Contribution of efflux pump activity to the delivery of pulmonary therapeutics.

To date, there are few in vitro models of the human lung that have been used to characterize multidrug resistant (MDR) efflux pump activity. It is expected that the presence of these protein transporter molecules, such as P-glycoprotein (Pgp) and the multidrug resistance protein associated protein-1 (MRP1), might play a role in limiting drug absorption through the pulmonary epithelium, as has been reported for other epithelial drug delivery barriers such as the intestine and brain. To date, the exact role of the lung resistance related protein (LRP) in MDR is unclear. In this article, we have summarized the biochemistry, function and in vitro/in vivo modulation of Pgp and MRP1. These topics are discussed in light of pulmonary delivery of therapeutic agents, with particular emphasis being placed on the bronchial region of human airways.

Modulation of P-glycoprotein activity in Calu-3 cells using steroids and beta-ligands.

The purpose of this work was to investigate if P-glycoprotein (Pgp) efflux pump activity could be inhibited in the sub-bronchial epithelial cell line, Calu-3, by glucocorticosteroids and beta-ligands. The Pgp modulation efficiency of each compound was determined by its ability to increase the accumulation of the Pgp substrate rhodamine 123 (Rh123) accumulation in these cells. Pgp inhibition was observed at > or =100 microM steroids and beta-ligand. The modulation effectiveness of the beta-ligands increased with increasing hydrophobicity (logP(octanol/aqueous)) whereas an obvious correlation was not obtained with the complete set of steroids tested. Steroidal Pgp substrates did not affect Rh123 accumulation (e.g. aldosterone, dexamethasone, 11beta,17alpha,21-OH progesterone). In contrast, two hydrophobic non-Pgp steroidal substrates (testosterone and progesterone) displayed different effects on Rh123 accumulation, with progesterone being the more potent modulator. The most hydrophobic beta-ligand, propranolol, a known Pgp substrate, gave the largest increase in Rh123 accumulation in this therapeutic class. The beta-ligand modulation efficiency could also be correlated to Pgp structural recognition elements such as hydrogen bonding potential, the presence of a basic nitrogen and planar aromatic ring. No effect on Rh123 accumulation was observed with the formulation additives tested (ethanol, glycerol and palmitoyl carnitine) at concentrations previously reported to be non-toxic to Calu-3 cells.

Multidrug resistance-associated protein-1 functional activity in Calu-3 cells.

The purpose of this work was to determine whether the in vitro bronchiolar epithelial cell model, Calu-3, possesses efflux pump activity by the multidrug resistance-associated protein-1 (MRP1). Reverse transcription-polymerase chain reaction demonstrated MRP1 gene expression in Calu-3 cells. Indirect fluorescence studies showed a basolateral membrane localization of MRP1 compared with P-glycoprotein (Pgp) that was found on the apical side of these cells. An increase in the rate of accumulation of the MRP1 substrate calcein was observed following treatment with the organic anion/MRP1 inhibitor indomethacin, the Pgp inhibitors cyclosporin A (CsA) and vinblastine, as well as conditions of energy depletion. Total calcein efflux was significantly decreased with the MRP1 inhibitors probenecid and indomethacin, while total efflux was unchanged following treatment with CsA. In the latter case, however, intracellular calcein levels postefflux were significantly greater. Probenecid and indomethacin increased calcein net secretion 2.4- and 3.5-fold, respectively. The efflux of etoposide, a known substrate for both Pgp and MRP1, was shown to be mainly Pgp-mediated by using the multidrug-resistant inhibitors quinidine (mixed Pgp/MRP1), CsA (Pgp), and MK571 (MRP1). Together, these data suggest that Calu-3 cells possess MRP1 functional activity that is subordinate to Pgp efflux. We present here kinetic analysis of calcein efflux from Calu-3 cells to support our findings.

P-glycoprotein efflux pump expression and activity in Calu-3 cells.

The purpose of this work was to determine if the sub-bronchial epithelial cell model, Calu-3, expresses the functionally active P-glycoprotein (Pgp) efflux pump. Calu-3 cells express lower levels of Pgp than both Caco-2 and A549 cells as determined by Western Blot analysis. In Calu-3 cells, accumulation of the Pgp substrates rhodamine 123 (Rh123) and calcein acetoxymethyl ester (calcein-AM) was increased in the presence of the specific Pgp inhibitors cyclosporin A (CsA), vinblastine, and taxol. Significant inhibition of Pgp activity was not observed until after 2 h in both cell lines. The organic anion/multidrug resistance associated protein-1 (MRP1) inhibitors, probenecid and indomethacin, did not affect Rh123 accumulation, whereas an increase in calcein accumulation was observed by both agents. The metabolic inhibitor sodium azide decreased the efflux of Rh123 out of Calu-3 cells to the same degree as CsA, supporting inhibition of an active, efflux pathway. The basolateral-to-apical transport of Rh123 was significantly higher than that in the reverse direction, indicating a secretory pathway of efflux that was inhibited 25-fold by CsA. Basolateral-to-apical transport of Rh123 was inhibited slightly with both MRP1 inhibitors; however, no significant effect of Rh123 net secretion was observed. Mixed inhibitor studies demonstrated that Rh123 efflux was mainly Pgp mediated. These results support an energy-dependent Pgp efflux pump pathway that is sensitive to inhibition with CsA in Calu-3 cells.

Co-administration of polyanions with a phosphorothioate oligodeoxynucleotide (CGP 69846A): a role for the scavenger receptor in its in vivo disposition.

The effects of co-administering polyanions on the pharmacokinetics of a 20-mer phosphorothioate oligodeoxynucleotide (CGP 69846A), and the role of scavenger receptors in its in vivo disposition, have been investigated. Following i.v. administration, CGP 69846A was rapidly cleared from the plasma and distributed amongst high (e.g. kidney, liver, spleen), low (e.g. skeletal muscle) and negligible (e.g. brain) accumulating tissues. In addition it was shown that: 1) dextran sulphate co-administration has a dose-dependent effect on the disposition of CGP 69846A; 2) CGP 69846A undergoes renal filtration and renal accumulation largely results from tubular reabsorption; 3) cross-inhibition studies are consistent with CGP 69846A being recognized by scavenger receptors in vitro and in vivo; and 4) the scavenger receptor may be an important determinant for the in vivo disposition of CGP 69846A in mice. These studies contribute toward an increased understanding of the mechanism underlying the pharmacokinetic behaviour of phosphorothioate oligodeoxynucleotides.

The transport of two iron chelators, desferrioxamine B and L1, across Caco-2 monolayers.

The transport of two iron chelators, desferrioxamine B (DFO) and L1 (1,2-dimethyl-3 hydroxypyridin-4-one) has been studied in vitro using the human adenocarcinoma cell line, Caco-2. The transport of DFO and L1 has also been compared with that of their iron-bound complexes, ferrioxamine (FO) and L1(3)-Fe, respectively. We report an apparent permeability coefficient (Papp) value for DFO of 0.170 x 10(-7) +/- 0.080 cm s-1. The Papp value of L1 was 1.297 x 10(-5) +/- 0.133 cm s-1. The Papp values of their iron bound complexes FO and L1(3)-Fe are 0.230 x 10(-7) +/- 0.065 cm s-1 and 2.356 x 10(-6) +/- 0.365 cm s-1, respectively. We have shown that the transport of DFO and FO is similar in the Caco-2 cell system. The transport of L1, however, is greatly reduced when complexed to iron. The value for total uptake after 60 min for DFO into the Caco-2 cells was 1.49 +/- 0.09 x 10(-3) nmol per filter. The values for total uptake after 60 min for L1 and L1(3)-Fe were 0.37 +/- 0.03 nmol per filter and 0.04 +/- 0.01 nmol per filter, respectively. Our results indicate that the poor oral bioavailability of DFO can be attributed to the low epithelial permeability of the molecule coupled with its size (mol wt 656). In contrast, the oral bioavailability observed with L1 is due to the high lipophilicity and low molecular weight (mol wt 139) of the molecule. We believe that these differences between the two molecules account for L1 being better orally absorbed than DFO.