Glucocorticoids modulate multidrug resistance transporters in the first trimester human placenta

Abstract The placental multidrug transporters, P‐glycoprotein (P‐gp, encoded by ABCB1) and breast cancer resistance protein (BCRP,ABCG2) protect the foetus from exposure to maternally derived glucocorticoids, toxins and xenobiotics. During pregnancy, maternal glucocorticoid levels can be elevated by stress or exogenous administration. We hypothesized that glucocorticoids modulate the expression of ABCB1/P‐gp and ABCG2/BCRP in the first trimester human placenta. Our objective was to examine whether dexamethasone (DEX) or cortisol modulate first trimester placental expression of multidrug transporters and determine whether cytotrophoblasts or the syncytiotrophoblast are/is responsible for mediating these effects. Three models were examined: (i) an ex‐vivo model of placental villous explants (7‐10 weeks), (ii) a model of isolated first trimester syncytiotrophoblast and cytotrophoblast cells and (iii) the BeWo immortalized trophoblast cell line model. These cells/tissues were treated with DEX or cortisol for 24 hour to 72 hour. In first trimester placental explants, DEX (48 hour) increased ABCB1 (P < .001) and ABCG2 (P < .05) mRNA levels, whereas cortisol (48 hour) only increased ABCB1 mRNA levels (P < .01). Dexamethasone (P < .05) and cortisol (P < .01) increased BCRP but did not affect P‐gp protein levels. Breast cancer resistance protein expression was primarily confined to syncytiotrophoblasts. BeWo cells, when syncytialized with forskolin, increased expression of BCRP protein, and this was further augmented by DEX (P < .05). Our data suggest that the protective barrier provided by BCRP increases as cytotrophoblasts fuse to form the syncytiotrophoblast. Increase in glucocorticoid levels during the first trimester may reduce embryo/foetal exposure to clinically relevant BCRP substrates, because of an increase in placental BCRP.


| INTRODUCTION
The placenta supplies nutrition to the growing foetus while simultaneously forming a barrier which protects the foetus from exposure to hormones, drugs and toxins (including glucocorticoids, organophosphate pesticides and endocrine disruptors) present in the maternal circulation. [1][2][3] This barrier function is supported by the ABC family of efflux transporters. 4 In this regard, the multidrug resistance transporters, P-glycoprotein (P-gp; encoded by the ABCB1 gene) and breast cancer resistance protein (BCRP [breast cancer resistance protein]; encoded by the ABCG2 gene) are primarily localized to the apical membrane of the syncytiotrophoblast. They efflux substrates from within the syncytiotrophoblast back into the maternal circulation hence preventing factors present in the maternal blood from entering the foetal compartment. 3 In normal gestation, placental P-gp levels are high during first trimester and decline with advancing gestation, whereas placental BCRP levels increase towards term. 3,5,6 In addition, altered expression of these transporters has been identified in reproductive disorders and in pathological pregnancies. 4 Therefore, the biodistribution of their substrates in the foetal compartment is gestational age dependent and may be altered in obstetric pathologies.
Maternal levels of glucocorticoids can be increased by stress in pregnancy and by administration of synthetic glucocorticoids (sGC) to pregnant women. In the early stages of pregnancy, sGC may be used in conjunction with other drugs as part of immunomodulatory treatments to improve the outcomes of in vitro fertilization 7 or given in early gestation to prevent virilization in female foetuses, where there is a risk of congenital adrenal hyperplasia (CAH). [8][9][10] The effects of sGC, such as Dexamethasone (DEX), are mediated via activation of the glucocorticoid receptor (GR). In contrast, endogenous glucocorticoids (eg, cortisol) act through the GR and the mineralocorticoid receptor (MR). Once activated, these receptors act as transcription factors and bind to the glucocorticoid response element (GRE) in the regulatory region of their target genes. 11,12 We have previously shown that sGC modulate expression of P-gp in the murine placenta. Expression of placental Abcb1a mRNA and P-gp protein was increased, whereas expression of placental Abcg2 mRNA was decreased and BCRP function inhibited in the mouse treated with sGC. [13][14][15] In the guinea pig, corticosteroid treatment induced P-gp function in the developing blood-brain barrier, 13 while betamethasone decreased placental Abcb1 mRNA and P-gp protein expression in late gestation, 16 demonstrating tissue-specific regulation. Furthermore, human third trimester preterm labor (PTL)-threatened pregnancies exposed to antenatal betamethasone therapy did not exhibit deranged ABCB1 and P-gp expression. 17 However, increased maternal distress was directly associated with altered term placental expression of ABCB1 and ABCG2 mRNA levels, 18 suggesting that glucocorticoids have the potential to modulate the expression of multidrug resistance transporters in the third trimester placenta in certain conditions. While evidence points to a regulatory action of glucocorticoids on placental multidrug resistance in the later stages of pregnancy, very little is known about the effect of glucocorticoids regulating P-gp/ABCB1 and BCRP/ABCG2 in the human first trimester placenta.
We hypothesized that glucocorticoids modulate the expression of P-gp and BCRP in the first trimester human placenta and that these effects are trophoblast lineage-specific. Therefore, in this study, we examined whether DEX or cortisol altered the placental expression of these multidrug transporters in the first trimester placenta. Further, we determined how trophoblast fusion into syncytium modifies transporter expression and if this is affected by subsequent glucocorticoid exposure.

| Placental villous explant culture
Placental villous explants were cultured as described previously, 3,19 with minor modifications. Briefly, placental specimens were placed into phosphate-buffered saline (1%; PBS) without Ca + and Mg + and transported to the laboratory. Tissues were dissected into villous clusters of approximately 15-30 mg, and 3 villous explants were cultured per well in 12-well plates that contained Dulbecco's modified Eagle's medium/F12, normocin antibiotic (Invivogen, San Diego, CA), and 1X insulin, transferrin and selenium-A (Invitrogen, Grand Island, NY) that was previously equilibrated at 8% O 2 (5% CO 2 , 37°C) for 24 hour. Explants were cultured for 24 hour and then randomly assigned into treatment groups. Explants were treated with DEX or cortisol (10 À8 or 10 À6 mol/L; Sigma-Aldrich, St. Louis, MO), or vehicle for either 24 hour or 48 hour. Explants were then collected and stored at À80°C for total RNA and protein extraction. The culture media was collected to measure lactate dehydrogenase (LDH) in order to assess tissue viability during culture (Roche Applied Science, Indianapolis, IN) as previously described. 3,19

Syncytialization of BeWo cells was induced by treatment with
| 3653 forskolin (25 lmol/L; Sigma-Aldrich) for 72 hour and subsequently these cells were treated with DEX (10 À6 mol/L) or vehicle for a further 72 hour at 2% O 2 (5% CO 2 , 37°C). Non-syncytialized BeWo cells (control) were treated with DEX (10 À6 mol/L) or vehicle for 72 hour at 2% O 2 . After treatments, cells were then collected and stored at À80°C for total RNA and protein extraction.

| First trimester human primary cytotrophoblast and syncytiotrophoblast cell isolation
Cytotrophoblast and syncytiotrophoblast cells were isolated from 7to 10-week placentae as previously described 22

| Immunohistochemistry
After culture with DEX or cortisol, explants were processed for immunohistochemical analysis as previously described. 3

| qPCR
Total RNA was isolated from the first trimester explants and placental cell line with RNeasy Plus Universal Mini Kit (Qiagen, Toronto, ON, Canada), as previously described. 3,19 RNA concentration and purity were assessed with NanoDrop1000 Spectrophotometer (Thermo Scientific, Wilmington, DE) and Experion RNA StdSens Analysis Kit (Bio-Rad, Mississauga, ON, Canada), respectively. RNA was reverse-transcribed into cDNA using the iScript Reverse Transcription Supermix (Bio-Rad). mRNA levels of the ABCB1, ABCG2, GR and MKP-1/DUSP1 (MAP kinase phosphatase-1) genes were measured by qPCR using SYBR Green reagent (Sigma-Aldrich) and the CFX 380 Real-Time system C1000 TM Thermal Cycler (Bio-Rad), with the following cycling conditions: initial denaturation at 95°C (2 minutes), followed by 39 cycles of denaturation at 95°C (5 seconds), combined annealing and extension at 60°C (20 seconds). Gene expression was normalized to the geometric mean of selected reference genes shown in Table 1. Reference genes which exhibited stable expression levels after DEX and cortisol treatments in the first trimester explants were as following: the zeta polypeptide (YWHAZ) and succinate-ubiquinone oxidoreductase (SDHA). The reference genes used following treatments in the Bewo cells were as follows: DNA topoisomerase 1 (TOP1) and 18S ribosomal RNA (18S).

| Protein extraction and immunoblotting
Western blot analysis was conducted as previously described. 3,5 Briefly, protein isolated from placental explants and cultured cells were extracted by sonication using lysis buffer (1 mol/L Tris-HCL pH 6.8, 2% SDS, 10% glycerol containing protease and phosphatase inhibitor cocktail; Thermo Scientific). Protein concentration was determined with the Pierce BCA Protein assay kit (Thermo Scientific).
Proteins were separated by electrophoresis (30 lg 100V, 1 hour) using SDS polyacrylamide gels (7%) and then transferred (10 minutes under UV using ChemiDoc TM MP Imaging system (Bio-Rad). The protein band intensity was quantified using Image Lab TM software and normalized against the ß-actin signal.

| Lactate dehydrogenase cytotoxic assay
Viability of the villous explants treated with DEX or cortisol (10 À8 , 10 À6 mol/L) was determined by measuring LDH leakage into the medium as previously described. 19,23 LDH was quantified with the Cytotoxicity Detection kit (Roche Applied Science) according to the manufacturer's instructions. A standard curve for the LDH assay was generated with LDH from rabbit muscle (Sigma-Aldrich), and absorbance was measured at 490 nm (BioTek Instruments Inc., Winooski, VT). The LDH concentration in the media was normalized to placental explant weight.

| Statistical analysis
Exploratory data analyses were performed with Prism version 6 (GraphPad Software Inc., San Diego, CA). Differences in explant protein levels were assessed by paired t tests in placental explants and by one-way ANOVA followed by the Newman-Keuls post hoc test in BeWo cells. Differences in mRNA levels in explants exposed to DEX and cortisol and culture medium LDH contents were assessed with two-way ANOVA, followed by Bonferroni' s test. Statistical differences were set at P < .05.

| DEX increases BCRP expression in BeWo cells after syncytialization
To better explore the hypothesis that syncytiotrophoblast cells are  Figure 4C) and MPK1/DUSP1 (P < .001; Figure 4D), an established marker of glucocorticoid responsiveness 24 but had no effect on ABCG2 mRNA ( Figure 4A).
By contrast, forskolin treatment significantly increased BCRP protein levels (P < .05), and the addition of DEX, further augmented BCRP protein expression (P < .05; Figure 4F). Dexamethasone exposure after forskolin treatment further increased the expression of MPK1/DUSP1 (P < .001; Figure 4D), but had no additional effect on SYN2 and GR mRNA, and had no effect on ABCG2 mRNA.

| DISCUSSION
This study is the first to show that glucocorticoids increase the expression of ABCB1 mRNA, ABCG2 mRNA and BCRP protein in the first trimester human placenta. Furthermore, we also demonstrated that BCRP is most highly expressed in the syncytiotrophoblast cells and that BCRP is up-regulated following syncytialization of tro- (term). 26 Further, DEX treatment of human trophoblast cell lines (JEG3 and BeWo) and human primary trophoblast cell cultures (derived from term placentae) also increased ABCB1 mRNA expression. 27,28 Together, these results show consistent effects of glucocorticoids on placental ABCB1 mRNA at different stages of pregnancy. However, we did not find changes in P-gp protein expression after glucocorticoid exposure, in the human first trimester placenta, suggesting that effects of glucocorticoids on placental P-gp expression depend upon treatment duration, species and/or gestational age. Furthermore, the We have previously demonstrated that chorioamnionitis induces ABCB1 mRNA expression but decreases P-gp protein levels. In this context, we have identified that miR-331-5p, involved in P-gp suppression, was concomitantly increased in chorioamnionitis and might explain the mismatch between mRNA levels and protein found in chorioamnionitis. 29 It is possible that glucocorticoids regulate expression of microRNAs involved in ABCB1 translation in the first trimester placenta. This possibility clearly warrants further investigation.
Placental ABCB1 and P-gp expression has been shown to be modulated by different factors. We have shown that ABCB1 mRNA and P-gp protein expression in first and third trimester human placental explants is altered by oxygen tension, lipopolysaccharide (LPS, bacterial antigen) and polyinosinic-polycytidylic acid (poly[I:C], viral antigen). 3,19 Importantly, pregnant mice (E15.5-mid to late pregnancy) exposed to LPS exhibited decreased placental P-gp activity. 30 In contrast, polyI:C did not alter P-gp activity, 31  DEX has also been shown to increase expression of other placental transporters, including the amino acid transport system A, and to promote syncytiotrophoblast differentiation and maturation at term. 25,33,34 One limitation of the present study is our inability to effectively assess of P-gp and BCRP function in placental explants; an efficient assay is not currently available.
Using immunohistochemistry, we and others have reported that BCRP is enriched in the syncytiotrophoblast, while there is lower expression in cytotrophoblast cells. 3,5,35 In the present study, we replicated these previous findings. By isolating cytotrophoblast and syncytiotrophoblast cells from first trimester placentae, we further confirmed that BCRP expression is enriched in syncytiotrophoblast cells. Residual expression of BCRP was also evident in isolated Western blot of BCRP protein following the various treatments and (F) corresponding quantification of BCRP protein levels relative to ERK2. *P < .05, **P < .01, ***P < .001 vs VEH; # P < .05 represents differences between forskolin vs forskilin + DEX treatments. Data are presented as mean AE SEM cytotrophoblast cells, which is consistent with our current and previous immunohistochemical studies. 20 As we observed increased placental BCRP expression following DEX treatment, which promotes syncytiotrophoblast differentiation and maturation in vitro, 4,25 we set out to investigate whether cytotrophoblasts, or syncytiotrophoblasts, both of which express GR, 36 are responsible for promoting BCRP up-regulation following glucocorticoid exposure. We utilized the in vitro cytotrophoblast fusogenic model "BeWo cell," which undergoes cell fusion and syncytialization following forskolin exposure. 37 Induction of BCRP expression was observed upon syncytialization with forskolin. In this context, human primary cytotrophoblast cells isolated from term pregnancies increased ABCG2 and BCRP expression as they syncytialize. 38 This suggests that the BCRP-mediated foetal protective barrier forms upon cytotrophoblast fusion and syncytialization.
Furthermore, here we show that DEX enhances BCRP expression following syncytialization, suggesting that glucocorticoids may represent an important inducer of development and differentiation of the trophoblast protective barrier throughout pregnancy, or at least, during the first trimester of gestation, a hypothesis that clearly requires further investigation.
We conclude that glucocorticoids exert selective effects on expression of key multidrug transporters in the human first trimester placenta. Specific effects depend on the type of glucocorticoid, dosage and length of exposure. Breast cancer resistance protein protein is highly expressed in the syncytiotrophoblasts, demonstrating that it is this trophoblast layer that provides BCRP-mediated protective barrier. Exposure of first trimester placental cells to glucocorticoids increases the expression of BCRP in syncytializing and/or syncytialized trophoblast cells, which enhances the placenta barrier to toxins, drugs and endogenous substrates. This suggests that maternal stress and or sGC administration during early stages of pregnancy has the potential to alter placental barrier permeability to a number of BCRP substrates present the maternal circulation. Our data offer evidence that syncytiotrophoblasts are already responsive to glucocorticoids at early stages of pregnancy and that glucocorticoids induce a further increase in the expression of drug transporter proteins and provide control over the factors entering the foetal compartment.