Proteomics Cardiac Metab

DOI: 10.1161/CIRCGENETICS.114.
000918
Proteomic Analysis of Human Pluripotent Stem Cell-Derived, Fetal and Adult

Ventricular Cardiomyocytes Reveals Pathways Crucial for Cardiac
Metabolism and Maturation
Running title: Poon et al.; Global proteomic profiling of hESC-VCMs
Ellen Poon, PhD1,2; Wendy Keung, PhD1,2; Y.M. Liang, MPhil3; Rajkumar Ramalingam, PhD3;
Bin Yan, PhD1,2; Shaohong Zhang, PhD4; Anant Chopra, PhD5,6; Jennifer Moore, PhD7;
Anthony Herren, BSc7; Deborah K. Lieu, PhD7.8; Hau San Wong, PhD
D9; Zhihui Weng, MB
MBBS1,2;
M
On Tik Wong, MS1,2; Yun Wah Lam, PhD3; Gordon F. Tomaselli, MD
D10; Christopher
Chri
Christ
ri stop
st ophe
op herr Chen,
he Chee MD,
Ch
PhD5,6; Kenneth R. Boheler, PhD1,2,10; Ronald A. Li, PhD
D1,
1,2,8
1,2,
2,88
2,
1
Stem Cell
elll & Regenerati
Regenerative
ivee M
Medicine
ed
dic
i in
inee Consortium,
Cons nso
ns m, 2De
ortiium Departments
Depar rtm
ment ntss off P
Physiology,
hyysio
ology y, LKS KS S Faculty
Faccul
ulty
t ofof Medicine,
Meedicinn
3
University off H
Hong
onng Kong; De
Department
epartme
meent ooff BBiology
iolog gy & Ch
Chemistry,
hem y, 9De
misttry Department
Dep parrtmmen nt of CComputer
omp mputter Sc
mp Science,
cien
ennce, Cit
City
ty Un
University
n
of Hong Kong,
o g, Hong
ong Ho o g; 4De
H ng Kong;
Kon Department
D epartmennt nt ooff Co
C
Computer
mp put
uter
er Sci
Science,
cien
cienncee, Gu
G
Guangzhou
angz
an g ho
gz h u UnUniv
University,
verrsity y, Gu
G
Guangzhou,
uan
angz
an gzh
gz hou,
u P
P.R.
.R
R. C
China;
5
Departmentt of Bioengineering
Bioengineering,
g, Bo
B
Boston University; 6Harvard
ston Un d Wyss Institute
Inssti
t tu
t te for Biologically
Biolo ogically Inspired Engin
Engineering,
n
7
A Department ooff Ce
Boston, MA;
A; Cell
ell BBiology
iol
olog
ogy & HumanH man
Hu man Anatomy,
ma An
A nat
atom
omy,
om y University
y, Univ verrsisity
ty ooff Ca
C
California,
liifo
lifo
forn ia,, Davis, CA; 8Cent
rnia
rn ia Center
t of
s
scular
Cardiovascular Research, Mount
Mountt S inai S
Sinai chhool off M
School eddic
i ine, N
Medicine, Newew York
York, k, N Y; 100Di
NY; Division
D viisiion of Cardiology, Johns
Jo
o
Hopkins
Hopk
Ho pk
pkin
kin University,
inss Un
Univ
n ver
ersi
sity,
sity
ty Baltimore,
y, Ba
B alt
lttim
ltim
imor
orre, M
ore, MD D
Correspondence:
Ronald Li, PhD
Center of Cardiovascular Research
Mount Sinai School of Medicine
Atran Berg Laboratory Building Floor 3rd, Room 323
1428 Madison Avenue
New York, NY 10029
Tel: 212-241-7369
Fax: 212- 241-4080
E-mail: ronald.li@mssm.edu
Journal Subject Codes: [105] Contractile function, [107] Biochemistry and metabolism
1
Downloaded from http://circgenetics.ahajournals.org/ at Emory University on March 17, 2015
DOI: 10.1161/CIRCGENETICS.114.000918
Abstract:
Background - Differentiation of pluripotent human embryonic stem cells (hESCs) to the cardiac
lineage represents a potentially unlimited source of ventricular cardiomyocytes (VCMs), but
hESC-VCMs are developmentally immature. Previous attempts to profile hESC-VCMs primarily
relied on transcriptomic approaches, but the global proteome has not been examined.
Furthermore, most hESC-CM studies focus on pathways important for cardiac differentiation,
rather than regulatory mechanisms for CM maturation. We hypothesized that gene products and
pathways crucial for maturation can be identified by comparing the proteomes of hESCs, hESC-
derived VCMs, human fetal (hF) and adult (hA) V and atrial (A) CMs.
Methods and Results - Using 2D-Differential-In-Gel Electrophoresis (DIGE),
(DIG
(D IGE)
IGE),, 12
E) differentially
1211 di
difff ren
ffer n
expressed (>1.5-fold, p<0.05) proteins were detected. The dataset implicated
ed a rrole
licated olee of th
ol thee
peroxisomee proliferator-activated
pro
roli
ro life
li fera
ferato
ra activated receptor alpha (PP
tor-ac
o ac (PPARA)
A A) signallingg in cardiac maturation
PPAR
PP maturation.
n
Consistently,
y, WY-14643,
WY-146433, a PPAR
PPARA
PP
PAR A agonist,
ARA agonisst, increased
gonis d ffatty
inccreassed ty ooxidative
attty
atty xiida
dati v eenzyme
tive
ti ve nzzym level,
ymee le
leve
veel,,
hyperpolarized
ized
ed m
mitochondrial
itocho
hondrriall m
ho membrane
em
mbr
branee po
potential
otenttiaal an
and
nd in
induced
nduce
ced a m
more
orre org
organized
rg
ganized morphology.
ed m
morpholo
orphholo
Along this line,
l treatment wi
with
with thyroid
t thee th
hyr hormone
y oidd ho
ormon triiodothyronine
o e triiod
doth
thyr
th o inee (T3)
yron 3) increased
(T3)
(T inccreased the dynamic
dyn
n
tension developed
v p d in engineered
velope engi
g neered hhuman
d hu ventricular
man vent riicula cardiac
l r ca microtissue
rddiac microti (hvCMT)
issue (h
hvC 3-folds,
CMT)) byy 3-fold
d
signifying their
the
heir
he maturation.
ir m atur
at urat
atio
tio
ion.
n.
Conclusions - We conclude that the PPARA and thyroid hormone pathways modulate the
metabolism and maturation of hESC-VCMs and their engineered tissue constructs. These results
may lead to mechanism-based methods for deriving mature chamber-specific CMs.
Key words: cardiac contractility and energetics, metabolism, proteomics, embryonic stem cell,
PPARA, PGC1A, Human embryonic stem cell derived cardiomyocytes, thyroid hormone
2
Introduction
Human (h) pluripotent stem cells such as embryonic stem cells (ESC) and induced pluripotent
stem cells (iPSC) self-renew; their differentiation to the cardiac lineage1-6 represents a potentially
unlimited source of ventricular (V) cardiomyocytes (CMs) for therapies and as experimental
models to investigate mechanisms involved in human cardiac development and disease
progression. Successful application of hESC-VCMs to myocardial repair or modeling requires
that the cells faithfully recapitulate the phenotype of adult CMs, especially in relation to their
electrophysiological, contractile and metabolic functions7. Indeed, hESC-CMs

SC-CMs are most ssimilar
i
im
to human embryonic or fetal CMs5, 8-11, and with a disorganised sarcomeric nt10
meric arrangement10,
0, 12
.
-baase
sedd ex
Microarray-based expe
perime
pe m nts have been previously
experiments ly pperformed
erformed to characterize
cha
harracterize the expressi
ha i
expression
profile of hESC-CMs,
ESC
C-CMs, pu
purified
uriified
d for
for vventricular
entrric
icuulaar li
lineage
ineeagge or no
not,
ot,
t ffor
or identifying
identtiffyin
in
ng signaling
sig
gnaliing pathways
pat
athw
w
implicated in
i thhei
eirr di
their enti ion13-17
ddifferentiation
fferrentiattion
tiat 17
17
.A
All
ll pr
profiling
rof
ofil
fil
ilin
i g at
in att
attempts
ttemp
temp
pts to
to date
datee almost
almo
lmost exclusively
excllussiv
ex
excl i el
ely
l relied
reli
rel e on
m data13-17 or ccell
transcriptomic
mic elll surf
surface
face pr i assessments188, but
pprotein
otein but the
th
he global
gllobball proteome
proteome of hE
hESC-
E
VCMs has nott been

be examined.
e amined
mii d F
Furthermore,
rthermore
th
h most
st hhESC
hESC-VCM
ESC
ES CV
VCM
CM studies
stt dies
di focus
foc
f s on pathways
path
th
h a s
important for cardiac differentiation, rather than mechanisms implicated in their developmental
maturation. In brief, previous results mostly indicate that hESC-CMs are relatively immature
compared to fetal and adult hearts, express lower levels of contractile and metabolic genes and
show differential expression of specific cardiac ion channels, complementary to known
functional data13-17. For obtaining a better understanding, here we employed a combinatorial
approach of performing proteomic, bioinformatics and functional analyses of hESCs, hESC-
derived VCMs, human (h) fetal (F, 18 weeks) and adult (A) VCMs to identify novel proteins and
pathways important for maturation.
3
Methods
Culturing of hESCs, cardiac differentiation and selection of hESC-VCMs
Undifferentiated hESC (HES2)19 were maintained at 37C and 5% CO2 on Matrigel (BD
Biosciences)-coated wells with mTeSR medium (Stem Cell Technologies). For cardiac
differentiation, growth factors were added at specific stages of differentiation20. The following
cytokines were used: days 01, BMP4 (0.5 ng/ml); days 1%03QJPO)*)QJPO
and activin A (3 ng/ml); days 48, DKK1 (150 ng/ml) and VEGF (10 ng/ml); after day 8, VEGF
(10 ng/ml), DKK1 (150 ng/ml) and bFGF (5 ng/ml). Cultures were maintained
aintained in a 5% CO2/5%
O2/90% N2 environment for the first 1012 days and were then transferred
erredd in
iinto
t a 55%
to % CO2//air
a
n After
nt.
environment. A te
Afterr ap
app
prox
oximately 12-14 days, card
ox
approximately dia
i c derivatives ma
cardiac ade
d up ~50% of the ce
made cell
population. For
Fo profiling ex
experiments,
xpeeriimeenntts, hhESC-VCMs
ESC
ES C-VCM
C s were
were
re pu
pur
purified
rified
ed usi
using
ing the
the LV-MLC2V-GFP-
LV-
V ML
V-MLC2
2V-G
G
T2A-ZEO repor
rreporter
rteer system
sy em tto
o av
avoi
o d am
avoid mbiguit
i ie
it ies due
ambiguities due to
to the
the presence
pre
ressenc
ncee off ccontaminating
nc onta
tami
taminati
mi ting
ti ng nnon-VCMs
on-V
on V
and non-cardiac
r
rdiac cells. We ddigested
ige
g sted
d 221-day
1-day
y old
ld hhESC-derived
ES
SC-
C derivedd cell
cells,
ls,, pplated
latedd th
them
hem onto matrigel-
matrig
g
coated surfaces
aaces and
d tr
transd
transduced
sdd ced
ed
d with
iith
thh recombinant
mbi
bin t LV
LV-MLC2V-GFP-T2A-ZEO
MLC2V
MLC2
MLC2VV GF
GFP
P T2
T2A
A ZE
ZEO
O particles
rtiiclle at an
MOI of 3. Transduced cells were cultured in 5% fetal bovine serum in CM media consisting of
DMEM supplemented with 0QRQHVVHQWLDODPLQRDFLGVP0JOXWDPLQH8PO
penicillin, 50 g/ml streptomycin (Invitrogen). Fluorescing GFP positive cells became visible
three days post-WUDQVGXFWLRQDQG]HRFLQJPOZDVDSSOLHGWRNLOOQRQ-VCMs. Please see
Figure S1 for characterisation of hESC-VCMs.
Isolation of fetal and adult CMs
Human fetal and adult CMs were isolated and experimented according to protocols approved by
the UC Davis IUPAC and IRB (Protocol #200614787-1 and # 200614594-1) (Table S1,
Supplemental Methods).
4
2D DIGE
Cells were lysed in 2-D lysis buffer. Each sample was labelled with CyDye-3 or CyDye-5 (GE
Healthcare/Amersham). As an internal standard, equal amounts of all cell lysates were mixed
and labelled with CyDye-2. CyDye-labeled protein samples were mL[HGDQGOZDVORDGHG
per IEF strip (GE Healthcare/Amersham). IEF was then done for a total of 25,000 V-h with
standard conditions using Ettan IPGPhore II. After the IEF, electrophoresis was performed at
16qC. The resulting 2-D gel was scanned using a Typhoon Trio scanner (GE
Healthcare/Amersham). Analyses were performed using the Profenesiss Samespot v2.3
(Nonlinear Dynamics). Selected protein spots were excised from the gel usi
using Ettan
ing E spot
pott picker
ttan spo
po pi
(GE Healthcare/Amersham),
h re
hcare
re/A
/Ame
/Amers
rsha digestion
s am), subjected to in-gel dige
gesstion and prot
ge protein
o einn IIDs
Ds were determinedd by
MS/MS analysis
alyssis using a MALDI-ToF/ToF
al MAL
LDI-To T F massspectrometer,
T F/To
To massssspecctrom
met
eter
e , AB
er SCIEX-4800
AB SC
CIEX
X-48000 (AB SCIEX).
AB SCI
CE
For more ddetails,

etails,
t please
pl methods.
see supplementall meth
hods.
d
Data analysis
ysiss
ysi
For hierarchical
hi l clustering,
l i we used
d average li
linkage
k bbased
d on E
Euclidean
lid distance to examine
di i
proteins with fold changes more than 1.5 (p<0.05). Principal component analysis of the
processed proteomic data was performed using the samespot software (Nonlinear Dynamics).
Functional annotation was performed based on Gene Ontology classifications (GO;
//www.geneontology.org). For differential protein abundance, proteins satisfying expression fold
difference of at least 1.5 and p value (t test) <0.05 were considered to be differentially expressed.
Pathway analysis of differentially expressed proteins
Differentially expressed proteins were imported into Ingenuity Pathway Analysis (IPA) 2013
(http://www.ingenuity.com/) to evaluate their association with corresponding canonical pathway
and upstream regulators. The enrichment analysis was performed by Fisher's right tailed exact
5
tests, where p value was below 0.05 based on false discovery rate (FDR) <5%.
Quantitative Real-time PCR (qRT-PCR)
CDNA was prepared using the QuantiTect Reverse Transcription Kit (Qiagen). QRT- PCR was
carried out using the KAPA SYBR FAST qPCR Kit (Kapa Biosystems) and gene expressions
were quantified using StepOnePlusTM Real-Time PCR system (Applied Biosystems). Primer
sequences are available upon request. Gene expression was normalized to GAPDH and is
presented as fold changes relative to DMSO treated control (meanSEM). A Students t-test was
employed to determine statistical significance. P values less than 0.05 were considered as
statistically significant.
Measurement
m t of m
ment mitochondriDOPHPEUDQHSRWHQWLDOP
ittoc
o hoondriDOPHPEUDQHSRWHQW
QWLD
QWL OP
PHDVXUHGXVLQJJ WKH ZHOOO--establ

b ishe
bl
PZDVPHDVXUHGXVLQJWKHZHOO-established
PH hed JC-1
JC-1 ddye
JC ye ((Invitrogen).
Invvitrog
In ogen). JJC-H[KLELWVP-
C-- H[KL P-
K ELWVV P
KL
P-&-11 forms
GHSHQGHQWDFFXPXODWLRQ$WKLJKP-&-1
D PXXOD
DFFXPX ODWL
WLRQQ$W
WL $W KLJK
$W LJK
P form
fo r s complexes
rm comp
comple
mp xess with
lexe
le
exe with intense
tense red
int
nte red fluorescence
d flu
uor
ores
esce
cenc
nc
&-11 remains
aQP$WORZP-&-1
$
$WORZP
-& i in the
remains he monomeric
th monomeriic form,, which
whi shows
hich sh y ggreen
hows only reen
IOXRUHVFHQFHaQP7KHUHIRUHWKHUDWLRRIUHGJUHHQVLJQDOLVDQLQGLFDWLRQRIP+(6&-

FH QP
7KHUHIRUH
7KH IR WK WKH
KH UDWLR
DWLLR RII UHGJUHHQ
G VLJQDO
VLL DOO LV
L DQ LQGLFDWLRQ
LQGL
GLFDWLLR RII P +(
+(
VCMs were incubated with 0.5M of JC-1 dye at 37C for 30 min in CM media. Cells were
imaged at green and red channels and signal intensities were analysed using ImageJ software to
FDOFXODWHPDVWKHUDWLRRIUHGJUHHQIOXRUHVFHQWVLJQDO
Quantification of contractile force in human ventricular cardiac microtissues (hvCMT)
HvCMTs were prepared as previously described21. In brief, hESC-VCMs were dissociated after
trypsin digestion. A cooled suspension of ~106 cells within reconstitution mixture, consisting of
1.5mg/mL liquid neutralized collagen I (BD Biosciences) and 0.5mg/mL fibrinogen (Sigma-
Aldrich), was added to the substrate and the entire assembly was centrifuged to drive the cells
into the micropatterned wells, where hESC-VCMs self-assembled into microtissues within 24 hr.
6
For quantifying microtissue forces, brightfield and fluorescence images were taken at 100Hz
with a fast CCD camera (Allied Vision), and an A-Plan 10X objective on a Nikon Eclipse Ti
(Nikon Instruments, Inc.) equipped with a live cell incubator. Only tissues that were uniformly
anchored to the tips of the cantilevers were included in the analysis. The displacement of
fluorescent microbeads at the top of the cantilevers was then tracked with using the SpotTracker
plug-in in ImageJ (National Institutes of Health). Microtissues were treated with the thyroid
hormone triiodothyronine (T3, 100nM) at the time of seeding for 6 consecutive days.
Measurements were made on day 6 of T3 treatment.
Measurements of action potential
Action potentials
e ial
enti a s in hhESC-VCMs
E C-
ES C VCMs treated with or without
with
wi t out T3 (100nM)
th M) for
for 6 days were analyzed
anall
by loading hhESC-VCMs
ES with
SC-VCMs with the
he vvoltage
h th
he oltage sensitive
nsittivve ddye
ge sen Di-8-ANEPPS
ye Di-88-A
AN S (5M)
NEPPS M) (Invitrogen)
(5 (In en) ffor 30
Inviitrrogen
min at 37C
C in DMEM/F12,
DME
MEM/
M/F1
M/ F12,
F1 2, followed
2, fol
ollo
llo imaging
l wedd by ima
magi
maging
gi
ing with
ithh a CMOS-based
n wit
it CMO
CM OS-basedd camera
b se
ba came
mera
me (MiCAM
ra (Mi
MiCA
Mi CAM
CAM
ULTIMA, SciMedia USA L

Ltd,
td
d, CA
CA,
A, US)
US) in
in Tyrodes
Tyr
y oddes
solu
solution
l tion containing
contain
i ingg (in
(iin mM): NaCl, 5
) 140 NaC
KCl, 1 CaCl
Cll2, 1 M
MgCl
gCl
Cl2, 10 glucose,
gll cose 10 HEPES,
H
HEPES
EPES
EP ES pH
H adjusted
adj
dj sted
tedd to 7.4
7 4 with
iith
thh NaOH.
NaOH
NaOH
OH Electrical
El triicall
stimulation was applied to evoke action potentials.
Results
Proteomic profiling was performed using 2D-Differential-In-Gel Electrophoresis (DIGE) on
protein extracts from undifferentiated hESCs, hESC-VCMs, hF-VCMs and hA-VCMs. This
technique was chosen such that protein abundance in multiple samples can be compared and
accurately quantitated. For comparison, human atrial (A) samples (i.e. hF-ACMs and hA-ACMs)
were also studied. Protein extracts were separated by 2D gels to detect individual protein spots.
As an example, Figure 1A shows the separation of proteins present in undifferentiated hESCs
and hESC-VCMs. Superimposition of the two images enabled the visualization of the differences
7
in protein expression: green and red protein spots corresponded to proteins that were over-
expressed in hESC and hESC-VCM cells, respectively. Yellow spots represented protein spots
that were equally expressed in both samples.
Figure 1B-C show the separation of proteins present in hF-ACMs, hF-VCMs, hA-ACMs
and hA-VCMs. Overall, a total of 554 individual protein spots were detected and matched to all
gels (Figure 2A). 316 spots were differentially expressed between any two samples (>1.5 fold,
p<0.05). 161 unambiguous IDs were obtained by mass spectroscopy, corresponding to 121
unique gene products. Cross analyses enabled us to examine the protein

in expression
exppression profile
ilee across
proffil
these samples. As an example (Figure 2B-C), the expression of MLC2V (MYL2)

2V (M
MYL
YL2)
2) was mo
most
robust in hA-VCMs
Ms and
A VCMs
A-VCM hF-VCMs but present at low
an hF lower hESC-VCMs
ower levels in hES
ow C VCMs and the atrial
SC- atriaa
samples (hF-ACMs
F-ACMs and hA-ACMs),
F- hA-A
hA ) and
ACMss), was
nd w virtually
ass vi absent
bseent in undifferentiated
irttualllyy ab undiff
ffeeren
ff entiated
en hESCs.
SCs. These
ed hES
ES Th
h
data were entirely

y consistent
entirely
ly consist entt with
stten wiith trends
treend
nds reported
d repo
port
po e ppreviously
rted
rt
ted slyy uusing
reviiouusl sin other
ng ot
the niiquues22
techniques
h r tech
chni
ch 22-25
2-25
25
,
validating the
t use of this approach
app
pproachh for
pp for id
iidentifying
entify
ify
f ingg novell proteomic
p oteomiic differences.
pr d ffferences.
di
Global proteomic
t i expression
i profiling
fili
fi li
We next performed hierarchical clustering and principal component analyses to examine the
proteomic landscape of our samples. Figure 3 shows that biologically independent replicates
clustered closely together, consistent with a high experimental reproducibility. Although
replicates of samples generally clustered according to their developmental stages and chamber-
specificities, the close distances between hF-VCMs and hF-ACMs, and between hA-VCMs and
hA-ACMs suggested that the proteomes of VCMs and ACMs from the same developmental
stages were similar. HESC-VCMs were preferentially grouped together with hF-VCMs and hF-
ACMs, indicating that the global proteome of hESC-VCMs was fetal-like, mirroring their
comparable global transcriptomes26.
8
Identification of proteomic changes in cardiac contraction and metabolism
To reveal specific protein changes associated with different stages of ventricular maturation, we
focused on the differential protein expression among hESC-VCM, hF-VCMs and hA-VCMs.
Among the proteins identified, 37 and 33 were significantly increased and decreased in hA-
VCMs compared to hF-VCMs, respectively, while 24 and 19 were significantly increased and
decreased in hF-VCMs compared to hESC-VCMs, respectively. Expression changes in
individual proteins are shown in Table S2. Table 1 summarizes the general trends in protein
expression among different functional groups in the three CM populations.

ions. Figure 4 shows
ows the
show
ow
spot intensities and topographical images of select key proteins for contraction, metabolism,
ntracttion, m ettabol
bol ism cell
olis
is
survival, ionn transport

trran
ansp ortt and
spor
sp nd regulation of mRNA and
an d protein
protein expression
on of CMs.
For ccontraction,
onntraction, as anticipated
as ant
ticcipaate
t d fr m ttheir
from
om mechanical
heiir levell off me
mech
han activities,
nical ac vities, hhESC-VCMs
ctiv ES VCM had
SC-V
the lowest OHYHOVRIVDUFRPHULFSURWHLQVVXFKDV-Myosin

O OV
OHYHOV
OV RI
I VDUFFRP
RPHU
HULF
L SURWHLQ
LF LQQV VXFK -M
FKK DV yosiin Heavy
Myo eavyy Chain
Hea Cha
h in
in 0<+DQG-actin
0 <+ DQGG -ac
0<+ actt
(ACTA1), followed
f hF-VCMs,
byy hF-VC s,, and
VCM andd then
h n hA-VCMs,
the hA-V
hA VCM ppotentially
CMs,, po tentia underlying
i llly unde
d rlyi
lyiingg the weaker
developed twitchh force

t iitch
tch f off engineered
gii d ventricular
ed entric tissue
triic llar ti
tiss triip that
i e strip thhat wee recentl
recently d27. F
tll reported
rted Fo
For
o
insights into cardiac metabolism, we next examined the abundances of proteins involved in FA
oxidation (e.g., very long-chain specific acyl-CoA Dehydrogenase, ACADVL) and glycolysis
(e.g., Cytosolic malate dehydrogenase, MDH1). These were significantly increased and
decreased, respectively, in hA-VCMs relative to hESC-VCMs and hF-VCMs whose levels were
comparable. 9 out of 33 known proteins involved in energy generation showed significant 1.6-
14.0-fold expression differences between hESC-VCMs and hF-VCMs. Indeed, glycogen
phosphorylase (PYGB), a rate-limiting step for glycogen breakdown, displayed the highest
difference by being >14-fold enriched in hF-VCMs compared to hESC-VCMs. In addition, we
also observed higher levels of proteins involved in energy transfer from the mitochondria to the
9
contractile apparatus and this includes four-and-a-half-LIM domain protein 2 (FHL2) (Figure 4),
creatine kinase (CKM) and mitochondrial CK (CKMT2) in hF-VCMs relative to hESC-VCMs.
CKMT2 protein level was increased further in hA-VCMs relative to hF-VCMs, consistent with
increased mitochondrial maturation.
Cardiac metabolism and maturation of hESC-VCMs
To seek further insights, we performed the complementary approach of combining clustering,
Ingenuity pathway and upstream regulator analyses of differentially expressed proteins, to
identify pathways or regulators important for CM maturation (Figure 5A-C; see Table S3A
S3A and
B for full details). IPA result shows that proteins differentially expressed hESC-VCMs,
sed among hE
hES
SC-V
C VC
hF-VCMs to
t hA-VCMs
hA-
A-VC s were
VCM we most significantly associated
ociated with metabolic
asssoc metab
abo ic processes (such as
bol
mitochondrial
r al dysfunction),
ria on)), raising
dysfunctio isingg the
raais iintriguing
the in trrig ng possibility
guing poossibi
bili
bi lity
li ty that metabolism
that m etab
abollissm an
ab andd CM
developmental
n ma
ntal maturation
m atu
turation
tu on ar related.
aree rel tedd. IPA
rela IP al so pprovides
also rov
ovid
id Upstream
des Upstr
ps treaam Re
tr Regulator
R torr Analysis
gulato
gu to Anal
allys i tto
ysis
is o
identify enriched
r
riched regulators
g lators that
regu thhat mayy be
b important
portant for
imp for regulating
gulatiing or aaffecting
reg fffectiingg gene/protein
g ne/p
ge protein
expression b
based
ed istii l analysis.
d on stati
statistical anal
anall sis
is Th
The id d regulators
identified
tifi
ified ato can bbe used
reg llators d tto predict
sed
ed dict whether
edi h
the regulatory cascades or target proteins are involved in the biological activities. As shown in
Figure 5C, we detected PPARGC1A (peroxisome proliferator-activated receptor gamma
coactivator 1-alpha)/ ESRRA(estrogen-related receptor alpha)/ PPARA (peroxisome proliferator-
activated receptor alpha) as a group of transcription factors (TFs) that regulates metabolic protein
expression, particularly those related to FA metabolism (e.g., ACADM and ACADVL etc), in
hESC-VCMs. Of the 11 proteins targeted by PPARGC1A, the transcripts of 9 were indeed
significantly upregulated in hA-VCMs and/or hF-VCMs relative to hESC-VCMs (Figure 5E).
Using the above pathway insights as a guide, the relationship between cardiac
metabolism and maturation of hESC-VCMs was further investigated. To seek functional
10
evidence, we next tested the pharmacological effect of WY-14643, a well-characterized PPARA
agonist28, RQP of hESC-VCMs using the JC-G\HP is an index of cellular metabolic
output and mitochondrial functionality and arises from the asymmetric distribution of charges
between the inner and outer sides of the inner mitochondrial membrane as a result of the electron
transport chain and its ATP production29. WY-14643 treatment significantly hyperpolarized
PE\LQK(6&-VCMs in the presence of oleic acid (p<0.05; Figure 6A); conversely,
WY-14643-treated hESC-VCMs cultured without oleic acid supplement had an Psimilar to
control (p>0.05). Along this line, WY-14643 (50M) significantly increased

reased the transcript
transcri
riipt
p levels
of genes involved in FA metabolism such as medium-/ long-/ very long-

g- chain
ch
haiin specific
specif
ifiic acyl-CoA
if acyl
n es ((ACADM,
nases
dehydrogenases ACAD
AC A M,
M ACADL and ACADVL)
L) ((Figure
Figure 6B). Alth
th
hough hESC-VCMs had
Although h
significantlyy lower
lo
ower levelss ooff bboth
oth ACADM
D (1.9
ACADM 0.1 fold)
(1.90.1 fold
d) and
and ACADVL
ACAD
AC DVL (3.80.6
(33.8
0.
0 6 fold)
foold))
o hA
compared to A-VC
VCM
VC Ms, WY-14643
hA-VCMs, WY-1146
WY 4 433 treatment
tre
reatment
ntt raised
rai
a seed AC
A ADVL
ADVL eexpression
ACADVL xpre
xp ress
ssio
i n to
t aduult llevels.
adult evel
els.
ls
mportant
mp
portant for FA transport,
CPT1B, important transpport,
t, was als
l o iincreased
also ncreasedd by
b 22.20.2
.220
0.22 ffold.
old.
d 1100M
00M
00 WY-14643
M
imilar
imil
il effects.
produced similar effects
ff ts CCollecti
oll
ll tii ell these
Collectively, th
h obser
ob
b ati
ations
io were
observations ere iin accordance
rdd ce with
iith
th
h a WY 1
WY-14643-
induced FA oxidation, which provides substrates for the electron transport chain. As for the
mitochondrial morphology, control hESC-VCMs had disorganised punctate and perinuclear
COXIV staining (Figure 6C). By contrast, WY-14643 treated hESC-VCMs showed stronger
COXIV staining. Furthermore, mitochondria became elongated and formed filamentous
networks between myofibril filaments, suggestive of a more developmentally mature
phenotype30. We also examined the staining pattern of FHL2, an anchoring protein that mediates
energy transfer between the mitochondria and the contractile apparatus31. Immunostaining of
hESC-VCMs indicated that WY-14643 improved the organisation and alignment of FHL2,
which colocalised ZLWK-actinin with a clearer striated staining pattern (Figure 6C). In addition,
11
the mitochondrial elongation factor GFM1 was also increased in WY-14643-treated cells relative
to control (Figure 6C).
Triiodothyronine (T3) treatment increases the contractile forces of engineered human
ventricular microtissues
Since PPARA/PPARGC1A is known to be regulated by or interact with the thyroid hormone
pathway32-35, which promotes the maturation of fetal VCMs and ESC-CMs36, 37, we then
examined the functional consequences of triiodothyronine (T3) treatment on contractile forces.
While shortening of single hESC-CMs or their clusters has been measured

ured as an surrog
surrogate
gat
atee iindex
for contractile forces38, 39, a multi-cellular 3D hvCMT system has beenn developed,
develo
l ped,
d where
whhere
re ttrue
r
n on developed
nsion
dynamic tension dev
evel
elop
el oped
op ed by the tissue in real time
me ccan d21
an be measured2
. Figure 7A shows tthat
hvCMT engineered
gineeered from aapproximately
gi pprroxim
matelyy 10
1000 hE
hESC-V
hESC-VCMs
VCM
CMss each
eaach
h ooff ~0.5mm
~0 mm inn lle
length
eng
ngtth aallowed
lllow
w
continuous measurement
measu
suure
rem
ment
ntt ooff th
their
eir
ei
ir sp
spontaneous
pon
onta
taneouus ass wel
ta well
elll aass stim
st
stimulated
timul
imul
ulat
l ted
d dynamic
dyn
ynaamic
ic twitch
twi
w tcch tension
tensiion
tens ion after
T3 (100nM)
M) treatment for 6 days.
M) d ys
da y . For
F r time-matched
Fo time-match
hed
d hhvCMTs
vC
CMT
Ts treated
d wi
with
ith
h T3,
T3,
3 the developed
developpe
twitch tension
iion increased
i d significantly
ed ifi ntll (by
significantl
ignifi b 33-fold,
((b ld P<0.05,
ffold
old P <00 0055 n=6).
P<0 n 66)) However,
Ho
H e er ththe spontaneous
h spontaneo
ta s ttwitch
frequencies did not reach statistical differences after T3 treatment (P>0.05) (Figure 7B). High-
resolution optical mapping of T3-treated single hESC-VCMs showed that their action potential
parameters were not different from those of time-matched untreated controls (Figure 7C and D).
Post-transcriptional regulation is crucial for CM maturation
We find that normalized transcript abundance relative to protein abundance during the
developmental continuum of CM differentiation (hESCs to hESC-VCMs) and maturation
(hESC-VCMs and hF-VCMs, hF-VCMs and hA-VCMs) is not always linear. Of the 121 proteins
detected here, transcripts for 116 could be detected by microarray26. Among proteins that showed
differential abundance between hESC-VCMs and hF-VCMs, 37% had significant transcriptomic
12
changes in the same direction. Only 19% of proteins differentially expressed between hF-VCMs
and hA-VCMs showed the same trend. Pearson correlation analysis confirmed that the
correlation between mRNA and protein expression changes during maturation was lower than
that during differentiation (Table S4). Thus, a majority of the protein changes identified here
during stages of development/differentiation could not be accounted for by transcriptional
regulation.
Discussion
The present study was the first to investigate cardiac maturation by profiling
ng the
ofiling the proteome
pro
rote
t om
teomee of
o
CMs isolated
ed from
m fetal and
nd adult ventricles and atria
fetaal an atria,
a, and comparingg to hESC-VCMs.
t that of hESC-VCM
hESC VCM
Our global pprofiling analysis

roofiling ana
aly indicates
lyssiss in
ndi
d ca
cate tthat
tes th
te he pproteome
att the me of hhESC-VCMs
roteeom ESC-
ES C-V
C- CMs are
VCM ar grossly
gros
gro sl
os similar
s y si
simmilaa to
those of hF-VCMs
-VCMss and
VCM an hF-ACMs.
nd hF
hF- Ms. Previous
-ACM P ev
Pr i uss global
evio
io globall profiling
loba
ba p of
profil
i in
il ng attempts
attte
t mpt primarily
pts pri
rima
mari
marily
ri ly involved
y in
nvo
volv
lveed
lv
microarray-based
-
-based transcriptomic
transcrip
pto
t mi
m c data13
data13-17
13--17
. A st
study
tud
u y by Van
Van Hoof
Hoo
of et al
al focused
focu
fo cuse
cusedd exclusively oon
se
mbrane
m bran
branee pr
plasma membrane prot
ottei
eins
proteinsns iinn CMs
CMs derived
deri
derive
ivedd from
from hESCs
hESCs
ESCs
C and
andd human
hum
uman
an fetal
fettall hhearts,
eart
earts,
s, aand
nd iidentified
identif
dent
dentiif
elastin microfibril interfacer 2 as a surface marker of CMs18. By examining the global proteomes,
here we report novel differences and similarities in contractile and metabolic enzymes among
our CM populations. In combination with bioinformatics and functional analyses, our
experiments further enable us to reveal the role of PPARA/PPARGC1A and related signaling
pathways in hESC-VCM maturation.
Metabolism is crucial for supplying energy for contraction and mutations in key
metabolic genes such as ACADVL can result in cardiomyopathy40. Chung et al has shown that
mitochondrial oxidative metabolism is crucial for the cardiac differentiation of mouse embryonic
stem cells41. It is therefore possible that potential differences in the mitochondrial capacities of
hESC-VCMs and hA-VCMs underlie the process of maturation. Specifically, we show that the
13
protein levels of important FA metabolic enzymes are low in hESC-VCMs relative to hA-VCMs
and that an established PPARA/PPARGC1A agonist WY-14643 significantly increases the
mRNA levels of FA metabolic genes, elevates P and improves the organization of the
mitochondria and the phosphocreatine system of hESC-VCMs. While a hyperpolarized P
drives ATP production and is an indication of enhanced mitochondrial/metabolic output, a
promoted alignment of mitochondria to the contractile apparatus and improved organization of
components of the phosphocreatine system further enables a more efficient transfer of energy
between the two units. These results in human cells are reminiscent off those that take place
plaace
c
during mouse embryonic development30, and are also consistent with those off Birket
Birkkett ett al w
who
a re
at
showed that epr
pres
essi
es s on off PP
si
repression P ARGC1A decreased m
PPARGC1A itochondrial conte
it
mitochondrial tennt and thereby
content
compromised
edd the
the capacity
y forr ccoping
opin
in
ng wi
wit
with
th en
energetic
nerg tresss42. IIndeed,
getic sstress nde
deed, ou
our
ur re
recent
ecentt tr
ttranscriptomic
ranscriiptom
m
x ime
xperi ment
n also
nt
profiling experiment alsso identifies
iden
id
denti
tifi
ti
ifi
fies PPARGC1A
PPA
PARG
GC1A ass a member
memb
member
mber off a cardiac
card
rdia
dia
i c TF network
netwo
work
rkk which
whhi
regulates genes
e
enes impo
important
p rtant fo
fforr he
hheart
art de
ddevelopment
velo
l pm unction26; similarly,
p ent andd ffunction siimila
il rlly, Xu
Xu et al reports
repo
p rts tthat
PPARGC1A/PPARA
A/PPARA
A/PPA
/PPARA
RA signaling
ig li regulate
reg llate
at genes enriched
riich
hed
d iin CM
CMs relative
relati
latii e to
t undifferentiated
ndifferentiated
diff
di ff tiat
i edd
hESCs14. Most interestingly, the thyroid hormone T3 significantly augments the contractile force
without affecting the electrical properties as gauged by their action potential parameters. In
neonatal rat cardiomyocytes, T3 has a biphasic effect on PPARGC1A level dependent on the
treatment duration32. In other systems such as hepatocarcinoma HepG2 cells, PPARA inhibits the
transcriptional activity of the thyroid hormone receptor by competing for the retinoid X
receptor33. How the different pathways interact and the intricate cause-and-effect relationships
among their components deserve additional investigations.
Here we show that transcriptomic and proteomic changes among CM samples are largely
discordant. This finding suggests that post-transcriptional mechanisms like RNA processing,
14
stability or transport, translation and post-translational pathways or protein stability must account
for at least some of these differences. This also demonstrates the importance of a proteomic
approach. In comparison to the results of Van Hoof et al18, who examined the plasma membrane
proteome of hESC-CM cultures and hF-CMs, 36 proteins were detected in both studies.(Table
S5). The small overlap maybe attributed to the precipitation of plasma membrane proteins during
DIGE experiments and are therefore mostly excluded from our analysis43 Gundry et al examined
published proteomic data on pluripotent stem cells and showed that overlap among different
proteomic studies are generally very small and this is consistent with what we observed44. O
Of the
ones that were found in both studies, only 31% showed concordant changes
anges bbetween
ettween hE
hESC-CMs
ESC
and hF-VCMs,
Ms, which
Ms whhic
i h could
coul
uld result from the unsorted
ul ed aand
nd MLC2v-pur
MLC2v-purified
riffied populations usedd in
Van Hoofss aand

nd our studi
studies.
ies.
Ourr lab
laboratory
bor
orat
ator
at o y ha
or hhass pr
prev
previously
eviiously
io ly rreported
eportted sseveral
ever
erall iion elss8, Ca
onn cchannels
hannnel
ha el Ca2+-h
-handling
han
a dlin
ng pr
prot in 45,
proteins
ottei
ein
microRNAss46 and epi

epigenetic
p ge ponents47 th
g netiic comp
components that
hat are ddeterminants
eterminants off the
th
he maturation of hESC
hESC-
C
CMs. Led by
b these
thhe data
dat andd other
othhe clues,
cll es wee ffo
found
o nd
d that
h induced
iind
ndd ced
ed
d contractions
tr tii via
iia electrical
ell triicall
stimulation increase the expression of contractile genes and improve the organization of
myofilaments8. Induced contractions place an increased demand on energy, which may be met
by driven metabolic maturation. Taken collectively, our data raise the intriguing possibility that
additive synergistic effects of such pro-maturation mechanisms as electrical conditioning8,
epigenetics priming47, T3 treatments, etc, exist for facilitated maturation of hESC/iPSC-VCMs.
Conclusion
Our present proteomic study identified new candidate proteins and pathways that are less
abundant in hESC-VCMs compared with their fetal and adult CMs. Specifically, we have
implicated a role for PPARA/PPARGC1A in hESC-CM metabolic and mitochondrial maturation
15
with T3 treatment significantly enhancing the contractile force. The successful use of hESC-CMs
as human heart disease models and cardiotoxicity screening tools depends on their ability to
recapitulate the properties of their adult counterparts. The results may provide new insight for
devising mechanism-based in vitro maturation strategies.
Funding Sources: This work was supported by the Research Grant Council of Hong Kong (T13-
706/11, HKU772913 and HKU17113514) and the Faculty Core of the University of Hong Kong.
Conflict of Interest Disclosures: None.
References:
1. Thomson
n JJA,
A, Itskovi
Itskovitz-Eldor
v tzz-E
-Eld
dor JJ,, Sh
Shapiro
hap
apir
i o SS
ir S
SS,, Wa
W
Waknitz
knnitz MA,
MA, Swiergiel
Swwie
ierg
rgie
rg iell JJ, Marshall
ie Mars
Ma rshaall VS,
rs VS, et al.
Embryonic stem
em cell lines derived
stem ffrom
derrivvedd fr m hhuman
om um blastocysts.
man bl SScience.
lasttocyssts.. Sc ieenc 1998;282:1145-1147.
nce. 199988;2 282::114
145--11447.
14
2. He JQ, Ma YY,, Le
Leee Y, T
Thomson
homson
homs o JA, A, Kamp
Kampp TJ.
TJ Human
Huma
Hu man em
embr
embryonic
mbr
bryoni niic st
stem
tem cel
cells
ells
el ls ddevelop
ev
vel
elop
lop iinto
nto
nt
t
multiple types
p of cardiac my
pes m
myocytes:
ocyt
oc y es
yt es: Ac
Acti
Action
t onn potential
ti pot
oten
e tiial
en al characterization.
cha
hara
r ct
cter
erriz
eriz
izat
a io
at ion.
n Ci
n. Circ
C rcc Res.
Res. 2003;93:32
2003;93:32-39.
2
3. Xu C, Police
olice S, Rao
o N,
Rao N, Carpenter
Carp
Carpen
rp terr MK.
ente
ente MK. Characterization
Char
Ch arac
aracteri
acte zaat onn and
teeriza
rizati andd enrichment
enric
nrric
ichm
hm entt of
hmen
men of cardiomyocytes
card
cardiomyocyy
rd
derived from
m hhuman embryonic
man embr
mb
b onicic stem
te cells.
cells Circ Res.
ll Ci R 20 2002;91:501-508.
2002;91:501
200202;91
91:50
5011 508
508
4. Kehat I, Kenyagin-Karsenti D, Snir M, Segev H, Amit M, Gepstein A, et al. Human

embryonic stem cells can differentiate into myocytes with structural and functional properties of
cardiomyocytes. J Clin Invest. 2001;108:407-414.
5. Xue T, Cho HC, Akar FG, Tsang SY, Jones SP, Marban E, et al. Functional integration of
electrically active cardiac derivatives from genetically engineered human embryonic stem cells
with quiescent recipient ventricular cardiomyocytes: Insights into the development of cell-based
pacemakers. Circulation. 2005;111:11-20.
6. Mummery C, Ward D, van den Brink CE, Bird SD, Doevendans PA, Opthof T, et al.
Cardiomyocyte differentiation of mouse and human embryonic stem cells. J Anat. 2002;200:233-
242.
7. Keung W, Boheler KR, Li RA. Developmental cues for the maturation of metabolic,
electrophysiological and calcium handling properties of human pluripotent stem cell-derived
cardiomyocytes. Stem Cell Res Ther. 2014;5:17.
8. Lieu DK, Fu JD, Chiamvimonvat N, Tung KC, McNerney GP, Huser T, et al. Mechanism-
16
based facilitated maturation of human pluripotent stem cell-derived cardiomyocytes. Circ

Arrhythm Electrophysiol. 2013;6:191-201.
9. Liu J, Fu JD, Siu CW, Li RA. Functional sarcoplasmic reticulum for calcium handling of
human embryonic stem cell-derived cardiomyocytes: Insights for driven maturation. Stem Cells.
2007;25:3038-3044.
10. Lieu DK, Liu J, Siu CW, McNerney GP, Tse HF, Abu-Khalil A, et al. Absence of transverse
tubules contributes to non-uniform ca(2+) wavefronts in mouse and human embryonic stem cell-
derived cardiomyocytes. Stem Cells Dev. 2009;18:1493-1500.
11. Poon E, Kong CW, Li RA. Human pluripotent stem cell-based approaches for myocardial
repair: From the electrophysiological perspective. Mol Pharm. 2011;8:1495-1504.
12. Snir M, Kehat I, Gepstein A, Coleman R, Itskovitz-Eldor J, Livne E, et al. Assessmen

Assessment
entt of the
en
ultrastructural and proliferative properties of human embryonic stem cell-derived
cel
elll-de
deri
rive
ri vedd
ve
cardiomyocytes. Am J Physiol Heart Circ Physiol. 2003;285:H2355-2363. 363.
13. Cao F, Wagner

Wag
Wag
agne nerr RA,
ne RA Wilson
A Wi
W lson KD, Xie X, Fu JD, Drukker
Drukker M, et al. T Transcriptional
ranscriptional and
functional pro
pprofiling
rofi
ro filing off hu
fi human
uma eembryonic
m n emmbr
b yo
yoni
nicc st
ni em ccell-derived
stem e l-de
el deriveed ca
de ccardiomyocytes.
rddio
i my
myoc
occ s. PL
ocytes PLoS
LoSS One.
Onee.
2008;3:e3474.
74.
14. Xu XQ,, Sooo S

SY,
Y Sun
Y, un W,
W, Zweigerdt
Zw ge
Zweige
gerddt R. Global
rdt
rd Glo b l expression
loba
ba expr
ex ion profile
presssi
sion proofi
fille of
ile hhighly
of hi ghly
gh ly enriched
enrric
i hed
hed
cardiomyocytes
c
cytes derived ffrom
rom
ro hhuman
m hu embryonic
m n em
ma embr
brryo
y ni stem
tem ccells.
nicc st Stem
e ls.. St
el tem C Cells.
e ls
el 2009;27:2163-2174.
ls. 20
2009
09;27:2163-21744
09
15. Asp J, Steel

Steel D, JoJonsson
Jons sonn M,
nsso
ns so M, Ameen
Ame
meen
en
nC C,, Dahlenborg
Dahl
Da
Dahl
hlen
enbo
enbo
borg
r K
rg K,, Je
Jeppsson
Jepp
ppss
ppss onn A,
sson A, et al.
l Cardiomyocyte
al. Carrdi
diom
o yocytee
clusters derivedd ffrom hhuman
rii ed embryonic
man embrbr onic
nii st lls share
stem cells
ell ha similarities
il itiie with
imil h human
iith
th h man hheart tissue.
rt ti
tiss
is J
Mol Cell Biol. 2010;2:276-283.
16. Synnergren J, Akesson K, Dahlenborg K, Vidarsson H, Ameen C, Steel D, et al. Molecular

signature of cardiomyocyte clusters derived from human embryonic stem cells. Stem Cells.
2008;26:1831-1840.
17. Synnergren J, Ameen C, Jansson A, Sartipy P. Global transcriptional profiling reveals

similarities and differences between human stem cell-derived cardiomyocyte clusters and heart
tissue. Physiol Genomics. 2012;44:245-258.
18. Van Hoof D, Dormeyer W, Braam SR, Passier R, Monshouwer-Kloots J, Ward-van

Oostwaard D, et al. Identification of cell surface proteins for antibody-based selection of human
embryonic stem cell-derived cardiomyocytes. J Proteome Res. 2010;9:1610-1618.
19. Reubinoff BE, Pera MF, Fong CY, Trounson A, Bongso A. Embryonic stem cell lines from
human blastocysts: Somatic differentiation in vitro. Nat Biotechnol. 2000;18:399-404.
20. Yang L, Soonpaa MH, Adler ED, Roepke TK, Kattman SJ, Kennedy M, et al. Human
cardiovascular progenitor cells develop from a kdr+ embryonic-stem-cell-derived population.
17
Nature. 2008;453:524-528.
21. Boudou T, Legant WR, Mu A, Borochin MA, Thavandiran N, Radisic M, et al. A

microfabricated platform to measure and manipulate the mechanics of engineered cardiac
microtissues. Tissue Eng Part A. 2012;18:910-919.
22. O'Brien TX, Lee KJ, Chien KR. Positional specification of ventricular myosin light chain 2
expression in the primitive murine heart tube. Proc Natl Acad Sci USA. 1993;90:5157-5161.
23. Franco D, Markman MM, Wagenaar GT, Ya J, Lamers WH, Moorman AF. Myosin light
chain 2a and 2v identifies the embryonic outflow tract myocardium in the developing rodent
heart. Anat Rec. 1999;254:135-146.
24. Chuva de Sousa Lopes SM, Hassink RJ, Feijen A, van Rooijen MA, Doevendans PA,
Tertoolen L, et al. Patterning the heart, a template for human cardiomyocyte development.
yocyte development. Dev
ent
Dyn. 2006;235:1994-2002.
25. Kaab S, Barth AS, Margerie D, Dugas M, Gebauer M, Zwermann L L, et al

al. Gl
Global
Glob al ggene
obal
ob ene
en
expression in human
huma
hu mann myocardium-oligonucleotide
ma m oc
my ocardium-oligonucleotide m microarray
i roarray analysis
ic iss of regional diversity and
transcriptional
ona regulation
tionn iin
nal regulati
ti heart
n hear
he
ear ffailure.
artt fa ilur
il ure.
ur Mol
e J Mo Med
ol Me
M d ((Berl).
Ber
erl)
rl) 22004;82:308-316.
l).. 20 0 ;8
04 ;82:
2:30
2: 308-31
30 316.
31 6.
26. Poon E,, Yan

Yan B, Zhanghanng S, Rushing
Zha
ha Rushhiningg S, Keung
Keuungg W,
W, Ren
Renn L,
L, et
et al.. Transcriptome-guided
Tran
nsccriipt
ptome- e-gu
e-guuid
ded
d
functional analy
aanalyses
ysees rreveal
eveeall nnovel biological
ovell biolo
logi
lo properties
gical pr
gi prop tiies aand
opeerti
op nd rregulatory
eg
gul
ulat
l tor
ory hierarchy
r hiierarch chyy off hhuman
ch uman
um an
embryonic stem cell-derived
cell-deriv e vventricular
ved e trric
en ullar ccardiomyocytes
icul ardi
ard om
di omyo
yocy
yo cyte
cytess cr
te crucial
ruc for
u iaal fo maturation.
or ma
matu
tura
tu tioon. PLoS One..
raati
2013;8:e77784.
784.
27. Turnbull
ll IC,
ll C Karakikes
IIC kik I,
K akik I Serrao
Se GW,
GW Backeris
GW B kerii PP, Lee JJ,
L JJJJ XXie C, et al
ie C al.
l AAdvancing
Ad cii ffunctional
d ancing nct
ct
engineered cardiac tissues toward a preclinical model of human myocardium. FASEB J.
2014;28:644-654.
28. Gilde AJ, van der Lee KA, Willemsen PH, Chinetti G, van der Leij FR, van der Vusse GJ, et
al. Peroxisome proliferator-activated receptor (ppar) alpha and pparbeta/delta, but not
ppargamma, modulate the expression of genes involved in cardiac lipid metabolism. Circ Res.
2003;92:518-524.
29. Venkataraman S, Jiang X, Weydert C, Zhang Y, Zhang HJ, Goswami PC, et al. Manganese
superoxide dismutase overexpression inhibits the growth of androgen-independent prostate
cancer cells. Oncogene. 2005;24:77-89.
30. Hom JR, Quintanilla RA, Hoffman DL, de Mesy Bentley KL, Molkentin JD, Sheu SS, et al.
The permeability transition pore controls cardiac mitochondrial maturation and myocyte
differentiation. Dev Cell. 2011;21:469-478.
31. Chung S, Dzeja PP, Faustino RS, Terzic A. Developmental restructuring of the creatine
kinase system integrates mitochondrial energetics with stem cell cardiogenesis. Ann N Y Acad
Sci. 2008;1147:254-263.
18
32. Xu W, Hou D, Jiang X, Lu Z, Guo T, Liu Y, et al. The protective role of peroxisome
proliferator-activated receptor gamma coactivator-1alpha in hyperthyroid cardiac hypertrophy. J
Cell Physiol. 2012;227:3243-3253.
33. Juge-Aubry CE, Gorla-Bajszczak A, Pernin A, Lemberger T, Wahli W, Burger AG, et al.
Peroxisome proliferator-activated receptor mediates cross-talk with thyroid hormone receptor by
competition for retinoid x receptor. Possible role of a leucine zipper-like heptad repeat. J Biol
Chem. 1995;270:18117-18122.
34. McClure TD, Young ME, Taegtmeyer H, Ning XH, Buroker NE, Lopez-Guisa J, et al.
Thyroid hormone interacts with pparalpha and pgc-1 during mitochondrial maturation in sheep
heart. Am J Physiol Heart Circ Physiol. 2005;289:H2258-2264.
35. Goldenthal MJ, Ananthakrishnan R, Marin-Garcia J. Nuclear-mitochondrial cross-talk in

cardiomyocyte t3 signaling: A time-course analysis. J Mol Cell Cardiol. 2005;39:319-326.
ol. 2005;39:319-32
26.
6
36. Chattergoon NN, Giraud GD, Louey S, Stork P, Fowden AL, Thornburg
rnburgg KL.
KL Thyroid
Thyr
Th y oi
oidd
hormone drives fetal cardiomyocyte maturation. FASEB J. 2012;26:397-408.
97 408
088
37. Lee YK,

K, Ng KM, C Chan
haan YC,
han YC C, Lai
L i WH,
La WH Au A KW,
KW,W HoHo CY,
CYY, et
et al.
l. TTriiodothyronine
r iodo
ri io
od thyr yron
yr onnin ppromotes
i e pr
promote
rom
om
mot
oe
cardiac differentiation
ferentiation
fe
erentiation and
annd maturation
maatu tiion of
urati of embryonic
em m cells
mbryyonnic stem cells vvia
iaa thee cclassical
laasssicall ggenomic
en pathway.
nomiic pa
J Clin Endocrinol
ocri
oc inol Metab.
rin tab.. 22010;95:4479.
Meta
ta 010
100;95::44
4479..
38. Sedan O, Dolnikov K,, Z Zeevi-Levin

eeevi
v -LLev
evin
in
nN N,, Le
L
Leibovich
ibov
ib ovic
ichh N, AAmit
mtM
mi M,, IItskovitz-Eldor
tsko
ts kovi
ko vitz
vi tzz-E 1,4,5-
-Eldor J, et al. 1,
inositol trisphosphate-operated
phospphate-opperated d iintracellular
ntracelllular ca(2+)
a((2+) stores and
ca( angiotensin-ii/endothelin-1
d ang gio
i tensiin-ii ii//endothelin-1
ii
signaling pathway
athway are functional
arre fu
func
ncction
ncti onna in
tional in human
hu mann embryonic
huma
uma embr
em nic stem
bryoni
bryo
yo
onic m cell-derived
s em
st cell
elll-d
-der
der ived cardiomyocytes.
eriv
ived carardi
diom
di omyo
om y cytes. SStem
Cells. 2008;26:3130-3138.
;26:3130
;26
26:313
31300 3138
3138
39. Dolnikov K, Shilkrut M, Zeevi-Levin N, Gerecht-Nir S, Amit M, Danon A, et al. Functional

properties of human embryonic stem cell-derived cardiomyocytes: Intracellular ca2+ handling
and the role of sarcoplasmic reticulum in the contraction. Stem Cells. 2006;24:236-245.
40. Mathur A, Sims HF, Gopalakrishnan D, Gibson B, Rinaldo P, Vockley J, et al. Molecular
heterogeneity in very-long-chain acyl-coa dehydrogenase deficiency causing pediatric
cardiomyopathy and sudden death. Circulation. 1999;99:1337-1343.
41. Chung S, Dzeja PP, Faustino RS, Perez-Terzic C, Behfar A, Terzic A. Mitochondrial
oxidative metabolism is required for the cardiac differentiation of stem cells. Nat Clin Pract
Cardiovasc Med. 2007;4 Suppl 1:S60-67.
42. Birket MJ, Casini S, Kosmidis G, Elliott DA, Gerencser AA, Baartscheer A, et al. Pgc-
1alpha and reactive oxygen species regulate human embryonic stem cell-derived cardiomyocyte
function. Stem Cell Reports. 2013;1:560-574.
43. Lilley KS, Friedman DB. All about dige: Quantification technology for differential-display
2d-gel proteomics. Expert Rev Proteomics. 2004;1:401-409.
19
44. Gundry RL, Riordon DR, Tarasova Y, Chuppa S, Bhattacharya S, Juhasz O, et al. A cell
surfaceome map for immunophenotyping and sorting pluripotent stem cells. Mol Cell Proteomics.
2012;11:303-316.
45. Liu J, Lieu DK, Siu CW, Fu JD, Tse HF, Li RA. Facilitated maturation of ca2+ handling
properties of human embryonic stem cell-derived cardiomyocytes by calsequestrin expression.
Am J Physiol Cell Physiol. 2009;297:C152-159.
46. Fu JD, Rushing SN, Lieu DK, Chan CW, Kong CW, Geng L, et al. Distinct roles of
microrna-1 and -499 in ventricular specification and functional maturation of human embryonic
stem cell-derived cardiomyocytes. PLoS One. 2011;6:e27417.
47. Chow MZ, Geng L, Kong CW, Keung W, Fung JC, Boheler KR, et al. Epigenetic regulation
of the electrophysiological phenotype of human embryonic stem cell-derived ventricular
cardiomyocytes: Insights for driven maturation and hypertrophic growth.
wth. Stem Cells Dev.
Dev
ev..
ev
2013;22:2678-2690.
20
Table 1: Summary of differential protein expression changes in different functional groups
Functional gropus General trends
Contraction Increase from hESC-VCM to hF-VCM to hA-VCM
Metabolism FAO: Modest increasee ffrom rom

om hhESC-VCM
ESC
ES C-VVC to hF-VCM,
large increase from om hF
hF-
hF-VCM
-VCM to to hA-VCM
Glycolysis:: MModest
odest
d increase
i ea
increa
ease from
f hhESC-VCM
hESC-
SC- to hF-VCM,
decr
de crrea
ease
decrease se ffrom
se r m hF
ro F-VCM to
hF-VCM to hA-VCM
hA
hA
Glycogen
Glyc
ycog
oggen
n metabolism:
metababol
ab oliism:
ol m PYGB
PYGB highly
hig
ghlly enriched
enrr
en in hF-VCMs,
low
loow in hESC-VCMs
hES
ESC-C VCVCMs
Ms and
and hA-VCMs
hA-
Energy
Ener
En ergy transfer:
ergy tra
rans
nsfe
ns fer:
fer: Increase
Incrrea
In e se from
fro
romm hESC-VCM
hESC
hE S -VCM to t hF-VCM to hA-VCM
Cell survival Similar between hESC-VCM to hF-VCM,

reduced in hA-VCMs.
Ion transport No general trend.

Differential expression of specific proteins.
Regulation of mRNA and Differential expression of specific proteins

protein expression eg MLL2 and GFM1
21
Figure Legends:
Figure 1: 2-D gels scanned for (A) hES2 (i.e. Cy3) and hESC-VCM (i.e. Cy5), (B) hF-ACM and
hF-VCM and (C) hA-ACM and hA-VCM. Overlay of the gels where Cy5 and Cy3 are red and
green, respectively. Proteins that were equally expressed in both samples are yellow.
Figure 2: (A) Image of preparatory gel. Differentially expressed spots identified are marked. (B)
Topographical and spot images of MLC2V (MYL2) in hESC, hESC-VCM,

VCM, hF-ACM, hF-VCM,
hF
F-V
hA-ACM and hA-VCM. (C) Normalised spot intensities of MLC2V (MYL2)

L2)) in
MYL2
L2 in the
the six CM
CM
populations.
s
s.
Figure 3: A)
A HHierarchical
iera
ierarc
ra r hica
rc clustering
call cl
clus ter ngg sshowing
lustteri howing
ho ng tthat
h t bi
ha bbiological
iollogiccall rreplicates
ogic epli
lica
li cate
cate ccluster
t s cl terr ttogether.
uste
te oge
geth
ther
th Relative
her. R e
gene expression
s
ssion is color-coded: d iindicates
color-codded: red upregulation
nddicates upr
p eg i n andd green
gullatio indicates
g een indi
gr downregulation.
d cates downregu
g latt B)
Principal component analysis

omponentt anal
all sis showing
is sho
ho iing th
that hESC-VCMs
hat hE
hESC
SC VCM are grouped
VCMs gro ped closely
lo ell with
d more closel iith
th fetal
h ffet
et
CMs.
Figure 4: Differential expression of selected proteins in hESC, hESC-VCM, hF-VCM, and hA-
VCM. Topographical images are shown. The graphs show normalised spot intensities.
Figure 5: Bioinformatic analysis of proteins upregulated during cardiac maturation. (A) Cluster
analysis of differentially expressed proteins. Maturation proteins (as indicated) was subject to
Ingenuity pathway analysis. (B) Summary of pathway analysis results. The p-val indicates the
likelihood that the association between the maturation proteins and a given process or pathway is
22
due to random chance. The smaller the p-val, the more significant the association. (C) Summary
of upstream regulator analysis. (D) Schematic of upstream regulation. Solid arrows indicate
regulatory relationships between upstream regulator and targets identified by Ingenuity pathway
analysis. Dotted line indicates interaction between upstream regulators reported in literature.
Dotted arrow indicates directional regulation between upstream regulators, as reported in
literature. (E) Predicted targets of PPARGC1A. Black/Red indicates proteins which were/were
not significantly enriched on the mRNA level in hA-VCMs and/or hF-VCMs relative to hESC-
VCMs. For full details, please see Table S3A-C.
Figure 6: Effec
Effect
E PPARA/PPARGC1A
ectt of P
ec P RA
PA RA/PPARGC1A activation
n oon
n hESC-VCMs.. A)
A Effect of WY-146
WY-14643
6
(50M) on m
mitochondrial
ittochondriall membrane
memb
mbrane
mb n pot
ne potential
oteenttiaal with
with and
an wi
w
without
thoout oleic acid
th acid supplement
d supp
supplem
eme
ement (0.2mM).
ent (0
(0.2m
.2m
* p<0.05. B) Effect
Eff
ffec
ff WY-14643
ectt of W
ec Y-14 643 onn the
1464
14 64 the expression
exppre
ress
s ion
ss ion off FA
FA metabolic
mettabo
me lic genes.
boolic gennes.
ge s WY-14643
WY-14
1464
14 6433 (100M
64 (10
(1
and 50M) was ap

applied
p ied to hhESC-VCMs
ppl ES
SC- CMs for
C VCM for four
f ur days.
fo ys. DMSO-treated
day DMS
SO-treatedd andd untreated hESC--
VCMs acted control.

d as control
troll Data
Data is
i presented
tedd as fold
fold
ld changes
ch
h relative
relati DMSO-treated
latii e tto DM
DMSO
SO tr
treated nt l cells
tedd control
(MeanSEM). Untreated cells were also assessed and gene expression was not significantly
different from DMSO treated cells. N=3. * p<0.05, ** p<0.01. (C) Effect of WY-14643 on
proteins important for mitochondrial morphology and energy transfer. COXIV, FHL2 and GFM1
staining are shown in red. Samples were also co-VWDLQHGZLWK-actinin antibody, in green.
Figure 7: Effect of T3 treatment on hESC-VCMs. Effect of T3 treatment on (A) dynamic
tension and (B) spontaneous contraction frequencies in hvCMTs. hvCMTs were treated with T3
(100nM) for 6 days after which spontaneous dynamic tensions were measured. Data represent
meanSEM. N=6 **p<0.05. (C) Representative tracing of action potential in control and T3
23
treated (100nM, 6 days) hESC-CMs at 1Hz and measured with Di-8-ANEPPS and a high
resolution optical mapping system. (D) Effect of T3 treatment (100nM, 6 days) on action
potential upstroke time and 50% decay time (APD50). Data represents meanSEM. N=6-8.
24
SUPPLEMENTAL MATERIALS
Isolation of fetal and adult CMs
Human fetal and adult CMs were isolated and experimented according to
protocols approved by the UC Davis IUPAC and IRB (Protocol #200614787-1
and # 200614594-1). All fetal hearts (18 weeks) were digested using the
Lagendorff system and adult (32-59 years) with a recirculatory system that
circulated the 37C collagenase solution until cells started to dissociate into
the enzyme solution. The fetal hearts typically took about 30 min and the adult
hearts took over 3 hours to digest. The collagenase solution had 200U/ml
collagenase II (Worthington Biochemical Corp), 4mg protease (Sigma) with 1%
BSA. After enzyme treatment, the hearts were chopped manually to release
the cells into high K+ solution. The fetal cells were plated for 1 hour in M199
with 5mM carnitine, 5mM creatine, 5mM taurine, 10% FBS and 1% pen/strep
to remove fibroblasts, then the medium was collected to retrieve the CMs still
in suspension. The adult cells were not plated but allowed to settle by gravity
for 15 min. The denser CMs at the bottom of the conical tubes were collected.
Detailed information is presented in Table S1.

2-D DIGE
The cells were lysed in 2-D lysis buffer containing 30 mM Tris-HCl, pH8.8, 7 M
Urea, 2 M thio-urea, and 4% CHAPS, and vigorously sonicated for 5 seconds
using VirSonic 100 (VirTis). After vigorous shaking at room temperature for 30
min, the protein lysates were cleared by high speed centrifugation for 30 min.
Supernatants were transferred to fresh eppendorf tubes, and protein
concentrations were adjusted to 5 mg/ml. In addition to the individual samples,
all protein samples were combined to form an internal standard that was also
resolved on each gel to minimize gel-to-gel differences.
CyDye labeling: Each sample was labeled with either CyDye-3 or CyDye-5.
The internal standard was labeled with Cydye-2. Briefly, CyDyes were diluted
1:25 with dimethyl formamide immediately before reaction and 25 g of protein
lysate was mixed with 0.7 l of diluted CyDye. The samples were then
incubated on ice for 30 min, followed by the addition of 0.7 l of 10 mM lysine
to each of the samples to stop the reaction.
First dimension IEF: CyDye-labeled protein samples were mixed in a fresh
vial and 100 l of rehydration buffer (7 M urea, 2 M thiourea, 4% CHAPS, 2

mg/ml DTT, 1% bromophenol blue, 1% Pharmalyte for IEF, pH 3-10) was
added, followed by 140 l destreak solution (7 M urea, 2 M thiourea, 4%
CHAPS, 1% bromophenol blue, 100 mM destreak, 2% Pharmalyte for IEF, pH
3-10) to a final volume of approximately 260 l. The samples were mixed well
and 250 l was loaded per IEF strip (GE Healthcare/Amersham). IEF was then
done for a total of 25000 volt-hours with standard conditions using Ettan
IPGPhore II (as recommended by GE Healthcare/Amersham).
Second dimension SDS-PAGE: After the IEF, each IEF strip was incubated
with 10 ml of 10 mg/ml DTT in equilibration solution (6 M urea, 30% Glycerol,
50 mM Tris-HCl, pH 8.8, 2% SDS) for 15 min with gentle shaking. Each IEF
strip was then incubated with 10 ml of 45 mg/ml Iodoacetamide in equilibration
solution (6 M urea, 30% Glycerol, 50 mM Tris-HCl, pH 8.8, 2% SDS) for 10 min
with gentle shaking to block SH groups. The strips were then rinsed in SDS gel
running buffer once, inserted into a 9-12% gradient SDS gel and covered with
0.5% agarose sealing solution. Electrophoresis was performed at 16C.
Gel scanning and data analysis: The resulting 2-D gel was scanned using a
Typhoon Trio scanner (GE Healthcare/Amersham). A total of 3 images were
generated for each gel at a resolution of 100m using the spectrally resolvable
dyes. These images were imported into Progenesis Samespots v2.3 software.
Using the DIGE mode of this software, the normalized volume ratio of each
CyDye-3 or CyDye5 labeled, individually identified protein spot on a given gel
was compared to the Cy2 labeled internal standard (corresponding to the
same spot) from the same gel to quantitatively determine the fold change in
the compared samples. The relative ratios of a given spot were determined
from each of the replicate gels.
Mass spectrometry:
Selected protein spots were excised from the gel using Ettan spot picker (GE
Healthcare/Amersham). Following in-gel digestion and sample cleaning, tryptic
digests were analyzed using matrix-assisted laser desorption/ionization
(MALDI) mass spectrometry. The matrix used was a-cyano, 4-hydroxy
cinnamic acid. A positive ion mass spectrum was obtained for the tryptic digest
using MALDI-TOF/TOF technology (4800 plus Proteomics Analyzer AB
SCIEX)). Mass accuracy was better than 50 ppm (usually 10 ppm). Peaks
were acquired in a reflector positive mode with mass range 900 to 3500. The
top 10 monoisotopic peaks were further subjected to MS/MS analysis using
TOF/TOF. Both MS and MS/MS data were submitted to the Protein pilot
software (AB SCIEX) for database search using the MASCOT search engine
v2.2 (MATRIX science). The NCBInr human database (2010-7, 227085
sequences) was used for searching and the mass accuracy was calculated
using trypsin autolytic peptide as a reference. Search parameters allowed for a
mass accuracy of 0.2 Da for MS and 0.5 Da foor Ms/MS, two missed cleavage
of trypsin, oxidation of methionine, and carbamido-methylation of cysteine.
Western Blotting
30g of protein lysate was separated on 12% SDS-polyacrylamide gels and
electrophoretically transferred onto PVDF membranes (Schleicher and
Schuell). After blocking the membranes for 30 min at room temperature in 5%
milk in PBS+1% Tween (PBST), primary antibodies were applied overnight at
4C. The membranes were washed three times with PBST, and horseradish
peroxidase-conjugated donkey anti-mouse (Abcam) or horseradish
peroxidase-conjugated donkey anti-rabbit (Abcam) antibodies were applied for
1 h. After washing with PBST, the membranes were developed using
SuperSignal West Pico Chemiluminescent Substrate (Thermo Scientific)
following the manufacturers instructions.

Antibody against FSCN1 (Abcam) was used to control for equal loading.
FSCN1 showed the least deviation among all our samples in the 2D-DIGE
experiment and was therefore considered to be a suitable loading control.
Immunofluorescence analysis
20-30 days after the start of differentiation, cardiospheres were dissociated
and plated onto matrigel-coated glass coverslips in DMEM (Invitrogen, Grand
Island, NY) supplemented with 5% fetal bovine serum (Invitrogen, Grand
Island, NY), 100 M nonessential amino acids, 2 mM glutamine, 50 U/ml
penicillin, 50 g/ml streptomycin (Invitrogen, Grand Island, NY). Cells were
fixed for 15 min at room temperature with 4% paraformaldehyde in
phosphate-buffered saline (PBS). After washing with PBS, cells were
permeabilized in PBS containing 1% Triton X-100 and subsequently blocked in
10% normal goat serum. Primary antibodies were diluted with 1% BSA at
1:100 and applied at 4C overnight. Alexa Fluor 488 goat anti-mouse/rabbit
IgG or alexa Fluor 555 anti-mouse/rabbit IgG (Invitrogen) was applied for 1 hr
at room temperature in dark. Coverslips were mounted onto glass slides in
Prolong Gold mounting medium (Invitrogen). Samples were imaged on a
Nikon Eclipse TiS microscope (Nikon) or Carl Zeiss LSM 700 confocal
microscope (Carl Zeiss). Antibodies against cardiac troponin-T, -actinin,
COXIV, FHL2, CACYBP were purchased from Abcam. Antibody against GFM1
was from Sigma.
Figure legends
Figure S1. Characterisation of hESC-VCMs.
Figure S2. Validation of protein expression. (A) Expression of FHL2, CACYBP
in hESC, hESC-VCM, hF-VCM and hA-VCM was examined by Western
blotting. 1=hESC, 2=hESC-VCMs, 3=hF-VCMs, 4-hA-VCMs. (B)
Immunofluorescent staining of hESC-CMs was performed and confocal
images are shown.
Table S1. Sample information
Table S2. Proteins that were differentially expressed among hESC,
hESC-VCM, hF-ACMs, hF-VCMs, and hA-ACMs and hA-VCMs were grouped
into functional categories based on gene ontology identifications. Pairwise

comparisons were performed as indicated and log transformed fold changes
(>1.5 fold, p<0.05) are shown.
Table S3. (A) Ingenuity pathway analysis, p<0.05. (B) Ingenuity upstream
regulator analysis, p<0.0001. (C) Predicted PPARGC1A targets. Proteomic
and transcriptomic expression changes relative to hESC-VCMs are shown,
fold change>1.5, p<0.05

Phase contrast MLC2V-GFP Merge
Cardiac Troponin T staining Representative action potential

V (mV)
100
500 ms
50
-50
-100
Figure S1. Characterisation of hESC-VCMs.

1 2 3 4
FHL2
CACYBP
GFM1
Figure S2: Validation of protein expression. Expression of FHL2, CACYBP and

GFM1 in hESC, hESC-VCM, hF-VCM and hA-VCM was examined by Western
blotting. 1=hESC, 2=hESC-VCMs, 3=hF-VCMs, 4-hA-VCMs.
Table S1. Sample information
Samples Details
HES2 Undifferentiated HES2 cells, female, passage 60
hESC-VCM D33 hESC-VCMs differentiated from HES2
hF-ACM 18 week gestation from 28 year old mother
hF-VCM 18 week gestation from 28 year old mother
hA-ACM 59 year old Caucasian female
hA-ACM 32 year old Caucasian male
hA-VCM 59 year old Caucasian female
hA-VCM 53 year old Caucasian male

Table S2. Proteins that were differentially expressed among hESC, hESC-VCM, hF-
ACMs, hF-VCMs, and hA-ACMs and hA-VCMs were grouped into functional
categories based on gene ontology identifications. Pairwise comparisons were
performed as indicated and log transformed fold changes (>1.5 fold, p<0.05) are
shown.
hF- hA- hF- hA-

hESC
VCM/ VCM VCM/ VCM
Spot -VCM/ Protein name SYMBOL
hESC / hF- hF- / hA-
hESC
-VCM VCM ACM ACM
Sarcomere
811 - - 1.9 - - Myosin light chain 2a MYL7
816 5.8 - - - - Myosin light chain 2a MYL7
827 - - 2.4 -6.7 -16.4 Myosin light chain 2a MYL7
764 - 2.6 -2.6 - - Myosin light chain 4 MYL4
766 9.3 1.6 -3.1 - -12.8 Myosin light chain 4 MYL4
358 - - 15.0 - - Desmin DES
506 1.9 - 1.8 - - Actin, alpha, cardiac ACTA1
482 3.1 - 2.2 - - Actin, alpha, cardiac ACTA1
535 - - - - - Actin, alpha, cardiac ACTA1
Myosin regulatory light chain 2,

838 7.6 9.7 - 10.3 - MYL2
slow ventricular muscle
Myosin regulatory light chain 2,

839 2.9 2.3 - 5.3 - MYL2
slow ventricular muscle
29 - - 2.9 - - Myosin binding protein C, cardiac MYBPC3
656 3.9 2.0 - 2.3 - Four and a half LIM domains 2 FHL2
432 2.1 2.1 2.3 - - Beta-myosin heavy chain MYH7

73 1.9 1.9 1.5 - - Beta-myosin heavy chain MYH7
176 - - - - - Beta-myosin heavy chain MYH7
770 - 3.2 4.1 3.2 2.8 Beta-myosin heavy chain MYH7
608 2.7 -2.0 -5.9 - - Alpha-myosin heavy chain MYH6
768 2.3 - - - -35.3 Alpha-myosin heavy chain MYH6
95 - - 1.8 - - Alpha-myosin heavy chain MYH6
780 1.6 1.9 - - -2.3 Alpha-myosin heavy chain MYH6
638 - 1.7 2.2 - - Tropomyosin 1 (alpha) TPM1
Other cytoskeletal proteins
260 - - -3.7 - - Myosin-14 MYH14
306 - -1.6 -2.7 -1.5 - Myosin-VIIb MYO7B
531 - - - - - Keratin 9 KRT9
386 - - - - - Spectrin beta chain, brain 4 SPTBN5
377 - - - - - FSCN1 protein FSCN1
268 -2.3 - -2.8 - - WD repeat-containing protein 1 WDR1
171 - - 2.3 - - Ezrin EZR
173 - -2.0 - - - Ezrin EZR
ARP1 actin-related protein 1

448 -2.4 -1.8 -2.3 - - ACTR1A
homolog A, centractin alpha
874 - -4.7 - - - Dihydropyrimidinase-like 3 DPYSL3
Metabolism - Fatty acid metabolism
Enoyl Coenzyme A hydratase 1,

678 2.5 - - - - ECH1
peroxisomal
Acyl-Coenzyme A dehydrogenase,
552 2.6 - - 2.0 - ACADS
C-2 to C-3 short
Short-chain specific acyl-CoA

534 2.2 - - - - ACADS
dehydrogenase, mitochondrial
Medium-chain specific acyl-CoA
499 - 2.5 1.5 - - ACADM
Very long-chain specific acyl-CoA

272 - - 5.2 - - ACADVL
Enoyl-Coenzyme A (Coa)
748 2.6 - - - - ECHS1
Hydratase Short Chain 1
Hydroxyacyl dehydrogenase,
756 - - 9.8 1.7 - HADHB
subunit B
Acetyl-CoA acetyltransferase,
518 - 2.7 - 1.7 - ACAT1
mitochondrial
522 1.6 1.7 - 1.6 - ACAT1
mitochondrial
524 - 2.0 -3.1 - - ACAT1
mitochondrial
351 - 3.6 - - - 3-oxoacid CoA transferase 1 OXCT1
346 1.5 - 1.6 - - 3-oxoacid CoA transferase 1 OXCT1
Metabolism - Electron transport chain and ATP synthesis
Succinate dehydrogenase complex,

252 2.4 - 2.0 - - SDHA
subunit A,
NADH dehydrogenase (ubiquinone)

181 - - - - - NDUFS1
Fe-S protein 1
NADH dehydrogenase (ubiquinone)

743 3.7 - - - - NDUFS3
Fe-S protein 3
Cytochrome b-c1 complex subunit

761 2.6 - - - - UQCRFS1
Rieske-like protein 1
ATP synthase subunit beta,

404 2.5 - 2.2 - - ATP5B
mitochondrial
ATP synthase, H+ transporting,

479 - - 2.5 - - ATP5A1
mitochondrial F1
Electron transfer flavoprotein
763 - - 1.5 - - ETFB
subunit beta
Manganese superoxide dismutase,

814 3.0 1.7 - - - SOD2
Q143a, Chain B
Metabolism - Tricarboxilic acid cycle
879 - - - - - Aconitase 2, precursor ACO2
897 -2.7 - - -1.7 - Aconitase 2, precursor ACO2
882 - - 2.1 - - Fumarate hydratase FH
Fumarate hydratase, isoform

457 - - - - - FH
CRA_d
Fumarate hydratase, isoform

466 - 2.2 4.0 - - FH
CRA_d
Isocitrate dehydrogenase 2
484 - - - - - IDH2
(NADP+), mitochondrial
Isocitrate dehydrogenase [NAD]

564 - - 1.9 - - IDH3A
subunit alpha, mitochondrial
Isocitrate dehydrogenase 3 (NAD+)

561 - 1.6 - - - IDH3B
beta
Metabolism - Glycolysis and glucose metabolism
430 -1.9 -1.9 -5.7 - - Enolase 1 ENO1
442 3.5 - - - 1.6 Enolase 3 ENO3
621 - - -5.8 2.4 - Cytosolic malate dehydrogenase MDH1
624 2.1 - -11.3 2.6 - Cytosolic malate dehydrogenase MDH1
633 - - -9.5 2.5 - Cytosolic malate dehydrogenase MDH1
Glyceraldehyde-3-phosphate
595 - - -6.7 - - GAPDH
dehydrogenase
596 - - -3.6 - - Glyceraldehyde-3-phosphate GAPDH

dehydrogenase
Glyceraldehyde-3-phosphate
591 - - -17.8 - - GAPDH
dehydrogenase
706 - - -3.1 - - Phosphoglycerate mutase 1 (brain) PGAM1
314 - - -9.0 - - Pyruvate kinase isozymes M1/M2 PKM2
329 - - -19.2 - - Pyruvate kinase isozymes M1/M2 PKM2
320 - - -8.9 - - Pyruvate kinase, muscle PKM2
769 -2.5 - - - - Triosephosphate isomerase 1 TPI1
762 - - -7.8 2.0 - Triosephosphate isomerase 1 TPI1
Chain A, Human Pyruvate

476 - - 1.7 - - PDHA1
Dehydrogenase
576 -6.0 - -2.1 - - Transaldolase 1 TALDO1
588 -2.0 - - - - Transaldolase 1 TALDO1
572 - -1.6 - - - Transaldolase 1 TALDO1
Glycogen phosphorylase, brain

109 - 14.0 -18.2 - - PYGB
form
Phosphoenolpyruvate
270 - - 7.4 - - PCK2
carboxykinase 2 (mitochondrial)
UTP-glucose-1-phosphate
390 - - 2.3 - - UGP2
uridylyltransferase
Metabolism - Energy transfer
884 - 2.2 2.5 2.0 - Creatine kinase, mitochondrial 2 CKMT2
505 4.4 2.5 - - - Creatine kinase, muscle CKM
500 1.7 - - - - Creatine kinase, muscle CKM
719 - - - - - Adenylate kinase 2 AK2
Other metabolic proteins
67 - - - -1.9 - Glucosidase II GANAB

Phosphoribosylaminoimidazole
carboxylase,
471 -7.4 - - - - PAICS
phosphoribosylaminoimidazole
succinocarboxamide synthetase
369 - - - - - Phosphoglycerate dehydrogenase PHGDH
354 -2.5 - 1.7 - - Phosphoglycerate dehydrogenase PHGDH
538 -3.9 -3.6 1.6 - - Phosphoserine aminotransferase 1 PSAT1
322 2.3 -2.9 - -2.0 - Prolyl 4-hydroxylase, beta subunit P4HB
481 -1.9 - - - - S-adenosylhomocysteine hydrolase AHCY
Trifunctional purine biosynthetic

406 - - 2.8 - - GART
protein adenosine-3
Regulation of expression - Gene expression
Histone-lysine N-methyltransferase
508 - 1.9 1.6 - - MLL2
MLL2
575 - -5.5 - - - TRIM28 protein TRIM28
556 - - - - - Zinc finger protein GLI2 GLI2
Regulation of expression - RNA Binding
Heterogeneous nuclear
619 2.5 - -15.8 - - HNRPDL
ribonucleoprotein D-like
298 - - - - - HNRPDL
ribonucleoprotein D-like
498 - - 3.4 - - HNRNPH1
ribonucleoprotein H1
371 - 1.7 - - - HNRNPH1
ribonucleoprotein H1
319 - -4.8 - - - HNRNPL
ribonucleoprotein L
896 - - -2.6 - - HNRNPL
ribonucleoprotein L
878 - 2.0 2.8 - - KH-type splicing regulatory protein KHSRP
212 - - - - - Splicing factor 3A subunit 1 SF3A1
Regulation of expression Protein translation and stability
507 - - - - - Laminin Receptor Precursor RPSA
112 - 3.7 - - - Glycyl-tRNA synthetase GARS
412 - 1.8 1.8 - - Glycyl-tRNA synthetase GARS
92 - - -2.7 - - Elongation factor 2 EEF2
159 - - 2.4 - - Elongation factor G, mitochondrial GFM1
Eukaryotic translation initiation

540 - - - 1.8 -1.8 EIF2S3
factor 2 subunit 3
740 - - 1.5 - - Growth-inhibiting gene 5 protein CACYBP
Ion transport
Transient receptor potential cation

32 - - 3.2 - - TRPM3
channel subfamily M member 3
655 1.7 1.7 - - - Voltage-dependent anion channel 2 VDAC2
Signal transduction
439 -2.4 - - - - GDP dissociation inhibitor 2 GDI2
Synapse defective 1, Rho GTPase,

580 - - - - - SYDE2
homolog 2
Tyrosine 3-
monooxygenase/tryptophan 5-
730 - - -3.8 - - YWHAG
monooxygenase activation protein,
gamma polypeptide
623 - - - -1.6 - G protein beta subunit GNB2
Guanine nucleotide binding protein

676 - - - - - GNB2L1
(G protein), beta
Guanine nucleotide binding protein

684 -3.2 - - - - GNB2L1
(G protein), beta
Phosphatidylethanolamine-binding
821 - - -6.6 - - PEBP1
protein 1
900 - - - - - Cystathionine-beta-synthase CBS
IQ calmodulin-binding motif-
255 - -1.9 5.1 -1.8 - IQCB1
containing protein 1
ROS and aldehyde metabolism
779 2.1 - 1.6 - - Peroxiredoxin 3 PRDX3
757 - - - - - Peroxiredoxin 6 PRDX6
797 -2.1 - -3.1 -1.7 - Peroxiredoxin-1 PRDX1
759 -1.8 - - - - Thioredoxin peroxidase PRDX4
795 - - -5.9 1.5 - Glutathione S-transferase P GSTP1
697 - -1.6 -2.4 - - Glutathione S-transferase omega-1 GSTO1
Glutathione-Dependent
545 -1.9 - -4.8 - - Formaldehyde Dehydrogenase ADH5
With Nad(H)
Aldehyde dehydrogenase 4A1,

321 - 2.8 -1.8 - - ALDH4A1
precursor
381 - - - - - Aldehyde dehydrogenase 7A1 ALDH7A1
Pro- and anti-apoptotic factors
666 - -2.0 - - - Caspase 3 CASP3
876 - -2.6 -2.0 - - Annexin A1 ANXA1
625 2.7 -1.9 -4.3 - - Annexin A2, isoform 2 ANXA2

341 - - - -1.5 - Protein disulfide-isomerase A3 PDIA3
345 - - -2.2 -1.7 - Protein disulfide-isomerase A3 PDIA3
234 - 2.2 -5.8 - - Albumin ALB
236 - - -6.9 - -4.8 Albumin ALB
237 - - -10.9 - - Albumin ALB
263 - - - - - Albumin ALB
Chaperonin containing TCP1,

334 - -1.5 - - - CCT7
subunit 7 (eta)
Heat shock-related 70 kDa protein

526 - - -2.1 - - HSPA2
2
Heat shock protein 90kDa alpha

130 -2.3 - - - - HSP90AB1
(cytosolic), class B member 1
Heat shock protein 90kDa beta

72 - - -2.3 -1.8 - HSP90B1
(Grp94), member 1
DNA repair
93 - - -3.4 - - Valosin-containing protein VCP
X-ray repair cross-complementing

225 -3.9 2.3 - - - XRCC6
protein
Protein degradation
132 -1.5 -1.7 - - - E3 ubiquitin-protein ligase HUWE1
Proteasome alpha 1 subunit,

886 - - - - - PSMA1
isoform 2
784 1.8 -1.7 - - - Proteasome subunit alpha type-2 PSMA2
802 -1.7 - -3.5 - - Proteasome subunit beta type-2 PSMB2
Microtubule
472 - 1.5 - - - Kinectin KTN1

167 - - 3.1 - - Pericentrin PCNT
161 -2.6 - - - - Outer dense fiber protein 2 ODF2
543 -6.5 - - - - Cortactin-binding protein 2 CTTNBP2
512 - - - - - Alstrom syndrome protein 1 ALMS1
Microtubule-associated protein,
667 -4.0 - - - - MAPRE1
RP/EB family, member 1
778 1.8 - - - - Es1 protein, isoform Ia precursor C21ORF33
Others
Leucine-rich repeat transmembrane

333 - - 2.0 - - LRRTM4
neuronal protein 4
Transmembrane protease serine TMPRSS1

523 -1.8 - - - -
13 3
Vacuolar protein sorting-associated

820 3.7 - 2.6 - - VPS13D
protein 13D
127 - - - - - Nipped-B-like protein NIPBL

Table S3. (A) Ingenuity pathway analysis, p<0.05. (B) Ingenuity upstream regulator
analysis, p<0.0001. (C) Predicted PPARGC1A targets. Proteomic and transcriptomic
expression changes relative to hESC-VCMs are shown, fold change>1.5, p<0.05
(A)
P-value of
Ingenuity Canonical Pathways Molecules
overlap
SDHA,PDHA1,PRDX3,NDUFS1,
Mitochondrial Dysfunction 2.95E-10 SOD2,ATP5B,ATP5A1,UQCRFS1,
NDUFS3
Ketolysis 2.29E-06 HADHB,ACAT1,OXCT1
Isoleucine Degradation I 1.23E-05 HADHB,ECHS1,ACAT1
MYH6,MYL2,TPM1,MYH7,ACTA1,
Calcium Signaling 1.74E-05
MYL7
RhoA Signaling 2.95E-05 KTN1,MYL2,EZR,ACTA1,MYL7
Cellular Effects of Sildenafil

4.37E-05 MYH6,MYL2,MYH7,ACTA1,MYL7
(Viagra)
TCA Cycle II (Eukaryotic) 4.68E-05 SDHA,FH,IDH3B
MYH6,MYL2,EZR,MYH7,ACTA1,
Actin Cytoskeleton Signaling 4.68E-05
MYL7
Creatine-phosphate
5.62E-05 CKMT2,CKM
Biosynthesis
Epithelial Adherens Junction

8.32E-05 MYH6,MYL2,MYH7,ACTA1,MYL7
Signaling
Tight Junction Signaling 1.07E-04 MYH6,MYL2,MYH7,ACTA1,MYL7
Fatty Acid Beta-Oxidation I 1.07E-04 HADHB,ECHS1,ACADM
ILK Signaling 2.57E-04 MYH6,MYL2,MYH7,ACTA1,MYL7
Glutaryl-CoA Degradation 5.01E-04 HADHB,ACAT1
Ketogenesis 5.01E-04 HADHB,ACAT1
Signaling by Rho Family

6.76E-04 MYL2,EZR,DES,ACTA1,MYL7
GTPases
Mevalonate Pathway I 8.32E-04 HADHB,ACAT1
Hepatic Fibrosis / Hepatic

8.91E-04 MYH6,MYL2,MYH7,MYL7
Stellate Cell Activation
Cardiomyocyte Differentiation
1.38E-03 MYL2,MYH7
via BMP Receptors
Valine Degradation I 1.38E-03 HADHB,ECHS1
Superpathway of
Geranylgeranyldiphosphate 1.38E-03 HADHB,ACAT1
Biosynthesis I (via Mevalonate)
Tryptophan Degradation III

1.70E-03 HADHB,ACAT1
(Eukaryotic)
RhoGDI Signaling 1.95E-03 MYL2,EZR,ACTA1,MYL7
Regulation of Actin-based
2.09E-03 MYL2,ACTA1,MYL7
Motility by Rho
Superpathway of Cholesterol
3.63E-03 HADHB,ACAT1
Biosynthesis
Integrin Signaling 2.24E-02 MYL2,ACTA1,MYL7
PAK Signaling 3.02E-02 MYL2,MYL7

Fc Receptor-mediated
Phagocytosis in Macrophages 3.39E-02 EZR,ACTA1
and Monocytes
(B)
Upstream P-Value of Number of

Molecule Type
Regulator Overlap Proteins
PPARGC1A transcription regulator 4.83E-14 11
MEF2C transcription regulator 1.90E-10 7
ligand-dependent
ESRRA 2.29E-09 8
nuclear receptor
HDAC5 transcription regulator 2.59E-08 5
TEAD1 transcription regulator 8.42E-08 4
TFAM transcription regulator 1.70E-07 4
GATA4 transcription regulator 2.82E-06 5
MYOCD transcription regulator 3.78E-06 4
ligand-dependent
PPARA 5.67E-06 8
nuclear receptor
HTT transcription regulator 8.10E-06 10
SRF transcription regulator 8.48E-06 7
(C)
Official Proteomic Transcriptomic
gene hF-VCM/ hA-VCM/ hF-VCM/ hA-VCM/
symbol hESC-VCMs hESC-VCMs hESC-VCMs hESC-VCMs
ACADM 2.5 3.7 2.4 -
2.2 3.4
ACADVL - 4.2 2.5 -
- -
1.5 -
ACAT1 2.7 - 7.1 7.1
1.7 2.0
2.0 -
ATP5B - 1.6 - -
CKMT2 2.2 5.5 6.7 10.8
MYH6 -2.0 -11.5 - -
- -
- 4.6
1.9 2.0
NDUFS1 - 4.4 3.6 -
OXCT1 3.6 3.4 13.5 10.8
- 1.7
PRDX3 - 1.7 - -
2.1 -
SOD2 1.7 3.3 - 2.4
- 2.5
SDHA - 2.0 3.0 -

2.4 3.9
Table S4. Pearson correlation coefficient (PCC) between fold changes in
mRNA and protein expression.
Sample pairs PCC
hESC/hESC-VCM 0.36
hF-VCM/hESC-VCM 0.31
hA-VCM/hF-VCM 0.11
Table S5. Comparison between protein expression changes in our study and Van
Hoof et al. Abundance changes between hESC-(V)CMs and hF-(V)CMs were
examined. Up/down indicates higher/lower level in hF-(V)CMs compared to
hESC-(V)CMs. indicates no significant difference or not concordant expression.
Comparison Expression changes
ACAT1 Concordant Up
ACO2 Not concordant -
ALB Not concordant -
ALDH4A1 Concordant Up
ALDH7A1 Concordant -
ANXA1 Not concordant -
ATP5A1 Not concordant -
ATP5B Not concordant -
CCT7 Concordant Down
DES Not concordant -
EEF2 Not concordant -
ETFB Not concordant -
FH Not concordant -
FHL2 Concordant Up
GANAB Not concordant -
GAPDH Not concordant -

GARS Not concordant -
GNB2 Not concordant -
GNB2L1 Concordant -
HADHB Not concordant -
HSP90AB1 Not concordant -
HUWE1 Concordant Down
IDH2 Not concordant -
IDH3A Not concordant -
KRT9 Not concordant -
KTN1 Not concordant -
MYBPC3 Not concordant -
MYH6 Not concordant -
MYH7 Concordant Up
PDHA1 Not concordant -
PHGDH Not concordant -
PKM2 Concordant -
PRDX1 Concordant -
UQCRFS1 Not concordant -
VCP Not concordant -
VDAC2 Concordant Up
Proteomic Analysis of Human Pluripotent Stem Cell-Derived, Fetal and Adult Ventricular
Cardiomyocytes Reveals Pathways Crucial for Cardiac Metabolism and Maturation
Ellen Poon, Wendy Keung, Y.M. Liang, Rajkumar Ramalingam, Bin Yan, Shaohong Zhang, Anant
Chopra, Jennifer Moore, Anthony Herren, Deborah K. Lieu, Hau San Wong, Zhihui Weng, On Tik
Wong, Yun Wah Lam, Gordon F. Tomaselli, Christopher Chen, Kenneth R. Boheler and Ronald A. Li
Circ Cardiovasc Genet. published online March 10, 2015;

Circulation: Cardiovascular Genetics is published by the American Heart Association, 7272 Greenville Avenue, Dallas,
TX 75231
Copyright 2015 American Heart Association, Inc. All rights reserved.
Print ISSN: 1942-325X. Online ISSN: 1942-3268
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circgenetics.ahajournals.org/content/early/2015/03/10/CIRCGENETICS.114.000918
Data Supplement (unedited) at:

http://circgenetics.ahajournals.org/content/suppl/2015/03/10/CIRCGENETICS.114.000918.DC1.html
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Circulation: Cardiovascular Genetics can be obtained via RightsLink, a service of the Copyright Clearance Center,
not the Editorial Office. Once the online version of the published article for which permission is being requested is
located, click Request Permissions in the middle column of the Web page under Services. Further information about
this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:

http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation: Cardiovascular Genetics is online at:

http://circgenetics.ahajournals.org//subscriptions/

Proteomics Cardiac Metab

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Proteomics Cardiac Metab

Uploaded by

Copyright:

Available Formats

DOI: 10.1161/CIRCGENETICS.114.

Proteomic Analysis of Human Pluripotent Stem Cell-Derived, Fetal and Adult

Running title: Poon et al.; Global proteomic profiling of hESC-VCMs

models to investigate mechanisms involved in human cardiac development and disease

progression. Successful application of hESC-VCMs to myocardial repair or modeling requires

electrophysiological, contractile and metabolic functions7. Indeed, hESC-CMs

VCMs has nott been

show differential expression of specific cardiac ion channels, complementary to known

functional data13-17. For obtaining a better understanding, here we employed a combinatorial

approach of performing proteomic, bioinformatics and functional analyses of hESCs, hESC-

pathways important for maturation.

Culturing of hESCs, cardiac differentiation and selection of hESC-VCMs

DMEM supplemented with 0QRQHVVHQWLDODPLQRDFLGVP0JOXWDPLQH8PO

three days post-WUDQVGXFWLRQDQG]HRFLQ JPO ZDVDSSOLHGWRNLOOQRQ-VCMs. Please see

Figure S1 for characterisation of hESC-VCMs.

Isolation of fetal and adult CMs

and labelled with CyDye-2. CyDye-labeled protein samples were mL[HGDQGOZDVORDGHG

Healthcare/Amersham). Analyses were performed using the Profenesiss Samespot v2.3

For more ddetails,

Functional annotation was performed based on Gene Ontology classifications (GO;

//www.geneontology.org). For differential protein abundance, proteins satisfying expression fold

Pathway analysis of differentially expressed proteins

(http://www.ingenuity.com/) to evaluate their association with corresponding canonical pathway

Quantitative Real-time PCR (qRT-PCR)

PHDVXUHGXVLQJJ WKH ZHOOO--establ

IOXRUHVFHQFH aQP 7KHUHIRUHWKHUDWLRRIUHGJUHHQVLJQDOLVDQLQGLFDWLRQRIP+(6&-

Quantification of contractile force in human ventricular cardiac microtissues (hvCMT)

Measurements were made on day 6 of T3 treatment.

Measurements of action potential

ULTIMA, SciMedia USA L

stimulation was applied to evoke action potentials.

Proteomic profiling was performed using 2D-Differential-In-Gel Electrophoresis (DIGE) on

As an example, Figure 1A shows the separation of proteins present in undifferentiated hESCs

that were equally expressed in both samples.

unique gene products. Cross analyses enabled us to examine the protein

these samples. As an example (Figure 2B-C), the expression of MLC2V (MYL2)

data were entirely

clustered closely together, consistent with a high experimental reproducibility. Although

comparable global transcriptomes26.

Identification of proteomic changes in cardiac contraction and metabolism

decreased in hF-VCMs compared to hESC-VCMs, respectively. Expression changes in

expression among different functional groups in the three CM populations.

survival, ionn transport

the lowest OHYHOVRIVDUFRPHULFSURWHLQVVXFKDV-Myosin

developed twitchh force

14.0-fold expression differences between hESC-VCMs and hF-VCMs. Indeed, glycogen

difference by being >14-fold enriched in hF-VCMs compared to hESC-VCMs. In addition, we

creatine kinase (CKM) and mitochondrial CK (CKMT2) in hF-VCMs relative to hESC-VCMs.

increased mitochondrial maturation.

Cardiac metabolism and maturation of hESC-VCMs

To seek further insights, we performed the complementary approach of combining clustering,

Ingenuity pathway and upstream regulator analyses of differentially expressed proteins, to

Figure 5C, we detected PPARGC1A (peroxisome proliferator-activated receptor gamma

coactivator 1-alpha)/ ESRRA(estrogen-related receptor alpha)/ PPARA (peroxisome proliferator-

hESC-VCMs. Of the 11 proteins targeted by PPARGC1A, the transcripts of 9 were indeed

significantly upregulated in hA-VCMs and/or hF-VCMs relative to hESC-VCMs (Figure 5E).

metabolism and maturation of hESC-VCMs was further investigated. To seek functional

evidence, we next tested the pharmacological effect of WY-14643, a well-characterized PPARA

agonist28, RQP of hESC-VCMs using the JC-G\HP is an index of cellular metabolic

PE\LQK(6&-VCMs in the presence of oleic acid (p<0.05; Figure 6A); conversely,

WY-14643-treated hESC-VCMs cultured without oleic acid supplement had an Psimilar to

control (p>0.05). Along this line, WY-14643 (50M) significantly increased

of genes involved in FA metabolism such as medium-/ long-/ very long-

mitochondrial morphology, control hESC-VCMs had disorganised punctate and perinuclear

DMEM supplemented with 0QRQHVVHQWLDODPLQRDFLGVP0JOXWDPLQH8PO

three days post-WUDQVGXFWLRQDQG]HRFLQJPOZDVDSSOLHGWRNLOOQRQ-VCMs. Please see

and labelled with CyDye-2. CyDye-labeled protein samples were mL[HGDQGOZDVORDGHG

IOXRUHVFHQFHaQP7KHUHIRUHWKHUDWLRRIUHGJUHHQVLJQDOLVDQLQGLFDWLRQRIP+(6&-

agonist28, RQP of hESC-VCMs using the JC-G\HP is an index of cellular metabolic

PE\LQK(6&-VCMs in the presence of oleic acid (p<0.05; Figure 6A); conversely,