Homeobox Gene HLX Is a Regulator of HGF/c-met-Mediated Migration of Human Trophoblast-Derived Cell Lines1

BIOLOGY OF REPRODUCTION 83, 676–683 (2010)
Published online before print 16 June 2010.
DOI 10.1095/biolreprod.109.078634


Homeobox Gene HLX Is a Regulator of HGF/c-met-Mediated Migration of Human
Trophoblast-Derived Cell Lines1
Gayathri Rajaraman,2,3,4 Padma Murthi,3,4 Shaun P. Brennecke,3,4 and Bill Kalionis3,4

Department of Obstetrics and Gynaecology,3 University of Melbourne, and Pregnancy Research Centre,4
Department of Perinatal Medicine, The Royal Women’s Hospital, Parkville, Victoria, Australia


ABSTRACT                                                                        INTRODUCTION
   Homeobox gene transcription factors play a critical role in                     Trophoblast cells of the placenta are critical for normal
normal placental development and are expressed in specialized                   placental development and a healthy pregnancy outcome.
trophoblast cells. Abnormal trophoblast function is associated                  Abnormal trophoblast cell functions are associated with
with clinically significant pregnancy disorders, including fetal                clinically significant pregnancy disorders such as fetal growth
growth restriction (FGR). Our previous studies demonstrated                     restriction (FGR) and preeclampsia [1, 2]. Trophoblast
that homeobox gene HLX is expressed in proliferating and                        functions are modulated in an autocrine/paracrine manner by
migrating (but not invading) human trophoblast cells and that                   growth factors, their binding proteins, and extracellular matrix
HLX expression is significantly decreased in human FGR. We                      components of the placenta [3]. Modulation of trophoblast cell
have also shown that HLX is a regulator of colony-stimulating-




                                                                                                                                                     Downloaded from www.biolreprod.org.
                                                                                functions involves various extracellular signals, signaling
factor-1-dependent trophoblast proliferation. Hepatocyte                        molecules, and subsequent receptor activation in the signaling
growth factor (HGF) activates trophoblast cell migration in a
paracrine fashion, and its receptor, c-met, is expressed on
                                                                                pathway. Disruption of these signaling pathways is associated
trophoblast cells. Given that HGF is a regulator of trophoblast                 with placental pathologies [4].
migration, we hypothesize that HLX is a mediator of HGF/c-met-                     Hepatocyte growth factor (HGF) is a well-established
dependent trophoblast migration but not invasion. Here we                       regulator of important trophoblast cell functions [5, 6]. HGF
investigated the potential role of HLX in HGF/c-met-mediated                    is a pleiotropic peptide of mesenchymal origin that was first
trophoblast migration and invasion in two human trophoblast-                    characterized by its ability to induce proliferation and
derived cell lines, SGHPL-4 and HTR-8/SVneo. Results showed                     migration of epithelial cells and mitogenesis of hepatocytes
that in cultured trophoblast cells, HGF stimulation significantly               [5, 7]. Cell paracrine interactions occur through HGF
increased HLX mRNA and protein expression. HLX inactivation                     interacting with its tyrosine kinase receptor, c-met. This
significantly decreased trophoblast migration but not invasion.                 HGF/c-met interaction plays a key role in epithelial gland
When HLX was inactivated in the presence of HGF stimulation,                    and tubule formation [8]. HGF is also produced in the villous
migration remained significantly decreased. SU11274-mediated                    core and mesenchymal cells of the placenta and in the maternal
inhibition of the receptor c-met significantly decreased HLX                    decidua [9].
mRNA and protein expression. In the presence of HGF                                HGF acts in a paracrine manner on the extravillous
stimulation, HLX expression remained significantly decreased                    trophoblast (EVT) cells, which express the c-met receptor
with c-met inhibition. This is the first study to show that                     [6], and this initiates cell signaling pathways that activate
homeobox gene HLX is a downstream effector gene of HGF, that                    trophoblast cell functions [7, 10]. Cartwright et al. [11] have
HLX regulates human trophoblast-derived cell migration, and                     demonstrated that HGF regulates trophoblast migration via the
that HGF, via receptor c-met, acts through HLX to control cell
                                                                                MAPK signaling pathway. Furthermore, mouse gene knockout
migration.
                                                                                studies reveal that mice mutant for either the Hgf gene or its
gene regulation, growth factors, HGF, HLX, migration, placenta,                 receptor c-met die in utero with severe fetal growth restriction
pregnancy, trophoblast                                                          effects and inadequate placentation [5].
                                                                                   Growth factor signaling pathways frequently lead to the
                                                                                activation, or repression, of transcription factors in the nucleus
                                                                                [12]. Of relevance to trophoblast function in this regard is the
                                                                                transcription factor homeobox gene HLX. Our previous
1
  Supported by NHMRC project grant 509140, the Lynne Quayle                     immunohistochemical study revealed that HLX is highly
Charitable Trust Fund (Equity Trustees), the Marian EH Flack Trust, and         expressed in proliferating and migrating EVT cells within the
a University of Melbourne Early Career Researcher Grant (to P.M.). G.R.
was supported by a Felix Myer Postgraduate Scholarship from the
                                                                                proximal region of the cell columns, as well as in endovascular
University of Melbourne and a Royal Women’s Hospital Postgraduate               trophoblast cells, which migrate in a retrograde fashion down
Scholarship.                                                                    the spiral arterioles [13]. However, expression of HLX was not
2
  Correspondence: Gayathri Rajaraman, University of Melbourne,                  detected in the nuclei of invading EVT in the distal regions of
Department of Obstetrics and Gynaecology, Level 7, The Royal                    the cell column [13] and, therefore, we predict that HLX does
Women’s Hospital, Locked Bag 300, Parkville, VIC 3052, Australia.               not play a direct role in trophoblast invasion.
FAX: 61 3 83453746; e-mail: gayaram@gmail.com
                                                                                   We have also demonstrated that HLX mRNA and protein
                                                                                expression is significantly reduced in FGR-affected human
Received: 3 May 2009.                                                           placentae compared with gestation age-matched controls [14].
First decision: 28 May 2009.
Accepted: 3 June 2010.
                                                                                These results are consistent with those of Somerset et al. [5],
Ó 2010 by the Society for the Study of Reproduction, Inc.                       who showed that HGF expression is substantially reduced in
eISSN: 1529-7268 http://www.biolreprod.org                                      FGR. Our most recent data [15] has identified downstream
ISSN: 0006-3363                                                                 target genes of HLX in the MAPK signaling pathway by
                                                                          676
HLX REGULATION OF TROPHOBLAST MIGRATION                                                                          677

employing the real-time PCR based RT2 profiler MAPK PCR                             supplemented with 10% fetal bovine serum (FBS), 100 U/ml penicillin, 100 lg/
array for gene profiling. This commercially available PCR                           ml streptomycin, 1.25 lg/ml Fungizone antimycotic (Invitrogen, Australia) and
                                                                                    0.1 mg/ml Tylosin (Sigma, Australia), as previously described [17]. HTR-8/
array encompassed the gene expression profiles of 84 genes                          SVneo cells were cultured in the same way using RPMI-1640 media
related to the MAPK signaling pathway. Members of the                               supplemented with 20% FBS. Cultures were maintained at 378C in an
MKKK, MKK, and MAPK families are represented on this                                atmosphere of 5% CO2 and 95% humidified air. Confluent cultures of cells
array as well as transcription factors and genes whose                              were seeded in 6-well and 24-well plates (Linbro Flow Laboratories, Australia)
expression is induced by MAPK signaling. This array                                 at a density of 2 3 105 and 5 3 104 cells per well, respectively. Cells seeded in
contained several cell cycle regulator genes, including CCNB1,                      6-well plates were harvested for RNA and protein analysis, and 24-well plates
                                                                                    were used for migration assays.
MYC, CDKN1C, and JUN, which have been previously
identified to be the targets of HLX in hematopoietic progenitor
cells [16]. Seven HLX candidate downstream target genes in                          Depletion of HLX mRNA Levels by siRNA
the MAPK pathway were identified as being RB1, MYC,                                     Inactivation of HLX by siRNA was performed as previously described [17].
EGR1, CDKN1C, ELK1, CCNB1, and JUN, which are genes                                 Briefly, cultured trophoblast cells were grown in Ham F-10 nutrient mixture/
known to be expressed in human trophoblast cells and                                RPMI-1640 media supplemented with 10% FBS for 48 h (5 3 104 cells per well
implicated in the regulation of cell migration and proliferation.                   in 24-well plates) and then transfected with the HLX siRNA [HLX-siRNA
                                                                                    sequence: r(UGAAUUUCUUGGGUUCGAG)d(TT)] using RNAiFect trans-
Furthermore, the identified HLX targets from the in vitro                           fection reagent (Qiagen, Australia). The siRNA used in this study was the most
trophoblast cell culture system were verified in human                              effective siRNA (HLX-si3 [17]) for optimal down regulation of HLX (80%
idiopathic FGR-affected placentae (n ¼ 25), which is associated                     reduction) out of four independent HLX-siRNA oligonucleotides that were
with abnormal trophoblast function, including migration and                         tested [17]. Following siRNA treatment, cells were returned to incubation at
proliferation, compared with gestation-age-matched control                          378C in 5% CO2 for 72 h. The controls used in the siRNA experiments were
placentae (n ¼ 25) obtained from uncomplicated human                                treated with a negative control siRNA consisting of a pool of enzyme-generated
                                                                                    siRNA oligonucleotides of 15–19 base pairs in length that were not specific for
pregnancies. Six out of the seven candidate HLX target genes,                       any known human gene (catalog #0652-13-7000; Superarray Bioscience
identified from cultured trophoblast cells, showed significant                      Corporation, Frederick, MD) and showed no significant sequence similarity to




                                                                                                                                                                        Downloaded from www.biolreprod.org.
gene expression changes in idiopathic FGR placental tissues,                        HLX (data not shown). The HLX siRNA oligonucleotide was designed to
consistent with cell line data. As we have previously shown                         exclude highly conserved regions of the homeobox and thereby avoid cross-
that HLX mRNA and protein expression is significantly                               reactivity [17]. The HLX siRNA showed no significant DNA sequence
decreased in human idiopathic FGR placentae [14], our results                       similarity to other genes in GenBank cDNA databases (data not shown).
showed that siRNA-mediated HLX inactivation in trophoblast
cell lines SGHPL-4 and HTR8/SVneo is a relevant in vitro                            HGF Stimulation
model for human idiopathic FGR. These data provided novel                               Confluent cultures of human EVT-derived cells were serum starved for 24 h
evidence that HLX directly or indirectly alters downstream                          in culture media supplemented with 0.5% bovine serum albumin (BSA;
target genes in the MAPK pathway, which may cause aberrant                          Sigma). Following starvation, the cells were stimulated with various
trophoblast functions associated with human idiopathic FGR                          concentrations (0.1–25 ng/ml) of recombinant human HGF (R&D Systems,
[15].                                                                               Minneapolis, MN) diluted in serum-free media for a further 24 h. The mock-
                                                                                    stimulated control for HGF stimulation experiments contained only serum-free
    Furthermore, we previously showed that the cytokine                             media without HGF added. For experiments involving HLX inactivation with
colony-stimulating factor-1 (CSF1) significantly upregulates                        HGF stimulation, cells were transfected with the HLX-siRNA oligonucleotide
the expression of HLX in cultured human trophoblast-derived                         or the negative control siRNA prior to serum starvation and HGF stimulation.
cells and that HLX is a regulator of CSF1-dependent
trophoblast proliferation [17].                                                     SU11274-Mediated c-met Inhibition
    Given that HGF is a regulator of trophoblast migration, we
hypothesize that HLX is a mediator of HGF/c-met-dependent                               Confluent cultures of human EVT-derived cells were grown in Ham F-10
                                                                                    nutrient mixture/RPMI-1640 media supplemented with 10% FBS for 48 h (2 3
trophoblast migration but is not involved in the regulation of                      105 cells per well in 6-well plates). Following this, the Met tyrosine kinase
trophoblast invasion. To test this hypothesis, the following                        inhibitor, SU11274 (Sigma; diluted in dimethyl sulfoxide [DMSO] according to
studies were undertaken. We measured the effect of HGF                              the manufacturer’s instructions), was added at a concentration of 6 lg per well
stimulation on HLX mRNA and protein expression in cultured                          and incubated for a further 24 h. The mock control contained only DMSO/
first trimester-derived human EVT cells. HLX gene expression                        media (1:5000 dilution) without SU11274 added. For experiments involving
was decreased by short interfering RNA (siRNA) oligonucle-                          SU11274-mediated c-met inhibition and HGF stimulation, cells were treated
                                                                                    with 6 lg per well SU11274, then stimulated with HGF (10 ng/ml). A
otides, and the effect on trophoblast cell migration and invasion                   concentration curve experiment for SU11274 was performed prior to c-met
was investigated. Finally, the dependence of HGF/c-met-                             inhibition. The SU11274 concentration tested ranged from 0–10 lg per well,
stimulated first trimester-derived human trophoblast migration                      and the optimum concentration was 6 lg per well (data not shown). A Trypan
on HLX expression was determined.                                                   blue exclusion assay for cell viability was also performed simultaneously for
                                                                                    the concentrations of SU11274 tested. SU11274-mediated c-met inhibition was
                                                                                    not toxic to the cultured trophoblast cells (data not shown).
MATERIALS AND METHODS
Cell Culture                                                                        Real-Time PCR
    This work was approved by the Research and Ethics Committees of The                 Total RNA was extracted from HGF-stimulated mock stimulated cells using
Royal Women’s Hospital, Melbourne, VIC, Australia. Well-characterized, first-       QIAShredder and an RNeasy Microkit (Qiagen) according to the manufactur-
trimester human EVT-derived cell lines SGHPL-4 [18] (a kind gift from               er’s instructions. First-strand synthesis was performed on 50 ng total RNA
Associate Professor Claire Roberts, University of Adelaide, Adelaide, SA,           using Superscript III ribonuclease H-reverse transcriptase (Invitrogen).
Australia) and HTR-8/SVneo [19] (a kind gift from Dr. Charles H. Graham,            Quantitation of HLX expression in cultured human EVT-derived cells was
Queen’s University, Kingston, Ontario, Canada) were used. These cell lines          performed in an ABI Prism 7700 (Applied Biosystems) using prevalidated
retain many features of normal EVT cells, including expression of cytokeratin-      Assays on Demand for HLX expression (consisting of 203 mix of unlabeled
7, human placental lactogen, human chorionic gonadotropin, and HLA-Class 1          HLX-PCR primers and TaqMan MGB probe (FAM dye labeled; HLX Assays
[20–22] and have been used in our previous functional studies [17]. The HTR-        on Demand catalog no. 4331182; Applied Biosystems) as previously described
8/SVneo cell line used in this study is a long-term cell line that senesces after   [14]. Gene expression quantitation was performed as the second step in a two-
about 30 passages. For this work, HTR-8/SVneo cells at passages 16–19 were          step RT-PCR protocol according to the manufacturer’s instructions. A total of
used. SGHPL-4 cells (passages 5–12) were cultured in Ham F-10 media                 20 ll PCR reaction mix containing TaqMan Universal PCR master mix,
678                                                                   RAJARAMAN ET AL.




FIG. 1. Real-time PCR of HLX mRNA expression with HGF stimulation.
Complementary DNA (2.5 ng) from SGHPL-4 cells stimulated with HGF
(0.1–25 ng/ml) and the no-stimulation mock control were amplified for 40
cycles, including denaturation at 958C for 15 sec and annealing/extension
at 608C for 60 sec (using prevalidated HLX Assays on Demand). Gene
expression quantitation for HLX was determined by coamplification with
the housekeeping gene GAPDH. Relative quantitation of HLX expression




                                                                                                                                                                    Downloaded from www.biolreprod.org.
in SGHPL-4 cells was normalized to GAPDH and calculated according to
the 2DDT method of Livak and Schmittgen [23], employing the mock
control as a calibrator. *P , 0.001, n ¼ 3. Error bars represent 6 SEM.


13 Assays on Demand gene expression assay mix, and 2.5 ng cDNA was used.
Amplification was carried out for 40 cycles, including denaturation at 958C for
15 sec and annealing/extension at 608C for 60 sec. Gene expression for the
housekeeping gene GAPDH was coamplified with HLX in the same reaction
                                                                                  FIG. 2. Effect of HGF stimulation on trophoblast migration. Trophoblast
well (catalog #4333762T; Applied Biosystems). Relative quantitation of HLX
                                                                                  migration was measured by performing Transwell migration assays on
expression normalized to GAPDH was calculated according to the 2DDCT
                                                                                  cultured trophoblast cells stimulated with 10 ng/ml HGF. Migrated
method of Livak and Schmittgen [23], employing the mock control as a
                                                                                  trophoblast cells on the underside of the membrane were stained and
calibrator according to the manufacturer’s instructions (ABI Prism 7700
                                                                                  counted. Representative fields of positively stained SGHPL-4 cells are
Sequence detection system, Applied Biosystems).
                                                                                  shown in A and B. A) The no-stimulation mock control. B) Migrated cells
                                                                                  in the presence of HGF stimulation. Bar ¼ 50 lm (A and B), magnification
Western Immunoblotting                                                            3100. C) Percentage of migrated cells with HGF stimulation compared
                                                                                  with the mock control. The pores of the filter can be readily distinguished
    Protein extraction from cultured human trophoblast-derived cells were         from the migrating trophoblast cells, which are larger and have a
carried out using homogenization buffer containing 0.1% Triton and 50 mM          fibroblast-like shape with prominent projections. The black and gray bars
glycine solution with 1 mM protease inhibitor cocktail AEBSF (4–2-                represent the SGHPL-4 and HTR-8/SVneo cell lines, respectively. *P ,
aminoethyl-benzenesulfonyl fluoride hydrochloride; Sigma), and cells were         0.001, n ¼ 4, t-test. Error bars represent 6 SEM.
homogenized by scraping. The protein concentration was calculated
spectrophotometrically by Pierce assay using Coomassie Blue, with BSA as
standard. Samples were electrophoresed on a Criterion 4%–15% gradient gel         inactivation in the presence of HGF, cells were transfected with siRNA for 72
(Biorad, Australia) in running buffer (250 mM Tris, 1.92 mM glycine) and          h, and then HGF diluted in serum-free media was added to the cells and
transferred electrophoretically to a nitrocellulose membrane at 100 V for 30      incubated for a further 24 h. The mock control wells received all reagents
min in transfer buffer (255 mM Tris, 1.92 mM glycine, 0.1% SDS). Since the        except HLX-siRNA or HGF. Transwell migration assays were then performed
51-kDa HLX product and the 40-kDa GAPDH product were of similar size,
                                                                                  according to the manufacturer’s instructions (catalog #3403, 8.0-lm pore size).
two parallel gels were prepared with the same protein load, one probed for
                                                                                  Briefly, cells subject to the experimental conditions were obtained as a
HLX and the other for GAPDH. The membranes were blocked with 5% (w/v)
                                                                                  suspension in 500 ll serum-free media, using the enzyme-free, PBS-based
nonfat milk powder/Tris-buffered saline (TBS) for 1 h. Incubation with a
                                                                                  dissociation solution (Chemicon, Australia) and added to the Transwell
rabbit polyclonal HLX antibody (1 lg/ml; catalog #P100839_T100; Aviva
                                                                                  migration inserts that contained a microporous membrane. Complete media
Biosys) or rabbit polyclonal GAPDH antibody (0.5 lg/ml; catalog #IMG-
6166K; Imgenex, San Diego, CA) was carried out overnight at 48C. The              (750 ll) was added to the wells containing the inserts to act as a
membranes were then washed in TBS and incubated with horseradish                  chemoattractant, and the migration assay plates were subjected to incubation
peroxidase-conjugated anti-rabbit secondary antibody (1.5 lg/ml; Zymed,           at 378C for 24 h. Following incubation, cells on the upper side of the membrane
Australia) in 5% (w/v) nonfat milk powder/TBS for 1 h at room temperature.        were gently swabbed off. Very few cells were left at the bottom of the wells.
Detection of protein bands was performed using the ECL-plus chemilumi-            The migrated trophoblast cells on the underside of the membrane were stained
nescence kit (Amersham Lifesciences, Australia) according to the manufac-         using Diff Quick stain kit, according to the manufacturer’s instructions (Grale
turer’s instructions. Competition experiments were performed to show the          Scientific, Australia). Migrated trophoblast cells were counted using the
specificity of HLX immunoreactivity.                                              Axiovision analysis software program version 3.5 (Zeiss, Australia). For cell
                                                                                  counting, four independent, non-overlapping regions were counted in each
                                                                                  insert, and the average was taken for the four replicate wells in each
Migration Assay                                                                   experimental condition tested.
    Trophoblast cell migration was assessed by Transwell migration assay
(Corning Life Sciences, Australia). The transwell migration assay used in this    Invasion Assay
study is a well-accepted and published method of measuring cell migration of
various cell types, including human trophoblast-derived cells [24–27]. SGHPL-         Human trophoblast-derived cell invasion was assessed by Matrigel invasion
4 and HTR-8/SVneo cells grown in 24-well plates were either stimulated with       assay according to the manufacturer’s protocol (catalog #354483; Becton and
HGF or transfected with the HLX-siRNA, as described above. For HLX                Dickinson, Franklin Lakes, NJ). The growth factor-reduced Bio-coat matrigel
HLX REGULATION OF TROPHOBLAST MIGRATION                                                                679




                                                                                                                                                             Downloaded from www.biolreprod.org.
                                                                                  FIG. 4. Trophoblast migration with HLX siRNA transfection and HGF
                                                                                  stimulation. Trophoblast migration was measured by performing Transwell
FIG. 3. Effect of HLX siRNA on trophoblast migration. Trophoblast                 migration assays on cultured trophoblast cells following the siRNA-
migration was measured by performing Transwell migration assays on                mediated decrease of HLX gene expression and HGF stimulation.
cultured trophoblast cells following the siRNA-mediated decrease of HLX           Migrated trophoblast cells on the underside of the membrane were
gene expression. Migrated trophoblast cells on the underside of the               stained and counted. Representative fields of positively stained SGHPL-4
membrane were stained and counted. Representative fields of positively            cells are shown in A and B. A) Migrated cells in the presence of HGF
stained SGHPL-4 cells are shown in A and B. A) The no-stimulation mock            stimulation alone. B) Migrated cells in the presence of HLX siRNA and
control. B) Migrated cells in the presence of HLX siRNA. Bar ¼ 50 lm (A           HGF stimulation. Bar ¼ 50 lm (A and B), magnification 3100. C)
and B), magnification 3100. C) Percentage of migrated cells with HLX              Percentage of migrated cells with HLX siRNA transfection and HGF
inactivation compared with the mock control. The black and gray bars              stimulation, compared with the HGF-stimulated cells alone. The black
represent the SGHPL-4 and HTR-8/SVneo cell lines, respectively. *P ,              and gray bars represent the SGHPL-4 and HTR-8/SVneo cell lines,
0.001, n ¼ 4, t-test. Error bars represent 6 SEM.                                 respectively. *P , 0.05, n ¼ 4, t-test. Error bars represent 6 SEM.


                                                                                  significantly influence HLX mRNA expression (P . 0.05, n ¼
membrane inserts (Becton and Dickinson) were rehydrated with the addition of
500 ll serum-free medium and incubated at room temperature for 30 min.
                                                                                  3, t-test); however, higher concentrations of HGF significantly
Following aspiration of the medium from the inserts, the protocol and cell        increased HLX mRNA levels in a dose-dependent manner, with
counting was performed as per the Transwell migration assay above.                HGF concentrations of 5.0–25 ng/ml, compared with the
                                                                                  mock-HGF-stimulated control without HGF added (P , 0.001,
Statistical Analysis                                                              n ¼ 3; Fig. 1). However, 10 ng/ml HGF resulted in maximal
                                                                                  upregulation of HLX (P , 0.001, n ¼ 3, t-test), which was not
    For PCR analysis, the experiments were repeated on three separate
occasions with triplicate wells for each sample in the experiment, whereas, for
                                                                                  significantly different to 25 ng/ml HGF stimulation (P ¼ 0.75,
migration and invasion assays, the experiments were repeated on three separate    data not shown). Therefore, HGF at 0 ng/ml was used for
occasions with quadruplicate wells for each sample in the experiment. The         further experimentation. This was consistent with previous
Student t-test was used for mRNA expression experiments and migration and         studies showing that HGF significantly stimulates trophoblast
invasion assays. A value of P , 0.05 was considered significant.                  migration at the same concentration [11].
RESULTS                                                                           HGF Stimulation Effect on Trophoblast-Derived
HGF Stimulation Effect on HLX mRNA Levels                                         Cell Migration
   Dose-response experiments were performed with HGF                                 Cells grown in 24-well plates were subjected to HGF
stimulation on SGHPL-4 cells, with HGF concentrations                             stimulation following serum starvation. Trophoblast cell
ranging from 0.1–25 ng/ml. Following RNA extraction, real-                        migration was calculated as the average number of positively
time PCR was performed to measure HLX mRNA expression                             stained migrated cells, relative to the appropriate mock-
relative to the housekeeping gene GAPDH (Fig. 1). Lower                           stimulated control, by using the Transwell migration assay.
concentrations of HGF (0.1, 1.0, and 2.5 ng/ml) did not                           SGHPL-4 cells stimulated with 10 ng/ml HGF showed a
680                                                               RAJARAMAN ET AL.




                                                                                                                                                          Downloaded from www.biolreprod.org.
                                                                             FIG. 6. Effect of HLX siRNA on trophoblast invasion. Trophoblast
                                                                             invasion was measured by performing Matrigel invasion assays on
                                                                             cultured trophoblast cells following siRNA-mediated HLX gene reduction.
                                                                             Invaded trophoblast cells on the underside of the membrane were stained
                                                                             and counted. Representative fields of positively stained SGHPL-4 cells are
                                                                             shown in A and B. A) The no-stimulation mock control. B) Trophoblast
                                                                             invasion in the presence of HLX siRNA. Bar ¼ 50 lm (A and B),
                                                                             magnification 3100. C) Percentage of migrated cells with HLX siRNA
                                                                             transfection compared with the mock control. The black and gray bars
                                                                             represent the SGHPL-4 and HTR-8/SVneo cell lines, respectively. P ¼ 0.83,
FIG. 5. A) HLX mRNA expression with SU11274-mediated c-met                   n ¼ 4, t-test. Error bars represent 6 SEM.
inhibition. Complementary DNA (2.5 ng) from SGHPL-4 and HTR8/
SVneo subjected to SU11274-mediated c-met inhibition. The mock
control had no added SU11274. Complementary DNA was amplified for            siRNA to reduce HLX mRNA and protein levels. This resulted
40 cycles, including denaturation at 958C for 15 sec and annealing/          in a significant decrease in SGHPL-4 and HTR-8/SVneo cell
extension at 608C for 60 sec (using prevalidated HLX Assays on Demand).      migration following HLX-siRNA transfection (Fig. 3). HLX
Gene expression quantitation for HLX was determined by coamplification
with the housekeeping gene GAPDH. Relative quantitation of HLX               inactivation resulted in a significant 75% decrease in SGHPL-
expression in SGHPL-4 and HTR8/SVneo cells normalized to GAPDH was           4 trophoblast migration, compared with the mock-transfected
calculated according to the 2DDT method of Livak and Schmittgen [23],       control (P , 0.001, n ¼ 4, t-test). HTR-8/SVneo cell
employing the mock control as a calibrator. *P , 0.05, n ¼ 3. Error bars     migration also significantly decreased by 65% with reduced
represent 6 SEM. B) HLX mRNA expression with SU11274-mediated c-             HLX levels relative to the mock control (P , 0.001, n ¼ 4, t-
met inhibition in the presence of HGF stimulation. Complementary DNA
(2.5 ng) from SGHPL-4 and HTR8/SVneo subjected to SU11274-mediated           test; Fig. 3).
c-met inhibition with HGF stimulation. The control was HGF stimulation
without added SU11274. Complementary DNA was amplified for 40                HGF Stimulation and HLX Down-Regulation Effect
cycles, including denaturation at 958C for 15 sec and annealing/extension
at 608C for 60 sec (using prevalidated HLX Assays on Demand). Gene           on Trophoblast-Derived Cell Migration
expression quantitation for HLX was determined by coamplification with          Transwell migration assays were performed on each cell line
the housekeeping gene GAPDH. Relative quantitation of HLX expression
in SGHPL-4 and HTR8/SVneo cells normalized to GAPDH was calculated           following HLX siRNA transfection in the presence of HGF
according to the 2DDT method of Livak and Schmittgen [23], employing        stimulation (Fig. 4). Reduced HLX mRNA levels had a
the mock control as a calibrator. *P , 0.05, n ¼ 3. Error bars represent 6   significant effect on SGHPL-4 and HTR-8/SVneo cell
SEM.                                                                         migration, even in the presence of HGF stimulation. Under
                                                                             these conditions, SGHPL-4 cell migration was reduced by 68%
significant increase in trophoblast-derived cell migration                   (P , 0.001, n ¼ 4, t-test), compared with HGF stimulation
                                                                             alone. HTR-8/SVneo cell proliferation was also significantly
compared with the mock-stimulated control (325%; P ,
                                                                             reduced by 55% relative to HGF stimulation alone (P , 0.001,
0.001, n ¼ 4, t-test), as did HTR-8/SVneo cells (215%; P ,                   n ¼ 4, t-test; Fig. 4).
0.001, n ¼ 4, t-test; Fig. 2).
                                                                             c-met Inhibition Effect on HLX mRNA Expression
HLX siRNA Effect on Trophoblast-Derived Cell Migration
                                                                                Real-time PCR for HLX mRNA expression was performed
  Transwell migration assays were performed on SGHPL-4                       on cDNA prepared from SGHPL-4- and HTR-8/SVneo-
and HTR-8/SVneo cells, following the addition of HLX-                        cultured cells following addition of 6 lg per well SU11274
HLX REGULATION OF TROPHOBLAST MIGRATION                                                                  681




                                                                                                                                                            Downloaded from www.biolreprod.org.
FIG. 7. Effect of HGF stimulation, HLX silencing, and c-met inhibition effects on HLX protein levels. Following protein extraction of SGHPL-4 and HTR-
8/SVneo cells subjected to HGF stimulation (10 ng/ml), HLX inactivation, HLX inactivation in the presence of HGF stimulation, c-met inhibition, and c-
met inhibition in the presence of HGF stimulation and control cells (n ¼ 3 pooled samples), protein assays were performed. Protein samples (25 lg) were
then electrophoresed and transferred to a nitrocellulose membrane. Immunoblotting was carried out using commercially available HLX and GAPDH
rabbit polyclonal antibodies. Composite images of chemiluminescence detection of HLX (70 kDa and 51 kDa) and GAPDH (40 Kda) are shown in A and
B, representing SGHPL-4 and HTR-8/SVneo cells, respectively. Lanes 1 and 2: mock and NC control samples; lane 3: 10 ng/ml HGF stimulation; lane 4:
siRNA-mediated HLX inactivation; lane 5: HLX inactivation in the presence of 10 ng/ml HGF; lane 6: SU11274-mediated c-met inhibition; lane 7:
SU11274 in the presence of 10 ng/ml HGF. C and D) The densitometric quantitation of HLX in (51-kDa band) SGHPL-4 and HTR-8/SVneo cells,
respectively, as the percentage densitometric values relative to GAPDH. Error bars represent 5% error in values. Where indicated by white dividing lines,
lanes 1–5 are from different areas of the same blot.


(Fig. 5A). SU11274-mediated c-met inhibition significantly                     HLX Silencing Effect on Trophoblast-Derived Cell Invasion
decreased HLX mRNA levels in both the cell lines, resulting in
an 80% decrease in SGHPL4 and 72% decrease in HTR-8/                              Following the addition of HLX-siRNA to reduce HLX
SVneo, compared with the mock control without added                            mRNA levels, Matrigel invasion assays were performed on
SU11274 (e.g., 2.1 6 0.32 SU11274 vs. 12.3 6 1.4 control,                      SGHPL-4 and HTR-8/SVneo cells. The results showed that
                                                                               HLX inactivation in SGHPL-4 and HTR-8/SVneo cells did not
densitometric units; P , 0.001, n ¼ 3).
                                                                               significantly affect trophoblast-derived cell invasion in either
                                                                               cell line, compared with the appropriate NC siRNA-transfected
c-met Inhibition and HGF Stimulation Effect on HLX                             controls (P ¼ 0.83, n ¼ 4, t-test; Fig. 6).
mRNA Expression
   Real-time PCR for HLX mRNA expression was performed                         HGF Stimulation, HLX Silencing, and c-met Inhibition
on cDNA prepared from SGHPL-4 and HTR-8/SVneo                                  Effects on HLX Protein Levels
trophoblast cell lines, with SU11274-mediated c-met inhibition                    Western immunoblotting for HLX protein expression was
and HGF stimulation (Fig. 5B). Under these conditions, HLX                     performed on protein lysates from SGHPL-4 and HTR-8/
mRNA expression remained significantly decreased in both the                   SVneo cells subjected to HGF stimulation (10 ng/ml), HLX
cell lines, with a 75% decrease in SGHPL-4 and 63% decrease                    inactivation, HLX inactivation in the presence of HGF
in HTR-8/SVneo compared with HGF stimulation alone (e.g.,                      stimulation, c-met inhibition, and c-met inhibition in the
8.02 6 1.3 SU11274 vs. 38.3 6 5.4 HGF, densitometric units;                    presence of HGF stimulation, as shown in Figure 7, A
P , 0.001, n ¼ 3).                                                             (SGHPL-4) and B (HTR-8/SVneo). Two bands were detected
682                                                         RAJARAMAN ET AL.

with a commercial HLX antibody (Aviva Biosys): the expected            trophoblast cell migration remained significantly decreased in
51-kDa HLX product [13, 14] and a higher molecular weight              both the cell lines. This suggests that HGF stimulation of
70-kDa product. Densitometric analysis of the expected 51-             trophoblast-derived cell migration is dependent on HLX
kDa band showed that HLX protein expression is significantly           expression.
increased with HGF stimulation in both the cell lines tested (P           HGF is produced in the placental villous core and acts in a
, 0.05, n ¼ 3, t-test). HLX protein expression is significantly        paracrine manner on human trophoblast-derived cells that
decreased with HLX inactivation, HLX inactivation in the               express the HGF receptor, c-met [6]. Studies have shown that
presence of HGF stimulation, c-met inhibition, and c-met               the small molecule inhibitor of c-met, SU11274, inhibited
inhibition in the presence of HGF stimulation (P , 0.05, n ¼ 3,        phosphorylation of c-met and downstream target genes,
t-test) in both the cell lines tested compared with the controls       resulting in decreased cell function, including migration [36,
(Fig. 7, C [SGHPL-4] and D [HTR-8/SVneo]). Densitometric               37]. In this study, SU11274-mediated c-met inhibition of
analysis of the 70-kDa band relative to GAPDH showed                   cultured human trophoblast-derived cells resulted in signifi-
essentially the same trends seen with the 51-kDa HLX product           cantly reduced HLX mRNA and protein expression. We
(data not shown). Further analyses on the specificity of the           showed that the inhibition of c-met activity resulted in the
immunoreactive HLX protein with the peptide competed out               inhibition of homeobox gene transcription in placental
both of the 70-kDa and 51-kDa HLX proteins (data not                   trophoblast cells. Furthermore, HLX mRNA and protein
shown). The 70-kDa band may represent a multimeric form of             expression remained significantly reduced with c-met inhibi-
HLX or a posttranslationally modified form of the HLX.                 tion in the presence of HGF stimulation, suggesting that HLX is
                                                                       an important downstream effector gene in the HGF/c-met
DISCUSSION                                                             signaling pathway of trophoblast-derived cell migration.
                                                                          Results from this study show that HLX is not a direct
    In this study we have used two independently derived, well-        regulator of trophoblast invasion, as siRNA-mediated inacti-
characterized first trimester-derived human EVT cell lines,            vation of HLX did not significantly affect trophoblast invasion
HTR-8/SVneo and SGHPL-4, to model first trimester EVT cell




                                                                                                                                                        Downloaded from www.biolreprod.org.
                                                                       (Fig. 6). This is consistent with our previous immunohisto-
migration and invasion. The results from this study were not           chemistry studies, which showed the expression pattern of HLX
cell-line specific, as consistent effects were seen in both the cell   in human trophoblast cells underwent a transition from nuclear
lines tested.                                                          expression in the actively proliferating and migrating cells of
    We verified the stimulatory effect of HGF on trophoblast           the proximal regions of the cell column to cytoplasmic
cell migration as demonstrated by others [11, 28] and showed           expression in the invasive EVT cells of the distal regions
that the same stimulatory effect on cell migration occurred in         [13]. The transition of HLX expression from nuclear to
the two trophoblast cell lines SGHPL-4 and HTR-8/SVneo.                cytoplasmic correlates with the declining proliferation potential
    HLX (also known as HB24 and HLX1) is an important                  of the EVT as they acquire an invasive phenotype.
regulator of cell proliferation and migration in hematopoietic            However, HLX is expressed in the nuclei of human EVT-
progenitor cells following stimulation by growth factors such          derived cell lines, including HTR-8/SVneo, JEG-3, JAR, and
as interleukin-3 and GM-CSF [16]. The data from this study             BeWo [13]. The HTR8/SVneo and SGHPL-4 trophoblast cells
show that growth factor-induced HLX stimulation of cell                lines are transformed trophoblast cell lines capable of
migration is not limited to the hematopoietic cell lineage but         proliferation, migration and invasion. Therefore, we conclude
also occurs in placental trophoblast cells. In this study, HGF         that nuclear expression of HLX is required for trophoblast-
stimulation of SGHPL-4 cells resulted in significantly                 derived cell proliferation and migration, but not for trophoblast
increased HLX mRNA levels. HGF stimulation of SGHPL-4                  invasion.
and HTR-8/SVneo cells resulted in significantly increased                 In summary, we have shown in cultured human trophoblast-
HLX protein levels. We have shown, therefore, that HGF is an           derived cells that the homeobox gene HLX is a downstream
activator of homeobox gene transcription in a reproductive cell        effector gene of HGF that HLX regulates trophoblast migration
type.                                                                  and that HGF, via its receptor c-met, acts through HLX to
    Targeted gene disruption in the mouse model system reveals         control cell migration. The results of this current study may
that many transcription factors, including homeobox genes, are         account for some aspects of the molecular mechanism for FGR:
important for murine placental development [29–33]. Howev-             decreased levels of HGF [5] result in lower levels of HLX
er, our understanding of the functional role of homeobox genes         mRNA and protein [14], and this in turn would deplete the pool
in the human placenta is rudimentary. We have previously               of proliferating EVT and reduce their ability to migrate. The
shown by immunohistochemistry that the HLX homeobox gene               reduction of EVT cell numbers is a characteristic of FGR.
is expressed primarily in the nuclei of proliferating and              However, we have shown that HLX does not regulate the
migrating cytotrophoblast cell types in early pregnancy human          invasive potential of EVT and, therefore, other mechanisms
placenta but not in the nuclei of invading distal cytotropho-          must be responsible for the shallow invasion, which is
blasts in the cell column [13]. Additionally, the changes in           associated with FGR.
expression of MMP2 and MMP9 along the cell column, and a
change in morphology of these cells as they reach the decidua,         REFERENCES
further suggest that the cells in the proximal zone are indeed
                                                                        1. Kaufmann P, Black S, Huppertz B. Endovascular trophoblast invasion:
migratory and not invasive [34, 35]. We have confirmed that                implications for the pathogenesis of intrauterine growth retardation and
HLX is an important regulator of CSF1-mediated trophoblast                 preeclampsia. Biol Reprod 2003; 69:1–7.
cell proliferation [17]. In this study, we showed that a decrease       2. Kadyrov M, Kingdom J, Huppertz B. Divergent trophoblast invasion and
in HLX gene expression levels in cultured trophoblast cells                apoptosis in placental bed spiral arteries from pregnancies complicated by
resulted in substantially decreased trophoblast cell migration.            maternal anemia and early-onset preeclampsia/intrauterine growth restric-
                                                                           tion. Am J Obst Gyn 2006; 194:557–563.
    We investigated whether HLX is a downstream effector gene           3. Chakraborty C, Gleeson LM, McKinnon T, Lala PK. Regulation of human
in the HGF signaling pathway of trophoblast migration. Upon                trophoblast migration and invasiveness. Can J Physiol Pharmacol 2002;
stimulation of human trophoblast-derived cells with HGF, in                80:116–124.
the presence of HLX siRNA, HLX protein expression and                   4. Xu G, Chakraborty C, Lala PK. Expression of TGF-beta signaling genes in
HLX REGULATION OF TROPHOBLAST MIGRATION                                                                          683

      the normal, premalignant, and malignant human trophoblast: loss of smad3             models of human trophoblast II: trophoblast cell lines—a workshop report.
      in choriocarcinoma cells. Biochem Biophys Res Commun 2001; 287:47–                   Placenta 2001; 22(suppl A):S104–S106.
      55.                                                                            21.   Lash GE, Cartwright JE, Whitley GS, Trew AJ, Baker PN. The effects of
 5.   Somerset DA, Li XF, Afford S, Strain AJ, Ahmed A, Sangha RK, Whittle                 angiogenic growth factors on extravillous trophoblast invasion and
      MJ, Kilby MD. Ontogeny of hepatocyte growth factor (HGF) and its                     motility. Placenta 1999; 20:661–667.
      receptor (c-met) in human placenta: reduced HGF expression in                  22.   Choy MY, Manyonda IT. The phagocytic activity of human first trimester
      intrauterine growth restriction. Am J Pathol 1998; 153:1139–1147.                    extravillous trophoblast cells. Hum Reprod 1998; 13:2941–2949.
 6.   Kauma SW, Bae-Jump V, Walsh SW. Hepatocyte growth factor stimulates            23.   Livak KJ, Schmittgen TD. Analysis of relative gene expression data using
      trophoblast invasion: a potential mechanism for abnormal placentation in             real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods
      preeclampsia. J Clin Endocrinol Metab 1999; 84:4092–4096.                            2001; 25:402–408.
 7.   Dokras A, Gardner LM, Seftor EA, Hendrix MJ. Regulation of human               24.   Mc Kinnon T, Chakraborty C, Gleeson LM, Chidiac P, Lala PK.
      cytotrophoblast morphogenesis by hepatocyte growth factor/scatter factor.            Stimulation of human extravillous trophoblast migration by IGF-II is
      Biol Reprod 2001; 65:1278–1288.                                                      mediated by IGF Type 2 receptor involving inhibitory G protein(s) and
 8.   Tsarfaty I, Resau JH, Rulong S, Keydar I, Faletto DL, Van de Woude GF.               phosphorylation of MAPK. J Clin Endocrinol Metab 2001; 86:3665–3674.
      The met proto-oncogene receptor and lumen formation. Science 1992;             25.   Rahnama F, Shafiei F, Gluckman PD, Mitchell MD, Lobie PE. Epigenetic
      257:1258–1261.                                                                       regulation of human trophoblastic cell migration and invasion. Endocrinol
 9.   Choy MY, Siu SS, Leung TN, Lau TK. Human decidual production of                      2006; 147:5275–5283.
      hepatocyte-growth factor is not influenced by trophoblastic invasion in        26.   Sahir KY, Schnegg CI, Raab-Traub N. EBV latent protein membrane-1
      vivo. Fertil Steril 2004; 82(suppl 3):1220–1225.                                     effects on plakoglobin cell growth and migration. Cancer Res 2008; 68:
10.   Nasu K, Zhou Y, McMaster MT, Fisher SJ. Upregulation of human                        6997–7005.
      cytotrophoblast invasion by hepatocyte growth factor. J Reprod Fertil          27.   Bernaciak TM, Zareno J, Parsons JT, Silva CM. A novel role for signal
      2000; 55(suppl):73–80.                                                               transducer and activator of transcription 5b (STAT5b) in b-1-integrin-
11.   Cartwright JE, Kenny LC, Dash PR, Crocker IP, Aplin JD, Baker PN,                    mediated human breast cancer cell migration. Breast Cancer Res 2009; 11:
                                                                                           R52; doi:10.1186/bcr2341.
      Whitley GS. Trophoblast invasion of spiral arteries: a novel in vitro model.
                                                                                     28.   Cartwright JE, Holden DP, Whitley GS. Hepatocyte growth factor
      Placenta 2002; 23:232–235.
                                                                                           regulates human trophoblast motility and invasion: a role for nitric oxide.
12.   Irving JA, Lala PK. Functional role of cell surface integrins on human
                                                                                           Br J Pharmacol 1999; 128:181–189.
      trophoblast cell migration: regulation by TGF-beta, IGF-II, and IGFBP-1.




                                                                                                                                                                         Downloaded from www.biolreprod.org.
                                                                                     29.   Simmons DG, Cross JC. Determinants of trophoblast lineage and cell
      Exp Cell Res 1995; 217:419–427.
                                                                                           subtype specification in the mouse placenta. Dev Biol 2005; 284:12–24.
13.   Rajaraman G, Murthi P, Quinn L, Brennecke SP, Kalionis B.
                                                                                     30.   Rinkenberger JL, Cross JC, Werb Z. Molecular genetics of implantation in
      Homeodomain protein HLX is expressed primarily in cytotrophoblast                    the mouse. Dev Genet 1997; 21:6–20.
      cell types in the early human placenta. Reprod Fertil Dev 2008; 20:357–        31.   Rossant J, Cross JC. Placental development: lessons from mouse mutants.
      367.                                                                                 Nat Rev Genet 2001; 2:538–548.
14.   Murthi P, Doherty V, Said J, Donath S, Brennecke SP, Kalionis B.               32.   Hemberger M, Cross JC. Genes governing placental development. Trends
      Homeobox gene HLX1 expression is decreased in idiopathic human fetal                 Endocrinol Metab 2001; 12:162–168.
      growth restriction. Am J Pathol 2006; 168:511–518.                             33.   Sapin V, Blanchon L, Serre AF, Lemery D, Dastugue B, Ward SJ. Use of
15.   Rajaraman G, Murthi P, Pathirage N, Brennecke SP, Kalionis B.                        transgenic mice model for understanding the placentation: towards clinical
      Downstream targets of homeobox gene HLX show altered expression in                   applications in human obstetrical pathologies? Transgenic Res 2001; 10:
      human idiopathic fetal growth restriction. Am J Pathol 2010; 176:278–                377–398.
      287.                                                                           34.   Isaka K, Usuda S, Ito H, Sagawa Y, Nakamura H, Nishi H, Suzuki Y, Li
16.   Kehrl JH, Deguchi Y. Potential roles for two human homeodomain                       YF, Takayama M. Expression and activity of matrix metalloproteinase 2
      containing proteins in the proliferation and differentiation of human                and 9 in human trophoblasts. Placenta 2003; 24:53–64.
      hematopoietic progenitors. Leuk Lymphoma 1993; 10:173–176.                     35.   Huppertz B, Kertschanska S, Demir AY, Hans-Georg F, Kaufmann P.
17.   Rajaraman G, Murthi P, Leo B, Brennecke SP, Kalionis B. Homeobox                     Immunohistochemistry of matrix metalloproteinases (MMP), their sub-
      gene HLX1 is a regulator of colony stimulating factor-1 dependent cell               strates, and their inhibitors (TIMP) during trophoblast invasion in the
      proliferation. Placenta 2007; 28:991–998.                                            human placenta. Cell Tissue Res 1998; 291:133–148.
18.   Tse WK, Whitley GS, Cartwright JE. Transforming growth factor-beta1            36.   Koon EC, Ma PC, Salgia R, Welch WR, Christensen JG, Berkowitz RS,
      regulates hepatocyte growth factor-induced trophoblast motility and                  Mok SC. Effect of a c-Met-specific, ATP-competitive small-molecule
      invasion. Placenta 2002; 23:699–705.                                                 inhibitor SU11274 on human ovarian carcinoma cell growth, motility, and
19.   Xia Y, Wen HY, Kellems RE. Angiotensin II inhibits human trophoblast                 invasion. Int J Gynecol Cancer 2007; 18:976–984.
      invasion through AT1 receptor activation. J Biol Chem 2002; 277:24601–         37.   Puri N, Ahmed S, Janamanchi V, Tretiakova M, Zumba O, Krausz T,
      24608.                                                                               Jagadeeswaran R, Salgia R. c-Met is a potentially new therapeutic target
20.   Shiverick KT, King A, Frank H, Cartwright JE, Schneider H. Cell culture              for treatment of human melanoma. Clin Cancer Res 2007; 13:2246–2253.
You can also read
Next part ... Cancel