The breast cancer resistance protein (BCRP) is abundant in the placenta and protects the fetus by limiting placental drug penetration. We hypothesize that pregnancy-specific hormones regulate BCRP expression. Hence, we examined the effects of progesterone (P4) and 17β-estradiol (E2) on BCRP expression in the human placental BeWo cells. P4 and E2 significantly increased and decreased BCRP protein and mRNA, respectively. Likewise, treatment with P4 and E2 increased and decreased, respectively, fumitremorgin C-inhibitable mitoxantrone efflux activity of BeWo cells. Reduction in BCRP expression by E2 was abrogated by the estrogen receptor (ER) antagonist ICI-182,780. However, the progesterone receptor (PR) antagonist RU-486 had no effect on P4-mediated induction of BCRP. P4 together with E2 further increased BCRP protein and mRNA compared with P4 treatment alone. This combined effect on BCRP expression was abolished by RU-486, ICI-182,780, or both. Further analysis revealed that E2 significantly decreased ERβ mRNA and strongly induced PRB mRNA in a dose-dependent manner but had no effect on PRA and ERα. P4 alone had no significant effect on mRNA of ERα, ERβ, PRA, and PRB. E2 in combination with P4 increased PRB mRNA, but the level of induction was significantly reduced compared with E2 treatment alone. Taken together, these results indicate that E2 by itself likely downregulates BCRP expression through an ER, possibly ERβ. P4 alone upregulates BCRP expression via a mechanism other than PR. P4 in combination with E2 further increases BCRP expression, presumably via a nonclassical PR- and/or E2-mediated synthesis of PRB.
- hormonal regulation
- ATP-binding cassette transporter
the breast cancer resistance protein (bcrp) is the second member (gene symbol ABCG2) of the subfamily G of the large ATP-binding cassette (ABC) transporter superfamily (1, 9, 25). BCRP is highly expressed in many normal tissues, including the epithelium of the small intestine and the liver canalicular membrane (22). Therefore, in addition to conferring resistance in cancer cells to chemotherapeutic agents such as mitoxantrone (MX), topotecan, and methotrexate (8, 9, 25, 36), BCRP has been shown to mediate apically-directed drug transport and play a significant role in absorption, distribution, and elimination of BCRP substrates (4, 19, 21, 32, 35). Of interest is that BCRP is also abundantly expressed in the apical membrane of placental syncytiotrophoblasts (22). Whereas the precise physiological role of BCRP in the placenta is still unclear, existing data suggest that BCRP may protect the fetus against toxic substances/drugs and metabolites by extruding them across the placental barrier. For example, Bcrp1, the murine homolog of BCRP, has been shown to significantly alter fetal distribution of topotecan, a BCRP substrate. The fetus/plasma ratio of topotecan was increased twofold in pregnant mice treated with the BCRP inhibitor GF-120918 compared with the vehicle-treatment control (19).
Distribution of drugs that are BCRP substrates across the placenta may, therefore, be altered by factors that can influence BCRP expression in the placenta. Several recent studies have shown that pregnancy can affect expression and function of ABC transporters (5, 13, 23). For instance, expression of P-glycoprotein (P-gp) protein in human placenta at early gestational stages (13–14 wk) was found to be 2–45 times higher than that at late gestational stages (38–41 wk) (13, 23). Expression and function of multidrug resistance protein 2 in the liver of pregnant rats decreased to 50% of that of nonpregnant control rats (5). Thus the protection of fetuses and drug disposition in general can be influenced by pregnancy through changing the expression and function of these transporters. A recent study by Mathias et al. (23) showed that BCRP expression in human placenta did not change significantly with gestational age. Because these studies were preliminary with limited tissue samples, and because substantial variation in BCRP expression (mRNA and protein) was observed, more detailed analysis is needed.
To date, little is known about the molecular mechanism by which expression of ABC transporters in the placenta is altered by pregnancy. Progesterone (P4) and 17β-estradiol (E2) are the two most important steroid hormones produced by the human placenta during pregnancy. Estrogens, including E2, play important roles in regulating the growth, development, and differentiation of many reproductive tissues. P4 is believed to be indispensable for the maintenance of pregnancy. Because the concentrations of E2 and P4 continuously increase throughout the course of pregnancy, we hypothesized that E2 and P4 play a significant role in regulating expression of ABC transporters in human placenta. Recent studies (10, 11, 17) have indeed demonstrated that E2 is an important determinant in the regulation of BCRP expression in cancer cells by transcriptional or posttranscriptional mechanisms. The effects of P4 and, particularly, the combined effects of E2 and P4 on BCRP expression have not been reported.
In the present study, we have systematically analyzed the effects of P4 and E2 on expression and efflux function of BCRP in the model human placental BeWo cells, which express high levels of endogenous BCRP (2). We found that E2 by itself decreased BCRP expression and P4 increased BCRP expression. P4 in combination with E2 further increased BCRP expression compared with P4 treatment alone. The effects of E2 and P4 on expression of progesterone receptor A (PRA), progesterone receptor B (PRB), estrogen receptor-α (ERα), and estrogen receptor-β (ERβ) have also been investigated to explore the possible contribution of these steroid hormone nuclear receptors in regulating BCRP expression in BeWo cells. These studies found that some of the steroid hormone nuclear receptors could be involved in the regulation of BCRP in BeWo cells. Our findings provide new insights into the regulation of BCRP in the human placenta by pregnancy.
MATERIALS AND METHODS
P4 (P-8783), E2 (E-2758), and 17β-hydroxy-11β-(4-dimethylaminophenyl)-17α-(1-propynyl)estra-4,9-dien-3-one (RU-486) were purchased from Sigma (St. Louis, MO). 7a,17b-[9-[(4,4,5,5,5-pentafluoropentyl)-sulfinyl]-nonyl]-estra-1,3,5(10)-triene-3,17-diol (ICI-182,780) was from Tocris Cookson (Ellisville, MO). Fumitremorgin C (FTC) was a kind gift from Dr. Susan Bates [National Cancer Institute (NCI), Bethesda, MD]. HPLC-grade DMSO was from Fisher Scientific (Pittsburgh, PA) and was used as the solvent to dissolve the above compounds. [3H]MX (1.5 Ci/mmol) was purchased from Moravek Biochemicals (Brea, CA). The Complete protease inhibitor cocktail was obtained from Roche Molecular Biochemicals (Mannheim, Germany). The Laemmli sample buffer and 2-mercaptoethanol were purchased from Bio-Rad (Hercules, CA). DNase I was obtained from Sigma. The BeWo cell line was from American Type Culture Collection (Manassas, VA). RPMI 1640 phenol red free and GIBCO Opti-MEM were from Gibco (Grand Island, NY). Phosphate-buffered saline (PBS) and fetal bovine serum (FBS) were from Invitrogen (Carlsbad, CA). Charcoal/dextran-stripped FBS was purchased from HyClone (Logan, UT).
Cell culture and whole cell lysate preparation.
The BeWo cells were maintained in RPMI 1640 phenol red free medium supplemented with 10% FBS and 2 mM l-glutamine at 37°C in a 5% CO2-humidified incubator. The medium was replaced with fresh medium every other day. To examine BCRP expression in the BeWo cells treated with E2, P4, or both, the cells were first cultured in RPMI 1640 phenol-red free medium supplemented with 5% charcoal/dextran-stripped FBS for at least 48 h to achieve 60–70% confluence. The medium was then replaced with fresh medium, and E2 or P4 at various concentrations was then added into the medium. Cell culture was continued for an additional 12–72 h with replacement of medium after 48 h. For studies in which cells were treated with a combination of E2 and P4, the cells were first primed with E2 at various concentrations for 24 h. The medium was then replaced with fresh medium, and the cells were incubated with E2 at the same concentrations in the presence of P4 for 72 h. The cells were then harvested for immunoblotting, mRNA isolation, or functional assays. Only cells within eight passages after purchase were used in these experiments. The concentration of DMSO used in all experiments was 0.1% (vol/vol). No effects of the vehicle on cell viability, BCRP protein and mRNA expression, the plasma membrane localization of the transporter, and MX efflux activity were observed at this concentration.
For whole cell lysate preparation, the BeWo cells grown in 10-cm dishes were washed once with ice-cold PBS after hormone treatment and then harvested by scraping the cell monolayer in ice-cold PBS. The suspended cells were centrifuged at 400 g for 5 min at 4°C. The cell pellet was resuspended in 200 μl of lysis buffer (1 M Tris·HCl, pH 7.5, 10% SDS, 5 mg/ml DNase I, 1 M MgCl2, 50 mg/ml PMSF, and protease inhibitor cocktail). The mixture was placed on ice for 1 h with gentle vortexing every 15 min, sonicated on ice by use of a tip-top sonicator for 20 s, and finally centrifuged at 15,100 g for 15 min at 4°C. The supernatant was immediately frozen in liquid N2 in aliquots and stored at −80°C until use. Protein concentrations were determined by the Bio-Rad DC protein assay kit (Bio-Rad), using bovine serum albumin as standard.
SDS-polyacrylamide gel electrophorsis and immunoblotting.
The protein samples of whole cell lysates (20 μg each lane) were subjected to immunoblotting by use of BXP-21 (1:500 dilution), a BCRP-specific monoclonal antibody (MAb) (Kamiya Biomedical, Seattle, WA), as previously described (15), with the exception that the secondary antibody, goat anti-mouse HRP-conjugated antibody (Bio-Rad), was used at 1:5,000 dilution. For detection of β-actin, an MAb specific for human β-actin (Sigma) was used as the primary antibody at 1:50,000 dilution, and the goat anti-mouse HRP-conjugated antibody (Bio-Rad) was used as secondary antibody at 1:25,000 dilution. Relative BCRP protein levels were determined by densitometric analysis of the immunoblots using the NIH Scion Image software (Scion, Frederick, MD). β-Actin was used as an internal control.
BeWo cells were seeded at ∼5 × 104 cells/well in a four-chamber glass slide (Falcon; BD Biosciences Discovery Labware, Bedford, MA). Cells were grown and treated with 10−5 M P4, 10−7 M E2, or vehicle control [0.1% (vol/vol) DMSO] for 72 h as described. After treatment, cells were washed twice with PBS at room temperature. Cells were then fixed with 4% paraformaldehyde in PBS for 30 min, washed twice with PBS, and incubated in permeabilization buffer (0.2% Triton X-100 in PBS) at room temperature for 10 min. Cells were then blocked for 90 min in blocking solution (0.1% Triton X-100–2% FBS) and incubated with BXP-21 (1:250 dilution in blocking solution) for 1 h at room temperature. After the cells were washed with blocking solution twice, Alexa fluor 488-conjugated goat anti-mouse IgG (H + L) (Fab')2 fragment (Molecular Probes) was added (1:1,000 dilution in blocking solution) and incubated in the dark for 1 h. Cells were then washed twice with PBS and mounted in Fluoromount G (Southern Biotechnology Associates, Birmingham, AL) and observed at 488-nm excitation and 519-nm emission wavelengths using a Leica TCS SPI MP multiphoton confocal microscope (Leica Microsystems, Exton, PA). The concentration of DMSO used in all experiments was 0.1% (vol/vol).
Total RNA isolation and quantitative real-time TaqMan RT-PCR analysis.
The effects of hormone treatment on mRNA expression of BCRP, PRA, PRB, ERα, or ERβ were quantified by TaqMan real-time reverse transcription-polymerase chain reaction (RT-PCR) as follows. After treatment of the BeWo cells with P4, E2, RU-486, or ICI-182,780, total cellular RNA was isolated from the cells by using the TRIzol reagent (Invitrogen) according to the manufacturer's instructions. To eliminate contamination of genomic DNA, all RNA samples were treated with DNase I (Promega, Madison, WI) and purified by ethanol precipitation before RT-PCR. The concentration of RNA was determined by measuring optical density at 260 nm. The OD260/OD280 nm ratios of all RNA samples were determined to be between 1.7 and 2.0 to ensure that all RNA samples are highly pure. RNA integrity was examined by agarose gel electrophorsis. Single-strand cDNA used for analysis of BCRP was then synthesized from 0.5 μg of purified total RNA using a TaqMan reverse transcription kit (Applied Biosystems, Branchberg, NJ), and single-strand cDNA used for analysis of PRA, PRB, ERα, or ERβ was synthesized from 2.5 μg of purified total RNA by use of a a high-capacity cDNA archive kit (Applied Biosystems, Foster City, CA), all in a volume of 25 μl. The synthesized cDNA was further purified by ethanol precipitation and dissolved in 25 μl of pure H2O. Real-time PCR reactions were then performed using a TaqMan universal PCR master mix on the ABI Prism 7000 Sequence Detection System (Applied Biosystems, Foster City, CA). All the primers and specific probes were synthesized by Applied Biosystems. Reactions were carried out in quadruplicates in a MicroAmp optical 96-well plate in a total volume of 20 μl. Each reaction mixture contained 10 μl of 2× TaqMan universal PCR master mix, 6.1 μl of sterile Millipore water, 0.47 μl of forward primer (235 nM), 0.47 μl of reverse primer (235 nM), 0.47 μl of probe (118 nM), and 2.5 μl of reverse-transcription products. PCR conditions were as follows: 50°C for 2 min, 95°C for 10 min, 95°C for 15 s, 60°C for 1 min (40 cycles). Quantification of relative mRNA levels was carried out by determining the threshold cycle (CT), which is defined as the cycle at which the 6-carboxyfluorescein reporter fluorescence exceeds by 10 times the standard deviation of the mean baseline emission for cycles 3 to 10. β-Actin was used as an internal control. The mRNA levels of BCRP, PRA, PRB, ERα, or ERβ were normalized to those of β-actin according to the following formula: CT (BCRP, PRA, PRB, ERα, or ERβ) − CT (β-actin) = ΔCT. Thereafter, the relative mRNA levels of these genes after hormone treatment were calculated using the ΔΔCT method: ΔCT (test hormone) − ΔCT (vehicle) = ΔΔCT (test hormone). The fold changes of mRNA levels of BCRP, PRA, PRB, ERα, or ERβ in BeWo cells upon treatment with respective hormones were expressed as 2. The primer pairs and probe for BCRP were 5′-CAGGTCTGTTGGTCAATCTCACA-3′ (forward), 5′-TCCATATCGTGGAATGCT GAAG-3′ (reverse), and 5′-CCATTGCATCTTGGCTGTCATGGCTT-3′(probe); the primer pairs and probe for PRA were 5′-AGAGCACTGGATGCTGTTGCT-3′ (forward), 5′-TGGCTTAGGGCTTGGCTTT-3′ (reverse), and 5′-CCACAGCCATTGGGCGTTCCAA-3′(probe); the primer pairs and probe for PRB were 5′-GCCAGACCTCGGACACCTT-3′(forward), 5′-CAGGGCCGAGGGAAGAGTAG-3′(reverse), and 5′-CCTGAAGTTTCGGCCATACCTATCTCCCT-3′ (probe); the primer pairs and probe for ERα were 5′-AGCACCCAGTGAAGCTACT-3′ (forward), 5′-TGAGGCACACAAACTCCT-3′ (reverse), and 5′-TGGCTACATCATCTCGGTTCCGCA-3′ (probe); the primer pairs and probe for ERβ were 5′-AAGAATATCTCTGTCAAGGCCATG-3′ (forward), 5′-GGCAATCACCCAAACCAAAG-3′(reverse), and 5′-TTGCTGAACGCCGTGACCGATG-3′ (probe). The primer pairs and probe for human β-actin were purchased from Applied Biosystems (Foster City, CA). The concentration of DMSO used in all experiments was 0.1% (vol/vol).
Intracellular MX accumulation assay.
Transport studies using [3H]MX were performed to examine whether treatment with P4 and E2 affects MX efflux activity of the BeWo cells. Briefly, the BeWo cells were seeded at a cell density of ∼2 × 105 per well in six-well plates and treated as described with P4 and/or E2 in the presence and absence of RU-486 and ICI-182,780 at concentrations indicated in Table 1. After 72 h of treatment, cells grown on the cell culture plates as a monolayer were washed once with prewarmed PBS and incubated in 1 ml per well of Opti-MEM for 30 min. In inhibition experiments, cells were first incubated with 10 μM FTC for 1 h. The experiments were then started by the addition of [3H]MX (20 nM) in the presence and absence of 10 μM FTC in 1 ml of Opti-MEM, and incubation was continued for 30 min to 90 min. The MX efflux was then stopped by washing the cells three times with ice-cold PBS. The cell monolayer was suspended in 1 ml of 2% (wt/vol) SDS for whole cell lysate preparation. The whole cell lysates (900 μl) were subjected to counting in a scintillation counter. Counts were normalized to the protein concentration that was measured by the Bio-Rad DC protein assay, using the remaining lysates. The intracellular MX concentrations were calculated on the basis of radioactivity associated with the cells and presented as picomoles of [3H]MX per milligram protein. The difference in intracellular MX concentrations in the presence and absence of FTC was used as a measure of FTC-inhibitable MX efflux activity of the BeWo cells. This FTC-inhibitable MX efflux activity should be attributable to BCRP. Only cells within eight passages after purchase were used in the experiments. The experiments were performed in triplicate at 37°C in a humidified incubator.
Data were analyzed for statistical significance using one-way ANOVA analysis or Student's t-test. Differences with P values of <0.05 were considered statistically significant.
P4 stimulates BCRP protein expression.
We first examined whether treatment with P4 or E2 can affect membrane localization of BCRP in BeWo cells with immunofluorescent confocal microscopy by use of the BCRP-specific mAb BXP-21. We found that BCRP was predominantly expressed on the plasma membrane of untreated BeWo cells with some intracellular expression (Fig. 1A). Treatment with 10−5 M P4 or 10−7 M E2 for 72 h had no qualitative effect on the plasma membrane localization of the transporter vs. the intracellular compartments (Fig. 1, B and C). Thus the levels of BCRP protein determined in the whole cell lysates should reflect the levels of BCRP protein on the plasma membrane. We therefore determined BCRP protein expression using whole cell lysates in all of the subsequent immunoblotting experiments.
To investigate the effect of P4 on BCRP protein expression, the BeWo cells were treated with P4 over a range of concentrations (10−9 − 10−5 M) for 48 h. BCRP protein expression was then analyzed by immunoblotting of whole cell lysates, using mAb BXP-21. Densitometric analysis of the immunoblots revealed that P4 at 10−6 M only slightly increased BCRP protein expression; however, P4 at 10−5 M significantly increased BCRP protein ∼2-fold compared with the vehicle control (Fig. 2A). P4 at concentrations below 10−6 M had no significant effect on BCRP expression. In contrast, the expression of β-actin (internal control) was not significantly affected by P4 at any of the concentrations used. When the effect of P4 on BCRP expression was analyzed at different treatment times, P4 at 10−5 M was found to increase BCRP expression at 48 and 72 h (Fig. 2B). The inductive effect of 10−5 M P4 on BCRP protein expression could not be reversed by the addition of the PR antagonist RU-486 at 10−5 M (Fig. 2C). RU-486 itself at 10−5 M had no significant effect on BCRP protein expression (data not shown). We found that the viability of BeWo cells was significantly reduced in the presence of RU-486 at 5 × 10−5 M or higher concentrations (data not shown). To further clarify the role of PR on BCRP expression, we performed experiments to examine the effect of RU-486 on BCRP expression at 10 times molar excess to P4. Thus BeWo cells were treated for 72 h with 2.5 × 10−6 M P4 in the presence and absence of 2.5 × 10−5 M RU-486. Similarly, P4 at 2.5 × 10−6 M increased BCRP protein and mRNA expression ∼1.7- and 1.5-fold, respectively, and the addition of 10 times molar excess RU-486 had no significant effect on P4-mediated induction of BCRP expression (Fig. 3). The immunoblots sometimes showed double bands (Fig. 2B). We treated the protein samples with PNGase F (New England Biolabs, Beverly, MA). This treatment led to an ∼10 kDa reduction in the apparent molecular mass of BCRP and the disappearance of double bands (data not shown). These data suggest that the double bands are most likely caused by multiple glycosylation on BCRP. Hence, the upper bands were always included in the densitometric analysis of BCRP expression.
E2 decreases BCRP protein expression.
To examine the effect of E2 on BCRP protein, BeWo cells were treated with E2 at various concentrations (10−11 − 10−7 M) and for the duration of treatment up to 72 h. E2 at 10−8 and 10−7 M significantly decreased BCRP protein expression after 48 h of treatment by ∼60% and 70%, respectively; however, β-actin expression was not affected by the same experimental conditions (Fig. 4A). E2 only slightly decreased BCRP protein expression at concentrations below 10−9 M. In time course studies, E2 at 10−7 M was found to significantly decrease BCRP protein expression by ∼60 and 75% at 48 and 72 h, respectively, but had no effect at 12 and 24 h (Fig. 4B). The inhibitory effect of E2 at 10−7 M after 72 h of treatment was significantly reversed by the addition of 10 times molar excess (10−6 M) of the ER antagonist ICI-182,780 (Fig. 4C). ICI-182,780 itself at 10−6 M had no significant effect on BCRP protein expression (data not shown).
BCRP protein expression was further increased by P4 in combination with E2.
To examine the combined effects of P4 and E2 on BCRP expression, BeWo cells were first primed with E2 at various concentrations (10−9 − 10−7 M) for 24 h. The cells were then switched to fresh medium and incubated with E2 at the same concentrations and P4 at 10−6 or 10−5 M for 72 h. P4 at 10−5 M in combination with E2 at 10−9 or 10−8 M further increased BCRP protein expression compared with P4 treatment alone (Fig. 5B). For example, P4 alone at 10−5 M stimulated BCRP protein expression ∼2-fold, whereas P4 at the same concentration, with 10−8 M E2, increased BCRP protein ∼3-fold (Fig. 5B). Although not statistically significant, this further stimulation of BCRP protein by 10−8 M E2 was greater than that by 10−9 M E2. However, a further increase of E2 concentration to 10−7 M decreased rather than increased BCRP expression. Similar effects of the combination of P4 with E2 on BCRP protein were observed for P4 at 10−6 M (Fig. 5A). The expression of β-actin was not influenced by any of these treatments. We then explored the combined effects of P4 and E2 on BCRP expression in the presence of 10−6 M ICI-182,780 and/or 10−5 M RU-486. The further stimulation of BCRP protein by the combination of 10−5 M P4 and 10−8 M E2 was completely abrogated by the addition of ICI-182,780, RU-486, or both. Treatment with ICI-182,780 or RU-486 decreased BCRP protein expression to 1.9-fold and 2.0-fold of the vehicle controls, respectively, and brought it down to the same level as for P4 (10−5 M) treatment alone (Fig. 5C). Likewise, treatment with ICI-182,780 and RU-486 decreased BCRP protein expression to 1.7-fold of the vehicle controls (Fig. 5C).
Effects of P4 and E2 on BCRP mRNA.
We then examined whether the effects of P4 and E2 on BCRP protein expression were due to changes on BCRP mRNA levels. Endogenous BCRP mRNA in BeWo cells could be readily detected by real-time RT-PCR at ∼23 cycles (data not shown). Quantitative real-time RT-PCR analyses revealed that treatment of BeWo cells with P4 alone at 10−5 M significantly increased BCRP mRNA ∼1.5-fold compared with the vehicle control (Fig. 6). Similar results were obtained after treating the BeWo cells for 24 (Fig. 6A) and 72 h (Fig. 6B). Although not statistically significant, 10−5 M P4 in combination with 10−8 M E2 further increased BCRP mRNA compared with P4 treatment alone. After 24 and 72 h of treatment, E2 by itself at 10−7 M significantly reduced BCRP mRNA by ∼40%. Although the ER antagonist ICI-182,780 at 10 times molar excess completely reversed the inhibitory effect of E2 on BCRP mRNA, the PR antagonist RU-486 at the same concentration (10−5 M) as P4 did not significantly affect the P4-mediated induction of BCRP mRNA (Fig. 6). RU-486 at 10 times molar excess to P4 (2.5 × 10−5 M RU-486 vs. 2.5 × 10−6 M P4) also did not influence the inductive effect of P4 on BCRP mRNA (Fig. 3C). RU-486 and ICI-182,780 themselves had no effect on BCRP mRNA expression (data not shown).
Effects of P4 and E2 on BCRP-mediated MX efflux activity.
To further examine whether the function of BCRP in BeWo cells is affected by treatment with P4 and/or E2, we investigated the effects of hormone treatment on MX efflux by the BeWo cells by using an MX accumulation assay. MX, a high-affinity BCRP substrate (29), was used as a model substrate to measure BCRP transport activity of the BeWo cells. To eliminate possible contribution of endogenous efflux transporters such as P-gp, a relatively specific BCRP inhibitor FTC was used to determine FTC-inhibitable MX efflux activity. Because 10 μM FTC used in the assay is sufficient to fully inhibit BCRP (29), the portion of MX efflux that can be inhibited by 10 μM FTC is attributable to BCRP expression. Similar FTC modulation of MX efflux has been used to detect BCRP expression in clinical leukemia samples (33). We first performed time course studies to find the optimal accumulation time for the efflux assay. The baseline FTC-inhibitable MX efflux activity of BeWo cells treated with the vehicle control after 60 min of accumulation was slightly greater than the activity after 30 min of accumulation; however, further increase of accumulation time to 90 min did not increase the activity (Fig. 7). With all three accumulation times, treatment with 10−5 M P4 or 10−7 M E2 significantly increased or decreased, respectively, the FTC-inhibitable MX efflux activity of the BeWo cells (Fig. 7), and accumulation for 60 min seems to produce the most significant difference compared with the vehicle control. Therefore, an accumulation time of 60 min was used in all of the subsequent efflux experiments. As shown in Table 1, the FTC-inhibitable MX efflux by BeWo cells treated with 10−7 M E2 was significantly reduced, by ∼30% compared with the vehicle control cells. This reduction of MX efflux by E2 treatment was completely reversed by the addition of 10−6 M ICI-182,780. Treatment with 10−5 M P4 or 10−5 M P4 in combination with 10−8 M E2 resulted in increases of ∼1.2-fold and 1.4-fold, respectively, in MX efflux. In particular, the MX efflux activity of the cells treated with 10−5 M P4 in combination with 10−8 M E2 was significantly greater than the activity of the cells treated with 10−5 M P4 alone. These findings are consistent with the protein and mRNA data (Figs. 5 and 6). The addition of 10−5 M RU-486 had no effect on P4-mediated stimulation of MX efflux (Table 1).
Effects of E2 and P4 on mRNA of PRA, PRB, ERα, and ERβ.
To investigate the possible mechanisms by which BCRP expression is regulated by P4 and/or E2 in BeWo cells, we examined the effects of P4 and/or E2 on mRNA levels of PRA, PRB, ERα, and ERβ by using quantitative real-time RT-PCR. To perform this study, BeWo cells were treated with E2, P4, or a combination of both, the same as described for determining protein expression. Endogenous expression of PRA, PRB, ERα, and ERβ mRNA in untreated BeWo cells was detected by real-time PCR at 37.1 ± 0.7, 37.6 ± 0.5, 35.0 ± 0.6, and 35.6 ± 0.4 cycles, respectively, under the present assay conditions. The relative changes of mRNA levels of these receptors upon hormone treatment were then determined and are summarized in Table 2. First, E2 by itself at 10−7 and 10−8 M significantly decreased ERβ mRNA by ∼40% compared with the vehicle control but had no significant effect on ERα mRNA. E2 at 10−7 M did not significantly influence PRA mRNA and increased PRB mRNA by 60%. However, E2 at 10−8 M strongly induced PRB mRNA ∼7.5-fold but had no significant affect on PRA mRNA. These effects of E2 on mRNA levels of ERβ and PRB were abrogated by the addition of 10−6 M ICI-182,780. For instance, the addition of ICI-182,780 almost completely reversed E2-mediated reduction of ERβ mRNA. In addition, the 7.5-fold increase in PRB mRNA by 10−8 M E2 was significantly reduced to only 2-fold by the addition of ICI-182,780. Second, when the effect of P4 alone was measured, P4 at 10−5 M was found to slightly decrease mRNA levels of PRA and PRB but had no effect on mRNA levels of ERα and ERβ. The addition of 10−5 M RU-486 did not influence the effect of P4 on mRNA levels of these receptors (data not shown). Third, we measured the combined effects of P4 and E2 on mRNA of PRA, PRB, ERα, and ERβ. A combination of 10−5 M P4 with 10−8 M E2 did not decrease ERβ mRNA, although E2 alone significantly attenuated ERβ mRNA. The combination of P4 and E2 also had no significant effect on ERα mRNA but slightly decreased PRA mRNA and significantly increased PRB mRNA ∼2.2-fold.
The present study examines the effects of P4 and E2 on BCRP expression in BeWo cells. We found that P4 significantly increased BCRP protein only at a relatively high concentration 10−5 M (Fig. 2A). The plasma P4 concentration at term was reported to be ∼0.7 × 10−6 M, and the intracellular P4 concentrations in placenta were about 12 times greater than those in maternal plasma (20). Hence, 10−5 M P4 could be achieved in the placenta at term. This concentration is much greater than the binding affinity of P4 to a classical PR (7). Moreover, RU-486, even at 10 times molar excess, did not inhibit the inductive effect of P4 (Fig. 3). These results suggest that it is unlikely that induction of BCRP by P4 is mediated by a classical PR. Several studies have demonstrated novel, nonclassical, membrane-bound forms of steroid receptors involving nongenomic actions of hormones (12, 27). For example, P4 at μM concentrations stimulated the expression of the steroidogenic acute regulatory protein in Leydig cells by a nonclassical PR (30). Therefore, upregulation of BCRP in BeWo cells by P4 at μM concentrations is possibly mediated by a nonclassical PR pathway.
The plasma E2 concentration during pregnancy increases steadily to around 0.8 × 10−7 M at term (3, 20). At concentrations observed during pregnancy (10−8 and 10−7 M), E2 significantly decreased BCRP protein (Fig. 4A). This decrease was abolished by ICI-182,780 (Fig. 4C), suggesting that downregulation of BCRP by E2 is mediated by ER. Male-predominant expression of Bcrp1 in rat kidney has been reported (34). The authors showed that castration had no effect on Bcrp1 mRNA in rat kidney; however, Bcrp1 mRNA in the kidneys of ovariectomized female rats was significantly higher than that of control females, indicating that male-predominant expression of Bcrp1 in rat kidneys is likely caused by the absence of the suppressive effects of female sex hormones such as E2. Male-predominant expression of human BCRP and mouse Bcrp1 in liver has also been demonstrated (24). These in vivo data seem to support our in vitro findings with respect to downregulation of BCRP by E2.
We demonstrated for the first time the combined effects of P4 and E2 on BCRP expression. It is of considerable interest that E2 at subthreshold doses (10−9 and 10−8 M) further increased P4-mediated induction of BCRP, even at 10−6 M P4, which by itself showed little effect (Figs. 2A and 5A). This finding suggests that placental BCRP expression could be affected by pregnancy, even at earlier gestational stages when the P4 concentrations are low. Because E2 at 10−8 M with 10−5 M P4 significantly induced PRB mRNA 2.2-fold (Table 2), this further increase in BCRP expression is possibly mediated by E2-induced synthesis of PRB. The combination of 10−7 M E2 with 10−5 M P4 decreased, rather than increased, BCRP expression to the levels of P4 treatment alone (Fig. 5). This could be explained by the fact that E2-induced synthesis of PRB was significantly diminished by 10−7 M E2 compared with 10−8 M E2 (Table 2). Hence, both a nonclassical PR (when P4 was used alone) and a classical PR (when P4 and E2 were used together) may be involved in P4-mediated upregulation of BCRP. E2 at 10−8 M induced PRB mRNA 7.5-fold. The decrease in E2-mediated induction of PRB in the presence of P4 is likely due to the widely observed suppressive effect of P4 on E2 action (28). The addition of RU-486, ICI-182,780, or both abolished this further increase in BCRP protein by the combination of P4 and E2 (Fig. 5C), further suggesting that endogenous expression of PRB in BeWo cells is low, and thus PRB exerts its function only after it is induced by E2 through ER.
The effects of P4 and E2 on BCRP mRNA in general corresponded well to the effects on BCRP protein (Fig. 6), suggesting that P4 and E2 regulate BCRP expression, at least in part, by a transcriptional mechanism. However, the possibility of a posttranscriptional mechanism cannot be excluded (17). We consistently observed a significant increase and decrease of the FTC-inhibitable MX efflux activity of BeWo cells treated with 10−5 M P4 and 10−7 M E2, respectively, compared with the vehicle controls (Fig. 7 and Table 1). ICI-182,780 completely reversed the inhibitory effect of E2, and RU-486 had no significant influence on the stimulatory effect of P4. The combination of P4 and E2 further increased MX efflux activity compared with P4 treatment alone. These activity data in general reflected well the BCRP protein (e.g., 122% increase in activity vs. 150–200% increase in protein by 10−5 M P4 and 30% decrease in activity vs. 60–70% decrease in protein by 10−7 M E2) and mRNA data (e.g., 122% increase in activity vs. 150% increase in mRNA by 10−5 M P4 and 30% decrease in activity vs. 40% decrease in mRNA by 10−7 M E2). Recently, Imai et al. (16) reported expression of endogenous MX efflux transporters other than BCRP in LLC-PK1 cells that can be inhibited by FTC. We also noticed the existence of other endogenous efflux transporters for the anti-HIV protease inhibitors ritonavir and saquinavir in human embryonic kidney cells inhibited by FTC (15). Therefore, the relatively smaller effects of P4 or E2 on MX efflux activity are most likely attributable to the endogenous transporters other than BCRP in BeWo cells, whose MX efflux activity can also be inhibited by FTC. Such endogenous MX efflux transporters would increase the background of the overall FTC-inhibitable MX efflux activity of the BeWo cells (which do not have enforced BCRP expression by transfection) and mask the changes in BCRP-specific MX efflux activity.
ERα has been detected in BeWo cells (18). This study, to the best of our knowledge, is the first to demonstrate expression of PRA, PRB, and ERβ in BeWo cells. E2 at 10−8 and 10−7 M significantly decreased ERβ mRNA (Table 2). Several studies (6, 14, 26, 31) also reported downregulation of ERα and/or ERβ by E2 in various tissues and cell lines. Because ICI-182,780 completely reversed the inhibitory effect of E2 on ERβ (Table 2), downregulation of BCRP by E2 is possibly mediated by a transcriptional mechanism via ERβ. A combination of 10−5 M P4 and 10−8 M E2 did not decrease ERβ mRNA, although E2 alone significantly attenuated ERβ mRNA. This finding further supports the notion that E2 alone suppresses BCRP, presumably via ERβ, and P4 in combination with E2 induces BCRP, possibly via PRB. Thus P4 and E2 seem to interact through PRB and ERβ for regulation of BCRP. Studies are now in progress in our laboratory to elucidate the molecular mechanisms by which P4 and E2 regulate BCRP expression in BeWo cells through PRB and ERβ.
Similar to the findings of this study, Imai et al. (17) demonstrated downregulation of BCRP by E2 in T-47D and MCF-7 breast cancer cells. In contrast, Ee and colleagues (10, 11) reported stimulation rather than suppression of BCRP by E2 in T-47D and BeWo cells. The reason for this apparent discrepancy is presently unknown. Genetic alterations may occur in cells after prolonged culture. Hence, BeWo cells within only eight passages after purchase were used in this study.
In summary, the present study suggests that 1) P4 and E2, respectively, upregulate and downregulate BCRP expression in BeWo cells; 2) the interaction between P4 and E2, through PRB and ERβ, may play a significant role in the regulation of BCRP in BeWo cells; and 3) steroid hormones, for example P4, may function through a classical or nonclassical PR, or both pathways, in response to specific endocrine status during pregnancy. Further studies are needed to elucidate the molecular mechanisms by which BCRP expression is regulated by P4 and E2 in BeWo cells. Such studies will help explain how pregnancy affects drug distribution across the placenta. It should be pointed out that the BeWo cell line is not exactly the same as the placental trophoblast with respect to the expression of ABC transporters, and therefore, care should be taken when extrapolating the data obtained in this cell line to in vivo human subjects.
We gratefully acknowledge financial support from National Institute of Child Health and Human Development Grant no. HD-044404 (to Q. Mao and J. D. Unadkat) and from the Department of Pharmaceutics, University of Washington.
We thank Drs. Robert W. Robey and Susan E. Bates (NCI) for providing FTC. We acknowledge Dr. Douglas Ross (University of Maryland, Baltimore, MD) and Dr. Virendra B. Mahesh (Medical College of Georgia, Augusta, GA) for their helpful comments on this study. We thank Dr. Ed Kelly, Dr. Carl Ton, Hiuxia Zhang, and the Center for DNA Sequencing and Gene Analysis (Department of Pharmaceutics, University of Washington, Seattle, WA) for technical assistance in real-time PCR and Greg Martin (Keck Imaging Center, Department of Pharmacology, University of Washington) for technical assistance in immunofluorescent confocal microscopy.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2006 by American Physiological Society