18162134.conll

BackgroundMuch O
of O
our O
current O
knowledge O
of O
the O
molecular O
expression O
profile O
of O
human B-Species
embryonic I-CellType
stem I-CellType
cells I-CellType
(hESCs) O
is O
based O
on O
transcriptional O
approaches. O
These O
analyses O
are O
only O
partly O
predictive O
of O
protein O
expression O
however, O
and O
do O
not O
shed O
light O
on O
post-translational O
regulation, O
leaving O
a O
large O
gap O
in O
our O
knowledge O
of O
the O
biology O
of O
pluripotent B-CellType
stem I-CellType
cells.ResultsHere O
we O
describe O
the O
use O
of O
two O
large-scale O
western O
blot O
assays O
to O
identify O
over O
600 O
proteins O
expressed O
in O
undifferentiated O
hESCs, O
and O
highlight O
over O
40 O
examples O
of O
multiple O
gel O
mobility O
variants, O
which O
are O
suspected O
protein O
isoforms O
and/or O
post-translational O
modifications. O
Twenty-two O
phosphorylation O
events O
in O
cell O
signaling O
molecules, O
as O
well O
as O
potential O
new O
markers O
of O
undifferentiated O
hESCs B-CellType
were O
also O
identified. O
We O
confirmed O
the O
expression O
of O
a O
subset O
of O
the O
identified O
proteins O
by O
immunofluorescence O
and O
correlated O
the O
expression O
of O
transcript O
and O
protein O
for O
key O
molecules O
in O
active O
signaling O
pathways O
in O
hESCs. O
These O
analyses O
also O
indicated O
that O
hESCs B-CellType
exhibit O
several O
features O
of O
polarized O
epithelia, O
including O
expression O
of O
tight O
junction O
proteins.ConclusionOur O
approach O
complements O
proteomic O
and O
transcriptional O
analysis O
to O
provide O
unique O
information O
on O
human B-Species
pluripotent I-CellType
stem I-CellType
cells, O
and O
is O
a O
framework O
for O
the O
continued O
analyses O
of O
self-renewal. O
Human B-Species
embryonic I-CellType
stem I-CellType
cells I-CellType
(hESCs) O
are O
pluripotent O
cells O
isolated O
from O
the O
inner B-Anatomy
cell I-Anatomy
mass I-Anatomy
of O
the O
blastocyst B-Anatomy
[1]. O
They O
can O
be O
maintained O
for O
prolonged O
periods O
in O
culture O
and O
differentiate O
to O
representatives O
of O
the O
three O
germ B-Anatomy
layers I-Anatomy
as O
well O
as O
trophoblasts B-CellType
and O
germ B-CellType
cells. O
This O
differentiation O
potential O
may O
be O
used O
to O
model O
certain O
aspects O
of O
human B-Species
embryogenesis, O
including O
the O
development O
and O
differentiation O
of O
pluripotent O
and O
other O
stem O
cell O
types O
during O
the O
processes O
of O
gastrulation, O
neurogenesis B-Anatomy
and O
organogenesis. O
Thus, O
hESCs B-CellType
provide O
a O
unique O
and O
powerful O
system O
to O
study O
otherwise O
intractable O
aspects O
of O
human B-Species
development. O
Furthermore, O
these O
approaches O
have O
the O
potential O
to O
provide O
differentiated O
cell O
types O
for O
cell O
replacement O
therapies O
of O
degenerative O
disorders O
such O
as O
Parkinson's O
disease O
and O
Type O
I O
diabetes O
[2,3]. O
Before O
these O
cell O
therapy O
applications O
are O
developed, O
an O
understanding O
of O
the O
molecular O
and O
cellular O
mechanisms O
that O
drive O
self-renewal O
and O
differentiation O
is O
required. O
Fundamental O
to O
this O
understanding O
is O
the O
elucidation O
of O
the O
transcriptome O
and O
proteome O
of O
hESCs, O
using O
approaches O
that O
lay O
a O
framework O
for O
functional O
analyses O
of O
the O
unique O
properties O
of O
these O
cells. O
Large-scale O
gene O
expression O
analyses O
such O
as O
microarray, O
massive O
parallel O
signature O
sequencing O
(MPSS), O
expressed O
sequenced O
tag O
(EST) O
enumeration, O
and O
serial O
analysis O
of O
gene O
expression O
(SAGE) O
have O
been O
used O
to O
compare O
multiple O
hESC O
lines O
[4-7]; O
hESCs B-CellType
to O
germ B-CellType
cell I-Anatomy
tumors I-Anatomy
[8]; O
or O
to O
differentiated O
derivatives O
in O
embryoid B-Anatomy
bodies I-Anatomy
[9-11] O
or O
neural B-CellType
populations I-CellType
[12]. O
These O
approaches O
have O
highlighted O
an O
expanded O
set O
of O
transcripts O
that O
mark O
the O
pluripotent O
state O
[4,13,14], O
cross-species O
commonalities O
in O
the O
molecular O
profile O
of O
ESCs B-CellType
[6,12,15], O
prominent O
receptors O
expressed O
by O
hESCs B-CellType
[8] O
and O
pathways O
that O
may O
play O
a O
role O
in O
the O
regulation O
of O
pluripotency O
[16,17]. O
Nevertheless, O
cataloguing O
the O
cellular O
transcriptome O
is O
only O
predictive O
of O
protein O
expression O
and O
typically O
does O
not O
shed O
light O
on O
post-transcriptional O
regulation. O
For O
example, O
while O
tens O
of O
thousands O
of O
transcripts O
can O
be O
followed O
simultaneously O
with O
SAGE, O
microarrays O
and O
MPSS, O
these O
methods O
do O
not O
routinely O
detect O
differences O
in O
transcript O
splice O
variants, O
or O
polyadenylation O
status. O
These O
differences O
may O
have O
profound O
effects O
on O
translation, O
as O
well O
as O
the O
isoform O
and O
function O
of O
the O
protein O
produced. O
Finally, O
numerous O
post-translational O
modifications O
are O
known O
to O
regulate O
protein O
function, O
including O
enzymatic O
cleavage, O
covalent O
coupling O
to O
other O
molecules, O
glycosylation, O
phosphorylation O
and O
ubiquitination. O
These O
issues O
all O
highlight O
potential O
shortfalls O
in O
our O
understanding O
of O
the O
hESC O
proteome. O
Several O
practical O
approaches O
for O
proteomic O
analyses O
are O
currently O
available, O
the O
most O
established O
of O
which O
is O
the O
2-dimensional O
(2D) O
separation O
of O
proteins O
by O
polyacrylamide O
gel O
electrophoresis O
(PAGE). O
HPLC-tandem O
mass O
spectrometry O
(HPLC-MS/MS) O
based O
technology O
is O
rapidly O
evolving O
and O
has O
recently O
been O
used O
to O
detect O
protein O
expression O
in O
multiple O
cell O
types. O
An O
alternate O
approach O
is O
the O
recent O
large-scale O
adaptation O
of O
standard O
western O
blotting O
[18]. O
In O
this O
procedure, O
a O
large O
well O
is O
used O
to O
separate O
the O
sample O
by O
PAGE O
and O
lanes O
are O
created O
on O
the O
membrane O
containing O
immobilized O
protein O
with O
the O
use O
of O
a O
manifold. O
Compatible O
combinations O
of O
primary O
antibodies O
are O
predetermined, O
with O
the O
criterion O
of O
being O
able O
to O
identify O
proteins O
that O
do O
not O
co-migrate. O
Different O
combinations O
of O
primary O
antibodies O
are O
added O
to O
each O
well, O
with O
appropriate O
dilutions O
of O
each O
primary O
antibody O
so O
that O
expressed O
proteins O
are O
detected O
in O
a O
single O
condition. O
The O
scalability O
of O
the O
system O
depends O
on O
defining O
suitable O
combinations O
of O
primary O
antibodies, O
with O
up O
to O
1000 O
antibodies O
in O
200 O
lanes O
being O
used O
in O
the O
largest O
screens O
thus O
far. O
Detection O
software O
is O
used O
to O
identify O
proteins O
based O
on O
their O
expected O
and O
observed O
gel O
mobility. O
Unlike O
2D O
PAGE O
and O
HPLC-MS/MS, O
large-scale O
western O
blotting O
only O
identifies O
proteins O
for O
which O
antibodies O
are O
already O
available. O
While O
this O
is O
not O
an O
appropriate O
screen O
for O
identifying O
uncharacterized O
proteins, O
it O
greatly O
simplifies O
the O
verification O
and O
functional O
analyses O
of O
proteins O
that O
are O
detected. O
In O
addition, O
this O
approach O
is O
highly O
flexible, O
and O
if O
desired O
can O
be O
focused O
to O
particular O
sets O
of O
proteins O
or O
protein O
function, O
such O
as O
cell O
signaling O
molecules. O
Importantly, O
the O
foundation O
of O
this O
approach O
is O
the O
large O
amount O
of O
data O
on O
individual O
antibodies, O
which O
are O
already O
available O
and O
characterized O
in O
the O
literature. O
More O
recently, O
two O
research O
groups O
have O
conducted O
proteomic O
analyses O
of O
hESCs B-CellType
using O
MS O
[19-22]. O
In O
the O
present O
study, O
we O
used O
two O
large-scale O
western O
blot O
systems O
to O
examine O
the O
expression O
of O
> O
1000 O
proteins O
in O
hESCs B-CellType
and O
detected O
> O
600 O
proteins O
that O
were O
grouped O
into O
18 O
functional O
classes. O
In O
addition, O
we O
identified O
42 O
examples O
of O
multiple O
bands O
for O
a O
single O
protein, O
likely O
to O
be O
protein O
isoforms O
and/or O
post-translational O
modifications, O
and O
22 O
phosphorylation O
events O
in O
cell O
signaling O
molecules. O
We O
correlated O
the O
expression O
of O
members O
of O
key O
active O
pathways O
in O
our O
transcriptional O
and O
proteomic O
databases O
and O
confirmed O
the O
validity O
of O
this O
approach. O
Using O
these O
approaches O
we O
identified O
new O
markers O
for O
undifferentiated O
hESCs B-CellType
and O
highlighted O
unrecognized O
epithelial O
characteristics O
of O
hESCs. O
Our O
data O
confirm O
the O
importance O
of O
proteomic O
analyses O
in O
complementing O
transcriptional O
profiling O
and O
provide O
a O
framework O
for O
continued O
analyses O
of O
the O
molecular O
and O
cellular O
biology O
of O
pluirpotent O
hESCs. O
PowerBlot O
analysis O
of O
hESCsWe O
first O
employed O
a O
large-scale O
western O
blot O
screen, O
the O
PowerBlot O
system, O
to O
profile O
protein O
expression O
in O
undifferentiated O
hESCs. O
This O
system O
used O
934 O
antibodies O
toward O
proteins O
representing O
22 O
diverse O
classes O
of O
function, O
such O
as O
transcription O
factors, O
the O
MAP B-GeneProtein
kinase I-GeneProtein
(MAPK) O
pathway, O
and O
apoptosis, O
among O
others. O
To O
expand O
a O
large-scale O
culture O
of O
BG01 B-CellLine
cells I-CellType
for O
this O
assay, O
a O
collagenase- O
and O
trypsin- O
based O
passaging O
method O
was O
used O
[23]. O
While O
these O
conditions O
have O
been O
associated O
with O
the O
accumulation O
of O
trisomies O
of O
chromosomes O
12, O
17 O
and O
X O
[24], O
the O
ease O
of O
use O
of O
these O
cultures O
and O
similarity O
in O
gene O
expression O
and O
differentiation O
potential O
to O
karyotypically O
normal O
BG01 B-CellLine
hESCs I-CellType
[11,24,25] O
make O
them O
suitable O
for O
such O
large O
scale O
applications. O
For O
the O
PowerBlot O
screen, O
whole O
cell O
lysate O
from O
BG01 B-CellLine
hESCs I-CellType
was O
separated O
on O
five O
4–15% O
gradient O
gels. O
Each O
blot O
contained O
size O
markers O
and O
39 O
lanes. O
Each O
lane O
was O
screened O
with O
1–8 O
antibodies O
in O
combinations O
that O
had O
been O
predetermined O
to O
enable O
accurate O
identification O
of O
well-separated O
proteins O
(Fig. O
1A–E). O
The O
gels O
and O
blots O
were O
performed O
in O
duplicate O
and O
expressed O
proteins O
were O
identified O
by O
their O
predicted O
size O
and O
verified O
by O
visual O
inspection.Figure O
1PowerBlot O
analysis O
of O
undifferentiated O
BG01 B-CellLine
hESCs. O
This O
large-scale O
western O
blot O
consisted O
of O
five O
gels O
run O
in O
duplicate O
and O
probed O
with O
934 O
antibodies. O
(A-E) O
One O
set O
of O
blots O
is O
shown O
at O
a O
contrast O
that O
highlights O
most O
bands. O
(F) O
A O
representative O
lane O
(gel O
C, O
lane O
24) O
aligned O
with O
protein O
markers O
used O
for O
band O
identification. O
(G) O
Scatter-plot O
of O
the O
normalized O
average O
intensity O
(i.u.) O
values O
for O
each O
protein O
indicating O
a O
linear O
relationship O
between O
duplicate O
blots. O
Datasets O
for O
this O
analysis O
are O
in O
Additional O
Tables O
1 O
and O
2.A O
total O
of O
545 O
antibodies O
detected O
bands O
of O
appropriate O
size, O
which O
could O
be O
compressed O
to O
529 O
proteins O
with O
unique O
SwissProt O
identification O
numbers O
(Fig. O
1A–E O
and O
Additional O
File O
1). O
An O
enlargement O
of O
a O
representative O
lane O
(lane O
24 O
of O
Blot O
C) O
alongside O
protein O
markers O
is O
shown O
in O
Fig. O
1F. O
Thirteen O
proteins O
including O
AKT, O
caveolin1 B-GeneProtein
and O
ERK1 B-GeneProtein
were O
detected O
in O
multiple O
lanes O
using O
the O
same O
or O
different O
antibodies. O
Information O
on O
the O
antibody O
catalogue O
number O
and O
dilution, O
band O
intensity O
for O
each O
repeat O
and O
the O
averaged O
value, O
description O
of O
protein O
function, O
and O
Entrez O
gene O
and O
SwissProt O
database O
identification O
numbers O
is O
shown O
in O
Additional O
File O
1. O
Three O
hundred O
and O
eighty O
three O
antibodies O
did O
not O
detect O
bands O
in O
this O
screen, O
indicating O
lack O
of O
expression, O
or O
possibly O
technical O
issues O
with O
detection O
under O
standard O
conditions O
(Additional O
File O
1).The O
size O
of O
the O
detected O
proteins O
ranged O
from O
15 O
kD O
(GS15) O
to O
280 O
kD O
(ABP-280). O
The O
average O
intensity O
of O
the O
detected O
proteins O
ranged O
from O
195 O
to O
117926 O
normalized O
intensity O
units O
(i.u.), O
with O
an O
average O
of O
5367 O
i.u. O
The O
proteins O
with O
the O
highest O
band O
intensity O
were O
the O
B2 B-GeneProtein
Bradykinin I-GeneProtein
Receptor I-GeneProtein
(117926 O
i.u.), O
Karyopherin B-GeneProtein
α I-GeneProtein
(80698 O
i.u.), O
and O
BiP B-GeneProtein
(74922 O
i.u.), O
whilst O
the O
proteins O
with O
the O
lowest O
intensity O
that O
could O
be O
verified O
by O
visual O
inspection O
were O
Inhibitor B-GeneProtein
2 I-GeneProtein
(247 O
i.u.), O
Caspase B-GeneProtein
8 I-GeneProtein
(201 O
i.u.), O
and O
OXA1Hs B-GeneProtein
(195 O
i.u.). O
Finally, O
the O
consistency O
of O
this O
assay O
was O
demonstrated O
by O
plotting O
the O
normalized O
average O
intensity O
values O
for O
each O
protein, O
which O
revealed O
a O
linear O
relationship O
between O
the O
duplicate O
samples O
(Fig. O
1G). O
Kinexus O
analysis O
of O
hESCsA O
more O
focused O
screen O
was O
used O
to O
profile O
expression O
of O
protein B-GeneProtein
kinases, O
phosphatases B-GeneProtein
and O
phosphorylated O
sites O
in O
cell O
signaling O
molecules O
in O
hESCs. O
The O
Kinexus O
assays O
contained O
140 O
antibodies O
to O
these O
related O
classes O
of O
proteins O
and O
phospho-sites. O
Karyotypically O
normal O
BG03 B-CellLine
hESCs B-CellType
grown O
on O
a O
fibronectin B-GeneProtein
matrix O
in O
MEF-CM O
[26] O
were O
used O
for O
this O
analysis, O
and O
whole O
cell O
lysate O
was O
separated O
on O
four O
12.5% O
gels O
for O
western O
blotting. O
Eighty O
five O
immunoreactive O
bands O
were O
identified, O
representing O
38 O
protein B-GeneProtein
kinases I-GeneProtein
and O
16 O
phosphatases, O
their O
isoforms, O
and O
22 O
phosphorylated O
sites O
in O
signaling O
molecules O
(Fig. O
2A–D, O
Additional O
File O
1). O
Sixty-four O
antibodies O
did O
not O
detect O
their O
corresponding O
antigen O
(Additional O
File O
1).Figure O
2Kinexus O
blots O
of O
undifferentiated O
BG03 B-CellLine
cells. O
Four O
blots O
were O
used O
to O
probe O
BG03 B-CellLine
lysate O
with O
(A, O
B) O
76 O
antibodies O
for O
protein B-GeneProtein
kinases, O
(C) O
27 O
antibodies O
for O
phosphatases B-GeneProtein
and O
(D) O
37 O
antibodies O
for O
phosphoylated O
sites O
in O
cell O
signaling O
molecules. O
Identified O
bands O
are O
indicated O
(*). O
Datasets O
for O
this O
analysis O
are O
in O
Additional O
Tables O
1 O
and O
2. O
Functional O
classification O
of O
proteins O
expressed O
in O
hESCsThe O
PowerBlot O
and O
Kinexus O
assays O
identified O
a O
diverse O
range O
of O
proteins O
expressed O
in O
hESCs. O
To O
further O
annotate O
these O
data, O
the O
detected O
proteins O
were O
ordered O
into O
18 O
subgroups O
based O
on O
protein O
function O
(Additional O
File O
2). O
For O
example, O
16 O
factors O
with O
known O
or O
implied O
roles O
in O
the O
regulation O
of O
self-renewal O
or O
pluripotency O
of O
mESCs B-CellType
or O
hESCs, O
such O
as O
Oct4 B-GeneProtein
[27], O
STAT3 B-GeneProtein
[28], O
members O
of O
the O
FGF B-GeneProtein
[29], O
PI3 B-GeneProtein
kinase I-GeneProtein
[30], O
Src B-GeneProtein
[31] O
or O
MAPK B-GeneProtein
pathways O
[32], O
and O
phosphorylated O
isoforms O
of O
GSK3, O
STAT3 B-GeneProtein
and O
p38 B-GeneProtein
MAPK, O
were O
grouped O
under O
"Pluripotency" O
(Fig. O
3A O
and O
Additional O
File O
2). O
Another O
functional O
group O
(Cell O
surface) O
consisted O
of O
20 O
transmembrane O
or O
cell B-GeneProtein
surface I-GeneProtein
proteins I-GeneProtein
(Additional O
File O
2). O
This O
included O
several O
receptors O
for O
peptides O
and O
growth O
factors, O
such O
as O
neurotensin B-GeneProtein
receptor I-GeneProtein
3, O
the O
B2 B-GeneProtein
bradykinin, O
endothelin B-GeneProtein
1, O
and O
thrombin B-GeneProtein
receptors, O
and O
the O
glial B-GeneProtein
derived I-GeneProtein
neurotrophic I-GeneProtein
factor I-GeneProtein
receptor I-GeneProtein
α I-GeneProtein
(Fig. O
3B). O
These O
molecules O
may O
be O
useful O
as O
targets O
for O
cell O
sorting O
experiments, O
and O
expression O
of O
these O
receptors O
could O
identify O
bioactive O
peptides O
or O
growth O
factors O
that O
may O
influence O
hESC O
self-renewal O
or O
differentiation.Figure O
3Functional O
classification O
and O
mobility O
variants O
of O
proteins O
detected O
in O
hESCs. O
(A) O
Proteins O
with O
known O
or O
suggested O
roles O
in O
self-renewal O
are O
shown, O
including O
Oct4, O
STAT3, O
Smad2/3 B-GeneProtein
and O
FGF2 B-GeneProtein
(Additional O
Table O
2, O
"Pluripotency"). O
Isoforms O
of O
FGF2, O
and O
phospho-GSK3 B-GeneProtein
are O
indicated O
(*). O
(B) O
Cell O
surface O
proteins O
are O
shown, O
including O
Connexin B-GeneProtein
43, O
E-Cad B-GeneProtein
and O
GDNFRα B-GeneProtein
(Additional O
Table O
2, O
"Cell O
Surface"). O
Other O
functional O
classes O
of O
proteins O
are O
indicated O
in O
Additional O
Table O
2. O
(C) O
A O
total O
of O
42 O
proteins, O
including O
FGF2, O
HSP70 B-GeneProtein
and O
ERK1, O
were O
found O
to O
have O
multiple O
bands O
in O
either O
the O
PowerBlot O
or O
Kinexus O
blots. O
These O
bands O
migrated O
closely O
but O
were O
sufficiently O
separated O
from O
other O
detected O
proteins. O
Bands O
predicted O
to O
be O
isoforms O
of O
the O
indicated O
protein O
are O
highlighted O
in O
some O
panels O
(*).Other O
functional O
classification O
of O
the O
proteins O
detected O
by O
the O
PowerBlot O
screen O
included: O
transcription O
factors O
(71 O
proteins), O
nucleus B-CellComponent
and O
nuclear B-CellComponent
transport O
(144), O
cytoskeleton B-CellComponent
(75), O
cell O
adhesion O
(45), O
MAP B-GeneProtein
kinase I-GeneProtein
pathway O
(24), O
protein B-GeneProtein
kinase I-GeneProtein
A I-GeneProtein
(13), O
protein B-GeneProtein
kinase I-GeneProtein
C I-GeneProtein
(20), O
tyrosine B-GeneProtein
kinases I-GeneProtein
(15), O
adaptors O
and O
tyrosine O
kinase O
substrates O
(51), O
protein B-GeneProtein
phosphatases I-GeneProtein
(17), O
GTPases B-GeneProtein
and O
regulators O
(42), O
calcium O
signaling O
(23), O
cell O
cycle O
(87), O
apoptosis O
(61), O
membrane B-CellComponent
research O
(62), O
and O
other O
functions O
(51) O
(Additional O
File O
1). O
Some O
proteins O
were O
included O
in O
multiple O
functional O
categories O
due O
to O
overlapping O
properties, O
such O
as O
AIM-1, O
which O
was O
included O
in O
the O
cell O
cycle O
as O
well O
as O
in O
the O
nucleus/nuclear O
transport O
categories. O
The O
Kinexus O
expression O
data O
was O
organized O
separately O
into O
cell O
signaling-related O
functional O
groups O
(Additional O
File O
1). O
In O
addition, O
35 O
proteins O
were O
detected O
by O
both O
the O
PowerBlot O
and O
Kinexus O
systems O
(Table O
1).Table O
1Proteins O
detected O
by O
both O
PowerBlot O
and O
Kinexus O
systemsProtein O
nameSwiss O
NrProtein O
nameSwiss O
NrBMXP51813MEK2P36506CaM O
Kinase I-GeneProtein
KinaseQ64572MKP2Q62767Casein O
Kinase I-GeneProtein
I I-GeneProtein
epsilonP49674p38 O
alpha/SAPK2aQ16539Casein O
Kinase I-GeneProtein
II I-GeneProtein
alpha/CK2aP19139PaxillinP49024Cdk1/Cdc2P06493PKA O
CP17612Cdk5Q00535PKC O
betaP05771Cdk7P50613PKC O
deltaQ05655DAP O
KinaseP53355PP2A O
Catalytic O
alphaP05323DAP3P51398PP5/PPTP53042ERK1Q63538PTP1BP18031ERK2P27703PTP1C/SHP1P29350FAKQ00944PTP1D/SHP2Q06124GSK-3 O
betaP18266RbP13405I O
kappa I-GeneProtein
B I-GeneProtein
alphaP25963RskQ15418IKK O
betaO14920Stat1A46159JAK1P23458Stat3P52631JNK1P45983VHRP51452MEK1Q02750 O
Detection O
of O
protein O
isoforms O
or O
post-translational O
variantsUnlike O
many O
cDNA-based O
gene O
expression O
assays, O
western O
blotting O
has O
the O
capacity O
to O
detect O
multiple O
protein O
isoforms O
due O
to O
translation O
of O
different O
mRNA O
splice O
variants, O
as O
well O
as O
post-translational O
modifications O
such O
as O
enzymatic O
cleavage, O
glycosylation, O
or O
phosphorylation. O
Examination O
of O
the O
blots O
described O
here O
identified O
42 O
examples O
of O
multiple O
banding O
for O
a O
single O
target O
antigen O
(Fig. O
3C). O
These O
candidates O
exhibited O
closely O
migrating O
multiple O
bands, O
which O
were O
close O
to O
their O
predicted O
size O
but O
were O
sufficiently O
separated O
from O
other O
proteins. O
For O
example, O
four O
closely O
migrating O
bands O
were O
observed O
for O
FGF2 B-GeneProtein
(Fig. O
3C, O
top O
panel), O
which O
may O
represent O
known O
glycosylation O
variants O
of O
this O
growth O
factor O
[33]. O
Other O
known O
examples O
of O
post-translational O
modifications O
included O
those O
of O
HSP70, O
IKKgamma B-GeneProtein
and O
ERK1. O
Verification O
of O
protein O
expression O
by O
immunocytochemistryThe O
PowerBlot O
and O
Kinexus O
assays O
identified O
proteins O
based O
on O
their O
expected O
and O
observed O
molecular O
weight, O
using O
combinations O
of O
antibodies O
that O
had O
been O
predetermined O
to O
detect O
proteins O
of O
sufficiently O
different O
sizes. O
Proteins O
known O
to O
be O
expressed O
by O
hESCs B-CellType
and O
also O
identified O
by O
these O
assays, O
included O
Oct4, O
E-CAD, O
Connexin B-GeneProtein
43 I-GeneProtein
and O
Hsp70. O
To O
verify O
expression O
using O
a O
complementary O
approach, O
we O
performed O
immunoflurorescent O
staining O
for O
10 O
proteins O
not O
previously O
reported O
to O
be O
expressed O
in O
hESCs B-CellType
by O
immunocytochemistry, O
using O
karyotypically O
normal O
BG01 B-CellLine
cultures O
(Fig. O
4A–K). O
These O
included O
ABP-280, O
a O
homodimeric O
actin-binding O
protein O
often O
associated O
with O
membrane O
glycoproteins; O
CtBP1 B-GeneProtein
and O
CtBP2, O
two O
C O
terminal O
binding O
proteins O
that O
are O
a O
class O
of O
transcription O
corepressors; O
GS-28, O
a O
golgi B-CellComponent
protein; O
HDJ-2, O
a O
member O
of O
the O
DnaJ-related O
Hsp40 B-GeneProtein
(heat O
shock I-GeneProtein
protein I-GeneProtein
40) O
subfamily; O
L-Caldesmon, O
a O
cytoplasmic O
actin-binding O
protein; O
Rabaptin, O
a O
GTP-binding O
protein; O
phosphorylated-p130 B-GeneProtein
Cas, O
a O
docking O
protein O
with O
an O
amino-terminal O
SH3 O
domain O
that O
may O
function O
as O
a O
molecular O
switch O
that O
regulates O
CAS O
(Crk-associated O
substrate) O
tyrosine O
phosphorylation; O
Ras-GAP B-GeneProtein
and O
phosphorylated B-GeneProtein
Ras-GAP I-GeneProtein
(p-Y460), O
a O
protein O
that O
down-regulates O
the O
signal O
transducer O
p21ras; O
and O
ShcC, O
a O
protein O
with O
an O
N-terminal O
phosphotyrosine-binding O
domain. O
These O
proteins O
were O
all O
expressed O
by O
hESCs, O
with O
the O
expected O
subcellular O
localization O
(Fig. O
4A–K). O
Oct4 B-GeneProtein
was O
used O
as O
a O
positive O
control O
(Fig. O
4L). O
These O
results O
suggested O
that O
most O
of O
the O
bands O
in O
the O
PowerBlot O
and O
Kinexus O
assays O
were O
likely O
to O
be O
correctly O
identified.Figure O
4Verification O
of O
protein O
expression O
using O
immunocytochemistry. O
(A-K) O
Ten O
proteins O
that O
were O
detected O
in O
undifferentiated O
hESCs B-CellType
by O
western O
blotting O
were O
also O
detected O
by O
immunofluorescence O
of O
BG01 B-CellLine
cells I-CellType
grown O
in O
MEF-CM. O
Ras-GAP B-GeneProtein
(pY460) O
is O
a O
phosphorylated O
form O
of O
Ras-GAP. O
The O
same O
antibodies O
were O
used O
in O
this O
analysis O
as O
in O
the O
PowerBlot O
assay, O
except O
phospho-p130 B-GeneProtein
Cas I-GeneProtein
(Tyr165). O
(L) O
Oct4 B-GeneProtein
was O
used O
as O
a O
positive O
control. O
(M-R) O
Oct4, O
TNIK B-GeneProtein
and O
p130 B-GeneProtein
Cas I-GeneProtein
as O
markers O
of O
undifferentiated O
hESCs. O
BG01 B-CellLine
cultures O
were O
partially O
differentiated O
by O
exposure O
to O
10% O
fetal O
bovine O
serum O
for O
3 O
days. O
(M) O
Oct4 B-GeneProtein
was O
expressed O
uniformly O
in O
undifferentiated O
cells, O
(P) O
but O
was O
downregulated O
in O
morphologically O
differentiated O
areas O
after O
3 O
days O
in O
serum O
(arrowhead). O
(N) O
TNIK B-GeneProtein
expression O
was O
localized O
to O
the O
cytoplasm, O
and O
(N, O
Q) O
expression O
appeared O
to O
be O
restricted O
to O
morphologically O
undifferentiated O
cells O
(arrowhead). O
(O) O
p130 B-GeneProtein
Cas I-GeneProtein
was O
detected O
in O
a O
membrane/peripheral-cytoplasmic O
pattern O
in O
undifferentiated O
cells, O
(R) O
but O
this O
distribution O
was O
substantially O
altered O
in O
differentiating O
cells O
with O
a O
flattened O
morphology, O
which O
exhibited O
a O
general O
cytoplasmic, O
or O
perinuclear B-CellComponent
profile. O
Scale O
bar O
for O
A-L: O
(A, O
L) O
200 O
μm; O
(C, O
D, O
F, O
H, O
I, O
J, O
K) O
100 O
μm; O
(B, O
E, O
G) O
50 O
μm. O
Scale O
bar O
for O
M-R: O
(M, O
N, O
P, O
Q): O
100 O
μm O
; O
(O, O
R): O
50 O
μm.Preliminary O
analyses O
also O
indicated O
that O
expression O
of O
some O
of O
these O
proteins O
was O
downregulated O
in O
differentiated O
cells, O
including O
p130 B-GeneProtein
Cas I-GeneProtein
and O
the O
Traf2- B-GeneProtein
and I-GeneProtein
Nck-interacting I-GeneProtein
kinase I-GeneProtein
(TNIK). O
TNIK B-GeneProtein
is O
known O
to O
be O
involved O
in O
the O
inhibition O
of O
cell O
spreading O
via O
disruption O
of O
F-actin B-CellComponent
[34,35]. O
Immunofluorescence O
was O
used O
to O
examine O
the O
expression O
of O
TNIK B-GeneProtein
and O
p130 B-GeneProtein
Cas I-GeneProtein
during O
early O
differentiation O
of O
hESCs. O
BG01 B-CellLine
cultures O
were O
partially O
differentiated O
by O
growth O
in O
serum O
containing O
media O
for O
3 O
days. O
This O
condition O
generated O
heterogeneous O
populations O
containing O
Oct4+ B-CellType
cells O
with O
characteristic O
hESC O
morphology O
and O
less O
tightly O
packed, O
and O
morphologically O
differentiated O
areas, O
lacking O
expression O
of O
Oct4 B-GeneProtein
(Fig O
4M, O
P). O
TNIK B-GeneProtein
was O
expressed O
highly O
in O
undifferentiated O
hESCs, O
and O
in O
the O
undifferentiated O
areas O
at O
day O
3, O
but O
was O
downregulated O
in O
areas O
undergoing O
morphological O
differentiation O
(Fig O
4N, O
Q). O
This O
may O
indicate O
that O
TNIK B-GeneProtein
is O
active O
in O
hESCs B-CellType
and O
degraded O
rapidly O
upon O
differentiation. O
p130 B-GeneProtein
Cas I-GeneProtein
was O
detected O
in O
a O
membrane/peripheral-cytoplasmic O
pattern O
in O
hESCs B-CellType
(Fig O
4O). O
The O
distribution O
of O
p130Cas B-GeneProtein
was O
substantially O
altered O
in O
differentiating O
cells O
with O
a O
flattened O
morphology, O
exhibiting O
a O
general O
cytoplasmic, O
or O
perinuclear B-CellComponent
profile O
(Fig O
4R). O
This O
could O
indicate O
an O
alteration O
in O
the O
function O
of O
p130 B-GeneProtein
Cas I-GeneProtein
as O
pluripotent B-CellType
cells O
differentiate. O
These O
analyses O
suggested O
that O
the O
change O
in O
expression O
or O
distribution O
of O
these O
proteins O
could O
be O
used O
as O
markers O
for O
undifferentiated O
hESCs. O
Comparison O
of O
proteomic O
and O
transcriptional O
profiles O
of O
hESCsWe O
have O
previously O
employed O
the O
Illumina O
Bead O
Array O
system O
for O
the O
large-scale O
profiling O
of O
gene O
expression O
in O
hESCs B-CellType
using O
24,000 O
transcript O
probes O
[11]. O
To O
compare O
proteomic O
and O
transcriptional O
analyses O
of O
hESCs, O
the O
levels O
of O
> O
600 O
proteins O
detected O
using O
large O
scale O
blotting O
were O
correlated O
with O
the O
levels O
of O
transcripts O
detected O
with O
the O
Illumina O
platform O
(Additional O
File O
3). O
In O
general, O
a O
close O
match O
between O
the O
expression O
level O
of O
transcript O
and O
protein O
was O
observed: O
transcripts O
for O
nearly O
all O
the O
detected O
proteins O
were O
also O
identified O
in O
the O
Illumina O
analysis, O
and O
most O
proteins O
expressed O
at O
high O
levels O
also O
exhibited O
high O
mRNA O
levels.We O
reasoned O
that O
a O
focused O
comparison O
of O
specific O
signaling O
pathways O
using O
a O
combination O
of O
proteomic O
and O
transcriptional O
data O
was O
likely O
to O
be O
much O
more O
informative O
than O
a O
global O
interrogation O
of O
hESCs. O
Several O
major O
signal O
pathways O
that O
have O
been O
suggested O
to O
be O
involved O
in O
self-renewal O
were O
examined O
to O
test O
this O
approach. O
These O
included O
the O
FGF, O
TGFβ, O
GSK3β/Wnt/β-catenin O
and O
Jak/Stat O
pathways O
[17,29,36-39], O
as O
well O
as O
the O
more O
recently O
suggested O
MAPK/ERK O
and O
Gap B-CellComponent
junction I-CellComponent
pathways O
[32,40]. O
Correlating O
transcriptional O
and O
proteomic O
data O
provided O
direct O
confirmation O
that O
these O
pathways O
were O
present O
and O
likely O
functional O
in O
hESCs B-CellType
(Table O
2). O
For O
example, O
FGF2 B-GeneProtein
protein O
was O
expressed O
highly O
in O
hESCs B-CellType
and O
expression O
of O
key O
members O
of O
the O
TGFβ, O
Wnt, O
Jak/Stat O
and O
Gap B-CellComponent
junction I-CellComponent
pathways, O
namely O
Stat1, O
SMADs, O
GSK3β, O
β-catenin B-GeneProtein
and O
Connexin B-GeneProtein
43, O
were O
detected O
in O
both O
transcriptional O
and O
proteomic O
databases.Table O
2Signal O
pathways O
that O
may O
be O
active O
in O
hESCsNameProteinmRNATGF O
βStat1++++++PAI-1/SERPINE1+++-Smad2/3++++Jun+++Smad4/DPC4+++Endoglin+-WntCtBP2+++++++PP2A O
Catalytic I-GeneProtein
alpha/PPP2CA+++++++EBP50/SLC9A3R1++++++beta-Catenin/Ctnnb1++++Cyclin O
D3/CCND3++++GSK-3 O
beta++++Jun+++Casein O
Kinase I-GeneProtein
II I-GeneProtein
alpha/CSNK2A1++++Jak-StatStat1++++++Crk++++Stat3/2+++++Stat6+++++PTP1B+++++JAK1++-Glucocorticoid O
R/NR3C1++-Thrombin O
Receptor/PAR1/F2R+++SHPS-1/PTPNS1++++MCM5+++++Smad2/3++++Tyk2++++Jun+++Bcl-x/BCL2L1++++Smad4/DPC4+++Stat5A++GPCRB2 O
Bradykinin B-GeneProtein
Receptor/BDKRB2++++-Neurotensin O
Receptor I-GeneProtein
3/SORT1+++-Endopeptidase O
3.4.24.16/NLN++++IP3R-3++++SHC++++Gap O
JunctionCdk1/Cdc2++++++GRB2++++++MEK1/MAP2K1++++++PKA O
C++-PKA O
RI I-GeneProtein
alpha++-PKC O
alpha++-C-Raf/RAF1++++ZO-1/TJP1+++++Connexin-43/GJA1++++IGFPKC O
iota++++++MEK1/MAP2K1++++++Rsk/RPS6KA1+++++GRB2++++++MEK2/MAP2K2++++++PI3Kinase/PIK3R1+++++pan O
ERK/MAPK1+++++Crk++++eIF-4E+++++ShcC+++-PAI-1/SERPINE1+++-C-Raf++++SHC+++++PKC O
beta/PRKCB1++++NCK++++PKB O
alpha/Akt+++GSK-3 O
beta++++Ercc-1++++Fatty O
Acid I-GeneProtein
Synthase/FASN+++++Jun+++RAFT1/FRAP++++PTP1D/SHP2/PTPN11++++SCAMP1++++Bcl-x/BCL2L1++++p70s6k/RPS6KB1+-PI3-Kinase O
p170/PIK3C2A++PTP1B/PTPN1+++Dok1/p62dok+++PI3-Kinase O
p110 I-GeneProtein
alpha/PIK3CA+-ERBBEphA4/Sek++++-ShcC/SHC3+++-c-erb-B2/ERBB2++++C-Raf/RAF1++++SHC/SHC1+++++GDNFI O
kappa I-GeneProtein
B I-GeneProtein
epsilon/NFKBIE++++++GRB2++++++MEK2++++++NCK+++++C-Raf++++Ras-GAP/RASA1++++SHC+++++GDNFR-alpha/Gfra1++-Jun+++IKK O
beta++++pan-JNK/SAPK1/MAPK10+++NBS1/ARTN++Dok1/p62dok+++Tight O
JunctionPTEN++++++PP2A O
Catalytic I-GeneProtein
alpha+++++++PKC O
iota++++++Sec8/SEC8L1+++++beta-Catenin/CTNNB1++++CDC42+++++AF6/MLLT4+++++PKC O
alpha++-Yes++++Rho/ARHA+++++ZO-1/TJP1+++++CASK+++Symplekin/SYMPK++-Ras/NRAS++++Casein O
Kinase I-GeneProtein
II I-GeneProtein
alpha/CSNK2A1++++VAP33/VAPA+++alpha-Catenin/Ctnna1+++MAPKpan O
ERK++++++MEK1++++++Rsk++++++ERK2++++++MEK2/Map2k2++++++MST3/STK25++++++ERK1+++++CDC42+++++C-Raf++++p38 O
alpha/SAPK2a++-G3BP+++++TFII-I/GTF2IRD1++++MST1/STK4++++MKP2/Dusp4++++Ras++++Phospho-p38MAPK O
(T180/Y182)+++pan-JNK/SAPK1+++Inhibitor2/PPP1R2++++ABP-280++++++14-3-3 O
epsilon/YWHAE++++MAPKAPK-5+++TAO1+*PBK+++MKK3b/Map2k3++Protein O
expression O
level: O
> O
10,000: O
++++; O
5,000–10,000: O
+++; O
1,000–5,000:++; O
100–1,000: O
+ O
mRNA O
gene O
expression O
level: O
> O
5,000:++++; O
1,000–5,000: O
+++; O
100–1,000: O
++; O
30–100: O
+*: O
not O
included O
in O
the O
gene O
expression O
arrayThis O
independent O
confirmation O
of O
known O
networks O
led O
us O
to O
examine O
other O
pathways O
that O
showed O
a O
similar O
correlation O
but O
have O
not O
been O
identified O
as O
key O
regulators O
of O
either O
self-renewal O
or O
differentiation, O
or O
suggest O
unappreciated O
characteristics O
of O
hESCs. O
Four O
signaling O
pathways O
(IGF, O
ERBB2, O
GPCR, O
and O
GDNF) O
and O
the O
tight B-CellComponent
junction I-CellComponent
complex I-CellComponent
were O
highlighted O
by O
this O
analysis O
(Table O
2), O
and O
expression O
of O
key O
proteins O
in O
these O
pathways O
was O
confirmed. O
A O
detailed O
study O
demonstrating O
the O
importance O
of O
the O
IGF B-GeneProtein
and O
ERBB2 B-GeneProtein
pathways O
in O
hESC O
self-renewal O
has O
been O
performed O
and O
enabled O
the O
development O
of O
a O
defined O
medium O
for O
hESC O
maintenance O
(TCS O
and O
AJR, O
submitted). O
Tight B-CellComponent
junctions O
are O
apical B-CellComponent
cell-cell I-CellComponent
junctions I-CellComponent
found O
in O
epithelia B-Anatomy
that O
establish O
a O
barrier B-CellComponent
to O
the O
extracellular B-CellComponent
environment O
and O
a O
border B-CellComponent
for O
apical-basolateral O
polarity. O
While O
hESCs B-CellType
grow O
in O
colonies O
that O
are O
highly O
reminiscent O
of O
epithelia, O
and O
have O
been O
shown O
to O
be O
coupled O
by O
gap B-CellComponent
junctions I-CellComponent
[40], O
the O
formation O
of O
tight B-CellComponent
junction I-CellComponent
complexes I-CellComponent
has O
not O
been O
described. O
hESCs B-CellType
expressed O
the O
ZO1 B-GeneProtein
and O
occludin B-GeneProtein
tight O
junction O
proteins O
along O
cell B-CellComponent
borders I-CellComponent
as O
expected O
in O
polarized B-Anatomy
epithelia. O
The O
distribution O
of O
ZO1 B-GeneProtein
expression O
changed O
dramatically O
as O
hESCs B-CellType
proliferated O
in O
culture. O
When O
tight O
junction O
complexes O
were O
disrupted O
by O
disaggreagation O
to O
single O
cells, O
only O
a O
subset O
of O
cells O
showed O
ZO1 B-GeneProtein
staining O
4 O
days O
after O
plating O
(Fig. O
5). O
Continued O
proliferation O
to O
a O
confluent O
monolayer O
on O
day O
7 O
was O
accompanied O
by O
widespread O
expression O
of O
ZO1, O
suggesting O
the O
formation O
of O
a O
general O
tight O
junction O
barrier. O
These O
cultures O
were O
undifferentiated O
and O
retained O
uniform O
expression O
of O
Oct4 B-GeneProtein
protein O
(not O
shown). O
ERBB2 B-GeneProtein
and O
3 O
are O
members O
of O
the O
epidermal B-Anatomy
growth O
factor O
(EGF)-receptor O
family, O
which O
regulate O
epithelial O
proliferation O
via O
EGF-family O
ligands. O
ERBB2 B-GeneProtein
and O
3 O
transcripts O
are O
expressed O
by O
hESCs B-CellType
[8], O
are O
known O
to O
function O
as O
a O
heterodimer O
[41], O
and O
transmit O
a O
strong O
proliferative O
signal O
for O
hESCs B-CellType
by O
Heregulin B-GeneProtein
1β I-GeneProtein
(an O
EGF-family O
ligand) O
(TCS O
and O
AJR, O
submitted). O
Immunofluorescence O
revealed O
general O
cell O
surface O
expression O
of O
ERBB2 B-GeneProtein
on O
hESCs. O
Conversely, O
ERBB3 B-GeneProtein
was O
highly O
localized O
to O
a O
concentrated O
area, O
and O
observed O
in O
cells O
that O
also O
expressed O
ZO1. O
Epithelial B-CellType
cells I-CellType
are O
known O
to O
localize O
ERBB B-GeneProtein
receptors I-GeneProtein
to O
the O
basolateral B-CellComponent
side O
of O
tight B-CellComponent
junctions, O
which O
serves O
to O
functionally O
separate O
receptors O
from O
ligands O
[42,43]. O
This O
is O
a O
basic O
epithelial B-CellType
wound O
healing O
mechanism, O
whereby O
disruption O
of O
the O
tight O
junction O
barrier O
by O
injury O
immediately O
exposes O
receptors O
to O
extracelluar O
ligands O
[43]. O
These O
staining O
patterns O
are O
also O
suggestive O
of O
basolateral O
sorting O
of O
ERBB3 B-GeneProtein
in O
hESCs. O
The O
pathways O
and O
complexes O
identified O
by O
these O
analyses O
lay O
a O
framework O
for O
future O
functional O
analyses O
of O
signaling O
networks O
in O
hESCs.Figure O
5Tight O
junction I-CellComponent
proteins O
and O
ERBB2/3 B-GeneProtein
expression O
in O
hESCs. O
BG01 B-CellLine
hESCs B-CellType
were O
disaggregated O
to O
single O
cells O
using O
accutase B-GeneProtein
[52] O
and O
cultured O
in O
defined O
conditions. O
(A) O
ZO1 B-GeneProtein
expression O
four O
and O
(B) O
seven O
days O
after O
plating, O
indicating O
progressive O
tight O
junction O
formation. O
(C) O
Occludin B-GeneProtein
expression O
5 O
days O
after O
plating. O
(D) O
General O
cell B-CellComponent
surface I-CellComponent
expression O
of O
ERBB2, O
in O
the O
same O
field O
of O
view O
as O
(A). O
(E) O
Localized O
expression O
of O
ERBB3, O
in O
the O
same O
field O
of O
view O
as O
(B). O
(F) O
Higher O
magnification O
of O
ERBB3 B-GeneProtein
localization O
in O
ZO1 B-GeneProtein
expressing O
BG01 B-CellLine
cells, O
5 O
days O
after O
plating. O
Nuclei B-CellComponent
were O
stained O
with O
DAPI. O
Attempts O
to O
harness O
the O
potential O
of O
hESCs B-CellType
for O
models O
of O
human B-Species
embryogenesis O
and O
cell O
therapy O
applications O
will O
be O
greatly O
enhanced O
by O
a O
detailed O
understanding O
of O
their O
molecular O
characteristics. O
This O
includes O
definition O
of O
the O
transcripts, O
splice O
variants, O
and O
protein O
isoforms O
expressed O
by O
these O
cells. O
Post-translational O
modifications O
such O
as O
phosphorylation O
and O
glycosylation, O
and O
the O
receptors O
and O
signaling O
pathways O
active O
in O
the O
pluripotent O
state, O
or O
during O
early O
differentiation, O
also O
need O
to O
be O
determined. O
This O
should O
also O
be O
complemented O
by O
an O
understanding O
of O
epigenetic O
characteristics O
of O
pluripotency, O
including O
methylation, O
imprinting O
and O
chromatin B-CellComponent
conformation. O
Such O
a O
comprehensive O
definition O
of O
the O
molecular O
state O
of O
hESCs B-CellType
will O
enable O
more O
accurate O
prediction O
and O
testing O
of O
the O
conditions O
used O
for O
growth O
and O
differentiation O
of O
hESCs, O
by O
precise O
genetic O
modification O
or O
application O
of O
specific O
growth O
factor O
cocktails O
and O
reagents. O
For O
example, O
a O
scalable, O
fully O
defined O
and O
GMP-certified O
culture O
system O
will O
need O
to O
be O
developed O
for O
the O
eventual O
development O
of O
hESC-based O
cellular O
therapies. O
Progress O
has O
been O
made O
in O
defining O
growth O
factor O
conditions O
that O
support O
self-renewal O
[44-46], O
and O
hESC O
lines O
have O
been O
isolated O
in O
the O
absence O
of O
mouse B-Species
embryonic I-CellType
fibroblasts I-CellType
and O
in O
animal O
protein O
free O
culture O
conditions O
[47,48]. O
A O
more O
refined O
understanding O
of O
the O
biology O
of O
hESCs O
has O
contributed O
the O
development O
of O
a O
defined O
medium O
utilizing O
ligands O
for O
IGF1R B-GeneProtein
and O
ERBB2/3 B-GeneProtein
receptors O
to O
promote O
in O
self-renewal O
(TCS O
and O
AJR, O
submitted). O
We O
and O
others O
have O
performed O
transcriptional O
analyses O
of O
hESCs, O
using O
cDNA B-CellComponent
and O
oligonucleotide O
microarrays, O
SAGE, O
MPSS O
and O
EST O
enumeration. O
These O
techniques O
have O
enabled O
the O
collation O
and O
comparison O
of O
transcriptional O
profiles O
from O
multiple O
hESC O
lines O
and O
their O
differentiated O
derivatives O
and O
have O
highlighted O
an O
expanded O
set O
of O
hESC O
specific O
markers O
and O
signaling O
pathways O
that O
may O
regulate O
self-renewal O
or O
differentiation. O
Using O
pathway O
analysis O
we O
were O
also O
able O
to O
identify O
key O
pathways O
that O
are O
active O
in O
ESCs B-CellType
(reviewed O
in O
[16]). O
While O
these O
efforts O
have O
been O
highly O
valuable O
in O
defining O
the O
transcriptional O
profile O
of O
undifferentiated O
hESCs, O
they O
are O
only O
predictive O
of O
translation O
and O
do O
not O
shed O
light O
on O
post-translational O
events O
in O
this O
unique O
cell O
type. O
These O
processes O
may O
also O
be O
highly O
regulated, O
which O
could O
contribute O
significantly O
to O
the O
overall O
conversion O
of O
genetic O
information O
to O
actual O
protein O
function. O
We O
report O
here O
a O
proteomic O
analysis O
of O
pluripotent O
hESCs B-CellType
by O
using O
two O
large-scale O
western O
blotting O
systems O
and O
highlight O
post-translational O
events O
in O
undifferentiated O
hESCs. O
The O
expression O
of O
545 O
bands O
was O
detected, O
potentially O
representing O
529 O
proteins, O
or O
their O
migratory O
isoforms. O
In O
addition, O
one O
hundred O
and O
forty O
phospho-specific O
antibodies O
were O
used O
to O
identify O
85 O
different O
phosphorylated O
sites, O
on O
76 O
proteins O
in O
these O
cells. O
The O
detected O
proteins O
were O
annotated O
into O
functional O
classes O
representing O
diverse O
cellular O
processes. O
For O
example, O
multiple O
proteins O
were O
detected O
that O
have O
been O
suggested O
to O
regulate O
the O
pluirpotent O
state O
in O
mouse B-Species
ESCs I-CellType
or O
hESCs. O
Defining O
the O
interplay O
of O
these O
multiple O
signaling O
pathways O
will O
be O
critical O
in O
understanding O
the O
self-renewal O
versus O
differentiation O
decisions O
of O
hESCs. O
Therefore, O
our O
data O
provide O
a O
powerful O
framework O
for O
the O
functional O
analysis O
of O
specific O
proteins, O
protein O
classes, O
or O
molecular O
pathways. O
In O
particular, O
the O
availability O
of O
antibodies O
for O
candidate O
proteins O
is O
a O
major O
benefit O
of O
this O
approach O
compared O
to O
2D-gel O
or O
HPLC-MS/MS O
based O
proteomics. O
Although O
these O
western O
blotting O
approaches O
are O
currently O
more O
limited O
in O
scope O
than O
most O
large-scale O
cDNA B-CellComponent
based O
assays, O
detecting O
up O
to O
1000 O
proteins O
compared O
to O
tens O
of O
thousands O
of O
transcripts, O
they O
have O
the O
potential O
to O
highlight O
translational O
events O
and O
post-translational O
modifications. O
By O
comparison, O
SAGE O
and O
MPSS O
are O
limited O
to O
detecting O
short O
sequence O
"tags" O
adjacent O
to O
the O
poly-A O
tail O
of O
transcripts, O
and O
may O
not O
distinguish O
splice O
variants O
with O
the O
same O
3' O
exon. O
We O
detected O
42 O
proteins O
with O
multiple O
closely O
migrating O
bands O
(Fig. O
3C), O
suggestive O
of O
closely O
related O
isoforms O
or O
post-translational O
modifications O
such O
as O
phosphorylation. O
These O
focused O
proteomic O
approaches O
are O
therefore O
likely O
to O
be O
highly O
complimentary O
to O
transcriptional O
analyses O
in O
investigating O
the O
functional O
expression O
of O
the O
genome O
in O
hESCs B-CellType
and O
during O
cellular O
differentiation. O
One O
potential O
issue O
with O
this O
approach O
is O
that O
multiple O
antibodies O
are O
included O
in O
each O
lane, O
which O
could O
possibly O
lead O
to O
misidentification O
of O
bands. O
To O
demonstrate O
that O
identified O
proteins O
were O
expressed O
in O
hESCs, O
the O
same O
antibodies O
used O
in O
the O
PowerBlot O
assay O
were O
used O
to O
confirm O
expression O
of O
10 O
representative O
proteins O
by O
immunofluorescence O
(Fig. O
4). O
Furthermore, O
13 O
proteins O
were O
detected O
with O
multiple O
different O
antibodies, O
and O
35 O
proteins O
(Table O
1) O
were O
detected O
in O
both O
the O
PowerBlot O
and O
Kinexus O
assays. O
This O
provided O
internal, O
or O
independent, O
confirmation O
of O
expression O
of O
these O
proteins. O
Other O
studies O
have O
also O
demonstrated O
the O
expression O
of O
several O
of O
the O
proteins O
we O
detected O
in O
hESCs. O
These O
include O
Oct4, O
a O
key O
marker O
of O
the O
pluripotent O
state, O
Connexin B-GeneProtein
43 I-GeneProtein
and O
GSK3β, O
confirming O
the O
reliability O
of O
large-scale O
western O
blotting. O
Finally, O
several O
proteins O
detected O
by O
our O
assays O
were O
also O
detected O
in O
hESCs B-CellType
by O
MS O
approaches O
including O
Karyopherin B-GeneProtein
α I-GeneProtein
[19]. O
Additionally, O
the O
PowerBlot O
assay O
was O
performed O
in O
duplicate, O
and O
was O
shown O
to O
be O
highly O
reproducible. O
This O
suggested O
that O
this O
approach O
should O
be O
informative O
when O
comparing O
hESCs O
to O
their O
differentiated O
derivatives. O
Two O
candidate O
proteins, O
TNIK B-GeneProtein
and O
p130 B-GeneProtein
Cas, O
were O
downregulated, O
or O
exhibited O
altered O
localization O
upon O
spontaneous O
differentiation O
of O
hESCs, O
respectively. O
This O
indicated O
that O
they O
were O
novel O
markers O
of O
undifferentiated O
cells O
and O
molecules O
that O
could O
be O
functionally O
involved O
with O
self-renewal. O
It O
is O
impossible O
in O
an O
initial O
manuscript O
to O
analyze O
and O
rigorously O
test O
all O
the O
predictions O
that O
could O
be O
made O
from O
comparing O
transcriptional O
and O
proteomic O
data O
sets. O
However, O
we O
did O
examine O
key O
features O
to O
illustrate O
the O
power O
of O
this O
methodology. O
Potential O
new O
markers O
for O
hESCs B-CellType
were O
identified, O
the O
expression O
and O
activation O
of O
proteins O
in O
key O
self-renewal O
pathways O
were O
confirmed, O
and O
a O
diverse O
range O
of O
proteins O
were O
detected O
and O
expression O
correlated O
with O
transcriptional O
analyses. O
In O
addition, O
we O
highlighted O
several O
candidate O
signaling O
pathways O
that O
may O
be O
relevant O
to O
self-renewal. O
Examination O
of O
tight B-GeneProtein
junction I-GeneProtein
protein I-GeneProtein
expression O
indicated O
that O
undifferentiated O
hESCs B-CellType
could O
form O
polarized B-Anatomy
epithelia, O
which O
has O
also O
been O
recently O
suggested O
by O
ultrastructural O
analyses O
[49]. O
Discrete O
localization O
of O
ERBB3 B-GeneProtein
may O
also O
suggest O
basolateral O
separation O
of O
this O
receptor O
from O
soluble O
ligand. O
These O
analyses O
highlight O
that O
predictions O
from O
a O
combination O
of O
transcriptional O
and O
proteomic O
approaches O
will O
serve O
to O
focus O
the O
investigation O
of O
hESCs B-CellType
in O
the O
future. O
In O
summary, O
we O
generated O
a O
focused O
proteome O
of O
hESCs B-CellType
using O
large-scale O
western O
blotting O
and O
sorted O
the O
detected O
proteins O
according O
to O
function O
and O
signaling O
pathways. O
This O
characterization O
provides O
important O
basic O
information O
on O
expressed O
proteins, O
their O
isoforms O
and O
post-translational O
modifications, O
and O
tools O
for O
the O
continued O
investigation O
of O
the O
underlying O
molecular O
characteristics O
of O
hESCs. O
Importantly, O
we O
provide O
a O
list O
of O
tools, O
in O
the O
form O
of O
commercially O
available O
antibodies, O
which O
can O
be O
used O
to O
interrogate O
the O
function O
of O
these O
molecules O
in O
self-renewal O
or O
differentiation. O
Culture O
of O
human B-Species
embryonic I-CellType
stem I-CellType
cellsFor O
the O
PowerBlot O
analysis, O
enzymatically O
passaged O
BG01 B-CellLine
hESCs I-CellType
were O
grown O
as O
described O
previously O
[23]. O
These O
conditions O
were O
necessary O
to O
scale O
up O
the O
culture O
to O
generate O
the O
milligram O
amounts O
of O
protein O
lysate O
required O
for O
this O
analysis. O
These O
conditions O
maintain O
cell O
populations O
that O
express O
the O
appropriate O
markers O
of O
pluripotency O
and O
can O
differentiate O
to O
representatives O
of O
all O
three O
germ O
layers, O
but O
may O
lead O
to O
eventual O
accumulation O
of O
trisomies O
for O
chromosomes O
12, O
17 O
or O
X O
[26]. O
For O
the O
Kinexus O
assays, O
BG03 B-CellLine
hESCs I-CellType
were O
maintained O
in O
MEF-conditioned O
medium O
(MEF-CM) O
without O
the O
accumulation O
of O
karyotypic B-CellComponent
abnormalities O
as O
described O
previously O
[14,26].hESCs O
were O
also O
maintained O
in O
a O
defined O
medium O
as O
indicated. O
These O
conditions O
are O
described O
in O
detail O
elsewhere O
(TCS, O
AJR, O
submitted). O
Briefly, O
the O
media O
consisted O
of O
DMEM/F12 O
(Invitrogen), O
2% O
fatty O
acid-free O
Cohn's O
fraction O
V O
BSA O
(Serologicals), O
1× O
nonessential O
amino O
acids, O
50 O
U/ml O
penicillin/streptomycin, O
50 O
μg/ml O
ascorbic O
acid, O
10 O
μg/ml O
bovine O
transferrin, O
0.1 O
mM O
β-mecaptoethanol O
(all O
from O
Invitrogen), O
1× O
Trace O
Elements O
A, O
B O
& O
C O
(Mediatech), O
10 O
ng/ml O
hergulin1β B-GeneProtein
(Peprotech), O
10 O
ng/ml O
activinA B-GeneProtein
(R&D O
Systems), O
200 O
ng/ml O
LR3-IGF1 B-GeneProtein
(JRH O
Biosciences), O
and O
8 O
ng/ml O
FGF2 B-GeneProtein
(R&D O
Systems). O
Cultures O
were O
passaged O
using O
Collagenase B-GeneProtein
IV I-GeneProtein
and O
plated O
on O
growth O
factor O
depleted O
Matrigel O
(BD O
Biosciences) O
diluted O
1:200. O
These O
cultures O
were O
karyotypically B-CellComponent
normal.To O
partially O
differentiate O
hESC O
cultures O
for O
immunostaining O
analysis, O
karyotypically B-CellComponent
normal O
BG01 B-CellLine
cells I-CellType
were O
plated O
on O
matrigel O
and O
grown O
for O
three O
days O
in O
DMEM/F12 O
containing O
10% O
fetal O
calf O
serum O
(HyClone), O
1× O
nonessential O
amino O
acids, O
20 O
mM O
L-glutamine, O
50 O
U/ml O
penicillin/streptomycin, O
and O
0.1 O
mM O
β-mecaptoethanol. O
PowerBlot O
assaysBG01 O
hESC O
lysate O
was O
prepared O
in O
10 O
mM O
Tris-HCl O
pH O
7.4, O
1 O
mM O
sodium O
orthovanadate O
and O
1% O
SDS, O
and O
the O
PowerBlot O
assays O
were O
performed O
by O
BD O
Biosciences O
(BD O
Biosciences). O
Briefly, O
200 O
μg O
of O
protein O
lysate O
was O
loaded O
in O
a O
single, O
gel-wide O
well, O
on O
a O
SDS-4–15% O
gradient O
polyacrylamide O
gel. O
The O
full O
PowerBlot O
screen O
consisted O
of O
five O
gels, O
which O
were O
blotted O
and O
probed O
with O
934 O
antibodies, O
and O
was O
performed O
in O
duplicate O
with O
the O
same O
cell O
lysate. O
The O
gel O
dimensions O
were O
130 O
× O
100 O
× O
0.5 O
mm, O
and O
proteins O
were O
separated O
at O
150 O
volts O
for O
1.5 O
hours, O
and O
transferred O
to O
an O
Immobilon-P O
membrane O
(Millipore). O
The O
membranes O
were O
blocked O
and O
clamped O
in O
a O
manifold O
that O
created O
40 O
lanes O
across O
each O
membrane. O
A O
mix O
of O
1 O
to O
8 O
mouse O
monoclonal O
primary O
antibodies O
was O
added O
to O
each O
lane, O
in O
dilutions O
and O
combinations O
that O
had O
been O
predetermined O
to O
enable O
accurate O
identification O
of O
well-separated O
proteins. O
The O
predicted O
sizes O
of O
detectable O
proteins O
in O
the O
blots O
ranged O
from O
10–540 O
kD, O
and O
the O
dilutions O
of O
the O
primary O
antibodies O
ranged O
from O
1:250 O
to O
1:15,000.The O
blots O
were O
removed O
from O
the O
manifolds, O
washed O
and O
incubated O
with O
goat O
anti-mouse O
secondary O
antibody O
conjugated O
to O
the O
Alexa680 O
fluorophore O
(Molecular O
Probes). O
The O
membranes O
were O
scanned O
using O
the O
Odyssey O
Imaging O
System O
(LI-COR). O
Molecular O
weight O
standards O
were O
generated O
by O
adding O
a O
cocktail O
of O
antibodies O
to O
P190 B-GeneProtein
(190 O
kD), O
Adaptin B-GeneProtein
beta I-GeneProtein
(106 O
kD), O
STAT-3 B-GeneProtein
(92 O
kD), O
PTP1D B-GeneProtein
(72 O
kD), O
Mek-2 B-GeneProtein
(46 O
kD), O
RACK-1 B-GeneProtein
(36 O
kD), O
GRB-2 B-GeneProtein
(24 O
kD) O
and O
Rap2 B-GeneProtein
(21 O
kD) O
to O
lane O
40 O
of O
gels O
A-D. O
Molecular O
standards O
for O
gel O
E O
were O
generated O
by O
adding O
a O
cocktail O
of O
antibodies O
to O
Exportin-1/CRM1 B-GeneProtein
(112 O
kD), O
MCM B-GeneProtein
(83 O
kD), O
Nucleoporin B-GeneProtein
p62 I-GeneProtein
(62 O
kD), O
α-tubulin B-GeneProtein
(55 O
kD), O
Actin B-GeneProtein
(42 O
kD), O
KNP-1/HES1 B-GeneProtein
(28 O
kD) O
and O
NTF2 B-GeneProtein
(15 O
kD) O
to O
lane O
16, O
and O
antibodies O
to O
p190 B-GeneProtein
(190 O
kD), O
Hip1R B-GeneProtein
(120 O
kD), O
Transportin B-GeneProtein
(101 O
kD), O
Calreticulin B-GeneProtein
(60 O
kD), O
Arp3 B-GeneProtein
(50 O
kD), O
eIF-6 B-GeneProtein
(27 O
kD) O
and O
Rap2 B-GeneProtein
(21 O
kD) O
to O
lane O
17.Bands O
were O
detected O
and O
raw O
signal O
intensity O
captured O
automatically O
using O
the O
PDQuest O
software O
(Bio-Rad). O
To O
normalize O
the O
signal O
intensities, O
the O
total O
raw O
quantity O
of O
each O
band O
was O
divided O
by O
the O
average O
intensity O
value O
of O
the O
molecular O
standards O
in O
that O
image O
and O
the O
normalized O
values O
for O
the O
duplicate O
samples O
were O
averaged O
and O
expressed O
as O
normalized O
intensity O
units O
(i.u.). O
These O
values O
represent O
the O
relative O
signal O
intensity O
observed O
for O
each O
identified O
protein O
band, O
rather O
than O
relative O
expression O
levels O
of O
different O
proteins, O
due O
to O
differences O
in O
the O
efficiencies O
of O
antibody O
binding O
and O
dilution O
of O
the O
primary O
antibodies O
used. O
Proteins O
were O
identified O
based O
on O
the O
similarity O
of O
expected O
and O
observed O
band O
migration O
profiles O
and O
bands O
that O
could O
not O
be O
identified O
were O
excluded O
from O
the O
analysis. O
All O
identified O
proteins O
were O
verified O
by O
visual O
inspection, O
and O
proteins O
exhibiting O
a O
low O
signal O
intensity, O
with O
an O
averaged O
signal O
of O
< O
1000 O
i.u., O
were O
verified O
by O
visual O
inspection O
using O
contrast O
enhancement O
in O
Adobe O
Photoshop. O
Bands O
with O
> O
800 O
i.u. O
could O
typically O
be O
observed O
without O
additional O
image O
enhancement. O
Microsoft O
Excel O
files O
were O
generated O
that O
contained O
information O
on: O
gel O
number, O
lane O
number, O
antibody O
catalogue O
number O
(BD O
Biosciences), O
protein O
name, O
expected O
size, O
observed O
size, O
repeat O
1 O
i.u. O
value, O
repeat O
2 O
i.u. O
value, O
averaged O
i.u. O
value, O
antibody O
dilution, O
outline O
of O
protein O
function, O
Entrez O
gene O
and O
SwissProt O
identification O
numbers. O
These O
tables O
were O
used O
to O
list O
expressed O
proteins O
(Additional O
File O
1). O
Kinexus O
assaysPreparation O
of O
the O
BG03 B-CellLine
cell O
lysate O
and O
western O
blotting O
was O
performed O
according O
to O
published O
protocols O
[50]. O
Briefly, O
cell O
lysate O
was O
prepared O
in O
20 O
mM O
MOPS O
pH O
7.0, O
2 O
mM O
EGTA, O
5 O
mM O
EDTA, O
30 O
mM O
sodium O
fluoride, O
40 O
mM O
β-glycerolphosphate O
pH O
7.2, O
20 O
mM O
sodium O
pyrophosphate, O
1 O
mM O
sodium O
orthovanadate, O
1 O
mM O
PMSF, O
3 O
mM O
benzamidine, O
5 O
μM O
pepstatin, O
10 O
μM O
leupeptin, O
0.5% O
nonidet O
P-40, O
with O
the O
final O
pH O
adjusted O
to O
7.2. O
The O
Kinexus O
assays O
for O
protein B-GeneProtein
kinases I-GeneProtein
(KPKS-1.2A O
and O
B O
[76 O
antibodies]), O
phosphatases B-GeneProtein
(KPPS-1.2 O
[27 O
antibodies]) O
and O
phosporylated O
sites O
in O
cell O
signaling O
molecules O
(KPSS-3.1 O
[37 O
antibodies]) O
were O
performed O
by O
Kinexus. O
The O
Bio-Rad O
Mini-PROTEAN O
3 O
electrophoresis O
system O
was O
used O
to O
separate O
proteins O
by O
SDS-PAGE. O
For O
each O
assay, O
250 O
μg O
of O
cell O
lysate O
was O
loaded O
in O
a O
single O
well O
spanning O
the O
width O
of O
the O
stacking O
gel, O
then O
separated O
through O
a O
12.5% O
SDS-Polyacrylamide O
gel O
and O
transferred O
to O
a O
PVDF O
membrane. O
A O
20-lane O
manifold O
was O
placed O
over O
the O
membrane O
and O
a O
different O
mixture O
of O
up O
to O
3 O
primary O
antibodies O
was O
added O
to O
each O
well. O
The O
combinations O
of O
primary O
antibodies O
had O
been O
predetermined O
to O
detect O
well-separated O
proteins, O
avoiding O
crossreaction O
to O
different O
proteins O
that O
co-migrate. O
The O
primary O
antibodies O
were O
rabbit B-Species
and O
goat B-Species
polyclonal, O
and O
mouse B-Species
monoclonal O
antibodies, O
diluted O
1:1000. O
After O
incubation O
with O
the O
primary O
antibodies, O
the O
membranes O
were O
removed O
from O
the O
manifolds, O
washed O
and O
incubated O
with O
a O
mix O
of O
the O
appropriate O
secondary O
antibodies. O
The O
secondary O
antibodies O
were O
donkey B-Species
anti-rabbit O
(at O
1:5000), O
sheep B-Species
anti-mouse O
(at O
1:10,000) O
and O
bovine B-Species
anti-goat O
(at O
1:10,000), O
all O
conjugated O
with O
horse B-Species
radish I-GeneProtein
peroxidase. O
The O
membranes O
were O
washed O
and O
immunoreactive O
bands O
detected O
by O
enhanced O
chemiluminescence O
(Amersham-Pharmacia) O
using O
a O
FluorS O
Max O
Multi-imager O
(Bio-Rad). O
Prestained O
size O
markers O
(201.5, O
156.8, O
106, O
79.7, O
48.4, O
37.8, O
23.3, O
and O
18.2 O
kD) O
and O
predetermined O
human-specific O
protein O
migration O
profiles O
were O
used O
to O
accurately O
identify O
proteins O
using O
the O
Kinexus O
immuno-reactivity O
identification O
system O
(IRIS) O
software. O
Detected O
proteins O
were O
verified O
by O
visual O
inspection. O
ImmunocytochemistryImmunocytochemistry O
and O
staining O
procedures O
were O
as O
described O
previously O
[51]. O
Briefly, O
cells O
were O
fixed O
with O
4% O
paraformaldehyde O
for O
half O
an O
hour, O
blocked O
in O
blocking O
buffer O
(5% O
goat B-Species
serum, O
1% O
BSA, O
0.1% O
Triton O
X-100) O
for O
1 O
hour O
followed O
by O
incubation O
with O
the O
primary O
antibody O
at O
4°C O
overnight. O
Appropriately O
coupled O
secondary O
antibodies O
(Molecular O
Probes) O
were O
used O
for O
single O
and O
double O
labeling. O
All O
secondary O
antibodies O
were O
tested O
for O
cross O
reactivity O
and O
non-specific O
immunoreactivity. O
The O
following O
antibodies O
were O
used: O
ABP-280 B-GeneProtein
(1:250, O
BD O
Biosciences O
610798), O
CtBP1 B-GeneProtein
(1:1000, O
BD O
Biosciences O
612042), O
CtBP2 B-GeneProtein
(1:1000, O
BD O
Biosciences O
612044), O
GS-28 B-GeneProtein
(1:2000, O
BD O
Biosciences O
611184), O
HDJ-2 B-GeneProtein
(1:100, O
BD O
Biosciences O
611872), O
L-Caldesmon B-GeneProtein
(1:2000, O
BD O
Biosciences O
610660), O
Rabaptin-5 B-GeneProtein
(1:500, O
BD O
Biosciences O
611080), O
phospho-p130 B-GeneProtein
Cas I-GeneProtein
(Tyr165) O
(1:50, O
Cell O
Signaling O
Technology O
4015), O
phospho-Ras-GAP B-GeneProtein
(pY460) O
(1:250, O
BD O
Biosciences O
612736), O
Ras-GAP B-GeneProtein
(1:250, O
BD O
Biosciences O
610043), O
Shc-C B-GeneProtein
(1:1000, O
BD O
Biosciences O
610642), O
Oct-4 B-GeneProtein
(Santa O
Cruz O
biotechnology, O
1:200 O
SC-8628), O
TNIK B-GeneProtein
(1:100, O
BD O
Biosciences, O
612250), O
p130 B-GeneProtein
Cas I-GeneProtein
(1:100, O
BD O
Biosciences, O
610272), O
ERBB2 B-GeneProtein
(1:100, O
Lab O
Vision, O
9G6.10), O
ERBB3 B-GeneProtein
(1:100, O
R&D O
Systems, O
MAB348), O
ZO1 B-GeneProtein
(1:100, O
Invitrogen, O
61–7300), O
or O
Occludin B-GeneProtein
(1:100, O
Invitrogen, O
71–1500). O
Hoechst O
(Invitrogen) O
or O
DAPI O
(Sigma) O
were O
used O
to O
identify O
nuclei, O
and O
Triton O
X-100 O
was O
omitted O
when O
staining O
for O
extracellular O
antigens O
(ZO1, O
occludin, O
ERBB2/3). O
Images O
were O
captured O
on O
an O
Olympus O
or O
Nikon O
fluorescence O
microscope. O
lllumina O
data O
and O
comparison O
to O
proteomic O
databaseExpression O
levels O
of O
proteins O
detected O
by O
the O
PowerBlot O
assay O
were O
compared O
to O
our O
previous O
published O
database O
of O
multiple O
hESC O
lines O
examined O
using O
the O
Illumina O
bead O
array O
platform O
(Liu O
et O
al., O
2006). O
Averaged O
transcript O
expression O
signals O
from O
the O
BG01, O
BG02 B-CellLine
and O
BG03 B-CellLine
cell O
lines O
were O
converted O
to O
a O
+/- O
format, O
based O
on O
the O
following O
criteria: O
A O
mean O
transcript O
detection O
level O
of O
> O
5,000 O
was O
designated O
as O
++++; O
1,000–5,000 O
as O
+++; O
100–1,000 O
as O
++; O
30–100 O
as O
+; O
and O
signals O
< O
30 O
was O
represented O
as O
-. O
In O
parallel, O
the O
protein O
expression O
levels O
were O
converted O
to O
a O
+/- O
format O
based O
on O
these O
criteria: O
i.u. O
> O
10,000 O
as O
++++; O
5,000–10,000 O
as O
+++; O
1,000–5,000 O
as O
++; O
100–1,000 O
as O
+. O
In O
addition, O
genes O
were O
categorized O
into O
the O
same O
functional/signaling O
pathways O
as O
per O
the O
western O
blot O
database. O