Expression and Activity of p-Glycoprotein Elevated by Dexamethasone in Cultured Retinal Pigment Epithelium Involve Glucocorticoid Receptor and Pregnane X Receptor

Abstract

Purpose.: To investigate whether dexamethasone has an effect on functional expression of p-glycoprotein in cultured human RPE and, if so, whether this occurs through interaction with glucocorticoid receptor (GR) and pregnane X receptor (PXR).

Methods.: The human RPE D407 was treated with increasing concentrations of dexamethasone and/or RU486 for various time periods up to 24 hours. Treated cells were collected for cell viability, expressions of p-glycoprotein and PXR, and rhodamine 123 accumulation assays. GR expression plasmid and rifampicin were chosen to investigate the relationship of GR/PXR activation and p-glycoprotein expression.

Results.: Significant increases in p-glycoprotein, as indicated by mRNA and protein levels, as well as by functional activity, were induced within 12 hours of dexamethasone treatment, persisted as long as 24 hours, and were dose-dependent and attenuable with coculture of RU486. In parallel, a dose-dependent upregulation of PXR was notable at both mRNA and protein levels by 24 hours of dexamethasone treatment, and was partially reversible with RU486 coculture. Additionally, transfection of GR expression plasmid increased the transitional expressions of PXR and p-glycoprotein in untreated cells, and enhanced PXR transcriptional expression in dexamethasone-treated cells. Further, PXR silencing inhibited the dexamethasone-induced p-glycoprotein alterations; however, rifampicin had no apparent effects on the dexamethasone-induced p-glycoprotein alterations.

Conclusions.: Our results suggest for the first time that expression and activation of p-glycoprotein involve GR and PXR in human RPE.

Introduction

RPE plays an essential role in protecting neural tissues from toxic materials and in maintaining vision and neural function in the retina by forming the outer blood-retinal barrier (BRB). It has been demonstrated that the outer BRB not only regulates the ionic environment of the subretinal space, secretes factors for structural integrity of the retina, phagocytizes shed outer segments of photoreceptors, and participates in the visual cycle, ¹ but also limits vitreal penetration of drugs administered by the systemic and transscleral routes. ^2–4 Efflux transport systems provide further barriers for the retina, by actively removing cytotoxic drugs and specific xenobiotic compounds from the retina and transferring them back into the systemic circulation. ^5,6 P-glycoprotein (P-gp), a 170-kDa protein encoded by the multiple drug resistance human MDR1 gene, is a member of the ABC superfamily of energy-dependent transport systems. ⁷ As a well-characterized efflux transporter, P-gp is strongly expressed by retinal vascular endothelial cells ⁸ and has recently been identified in human RPE. ⁹ P-gp displays broad specificity, accepting many structurally, functionally, and mechanistically unrelated compounds, ¹⁰ and its role in limiting drug penetration across biological barriers is well established. Although efflux pumps such as P-gp are best known as major barriers to drug delivery in brain and gut, ^11–13 they have also received attention for their potential roles in barrier maintenance at the outer BRB. Kennedy and Mangini ⁹ have demonstrated that P-gp is expressed on both the apical as well as basal membranes of the RPE cells. P-gp on the RPE cells may thus affect permeation of substrates from the vitreous humor into the systemic circulation and vice versa ^3,14,15 and could be a major factor behind the inability of systemic and transscleral routes of administration to generate and maintain therapeutic concentrations of P-gp substrates in the retina. Thus, factors/agents that can modulate the efflux activity of RPE P-gp could probably alter ocular pharmacokinetics of P-gp substrates; however, published reports of the role and the mediation of P-gp at the outer BRB remain scarce. 

Because of its rapid and beneficial clinical effects, dexamethasone, a synthetic glucocorticoid, is widely used as an immunosuppressive and anti-inflammatory agent in the management of ocular diseases secondary to the breakdown of BRB, such as age-related macular degeneration, vitreoretinopathies, and uveitis. ^16–18 Dexamethasone has been shown to differentially alter the expression of efflux transporters such as P-gp in the barriers of brain, gut, liver, and lung ^19–22 ; however, it remains unclear whether dexamethasone also has an effect on the functional expression of efflux transporters at the BRB. Because dexamethasone is often a concomitant agent for the treatment of ocular disease secondary to the BRB breakdown, ^17,18,23 it is important to determine how dexamethasone affects the functional expression of efflux transporters in RPE cells. Using cultured human RPE cells, we set to investigate the effect of dexamethasone on the expression and functional activity of P-gp in cultured human RPE cells. In addition, because glucocorticoid receptor (GR) and pregnane X receptor (PXR) are known major nuclear transcription factors in regulating drug efflux transporters and metabolic enzymes when adjuvant therapy with corticosteroids, ^19,24–28 and on activation these nuclear transcription factors can stimulate or inhibit gene expressions of most drug transporters by binding to relative response elements on DNA, we also investigated the relative contributions of GR and PXR in regulating the dexamethasone-induced alteration of P-gp in the present culture system. 

Materials and Methods

Cell Culture and Treatment

The human RPE cell line D407 was generously provided by Richard Hunt (Department of Immunology and Pathology, University of South Carolina Medical School, Columbia, SC). These cells were shown to possess most metabolic and morphologic characteristics of RPE cells in vivo ²⁹ and express functional P-gp. ⁹ D407 cells were grown in Dulbecco's modified Eagle's medium (Gibco, Grand Island, NY) supplemented with 2 mM glutamine, 100 units/mL penicillin, 0.1 mg/mL streptomycin, and 10% fetal bovine serum. Cells were maintained in a 37°C humidified incubator in the presence of 5% CO₂. The culture medium was replaced with fresh medium every 2 to 3 days. Phase-contrast microscopy confirmed that confluent cultures formed a monolayer with the typical morphological characteristics of cultured human RPE cells. 

Treatment of D407 cells for all assays consisted of incubation with fresh medium containing different concentrations of dexamethasone (Sigma, St. Louis, MO) or RU486 (Sigma), or equivalent volumes of vehicle (ethanol, final concentration less than 0.04%). For the transfection experiments, six-well tissue culture plates were used and several plasmid and small interfering RNA (siRNA) vectors were transfected into the cultured D407 cells, respectively. In the P-gp function experiment, cells were pretreated with the dexamethasone, RU486, or control or PXR siRNA, and then subjected to rhodamine 123 accumulation assays. 

Cell Viability Assay

The cell survival assay of RPE cells (10⁴ cells/mL each) was performed up to 24 hours in all experiments. Cells were stained with Trypan blue, and percentages of viable cells were calculated by dividing the number of live cells by the total cell number. Changes in the percentages of viable cells were calculated on the basis of the number of cells added to the culture of hour 0. 

Glucocorticoid Receptor Expression Plasmid Transfection

In humans, the GR has two major forms: GR-α and -β. The β-form is known to function as a dominant negative regulator against the α-form. ³⁰ Throughout this article, unless otherwise specified, the GR is used to describe the α-form. The GR expression plasmid was generated by replacing the DNA fragment between NheI and XbaI sites containing the coding sequence of Renilla luciferase of the pRL-SV40 vector (Promega, Madison, WI) with a fragment containing the distal 940 bp of the human GR-α coding region and the 2322 bp human GR-α 3′UTR. D407 cells were cultured in Waymouth medium and transfected using TransPass D1 Transfection Reagent (New England Biolabs, Hercules, CA). Transfection mixtures consisted of Waymouth medium, empty plasmid, or nuclear receptor expression plasmid and TransPass D1 at 2 mL, 5 μg, and 5 μL, respectively. Transfection continued for 6 hours, and the medium was changed. The cells were then treated with dexamethasone at various concentrations in fresh medium after a further 24-hour incubation. Luciferase and β-galactosidase assays were performed according to the specifications of the manufacturer (Promega). 

PXR Silencing Assay

The human PXR siRNA cocktail (siFECTOR; B-Bridge International Inc., Mountain View, CA) contains 3 siRNAs: first sequence ggacaaggccacuggcuau (sense), auagccaguggccuugucc (antisense), second sequence agccgacaguggcgggaaa (sense), uuucccgccacugucggcu (antisense), and third sequence gggccaagacagauggaca (sense), uguccaucugucuuggccc (antisense). Negative control cocktail (Cat# C6A-0126) and liposome for siRNA transfection were also products from B-Bridge International Inc. Cells were transfected with PXR siRNA or control siRNA using siFECTOR for 48 hours with a change of fresh medium at 24 hours according to the manufacturer's protocol. The transfected cells then underwent the dexamethasone treatment. 

Quantitative Real-Time PCR

Total cellular RNA was extracted from D407 cells using TRIzol reagent (Invitrogen, Guangzhou, China) according to the manufacturer's instructions. Purified RNA was reverse-transcribed with an ABI PRISM 7000 Sequence Detection System using SYBR Green I as the reporter dye (TaKaRa Corp., Dalian, China). The comparative Ct method was used whereas the relative quantity of the target gene mRNA, normalized to GAPDH as internal control and relative to the calibrator, was expressed as fold change = 2^−ΔΔCt. The primers used for amplification were from different exons and their sequences were as follows: MDR1, 5′-TGGCACCCAGCACAATGAA-3′ and 5′-CTAAGTCATAGTCCGCCTAGAAGCA-3′; PXR, 5′-GGCCACTGGCTATCACTTCAA-3′ and 5′-GTTTCATGGCCCTCCTGAAA-3′; GAPDH, 5′-GCA CCG TCA AGG CTG AGA AC-3′ and 5′-TGG TGA AGA CGC CAG TGG A-3′. Duplicate PCR reactions were tested using the following amplification protocol: 95°C for 10 seconds followed by 30 cycles at 95°C for 5 seconds and at 60°C for 31 seconds.  

Western Blot Analysis

Proteins were separated using 12% SDS-polyacrylamide gels. The resolved proteins were transferred electrically to PVDF membranes and incubated with 5% skim milk in Tris-buffered saline with 0.05% Tween-20. The membrane was probed with primary antibodies for P-gp (C219, Calbiochem, San Diego, CA), or PXR (A-20, Santa Cruz Biotechnology, Santa Cruz, CA), or ACTB (Sigma) in blocking buffer overnight at 4°C. It was then incubated with horseradish peroxidase–conjugated secondary antibodies (Santa Cruz Biotechnology) for 1 hour at room temperature. The proteins were detected with chemiluminescence reagents. The density of the bands was quantified using a laser densitometer (ATTO densitograph 4.0; Fujifilm, Tokyo, Japan). For P-gp expression, mouse fibroblast cell line NIH/3T3 (American Type Culture Collection, CRL-1658) known for expressing low/no P-gp was used as a negative control. Each experiment was repeated three times. 

Rhodamine 123 Accumulation Assay

P-gp function was determined by rhodamine 123 accumulation assay. Fluorescence intensity of intracellular rhodamine 123 (Invitrogen) was determined by flow cytometry. D407 cells (5 × 10⁵) were collected by trypsinization, washed in PBS, and then incubated for 1 hour at 37°C in the dark with rhodamine 123. After 1-hour incubation, cells were washed and fed with rhodamine 123–free culture medium. To exclude dead cells, cells were stained with 5 μg/mL propidium iodide (PI), a nonpermanent dye that cannot stain living cells, for 10 minutes. The cultured cells were pelleted at 200g for 5 minutes, washed twice in PBS, and then analyzed immediately by flow cytometry analysis using a BD FACS AriaTM flow cytometer and BD FACSDiVa software (Becton Dickinson, Mountain View, CA). Photomultiplier settings were adjusted to detect green fluorescence (λem = 488 nm) of rhodamine 123, and detect red fluorescence of PI at an emission wavelength of 620 nm on the filter detector. In each experiment, at least 20,000 events were analyzed. All experiments used six wells per condition and were repeated on two to three separate occasions. The mean fluorescence intensity in arbitrary units was used for data presentation. 

Statistical Analysis

All data were expressed as mean ± SE. Comparisons between groups were performed using one-way ANOVA and a post hoc Tukey test, and differences were considered statically significant at P < 0.05. The individual studies described in the results section are representative of at least three independent studies. 

Results

Upregulation of P-gp by Dexamethasone Is Dose- and Time-Dependent, and Associated with GR

To assess whether dexamethasone altered drug transporters at the outer BRB, a human RPE cell line (D407) was cultured with increasing concentration of dexamethasone (10⁻⁷–10⁻³ M) for 24 hours, and was then assessed for P-gp expression in mRNA and protein levels. As shown in Figure 1A, the transcript levels of MDR1 were significantly induced by treatment with dexamethasone in a concentration-dependent manner, and a maximal induction was attained with 10⁻⁵ M dexamethasone (P < 0.01 compared with cells incubated in controls). No further increase was observed with higher dexamethasone concentrations (up to 10⁻³ M). We thus used the concentration of 10⁻⁵ M dexamethasone in the following steps of our study. To assess the kinetics of upregulation of P-gp, D407 cells were cultured with 10⁻⁵ M dexamethasone for varying time periods up to 24 hours. An increase in transcriptional levels of the protein (MDR1 mRNA) was observed as early as 12 hours, and was elevated as long as 24 hours (Fig. 1B). As shown in Figure 1C, the alterations of translational levels were consistently occurred. Compared with negative controls (ethanol), only a trace of P-gp was detected in untreated control cells, while a strong signal was found in dexamethasone-treated cells. Consistently, the protein level of P-gp was also increased in dexamethasone-treated cells, and a concentration-dependent trend was found within a certain range of concentration (10⁻⁷ to 10⁻³ M, Fig. 1D). This correlated well with the mRNA levels of MDR1 shown in Figure 1A. To exclude the attribution of increased viability of cultured cells, cell survival assay showed that treatment with dexamethasone concentrations up to 10⁻³ M for 24 hours had no effect on RPE cell viability (data not shown). 

Figure 1. 

 

Upregulation of P-gp by dexamethasone is dose- and time-dependent, and associated with GR. (A) The levels of MDR1 mRNA were determined by quantitative PCR. D407 cells were cultured with increasing concentration of dexamethasone (10⁻⁷–10⁻³ M) in the absence or presence of RU486 (10⁻⁷ M), and then total RNA was prepared for PCR. GAPDH was used as the internal loading control. (B) Time course study on the levels of MDR1 mRNA after dexamethasone treatment. (C) Time course study on the expression of P-gp after dexamethasone treatment. NIH/3T3 cell line was used as a negative control. (D) The presence of P-gp was detected by Western blots in cell lysates obtained from the cultured D407 cells treated with dexamethasone and RU486. (E) Transfection of GR expression plasmid elevated the expression of P-gp in untreated cells. The expression of P-gp was detected by Western blots in cell lysates from the cultured D407 cell transfected with empty or GR expression plasmid. For protein expression, ACTB was used as the internal loading control. Each column represents the mean ± SE. Data presented in this figure were assembled from three independent experiments. *P < 0.05; **P < 0.01 versus untreated controls; #P < 0.05 vs. 10⁻⁷ M DEX + RU486; $P < 0.05 vs. 10⁻⁵ M DEX; ▵P < 0.05 vs. 10⁻³ M DEX. DEX, dexamethasone; CTL, negative control.

The GR is the major receptor transmitting the signal of glucocorticoids, particularly at low concentrations (< 10⁵ M). ^26,27 To investigate the possible role of GR in the P-gp alterations, we then examined indirectly whether RU486, a typical GR antagonist, could inhibit P-gp after dexamethasone exposure. As shown in Figures 1A and 1D, the addition of RU486 (10⁻⁵ M) partially reversed the elevated expression of P-gp at both the mRNA and the protein levels. We next overexpressed GR directly with plasmid transfection, and found that the GR overexpression did increase the expression of P-gp (Fig. 1E). These results indicate that GR is associated with the dexamethasone-induced P-gp alteration. 

Dexamethasone-Induced Activity of P-gp Was Partially Reversed with RU486

Dexamethasone has been demonstrated to functionally regulate the activity of P-gp at the blood brain barrier. ^31,32 To assess the effects of dexamethasone on functional capacity of P-gp at the outer BRB, D407 cells were next incubated with rhodamine 123 and the intracellular rhodamine 123 concentration was measured by flow cytometry analysis. Rhodamine 123, a fluorescent substrate for efflux transporters, has been used as a typical probe to assess in vitro and in vivo P-gp function in various multidrug-resistant cells and various normal tissues including the human outer BRB. ^5,9 After dexamethasone treatment (10⁻⁵ M), we detected a significant decrease in the fluorescence intensity of intracellular rhodamine 123, indicative of an increased P-gp function. As expected, the rhodamine 123 retention was increased with coculture of RU486 when compared with dexamethasone alone, but did not fully return to the level of the control, untreated cultures (Fig. 2). These results are consistent with previous findings at the blood brain barrier after dexamethasone administration. ¹⁹  

Figure 2. 

 

Dexamethasone-induced activity of P-gp involves GR in cultured pigment epithelium. Cells were treated with ethanol (control) or dexamethasone (10⁻⁶ M) alone or in combination with RU486 (10⁻⁶ M) for 24 hours. The activity of P-gp was determined by measuring intracellular rhodamine 123 accumulations in cells. Following the dexamethasone or RU486 treatment, cells were then loaded with rhodamine 123 (10 μg/mL). The mean fluorescence intensity of intracellular rhodamine 123 was determined by flow cytometry. Data are means ± SE of three independent experiments. **P < 0.01 versus controls. CTL, control; DEX, dexamethasone.

Upregulation of PXR by Dexamethasone Is Dose-Dependent, and Involves GR

To investigate possible involvement of PXR in the present culture system, we first assessed the expression of PXR after incubation with various concentrations (10⁻⁷ to 10⁻⁵ M) of dexamethasone in the absence and presence of RU486 (10⁻⁵ M). Treatment with dexamethasone concentration-dependently increased expression of PXR both at mRNA (Fig. 3A) and at protein levels (Fig. 3B). Coculture of RU486 partially antagonized the dexamethasone-induced upregulation of PXR (Fig. 3C). Because RU486 is also a known agonist for PXR, ³³ the net effect of attenuated expression of PXR suggests that there must be some mechanism other than the agonist activity of RU486. Its anti-GR property is certainly a possibility. 

Figure 3. 

 

Upregulation of PXR by dexamethasone is dose-dependent, and associated with GR. (A) The levels of PXR mRNA were determined by quantitative PCR. D407 cells were cultured with increasing concentration of dexamethasone (10⁻⁷–10⁻⁵ M) for 24 hours, and then total RNA was prepared for PCR. GAPDH was used as the internal loading control. Each column represents the mean ± SE. **P < 0.01 versus untreated controls. (B) The expression of PXR was detected by Western blots in cell lysastes obtained from the cultured D407 cells treated with dexamethasone. ACTB was used as the internal loading control. (C) RU486 partially attenuated the expression of PXR induced by dexamethasone (10⁻⁵ M). All other details are described in Materials and Methods. Data presented in this figure were assembled from three independent experiments. DEX, dexamethasone.

To further establish the role of GR mediation, we next evaluated if the upregulation of GR per se is able to produce the PXR induction in cultured RPE cells. As shown in Figure 4, the GR expression plasmid transfection induced a significant increase in the expression of PXR at both mRNA (Fig. 4A) and protein levels (Fig. 4B). Additionally, when the GR expression plasmid was cotransfected, dexamethasone treatment at 10⁻⁶ up to 10⁻⁵ M considerably caused a maximum 40-fold increase in PXR expression over the level attained by empty plasmid (Fig. 4A). Thus, the effect of GR transfection in untreated cells and the synergistic effect with dexamethasone exclude the possibility that GR is just a cofactor, and further support a GR-mediated PXR expression mechanism in the present culture system. 

Figure 4. 

 

GR expression plasmid transfection elevated PXR expression. (A) Transfection of GR expression plasmid elevated PXR translational expression in untreated cells. The expression of PXR was detected by Western blots in cell lysastes obtained from the cultured D407 cell transfected with empty or GR expression plasmid. ACTB was used as the internal loading control. Data presented in this figure were assembled from three independent experiments. (B) Cotransfection of GR expression plasmid enhanced dexamethasone-induced expression of PXR at transcriptional level. D407 cells were transfected with control vector plasmid (empty plasmid) or GR expression plasmid. Six hours later, cells were then treated with dexamethasone at 10⁻⁷ to 10⁻⁵ M for 24 hours. Expression of the PXR mRNAs was evaluated by quantitative PCR. Each mRNA was normalized to the level of GAPDH and is shown relative to those in empty plasmid-transfected controls. Each column represents the mean ± SE. *P < 0.05; **P < 0.01 versus transfected with empty plasmid untreated cells; #P < 0.05 versus transfected with GR expression plasmid untreated cells; $P < 0.05 versus transfected with empty plasmid treated 10⁻⁷ M dexamethasone cells; ▵P < 0.05 versus transfected with empty plasmid treated with 10⁻⁶ M dexamethasone cells. DEX, dexamethasone.

PXR Silencing Attenuated the Upregulation of P-gp by Dexamethasone

In mammalian systems, PXR has been recognized as a master xenobiotic-sensor that can upregulate the functional expression of drug transporters including P-gp. ^34,35 To determine the role of PXR in the dexamethasone-induced alterations of P-gp, a PXR siRNA knockdown experiment was then performed. The efficacy of PXR siRNA for the knockdown of PXR expression in D407 cells was confirmed by Western blots (Fig. 5A). In cells cotransfected with PXR siRNA, no significant increases were seen in P-gp expression (Figs. 5B, 5C) and activity (Fig. 5D) in the presence of dexamethasone, when compared with cells treated with control siRNA, suggesting a role of PXR in regulating the alteration of P-gp by dexamethasone. 

Figure 5. 

 

PXR silencing attenuated the dexamethasone-induced P-gp alteration. (A) PXR expression in D407 cells was inhibited by PXR siRNA. Cells were transfected with PXR or control siRNA for 48 hours. PXR and ACTB protein levels in whole cell extracts were determined by Western blots using anti-PXR and -ACTB antibodies. Each column represents the mean ± SE. (B) Representative Western blots and (C) graphic analysis show that P-gp expression was inhibited by PXR siRNA. After transfected with PXR or control siRNAs and treated with dexamethasone or ethanol (control), cells were lysates for Western blots using anti–P-gp or –ACTB antibodies. ACTB was used as loading control. (D) P-gp activity was inhibited by PXR siRNA. Data presented in this figure were assembled from three independent experiments. **P < 0.01 versus untreated controls transfected with control siRNA; ##P < 0.01 versus untreated controls transfected with PXR siRNA. DEX, dexamethasone.

Dexamethasone-Induced P-gp Alteration Is Not Due to Direct Activation on PXR

Although the studies with RU486 and transfection of GR collectively establish that dexamethasone induces PXR through a GR mechanism, we cannot exclude the possibility that dexamethasone also directly activates PXR expression, which therefore constitutes the aim of our next experiment. Using rifampicin, a typical ligand of human PXR, we found that the activation of PXR did increase both the transcriptional and transitional levels of P-gp in cultured RPE cells. This result is consistent with earlier observations using rifampicin in cell culture models of intestine, ³⁶ liver, ³⁷ and lymphocytes. ³⁸ However, the coculture of rifampicin failed to enhance the dexamethasone-induced expression of P-gp at either mRNA or protein levels (Fig. 6). If the P-gp response to dexamethasone is due to a direct activation on PXR, not of GR, it is anticipated that the addition of ligand of PXR will cause an enhanced response to the dexamethasone treatment. Therefore, our findings suggested that dexamethasone acts through an indirect PXR-mediated activation of P-gp in the present culture system. Taken together, a GR-PXR–dependent regulation mechanism of P-gp is presented in this study. 

Figure 6. 

 

Rifampicin does not influence the dexamethasone-induced expression of P-gp. (A) The expression of MDR1 mRNA was detected by quantitative PCR. D407 cells were treated with dexamethasone (10⁻⁷ to 10⁻⁵ M) in the absence or presence of rifampicin (10⁻⁷ M) for 24 hours, and then total RNA was prepared for PCR with GADPH as internal control. Each column represents the mean ± SE. *P < 0.05; **P < 0.01 versus untreated controls. (B) The expression of P-gp was detected by Western blots in cell lysastes obtained from the cultured D407 cell treated with dexamethasone and rifampicin. ACTB was used as the internal control. Data presented in this figure were assembled from three independent experiments. CTL, control; DEX, dexamethasone.

Discussion

In recent years, efflux transport proteins that serve as barriers to substrate distribution into organs, ³⁹ tissues, ⁴⁰ or cellular spaces ⁴¹ have garnered significant attention. Although a variety of efflux transport proteins are expressed in mammalian tissues, the most well-characterized, and perhaps the most important, of these systems is P-gp. ⁷ At the outer BRB, dexamethasone is frequently used for antiproliferative and/or anti-inflammatory effects. ^42,43 However, dexamethasone is also a physiologic substrate of P-gp, and a recognized inducer of drug efflux proteins including P-gp. ^20,44 The fact that dexamethasone-induced P-gp plays a critical role in barrier cells ^19–22 led us to address it particularly in human RPE cells. The main finding of this study is that dexamethasone significantly elevated the functional expression of P-gp in cultured human RPE cells. 

This finding may have significant clinical implications. Dexamethasone-induced alterations in barrier transporter levels can lead to altered pharmacokinetics and variable drug disposition of concomitant chemotherapeutic agents. More specifically, upregulation of a drug-efflux pump, such as P-gp, at the outer BRB may further restrict the penetration of therapeutic agents targeted to specific areas of the retinal neurons and photoreceptor cells (rods and cones), and therefore decrease their therapeutic efficacy across the barrier. This affects numerous drugs that are substrates for metabolizing enzymes and efflux transporters. On the other hand, however, P-gp is also responsible for the efflux of neurotoxic chemicals and metabolites from the retina, and increased pump expression should provide increased neuroprotection. The consequences of adjuvant therapy, such as the use of corticosteroids, have not been extensively investigated to determine whether such treatment would be detrimental to the penetration of other drugs. ¹⁹ Our finding suggests that corticosteroids may discount the effectiveness of concomitant drugs that are required to penetrate the outer BRB. Ineffective therapy or even therapy resistance could be the outcome. In the context of this present study, it is noteworthy to point out that the regulation of dexamethasone for P-gp is performed differentially in a tissue- or cell-specific manner, because some literature reported that dexamethasone induced increases of P-gp in brain, small intestine, and lung, ^{19–21,31,36,45} whereas others reported no change or decreased expression in the colon, kidney, and liver. ^21,22 Another in vivo study showed that viral protease inhibitor ritonavir (20 mg/kg) and dexamethasone (80 mg/kg), once daily for 3 days, induced a similar upregulation of P-gp in both the blood brain barrier and the gut. ²⁰  

Dexamethasone is known to modulate the expression of a multitude of genes involved in drug metabolism and disposition via two major receptors: GR and PXR. GR, one of the major members of the steroid nuclear receptor family, is a ligand-induced transcription factor, and ligand-activated GR is translocated from the cytoplasm to the nucleus, wherein it influences gene transcription by binding at specific glucocorticoid response elements or interacts with some transcription factors. ^36,46 GRs are widely distributed in mammalian tissues and have been demonstrated in human RPE cells. ⁴⁷ To investigate the possible role of GR in the alterations of P-gp by dexamethasone in cultured human RPE, we found here that the dynamic alterations of P-gp by dexamethasone could be attenuated with the GR antagonist RU486. This suggests that dexamethasone induces expression and activity of the efflux transporter, P-gp, through a direct interaction with GR. 

We have also shown that the mechanism does involve PXR, which is well known for its role in regulating the expressions of drug metabolizing enzymes and efflux transporters. It has been demonstrated that PXR is activated by a number of xenobiotics and corticosteroids. ⁴⁸ More specifically, PXR can mediate the dexamethasone-upregulated P-gp expression and transport function at the blood brain barrier. ^12,34 To date, studies aimed at elucidating the precise mechanism of efflux transporter activation by glucocorticoids have predominantly focused on drug-activated nuclear receptors, including PXR, CAR, VDR, and GR, particularly the cross-talk between these pathways. ^{12,19,27,34,48–51} In the present study, however, our GR antagonist and transfection experiments suggest a GR-mediated PXR expression mechanism. Consistent with this finding, similar regulatory cascades have been shown to exist in hepatocytes ^26,27 and at the blood brain barrier. ¹⁹ However, in addition to the GR-PXR signaling pathway, Narang et al. ¹⁹ described a direct interaction between dexamethasone and PXR involved in the transporter expression at the blood-brain barrier. Pascussi et al. ^26,27 found that only low concentrations (nanomolar) of dexamethasone activated the GR-PXR pathway, whereas higher concentrations (micromolar) directly activated PXR in hepatocytes. In the present culture system, however, although the GR-PXR pathway is confirmed, the absence of synergistic effect from the addition of ligand of PXR does not support the possibility of direct action on PXR by dexamethasone. A possible explanation for this discrepancy is that P-gp in different tissue barriers may have tissue-specific characteristic responses to the use of corticosteroids, such as dexamethasone. Our findings help clarify and provide evidence that these two nuclear receptors are functionally expressed at the outer BRB and can serve as potential sites for drug-receptor interactions and regulation of drug transporters and metabolic enzymes.  

As an established xenobiotic receptor, many of PXR's target genes are involved in the biotransformation and homeostasis of both endogenous and exogenous compounds. ²⁵ This may affect both normal and diseased physiological processes. As such, PXR might be a viable therapeutic target for various diseases. ^25,52 Our finding of the restored expression of P-gp after dexamethasone suggests that PXR inhibition has a beneficial effect on the pharmacokinetics and variable drug disposition of concomitant chemotherapeutic agents at the outer BRB, and further supports the concept of regulating PXR pathway as a potential therapeutic strategy that could be used to modulate P-gp expression/activity in real-world practice. 

Interestingly, the fact that RU486 coculture and PXR silencing did not completely recover the dexamethasone alteration indicates that some pathways other than the GR-PXR cascade may also be involved in the regulation of P-gp. A potential explanation for this may relate to inflammatory mediators such as TNF-α and endothelin-1 (ET-1) suppressed in response to dexamethasone administration. ⁵⁰ Some studies, for example, have reported that dexamethasone-suppressed inflammatory pathways mediate increase in ET-1 levels in the medium of cultured RPE cells. ⁵³ Binding of ET-1 to both the ETA receptor and the ETB receptor was believed to initiate a signal transduction cascade resulting in the upregulation of P-gp transport activity and protein expression, and that ET-1 signaled through nuclear factor–κB, a transcription factor, may also have a role in this regulation. ^50,54 The notion of inflammatory response pathways contributing to the dysfunction of BRB is also supported by a previous study. Penfold et al. ⁵⁵ found that downregulation of inflammatory stimuli at the outer BRB is a significant effect of intravitreal triamcinolone acetonide, another synthetic glucocorticoid corticosteroid with marked anti-inflammatory action, in vivo. Nevertheless, we have demonstrated in the present culture system the involvement of the GR pathway in the regulation of P-gp. Further experiments are thus needed to explore the precise mechanisms regarding the inducible effect of dexamethasone on the P-gp, both by the GR-PXR cascade and by other signaling pathways. 

In summary, we showed that expression and activation of P-gp elevated by dexamethasone in cultured RPE involve GR and PXR. The exact role of P-gp still needs to be determined, but this finding is significant, as it is the first to suggest a link between dexamethasone application and functional expression of P-gp in human RPE cells. 

Footnotes

 Supported in part by Natural Science Foundation of Guangdong Province Grant S2011010000587, Natural Science Foundation of China Grant 30700786, and Bureau of Health of Guangzhou Grants 20121A021007 and 210121A011013.

Footnotes

 Disclosure: Y. Zhang, None; M. Lu, None; X. Sun, None; C. Li, None; X. Kuang, None; X. Ruan, None