I Blockade Contributes Importantly to Drug-Induced...

DOI: 10.1161/CIRCEP.113.000239

1

IKs Blockade Contributes Importantly to Drug-Induced Long QT Syndrome

Running title: Veerman et al.; Drug-induced long QT syndrome by IKs blockade

Christiaan C. Veerman, MD1*; Arie O. Verkerk, PhD2*; Marieke T. Blom, MA, MsE1;

Christine A. Klemens, BSc1; Pim N.J. Langendijk, PharmD4,5; Antoni C.G. van Ginneken, PhD2;

Ronald Wilders, PhD2; Hanno L. Tan, MD, PhD1,3

1Heart Center, 2Dept of Anatomy, Embryology & Physiology, 3Dept of Cardiology, 4Dept of Hospital Pharmacy, Academic Medical Center, University of Amsterdam, Amsterdam; 5Dept of

Hospital Pharmacy, Reinier de Graaf Group Hospitals, Delft, The Netherlands*contributed equally

Correspondence to:

Hanno L. Tan, MD, PhD

Heart Center, room K2-109

Academic Medical Center

Meibergdreef 15, 1105AZ

(P.O. Box 22700)

1100DE Amsterdam

The Netherlands

Tel: +31-20-5663264

Fax: +31-20-6975458

E-mail: [email protected]

Journal Subject Codes: [132] Arrhythmias - basic studies, [5] Arrhythmias, clinical electrophysiology, drugs

DDD11,1,1 3333

C pP DH

Ceeeentttter, 2Dept ofofoff AAAnananan totototomymymymy,,, EmEmEmEmbrbrbrb yoyoy looogygygy && PPhyyyysisiisiolololologoggy,y,, 3333DeDeDeepppt offff CCCCararardididid ololoo oogo y,y,y,y 4444DeDDD pPhahahaharrrmr acy, Acaaddemmmicc Meedddid cal Ceeennterr, UUUniviversiiitttty yyy ofoofo AAAmmmsterdrdammm, Amsstss erere dadam;;; 555DHospspsppitititalaal PPPPhahahaharmrmmaaca y,y,y,y, RRRReinininiiererer dddeee GGrararaafafaff GGGrorr upupu HHHHossspipipiitaatat lslsls, , DeDeDeD lflflftt, Theheheh NNNNetetetthhehh rlrlrlaana dssss

*contributuuu eddd eeequqqq ally

by guest on May 14, 2018

http://circep.ahajournals.org/D

ownloaded from

by guest on M

ay 14, 2018http://circep.ahajournals.org/

Dow

nloaded from



ownloaded from

by guest on M


Dow

nloaded from



ownloaded from

by guest on M


Dow

nloaded from



ownloaded from

http://circep.ahajournals.org/







DOI: 10.1161/CIRCEP.113.000239

2

Abstract

Background - Drug-induced long QT syndrome (LQTS) is generally ascribed to inhibition of the

cardiac rapid delayed rectifier potassium current, IKr. Effects on the slow delayed rectifier

potassium current IKs are less recognized. Triggered by a patient who carried the K422T

mutation in KCNQ1 (encoding the -subunit of the IKs channel), who presented with excessive

QT prolongation and high serum levels of norfluoxetine, we investigated the effects of fluoxetine

and its metabolite norfluoxetine on IKs.

Methods and Results - ECG data from mutation carriers and non-carriers revealed that the

K422T mutation per se had mild clinical effects. Patch-clamp studies, performed on HEK293

cells, showed that heterozygously expressed K422T KCNQ1/KCNE1 channels had a positive

shift in voltage-dependence of activation and an increase in deactivation rate. Fluoxetine and its

metabolite norfluoxetine both inhibited KCNQ1/KCNE1 current, with norfluoxetine being the

most potent. Moreover, norfluoxetine increased activation and deactivation rates. Computer

simulations of the effects of norfluoxetine on IKs and IKr demonstrated significant action potential

prolongation, to which IKs block contributed importantly. While the effects of the mutation per se

were small, additional IKs blockade by norfluoxetine resulted in more prominent QTc

prolongation in mutation carriers than in non-carriers, demonstrating synergistic effects of innate

and drug-induced IKs blockade on QTc prolongation.

Conclusions - IKs blockade contributes importantly to drug-induced LQTS, especially when

repolarization reserve is reduced. Drug safety tests might have to include screening for IKs

blockade.

Key words: torsade de pointes, long QT syndrome, potassium channels, drug-induced long QT syndrome, fluoxetine

p

chchchchanananannenenenelslslsls hhhhadadadad aaaa ppppososssitiititivivvv

on ratttte.e.e.e FFFFlulululuoxoxoxoxetetetetinininneeee ana

t

nt. Moreover, norfluoxetine increased activation and deactivation rates. Compute

s o

o

on in mutation carriers than in non-carriers, demonstrating synergistic effects of i

nororororflflflfluououou xexexexetititit neee bbbboth inhibited KCNQ1/KCCCNENN 1 current, wwwith hhh nnon rfluoxetine being t

nttt.t MMMMoreover, nnnoorflfllfluouououoxexexexetiiitinenn iiiincncncncreeeaasedededd aaactivvaatioooonn n annnnd dd ddedd aaactitititivavav tionononon ratatatesesess. CoCoCoC mpmpmpmpuutuu e

s ofofoff ttthe efffffeeeectss oof nnnoorfluououoxexx tinennn onn n IKKsKII andd IIIKrrrII dddemmmmoonssstrrrar tedd ssigngngngnifiii icanananantt t aacttionnn pppor

on, tooo whichchchh IKsKsKs II blblblocoo k cococoontnnntributed immmmpopp rtrtrtrtanananntly.yy WWWhile the eeeffects oooof thhhhe e mutation

,,, adddddididi itiionalll IIIKsKK II blblblblockkak ddde bbbby yy no ffrflull oxetine resultllt ddedd iiin momooore ppprorrr imiinent QTQTQTQ c ccc

onon iin n mumutatatat titititionon ccararririiierererss thththhanan iiiin n nonon-n caacarrrrieiei rsrs,, dedded mooomonsnsttrtrt atatiniini g g sysynenergrgisisii tititic c efefffefectcts s ofof ii by guest on May 14, 2018


ownloaded from


DOI: 10.1161/CIRCEP.113.000239

3

Introduction

The ability of drugs to prolong the QT interval on the ECG is one of the leading causes of drug

withdrawal from the market.1 Similar to its hereditary forms, drug-induced long QT syndrome

(LQTS) may culminate in polymorphic ventricular tachycardia (Torsade de Pointes, TdP) and

ventricular fibrillation (VF), which cause syncope and sudden death.2,3 Drug-induced LQTS is

commonly ascribed to blockade of the fast component of the delayed rectifier current (IKr) as a

result of a pharmacodynamic interaction of the drug with cardiac potassium-channel proteins.4,5

Accordingly, compounds under drug development are routinely screened for their (undesired)

ability to block IKr.6 Yet, anecdotal reports indicate that inhibition of slow delayed rectifier

current (IKs) may also contribute to drug-induced LQTS.7,8 This possibility has so far received

less recognition.

Fluoxetine, a widely prescribed antidepressant, is known as a potential LQTS-inducing

drug,9 particularly at high serum levels.10,11 Fluoxetine not only blocks IKr acutely, but also

inhibits protein trafficking of the IKr channel, thus further reducing IKr. Norfluoxetine, the active

metabolite of fluoxetine, has similar effects.9 However, the effects of fluoxetine and

norfluoxetine on IKs are unknown. To establish whether IKs blockade by these drugs contributes

to LQTS, we conducted the present study. This study was triggered by a patient who presented

with TdP at mildly elevated serum levels of norfluoxetine. Genetic analysis revealed a mutation

(K422T) in KCNQ1, which encodes the major protein that carries IKs. This mutation was

identified previously, but not studied in detail, in a clinical study on LQTS patients.12 Employing

ECG analysis, cytochrome P450 genotyping, patch-clamp analysis, and modeling studies, we

provide evidence that IKs blockade by fluoxetine and norfluoxetine contributes importantly to the

TdP-inducing potential of these drugs.

ned for their ((undeeeesisisisirer

low ddddelelllayayeddedd rrecectitiiififififieeerer

s7 8 v

n

o c

i

otein trafficking of the I channel th s f rther red cing I Norfl o etine the a

s) mamamamayy yy alalallsososos conononnttrt ibute to drug-induced LQLQLQTTS.7,8 This popopossibibibbilii ity has so far receiv

niiiitiooono .

oxetititiinennne, a wiiiideddd llly pprescribebebebedddd antiiiidedededeprprpreessa ttnt, is kkkknononowwwwn aaassss a potttet nttttiaiaiai l lll LQLQLQLQTSTSTS-iiinddduc

iculalll rllly yy at hhhigigigh hhh serum lell vellsll .1000,111111 FFFlul oxetiiine nott onlylyly bbblolooockkksss IIIIKrII acuttellly,y,y bbbbutututu aaaalso

tteiin tt ffffiickiki ff thth II hha lel thth ff thth ded iin II NNo frfll tetiin tthhe



ownloaded from


DOI: 10.1161/CIRCEP.113.000239

4

Methods

A detailed description of the methods can be found in the Online Supplement.

Clinical studies

ECGs were obtained from the index patient (screened for mutations in KCNQ1 and KCNH2) and

7 relatives (screened for the K422T mutation in KCNQ1). The index patient underwent analysis

of the serum concentrations of fluoxetine and norfluoxetine as well as genotyping of CYP2D6

with particular interest in determining the presence of CYP2D6 *3, *4, *5, *6, *7 or *8 alleles,

or other CYP2D6 deletions or duplications. Written informed consent was obtained from all

participants.

cDNA constructs, mutagenesis, and heterologous expression

Constructs of wild-type KCNQ1 and -subunit KCNE1 cloned in pSP64 (kindly provided by Dr.

M.C. Sanguinetti, University of Utah, USA) were subcloned into the mammalian expression

vector pCGI. The KCNQ1 K422T mutation was created by using the QuikChange site-directed

mutagenesis kit (Stratagene, USA), according to manufacturer’s protocol.

HEK293 cells were transfected with 1 g of wild-type KCNQ1 cDNA and 1 g KCNE1

cDNA (WT-KCNQ1/KCNE1), or, to recapitulate a heterozygous state, 0.5 g of both wild-type

and mutant KCNQ1 in addition to 1 g KCNE1 (HET-KCNQ1/KCNE1).

Electrophysiological experiments

KCNQ1/KCNE1 currents were recorded at 36±0.2°C by the amphotericin-perforated patch-clamp

technique. Activation and deactivation kinetics of KCNQ1/KCNE1 currents were determined by

voltage clamp protocols diagrammed in the accompanying figures. Current densities were

calculated by dividing current amplitudes by cell membrane capacitance. Voltage-dependence of

activation and the time course of current (de)activation. were analyzed as detailed in the Online

was obtained from m m m ala

s

b

uinetti, University of Utah, USA) were subcloned into the mammalian expressio

G c

is kit (Stratagene USA) according to man fact rer’s protocol

strururuructctctctsss, mmmmutu aggggeenee esis, and heterologousss eeexpxpression

oooof ff wild-type KCKCKCNQNQNQ1 andndndd -ssubububunnit KCKK NENE1 ccclololonenenenedd innn pppSPP644 ((((kikikik ndlyyyy ppproovvideeeddd b

uinetettttititti, UnUnUnU iver iisiittytyt offff Utahahahah, UUUUSA)A)AA) wwwwererere subbbclooneneneneddd d inininnto tttthheheh mammamamamalililil an expressiioii

GI. TTTThehh KCKCKCNQNQNQQ111 1 K4K4K4K 222222TTT mutatiiiion was createdddd bbby yy using gg ththththe QuQuQuuikikikkChChChangegg ssssitititite-e-e-e dddid rec

ii kikitt (S(Sttr tat UUSASA)) didi tt ffa tct ’’s tto ll



ownloaded from


DOI: 10.1161/CIRCEP.113.000239

5

Supplement. Effects of forskolin, fluoxetine, norfluoxetine, tramadol, and codeine (all Sigma-

Aldrich, USA) on KCNQ1/KCNE1 currents were tested in paired measurement 5 minutes after

onset of bath application at a membrane potential of +110 mV. At this potential, possible effects

on voltage dependency of activation do not interfere with effects on current density. Dose-

response curves were fitted to the Hill equation (see Online Supplement).

Computer simulations

Functional differences between WT-KCNQ1/KCNE1 and HET-KCNQ1/KCNE1 were tested by

computer simulations using the midmyocardial version of the ten Tusscher et al. human

ventricular cell model,13 as updated by ten Tusscher and Panfilov.14 The experimentally observed

mutant- and drug-induced changes in electrophysiological properties of KCNQ1/KCNE1 current

were implemented as changes in IKs. Previously reported norfluoxetine-induced IKr blockade was

also implemented, resulting in 72% decrease in IKr current density at a norfluoxetine

15 were not taken into

account.

Statistics

Data are presented as mean±SEM. Statistical comparisons were made as detailed in the Online

Supplement. P<0.05 defines statistical significance.

Results

Clinical studies

A 62-year-old woman presented with repeated syncope. ECGs revealed severe QT prolongation

(maximal rate-corrected QT interval [QTc, Bazett’s formula] 620 ms, Figure 1A), TdP (Figure

1B), and VF. Her history revealed ventricular tachycardia after minor surgery at age 54, which

was not explored further. She had hypertension and diabetes. Physical examination and serum

scher et al. human n

he exxpepep iiririmementtntalallylyyy oooobbbsbs

d u

e d

m

d drdrdrugugugug-iiindndndnducuu edededd changes in electrophysiolololologgical properttties ofofof KCNQ1/KCNE1 cu

ememememented as chanangesss iiin IKsKsKsIIII . Preve iooouuslyy rrrepororted ddd nonononorfrrfr lluoxoxoxetinne-innnndudducedd dd IIIIKrII bbloccckkkad

menttttedededed, rrrresultltltltiiing iniii 77772%%% ddddecececrease eee ininin IIIIKrII currentntntnt dddenenenensitytytyy at ttt a norffflulululuoxoxoxo etiniii e

151515 wwwwere not ttakekekeken nnn inininnto



ownloaded from


DOI: 10.1161/CIRCEP.113.000239

6

electrolytes were unremarkable. Echocardiography revealed mild concentric left ventricular

hypertrophy, but was otherwise unremarkable. There were no signs of ischemic heart disease or

heart failure. There was a family history of sudden death in one relative (father, aged 52). Other

relatives had no history of LQTS-related symptoms (Figure 2). Her medication included

fluoxetine (20mg OD), which she had used for >2 years, and no other QT prolonging drugs.

Serum levels of fluoxetine and norfluoxetine at admission were 70 and 1,230 ng/mL, the latter

reaching 1,490 ng/mL one day after admission (therapeutic ranges are 100-450 ng/mL and 50-

350 ng/mL,16 respectively). The increased serum levels of norfluoxetine were ascribed to the fact

that she had recently started using codeine (15 mg OD) and tramadol (100 mg OD). Both drugs

are substrates of cytochrome CYP2D6, and compete with fluoxetine and norfluoxetine as

substrates for CYP2D6.17,18 Upon discontinuation of fluoxetine, codeine, and tramadol, QTc

duration normalized to 434 ms, and TdP did not recur. Genetic screening for CYP2D6 revealed

that she was an extensive metabolizer of CYP2D6 with genotype CYP2D6*1/*1. No additional

mutations, deletions or duplications in these CYP2D6 alleles were found.

Initial genetic screening for mutations in KCNH2 revealed no mutations. However, she

was found to be a heterozygous carrier of a missense mutation in KCNQ1 (K422T), which

encodes the pore-forming subunit of the IKs channel. This variant, previously associated with

LQTS12, was not found in 400 alleles of control individuals. To assess the clinical consequences

of the K422T mutation, 8 family members of the index patient were invited for genetic screening

and functional testing (ECGs at rest and exercise ECGs). Of these family members, 7 accepted to

undergo genetic screening and ECGs at rest, while 5 also accepted to undergo exercise testing.

Genetic analysis revealed 3 more carriers of the mutation (Figure 2). At rest, QTc durations were

not prolonged in mutation carriers (<450 ms for males and <470 ms for females3). Within

ne were ascribed ttto o o o tht

(100 mmgg ODODODOD).))) BBBBotototthh h h ddd

t

f T

ormalized to 434 ms, and TdP did not recur. Genetic screening for CYP2D6 reve

as an extensive metabolizer of CYP2D6 with genotype CYP2D6*1/*1. No additi

deletions or d plications in these CYP2D6 alleles ere fo nd

tessss oooof fff cycycytototot chccc rororoommemm CYP2D6, and compeeetetete wwith fluoxetiine aaaannnd norfluoxetine as

foooro CYP2D6.1171 ,118 UUUppon diddid sconontinunnuationnn off ffluoxoxoxetetetetinne, codeiinne, annd trrrramamamadadol, QQQT

ormalillilizezezedddd to 44443434344 ms, andddd TTTTdPdPdPdP didddd nnnnototot recur. GeGeGeGennnnetititit cccc scccrererereeniiing forrrr CYCYCYCYP2P2P2P D6D6D6D6 reve

as an extensivivive eee metabobb lililizer offf CCCYPYPYP2D2D2D2 6 66 iiwi hththh gggen totypypype CYCYCYC P2P2P22D6D6D6D6*1*1* /*/*// 111. NNNNoooo adadaddditi

dd lel tetiio dd lpliic tatiio ii thth CYCYP2P2D6D6 llll lel ffo dd



ownloaded from


DOI: 10.1161/CIRCEP.113.000239

7

generation II, exercise ECGs were analyzed in 2 carriers and the index patient demonstrating

mild QTc prolongation. QTc intervals remained within normal range in all family members (1

carrier and 2 non-carriers) in generation III. However, given the small family size, the unequal

distribution of carriers and non-carriers over the generations and the difference in genetic

relations between carriers and non-carriers, it was impossible to complement these clinical

observations with an adequate statistical comparison.

Electrophysiological studies

Effects of the K422T mutation on KCNQ1/KCNE1 current

Figure 3A shows typical currents of WT-KCNQ1/KCNE1 and HET-KCNQ1/KCNE1 channels,

respectively. The steady-state currents of both channel types did not differ significantly (Figure

3B). Compared to WT-KCNQ1/KCNE1, HET-KCNQ1/KCNE1 channels displayed a +13.3 mV

shift (P=0.005) of steady-state activation (Figure 3C). Moreover, in the physiological range from

-20 to - –25% faster (P<0.05) in HET-KCNQ1/KCNE1 channels

(Figure 3E). The reversal potentials (Figure 3D) and activation time constants (Figure 3E) were

not significantly different. Because mild QTc prolongation during exercise, found in 3 out of 4

mutation carrie -adrenergic stimulation between

HET-KCNQ1/KCNE1 channels and WT-KCNQ1/KCNE1 channels, we studied the effects of

forskolin (10 μM), a stimulator of cAMP production, using the protocol shown in Figure 3A.

Forskolin significantly increased current density, decreased activation time constant, and shifted

the V1/2. However, these effects did not differ significantly between HET-KCNQ1/KCNE1 and

WT-KCNQ1/KCNE1 (Figure 4, A and F).

Effects of fluoxetine and norfluoxetine on KCNQ1/KCNE1 current

Next, we studied the effects of fluoxetine and norfluoxetine on current density and kinetics of

CNQQQQ1/1/1/1/KCKCKCKCNENENENE1111 chahahahannnnn

y

p

005) of steady-state activation (Figure 3C). Moreover, in the physiological range

e

) The re ersal potentials (Fig re 3D) and acti ation time constants (Fig re 3E)

y. ThThhThee e sttsteaeaeae dydd -s-s-s-stat te currents of both channnnenen l types did nonn t dididid ffff er significantly (Fi

paaaareeeed to WT-KCKCNQQQ111/KCCCCNNENN 1,111 HEHEET-KCKKCNQNQ1/KCKCKCK NENNEN 1 hchchannenels ddddiisi plaayayayedede aa +131313.3

005))) ofofofof sssstttet addddy- tttstattte activaaaatititition (FiFiFiFigugugug rerere 333C)C)C). MoMoMoMorrrreovovovover, inininin ttthehh phyysisisis olololologogicii llall range

–2252 %%%% ffaf tster ((((PPPP 00<0.0005)5)5)) iiin HEHEHEETTTT-KCKCKCNQNQNQNQ1/1/1/KCKCKCKCNENENEE1111 chchchc aaana ne

)) TThhe ll tte tntiialls ((FiFi 33D)D) dnd tctii tatiio titi tst tts ((FiFi 33E)E)



ownloaded from


DOI: 10.1161/CIRCEP.113.000239

8

WT-KCNQ1/KCNE1 and HET-KCNQ1/KCNE1 channels using the protocols shown in Figure

3A. The index patient experienced QTc prolongation with serum levels of norfluoxetine reaching

1,490 ng/mL (~5 μM). Accordingly, this concentration was used to assess the effects of this drug,

while the effects of fluoxetine were examined using a supratherapeutic concentration of 10 μM.

Figure 4, B and C, shows typical examples of the effects of both drugs on WT-KCNQ1/KCNE1

and HET-KCNQ1/KCNE1 currents, elicited by a voltage clamp step to +110 mV from a holding

fluoxetine resulted in 25% blockade, while norfluoxetine inhibited IKs by 55% at 5 μM. Both

effects did not differ significantly between WT-KCNQ1/KCNE1 and HET-KCNQ1/KCNE1

channels. Norfluoxetine also decreased activation and deactivation time constants at membrane

potentials of +110 and -

Again, these effects did not differ between WT-KCNQ1/KCNE1 and HET-KCNQ1/KCNE1.

To study whether the mutation changed the sensitivity to IKs blockade by norfluoxetine, a

dose-response curve was obtained by using concentrations between 0.01 and 100 μM (Figure 5).

We found that IC50 values for blocking effects of norfluoxetine on WT-KCNQ1/KCNE1 and

HET-KCNQ1/KCNE1 channels were 5.3±0.8 and 4.9±0.8 μM, with Hill coefficients of

0.78±0.10 and 0.89±0.14, respectively. All effects were reversible upon wash-out (data not

shown).

Effects of tramadol and codeine on KCNQ1/KCNE1 current

The occurrence of TdP in the index patient coincided with her recent use of tramadol and

codeine. Tramadol is not associated with QTc prolongation,19 while codeine and its metabolite

morphine are known to have only weak potential of blocking IKr current, with IC50 levels >100

μM above the therapeutic limit of 0.8 μM.20 To exclude any QTc prolonging effects of these

by y 55% at 5 μμM.. BBBBoto

HET--KCKCKCKCNQNQNQNQ1/1/1/1/KCKCKCKCNENENENE11

No oxet e so ecreas act at n an ac va on t e constants at memb

o

s 1

s t

nse c r e as obtained b sing concentrations bet een 0 01 and 100 M (Fig

Norfrfrflululul oxooxetetete ininini e alalallso decreased activation ananand ddd deactivationn timmmee e constants at memb

of fff +++1+ 10 and -

se effffffffeceectsststs dddidddd n tttot ddddiffififffffer bebebetwtwtwtween WTWTWTWT-KCKCKCKCNQNQNQQ1////KCKCKCKCNENENENE1 anananand ddd HEHEHEET-KCKCKCKCNQNQNQNQ1/1/1/KCKCKCKCNENENENE1

studydydyd whehh thhherererr thehhh mutatioii n chhhhangegg d ddd the sensiiitivii itiity yy to IIIKKKsKIIIII blblblocococckakkk dedd bbbbyyy norfrfrfrflulululuooxet

bobttaiin ded bb iin ttr tatiio bb tet 00 0011 dd 101000 MM ((FiFi



ownloaded from


DOI: 10.1161/CIRCEP.113.000239

9

drugs in the index patient, we evaluated the effects of tramadol and codeine on IKs. Neither

tramadol (30 μM; Figure 4, D and F), nor codeine (30 μM; Figure 4, E and F) altered the

electrophysiological properties of WT-KCNQ1/KCNE1 or HET-KCNQ1/KCNE1 channels.

Modeling studies

Having found that HET-KCNQ1/KCNE1 channels have altered kinetic properties which are

predicted to reduce net IKs, and that norfluoxetine also significantly reduced IKs (in addition to

IKr), we next assessed the physiological implications of these changes by conducting modeling

studies. Figure 6 shows the simulated action potential at 1Hz and corresponding IKr and IKs

currents of a non-carrier (Figure 6, A–C, black lines) and a heterozygous K422T mutation carrier

(Figure 6, D–F, black lines). The mutation prolonged the action potential by 28 ms (Figure 6, A

and D, black lines) as a result of reduction in IKs (Figure 6, C and F, black lines). This effect was

similar in simulations at higher frequencies (data not shown).

We next studied the effects of inhibition of IKr and IKs by 5 μM norfluoxetine. In non-

carriers, IKr inhibition by norfluoxetine prolonged action potential duration by 68 ms (Figure 6A,

blue dashed line). Norfluoxetine-induced IKs inhibition also caused action potential prolongation,

albeit less than IKr inhibition (28 ms, Figure 6A, red dotted line). However, combined IKr and IKs

inhibition resulted in synergistic effects, prolonging action potential duration by 118 ms (Figure

6A, arrow). In mutation carriers, where IKs is already reduced by the mutation, inhibition of IKr

by norfluoxetine caused more action potential prolongation than in non-carriers (+94 ms),

whereas inhibition of IKs had similar effects (+25 ms, Figure 6D, blue dashed line and red dotted

line). Here, inhibition of both IKr and IKs caused even larger synergistic effects than in non-

carriers, resulting in action potential prolongation by 154 ms (Figure 6D, arrow).

espop nding g IKrII anddr IIIIKsKsKKIIIIII

ous K4K4K4K422222222TTT T mumuttatat titititionononon c

D

c c

s

inhibition b norfl o etine prolonged action potential d ration b 68 ms (Fig r

D–F,F,F,F, bbbblalalalackckckck linnnesesee ). The mutation prolongegegedd dd the action pppotoo ennnntititial by 28 ms (Figure

ckkkk llllines) as a rressulttt oof redededductititiioon inn IKKsII (Figugure 666,,,, CCCC anddd F, bblaackkkk llliness).).)) Thihis eeeffffec

simululllatataatioiooions attt hihihighghhher fffreququuuenenencies ((((dadadadatatata n ttot shohohh wwnwnwn).)))

next studidd eddd tttthehehehe effffffects offf f iini hihihih bibibi itiion of fff IIIKrII a ddndr IIIKsII by yy 55 μMμMμM nnnnorflflfluoxetiitiineneee. InInInn no

ii hnhibibititiio bb flfl titi llo ded tctiio tte tntiiall dd titi bb 6868 (F(Fiig



ownloaded from


DOI: 10.1161/CIRCEP.113.000239

10

Discussion

We found that IKs blockade contributes importantly to pathologic drug-induced QT prolongation.

IKs blockade by norfluoxetine, not previously recognized, although mild in isolation, acted

synergistically with inhibition of IKr to cause marked action potential prolongation. This effect

was particularly large in the index patient, who had increased vulnerability to these effects

because she carried a KCNQ1 mutation that reduced IKs. Our modeling studies indicated that,

while IKr blocking effects of norfluoxetine alone would increase action potential duration by only

68 ms, the additional presence of IKs block by norfluoxetine and IKs reduction by the mutation

resulted in 154 ms prolongation.

We studied the effects of elevated norfluoxetine concentrations that corresponded to the

serum levels found in the index patient. Many situations are conceivable in which the therapeutic

levels of fluoxetine and norfluoxetine are exceeded. This report deals with a situation which may

occur relatively commonly, i.e., drug competition for metabolizing enzymes (CYP2D6). In the

index patient, recent concomitant use of tramadol and codeine, drugs that compete with

fluoxetine and norfluoxetine as substrates for CYP2D6, presumably increased serum

concentrations of norfluoxetine to above-therapeutic levels. Given that neither tramadol nor

codeine affected repolarizing currents, it is likely that it was this increase in serum norfluoxetine

levels which caused LQTS and TdP/VF in a patient who had used fluoxetine for >2 years

without arrhythmia. This notion was supported by CYP2D6 genotyping, which indicated that she

was an extensive metabolizer, not an ultrarapid metabolizer, in whom addition of extra CYP2D6

substrates such as tramadol and codeine would be unlikely to cause relevant increases in

norfluoxetine serum levels. Conversely, CYP2D6 genotyping also indicated that this competition

may be disease-causing even in extensive metabolizers such as the index patient. However,

duction by y the muuuutatatatatit

studied the effects of elevated norfluoxetine concentrations that corresponded to

a

uoxetine and norfluoxetine are exceeded. This report deals with a situation which

ively commonly, i.e., drug competition for metabolizing enzymes (CYP2D6). In

ent recent concomitant se of tramadol and codeine dr gs that compete ith

stuuuudidididiededede tttthehehehe efffffffeece ts of elevated norfluoxeexetitititinne concentraaationnnnss s s that corresponded to

lssss ffffound in thee iindddexx paaatititiient. MMMManaany ssitttuattioions ararrreee conceeeivablble ininini whiiiichhh tthehe thehehera

uoxetitittineenene andddd norflfllfluoxetinenenee are excxcxcceeeeeedededed d.ddd TTTThihhh ss rerererepopopoporrt ddddeaeaeae lslll witiii h a aaa sisisisitttuattttioii n whhhhiiiich

ivelylylyl commonlnlnlnly,y,y iii.e., dddrug gg compppetiititiioii n fffor metat bbob llil ziiiingngngg enzzzymymymy es ((((CYCYCYC P2P2P2P2D6D6D6D ).).).) In

tt tt imitta tnt ff ttr dad lol dnd dod iei ddr thth tat tte itithh



ownloaded from


DOI: 10.1161/CIRCEP.113.000239

11

although concomitant use of fluoxetine, tramadol, and codeine is probably common, drug-

induced LQTS by this drug combination is not commonly reported. We demonstrate that the

occurrence of LQTS/TdP required the presence of increased susceptibility to the LQTS-inducing

potential of these drugs.21,22 In the index patient, such susceptibility was provided by the K422T

mutation in KCNQ1.

In the absence of QT prolonging drugs, the biophysical derangements caused by the

K422T mutation were small and did not cause appreciable QTc prolongation. Although slight

QTc prolongation was observed during exercise in 3 mutation carriers, no hard conclusions can

be drawn on the clinical effects of the mutation per se. QTc interval is known to increase with

age,23 which is a likely explanation for the observed effects here, especially because no QTc

prolongation was observed in the investigated carrier in generation III. Also, our

electrophysiological studies with forskolin did -adrenergic

stimulation in mutant channels. The hastened deactivation time course demonstrated in K422T

channels, could, however, lead to small action potential prolongation at fast heart rates, as a

consequence of less accumulation of IKs,24

While the QTc prolonging effects of the K422T mutation per se were limited, our

modeling studies indicated that they reduced the repolarization reserve25-27 sufficiently to result

in more severe QTc prolongation in the presence of supratherapeutic norfluoxetine levels. These

modeling studies also revealed that the IKs blocking effects of norfluoxetine contribute

importantly to the pathologic QT prolongation observed here and act synergistically with the

well-known IKr blocking effects of this drug. Taken together, these findings highlight the role of

IKs blockade in the occurrence of drug-induced LQTS. Confirmation in future studies that

disease-causing QT prolongation by drugs involves their IKs blocking properties may have

, no hard conclusiiionononons

knoownwn ttto o iinini crcreeasesesese ww

ch is a likely explanation for the observed effects here, especially because no QT

o

s e

n 2

co ld ho e er lead to small action potential prolongation at fast heart rates as a

ch isisis aaaa llllikikikkelelelelyy exexexxplpp anation for the observedededed eeffects here, eeespececececially because no QT

onnnn wwwwas observedd innn tttthhe ininini vesttttigigigi aaateed ccarrrrieer in gggenenenenereere atiioionnn IIII. Alllsl oo, ouuruu

siologogggicicicalalalal studididies wiiiiththth forrskskskskolin ddddidididid - daddrene

n in mutant chchchaaana nelsll . ThThThe hhahh steneddd ddddeactiiivatiiion tiitiime couooo rse e dedededemonstrattedededd iiiin nnn K42

lldd hh lle dad tt lalll titi tot titi lal llo tatiio tat ff tt hhe tt tte



ownloaded from


DOI: 10.1161/CIRCEP.113.000239

12

important clinical implications. For instance, in vitro testing in development programs for novel

drugs may have to include screening for IKs blockade liability.28,29 IKs blockade by such drugs

may cause disease-causing QT prolongation in the added presence of common factors that

reduce repolarization reserve, e.g., gene variants, electrolyte imbalance, or concomitant cardiac

disease.

Remarkably, while supratherapeutic levels of fluoxetine and norfluoxetine are clinically

associated with QTc prolongation,9-11 previous electrophysiological studies showed action

potential shortening in canine, guinea pig, and rabbit isolated myocytes, whereas studies in rat

myocytes showed lengthening of the action potential by fluoxetine.15,30-32 These contradictory

effects may be explained by species-differences in electrophysiological properties. For example,

action potential prolongation by inhibition of IKr is clearly species dependent, as shown in

simulations and experimental studies.33 The observed differences might also be attributed to the

long-term effects of norfluoxetine on KCNH2 channel expression, which are obviously not

included in the studies into the acute effects on action potential duration. Whether long-term

effects on expression of other channels occur is not known and requires further investigation.

This study has some limitations. First, the concentration of drugs at the cellular level is

hard to predict: while the fact that norfluoxetine is highly protein-bound would obviously lead to

lower active levels,34 conversely, tissue accumulation in the heart occurs, resulting in

heart/serum concentration ratios of 7±2.35 Second, the computer modeling studies should be

interpreted as a qualitative, rather than a quantitative illustration. As we were interested in the

contribution of IKs blockade to action potential prolongation, the effects of norfluoxetine on

sodium current, L-type calcium current and transient outward current, previously shown in

isolated canine ventricular myocytes,15 were not taken into account. Furthermore, given the

es, whereas studiees s s s ininii

0-32 ThThThThesesee cocontttntrar didididiccctcto

y m

e

s

e

n the st dies into the ac te effects on action potential d ration Whether long term

y bebebe eeeexpxxplalalalainininineddd bbbby species-differences in eeelelell ctrophysiolooogicacacaalll properties. For exam

ennnntiala prolongaatiion bybby inhnhnhhibitttioioioi n oof IKKrII iis ccleearlylylyly sssspeppep cieeses deppeendededed nnt, assss ssshowwn innn

s anddd d exexexpppperime tttntallll stttudiessss.333333 The oooobsbbsbserveddd diddd ffffffffeeeerenenenencces s mimimimighghhttt alsoooo bbbbe atttrttt ibibibb tutt ddded t

effects foff norrflflflfluoxetiiine on KCKCKCNHNHNHN 222 chhah nn lell expppression,n whihihichchchch are obvbb ioiii ususususlylylyly not

thth tst ddiie iintto tthhe tte ffff tts titi tot titi lal dd tatiio WWhhethth ll tte



ownloaded from


DOI: 10.1161/CIRCEP.113.000239

13

uncertainties in the electrophysiology of human ventricular myocytes and, consequently, in

mathematical models of such myocytes, we refrained from carrying out tissue simulations in

order to extrapolate APD data to ECG. Nevertheless, our modeling studies clearly showed an

important contribution of IKs blockade to action potential prolongation and revealed that K422T

mutation carriers are more prone to this effect. Third, while we assumed that norfluoxetine serum

levels were increased by concomitant use of tramadol and codeine, we could not prove this

assumption, because serum levels of norfluoxetine levels before tramadol/codeine use were not

available. Finally, although 90% of the LQTS cases is caused by reduction in either IKr or IKs,4

other gene mutations associated with prolonged QT interval cannot be excluded.

In conclusion, we demonstrate here, on the basis of a patient with IKs reduction due to a

KCNQ1 mutation, that the IKs blocking effects of the noncardiac drug norfluoxetine contribute

importantly to the occurrence of lethal cardiac arrhythmias. These findings highlight the

importance of IKs blockade in drug-induced LQTS.

Acknowledgments: The authors thank J.G. Zegers for kindly providing the data-acquisition software, and B. de Jonge for valuable biotechnical assistance.

Funding Sources: Dr. Tan was supported by the Netherlands Organization for Scientific Research (NWO, grant ZonMW Vici:918.86.616), the Dutch Medicines Evaluation Board (MEB/CBG), and the European Community's Seventh Framework Programme (FP7/2007-2013) under grant agreement nr.:241679-the ARITMO-project.

Conflict of Interest Disclosures: None.

References:

1. Darpo B, Nebout T, Sager PT. Clinical evaluation of QT/QTc prolongation and proarrhythmic potential for nonantiarrhythmic drugs: the International Conference on Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use E14 guideline. JClin Pharmacol. 2006;46:498-507.

ction in either IKr II orororor IIKII

excclulull dedddedddd.

c

u b

y

e

connnnclcllclusususu ioioioon,n,n,n weeee deddd monstrate here, on the bbbbassis of a patienee t wiwiwithtt IKsII reduction due

uuuutaaaatttion, that thehe IKsssII bbblockckckking eeeffeeeccts ofoff thee nnonnncacaaardrdrdrdiiaccc dddrug nnorrrflflflf uoxeeeetitiinee cconnntrrrib

y to tttthehehehe ooooccurrence ffof lethahahahalll carddddiaiaiaaccc ararar hhrhhyttythhhmh iaiaiaiassss. TTTThhheh sesee fffiniii dididings hihihihighghghghlighhhttt ththththe

e of fff IIIKsII bbbloll ckkkadadadde iiini dddrug-gg iiinddud ceddd LQLQLQQTSTSTSS.



ownloaded from


DOI: 10.1161/CIRCEP.113.000239

14

2. Roden DM. Drug-induced prolongation of the QT interval. N Engl J Med. 2004;350:1013-1022.

3. Zareba W, Cygankiewicz I. Long QT syndrome and short QT syndrome. Prog Cardiovasc Dis.2008;51:264-278.

4. De Ponti F, Poluzzi E, Cavalli A, Recanatini M, Montanaro N. Safety of non-antiarrhythmic drugs that prolong the QT interval or induce torsade de pointes: an overview. Drug Saf.2002;25:263-286.

5. Mitcheson JS, Chen J, Lin M, Culberson C, Sanguinetti MC. A structural basis for drug-induced long QT syndrome. Proc Natl Acad Sci U S A. 2000;97:12329-12333.

6. Lindgren S, Bass AS, Briscoe R, Bruse K, Friedrichs GS, Kallman MJ, Markgraf C, Patmore L, Pugsley MK. Benchmarking Safety Pharmacology regulatory packages and best practice. JPharmacol Toxicol Methods. 2008;58:99-109.

7. Letsas KP, Alexanian IP, Pappas LK, Kounas SP, Efremidis M, Sideris A, Kardaras F. QT interval prolongation and torsade de pointes associated with indapamide. Int J Cardiol.2006;112:373-374.

8. Camm AJ, Savelieva I. New antiarrhythmic drugs for atrial fibrillation: focus on dronedarone and vernakalant. J Interv Card Electrophysiol. 2008;23:7-14.

9. Rajamani S, Eckhardt LL, Valdivia CR, Klemens CA, Gillman BM, Anderson CL, Holzem KM, Delisle BP, Anson BD, Makielski JC, January CT. Drug-induced long QT syndrome: hERG K+ channel block and disruption of protein trafficking by fluoxetine and norfluoxetine. Br J Pharmacol. 2006;149:481-489.

10. Varriale P. Fluoxetine (Prozac) as a cause of QT prolongation. Arch Intern Med.2001;161:612.

11. Nykamp DL, Blackmon CL, Schmidt PE, Roberson AG. QTc prolongation associated with combination therapy of levofloxacin, imipramine, and fluoxetine. Ann Pharmacother.2005;39:543-546.

12. Itoh H, Shimizu W, Hayashi K, Yamagata K, Sakaguchi T, Ohno S, Makiyama T, Akao M, Ai T, Noda T, Miyazaki A, Miyamoto Y, Yamagishi M, Kamakura S, Horie M. Long QT syndrome with compound mutations is associated with a more severe phenotype: a Japanese multicenter study. Heart Rhythm. 2010;7:1411-1418.

13. ten Tusscher KHWJ, Noble D, Noble PJ, Panfilov AV. A model for human ventricular tissue. Am J Physiol Heart Circ Physiol. 2004;286:H1573-H1589.

14. ten Tusscher KHWJ, Panfilov AV. Alternans and spiral breakup in a human ventricular tissue model. Am J Physiol Heart Circ Physiol. 2006;291:H1088-H1100.

ageg s and best ppractctctcticicicice

eris AAAA KaKaKaKardrdrdrdararararasasasas FFFF Qo3

AJ, Savelieva I. New antiarr thmic dr s for atrial fibrillation: focus on dronedak

ni S, Eckhardt LL, Valdivia CR, Klemens CA, Gillman BM, Anderson CL, Holzle BP, Anson BD, Makielski JC, January CT. Drug-induced long QT syndrome:l block and disr ption of protein trafficking b fl o etine and norfl o etine B J

olooongngngngatatatioioioon n nn aand dd ttot rsade de pointes associaiaiaiatetted with indappppammmidididi e. Int J Cardiol.377377 -374.

AJJJ,,, SSSavelievvvva II. Newew antttiaaarrrr hyhyyhythththt mimm c drdrdrugss fforrrr aata rialaala fibbbrrrir llattioon:::: ffffocoo ussss ooonn ddronnneeedakalanttt.t JJJ InInInI terv CCCCarddd ElEEE ectrtrtrrooophysiiiolololo . 202020200808080 2;2223:333 7-777 14141414.

ni S, EEE kckkhahh drddt tt LLLLLLLL, VaVV ldldldivii iaiii CR,RRR KKKlllemens CCCA,AAA GGGGilililllmll ann BBBBM,MM AAAA dndderson CLCLCLCL, HoHHH lzle BP, Annnsososoon n nn BDBDBDBD, , , MaMaMaM kikkikieleee skskski ii i JCJCJCJC,,, JaJaJaJanununuuarararary y y y CTCTCTCT. DrDrDrDrugugugug-i-i-indndnducucucucedededed llllonononong g g g QTQTQTQT sssynyyy drome:ll bbllo kck dnd ddiis tptiio fof tteiin tt ffffiickiki bb flfl titi dnd flfl titi BB JJ



ownloaded from


DOI: 10.1161/CIRCEP.113.000239

15

15. Magyar J, Szentandrássy N, Bányász T, Kecskeméti V, Nánási PP. Effects of norfluoxetine on the action potential and transmembrane ion currents in canine ventricular cardiomyocytes. Naunyn Schmiedebergs Arch Pharmacol. 2004;370:203-210.

16. Orsulak PJ, Kenney JT, Debus JR, Crowley G, Wittman PD. Determination of the antidepressant fluoxetine and its metabolite norfluoxetine in serum by reversed-phase HPLC with ultraviolet detection. Clin Chem. 1988;34:1875-1878.

17. Vree TB, Verwey-van Wissen CP. Pharmacokinetics and metabolism of codeine in humans. Biopharm Drug Dispos. 1992;13:445-460.

18. Hiemke C, Härtter S. Pharmacokinetics of selective serotonin reuptake inhibitors. Pharmacol Ther. 2000;85:11-28.

19. Fanoe S, Jensen GB, Sjøgren P, Korsgaard MP, Grunnet M. Oxycodone is associated with dose-dependent QTc prolongation in patients and low-affinity inhibiting of hERG activity in vitro. Br J Clin Pharmacol. 2009;67:172-179.

20. Deisemann H, Ahrens N, Schlobohm I, Kirchhoff C, Netzer R, Möller C. Effects of common antitussive drugs on the hERG potassium channel current. J Cardiovasc Pharmacol.2008;52:494-499.

21. Makita N, Horie M, Nakamura T, Ai T, Sasaki K, Yokoi H, Sakurai M, Sakuma I, Otani H, Sawa H, Kitabatake A. Drug-induced long-QT syndrome associated with a subclinical SCN5Amutation. Circulation. 2002;106:1269-1274.

22. Itoh H, Sakaguchi T, Ding WG, Watanabe E, Watanabe I, Nishio Y, Makiyama T, Ohno S, Akao M, Higashi Y, Zenda N, Kubota T, Mori C, Okajima K, Haruna T, Miyamoto A, Kawamura M, Ishida K, Nagaoka I, Oka Y, Nakazawa Y, Yao T, Jo H, Sugimoto Y, Ashihara T, Hayashi H, Ito M, Imoto K, Matsuura H, Horie M. Latent genetic backgrounds and molecular pathogenesis in drug-induced long-QT syndrome. Circ Arrhythm Electrophysiol. 2009;2:511-523.

23. Mangoni AA, Kinirons MT, Swift CG, Jackson SH. Impact of age on QT interval and QT dispersion in healthy subjects: a regression analysis. Age Ageing. 2003;32:326-331.

24. Stengl M, Volders PGA, Thomsen MB, Spätjens RLHMG, Sipido KR, Vos MA. Accumulation of slowly activating delayed rectifier potassium current (IKs) in canine ventricular myocytes. J Physiol. 2003;551:777-786.

25. Roden DM, Yang T. Protecting the heart against arrhythmias: potassium current physiology and repolarization reserve. Circulation. 2005;112:1376-1378.

26. Jost N, Virág L, Bitay M, Takács J, Lengyel C, Biliczki P, Nagy Z, Bogáts G, Lathrop DA, Papp JG, Varró A. Restricting excessive cardiac action potential and QT prolongation: a vital role for IKs in human ventricular muscle. Circulation. 2005;112: 1392-1399.

ododododonononone ee e isisisis aaaassssssssocococociaiaiaiateted d d d wwwwng offff hhhhERERERERGGG G acacttttivivivivitytytyty ii

mann H, Ahrens N, Schlobohm I, Kirchhoff C, Netzer R, Möller C. Effects of com

9

N, Horie M, Nakamura T, Ai T, Sasaki K, Yokoi H, Sakurai M, Sakuma I, OtanK NC

maaana nnnn H, Ahrhrhrh enenenssss N,,N, SSSchchchhlolll boboboohmhmhmhm III,,, KiKiKircrcrchhhh offf C,,, NNNNetetetetzez r rrr R,R,R,R MMMölölölö ler C.C.C.C EEEEffff ececectstststs ooofff f coc mdddrd uuugu s on the hhEERG G G potatatassium m chcchanneneel cuurrrennntt. J J J J CCardrdrdiovaascc PPPPhhah rmmmmacccool.

94-4-4-4 4949494 9.

N, HoHoHoH riririr e ee M,M,M, NNNNakakakakamamamamururururaa a T,T,T,T, AAAi i T,T,T,T SSSSasasasasaakiii KKK, YYYYokokkokoioioii HHHH,, SaSaSaSakukukurararar i i i M,MM,M SSSakakaka umumumma aa a I,I,I, OOOOtanKitababbb takkke AAA. DrDDD uggg ii-inddduc deddd lllong-gg QTQTQTT synyy dddrome associiiiatatededede wwwitititthh h a subbcbb liiliininininicacacacal ll SSSCS NCirculatioooonnnn.. 202202002020202;111;10600 :1:1:1:126226269-9-9-1212121274747474... by guest on M


Dow

nloaded from


DOI: 10.1161/CIRCEP.113.000239

16

27. Varró A, Baláti B, Iost N, Takács J, Virag L, Lathrop DA, Csaba L, Tálosi L, Papp JG. The role of the delayed rectifier component IKs in dog ventricular muscle and Purkinje fibre repolarization. J Physiol. 2000;523 Pt 1:67-81.

28. Towart R, Linders JTM, Hermans AN, Rohrbacher J, van der Linde HJ, Ercken M, Cik M, Roevens P, Teisman A, Gallacher DJ. Blockade of the IKs potassium channel: an overlooked cardiovascular liability in drug safety screening? J Pharmacol Toxicol Methods. 2009;60:1-10.

29. Varró A, Baczkó I. Cardiac ventricular repolarization reserve: a principle for understanding drug-related proarrhythmic risk. Br J Pharmacol. 2011;164:14-36.

30. Park KS, Kong ID, Park KC, Lee JW. Fluoxetine inhibits L-type Ca2+ and transient outward K+ currents in rat ventricular myocytes. Yonsei Med J. 1999;40:144-151.

31. Pacher P, Magyar J, Szigligeti P, Bányász T, Pankucsi C, Korom Z, Ungvári Z, Kecskeméti V, Nánási PP. Electrophysiological effects of fluoxetine in mammalian cardiac tissues. Naunyn Schmiedebergs Arch Pharmacol. 2000;361:67-73.

32. Magyar J, Rusznák Z, Harasztosi C, Körtvély A, Pacher P, Bányász T, Pankucsi C, Kovács L, Szúcs G, Nánási PP, Kecskeméti V. Differential effects of fluoxetine enantiomers in mammalian neural and cardiac tissues. Int J Mol Med. 2003;11:535-542.

33. O'Hara T, Rudy Y. Quantitative comparison of cardiac ventricular myocyte electrophysiology and response to drugs in human and nonhuman species. Am J Physiol Heart Circ Physiol. 2012;302:H1023-H1030.

34. Guo B, Li C, Wang G, Chen L. Rapid and direct measurement of free concentrations of highly protein-bound fluoxetine and its metabolite norfluoxetine in plasma. Rapid Commun Mass Spectrom. 2006;20:39-47.

35. Johnson RD, Lewis RJ, Angier MK. The distribution of fluoxetine in human fluids and tissues. J Anal Toxicol. 2007;31:409-414.

Figure Legends:

Figure 1. Twelve-lead ECG of index patient at admission. Prominent QTc prolongation

(maximal QTc interval 620 ms) (A) culminating in an episode of Torsade de Pointes (25 mm/s,

10 mm/mV) (B).

ZZZZ,, UnUnUnUngvgvgvgvárárárári i i i Z,Z,Z,Z, KKKKeecece skskskskeeeen cardrdddiaiiiacc titititissssueuess. NaNaNaNau

r J, Rusznák Z, Harasztosi C, Körtvély A, Pacher P, Bányász T, Pankucsi C, KoNánási PP, Kecskeméti V. Differential effects of fluoxetine enantiomers in mamm

s eo

r JJJJ, Rusznáááákk k Z,Z,Z, Harararasasasztztztz ossi i ii C,CC,C KKKörörörtvvvélélély yy A, PPacccheheheher r r r P,P BBBBánánáná yáyáyászszszs T, , , PaPaPaPanknknkn ucccsisisii CCC,, KoKNánánánánásá i PP, Keccskeméméméti VVV. Diiifffffffferrrenntiaal effecects ooof f f flflflfluuoxxetttine ennaantntntiomeeeerrs iinn maaammmm

caara didd ac tisssususus es. Inttt JJ MoMoMolll MeMeMedddd. 2020202 033;1;1;11:53535--5454545 2..

T, RuRuRuR dydydydy YYY. QuQuQuQuananantititititatataatitititivevevev ccccomomommpapapapaririririsosososonn n offff cccarararrdidididiacacacac vvvenenenntrtrtrriciciciculululu arararr mmmmyoyoyoyocycycyytetetet siolllogggy yy and dd reeespspsps onse to drdd ugggs iiini hhhuman a ddnd nonhuhhh man n spppecccieieieies. AmAA JJJ PPPPhyhyhyhysisisis oloolo Heol. 2012;33302020202:H:H:H:H10100102323223-HH-H-H1001010303030... by guest on M


Dow

nloaded from


DOI: 10.1161/CIRCEP.113.000239

17

Figure 2. Pedigree showing mutation status and QTc intervals at baseline and during exercise.

NA=not available; SD=sudden death.

Figure 3. Electrophysiological properties of wild-type (WT) and heterozygously expressed

(HET) KCNQ1/KCNE1 channels. A. Typical examples of currents through WT-KCNQ1/KCNE1

(left) and HET-KCNQ1/KCNE1 channels (right) in response to a voltage-clamp step to test

potentials ranging

clamp protocol used. B. Current-voltage relationship of steady-state WT-KCNQ1/KCNE1 and

HET-KCNQ1/KCNE1 channels C. Voltage-dependence of steady-state activation of WT-

KCNQ1/KCNE1 and HET-KCNQ1/KCNE1 channels. Solid lines: Boltzmann fits of average data.

Half-maximum activation voltage (V1/2) was more positive in HET-KCNQ1/KCNE1 channels

(P<0.05; unpaired t-test). D. Fully-activated currents as determined by the protocol shown in the

inset. E. -

KCNQ1/KCNE1 and HET-KCNQ1/KCNE1 currents. *P<0.05 (RM-ANOVA).

Figure 4. Effects of forskolin, fluoxetine, norfluoxetine, tramadol, and codeine on WT-

KCNQ1/KCNE1 and HET-KCNQ1/KCNE1 channels. A-E. Typical examples of WT-

KCNQ1/KCNE1 and HET-KCNQ1/KCNE1 currents, activated by depolarizing voltage-clamp

step to +110 mV, in absence and presence of forskolin (10 μM; A), fluoxetine (10 μM; B),

norfluoxetine (5 μM; C), tramadol (30 μM; D), and codeine (30 μM; E) F. Average effects of

forskolin (WT, n=7; HET, n=5), fluoxetine (WT, n=7; HET, n=6), norfluoxetine (WT, n=6; HET,

n=6), tramadol (WT, n=6; HET, n=4) and codeine (WT, n=5; HET, n=5) on current density, V1/2

and slope factor (k) of the steady-state activation curve, activation time constant (measured

WT-KCNQQ1/KCNENEEE1111 a

e actitiitivavatitiitionon ooffff WTWTWTWT--

C g

m n

n

CNE1 and HET KCNQ1/KCNE1 c rrents *P<0 05 (RM ANOVA)

CNENENENE1111 ananand dd d HHHETETETT-KCNQ1/KCNE1 channeeelslsls.. Solid lines: Bolllltztztzt mann fits of averag

mmmum mmm m activation volltaage (((VVVV1/2)))) waass moorrre ppositivivivive e e e iniini HETETET-KCKCNQQNQQ11/1 KCCCCNENEN 1 chhhananann

npaiiirerereredddd ttt-t testttt).)) DDD.D FFFullll y-actctctc iviivivatedddd ccccurururrenttts as ddddeteteteterrrmmmim neneneeddd d bybybb ttthehhh pprorororotttocolll hhshhown

---

CCNENE11 dd HEHETT KCKCNQNQ1/1/KCKCNENE11 tts **PP<0<0 0055 (R(RMM ANANOVOVA)A)



ownloaded from


DOI: 10.1161/CIRCEP.113.000239

18

during the depolarizing voltage-clamp step to +110 mV) and deactivation time constant

(measured at -60 mV after the depolarizing voltage-clamp step to +110 mV). *P<0.05 (paired t-

test).

Figure 5. Concentration-dependence of blockade of WT-KCNQ1/KCNE1 and HET-

KCNQ1/KCNE1 current by norfluoxetine. Currents were recorded during a depolarizing voltage-

clamp step to +110 mV at norfluoxetine concentrations between 0.01 and 100 μM and

normalized to the current before application of the drug. Fitting a Hill equation to the normalized

data revealed IC50 values of 5.3±0.8 and 4.9±0.8 μM and Hill coefficients of 0.78±0.10 and

0.89±0.14 for WT-KCNQ1/KCNE1 and HET-KCNQ1/KCNE1 current, respectively. Numbers

near symbols indicate the number of cells measured at a given concentration.

Figure 6. Simulated effects of norfluoxetine on the action potential and major repolarizing

currents in heterozygous K422T mutation carriers and non-carriers. Membrane potential (A, D),

IKr (B, E) and IKs (C, F) during 1Hz stimulation in non-carriers (A–C) and mutation carriers (D–

F). Black solid lines indicate control. The effects of norfluoxetine (5 μM) through IKr inhibition

per se, IKs inhibition per se, and inhibition of both IKr and IKs are shown by blue dashed lines, red

dotted lines, and green solid lines, respectively. Arrows indicate the increase in action potential

duration.

equq ation to the nnorororormmmm

ents ofofff 0000 777.78±8±8±8±0000.101000 aaaannndnd

f b

o

S

hetero go s K422T m tation carriers and non carriers Membrane potential (A

for r r r WTWTWTW -KCKCKCK NQNQQQ1/11 KCNE1 and HET-KCNQNQNQ1/1 KCNE1 curuu rentntntnt,,, respectively. Numb

olllsl indicate thee numbmbmber oooff celllllllls memm assurrred aat a gigigivevevevenn cooonnncenntrratttioioioi n.

Simulal ted dd effffefefeccctc s ffof norflflfluoxe ititine on thhhe actitiiion ppp totentitialalala andndndd majajajor repppololololararararizizizzing gg

hhette KK42422T2T tt tatiio iie dnd iie MMe bmb tte tntiiall ((AA



ownloaded from




ownloaded from


Langendijk, Antoni C.G. van Ginneken, Ronald Wilders and Hanno L. TanChristiaan C. Veerman, Arie O. Verkerk, Marieke T. Blom, Christine A. Klemens, Pim N.J.

Blockade Contributes Importantly to Drug-Induced Long QT SyndromeKsI

Print ISSN: 1941-3149. Online ISSN: 1941-3084 Copyright © 2013 American Heart Association, Inc. All rights reserved.

Dallas, TX 75231is published by the American Heart Association, 7272 Greenville Avenue,Circulation: Arrhythmia and Electrophysiology

published online August 31, 2013;Circ Arrhythm Electrophysiol.

http://circep.ahajournals.org/content/early/2013/08/31/CIRCEP.113.000239World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://circep.ahajournals.org/content/suppl/2013/08/31/CIRCEP.113.000239.DC1Data Supplement (unedited) at:

http://circep.ahajournals.org//subscriptions/

is online at: Circulation: Arrhythmia and Electrophysiology Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. Permissions and Rights Question and Answerinformation about this process is available in the

requested is located, click Request Permissions in the middle column of the Web page under Services. FurtherCenter, not the Editorial Office. Once the online version of the published article for which permission is being

can be obtained via RightsLink, a service of the Copyright ClearanceCirculation: Arrhythmia and Electrophysiology Requests for permissions to reproduce figures, tables, or portions of articles originally published inPermissions:



ownloaded from

http://circep.ahajournals.org/content/early/2013/08/31/CIRCEP.113.000239

http://circep.ahajournals.org/content/suppl/2013/08/31/CIRCEP.113.000239.DC1

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://circep.ahajournals.org//subscriptions/


1

SUPPLEMENTAL MATERIALS

Supplemental methods

ECG analysis

Twelve-lead ECGs were obtained from the index patient during and after admission, and from

7 relatives. Written informed consent was obtained from all participants. Exercise testing was

conducted in 6 individuals, including the index patient. QT durations were corrected for heart

rate according to Bazett’s formula.

Genetic analysis

KCNQ1 and KCNH2 were screened for mutations as described previously.1, 2

Genotyping

analysis of CYP2D6, with particular interest in determining the presence of CYP2D6 *3, *4,

*5, *6, *7 or *8 alleles, or other CYP2D6 deletions or duplications, was performed by means

of TaqMan® SNP Genotyping Assays, according to manufacturer’s protocol (Applied

Biosystems, Foster City, CA).

cDNA constructs and mutagenesis

Constructs of wild-type KCNQ1 and β-subunit KCNE1 cloned in pSP64 were kindly provided

by Dr. Michael C. Sanguinetti, University of Utah, Salt Lake City, UT. To transfect and

express these constructs, wild-type KCNQ1 and wild-type KCNE1 were excised from this

vector and subcloned into the mammalian expression vector pCGI. The KCNQ1-K422T

mutation was created by using the QuikChange site-directed mutagenesis kit (Stratagene,

Santa Clara, CA), according to manufacturer’s protocol. Mutant KCNQ1 was constructed by

means of PCR, amplified with primers containing the point mutation K422T. The following

oligonucleotides complementary to wild-type gene were used (boldface letters indicate

mutation): 5’GTGGTGGTAAAGAAAAAAACGTTCAAGCTGGAC 3’ (forward) and

5’GTCCAGCTTGAACGTTTTTTTCTTTACCACCAC 3’ (reverse). The mutation was

verified and confirmed by DNA sequencing. To exclude point mutations within the plasmid,

KCNQ1-K422T was excised and cloned back into wild-type pCGI-KCNQ1 using HindIII and

EcoRI restriction sites. The complete KCNQ1 gene was sequenced to confirm correctness.

Cell preparation and heterologous expression

HEK293 cells were cultured in Minimal Essential Medium (MEM) supplemented with 10%

2

fetal bovine serum, penicillin, streptomycin, and non-essential amino acids. To express the

KCNQ1/KCNE1 potassium current of wild-type channels, cells were transiently transfected

with lipofectamine using 1 µg of wild-type KCNQ1 cDNA and 1 µg KCNE1 cDNA (WT-

KCNQ1/KCNE1). To recapitulate a heterozygous state, 0.5 µg of both wild-type and mutant

KCNQ1, in addition to 1 µg KCNE1, were co-transfected (HET-KCNQ1/KCNE1).

Coexpression of green fluorescent protein (GFP) was realized by an internal ribosomal entry

site cassette in the vector. Transfected cells were identified by an epifluorescent microscope.

After transfection, the cells were incubated in 5% CO2 at 37ºC for 24–48 hours.

Electrophysiological experiments

KCNQ1/KCNE1 currents were recorded at 36±0.2°C by the amphotericin-perforated patch-

clamp technique using an Axopatch 200B amplifier (Molecular Devices, Sunnyvale, CA).

Cells were superfused with solution containing (in mM): NaCl 140, KCl 5.4, CaCl2 1.8, MgCl2

1, glucose 5.5, HEPES 5, pH 7.4 (NaOH). Pipettes of borosilicate glass were filled with

solution containing (in mM): K-gluc 125, KCl 20, NaCl 10, amphotericin-B 0.88, HEPES 10,

pH 7.2 (KOH) and had resistances of 1.5–2 MΩ. Series resistance was compensated by ≥80%,

and potentials were corrected for the estimated liquid junction potential. Signals were low-pass

filtered (cut-off frequency: 2 kHz) and digitized at 2 kHz. Voltage control, data acquisition,

and data analysis were accomplished using custom software. Cell membrane capacitance was

estimated by dividing the time constant of the decay of the capacitive transient in response to 5

mV hyperpolarizing voltage clamp steps from –60 mV by the series resistance, and amounted

to 9.8±0.4 pF (mean±SEM, n=124).

The activation and deactivation kinetics of the KCNQ1/KCNE1 currents were

determined by the voltage clamp protocols diagrammed in Figure 3, A and D, respectively. For

both protocols, holding potential was –80 mV and the cycle interval 6 seconds. Voltage-

dependence of activation was determined from tail currents, which were fitted with a

Boltzmann function I/Imax=A/1.0+exp[(V1/2–V)/k] to determine the half-maximum activation

voltage (V1/2) and slope factor (k). The time course of current activation and deactivation were

fitted by the monoexponential equations I/Imax=A×[1–exp(–t/τ)] and I/Imax=A×exp(–t/τ),

respectively. Current densities were calculated by dividing current amplitudes by the cell

membrane capacitance.

3

Drugs

The effects of forskolin, fluoxetine, norfluoxetine, tramadol, and codeine on KCNQ1/KCNE1

currents were tested 5 minutes after the onset of bath application. Forskolin was prepared as

10 mM stock solution in ethanol. Fluoxetine, norfluoxetine, tramadol, and codeine were

prepared as 5 mM stock solutions dissolved in distilled water. All stock solutions were diluted

appropriately before use. The effects of the drugs were evaluated in paired measurements at a

membrane potential of +110 mV. At this membrane potential the drug effects on density

could be assessed without interference by the effects on voltage-dependency of activation. All

drugs were obtained from Sigma-Aldrich, St.Louis, MO. Dose-response curves were fitted to

the Hill equation Idrug/Icontrol=1/[1+(dose/IC50)n], where Idrug/Icontrol is the normalized IKs

current at a membrane potential of +110 mV, dose is the bath concentration of the drug, IC50

is the dose required for 50% current block, and n is the Hill coefficient. Serum concentrations

of fluoxetine and norfluoxetine were assayed according to a previously described method.3

Computer simulations

Functional differences between WT-KCNQ1/KCNE1 and HET-KCNQ1/KCNE1 were tested by

computer simulations using the midmyocardial version of the ten Tusscher et al. human

ventricular cell model,4 as updated by ten Tusscher and Panfilov.

5 This midmyocardial version

of the model has a 4-fold lower IKs density, but is otherwise identical to the generic epicardial

version of the model. We selected the midmyocardial version in order to limit the effects of a

possible overscaling of IKs in the ten Tusscher et al. model,4 as discussed by Grandi et al.

6 and

O’Hara et al.7

The experimentally observed mutant- and drug-induced changes in electrophysiological

properties of KCNQ1/KCNE1 current were implemented as changes in IKs. The effect of the

K422T mutation was implemented as a +13.3 mV shift in the IKs steady-state activation curve,

through a +13.3 mV shift in the xs,∞-Vm relation of the model, and a 20% increase in

deactivation rate βxs. The inhibitory effect of 5 µM norfluoxetine was implemented as a 55%

decrease in current density GKs, a 25% increase in activation rate αxs, and a 40% increase in

deactivation rate βxs.

Previously reported norfluoxetine-induced IKr block was also implemented. When fitted

to the Hill equation, norfluoxetine showed an IC50 value of 2.5 µM, with a Hill coefficient of

1.3.8 Accordingly, the effect of 5 µM norfluoxetine on IKr was implemented as a 72% decrease

4

in current density GKr. Effects of norfluoxetine on other transmembrane ion currents9 were not

taken into account.

Software was compiled as a 32-bit Windows application using Compaq Visual Fortran

6.6C and run on an Intel Xeon processor based workstation. For numerical integration of

differential equations we applied a simple and efficient Euler-type integration scheme with a 5-

µs time step. All figures show steady-state action potential characteristics, obtained at 2 min

after onset of 1-Hz stimulation with a 1-ms, 9.2-nA stimulus.

Statistics

Data are presented as mean±SEM. Statistical comparison between WT-KCNQ1/KCNE1 and

HET-KCNQ1/KCNE1 was performed by using unpaired t-test, except for the comparison of

time constants of activation and deactivation between these groups, which were evaluated by

means of Two-Way Repeated Measures ANOVA (RM-ANOVA) followed by pairwise

comparison using the Student-Newman-Keuls test. To assess the effects of drugs on WT-

KCNQ1/KCNE1 and HET-KCNQ1/KCNE1 channels, paired t-tests were used. P<0.05 defines

statistical significance.

Supplemental references

1. Bellocq C, van Ginneken ACG, Bezzina CR, Alders M, Escande D, Mannens MMAM,

Baro I, Wilde AAM. Mutation in the KCNQ1 gene leading to the short QT-interval

syndrome. Circulation. 2004;109:2394-2397.

2. Verkerk AO, Wilders R, Schulze-Bahr E, Beekman L, Bhuiyan ZA, Bertrand J, Eckardt

L, Lin D, Borggrefe M, Breithardt G, Mannens MM, Tan HL, Wilde AA, Bezzina CR.

Role of sequence variations in the human ether-a-go-go-related gene (HERG, KCNH2)

in the Brugada syndrome. Cardiovasc Res. 2005;68:441-453.

3. Holladay JW, Dewey MJ, Yoo SD. Quantification of fluoxetine and norfluoxetine serum

levels by reversed-phase high-performance liquid chromatography with ultraviolet

detection. J Chromatogr B Biomed Sci Appl. 1997;704:259-263.

4. ten Tusscher KHWJ, Noble D, Noble PJ, Panfilov AV. A model for human ventricular

tissue. Am J Physiol Heart Circ Physiol. 2004;286:H1573-H1589.

5

5. ten Tusscher KHWJ, Panfilov AV. Alternans and spiral breakup in a human ventricular

tissue model. Am J Physiol Heart Circ Physiol. 2006;291:H1088-H1100.

6. Grandi E, Pasqualini FS, Bers DM. A novel computational model of the human

ventricular action potential and Ca transient. J Mol Cell Cardiol. 2010;48:112-121.

7. O'Hara T, Virág L, Varró A, Rudy Y. Simulation of the undiseased human cardiac

ventricular action potential: model formulation and experimental validation. PLoS

Comput Biol. 2011;7:e1002061.

8. Rajamani S, Eckhardt LL, Valdivia CR, Klemens CA, Gillman BM, Anderson CL,

Holzem KM, Delisle BP, Anson BD, Makielski JC, January CT. Drug-induced long QT

syndrome: hERG K+ channel block and disruption of protein trafficking by fluoxetine

and norfluoxetine. Br J Pharmacol. 2006;149:481-489.

9. Magyar J, Szentandrássy N, Bányász T, Kecskeméti V, Nánási PP. Effects of

norfluoxetine on the action potential and transmembrane ion currents in canine

ventricular cardiomyocytes. Naunyn Schmiedebergs Arch Pharmacol. 2004;370:203-

210.

I Blockade Contributes Importantly to Drug-Induced...

Documents

Transcript of I Blockade Contributes Importantly to Drug-Induced...