PI-022 - A CRISPR/CAS9-DERIVED ABCB1/ABCG2 KNOCKOUT PLATFORM FOR SUBSTRATE CHARACTERIZATION.

A. Kasten¹, S. Neef², U. Hoffmann², S. Oswald³, I. Vater⁴, M. Spielmann⁵, M. Schwab⁶, I. Cascorbi⁷, M. Kaehler⁷; ¹University of Kiel, University Hospital Schleswig-Holstein Kiel, Germany, ²Dr. Margarete Fischer-Bosch Institute, Dr. Margarete Fischer-Bosch Institute, Stuttgart, Germany, ³University Medical Center Rostock, University Medical Center Rostock, Germany, ⁴University Hospital-Schleswig-Holstein, University Hospital-Schleswig-Holstein, Germany, ⁵University Hospital Schleswig-Holstein, University Hospital Schleswig-Holstein, Luebeck, Germany, ⁶Dr. Margarete Fischer-Bosch Institute of Clinical Pharmacology, Dr. Margarete Fischer-Bosch Institute of Clinical Pharmacology, Germany, ⁷University Hospital Schleswig-Holstein, University Hospital Schleswig-Holstein, Kiel, Germany.

Background: ATP-binding cassette (ABC) transporters have a crucial impact on biodistribution of a large number of drugs. It is therefore necessary to determine transport properties in the preclinical development of new drugs. Currently used in-vitro models are subject to limitations, e.g. overlapping transporter specificities or lack of specific inhibitors. In order to get more specific insights, our aim was to establish a knockout platform of a human cell line for the two major drug transporters, ABCB1 (P-glycoprotein, P-gp) and ABCG2 (breast cancer resistance protein, BCRP).

Methods: ingle and double knockouts were generated in Caco-2 cells by CRISPR/Cas9-genome editing and confirmed on genomic, protein and functional level. Bidirectional transport studies with established and putative ABCB1 and ABCG2 substrates were conducted followed by HPLC-MS/MS. Cytotoxicity of anti-cancer drugs was assessed using cell viability assays.

Results: CRISPR/Cas9-genome editing successfully created ABCB1 and ABCG2 single and double knockouts characterized by frameshift deletions with premature stop codons. ABCB1- or ABCG2-mediated transport was significantly diminished compared to wild-type cells for fluorescent dyes (rhodamine-123: p < 0.001; BODIPY-prazosin: p < 0.001). In case of sulfasalazine and digoxin, transport activity was hardly detectable in single and double knockouts, whereas for loperamide, knockout cells showed a residual transport activity. Rosuvastatin transport was lower in the double compared to the ABCG2 single knockout (p < 0.001). Doxorubicin had a significantly higher cytotoxicity in the ABCB1 single and double knockout compared to wild-type cells (-32.5/-44.4 %, p < 0.001), while loss of ABCG2 led to a significant increase in cytotoxicity for imatinib (-11.9/-34.6 %, p < 0.001).

Conclusion: This successfully established ABCB1 and ABCG2 knock-out platform exhibited no residual transporter activity, thus allowing more specific determination of transporter affinities when comparing wild-type and knock-out cells in relation to the use of inhibitors. Further potential applications and developments of this model include the identification of novel substrates, the analysis of pharmacogenetic variants and the extension to other drug transporters.