1521-0111/95/2/210–221$35.00 h ps://doi.o g/10.1124/mol.118.113696
MOLECULAR PHARMACOLOGY Mol Pha macol 95:210–221, Feb ua y 2019
Copy igh ª2019 The Au ho (s).
This is an open access a icle dis ibu ed unde he CC BY A ibu ion 4.0 In e na ional license.
Ginsenosides Ac As Posi i e Modula o s o P2X4 Recep o s s
Kshi ija Dhuna, Ma hew Felga e, S e an M. Bidula, Samuel Walpole, Lucka Bibic,
B e A. C ome , Jesus Angulo, Julie Sande son, Ma in J. S ebbing, and Leanne S okes
School o Heal h and Biomedical Sciences, RMIT Uni e si y, Bundoo a, Vic o ia, Aus alia (K.D., B.A.C., M.J.S., L.S.); School o
Pha macy, Uni e si y o Eas Anglia, No wich Resea ch Pa k, No wich, Uni ed Kingdom (M.F., S.M.B., S.W., L.B., J.A., J.S.,
L.S.); Depa men o Chemis y and Bio echnology, Swinbu ne Uni e si y o Technology, Haw ho n, Vic o ia, Aus alia (B.A.C.);
and Flo ey Ins i u e o Neu oscience and Men al Heal h, Depa men o Ana omy and Neu oscience, Uni e si y o Melbou ne,
Vic o ia, Aus alia (M.J.S.)
Recei ed July 23, 2018; accep ed Decembe 10, 2018
ABSTRACT
We in es iga ed he selec i i y o p o opanaxadiol ginsenosides
om Panax ginseng ac ing as posi i e allos e ic modula o s on
P2X ecep o s. ATP-induced esponses we e measu ed in s able
cell lines o e exp essing human P2X4 using a YOPRO-1 dye
up ake assay, in acellula calcium measu emen s, and whole-
cell pa ch-clamp eco dings. Ginsenosides CK and Rd we e
demons a ed o enhance ATP esponses a P2X4 by ∼ wo old,
simila o po en ia ion by he known posi i e modula o i e mec-
in. In es iga ions in o he ole o P2X4 in media ing a cy o oxic
e ec showed ha only P2X7 exp ession in HEK-293 cells
induces cell dea h in esponse o high concen a ions o ATP,
and ha ginsenosides can enhance his p ocess. Gene a ion o a
P2X7-de icien clone o BV-2 mic oglial cells using CRISPR/
Cas9 gene edi ing enabled an in es iga ion o endogenous P2X4
in a mic oglial cell line. Compa ed wi h pa en al BV-2 cells, P2X7-
de icien BV-2 cells showed mino po en ia ion o ATP e-
sponses by ginsenosides, and insensi i i y o ATP
2
o ATP
1
ginsenoside-induced cell dea h, indica ing a p ima y ole o
P2X7 ecep o s in bo h o hese e ec s. Compu a ional docking
o a homology model o human P2X4, based on he open s a e o
z P2X4, yielded e idence o a pu a i e ginsenoside binding si e
in P2X4 in he cen al es ibule egion o he la ge ec odomain.
In oduc ion
P2X ecep o s a e a amily o ATP-ga ed nonselec i e
ca ion channels o which he e a e se en known subuni s
(P2X1–7) wi h a ying exp ession pa e ns (No h, 2002).
Thei physiological oles ange om he egula ion o mem-
b ane po en ial and in acellula calcium concen a ion (all
P2X ecep o s) o he egula ion o media o sec e ion such as
in e leukin 1b(IL-1b) and b ain-de i ed neu o ophic ac o
(BDNF) by P2X7 and P2X4, espec i ely. In es iga ions in o
d ugs ha can modula e he ac i i y o P2X ecep o s ha e
been qui e in ensi e in he las decade, wi h success ul
iden i ica ion o selec i e an agonis s o many o he P2X
amily, as e iewed in Ba le e al. (2014) and S okes e al.
(2017). We ecen ly epo ed he iden i ica ion o a se ies o
posi i e allos e ic modula o s o P2X7 ound in ex ac s o
he adi ional Chinese medicinal plan , Panax ginseng
(Helliwell e al., 2015). In his wo k, we ha e u he in es-
iga ed he selec i i y o ginsenosides o P2X7 wi hin he
P2X amily, ocusing on pu ine gic ecep o s ypically coex-
p essed wi h P2X7 in immune cells, namely P2X4, P2Y1, and
P2Y2 (Bowle e al., 2003).
P2X4 is one o he mos ubiqui ously exp essed P2X
ecep o s (So o e al., 1996) and has been implica ed in se e al
physiological pa hways in di e en issues. P ominen ex-
p ession o P2X4 has been demons a ed in endo helial cells,
immune cells, and neu ons, as e iewed in S okes e al. (2017).
An impo an ole o P2X4 in asodila ion esponses o shea
s ess was elucida ed in 2000 (Yamamo o e al., 2000), and
ansgenic mice lacking P2X4 la e con i med a ole in ni ic
oxide p oduc ion and essel emodelling (Yamamo o e al.,
2006). In he cen al ne ous sys em (CNS), P2X4 has been
implica ed in long- e m po en ia ion (Sim e al., 2006) and in
he pa hophysiology associa ed wi h neu opa hic pain (Tsuda
e al., 2003; Coull e al., 2005). P2X4 exp essed on spinal co d
mic oglia is in ol ed in ac i a ion o mic oglia and elease o
media o s, including BDNF, which al e senso y neu onal pain
ansmission pa hways (Coull e al., 2005; Ulmann e al., 2008).
Da a sha ing
Reques s o aw expe imen al da a om his s udy should be di ec ed o
he co esponding au ho .
This wo k was pa ly suppo ed by a s a egic g an om RMIT Uni e si y
Heal h Inno a ions Resea ch Ins i u e, a RMIT Uni e si y Vice Chancello s
Resea ch Fellowship, a Uni e si y o Eas Anglia School o Pha macy s a -up
und, and a Bio echnology and Biological Sciences Resea ch Council p ojec
g an awa ded o L.S. [BB/N018427/1]. K.D. was suppo ed by a RMIT
Uni e si y PhD schola ship (2014–2018). J.A. and S.W. a e suppo ed by a
Bio echnology and Biological Sciences Resea ch Council p ojec g an [BB/
P010660/1] and a Bio echnology and Biological Sciences Resea ch Council–
unded NRPDTP s uden ship [1654460], espec i ely.
h ps://doi.o g/10.1124/mol.118.113696.
sThis a icle has supplemen al ma e ial a ailable a molpha m.
aspe jou nals.o g.
ABBREVIATIONS: ANOVA, analysis o a iance; BDNF, b ain-de i ed neu o ophic ac o ; BSA, bo ine se um albumin; CK, compound K; CNS,
cen al ne ous sys em; DMEM, Dulbecco’s modi ied Eagle’s medium; DMSO, dime hylsul oxide; MTS, 3-(4,5-dime hyl hiazol-2-yl)-5-(3-
ca boxyme hoxyphenyl)-2-(4-sul ophenyl)-2H- e azolium, inne sal ; PAM, posi i e allos e ic modula o ; PBS, phospha e-bu e ed saline; PPD,
p o opanaxadiol; RMSD, oo -mean-squa e de ia ion; TBST, T is-bu e ed saline/Tween 20.
210
h p://molpha m.aspe jou nals.o g/con en /suppl/2018/12/13/mol.118.113696.DC1
Supplemen al ma e ial o his a icle can be ound a :
a ASPET Jou nals on June 8, 2020molpha m.aspe jou nals.o gDownloaded om
Also in he CNS, a ole o P2X4 has been desc ibed in
alcohol-in ake beha io due o egula ion o he dopamine
ewa d pa hway in he b ain (Asa yan e al., 2011; F anklin
e al., 2015; Khoja e al., 2016). Finally, in he immune
sys em, P2X4 plays a ole in he egula ion o CXCL5
p oduc ion and sec e ion om monocy es and mac ophages
(Layhadi e al., 2018).
Many o he oles o P2X4 ha e been elucida ed using
ansgenic P2X4
2/2
mice o sho hai pin RNA knockdown o
he ecep o because selec i e and po en an agonis s o
P2X4 ha e only ecen ly been desc ibed. These include
PSB-12062, BX430, NP-1815-PX, and 5-(3-B omophenyl)-
1,3-dihyd o-2H-benzo u o[3,2-e]-1,4-diazepin-2-one (5-BDBD)
(He nandez-Olmos e al., 2012; Balázs e al., 2013; Ase e al.,
2015; Ma sumu a e al., 2016; S okes e al., 2017). In con as
o an agonis s, ela i ely ew posi i e allos e ic modula o s
(PAMs) ha e been desc ibed o P2X ecep o s. Possibly he
bes known PAM o P2X ecep o s is i e mec in, which has
mos ac i i y a P2X4 (Khakh e al., 1999b; P iel and Silbe be g,
2004), al hough i also has some epo ed posi i e modula o
ac i i y on human P2X7 (Nö enbe g e al., 2012). O he han
i e mec in, cibac on blue, enidap, clemas ine, p oges e one,
and e ahyd odeoxyco icos e one ha e been iden i ied as
posi i e modula o s o P2X4, P2X7, and P2X2, espec i ely
(Mille e al., 1998; Sanz e al., 1998; De Roo e al., 2010;
Nö enbe g e al., 2011). In addi ion, ace me als such as
zinc and coppe ha e PAM ac i i y a se e al P2X ecep-
o s, including P2X2 and P2X4, as e iewed by Coddou
e al. (2011a,b).
Ginsenosides a e i e penoid saponins ound in he oo
ex ac o plan s belonging o P. ginseng. Two subclasses o
ginsenosides exis , designa ed p o opanaxadiols (PPD) and
p o opanax iols (PPT) due o he posi ioning o glycoside
a achmen s on he molecules. Ginsenosides ha e been e-
po ed o ac on a numbe o ion channels such as ol age-
ga ed Na
1
and K
1
channels, GABA
A
, and 5-HT
3
ecep o s
(Nah, 2014). O he cons i uen s o ginseng such as he
glycolipop o ein gin onin ha e been shown o po en ia e
P2X1 ecep o s (Choi e al., 2013). In ou p e ious s udy
(Helliwell e al., 2015), we demons a ed ha he majo
ginsenoside me aboli e gene a ed in i o, compound K (CK),
in addi ion o p o opanaxadiol ginsenosides Rd, Rb1, and Rh2,
ac s on P2X7 o po en ia e he ac ion o ATP. Ginsenosides
educed he EC
50
o he physiologic agonis , ATP, a mouse
and human P2X7 and enhanced ATP-induced cell dea h in he
J774 mouse mac ophage cell line (Helliwell e al., 2015). In
his s udy, we demons a e ha he p o opanaxadiol ginseno-
sides Rd and CK ha e a simila po en ia ing ac ion on P2X4,
mos likely h ough a highly simila binding si e iden i ied in
he cen al es ibule egion; howe e , he magni ude o he
po en ia ing e ec on P2X4 is less han a P2X7.
Ma e ials and Me hods
Ma e ials. ATP (A7699; Sigma-Ald ich, S . Louis, MO) was p e-
pa ed in dis illed wa e and adjus ed o pH 7.4 wi h 5 M NaOH.
Aliquo s we e s o ed a 220°C and used only once. 5-(3-B omophenyl)-
1,3-dihyd o-2H-benzo u o[3,2-e]-1,4-diazepin-2-one (5-BDBD) (Toc is
Biosciences, Bio-Techne L d., Abingdon, U.K.) was p epa ed in
dime hylsul oxide (DMSO) a 10 mM and s o ed a 220°C. PSB-
12062 (SML0753; Sigma-Ald ich) was p epa ed in dis illed wa e o
10 mM. I e mec in (Toc is Biosciences) was p epa ed a 30 mM in
DMSO and s o ed a 220°C. Ginsenosides CK, Rd, Rb1, Rh2, and
p o opanaxadiol (PPD) (99% pu i y) we e pu chased om Shanghai
Richem In e na ional L d., Shanghai, China, and 10 mM s ocks we e
p epa ed in s e ile DMSO (Sigma-Ald ich) in glass ials. S ock
solu ions we e s o ed a 220°C. The G115 s anda dized ginseng
ex ac was a gi o P o esso C. C. Xue (RMIT Uni e si y, Aus alia)
and was p epa ed in wa e , as desc ibed in Helliwell e al. (2015).
Cell Cul u e. S ably ans ec ed HEK-293 cells we e main ained
in Dulbecco’s modi ied Eagle’s medium (DMEM):F12 media con ain-
ing L-glu amine (ca . no. 11320-034; Li e Technologies, Fishe Scien-
i ic, Wal ham, MA) supplemen ed wi h 10% e al bo ine se um
(US o igin, ca . no. F2442; Sigma-Ald ich) and 100 U/ml penicillin
plus 100 mg/ml s ep omycin (Fishe Scien i ic, Wal ham, MA). S ably
ans ec ed 1321N1 cells we e main ained in DMEM high-glucose
media supplemen ed wi h 10% e al bo ine se um and 100 U/ml
penicillin plus 100 mg/ml s ep omycin. To main ain s able exp ession
o P2X cons uc s, 400–800 mg/ml gene icin (Gibco, Fishe Scien i ic)
was added o he cul u e media. Ini ial ansien ans ec ions o
HEK-293 cells we e pe o med in 35-mm pe i dishes using 0.1–1mg
plasmid DNA (Glu-Glu- agged hP2X4 o hP2X7 in pcDNA3) and
Lipo ec amine 2000 (3 ml/mg DNA) complexes in Op iMem media
(Fishe Scien i ic). Cells we e pla ed in o 96-well pla es (ca . no.
167008; Nunc, Fishe Scien i ic, Wal ham, MA) p ecoa ed wi h 50 mg/ml
poly-D-lysine (Me ck Millipo e, Bu ling on, MA), and pla es we e
incuba ed o e nigh be o e expe imen s.
Mouse mic oglial cell line BV-2 (a gi o D . B. Gu, Flo ey Ins i u e
o Neu osciences, Melbou ne) was main ained in DMEM:F12 media
con aining L-glu amine (ca . no. 11320-034; Li e Technologies) sup-
plemen ed wi h 10% e al bo ine se um (Sigma-Ald ich US o igin,
F2442) and 100 U/ml penicillin plus 100 mg/ml s ep omycin (Li e
Technologies).
CRISPR/Cas9 Gene Edi ing. Complemen a y RNA was designed
agains exon 2 o mouse P2RX7, and wo RNA cons uc s we e
pu chased om Dha macon (GE Heal hca e, Chicago, IL). BV-2 cells
(120,000 cells/well) in a six-well pla e we e ans ec ed wi h endo oxin-
ee 1 mg Edi R-Cas9 plasmid wi h blas icidin esis ance (GE Heal h-
ca e Dha macon, La aye e, CO) plus 25 nM a ge RNA and 25 nM
complemen a y RNA o P2RX7 using Dha ma ec DUO eagen (4 ml).
Following 24-hou incuba ion, media we e eplaced, and, a e an
addi ional 72-hou incuba ion, blas icidin was added o a inal concen-
a ion o 1 mg/ml. An ibio ic selec ion was pe o med o 5 days, and
emaining cells we e le o expand o analysis o gene edi ing. Cells
we e es ed o lack o P2X7 exp ession using low cy ome y. Success ul
wells showing lack o P2X7 exp ession unde wen single-cell cloning in
96-well pla es, and mul iple clones we e sc eened o lack o exp ession.
Genomic DNA was ex ac ed om success ul clones and ampli ied by
polyme ase chain eac ion o he P2RX7 exon 2 egion. Polyme ase
chain eac ion p oduc s we e sen o sequencing o e i y mu a ions in
his egion (Eu o ins Genomics, Ebe sbe g, Ge many).
Flow Cy ome y and Immuno luo escence. BV-2 cells (5 10
5
cells) we e s ained wi h a an i-mouse P2X7 an ibody (Hano43; Enzo
Li e Sciences UK L d, Exe e , U.K.) a 1:10 dilu ion in cold phospha e-
bu e ed saline (PBS)/0.5% bo ine se um albumin (BSA) bu e . S ain-
ing was pe o med on ice o 1 hou . Cells we e washed wi h PBS/0.5%
BSA bu e and s ained wi h a goa an i- a IgG Alexa488 seconda y
an ibody (Fishe Scien i ic) a 1:200 dilu ion o 1 hou on ice. Following
washing wi h PBS/0.5% BSA bu e , cells we e esuspended in 300 ml
PBS/0.5% BSAbu e o acquisi ion on a FACSCalibu o on a Cy o lex
low cy ome e .
Fo immuno luo escence, cells we e g own o e nigh on 13-mm
glass co e slips and washed wi h PBS p io o ixa ion wi h 4%
pa a o maldehyde (Sigma-Ald ich). Cells we e pe meabilized using
PBS/0.5% BSA bu e con aining 0.1% saponin (Sigma-Ald ich) and
blocked o 1 hou . Rabbi an i-P2X4 (Alomone Labo a o ies, Je usalem,
Is ael) was used a 1:200 dilu ion in PBS/0.5% BSA bu e , and cells
we e s ained o e nigh . Following washing, an i- abbi IgG Alexa
568 was used a 1:200 dilu ion in PBS/0.5% BSA bu e . Co e slips
we e moun ed on o glass slides using P olong Gold an i ade con aining
Posi i e Modula ion o P2X4 211
a ASPET Jou nals on June 8, 2020molpha m.aspe jou nals.o gDownloaded om
49,69-diamidino-2-phenylindole (Fishe Scien i ic). A Zeiss AxioPlan
2ie up igh luo escen mic oscope was used o ake he images.
Wes e n Blo ing. Whole-cell lysa es we e p epa ed in adio-
immunop ecipi a ion assay lysis bu e (Fishe Scien i ic) con aining
p o ease inhibi o s (Fishe Scien i ic). Cell pelle s we e dis up ed
wi h a pipe e ip and lysed on ice o 30 minu es. Lysa es we e hen
clea ed by cen i uga ion a 10,000g o 10 minu es, and supe na-
an s we e ans e ed in o clean ubes. P o ein was measu ed
using a bicinchoninic acid assay (Pie ce, Fishe Scien i ic, Wal ham,
MA) wi h BSA as s anda d (2 mg/ml o 20 ng/ml). Abso bance a
562 nm was measu ed using a Flexs a ion 3 pla e eade (Molecula
De ices, San Jose, CA). A quan i y amoun ing o 25 mg o al p o ein
was loaded pe lane on o a Bol 4%–12% Bis–T is gel (Fishe
Scien i ic) using he supplied gel-loading bu e plus educing
agen . Elec opho esis was pe o med using 1MES bu e . P o ein
was ans e ed on o poly inylidene di luo ide memb ane (Immo-
bilon P; Fishe Scien i ic) using semid y ans e o 60 minu es.
Memb anes we e blocked wi h 5% non a milk solu ion in T is-
bu e ed saline/Tween 20 (TBST) o e nigh a 4°C. P ima y an i-
bodies we e incuba ed o 2 hou s a oom empe a u e in 5% non a
milk solu ion in TBST. An i-P2X4 (APR-004; Alomone Labo a o-
ies) and an i-P2X7 C- e minal an ibody (APR-002; Alomone Lab-
o a o ies) we e bo h used a 1:2000 dilu ion. A goa an i- abbi
IgG-ho se adish pe oxidase seconda y (Sigma-Ald ich) was used a
1:2000 dilu ion. Blo s we e s ipped, washed, eblocked, and hen
ep obed o b-ac in using a mouse monoclonal an i–b-ac in an i-
body (A5316; Sigma-Ald ich) a 1:2000 dilu ion in 5% non a milk
solu ion in TBST. An an i-mouse IgG-ho se adish pe oxidase
(Sigma-Ald ich) seconda y was used a 1:10,000 dilu ion. Lumina a
chemiluminescen subs a e (Me ck Millipo e) was used o de-
elop blo s, and images we e aken using an Image Quan LAS
4000 image .
Dye Up ake Expe imen s. Fo YOPRO-1 dye up ake expe i-
men s, cells we e pla ed a a densi y o 2 10
4
cells/well in comple e
DMEM:F12 media (100 ml pe well) in poly-D-lysine–coa ed 96-well
pla es. Media we e emo ed using a manual mul ichannel pipe e and
eplaced wi h a low di alen ca ion bu e (145 mM NaCl, 2 mM KCl,
13 mM D-glucose, 10 mM HEPES, and 0.1 mM CaCl
2
, pH 7.3)
con aining 2 mM YO-PRO-1 iodide (ca . no. Y3663; Li e Technologies).
Fo p eincuba ion expe imen s, he cells we e ea ed wi h ginseno-
sides (10 mM) in YO-PRO–con aining low di alen ca ion bu e . Fo
he majo i y o expe imen s, ginsenosides we e coinjec ed simul a-
neously wi h he agonis using a Flexs a ion 3 mic opla e eade
(Molecula De ices). Ginsenosides and agonis we e p epa ed a 10
inal concen a ion in he compound pla e. Dye up ake o e ime was
eco ded using an exci a ion wa eleng h o 488 nm and an emission
wa eleng h o 520 nm on he Flexs a ion 3 (six eads/well, pho o-
mul iplie ube (PMT) se ing medium). Basal luo escence measu e-
men s we e acqui ed o 40 seconds, ollowed by au oma ic injec ion o
agonis , and he kine ic measu emen o luo escence in ensi y was
pe o med o 300 seconds using So max P o 5.4 so wa e. Dye
up ake esponses we e calcula ed as a ea unde he cu e om 50 o
300 seconds using ze o baseline no malized da a.
In acellula Calcium Measu emen s. HEK-293, HEK-hP2X4
cells, and 1321N1-hP2X4 cells in poly-D-lysine–coa ed 96-well pla es
we e loaded wi h 2 mM Fu a-2 ace oxyme hyl es e (Fishe Scien i ic
o HelloBio, B is ol, U.K.) in Hanks’balanced sal solu ion bu e
con aining 250 mM sul inpy azone (Sigma-Ald ich) o 40–60 minu es
Fig. 1. P o opanaxadiol ginsenosides and G115 inc ease hP2X4-media ed YOPRO-1 up ake esponses. (A) Rep esen a i e da a showing YOPRO-1
up ake in esponse o ATP (5 mM) in he p esence o i e mec in (IVM, 3 mM) o 5-BDBD (20 mM). (B) Summa y a ea unde cu e da a om no malized
YOPRO-1 up ake in HEK-hP2X4 cells (n= 3 independen expe imen s). E o ba s ep esen S.D.; * ep esen s P,0.05 by one-way ANOVA wi h
Dunne ’s mul iple compa ison pos hoc es . (C) YOPRO-1 up ake esponses in HEK-hP2X4 cells o ATP (5 mM) in low di alen bu e solu ion (black) o
ATP plus PPD (pu ple; 10 mM) and ATP plus CK ( ed; 10 mM). Ginsenosides we e coinjec ed wi h he agonis ATP. (D) Da a ha e been quan i ied as
a ea unde dye up ake cu e (50–180 seconds). Ba g aphs show da a om h ee o i e independen expe imen s. E o ba s ep esen S.D.; * ep esen s
P,0.05 by one-way ANOVA wi h Dunne ’s mul iple compa ison pos hoc es , ns ep esen s no signi ican .
212 Dhuna e al.
a ASPET Jou nals on June 8, 2020molpha m.aspe jou nals.o gDownloaded om
a 37°C. Following loading, bu e was emo ed using a mul ichannel
pipe e and eplaced wi h s anda d ex acellula bu e o low di alen
bu e . Cells we e wa med o 10 minu es be o e measu emen s we e
s a ed. Fu a-2 was measu ed a exci a ion wa eleng hs 340 and
380 nm wi h emission wa eleng h 520 nm using a Flexs a ion 3 pla e
eade . Sampling in e al was 3.5 seconds and h ee eads/well. Fu a-2
a io was calcula ed using So max P o 5.4, and esponses we e
measu ed using a ea unde cu e kine ic educ ion.
Calcium esponses in pa en al and P2X7-de icien BV-2 cells we e
pe o med using a Fu a-2-QBT ki (Molecula De ices). Reagen was
p epa ed in low di alen bu e con aining 250 mM sul inpy azone and
added o cells pla ed in o poly-D-lysine–coa ed 96-well pla es o
60 minu es (180 ml/well). Using he Fu a-2-QBT ki elimina ed he
washing s ep om he p ocedu e, ensu ing ha no cells we e los om
he pla e. Fu a-2 luo escence was measu ed a exci a ion wa e-
leng hs 340 and 380 nm wi h emission wa eleng h 520 nm using a
Flexs a ion 3 pla e eade . Sampling in e al was 3.5 seconds and
h ee eads/well. Fu a-2 a io was calcula ed using So max P o 5.4,
and esponses we e measu ed using a ea unde cu e kine ic e-
duc ion. Condi ions we e applied in iplica e, and expe imen s we e
pe o med ou independen imes.
Cell Viabili y. Viabili y expe imen s we e pe o med using he
CellTi e 96 Aqueous Non-Radioac i e Cell P oli e a ion assay (G3580,
P omega, Madison, WI), a colo ime ic me hod o de e mining he
numbe o iable cells. HEK-293 cells we e pla ed a 5 10
4
cells/well
in comple e DMEM:F12 media and le o 24 hou s unde no -
mal g ow h condi ions. T ea men s we e hen added o he cells
(2 inal concen a ion) and incuba ed o an addi ional 24 hou s.
3-(4,5-dime hyl hiazol-2-yl)-5-(3-ca boxyme hoxyphenyl)-2-(4-sul o-
phenyl)-2H- e azolium, inne sal (MTS) solu ion (20 ml) was added
o media in each well 1–4 hou s be o e he s ipula ed end poin ime.
Abso bance was measu ed a 490 nm using a Cla ios a pla e eade
(BMG Lab ech L d., Aylesbu y, U.K.) o a Flexs a ion 3 pla e eade .
Pa ch Clamping. HEK-293 cells s ably exp essing hP2X4 we e
pla ed on o 13-mm glass co e slips 24 hou s be o e eco ding ATP-
induced memb ane cu en s in he whole-cell pa ch clamp con igu a-
ion. An EPC10 ampli ie (HEKA Elek onik, Ha a d Bioscience
Inc., Lamb ech /P alz, Ge many) was used o ol age clamp cells
a 260 mV. Bo osilica e glass elec odes (Wo ld P ecision Ins u-
men s, Hi chin, U.K.) had a esis ance o 5–8MVwhen illed wi h
s anda d in e nal solu ion (145 mM NaCl, 10 mM HEPES, 10 mM
EGTA, pH 7.3, wi h 5 M NaOH). Cells we e con inually pe used by a
g a i y eed wi h s anda d ex acellula bu e solu ion (145 mM
NaCl, 5 mM KCl, 2 mM CaCl
2
, 1 mM MgCl
2
, 13 mM glucose, 10 mM
HEPES, pH 7.3) p io o seal o ma ion. ATP and ginsenosides we e
applied using a compu e -con olled as - low sys em (Bio-Logic
Science Ins umen s, Seyssine -Pa ise , F ance) wi h he pe usion
capilla ies placed in close p oximi y o he cell unde in es iga ion.
Cells we e dialyzed wi h in e nal solu ion o 1 minu e, and ATP in
he absence o p esence o ginsenoside was applied e e y minu e o
ou applica ions.
Molecula Modeling and Docking. The coo dina es o zeb a ish
P2X4 in he ATP-bound open s a e (Ha o i and Gouaux, 2012)
(P o ein Da a Bank: 4DW1) we e used as a empla e. The sequence
o human P2X4 was aligned o he empla e using he Clus alW
algo i hm (La kin e al., 2007). The Sch ödinge P ime so wa e
(www.sch odinge .com) was used o cons uc an ene gy-based all-
a om model wi h he OPLS3 o ce ield keeping he ATP coo dina es
om he empla e. Th ee-dimensional models o he ginsenosides CK
and Rd we e gene a ed using LigP ep so wa e wi hin he Sch ö-
dinge Maes o sui e (www.sch odinge .com). The OPLS3 o ce ield
was used o gene a e 32 low-ene gy con o me s o each ginsenoside.
Induced i docking was pe o med using he au oma ed ex ended
sampling p o ocol in he Sch ödinge Maes o sui e (She man e al.,
2006). This i s pe o ms se e al ini ial docking uns in which side
chains a e ei he immed, o hei Van de Waals po en ials so ened
acco ding o hei lexibili y. Side chains a e hen ebuil , and hose
wi hin 5 Å o he ligand a e op imized using P ime. Ligands we e hen
edocked o he new ecep o s uc u e using he Glide SP algo i hm
(F iesne e al., 2004) and s anda d po en ials. S uc u es wi hin
30 kcal mol
21
o he lowes ene gy s uc u e we e e ained. The
ecep o g id was cen e ed on he highes -sco ing po en ial binding
si e, ound using Si emap (Halg en, 2007), and had cubic box
dimensions o 30 Å. Each o he 32 con o me s was docked using he
ex ended sampling p o ocol. Fo each ginsenoside, he esul ing poses
we e clus e ed by hea y a om oo -mean-squa e de ia ion (RMSD)
using he a e age-linkage me hod, and a ep esen a i e s uc u e was
chosen om he model closes o he cen oid o he mos popula ed
clus e . Fo CK and Rd, he mos popula ed clus e made up 46% and
82% o all solu ions, espec i ely.
S a is ical Analysis. G aphs we e plo ed using G aphPad P ism
e sions 6 o 7 (La Jolla, CA). Concen a ion– esponse cu es we e
i ed using a log (agonis ) e sus esponse– a iable slope ( ou
pa ame e ) bes - i equa ion. Da a we e analyzed o s a is ical signi -
icance using one-way analysis o a iance (ANOVA) o wo-way ANOVA
wi h pos hoc es s as app op ia e (G aphPad P ism). Signi icance was
aken as P,0.05.
Resul s
Fo ou in es iga ion in o he selec i i y o p o opanaxadiol
ginsenosides o P2X7, we chose he closely ela ed P2X4
ecep o because his is known o be coexp essed wi h P2X7 in
mac ophages and o he immune cells. We i s gene a ed a
HEK-293 cell line s ably exp essing a Glu-Glu C- e minal
Fig. 2. Ginsenosides enhance ATP esponses a ecombinan hP2X4. (A)
Va ious concen a ions o ATP (100 nM o 100 mM) we e used o ac i a e
hP2X4, and YOPRO-1 up ake o e 200 seconds was measu ed on a
Flexs a ion 3. Concen a ion esponse o ATP is shown in black, ATP in
he p esence o 10 mM CK is shown in ed, and ATP in he p esence o
i e mec in (IVM; 5 mM) is shown in o ange. Da a a e aken om h ee
independen expe imen s, and e o ba s ep esen S.E.M. T ans o med
da a we e cu e i using nonlinea eg ession, and EC
50
alues we e
1.92 mM ATP, 1.22 mM o ATP + CK, and 1.3 mM o ATP + IVM. (B)
Concen a ion esponse ela ionship o selec ginsenosides on YOPRO-1
dye up ake o 5 mM ATP in HEK-hP2X4 cells. Ginsenosides we e es ed
o e he ange 0.1–50 mM. EC
50
alues we e 7.5 mM Rd, 8.5 mM CK, and
10.5 mM Rb1. Da a we e aken om h ee independen expe imen s.
Posi i e Modula ion o P2X4 213
a ASPET Jou nals on June 8, 2020molpha m.aspe jou nals.o gDownloaded om
agged human P2X4 (HEK-hP2X4) and es ablished an assay
o measu e P2X4-media ed seconda y po e esponses ia
YOPRO-1 dye up ake using a Flexs a ion 3 pla e eade .
The ATP-induced YOPRO-1 dye up ake esponse could be
inc eased by i e mec in (3–5mM) and educed by he known
P2X4 an agonis 5-BDBD (20 mM) (Fig. 1). This con i ms ha
P2X4 ecep o s can elici a la ge po e esponse in HEK-293
cells (Be nie e al., 2012). The s anda dized ginseng ex ac
G115 (100 mg/ml) was also ound o inc ease he P2X4 dye
up ake esponse (Fig. 1B). We in es iga ed he e ec o he
indi idual pu i ied ginsenoside chemicals on P2X4-media ed
YOPRO-1 dye up ake esponses in HEK-293 cells. To do his,
we coinjec ed 10 mM o he ollowing ginsenosides—CK, Rd,
Rb1, Rh2, o he aglycone PPD— oge he wi h ATP. Simila o
he epo ed po en ia ion o P2X7 (Helliwell e al., 2015), he
ginsenosides also po en ia ed P2X4 esponses (Fig. 1, C and
D). The da a sugges ha Rd, CK, and Rb1 can po en ia e
P2X4 (P,0.05), whe eas Rh2 and PPD had no e ec on
ATP-induced P2X4 po e esponses in HEK-293 cells.
We in es iga ed whe he ginsenoside CK could shi he
concen a ion– esponse cu e o ATP and al e he sensi i i y
o he P2X4 ecep o o he agonis . We p e iously epo ed ha
ginsenosideCK educed heEC
50
o ATPa P2X7 in addi ion o
inc easing he maximum esponse elici ed by ATP (Helliwell
e al., 2015). Using heYOPRO-1 up ake assay, 10 mMCKcould
inc ease he maximum esponse induced by ATP, simila o he
e ec o he posi i e modula o i e mec in (5 mM) (Fig. 2A). The
EC
50
alue o ATP was 1.92 mM (95% con idence in e al 1.03
o 3.56 mM) compa ed wi h 1.22 mM (con idence in e al 0.49 o
2.99 mM) in he p esence o CK, sugges ing ha CK does no
d ama ically a ec sensi i i y o agonis . We also de e mined
he EC
50
alue o each ginsenoside on po en ia ion o hP2X4
esponses (Fig. 2B), which we e Rd 7.5 mM (con idence in e al
5.74–9.91 mM), CK 8.5 mM (con idence in e al 5.88–12.4 mM),
and Rb1 10.5 mM (con idence in e al 8.46–13.23 mM).
Ou nex s ep was o in es iga e he e ec s o ginsenosides
on P2X4-media ed calcium esponses using he HEK-hP2X4
s able cell line, as his is mo e physiologically ele an . Fu a-2
ace oxyme hyl es e –loaded HEK-hP2X4 cells we e chal-
lenged wi h ATP (1 mM) in he absence o p esence o 10 mM
CK, and calcium esponses we e measu ed o 300 seconds
(Fig. 3A). Endogenous P2Y ecep o s a e known o con ibu e
o he ATP-induced calcium esponses in HEK-293 cells;
howe e , a sepa a e se o expe imen s con i med ha he
ginsenosides did no po en ia e ADP- o UTP-media ed
calcium esponses (Supplemen al Fig. 1). Bo h CK and Rd
Fig. 3. Ginsenosides enhance Ca
2+
in lux esponses in cells exp essing hP2X4. In acellula Ca
2+
esponses we e measu ed in (A) Fu a-2 ace oxyme hyl
es e –loaded HEK-hP2X4 cells o (C) Fu a-2 ace oxyme hyl es e –loaded 1321N1-hP2X4 cells. Baseline alues we e eco ded o 15–20 seconds,
and hen ATP was au oma ically injec ed a 30 seconds (a ow indica es agonis addi ion). Rep esen a i e ATP esponses a e shown in black; ATP
plus CK (10 mM) a e shown in ed. E o ba s a e S.D. (B) Ca
2+
esponses in HEK-hP2X4 we e calcula ed as a e age Fu a-2 a io o e he inal
150 seconds. CK and Rd (10 mM) signi ican ly inc ease hP2X4-media ed Ca
2+
in lux. *P,0.05, one-way ANOVA wi h Dunne ’s mul iple compa ison
pos hoc es . (D) Ca
2+
esponses in 1321N1-hP2X4 cells we e measu ed as a ea unde he cu e o ze o baseline; e o ba s a e S.D.; da a a e om ou
independen expe imen s.
214 Dhuna e al.
a ASPET Jou nals on June 8, 2020molpha m.aspe jou nals.o gDownloaded om

caused an inc ease in he sus ained calcium esponse
media ed by hP2X4 in HEK-293 cells (Fig. 3B) (P,0.05).
We also gene a ed a s able 1321N1 as ocy oma cell line
exp essing hP2X4 and con i med ha he ginsenosides
CK and Rd po en ia ed ATP-induced calcium esponses in
his second cell ype (Fig. 3, C and D). Finally, o con i m
ginsenoside po en ia ion o hP2X4 channel esponses, we
used whole-cell pa ch clamp eco dings in HEK-hP2X4 cells.
As o he s ha e shown, mul iple addi ions o ATP wi h a
washou pe iod in be ween cause a undown e ec on
esponses (Fig. 4A), in which he second esponse was 73%
65.8% o he ini ial esponse (n57 cells). Figu e 4B
demons a es he e ec o coadminis e ing ATP in he p es-
ence o CK o Rd as he second applica ion, showing ha he
ginsenosides inc ease he ATP esponses. Figu e 4C sum-
ma izes he pa ch clamp da a as no malized da a (pe cen -
age o second ATP applica ion) (n57–10 cells).
By measu ing indi idual P2X ecep o s he e ologously
exp essed in HEK-293 and 1321N1 cells using a a ie y o
expe imen al echniques, we ha e demons a ed ha se-
lec ed ginsenosides, namely CK and Rd, can po en ia e
P2X4 esponses. In ou p e ious s udy, we in es iga ed he
downs eam consequences o po en ia ing P2X7 ecep o s
wi h he ginsenoside CK by de e mining i s e ec on ATP-
induced cell dea h in mac ophages (Helliwell e al., 2015).
The e o e, we in es iga ed whe he P2X4 ecep o s play any
ole in he induc ion o cell dea h and whe he ginsenosides
could enhance his. We pe o med cell iabili y expe imen s
using cell iabili y MTS abso bance assays in HEK-293 cells
s ably exp essing ei he hP2X4 o hP2X7. In HEK-hP2X7
he e was a educ ion in cell iabili y wi h inc easing concen-
a ions o ATP wi h concen a ions g ea e han 1 mM causing
cell dea h (Fig. 5). The selec i e P2X7 an agonis AZ10606120
(10 mM) could e ec i ely block he cell dea h induced by hese
high concen a ions o ATP (Fig. 5). The p esence o CK
enhanced he le hal e ec o 500 mM ATP on P2X7 in hese
expe imen s (Fig. 5). Con e sely, in HEK-hP2X4 cells, he e
was no meaning ul educ ion in cell iabili y wi h inc easing
concen a ions o ATP (Fig. 5); cells ea ed wi h 3 mM ATP
we e ully iable (104% o con ol). Viabili ywas educed o 86%
o con ol when 5-BDBD was in oduced in combina ion wi h
high concen a ions o ATP (1 and 3 mM) and iabili y was
educed o 76% o con ol when CK was in oduced wi h 3 mM
ATP (Fig. 5, P,0.05).
To in es iga e he con ibu ion o ginsenoside po en ia ion
o P2X4 esponses e sus ginsenoside po en ia ion o P2X7
esponses in immune cells, we used he mouse mic oglia BV-2
cell line. These cells a e known o exp ess bo h P2X4 and P2X7
ecep o s (Gilbe e al., 2016; Asa yan e al., 2018). We
gene a ed a P2X7-de icien BV-2 cell line using CRISPR/Cas9
gene-edi ing echnology di ec ed agains exon 2 o mouse
P2X7. Single-cell clones o P2X7-de icien BV-2 cells we e
gene a ed and cha ac e ized using low cy ome y and Wes -
e n blo ing o de ec su ace and o al P2X7 p o ein, e-
spec i ely (Fig. 6, A and B). P2X4 p o ein exp ession was no
a ec ed in BV-2 cells lacking P2X7 (Fig. 6B), and he
dis ibu ion o P2X4 appea ed simila in BV-2 and P2X7-
de icien BV-2 cells (Fig. 6C). ATP (25 mM)-induced calcium
esponses we e simila in he BV-2 and P2X7-de icien BV-2
cells and displayedP2X4-like pha macological p ope ies such
as po en ia ion by i e mec in and educ ion in esponses
by wo P2X4 an agonis s 5-BDBD and PSB-12062 (Fig. 6D,
P,0.05). In he pa en al BV-2 cells, bo h ginsenosides
CK and Rd ga e a s a is ically signi ican po en ia ion o he
ATP-induced calcium esponse by 5.6- old and 6.05 old,
espec i ely (Fig. 7A, P,0.05 by wo-way ANOVA); howe e ,
his e ec was d ama ically educed in cells lacking P2X7
ecep o s o 1.6- old (CK) and 1.1- old (Rd) po en ia ion (Fig. 7,
A and B). This sugges s ha he majo i y o he ginsenoside
e ec is due o hei ac ion on P2X7. We also demons a e ha
Fig. 4. Pa ch clamp analysis con i ms CK, and Rd en-
hances ATP-induced cu en s. HEK-hP2X4 cells we e ol -
age clamped a 260 mV, and (A) ATP (5 mM) was apidly
applied o 5 seconds (pulse deno ed by black ba s), ollowed
by a 45-second washou wi h s anda d ex acellula solu-
ion. In con ol cells, u he applica ions o ATP we e
applied e e y 1 minu e o a o al o ou applica ions.
Human P2X4 esponses we e obse ed o display consis en
undown. (B) In ginsenoside- ea ed cells, an ini ial con ol
applica ion o ATP was adminis e ed and hen he second
applica ion was ATP plus CK o ATP plus Rd (bo h a
10 mM). (C) No malized cu en ampli udes (pA/pF) show-
ing ATP undown o e he i s ou applica ions. Po en i-
a ion o ATP esponses by CK ( ed) and Rd (blue) exp essed
as pe cen age o second ATP esponse.
Posi i e Modula ion o P2X4 215
a ASPET Jou nals on June 8, 2020molpha m.aspe jou nals.o gDownloaded om
pa en al BV-2 cells we e success ully killed by 3 mM ATP,
whe eas P2X7-de icien BV-2 cells we e p o ec ed om ATP-
induced cell dea h (Fig. 7C). The addi ion o CK (10 mM)
oge he wi h ATP did no induce o enhance any cell dea h in
P2X7-de icien BV-2 cells (Fig. 7C).
We ha e in es iga ed pu a i e binding si es on P2X4 and
P2X7 ecep o s o he ginsenosides using a homology model o
hP2X4 gene a ed om he c ys al s uc u e o z P2X4 in he
ATP-bound open s a e (P o ein Da a Bank: 4DW1). The
docking poses o bo h CK (Supplemen al Ma e ial 1) and Rd
(Supplemen al Ma e ial 2) on hP2X4 we e simila o hose
obse ed o hP2X7 (Fig. 8). The e a e no able di e ences in
he leng h o he in e nal loop a he op bounda y o he
cen al es ibule, wi h his sec ion being longe in hP2X4 and
con aining wo nega i ely cha ged glu amic acid esidues
(-QEENS-) in compa ison wi h neu al sequence -QGNS- in
hP2X7. This addi ional s e ic epulsion o he longe in e -
nal loop in P2X4, as well as elec os a ic epulsion o he
diglu ama e mo i , may al e he binding mode o he ginseno-
sides compa ed wi h hP2X7. Figu e 8, B and E, shows o e lays
o he p edic ed docking poses o CK and Rd on hP2X7 and
hP2X4. Se e al in e ac ing esidues ha e been highligh ed,
including E96 in he in e nal loop, S62 and D58 in he b-2
shee , plus D320 in he b-14 shee . Ginsenoside Rd binds in an
in e ed o ien a ion wi h espec o CK, which is simila o he
docking pose obse ed on hP2X7 (unpublished da a).
Discussion
In his s udy, we ha e desc ibed a no el posi i e allos e ic
modula o o P2X4 and highligh ed a pu a i e binding si e o
posi i e allos e ic modula o s ha may be conse ed be ween
P2X4 and P2X7 ecep o s. We ha e p esen ed pha macolog-
ical da a o posi i e modula o ac i i y a P2X4 and ha e
desc ibed impo an di e ences be ween ginsenoside ac ion a
P2X4 and P2X7.
To measu e P2X4 esponses, we used a YOPRO-1 dye
up ake assay, which is mos commonly used o measu e
P2X7 esponses. Ea lie wo ks sugges ed ha dye up ake
esponses could also be induced by o he pu ine gic P2X
ecep o s, including P2X4 (Khakh e al., 1999a; Vi ginio
e al., 1999), and his was con i med by Be nie e al. (2012)
in bo h ans ec ed HEK-293 cells and mic oglia. This migh
sugges ha dye up ake is an in insic ea u e o P2X
ecep o channels. Rega dless o he mechanism unde lying
pe mea ion o la ge dye molecules, his is a po en ially
use ul way o measu ing P2X4 esponses in a medium- o
high- h oughpu sys em. We show ha dye up ake was
induced by low concen a ions o ATP, was po en ia ed by
i e mec in, and was educed by p e ea men o he cells
wi h 5-BDBD, all cha ac e is ics o a P2X4-media ed e-
sponse. In hese expe imen s, CK appea ed o be mo e
e ec i e a enhancing ATP-induced esponses han i e -
mec in (Figs. 1 and 2).
Tes ing he p o opanaxadiol ginsenosides on P2X4 in
YOPRO-1, calcium measu emen s and pa ch clamp eco dings
e ealed some di e ences o P2X7; he mos s iking di e ence
was ha Rd was equi alen o CK in he le el o po en ia ion i
induced on P2X4, whe eas CK is mo e e ec i e a P2X7.
O e all, he magni ude o po en ia ion obse ed o P2X4 was
much lowe han on P2X7; ginsenosides ypically inc eased
P2X4 esponses by ∼ wo old, whe eas his eached .20- old in
P2X7 esponses by pa ch clamp analysis (Helliwell e al., 2015).
The simila PAM ac ion on P2X4 and P2X7 p omp ed an
in es iga ion in o whe he he p edic ed binding si e o
ginsenosides would be he same. The e is a high deg ee o
simila i y be ween hese channels, and we used he z P2X4
c ys al s uc u e o p edic he ATP-bound open s a es o
hP2X4 and hP2X7. Di e ences in he binding poses we e
obse ed and in amino acid esidues p edic ed o in e ac wi h
bound ginsenosides, he oles o which will be in es iga ed in a
mo e de ailed s udy. I is no able ha he p edic ed ginsenoside
si e does no o e lap wi h he i e mec in binding si e on P2X4,
which is hough o be loca ed wi hin he ansmemb ane
domains (Jelínko á e al., 2006; Silbe be g e al., 2007; Popo a
e al., 2013), sugges ing ha wo dis inc posi i e modula o
binding si es exis o P2X4.
In his s udy, we ha e paid pa icula a en ion o he
cy o oxic ac ion o ATP. I is well known ha sus ained
ac i a ion o P2X7 leads o cell dea h (O ioli e al., 2017),
and we ha e shown ha ginsenosides can enhance his
(Helliwell e al., 2015). The ole o o he pu ine gic ecep o s
in cell dea h is no so clea , and he e o e we in es iga ed he
in ol emen o P2X4 in his p ocess. We ound ha sus ained
ac i a ion o hP2X4 o e exp essed in HEK-293 cells did no
cause cell dea h (Fig. 5), and u he mo e, high concen a ions
o ATP we e no cy o oxic in P2X7-de icicen BV-2 mic oglial
Fig. 5. Ginsenoside CK does no enhance cell dea h in HEK-hP2X4 cells.
HEK-293 cells s ably exp essing hP2X4 o hP2X7 (50,000 cells/well) we e
pla ed in o 96-well pla es o 24 hou s p io o s imula ion wi h ATP.
Cells we e p e ea ed wi h an agonis s AZ10606120 (AZ) o 5-BDBD o
10 minu es be o e ATP was added. Cells we e exposed o he indica ed
concen a ions o ATP o ATP + CK (10 mM) o 24 hou s, and iabili y was
measu ed using MTS abso bance. MTS was added o he inal 4 hou s o
incuba ion, and abso bance ead using a Flexs a ion 3 eade . Da a we e
no malized o cells ea ed wi h ehicle con ol and a e exp essed as
pe cen age o con ol. Da a a e om ou independen expe imen s.
216 Dhuna e al.
a ASPET Jou nals on June 8, 2020molpha m.aspe jou nals.o gDownloaded om
Fig. 6. P2X4 esponses in gene-edi ed BV-2 cells o elimina e P2X7 ecep o s. BV-2 cells de icien o P2X7 we e gene a ed using CRISPR/Cas9 edi ing.
(A) Flow cy ome y da a indica ing pa en al BV-2 cells exp ess P2X7 (black his og am), whe eas P2X7-de icien BV-2 cells show no P2X7-posi i e s ain
(g een his og am) using Hano43 an ibody labeling. Red his og am indica es nega i e con ol, cells labeled wi h seconda y an ibody only. (B) Wes e n
blo s showing no an i-P2X7 eac i e p o ein band in P2X7-de icien BV-2 cells compa ed wi h pa en al BV-2 and J774 mac ophages. Equal amoun s o
p o ein (25 mg) we e loaded pe lane as indica ed by he b-ac in loading con ol on s ipped and ep obed blo s. P2X4 p o ein le els we e unchanged
ollowing knockou o P2X7. (C) S aining o P2X4 ecep o s was pe o med using a ix/pe m me hod wi h cells g own on 13-mm glass co e slips. Rabbi
an i-P2X4 (1:200) was used o label all P2X4 ecep o s in BV-2 pa en al and P2X7-de icien BV-2 cells. Images a e ep esen a i e o wo independen
expe imen s. (D) In acellula calcium measu emen s in Fu a-2 ace oxyme hyl es e –loaded BV-2 and P2X7-de icien BV-2 show iden ical ATP (25 mM)
esponses. ATP + i e mec in (3 mM) esponses we e educed by p e ea men o cells wi h 5-BDBD (10 mM) o PSB-12062 (1 mM). E o ba s a e S.D. and
*P,0.05, one-way ANOVA wi h Dunne ’s mul iple compa ison pos hoc es .
Posi i e Modula ion o P2X4 217
a ASPET Jou nals on June 8, 2020molpha m.aspe jou nals.o gDownloaded om
cells (Fig. 7), sugges ing ha ac i a ion o P2X4 does no cause
cell dea h. In HEK-hP2X7 cells, CK was able o enhance cell
dea h o nonle hal concen a ions o ATP (500 mM), simila o
J774 cells (Helliwell e al., 2015). This is likely due o he
inc eased ecep o sensi i i y o ATP in he p esence o CK,
allowing cells o o e come a h eshold le el o signaling o
he ini ia ion o apop osis. CK was no as e ec i e a
enhancing cell dea h in esponse o 1 o 3 mM ATP; howe e ,
his may be due o maximum cell dea h being induced by ull
ac i a ion o P2X7.
Posi i e allos e ic modula o s o P2X4 a e hough o be
use ul in ea men o alcohol-use diso de s due o a compe -
i i e e ec on channel ac i i y (Asa yan e al., 2010; Ya dley
e al., 2012). E hanol has an inhibi o y e ec on P2X4
esponses and p e en s he ac ion o ATP, which in u n
a ec s he modula ion o neu o ansmission in a ious e-
gions o he b ain (F anklin e al., 2014). The binding o
i e mec in o P2X4 is pos ula ed o coun e ac he binding o
e hanol, hus p e en ing i s inhibi o y e ec on channel
ac i a ion. I would be in e es ing o de e mine whe he
posi i e allos e ic modula o s ha bind o o he egions o
he channel will ha e a simila e ec on elimina ion o e hanol
e ec s on P2X4. In addi ion o alcohol-use diso de s, P2X4 has
been implica ed in he egula ion o dopamine-dependen
beha io s (Khoja e al., 2016), sugges ing ha modula ion
o P2X4 in he CNS may also be bene icial in dopamine -
gic diso de s such as Pa kinson’s disease. In e es ingly,
P. ginseng ex ac is epo ed o ha e some bene icial e ec s
in animal models o Pa kinson’s disease (Van Kampen e al.,
2014). Howe e , due o hei chemical p ope ies, i is
unclea how easily he ginsenosides c oss he in ac blood-
b ain ba ie . A ecen s udy sugges s ginsenosides can be
de ec ed in a ascula endo helial cells and as ocy es
ollowing o al adminis a ion o high pu i y ginseng o al
saponins (Zhao e al., 2018). The la ge ginsenosides wi h
mul iple suga moie ies a e much less likely o be abso bed
om he in es ine in o he bloods eam, bu he smalle
ginsenosides and me aboli es such as CK a e ound a
app eciable concen a ions in plasma and li e . Peak plasma
concen a ions o CK a e epo ed o be 8.35 ng/ml (low nM
ange) in humans ollowing o al adminis a ion o Ko ean
ed ginseng ex ac (Kim, 2013), al hough a mo e ecen
s udy ound highe plasma concen a ions (1183.2 6445.1 ng/ml
ollowing a single 800 mg dose) in olun ee s ecei ing pu e
Ginsenoside CK able s (Chen e al., 2018). Fu he mo e, he
sou ce o he ginsenosides, whe he om a na u al ex ac ,
Fig. 7. Ginsenosides show li le po en ia ion o ATP-induced esponses in BV-2 mic oglia in he absence o P2X7. (A and B) In acellula calcium
measu emen s we e pe o med using a no-wash Fu a-2-QBT assay on pa en al BV-2 cells and P2X7-de icien BV-2 mic oglia cells. ATP (200 mM) was
injec ed in he absence o p esence o a ious ginsenosides (all 10 mM). Calcium esponses we e calcula ed as a ea unde cu e using So max P o
so wa e. Signi ican po en ia ion was only e iden in cells exp essing P2X7. Da a a e om h ee independen expe imen s, e o ba s ep esen S.D.,
and *indica es P,0.05 using wo-way ANOVA wi h Bon e oni’s mul iple compa ison pos hoc es . (C) Cell iabili y was assessed using MTS
eagen .BV-2cellswe epla ed o 24hou sp io os imula ionwi hATPin heabsenceo p esenceo ginsenosideCK(10mM). Cells we e ea ed o
24 hou s, and MTS was added o he inal 1 hou . Da a a e ep esen ed as mean pe cen age o con ol alues in which con ol was ehicle- ea ed
cells. Da a a e om h ee independen expe imen s, e o ba s ep esen S.D., and *indica es P,0.05 using one-way ANOVA wi h Dunne ’s
mul iple compa ison pos hoc es .
218 Dhuna e al.
a ASPET Jou nals on June 8, 2020molpha m.aspe jou nals.o gDownloaded om