Resea ch A icle
The Rocke elle Uni e si y P ess
J. Gen. Physiol. Vol. 149 No. 8 813–831
h ps://doi.o g/10.1085/jgp.201711802
The Jou nal o Gene al Physiology
813
In Roduc Ion
In elec ically exci able o gans, such as he hea , b ain,
and skele al muscle, ol age-ga ed Na+ (NaV) channels
cause he ini ia ion and p opaga ion o ac ion po en-
ials by conduc ing a la ge and apid inwa d Na+ lux.
Wi hin he cells o hese issues, NaV channels o m mac-
omolecula signaling complexes (Ab iel, 2010) whose
pa s wo k in conce o egula e channel unc ion.
The NaV β subuni membe s o his complex ha e been
shown o egula e cell adhesion (Isom e al., 1995; Mal-
ho a e al., 2000; Yu e al., 2003) and signaling in ad-
di ion o a ec ing channel densi y (Calhoun and Isom,
2014), ga ing kine ics (Fahmi e al., 2001; Wa anabe e
al., 2009; Calhoun and Isom, 2014), and pha macology
(Lenkowski e al., 2003; Uebachs e al., 2010). Howe e ,
he mechanisms whe eby he β subuni s in e ac wi h
he NaV channel α subuni o exe hei in luence on
ga ing emain undisco e ed.
Fi e ypes o NaV β subuni s ha e been iden i ied: β1,
β2, β3, β4, and β1b (Ha sho ne and Ca e all, 1984;
Messne and Ca e all, 1985; Kazen-Gillespie e al.,
2000; Mo gan e al., 2000; Yu e al., 2003). β2 and β4
o m co alen disul ide bonds wi h he α subuni (Isom
e al., 1995; Yu e al., 2003), whe eas β1 and β3 in e ac
non-co alen ly (Isom e al., 1992; Mo gan e al., 2000).
Wi h he excep ion o he β1b splice a ian (Pa ino e
al., 2011), he β subuni s comp ise a single ansmem-
b ane domain ha is e he ed o an ex acellula Ig
loop and a cy oplasmic C e minus (Calhoun and Isom,
2014). Ve y ecen ly, he co alen ly bound β2 and β4
subuni s we e c ys allized (Gilch is e al., 2013; Das e
al., 2016), and a c ucial disul ide bond o med by 55Cys
in β2 and 910Cys in he DII po e loop was iden i ied (Das
e al., 2016). Howe e , 910Cys is no p esen in NaV1.5,
and ins ead he homologous posi ion is 868Leu.
The β subuni s a e widely exp essed in many is-
sues, including he cen al and pe iphe al ne ous
sys em, he hea , and skele al muscle (Calhoun and
Isom, 2014). Despi e he sequence homology be ween
non-co alen ly associa ed β1 and β3 subuni s, hei ex-
p ession p o ile ac oss o gans di e s. Fo ins ance, β1,
bu no β3, is highly exp essed in skele al muscles (The
Human P o ein A las). In iguingly, e en in he same
o gan, β subuni localiza ion can di e (Fahmi e al.,
2001; Calhoun and Isom, 2014; Yuan e al., 2014). Fo
example, he β1 and β3 subuni s ha e been shown o
di e en ially exp ess in he a ia and en icles (Fahmi
e al., 2001; Wa anabe e al., 2009; Yuan e al., 2014),
sugges ing ha hey may speci ically ailo NaV channel
unc ion acco ding o cell ype. Mo eo e , β1 and β3
also ha e a a ied empo al exp ession p o ile du ing
hea de elopmen . β1 exp ession has been shown o
inc ease (Domínguez e al., 2005), whe eas β3 has been
shown o dec ease h ough emb yonic de elopmen
(Oka a e al., 2016). The dynamic exp ession pa e ns
o β1 and β3 sugges ha hese wo subuni s play dis inc
oles in he egula ion o NaV channel unc ion and he
ac ion po en ial.
Vol age-ga ed Na+ (NaV) channels comp ise a mac omolecula complex whose componen s ailo channel unc-
ion. Key componen s a e he non-co alen ly bound β1 and β3 subuni s ha egula e channel ga ing, exp ession,
and pha macology. He e, we p obe he molecula basis o his egula ion by applying ol age clamp luo ome y
o measu e how he β subuni s a ec he con o ma ional dynamics o he ca diac NaV channel (NaV1.5) ol -
age-sensing domains (VSDs). The po e- o ming NaV1.5 α subuni con ains ou domains (DI–DIV), each wi h a
VSD. Ou esul s show ha β1 egula es NaV1.5 by modula ing he DIV-VSD, whe eas β3 al e s channel kine ics
mainly h ough DIII-VSD in e ac ion. In oduc ion o a quenching yp ophan in o he ex acellula egion o he
β3 ansmemb ane segmen in e ed he DIII-VSD luo escence. Addi ionally, a luo opho e e he ed o β3 a he
same posi ion p oduced ol age-dependen luo escence dynamics s ongly esembling hose o he DIII-VSD.
Toge he , hese esul s p o ide compelling e idence ha β3 binds p oximally o he DIII-VSD. Molecula -le el
di e ences in β1 and β3 in e ac ion wi h he α subuni lead o dis inc ac i a ion and inac i a ion eco e y kine ics,
signi ican ly a ec ing NaV channel egula ion o cell exci abili y.
Mechanisms o nonco alen β subuni egula ion o NaV channel ga ing
WandiZhu,1 Taylo L.Voelke ,1 Zol anVa ga,4 AngelaR.Schube ,1 JeanneM.Ne bonne,2,3 and
Jona hanR.Sil a1
1Depa men o Biomedical Enginee ing, 2Depa men o De elopmen al Biology, and 3Depa men o In e nal Medicine,
Washing on Uni e si y in S . Louis, S . Louis, MO
4MTA-DE-NAP B Ion Channel S uc u e-Func ion Resea ch G oup, RCMM, Uni e si y o Deb ecen, Deb ecen, Hunga y
© 2017 Zhu e al. This a icle is a ailable unde a C ea i e Commons License (A ibu ion 4.0
In e na ional, as desc ibed a h ps ://c ea i ecommons .o g /licenses /by /4 .0 /).
Co espondence o Jona han R. Sil a: [email p o ec ed]
Abb e ia ions used: F-V, luo escence agains ol age; SSI, s eady-s a e inac i a-
ion; VCF, ol age clamp luo ome y; VSD, ol age-sensing domain.
on Janua y 31, 2018jgp. up ess.o gDownloaded om
h p://doi.o g/10.1085/jgp.201711802
Supplemen al ma e ial can be ound a :
Mechanisms o Na+ channel egula ion by β1 and β3 | Zhu e al.814
The po e- o ming NaV channel α subuni is com-
posed o ou homologous domains (DI–DIV) con-
nec ed by cy oplasmic linke s (Gellens e al., 1992).
Each domain is o med by six α helical ansmem-
b ane segmen s (S1–S6). The ou h segmen s (S4)
con ain mul iple posi i ely cha ged esidues ha mo e
ac oss he memb ane in esponse o changes in mem-
b ane po en ial. S4, oge he wi h S1–S3, o m he
ol age-sensing domains (VSDs) and a e coupled o
he S5 and S6, which o m he channel po e. Upon
memb ane depola iza ion, he S4 segmen s wi hin he
VSDs o DI–DIII a e p opelled ou wa d o open he
channel wi hin a millisecond; his is known as chan-
nel ac i a ion (Chanda and Bezanilla, 2002). Sho ly
he ea e , channels apidly close—a p ocess e med
“ as inac i a ion” ha is media ed by he in acellula
DIII–DIV linke and he DIV-VSD (Wes e al., 1992).
Bo h ac i a ion and inac i a ion ga ing ha e been
p e iously shown o be modula ed by he β1 and β3
subuni s (Mo gan e al., 2000; Fahmi e al., 2001; Wa a-
nabe e al., 2009; Calhoun and Isom, 2014).
Much mechanis ic insigh in o NaV channel ga ing
has been ecen ly p o ided by applying he ol age
clamp luo ome y (VCF) p o ocol, which is used o
luo escen ly ack VSD con o ma ion and co ela e
VSD kine ics wi h ionic cu en ga ing. Fo many yea s,
his p o ocol has been applied o s udy he skele al
muscle iso o m NaV1.4, and i has p o ided g ea in-
sigh in o he VSD oles in de e mining ac i a ion and
inac i a ion ga ing kine ics (Cha e al., 1999; Chanda
and Bezanilla, 2002; Sil a and Golds ein, 2013a,b), he
mechanisms o local anes he ic egula ion o he VSDs
(Mu oi and Chanda, 2009; A cisio-Mi anda e al., 2010),
and de ails o how oxins pa hologically a ec VSD ac-
i a ion (Campos e al., 2007, 2008). We ha e ecen ly
b oadened his app oach by de eloping VCF cons uc s
o ack VSD con o ma ions o all ou domains in he
ca diac pa alog, NaV1.5 (Va ga e al., 2015; Zhu e al.,
2016), whose ionic cu en modula ion by he β subuni
in oocy es mi o s he mammalian cell pheno ype.
We hypo hesized ha he non-co alen ly bound
β1 and β3 subuni s would modula e NaV1.5 ionic cu -
en kine ics by al e ing he ac i a ion o one o mo e
VSDs. In his s udy, we es his hypo hesis by applying
VCF o obse e he β subuni e ec s on he VSD o
each NaV1.5 domain.
MA eRIAls And Me hods
Molecula biology
cDNA encoding he human NaV β3 (UniP o KB/
Swiss-P o unde accession no. Q9NY72) subuni
was cus om syn hesized by Li e Technologies and
inse ed in o he pBSTA plasmid. cRNAs o he
human β1 subuni (UniP o KB/Swiss-P o unde ac-
cession no. Q07699.1) and α subuni NaV1.5 (acces-
sion no. Q14524.1) we e p oduced om he pBSTA
and pMAX ec o s, espec i ely. All mu agenesis
was accomplished using he QuikChange II si e-di-
ec ed mu agenesis ki (Agilen ), wi h p ime s om
Sigma-Ald ich. Mul iple colonies we e picked, and
plasmids we e isola ed using he NucleoSpin plas-
mid minip ep ki (Mache ey-Nagel). A e samples
we e con i med wi h sequencing (Genewiz), a single
clone was selec ed o a Midip ep p epa a ion (Nu-
cleoBond X a Midi; Mache ey-Nagel). Each plasmid
was hen linea ized wi h he No I o EcoRI es ic ion
enzyme and pu i ied wi h he NucleoSpin Gel and
PCR Clean-up ki (Mache ey-Nagel). Finally, capped
mRNA was syn hesized in i o using he mMES SAGE
mMAC HINE T7 T ansc ip ion Ki (Li e Technolo-
gies), pu i ied ia phenol–chlo o o m ex ac ion,
and econs i u ed o a concen a ion o ∼1 µg/µl.
Cu -open oocy e eco ding
mRNAs o he human α subuni NaV1.5 and β1 o β3
subuni s we e injec ed a a 3:1 mola a io (50–56 ng
pe cell o al) in o
Xenopus
oocy es. Oocy es we e hen
incuba ed a 18°C in ND93 solu ion (93 NaCl mM,
5 KCl mM, 1.8 CaCl2mM, 1 MgCl2mM, 5 HEP ES mM,
2.5 Na py u a e mM, and 1% penicillin–s ep omycin,
pH 7.4). 3–7 d a e injec ion, cu -open eco dings
(S e ani and Bezanilla, 1998; Rudokas e al., 2014)
we e pe o med using a cu -open ampli ie (CA-1B;
Dagan Co po a ion) coupled o an A/D con e e
(Digida a 1440; Molecula De ices). Clampex so wa e
( 10; Molecula De ices) was used o da a acquisi ion.
Du ing eco ding, he empe a u e was main ained a
19°C wi h a con olle (HCC-100A; Dagan Co po a-
ion). The in e nal eco ding solu ion was composed
o 105 NMG-Mes mM, 10 Na-Mes mM, 20 HEP ES mM,
and 2 EGTA mM, a a pH le el o 7.4, and he ex e -
nal solu ion was composed o 25 NMG-Mes mM, 90
Na-Mes mM, 20 HEP ES mM, and 2 Ca-Mes2mM, a a
pH le el o 7.4.
Be o e eco ding, he memb ane capaci ance com-
pensa ion and P/–8 leak sub ac ion we e applied.
The ionic cu en s we e eco ded using he s anda d
I-V p o ocol. F om a holding po en ial o −120 mV,
cells we e s epped o a 100-ms p epulse o −120 mV
and hen s epped o es po en ials anging om −120
o 60 mV wi h a 10-mV inc emen , p eceded by a 100-
ms pos pulse o −120 mV. Fo s eady-s a e inac i a ion
(SSI), cells we e held a es po en ial o 200 ms; a ail-
abili y was hen es ed using a depola izing pulse o
−20 mV. Ga ing cu en s we e eco ded du ing es
pulses om −150 o 50 mV om a holding po en ial
o −120 mV. Capaci ance and leak we e compensa ed
by P/4 leak sub ac ion wi h a subsweep po en ial o
40 mV. Ga ing cha ge– ol age (Q-V) cu es we e con-
s uc ed by in eg a ing ga ing cu en s o e 7 ms a e
he ol age s ep.
on Janua y 31, 2018jgp. up ess.o gDownloaded om
815JGP Vol. 149, No. 8
Vol age clamp luo ome y
Be o e eco ding, oocy es we e labeled wi h 10 µmol/L
me hane hiosul ona e-ca boxy e ame hyl hodamine
(MTS-TAM RA; San a C uz Bio echnology) in a depo-
la izing solu ion (in mM: 110 KCl, 1.5 MgCl2, 0.8 CaCl2,
0.2 EDTA, and 10 HEP ES, pH 7.1) o 30 min on ice.
Fluo escence da a we e collec ed simul aneously wi h
ionic cu en on a cus om ig (Va ga e al., 2015), com-
bining he cu -open ol age clamp and an epi luo es-
cence up igh mic oscope (FN1; Nikon), using a 40×
wa e -imme sion objec i e wi h 0.8 NA (CFI Plan Fluo ;
Nikon). A g een, high-powe ed LED (Luminus; PT-
121) was used o illumina ion, con olled by a d i e
(Lumina Powe ; LDPC-30-6-24VDC) by Clampex so -
wa e. The emission ligh was measu ed wi h a pho odi-
ode (PIN-040A; Uni ed De ec o Technology) moun ed
on he mic oscope epi luo escence po . The pho ocu -
en s gene a ed by he pho odiode we e hen ampli ied
by a pa ch clamp ampli ie (Axopa ch-200A; Molecula
De ices). Each luo escence ace is a mean o 7–10 lu-
o escence eco dings o he same cell.
Da a analyses
Da a analyses we e pe o med using Clamp i ( 10; Mo-
lecula De ices), MAT LAB (R2012a; MAT LAB), and Excel
(Mic oso ). Fo luo escence da a, signals we e low-pass
il e ed a 1 kHz o line be o e analysis. To co ec o pho-
obleaching, he baseline luo escence ace, which has no
change in ol age, was i and sub ac ed om he aces
eco ded when he ol age p o ocol was applied.
S eady-s a e ol age dependence cu es (G-V, luo-
escence agains ol age [F-V], SSI) we e quan i ied
by i ing a Bol zmann unc ion: y = 1/(1 + exp[(V −
V1/2)/k]). Sample sizes we e chosen so ha he s an-
da d e o o mean was less han 0.1 o each da a poin ,
and a minimum sample size o h ee was de e mined o
calcula e he SD. Each da a poin shown e lec s
n
= 3
o mo e om wo o mo e ba ches o oocy es. S a is ics
o compa ison be ween di e en cons uc s we e pe -
o med using an independen
es (Mic oso Excel).
The ± symbols in he ex and able and he e o ba s
in he igu es ep esen he SEMs.
Online supplemen al ma e ial
The supplemen al ma e ial con ains da a o channel
cu en s p ope ies ha a e no depic ed in he main
igu es and o he con ol da a. Figs. S1 and S2 show
channel ac i a ion and inac i a ion wi h o wi hou β1
and β3 o NaV1.5 exp essed in HEK 293T cells and
o ou VCF cons uc s. Fig. S3 shows he ol age de-
pendence o luo escence and ga ing cha ges o he
decoupling mu a ions, A1330W and N1759A. Fig. S4
shows he luo escence da a om all ou domains wi h
S156W β1 o S155W β3. Fig. S5 shows DIII and DIV lu-
o escence deac i a ion kine ics compa isons o NaV1.5
exp essed wi h β1/β3 chime as.
Resul s
β1 egula es channel inac i a ion by al e ing ol age-
dependen DIV-VSD ansi ions
We coexp essed he human β1 subuni wi h he
po e- o ming NaV1.5 α subuni in
Xenopus
oocy es by
coinjec ing β1 and α subuni mRNA a a mola a io o
3:1. β1 coexp ession had no signi ican e ec on he
ol age dependence o WT channel ac i a ion, as shown
by he conduc ance- ol age (G-V) cu e (Fig.1b and
Table1) bu caused a depola izing shi in he chan-
nel s eady-s a e inac i a ion (SSI) cu e compa ed wi h
WT alone (ΔV1/2 = 12.2 ± 1.4 mV, P = 0.02; Fig.1b).
The igh -shi ed SSI cu e implies ha mo e channels
a e a ailable o open a highe po en ials. Mo eo e ,
β1 u he inc eased channel opening by accele a ing
channel eco e y om inac i a ion (Fig.1d). To ensu e
ha he changes in he β1 egula ion mechanism ha
we obse ed we e consis en ac oss di e en exp es-
sion sys ems, we also used iden ical p o ocols o assess
β1 e ec s on NaV1.5 cu en s in HEK 293T cells and
obse ed simila beha io (Fig. S1 a). The β1-induced
depola iza ion o SSI we obse ed is also consis en wi h
p e ious esul s in HEK 293 and HEK 293T cells (An
e al., 1998; Malho a e al., 2001; Mal se e al., 2009).
We in es iga ed how β1 modula es inac i a ion by
i s measu ing ga ing cu en s, which e lec cha ge
ansloca ion o all ou VSDs. To be able o measu e
he ga ing cu en o he NaV1.5 channel, we used
he WT LFS cons uc , which con ains C373Y mu a-
ion ha inc eases channel sensi i i y o TTX, and he
Y1977A mu a ion, which p e en s ubiqui ina ion o he
channels o inc ease exp ession (Va ga e al., 2015).
Compa ison be ween he ga ing cha ge– ol age de-
pendence (Q-V) o WT LFS channels coexp essed wi h
and wi hou he β1 subuni (Fig.1e) e ealed ha β1
caused a depola izing shi in he Q-V cu e a nega i e
po en ials, esul ing in a s eepe Q-V ela ionship (Δk
= −10.1 ± 4.2 mV, P = 0.04). This esul sugges s ha
in he p esence o he β1 subuni , one o mo e o he
VSDs equi es highe po en ials o ac i a e. To iden i y
which VSD was a ec ed, NaV1.5 VCF cons uc s we e
coexp essed wi h he β1 subuni . We ha e p e iously
shown ha MTS-TAM RA–labeled NaV1.5 channels ac i-
a e and inac i a e simila ly o WT channels (Va ga e
al., 2015). Coexp ession o he β1 subuni wi h he VCF
cons uc s caused a shi in he SSI cu es ha is simila
o he shi caused by β1 in WT channels (Fig. S1 b).
The ol age dependence o ac i a ion o each VSD can
be desc ibed by plo ing he s eady-s a e luo escence
agains ol age (F-V) cu e. In compa ison o α alone,
he β1 subuni did no signi ican ly al e he DI, DII,
o DIII F-V cu es (Fig.1, –h), bu induced a s ong
depola izing shi in he DIV F-V cu e (ΔV1/2 = 31.3 ±
1.7 mV, P = 0.02; Fig.1i). Thus, in he p esence o he
β1 subuni , he DIV-VSD equi es highe po en ials o
on Janua y 31, 2018jgp. up ess.o gDownloaded om
Mechanisms o Na+ channel egula ion by β1 and β3 | Zhu e al.816
Figu e 1. naV β1 egula es naV1.5 inac i a ion by al e ing dIV-Vsd ac i a ion. NaV1.5 ionic cu en s we e measu ed using he
cu -open ol age clamp echnique o esol e as Na+ channel kine ics. Changes in si e-speci ic luo escence o NaV1.5 a e epo ed
by ou VCF cons uc s (V215C, S805C, M1296C, and S1618C) a e conjuga ing o MTS-TAM RA (Va ga e al., 2015). The mean ±
SEM is epo ed o g oups o h ee o eigh cells. The e o ba s ep esen he SEMs. Some e o ba s a e no isible due o small
SEMs. (a) Topology o NaV1.5 and NaV β1 subuni s on plasma memb ane. The β1 subuni is a single ansmemb ane p o ein con-
aining an ex acellula Ig domain and a sho in acellula C e minus. (b) Vol age dependence o ac i a ion (G-V) and s eady-s a e
inac i a ion (SSI) o WT NaV1.5 wi h β1 (α + β1, squa e) o wi hou β1 (α, ci cle). The G-V cu e is cons uc ed by measu ing he peak
cu en du ing es pulses om −120 o 20 mV om a holding po en ial o −120 mV and hen di iding by he d i ing o ce ( es
pulse po en ial minus e e sal po en ial). The e e sal po en ial o each cell is de e mined indi idually. Fo he SSI cu e, cells a e
held om −150 o 20 mV wi h a 10-mV inc emen o 200 ms. A ailabili y is hen measu ed by he peak cu en ha esul s om a
es pulse o −20 mV. Da a a e i wi h a Bol zmann equa ion (solid lines), and pa ame e s a e epo ed in Table1. (c) Rep esen a i e
cu en aces o WT channel wi h β1 (black) o wi hou β1 (g ay) in esponse o he depola izing pulse o 0 mV om −120 mV. Cu en
aces shown a e cons uc ed as a mean o h ee no malized cu en aces. Channels wi h o wi hou β1 show compa able ac i a ion
and inac i a ion kine ics. (d) Time dependence o ac ion o cu en eco e ed o channels wi h β1 (α + β1, squa e) o wi hou β1 (α,
ci cle). Cells we e i s depola ized o −20 mV o 200 ms o induce inac i a ion; hen, a e a ious eco e y du a ions a −120 mV,
cells we e depola ized o −20 mV o es he ac ion o cu en eco e ed. (e) Ga ing cha ge– ol age (Q-V) cu es o he WT LFS
channel wi h β1 (α + β1, squa e) o wi hou β1 (α, ci cle). Ga ing cu en s we e eco ded du ing es pulses om −150 o 50 mV om
a holding po en ial o −120 mV. Capaci ance and leak we e compensa ed by P/4 leak sub ac ion wi h a subsweep po en ial o 40
on Janua y 31, 2018jgp. up ess.o gDownloaded om
817JGP Vol. 149, No. 8
accomplish i s ac i a ion ansi ion, consis en wi h he
ga ing cha ge shi (Fig.1e).
P e iously, DIV-VSD ac i a ion was shown o be mo e
closely linked o NaV channel inac i a ion han ac i a-
ion (Cha e al., 1999; Capes e al., 2013). Speci ically,
he DIV-VSD was obse ed o be immobilized by as
inac i a ion (Cha e al., 1999), and DIV-VSD ac i a ion
was shown o be he a e-limi ing s ep o as inac i a-
ion (Capes e al., 2013). Hence, changes in he ol age
dependence o DIV-VSD ac i a ion o DIV-VSD kine -
ics would be expec ed o cause co ela ed changes in
channel SSI o inac i a ion kine ics. No ably, DIV-VSD
deac i a ion kine ics a e also as e wi h β1 ( 100–10% =
4.5 ± 0.6 ms a −160 mV a e 0-mV pulse) compa ed
wi h α alone ( 100–10% = 13.2 ± 0.4 ms a −160 mV a e
0-mV pulse, P = 0.0003). Thus, ou esul s imply ha he
β1 subuni egula es inac i a ion by al e ing DIV-VSD
ansi ions. This inding is consis en wi h he esul s
o p e ious s udies sugges ing ha β1 binds o he C
e minus o NaV1.1 (Spampana o e al., 2004) and he
S5–S6 linke o DIV o NaV1.4 (Maki a e al., 1996). Ou
esul s build on hese p e ious indings by connec ing
VSD egula ion o al e ed inac i a ion kine ics.
E en hough β1 does no a ec he ol age depen-
dence o DIII-VSD ac i a ion, compa ison o DIII-VSD
deac i a ion kine ics in he p esence o β1 shows ha
DIII-VSD eco e y o he es ing posi ion upon epo-
la iza ion is as e and mo e comple e ( 100–10% = 16.5 ±
1.6 ms a 0 mV) in con as o α alone ( 100–10% = 23.9 ±
2.3 ms a 0 mV, P = 0.05; Fig.1h, igh ). In p e ious
s udies, as inac i a ion was shown o immobilize he
ga ing cha ge displaced by DIII and DIV (A ms ong
and Bezanilla, 1977; Cha e al., 1999), pa icula ly DIII-
VSD (Shee s and Hanck, 2005; Va ga e al., 2015). Ou
esul s sugges ha β1 allows he DIII and DIV VSDs o
eco e o he es ing s a e mo e quickly. Gi en he link
o inac i a ion, his mo e apid eco e y o he VSDs
will allow channels o eco e mo e quickly om inac i-
a ion (Fig.1d) and become a ailable o exci a ion in
a sho e amoun o ime.
β3 al e s channel ac i a ion and inac i a ion by
modula ing DIII and DIV VSD kine ics
We coexp essed he β3 subuni wi h NaV1.5 using he
same p o ocols ha we e used o β1 (Fig.2a). As wi h
he β1 subuni , β3 had no appa en e ec on he ol age
dependence o channel ac i a ion (G-V; Fig.2b), bu
slowed ionic cu en ac i a ion and inac i a ion kine ics
(Fig.2c). I also caused a depola izing shi (ΔV1/2 =
8.7 ± 1.5 mV, P = 0.02) in SSI (Fig.2b and Table1),
implying ha β3 exp ession inc eases NaV1.5 channel
a ailabili y a highe po en ials. A simila β3-induced
SSI shi was also p esen in HEK cells eco ded wi h
iden ical p o ocols (Fig. S2 a). Unlike β1, β3 does no
signi ican ly al e channel eco e y kine ics (Fig.2d).
When we coexp essed β3 wi h he ou VCF cons uc s,
he ga ing e ec s o β3 we e p ese ed (Fig. S2 b), ex-
cep wi h he DII LFS cons uc , whe e he shi in SSI
induced by β3 is less p onounced. Ou obse a ions o
he ionic cu en changes induced by β3 coexp ession
a e consis en wi h he ga ing e ec s shown p e iously
in oocy es (Fahmi e al., 2001) and he
scn3b
knockou
mouse pheno ype (Hakim e al., 2008).
Like β1, β3 caused a depola izing shi in he Q-V
cu e a nega i e po en ials and a s eepe Q-V ela ion-
ship (Δk = 8.6 ± 3.0 mV, P = 0.05), showing ha β3 also
al e s he ol age dependence o VSD ac i a ion. Con-
side ing he homology be ween he β1 and β3 subuni s,
he simila Q-V cu es a e no su p ising. Howe e ,
compa ison o he F-V cu es o α alone and α wi h β3
shows ha β3 induces a depola izing shi in he DIII
F-V cu e (DIII F-V: ΔV1/2 = 20.7 ± 3.9 mV, P = 0.01) in
addi ion o i s depola izing e ec on he DIV F-V cu e
(DIV F-V: ΔV1/2 = 25.0 ± 7.7 mV, P = 0.01; Fig.2, h and
i; and Table1), causing bo h he DIII and DIV VSDs o
ac i a e a highe po en ials. The DIV F-V depola izing
shi occu s o e he same po en ial ange as he shi
in SSI, consis en wi h he indings ha DIV-VSD ac i-
a ion s ongly co ela es wi h channel inac i a ion and
wi h he a o emen ioned expe imen s wi h β1.
Channel opening is known o be egula ed by DIII-
VSD ac i a ion (Mu oi e al., 2010; Wang e al., 2016).
Ye , β3 induced depola iza ion o DIII-VSD ac i a ion
wi hou a ec ing he channel ol age dependence o ac-
i a ion (G-V), which may be due o DIII-VSD ac i a ion
a e y nega i e po en ials in he NaV1.5 pa alog. S ill,
β3 slowed ionic cu en ac i a ion kine ics (α alone: dI/
d max = 1.6 ± 0.1 ms−1; α + β3: dI/d max = 1.1 ± 0.1 ms−1,
P = 0.04; Fig.2c). Slowe inac i a ion a es can also esul
in slowe ac i a ion kine ics when no malized cu en s
a e compa ed because channel ac i a ion and inac i a-
ion a e igh ly coupled (Ald ich e al., 1983). Thus, he
slowed ac i a ion kine ics we obse ed wi h β3 could al-
e na i ely be caused by slowed inac i a ion kine ics. In
mV. Q-V cu es we e cons uc ed by in eg a ing ga ing cu en s o e 7 ms a e he ol age s ep. ( –i, le ) Vol age dependence o
s eady-s a e luo escence (F-V cu e) om all ou domains—( ) DI-S216C, (g) DII-S805C, (h) DIII-M1296C, and (i) DIV-S1618C—co-
exp essed wi h β1 (α + β1, squa es) o wi hou β1 (α, ci cles). F-V cu es a e measu ed wi h 50-ms depola izing pulses, anging om
−180 o 20 mV, wi h a 20-mV inc emen . The luo escence change a each po en ial, ΔF, is de e mined by aking he mean o he
signal ampli ude a e i eaches s eady s a e. β1 coexp ession causes a depola izing shi in he DIV F-V cu e wi hou signi ican ly
a ec ing o he domains. ( –i, igh ) Rep esen a i e luo escence signals showing he kine ics o VSD ac i a ion om each domain e-
sul ing om 50-ms depola izing pulses anging om −160 o 40 mV wi h 20-mV inc emen s a e a p epulse o −120 mV. Fo cla i y,
only ou aces a e shown o each cons uc . Pe cen age o luo escence change (ΔF/F) is epo ed.
on Janua y 31, 2018jgp. up ess.o gDownloaded om

Mechanisms o Na+ channel egula ion by β1 and β3 | Zhu e al.818
Figu e 2. naV β3 subuni a ec s naV1.5 inac i a ion by modi ying bo h dIII and dIV Vsd ac i a ion. NaV1.5 ionic cu -
en s and si e-speci ic luo escence changes a e measu ed as desc ibed in Fig.1. G oups o 3–10 cells a e epo ed as mean
± SEM. (a) Topology o NaV1.5 and NaV β3 subuni s on plasma memb ane. The NaV β3 subuni s uc u e is homologous o he
β1 subuni , and i has also been shown o exp ess in he myoca dium (Hu e al., 2012). (b) Vol age dependence o ac i a ion
(G-V) and s eady-s a e inac i a ion (SSI) o WT NaV1.5 wi h β3 (α + β3, squa es) o wi hou β3 (α, ci cles). The G-V cu e and
SSI cu e we e cons uc ed and eco ded as shown in Fig.1. Bol zmann i pa ame e s a e lis ed in Table1. (c) Rep esen a i e
cu en aces o WT channel wi h β3 (black) o wi hou β3 (g ay) in esponse o depola izing he pulse o 0 om −120 mV.
Channels wi h β3 show slowe ac i a ion and deac i a ion kine ics compa ed wi h α alone. (d) Time dependence o ac ion o
cu en eco e ed o channels wi h β3 (α + β3, black squa es) o wi hou β3 (α, g ay ci cles). The same p o ocol was used as
shown in Fig.1. (e) Ga ing cha ge– ol age (QV) cu e o WT LFS NaV1.5 wi h β3 (α + β3, squa es) o wi hou β3 (α, ci cles). The
Q-V cu e was eco ded and cons uc ed as shown in Fig.1. Bol zmann i pa ame e s a e lis ed in Table1. ( –i, le ) Vol age
dependence o luo escence (F-V cu e) om ou VCF cons uc s—( ) DI-V215C, (g) DII-S805C, (h) DIII-M1296C, and (i) DIV-
S1618C—coexp essed wi h β3 (α + β3, squa es) o wi hou β3 (α, ci cles). F-V cu es a e cons uc ed and eco ded as shown in
Fig.1. β3 coexp ession causes depola izing shi s in DIII and DIV F-V wi hou signi ican ly a ec ing he o he wo domains. ( –i,
igh ) Rep esen a i e luo escence signals ep esen ing he kine ics o each VSD ac i a ion. Fo cla i y, only ou aces wi h a
40-mV in e al a e shown o each cons uc .
on Janua y 31, 2018jgp. up ess.o gDownloaded om
819JGP Vol. 149, No. 8
con as o β1, β3 only accele a es DIII-VSD deac i a ion
(α + β3: 100–10% = 9.2 ± 1.3 ms; α alone: 100–10% = 23.9 ±
2.3 ms, P = 0.005), bu no DIV-VSD deac i a ion (see
Table4). In he NaV1.5 channel, he DIII-VSD ac i a es
a e y nega i e po en ials (∼160 mV). Thus, i is a ech-
nical challenge o acqui e he nega i e baseline o he
DIII F-V cu e o he mo e hype pola ized shi ed con-
s uc s (α alone and α + β1). Despi e his limi a ion, we
expec ha he hype pola ized shi ed cons uc s would
ha e mo e nega i e V1/2 i we we e able o eco d o
he baseline, sugges ing ha he DIII depola izing shi
induced by β3 is e en la ge han epo ed.
Despi e being highly homologous o β1, β3 is unique
in al e ing he VSD ansi ions o bo h DIII and DIV.
The DIV-VSD e ec s induced by β3 a e simila o hose
induced by β1, causing a depola izing SSI shi . The β3
e ec on he DIII-VSD, which shi s DIII-VSD’s ac i a-
ion o highe po en ials, slows ionic cu en ac i a ion
and inac i a ion kine ics. Two α–β3 in e ac ion mecha-
nisms could explain he changes in he VSD mo emen s
ha we obse ed. One possibili y is ha β3 can in e ac
wi h bo h DIII and DIV. A second plausible mechanism
is ha β3 mainly in e ac s wi h he DIII-VSD, which can
allos e ically modi y he adjacen DIV-VSD ac i a ion.
In he ollowing sec ions, esul s om β1/β3 chime a
and α–β3 quenche luo opho e pai expe imen s sup-
po he la e mechanism.
High exp ession o β3 sepa a es DIII VSD
ac i a ion in o wo s eps
To ensu e ha he VSD al e a ions we obse ed we e
uly caused by he exp ession o β subuni s and ha
he amoun o β subuni s exp essed on he memb ane
sa u a ed he modula ion e ec s o NaV1.5 channels,
we es ed di e en exp ession le els o β subuni s. We
al e ed β subuni exp ession le els by injec ing mRNAs
encoding α and β subuni s a di e en mola a ios, ob-
se ing hei e ec s on ionic cu en and VSD ac i a ion.
Fo he β1 subuni , as we inc eased he mRNA mola
a io om 1:1 o 1:2, he DIV F-V cu e shi ed o mo e
depola ized po en ials (1:1 α:β1: V1/2 = −70.1 ± 5.2 mV,
1:2 α:β1: V1/2 = −56.8 ± 5.0 mV). Fu he , when he α:β1
mRNA mola a io was inc eased o 1:4, he DIV F-V
cu e o e lapped wi h he F-V cu e o a 1:2 α:β1 mRNA
mola a io (Fig.3a), sugges ing ha β1 modula ion o
DIV-VSD sa u a ed a a 1:2 α:β1 a io. Consis en ly, he
β1 al e a ion o channel SSI ollowed a simila sa u a-
ion pa e n (Fig.3b). This esul u he suppo s he
idea ha β1 egula es channel inac i a ion by al e ing
DIV-VSD ac i a ion, an e ec ha sa u a es a a 1:2 a io.
Table1. Pa ame e s o Bol zmann i o G-V, ssI, and F-V cu es o W naV1.5 o VcF cons uc s exp essed wi h o wi hou
W β1 o β3
Pa ame e dI dI + β1 dI + β3 dII dII + β1 dII + β3
G-V
V1/2 −42.3 ± 1.7 −50.2 ± 1.0 −37.1 ± 1.8 −33.9 ± 2.2 −38.9 ± 3.0 −43.5 ± 2.2
k [n] 8.16 ± 0.7 [5] 8.3 ± 0.6 [4] 9.7 ± 0.5 [4] 9.5 ± 0.2 [4] 8.5 ± 1.2 [9] −7.0 ± 0.5 [4]
ssI
V1/2 −96.3 ± 4.55 −78.6 ± 2.1 −79.6 ± 2.4 −88.8 ± 1.3 −79.2 ± 4.5 −86.4 ± 0.7
k [n] −11.0 ± 1.1 [4] −6.4 ± 0.2 [8] −7.0 ± 0.4 [4] −8.2 ± 0.4 [4] −6.5 ± 0.9 [7] −6.1 ± 0.1 [4]
F-V
V1/2 −111.5 ± 1.0 −92.1 ± 11.4 −75.7 ± 4.6 −48.4 ± 2.7 −51.1 ± 3.5 −45.5 ± 5.0
k [n] 21.3 ± 2.4 [4] 18.6 ± 2.3 [4] 15.3 ± 2.5 [4] 19.4 ± 0.7 [4] 23.1 ± 2.2 [6] 19.3 ± 0.8 [4]
dIII dIII + β1 dIII + β3 dIV dIV + β1 dIV + β3
G-V
V1/2 −43.7 ± 1.9 −40.0 ± 4.4 −39.2 ± 1.4 −36.8 ± 1.6 −34.6 ± 3.4 −38.2 ± 1.3
k [n] 7.8 ± 0.6 [5] 9.4 ± 0.7 [13] 7.4 ± 0.5 [6] 8.9 ± 0.9 [4] 9.2 ± 0.6 [19] 7.2 ± 0.5 [5]
ssI
V1/2 −94.7 ± 1.9 −76.2 ± 2.5 −86.0 ± 1.6 −91.7 ± 3.4 −74.2 ± 3.1 −78.0 ± 1.6
k [n] −9.8 ± 0.7 [4] −6.7 ± 0.4 [7] −7.6 ± 0.4 [4] −12.6 ± 0.9 [5] −10.3 ± 0.9 [13] −9.2 ± 0.5 [5]
F-V
V1/2 −120.7 ± 4.8 −122.1 ± 1.4 −98.0 ± 2.6 −88.2 ± 5.3 −56.8 ± 6.6 −63.2 ± 4
k [n] 25.6 ± 0.7 [4] 24.3 ± 1.3 [5] 26.4 ± 2.0 [5] 23.6 ± 3.1 [4] 14.5 ± 2.5 [6] 14.4 ± 0.3 [4]
W αW α + β1 W α + β3
G-V
V1/2 −35.8 ± 1.4 −34.6 ± 1.9 −30.4 ± 1.9
k [n] 6.8 ± 0.2 [4] 7.2 ± 0.7 [5] 7.4 ± 0.3 [3]
ssI
V1/2 −84.8 ± 2.5 −74.0 ± 2.4 −75.2 ± 1.9
k [n] −8.9 ± 0.7 [5] −5.7 ± 0.3 [4] −6.2 ± 0.3 [3]
W lFs αW lFs α + β1 W lFs α + β3
Q-V
V1/2 −60.7 ± 6.9 −49.8 ± 3.1 −55.1 ± 2.3
k [n] 25.9 ± 2.8 [4] 17.3 ± 1.2 [3] 17.3 ± 2.5 [8]
on Janua y 31, 2018jgp. up ess.o gDownloaded om
Mechanisms o Na+ channel egula ion by β1 and β3 | Zhu e al.820
Figu e 3. high exp ession o β3 subuni s sepa a e dIII-Vsd mo emen s in o wo componen s. G oups o h ee o i e cells a e
epo ed as mean ± SEM. (a and b) The DIV VCF cons uc (α) was coinjec ed wi h β1 mRNA a mola a ios o 1:1, 1:2, and 1:4 o
wi hou β. (a) The DIV F-V cu e was cons uc ed o each mola a io o α:β1. The β1-induced depola izing shi o DIV F-V sa u a es
when mola a io o α:β1 eaches 1:2. (b) Channel s eady-s a e inac i a ion (SSI) cu es we e cons uc ed o he same mola a ios o
α and β1. The depola izing shi o he SSI cu e caused by β1 sa u a es when he mola a io o α:β1 eaches 1:2. (c and d) The DIII
VCF cons uc (α) was coinjec ed wi h β3 mRNA a mola a ios o 1:1, 1:2, 1:4, and 1:6 o wi hou β. (c) DIII F-V was cons uc ed o
each mola a io o α:β3. When α:β3 is g ea e han 1:4, DIII F-V s a s o exhibi wo ac i a ion s eps. (d) Channel SSI cu es we e
on Janua y 31, 2018jgp. up ess.o gDownloaded om
821JGP Vol. 149, No. 8
Fo β3, he sa u a ion beha io was mo e complex.
When we inc eased he mRNA mola a io om 1:1 o
1:2, he DIII F-V cu e shi ed o depola ized po en ials
(1:1 α:β3: V1/2 = −108.6 ± 5.6 mV; 1:2 α:β3: V1/2 = −97.9
± 2.6 mV; Fig.3c). In iguingly, when he mola a io o
α:β3 was inc eased o 1:4 o highe , he DIII F-V cu e
s a ed o exhibi wo componen s ha could no lon-
ge be i wi h a single Bol zmann unc ion (Fig.3c).
The cu e was well i wi h wo Bol zmann cu es, one
a e y nega i e po en ials (−180 o −80 mV), and he
o he wi hin he channel ac i a ion ol age ange (−80
o 20 mV). Co espondingly, he DIII luo escence ki-
ne ics also ollowed wo s eps, which also we e no well
i wi h a single exponen ial. The i s apid ansi ion
occu ed wi hin 2 ms a e depola iza ion, deno ed by
F1, ollowed by a e y slow componen , deno ed by F2,
o e a ime pe iod o 60 ms (Fig.3g).
The sepa a ion o wo componen s in DIII-VSD ac i a-
ion caused unusual eco e y om inac i a ion, whe e
he peak cu en magni ude du ing he es pulse was
la ge han ha du ing he con ol pulse a e eco e y
imes o 10 o 300 ms (Fig.3, e and ). A e 500- o
1,000-ms eco e y a −120 mV, he peak cu en du ing
he es pulse e u ned o he magni ude o he con ol
pulse. Typically, Na+ cu en s exhibi mono onic beha -
io du ing his p o ocol.
We assessed he ela ionship be ween he wo DIII-
VSD ac i a ion componen s and channel eco e y om
inac i a ion by aligning cu en du ing he i s con-
ol pulse wi h ha o he second es pulse a e 10-
o 1,000-ms eco e y (Fig.3i, bo om le , op igh ).
Fo channels exp essed wi hou he β subuni , cu en
du ing he es pulse a e 10-ms eco e y ac i a ed a
he same a e as he con ol pulse (Fig.3i, op le ).
Fo channels coexp essed wi h he β3 subuni a a 1:4
mola a io, cu en du ing he es pulse a e 10-ms
eco e y ac i a ed mo e quickly in compa ison wi h he
con ol pulse (Fig.3i, bo om le ). In con as , he cu -
en du ing he es pulse a e 1,000-ms eco e y ose
a he same a e as ha o he con ol pulse (Fig.3i,
op igh ). Fas e channel ac i a ion kine ics can signi i-
can ly inc ease peak cu en . High β3 exp ession causes
as e ac i a ion kine ics du ing a es pulse ha ollows
a sho eco e y in e al (10–300 ms), esul ing in peak
cu en ha exceeds he con ol pulse cu en .
To accoun o his beha io , we i s suppose ha
he DIII-VSD ac i a es in wo s eps om es ing (R) o
in e media e ac i a ed (A1) and ac i a ed (A2) s a es
(Fig. 3 h). We hen assume ha he ansi ion om
R o A1 is as and is desc ibed by he F1 componen ,
whe eas he ansi ion om A1 o A2 is slow, as shown
by he F2 componen . In his model, po e opening is
acili a ed by he ansi ion o he DIII-VSD o he A1
s a e and u he encou aged by en y in o he A2
s a e. Du ing he i s 200-ms pulse, mos DIII-VSDs a e
b ough o A2. Because he ansi ion om A2 o A1 is
slow, when channels we e gi en 10–300 ms o eco e
a −120 mV, he ime was oo sho o DIII VSD o e-
co e o A1, esul ing in mos o he DIII-VSDs being
apped in he A2 s a e. As mos o he DIII-VSDs we e
s ill in he A2 s a e and i g ea ly acili a es po e open-
ing, channel ac i a ion was as e o he second pulse.
I his scheme is co ec , we would p edic ha i we
only allow he DIII-VSD o en e A1 by applying a sho
2-ms depola izing pulse as he con ol pulse, he second
pulse will no ha e as e ising kine ics compa ed wi h
he con ol pulse. Indeed, compa ison o he con ol
and es pulses shows ha bo h pulses comple ely o e -
lap (Fig.3i, bo om igh ).
I is unlikely ha he physiological assembly o α–β3
will each a a io high enough o sepa a e DIII-VSD
mo emen in o wo componen s (Yuan e al., 2014). I is
possible ha o e exp ession o β3 will o ce some o he
β3 subuni s in o a seconda y low-a ini y binding si e.
Consequen ly, i β3 is locally exp essed a e y high le -
els, his g oup o cells will ha e a ela i ely sho e ac-
cons uc ed o he same mola a ios o α and β3. β3-induced SSI depola iza ion sa u a es a a 1:2 α:β3 mola a io. (e) Compa ison
be ween channel eco e y om inac i a ion cu es o α:β3 a 1:2 and 1:6. To assess he ime dependence o channel eco e y
om inac i a ion om a holding po en ial o −120 mV, channels we e s epped and held a −20 mV o 200 ms, s epped back o
−120 mV o a di e en amoun o ime (2–1,000 ms), and hen s epped up o 20 mV o es a ailabili y. ( ) Rep esen a i e channel
eco e y om inac i a ion cu en aces a a 1:6 α:β3 mola a io. The do ed line ep esen s he i s pulse peak ampli ude. (g) DIII
luo escence ace a a 1:6 α:β3 mola a io in esponse o 0 mV depola izing po en ial. The luo escence shown is a mean o aces
om h ee cells. In pa allel o DIII luo escence ol age dependence, he DIII luo escence ac i a ion kine ics also display wo com-
ponen s: F1 and F2. (h) Schema ic model o DIII VSD mo emen s when channels a e coassembled wi h high exp ession le el o β3.
(i) Compa ison o he cu en ac i a ion kine ics shows ha wi h high exp ession o he β3 subuni , he second cu en pulse a e a
sho eco e y ime has as e ac i a ion kine ics. (i, op le ) Compa ing he i s and second cu en aces e oked by depola izing
pulse a −20 mV wi h a 10-ms eco e y ime be ween hese wo pulses o channels exp essed wi hou he β subuni . The second
pulse has he same ise ime bu as e inac i a ion. (i, bo om le ) Channels we e o e exp essed wi h he β3 subuni a a 1:4 α:β3
mola a io. Compa ison o he i s and second cu en aces e oked by depola izing pulse a −20 mV wi h a 10-ms eco e y ime
be ween hese wo pulses o channels o e exp essed wi h he β3 subuni . The second cu en pulse has as e ac i a ion kine ics in
con as o α alone. (i, op igh ) Channels we e o e exp essed wi h he β3 subuni a a 1:4 α:β3 mola a io. Compa ison o he i s
and second cu en aces e oked by depola izing pulse a −20 mV wi h a 1,000-ms eco e y ime in be ween. The i s and second
pulses ha e simila ac i a ion kine ics. (i, bo om igh ) Fo channels o e exp essed wi h he β3 subuni a a 1:4 α:β3 mola a io,
he i s ol age pulse was sho ened o 2 ms. Compa ing he i s and second pulses a e 10 ms o eco e y, he ac i a ion kine ics
emain he same.
on Janua y 31, 2018jgp. up ess.o gDownloaded om
Mechanisms o Na+ channel egula ion by β1 and β3 | Zhu e al.828
e ac ion be ween β1 and o he membe s o he mac-
omolecula NaV channel complex in na i e cells, such
as anky in G (Malho a e al., 2002). Thus, he na u e
o α–β1 in e ac ion may a y by exp ession sys em, p e-
cluding he iden i ica ion o a uni e sal pheno ype.
Se e al α–β1 in e ac ion si es ha e been p oposed. On
he α subuni , a C- e minal mu a ion was able o elim-
ina e β1 egula ion o NaV1.1 cu en kine ics (Spam-
pana o e al., 2004), and NaV1.4/NaV1.5 chime as show
ha he S5–S6 linke o DIV plays a ole in α–β1 in e -
ac ion (Maki a e al., 1996). Bo h esul s sugges ha
binding occu s nea he DIV domain. Aside om he
consequences o di ec binding, he β1 subuni has also
been shown o in oduce he su ace cha ges ha elec-
os a ically a ec channel ga ing (Fe e a and Mo an,
2006). We obse ed ha WT β1 mainly a ec s he ol -
age dependence o DIV-VSD ac i a ion and i s deac i-
a ion kine ics h ough possible di ec in e ac ion wi h
he DIV-VSD, which sugges s β1 p oximi y o he DIV-
VSD. Toge he , hese esul s suppo he hypo hesis
ha β1 modula es inac i a ion by al e ing DIV-VSD ac-
i a ion. We in e ha β1 mos likely esides in he cle
be ween he DIII and DIV VSDs (Fig.8, a and b).
The β3 subuni is homologous o β1 and also in e -
ac s wi h NaV channels nonco alen ly. Less is known
abou he α–β3 in e ac ion. Because he β1 and β3
subuni s a e homologous (50%), i has been gene ally
supposed ha β3 in e ac s wi h he channel ia he
same mechanism as β1 (Namadu ai e al., 2015). In
con as , we ound ha he WT β3 subuni caused a
la ge depola izing shi in DIII-VSD ac i a ion in addi-
ion o he DIV shi . The depola iza ion o DIII-VSD
ac i a ion slowed down ionic cu en ac i a ion and
inac i a ion kine ics, allowing he DIII-VSD o play a
mo e p ominen ole in egula ing his ga ing o e a
physiological ange o po en ials. Fu he , we demon-
s a ed β3 p oximi y o he DIII S4 segmen , as we ob-
se ed ha a yp ophan mu a ion on op o he β3
subuni s ongly quenches he luo opho e a ached
o he DIII S3–S4 linke . I is mos likely ha β3 is ad-
jacen o he DIII S4 segmen in he cle be ween he
DII and DIII VSDs, which allows β3 o di ec ly in e -
ac wi h he DIII-VSD. Al e ed DIII-VSD ac i a ion can
hen allos e ically a ec DIV-VSD ac i a ion. Ye , we
obse ed ha β3 s ill depola ized he DIV-VSD when
he DIV VSD–po e coupling ia he S4–S5 linke was
abolished by he N1759A mu a ion, sugges ing ha
his coupling o he DIV VSD akes place ia al e na-
i e mechanisms, such as he DIII–DIV linke .
The impo ance o β3 in main aining no mal ca diac
unc ion has been highligh ed by
scn3b
knockou mice.
These mice exhibi slowed sinoa ial and a io en ic-
ula conduc ion, bu s pacing–induced a ial achyca -
dia, ib illa ion, and en icula achyca dia (Hakim e
al., 2008, 2010). Consis en wi h ou esul s (Fig. 2),
knocking ou β3 shi s NaV SSI o nega i e po en ials, e-
ducing peak Na+ cu en in he en icle, which causes
slowed conduc ion and dec eased ac ion po en ial du-
a ion in he endoca dium and epica dium. Gi en he
consis ency o he knockou mouse pheno ype wi h ou
esul s, we in e ha β3 egula ion o he NaV1.5 DIII
Figu e 8. P oposed model o β1 and β3 assembly wi h
naV1.5 channel. (a) Ex acellula iew o he NaV1.5 channel
based on he NaVab s uc u e (Payandeh e al., 2011). Each do-
main is colo -coded as shown in Figs. 1, 2, 3, 4, 5, and 6. The β1
and β3 loca ions sugges ed by ou esul s a e shown. (b) Side
iew o NaV1.5 coassembled wi h β1 subuni . Only DIII (blue)
and DIV (pink) a e shown o cla i y. Ou model sugges s ha β1
is loca ed in he cle be ween he DIII VSD and DIV VSD, allow-
ing i o in e ac wi h he DIV VSD, he C e minus, and po en-
ially he S4–S5 linke o DIV o modi y DIV VSD mo emen s. (c)
Side iew o NaV1.5 ( o a ed 90°) coassembled wi h β3 subuni .
Ou model sugges s ha β3 is loca ed in he cle be ween he
DIII VSD and DII VSD, nex o he S4 segmen o DIII, allowing
i o s ongly modi y DIII VSD ac i a ion and a ec ing DIII VSD–
po e coupling by in e ac ing wi h he hinge connec ing he S4
and S4–S5 linke o DIII.
on Janua y 31, 2018jgp. up ess.o gDownloaded om

829JGP Vol. 149, No. 8
and DIV VSDs signi ican ly de e mines ac ion po en ial
mo phology and conduc ion.
When we exp essed β1 and β3 oge he , we obse ed
enhancemen o he depola izing DIII-VSD shi and an
exclusi e e ec o β1 on he DIV-VSD (Fig.6). These
esul s imply ha he subuni s do no in e ac wi h he
channel independen ly. P e ious wo k has shown ha
he e ophilic in e ac ion be ween he β1 and β3 sub-
uni s can occu ia hei espec i e Ig domains (Ye eddi
e al., 2013), which could possibly al e hei in e ac ion
wi h each VSD. β3 binding o β1 may a ec i s in e -
ac ion wi h he DIII-VSD. Ou p oposed localiza ion
o β1 and β3 would b ing he ex acellula domains
o bo h subuni s close o he DIII S4 (Fig.8). In sum,
ou esul s show ha bo h he β1 and β3 subuni s a e
likely able o coassemble wi h a single NaV1.5 channel
complex and ha his coassembly signi ican ly a ec s
channel unc ion.
Each domain o β1 and β3 has dis inc in e ac ions wi h
he NaV1.5 channel
By measu ing VSD con o ma ion in he p esence o
β1/β3 chime as, we showed ha bo h he ex acellula
and ansmemb ane domains o β3 a e necessa y o β3
depola iza ion o he DIII-VSD. When channels we e
coexp essed wi h β1/β3 chime as con aining he β3
ex acellula domain and he β1 ansmemb ane and
in acellula domain, DIII-VSD ac i a ion was no depo-
la ized, sugges ing ha he ansmemb ane domain o
β3 is c i ical o localizing he β3 subuni o his loca ion
in NaV1.5, allowing he ex acellula Ig domain o β3
o in e ac wi h he DIII-VSD. We also showed ha he
C e minus o he β1 subuni is necessa y o elie ing
DIV-VSD immobiliza ion and ha he β3 C e minus is
impo an o elie ing DIII-VSD immobiliza ion om
as inac i a ion (Fig. S5, c and e). P e iously, we showed
ha he in e ac ion be ween he as inac i a ion ga e
(IFM) and he N1659 esidue on he DIV S4–S5 linke
plays an impo an ole in immobilizing he DIII and
DIV VSD. I is plausible ha he β1 C e minus can in-
e ac wi h he in acellula DIV S5 segmen , al e ing
his in e ac ion. These esul s show ha β1 and β3 could
in e ac wi h he NaV channel h ough mul iple in e ac-
ion si es and ha bo h he ex acellula domain and
he C e minus play impo an oles in de e mining
hei ga ing p ope ies.
Two-s ep mo emen s o DIII VSD e ealed by high
exp ession o β3
As wi h KV channels, wo-s ep ansi ions o NaV channel
VSDs ha e been p oposed based on p e ious obse a-
ions ha pulses o inc easing du a ion al e he DIII-VSD
deac i a ion a e (Va ga e al., 2015; Hsu e al., 2017).
Ou esul s suppo his sugges ion by showing ha high
le els o β3 subuni exp ession cause wo p ominen
componen s o DIII VSD ac i a ion. This phenomenon is
simila o KCNQ1 channels ha show wo-s ep ac i a ion
in he p esence o KCNE1, which s abilizes he ac i a ed
VSD closed-po e s a e and slows ionic cu en ac i a ion
(Ba o-So ia e al., 2014). Simila ly, he ac i a ion o Na+
cu en is slowe when β3 sepa a es he DIII-VSD an-
si ion in o wo componen s. Addi ionally, because NaV
channels ha e e y as and p ominen inac i a ion, he
sepa a ion o DIII-VSD mo emen will inc ease he am-
pli ude o he Na+ cu en as he channels a e exci ed
epe i i ely a a ela i ely high equency (>3 Hz).
Ou s udy shows ha β1 and β3’s di e en ial in e ac-
ions wi h he DIII and DIV VSDs de e mine hei egu-
la ion o Na+ cu en and cell exci abili y. These dis inc
egula o y mechanisms a e essen ial o unde s anding
how β subuni s egula e exci able cells and how mu an
β subuni s cause disease.
AcknoWledGMen s
We hank D . S acey Ren schle , D . Hen y Colec a , James Bal-
la d, D . Panpan Hou, and D . Roman Slou sky o many help ul
discussions and ad ice.
This s udy was unded by he e Fund Ca ee Awa d om he
Scien i ic In e ace (g an 1010299); he Na ional Ins i u es o
Heal h (g an R01 HL136553 awa ded o J.R. Sil a); an Ame ican
Hea Associa ion ellowship (g an 15PRE25080073 awa ded o
W. Zhu); he Hunga ian Academy o Sciences g an KTIA_
NAP_13-2-2015-0009; a Bolyai ellowship (awa ded o Z. Va ga);
and he Na ional Hea , Lung, and Blood Ins i u e o he Na-
ional Ins i u es o Heal h (g an HL-034161 awa ded o
J.M. Ne bonne).
The au ho s decla e no compe ing inancial in e es s.
Richa d W. Ald ich se ed as edi o .
Submi ed: 7 Ap il 2017
Accep ed: 26 June 2017
ReFeRences
Ab iel, H. 2010. Ca diac sodium channel Na 1.5 and in e ac ing
p o eins: Physiology and pa hophysiology. J. Mol. Cell. Ca diol.
48:2–11. h p ://dx .doi .o g /10 .1016 /j .yjmcc .2009 .08 .025
Ald ich, R.W., D.P. Co ey, and C.F. S e ens. 1983. A ein e p e a ion
o mammalian sodium channel ga ing based on single channel
eco ding. Na u e. 306:436–441. h p ://dx .doi .o g /10 .1038
/306436a0
An, R.H., X.L. Wang, B. Ke em, J. Benho in, A. Medina, M. Goldmi ,
and R.S. Kass. 1998. No el LQT-3 mu a ion a ec s Na+ channel
ac i i y h ough in e ac ions be ween α- and β1-subuni s. Ci c. Res.
83:141–146. h p ://dx .doi .o g /10 .1161 /01 .RES .83 .2 .141
A cisio-Mi anda, M., Y. Mu oi, S. Chowdhu y, and B. Chanda.
2010. Molecula mechanism o allos e ic modi ica ion o ol age-
dependen sodium channels by local anes he ics. J. Gen. Physiol.
136:541–554. h p ://dx .doi .o g /10 .1085 /jgp .201010438
A ms ong, C.M., and F. Bezanilla. 1977. Inac i a ion o he sodium
channel. II. Ga ing cu en expe imen s. J. Gen. Physiol. 70:567–
590. h p ://dx .doi .o g /10 .1085 /jgp .70 .5 .567
Ba o-So ia, R., S. Rebolledo, S.I. Liin, M.E. Pe ez, K.J. Sampson,
R.S. Kass, and H.P. La sson. 2014. KCNE1 di ides he ol age
senso mo emen in KCNQ1/KCNE1 channels in o wo s eps.
Na . Commun. 5:3750. h p ://dx .doi .o g /10 .1038 /ncomms4750
Calhoun, J.D., and L.L. Isom. 2014. The ole o non-po e- o ming
β subuni s in physiology and pa hophysiology o ol age-ga ed
on Janua y 31, 2018jgp. up ess.o gDownloaded om
Mechanisms o Na+ channel egula ion by β1 and β3 | Zhu e al.830
sodium channels. Handb. Exp. Pha macol. 221:51–89. h p ://dx
.doi .o g /10 .1007 /978 -3 -642 -41588 -3 _4
Campos, F.V., B. Chanda, P.S.L. Bei ão, and F. Bezanilla. 2007.
β-Sco pion oxin modi ies ga ing ansi ions in all ou ol age
senso s o he sodium channel. J. Gen. Physiol. 130:257–268. h p
://dx .doi .o g /10 .1085 /jgp .200609719
Campos, F.V., B. Chanda, P.S.L. Bei ão, and F. Bezanilla. 2008.
α-Sco pion oxin impai s a con o ma ional change ha leads
o as inac i a ion o muscle sodium channels. J. Gen. Physiol.
132:251–263. h p ://dx .doi .o g /10 .1085 /jgp .200809995
Capes, D.L., M.P. Goldschen-Ohm, M. A cisio-Mi anda, F. Bezanilla,
and B. Chanda. 2013. Domain IV ol age-senso mo emen is
bo h su icien and a e limi ing o as inac i a ion in sodium
channels. J. Gen. Physiol. 142:101–112. h p ://dx .doi .o g /10 .1085
/jgp .201310998
Cha, A., P.C. Ruben, A.L. Geo ge J ., E. Fujimo o, and F. Bezanilla.
1999. Vol age senso s in domains III and IV, bu no I and II, a e
immobilized by Na+ channel as inac i a ion. Neu on. 22:73–87.
h p ://dx .doi .o g /10 .1016 /S0896 -6273(00)80680 -7
Chanda, B., and F. Bezanilla. 2002. T acking ol age-dependen
con o ma ional changes in skele al muscle sodium channel
du ing ac i a ion. J. Gen. Physiol. 120:629–645. h p ://dx .doi .o g
/10 .1085 /jgp .20028679
Das, S., J. Gilch is , F. Bosmans, and F. Van Pe egem. 2016. Bina y
a chi ec u e o he Na 1.2-β2 signaling complex. eLi e. 5:1–21.
h p ://dx .doi .o g /10 .7554 /eLi e .10960
Domínguez, J.N., F. Na a o, D. F anco, R.P. Thompson, and A.E.
A ánega. 2005. Tempo al and spa ial exp ession pa e n o β1
sodium channel subuni du ing hea de elopmen . Ca dio asc.
Res. 65:842–850. h p ://dx .doi .o g /10 .1016 /j .ca dio es .2004 .11
.028
Fahmi, A.I., M. Pa el, E.B. S e ens, A.L. Fowden, J.E. John III, K. Lee,
R. Pinnock, K. Mo gan, A.P. Jackson, and J.I. Vandenbe g. 2001.
The sodium channel β-subuni SCN3b modula es he kine ics o
SCN5a and is exp essed he e ogeneously in sheep hea . J. Physiol.
537:693–700. h p ://dx .doi .o g /10 .1113 /jphysiol .2001 .012691
Fe e a, L., and O. Mo an. 2006. β1-subuni modula es he Na 1.4
sodium channel by changing he su ace cha ge. Exp. B ain Res.
172:139–150. h p ://dx .doi .o g /10 .1007 /s00221 -005 -0323 -4
Gellens, M.E., A.L. Geo ge J ., L.Q. Chen, M. Chahine, R. Ho n, R.L.
Ba chi, and R.G. Kallen. 1992. P ima y s uc u e and unc ional
exp ession o he human ca diac e odo oxin-insensi i e ol age-
dependen sodium channel. P oc. Na l. Acad. Sci. USA. 89:554–
558. h p ://dx .doi .o g /10 .1073 /pnas .89 .2 .554
Gilch is , J., S. Das, F. Van Pe egem, and F. Bosmans. 2013.
C ys allog aphic insigh s in o sodium-channel modula ion by he
β4 subuni . P oc. Na l. Acad. Sci. USA. 110:E5016–E5024. h p ://
dx .doi .o g /10 .1073 /pnas .1314557110
Hakim, P., I.S. Gu ung, T.H. Pede sen, R. Th eshe , N. B ice, J.
Law ence, A.A. G ace, and C.L. Huang. 2008.
Scn3b
knockou
mice exhibi abno mal en icula elec ophysiological
p ope ies. P og. Biophys. Mol. Biol. 98:251–266. h p ://dx .doi .o g
/10 .1016 /j .pbiomolbio .2009 .01 .005
Hakim, P., N. B ice, R. Th eshe , J. Law ence, Y. Zhang, A.P.
Jackson, A.A. G ace, and C.L. Huang. 2010.
Scn3b
knockou mice
exhibi abno mal sino-a ial and ca diac conduc ion p ope ies.
Ac a Physiol. (Ox .). 198:47–59. h p ://dx .doi .o g /10 .1111 /j .1748
-1716 .2009 .02048 .x
Ha sho ne, R.P., and W.A. Ca e all. 1984. The sodium channel
om a b ain. Pu i ica ion and subuni composi ion. J. Biol.
Chem. 259:1667–1675.
Hsu, E.J., W. Zhu, A.R. Schube , T. Voelke , Z. Va ga, and J.R. Sil a.
2017. Regula ion o Na+ channel inac i a ion by he DIII and DIV
ol age-sensing domains. J. Gen. Physiol. 149:389–403. h p ://dx
.doi .o g /10 .1085 /jgp .201611678
Hu, D., H. Ba ajas-Ma ínez, A. Medei os-Domingo, L. C o i, C.
Vel mann, R. Schimp , J. U u ia, A. Alday, O. Casis, R. P ei e ,
e al. 2012. A no el a e a ian in SCN1Bb linked o B ugada
synd ome and SIDS by combined modula ion o Na 1.5 and K 4.3
channel cu en s. Hea Rhy hm. 9:760–769. h p ://dx .doi .o g /10
.1016 /j .h hm .2011 .12 .006
Isom, L.L., K.S. De Jongh, D.E. Pa on, B.F. Rebe , J. O o d, H.
Cha bonneau, K. Walsh, A.L. Goldin, and W.A. Ca e all. 1992.
P ima y s uc u e and unc ional exp ession o he β 1 subuni o
he a b ain sodium channel. Science. 256:839–842. h p ://dx .doi
.o g /10 .1126 /science .1375395
Isom, L.L., D.S. Ragsdale, K.S. De Jongh, R.E. Wes enb oek,
B.F.X. Rebe , T. Scheue , and W.A. Ca e all. 1995. S uc u e
and unc ion o he β 2 subuni o b ain sodium channels, a
ansmemb ane glycop o ein wi h a CAM mo i . Cell. 83:433–442.
h p ://dx .doi .o g /10 .1016 /0092 -8674(95)90121 -3
Kazen-Gillespie, K.A., D.S. Ragsdale, M.R. D’And ea, L.N. Ma ei,
K.E. Roge s, and L.L. Isom. 2000. Cloning, localiza ion, and
unc ional exp ession o sodium channel β1A subuni s. J. Biol.
Chem. 275:1079–1088. h p ://dx .doi .o g /10 .1074 /jbc .275 .2 .1079
Lenkowski, P.W., B.S. Shah, A.E. Dinn, K. Lee, and M.K. Pa el. 2003.
Lidocaine block o neona al Na 1.3 is di e en ially modula ed by
co-exp ession o β1 and β3 subuni s. Eu . J. Pha macol. 467:23–30.
h p ://dx .doi .o g /10 .1016 /S0014 -2999(03)01595 -4
Maki a, N., P.B. Benne , and A.L. Geo ge J . 1996. Molecula de e -
minan s o β 1 subuni -induced ga ing modula ion in ol age-de-
penden Na+ channels. J. Neu osci. 16:7117–7127.
Malho a, J.D., K. Kazen-Gillespie, M. Ho sch, and L.L. Isom. 2000.
Sodium channel β subuni s media e homophilic cell adhesion
and ec ui anky in o poin s o cell-cell con ac . J. Biol. Chem.
275:11383–11388. h p ://dx .doi .o g /10 .1074 /jbc .275 .15 .11383
Malho a, J.D., C. Chen, I. Ri ol a, H. Ab iel, R. Malho a,
L.N. Ma ei, F.C. B osius, R.S. Kass, and L.L. Isom. 2001.
Cha ac e iza ion o sodium channel α- and β-subuni s in a and
mouse ca diac myocy es. Ci cula ion. 103:1303–1310. h p ://dx
.doi .o g /10 .1161 /01 .CIR .103 .9 .1303
Malho a, J.D., M.C. Koopmann, K.A. Kazen-Gillespie, N. Fe man,
M. Ho sch, and L.L. Isom. 2002. S uc u al equi emen s o
in e ac ion o sodium channel β 1 subuni s wi h anky in. J.
Biol. Chem. 277:26681–26688. h p ://dx .doi .o g /10 .1074 /jbc
.M202354200
Mal se , V.A., J.W. Kyle, and A. Und o inas. 2009. La e Na+ cu en
p oduced by human ca diac Na+ channel iso o m Na 1.5 is
modula ed by i s β1 subuni . J. Physiol. Sci. 59:217–225. h p ://dx
.doi .o g /10 .1007 /s12576 -009 -0029 -7
Mansoo , S.E., H.S. McHaou ab, and D.L. Fa ens. 2002. Mapping
p oximi y wi hin p o eins using luo escence spec oscopy. A
s udy o T4 lysozyme showing ha yp ophan esidues quench
bimane luo escence. Biochemis y. 41:2475–2484. h p ://dx .doi
.o g /10 .1021 /bi011198i
Mansoo , S.E., M.A. Dewi , and D.L. Fa ens. 2010. Dis ance
mapping in p o eins using luo escence spec oscopy: The
yp ophan-induced quenching (T IQ) me hod. Biochemis y.
49:9722–9731. h p ://dx .doi .o g /10 .1021 /bi100907m
Messne , D.J., and W.A. Ca e all. 1985. The sodium channel om
a b ain. Sepa a ion and cha ac e iza ion o subuni s. J. Biol.
Chem. 260:10597–10604.
Mo gan, K., E.B. S e ens, B. Shah, P.J. Cox, A.K. Dixon, K. Lee, R.D.
Pinnock, J. Hughes, P.J. Richa dson, K. Mizuguchi, e al. 2000.
β3: An addi ional auxilia y subuni o he ol age-sensi i e sodium
channel ha modula es channel ga ing wi h dis inc kine ics.
P oc. Na l. Acad. Sci. USA. 97:2308–2313. h p ://dx .doi .o g /10
.1073 /pnas .030362197
Mu oi, Y., and B. Chanda. 2009. Local anes he ics dis up ene ge ic
coupling be ween he ol age-sensing segmen s o a sodium
on Janua y 31, 2018jgp. up ess.o gDownloaded om
831JGP Vol. 149, No. 8
channel. J. Gen. Physiol. 133:1–15. h p ://dx .doi .o g /10 .1085 /jgp
.200810103
Mu oi, Y., M. A cisio-Mi anda, S. Chowdhu y, and B. Chanda. 2010.
Molecula de e minan s o coupling be ween he domain III
ol age senso and po e o a sodium channel. Na . S uc . Mol.
Biol. 17:230–237. h p ://dx .doi .o g /10 .1038 /nsmb .1749
Namadu ai, S., N.R. Ye eddi, F.S. Cusdin, C.L.H. Huang, D.Y.
Chi gadze, and A.P. Jackson. 2015. A new look a sodium channel
β subuni s. Open Biol. 5:140192. h p ://dx .doi .o g /10 .1098 / sob
.140192
Oka a, S., S. Yuasa, T. Suzuki, S. I o, N. Maki a, T. Yoshida, M. Li,
J. Ku okawa, T. Seki, T. Egashi a, e al. 2016. Emb yonic ype
Na+ channel β-subuni ,
SCN3B
masks he disease pheno ype o
B ugada synd ome. Sci. Rep. 6:34198. h p ://dx .doi .o g /10 .1038
/s ep34198
Olesen, M.S., T. Jespe sen, J.B. Nielsen, B. Liang, D.V. Molle , P.
Hedley, M. Ch is iansen, A. Va ó, S.P. Olesen, S. Haunsø, e
al. 2011. Mu a ions in sodium channel β-subuni SCN3B a e
associa ed wi h ea ly-onse lone a ial ib illa ion. Ca dio asc. Res.
89:786–793. h p ://dx .doi .o g /10 .1093 /c /c q348
Pan azis, A., and R. Olcese. 2012. Rela i e ansmemb ane segmen
ea angemen s du ing BK channel ac i a ion esol ed by
s uc u ally assigned luo opho e-quenche pai ing. J. Gen.
Physiol. 140:207–218. h p ://dx .doi .o g /10 .1085 /jgp .201210807
Pa ino, G.A., W.J. B ackenbu y, Y. Bao, L.F. Lopez-San iago, H.A.
O’Malley, C. Chen, J.D. Calhoun, R.G. La eniè e, P. Cosse e,
G.A. Rouleau, e al. 2011. Vol age-ga ed Na+ channel β1B: A
sec e ed cell adhesion molecule in ol ed in human epilepsy. J.
Neu osci. 31:14577–14591. h p ://dx .doi .o g /10 .1523 /JNE URO
SCI .0361 -11 .2011
Payandeh, J., T. Scheue , N. Zheng, and W.A. Ca e all. 2011. The
c ys al s uc u e o a ol age-ga ed sodium channel. Na u e.
475:353–358. h p ://dx .doi .o g /10 .1038 /na u e10238
Rudokas, M.W., Z. Va ga, A.R. Schube , A.B. Asa o, and J.R. Sil a.
2014. The
Xenopus
oocy e cu -open aseline gap ol age-clamp
echnique wi h luo ome y. J. Vis. Exp. 85:1–11. h p ://dx .doi
.o g /10 .3791 /51040
Shee s, M.F., and D.A. Hanck. 2005. Cha ge immobiliza ion o he
ol age senso in domain IV is independen o sodium cu en
inac i a ion. J. Physiol. 563:83–93. h p ://dx .doi .o g /10 .1113 /
jphysiol .2004 .077644
Shee s, M.F., H.A. Fozza d, and D.A. Hanck. 2015. Impo an ole
o aspa agines in coupling he po e and ol age-senso domain
in ol age-ga ed sodium channels. Biophys. J. 109:2277–2286. h p
://dx .doi .o g /10 .1016 /j .bpj .2015 .10 .012
Shen, H., Q. Zhou, X. Pan, Z. Li, J. Wu, and N. Yan. 2017. S uc u e
o a euka yo ic ol age-ga ed sodium channel a nea -a omic
esolu ion. Science. 355:eaal4326. h p ://dx .doi .o g /10 .1126 /
science .aal4326
Sil a, J.R., and S.A. Golds ein. 2013a. Vol age-senso mo emen s
desc ibe slow inac i a ion o ol age-ga ed sodium channels I:
Wild- ype skele al muscle NaV1.4. J. Gen. Physiol. 141:309–321.
h p ://dx .doi .o g /10 .1085 /jgp .201210909
Sil a, J.R., and S.A. Golds ein. 2013b. Vol age-senso mo emen s
desc ibe slow inac i a ion o ol age-ga ed sodium channels II:
A pe iodic pa alysis mu a ion in NaV1.4 (L689I). J. Gen. Physiol.
141:323–334. h p ://dx .doi .o g /10 .1085 /jgp .201210910
Spampana o, J., J.A. Kea ney, G. de Haan, D.P. McEwen, A. Escayg,
I. A adi, B.T. MacDonald, S.I. Le in, I. Sol esz, P. Benna, e al.
2004. A no el epilepsy mu a ion in he sodium channel
SCN1A
iden i ies a cy oplasmic domain o β subuni in e ac ion. J.
Neu osci. 24:10022–10034. h p ://dx .doi .o g /10 .1523 /JNE URO
SCI .2034 -04 .2004
S e ani, E., and F. Bezanilla. 1998. Cu -open oocy e ol age-clamp
echnique. Me hods Enzymol. 293:300–318. h p ://dx .doi .o g /10
.1016 /S0076 -6879(98)93020 -8
Uebachs, M., T. Opi z, M. Royeck, G. Dickho , M.-T. Ho s mann,
L.L. Isom, and H. Beck. 2010. E icacy loss o he an icon ulsan
ca bamazepine in mice lacking sodium channel β subuni s ia
pa adoxical e ec s on pe sis en sodium cu en s. J. Neu osci.
30:8489–8501. h p ://dx .doi .o g /10 .1523 /JNE URO SCI .1534 -10
.2010
Va ga, Z., W. Zhu, A.R. Schube , J.L. Pa dieck, A. K umholz,
E.J. Hsu, M.A. Zaydman, J. Cui, and J.R. Sil a. 2015. Di ec
measu emen o ca diac Na+ channel con o ma ions e eals
molecula pa hologies o inhe i ed mu a ions. Ci c A hy hm
Elec ophysiol. 8:1228–1239. h p ://dx .doi .o g /10 .1161 /CIR CEP
.115 .003155
Wang, H.-G., W. Zhu, R.J. Kan e , J.R. Sil a, C. Honeywell, R.M.
Gow, and G.S. Pi . 2016. A no el NaV1.5 ol age senso mu a ion
associa ed wi h se e e a ial and en icula a hy hmias. J. Mol.
Cell. Ca diol. 92:52–62. h p ://dx .doi .o g /10 .1016 /j .yjmcc .2016
.01 .014
Wa anabe, H., D. Da ba , D.W. Kaise , K. Ji amongkolchai, S.
Chop a, B.S. Donahue, P.J. Kannanke il, and D.M. Roden. 2009.
Mu a ions in sodium channel β1- and β2-subuni s associa ed wi h
a ial ib illa ion. Ci c A hy hm Elec ophysiol. 2:268–275. h p ://
dx .doi .o g /10 .1161 /CIR CEP .108 .779181
Wes , J.W., D.E. Pa on, T. Scheue , Y. Wang, A.L. Goldin, and W.A.
Ca e all. 1992. A clus e o hyd ophobic amino acid esidues
equi ed o as Na(+)-channel inac i a ion. P oc. Na l. Acad. Sci.
USA. 89:10910–10914. h p ://dx .doi .o g /10 .1073 /pnas .89 .22
.10910
Ye eddi, N.R., F.S. Cusdin, S. Namadu ai, L.C. Packman, T.P.
Monie, P. Sla ny, J.J. Cla e, A.J. Powell, and A.P. Jackson. 2013.
The immunoglobulin domain o he sodium channel β3 subuni
con ains a su ace-localized disul ide bond ha is equi ed o
homophilic binding. FAS EB J. 27:568–580. h p ://dx .doi .o g /10
.1096 / j .12 -209445
Yu, F.H., R.E. Wes enb oek, I. Silos-San iago, K.A. McCo mick,
D. Lawson, P. Ge, H. Fe ie a, P.S. DiS e ano, W.A. Ca e all, T.
Scheue , e al. 2003. Sodium channel β4, a new disul ide-linked
auxilia y subuni wi h simila i y o β2. J. Neu osci. 23:7577–7585.
Yuan, L., J.T. Koi umäki, B. Liang, L.G. Lo en zen, C. Tang, M.N.
Ande sen, J.H. S endsen, J. T el -Hansen, M. Malecka , N.
Schmi , e al. 2014. In es iga ions o he Na β1b sodium channel
subuni in human en icle; unc ional cha ac e iza ion o he
H162P B ugada synd ome mu an . Am. J. Physiol. Hea Ci c.
Physiol. 306:H1204–H1212. h p ://dx .doi .o g /10 .1152 /ajphea
.00405 .2013
Zhu, W., Z. Va ga, and J.R. Sil a. 2016. Molecula mo ions ha
shape he ca diac ac ion po en ial: Insigh s om ol age clamp
luo ome y. P og. Biophys. Mol. Biol. 120:3–17. h p ://dx .doi .o g
/10 .1016 /j .pbiomolbio .2015 .12 .003
on Janua y 31, 2018jgp. up ess.o gDownloaded om