Abstract
Background Several mechanisms of glucocorticoid resistance in asthma have been proposed. P-glycoprotein (P-gp), a ubiquitous efflux transport protein, is associated with variability in the disposition of many drugs and interindividual variability in drug treatment response. This study was undertaken to determine the effect of P-gp expression on glucocorticoid efflux from airway epithelial cells.
Hypothesis Decreasing respiratory epithelial P-gp expression in dexamethasone-exposed airway epithelial cells in vitro will increase intracellular dexamethasone concentration.
Methods A549 lung epithelial cells, transfected with small interfering RNA (siRNA) targeted at messenger RNA for the gene encoding P-gp, were exposed to 100-nM dexamethasone for 15 minutes. Transfection efficiency of siRNA, P-gp expression, and intracellular dexamethasone were measured with flow cytometry.
Results Cells transfected with both negative siRNA and siRNA targeted at P-gp exhibited a positive correlation of P-gp expression with intracellular dexamethasone. The mean ± SEM correlation coefficients were 0.78 ± 0.07 for cells transfected with negative siRNA and 0.79 ± 0.08 for cells transfected with siRNA targeted at P-gp.
Discussion Contrary to our hypothesis, the positive correlation between P-gp expression and intracellular dexamethasone suggests that P-gp is not a primary transporter of glucocorticoids from airway epithelial cells. Increased P-gp expression is unlikely to be an important mechanism of glucocorticoid resistance in asthma.
Glucocorticoids are a mainstay for acute and maintenance therapy for asthma, but 5% of patients with asthma have "difficult asthma," defined as a lack of response to long-term treatment with inhaled glucocorticoids and symptomatic treatment with a short-acting β-agonist.1Patients with asthma who show no clinical improvement after treatment with 40 mg of prednisolone or equivalent steroid dosages for 2 weeks are considered glucocorticoid resistant.2These patients with difficult asthma account for 50% of the total health care costs of asthma.1,3Understanding the mechanisms of glucocorticoid resistance in asthma may improve treatment for these patients with refractory symptoms and for patients with other glucocorticoid-resistant inflammatory diseases.4
Several mechanisms of glucocorticoid resistance in asthma have been described including glucocorticoid-resistant genes, defective glucocorticoid receptor binding and nuclear translocation, and environmentally altered nuclear factors.2,4-6Another proposed mechanism is the increased levels of efflux transporters such as P-glycoprotein (P-gp). P-glycoprotein, a member of the ATP-binding cassette transporter superfamily, is a 170-kd ATP-dependent efflux transporter initially identified as a surface glycoprotein that increases the efflux of chemotherapeutic drugs from cancer cells, a result known as multidrug resistance (MDR).7-9In addition to its role in the transport of chemotherapeutic agents, P-gp has been shown to efflux glucocorticoids in intestinal, brain, and adrenal epithelial cells in vitro.10-12High levels of P-gp expression have been identified on lymphocytes and intestinal epithelial cells of patients with glucocorticoid-resistant inflammatory bowel disease and on lymphocytes of patients with refractory rheumatoid arthritis.13-15However, despite the known expression of P-gp on airway epithelium, it remains unclear if the variable expression of P-gp contributes to glucocorticoid resistance in respiratory inflammatory diseases, particularly glucocorticoid-resistant asthma.16-21
The purpose of this study, therefore, was to examine the effect of P-gp expression on intracellular glucocorticoid concentrations in airway epithelial cells in vitro. We hypothesized that decreasing respiratory epithelial P-gp expression in dexamethasone-exposed airway epithelial cells in vitro will increase intracellular dexamethasone concentration. We decreased P-gp expression in lung epithelial cells with small interfering RNA (siRNA) targeted to MDR1, the gene encoding P-gp. After exposing these cells to fluorescein-labeled dexamethasone, we used flow cytometry to simultaneously measure intracellular dexamethasone and surface P-gp expression. Contrary to our hypothesis, our data showed a positive correlation between P-gp expression and intracellular dexamethasone levels.
MATERIALS AND METHODS
Cell Culture
CCL-185 A549 cells (ATCC Cell Biology Collection, Manassas, VA) were maintained and grown in medium containing Dulbecco modified Eagle medium (Gibco/Invitrogen, Carlsbad, CA), 10% fetal bovine serum (Invitrogen), and 1% penicillin/streptomycin (Sigma-Aldrich, St Louis, MO), at 37°C in a humidified incubator. We maintained cells in a monolayer and passaged them at 70% confluence using phosphate-buffered saline (Fisher, Houston, TX) and trypsin (Sigma-Aldrich).
Transfection
The design and synthesis of 3'-Alexa Fluor 647-modified MDR1 siRNA and AllStars Negative siRNA were completed by Qiagen (Valencia, CA). See Table 1 for the nucleotide sequences of sense MDR1 siRNAs. An RNAi Human/Mouse Starter kit containing HiPerFect Transfection Reagent, siRNA Suspension Buffer, AllStars Negative Control siRNA (Alexa Fluor 488-labeled), and Hs/Mm_MAPK1 Control siRNA was also purchased from Qiagen. We diluted the siRNA with a negative control suspension buffer per the manufacturer's recommendations and stored it at −20°C. We incubated the cells for 5 to 10 minutes in 3 μL of HiPerFect reagent to allow for formation of transfection complexes before incubating with 3 μL of siRNA suspended in 100 μL of serum-free medium. We incubated the cells for 24 hours.
Dexamethasone Exposure
After 24 hours of incubation, we changed the cell culture medium and added 100 nM dexamethasone fluorescein (catalog no. D-1383; Invitrogen) to all wells. After preliminary studies indicated no significant difference in dexamethasone uptake at 60, 30, and 15 minutes, cells were incubated with dexamethasone fluorescein for 15 minutes. The secretions were discarded. We applied trypsin-ethylenediaminetetraacetic acid (EDTA) solution (Sigma-Aldrich) to the cells for 5 minutes to achieve a single cell suspension. Cells were subsequently fixed with 1% paraformaldehyde.
Flow Cytometry
We purchased phycoerythrin (PE)-labeled monoclonal mouse antihuman P-gp (clone 17F9) and appropriate isotype control antibodies from BD Pharmingen (Franklin Lakes, NJ). We centrifuged the cells for 5 minutes at 2200 rpm at 4°C and then aspirated the supernatant. The cells were resuspended in 100 μL of fixation solution and incubated in the dark for 20 minutes. We washed the cells 3 times with 1 mL of permeabilization buffer before the addition of 20 μL of pretitrated intracellular antibodies or appropriate isotype controls. The cells were incubated in the dark for 20 minutes and then washed again with 1 mL of permeabilization buffer. We resuspended the cells in 500 μL of flow cytometry staining buffer (eBioscience, San Diego, CA) and stored them on ice in the dark before analysis on a FACS Calibur System (BD Biosciences, San Jose, CA).
Statistical Analysis
We tested a total of 9 experimental replicates for each of the 2 conditions, negative control siRNA, and MDR1 siRNA-treated cells. Using FlowJo 7.5 Software (Tree Star, Inc, Ashland, OR), we used correlation analysis to determine the correlation between intracellular dexamethasone positivity and P-gp positivity for each condition. We calculated the mean correlation coefficient and SEM.
RESULTS
We tested 4 MDR1 siRNA duplexes targeted to the MDR1 gene on chromosome 7q21.12 and modified with sense 3'-Alexa Fluor 488 for gene-silencing efficacy. Transfections of all 4 MDR1 siRNAs were successful, with varying degrees of P-gp protein expression knockdown ranging from 20% to 65% as measured with flow cytometry (Table 1). Because of its high knockdown efficiency in this initial studies, siRNA #2 was chosen for further analysis. A subsequent FACS analysis determined that transfection with MDR1 siRNA #2 yielded some knockdown of P-gp in 95% of cells and a 20-U overall decrease in the geometric mean expression of P-gp (baseline P-gp: geometric mean ± SEM = 389 ± 64). As a result, MDR1 siRNA #2 was selected for further experimentation.
RNA was isolated from cells transfected with MDR1 siRNA #2, and MDR1 messenger RNA (mRNA) amplification was measured with reverse transcription-polymerase chain reaction using 3 primers targeted at different regions of the MDR1 gene. There was no amplification of MDR1 mRNA for any of these primers; however, there was no alteration in the amplification of the control housekeeping gene β2-microglobulin mRNA.
Cells transfected with MDR1 siRNA #2 and cells transfected with negative control siRNA were exposed to 100-nM dexamethasone for 15 minutes and subjected to flow cytometry (Fig. 1). Cells transfected with both negative siRNA and siRNA targeted at P-gp exhibited a positive correlation of P-gp expression with intracellular dexamethasone. The mean correlation coefficients were 0.78 ± 0.07 for cells transfected with negative siRNA and 0.79 ± 0.08 for cells transfected with siRNA targeted at P-gp.
DISCUSSION
This study was undertaken to determine the effect of altered P-gp expression on glucocorticoid efflux in respiratory epithelial cells. We hypothesized that knocking down P-gp expression in respiratory epithelial cells would result in lower dexamethasone efflux in vitro. However, our results demonstrate the contrary: a positive correlation between P-gp expression and intracellular dexamethasone. This positive correlation is constant with varying levels of P-gp expression.
Glucocorticoids freely diffuse across cell membranes and exert their effect by binding to a cytoplasmic glucocorticoid receptor, which translocates to the nucleus where it functions as a transcription factor.22,23Patients with glucocorticoid-resistant asthma are still susceptible to the Cushingoid adverse effects of glucocorticoids, suggesting that the mechanism of resistance is different from that seen in cases of primary cortisol resistance, which has been attributed to a variety of disorders of glucocorticoid receptor number, translocation, and function.24,25
P-glycoprotein has been identified, and it seems to be functional in both bronchial and alveolar epithelial cells.16,26-28Mutations in similar transmembrane transporters in the lung are found in disease states such as cystic fibrosis, Tangier disease, and chronic obstructive pulmonary disease.21,29-31Most likely, in the respiratory system, P-gp functions to rid the lungs of pathogens, irritants, pollutants, and other agents that produce oxidative stress.27
The relationship between P-gp and glucocorticoids may be complicated by the fact that P-gp expression is upregulated by cellular stress, cytokine release, and inflammation.26,29,32In addition, the significance of P-gp expression and function as a determinant of inhaled glucocorticoid delivery is debatable. Multiple studies attempting to characterize P-gp expression and function in the respiratory epithelium in vitro have generated contradictory results. Some studies suggest that P-gp activity in respiratory epithelium has an insignificant effect on drug absorption in vitro, particularly compared with intestinal models.18,19,33-35Further, inhaled glucocorticoids have been shown to decrease P-gp expression by reducing MDR1 gene activity in lymphocytes of children with asthma.36In our experiments, a brief exposure to dexamethasone is unlikely to induce de novo P-gp synthesis. However, it seems plausible that glucocorticoids could increase surface P-gp expression by inducing its translocation from the cytosol to the cell surface.
Our study is limited by the epithelial cell model used. Although A-549 cells have many properties in common with human respiratory epithelial cells, they may have different drug permeability. In addition, they lack the polarization of human respiratory epithelial cells. The population of A549 cells may contain various cell sizes that could be correlated with different amounts of cell surface P-gp expression and different amounts of dexamethasone uptake. This could contribute to the direct relationship observed between P-gp expression and intracellular dexamethasone. In addition, the dose of dexamethasone used in our experiment may not reflect physiologic dosing and the 15-minute exposure may be too long or too short to accurately reflect the duration of cellular drug exposure in vivo. Our study may also not reflect physiologic conditions in that, because of its higher molecular weight, dexamethasone fluorescein is incorporated into cells at a slightly slower rate than dexamethasone.
Our results suggest that, although it is likely that there are multiple molecular mechanisms of glucocorticoid insensitivity, glucocorticoid efflux by P-gp is unlikely to be a significant mechanism of glucocorticoid resistance in subjects with asthma. Understanding these mechanisms may ultimately help guide the management of subgroups of patients with glucocorticoid-resistant asthma.