| dc.description.abstract | Over the last decade, the incidence of lung-related diseases such as asthma and chronic obstructive pulmonary disease and their associated morbidity have continuously risen. A heterogeneity in treatment response to commonly prescribed inhaled corticosteroids, ?2 adrenergic agonists and anticholinergic drugs can be observed, yet the cause for this variability is currently unclear. Membrane transporters in the pulmonary epithelial barrier have been suspected to influence the rate and extent of pulmonary absorption of these drugs, but unfortunately, their exact role in the lung is less well understood than in other biological barriers like the liver, intestine or brain. It was the aim of this project to further elucidate the role of three major adenosine triphosphate-binding cassette (ABC) transporters (i.e., MRP1 (ABCC1), BCRP (ABCG2) and P-gp (MDR1/ABCC1)), on the disposition of inhaled drugs. Furthermore, to assess whether different in vitro models of the pulmonary epithelial barriers show distinct expression and transport profiles. Until recently, due to a lack of substrates and inhibitors selective for single transporters, it had been challenging to scrutinise carriers of interest.
With the help of the CRISPR/Cas technology, knockout models based on the NCI H441 cell line were generated, which lacked a specific membrane transporter, i.e., MRP1 or BCRP, while otherwise behaving like their wildtype counterpart. Characterisation of the KO models took place on a molecular biological level via qPCR and immunoblots, as well as on a genomic level via Sanger sequencing. To fully validate each knockout model, functional transport studies were carried out using model substrates of the respective transporter. During this procedure and using the MRP1?/? clone, it was shown, that the frequently used substrate carboxyfluorescein had been incorrectly assumed to be a substrate of MRP1 and that it should be exchanged for more suitable options such as BMP/MPG in organotypic models.
In a comparative study, differences in transporter expression levels of the three aforementioned transporters as well as other MRPs and organic anion transporters (OATs) were evaluated between the established NCI-H441 cell line, the commercially available MucilAir system and freshly isolated AT2/AT1-like cells in primary culture. Differences found on a transcript level were shown to have no significant influence on transport rates of the MRP1 substrate MPG, nor on the P-gp substrate Rh-123, which were compared between NCI-H441 and MucilAir cells. Therefore, the MucilAir model as well as AT1-like cells showed no advantage over NCI-H441 cells justifying the additional costs and other requirements associated with them. Instead, only the NCI-H441 cell line allowed for an assessment of the interaction profile of 5 tested (pre )clinical drugs with the three transporters. This was due to the combined employment of wildtype NCI-H441 cells with high transporter expression levels and their transporter lacking knockout variants.
At the final stage, an array of all above-described models was utilised to study the influence of MRP1, BCRP and P-gp on the bi-directional flux of four clinically relevant pharmaceuticals for inhalation (i.e., tiotropium, umeclidinium, budesonide and vilanterol) and one experimental drug (PF-06263276). In this way it was possible to compare the knockout models with the chemical transporter inhibitors reversan, Ko-143 and zosuquidar, which are described to be decently specific for MRP1, BCRP and P-gp, respectively. While hypothesised to play an important role in drug disposition, MRP1 did not interact with any of the 5 compounds in a major way. Tiotropium, umeclidinium and budesonide did not appear to be influenced by any of the three investigated transporters. Based on distinctly different transport rates in NCI-H441 and MucilAir cells and their significantly different transporter expression levels, MRP3, MRP6 and MRP7 were identified as potential candidates responsible for the observed differences. Vilanterol, described to be a substrate of P-gp, showed no such behaviour in any of the employed cell models. However, this is likely due to low expression levels of the transporter at the pulmonary barrier, which was confirmed by data from this study as well as other recently published investigations of alveolar primary cells. Only the compound PF-06263276 was proven to be majorly transported by BCRP, which was confirmed by a direct comparison between wildtype, chemically inhibited and genetically BCRP lacking NCI-H441 cells.
In summary, the transporter knockout models developed in this project will allow for an assessment of drug?transporter interactions more precise than possible with chemical inhibitors. The organotypic NCI-H441 cell line was shown to be an optimal host for the genetic modifications and studies with clinically relevant compounds yielded valuable results. | en |
