-
Cellular Biology
Long QT Syndrome–Associated Mutations in KCNQ1 andKCNE1 Subunits
Disrupt Normal Endosomal Recycling of
IKs ChannelsGuiscard Seebohm, Nathalie Strutz-Seebohm, Oana N.
Ureche, Ulrike Henrion, Ravshan Baltaev,Andreas F. Mack, Ganna
Korniychuk, Katja Steinke, Daniel Tapken, Arne Pfeufer, Stefan
Kääb,
Cecilia Bucci, Bernard Attali, Jean Merot, Jeremy M. Tavare, Uta
C. Hoppe,Michael C. Sanguinetti, Florian Lang
Abstract—Physical and emotional stress is accompanied by release
of stress hormones such as the glucocorticoidcortisol. This hormone
upregulates the serum- and glucocorticoid-inducible kinase (SGK)1,
which in turnstimulates IKs, a slow delayed rectifier potassium
current that mediates cardiac action potential
repolarization.Mutations in IKs channel � (KCNQ1, KvLQT1, Kv7.1) or
� (KCNE1, IsK, minK) subunits cause long QTsyndrome (LQTS), an
inherited cardiac arrhythmia associated with increased risk of
sudden death. Together withthe GTPases RAB5 and RAB11, SGK1
facilitates membrane recycling of KCNQ1 channels. Here, we show
alteredSGK1-dependent regulation of LQTS-associated mutant IKs
channels. Whereas some mutant KCNQ1 channels hadreduced basal
activity but were still activated by SGK1, currents mediated by
KCNQ1(Y111C) or KCNQ1(L114P)were paradoxically reduced by SGK1.
Heteromeric channels coassembled of wild-type KCNQ1 and
theLQTS-associated KCNE1(D76N) mutant were similarly downregulated
by SGK1 because of a disruptedRAB11-dependent recycling.
Mutagenesis experiments indicate that stimulation of IKs channels
by SGK1 dependson residues H73, N75, D76, and P77 in KCNE1.
Identification of the IKs recycling pathway and its modulation
bystress-stimulated SGK1 provides novel mechanistic insight into
potentially fatal cardiac arrhythmias triggered byphysical or
psychological stress. (Circ Res. 2008;103:1451-1457.)
Key Words: kinase � trafficking � PIKfyve � LQT � stress
Physical and emotional stress may trigger cardiac arrhyth-mia
and sudden death in susceptible individuals.1–4 Thestress reaction
involves the release of stress hormones such asthe glucocorticoid
cortisol via the hypothalamic–pituitary–adrenal axis.5 Cortisol
regulates the expression of severalgenes, including the serum- and
glucocorticoid-induciblekinase (SGK)16,7 that is abundant in
cardiac tissue.8 Accord-ing to in vitro experiments SGK1 stimulates
a slow delayedrectifier K� current (IKs)9 that mediates cardiac
repolariza-tion. IKs is conducted by channels composed of KCNQ1
�subunits and KCNE1 � subunits.10,11 SGK1 phosphorylatesand thereby
activates phosphoinositide 3-phosphate 5-kinase(PIKfyve), which
generates PI(3,5)P2, which in turn enhancesRAB11-dependent
insertion of KCNQ1/KCNE1 (Q1/E1)
channels into the plasma membrane.12 Accordingly,
gain-of-function mutations of the genes encoding either SGK1or
Q1/E1 are associated with shortening of the QT interval,an
electrocardiographic measure of ventricular repolariza-tion
time,13–15 whereas loss-of-function mutations lead toprolongation
of the QT interval, causing long QT syn-drome (LQTS). Here, we
study the ability of SGK1 torecover loss-of-function LQTS mutant
channels and deter-mine the molecular requirements of SGK1
sensitivity.Stress-dependent stimulation of SGK1-mediated
channelregulation might be of particular clinical importance
forpatients with KCNQ1 or KCNE1 mutations who arepredisposed to
potentially fatal cardiac arrhythmias trig-gered by physical and/or
psychological stress.1,2
Original received August 14, 2007; resubmission received April
8, 2008; revised resubmission received October 9, 2008; accepted
November 3, 2008.From the Department of Physiology I (G.S.,
N.S.-S., O.N.U., U.H., R.B., A.F.M., G.K., F.L.), University of
Tuebingen, Germany; Department of
Biochemistry I (G.S., N.S.-S., U.H., K.S., D.T.), Receptor
Biochemistry, Ruhr University Bochum, Germany; Institute of Human
Genetics (A.P., S.K.),Technical University Munich, Germany;
Institute of Human Genetics (A.P., S.K.), National Research Center
of Environment and Health, Neuherberg,Germany; Dipartimento di
Scienze e Tecnologie Biologiche ed Ambientali (C.B.), Università
di Lecce, Italy; Department of Physiology andPharmacology (B.A.),
Sackler Medical School, Tel Aviv University, Israel; INSERM U533
(J.M.), Institut du Thorax, Faculté de Médecine, Nantes,France;
Department of Biochemistry (J.M.T.), School of Medical Sciences,
University of Bristol, England; Department of Internal Medicine III
(U.C.H.),Center for Molecular Medicine, University of Cologne,
Germany; and Department of Physiology and Nora Eccles Harrison
Cardiovascular Research &Training Institute (M.C.S.),
University of Utah, Salt Lake City.
This manuscript was sent to Harry Fozzard, Consulting Editor,
for review by expert referees, editorial decision, and final
disposition.Correspondence to Prof Dr Guiscard Seebohm,
Biochemistry I, Cation Channel Group, Room NC6/132, Ruhr University
Bochum, Universitätsstr. 150,
D-44780 Bochum, Germany. E-mail [email protected]© 2008
American Heart Association, Inc.
Circulation Research is available at
http://circres.ahajournals.org DOI:
10.1161/CIRCRESAHA.108.177360
1451 at TEL AVIV UNIV on March 4, 2010
circres.ahajournals.orgDownloaded from
http://circres.ahajournals.org
-
Materials and MethodsWestern Blot, Immunocytochemistry,
andMolecular BiologyWestern blot of plasma membrane proteins and
molecular biologywas performed as reported earlier.12 Cloning of
RAB5, RAB7,RAB11, and FLAG-tagged KCNQ1 have been described
previous-ly.16–19 Further details are available in the online data
supplement athttp://circres.ahajournals.org.
ElectrophysiologyXenopus laevis oocytes were obtained according
to German law asdescribed previously.12 Ovary lobes were digested
with collagenase(type II; Worthington), and stage 5 oocytes were
collected andinjected with 20 to 60 nL of cRNA. Oocytes were
injected with 1 ngor 5 ng of KCNQ1 cRNA alone or with 1 ng KCNQ1
cRNA plus 1ng of KCNE1 cRNA or 5 ng SGK1, RAB5/7/11 cRNA.
Oocyteswere stored for 3 to 4 days at 17°C in ND96 solution (in
mmol/L: 96NaCl, 4 KCl, 1.8 MgC12, 1.0 CaC12, 5 HEPES; and 50
mg/Lgentamicin; pH 7.6). For voltage-clamp experiments the
oocyteswere bathed in ND96 solution. A TurboTEC-10 amplifier
(npielectronic, Tamm, Germany) was used to record currents at 24°C
inoocytes 3 to 4 days after injection with cRNA using
standard2-electrode voltage-clamp techniques. Data acquisition was
per-formed using a Pentium IV computer, a Digidata 1322 A/D
interface,and pClamp 8 software (Axon Instruments).
ResultsStructural Requirements for the Modulation of IKsChannels
by SGK1Previous experiments from our laboratory indicated thatSGK1
enhances IKs by increasing the insertion of Q1/E1channels into the
plasma membrane.12 The effect did notrequire the presence of KCNE1
� subunits (Figure 1a). Bycontrast, deletion of the N-terminal
residues 1 to 81 inKCNQ1 (resulting in KCNQ1�N-term) completely
abolishedthe stimulation by SGK1 (Figure 1b). To determine whethera
specific region of KCNE1 modulates SGK1-sensitive recy-cling, we
deleted its intracellular C terminus. Deletion of theC-terminal
residues 73 to 129 within KCNE1 (resulting inKCNE1�C-term)
abolished the stimulation of IKs by SGK1(Figure 1b). Thus, the
enhanced plasma membrane insertionof Q1/E1 channels depends on the
presence of both theKCNQ1 N terminus and the KCNE1 C terminus.
Furthertruncations identified the 7-aa stretch from residues 73 to
79of KCNE1 as important for SGK1-dependent regulation ofQ1/E1
channels (Figure 1c). Cysteine-scanning mutagenesisof this region
identified residues crucial for SGK1 activation.When H73, N75, or
D76 were mutated to Cys, coexpressionof SGK1 had no effect or even
reduced the current. Bycontrast, the P77C mutation facilitated the
SGK1-mediatedstimulation of IKs (Figure 1c). These results
demonstrate theimportance of the C-terminal HxNDP-containing region
ofKCNE1 for targeted Q1/E1 vesicular transport to the
plasmamembrane.
LQTS-Associated Mutations in KCNQ1 or KCNE1Can Disrupt
SGK1-Dependent Modulation of IKsNext, we characterized the
mechanism of SGK1-dependentmodulation of several LQTS-associated
mutant IKs channels.Two common LQTS-associated missense mutations
inKCNE1 are located within the 73 to 79 region, namely S74Land
D76N.20 To characterize the mechanism of SGK1-
dependent modulation of these LQTS-associated mutant IKschannels
we studied heteromeric channels coassembled ofKCNQ1 and KCNE1(S74L)
(Q1/S74L) or KCNE1(D76N)(Q1/D76N) subunits. Q1/S74L channels were
activated bySGK1 (Figure 5), whereas currents mediated by
Q1/D76Nchannels were reduced by SGK1 (Figure 2a). Changes inplasma
membrane-associated KCNQ1 protein suggestedthat this functional
reduction in mutant IKs was caused bya trafficking defect. SGK1
increased wild-type Q1/E1 butnot Q1/D76N channel abundance in the
plasma membrane,as assayed by Western blot and chemiluminescence
anal-ysis of surface protein (Figure 2b and 2c).12 Furthermore,a
chemiluminescence assay of oocytes injected withcRNAs encoding
Myc-tagged KCNQ1 and KCNE1(D76N)showed that constitutively active
SGK1(S422D) but not
Figure 1. Deletion of the KCNQ1 N terminus and disruption of
amotif in KCNE1 impair SGK1-dependent stimulation. a, KCNQ1(Q1)
channels are stimulated when coexpressed with SGK1 inoocytes.
Channels were activated by 7-second pulses to 60 mV.Example traces
are shown overlaid. Horizontal scale bar, 1 sec-ond; vertical scale
bar, 1 �A. KCNQ1/KCNE1 (Q1/E1) coexpressedwith SGK1 yielded larger
currents than Q1/E1 expressed alone inXenopus oocytes. Horizontal
scale bar, 2 seconds; verticalscale bar, 3 �A. b, Deletion of the
KCNQ1 residues 1 to 81(KCNQ1�N-term) and deletion of KCNE1 residues
73 to 129(KCNE1�C-term) render IKs channels insensitive to SGK1
(n�8 to18). To determine the potentiation by SGK1, current
amplitudes inthe presence and absence of SGK1 were analyzed at the
end of7-second pulses to 60 mV and the ratio was calculated.
Drawingsindicate structure of channel subunits expressed. c,
Deletion ofresidues 73 to 79 but not of residues 80 to 129 of KCNE1
abol-ishes stimulation by SGK1 at 60 mV. The residues 73 to 80
wereindividually mutated to Cys, and the resulting mutant
channelswere tested for stimulation by SGK1 as described in b. A
uniquemotif (HxNDP) was required for the SGK1 effects (n�7 to 15).
TheKCNE1 constructs are illustrated on the left, and the numbers
indi-cate deleted residues or the position of individual Cys
substitu-tions. d, KCNQ1 and KCNE1 may interact at different sites
witheach other, and correct interaction at these sites is a
prerequi-site for stimulation by SGK1. Another requirement for this
stimu-lation is intact Ser27, a target of PKA phosphorylation
(supple-mental Figure I). Binding of �-tubulin to the KCNQ1 N
terminus(possibly influenced by Ser27 and allowed by correct
KCNQ1-KCNE1 interaction) may be a molecular linker to the
cytoskele-ton and may allow specific and efficient sorting of KCNQ1
pro-teins into early endosomes.
1452 Circulation Research December 5, 2008
at TEL AVIV UNIV on March 4, 2010
circres.ahajournals.orgDownloaded from
http://circres.ahajournals.org
-
inactive SGK1(K127N) decreased KCNQ1 protein in theplasma
membrane (Figure 2c). In COS-7 cells cotrans-fected with Q1/D76N
and SGK1(S422D), no increase inplasma membrane immunofluorescent
staining of KCNQ1was observed (Figure 2d). By contrast, we
previouslyreported that constitutively active RAB11 increasedplasma
membrane abundance of heterologously expressedwild-type KCNQ1
protein in COS-7 cells.12
The small G proteins RAB5 and RAB11 are expressed incardiac
tissue and oocytes, where they constitute centralcomponents of
recycling specificity and efficiency for vesi-cles containing
wild-type IKs channels.12,20 RAB5 has beenimplicated in the
regulation of early steps in the endocyticpathway, whereas RAB11 is
localized at the trans-Golginetwork, post-Golgi vesicles, and the
recycling endosome.21
Hydrolysis of GTP activates RAB-dependent vesicle
traffick-ing.16,22–26 The IKs recycling pathway can be assayed
by
injecting GTP into oocytes where RAB-dependent pathwaysare
functionally impaired by overexpression of mutant formsof RAB5 and
RAB11.12 Here, we used a similar approach toassay for the recycling
pathway of mutant IKs channels.Oocytes expressing Q1/D76N channels
were microinjectedwith GTP during voltage clamp, and ionic currents
wererecorded. Q1/D76N-mediated currents were increased byGTP when
channels were coexpressed with dominant-negative RAB11(S25N) or the
switch2 domain mutantRAB11(T77A).27 However, no change in
Q1/D76N-mediatedcurrents was noted when channels were coexpressed
withGTP-insensitive RAB5(N133I) alone or in combination witheither
of the RAB11 mutants (Figure 3a). Wild-type IKschannels colocalize
with RAB1112; however, using greenfluorescent protein (GFP)-tagged
KCNQ1 and DsRed-taggedRAB11 constructs, we observed no
colocalization of RAB11with Q1/D76N channels in COS-7 cells (Figure
3c). Taken
Figure 2. SGK1 decreasesQ1/D76N current density byreducing
plasma membraneabundance of the channels. a,KCNQ1 was coexpressed
withKCNE1 carrying the LQTS-associated mutation D76N,which is
localized in the regionimportant for SGK1 effects.Coexpression of
the mutantchannels with SGK1 resulted indecreased currents.
Channelswere activated by 7-secondpulses to varying potentials
(�80to 60 mV in 20 mV steps; n�17to 20; horizontal scale bar,
1second; vertical scale bar, 3 �A).b, Biotinylation Western
blotrevealed an increase of plasmamembrane KCNQ1 protein bySGK1 for
wild-type Q1/E1 and adecrease for Q1/D76N. FourWestern blots were
densito-metrically analyzed using Scionimage software. c,
Chemilumi-nescence assay of KCNQ1(Myc-tagged between S1 andS2)
coexpressed with wild-typeKCNE1 (E1) or KCNE1(D76N)(D76N) in the
absence or pres-ence of constitutively activeSGK1(SD) or inactive
SGK1(KN)mutant kinases. Data for wild-type KCNE1 are depicted
inblack; data for KCNE1(D76N) arein gray. d, KCNQ1(FLAG)/D76Nwas
coexpressed with a GFP-tagged constitutively activemutant
SGK1(S422D) in COS-7cells (SGK1-expressing cells aregreen).
KCNQ1(FLAG) wasprobed by immunostaining withan anti-FLAG antibody
(red).Visual observation and 2D pixel-intensity analysis using
ImageJsoftware suggest that theSGK1(S422D) mutant does not
markedly increase plasma membrane expression of KCNQ1(FLAG)/D76N
channels (position and direction of analyzed areas are indi-cated
by yellow arrows). The lower graph shows control data from
KCNQ1(FLAG)/KCNE1(wt) channels (replotted from our
previousstudy12). Here, the increased fluorescence in the plasma
membrane can be observed. Error bars indicate �SEM. *Significant
differ-ences (P�0.05).
Seebohm et al Disrupted IKs Trafficking in LQTS 1453
at TEL AVIV UNIV on March 4, 2010
circres.ahajournals.orgDownloaded from
http://circres.ahajournals.org
-
together, these results indicate that Q1/D76N channels
areendocytosed by RAB5 and reinserted into the plasma mem-brane by
a RAB11-independent mechanism.
RAB7 is a protein that has been implicated in the regula-tion of
late endosomal steps in the endocytic and lysosomalpathways.21 To
understand where Q1/D76N channels accu-mulate in the cell, we
coexpressed mutant channels withRAB7 and the dominant-negative RAB7
mutant T22N.RAB7(T22N) increased only Q1/D76N-mediated currentsbut
not wild-type IKs (Figure 4a). According to chemilumi-nescence,
coexpression of Q1/D76N channels withRAB7(T22N) increased the
amount of Q1/D76N plasmamembrane protein (Figure 4b).
Cotransfection of wild-typeand mutant Q1/E1 channels with
RAB7(T22N) in COS-7cells showed that Q1/D76N but not wild-type
channelscolocalize with RAB7-positive late endosomal vesicles
(Fig-ure 4c). These data suggest a close relationship of RAB7
withQ1/D76N but not wild-type channels.
To determine whether SGK1 modulates channels harbor-ing a
LQTS-associated mutation, we examined 8 previously
characterized KCNQ1 mutants and the S74L mutant ofKCNE1. Six of
the 9 mutant channels were stimulated bycoexpression with SGK1
(Figure 5). By contrast,Q1(P117L)/E1 channels were insensitive, and
Q1(Y111C)/E1- and Q1(L114P)/E1-mediated currents were reduced
bySGK1. These mutations were recently reported to disruptnormal
trafficking.28 Interestingly, Q1(L114P) expressedwithout E1 was
downregulated on SGK1 coexpression aswell, suggesting that E1 is
not required for the inversed SGK1sensitivity (Figure I in the
online data supplement). LikeQ1/E1(D76N) channels, Q1(L114P)/E1
channels were mod-ulated by RAB7(T22N) (Figure 4 and supplemental
Figure II)and were mistargeted in transfected cardiomyocytes
(supple-mental Figure III).28 Furthermore, Q1(Y111C)/E1
andQ1(L114P)/E1 channels colocalized with RAB7 but not
withRAB7(T22N) in COS7 cells (supplemental Figure IV). Sim-ilar to
KCNE1, the KCNQ1 residues required for activationby SGK1 are
located in an N-terminal juxtamembranousregion.28 This raises the
possibility that these regions ofKCNQ1 and KCNE1 interact to
promote trafficking of theheteromeric channel complex (Figure 5,
inset).
DiscussionIn a previous study, we showed that SGK1 enhances
theinsertion of Q1/E1 channels into the plasma membrane.12
However, structural prerequisites have not been studied
untilnow. The N terminus of KCNQ1 contains the importanttrafficking
motif LEL and a critical tyrosine residue (Tyr51),both of which are
required for Q1/E1 channel trafficking tobasolateral membranes in
MDCK cells.29 The LEL motif, aswell as Tyr51, might be involved in
the RAB5/11-dependentand SGK1-sensitive targeted recycling of IKs
channels.12
Furthermore, the N terminus of KCNQ1 contains a PXXPsequence
that may facilitate interactions with SH3 domains.Recently, direct
interaction of the N terminus of KCNQ1 with�-tubulin in transfected
COS-7 cells and in guinea pigcardiomyocytes was reported.30 Both
the interaction with�-tubulin and the phosphorylation of Ser27 in
KCNQ1 areprerequisites for protein kinase (PK)A-mediated activation
ofthe channels.30,31 Interestingly, mutation of Ser27 or deletionof
the N-terminal residues 1 to 81 in KCNQ1 (resulting inKCNQ1�N-term)
disrupt its stimulation by SGK1 (Figure 1band supplemental Figure
V). These data raise the possibilitythat stimulations by SGK1 or
PKA require both the integrityof a macromolecular complex and an
intact interaction withthe cytoskeleton.12,30,31 Misfolding of the
� subunit KCNE1as a result of deletion of the intracellular domain
or of specificsingle amino acid substitutions might disturb the
integrity ofa macromolecular complex and/or cytoskeleton–KCNQ1
in-teractions, disrupting correct intracellular sorting to
vesiclesthat are subject to SGK1-stimulated exocytosis.
However,stimulation by SGK1 involves increased trafficking to
theplasma membrane, whereas PKA-mediated stimulation ofKCNQ1 seems
not to be related to trafficking events.12,30
Here, we show that stimulation by SGK1 does not require
thepresence of KCNE1 � subunits (Figure 1a). However, whenKCNE1 is
present, a short stretch (73 to 79 region) within theKCNE1
intracellular C terminus is required for stimulationby SGK1 (Figure
1b and 1c). Within this region, we identi-
Figure 3. Q1/E1(D76N) is endocytosed by a RAB5-dependentand
recycled back to the plasma membrane by a RAB11-independent
SGK1-sensitive pathway. a, Q1/D76N expressedalone, with GTP
binding–insufficient RAB5(N133I), withdominant-negative
RAB11(S25N), with switch2 domain mutantRAB11(T77A), or with
combinations of the constructs. Oocytesexpressing Q1/D76N were
injected with 0.23 nmol GTP throughglass pipettes while currents
were recorded continuously by2-electrode voltage clamp (see inlay
to the right). Injection ofGTP increased Q1/D76N-mediated currents
when RAB11(S25N)or RAB11(T77A) were present (filled symbols). In
oocytesexpressing Q1/D76N together with
dominant-negativeRAB5(N133I) alone or in combinations with
RAB11(S25N) orRAB11(T77A) (open symbols), GTP had no effect on
Q1/D76Ncurrent amplitudes (n�3 to 8). These results indicate that
inhibi-tion of RAB5(N133I) may block accumulation of Q1/D76N at
theplasma membrane by RAB11-uncoupled exocytosis. b, GFP-tagged
Q1/D76N (green) was coexpressed with DsRed-fusedRAB11 (red) or
RAB11(S25N) (red) in COS-7 cells. Visual obser-vation and 2D
pixel-intensity analysis using ImageJ software didnot suggest
colocalization of the mutant channels with RAB11or RAB11(S25N)
[position and direction of analyzed areas areindicated by yellow
arrows, results of Q1/D76N scans are repre-sented as green curves,
results of RAB11/RAB11(S25N) scansas red curves]. Error bars
indicate �SEM.
1454 Circulation Research December 5, 2008
at TEL AVIV UNIV on March 4, 2010
circres.ahajournals.orgDownloaded from
http://circres.ahajournals.org
-
fied 4 residues (H73, N75, D76, and P77) that are critical
forthe normal effect of SGK1 on Q1/E1 channels (Figure 1c).These
results indicate that the C-terminal
juxtamembranousHxNDP-containing region of KCNE1 is important for
tar-geted Q1/E1 vesicular transport to the plasma membrane.
Theintact intracellular KCNE1 C terminus was shown to interactwith
the sarcomeric protein T-cap, suggesting a T-tubule–myofibril
linking system.32 Thus, Q1/E1 channel complexescontain several
molecular components allowing for physicallinkage to cytoskeletal
compartments, which may allowspecific and efficient trafficking
along the cytoskeleton (Fig-ure 1d).
Two common LQTS-associated missense mutations inKCNE1 are
located within the 73 to 79 region, namely S74Land D76N.33 Q1/S74L
channels were activated by SGK1(Figure 5), whereas Q1/D76N-mediated
currents were re-duced by active SGK1 (Figure 2a) possibly as a
result ofreduced plasma membrane-associated KCNQ1 protein
asdemonstrated by Western blot and chemiluminescence anal-ysis of
surface protein (Figure 2b and 2c).12 By analysis of 9previously
characterized LQT1 mutants, we identified 6mutant channels that
were stimulated by coexpression withSGK1 (Figure 5). By contrast, 3
mutant channels were eitherinsensitive or inhibited by SGK1. These
3 mutations (P117L,Y111C, and L114P in KCNQ1) were recently
reported todisrupt normal trafficking.28 We previously reported
that con-stitutively active SGK1 increased plasma membrane
abundanceof heterologously expressed wild-type KCNQ1 protein
inCOS-7 cells.12 This effect is absent in COS-7 cells
cotransfected
with Q1/D76N and constitutively active SGK1(S422D) (Figure2d).
Thus, the SGK1-stimulated plasma membrane insertionof Q1/E1 is
disrupted in several LQT1 mutant channels and 1LQT5 mutant Q1/E1
channel. Similar to KCNE1, theKCNQ1 residues required for
activation by SGK1 are locatedin an N-terminal juxtamembranous
region.28 This raises thepossibility that these regions of KCNQ1
and KCNE1 interactto promote trafficking of the heteromeric channel
complex(Figure 5, inset).
RAB5 has been implicated in the regulation of early stepsin the
endocytic pathway, whereas RAB11 is localized at thetrans-Golgi
network, post-Golgi vesicles, and the recyclingendosome.21 The D76N
mutation in KCNE1 uncouples IKschannels from normal RAB11-dependent
endosome recy-cling to the plasma membrane and induces a distinct,
RAB11-independent recycling pathway (Figure 3a). On the
contrary,wild-type IKs channels colocalize with RAB11.12
Takentogether, these results indicate that Q1/D76N channels
areendocytosed by RAB5 and reinserted into the plasma mem-brane by
a RAB11-independent mechanism. The lack ofacute functional effects
of RAB11(S25N) or RAB11(T77A)(Figure 3a) and the lack of
colocalization with RAB11(Figure 3b) suggest that RAB11-dependent
vesicle recyclingto the plasma membrane might be disrupted in
Q1/D76Nchannels. Consequently, Q1/D76N channels may
escapeRAB11-dependent recycling, and formation of storage vesi-cles
(as was observed for wild-type IKs channels) may becompromised.
RAB7 is a protein implicated in the regulation of lateendosomal
steps in the endocytic and lysosomal pathways.21
Figure 4. RAB7 modulates cur-rent density and plasma mem-brane
abundance ofQ1/E1(D76N) channels. a, Q1/E1and Q1/D76N were
expressed inoocytes in the absence or pres-ence of either wild-type
RAB7 orthe dominant-negative mutantRAB7(T22N). Q1/E1 currents
andQ1/D76N currents were analyzedat the end of a 7-second pulseto
60 mV and normalized to theQ1/E1 current and Q1/D76N cur-rent,
respectively (n�12 to 47).Data are represented asmeans�SEM. b,
Oocytesexpressing KCNQ1-Myc/E1 orKCNQ1-Myc/D76N were injectedwith
RAB7 cRNA or RAB7(T22N)cRNA. After 3 days, plasmamembrane
expression of Myc-tagged protein was analyzed bya chemiluminescence
assay. Theresults were normalized to theQ1/E1 and Q1/D76N
values,respectively. Data are represent-ed as means�SEM. c,
GFP-tagged Q1/E1 and Q1/D76Nwere coexpressed with DsRed-
fused RAB7 or RAB7(T22N) in COS-7 cells. Q1/E1 channels were
expressed intracellularly and in the plasma membrane and did
notcolocalize with intracellularly expressed RAB7 or RAB7(T22N).
However, GFP-tagged Q1/D76N colocalized to some degree
withDsRed-fused RAB7 but not with RAB7(T22N), as suggested by
visual observation and 2D pixel-intensity analysis using ImageJ
soft-ware. The position and direction of analyzed areas are
indicated by yellow arrows, results of GFP-Q1/E1 and GFP-Q1/D76N
scans arerepresented as green curves, and results of DsRed-RAB7 and
DsRed-RAB7(T22N) scans as red curves. Interestingly,
colocalizationwas mostly detected in intracellular compartments.
Error bars indicate �SEM, and significant differences (P�0.05) are
marked by anasterisk (*).
Seebohm et al Disrupted IKs Trafficking in LQTS 1455
at TEL AVIV UNIV on March 4, 2010
circres.ahajournals.orgDownloaded from
http://circres.ahajournals.org
-
The dominant-negative RAB7 mutant T22N increased
onlyQ1/D76N-mediated currents but not wild-type IKs (Figure 4a)by
increasing the amount of Q1/D76N plasma membraneprotein (Figure
4b). Like Q1/E1(D76N) channels,Q1(L114P)/E1 channels were modulated
by RAB7(T22N)(Figure 4 and supplemental Figure II). Furthermore,
cotrans-fection of wild-type and mutant Q1/E1 channels
withRAB7(T22N) in COS-7 cells showed that Q1/D76N but notwild-type
channels colocalize with RAB7-positive late endo-somal vesicles
(Figure 4c). Furthermore, Q1(Y111C)/E1 andQ1(L114P)/E1 channels
colocalize with RAB7 but not withRAB7(T22N) in COS7 cells
(supplemental Figure IV). Threefindings suggest that the
disease-associated mutant channelsmay be stored in late endosomes
and possibly the endoplas-mic reticulum (ER): (1) Q1/D76N channels
do not colocalizewith RAB11 and can be trafficked back to the
plasmamembrane by a GTP-dependent but RAB11-independentprocess
(Figure 3a and 3c); (2) Q1/D76N channels [andQ1(Y111C)/E1 and
Q1(L114P)/E1 channels] colocalize in anintracellular compartment
with RAB7 and are modulated byRAB7(T22N) (Figure 4); and (3) SGK1
treatment does notresult in additional fractional bands in Western
blots, indicat-ing that Q1/D76N channels are not trafficked to
lysosomesand digested by enzymes.
SGK1-mediated phosphorylation of PIKfyve and subse-quent
PI(3,5)P2 production act to regulate channel activity
viaRAB11-dependent vesicle exocytosis (Figure 6). Taken to-gether,
our findings suggest that Q1/D76N, Q1(Y111C)/E1,and Q1(L114P)/E1
channels are trafficked forward to RAB7-dependent late endosomal
vesicles and/or the endoplasmicreticulum. Indeed, Q1(Y111C)/E1 and
Q1(L114P)/E1 were
reported to be enriched in the ER.28 The localization
ofKCNE1(D76N) seems to be altered in stem cell–derivedventricular
myocytes as well (supplemental Figure V). Analtered localization of
Q1(Y111C)/E1, Q1(L114P)/E1, andQ1(P117L)/E1 channels has been
reported for cardiac myo-cytes.28 Stimulation of RAB11-dependent
exocytosis willresult in increased endocytosis of plasma membrane
contain-ing mutant Q1/E1 channels, reducing channel density in
theplasma membrane and thereby current amplitudes (Figure 6).Mutant
channels stored in late endosomes, ER, and possiblyGolgi apparatus
can potentially be trafficked to the plasmamembrane, and
stimulation of this ER export forward traf-ficking by GTP would
explain the RAB11-independent stim-ulation of Q1/D76N-mediated
currents (Figure 3a).
The present findings bear potential clinical
significance.Carriers of the E1(D76N), Q1(Y111C), or Q1(L114P)
muta-tions may benefit from avoidance of situations (eg,
sustainedstress, dexamethasone treatment, and excessive blood
insulinlevels) that might stimulate SGK1 and lead to an even
greaterdecrease in IKs and prolongation of QT intervals.
In summary, our studies demonstrate a link between
alteredvesicle recycling of disease-associated mutant IKs
channelsand the stress-dependent kinase SGK1.
Sources of FundingThis work was supported by the Deutsche
Forschungsgemeinschaft(La315/4-5 and SFB), a stipend from the
Gottlieb Daimler-und KarlBenz-Stiftung (to G.S.), a stipend from
the Erwin-Riesch-Stiftung (toG.S.) and MIUR-PRIN2004 (to C.B.), and
a Deutsche Forschungs-gemeinschaft stipend (GRK 1302/1) (to U.H.).
S.K. was funded by
Figure 5. Differential SGK1 sensitivity of Q1/E1 channels
con-taining LQTS1-associated mutations. Several
LQTS-associatedmutations in Q1 reduced IKs in oocytes by 40% to 70%
com-pared to wild-type Q1/E1. SGK1 partially recovered function
ofmost mutant channels (n�7 to 20). The pulse protocol used
isdescribed in Figure 1. Error bars indicate �SEM (*P�0.05).Inset,
Approximate positions of LQTS-associated point muta-tions studied
here are indicated (circles). Location of mutationsleading to
SGK1-mediated reduction in IKs are shown as lightgray [Q1(Y111C),
Q1(L114P)] or dark gray [KCNE1(D76N)] filledcircles.
Figure 6. Diagram of Q1/E1 and Q1/D76N channel
recycling.Wild-type Q1/E1 channels are endocytosed by a
RAB5-dependent mechanism and reinserted/recycled by a
RAB11-dependent mechanism. RAB11-dependent Q1/E1 exocytosis
isenhanced by SGK1, an effect involving the phosphorylation
andactivation of PIKfyve and the generation of PI(3,5)P2. This
mech-anism is disrupted in Q1/D76N channels. Q1/D76N channels
aresimilarly endocytosed via a RAB5-dependent endocytosis butare
forward-trafficked to RAB7-enclosing late endosomal vesi-cles and
possibly the ER and Golgi apparatus. Stimulation ofRAB11-dependent
exocytosis leads to increased membrane fluxinto the plasma
membrane, resulting in increased endocytosis.Because
KCNE1(D76N)-containing channels are not enriched inRAB11 vesicles,
their exocytosis is not stimulated, but they areendocytosed,
resulting in reduced functional expression of
thesedisease-associated channels. However, injection of GTP
stimu-lates trafficking of Q1/D76N channels from the late
endosomesand the ER or Golgi apparatus independent of RAB11
function.
1456 Circulation Research December 5, 2008
at TEL AVIV UNIV on March 4, 2010
circres.ahajournals.orgDownloaded from
http://circres.ahajournals.org
-
German National Genome Research Network grant 01GS0838 andthe
Leducq Fondation.
DisclosuresNone.
References1. Eliot RS, Buell JC. Role of emotions and stress in
the genesis of sudden
death. J Am Coll Cardiol. 1985;5:95B–98B.2. Schwartz PJ, Zaza A,
Locati E, Moss AJ. Stress and sudden death. The
case of the long QT syndrome. Circulation. 1991;83(suppl
II):II-71–II-80.
3. Paavonen KJ, Swan H, Piippo K, Hokkanen L, Laitinen P,
Viitasalo M,Toivonen L, Kontula K. Response of the QT interval to
mental andphysical stress in types LQT1 and LQT2 of the long QT
syndrome. Heart.2001;86:39–44.
4. Towbin JA, Wang Z, Li H. Genotype and severity of long QT
syndrome.Drug Metab Dispos. 2001;29:574–579.
5. Kern W, Perras B, Wodick R, Fehm HL, Born J. Hormonal
secretionduring nighttime sleep indicating stress of daytime
exercise. J ApplPhysiol. 1995;79:1461–1468.
6. Lang F, Cohen P. Regulation and physiological roles of serum-
andglucocorticoid-induced protein kinase isoforms. Sci STKE.
2001;2001:RE17.
7. Lang F, Henke G, Embark HM, Waldegger S, Palmada M, Bohmer
C,Vallon V. Regulation of channels by the serum and
glucocorticoid-inducible kinase - implications for transport,
excitability and cell prolif-eration. Cell Physiol Biochem.
2003;13:41–50.
8. Waldegger S, Barth P, Raber G, Lang F. Cloning and
characterization ofa putative human serine/threonine protein kinase
transcriptionallymodified during anisotonic and isotonic
alterations of cell volume. ProcNatl Acad Sci U S A.
1997;94:4440–4445.
9. Embark HM, Böhmer C, Vallon V, Luft F, Lang F. Regulation of
KCNE1dependent K�current by the serum and glucocorticoid-inducible
kinase(SGK) isoforms. Pflugers Arch. 2003;445:601–606.
10. Barhanin J, Lesage F, Guillemare E, Fink M, Lazdunski M,
Romey G.K(V)LQT1 and lsK (minK) proteins associate to form the
I(Ks) cardiacpotassium current. Nature. 1996;384:78–80.
11. Sanguinetti MC, Curran ME, Zou A, Shen J, Spector PS,
Atkinson DL,Keating MT. Coassembly of K(V)LQT1 and minK (IsK)
proteins to formcardiac I(Ks) potassium channel. Nature.
1996;384:80–83.
12. Seebohm G, Strutz-Seebohm N, Birkin R, Dell G, Bucci C,
Spinosa MR,Baltaev R, Mack AF, Korniychuk G, Choudhury A, Marks D,
Pagano RE,Attali B, Pfeufer A, Kass RS, Sanguinetti MC, Tavare JM,
Lang F.Regulation of endocytic recycling of KCNQ1/KCNE1
potassiumchannels. Circ Res. 2007;100:686–692.
13. Busjahn A, Seebohm G, Maier G, Toliat MR, Nurnberg P, Aydin
A, LuftFC, Lang F. Association of the serum and glucocorticoid
regulated kinase(sgk1) gene with QT interval. Cell Physiol Biochem.
2004;14:135–142.
14. Chen YH, Xu SJ, Bendahhou S, Wang XL, Wang Y, Xu WY, Jin
HW,Sun H, Su XY, Zhuang QN, Yang YQ, Li YB, Liu Y, Xu HJ, Li XF,
MaN, Mou CP, Chen Z, Barhanin J, Huang W. KCNQ1
gain-of-functionmutation in familial atrial fibrillation. Science.
2003;299:251–254.
15. Hong K, Piper DR, Diaz-Valdecantos A, Brugada J, Oliva
A,Burashnikov E, Santos-de-Soto J, Grueso-Montero J, Diaz-Enfante
E,Brugada P, Sachse F, Sanguinetti MC, Brugada R. De novo
KCNQ1mutation responsible for atrial fibrillation and short QT
syndrome inutero. Cardiovasc Res. 2005;68:433–440.
16. Hoekstra D, Tyteca D, Van IJzendoorn SC. The subapical
compartment:a traffic center in membrane polarity development. J
Cell Sci. 2004;117:2183–2192.
17. Choudhury A, Dominguez M, Puri V, Sharma DK, Narita K,
WheatleyCL, Marks DL, Pagano RE. Rab proteins mediate Golgi
transport ofcaveola-internalized glycosphingolipids and correct
lipid trafficking inNiemann-Pick C cells. J Clin Invest.
2002;109:1541–1550.
18. Sharma DK, Choudhury A, Singh RD, Wheatley CL, Marks DL,
PaganoRE. Glycosphingolipids internalized via caveolar-related
endocytosisrapidly merge with the clathrin pathway in early
endosomes and formmicrodomains for recycling. J Biol Chem.
2003;278:7564–7572.
19. Kanki H, Kupershmidt S, Yang T, Wells S, Roden DM. A
structuralrequirement for processing the cardiac K� channel KCNQ1.
J BiolChem. 2004;279:33976–33983.
20. Kessler A, Tomas E, Immler D, Meyer HE, Zorzano A, Eckel J.
Rab11is associated with GLUT4-containing vesicles and redistributes
inresponse to insulin. Diabetologia. 2000;43:1518–1527.
21. Grosshans BL, Ortiz D, Novick P. Rabs and their effectors:
achievingspecificity in membrane traffic. Proc Natl Acad Sci U S A.
2006;103:11821–11827.
22. Marks DL, Pagano RE. Endocytosis and sorting of
glycosphingolipids insphingolipid storage disease. Trends Cell
Biol. 2002;12:605–613.
23. Maxfield FR, McGraw TE. Endocytic recycling. Nat Rev Mol
Cell Biol.2004;5:121–132.
24. Prekeris R. Rabs, Rips, FIPs, and endocytic membrane
traffic. Sci World J.Available at:
http://www.ncbi.nlm.nih.gov/sites/entrez.
25. Ullrich O, Horiuchi H, Bucci C, Zerial M. Membrane
association of Rab5mediated by GDP-dissociation inhibitor and
accompanied by GDP/GTPexchange. Nature. 1994;368:157–160.
26. Pasqualato S, Senic-Matuglia F, Renault L, Goud B, Salamero
J, CherfilsJ. The structural GDP/GTP cycle of Rab11 reveals a novel
interfaceinvolved in the dynamics of recycling endosomes. J Biol
Chem. 2004;279:11480–11488.
27. Pfeffer SR. Structural clues to Rab GTPase functional
diversity. J BiolChem. 2005;280:15485–15488.
28. Dahimene S, Alcolea S, Naud P, Jourdon P, Escande D,
Brasseur R,Thomas A, Baro I, Merot J. The N-terminal
juxtamembranous domain ofKCNQ1 is critical for channel surface
expression: implications in theRomano-Ward LQT1 syndrome. Circ Res.
2006;99:1076–1083.
29. Jespersen T, Rasmussen HB, Grunnet M, Jensen HS, Angelo K,
DupuisDS, Vogel LK, Jorgensen NK, Klaerke DA, Olesen SP.
Basolaterallocalization of KCNQ1 potassium channels in MDCK cells:
molecularidentification of an N-terminal targeting motif. J Cell
Sci. 2004;117:4517–4526.
30. Nicolas CS, Park KH, El Harchi A, Camonis J, Kass RS,
Escande D,Mérot J, Loussouarn G, Le Bouffant F, Baró I. IKs
response to proteinkinase A-dependent KCNQ1 phosphorylation
requires direct interactionwith microtubules. Cardiovasc Res.
2008;79:427–435.
31. Marx SO, Kurokawa J, Reiken S, Motoike H, D’Armiento J,
Marks AR,Kass RS. Requirement of a macromolecular signaling complex
for betaadrenergic receptor modulation of the KCNQ1-KCNE1
potassiumchannel. Science. 2002;295:496–499.
32. Furukawa T, Ono Y, Tsuchiya H, Katayama Y, Bang ML, Labeit
D, et al.Specific interaction of the potassium channel beta-subunit
minK with thesarcomeric protein T-cap suggests a T-tubule-myofibril
linking system. JMol Biol. 2001;313:775–784.
33. Splawski I, Tristani-Firouzi M, Lehmann MH, Sanguinetti MC,
KeatingMT. Mutations in the hminK gene cause long QT syndrome and
suppressIKs function. Nat Genet. 1997;17:338–340.
Seebohm et al Disrupted IKs Trafficking in LQTS 1457
at TEL AVIV UNIV on March 4, 2010
circres.ahajournals.orgDownloaded from
http://circres.ahajournals.org
-
CIRCRESAHA/2008/177360/R2
1Su
Supplement Material Online Figure I SGK1 inhibits Q1(L114P)
currents in the absence of KCNE1.
Large amounts of cRNA (12 ng per oocyte) encoding the
LQTS-associated mutant KCNQ1(L114P) were injected into Xenopus
laevis oocytes. Coexpression of these mutant channels with SGK1 (5
ng of cRNA per oocyte) resulted in decreased currents measured at
60 mV (KCNQ1(L114P), n = 12-16). Data are represented as mean +
SEM.
Online Figure II Mutant Q1/E1 channels are modulated by
RAB7(T22N).
Heteromeric channels composed of KCNE1 and either the Y111C or
the L114P mutant of KCNQ1 were expressed in oocytes in the absence
or presence of RAB7 or RAB7(T22N). Q1/E1-current amplitudes were
determined at the end of a 7-s pulse to 60 mV and normalized to the
respective amplitudes in the absence of RAB7 (n = 23-29). Data are
represented as mean + SEM.
at TEL AVIV UNIV on March 4, 2010
circres.ahajournals.orgDownloaded from
http://circres.ahajournals.org
-
CIRCRESAHA/2008/177360/R2
2
Online Figure III Localization of KCNE1-EGFP and
KCNE1(D76N)-EGFP in stem cell-derived murine cardiac myocytes.
EGFP-tagged wild-type KCNE1 or KCNE1(D76N) were transfected into
stem cell-derived murine ventricular myocytes, and the green EGFP
fluorescence was detected by confocal microscopy. Three
representative examples of each transfection are shown. KCNE1-EGFP
seemed to be relatively evenly distributed in the cells, whereas
KCNE1(D76N)-EGFP localized to an unidentified fibrous structure, as
suggested by visual observation and 2D-pixel intensity analysis
using ImageJ-software. The position and direction of analysed areas
are indicated by yellow bars, results of scans are presented.
Analysis of frequencies of pixel intensity peaks revealed an
increase in narrow peaks. These peaks indicate an enrichment of
EGFP-tagged KCNE1(D76N) and to a lesser extent of EGFP-tagged
KCNE1wt to intracellular fibre-like structures. Data are
represented as mean ± SEM.
at TEL AVIV UNIV on March 4, 2010
circres.ahajournals.orgDownloaded from
http://circres.ahajournals.org
-
CIRCRESAHA/2008/177360/R2
3
Online Figure IV Colocalisation of RAB7 and LQTS-associated
mutant Q1/E1 channels.
VSV-tagged mutant Q1/E1 concatemers as described by Dahimene et
al. 1 were coexpressed in COS-7 cells with DsRed-fused RAB7 or
RAB7(T22N). VSV-tagged mutant Q1/E1 concatemers partially
colocalized with DsRed-fused RAB7 but not with RAB7(T22N), as
suggested by visual observation and 2D-pixel intensity analysis
using ImageJ-software. The position and direction of analyzed areas
are indicated by yellow bars, results of VSV-tagged mutant-Q1/E1
concatemer scans are represented as green curves, results of
DsRed-RAB7 and DsRed-RAB7(T22N) scans as red curves. Interestingly,
the colocalization is mostly seen in restricted intracellular
compartments.
at TEL AVIV UNIV on March 4, 2010
circres.ahajournals.orgDownloaded from
http://circres.ahajournals.org
-
CIRCRESAHA/2008/177360/R2
4
Online Figure V SGK1 inhibits function of
phosphorylation-deficient mutant Q1(S27A/D)/E1 channels.
Ser27 in KCNE1 (orange circle) is located in the N-terminal
region and is subject to PKA-mediated phosphorylation 2. Q1/E1
channels were expressed in oocytes in the absence or presence of
SGK1. Amplitudes of Q1/E1- and Q1(S27A/D)/E1-mediated currents were
determined at the end of a 7-s pulse to 60 mV and normalized to the
respective amplitudes in the absence of SGK1 (n = 8-11). Data are
represented as mean + SEM.
Methods:
Stem cell-derived ventricular cardiac myocytes - Cor.Ve murine
ventricular cardiomyocytes
(CellSystems Biotechnologie Vertrieb GmbH, St. Katharinen,
Germany) are derived from
transgenic mouse embryonic stem cells. These cells are a set of
ESC-derived 99.9% pure
ventricular cardiomyocytes that exhibit normal morphology and
physiological behavior. They
can be used for electrophysiology (patch clamp), cardiotoxicity,
and other functional studies
(http://www.axiogenesis.com/cms/front_content.php?client=1&lang=1&idcat=3).
The cells
possess puromycin resistance and GFP reporter genes driven by a
cardiac-specific promoter
(rlc2v promoter) and are obtained by in vitro differentiation
and puromycin selection of
mouse ES cells. The cells were thawed, seeded onto
fibronectin-coated dishes, and then
incubated for 48 h for complete recovery. Subsequently,
transfection of Cor.Ve ventricular
myocytes with KCNE1-EGFP or KCNE1(D76N)-EGFP was performed by
Fugene 6
transfection according to the manufacturer’s protocol.
Fluorescence was detected using a
at TEL AVIV UNIV on March 4, 2010
circres.ahajournals.orgDownloaded from
http://circres.ahajournals.org
-
CIRCRESAHA/2008/177360/R2
5
confocal microscope (LSM 510, Zeiss) with adequate filter sets.
The settings of confocal
imaging (filter set, detection intensity) of all images were
identical.
Western Blot – Western blot of plasma membrane proteins was
performed as reported earlier
3. Intact healthy oocytes were incubated in 1 mg/ml
Sulfo-NHS-LC-Biotin (Pierce, USA) for
30 min at room temperature. After washing five times in ND96, 20
intact oocytes were
homogenized in 400 µl H-buffer (in mM: 100 NaCl, 20 Tris-HCl, pH
7.4, 1% Triton X-100,
plus a mixture of protease inhibitors, CompleteTM, Roche,
Germany) and kept for 1 h at 4°C
on a rotator. Thereafter, the lysed oocytes were centrifuged for
1 min at 16,000 x g. The
supernatants were supplemented with 25 µl NeutrAvidin
Biotin-Binding Protein (Pierce,
USA) and incubated for 3 h at 4°C on a rotator. The beads were
then pelleted by
centrifugation for 2 min at 1600 x g and washed three times in
H-buffer. The pellets were
boiled in 40 µl SDS-PAGE loading buffer (sodium dodecyl
sulfate-polyacrylamide gel
electrophoresis, 0.8 M 2-mercaptoethanol, 6% SDS, 20% glycerol,
25 mM Tris-HCl, pH 6.8,
0.1% bromophenol blue). Finally, the samples were
Western-blotted and probed with primary
KCNQ1 antibody (anti-KCNQ1, 1:100 dilution, Santa Cruz:
sc-10646) and secondary
antibody (anti-rabbit, Santa-Cruz).
Molecular Biology – The molecular biological procedures were the
same as previously
described 3. Human KCNQ1 and SGK1 were subcloned into oocyte
expression vectors psp64,
a modified pcDNA3.1 vector, or pSGEM. The clones were mutated at
the positions
mentioned in the text by site-directed mutagenesis using PCR
with cloned Pfu-polymerase
(Invitrogen, Germany). Cloning procedures of wt and mutant RAB5,
RAB7, RAB11, and
FLAG-tagged KCNQ1 have been described previously 4-7.
SGK1(S422D) and SGK1(K127N)
were subcloned into pIRES2-EGFP. All constructs were confirmed
by sequencing. In vitro
at TEL AVIV UNIV on March 4, 2010
circres.ahajournals.orgDownloaded from
http://circres.ahajournals.org
-
CIRCRESAHA/2008/177360/R2
6
synthesis of cRNA was performed with SP6 and T7 mMessage
mMachine kits (Ambion via
Applied Biosystems, Germany).
Immunocytochemistry – COS-7 cells were grown on glass coverslips
and fixed with 4%
paraformaldehyde 3 days after transfection. Cells were
subsequently stained with an anti-
FLAG antibody (anti-FLAG polyclonal antibody from rabbit, F7425,
Sigma, Germany) to
detect the FLAG-tagged KCNQ1 as described before 3.
Immunostaining of VSV-tagged
mutant Q1/E1 concatemers was performed as described by Dahimene
et al. 2 using the same
anti-VSV antibody (1:500 dilution, Sigma). Fluorescence was
detected using a confocal
microscope (LSM 510, Carl Zeiss, Germany) with adequate filter
sets.
Chemiluminescence assay – Experiments were performed as
described previously 3. Oocytes
expressing KCNQ1 with an extracellular myc-tag (between S1 and
S2) were incubated for
30 min in ND96 with 1% bovine serum albumin (BSA) at 4°C.
Oocytes were subsequently
incubated with rat monoclonal anti-myc antibody (Roche, 100
µg/ml, dilution: 1:100 in ND96
+ 1% BSA) for 1 h at 4°C, washed 5 times at 4°C with ND96 + 1%
BSA, and incubated with
2 µg/ml peroxidase-conjugated affinity-purified F(ab)2 fragment
goat anti-rat IgG antibody
(Jackson ImmunoResearch, England) in ND96 + 1% BSA for 1 h.
Oocytes were washed
thoroughly for 5 min at 4°C with ND96 + 1% BSA and then 5 times
for 5 min at 4°C with
ND96. Individual oocytes were transferred to 100 µl Power Signal
Elisa solution (Pierce,
USA), and chemiluminescence was measured with a multilabel
counter (Wallac Victor,
Perkin Elmer, Germany). The results from 20 oocytes were
averaged and are presented in
relative light units (RLU).
at TEL AVIV UNIV on March 4, 2010
circres.ahajournals.orgDownloaded from
http://circres.ahajournals.org
-
CIRCRESAHA/2008/177360/R2
7
Online data reference List:
1. Dahimène S, Alcoléa S, Naud P, Jourdon P, Escande D, Brasseur
R, Thomas A, Baró
I, Mérot J. The N-terminal juxtamembranous domain of KCNQ1 is
critical for channel
surface expression: implications in the Romano-Ward LQT1
syndrome. Circ Res.
2006;99:1076-83.
2. Marx SO, Kurokawa J, Reiken S, Motoike H, D'Armiento J, Marks
AR, Kass RS.
Requirement of a macromolecular signaling complex for beta
adrenergic receptor
modulation of the KCNQ1-KCNE1 potassium channel. Science.
2002;295:496-
9.
3. Seebohm G, Strutz-Seebohm N, Birkin R, Dell G, Bucci C,
Spinosa MR, Baltaev R,
Mack AF, Korniychuk G, Choudhury A, Marks D, Pagano RE, Attali
B, Pfeufer A,
Kass RS, Sanguinetti MC, Tavare JM, Lang F. Regulation of
endocytic recycling of
KCNQ1/KCNE1 potassium channels. Circ Res. 2007;100:686-692.
4. Hoekstra D, Tyteca D, Van IJzendoorn SC. The subapical
compartment: a traffic
center in membrane polarity development. J Cell Sci.
2004;117:2183-2192.
5. Choudhury A, Dominguez M, Puri V, Sharma DK, Narita K,
Wheatley CL, Marks
DL, Pagano RE. Rab proteins mediate Golgi transport of
caveola-internalized
glycosphingolipids and correct lipid trafficking in Niemann-Pick
C cells. J Clin Invest.
2002;109:1541-1550.
6. Sharma DK, Choudhury A, Singh RD, Wheatley CL, Marks DL,
Pagano RE.
Glycosphingolipids internalized via caveolar-related endocytosis
rapidly merge with
the clathrin pathway in early endosomes and form microdomains
for recycling. J Biol
Chem. 2003;278:7564-7572.
7. Kanki H, Kupershmidt S, Yang T, Wells S, Roden DM. A
structural requirement for
processing the cardiac K+ channel KCNQ1. J Biol Chem.
2004;279:33976-33983.
at TEL AVIV UNIV on March 4, 2010
circres.ahajournals.orgDownloaded from
http://circres.ahajournals.org