Reproductive Biology and Endocrinology Ets-2 and P53 Mediate Camp-induced Mmp-2 Expression, Activity and Trophoblast Invasion

Background: We have previously shown that Matrix metalloproteinase (MMP)-2 is a key-enzyme in early trophoblast invasion and that Protein Kinase A (PKA) increases MMP-2 expression and trophoblast invasion. The aim of this study was to examine MMP-2 regulation by PKA in invasive trophoblasts: JAR choriocarcinoma cell-line and 6-8 w first trimester trophoblasts.


Background
During implantation the trophoblast exhibit behavior similar to tumor cells, including tissue invasiveness, metastasis, loss of contact inhibition, escape from immune surveillance and massive proliferation ability [1]. However, in sharp contrast to aggressive tumors, trophoblast invasion process is a spatial and temporal strictly regulated process, controlled by multiple interactions between cells and environment, mediated by several factors including growth factors, hormones and cytokines [2,3].
Trophoblast are highly invasive due to the secretion of extracellular proteases, like Matrix Metalloproteinases (MMPs), mediating degradation of the extracellular matrix (ECM), and of their balancing inhibitors Tissue inhibitors of Metalloproteinases (TIMPs) [4][5][6]. The two gelatinases, MMP-2 and MMP-9, digest type IV collagen, the main component of the basement membrane and are therefore regarded as key enzymes in trophoblast invasion [5,[7][8][9]. MMP-2 and MMP-9 expression is differential during the first trimester of pregnancy. MMP-2 production is dominant until 8 weeks of gestation and then declining, whereas MMP-9 production significantly increases from 8 weeks, causing a shift of dominant gelatinase from MMP-2 to MMP-9 from 9 week of gestation [8,9]. The implantation process occurs by two waves of invasion by the trophoblast, the first strictly controlled to the implantation site and completed at 8 weeks of gestation, and the second wave deep into the uterine wall at the end of first trimester [10,11]. MMP-2 is likely to be the key-enzyme in the first physiological invasion wave, whereas MMP-9 may be the key-enzyme in the second invasion wave.
The JAR cell line, obtained from choriocarcinoma, is comprised of highly invasive trophoblast, expressing mainly MMP-2 and to a much lesser extent MMP-9 [8,12,13]. JAR cells were shown in our lab to resemble early 1 st trimester trophoblast in gelatinase profile and invasiveness, and can therefore be used as a good model for studying early invasive trophoblast regulation [8]. Several factors with importance in embryo implantation act via cyclic adenosine monophosphate (cAMP)-protein kinase A (PKA) signal transduction pathway, including human chorion gonadotropin (hCG), the primary signal of an implanting pregnancy [14,15]. hCG was found to increase collagenolytic activity, trophoblast invasion and migration [15][16][17], and MMP-9 [18]. Forskolin is a prototypical stimulator of the cAMP pathway by direct activation of adenylate cyclase [19]. In a previous study, we showed that Forskolin (PKA) induces MMP-2 and MMP-9 and trophoblast invasiveness in both JAR cells and 1 st trimester trophoblasts [8]. Forskolin has been used also in other studies to induce invasion or invasion-related molecules [19,20]. Additionally, Forskolin/cAMP has been used to induce syncytiotrophoblast formation [21,22]. The use of celllines of different origin as well as primary or term trophoblast may be the cause of these differential effects.
Most MMPs are regulated at the transcriptional level, by several cytokines and growth factors, which influence multiple signaling pathways [23]. The MMP gene promoter contains several cis-regulatory elements, often acting synergistically, with varying importance and effect, depending upon cell-type and inducer [6]. The MMP -2 promoter lacks a typical TATA-box, but is not refractory to modulations as previously thought [24,25]. The transcriptional regulation on MMP-2 in trophoblasts is not yet clear. The aim of this study was to investigate the transcriptional regulation of MMP-2 in cAMP-activated invasive trophoblast. Other MMP regulation levels include mRNA stabilization, protein activation or inhibition, surface localization and secretion; however these levels were not part of the present study.

Cell culture
The JAR (Jar, HTB 144, ATCC) human choriocarcinoma line was a generous gift from Pr. Hochberg, Hebrew University, Israel. JAR cells were cultured in M-199 medium containing 10% Fetal Calf Serum (FCS) and penicillin/ streptomycin (all Beit-Ha'Emek, Israel) at 37°C in a humidified, 5% CO 2 atmosphere. After 24 h, medium was changes to M-199 medium with 1.5% serum, in the presence or absence of Forskolin 10 μM (Sigma, ST Louis, MO, USA).

Isolation and cultivation of human cytotrophoblast
Human trophoblast cells were obtained from legal abortions (6 to 8 weeks), with the approval of the local ethical committee (in compliance with the Helsinki Declaration) and the consent of the participating patients. Trophoblast cells were isolated according to a protocol generously provided by Desoye G, clinic of Obstetrics and Gynecology, Austria. Briefly, placenta were washed in saline, then the minced trophoblastic villi were digested by 0.25% trypsin and DNase I (both Sigma, ST Louis, MO, USA), and CTB separated from blood cells and decidua on a discontinuous Percoll gradient (Sigma, ST Louis, MO, USA). Contaminating leukocytes were removed by immunopurification with anti-CD45RB (DAKO, Glostrup, Denmark) coupled to magnetic particle. This method supplies a 95-98% purity of trophoblast, including all trophoblast sub-groups [26]. Cells were cultured in M-199 medium supplemented with 1.5% FCS and 1% penicillin/streptomycin for the various experiments. molecular weight and commercial standards (data not shown). Bands on each gel were compared to their control band and quantified with a densitometer system, endowed with Bio-Capt and TINA software (Raytest, Staubenhardt, Germany).

Matrigel invasion assay
Matrigel (1 mg/ml) (BD Biosciences, Franklin Lakes, NJ, USA) diluted in serum free media was added to upper chamber of 24-well transwell plate (8 μm pores, Corning, MA, USA). 10 5 cells in 100 μl media were added to the upper chamber, and 500 μl media added to the lower chamber. 10 μM Forskolin was added to medium in upper and lower wells and cells incubated at 37°C for 48-72 hours. Cells (10 5 /well) seeded simultaneously in the same media in a well without transwell, served as reference of total seeded cells. After incubation non-invaded cells on top of the transwell were scraped off with a cotton swab and the amount of invaded cells in the lower well as well as the amount of total seeded cells was evaluated with XTT Reagent kit. The percent of invasion was calculated as: Invasion was expressed as Invasion Index (Percent of control).

RNA extraction
Total cellular RNA was extracted using a total RNA isolation kit EZ-RNA (Beit HaEmek, Israel) according to manufacturer's instructions. RNA concentration was determined spectro-photo-metrically.

Northern blot analysis
15 μg of total RNA was electrophoresed on a 1% denaturating agarose gel and transferred to a nitrocellulose membrane. The membrane was hybridized with 2 μg/lane complimentary digoxigenin-labeled DNA probe for MMP-2 or GAPDH (Biognostik, Göttingen, Germany) using the digoxigenin (DIG) Easy Hyb system, according to manufacturer's instructions (Roche, Basel, Switzerland). The probes were previously labeled with digoxigenin by terminal transferase labeling kit (Roche, Basel, Switzerland). The membrane was then rinsed in 2× SSC/ SDS and in 0.1× SSC/SDS. Bands were detected with antidigoxigenin-AP antibody and the chemiluminescent substrate CSPD (both Roche, Basel, Switzerland) on X-ray film.

Preparation of nuclear extract
JAR cells or 1 st trimester trophoblasts were incubated in serum-free medium for 24 hours, with or without 10 μM Forskolin. Cells were lysed with swelling buffer (10 mM Hepes, 10 mM KCL, 1 mM EDTA, 1 mM DTT, 1 mM PMSF, 10 μg/μl Leupeptin, 50 μg/ml Aprotinin) for 15 min at 4°C, then 10% NP-40 was added for lysis and cells centrifuged at 4°C for 3 min at 3000 RPM. The cytosol protein extract was collected and the remaining pellet resuspended in extraction buffer (20 mM Hepes, 200 mM NaCl, 1 mM EDTA, 1 mM DTT, 1 mM PMSF, 10 μg/μl Leupeptin, 50 μg/ml Aprotinin), vortexed and kept at 4°C for 15 min, then centrifuged at 4°C for 5 min at 12000 RPM. The protein concentration was determined with Bradford reagent assay (Bio-Rad laboratories, Washington, DC). Nuclear extract was assayed for the presence of transcription factors, entering the nucleus in response to stimulus.

Statistical methods
All data are expressed as mean ± SEM. Statistical analysis of the data was performed using Student's t-test. P < 0.05 was considered significant.

The effect of forskolin on MMP-2 expression in JAR and 6-8 w trophoblast
In order to explore the effect of Forskolin upon MMP-2 expression in JAR cells both Northern blot and RT-PCR was performed using the same RNA samples. In order to compare the results of the cell-line with primary cells RT-PCR was performed with 6-8 w 1 st trimester trophoblast. Forskolin was added to JAR or 6-8 w trophoblast culture for various length of time, in order to find the optimal exposure. Figure 1 summarizes the results. The mRNA of MMP-2 was increased in JAR cells by Forskolin (10 μM), reaching a maximum after 24 hours of incubation (382% ± 60, P = 0.02, Fig. 1A, 363% ± 90, P = 0.03, Fig. 1B). In 6-8 w trophoblast a maximum was reached within 1 hour after adding Forskolin (147% ± 17, P = 0.03, Fig. 1C). These exposures were chosen for further studies as optimal exposure for MMP-2 induction.

Effect of forskolin on JAR nuclear protein binding to oligonucleotides sequences resembling binding sites found in the MMP-2 promoter
In order to detect transcription factors possibly involved in cAMP induction of MMP-2 in trophoblasts, we performed Gelshift assay (EMSA) of nuclear extract obtained from cells cultured for 24 hours in the presence or absence Effect of forskolin on MMP-2 expression in JAR and 6-8 w trophoblast of 10 μM Forskolin. The binding of nuclear proteins to labeled oligonucleotides (Table 1), with sequences resembling binding sites found in the MMP-2 promoter, was explored. This method has been commonly used to detect possible binding sites and factors involved in transcriptional regulation [27][28][29]; Figure 2 summarizes the results. Forskolin increased binding to ETS, C/EBP, AP-2 and p53 DNA binding sites ( Fig. 2A). No increase was found in binding to oligonucleotides SP-1, NF k B, AP-1 and CREB (data not shown). In order to assure binding specificity, competing non-labeled 8x excess oligonucleotide was added to reaction mixture. The non-labeled excess oligonucleotide strongly reduced the gelshift band, hereby ensuring binding specificity. Additionally, in order to assure involvement of the PKA signaling pathway, specific PKA inhibitor H89 (100 μm) was added to cell culture and gelshift performed ( Fig 2B). H89 reduced binding activity and thereby confirmed the involvement of the PKA pathway (lane 3, 6,9,12).

Identification of transcription factors binding to oligonucleotides involved in forskolin-induced MMP-2 expression
In order to explore transcription factors involved in Forskolin-induced MMP-2 expression, supershift EMSA was performed. In this assay, antibodies that recognize and identify the specific protein, bind to the protein-oligonucleotide complex, resulting in a higher, heavier band. Specific antibodies belonging to the families of transcription factors capable of binding to the relevant oligonucleotides were added to nuclear extract, prior to the binding reaction with oligonucleotides. Figure 2C shows the results.

Detection of transcription factors induced by forskolin and possibly involved in MMP-2 regulation
EMSA can only show which transcription factors perform increased binding to their respective binding sequence under stimulation of Forskolin but not whether these factors indeed bind to the MMP-2 promoter. In order to further verify our results we performed a complementary study, using western blot to detect transcription factors increased in the nucleus by Forskolin. Nuclear cell extract was prepared from JAR cells cultured 24 hours in the presence or absence of 10 μM Forskolin and western blot performed with antibodies against the same transcription factors previous tested with EMSA. Figure 3 shows the results. Forskolin increased significantly nuclear expression of AP-2 by 123% ± 11, P = 0.043 (Fig. 3A), of C/EBPε by 150% ± 24, P = 0.049 (Fig. 4B), of C/EBPλ by 159% ± 22, P = 0.007 (Fig. 3B), of phospho-p53 by 169% ± 34, P = 0.021 (Fig. 3C) and of Ets-2 by 151% ± 22, P = 0.03 (Fig.  3D).

Involvement of Ets-2 in forskolin-induced MMP-2 expression and trophoblast invasiveness
In order to further examine the involvement of Ets-2 in Forskolin-induced MMP-2 transcriptional regulation, we performed a silencing study using antisense against Ets-2.
Ets-2 was inhibited by transfection of JAR cells with a specific antisense against Ets-2. Cells were then cultured in the absence or presence of Forskolin and four different parameters were analyzed. Figure 4A-F shows the results. Fluorescent antibody staining against Ets-2 and western blot analysis confirmed the inhibitory effect of the antisense. The transient transfection with Ets-2 antisense caused a 49% decrease in Ets-2 expression in non treated cells (100% vs.51% ± 8, P < 0.05), and a 65% decrease in Forskolin-treated cells (139% vs. 74% ± 6, P < 0.05) (Fig.  4A, B, C).

The effect of forskolin on Ets-2 and p53 in 6-8 w 1st trimester trophoblasts -verification of similarity with cellline results
Trophoblast cell-lines are excellent models for studying the mechanism of implantation, but results should always be verified with primary cells. We therefore performed EMSA and immunofluorescent staining of Forskolintreated 6-8 w 1 st trimester trophoblastic cells in order to verify the presence and involvement of Ets-2 and p53 in these primary cells. 6-8 w 1 st trimester trophoblast were cultured in the presence or absence of 10 μM Forskolin, then cells were fixed and stained with fluorescent antibodies against Ets-2, p53 and MMP-2, or total nuclear protein was extracted and EMSA performed with oligonucleotides Ets and p53. Figure 6 summarizes the results. Forskolin increased Ets-2, p53 and MMP-2 expression significantly by 164 ± 7, 125 ± 12 and 162 ± 13, respectively (Fig. 6A, B), and increased binding activity to oligonucleotides p53 and Ets in 6-8 w 1 st trimester trophoblasts, like previously found with JAR cells (Fig. 6C). This confirms the resemblance between 6-8 w 1 st trimester trophoblast and JAR cells and indicates that the results obtained with JAR cellline may be assumed to apply to early trophoblast with regards of gelatinolytic activity, regulation and invasiveness.

Discussion
We have found that Forskolin increases both trophoblast MMP-2 activity and MMP-2 mRNA expression. The maximal mRNA response in JAR cells appeared only after 24 hours of incubation, suggesting that Forskolin may stimulate MMP-2 by inducing other, early response genes. The MMP-9 gene has similarly been reported to be induced within 24 h in trophoblasts by 12-O-tetradecanoyl phorbol 13-acetate (TPA) and TNF, through transcription of early response genes [29]. In 6-8 w trophoblasts, in contrast, Forskolin induced MMP-2 mRNA within 1 h, indicating that Forskolin apparently can stimulate MMP-2 expression more directly, probably through initiating phosphorylation and migration of already present transcription factors. The controversy found here between JAR cells and 6-8 w trophoblasts, in regards of time for maximal mRNA response to Forskolin, underlines that some differences are to be expected between primary cells and a cell-line. One can speculate that in choriocarcinoma cells like JAR, an altered presence of transcription factors or activation of phosphorylation cascades may be part of the malignancy of these cells. The difference observed in ultimate magnitude between JAR cells and 1 st trimester trophoblasts in response to Forskolin is consistent with our previous findings, showing the invasive choriocarcinoma cell-line to be highly expressing MMP-2, both in response to Forskolin and EGF [8]. The primary trophoblast are less invasive and express less MMP-2 than the JAR cells.
Forskolin increased the binding of nuclear proteins to oligonucleotides corresponding to the consensus sequence of AP-2, C/EBP, p53 and Ets response elements in the MMP-2 promoter. These binding activities were induced specifically through PKA, since specific PKA-inhibitor H89 reduced this effect. The supershift EMSA assay revealed, that transcription factors AP-2α, phospho-p53, Ets-2 and C/EBPε and -λ bind to their respective binding sequence under stimulation of Forskolin. Forskolin also increased the protein expression of Ets-2, C/EBPε, C/EBPλ and AP-2α, and the phosphorylation of p53 as shown by western blot. These transcription factors may therefore be involved in transregulation of the MMP-2 promoter by Forskolin.
Several transcription factors have been found to be involved in MMP-2 regulation in various cells, including AP-2, AP-1 and p53 [28,35,36]. Only a few studies of MMP-2 transcriptional regulation in trophoblast have been published, and have focused mainly on the signaling pathway involved. In 1 st trimester trophoblast the MAPK/ Erk1/2 pathway was found to be involved in induction of MMP-2 and MMP-9 by Cyclosporin A [37]. A cytoplasmatic tyrosine kinase named focal adhesion kinase was found to be involved in cell differentiation, adhesion, proliferation and MMP-2 activity in early trophoblast [38]. In another study from our lab, progesterone was found to regulate MMP-2 through differences in progesterone-receptor isoforms, apparently by binding to coactivators or transcription factors, since there is no progesterone response element in the MMP-2 promoter [39]. hCG, which like Forskolin acts through PKA, induces MMP-2 expression and trophoblast invasion through ERK 1/2 and PI3K/AKT pathways [40]. Our study is the first to focus on transcription factors, which mediate MMP-2 transcriptional regulation in trophoblasts. We chose to focus our further investigation upon two transcription factors, Ets-2 and p53, which have previously been suggested to be involved in the implantation process. We showed that that Ets-2 mediates Forskolin-induced MMP-2 activity, since Forskolin increased Ets-2 expression in trophoblast and since antisense inhibition of Ets-2 reduced the Forskolin-induced MMP-2 gelatinolytic activity and trophoblast invasion. Inhibition of Ets-2 did not significantly reduce MMP-2 transcription, indicating that this transcription factor alone, although essential to MMP-2 activity and function, is not sufficient to MMP-2 transcriptional regulation. Based on our results we hypothesize that Ets-2 may effect MMP-2 regulation on a post-tran-scriptional level, however further investigation is necessary to elucidate the exact mechanism.
The ETS transcription factor family is characterized by an highly evolutionary-conserved DNA-binding domain, that binds to a purine-rich GGA(A/T) core sequence [41,42]. Ets members direct cytoplasmic signals to control gene expression and mediate multiple signaling pathways such as the Ras-Map Kinases, Erk1,2, p38, JNK and the PI3 kinases, which control their activity, protein partnership and specification of downstream target genes [41-

43
]. Ets members have been shown to interact with several transcription factors including AP-1, pit-1, NF k B and SP-1 [34,42,44]. Ets proteins play important roles in cell development, differentiation and proliferation, and participate in malignancy of tumor cells, including invasion and metastasis, by activating the transcription of several cancer-related genes such as proteases and angiogenesisrelated genes [41]. Binding sites for the ETS family are found in promoters of the MMP family [45], and regulation of matrix degrading proteases by Ets factors in tumor invasion and metastasis is well established [46]. Ets-2 is expressed in many cell types in developing mouse embryos. Targeted deletion of Ets-2 in mice results in retardation and death of the mouse embryo before 8.5 days of development [34]. These embryos show defects in the extraembryonic tissue gene expression and function, including deficient expression of MMP-9, MMP-13 and MMP-3 and die as a result of trophoblast failure [34,47]. Apparently. Ets-2 is not essential for specifying trophoec-toderm before the morula-blastocyst transition, since blastocyst do form and initiate implantation. Properly Ets-2 coordinates early expression of genes providing full characteristics to cytotrophoblast, including the production of key enzymes and hormones [47]. MMP-2 may be one of these genes. Ets proteins mediate the control of several genes characteristically up-regulated in human trophoblast and involved in trophoblast differentiation and placenta development [32]. This includes the hCG-α and hCG-β subunit gene, placental lactogen II, p-450 side chain cleavage enzyme, decidual/trophoblast prolactin related gene and MMP-1, -3 [48][49][50][51]. The hCG subunits are activated in stimulation to cAMP through interactions between Ets and CREB [32,47].
We also found that Forskolin increases p53 expression, and that p53 mediates Forskolin-induced MMP-2 expression, gelatinolytic activity and both basic and PKAinduced trophoblast invasion. In contrast to Ets-2, tran- sient transfection of antisense mRNA to p53 into trophoblasts decreased MMP-2 transcription, both in control and in Forskolin-treated cells. This indicates, that p53 may be involved in the constitutive expression of MMP-2, and additionally mediates the response to PKA. We have recently published that p53 is involved in MMP-2 regulation and trophoblast invasion by EGF [52]. p53 was among the transcription factors increased under EGF stimulation (p53, SP-1, AP-2α and -γ, C/EBPε and-λ). p53 antisense inhibition was found to decrease EGF-induced MMP-2 mRNA expression, secretion and trophoblast invasiveness. Together our two studies emphasize the con-clusion that p53 is involved in MMP-2 regulation in trophoblast. Since AP-2α and C/EBPε and -λ were also increased by both stimulators, they should be further examined by functional assays. Other transcription factors, like ETS-2, may be specific for a certain pathway.

Inhibitory effect of Ets-2 antisense upon forskolin-induced MMP-2 expression and secretion and trophoblast invasiveness
p53 is a 53 kDa nuclear transcription factor, that recognizes and binds to a specific DNA consensus sequence consisting of two copies of the 10 bp motif 5'-PuPu-PuC(A/T)(T/A)GPyPyPy-3', separated by 0 to 13 bp [53]. p53 acts as an activator by binding to this sequence and can also act as a transcriptional repressor of promoters Inhibitory effect of p53 antisense upon forskolin-induced MMP-2 expression and secretion and trophoblast invasiveness lacking a p53 binding site [54][55][56]. p53 activates genes involved in DNA repair [30]. Biologically p53 induces G1 arrest and apoptosis following DNA damage [57], and is therefore known as a common tumor suppressor [58]. Many p53 functions are mediated through the activation or repression of a series of genes, and this includes the MMP-2 gene [36]. Mutation of p53 may inactivate the growth regulatory function and cause a loss of tumor-sup-pressive activity [59]. Mutation at the specific DNA binding domain have been identified in nearly 50% of all human cancers [60]. p53 is present in the preimplantation human embryo [61,62], and is expressed throughout mouse implantation development. The p38 MAPK pathway was found to regulate p53 expression in 2-8 cell mouse embryos [63]. p53-deficient mice are viable and reproductive, however in p53-null (p53(-/-) the placenta The effect of forskolin on Ets-2 and p53 expression in 6-8 w 1 st trimester trophoblasts Figure 6 The effect of forskolin on Ets-2 and p53 expression in 6-8 w 1 st trimester trophoblasts. A, Representative photos taking with a confocal microscope of 6-8 w 1 st trimester trophoblast, magnification × 60. 6-8 w 1 st trimester trophoblast, cultured with or without Forskolin (10 μM), were fixed and stained with a fluorescent antibody against Ets-2, p53 or MMP-2. B, Density of p53, Ets-2 and MMP-2 staining was quantified with Image software and presented by Bar graph, left-Ets-2, middle-p53, right-MMP-2. C, Binding of nuclear proteins extracted from 6-8 w 1 st trimester trophoblast cells treated with or without Forskolin (10 μM) to dig-labeled oligonucleotides Ets and p53 was assessed with EMSA. Results represent 2 different experiments performed in triplicates.
is altered [64]. In rodent placenta, a decrease in p53 is part of the transition of trophoblast from a state of proliferation to differentiation in trophoblast [33]. p53 has a short half-live, and is therefore normally present at very low concentrations in cells [30]. However, p53 is detectable in the nucleus and cytoplasm of cytotrophoblasts and syncytiotrophoblasts [30,65,66], and this apparently overexpression of p53 in trophoblasts may be controlling excessive proliferation in normal placentation [67]. It appears that most p53 protein forms complexes in the cytoplasm and tetramers in the nuclei in cytotrophoblasts, suggested to stabilize p53 and hereby prolong the half-life [30,31]. p53 is correlated with enhanced apoptosis in trophoblast [68], and can be induced by hypoxia [69]. Increased p53 and apoptosis was found in different pregnancy complications, such as preeclampsia [69,70], fetal growth restriction [68] and Gestational Trophoblastic diseases (GTDs) [71]. In invasive hydatidiform moles, characterized by abnormal growth of chorionic tissues with various degrees of local invasion and metastasis, MMP-2 and mutant p53, but not wild type p53 (wtp53), were increased, as compared to normal placenta, and may be involved in this pathology [72]. However, since mutant p53 may still perform normal suppressive activity, depending on the localization of the mutation, and since other reports found wtp53 to be elevated in hydatidiform moles [73][74][75], the role of wtp53 in this pathology is not clear. A recent article described that exogenous p53, but not endogenous p53, decrease MMP-9 in cytotrophoblasts [76]. In their report, in contrast to our results, p53 inhibition had no effect on either MMP-2 or MMP-9, and the authors suggested that p53 is inactivated in trophoblasts, maybe through formation of high molecular weight complexes. These controversial results regarding MMP-2 may result from differences in experimental design, including the use of JAR cells as a model as well as primary trophoblast of different gestational ages (early versus late 1 st trimester trophoblast).
The examined transcription factors in this study are involved in transactivation and transrepression of many genes, thus the EMSA does not allow us to conclude, that MMP-2 is one of them. However, the studies using transfection of p53 and Ets-2 antisense in JAR demonstrate that at least p53 is involved in Forskolin-induced MMP-2 expression, while Ets-2 may effect post-transcriptional regulation. Our results suggest that MMP-2 regulation in trophoblast is mediated by several transcription factors, including p53 and Ets-2, however since MMP-2 transcription, secretion and trophoblast invasion were not completely diminished, other transcription factors must be involved. This may include C/EBP-ε, C/EBPλ and AP-2α, which should be subject to further study.