Effects of the ABCB1 (1199G > A) Polymorphism on Steroid Sex Hormone-Induced P-Glycoprotein Expression, ATPase Activity, and Hormone Efflux

This study examined how the 1199G > A polymorphism in the ABCB1 gene encoding P-glycoprotein (P-gp) affects the protein’s expression, ATPase activity, and ability to pump female steroid sex hormones out of LLC-PK1 cells. The ABCB1 (1199G) or ABCB1 (1199A) allele was transfected into cells, which were incubated for 48 h with various hormone concentrations, then analyzed by Western blotting to examine expression of P-gp protein and by reverse transcription-polymerase chain reaction (RT-PCR) to examine expression of mRNA. Cells were also compared in terms of their transepithelial permeability to steroid sex hormones in the presence and absence of the specific P-gp inhibitor GF120918. P-gp ATPase activity induced by steroid sex hormones was also assayed. Estriol and ethynyl estradiol up-regulated levels of ABCB1 mRNA in a concentration-dependent manner, with ABCB1 (1199A) mRNA showing greater up-regulation than ABCB1 (1199G) mRNA. Estrone, estriol, and ethynyl estradiol were substrates of both types of P-gp in transepithelial permeability assays, and the ABCB1 (1199A) protein showed a significantly higher net efflux ratio for estrone (13.4 vs. 7.4, p < 0.005), estriol (5.6 vs. 3.3, p < 0.05), and ethynyl estradiol (12.7 vs. 5.3, p < 0.005). Induction of P-gp ATPase activity by ethynyl estradiol and progesterone increased with increasing hormone concentration, and the magnitude of stimulation was greater for ABCB1 (1199A) P-gp than for ABCB1 (1199G) P-gp. These results indicate that the ABCB1 (1199G > A) polymorphism influences steroid sex hormone-induced expression and function of P-gp, which may help to explain inter-patient differences in P-gp-mediated chemotherapy resistance in vivo.

hormone regulation is likely to be important for human chemotherapy, given that it may strongly influence the efficacy of orally-administered synthetic derivatives of estrogens and progestins. Therefore, the present study examined whether the ABCB1 1199G > A polymorphism affects the ability of P-gp to recognize steroid sex hormones, such as estrogen and progesterone, as substrates, as well as the ability of such hormones to up-regulate P-gp expression and activity.

Cell Culture
Stably transfected ABCB1 (1199G)-LLC-PK1 and ABCB1 (1199A)-LLC-PK1 cells, as well as untransfected LLC-PK1 cells, were grown in six-well plates (3 × 10 5 cells) in medium mixed with FBS and A-A. Transfected cultures were cultured in the presence of 2 μg/mL puromycin to select for cells overexpressing P-gp. All three types of cultures were then incubated in the presence or absence of the indicated steroid sex hormones at 37 °C in a moderately humid atmosphere of 5% CO2. Before bidirectional transport studies, transepithelial electrical resistance (TEER) was measured in each well to confirm the confluence and integrity of the cell monolayer; values ranged from 330 to 400 Ω/cm 2 across the three types of cultures.

Flow Cytometry
P-gp expression on the surface of stably-transfected ABCB1 (1199G)-LLC-PK1 and ABCB1 (1199A)-LLC-PK1 cells was determined by flow cytometry as described. Briefly, LLC-PK1 cells were washed twice with ice-cold, 0.22 μm-filtered Hank's buffer containing 3% decomplemented FBS and 20 mM NaN3 (HAFA), then immobilized using 1% formaldehyde for 15 min at room temperature. Cells were rinsed three times with HAFA, then incubated in HAFA containing rabbit anti-P-gp monoclonal antibody (1:10; Abcam, AB129450) for 45 min on ice in the dark. Cells were again washed twice with HAFA and incubated with FITC-conjugated goat anti-rabbit IgG (1:50; Abcam, AB6717) for 30 min on ice in the dark. Cells were re-suspended in HAFA and analyzed by flow cytometry on a FACSCanto II device (BD Biosciences, San Jose, CA, USA).

Western Blot Analysis
Protein (40 μg) was fractionated on 8% polyacrylamide-SDS gels, transferred to nitrocellulose membranes and sealed for 1 h in nonfat dried milk in TBST. Membranes were probed first with rabbit anti-P-gp monoclonal antibody (1:3000; Abcam, AB129450), followed by HR dye-conjugated goat anti-rabbit IgG (1:10,000; Abcam, AB6532); both incubations were for 1 h at room temperature. Antibody binding was detected, and band intensity quantified, using the Odyssey IR imaging system (LI-COR Biosciences, Lincoln, NE, USA).

RT-PCR
ABCB1 (1199G)-LLC-PK1 and ABCB1 (1199A)-LLC-PK1 cell lines were cultured in six-well dishes (1.0 × 10 6 cells/well), and total RNA was extracted using TRIZOL (Fermentas, Walthma, MA, USA). Purity and concentration of RNA extracts were checked using a UV spectrophotometer; extracts were used only if the absorbance ratio (260/280 nm) was 1.8-2.0. Extracts were diluted and stored at −80 °C until they were used to generate cDNA by reverse transcription RT-PCR as described. RT-PCR was performed on the Prism 7700 Sequence Detection System (ABI, Foster City, CA, USA), and it involved one step at 50 °C for 2 min, one step at 95 °C for 10 min, and then 40 cycles at 95 °C for 30 s and 60 °C for 30 s. The sense ABCB1 primer was 5′-ATGTTTCCGGTTTGGAGCCT-3′; the anti-sense primer, 5′-TCCTTCCAATGTGTTCGGCA-3′. SYBR Green/Fluorescein qPCR Master Mix (2X; Fermentas, catalog No. ko242) was used. Levels of ABCB1 mRNA were normalized to levels of β-actin transcript.

Transepithelial Permeability Assay
Bidirectional transport across monolayers of ABCB1 (1199G)-LLC-PK1, ABCB1 (1199A)-LLC-PK1, and untransfected LLC-PK1 cells were assayed as described [15], with a slight modification. Cells were plated in six-well Transwell dishes with inserts and TEER was measured as described in Section 2.3. Then, cells were washed with Hank's Balanced Salt Solution containing 1% FBS and 22.5 mM HEPES (pH 7.4) (HBSS-GH) and incubated with shaking for another 30 min at 37 °C. Transport of potential P-gp substrates (estrone, estriol, and ethynyl estradiol) from the basolateral to apical side of the insert was measured by adding various concentrations of substrate to the basolateral chamber and HBSS-GH to the apical chamber. Cells were incubated with shaking at 37 °C, and 200-μL isometric medium were removed from the opposite side at 0.5, 1, 1.5, and 2 h. The analogous procedure was followed to measure transport from the apical to basolateral side of the insert. After sampling, 200 μL of fresh medium was added to the appropriate chamber to maintain the same initial volume. In some experiments, the HBSS-GH in both chambers was supplemented with 1.0 μM GF120918 to inhibit P-gp-mediated transport.
Steroid sex hormones in the harvested HBSS-GH were extracted with organic solvent, evaporated to dryness and redissolved. The supernatant was analyzed using high-performance liquid chromatography (HPLC). Apparent permeability (Papp) was obtained from the equation where ∆Ǫ/∆t is the linear rate of accumulation of steroid sex hormone in the receiver chamber, C0 represents the initial concentration of hormone, and A is the surface area of the osmotic membrane. Directionality of transport was quantified by calculating the net efflux ratio, defined as the ratio Papp(basolateral → apical)/Papp(apical → basolateral). The minimum efflux ratio for P-gp-mediated efflux was defined to be 2.0.

P-gp ATPase Activity Assay
Membrane fractions were prepared and evaluated for P-gp ATPase activity as described [16], with slight modifications. In parallel, human P-gp and control membranes (Gentest) were also evaluated. Membranes were incubated at 37 °C for 20 min in the presence of buffer (50 mM Tris-MES, 50 mM KCl, 2 mM dithiothreitol, 3 mM MgATP, 2 mM EGTA, 5 mM sodium azide) supplemented with 20 μM verapamil (positive control) or increasing concentrations of steroid sex hormones. The same reactions were set up in parallel in the presence of the ATPase inhibitor sodium orthovanadate (100 μM) as a negative control. 10% SDS was added to the solution in order to terminate the reactions, after which zinc acetate and ammonium molybdate were added to final respective concentrations of 15 and 35 mM. Over the next 20 min, absorbance at 800 nm was continuously monitored at room temperature to track the release of inorganic phosphate. Phosphate amounts were quantified against a phosphate standard curve.

Steroid Sex Hormone-Induced Expression of P-gp Protein
Stably-transfected LLC-PK1 cells expressing the ABCB1 (1199G) or ABCB1 (1199A) allele of P-gp, together with untransfected LLC-PK1 cells as a control, were treated for 48 h with various concentrations of estrone, estriol, and ethynyl estradiol (Figure 2). Cells were lysed and fractionated by SDS-PAGE, then analyzed by Western blotting to examine P-gp levels. All three sex hormones up-regulated the levels of P-gp. In contrast, progesterone, estradiol, or norethisterone did not significantly affect levels of either type of P-gp (data not shown).

Steroid Sex Hormone-Induced Expression of P-gp mRNA
Stably-transfected LLC-PK1 cells expressing the ABCB1 (1199G) or ABCB1 (1199A) allele of P-gp, together with untransfected LLC-PK1 cells as a control, were treated for 48 h with various concentrations of estrogens and progestins. Then total cellular RNA was extracted and levels of P-pg mRNA were quantified using RT-PCR ( Figure 3). All three hormones up-regulated levels of both mRNAs in a concentration-dependent manner. At a concentration of 0.8 μM, estriol treatment up-regulated levels of ABCB1 (1199G) mRNA 6.0-fold (over the control); ethynyl estradiol, 6.5-fold; and estrone, 7.4-fold. The corresponding fold increases in levels of ABCB1 (1199A) mRNA were 10.3, 10.7 and 7.9. Statistical analysis showed that estriol and ethynyl estradiol up-regulated ABCB1 (1199A) mRNA levels significantly more than ABCB1 (1199G) mRNA levels ( Table 1). In contrast, estrone up-regulated the two mRNAs to a similar extent.   Results are the mean ± SD of three replicates. * p < 0.05, ** p < 0.001 between the two mRNAs.
Norethindrone at 0.2 μM up-regulated ABCB1 (1199G) mRNA 3.2-fold and ABCB1 (1199A) mRNA 2.8-fold, which was greater than the up-regulation observed at higher concentrations. This may reflect variations in the amount of template cDNA. Progesterone at 0.4 μM up-regulated levels of both types of mRNA three-fold, while estradiol caused only slight increases in the level of either mRNA.
These results suggest that estrogens and progestins can regulate ABCB1 transcription, and that this regulation is significantly affected by the 1199G > A polymorphism in the case of some hormones but not others.

Transepithelial Permeability of LLC-PK1 Monolayers to Steroid Sex Hormones in the Presence of ABCB1 (1199G)-or ABCB1 (1199A)-Encoded P-gp
Natural and synthetic estrogen and progesterone were evaluated to determine whether they can be transported by ABCB1 (1199G) and ABCB1 (1199A) P-gp across an LLC-PK1 cell monolayer system. Estriol at 0.8 μM was transported in both basolateral-to-apical and apical-to-basolateral directions (Figure 4), and Papp from the basolateral side was much higher than that from the apical side, giving a net efflux ratio of 3.3 for cells expressing ABCB1 (1199G) P-gp and 5.6 for cells expressing ABCB1 (1199A) P-gp ( Table 2). The corresponding net efflux ratio was only 1.1 in untransfected cells. These results suggest that estriol is a good substrate for both types of P-gp.
Transport of estradiol and estrone was greater in cells expressing either type of P-gp than in control cells, and net efflux ratios were >2.0, suggesting that these hormones are also good substrates for the transporter ( Table 2). The net efflux of ethynyl estradiol was significantly higher in cells expressing ABCB1 (1199A) P-gp than in cells expressing ABCB1 (1199G) P-gp (12.7 vs. 5.3, p < 0.005). The same was true for the net efflux of estrone (13.4 vs. 7.4, p < 0.005). These results suggest that the change from 1199G to A leads to significantly greater ability to transport ethynyl estradiol and estrone.
Net efflux ratios were much lower than two for progesterone, norethindrone, and estradiol. Thus, these hormones are unlikely to be P-gp subtrates. Values are mean ± SD of three replicates. * p < 0.05, ** p < 0.005 between cells expressing the ABCB1 (1199G) or ABCB1 (1199A) allele.
To confirm the results for ethynyl estradiol, estrone, and estriol, we re-measured net efflux ratios of these three hormones in the presence of the specific P-gp inhibitor GF120918 (1.0 μM). Addition of GF120918 to both apical and basolateral chambers completely inhibited efflux of all three hormones in cells expressing the ABCB1 (1199G) or ABCB1 (1199A) allele.

Steroid Sex Hormone-Induced ATPase Activity in P-Gp
Membranes from LLC-PK1 cells expressing the ABCB1 (1199G) or ABCB1 (1199A) allele were used to examine whether steroid sex hormones affect P-gp ATPase activity. Both ethynyl estradiol and progesterone (0.1-0.8 μM) induced ATPase activity in a concentration-dependent manner in cells expressing the ABCB1 (1199G) or ABCB1 (1199A) allele ( Figure 5). ATPase induction was much higher in cells expressing the ABCB1 (1199A) allele. Control experiments in the presence of vanadate confirmed that the observed ATPase activity was due to P-gp. None of the other hormones appreciably stimulated P-gp ATPase activity. P-gp ATPase activity in the absence of any hormones was similar between cells expressing the ABCB1 (1199G) or ABCB1 (1199A) allele. In the presence of progesterone or ethynyl estradiol, significantly greater ATPase activity was induced in cells expressing ABCB1 (1199G) P-gp than in cells expressing ABCB1 (1199A) P-gp (Table 3). These results suggest that the change from 1199G to A leads to significantly greater ATPase induction by ethynyl estradiol and progesterone. Results are the mean ± SD of three replicates. * p < 0.05, ** p < 0.005 between ABCB1 (1199G) and ABCB1 (1199A) P-gp.

Discussion
Flow cytometry experiments indicated that surface expression of P-gp was similar in LLC-PK1 cells transfected with either the ABCB1 (1199G) or ABCB1 (1199A) allele. This suggests that the results obtained in this study do not reflect differences in baseline P-gp expression between the two cell lines. As a result, we were able to compare steroid sex hormone net efflux for both ABCB1 (1199G) and ABCB1 (1199A) P-gp.
Earlier studies [17,18] reported that P-gp polymorphisms played an important role in the development of steroid resistance in children with nephrotic syndrome. Cho et al. [19] showed an influence of the G3435T polymorphism in the ABCB1 gene on the therapeutic efficacy of corticoids in patients with Crohn's disease. Here we provide evidence that the ABCB1 1199G > A polymorphism significantly affects how much steroid sex hormones up-regulate P-gp mRNA and protein, and how much they activate P-gp ATPase activity in vitro. However, the influence of ABCB1 1199G > A polymorphism on hormones up-regulation P-gp mRNA, protein and ATPase activity in vivo were not reported and need further investigations. These findings show that modulation of P-gp activity by steroid sex hormones may give rise to different pharmacokinetics and outcomes of chemotherapy in patients, depending on their genotype at the ABCB1 1199G > A polymorphism. Our results provide testable hypotheses for explaining why the ABCB1 (1199A) allele displays stronger resistance to anticancer agents compared to the 1199G allele [12,19]. The strong effects of the 1199G > A substitution in the present study are consistent with the fact that it changes Ser400 to Asn near the substrate binding site in P-gp [20,21].
Steroid hormones and their metabolites are already known to regulate expression and activity of ABCB1 (1199G) P-gp, but whether estrogen and progesterone can regulate ABCB1 (1199A) P-gp is unclear. Here we provide evidence that estriol, ethynyl estradiol, and estrone are substrates of wild-type and variant P-gp. In addition, ethynyl estradiol and progesterone significantly stimulate ATPase activity in both types of P-gp. The present in vitro study justifies in vivo experiments to elucidate in greater detail how these hormones modulate P-gp expression and activity, and how this modulation is influenced by various ABCB1 polymorphisms. For example, our results may be relevant to a report linking the ABCB1 G3435T polymorphism to efficacy of corticoid therapy in patients with Crohn's disease [22].
In our bidirectional transport assay in parental LLC-PK1, ABCB1 (1199G)-LLC-PK1 and ABCB1 (1199A)-LLC-PK1 cells, ABCB1 (1199A) P-gp gave much higher net efflux ratios for estriol, ethynyl estradiol, and estrone than did ABCB1 (1199G) P-gp, suggesting that the three hormones are better substrates for the variant protein. This difference between the P-gp proteins reflects, in part, differences in how much their expression is up-regulated by the three hormones. Progesterone and norethindrone, in contrast, do not appear to be substrates for wild-type or variant P-gp. Repeating the transport assays in the presence of the specific P-gp inhibitor GF120918 led to net efflux ratios below 2.0, defined as the lower limit of P-gp transport.
The present work is the first, to our knowledge, that demonstrates several effects of the 1199G > A SNP on P-gp-dependent efflux of steroid sex hormones. In-depth studies must be performed to demonstrate the complex interactions between steroid sex hormones and pleomorphic P-gp, as well as explore how these interactions may affect clinical outcomes. The ability of steroid hormones to induce P-gp expression and transport activity in human tissue needs to be examined further, such as during the menstrual cycle, when changes in hormone levels alter P-gp expression and activity. Greater insights into how hormones influence the pharmacokinetics and P-gp-mediated efflux of drugs may improve our ability to predict complications and poor efficacy and, thereby, personalize treatment, particularly for women on steroid hormone therapy.

Conclusions
This is the first study to evaluate the effects of the ABCB1 (1199G > A) polymorphism on steroid sex hormone-induced P-gp expression, ATPase activity, and hormone efflux. Our findings suggest that estrone, estriol, and ethynyl estradiol are all substrates for both two types of P-gp, and these hormones are able to stimulate the expression levels of P-gp proteins and mRNA in vitro. Two types of P-gp ATPase catalytic activity were induced by ethynyl estradiol and progesterone, but also influenced by ABCB1 (1199G > A) polymorphism. Therefore, it should be widely considered that SNP may be a important factor to affect the pharmacokinetics and efficacy of chemotherapeutics in patients.