CXCL14 Inhibits Trophoblast Outgrowth Via a Paracrine/Autocrine Manner During Early Pregnancy in Mice
CXCL14, a member of chemokine family, was previously known to participate in many pathophysiological events, such as leukocytes recruitment and tumor suppression. However, it remained largely unknown whether CXCL14 is a physiological player during early pregnancy. In this regard, our recent global gene microarray analysis has observed an implantation-specific expression profile of CXCL14 mRNA during early pregnancy in mice, showing its higher levels at implantation sites compared to inter-implantation sites, implicating a potential role of CXCL14 in the periimplantation events. In the present investigation, using Northern blot, in situ hybridization and immunostaining, we further demonstrated that uterine CXCL14 expression was specifically induced at embryo implantation site and expanded with subsequent decidualization process in a spatiotemporal manner. The implanting embryo also showed a highlighted expression of CXCL14 in the blastocyst trophectoderm and its derived ectoplacental cones (EPCs) during postimplantation development. In vitro functional study revealed that CXCL14 could significantly inhibit both primary and secondary trophoblast attachment and outgrowth, correlated with a stage-dependant downregulation of MMP-2 and/or MMP-9 activity. Moreover, it was found that biotinylated CXCL14 could specifically bind to trophoblast cells in vitro and in vivo, suggesting trophoblast cell, perhaps expressing the unidentified CXCL14 receptor, is a bioactive target of CXCL14. Collectively, our findings provide evidences supporting the contention that CXCL14 is an important paracrine/autocrine modulator regulating trophoblast outgrowth at the maternal-fetal interface during the process of pregnancy establishment. This study is clinically related since CXCL14 is also highly expressed in human receptive endometrium and trophoblasts.
CXCL14 Inhibits Trophoblast
Outgrowth Via a Paracrine/
Autocrine Manner During Early
Pregnancy in Mice
AND ENKUI DUAN
State Key Laboratory of Reproductive Biology, Institute of Zoology, Chinese Academy of Sciences, Beijing, P.R. China
Graduate School of the Chinese Academy of Sciences, Beijing, P.R. China
Department of Physiology, School of Medicine, Nanchang University,
Nanchang, P.R. China
CXCL14, a member of chemokine family, was previously known to participate in many pathophysiological events, such as leukocytes
recruitment and tumor suppression. However, it remained largely unknown whether CXCL14 is a physiological player during early
pregnancy. In this regard, our recent global gene microarray analysis has observed an implantation-speciﬁc expression proﬁle of CXCL14
mRNA during early pregnancy in mice, showing its higher levels at implantation sites compared to inter-implantation sites, implicating a
potential role of CXCL14 in the periimplantation events. In the present investigation, using Northern blot, in situ hybridization and
immunostaining, we further demonstrated that uterine CXCL14 expression was speciﬁcally induced at embryo implantation site and
expanded with subsequent decidualization process in a spatiotemporal manner. The implanting embryo also showed a highlighted
expression of CXCL14 in the blastocyst trophectoderm and its derived ectoplacental cones (EPCs) during postimplantation development.
In vitro functional study revealed that CXCL14 could signiﬁcantly inhibit both primary and secondary trophoblast attachment and
outgrowth, correlated with a stage-dependant downregulation of MMP-2 and/or MMP-9 activity. Moreover, it was found that biotinylated
CXCL14 could speciﬁcally bind to trophoblast cells in vitro and in vivo, suggesting trophoblast cell, perhaps expressing the unidentiﬁed
CXCL14 receptor, is a bioactive target of CXCL14. Collectively, our ﬁndings provide evidences supporting the contention that CXCL14
is an important paracrine/autocrine modulator regulating trophoblast outgrowth at the maternal–fetal interface during the process of
pregnancy establishment. This study is clinically related since CXCL14 is also highly expressed in human receptive endometrium and
J. Cell. Physiol. 221: 448–457, 2009. ß 2009 Wiley-Liss, Inc.
The initiation of implantation requires synchronized
development of the embryo to the blastocyst stage, and the
well-differentiated endometrium into a receptive state. As soon
as the blastocyst attachment reaction happens, the primary
trophoblast begin to invade through the epithelial layer of
uterine wall, followed by decidualization of uterine stroma and
further sophisticated interactions between invading
trophoblast and decidualizing cells towards successful
placentation (Wang and Dey, 2006). This well-controlled
invading processes between trophoblast and uterine cells are
subtly balanced by groups of positive and negative regulators
locally produced at the maternal–fetal interface. Among these
numerous regulators, the chemokine family, which is previously
well known by their role in leukocyte chemotaxis, has been
recently found to play important roles during early pregnancy,
such as governing the migration of trophoblast (Salamonsen
et al., 2007).
Recent global gene expression studies in human menstrual
cycle have shown that among numerous chemokine members,
the CXCL14, also known as BRAK (breast and kidney expressed
chemokine), has a surge up-regulation in the mid-secretory
phase of endometrium, which coincides the predicted time of
embryo implantation (Talbi et al., 2006). Similarly, our own
microarray data (Mouse 430 2.0, Affymetrix, Santa Clara, CA)
on day 5 of mouse pregnancy also revealed that the CXCL14 was
signiﬁcantly up-regulated in the implantation sites compared to
the inter-implantation sites. The similar expressional ﬁnding
in both human and mice species, to our knowledge, have
suggested an important role of CXCL14 during the window of
embryo implantation, and therefore become the interest of
CXCL14 belongs to a large family of the chemokines, which
were a group of structurally related molecules that regulate
activation and trafﬁcking of leukocytes via members of the
seven-transmembrane G protein-coupled receptors
(Fernandez and Lolis, 2002). CXCL14 was ﬁrst found in human
normal breast and kidney tissues, and further found to express
Haibin Kuang and Qi Chen contributed equally to this work.
Additional Supporting Information may be found in the online
version of this article.
Contract grant sponsor: The National Basic Research Program of
Contract grant numbers: 2006CB944006, 2006CB504006.
Contract grant sponsor: CAS Knowledge Innovation Program;
Contract grant number: KSCX2-YW-R-080.
Contract grant sponsor: National Natural Science Foundation of
Contract grant number: 30670785.
*Correspondence to: Enkui Duan, State Key Laboratory of
Reproductive Biology, Institute of Zoology, Chinese Academy of
Sciences, Beijing, P.R. China. E-mail: firstname.lastname@example.org
Received 27 February 2009; Accepted 11 June 2009
Published online in Wiley InterScience
(www.interscience.wiley.com.), 22 July 2009.
ß 2009 WILEY-LISS, INC.
in many other normal tissues, and also in some kinds of
malignant tissues (Hromas et al., 1999; Frederick et al., 2000;
Augsten et al., 2009; Pelicano et al., 2009). As a unique member
of chemokine family, CXCL14 is currently suggested as a
chemoattractant for dendritic cells (DC), activated monocytes,
neutrophils, and natural killer cells (NK) under different
pathophysiological conditions (Kurth et al., 2001; Shellenberger
et al., 2004; Shurin et al., 2005; Starnes et al., 2006; Salogni et al.,
2009). It has been also reported that CXCL14 was a potent
mediators of neoangiogenesis and tumor growth, invasion,
metastasis (Schwarze et al., 2005; Ozawa et al., 2006; Augsten
et al., 2009), which shares many similarities with the process of
trophoblast invasion during early pregnancy (Fitzgerald et al.,
2008). To date, the speciﬁc receptor of CXCL14 has not yet
been found. Recently, targeted deletion of CXCL14 in mice has
resulted in compromised fertility (Meuter et al., 2007 and
personal communication with authors). As the homozygous
pairs were reported to produce reduced litter size and some
knockout females did not produce newborns. However, the
underling mechanisms of this phenomenon are still unexplored.
The observed reproductive phenotype in CXCL14 null mice
coupled with its potential role in cell migration has encouraged
us to hypothesize that CXCL14 might play important roles in
trophoblast function during early pregnancy, since a lot of
pregnant failures were associated with abnormal trophoblast
invasion at periimplantation stage (Wilcox et al., 1999; Song
et al., 2002). In this investigation, we showed a site-strict and
cell-speciﬁc expression of CXCL14 within the implantation sites
at both maternal and embryonic compartment. Further
functional study revealed CXCL14 as an inhibitory factor
against trophoblast outgrowth through downregulating MMP-2
and 9. Also, binding assays of CXCL14 were performed at the
maternal–fetal interface to predict the potential site of
unidentiﬁed CXCL14 receptor.
Materials and Methods
Animals and treatments
Kunming white strain mice (Experimental Animal Center, Institute
of Genetics and Development, Chinese Academy of Sciences) were
maintained in the animal facility of the State Key Laboratory of
Reproductive Biology, Institute of Zoology, Chinese Academy of
Sciences. The Guidelines for the Care and Use of Animals in
Research were followed. Mice were allowed free access to water
and food with a constant photoperiod (12L:12D). Adult female
mice (25–30 g, 7–8 weeks old) in estrus were mated with fertile
males of the same strain at room temperature (258C). The
following morning of ﬁnding a vaginal plug was designated as day 1 of
DNA microarray analyses
Implantation (IS) and inter-implantation (IIS) sites were monitored
by an intravenous injection of a blue dye and divided by sharp
dissection at 0800–0900 h a.m. on day 5 (n ¼ 10 mice). Uterine
tissues were ﬂash frozen and stored at 808C. Total RNA was
extracted with Trizol reagent (Invitrogen, Eugene, OR). cDNA
and Biotinylated cRNA were prepared according to Affymetrix
protocols. Samples were hybridized overnight to high-density
Mouse Genome 430 2.0 array (Affymetrix), at the Shanghai Hujing
Biotech Co., Ltd (Shanghai, China). Then the chips were scanned
using a Scanner3000, and the data were extracted using the
Affymetrix GeneChip Operating Software version 1.2
(Affymetrix). Each sample was hybridized on two chips for reducing
the false positive rate.
Northern blot analyses
Total uterine RNAs were extracted with Trizol reagent
(Invitrogen), and were quantiﬁed by absorbance at 260 nm as well
as by ethidium bromide staining after electrophoresis through
agarose gels. Samples of total RNA (20 mg) were then blotted
overnight to Hybond-N membrane by electrical transfer. The
cDNA probe (bases 442–1058 of Mus musculus CXCL14:
GenBank Accession No. NM_019568) was labeled with
hybridization was carried out overnight at 608C. The blot was
subjected to autoradiography at 708C using BioMax MS ﬁlm
(Amersham, UK). To determine the relative amounts of RNA
transferred to the membrane, blots were stripped and hybridized
P-labeled 18S RNA probe.
RNA extraction and RT-PCR
Total RNA was extracted using Trizol Reagents and reverse
transcribed in 25 ml of reaction mixture containing 30 U avian
myeloblastosis virus reverse transcriptase (Promega, Madison,
WI). The PCR was conducted in a total volume of 50 ml for
30 cycles of denaturation at 948C of 30 sec, annealing at 568C for
30 sec, and extension at 728C for 45 sec, with a ﬁnal extension step
of 10 min at 728C. The primers used in this study include CXCL14
(5-GGG TCC AAG TGT AAG TGT TC-3; 5-GTA GTG CTG TGA
ACG GTC TC-3) and 18S (5-AAT CAG GGT TCG ATT CCG GA-
3; 5-CCA AGA TCC AAC TAC GAG CT-3). The ampliﬁed
products were analyzed by electrophoresis on 1% agarose gels
stained with ethidium bromide. Images of the RT-PCR agarose gels
were acquired with a High Performance CCD camera and
quantiﬁcation of the bands was performed by ChemiDocXRS
(BioRAD, Hercules, CA). Band intensity of the genes determined
was compared with the band intensity of 18S as an internal control,
and the relative level was acquired.
Preimplantation embryos at different developmental stages were
ﬂushed from the oviduct or uterus on days 2–4 of pregnancy. All
embryos were then ﬁxed in 4% paraformaldehyde (30 min, room
temperature). After ﬁxation, embryos were washed three times in
PBS and permeabilized in PBS containing 0.2% TritonX-100
(12 min, room temperature). After rinsing three times in PBS,
embryos were incubated in 5% BSA for 30 min at room
temperature to block nonspeciﬁc binding of the antibodies. Then,
embryos were incubated with anti-CXCL14mAb (R&D Systems,
Minneapolis, MN) at 48C overnight followed by secondary antibody
(ﬂuorescein isothiocyanate-labeled anti-rat IgG) for 1 h at 378C.
Nuclei were stained with 5 mg/ml of propidium iodide (Sigma,
St. Louis, MO) for 10 min. Embryos were viewed under a
laser-scanning confocal microscope (Leica, Heidelberg, Germany).
Rat preimmune IgG was used as a negative control.
In situ hybridization
In situ sense and antisense probe templates were prepared by
amplifying a 351bp fragment of CXCL14 cDNA (GenBank
accession no: NM_019568, nucleotides 334–684), using sense and
antisense primers modiﬁed with either T7 or SP6 sequences. The
resulting templates were then transcribed with the
digoxigenin-labeling kit (Roche Molecular Biomedicals, Mannheim,
Germany) according to the manufacturer’s protocol. After
deparafﬁnization and proteinase K treatment, the tissue sections
were prehybridized and then incubated in the hybridization buffer
containing 1 mg/ml of digoxigenin-labeled sense or antisense BRAK
cRNA probe overnight. After serial washing and RNase treatment,
samples were blocked and incubated with alkaline phosphatase-
conjugated antidigoxigenin antibody (Roche Molecular
Biochemicals; 1:2,000 dilution) in the blocking buffer at room
temperature for 1 h. After intense washing, signal was visualized
through the use of nitro blue tetrazolium and
5-bromo-4-chloro-3-indolyl-phosphate (Promega) until the color
developed to the desired extent.
JOURNAL OF CELLULAR PHYSIOLOGY
CXCL14 INHIBITS TROPHOBLAST OUTGROWTH
Blastocyst attachment and outgrowth assays in vitro
Blastocyst attachment and outgrowth assays were performed as
described previously (Qin et al., 2005). In brief, blastocysts were
obtained by ﬂushing the uterine horns with Ham’s F-12 medium on
day 4 morning of pregnancy and transferred in 96-well plates
precoated with ﬁbronectin (10 mg/ml; Sigma), containing Ham’s
F-12 (supplemented with 0.4% BSA) plus 0, 1, 10, 50, 100, 200,
1,000 ng/ml of recombinant mouse CXCL14 (rmCXCL14) protein
(R&D Systems). It was determined that the recombinant CXCL14 at
the concentration of 100 ng/ml has an optimal effect for blastocyst
attachment and outgrowth at 48 h after initiation of culture.
The blastocysts were then divided into four treatment groups:
the medium only (control group, n ¼ 40, four wells), the medium
plus 100 ng/ml of rmCXCL14 (rmCXCL14 group, n ¼ 40, four
wells), the medium plus 30 mg/ml of rat anti-mouse CXCL14 IgG
(anti-CXCL14 group, n ¼ 40, four wells), and medium plus 30 mg/
ml rat IgG (IgG group, n ¼ 40, four wells). When primary giant
trophoblast cells were visible around the attachment site of the
attached blastocysts, we designated the blastocysts as outgrowth.
Blastocyst attachment was examined at 48 h of culture. Blastocyst
attachment was only examined once, because gentle pipetting was
required to determine whether the embryo would detach from the
bottom of plates. The ratios of blastocysts with attachment and
outgrowth relative to the total number of embryos were
Outgrowth assays of EPCs
Ectoplacental cones (EPCs) from the mice on day 8 of pregnancy
were dissected out under sterile conditions and transferred to
4-well (30 EPCs per well) plates precoated with Matrigel (8 mg/ml;
BD). The EPCs were cultured in Ham’s F-12 (supplemented with
0.4% BSA). Recombinant mouse CXCL14 (100 ng/ml), rat
anti-mouse CXCL14 IgG (30 mg/ml), or normal rat IgG was
supplemented to freshly isolated EPCs, respectively. Outgrowth
assays in vitro were performed according to the methods
established by our group (Dai et al., 2003; Liu et al., 2008). When
secondary trophoblast giant cells (sTGCs) were visible around the
attachment site of the attached blastocysts, we designated the
blastocysts as outgrowth. The percentage of EPCs outgrowth was
evaluated with outgrown EPCs to the number of total EPCs. The
outgrowth area, which is occupied by sTGC, was recorded by
inverted microscopy and measured with NIH ImageJ software
Detection of CXCL14 binding sites
Recombinant CXCL14 protein was biotinylated by EZ-link
biotinylation reagent (Pierce). Proteins were incubated with 1 ml of
0.3 mg/ml EZ-link biotinylation reagent in PBS on ice for 2 h. After
incubation, samples were dialyzed to remove the free biotin and
measured their protein concentrations. Trophoblast giant cells
from EPCs and frozen tissues sectioned were ﬁxed in 4%
paraformaldehyde (30 min, room temperature). Sample were
blocked with Block Ace at room temperature for 1 h and then
incubated with biotinylated protein (100 mg/ml) at room
temperature for 1 h. After incubation, the samples were incubated
with avidin-FITC (1:1,000, Santa Cruz Biotechnology, Inc.,
Santa Cruz, CA) at room temperature for 1 h. Nuclei were
counterstained with 5 m g/ml of propidium iodide (Sigma) for
10 min. Samples were viewed under a ﬂuorescence inverted
microscope (Nikon, Tokyo, Japan). Nonlabeled CXCL14 and
biotin were used as negative controls.
The conditioned medium of blastocysts and EPCs culture was
collected at 24 and 48 h. Protein content of conditioned media was
measured according to the method of Bradford, and 250 ng protein
mixed with 4 sample buffer (8% SDS (w/v), 0.04% Bromophenol
Blue (w/v), 40% glycerol (v/v), and 0.25M Tris) and then subjected
to electrophoresis in a 10% polyacrylamine gel containing 0.5 mg/ml
gelatin (Sigma). The gel was washed in 2.5% Triton X-100 and
50 mM Tris/HCl, at pH 7.5 for 1 h to remove the SDS and incubated
for 16–18 h in calcium assay buffer (50 mM Tris, 200 mM NaCl, and
10 mM CaCl
, pH 7.5) at 378C. After staining with 0.2% Coomassie
Brilliant Blue R250 in 50% methanol, and 10% acetic acid, the gel
was destained with 10% acetic acid. The lytic bands were quantiﬁed
by computer-aided densitometry. Each experiment was repeated
All experimental treatments were carried out in triplicate and each
experiment was repeated at least three times. Values were
presented as means SEM. Statistical analysis was performed using
one-way ANOVA followed by LSD’s post hoc test, and a value of
P < 0.05 was considered to be statistically signiﬁcant. All statistical
analyses were performed using SPSS 11.5.
CXCL14 expression is higher at implantation site
with the onset of embryo implantation and
Our whole-genome microarray study comparing mRNAs from
mouse implantation sites and inter-implantation sites on day 5
of pregnancy has shown that the expression intensity of
CXCL14 mRNA in implantation sites was 4.1-fold higher
compared with inter-implantation sites (Fig. 1A). This result
was further conﬁrmed by Northern blot analysis as showed in
Figure 1B. The changing proﬁles of CXCL14 mRNA during
periimplantation period were next performed by RT-PCR as
showed in Figure 1C,D. From D1 to D4 of pregnancy, the
CXCL14 expression was generally low or undetectable, the
slightly increased expression observed on D1 and D4 may
reﬂect an estrogen regulation of this gene (proestrous estrogen
on D1 and preimplantation estrogen surge from corpus luteum
on D4), as demonstrated in Supplementary Figure 1. In contrast
with D1–D4, the expression level of CXCL14 dramatically
increased with the initiation of implantation from D5 onward,
but this increased expression were only restricted within the
implantation sites, while the level of inter-implantation sites
remained low or undetectable (Fig. 1C,D).
CXCL14 spatiotemporally localized on both mouse
uterus and embryo in a cell-speciﬁc manner
The dynamic changing level of CXCL14 has next led us checking
the site-speciﬁc localization of this gene during mouse
periimplantation. By using in situ hybridization, here we found
that from D1 to D4 of pregnancy, CXCL14 were located in the
epithelial (both luminal and glandular) and subepithelial region
with low or undetectable level in accordance with RT-PCR
results (Fig. 2A–C). While on day 5, the expression of CXCL14
signiﬁcantly increased at the primary decidualization zone
(PDZ) surrounding the site of embryo implantation. This
expression pattern in the decidua, however, ceased to grow
with the expanding of decidualization, and began to decrease
from D6 onward, then disappeared from the PDZ on D7 and
D8 while remained in outer region of PDZ, which is termed as
secondary decidualization zone (SDZ; Fig. 2D–G). During D7
and D8, an increasing level of CXCL14 was also observed at the
mesometrial part of decidualizing stroma as shown in
Figure 2F,G, a site considered for future placenta formation.
Notablely, through D5–D8 of pregnancy (Fig. 2D–G), the
growing embryo also showed an evidenced expression of
CXCL14, with particularly strong signal in the region of EPCs on
D7 and D8 (Fig. 2F,G), which is a major source of trophoblast
JOURNAL OF CELLULAR PHYSIOLOGY
450 KUANG ET AL.
cells within the implantation sites. This ﬁnding has inspired us
further examine whether CXCL14 was also expressed in
preimplantation embryos. And indeed, our
immunoﬂuorescence staining showed that the CXCL14 were
expressed in the D4 blastocyst with strong signal in the
trophectoderm (Fig. 3B), whereas the morula showed very
weak or undetectable signals (Fig. 3A), suggesting a potential
role of CXCL14 in trophoblast function.
CXCL14 inhibits trophoblast outgrowth in vitro
Given the abundant expression of CXCL14 on trophectoderm
and trophoblast of periimplantation embryo, and several
reports showing CXCL14 as a tumor migration regulator
(Schwarze et al., 2005; Ozawa et al., 2006; Pelicano et al., 2009),
here we want to address the question as to whether CXCL14
participated in regulating trophoblast invasion during embryo
implantation, a process similar to tumor invasion in many
aspects. Here an in vitro culture system was utilized to assess
the ability of blastocyst attachment and outgrowth in the
presence of rmCXCL14 or CXCL14 antibody. Blastocyst
attachment status was checked at 48 h after initiation of in vitro
culture. As shown in Figure 4A, compared with the control
groups, the attachment rate of blastocysts onto the ﬁbronectin
substrate were signiﬁcantly inhibited by rmCXCL14 while
promoted by CXCL14 antibody after 48 h of culture ( P < 0.01).
There’s no signiﬁcant differences observed between control
and rat IgG group. The extent of blastocyst outgrowth was
Fig. 2. Localization of CXCL14 mRNA in mouse uteri during periimplantation. In situ hybridization shows localization of CXCL14 mRNA as
blue-purple precipitates. Uterine cross-sections from days 1(A), 3(B), 4(C), 5(D), 6(E), 7(F), and 8(G) of pregnancy were subjected to in situ
hybridization using a digoxigenin-labeled CXCL14 antisense cRNAprobes. H: Negative control: day 6 of pregnancy, sense probes of CXCL14 were
used. Similar results were obtained in 2–3 mice for each checkpoint. GE, glandular epithelium; LE, luminal epithelium; S, stroma; EM, embryo; AM,
antimesometrial pole; M, mesometrial pole; PDZ, primarydecidualzone; SDZ, secondary decidual zone; DE, decidua.Originalmagniﬁcation, 40T.
Fig. 1. Levels of CXCL14 mRNA spatiotemporally expressed in periimplantation mouse uteri. A: Relative expression intensity of CXCL14 mRNA
was analyzed by DNA microarray comparing implantation site (IS) and inter-implantation site (IIS) of day 5 pregnancy mouse uterus (n U 10).
Values for expression intensity were derived from integration of hybridization signals. B: Northern blot conﬁrmation of DNA microarray results.
Autoradiography of a membrane probed sequentially with a
P-labeled cDNA probe for CXCL14 and 18S rRNA. C: RT-PCR analysis of CXCL14
mRNA in mouse uteri (n U 3 per day) from days 1 to 7 of pregnancy. D: Optical indensity of CXCL14 mRNA (CXCL14/18S mRNA) during
periimplantation uteri. The results (mean W SEM, n U 3) were calculated from values of CXCL14 mRNA relative to that of 18S mRNA (internal
JOURNAL OF CELLULAR PHYSIOLOGY
CXCL14 INHIBITS TROPHOBLAST OUTGROWTH
checked at three time points (24, 36, and 48 h). As shown in
Figure 4B, after 36 and 48 h in vitro culture, an inhibitory effect
on blastocyst outgrowth was evidently observed in
rmCXCL14-treated groups, while a promontory effect was
observed at 48 h when cultured with CXCL14 antibody
( P < 0.01). We have further performed Ki67 staining of
blastocysts after cultured with control medium and medium
plus CXCL14, as to examine whether CXCL14 affect the
proliferation ability of the blastocyst. As shown in
Supplementary Figure 3. CXCL14 (100 and 200 ng/ml)
treatment group does not show obvious differences in regards
to ki67 staining, also, the blastocysts seems morphologically
normal compared with control group. These results suggested
that the reduced outgrowth performance after CXCL14
treatment is not due to reduced proliferation.
To further study the role of CXCL14 during trophoblast cell
outgrowth and migration, we isolated EPCs from day 8 mouse
pregnant uteri, a structure predominantly composed of invasive
sTGCs derived from blastocyst trophectoderm (Liu et al.,
2008), EPCs were cultured in the presence of either
rmCXCL14, CXCL14 antibody, or normal rat IgG. The results
showed that the trophoblastic outgrowth capacity were
signiﬁcantly reduced in the CXCL14-treated group, as
demonstrated by the outgrowth rate and outgrowth area of
EPCs, while the CXCL14 antibody shows an opposite effect as
shown in Figure 5A–C. These results demonstrated that
CXCL14 has an inhibitory role on both primary and secondary
CXCL14 inhibits trophoblast outgrowth via MMP-2
and 9 downregulation in vitro
To shed additional light on the mechanisms by which CXCL14
regulates trophoblast outgrowth, we next examined whether
CXCL14 could regulate MMP-2 and 9 production of EPCs,
which have been considered as two key factors regulating
trophoblast invasion (Staun-Ram and Shalev, 2005). After
24 and 48 h of in vitro culture, EPCs-conditioned medium was
collected and analyzed for collagenase activity by gelatin
zymography. At 24 h checkpoint, no signiﬁcant differences were
Fig. 3. Immunoﬂuorescence detection of CXCL14 in mouse preimplantation embryos. Green signal represents CXCL14 staining with FITC-
conjugated secondary antibody and red signal indicates nuclear staining with PI. A: Morula. B: Blastocyst. Note the highlighted expression of
rat IgG. Scale bars, 50 mm.
JOURNAL OF CELLULAR PHYSIOLOGY
452 KUANG ET AL.
observed between three groups (data not shown). While after
48 h culture, MMP-2 and 9 levels were signiﬁcantly suppressed
in CXCL14-treated group and promoted in CXCL14
antibody-treated group ( P < 0.05 and P < 0.01; Fig. 6A–C).
These results have demonstrated a direct or indirect effect of
CXCL14 on regulating trophoblast MMPs production during
the process of trophoblast outgrowth.
CXCL14 could speciﬁcally bind to trophoblast cells
both in vitro and in vivo
Since CXCL14’s-speciﬁc receptor(s) has not been identiﬁed to
date, binding assays were performed to test whether the
trophoblast cell is the direct target of secreted CXCL14 and to
explore the binding site of biotinylated CXCL14. As shown in
Figure 7A–C, the biotinylated CXCL14 uniformly stained the
trophoblast giant cells grown out from EPCs in culture, while
biotin and no labeled CXCL14 group did not show positive
signals. Furthermore, in vivo binding assay of biotinylated
CXCL14 was performed on frozen tissues section of day 8
pregnant mouse uteri. As demonstrated in Figure 7D, the
biotinylated CXCL14 could speciﬁcally bind to the EPC region
of D8 implantation site, while the surrounding decidual cells
shows no or very weak binding signal.
The current study has established the spatialtemporal
expression pattern of CXCL14 during mouse periimplantation.
The cell-speciﬁc expression of CXCL14 at implantation sites on
both uterine and embryonic compartment has led us into two
lines of ﬁndings. At one side, our in vitro functional study
demonstrated that recombinant CXCL14-inhibited
trophoblast cell attachment and outgrowth while the CXCL14
antibody had an opposite effects. This inhibitory effect of
CXCL14 was further found to associate with downregulation of
trophoblast MMP-2 and 9 activity. On the other side, by using
biotinylated recombinant CXCL14 binding experiment, we ﬁnd
that at the maternal–fetal interface, the major binding site of
secreted CXCL14 was trophoblast cells derived from
blastocyst, but not the stroma cells from the uterus, indicating
that the trophoblast cells were the bioactive target of CXCL14
in the context of early pregnancy, and would express the
unidentiﬁed CXCL14 receptor(s).
In early pregnant uterus, the process of implantation requires
a highly coordinated and complex dialogue between the
embryo and maternal tissue, numerous locally produced
modulators and signaling events are involved during this
reciprocal interaction. Recent evidences have shown that,
several members of the chemokine family could regulate
trophoblast migration in a way similar to that of leukocyte
recruitment (Hannan and Salamonsen, 2007), which exerts
their function via binding to the cognate receptors within the
context of maternal–fetal interface. To our best knowledge,
most of the currently studied chemokines with a regulatory
role during early pregnancy shows a promontory effect on
trophoblast migration. However, our results showed that
CXCL14 played an evidenced inhibitory role on trophoblast
attachment and outgrowth, this unique feature of CXCL14 on
trophoblast migration were also in accordance with several
reports that CXCL14 have a tumor suppressive function
(Schwarze et al., 2005; Ozawa et al., 2006). Furthermore, our
parallel study in human pregnancy also showed an inhibitory
effect of CXCL14 when culturing human trophoblast in vitro
(unpublished data in our lab). These data has suggested a unique
role of CXCL14 in balancing the invasive ability of trophoblast
against other promoting chemokines. Since the receptor of
CXCL14 has not been found to date by testing several cognate
receptors of the chemokine family (CXCR1, 2, 3, 4) (Cao et al.,
2000; Sleeman et al., 2000), it is possible that CXCL14 might has
an unique receptor beyond the currently known receptors of
chemokine family (Hannan and Salamonsen, 2007; Hess et al.,
2007). In this regard, using in vitro binding assay to predict the
potential site of CXCL14 receptor would be sound and ensures
further identiﬁcation of its true face. Here our results showed
that the biotinylated CXCL14 uniformly stained sTGCs of EPCs
in vitro and in vivo, demonstrating the functional binding sites of
its potential receptor(s). These results has also strengthen our
belief that CXCL14 secreted form both the trophoblast and
surrounding uterine tissue could directly act on trophoblast cell
in a paracrine/autocrine manner. However, it should also be
mentioned that there are also studies demonstrating that
CXCL14 promote cell migration in other speciﬁc systems
(Augsten et al., 2009; Pelicano et al., 2009) Implicating that
CXCL14 might have potentially diverse roles under different
During early pregnancy, trophoblast invasion are closely
correlative with the expression of MMPs (Staun-Ram and
Shalev, 2005; Cohen and Bischof, 2007; Ferretti et al., 2007),
which are capable of degrading extra cellular matrix. According
to the existing literature, at the stage of early pregnancy, MMP-2
and 9 were the major participants during embryo implantation,
which due to their ability to degrade Collagen IV, the main
component of the basement membrane, thus, enabling the
Fig. 4. CXCL14-inhibited attachment and outgrowth of blastocyst
in vitro. Blastocyst attachment and outgrowth assays were
performed in 96-well plates precoated with 10 mg/ml of FN. A: Rate of
blastocyst attachment. Recombinant mouse CXCL14 (rmCXCL14)
added exogenously to the culture media-inhibited blastocyst
attachment at 48 h culture (top), whereas blastocyst attachment was
promoted by the treatment with anti-mouse CXCL14 antibody
(anti-CXCL14) (bottom). B: Rate of blastocyst outgrowth. Rate of
blastocyst outgrowth was reduced with the use of rmCXCL14 at
36 and 48 h after initiation of culture (top), whereas outgrowth
increased by the treatment with anti-CXCL14 antibody (bottom).
Blastocysts incubated in normal cultural media were served as
controls. Results were shown as mean W SEM (n U 4).
P < 0.01
compared with the control group.
JOURNAL OF CELLULAR PHYSIOLOGY
CXCL14 INHIBITS TROPHOBLAST OUTGROWTH
invasion of the trophoblast cells through the decidua and into
the maternal vasculature (Staun-Ram and Shalev, 2005). Since
CXCL14 could inhibit migration of both blastocyst (rich in
primary trophoblast giant cells) and EPCs outgrowth (rich in
sTGCs), which were previously reported to have different
predominance in MMP-2 and 9 content (Hulboy et al., 1997).
We tested the effect of CXCL14 on MMP-2, and 9 production
in both targets. Our data indicated that CXCL14 could
downregulate MMP-2 in the blastocyst (data not shown) while
downregulate both MMP2, and 9 production in the EPCs. These
results are in correspondance with previously reported
relationship between chemokine and MMPs secretion in other
systems (Cross and Woodroofe, 1999; Van and Libert, 2007).
The relative abundance of MMP-2, and 9 in blastocyst and EPCs
Fig. 5. Effect of CXCL14 on EPCs outgrowth in vitro. Outgrowth assays of EPCs were performed in 4-well plates precoated with Matrigel
(8 mg/ml). A: Outgrowth percentage of EPCs. rmCXCL14 (top) and anti-CXCL14 antibody (bottom) were added into the culture medium and
cultured with EPCs for 24, 36, and 48 h, and outgrowth percentage of EPCs was recorded at each checkpoint. B: Outgrowth area of EPCs.
Outgrowth area of EPCs was reduced with the treatment of rmCXCL14 at 48 h culture (top), whereas was promoted by the treatment with anti-
CXCL14 (bottom). Data represented as fold change in the outgrowth area of EPCs (WSEM, n U 30) compared to controls.
P < 0.05,
P < 0.01
compared with the control group. C: Demonstrative photos showed morphologic observation of EPCs in the presence of rmCXCL14 or
anti-CXCL14 antibody or rat IgG. Photos were taken under the light microscope after 24 and 48 h of culture. Original magniﬁcation, 40T.
JOURNAL OF CELLULAR PHYSIOLOGY
454 KUANG ET AL.
also support previous ﬁndings that MMP-2 is the main gelatinase
in primary trophoblast giant cells while MMP-9 is the main
gelatinase in sTGCs (Staun-Ram and Shalev, 2005). The
underlining mechanisms of how CXCL14 achieve this effect still
kept unknown, mainly due to the lack of knowledge on its
receptor and subsequent signaling pathway. Future
identiﬁcation of CXCL14 receptor may further facilitate
understanding of these detailed processes.
In addition to the evidenced role of CXCL14 on trophoblast
migration, the highly uterine expression of CXCL14 at embryo
implantation on D5 might also suggest that CXCL14 could
contribute to the establishment of PDZ at this time. While from
D7 to D8, when the SDZ were establishing in the outer region
of PDZ, the CXCL14 again showed a timely increased
expression at these parts of decidua (Fig. 2). Interestingly, this
expression pattern of CXCL14 is somewhat similar with
another immunoregulator gene PTX3, and the PTX3 null female
mice do show a compromised implantation and decidualization
(Tranguch et al., 2007). Similar expression pattern of CXCL14 in
the decidualizing stroma was also observed in the oil-induced
deciduoma from D5 to D8 (Supplementary Fig. 2), suggesting its
close relationship with uterine decidualization. Alternatively,
the uterine expression of CXCL14 might also play a role in uNK
cell recruitment thus regulate early pregnancy (Hanna et al.,
2006; Carlino et al., 2008), because there have been reports
about CXCL14’s role on NK cell recruitment (Starnes et al.,
2006). Our lab also found an overlapped distribution of
CXCL14 and uNK cells in the uterus section from days 7 to
9 postimplantation (data not shown). However, a clear map of
CXCL14’s effects on uNK cell recruitment in the context of
uterus were still vague, since a recent report has showed
opposite results against its chemoattractant roles on many
types of leukocytes based on knockout models (Meuter et al.,
To date, the CXCL14 knockout mice have been generated
by two independent labs. Although a comprehensive analysis of
reproductive phenotype has not been performed, both labs
has reported different degree of subfertility phenotype of
CXCL14 null mice (Meuter et al., 2007; Nara et al., 2007). It
should be noted that a reduced Mendelian frequency of
mice was reported by both labs when
intercrossing heterozygous mice. And one lab has reported
that the littersize of a knockout knockout breeding pair was
generally small while some knockout females could not
produce newborns. [personal communication with the
authors], this phenotype is particularly interesting to us,
because our current result has suggested that the deletion of
CXCL14 might result in enhanced abnormal trophoblast
invasion and cause overdestruction of newly formed decidual
cells, which further hamper the process of placentation and
subsequent pregnancy. However, detailed investigation in
regards to when and how the deletion of CXCL14 cause an
compromised fertility were further needed by using CXCL14
In conclusion, the current study has for the ﬁrst time showed
a map of expressional proﬁles of CXCL14 during mouse
Fig. 6. CXCL14-inhibited MMP-2 and 9 production in cultured EPCs. A: Representative gelatin zymography showed the relative secretion of
MMP-2 (72-kDa) and MMP-9 (92-kDa) in EPCs culture medium with treatment of rmCXCL14, anti-CXCL14 antibody and rat IgG at 48 h culture.
B,C: Relative densitometric analysis showed that the MMP-2 and 9 production were inhibited in rmCXCL14-treated group (B), while promoted in
anti-CXCL14-treated group (C). Resultsrepresented asfoldchangeintheproductionofcollagenase(WSEM,n U 3) compared to controls.
P < 0.05,
P < 0.01 compared with the control group.
JOURNAL OF CELLULAR PHYSIOLOGY
CXCL14 INHIBITS TROPHOBLAST OUTGROWTH
periimplantation. It is demonstrated that the CXCL14 plays an
inhibitory role on trophoblast outgrowth during early
pregnancy in a paracrine/autocrine manner. Our data has
provided evidence that different member of chemokine family
could play opposing effect within the context of maternal–fetal
interface, ensuring an optimized environment for the
well-controlled process of trophoblast invasion. This study also
provided future possibilities to identify the currently unknown
CXCL14 receptor(s) and warrants future clinical research on
CXCL14, because CXCL14 is also speciﬁcally expressed in
Fig. 7. Biotinylated CXCL14 speciﬁcally bound to mouse trophoblast of EPCs in vitro and in vivo. Recombinant CXCL14 proteins were
biotinylated using the EZ-link biotinylation reagent. Biotinylated CXCL14 (A), no labeled CXCL14 (B), and biotin (C) were incubated with
the trophoblast cells grown out from EPCs for 1 h and visualized with streptavidin-FITC and PI for nuclear counterstaining. Biotinylated CXCL14
(D) and biotin (E) was incubated with frozen tissue sections of day 8 implantation site. Note the EPCs region showed strong binding signal of
biotinylated CXCL14 while the surrounding decidual cells showedweakornosignal. AM, anti-mesometrial pole; DE, decidua; M,mesometrialpole,
Scale bar, 100 mM.
JOURNAL OF CELLULAR PHYSIOLOGY
456 KUANG ET AL.
human endometrium and trophoblast (Red-Horse et al., 2001,
2004; Talbi et al., 2006).
We thank Dr. Haibin Wang at Institute of Zoology, Chinese
Academy of Sciences for proofreading of this manuscript.
Augsten M, Hagglof C, Olsson E, Stolz C, Tsagozis P, Levchenko T, Frederick MJ, Borg A,
Micke P, Egevad L, Ostman A. 2009. CXCL14 is an autocrine growth factor for ﬁbroblasts
and acts as a multi-modal stimulator of prostate tumor growth. Proc Natl Acad Sci USA
Cao X, Zhang W, Wan T, He L, Chen T, Yuan Z, Ma S, Yu Y, Chen G. 2000. Molecular cloning and
characterization of a novel CXC chemokine macrophage inﬂammatory protein-2 gamma
chemoattractant for human neutrophils and dendritic cells. J Immunol 165:2588–2595.
Carlino C, Stabile H, Morrone S, Bulla R, Soriani A, Agostinis C, Bossi F, Mocci C, Sarazani F,
Tedesco F, Santoni A, Gismondi A. 2008. Recruitment of circulating NK cells through
decidual tissues: A possible mechanism controlling NK cell accumulation in the uterus
during early pregnancy. Blood 114:3108–3115.
Cohen M, Bischof P. 2007. Factors regulating trophoblast invasion. Gynecol Obstet Invest
Cross AK, Woodroofe MN. 1999. Chemokine modulation of matrix metalloproteinase and
TIMP production in adult rat brain microglia and a human microglial cell line in vitro. Glia
Dai B, Cao Y, Liu W, Li S, Yang Y, Chen D, Duan E. 2003. Dual roles of progesterone in
embryo implantation in mouse. Endocrine 21:123–132.
Fernandez EJ, Lolis E. 2002. Structure, function, and inhibition of chemokines. Annu Rev
Pharmacol Toxicol 42:469–499.
Ferretti C, Bruni L, Dangles-Marie V, Pecking AP, Bellet D. 2007. Molecular circuits shared by
placental and cancer cells, and their implications in the proliferative, invasive and migratory
capacities of trophoblasts. Hum Reprod Update 13:121–141.
Fitzgerald JS, Poehlmann TG, Schleussner E, Markert UR. 2008. Trophoblast invasion: The
role of intracellular cytokine signalling via signal transducer and activator of transcription 3
(STAT3). Hum Reprod Update 14:335–344.
Frederick MJ, Henderson Y, Xu X, Deavers MT, Sahin AA, Wu H, Lewis DE, El-Naggar AK,
Clayman GL. 2000. In vivo expression of the novel CXC chemokine BRAK in normal and
cancerous human tissue. Am J Pathol 156:1937–1950.
Hanna J, Goldman-Wohl D, Hamani Y, Avraham I, Greenﬁeld C, Natanson-Yaron S, Prus D,
Cohen-Daniel L, Arnon TI, Manaster I, Gazit R, Yutkin V, Benharroch D, Porgador A,
Keshet E, Yagel S, Mandelboim O. 2006. Decidual NK cells regulate key developmental
processes at the human fetal-maternal interface. Nat Med 12:1065–1074.
Hannan NJ, Salamonsen LA. 2007. Role of chemokines in the endometrium and in embryo
implantation. Curr Opin Obstet Gynecol 19:266–272.
Hess AP, Hamilton AE, Talbi S, Dosiou C, Nyegaard M, Nayak N, Genbecev-Krtolica O,
Mavrogianis P, Ferrer K, Kruessel J, Fazleabas AT, Fisher SJ, Giudice LC. 2007. Decidual
stromal cell response to paracrine signals from the trophoblast: Ampliﬁcation of immune
and angiogenic modulators. Biol Reprod 76:102–117.
Hromas R, Broxmeyer HE, Kim C, Nakshatri H, Christopherson K II, Azam M, Hou YH. 1999.
Cloning of BRAK, a novel divergent CXC chemokine preferentially expressed in normal
versus malignant cells. Biochem Biophys Res Commun 255:703–706.
Hulboy DL, Rudolph LA, Matrisian LM. 1997. Matrix metalloproteinases as mediators of
reproductive function. Mol Hum Reprod 3:27–45.
Kurth I, Willimann K, Schaerli P, Hunziker T, Clark-Lewis I, Moser B. 2001. Monocyte
selectivity and tissue localization suggests a role for breast and kidney-expressed
chemokine (BRAK) in macrophage development. J Exp Med 194:855–861.
Liu Q, Zhang B, Zhao X, Zhang Y, Liu Y, Yan X. 2008. Blockade of adhesion molecule CD146
causes pregnancy failure in mice. J Cell Physiol 215:621–626.
Meuter S, Schaerli P, Roos RS, Brandau O, Bosl MR, von AU, Moser B. 2007. Murine CXCL14
is dispensable for dendritic cell function and localization within peripheral tissues. Mol Cell
Nara N, Nakayama Y, Okamoto S, Tamura H, Kiyono M, Muraoka M, Tanaka K, Taya C,
Shitara H, Ishii R, Yonekawa H, Minokoshi Y, Hara T. 2007. Disruption of CXC motif
chemokine ligand-14 in mice ameliorates obesity-induced insulin resistance. J Biol Chem
Ozawa S, Kato Y, Komori R, Maehata Y, Kubota E, Hata R. 2006. BRAK/CXCL14 expression
suppresses tumor growth in vivo in human oral carcinoma cells. Biochem Biophys Res
Pelicano H, Lu W, Zhou Y, Zhang W, Chen Z, Hu Y, Huang P. 2009. Mitochondrial
dysfunction and reactive oxygen species imbalance promote breast cancer cell motility
through a CXCL14-mediated mechanism. Cancer Res 69:2375–2383.
Qin J, Diaz-Cueto L, Schwarze JE, Takahashi Y, Imai M, Isuzugawa K, Yamamoto S, Chang KT,
Gerton GL, Imakawa K. 2005. Effects of progranulin on blastocyst hatching and subsequent
adhesion and outgrowth in the mouse. Biol Reprod 73:434–442.
Red-Horse K, Drake PM, Gunn MD, Fisher SJ. 2001. Chemokine ligand and receptor
expression in the pregnant uterus: Reciprocal patterns in complementary cell subsets
suggest functional roles. Am J Pathol 159:2199–2213.
Red-Horse K, Drake PM, Fisher SJ. 2004. Human pregnancy: The role of chemokine networks
at the fetal-maternal interface. Expert Rev Mol Med 6:1–14.
Salamonsen LA, Hannan NJ, Dimitriadis E. 2007. Cytokines and chemokines during human
embryo implantation: Roles in implantation and early placentation. Semin Reprod Med
Salogni L, Musso T, Bosisio D, Mirolo M, Jala VR, Haribabu B, Locati M, Sozzani S. 2009. Activin
A induces dendritic cell migration through the polarized release of CXCL12 and CXCL14.
Schwarze SR, Luo J, Isaacs WB, Jarrard DF. 2005. Modulation of CXCL14 (BRAK) expression
in prostate cancer. Prostate 64:67–74.
Shellenberger TD, Wang M, Gujrati M, Jayakumar A, Strieter RM, Burdick MD, Ioannides CG,
Efferson CL, El-Naggar AK, Roberts D, Clayman GL, Frederick MJ. 2004. BRAK/CXCL14 is
a potent inhibitor of angiogenesis and a chemotactic factor for immature dendritic cells.
Cancer Res 64:8262–8270.
Shurin GV, Ferris RL, Tourkova IL, Perez L, Lokshin A, Balkir L, Collins B, Chatta GS, Shurin
MR. 2005. Loss of new chemokine CXCL14 in tumor tissue is associated with low
inﬁltration by dendritic cells (DC), while restoration of human CXCL14 expression
in tumor cells causes attraction of DC both in vitro and in vivo. J Immunol 174:5490–
Sleeman MA, Fraser JK, Murison JG, Kelly SL, Prestidge RL, Palmer DJ, Watson JD, Kumble
KD. 2000. B cell- and monocyte-activating chemokine (BMAC), a novel non-ELR alpha-
chemokine. Int Immunol 12:677–689.
Song H, Lim H, Paria BC, Matsumoto H, Swift LL, Morrow J, Bonventre JV, Dey SK. 2002.
Cytosolic phospholipase A2alpha is crucial [correction of A2alpha deﬁciency is crucial] for
‘on-time’ embryo implantation that directs subsequent development. Development
Starnes T, Rasila KK, Robertson MJ, Brahmi Z, Dahl R, Christopherson K, Hromas R. 2006.
The chemokine CXCL14 (BRAK) stimulates activated NK cell migration: Implications for
the downregulation of CXCL14 in malignancy. Exp Hematol 34:1101–1105.
Staun-Ram E, Shalev E. 2005. Human trophoblast function during the implantation process.
Reprod Biol Endocrinol 3:56.
Talbi S, Hamilton AE, Vo KC, Tulac S, Overgaard MT, Dosiou C, Le SN, Nezhat CN, Kempson
R, Lessey BA, Nayak NR, Giudice LC. 2006. Molecular phenotyping of human
endometrium distinguishes menstrual cycle phases and underlying biological processes in
normo-ovulatory women. Endocrinology 147:1097–1121.
Tranguch S, Chakrabarty A, Guo Y, Wang H, Dey SK. 2007. Maternal pentraxin 3 deﬁciency
compromises implantation in mice. Biol Reprod 77:425–432.
Van LP, Libert C. 2007. Chemokine and cytokine processing by matrix metalloproteinases
and its effect on leukocyte migration and inﬂammation. J Leukoc Biol 82:1375–1381.
Wang H, Dey SK. 2006. Roadmap to embryo implantation: Clues from mouse models. Nat
Rev Genet 7:185–199.
Wilcox AJ, Baird DD, Weinberg CR. 1999. Time of implantation of the conceptus and loss of
pregnancy. N Engl J Med 340:1796–1799.
JOURNAL OF CELLULAR PHYSIOLOGY
CXCL14 INHIBITS TROPHOBLAST OUTGROWTH