TJP1 suppresses trophoblast cell invasion by expressing E2F8 in the human placenta ☆

Adequate extravillous trophoblast (EVT) invasion into the maternal decidua is important for human placental development. We identified that E2F transcription factor 8 (E2F8) suppresses EVT invasion, and that tight junction protein-1 ( TJP1 ) is a potential downstream target gene of E2F8. We investigated the role of TJP1 in the human placenta and regulation of TJP1 expression by E2F8. TJP1 expression decreased in E2F8 knockdown HTR-8/SVneo cells. TJP1 and E2F8 were co-expressed in villi in the first-trimester placenta and in EVTs and villi in the third-trimester placenta. TJP1 was significantly increased in the pre-eclamptic compared with control placenta. TJP1 knockdown increased the invasion of HTR-8/SVneo cells, while TJP1 overexpression inhibited cell invasion. Halo-E2F8 overexpression significantly increased TJP1 expression and TJP1 transcription compared with control placenta. Our findings suggest that E2F8 promotes TJP1 transcription, and that TJP1 expression by E2F8 inhibits EVT invasion. TJP1 and E2F8 may be related to pre-eclampsia pathogenesis.


Introduction
Pre-eclampsia (PE) is the leading cause of mortality and morbidity among mothers and neonates (Lees et al., 2013).PE is associated with several factors, including maternal age, pre-pregnancy body mass index, and smoking.Although the pathogenesis of PE is not well understood, it is most commonly supported by the two-step theory (Staff, 2019).During placental development, cytotrophoblasts (CTBs) in cell columns differentiate into extravillous trophoblasts (EVTs).EVTs can migrate away from the CTB cell columns of anchoring chorionic villi, invade the maternal decidua, and replace the endothelial cells and remodel the spiral arteries in the myometrium, thereby allowing appropriate vessel dilation and increasing blood flow from the uterus to the placenta.In PE pathophysiology, the first step is the failure of EVT invasion resulting in high vascular resistance in the uterine arteries.The second step leads to the release of various molecules from syncytiotrophoblasts, resulting in the development of the characteristic symptoms of PE, including hypertension, proteinuria, and edema, via vascular endothelial damage.Furthermore, a small placenta owing to poor placentation results in fetal growth restriction, which is associated with poor prognosis in children born to mothers with PE (Liu et al., 2021).
TJP1 is an epithelial marker for epithelial-mesenchymal transition (EMT).EMT is a key process in cancer progression and metastasis and is characterized by the loss of epithelial markers and upregulation of mesenchymal markers.EVT invasion is regulated by several factors, including growth factors, nutrients, and extracellular matrix proteins, that are associated with EMT (Lamouille et al., 2014).EMT-associated gene change is observed during the differentiation of CTBs into EVTs (DaSilva-Arnold et al., 2015).It was reported that TJP1 is required for cell-cell fusion in human trophoblasts (Pidoux et al., 2010), and that TJP1 mRNA expression is elevated in the human PE placenta (Du et al., 2017).Although Tjp1 null mice exhibited normal fertility, delayed growth and development were evident from embryonic day 8.5, resulting in defects in vascular development (Katsuno et al., 2008).During pre-implantation, TJP1 is expressed in the trophectoderm, which forms trophoblasts, and is involved in morula-blastocyst formation in the next step (Wang et al., 2008;Paria et al., 1999;Ghassemifar et al., 2003).However, the role of TJP1 in EVT function remains unclear.
Based on these findings, we hypothesized that E2F8 suppression may downregulate TJP1 and that TJP1 inhibits the invasion of EVTs into the maternal uterus.Therefore, in this study, we elucidated the role of TJP1 in EVT invasion using HTR-8/SVneo cells and examined TJP1 expression and localization in the human placenta.Therefore, we investigated whether E2F8 regulates TJP1 expression in EVTs.

Subjects
This study was approved by the Ethics Committee of Nagoya University Hospital (approval number: 2008-0648, 2017-0302).For the use of first-trimester and third-trimester placental samples, the need for written informed consent was waived owing to the retrospective nature of these experiments (Mizuno et al., 2019).The study participants were pregnant women with a singleton delivery via cesarean section at Nagoya University Hospital.The control group comprised women with neonates with an appropriate gestational age (control group).PE was defined as hypertension (BP ≥ 140 mm Hg systolic and/or ≥ 90 mm Hg diastolic) and either proteinuria, other maternal organ dysfunction, or uteroplacental dysfunction after 20 weeks of gestation; this definition is consistent with the international definition.The inclusion criteria were strictly defined to decrease clinical variance in immunohistochemical (IHC) staining.Table 1 shows the clinical information of the participants who provided samples for IHC staining.

IHC staining and immunofluorescence analysis
IHC staining and immunofluorescence analysis were performed as described previously (Mizuno et al., 2019).Briefly, paraffin sections were deparaffinized and rehydrated, and antigens were retrieved using 10 mM sodium citrate buffer (pH 6.0) in a microwave oven and then treated with 0.3% H 2 O 2 in methanol for 20 min.Thereafter, the sections were treated with 10% normal goat serum, and incubated at 4 • C overnight with 1.87 μg/ml rabbit anti-human TJP1 antibody (21773-1-AP, Proteintech, Rosemont IL).After washing with phosphate-buffered saline (PBS), the sections were incubated with a secondary antibody (Histofine SAB-PO(R) kit; Nichirei Bioscience Inc., Tokyo, Japan).Sections were visualized using diaminobenzidine (Nichirei Bioscience Inc.) and counterstained with Mayer's hematoxylin (Wako Pure Chemical Industries, Ltd., Osaka, Japan).The immunoreactivity intensity was evaluated on a four-level scale (0: none, 1: weak, 2: moderate, 3: strong).Three representative fields per section were randomly selected and evaluated by two examiners independently, and their values were averaged.
For immunofluorescence analysis, the sections were incubated at nheim, Germany).The slides were mounted using Fluoromount™ (Diagnostic BioSystems Inc., Pleasanton, CA USA).The slides prepared for IHC and immunofluorescence were visualized under the BX43 microscope with a DP21 camera (Olympus, Tokyo, Japan) and the Fv1000 laser confocal microscope (Olympus).
The human 3′-terminal of the TJP1 fragment without a stop codon was amplified using human TJP1 pTOPO (MHS6278-211690364, Horizon Discovery, Cambridge, UK) as a template using Tks Gflex™ DNA Polymerase (TAKARA).The primer sequences used were as follows: forward primer 5′-CACCAGCGCGTCTCTCCACA-3′ and MluI site reverse primer (without a stop codon) 5′-CGCACGCGTAAAGTGGTCAA-TAAGGACAGAAACACAGTTTGCTCC-3′.The human TJP1 pTOPO and the 3′-terminal of the TJP1 PCR fragment were digested with EcoRI/ClaI and ClaI/MluI, respectively.Human TJP1 fragments were ligated into the pQCXIP vector (Clontech, Mountain View, CA, USA) with a C-terminal MluI-Flag.The inserts of the expression constructs were confirmed by DNA sequencing.

Total RNA isolation and real-time PCR
Total RNA was isolated using the RNeasy mini kit, followed by treatment with RNase-free DNase I (QIAGEN Inc., Hilden, Germany) according to the manufacturer's instructions.Total RNA was reverse transcribed into cDNA using the High-Capacity Reverse Transcription Kit and an RNase Inhibitor (Thermo Fisher Scientific Inc.).Real-time PCR was performed using Quant Studio3 (Thermo Fisher Scientific Inc.) to measure mRNA expression using the Fast SYBR® Green Master Mix (Thermo Fisher Scientific Inc.).Relative mRNA expression was calculated using the comparative Ct (ΔΔCT) method described previously (Livak and Schmittgen, 2001).GAPDH was used as the normalizing gene.Table 2 lists the primer sequences used in PCR.
The membranes were incubated with horseradish peroxidase-conjugated sheep anti-rabbit secondary antibody (Amersham; diluted 1:50000) for 1 h at room temperature.Antibody binding was visualized using ECL Plus (GE Healthcare, UK) and ImageQuant LAS 4010 (GE Healthcare).Band intensity was measured using ImageJ software (National Institutes of Health, Bethesda, MD, USA).

Cell migration and invasion assays
Cell migration and invasion assays were performed using Transwell (8-μm pores, 6.5-mm filter; Corning Costar, Inc., Corning, NY, USA) coated with or without matrigel (BD Biosciences, CA, USA.).The shControl, shTJP1-1 and shTJP1-2 cells (3 × 10 4 cells/well) or TJP1-Flag and Flag cells (6 × 10 4 cells/well) were added to the upper chamber of the Transwell and 700 μl of RPMI 1640 containing 10% FCS was added to the lower chamber.After incubation for 16 h at 37 • C under a 5% CO 2 atmosphere, the cells on the upper surface of the base membrane were removed using a sterile cotton swab and then stained with Giemsa Stain Solution (Nacalai Tesque, Inc.).The chamber membranes were photographed using the BX43 microscope with a DP21 camera and cellSence software (Olympus).

Luciferase reporter assay
The human TJP1 promoter fragment was amplified from the bacterial artificial chromosome clone RP-11-680F8 (Advanced Geno Techs Co., Tsukuba, Japan) using PCR.The primer sequences used were as follows: forward 5′-GGGCAACTTGTAAAAGTGACAAATAAGG-3′ and reverse 5′-CTTGTCTCTCTCCAGCGCCGCGCGA-3'.PCR was performed using Tks Gflex™ DNA Polymerase (Takara) under the following conditions: 2 min at 95 • C; 30 cycles for 40 s at 95 • C, 60 s at 60 • C; 2 min at 72 • C; and 7 min at 72 • C. A fragment of the human TJP1 promoter was inserted into the EcoRV site of the firefly luciferase pGL4.10 vector (Promega Corp, Madison, WI).The HaloTag, Halo-E2F8, shDsRed, or shE2F8-DsRed expressing HTR-8/SVneo cells were seeded into a 24-well plate at a density of 0.5 × 10 5 cells/well and incubated at 37 • C in a 5% CO 2 incubator.One day after seeding, the cells were co-transfected with the human TJP1 promoter-pGL4.10 vector and the internal control pRenilla-TK (Promega).After 24 h of transfection, luciferase activity was measured using ARVO X (PerkinElmer, Waltham, MA, USA) and the Dual-Luciferase Reporter Assay Kit (Promega) according to the manufacturer's instructions.Luciferase activity was normalized to the corresponding Renilla luciferase activity to determine transfection efficiency.

Statistical analysis
All experiments were performed at least in triplicates.R software (version 4.2.3) was used for statistical analysis.Data were expressed as the mean ± standard error of the mean (SEM) or medians [interquartile ranges].Statistical analysis was performed using the two-tailed unpaired Student's t-test, Wilcoxon rank-sum test, one-way ANOVA followed by Tukey's multiple comparison test, Kruskal-Wallis test followed by the steel test, or two-way ANOVA followed by Dunnett's multiple comparisons test based on whether or not equal variance was assumed or not assumed.Differences were considered statistically significant at a P-value of <0.05.

TJP1 immunolocalization in the cell column of first-trimester human placenta
Human placentas were analyzed to determine TJP1 expression.In the first-trimester placentas, TJP1 staining was observed on a part of the CTBs in the cell column (Fig. 2B, magenta arrowheads) and syncytiotrophoblasts (Fig. 2B, yellow arrows).Furthermore, TJP1-positive cells were present in the decidua (Fig. 2C, black arrows).To determine the location of TJP1, we performed immunofluorescence analysis using TJP1 and E2F8 with KRT7 as an epithelial marker including trophoblast.TJP1-positive cells were detected in CTBs and syncytiotrophoblasts (Fig. 3A, white arrows).In addition, E2F8 (Fig. 3B) was detected in CTBs, syncytiotrophoblasts (white arrows), and the cell column (Fig. 3L, asterisks).In the first-trimester placenta, TJP1 and E2F8 were coexpressed in the cell column, which was positive for KRT7 (Fig. 3K and L, asterisks).Nevertheless, TJP1-and E2F8-positive cells were not observed in KRT7-positive cells in the decidua (Fig. 3A-E, 3M, and 3N, yellow arrowheads).
TJP1-weakly positive and HLA-G-positive cells were observed in the cell column (Fig. 3O-S), but not detected in the decidua.These results showed that TJP1 and E2F8 are expressed in the CTB in first-trimester placenta.

TJP1-positive cells were co-stained with E2F8 in the third-trimester placenta
In the control placenta from the third-trimester, TJP1-positive cells were observed in the EVTs in the decidua (Fig. 4A, 4M, and 4N, yellow arrowheads) and the chorionic villi (Fig. 4A, K, and 4L, white arrows), which were stained with KRT7 (Fig. 4C, K, and 4M).E2F8 was also detected in EVTs (Fig. 4B and 4N, yellow arrowheads) and chorionic villi (Fig. 4B and 4L, white arrows).TJP1 was colocalized with E2F8 and KRT7 in the third-trimester placenta (Fig. 4E and 4M-N, yellow arrows).Moreover, TJP1-positive cells were observed in HLA-G positive cells, which served as a marker for EVT (Fig. 4O-R, and 4X).We performed IHC staining of TJP1 in ten placental specimens with PE and ten matched adjacent control placentas.TJP1 expression was significantly higher in EVTs in PE placentas (Fig. 5E and 5F, yellow arrowheads) than in control placentas (Fig. 5I, A, and 5B, yellow arrowheads, P < 0.01).TJP1 expression was higher in the chorionic villi in PE placentas (Fig. 5J and 5G, black arrows, P < 0.01) than in control placentas (Fig. 5C, black arrows).Moreover, E2F8 expression was significantly higher in EVTs in PE placentas (Fig. 6E and 6F, yellow arrowheads) than in control placentas (Fig. 6I, A, and 6B, yellow arrowheads, P < 0.01).E2F8 expression in villi increased in PE placentas (Fig. 6G, black arrows) compared with control placenta, but E2F8 staining score was not significantly different (Fig. 6J, P = 0.109).

TJP1 inhibition promotes HTR8/SVneo cell migration and invasion
To examine the role of TJP1 in the placenta, we knocked down TJP1 using two shRNA sequences (shTJP1-1 and shTJP1-2) in HTR-8/SVneo cells.TJP1 mRNA expression was significantly decreased in shTJP1-1 and shTJP1-2 cells compared with that in shControl cells (Fig. 7A, P < 0.01).To examine the effects of TJP1 on EVT functions, we performed migration and invasion assays.We observed a significantly increased number of migrated cells in shTJP1-1 and shTJP1-2 HTR-8/SVneo cells compared to that in shControl HTR-8/SVneo cells (Fig. 7B, P < 0.01 and P < 0.05, respectively).Moreover, the numbers of invading shTJP1-1 and shTJP1-2 cells were significantly higher than that of shControl cells (Fig. 7C, P < 0.05).These results suggest that TJP1 expression is negatively associated with cell migration and invasion in the placenta.To assess the proliferative ability of shTJP1-transfected cells, we examined cell growth using the CCK-8 assay.Decreased TJP1 expression did not affect the proliferation of HTR-8/SVneo cells (Fig. 7D).

TJP1 overexpression decreases HTR-8/SVneo cell migration and invasion
We determined the effects of increased TJP1 on cell migration and invasion.TJP1 expression was significantly increased in HTR-8/SVneo cells after infection with TJP1-Flag compared with Flag control cells (Fig. 7E, P < 0.05).However, the overexpression of TJP1-Flag did not affect E2F8 expression (Fig. 7E).Nevertheless, exogenous overexpression of TJP1-Flag resulted in a significant decrease in the migration of HTR-8/SVneo cells compared to Flag control HTR-8/SVneo cells (Fig. 7F, P < 0.01).Further, the invasive capacity of TJP1-Flag cells was lower than that of Flag control cells (Fig. 7G, P < 0.01).

Discussion
In the present study, we demonstrated for the first time that TJP1 expression is regulated by E2F8 in the human placenta and determined TJP1 function in human EVTs.E2F1 expression is lower in the PE placenta than in the control placenta (Kaitu'u-Lino et al., 2015), while E2F1 upregulation enhances the proliferation and invasion of EVTs through ZEB1 expression (Gong et al., 2022).Recently, several studies showed that EMT-related genes induced the downregulation of TJP1 in several tissues and tumors (Lamouille et al., 2014).However, the expression of EMT-related genes, such as SNAI1, SNAI2, TWIST1, ZEB1, and ZEB2, remained unchanged (Supplementary Fig. 1).These results suggest that E2F8 directly binds to the TJP1 promoter without upregulation of EMT-related genes and increases TJP1 expression in the placenta.Further investigation is required to determine the regulatory relationship between E2F8 and other factors.
TJP1 is expressed at low levels in several malignancies and is involved in tumor cell proliferation, invasion, and metastasis (Diaz-Coranguez et al., 2019;Martin, 2014).TJP1 is expressed in the proximal CTBs of the cell column in the first-trimester, suggesting that TJP1 downregulation by decreasing E2F8 expression is involved in EVT invasion from the cell column.Furthermore, we observed that TJP1 expression was upregulated in PE placentas compared with that in control placentas.EMT suppression markers, including TJP1, are significantly higher in PE placentas than in control placentas (Du et al., 2017).Moreover, the Illumina PE cohort study (NCT02808494) showed that cell free E2F8 RNA and TJP1 RNA were significantly increased in PE maternal plasma.These findings suggest that E2F8 and TJP1 may be related to the pathogenesis of PE placentas, although further investigation is needed.
In the present study, we showed that TJP1 and E2F8 significantly increased in the EVTs in PE placenta compare with those in control placenta.The differentiation of CTB into EVT in the first-trimester is accomplished via EMTs, with EVT displaying a metastable mesenchymal phenotype (DaSilva-Arnold et al., 2015;DaSilva-Arnold et al., 2018).However, the invasiveness of EVTs is lost in the third-trimester.It is often associated with placenta accreta, and is thought to be caused by a defect in the endometrium owing to inflammation of the endometrium or previous injury (Illsley et al., 2020;Abumaree et al., 2012;Zhou et al., 2022).Nevertheless, EVTs isolated from placenta accreta in non-adherent sites of decidua exhibit cell invasiveness and continue to express EMT-associated factors compared to normal third-trimester EVTs (Afshar et al., 2024;Ma et al., 2022;McNally et al., 2020).These findings suggest that the downregulation of TJP1, which is induced by decreased E2F8 activity, is related to the loss of invasiveness in the third-trimester placenta, although further investigation is warranted.
This study has several limitations.The patient sample size is small, and these findings should be replicated in a larger cohort to validate the present results.A major limitation of this study is that we did not perform the experiments using primary EVTs from first-trimester    Data are expressed as means ± SEM.The P-value was calculated using the Wilcoxon rank-sum test.*P < 0.01.placenta.Furthermore, the patient samples were derived from the placenta after birth and at the end of placentation.Thus, the high expression of TJP1 that was observed in PE placentas using IHC may show the result of placental hypoplasia rather than its cause.Therefore, further studies should be performed to confirm these results.

Conclusion
TJP1 is expressed in CTBs and syncytiotrophoblasts in the firsttrimester and EVTs and villi in the third-trimester.E2F8 regulates the TJP1 gene transcription in EVTs, while TJP1 expression inhibits the migration and invasion of EVTs.Furthermore, TJP1 is increased in PE placentas, and E2F8 and TJP1 may correlate with the pathogenesis of pre-eclampsia.

Compliance with ethical standards
This study was approved by the Ethics Committee of Nagoya University Hospital (approval number: 2008-0648, 2017-0302).Informed consent for sample collection was obtained from each patient.For the use of first-trimester, control, and PE placental samples, written informed consent was waived due to the retrospective nature of these experiments.

Declaration of competing interest
All the authors declare that they have no competing interests.

Fig. 1 .
Fig. 1.Effects of E2F8 knockdown on mRNA expression and protein levels of TJP1.(A-B) Real-time PCR revealed that the expression of E2F8 (A) (n = 5) and TJP1 (B) (n = 5) decreased in shE2F8 cells.(C) Western blotting results showing TJP1 expression in shE2F8 and shControl cells (n = 5).The bar graph shows TJP1 band intensity normalized to that of beta Actin.Statistical bars in the graph represent the mean ± SEM.The P-value was calculated using the Student's t-test in (A) and the Wilcoxon rank-sum test in (B) and (C).*P < 0.01.

Fig. 2 .
Fig. 2. Immunolocalization of TJP1 in the first-trimester human placenta.(A-C) Immunohistochemical staining of TJP1 in human placenta in the first-trimester.(B and C) The higher-magnification images of the square box in A. (B) TJP1 was detected in syncytiotrophoblasts (yellow arrows) and cytotrophoblasts in the proximal side of the cell column (magenta arrowheads) in the first-trimester.(C) Black arrows represent TJP1 positive cells in the decidua.(D) The negative control for the image is shown in the right panel.Scale bar, 200 μm.

Fig. 3 .
Fig. 3. Localization of TJP1 and E2F8 in the first-trimester human placenta.(A-E) Immunofluorescence analysis of human placenta in the first-trimester.Staining was performed for cytokeratin 7 (KRT7) (C, magenta), a marker for EVTs and cytotrophoblasts, E2F8 (B, orange), TJP1 (A, green), and DAPI (D, blue), a marker for the nucleus.(F-J) The lower figures showed no staining with normal donkey serum used as the negative control.(K-N) The higher-magnification images of the square boxes in E and J. (O-R) Staining was performed for HLA-G (P, red), a marker for EVTs, TJP1 (O, green), and DAPI (Q, blue), a marker for the nucleus.(S) The higher-magnification images of the square box in R. (T-W) No staining was observed with normal donkey serum used as the negative control.White arrows, yellow arrowheads, and asterisks represent syncytiotrophoblasts, EVTs in the decidua, and EVTs in the cell column, respectively.Scale bar, 100 μm.

Fig. 4 .
Fig. 4. Localization of TJP1 and E2F8 in the third-trimester human placenta.Triple immunofluorescence staining of the control placenta in the third-trimester.(A-E) Staining was performed for cytokeratin 7 (KRT7) (C, magenta), a marker for EVTs and cytotrophoblasts, E2F8 (B, orange), TJP1 (A, green), and DAPI (D, blue), a marker for the nucleus.(F-J) The lower panels show no staining with normal donkey serum used as a negative control.(K-N) The higher-magnification images of the square boxes in E and J. (O-R) Staining was performed for HLA-G (P, red), a marker for EVTs, TJP1 (O, green), and DAPI (Q, blue), a marker for the nucleus.(S-V) No staining with normal donkey serum used as a negative control.(W and V) The higher-magnification images of the square box in R. White arrows and yellow arrowheads represent villi and EVTs, respectively.Scale bar, 100 μm.

Fig. 5 .
Fig. 5. Immunolocalization of TJP1 in the third-trimester human placenta.Representative immunohistochemistry (IHC) staining results of TJP1 in the PE placentas (E-H) compared to those in the Control placentas (A-D).TJP1 was detected in villi (arrows) and EVTs (yellow arrowheads).(D and H) The negative control for the image is shown in D and H. (B, C, F, and G) The higher-magnification images of the square boxes in A and E. Scale bar, 200 μm.(I and J) The TJP1 IHC score for EVTs (I) and villi (J) was higher in the PE group (n = 10) than in the Control group (n = 10).IHC results are shown as IHC scores of 0-3.The square boxes in the upper images indicated the location of the lower higher-magnification images.Scale bar, 100 μm.Data are expressed as means ± SEM.The P-value was calculated using the Wilcoxon rank-sum test.*P < 0.01.

Fig. 6 .
Fig. 6.The comparison of E2F8 between PE and Control placentas.Representative immunohistochemistry (IHC) staining results of E2F8 in the PE placentas (E-G) compared to that in the Control placentas (A-C).E2F8 was detected in villi (arrows) and EVTs (yellow arrowheads).(D and H) The negative control for the image is shown in D and H. (B, C, F, and G) The higher-magnification images of the square boxes in A and E. Scale bar, 200 μm.The E2F8 IHC score for EVTs (I) was higher in the PE group (n = 10) than in the Control group (n = 10), and villi (J) staining showed no significant difference.The square boxes in the upper images indicated the location of the lower higher-magnification images.Scale bar, 100 μm.

Fig. 7 .
Fig. 7. Effects of TJP1 knockdown on TJP1 and E2F family member mRNA expression in HTR-8/SVneo cells.(A) TJP1 mRNA expression was significantly decreased in shTJP1 cells compared with shControl cells (n = 8).(B) Migrated cell numbers per mm 2 were significantly increased in shTJP1-1 and shTJP1-2 cells.Representative images of the Transwell migration assay performed using shControl, shTJP1-1, and shTJP1-2 cells are shown (n = 6).(C) Effects of TJP1 knockdown on the invasive ability of HTR-8/SVneo cells were examined using the Transwell matrigel invasion assay (n = 6).(D) The time course of cell proliferation was measured using the Cell Counting Kit-8 assay.Cell growth curves did not differ significantly between shControl and shTJP1 cells (n = 4).shControl, shTJP1-1, and shTJP1-2 cells are represented as a solid line with red circles, a dashed line with blue triangles, and a solid line with green squares, respectively.(E) mRNA expression levels of TJP1 and E2F8 were determined using real-time PCR analysis of Flag control and TJP1-Flag cells (n = 4).(F) Migrated cell numbers per mm 2 were significantly decreased in TJP1-Flag cells (n = 4).(G) Effects of TJP1 expression on the invasive ability of HTR-8/SVneo cells were examined using the Transwell matrigel invasion assay (n = 4).Scale bar, 500 μm.Bar graphs were determined as mean ± SEM.The Kruskal-Wallis test was used in (A), (B), (C), and (E), two-way ANOVA was used in (D), and the Student's t-test was used in (F), and (G).*P < 0.05, **P < 0.01.

Table 1
Clinical characteristics of the study population.
Systolic blood pressure was indicated as the highest value in childbirth hospitals and diastolic blood pressure was indicated as the value measured at the same time point.Continuous variables are presented as medians [interquartile ranges] or number (percentage).P-values were calculated using the Student's t-test, † Wilcoxon rank-sum test, or ‡ Fisher's exact test.

Table 2
Primers used for Real-Time PCR.