HER1 Signaling Mediates Extravillous Trophoblast Differentiation in ...

BIOLOGY OF REPRODUCTION 83, 1036–1045 (2010) Published online before print 25 August 2010. DOI 10.1095/biolreprod.109.083246

HER1 Signaling Mediates Extravillous Trophoblast Differentiation in Humans1 J.K. Wright,3,4,5 C.E. Dunk,2,3,4 H. Amsalem,4 C. Maxwell,6 S. Keating,7 and S.J. Lye4,5,6 Women’s and Infants’ Health Research Centre,4 Samuel Lunenfeld Research Institute, Mount Sinai Hospital, Departments of Physiology,5 Obstetrics and Gynecology,6 and Pathology,7 University of Toronto, Toronto, Canada ABSTRACT

decidua, EGF, epidermal growth factor receptor, HBEGF, HER1, invasion, signal transduction, trophoblast, trophoblast invasion

INTRODUCTION Extravillous trophoblast (EVT) migration and invasion into the maternal decidua are critical aspects of normal human placentation. The process of EVT invasion leads to the transformation of the maternal spiral arteries into largediameter, low-resistance, high-flow vessels capable of supplying adequate blood into the intervillous space to nourish the growing conceptus. In normal pregnancies the depth of trophoblast invasion is strictly regulated to ensure adequate access of placental cells to the maternal spiral arteries [1]. The importance of this regulation is underscored by the pathologies 1

Supported by the CIHR IHD 165436. Correspondence: C.E. Dunk, Women’s and Infants’ Health Research Centre, Samuel Lunenfeld Research Institute, Mount Sinai Hospital, 25 Orde Street, Rm. 6-1025, Toronto, ON, Canada M5T 3H7. FAX: 416 586 5116; e-mail: [email protected] 3 These authors contributed equally to this manuscript. 2

Received: 22 December 2009. First decision: 22 January 2010. Accepted: 4 August 2010. Ó 2010 by the Society for the Study of Reproduction, Inc. eISSN: 1529-7268 http://www.biolreprod.org ISSN: 0006-3363

1036

Downloaded from www.biolreprod.org.

This study examines the role of HER1 signaling in the differentiation of proliferative extravillous trophoblast (EVT) into invasive EVT. Using the JAR choriocarcinoma cell line and placental villous explants as experimental models and immunohistochemical assessment of protein markers of EVT differentiation (downregulation of HER1 and Cx40 and upregulation of HER2 and alpha1 integrin), we show that the ability of decidual conditioned medium (DCM) to induce HER1/2 switching was abrogated in the presence of the HER1 antagonist, AG1478. Similarly, epidermal growth factor (EGF) treatment resulted in the downregulation of HER1 and an upregulation of HER2 expression, whereas co-incubation of EGF with AG1478 inhibited this response. However, EGF did not downregulate Cx40 or induce migration of EVT. In contrast, heparin-binding epidermal-like growth factor (HBEGF) stimulated dose-dependent JAR cell migration, which was inhibited by both AG1478 and AG825 (HER2 antagonist). Western blot analysis of HER1 activation demonstrated that HBEGF-mediated phosphorylation of the HER1 Tyr992 and Tyr1068 sites, while EGF activated the Tyr1045 site. Moreover, HBEGF induced a stronger and more sustained activation of both the mitogen-activated protein kinase and phosphoinositol 3 kinase (PIK3) signaling pathways. Migration assays using a panel of signaling pathway inhibitors demonstrated that the HBEGF-mediated migration was dependent on the PIK3 pathway. These results demonstrate that HBEGF-mediated HER1 signaling through PIK3 is an important component of EVT invasion.

of pregnancy that arise from aberrant invasion; for instance, preeclampsia is marked by hypoinvasion of the trophoblast into the decidua and a failure of vascular remodeling [2, 3], whereas choriocarcinoma, invasive moles, and placenta accreta are characterized by hyperinvasion of the trophoblast into maternal tissues [4]. The EVT population of the placenta is phenotypically heterogeneous, comprised of both proliferative and invasive cells. When the proliferative EVT of the placental cell column contact the maternal decidua, they become invasive cells capable of moving through and colonizing the decidual stroma. This differentiation event is tightly regulated both temporally and spatially such that the EVT of the cell columns are proliferative but noninvasive, and the invasive EVT cells within the decidual stroma have no proliferative capacity [1]. The transition between the proliferative and invasive EVT phenotypes is associated with changes in the expression profiles of several proteins: connexin-40 (Cx40), MKI67, EGF receptor (HER1), and the a5 integrin are expressed by proliferative EVT; the EGF receptor c-erb-B2 (HER2) and the a1 integrin are expressed by invasive EVT [4–8]. Previous investigations by our group into the mechanisms regulating the differentiation of EVT demonstrated that treatment of villous explants with decidua-conditioned media (DCM) initiated a complete differentiation of EVT from the proliferative to the invasive phenotype. Under these conditions, EVT not only downregulated MKI67, Cx40, and a5 integrin and upregulated a1 integrin expression, but also downregulated HER1 expression and began to express HER2 [9]. In JAR cells, DCM treatment downregulated Cx40 expression and induced cell migration. Notably, AG1478, a potent and specific inhibitor of HER1 tyrosine kinase activity [10], blocked the DCM-induced differentiation effects on JAR cells [9]. These findings suggested that activation of HER1 and its downstream signaling pathways may be important for initiating the differentiation of invasive EVT. However, when we treated JAR cells with EGF (one of many HER1 ligands) in a transwell migration assay, it did not induce any JAR cell migration, suggesting that specific HER1 activation, via another HER1 ligand, was required to prompt JAR cell migration. These data, together with the phenotype-specific expression profile of HER isoforms in EVT, suggests that different members of the EGF family may be capable of initiating distinct signaling events in proliferative and invasive EVT. A wealth of studies has established roles for HER signaling in a variety of cellular functions [11–13]. In particular, HER1 has been reported to be an important signal mediator of mitotic events, and a growing number of studies report that HER2 signaling is involved in invasive processes such as those underlying breast cancer metastasis [7, 14, 15]. The involvement of HER1 and HER2 in each of these respective processes, their EVT phenotype-specific expression pattern, and the detection of many EGF family ligands at the maternal/fetal interface, such as HBEGF, transforming growth factor-a

1037

HER1 AND EVT DIFFERENTIATION TABLE 1. Antibodies, dilutions, and antigen retrieval for placental villous explant immunohistochemistry. Antibody Cx40 a1 integrin HER1 HER2 Cy-3 anti-mouse FITC anti-rabbit Biotinylated anti-mouse Biotinylated anti-rabbit Alexa 488-conjugated streptavidin a

Host

Dilution

Source

Unmaskinga

Rabbit Mouse Rabbit Mouse Rabbit

1:125 1:100 1:500 1:250 1:300 1:200 1:200 1:300 1:300 1:1000

Invitrogen Abcam Cedarlane Santa Cruz (CA) DAKO Jackson Laboratories Jackson Laboratories DAKO DAKO Invitrogen

Triton 0.02% Triton 0.02% Proteinase K (10 lg/ml) Trypsin 0.125% 10 mM Sodium citrate pH 6, heat NA NA NA NA NA

NA, not applicable.

MATERIALS AND METHODS Tissue Collection First trimester placentae and decidua were obtained at the time of elective terminations of pregnancy. Informed consent was obtained from each patient, and collections were approved by the Mount Sinai Hospital’s Review Committee on the Use of Human Subjects. Tissue was collected into ice-cold PBS for villous explant or decidual cell culture.

modified Eagle medium-Ham F-12 media (400 ll; Invitrogen, Burlington, ON) supplemented with 100 lg/ml streptomycin (Sigma, St. Louis, MO), 100 U/ml penicillin (Sigma), pH7.4 was placed in the well surrounding the outside of the culture insert. Explants were allowed to attach to the matrigel and were covered in 200 ll Explant Media. Explants were treated 48 h later once the villous outgrowth was established. Patient-matched decidua parietalis was dissected to 2-mm2 pieces and placed in a 24-well plate, with approximately 100 mg of tissue per well. Decidua explants were cultured for 48 h at 3% O2/5% CO2 and 378C in Explant Media. DCM was collected, centrifuged at 2000 rpm for 5 min, filtered to remove contaminating cells, and supplemented with 2 mM Lglutamine (Invitrogen) prior to treatment of established EVT outgrowths. For experimental treatments, DCM was supplemented with vehicle (0.0025% dimethyl sulfoxide [DMSO]) or HER1 antagonist AG1478 (2 lM in 0.0025% DMSO; Cedarlane, Burlington, ON). Established outgrowths were also treated with 10 ng/ml EGF (Sigma) 6 2 lM AG1478. Explants from a single placenta were cultured in triplicate for each treatment point. Each experiment was repeated with at least three placentae.

Fluorescent Immunohistochemistry Placental villous explants were fixed and processed to paraffin blocks from which 5-lm sections were cut and adhered to Superfrost Plus glass slides (VWR, Mississauga, ON). Sections were deparaffinized in xylene and rehydrated through a descending concentration gradient of ethanol. Antigen retrieval was performed using either microwave pretreatment in 10 mM sodium citrate buffer (pH 6; Sigma), 0.02% Triton X100 (Sigma), 0.125% Trypsin (Sigma), or 10 lg/ml proteinase K (Roche, Montreal, QC, Canada) at 378C (Table 1). Slides used for immunofluorescence detection were rapidly exposed to 0.1% Sudan Black in 70% ethanol to prevent autofluorescence (Sigma) and washed prior to blocking for nonspecific binding in serum-free DAKO protein block (DAKO, Mississauga, ON, Canada) for 1 h. All primary and secondary antibodies used in these procedures are detailed in Table 1. Primary antibodies were prepared in DakoCytomation Antibody Diluent with BackgroundReducing Components (DAKO) and were incubated on sections overnight at 48C. In control experiments, primary antibodies were replaced with DAKO blocking solution or mouse or rabbit IgG at the same concentration as the primary antibody. Secondary biotinylated antibodies were prepared in PBS þ 0.04% Azide (Sigma) þ 0.008% gelatin (Sigma), used at 1:300, and detected using Streptavidin-Alexa488 (1:1000; Invitrogen); cells were counterstained with the nuclear stain Hoescht 33258 (1 lg/ml, 1 h; Sigma). All incubations were performed in a light-protected incubation chamber. Slides were washed in PBS and mounted in 50% glycerol/50% PBS for deconvolution microscopy. Immunofluorescent images were captured using a Sony Interline ICX285ER Progressive scan camera and an Olympus IX70 microscope (Olympus America Inc., Melville, NY). Images were collected using Resolve3D Image acquisition software and deconvolved using Deltavision softWoRx 2.50 software (Applied Precision, Issaquah, WA).

Placental Villous and Decidual Explant Culture and Treatment

Maintenance of Cell Lines

Villous explant cultures were established from first trimester human placentae using the method of Caniggia et al. [24]. Small fragments (15–20 mg wet wt) of placental villi from 6- to 8-wk gestation placentae were dissected from the placenta, teased apart, and placed on Millicell-CM culture dish inserts (pore size 0.4 lm; Millipore Corp., MD), precoated with 0.2 ml undiluted phenol red-free Matrigel substrate, and cultured at 3% O2/5% CO2 and 378C (Becton Dickinson, Mississauga, ON). Explant Media serum-free Dulbecco

The human placental choriocarcinoma cell line JAR was obtained from American Type Culture Collection (Manassas, VA). Cells were maintained at 378C under atmospheric O2 plus 5% CO2 in 10% FBS RPMI, supplemented with 100 lg/ml amphotericin B, 100 lg/ml streptomycin, and 100 U/ml penicillin. Prior to treatment, JAR cells were serum-starved for 24 h in 0.2% BSA RPMI media supplemented with Normocin (Cedarlane) serum free media (SFM).


(TGF), and amphiregulin (AREG), support a role for these proteins in the expression of EVT phenotype [16–20]. In the present study, we examine the role of HER1 signaling in mediating the differentiation cascade by which proliferative EVT transform into invasive EVT. We used two models of trophoblast differentiation. First, we used the JAR choriocarcinoma cell line as a model of proliferative EVT on the basis of its common expression of many EVT phenotype markers, as found on EVT in vivo. While this cell line is derived from a term trophoblast tumor, these cells express Cx40 [21] and HER1 [22] just as proliferative EVT do; moreover, under serum-free medium conditions, JAR cells are proliferative and nonmigratory [9]. However, just as placental villous EVT become invasive when treated with DCM, so too do JAR cells become migratory under these conditions [9]. Secondly, we confirmed our observations from JAR cultures using cultures of first trimester placental villous explant outgrowths [23] since this model more closely approximates the in vivo milieu. To assess the role of HER1 in the differentiation of invasive EVT, we treated JAR cells and placental explants with either DCM or EGF and assessed the expression of HER receptors. Furthermore, we attenuated these HER1-mediated differentiation effects using AG1478. We also assessed JAR cell migration in response to an alternative HER1 ligand, HBEGF, and determined the role of HER1 and HER2 in this process using the receptor-specific inhibitors, AG1478 and AG825, as well as specific inhibitors of the mitogen-activated protein kinase (MAPK) pathway, protein kinase C (PKC) pathway, and phosphoinositol 3 kinase (PIK3) pathway. Finally, we investigated HER1 phosphorylation status and downstream MAPK and PIK3 signaling pathway activation following treatment with either EGF or HBEGF.

1038

WRIGHT ET AL.

JAR Cell Migration Assays JAR choriocarcinoma cells were seeded in fibronectin-coated 8-lm pore transwell cell culture inserts (100 000 cells per well in SFM; Becton Dickinson) and allowed to attach for 8 h under the culture conditions described above. The protocol for migration experiments involving the HER1 antagonist (2 lM AG1478), HER2 antagonist (5 lM AG825), or HER1 signaling pathway inhibitors (1 lM PD90589, 5 lM SB203580, 100 or 50 nM Ro318220, or 10 or 1 lM LY294002; CedarLane) involved a 30-min pretreatment of JAR cells with the antagonist/inhibitor prior to placing the JAR cell-containing inserts into culture wells containing HBEGF (Sigma)-supplemented SFM. Serum-free vehicle controls were also performed by the addition of 0.0025% DMSO. After 48 h of exposure to HBEGF, migrated JAR cells were fixed, stained with DiffQuik (Dade Behring, Milton Keynes, U.K.), and mounted in 90% glycerol and 10% PBS. Four random fields, each 1 cm2 per filter, were counted at 2003 magnification to obtain a mean value for each filter. Each experiment was carried out in triplicate and performed as four independent experiments.

Western Blot Analysis

Treatment of Cell Lines and Fluorescent Immunocytochemistry After a period of 24 h of serum starvation (above), JAR cells were preincubated with either vehicle (0.0025% DMSO) or AG1478 (2 lM in 0.0025% DMSO) for 30 min. JAR cells were then stimulated with either DCM þ vehicle, EGF þ vehicle, or þ/ AG1478 (2 lM in 0.0025% DMSO) as indicated in each experiment for 24 h. DCM was collected from primary first trimester decidual cell cultures as previously described [9]. Briefly, decidual cells were serum-starved in 0.2% BSA (Sigma) RPMI supplemented with Normocin (serum-free DCM) for 48 h. Following the treatment period, immunocytochemistry was performed on confluent JAR monolayers. JAR cells were fixed with 4% paraformaldehye, permeabilized using 0.02% triton X100, and quickly exposed to Sudan Black. All primary and secondary antibodies used in these procedures are detailed in Table 1. Primary and secondary antibodies were prepared and used as described above. Slides were coverslipped using 90% glycerol. All incubations were performed in a light-protected incubation chamber. Fluorescent images were captured as described above.

Bioplex ELISA Lysates from JAR cells stimulated with either EGF or HBEGF were prepared as above in Bio-Rad cell lysis buffer and diluted to 0.7 lg/ml. To detect downstream HER1 targets we performed a multiplex phosphoprotein assay for five proteins (MAP2K1, also known as MEK1; Ser217/Ser221), extracellular signal-regulated kinase 1 and 2 (MAPK3/1, also known as ERK1/ 2; Thr202/Tyr204, Thr185/Tyr187), 90-kDa ribosomal S6 kinase (RPS6KA, also known as p90RSK; Thr359/Ser363), protein kinase B (AKT1, also known as [PKB]/Akt; Ser473), and glycogen synthase kinase 3 (GSK3A/B; Ser21/ Ser9). A corresponding multiplex total protein assay for MAP2K1, MAPK3/1, RPS6KA, and AKT1 was performed on matching samples. Samples were run in duplicate and according to the manufacturer’s instructions (Bio-Rad). Sample data is presented as the median fluorescence intenstity of the phosphoprotein over the corresponding total protein and relative to the time 0 m control for each experiment.

Statistical Analysis Statistical analysis of data was performed using Prism software on normally distributed data using a one-way or two-way ANOVA where appropriate with either a Dunnet or Bonferroni multiple-comparison test respectively. Error bars represent the SD of the mean of three independent experiments performed in triplicate. P , 0.05 as compared with respective controls are considered significant.

RESULTS DCM and EGF Promote EVT Differentiation Through HER1-Dependent Mechanisms The effect of DCM and 10 ng/ml EGF on HER isoform expression in JAR cells was examined using dual immunoflu-


FIG. 1. DCM and EGF induce a switch in HER isoform expression. Confluent, serum-starved JAR monolayers were treated with either (C) DCM, (E) EGF (10 ng/ml), (D) DCM þ AG1478 (2 lM), or (F) EGF þ AG1478 (2 lM) for 24 h. Under serum-free media conditions, JAR cells predominantly express (A) HER1 (red), whereas both DCM and EGF treatments resulted in the (C) downregulation of HER1 and (E) upregulation of HER2 (green). Co-incubation of treatment media with AG1478 prevented this switch in HER isoform expression (D and F). AG1478 treatment alone did not alter HER isoform expression (B) as compared with serum-free controls (A) (n ¼ 3). Rabbit IgG and mouse IgG negative controls are displayed in panels G and H, respectively. Cell nuclei were stained with Hoescht. Bars ¼ 25 lM.

Serum-starved JAR cells were stimulated with either EGF or HBEGF (10 ng/ml) over a 4-h time course. Following treatment, cells were washed with icecold PBS and lysed using a Bio-Rad cell lysis kit according to the manufacturer’s instructions (Bio-Rad, Mississauga, ON). Equal amounts of JAR cell proteins per lane were analyzed by Western blot analysis. Briefly, 60 lg of total protein per sample were added to NUPAGE LDS sample buffer (Invitrogen) with 2.5% b-mercaptoethanol and boiled for 5 min. Proteins were run on precast 3%–8% Tris-Glycein gels (Invitrogen) and transferred to polyvinylidene fluoride membrane (Millipore) at 48C overnight. Membranes were blocked with 5% nonfat milk in Tris-buffered saline with Tween (TBS-T; 10 mM Tris [pH 7.5], 100 mM NaCl, and 0.1% Tween 20) for 1 h at room temperature. Membranes were incubated with either rabbit polyclonal antiphospho HER1, Tyr992, Tyr1045, Tyr1068, or anti-phospho p85 PIK3 (New England Biolabs, ON) at 48C overnight, and then washed and incubated with anti-rabbit horseradish peroxidase (1:2000)-linked secondary antibody (New England Biolabs) for 1 h at room temperature. Membranes were then stripped and reprobed with antibodies against the respective nonphosphorylated antibody (anti-HER1 and Anti p85 PIK3, 1:1000; New England Biolabs). To further assess equal loading of protein samples, blots were stripped for a second time and reprobed with an antibody against the housekeeping protein b-actin (Abcam, Cambridge, MA). Antibody reactions were detected using Bio-Rad WesternC chemiluminescence detection kit (Bio-Rad), followed by detection of chemiluminescence and band analysis using the VersaDoc 6 gel documentation system and Quantity One Software (Bio-Rad).

HER1 AND EVT DIFFERENTIATION

1039 FIG. 2. AG1478 inhibits DCM-mediated HER1/2 regulation. Representative immunofluorescent images from established first trimester placental villous outgrowths treated for 48 h with control SFM (left panel), DCM (middle panel), or DCM þ AG1478 (2 lM; right panel). Serial sections were stained with a panel of antibodies identifying EVT phenotype markers. HLAG was expressed in explant outgrowths (A–C) marking the presence of EVT. DCM downregulated HER1 in the EVT (E) as compared with control explants (D); these effects were prevented by co-incubation with AG1478 (F). Furthermore, DCM treatment upregulated EVT expression of HER2 (H) as compared with control explants (G), and AG1478 inhibited this upregulation (I). Cell nuclei were stained with Hoescht. Negative controls were performed using mouse IgG (J) or rabbit IgG (K) followed by anti-mouse/ anti-rabbit biotinylated antibody and Streptavidin-Alexa 488. Bars ¼ 100 lM. EVT, extravillous trophoblast outgrowth; V, villous. n ¼ 3.

EGF, a HER1 Ligand, Induces Partial Differentiation of EVT The effect of EGF treatment on the expression of EVT phenotype markers was also assessed in the placental villous explant model. Serial sections of control outgrowths marked by HLAG (Fig. 3A) demonstrate high levels of Cx40 throughout the outgrowth (Fig. 3D), restriction of HER1 staining to the

proximal EVT (Fig. 3G), and low levels of HER2 and a1 integrin staining throughout the outgrowth (Fig. 3, J and M). Treatment with EGF had little effect on Cx40 or HER1 expression (Fig. 3, E and H) but resulted in the strong upregulation of both HER2 and a1 integrin throughout the EVT outgrowth (Fig. 3, K and N). The EGF-treated EVT outgrowths extended across the matrigel surface as columns of EVT that maintained cell-cell contacts. Co-incubation of EGF with AG1478 prevented the upregulation of both HER2 and a1 integrin as compared to EGF alone (Fig. 3, L and O; n ¼ 3). HBEGF-Stimulated JAR Cell Migration Is Mediated by HER1 and the PI3K Pathway JAR cell migration assays were conducted using increasing concentrations of HBEGF as chemoattractant in the lower well of a transwell migration assay. HBEGF stimulated a dosedependent and significant 3- to 6-fold increase in migrated JAR cells at 10 and 100 ng/ml (Fig. 4A, P , 0.001; n ¼ 4). Preincubation of the JAR cells with AG1478 resulted in complete inhibition of HBEGF-induced JAR cell migration (Fig. 4B, striped bar, P , 0.001). In contrast, preincubation of JAR cells with AG825 only had a partial inhibitory effect on HBEGF-mediated migration (Fig. 4B, shaded bar, P , 0.01; n ¼ 4). To further assess the downstream effects of the signaling pathways involved in HBEGF-stimulated JAR cell migration, we preincubated JAR cells with a panel of signaling pathway inhibitors against the ERK and MAPK14 (p38), PKC, and PIK3 pathways for 30 min prior to and during 48-h exposure to HBEGF as the chemotactic stimulus in a migration assay. Only


orescence. These data demonstrate that 24-h incubation with either DCM (Fig. 1C) or EGF (Fig. 1E) resulted in downregulation of HER1 (red) and upregulation of HER2 (green) expression as compared to SFM control (Fig. 1A). Coincubation of either DCM (Fig. 1D) or EGF (Fig. 1F) with the HER1 antagonist AG1478 inhibited this switch in HER expression. Incubation of the cells with AG1478 alone did not influence HER isoform expression (Fig. 2B; n ¼ 3). Using the placental villous explant model we show that after 4 days, culture control explants possess extensive EVT outgrowths that are marked by HLAG expression (Fig. 2A). These explants express high levels of HER1 in the villous cytotrophoblast and early proliferative EVT (Fig. 2D), while HER2 is undetectable (Fig. 2G). Treatment of placental villous explants with DCM resulted in the loss of HER1 expression in the EVT outgrowth (Fig. 2E) and upregulation of HER2 (Fig. 2H). Co-incubation of the explants with DCM plus AG1478 inhibited the switch in HER isoform such that cells of the placental villous outgrowth retained HER1 (Fig. 2F) and did not upregulate HER2 (Fig. 2I) as compared to DCM treatment alone. Negative controls using either mouse IgG or rabbit IgG showed no staining (Fig. 2, J and K; n ¼ 3).

1040

WRIGHT ET AL.

inhibition of the PIK3 pathway using 1 lM LY294002 (Fig. 4 C, bold striped bar, P , 0.001; n ¼ 3) had a significant inhibitory effect on HBEGF-mediated JAR cell migration (gray bar; Fig. 4C). EGF and HBEGF Differentially Activate HER1 The time-dependent activation of HER1 signaling following stimulation with either EGF or HBEGF (10 ng/ml) was assessed by Western blotting using a panel of antibodies against phosphorylated tyrosine residues in HER1 (Fig. 5). Blots were probed with either anti-HER1 (Tyr992), anti-HER1


FIG. 3. AG1478 inhibits the effects of EGF on the differentiation of EVT. Representative micrographs of established first trimester placental villous outgrowths treated for 48 h with either control SFM (left panel), EGF (10 ng/ml; middle panel), or EGF þ AG1478 (2 lM; right panel). Serial sections were stained with a panel of antibodies identifying EVT phenotype marker proteins. HLAG was expressed in all explant outgrowths (A– C), marking the presence of EVT. EGF upregulated EVT expression of HER2 (K) and a1 integrin (N) as compared with respective control explants (J and M). AG1478 inhibited this EGF-mediated upregulation (L and O). Neither EGF nor EGF þ AG1478 treatments had significant effects on placental villous outgrowth expression of Cx40 or HER1 compared with control explants (D–I). Cell nuclei were stained with Hoescht. Negative controls were performed by omission of primary antibody followed by anti-mouse or anti-rabbit biotinylated antibody and Streptavidin-Alexa 488 (P and Q, respectively). Bars ¼ 100 lM. EVT, extravillous trophoblast outgrowth; V, villous. n ¼ 3.

(Tyr1045), or anti-HER1 (Tyr1068); each of these respective sites is associated with activation of specific downstream signaling pathways. Blots were also stripped and reprobed with anti-HER1 antibody that detected a single 170-kDa band corresponding to total nonphosphorylated HER1 in all samples (Fig. 5A). Interestingly, EGF (black bars) and HBEGF (gray bars) stimulated differential and time-dependent phosphorylation of the three tyrosine residues. At the tyrosine 992 site, only HBEGF induced band intensity that peaked between 10 and 30 min of stimulation (P , 0.001 vs. EGF, Fig. 5B). In contrast, EGF was the stronger activator of the tyrosine 1045 site where


1041


FIG. 4. A and B) HBEGF-stimulated JAR cell migration is mediated by HER1. A) JAR cells were exposed to increasing concentrations (1–100 ng/ ml) of HBEGF (gray bars) that stimulated a significant increase in numbers of migrated cells. B) Preincubation of the JAR cells for 30 min with either AG4178 (2 lM, striped bar) or AG825 (5 lM, shaded bar) prior to exposure to HBEGF (10 ng/ml gray bar) resulted in inhibition of HBEGFmediated JAR cell invasion. ***P , 0.001, **P , 0.01, *P , 0.05, n ¼ 4. C) Effect of the PI3K inhibitor on HBEGF-stimulated JAR cell migration. JAR cells were preincubated with either 5 lM PD90589 (PD), 100 nM and 50 nM Ro318220 (Ro), 10 lM and 1 lM LY294002 (Ly), or 1 lM SB203580 (SB) prior to 48-h exposure to HBEGF (10 ng/ml) in the lower chamber of a migration assay. HBEGF-mediated migration of JAR cells was significantly inhibited by the PI3K inhibitor LY294002 alone. ***P , 0.001, **P , 0.01, *P , 0.05, n ¼ 4.

band intensity increased between 1 and 4 h (P , 0.001 vs. HBEGF, Fig. 5C). Though the tyrosine 1068 site was phosphorylated by both EGF and HBEGF, as demonstrated by an increase in band intensity above their respective time zero controls, HBEGF was the stronger activator and led to more sustained phosphorylation across the time course (15 min: P , 0.05, 1 h: P , 0.001, Fig. 5D; n ¼ 3).

FIG. 5. Differential tyrosine phosphorylation of HER1 by EGF and HBEGF. JAR cells were stimlulated with either EGF (10 ng/ml) or HBEGF (10 ng/ml) across a 4-h time course. A) Representative photomicrographs of Western blotting using a anti-total HER1 antibody detected a 170-kDa band corresponding to the known molecular weight of HER1 that did not change in intensity across the time course in response to EGF or HBEGF treatment. In contrast, the anti-phosphotyrosine antibodies against the Tyr 992, Tyr 1045, or the Tyr 1068 phosphorylation sites of HER1 detected bands that displayed time-dependent phosphorylation patterns that differed between EGF (black bars) and HEGF (gray bars) treatment. B) When expressed as a ratio over the corresponding total nonphosphorylated HER1 bands, only HBEGF induced a significant increase in phosphorylation of the Tyr 992 site between 10 and 30 min. C) At the Tyr 1045 site, only EGF increased phosphorylation status above the respective control at 1 h and 4 h. D) While both EGF and HBEGF mediated an increase in phosphorylation of the Tyr 1068 site above the time 0 control, HBEGF lead to a stronger and more sustained activation at 1 h. *P , 0.05, **P , 0.01, ***P , 0.001, n ¼ 3.

EGF and HBEGF Differentially Activate the MAPK and PIK3 Signaling Pathways Next we assessed the ability of EGF and HBEGF to induce activation of either the MAPK or PIK3 signaling pathways that are known to be downstream of the HER1 tyrosine 992 or

1042

WRIGHT ET AL.

tyrosine 1068 phosphorylation sites. Bioplex multiplex phosphorylation ELISA assays were performed using lysates of JAR cells that had been stimulated with either EGF or HBEGF over a 4-h time course to simultaneously assess the phosphorylation of MAP2K1 (Ser217/Ser221), MAPK3/1 (Thr202/Tyr204, Thr185/Tyr187), RPS6KA (Thr359/Ser363), AKT1 (Ser473), and GSK3A/B (Ser21/Ser9). Multiplex assays for the total nonphosphorylated proteins were also performed on matching samples except for GSK3A/B, for which a Western blot for b-actin was used to standardize phosphoprotein levels. PIK3 activity was also assessed by Western blotting using anti-phospho-p85 PIK3 and total p85 PIK3 antibodies (see Supplemental Fig. S1, available online at www.biolreprod. org, for representative blot). Figures 6 and 7 demonstrate the sequential activation of the MAPK pathway and the PIK3 pathway, respectively. Activation of the MAPK pathway was achieved within 2 min of treatment with EGF and HBEGF, as

FIG. 7. EGF and HBEGF differentially activate the PIK3 pathway. JAR cells were stimulated with either EGF (10 ng/ml) or HBEGF (10 ng/ml) across a 4-h time course. A) Activation of PIK3 was assessed by Western blotting using a phospho p85 PIK3 antibody that detected a single 85-kDa (Tyr458) band (Supplemental Fig. S1) that demonstrated a significant increase in optical density by 30 min of HBEGF treatment and was sustained over the 1 h and 4 h time points when corrected for total p85 PIK3 levels. EGF had no effect above the time 0 control. B) Phospho AKT (Ser473) and C) phospho GSK3A/B (Ser21/9) levels were assessed by Bioplex multiplex phosphoprotein and total ELISA assay and demonstrated a significant increase above time 0 control levels across the 2 min to 1 h time points following either EGF or HBEGF treatment. However, HBEGF stimulated a stronger activation of AKT and a more sustained activation of GSK3A/B at 1 h and 4 h. GSK3A/B levels are expressed as a ratio of the optical density of b-actin levels in the same samples as assessed by Western blotting, as there is no available total protein control in the Bioplex assay. *P , 0.05, **P , 0.01, ***P , 0.001, n ¼ 3.

shown by the phosphorylation of MAP2K1, MAPK3/1, and RPS6KA (Fig. 6, A–C). EGF-induced activation of this pathway was transient, reaching a maximum at 2 min for MAP2K1 and MAPK3/1 and 10 min for RPS6KA; all decreased at 1 and 4 h. In contrast, HBEGF induced a stronger activation of MAP2K1 from 10 min through to 4 h of treatment (Fig. 6A, P , 0.01); similarly, levels of both MAPK3/1 and RPS6KA phosphorylation were maintained at the 1-h and 4-h time points (Fig. 6, B and C, P , 0.001). The p85 PIK3 pathway was also more strongly activated by HBEGF than EGF (Fig. 7). Only HBEGF significantly increased levels of phosphorylated PIK3 p85 protein between 30 min and 4 h (Fig. 7A, P , 0.001) and had a more pronounced and sustained effect on the phosphorylation status of AKT1 between the 2 min and 1 h time points (Fig. 7B, P , 0.001) and GSK3A/B at 1 h and 4 h (Fig. 7C, P , 0.05) (n ¼ 3).


FIG. 6. EGF and HBEGF differentially activate the MAPK pathway. JAR cells were stimulated with either EGF (10 ng/ml) or HBEGF (10 ng/ml) across a 4-h time course. Activation of the MAPK pathway was assessed by Bioplex multiplex phosphoprotein and total ELISA for MAP2K1, MAPK3/1, and RPS6KA. Both EGF (black bars) and HBEGF (gray bars) induced a rapid increase in the phosphorylation of MAP2K1 (A), MAPK3/1 (B), and RPS6KA (C) above the time 0 control. EGF-mediated phosphorylation of MAP2K1 and MAPK3/1 peaked at 2 min whereas RPS6KA reached a peak at 10 min; all levels gradually decreased to control levels by 4 h. In contrast, HBEGF mediated a stronger and more sustained activation of the MAP2K1, MAPK3/1, and RPS6KA proteins that increased above EGF levels at both the 1 h and 4 h time points. Data is expressed as the median fluorescence intensity and is relative to the respective time 0 control for either EGF or HBEGF. *P , 0.05, **P , 0.01, ***P , 0.001, n ¼ 3.


DISCUSSION

further suggested that differentiation to the full invasive EVT phenotype involved other members of the EGF family found in DCM that can bind HER1, such as HBEGF, AREG, or TGF (data not shown). Indeed, when JAR cells were exposed to increasing concentrations of HBEGF as the chemotactic ligand, we observed a significant dose-dependent increase in cell migration. Moreover, this HBEGF-induced migration, similar to that induced by DCM, was completely blocked by AG1478 inhibition of HER1 signaling. Interestingly, the HER2 antagonist AG825 only affected a partial inhibitory effect on JAR cell migration, suggesting that HER1 may dimerise with another receptor, such as HER4, to mediate EVT invasion. This is supported by the demonstration by Leach et al. [31] that HBEGF promoted EVT differentiation in a manner that is mediated by either HER1 or HER4, as only a combination of both antibodies inhibited integrin switching. Ligand binding of HERs results in receptor dimerization, autophosphorylation, activation of downstream signaling, and lysosomal degradation [32, 33]. Our investigation of the differential phosphorylation status of HER1 in response to stimulation with either EGF or HBEGF demonstrated rapid and specific phosphorylation of the HER1 Tyr992 residue by HBEGF alone. Phosphorylation of this site is known to recruit the SH2 domain of PLCG and lead to downstream signaling through MAPK/ERK [34]. Indeed, we further demonstrated that HBEGF treatment led to the sequential activation of the MAP2K1, MAPK3/1, and RPS6KA pathway. Though EGF also activated this pathway, its activity was transient and not maintained past 1 h. In contrast, EGF treatment did lead to a time-dependent increase in phosphorylation of the Tyr1045 site, which is associated with recruitment of CBL and receptor ubiquitination and degradation [35]. This result is supported by the loss of HER1 immunostaining following EGF treatment of either JAR cells or primary EVT outgrowths. Both EGF and HBEGF also induced phosphorylation of the HER1 Tyr1068 site, although again HBEGF was the stronger activator. This site is known to recruit the second messenger GRB2 and lead to the downstream activation of PIK3 and AKT1 [36]. Indeed, our data showed that HBEGF alone stimulated a significant and sequential time-dependent increase in the p85 (Tyr458) subunit of PIK3 at 30 min, which was sustained over the 4-h treament point and translated to a sustained downstream phosphorylation of AKT1 (Ser473) and GSK3A/B at 4 h. AKt1 and phosphoGSK have both been localized to the leading edge of migrating cells and have been shown to influence the motility of a number of cell types [37, 38]. Interestingly, when we assessed the effect of either MAPK, PKC, or PIK3 signaling pathway inhibitors on HBEGF-mediated JAR cell migration, only the PIK3 inhibitor (LY294002) had a significant inhibitory effect. These data are supported by other studies similarly showing that EGF treatment of either HTR8 SVneo or SGHPL-4 EVT cell lines also resulted in PIK3-dependent trophoblast migration [39, 40]. However, these studies also showed EGF stimulated activation of the ERK and MAPK14 (p38) pathways and indicated their involvement in cell motility [39, 40]. One event that should be noted in the current study is that while the phospho p85 PIK3 increased at 30 min, phospho AKT1 levels rapidly increased above baseline within 2 min of either EGF or HBEGF treatment. This disparity may be due to the differences in sensitivity of the detection methods used, as Western blot is less sensitive than ELISA-based analysis. However, there is a possibility of alternative noncanonical activation of AKT through crosstalk with the MAP kinase pathways. It has been shown that both MAP2K1 and MAPK14 can activate AKT on the ser473 residue in ovarian cancer cell lines [41]. Similarly, inhibition of MAPK14 inhibited EGF-mediated AKT activa-


The results from this study support an integral role for HER1 signaling in EVT differentiation. Using both JAR cell and placental villous explant models, we demonstrate that inhibition of HER1 signaling using the specific antagonist AG1478 prevented DCM-induced changes in protein expression associated with EVT differentiation, such as the switch from HER1 to HER2 expression. These results led us to suggest that HER1 signaling constitutes an important aspect of a cascade of events leading to EVT differentiation. From these findings, we expected that we would be able to replicate the differentiation effects of DCM by inducing HER1 signaling directly with EGF. Indeed we show that EGF, like DCM, induced the downregulation of HER1 and upregulation of HER2 in JAR cells, and this was dependent on HER1 activation, as the HER1 antagonist completely blocked the EGF effect. However, EGF treatment (10 ng/ml) of placental villous explants induced only a subset of the changes associated with the differentiation of invasive EVT: HER1 expression was downregulated and both HER2 and a1 integrin expression were upregulated. Notably, however, EGF treatment neither resulted in the downregulation of Cx40 protein in the placental villous outgrowth nor did it produce the outgrowth morphology characteristic of DCM-treated explants: rather than detaching from the outgrowth and extending along the surface of the matrigel, the cells of EGF-treated placenta outgrowths formed continuous streams and did not separate from one another. Interestingly, this observation correlates to our previous findings that EGF, unlike DCM, is unable to induce JAR cell invasion in a migration assay [9]. These latter observations appear to contradict our earlier findings that loss of Cx40 expression is a prerequisite for trophoblast cells to become migratory. However, there are systems in which the coordinated migration of a population of cells depends on connexin expression. One such example is murine heart development, in which cardiac crest cells express Cx43 as they migrate from the neural crest to their destination in the heart. Lo et al. [25], Huang et al. [26], and Waldo et al. [27] have demonstrated that deletion of Cx43 results in embryonic lethality due to major heart defects resulting from the failure of neural cardiac crest cell migration. It is possible that the retention of Cx40 expression in EGF-treated explants is responsible for their specific migratory population phenotype since cells connected by gap junction channels may not be able to ‘‘let go’’ of each other. The downregulation of Cx40 may, therefore, be the crucial physiological event for the cells to begin to migrate away from each other and penetrate the maternal decidua. Nevertheless, it is notable that despite the failure to downregulate Cx40, EGF treatment was still capable of inducing a1 integrin expression in a HER1-dependent manner in the EVT outgrowth of the placental villous explant model. It appears that integrin and HER1 signaling are closely linked; indeed, integrin-induced activation of HER1 and phosphorylation of a subset of tyrosine phospho-acceptor sites on HER1 have been shown to lead to recruitment and tyrosine phosphorylation of SHC, PLCG, the p85 subunit of PIK3, and CBL, and leads to subsequent activation of the downstream targets ERK and AKT1 in epithelial cells [15]. It is, therefore, possible that EVT differentiation is regulated by more than one HER signaling cascade. The HER signaling system is also capable of pleiotropic responses dependent on the identity of the activating HER ligand [28–30]. Our data showing that DCM-mediated upregulation of HER-2 in JAR cells was not completely blocked by AG4178 treatment, whereas that of EGF was,

1043

1044

WRIGHT ET AL.

ACKNOWLEDGMENTS The authors thank the donors and the Research Centre for Women’s and Infants’ Health BioBank Program of the CIHR Group in Development and Fetal Health, the Samuel Lunenfeld Research Institute, and the MSH/ UHN Department of Obstetrics & Gynecology for the human specimens used in this study.

REFERENCES 1. Benirschke K. Pathology of the Human Placenta. New York: Springer; 2006. 2. Naicker T, Khedun SM, Moodley J, Pijnenborg R. Quantitative analysis of trophoblast invasion in preeclampsia. Acta Obstet Gynecol Scand 2003; 82:722–729. 3. Brosens JJ, Pijnenborg R, Brosens IA. The myometrial junctional zone spiral arteries in normal and abnormal pregnancies: a review of the literature. Am J Obstet Gynecol 2002; 187:1416–1423. 4. Chakraborty C, Gleeson LM, McKinnon T, Lala PK. Regulation of human trophoblast migration and invasiveness. Can J Physiol Pharmacol 2002; 80:116–124. 5. Winterhager E, Von Ostau C, Gerke M, Gruemmer R, Traub O, Kaufmann P. Connexin expression patterns in human trophoblast cells during placental development. Placenta 1999; 20:627–638. 6. Enders AC, Blankenship TN, Fazleabas AT, Jones CJ. Structure of anchoring villi and the trophoblastic shell in the human, baboon and macaque placenta. Placenta 2001; 22:284–303. 7. Jokhi PP, King A, Loke YW. Reciprocal expression of epidermal growth factor receptor (EGF-R) and c-erbB2 by non-invasive and invasive human trophoblast populations. Cytokine 1994; 6:433–442. 8. Damsky CH, Librach C, Lim KH, Fitzgerald ML, McMaster MT, Janatpour M, Zhou Y, Logan SK, Fisher SJ. Integrin switching regulates normal trophoblast invasion. Development 1994; 120:3657–3666. 9. Wright JK, Dunk CE, Perkins JE, Winterhager E, Kingdom JC, Lye SJ. EGF modulates trophoblast migration through regulation of Connexin 40. Placenta 2006; 27(suppl A):S114–S121. 10. Ellis AG, Doherty MM, Walker F, Weinstock J, Nerrie M, Vitali A, Murphy R, Johns TG, Scott AM, Levitzki A, McLachlan G, Webster LK, et al. Preclinical analysis of the analinoquinazoline AG1478, a specific small molecule inhibitor of EGF receptor tyrosine kinase. Biochem Pharmacol 2006; 71:1422–1434. 11. Carpenter G. Epidermal growth factor: biology and receptor metabolism. J Cell Sci Suppl 1985; 3:1–9.

12. Carpenter G. EGF: new tricks for an old growth factor. Curr Opin Cell Biol 1993; 5:261–264. 13. Carpenter G. The EGF receptor: a nexus for trafficking and signaling. Bioessays 2000; 22:697–707. 14. Yu D, Hamada J, Zhang H, Nicolson GL, Hung MC. Mechanisms of cerbB2/neu oncogene-induced metastasis and repression of metastatic properties by adenovirus 5 E1A gene products. Oncogene 1992; 7:2263– 2270. 15. Bill HM, Knudsen B, Moores SL, Muthuswamy SK, Rao VR, Brugge JS, Miranti CK. Epidermal growth factor receptor-dependent regulation of integrin-mediated signaling and cell cycle entry in epithelial cells. Mol Cell Biol 2004; 24:8586–8599. 16. Lala PK, Hamilton GS. Growth factors, proteases and protease inhibitors in the maternal-fetal dialogue. Placenta 1996; 17:545–555. 17. Bass KE, Morrish D, Roth I, Bhardwaj D, Taylor R, Zhou Y, Fisher SJ. Human cytotrophoblast invasion is up-regulated by epidermal growth factor: evidence that paracrine factors modify this process. Dev Biol 1994; 164:550–561. 18. Chobotova K, Spyropoulou I, Carver J, Manek S, Heath JK, Gullick WJ, Barlow DH, Sargent IL, Mardon HJ. Heparin-binding epidermal growth factor and its receptor ErbB4 mediate implantation of the human blastocyst. Mech Dev 2002; 119:137–144. 19. Lysiak JJ, Han VK, Lala PK. Localization of transforming growth factor alpha in the human placenta and decidua: role in trophoblast growth. Biol Reprod 1993; 49:885–894. 20. Lysiak JJ, Johnson GR, Lala PK. Localization of amphiregulin in the human placenta and decidua throughout gestation: role in trophoblast growth. Placenta 1995; 16:359–366. 21. Hellmann P, Winterhager E, Spray DC. Properties of connexin40 gap junction channels endogenously expressed and exogenously overexpressed in human choriocarcinoma cell lines. Pflugers Arch 1996; 432: 501–509. 22. Filla MS, Kaul KL. Relative expression of epidermal growth factor receptor in placental cytotrophoblasts and choriocarcinoma cell lines. Placenta 1997; 18:17–27. 23. Genbacev O, Schubach SA, Miller RK. Villous culture of first trimester human placenta—model to study extravillous trophoblast (EVT) differentiation. Placenta 1992; 13:439–461. 24. Caniggia I, Grisaru-Gravnosky S, Kuliszewsky M, Post M, Lye SJ. Inhibition of TGF-beta 3 restores the invasive capability of extravillous trophoblasts in preeclamptic pregnancies. J Clin Invest 1999; 103:1641– 1650. 25. Lo CW, Waldo KL, Kirby ML. Gap junction communication and the modulation of cardiac neural crest cells. Trends Cardiovasc Med 1999; 9: 63–69. 26. Huang GY, Cooper ES, Waldo K, Kirby ML, Gilula NB, Lo CW. Gap junction-mediated cell-cell communication modulates mouse neural crest migration. J Cell Biol 1998; 143:1725–1734. 27. Waldo KL, Lo CW, Kirby ML. Connexin 43 expression reflects neural crest patterns during cardiovascular development. Dev Biol 1999; 208: 307–323. 28. Sweeney C, Lai C, Riese DJ II, Diamonti AJ, Cantley LC, Carraway KL III.. Ligand discrimination in signaling through an ErbB4 receptor homodimer. J Biol Chem 2000; 275:19803–19807. 29. Heldin CH. Dimerization of cell surface receptors in signal transduction. Cell 1995; 80:213–223. 30. Moghal N, Sternberg PW. Multiple positive and negative regulators of signaling by the EGF-receptor. Curr Opin Cell Biol 1999; 11:190–196. 31. Leach RE, Kilburn B, Wang J, Liu Z, Romero R, Armant DR. Heparinbinding EGF-like growth factor regulates human extravillous cytotrophoblast development during conversion to the invasive phenotype. Dev Biol 2004; 266:223–237. 32. Hackel PO, Zwick E, Prenzel N, Ullrich A. Epidermal growth factor receptors: critical mediators of multiple receptor pathways. Curr Opin Cell Biol 1999; 11:184–189. 33. Zwick E, Hackel PO, Prenzel N, Ullrich A. The EGF receptor as central transducer of heterologous signalling systems. Trends Pharmacol Sci 1999; 20:408–412. 34. Emlet DR, Moscatello DK, Ludlow LB, Wong AJ. Subsets of epidermal growth factor receptors during activation and endocytosis. J Biol Chem 1997; 272:4079–4086. 35. Levkowitz G, Waterman H, Ettenberg SA, Katz M, Tsygankov AY, Alroy I, Lavi S, Iwai K, Reiss Y, Ciechanover A, Lipkowitz S, Yarden Y. Ubiquitin ligase activity and tyrosine phosphorylation underlie suppression of growth factor signaling by c-Cbl/Sli-1. Mol Cell 1999; 4:1029– 1040.


tion in the SGHPL4 cell line [40]. This suggests that there may be cross talk between these pathways in the early stages of signaling activation in the Jar cell line as well. However, as only inhibition of the PIK3 pathway had any effect of HBEGFmediated JAR cell, migration we conclude that HBEGFmediated JAR cell migration is mediated by HER1 tyr1068 phosphorylation and sustained downstream activation of the PIK3 and AKT1 signaling pathways. Recently it has also been shown that in HTR8 SVneo and SW.71 trophoblast cell lines, HBEGF-mediated cell migration is mediated by PIK3, MAPK14, ERK, and JNK pathways but with no apparent cross talk between pathways [42]. However, no direct assessment of either MAP2K1 or PIK3 activation was made in this study, and the earliest time point studied was 15 min [42]. The differences of these studies with our findings may be due to the demonstrated differential activities of both EGF and HBEGF and the inherent differences in the invasive capability and/or relative levels of HER1 expression of the trophoblast cell lines used in each study. In our studies we are unable to induce either JAR or primary EVT cell motility with EGF alone and require either HBEGF or DCM treatment to induce JAR cell migration or primary EVT invasion. In conclusion, using both explant EVT and JAR cell models and distinct HER1 ligands we have dissected the HER1 signaling pathway and shown that HBEGF-mediated trophoblast migration is specifically mediated by HER1 (Tyr1068) activation of the PIK3/AKT1 pathway.

HER1 AND EVT DIFFERENTIATION 36. Sampaio C, Dance M, Montagner A, Edouard T, Malet N, Perret B, Yart A, Salles JP, Raynal P. Signal strength dictates phosphoinositide 3-kinase contribution to Ras/extracellular signal-regulated kinase 1 and 2 activation via differential Gab1/Shp2 recruitment: consequences for resistance to epidermal growth factor receptor inhibition. Mol Cell Biol 2008; 28:587– 600. 37. Byzova TV, Goldman CK, Pampori N, Thomas KA, Bett A, Shattil SJ, Plow EF. A mechanism for modulation of cellular responses to VEGF: activation of the integrins. Mol Cell 2000; 6:851–860. 38. Kim D, Kim S, Koh H, Yoon SO, Chung AS, Cho KS, Chung J. Akt/PKB promotes cancer cell invasion via increased motility and metalloproteinase production. FASEB J 2001; 15:1953–1962. 39. Qiu Q, Yang M, Tsang BK, Gruslin A. Both mitogen-activated protein kinase and phosphatidylinositol 3-kinase signalling are required in

1045

epidermal growth factor-induced human trophoblast migration. Mol Hum Reprod 2004; 10:677–684. 40. LaMarca HL, Dash PR, Vishnuthevan K, Harvey E, Sullivan DE, Morris CA, Whitley GS. Epidermal growth factor-stimulated extravillous cytotrophoblast motility is mediated by the activation of PI3-K, Akt and both p38 and p42/44 mitogen-activated protein kinases. Hum Reprod 2008; 23:1733–1741. 41. Baudhuin LM, Cristina KL, Lu J, Xu Y. Akt activation induced by lysophosphatidic acid and sphingosine-1-phosphate requires both mitogen-activated protein kinase kinase and p38 mitogen-activated protein kinase and is cell-line specific. Mol Pharmacol 2002; 62:660–671. 42. Jessmon P, Kilburn BA, Romero R, Leach RE, Armant DR. Functionspecific intracellular signaling pathways downstream of heparin-binding EGF-like growth factor utilized by human trophoblasts. Biol Reprod 2010; 82:921–929.
