Influences of Extracellular Matrix and of Conditioned Media on Differentiation and Invasiveness of Trophoblast Stem Cells

Placenta 28 (2007) 14e21

Influences of Extracellular Matrix and of Conditioned Media on Differentiation and Invasiveness of Trophoblast Stem Cells T. Lei a,b, H.-P. Hohn a, R. Behr a, H.-W. Denker a,* a b

Institut fu¨r Anatomie, Universita¨tsklinikum Essen, Hufelandstrasse 55, 45122 Essen, Germany State Key Laboratory of Agricultural Microbiology, College of Life Science and Technology, Huazhong Agricultural University, 430070 Wuhan, China Accepted 28 January 2006

Abstract Embryo implantation in the human and rodents relies on the trophoblast’s ability to invade into the uterine stroma, partly depending on proteinases degrading components of basement membrane and underlying extracellular matrix (ECM). We have utilized mouse trophoblast stem (TS) cells (Science, 1998, 282:2072) to study trophoblast invasion and trophoblasteECM interactions in vitro. On plastic in fibroblastconditioned medium containing fibroblast growth factor (FGF)-4 and heparin, the cells remain proliferative but display increased differentiation in media without these components. Marker gene expression (Eomes, Pl-1, Tpbp) and invasion assays showed that TS cells exhibit increased invasive capacity when differentiating into giant cells and spongiotrophoblasts in unconditioned media without FGF-4 and heparin. Concomitantly, an up-regulation of matrix metalloproteinases (MMP)-9 and -14 was observed. Culture on gels of the basement membrane-like Matrigel resulted in striking changes in morphology and gene expression. Differentiating TS cells invaded into this ECM in a three-dimensional culture, while in turn ECM contact enhanced differentiation of TS cells and up-regulated the expression of MMP-9 and its tissue inhibitor (TIMP)-3. These findings implicate that the TS cell culture system used in this study can be utilized as a model for studying the regulation of trophoblasteECM interactions, differentiation, and invasion in vitro. Ó 2006 Elsevier Ltd. All rights reserved. Keywords: Trophoblast stem cell; Extracellular matrix; Differentiation; Invasion; Matrix metalloproteinase

1. Introduction The trophoblast cell lineage arises at the blastocyst stage and contributes exclusively to the extraembryonic structures [1]. In addition to endocrine and transport functions, trophoblast cells play a central role in embryo implantation. In mice, four differentiated trophoblast cell phenotypes can be identified: (1) trophoblast giant cells (TGCs), (2) spongiotrophoblast (SpT), (3) glycogen trophoblast (GlyT), and (4) syncytiotrophoblast (SynT) [2]. TGCs are identified as those cells that have characteristic giant nuclei due to DNA * Corresponding author. Institut fu¨r Anatomie, Universita¨tsklinikum Essen, Hufelandstrasse 55, 45122 Essen, Germany. Tel.: þ49 201 7234380; fax: þ49 201 7235916. E-mail address: [email protected] (H.-W. Denker). 0143-4004/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.placenta.2006.01.020

endoreduplication [3]. This cell type exhibits invasive characteristics [4] and also possesses significant endocrine activities [5]. SpT forms the middle layer of the placenta between the outermost TGCs and the innermost labyrinth. In addition to having a structural role, SpT is also a source of precursors of TGCs and GlyT. GlyT appears to be a differentiated derivative of SpT because it first appears within the SpT layer and expresses the same marker genes as SpT [4]. At late stages of gestation, GlyT penetrates into the decidual stroma although it does not establish close contact with the maternal blood vessels. SynT cells are multinucleated cells that form by celle cell fusion and play a role in bi-directional transport between the mother and fetus. The cascade of cellular events which finally lead to interstitial implantation in the mouse includes the initial attachment between the apical surface of trophoblast cells and the luminal

T. Lei et al. / Placenta 28 (2007) 14e21

epithelium, followed by trophoblast penetration through the apoptosing epithelial lining and finally invasion of the underlying decidua [6]. The maternalefetal interactions are accompanied by proliferation and differentiation of trophoblast and, on the other hand, endometrial decidualization [7]. Trophoblast invasion involves degradation and remodeling of uterine extracellular matrix (ECM), partly regulated by matrix metalloproteinases (MMPs) and their tissue inhibitors (TIMPs) [8]. An in vitro culture model, the mouse trophoblast stem (TS) cell lines, has been established from the 3.5 days post coitum (dpc) blastocysts or the extraembryonic ectoderm of 6.5 dpc embryos [9]. TS cells can be maintained in a proliferative state or be induced to differentiate into all the trophoblast subtypes found in vivo. We have exploited the TS cell culture for studying trophoblast differentiation and invasion monitored under different culture conditions and in ECM penetration model systems.

2. Materials and methods 2.1. Cell culture The TS cell line (kindly provided by Dr. Rossant, Samuel Lunenfeld Research Institute, Toronto, Canada), derived from a 3.5 dpc blastocyst of B5/EGFP transgenic mice, was cultured according to methods reported previously [9]. The basal culture medium (TS-M) consisted of RPMI 1640 medium supplemented with 20% fetal calf serum, glutamine, sodium pyruvate, betamercaptoethanol, and penicillin/streptomycin (Invitrogen, UK). For stem cell culture, TS-M was supplemented with 70% TS-M preconditioned for 3 days by confluent cultures of radiation-treated primary mouse embryo fibroblasts (MEF), and supplemented with FGF-4 (25 ng/ml) and heparin (1 mg/ml, Sigma, USA) [9]. This supplemented medium that suppresses TS cell differentiation in cultures on plastic will be referred to as TSþM. To induce differentiation, TS cells were cultured in TS-M, i.e. without FGF-4, heparin, and MEF-conditioned medium (MEF-CM). For the analysis of RNA cells were harvested 3e4 days after passage. For the culture on gels of Matrigel, 96-well tissue culture plates were covered with undiluted Matrigel (BD Biosciences, 50 ml/well) and left at 37  C for at least 2 h to allow the Matrigel to solidify and form gels [10]. These gels were not allowed to dry but were always kept covered with culture medium so that a relatively thick and soft matrix is preserved. The use of this type of substrate is referred to as ‘‘culture on Matrigel’’ or on ‘‘gels of Matrigel’’. In contrast, the terms ‘‘cultures on plastic’’ and ‘‘uncoated plastic’’, respectively, both express that the tissue culture plastic was not modified by either adsorbing molecules (e.g. matrix molecules) to it or covering the plastic with any gel. Fifty thousand cells per well were seeded and grown for 6 and 9 days in TSþM and TS-M, respectively. The medium was changed daily.

2.2. Morphology The morphology of TS cells cultured on plastic and Matrigel was observed by phase contrast microscopy. Semi-thin sections were taken after fixation with 2.5% glutaraldehyde and 1% osmium tetroxide and embedding in Epon, and were stained with toluidine blue according to standard protocols.

2.3. Invasion assay The invasiveness of TS cells was studied using Transwell chambers (Costar, USA) with 8 mm pore filters as previously described [11]. Briefly, the filters were covered with 100 ml diluted Matrigel (1:20 in RPMI 1640) and allowed to dry overnight in a laminar flow hood. TS cells cultured in TSþM were harvested and suspended either in TSþM or in TS-M. Fifty

15

thousand cells in 200 ml of TSþM and TS-M, respectively, were seeded in the upper chamber, and the lower chamber was filled with 800 ml of the same medium. The medium was changed daily. After 96 h incubation at 37  C in a humidified atmosphere of 5% CO2 in air, the contents of the upper chamber were aspirated followed by washing twice with PBS, and the layer of Matrigel with the cells on the upper surface of the filters was removed with a cotton swab. In order to assess cell motility, cells were transferred to plain Transwell chambers, where the filters had not been covered with a layer of Matrigel [11]. After incubation the cells on the surface of the filters were removed with a cotton swab. In both types of experiments, the cells that had migrated through the filter and reached the lower surface were fixed with ice-cold methanol and stained with hematoxylin. For quantitation, the filters were cut off the chamber and mounted on slides, and cells were then counted under a microscope at 100.

2.4. RNA isolation and reverse transcription-polymerase chain reaction (RT-PCR) Total RNA was prepared using the peqGOLD Trifast solution (peqLab, Germany) followed by treatment with RNase-free DNase (Promega, USA), and quantified by GeneQuant (Pharmacia Biotech, Germany). For RT-PCR, 2 mg RNA were reverse transcribed into cDNA with the Omniscript RT Kit (Qiagen, Germany) and oligo(dT) primers. One tenth of RT products were then analyzed in a 30 ml PCR system containing 3 ml 10 Biotherm PCR buffer (Genecraft, Germany), 0.5 ml of each primer in 10 pM, 0.5 ml of 10 mM dNTPs, and 2.5 U BioTherm DNA polymerase (Genecraft, Germany). The primer pairs and PCR reaction parameters are shown in Table 1. For semiquantitative PCR, the number of cycles was optimized for each target to avoid saturation effects. RT-PCR products were resolved on 1.6% agarose gel containing ethidium bromide, and the intensities of the bands were quantified using the Gelscan software (BioSciTec, Germany). In order to normalize the mRNA levels in different samples, the intensity of the band corresponding to each mRNA was divided by the intensity of the band corresponding to the b-actin mRNA used as an internal control. The identities of PCR products were further confirmed by sequencing.

Table 1 Primers for RT-PCR Genes

Primers 50 e30

Tm Cycles Ref. (  C)

Eomes

Fw: CAA ATT CCA CCG GCA CCA AAC Re: GGG GTT ATG (AG)TC (AG)AT CTT TAG

58

28

[31]

Pl-1

Fw: CTC CAA ACC AAC TGA CAT (GT)GT 59 Re: GTC CCT TTT AA(AT) GCA GCG GCA

25

[32]

Tpbp

Fw: CAG CAC AGC TTT GGA CAT CAC Re: GGA AGT CCC TGC TGT TTT TGC

59

28

[14]

MMP-9

Fw: TTG AGT CCG GCA GAC AAT CC Re: CCT TAT CCA CGC GAA TGA CG

59

30

[33]

MMP-14 Fw: GAA GTG CCC TAT GCC TAC ATC Re: GTT CTC TGT GTC CAT CCA CTG

59

30

[34]

TIMP-1

Fw: CGC AGA TAT CCG GTA CGC CTA Re: CAC AAG CCT GGA TTC CGT GG

59

28

[27]

TIMP-2

Fw: CTC GCT GGA CGT TGG AGG AA Re: CAC GCG CAA GAA CCA TCA CT

59

32

[27]

TIMP-3

Fw: CTT GTC GTG CTC CTG AGC TG Re: CAG AGG CTT CCG TGT GAA TG

59

28

[27]

b-Actin

Fw: ACC TTC AAC ACC CC(AC) GCC 60 ATG TACG Re: CT(AG) ATC CAC ATC TGA TGG AAG GTG G

28

16

T. Lei et al. / Placenta 28 (2007) 14e21

2.5. Data analysis Data were represented as means  SEM and analyzed by Student’s t-test. A P-value of less than 0.05 was considered to be statistically significant.

observed in serial sections (not shown), at least some of the cell clusters found deeper in the gels were not connected with colonies on the surface, but were still surrounded by intact gel (Fig. 2D). There was no indication of massive degradation of the gel in spite of the presence of invasive cells.

3. Results 3.1. TS cell morphology on Matrigel 3.2. Effect of Matrigel on TS cell differentiation Cells were cultured on gels of Matrigel for up to 9 days in TSþM and TS-M, respectively. As compared to the same cultures in TSþM on plastic, where they formed confluent monolayers within 3 days (Fig. 1A), TS cells formed scattered colonies on the gels of Matrigel (Fig. 1C). In TS-M cultures, TS cells on plastic differentiated into trophoblast giant cells (Fig. 1B) whereas on Matrigel the cells continued to grow as scattered colonies (Fig. 1D). As can be seen on semi-thin sections, TS cells cultured on gels of Matrigel in TSþM grew as monolayers of irregularly shaped but mostly cuboidal cells (Fig. 2A, B) at the surface of the gel. After 6 days of culture they appeared to have formed pits in the gel in which, occasionally, remnants of degenerated cells were accumulated (Fig. 2A, B). After 9 days in culture, these pits seemed to have developed into solid or hollow tongues of piled-up/multilayered cells mostly without lumen between them, and intruded the gel (Fig. 2C). In TS-M (Fig. 2D, E), in contrast, cells did not form a monolayer on the Matrigel but rather multicellular aggregates, and they penetrated into Matrigel either as solid chords of cells still in contact with the aggregate on the surface of the gel (Fig. 2D) or as isolated clusters of cells found deep in the gel after 6 days in culture (Fig. 2D). As

Cells grown on plastic for 96 h in TSþM expressed the stem cell marker Eomes and the giant cell marker Pl-1 (Fig. 3). This is in agreement with the previous report that TS cells have a spontaneous tendency to differentiate and there is always a small portion of giant cells in the culture even under differentiation-suppressing conditions [9], which was also observed in our experiments (Fig. 1A). Weak expression of the spongiotrophoblast marker Tpbp was also detected. In TS-M on plastic Eomes expression was reduced up to 10-fold (P < 0.05) while Pl-1 was up-regulated, and the expression of Tpbp was even more highly increased (10-fold, P < 0.05) as compared to that in TSþM. When the cells were grown on Matrigel in TSþM Eomes expression was reduced (2fold, P < 0.05), while Tpbp was up-regulated (6-fold, P < 0.05) as compared to cells grown on plastic. In TS-M on Matrigel the expression of Eomes was lost in the same way as it was in this medium on plastic. Under these conditions, the expression of Tpbp reached a high level that was comparable to that observed on plastic in TS-M. The levels of Pl-1 expression on gels of Matrigel were comparable to those on plastic (Fig. 3).

Fig. 1. Phase contrast morphology of TS cells growing on plastic and Matrigel. Equal numbers (50,000 per well) of TS cells were plated on 96-well tissue culture plates that were either uncoated (A, B) or covered with a gel of Matrigel (C, D) and incubated for 3 days (A, B) and 6 days (C, D) in TSþM (A, C) and TS-M (B, D), respectively. Trophoblast giant cells in cultures on plastic are marked by arrows (A, B). Scale bars represent 10 mm (A, B) and 100 mm (C, D), respectively.

T. Lei et al. / Placenta 28 (2007) 14e21

17

Fig. 2. Histology of TS cells growing on Matrigel in differentiation-inhibiting or promoting media. TS cells were cultured on gels of Matrigel for 6 and 9 days in (AeC) TSþM (A, B ¼ 6 days, C ¼ 9 days) and (D, E) TS-M (D ¼ 6 days, E ¼ 9 days), respectively. Semi-thin sections were made and stained with toluidine blue for light microscopy. Note the degenerated cells (arrowheads in A, B) and the tongues of piled-up/multilayered cells (arrows in C). Chords of cells and isolated clusters of cells invading into the gel in (D) are indicated by arrows and arrowheads, respectively. Ma ¼ Matrigel. Scale bars represent 20 mm.

3.3. Invasiveness Cells were cultured on uncovered (motility assay) or Matrigel-covered (invasion assay) Transwell inserts for up to 96 h and the number of cells that had reached the lower side of the filter was quantified. The composition of culture media had no significant effect on the motility of TS cells in Transwell chambers without a barrier of Matrigel (Fig. 4). In the Matrigel invasion system, there were no cells on the lower surface after 48 h of culture and few cells appeared after 72 h (without apparent difference between the different culture conditions, data not shown). A significant difference was observed first after 96 h of culture when the cell number was at least 200 in TS-M cultures (but no more than 400e500 so that accurate counting was still possible) (Fig. 4). In TSþM, only a few cells were found on the lower side of the filter (Fig. 4). By contrast, the number of penetrating cells significantly increased in TS cells grown in TS-M (Fig. 4). In order to identify the types of the invading cells at a molecular level, the cells from the bottom of the filters were collected and used for RT-PCR analysis (the number of PCR cycles was increased up to 40 in order to compensate for the low number of cells) (Fig. 5). The stem cell marker Eomes could not be detected in these cells while the giant cell marker Pl-1 was strongly expressed. There was also a weak expression of spongiotrophoblast marker Tpbp.

3.4. Up-regulation of MMPs and TIMPs during TS cell differentiation Using RT-PCR, the expression of MMP-9, MMP-14 and their inhibitors TIMP-1, -2, and -3 in TS cells cultured under different conditions was examined. All the MMPs and TIMPs checked were up-regulated along with TS cell differentiation stimulated in TS-M (Figs. 6, 7). In TS-M, the expression pattern was similar in cultures on plastic and those on gels of Matrigel. In TSþM, Matrigel contact slightly up-regulated the expression of MMP-9 (2-fold, P < 0.05) and TIMP-3 (3-fold, P < 0.05), as compared to that on plastic. In addition, the expression level of TIMP-3 in TS cells cultured in TS-M on Matrigel was similar to that of the same culture in TSþM, whereas on plastic TS-M stimulated expression of TIMP-3 by over 3-fold (P < 0.05). 4. Discussion The invasive behavior of differentiating TS cells was demonstrated during growth on gels of Matrigel (Fig. 2) and using Matrigel invasion chambers (Fig. 4). These results are consistent with a previous report where the influence of FGF on TS cell invasion was shown [12]. In the present communication we are reporting additional observations on expression of matrix metalloproteinases and their inhibitors and on effects of matrix contact. As the RT-PCR results suggest, the main

18

T. Lei et al. / Placenta 28 (2007) 14e21

Fig. 3. Analysis of marker gene expression in TS cells cultured under various conditions. TS cells were cultured on uncoated plastic and gels of Matrigel in TSþM and TS-M, respectively. After 96 h, total RNA was prepared from each sample and subjected to semi-quantitative RT-PCR; b-actin was included as a loading standard and PCR without RT () as negative controls. M: molecular weight markers. (A) Representative RT-PCR analysis; (B) relative expression levels of each mRNA were determined by densitometry and normalized to corresponding b-actin expression. Data were represented as means  SEM from three independent experiments.

Fig. 4. Quantitation of cell mobility and invasion. Invasion assay in Transwell chambers covered with Matrigel compared with motility assay (in which the filters were uncovered). Fifty thousand cells (harvested from maintenance cultures in TSþM and resuspended in TSþM and TS-M, respectively) were seeded into the upper chamber and cultured accordingly in TSþM or TS-M for 96 h. Numbers of penetrated cells were determined by counting the whole field of the lower surface of the filter under 100 magnification. Data were represented as means  SEM from three independent experiments.

Fig. 5. Gene expression in cells harvested from the lower surface of filters in invasion chambers. The cells that had moved to the lower surface of the filters (see Fig. 4) in TSþM and TS-M, respectively, were collected and subjected to RTPCR analysis of TS (Eomes), spongiotrophoblast (Tpbp), and giant cell (Pl-1) marker gene expression. The number of PCR cycles had to be increased up to 40 in order to compensate for the low numbers of penetrated cells. b-actin was included as control and PCR without RT () was performed as a negative control. RT-PCR for Eomes in TS cells grown separately in TSþM performed as a control gave a strong and clear signal (data not shown). M: molecular weight markers.

trophoblast subtype penetrating the ECM layer appears to be trophoblast giant cells. A few spongiotrophoblasts and/or glycogen trophoblast cells also may be involved, although actual data on the numbers of cells belonging to the different subpopulations are not available. In mice, trophoblast cells are found invading the uterus in two patterns [4]. The first type of invasion occurs soon after embryo implantation, and the cell population responsible for it is TGCs that invade the endometrium in a pattern tightly associated with spiral arteries. A second type of invasion appears at a later stage and is dominated by GlyT cells, which penetrate into the decidual stroma without obvious association with maternal blood vessels. In our in vitro invasion experiment, both Pl-1- and Tpbppositive cells, which represent the TGCs [13] and SpT/GlyT [14] cells, respectively, were found on the lower side of the invasion chamber filters. Thus, the two phases of invasion found in vivo (see above) may be recapitulated in vitro, although indications for two separate phases of invasion could not be

T. Lei et al. / Placenta 28 (2007) 14e21

Fig. 6. Analysis of MMP expression in TS cells cultured under various conditions. TS cells were cultured on plastic and gels of Matrigel, in TSþM and TS-M, respectively. After 96 h, total RNA was prepared from each sample and subjected to semi-quantitative RT-PCR; b-actin was included as a loading standard and PCR without RT () as negative controls. M: molecular weight markers. (A) Representative RT-PCR analysis; (B) relative expression levels of each mRNA were determined by densitometry and normalized to corresponding b-actin expression. Data were represented as means  SEM from three independent experiments.

assessed with this invasion assay. In our experiments, practically no cells were detected on the lower surface of the filters of invasion chambers before 96 h with cell numbers being still rather low at that time. Therefore, the penetrating cells found in our experiment appear to be true invading cells, rather than cells that bypass the Matrigel layer and migrate through the pores of the filter with the help of ‘‘pioneer’’ cells that degrade the ECM and thus may create paths in the Matrigel layer. The ability of differentiating TS cells to degrade ECM is also illustrated morphologically by the fact that in the culture on the gels of Matrigel (Figs. 1, 2), differentiated TS cells deeply invade into the gel when grown in TS-M. As trophoblasts invade the endometrium in vivo, they are confronted with various basement membranes and matrix substrates, which can influence the adhesion, spreading, migration and differentiation of the invading trophoblast cells [15]. Matrigel, a composite ECM rich in laminin and type IV collagen, may indeed be a model for matrix present around the invading trophoblast in the decidua [16]. Differentiated, i.e. potentially invasive, TS cells tend to form scattered colonies on Matrigel, rather than the monolayers typically found on plastic. It has been reported that the proliferation of TS cells was decreased on substrates of laminin [17], which is the major component of the basement membrane-like Matrigel [16]. However, further investigation is still needed to evaluate the effects of gels of

19

Fig. 7. Analysis of TIMP expression in TS cells cultured under various conditions. TS cells were cultured on plastic and gels of Matrigel, in TSþM and TS-M, respectively. After 96 h, total RNA was prepared from each sample and subjected to semi-quantitative RT-PCR; b-actin was included as a loading standard and PCR without RT () was used as negative controls. M: molecular weight markers. (A) Representative RT-PCR analysis; (B) relative expression levels of each mRNA were determined by densitometry and normalized to corresponding b-actin expression. Data were represented as means  SEM from three independent experiments.

Matrigel on the growth of TS cells in our system. The results of RT-PCR reveal that, compared to their counterparts on plastic, TS cells maintained on Matrigel in TSþM show a downregulation of Eomes and an up-regulation of Tpbp. Since Eomes and Tpbp are marker genes for trophoblast stem cells [18] and spongiotrophoblast cells [14], respectively, Matrigel appears to promote the differentiation of TS cells, even in culture conditions (TSþM) that favor the proliferation of stem cells when grown on plastic. As a result the proportion of spongiotrophoblast-type cells may be increased (although it must be pointed out that direct counts of the individual subpopulation cell numbers are not available so that this interpretation needs to be made cautiously). Besides FGF-4 and unidentified factors in MEF-CM [9], several other factors have been found to regulate TS cell differentiation [19,20]. Ectopic expression and overexpression of specific transcription factors can also override the effects of FGF-4 and MEF-CM and promote the terminal differentiation of TS cells [21]. In the culture, ECM can function as a scaffold for tissue morphogenesis, and provide cues for the regulation of stem cell proliferation and differentiation [22]. The differentiation-inducing components of Matrigel are

20

T. Lei et al. / Placenta 28 (2007) 14e21

unclear, but laminin should be an interesting candidate since it can regulate the differentiation of trophoblast cells in vitro [17]. In addition, various growth factors present in Matrigel may also have effects on trophoblast differentiation [23]. Degradation and remodeling of ECM in the endometrium is a crucial task for trophoblast cells. MMPs are instrumental in matrix-degradation by mouse trophoblast [24,25], they are a growing group of zinc-dependent enzymes that can degrade all the components in ECM. Their functions can be inhibited by TIMPs [8]. MMP-9 and MMP-14 have been reported to be highly expressed in the invading giant cells [26e28], and embryo-derived MMP-9 and uterus-derived TIMP-3 appear to be key regulators of ECM degradation by mouse trophoblast [29]. Based on the results of RT-PCR, we have found that the expression of both MMP-9 and MMP-14 is up-regulated along with the differentiation of TS cells, and also parallels the acquisition of invasiveness by differentiating TS cells. The expression levels of TIMP-1, -2 and -3 are also increased when the cells are induced to differentiate on plastic. This suggests that the invasive behavior of trophoblast is regulated in an autocrine manner, since the correct balance between the activity of MMPs and their inhibitors is crucial in the regulation of cell invasion [8]. When TS cells were exposed to Matrigel, the expression of MMPs and TIMPs was slightly increased, although in TSM there was no significant difference in the expression patterns between the cultures on plastic and Matrigel. When the cells were grown in TSþM, Matrigel appeared to up-regulate the expression of MMP-9 and TIMP-3 significantly. Besides, the exposure to Matrigel decreased the difference in the level of TIMP-3 between the TS cells cultured in TSþM and TS-M as observed on plastic. This is in agreement with the highly invasive characteristics shown in Figs. 1 and 2. During invasion into Matrigel, the expression of MMP-9 was also reported to be up-regulated in the differentiation of Rcho-1 cells, a rat trophoblast stem cell line which can only differentiate into giant cells [30], and appears to be regulated at the transcriptional level by factors involved in the giant cell differentiation. In our culture system, Matrigel promotes the differentiation of TS cells, possibly towards a spongiotrophoblast fate, correlated with the expression of MMP-9. This is consistent with the fact that ectoplacental cone is a source of MMP-9 [27]. In conclusion, the TS cell line used in this study provides an attractive tool to experimentally investigate the interrelationship between differentiation and invasiveness of trophoblast cells, as well as properties constituting the invasive phenotype like expression of MMPs and TIMPs, and should open possibilities for the study of their regulation. Acknowledgements We thank Dr. Rossant for TS cells, R. Brand, B. MarancaHuewel, and B. Gobs-Hevelke for cell culture and sample preparation, I. Kromberg for RNA manipulation, and D. Kittel for help with image processing. T.L. was financially supported by a scholarship from Deutscher Akademischer Austausch Dienst (DAAD) and China Scholarship Council (CSC).

References [1] Rossant J, Cross JC. Placental development: lessons from mouse mutants. Nat Rev Genet 2001;2:538e48. [2] Cross JC, Baczyk D, Dobric N, Hemberger M, Hughes M, Simmons DG, et al. Genes, development and evolution of the placenta. Placenta 2003;24:123e30. [3] Zybina EV, Zybina TG. Polytene chromosomes in mammalian cells. Int Rev Cytol 1996;165:53e119. [4] Adamson SL, Lu Y, Whiteley KJ, Holmyard D, Hemberger M, Pfarrer C, et al. Interactions between trophoblast cells and the maternal and fetal circulation in the mouse placenta. Dev Biol 2002;250:358e73. [5] Cross JC. Genetic insights into trophoblast differentiation and placental morphogenesis. Semin Cell Dev Biol 2000;11:105e13. [6] Carson DD, Bagchi I, Dey SK, Enders AC, Fazleabas AT, Lessey BA, et al. Embryo implantation. Dev Biol 2000;223:217e37. [7] Bischof P, Meisser A, Campana A. Mechanisms of endometrial control of trophoblast invasion. J Reprod Fertil Suppl 2000;55:65e71. [8] Vu TH, Werb Z. Matrix metalloproteinases: effectors of development and normal physiology. Genes Dev 2000;14:2123e33. [9] Tanaka SK, Kunath T, Hadjantonakis AK, Nagy A, Rossant J. Promotion of trophoblast stem cell proliferation by FGF4. Science 1998;282: 2072e5. [10] Hohn HP, Parker Jr CR, Boots LR, Denker HW, Hook M. Modulation of differentiation markers in human choriocarcinoma cells by extracellular matrix: on the role of a three-dimensional matrix structure. Differentiation 1992;51:61e70. [11] Hohn HP, Denker HW. Experimental modulation of cellecell adhesion, invasiveness and differentiation in trophoblast cells. Cells Tissues Organs 2002;172:218e36. [12] Hemberger M, Hughes M, Cross JC. Trophoblast stem cells differentiate in vitro into invasive trophoblast giant cells. Dev Biol 2004;271:362e71. [13] Carney EW, Prideaux V, Lye SJ, Rossant J. Progressive expression of trophoblast-specific genes during formation of mouse trophoblast giant cells in vitro. Mol Reprod Dev 1993;34:357e68. [14] Lescisin KR, Varmuza S, Rossant J. Isolation and characterization of a novel trophoblast-specific cDNA in the mouse. Genes Dev 1988;2: 1639e46. [15] Armant DR, Kameda S. Mouse trophoblast cell invasion of extracellular matrix purified from endometrial tissue: a model for peri-implantation development. J Exp Zool 1994;269:146e56. [16] Kleinman HK, McGarvey ML, Hassell JR, Star VL, Cannon FB, Laurie GW, et al. Basement membrane complexes with biological activity. Biochemistry 1986;25:312e8. [17] Klaffky E, Williams R, Yao CC, Ziober B, Kramer R, Sutherland A. Trophoblast-specific expression and function of the integrin alpha 7 subunit in the peri-implantation mouse embryo. Dev Biol 2001;239:161e75. [18] Hancock SN, Agulnik SI, Silver LM, Papaioannou VE. Mapping and expression analysis of the mouse ortholog of Xenopus eomesodermin. Mech Dev 1999;81:205e8. [19] Yan J, Tanaka S, Oda M, Makino T, Ohgane J, Shiota K. Retinoic acid promotes differentiation of trophoblast stem cells to a giant cell fate. Dev Biol 2001;235:422e32. [20] Tremblay GB, Kunath T, Bergeron D, Lapointe L, Champigny C, Bader JA, et al. Diethylstilbestrol regulates trophoblast stem cell differentiation as a ligand of orphan nuclear receptor ERR beta. Genes Dev 2001;15:833e8. [21] Hughes M, Dobric N, Scott IC, Su L, Starovic M, St-Pierre B, et al. The Hand1, Stra13 and Gcm1 transcription factors override FGF signaling to promote terminal differentiation of trophoblast stem cells. Dev Biol 2004;271:26e37. [22] Gerami-Naini B, Dovzhenko OV, Durning M, Wegner FH, Thomson JA, Golos TG. Trophoblast differentiation in embryoid bodies derived from human embryonic stem cells. Endocrinology 2004;145:1517e24. [23] Aplin JD, Lacey H, Haigh T, Jones CJ, Chen CP, Westwood M. Growth factoreextracellular matrix synergy in the control of trophoblast invasion. Biochem Soc Trans 2000;28:199e202.

T. Lei et al. / Placenta 28 (2007) 14e21 [24] Behrendtsen O, Alexander CM, Werb Z. Metalloproteinases mediate extracellular matrix degradation by cells from mouse blastocyst outgrowths. Development 1992;114:447e56. [25] Sharma RK. Mouse trophoblastic cell lines: II e Relationship between invasive potential and proteases. In Vivo 1998;12:209e17. [26] Alexander CM, Hansell EJ, Behrendtsen O, Flannery ML, Kishnani NS, Hawkes SP, et al. Expression and function of matrix metalloproteinases and their inhibitors at the maternal-embryonic boundary during mouse embryo implantation. Development 1996;122:1723e36. [27] Harvey MB, Leco KJ, Arcellana-Panlilio MY, Zhang X, Edwards DR, Schultz GA. Proteinase expression in early mouse embryos is regulated by leukaemia inhibitory factor and epidermal growth factor. Development 1995;121:1005e14. [28] Tanaka SS, Togooka Y, Sato H, Seiki M, Tojo H, Tachi C. Expression and localization of membrane type matrix metalloproteinase-1 (MT1-MMP) in trophoblast cells of cultured mouse blastocysts and ectoplacental cones. Placenta 1998;19:41e8. [29] Whiteside EJ, Jackson MM, Herington AC, Edwards DR, Harvey MB. Matrix metalloproteinase-9 and tissue inhibitor of metalloproteinase-3

[30]

[31]

[32]

[33]

[34]

21

are key regulators of extracellular matrix degradation by mouse embryos. Biol Reprod 2001;64:1331e7. Peters TJ, Albieri A, Bevilacqua E, Chapman BM, Crane LH, Hamlin GP, et al. Differentiation-dependent expression of gelatinase B/matrix metalloproteinase-9 in trophoblast cells. Cell Tissue Res 1999;295:287e96. Yi CH, Terrett JA, Li QY, Ellington K, Packham EA, ArmstrongBuisseret L, et al. Identification, mapping, and phylogenomic analysis of four new human members of the T-box gene family: EOMES, TBX6, TBX18, and TBX19. Genomics 1999;55:10e20. Colosi P, Talamantes F, Linzer DI. Molecular cloning and expression of mouse placental lactogen I complementary deoxyribonucleic acid. Mol Endocrinol 1987;1:767e76. Ku SY, Choi YM, Suh CS, Kim SH, Kim JG, Moon SY, et al. Effect of superovulation on the expression of tissue inhibitor of metalloproteinase3 in the murine endometrium. Gynecol Obstet Invest 2003;55:1e6. Apte SS, Fukai N, Beier DR, Olsen BR. The matrix metalloproteinase-14 (MMP-14) gene is structurally distinct from other MMP genes and is co-expressed with the TIMP-2 gene during mouse embryogenesis. J Biol Chem 1997;272:25511e7.