Anti-arrhythmic effects of atrial specific IKur block: A simulation study

Abstract

The ultra rapid potassium current (I_Kur) is an attractive pharmacological target in atrial fibrillation (AF) management due to its atrial specific nature. An experimentally based 78% I_Kur reduction was incorporated into a human atrial action potential (AP) model under sinus rhythm (SR) and atrial fibrillation (AF) conditions. Its effects on cell and tissue level electrical activity were simulated. I_Kur block reduced AP duration (APD) and effective refractory period (ERP) under SR conditions, but prolonged APD and ERP under AF conditions. I_Kur block modulated tissue's ability to sustain high pacing rate conduction under SR and AF conditions. Vulnerability window (VW) was augmented under SR, and reduced under AF conditions. I_Kur block did not effect on re-entrant waves in 2D and 3D simulations. Simulations show pro-arrhythmic effects in SR, but anti-arrhythmic effects in AF case due to I_Kur block.

Full text

Page 1

Anti-arrhythmic Effects of Atrial Specific IKur Block: A Simulation Study
P Law, S Kharche, J Higham, H Zhang
School of Physics and Astronomy, University of Manchester, Manchester, UK
Abstract
The ultra rapid potassium current (IKur) is an attractive
pharmacological target in atrial fibrillation (AF)
management due to its atrial specific nature. An
experimentally based 78%
incorporated into a human atrial action potential (AP)
model under sinus rhythm (SR) and atrial fibrillation
(AF) conditions. Its effects on cell and tissue level
electrical activity were simulated.
IKur block reduced AP duration (APD) and effective
refractory period (ERP) under SR conditions, but
prolonged APD and ERP under AF conditions. IKur block
modulated tissue’s ability to sustain high pacing rate
conduction under SR and AF conditions. Vulnerability
window (VW) was augmented under SR, and reduced
under AF conditions. IKur block did not effect on re-
entrant waves in 2D and 3D simulations. Simulations
show pro-arrhythmic effects in SR, but anti-arrhythmic
effects in AF case due to IKur block.


1. Introduction
IKur reduction was
Atrial fibrillation (AF) is the most common arrhythmia
and affects a large part of the elderly population in
developed nations. Clinical management of atrial
fibrillation (AF) is largely rhythm control and achieved
by administration of Class I and III pharmacological
agents [1, 2]. While interventions with such agents are
common, their medium to long term effectiveness in
maintaining sinus-rhythm (SR) is limited with suboptimal
efficacy and adverse side effects [3]. Pharmacological
agents aim to alleviate AF by blocking key atrial ion
currents responsible for the atrial myocyte repolarisation
thereby lengthening the action potential duration (APD)
and effective refractory periods (ERP). Increased ERP
lengthens the wavelength at which the tissue can sustain a
re-entrant circuit and therefore limits the tissues
susceptibility to the genesis of AF. However, the
ventricles share many of the same ion channels as the
atria therefore it is not possible to alter ion channel
properties in the atria with out also altering the electro-
physical properties of ventricular cells. This leads to
adverse side effects of ventricular pro-arrhythmia.
Current drug development is increasingly focused on
developing atrial specific pharmacological agents [4].


Figure 1 A: Top panel shows AP profiles under SR (solid
line) and IKur blocked under SR (dashed line) conditions.
Bottom panel shows the corresponding IKur during the AP.
B: Top panel shows AP profiles under AF (gray line) and
bottom panel shows the corresponding IKur traces during
the AP.

The benefits of blocking IKur are being increasingly
identified with pharmacological agents being developed
to provide a robust AF therapy with minimal risk [5, 6]. It
has been experimentally shown by Wettwer et al. [7] that
low concentrations (~ 10 – 50 ?M/L) of 4-aminopyridine
(4-AP) selectively blocked IKur in human atrial myocytes
without affecting properties of other ionic currents. They
reported that an IKur block resulted in a shortening of
APD. In contrast the application of 4-AP to AF cells was
observed to give a modest increase in APD [7].
While previous experimental data [7] and simulation
[8] studies have identified that blocking of IKur results in a
shortening of APD, the full effects have not been studied
in detail. The present multi-physics simulation study
focused on quantifying the anti-arrhythmogenic effects of
blocking IKur at cellular, tissue and organ levels under SR
and AF conditions. This case study uses previously
developed simulation environments [9].

2. Methods
The biophysically detailed model for human atrial
ISSN 0276−6574
429
Computing in Cardiology 2010;37:429−432.

Page 2

action potential (AP) developed by Courtemanche et al.
[8] (CRN model) was implemented in the simulations.
The model is well established and has been used in
previous simulation studies [10-13]. The model was
modified to simulate APs under normal SR conditions
(standard CRN model), IKur block under SR conditions
(SRB), AF due to ion channel remodelling, and IKur block
under AF conditions (AFB). The IKur block was
implemented as a 78% reduction of the IKur current based
on experimental data [7]. AF was simulated as described
in our previous studies [12, 14]. In brief, simulating AF
consists of a 98% increase in the inward rectifying
potassium current (IK1), a 61% reduction in the transient
outward current (Ito), a 68% reduction in the in the L-type
calcium current (ICaL), a 52% reduction in the IKur current,
and a 67% increase in the sodium-calcium exchanger
pump (INaCa). IKur block was simulated as a further 78%
blocking of IKur under the SR or AF conditions. APs were
evoked by a series of 10 conditioning supra-threshold
stimuli at pacing cycle length (PCL) of 1 s, with strength
2 nA/pF and duration 2 ms. Such a conditioning was
deemed sufficient to elicit further stable APs. The 11th AP
was then noted for further analysis. APD was defined as
the time interval taken from the 11th stimulus to the time
when the evoked AP reached 90% repolarization. APD
restitution (APDr) were computed using standard S1-S2
protocol where a premature stimulus (S2) was applied at a
given time after the 10th conditioning stimulus of S1 (PCL
= 1 s), as in our previous study [12]. Diastolic interval
(DI) was defined as the time interval between 90%
repolarization of the previous AP and the upstroke of the
final AP. A plot of the measured APD against DI gave the
APDr curves. Maximal slopes of the curves were
determined. ERP was defined as the minimum S1-S1
stimulus interval that produced an AP with peak potential
over 80% of that of the final S1 evoked AP [15]. ERP
was simulated over a range of PCLs and ERP restitution
(ERPr) curves were constructed.
The cell models were then incorporated into a reaction-
diffusion parabolic partial differential equation (PDE) to
construct 1D, 2D and 3D mono-domain models of
spatially extended homogeneous atrial tissue. The
parabolic PDE has the form
 
2
/
mion
CutDuI

(1)
where Cm is cell membrane capacitance, D is the uniform
electrotonic diffusive coupling
intercellular gap junctional electrical coupling, u is the
membrane potential, and Iion is the total membrane
current. In the models, D was set to 0.031 mm2/ms to
produce a physiological conduction velocity (CV) of
0.27 mm/ms in a solitary excitation wave under SR
conditions [11]. In the 1D and 2D tissue models, the
inter-cellular distance was taken to be 0.1 mm. Such a
space step is close to the length of human atrial cells and
is also sufficiently small to give stable numerical
that models the
solutions. The 1D strand model of homogenous atrial
fibres had a length 20 mm, discretized to consist of 200
coupled cells. Using the 1D model, CV restitution (CVr)
and temporal vulnerable window (VW) of the atrial tissue
were computed using methods described in a previous
study [11]. Re-entrant wave dynamics were studied using
2D and 3D models. Using stimulation protocols as in
previous study [9], re-entrant excitation scroll waves were
initiated. The anatomically detailed 3D model of human
atria was developed in our previous study [13]. The 3D
model has a spatial resolution of 0.33 mm x 0.33 mm x
0.33 mm. The models were numerically solved using the
forward marching explicit Euler single step method with
a constant time step of 0.005 ms, which gives accurate
solutions without compromising computation time [9].



Figure 2 Effects of IKur block on cell and 1D model
properties. A: APDr curves; B: Maximum slopes of APDr
curves; C: ERPr curves; D: CVr curves.

3. Results
Solitary APs show an abbreviation due to IKur block
under SR conditions, and a prolongation under AF
conditions as shown in Figure 1. The SR APD of 314.13
ms was abbreviated to 299.2 ms under SRB conditions,
and the characteristically short AF APD of 155.5 ms was
prolonged to 180.6 ms under AFB conditions. APDr
curves (Figure 2, A) show an increased maximal slope as
shown in Figure 2, B. The maximal slopes under SR
conditions were observed to be 2, to be 1.2 under SRB
conditions, 2.15 under AF and 2.36 under AFB
conditions. The rate dependence of the ERP is shown in
Figure 2, C. Under SR conditions the ERP at a PCL of 1 s
was measured to be 327.9 ms. Under SRB conditions the
ERP was reduced by 2.5% to 319.8 ms indicating that IKur
blocking increases the tissue’s susceptibility to AF.
However, a sudden dramatic reduction in ERP was
observed at high PCLs and the cut off PCL is measured to
be 417.7 ms, 27.1% higher than under SR conditions.
430

Page 3

This indicates that blocking IKur under SR conditions
provides an arrhythmogenic effect by reducing the ERP.
Under AF conditions the ERP is reduced to be 167.7 ms
which is 48.8% shorter then the SR case and the cut off
BCL was measured to be 312.0 ms, a 5.1% reduction
from the SR case. In the AFB case the ERP was found to
be reduced by 41% at 193.5 ms and the cut off BCL was
found to be reduced by 19.4% to 265.0 ms. The CVr
curves are shown in Figure 2, D. Solitary wave CV under
SR and SRB were measured to be 0.26 mm/ms, and CV
in under AF and AFB was measured to be 0.25 mm/ms
indicating that the IKur block did little to alter the CV at
low pacing rates. However, CVr shows that SRB shifted
the restitution curve in the negative direction, indicating
an increased propensity to sustained wave propagation at
high pacing rates, whereas AFB results in a shift of the
CVr curve to the positive direction from the AF case
indicating a loss of excitation propagation at high pacing
lengths. This suggests that under AF conditions, blocking
IKur has an anti-arrhythmogenic effect.
2D re-entry simulations are shown in Figure 3. Under
SR conditions the lifespan (LS) of spiral waves was
measured to be 1.4 s [9]. The dominant frequency (DF) of
localized excitations was measured at 3 Hz and the tip
meander area was 5.25 cm2. In contrast under AF
conditions, the stable spiral wave persisted for the total
simulated duration of 10 s. DF was increased to 8.2 Hz
and the tip meander area was reduced to 4 cm2. Under
SRB conditions, LS of the spiral wave was around 1 s.
The measured DF was 4.6 Hz. However, the tip meander
area was reduced substantially due to IKur block. Under
AFB conditions the stability and persistence of the spiral
wave was unaffected as compared to the AF case.
3D simulations along with time traces and dominant
frequency are shown in Figure 4. Under SR conditions,
scroll waves self-terminated after 4.2 s and the DF was
measured to be 3 Hz. Under SRB conditions, a moderate
reduction of LS and DF from SR conditions to 4.1 s and
2.6 Hz respectively. In the AF case scroll waves persisted
for the simulated period of 10 s, and had a DF of 7.4 Hz.
In case of AFB, the re-entrant wave persisted for the
period of the simulation and the DF was also reduced
from AF conditions to 6 Hz. The anti-arrhythmic
properties of the IKur blocking were observed in a
reduction of the power of the re-entrant wave.


4. Conclusions and discussion
This simulation study shows that IKur channel blocking
under normal SR conditions results in a reduction of APD
and ERP which increases arrhythmogenicity. However,
blocking of IKur under AF conditions prolongs AP and
increases ERP. While simulations in the single cell
indicate the anti-arrhythmic effects of blocking the IKur
current, re-entrant dynamics in the 2D and 3D simulations
show that blocking IKur does not play a significant role in
the dissipation of re-entrant spiral waves. However, in the
2D simulation the tip-meander area is reduced
substantially. This suggests that the blocking of the IKur
channel alone is not effective in alleviating the onset of
AF and surgical intervention is required.



Figure 3 Re-entry in 2D homogenous sheets of human
atrial tissue. Data for SR, SRB, AF, and AFB are shown
in top, second, third, and bottom panels respectively.
Columns I, II, and III show illustrative frames from the
2D simulations. Column VI shows trace of spiral wave
tip. Column V shows AP traces from representative
locations in the 2D sheet, and Column VI shows the
frequency spectrum in the AP traces.

Atrial selective drugs which target atrial specific ion
channels, such as IKur, provide an attractive target for drug
therapy in the treatment of AF without adversely
affecting the electrophysiological function of the
ventricles. However, while blocking of the IKur has been
shown experimentally, and in our simulations to prolong
APD in cardiac myocytes which have undergone AF
induced electrical remodelling, current pharmacological
agents which block IKur (e.g. Vernakalant, AZD7009, and
AVE0118) also block the sodium current (INa) and the
transient outward current (Ito), resulting in long term
complications if administered for extended durations. Our
simulations confirmed that the blockage for patients who
have undergone electrical remodelling may limit the
occurrence of AF and aid in rhythm control but is not
effective once a spiral wave has been initiated.
The potential limitations of the CRN model used in
this study have been discussed previously [8, 16]. A
limitation of our 3D simulations is the absence of
electrical heterogeneity in the atria, and only consists of
atrial cell types while it is known that cells in different
regions have different electrophysiological properties
[13]. The spatial heterogeneities due to fiber orientation
are also neglected in our simulations and we have
implemented a uniform diffusion tensor in our
431

Page 4

simulations.


Figure 4 Scroll waves in 2D homogenous human atria.
Data for SR, SRB, AF, and AFB are shown in top,
second, third, and bottom panels respectively. Column I,
II, and III illustrate the simulation. Column IV shows AP
traces from representative locations in the 3D model, and
Column VI shows the frequency spectrum in the AP
traces.

Acknowledgements
This work was supported by EPSRC and Wellcome
Trust (UK) grants (WT/081809/Z/06/Z).

References
[1] Nattel S, Singh BN. Evolution, mechanisms, and
classification of antiarrhythmic drugs: focus on class III
actions. Am J Cardiol. 1999; 84(9A):11R-9R.
[2] Tamargo J, Valenzuela C, Delpon E. New insights into the
pharmacology of sodium channel blockers. Eur Heart J.
1992; 13 Suppl F:2-13.
[3] Tamargo J, Caballero R, Delpon E. Pharmacological
approaches in the treatment of atrial fibrillation. Curr Med
Chem. 2004; 11(1):13-28.
[4] Tamargo J, Caballero R, Gomez R, Delpon E. I(Kur)/Kv1.5
channel blockers for the treatment of atrial fibrillation.
Expert Opin Investig Drugs. 2009; 18(4):399-416.
[5] Brendel J, Peukert S. Blockers of the Kv1.5 channel for the
treatment of atrial arrhythmias. Curr Med Chem Cardiovasc
Hematol Agents. 2003; 1(3):273-87.










[6] Ford JW, Milnes JT. New drugs targeting the cardiac ultra-
rapid delayed-rectifier current
pharmacology and evidence for potential therapeutic value. J
Cardiovasc Pharmacol. 2008; 52(2):105-20.
[7] Wettwer E, Hala O, Christ T, Heubach JF, Dobrev D, Knaut
M, et al. Role of IKur in controlling action potential shape
and contractility in the human atrium: influence of chronic
atrial fibrillation. Circulation. 2004; 110(16):2299-306.
[8] Courtemanche M, Ramirez RJ, Nattel S. Ionic mechanisms
underlying human atrial action potential properties: insights
from a mathematical model. Am J Physiol. 1998; 275(1 Pt
2):H301-21.
[9] Kharche S, Seemann G, Margetts L, Leng J, Holden AV,
Zhang H. Simulation of clinical electrophysiology in 3D
human atria: a high-performance computing and high-
performance visualization application. Concurrency and
Computation: Practice and Experience. [Journal Article].
2008; 20(11):10.
[10] Zhang H, Garratt CJ, Zhu J, Holden AV. Role of up-
regulation of IK1 in action potential shortening associated
with atrial fibrillation in humans. Cardiovasc Res. 2005;
66(3):493-502.
[11] Kharche S, Garratt CJ, Boyett MR, Inada S, Holden AV,
Hancox JC, et al. Atrial proarrhythmia due to increased
inward rectifier current (I(K1)) arising from KCNJ2
mutation--a simulation study. Prog Biophys Mol Biol. 2008;
98(2-3):186-97.
[12] Kharche S, Zhang H. Simulating the effects of atrial
fibrillation induced electrical remodeling: a comprehensive
simulation study. Conf Proc IEEE Eng Med Biol Soc. 2008;
2008:593-6.
[13] Seemann G, Hoper C, Sachse FB, Dossel O, Holden AV,
Zhang H. Heterogeneous three-dimensional anatomical and
electrophysiological model of human atria. Philos Transact
A Math Phys Eng Sci. 2006; 364(1843):1465-81.
[14] Kharche S, Seemann G, Leng J, Holden AV, Garratt CJ,
Zhang H. Scroll Waves in 3D Virtual Human Atria: A
Computational Study. In: Seemann FBSaG, editor. LNCS;
2007. p. 129–38.
[15] Workman AJ, Kane KA, Rankin AC. The contribution of
ionic currents to changes in refractoriness of human atrial
myocytes associated with chronic atrial fibrillation.
Cardiovasc Res. 2001; 52(2):226-35.
[16] Cherry EM, Evans SJ. Properties of two human atrial cell
models in tissue: restitution, memory, propagation, and
reentry. J Theor Biol. 2008; 254(3):674-90.

Address for correspondence:
Name: Dr. Sanjay Kharche
Full postal address: School of Physics and Astronomy,
University of Manchester, Manchester, UK, M13 9PL
E-mail address: Sanjay.Kharche@manchester.ac.uk
(I Kur): rationale,
432