ETS1 Induces Human Trophoblast Differentiation

Abstract

A possible role for the transcription factor v-ets avian erythroblastosis virus E26 oncogene homolog 1 (ETS1) in human trophoblast cell differentiation was examined using a highly enriched fraction of human mononuclear cytotrophoblast cells (CTBs) that differentiate spontaneously in vitro to a multinucleated syncytiotrophoblast cell (STB) phenotype. ETS1 mRNA and protein levels were abundant in freshly isolated CTBs and decreased as the cells differentiated. Silencing of ETS1 expression in freshly prepared CTBs markedly attenuated syncytialization, as demonstrated by desmoplakin staining, and blocked the induction of syncytin, the transcription factor activator protein-2α, placental lactogen, and other STB-specific genes. Conversely, overexpression of ETS1 in primary trophoblast cells induced STB marker gene mRNAs and transactivated each of the gene proximal promoters. Taken together, these findings strongly suggest a critical role for ETS1 in the induction of human villus CTB differentiation. The effect of ETS1 on syncytialization likely results, at least in part, from inhibition of syncytin expression, whereas the induction of STB marker genes likely results in part from transactivation by activator protein-2α.

The trophoblast layer of the human placental villus is composed of mononucleated cytotrophoblast (CTB) cells and multinucleated syncytiotrophoblast (STB) cells (for review see reference ¹). The CTB cells, which are located between the STBs and the basement membrane of the trophoblast layer, function as precursors of the STB (1). The STBs, which form the outermost cells of the villus, are in direct contact with maternal blood. The differentiated STB perform a wide variety of critical functions throughout pregnancy, including the exchange of substrates, gases, and other factors between the maternal and fetal circulations and the synthesis and secretion of protein and steroid hormones and growth factors.

The differentiation of villus CTBs to a syncytiotrophoblast phenotype has been studied extensively using primary cultures of villus CTBs and human choriocarcinoma cells as model systems. Primary human CTB cells prepared by enzymatic digestion of whole placental tissue spontaneously aggregate and fuse to form a STB phenotype that expresses abundant amounts of placental lactogen (PL), CRH, and other STB-specific genes (for review see references 2 and 3). By day 3 of culture, about 50% of the mononuclear CTB cells have fused to form a syncytium; by day 6, 80% or more have fused. The in vitro differentiation results in the induction of genes involved in substrate transport, hormone synthesis and secretion, metabolism, and other functions of terminally differentiated STB. Choriocarcinoma cells, which fuse to form a syncytium upon exposure to cAMP, have also been used as a model to study trophoblast differentiation. However, these cells do not synthesize or release many of the proteins expressed by cultured primary trophoblast cells.

Although microarray and other studies have identified genes induced and repressed during villus CTB differentiation, the transcription factors, signaling molecules, and other factors that regulate trophoblast differentiation are incompletely understood (for review see reference ⁴). A better understanding of the regulation of human villus differentiation is important because abnormalities in differentiation are present in many pathologic conditions of pregnancy associated with increased maternal and/or fetal morbidity, including preeclampsia and fetal growth restriction.

For many years, our laboratory has examined the roles for syncytin and the transcription factor activator protein-2 (AP-2)-α (also known as TFAP2A) in the differentiation of villus CTB cells. Syncytin (endogenous retrovirus group W, member 1), a glycosylated transmembrane protein derived from the envelope protein of human retrovirus human endogenous retrovirus-W (5, 6), is critical for the induction of syncytialization. The transcription factor AP-2α is critical for the induction of human PL, human chorionic gonadotropin (CG)-β, CRH, human GH (hGH) variant, pregnancy-specific glycoprotein 1 (PSG1), and other STB-specific genes but has no effect on syncytialization (7).

Several lines of evidence suggest that the transcription factor v-ets avian erythroblastosis virus E26 oncogene homolog 1 (ETS1) is involved in villus CTB differentiation. ETS1 is expressed in first- and third-trimester human placenta (8), and DNA microarray studies have demonstrated that ETS1 mRNA is expressed in human CTB undergoing in vitro differentiation (2). ETS expression is evident in many other developing tissues, including development of the lymphoid and myeloid lineages, brain and central nervous system, bone, and mammary gland (9). ETS transcriptions factors, such as ETS1, regulate numerous genes that are involved in stem cell development, cell senescence and death, and tumorigenesis (10–12). Furthermore, ETS domain proteins, which function as either transcriptional activators or repressors, are regulated by MAPK pathways (13), which are known to regulate the expression of PL and other genes induced during villus trophoblast differentiation.

In this study, we have examined the role for ETS1 in the regulation of human villus trophoblast differentiation using an in vitro model developed in our laboratory. ETS1 expression during differentiation was knocked down with ETS short hairpin RNAs (shRNAs) and an small interfering RNA (siRNA), and the effect of ETS1 silencing on syncytialization and STB marker genes was assessed. Syncytialization was evaluated by immunohistochemistry with an antiserum to desmoplakin, a cytoskeleton protein that has been used extensively to study cell fusion. During cell fusion, the cytoskeleton undergoes rearrangement and cell junctions disappear. The expressions of syncytin, AP-2α, and other STB marker genes were examined by real-time PCR and Western blot analyses.

Materials and Methods

Localization of ETS1 in placental villi by immunohistochemistry

The cellular localization of ETS1 protein in term placental tissue was performed by immunohistochemistry of placental paraffin blocks from grossly unremarkable areas of placenta previously fixed in 10% buffered formaldehyde solution using double immunostaining with anti-ETS1 and anti-E-cadherin antiserums (details of all antibodies shown in Table 1). Briefly, for dual immunostaining, 0.4-μm sections were mounted on superfrost plus slides and then stained using an automated Ventana immunostainer (Ventana Medical Systems). The following protocol was used: deparaffinization, cell conditioning with EDTA for 30 minutes, incubation with anti-ETS1 antiserum (Genway) for 8 minutes, and E-cadherin antiserum (Cell Marque) for 52 minutes, followed by basic Ultraview DAB and RedMap detection systems (Ventana Medical Systems). Slides were then rinsed with reaction buffer and counterstained with hematoxylin for 4 minutes, followed by postcounterstaining with bluing reagent for 4 minutes. Appropriate positive and negative controls were used. E-cadherin staining distinguishes between trophoblastic cells, which stain positively, and fibroblast, endothelial, and Hofbauer cells, which did not stain. Color deconvolution of selected dual stained microscope images was performed to distinguish between ETS1 and E-cadherin immunostaining using ImageJ morphometry software (National Institutes of Health, Bethesda, Maryland) selecting the region of interest sampling tool (14).

Table 1.

Antibodies

Peptide/Protein Target	Name of Antibody	Manufacturer, Catalog Number, and/or Name of Individual Providing the Antibody	Species Raised (Monoclonal or Polyclonal)	Dilution Used
ETS1	Human ETS1 (Ab-38) antibody	Genway; GWB-ASC17	Rabbit, polyclonal	1:50 (IHC/WB)
E-cadherin	Anti-E-cadherin	Cell Marque; EP700–4497	Rabbit monoclonal	1:50
Desmoplakin	Antidesmoplakin I + II antibody (2Q400)	Abcam; ab16434	Mouse monoclonal	1:50
IgG	Antimouse IgG	Life Technologies; A-21203	Donkey polyclonal	0.111 111 111
β-Actin	Anti-β-actin	Dr James Lessard (Cincinnati Children's	Mouse monoclonal	1:10 000
		Hospital Medical Center, Cincinnati, Ohio)

Abbreviations: IHC, immunohistochemistry; WB, Western blot.

Details of the antibodies used in this study are included in Table 1.

Time course of ETS1 mRNA and protein levels during trophoblast differentiation

The time course of ETS1 mRNA and protein levels during trophoblast differentiation was determined in cultured human trophoblast cells by real-time PCR and Western blot analysis, respectively. The cells were prepared by enzymatic dispersion of term placentas with trypsin and deoxyribonuclease followed by isopycnic centrifugation through 40% Percoll (2, 15, 16). The isolation procedure resulted in an enriched fraction of CTB that was greater than 95% purified. The cells were then cultured in DMEM with 10% fetal bovine serum for up to 6 days with frequent medium changes every 2 days. The protocol for obtaining placentas was approved by the Human Investigation Committees of the University of Cincinnati and the Cincinnati Children's Hospital Medical Center.

Total RNA was extracted from the cells using Trizol (Invitrogen) according to the manufacturer's specifications. Two micrograms of total RNA were reverse transcribed using SuperScript II reverse transcriptase (Invitrogen) or the high-capacity RNA to cDNA kit (Life Technologies). Real-time PCRs were performed using a Stratagene Mx3000P qPCR machine (Stratagene). Quantitative PCR amplifications were performed using the Taqman gene expression master mix that contains Rox as a passive internal reference dye (Life Technologies). The TaqMan primers contained internal probes and were used according to the manufacturer's instructions with a 20-μL final volume. The PCR program consisted of a 2-minute incubation at 50°C, followed by a 10-minute incubation at 95°C, and then 40 cycles at 95°C for 15 seconds and then 60°C for 1 minute. The primers used for quantitative PCR are shown in Supplemental Table 1.

Western blot analyses were performed as previously described (17) using an anti-ETS1 antiserum (Genway).

The effects of ETS1 knockdown on syncytialization

In immunocytochemistry studies, freshly prepared human CTBs were exposed to an ETS1 siRNA [Silencer Select Validated siRNA for Ets1 (catalog number 4392420, clone s4848)] or a nonsilencing siRNA control (Silencer Select negative control number 1 siRNA, both purchased from Life Technologies) and cultured as above except that the cells were cultured on glass four-well chamber slides (Lab Tek, Nalge Nunc). After the fourth day of culture, the cells were fixed in ice-cold methanol for 20 minutes at −20°C. The methanol was removed, and the cells were rehydrated for 5 minutes with PBS containing 0.1% saponin. After the washing, the cells were blocked with 5% goat serum in PBS-saponin for 30 minutes at room temperature. The cells were immunostained with a mouse monoclonal antihuman desmoplakin antibody (1:50 final dilution; Abcam) or normal mouse serum and incubated overnight at 4°C in a humidified box. The next day the slides were washed three times with PBS-S, and then a secondary antibody, Alexa Fluor 594 donkey antimouse IgG (H+L) (Life Technologies) at a 1:100 dilution, was added and incubated for 2 hours at room temperature. Finally, the wells were washed as above and then ProLong Gold reagent with 4′6-diamidine-2-phenylindole, dihydrochloride (DAPI; Life technologies) was added to visualize the nuclei, and the slide was coverslipped and allowed to cure overnight in the dark at 4°C. No staining was detected with the normal mouse serum. Two microscope slides of the ETS1 siRNA-exposed cells and two slides of the control cells (20 ×10 fields per slide) were examined to determine the relative amounts of mononuclear CTB and multinucleated STB in the two cell groups.

Studies of ETS1 knockdown on gene expression

ETS1 expression in cultured trophoblast cells was knocked down by both siRNA and shRNA methods. The siRNA study used the same ETS1 siRNA and control siRNA used above. At the end of day 4 of culture, RNA was extracted from the cells and analyzed for ETS1 and STB marker gene mRNA levels as described. For the shRNA studies, three lentiviruses were constructed in the Cincinnati Children's Hospital Medical Center Viral Vector Core Laboratory, each expressing a different human ETS1 shRNA (LV-ETS1shRNA-1, -2, and -3). The human ETS1 shRNAs (accession number BC036704; Open Biosystem Inc) were cloned into pWPT-GFP (Addgene Inc). After the sequence and orientation of the ETS1 shRNAs were confirmed, the pWPT-GFP-ETS1 and the empty vector pWPT-GFP were packaged with three helper plasmids, gag/pol, rev, and a VSV-G envelope. The lentivirus particles were titered by transfecting human HT1080 cells in the presence of polybrene for 4 hours and then analyzed at day 3 of culture for expression of green fluorescent protein (GFP). These studies indicated that approximately 90% of the trophoblast cells that were infected with each LV-ETS1shRNA or control GFP-containing lentivirus stained positively for GFP.

In the shRNA knockdown experiments, highly purified trophoblast cells (2.5 × 10⁶ cells) in six-well plates were infected [8.0 × 10⁵ ISilencer Select negative control number 1 siRNA (purchased from Life Technologies) per milliliter medium] separately with each of the LV-ETS1shRNAs or a control lentivirus containing a nonsilencing RNA insert in place of the ETS1 shRNA (LV-control) for 4 days in RPMI 1640 medium with 10% fetal bovine serum. The medium was changed at 3 days and replaced with fresh medium without lentivirus.

The effect of ETS1 overexpression on STB marker genes

Human trophoblast cells were prepared as described above and plated in 12-well culture plates at a density of 1.2 × 10⁶ cells/well. Two hours later, 1 μg of the ETS1 expression plasmid or the empty control plasmid (five wells per treatment) using Lipofectamine 3 (Life Sciences) were performed as described by the manufacturer. The cultures were stopped after 2.5 days, and total RNA was extracted from the cells. The relative amounts of selected STB-specific marker genes (relative to actin mRNA) were then determined by quantitative PCR (qPCR) as described above.

The effects of ETS1 on promoter activity of STB marker genes

Transient transfection studies were performed in freshly prepared primary human trophoblast cells and JEG-3 choriocarcinoma cells using plasmids containing proximal promoter fragments linked to a luciferase reporter gene in pGL3-Basic (Promega Corp). The plasmids included pGL3B-syncytin (−1054/+180), pGL3-hPL (−1078/+2)-Luc (3), pGL3B-AP-2α (−1027/+287)-Luc, pGL3-hGH-v (−158/+157), and pGL3-hCRH (−663/+127). Each promoter contains one or more putative ETS sites [GGA(A/T)]. An expression plasmid for ETS1 (pSG5-ETS1) was kindly provided by Dr Arthur Guiterrez-Hartman (University of Colorado Health Science Center, Denver, Colorado). The transient transfections were performed in triplicate in six-well plates by the liposome method previously described by our laboratory (18) or with Lipofectamine 3000 according to the manufacturer's directions (Life Technologies). The cells (2.5 × 10⁶ cells/well) were harvested 48 hours after transfection in 1× passive lysis buffer (Promega). Luciferase activity in each well was normalized to renilla luciferase activity and the results presented as the mean ± SEM of the normalized luciferase activity. The normalized luciferase activity was then compared in each experiment with a control transfection in which the expression plasmids were cotransfected with the appropriate empty reporter plasmid lacking an inserted promoter fragment.

Statistical analysis

Multiple comparisons were performed by one-way ANOVA with post hoc pair-wise comparisons (Dunnett's test). The statistical difference between the distribution of cell nuclei in CTB and STB of control and ETS1-exposed trophoblast cells was determined by χ² analysis. All values are expressed as the mean ± SEM, and P < .05 was considered statistically significant.

Results

ETS1 expression in human placental villi and cultured trophoblast cells

To determine the specific cell types in term human placenta villi that express ETS1 protein, immunohistochemical analyses were performed using specific antisera to ETS1 and E-cadherin, which distinguishes mononuclear CTB from multinucleated STB. As shown in Figure 1, positive staining for ETS1 protein was detected in CTBs and STBs as well as lymphocytes and stromal cells. The immunostaining for ETS1 protein in CTBs was relatively intense in the nuclei, with much weaker staining in the cytoplasm. In STBs, on the other hand, ETS1 staining was primarily detected in the cytoplasm with little or no immunostaining in the nuclei.

Figure 1.

Localization of ETS1 protein in human placenta. ETS1 protein was detected in paraffin sections of human term placenta using a specific antiserum to ETS1 (red color) as described in Materials and Methods. The CTBs and STBs were distinguished from other cell types using an antiserum to E-cadherin (brown color), and nuclei were counterstained with hematoxylin (blue). Each panel represents villi from two different placentas at ×40 magnification. The blue arrows point to CTBs, and the black arrows point to STBs. Placenta sections treated with normal rabbit serum in place of the ETS1 antiserum showed no immunostaining. Altogether three sections from each placenta (multiple fields per section) were examined.

To examine the pattern of ETS1 expression during villus differentiation, we determined ETS1 protein and mRNA levels in highly purified preparations of human CTBs undergoing spontaneous differentiation to a multinucleated syncytium that expresses abundant amounts of STB marker genes such as PL and CGβ. As shown in Figure 2, the abundance of ETS1 mRNA (lower panel) and ETS1 protein (top panel) decreased as the CTBs began to syncytialize and express STB marker genes. ETS1 mRNA content of trophoblast cells from three different placentas decreased by 37%, 38%, and 60% during the first 3 days of differentiation and decreased even further over the remaining 3 days, when nearly all of the CTBs had fused to form a syncytium (Figure 2, lower panel). ETS1 protein levels in villus trophoblast cells also decreased markedly (>80%-90%) during the differentiation process. Similar changes in ETS1 protein levels were observed in trophoblast cells from two other placentas.

Figure 2.

Relative ETS1 mRNA and protein levels during in vitro differentiation of villus human CTB to a STB phenotype. The mRNA levels for ETS1 were determined by RT-PCR in primary cultures of human trophoblast cells from three different placentas undergoing spontaneous differentiation (bottom panel). ETS1 and actin protein levels during in vitro differentiation from a representative placenta are shown in the top panel.

Effect of ETS1 knockdown on syncytialization

Because ETS1 is known to regulate differentiation of several cell types, we next examined whether ETS1 expression is critical for syncytialization. Freshly isolated CTBs were exposed for 4 days to an ETS1 siRNA or a nonsilencing control siRNA. Preliminary studies indicated that the siRNA down-regulated ET1 mRNA levels by 68%, and the control siRNA had no significant effect (Supplemental Figure 1). Syncytialization was then assessed by immunocytochemical analysis using an antidesmoplakin antiserum and by reaction with DAPI, which stains cell nuclei (Figure 3). Studies from numerous laboratories had demonstrated that microscopic examination of cultured human trophoblast cells that had been stained with antidesmoplakin antiserum and DAPI clearly distinguishes mononuclear CTBs from multinucleated STBs (19, 20). Desmoplakin-DAPI staining of representative fields of ETS1 siRNA-exposed trophoblast cells (Figure 3, right panel) and control cells exposed to a nonsilencing siRNA (Figure 3, left panel) are shown in Figure 3 (top panels). Most of the trophoblast cells exposed to the ETS1 siRNA remained mononuclear, whereas most of the trophoblast cells exposed to the control siRNA formed syncytia. Quantitative assessment of the effects of the ETS1 siRNA on syncytialization (Figure 3, lower panel) indicated that cultures exposed to the ETS1 siRNA contained significantly more mononuclear cells (44.3% vs 33.2%) and cells containing two to four nuclei (33.3% vs 21.8%) than the cultures exposed to the control siRNA. Furthermore, the ETS1 siRNA-exposed cultures contained significantly fewer cells containing more than 10 nuclei/cell than control cultures (5.2% vs. 27.3%). Comparison of the distribution of the nuclei in the control and ETS1 siRNA-treated groups by χ² analysis revealed a value of P < .001. Taken together, these findings indicate that syncytialization is attenuated in ETS1 siRNA-exposed cells, with fewer CTBs forming STBs containing greater than 10 nuclei/cell.

Figure 3.

Silencing of ETS1 with an ETS1 siRNA attenuates syncytialization of human villus CTBs. A freshly prepared, highly enriched fraction of human villus CTB was cultured for 4 days in the presence of a ETS1 shRNA (right panel) or a control nonsilencing RNA (left panel). As shown in the top panel, cell membranes were detected by immunostaining stained with an antidesmoplakin antiserum (red), and cell nuclei were detected by staining with DAPI (blue). The cultures exposed to the ETS1 siRNA contained more mononuclear CTBs and fewer multinucleated STBs than the control cells. Quantitative analysis of the results is shown in the lower panel. The analysis was performed by determining the distribution of the nuclei in mononuclear cells, cells containing two to four nuclei, cells containing 5–10 nuclei, and cells containing more than 10 nuclei. A total of 1000 nuclei were examined in each group of cells. The scale bar encloses 30 μM. A χ² analysis indicated that the distribution of the nuclei in the ETS1 siRNA- and control RNA-treated cells was significantly different (P < .001). Similar results were observed with villous trophoblast cells from a different placenta.

Effect of ETS1 knockdown on gene expression during trophoblast differentiation

Knockdown of ETS1 gene expression in CTBs by ETS1 shRNAs and an ETS1 siRNA (same siRNA as used in above immunohistochemistry study) also significantly attenuated the induction of genes that are normally induced in vivo during villus CTB differentiation. In the experiment depicted in Figure 4, freshly isolated CTBs were cultured for 4 days with lentiviruses that express ETS1 shRNAs (LV-ETS1shRNA-1, LV-ETS1shRNA-2, and LV-ETS1shRNA-3) or a control nonsilencing RNA (LV-RNA control). After day 4 of culture, the trophoblast cells exposed separately to LV-ETS1shRNA-1, LV-ETS1shRNA-2, and LV-ETS1shRNA-3 expressed 52.6%, 85.8%, and 86.9% less ETS1 mRNA, respectively (Figure 4B) and 60%-85% less ETS1 protein (Figure 4A) than the LV-shRNA control. LV-ETS1shRNA-1, which blocked ETS1 mRNA expression by only 52.6%, had a relatively small effect on the expression of the STB marker gene mRNAs except for human PL mRNA. In contrast, LV-ETS1shRNA-2 inhibited each of the marker gene mRNAs by 40%-60%, and LV-ETS1shRNA-3 inhibited the mRNAs by 60%-80%. Inhibition of STB marker genes during differentiation was also observed in two other experiments with LV-ETS1shRNA-3 (data not shown). Down-regulation of PSG1, CRH, and PL was also observed in the trophoblast cells exposed during differentiation to the ETS1 siRNA (Supplemental Figure 1). As shown in Figure 4, villus trophoblast cells exposed to the ETS1 siRNA for 4 days expressed 20%-60% less mRNAs for PSG1, CRH, and PL than control cells exposed to a nonsilencing RNA.

Figure 4.

Silencing of ETS1 with ETS1 shRNAs attenuate the induction of STB marker genes. Freshly isolated villus CTBs were exposed for 4 days to one of three different shRNAs linked to a lentivirus (LV-ETS1 shRNA-1, -2, and -3) or a nonsilencing control RNA linked to a lentivirus (LV-ns shRNA). At the end of day 4, total protein (two wells) and RNA (three wells) was extracted from each culture group, and the relative amounts of ETS1 protein was determined by Western blot analysis and the mRNAs for selected STB marker genes were determined by qPCR. The Western blot analyses are shown in the upper panel (A), and the qPCR results for ETS1 and marker gene mRNAs are shown in the lower panel (B). Each bar represents the mean + SEM of three wells. **, P < .01, ***; P < .001; ****, P < .0005. Similar results were observed with LV-ETS1 shRNA-3 in two other placentas.

Effect of ETS1 overexpression on CTB differentiation

Overexpression of ETS1 significantly induced the STB marker genes. As shown in Figure 5, CTB cultures (1.2 × 10⁶ cells/well) exposed to the plasmid pSG5-ETS1 (1 μg) expressed 5.6-fold more ETS1 mRNA and 3.9- to 5.5-fold more of the mRNAs for PSG1, PL, CRH, GH variant, syncytin, and AP-2α than CTB exposed to the empty plasmid (pSG5) alone (P < .01 for each gene).

Figure 5.

Overexpression of ETS1 induces STB marker protein expression. Freshly isolated trophoblast cells were exposed for 2.5 days to the expression plasmid pSG5-ETS1 or the control plasmid pSG5. Total RNA was extracted from the cells, and mRNA levels determined by qPCR. Each bar represents the mean of five wells, and the brackets enclose the SEM. **, P < .01; ****, P < .0005. Nearly identical results were observed in another experiment in which cells were exposed to the expression plasmid for 4 days.

The effects of ETS1 overexpression on the promoter activities of STB marker genes

To examine whether the induction of the STB-specific genes is due, at least in part, to transactivation of the gene promoters, transient transfections were performed in primary trophoblast cells and JEG-3 cells cotransfected with pSG5-ETS1 and plasmids containing fragments of the different STB-specific gene promoters. As shown in Figure 6, overexpression of ETS1 in both cell types transactivated the promoters of, PL, CRH, GH variant, syncytin, and AP-2α, whereas the empty plasmid (pSG5) had no effect on each promoter activity. The magnitude of promoter activation by pSG5-ETS1 (2.0 μg) ranged from 2.1- to 13.0-fold (P < .001 in each instance). Transactivation of each marker gene promoter was also observed in the JEG3 cells (data not shown).

Figure 6.

ETS1 transactivates the human PL, CRH, human GH variant, syncytin, and AP-2α promoters. Primary human trophoblast cells were transiently transfected with plasmids containing the proximal promoter regions of human PL (pGL3B-hPL-Luc), human CRH (pGL3B-hCRH-Luc), human GH variant (GHv; pGL3-hGH-V-Luc), syncytin (pGL3-syn-Luc), or AP-2α (pGL3-AP-2α-Luc) (2 μg each) along with the expression plasmid for ETS1 (pSG5-ETS1; 0.3–2.0 μg). The results were compared with transfection with the ETS1 empty vector, pSG5. The luciferase activity in each sample was normalized to Renilla luciferase activity in the same sample. The results are expressed as the mean ± SEM of triplicate culture wells. Nearly identical results were observed in two other experiments. *, P ≦ .05, ***, P ≦ .001, P vs control.

Discussion

The results of this study indicate that ETS1 protein is abundantly expressed in CTB nuclei of term human placental villi but only weakly expressed in STB nuclei. Both ETS1 protein and mRNA levels are also abundantly expressed in freshly isolated CTBs. The abundance of ETS1 protein and mRNA, however, significantly decreases as the cells undergo spontaneous syncytialization and express STB marker genes. After 6 days in culture, when most of the cells have formed a syncytium, the levels of ETS1 protein and mRNA are less than 20% those of freshly isolated CTBs (time 0).

Earlier studies have shown that ETS1 is also expressed in first-trimester placentas, in which it is strongly expressed in the cytoplasm (8, 21). ETS1 mRNA is transcribed in both the endothelial cells of villous trophoblast and extravillous trophoblastic cells invading uterine vessels, with the strongest expression observed in cell columns that invade the endometrium. ETS1 is detected in both the cytoplasm and nucleus of the invading trophoblast in the basal plate and in the amniotic membrane. Because ETS1 is expressed during the invasive process of the endometrium and maternal vessels by trophoblastic cells, Luton et al (21) suggest a possible role for ETS1 in the regulation of metalloproteinase gene transcription, a gene family known to be a target for ETS protein. Michaelis et al (22) observed that ETS1 is also expressed in human fetal membranes. The expression is up-regulated with premature rupture of the membranes, suggesting a role for ETS1 in extracellular matrix remodeling of the membranes.

Knockdown and overexpression studies indicate that ETS1 regulates the induction of STB marker genes during human villous trophoblast differentiation. Knockdown of ETS1 levels in term trophoblast cells was performed by lentiviruses that express ETS1 shRNAs and by a siRNA, each directed to a different region of the 3′ end of the ETS1 gene. Both the shRNAs and the siRNA significantly attenuated gene expression as compared with nonsilencing controls. The knockdown of ETS1 in each experiment was confirmed by analysis of ETS1 protein and mRNA levels before and after exposure to the inhibitory agents. Compared with controls, two of the three ETS1 shRNAs (LV-ETS1 shRNA-2 and -3) and the ETS1 siRNA attenuated ETS1 levels by greater than 65% and blocked the induction of STB marker genes by 40%-80%. LV-ETS1 shRNA-1, which attenuated ETS1 expression by only 40%, had only a small effect on gene expression. The fact that the shRNAs and the siRNA each blocked ETS1 expression and the expression of STB-specific target genes strongly indicate the attenuation of the marker genes is not the result of a nonspecific effect. In the overexpression studies, the overexpression of ETS1 in the primary trophoblast cells induced the mRNAs for STB marker genes by 3.6- to 7.0-fold. Taken together, these knockdown experiments strongly suggest that ETS1 is critical for maximal expression of genes normally induced during trophoblast differentiation.

The desmoplakin immunocytochemistry studies also indicate a role for ETS1 in the syncytialization of CTB. At the end of the fourth day of differentiation, most of the control CTB cells had aggregated and fused to form a syncytium, with 27.3% of the nuclei in cells containing more than 10 nuclei. In contrast, only 5.2% of the nuclei of the cells exposed to the ETS1 siRNA were present in STB with more than 10 nuclei/cell. Because numerous studies indicate that the transmembrane glycoprotein syncytin is critical for syncytialization of CTBs (5, 6, 23, 24), the attenuation of syncytialization in response to the knockdown of ETS1 appears to be due, at least in part, to the partial knockdown (∼50%) of syncytin expression. Because ETS1 knockdown does not completely attenuate the induction of syncytin expression, the reduced amount of syncytin not affected by the knockdown of ET1 appears to be sufficient for CTB to form STB with fewer than 10 nuclei/cell but not with more than 10 nuclei/cell.

The effects of ETS1 on syncytin and the other genes are due, at least in part, to transactivation of the promoters for these genes. Because silencing of ETS1 expression attenuates the early induction of syncytialization and syncytin mRNA levels, it appears that ETS1 plays a critically important role in the regulation of both stages of villus CTB cell differentiation. The effect of ETS1 on PL, CGβ, and other STB-specific genes may be due to a direct effect of ETS1 on the promoters of these genes as well as to an indirect effect mediated by activation of AP-2α.

Recent studies from our laboratory demonstrated by immunohistochemistry that AP-2a protein expression in villous STB is decreased in mild preeclampsia, diabetes mellitus, hypertension, and fetal growth restriction compared with gestational age-matched placentas (P < .0001 in each instance) (25). Because AP-2α has been shown to be critical for the differentiation of villous CTB to a STB phenotype, these findings suggest that abnormalities in the AP-2α cascade of transcription factors and/or signaling molecules may contribute to the pathogenesis of the abnormal maturation in placentas in certain types of high-risk pregnancies. Because the knockdown of ETS1 blocks the expression of AP-2a, it is possible that the decreased AP-2a expression in pathological pregnancies may result, at least in part, from a decrease in ETS1 expression.