Expression of CPPED1 in human trophoblasts is associated
with timing of term birth
Antti M. Haapalainen a, b, * , Minna K. Karjalainen a, b, Ravindra Daddali a, b, Steffen Ohlmeier c,
Julia Anttonen a, b, Tomi A. M€a€att€a a, b, Annamari Salminen a, b, Mari Mahlman a, b,
Ulrich Bergmann c, Kaarin M€akikallio a, d, e, Marja Ojaniemi a, b, Mikko Hallman a, b, #,
Mika R€amet a, b, f, #
a PEDEGO Research Unit and Medical Research Center Oulu, University of Oulu, Oulu, Finland
b Department of Children and Adolescents, Oulu University Hospital, Oulu, Finland
c Proteomics Core Facility, Biocenter Oulu, Faculty of Biochemistry and Molecular Medicine, University of Oulu, Oulu, Finland
d Department of Obstetrics and Gynecology, Oulu University Hospital, Oulu, Finland
e Department of Obstetrics and Gynecology, Turku University Hospital and University of Turku, Turku, Finland
f BioMediTech Institute and Faculty of Medical and Life Sciences, University of Tampere, Tampere, Finland
Received: December 2, 2016; Accepted: August 29, 2017
Abstract
Understanding of timing of human parturition is incomplete. Therefore, we carried out proteomic analyses of full-term placentas from uncompli-
cated pregnancies to identify protein signatures associated with the onset of spontaneous delivery. We found quantitative associations of 10
proteins with spontaneous term birth, evident either in the basal or in the chorionic plates or in both. Additional 18 proteins were associated
according to the location within placenta indicating local variations in protein amounts. Calcineurin-like phosphoesterase domain-containing 1
(CPPED1), a phosphatase previously suggested dephosphorylating AKT1/PKB, was one of the identified proteins. qRT-PCR revealed the mRNA
level of CPPED1 was higher in elective caesarean deliveries than in spontaneous births, while immunohistochemistry showed CPPED1 in
cytotrophoblasts, syncytiotrophoblasts and extravillous trophoblasts. Noteworthy, phosphorylation status of AKT1 did not differ between pla-
centas from elective caesarean and spontaneous deliveries. Additionally, analyses of samples from infants indicated that single-nucleotide poly-
morphisms rs11643593 and rs8048866 of CPPED1 were associated with duration of term pregnancy. Finally, post-transcriptional silencing of
CPPED1 in cultured HTR8/SVneo cells by siRNAs affected gene expression in pathways associated with inflammation and blood vessel develop-
ment. We postulate that functions regulated by CPPED1 in trophoblasts at choriodecidual interphase have a role in the induction of term labour,
but it may be independent of AKT1.
Keywords: CPPED1
 genetics
 placenta
 proteomics
 trophoblast
Introduction
Despite extensive research, our understanding of the timing of human
parturition at the molecular level remains incomplete. Multiple factors
may initiate labour [1], and the signal can originate from the mother,
foetus, uteroplacental unit or a combination of these. Some hormones
are known to influence the maintenance of pregnancy. These include
steroid hormones such as progesterone, oestrogens, and androgens,
as well as the peptide hormones chorionic gonadotropin, chorionic
somatomammotropin, relaxin and oxytocin, which further interact
with the immune system to promote pregnancy [2, 3]. Progesterone
is essential for uterine quiescence; oestrogen, on the other hand,
increases myometrial sensitivity to oxytocin, thereby inducing con-
tractions [2]. Strong evidence from animal studies also suggests that
the foetal hypothalamic–pituitary–adrenal (HPA) axis controls secre-
tion of corticotrophin-releasing hormone (CRH), leading to increased
induction of foetal maturity by corticosteroids and increased secretion
of labour-inducing hormones [4]. In the placenta, CRH expression is
negatively regulated by progesterone and oestrogen and positively
regulated by various neuropeptides [1, 5–7]. Other known factors,
such as uterine stretch, inflammatory cytokines, prostaglandins and
nitric oxide, have known or proposed roles in the initiation of labour.
#These authors made equal contributions.
*Correspondence to: Antti M. HAAPALAINEN
E-mail: antti.haapalainen@oulu.fi
ª 2017 The Authors.
Journal of Cellular and Molecular Medicine published by John Wiley & Sons Ltd and Foundation for Cellular and Molecular Medicine.
This is an open access article under the terms of the Creative Commons Attribution License, which permits use,
distribution and reproduction in any medium, provided the original work is properly cited.
doi: 10.1111/jcmm.13402
J. Cell. Mol. Med. Vol 22, No 2, 2018 pp. 968-981
CRH and foetal fibronectin (FFN) are additional key proteins that
have been used to monitor the timing of parturition in humans [1]. In
maternal plasma, CRH levels increase exponentially during pregnancy
[1]. Because there are large individual variations, the rate of increase
rather than a single CRH measurement should be determined [1, 8]. By
contrast, FFN is present mainly in amniotic fluid and foetal membranes
and in placental tissue. FFN may be released into cervicovaginal secre-
tions through mechanical and infection-mediated damage to the foetal
membranes or placenta prior to birth [9]. Elevated FFN levels in cervical
fluids indicate an increased likelihood of delivery; thus, they are a pre-
dictor of term and preterm birth [9, 10]. Unlike in experimental animals
such as sheep and small rodents, in human pregnancy blood levels of
progesterone increase rather than decrease during labour. It has been
proposed that functional withdrawal of progesterone upon labour in
humans may occur, as expression of a low-affinity variant of proges-
terone receptor increases towards term [11]. In human placenta, tro-
phoblasts produce and secrete progesterone [12].
Some studies have reported crosstalk between progesterone and
oestrogen receptors and RAC-alpha serine/threonine-protein kinase
(AKT1) in various diseases [13]. AKT1 is part of the phosphatidylinos-
itol-3-kinase (PI3K) signalling pathway [14, 15], and in some cancers,
the PI3K/AKT pathway is hyperactivated [13]. Phosphorylation of
Ser473 is required for full activation of AKT1; this is accomplished by
3-phosphoinositide-dependent protein kinase 2 (PDK2) [16–18]. Cal-
cineurin-like phosphoesterase domain-containing 1 (CPPED1) is a
newly characterized phosphatase. It inactivates AKT1 by removing the
critical phosphate from Ser473 of AKT1 [19].
Recent epigenetic, transcriptomic, proteomic and exosome analy-
ses of human placenta and the placental microbiome have revealed
new data, not only about placental development but also about labour.
Placenta-derived exosomes, released to the maternal circulation,
appear to have roles in regulating diverse functions, including placen-
tal development and immunotolerance [20]. Placental transcriptomic
differences have been identified in the decidua, amnion and chorion,
suggesting various functionally distinct subregions [21, 22]. More-
over, epigenetic variation in the placenta may be associated with
pregnancy outcome [23].
Several proteomic studies have reported biomarkers of term and
preterm births [10, 24, 25]. Over a hundred biomarkers of spontaneous
preterm birth have been identified for various types of biochemical
pathways [26]. More than half of these biomarkers have been classified
as being involved in immune function and inflammation. Proteomic
analyses of placental blood plasma [27] and cervicovaginal fluids [28]
from normal term deliveries have identified proteins linked to immune
and defence responses and inflammation and oxidative stress, respec-
tively. Proteomic analyses of placenta have also focused on foetal
growth restriction [29], early pregnancy loss [30–32], pre-eclampsia
[32, 33], down syndrome [34], differences between first- and third-tri-
mester human placentas [35] and low molecular weight components
in the chorionic and basal plates of placenta [36].
The majority of placental studies have focused on different patho-
logical conditions, while only few proteomic studies have investigated
changes associated with normal term birth. The aim of this study was
to identify placental proteins associated with gestational age and
labour in uncomplicated term pregnancies. To do so, we analysed the
proteome of the chorionic and basal plates of human placenta to iden-
tify proteins involved in the normal labour process. We found that the
extraembryonic protein CPPED1 is not only a promising marker for
spontaneous term birth but that a polymorphism of the CPPED1 gene
is associated with the length of normal uncomplicated pregnancy. We
also provide evidence that in human term placenta, CPPED1 levels
are not associated with the phosphorylation status of AKT1.
Materials and methods
Placenta samples
The study was approved by the ethics committee of Oulu University
Hospital, and all mothers provided written informed consent. Placenta
tissues were collected at Oulu University Hospital in 2012–2014. Sam-
ples were harvested as described previously [37], including tissue biop-
sies from both basal and chorionic plates. All placental specimens were
from uncomplicated term pregnancies. Of the cases (n = 31), 18 were
spontaneous vaginal deliveries and 13 infants were delivered electively
by Caesarean section without signs or symptoms of labour. Of the
spontaneous term deliveries, the gestational age (GA) varied from
39 weeks + 6 days to 41 weeks + 3 days (average 40 week + 1 day
GA). In the elective caesarean term birth group, the placentas were
from pregnancies with GAs of 38 weeks + 3 days to 42 weeks + 0 day
(average 39 weeks + 2 days GA). The length of pregnancy was based
on foetal ultrasound examination before 16 weeks of pregnancy.
Two-dimensional minimal difference gel
electrophoresis and mass spectrometry
For the proteomic study, we included placenta samples collected after
either elective caesarean or spontaneous term birth from the chorionic
and basal plates (n = 6 per group). For protein spots to be considered
significant, we used a  1.5-fold cut-off ratio. A P value of ≤0.05 accord-
ing to Student’s t-test was considered statistically significant. Theoretical
spot positions in the 2D gel according to known protein sequences were
calculated with the Compute pI/Mw tool (http://ca.expasy.org/tools/
pi_tool.html). Associations between proteomic changes and GO biologi-
cal processes and molecular functions were identified with the Database
for Annotation, Visualization and Integrated Discovery (DAVID), v 6.7
[38, 39]. In mass spectrometry, to identify proteins, we ran additional 2D
gels with larger amounts of unlabelled protein (400–800 lg) combined
with Cyanine dye (Cy) 2-labelled internal standard. Further details regard-
ing 2D-gel and MS analyses are presented in the Appendix S1.
Study population for the genetic study of
gestational age
The study population was collected prospectively at Oulu University Hospi-
tal in 2004–2007 and 2014. Altogether, we included 342 infants born spon-
taneously at term (GA from 38 weeks + 0 days to 41 weeks + 6 days; GA
mean  standard deviation [S.D.] 40.1  0.9 weeks). All study subjects
were of Finnish origin. Umbilical cord blood or umbilical cord tissue was
ª 2017 The Authors.
Journal of Cellular and Molecular Medicine published by John Wiley & Sons Ltd and Foundation for Cellular and Molecular Medicine.
969
J. Cell. Mol. Med. Vol 22, No 2, 2018
collected for DNA extraction. Specific inclusion criteria are presented in the
Appendix S1.
Genetic study and genotyping
Genes encoding proteins that we identified as up- or down-regulated in
the placental proteome from spontaneous term deliveries were analysed
for their association with GA. Altogether, we investigated 77 SNPs: ACTB,
three SNPs; A2M, nine SNPs; B2M, one SNP; CPPED1, 16 SNPs; CYB5A,
11 SNPs; HBG2, five SNPs; KRT8, 12 SNPs; KRT19, five SNPs;
PRDX2, three SNPs; and SERPINB2, 12 SNPs (Table S4). Associations
between these SNPs and GA were assessed by the Wald test with PLINK,
v. 1.07 [40]. Under strict Bonferroni correction, the significance level was
P < 6.5 9 104.
We focused in depth on genotypes of the two GA-associated CPPED1
SNPs rs11643593 and rs8048866 in the investigated infants. Further
details are presented in the Appendix S1.
Western blotting
To quantitate the amount of AKT1 phosphorylated at Ser473 compared to
the total amount of AKT1, we used rabbit anti-human AKT1 pS473 antibody
(2967S, 1:1000 dilution; Cell Signaling Technology, Leiden, The Nether-
lands) and mouse anti-human AKT1 antibody (9018S, 1:1000 dilution; Cell
Signaling Technology), respectively. Secondary antibodies were goat anti-
rabbit IgG Dylight 680 conjugate (611-144-002-0.5, 1:10000 dilution; Rock-
land, Limerick, PA, USA) and goat antimouse IgG Dylight 800 conjugate
(610-145-002-0.5, 1:10000 dilution; Rockland). Detection was done with
the Odyssey Infrared Imaging System (LI-COR Biosciences, Lincoln, NE,
USA). We used 87 ng of human recombinant AKT1 (009-001-P21; Rock-
land) as a positive control.
To quantitate the amount of FOXO1 phosphorylated at Ser256 compared
to the total amount of FOXO1, we used rabbit anti-human FoxO1 pSer256
antibody (NB100-81927, 1:1000 dilution; Novus Biologicals, Abingdon,
Oxon, UK) and mouse anti-human FOXO1 antibody (F6928, 1:1000 dilution;
Sigma-Aldrich, St. Louis, MO, USA), respectively. The secondary antibody
and detection method were the same as for AKT1. Further details are pre-
sented in the Appendix S1.
Immunohistochemistry
For CPPED1 detection, samples were incubated in rabbit anti-human
CPPED1 antibody (HPA040938, 1:250 dilution; Sigma-Aldrich) for 1 hr,
followed by detection of the bound antibodies with the two-step Envi-
sion kit (K5007; DAKO). For FOXO1 and FOXO3 detection, samples were
incubated in rabbit anti-human phospho-FOXO1 pSer256 antibody
(SAB4300094, 1:4000 dilution; Sigma-Aldrich) and rabbit anti-human
phospho-FOXO3 pSer253 antibody (PA5-37578, 1:10000 dilution;
Thermo Fisher Scientific, Waltham, MA, USA), respectively. For negative
controls, non-immune rabbit IgG was used to confirm the specificity of
the primary antibody. Further details are presented in the Appendix S1.
Quantitative PCR
We included samples from the basal and chorionic plates of placentas
from 13 elective caesarean and 18 spontaneous term births for qRT-
PCR analysis. RNA isolation, cDNA synthesis and quantitative PCR anal-
yses were done as described earlier [37]. The High Pure RNA Tissue Kit
(Roche, Branchburg, NJ, USA) and RNeasy Micro Kit (Qiagen, Hilden,
Germany) were used for RNA isolation. Relative quantifications of
CPPED1 mRNA levels from chorionic and basal plates were done with
the LightCycler96 instrument (Roche). The experiments were designed
as intron spanning assays with cytochrome c1 (CYC1) mRNA as a ref-
erence. Further details regarding quantitative PCR are presented in the
Appendix S1.
Transfection with small interfering RNAs
HTR8/SVneo cells (ATCC, CRL-3271) were initially transfected with
either 30 nM of negative control siRNA or 100 nM of pooled CPPED1
siRNAs (siRNA pairs 1–3) in suspensions and then added into 12-well
plate. After 24 hrs of incubation, adherent cells were re-transfected with
siRNA pairs 1–3 at the same concentration as transfection performed in
suspension and cells harvested after 48 hrs of second transfection. Fur-
ther details regarding transfection are presented in the Appendix S1.
Transcriptomes of siRNA samples
Prior to transcriptomic analysis of siRNA samples, the RNA quality was
determined by Agilent 2100 Bioanalyzer system at Biocenter Oulu
Sequencing Center, Finland. The transcriptomes of the CPPED1 silenced
cells and negative control cells were determined using Illumina HiSeq
high-throughput sequencing system at the Finnish Functional Genomics
Centre, Finland. The sequencing data were analysed by the Bioinformat-
ics Unit Core Service at the Turku Centre for Biotechnology, Finland.
Results
Protein signatures associated with spontaneous
term deliveries
To explore the molecular mechanisms associated with spontaneous
term delivery, we compared protein quantities in human placentas from
uncomplicated pregnancies with spontaneous and elective caesarean
term deliveries. We likewise investigated differences within the placenta
proteome by studying tissue samples collected from basal and chori-
onic plates. Gel-based proteomics with difference gel electrophoresis
(DIGE) approach was used to study not only protein levels but also
modification and processing of intact proteins. In the first step, sample
preparation, protein labelling and separation were optimized for pla-
centa tissue, which resulted in the reproducible detection of about
2,400 spots (Fig. S1). The applied software for two-dimensional (2D)
gel analysis allows 100% spot matching; therefore, all spots were com-
pared among the four study groups, which comprised spontaneous or
elective caesarean birth with samples collected from the basal and
chorionic plates of the same placenta (n = 6 per group). Comparisons
of both delivery types as well as placenta locations revealed 49 signifi-
cantly changed spots (cut-off ratio: 1.5-fold, P ≤ 0.05; Fig. S1),
970 ª 2017 The Authors.
Journal of Cellular and Molecular Medicine published by John Wiley & Sons Ltd and Foundation for Cellular and Molecular Medicine.
Ta
bl
e
1
B
ir
th
ph
en
ot
yp
e-
an
d
lo
ca
tio
n-
de
pe
nd
en
t
ch
an
ge
s
in
th
e
hu
m
an
pl
ac
en
ta
pr
ot
eo
m
e
S
po
t
P
ro
te
in
U
ni
P
ro
t-
K
B
D
es
cr
ip
ti
on
Th
eo
re
ti
ca
l
pI
/M
W
(k
D
)
D
et
ec
te
d
pI
/M
W
(k
D
)
R
at
io
E
ve
rs
us
S
B
a
ve
rs
us
C
h
B
a
C
h
E
S
El
ec
tiv
e
(E
)
ve
rs
us
sp
on
ta
ne
ou
s
(S
)
bi
rt
h
46
*
A
C
TB
P
60
70
9
A
ct
in
,
cy
to
pl
as
m
ic
1
(f
ra
gm
en
t)
5.
29
/4
1.
7
(5
.2
9/
41
.6
)
4.
93
18
1,
57
(1
,1
3)
2,
17
(1
,5
4)
1
A
2M
P
01
02
3
a-
2-
m
ac
ro
gl
ob
ul
in
6.
03
/1
63
.3
5.
51
14
2
1
,5
7
(
1,
45
)
(1
,1
2)
(1
,2
2)
2
(5
.9
8/
16
0.
8)
5.
54
14
2
1
,6
2
1
,5
9
(1
,0
9)
(1
,1
3)
3
5.
57
14
2
1
,5
5
1
,6
7
(
1,
04
)
(
1,
11
)
49
B
2M
P
61
76
9
b-
2-
m
ic
ro
gl
ob
ul
in
6.
06
/1
3.
7
(6
.0
7/
11
.7
)
6.
07
14
(1
,1
2)
1,
59
(
1,
45
)
(
1,
04
)
39
C
P
P
ED
1
Q
9B
R
F8
S
er
in
e/
th
re
on
in
e-
pr
ot
ei
n
ph
os
ph
at
as
e
(i
so
fo
rm
1)
5.
79
/3
5.
5
5.
75
31
1
,7
1
1
,7
9
(
1,
14
)
(
1,
19
)
44
C
Y
B
5A
P
00
16
7
C
yt
oc
hr
om
e
b5
(i
so
fo
rm
1
or
2)
4.
86
/1
5.
31
4.
72
18
1
,7
4
1
,5
9
(1
,0
7)
(1
,1
7)
47
H
B
G
2
P
69
89
2
H
ae
m
og
lo
bi
n
su
bu
ni
t
c-
2
6.
64
/1
6.
1
6.
71
16
2
,5
7
2
,2
6
(1
,0
3)
(1
,1
8)
48
(6
.7
1/
15
.9
)
6.
54
16
2
,3
3
2
,0
3
(1
,1
3)
(1
,2
9)
13
K
R
T8
P
05
78
7
K
er
at
in
,
ty
pe
II
cy
to
sk
el
et
al
8
(i
so
fo
rm
1
or
2)
5.
52
/5
3.
71
5.
24
49
(1
,1
6)
1,
71
(
1,
14
)
(1
,3
0)
14
is
of
or
m
1
or
2
5.
52
/5
3.
71
5.
24
47
(1
,0
8)
1,
73
(
1,
29
)
(1
,2
4)
15
is
of
or
m
1
or
2
5.
52
/5
3.
71
5.
33
50
(1
,5
5)
2,
09
(1
,0
9)
(1
,4
6)
30
*
fr
ag
m
en
t
of
is
of
or
m
1
or
2
5.
52
/5
3.
7
4.
79
37
(1
,8
1)
2,
39
(1
,5
9)
(2
,0
9)
31
*
fr
ag
m
en
t
of
is
of
or
m
1
or
2
5.
52
/5
3.
7
4.
85
38
(1
,8
8)
2,
77
(1
,0
7)
(1
,5
6)
21
K
R
T1
9
P
08
72
7
K
er
at
in
,
ty
pe
I
cy
to
sk
el
et
al
19
5.
05
/4
4.
2
4.
85
41
(
1,
09
)
1,
59
(
1,
15
)
(1
,4
8)
29
*
C
-t
er
m
in
al
K
R
T1
9
fr
ag
m
en
t
5.
05
/4
4.
2
4.
74
37
(1
,1
9)
1,
62
(1
,1
2)
(1
,5
4)
42
P
R
D
X
2
P
32
11
9
P
er
ox
ir
ed
ox
in
-2
(i
so
fo
rm
1)
5.
66
/2
1.
9
(5
.6
7/
21
.8
)
5.
52
22
1
,6
5
(
1,
49
)
(1
,1
0)
(1
,2
1)
22
S
ER
P
IN
B
2
P
05
12
0
P
la
sm
in
og
en
ac
tiv
at
or
in
hi
bi
to
r
2
5.
46
/4
6.
6
5.
51
42
(
1,
14
)
1
,9
2
(
1,
18
)
(
1,
99
)
ª 2017 The Authors.
Journal of Cellular and Molecular Medicine published by John Wiley & Sons Ltd and Foundation for Cellular and Molecular Medicine.
971
J. Cell. Mol. Med. Vol 22, No 2, 2018
Ta
bl
e
1.
C
on
tin
ue
d
S
po
t
P
ro
te
in
U
ni
P
ro
t-
K
B
D
es
cr
ip
ti
on
Th
eo
re
ti
ca
l
pI
/M
W
(k
D
)
D
et
ec
te
d
pI
/M
W
(k
D
)
R
at
io
E
ve
rs
us
S
B
a
ve
rs
us
C
h
B
a
C
h
E
S
B
as
al
pl
at
e
(B
a)
ve
rs
us
ch
or
io
ni
c
pl
at
e
(C
h)
46
*
A
C
TB
P
60
70
9
A
ct
in
,
cy
to
pl
as
m
ic
1
(f
ra
gm
en
t)
5.
29
/4
1.
7
(5
.2
9/
41
.6
)
4.
93
18
1,
57
(1
,1
3)
2,
17
(1
,5
4)
38
A
N
X
A
3
P
12
42
9
A
nn
ex
in
A
3
5.
62
/3
6.
4
(5
.6
3/
36
.2
)
5.
58
31
(
1,
35
)
(1
,3
9)
(1
,5
2)
2,
83
45
*
A
N
X
A
5
P
08
75
8
A
nn
ex
in
A
5
(f
ra
gm
en
t)
4.
93
/3
5.
9
(4
.9
3/
35
.8
)
4,
83
18
(1
,3
9)
(
1,
04
)
1,
55
(1
,0
8)
41
A
P
C
S
P
02
74
3
S
er
um
am
yl
oi
d
P
-c
om
po
ne
nt
6.
10
/2
5.
4
(6
.1
2/
25
.3
)
5.
52
25
(1
,1
3)
(1
,6
7)
(1
,8
9)
2,
75
32
C
LU
P
10
90
9
C
lu
st
er
in
(i
so
fo
rm
3
or
fr
ag
m
en
ts
of
is
of
or
m
s
1,
2,
4,
5)
5.
88
/5
2.
51
4.
76
36
(1
,1
3)
(1
,6
5)
(1
,9
8)
2,
87
33
is
of
or
m
3
or
fr
ag
m
en
ts
of
is
of
or
m
s
1,
2,
4,
5
(5
.8
9/
50
.1
1
)
4.
82
35
(1
,1
3)
(1
,5
7)
2,
16
2,
98
34
is
of
or
m
3
or
fr
ag
m
en
ts
of
is
of
or
m
s
1,
2,
4,
5
4.
88
34
(
1,
11
)
(1
,3
8)
2,
43
3,
69
5
EE
F2
P
13
63
9
El
on
ga
tio
n
fa
ct
or
2
6.
41
/9
5.
3
(6
.4
2/
95
.2
)
6.
42
95
(
1,
02
)
(
1,
77
)
(
1,
34
)
2
,3
2
23
*
FG
B
P
02
67
5
Fi
br
in
og
en
b
(C
-t
er
m
in
al
fr
ag
m
en
t
in
D
di
m
er
)
8.
54
/5
5.
9
5.
51
38
(1
,4
3)
(1
,5
7)
2,
25
2,
45
24
*
C
-t
er
m
in
al
fr
ag
m
en
t
in
D
di
m
er
(7
.9
5/
50
.8
)
5.
60
39
(1
,2
8)
(1
,6
3)
2,
49
3,
17
25
*
C
-t
er
m
in
al
fr
ag
m
en
t
in
D
di
m
er
5.
71
39
(1
,1
6)
(1
,5
6)
2,
54
3,
42
26
*
C
-t
er
m
in
al
fr
ag
m
en
t
in
D
di
m
er
5.
85
38
(1
,5
6)
(1
,6
1)
1,
73
(1
,8
1)
6
FG
G
P
02
67
9
Fi
br
in
og
en
c
(i
so
fo
rm
ga
m
m
a-
A
or
–B
)
5.
37
/5
1.
5A
5.
46
94
(2
,5
5)
(1
,6
8)
2,
07
(1
,3
6)
7
is
of
or
m
c-
A
or
–B
(5
.2
4/
48
.5
A
)
5.
54
94
(2
,1
8)
(1
,5
6)
2,
15
(1
,5
3)
16
is
of
or
m
c-
A
or
–B
5.
54
50
(
1,
13
)
(1
,1
6)
1,
58
2,
05
43
FT
L
P
02
79
2
Fe
rr
iti
n
lig
ht
ch
ai
n
5.
50
/2
0.
0
(5
.5
0/
19
.9
)
5.
50
20
(
1,
03
)
(
1,
47
)
(
1,
97
)
2
,7
5
4
G
S
N
P
06
39
6
G
el
so
lin
(i
so
fo
rm
1)
5.
90
/8
5.
7
(5
.7
2/
82
.9
)
5.
82
95
(
1,
58
)
(1
,3
7)
(1
,1
9)
2,
59
40
H
IB
A
D
H
P
31
93
7
3-
hy
dr
ox
yi
so
bu
ty
ra
te
de
hy
dr
og
en
as
e
8.
38
/3
5.
3
(5
.5
4/
31
.5
)
5.
62
29
(
1,
01
)
(
1,
42
)
(
1,
15
)
1
,6
5
972 ª 2017 The Authors.
Journal of Cellular and Molecular Medicine published by John Wiley & Sons Ltd and Foundation for Cellular and Molecular Medicine.
Ta
bl
e
1.
C
on
tin
ue
d
S
po
t
P
ro
te
in
U
ni
P
ro
t-
K
B
D
es
cr
ip
ti
on
Th
eo
re
ti
ca
l
pI
/M
W
(k
D
)
D
et
ec
te
d
pI
/M
W
(k
D
)
R
at
io
E
ve
rs
us
S
B
a
ve
rs
us
C
h
B
a
C
h
E
S
11
H
N
R
N
P
K
P
61
97
8
H
et
er
og
en
eo
us
nu
cl
ea
r
ri
bo
nu
cl
eo
pr
ot
ei
n
K
(i
so
fo
rm
1,
2
or
3)
5.
39
/5
0.
91
5.
39
56
(1
,2
7)
(
1,
74
)
(
1,
10
)
2
,4
5
27
H
S
D
17
B
1
P
14
06
1
Es
tr
ad
io
l
17
-b
-d
eh
yd
ro
ge
na
se
1
5.
46
/3
4.
9
5.
46
35
(1
,1
1)
(
1,
40
)
(
1,
25
)
1
,9
6
28
(5
.4
7/
34
.8
)
5.
59
35
(1
,0
9)
(
1,
43
)
(
1,
32
)
2
,0
5
10
LM
N
B
1
P
20
70
0
La
m
in
-B
1
5.
11
/6
6.
4
(5
.1
1/
65
.9
)
5.
11
66
(
1,
01
)
(
1,
36
)
(
1,
13
)
1
,5
3
17
LU
M
P
51
88
4
Lu
m
ic
an
6.
16
/3
8.
4
4.
45
53
(
1,
21
)
(1
,4
8)
2,
81
5,
02
18
(6
.1
7/
36
.7
)
4.
51
53
(
1,
24
)
(1
,5
2)
(2
,3
8)
4,
52
19
4.
61
53
(
1,
29
)
(2
,2
2)
(2
,6
5)
7,
59
20
4.
71
52
(
1,
01
)
(2
,3
4)
(2
,2
5)
5,
23
36
TP
M
1
P
09
49
3
Tr
op
om
yo
si
n
a
-1
ch
ai
n
(i
so
fo
rm
8)
4.
71
/3
2.
8
4.
71
33
(
1,
06
)
(1
,3
6)
2,
10
3,
03
37
is
of
or
m
1
or
9
4.
69
/3
2.
71
4.
69
33
(
1,
09
)
(1
,3
2)
(2
,0
1)
2,
90
35
TP
M
2
P
07
95
1
Tr
op
om
yo
si
n
b
ch
ai
n
(i
so
fo
rm
2)
4.
63
/3
2.
9
4.
62
35
(
1,
04
)
(1
,5
4)
2,
45
3,
91
8
V
P
S
35
Q
96
Q
K
1
V
ac
uo
la
r
pr
ot
ei
n
so
rt
in
g-
as
so
ci
at
ed
pr
ot
ei
n
35
5.
32
/9
1.
7
5.
36
82
(
1,
02
)
(
1,
91
)
(
1,
60
)
2
,9
9
9
5.
41
82
(1
,0
2)
(
1,
54
)
(
1,
38
)
2
,1
7
12
W
A
R
S
P
23
38
1
Tr
yp
to
ph
an
-t
R
N
A
lig
as
e
(i
so
fo
rm
1
or
2)
5.
83
/5
3.
21
(5
.8
3/
53
.0
1
)
5.
83
55
(1
,0
9)
(
1,
45
)
(
1,
42
)
2
,2
9
S
po
t
nu
m
be
rs
ar
e
ac
co
rd
in
g
to
Fi
g.
S
1.
D
es
cr
ip
tio
ns
an
d
U
ni
P
ro
t
ac
ce
ss
io
n
nu
m
be
rs
fo
r
th
e
id
en
tifi
ed
sp
ot
s
ar
e
sh
ow
n.
S
po
ts
be
lo
ng
in
g
to
th
e
sa
m
e
pr
ot
ei
n
ar
e
cl
us
te
re
d.
P
ro
te
in
fr
ag
-
m
en
ts
ar
e
in
di
ca
te
d
w
ith
as
te
ri
sk
s.
S
pe
ci
fic
pr
ot
ei
n
is
of
or
m
s
w
er
e
va
lid
at
ed
ac
co
rd
in
g
to
sp
ot
po
si
tio
ns
in
th
e
ge
l
an
d
th
e
co
ve
re
d
pr
ot
ei
n
se
qu
en
ce
as
w
el
l
as
is
of
or
m
-s
pe
ci
fic
pe
pt
id
es
ob
ta
in
ed
by
m
as
s
sp
ec
tr
om
et
ry
.
Th
eo
re
tic
al
sp
ot
id
en
tifi
ca
tio
ns
w
er
e
ca
lc
ul
at
ed
ac
co
rd
in
g
to
fu
ll
or
m
at
ur
e
(i
n
pa
re
nt
he
se
s)
pr
ot
ei
n
se
qu
en
ce
s,
w
he
re
as
de
te
ct
ed
sp
ot
po
si
tio
ns
in
th
e
ge
l
w
er
e
de
te
rm
in
ed
ac
co
rd
in
g
to
se
le
ct
ed
m
ar
ke
r
sp
ot
s.
If
th
e
id
en
tifi
ed
sp
ot
al
lo
w
ed
fo
r
th
e
pr
es
en
ce
of
se
ve
ra
l
is
of
or
m
s,
th
eo
re
tic
al
va
lu
es
w
er
e
in
di
ca
te
d
fo
r
th
e
m
os
t
co
m
m
on
is
o-
fo
rm
1
or
A
.
R
at
io
sh
ow
s
th
e
ch
an
ge
in
m
ea
n
no
rm
al
iz
ed
sp
ot
vo
lu
m
es
be
tw
ee
n
el
ec
tiv
e
(E
)
an
d
sp
on
ta
ne
ou
s
(S
)
te
rm
bi
rt
h
or
be
tw
ee
n
ba
sa
l
(B
a)
an
d
ch
or
io
ni
c
(C
h)
pl
at
es
of
th
e
pl
a-
ce
nt
a.
C
ha
ng
es
w
ith
a
to
o-
lo
w
ra
tio
(<
1.
5-
fo
ld
)
or
no
si
gn
ifi
ca
nc
e
(P
>
0.
05
)
ar
e
de
no
te
d
by
pa
re
nt
he
se
s.
O
nl
y
th
e
am
ou
nt
of
on
e
pr
ot
ei
n
(A
C
TB
)
ch
an
ge
d
si
gn
ifi
ca
nt
ly
in
bo
th
co
m
pa
ri
so
ns
(E
ve
rs
us
S
an
d
B
a
ve
rs
us
C
h)
.
Fo
r
m
or
e
de
ta
ils
ab
ou
t
pr
ot
ei
n
le
ve
ls
,
st
at
is
tic
al
si
gn
ifi
ca
nc
e
an
d
pr
ot
ei
n
id
en
tifi
ca
tio
n,
se
e
Ta
bl
es
S
1
an
d
S
2.
ª 2017 The Authors.
Journal of Cellular and Molecular Medicine published by John Wiley & Sons Ltd and Foundation for Cellular and Molecular Medicine.
973
J. Cell. Mol. Med. Vol 22, No 2, 2018
which were identified by mass spectrometry (MS) analyses to corre-
spond with 28 proteins (Tables 1 and S1).
Ten proteins changed based on the mode of delivery (Fig 1;
Table 1). Spontaneous delivery compared to elective caesarean birth
was associated with up-regulation of four proteins: ACTB, b-2-micro-
globulin (B2M), keratin type II cytoskeletal 8 (KRT8) and 19 (KRT19).
Six proteins were down-regulated in spontaneous delivery compared
to elective caesarean birth: a-2-macroglobulin (A2M), CPPED1, cyto-
chrome b5 (CYB5A), haemoglobin subunit c-2 (HBG2), peroxire-
doxin-2 (PRDX2) and plasminogen activator inhibitor 2 (SERPINB2).
We detected statistically significant differences for A2M, CPPED1,
CYB5A and HGB2 in both the chorionic and the basal plates. By con-
trast, ACTB and PRDX2 were only significantly changed in the basal
plate and B2M, KRT8, KRT19 and SERPINB2 were only significantly
changed in the chorionic plate (Fig. 1).
Different proteomic profiles in placental
chorionic and basal plates
We took samples of placenta from two opposite locations within the
same tissue (chorionic and basal plates) to additionally determine
whether there were local variations in protein levels independent of
birth phenotype. Totally, 19 proteins were altered based on location in
the chorionic and basal plates of the same placenta after spontaneous
and elective caesarean term delivery (n = 6 per group; Fig. 2;
Table 1).
We found higher protein levels in the chorionic plate compared to
the basal plate for 11 proteins: ACTB, annexin A3 (ANXA3), annexin
A5 (ANXA5), serum amyloid P-component (APCS), clusterin (CLU),
fibrinogen b (FGB), fibrinogen c (FGG), gelsolin (GSN), lumican
(LUM), tropomyosin a-1 chain (TPM1) and tropomyosin b chain
(TPM2). The change was statistically significant after spontaneous
and elective caesarean delivery for CLU, FGB, FGG, LUM and TPM1
and 2. ACTB and ANXA5 were significantly changed after elective cae-
sarean delivery, and ANXA3, APCS and GSN were significantly chan-
ged after spontaneous delivery.
Eight proteins were present at lower levels in the chorionic plate
compared to the basal plate: elongation factor 2 (EEF2), ferritin
light chain (FTL), 3-hydroxyisobutyrate dehydrogenase (HIBADH),
heterogeneous nuclear ribonucleoprotein K (HNRNPK), estradiol
17-b-dehydrogenase 1 (HSD17B1), lamin-B1 (LMNB1), vacuolar
protein sorting-associated protein 35 (VPS35) and tryptophan-
tRNA ligase (WARS). These changes were only significant after
spontaneous delivery (Fig. 2; Table 1). Our findings indicate that
there are functionally different subregions within the placenta,
Fig. 1 Proteomic changes between elective (E) and spontaneous (S) term birth in basal (Ba) and chorionic (Ch) plates of the placenta. Sections of
the gel represent the exact spot positions and expression profiles of the protein level according to the mean normalized spot volumes for the spots
shown in Fig. S1. Spots corresponding to the same protein are clustered with a vertical line. Protein fragments are indicated with asterisks. Statisti-
cally significant changes (P < 0.05) in expression profiles are shown by horizontal brackets. In the comparisons, six placenta samples were included
per group.
974 ª 2017 The Authors.
Journal of Cellular and Molecular Medicine published by John Wiley & Sons Ltd and Foundation for Cellular and Molecular Medicine.
highlighting the importance of carefully collecting samples from
the same location.
Delivery type and tissue location alter specific
protein variants
We detected several proteins in multiple spots and/or at unexpected
positions in the 2D gel. This suggests that additional regulatory
mechanisms such as alternative splicing, protein processing, frag-
mentation and other modifications are taking place. Therefore, the
theoretical spot positions calculated with Compute pI/Mw and the
detected spot positions in the gel (Fig. S1; Table 1) were compared.
In addition, sequence coverage and spot-specific peptides obtained
by MS (Table S2) were also examined. We found specific isoforms
associated with the delivery type (e.g. CPPED1 isoform 1 and PRDX2
isoform 1) and placenta location (e.g. GSN isoform 1, TPM1 isoform
8 and TPM2 isoform 2). Likewise, numerous protein fragments were
Fig. 2 Proteomic changes between basal (Ba) and chorionic (Ch) plates of the placenta. Samples were collected after elective (E) or spontaneous
(S) term birth. Gel parts represent the exact spot positions and expression profiles of the protein level according to the mean normalized spot vol-
umes for the spots shown in Fig. S1. For spots 17–20, the exact positions are shown in gel parts representative of the chorionic plate. Spots corre-
sponding to the same protein are clustered with a vertical line. Protein fragments are indicated with asterisks. Statistically significant changes
(P < 0.05) in expression profiles are shown by horizontal brackets. In the comparisons, six placenta samples were included per group.
ª 2017 The Authors.
Journal of Cellular and Molecular Medicine published by John Wiley & Sons Ltd and Foundation for Cellular and Molecular Medicine.
975
J. Cell. Mol. Med. Vol 22, No 2, 2018
associated with delivery type (KRT8 and KRT19), placenta location
(ANXA5 and FGB) or both (ACTB). However, all multiple spots of
A2M, HBG2, KRT8, KRT19, as well as CLU, FGB, FGG, HSD17B1,
LUM, TPM1 and VPS35 showed changes similar to those of the cor-
responding protein. Thus, the detected protein signatures are likely a
result of altered gene expression rather than specific processing.
Protein changes during spontaneous term
delivery mostly associated with immune system
and structural/cytoskeletal organization
We further clustered proteins associated with the delivery type
(Table 1) according to their gene ontology (GO) biological processes
and molecular functions (Table S3). The majority of the GO terms
were linked to either the immune system or to structural/cytoskeletal
organization. Biological processes comprising at least three changed
proteins included responses to inorganic substance (ACTB, A2M and
PRDX2), organic substance (A2M, B2M, KRT19 and PRDX2) and
wounding (A2M, PRDX2 and SERPINB2). Structural molecule activity
(ACTB, KRT8 and KRT19) represented the major molecular function.
In addition, A2M and SERPINB2 were in the same blood coagulation
pathway. CPPED1 was not associated with any of the enriched GO
terms. As a phosphatase, CPPED1 is thought to be involved in block-
ing cell cycle progression, promoting apoptosis and affecting glucose
metabolism [19, 41].
CPPED1 polymorphism associated with
gestational age
Changes in protein levels are expected during labour. To identify pro-
teins that are important in initiating labour rather than affected by it,
we investigated whether single-nucleotide polymorphisms (SNPs)
within 20 kb of the genes encoding the aforementioned proteins
(ACTB, A2M, B2M, CPPED1, CYB5A, HBG2, KRT8, KRT19, PRDX2
and SERPINB2) were associated with gestational age (GA) in infants
born at term. We analysed 77 SNPs from 342 samples from infants
born between 38 and 42 weeks of gestation. Of all the analysed SNPs,
only two intronic SNPs within CPPED1 were associated with GA at the
Bonferroni-corrected significance level (P < 6.5 9 104):
rs11643593 (P = 2.7 9 104) and rs8048866 (P = 3.3 9 104;
Table S4). The minor allele of rs11643593 (A) and major allele of
rs8048866 (A) were associated with higher GA (Table S5). This sug-
gests that CPPED1 may be involved in the initiation of term delivery
and prompted us to study its expression in the placenta during labour
in more detail.
CPPED1 expressed in trophoblasts of full-term
placentas
To determine which cell types express CPPED1 in the placenta, we
immunostained placentas from spontaneous and elective caesarean
term deliveries with anti-human CPPED1 antibody. As shown in
Figure 3, a CPPED1 signal was obtained from samples of both basal
and chorionic plates of spontaneous and elective term placentas.
CPPED1 was detected in the cytoplasm and cell membranes of
cytotrophoblasts and syncytiotrophoblasts (Fig. 3). CPPED1 staining
was also positive in the epithelium of the amniotic membranes. In the
decidua, extravillous trophoblasts were positive for CPPED1. The cap-
illary endothelia of the villi showed only faint and patchy staining,
whereas intracapillary and intervillous leucocytes were highly positive
(Fig. 3).
Lower levels of CPPED1 mRNA in spontaneous
term deliveries
To investigate whether the decreased levels of CPPED1 protein that
we observed in spontaneous delivery were the result of altered tran-
scription, we performed quantitative PCR (qPCR) analysis. Figure 4
Fig. 3 Placental localization of CPPED1 in spontaneous and elective
births. Immunostaining of CPPED1 in full-term placentas from sponta-
neous (n = 6) or elective births (n = 6). Representative placentas
immunostained with anti-human CPPED1 antibodies. Immunostaining is
indicated by filled large arrows in cytotrophoblasts, unfilled large arrows
in syncytiotrophoblasts, filled small arrows in endothelium, unfilled
arrowheads in leucocytes and filled arrowheads in decidual tro-
phoblasts. Original magnification is 209 in all figures. Control repre-
sents isotype controls for immunostaining. Scale bar, 100 lm.
976 ª 2017 The Authors.
Journal of Cellular and Molecular Medicine published by John Wiley & Sons Ltd and Foundation for Cellular and Molecular Medicine.
shows a comparison of CPPED1 mRNA levels after spontaneous
(n = 18) and elective caesarean (n = 13) deliveries in the basal and
chorionic plates of the placenta. CPPED1 mRNA levels were signifi-
cantly lower in spontaneous term deliveries compared to elective cae-
sarean term births in the basal (2.3-fold, P = 0.000062) and
chorionic (1.5-fold, P = 0.041) plates. The genotypes of the two
CPPED1 SNPs (rs11643593 and rs8048866) associated with GA in
term infants were not significantly associated with mRNA levels.
However, the change in CPPED1 mRNA levels was in line with what
we observed for CPPED1 protein levels.
CPPED1 does not affect phosphorylation of AKT1
and FOXO1
CPPED1 has been shown to directly dephosphorylate Ser473 of AKT1
in cancer tissue [19]. Furthermore, a downstream target of AKT1 is
forkhead box protein O1 (FOXO1, also known as forkhead homologue
in rhabdomyosarcoma [FKHR]) [42], which is located in the same
transcriptional complex as progesterone receptor B [43, 44] and reg-
ulates decidual transcription of certain genes [45, 46]. Therefore, we
investigated the phosphorylation status of AKT1 at Ser473 and FOXO1
at Ser256 in the human placenta by Western blotting to find out
whether lower levels of CPPED1 during spontaneous labour are asso-
ciated with increased phosphorylation of AKT1 and FOXO1. When we
quantitated and compared samples from both spontaneous and elec-
tive caesarean labour, we found no detectable changes in the degree
of phosphorylation at either Ser473 of AKT1 (P = 0.211) or Ser256 of
FOXO1 (P = 0.61; Fig. 5).
Previous reports have indicated that phosphorylation of FOXO1
affects its function and localization in cells [47–49] and that AKT1 is
additionally able to phosphorylate FOXO3 (also known as forkhead
homologue in rhabdomyosarcoma-like 1 [FKHRL1])[50]. Therefore,
we used immunohistochemistry to examine the expression of FOXO1
and FOXO3 in human placenta. We detected phosphorylated FOXO1
and phosphorylated FOXO3 in syncytio- and cytotrophoblasts as well
as in decidual trophoblasts (Fig. 6). There was strong phosphorylated
FOXO1 and FOXO3 staining in the nuclei of trophoblasts and moder-
ate amounts in the cytosol. However, we did not detect any differ-
ences in the amount or subcellular localization of phosphorylated
FOXO1 and FOXO3 when we compared placenta samples from spon-
taneous and elective caesarean deliveries. These results suggest that
the role of CPPED1 in the timing of spontaneous birth is independent
of the phosphorylation status of AKT1 and FOXO1.
Silencing of CPPED1 affects gene expression
related to inflammation and blood vessel
development
To get initial insight into the function of CPPED1 in placenta, we
silenced CPPED1 post-transcriptionally with small interfering RNA
(siRNA) in HTR-8/SVneo cells. These cells are of human origin and
they have extravillous invasive trophoblast characteristic. Based on
qRT-PCR, the silencing of CPPED1 at mRNA level was 68% (Fig. S2).
When gene expression levels of HTR-8/SVneo cells were analysed
using RNA sequencing, there were 147 genes that were either up
(n = 77)- or down (n = 70)-regulated by CPPED1 silencing com-
pared to control cells. When the differentially expressed genes
(Tables S6 and S7) were grouped according to their GO biological
processes and molecular functions (Table S8), the inflammation and
blood vessel and vasculature development-related pathways were
Fig. 4 The mRNA levels of CPPED1 in spontaneous and elective term placentas. Relative mRNA levels from the basal or chorionic plates were nor-
malized to mRNA levels of the housekeeping gene CYC1. Differences were analysed with non-parametric Mann–Whitney U-test. Statistically signifi-
cant changes according to the non-parametric Mann–Whitney U-test are indicated by one (P < 0.05) and four (P < 0.0001) asterisks. Solid lines
denote median values.
ª 2017 The Authors.
Journal of Cellular and Molecular Medicine published by John Wiley & Sons Ltd and Foundation for Cellular and Molecular Medicine.
977
J. Cell. Mol. Med. Vol 22, No 2, 2018
among the top affected pathways. For example, in the GO molecular
functions analysis (Table S8), cytokine activity (GO:0005125) had the
lowest P-value (8.9 9 1011) and included differentially expressed
genes: CXCL8, CXCL1, IL6, CCL2, CCL7, IL1b, GDF15, VSTM1,
CXCL2, AREG, IL24, IL1a, CXCL6, EDN1 and IL18 (Tables S6 and
S7). Moreover, insulin-like growth factor binding protein 5 (IGFBP5)
Fig. 5 Phosphorylation of AKT1 and FOXO1 in spontaneous and elective births. The phosphorylation at Ser473 of AKT1 (pAKT1) was quantitated and
compared to total AKT1 (spontaneous term birth n = 18, elective term birth n = 13) (A) and phosphorylation at Ser256 of FOXO1 (pFOXO1) was
quantitated and compared to total FOXO1 (spontaneous term birth n = 16, elective term birth n = 12) (B) by Western blotting. NS means not signif-
icant change according to Mann–Whitney U-test. Solid lines denote median values.
Fig. 6 Placental localization of phosphorylated FOXO1 and phosphorylated FOXO3 in spontaneous and elective births. Immunostaining of phosphory-
lated FOXO1 (A) and phosphorylated FOXO3 (B) in full-term placentas after spontaneous and elective delivery. Immunostaining is indicated by filled
large arrows in cytotrophoblasts, unfilled large arrows in syncytiotrophoblasts and filled arrowheads in decidual trophoblasts. Original magnification
is 409 in samples and 209 in controls. Controls represent isotype controls for immunostaining. Scale bars, 16 lm and 25 lm for samples and
controls, respectively.
978 ª 2017 The Authors.
Journal of Cellular and Molecular Medicine published by John Wiley & Sons Ltd and Foundation for Cellular and Molecular Medicine.
and phosphoinositide-3-kinase interacting protein 1 (PIK3IP1) were
significantly up-regulated, and phosphatidylinositol-4,5-bisphosphate
3-kinase catalytic subunit gamma (PIK3CG) was significantly down-
regulated in the case of lowered levels of CPPED1. These changes
putatively inhibit PI3K/AKT signalling pathway.
Discussion
Our aim in the present study was to identify placental proteins that
have functional roles in the initiation of term labour. We compared
the human placenta proteome after spontaneous and elective cae-
sarean term delivery at two tissue locations, the basal and chorionic
plates, to identify biochemical pathways associated with spontaneous
term birth. We identified statistically significant differences in the
quantity of ten proteins between spontaneous and elective caesarean
term deliveries after uncomplicated pregnancies. The quantitative
changes were significant in either the basal or the chorionic plates or
in both.
A secondary aim of our study was to compare the proteomes of
the chorionic and basal plates to identify specific protein signatures
for the foetal (chorionic plate) and the maternal (basal plate) sides of
the placenta and to verify spontaneous term birth-dependent changes
in different placenta locations. We identified 19 location-dependent
proteins. Among these proteins, 11 were present at higher levels in
the chorionic plate. The remaining eight proteins were present at
higher levels in the basal plate, suggesting that they may have a role
in the placenta–uterine wall interface. To the best of our knowledge,
thus far only one other proteomic study has compared placenta sides
[36]. Kedia et al. analysed uncomplicated vaginal and caesarean
deliveries with particular emphasis on low molecular weight proteins
and peptides. However, none of the proteins identified in our study
were linked to a specific placenta region by Kedia et al. Furthermore,
our finding that levels of these proteins differed depending on location
indicates that standardized collection of tissue samples from the
same placental location is essential for analyses.
One of our identified proteins was the recently characterized
CPPED1 [19, 41]. Additionally, we detected a genetic association
between the length of term pregnancy and CPPED1 but not for any
other genes encoding detected proteins. Therefore, we investigated
CPPED1 and its potential function during spontaneous term birth in
more detail. We found that the lower levels of CPPED1 in sponta-
neous term birth compared to elective caesarean term delivery were
associated with a lower rate of CPPED1 transcription. Proteomics
revealed similar CPPED1 levels in the basal and chorionic plates, indi-
cating that CPPED1 is not restricted to either the foetal or maternal
side of the placenta. The reduced CPPED1 levels that we observed in
association with spontaneous term birth compared to elective term
birth suggest that its major function takes place until onset of labour
but not during labour.
Zhuo et al. [19] detected decreased levels of CPPED1 mRNA in
bladder cancer but not in pancreatic, liver, stomach, colon or renal
cancer. They also found that CPPED1 has serine/threonine-protein
phosphatase activity and can inactivate AKT1 through specific
dephosphorylation of Ser473. AKT1 is a critical component of the
PI3K/AKT signalling pathway, which regulates multiple processes
including metabolism, protein synthesis, cell cycle, proliferation,
angiogenesis, apoptosis and survival [13–15]. Downstream targets of
AKT1 include FOXO1 [42] and FOXO3 [50, 51]. FOXO1 acts as a DNA-
binding transcription factor [52, 53] and regulates cell cycle and
inflammation genes [52]. Phosphorylation of FOXO1 by AKT1 may
cause localization to the cytoplasm and subsequent degradation of
FOXO1 [54, 55]. In addition to regulation by phosphorylation, FOXO1
and FOXO3 are also regulated by various types of post-translational
modifications such as acetylation. Deacetylation by NAD-dependent
protein deacetylase sirtuin-1 (SIRT1) results in the nuclear retention
of FOXO1 and FOXO3 [48,56].
A previous study detected AKT1 in both villous and extravillous
trophoblasts of human placenta [57]. We found that not only CPPED1
but also FOXO1 and FOXO3 were expressed in cytotrophoblasts, syn-
cytiotrophoblasts and decidual extravillous trophoblasts in human
placenta. Interestingly, the majority of phosphorylated FOXO1 and
FOXO3 was detected in the nuclei of these cells. Due to their nuclear
localization, these phosphorylated FOXOs were presumably function-
ally active and deacetylated [48]. In the present study, immunohisto-
chemistry revealed no differences in subcellular localization and
amounts of phosphorylated FOXO1 and FOXO3 between the placentas
of spontaneous and elective caesarean term births. Nor did we
observe any changes in phosphorylation levels of AKT1 and FOXO1
associated with lower CPPED1 levels. This suggests that CPPED1
does not a regulate AKT1 in AKT-mediated signalling pathway in pla-
centa. Interestingly, when we used siRNAs in reducing the CPPED1
levels in immortalized extravillous invading trophoblasts, we noticed
up-regulation of the negative regulators of PI3K/AKT signalling path-
way upstream of AKT1. However, the most significant affected path-
ways involved the cytokine activity and blood vessel development. It
is known, that tumour invasion and trophoblastic invasion share the
same biochemical mediators, such as matrix metalloproteinases,
cytokines and growth factors, and that trophoblast invasion and vas-
cular remodelling are essential during pregnancy [58]. Moreover, tro-
phoblasts can alter the local immune environment of the decidua by
allowing proinflammatory environment in early invasive phase of for-
mation of placental vessels [59]. Possibly, CPPED1 may also have a
role in placentation and promote this proinflammatory effect.
In conclusion, the labour-associated molecular mechanisms that
take place in the placenta before and during spontaneous term birth
are poorly understood. We have described protein signatures that
were characteristic for the delivery type as well as the placental loca-
tion, thereby providing important information about local and general
changes in the placenta during parturition. We identified CPPED1 as a
promising marker for spontaneous term delivery, confirmed a highly
significant genetic association between intronic SNPs in the CPPED1
gene and GA and identified proinflammatory pathways affected by
lowered CPPED1 levels. To date, CPPED1 has been shown to have dif-
ferent effects depending on the cell type [19, 41]. Accordingly,
CPPED1 might affect differently in early and late pregnancy tro-
phoblasts. Therefore, additional studies are required to further specify
the role of CPPED1 in regulation of inflammatory mediators in tro-
phoblasts and to identify the downstream targets of CPPED1 at the
onset of spontaneous labour.
ª 2017 The Authors.
Journal of Cellular and Molecular Medicine published by John Wiley & Sons Ltd and Foundation for Cellular and Molecular Medicine.
979
J. Cell. Mol. Med. Vol 22, No 2, 2018
Acknowledgements
We acknowledge Maarit Haarala and Leila Polus for technical assistance, as
well as Riitta Vikev€ainen, Sonja Eeli and Miia Lehto for sample collection. This
work was supported by the Jane and Aatos Erkko Foundation (MH and MR),
Sigrid Juselius Foundation (MH), Competitive State Research Financing of the
Expert Responsibility Area of Oulu University Hospital (MR), Emil Aaltonen
Foundation (MO), KA Snellman Foundation (MO and MR) and Pediatric
Research Foundation (AMH).
Conflict of interest
The authors confirm that there are no conflicts of interest.
Supporting information
Additional Supporting Information may be found online in the
supporting information tab for this article:
Appendix S1 Materials and methods.
Table S1. Protein levels and statistical significance.
Table S2. Protein identification.
Table S3. Functional classification of spontaneous birth–associ-
ated placental proteins.
Table S4. Analysis of SNPs within and near ACTB, A2M, B2M,
CPPED1, CYB5A, HBG2, KRT8, KRT19, PRDX2, and SERPINB2
for associations with gestational age in infants born at term.
Table S5. Means and medians of gestational age for infants
with each genotype for the CPPED1 SNPs associated with ges-
tational age.
Table S6. Downregulated genes after CPPED1 silencing in
HTR8/SVneo cells.
Table S7. Upregulated genes after CPPED1 silencing in HTR8/
SVneo cells.
Table S8. Functional classification of differentially expressed
genes after post-transcriptional silencing of CPPED1 in HTR8/
SVneo cells.
Figure S1 Representative 2D gel of human placenta tissue after sponta-
neous term birth. Placenta proteins (50 lg) collected from the basal plate
of the placenta were labeled with Cy5 (minimal difference gel electro-
phoresis) and separated by isoelectic focusing (pH4–7, 24 cm) and SDS-
PAGE. Positions of spots that were significantly changed are indicated.
Figure S2 The effect of siRNA on CPPED1 mRNA levels. CPPED1 was
post-transcriptionally silenced in HTR8/SVneo cells that is a human
placental trophoblast continuous cell line. RNA was isolated from
CPPED1 silenced cells and compared to RNA from control cells. In
the figure, CPPED1 expression levels are shown as determined by
qRT-PCR (A) and high throughput RNA sequencing (B). Relative
mRNA levels were normalized to mRNA levels of the housekeeping
gene CYC1 (A). Reads per kilobase of exon per million reads mapped
(RKPM) is a normalized gene counts value determined in the
transcriptomic analysis (B). The columns show the mean value of
triplicate samples; the maximum and minimum values are also
indicated.
References
1. Smith R. Parturition. N Engl J Med. 2007;
356: 271–83.
2. Nair RR, Verma P, Singh K. Immune-endo-
crine crosstalk during pregnancy. Gen Comp
Endocrinol. 2016; 242: 18–23.
3. Casey ML, MacDonald PC. The endocrinol-
ogy of human parturition. Ann N Y Acad Sci.
1997; 828: 273–84.
4. McLean M, Smith R. Corticotropin-releasing
hormone in human pregnancy and parturition.
Trends Endocrinol Metab. 1999; 10: 174–8.
5. Ni X, Hou Y, Yang R, et al. Progesterone
receptors A and B differentially modulate
corticotropin-releasing hormone gene
expression through a cAMP regulatory ele-
ment. Cell Mol Life Sci. 2004a; 61: 1114–22.
6. Ni X, Hou Y, King BR, et al. Estrogen recep-
tor-mediated down-regulation of corti-
cotropin-releasing hormone gene expression
is dependent on a cyclic adenosine 30,50-
monophosphate regulatory element in
human placental syncytiotrophoblast cells. J
Clin Endocrinol Metab. 2004b; 89: 2312–8.
7. Petraglia F, Sutton S, Vale W. Neurotrans-
mitters and peptides modulate the release of
immunoreactive corticotropin-releasing fac-
tor from cultured human placental cells. Am
J Obstet Gynecol. 1989; 160: 247–51.
8. Leung TN, Chung TK, Madsen G, et al. Rate
of rise in maternal plasma corticotrophin-
releasing hormone and its relation to gesta-
tional length. BJOG. 2001; 108: 527–32.
9. Koullali B, Oudijk MA, Nijman TA, et al.
Risk assessment and management to pre-
vent preterm birth. Semin Fetal Neonatal
Med. 2016; 21: 80–8.
10. Kiss H, Ahner R, Hohlagschwandtner M,
et al. Fetal fibronectin as a predictor of term
labor: a literature review. Acta Obstet Gyne-
col Scand. 2000; 79: 3–7.
11. Mesiano S, Chan EC, Fitter JT, et al. Pro-
gesterone withdrawal and estrogen
activation in human parturition are coordi-
nated by progesterone receptor A expres-
sion in the myometrium. J Clin Endocrinol
Metab. 2002; 87: 2924–30.
12. Tuckey RC. Progesterone synthesis by the
human placenta. Placenta. 2005; 26: 273–81.
13. Lee II, Kim JJ. Influence of AKT on proges-
terone action in endometrial diseases. Biol
Reprod. 2014; 91: 1–10.
14. Toker A, Cantley LC. Signalling through the
lipid products of phosphoinositide-3-OH
kinase. Nature. 1997; 387: 673–6.
15. Stephens L, Anderson K, Stokoe D, et al.
Protein kinase B kinases that mediate phos-
phatidylinositol 3,4,5-trisphosphate-depen-
dent activation of protein kinase B. Science.
1998; 279: 710–4.
16. Liu P, Cheng H, Roberts TM, et al. Targeting
the phosphoinositide 3-kinase pathway in
cancer. Nat Rev Drug Discov. 2009; 8:
627–44.
980 ª 2017 The Authors.
Journal of Cellular and Molecular Medicine published by John Wiley & Sons Ltd and Foundation for Cellular and Molecular Medicine.
17. Vivanco I, Sawyers CL. The phosphatidyli-
nositol 3-Kinase AKT pathway in human can-
cer. Nat Rev Cancer. 2002; 2: 489–501.
18. Jacinto E, Facchinetti V, Liu D, et al. SIN1/
MIP1 maintains rictor-mTOR complex integ-
rity and regulates Akt phosphorylation and
substrate specificity. Cell. 2006; 127: 125–37.
19. Zhuo DX, Zhang XW, Jin B, et al. CSTP1, a
novel protein phosphatase, blocks cell cycle,
promotes cell apoptosis, and suppresses
tumor growth of bladder cancer by directly
dephosphorylating Akt at Ser473 site. PLoS
ONE. 2013; 8: e65679.
20. Mitchell MD, Peiris HN, Kobayashi M,
et al. Placental exosomes in normal and
complicated pregnancy. Am J Obstet Gyne-
col. 2015; 213: S173–81.
21. Saben J, Zhong Y, McKelvey S, et al. A com-
prehensive analysis of the human placenta
transcriptome. Placenta. 2014; 35: 125–31.
22. Kim J, Zhao K, Jiang P, et al. Transcrip-
tome landscape of the human placenta. BMC
Genom. 2012; 13: 115–136.
23. Januar V, Desoye G, Novakovic B, et al.
Epigenetic regulation of human placental
function and pregnancy outcome: considera-
tions for causal inference. Am J Obstet
Gynecol. 2015; 213: S182–96.
24. Kacerovsky M, Lenco J, Musilova I, et al.
Proteomic biomarkers for spontaneous pre-
term birth: a systematic review of the litera-
ture. Reprod Sci. 2014; 21: 283–95.
25. Georgiou HM, Di Quinzio MK, Permezel M,
et al. Predicting preterm labour: current sta-
tus and future prospects. Dis Markers.
2015; 2015: 435014.
26. Menon R, Torloni MR, Voltolini C, et al.
Biomarkers of spontaneous preterm birth:
an overview of the literature in the last four
decades. Reprod Sci. 2011; 18: 1046–70.
27. Yuan W, Heesom K, Phillips R, et al. Low
abundance plasma proteins in labour.
Reproduction. 2012; 144: 505–18.
28. Heng YJ, Liong S, Permezel M, et al.
Human cervicovaginal fluid biomarkers to
predict term and preterm labor. Front Phys-
iol. 2015; 6: 151–169.
29. Miao Z, Chen M, Wu H, et al. Comparative
proteomic profile of the human placenta in
normal and fetal growth restriction subjects.
Cell Physiol Biochem. 2014; 34: 1701–10.
30. Ni X, Li X, Guo Y, et al. Quantitative pro-
teomics analysis of altered protein expres-
sion in the placental villous tissue of early
pregnancy loss using isobaric tandem mass
tags. Biomed Res Int. 2014; 2014: 647143.
31. Liu AX, Jin F, Zhang WW, et al. Proteomic
analysis on the alteration of protein expression
in the placental villous tissue of early preg-
nancy loss. Biol Reprod. 2006; 75: 414–20.
32. Khan GH, Galazis N, Docheva N, et al.
Overlap of proteomics biomarkers between
women with pre-eclampsia and PCOS: a
systematic review and biomarker database
integration. Hum Reprod. 2015; 30: 133–48.
33. Yang JI, Kong TW, Kim HS, et al. The pro-
teomic analysis of human placenta with pre-
eclampsia and normal pregnancy. J Korean
Med Sci. 2015; 30: 770–8.
34. Chen CP, Chen YH, Chern SR, et al. Pla-
centa proteome analysis from Down syn-
drome pregnancies for biomarker discovery.
Mol BioSyst. 2012; 8: 2360–72.
35. Gharesi-Fard B, Zolghadri J, Kamali-Sar-
vestani E. Proteome differences in the first-
and third-trimester human placentas.
Reprod Sci. 2015; 22: 462–8.
36. Kedia K, Nichols CA, Thulin CD, et al. Novel
“omics” approach for study of low-abundance,
low-molecular-weight components of a complex
biological tissue: regional differences between
chorionic and basal plates of the human pla-
centa. Anal Bioanal Chem. 2015; 407: 8543–56.
37. Karjalainen MK, Ojaniemi M, Haapalainen
AM, et al. CXCR3 polymorphism and
expression associate with spontaneous pre-
term birth. J Immunol. 2015; 195: 2187–98.
38. da Huang W, Sherman BT, Lempicki RA.
Systematic and integrative analysis of large
gene lists using DAVID bioinformatics
resources. Nat Protoc. 2009a; 4: 44–57.
39. da Huang W, Sherman BT, Lempicki RA.
Bioinformatics enrichment tools: paths
toward the comprehensive functional analy-
sis of large gene lists. Nucleic Acids Res.
2009b; 37: 1–13.
40. Purcell S, Neale B, Todd-Brown K, et al.
PLINK: a tool set for whole-genome associa-
tion and population-based linkage analyses.
Am J Hum Genet. 2007; 81: 559–75.
41. Vaittinen M, Kaminska D, Kakela P, et al.
Downregulation of CPPED1 expression
improves glucose metabolism in vitro in adi-
pocytes. Diabetes. 2013; 62: 3747–50.
42. Rena G, Guo S, Cichy SC, et al. Phosphory-
lation of the transcription factor forkhead
family member FKHR by protein kinase B. J
Biol Chem. 1999; 274: 17179–83.
43. Diep CH, Charles NJ, Gilks CB, et al. Pro-
gesterone receptors induce FOXO1-depen-
dent senescence in ovarian cancer cells. Cell
Cycle. 2013; 12: 1433–49.
44. Kim JJ, Buzzio OL, Li S, et al. Role of
FOXO1A in the regulation of insulin-like growth
factor-binding protein-1 in human endometrial
cells: interaction with progesterone receptor.
Biol Reprod. 2005; 73: 833–9.
45. Vasquez YM, Mazur EC, Li X, et al. FOXO1
is required for binding of PR on IRF4, novel
transcriptional regulator of endometrial
stromal decidualization. Mol Endocrinol.
2015; 29: 421–33.
46. Takano M, Lu Z, Goto T, et al. Transcriptional
cross talk between the forkhead transcription
factor forkhead box O1A and the progesterone
receptor coordinates cell cycle regulation and
differentiation in human endometrial stromal
cells. Mol Endocrinol. 2007; 21: 2334–49.
47. Zhao X, Gan L, Pan H, et al. Multiple ele-
ments regulate nuclear/cytoplasmic shuttling
of FOXO1: characterization of phosphoryla-
tion- and 14-3-3-dependent and -independent
mechanisms. Biochem J. 2004; 378: 839–49.
48. Daitoku H, Sakamaki J, Fukamizu A. Regula-
tion of FoxO transcription factors by acetyla-
tion and protein-protein interactions. Biochim
Biophys Acta. 2011a; 1813: 1954–60.
49. Matsuzaki H, Daitoku H, Hatta M, et al.
Acetylation of Foxo1 alters its DNA-binding
ability and sensitivity to phosphorylation. Proc
Natl Acad Sci USA. 2005; 102: 11278–83.
50. Brunet A, Bonni A, Zigmond MJ, et al. Akt
promotes cell survival by phosphorylating
and inhibiting a Forkhead transcription fac-
tor. Cell. 1999; 96: 857–68.
51. Tzivion G, Dobson M, Ramakrishnan G.
FoxO transcription factors; Regulation by
AKT and 14-3-3 proteins. Biochim Biophys
Acta. 2011; 1813: 1938–45.
52. Burgering BM, Kops GJ. Cell cycle and death
control: long live Forkheads. Trends Bio-
chem Sci. 2002; 27: 352–60.
53. Kaestner KH, Knochel W, Martinez DE. Uni-
fied nomenclature for the winged helix/fork-
head transcription factors. Genes Dev. 2000;
14: 142–6.
54. Matsuzaki H, Daitoku H, Hatta M, et al. Insu-
lin-induced phosphorylation of FKHR (Foxo1)
targets to proteasomal degradation. Proc Natl
Acad Sci USA. 2003; 100: 11285–90.
55. Rena G, Woods YL, Prescott AR, et al. Two
novel phosphorylation sites on FKHR that
are critical for its nuclear exclusion. EMBO J.
2002; 21: 2263–71.
56. Frescas D, Valenti L, Accili D. Nuclear trap-
ping of the forkhead transcription factor
FoxO1 via Sirt-dependent deacetylation pro-
motes expression of glucogenetic genes. J
Biol Chem. 2005; 280: 20589–95.
57. Fabi F, Asselin E. Expression, activation,
and role of AKT isoforms in the uterus.
Reproduction. 2014; 148: R85–95.
58. Bischof P, Campana A. A putative role for
oncogenes in trophoblast invasion? Hum
Reprod. 2000; 15(Suppl 6): 51–8.
59. Hess AP, Hamilton AE, Talbi S, et al. Decid-
ual stromal cell response to paracrine sig-
nals from the trophoblast: amplification of
immune and angiogenic modulators. Biol
Reprod. 2007; 76: 102–17.
ª 2017 The Authors.
Journal of Cellular and Molecular Medicine published by John Wiley & Sons Ltd and Foundation for Cellular and Molecular Medicine.
981
J. Cell. Mol. Med. Vol 22, No 2, 2018