Placenta
Volume 31, Issue 3 , Pages 213-221, March 2010

Decidual NK cell-derived conditioned medium enhances capillary tube and network organization in an extravillous cytotrophoblast cell line☆

  • Y. Hu

      Affiliations

    • Department of Obstetrics and Gynaecology, University of British Columbia, 2H30-4500 Oak Street, Vancouver, BC V6H 3N1, Canada
    • The Child and Family Research Institute, University of British Columbia, 950 28th Avenue, Vancouver, BC V5Z 4H4, Canada
    • ,
  • G. Eastabrook

      Affiliations

    • Department of Obstetrics and Gynaecology, University of British Columbia, 2H30-4500 Oak Street, Vancouver, BC V6H 3N1, Canada
    • The Child and Family Research Institute, University of British Columbia, 950 28th Avenue, Vancouver, BC V5Z 4H4, Canada
    • ,
  • R. Tan

      Affiliations

    • Department of Pathology and Laboratory Medicine, University of British Columbia, 950 28th Avenue, Vancouver, BC V5Z 4H4, Canada
    • The Child and Family Research Institute, University of British Columbia, 950 28th Avenue, Vancouver, BC V5Z 4H4, Canada
    • ,
  • C.D. MacCalman

      Affiliations

    • Department of Obstetrics and Gynaecology, University of British Columbia, 2H30-4500 Oak Street, Vancouver, BC V6H 3N1, Canada
    • The Child and Family Research Institute, University of British Columbia, 950 28th Avenue, Vancouver, BC V5Z 4H4, Canada
    • ,
  • J.P. Dutz

      Affiliations

    • Department of Dermatology and Skin Science, University of British Columbia, 950 28th Avenue, Vancouver, BC V5Z 4H4, Canada
    • The Child and Family Research Institute, University of British Columbia, 950 28th Avenue, Vancouver, BC V5Z 4H4, Canada
    • ,
  • P. von Dadelszen

      Affiliations

    • Department of Obstetrics and Gynaecology, University of British Columbia, 2H30-4500 Oak Street, Vancouver, BC V6H 3N1, Canada
    • The Child and Family Research Institute, University of British Columbia, 950 28th Avenue, Vancouver, BC V5Z 4H4, Canada
    • Corresponding Author InformationCorrespondence to: P. von Dadelszen, Department of Obstetrics and Gynaecology, University of British Columbia, 2H30-4500 Oak Street, Vancouver, BC V6H 3N1, Canada. Tel.: +1 604 875 3054; fax: +1 604 875 2725.

Accepted 10 December 2009. published online 18 January 2010.

Article Outline

Abstract 

Extravillous cytotrophoblast (EVT) migration, invasion and endovascular differentiation are regulated by a variety of growth factors, cytokines and adhesion molecules. Decidual natural killer cells (dNK) and their secreted cytokines probably modulate these processes. In this study, we used dNK-derived conditioned medium (dNK-CM) to investigate whether or not (i) dNK-CM was able to enhance capillary tube and network formation of an EVT cell line, HTR8/SVneo, on Matrigel, (ii) PI3K/AKT pathway and p38 MAPK pathway activation were involved, and (iii) HTR8/SVneo surface ICAM-1 played a role in the process of HTR8/SVneo endovascular differentiation. The results demonstrated that HTR8/SVneo constitutively form ‘vascular’ tubes and networks after culture on Matrigel. dNK-CM enhanced and maintained tube and network formation, acquiring an endothelium-like angiogenic morphology followed by increased VEGF-C production. HTR8/SVneo cell expression level of VE-cadherin, PECAM-1, VCAM-1 and αvβ3 was unaltered by dNK-CM, whereas ICAM-1 expression level was increased. Anti-human ICAM-1 blocking antibody inhibited HTR8/SVneo migration and partially reversed dNK-CM-mediated enhancement of HTR8/SVneo tube and network formation. PI3K/AKT and p38 MAPK pathways were activated in dNK-CM-mediated enhancement of HTR8/SVneo tube and network formation. The PI3K/AKT and p38 MAPK pathway inhibitors (LY294002 and SB202190, respectively) decreased dNK-CM-stimulated ICAM-1 induction, HTR8/SVneo migration, and reversed tube and network formation. The results suggest that dNK cell-secreted growth factors and cytokines participate in the regulation of HTR8/SVneo endothelium-like tube formation. Adhesion molecules, particularly ICAM-1, expressed on EVT may participate in the process. To our knowledge, this is the first report of a role for ICAM-1 in EVT angiogenesis, as previously reported for endothelial cells.

Keywords: Decidual natural killer cells, Extravillous cytotrophoblast, Angiogenesis, Capillary tube and network formation, VEGF and ICAM-1

Abbreviations: dNK cells, decidual natural killer cells, EVT, extravillous cytotrophoblast, dNK-CM, dNK-derived conditioned medium, CTB, cytotrophoblast

 

Back to Article Outline

1. Introduction 

Decidual NK cells (dNK) account for more than 40% of cells in decidua during early pregnancy and are found in direct contact with fetal trophoblast. dNK regulate trophoblast invasion and migration [1], [2], [3], and are involved in the process of spiral artery remodeling [4], [5]. dNK produce angiogenic factors such as vascular endothelial growth factor-C (VEGF-C), angiopoietin-1, angiopietin-2, placental growth factor (PLGF) and TGF-beta-1 (TGF-β1) [6], [7]. The secretion of these growth factors by NK has been suggested to initiate “decidual-associated remodeling of spiral artery”, preparing vessels for trophoblast infiltration and the final stages of spiral artery remodeling [6]. dNK also produce cytokines such as tumor necrosis factor-α (TNF-α) and interleukin-8 (IL-8); these cytokines regulate angiogenesis in a variety of model systems [8], [9], [10]. Thus, dNK may play an important role in modulating angiogenesis in early pregnancy.

Extravillous cytotrophoblasts (EVT) are highly invasive cells. They invade into uterine wall (interstitial invasion) and blood vessels (endovascular invasion), where they replace endothelial cells and remodel spiral arteries. During endovascular differentiation, EVT adopt endovascular phenotypes, expressing VE (endothelial cell)-cadherin, platelet–endothelial adhesion molecule-1 (PECAM-1), vascular endothelial adhesion molecule-1 (VCAM-1), and integrin alpha-v beta-3 (αvβ3) [11]. Poor trophoblast invasion of uterine tissue has been suggested to play an important role in pre-eclampsia, in which cytotrophoblast fails to express most of these endothelial markers [11], [12]. It is reasonable to assume that EVT may mimic the behavior of endothelial cells in angiogenesis during endovascular differentiation. The cellular and molecular mechanisms responsible for spiral artery remodeling during early pregnancy are under investigation.

The VEGF family, including VEGF-A and VEGF-C, comprises key growth factors that modulate angiogenesis, and they mediate their action by signaling through their cognate receptors such as VEGFR-1 and VEGFR-2 [13]. The EVT cell line, HTR8/SVneo, produces both VEGF-A and VEGF-C, and predominately expresses VEGFR-2. The HTR8/SVneo cells synchronized with the endothelial cells in a capillary network in an endothelial–trophoblast co-culture model. Term primary trophoblast and the cell line, TCL1, mainly produced VEGF-A, minimally expressed VEGFRs and failed to respond the signal from endothelial cells to form vascular structures. High expression of VEGF-C and VEGFR-2 on HTR8/SVneo cells may play a role in endothelial cell-induced HTR8/SVneo capillary tube structure formation in co-culture system [14].

Intercellular adhesion molecule-1 (ICAM-1) is an inducible surface glycoprotein belonging to the immunoglobulin superfamily. ICAM-1 is expressed in low level on the cell surface of a variety of cell types such as fibroblasts, leukocytes, endothelial cells and epithelial cells, and it is up-regulated in response to a number of inflammatory mediators like pro-inflammatory cytokines [15], [16], [17].

Cellular adhesion molecules mediate angiogenesis. ICAM-1 expression appears to regulate endothelial cell migration and angiogenesis, as loss of ICAM-1 expression results in reduced endothelial cell migration, with decreased AKT Thr308 synthesis, endothelial nitric-oxide synthesis, Ser1177 phosphorylation and NO bioavailability [18]. In vivo, VEGF-induced angiogenesis is attenuated in mice genetically deficient in ICAM-1 [19]. Ets-1 is a transcription factor that regulates protease gene expression and mediates angiogenesis [20], [21]. Polymorphonuclear leukocytes (PMN) have been reported to induce angiogenesis by stimulating ICAM-1 through Ets-1 expression on endothelial cells [22].

Villous trophoblast ICAM-1 is overexpressed in placentitis, which is characterized by the accumulation of leukocytes in the villi or intervillous space [23]. However, there is no direct evidence of ICAM-1 involvement in EVT migration and angiogenesis.

Recent studies have shown that mitogen-activated protein kinases (MAPK) are involved in angiogenesis through regulation of crucial processes such as endothelial cell proliferation, mobility, and differentiation into capillary-like structures [24], [25], [26], [27]. P38alpha mutant mice suffer defective placental angiogenesis suggesting an important role for MAPK in angiogenesis [28].

Phosphoinositide 3-kinase (PI3K)/AKT signaling axis is activated by a variety of stimuli in endothelial cells and regulates multiple critical steps in angiogenesis including endothelia cell survival, migration, and capillary-like structure formation [29]. Various studies have demonstrated the involvement of PI3K/AKT in VEGF-induced rat aorta angiogenesis [30], green tea component epigallocatechin-3-gallate (EGCG)-induced angiogenesis [31], soluble E-selectin-induced angiogenesis [32], and statin-induced angiogenesis [33].

In this study, we observed that dNK-derived conditioned medium (dNK-CM) directly influenced EVT cell line, HTR8/SVneo, promoting proliferation, migration, and formation of capillary tube and network structure on Matrigel. These findings were associated with dNK-CM-enhanced VEGF-C production. Anti-ICAM-1 blocking antibody decreased HTR8/SVneo migration and partially reversed dNK-CM-mediated enhancement of vascular structure formation. Both the PI3K/AKT and the p38 MAPK pathways are involved in dNK-CM-mediated HTR8/SVneo capillary tube and network formation. ICAM-1 up-regulation induced by dNK-CM was attenuated by PI3K/AKT inhibitor LY294002 and p38 kinase inhibitor SB202190. This suggests that (i) HTR8/SVneo cells mimic endothelial cells in angiogenesis; (ii) the process is facilitated by dNK-CM; and (iii) down-regulation of ICAM-1 expression attenuates their migration, with subsequent inhibition of capillary tube and network formation. ICAM-1 induction is regulated by both PI3K and p38 MAPK pathways, and ICAM-1 up-regulation appears to regulate dNK-CM-mediated enhancement of HTR8/SVneo angiogenesis.

Back to Article Outline

2. Materials and methods 

2.1. Preparation of dNK-CM and culture of primary cytotrophoblast (CTB) and HTR8/SVneo cells 

Decidual tissues were obtained from healthy women undergoing elective pregnancy termination of a normal pregnancy at 6–10 weeks gestation after UBC Clinical Research Ethics Board-approved informed consent. dNK were separated and purified, as described [1]. dNK were seeded at a density of 5 × 105 cells/ml in the DMEM/F12 medium (Invitrogen, Burlington, ON) supplemented with 5 ng/ml of IL-15 and 10% FBS. The dNK-CM was harvested every 48 h during a culture period of up to 28 days; dNK-CM samples from each donor were then pooled. The pools of conditioned media from different donors were combined and aliquoted. dNK-CM containing 10% FBS was used for all assays other than Western Blots.

Serum-free dNK-CM was also collected. dNK-CM was harvested after 48 h incubation and dNK cells were then discarded. CM from different donors were pooled and saved for the Western Blots. Serum-free dNK-CM was also tried in HTR8/SVneo cell assays on proliferation, migration and capillary tube and network formation, with similar effect as serum-containing dNK-CM (data not shown).

CTB were isolated from first-trimester elective termination samples, as described, with minor modifications [34]. In brief, the CTB-containing cell suspension was collected from 0.1% trypsin digested villi tissue and centrifuged at 1600 × g for 20 min through a 40% Percoll solution. CTB-enriched cells were collected from the Percoll/medium interface. The collected cells were then added to dishes coated with 50 μg/ml of Matrigel overnight. The non-attached cells were removed next day, and the remaining attached cells were largely cytokeratin positive CTB (∼80–90%). The CTB obtained were cultured for 4–5 days prior to each assay. The CTB were only used in angiogenesis assay in thick layer Matrigel for initial confirmatory observations. HTR8/SVneo cells were used otherwise.

HTR8/SVneo cells were a gift from Dr Charles H. Graham (Queen's University, Kingston, ON, Canada) and have been described previously [35]. The cells were routinely maintained in RPMI-1640 containing 5% FBS. HTR8/SVneo cells were always grown in serum-free DMEM/F12 medium overnight before setting up the assays described below.

2.2. HTR8/SVneo proliferation assay 

The effect of dNK-CM on HTR8/SVneo proliferation was measured by mitochondrial lactate dehydrogenase-based MTS staining (Promega, Madison, WI) [1]. After serum-starvation, HTR8/SVneo cells were harvested and resuspended in serum-free DMEM/F12. Cells were then seeded into 96 well plates at a density of 1 × 104 cells/100 μl/well in the absence or presence of 100 μl of 1:2–1:8 diluted dNK-CM. The plates were incubated for 72 h at 5% CO2 and 37 °C temperature, after which time 50 μl of MTS was added to each well. Normally the plate was incubated for 4–6 h after adding MTS and then read at 490 nm. The results were expressed as % of control based on this formula: % of control = OD490 nm [assay group] × 100/OD490 nm [control medium group]. The assay was repeated for a total of 4 times.

2.3. HTR8/SVneo migration assay 

HTR8/SVneo migration assay was performed in 24 well format Transwell inserts with PET track-etched membranes; the pore size was 8 μm (Becton Dickinson Labware, Franklin Lakes, NJ). The inserts were coated with Matrigel (BD Biosciences) at 5 μg/insert for 2 h, and briefly washed with PBS. 2 × 105 cells in serum-free DMEM/F12 media were loaded to the top of the inserts. Either DMEM/F12 containing 5% FBS or 1:2–1:8 diluted dNK-CM with 5% final concentration of FBS was added to the bottom of the inserts. After overnight incubation, the media from the bottom of the inserts were taken off, and replaced with 500 μl of serum-free medium containing 5 μg/ml calcein AM (Invitrogen). On completion of the 1 h calcein AM incubation the inserts were transferred to new plates that contained 0.25% trypsin solution. The plate was incubated with shaking for 10 min. Thereafter, the inserts were discarded and the plate was read in a fluorescence plate reader [excitation 488 nm; emission 530 nm]. The results were expressed as % of control using this formula: % of control = RFU [assay group] × 100/RFU [control medium group]. The assay was repeated for a total of 4 times.

To study whether or not anti-human ICAM-1 antibody (R&D Systems, Minneapolis, MN), PI3K inhibitor (LY294002; Biomol International, Burlington, ON), or p38 MAPK inhibitor (SB202190; Biomol International) were able to modify basal HTR8/SVneo migration or dNK-CM-mediated HTR8/SVneo migration, HTR8/SVneo cells were pre-treated with either 10 μg/ml anti-ICAM-1 antibody, 20 μM LY294002, or 4 μM SB202190 for 1 h in inserts before the bottom of the inserts was loaded with medium; the remainder of the steps was performed, as described above.

2.4. HTR8/SVneo capillary tube and network formation assay on Matrigel 

300 μl of Matrigel diluted 1:1 with serum-free medium was added to 24 well plates and incubated at least for 1 h for gelling (thick layer Matrigel). 1 × 105 serum-starved HTR8/SVneo cells were added to the pre-solidified Matrigel. Equal volume of 1:2–1:8 diluted serum-containing dNK-CM with 5% FBS final concentration were added to cells. HTR8/SVneo cells started the process to form capillary tubes and networks once seeded on Matrigel. Overnight incubation revealed a consistent, and the most pronounced, difference between control medium-treated and dNK-CM-treated groups in terms of capillary tube and network morphology.

Digital images (40× magnification) were taken under all assay conditions, at least 5 different fields per well were imaged, and image analysis undertaken using Scion image software (Beta 4.0.2, Scion Corporation, Frederick, MD). Quantification of network complexity was made by measuring the total length of the tubes per 1 mm2. The final results were expressed as % of control using this formula: % of control = the total length of tubes [assay group] × 100/the total length of tubes [control medium group].

To study whether or not anti-human ICAM-1 blocking antibody, a PI3K inhibitor (LY294002) and a p38 MAPK inhibitor (SB202190) influenced capillary tube and network formation, serum-starved HTR8/SVneo cells were pre-incubated with the 10 μg/ml anti-human ICAM-1 antibody, 20 μM LY294002, or 40 μM SB202190 for 1 h, and then added to pre-solidified Matrigel for the assays.

To investigate if human recombinant VEGF-A (VEGF-165 R&D Systems) and VEGF-C (R&D Systems) had an effect on HTR8/SVneo capillary tube and network, 20–100 μg/ml VEGF-A and 200–400 μg/ml VEGF-C were added to HTR8/SVneo cells on pre-solidified Matrigel.

2.5. FACS staining 

To investigate the changes of vascular phenotypes and adhesion molecule expressions on HTR8/SVneo cells loaded on thick layer Matrigel, cells were harvested by 0.25% EDTA and processed for FACS staining.

To determine if either LY294002 or SB202190 modified dNK-CM-induced ICAM-1 expression, HTR8/SVneo cells were pre-treated with 20 μM LY294002 or 40 μM SB202190 for 1 h and then loaded to the thick layer Matrigel. dNK-CM was added to both compound treated or untreated cells overnight. Cells were harvested using 0.25% EDTA solution.

For FACS staining, cells were incubated with primary antibodies (5 μg/ml) on ice. After washing with PBS, cells were exposed to the relevant FITC-conjugated secondary antibody (BD Bioscience). Cells were analyzed in FACScan flow cytometry (BD Biosciences). Primary antibodies used included anti-human VE-cadherin (CD144) (R&D Systems); anti-human αvβ3 integrin (CD51/CD61) (R&D Systems); anti-human VCAM-1 (CD106) (R&D Systems); anti-human PECAM-1 (CD31) (R&D Systems); and anti-human ICAM-1 (CD54) (BD Biosciences).

2.6. Immunocytochemistry staining 

The expressions of vascular phenotype and adhesion molecules on primary CTB were performed by immunocytochemistry staining as described in our previous study [1]. Primary CTB were collected from explant cultures treated with medium control or dNK-CM [1] and then fixed with cold 70% methanol. The primary antibodies against VE-cadherin, αvβ3 integrin, VCAM-1, PECAM-1 and ICAM-1 were investigated.

2.7. ELISA 

VEGF-A and VEGF-C in the dNK-CM and HTR8/SVneo supernatants were measured by Duoset ELISA kit and Quantikine ELISA kit respectively following manufacturer's instruction (R&D Systems).

2.8. Western blots 

Serum-starved HTR8/SVneo cells were seeded in 6 well plates pre-coated with 2-fold diluted thick layer Matrigel. Cells were treated with serum-free medium as a control or 1:2 diluted serum-free dNK-CM overnight. The cells were then washed briefly with PBS to avoid damage to the Matrigel layer. The HTR8/SVneo cells from angiogenesis assays were incubated with cell lysis buffer 20 mM Tris, pH 7.5 containing 150 mM NaCL, 1% NP-40, 10 mM EDTA, 1 mM PMSF, 50 mM NaF, 40 mM β-glycerophosphate, 200 μM sodium orthovanadate, 40 mM benzamidine, and protease inhibitor cocktail (EMD, Gibbstown, NJ) on ice for 20 min with gentle shaking. The lysates were collected and centrifuged at 12,000 × g for 5 min at 4 °C. The supernatants were transferred to new tubes, mixed with loading buffer, and boiled. Aliquots of 30 μg protein were resolved by SDS-PAGE. Following electrophoresis, the proteins on the gels were transferred onto nitrocellulose membrane (Bio-Rad, Hercules, CA), which then blocked with TBS containing 0.05% Tween-20 (TTBS) and 5% skim milk powder. After washing, the membranes were incubated with primary antibodies (concentration: 1 μg/ml) to AKT (Upstate, Billerica, MA), phospho-AKT (Ser473, Cell Signaling Technologies [CST], Beverly, MA), p38 MAPK (Upstate), phospho-p38 MAPK (Thr180/Tyr182, Santa Cruz Biotechnology), p44/42 MAPK (ERK) (CST) and phospho-ERK (Thr202/Tyr204, CST), with HTR-conjugated secondary antibodies (Bio-Rad). After washing three times with TTBS, the bound antibodies were developed using an ECL Western Blotting Detection System (Amersham Biosciences, Buckinghamshire, England).

2.9. Statistics 

Groups were compared using either parametric one-way analysis of variance (ANOVA), with Bonferroni's multiple comparison tests, non-parametric Kruskal–Wallis ANOVA, with Dunn's post-test, or Mann–Whitney U test, as appropriate using GraphPad Prism 4.0 software (San Diego, CA). Statistical significance was set at p < 0.05.

Back to Article Outline

3. Results 

3.1. dNK-CM increases HTR8/SVneo proliferation and migration 

dNK-CM increased HTR8/SVneo proliferation over 72 h incubation period, as measured by MTS staining, in a dose-dependent manner (Fig. 1a).

  • View full-size image.
  • Fig. 1 

    dNK-CM increased HTR8/SVneo proliferation (a; MTS assay) and migration (b; calcein AM-based Transwell plate migration assay) in a dose-dependent manner. For MTS assays, serum-starved HTR8/SVneo were seeded in serum-free DMEM/F12 in the absence or presence of dNK-CM, and incubated for 72 h, after which time MTS was added. The results were expressed as % of control = OD490 nm [assay group] × 100/OD490 nm [control medium group]. For the migration assay, HTR8/SVneo were loaded to Matrigel-coated Transwell inserts and incubated in either DMEM/F12/5% FBS or dNK-CM/5% FBS in the bottom overnight, followed by 1 h incubation with calcein AM. Migrated and calcein AM-labeled cells were then shaken off from the bottom of the plate by trypsinization. The plates with labeled cells were read and the results expressed as % of control = RFU [assay group] × 100/RFU [control medium group]. The assay was repeated for a total of 4 times. Dotted line and p value: Kruskal–Wallis non-parametric ANOVA. Dunn's multiple post-hoc test: *p < 0.05; **p < 0.01; ***p < 0.001, compared with control. Results normalized to mean control result (100%) for each experiment.

Serum-containing medium promoted HTR8/SVneo migration; this migration was significantly increased by dNK-CM, in a dose-dependent manner (Fig. 1b).

As dNK-CM increased HTR8/SVneo proliferation and migration, we investigated whether or not dNK-CM was able to enhance capillary tube and network formation of HTR8/SVneo cells on thick layer Matrigel.

3.2. dNK-CM enhances HTR8/SVneo capillary tube and network organization 

HTR8/SVneo cells had an intrinsic capacity to form capillary tubes and networks once cells were cultured on thick layer Matrigel. After overnight incubation, some network arms became disrupted and most of HTR8/SVneo cells aggregated into clumps. dNK-CM enhanced capillary tube and network formation with morphology typical of angiogenesis (Fig. 2a). The total length of tubes was increased more than 2-fold when 1:2 diluted dNK-CM was used compared with control medium alone (Fig. 2b). Neither bare nor collagen gel-coated plates supported HTR8/SVneo to form either typical capillary structures or networks as observed on Matrigel, even after exposure to dNK-CM (data not shown). dNK-CM collected and pooled from 28 day-cultured dNK cells in the system contained approximately 12 pg/ml of VEGF-A and 5 pg/ml of VEGF-C. Recombinant VEGF-A at the range of 20–100 ng/ml did not show influence on HTR8/SVneo capillary structure formation; VEGF-C at 200–400 ng/ml tended to increase capillary tube and network formation (data not shown). HTR8/SVneo on Matrigel produced both VEGF-A and VEGF-C, and dNK-CM increased HTR8/SVneo production VEGF-A and VEGF-C by 118.9% and 178.8%, respectively compared with control medium.

  • View full-size image.
  • Fig. 2 

    dNK-CM enhanced HTR8/SVneo, and primary CTB, capillary tube and network organization. Serum-starved HTR8/SVneo cells were added to pre-solidified Matrigel. Overnight incubation with either dNK-CM/5% FBS or DMEM/F12/5% FBS (control) revealed differences between control medium-treated and dNK-CM-treated groups in terms of capillary tube and network morphology (a) and total network tube length (b), using image analysis software. These concentration-dependent findings were confirmed using primary first-trimester CTB (c and d). Dotted line and p value: Kruskal–Wallis non-parametric ANOVA. Dunn's multiple post-hoc test: *p < 0.05; **p < 0.01; ***p < 0.001, compared with control (b); Mann–Whitney U test (d). Results normalized to mean control result (100%) for each experiment. Scale bar: 250 μm.

To confirm the direct relevance of the cell line data, we determined that capillary tube and network formation by first-trimester primary CTB was increased by treatment with dNK-CM after 4–6 h incubation (Fig. 2c). The total length of tubes was increased by 838% (Fig. 2d).

Therefore, dNK-CM promoted HTR8/SVneo cells capillary tube and network formation. We examined whether or not dNK-CM modified the HTR8/SVneo vascular phenotype and adhesion molecule expression (e.g., αvβ3 integrin, ICAM-1, PECAM-1, VCAM-1, and VE-cadherin).

3.3. dNK-CM-mediated changes of vascular phenotype and adhesion molecule expression by HTR8/SVneo 

Both control medium- and dNK-CM-treated HTR8/SVneo cells expressed high levels of VE-cadherin; no difference was observed between groups. HTR8/SVneo cells either did not express, or expressed very low level of VCAM-1, PECAM-1 and αvβ3 integrin, and no significant differences were noted between control medium- and dNK-CM-treated cells. HTR8/SVneo expressed a relatively high level of ICAM-1; dNK-CM increased its expression level by 287% (Fig. 3). This observation was confirmed using primary CTB (Fig. 3). Therefore, the involvement of ICAM-1 in dNK-CM-mediated enhancement of HTR8/SVneo migration and capillary tube and network formation was investigated.

  • View full-size image.
  • Fig. 3 

    Vascular phenotype and adhesion molecule expressions on HTR8/SVneo cells, by flow cytometry with FITC-conjugated antibodies (upper panels) and by confirmatory immunohistochemistry staining on primary CTB cells (lower panels). Flow cytometry: red: isotype control; green: control medium-treated; blue: dNK-CM-treated. dNK-CM increased ICAM-1 expression by HTR8/SVneo and primary CTB cells compared with control serum Both control medium- and dNK-CM-treated HTR8/Svneo and primary CTB cells expressed VE-cadherin with no difference between groups. Other vascular and adhesion molecules were poorly expressed. Scale bar: 70 μm.

3.4. Anti-ICAM-1 antibody blocks dNK-CM-mediated enhancement of HTR8/SVneo migration and capillary tube formation 

Serum-starved HTR8/SVneo cells were loaded onto 24 well Transwell inserts. Cells were pre-treated with anti-human ICAM-1 antibody prior to filling up the bottom of the assay plate with media. According to the product data sheet, the anti-ICAM-1 antibody inhibits cell adhesion to recombinant ICAM-1-coated surfaces. HTR8/SVneo migration was reduced to 67.5% compared with control medium (Fig. 4a). dNK-CM increased HTR8/SVneo migration by 131.8% and it was reduced to 21.3% compared with control medium in the presence of anti-ICAM-1 antibody (Fig. 4a).

  • View full-size image.
  • Fig. 4 

    Anti-ICAM-1 antibody blocked dNK-CM-mediated enhancement of HTR8/SVneo migration (a) and capillary tube formation (b and c). Serum-starved HTR8/SVneo cells were loaded into top of the inserts and pre-incubated with anti-ICAM-1 antibody prior to filling bottom of the plate with control medium or dNK-CM. The remainder of the procedures was performed by regular migration assay. The HTR8/SVneo migration was increased by dNK-CM, and the enhanced effect was reduced by anti-ICAM-1 antibody. For the capillary tube formation on Matrigel, serum-starved HTR8/SVneo cells were pre-incubated with anti-ICAM-1 antibody before adding to Matrigel. dNK-CM-induced enhancement of capillary tube formation was partially reversed by anti-ICAM-1 antibody. Dotted line and p value: Kruskal–Wallis non-parametric ANOVA. Dunn's multiple post-hoc test: *p < 0.05; **p < 0.01; ***p < 0.001, compared with control; †p < 0.05; †††p < 0.001, compared with dNK-CM. Results normalized to mean control result (100%) for each experiment. Scale bar: 250 μm.

Therefore, anti-ICAM-1 antibody disproportionately inhibited dNK-CM-induced migration compared with intrinsic HTR8/SVneo migration. Anti-ICAM-1 antibody pre-treated or untreated cells were loaded to pre-solidified thick layer Matrigel in the absence and presence of dNK-CM. dNK-CM-mediated angiogenic effects on HTR8/SVneo were partially inhibited (Fig. 4b). 1:2 diluted dNK-CM increased total length of tubes by 228.5% compared with control medium. Total tube length was only increased by 141.8% in the presence of anti-ICAM-1 antibody (Fig. 4c). The total tube length in the medium control group was not significantly affected by anti-ICAM-1 antibody (Fig. 4c).

The PI3K and MAPK pathways are important in the process of angiogenesis (26, 27, 29–33). We examined whether or not these pathways were activated by dNK-CM in HTR8/SVneo angiogenesis on Matrigel.

3.5. dNK-CM-mediated PI3K/AKT and p38 MAPK pathway activation in HTR8/SVneo cells 

HTR8/SVneo cells were seeded into 6 well plates pre-coated with thick layer Matrigel in the presence or absence of dNK-CM. DMEM/F12 medium containing serum stimulated the phosphorylation of AKT and p38. To exclude the serum-stimulation effect and to elucidate the role of CM derived from dNK on pathway activation and kinase phosphorylation, serum-free DMEM/F12 medium was used as a control and serum-free dNK-CM was used for treatment. After 24 h incubation, cells were harvested and processed for Western Blot. Phosphorylation of AKT, p38 and ERK (pAKT, pp38 and p44/p42 ERK) were examined.

High levels of p44/p42 ERK were observed in both control medium- and dNK-CM-treated HTR8/SVneo cells (Fig. 5). While pAKT and pp38 were low in serum-free control medium group, higher levels of pAKT and pp38 were observed in HTR8/SVneo cells exposed to serum-free dNK-CM (Fig. 5).

  • View full-size image.
  • Fig. 5 

    dNK-CM activated PI3K/AKT and p38 MAPK pathways in HTR8/SVneo. HTR8SVneo were plated on Matrigel and exposed to serum-free DMEM/F12 (control) and serum-free dNK-CM. After 24 h incubation, cells were harvested for Western Blot analysis. High levels of p44/p42 ERK were observed in both control medium- and dNK-CM-treated HTR8/SVneo. While pAKT and pp38 were low in the control group, higher levels of pAKT and pp38 were observed in HTR8/SVneo exposed to serum-free dNK-CM.

PI3K/AKT and p38 MAPK pathways were activated after exposure to dNK-CM. Therefore, we studied whether or not pathway inhibitors influenced HTR8/SVneo ICAM-1 expression, migration, and capillary tube and network formation.

3.6. PI3K/AKT and p38 MAPK pathway inhibitors reduce HTR8/SVneo ICAM-1 expression, migration and capillary tube and network formation 

LY294002 and SB202190 reduced ICAM-1 expression on HTR8/SVneo cells induced by dNK-CM treatment (Fig. 6a). After stimulation with dNK-CM, ICAM-1 expression was increased by 413.8% compared with control medium. ICAM-1 expression was only increased by 263.4% and 318.5% in the presence of 20 μM LY294002 and 40 μM SB202190, respectively, compared with control medium. Neither inhibitor significantly reduced constitutive ICAM-1 expression (data not shown). LY294002 and SB202190, at non-toxic doses, reduced basal HTR8/SVneo migration and dNK-CM-mediated enhancement of HTR8/SVneo migration (Fig. 6b). LY294002 and SB202190 reduced basal migration by 71.9% and 62.6% respectively compared to medium control; and they reduced dNK-CM-mediated migration by 69.5% and 63.9% respectively compared with dNK-CM. The enhanced vascular structure formation induced by dNK-CM was also reversed by LY294002 and SB202190 (Fig. 6c). dNK-CM increased total length of tubes to 353.7% of control medium; however, this was 76.9% in the presence of 20 μM LY294002 and 87.4% in the presence 40 μM SB202190 (Fig. 6d).

  • View full-size image.
  • Fig. 6 

    PI3K/AKT and p38 MAPK inhibitors (LY294002 and SB202190, respectively) reduced HTR8/SVneo ICAM-1 expression (a), migration (b), and capillary tube and network formation (c and d). HTR8/SVneo cells were pre-treated with either LY294002 or SB202190, and then processed for capillary tube and network formation on Matrigel. The capillary tube and network formation was enhanced by dNK-CM, and the enhanced effect was reversed by either LY294002 or SB202190 (c and d). HTR8/SVneo cells were harvested for FACS analysis to assess ICAM-1 expression. ICAM-1 expression was enhanced by dNK-CM, and the enhanced effect was reduced either by LY294002 or SB202190 (a). For migration assay, serum-starved HTR8/SVneo cells were loaded into top of the inserts and pre-treated with LY294002 or SB202190 prior to filling bottom of the plate with medium or dNK-CM. dNK-CM was able to increase HTR8/SVneo migration, this effect was inhibited by either LY294002 or SB202190. In (a), red: isotype control; green: control medium-treated; blue: dNK-CM-treated; brown: dNK-CM and inhibitor treated (LY or SB). In (b) and (d), the dotted line and p value: Kruskal–Wallis non-parametric ANOVA. Dunn's multiple post-hoc test: *p < 0.05, compared with control; ††p < 0.01; †††p < 0.001, compared with dNK-CM. Results normalized to mean control result (100%) for each experiment. Scale bar: 250 μm.

Back to Article Outline

4. Discussion 

The dominant immune cells in the first half of human pregnant decidua are CD56bright CD16− NK cells that have low and inefficient cytotoxic function but possess potent cytokine producing ability. Hanna and his group showed that dNK cells regulated trophoblast invasion both in vitro and in vivo by production of the interleukin-8 and interferon-inducible protein-10 chemokines [36]. dNK cells were potent secretors of an array of angiogenic factors and induced the endothelial cell line, HUVEC, to form network-like structure [36]. Fukushima and his group used an EVT cell line, TCL1 cells to study the regulation of endovascular differentiation [8]. They found that TCL1 cells could form tube-like structures that specifically recall endothelial cell morphological changes. TNF-α, VEGF, and ECM collaboratively regulated EVT behavior, including cell survival and endovascular differentiation. In this study, we investigated the angiogenic effect of CM derived from dNK cells mainly on the EVT cell line, HTR8/SVneo.

We found that dNK-CM increased HTR8/SVneo proliferation and migration (Fig. 1a and b). HTR8/SVneo cells had intrinsic capacity to form tubes and network structures. Matrigel was required to support HTR8/SVneo capillary structure and network formation. However, tube and network structures were disrupted, and most of the HTR8/SVneo cells were aggregated into clumps, after overnight incubation in control medium. Stable capillary tube and network structures, which mimic endothelial cells, were observed in the presence of dNK-CM (Fig. 2a and b). Kalkunte's group found that HTR8/SVneo failed to form capillary-like tube formation when cultured alone but did in response to endothelial cell stimulation, and synchronized with endothelial cells in a capillary network [14]. Aldo's group reported that HTR8/SVneo cells migrated towards endothelial cell capillary network and eventually replaced endothelial cells [37]. The findings suggest that signal stimulation released from endothelial cells is important for EVT endovascular differentiation. The angiogenic factors contained in dNK-CM may possess similar stimulation effect on HTR8/SVneo endovascular differentiation. We observed similar dNK-CM effects on primary first-trimester CTB (Fig. 2c and d). These data suggest that, in early pregnancy, dNK have the potential to promote EVT endovascular differentiation.

During endovascular differentiation, human cytotrophoblasts transform their adhesion receptor phenotype so as to resemble the endothelial cells that they replace. The vascular phenotypes that cytotrophoblast express include VE-cadherin, PECAM-1, VCAM-1, and avβ3 integrin [12]. HTR8/SVneo cells cultured on thick layer Matrigel expressed high levels of VE-cadherin, which is a critical molecule for trophoblast–endothelial cell interaction in decidual spiral arteries [38] but minimally expressed PECAM-1, VCAM-1, and avβ3 integrin; no significant differences were observed between the control medium- and dNK-CM-treated groups (Fig. 3).

Angiogenesis involves rearrangement of endothelial junctions and implicates modulation of cell polarity. In VEGF-stimulated endothelial cell angiogenesis, surface expression level of junction adhesion molecule-C (JAM-C) did not change, however redistribution of JAM-C occurred within minutes [39]. Similar observations were reported in angiotension-II stimulated endothelial angiogenesis [40]. In our study, we did not observe redistribution changes in the molecules studied; it would be of great interest to investigate whether or not molecular redistribution occurs in dNK-CM-induced EVT angiogenesis.

HTR8/SVneo cells expressed a relatively high level of ICAM-1 and its expression was greatly increased by dNK-CM (Fig. 3a). The observation was confirmed on primary CTB cells (Fig. 3b). ICAM-1 in cytotrophoblasts is overexpressed during placenta inflammation [23]. dNK-CM contains a variety of ICAM-1-inducing inflammatory cytokines [15], [16], [41]. ICAM-1 mediates the migration of endothelial cells [18], [19], [42]. We found a similar functionality of ICAM-1 in HTR8/SVneo cells as occurs with endothelial cells (Fig. 4a). To our knowledge, this is the first report to show an effect of ICAM-1 in an EVT cell line.

Anti-ICAM-1 antibody significantly inhibited the basal migration of HTR8/SVneo cells (Fig. 4a). With dNK-CM stimulation, HTR8/SVneo migration was increased; the mobility induced by dNK-CM was much more inhibited by ICAM-1 antibody compared with control medium (Fig. 4a). This suggests that anti-ICAM-1 antibody primarily modified inducible ICAM-1, and, to a smaller extent, constitutive ICAM-1. ICAM-1 deficiency in endothelial cells results in decreased AKT phosphorylation, increased actin stress fiber formation, and a lack of distinct cell polarity [18].

ICAM-1 induction in HTR8/SVneo cells was regulated by PI3K/AKT and p38 MAPK pathways. The ICAM-1 expression was inhibited significantly in the presence of LY294002 and SB202190, respectively (Fig. 6a). LY294002 revealed higher inhibition compared with SB202190 (Fig. 6a). The tube and network formation of HTR8/SVneo enhanced by dNK-CM was reduced by the application of anti-ICAM-1 antibody (Fig. 4b and c); and it was also reversed by the presence of PI3K/AKT inhibitor LY294002 and p38 MAPK inhibitor SB202190 (Fig. 6c and d). The results imply that activation of the PI3K/AKT and p38 MAPK pathways are critical in the system. ICAM-1 induction is regulated by both PI3K and MAPK pathways. However, we did not exclude the possibility that dNK-CM might activate ICAM-1, which in turn might activate PI3K/AKT and p38 MAPK pathways. This phenomenon was reported in fibrinogen-induced migration of human vascular smooth muscle cells, in which ICAM-1 blocking antibody reduced phosphorylation of AKT and p38 upon fibrinogen stimulation [43] and ICAM-1 induction at least partially involved in the process of HTR8/SVneo cell angiogenesis. Clearly, studying the effect of ICAM-1 on PI3K/AKT and p38 MAPK phosphorylation is a next step that might strengthen these findings.

dNK-derived CM was chosen for the study. As reported elsewhere, a variety of cytokines, chemokines, and growth factors secreted by dNK cells are claimed to promote trophoblast endovascular differentiation and angiogenesis [6], [8], [9], [10]. VEGF-A and VEGF-C are well-known angiogenic factors on endothelial cells. It promotes angiogenesis through interacting with two VEGF receptors, VEGFR-1 and VEGFR-2 [13]. HTR8/SVneo express these two receptors [14], [44]. We found that dNK-CM in the system contained low levels of VEGF-A and VEGF-C, although higher level of VEGF-C was observed in fresh isolated and short-term cultured dNK-CM (data not shown). HTR8/SVneo cells produced higher level of VEGF-C compared to VEGF-A upon dNK-CM stimulation on Matrigel. Up to 200–400 ng/ml of recombinant VEGF-C but not VEGF-A was observed to facilitate the capillary structure formation of HTR8/SVneo cells. The result agreed with the finding from Kalkunte's group [14]. VEGF-A was reported to stimulate proliferation but not migration or invasiveness in human extravillous trophoblast [44]. Cytokines, which have potential to induce ICAM-1 expression, also seem to play an important role in regulating the mobility and angiogenesis of HTR8/SVneo cells. HTR8/SVneo cells are immortalized EVT cells, and they have most of the characteristics of EVT cells [35].

Advanced interpretation of the role of ICAM-1 on EVT endovascular differentiation and angiogenesis would require investigation with primary EVT cells to expand these observations that we derived using an immortalized cell line.

In summary, dNK cells secreted angiogenic factors to participate in the regulation of HTR8/SVneo endothelium-like tube formation. Both PI3K/AKT and p38 MAPK pathways on HTR8/SVneo cells were activated upon dNK-CM stimulation. Adhesion molecules, particularly ICAM-1, expressed on HTR8/SVneo cells are involved in their migration and capillary tube and network formation on Matrigel.

Back to Article Outline

Acknowledgements 

We thank the CARE Program, BC Women's Hospital and Health Centre, for their assistance in gaining access to reproductive tissue, and Drs Geoff Hammond and Peter Leung for access to laboratory equipment.

Back to Article Outline

References 

  1. Hu Y, Dutz JP, MacCalman CD, Yong P, Tan R, von Dadelszen P. Decidual natural killer cells alter in vitro first trimester extravillous cytotrophoblast migration; a role for interferon-gamma. J Immunol. 2006;177:8522
  2. King A, Jokhi PP, Burrows TD, Gardner L, Sharkey AM, Loke YW. Functions of human decidual NK cells. Am J Reprod Immunol. 1996;35:258
  3. Koopman LA, Kopcow HD, Rybalov B, Boyson JE, Orange JS, Schatz F, et al. Human decidual natural killer cells are a unique NK cell subset with immunomodulatory potential. J Exp Med. 2003;198:1201
  4. Croy BA, Ashkar AA, Foster RA, DiSanto JP, Magram J, Carson D, et al. Histological studies of gene-ablated mice support important functional roles for natural killer cells in the uterus during pregnancy. J Reprod Immunol. 1997;35:111
  5. Guimond MJ, Wang B, Croy BA. Engraftment of bone marrow from severe combined immunodeficient (SCID) mice reverses the reproductive deficits in natural killer cell-deficient tg epsilon 26 mice. J Exp Med. 1998;187:217
  6. Lash GE, Schiessl B, Kirkley M, Innes BA, Cooper A, Searle RF, et al. Expression of angiogenic growth factors by uterine natural killer cells during early pregnancy. J Leukoc Biol. 2006;80:572
  7. Li XF, Charnock-Jones DS, Zhang E, Hiby S, Malik S, Day K, et al. Angiogenic growth factor messenger ribonucleic acids in uterine natural killer cells. J Clin Endocrinol Metab. 1823;2001:86
  8. Fukushima K, Miyamoto S, Tsukimori K, Kobayashi H, Seki H, Takeda S, et al. Tumor necrosis factor and vascular endothelial growth factor induce endothelial integrin repertories, regulating endovascular differentiation and apoptosis in a human extravillous trophoblast cell line. Biol Reprod. 2005;73:172
  9. Heidemann J, Ogawa H, Dwinell MB, Rafiee P, Maaser C, Gockel HR, et al. Angiogenic effects of interleukin 8 (CXCL8) in human intestinal microvascular endothelial cells are mediated by CXCR2. J Biol Chem. 2003;278:8508
  10. Lee CC, Liu KJ, Chen LL, Wu YC, Huang TS. Tumor necrosis factor-a, interleukin-8 and interleukin-6 are involved in vascular endothelial cell capillary tube and network formation induced by tumor-associated macrophages. J Cancer Mol. 2006;2:155
  11. Zhou Y, Fisher SJ, Janatpour M, Genbacev O, Dejana E, Wheelock M, et al. Human cytotrophoblasts adopt a vascular phenotype as they differentiate. A strategy for successful endovascular invasion?. J Clin Invest. 1997;2139:99
  12. Zhou Y, Damsky CH, Fisher SJ. Preeclampsia is associated with failure of human cytotrophoblasts to mimic a vascular adhesion phenotype. One cause of defective endovascular invasion in this syndrome?. J Clin Invest. 1997;2152:99
  13. Ferrara N, Gerber HP, LeCouter J. The biology of VEGF and its receptors. Nat Med. 2003;9:669
  14. Kalkunte S, Lai Z, Tewari N, Chichester C, Romero R, Padbury J, et al. In vitro and in vivo evidence for lack of endovascular remodeling by third trimester trophoblasts. Placenta. 2008;29:871
  15. Chang YJ, Holtzman MJ, Chen CC. Interferon-gamma-induced epithelial ICAM-1 expression and monocyte adhesion. Involvement of protein kinase C-dependent c-Src tyrosine kinase activation pathway. J Biol Chem. 2002;277:7118
  16. Chen C, Chou C, Sun Y, Huang W. Tumor necrosis factor alpha-induced activation of downstream NF-kappaB site of the promoter mediates epithelial ICAM-1 expression and monocyte adhesion. Involvement of PKCalpha, tyrosine kinase, and IKK2, but not MAPKs, pathway. Cell Signal. 2001;13:543
  17. van de Stolpe A, van der Saag PT. Intercellular adhesion molecule-1. J Mol Med. 1996;74:13
  18. Kevil CG, Orr AW, Langston W, Mickett K, Murphy-Ullrich J, Patel RP, et al. Intercellular adhesion molecule-1 (ICAM-1) regulates endothelial cell motility through a nitric oxide-dependent pathway. J Biol Chem. 2004;279:19230
  19. Kasselman LJ, Kintner J, Sideris A, Pasnikowski E, Krellman JW, Shah S, et al. Dexamethasone treatment and ICAM-1 deficiency impair VEGF-induced angiogenesis in adult brain. J Vasc Res. 2007;44:283
  20. Iwasaka C, Tanaka K, Abe M, Sato Y. Ets-1 regulates angiogenesis by inducing the expression of urokinase-type plasminogen activator and matrix metalloproteinase-1 and the migration of vascular endothelial cells. J Cell Physiol. 1996;169:522
  21. Oda N, Abe M, Sato Y. ETS-1 converts endothelial cells to the angiogenic phenotype by inducing the expression of matrix metalloproteinases and integrin beta3. J Cell Physiol. 1999;178:121
  22. Yasuda M, Shimizu S, Ohhinata K, Naito S, Tokuyama S, Mori Y, et al. Differential roles of ICAM-1 and E-selectin in polymorphonuclear leukocyte-induced angiogenesis. Am J Physiol Cell Physiol. 2002;282:C917–C925
  23. Juliano PB, Blotta MH, Altemani AM. ICAM-1 is overexpressed by villous trophoblasts in placentitis. Placenta. 2006;27:750
  24. McMullen ME, Bryant PW, Glembotski CC, Vincent PA, Pumiglia KM. Activation of p38 has opposing effects on the proliferation and migration of endothelial cells. J Biol Chem. 2005;280:20995
  25. Shin EY, Kim SY, Kim EG. c-Jun N-terminal kinase is involved in motility of endothelial cell. Exp Mol Med. 2001;3:276
  26. Yang B, Cao DJ, Sainz I, Colman RW, Guo YL. Different roles of ERK and p38 MAP kinases during tube formation from endothelial cells cultured in 3-dimensional collagen matrices. J Cell Physiol. 2004;200:360
  27. Ye J, Yuan L. Inhibition of p38 MAPK reduces tumor conditioned medium-induced angiogenesis in co-cultured human umbilical vein endothelial cells and fibroblasts. Biosci Biotechnol Biochem. 2007;71:1162
  28. Mudgett JS, Ding J, Guh-Siesel L, Chartrain NA, Yang L, Gopal S, et al. Essential role for p38alpha mitogen-activated protein kinase in placental angiogenesis. Proc Natl Acad Sci U S A. 2000;97:10454
  29. Shiojima I, Walsh K. Role of Akt signaling in vascular homeostasis and angiogenesis. Circ Res. 2002;90:1243
  30. Zhu WH, MacIntyre A, Nicosia RF. Regulation of angiogenesis by vascular endothelial growth factor and angiopoietin-1 in the rat aorta model: distinct temporal patterns of intracellular signaling correlate with induction of angiogenic sprouting. Am J Pathol. 2002;161:823
  31. Shankar S, Chen Q, Srivastava RK. Inhibition of PI3K/AKT and MEK/ERK pathways act synergistically to enhance antiangiogenic effects of EGCG through activation of FOXO transcription factor. J Mol Signal. 2008;3:7
  32. Kumar P, Amin MA, Harlow LA, Polverini PJ, Koch AE. Src and phosphatidylinositol 3-kinase mediate soluble E-selectin-induced angiogenesis. Blood. 2003;101:3960
  33. Urbich C, Dernbach E, Zeiher AM, Dimmeler S. Double-edged role of statins in angiogenesis signaling. Circ Res. 2002;90:737
  34. Hills FA, Elder MG, Chard T, Sullivan MH. Regulation of human villous trophoblast by insulin-like growth factors and insulin-like growth factor-binding protein-1. J Endocrinol. 2004;183:487
  35. Graham CH, Hawley TS, Hawley RG, MacDougall JR, Kerbel RS, Khoo N, et al. Establishment and characterization of first trimester human trophoblast cells with extended lifespan. Exp Cell Res. 1993;206:204
  36. Hanna J, Goldman-Wohl D, Hamani Y, Avraham I, Greenfield C, Natanson-Yaron S, et al. Decidual NK cells regulate key developmental processes at the human fetal–maternal interface. Nat Med. 2006;12:1065
  37. Aldo PB, Krikun G, Visintin I, Lockwood C, Romero R, Mor G. A novel three-dimensional in vitro system to study trophoblast–endothelium cell interactions. Am J Reprod Immunol. 2007;58:98
  38. Bulla R, Villa A, Bossi F, Cassetti A, Radillo O, Spessotto P, et al. VE-cadherin is a critical molecule for trophoblast–endothelial cell interaction in decidual spiral arteries. Exp Cell Res. 2005;303:101
  39. Lamagna C, Hodivala-Dilke KM, Imhof BA, Urrand-Lions M. Antibody against junctional adhesion molecule-C inhibits angiogenesis and tumor growth. Cancer Res. 2005;65:5703
  40. Dominguez-Jimenez C, Yanez-Mo M, Carreira A, Tejedor R, Gonzalez-Amaro R, Alvarez V, et al. Involvement of alpha3 integrin/tetraspanin complexes in the angiogenic response induced by angiotensin II. FASEB J. 2001;15:1457
  41. Li H, Nord EP. IL-8 amplifies CD40/CD154-mediated ICAM-1 production via the CXCR-1 receptor and p38-MAPK pathway in human renal proximal tubule cells. Am J Physiol Renal Physiol. 2009;296:F438
  42. Radisavljevic Z, Avraham H, Avraham S. Vascular endothelial growth factor up-regulates ICAM-1 expression via the phosphatidylinositol 3 OH-kinase/AKT/Nitric oxide pathway and modulates migration of brain microvascular endothelial cells. J Biol Chem. 2000;275:20770
  43. Rauch BH, Müschenborn B, Braun M, Weber A-A, Schrör K. ICAM-1 and p38 MAPK mediate fibrinogen-induced migration of human vascular smooth muscle cells. Eur J Pharmacol. 2007;577:54
  44. Athanassiades A, Hamilton GS, Lala PK. Vascular endothelial growth factor stimulates proliferation but not migration or invasiveness in human extravillous trophoblast. Biol Reprod. 1998;59:643

☆ The study was funded by a pilot grant from the Infection and Immunity Program, Child and Family Research Institute (CFRI). YH is funded through grants from the Canadian Institutes for Health Research, from whom PvD receives salary support. CDMacC and PvD receive salary support from CFRI. PvD, JPD, and RT receive salary support from the Michael Smith Foundation for Health Research.

PII: S0143-4004(09)00401-9

doi:10.1016/j.placenta.2009.12.011

Placenta
Volume 31, Issue 3 , Pages 213-221, March 2010

Article Tools

* Image Usage & Resolution