Placenta 35 (2014) 1089e1094

Contents lists available at ScienceDirect

Placenta journal homepage: www.elsevier.com/locate/placenta

Oxygen regulates human cytotrophoblast migration by controlling chemokine and receptor expression A. Schanz a, b, *, K. Red-Horse a, d, A.P. Hess b, D.M. Baston-Büst b, C. Heiss c, J.S. Krüssel b a

Department of Cell and Tissue Biology, University of California (UCSF), San Francisco, 513 Parnassus Ave, CA 94143, USA University Düsseldorf, Medical Faculty, Department of Obstetrics, Gynecology and REI (UniKiD), Moorenstrasse 5, 40225 Düsseldorf, Germany c University Düsseldorf, Medical Faculty, Division of Cardiology, Pulmonology and Vascular Medicine, Moorenstrasse 5, 40225 Düsseldorf, Germany d Department of Biology, School of Medicine, Stanford University, Stanford, CA 94305, USA b

a r t i c l e i n f o

a b s t r a c t

Article history: Accepted 17 September 2014

Introduction: Placental development involves the variation of oxygen supply due to vascular changes and cytotrophoblast invasion. Chemokines and their receptors play an important role during placental formation. Herein, the analysis of the chemokine/receptor pair CXCL12/CXCR4 and further chemokine receptors, such as CCR1, CCR7 and CXCR6 expression in human cytotrophoblasts was conducted. Methods: Human cytotrophoblasts were examined directly after isolation or after incubation with different oxygen tensions and a chemical HIF-stimulator for 12 h with realtime PCR, immunoblot, immunohistochemistry. Conditioned media of placental villi, decidua, and endothelial cells was used for ELISA analysis of CXL12. Cytotrophoblast migration assays were conducted applying conditioned media of endothelial cells, a CXCL12 gradient, and different oxygen level. Endometrial and decidual tissue was stained for CXCL12 expression. Results: An upregulation of CXCL12, CXCR4, CCR1, CCR7 and CXCR6 was observed after cytotrophoblast differentiation. Low oxygen supply upregulated CXCR4, CCR7 and CXCR6, but downregulated CXCL12 and CCR1. In contrast to the HIF associated upregulation of the aforementioned proteins, downregulation of CXCL12 and CCR1 seemed to be HIF independent. Cytotrophoblast migration was stimulated by low oxygen, the application of a CXCL12 gradient and endothelial cell conditioned media. CXCL12 was detected in endometrial vessels, glands and conditioned media of placental and decidual tissue, but not decidual vessels. Discussion/conclusion: Taken together, oxygen supply and cytotrophoblast differentiation seem to be regulators of chemokine and receptor expression and function in human cytotrophoblasts. Therefore, this system seems to be involved in placental development, directed cytotrophoblast migration in the decidual compartment and a subsequent sufficient supply of the growing fetus. © 2014 Elsevier Ltd. All rights reserved.

Keywords: Chemokine Chemokine receptor Hypoxia Placenta Cytotrophoblast

1. Introduction Human placentation requires trophoblast proliferation, migration, differentiation, and invasion. In the beginning of pregnancy, trophoblast cells ‘plug’ uterine arteries and minimize the maternal

Abbreviations: CM, conditioned media; CM EC, conditioned media of endothelial cells; DFX, desferrioxamine; GFM, growth factor media; HIF, hypoxia inducible factor; O2, oxygen; SFM, serum free media. * Corresponding author. University Düsseldorf, Medical Faculty, Department of Obstetrics, Gynecology and REI (UniKiD), Moorenstrasse 5, 40225 Düsseldorf, Germany. Tel.: þ49 211 8104060; fax: þ49 211 8116787. E-mail addresses: [email protected] (A. Schanz), [email protected] (K. Red-Horse), [email protected] (A.P. Hess), [email protected] (D.M. BastonBüst), [email protected] (C. Heiss), [email protected] (J.S. Krüssel). http://dx.doi.org/10.1016/j.placenta.2014.09.012 0143-4004/© 2014 Elsevier Ltd. All rights reserved.

blood flow to the placenta, which leads to a physiologically low oxygen environment [1].The partial pressure of oxygen in the intervillous space and within the endometrium at this time is estimated to be as low as 18 and 40 mmHg [2,3], compared to the exposure of the placenta to the maternal blood which exhibits 90e100 mmHg at 10e12 weeks of gestation [4]. The differentiation process involves fusion to the syncytiotrophoblast, but also development into invasive, extravillous cytotrophoblasts which conquer the maternal vessels and lead the maternal blood into the intervillous space. Insufficient trophoblast invasion and remodeling of the uterine arteries in the first trimester can lead to increased blood pressure with high velocity of the blood flow, which can damage the placental architecture and subsequently be followed by ischemia due to impaired placental perfusion and disturbed pregnancy development [5,6].

1090

A. Schanz et al. / Placenta 35 (2014) 1089e1094

In addition, to the regulation of placental growth according to oxygen levels [7], chemokines and chemokine receptors can regulate human placentation. Human cytotrophoblasts express the chemokine receptors CXCR4, CXCR6, CCR1, CCR5, CCR7 and the ligands CXCL12, HCC-1, and CXCL16 [8e11]. CXCL12, CXCL16, and CCL21 stimulate cytotrophoblast migration in vitro [12]. When considering a link between chemokine and chemokine receptor expression with oxygen dependent signaling, it is important that chemokines/receptors, such as CXCL12 and CXCR4 are target genes of the hypoxia inducible factors (HIFs) [13]. It is known that HIFs mediate response to hypoxia [14] and are expressed in the placenta with expression peaking at 7e10 weeks of gestation [15,16], also showing dynamic modulation between the 9e12 week of gestation when the maternal blood flow is slowly establishing a circulatory network [17]. CXCR4 transcription is increased by low oxygen in several different cell types (e.g. monocytes, endothelial, or cancer cells) [18,19], which is paralleled by elevated chemotactic responsiveness to its ligand, CXCL12 [13]. Further, CXCR6 and CCR7 are also known to be involved in breast- and lung cancer cell migration and seem to experience an oxygen and HIF-dependent upregulation [20,21]. In contrast, CCR1 is scarcely expressed by cytotrophoblasts cultured in low oxygen [8]. In the present study, we aimed at investigating the effect of cell differentiation and oxygen tension on the chemokine and receptor pair CXCL12 and CXCR4 in human cytotrophoblasts. Further, the investigation of additional chemokine receptors, such as CXCR6, CCR1, and CCR7 were included to show an overall regulation pattern, which might be of importance for placenta development. In addition, the source of CXCL12 at the maternal fetal interface was analyzed suggesting a CXCL12 gradient, which might influence the chemotactic behavior of cytotrophoblasts, which is further modulated by chemokine receptor expression and the existing oxygen supply. We hypothesize that the regulation of chemokine and receptor expression due to changes in oxygen supply at the maternalefetal interface and cytotrophoblast differentiation to the invasive, extravillous phenotype modulate their chemotactic responses and that specific CXCL12 expression as chemoattractant are implemented during this process. These regulatory events might be of great importance for placental development and growth. 2. Material and methods 2.1. Human tissue collection Tissue collection was approved by the University's ethical board and the patients gave informed consent. Placentas from elective terminations of pregnancy (5e12 weeks) were prepared for further cytotrophoblast isolation and placenta villous explant culture. Additionally, these tissues (villi and decidua) and also endometrial samples throughout the menstrual cycle were fixed and embedded in paraffin [22]. Endometrial samples were obtained from regularly cycling women undergoing hysterectomy or endometrial biopsy for benign conditions. The samples were correlated to the proliferative (n ¼ 3) or secretory phase (n ¼ 3) (6 different women) evaluated by the cycle day and by a pathologist. 2.2. Cytotrophoblast isolation and culture Cytotrophoblasts were isolated from the placentas (n ¼ 6) as published before [23,24]. Placental villi were subjected to a series of enzymatic digests, which detached cytotrophoblasts progenitors from the villi cores. Afterwards purification with a Percoll gradient was performed. Cytotrophoblasts were cultured in serum free medium (SFM; Dulbecco's modified Eagle's medium, 4.5 g/l glucose (Sigma, St. Louis, MO, US) with 2% Nutridoma (Roche, Indianapolis, IN, US), 100 mg/ml penicillin/ streptomycin, 1% sodium pyruvate, 1% HEPES and 1% gentamicin, UCSF Cell Culture Facility). Except for the chemotaxis and migration assays, remaining leukocytes were removed by using CD45 coupled to magnetic beads (Dynal beads, Invitrogen, Grand Island, NY, US) [25]. Cytotrophoblasts were maintained under standard tissue culture conditions (20%O2/5%CO2/95%air) or placed in a Bactron anaerobic incubator (Sheldon Manufacturing Inc., Cornelius, US) with a 2%O2/93%N2/5%CO2 environment for 12 h.

Desferrioxamine (DFX, Sigma) at a concentration of 160 mM was used as a chemical HIF stimulator for 12 h [26]. For the differentiation assay, which provides information about the expression patterns of the invasive, extravillous phenotype, cytotrophoblasts were collected for protein analysis immediately after isolation and after 12 h of culture on Matrigel® (BD biosciences, Sparks, MD, US) in standard culture conditions [27]. For the analysis of the oxygen influence cells were collected for RNA and protein analysis after 12 h of culture in standard conditions (control value) or low oxygen condition or after treatment with DFX (n ¼ 6). 2.3. Immunoblotting The cells were lysed in a modified RIPA buffer (1% DOC, 0.1% SDS, 1% NP-40, 150 mM NaCl, 10 mM Tris/HCl, pH 7.4, 1 mM EDTA, 10% glycerol) containing protease inhibitors (Pierce, Rockford, IL USA). Protein concentration was measured using DC Protein assay kit (BIO-RAD, Hercules, CA, US). Equal amounts of proteins were separated on 10% TriseHCl gels and transferred to nitrocellulose (BIO-RAD). The membrane was blocked with 5% BSA/PBS-Tween for 15 min and probed using the following antibodies: polyclonal rabbit anti-human CXCR4 (1 mg/ml; Abcam), polyclonal goat anti-human CXCL12 (0.1 mg/ml), monoclonal mouse anti-human, CCR7 (1 mg/ml), CXCR6 (1 mg/ml) (R&D Systems), polyclonal rabbit anti-human CCR1 (1 mg/ml; Abcam), mouse monoclonal anti-human HIF-2a (1 mg/ml; Novus Biologicals, Littleton, CO, US), rabbit polyclonal anti human HIF-1a (0.1 mg/ml; Bethyl Laboratories, Montgomery, TX, US) and monoclonal mouse-anti-human actin (0.1 mg/ml; Sigma). Lysates of a Jurkat cell line (ATCC, Manassas, VA, US) and HUVECs (ATCC) were the positive control. Non-specific mouse or rabbit IgG was used as negative control. Enhanced chemiluminescence (ECL) detection reagents and Hyperfilm ECL (Amersham Biosciences Pittsburgh, PA, US) were used for visualization. 2.4. Reverse transcription-PCR/real-time PCR RNA was extracted after lyzing cytotrophoblasts with Trizol®. For Reverse Transcription-PCR, AMV reverse transcriptase (Invitrogen) was used to produce cDNA. RTmastermix contained RT buffer, dNTPs (each 100 mM), RT-random primer and DEPC-treated dH2O ad 18 mL (High capacity cDNA Archive Kit 432217, Applied Biosystems, Foster City, CA, USA). CDNA corresponding to 50 ng total RNA was used as a template in the PCR reaction. This consisted of ABI MasterMix (Applied Biosystems, Carlsbad, CA, USA) and pre-designed CXCR4, CXCL12, CXCR6, CCR1 and CCR7 TaqMan® gene expression systems which included the primers (Applied Biosystems). ABI Prism 7900HT real-time quantitative PCR instrument detected the accumulation of the PCR product. The endogenous control was eukaryotic 18S rRNA (Applied Biosystems). The condition for the PCR reaction was chosen following the instructions (http://tools. invitrogen.com/content/sfs/manuals/cms_041280.pdf, page 18). 2.5. In vitro organ culture of placental anchoring villus explants Organ cultures of placental anchoring villi (n ¼ 4) were set up as previously described [28]. First trimester anchoring villi were obtained by microdissection and cut into pieces no thicker then 5 mm to ensure even distribution of the oxygen levels in the tissue [29]. The villi were then placed on a Matrigel (BD Biosciences, San Jose, CA, US) coated 12 mm Millicell-CM culture dish inserts (Millipore, Billerica, MA, US). These were cultured for 48 h in Ham's F-12/Dulbecco's modified Eagle's medium (1:1, vol/vol) containing antibiotics/antimycotics and 10% FCS (UCSF Cell Culture Facility). For another 48 h the explants were maintained under standard or low oxygen conditions, then fixed in 4% paraformaldehyde, washed in PBS, infiltrated with increasing concentrations of sucrose (5e15%) followed by OCT compound (Miles Scientific, Princeton, MN, US) and frozen. 2.6. Immunohistochemistry Placenta villus explants (n ¼ 4) were triple stained with CXCR4 or CXCL12 (both 10 mg/ml) (R&D Systems, Minneapolis, MN, US) primary antibody, a rat anti-human cytokeratin-7 (CK7) antibody (1:100; 7D3, UCSF) and nuclear stain (Vectashield mounting medium with DAPI, Vector Laboratories, Burlingname, CA, US). Secondary antibodies were FITC- and Rhodamine-conjugated (Jackson Immuno Research, West Grove, PA, US). As a negative control, sections were incubated with non-immune serum. Staining was detected with constant exposure times with a Leica CTR5000 upright microscope and a DFC480 color camera. Decidual (n ¼ 6) and endometrial (n ¼ 6) paraffin sections (both from 6 different women) were processed using the manufacturers protocol (Vector) and CXCL12 (R&D Systems), vWF and CK7 (Dako, Clostrup, Denmark) antibodies. 2.7. Conditioned media of first trimester placental villi, decidua and endothelial cells Placental villi and decidua from the same placenta (n ¼ 4) were dissected in 2e3 mm3 big pieces, weighed (wet weight) and cultured in SFM (1 ml/g). After 24 h the conditioned media (CM) was collected and cell debris removed by centrifugation. CM of uterine microvascular endothelial cells (CM EC; n ¼ 3) (Cambrex, East Rutherford, NJ, US, now the company is LONZA) was generated by incubating 300,000 cells/ml in endothelial cell growth medium (EGM-2; Cambrex; growth factor medium, GFM) containing fetal calf serum, human epidermal growth factor,

A. Schanz et al. / Placenta 35 (2014) 1089e1094 vascular endothelial growth factor (VEGF), human fibroblast growth factor-B, and insulin-like growth factor-1 for 24 h. 2.8. Enzyme-linked immunosorbent assay CXCL12 was quantified via ELISA (Quantikine human CXCL12 kit (R&D Systems) in CM of the placenta villi, decidua, and CM EC. The optical density was determined by a microplate reader, mQuant (Bio-Tek Instruments, UK), set to 450 nm and a wavelength correction set to 540 nm. 2.9. Chemotaxis/migration assay The bottom of transwell inserts (8 mm pore size, 6.5 mm diameter, Corning, Union City, CA, US) of a modified Boyden-chamber assay were incubated with 10 mg/ ml human plasma fibronectin (Invitrogen) and washed with PBS. Cytotrophoblasts (n ¼ 6, 100,000 cells in 100 ml SFM) were added to the upper compartments. The lower compartment contained 500 ml medium with either vehicle alone (PBS/0.1% BSA), CXCL12 (1 mg/ml) (R&D Systems), or CM EC. CXCL12 blocking antibody (10 mg/ml; R&D Systems) was added to the lower chamber to neutralize bioactivity. Concentrations were chosen following a publication [12]. GFM was utilized as positive control. The cells were incubated for 12 h in standard (control value) or low oxygen conditions, washed in PBS and fixed in 3% paraformaldehyde. Cytotrophoblasts were immunostained for CK7 on the bottom of the filter and quantified in three randomly chosen fields using the Openlab software (Improvision, Waltham, US) region-of-interest tool. 2.10. Statistical analyses All statistical analyses, mean values, standard deviation, and ANOVA (non significant (ns) p > 0.05, significant (s)*p < 0.05, s**p < 0.001) were performed applying Excel software (Microsoft, USA) and SPSS (IBM, Statistics 20, Germany). The control values in the densitometric, RNA and migration analysis were set at 100% and results after treatment were expressed as a percentage of control values.

3. Results 3.1. Chemokine and receptor regulation by cell differentiation and different oxygen tension in human cytotrophoblasts The culture in 20% oxygen over 12 h on a thin layer of Matrigel® supports in vitro differentiation of cytotrophoblasts into the invasive, extravillous phenotype and was analyzed on the protein level (n ¼ 3) (Fig. 1A) [27]. Expression of CXCL12, CXCR4, CXCR6, CCR7, HIF-1a and HIF-2a increased in the invasive, extravillous cytotrophoblast during the differentiation process (Fig. 1A). Protein and mRNA levels of the above mentioned molecules were determined in first trimester cytotrophoblasts after 12 h in culture in different oxygen conditions and after DFX stimulation (n ¼ 6) (Fig. 1BeD). 2% O2 and DFX stimulation significantly increased CXCR4, CXCR6, and CCR7. In contrast, low oxygen decreased CXCL12 and CCR1 protein and mRNA levels. DFX only downregulated mRNA (Fig. 1D), but not the protein expression of CXCL12 and CCR1 (Fig. 1B,C; n ¼ 6). RNA data are summarized in Fig. 1D.

1091

The HIF-1a and HIF-2a increase [30] and the CCR1 decrease [8] have been described before and served as an internal control (Fig. 1BeD). Verifying the immunoblot results for receptor and chemokine pair CXCR4 and CXCL12 immunostaining in villous explant cultures were conducted. The tissue samples were maintained in standard (20% oxygen) or low (2%) oxygen conditions for 48 h (Fig. 2AeL). In agreement with the PCR and immunoblot analysis, the intensity of CXCR4 staining increased in hypoxic explant cultures, which involved progenitor and invasive, extravillous cytotrophoblasts as well as the syncytiotrophoblasts (Fig. 2J). CXCL12 levels also recapitulated the PCR and immunoblot analysis showing a decrease in the staining detected in progenitor and invasive, extravillous cytotrophoblasts in low oxygen condition (Fig. 2D). 3.2. CXCL12 localization in the uterine wall To identify whether maternal tissues were a source of CXCL12, immunolocalization on adjacent endometrium (proliferative phase n ¼ 3, secretory phase n ¼ 3) and first trimester decidua (n ¼ 6) samples was performed. In the endometrium, CXCL12 was detected and localized to the glandular epithelium as well as endothelial cells showing no changes of localization and intensity during the cycle (Fig. 3A, proliferative endometrium). In the decidua, glandular epithelial stained positive and low levels were detected in the stromal compartment (decidual cells). Vascular staining was absent (Fig. 3B, 6 weeks of gestation). ELISA showed significantly higher levels of CXCL12 protein secretion in the CM of decidual than the CM of uterine endothelial cells or placenta villous tissues (Fig. 3C). 3.3. Low oxygen levels enhance CXCL12-stimulated cytotrophoblast migration Migration towards CXCL12 increased up to 30% in standard culture condition in comparison to the control condition with SFM (set as 100%). In 2% oxygen a 50% increase of undirected cytotrophoblast migration, and an up to 70% increase with the CXCL12 stimulus was observed which was reversed by the addition of a CXCL12 blocking antibody (Fig. 4A). GFM was employed as a positive control and displayed a 50% increase compared to the SFM, which was not affected by a CXCL12-blocking antibody. 3.4. Endothelial derived molecules enhance cytotrophoblast migration A physiological source of chemotactic molecules, CM EC, stimulated a significant increase of cytotrophoblast migration up to

Fig. 1. The effect of cytotrophoblast differentiation and exposure to different oxygen levels on CXCR4, CXCL12, CCR1, CXCR6, CCR7, Hif-2a and HIF-1a protein and RNA expression. (A) Immunoblot analyses of human first trimester cytotrophoblast lysates immediately after isolation (0 h), after 12 h of differentiation (n ¼ 3), (B) after 12 h in standard culture on Matrigel® (20% oxygen, control value set as 100%), low oxygen (2%) condition and coincubation with DFX (20% oxygen) for CXCL12, CXCR4, CCR1, CXCR6, CCR7, HIF-2a and HIF-1a (n ¼ 6). Actin expression was used as a loading control. (C) Densitometric analysis of the immunoblots in different oxygen conditions and DFX treatment (s* ¼ p < 0.05). (D) RNA analyses of human first trimester cytotrophoblast after 12 h in standard culture (20% oxygen, control value set as 100%), low oxygen (2%) condition and coincubation with DFX (20% oxygen) for CXCL12, CXCR4, CCR1, CXCR6, CCR7, HIF-2a and HIF-1a (n ¼ 6).

1092

A. Schanz et al. / Placenta 35 (2014) 1089e1094

Fig. 2. The effect of low oxygen level on CXCR4 and CXCL12 expression in human first trimester explant villus cultures in vitro. Placental explants (n ¼ 4) were cultured either in standard culture conditions (20% oxygen) (A, B, C, G, H, I) or in low oxygen (2%) (D, E, F, J, K, L) for 48 h, followed by triple immunofluorescence for CXCR4 or CXCL12, respectively (red) and CK7 as a cytotrophoblast marker (green). DAPI was included as a nuclear stain (blue). The arrow in (A) shows the direction of cytotrophoblast invasion. Scale bar: 50 mm.

140% compared to only GFM (control condition set as 100%, Fig. 4B, n ¼ 3). This effect was significantly decreased, but not completely blocked, by the presence of a CXCL12 blocking antibody, showing a specific CXCL12 secretion of the endothelial cells. This result was

Fig. 3. CXCL12 expression in human endometrium, decidua and conditioned media. (A) Immunohistochemistry for CXCL12, CK7 and vWF in human endometrium (n ¼ 6) in the proliferative phase and in (B) human first trimester decidua (n ¼ 6) of a 6 week pregnancy. Scale bar: 100 mm (C) ELISA for CXCL12 in conditioned media (CM) of uterine endothelial cells, first trimester decidua and placental villi from the same donor (s ¼ **p < 0.001, ns ¼ p > 0,05). Error bars show standard deviation.

confirmed via ELISA for CXCL12 in CM EC (Fig. 3C). The blocking antibody in the GFM alone did not affect migration. 4. Discussion In the present study, in vitro cytotrophoblast differentiation into the invasive, extravillous phenotype lead to an upregulation of the chemokine/receptor pair, CXCL12 and CXCR4, but also further receptors as CXCR6, CCR7 and CCR1. Low oxygen stimulated an increase of CXCR4, CXCR6, and CCR7, but decrease of CXCL12 and CCR1. Functionally, low oxygen-induced CXCR4 expression led to increased cytotrophoblast chemotaxis towards CXCL12. CXCR4 and CXCL12 were analyzed more extensively since oxygen and HIF dependent upregulation of which has been reported in other cell types such as endothelial or cancer cells [13,31]. Herein, the results showed also a CXCR4 upregulation in cytotrophoblasts, but a CXCL12 decrease indicating a cell-type specific mechanism. In addition, the receptor CCR1 was also downregulated under these conditions, a result consistent with previous reports [8]. CXCL12 and CCR1 were not regulated by DFX on the protein level, but on mRNA level, indicating a HIF independent mechanism. In breast cancer metastasis, hypermethylation of CXCL12 DNA plays a role in CXCL12 downregulation, which mediates selective migration to target organs where the ligand is highly secreted [30]. The significance of CXCL12 and CCR1 downregulation in cytotrophoblasts due to low oxygen conditions is unclear at present. However, mirroring breast cancer metastasis, this regulation in cytotrophoblasts might ensure sufficient invasion in the maternal wall in early pregnancy when fetal oxygen tensions are low. In contrast, DFX led to an increase of CXCR4, CXCR6, and CCR7, suggesting a HIF-dependent mechanism. Supporting these data, HIF-dependent CXCR6 and CCR7 regulation was also shown in breast and lung cancer cells [20,21]. Using DFX, which is a well-

A. Schanz et al. / Placenta 35 (2014) 1089e1094

Fig. 4. Cytotrophoblast migration in low oxygen, towards a chemotactic gradient of CXCL12 and endothelial cells conditioned media in vitro. (A) Transwell migration of cytotrophoblasts in the presence of standard (20%) or low oxygen (2%) in serum free media (SFM), with CXCL12 or growth factor media (GFM, control) added to the lower chamber (for 12 h, n ¼ 6). CXCL12 blocking antibody was added for neutralization of bioactivity. Undirected migration in SFM at 20% oxygen was defined as basic migration without stimulus and set to 100%. (B) Cytotrophoblast migration (n ¼ 3) towards GFM in the absence (set to 100%) or presence of CXCL12 blocking antibody or conditioned media from endothelial cells (CM EC) in the absence or presence of CXCL12 blocking antibody. All values are shown in percent control. Error bars indicate standard deviation. *p < 0.05; **p < 0.001.

known modulator von HIF and subsequent HIF dependent proteins, such as VEGF and erythropoietin [26] an association between HIF regulation and chemokine and chemokine receptor regulation in human cytotrophoblast can be described. Nevertheless, silencing of HIF would have given more proof to the claim and will be subject of further experiments. Taken into account that the choice of the oxygen levels has a great influence on the expected results, we need to address that choosing 2 and 20% oxygen only shows to some extend the in vivo situations. The application of 20% oxygen was discussed as being hyperoxic [32], but according to the literature the partial pressure of oxygen in the culture conditions is also dependent on the density of the cells [33,34] or the thickness of the placental explants [29], which can lead to a partial pressure of 80e120 mmHg in the culture conditions, which would reflect the arterial oxygen partial pressure. During cytotrophoblast differentiation into the invasive, extravillous phenotype, multiple genes, such as integrins avb3 and a1b1 and members of VEGF family are elevated [35]. Chemokines and receptors analyzed in the present study showed the same regulation pattern, underlining that the differentiation process of the cytotrophoblasts exhibits multiple regulations. Taken into account that cytotrophoblasts move towards maternal vessels, endothelial cells themselves might be the origin of chemotactic molecules. Here, cell medium containing endothelium derived factors (CM EC) strongly activated cytotrophoblast

1093

migration in vitro, an effect that was reversed by CXCL12-blocking only to some extent. Thus, uterine endothelial cells have the ability to attract cytotrophoblasts via CXCL12 secretion but might also use other factors e.g. VEGF. Interestingly, CXCL12 expression in endothelial cells is turned off in decidual vessel, which indicates a specific process regulating endothelial CXCL12 expression during pregnancy [10] and further suggests, that endothelial CXCL12 might only play a role in very early pregnancy, when CXCL12 might still be present. Another source of CXCL12 is the decidua, which exhibits much higher amounts of CXCL12 as the placental villi. This might be due to CXCL12 secretion of decidual glands and stromal cells. This finding supports the notion that a chemotactic gradient from the placenta to the uterine wall may guide cytotrophoblasts towards maternal tissues and the bloodstream, which shows the highest amount of CXCL12 [36]. Reduced oxygen further led to increased undirected cytotrophoblast migration, but also to an increase in the directed response to CXCL12, which could be attributed to elevated CXCR4 levels. Growth factor receptors including VEGFR2 can also upregulate in low oxygen conditions [37], the ligand of which is included in the control media (GFM) and might explain the increased migratory response to the GFM. The observation that low oxygen stimulates undirected migration supports the concept that autocrine loops are involved [38,39], but could also exhibit to some extend a proliferative response to the reduced oxygen [1]. In the present data set, cytotrophoblast migration was analyzed to show the possible functional relevance of chemokine and chemokine receptor regulation. Cytotrophoblast migration can be observed in the cell column and during the remodeling of the spiral arteries, but is also coupled closely to cytotrophoblast invasion. Using different assays for migration or invasion examination it must be noted, that diverse results might be detected. For example results for Interleukin-11 showed an inhibition of human cytotrophoblast invasion [40], but had no effect on the migratory response [41], suggesting diverse regulatory mechanisms for these cell movements. Nevertheless, showing an increase in directed cytotrophoblast migration due to low oxygen levels and CXCL12 treatment in the present data, shows that the described regulatory pattern might be of functional importance. Taken together, chemokine ligand and receptor regulation in human cytotrophoblasts by oxygen levels and differentiation to the invasive, extravillous phenotype might play a crucial role in placental development and specific changes CXCL12 expression at the maternalefetal interface are implemented during this process. Defects in this regulation might be followed by insufficient cytotrophoblast invasion, which could lead to either loss of pregnancy or to a later onset of preeclampsia due to insufficient placentation. Conflict of interest The authors declare no conflict of interest. Acknowledgments This work was supported by the Deutsche Forschungsgemeinschaft (DFG) grant to SCHA 1248/2-1, Germany. Placental, decidual, and endometrial tissue was generously provided by Prof. Susan Fisher, UCSF, CA, USA. References [1] Genbacev O, Joslin R, Damsky CH, Polliotti BM, Fisher SJ. Hypoxia alters early gestation human cytotrophoblast differentiation/invasion in vitro and models

1094

[2]

[3] [4] [5]

[6]

[7] [8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16] [17]

[18]

[19]

[20]

[21]

A. Schanz et al. / Placenta 35 (2014) 1089e1094 the placental defects that occur in preeclampsia. J Clin Invest 1996;97(2): 540e50. Rodesch F, Simon P, Donner C, Jauniaux E. Oxygen measurements in endometrial and trophoblastic tissues during early pregnancy. Obstet Gynecol 1992;80(2):283e5. Hustin J, Jauniaux E, Schaaps JP. Histological study of the materno-embryonic interface in spontaneous abortion. Placenta 1990;11(6):477e86. Burton GJ, Caniggia I. Hypoxia: implications for implantation to delivery-a workshop report. Placenta 2001;22(Suppl. A):S63e5. Burton GJ, Woods AW, Jauniaux E, Kingdom JC. Rheological and physiological consequences of conversion of the maternal spiral arteries for uteroplacental blood flow during human pregnancy. Placenta 2009;30(6):473e82. Chaddha V, Viero S, Huppertz B, Kingdom J. Developmental biology of the placenta and the origins of placental insufficiency. Semin Fetal Neonatal Med 2004;9(5):357e69. Genbacev O, Zhou Y, Ludlow JW, Fisher SJ. Regulation of human placental development by oxygen tension. Science 1997;277(5332):1669e72. Sato Y, Higuchi T, Yoshioka S, Tatsumi K, Fujiwara H, Fujii S. Trophoblasts acquire a chemokine receptor, CCR1, as they differentiate towards invasive phenotype. Development 2003;130(22):5519e32. Drake PM, Red-Horse K, Fisher SJ. Reciprocal chemokine receptor and ligand expression in the human placenta: implications for cytotrophoblast differentiation. Dev Dyn: Off Publ Am Assoc Anatomists 2004;229(4):877e85. Hanna J, Wald O, Goldman-Wohl D, Prus D, Markel G, Gazit R, et al. CXCL12 expression by invasive trophoblasts induces the specific migration of CD16human natural killer cells. Blood 2003;102(5):1569e77. Ren L, Liu Y-Q, Zhou W-H, Zhang Y-Z. Trophoblast-derived chemokine CXCL12 promotes CXCR4 expression and invasion of human first-trimester decidual stromal cells. Hum Reprod 2012;27(2):366e74. Red-Horse K, Kapidzic M, Zhou Y, Feng KT, Singh H, Fisher SJ. EPHB4 regulates chemokine-evoked trophoblast responses: a mechanism for incorporating the human placenta into the maternal circulation. Development 2005;132(18): 4097e106. Ceradini DJ, Kulkarni AR, Callaghan MJ, Tepper OM, Bastidas N, Kleinman ME, et al. Progenitor cell trafficking is regulated by hypoxic gradients through HIF1 induction of SDF-1. Nat Med 2004;10(8):858e64. Pringle KG, Kind KL, Sferruzzi-Perri AN, Thompson JG, Roberts CT. Beyond oxygen: complex regulation and activity of hypoxia inducible factors in pregnancy. Hum Reprod Update 2010;16(4):415e31. Genbacev O, Krtolica A, Kaelin W, Fisher SJ. Human cytotrophoblast expression of the von Hippel-Lindau protein is downregulated during uterine invasion in situ and upregulated by hypoxia in vitro. Dev Biol 2001;233(2): 526e36. Ietta F, Wu Y, Winter J, Xu J, Wang J, Post M, et al. Dynamic HIF1A regulation during human placental development. Biol Reprod 2006;75(1):112e21. Rolfo A, Many A, Racano A, Tal R, Tagliaferro A, Ietta F, et al. Abnormalities in oxygen sensing define early and late onset preeclampsia as distinct pathologies. PloS One 2010;5(10):e13288. Kanzler I, Tuchscheerer N, Steffens G, Simsekyilmaz S, Konschalla S, Kroh A, et al. Differential roles of angiogenic chemokines in endothelial progenitor cell-induced angiogenesis. Basic Res Cardiol 2012;108(1):1e14. Oh YS, Kim HY, Song IC, Yun HJ, Jo DY, Kim S, et al. Hypoxia induces CXCR4 expression and biological activity in gastric cancer cells through activation of hypoxia-inducible factor-1alpha. Oncol Rep 2012;28(6):2239e46. Huang X, Su K, Zhou L, Shen G, Dong Q, Lou Y, et al. Hypoxia preconditioning of mesenchymal stromal cells enhances PC3 cell lymphatic metastasis accompanied by VEGFR-3/CCR7 activation. J Cell Biochem 2013;114(12): 2834e41. Li Y, Qiu X, Zhang S, Zhang Q, Wang E. Hypoxia induced CCR7 expression via HIF-1alpha and HIF-2alpha correlates with migration and invasion in lung cancer cells. Cancer Biol Ther 2009;8(4):322e30.

[22] Damsky CH, Fitzgerald ML, Fisher SJ. Distribution patterns of extracellular matrix components and adhesion receptors are intricately modulated during first trimester cytotrophoblast differentiation along the invasive pathway, in vivo. J Clin Invest 1992;89(1):210e22. [23] Fisher SJ, Cui TY, Zhang L, Hartman L, Grahl K, Zhang GY, et al. Adhesive and degradative properties of human placental cytotrophoblast cells in vitro. J Cell Biol 1989;109(2):891e902. [24] Kliman HJ, Nestler JE, Sermasi E, Sanger JM, Strauss 3rd JF. Purification, characterization, and in vitro differentiation of cytotrophoblasts from human term placentae. Endocrinology 1986;118(4):1567e82. [25] Kovats S, Main EK, Librach C, Stubblebine M, Fisher SJ, DeMars R. A class I antigen, HLA-G, expressed in human trophoblasts. Science 1990;248(4952): 220e3. [26] Wang GL, Semenza GL. Desferrioxamine induces erythropoietin gene expression and hypoxia-inducible factor 1 DNA-binding activity: implications for models of hypoxia signal transduction. Blood 1993;82(12):3610e5. [27] Librach CL, Werb Z, Fitzgerald ML, Chiu K, Corwin NM, Esteves RA, et al. 92-kD type IV collagenase mediates invasion of human cytotrophoblasts. J Cell Biol 1991;113(2):437e49. [28] Genbacev O, Jensen KD, Powlin SS, Miller RK. In vitro differentiation and ultrastructure of human extravillous trophoblast (EVT) cells. Placenta 1993;14(4):463e75. [29] Burton GJ, Charnock-Jones DS, Jauniaux E. Working with oxygen and oxidative stress in vitro. Methods Mol Med 2006;122:413e25. [30] Maltepe E, Krampitz GW, Okazaki KM, Red-Horse K, Mak W, Simon MC, et al. Hypoxia-inducible factor-dependent histone deacetylase activity determines stem cell fate in the placenta. Development 2005;132(15):3393e403. [31] Liekens S, Schols D, Hatse S. CXCL12-CXCR4 axis in angiogenesis, metastasis and stem cell mobilization. Curr Pharm Des 2010;16(35):3903e20. [32] Tuuli MG, Longtine MS, Nelson DM. Review: oxygen and trophoblast biologyea source of controversy. Placenta 2011;32(Suppl. 2):S109e18. [33] Metzen E, Wolff M, Fandrey J, Jelkmann W. Pericellular PO2 and O2 consumption in monolayer cell cultures. Respir Physiol 1995;100(2):101e6. [34] Newby D, Marks L, Lyall F. Dissolved oxygen concentration in culture medium: assumptions and pitfalls. Placenta 2005;26(4):353e7. [35] 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;99(9): 2139e51. [36] Schanz A, Winn VD, Fisher SJ, Blumenstein M, Heiss C, Hess AP, et al. Preeclampsia is associated with elevated CXCL12 levels in placental syncytiotrophoblasts and maternal blood. Eur J Obstet Gynecol Reprod Biol 2011;157(1):32e7. [37] Gerber HP, Condorelli F, Park J, Ferrara N. Differential transcriptional regulation of the two vascular endothelial growth factor receptor genes. Flt-1, but not Flk-1/KDR, is up-regulated by hypoxia. J Biol Chem 1997;272(38): 23659e67. [38] Red-Horse K, Drake PM, Fisher SJ. Human pregnancy: the role of chemokine networks at the fetal-maternal interface. Expert Rev Mol Med 2004;6(11): 1e14. [39] Zhou Y, McMaster M, Woo K, Janatpour M, Perry J, Karpanen T, et al. Vascular endothelial growth factor ligands and receptors that regulate human cytotrophoblast survival are dysregulated in severe preeclampsia and hemolysis, elevated liver enzymes, and low platelets syndrome. Am J Pathol 2002;160(4): 1405e23. [40] Wulf H. Gas exchange in the mature placenta in man. II. The causes of uterineumbilical oxygen and carbon dioxide tension gradient. Z Geburtshilfe Gynakol 1962;158:269e319. [41] Ghilain A. The consumption of oxygen by the human placenta in vitro. II. Normal pregnancy. Bull Soc R Belge Gynecol Obstet 1962;32:451e9.