Placenta
Volume 31, Issue 1 , Pages 67-74, January 2010

EGF stimulates proliferation in the bovine placental trophoblast cell line F3 via Ras and MAPK

Department of Anatomy, University of Veterinary Medicine Hannover, Bischofsholer Damm 15, 30173 Hannover, Germany

Accepted 22 October 2009. published online 16 November 2009.

Article Outline

Abstract 

In the bovine placenta, multinucleate trophoblast giant cells (TGC), evolving from uninucleate trophoblast cells, are crucial for feto-maternal interaction as they show endocrine activity and the ability to migrate and fuse with caruncular epithelial cells. In contrast to caruncular epithelial cells, the isolation and culture of bovine trophoblast cells is complicated because they cease to express their specific products, like placental lactogen (PL), during prolonged culture. In the present study, we aimed to establish a bovine cotyledonary trophoblast cell line targeting our long term goal to develop an in vitro model for the bovine placenta. Therefore, the functional activity of important signalling pathways was tested. Primary trophoblast cells were isolated from a bovine cotyledon of a male fetus and successfully subcultured and cryopreserved. The obtained cell line, termed F3, showed epithelial morphology and characteristic binuclear giant cells in small numbers through all passages. The trophoblastic origin of F3 cells was verified by amplification of a Y-chromosome specific DNA-sequence and the presence of PL mRNA. Immunofluorescence demonstrated that F3 cells were continuously positive for zonula occludens-2 (ZO-2), cytokeratin and vimentin, whereas they expressed the TGC specific marker PL only in the first two passages. F3 cell growth was accelerated in medium supplied with epidermal growth factor (EGF). EGF-stimulated proliferation was mediated through activation of Ras and the phosphorylation of mitogen-activated protein kinase (MAPK) 42 and 44. In conclusion, the F3 cell line shows several in vivo characteristics of bovine cotyledonary trophoblast cells. The response to EGF stimulation indicates that EGF plays a role during bovine placentation, and illustrated that F3 cells may provide a valuable tool for further mechanistic studies elucidating the feto-maternal interplay.

Keywords: Bovine, Trophoblast cell line, Epidermal growth factor, Ras, Mitogen-activated protein kinase

 

Back to Article Outline

1. Introduction 

In the cow, the main sites of feto-maternal interactions are placentomes, which are formed by maternal caruncles and fetal cotyledons. A unique feature of the synepitheliochorial bovine placenta are trophoblast giant cells (TGC) which are formed by acytokinetic mitosis [1]. These cells are crucial for the feto-maternal interaction as they migrate and fuse with single uterine epithelial cells, releasing a variety of products, like placental lactogen, into the maternal compartment. The differentiation, genome multiplication, endocrine activities, migration and fusion are common features of bovine binucleate cells and trophoblastic giant cells in other species [2]. To date, the control mechanisms of these highly regulated processes are poorly understood. To develop a suitable in vitro model, cell culture systems are needed for both, uterine epithelial and trophoblast cells. While the isolation, characterisation and long-term culture of caruncular epithelial cells have been successfully achieved [3], [4], the establishment of a trophoblastic cell line has proven to be more complex. Even though previous studies have dealt with the isolation of bovine trophoblast cells from either cotyledons [5], [6] or blastocysts [7], [8] only the latter two groups established permanent lines. Both were shown to express bIFN-tau as a marker for trophoblast cells. Additionally, the bovine blastocyst-derived trophoblast cell line 1 (BT-1) [8] displayed a small number of binucleated, PL positive cells in each passage, suggesting spontaneous [8] and substrate dependent differentiation of BT-1 cells into TGC [9]. Nevertheless, trophoblast cells derived from blastocysts could have other properties than the more differentiated ones from the bovine cotyledon. As far as the isolation of the latter cells is concerned it has been demonstrated that uni- and binucleated bovine trophoblasts can be successfully isolated and cultured over several days [3], [5]. However, these studies failed to properly characterise the obtained cells since a complete manual separation of fetal cotyledon and maternal caruncle has been shown to be impossible [3], and later on in culture trophoblast cells cease to express their specific markers like certain hormonal products [5]. Furthermore, characterisation of isolated cotyledonary cells was done solely according to the phenotype and the appearance of binucleated cells [10] or the isolated cells were not correlated to a specific cell type at all [11]. The phenotype is an insufficient criterion since both maternal caruncular epithelium and trophoblast cells are indeed epithelial cells which express the same markers [12]. Because of the obvious lack of appropriate trophoblast-specific markers it has been suggested to utilize the Y-chromosome. This is an elegant solution requiring that the donor fetus is male. The origin of the isolated cells can be efficiently demonstrated by detecting the Y-chromosome by fluorescence in situ hybridization [3] or PCR.

In addition to identification of the cell type, functional cell properties are an important issue. An earlier study demonstrated the positive influence of epidermal growth factor (EGF) on the growth of isolated bovine trophoblast cells [6]. EGF is a multifunctional growth factor which binds to cell surface receptors with intrinsic tyrosine kinase activity, thereby initialising a variety of signalling pathways leading to proliferation, migration and differentiation depending on the cell type and condition [13], [14]. The role in implantation and placentation is supported by reports describing EGF production in the preimplantative uterus in different species [15], [16]. Besides, EGF signalling is involved in the outgrowth of murine trophoblasts [17] and enhances maturation of human first trimester cytotrophoblasts [14]. Furthermore EGF reduces apoptosis in these cells through activation of the mitogen-activated protein kinases 42 and 44 (MAPK42/44) and the Phosphoinositide 3-kinase (PI-3K) pathway [18]. According to these findings the activation of pro-proliferative signalling pathways by EGF can be considered as an indication for a functional trophoblast cell population.

In this study we report on the establishment of a bovine placental trophoblast cell line (F3) of confirmed origin. As we have demonstrated that the proliferative effect observed upon EGF stimulation in vitro is due to an activation of the Ras–MAPK pathway, we hypothesized that EGF in vivo is an important factor during bovine placentation. Thus F3 cells may serve as a valuable tool to elucidate the specific role of different factors during bovine trophoblast growth and differentiation.

Back to Article Outline

2. Methods 

2.1. Materials 

Hank's Salt Solution without Ca2+ and Mg2+, Trypsin, Dulbecco's Modified Eagle Medium (DMEM)/Ham's F12, Penicillin/Streptomycin and Glutamine were obtained from PAA (Cölbe, Germany), fetal calf serum (FCS) and Percoll from Biochrom (Berlin, Germany), human recombinant EGF from Biomol (Hamburg, Germany) and protease and phosphatase inhibitor cocktail from Sigma (Munich, Germany). Cell culture dishes and flasks were obtained from TPP (Trasadingen, Switzerland). Antibodies, used in the indicated dilution, were obtained from the following sources: mouse anti-phosphoMAPK (M8159, 1:12,500) and mouse anti-beta-tubulin (A5441, 1:200) – Sigma (Munich, Germany); rabbit anti-Ras (3965, 1:2000) and rabbit anti-ZO-2 (2847, 1:100) – New England Biolabs (Beverly, USA); rabbit anti-pan-cytokeratin (Z0622, 1:800) – Dako (Hamburg, Germany); mouse anti-beta-actin (sc-47778, 1:5000) – Santa Cruz (Heidelberg, Germany); anti-mouse horse radish peroxidise (HRP; 1:750) and anti-rabbit HRP (1:750) – Pierce (Rockford, USA); Hoechst 33342 (H1399, 1:10,000), anti-mouse Alexa 488 (1:2000) and anti-rabbit Alexa 594 (1:2000) – Invitrogen (Karlsruhe, Germany). Rabbit Anti-oPL was a kind gift from Jean-Luc Servely, INRA, Jouy-en-Josas, France.

Primers were purchased form Eurofins MWG (Ebersberg, Germany) and PCR kits and reagents from Promega (Mannheim, Germany). All other agents, if not otherwise indicated, were obtained from Sigma (Munich, Germany).

2.2. Isolation and culture of bovine trophoblast cells 

A pregnant uterus (bos taurus) was obtained from the local abattoir. By measuring crown rump length the age of the fetus was determined to be approx. 5 months of gestation. Placentomes were manually separated into fetal cotyledon and maternal caruncle. The fetal tissue was tied up with villi facing outward, washed twice in PBS and immersed in 10ml trypsin solution (0.25% trypsin, 0.01mg/ml DNase1 in Hank's Salt Solution without Ca2+ and Mg2+). After the first incubation (10min, 37°C, shaking) the tissue was transferred into a new batch of trypsin solution while the digested cells were kept on ice. In total the tissue was digested three times each for 10min. The dispersed cells were pooled, washed (centrifugation for 5min at 200g) with full supplemented medium (FSM; Dulbecco's Modified Eagle Medium (DMEM)/Ham's F12 containing 10% fetal calf serum (FCS) 100IU/ml Penicillin, 100μg/ml Streptomycin and 2mM Glutamine) and filtered through 100μm meshes (BD, Heidelberg, Germany). The filtrate was layered on a Percoll cushion (1077g/ml) and centrifuged for 20min at 800g (25°C) without brake to eliminate the erythrocytes. Cells at the interface were collected and washed three times with FSM. The suspension was plated on uncoated culture flasks and kept at 37°C in an atmosphere of 5% CO2. Half of the medium was replaced every 2 days. A further purification of trophoblast cells was achieved by controlled incubation with trypsin/EDTA (0.25%/2mM). Subculturing was performed as soon as confluence was reached using trypsin/EDTA until passage 45. For cryopreservation, cells were resuspended in a solution containing 10% dimethylsulphoxide (DMSO), 20% FCS and 70% DMEM/Ham's F12.

2.3. Collagen gels 

For collagen gel preparation type 1 rat tail collagen (SERVA Electrophoresis, Heidelberg) was used according to the manufacturer's information. The final collagen concentration in the gel was 1.7mg/ml.

2.4. Fluorescence-activated cell sorting (FACS) 

Detection of apoptosis by FACS was performed using the Annexin V-FITC apoptosis detection kit 1 (BD, Heidelberg, Germany). The staining was performed according to the producer's manual. Analysis was run on an FACS Calibur instrument and CellQuest software (Becton Dickinson, Germany, Heidelberg). Ten thousand events were collected for each sample and the analysis of whole cells was performed using appropriate scatter gates to exclude cellular debris and aggregates. The analysis was performed with cells derived from three different isolations.

2.5. Immunofluorescence 

For immunofluorescence, 2×104 cells were seeded onto cover slips and cultured until 50% confluence was reached. Cells were fixed with 4% paraformaldehyde in TBS containing 1% sucrose for 20min. Cryosections of placentomes served as positive controls and were fixed in 100% Methanol (10min, −20°C). Unspecific antibody binding and autofluorescence were blocked by 3% BSA in TBS (10mM Tris pH 8.0, 150mM NaCl) with 0.05% Tween20 (TBST) for 1h. All washing steps were performed in TBST, while incubation with primary antibodies was carried out according to supplier's suggestions. As secondary antibodies mouse Alexa 488 and rabbit Alexa 546 were used and the nuclei were stained for 5min with Hoechst. After the final washing steps, cells and sections were mounted onto slides with ProLong Antifade (Invitrogen, Karlsruhe, Germany) and viewed under a Zeiss Axiovert 200M fluorescence microscope (Jena, Germany). All stainings were carried out in triplicates.

2.6. DIL-Ac-LDL-uptake and isolation of bovine umbilical vein endothelial cells (BUVEC) 

Cells used for LDL labelling were seeded into 24-well plates containing cover slips. After 2 days, the cells were incubated for 4h (37°C) with regular culture medium containing 10μg/ml human acetylated LDL labeled with the fluorescent probe, 1,1′-dioctaldecyl-3,3,3′,3′-tetramethylindocarbocyanine perchlorate (Dil-Ac-LDL; Harbor Bio-Products, Norwood, U.S.A.). After washing with PBS, the cells were fixed with 2% Paraformaldehyde (in PBS), stained with Hoechst and mounted as described above. As a positive control bovine umbilical vein endothelial cells were isolated and cultured as described elsewhere [19]. The experiment was repeated three times.

2.7. Scanning electron microscopy (SEM) 

SEM was performed with confluent cultures of F3 and BCEC-1 cells as described in Ref. [4]. The monolayer was fixed in buffer containing 0.1M cacodylate, 4% paraformaldehyde and 3.5% glutaraldehyde. After dehydration in ethanol followed by isoamylacetate, the cells were Critical Point dried and sputter-coated with gold. The membrane was mounted to the sample holder and viewed using a scanning electron microscope (Zeiss DSM 940, Jena, Germany). Samples were assessed for the presence of apical microvilli.

2.8. Y-chromosome specific PCR 

Genomic DNA was isolated from F3 cells, the bovine caruncular epithelial cell line (BCEC-1) and tissue with Wizard Genomic DNA purification kit according to the manufacturer's protocol. As positive control genomic DNA from the liver of a male fetus was used, maternal DNA (from BCEC-1 cells) served as negative control. For amplification of a 163-bp region of a Y-chromosome specific sequence the following primers were used: 5′-TGAACGCTTTCATTGTGTGGTC-3′ (5′ upstream primer) and 5′-GCCAGTAGTCTCTGTGCCTCCT-3′ (3′ downstream primer) [20]. The PCR was carried out with GoTaq® Hot Start Green Master Mix Kit. PCR conditions were as follows: heating top 95°C, 95°C for 4min, 40 cycles of denaturation at 94°C (45 s), annealing at 59°C (45s) and extension at 72°C (45s), final extension at 72°C for 10min. PCR products were separated using a 2% agarose gel and visualized by ethidium bromide. The experiment was repeated three times.

2.9. Reverse Transcription Polymerase Chain Reaction (RT-PCR) 

Specific mRNAs of PL and GAPDH were amplified from bovine cotyledon homogenate (positive control) as well as from F3 cells and mesenchymal cells derived from the bovine cotyledon (negative control) by a Two-Step RT-PCR. Total RNA was extracted from tissue and cells using the SV Total RNA Isolation System in accordance with the manufacturer's protocol. cDNA was synthesized using ImProm-II reverse transcriptase. For amplification of GAPDH 1μl and for PL 4μl of cDNA were added to the GoTaq® Hot Start Green Master Mix according to manufacturer's instructions. Primers for GAPDH cDNA amplification were: 5′-GTCTTCACTACCATGGAGAAGG-3′ and 5′-TCATGGATGACCTTGGCCAG-3 (GenBank accession number NM_001034034.1), and for PL primers were described in Ref. [8]. PCR conditions were as follows: heating top 95°C, 95°C for 4min, 45 cycles of denaturation at 94°C (45s), annealing at 60°C for GAPDH and 61.5°C for PL (45s) and extension at 72°C (45s), final extension at 72°C for 10min. PCR products were separated using a 2% agarose gel and visualized by ethidium bromide. The analysis was carried out in triplicates.

2.10. Growth response to EGF 

Cells were seeded in 24-well culture plates (18,000 cells/well) and incubated 24h for attachment in FSM. Prior to stimulation cells were serum starved for 4h and pre-treated for 45min with 50μM PD98059 to inhibit MAPK activation. Afterwards the cells were stimulated with 10% FCS or 50ng/ml EGF for 48h. An MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] assay (Sigma, Munich, Germany) was used to determine cell proliferation. Data were analyzed and statistical significance was determined by the Tukey's HSD (Honestly Significant Difference) test using SAS® software. A p value of <0.001 was defined as highly significant.

2.11. Stimulation, pull-down experiment and Western blot analysis 

Cells were grown to 80% confluence in FSM, serum starved for 4h and then stimulated by adding EGF. After stimulation cells were placed on ice, washed once with PBS and were lysed with a buffer containing 50mM Tris–HCl at pH 7.4, 150mM NaCl, 40mM NaF, 5mM EDTA, 1% (v/v) Nonidet P40, 0.1% (w/v) sodium deoxycholate and 0.1% (w/v) sodium dodecyl sulfate (SDS) supplemented with protease and phosphatase inhibitor cocktail. Lysates were centrifuged for 5min at 13,000g (4°C), and protein concentration of the supernatant was determined. For stimulation experiments, equal amounts of protein (15μg/slot) were denatured in sample buffer (5min at 95°C) and analyzed with SDS-PAGE. Ras activation was measured as described previously [21] with slight modifications. Briefly, a Ras-binding domain of c-Raf GST fusion protein (Raf-RBD-GST) was expressed in Escherichia coli. The bacterial pellet was washed with buffer 1 (50mM Tris pH 7.6, 2mM EDTA and 100mM NaCl). Afterwards cells were lysed in 10ml per 1g moist mass buffer 2 (buffer 1 with 10mM DTT, 2mg/ml lysozyme supplemented with protease inhibitor cocktail) by constant shaking for 60min. MgCl2 and DNase1 were added to a final concentration of 10mM and 10μg/ml respectively. After digestion of genomic DNA (30min, 4°C) the lysate was centrifuged at 12,000g for 20min at 4°C and supernatant was snap-frozen in aliquots for storage. When needed, 15μl of bacterial lysate per sample was incubated with 20μl Glutathion-Cellulose slurry (Carl Roth, Karlsruhe, Germany) for 1h at 4°C. The beads were washed three times in Ras buffer (50mM Tris pH 7.6, 2mM MgCl2, 100mM NaCl, 2% NP40, 10% glycerol). For pull-down experiments, the cells were stimulated and lysed with Ras buffer supplemented with protease and phosphatase inhibitor cocktail. Nucleus-free supernatants containing 800μg total protein were incubated with Raf-RBD-GST-Glutathion-Cellulose for 30min under agitation. The beads were subsequently washed three times in Ras buffer before being resuspended in 25μl of 2×SDS-PAGE loading buffer and boiled for 5min prior to electrophoretic analysis. Afterwards Western blots were performed as suggested by the suppliers of the antibodies and detected with chemiluminescence (SuperSignal West Pico, Pierce, Rockford, USA). In case of pMAPK detection, blots were analyzed by densitometric measurement and quantification (Bio 1D, Vilber Lourmat, Germany). Data were analyzed and statistical significance was determined by the Mann–Whitney-Test using SAS® software. A p value of <0.001 was defined as highly significant. The stimulation experiment was repeated five times and the pull down experiment three times.

Back to Article Outline

3. Results 

3.1. Isolation, culture and characterisation of F3 cells 

We established a placental bovine trophoblast cell line (F3) by continuous subculture of cotyledonary cells. The initial primary cells were obtained from trypsin-dispersed cotyledonary tissue using discontinuous density centrifugation. To examine whether this isolation method was superior to the usage of collagenase 1 solution, we evaluated the effect of the enzymatic digest on the viability of cells. By staining with FITC-Annexin V and propidium iodide and subsequent FACS analysis (Fig. 1) we showed for three individual isolations that trypsin digestion resulted in 54, 58 and 60 percent vital cotyledonary cells, whereas collagenase 1 yielded only 6, 9 and 11 percent vital cells. Thus the isolation with trypsin was preferred.

  • View full-size image.
  • Fig. 1 

    Contour diagram of FITC-Annexin V/PI flow cytometry of cotyledonary cells after digestions with collagenase 1 (A) or trypsin (B). The lower left quadrant of each panel (R1) shows the viable cells, which exclude PI and are negative for FITC-Annexin V binding. The right quadrants (R2+R3) contain non-viable cells which are either in later stages of apoptosis or necrotic (upper right, positive for Annexin V and PI) or early apoptotic (lower right, positive for Annexin V and negative for PI). One representative experiment out of three is shown.

The obtained fetal F3 cells grew in Ham's F12/DMEM medium supplemented with 10% FCS, 2mM glutamine, 100IU/ml penicillin and 100μg/ml streptomycin (FSM). F3 cells became senescent after extended culture beyond 45 passages. Optimal cell growth was achieved by passaging cells at the ratio of 1:8. Upon cultivation on uncoated tissue flasks the established F3 cell line, similar to the primary culture, formed monolayer composed of approximately polygonal epithelial cells. In general, these cells were closely packed together, showing a few areas where cells were flatter and more spread out and therefore bigger (Fig. 2B). Morphologically they were clearly distinguishable from maternal caruncular epithelial cells (BCEC-1, Fig. 2A) which displayed curl-like morphology within the colonies. When cultured on collagen gels, F3 cells grow as a very compact monolayer with clearly defined brims (Fig. 2C) whereas BCEC-1 cells did not adhere to collagen gels at all. In addition, SEM analysis confirmed the presence of apical microvilli and a polarised morphology (Fig. 2D) similar to BCEC-1 (Fig. 2E).

  • View full-size image.
  • Fig. 2 

    Bovine caruncular cell line BCEC-1 (A, E) and cotyledonary trophoblast cell line F3 on different substrata (B–D) in culture. (A) BCEC-1 at passage 21 (P21) shows curve-like morphology when grown on uncoated tissue flasks. (B) F3 cells (P24) exhibit a cobblestone-like phenotype with weakly confined colonies and some cells that display an outspread morphology. (C) After 13 days on collagen gel F3 (P24) colonies are clearly defined with no outspread cells present. (D, E) Scanning electron microscopy (SEM) of F3 (D, P24) and BCEC-1 cells (E, P21) grown on uncoated tissue flasks. Each image shows a single cell with dense apical microvilli (arrowheads).

Characterisation of F3 cells at various time points during culture via immunofluorescence revealed that the larger binucleated cells were positive for the TGC exclusive product PL in a passage dependent way (Fig. 3C). The specificity of the antibody, which was raised against the ovine placental lactogen, was evaluated on primary isolated TGC and on cryosections of bovine placentomes (Fig. 3A, B) and additionally confirmed by detection of the recombinant bovine protein in western blot (Data not shown). To investigate the expression of cytokeratin and the TGC specific product PL over the time of culture, we performed double-label immunostaining. Until the second passage, which corresponds to 70 days in culture (Fig. 3C), the PL antibody stained the cytoplasm of a small number of binucleated cells in the colony. However, these positive cells were mostly devoid of cytokeratin staining (Fig. 3C). During continuous culture, F3 cells lost the PL staining, but became positive for cytokeratin filaments (Fig. 3D). Furthermore the cells were continuously positive for zonula occludens-2 (ZO-2) and tubulin (Fig. 3E). F3 cells did not take up Dil-Ac-LDL indicating the cell line was not of endothelial origin (Fig. 3G). Since the corresponding fetus was male, the origin of the F3 cell line could be confirmed by Y-chromosome PCR. The predicted amplicon of 163bp was obtained with genomic DNA from F3 cells and fetal liver (positive control). As expected, no specific product resulted with the use of maternal DNA derived from BCEC-1 cells as template (Fig. 4A). To further confirm the existence of PL in the F3 cell line, we additionally detected PL mRNA using RT-PCR. The predicted cDNA product of 230bp was amplified from F3 cells and cotyledon homogenate serving as control (Fig. 4B). GAPDH was used as reference gene. The mRNA expression of PL was detectable throughout cultivation, not showing the decline that was observed in the protein expression. No differences in protein or mRNA expression could be observed when using different passages or cells which were previously cryopreserved (Data not shown).

  • View full-size image.
  • Fig. 3 

    Characterisation of F3 via immunofluorescence at different time points in culture. Directly after isolation, around 80% of yielded cells stain positive with an antibody against ovine placental lactogen (A, red). On cryosections of placentomes (B, in vivo control) this antibody stains exclusively TGC (red). The nuclei are stained with Hoechst (A–H, blue). Double labelling of F3 cells in culture (C–F) with cytokeratin (C–D, green) and ovine placental lactogen antibody (C–D, red) shows binucleate F3 cells (C, P2) still expressing TGC specific products while no cytokeratin is detectable. At higher passages (D, P18) similar cells express cytokeratin, but no TGC specific products are present anymore. F3 cells constantly express beta-tubulin (E, green) and tight junctional zonula occludens-2 (Zo-2) protein (E, red; P20). Representative isotype control (F; P20). Low density lipoprotein (Dil-AC-LDL, red) labelling shows no uptake of acetylated LDL in F3 cells (G; P20), whereas bovine umbilical vein endothelial cells (H, positive control) were positive. Scale bar is 50μm.

  • View full-size image.
  • Fig. 4 

    Identification of F3 cells according to genomic DNA or mRNA expression. (A) The presence of a Y-chromosome specific sequence (163bp) in F3 cells (lane 3, passage 3) confirmed the fetal origin. The liver of a male fetus (lane 2) served as positive control and maternal BCEC-1 cells as negative control (lane 1). Band sizes (lane M, top to bottom) are 500, 400, 300, 200 and 100bp. Lane 4 represents the negative PCR control. (B) Trophoblast F3 cells (lane 3) and bovine cotyledonary homogenate (lane 2) showed specific bands for PL at 230bp, while cultured fetal mesenchyme cells were negative (lane 1). GAPDH was used as a housekeeping gene. Specific products (197bp) for GAPDH were amplified from cultured fetal mesenchymal cells, cotyledonary homogenate and F3 cells (lanes 5–7, respectively). Lane 4 shows the negative PCR control.

3.2. Effect of EGF on F3 proliferation and signalling 

After establishment, the cell line had a population doubling time of approximately 24h in FSM, as estimated by direct cell count (Fig. 5A). A significant growth-promoting effect of EGF without the presence of FCS was shown by an in vitro proliferation assay. Compared to cells grown in serum-free Ham's F12/DMEM medium (SF) the addition of 50ng/ml EGF enhanced F3 proliferation by about 52% (p>0.001) while 10% FCS lead to an increase of 25% (p>0.001). In contrast, 10% FCS notably induced maternal BCEC-1 proliferation (21%; p>0.001) whereas EGF did not stimulate cell growth. The enhanced mitogenic response of F3 cells to EGF could be inhibited by pre-treatment with the MEK1/2 inhibitor PD98059 (Fig. 5B). We also confirmed the expression of the corresponding EGF receptor (EGF-R) by RT-PCR and western blotting in both cell lines (data not shown).

  • View full-size image.
  • Fig. 5 

    (A) Daily growth curve of F3 trophoblast cells. 1×105 cells were seeded in 25cm2 culture flasks and grown in medium containing 10% FCS, 2mM Glutamine and antibiotics (full supplemented medium). The medium was changed every 48h and cell number was determined every 24h by counting four random fields per time on a haemacytometer. The experiment was repeated 8 times (n=8). (B) Growth response of F3 and BCEC-1 cells after incubation with EGF (50ng/ml) for 48h. Proliferation was determined by MTT assay. Results were normalized relative to the control (SF=serum-free medium) and then averaged (means±SEM, F3 n=20, BCEC-1 n=5). Asterisks represent p<0.001 compared to control. 10% FCS=SF+10% FCS; SF+EGF=SF+EGF 50ng/ml; SF+EGF+PD=SF+EGF 50ng/ml+50μM PD98059 MEK inhibitor.

The signalling events involved in the enhanced proliferation by EGF were analyzed by Ras pull-down assay and western immunoblotting. EGF elevated the level of activated Ras within 2.5min in F3 cells (detected by pull-down assay, using Raf-RBD as bait; Fig. 6A). Consistent with the results from the proliferation assay, a stimulation of Ras was absent in the BCEC-1 cells. EGF (50ng/ml) stimulated phosphorylation of MAPK42 and MAPK44 in a time-dependent manner reaching a maximum at 5–15min after treatment. As expected, this activation was limited to F3 cells (Fig. 6B). The typical time course of the specific MAPK activity is shown in a representative histogram (Fig. 6C). The MAPK activities significantly increased within 5min followed by a steady decrease to about 4-fold of control value after 30min (p>0.001).

  • View full-size image.
  • Fig. 6 

    Ras activation and MAPK phosphorylation in F3 or BCEC-1 cells upon stimulation with EGF. (A) Cells were stimulated for the indicated times with 50ng/ml EGF and analyzed for activated Ras (using a pull-down assay with Raf-RBD as bait, 21kDa). A representative western immunoblot probed with anti-Ras antibody shows that EGF leads to Ras activation only in F3 cells, while no activation is induced in maternal BCEC-1 cells. (B) Representative western blot of F3 and BCEC-1 protein extracts after stimulation with EGF (50ng/ml) for indicated times. Analysis of probes for MAPK phosphorylation (pMAPK42/44, 42 and 44kDa) revealed that a phosphorylation occurred in F3 cells, in contrast BCEC-1 cells showed no significant phosphorylation. Actin (42kDa) was used as loading control. (C) Densitometric data from five experiments (mean±SD) describing the time course of EGF induced MAPK phosphorylation in F3 cells. Optical densities were evaluated with the program Bio 1D (Vilber Lourmat) and corrected for actin as loading control. Asterisks represent p<0.001 compared to control (time 0). A maximum stimulation is present after 5min and decreases until 30min.

Back to Article Outline

4. Discussion 

In order to study trophoblast physiology in the synepitheliochorial bovine placenta many attempts have been made to culture bovine placental trophoblast cells. The success of these attempts was limited, firstly because the trophoblast cell cultures turned out to be uncryopreservable and secondly, because the characterisation of the isolated putative trophoblast cell cultures left doubts concerning the cellular origin.

In the present study we describe the establishment and characterisation of the first bovine placental trophoblast cell line (F3) of confirmed identity. The F3 cell line was obtained by trypsin digestion of cotyledonary tissue after manual separation of bovine placentomes from a male fetus. Expression of cytokeratin and the trophoblast specific marker PL, the presence of the Y-chromosome, the epitheloid morphology and the simultaneous lack of Dil-Ac-LDL uptake confirmed the trophoblastic origin of F3. Proliferation and activation of the MAPK signalling cascade in reaction to EGF stimulation indicates that EGF plays a role during bovine placentation and suggests that F3 cells may be used for further functional studies elucidating the regulatory mechanisms of placental growth and function.

4.1. Implications of methodological approach and characterisation methods 

As the commonly used protocol for the isolation of placentomal trophoblast cells [3], [5], [10], [12] caused rather devastating effects on cellular viability in our hands (viability of 10% with collagenase 1 versus 60% with trypsin), we applied trypsin digestion, which was used successfully before for isolation of human trophoblast cells [22]. The fact that collagenase treatment leads to satisfactory results in other studies may be due to differences in the preparation of enzyme, the biological activity and purity (i.e. contamination with additional enzymes). Consistent with other researchers [5] we found that Percoll gradient centrifugation is suitable to reduce the amount of erythrocytes and cell debris. Nevertheless, an enrichment of trophoblast cells with this method was not possible.

A stable cell line can be generated with a range of techniques such as gamma rays or different viral systems. However, viral transfection can result in a deficiency in specific cell properties which was demonstrated by the loss of TGFbeta responsiveness in human trophoblast [23]. To bypass these complications we decided to obtain the F3 cell line through spontaneous immortalization. Even though many mouse cell lines undergo spontaneous immortalization or transformation in tissue culture readily [24], [25], cells from higher species (i.e. human cells) rarely display spontaneous immortalization [26]. The low spontaneous immortalization rate and the low cell viability after the widely used enzymatic digestion with collagenase are most likely the main reasons why a stable bovine trophoblast cell line was not reported until now.

Our isolation method yielded a primary culture with predominantly epitheloid cells. Even though a small fraction of stromal cells was present in the beginning, no overgrowth was observed. In fact, the presence of fibroblasts may exert a positive influence on epithelial cells as described for human endometrial cells [27]. However, the stromal cells were eliminated after the fifth passage due to differential trypsinization that takes advantage of the different adherent properties of these cell types. Once in constant culture, F3 showed similar morphological characteristics as primary cells directly after attachment. In addition, their appearance resembled that of the cells described by Munson et al. [6] and the in vitro-produced blastocyst derived cell lines CT-1/CT-5 [7] and BT-1 [8]. F3 cells grew without the necessity of a substrate like collagen or a feeder layer (in contrast to the reports of others [8], [28]). The cell doubling time remained stable over the whole duration of culture with a progressing decrease above passage 45. Furthermore, the ability to adhere to collagen gel was one quality that distinguished F3 from maternal BCEC-1 cells which did not attach to collagen gel. To identify distinct differences between trophoblast and caruncular epithelial cells was of particular importance because isolated cotyledonary cells are frequently contaminated with maternal epithelial cells [3]. To ensure the origin of the F3 cell line, we took advantage of the fact that F3 was derived from a male fetus and produced an absolute proof, the PCR-amplification of a Y-chromosome specific sequence. This proof is lacking in other studies dealing with placentomal trophoblast cells [5], [6]. Therefore, so far only blastocyst derived trophoblast cell cultures can be considered as trophoblast for sure.

To further expand the evidence that the cells were of trophoblast origin and retained the in vivo characteristics, we localized cytoskeletal filaments and associated proteins in F3 by immunofluorescence. It turned out that fetal F3 and maternal cells (BCEC-1) share many properties in vitro. Both cell types are polarised as they show apical microvilli and expression of the tight junctional zonula occludens protein as well as epithelial cytokeratin and mesenchymal vimentin. The expression of this mesenchymal marker in endometrial cell cultures has been described previously for several species [29], [30]. In contrast to the maternal BCEC-1 cells, F3 showed an increased tendency to form binucleated cells. Such cells are more often observed among cultured cells in senescence; however, their emergence in F3 cultures was not altered over passages.

Right after isolation binucleated F3 cells expressed PL and were almost devoid of cytokeratin filaments. The expression of PL is exclusively observed in TGC in the bovine placenta in vivo [2] providing direct proof of the cells trophoblast origin. The observed decrease of this marker in cultured bovine trophoblast cells could be characteristic for a mixed culture of uninucleate and binucleate cells since in vivo binucleate cells undergo atrophy and die once they fuse with uterine epithelial cells [31]. In addition Nakano et al. (2001) demonstrated an inverse relationship between the expression of PL and cytokeratin. Primary TGC that produce PL showed no cytokeratin filaments, while in prolonged culture TGC cease to express PL but gain cytokeratin expression [28]. A reestablishment of the PL expression like described for BT-1 cells when grown on collagen gel [9] could not be observed in F3 cells. However, RT-PCR revealed that PL mRNA expression in F3 cells was not considerably altered while the protein expression ceased over time, suggesting a regulation on the translational level.

4.2. Implications of EGF signalling 

Treatment of F3 cells with EGF leads to a strong increase in cell proliferation and induction of the MAPK signalling cascade as binding of EGF to its receptor (EGF-R) activates different signalling pathways which regulate a variety of cellular responses, including cell proliferation, migration and differentiation [14], [32]. The presence of EGF-R in trophoblast and maternal epithelium [33], [34], [35] allows a pivotal role of the EGF system in the embryonic and/or uterine development, which has been proposed before [36], [37]. EGF is a potent inhibitor of apoptosis in primary human cytotrophoblast (CT) [38] and also promotes syncytial formation, thereby mediating differentiation of cytotrophoblasts [14], [39]. Additionally, EGF can upregulate extravillous trophoblast invasion [40] as well as enhance CT motility [41] or modulate gap junction protein expression [42]. EGF also has a positive effect on cultures of primary trophoblast cells from the cow [6], mouse [43] and man [44]. Based on our finding that EGF strongly enhanced F3 cell growth we hypothesized that the observed proliferative effect could be due to an activation of the Ras/MAPK pathway because it is the major pathway involved in EGF induced mitogenic activation ([45] for review). Indeed we were able to demonstrate that EGF leads to an activation of the small GTPase Ras as well as to a phosphorylation of MAPK42/44. An inhibition of MAPK activation by PD98059 almost completely prevented EGF induced cell proliferation and MAPK42/44 phosphorylation (data not shown). In contrast, the same treatments of the caruncular epithelial cell line BCEC-1 neither lead to an activation of the Ras/MAPK pathway nor to a growth promoting effect. However, we were able to detect EGF-R mRNA and protein in both cell lines (F3 and BCEC-1; data not shown). Even though EGF did not stimulate the growth of BCEC-1 cells under the reported conditions it is likely that other yet unknown effects are mediated by EGF since EGF-R is expressed in the caruncular epithelium in vivo throughout gestation [33]. To discover these other biological effects in future studies will be of particular interest.

In summary, we are the first to establish a long-term stable bovine cotyledonary trophoblast cell line (F3) of confirmed origin and functional activity of the EGF signalling cascade. Since the specific components of this cascade are also expressed in the bovine placenta in vivo, we can conclude that EGF signalling via Ras/MAPK pathway seems to play a role during synepitheliochorial bovine placentation. In the present study we have focussed on proliferative effects, however, the described trophoblast F3 cell line provides us with the opportunity to investigate the effects of different growth factors on other biological processes such as differentiation or invasion.

Back to Article Outline

Acknowledgements 

We thank Prof. C. Herrmann for providing us with the plasmid pGEX-RBD. The technical assistance of Mrs. I. Blume is gratefully acknowledged. This work was supported by a grant from the Deutsche Forschungsgemeinschaft (DFG).

Back to Article Outline

References 

  1. Klisch K, Pfarrer C, Schuler G, Hoffmann B, Leiser R. Tripolar acytokinetic mitosis and formation of feto-maternal syncytia in the bovine placentome: different modes of the generation of multinuclear cells. Anat Embryol (Berl). 1999;200(2):229–237
  2. Wooding FB. Current topic: the synepitheliochorial placenta of ruminants: binucleate cell fusions and hormone production. Placenta. 1992;13(2):101–113
  3. Bridger PS, Haupt S, Klisch K, Leiser R, Tinneberg HR, Pfarrer C. Validation of primary epitheloid cell cultures isolated from bovine placental caruncles and cotyledons. Theriogenology. 2007;68(4):592–603
  4. Bridger PS, Menge C, Leiser R, Tinneberg HR, Pfarrer CD. Bovine caruncular epithelial cell line (BCEC-1) isolated from the placenta forms a functional epithelial barrier in a polarised cell culture model. Placenta. 2007;28(11–12):1110–1117
  5. Vanselow J, Furbass R, Tiemann U. Cultured bovine trophoblast cells differentially express genes encoding key steroid synthesis enzymes. Placenta. 2008;29(6):531–538
  6. Munson L, Chandler SK, Schlafer DH. Long-term culture of bovine trophoblastic cells. J Tissue Cult Methods. 1988;1988(11):123–128
  7. Talbot NC, Caperna TJ, Edwards JL, Garrett W, Wells KD, Ealy AD. Bovine blastocyst-derived trophectoderm and endoderm cell cultures: interferon tau and transferrin expression as respective in vitro markers. Biol Reprod. 2000;62(2):235–247
  8. Shimada A, Nakano H, Takahashi T, Imai K, Hashizume K. Isolation and characterization of a bovine blastocyst-derived trophoblastic cell line, BT-1: development of a culture system in the absence of feeder cell. Placenta. 2001;22(7):652–662
  9. Nakano H, Shimada A, Imai K, Takezawa T, Takahashi T, Hashizume K. Bovine trophoblastic cell differentiation on collagen substrata: formation of binucleate cells expressing placental lactogen. Cell Tissue Res. 2002;307(2):225–235
  10. Landim LP, Miglino MA, Pfarrer C, Ambrosio CE, Garcia JM. Culture of mature trophoblastic giant cells from bovine placentomes. Anim Reprod Sci. 2007;98(3–4):357–364Epub 2006 Apr 28
  11. Feng S, Peter AT, Asem E. Establishment of a pure culture of distinct cell types from bovine placental cotyledon. Methods Cell Sci. 2000;22(2–3):101–106
  12. Zeiler M, Leiser R, Johnson GA, Tinneberg HR, Pfarrer C. Development of an in vitro model for bovine placentation: a comparison of the in vivo and in vitro expression of integrins and components of extracellular matrix in bovine placental cells. Cells Tissues Organs. 2007;186(4):229–242
  13. Zwick E, Hackel PO, Prenzel N, Ullrich A. The EGF receptor as central transducer of heterologous signalling systems. Trends Pharmacol Sci. 1999;20(10):408–412
  14. Morrish DW, Dakour J, Li H, Xiao J, Miller R, Sherburne R, et al. In vitro cultured human term cytotrophoblast: a model for normal primary epithelial cells demonstrating a spontaneous differentiation programme that requires EGF for extensive development of syncytium. Placenta. 1997;18(7):577–585
  15. Kliem A, Tetens F, Klonisch T, Grealy M, Fischer B. Epidermal growth factor receptor and ligands in elongating bovine blastocysts. Mol Reprod Dev. 1998;51(4):402–412
  16. Evain-Brion D, Alsat E. Epidermal growth factor receptor and human fetoplacental development. J Pediatr Endocrinol. 1994;7(4):295–302
  17. Machida T, Taga M, Minaguchi H. Effects of epidermal growth factor and transforming growth factor alpha on the mouse trophoblast outgrowth in vitro. Eur J Endocrinol. 1995;133(6):741–746
  18. Johnstone ED, Mackova M, Das S, Payne SG, Lowen B, Sibley CP, et al. Multiple anti-apoptotic pathways stimulated by EGF in cytotrophoblasts. Placenta. 2005;26(7):548–555
  19. Hermosilla C, Zahner H, Taubert A. Eimeria bovis modulates adhesion molecule gene transcription in and PMN adhesion to infected bovine endothelial cells. Int J Parasitol. 2006;36(4):423–431
  20. Pomp D, Good BA, Geisert RD, Corbin CJ, Conley AJ. Sex identification in mammals with polymerase chain reaction and its use to examine sex effects on diameter of day-10 or -11 pig embryos. J Anim Sci. 1995;73(5):1408–1415
  21. de Rooij J, Bos JL. Minimal Ras-binding domain of Raf1 can be used as an activation-specific probe for Ras. Oncogene. 1997;14(5):623–625
  22. Ho CK, Chiang H, Li SY, Yuan CC, Ng HT. Establishment and characterization of a tumorigenic trophoblast-like cell line from a human placenta. Cancer Res. 1987;47(12):3220–3224
  23. Khoo NK, Bechberger JF, Shepherd T, Bond SL, McCrae KR, Hamilton GS, et al. SV40 Tag transformation of the normal invasive trophoblast results in a premalignant phenotype. I. Mechanisms responsible for hyperinvasiveness and resistance to anti-invasive action of TGFbeta. Int J Cancer. 1998;77(3):429–439
  24. Sanford KK, Evans VJ. A quest for the mechanism of “spontaneous” malignant transformation in culture with associated advances in culture technology. J Natl Cancer Inst. 1982;68(6):895–913
  25. Pringproa K, Kumnok J, Ulrich R, Baumgartner W, Wewetzer K. In vitro characterization of a murine oligodendrocyte precursor cell line (BO-1) following spontaneous immortalization. Int J Dev Neurosci. 2008;26(3–4):283–291
  26. DiPaolo JA. Relative difficulties in transforming human and animal cells in vitro. J Natl Cancer Inst. 1983;70(1):3–8
  27. Arnold JT, Kaufman DG, Seppala M, Lessey BA. Endometrial stromal cells regulate epithelial cell growth in vitro: a new co-culture model. Hum Reprod (Oxford, England). 2001;16(5):836–845
  28. Nakano H, Takahashi T, Imai K, Hashizume K. Expression of placental lactogen and cytokeratin in bovine placental binucleate cells in culture. Cell Tissue Res. 2001;303(2):263–270
  29. Johnson GA, Burghardt RC, Newton GR, Bazer FW, Spencer TE. Development and characterization of immortalized ovine endometrial cell lines. Biol Reprod. 1999;61(5):1324–1330
  30. Zhang Z, Paria BC, Davis DL. Pig endometrial cells in primary culture: morphology, secretion of prostaglandins and proteins, and effects of pregnancy. J Anim Sci. 1991;69(7):3005–3015
  31. Wooding FB. The role of the binucleate cell in ruminant placental structure. J Reprod Fertility. 1982;31:31–39
  32. Schlessinger J. Ligand-induced, receptor-mediated dimerization and activation of EGF receptor. Cell. 2002;110(6):669–672
  33. Weise S. Untersuchungen zum Vorkommen von Wachstumsfaktoren (aFGF, bFGF, TGF-α) und deren Rezeptoren (FGF-R, EGF-R) in der Plazenta des Rindes. Dissertation; 2001.
  34. Duello TM, Bertics PJ, Fulgham DL, Van Ess PJ. Localization of epidermal growth factor receptors in first- and third-trimester human placentas. J Histochem Cytochem. 1994;42(7):907–915
  35. Muhlhauser J, Crescimanno C, Kaufmann P, Hofler H, Zaccheo D, Castellucci M. Differentiation and proliferation patterns in human trophoblast revealed by c-erbB-2 oncogene product and EGF-R. J Histochem Cytochem. 1993;41(2):165–173
  36. Hofmann GE, Scott RT, Bergh PA, Deligdisch L. Immunohistochemical localization of epidermal growth factor in human endometrium, decidua, and placenta. J Clin Endocrinol Metab. 1991;73(4):882–887
  37. Pöhland R, Tiemann U. Immunhistochemical localization of the epidermal growth factor and its binding sites in the bovine female reproductive tract. J Reprod Fertility. 1994;14:21;Abstract Series
  38. Smith S, Francis R, Guilbert L, Baker PN. Growth factor rescue of cytokine mediated trophoblast apoptosis. Placenta. 2002;23(4):322–330
  39. Morrish DW, Bhardwaj D, Dabbagh LK, Marusyk H, Siy O. Epidermal growth factor induces differentiation and secretion of human chorionic gonadotropin and placental lactogen in normal human placenta. J Clin Endocrinol Metab. 1987;65(6):1282–1290
  40. Bass KE, Morrish D, Roth I, Bhardwaj D, Taylor R, Zhou Y, et al. Human cytotrophoblast invasion is up-regulated by epidermal growth factor: evidence that paracrine factors modify this process. Dev Biol. 1994;164(2):550–561
  41. LaMarca HL, Dash PR, Vishnuthevan K, Harvey E, Sullivan DE, Morris CA, et al. 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 (Oxford, England). 2008;23(8):1733–1741
  42. 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
  43. Iguchi T, Tani N, Sato T, Fukatsu N, Ohta Y. Developmental changes in mouse placental cells from several stages of pregnancy in vivo and in vitro. Biol Reprod. 1993;48(1):188–196
  44. Li RH, Zhuang LZ. The effects of growth factors on human normal placental cytotrophoblast cell proliferation. Hum Reprod (Oxford, England). 1997;12(4):830–834
  45. Wells A. EGF receptor. Int J Biochem Cell Biol. 1999;31(6):637–643

PII: S0143-4004(09)00334-8

doi:10.1016/j.placenta.2009.10.011

Placenta
Volume 31, Issue 1 , Pages 67-74, January 2010

Article Tools

* Image Usage & Resolution