fibrillation. Circ Res 1997;81:512-25.
4. Franz MR, Karasik PL, Li C, Moubarak J, Chavez M. Electrical remodeling of the human atrium: similar
effects in patients with chronic atrial fibrillation and atrial flutter. J Am Coll Cardiol 1997;30:1785-92.
5. Pandozi C, Bianconi L, Villani M, et al. Electrophysiological characteristics of the human atria after
cardioversion of persistent atrial fibrillation. Circulation 1998;98:2860-5.
6. Yue L, Melnyk P, Gaspo, Wang Z, Nattel S. Molecular mehanisms underlying ionic remodeling in a dog
model of atrial fibrillation. Circ Res 1999;84:776-84.
7. Gaspo R, Bosch RF, Bou-Abboud E, Nattel S. Tachycardia-induced changes in Na+ current in a chronic
dog model of atrial fibrillation. Circ Res 1997;81:1045-52.
8. Van Wagoner DR, Pond AL, McCarthy PM, Trimmer JS, Nerbonne JM. Outward K+ current densities and
Kv1.5 expression are reduced in chronic human atrial fibrillation. Circ Res 1997;80:1-10.
9. Van Wagoner DR, Pond AL, Lamorgese M, Rossie SS, McCarthy PM, Nerbonne JM. Atrial L-Type Ca2+
Currents and Human Atrial Fibrillation. Circ Res 1999;85:428-36.
10. Van Gelder IC, Brundel BJJM, Henning RH, et al. Alterations in gene expression of proteins involved in
the calcium handling in patients with atrial fibrillation. J Cardiovasc Electrophysiol 1999;10:552-60.
11. Brundel BJJM, Van Gelder IC, Henning RH, et al. Gene expression of proteins influencing the calcium
homeostasis in patients with persistent and paroxysmal atrial fibrillation. Cardiovasc Res 1999;42:443-54.
12. Dixon JE, Shi W, Wang HS, et al. Role of the Kv4.3 K+ channel in ventricular muscle. A molecular corre-
late for the transient outward current [published erratum appeared in Circ Res 1997 Jan;80(1):147]. Circ
Res 1996;79:659-68.
13. Curran ME, Splawski I, Timothy KW, Vincent GM, Green ED, Keating MT. A molecular basis for cardiac
arrhythmia: HERG mutations cause long QT syndrome. Cell 1995;80:795-803.
14. Wang Z, Fermini B, Nattel S. Sustained depolarization-induced outward current in human atrial myocytes:
evidence for a novel delayed rectifier K+ current similar to Kv1.5 cloned channel currents. Circ Res
1993;73:1061-76.
15. Fedida D, Wible B, Wang Z, et al. Identity of a novel delayed rectifier current from human heart with a
cloned K+ channel current. Circ Res 1993;73:210-6.
16. Krapivinsky G, Gordon EA, Wickman K, Velimirovic B, Krapivinsky L, Clapham DE. The G-protein-
gated atrial K+ channel IKACh
is a heteromultimer of two inwardly rectifying K(+)-channel proteins. Nature
1995;374:135-41.
17. Inagaki N, Gonoi T, Clement IV J, et al. Reconstitution of IKATP
: An inward rectifier subunit plus the sulfo-
nylurea receptor. Science 1995;270:1166-70.
18. Gallagher MM, Camm AJ. Classification of atrial fibrillation. Pacing Clin Electrophysiol 1997;20:1603-5.
19. Tieleman RG, De Langen CDJ, Van Gelder IC, et al. Verapamil reduces tachycardia-induced electrical
remodeling of the atria. Circulation 1997;95:1945-53.
67
Alterations in potassium channel gene expression in atria
20. Goette A, Honeycutt C, Langberg JJ. Electrical remodeling in atrial fibrillation. Time course and mecha-
nisms. Circulation 1996;94:2968-74.
21. Daoud EG, Knight BP, Weiss R, et al. Effect of verapamil and procainamide on atrial fibrillation-induced
electrical remodeling in humans. Circulation 1997;96:1542-50.
22. Yu WC, Chen SA, Lee SH, et al. Tachycardia-induced change of atrial refractory period in humans. Rate
dependency and effects of antiarrhytmic drugs. Circulation 1998;97:2331-7.
23. Bosch RF, Zeng X., Grammer JB, Popovic K, ewis C, ühlkamp V. Ionic mechanisms of electrical remodel-
ing in human atrial fibrillation. Cardiovasc Res 1999;44:121-31.
24. Shaw RM, Rudy Y. Electrophysiologic effects of acute myocardial ischemia. A theoretical study of altered
cell excitability and action potential duration. Cardiovasc Res 1997;35:256-72.
25. Akao M, Otani H, Horie M, et al. Myocardial ischemia induces differential regulation of KATP
channel gene
expression in rat hearts. J Clin Invest 1997;100:3053-9.
26. White C, Holida M, Marcus M. Effects of acute atrial fibrillation on the vasodilator reserve of the canine
atrium. Cardiovasc Res 1986;20:683-9.
27. Ausma J, Wijffels M, Thone F, Wouters L, Allessie M, Borgers M. Structural changes of atrial myocardium
due to sustained atrial fibrillation in the goat. Circulation 1997;96:3157-63.
28. Yue L, Feng J, Wang Z, Nattel S. Adrenergic control of the ultrarapid delayed rectifier current in canine
atrial myocytes. Journal of Physiology 1999;516:385-98.
29. Sun H, Leblanc N, Nattel S. Effects of atrial tachycardia on intracellular Ca2+ and cellular contractility.
Circulation 1999;100:I-200
30. Morillo CA, Klein GJ, Jones D, Guiraudom CM. Chronic rapid atrial pacing. Structural, functional, and
electrophysiological characteristics of a new model of sustained atrial fibrillation. Circulation 1995;91:1588-
95.
31. Medina R, Wing SS, Goldberg AL. Increase in levels of polyubiquitin and proteasome mRNA in skeletal
muscle during starvation and denervation atrophy. Biochem J 1995;307:631-7.
32. Bartus RT, Elliott PJ, Hayward NJ, et al. Calpain as a novel target for treating acute neurodegenerative
disorders. Neurol Res 1995;17:249-58.
33. Atsma DE, Bastiaanse EM, Jerzewski A, Van Der Valk LJ, Van Der Laarse A. Role of calcium-activated
neutral protease (calpain) in cell death in cultured neonatal rat cardiomyocytes during metabolic inhibition.
Circ Res 1995;76:1071-8.
34. Gorza L, Menabo R, Di Lisa F, Vitadello M. Troponin T cross-linking in human apoptotic cardiomyocytes.
Am J Pathol 1997;150:2087-97.
35. Gorza L, Menabo R, Vitadello M, Bergamini CM, Di Lisa F. Cardiomyocyte troponin T immunoreactivity
is modified by cross-linking resulting from intracellular calcium overload. Circulation 1996;93:1896-904.
36. Gao WD, Atar D, Liu Y, Perez NG, Murphy AM, Marban E. Role of troponin I proteolysis in the pathogen-
esis of stunned myocardium. Circ Res 1997;80:393-9.
37. Mykles DL. Intracellular proteinases of invertebrates: calcium-dependent and proteasome/ubiquitin-de-
pendent systems. International Review of cytology 1998;184:157-289.
38. Ausma J, Dispersyn GD, Duimel H, et al. Changes in ultrastructural calcium distribution in goat atria
during atrial fibrillation. J Mol Cell Cardiology 2000;32:355-64.
69
Ion channel remodeling is related to intra operative atrial refractory periods in patients
Chapter 5
Ion Channel Remodeling is Related to Intra- Operative
Atrial Effective Refractory Periods in Patients with
Paroxysmal and Persistent Atrial Fibrillation
Bianca J. J. M. Brundel, MSc1,2; Isabelle C. Van Gelder, MD1;
Robert H. Henning, MD2; Robert G. Tieleman, MD1; Anton E. Tuinenburg,
MD1; Mirian Wietses2; Jan G. Grandjean, MD3, Wiek H. Van Gilst, PhD2;
Harry J. G. M. Crijns, MD1.
Departments of Cardiology1, Clinical Pharmacology2 and Thoracic Surgery3,
Thoraxcenter University Hospital Groningen, The Netherlands.
Circulation: in press
Abstract
Background Sustained shortening of the atrial effective refractory period (AERP),
probably due to reduction in L-type calcium current, is a major factor in the initiation and
maintenance of AF. We investigated underlying molecular changes by studying the rela-
tion between gene expression of the L-type calcium channel and potassium channels and
the AERP in patients with AF. Methods and Results mRNA and protein expression were
determined in left and right atrial appendages of 13 patients with paroxysmal AF, 16 with
persistent AF and 13 controls in sinus rhythm, by RT-PCR and slot-blot, respectively. The
mRNA content of almost all investigated ion-channel genes was reduced in persistent but
not in paroxysmal AF. Protein levels for L-type Ca2+ channel and five potassium channels
(Kv4.3, Kv1.5, HERG, minK and Kir3.1) were significantly reduced in both persistent
and paroxysmal AF. Furthermore, AERPs were determined intra-operatively with pro-
grammed electrical stimulation at 5 basic cycle lengths (BCLs) (between 250 and 600
ms). Patients with persistent and paroxysmal AF displayed significant shorter AERPs.
Protein levels of all ion-channels investigated correlated positively with the AERP (at
BCL: 600, 500, 400 and 300 ms) and with the rate adaptation of AERP. Patients with
reduced ion-channel protein expression revealed shorter AERP duration and poorer rate
adaptation.
70
Chapter 5
Conclusions AF is predominantly accompanied by decreased protein contents of the L-
type Ca2+ channel and several potassium channels. Reductions in L-type Ca2+ channel
correlated with AERP and rate adaptation, and represent a probable explanation for the
electrophysiological changes during AF.
Introduction
Atrial fibrillation (AF) is a common arrhythmia affecting millions of people world-
wide.1 AF has the tendency to become more persistent and increasingly difficult to treat
over time. During recent years, experimental and human studies showed that rapid short-
ening of the atrial effective refractory period (AERP) is an important factor contributing
to the maintenance of AF.2-6 Rapid shortening of the AERP in AF involves functional
changes in ion channels. Animal experimental data revealed that the L-type Ca2+ channel
plays a main role in shortening of AERP and action potential duration (APD).7,8 These
observations are supported by blocking of AERP shortening with the L-type Ca2+ antago-
nist, verapamil, in other experimental studies.9,10 In addition, human data on AF have dem-
onstrated reductions in ICaL
11,12 and gene expression of L-type Ca2+ channel.13
However, shortening of AERP could also be explained by an increase in (repolariz-
ing) K+ channel activity. Indeed, one study found increased IKACh
and IK1
in isolated human
atrial cells of patients with persistent AF.11 In contrast, other studies support a decrease in
K+ channels in AF. In human atrial myocytes reductions in ITo
and IKsus
and a reduced gene
expression of Kv1.5, Kv4.3, Kir3.1, Kir3.4 and Kir6.213-15 were found.
Until now, the relationship between changes in AERP and ion channel gene expres-
sion has not been investigated in human tissue of patients with AF. The aim of the present
study was to investigate the regulation of L-type Ca2+ channel and K+ channels and its
relation to AERP in patients with persistent and paroxysmal AF. We included both patients
with lone AF and with mitral valve disease (MVD), since the occurrence of MVD seems
to prolong the AERP.16,17
Materials and Methods
Patients and atrial tissue collecting
Prior to surgery, one investigator assessed the clinical characteristics of patients
(Table 1). The patient’s arrhythmia history was classified according to Gallagher.18 The
persistent and paroxysmal AF group contained patients with lone AF or AF with underly-
ing MVD. All patients underwent MAZE surgery, were euthyroid and had normal left
ventricular function. Coumarin therapy was interrupted 3 days before surgery and class I
and III anti-arrhythmic drugs were discontinued for at least 5 half-times. During surgery
the AERPs were determined with use of temporary epicardial pacing leads. AERPs were
measured intra-operatively at 5 different basic cycle lengths (BCL; 600, 500, 400, 300 and
71
Ion channel remodeling is related to intra operative atrial refractory periods in patients
250 ms) at the right atrial appendage (RAA) and left atrial appendage (LAA) using pro-
grammed electrical stimulation.
LAAs and RAAs were obtained except for the control patients undergoing CABG
from whom only the RAA was gathered. After excision, the RAAs and LAAs were imme-
diately snap-frozen in liquid nitrogen and stored at -85 °C. The study was approved by the
Institutional Review Board and written informed consent was given by all patients.
Table 1. Baseline characteristics of patients with lone paroxysmal AF, lone persistent AF and control patients
in sinus rhythm
Lone AF AF with MVD
SR (CABG) PAF CAF SR(MVD) PAF CAF
N 9 7 7 4 6 9
Age 61±8 48±7 50±7 60±9 47±9 56±10
Previous duration of AF
(median, range (months)) - - 13.6 (0.1-56) - - 8(0.4-32)
Duration SR before surgery
(median, range (days)) - 2 (0.5-12) - - 75(10-210) -
Underlying heart disease (n)
and /surgical procedure
Coronary artery disease/CABG 9 0 0 0 0 0
Lone AF / MAZE 0 7 7 0 0 0
MVD /MV replacement/repair 0 0 0 4 6 9
New York Heart Association for
exercise tolerance
Class I 9 6 4 1 0 0
Class II 0 1 3 3 3 4
Class III 0 0 0 0 3 5
Echocardiography
Left atrial diameter (parasternal) 36± 5 39± 5 45± 8 47±5 45±9 51±10
Left ventricular end-diastolic diam
eter (mm) 37± 7 50± 4 49± 8 60±7 54±8 54±5
Left ventricular end-systolic diam
eter (mm) 29± 8 37± 4 29± 13 38±6 38±6 38±7
Medication (n)
Ace-inhibitors 0 0 0 4 5 7
Digitalis 0 1 4 0 0 2
Verapamil 2 2 3 2 1 0
Beta-blocker 4 2 2 2 1 2
Values are presented as mean value ± SD or number of patients. ACE inhibitor indicates Angiotensin Con-
verting Enzyme; AF, atrial fibrillation; CABG, Coronary Artery Bypass Grafting; CAF, chronic persistent
atrial fibrillation; PAF, paroxysmal atrial fibrillation; SR, control patients in sinus rhythm.
72
Chapter 5
RNA isolation and cDNA synthesis
Total RNA was isolated and processed as described previously.13 Briefly, first strand
cDNA was synthesized by incubation of 1 µg of total RNA in reverse transcription 10x
buffer, 200 ng of random hexamers with 200 units of Moloney Murine Leukemia Virus
Reverse Transcriptase, 1mM of each dNTP and 1 unit of RNase inhibitor (Promega, The
Netherlands) in 20 µl. Synthesis reaction was performed for 10 minutes at 20 °C, 20
minutes at 42 °C, 5 minutes at 99 °C and 5 minutes at 4 °C. All the products were checked
for contaminating DNA.
Semi quantitative PCR analyses
We described and validated these methods before.13 In short, the cDNA of interest
and the cDNA of the ubiquitously expressed housekeeping gene glyceraldehyde-3-phos
Table 2. The sequence for the primers.
protein sequence cycles ann. temp(°C)
GAPDH: F 5'-CCC ATC ACC ATC TTC CAG GAG CG-3', var. var.
R 5'-GGC AGG GAT GAT GTT CTG GAG AGC C-3'.
Na-channel: F 5'-ATG CAG CTG TGG ACT CCA GG-3' 27 56
SCN5A R 5'-CAG GCG GAT GAC TCG GAA GA-3'
L-type Ca2+ F 5'-CTG GAC AAG AAC CAG CGA CAG TGC G-3' 30 56
channel: R 5'-ATC ACG ATC AGG AGG GCC ACA TAG GG-3'
To1: F 5’-CAG CAA GTT CAC AAG CAT CC-3’ 31 52
Kv4.3 R 5’-AGC TGG CAG GTT AGA ATT GG-3’
Ks: F 5'-AGC AGA AGC AGA GGC AGA AG-3' 28 58
KvLQT1 R 5'-GAC GGA GAT GAA CAG TGA GG-3'
Kr: F 5’-GTC AAT GCC AAC GAG GAG GT-3’, 27 58
HERG R 5’-CTG GTG GAA GCG GAT GAA CT-3’
Kur: F 5’-AAC GAG TCC CAG CGC CAG GT-3’ 32 64
Kv1.5 R 5’-AGG CGG ATG ACT CGG AGG AT-3’
KACh F 5’-CAC CCT GGT GGA CCT CAA GTG GCG C-3’ 30 56
Kir3.4 R 5’-AGC TCC GGG CTT GGC AGG TCA TGC-3’
KATP F 5’-CAC GCT GGT GGA CCT CAA GT-3’ 29 59
Kir6.2 R 5’-CAC GAT GAG GCT CAG GAT GG-3’
GAPDH, glyceraldehyde-3-phosphate dehydrogenase; SCN5A, gene encoding Na-channel; l-type Ca2+ chan-
nel, voltage gated L type calcium channel α1 subunit; Kv4.3, gene probably encoding the calcium indepen-
dent transient outward current Ito1
; KvLQT1, gene encoding the slow delayed rectifier current together with
minK; HERG, gene encoding the rapid component of the delayed rectifier; Kv1.5, gene encoding the ultra-
rapid component of the delayed rectifier IKur
; Kir3.4, gene encoding part of the heterotetrameric complex of
this gene together with Kir3.1 which forms the acetylcholine dependent potassium current, IKACh
; Kir6.2,
gene encoding the inward rectifier K+ current, forming IKATP
with sulfonylurea receptor 2.
73
Ion channel remodeling is related to intra operative atrial refractory periods in patients
phate dehydrogenase (GAPDH) were co-amplified in a single PCR. Primers (Eurogentec,
Belgium) were designed for SCNA1, L-type calcium channel, Kv4.3, HERG, Kv1.5, Kir3.4
and Kir6.2 and the housekeeping gene GAPDH (Table 2).
The PCR products were separated on agarose gel by electrophoresis and stained with
ethidium bromide. The density of the PCR products was quantified by densitometry. Lin-
earity of the PCR was established by a good correlation between the number of cycles and
the density of gene of interest and GAPDH (data not shown).
Protein Preparation and Slot Blotting
Frozen atrial appendages of patients in sinus rhythm, patients with paroxysmal AF
and patients with persistent AF were homogenized in RadioImmunoPrecipitationAssay
(RIPA) buffer as described before.13 The homogenate was centrifuged at 14.000 rpm for
20 minutes at 4°C. The supernatant was used for protein concentration measurement ac-
cording to the Bradford method (Bio-Rad, The Netherlands) with bovine albumin used as
a standard. Samples of 10 µg heat denatured protein were spotted on nitrocellulose mem-
branes (Stratagene, The Netherlands) and checked by staining with Ponceau S solution
(Sigma, The Netherlands). Blocking was performed for 20 minutes in blocking buffer (5%
nonfat milk, TBS and 0.1% Tween 20). After washing three times in TBS with 0.1%
Tween 20 the membranes were incubated with primary antibody against GAPDH (Affin-
ity Reagents, USA), L-type calcium channel α1 subunit, Kv4.3, HERG, minK, Kir3.1
and Kv1.5 (all Alomone Labs, Israel). Immunodetection of the primary antibody was per-
formed with peroxidase conjugated secondary antibody anti-mouse IgG (Santa Cruz Bio-
technology, The Netherlands). The blot was incubated with the ECL-detection reagent
(Amersham, The Netherlands) for 1 minute, and exposed to an X-OMAT x-ray film (Kodak,
The Netherlands) for 15 to 90 seconds. The band densities were evaluated by densitomet-
ric scanning using a Snap Scan 600 (Agfa, The Netherlands). The amount of protein cho-
sen was in the linear immunoreactive signal area and the specificity of the antibody was
checked by SDS-PAGE and pre-incubation with the control peptide antigen.
Rate adaptation coefficient
To quantify the change in AERP at the different BCLs, we calculated the rate adapta-
tion coefficient for individual patients as the slope of the linear regression after logarith-
mic transformation of BCL. Three patients were excluded, because AERP was obtained at
less than 4 BCLs.
Statistical Analysis
All PCR and slot-blotting procedures were performed in duplo series and mean val-
ues were used for statistical analysis. Comparison between groups for normally distrib
74
Chapter 5
uted variables was performed by one-way ANOVA and for skewed variables by Wilkoxon
two-sample test. For determination of correlations the Spearman correlation test was used.
The Mann-Whitney U-test was performed for group to group comparisons of small num-
bers. All p-values are two-sided, a p-value <0.05 was considered statistically significant.
SPSS version 8.0 was used for all statistical evaluations.
Results
mRNA Remodeling
Changes in transcription of the gene of interest were determined by comparison of
gene-of-interest/GAPDH ratios between patients with persistent AF, paroxysmal AF and
their controls in sinus rhythm (Table 3). No differences in GAPDH amount between the
groups were found for all the genes investigated (data not shown). Persistent, lone AF was
associated with reductions in mRNA amount of Kv4.3, L-type Ca2+ channel and Kir3.4.
The mRNA amounts of HERG and KvLQT1 showed an additional reduction in persistent
AF with MVD. In general the reduction in mRNA expression was less pronounced in
paroxysmal AF.
Table 3. Percentage ion-channel remodeling: comparison of lone AF versus AF with MVD
mRNA Protein
lone +MVD lone +MVD lone +MVD lone +MVD
PAF PAF CAF CAF PAF PAF CAF CAF
Na-channel +35 ±6 -* - - na na
Kv4.3 -20 ±4 -20 ±5 -19 ±5 -25 ±3 -46 ±10-75 ±6* -33 ±8 -64 ±4*
L-type calcium channel -13 ±5 - -27 ±4 -20 ±4 -48 ±8 -65 ±5 -55 ±6 -57 ±6
HERG - - - -22 ±4 -28 ±10 -38 ±5 -25 ±8 -42 ±6
Kv1.5 - - - - -46 ±10 -69 ±5 -47 ±8 -61 ±7
KvLQT1/minK - +82 ±8* - +56 ±6* -35 ±5 -44 ±3 -50 ±6 -39 ±7
Kir3.4/Kir3.1 - - -27 ±5 -34 ±5 -41 ±7 -67 ±5* -42 ±6 -62 ±6*
Kir6.2 +28 ±4 -* - -42 ±5* na na
Only significant changes (p<0.05) are given for the mRNA or protein content of interest/GAPDH. CAF means
patients with chronic, persistent AF, PAF means paroxysmal AF, MVD means mitral valve disease; - means
not significant; na means not available. *, means significant differences between lone PAF and PAF with
MVD or lone CAF and CAF with MVD.
75
Ion channel remodeling is related to intra operative atrial refractory periods in patients
Protein Remodeling
Proteins were isolated from RAA and LAA and used for immunological detection of
L-type Ca2+ channel, Kv4.3, HERG, Kv1.5, minK and Kir3.1. Changes in protein expres-
sion were studied in relation to protein levels of GAPDH and to total amount of protein
spotted on the membrane. Because the GAPDH density and total protein amount density
showed a highly significant positive correlation (r=0.92, p<0.001), we used the protein of
interest/GAPDH ratio for further investigation. The protein expression of L-type Ca2+ chan-
nel, Kv4.3, Kv1.5, HERG, Kir3.1 and minK was reduced in both patients with both persis-
tent and paroxysmal AF (Figure 1 and 2, Table 3). Furthermore, ion-channel protein levels
did not correlate with mRNA levels, duration of persistent AF or with the duration of sinus
rhythm before surgery (data not shown).
Significant differences in protein remodeling between the lone AF group and the AF
group with MVD were observed. Reductions in protein expression of Kv4.3, minK and
Kir3.1 were more pronounced in patients with underlying MVD (Table 3).
Figure 1. Slot blot analysis of 10
µg of protein homogenates of 6
patients in sinus rhythm (SR), 6
patients with paroxysmal AF
(PAF) and 6 patients with
chronic, persistent AF (CAF).
The immunoblots were done for
(A) anti GAPDH, (B) anti L-type
calcium channel α1 subunit, (C)
anti Kv4.3, (D) anti-Kv1.5, (E)
anti HERG, (F) anti minK and
(G) anti-Kir3.1.
SRPAFCAF
SR
SR
PAF
PAF
CAF
CAF
A
B
C
SR
SR
SR
PAF
PAF
PAF
CAF
CAF
CAF
F
E
D
CAFPAFSRG
76
Chapter 5
Atrial Effective Refractory Period and Protein Remodeling
The AERP at 5 different basic cycle lengths (BCLs: 600, 500, 400, 300 and 250 ms)
was determined in the RAA and LAA of patients during surgery. Patients with persistent
and paroxysmal AF had significantly shorter AERPs than patients in sinus rhythm (Table
4). The relation between AERP and the amount of ion-channel protein was investigated,
because protein amounts are anticipated to represent the amount of functional ion-channel
closer than mRNA levels. A significant positive correlation was found at BCLs of 600,
500, 400 and 300 ms for all the proteins investigated in patients with AF (Figure 3, Table
5). Patients with reduced ion-channel protein expression exhibited the shortest AERP.
Furthermore, no significant correlation was found between the GAPDH amount and AERP
(data not shown).
Relation Rate Adaptation and Protein Remodeling
The rate adaptation coefficient was determined for every RAA and LAA. The rate
adaptation coefficient was significantly reduced by 32% in persistent AF compared to
sinus rhythm (mean persistent AF: 104 ± 53; paroxysmal AF: 133 ± 62 and sinus rhythm:
Figure 2. Protein ratios for (A) L-type calcium channel, (B) Kv4.3, (C) Kv1.5, (D) HERG, (E) minK and (F)
Kir3.1 for patients in sinus rhythm (SR), with paroxysmal AF (PAF) and with chronic, persistent AF (CAF). *,
p<0.01, **, p<0.001
SR PAF CAF0
1
2
SR PAF CAF0,0
0,5
1,0
SR PAF CAF0,0
0,5
1,0
1,5
SR PAF CAF0,0
0,5
1,0
1,5
SR PAF CAF0,0
0,5
1,0
SR PAF CAF
Ion-channel protein expression
0,0
0,5
1,0
1,5
* ** ** **
****
****
Ion-
chan
nel p
rote
in e
xpre
ssio
n
A B
C D
E F
77
Ion channel remodeling is related to intra operative atrial refractory periods in patients
153 ± 32), indicating a poorer adaptation to higher heart rates in patients with AF. Signifi-
cant positive correlations were observed between ion-channel protein expression and the
adaptation coefficient (Figure 4). AF patients with reduced ion-channel protein expression
demonstrated poorer rate adaptation.
Furthermore, significant differences were observed between lone paroxysmal AF and
patients with paroxysmal AF and MVD. Lone paroxysmal AF demonstrated a poorer rate
adaptation compared to paroxysmal AF with MVD (109 ± 40 and 164 ± 76, p=0.04,
respectively).
Discussion
Both experimental and human AF is accompanied by electrical remodeling2,4-6,10,19
and ion-channel remodeling.7,8,11,12-15,20 This is the first study which demonstrates in human
paroxysmal and persistent AF (1) a positive correlation between the ion-channel protein
remodeling and the AERPs, irrespective of the underlying heart disease, (2) a correlation
between ion-channel protein remodeling and changes in rate adaptation and (3) dis-
Figure 3. Correlation between
the ion-channel protein expres-
sion of (A) L-type calcium chan-
nel, (B) Kv4.3, (C) Kv1.5, (D)
HERG, (E) minK and (F) Kir3.1
and the AERP measured at BCL
of 500 ms in RAA and LAA.
( ) represents control patients
in sinus rhythm undergoing
CABG, ( ) patients with lone
paroxsymal AF, ( ) patients
with lone persistent AF, ( ) pa-
tients in sinus rhythm with un-
derlying MVD, ( ) patients
with paroxysmal AF and MVD,
( ) patients with persistent AF
and MVD.
0 1 2 3
AERP(msBCL500ms)
150200250300350
0 1 2150200250300350
r=0.77, p<0.001
r=0.65, p<0.001
L-type Ca2+ channel
Kv1.5
0,0 0,5 1,0 1,5r=0.43, p<0.001
Kv4.3
HERG0 1 2r=0.52, p<0.001
Kir3.10 1 2 3minK0,0 0,5 1,0 1,5150200250300350
r=0.55, p<0.001r=0.54, p<0.001
AE
RP
(m
s, B
CL
500 m
s)
A B
C D
E F
78
Chapter 5
Table 4. AERP measured at the different BCLs.
AERP (ms)
lone AF AF with MVD
BCL (ms) SR (CABG) PAF CAF SR (MVD) PAF CAF
600 291±53 222±15* 208±39* 271±19 274+25 226±42
500 277±42 224±24* 207±29* 268±22 259±35 228±34
400 252±34 216±24* 203±25* 256±20 243±32 217±33*
300 224±16 202±20* 189±24* 226±21 217±33 200±40
250 184±5 185±19 172±17 180±5 187±28 170±11
Data expressed as mean ± SD
*, means p<0.05
Table 5. Relation AERP and protein remodeling for the different basic cycle lengths (BCL).
BCL L-type Ca2+ Kv4.3 Kv1.5 HERG Kir3.1 minK
channel
r p-value r p-value r p-value r p-value r p-value r p-value
600 0.67 <0.001 0.32 0.001 0.57 <0.001 0.34 0.008 0.39 0.003 0.4 0.002
500 0.77 <0.001 0.43 <0.001 0.65 <0.001 0.53 <0.001 0.55 <0.001 0.54 <0.001
400 0.68 <0.001 0.35 0.004 0.61 <0.001 0.47 <0.001 0.45 <0.001 0.48 <0.001
300 0.53 <0.001 0.29 0.04 0.47 0.009 0.32 0.009 0.34 0.006 0.33 0.008
250 0.47 0.001 ns 0.42 0.004 0.3 0.03 ns 0.310.02
ns, not significant; r, regression coefficient
crepancies between mRNA and protein remodeling. These data suggest ion-channel protein
remodeling represent an important adaptation mechanism during AF, that may contribute
to intractability of AF and inactivity of antiarrhythmic drugs instituted for the prevention
of AF.
Relation ion channel remodeling and AERP
The observed ion-channel protein remodeling in this study is associated with the
occurrence of AF. Patients with paroxysmal and persistent AF showed marked reductions
in ion-channel protein expression of both L-type Ca2+ channel and several K+ channels.
Furthermore, low ion-channel protein levels were associated with short AERP and poor
79
Ion channel remodeling is related to intra operative atrial refractory periods in patients
rate adaptation. This indicates that electrical remodeling2 and structural remodeling21 are
paralleled by ion-channel protein remodeling as part of the adaptation mechanisms during
AF. Furthermore, patients with paroxysmal AF showed a reduction in ion-channel protein
expression comparable to persistent AF in the absence of mRNA reductions, suggesting
that paroxysms of AF are able to induce changes in ion-channel protein expression via
activation of a proteolytic system. Indeed, we have observed activation of the calpain
system in human paroxysmal and persistent AF (Brundel et al., submitted).
As stated above, AF is accompanied by shortening of the AERP and action potential
duration (APD). It has been suggested that the short-term decrease of APD and its reduced
rate adaptation is mainly due to a ± 70 % reduction of the L-type calcium current in
animal experimental studies and human AF.7,8,11,12 This assumption is further supported by
the observation that administration of the L-type Ca2+ channel agonist Bay K 8644 largely
restored the plateau phase of the action potential in remodeled cells.22 If the main role for
Kir3.10 1 2 3
Kv1.50 1 2
0100200300
Kv4.30,0 0,5 1,0 1,5L-type Ca2+ channel 0 1 2 3
adaptationcoefficient
0
100
200
300
r=0.50, p<0.001 r=0.34, p=0.004
r=0.34, p=0.007
minK0,0 0,5 1,0 1,5
0100200300
HERG0 1 2
r=0.37, p=0.003
r=0.36, p=0.003 r=0.28, p=0.02
adapta
tion c
oeff
icie
nt
Figure 4. Correlation between
the ion-channel protein expres-
sion and the rate adaptation
coefficient for (A) L-type cal-
cium channel, (B) Kv4.3, (C)
Kv1.5, (D) HERG, (E) minK
and (F) Kir3.1. ( ) represents
control patients in sinus rhythm
undergoing CABG, ( ) pa-
tients with lone paroxsymal AF,
( ) patients with lone persist-
ent AF, ( ) patients in sinus
rhythm with underlying MVD,
( ) patients with paroxysmal
AF and MVD, ( ) patients
with persistent AF and MVD.
80
Chapter 5
L-type Ca2+ channels in APD is correct, the observed reduction in protein expression of L-
type Ca2+ channel in this study explains the present AERP shortening and decrease in its
adaptation to rate.
The other possibility that may mediate AERP shortening is an increase in (repolariz-
ing) K+ channel gene products and/or activity. However, we observed a reduction of K+
channel gene expression. Similar results were obtained in animal experimental studies
showing reductions in ITo
and Kv4.3 mRNA amount without reductions in delayed inward
rectifier K+ current and Kir2.1 expression.7 The group of Van Wagoner et al. and our group
examined the adaptation in gene expression of several potassium channels in patients with
AF.13-15 The current of ITo
and the protein expression of Kv1.5 were reduced rather than
elevated during persistent AF.15 Our previous study, in a different patient group, showed
reductions in gene expression of Kv4.3, Kv1.5, Kir3.1 and Kir6.2.14 Only one study in
isolated RAA cells of patients with persistent AF showed that shortening of the human
action potential by AF was related to a 70% reduction in ICaL
and ITo
and a 30% increase in
IK1
and IKACh
.11 The downregulation of potassium channel protein amounts observed in our
study are in contrast with the few reports on the electrophysiological level. This possible
inconsistency between decrease in protein level and increase in current density may be
explained by a change in single channel properties in patients with persistent AF, such as
an increase of mean open-time, an increase in channel conductance or a change in voltage
dependency. Thus, a reduced expression of L-type Ca2+ channels probably plays a main
role in AERP shortening. Secondary to this process, the myocardial cell may further adapt
to high rate by reducing the expression of potassium channels to counteract the shortening
of the AERP.
We did not find differences in ion-channel protein expression between AF patients
with lone AF and AF with underlying MVD. Nevertheless, AERP was prolonged in MVD,
as previously reported in experimental studies.6,16,17 Also an association between AF with
MVD and severe cellular degeneration was observed.23 The results indicate that other
factors beside AF are probably involved in the regulation of the duration of the effective
refractory period. One of most likely candidates would be morphological changes, as AF
is promoted by structural changes induced during experimental heart failure, which cause
important local conduction abnormalities that could play an additional role in the vulner-
ability of AF.24,25
Post-transcriptional regulation?
The observed discrepancy between alterations in mRNA and protein expression in
patients with paroxysmal AF suggests the activation of proteolysis. Recently, we found
that activation of the calpain system in human persistent and paroxysmal AF, in the ab-
sence of activation of the proteasome pathway (Brundel et al., submitted). As calpain are
81
Ion channel remodeling is related to intra operative atrial refractory periods in patients
activated by calcium overload in the myocard cell26,27, calpain activation would serve to
protect the cells to additional damage by down-regulation of multiple ion-channels. How-
ever, this would be at the cost of proteolysis of several cytoskeletal, membrane-associated
and regulatory proteins26,28-32 Whether interference with the calpain system represents a
valuable therapeutic strategy in AF remains to be investigated.
In conclusion, the observed correlation between ion-channel protein amounts and
AERP strongly suggest that ion-channel protein remodeling, beside the electrical remod-
eling and structural remodeling33 may play an important role in the vulnerability of AF
after restoration of sinus rhythm.
Limitations of the Study
The patients with lone AF included in this study represent patients who were difficult
to treat and underwent finally MAZE surgery. Therefore, the present data cannot be ex-
trapolated uncritically to all AF patients. Furthermore, it should be noted that in all groups
the number of patients was small.
Acknowledgments
Dr. Van Gelder was supported by Grant 94.014 of the Netherlands Heart Foundation,
The Hague, The Netherlands. The study was supported by Grant 96.051 of The Nether-
lands Heart Foundation, The Hague, The Netherlands.
References
1. Kannel WB, Abbott RD, Savage DD, et al. Epidemiologic features of chronic atrial fibrillation: the
Framingham study. N Engl J Med 1982; 306:1018-1022.
2. Wijffels MC, Kirchhof CJ, Dorland R, et al. Atrial fibrillation begets atrial fibrillation. A study in awake
chronically instrumented goats. Circulation 1995; 92:1954-1968.
3. Pandozi C, Bianconi L, Villani M, et al. Electrophysiological characteristics of the human atria after
cardioversion of persistent atrial fibrillation. Circulation 1998; 98:2860-2865.
4. Franz MR, Karasik PL, Li C, et al. Electrical remodeling of the human atrium: similar effects in patients
with chronic atrial fibrillation and atrial flutter. J Am Coll Cardiol 1997; 30:1785-1792.
5. Yu WC, Chen SA, Lee SH, et al. Tachycardia-induced change of atrial refractory period in humans. Rate
dependency and effects of antiarrhytmic drugs. Circulation 1998; 97:2331-2337.
6. Tieleman RG, Van Gelder IC, Tuinenburg AE, et al. Intra- and post-operative atrial refractory periods in
relation to atrial arrhythmia history and the presence of mitral regurgitation. Circulation 1999; 100:I-361
7. Yue L, Melnyk P, Gaspo, et al. Molecular mehanisms underlying ionic remodeling in a dog model of atrial
fibrillation. Circ Res 1999; 84:776-784.
8. Yue L, Feng J, Gaspo R, et al. Ionic remodeling underlying action potential changes in a canine model of
atrial fibrillation. Circ Res 1997; 81:512-525.
9. Tieleman RG, De Langen CDJ, Van Gelder IC, et al. Verapamil reduces tachycardia-induced electrical
remodeling of the atria. Circulation 1997; 95:1945-1953.
10. Goette A, Honeycutt C, Langberg JJ. Electrical remodeling in atrial fibrillation. Time course and mecha-
nisms. Circulation 1996; 94:2968-2974.
11. Bosch RF, Zeng X., Grammer JB, et al. Ionic mechanisms of electrical remodeling in human atrial fibril-
lation. Cardiovasc Res 1999; 44:121-131.
82
Chapter 5
12. Van Wagoner DR, Pond AL, Lamorgese M, et al. Atrial L-Type Ca2+ Currents and Human Atrial Fibrilla-
tion. Circ Res 1999; 85:428-436.
13. Brundel BJJM, Van Gelder IC, Henning RH, et al. Gene expression of proteins influencing the calcium
homeostasis in patients with persistent and paroxysmal atrial fibrillation. Cardiovasc Res 1999; 42:443-
454.
14. Brundel BJJM, Van Gelder IC, Henning RH, et al. Changes in mRNA and protein content of ion channels
in patients with paroxysmal and persistent atrial fibrillation. PACE 1999; 22:II-753
15. Van Wagoner DR, Pond AL, McCarthy PM, et al. Outward K+ current densities and Kv1.5 expression are
reduced in chronic human atrial fibrillation. Circ Res 1997; 80:1-10.
16. Boyden PA, Tilley LP, Albala A, et al. Mechanisms for atrial arrhythmias associated with cardiomyopa-
thy: a study of feline hearts with primary myocardial disease. Circulation 1984; 69:1036-1047.
17. Boyden PA, Tilley LP, Pham T, et al. Effects of left atrial enlargement on atrial transmembrane potentials
and structure in dogs with mitral valve fibrosis. Am J Cardiol 1982; 49:1896-1908.
18. Gallagher MM, Camm AJ. Classification of atrial fibrillation. Pacing Clin Electrophysiol 1997; 20:1603-
1605.
19. Daoud EG, Knight BP, Weiss R, et al. Effect of verapamil and procainamide on atrial fibrillation-induced
electrical remodeling in humans. Circulation 1997; 96:1542-1550.
20. Gaspo R, Bosch RF, Bou-Abboud E, et al. Tachycardia-induced changes in Na+ current in a chronic dog
model of atrial fibrillation. Circ Res 1997; 81:1045-1052.
21. Ausma J, Wijffels M, Thone F, et al. Structural changes of atrial myocardium due to sustained atrial
fibrillation in the goat. Circulation 1997; 96:3157-3163.
22. Leistad E, Aksnes G, Verburg E, et al. Atrial contractile dysfunction after short-term atrial fibrillation is
reduced by verapamil but increased by BAY K8644. Circulation 1996; 93:1747-1754.
23. Thiedemann KU, Ferrans VJ. Left atrial ultrastructure in mitral valvular disease. Am J Pathol 1977;
89:575-604.
24. Li D, Melnyk P, Feng J, et al. Effects of experimental heart failure on atrial cellular and ionic electrophysi-
ology. Circulation 2000; 101:2631-2638.
25. Li D, Fareh S, Leung TK, et al. Promotion of atrial fibrillation by heart failure in dogs; atrial remodeling
of a different sort. Circulation 1999; 100:87-95.
26. Sun H, Leblanc N, Nattel S. Effects of atrial tachycardia on intracellular Ca2+ and cellular contractility.
Circulation 1999; 100:I-200
27. Ausma J, Dispersyn GD, Duimel H, et al. Changes in ultrastructural calcium distribution in goat atria
during atrial fibrillation. J Mol Cell Cardiology 2000; 32:355-364.
28. Bartus RT, Elliott PJ, Hayward NJ, et al. Calpain as a novel target for treating acute neurodegenerative
disorders. Neurol Res 1995; 17:249-258.
29. Atsma DE, Bastiaanse EM, Jerzewski A, et al. Role of calcium-activated neutral protease (calpain) in cell
death in cultured neonatal rat cardiomyocytes during metabolic inhibition. Circ Res 1995; 76:1071-1078.
30. Gorza L, Menabo R, Di Lisa F, et al. Troponin T cross-linking in human apoptotic cardiomyocytes. Am
J Pathol 1997; 150:2087-2097.
31. Gorza L, Menabo R, Vitadello M, et al. Cardiomyocyte troponin T immunoreactivity is modified by cross-
linking resulting from intracellular calcium overload. Circulation 1996; 93:1896-1904.
32. Gao WD, Atar D, Liu Y, et al. Role of troponin I proteolysis in the pathogenesis of stunned myocardium.
Circ Res 1997; 80:393-399.
33. Ausma J, Lenders MH, Duimel H, et al. Time course of structural changes in atria after atrial fibrillation
in goats: existence of hibernating myocardium. in press 2000
85
Gene expression of the natriuretic peptide system in atrial tissue
Chapter 6
Gene Expression of the Natriuretic Peptide system in
Atrial Tissue of Patients with Paroxysmal and
Persistent Atrial Fibrillation
Anton E. Tuinenburg, Bianca J.J.M. Brundel, Isabelle C. Van Gelder,
Robert H. Henning*, Maarten P. Van Den Berg, Cécile Driessen*,
Jan G. Grandjean†, Wiek H. Van Gilst*, Harry J.G.M. Crijns
From the Departments of Cardiology, Clinical Pharmacology*, and Thoracic Surgery†,
Thoraxcenter, University Hospital Groningen, Groningen, The Netherlands
J Cardiovasc Electrophysiol 1999;10:827-835
Abstract
Introduction: Circulating cardiac natriuretic peptides play an important role in
maintaining volume homeostasis, especially during conditions affecting hemodynamics.
During atrial fibrillation (AF), plasma atrial natriuretic peptide (ANP) becomes elevated.
It was the aim of the study to gather information about gene expression of the natriuretic
peptide system on the level of the atrium in patients with AF.
Methods and Results: Right atrial appendages of 36 patients with either paroxysmal or
persistent AF were compared with 36 case matched controls in sinus rhythm for mRNA
expression of pro- atrial natriuretic peptide (pro-ANP), pro-brain natriuretic peptide (pro-
BNP), and their natriuretic peptide receptor type-A (NPR-A). We investigated patients
without (n=36) and with (n=36) valvular disease. Persistent AF was associated with higher
mRNA expression of pro-BNP (+66%, p=0.04, in patients without valvular disease, and
+69%, p<0.01, in patients with valvular disease) and lower mRNA expression of NPR-A
(-58%, p=0.02, in patients without valvular disease, and -62%, p<0.01, in patients with
valvular disease). The mRNA content of pro-ANP was only increased in patients with
valvular disease (+12%, p=0.03). No changes were observed in patients with paroxysmal
AF. Conclusion: This study demonstrates that persistent AF, but not paroxysmal AF, induces
alterations in gene expression on the level of the atrium of pro-BNP and NPR-A. Although
AF is generally associated with an increase of plasma ANP, a change in mRNA content of
pro-ANP is only observed in the presence of concomitant valvular disease and is of minor
magnitude.
86
Chapter 6
Introduction
The pathophysiological mechanisms contributing to the perpetuation of atrial
fibrillation (AF), e.g. electrical remodeling and (ultra)cellular changes, are slowly being
unravelled.1-5 Abnormal calcium handling by the atrial myocyte,6-10 in response to
tachycardia induced intracellular calcium overload,11-13 is of pivotal importance in this
respect. However, the signaling pathways involved in these processes remain to be clarified.
The cardiac natriuretic peptide system plays an important role in
maintaining volume homeostasis via renal and cardiovascular actions, especially in
conditions that affect hemodynamics.14,15 In response to cardiac volume or pressure overload,
either caused by valvular disease, left ventricular dysfunction, hypertension, or AF, resultant
atrial stretch induces cardiac natriuretic peptide production. Data in AF, however, pertain
mainly to circulating atrial natriuretic peptide (ANP).16-19 Studies about brain natriuretic
peptide (BNP) in relation to AF are sparse and observational,20,21 merely stating that AF
(without specification) is one of the factors explaining plasma BNP level. ANP and BNP
modulate cardiac calcium handling indirectly via the autonomic nervous system,22 and
directly through cyclic guanosinemonophosphate (cGMP) mediated pathways (via the
cardiac natriuretic receptor), leading to reduced intracellular concentrations of cyclic
adenosinemonophosphate (cAMP), in turn lowering cAMP-dependent ion channel
activation.23 Among others, the L-type calcium channel becomes inactivated, whilst the
acetylcholine dependent potassium channel (KACh) becomes potentiated. Furthermore, ANP
stimulates calcium efflux from isolated atrial myocytes,24 underscoring the involvement
of natriuretic peptides in cellular calcium handling.
We hypothesized that AF would enhance gene expression of pro-atrial natriuretic
peptide (pro-ANP) and pro-brain natriuretic peptide (pro-BNP), and attenuate gene
expression of the cardiac natriuretic peptide receptor type-A (NPR-A) on the level of the
atrium. Therefore, it was our aim to investigate alterations in mRNA (messenger ribonucleic
acid) expression of pro-ANP, pro-BNP, and their NPR-A in the right atrial appendage
(RAA) of patients with paroxysmal and persistent AF undergoing cardiac surgery. To single
out the effect of AF, case matched control patients in SR were used. Since hemodynamic
overload of the heart per se is known to cause changes in natriuretic peptides,25-27 patients
with and without valvular disease were investigated.
Methods
Study patients
Patients were enrolled by using the in-house waiting list for cardiac surgery. The day
before elective cardiac surgery, the clinical characteristics of each patient (including
medication use and exercise tolerance according to the New York Heart Association
classification) were assessed by one investigator (AET). Patients with renal dysfunction
87
Gene expression of the natriuretic peptide system in atrial tissue
(serum creatinine > 150 µmol/L) were excluded from the study. Presence, type, and duration
of AF were assessed based on the patient’s history and previous electrocardiograms.
Echocardiographic data were obtained while patients were on the waiting list, but within
3 months prior to cardiac surgery. RAA was obtained from 36 patients with (paroxysmal
or persistent) AF and from 36 controls in sinus rhythm who were optimally case matched
for age, sex, and left ventricular function. During cardiac surgery, RAA was removed,
immediately snap-frozen in liquid nitrogen, and stored at -850C. The RAA’s were analyzed
separately in two series: one series consisting of patients without valvular disease, and
another series with patients with mitral or aortic valve disease of hemodynamic significance
to such extent that valvular surgery was deemed indicated. The investigation conforms to
the principles outlined in the Declaration of Helsinki, was approved by the Institutional
Review Board, and written informed consent was given by all patients.
RNA isolation and cDNA synthesis
Total RNA (ribonucleic acid) was isolated from RAA’s using the method of acid
guanidinium thiocyanate/phenol/chloroform extraction followed by a RNeasy kit for RNA
minipreps for tissues (Qiagen). The amount of RNA was evaluated by absorption at 260
nm, using a GeneQuant II (Pharmacia LKB Biotechnology, The Netherlands). The ratio
of absorption (260-280 nm) of all preparations was between 1.8 and 2.0. First strand
cDNA (copy-deoxyribonucleic acid) was synthesized by incubation of 1 µg of total RNA,
reverse transcription 10x buffer and 200 ng of random hexamers with 200 units of Moloney
Murine Leukemia Virus Reverse Transcriptase, 1mM of each dNTP
(dinucleotidetriphosphate) and 1 unit of ribonuclease (RNAse) inhibitor (Promega, The
Netherlands) in 20 µl. The synthesis reaction lasted 10 minutes at 20°C, 20 minutes at
42°C, 5 minutes at 99°C and 5 minutes at 4°C, respectively. All the products were checked
on contaminating DNA (data not shown).
Semiquantitative polymerase chain reaction analyses
We decribed these methods before.28,29 Validation of the present semi-quantitative
polymerase chain reaction (PCR) was performed by determination of the absolute alterations
of mRNA.28,29 In short, the cDNA of interest and the cDNA of the ubiquitously expressed
housekeeping enzyme glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were
coamplified in a single PCR. Primers were designed for pro-ANP, pro-BNP, the NPR-A
receptor, and the housekeeping gene GAPDH (Table 1). The oligonucleotides were
synthesized by Eurogentec (Belgium).
For the semi-quantitative PCR co-amplification 1 µl of cDNA mixture, 0.5 unit of
Taq polymerase (Eurogentec, Belgium) was added to 17.5 nM of dNTP’s, 10x PCR buffer
provided with Taq polymerase, 2.5 mM MgCl2
, 40 pmol of sense and antisense primer for
88
Chapter 6
the gene of interest, 40 pmol of sense and antisense GAPDH primer and water to bring the
final volume to 50 µl. All reaction mixtures were overlaid with 50 µl of mineral oil (Sigma,
The Netherlands). After 3 min denaturation at 94°C, n cycles of amplification (Table 1)
were performed, each for 1 min at 94°C , 1 min at annealing temperature (Table 1), 1 min
at 72°C, using the thermocycler Perkin Elmer 480 (The Netherlands). After the last cycle,
the 72°C elongation step was extended to 5 min. The PCR products were separated on a
1.5% agarose gel by gel-electrophoresis and stained with ethidium bromide. The densities
of the PCR products were quantified by densitometry (Aldus PhotoStyler 2.0, Graphic
Workshop and ImageQuant Version 3.3). Linearity for the PCR reactions was established
by making a correlation between the number of cycles and the ratio of the densities of
gene of interest / GAPDH (data not shown).
Table 1. Primer sequences
Sequence Cycles Annealing
Temperature
Glyceraldehyde-3-phosphate dehydrogenase:
F 5'-CCC ATC ACC ATC TTC CAG GAG CG-3' 26 -
R 5'-GGC AGG GAT GAT GTT CTG GAG AGC C-3'
Pro-ANP:
F 5'-CCA TGT ACA ATG CCG TGT CC-3' 26 56 °CR 5'-GCT CCA ATC CTG TCC ATC CT-3'
pro-BNP:
F 5'-GTT ACA GGA GCA GCG CAA CC-3' 26 56 °CR 5'-AGG CCA CTG GAG GAG CTG AT-3'
NPR-A receptor:
F 5'-CTT GCT CGG CAT TCT GAT TG-3' 26 56 °CR 5'-CAC GCA GTT GGA TGA CTT GA-3
ANP= atrial natriuretic peptide, BNP= brain natriuretic peptide,
NPR-A= natriuretic receptor type-A
Definitions
Persistent AF: continuous presence of AF until the moment of cardiac surgery, i.e.
at least two consecutive electrocardiograms with AF more than 24 hours apart, without
intercurrent SR. Persistent AF does not spontaneously convert to SR, but is considered
cardiovertible.30 Previously, this type of AF was classified as chronic AF.
Paroxysmal AF: AF typically occurring in episodes of shorter duration than 24 hours
(though paroxysms may occasionally last longer) with intermittent sinus rhythm.
Paroxysmal AF either converts spontaneously or can be terminated with an intravenously
administered antiarrhythmic
89
Gene expression of the natriuretic peptide system in atrial tissue
drug.30 Due to its spontaneous character, paroxysmal AF might be present at the moment
of cardiac surgery. The intensity of the arrhythmia was scored using a recently proposed
classification.31
Statistical analysis
All PCRs were performed twice. Mean values of the ratios are presented. Unless
stated otherwise, mean values and standard deviations are reported. For the comparison
between groups, a Student’s t-test was used for normally distributed variables and a
Wilcoxon-Mann-Whitney test for non-normally distributed variables. In case of categorical
variables, a chi-square test with continuity correction or a Fisher’s exact test was used,
when appropriate. For determination of correlation, the Spearman correlation test was
used. A two-sided probability level < 0.05 was considered to indicate statistical significance.
The analysis was performed by SAS statistical software (SAS, version 6.12, Cary, NC).
Results
Patients without valvular disease
Eight patients with paroxysmal AF and 10 patients with persistent AF were included.
These two groups were compared with their case matched control groups in sinus rhythm
(Table 2). The distribution of underlying heart disease and type of surgery in the AF groups
differed from the control groups; most AF patients (6 and 6 patients, respectively) underwent
atrial arrhythmia surgery (Cox’s MAZE III procedure32) for intractable, symptomatic AF
whilst all control patients underwent coronary bypass surgery. Despite these unavoidable
differences, patient groups were otherwise comparable for left ventricular function, atrial
dimensions, and functional class for exercise tolerance according to the New York Heart
Association classification (NYHA class I and II).
In the persistent AF group, the median duration of AF before surgery was 16 months,
with a range of 8 months to 64 months. In the paroxysmal AF group, the median duration
of sinus rhythm before surgery was 1.5 days, and the median frequency of paroxysms was
once a day with a median duration
of 3 hours. These patients can be categorized as the most severe type of
paroxysmal AF (type IIIc; symptomatic, > 1 attack / 3 months under treatment). Of note,
3 patients with paroxysmal AF were in AF at the moment of harvesting of the RAA during
surgery.
Changes in transcription of the genes of interest were determined by comparison of
gene of interest / GAPDH cDNA ratios between the paroxysmal AF group and the control
group (SR1), and between the persistent AF group and the control group (SR2) (Figure 1).
Patients with persistent AF showed a significant change of the cDNA ratios of pro-BNP /
GAPDH (increase, +66%, p=0.04, Figure 1B) and NPR-A / GAPDH (decrease, -58%,
90
Chapter 6
p=0.02, Figure 1C). No significant correlation was found between the duration of persistent
AF and the cDNA ratios of pro-BNP / GAPDH or NPR-A / GAPDH. In contrast to the
patients with persistent AF, no changes were observed in the paroxysmal AF group.
Patients with valvular disease
By coincidence, no patients with paroxysmal AF were available. Eighteen patients
with persistent AF were included, and compared with their case matched control patients
Table 2. Characteristics of the 36 patients without valvular disease
Chacteristic PaAF SR1 PeAF SR2
Patient number 8 8 10 10
Male / female (n) 6/2 6/2 6/4 6/4
Age (years) 51±7 56±11 63±11 65±8
Cardiac surgery
Coronary bypass grafting (n) 2 † 8 4 10
Cox’s MAZE III procedure (n) 6 † 0 6 0
Underlying heart disease **
Coronary artery disease (n) 2 † 8 4 10
Hypertension (n) 1 1 3 2
Lone AF (n) 6 † - 5 -
Concomitant systemic disease
Diabetes (n) 0 1 0 3
COPD (n) 1 2 1 4
Rhythm characteristics
Duration PeAF (months) * - - 16 (8-64) -
Duration SR (days) * 1.5 (0-30) - - -
NYHA class I / II / III
For exercise tolerance (n) 7/1/0 5/3/0 6/4/0 5/5/0
For angina (n) 6/0/2 † 1/2/5 6/1/3 † 1/4/5
Echocardiographic parameters
LA long axis view (mm) 43±7 41±3 45±7 44±5
LA apical view (mm) 60±6 64±3 63±4 64±6
RA long axis view (mm) 54±9 54±4 62±7 57±4
LVEDD (mm) 48±4 49±8 53±3 53±6
LVESD (mm) 35±4 35±7 33±6 35±4
Medication
ACE inhibitor (n) 0 1 4 2
Betablocker (n) 1 † 5 3 6
Calcium entry blocker (n) 0 3 3 3
Digoxin (n) 0 1 5 3
* = values are presented as median with range. ** = per patient, more than one underlying disease might
have been present. † = p value < 0.05 compared to the control group.
ACE = angiotensin converting enzyme, ASD = atrial septal defect, COPD = chronic obstructive pulmonary
disease, LA = left atrium, LVEDD = left ventricular end diastolic diameter, LVESD = left ventricular end
systolic diameter, ND = not done, NYHA = New York Heart Association, PaAF = paroxysmal atrial fibrilla-
tion, PeAF = persistent atrial fibrillation, RA = right atrium, SR = sinus rhythm, SR1 = are sinus rhythm
control patients for PaAF, SR2 = are sinus rhythm control patients for PeAF.
91
Gene expression of the natriuretic peptide system in atrial tissue
Figure 1. cDNA ratios for pro-ANP / GAPDH (Figure 1a), pro-BNP / GAPDH (Figure 1b), and NPR-A /
GAPDH (Figure 1c) of individual patients without valvular disease. SR1 are sinus rhythm control patients for
the patients with PaAF (paroxysmal atrial fibrillation), SR2 are sinus rhythm control patients for the patients
with PeAF (persistent atrial fibrillation). All data are represented in density units / density units. • = individual
value, = mean value of group ± standard error of the mean.
Figure 2. cDNA ratios for pro-ANP / GAPDH (Figure 2a), pro-BNP / GAPDH (Figure 2b), and NPR-A /
GAPDH (Figure 2c) of individual patients with valvular disease. SR are sinus rhythm control patients, PeAF are
the patients with persistent atrial fibrillation. All data are represented in density units / density units. • = indi-
vidual value, = mean value of group ± standard error of the mean.
Figure 3. Typical agarose gel showing the cDNA levels of pro-BNP and GAPDH of patients with valvular
disease. SR are sinus rhythm control patients for the patients with PeAF (persistent atrial fibrillation).
92
Chapter 6
Characteristic PeAF SR
Patient number 18 18
Male / female (n) 11/7 14/4
Age (years) 70 ±9 66 ±11
Cardiac surgery **
Valvular surgery (n) 18 18
Coronary bypass grafting (n) 7 6
ASD closure (n) 2 0
Primary heart disease
Aortic stenosis (n) 6 8
Aortic regurgitation (n) 0 1
Aortic stenosis & regurgitation (n) 3 0
Mitral stenosis (n) 0 0
Mitral regurgitation (n) 7 3
Mitral stenosis & regurgitation (n) 1 3
Aortic and mitral valve disease (n) 1 3
Concomitant heart disease **
Coronary artery disease (n) 7 6
ASD (n) 2 0
Hypertension (n) 6
Concomitant systemic disease **
Diabetes (n) 4 2
COPD (n) 3 4
Rhythm characteristics
Duration PeAF (months) * 6 (0.5-240) -
NYHA class I-III
For exercise tolerance (n) 1/1/16 1/6/11
For angina (n) 13/3/2 10/4/4
Echocardiographic parameters
LA long axis view (mm) 52±10 ‡ 44±5
LA apical view (mm) 72±9 66±7
RA long axis view (mm) 64±5 57±4
LVEDD (mm) 54±9 60±10
LVESD (mm) 38±7 43±15
Hemodynamic parameters
LV pressure max (mmHg) 167±40 178±64
LV pressure min (mmHg) 5±9 3
Wedge pressure (mmHg) 16±6 12±9
RA pressure (mmHg) 5±5 3±2
LV function I-III (n) † 13/1/3 14/0/4
Medication
ACE inhibitor (n) 11 ‡ 1
Betablocker (n) 3 4
Calcium entry blocker (n) 4 2
Digoxin (n) 14 ‡ 3
* = values are presented as median with range. ** = per patient, more than one surgical procedure might
have been performed or more than one concomitant disease might have been present. † = I indicates
normal, II indicates slightly impaired, and III indicates moderately impaired. ‡ = p value < 0.05 compared
to the control group.
LV = left ventricular, SR = are sinus rhythm control patients for PeAF. Other abbreviations are the same as
in Table 2.
Table 3. Characteristics of the 36 patients with valvular disease
93
Gene expression of the natriuretic peptide system in atrial tissue
in sinus rhythm (Table 3). Types of underlying heart disease, concomitant heart disease,
concomitant systemic disease, and surgery were equally distributed among both groups.
Patients were also comparable for NYHA functional class, left ventricular function, and
hemodynamic parameters. However, in the persistent AF group left atrial dimension (long
axis view) was larger, and angiotensin converting
enzyme inhibitors and digoxin were used more frequently. Clearly, the latter was used for
rate-control during AF. The median duration of AF before surgery was 6 months, with a
range of 0.5 months to 240 months.
Changes in transcription of the genes of interest were determined by comparison of
gene of interest / GAPDH cDNA ratios between the persistent AF group and the control
group (SR) (Figure 2). A representative agarose gel with cDNA levels of pro-BNP and
GAPDH is shown in Figure 3. Patients with persistent AF showed a significant change of
the cDNA ratios of all the genes of interest; a minor increase of pro-ANP / GAPDH (+12%,
p=0.03, Figure 2A), a substantial increase of pro-BNP / GAPDH (+69%, p<0.01, Figure
2B) and substantial decrease of NPR-A / GAPDH (-62%, p<0.01, Figure 2C). Also for the
patients with valvular disease, no significant correlation was found between the duration
of persistent AF and the cDNA ratios of pro-ANP / GAPDH, pro-BNP / GAPDH or NPR-
A / GAPDH.
Discussion
The main findings of the study are that persistent AF induces evident changes in
mRNA content of pro-BNP and NPR-A. The extent to which these changes occur are of
the same magnitude for patients with and without valvular disease. A change in mRNA
content of pro-ANP is only observed in the presence of concomitant valvular disease and
is of minor magnitude. Paroxysmal AF, although almost occurring daily in the present
patient group and therefore of clinical importance (6 of the 8 patients with paroxysmal AF
underwent Cox’s MAZE III procedure), was not associated with any change in expression
of the genes of interest.
The cardiac natriuretic peptides ANP (28-amino-acid peptide) and BNP (32-amino-
acid peptide) are produced in myocardium in the precursor forms pro-ANP and pro-BNP
that are spliced into an inactive and active peptide (N-terminal ANP / ANP and N-terminal
BNP / BNP respectively), which are released into the circulation.14 The third, more recently
discovered member of the natriuretic peptide family, C-type natriuretic peptide or CNP
(22-amino-acid peptide) is not considered to be produced in significant amounts in the
myocardium (not even in disease states),33 and was therefore not investigated in the present
study. ANP is generally considered to be primarily of atrial origin.14 BNP, on the other
hand, is mainly produced in the ventricles, and is considered to reflect left ventricular
function.34,35 ANP and BNP exercise their effects on myocardial electrophysiology and
94
Chapter 6
contractility indirectly via the autonomic nervous system, and directly via the natriuretic
receptor type-A (NPR-A), which is the principal receptor for natriuretic peptides in the
heart.22,23,36 NPR-A activates cGMP release, which in turn initiates activation of intracellular
protein kinase G.23 The pathways by which the natriuretic peptides modulate cellular
electrophysiology, calcium handling, and contractility are complex.23,24,37,38 The eventual
effects are most well described for ANP, being a decrease in intracellular calcium loading,
shortening of the action potential, an increase in conduction velocity, and a decrease in
phase 4 atrial depolarization, thus enhancing myocyte relaxation properties.22
mRNA expression of pro-ANP was unchanged in patients without valvular disease,
although the occurrence of AF is associated with a rise in plasma ANP.39,40 It should be
emphasized, however, that these observations pertain to acute AF. It is conceivable that
enhanced release of ANP from the atria, but not an altered gene expression, is responsible
for the rise in plasma ANP. A minor augmentation in mRNA content of pro-ANP was
observed only in the presence of valvular disease. In these patients, the hemodynamic
data of the AF group were comparable with the control group, but echocardiography showed
a significantly larger left atrium in the AF group, indicating higher atrial volume. This
difference was not observed in the patients without valvular disease, and might explain
the increased mRNA expression of pro-ANP in patients with valvular disease. This
possibility is supported by studies in patients with heart failure in sinus rhythm that indicate
the importance of atrial volume (stretch) for ANP release into the circulation.27 It should
also be noted that the (median) duration of AF was shorter in the presence of valvular
disease (shorter than 8 months in 10 of the 18 patients), as compared to group without
valvular disease (at least 8 months in all patients). Temporal depletion effects of plasma
ANP during the time course of longstanding AF (in a matter of months) have been
described.16,18 Pro-ANP gene expression might have been influenced the same way, thus
contributing to the differences in gene expression between patients with and without valvular
disease.
In contrast to pro-ANP, mRNA content of pro-BNP was increased irrespective of
valvular disease (+66% versus +69%, respectively). This is a surprising finding suggesting
that pro-BNP production in RAA is affected by the fibrillatory activity per se (a “load-
independent mechanism”), rather than by atrial stretch resulting from hemodynamic
overload. Analogous, during heart failure, the presence of atrial tissue-BNP is much more
pronounced as compared to ANP.26 Therefore, independent of the type of heart disease,
the atria might have the capability to vastly enhance BNP production, but not ANP
production. Another issue is that BNP seems to have a different atrial processing and
release in the circulation, as compared to ANP, especially during heart failure.26,41 The
latter might also be partly responsible for the differences in gene expression of pro-BNP
versus pro-ANP in response to AF. Only recently, a correlation between mRNA levels of
95
Gene expression of the natriuretic peptide system in atrial tissue
BNP and ANP, changing concomitantly with mean right atrial pressure, was reported for
right atrial appendage specimens of patients undergoing cardiac surgery, suggesting a
common regulation of tissue BNP and ANP.42 It should be noted, however, that patients
with and without valvular disease were mixed in this study, and that no data were given on
atrial rhythm.
Our study demonstrates that mRNA content of NPR-A is downregulated in patients
with persistent AF, either in the absence or presence of hemodynamic overload, i.e. valvular
disease. Receptor downregulation could be a result of an increase in signaling agonists, in
this case plasma ANP and BNP. mRNA expression of pro-BNP is clearly increased in the
present study, which would be in line with the observed downregulation of mRNA content
of NPR-A. Similar to the alterations of mRNA expression of pro-BNP, AF itself is likely
to be responsible for changes in mRNA content of NPR-A, because the magnitude of
change is comparable for patients with and without valvular disease (-58% versus -62%,
respectively).
In patients with frequently recurring attacks of paroxysmal AF under antiarrhythmic
treatment, to be assessed as a relevant arrhythmia burden, no changes in gene expression
of pro-ANP, pro-BNP, or NPR-A were found. It should be noted that most patients were in
sinus rhythm at the moment of the operation, with a median duration of 1.5 days (range 0-
30 days). These data suggest that intercurrent sinus rhythm between the attacks of
paroxysmal AF was enough to protect against gene alterations in the natriuretic peptide
system.
Limitations of the study
The present study has several important limitations. Firstly, atrial mRNA expression
was investigated without protein expression, binding experiments, ventricular mRNA
expression, or plasma level determinations. Such additional data might have provided
more insight in the pathophysiology. Secondly, patients with valvular disease used
“unloading” drugs, aimed at lowering of cardiac pressures and clinical stabilization. These
drugs, and differences in use of these drugs, might have interfered with the results, but the
clinical condition of most patients did not allow discontinuation (of more than 5x half-
time life) for study purposes.
Acknowledgement
Dr. Van Gelder was supported by Grant 94.014 from The Netherlands Heart
Foundation, The Hague, The Netherlands. The study was supported by Grant 96.051 from
The Netherlands Heart Foundation, The Hague, The Netherlands. We are indebted to Pieter
J. de Kam for helping us with the statistical analysis of the data.
96
Chapter 6
References
1. Wijffels MC, Kirchhof CJ, Dorland R, Allessie MA: Atrial fibrillation begets atrial fibrillation. A study in
awake chronically instrumented goats. Circulation 1995;92:1954-1968.
2. Goette A, Honeycutt C, Langberg JJ: Electrical remodeling in atrial fibrillation. Time course and mechanisms.
Circulation 1996;94:2968-2974.
3. Wijffels MC, Kirchhof CJ, Dorland R, Power J, Allessie MA: Electrical remodeling due to atrial fibrillation
in chronically instrumented conscious goats: roles of neurohumoral changes, ischemia, atrial stretch, and
high rate of electrical activation. Circulation 1997;96:3710-3720.
4. Ausma J, Wijffels MC, Thone F, Wouters L, Allessie M, Borgers M: Structural changes of atrial myocardium
due to sustained atrial fibrillation in the goat. Circulation 1997;96:3157-3163.
5. Ausma J, Wijffels MC, van Eys G, Koide M, Ramaekers F, Allessie M, Borgers M: Dedifferentiation of
atrial cardiomyocytes as a result of chronic atrial fibrillation. Am J Pathol 1997;151:985-997.
6. Leistad E, Aksnes G, Verburg E, Christensen G: Atrial contractile dysfunction after short-term atrial
fibrillation is reduced by verapamil but increased by BAY K8644. Circulation 1996;93:1747-1754.
7. Van Wagoner DR, Lamorgese M, Kirian P, Cheng Y, Efimov IR, Mazgalev TN: Calcium current density is
reduced in atrial myocytes isolated from patients in chronic atrial fibrillation. Circulation
1997;96:I-180(Abstract)
8. Daoud EG, Knight BP, Weiss R, Bahu M, Paladino W, Goyal R, Man KC, Strickberger SA, Morady F:
Effect of verapamil and procainamide on atrial fibrillation-induced electrical remodeling in
humans.Circulation 1997;96:1542-1550.
9. Yue L, Feng J, Gaspo R, Li GR, Wang Z, Nattel S: Ionic remodeling underlying action potential changes in
a canine model of atrial fibrillation. Circ Res 1997;81:512-525.
10. Tieleman RG, De Langen CDJ, Van Gelder IC, De Kam PJ, Grandjean JG, Bel KJ, Wijffels MC, Allessie
MA, Crijns HJGM: Verapamil reduces tachycardia-induced electrical remodeling of the atria. Circulation
1997;95:1945-1953.
11. Lee HC, Clusin WT: Cytosolic calcium staircase in cultured myocardial cells. Circ Res 1987;61:934-939.
12. Schouten VJ, Morad M: Regulation of Ca2+ current in frog ventricular myocytes by the holding potential,
c-AMP and frequency. Pflugers Arch 1989;415:1-11.
13. De Pauw M, Borgers M, Heyndrickx GR: Ultrastructural calcium distribution in cardiac myocytes after
48h of rapid pacing in dogs (abstr). Circulation 1996;94:I-604
14. Levin ER, Gardner DG, Samson WK: Natriuretic peptides. N Engl J Med 1998;339:321-328.
15. Wilkins MR, Redondo J, Brown LA: The natriuretic-peptide family. Lancet 1997;349:1307-1310.
16. Seino Y, Shimai S, Ibuki C, Itoh K, Takano T, Hayakawa H: Disturbed secretion of atrial natriuretic peptide
in patients with persistent atrial standstill: endocrinologic silence. J Am Coll Cardiol 1991;18:459-463.
17. Roy D, Paillard F, Cassidy D, Bourassa MG, Gutkowska J, Genest J, Cantin M: Atrial natriuretic factor
during atrial fibrillation and supraventricular tachycardia. J Am Coll Cardiol 1987;9:509-514.
18. Van den Berg MP, Crijns HJGM, Van Veldhuisen DJ, Van Gelder IC, De Kam PJ, Lie KI: Atrial natriuretic
peptide in patients with heart failure and chronic atrial fibrillation: role of duration of atrial fibrillation. Am
Heart J 1998;135:242-244.
19. Tuinenburg AE, Van Veldhuisen DJ, Boomsma F, Van den Berg MP, De Kam PJ, Crijns HJGM: Compari-
son of plasma neurohormones in congestive heart failure patients with atrial fibrillation versus patients
with sinus rhythm. Am J Cardiol 1998;81:1207-1210.
20. Wallen T, Landahl S, Hedner T, Nakao K, Saito Y: Brain natriuretic peptide predicts mortality in the eld-
erly. Heart 1997;77:264-267.
21. Niinuma H, Nakamura M, Hiramori K: Plasma B-type natriuretic peptide measurement in a multiphasic
health screening program. Cardiology 1998;90:89-94.
22. Clemo HF, Baumgarten CM, Ellenbogen KA, Stambler BS: Atrial natriuretic peptide and cardiac electro-
physiology: autonomic and direct effects. J Cardiovasc Electrophysiol 1996;7:149-162.
23. Koller KJ, Goeddel DV: Molecular biology of the natriuretic peptides and their receptors. Circulation
1992;86:1081-1088.
24. Yoshizumi M, Houchi H, Tsuchiya K, Minakuchi K, Horike K, Kitagawa T, Katoh I, Tamaki T: Atrial
natriuretic peptide stimulates Na+-dependent Ca2+ efflux from freshly isolated adult rat cardiomyocytes.
FEBS Lett 1997;419:255-258.
25. Burnett JC, Jr., Kao PC, Hu DC, Heser DW, Heublein D, Granger JP, Opgenorth TJ, Reeder GS: Atrial
natriuretic peptide elevation in congestive heart failure in the human. Science 1986;231:1145-1147.
26. Wei CM, Heublein DM, Perrella MA, Lerman A, Rodeheffer RJ, McGregor CG, Edwards WD, Schaff HV,
Burnett JC, Jr. Natriuretic peptide system in human heart failure. Circulation 1993;88:1004-1009.
97
Gene expression of the natriuretic peptide system in atrial tissue
27. Globits S, Frank H, Pacher B, Huelsmann M, Ogris E, Pacher R: Atrial natriuretic peptide release is more
dependent on atrial filling volume than on filling pressure in chronic congestive heart failure. Am Heart J
1998;135:592-597.
28. Van Gelder IC, Brundel BJJM, Henning RH, Tuinenburg AE, Tieleman RG, Deelman LE, Grandjean JG,
De Kam PJ, van Gilst WH, Crijns HJGM: Alterations in gene expression of proteins involved in the cal-
cium handling in patients with atrial fibrillation. J Cardiovasc Electrophysiol 1999;10:552-560.
29. Brundel BJJM, Van Gelder IC, Henning RH, Tuinenburg AE, Deelman LE, Tieleman RG, Grandjean JG,
van Gilst WH, Crijns HJGM: Gene expression of proteins influencing the calcium homeostasis in patients
with persistent and paroxysmal atrial fibrillation. Cardiovasc Res 1999;42:443-454.
30. Gallagher MM, Camm AJ: Classification of atrial fibrillation. Pacing Clin Electrophysiol 1997;20:1603-1605.
31. Levy S, Breithardt G, Campbell RW, Camm AJ, Daubert J-C, Allessie M, Aliot E, Capucci A, Cosio F,
Crijns HJGM, Jordaens L, Hauer RN, Lombardi F, Luderitz B: Atrial fibrillation: current knowledge and
recommendations for management. The Working Group on Arrhythmias of the European Society of Cardi-
ology. Eur Heart J 1998;19:1294-1320.
32. Cox JL, Jaquiss RD, Schuessler RB, Boineau JP: Modification of the maze procedure for atrial flutter and
atrial fibrillation. II. Surgical technique of the maze III procedure. J Thorac Cardiovasc Surg
1995;110:485-495.
33. Takahashi T, Allen PD, Izumo S: Expression of A-, B-, and C-type natriuretic peptide genes in failing and
developing human ventricles. Correlation with expression of the Ca2+-ATPase gene. Circ Res 1992;71:9-17.
34. Richards AM, Nicholls MG, Yandle TG, Frampton C, Espiner EA, Turner JG, Buttimore RC, Lainchbury
JG, Elliott JM, Ikram H, Crozier IG, Smyth DW: Plasma N-terminal pro-brain natriuretic peptide and
adrenomedullin: new neurohormonal predictors of left ventricular function and prognosis after myocardial
infarction. Circulation 1998;97:1921-1929.
35. Nagaya N, Nishikimi T, Goto Y, Miyao Y, Kobayashi Y, Morii I, Daikoku S, Matsumoto T, Miyazaki S,
Matsuoka H, Takishita S, Kangawa K, Matsuo H, Nonogi H: Plasma brain natriuretic peptide is a bio-
chemical marker for the prediction of progressive ventricular remodeling after acute myocardial infarction.
Am Heart J 1998;135:21-28.
36. Lin X, Hanze J, Heese F, Sodmann R, Lang RE: Gene expression of natriuretic peptide receptors in myo-
cardial cells. Circ Res 1995;77:750-758.
37. Le Grand B, Deroubaix E, Couetil JP, Coraboeuf E: Effects of atrionatriuretic factor on Ca2+ current and ICa
-
independent transient outward K+ current in human atrial cells. Pflugers Arch 1992;421:486-491.
38. Rebsamen MC, Church DJ, Morabito D, Vallotton MB, Lang U: Role of cAMP and calcium influx in
endothelin-1-induced ANP release in rat cardiomyocytes. Am J Physiol 1997;273:E922-31.
39. Oliver JR, Twidale N, Lakin C, Cain M, Tonkin AM: Plasma atrial natriuretic polypeptide concentrations
during and after reversion of paroxysmal supraventricular tachycardias. Br Heart J 1988;59:458-462.
40 Christensen G, Leistad E: Atrial systolic pressure, as well as stretch, is a principal stimulus for release of
ANF. Am J Physiol 1997;272:H820-6.
41. Suzuki E, Hirata Y, Kohmoto O, Sugimoto T, Hayakawa H, Matsuoka H, Kojima M, Kangawa K, Minamino
N: Cellular mechanisms for synthesis and secretion of atrial natriuretic peptide and brain natriuretic peptide
in cultured rat atrial cells. Circ Res 1992;71:1039-1048.
42. Doyama K, Fukumoto M, Takemura G, Tanaka M, Oda T, Hasegawa K, Inada T, Ohtani S, Fujiwara T, Itoh
H, Nakao K, Sasayama S, Fujiwara H: Expression and distribution of brain natriuretic peptide in human
right atria. J Am Coll Cardiol 1998;32:1832-1838.
99
Endothelin-1 mRNA is upregulated in patients
Chapter 7
Endothelin 1 mRNA is Upregulated in Human Persistent
Atrial Fibrillation with Underlying Valve Disease
Brundel, Short title: Endothelin system in atrial fibrillation
Bianca J. J. M. Brundel, MSc1,2; Isabelle C. Van Gelder, MD1;
Anton E. Tuinenburg, MD1; Mirian Wietses2, Dirk J Van Veldhuisen, MD1;
Wiek H. Van Gilst, PhD2; Harry J. G. M. Crijns, MD1; Robert H. Henning, MD2
Departments of Cardiology1and Clinical Pharmacology2, Thoraxcenter University
Hospital Groningen, The Netherlands.
Submitted Journal of Cardiovascular ElectrophysiologyAbstract
Background: Activation of the endothelin system is an important compensatory
mechanism that is activated during left ventricular dysfunction. Whether this system also
plays a role at the atrial level during AF has not been thoroughly examined. The purpose
of this study was to investigate mRNA and protein expression levels of the endothelin
system in AF patients with and without concomitant underlying heart disease. Metodsand results: Right atrial appendages of 36 patients with either paroxysmal or persistent
AF were compared with 36 controls in sinus rhythm. The mRNA amounts of pro-endothelin-
1 (ET-1), endothelin receptors A (ET-A) and B (ET-B) were studied by semi-quantitative
PCR. Protein amounts of the receptors were investigated by slot-blot analysis. The
endogenous mRNA production of pro-ET-1 was induced (+ 40%, p=0.002) in patients
with persistent AF and underlying valve disease. Furthermore, the ET-A and ET-B receptor
protein amounts were significantly reduced in paroxysmal AF (-39% and –47%,
respectively) and persistent AF with (-28% and –30%, respectively) and without (-20%
and –40%, respectively) underlying valve disease. Moreover, the mRNA amounts for pro-
ET-1 and ET-A were not different in AF patients, in contrast to ET-B mRNA amounts
which were significantly reduced in persistent AF with (-30%, p<0.001) and without (-
30%, p=0.04) underlying valve disease. Conclusions: Alterations in gene expression of
the endothelin system occur in the atria during AF, especially in the presence of underlying
valve disease. These results suggest that the endothelin system might play a role in adaptive
mechanisms in these patients.
100
Chapter 7
Introduction
There exists a reciprocal relation between congestive heart failure and atrial fibrillation
(AF) (1). Congestive heart failure may lead to AF. Also, AF may promote the occurrence
of heart failure. The endothelin (ET) system plays a role in the pathophysiology of heart
failure (2). Although, it is unknown yet whether the ET system plays a role in the
pathophysiology of AF and may mediate the atrial remodeling proces. Apart from their
direct inotropic effect on the myocardium, ET-1 increases intracellular calcium
concentrations, via activation of the L-type Ca2+ channel, and cell growth (3,4). Several
studies demonstrated that intracellular calcium overload played a crucial role in the atrial
electrical and contractile remodeling (5-7). Moreover, experimental data revealed that the
L-type Ca2+ channel plays a main role in shortening of AERP and action potential duration
(APD) (8,9). There are indications that endothelin plays also a role in human AF, as
suggested by elevated ET-1 plasma concentrations found in patients with AF with
concomitant heart failure (10,11). However, the atrial expression levels of ET-1 and the
endothelin receptor A and B in human AF has not been thoroughly examined.
The present study was designed to investigate the mRNA and protein amounts of
endogenous pro ET-1 and their receptors ET-A/ET-B, in right atrial appendages of patients
with AF. Included were patients with paroxysmal and persistent AF with and without
underlying valve disease and compared to patients in sinus rhythm undergoing cardiac
surgery.
Materials and Methods
Patient selection and atrial tissue collecting
The day before surgery, one investigator (AET) assessed the clinical characteristics
of the patient as previously described (12). Presence, type and duration of AF were assessed
by patient’s complaints and previous electrocardiograms. In addition, medication use and
exercise tolerance (according to the NYHA classification) was determined. Right atrial
appendages (RAAs) were gained from 10 patients with persistent AF without valvular
disorders and from 8 patients with paroxysmal AF. All patients were euthyroid. As controls
18 clinically stable patients in sinus rhythm undergoing CABG were used. The study was
approved by the Institutional Review Board and written informed consent was given by
all patients. Immediately after excision, the RAAs were snap-frozen in liquid nitrogen and
stored at -85 °C.
RNA isolation and cDNA synthesis
Total RNA was isolated and processed as described previously (12). Briefly, first
strand cDNA was synthesized by incubation of 1 µg of total RNA in reverse transcription
10x buffer, 200 ng of random hexamers with 200 units of Moloney Murine Leukemia
101
Endothelin-1 mRNA is upregulated in patients
Virus Reverse Transcriptase, 1mM of each dNTP and 1 unit of RNase inhibitor (Promega,
The Netherlands) in 20 µl total volume. Synthesis reaction was performed for 10 minutes
at 20 °C, 20 minutes at 42 °C, 5 minutes at 99 °C and 5 minutes at 4 °C. All the products
were checked for contaminating DNA.
Semi quantitative PCR analyses
We described and validated these methods before (12). In short, the cDNA of interest
and the cDNA of the ubiquitously expressed housekeeping gene glyceraldehyde-3-
phosphate dehydrogenase (GAPDH) were co-amplified in a single PCR. Primers
(Eurogentec, Belgium) were designed for pro-ET-1, ET-A and ET-B receptors and the
housekeeping gene GAPDH (Table 1).
The PCR products were separated on agarose gel by electrophoresis and stained with
ethidium bromide. The density of the PCR products was quantified by densitometry.
Linearity of the PCR was established by a good correlation between the number of cycles
and the density of gene of interest and GAPDH (data not shown).
Protein Preparation and Slot Blotting
From a number of patients there was enough tissue to perform protein isolation and
slot-blot analysis. Two different patient groups were made. The first group consisted of 6
patients with paroxysmal AF without valve disease, 9 patients with persistent AF without
valve disease and 6 controls in sinus rhythm. The second group consisted of 7 patients
Table 1. The sequence for the primers.
protein sequence cycles annealing
temp (°C)
GAPDH F 5'-CCC ATC ACC ATC TTC CAG GAG CG-3', var. var.
R 5'-GGC AGG GAT GAT GTT CTG GAG AGC C-3'.
Pro-ET-1 F 5’-TAC TTC TGC CAC CTG GAC AT-3’ 30 56
R 5’-CTT CCT CTC ACT AAC TGC TG-3’
ET-A F 5’-CAC GAT GAG GCT CAG GAT GG-3’ 29 56
R 5’-CTT CCT CTC ACT AAC TGC TG -3’
ET-B F 5'-CCG CAG AGA TAA TGA CGC CA-3' 29 56
GAPDH, glyceraldehyde-3-phosphate dehydrogenase; Pro-ET-1, gene encoding the endothelin pro-peptide;
ET-A, gene encoding the endothelin receptor type A; ET-B, gene encoding the endothelin receptor type B.
102
Chapter 7
with persistent AF with valve disease and 5 control patients in sinus rhythm with valve
disease. The frozen RAAs were homogenized in RadioImmunoPrecipitationAssay (RIPA)
buffer as described before (12). The homogenate was centrifuged at 14.000 rpm for 20
minutes at 4°C. The supernatant was used for protein concentration measurement according
to the Bradford method (Bio-Rad, The Netherlands) with bovine albumin used as a standard.
Samples of 10 µg heat denatured protein were spotted on nitrocellulose membranes
(Stratagene, The Netherlands) and checked by staining with Ponceau S solution (Sigma,
The Netherlands). After blocking in blocking buffer (5% nonfat milk, TBS and 0.1% Tween
20) and washing in TBS with 0.1% Tween 20 the membranes were incubated with primary
antibody against GAPDH (Affinity Reagents, USA), ET-A and ET-B (Research Diagnostics
Inc., USA). Immunodetection of the primary antibody was performed with peroxidase
conjugated secondary antibody anti mouse (for GAPDH) or anti sheep IgG (for ET-A and
ET-B, Santa Cruz Biotechnology, The Netherlands). The blot was incubated with the ECL-
detection reagent (Amersham, The Netherlands) for 1 minute, and exposed to an X-OMAT
x-ray film (Kodak, The Netherlands) for 15 to 90 seconds. The band densities were evaluated
by densitometric scanning using a Snap Scan 600 (Agfa, The Netherlands). The amount of
protein chosen was in the linear immunoreactive signal area and the specificity of the
antibody was checked by SDS-PAGE.
Statistical Analysis
All PCR and slot-blot procedures were performed in duplo series and mean values
were used for statistical analysis. Comparison between groups for normally distributed
variables was performed by one-way ANOVA. For determination of correlations the
Spearman correlation test was used. The Mann-Whitney U-test was performed for group
to group comparisons of small numbers. All p-values were two-sided, a p-value <0.05 was
considered statistically significant. SPSS version 8.0 was used for all statistical evaluations.
Results
Patients
Table 2 and 3 show the characteristics of the patients. The distribution of underlying
heart disease and type of surgery in the AF groups differed from the control groups; most
AF patients underwent atrial arrhythmia surgery (Cox maze III procedure) for intractable
symptomatic AF, whereas control patients underwent coronary bypass surgery and valvular
surgery. Other than these unavoidable differences, patient groups were comparable for left
ventricular function, atrial dimensions, and functional class for exercise tolerance according
to the NYHA classification. Of note, three patients with paroxysmal AF were in AF at the
moment of harvesting the RAA during surgery. Unfortunately no patients with underlying
valve disease and a history of paroxysmal AF could be included in this study.
103
Endothelin-1 mRNA is upregulated in patients
Table 2. Characteristics of the 36 patients without valvular disease
PAF SR1 CAF SR2
Male/female (n) 6/2 6/2 6/4 6/4
Age (years) 51 ± 7 56 ± 11 63 ± 11 65 ± 8
Cardiac surgery
Coronary bypass grafting 2 8 4 10
Cox’s maze III procedure 6 0 6 0
Underlying heart disease
Coronary artery disease 3 8 4 10
Lone AF 6 - 6 -
Rhythm characteristics
Duration CAF (months*) - - 16 (8-64) -
Duration SR (days*) 1.5 (0-30) - - -
NYHA Class I/II/III
For exercise tolerance 7/1/0 5/3/0 6/4/0 5/5/0
Echocardiographic parameters
LA long-axis view (mm) 43 ± 7 41 ± 3 45 ± 7 44 ± 5
LA apical view (mm) 60 ± 6 64 ± 3 63 ± 4 64 ± 6
RA long-axis view (mm) 54 ± 9 54 ± 4 62 ± 7 57 ± 4
LVEDD (mm) 48 ± 4 49 ± 8 53 ± 3 53 ± 6
LVESD (mm) 35 ± 4 35 ± 7 33 ± 6 35 ± 4
Medication
ACE inhibitor (n) 0 1 4 2
Beta blocker (n) 1 5 3 6
Calcium entry blocker (n) 0 3 3 3
Digoxin (n) 0 1 5 3
* values are presented as median (range). La = left atrium; LVEDD = left ventricular end diastolic diameter;
LVESD = left ventricular end-systolic diameter; NYHA = New York Heart Association; CAF = chronic,
persistent atrial fibrillation; PAF = paroxysmal atrial fibrillation; RA = right atrium; SR1 = sinus rhythm
control patients for PAF; SR2 = sinus rhythm control patients for CAF.
Endothelin system mRNA levels
Changes in transcription of the genes of interest were determined by comparison
gene of interest/GAPDH mRNA ratios between paroxysmal AF group and control group
(SR1), between persistent AF group and control group (SR2) and between persistent AF
group with valve disease and controls with valve disease (SR VD).
Pro-ET-1 mRNA contents were significantly increased (+40%, p=0.002) only in
patients with persistent AF and concomitant valve disease compared to control patients in
sinus rhythm with valve disease (Table 4 and Figure 1A and B).
The ET-B mRNA contents were reduced in patients with persistent AF and valve
disease (-30%, p<0.001) and without valve disease (-30%, p=0.04, Figure 2C,D, Table 4).
104
Chapter 7
Furthermore the mRNA amounts of ET-A were not different between the patient groups
(Figure 2A and B, Table 4).
Paroxysmal AF patients without valve disease showed also no differences in mRNA
amounts for pro-ET-1, ET-A and ET-B (Figure 1B and 2A,B).
Table 3. Characteristics of the 36 patients with valvular disease
CAF SR
Male/female 11/7 14/4
Age (years) 70 ± 9 66 ± 11
Cardiac surgery
Valvular surgery 18 18
Coronary bypass grafting 7 6
Underlying heart disease
Aortic stenosis 6 8
Aortic regurgitation 0 1
Aortic stenois and regurgitation 3 0
Mitral regurgitation 7 3
Mitral stenosis and regurgitation 1 3
Coronary artery disease 7 6
Rhythm characteristics
Duration CAF (months)* 6 (0.5 - 240)
NYHA Class I-III
For exercise tolerance 1/1/16 1/6/11
Echocardiographic parameters
LA long axis view (mm) 52 ± 10** 44 ± 5
LA apical view (mm) 72 ± 9 66 ± 7
RA long axis view (mm) 64 ± 5 57 ± 4
LVEDD (mm) 54 ± 9 60 ± 10
LVESD (mm) 38 ± 7 43 ± 15
Medication
ACE inhibitor (n) 11** 1
Beta blocker (n) 3 4
Calcium entry blocker (n) 4 2
Digoxin (n) 14** 3
* Values are presented as median (range). ** p<0.05 compared to the control group. ASD = atrial septal
defect; LA = left atrium; LVEDD = left ventricular end diastolic diameter; LVESD = left ventricular end-
systolic diameter; NYHA = New York Heart Association; CAF = chronic, persistent atrial fibrillation; RA =
right atrium; SR = sinus rhythm control patients for CAF.
105
Endothelin-1 mRNA is upregulated in patients
Figure 1.(A) Typical example of an agarose gel. Here the pro-ET-1 and GAPDH are shown of two patients with
persistent AF and valve disease (CAF VD), one patient with persistent AF (CAF), one patient with paroxysmal
AF (PAF) and controls in sinus rhythm (SR). Individual cDNA ratios for pro-ET-1/GAPDH are given for pa-
tients without valve disease (B) and with valve disease (C). SR1 and SR2 are the control groups in sinus rhythm
for persistent AF and paroxysmal AF, respectively. SR VD is the sinus rhythm control group with valve disease
for persistent AF with valve disease. Values are mean ± SEM.
Figure 2.Individual cDNA ratios are given for ET-A/GAPDH for patients without valve disease (A) and with
underlying valve disease (B). Also the individual cDNA ratios for ET-B/GAPDH of patients without valve dis-
ease (C) and with valve disease (D) are shown. SR1 and SR2 are the control groups in sinus rhythm for persistent
AF (CAF) and paroxysmal AF (PAF), respectively. SR VD is the sinus rhythm control group with valve disease
for persistent AF with valve disease (CAF VD). Values are given as mean ± SEM.
106
Chapter 7
Protein remodeling
Changes in protein expression were studied in relation to protein levels of GAPDH
and the density of total amount of protein spotted on the membrane. Because the GAPDH
density and total protein amount density showed a significant correlation, we used the
protein of interest/GAPDH ratio for further investigation.
The protein expression of ET-A and ET-B were significantly reduced in paroxysmal
(-39%, p=0.02 and –47%, p=0.02, respectively) and persistent AF without underlying
valve disease (-20%, p=0.04 and –40%, p=0.03) and persistent AF with underlying valve
disease (-28%, p=0.03 and –30%, p=0.03, respectively) (Table 4, Figure 3A and B).
Discussion
The main findings of the study are that 1) endogenous mRNA production of pro-ET-
1 is induced in patients with persistent AF with underlying valve disease, 2) ET-A and ET-
B receptor protein amounts are reduced in paroxysmal and persistent AF, irrespective to
the presence of underlying valve disease and finally 3) a discrepancy between ET-A protein
reductions in the presence of unchanged mRNA amounts is found. The results reveal that
in the atria of patients with AF, alterations in the endothelin system occur. Especially the
induction of pro-ET-1 in AF patients with valve disease may be of significant importance.
This finding can give new insights in the adaptation mehanisms during AF with respect to
the underlying valve disease reflecting the role of heart failure in AF.
Endothelin-1
Cardiac myocytes (13) as well as vascular endothelial cells (14) produce ET-1, a 21
amino acid peptide, a potent vasoconstricting peptide and exerts important cardiac effects
(2). These include positive inotropic and chronotropic effects in the heart of various species,
including humans, and growth-promoting properties. Experimental studies showed that
Table 4. mRNA and protein remodeling in patients with paroxysmal AF (PAF) and persistent AF (CAF)
without valve disease (VD) and persistent AF (CAF) with valve disease.
AF without valve disease AF with valve disease
PAF CAF CAF
mRNA protein mRNA protein mRNA protein
ET-1 ns - ns - +40% -
ET-A ns -39% ns -20% ns -28%
ET-B ns -47% -30% -40% -30% -30%
ns means not significant
107
Endothelin-1 mRNA is upregulated in patients
ET-1 trigger the increase of intracellular calcium (3), likely via the L-type calcium channel
in the atrial myocyte (4). It is well known that calcium overload plays a key role in the
pathophysiology of AF (5-7).
Elevated plasma concentrations of ET-1 were observed in patients with AF in the
setting of advanced heart failure (10,11). The found increase of endogenous pro-ET-1
mRNA levels in RAA of patients with persistent AF with underlying valve disease is in
accordance with the elevated plasma levels. Moreover this finding indicates that activation
of endothelin plays a role especially in patients with AF and concomitant moderate heart
failure. At the atrial myocytes the ET-1 could trigger elevation of intracellular calcium
amounts leading to AF induced contractile dysfunction (15,16) and electrical remodeling
(6,9,17) probably via the activation of calcium overload induced protease calpain (Brundel
et al., submitted)
A reduction of ET-B mRNA and no changes in ET-A mRNA amounts were found
during persisent AF irrespective to the underlying valve disease. This result might suggest
that the elevated endothelin acts on the ET-B receptor leading to reduction of its mRNA
SR PAF CAF SR VD CAF VD0
1
2
SR PAF CAF SR VD CAF VD0
1
2
ET-A protein expressionET-B protein expression
p=0.04p=0.02
p=0.03
p=0.03p=0.02
p=0.03
A
B
SR
CAF
PAF
SR
CAF
PAF
ET-
B p
rote
in e
xpre
ssio
nE
T-A
pro
tein
exp
ress
ion
Figure 3. Protein ratios of ET-A/GAPDH (A) and ET-B/GAPDH (B) of the individual patients. The top of each
panel shows a typical slot blot analysis of 10 µg protein of control patients (SR), persistent AF (CAF) and
paroxysmal AF (PAF). Values are given as mean ± SEM.
108
Chapter 7
expression. In this case the ET-B receptor seemed to play a dominant role in the RAA
compared to ET-A. However the ET-A receptor predominates in normal myocardium and
was found to play an important role during heart failure (2). This finding suggests that the
ET-A and ET-B receptors are differentially regulated on mRNA level during AF and heart
failure. Furthermore, the ET-B mRNA reduction in patients without valve disease can not
be explained by the endogenous elevation of pro-ET-1 and plasma ET-1 should be measured
to elucidate a possible exogenous induction of ET-1, which could trigger the ET-B mRNA
reduction.
Post-transcriptional regulation?
The observed discrepancy between alterations in mRNA and protein expression mainly
in patients with paroxysmal AF suggests the activation of proteolysis. Recently, we found
that activation of the calpain system in human persistent and paroxysmal AF, in the absence
of activation of the proteasome pathway (Brundel et al., submitted). As calpain are activated
by calcium overload in the myocard cell (5), calpain activation would serve to protect the
cells to additional damage by down-regulation of proteins. However, this would be at the
cost of proteolysis of several cytoskeletal, membrane-associated and regulatory proteins
(18,19). Whether interference with the calpain system represents a valuable therapeutic
strategy in AF remains to be investigated.
In conclusion
This report describes increased levels of pro-ET-1 mRNA in patients with persistent
AF with underlying valve disease in combination with reductions in protein levels of the
ET-A and ET-B receptor. This finding indicates that alterations in gene expression of the
endothelin system occur in the atria during AF in the presence of moderate heart failure.
The endothelin system could play a significant role in adaptive mechanisms in these patients.
Moreover, the observed discrepancy between mRNA and protein levels of these receptors,
mainly in patients with paroxysmal AF, suggest the influence of proteolytic system.
109
Endothelin-1 mRNA is upregulated in patients
References
1. Murgatroyd F, Camm AJ. Atrial arrhythmias. Lancet 1993;341:1317-22.
2. Ponicke K, Vogelsang M, Heinroth, Becker K, Zolk O, Bohm M, Zerkowski HR, Brodde OE. Endothelin
receptors in the failing and nonfailing human heart. Circulation 1998;97:744-51.
3. Alvarez BV, Perez NG, Ennis IL, Camilion de Hurtado M, Cingolani HE. Mechanisms underlying the
increase in force and Ca(2+) transient that follow stretch of cardiac muscle: a possible explanation of the
Anrep effect. Circ Res 1999;85:716-22.
4. He JQ, Pi Y, Walker J, Kamp TJ. Endothelin-1 and photoreleased diacylglycerol increase L-type Ca2+
current by activation of protein kinase C in rat ventricular myocytes. J Physiol (Lond) 2000;524:807-20.
5. Ausma J, Dispersyn GD, Duimel H, Thone F, Ver Donck L, Allessie M, Borgers M. Changes in ultrastruc-
tural calcium distribution in goat atria during atrial fibrillation. J Mol Cell Cardiology 2000;32:355-64.
6. Goette A, Honeycutt C, Langberg JJ. Electrical remodeling in atrial fibrillation. Time course and mecha-
nisms. Circulation 1996;94(11):2968-74.
7. Leistad E, Aksnes G, Verburg E, Christensen G. Atrial contractile dysfunction after short-term atrial fibril-
lation is reduced by verapamil but increased by BAY K8644. Circulation 1996;93(9):1747-54.
8. Yue L, Melnyk P, Gaspo, Wang Z, Nattel S. Molecular mehanisms underlying ionic remodeling in a dog
model of atrial fibrillation. Circ Res 1999;84:776-84.
9. Yue L, Feng J, Gaspo R, Li GR, Wang Z, Nattel S. Ionic remodeling underlying action potential changes in
a canine model of atrial fibrillation. Circ Res 1997;81(4):512-25.
10. Tuinenburg AE, Van Veldhuisen DJ, Boomsma, Van Den Berg M, De Kam PJ, Crijns HJGM. Comparison
of plasma neurohormones in congestive heart failure patients with atrial fibrillation versus patients with
sinus rhythm. Am J Cardiol 1998;81:1207-10.
11. Masson S, Gorini M, Salio M, Lucci, Latini R, Maggioni AP. Clinical correlates of elevated plasma natri-
uretic peptides and Big endothelin-1 in a population of ambulatory patients with heart failure. Ital Heart J
2000;1:282-8.
12. Brundel BJJM, Van Gelder IC, Henning RH, Tuinenburg AE, Deelman LE, Tieleman RG, Grandjean JG,
Van Gilst WH, Crijns HJGM. Gene expression of proteins influencing the calcium homeostasis in patients
with persistent and paroxysmal atrial fibrillation. Cardiovasc Res 1999;42:443-54.
13. Suzuki T, Kumazaki T, Mitsui Y. Endothelin-1 is produced and secreted by neonatal rat cardiac myocytes
in vitro. Biochem Biophys Res Commun 1993;193:823-30.
14. Yanagisawa M, Kurihara H, Kimura S, Tomobe Y, Kobayashi M, Mitsui Y, Yazaki Y, Goto K, Masaki T. A
novel potent vasocontrictor peptide produced by vascular endothelial cells. Nature 1988;332:411-5.
15. Daoud EG, Marcovitz P, Knight B, Goyal R, Ching Man K, Strickberger A, Armstrong WF, Morady F.
Short-term effect of atrial fibrillation on atrial contractile function in humans. Circulation 1999;99:3024-
7.
16. Manning WJ, Silverman DI, Katz SE, Riley MF, Come PC, Doherty RM, Munson JT, Douglas PS. Im-
paired left atrial mechanical function after cardioversion: relation to the duration of atrial fibrillation. J Am
Coll Cardiol 1994;23(7):1535-40.
17. Tieleman RG, De Langen CDJ, Van Gelder IC, De Kam PJ, Grandjean JG, Bel KJ, Wijffels MC, Allessie
MA, Crijns HJGM. Verapamil reduces tachycardia-induced electrical remodeling of the atria. Circulation
1997;95(7):1945-53.
18. Gorza L, Menabo R, Vitadello M, Bergamini CM, Di Lisa F. Cardiomyocyte troponin T immunoreactivity
is modified by cross-linking resulting from intracellular calcium overload. Circulation 1996;93(10):1896-
904.
19. Gao WD, Atar D, Liu Y, Perez NG, Murphy AM, Marban E. Role of troponin I proteolysis in the pathogen-
esis of stunned myocardium. Circ Res 1997;80(3):393-9.
113
Activation of proteolysis by calpain
Chapter 8
Activation of Proteolysis by Calpains in Human
Paroxysmal and Persistent Atrial Fibrillation
Brundel, Short title: Calpain activation in Atrial Fibrillation
Bianca J.J.M. Brundel, MSc1,2, Isabelle C. van Gelder, MD2, Harry J.G.M.
Crijns MD2, Wiek H. van Gilst PhD1, Robert H. Henning, MD1.
Department of Clinical Pharmacology1, Groningen University Institute for Drug
Exploration (GUIDE) and Department of Cardiology2, Thoraxcenter University Hospi-
tal, Groningen, The Netherlands.
Submitted Circulation Research
Abstract
Background: In human paroxysmal atrial fibrillation (AF), we observed reductions
of ion-channel proteins in the presence of unchanged mRNA levels. As this suggests
activation of proteolysis, we investigated two main proteolytic pathways, calpain and the
proteasome in atrial tissue of AF patients. Methods and Results: Right atrial appendages
were obtained from patients with paroxysmal (n=7) or persistent (n=10) lone AF and
compared to controls (n=10) in sinus rhythm undergoing coronary artery bypass grafting
(CABG). Proteolysis was measured using a fluorogenic substrate and expression of calpain
I and II was determined by Western-blot. Proteolytic activity was significantly increased
in paroxysmal and persistent AF, and abolished by the calpain inhibitor E-64, but unaffected
by the proteasome inhibitor lactacystin. Protein expression of calpain I was increased by
35% in persistent AF, while expression of calpain II was unchanged in AF. Tissue calpain
activity showed a significant correlation with protein levels of calpain I (r=0.61, p<0.001),
but not with calpain II levels.
Conclusions: Increased proteolytic activity during paroxysmal and persistent lone AF is
due to activation of the calpain pathway, especially of calpain I. Calpain activation may
represent an important mechanism conveying cellular changes underlying
electrophysiological, contractile and structural remodeling in AF.
114
Chapter 8
Introduction
Human atrial fibrillation (AF) is characterized by heterogeneity in electrical activation
pattern and the loss of contractile function of atrial tissue. Recent experimental research
has demonstrated AF to induce changes at the electrophysiological, protein and
morphological level, which increase the vulnerability to AF.1-3 Still, the molecular
mechanisms underlying these changes are poorly characterized. In a previous human study
we observed a reduction in protein expression of several plasma membrane ion channels
in the absence of changes in mRNA in patients with lone paroxysmal AF.4 Therefore, we
hypothesize activation of a protein degradation mechanism during AF. Different proteolytic
pathways may be involved. First, as cytosolic calcium is increased during AF2, proteolysis
may be invoked by calcium dependent neutral proteases, calpain I and calpain II, whose
activity was demonstrated in animal models of metabolic inhibition, cardiac stunning and
calcium overload.5-7 Alternatively, activation of the proteasome may underlie increased
proteolysis in cardiac cells, as shown in rat for degradation of myosin heavy chain and
connexin43 gap junctions.8,9
To examine proteolytic activation in human AF, we determined the activity of calpains
and the proteasome in atrial tissue of patients with paroxysmal and persistent lone AF and
of controls in sinus rhythm. As an increased calpain-mediated proteolysis was found in
AF, protein levels of calpain I and II were determined.
Materials and Methods
Patients and tissue collection
Right atrial appendages (RAAs) were obtained as described before10 from patients
with normal left ventricular function. Patients with paroxysmal (n=7) and persistent (n=10)
lone AF undergoing MAZE surgery were included and matched to clinically stable control
patients in sinus rhythm undergoing CABG (n=10, Table 1). The Institutional Review
Board approved of the study and patients gave written informed consent.
Protein Extraction
For analysis of proteolysis, frozen RAAs were homogenized in buffer (100 mM Tris-
HCl, 145 mM NaCl, pH=7.3) and centrifuged at 26.000 x g (30 min, 4°C). For Western-
blot analysis, parts of the same RAAs were homogenized in Radio-Immuno-Precipitation-
Assay (RIPA) buffer.10 Protein concentration was determined using the DC assay (Bio-
Rad, Netherlands) with a bovine albumin standard.
Proteolytic Assay
Suc-Leu-Leu-Val-Tyr-7-amino-4-methyl-coumarin (AMC, Sigma, Netherlands) was
used as substrate. Twenty-five µg protein extract was added to 20 µM AMC in 300 µl
115
Activation of proteolysis by calpain
Tris-buffered saline. AMC release was measured by fluorometry (360-nm excitation; 430-
nm emission, Spectrometer LS50B, Perkin Elmer, Netherlands) after incubation for 30
min at 25°C. Standard curves were generated using known concentrations of 7-amino-4
methyl-coumarin (Sigma, Netherlands) and 25 µg heat denatured protein. Maximal calpain
activation was assessed after reconstitution of calcium at 1 mM. E-64 (10-4 M, Roche, The
Netherlands), calpain I inhibitor (N-Acetyl-Leu-Leu-norleucinal, 10-4 M, Sigma, The
Netherlands) and calpain II inhibitor (N-acetyl-Leu-Leu-methioninal, 10-4 M Sigma, The
Netherlands) were used to assess calpain activation. Lactacystin (10-4 M, Calbiochem,
Netherlands) was used to investigate proteasome activity. Assays were conducted in
triplicate.
Western-Blot Analysis
Protein expression was determined by Western-blot and expressed as ratio to levels
of GAPDH, as described previously.10 Denatured protein (10 µg) was separated by SDS-
PAGE, transferred to nitrocellulose membranes (Stratagene, Netherlands) and incubated
with primary antibodies against GAPDH (Affinity Reagents, USA), calpain I and calpain
II (Research Diagnostics, USA). Horseradish peroxidase-conjugated anti-mouse or anti-
rabbit IgG (Santa-Cruz Biotechnology, Netherlands) was used as secondary antibody.
Signals were detected by the ECL-detection method (Amersham, Netherlands) and
quantified by densitometry.
Table 1. Baseline characteristics of patients with lone paroxysmal AF (PAF), lone persistent AF (CAF)
and control patients in sinus rhythm (SR).
Patient characteristics SR PAF CAF
N 10 7 10
Age 60±7 50±7 53±8
Previous duration AF - Median, range (months) - - 12.5 (3-56)
Duration SR before surgery - Median, range (days) - 2 (0.5-12) -
Duration last paroxysm - Median, range (hours) 9 (1-24)
Exercise tolerance
• NYHA Class I 10 6 7
• NYHA Class II 0 1 3
Medication
• Digitalis 0 1 2
• Calcium antagonists 4 2 2
• Beta-blockers 6 1* 1*
* p<0.05 compared to SR
116
Chapter 8
Statistical Analysis
Results are expressed as mean ± SD. One-way ANOVA was used for multiple group
comparisons. Correlation was determined with Spearman correlation test (SPSS 8.0). p<0.05
was considered statistically significant.
Results
Proteolytic activity
Proteolytic activity was significantly increased in tissue from patients with paroxysmal
and persistent AF compared to patients in sinus rhythm (Figure 1A, Table 2).
Proteolytic activity was measured in the presence of inhibitors to assess the
involvement of different pathways (Table 2). The non-selective calpain inhibitor E-64 (10-
4 M) and calpain I inhibitor (10-4 M) significantly reduced both tissue and maximal
proteolytic activity to a similar level in all groups. Calpain II inhibitor (10-4 M) reduced
proteolysis, but only partially attenuated the increased proteolytic activity observed in AF.
In contrast, the proteasome inhibitor lactacystin (10-4 M) did not reduce proteolytic activity
in any group. Thus, increased proteolysis in AF is due to activation of calpains.
To assess maximal calpain activation, proteolytic activity was determined in the
presence of 1 mM calcium. Under these conditions, a similar increase in proteolytic activity
was found in patients with AF and controls compared with experiments in the absence of
calcium (Figure 1B, Table 2). The extra activation due to addition of calcium was abolished
by E-64, calpain I inhibitor and calpain II inhibitor, but unaffected by lactacystin (Table
2). Further, tissue calpain activity across all groups correlated significantly with the maximal
calpain activity (r=0.89, p<0.001; Figure 1C).
SR PAF CAF
Calpain activity (nM AMC/mg/30min)
01020304050607080
SR PAF CAF0
1020304050607080
0 10 20 30 40 50 600
1020304050607080
r=0.89, p<0.001
A CB
** *
*
Figure 1.
Significant increase of tissue calpain activity (A) and maximal calpain activity (B) in RAAs of patients with
sinus rhythm (SR; o), paroxysmal AF (PAF; ) and persistent AF (CAF; •). * p<0.01 compared to SR. (C)
Significant correlation for maximal and tissue calpain activity. Calpain activity was expressed as nM AMC/mg
protein/30 min.
Calp
ain
acti
vit
y (
nM
AM
C/m
g/3
0m
in)
117
Activation of proteolysis by calpain
Calpain I and II Protein Levels
To examine the relation between tissue calpain activity and calpain expression, protein
levels of calpain I and II were determined by Western-blotting and expressed as ratio to
the housekeeping enzyme GAPDH (Figure 2A,D). Protein levels of calpain I were
significantly increased with 35% in patients with persistent AF, compared to controls,
whereas calpain II levels were similar in all groups (Figure 2B,E). GAPDH levels did not
differ between the groups (sinus rhythm: 1139 ± 150, paroxysmal AF: 1149 ± 175, persistent
AF: 1317 ± 190, arbitrary OD units). A positive correlation was found between tissue
calpain activity and protein expression of calpain I (r=0.61, p<0.001, Figure 2C). In contrast,
no correlation was found between calpain II expression and tissue calpain activity (r=-0.2,
p=0.33, Figure 2F).
Discussion
In this study, activation of proteolysis was found in the atrial tissue of patients with
paroxysmal and persistent lone AF. Increased proteolytic activity in AF is mediated
exclusively by calpains, as the non-selective calpain inhibitor E-64 reduced proteolysis to
similar levels in all groups, even under high calcium conditions known to activate calpain
II. In addition, the selective inhibitor of the proteasome, lactacystin, did not influence
proteolysis in any group. Thus these findings represent, to the best of our knowledge, the
first report demonstrating the activation of the calpain pathway in human cardiac disease.
The relative contribution of calpain I and II is difficult to establish by using inhibitors,
due to their partial selectivity. However, calpain I inhibitor completely attenuated the
increased proteolysis in AF both under basal and high calcium conditions, whereas calpain
II inhibitor did not. In addition, Western-blot demonstrated increased protein levels of
Table 2. Proteolytic activity (nM AMC/mg protein/30min) in atrial tissue of patients with lone paroxys-
mal AF (PAF), lone persistent AF (CAF) and control patients in sinus rhythm (SR).
no calcium added + 1 mM calcium
SR PAF CAF SR PAF CAF
None 33 ± 10 53 ± 13* 58 ± 17* 41 ± 13 72 ± 17* 82 ± 16*
E-64 (10-4 M) 19 ± 4 22 ± 4 23 ± 6 20 ± 6 23 ± 8 25 ± 5
Calpain I inhibitor (10-4 M) 23 ± 3 24 ± 3 25 ± 2 25 ± 4 26 ± 3 27 ± 4
Calpain II inhibitor (10-4 M) 21 ± 5 27 ± 6 33 ± 1* 23 ± 7 34 ± 11* 37 ± 13*
Lactacystin (10-4 M) 35 ± 12 47 ± 15 69 ± 12* 32 ± 11 60 ± 16* 72 ± 16*
* p<0.05 compared to SR
118
Chapter 8
calpain I in persistent AF, but unchanged levels of calpain II. Finally, calpain I protein
levels, but not calpain II levels, correlated with increased calpain activity. Taken together,
these results suggest that the increased proteolysis in AF is mainly due to activation and
up-regulation of calpain I, rather than calpain II.
Calpain activation represents a likely candidate to mediate important cellular changes
in AF. While calpains are readily activated by elevated cellular calcium levels11, calcium
overload plays a key role in the pathogenesis of AF2,12. Moreover, calpains have been
demonstrated to degrade cytoskeletal13, contractile6,7 and L-type Ca2+ channel14 proteins.
Reduction in expression of the L-type Ca2+ channel is thought to play a major role in the
electrical remodeling in AF.3 Finally, calpain activation would explain the apparently general
reduction in protein expression of multiple ion-channels in the absence of changes in
Figure 2.
Protein levels of calpain I (left panels) and calpain II (right panels) in patients with sinus rhythm (SR; o), parox-
ysmal AF (PAF; ) and persistent AF (CAF; •) expressed as ratio to GAPDH. (A,D) typical Western-blots of 10
µg protein. (B,E) group protein ratios. (C,F) Significant correlation of calpain activity and calpain I levels,
whereas absence of correlation with calpain II levels. * p<0.01 compared to SR.
Calpain I expression 0,5 1,0 1,5 2,0Calpain activity (nM
0102030405060
r = 0.61, p<0.001
Calpain II expression 0,5 1,0
0102030405060
r = -0.2, p=0.33
SR PAF CAF
Calpain expression
0,60,81,01,21,41,6
SR PAF CAF0,60,81,01,21,41,6*
Calpain I Calpain II
SR PAF CAF SR PAF CAF
Calpain I
GAPDH
Calpain II
GAPDH
A
B
C
D
E
F
Calp
ain
expre
ssio
nC
alp
ain
acti
vit
y (
nM
AM
C/m
g/3
0m
in)
119
Activation of proteolysis by calpain
mRNA, as observed in human paroxysmal AF.4 Thus, interference with the calpain pathway
may both represent an important tool to unravel the sequence of molecular events in AF,
and a possible future alternative for pharmacological intervention.
References
1. Wijffels MC, Kirchhof CJ, Dorland R, et al. Atrial fibrillation begets atrial fibrillation. A study in awake
chronically instrumented goats. Circulation 1995; 92:1954-1968.
2. Ausma J, Dispersyn GD, Duimel H, et al. Changes in ultrastructural calcium distribution in goat atria
during atrial fibrillation. J Mol Cell Cardiology 2000; 32:355-364.
3. Yue L, Melnyk P, Gaspo, et al. Molecular mehanisms underlying ionic remodeling in a dog model of atrial
fibrillation. Circ Res 1999; 84:776-784.
4. Brundel BJJM, Van Gelder IC, Henning RH, et al. Alterations in potassium channel gene expression in
atria of patients with persistent and paroxysmal atrial fibrillation. J Am Coll Cardiol 2000; in press
5. Atsma DE, Bastiaanse EM, Jerzewski A, et al. Role of calcium-activated neutral protease (calpain) in cell
death in cultured neonatal rat cardiomyocytes during metabolic inhibition. Circ Res 1995; 76:1071-1078.
6. Gorza L, Menabo R, Vitadello M, et al. Cardiomyocyte troponin T immunoreactivity is modified by cross-
linking resulting from intracellular calcium overload. Circulation 1996; 93:1896-1904.
7. Gao WD, Atar D, Liu Y, et al. Role of troponin I proteolysis in the pathogenesis of stunned myocardium.
Circ Res 1997; 80:393-399.
8. Eble DE, Spragia ML, Ferguson A, et al. Sarcomeric myosin heavy chain is degraded by the proteasome.
Cell Tissue Res 1999; 296:541-548.
9. Laing JG, Tadros PN, Saffitz J, et al. Proteolysis of connexin43-containing gap junctions in normal and
heat-stressed cardiac myocytes. Cardiovasc Res 1998; 38:711-718.
10. Brundel BJJM, Van Gelder IC, Henning RH, et al.Gene expression of proteins influencing the calcium
homeostasis in patients with persistent and paroxysmal atrial fibrillation. Cardiovasc Res 1999; 42:443-
454.
11. Suzuki K, Imajoh S, Emori Y, et al. Calcium-activated neutral protease and its endogenous inhibitor.
Activation at the cell membrane and biological function. FEBS Letters 1987; 220:271-277.
12. Goette A, Honeycutt C, Langberg JJ. Electrical remodeling in atrial fibrillation. Time course and mecha-
nisms. Circulation 1996; 94:2968-2974.
13. Papp Z, Van Der Velden J, Stienen G. Calpain-I induced alterations in the cytoskeletal structure and
impaired mechanical properties of single myocytes of rat heart. Cardiovasc Res 2000; 45:981-993.
14. Belles B, Hescheler J, Trautwein W, et al. A possible physiological role of the Ca-dependent protease
calpain and its inhibitor calpastatin on the Ca current in guinea pig myocytes. Pflugers Arch 1988; 412:554-
556.
121
Calpain activity is related to ion-channel, structural and electrical remodeling
Chapter 9
Calpain Activity is related to Ion-Channel, Structural and
Electrical Remodeling in human Paroxysmal and
Persistent Atrial Fibrillation
Brundel, Short title: Calpain activation during Atrial Fibrillation
Bianca J. J. M. Brundel, MSc1,2, Jannie Ausma, PhD4, Isabelle C. Van Gelder,
MD2, Harry J. G. M. Crijns, MD2, Wiek H. van Gilst, PhD1,
Robert G. Tieleman, MD2 Johan J.L. Van Der Want, PhD3;
Robert H. Henning, MD1.
Department of Clinical Pharmacology1, Groningen University Institute for Drug
Exploration (GUIDE), Department of Cardiology2, Thoraxcenter University Hospital,
Groningen, Department of Cell Biology and Electron Microscopy3, University of
Groningen and Department of Physiology4, Cardiovascular Research Institute
Maastricht, University of Maastricht, The Netherlands
Submitted CirculationAbstract
Background: Atrial fibrillation (AF) is accompanied by electrical, structural and
ion-channel protein remodeling. We tested whether calcium activated neutral protease,
calpain, is involved in AF-induced remodeling. Therefore, calpain activity and localization
in atrial tissue of patients with paroxysmal and persistent AF were determined, and correlated
with remodeling processes. Methods and Results: Right atrial appendages were obtained
from patients with paroxysmal (n=7, PAF), persistent (n=10, CAF) lone AF and controls
in sinus rhythm (n=10) and used for measuring tissue calpain activity with a fluorogenic
calpain specific substrate. Calpain I localization was studied by immunohistochemistry.
Ion-channel protein expression was determined by slot-blot analysis. Structural changes
were quantified by counting atrial myocytes showing degeneration or hibernation. Rate
adaptation was calculated from atrial effective refractory periods (AERPs) at 5 basic cycle
lengths. Tissue calpain activity was significantly increased in PAF (2-fold, p<0.001) and
CAF (3-fold, p<0.001), mainly due to calpain I activation. Calpain I was localized in the
cytosol, intercalated discs and nucleus of cardiomyocytes. Patients with AF showed
122
Chapter 9
decreased protein expression and rate adaptation, and increased structural changes compared
to controls. Tissue calpain activity correlated with ion-channel protein amounts (L-type
Ca2+ channel, r=-0.73; Kir3.1, r=-0.75; Kv1.5, r=-0.74 and minK, r=-0.79, all p<0.001),
the degree of structural changes (r=0.9, p<0.001) duration of AERP (BCL 500 ms, r=-0.6,
p<0.001) and rate adaptation of AERP (r=-0.80, p<0.001). Conclusions: Calpain activity
is induced during AF and correlates with parameters of ion-channel protein, structural
and electrical remodeling. The results strongly suggest that calpain activation represents
an important mechanism linking calcium overload to cellular adaptation mechanisms in
human AF.
Introduction
Atrial fibrillation (AF) is characterized by a heterogenic electrical activation pattern
and the loss of contractile function of atrial tissue. Clinical and animal experimental studies
indicate that atrial fibrillation has a strong tendency to promote itself.1,2 Research has been
directed at obtaining insight in the underlying mechanisms. Electrophysiological studies
have identified important factors promoting vulnerability to AF, including shortening of
atrial effective refractory period (AERP) and reduction of its rate adaptation.1,3,4 Recent
studies examining function and expression of various plasma membrane ion-channels
indicate that a reduction in the expression of the L-type Ca2+ channel plays a crucial role in
the electrical remodeling during AF.5-8 Interestingly, AF-induced electrical remodeling is
attenuated by application of calcium channel antagonists during the induction of the
arrhythmia.9,10 Therefore, the down-regulation of the L-type Ca2+ channel most likely
represents a feedback mechanism in response to the early calcium overload observed in
AF11, however, at the expense of electrophysiological changes promoting its maintenance.
Although electrical remodeling favors arrhythmia maintenance additional factors
increasing the vulnerability to AF seem to be involved, as electrical remodeling is
accomplished well in advance of the AF-promoting effect.1,12 Structural changes13 of atrial
tissue represent a likely candidate, as this could lead to conduction slowing3,12,14 and spatial
AERP heterogeneity15,16 thereby increasing the number of functional re-entry circuits in
the atrium.
The cellular mechanisms underlying the remodeling processes are not well
characterized. Two observations suggest that activation of a proteolytic pathway may
mediate the molecular changes underlying the AF-induced remodeling processes. Firstly,
we observed a discrepancy between changes in protein and mRNA expression of plasma
membrane ion channels in patients with lone paroxysmal AF.17 Whereas protein levels of
L-type Ca2+ channel, Kv1.5, Kir3.1 and minK were substantially decreased, the mRNA
levels were essentially unaffected in paroxysmal AF, suggesting activation of a protein
degradation mechanism at the post-transcriptional level. Secondly, we found increased
123
Calpain activity is related to ion-channel, structural and electrical remodeling
proteolysis in atrial tissue of patients with paroxysmal and persistent lone AF, which was
mainly due to activation of the calcium-activated neutral protease, calpain I (Brundel et al.
submitted).
In this study we sought to further explore the potential role of calpain activation in
the cellular adaptation mechanisms underlying AF. Therefore, we first localized the calpain
I protein in the atrial tissue by immunohistochemistry. Furthermore, we determined ion-
channel protein amounts, structural changes, AERPs and rate adaptation of AERP in atrial
tissue of AF patients and correlated these with the activity of calpain I.
Materials and Methods
Patients and electrophysiological measurements
RAAs were obtained as described before.7 In short, patients with paroxysmal (n=7)
and persistent (n=10) lone AF undergoing MAZE surgery were included and compared to
clinically stable control patients in sinus rhythm undergoing coronary bypass grafting
(CABG, n=10, Table 1). During surgery, the AERPs were determined in 8 patients with
persistent AF, 7 with paroxysmal AF and 8 control patients, with use of temporary epicardial
pacing leads. AERPs were measured at five different basic cycle lengths (BCL: 600, 500,
400, 300 and 250 ms) at the RAA using programmed electrical stimulation (Table 2). To
quantify the change in AERP at the different BCLs, we calculated the rate adaptation
Table 1. Baseline characteristics of patients with lone paroxysmal AF (PAF), lone persistent AF (CAF)
and control patients in sinus rhythm (SR).
SR PAF CAF
N 10 7 10
Age 60±7 50±7 53±8
Previous duration of AF (median, range (months) - - 12.5 (3-56)
Duration of SR before surgery (median, range (days) - 2 (0.5-12) -
Duration of last paroxysm (median, range (hours) - 9 (1-24) -
Surgical procedure
•CABG 10 0 0
•MAZE 0 7 10
Exercise tolerance
•NYHA Class I 10 6 7
•NYHA Class II 0 1 3
Medication
Digitalis 0 1 2
Calcium antagonists 4 2 2
Beta blockers 6 1* 1*
* p<0.05 compared to SR
124
Chapter 9
coefficient for individual patients as the slope of the linear regression after logarithmic
transformation of BCL.
Although AF groups and their controls differed with respect to the underlying heart
disease, all had normal left ventricular function (Table 1). The Institutional Review Board
approved of the study and all patients gave written informed consent.
Protein Extraction
For analysis of calpain activity, frozen RAAs were homogenized in buffer (100 mM
Tris-HCl, 145 mM NaCl, pH=7.3) and centrifuged at 26.000 x g (30 min, 4°C). For slot-
blotting analysis parts of the same RAAs were homogenized in Radio-
ImmunoPrecipitationAssay (RIPA) buffer.7 Protein concentration was determined using
the Bio-Rad DC assay (Bio-Rad, The Netherlands) with a bovine albumin standard.
Calpain measurements
Calpain activity was measured as described previously (Brundel et al. submitted).
Immunohistochemistry was performed for six patients with paroxysmal AF, eight with
persistent AF and eight control patients in sinus rhythm. Two groups of 5 µm thick frozen
RAA sections were made. One group was only air dried before use, whereas in a second
group the sections were immediately fixed for 10 minutes in 4% paraformaldehyde (in
PBS). After three times washing in PBS for 10 minutes and 30 minutes blocking with 1%
BSA in PBS, all sections were incubated with anti-calpain I antibodies (1:100) (Research
Diagnostics, USA) overnight at room temperature. After washing the sections three times
for 10 minutes with PBS they were incubated with secondary antibody peroxidase
conjugated rabbit-anti-mouse IgG (DAKO A/S, Glostrup, Denmark) for one hour. After
three times washing with PBS for 10 minutes, peroxidase activity was detected with 3-
amino-9-ethylcarbazole (AEC, Sigma, The Netherlands). To this end, 40 mg of AEC was
dissolved in 10 ml N,N dimethylformamide (Merck, The Netherlands) and added to 190
Table 2. AERP measured at the different BCLs.
AERP (ms) correlation calpain activity
BCL (ms) SR PAF CAF r p
600 291±53 222±15* 208±39** -0.52 0.007
500 277±42 224±24* 207±29** -0.6 <0.001
400 252±34 216±24* 203±25** -0.55 0.001
300 224±16 202±20 189±24* -0.54 0.002
250 184±5 185±19 172±17 -0.33 NS
*, p<0.05; **, p<0.01; NS, not significant
125
Calpain activity is related to ion-channel, structural and electrical remodeling
ml of 0.05 M sodium actetate buffer (pH 4.95). Hydrogen peroxide was added to a final
concentration of 0.01% (v/v). After 10 minutes staining, the sections were rinsed with
water, counterstained with haematoxylin (Sigma, The Netherlands) and mounted with
Kaiser’s glycerol gelatin (Merck, The Netherlands).
Slot-Blot Analysis
Protein expression was determined by slot-blot analysis and expressed as ratio to
levels of GAPDH as described previously.7 Denatured protein (10 µg) was spotted on
nitrocellulose membranes (Bio-Rad, The Netherlands) and incubated with primary
antibodies against GAPDH (Affinity Reagents, USA), L-type calcium channel α1 subunit,
minK, Kir3.1 and Kv1.5 (Alomone Labs, Israel). The specificity of antibodies was
confirmed by preincubation with control peptide antigen and by SDS-PAGE. Anti-mouse
IgG (Santa-Cruz Biotechnology, The Netherlands) was used as secondary antibody. Signals
were detected by the ECL-detection method (Amersham, The Netherlands) and quantified
by densitometry. GAPDH levels did not differ between the groups (sinus rhythm: 1082 ±292, paroxysmal AF: 1190 ± 181; persistent AF: 1177 ± 222, arbitrary OD units).
Morphological evaluation
For morphological evaluation by light microscopy, a biopsy was taken from all the
RAAs and immediately fixed for at least 2 hours at 4°C in 2% glutaraldehyde (in 0.1 M
cacodylate buffer, pH 7.4). Post-fixation was performed for 2 hours in 1% osmium tetroxide
(supplemented with 1.5% K4Fe(CN)
6 in cacodylate buffer, pH 7.4) at 4°C. After dehydration
in ethanol, biopsies were embedded in Epon and semi-thin sections (1 µm) were cut and
stained with 1% toluidine blue. The degree of cellular changes was evaluated by light
microscopy in cells where the nucleus was visible in the plane of the section. To quantify
the structural changes a minimum of 300 cells from six randomly chosen regions of the
RAA were evaluated by an investigator blinded for patient groups. Two types of structural
changes were present in our biopsies, i.e. hibernation or degeneration of atrial myocytes.
Hibernating cells show areas with loss of sarcomeres and contain pale nuclei. An atrial
myocyte was defined as hibernating when >10% of the cell surface was free from
sarcomeres.18 Degenerating cells display the following characteristics: contraction band
necrosis, pyknotic nuclei with intense stained chromatin, secondary lysosomal structures
(inclusion bodies) and/or vacuoles. Myocytes were scored positive for degeneration if
>10% of the sarcomere surface contained intensely stained contraction bands.19 To verify
the structural changes on the ultrastructural level, electron microscopy was performed.
Ultrathin sections (60 nm) of the RAAs were stained with uranylacetate and lead citrate
and examined in a Philips 201 electron microscope operating at 60 kV.
126
Chapter 9
Figure 1.
Immunohistochemical localization of calpain I in myocytes. (A) at the nucleus (arrow) and (B) at the intercalated
discs (arrow) (magnification A x 400, B x 300).
Statistical Methods
Results are expressed as mean ± SEM. Parametric and non-parametric ANOVA
(Kruskal-Wallis test) were used for multiple group comparisons. Correlation was determined
using the Spearman correlation test. P<0.05 was considered statistically significant. SPSS
version 8.0 was used for all statistical evaluations.
Results
Calpain activity and localization
Calpain activity was significantly increased in tissue from patients with paroxysmal
and persistent AF compared to patients in sinus rhythm (Table 3). Both the non-selective
calpain inhibitor E64 and calpain I inhibitor significantly reduced tissue calpain activity to
a similar level in all groups.
Localization of calpain I was performed by immunohistochemical staining. Irrespective
of the fixation method used, calpain I was predominantly localized in the atrial myocytes
and to a less extend in interstitial cells. In non-fixed sections, calpain I was detected both
at the nucleus and in the cytoplasm (Figure 1A), whereas calpain I was located at the
intercalated discs and in the cytoplasm in paraformaldehyde fixed sections (Figure 1B).
A B
Table 3. Calpain activity (nM AMC/mg protein/30min) in atrial tissue of patients with lone paroxysmal
AF (PAF), lone persistent AF (CAF) and control patients in sinus rhythm (SR).
SR PAF CAF
None 33 ± 10 53 ± 13* 58 ± 17*
E-64 (10-4 M) 19 ± 4 22 ± 4 23 ± 6
Calpain I inhibitor (10-4 M) 23 ± 3 24 ± 3 25 ± 2
* p<0.05 compared to SR
127
Calpain activity is related to ion-channel, structural and electrical remodeling
Figure 2.
Significant correlation between tissue
calpain activity and expression of vari-
ous plasma membrane ion-channels.
The top of each panel shows a typical
slot blot analysis of 10 µg protein of
control patients (SR), paroxysmal AF
(PAF) and persistent AF (CAF). The
immunoblots were done for (A) anti-
L-type calcium channel, (B) anti-
Kir3.1, (C) anti-Kv1.5 and (D) anti-
minK. ( ) indicates patients with PAF,
(•) CAF and (o) SR.
Kir3.1/GAPDH0 1 2 3
,
-100
10203040
L-type Ca2+ channel/GAPDH0 1 2 3calpain activity (nM AMC/mg)
-100
10203040
Kv1.5/GAPDH0 1 2
,
-100
10203040
minK/GAPDH0 1 2,
-100
10203040
r=-0.73, p<0.001 r=-0.75, p<0.001
r=-0.74, p<0.001 r=-0.79, p<0.001
SR PAFCAF
SR PAFCAF
PAF PAFCAFCAF
SR SR
A B
C D
Calp
ain
acti
vit
y (
nM
AM
C/m
g)
Figure 3.
(A) Atrial myocardium from a patient in sinus rhythm stained with toluidine blue showing normal structural
myocytes without myolysis and degenerative features. (B) electron microscopic detail of a normal atrial myo-
cyte without myolysis and normal nucleus. (C) Atrial myocytes from a patient show degenerative changes as
contraction band necrosis (arrows) (D) Electron microscopic detail of an atrial myoctye from a patient with PAF
showing degenerative featrues as clumping of nuclear chromatin and contraction band necrosis (arrows). (E)
Structural changes observed after persistent AF. Extensive myolysis is present perinuclear and pale nuclei were
observed. (F) Electron microscopy of a persistent myocyte showing myolysis. In this hibernating atrial myocyte
only a rim of sarcomeres is present at the border of the cell. Glycogen (g) is visual and the nucleus has a homog-
enous dispersion of chromatin. (Magnification A, C and E x 250, B x 7000, D and F x 4500)
g
A
B
C
D
E
F
128
Chapter 9
Table 4. Immunohistochemical localization of calpain I in myocytes. Intensity of the calpain I staining in
patients with paroxysmal (PAF, n=6), chronic persistent AF (CAF, n=8) and sinus rhythm (SR, n=8).
Intensity calpain I staining
SR PAF CAF
Localization -/+ + ++ +++ - + ++ +++ P - + ++ +++ P
Intercalated discs 5 3 0 0 0 2 4 0 0.04 0 0 2 6 <0.01
Nucleus 2 4 2 0 0 1 4 1 0.04 0 0 3 5 0.02
Cytoplasm 0 7 1 0 0 4 2 0 NS 0 8 0 0 NS
+/- low staining - +++ intense staining
To obtain an indication about the amount of calpain I protein in the myocytes, the intensity
of staining was scored semi-quantitatively (Table 4). The intensity of staining of the nucleus
and the intercalated discs increased significantly from sinus rhythm, via paroxysmal AF to
reach a maximum in persistent AF. In contrast, the intensity of staining in the cytoplasm
was not different between the groups.
Ion-channel Proteins
To examine the relation between tissue calpain activity and ion-channel protein
expression, protein levels of L-type calcium channel, Kir3.1, Kv1.5 and minK
weredetermined by slot-blot analysis (Figure 2). Protein levels of these ion-channels were
all significantly reduced in patients with AF as compared to controls (Table 5).
Importantly, a significant negative correlation was observed between tissue calpain
activity and protein levels of all ion-channels examined (L-type Ca2+ channel, r=-0.73;
Kir3.1, r=-0.75; Kv1.5, r=-0.74 and minK, r=-0.79, all p<0.001, Figure 2). In contrast, no
correlation was found between GAPDH densities and tissue calpain activity (r=0.32,
p>0.05).
Structural changes
The amount of structural and ultra-structural changes in the atrial tissue was examined
by light microscopy and electron microscopy, respectively. Sections were scored for definite
signs of cellular degeneration or hibernation (see Materials and Methods). Atrial
myocardium of sinus rhythm patients showed mainly normal structured myocytes without
myolysis and degenerative features (Figure 3A). At the ultrastructural level, these control
patients showed a highly organized sarcomeric structure with mitochondria in between. A
129
Calpain activity is related to ion-channel, structural and electrical remodeling
typical distribution of heterochromatin in the form of clusters mainly at the cardiomyocyte
nuclear membrane was observed (Figure 3B). In contrast, contraction band necrosis was
abundantly present in patients with paroxysmal AF (Figure 3C). At the ultrastructural
level, their myocytes showed contraction band necrosis and pyknotic nuclei (Figure 3D).
Patients with persistent AF showed contracting myocytes with band necrosis or hibernation
(Figure 3E). Hibernating cells were only extensively present in patients with persistent AF
(Figure 3F).
The number of myocytes with contraction band necrosis was increased 3-fold both in
patients with paroxysmal and persistent AF compared to patients in sinus rhythm (Figure
4A). However, the amount of hibernating cells was only significantly increased in patients
with persistent AF compared to the other groups (Figure 4A). Furthermore, a good
correlation was found between the calpain activity and the total number of affected cells
(either contraction band necrosis or hibernation; r=0.71, p<0.001; Figure 4B). Thus,
atrialtissue of patients with increased calpain activity showed an increased number of
myocytes with structural changes.
When the type of structural change, i.e. contraction band necrosis or hibernation, in
patients with persistent AF was plotted against the duration of AF an intriguing relationship
was revealed (Figure 4C). Atrial myocytes of patients with the shortest duration of AF (<
10 months) showed high amounts of contraction band necrosis and low amounts of
hibernating cells, whereas the opposite pattern was found in patients with the longest
duration of AF (> 10 months).
AERP and Rate adaptation
AERPs were measured at five BCLs (600, 500, 400, 300 and 250 ms). Patients with
persistent and paroxysmal AF had significantly shorter AERPs than patients in sinus rhythm
(Table 2). Calpain activity showed significant negative correlations with AERP measured
at BCL 600, 500, 400 and 300ms (Figure 5A, Table 2).
Table 5. Ion-channel protein amounts in patients with paroxysmal (PAF) and chronic persistent AF (CAF)
and patients in sinus rhythm (SR).
Protein (ratio to GAPDH, arbitrary OD units) SR PAF CAF
L-type calcium channel 1.21 ± 0.13 0.63 ± 0.11* 0.55 ± 0.07*
Kv1.5 1.18 ± 0.14 0.67 ± 0.13* 0.60 ± 0.11*
Kir3.1 1.30 ± 0.15 0.80 ± 0.11* 0.63 ± 0.10*
MinK 0.83 ± 0.07 0.51 ± 0.03* 0.49 ± 0.03*
130
Chapter 9
Figure 4.
(A) Percentage structural changes of atrial cells affeted by contaction band necrosis or hibernation in patients
with sinus rhythm (SR), paroysmal AF (PAF) and chronic, persistent AF (CAF). (B) Significant correlation
between the total amount of affected cells and calpain activity. (C) Significant correlation between the type of
structural change and the duration of persistent AF. (•) contraction band necrosis and ( ) hibernation.
tissue calpain activity (nM AMC/mg/30 min)0 5 10 15 20 25 30 35 40
% affected cells
020406080
r=+/- 0.9, p<0.001
SR PAF CAF
% affected cells
0102030405060
* * #= contraction band necrosis= hibernation
duration of persistent AF (months)0 10 20 30 40 50
% affected
0
20
40
60
r=0.71, p<0.001
r= +/- 0.9, p<0.001
A
B
C
% a
ffecte
d c
ell
s%
aff
ecte
d c
ell
s%
aff
ecte
d c
ell
s
From the AERP obtained at different basal cycle lengths, the rate adaptation coefficient
was calculated. The rate adaptation coefficient was significantly reduced by 36% in
persistent AF compared to sinus rhythm (persistent AF: 95 ± 13, paroxysmal AF: 121 ± 9
and sinus rhythm: 148 ± 9), indicating a poorer adaptation to higher heart rates in patients
with AF. Furthermore, a significant negative correlation was observed between the calpain
activity and the rate adaptation coefficient (r=-0.80, p<0.001, Figure 5B).
131
Calpain activity is related to ion-channel, structural and electrical remodeling
Discussion
In this study we examined a variety of changes in atrial tissue from patients with
paroxysmal and persistent lone AF and related them to the level of calpain activity.
Immunohistochemical detection of calpain I demonstrated increased staining at the
intercalated disk and in the nucleus of atrial myocytes of AF patients, but not in the cytosol.
Accordingly, calpain activity was increased in patients with AF. Furthermore, an increased
number of degenerative myocytes was observed in both patient groups with AF. Hibernating
myocytes were only present in persistent AF and numbers increased with the duration of
AF. Finally, calpain activity correlated with the expression levels of ion-channel proteins,
the degree of structural changes, duration of AERP and the rate adaptation coefficient of
AERP. This study strongly indicates that calpain activation represents an important adaptive
molecular mechanism occurring in human AF. Consequently, interference with the calpain
pathway may both represent an important tool to investigate molecular events in AF, and
a possible future alternative for pharmacological intervention.
calpain activity (nM AMC/mg/30 min)0 10 20 30 40 50 60
adaptation coef
50
100
150
200
r=-0.80, p<0.001
calpain activity (nM AMC/mg/30 min)0 10 20 30 40 50 60
AERP (ms, BCL 500 ms)
150
200
250
300
r=-0.6, p<0.001
A
B
AE
RP
(m
s, B
CL
500 m
s)
adepta
tion c
oeff
icie
nt
Figure 5.
(A) Significant correlation between AERP
measured at BCL 500 ms and tissue calpain
activity. (B) Significant correlation between
rate adaptation coefficient and tissue
calpain activity. ( ) indicates patients with
PAF, (•) CAF and (o) SR.
132
Chapter 9
Localization of calpain I
Previously, we demonstrated induction of the calpain activity in atrial tissue of patients
with paroxysmal and persistent lone AF, mainly due to increased activation of the calpain
I protein (Brundel et al. submitted). Now we found that calpain I was localized in the
cytosol, intercalated discs and nucleus of atrial myocytes. Its intensity at the intercalated
discs and nucleus was increased in patients with AF, in accordance with the increased
calpain activity measured in AF patients. The change in cellular distribution in AF suggests
that calpain I activation mediates underlying molecular changes. Under normal conditions,
calpain is localized diffusely in the cytosol. After an increase in intracellular calcium,
calpain rapidly translocates to the inner surface of the plasma membrane, aggregates at the
intercalated discs, which is followed by its activation.20 At the intercalated discs, activated
calpain I may degrade important ion-channels, like the Na+-channel21 and Kv1.522, but
also proteins involved in excitation-contraction coupling23 and conduction.13 The increased
expression of calpain I at the nucleus is in agreement with its role in promoting necrosis
and apoptosis24, suggesting that calpain I could play a role in the degeneration observed in
AF.25 Thus, the increased calpain I expression in AF seems confined to cellular areas that
are vital for action potential conduction and structural integrity of the atrial myocytes.
Structural changes in myocytes
We systematically analyzed the amount of myocytes affected by degeneration and
hibernating in microscopic sections of right atrial appendages. Increased degeneration
was observed both in patients with persistent and paroxysmal AF, as evidenced by an
equal increase in contraction band necrosis in both groups. In contrast, hibernating cells
were only increased in patients with persistent AF. The latter is in agreement with
observations in an experimental goat model for AF, in which hibernation develops only at
a relatively late stage of the arrhythmia.18 Thus, these findings suggest that degeneration
represents an early structural abnormality in human lone AF, whereas hibernation is
dependent on protracted periods of sustained AF. Similar alterations in myocardial structure
were described in patients with atrial arrhythmias of various aetiology.26 In adition to
degenerative changes, part of the myocytes showed loss of myofibrils and presence of
glycogen granules as in our human study.
Further, in patients with persistent AF, hibernation increased with the duration of AF,
while degeneration decreased. Possibly, this is related tothis is due to a protection from
degeneration by the hibernation, as described in ischemic preconditioning.27 Taken together,
the observed structural changes are indicative of a substantial deterioration of normal
tissue architecture, likely to promote AF through heterogeneity of atrial refractoriness15,16
and slowed atrial conduction.3,12,14
133
Calpain activity is related to ion-channel, structural and electrical remodeling
Interpretation of correlation
Our data demonstrate that human paroxysmal and persistent lone AF is accompanied
by substantial electrical28, structural and ion-channel protein17 remodeling, as indicated by
shortening of AERP, decreased rate adaptation, increased numbers of cells showing
degeneration or hibernation and decreased expression of four ion-channel proteins. These
parameters all correlated significantly with calpain activity measured in the atrial tissue.
Accordingly, these parameters generally show a good correlation with each other. Therefore,
the question remains whether calpain activation causes the molecular changes as observed
in AF, or merely reflects gross cellular damage. There are strong indications that calpain
causes the molecular changes. First as cytosolic calcium is rapidly increased during AF11,
increased calpain activation may occur already early after AF. Secondly, calpain I and not
calpain II or proteasome showed increased activity (Brundel et al. submitted). Third, this
increased calpain activity was due to elevation of calpain I and not calpain II protein
expression. Finally we found increased staining intensity of calpain I at vital areas of atrial
myocytes. One can speculate that the rapidly induced calpain acitivity immediately
proteolyse the ion-channel proteins leading to the loss of physiological rate adaptation and
structural changes. In this way the fibrillating atria can no longer be expected to reverse
their action potentials when sinus rhythm is restored and this may explain the vulnerability
to AF.
Possible clinical relevance
Successful chemical cardioversion and maintenance of sinus rhythm after
cardioversion is dependent on the duration of AF.29 This clinically observed diminished
efficacy of chemical cardioversion after long term AF cannot only be explained by the
occurrence of electrical remodeling. The protein remodeling and structural remodeling
could also affect the success rate of cardioversion. In patients with persistent AF, there is
a correlation between the duration of AF before cardioversion and the time needed to
recover atrial contractile function thereafter.30 The increase in calpain activity and correlated
structural remodeling of the atrial myocytes might give an explanation for this delay in
recovery of atria contractile function after conversion to sinus rhythm. After restoration of
normal sinus rhythm it may take the cardiomyocytes a certain period to rebuild a normal
amount of sarcomeres, if this is possible at all.31 Our study implicates that it is necessary to
bring AF as soon as possible back to sinus rhythm, thereby preventing the continuation of
the atrial structural, ion-channel protein and electrical remodeling.
Acknowledgements
The authors would like to thank Bert Blaauw and Hans Duimel for their excellent
technical assistance.
134
Chapter 9
References
1. Wijffels MC, Kirchhof CJ, Dorland R, et al. Atrial fibrillation begets atrial fibrillation. A study in awake
chronically instrumented goats. Circulation 1995; 92:1954-1968.
2. Godtfredsen J. Etiology, course and prognosis. A follow-up study of 1212 cases. Copenhagen: University
of Copenhagen. Thesis 1975;
3. Morillo CA, Klein GJ, Jones D, et al. Chronic rapid atrial pacing. Structural, functional, and
electrophysiological characteristics of a new model of sustained atrial fibrillation. Circulation 1995; 91:1588-
1595.
4. Daoud EG, Bogun F, Goyal R, et al. Effects of atrial fibrillation on atrial refractoriness in humans. Circu-
lation 1996; 94:1600-1606.
5. Yue L, Feng J, Gaspo R, et al. Ionic remodeling underlying action potential changes in a canine model of
atrial fibrillation. Circ Res 1997; 81:512-525.
6. Van Wagoner DR, Pond AL, Lamorgese M, et al. Atrial L-Type Ca2+ Currents and Human Atrial Fibrilla-
tion. Circ Res 1999; 85:428-436.
7. Brundel BJJM, Van Gelder IC, Henning RH, et al. Gene expression of proteins influencing the calcium
homeostasis in patients with persistent and paroxysmal atrial fibrillation. Cardiovasc Res 1999; 42:443-
454.
8. Bosch RF, Zeng X., Grammer JB, et al. Ionic mechanisms of electrical remodeling in human atrial fibril-
lation. Cardiovasc Res 1999; 44:121-131.
9. Tieleman RG, De Langen CDJ, Van Gelder IC, et al. Verapamil reduces tachycardia-induced electrical
remodeling of the atria. Circulation 1997; 95:1945-1953.
10. Goette A, Honeycutt C, Langberg JJ. Electrical remodeling in atrial fibrillation. Time course and mecha-
nisms. Circulation 1996; 94:2968-2974.
11. Ausma J, Dispersyn GD, Duimel H, et al. Changes in ultrastructural calcium distribution in goat atria
during atrial fibrillation. J Mol Cell Cardiology 2000; 32:355-364.
12. Gaspo R, Bosch RF, Talajic M, et al. Functional mechanisms underlying tachycardia-induced sustained
atrial fibrillation in a chronic dog model. Circulation 1997; 96:4027-4035.
13. Van der Velden HMW, Ausma J, Rook MB, et al. Gap junctional remodeling in relation to stabilization of
atrial fibrillation in the goat. Cardiovasc Res 2000; 46:476-486.
14. Elvan A, Wylie K, Zipes DP. Pacing-induced chronic atrial fibrillation impairs sinus node function in dogs:
electrophysiological remodeling. Circulation 1996; 94:2953-2960.
15. Ramanna H, Hauer RNW, Wittkampf FHM, et al. Identification of the substrate of atrial vulnerability in
patients with idiopathic atrial fibrillation. Circulation 2000; 101:995-1001.
16. Fareh S, Villemaire C, Nattel S. Importance of refractoriness heterogeneity in the enhanced vulnerability
to atrial fibrillation induction caused by tachycardia-induced atrial electrical remodeling. Circulation 1998;
83:2202-2209.
17. Brundel BJJM, Van Gelder IC, Henning RH, et al. Alterations in potassium channel gene expression in
atria of patients with persistent and paroxysmal atrial fibrillation. J Am Coll Cardiol 2000; in press:
18. Ausma J, Wijffels M, Thone F, et al. Structural changes of atrial myocardium due to sustained atrial
fibrillation in the goat. Circulation 1997; 96:3157-3163.
19. Borgers M, Guo Shu L, Xhonneux R, et al. Changes in ultrastructure and Ca2+ distribution in the isolated
working rabbit heart after ischemia. A time-related study. Am J Pathol 1987; 126:92-102.
20. De Tullio R, Passalacqua M, Averna, et al. Changes in intracellular localization of calpastatin during
calpain activation. Biochem J 1999; 343:467-472.
21. Cohen SA. Immunocytochemical localization of rH1 sodium channel in adult rat heart atria and ventricle.
Presence in terminal intercalated disks. Circulation 1996; 94:3083-3086.
22. Mays DJ, Foose JM, Philipson LH, et al. Localization of the Kv1.5 K+ channel protein in explanted
cardiac tissue. J.Clin.Invest. 1995; 96:282-292.
23. Laflamme MA, Becker PL. G(s) and adenylylcyclase in transverse tubules of heart: implications for cAMP-
dependent signaling. Am.J.Physiol. 1999; 277:H1841-H1848
24. Yuen PW, Wang KKW. Calpain inhibitors, novel neuroprotectants and potential anticataractic agents.
Drugs Future 1998; 23:741-749.
25. Aime-Sempe C, Folliguet T, Rucker-Martin, et al. Myocardial cell death in fibrillating and dilated human
right atria. J Am Coll Cardiol 1999; 34:1577-1586.
26. Mary-Rabine L, Pham TD, Hordof A, et al. The relationship of human atrial cellular electrophysiology to
clinical function and ultrastructure. Circ.Res. 1983; 52:188-199.
27. Tanaka M, Fujiwara H, Yamasaki K, et al. Expression of heat shock protein after ischemic precondition-
ing
135
Calpain activity is related to ion-channel, structural and electrical remodeling
in rabbit heart. Jpn Circ J 1998; 62:512-516.
28. Tieleman RG, Van Gelder IC, Tuinenburg AE, et al. Intra- and post-operative atrial refractory periods in
relation to atrial arrhythmia history and the presence of mitral regurgitation. Circulation 1999; 100:I-361
29. Van Gelder IC, Crijns HJGM, Tieleman RG, et al. Value and limitation of electrical cardioversion in
patients with chronic atrial fibrillation - importance of arrhythmia risk factors and oral anticoagulation.
Arch Intern Med 1996; 156:2585-2592.
30. Manning WJ, Silverman DI, Katz SE, et al. Impaired left atrial mechanical function after cardioversion:
relation to the duration of atrial fibrillation. J Am Coll Cardiol 1994; 23:1535-1540.
31. Ausma J, Duimel H, Wouters L, et al. Structural atrial changes induced in the goat by 16 weeks of atrial
fibrillation are still present 8 weeks after cardioversion. Europace 2000; 1:B12
137
General discussion
General Discussion
Calcium Homeostasis
Potassium Channels
L-type Calcium Channel
Neurohumoral Changes
Post-Transcriptional Regulations
Integrated Model for Remodeling Processes in AF
Evidence for calcium overload
Time course structural remodeling
Electrophysiological properties
Future Perspectives
Pre-conditioning
Possible clinical relevance
New experimental model for AF
Calcium Homeostasis
Studies showed that AF has the tendency to become more persistent over time. A
large percentage of patients with paroxysmal AF will develop persistent AF.1 Also,
pharmacological and electrical cardioversion and maintenance of sinus rhythm thereafter
become more difficult the longer the arrhythmia exists.2 Therefore it is important to study
the underlying mechanisms which play a role in the vulnerability to AF.
Experimental studies showed that electrical and contractile remodeling occur early
after the onset of AF (Figure 1A).3-5 Both remodeling processes were attenuated by blocking
of the L-type Ca2+ channel indicating that changes in the calcium homeostasis triggered by
tachycardia induced intracellular calcium overload, play a pivotal role in the induction of
atrial electrical remodeling and contractile dysfunction.
We first investigated the AF induced contractile dysfunction by studying the molecular
remodeling of proteins which influence calcium homeostasis (Chapter 2 and 3). The main
finding was reductions in mRNA and protein expression of the L-type calcium channel
and sarcoplasmatic reticulum Ca2+ ATPase (SR Ca2+ ATPase), predominantly in patients
with persistent AF. Also, an increased reduction in mRNA expression was found the longer
the duration of persistent AF existed. Patients with >6 months AF revealed reductions in
mRNA of L-type Ca2+ channel, in contrast to patients with <6 months duration of AF, in
whom no changes in mRNA expression were seen (Figure 1B and 2). This finding indicates
that changes in mRNA expression are the consequence rather than the cause of AF. The
results described in chapter 2 and 3 were confirmed by other studies which also showed
that the L-type Ca2+ channel6 and SR Ca2+ ATPase6,7 were both reduced in AF. Unfortunately,
138
Chapter 10
in one study no time dependent changes in mRNA expression could be investigated since
the duration of AF was not known.7 Both studies were limited by differences between AF
and controls with respect to the underlying heart disease, which could have influenced
mRNA expression. Another limitation was that in both studies only mRNA expression of
L-type Ca2+ channel and SR Ca2+ ATPase was measured and no protein levels, which are
anticipated to represent the amount of functional proteins more adequately.
Taken together these studies indicate that changes in gene expression of proteins
influencing the calcium homeostasis occur in persistent AF. These changes probably are a
contributory factor for the atrial contractile dysfunction in AF.
Figure 1A.
Overview AF induced adaptations in experimental studies
Figure 1B.
Overview AF induced adaptations in human AF
Electrical remodeling/contractile dysfunction
Changes in ion channel protein amounts
Structural changes (myolysis)
0 1 10 50 weeks
Start AF
Changes mRNA expression
Structural changes (degeneration)
Start induction calpain activity?
0 1 10 100 days
Start inductionof AF bypacing
Electrical remodeling / Contractile dysfunction
Functional changes ion channels
Changes in ion channel protein amounts
Calcium overload
Start induction calpain activity?
Structural changes (myolysis)
Changes in mRNA expression
Persistent AF
A
B
139
General discussion
Ion-channel remodeling related to atrial refractory periods
It is well known that an abrupt increase in heart frequency, like in AF, causes an
immediate (within one action potential) and then a gradual (reaching steady state over
several minutes) decrease in action potential duration (APD).8 These alterations in APD
reduce atrial effective refractory period (AERP) and shorten the wavelength for reentry,
which will facilitate the occurrence and maintenance of reentrant arrhythmias like AF.
The rapid nature of these changes suggests that the short-term APD adaptation to rate is
due to functional changes in ion channels. With longer periods of sustained atrial
tachycardia, changes develop over the course of hours to days.4,9,10 The latter alterations
appear to concern mainly ion channel density and are due to modified gene expression
(Figure 1A,B).11,12 Most studies in changes of ion-channel protein expression have been
performed in animal experimental settings and gradually, studies have revealed that the
rapid shortening of the AERP in animal experimental AF mainly involves functional changes
in the L-type Ca2+ channel.11,12 In human AF the relationship between changes in AERP
and ion channel gene expression has not been investigated previously. We studied the
regulation of L-type Ca2+ channel and K+ channels and their relation to AERP in patients
with persistent and paroxysmal AF (Chapter 5). We demonstrated a positive correlation
between the ion-channel protein expression of L-type Ca2+ channel, Kv4.3, Kv1.5, HERG,
minK and Kir3.1 and the AERP but also with the rate adaptation to AERP in patients with
persistent and paroxysmal AF. Low ion-channel protein levels were associated with short
AERP and poor rate adaptation. This indicates that electrical remodeling is paralleled by
general ion-channel protein reductions as part of the adaptation mechanisms during AF.
Since reduced ion-channel protein expression occurred due to AF we called this
phenomenon ion-channel remodeling. The ion-channel protein remodeling could play an
important role in the susceptibility to AF after restoration of sinus rhythm. Since shortening
of AERP can be explained by decrease in L-type Ca2+ channel and increase in K+ channel
gene expression or activity, the reductions in L-type Ca2+ channel could represent an
explanation for the electrophysiological changes during AF.
As noted the shortening in AERP can also be explained by an increase in K+ channel
activity and expression, we investigated the contribution of potassium channels in
paroxysmal and persistent AF (Chapter 4 and 5). Reductions in mRNA and protein levels
were found for several K+ channels in patients with persistent AF. In patients with
paroxysmal AF these reductions were observed predominantly at the protein level and not
at the mRNA level, suggesting the activation of a proteolytic system. The reductions in
mRNA and protein amount of K+ channels do not explain the shortening of AERP, however
other studies have also reported the decrease in K+ channels in AF.13 One study found
increased IKACh
and IK1
in isolated human atrial cells of patients with persistent AF due to
different underlying heart diseases.14 The apparent inconsistency between protein levels
140
Chapter 10
and current density can only be explained by assuming a change in single channel properties
in patients with persistent AF, such as an increase of mean open-time and increase in
channel conductance or a change in voltage dependency.
Unfortunately in human studies it is difficult to make a time course for remodeling
processes, but experimental data can elucidate the ion-channel remodeling in more detail.
Human and experimental studies showed that brief periods of experimental AF (<1 h)
abbreviate AERP and favor AF induction via functional changes, including Ca2+ overload
induced L-type Ca2+ current (ICaL
) inactivation, that cause APD shortening (Figure 1A,B).15,16
With longer periods of sustained atrial tachycardia adaptations appear to involve alterations
mainly in ion channel density that are due to modified gene expression (Figure 1A,B and
2).11,12 An examination of ionic current changes in atrial myocytes from dogs subjected to
rapid atrial pacing for 7 and 42 days 11,12 indicated that high rate stimulation of atrial
myocytes does not change a variety of currents, including inward and delayed rectifier K+
currents, T-type Ca2+ current and Ca2+ dependent Cl- current. Currents that show important
alterations were the transient outward K+ current (ITo
) and L-type Ca2+ current (ICaL
), both
of which are reduced by about 70% after 6 weeks of rapid atrial pacing due to reductions
in protein amount.12,17 Other properties of ICaL
and ITo
, like voltage, time and frequency
Figure 2. Overview L-type calcium channel alterations
141
General discussion
dependence are unchanged. This observation suggests that the changes observed are due
to a reduction in the number of functional channels in the membrane rather than to a
change in basic channel properties. The use of pharmacological probes to mimic the effects
of reduced ICaL
and ITo
on the action potential showed that reductions in ICaL
are likely to
play the central role in the APD alterations caused by atrial tachycardia with the changes
in ITo
being of much less importance,11 despite the quantitatively similar reduction.
Thus, experimental and human AF studies reported important observations concerning
ion-channel remodeling. Although a difference in time course was found between human
and experimental AF. Changes in mRNA expression were observed in animal experiments
around 1 week of AF, in human AF significant changes were observed only after > 6
months (Chapter 3) and >3 months6 (Figure 2), indicating that other factors play a role in
the adaptation mechanisms in human AF, for example preconditioning.
Furthermore in AF series of changes were found, involving rapid functional alterations
and slower changes in gene expression that cause APD reduction and reduced cellular
calcium loading. These changes can be considered to reduce ICaL
and thereby protect the
cell against potentially lethal Ca2+ overload resulting from an increase in rate of action
potential generation between resting sinus rhythm and AF. This protective effect occurs,
however, at the expense of electrophysiological changes that promote the maintenance of
AF.
Neurohumoral changes
The cardiac natriuretic peptide system and the endothelin system play an important
role in maintaining volume homeostasis especially in conditions that affect
hemodynamics.18,19 In Chapter 6 and 7 the local gene expression of these systems in atrial
tissue of patients with AF was studied. Persistent AF was associated with evident expression
of ANP and BNP mRNA and also endothelin-1 mRNA contents. The extent of these changes
was more pronounced in patients with concomitant valvular heart diseases, indicating that
these systems play a role in human AF and in particular in the presence of atrial pressure
or volume overload. Furthermore reductions were found in the protein expression of the
endothelin type A and B receptors during paroxysmal and persistent AF in contrast to
unchanged expression of mRNA amounts of these receptors suggesting post-transcriptional
regulation.
Post-transcriptional regulations
A remarkable finding during the study of mRNA and ion-channel protein remodeling
was a discrepancy between changes in mRNA and protein levels in patients with paroxysmal
AF (Chapter 4, 5 and 7). Whereas ion-channel protein levels of L-type Ca2+ channel, Kv1.5,
Kir3.1 and minK, but also ET-A were substantially decreased, the mRNA levels were
142
Chapter 10
essentially unaffected in paroxysmal AF. This discrepancy was also observed in other
studies20,21 and prompted us to explore the role of an adaptative mechanism of which the
influence in AF was unknown: the activation of a proteolytic system. Different proteolytic
pathways could be involved in AF. Since cytosolic calcium is increased during AF22,23,
proteolysis may be invoked by calcium dependent neutral proteases, calpain I and II.
Calpains are proteases which cleave mainly cytoskeletal and membrane-associated proteins
into ‘limited fragments’ without further degradation.24 In cardiac cells, calpains mediate
cell death in metabolically inhibited cultured rat cardiomyocytes and are involved in
troponin proteolysis and cross-linking following cardiac stunning and calcium overload.25-
27 In Chapter 8 an increased proteolytic activity in atrial tissue of patients with paroxysmal
and persistent lone AF was described. This increase was predominantly due to elevation
of calpain I activity and expression. Furthermore we observed that calpain I protein was
mainly localized at the nucleus and intercalated discs of atrial myocytes (Chapter 9). The
intensity of staining was low in sinus rhythm higher in paroxysmal AF and reached a
maximum in persistent AF. At the intercalated discs calpain can interact with Ca2+ and
thereby become an active proteinase and can degradate some important ion-channels like
the Na-channel28 and Kv1.519, but also several proteins directly involved in excitation-
contraction coupling.29 At the nucleus30 calpain can induce degenerative features leading
to apoptosis, which is observed in human AF (Figure 3).31
The role of calpain in cellular changes underlying the electrophysiological (Chapter
5), ion-channel (Chapter 5) and structural remodeling (Chapter 9) was examined. The
amount of structural and ultra-structural changes in the atrial tissue was examined by light
microscopy and electron microscopy, respectively. Calpain activity correlated with the
AF
calcium overload
induction calpain activity
degeneration proteins
necrosis/apoptosis
Figure 3.
143
General discussion
expression levels of ion-channel proteins, the degree of structural changes, measured AERP
and the rate adaptation coefficient of AERP. The results suggest that induction of calpain
activation represents a missing link between the calcium overload observed in AF22 and
remodeling of atrial myocytes during AF (Figure 3).
Integrated model for remodeling processes in AF
Evidence for AF induced Calcium overloadWhile studying the ion-channel gene expression adaptation mechanisms in human
AF, it became clear that the central feature in all these processes is calcium, in particular
calcium overload. Most investigations recognize the direct and indirect role of calcium
and our studies support the notion of calpain activation in AF. This notion is appealing
since it provides the molecular link between AF induced calcium overload and remodeling
processes, which was still not identified. Several lines of evidence for the crucial role of
AF induced calcium overload and subsequent calpain induction are reported below.
It has been proposed that atrial contractile dysfunction occurs after short-term and
chronic AF32-34 (Figure 1A,B). Contractile dysfunction after chronic AF is most likely
related to the cellular alterations in atrial myocytes35,36 reflected by structural alterations,
probably induced by proteolysis. The explanation for the atrial dysfunction after short-
term AF might be an increase in cytosolic Ca2+ due to the high rate of atrial activation. Fast
successive action potentials inhibit a proper sarcoplasmic reticulum Ca2+ re-uptake, resulting
in elevated cytosolic Ca2+, possibly impairing the excitation-contraction coupling and
contractile function.3,5,15,37,38 Ausma and coworkers showed that sarcolemma-bound Ca2+
and Ca2+ deposits in mitochondria increased markedly up to 2 weeks in experimental AF
and tends to regress towards normal levels at 4 and 8 weeks of AF (Figure 1A).22
Unfortunately, for Ca2+ localization they used antimonate based methods, which limit the
visualization of overall Ca2+ load at subcellular sites like the sarcoplasmic reticulum. Other
preliminary data showed that atrial tachycardia causes an immediate increase in cytoplasmic
Ca2+ concentration, which results in impaired Ca2+ release and cellular contractile
dysfunction after the cessation of tachycardia.23
Also other signaling pathway(s) by which AF leads to changes in atrial calcium
handling are involved. Recently completed experimental work suggests that T-type Ca2+
channels may mediate atrial tachycardia-induced electrical remodeling, because the T-
type Ca2+ channel blocker mibefradil limits both the ERP changes and AF promotion caused
by one week of rapid atrial pacing. Also in this case calcium overload would be prevented
by blocking a calcium channel.39
The similarity between the cellular ultrastructural changes caused by sustained AF
and those seen in hibernating myocardium40 have led to a suggestion that atrial ischemia
may play a role in triggering remodeling caused by AF. Whether ischemia occurs in AF is
144
Chapter 10
still debatable, but a reduced atrial blood flow in dogs with rapid pacing induced AF was
found and could result in atrial ischemia.41 A potential role for atrial ischemia is consistent
with the protective effect observed with blockade of the Na+/H+ exchanger in short term
tachycardia-induced atrial remodeling.42 In this model ischemia would give rise to a decrease
in intracellular pH, which leads to an exchange of intracellular hydrogen ions for
extracellular Na+ ions. Such an increase in intracellular Na+ results in a lower, or even
negative, equilibrium potential for the Na+/Ca2+ exchanger, thereby leading to a greater
magnitude of ‘reverse-mode’ functioning of the Na+/Ca2+ exchanger and therefore an influx
of Ca2+ ions.43 Alternatively, inhibition of the Na+/H+ exchanger may alter cellular ionic
homeostasis and combat calcium overload induced by ischemia.
In chapter 7 a mechanism that potentially modulates calcium overload in AF was
described. Studies on gene-expression of the endothelin system revealed that these systems
play a role in AF induced remodeling especially in patients with underlying valve disease.
We found that mRNA amounts of endothelin-1 are induced predominantly in persistent
AF with underlying valve disease. It is known that elevated endothelin-1 levels increase
intracellular calcium levels via the L-type Ca2+ channel44, indicating that calpain activation
also could play a role in AF with underlying valve disease, via different signal transduction
pathways. Moreover, increased amounts of BNP (Chapter 6) could be a compensatory
mechanism to reduce the intracellular calcium overload and thereby leading to relaxation
of the myocardial cell. BNP modulates cardiac calcium homeostasis via reduced intracellular
concentrations of cyclic adenosine monophosphate (cAMP), which inactivate the L-type
Ca2+ channel and activate the acetylcholine-dependent potassium channel leading to
repolarization of the action potential.18
Thus, several lines of evidence point to a central role of cellular calcium overload in
AF induced remodeling. Our work now reveals the potential role of calcium sensitive
processes that lead to changes in gene-expression and structural changes. Because elevated
levels of intracellular Ca2+ are known to activate proteolysis, this could result in increased
breakdown of myofilaments27,45 and ion-channel proteins (Chapter 9). In turn this could be
responsible for decreased contractility as well as for the vulnerability to AF.
Time course structural remodelingIn addition to electrophysiological, functional ion-current and ion-channel gene
expression changes, AF is associated with alterations in morphology. In Chapter 9 we
described an increase in degenerative contraction band necrosis observed in patients with
persistent and paroxysmal AF. Furthermore, we observed an increase in myocardial
hibernation (loss of sarcomeres and pale nuclei) only in patients with persistent AF, which
positively correlated with the duration of AF. This indicates that in human persistent AF
hibernation could be the specific structural change due to AF, in accordance with
145
General discussion
development of hibernation in the goat model for AF.40 We observed abundant degenerative
features in lone, paroxysmal AF. These could represent the prelude to the vulnerability to
AF by inducing dispersion of conduction (Figure 1B). Once persistent AF has been
developed, hibernation (which depends on prolonged periods of sustained AF) is more
abundant considered that cells liable to degeneration have now disappeared. These notions
are supported by the finding that in our patients with persistent AF, hibernation increased
with the duration of AF, while degeneration decreased. Possibly, hibernating myocardium
is protected against degeneration, as found after ischemic preconditioning.46 The reported
structural changes in human AF are in accordance with other studies. In humans structural
changes occur in atrial myocytes in patients with persistent AF.47 In patients with atrial
arrhythmias, myolysis and glycogen storage were only observed in a small number of
cells and that changes were frequently accompanied by lysosomal degeneration. In
experimental models these structural abnormalities appeared to be more pronounced when
the underlying pathology was aggravated by sustained AF.48,49 The occurrence of
degenerative myocardium could lead to increased dispersion of refractoriness and
conduction, which was found to enhance the inducibility and spontaneous occurrence of
idiopathic human AF.50 In addition to these defined changes in structural features, the
myocytes of AF patients displayed increased heterogeneity of cell size. Taken together,
the observed structural changes are indicative of a substantial deterioration of normal
tissue architecture, likely to promote AF through heterogeneity of atrial refractoriness50,51
and slowed atrial conduction.9,10,52 Since extreme physical stress, in combination with
sustained elevated cytosolic calcium levels, as in experimental AF22 often results in necrosis,
calpain could play an important role in this condition.53
In their model Ausma and coworkers noted mitochondrial enlargement, glycogen
accumulation, loss of sarcoplasmic reticulum and contractile elements in the atria of goats
subjected to chronic AF for upto 23 weeks.40 These changes resemble those observed in
the hibernating myocardium of the patients we investigated (Chapter 9). Recently the time
course of structural changes during AF in goats after 1-16 weeks of AF was studied.54
Here, the structural changes appeared to develop progressively, the earliest changes having
been noted after 1 week of AF and related to nuclear redistribution of heterochromatin
(Figure 1A). The nuclei showed a homogeneous distribution of chromatin, resembling
that found in embryonic/neonatal cardiomyocytes.40,55 After 4 weeks and at later times, AF
affected sarcomeres, glycogen, mitochondria and sarcoplasmatic reticulum simultaneously.
The loss of sarcoplasmatic reticulum and contractile proteins surely cause a decrease in
contractile force and hence atrial stunning54, which could be mediated by calpain.25,27
Since AF is promoted by slow conduction9,10,52,56 studies investigated the gap-junction
proteins, connexins, which play an important role in homogenous wavefront propagation
and conduction velocities in the heart.9,21,57,58 Gap-junctions are clusters of channels which
146
Chapter 10
span the closely apposed plasma membranes, forming cell-to-cell pathways. Connexins
are permeable to ions and small molecules up to 1 kDa in molecular mass, like second
messengers such as inositol triphosphate, cyclic AMP and calcium.
The initial data presented on changes in intercellular connexins were contradictory.
One study in the dog showed that AF increases connexin43 expression, the most abundant
connexin9 and another in the goat suggested that connexin43 is unaltered, but the distribution
of connexin40, mainly present in atrium was altered.57 In a recent study the gap junctional
changes in relation to stabilization of AF were studied.21 In goats that were in sinus rhythm
the distribution of connexin40, a connexin that gives high conductance, was homogeneous.
After 2 weeks in AF, which was the time associated with markedly increased intracellular
Ca2+ deposition22 and just before AF became sustained, heterogeneity in the connexin40
distribution was observed. The connexin40 distribution pattern correlated with the
occurrence of structural changes (myolysis) in atrial myocytes .
The structural changes, myolysis and heterogeneity of connexin40 distribution,
possibly relate to calcium induced calpain activity and explain the slow recovery (weeks
to months) after cardioversion of AF in patients33,59, the contractile dysfunction5,34 and the
electrophysiological changes during AF.4
Electrophysiological propertiesOver the past several years, AF-induced electrical remodeling and its underlying
mechanisms have been studied in substantial detail. In experimental studies part of the
underlying electrophysiological changes explaining the progressive nature of AF were
demonstrated.4,52 The increased tendency of the atria to fibrillate was paralleled by a
progressive shortening of the atrial effective refractory period (AERP) and loss of the
physiological rate adaptation of the refractory period which was termed atrial electrical
remodeling.4 The reduction in rate adaptation of the AERP is also observed in patients
with AF.60 All studies have shown that sustained atrial tachycardia decreases AERP and
changes occur over a period of days to weeks4,9,10,52, but AF can decrease AERP over a time
interval as short as several minutes (Figure 1A,B).15 Although the AERP reduction caused
by AF favors arrhythmia maintenance, it seems not be the only factor involved because
AF-induced AERP alterations become maximal well before AF-promoting effects
stabilize.4,10 One of the AF-promoting effects is tachycardia induced atrial conduction
slowing.9,10,52 It has a slower time course than AERP changes, probably due to delayed
onset of structural changes in the gap junctional remodeling21,57 and could account for at
least a part of the continued development of AF promotion after AERP changes have
stabilized. Whether gap junctional remodeling is caused by calpain induction is unknown,
but it is known that at least proteasome activity underlies a connexin43 degradation.61
147
General discussion
In addition to changes in the absolute value of AERP, atrial tachycardia alters the spatial
distribution of AERP. The spatial heterogeneity of AERP appears to be an important
determinant in the maintainance ot AF50,51,62,63 and there are indications that changed atrial
autonomic innervation, i.e. norephinephrine induced atrial sympathetic innervation, plays
an important role.64 Since norephinephrine causes elevation of the intracellular calcium
concentration in atrial myocytes, calpains might be activated and represent the causal link
to the maintenance of AF.
The combination of electrophysiological changes caused by sustained atrial tachycardia
i.e. reduced AERP, diminished or reversed adaptation to rate, slowed conduction and
increased spatial AERP heterogeneity, and the underlying structural changes caused by
calcium overload induced calpain ativity would be expected to promote AF maintenance
by enhancing the number of functional reentry circuits during AF.
Future perspectives
Pre-conditioningThe specific structural change induced by AF (Chapter 9) seemed to resemble chronic
hibernating myocardium.40 It is generally believed that by down-regulating their function,
cardiomyocytes adapt to a lowered oxygen availability and thereby restore the oxygen
supply/demand ratio. Part of the atrial cardiomyocytes acquired a dedifferentiated
phenotype, by re-expression of typical embryonic proteins. Furthermore, there is indirect
evidence that dedifferentiated,55 hibernating cardiomyocytes tolerate ischemia better than
non-dedifferentiated cardiomyocytes.65 It could be hypothesised that endogenous protective
mechanisms, such as an increased expression of certain heat shock proteins (stress induced
proteins which protects the cell against damage)66, are up-regulated in AF induced
hibernating myocytes. Although direct evidence of such up-regulation is missing, it is
known that ischemic preconditioning induces the efficient translation of stress proteins.46
Several of these proteins are subsequently translocated to the nucleus, possibly to protect
against degradation of DNA that has become more susceptible to degradation due to a
transformation of the chromatin organisation into a nuclease-sensitive conformation (as is
the case in apoptosis).67,68 Heat shock proteins, such as Hsp70, Hsp27 and αB-crystallin
are known to protect against ischemic cardiac damage. Unlike ischemic preconditioning,
which also attenuates apoptotic cell death induced by ischemia/reperfusion in a pig model
of short-term hibernation69, mRNA expression of Hsp70 and several other apoptosis-
modulating proteins was not altered in the ventricle during coronary stenosis nor during
subsequent stunning.70 Still it could be worthwhile investigating the role of protective heat
shock proteins in AF.
148
Chapter 10
Figure 4.
Example of an agarose gel. Here the L-type calcium channel α1 subunit and the GAPDH are given. H9c2 cells
were stimulated 0, 12, 24 and 36 hours and 36 hours stimuation combined with 48 hours recovery. The amount
of L-type Ca2+ channel increased during the different stimulation protocols.
0 12 24 36 recovery M
GAPDH
L-type Ca2+ channel
Figure 4
Possible clinical relevanceThe chance of successful chemical cardioversion and/or prevention of AF is dependent
on the duration of AF. This clinically observed diminished efficacy of cardioversion therapy
after long term AF cannot only by explained by the occurrrence of electrical remodeling.
The ion-channel protein remodeling and structural remodeling probably also affect the
electrophysiological function of the atrial myocardium.
In patients with persistent AF, there is a correlation between the duration of AF and
the time needed to recover atrial contractile function after cardioversion.33,71 The increase
in calpain activity which could lead to structural remodeling of the atrial myocytes might
give an explanation for the delay in recovery of contractile function in the atria after
conversion to sinus rhythm as seen in patients with persistent AF. Interference with the
calpain pathway by pharmacological intervention might represent an important new
therapeutic strategy to decrease protein degradation and thereby reduce the vulnerability
to AF. Calpain inhibitors as therapeutic agents are already used in nerve and muscle
degeneration72, but their potential benefit in heart diseases is not studied yet. After restoration
of normal sinus rhythm it may take the cardiomyocytes a certain period to rebuild a normal
amount of sarcomeres, if that is still possible at all.73 Data describing the recovery of ion-
channel protein expression are lacking, but a few reports describe, in contrast to the structural
remodeling, reversal of electrical remodeling in human AF after cardioversion.74,75 Since
AF induces remodeling in the atria it is essential to restore sinus rhythm as soon as possible,
thereby preventing the continuation of the atrial structural, ion-channel protein and electrical
remodeling.
Furthermore, differences in adaptation mechanisms which were found between patients
with lone AF and AF with underlying heart disease suggest the need for different
pharmacological treatment. For example, patients with elevated levels of endothelin could
be treated with an endothelin receptor antagonist, which has been shown to normalize
alterations in expression of various cardiac genes (like normalization of ryanodine receptor,
sarcoplasmic reticulum Ca2+ ATPase, angiotensin-converting enzyme, angiotensin II type
I receptor and prepro-endothelin 1) in failing myocardium.76
149
General discussion
Figure 5.
Calpain activation measurement in H9c2 cells.
The cells were stimulated 0, 20 and 30 hours.
An increase in calpain activity was found in
stimulated cells and inhibited by calpain I
inhibitor by 65%.hours of electrical stimulation0 5 10 15 20 25 30
calpain activation (arbitrary units)
0
100
200
300
400
500
calp
ain
acti
vati
on (
arb
itra
ry u
nit
s)
New Experimental model for AFTo mimic human AF experimental models were developed. AF is studied in more
detail in different animal models. The dog10,38,52,77 and goat models4 for AF are well studied
and characterized. For studying molecular and cellular mechanisms the animal models
have disadvantages. First, in the animal a lot of (unknown) parameters can interfere with
the results. Second, animal models take time and are expensive and the third important
disadvantage is for the animal itself. These are reasons to think of a different experimental
model for AF, especially for studying the fundamental calcium sensitive pathways. A pos-
sible new model could be the electrical stimulation of myocardium cells. Pilot experi-
ments have been done for H9c2 cells, which are rat myoblast cells and are able to differen-
tiate into myotubes. H9c2 cells appear to be unique in that they express the cardiac isoforms
of the L-type Ca2+ channel alpha 1-subunit mRNA (data not shown). Another good candi-
date cell line is the immortalised HL-1 cells. These are contracting mouse atrial cells and
when electrically stimulated could mimic human AF.
For the electrical stimulation experiment a special culture flask was developed by
Popta and Henning. Pilot experiments were done to investigate changes in L-type Ca2+
channel mRNA expression and for measuring the calpain activation during electrical stimu-
lation in H9c2 myotubes. Therefore, the cells were stimulated for 12, 24 and 36 hours
followed by three days recovery (0.5 Hz, 25V). Figure 4 shows that in stimulated H9c2
cells the mRNA expression of the L-type Ca2+ a1 subunit was increased compared to
unstimulated H9c2 cells. Thus the H9c2 cell line may prove to be useful when studying
the regulation of subtype-specific Ca2+ channel gene expression. The second set of experi-
ments were done to measure the calpain activity with a calpain specific fluorogenic sub-
strate. The calpain activity increased significantly after 20 and 30 hours of stimulation
(Figure 5). This activity could be reduced by 65% by a specific inhibitor of calpain I.
These two sets of pilot experiments reveal that these cell models could represent excellent
models for studying the signaling pathways activated by electrical stimulation and the
tential benefits of pharmacological intervention in AF.
150
Chapter 10
In conclusion, the combination of electrophysiological, ion-channel protein and struc-
tural remodeling caused by sustained AF would be expected to promote AF maintenance
by increasing the calcium overload induced calpain activity leading to induction of the
number of functional reentry circuits during AF. New cell models could be benificial for
studying the signaling pathways activated by electrical stimulation and pharmacological
intervention in AF.
Reference List
1. Godtfredsen J. Etiology, course and prognosis. A follow-up study of 1212 cases. Copenhagen: University
of Copenhagen. Thesis 1975;
2. Van Gelder IC, Crijns HJGM, Tieleman RG, et al. Value and limitation of electrical cardioversion in
patients with chronic atrial fibrillation - importance of arrhythmia risk factors and oral anticoagulation.
Arch Intern Med 1996; 156:2585-2592.
3. Tieleman RG, De Langen CDJ, Van Gelder IC, et al. Verapamil reduces tachycardia-induced electrical
remodeling of the atria. Circulation 1997; 95:1945-1953.
4. Wijffels MC, Kirchhof CJ, Dorland R, et al. Atrial fibrillation begets atrial fibrillation. A study in awake
chronically instrumented goats. Circulation 1995; 92:1954-1968.
5. Leistad E, Aksnes G, Verburg E, et al. Atrial contractile dysfunction after short-term atrial fibrillation is
reduced by verapamil but increased by BAY K8644. Circulation 1996; 93:1747-1754.
6. Lai LP, Su MJ, Lin J, et al. Down-regulation of L-type calcium channel and sarcoplasmic reticular Ca2+ -
ATPase mRNA in human atrial firbillation without significant change in the mRNA of ryanodine receptor,
calsequestrin and phospholamban: an insight into the mechanism of atrial electrical remodeling. J Am Coll
Cardiol 1999; 33:1231-1237.
7. Ohkusa T, Ueyama T, Yamada J, et al. Alterations in cardiac sarcoplasmic reticulum Ca2+ regulatory
proteins in the atrial tissue of patients with chronic atrial fibrillation. J Am Coll Cardiol 1999; 34:255-263.
8. Boyett MR, Jewell BR. Analysis of the effects of changes in rate and rhythm upon electrical activity in the
heart. Prog Biophys Molec Biol 1980; 36:903-923.
9. Elvan A, Wylie K, Zipes DP. Pacing-induced chronic atrial fibrillation impaires sinus node function in
dogs: electrophysiological remodeling. Circulation 1996; 94:2953-2960.
10. Gaspo R, Bosch RF, Talajic M, et al. Functional mechanisms underlying tachycardia-induced sustained
atrial fibrillation in a chronic dog model. Circulation 1997; 96:4027-4035.
11. Yue L, Feng J, Gaspo R, et al. Ionic remodeling underlying action potential changes in a canine model of
atrial fibrillation. Circ Res 1997; 81:512-525.
12. Yue L, Melnyk P, Gaspo, et al. Molecular mehanisms underlying ionic remodeling in a dog model of atrial
fibrillation. Circ Res 1999; 84:776-784.
13. Van Wagoner DR, Pond AL, McCarthy PM, et al. Outward K+ current densities and Kv1.5 expression are
reduced in chronic human atrial fibrillation. Circ Res 1997; 80:1-10.
14. Bosch RF, Zeng X., Grammer JB, et al. Ionic mechanisms of electrical remodeling in human atrial fibril-
lation. Cardiovasc Res 1999; 44:121-131.
15. Daoud EG, Knight BP, Weiss R, et al. Effect of verapamil and procainamide on atrial fibrillation-induced
electrical remodeling in humans. Circulation 1997; 96:1542-1550.
16. Yu WC, Chen SA, Lee SH, et al. Tachycardia-induced change of atrial refractory period in humans. Rate
dependency and effects of antiarrhythmic drugs. Circulation 1998; 97:2331-2337.
17. Gaspo R, Sun H, Fareh S, et al. Dihydropyridine and beta adrenergic receptor binding in dogs with
tachycardia-induced atrial fibrillation. Cardiovasc Res 1999; 42:434-442.
18. Clemo HF, Baumgarten CM, Ellenbogen K, et al. Atrial natriuretic peptide and cardiac electrophysiology:
autonomic and direct effects. J Cardiovasc Electrophysiol 1996; 7:149-162.
19. Mays DJ, Foose JM, Philipson LH, et al. Localization of the Kv1.5 K+ channel protein in explanted
cardiac tissue. J.Clin.Invest. 1995; 96:282-292.
20. Goette A, Arndt M, Röcken C, et al. Regulation of angiotensin II receptor subtypes during atrial fibrilla-
tion in humans. Circulation 2000; 101:2678-2781.
21. Van der Velden HMW, Ausma J, Rook MB, et al. Gap junctional remodeling in relation to stabilization of
atrial fibrillation in the goat. Cardiovasc Res 2000; 46:476-486.
22. Ausma J, Dispersyn GD, Duimel H, et al. Changes in ultrastructural calcium distribution in goat atria
during atrial fibrillation. J Mol Cell Cardiology 2000; 32:355-364.
151
General discussion
23. Sun H, Leblanc N, Nattel S. Effects of atrial tachycardia on intracellular Ca2+ and cellular contractility.
Circulation 1999; 100:I-200
24. Suzuki K, Imajoh S, Emori Y, et al. Calcium-activated neutral protease and its endogenous inhibitor.
Activation at the cell membrane and biological function. FEBS Letters 1987; 220:271-277.
25. Atsma DE, Bastiaanse EM, Jerzewski A, et al. Role of calcium-activated neutral protease (calpain) in cell
death in cultured neonatal rat cardiomyocytes during metabolic inhibition. Circ Res 1995; 76:1071-1078.
26. Gorza L, Menabo R, Vitadello M, et al. Cardiomyocyte troponin T immunoreactivity is modified by cross-
linking resulting from intracellular calcium overload. Circulation 1996; 93:1896-1904.
27. Gao WD, Atar D, Liu Y, et al. Role of troponin I proteolysis in the pathogenesis of stunned myocardium.
Circ Res 1997; 80:393-399.
28. Cohen SA. Immunocytochemical localization of rH1 sodium channel in adult rat heart atria and ventricle.
Presence in terminal intercalated disks. Circulation 1996; 94:3083-3086.
29. Laflamme MA, Becker PL. G(s) and adenylylcyclase in transverse tubules of heart: implications for cAMP-
dependent signaling. Am.J.Physiol. 1999; 277:H1841-H1848
30. Lane RD, Allan DM, Mellgren R. A comparison of the intracellular distribution of mu-calpain, m-calpein
and calpastatin in proliferating human A431 cells. Exp Cell Res 1992; 203:5-16.
31. Aime-Sempe C, Folliguet T, Rucker-Martin, et al. Myocardial cell death in fibrillating and dilated human
right atria. J Am Coll Cardiol 1999; 34:1577-1586.
32. Manning WJ, Leeman DE, Gotch P, et al. Pulsed evaluation of atrial mechnical function after electrical
cardioversion of atrial fibrillation. J Am Coll Cardiol 1989; 13:617-623.
33. Manning WJ, Silverman DI, Katz SE, et al. Impaired left atrial mechanical function after cardioversion:
relation to the duration of atrial fibrillation. J Am Coll Cardiol 1994; 23:1535-1540.
34. Daoud EG, Marcovitz P, Knight B, et al. Short-term effect of atrial fibrillation on atrial contractile func-
tion in humans. Circulation 1999; 99:3024-3027.
35. Van Gelder IC, Brundel BJJM, Henning RH, et al. Alterations in gene expression of proteins involved in
the calcium handling in patients with atrial fibrillation. J Cardiovasc Electrophysiol 1999; 10:552-560.
36. Brundel BJJM, Van Gelder IC, Henning RH, et al.Gene expression of proteins influencing the calcium
homeostasis in patients with persistent and paroxysmal AF. Cardiovasc Res 1999;42:443-454.
37. Tieleman RG, Van Gelder IC, Crijns HJ, et al. Early recurrences of atrial fibrillation after electrical
cardioversion: A result of fibrillation induced electrical remodeling of the atria? J Am Coll Cardiol 1998;
31:167-173.
38. Goette A, Honeycutt C, Langberg JJ. Electrical remodeling in atrial fibrillation. Time course and mecha-
nisms. Circulation 1996; 94:2968-2974.
39. Fareh S, Bénardeau A, Thibault, et al. The T-type Ca2+ channel blocker mibefradil prevents the develop-
ment of a substrate for atrial fibrillation by tachycardia-induced atrial remodeling in dogs. Circulation
1999; 100:2191-2197.
40. Ausma J, Wijffels M, Thone F, et al. Structural changes of atrial myocardium due to sustained atrial
fibrillation in the goat. Circulation 1997; 96:3157-3163.
41. Jayachandran JV, Winkle W, Sih HJ, et al. Chronic atrial fibrillation from rapid atrial pacing is associated
with reduced atrial blood flow: a positron emission tomography study. Circulation 1998; 98:I-209
42. Jayachandran JV, Zipes DP, Weksler J, et al. Role of the Na+/H+ exchanger in short-term atrial electro-
physiological remodeling. Circulation 2000; 101:1861-1866.
43. Levi AJ, Dalton GR, Hancox JC, et al. Role of intracellular sodium overload in the genesis of cardiac
arrhythmias. J Cardiovasc Electrophysiol 1997; 8:700-721.
44. He JQ, Pi Y, Walker J, et al. Endothelin-1 and photoreleased diacylglycerol increase L-type Ca2+ current
by activation of protein kinase C in rat ventricular myocytes. J Physiol (Lond) 2000; 524:807-820.
45. Gorza L, Menabo R, Di Lisa F, et al. Troponin T cross-linking in human apoptotic cardiomyocytes. Am
J Pathol 1997; 150:2087-2097.
46. Tanaka M, Fujiwara H, Yamasaki K, et al. Expression of heat shock protein after ischemic precondition-
ing in rabbit heart. Jpn Circ J 1998; 62:512-516.
47. Mary-Rabine L, Pham TD, Hordof A, et al. The relationship of human atrial cellular electrophysiology to
clinical function and ultrastructure. Circ.Res. 1983; 52:188-199.
48. Boyden PA, Tilley LP, Albala A, et al. Mechanisms for atrial arrhythmias associated with cardiomyopa-
thy: a study of feline hearts with primary myocardial disease. Circulation 1984; 69:1036-1047.
49. Boyden PA, Tilley LP, Pham T, et al. Effects of left atrial enlargement on atrial transmembrane potentials
and structure in dogs with mitral valve fibrosis. Am J Cardiol 1982; 49:1896-1908.
50. Ramanna H, Hauer RNW, Wittkampf FHM, et al. Identification of the substrate of atrial vulnerability in
patients with idiopathic atrial fibrillation. Circulation 2000; 101:995-1001.
51. Fareh S, Villemaire C, Nattel S. Importance of refractoriness heterogeneity in the enhanced vulnerability
152
Chapter 10
to atrial fibrillation induction caused by tachycardia-induced atrial electrical remodeling. Circulation 1998;
83:2202-2209.
52. Morillo CA, Klein GJ, Jones D, et al. Chronic rapid atrial pacing. Structural, functional, and electrophysi-
ological characteristics of a new model of sustained atrial fibrillation. Circulation 1995; 91:1588-1595.
53. Majno G, Joris I. Apoptosis, oncosis and necrosis. An overview of cell death. Am J Pathol 1995; 146:3-
15.
54. Ausma J, Lenders MH, Duimel H, et al. Time course of structural changes in atria after atrial fibrillation
in goats: existence of hibernating myocardium. in press 2000;
55. Ausma J, Wijffels M, Van Eys G., et al. Dedifferentiation of atrial cardiomyocytes as a result of chronic
atrial fibrillation. Am J Pathol 1997; 151:985-997.
56. Franz MR, Karasik PL, Li C, et al. Electrical remodeling of the human atrium: similar effects in patients
with chronic atrial fibrillation and atrial flutter. J Am Coll Cardiol 1997; 30:1785-1792.
57. Van der Velden HMW, Van Kempen MJA, Wijffels MCEF, et al. Altered pattern of connexin40 distribu-
tion in persistent atrial fibrillation in the goat. J Cardiovasc Electrophysiol 1998; 9:596-607.
58. Britz-Cunningham SH, Shah MM, Zuppan CW, et al. Mutations of the connexin43 gap-junction gne in
patients with heart malformations and defects of laterality. N Engl J Med 1995; 332:1323-1329.
59. Van Gelder IC, Crijns HJ, Blanksma PK, et al. Time course of hemodynamic changes and improvement of
exercise tolerance after cardioversion of chronic atrial fibrillation unassociated with cardiac valve disease.
Am J Cardiol 1993; 72:560-566.
60. Attuel P, Childers R, Cauchemez B, et al. Failure in the rate adaptation of the atrial refractory period: its
relationship to vulnerability. Intern J Cardiol 1982; 2:179-197.
61. Laing JG, Tadros PN, Saffitz J, et al. Proteolysis of connexin43-containing gap junctions in normal and
heat-stressed cardiac myocytes. Cardiovasc Res 1998; 38:711-718.
62. Liu L, Nattel S. Differing sympathetic and vagal effects on atrial fibrillation in dogs: role of refractoriness
heterogeneity. Am J Physiol 1997; 273:H805-H816
63. Wang J, Liu L, Feng, et al. Regional and functional factors determining induction and maintenance of
atrial fibrillation in dogs. Am J Physiol 1996; 271:H148-H158
64. Jayachandran JV, Sih HJ, Winkle W, et al. Atrial fibrillation produced by prolonged rapid atrial pacing is
associated with heterogeneous changes in atrial sympathetic innervation. Circulation 2000; 101:1185-
1191.
65. Vanoverschelde JLJ, Wijns W, Depré, et al. Mechanisms of chronic regional postischemic dysfunction in
humans: new insights from the study on non-infarcted collateral dependent myocardium. Circulation 1993;
87:1513-1523.
66. Martin JL, Mestril R, Hilal-Dandan R, et al. Small heat shock proteins and protection against ischemic
injury in cardiac myocytes. Circulation 1997; 96:4343-4348.
67. Arrigo AP, Suhan JP, Welch WJ. Dynamic changes in the structure and intracellular locale of the mamma-
lian low-molecular-weight heat shock protein. Mol Cell Biol 1988; 8:5059-5071.
68. McMillan DR, Xiao X, Shao L, et al. Targeted disruption of heat shock transcription factor 1 abolished
thermotolerance and protection against heat-inducible apoptosis. J Biol Chem 1998; 273:7523-7528.
69. Maulik N, Yoshida T, Engelman RM, et al. Ischemic preconditioning attenuates apoptotic cell death
associated with ischemia/reperfusion. Mol Cell Biochem 1998; 186:139-145.
70. Bartling B, Hoffmann J, Holtz J, et al. Quantification of cardioprotective gene expression in porcine short-
term hibernating myocardium. J Mol Cell Cardiology 1999; 31:147-158.
71. Van Gelder IC, Crijns HJ, Van Gilst WH, et al. Prediction of uneventful cardioversion and maintenance of
sinus rhythm from direct-current electrical cardioversion of chronic atrial fibrillation and flutter.
Am.J.Cardiol. 1991; 41
72. Stracher A. Calpain inhibitors as therapeutic agents in nerve and muscle degeneration. Ann N Y Acad Sci
1999; 28:52-59.
73. Ausma J, Duimel H, Wouters L, et al. Structural atrial changes induced in the goat by 16 weeks of atrial
fibrillation are still present 8 weeks after cardioversion. Europace 2000; 1:B12
74. Hobbs WJC, Fynn S, Todd D, et al. Reversal of atrial electrical remodeling after cardioversion of persis-
tent AF in humans. Circulation 2000; 101:1145-1151.
75. Pandozi C, Bianconi L, Villani M, et al. Electrophysiological characteristics of the human atria after
cardioversion of persistent atrial fibrillation. Circulation 1998; 98:2860-2865.
76. Sakai S, Miyauchi T, Yamaguchi. Long-term endothelin receptor antagonist administration improves al-
terations in expression of various cardiac genes in failing myocardium of rats with heart failure. Circula-
tion 2000; 101:2849-2853.
77. Yue L, Feng J, Li GR, et al. Transient outward and delayed rectifier currents in canine atrium: properties
and role of isolation methods. Am J Physiol 1996; 270:H2157-H2168
153
Summary
Summary
Clinical and experimental studies showed that electrical and contractile remodeling
occurred early after onset of atrial fibrillation. Both processes could be reduced by blocking
the L-type Ca2+ channel suggesting the notion that changes in the calcium homeostasis
triggered by tachycardia induced intracellular calcium overload, play a pivotal role in the
induction of these remodeling processes. To obtain insight in the underlying molecular
mechanisms we first studied the molecular remodeling of proteins, which influence the
calcium homeostasis in a heterogeneous group of AF patients (Chapter 2 and 3). We found
that reductions in mRNA and protein expression of the L-type Ca2+ channel and sarcoplasmic
reticulum Ca2+ ATPase occurred predominantly in patients with persistent AF. Furthermore,
mRNA expression was found to be dependent on the duration of persistent AF. Patients
with >6 months duration of AF revealed reductions in mRNA in contrast to patients with
<6 months duration of AF. In these patients no changes in mRNA expression were found.
Secondly, we investigated molecular changes in ion-channels that contribute
importantly to the action potential duration. Apart from the L-type Ca2+ channel, we
investigated the contribution of potassium channel gene expression in right atrial appendages
in paroxysmal and persistent AF, since an increase in potassium channel amount or activity
could explain the electrical remodeling (Chapter 4). Reductions in mRNA and protein
levels were found for several K+ channels in patients with persistent AF. In patients with
paroxysmal AF these reductions were observed predominantly at the protein level and not
at the mRNA level. In addition, the regulation of L-type Ca2+ channel and several K+
channels and its relation to AERP in patients with persistent and paroxysmal AF was
studied (Chapter 5). We demonstrated a positive correlation between the ion-channel protein
expression of L-type Ca2+ channel, Kv4.3, Kv1.5, HERG, minK and Kir3.1 and the AERP
but also with the rate adaptation of AERP in patients with persistent and paroxysmal AF.
The correlation between ion-channel protein amounts and AERP indicate that ion-channel
protein remodeling, beside the electrical remodeling plays an important role in the
vulnerability to AF. The reductions in L-type Ca2+ channel could represent a possible
explanation for the electrophysiological changes during AF. Furthermore, the data indicate
that reduced ion-channel protein expression occurres due to high atrial rate. We called this
phenomenon ion-channel remodeling to describe the AF induced changes in ion-channel
protein expression.
The impact of other compounds such as the natriuretic peptide system (Chapter 6)
and the endothelin system (Chapter 7) were also studied. These studies revealed the
influence of concomitant valvular disease in patients with persistent AF on mRNA
expression of neurohumones. The right atrial appendage of these patients showed increased
levels of ANP, BNP and endothelin-1 in combination with reduced expression of their
receptors. Possibly the increased ANP and BNP mRNA enhance the myocyte relaxation
properties by combatting the calcium overload. An increase in endothelin-1 might generate
an additional increase in intracellular calcium concentrations by activation of the L-type
154
Ca2+ channel.
A remarkable finding during the study of mRNA and ion-channel protein remodeling
was the discrepancy between changes in mRNA and protein levels in patients with
paroxysmal AF (Chapter 4 and 5). Whereas ion-channel protein levels of the examined L-
type Ca2+ channel, Kv1.5, Kir3.1 and minK were substantially decreased, the mRNA
levels were essentially unaffected in paroxysmal AF. This discrepancy prompted us to the
role of an adaptive mechanism which influence in AF was previously unknown, i.e. the
activation of a proteolytic system. Different proteolytic pathways could be involved in AF.
Since cytosolic calcium is increased during AF, proteolysis may be invoked by calcium
dependent neutral proteases, calpain I and II. In Chapter 8 the increased proteolytic activity
during paroxysmal and persistent lone AF due to activation of the calpain pathway was
described. This increase seemed to be predominantly due to elevated activation and
expression of calpain I.
In Chapter 9 we examined a variety of molecular changes in atrial tissue from patients
with paroxysmal and persistent lone AF and related them to the level of calpain activity.
Immunohistochemical detection of calpain I demonstrated increased staining at the
intercalated disk and in the nucleus of atrial myocytes of AF patients. Accordingly, calpain
activity was increased in patients with AF. Furthermore, an increased number of
degenerative myocytes was observed in both patient groups with AF. Hibernating myocytes
were only present in persistent AF and numbers increased with the duration of AF. Finally,
calpain activity correlated inversely with the expression levels of ion-channel proteins,
the degree of structural changes and the rate adaptation coefficient of AERP. These results
strongly suggest that induction of calpain activation represents the missing link between
the calcium overload observed in AF and remodeling of atrial myocytes during AF.
The incidence of AF increases due to ageing of the population and AF has the tendency
to promote itself. Currently the arrhythmogenic electrophysiological changes (electrical
remodeling) are well known, but can not explain by themselves that ‘AF begets AF’. This
thesis shows significant molecular and ultrastructural changes related to cellular hibernation.
The latter support the notion of a second factor in AF promotion, since these - in part -
form the basis for conduction slowing and dispersion of conduction and refractoriness.
This thesis also shows a novel mechanism of protein remodeling taking place very early
after AF onset. This mechanism is driven by calpain I activation. Future studies on
pharmacological interventention in AF induced protein remodeling, like calpain activation,
may prove effective in preventing atrial damage after AF onset. Such interventions may
also break the chain of events by which AF tends to beget AF.
155
Samenvatting
Samenvatting
Boezemfibrilleren is een hartritmestoornis welke resulteert in een sterke toename
van de hoeveelheid contracties in de boezemcel. Dierexperimentele en klinische studies
hebben laten zien dat snel na de inductie van boezemfibrilleren verkorting van de atriale
effectieve refractaire periode (AERP) en ook de actie potentiaal duur plaatsvindt. Daarnaast
treedt verminderde contractiliteit van de boezem op nadat boezemfibrilleren is teruggebracht
in normaal sinus ritme. Beide processen zijn te verbeteren door toediening van een L-type
calciumantagonist voorafgaand aan boezemfibrilleren, deze voorkomt namelijk dat cal-
cium de cel instroomt en kan leiden tot contractie van de boezem. Dit effect van een L-
type calciumantagonist suggereert dat een verstoring in de calciumhuishouding,
waarschijnlijk een verhoging van de hoeveelheid calcium in de boezemcel, ten grondslag
ligt aan beide processen.
Als eerste wilden we inzicht verkrijgen in de moleculaire mechanismen welke ten
grondslag liggen aan een verstoorde calcium huishouding. Daarvoor hebben we
veranderingen in genexpressie van eiwitten, die een belangrijke rol in de
calciumhuishouding spelen bestudeerd in hartoren van patiënten met paroxysmaal en
persisterend boezemfibrilleren (Hoofdstuk 2 en 3). We beschrijven dat de mRNA
(boodschapper RNA welke de informatie bevat voor de volgorde van de aminozuren in
het te vormen eiwit) en de eiwit expressie van het L-type calcium kanaal en van het
sarcoplasmatisch reticulum calcium ATPase zijn verminderd. Beide verminderingen zijn
het meest uitgesproken in patiënten met persisterend boezemfibrilleren (Hoofdstuk 2).
Daarnaast zien we een relatie tussen de mate van reductie in mRNA expressie en de duur
van boezemfibrilleren. Patiënten met boezemfibrilleren van langer dan 6 maanden geven
de meeste mRNA reducties en deze nemen af naarmate de duur van boezemfibrilleren
korter is (Hoofdstuk 3).
Ten tweede hebben we de moleculaire veranderingen in genexpressie bestudeerd van
ion-kanalen welke een belangrijke rol spelen in het bepalen van actie potentiaal duur.
Naast het L-type calcium kanaal hebben we diverse kalium kanalen betudeerd, aangezien
activatie van deze kanalen resulteert in repolarisatie en dus verkorting van de actiepotentiaal.
In hoofdstuk 4 beschrijven we reducties in mRNA en eiwitexpressie van verschillende
kalium kanalen in patiënten met persisterend boezemfibrilleren. In patiënten met
paroxysmaal boezemfibrilleren zagen we ook vermindering in kalium kanaal expressie
maar dan alleen in eiwit en niet mRNA hoeveelheid. Verder zijn in een groep patiënten
met boezemfibrilleren de AERP’s gemeten en deze correleren positief met de hoeveelheden
eiwit van het L-type calcium kanaal en de diverse kalium kanalen. Dit houdt in dat patiënten
met boezemfibrilleren minder ion-kanaal eiwitexpressie hebben gecombineerd met een
verkorting van de AERP. Dit resultaat laat zien dat boezemfibrilleren niet alleen leidt tot
verkorting van de AERP, maar parallel hieraan ook leidt tot vermindering in ion-kanaal
eiwitexpressie.
Naast de genexpressie van de calcium huishouding eiwitten en ion-kanaal eiwitten
156
hebben we ook gekeken naar de genexpressie van andere eventueel belangrijke systemen
in boezemfibrilleren; het natriuretische systeem (ANP en BNP) (Hoofdstuk 6) en het
endotheline systeem (Hoofdstuk 7). Deze studies laten zien dat de genoemde systemen in
patiënten met boezemfibrilleren in combinatie met hartfalen een rol spelen. De hartoren
van deze groep patiënten gaven hogere expressie hoeveelheden van ANP, BNP en
endotheline-1, maar de corresponderende receptoren waren verminderd in expressie. Een
verhoging in ANP en BNP hoeveelheden zullen leiden tot relaxatie van de boezemcel. Een
verhoging van de hoeveelheid endotheline-1 in de boezemcel zal leiden tot meer contractie,
waarschijnlijk door een stimulerende werking op het l-type calcium kanaal.
Een in het oog springende bevinding tijdens deze studies is de discrepantie tussen
mRNA- en eiwitexpressie van ion-kanaal eiwitten in patiënten met paroxysmaal
boezemfibrilleren (Hoofdstukken 4 en 5). We zien dat de eiwitexpressie van het L-type
calcium kanaal en de diverse kalium kanalen was verminderd in deze patiënten. Echter de
hoeveelheid mRNA van deze ion-kanalen bleef onveranderd. Deze discrepantie zorgde
ervoor dat we de rol van een tot dan toe onbekend adaptatie mechanisme, namelijk het
eiwit-afbraak (proteolyse) systeem, in boezemfibrilleren zijn gaan bestuderen. We hebben
de mogelijke invloed van verschillende proteolytische systemen in boezemfibrilleren
onderzocht (Hoofdstuk 8). We vonden dat de calcium afhankelijke protease, calpaine I, is
geactiveerd tijdens idiopatisch paroxysmaal en persisterend boezemfibrilleren. De
verhoogde activiteit in patiënten met persisterend boezemfibrilleren kan verklaard worden
door een toename in hoeveelheid van het calpaine I eiwit.
In hoofdstuk 9 laten we zien waar in de boezemcel het calpaïne I voorkomt. We zien
dat calpaïne I zich bevindt in de zeer nauwe opening tussen boezemcellen (de intercalated
disk) en in de kern van de boezemcel en de hoeveelheden zijn het hoogst in patiënten met
persisterend boezemfibrilleren. Daarnaast beschrijven we de relaties tussen de verhoogde
calpaïne I activiteit en verschillende moleculaire veranderingen veroorzaakt door
boezemfibrilleren. De mate van structurele verandering, degeneratief en hibernerend (in
‘winterslaap’), is gequantificeerd en nam toe in hartoren van patiënten met idiopatisch
paroxysmaal en persisterend boezemfibrilleren. Daarnaast observeerden we dat
hibernerende boezemcellen alleen voorkomen in persisterend boezemfibrilleren en dat de
hoeveelheid toeneemt met de duur van het boezemfibrilleren. Verder correleerde de mate
van structurele, maar ook de mate van ion-kanaal veranderingen en de duur van de AERP
met de calpaïne activiteit. Samengevat laten deze resultaten zien dat het calpaïne systeem
de ontbrekende schakel kan zijn tussen de calcium overmaat en de verschillende moleculaire
veranderingen tijdens boezemfibrilleren in de boezemcel.
Kortom, de gepresenteerde studies laten zien dat naast de electrische veranderingen
er zich ook belangrijke veranderingen voordoen op eiwit en structureel niveau in
boezemcellen van patiënten met boezemfibrilleren. Deze veranderingen kunnen de
waargenomen verminderde contractiliteit van de boezem verklaren. Er zijn duidelijke
aanwijzingen dat het calpaïne systeem hier een essentiële rol in speelt.
157
Dankwoord
Dankwoord
Ik hoor het mezelf nog regelmatig zeggen: “Ik ga nooooooit naar Groningen”. Dat was
ongeveer vijf jaren geleden in Amsterdam. Een paar maanden later waren we spullen aan
het inpakken vanwege de verhuizing naar Groningen. Ik had een onderzoeksplaats
geaccepteerd waar ik nooit direct op had gesolliciteerd. Ik ging werken bij de Klinische
Farmacologie. De afdeling was juist oververhuist van de oudbouw naar de nieuwbouw en
niet 1 of 2 maar 3 nieuwe hoogleraren (Wiek van Gilst, Dick de Zeeuw en Pieter de Graaff)
waren er aangesteld. Maar eigenlijk werkte ik niet voor de klinische farmacologie maar
voor de Cardiologie waar ook net een nieuwe hoogleraar was aangesteld (Harry Crijns).
Daarnaast had ik niet één maar twee begeleiders (Isabelle van Gelder en Rob Henning).U
kunt zich waarschijnlijk voorstellen dat dit wat verwarrend was in de beginfase.
Ik ging werken in een moleculair lab waar nog een andere AIO rondhing. Leo Deelman,
King of the Lab, vastgeplakt aan zijn 486-er zat hij dagelijks computerspelletjes te spelen
met Jepe, een vervangende dienstplicht vervullende bioloog, of illegale
computerprogramma’s van het internet te plukken. Isabelle van Gelder kwam toen nog
wekelijks langs om te kijken of er niet gelanterfant werd en de experimenten wel de gewenste
resultaten gaven. En resultaten waren er (gelukkig). Dit is mede mogelijk gemaakt doordat
er een enorme humane hartoor vooraad was aangelegd en ik meteen aan de slag kon.
Isabelle heeft in deze begin fase een belangrijke en essentiële rol gespeelt. Isabelle, ik wil
je dan ook hartelijk danken voor je enorme inzet om het onderzoek zo goed mogelijk uit te
laten voeren. Als er problemen waren dan reageerde je snel en accuraat, factoren welke
onmisbaar zijn voor het nabehoren uitvoeren van onderzoek. Verder was je een stimulerende
factor om resultaten op papier te krijgen in de vorm van een abstract dan wel een artikel.
Dit laatste heeft weleens tot strubbelingen geleid, maar toch heeft dat onze verdere
samenwerking niet kunnen schaden. Verder heb ik onze persoonlijke gesprekken erg kunnen
waarderen.
Aangezien de experimenten naar behoren verliepen kwam er plaats voor analytische
ondersteuning en wel eerst in de vorm van Marion Franke later Cecile Driessen en nog
later Mirian Wietses. Ik wil jullie alle drie ontzettend bedanken voor jullie inzet en
gezelligheid. Mirian en Simone Gschwend wil ik nog extra bedanken niet alleen omdat ze
mijn paranimfen wilden zijn, maar ook vanwege de prettige omgang en hun organisatie
vermogen.
Regelmatig had ik promotie besprekingen met Harry Crijns, Wiek van Gilst, Rob Henning
en Isabelle van Gelder. Harry, ik wil je bedanken voor je steun en inlevingsvermogen.
Moest ik tijdens de eerste promotiebesprekingen nog uitleggen wat DNA, RNA en eiwitten
waren, je hebt altijd de moeite genomen om mee-te-denken in het onderzoek ondanks je
niet zo rooskleurige visoenen (de eerste twee jaren van het onderzoek). Wiek van Gilst zei
nooit erg veel tijdens promotie- en werkbesprekingen, maar wat hij zei heb ik altijd als
waardevol ervaren. Verder heb ik me verbaasd over de danskwaliteiten van Wiek (nachtclub
158
in Stockholm tijdens congres 1997). Ik hoop dat ik tijdens mijn promotiefeest er nog een
keer van mag genieten!
Dan hebben we nog Rob Henning, ja ja ja ja. Rob heeft een gave en dat is om de meest
negatieve resultaten om te buigen naar iets briljants. Een andere niet geheel onbelangrijke
kwaliteit is de positieve instelling en de respectvolle benadering van ieder individu. Ik heb
met name in de laatste twee jaren van mijn promotie onderzoek dankbaar gebruik mogen
maken van je kwaliteiten. Ik besef me dat jij de succesfactor bent achter een aantal
belangrijke artikelen. Bedankt Rob en ik hoop in de toekomst weer met je samen te werken.
Naast de directe medewerkers in het lab waren er ook buiten het lab mensen aan het werk
voor dit onderzoek. Bij de cardiologie was vooral Anton Tuinenburg druk om nog meer
hartoren te verzamelen. Dankzij de door Robert Tieleman en Anton gemeten ‘actie potentiaal
duur’ waarden in de hartoren van patiënten kon er een brug geslagen worden tussen klinische
en laboratorium meetwaarden. Heren bedankt voor jullie inzet en gezelligheid. Robert wil
ik nog bedanken voor mijn inwijding in het mega congres gebeuren en de gezelligheid
aldaar (American College of Cardiology, Los Angeles, 1997).
Verder wil ik de mensen van het mollab bedanken voor de gezelligheid welke ertoe heeft
bijgedragen voor het succesvol afronden van dit boekje: Marry, bedankt voor de goede
gesprekken; Leo, bedankt voor je kennis maar zeker ook voor je humor en het grote avontuur
in Keulen (we weten nu waar de het gezegde ‘Het in Keulen horen donderen’ vandaan
komt); Marion; Cecile; Mirian; Edith; Marjolein; Rudolf (nooit gedacht dat jongens zo
goed konden roddelen); Soesja; Anton; Ulricke; Sara en Kaska.
Daarnaast wil ik alle medewerk(st)ers van de Klinische Farmacologie bedanken voor de
prettige sfeer. Dick de Zeeuw, je manier van benaderen heeft me gescherpt, dedankt
daarvoor. Ad Nelemans, mocht je ooit ons ‘cannabis’ receptor artikel afkrijgen, je weet me
te vinden! Ook de secretaresses van de klinische farmacologie (Alexandra, Ardy en Paula),
maar vooral ook van de Cardiologie (Gretha) wil ik bedanken voor hun onmisbare
ondersteuning.
Tijdens mijn promotie onderzoek heb ik samengewerkt met een paar andere vakgroepen.
Met regelmaat liep ik rond in het Laboratorium voor Celbiologie en Electronenmicroscopie.
Ik wil alle medewerkers van deze afdeling bedanken maar expliciet de vakkundige analist
EHB Blaauw (waar stond EHB ook alweer voor?). Bert heeft menig stukje weefsel van
patiënten en geiten in superdunne plakjes gesneden. Bert ik wil je nog hartelijk danken
voor je geweldige inzet en interesse in het onderzoek, maar ook voor de prettige sfeer bij
jullie op de afdeling. Dan kom ik meteen bij Han van der Want terecht. Han ook jij bedankt
voor je inzet, openhartigheid en je creatieve oplossingen. Verder wil ik nog Peter van der
Syde bedanken voor de opmaak van het proefschrift. Ik hoop in de toekomst ook weer met
jullie samen te werken.
In de eindfase van mijn onderzoek kwam ik Jannie Ausma tegen tijdens een congres in
Atlanta U.S.A.. Ik vertelde Jannie wat voor experimenten ik graag zou willen doen en
Jannie stelde meteen voor om eens in Maastricht te komen. Ik ben nog nooit iemand
tegengekomen met zo’n toewijding aan het onderzoek. Jannie, het was voor mij dan ook
erg prettig dat je me zo goed hielp met de experimenten, maar vooral ook met het schrijven
van het artikel (hoofdstuk 9). Ik ben me ervan bewust dat dit artikel zonder je kennis nooit
het huidige niveau zou hebben gehaald. Bedankt Jannie! Ik ben tweemaal in Maastricht
geweest en daar heeft Hans Duimel me geholpen bij de uitvoer van de experimenten. Hans
hartelijk bedankt voor je inzet en deskundige begeleiding gecombineerd met je (Brabantse)
gezelligheid.
Tsja, en dat ik na mijn promotie-onderzoek, het ter wereld brengen en (op)voeden van drie
kinderen en de hectiek waarmee dit gepaard gaat nog niet overspannen ben komt vooral
door mijn fantastische partner Marcus. Marcus, Myrthe, Joachim & Jona, ik wil jullie
bedanken voor het feit dat jullie me iedere dag laten zien waar het leven eigenlijk om
draait......