p63 Inhibits Extravillous Trophoblast Migration and Maintains Cells in a Cytotrophoblast Stem Cell-Like State

Abstract

Proper differentiation of placental epithelial cells, called trophoblast, is required for implantation. Early during placentation, trophoblast cell columns help anchor the developing embryo in the uterine wall. Although proximally continuous with villous cytotrophoblast (CTB) distally, these cells differentiate into invasive extravillous trophoblast. We previously reported that p63, a p53 family member, is highly expressed in proliferative villous CTB and required for induction of the trophoblast lineage in human pluripotent stem cells. We now further explore its function in human trophoblast by using both primary CTB from the early placenta and established trophoblast cell lines. We show that p63 is expressed in epidermal growth factor receptor-positive CTB and that its expression decreases with differentiation into HLA-G⁺ extravillous trophoblast. In trophoblast cell lines, p63 is expressed in JEG3 cells but absent from HTR8 cells. Overexpression of p63 in both cell lines enhances cell proliferation and significantly reduces cell migration; conversely, down-regulation of p63 in JEG3 cells reduces cell proliferation and restores cell migration. Analysis of epithelial-to-mesenchymal transition, cell adhesion, and matrix degradation pathways shows that p63 blocks epithelial-to-mesenchymal transition, promotes a CTB-specific cell adhesion profile, and inhibits expression of matrix metalloproteinases. Taken together, these data show that p63 maintains the proliferative CTB state, at least partially through regulation of epithelial-to-mesenchymal transition, cell adhesion, and matrix degradation pathways.

CME Accreditation Statement: This activity (“ASIP 2014 AJP CME Program in Pathogenesis”) has been planned and implemented in accordance with the Essential Areas and policies of the Accreditation Council for Continuing Medical Education (ACCME) through the joint sponsorship of the American Society for Clinical Pathology (ASCP) and the American Society for Investigative Pathology (ASIP). ASCP is accredited by the ACCME to provide continuing medical education for physicians.

The ASCP designates this journal-based CME activity (“ASIP 2014 AJP CME Program in Pathogenesis”) for a maximum of 48 AMA PRA Category 1 Credit(s)™. Physicians should only claim credit commensurate with the extent of their participation in the activity.

CME Disclosures: The authors of this article and the planning committee members and staff have no relevant financial relationships with commercial interests to disclose.

Proper fetal development in utero requires an appropriately functioning placenta, able to deliver oxygen and nutrients to the growing fetus. Functional epithelial cells of the placenta are called trophoblast, which develop from the trophectoderm of the blastocyst.^1,2 Early during human placental development, proliferative cytotrophoblast (CTB) forms cell columns, which help anchor the placenta to the uterus; in the distal parts of such anchoring villi, the CTBs differentiate into a migratory phenotype, the extravillous trophoblast (EVT).¹

EVT invades the uterus, establishing blood flow to the fetoplacental unit through remodeling of maternal spiral arterioles.¹ The CTB-EVT differentiation is characterized by an integrin switch, from α₆β₄ in the villous CTB to α₅β₁ in the cell columns and α₁β₁ in the uterine wall.³ Changes in integrin expression are accompanied by changes in cell adhesion properties and an increase in autophosphorylation of focal adhesion kinase (FAK) in EVT.⁴

Mature EVTs are characterized by loss of expression of epidermal growth factor receptor (EGFR) on their surface, gaining instead surface expression of HLA-G and melanoma cell adhesion molecule.^5–7 EVT differentiation also resembles, in part, the process of epithelial-to-mesenchymal transition (EMT), with reduction of E-cadherin expression in the cell columns.⁸ Finally, EVT differentiation was also compared with cancer cell invasion, because these cells acquire the ability to secrete matrix metalloproteinases.⁹

Despite the above knowledge about markers of CTB and EVT, little is known about transcription factors that regulate each phenotype in the human placenta. We have previously described expression of p63, a nuclear protein and transcriptional regulator in the p53 family, in the human placenta and have shown that it is expressed only in proliferative CTB and completely excluded from both syncytiotrophoblast and EVT.⁷ The α isoform of the N-terminally truncated p63 (ΔNp63α) has been shown to be involved in maintaining the stem cell state in stratified epithelia, including skin.^10,11 p63 is also known to regulate cell adhesion in mammary epithelium, promoting adhesion-dependent protection against cell death.¹² In addition, both in bladder and prostate cancer cell models, loss of ΔNp63α has been shown to promote EMT, leading to greater invasive potential.^13,14

We have recently determined that bone morphogenetic protein-4–induced trophoblast differentiation of human pluripotent stem cells occurs through a p63⁺/KRT7⁺ intermediate, likely representing a CTB stem cell state.¹⁵ We also observed that forced expression of p63, specifically ΔNp63α, in cultured term CTB maintained cyclin B expression and inhibited human chorionic gonadotropin (hCG) secretion. Here, we dissect the role of p63 in more detail during the CTB-to-EVT transition, evaluating changes in marker expression and adhesive and migratory functions, using both first-trimester CTB and representative human trophoblast cell lines.

Materials and Methods

Human Placental Samples and Isolation of Primary Trophoblasts

Human placental tissues were collected under a protocol approved by the Human Research Protections Program Committee of the University of California San Diego Institutional Review Board; all patients gave informed consent for collection and use of these tissues. First-trimester trophoblasts were isolated from placentas of 6 to 12 weeks’ gestation. Chorionic villi were minced, washed in phosphate-buffered saline (PBS), and subjected to three sequential digestions, digestion I: 300 U/mL DNase I (Sigma-Aldrich, St. Louis, MO) and 0.125% trypsin (Gibco, Carlsbad, CA); digestions II and III: 0.25% trypsin (Gibco) and 300 U/mL DNase I. The pelleted cells from the second and third digests were pooled, resuspended in Hanks’ balanced salt solution, and separated on a Percoll gradient. Cells were plated on fibronectin-coated plates in Dulbecco’s modified Eagle’s medium/F12 with 10% fetal bovine serum (FBS), 1× penicillin/streptomycin, and 50 μg/mL gentamicin.

Cell Lines and Culture Conditions

The human choriocarcinoma cell line JEG3 was obtained from the ATCC (Manassas, VA). The human immortalized EVT cell line HTR8 was a kind gift from Dr. Charles Graham (Queen's University, Kingston, ON, Canada). JEG3 cells were grown in Dulbecco's modified Eagle's medium (Cellgro, MediaTech Inc., Manassas, VA), supplemented with 10% FBS (Sigma-Aldrich) and penicillin-streptomycin (Life Technologies, Carlsbad, CA). HTR8 cells were grown in RPMI 1640 medium (Cellgro) also supplemented with 10% FBS and penicillin-streptomycin (Gibco).

Flow Cytometric Analysis and FACS

For flow cytometric analysis, cells were fixed and permeabilized with 100% ice-cold methanol for 30 minutes, then incubated with primary antibodies on ice for 1 hour, followed by Alexa Fluor 488-conjugated secondary antibodies (Invitrogen, Carlsbad, CA) on ice for 1 hour. Cells were washed and resuspended in 1 mL of fluorescence-activated cell sorting (FACS) buffer (0.5% bovine serum albumin and 1% FBS in PBS). Flow cytometric analysis was performed with a BD FACS-Canto Flow Cytometer (BD Biosciences, San Jose, CA). Primary antibodies used for flow cytometry included rat anti–E cadherin (Abcam, Cambridge, MA), rat anti–integrin-β4 (Novus Biologicals, Littleton, CO), rabbit anti-EGFR (Santa Cruz Biotechnology, Santa Cruz, CA). Alexa Fluor 488-conjugated rat or rabbit IgG (Invitrogen) was used as isotype control.

For FACS of CTB, after isolation, the cells were washed three times in PBS supplemented with 10% FBS. Cells were then incubated in PBS plus serum at room temperature for 1 hour with an allophycocyanin anti–human EGFR antibody (clone AY13; BioLegend, San Diego, CA) and a phosphatidylethanolamine anti–human HLA-G antibody (clone 87G; BioLegend) followed by FACS with the use of a BD Influx cell sorter (BD Biosciences). The sorted cells were collected, resuspended in RNA lysis buffer (mirVana miRNA Isolation Kit; Ambion, Austin, TX), and stored at −80°C until processed for RNA extraction.

Immunostaining of Cells and Tissues

For immunofluorescence staining, cells grown on coated coverslips were fixed with 4% paraformaldehyde in PBS at room temperature for 10 minutes. Cells were then permeabilized with 0.3% Triton X-100 for 10 minutes and incubated with primary antibodies, including rabbit anti–phospho-FAK (Y397; BD Biosciences) and mouse anti–β-catenin (BD Biosciences), and visualized by Alexa Fluor 488- or Alexa Fluor 595-conjugated secondary antibodies (Invitrogen). For nuclear staining, cells were incubated with DAPI (Invitrogen) for 5 minutes.

For immunohistochemistry on formalin-fixed, paraffin-embedded human placenta tissue, serial sections were stained with mouse anti-p63 (4A4 clone; Sigma-Aldrich), rabbit anti-Ki67 (Abcam), mouse anti-CK7 (Abcam), or mouse anti–HLA-G (4H84 clone; Abcam), using a Ventana Discovery Ultra automated immunostainer (Ventana Medical Systems, Tucson, AZ) with standard antigen retrieval and reagents per manufacturer's protocol.

Cell Proliferation Assay

Cell proliferation was measured with the bromodeoxyuridine/7aminoactinomycin D (BrdU/7-AAD) Flow Kit (BD Biosciences). Cells were labeled with 10 μmol/L BrdU for 30 minutes at 37°C, then trypsinized and fixed/permeabilized with BD Cytofix/Cytoperm buffer. Cells were treated with DNase for 1 hour at 37°C and stained with fluorescein isothiocyanate–conjugated anti-BrdU antibody. Total DNA was determined by 7-AAD staining and analyzed by flow cytometry.

Quantitative Real-Time PCR

Total RNA was extracted with the mirVana RNA Isolation Kit (Ambion) according to the manufacturer's protocol. Purity and concentration of the samples were assessed with NanoDrop ND-1000 Spectrophotometer (Thermo Scientific Inc., Waltham, MA). cDNA was prepared from 1 μg of RNA by using iScript (Bio-Rad, Hercules, CA) in a 20-μL reaction and diluted 1:5 with nuclease-free water. Quantitative real-time PCR (qPCR) was performed with 4 μL of the diluted cDNA, along with 625 nmol/L of each primer and POWER SYBR Green PCR master mix (Applied Biosystems, Foster City, CA) in a total reaction volume of 20 μL. qPCR was performed with a System 7300 instrument (Applied Biosystems) and a one-step program: 95°C, 10 minutes; 95°C, 30 seconds; 60°C, 1 minute; for 40 cycles. Unless otherwise stated, each experiment was performed in triplicate, and all results were normalized against 18S rRNA. Relative mRNA expression levels, compared with 18S rRNA, were determined by the ΔΔC_T method. All primer pairs (Table 1) were checked for specificity by using BLAST analysis and were checked by both agarose gel electrophoresis and thermal dissociation curves to ensure amplification of a single product with the appropriate size and melting temperature.

Table 1.

List of Quantitative Real-Time PCR Primers

Primer name	Primer sequence
TAp63
Forward	5′-AAACAAGATTGAGATTAGCATGGA-3′
Reverse	5′-AGAGAGCATCGAAGGTGGAG-3′
ΔNp63
Forward	5′-CTGGAAAACAATGCCCAGA-3′
Reverse	5′-AGAGAGCATCGAAGGTGGAG-3′
EGFR
Forward	5′-CTAAGATCCCGTCCATCGCC-3′
Reverse	5′-GGAGCCCAGCACTTTGATCT-3′
CDH1
Forward	5′-CGAGAGCTACACGTTCACGG-3′
Reverse	5′-GGGTGTCGAGGGAAAAATAGG-3′
SNAI1
Forward	5′-TCGGAAGCCTAACTACAGCGA-3′
Reverse	5′-AGATGAGCATTGGCAGCGAG-3′
CSH1/2
Forward	5′-AATCCAATCTAGAGCTGCTCCGCA-3′
Reverse	5′-TGTCATACACCAGGTTGTTGGCGA-3′
ERVW-1
Forward	5′-GTCACTGTCTGTTGGACTTACT-3′
Reverse	5′-CGGCTGAGTTGGGAGATTAC-3′
CGB
Forward	5′-TGAGATCACTCACCGTGGTCTCC-3′
Reverse	5′-TTTATACCTCGGGGTTGTGGGG-3′
ITGA6
Forward	5′-GGCGGTGTTATGTCCTGAGTC-3′
Reverse	5′-AATCGCCCATCACAAAAGCTC-3′
ITGA5
Forward	5′-GGCTTCAACTTAGACGCGGAG-3′
Reverse	5′-TGGCTGGTATTAGCCTTGGGT-3′
ITGB1
Forward	5′-CCTACTTCTGCACGATGTGATG-3′
Reverse	5′-CCTTTGCTACGGTTGGTTACATT-3′
ITGB4
Forward	5′-CTCCACCGAGTCAGCCTTC-3′
Reverse	5′-CGGGTAGTCCTGTGTCCTGTA-3′
MMP2
Forward	5′-CCAAGGAGAGCTGCAACCTG-3′
Reverse	5′-TGGGCTTGCGAGGGAAGAAG-3′
MMP3
Forward	5′-CGGTTCCGCCTGTCTCAAG-3′
Reverse	5′-CGCCAAAAGTGCCTGTCTT-3′
MMP9
Forward	5′-AGACCTGGGCAGATTCCAAAC-3′
Reverse	5′-CGGCAAGTCTTCCGAGTAGT-3′
MMP12
Forward	5′-GGAATCCTAGCCCATGCTTTT-3′
Reverse	5′-CATTACGGCCTTTGGATCACT-3′
18S
Forward	5′-CGCCGCTAGAGGTGAAATTCT-3′
Reverse	5′-CGAACCTCCGACTTTCGTTCT-3′

Standard RT-PCR for various TP63 isoforms was performed with the indicated primers (Table 2).

Table 2.

List of RT-PCR Primers Isoform-Specific Primers

Gene abbreviation	Primer sequence
α Isoform
Forward	5′-GAGGTTGGGCTGTTCATCAT-3′
Reverse	5′-AGGAGATGAGAAGGGGAGGAGA-3′
β Isoform
Forward	5′-GAGGTTGGGCTGTTCATCAT-3′
Reverse	5′-TCAGACTTGCCAGATCCTGACA-3′
γ Isoform
Forward	5′-GAGGTTGGGCTGTTCATCAT-3′
Reverse	5′-GCTCCACAAGCTCATTCCTGAA-3′
GAPDH
Forward	5′-GAAGGTGAAGGTCGGAGTC-3′
Reverse	5′-GAAGATGGTGATGGGATTTC-3′

p63 Lentiviral Constructs and Generation of Stable Cell Lines

Full-length ΔNp63α cDNA was PCR-amplified and cloned into pCMV–lentiviral-based vector as previously described.¹⁵ pLKO.1-based scrambled and p63-specific shRNA lentiviral constructs were obtained and used as previously described.¹⁵ Lentiviral supernatants were concentrated with PEG-it virus precipitation solution (System Biosciences, Mountain View, CA). The concentrated viral particles were then incubated with target cells in the presence of 5 μg/mL polybrene (Sigma-Aldrich). After 3 days of infection, cells were treated with 5 μg/mL puromycin (Sigma-Aldrich) to select stably transduced cells. Overexpression and knockdown efficiency were determined by Western blot analysis.

Cell Migration Assay

Transwell membranes (8.0-μm pore inserts in 24-well BioCoat Chambers; BD Biosciences) were used to study cell migration in vitro. Cells were trypsinized, washed in cold PBS, and resuspended in serum-free media with 5 × 10⁴ cells per well in the upper chamber and medium with 10% fetal bovine serum in the lower chamber. After 16 to 24 hours of incubation, cells on the upper surface were removed by scrubbing with a cotton swab; filters were then stained with Diff-Quik Stain Set (Dade Behring, Newark, DE). Three independent ×20 fields for each well were used for quantitation.

Cell Adhesion Assay

CyQUANT cell proliferation assay kit (Invitrogen) was used to study cell adhesion. Two different densities of cells were plated in a 96-well black plate (Greiner Bio-One, Longwood, FL). The cell suspensions were incubated at 37°C for 1 hour to allow the cells to attach. The wells were then washed three times gently with prewarmed PBS to remove nonadherent cells. The fluorescence of the samples was measured by using a microplate reader at 480 nm/520 nm, according to the manufacturer's protocol.

hCG Hormone Secretion Assay

Cell culture supernatants were collected and stored at −80°C until use. Levels of total hCG were quantified with the hCG ELISA Kit (Calbiotech, Spring Valley, CA) according to the manufacturer's protocol. The results were normalized to DNA content, quantified by DNeasy (Qiagen, Valencia, CA).

Western Blot Analysis

After washing with ice-cold PBS, cells were lyzed with 2× SDS-PAGE sample buffer (20 mmol/L Tris, pH 8.0, 2% SDS, 2 mmol/L dithiothreitol, 1 mmol/L Na₃VO₄, 2 mmol/L ethylenediamine tetraacetic acid, and 20% glycerol) and boiled for 10 minutes. Protein concentration of each sample was determined with a bicinchoninic acid protein assay reagent (Thermo Scientific Inc.). Total cellular protein (30 μg) was separated by 10% SDS-PAGE and then transferred to polyvinylidene difluoride membranes. The membranes were blocked for 1 hour at room temperature in 20 mmol/L Tris, pH 8.0, 150 mmol/L NaCl, and 0.05% Tween 20 (TBS-T) that contained 5% nonfat dried milk. The membranes were then incubated with the primary antibody overnight at 4°C. Primary antibodies included mouse anti–β-catenin (BD Biosciences), mouse anti–β-actin (Sigma-Aldrich), mouse anti-p63 (4A4 clone; Sigma-Aldrich), mouse anti–phospho-FAK (Y397; BD Biosciences), and mouse anti-total FAK (BD Biosciences). After three washes in TBS-T, the membranes were incubated with the appropriate horseradish peroxidase–conjugated secondary antibodies (Jackson ImmunoResearch Laboratories Inc., West Grove, PA) for 1 hour at room temperature and then washed with TBS-T three times. The labeled proteins were visualized with the enhanced chemiluminescence kit (Thermo Scientific Inc.).

Statistical Analysis

Unless otherwise stated, data presented are means ± SD of technical triplicates; the results shown are representative of three to five separate experiments, each performed with a different preparation of cells. Student's t-test was performed, and P < 0.05 was considered statistically significant.

Results

Localization of p63 in the Early Placenta Tissue

Previously, we have shown that, in placentas across gestation, p63 is expressed in all Ki-67⁺ villous CTBs but is not expressed in villous syncytiotrophoblast and invasive EVT.⁷ To expand on this finding, we first performed immunohistochemistry on the earliest human placental tissue we had, a blastocyst 16 days after implantation serendipitously discovered in an archived hysterectomy specimen in which the indication for the procedure was uterine prolapse (Figure 1A). We confirmed that the majority of p63 staining was localized to the villous CTB layer, immediately adjacent to the villous stroma. p63 was absent from the syncytiotrophoblast. In addition, although rare proximal cell column trophoblasts were p63⁺, the distal column and invasive EVT at the implantation site were p63⁻. All p63⁺ cells were also Ki-67 positive, although the Ki-67⁺ cells located in the distal cell column were in fact negative for p63. HLA-G stained a subset of EVT in the distal cell column and invasive EVT at the implantation site. Finally, p63 staining overlapped with p57 only in CTB (rare ones in the embryo and almost all CTBs in later first-trimester tissue), whereas the p57⁺ cell column and mature EVT were negative for p63 (Figure 1B).

Figure 1.

Marker expression in early placental tissue and isolated primary cytotrophoblast (CTB). A: Human placenta at 4 weeks’ gestational age shows p63 staining localized exclusively to the villous CTB layer and HLA-G staining in the distal cell column and invasive extravillous trophoblast (EVT) at the implantation site. p57 staining is seen in rare villous CTB, villous mesenchymal cells, and distal cell column and invasive EVT. Note that all p63⁺ cells are also Ki-67⁺; Ki-67⁺ trophoblast extends well into the cell column. B: p63 and p57 overlap significantly more in the CTB of a human placenta at later gestation (8 weeks). C: Real-time quantitative PCR of CTB- and EVT-specific genes in primary CTB and EVT cells isolated from placentas at 6 to 8 weeks’ gestation. Data are normalized to 18S. Data are expressed as fold change relative to CTB; error bars indicate SD of technical triplicates. ^∗P < 0.05 in comparison with CTB values.

To confirm these in vivo data, we isolated CTB from placentas at 6 to 8 weeks of gestation and fractionated them into villous CTB and EVT by FACS, based on surface expression of EGFR and HLA-G, respectively. RNA was isolated from both cell fractions and qPCR was performed (Figure 1C). The results suggested the EGFR⁺ CTB fraction was highly enriched in p63, EGFR, CDH1, ITGA6, and ITGB4, whereas the EVT fraction was highly enriched in HLA-G, SNAI1, ITGA5, and multiple matrix metalloproteinases (MMP)-2, -3, -9, and -12.

It has been shown that p63 has six isoforms that are expressed in a tissue-specific manner.¹⁶ On the basis of our previous study, ΔNp63α was the most abundant isoform in CTBs isolated from term human placental tissues.¹⁵ To determine which p63 isoform(s) is (are) expressed in first-trimester CTB, we first performed qPCR with TAp63- and ΔNp63-specific primers and found ΔNp63 to be the major isoform. With the use of C-terminal–specific primers in a standard real-time PCR, we found the α isoform to be the main isoform, with only a minute amount of the β and γ isoforms also detected (Supplemental Figure S1).

p63 Promotes Trophoblast Cell Proliferation

Primary isolated CTB quickly lose their proliferative potential in vitro. Therefore, to assess the role of p63 in human trophoblast proliferation, we turned to available cell lines for modeling EVT differentiation: JEG3 and HTR8 cells. Of these, only JEG3 cells expressed p63. We checked p63 isoform expression in JEG3 cells and confirmed that, similar to first-trimester CTB, JEG3 cells mainly express ΔNp63α (Supplemental Figure S1).

We generated a lentiviral expression construct with ΔNp63α downstream of the cytomegalovirus promoter. First, we infected the HTR8 cell line, an immortalized EVT cell line, with undetectable levels of endogenous p63. We confirmed p63 overexpression by Western blot analysis; comparison with primary CTB found higher p63 protein levels in the HTR8 cells that overexpressed p63 (Figure 2A). Then we checked cell proliferation with BrdU/7-AAD double staining by flow cytometry. p63-overexpressing HTR8 cells indicated a higher percentage of cells in S-phase compared with control mock-infected cells (Figure 2B).

Figure 2.

p63 promotes proliferation in the HTR8 EVT cell line. A: Confirmation of p63 overexpression by Western blot analysis, compared with endogenous levels in two different preparations of first-trimester CTB. β-actin served as loading control. B: Cell proliferation assay with BrdU/7-AAD double staining, followed by flow cytometric analysis. p63-overexpressing cells show a higher percentage of cells in S-phase (P2 region) compared with cells transduced with an empty lentivirus. Data are expressed as means ± SD of triplicate cell samples (B). n = 3 independent experiments (B). ^∗P < 0.05 in comparison to mock cells.

Because HTR8 cells did not have endogenous p63, we also used JEG3 cells, first to confirm the effects of p63 overexpression, and then proceeded to p63 knockdown experiments. We used the same construct as above to overexpress ΔNp63α and previously generated shRNA-lentiviral cocktail to knockdown its expression in JEG3 cells.¹⁵ We confirmed overexpression and knockdown by Western blot analysis (Figure 3A). Similar to HTR8 cells, JEG3 cells overexpressing p63 indicated an increase in proportion of cells in S-phase, compared with control mock-infected cells. Conversely, this population was reduced in p63-knockdown cells, indicating a clear role for p63 in regulation of trophoblast proliferation (Figure 3B).

Figure 3.

p63 promotes proliferation of JEG3 choriocarcinoma cells. A: Confirmation of p63 overexpression and knockdown by Western blot analysis. β-Actin served as loading control. B: Cell proliferation assay with BrdU/7-AAD double staining by flow cytometric analysis. p63-overexpressing cells show a higher percentage of cells in S-phase and lower percentage of cells in G₀/G₁ phase compared with control cells. Conversely, p63-knockdown cells show a lower percentage of cells in S-phase and higher percentage of cells in G₀/G₁ phase compared with scramble control cells. Data are expressed as means ± SD of triplicate cell samples (B). ^∗P < 0.05 in comparison with mock-infected or scramble shRNA-infected values.

p63 Maintains the CTB Phenotype of Both Human Trophoblast Cell Lines and Primary CTB

Next, we performed functional analysis of trophoblast differentiation in comparison with changes in gene expression after manipulation of p63 expression. We assessed cell migration, a characteristic feature of EVT, in both JEG3 and HTR8 cell lines; we also assessed hormone (hCG) secretion, a characteristic feature of syncytiotrophoblast, which also occurred in JEG3 cells. p63-overexpressing HTR8 cells exhibited reduced cell migration compared with control cells in the Transwell migration assay (Figure 4A). These cells also up-regulated expression of EGFR and ITGB4 and down-regulated expression of ITGB1, SNAI1, MMP-2, and MMP-3 (Figure 4B). Although CDH1 mRNA remained unchanged (data not shown), expression of EGFR, CDH1, and ITGB4 increased according to flow cytometric analysis in p63-overexpressing HTR8 cells (Figure 4C). Immunostaining confirmed EGFR expression to be enhanced, particularly in p63^high HTR8 cells (Figure 4D).

Figure 4.

p63 overexpression shifts the EVT cell line, HTR8, toward a CTB phenotype. A: Transwell migration assay of p63-overexpressing HTR8 and mock control (empty lentivirus-transduced) cells. B: Real-time quantitative PCR for p63, EGFR, ITGB1, ITGB4, SNAI1, and MMP-2, -3, and -9 in p63-overexpressing HTR8 cells. Data are normalized to 18S. C: Surface expression of EGFR, CDH1, and ITGB4 in mock (red) and p63-overexpressing (blue) cells by flow cytometric analysis. D: HTR8 cells, infected with either empty or p63-expressing lentivirus, are immunostained for p63 (red) and EGFR (green), with numerous cells showing costaining. Data are expressed as means ± SD of triplicate filters (A); fold change relative to the mock cells of technical triplicates (B). n = 3 independent ×20 fields for each well (A). ^∗P < 0.05 in comparison to mock cells.

In JEG3 cells, overexpressing p63 also inhibited cell migration; however, cells with p63 knockdown did not show a further elevation in migration (Figure 5A). In terms of hormone production, JEG3 cells overexpressing p63 had decreased hCG secretion, whereas cells with p63 knockdown indicated a small, albeit statistically significant, increase in hormone secretion compared with the scrambled control (Figure 5B). Gene expression changes in p63-overexpressing JEG3 cells were consistent with inhibition of terminal differentiation, showing a reduction in SNAI1, ERVW-1 (syncytin), CGB, and CSH1 (Figure 5C). Gene expression changes in p63-knockdown JEG3 cells were more modest, showing statistically significant increases only in CGB and CSH1 (Figure 5C).

Figure 5.

p63 maintains the CTB phenotype of human trophoblast cell line, JEG3. A: Transwell migration assay of mock, p63-overexpressing, scramble control, and p63-shRNA JEG3 cells. B: hCG ELISA of mock, p63-overexpressing, scramble control, and p63 shRNA JEG3 cells; hCG secretion was normalized to DNA content. C: Gene expression changes in p63-overexpressing and -knockdown JEG3 cells by real-time quantitative PCR. Data are expressed as means ± SD of triplicate filters (A); of technical triplicates (B and C). n = 3 ×20 fields for each well (A). ^∗P < 0.05 in comparison with mock-infected or scramble shRNA-infected values.

Finally, in primary first-trimester CTB, we tested the effect of p63 overexpression, because these cells rapidly lost p63 after plating. Infection with a green fluorescent protein–expressing lentivirus confirmed maximal gene expression at day 4 after infection (Figure 6A). p63-overexpressing CTB did not show altered expression of EGFR or HLA-G by flow cytometry or hCG secretion by ELISA (data not shown). However, qPCR found significantly decreased expression of EVT markers, including MMP-2, -3, and -12 and ITGA1, ITGA5, and ITGB1 (Figure 6B) at day 6 after infection.

Figure 6.

p63 overexpression in primary first-trimester CTB. A: Infection of primary CTB with a green fluorescent protein–expressing lentivirus; note that transgene expression is sparse before day 3 and decreases after day 4. B: Real-time quantitative PCR for p63, MMP-2, -3, -12, integrin beta (ITGA)1, ITGA5, and ITGB1 in p63-overexpressing cells. Data are normalized to 18S. Data are expressed as fold change, relative to the mock cells on day 4, as means ± SD of technical triplicates (B). ^∗P < 0.05 in comparison to mock cells of the same day.

p63 Promotes a CTB-Like Cell Adhesion Profile

Because CTB differentiated into EVT during early placental development, they decrease cell-cell adhesion and instead develop stronger cell-matrix adhesive properties, with greater focal adhesion turnover required for cell migration. To determine whether the effect of p63 on trophoblast migration may be mediated through changes in adhesive properties, we evaluated cell adhesion in HTR8 cells. We measured cell adhesion 1 hour after plating HTR8 cells, with or without p63 expression, at two different densities. We found that overexpression of p63 decreases cell adhesion to tissue culture plastic in a cell density-independent manner (Figure 7A). We next evaluated focal adhesion size and turnover, using phosphorylated FAK (pY397FAK). Immunocytochemical staining found smaller, more peripheral focal adhesion sites in p63-overexpressing cells, compared with the larger, more central focal adhesions in the control cells (Figure 7B). Western blot analysis confirmed this phenotype, showing FAK phosphorylation to be lower at the basal state and increasing more slowly after adhesion in p63-overexpressing cells compared with control cells (Figure 7C). Conversely, β-catenin expression was higher in p63-overexpressing cells (Figure 7C), localizing in abundance in cell-cell junctions (Figure 7D). These data indicate that p63 alters both cell migration and cell adhesion during trophoblast differentiation.

Figure 7.

p63 regulates cell adhesion properties. A: p63-overexpressing cells show decreased adhesion to tissue culture plastic 1 hour after plating. B: Immunocytochemical staining of pY397FAK. p63-overexpressing cells show more peripheral focal adhesions compared with the larger, more central focal adhesions in the mock control cells. C: Western blot analysis of pY397FAK, total FAK, and β-catenin at 0, 1, 8, and 24 hours after plating. Numbers below each blot reflect the amount of pY397FAK (normalized to total FAK) and β-catenin (normalized to β-actin), relative to mock cells at time 0. FAK phosphorylation is lower at the basal state and increases more slowly after plating in p63-overexpressing cells compared with mock control cells. β-catenin expression was higher in p63-overexpressing cells. β-actin served as loading control. D: Immunocytochemical staining of β-catenin. p63-overexpressing cells show more β-catenin in their cell-cell junctions compared with mock cells. Data are expressed as means ± SD of triplicate wells (A). ^∗P < 0.05 in comparison with mock-infected cells.

Discussion

Proper differentiation and function of EVT are crucial to pregnancy success, because these cells are responsible for establishing the maternal-fetal interface early on after implantation.¹ In this study, we have determined a role for p63, specifically ΔNp63α, in maintaining cells in a CTB stem cell-like state, based on both gene expression and functional evaluation.

Both immunohistochemistry and qPCR found that p63 expression is lost quickly on differentiation into the EVT lineage. Few other transcription factors show such a dramatic down-regulation going from villous stroma-adjacent CTB into the cell column. ID-2 has perhaps the most similar localization, but it is still expressed in the proximal column trophoblast, slightly removed from the villous stroma.¹⁷ Although p63 expression was not found in all proliferative trophoblast in the early human placenta, all p63⁺ CTBs were Ki-67⁺; in fact, p63 overexpression promoted the G₁- to S-phase transition in both human trophoblast cell lines tested. This is consistent with our previous data, showing that forced expression of p63 in term CTB inhibits hCG secretion and maintains cyclin B1 expression¹⁵; the latter has also been noted with ID-2 overexpression,¹⁷ and, at least in this respect, these two proteins appear to be functionally equivalent. Subsequent analysis, however, found that p63 function diverges from ID-2, particularly in terms of its effects on cell migration.

In fact, overexpression of ΔNp63α had multiple differentiation-inhibitory effects, decreasing migration in both cell lines and hCG secretion in JEG3 cells. Gene expression changes were consistent with this phenotype, showing that p63 increased markers associated with CTB, including EGFR and ITGB4, and decreased markers associated with EVT, including SNAI1, ITGB1, and several MMPs. Because MMPs are responsible for matrix remodeling,⁹ it is likely that modulation of their expression contributed, at least in part, to the p63-overexpression phenotype of blunted migration. Knockdown of p63 in JEG3 cells had a significantly milder phenotype but was nevertheless consistent with a role for p63 in maintaining the CTB state, with a slight increase in hCG secretion and associated gene expression changes. The rather mild knockdown phenotype can be attributed to a relatively low basal expression of p63 protein in JEG3 cells; however, it is also possible that p63 knockdown alone is not sufficient to induce trophoblast differentiation.

p63 overexpression in the primary CTB also interfered with EVT-associated gene expression, although the cells did not have significantly different surface expression of HLA-G or hCG secretion. This may have been partly due to the delayed expression of the transgene not being uniformly expressed until 3 days after infection (Figure 6A), leading to initiation of differentiation before significant transgene expression. This is supported by the fact that most decreases in EVT-associated gene expression were not seen until day 6 after infection (Figure 6B).

How p63 functions to maintain the CTB stem cell state is likely to be complex. p63 is a nuclear protein and transcription factor, known to have numerous target genes according to data from chromatin immunoprecipitation and knockout mouse models.¹⁸ Interestingly, some of the target genes, specifically of ΔNp63α, include CTB-associated markers EGFR and integrin β4 and β-catenin,¹⁸ some of the genes which we noted to be up-regulated in ΔNp63α-overexpressing HTR8 cells. ΔNp63α is thought to regulate cell adhesion through modulation of some of the same genes.¹² However, although overexpression of ΔNp63α induces FAK phosphorylation in primary mammary epithelial cells,¹² we noted lower basal levels and a slower induction of phosphorylated FAK in p63-overexpressing HTR8 cells after plating, although at later time points FAK phosphorylation appeared to catch up (Figure 7C). Our data are consistent with in vivo levels of pY397FAK in the placenta, with higher levels noted in invasive EVT.⁴ These data are significant because FAK phosphorylation/activity is required for focal adhesion turnover and cell migration.¹⁹ However, the change in FAK phosphorylation, in and of itself, is likely too small to account for the reduced migratory ability of p63-overexpressing cells. Rather, other changes, including reduced expression of MMPs after p63 overexpression, likely contribute to this phenotype.

Finally, p63 is known to regulate EMT, a process that has been studied in most detail in the context of metastasis. p63 targets include E-cadherin itself¹² and multiple ligands, receptors, and effectors of transforming growth factor-β/bone morphogenic protein signaling, a crucial regulator of EMT.¹⁸ In addition, a recent study in bladder cancer reported a role for miRNA miR-205 downstream of inhibition of EMT by ΔNp63α.¹⁴ In the context of the placenta, EMT remains somewhat controversial^20,21; for example, although E-cadherin expression has been shown to be reduced in the cell column,⁸ Floridon et al²² have noted E-cadherin to be re-expressed in a subpopulation of invasive EVT. Nevertheless, the loss of apico-basal polarity and gain of migratory and invasive potential during CTB-to-EVT differentiation are consistent with the general concept of EMT. Consistent with inhibition of EMT, we found a reduction in SNAI1 expression in both HTR8 and JEG3 cells overexpressing ΔNp63α; however, the reciprocal increase in E-cadherin expression was only noted at the level of protein. Nevertheless, functional data, including the decrease in migratory ability in ΔNp63α-overexpressing cells, were also consistent with inhibition of EMT.

Conclusion

Our data indicate a specific role for ΔNp63α, the main isoform of p63 in the human placenta, in inhibition of EVT migration and maintenance of the undifferentiated state in CTB. Although the direct targets of ΔNp63α in these cells remain to be further evaluated, changes in cell adhesion properties, accompanied by specific alterations in both gene expression and post-translational modification of adhesion-related proteins, at least partly explain its mechanism of action in maintaining the CTB stem cell state.

Supplemental Data

Supplemental Figure S1

p63 isoforms in primary first-trimester CTB and JEG3 cells. A: Real-time quantitative PCR for TA and ΔN isoforms shows ΔN isoform to predominate in both cell types. Error bars indicate SD of technical triplicates. B: Standard real-time PCR for α, β, and γ isoforms shows the α isoform to be the dominant one, with lesser amounts of β and γ isoforms. ^∗P < 0.05 in comparison with TA isoform values.