Impac o ansien down- egula ion o DREAM in human emb yonic
s em cell plu ipo ency☆
The ole o DREAM in he main enance o hESCs
A. Fon án-Lozano
a,1
, V. Capilla-Gonzalez
a
,Y.Aguile a
a
, N. Mellado
a
,A.M.Ca ión
b
,B.So ia
a,c
,A.Hmadcha
a,c,
⁎
a
Cen o Andaluz de Biología Molecula y Medicina Regene a i a (CABIMER), Se illa 41092, Spain
b
Di isión de Neu ociencias, Uni e sidad Pablo de Ola ide de Se illa, Se illa 41013, Spain
c
CIBER de Diabe es y En e medades Me abólica asociada (CIBERDEM), Mad id 28029, Spain
abs ac a icle in o
A icle his o y:
Recei ed 13 Ap il 2015
Recei ed in e ised o m 25 Feb ua y 2016
Accep ed 3 Ma ch 2016
A ailable online 4 Ma ch 2016
Li le is known abou he unc ions o downs eam egula o y elemen an agonis modula o (DREAM) in emb y-
onic s em cells (ESCs). Howe e , DREAM in e ac s wi h cAMP esponse elemen -binding p o ein (CREB) in a
Ca
2+
-dependen manne , p e en ing CREB binding p o ein (CBP) ec ui men . Fu he mo e, CREB and CBP a e
in ol ed in main aining ESC sel - enewal and plu ipo ency. Howe e , a p e ious knockou s udy e ealed he
p o ec i e unc ion o DREAM deple ion in b ain aging degene a ion and ha aging is accompanied by a p og es-
si e decline in s em cells (SCs) unc ion. In e es ingly, we ound ha DREAM is exp essed in di e en cell ypes,
including human ESCs (hESCs), human adipose-de i ed s omal cells (hASCs), human bone ma ow-de i ed
s omal cells (hBMSCs), and human newbo n o eskin fib oblas s (hFFs), and ha ansi o y inhibi ion o
DREAM in hESCs educes hei plu ipo ency, inc easing di e en ia ion. We s ipula e ha hese changes a e pa ly
media ed by inc eased CREB ansc ip ional ac i i y. O e all, ou da a indica es ha DREAM ac s in he egula ion
o hESC plu ipo ency and could be a a ge o p omo e o p e en di e en ia ion in emb yonic cells.
© 2016 The Au ho s. Published by Else ie B.V. This is an open access a icle unde he CC BY license
(h p://c ea i ecommons.o g/licenses/by/4.0/).
Keywo ds:
Human emb yonic s em cells
Plu ipo ency
Di e en ia ion
DREAM
KChIP-3
Calsenilin
CREB
1. In oduc ion
Downs eam egula o y elemen an agonis modula o (DREAM) is
a mul i unc ional Ca
2+
-binding p o ein, in he EF-hand sub amily
o neu onal calcium senso s, wi h specific oles in di e en cell
compa men s (Ca ion e al., 1999; An e al., 2000). In he nucleus,
DREAM ac s as a Ca
2+
-dependen ansc ip ional ep esso (Ca ion
e al., 1999; Campos e al., 2003; Ledo e al., 2000, 2002; Ri as e al.,
2004) and a ansc ip ion ac o in e ac ing di ec ly wi h DNA a a
specific sequence, he down egula o y elemen (DRE) (Ca ion e al.,
1999; Campos e al., 2003). Ou side o he nucleus, DREAM in e ac s
wi h K 4 po assium channels, di ec ing hei a ficking o he plasma
memb ane and egula ing channel ga ing p ope ies (An e al., 2000;
Takimo o e al., 2002; Fon an-Lozano e al., 2011). In addi ion, DREAM
appea s o di ec ly o indi ec ly a ec synap ic plas ici y by modula ing
N-me hyl-D-aspa a e (NMDA) ecep o s (Fon an-Lozano e al., 2011;
Wu e al., 2010; Zhang e al., 2010). In he cy osol, DREAM binds o
p esenilins (Buxbaum e al., 1998), blocks he elease o Ca
2+
om
he endoplasmic e iculum, and induces apop osis o p esenilin mu an s
associa ed wi h Alzheime disease (Lilliehook e al., 2002).
Exp ession o DREAM is also upon cyclic adenosine monophospha e
(cAMP) signaling (Ca ion e al., 1999). I has been epo ed ha
DREAM in e ac s wi h CREB (cAMP esponse elemen -binding p o ein)
wi hin i s kinase-inducible domain (KID) (Ledo e al., 2002). CREB binds
o cAMP- esponse elemen sequences (CRE si e) (Zhang e al., 2005). To
a ain ull ansc ip ional ac i i y, CREB in e ac s wi h he his one ace yl
ans e ase enzyme CBP dependen on Se -133 phospho yla ion (May
and Mon miny, 2001). Unphospho yla ed CREB in e ac s wi h DREAM
o abolish CREB–CBP in e ac ion in a Ca
2+
-dependen manne (Ledo
e al., 2002).
Emb yonic s em cells (ESCs) display wo impo an cha ac e is ics:
sel - enewal, which allows hem o be indefini ely expanded while
S em Cell Resea ch 16 (2016) 568–578
Abb e ia ions: SCs, s em cells; hESCs, human emb yonic s em cells; hASCs, human
adipose-de i ed s omal cells; hBMSCs, human bone ma ow-de i ed s omal cells;
hFFs, human newbo n o eskin fib oblas s; DREAM, downs eam egula o y elemen an-
agonis modula o (also e med KChIP-3, po assium channel in e ac ing p o ein-3, o
calsenilin); CREB, cAMP esponse elemen -binding p o ein; pCREB, phospho yla ed
CREB; CBP, CREB binding p o ein.
☆Au ho con ibu ions in he s udy a e as ollows: A. Fon án-Lozano and A. Hmadcha
designed he s udy, analyzed he da a, and w o e he manusc ip ; A. Fon án-Lozano pe -
o med mos o he expe imen s wi h he help o V. Capilla-Gonzalez; Y. Aguile a and N.
Mellado pe o med he cul u e o hESCs; A. M. Ca ión assis ed in some expe imen s;
and B. So ia and A. Hmadcha di ec ed he s udy.
⁎Co esponding au ho s a : Depa men o S em Cells, Andalucian Cen e o Molecula
Biology and Regene a i e Medicine (CABIMER), A da. Amé ico Vespucio s/n, Pa que
Cien i íco y Tecnológico Ca uja 93, Se illa 41092, Spain.
E-mail add ess: ka im.hmadcha@cabime .es (A. Hmadcha).
1
P esen add ess: Depa men o Molecula , Cellula and De elopmen al Neu oscience,
Cajal Ins i u e —CSIC, Mad id, 28002, Mad id, Spain.
h p://dx.doi.o g/10.1016/j.sc .2016.03.001
1873-5061/© 2016 The Au ho s. Published by Else ie B.V. This is an open access a icle unde he CC BY license (h p://c ea i ecommons.o g/licenses/by/4.0/).
Con en s lis s a ailable a ScienceDi ec
S em Cell Resea ch
jou nal homepage: www.else ie .com/loca e/sc
main aining he undi e en ia ed s a e, and plu ipo ency, which allows
hem o di e en ia e in o almos all cell ypes. ESCs a e able o di e en i-
a e in o all de i a i es o he h ee p ima y ge m laye s (ec ode m,
endode m, and mesode m). Plu ipo ency dis inguishes ESCs om he
s em cells (SCs) in an adul o ganism, hus adul SCs a e mul ipo en
and di e en ia e in o a limi ed numbe o cell ypes. Fu he mo e, ESCs
in cul u e can di e en ia e spon aneously and gi e ise o cell ypes ha
de i e om di e en p ima y ge m cell laye s (Hmadcha e al., 2009).
ESC di e en ia ion equi es he ep ession o ansc ip ion ac o s in-
ol ed in main aining plu ipo ency and he ac i a ion o de elopmen al
genes (Ho illo e al., 2013). Li le is known abou he ela ionship
be ween CREB/CBP and SCs. A p e ious s udy showed ha leukemia
inhibi o y ac o (LIF) in mouse ESCs p oduces LIF-dependen phospho -
yla ion o CREB. This p ocess is pa ially unde he con ol o he RSK2
kinase and as a consequence CREB egula es plu ipo ency and su i al
genes in mouse ESCs (Boeu e al., 2001). CREB is also in ol ed in
di e en ia ion and su i al p ocess o neu al SCs in he e olu iona y
scale (Dwo kin and Hea h, 2007; Dwo kin e al., 2009). Dwo kin e al.
(2009) ha e shown a specific ole o CREB in mammalian emb yonic
neu ogenesis. Ano he s udy associa ed CREB wi h ea ly neu ogenesis
(Pel ie e al., 2007) bu p esen ed only indi ec e idence o PI3K/Ak
s imula ion ac i a ing bo h CREB and neu ogenesis. Phospho yla ion o
CREB a se ine 133 (Se 133) is a equi ed s ep o inducing he ansc ip-
ion o mul iple genes h ough he cAMP esponse elemen (CRE).
Se e al kinases possess he capabili y o phospho yla ing CREB a his
si e, including p o ein kinases A and C (Manie e al., 2001), MAP kinase
ac i a ed p o ein kinase-2 (Xing e al., 1998), and Ak (Pel ie e al.,
2007; Ca a a a e al., 2008). In pa icula , he kinase Ak and i s ac i a o
PI3K play a significan ole in mul iple cellula unc ions, such as cell su -
i al, p oli e a ion, and cy oskele al ea angemen s (Can ley, 2002;
Engelman e al., 2006; Xue and Hemmings, 2013). While he ole o he
PI3K/Ak signaling pa hway in he unc ion o adul neu al s em o p o-
geni o cells is s ill unclea , i s ole du ing de elopmen has been widely
demons a ed. Fo ins ance, Ak is implica ed in main aining sel -
enewal o emb yonic co ical p ogeni o s (Sino and Lillien, 2004)and
ESC lines (Paling e al., 2004; Wa anabe e al., 2006). Fu he mo e, he
PI3K an agonis phospha ase and ensin homolog (PTEN) has been
shown o nega i ely egula e p oli e a ion o emb yonic neu al SCs
(G osze e al., 2006). The ac i a ion o he PI3K/Ak signaling pa hway
is also associa ed wi h neu ogenesis in p ogeni o cells de i ed om
he sub en icula zone and in ol ac o y SCs (Voj ek e al., 2003). In
line wi h hese epo s, Ak 3-deficien mice showed educed b ain size
and weigh (Tschopp e al., 2005), e ealing ha Ak is c ucial o he co -
ec b ain de elopmen . All he abo e demons a e a mechanis ic link
be ween PI3K/Ak , CREB, and neu ogenesis.
ESCs and cance cells sha e sel - enewal capaci y, which allows
hem o be indefini ely expanded. Mo eo e , he ecen ecogni ion
ha CREB may ha e oncogenic p ope ies leads us o hypo hesize ha
o e -ac i a ion o CREB could con ibu e o he de elopmen o some
ypeso umo s(Ghosh e al., 2007; Ab amo i ch e al., 2004; Linne h
e al., 2005; Shanka e al., 2005). This sugges s ha CREB could ha e a
ole in egula ing genes ela ed o sel - enewal. A a mechanis ic le el,
CREB is pe haps one o he bes -unde s ood phospho yla ion-
dependen ansc ip ion ac o s. In p inciple, phospho yla ion o CREB
a Se 133 seems o be su ficien o p omo e a ge gene ac i a ion
h ough ec ui men o CBP
14
. On he o he hand, unphospho yla ed
CREB in e ac s wi h DREAM o abolish he CREB–CBP in e ac ion in a
Ca
2+
-dependen manne (Ledo e al., 2002; Fon an-Lozano e al., 2009).
CBP is implica ed in di e en biological unc ions wi h an essen ial
ole in main aining no mal ge m cells de elopmen . CBP is exp essed
in p imo dial ge m cells (PGCs) and in ESCs bu is highly exp essed in
PGCs compa ed wi h ESCs. P ope le els o CBP ha e been shown o
be impo an o hema opoie ic SCs; CBP deficien hema opoie ic SCs
ail o sel - enew (Rebel e al., 2002). In e es ingly, he dele ion o
CREB, which could a ec CBP binding o DNA-associa ed ansc ip ion
ac o s, esul s in apop osis in nume ous cell ypes (Zhang e al., 2002;
Ba on e al., 1996; Jawo ski e al., 2003; Reusch and Klemm, 2002;
Dwo e and Meinko h, 2006; Scobey e al., 2001).
In con as , only wo s udies associa e DREAM wi h SCs. Fi s , Sanz
e al. (2001) ound ha DREAM egula es he exp ession o he apop o-
ic p o ein H k wi hin he hema opoie ic sys em. Second, Cebolla e al.
(2008) implica ed DREAM in he di e en ia ion o neu al p ogeni o s
in o as ocy es. Mo e ecen ly, ou eam desc ibed ha he loss o
DREAM p o ec s he b ain om degene a ion du ing aging (Fon an-
Lozano e al., 2009). DREAM could be di ec ing a deple ion o s em/
p ecu so cell ese oi s. This may indica e a possible ela ionship
be ween DREAM and SCs; ou esul s unco e a hi he o unknown
ole o DREAM in egula ing human ESCs (hESCs) plu ipo ency and
di e en ia ion.
2. Ma e ials and me hods
2.1. E hics o he hESCs
The HS-181 hESC line was ob ained om he Ka olinska Ins i u e
cou esy o D . Ou i Ho a a. The cell line was de i ed om a no mal
and heal hy blas ocys dona ed o esea ch in acco dance wi h he
legal equi emen s o he coun y o o igin and he dono s ga e w i en
in o med consen . The HS-181 line is included in he Eu opean Union
hESC egis y (h p://www.hesc eg.eu/).
2.2. Cell line cul u e
HS-181 was cul u ed, as desc ibed by Ho a a e al. (2003), on o BD
Ma igel™hESC-Qualified Ma ix coa ed flasks (BD Biosciences, San
Diego, CA, USA), in human eede -condi ioned medium consis ing o
knockou DMEM (Gibco, G and Island, NY, USA) supplemen ed wi h
20% se um eplacemen (Gibco), 2 mM L-glu amine (Gibco), 1% non-
essen ial amino acids (Gibco), 50 U/ml penicillin (Gibco), 50 μg/ml s ep-
omycin (Gibco), 0.1 mM β-me cap oe hanol (Gibco), and 4 ng/ml basic
fib oblas g ow h ac o (bFGF) (R&D Sys ems, Minneapolis, MN, USA).
To p omo e in i o spon aneous di e en ia ion, undi e en ia ed colo-
nies (hESCs a day 0: 0 d) we e de ached by ea men wi h Accu ase
(Gibco, G and Island, NY, USA), bFGF was wi hd awn om he medium,
and cells we e subjec ed o emb yoid body (EB) o ma ion using he
hanging d op me hod o 48 h. Cells we e hen pla ed and incuba ed as
floa ing agg ega es o 13 days (hESCs a day 15: 15 d) in ul a-low
a ach-flasks wi hou bFGF.
Human adipose-de i ed s omal cells (hASCs) (ATCC®, Manassas,
VA, USA; PCS-500-011™) and human bone ma ow-de i ed s omal
cells (hBMSCs) (Lonza, Basilea, Swi ze land; PT-2501) we e pu chased
and used in expe imen s because hey a e known o con ain subse s
o s em cells. hASCs and hBMSCs we e cul u ed in low glucose DMEM
supplemen ed wi h 10% FBS and we e cha ac e ized by hei su ace
ma ke s (CD13, CD29, CD31, CD34, CD45, CD73, CD90, HLA-II, and
CD105) and di e en ia ion po en ial a e cell expansion (da a no
shown).
The human newbo n o eskin fib oblas s (hFFs) (ATCC® CRL-2429)
we e cul u ed in DMEM supplemen ed wi h 10% FBS.
2.3. Quan i a i e RT-PCR
To al RNA was ex ac ed wi h TRIzol® Reagen (In i ogen,
Ca lsbad, CA, USA) acco ding o he manu ac u e 's ins uc ions. cDNA
was syn hesized om 2 μg o al RNA using MMLV e e se ansc ip ase
(P omega, Madison, WI, USA). Quan i a i e eal- ime RT-PCR (qRT-
PCR) analysis was pe o med on an ABI 7500 FAST Sequence De ec ion
Sys em (Applied Biosys ems, Fos e Ci y, CA, USA) using he SensiFAST
SYBR Lo-ROX Ki (Bioline, London, UK). The exp ession le el o each
gene was no malized o TATA box binding p o ein (TBP) mRNA le els
(D101-D136, TATAA Biocen e AB, Gö ebo g, Sweden). The esul s
we e exp essed using he ΔΔC me hod as he numbe o he old
569A. Fon án-Lozano e al. / S em Cell Resea ch 16 (2016) 568–578
change ela i e o he exp ession o con ol samples. P ime sequences
a e lis ed in Supplemen a y Table S1.
2.4. Flow cy ome y
2.4.1. Cell p epa a ions
Undi e en ia ed cells (0 d) we e disagg ega ed by incuba ion in
Accu ase (5 min, 37 °C). The eac ion was s opped wi h human
eede -condi ioned medium and cells we e hen cen i uged and
esuspended in PBS. Di e en ia ed cells (15 d) we e disagg ega ed by
incuba ion in ypsin (5 min, 37 °C). The eac ion was s opped wi h
media supplemen ed wi h FBS and cells we e hen cen i uged and
esuspended in PBS. A single-cell suspension was ob ained.
2.4.2. Cell cycle analysis
Cells we e fixed and pe meabilized o e nigh a −20 °C in cold 70%
e hanol. Cells we e collec ed by cen i uga ion and esuspended in PBS
wi h 1% FBS and 10 mM HEPES. Samples we e labeled wi h p opidium
iodide (PI) and subjec ed o analysis using FL-2A o sco e he DNA
con en o he cells' PI Fluo escence (FL2) using a flow cy ome e (BD
FACSCalibu Cy ome y Sys em, San Jose, CA, USA). Da a we e p ocessed
wi h CellQues P o 5.2.1. so wa e (BD, San Jose, CA, USA) and
analyzed in he ModFi p og am.
2.4.3. Apop osis analysis
To analyze he apop osis le els p o oked by DREAM inhibi ion, we
used he FITC Annexin V Apop osis De ec ion Ki I (BD Pha mingen).
Analysis was pe o med using flow cy ome y, conside ing PI
Fluo escence (FL2) and Annexin V-FITC fluo escence (FL1) in he flow
cy ome e (BD FACSCalibu Cy ome y Sys em). Da a we e p ocessed
wi h CellQues P o 5.2.1. so wa e (BD). We calcula ed he pe cen age
o cells a each s age o apop osis. Ea ly apop osis was defined as
cells ha a e posi i e o FITC Annexin V and nega i e o PI (Annexin
V
+
/PI
−
); nec osis was defined as cells ha a e nega i e o FITC Annexin
V and posi i e o PI (Annexin V
−
/PI
+
); la e apop osis was defined as
cells ha a e posi i e o FITC Annexin V and PI (Annexin V
+
/PI
+
).
To al apop osis was exp essed as he sum o ea ly and la e apop osis.
2.4.4. SSEA4 and TRA-1-60 exp ession analysis
Cells labeled wi h he fluo escen an ibodies FITC Mouse an i-
Human TRA-1-60 An igen (BD Pha mingen) o FITC Mouse an i-SSEA-
4 (BD Pha mingen) we e analyzed using Flow Cy ome y (BD
FACSCalibu Cy ome y Sys em). The da a we e p ocessed wi h
CellQues P o 5.2.1. so wa e (BD).
2.5. Human plu ipo en s em cell p o eome a ay
The p o ein exp ession p ofiles we e assayed using he specific
Human Plu ipo en S em Cell A ay Ki ‘P o eome P ofile A ay’(R&D
Sys ems Eu ope, Abingdon, UK; Ca alog No. ARY010) ollowing he
manu ac u e 's ins uc ions. This a ay allowed us o simul aneously
de ec 15 di e en SCs ma ke s ela ed o plu ipo ency and di e en ia-
ion. The densi ome y analysis o he esul ing spo s was ca ied ou
using ImageJ 1.41 so wa e (Wayne Rasband, NIH, Be hesda, MD, USA)
and exp essed as a bi a y uni s (a.u.) o spo densi y.
2.6. Immunofluo escence
Cells we e cul u ed in Ma igel™-coa ed con ocal pla es. Cells we e
fixed o 20 min in 4% pa a o maldehyde solu ion, washed h ee imes
wi h 0.05% PBS-Tween, and pe meabilized o 1 h wi h PBS con aining
0.1% T i on-X 100. A e 1 h o blocking incuba ion wi h PBS supple-
men ed wi h 5% goa se um and 0.05% Tween, cells we e incuba ed
wi h he p ima y an ibody dilu ed on blocking solu ion o e nigh a
4 °C. Then, cells we e washed h ee imes wi h PBS and incuba ed
wi h seconda y an ibodies o 1 h a oom empe a u e. Nex , cells
we e washed h ee imes wi h PBS and incuba ed wi h 1 mg/ml
Hoechs 33342 (DNA dye) o 5 min a RT, and finally washed h ee
imes wi h PBS. Digi al images we e ob ained using a Leica SP5 con ocal
mic oscope (Leica, Mannheim, Ge many) o an Olympus IX71 in e ed
mic oscope (Olympus, Tokyo, Japan). The images we e quan ified by
ImageJ 1.41 so wa e (Wayne Rasband, NIH, Be hesda, MD, USA). The
an ibodies used in his s udy a e lis ed in Supplemen a y Table S2.
2.7. Wes e n blo and immunoblo ing
Ha es ed cells we e lysed o p o ein ex ac ion using RIPA bu e
(Sigma-Ald ich), supplemen ed wi h p o ease and phospha ase inhibi-
o cock ail. Then, p o eins om whole cell lysa es we e esol ed
using 10% T is-Glycine gel elec opho esis and ans e ed o a
poly inylidene difluo ide memb ane (GE Heal hca e Li e Sciences,
Buckinghamshi e, UK). Memb anes we e hen blocked wi h 5% non-
a milk and p ima y an ibodies we e p obed (see Supplemen a y
Table S2 o an ibody in o ma ion). De ec ion was pe o med wi h he
app op ia e ho se adish-pe oxidase conjuga ed seconda y an ibodies
(Supplemen a y Table S2) using he enhanced chemiluminescence
eagen (GE Heal hca e Li e Sciences). Densi ome y analyses o he
blo s we e pe o med using ImageJ so wa e ( e sion 1.4 ; Na ional
Ins i u e o Heal h, Be hesda, MD, USA).
2.8. Small in e e ing RNA (siRNA)
Non-specific con ol siRNA (San a C uz Bio echnology, sc-37007,
San a C uz, CA, USA) and a ecommended DREAM siRNA (San a C uz
Bio echnology, sc-42398) we e used o ensu e silencing o DREAM ex-
p ession. Cell ans ec ions we e ca ied ou wi h siRNA and Lipo ec a-
mine 2000 Reagen (Li e Technologies, Ca lsbad, CA, USA) ollowing he
manu ac u e 's ins uc ions. A e es ing di e en concen a ions o
DREAM siRNA (20, 40, and 80 pmol), he concen a ion o 40 pmol was
selec ed as e ec i e o DREAM deple ion because 90% o inhibi ion was
de ec ed by qRT-PCR 24 h a e ans ec ion o hESCs (da a no shown).
2.9. S a is ical analysis
All expe imen s we e epea ed a leas in iplica e (n ≥3 indepen-
den expe imen s) and da a we e analyzed using he SPSS package o
Windows (SPSS, Chicago, IL). Values a e p esen ed as he mean ±
SEM. S a is ical significance was calcula ed using ANOVA o - es . The
Mann–Whi ney Rank Sum Tes was used in samples ha we e no
no mally dis ibu ed o ha had unequal a iance.
3. Resul s
3.1. DREAM is exp essed in human emb yonic s em cells
The lack o DREAM is implica ed in he delay o b ain aging and in
p e en ion o he loss o SCs wi h age. In his s udy, he esul s demon-
s a ed ha DREAM CREB and CBP mRNA a e exp essed in hESCs, and
ha he exp ession o DREAM and CBP inc eased when cells ini ia ed
spon aneous di e en ia ion (Fig. 1A). In con as , no changes in gene
exp ession o CREB we e de ec ed and Oc 4 exp ession was dec eased
when cells we e cul u ed unde condi ions ha p omo e di e en ia ion.
We also de ec ed exp ession o DREAM, CREB,and CBP in he o he e al-
ua ed cell ypes. Oc 4 was only exp essed in undi e en ia ed hESCs
(0 d) and in hFFs (Fig. 1A). DREAM signaling and i s ela ed p o eins
we e u he analyzed; immunofluo escence esul s e ealed ha
hESCs (0 d), hASCs, hBMSCs, and hFFs we e posi i e o DREAM, CREB,
pCREB, and CBP (Fig. 1B). The exp ession o DREAM coincided wi h a
hESC specific ma ke , TRA-1-60 (Fig. 1C). Un o una ely, despi e using
di e en an ibodies, we ha e been unable o de ec DREAM by Wes e n
blo . Toge he , hese da a sugges ha DREAM and i s ela ed genes a e
exp essed in di e en cell ypes and indica e a possible ole o DREAM in
570 A. Fon án-Lozano e al. / S em Cell Resea ch 16 (2016) 568–578
hESCs when hei plu ipo ency is diminished and cells begin o
di e en ia e.
3.2. Spon aneous di e en ia ion o hESCs a ec s DREAM exp ession
We e alua ed he exp ession o DREAM and i s ela ed p o eins du -
ing spon aneous di e en ia ion. We fi s e alua ed changes in DREAM
gene exp ession by qRT-PCR in undi e en ia ed (0 d) and spon aneous-
ly di e en ia ed (15 d) hESCs (Fig. 2A). We obse ed a clea inc ease o
DREAM and CBP and no significan changes o CREB exp ession in
di e en ia ed cells compa ed wi h undi e en ia ed cells. Mo eo e , o
confi m hese changes we pe o med immunofluo escence in bo h
g oups o de ec he DREAM p o ein and i s ela ed p o eins, CREB,
pCREB, and CBP (Fig. 2B), and quan ified he fluo escence in ensi y
(Fig. 2C). We obse ed ha DREAM and pCREB p o ein le els signifi-
can ly inc eased in spon aneous di e en ia ion (15 d) hESCs compa ed
wi h undi e en ia ed (0 d) cells, while CREB and CBP exp ession
dec eased. We no ed ha mRNA le els o CREB and CBP we e no
eflec ed in he fluo escen in ensi y as a su oga e o p o ein exp es-
sion, which could be caused by pos - ansla ional e en s. To confi m
ha di e en ia ion p o okes an inc ease in pCREB exp ession, we
analyzed he phospho yla ion o CREB a Se 133 by Wes e n blo
and ound a clea inc ease in pCREB (Se 133) in spon aneously di e en-
ia ed cells (Fig. 2D). These esul s sugges a possible ole o DREAM
in egula ion o plu ipo ency and di e en ia ion, possibly h ough
modula ion o CREB–CBP.
3.3. DREAM deple ion in hESCs (DREAM-siRNA)
The sligh inc ease in DREAM obse ed upon hESCs en y in o di e -
en ia ion may sugges a ole o his p o ein in di e en ia ion. We
p oceeded o ansien ly inhibi DREAM exp ession using small in e e -
ing RNA (siRNA) agains DREAM (a pool o 3 a ge -specific19–25 n
siRNAs designed o knock down DREAM gene exp ession) and s udied
he knockdown e ec s 24 h a e ans ec ion in spon aneously di e en-
ia ed cells. As shown in Fig. 3A, we ob ained a educed exp ession o
DREAM mRNA in DREAM-siRNA ans ec ed cells compa ed wi h
Con ol cells and cells ea ed wi h Con ol-siRNA (a sc ambled
sequence ha does no lead o specific deg ada ion o any known
cellula mRNA). To confi m he inhibi ion o DREAM, we de ec ed i s
exp ession by immunofluo escence and quan ified he esul ing images
(ImageJ so wa e). Immunofluo escence images e ealed a weak
s aining o DREAM and he SCs ma ke TRA-1-60 in DREAM-siRNA
cells (Fig. 3B). Quan ifica ion e ealed a significan dec ease in o al in-
ensi y o DREAM s aining in he ans ec ed g oup (DREAM-siRNA)
compa ed wi h o al in ensi y o Con ol and Con ol-siRNA s aining
(Fig. 3C). As DREAM is a mul i unc ional p o ein wi h specific oles in
di e en cell compa men s, we e alua ed di e en ial DREAM exp es-
sion in he cy oplasm and nuclei by analyzing he immunofluo escence
con ocal images. Fo cells ans ec ed wi h DREAM-siRNA, we ob ained
a significan dec ease in nuclea and cy oplasmic p o ein fluo escence
in ensi y (exp ession) compa ed wi h Con ol and Con ol-siRNA cells
(Fig. 3D). These esul s confi m he e ficacy o DREAM-siRNA used in
he ansien inhibi ion o DREAM exp ession in hESCs.
3.4. Cellula e ec s o DREAM inhibi ion (DREAM-siRNA)
Cell cycle-con ol mechanisms a e unc ionally linked o sel -
enewal and di e en ia ion in hESCs. Consequen ly, we used flow
cy ome y o e alua e i DREAM-siRNA p o oked changes in a hESC
cell cycle. Fig. 4A ep esen s he pe cen age o cells in di e en cell
cycle phases: G1, S, G2m (G2 and mi osis) and apop osis in Con ol,
Con ol-siRNA, and DREAM-siRNA ans ec ed cells. No significan
di e ences we e ound in he cell cycle, and we only obse ed a
end owa d inc eased apop osis in DREAM-siRNA ea ed cells. As
DREAM p o ein has been conside ed p o-apop o ic (Lilliehook
e al., 2002), we also e alua ed i DREAM inhibi ion in hESCs could
induce apop osis using a FITC Annexin V Apop osis De ec ion Ki I,
which disc imina es ea ly apop osis om la e apop osis. Fig. 4B
ep esen s he pe cen age o cells in ea ly, la e, o o al apop osis
and nec osis o Con ol, Con ol-siRNA, and DREAM-siRNA cells.
We obse ed an inc ease o ea ly and o al apop osis in DREAM-
siRNA ans ec ed cells compa ed wi h Con ol and Con ol-siRNA
cells. These da a indica e ha DREAM inhibi ion could a ec hESC
su i al.
Fig. 1. DREAM p o ein is exp essed in s em cells: A. qRT-PCR analysis o ela i e gene exp ession le els o DREAM, CREB, CBP, and Oc 4 in undi e en ia ed hESC (0 d), spon aneously
di e en ia ed (15 d) hESCs, hASCs, hBMSCs, and hFFs. No e he s a is ically significan di e ence o undi e en ia ed hESCs (0 d) compa ed wi h di e en ia ed hESC (15 d) g oups o
DREAM and CBP (**p ≤0.01 and ***p ≤0.001). B. Immunofluo escence de ec ion o DREAM, CREB, pCREB, and CBP ( ed) exp ession in cul u ed hESCs, hASCs, hBMSCs, and hFFs. β-
ac in (g een) was used as a fiducial ma ke and nuclei we e s ained wi h Hoechs 33342 (blue). No e ha all ou ma ke s we e exp essed in all assessed cells ypes. The scale ba is
250 μm. C. Immunofluo escence analysis o DREAM ( ed) and TRA-1-60 (g een) exp ession in hESCs. Nuclei a e s ained wi h Hoechs 33342 (blue). The scale ba is 40 μm.
571A. Fon án-Lozano e al. / S em Cell Resea ch 16 (2016) 568–578
3.5. DREAM-siRNA induces modifica ions in hESC plu ipo ency
To es he e ec o DREAM inhibi ion upon hESC en y in o
di e en ia ion, we pe o med qRT-PCR o e alua e changes in plu-
ipo en gene ma ke s (Oc 4, Nanog, and Sox2) and di e en ia ion
gene ma ke s (Pax6, N g1, e bB2, e bB3, and e bB4) and s udied
he e ec s o siRNA ans ec ion in undi e en ia ed (0 d) and spon-
aneously di e en ia ed cells (15 d). As shown in Fig. 5A, plu ipo en
gene exp ession was dec eased in all cell g oups (Con ol, Con ol-
siRNA, and DREAM-siRNA) a e spon aneous di e en ia ion. In
con as , we obse ed ha DREAM-siRNA ans ec ion p o oked a
significan inc ease in he exp ession o he di e en ia ion gene
Pax6 in spon aneously di e en ia ed cells (15 d). Fu he mo e, we
e alua ed changes in N g1 and i s ecep o s (e bB2, e bB3, and
e bB4) and ound ha N g1 and e bB4 exp ession significan ly in-
c eased wi h ans ec ion (DREAM-siRNA) o di e en ia ed cells
(15 d). O e all, he exp ession o plu ipo ency ma ke s dec eased
in a DREAM-independen ashion and exp ession o di e en ia ion
ma ke s Pax6, N g1, and e bB4 was enhanced by DREAM knockdown
du ing di e en ia ion.
Fig. 2. Spon aneous di e en ia ion esul s in changes in DREAM exp ession le els: A. qRT-PCR analysis o ela i e gene exp ession le els o DREAM, CREB and CBP o undi e en ia ed
hESCs (0 d) and di e en ia ed hESCs (15 d). No e he s a is ically significan inc ease o DREAM and CBP in di e en ia ed hESCs (15 d) compa ed wi h undi e en ia ed hESCs (0 d)
(**p ≤0.01 and ***p ≤0.001). B. Di e en ial exp ession o DREAM, CREB, pCREB, and CBP ( ed) in cul u ed undi e en ia ed (0 d) and spon aneously di e en ia ed (15 d) hESCs. β-
ac in (g een) was used as a fiducial ma ke and nuclei we e s ained wi h Hoechs 33,342 (blue). The scale ba is 250 μm. C. Ba g aph ep esen ing he fluo escence in ensi y in
a bi a y uni s (a.u.) o immunos aining images in A. No e ha DREAM and pCREB le els we e significan ly inc eased in spon aneously di e en ia ed (15 d) hESCs compa ed wi h
undi e en ia ed (0 d) cells, while CREB and CBP le els we e dec eased (**p ≤0.01 and ***p ≤0.001). D. The phospho yla ion o CREB was analyzed by Wes e n blo in undi e en ia ed
(0 d) and spon aneously di e en ia ed (15 d) hESCs; GAPDH was used as he in e nal con ol. Ba g aph ep esen s he esul ing densi ome y quan ifica ion o Wes e n blo s
(***p ≤0.001).
572 A. Fon án-Lozano e al. / S em Cell Resea ch 16 (2016) 568–578

The analysis o p o ein exp ession le els wi h a hESC-specific
p o ein a ay (Fig. 5B) e ealed ha spon aneous di e en ia ion
(15 d) esul s in a dec ease in exp ession o OCT4, NANOG, SOX2,
GATA4, and OTX2 in all cell g oups (Con ol, Con ol-siRNA, and
DREAM-siRNA). In e es ingly, he e was a clea and significan
dec ease o p o ein exp ession o OCT4, NANOG, SOX2, SOX17,
OTX2, TP63/73L, GSC, SNAIL, and VEGF/R2 in DREAM-siRNA depen-
den manne . Howe e , when compa ing p o ein le els be ween
undi e en ia ed cells (0 d) and spon aneously di e en ia ed cells
(15 d), no significan di e ences o p o ein exp ession we e de ec ed
o E-CADHERIN, AFP, GATA4, FOXA2, IFP1, and HCG in all g oups
(Con ol, Con ol-siRNA, and DREAM-siRNA). The p o ein dec ease
induced by DREAM-siRNA in OCT4, NANOG and SOX2 shown in
Fig. 5Bwasno eflec ed in he lack o change in mRNA le els as
shown in Fig. 5A.
In addi ion, we e alua ed he pe cen age o cells ha exp ess SSEA4
and TRA-1-60 p o eins by flow cy ome y (Fig. 5C). Consis en wi h ou
esul s o o he plu ipo ency ma ke s (OCT4, NANOG, and SOX2), we
obse ed ha spon aneous di e en ia ion (15 d) esul s in a clea de-
c ease in exp ession o SEEA4 and TRA-1-60 in all cell g oups (Con ol,
Con ol-siRNA, and DREAM-siRNA). Rema kably, he e was a significan
dec ease in exp ession upon educ ion o DREAM (DREAM-siRNA).
These da a indica e ha he e is a a iable exp ession o plu ipo ency
and di e en ia ion ma ke s in a DREAM-independen ashion du ing
ea ly di e en ia ion o hESCs.
3.6. DREAM-siRNA induces changes in CREB ansc ip ion ac i i y
To de e mine whe he CREB is implica ed in inc eased exp ession o
some di e en ia ion ma ke s in hESCs ea ed wi h DREAM-siRNA
Fig. 3. siRNA-DREAM induces dec eased DREAM exp ession in hESCs: A. Rela i e gene exp ession analysis o DREAM 24 h a e ans ec ion o Con ol, Con ol-siRNA, and DREAM-siRNA
ans ec ed hESCs. DREAM exp ession was significan ly educed (***p ≤0.001). B. Immunofluo escence analysis o DREAM ( ed) and TRA-1-60 (g een) exp ession in Con ol, Con ol-
siRNA, and DREAM-siRNA ea ed hESCs 24 h a e ans ec ion. Nuclei a e s ained wi h Hoechs 33342 (blue). No e he significan dec ease in o al in ensi y o DREAM and TRA-1-60
in he ans ec ed g oup (DREAM-siRNA) compa ed wi h he Con ol and Con ol-siRNA g oups. The scale ba is 50 μm. C. Ba g aph ep esen ing he o al fluo escence in ensi y o
DREAM o he immunos ained images in B (***p ≤0.001). The esul s a e exp essed as op ical densi y (O.D.). D. Rela i e fluo escence in ensi y in a bi a y uni s (a.u.) o DREAM
p o ein signal in nuclei and cy oplasm compa men s o he immunos aining images in B. The e we e s a is ically significan di e ences be ween condi ions and g oups (***p ≤0.001
and
+++
p≤0.001, espec i ely).
573A. Fon án-Lozano e al. / S em Cell Resea ch 16 (2016) 568–578
du ing di e en ia ion, we fi s e alua ed changes in ela i e gene ex-
p ession o DREAM, CREB, and CBP in undi e en ia ed (0 d) Con ol,
Con ol-siRNA, and DREAM-siRNA cells compa ed wi h di e en ia ed
(15 d) Con ol, Con ol-siRNA, and DREAM-siRNA cells (Fig. 6A). We
obse ed ha DREAM exp ession inc eased in Con ol and Con ol-
siRNA cells wi h di e en ia ion (15 d), bu , as expec ed, DREAM-
siRNA esul ed in a dec ease in DREAM exp ession in di e en ia ed
cells. We also obse ed an inc ease in CBP gene exp ession in di e en i-
a ed cells (15 d) bu no o cells ea ed wi h DREAM-siRNA. A non-
significan inc easing end o CREB gene exp ession was de ec ed
be ween undi e en ia ed and di e en ia ed cells o all condi ions.
We also examined he phospho yla ion o he CREB p o ein in undi -
e en ia ed cells (0 d) and di e en ia ed cells (15 d). Immunofluo es-
cence esul s demons a ed ha he CREB signal was simila o all
condi ions (Fig. 6B); howe e , pCREB significan ly inc eased when
cells ini ia ed spon aneous di e en ia ion and in e es ingly ansien
inhibi ion o DREAM (DREAM-siRNA) significan ly inc eased pCREB
(Fig. 6C). Taken oge he , hese esul s indica e ha he ansien inhibi-
ion o DREAM in di e en ia ed hESCs induces an inc ease o CREB
phospho yla ion.
4. Discussion
Aging is a p ocess ha depends on di e se molecula and cellula
mechanisms, such as genome main enance and inflamma ion. Mecha-
nisms o main ain genomic s abili y a e hough o coun e ac he
aging p ocess, whe eas inflamma ion is conside ed a d i ing o ce o
human aging (T oen, 2003). O ganisms ha e di e en an i-aging mech-
anisms o main ain genome in eg i y, such as DNA epai and cell cycle
con ol, and o emo e and ecycle hea ily damaged cells om he body,
such as apop osis and cellula senescence. E en i hese mechanisms
may be e y e ficien , hey canno cope wi h all genomic damage,
leading o a g adual accumula ion o DNA damage and mu a ions, hus
con ibu ing o o ganismal aging (Ga inis e al., 2008). DREAM is
conside ed a p o-inflamma o y gene (Ti uppa hi e al., 2014) and has
a ole in he modula ion o inflamma o y pain (Cheng e al., 2002; Jin
e al., 2012), sugges ing a possible implica ion in he aging p ocess.
Mo eo e , in an in i o s udy (18-mon h-old d eam
−/−
mice), we
ound ha DREAM is a key egula o o memo y and b ain aging; hus,
DREAM could be a po en ial a ge o no only educe pain bu also
o con ol decline o neu ons in aging (Fon an-Lozano e al., 2009),
indica ing a possible ela ionship be ween DREAM and SCs. Howe e ,
he ex i o da a p esen ed he e indica es ha loss o DREAM enhanced
he loss o s em cell cha ac e is ics and inc eased he le el o some
di e en ia ion ma ke s.
Ch onic inflamma ion has been associa ed wi h age- ela ed decline
in he unc ion o hema opoie ic SCs and issue-specific s em/p ogeni o
cells (Chambe s e al., 2007; Leppe dinge , 2011) and has been impli-
ca ed as a media o o almos all aging-associa ed diseases (Gadalla
e al., 2015), such as ascula diseases, diabe es, neu odegene a i e
diseases, and cance (Medzhi o , 2010; F eund e al., 2010; De
Ma inis e al., 2005; Sa ka and Fishe , 2006). Aging is accompanied
by a p og essi e decline in SC unc ion, which esul s in less e ec i e
issue homeos asis and epai in mammals (Jones and Rando, 2011;
Capilla-Gonzalez e al., 2013). Fo hese easons, we checked o a
possible ole o DREAM in SCs. As a fi s app oxima ion, we de ec ed
DREAM exp ession in di e en cell ypes, including hESCs, hASCs, and
hBMSCs. DREAM in e ac ed wi h CREB in a Ca
2+
-dependen manne ,
p e en ing CBP ec ui men . We likewise de ec ed exp ession o CREB
and CBP in hESCs, hASCs, and hBMSCs. All hese findings poin owa d
apossible oleo DREAMinSCbiology.
Aging o SCs leads o impai ed sel - enewal and abe an di e en i-
a ion po en ial (Jones and Rando, 2011). Mechanisms con olling sel -
enewal and he p oli e a i e capaci y o SCs a e essen ial o main ain-
ing unc ional SCs du ing aging. I emains unknown whe he declines
in SCs unc ionali y du ing aging influences o ganismal longe i y. To
elucida e he ole(s) o DREAM in SC plu ipo ency, we s udied he inhi-
bi ion e ec s o DREAM on spon aneous di e en ia ion o hESCs. Du ing
spon aneous di e en ia ion, he exp ession o DREAM inc eased, no
changes we e obse ed in CREB, and he e was an inc ease in he
exp ession o CBP. In addi ion, we de ec ed a dec ease in CBP p o ein
and an inc ease o Se 133-CREB phospho yla ion. I would appea ha
CREB is phospho yla ed in esponse o DREAM inhibi ion. Al hough
CBP ec ui men is known o equi e phospho yla ion o Se 133 in
Fig. 4. Cellula e ec s o siRNA-DREAM: A. Ba g aph ob ained om cell cycle analysis o Con ol, Con ol-siRNA, and DREAM-siRNA ans ec ed hESCs showing he pe cen age o cells in
di e en phases o cell cycle (G1, S, G2m [G2 and mi osis], and apop osis). B. Flow cy ome y analysis o he pe cen age o cells wi h ea ly, la e, o o al apop osis and nec osis o Con ol,
Con ol-siRNA, and DREAM-siRNA ans ec ed hESCs. No e he s a is ically significan inc ease o ea ly and o al apop osis in DREAM-siRNA ans ec ed hESCs (*p ≤0.05).
Fig. 5. siRNA-DREAM al e s he exp ession o plu ipo ency ma ke s: A. Rela i e gene exp ession o plu ipo ency genes (Oc 4, Nanog, and Sox2) and di e en ia ion genes (Pax6, N g1,
e bB2, e bB3, and e bB4) in undi e en ia ed (0 d) and di e en ia ed (15 d) Con ol, Con ol-siRNA, and DREAM-siRNA ans ec ed hESCs. No e he s a is ically significan dec ease o
plu ipo ency genes in all di e en ia ed (15 d) cell g oups compa ed wi h undi e en ia ed (0 d) cell g oups and an inc ease o Pax6, N g1, and e bB4 in di e en ia ed (15 d) DREAM-
siRNA ans ec ed hESCs (***p ≤0.001; **p ≤0.01). B. Rela i e p o ein exp ession le els o plu ipo ency ma ke s in undi e en ia ed hESCs (0 d) and di e en ia ed (15 d) Con ol,
Con ol-siRNA, and DREAM-siRNA hESCs e alua ed by he human plu ipo en s em cell an ibody a ay. **p ≤0.01 deno es a s a is ically significan di e ence o OCT4, NANOG, SOX2,
SOX17, OTX2, TP63/73L, GSC, SNAIL, and VEGF/R2 p o eins. C. Quan ifica ion o he pe cen age o cells exp essing he su ace plu ipo en ma ke s SSEA4 and TRA-1-60 in
undi e en ia ed (0 d) and di e en ia ed (15 d) Con ol, Con ol-siRNA, and DREAM-siRNA ea ed cells. Significan educ ion in SSEA4 and TRA-1-60 exp ession was de ec ed in
DREAM-siRNA cells (**p ≤0.01).
574 A. Fon án-Lozano e al. / S em Cell Resea ch 16 (2016) 568–578
575A. Fon án-Lozano e al. / S em Cell Resea ch 16 (2016) 568–578
CREB, he e mus be addi ional mechanisms (pe haps u he co ac o
in e ac ions) ha a e also equi ed o CBP ec ui men (Zhang e al.,
2005; Ooi and Wood, 2008). Howe e , we mus keep in mind ha
CBP is essen ial in main aining no mal ge m cell de elopmen (Ellio
e al., 2007) and is necessa y o neu al di e en ia ion (Rebel e al.,
2002). These da a could sugges he exis ence o mechanisms con e -
gen on DREAM and CREB p o ein phospho yla ion, howe e he ole
o CBP wi hin he hESC di e en ia ion emains o be elucida ed.
Keeping SCs ou o he ac i e cell cycle phase and minimizing he
isk o DNA damage may be especially impo an in aging and has led
o a SC hypo hesis o aging (Sha pless and DePinho, 2007; Beausejou
and Campisi, 2006). We succeeded in ansien ly inhibi ing DREAM
exp ession 24 h a e siRNA ans ec ion. This inhibi ion o DREAM
does no a ec he cell cycle bu p o okes an inc ease in apop osis.
Taking his in o accoun , p og ammed cell dea h is a undamen al p o-
cess h oughou mammalian de elopmen (Jacobson e al., 1997), and
Caspase-3 has been shown o be c ucial o mouse ESC di e en ia ion
ia Nanog deac i a ion (Fuji a e al., 2008). These ac s con inue o
sugges a possible ole o DREAM in hESC su i al, cell p oli e a ion,
and di e en ia ion.
We ound ha he lack o DREAM was associa ed wi h a dec ease in
p o ein le els o plu ipo ency ma ke s and e ec ed di e en ia ion
ma ke s di e en ly. CREB signaling di ec ly egula es Pax6 and is
known o con ol neu ogenesis (He old e al., 2011). In ou case, an-
sien inhibi ion o DREAM in hESCs inc eased he exp ession o Pax6,
N g1, and e bB4 in spon aneously di e en ia ed cells and p omo ed
he phospho yla ion o CREB. These esul s sugges ha di e en ia ion
o hESCs could be ela ed o DREAM inhibi ion, which in u n ac i a es
CREB h ough CREB phospho yla ion.
To he bes o ou knowledge, his is he fi s s udy ha includes
DREAM in he ne wo k hie a chy con olling he plu ipo ency o hESCs.
Ou esul s p o ide new mechanis ic insigh s ega ding he unc ion o
DREAM du ing ea ly hESC di e en ia ion. Howe e , i is c ucial o dissec
he addi ional pe spec i es aside om gene exp ession o molecula
mechanisms unde lying he ole o DREAM in cellula plu ipo ency.
Eme ging s udies show ha pos - ansla ional modifica ions influence
Fig. 6. siRNA-DREAM induces changes in CREB ansc ip ion ac i i y: A. Rela i egene exp ession o DREAM, CREB, and CBP genes in undi e en ia ed (0 d) and di e en ia ed (15 d) Con ol,
Con ol-siRNA, and DREAM-siRNA ans ec ed cells. **p ≤0.01 and ***p ≤0.001 indica e a s a is ically significan di e ence. B. Rep esen a i e immunofluo escence images o CREB p o ein
de ec ion in undi e en ia ed (0 d) and di e en ia ed (15 d) Con ol, Con ol-siRNA, and DREAM-siRNA ans ec ed cells. The scale ba is 50 μm. The quan ifica ion o fluo escence in ensi y
in a bi a y uni s (a.u.) o hese images showed non-significan di e ences. C. Rep esen a i e immunofluo escence images o pCREB p o ein de ec ion in undi e en ia ed (0 d) and
di e en ia ed (15 d) Con ol, Con ol-siRNA, and DREAM-siRNA ans ec ed cells. The scale ba is 50 μm. The quan ifica ion o fluo escence in ensi y in a bi a y uni s (a.u.) e ealed
ha pCREB was significan ly inc eased in spon aneously di e en ia ed (15 d) hESCs compa ed wi h undi e en ia ed (0 d) cells and in DREAM-siRNA ans ec ed hESCs (**p ≤0.01 and
***p ≤0.001).
576 A. Fon án-Lozano e al. / S em Cell Resea ch 16 (2016) 568–578