Extravillous Trophoblast and Endothelial Cell Crosstalk Mediates Leukocyte Infiltration to the Early Remodeling Decidual Spiral Arteriole Wall

Ruhul H. Choudhury, Caroline E. Dunk, Stephen J. Lye, John D. Aplin, Lynda K. Harris and Rebecca L. Jones
J Immunol May 15, 2017, 198 (10) 4115-4128; DOI: https://doi.org/10.4049/jimmunol.1601175

Abstract

Decidual spiral arteriole (SpA) remodeling is essential to ensure optimal uteroplacental blood flow during human pregnancy, yet very little is known about the regulatory mechanisms. Uterine decidual NK (dNK) cells and macrophages infiltrate the SpAs and are proposed to initiate remodeling before colonization by extravillous trophoblasts (EVTs); however, the trigger for their infiltration is unknown. Using human first trimester placenta, decidua, primary dNK cells, and macrophages, we tested the hypothesis that EVTs activate SpA endothelial cells to secrete chemokines that have the potential to recruit maternal immune cells into SpAs. Gene array, real-time PCR, and ELISA analyses showed that treatment of endothelial cells with EVT conditioned medium significantly increased production of two chemokines, CCL14 and CXCL6. CCL14 induced chemotaxis of both dNK cells and decidual macrophages, whereas CXCL6 also induced dNK cell migration. Analysis of the decidua basalis from early pregnancy demonstrated expression of CCL14 and CXCL6 by endothelial cells in remodeling SpAs, and their cognate receptors are present in both dNK cells and macrophages. Neutralization studies identified IL-6 and CXCL8 as factors secreted by EVTs that induce endothelial cell CCL14 and CXCL6 expression. This study has identified intricate crosstalk between EVTs, SpA cells, and decidual immune cells that governs their recruitment to SpAs in the early stages of remodeling and has identified potential key candidate factors involved. This provides a new understanding of the interactions between maternal and fetal cells during early placentation and highlights novel avenues for research to understand defective SpA remodeling and consequent pregnancy pathology.

Introduction

During the first half of pregnancy, uterine spiral arterioles (SpAs) undergo remodeling to optimize maternal perfusion of the placenta, ensuring that nutrient and oxygen supply can increase to meet fetal needs in later gestation (1). Disruption and loss of extracellular matrix, vascular endothelium, and smooth muscle cells allow irreversible expansion of the arterial channels with loss of vasoreactivity (2, 3). Failure of remodeling is associated with fetal growth restriction, late miscarriage, and/or preeclampsia (4–6).

The highly specialized immune environment of the uterus reflects its dual role in protecting the mother from blood-borne and mucosally transmitted pathogens, while supporting the development of the hemiallogeneic conceptus. The placenta is not immunologically silent; rather its active recognition by the maternal immune system is central to successful pregnancy outcome (7). Thus immune cells are abundant in the decidua, comprising almost 40% of the total (8). They include decidual NK (dNK) cells, macrophages, dendritic cells, and T cells, the former two comprising almost 90% of the total leukocyte population (9). dNK cells are phenotypically distinct from peripheral blood NK (PBNK) cells, coupling reduced cytotoxicity with an ability to secrete cytokines and angiogenic factors (10).

SpA remodeling is now understood to result from the cooperative activity of placental-derived extravillous trophoblasts (EVTs) and decidual immune cells (11–13). Although EVTs actively contribute to extracellular matrix breakdown and vascular cell loss (3, 10), recent evidence shows SpA remodeling may be initiated by infiltration of decidual leukocytes into arteriolar walls before migrating endovascular (v)EVTs reach the immediate locality (13, 14). Mice deficient in dNK and T cells exhibit impaired SpA remodeling and subsequently develop a preeclampsia-like disorder (15), and impaired regulation of macrophages is also associated with poor SpA remodeling (16, 17). However, the trigger that initiates immune cell infiltration is unknown (13).

Using a human placenta-decidua coculture model in which EVTs detach from anchoring villus tips and invade decidual arteries, we previously demonstrated that leukocyte infiltration of SpAs depends on the presence of EVTs in adjacent tissue (18). This is consistent with histological evidence showing that SpA remodeling takes place in the decidua basalis, where EVTs invade the decidual stroma (as interstitial EVTs) and plug (as vEVTs) the SpAs (19), but not in parts of the decidua distant from the placental site. These findings suggest an indirect interaction between EVTs and immune cells that drives immune cell infiltration of the SpA media.

EVTs produce a large repertoire of paracrine factors including cytokines, growth factors, and proteases, many of which have the potential to alter endothelial cell function. Chemokines have established roles in decidual function and leukocyte recruitment during pregnancy (20, 21, 23, 24). We therefore hypothesized that cytokines released by invading EVTs activate endothelial cells to upregulate the expression of dNK cells and macrophage chemoattractants, thus inducing infiltration of the SpA wall from the surrounding decidual stroma.

Materials and Methods

Primary tissues

First trimester placental tissue (5–9 wk gestation) (n = 16) was collected from women undergoing elective medical and surgical termination of pregnancy. Decidual tissue (6–12 wk gestation) (n = 22) was collected following surgical termination of pregnancy only. Placental and decidual tissues were dissected as described previously and processed for cell and tissue culture (13). Approval was obtained from North West Research Ethics Committee (08/H1010/28). Written informed consent was obtained from all women undergoing termination of pregnancy.

EVT outgrowths and production of conditioned medium

EVT conditioned medium (EVT-CM) (n = 9) was generated as described previously (25). Briefly, the terminal portions of villi from <9 wk placentae were dissected out and laid over a collagen I (Corning, Bedford, U.K.) surface in a 24-well plate. The explants were cultured in DMEM/F12 (Sigma, Gillingham, Dorset) with 10% FBS in 5% CO2 and 20% O2. After 48 h, outgrowths with 30% or more EVT coverage were washed in PBS and cultured in serum-free DMEM/F12 for a further 48 h. EVT-CM was then collected and centrifuged at 3000 × g for 5 min to remove any debris and stored at −80°C for further experiments.

Endothelial cell culture

Human uterine microvascular endothelial cells (HUtMvECs) (Lonza, Cologne, Germany) were cultured as described previously (26). Upon reaching confluency (passages 8–10), 1 × 105 cells were seeded per well in a six-well plate. After culturing in serum-containing EBM-2 (Lonza) for 24 h, the medium was removed and the cells were washed with PBS. The cells were cultured in a 1:1 mixture of serum-free DMEM/F12 and EBM-2 for a further 24 h, then treated with a 1:1 mixture of serum-free EBM-2 and either control medium (DMEM/F12) or EVT-CM for 24 h. After washing in PBS, RNA was extracted using the RNeasy mini kit (Qiagen, Manchester, U.K.), followed by treatment with the DNase kit (Ambion, Austin) to remove contaminating genomic DNA, both according to the manufacturer’s instructions. The RNA sample was assessed for purity and concentration using spectrophotometry and a Nanodrop (2000c; Thermo Scientific).

Ab array.

A proteome profiler cytokine array (ARY022; R&D Systems, Minneapolis) was carried out according to the manufacturer’s instructions on pooled EVT-CM (n = 6) to identify EVT-secreted factors. The presence of candidate proteins identified in the array was validated using individual samples; protein concentration was measured by ELISA, as described (27). Macrophage inhibitory cytokine-1 (MIC-1) concentration in EVT-CM was quantified using ELISA (R&D Systems, Abingdon, U.K.) according to the manufacturer’s instructions.

Endothelial cell culture with IL-6, CXCL8, and MIC-1 neutralizing Abs

HUtMvECs were treated with EVT-CM as described above, with the addition of neutralizing Abs for IL-6 and CXCL8 (6708 and 48311, mouse monoclonal; R&D Systems), both singularly and in combination, at an excess of 10 × (1 ng/ml) the original cytokine concentration. The experiment was also repeated using MIC-1 neutralizing Abs (20 ng/ml, (147627, mouse monoclonal; R&D Systems). HUtMvECs were treated with EVT-CM containing mouse IgG at a matching concentration as a negative control.

Gene expression analysis

RNA extracted from HUtMvECs treated with EVT-CM for 24 h (EVT-CM from nine different placentae), and untreated control cells were separately pooled and gene profiling performed using Affymetrix Human Genome U133 Plus 2.0 microarray. Quality control, normalization, and expression analysis in control and EVT-CM–treated groups were as previously described (28–30). Using positional update and matching algorithms, the probability of positive log ratio (PPLR) was calculated (30). Thresholds for significant changes in expression were predefined as PPLR scores of >0.9 (for upregulated genes) and <0.1 (for downregulated genes) and were used to select candidate chemokines. Chemokine expression was considered to be above background levels at a signal intensity of >50 arbitrary unit (AU).

Reverse transcription, real-time PCR, and ELISA

To validate the changes in mRNA expression of candidate chemokines detected by gene array, real-time PCR was carried out on individual RNA samples from EVT-CM–treated and untreated HUtMvECs (n = 8), as described previously (21). A total of 250 ng of RNA was used for reverse transcription using an AffinityScript Multi Temperature cDNA synthesis kit (Agilent, Berkshire, U.K.). PCR was performed using specific primers (Table I) with Brilliant III Ultra-Fast SYBR Green QPCR Master Mix (Agilent), and an annealing temperature of 60°C. The PCR data were normalized using the geometric mean of three housekeeping genes: YWHAZ, HMBS, and ZNF8232 (Table I) (22, 31). An ELISA (RayBiotech, Wembley, U.K.) was carried out according to the manufacturer’s instructions to quantify CXCL6 and CCL14 protein expressions in cell lysates and conditioned medium from EVT-CM–treated and untreated HUtMvECs (n = 6).

View this table:
Table I. Details of primer sequences for real-time PCR

Isolation of primary dNK cells and decidual macrophages

Primary dNK cells and decidual macrophages were isolated from decidual samples as described (32). Briefly, decidual samples were washed, minced, and shaken at 37°C for 30 min before passing through a 200-μm sieve. Stromal cells and leukocytes were separated from RBCs using a Lymphoprep density gradient (Cambridge, U.K.). CD56+ dNK cells and CD14+ macrophages were isolated by positive selection using magnetic microbeads (MACS Miltenyi Biotec, Surrey, U.K.) from each decidua. The purity of the cell isolation was measured using flow cytometry after immunostaining for CD56 and CD14, as previously described (32) (Supplemental Fig. 3).

Chemotaxis assay

Isolated dNK cells and macrophages were used for in vitro migration assays with recombinant human (rh) CCL14 (Canton) and CXCL6 (San Diego). CD56+ dNK cells were labeled with the nuclear fluorophore YOYO1 (Invitrogen) at 1:1000 in serum-free medium for 15 min at 37°C. The chemotaxis assay was carried out in 24-well plates containing transwell inserts with 3-μm pores. rhCCL14 and rhCXCL6 were added to the bottom of the wells at 10 ng/ml and 100 ng/ml, and 200,000 immune cells were added to the inserts. Ten percent FBS and serum-free medium (Life Technologies RPMI 1640, Leicestershire, U.K.) were used as positive and negative controls, respectively. After an overnight incubation, cells that had migrated through the filter were collected and their fluorescence read using a M200 plate reader (Tecan) with excitation at 480 nm and measurement at 525 nm. The cell number was determined using a standard curve. The assay was also repeated using culture medium from EVT-CM–treated HUtMvECs and in the presence of neutralizing Abs to CCL14 (1.5 ng/ml) (R&D Systems) and CXCL6 (1 ng/ml) (R&D Systems) at an excess of 10 × the original chemokine concentration. Chemotaxis assays with decidual CD14+ macrophages were also performed using the same method except for the transwell inserts, which contained 8-μm pores.

Primary decidual endothelial cell isolation and culture

Primary decidual endothelial cells were isolated as previously described (33). The cells were seeded on collagen (250 μg/ml)-coated six-well plates and cultured with EVT-CM or control DMEM/F12 medium as described above (n = 6). RNA from pooled and individual samples was collected for PCR prior to real-time PCR for chemokines, as above.

Immunohistochemistry and dual immunofluorescence

Immunohistochemistry was carried out on formalin-fixed, paraffin-embedded human first trimester decidua basalis (6–12 wk gestation), using colorimetric detection as described previously (13). Serial tissue sections (5 μm) were immunostained for vascular cells (α-smooth muscle actin [α-SMA] and CD31 [endothelial cells]), EVTs (HLA-G), and leukocyte common Ag (CD45) using mouse monoclonal Abs (Dako, Cambridgeshire, U.K.), and for CCL14 and CXCL6 using rabbit polyclonal Abs (Thermo Fisher Scientific, Cheshire, U.K.). Serial tissue sections were also immunostained for CXCL8 (goat anti-human polyclonal; Santa Cruz, Wembley U.K.) and IL-6 (mouse anti–IL-6 monoclonal; Novus Biologicals, Abingdon, U.K.). Immunocytochemistry was also carried out on 5 × 104 HUtMvECs seeded on 13-mm cover glass (Thermo Fisher Scientific) for CD31 and α-SMA (Supplemental Fig. 1) to assess the morphology of the cells.

For dual immunofluorescence, 5-μm formalin-fixed, paraffin-embedded decidua basalis samples were dual-stained for dNK cells or macrophages using mouse monoclonal Abs against CD56 (Invitrogen) or CD163 (AbD Serotech, Kidlington, U.K.) and the chemokine receptors CCR1, CCR5 (goat polyclonals; Santa Cruz), CXCR1, or CXCR2 (mouse monoclonals; R&D Systems). As Abs for CXCR1, CXCR2, and CD56 were raised in the same species, tissues sections were incubated with 10 μg/ml of unlabeled goat anti-mouse IgG after addition of the first primary Ab for 30 min at 37°C, to eliminate cross-reactivity. For immunofluorescence detection, donkey anti-mouse Alexa Fluor 488, donkey anti-mouse Alexa Fluor 568, and rabbit anti-goat Alexa Fluor 568 (Life Technologies, Cheshire, U.K.) conjugates were used. Whole mount immunofluorescence was also carried out using the same method on EVT outgrowth (n = 3) for EVT identity and cell death using monoclonal Abs against HLA-G and M30 (Roche, West Sussex, U.K.) (Supplemental Fig. 2), respectively.

Quantification of chemokine and cytokine staining

Intensity of CCL14, CXCL6, CXCL8, and IL-6 immunostaining in decidual sections was quantified using unbiased HistoQuest image analysis software version 3.5 (TissueGnostics, Vienna, Austria) as previously described (34). Briefly, seven images of remodeling SpAs, unremodeled SpAs from the decidua basalis and decidua parietalis, and veins from five different tissues were analyzed for mean DAB staining intensities in endothelial cells and the rest of the decidual cells. Mean DAB staining intensities for CXCL8 (n = 7, five individual tissues) and IL-6 (n = 7, five individual tissues) in vEVTs and surrounding total decidual cells were also analyzed using the same methods.

Statistical analysis

The Wilcoxon matched-pairs signed rank test was carried out to assess the difference between EVT-CM–treated and control HUtMvEC groups. The Wilcoxon matched-pairs signed rank test was also used for DAB intensity analysis in chemokine and cytokine staining. The Friedman test and Dunn multiple comparisons post hoc test were used to analyze the difference between groups following chemokine or cytokine neutralization experiments.

Results

Treatment of HUtMvECs with EVT-CM alters chemokine mRNA expression profile

Gene profiling was performed on HUtMvECs following stimulation with EVT-secreted factors. A total of 1179 genes were upregulated and 1140 genes were downregulated by EVT-CM using PPLR thresholds of >0.9 and <0.1 to define significant differential expression (Accession number E-MTAB-5467, ArrayExpress, https://www.ebi.ac.uk/arrayexpress/). Most of the altered genes are related to cell cycle regulation, dTMP de novo biosynthesis pathway and cancer signaling. For the purpose of this study, alterations in chemokine expression were prioritized. Six candidates with signal intensity >50 AU were selected for validation: CCL14, CCL20, CXCL2, CXCL6, CX3CL1 and CXCR7 (Tables I, II).

View this table:
Table II. Chemokine ligands/receptors differentially expressed in EVT-CM–treated HUtMvECs

CCL14 and CXCL6 are upregulated by HUtMvECs following stimulation with EVT-CM

As predicted by the microarray, all six chemokine/receptor candidate genes were confirmed by real-time PCR to be altered at the mRNA level by EVT-derived factors (p < 0.05 and p <0.01; Fig. 1A–F). CCL14 and CXCL6 were selected for further study, in line with our hypothesis that EVT-CM stimulates production of leukocyte chemoattractants by uterine vascular endothelial cells. Both chemokines were also upregulated at the protein level in HUtMvEC lysates (Fig. 1G, 1H; p < 0.05) and in HUtMvEC conditioned medium (HUtMvEC-CM) (p < 0.05; Fig. 1I, 1J), following treatment with EVT-CM. The chemokine levels in the EVT-CM were also measured; EVTs secreted a low concentration of CCL14 (4.0 pg/ml) and CXCL6 (8.0 pg/ml) compared with the higher mean concentration in the EVT-CM–treated HUtMvEC-CM (129 and 85 pg/ml respectively).

FIGURE 1.
FIGURE 1.

Alterations in chemokine mRNA and protein expression by HUtMvECs in response to EVT-CM. Real-time PCR verified that EVT-CM treatment significantly increased mRNA expression of (A) CCL14, (B) CXCL6, and (F) CXCR7 and decreased the expression of (C) CXCL2, (D) CCL20, and (E) CX3CL1 (n = 8). ELISA demonstrated increased concentrations of CCL14 and CXCL6 in EVT-CM–treated HUtMvEC lysates (G and H) and conditioned medium (I and J) (n = 6). Data are median ± interquartile range. *p < 0.05, **p < 0.01, Wilcoxon matched-pairs test.

CCL14 and CXCL6 are observed in endothelial cells of decidual SpAs and are upregulated following stimulation with EVT-CM

HUtMvECs were used as a model of decidual SpA endothelial cells. To assess whether SpA endothelial cells in the decidua basalis express CCL14 and CXCL6 in vivo, immunohistochemistry was performed. Unremodeled and early remodeling SpAs were identified by serial section immunostaining for vascular, immune cell, and trophoblast markers. Endothelial cells in both unremodeled (Fig. 2G, 2H) and early remodeling (Fig. 2C, 2D) SpAs expressed CCL14 and CXCL6. Both chemokines were additionally localized to endothelial cells in unremodeled SpAs in the decidua parietalis (Fig. 2K, 2L), venous endothelial cells (Fig. 2O, 2P), and decidual stromal cells, but were markedly absent from SpA smooth muscle cells (20). Quantitative analysis of immunostaining intensity using HistoQuest software demonstrated that CCL14 and CXCL6 protein expressions were significantly higher in endothelial cells of remodeling arteries (p < 0.05, Fig. 3A, 3B) compared with surrounding decidual cells (34). There was no difference in mean immunostaining intensity of either chemokine between endothelial cells of unremodeled SpAs and decidual cells (Fig. 3C–F). However, immunostaining intensity for CCL14, but not CXCL6, (Fig. 3G, 3H) was lower in venous endothelial cells than decidual cells (p < 0.01). The immunostaining intensity for CCL14 was significantly higher in endothelial cells in remodeling arteries compared with those in unremodeled arteries (Fig. 3I, p < 0.05). A similar pattern was detected for CXCL6 but this did not reach significance (Fig. 3J).

FIGURE 2.
FIGURE 2.

CCL14 and CXCL6 are expressed by decidua basalis spiral arteriolar endothelial cells. (A–D) Representative images of remodeling SpAs. (A) Disorganized smooth muscle cell layer and (B) disorganized endothelium. (C and D) Endothelial cells are positive for CCL14 and CXCL6 (black arrow). (E–L) Representative images of an unremodeled SpA in (E–H) decidua basalis and (I–L) decidua parietalis; endothelial cells are positive for CCL14 and CXCL6. (O and P) Representative images of vein in decidua, endothelial cells are positive for CCL14 and CXCL6. (M) Mouse IgG and (N) Rabbit IgG acted as negative controls. Original magnification ×20. Scale bar, 50 μm. α-SMA stains for smooth muscle cells and CD31 stains for endothelial cells.

FIGURE 3.
FIGURE 3.

Quantification of chemokine expression by vascular endothelial cells and decidual cells. (A and B) Mean immunostaining intensity for CCL14 and CXCL6 was significantly higher in endothelial cells of remodeling SpAs than surrounding decidual cells. Mean immunostaining intensity for CCL14 and CXCL6 in arteriole endothelial cells of (C and D) decidua basalis and (E and F) decidua parietalis was not significantly different from the surrounding decidual cells. Immunostaining intensity for (G) CCL14, but not (H) CXCL6 was higher in decidual cells than vein endothelial cells. (I) CCL14, but not (J) CXCL6 immunostaining intensity was significantly higher in endothelial cells in remodeling arteries compared with those in unremodeled arteries. Data are median ± interquartile range (n = 7). *p < 0.05, **p < 0.01, Wilcoxon matched-pairs test.

To verify the relevance of the EVT-induced changes in chemokine expression identified in HUtMvECs, real-time PCR analysis was performed on EVT-CM–treated primary endothelial cells isolated from decidual tissues. EVT-CM treatment induced increased expression of CCL14 (3.9-fold change) and CXCL6 (2.2-fold change) in pooled samples (n = 6). Analysis of individual samples (n = 4) demonstrated the consistency of the response, although the sample size was too small for robust statistical analysis (Fig. 4).

FIGURE 4.
FIGURE 4.

Alterations in chemokine mRNA expression by primary decidual endothelial cells in response to EVT-CM. real-time PCR verified that EVT-CM treatment increased mRNA expression of (A) CCL14, (B) CXCL6, and (F) CXCR7 and decreased mRNA expression of (C) CXCL2, (D) CCL20, and (E) CX3CL1 in all four individual samples. (n = 4). Wilcoxon matched-pairs test. The changes are not significant with p > 0.05.

CCL14 and CXCL6 chemoattract dNK cells

To investigate whether CCL14 and CXCL6 act as chemoattractants for dNK cells, in vitro migration assays were carried out. Primary human dNK cells migrated toward rhCCL14 and rhCXCL6 (Fig. 5A, 5C) at 100 ng/ml (p < 0.05). dNK cells also migrated toward HUtMvEC-CM when compared with serum-free medium (p < 0.05, Fig. 5E), and this migration could be effectively inhibited by cotreatment with CCL14- and CXCL6-neutralizing Abs in combination (p < 0.05, Fig. 5E).

FIGURE 5.
FIGURE 5.

Chemotaxis of dNK cells and decidual macrophages toward recombinant chemokines and HUtMvEC-CM. (A) CCL14 and (C) CXCL6 were chemoattractants for dNK cells at 100 ng/ml; (B) CCL14 but not (D) CXCL6 was a chemoattractant for decidual macrophages; (E) dNK cells and (F) decidual macrophages migrate toward conditioned medium from HUtMvECs stimulated by EVT-secreted factors. dNK cell and macrophage migration can be reduced significantly by cotreatment with neutralizing Abs against (E) CCL14 and CXCL6, and (F) CCL14, respectively. Data are median ± interquartile range (n = 8). * and ++ represent significant differences compared to serum-free and HUtMvEC CM groups, respectively. *p < 0.05 Friedman test with Dunn’s posthoc test.

CCL14, but not CXCL6, is chemoattractant to decidual macrophages

Primary decidual macrophages migrated in response to rhCCL14 (p < 0.05, Fig. 5B), but were unresponsive to rhCXCL6 (Fig. 5D). Decidual macrophages also migrated toward HUtMvEC-CM (p < 0.05, Fig. 5F) and cotreatment with CCL14 neutralizing Ab significantly reduced their migration (p < 0.05, Fig. 5F).

Decidual NK cells express receptors for CXCL6 and CCL14

Dual immunofluorescence of the decidua basalis revealed expression of CCR1 and CCR5 (CCL14 receptors), and CXCR1 and CXCR2 (CXCL6 receptors) by dNK cells (Fig. 6A–H). CXCR1 and a low level of CXCR2 were detectable on the majority of CD56+ NK cells, whereas only a subset expressed CCR1 and CCR5. Decidual stromal cells also express these receptors.

FIGURE 6.
FIGURE 6.

dNK cells and macrophages in decidua basalis express receptors for chemokines. Dual immunofluorescence of dNK cells (green) with (A and B) CXCR1, (C and D) CXCR2, (E and F) CCR1, and (G and H) CCR5 receptors (all red). Representative immune cells expressing receptors are indicated by white arrows. Dual immunofluorescence of decidual macrophages (green) with (I and J) CCR1 and (K and L) CCR5 receptors (all red). Negative controls: (N) mouse or (O) goat IgG. (M) Liver was used as a positive control.

Decidual macrophages express CCR1 and CCR5

Almost all decidual macrophages strongly expressed CCR1 and CCR5 (CXCL6 receptors) (Fig. 6I–L). CCL14 and CXCL6 receptor expressions in dNK cells and macrophages were consistent at 6–12 wk gestation.

EVTs secrete multiple cytokines and growth factors

An Ab array was employed to identify cytokines and growth factors present in EVT-CM. These included IL-6 and CXCL8, macrophage migration inhibitory factor (MIF), MIC-1 and CCL2 (Table III). The concentrations of MIC-1, CXCL8, and IL-6 were measured by ELISA in individual samples. High and consistent secretion of MIC-1 was observed in all samples (Table IV). CXCL8 and IL-6 secretions varied between samples but no gestation-dependent trend was observed (Table IV). MIC-1, IL-6, and CXCL8 were selected for further study due to their ability to act on endothelial cells (35, 36) and induce chemokine expression in other cell types (37, 38). To assess whether EVTs in the decidua basalis express IL-6 and CXCL8, immunohistochemistry was performed; both were expressed by EVTs in anchoring columns and vEVTs in remodeling decidual SpAs, although IL-6 immunoexpression was weaker than IL-8, consistent with the ELISA data (Fig. 7A–H, Table IV). The mean immunostaining intensity for CXCL8 was significantly (p < 0.05) higher in vEVTs (Fig. 7I) compared with surrounding decidual cells. However, there was no difference in mean IL-6 expression between vEVTs and decidual cells (Fig. 7J). Furthermore, the original microarray performed on unstimulated HUtMvECs showed that these cells expressed receptors for IL-6: IL-6R (311 AU), IL-6-ST (9586 AU) and CXCL8 (CXCR1 (102 AU).

View this table:
Table III. Analysis of EVT-CM by cytokine Ab array
View this table:
Table IV. EVTs differentially secrete MIC-1, CXCL8, and IL-6
FIGURE 7.
FIGURE 7.

CXCL8 and IL-6 are expressed by vEVTs. Serial sections of first trimester decidua basalis immunostained for (A and E) HLA-G (EVT marker), (D) α-SMA (smooth muscle cell marker), (B and F) CXCL8, and (C and G) IL-6. (A–C) Invading EVTs from anchoring villus are positive for CXCL8 and IL-6 (black arrow). (D) Extensive disorganization of smooth muscle cell layer (black arrow) in a remodeling arteriole. (F and G) Endovascular EVTs are positive for CXCL8 and IL-6. (H) Mouse IgG was used as a negative control. (A–C) Original magnification ×10 and (D–H) original magnification ×20. Scale bars, 50 μm. (I) CXCL8 immunostaining intensity is significantly higher in vEVTs than surrounding decidual cells. (J) IL-6 immunostaining intensity is similar in vEVTs and decidual cells. Data are median ± interquartile range (n = 7). *p < 0.05, Wilcoxon matched-pairs test.

Neutralizing IL-6 and CXCL8 in EVT-CM reduces the stimulation of CCL14 and CXCL6 expressions in HUtMvECs, but MIC-1 neutralization has no effect on chemokine expression

Addition of IL-6 and CXCL8 neutralizing Abs in combination to EVT-CM significantly reduced its ability to upregulate expressions of CCL14 and CXCL6 mRNA in HUtMvECs (p < 0.001; Fig. 8A, 8B). Treatment with IL-6 neutralizing Ab alone had a similar effect (p < 0.01), although this effect did not reach statistical significance when CXCL8 neutralizing Ab was used alone (p = 0.067). Neutralization of either cytokine alone in EVT-CM significantly reduced HUtMvEC expression of CXCL6 (p < 0.01, Fig. 8B). Addition of CXCL8 neutralizing Ab alone or in combination with IL-6 neutralizing Ab had no effect on expression of CXCL2, CCL20, CX3CL1, and CXCR7 mRNA (Fig. 8C–F). Neutralization of MIC-1 did not affect EVT-CM induced up- or downregulation of CCL14, CXCL6, CXCL2, CCL20, CX3CL1, and CXCR7 (Fig. 9A–F). The addition of a nonspecific IgG had no inhibitory effect. Findings were assembled into a hypothetical model of the initiation of SpA remodeling (Fig. 10).

FIGURE 8.
FIGURE 8.

Chemokine expression by HUtMvECs in response to EVT-CM in the presence and absence of IL-6 and IL-8 neutralizing Abs. (A) EVT-CM stimulated CCL14 mRNA expression compared with control medium. This effect was blocked by cotreatment with neutralizing Ab against IL-6, alone or in combination with anti–IL-8. (B) EVT-CM stimulated CXCL6 mRNA expression compared with control medium. This effect was blocked by cotreatment with anti–IL-8 or anti–IL-6, or both in combination. EVT-CM also reduced (C) CXCL2, (D) CCL20, and (E) CX3CL1, and stimulated (F) CXCR7 mRNA expression compared with control medium. Cotreatment with neutralizing Ab against IL-6 and/or IL-8 had no effect in the mRNA expression of these chemokines/receptors. Mouse IgG served as a negative control. Data are median ± interquartile range (n = 8). Friedman test with Dunn post hoc test. +p < 0.05, ++p < 0.01 compared with control. *p < 0.05, **p < 0.01, ***p < 0.001 compared with EVT treatment alone.

FIGURE 9.
FIGURE 9.

Chemokine expression by HUtMvECs in response to EVT-CM in the presence and absence of MIC-1 neutralizing Abs. EVT-CM treatment significantly increased expression of (A) CCL14, (B) CXCL6, and (F) CXCR7 and decreased expression for (C) CXCL2, (D) CCL20, and (E) CX3CL1. Cotreatment with neutralizing Ab against MIC-1 had no effect on the expression of the chemokines/receptors. Mouse IgG served as a negative control for the neutralizing Abs. Data are median ± interquartile range (n = 8). Friedman test with Dunn post hoc test. +p < 0.05, ++p < 0.01, compared with control.

FIGURE 10.
FIGURE 10.

Hypothetical model of EVT-endothelial interactions regulating immune cell infiltration of decidual SpAs based on the results of the study. (A) vEVTs (light blue) secrete IL-6 and CXCL8 that induce spiral artery endothelial cells (blue) to upregulate and secrete (B) CCL14 and CXCL6, which subsequently induce (C) recruitment of leukocytes (red) in the vessel. (D) Disorganization of smooth muscle cells and endothelial cells as leukocytes infiltrate the arteries and initiate remodeling.

Discussion

The findings from this study support the hypothesis that EVTs interact indirectly with SpA endothelial cells to facilitate the recruitment of decidual immune cells into the SpA wall during the early stages of remodeling. We identified two candidate endothelial chemokines, CCL14 and CXCL6, which are upregulated in response to EVT-secreted factors and induce the migration of dNK cells and decidual macrophages. Inhibition of IL-6 and CXCL8 reduced the stimulatory effect of EVT-CM on the expression of these chemokines indicating that they are two key EVT-derived activating factors. Thus we propose a model whereby factors including IL-6 and CXCL8, derived from EVTs in the upper segments of decidual SpAs, activate endothelial cells to release CCL14 and CXCL6, which then attract dNK cells and macrophages from the surrounding tissue into the SpAs to initiate remodeling (Fig. 10). To our knowledge, the in vitro findings suggest, for the first time, a mechanism by which decidual immune cells are stimulated to infiltrate the SpAs and provide evidence of complex interactions between fetal EVTs and maternal vascular and immune cells in coordinating the process. It may also explain why arteriolar walls in the decidua parietalis, despite being surrounded by tissue rich in dNK cells and macrophages, remain free of infiltrating immune cells.

Decidual stromal cells and endothelial cells in unremodeled vessels and veins also expressed both of the chemokines of interest. This is not surprising, as previous studies have demonstrated that decidual cells express a wide range of chemokines, presumed to be responsible for inducing immune cell recruitment into the decidua and creating the unique decidual immunological environment (20, 39–41). Moreover, our in vitro studies demonstrated that untreated HUtMvECs and primary decidual endothelial cells express CCL14 and CXCL6 mRNA, with increased expression upon EVT-CM treatment. These findings were reproduced when immunostaining intensity of the chemokines was examined in the decidua basalis, with greater staining intensity detected in the endothelial cells of remodeling SpAs than in surrounding decidual stromal cells and endothelial cells in unremodeled SpAs. This reinforces the concept of a localized chemokine gradient in remodeling arteries that regulates focal SpA leukocyte infiltration.

Our findings using recombinant chemokine ligands and neutralizing Abs demonstrate that CCL14 and CXCL6 are endothelial chemoattractants for dNK cells in vitro. CCL14 is a known monocyte chemoattractant, whereas CXCL6 classically acts on neutrophils, acting through the receptors CXCR1 and CXCR2 (42). We identified expression of the cognate receptors (CXCR1, CXCR2, CCR1, and CCR5) for both chemokines by dNK cells. Previous studies indicated low mRNA expression of CCR1, CCR5, and CXCR1 by dNK cells (43); however, at the protein level CXCR1 was relatively abundant (41). Furthermore, coculture of PBNK cells with primary human decidual cells induces a dNK cell–like phenotype and this is accompanied by a significant increase in CCR1, CCR5, and CXCR1 protein expression (44). These results support the receptor expression patterns observed in our study and also suggest dNK cells express these receptors at higher levels than PBNK cells due to the local decidual microenvironment. CCL14 induced chemotaxis of decidual macrophages, consistent with its established actions in peripheral monocytes via CCR1 and CCR5 receptors (45). To our knowledge, expression of these receptors by decidual macrophages has not been reported previously.

The chemokine receptors studied (CCR1, CCR5, CXCR1, and CXCR2) have multiple ligands, including chemokines such as CCL3 and CXCL8 that have been previously reported in decidual stromal cells (23, 44) or vascular cells (CCL16) (18, 20, 46, 47). These chemokines, and others, may contribute to recruitment of immune cells to the decidual stroma; however, in the current study we found no support for their involvement in EVT-stimulated homing to remodeling SpAs because they were not identified in our unbiased gene array. As well as decidual leukocytes, EVTs also express the receptors for CCL14 and CXCL6 (46, 48), hence these chemokines may have further actions in promoting colonization and remodeling of SpAs by EVTs. It is important to note that the corresponding receptors are not specific to dNK cells or decidual macrophages; T cells express receptors for both chemokines. However, very few T cells have been observed in the remodeling artery wall, presumably because of their relative rarity in the decidua (9, 14).

EVTs plug the openings of SpAs into the intervillous space very early in pregnancy and also invade interstitially into the decidua. EVTs are a rich source of cytokines and growth factors, including IL-6, CXCL8, M-CSF, TGF-β, TNF and IL-13 (49–51). CXCL8 and IL-6 in EVT-CM can activate HUtMvECs to secrete CCL14 and CXCL6. This is consistent with our observation that IL-6 and CXCL8 are expressed by vEVTs in first trimester decidua and their receptors CXCR1, IL-6R, and gp130 by HUtMvECs. Although decidual cells express both these cytokines, EVTs from anchoring villi and vEVTs express higher levels of CXCL8. Moreover, their anatomical location within the SpA lumen places them in closer proximity to the endothelium, suggesting local interactions between vEVT-secreted cytokines and endothelial cells. Both cytokines have established actions on the vascular endothelium, signaling via Jak/stat pathways to activate inflammatory mediators, including NF-κB, adhesion molecules such as ICAM-1 and VCAM-1, angiogenic factors, and chemokines (including CXCL6) (52–56).

Studies of spontaneous miscarriage have identified reduced IL-6 and CXCL8 in EVTs and lesser expression of their receptors by decidual endothelial cells (57); this may impair EVT-endothelial cell crosstalk, leading to dysregulation of endothelial cell-derived chemokine production and reduced leukocyte recruitment. There is evidence for altered levels of cytokines, including CXCL8 and IL-6, and chemokines (CCL2, CCL4, CCL7, CXCL10, and CXCL12), in patients with preeclampsia (58–63). However, these measurements are restricted to maternal serum or third trimester decidua and were made after diagnosis of preeclampsia, and thus may reflect systemic endothelial dysfunction and inflammation associated with the disease state. Analyzing cytokine and chemokine expressions in the decidua of patients with a high risk of developing preeclampsia, such as those presenting with high resistance Doppler indices of uterine arteries in the first trimester (62), may help to identify whether defects in EVT-endothelial signaling contribute to impaired remodeling in vivo.

Other factors secreted by EVTs that could affect endothelial cell phenotype and behavior (64–67) were identified in our analysis, including MIF, MIC-1, CD147, IL-1α and CCL2, all potential candidates for contributing to EVT-SpA crosstalk. A low concentration of MIC-1 in maternal circulation is a strong predictor of miscarriage (68, 69). However, despite being present in high concentrations in EVT-CM, we did not observe any effect of MIC-1 on endothelial expression of the chemokines of interest.

A strength of this study is the use of primary EVT cultures, generated using the established EVT outgrowth model (25, 70), thus closely resembling physiological conditions in pregnancy. Similar studies of EVT-endothelial cell interactions have used EVT cell lines and have identified upregulation of CXCL10 in response to EVT-CM exposure (71). Although CXCL10 expression was identified in EVT-primed endothelial cells, the change in expression (PPLR = 0.55) did not meet our threshold for further analysis. This may be due to differences in endothelial cell lines used, or in the cytokine repertoire of primary EVTs compared with SGHPL-4 cells. HUtMvECs are derived from the uterus and have been widely used in similar studies (72). In addition, we were able to replicate the upregulation of CXCL6 and CCL14 expression by EVT-CM in primary decidual endothelial cells. Furthermore, immunostaining of the decidua basalis revealed CCL14 and CXCL6 on SpA endothelial cells, supporting our in vitro observations. We were unable to pair leukocytes and EVT outgrowths from the same samples because EVT outgrowth takes a total of 96 h, and it is not possible to culture decidual leukocytes for that long without affecting their phenotype and viability. EVT-CM and HUtMvEC-CM were cleared of cellular debris so the medium contained only soluble proteins. NK cells and macrophages were freshly isolated for the chemotaxis assay to minimize phenotypic drift. Migrated cells were in HUtMvEC-CM for a maximum of 16 h before collection and counting. Cytokine production reflecting NK cell/macrophage activation due to a possible allogeneic response was not measured as the study focused on migration only. The purity of the dNK cells and macrophage populations used for the chemotaxis assay was 80%, suggesting a strong enrichment for the specific immune cells of interest, but the presence of small numbers of other decidual cells. One further caution for data interpretation is the use of a standard 20% oxygen tension for EVT outgrowths. The maternofetal interface is physiologically hypoxic in the first trimester (73), but oxygen tension in the decidua basalis is unknown. As previously described (74), we found high rates of EVT outgrowth at 20% oxygen, allowing us to obtain conditioned medium in a 48-h time frame, prior to EVT-induced matrix digestion and loss of cell viability. Furthermore, we validated observations in first trimester decidua basalis tissue sections.

In summary, this study demonstrates a complex localized communication between EVTs, SpA endothelial cells, and decidual leukocytes that potentially regulates leukocyte recruitment to SpAs in early pregnancy and identifies key candidate chemokines and cytokines involved. Most studies have focused on EVT migration to the SpA wall and the consequences of poor EVT invasion. We suggest initiation of SpA remodeling depends on activation of innate immune cells by paracrine factors released from trophoblasts. The findings identify novel avenues for research and potential therapeutic interventions to treat inadequate SpA remodeling in pathological pregnancies.

Disclosures

The authors have no financial conflicts of interest.

Acknowledgments

We thank all the research nurses in St. Mary’s Hospital and Mount Sinai Hospital for efforts in obtaining consents and with collecting tissues. We also thank all the donors for consenting to participate and making this study possible. We would like to acknowledge Dr. Samantha D Smith, Dr. Leo Zeef and Bona Kim for assistance with experiments.

Footnotes

  • This work was supported by a grant from the Canadian Institutes for Health Research. The Maternal and Fetal Health Research Centre is supported by Tommy’s Fund, an Action Medical Research Endowment, and the Greater Manchester Clinical Research Network. L.K.H. was supported by the Biotechnology and Biological Sciences Research Council’s David Phillips Research Fellowship (BB/H022627/1).

  • The gene expression data presented in this article have been submitted to ArrayExpress (https://www.ebi.ac.uk/arrayexpress/) under accession number E-MTAB-5467.

  • The online version of this article contains supplemental material.

  • Abbreviations used in this article:

    AU
    arbitrary unit
    dNK
    decidual NK
    EVT
    extravillous trophoblast
    EVT-CM
    EVT conditioned medium
    HUtMvEC
    human uterine microvascular endothelial cell
    HUtMvEC-CM
    HUtMvEC conditioned medium
    MIC-1
    macrophage inhibitory cytokine-1
    PBNK
    peripheral blood NK
    PPLR
    probability of positive log ratio
    rh
    recombinant human
    α-SMA
    α-smooth muscle actin
    SpA
    spiral arteriole
    vEVT
    endovascular EVT.

  • Received July 5, 2016.
  • Accepted March 14, 2017.

References

  1. ↵
    1. Burton G. J.,
    2. A. W. Woods,
    3. E. Jauniaux,
    4. J. C. Kingdom
    . 2009. Rheological and physiological consequences of conversion of the maternal spiral arteries for uteroplacental blood flow during human pregnancy. Placenta 30: 473–482.
    OpenUrlCrossRefPubMed
  2. ↵
    1. Harris L. K.
    2010. Review: trophoblast-vascular cell interactions in early pregnancy: how to remodel a vessel. Placenta 31(Suppl.): S93–S98.
    OpenUrlCrossRefPubMed
  3. ↵
    1. Smith S. D.,
    2. R. H. Choudhury,
    3. P. Matos,
    4. J. A. Horn,
    5. S. J. Lye,
    6. C. E. Dunk,
    7. J. D. Aplin,
    8. R. L. Jones,
    9. L. K. Harris
    . 2016. Changes in vascular extracellular matrix composition during decidual spiral arteriole remodeling in early human pregnancy. Histol. Histopathol. 31: 557–571.
    OpenUrl
  4. ↵
    1. Brosens I. A.,
    2. W. B. Robertson,
    3. H. G. Dixon
    . 1972. The role of the spiral arteries in the pathogenesis of preeclampsia. Obstet. Gynecol. Annu. 1: 177–191.
    OpenUrlPubMed
    1. Lyall F.,
    2. S. C. Robson,
    3. J. N. Bulmer
    . 2013. Spiral artery remodeling and trophoblast invasion in preeclampsia and fetal growth restriction: relationship to clinical outcome. Hypertension 62: 1046–1054.
    OpenUrlCrossRef
  6. ↵
    1. Ball E.,
    2. J. N. Bulmer,
    3. S. Ayis,
    4. F. Lyall,
    5. S. C. Robson
    . 2006. Late sporadic miscarriage is associated with abnormalities in spiral artery transformation and trophoblast invasion. J. Pathol. 208: 535–542.
    OpenUrlCrossRefPubMed
  7. ↵
    1. Moffett A.,
    2. F. Colucci
    . 2015. Co-evolution of NK receptors and HLA ligands in humans is driven by reproduction. Immunol. Rev. 267: 283–297.
    OpenUrlCrossRefPubMed
  8. ↵
    1. Bulmer J. N.,
    2. L. Morrison,
    3. M. Longfellow,
    4. A. Ritson,
    5. D. Pace
    . 1991. Granulated lymphocytes in human endometrium: histochemical and immunohistochemical studies. Hum. Reprod. 6: 791–798.
    OpenUrlAbstract/FREE Full Text
  9. ↵
    1. Bartmann C.,
    2. S. E. Segerer,
    3. L. Rieger,
    4. M. Kapp,
    5. M. Sütterlin,
    6. U. Kämmerer
    . 2014. Quantification of the predominant immune cell populations in decidua throughout human pregnancy. Am. J. Reprod. Immunol. 71: 109–119.
    OpenUrlCrossRefPubMed
  10. ↵
    1. Wallace A. E.,
    2. R. Fraser,
    3. J. E. Cartwright
    . 2012. Extravillous trophoblast and decidual natural killer cells: a remodelling partnership. Hum. Reprod. Update 18: 458–471.
    OpenUrlAbstract/FREE Full Text
  11. ↵
    1. Harris L. K.,
    2. R. J. Keogh,
    3. M. Wareing,
    4. P. N. Baker,
    5. J. E. Cartwright,
    6. J. D. Aplin,
    7. G. S. Whitley
    . 2006. Invasive trophoblasts stimulate vascular smooth muscle cell apoptosis by a fas ligand-dependent mechanism. Am. J. Pathol. 169: 1863–1874.
    OpenUrlCrossRefPubMed
    1. Keogh R. J.,
    2. L. K. Harris,
    3. A. Freeman,
    4. P. N. Baker,
    5. J. D. Aplin,
    6. G. S. Whitley,
    7. J. E. Cartwright
    . 2007. Fetal-derived trophoblast use the apoptotic cytokine tumor necrosis factor-alpha-related apoptosis-inducing ligand to induce smooth muscle cell death. Circ. Res. 100: 834–841.
    OpenUrlAbstract/FREE Full Text
  13. ↵
    1. Smith S. D.,
    2. C. E. Dunk,
    3. J. D. Aplin,
    4. L. K. Harris,
    5. R. L. Jones
    . 2009. Evidence for immune cell involvement in decidual spiral arteriole remodeling in early human pregnancy. Am. J. Pathol. 174: 1959–1971.
    OpenUrlCrossRefPubMed
  14. ↵
    1. Lash G. E.,
    2. H. Pitman,
    3. H. L. Morgan,
    4. B. A. Innes,
    5. C. N. Agwu,
    6. J. N. Bulmer
    . 2016. Decidual macrophages: key regulators of vascular remodeling in human pregnancy. J. Leukoc. Biol. 100: 315–325.
    OpenUrlAbstract/FREE Full Text
  15. ↵
    1. Croy B. A.,
    2. A. A. Ashkar,
    3. R. A. Foster,
    4. J. P. DiSanto,
    5. J. Magram,
    6. D. Carson,
    7. S. J. Gendler,
    8. M. J. Grusby,
    9. N. Wagner,
    10. W. Muller,
    11. M. J. Guimond
    . 1997. Histological studies of gene-ablated mice support important functional roles for natural killer cells in the uterus during pregnancy. J. Reprod. Immunol. 35: 111–133.
    OpenUrlCrossRefPubMed
  16. ↵
    1. Schonkeren D.,
    2. M. L. van der Hoorn,
    3. P. Khedoe,
    4. G. Swings,
    5. E. van Beelen,
    6. F. Claas,
    7. C. van Kooten,
    8. E. de Heer,
    9. S. Scherjon
    . 2011. Differential distribution and phenotype of decidual macrophages in preeclamptic versus control pregnancies. Am. J. Pathol. 178: 709–717.
    OpenUrlCrossRefPubMed
  17. ↵
    1. Gustafsson C.,
    2. J. Mjosberg,
    3. A. Matussek,
    4. R. Geffers,
    5. L. Matthiesen,
    6. G. Berg,
    7. S. Sharma,
    8. J. Buer,
    9. J. Ernerudh
    . 2008. Gene expression profiling of human decidual macrophages: evidence for immunosuppressive phenotype. PLoS One 3(4): e2078.
    OpenUrlCrossRefPubMed
  18. ↵
    1. Hazan A. D.,
    2. S. D. Smith,
    3. R. L. Jones,
    4. W. Whittle,
    5. S. J. Lye,
    6. C. E. Dunk
    . 2010. Vascular-leukocyte interactions: mechanisms of human decidual spiral artery remodeling in vitro. Am. J. Pathol. 177: 1017–1030.
    OpenUrlCrossRefPubMed
  19. ↵
    1. Kam E. P.,
    2. L. Gardner,
    3. Y. W. Loke,
    4. A. King
    . 1999. The role of trophoblast in the physiological change in decidual spiral arteries. Hum. Reprod. 14: 2131–2138.
    OpenUrlAbstract/FREE Full Text
  20. ↵
    1. Jones R. L.,
    2. N. J. Hannan,
    3. T. J. Kaitu’u,
    4. J. Zhang,
    5. L. A. Salamonsen
    . 2004. Identification of chemokines important for leukocyte recruitment to the human endometrium at the times of embryo implantation and menstruation. J. Clin. Endocrinol. Metab. 89: 6155–6167.
    OpenUrlCrossRefPubMed
  21. ↵
    1. Hamilton S. A.,
    2. C. L. Tower,
    3. R. L. Jones
    . 2013. Identification of chemokines associated with the recruitment of decidual leukocytes in human labour: potential novel targets for preterm labour. PLoS One 8(2): e56946.
    OpenUrlCrossRefPubMed
  22. ↵
    1. Higgins L. E.,
    2. N. Rey de Castro,
    3. N. Addo,
    4. M. Wareing,
    5. S. Greenwood,
    6. R. L. Jones,
    7. C. P. Sibley,
    8. E. D. Johnstone,
    9. A. E. Heazell
    . 2015. Placental features of late-onset adverse pregnancy outcome. PLoS One 10(6): e0129117.
    OpenUrlCrossRefPubMed
  23. ↵
    1. Red-Horse K.,
    2. P. M. Drake,
    3. S. J. Fisher
    . 2004. Human pregnancy: the role of chemokine networks at the fetal-maternal interface. Expert Rev. Mol. Med. 6: 1–14.
    OpenUrlCrossRefPubMed
  24. ↵
    1. Ramhorst R.,
    2. E. Grasso,
    3. D. Paparini,
    4. V. Hauk,
    5. L. Gallino,
    6. G. Calo,
    7. D. Vota,
    8. C. Pérez Leirós
    . 2016. Decoding the chemokine network that links leukocytes with decidual cells and the trophoblast during early implantation. Cell Adhes. Migr. 10: 197–207.
    OpenUrl
  25. ↵
    1. Wright J. K.,
    2. C. E. Dunk,
    3. H. Amsalem,
    4. C. Maxwell,
    5. S. Keating,
    6. S. J. Lye
    . 2010. HER1 signaling mediates extravillous trophoblast differentiation in humans. Biol. Reprod. 83: 1036–1045.
    OpenUrlAbstract/FREE Full Text
  26. ↵
    1. Amsalem H.,
    2. M. Kwan,
    3. A. Hazan,
    4. J. Zhang,
    5. R. L. Jones,
    6. W. Whittle,
    7. J. C. Kingdom,
    8. B. A. Croy,
    9. S. J. Lye,
    10. C. E. Dunk
    . 2014. Identification of a novel neutrophil population: proangiogenic granulocytes in second-trimester human decidua. J. Immunol. 193: 3070–3079.
    OpenUrlAbstract/FREE Full Text
  27. ↵
    1. Lee Y. H.,
    2. O. Shynlova,
    3. S. J. Lye
    . 2015. Stretch-induced human myometrial cytokines enhance immune cell recruitment via endothelial activation. Cell. Mol. Immunol. 12: 231–242.
    OpenUrlCrossRefPubMed
  28. ↵
    1. Li C.,
    2. W. H. Wong
    . 2001. Model-based analysis of oligonucleotide arrays: expression index computation and outlier detection. Proc. Natl. Acad. Sci. USA 98: 31–36.
    OpenUrlAbstract/FREE Full Text
    1. Liu X.,
    2. M. Milo,
    3. N. D. Lawrence,
    4. M. Rattray
    . 2005. A tractable probabilistic model for Affymetrix probe-level analysis across multiple chips. Bioinformatics 21: 3637–3644.
    OpenUrlAbstract/FREE Full Text
  30. ↵
    1. Pearson R. D.,
    2. X. Liu,
    3. G. Sanguinetti,
    4. M. Milo,
    5. N. D. Lawrence,
    6. M. Rattray
    . 2009. puma: a Bioconductor package for propagating uncertainty in microarray analysis. BMC Bioinformatics 10: 211.
    OpenUrlCrossRefPubMed
  31. ↵
    1. Meller M.,
    2. S. Vadachkoria,
    3. D. A. Luthy,
    4. M. A. Williams
    . 2005. Evaluation of housekeeping genes in placental comparative expression studies. Placenta 26: 601–607.
    OpenUrlCrossRefPubMed
  32. ↵
    1. Zhang J.,
    2. C. E. Dunk,
    3. M. Kwan,
    4. R. L. Jones,
    5. L. K. Harris,
    6. S. Keating,
    7. S. J. Lye
    . 2015. Human dNK cell function is differentially regulated by extrinsic cellular engagement and intrinsic activating receptors in first and second trimester pregnancy. Cell. Mol. Immunol. 14: 203–213.
    OpenUrl
  33. ↵
    1. Ashton S. V.,
    2. G. S. Whitley,
    3. P. R. Dash,
    4. M. Wareing,
    5. I. P. Crocker,
    6. P. N. Baker,
    7. J. E. Cartwright
    . 2005. Uterine spiral artery remodeling involves endothelial apoptosis induced by extravillous trophoblasts through Fas/FasL interactions. Arterioscler. Thromb. Vasc. Biol. 25: 102–108.
    OpenUrlAbstract/FREE Full Text
  34. ↵
    1. Derricott H.,
    2. R. L. Jones,
    3. S. L. Greenwood,
    4. G. Batra,
    5. M. J. Evans,
    6. A. E. Heazell
    . 2016. Characterizing villitis of unknown etiology and inflammation in stillbirth. Am. J. Pathol. 186: 952–961.
    OpenUrl
  35. ↵
    1. Watson C.,
    2. S. Whittaker,
    3. N. Smith,
    4. A. J. Vora,
    5. D. C. Dumonde,
    6. K. A. Brown
    . 1996. IL-6 acts on endothelial cells to preferentially increase their adherence for lymphocytes. Clin. Exp. Immunol. 105: 112–119.
    OpenUrlCrossRefPubMed
  36. ↵
    1. Schraufstatter I. U.,
    2. J. Chung,
    3. M. Burger
    . 2001. IL-8 activates endothelial cell CXCR1 and CXCR2 through Rho and Rac signaling pathways. Am. J. Physiol. Lung Cell. Mol. Physiol. 280: L1094–L1103.
    OpenUrlAbstract/FREE Full Text
  37. ↵
    1. Harada A.,
    2. N. Sekido,
    3. T. Akahoshi,
    4. T. Wada,
    5. N. Mukaida,
    6. K. Matsushima
    . 1994. Essential involvement of interleukin-8 (IL-8) in acute inflammation. J. Leukoc. Biol. 56: 559–564.
    OpenUrlAbstract
  38. ↵
    1. Romano M.,
    2. M. Sironi,
    3. C. Toniatti,
    4. N. Polentarutti,
    5. P. Fruscella,
    6. P. Ghezzi,
    7. R. Faggioni,
    8. W. Luini,
    9. V. van Hinsbergh,
    10. S. Sozzani,
    11. et al
    . 1997. Role of IL-6 and its soluble receptor in induction of chemokines and leukocyte recruitment. Immunity 6: 315–325.
    OpenUrlCrossRefPubMed
  39. ↵
    1. Red-Horse K.,
    2. P. M. Drake,
    3. M. D. Gunn,
    4. S. J. Fisher
    . 2001. Chemokine ligand and receptor expression in the pregnant uterus: reciprocal patterns in complementary cell subsets suggest functional roles. Am. J. Pathol. 159: 2199–2213.
    OpenUrlCrossRefPubMed
    1. Bulmer J. N.,
    2. P. J. Williams,
    3. G. E. Lash
    . 2010. Immune cells in the placental bed. Int. J. Dev. Biol. 54: 281–294.
    OpenUrlCrossRefPubMed
  41. ↵
    1. Lockwood C. J.,
    2. S. J. Huang,
    3. C. P. Chen,
    4. Y. Huang,
    5. J. Xu,
    6. S. Faramarzi,
    7. O. Kayisli,
    8. U. Kayisli,
    9. L. Koopman,
    10. D. Smedts,
    11. et al
    . 2013. Decidual cell regulation of natural killer cell-recruiting chemokines: implications for the pathogenesis and prediction of preeclampsia. Am. J. Pathol. 183: 841–856.
    OpenUrlCrossRefPubMed
  42. ↵
    1. Gijsbers K.,
    2. M. Gouwy,
    3. S. Struyf,
    4. A. Wuyts,
    5. P. Proost,
    6. G. Opdenakker,
    7. F. Penninckx,
    8. N. Ectors,
    9. K. Geboes,
    10. J. Van Damme
    . 2005. GCP-2/CXCL6 synergizes with other endothelial cell-derived chemokines in neutrophil mobilization and is associated with angiogenesis in gastrointestinal tumors. Exp. Cell Res. 303: 331–342.
    OpenUrlCrossRefPubMed
  43. ↵
    1. Hanna J.,
    2. O. Wald,
    3. D. Goldman-Wohl,
    4. D. Prus,
    5. G. Markel,
    6. R. Gazit,
    7. G. Katz,
    8. R. Haimov-Kochman,
    9. N. Fujii,
    10. S. Yagel,
    11. et al
    . 2003. CXCL12 expression by invasive trophoblasts induces the specific migration of CD16- human natural killer cells. Blood 102: 1569–1577.
    OpenUrlAbstract/FREE Full Text
  44. ↵
    1. Carlino C.,
    2. H. Stabile,
    3. S. Morrone,
    4. R. Bulla,
    5. A. Soriani,
    6. C. Agostinis,
    7. F. Bossi,
    8. C. Mocci,
    9. F. Sarazani,
    10. F. Tedesco,
    11. et al
    . 2008. Recruitment of circulating NK cells through decidual tissues: a possible mechanism controlling NK cell accumulation in the uterus during early pregnancy. Blood 111: 3108–3115.
    OpenUrlAbstract/FREE Full Text
  45. ↵
    1. Katschke K. J., Jr..,
    2. J. B. Rottman,
    3. J. H. Ruth,
    4. S. Qin,
    5. L. Wu,
    6. G. LaRosa,
    7. P. Ponath,
    8. C. C. Park,
    9. R. M. Pope,
    10. A. E. Koch
    . 2001. Differential expression of chemokine receptors on peripheral blood, synovial fluid, and synovial tissue monocytes/macrophages in rheumatoid arthritis. Arthritis Rheum. 44: 1022–1032.
    OpenUrlCrossRefPubMed
  46. ↵
    1. Hannan N. J.,
    2. R. L. Jones,
    3. C. A. White,
    4. L. A. Salamonsen
    . 2006. The chemokines, CX3CL1, CCL14, and CCL4, promote human trophoblast migration at the feto-maternal interface. Biol. Reprod. 74: 896–904.
    OpenUrlAbstract/FREE Full Text
  47. ↵
    1. Berahovich R. D.,
    2. N. L. Lai,
    3. Z. Wei,
    4. L. L. Lanier,
    5. T. J. Schall
    . 2006. Evidence for NK cell subsets based on chemokine receptor expression. J. Immunol. 177: 7833–7840.
    OpenUrlAbstract/FREE Full Text
  48. ↵
    1. Zhang H.,
    2. L. Hou,
    3. C. M. Li,
    4. W. Y. Zhang
    . 2013. The chemokine CXCL6 restricts human trophoblast cell migration and invasion by suppressing MMP-2 activity in the first trimester. Hum. Reprod. 28: 2350–2362.
    OpenUrlAbstract/FREE Full Text
  49. ↵
    1. Naruse K.,
    2. B. A. Innes,
    3. J. N. Bulmer,
    4. S. C. Robson,
    5. R. F. Searle,
    6. G. E. Lash
    . 2010. Secretion of cytokines by villous cytotrophoblast and extravillous trophoblast in the first trimester of human pregnancy. J. Reprod. Immunol. 86: 148–150.
    OpenUrlCrossRefPubMed
    1. Tabibzadeh S.,
    2. Q. F. Kong,
    3. A. Babaknia,
    4. L. T. May
    . 1995. Progressive rise in the expression of interleukin-6 in human endometrium during menstrual cycle is initiated during the implantation window. Hum. Reprod. 10: 2793–2799.
    OpenUrlAbstract/FREE Full Text
  51. ↵
    1. Champion H.,
    2. B. A. Innes,
    3. S. C. Robson,
    4. G. E. Lash,
    5. J. N. Bulmer
    . 2012. Effects of interleukin-6 on extravillous trophoblast invasion in early human pregnancy. Mol. Hum. Reprod. 18: 391–400.
    OpenUrlAbstract/FREE Full Text
  52. ↵
    1. Waugh D. J. J.,
    2. C. Wilson
    . 2008. The interleukin-8 pathway in cancer. Clin. Cancer Res. 14: 6735–6741.
    OpenUrlAbstract/FREE Full Text
    1. Tsou P. S.,
    2. B. J. Rabquer,
    3. B. Balogh,
    4. A. Kendzicky,
    5. B. Kahaleh,
    6. E. Schiopu,
    7. D. Khanna,
    8. A. E. Koch
    . 2012. Primary human scleroderma dermal endothelial cells exhibit defective angiogenesis. Arthritis Rheum. 64(10): S645.
    OpenUrl
    1. Ha J.,
    2. H. S. Choi,
    3. Y. Lee,
    4. H. J. Kwon,
    5. Y. W. Song,
    6. H. H. Kim
    . 2010. CXC chemokine ligand 2 induced by receptor activator of NF-kappa B ligand enhances osteoclastogenesis. J. Immunol. 184: 4717–4724.
    OpenUrlAbstract/FREE Full Text
    1. Sansone P.,
    2. J. Bromberg
    . 2012. Targeting the interleukin-6/Jak/stat pathway in human malignancies. J. Clin. Oncol. 30: 1005–1014.
    OpenUrlAbstract/FREE Full Text
  56. ↵
    1. Zhang S.,
    2. R. Hwaiz,
    3. L. Luo,
    4. H. Herwald,
    5. H. Thorlacius
    . 2015. STAT3-dependent CXC chemokine formation and neutrophil migration in streptococcal M1 protein-induced acute lung inflammation. Am. J. Physiol. Lung Cell. Mol. Physiol. 308: L1159–L1167.
    OpenUrlAbstract/FREE Full Text
  57. ↵
    1. Pitman H.,
    2. B. A. Innes,
    3. S. C. Robson,
    4. J. N. Bulmer,
    5. G. E. Lash
    . 2013. Altered expression of interleukin-6, interleukin-8 and their receptors in decidua of women with sporadic miscarriage. Hum. Reprod. 28: 2075–2086.
    OpenUrlAbstract/FREE Full Text
  58. ↵
    1. Tosun M.,
    2. H. Celik,
    3. B. Avci,
    4. E. Yavuz,
    5. T. Alper,
    6. E. Malatyalioğlu
    . 2010. Maternal and umbilical serum levels of interleukin-6, interleukin-8, and tumor necrosis factor-alpha in normal pregnancies and in pregnancies complicated by preeclampsia. J. Matern. Fetal Neonatal Med. 23: 880–886.
    OpenUrlCrossRefPubMed
    1. Szarka A.,
    2. J. Rigo Jr..,
    3. L. Lazar,
    4. G. Beko,
    5. A. Molvarec
    . 2010. Circulating cytokines, chemokines and adhesion molecules in normal pregnancy and preeclampsia determined by multiplex suspension array. BMC Immunol. 11: 59.
    OpenUrlCrossRefPubMed
    1. Liu X.,
    2. L. I. Dai,
    3. R. Zhou
    . 2015. Association between preeclampsia and the CXC chemokine family (Review). Exp. Ther. Med. 9: 1572–1576.
    OpenUrl
    1. Schanz A.,
    2. V. D. Winn,
    3. S. J. Fisher,
    4. M. Blumenstein,
    5. C. Heiss,
    6. A. P. Hess,
    7. J. S. Kruessel,
    8. M. Mcmaster,
    9. R. A. North
    . 2011. Pre-eclampsia is associated with elevated CXCL12 levels in placental syncytiotrophoblasts and maternal blood. Eur. J. Obstet. Gynecol. Reprod. Biol. 157(1): 32–37.
    OpenUrlCrossRefPubMed
  62. ↵
    1. Fraser R.,
    2. G. S. Whitley,
    3. A. P. Johnstone,
    4. A. J. Host,
    5. N. J. Sebire,
    6. B. Thilaganathan,
    7. J. E. Cartwright
    . 2012. Impaired decidual natural killer cell regulation of vascular remodelling in early human pregnancies with high uterine artery resistance. J. Pathol. 228: 322–332.
    OpenUrlCrossRefPubMed
  63. ↵
    1. Boij R.,
    2. J. Svensson,
    3. K. Nilsson-Ekdahl,
    4. K. Sandholm,
    5. T. L. Lindahl,
    6. E. Palonek,
    7. M. Garle,
    8. G. Berg,
    9. J. Ernerudh,
    10. M. Jenmalm,
    11. L. Matthiesen
    . 2012. Biomarkers of coagulation, inflammation, and angiogenesis are independently associated with preeclampsia. Am. J. Reprod. Immunol. 68: 258–270.
    OpenUrlCrossRefPubMed
  64. ↵
    1. Cheng Q.,
    2. S. J. McKeown,
    3. L. Santos,
    4. F. S. Santiago,
    5. L. M. Khachigian,
    6. E. F. Morand,
    7. M. J. Hickey
    . 2010. Macrophage migration inhibitory factor increases leukocyte-endothelial interactions in human endothelial cells via promotion of expression of adhesion molecules. J. Immunol. 185: 1238–1247.
    OpenUrlAbstract/FREE Full Text
    1. Voigt H.,
    2. C. S. Vetter-Kauczok,
    3. D. Schrama,
    4. U. B. Hofmann,
    5. J. C. Becker,
    6. R. Houben
    . 2009. CD147 impacts angiogenesis and metastasis formation. Cancer Invest. 27: 329–333.
    OpenUrlCrossRefPubMed
    1. Stamatovic S. M.,
    2. R. F. Keep,
    3. M. Mostarica-Stojkovic,
    4. A. V. Andjelkovic
    . 2006. CCL2 regulates angiogenesis via activation of Ets-1 transcription factor. J. Immunol. 177: 2651–2661.
    OpenUrlAbstract/FREE Full Text
  67. ↵
    1. Fan F.,
    2. O. Stoeltzing,
    3. W. Liu,
    4. M. F. McCarty,
    5. Y. D. Jung,
    6. N. Reinmuth,
    7. L. M. Ellis
    . 2004. Interleukin-1beta regulates angiopoietin-1 expression in human endothelial cells. Cancer Res. 64: 3186–3190.
    OpenUrlAbstract/FREE Full Text
  68. ↵
    1. Jin Y. J.,
    2. J. H. Lee,
    3. Y. M. Kim,
    4. G. T. Oh,
    5. H. Lee
    . 2012. Macrophage inhibitory cytokine-1 stimulates proliferation of human umbilical vein endothelial cells by up-regulating cyclins D1 and E through the PI3K/Akt-, ERK-, and JNK-dependent AP-1 and E2F activation signaling pathways. Cell. Signal. 24: 1485–1495.
    OpenUrlCrossRefPubMed
  69. ↵
    1. Tong S.,
    2. B. Marjono,
    3. D. A. Brown,
    4. S. Mulvey,
    5. S. N. Breit,
    6. U. Manuelpillai,
    7. E. M. Wallace
    . 2004. Serum concentrations of macrophage inhibitory cytokine 1 (MIC 1) as a predictor of miscarriage. Lancet 363: 129–130.
    OpenUrlCrossRefPubMed
  70. ↵
    1. van Dijk M.,
    2. J. van Bezu,
    3. D. van Abel,
    4. C. Dunk,
    5. M. A. Blankenstein,
    6. C. B. M. Oudejans,
    7. S. J. Lye
    . 2010. The STOX1 genotype associated with pre-eclampsia leads to a reduction of trophoblast invasion by alpha-T-catenin upregulation. Hum. Mol. Genet. 19: 2658–2667.
    OpenUrlAbstract/FREE Full Text
  71. ↵
    1. Wallace A. E.,
    2. J. E. Cartwright,
    3. R. Begum,
    4. K. Laing,
    5. B. Thilaganathan,
    6. G. S. Whitley
    . 2013. Trophoblast-induced changes in C-x-C motif chemokine 10 expression contribute to vascular smooth muscle cell dedifferentiation during spiral artery remodeling. Arterioscler. Thromb. Vasc. Biol. 33: e93–e101.
    OpenUrlAbstract/FREE Full Text
  72. ↵
    1. Kitaya K.,
    2. T. Yasuo,
    3. T. Yamaguchi,
    4. S. Fushiki,
    5. H. Honjo
    . 2007. Genes regulated by interferon-gamma in human uterine microvascular endothelial cells. Int. J. Mol. Med. 20: 689–697.
    OpenUrlPubMed
  73. ↵
    1. Chen Q.,
    2. X. L. Liversidge,
    3. B. Liu,
    4. P. Stone,
    5. L. W. Chamley
    . 2011. Does oxygen concentration affect shedding of trophoblastic debris or production of inflammatory mediators from first trimester human placenta? Placenta 32: 362–366.
    OpenUrlCrossRefPubMed
  74. ↵
    1. Aplin J. D.,
    2. T. Haigh,
    3. C. J. Jones,
    4. H. J. Church,
    5. L. Vićovac
    . 1999. Development of cytotrophoblast columns from explanted first-trimester human placental villi: role of fibronectin and integrin alpha5beta1. Biol. Reprod. 60: 828–838.
    OpenUrlAbstract/FREE Full Text
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section