(T1430-11-62) MRP1 Knockdown in NCI-H441 Cells to Generate an Organotypic In Vitro Model for Pulmonary Drug Disposition Studies

Purpose: Multidrug resistance-associated protein 1 (MRP1/ABCC1) is a 190 kDa efflux transporter and member of the ATP-binding cassette transporter family. It is ubiquitously expressed throughout the human body, with high levels found in lungs, kidneys and brain. Pulmonary MRP1 is localised to the basolateral membrane of epithelial cells and involved in cellular detoxification of xenobiotics and endogenous compounds. Typical substrates include endogenous compounds such as glutathione, glucuronide and sulphate conjugates of chemically unrelated substances and drugs [1]. Besides its association with diseases such as COPD and cystic fibrosis, an interaction of MRP1 with drugs approved for inhalation [2] and a contribution of MRP1 to pulmonary drug absorption in vivo [3] has also been observed. To study pulmonary MRP1 function in vitro, an organotypic knockout model based on the human NCI-H441 distal lung epithelial cell line was generated and characterised.

Methods: CRISPR/Cas9 technique using the knock-in (KI)/promoter trap method was employed. Pre-assembled spCas9 protein/gRNA construct and repair template were delivered to NCI-H441 cells by electroporation. Repair templates contained puromycin resistance and mCherry fluorescence genes. The successful KI was evaluated by puromycin selection and mCherry detection via flow cytometry. In the thus generated clones ABCC1/MRP1 expression was compared to wild-type (WT) cells by qPCR and Western blot, respectively. Transporter activity was studied using 6-bromo-7-methylpurine (BMP), a prodrug of the MRP1-specific substrate S-(6-(7-methylpurinyl))glutathione (MPG). Transepithelial electrical resistance (TEER) of Transwell-grown KO cells was determined to assess their suitability for transport studies.

Results: Initially, nine KO clones (M1–M9) were generated, of which all but M9 showed mCherry expression as detected via flow cytometry with varying purity (data not shown). Clones M3, M5, M6 and M9 were discarded due to low purity, unnatural morphology or cell senescence. The remaining clones were analysed by qPCR analysis, which revealed that only M1 and M2 showed significantly (p ≤ 0.001) reduced ABCC1 mRNA levels of 34 and 4%, respectively, compared to WT cells. Absence of MRP1 on protein level was investigated by Western blot. In contrast to qPCR results, protein was undetectable in all clones. For functional studies, only M2 was chosen based on the low ABCC1 expression level. Quantities of MPG effluxed were significantly (p ≤ 0.001) lower than in WT cells. The inhibitory effect was comparable to MK-571. Intracellular concentrations of MPG were significantly (p ≤ 0.001) higher in M2 than in WT cells after 1 h loading with BMP. Lastly, the ability to form electrically tight monolayers was assessed in the M2 clone. Cells reached TEER values of over 800 Ω×cm2 after 4 days in culture, confirming their suitability for transport studies. The design of the repair template should, in theory, have prevented any amplification of RNA with the employed primers in a qPCR experiment. However, the almost normal ABCC1 levels in 3 of 5 of the analysed clones can be explained by either, a heterogeneous KO or by a non-homologous end-joining repair in the second allele. In contrast to qPCR results, evaluation of protein abundance by Western blot allowed for a clear differentiation between WT and KO clones, indicating a complete KO of the ABCC1 gene. The absence of MRP1 on functional level in the investigated M2 clone was proven with the BMP/MPG pro-dug/substrate setup. In addition to significantly reduced efflux in the KO clone, this inhibition was similar to that in WT cells incubated with MK-571. The higher intracellular concentration of MPG in the M2 clone after the 1 h loading phase is likely caused by the inhibited efflux during this period. After confirming ABCC1/MRP1 absence on gene, protein and functional level, the ability to form confluent monolayers (determined by measuring TEER) is a prerequisite for cells being suitable as in vitro screens in drug transport studies. The presence of high resistance values in this case is sufficient to confirm the suitability of clone M2.

Conclusion: We successfully generated and characterised an MRP1 KO clone, which can be used to study the potential influence of MRP1 on pulmonary drug disposition and physiology on mechanistic level in vitro. This has previously been challenging due to unspecific pharmacological transporter substrates and inhibitors. Our KO clone also has retained the ability to form electrically tight monolayers, a prerequisite for transport studies.

References: [1] S.P.C. Cole, J. Biol. Chem. 289 (2014) 30880–30888.
[2] M.A. Selo et al., Front. Bioeng. Biotechnol. 8 (2020) 1–12.
[3] S. Mairinger et al., J. Nucl. Med. 61 (2020) 1650–1657.

ABCC1 and MRP1 expression analysis. A Relative expression levels of ABCC1 RNA in WT and ABCC1 KO NCI-H441 cells. qPCR analysis shows significantly reduced mRNA levels in clones M1, M2 and M7. B Analysis of MRP1 protein abundance in WT and MRP1 KO NCI-H441 cells by immunoblot. The absence of protein bands in cell lysates of all clones indicated a successful KO of MRP1. qPCR data represent means ± SD, n = 3, **p ≤ 0.01, ***p ≤ 0.001. Significance of differences between group means and the control were analysed by one-way ANOVA, followed by Dunnett’s multiple comparison test.



Amount of MPG released over 90 min from WT and KO cells incubated with or without the MRP1 inhibitor MK-571. A Release from WT cells was significantly higher than from the M2 clone. No difference was observed between KO cells and WT cells incubated with MK-571. B Intracellular amounts of MPG were significantly higher after 90 min in WT cells incubated with MK-571. Intracellular amounts at T0 (i.e., after 1 h loading with BMP) were significantly higher in all KO cell groups. Data represent means ± SD, n = 3, ns: not significant, **p ≤ 0.01, ***p ≤ 0.001. Significance of differences between all group means was analysed by two-way (A) or one-way (B) ANOVA, followed by Tukey’s multiple comparison test.



TEER curve of WT and M2 NCI-H441 cells on Transwell inserts. Both cell types reached TEER values higher than 800 Ω×cm2 on day 4 post seeding. From there on, TEER values declined steadily in all groups.