- Email: [email protected]
A CRISPR-Cas9 Generated MDCK Cell Line Expressing Human MDR1 Without Endogenous Canine MDR1 (cABCB1): An Improved Tool for Drug Efflux Studies
Accepted Manuscript A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an improved tool for drug efflux studies Maria Karlgren, Ivailo Simoff, Maria Backlund, Christine Wegler, Markus Keiser, Niklas Handin, Janett Müller, Patrik Lundquist, Anne-Christine Jareborg, Stefan Oswald, Per Artursson PII:
S0022-3549(17)30251-4
DOI:
10.1016/j.xphs.2017.04.018
Reference:
XPHS 741
To appear in:
Journal of Pharmaceutical Sciences
Received Date: 15 February 2017 Revised Date:
11 April 2017
Accepted Date: 12 April 2017
Please cite this article as: Karlgren M, Simoff I, Backlund M, Wegler C, Keiser M, Handin N, Müller J, Lundquist P, Jareborg AC, Oswald S, Artursson P, A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an improved tool for drug efflux studies, Journal of Pharmaceutical Sciences (2017), doi: 10.1016/j.xphs.2017.04.018. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Karlgren
A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an improved tool for drug efflux studies ACCEPTED MANUSCRIPT
1 (14)
A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an
RI PT
improved tool for drug efflux studies
Karlgren Maria1,2*, Simoff Ivailo1, Backlund Maria1,2, Wegler Christine1,3, Keiser Markus4,
SC
Handin Niklas1, Janett Müller4, Lundquist Patrik1, Jareborg Anne-Christine5, Oswald Stefan4, Artursson Per1,2 1
2
M AN U
Department of Pharmacy, Uppsala University, Uppsala, Sweden
Uppsala University Drug Optimization and Pharmaceutical Profiling Platform (UDOPP), Science for
Life Laboratory, Uppsala, Sweden 3
Cardiovascular Metabolic Diseases DMPK, AstraZeneca R&D, Mölndal, Sweden
4
Department of Clinical Pharmacology, Center of Drug Absorption and Transport (C_DAT), University
5
TE D
Medicine of Greifswald, Germany
Department of immunology, genetics and pathology, Science for Life Laboratory, Uppsala University,
EP
Uppsala, Sweden
*Address correspondence to:
AC C
Maria Karlgren, PhD
Department of Pharmacy Uppsala University Box 580
SE-751 23 Uppsala, Sweden Email: [email protected] Phone: +46 – 18 471 41 49 Fax: +46 – 18 471 42 23
Karlgren
A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an improved tool for drug efflux studies ACCEPTED MANUSCRIPT
2 (14)
Abstract Madin-Darby canine kidney (MDCK) II cells stably transfected with transport proteins are commonly used models for drug transport studies. However, endogenous expression of especially canine MDR1 (cMDR1) confounds the interpretation of such studies. Here we have established an MDCK cell line
RI PT
stably overexpressing the human MDR1 transporter (hMDR1; P-glycoprotein), and used CRISPRCas9 gene editing to knock out the endogenous cMDR1. Genomic screening revealed the generation of a clonal cell line homozygous for a four-nucleotide deletion in the canine ABCB1 gene leading to a
SC
frameshift and a premature stop codon. Knockout of cMDR1 expression was verified by quantitative protein analysis and functional studies showing retained activity of the human MDR1 transporter.
M AN U
Application of this cell line allowed unbiased reclassification of drugs previously defined as both substrates and non-substrates in different studies using commonly used MDCK-MDR1 clones. Our new MDCK-hMDR1 cell line, together with a previously developed control cell line, both with identical deletions in the canine ABCB1 gene and lack of cMDR1 expression represent excellent in
TE D
vitro tools for use in drug discovery.
Keywords: MDCK cells, ABC transporters, Efflux pumps, P-glycoprotein, Permeability, Membrane
AC C
EP
transporter, Drug transport, Genomics, Proteomics, Multidrug resistance transporters
Karlgren
A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an improved tool for drug efflux studies ACCEPTED MANUSCRIPT
3 (14)
Introduction Multidrug resistance protein 1 (protein name: multidrug resistance protein 1 (MDR1) often referred to as P-glycoprotein (Pgp); gene name: ABCB1), discovered in 19761 modifies the pharmacokinetics of drugs by affecting their intestinal absorption and/or their biliary or renal secretion as well as their
RI PT
tissue distribution (e.g. across the blood-brain barrier). As MDR1 has a very broad substrate specificity it is routinely studied in drug discovery. Regulatory agencies such as the EMA and the FDA provide guidelines for in vitro investigations of putative substrates and inhibitors of MDR1 that
SC
may cause drug-drug interactions (DDI) with MDR1 substrates.2-5
Madine-Darby Canine Kidney II (MDCK) cells, a monolayer-forming epithelial cell line, is the most
M AN U
common model system for studying the influence of human efflux transporters on drug disposition. These cells possess the correct localization of clinically relevant transporters such as MDR1 and MRP2 and are therefore frequently used for heterologous overexpression of human transporters, singly or (as first shown by Keppler and Sugiyama) in combination with each other.6, 7 Confluent MDCK cells on filter supports form polarized, tight monolayers allowing studies of bidirectional
TE D
transepithelial transport.7-11 As human MDR1 (hMDR1) is sorted to the apical plasma membrane in these cells, the transport rate of a typical MDR1 substrate is lower from the apical to the basolateral side of the cells than in the opposite direction. The ratio between the permeability of these two
EP
transport directions, the so called efflux ratio (ER), is the simplest approach for identifying efflux substrates in various tissue barriers including the blood-brain-barrier.12
AC C
MDCK cells express endogenous canine kidney transporters like canine MDR1 (cMDR1), which was shown to influence transport of compounds.13 Therefore, contribution of cMDR1 in MDCK-hMDR1 cells (expressing both hMDR1 and cMDR1) needs to be subtracted by including MDCK wild type cells (expressing cMDR1) as a negative control in the experiment. Alternative approaches to overcome the endogenous background have included selection of clones with low levels of cMDR1, suppression of expression by inhibitory RNA, and knockouts using zinc-finger nucleases.14-17 These approaches have not been completely satisfying as background correction is necessary even for knockout cells using the zink-finger nuclease technology. Therefore, we recently used the CRISPR-Cas9 gene editing
Karlgren
A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an improved tool for drug efflux studies ACCEPTED MANUSCRIPT
4 (14)
technology to develop a MDCK cell line completely lacking endogenous cMDR1 background.18 In this cell line, all copies of the canine ABCB1 gene were modified, leading to a truncated protein without cMDR1 activity, which was confirmed using prototypic MDR1 substrates. By using the same CRISPR-Cas9 gene editing approach18 and a previously established MDCK-
RI PT
hMDR1 cell line19 as starting point, we now present the development of a MDCK-hMDR1cMDR1-KO cell line, which completely lacks cMDR1 background but retains hMDR1 activity. This MDCK-
hMDR1cMDR1-KO cell line, which we propose can be used without background correction, represents an
AC C
EP
TE D
M AN U
SC
improved tool for transport and DDI studies in vitro.
Karlgren
A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an improved tool for drug efflux studies ACCEPTED MANUSCRIPT
5 (14)
Methods Cell culture MDCK cells (obtained from American Type Culture Collection (ATCC CRL-2936, Rockville, MD) and MDCKcMDR1-KO cells18, were cultured as previously described.18 For MDCK cells overexpressing
RI PT
hMDR1 transporter, i.e.MDCK-hMDR119 and MDCK-hMDR1cMDR1-KO cells, culturing media was further supplemented with 375µg/ml Hygromycin B. Cell culture media and supplements were from Invitrogen (Carlsbad, CA), Sigma-Aldrich (St. Louis, MO) or from PAN Biotech (Aidenbach,
SC
Germany). Cells were cultured at 37°C, 95% humidity and 5% CO2 and sub-cultured twice a week. Knockout of canine MDR1
M AN U
Knockout of the canine ABCB1 gene using canine-specific CRISPR-Cas9 vectors was performed as described in18 and in the supplementary material. To confirm gene editing, the target regions of genomic DNA from individual cell clones were amplified by PCR and the PCR product was sequenced by Sanger sequencing at Uppsala Genome Center, Uppsala University.
TE D
Protein quantification of drug transporters
Samples from MDCK wildtype, MDCKcMDR1-KO, MDCK-hMDR1, and MDCK-hMDR1cMDR1-KO were subjected to protein quantification to determine expression levels of cMDR1 and hMDR1 as well as
EP
eight other transport proteins using two different types of sample preparation protocols prior to targeted proteomics (see supplementary material).
AC C
Transport assays
Transport experiments with 15 known or putative MDR1 substrates were performed at 1µM as previously described.18 Experiments were partly verified using radiolabeled compounds (see supplementary material). Digoxin experiments were performed using 2.5 µCi/ml HBSS buffer of 3Hdigoxin (26.3 Ci/mmol) (Perkin Elmer, Wiltham, MA) and 2 µM unlabeled digoxin (Sigma-Aldrich, St Louis, MO) with or without addition of 2 µM elacridar. For details see Simoff et al.18 Student’s ttest was used for statistical comparisons.
Karlgren
A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an improved tool for drug efflux studies ACCEPTED MANUSCRIPT
6 (14)
Results Knockout of canine MDR1 in MDCK-hMDR1 cells Genetic analysis revealed that homozygous indels occurred in four MDCK-hMDR1 clones out of 14 clones analyzed by genomic sequencing. Remaining clones exhibited an unaltered genotype or were
RI PT
heterozygous at exon 4 and/or 13. The clone selected for further analysis, here referred to as MDCKhMDR1cMDR1-KO, showed the same genotype as the previously established MDCKcMDR1-KO clone.18 Thus, both clones had a four-nucleotide deletion in exon 4 (Figure 1), leading to a shift of the reading
Protein quantification using surrogate peptides
SC
frame and three consecutive stop codons, resulting in a truncated protein.
M AN U
The knock-out of cMDR1 and overexpression of hMDR1 in MDCK cells were verified by quantification of MDR1 proteins in filter-grown cells with targeted proteomics using a human-specific peptide and a shared peptide detecting both cMDR1 and hMDR1 (Table 1). As expected, cMDR1 could only be quantified with the shared peptide (recognizing both cMDR1 and hMDR1) in the MDCK wildtype cells (on average 1.5 fmol/µg total protein). Further, cMDR1 was not detected in the
TE D
MDCKcMDR1-KO cells, with either peptide, demonstrating a successful knock-out. MDR1 was quantified in MDCK-hMDR1 as well as in MDCK-hMDR1cMDR1-KO cells with both peptides (on average 8.2 fmol/µg total protein; range 4.7-14.3 between the two samples, measured by the two peptides),
EP
confirming overexpression of hMDR1 in the cells (Table 1). The fold difference between the levels determined with the shared peptide and the human-specific peptide, as well as the absolute levels
AC C
measured with the shared peptide, was significantly higher in MDCK-hMDR1 than for MDCKhMDR1cMDR1-KO cells (p < 0.05, two tailed Mann-Whitney test, and p < 0.01, one-way ANOVA followed by Tukey's multiple comparisons test, respectively)(see supplementary table 3). This indicated a complete knock-out of cMDR1 in the MDCK-hMDR1cMDR1-KO cells. Successful knockout of cMDR1 and over-expression of hMDR1 was also confirmed in enriched crude membrane fractions of undifferentiated cells grown in cell culture flasks (Table 1). Further, quantification of other transporters showed only minor differences between the four cell lines (Table 1).
Karlgren
A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an improved tool for drug efflux studies ACCEPTED MANUSCRIPT
7 (14)
MDR1 mediated transport Transport experiments with the MDR1 substrate digoxin showed that hMDR1 function was retained in MDCK-hMDR1cMDR1-KO and parenteral MDCK-hMDR1 cells, giving higher Papp in the basolateral-to-
RI PT
apical (BA) than in the apical-to-basolateral (AB) direction (Figure 2a). Moreover, MDCK wildtype cells showed significant efflux. The MDR1 inhibitor elacridar reduced the ERs of these three cell lines close to unity. In contrast, for the MDCKcMDR1-KO cells, digoxin ER was close to unity both in presence
SC
and absence of the MDR1 inhibitor.
Nine prototypical MDR1 substrates were chosen for functional studies in the cMDR1 knock-out cell
M AN U
lines (Figure 2b). Seven of these substrates showed ERs close to unity while two compounds showed some residual efflux (ER close to 2) presumably via other endogenous transporters in the MDCKcMDR1KO
cells, whereas ERs ranged from 5 (for saquinavir) up to 362 (for ritonavir) in the MDCK-
hMDR1cMDR1-KO cells. To further test the robustness of these cell lines, loperamide, ritonavir, talinolol, and verapamil were investigated using radiolabeled compounds in a second laboratory. Also here, all
TE D
compounds were classified as MDR1 substrates, but at lower resolution, which was probably a result of the indirect determination of radioactivity (see supplementary material). To further test the applicability of the CRISPR generated cMDR1 knockout cell lines, six compounds
EP
with contradicting classifications using MDCK cell lines overexpressing hMDR1 developed by Prof. Piet Borst (from the Netherlands Cancer Institute; NKI) or by Pastan et al.20 (from the National
AC C
Institutes of Health; NIH)(cf. Broccatelli et al.21), were investigated using MDCKcMDR1-KO and MDCKhMDR1cMDR1-KO cells. In our study, four of these compounds showed a clear difference in ERs in the MDCKcMDR1-KO and the MDCK-hMDR1cMDR1-KO cells, respectively (Figure 2c and d). For two compounds, i.e. olanzapine and clozapine, no difference was observed between MDCKcMDR1-KO and MDCK-hMDR1cMDR1-KO cells. Taken together, four of these six compounds could be (re)classified as MDR1 substrates.
Karlgren
A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an improved tool for drug efflux studies ACCEPTED MANUSCRIPT
8 (14)
Discussion Different approaches have been used to correct for endogenous cMDR1 transport activity in MDCK cells (see for example recommendations by the International Transporter Consortium4, 5) in order to avoid misclassification and unnecessary follow-up studies of MDR1 substrates. All currently used
RI PT
approaches necessitate background correction in MDCK wildtype cells. Despite that it is well known that overexpression of transport proteins can alter the expression or function of endogenous
transporters13, 22, cMDR1expression levels in the wildtype and the overexpressing cells are generally
SC
considered equivalent. Furthermore, expression levels can vary from laboratory to laboratory, due to culturing conditions, or as a result of clonal selection during the generation of the overexpressing cell
M AN U
line.23
In a previous study we have used CRISPR-Cas9 to generate a MDCK cell line completely lacking endogenous cMDR1 expression.18 As the CRISPR-Cas9 guide RNAs (gRNAs) in that study were designed to be sequence-specific to canine ABCB1 only, herein we used these gRNAs to knock out
TE D
cMDR1 in a MDCK cell line already overexpressing hMDR1.
The sequences for cMDR1 and hMDR1 are very similar at both the genomic and proteomic levels,
EP
which makes it difficult to design species-specific CRISPR-Cas9 gRNAs and especially peptides for proteomics quantification. Therefore, no suitable canine-specific peptide could be identified. Hence,
AC C
we took another approach and used one hMDR-specific and one shared peptide that identified both hMDR1 and cMDR1. This allowed indirect quantification of cMDR1 expression by comparing the ratio of the two peptides in the different cell lines. When comparing expression levels determined with the shared and the human-specific peptide we could see that the protein quantification ratio between these peptides in the MDCK-hMDR1cMDR1-KO cells were identical to ratios obtained in human hepatocytes and human jejunum (i.e. tissues only expressing hMDR1; see supplementary table 3). In contrast, the ratio of the two peptides in MDCK-hMDR1 was significantly higher, reflecting expression of both cMDR1 and hMDR1 in this cell line. Furthermore, the high expression of hMDR1
Karlgren
A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an improved tool for drug efflux studies ACCEPTED MANUSCRIPT
9 (14)
in the MDCK-hMDR1cMDR1-KO cell line verified that the gRNAs were indeed canine-specific. Correct localization of hMDR1 in the apical membrane was verified by immunofluorescence staining (see supplementary material). Despite the complete knock-out of cMDR1, no compensatory up regulation
RI PT
of other efflux transporters could be detected.
Transwell experiments with prototypical MDR1 substrates further verified retained activity of the hMDR1, with the MDCK-hMDR1cMDR1-KO cells having significantly higher ERs compared to
MDCKcMDR1-KO cells for all investigated substrates. In addition, experiments with e.g. digoxin
SC
indicated that we may have selected a clone with high hMDR1 activity during the CRISPR-Cas9
to the parental MDCK-hMDR1 cells.
M AN U
clonal selection process, as higher ER was observed in the MDCK-hMDR1cMDR1-KO cells as compared
To further test the performance of our knockout models we examined six compounds, as identified by Broccatelli et al.21 who presented contradicting results on MDCK-hMDR1 cells of different origin.
TE D
Our hypothesis was that it should be possible to use our two knockout cell lines, MDCKcMDR1-KO and MDCK-hMDR1cMDR1-KO, in a standard screening set-up to obtain conclusive and easily interpretable results for these compounds. Indeed we observed a significant difference in the ERs of the
EP
MDCKcMDR1-KO and MDCK-hMDR1cMDR1-KO cells for four of these compounds. Our results were in better agreement with those obtained using the NIH cells than the NKI cells, either when using the
AC C
classical ER cutoff ≥2 or ≥8.5 as suggested by Broccatelli et al.21 for NIH cells. At an ER cutoff of 2, the classification of olanzapine was different in our cells as compared to the NIH cells (Figure 2d). Olanzapine has previously shown no or moderate affinity for MDR1 using different in vitro assays.24-26 In addition, olanzapine has previously displayed a high permeability and a low ER around the ER cutoff in MDCK-hMDR1 cells.27 The difference in classifications is therefore not surprising. However, if an ER cutoff of ≥8.5 (as suggested by Broccatelli et al.21) is used, olanzapine is classified as a non-substrate, both in our cell line and in the NIH cell line. In general it is difficult to determine in vitro if compounds with very high permeability are substrates of efflux transporters.28, 29
Karlgren
A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an improved tool for drug efflux studies ACCEPTED MANUSCRIPT
10 (14)
At an ER cutoff at 8.5, the classification of the more hydrophilic drug ranitidine differed in our cells and the NIH cells (Figure 2d). This discrepancy might be explained by differences in the experimental conditions as in the previous study ranitidine was tested at a concentration of 50µM30 whereas in this study, we used 1µM. We speculate that the high concentration (50µM) could partly have saturated the
RI PT
transporter and thereby led to false negative results.
Earlier attempts to obtain MDCK cells with lower or no cMDR1 background did not conclusively establish that the cell lines were devoid of endogenous background. CRISPR-Cas9 has emerged as the
SC
method of choice for controlled and precise generation of knock-out of specific genes in cell lines. After our previous publication of a MDCK cell line completely lacking cMDR1 expression, others
M AN U
have used similar CRISPR-Cas9 approaches to knockout hMDR1 in multi-drug resistant human cell lines31 and cMDR1 in MDCK32. Although functional studies indicate that hMDR1/cMDR1 has been knocked down/out in these studies, to our knowledge, MDR1 protein quantification has not been
TE D
performed to verify the knockouts.
In conclusion we here describe the development of an MDCK-hMDR1cMDR1-KO cell line. This cell line, and its matching control cell line MDCKcMDR1-KO, lack endogenous cMDR1 expression, as verified
EP
using genomic screening and proteomics analysis. Functional validation further showed no MDR1 activity in the control cell line MDCKcMDR1-KO, whereas high hMDR1 activity was retained in the
AC C
MDCK-hMDR1cMDR1-KO cell line.
Karlgren
A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an improved tool for drug efflux studies ACCEPTED MANUSCRIPT
11 (14)
Acknowledgements We are grateful to Maria Mastej and Elin Khan for assistance with clonal expansion of the cells and Maria Mastej for cell culturing. We acknowledge the staff at the BioVis platform, Science for Life Laboratory, Uppsala University, for assistance with FACS, microscopy and imaging. This work was
RI PT
supported by the Swedish Fund for Research without Animal Experiments, the Åke Wiberg foundation, the Magnus Bergvall foundation, the Swedish Research Council (approval number 5715), and the German Federal Ministry of Education and Research (BMBF, InnoProfile-Transfer, grant
AC C
EP
TE D
M AN U
SC
number 03IPT612X).
Karlgren
A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an improved tool for drug efflux studies ACCEPTED MANUSCRIPT
12 (14)
References
7.
8. 9.
10.
11.
12.
13.
14. 15. 16.
17. 18.
19.
RI PT
SC
6.
M AN U
5.
TE D
3. 4.
EP
2.
Juliano RL, Ling V. A surface glycoprotein modulating drug permeability in Chinese hamster ovary cell mutants. Biochim Biophys Acta. 1976;455(1):152-162. Guidance for Industry Drug Interaction Studies — Study Design, Data Analysis, Implications for Dosing, and Labeling Recommendations. U.S. Dep of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research; 2012. Guideline on the Investigation of Drug Interactions. European Medicines Agency; 2013. Giacomini KM, Huang SM, Tweedie DJ, Benet LZ, Brouwer KL, et al. Membrane transporters in drug development. Nature reviews Drug discovery. 2010;9(3):215-236. Tweedie D, Polli JW, Berglund EG, Huang SM, Zhang L, et al. Transporter studies in drug development: experience to date and follow-up on decision trees from the International Transporter Consortium. Clin Pharmacol Ther. 2013;94(1):113-125. Cui Y, Konig J, Keppler D. Vectorial transport by double-transfected cells expressing the human uptake transporter SLC21A8 and the apical export pump ABCC2. Mol Pharmacol. 2001;60(5):934-943. Sasaki M, Suzuki H, Ito K, Abe T, Sugiyama Y. Transcellular transport of organic anions across a double-transfected Madin-Darby canine kidney II cell monolayer expressing both human organic anion-transporting polypeptide (OATP2/SLC21A6) and Multidrug resistanceassociated protein 2 (MRP2/ABCC2). J Biol Chem. 2002;277(8):6497-6503. Gottesman MM, Fojo T, Bates SE. Multidrug resistance in cancer: role of ATP-dependent transporters. Nat Rev Cancer. 2002;2(1):48-58. Hammerle SP, Rothen-Rutishauser B, Kramer SD, Gunthert M, Wunderli-Allenspach H. PGlycoprotein in cell cultures: a combined approach to study expression, localisation, and functionality in the confocal microscope. Eur J Pharm Sci. 2000;12(1):69-77. Konig J, Zolk O, Singer K, Hoffmann C, Fromm MF. Double-transfected MDCK cells expressing human OCT1/MATE1 or OCT2/MATE1: determinants of uptake and transcellular translocation of organic cations. Br J Pharmacol. 2011;163(3):546-555. Schinkel AH, Mayer U, Wagenaar E, Mol CA, van Deemter L, et al. Normal viability and altered pharmacokinetics in mice lacking mdr1-type (drug-transporting) P-glycoproteins. Proc Natl Acad Sci U S A. 1997;94(8):4028-4033. Kalvass JC, Pollack GM. Kinetic considerations for the quantitative assessment of efflux activity and inhibition: implications for understanding and predicting the effects of efflux inhibition. Pharm Res. 2007;24(2):265-276. Kuteykin-Teplyakov K, Luna-Tortos C, Ambroziak K, Loscher W. Differences in the expression of endogenous efflux transporters in MDR1-transfected versus wildtype cell lines affect P-glycoprotein mediated drug transport. Br J Pharmacol. 2010;160(6):1453-1463. Celius T, Garberg P, Lundgren B. Stable suppression of MDR1 gene expression and function by RNAi in Caco-2 cells. Biochem Biophys Res Commun. 2004;324(1):365-371. Di L, Whitney-Pickett C, Umland JP, Zhang H, Zhang X, et al. Development of a new permeability assay using low-efflux MDCKII cells. J Pharm Sci. 2011;100(11):4974-4985. Gartzke D, Delzer J, Laplanche L, Uchida Y, Hoshi Y, et al. Genomic Knockout of Endogenous Canine P-Glycoprotein in Wild-Type, Human P-Glycoprotein and Human BCRP Transfected MDCKII Cell Lines by Zinc Finger Nucleases. Pharm Res. 2015;32(6):20602071. Gartzke D, Fricker G. Establishment of optimized MDCK cell lines for reliable efflux transport studies. J Pharm Sci. 2014;103(4):1298-1304. Simoff I, Karlgren M, Backlund M, Lindstrom AC, Gaugaz FZ, et al. Complete Knockout of Endogenous Mdr1 (Abcb1) in MDCK Cells by CRISPR-Cas9. J Pharm Sci. 2016;105(2):1017-1021. Siegmund W, Modess C, Scheuch E, Methling K, Keiser M, et al. Metabolic activation and analgesic effect of flupirtine in healthy subjects, influence of the polymorphic NAT2, UGT1A1 and GSTP1. Br J Clin Pharmacol. 2015;79(3):501-513.
AC C
1.
Karlgren
26.
27.
28.
29. 30.
31. 32.
RI PT
25.
SC
24.
M AN U
23.
TE D
22.
EP
21.
13 (14)
Pastan I, Gottesman MM, Ueda K, Lovelace E, Rutherford AV, et al. A retrovirus carrying an MDR1 cDNA confers multidrug resistance and polarized expression of P-glycoprotein in MDCK cells. Proc Natl Acad Sci U S A. 1988;85(12):4486-4490. Broccatelli F, Larregieu CA, Cruciani G, Oprea TI, Benet LZ. Improving the prediction of the brain disposition for orally administered drugs using BDDCS. Adv Drug Deliv Rev. 2012;64(1):95-109. Li J, Wang Y, Hidalgo IJ. Kinetic analysis of human and canine P-glycoprotein-mediated drug transport in MDR1-MDCK cell model: approaches to reduce false-negative substrate classification. J Pharm Sci. 2013;102(9):3436-3446. Dukes JD, Whitley P, Chalmers AD. The MDCK variety pack: choosing the right strain. BMC Cell Biol. 2011;12:43. Boulton DW, DeVane CL, Liston HL, Markowitz JS. In vitro P-glycoprotein affinity for atypical and conventional antipsychotics. Life Sci. 2002;71(2):163-169. Wang JS, Zhu HJ, Markowitz JS, Donovan JL, DeVane CL. Evaluation of antipsychotic drugs as inhibitors of multidrug resistance transporter P-glycoprotein. Psychopharmacology (Berl). 2006;187(4):415-423. Nagasaka Y, Oda K, Iwatsubo T, Kawamura A, Usui T. Effects of aripiprazole and its active metabolite dehydroaripiprazole on the activities of drug efflux transporters expressed both in the intestine and at the blood-brain barrier. Biopharm Drug Dispos. 2012;33(6):304-315. Wager TT, Chandrasekaran RY, Hou X, Troutman MD, Verhoest PR, et al. Defining desirable central nervous system drug space through the alignment of molecular properties, in vitro ADME, and safety attributes. ACS Chem Neurosci. 2010;1(6):420-434. Artursson P, Matsson P, Karlgren M. In vitro characterization of interactions with drug transporting proteins. In: Sugiyama T, Steffensen B, editors. Transporters in Drug Development. AAPS Advances in the Pharmaceutical Sciences Springer; 2014. Polli JW, Wring SA, Humphreys JE, Huang L, Morgan JB, et al. Rational use of in vitro Pglycoprotein assays in drug discovery. J Pharmacol Exp Ther. 2001;299(2):620-628. Wang Q, Rager JD, Weinstein K, Kardos PS, Dobson GL, et al. Evaluation of the MDRMDCK cell line as a permeability screen for the blood-brain barrier. Int J Pharm. 2005;288(2):349-359. Yang Y, Qiu JG, Li Y, Di JM, Zhang WJ, et al. Targeting ABCB1-mediated tumor multidrug resistance by CRISPR/Cas9-based genome editing. Am J Transl Res. 2016;8(9):3986-3994. Xiao Y, McCollum B, Pratt J, Hauser S, Bourner M, et al. Better Understand ADME Properties of Drugs Using MDR1 Knock In/Out in MDCKII Cells. Poster at the AAPS annual meeting and exposition. 2016.
AC C
20.
A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an improved tool for drug efflux studies ACCEPTED MANUSCRIPT
Karlgren
A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an improved tool for drug efflux studies ACCEPTED MANUSCRIPT
14 (14)
Figure legends Fig 1. The CRISPR-Cas9 target region in exon 4 of the canine ABCB1 gene. Chromatograms for the parental MDCK-hMDR1 cell line and the CRISPR-Cas9 generated MDCK-hMDR1cMDR1-KO (a). The position of the four-nucleotide deletion which leads to a frameshift missense mutation and premature
RI PT
stop codons downstream in exon 4 is indicated with a box /line. Alignment of MDCK-hMDR1cMDR1-KO and corresponding sequences for the cMDR1 reference sequence (gene ID 403879) and the previously
SC
published MDCKcMDR1-KO 18 (b).
Fig 2. Efflux ratios for digoxin alone and in the presence of the MDR1 inhibitor elacridar in the
M AN U
MDCK wildtype (wt), MDCKcMDR1-KO, MDCK-hMDR1cMDR1-KO, and MDCK-hMDR1 cells (a). Efflux ratios for prototypic MDR1 substrates in the MDCKcMDR1-KO and in the MDCK-hMDR1cMDR1-KO cell lines (b). Efflux ratios for six compounds (as selected from Broccatelli et al.21 in the MDCKcMDR1-KO and in the MDCK-hMDR1cMDR1-KO cell lines (c). The dashed lines indicate an efflux ratio of 1. Data are presented as mean ± standard deviation for one representative experiment performed in triplicate.
TE D
*p < 0.05; using student’s t-test. hMDR1 substrate classification for the six compounds in Figure c using NIH or NKI MDCK-hMDR1 cells as compiled and reported by Broccatelli et al.21 as well as hMDR1 substrate classification using the MDCK-hMDR1cMDR1-KO cells described in this paper (d).
EP
Efflux ratio cutoff values are shown in parentheses. A compound classified as an MDR1 substrate is
AC C
indicated with + and a compound classified as a non-substrate is indicated with -.
ACCEPTED MANUSCRIPT
Table 1. Transporter expression levels as determined using surrogate peptides in the MDCK wildtype (wt), MDCKcMDR1-KO, MDCK-hMDR1cMDR1-KO, and MDCK-hMDR1 cells. MDCK wt
MDCKcMDR1-KO
MDCK cells from filtersa cMDR1 + hMDR1 hMDR1 cBCRP + hBCRP
AGAVAEEVLAAIR FYDPLAGK SSLLDVLAAR
1.4±0.02** nd nd
nd nd nd
MDCK-hMDR1cMDR1-KO
MDCK-hMDR1
LOQ
9.5±2.3** 4.7±1.2 nd
14±2.8** 7.1±2.7 nd
0.054 0.010 0.054
RI PT
Peptide
SC
Protein
AC C
EP
TE D
M AN U
MDCK cells from flasksb cMDR1 + hMDR1 AGAVAEEVLAAIR 0.43
ED
M AN
CC EP TE D
SC
M AN U
RI P
S0022-3549(17)30251-4
DOI:
10.1016/j.xphs.2017.04.018
Reference:
XPHS 741
To appear in:
Journal of Pharmaceutical Sciences
Received Date: 15 February 2017 Revised Date:
11 April 2017
Accepted Date: 12 April 2017
Please cite this article as: Karlgren M, Simoff I, Backlund M, Wegler C, Keiser M, Handin N, Müller J, Lundquist P, Jareborg AC, Oswald S, Artursson P, A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an improved tool for drug efflux studies, Journal of Pharmaceutical Sciences (2017), doi: 10.1016/j.xphs.2017.04.018. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Karlgren
A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an improved tool for drug efflux studies ACCEPTED MANUSCRIPT
1 (14)
A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an
RI PT
improved tool for drug efflux studies
Karlgren Maria1,2*, Simoff Ivailo1, Backlund Maria1,2, Wegler Christine1,3, Keiser Markus4,
SC
Handin Niklas1, Janett Müller4, Lundquist Patrik1, Jareborg Anne-Christine5, Oswald Stefan4, Artursson Per1,2 1
2
M AN U
Department of Pharmacy, Uppsala University, Uppsala, Sweden
Uppsala University Drug Optimization and Pharmaceutical Profiling Platform (UDOPP), Science for
Life Laboratory, Uppsala, Sweden 3
Cardiovascular Metabolic Diseases DMPK, AstraZeneca R&D, Mölndal, Sweden
4
Department of Clinical Pharmacology, Center of Drug Absorption and Transport (C_DAT), University
5
TE D
Medicine of Greifswald, Germany
Department of immunology, genetics and pathology, Science for Life Laboratory, Uppsala University,
EP
Uppsala, Sweden
*Address correspondence to:
AC C
Maria Karlgren, PhD
Department of Pharmacy Uppsala University Box 580
SE-751 23 Uppsala, Sweden Email: [email protected] Phone: +46 – 18 471 41 49 Fax: +46 – 18 471 42 23
Karlgren
A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an improved tool for drug efflux studies ACCEPTED MANUSCRIPT
2 (14)
Abstract Madin-Darby canine kidney (MDCK) II cells stably transfected with transport proteins are commonly used models for drug transport studies. However, endogenous expression of especially canine MDR1 (cMDR1) confounds the interpretation of such studies. Here we have established an MDCK cell line
RI PT
stably overexpressing the human MDR1 transporter (hMDR1; P-glycoprotein), and used CRISPRCas9 gene editing to knock out the endogenous cMDR1. Genomic screening revealed the generation of a clonal cell line homozygous for a four-nucleotide deletion in the canine ABCB1 gene leading to a
SC
frameshift and a premature stop codon. Knockout of cMDR1 expression was verified by quantitative protein analysis and functional studies showing retained activity of the human MDR1 transporter.
M AN U
Application of this cell line allowed unbiased reclassification of drugs previously defined as both substrates and non-substrates in different studies using commonly used MDCK-MDR1 clones. Our new MDCK-hMDR1 cell line, together with a previously developed control cell line, both with identical deletions in the canine ABCB1 gene and lack of cMDR1 expression represent excellent in
TE D
vitro tools for use in drug discovery.
Keywords: MDCK cells, ABC transporters, Efflux pumps, P-glycoprotein, Permeability, Membrane
AC C
EP
transporter, Drug transport, Genomics, Proteomics, Multidrug resistance transporters
Karlgren
A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an improved tool for drug efflux studies ACCEPTED MANUSCRIPT
3 (14)
Introduction Multidrug resistance protein 1 (protein name: multidrug resistance protein 1 (MDR1) often referred to as P-glycoprotein (Pgp); gene name: ABCB1), discovered in 19761 modifies the pharmacokinetics of drugs by affecting their intestinal absorption and/or their biliary or renal secretion as well as their
RI PT
tissue distribution (e.g. across the blood-brain barrier). As MDR1 has a very broad substrate specificity it is routinely studied in drug discovery. Regulatory agencies such as the EMA and the FDA provide guidelines for in vitro investigations of putative substrates and inhibitors of MDR1 that
SC
may cause drug-drug interactions (DDI) with MDR1 substrates.2-5
Madine-Darby Canine Kidney II (MDCK) cells, a monolayer-forming epithelial cell line, is the most
M AN U
common model system for studying the influence of human efflux transporters on drug disposition. These cells possess the correct localization of clinically relevant transporters such as MDR1 and MRP2 and are therefore frequently used for heterologous overexpression of human transporters, singly or (as first shown by Keppler and Sugiyama) in combination with each other.6, 7 Confluent MDCK cells on filter supports form polarized, tight monolayers allowing studies of bidirectional
TE D
transepithelial transport.7-11 As human MDR1 (hMDR1) is sorted to the apical plasma membrane in these cells, the transport rate of a typical MDR1 substrate is lower from the apical to the basolateral side of the cells than in the opposite direction. The ratio between the permeability of these two
EP
transport directions, the so called efflux ratio (ER), is the simplest approach for identifying efflux substrates in various tissue barriers including the blood-brain-barrier.12
AC C
MDCK cells express endogenous canine kidney transporters like canine MDR1 (cMDR1), which was shown to influence transport of compounds.13 Therefore, contribution of cMDR1 in MDCK-hMDR1 cells (expressing both hMDR1 and cMDR1) needs to be subtracted by including MDCK wild type cells (expressing cMDR1) as a negative control in the experiment. Alternative approaches to overcome the endogenous background have included selection of clones with low levels of cMDR1, suppression of expression by inhibitory RNA, and knockouts using zinc-finger nucleases.14-17 These approaches have not been completely satisfying as background correction is necessary even for knockout cells using the zink-finger nuclease technology. Therefore, we recently used the CRISPR-Cas9 gene editing
Karlgren
A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an improved tool for drug efflux studies ACCEPTED MANUSCRIPT
4 (14)
technology to develop a MDCK cell line completely lacking endogenous cMDR1 background.18 In this cell line, all copies of the canine ABCB1 gene were modified, leading to a truncated protein without cMDR1 activity, which was confirmed using prototypic MDR1 substrates. By using the same CRISPR-Cas9 gene editing approach18 and a previously established MDCK-
RI PT
hMDR1 cell line19 as starting point, we now present the development of a MDCK-hMDR1cMDR1-KO cell line, which completely lacks cMDR1 background but retains hMDR1 activity. This MDCK-
hMDR1cMDR1-KO cell line, which we propose can be used without background correction, represents an
AC C
EP
TE D
M AN U
SC
improved tool for transport and DDI studies in vitro.
Karlgren
A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an improved tool for drug efflux studies ACCEPTED MANUSCRIPT
5 (14)
Methods Cell culture MDCK cells (obtained from American Type Culture Collection (ATCC CRL-2936, Rockville, MD) and MDCKcMDR1-KO cells18, were cultured as previously described.18 For MDCK cells overexpressing
RI PT
hMDR1 transporter, i.e.MDCK-hMDR119 and MDCK-hMDR1cMDR1-KO cells, culturing media was further supplemented with 375µg/ml Hygromycin B. Cell culture media and supplements were from Invitrogen (Carlsbad, CA), Sigma-Aldrich (St. Louis, MO) or from PAN Biotech (Aidenbach,
SC
Germany). Cells were cultured at 37°C, 95% humidity and 5% CO2 and sub-cultured twice a week. Knockout of canine MDR1
M AN U
Knockout of the canine ABCB1 gene using canine-specific CRISPR-Cas9 vectors was performed as described in18 and in the supplementary material. To confirm gene editing, the target regions of genomic DNA from individual cell clones were amplified by PCR and the PCR product was sequenced by Sanger sequencing at Uppsala Genome Center, Uppsala University.
TE D
Protein quantification of drug transporters
Samples from MDCK wildtype, MDCKcMDR1-KO, MDCK-hMDR1, and MDCK-hMDR1cMDR1-KO were subjected to protein quantification to determine expression levels of cMDR1 and hMDR1 as well as
EP
eight other transport proteins using two different types of sample preparation protocols prior to targeted proteomics (see supplementary material).
AC C
Transport assays
Transport experiments with 15 known or putative MDR1 substrates were performed at 1µM as previously described.18 Experiments were partly verified using radiolabeled compounds (see supplementary material). Digoxin experiments were performed using 2.5 µCi/ml HBSS buffer of 3Hdigoxin (26.3 Ci/mmol) (Perkin Elmer, Wiltham, MA) and 2 µM unlabeled digoxin (Sigma-Aldrich, St Louis, MO) with or without addition of 2 µM elacridar. For details see Simoff et al.18 Student’s ttest was used for statistical comparisons.
Karlgren
A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an improved tool for drug efflux studies ACCEPTED MANUSCRIPT
6 (14)
Results Knockout of canine MDR1 in MDCK-hMDR1 cells Genetic analysis revealed that homozygous indels occurred in four MDCK-hMDR1 clones out of 14 clones analyzed by genomic sequencing. Remaining clones exhibited an unaltered genotype or were
RI PT
heterozygous at exon 4 and/or 13. The clone selected for further analysis, here referred to as MDCKhMDR1cMDR1-KO, showed the same genotype as the previously established MDCKcMDR1-KO clone.18 Thus, both clones had a four-nucleotide deletion in exon 4 (Figure 1), leading to a shift of the reading
Protein quantification using surrogate peptides
SC
frame and three consecutive stop codons, resulting in a truncated protein.
M AN U
The knock-out of cMDR1 and overexpression of hMDR1 in MDCK cells were verified by quantification of MDR1 proteins in filter-grown cells with targeted proteomics using a human-specific peptide and a shared peptide detecting both cMDR1 and hMDR1 (Table 1). As expected, cMDR1 could only be quantified with the shared peptide (recognizing both cMDR1 and hMDR1) in the MDCK wildtype cells (on average 1.5 fmol/µg total protein). Further, cMDR1 was not detected in the
TE D
MDCKcMDR1-KO cells, with either peptide, demonstrating a successful knock-out. MDR1 was quantified in MDCK-hMDR1 as well as in MDCK-hMDR1cMDR1-KO cells with both peptides (on average 8.2 fmol/µg total protein; range 4.7-14.3 between the two samples, measured by the two peptides),
EP
confirming overexpression of hMDR1 in the cells (Table 1). The fold difference between the levels determined with the shared peptide and the human-specific peptide, as well as the absolute levels
AC C
measured with the shared peptide, was significantly higher in MDCK-hMDR1 than for MDCKhMDR1cMDR1-KO cells (p < 0.05, two tailed Mann-Whitney test, and p < 0.01, one-way ANOVA followed by Tukey's multiple comparisons test, respectively)(see supplementary table 3). This indicated a complete knock-out of cMDR1 in the MDCK-hMDR1cMDR1-KO cells. Successful knockout of cMDR1 and over-expression of hMDR1 was also confirmed in enriched crude membrane fractions of undifferentiated cells grown in cell culture flasks (Table 1). Further, quantification of other transporters showed only minor differences between the four cell lines (Table 1).
Karlgren
A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an improved tool for drug efflux studies ACCEPTED MANUSCRIPT
7 (14)
MDR1 mediated transport Transport experiments with the MDR1 substrate digoxin showed that hMDR1 function was retained in MDCK-hMDR1cMDR1-KO and parenteral MDCK-hMDR1 cells, giving higher Papp in the basolateral-to-
RI PT
apical (BA) than in the apical-to-basolateral (AB) direction (Figure 2a). Moreover, MDCK wildtype cells showed significant efflux. The MDR1 inhibitor elacridar reduced the ERs of these three cell lines close to unity. In contrast, for the MDCKcMDR1-KO cells, digoxin ER was close to unity both in presence
SC
and absence of the MDR1 inhibitor.
Nine prototypical MDR1 substrates were chosen for functional studies in the cMDR1 knock-out cell
M AN U
lines (Figure 2b). Seven of these substrates showed ERs close to unity while two compounds showed some residual efflux (ER close to 2) presumably via other endogenous transporters in the MDCKcMDR1KO
cells, whereas ERs ranged from 5 (for saquinavir) up to 362 (for ritonavir) in the MDCK-
hMDR1cMDR1-KO cells. To further test the robustness of these cell lines, loperamide, ritonavir, talinolol, and verapamil were investigated using radiolabeled compounds in a second laboratory. Also here, all
TE D
compounds were classified as MDR1 substrates, but at lower resolution, which was probably a result of the indirect determination of radioactivity (see supplementary material). To further test the applicability of the CRISPR generated cMDR1 knockout cell lines, six compounds
EP
with contradicting classifications using MDCK cell lines overexpressing hMDR1 developed by Prof. Piet Borst (from the Netherlands Cancer Institute; NKI) or by Pastan et al.20 (from the National
AC C
Institutes of Health; NIH)(cf. Broccatelli et al.21), were investigated using MDCKcMDR1-KO and MDCKhMDR1cMDR1-KO cells. In our study, four of these compounds showed a clear difference in ERs in the MDCKcMDR1-KO and the MDCK-hMDR1cMDR1-KO cells, respectively (Figure 2c and d). For two compounds, i.e. olanzapine and clozapine, no difference was observed between MDCKcMDR1-KO and MDCK-hMDR1cMDR1-KO cells. Taken together, four of these six compounds could be (re)classified as MDR1 substrates.
Karlgren
A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an improved tool for drug efflux studies ACCEPTED MANUSCRIPT
8 (14)
Discussion Different approaches have been used to correct for endogenous cMDR1 transport activity in MDCK cells (see for example recommendations by the International Transporter Consortium4, 5) in order to avoid misclassification and unnecessary follow-up studies of MDR1 substrates. All currently used
RI PT
approaches necessitate background correction in MDCK wildtype cells. Despite that it is well known that overexpression of transport proteins can alter the expression or function of endogenous
transporters13, 22, cMDR1expression levels in the wildtype and the overexpressing cells are generally
SC
considered equivalent. Furthermore, expression levels can vary from laboratory to laboratory, due to culturing conditions, or as a result of clonal selection during the generation of the overexpressing cell
M AN U
line.23
In a previous study we have used CRISPR-Cas9 to generate a MDCK cell line completely lacking endogenous cMDR1 expression.18 As the CRISPR-Cas9 guide RNAs (gRNAs) in that study were designed to be sequence-specific to canine ABCB1 only, herein we used these gRNAs to knock out
TE D
cMDR1 in a MDCK cell line already overexpressing hMDR1.
The sequences for cMDR1 and hMDR1 are very similar at both the genomic and proteomic levels,
EP
which makes it difficult to design species-specific CRISPR-Cas9 gRNAs and especially peptides for proteomics quantification. Therefore, no suitable canine-specific peptide could be identified. Hence,
AC C
we took another approach and used one hMDR-specific and one shared peptide that identified both hMDR1 and cMDR1. This allowed indirect quantification of cMDR1 expression by comparing the ratio of the two peptides in the different cell lines. When comparing expression levels determined with the shared and the human-specific peptide we could see that the protein quantification ratio between these peptides in the MDCK-hMDR1cMDR1-KO cells were identical to ratios obtained in human hepatocytes and human jejunum (i.e. tissues only expressing hMDR1; see supplementary table 3). In contrast, the ratio of the two peptides in MDCK-hMDR1 was significantly higher, reflecting expression of both cMDR1 and hMDR1 in this cell line. Furthermore, the high expression of hMDR1
Karlgren
A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an improved tool for drug efflux studies ACCEPTED MANUSCRIPT
9 (14)
in the MDCK-hMDR1cMDR1-KO cell line verified that the gRNAs were indeed canine-specific. Correct localization of hMDR1 in the apical membrane was verified by immunofluorescence staining (see supplementary material). Despite the complete knock-out of cMDR1, no compensatory up regulation
RI PT
of other efflux transporters could be detected.
Transwell experiments with prototypical MDR1 substrates further verified retained activity of the hMDR1, with the MDCK-hMDR1cMDR1-KO cells having significantly higher ERs compared to
MDCKcMDR1-KO cells for all investigated substrates. In addition, experiments with e.g. digoxin
SC
indicated that we may have selected a clone with high hMDR1 activity during the CRISPR-Cas9
to the parental MDCK-hMDR1 cells.
M AN U
clonal selection process, as higher ER was observed in the MDCK-hMDR1cMDR1-KO cells as compared
To further test the performance of our knockout models we examined six compounds, as identified by Broccatelli et al.21 who presented contradicting results on MDCK-hMDR1 cells of different origin.
TE D
Our hypothesis was that it should be possible to use our two knockout cell lines, MDCKcMDR1-KO and MDCK-hMDR1cMDR1-KO, in a standard screening set-up to obtain conclusive and easily interpretable results for these compounds. Indeed we observed a significant difference in the ERs of the
EP
MDCKcMDR1-KO and MDCK-hMDR1cMDR1-KO cells for four of these compounds. Our results were in better agreement with those obtained using the NIH cells than the NKI cells, either when using the
AC C
classical ER cutoff ≥2 or ≥8.5 as suggested by Broccatelli et al.21 for NIH cells. At an ER cutoff of 2, the classification of olanzapine was different in our cells as compared to the NIH cells (Figure 2d). Olanzapine has previously shown no or moderate affinity for MDR1 using different in vitro assays.24-26 In addition, olanzapine has previously displayed a high permeability and a low ER around the ER cutoff in MDCK-hMDR1 cells.27 The difference in classifications is therefore not surprising. However, if an ER cutoff of ≥8.5 (as suggested by Broccatelli et al.21) is used, olanzapine is classified as a non-substrate, both in our cell line and in the NIH cell line. In general it is difficult to determine in vitro if compounds with very high permeability are substrates of efflux transporters.28, 29
Karlgren
A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an improved tool for drug efflux studies ACCEPTED MANUSCRIPT
10 (14)
At an ER cutoff at 8.5, the classification of the more hydrophilic drug ranitidine differed in our cells and the NIH cells (Figure 2d). This discrepancy might be explained by differences in the experimental conditions as in the previous study ranitidine was tested at a concentration of 50µM30 whereas in this study, we used 1µM. We speculate that the high concentration (50µM) could partly have saturated the
RI PT
transporter and thereby led to false negative results.
Earlier attempts to obtain MDCK cells with lower or no cMDR1 background did not conclusively establish that the cell lines were devoid of endogenous background. CRISPR-Cas9 has emerged as the
SC
method of choice for controlled and precise generation of knock-out of specific genes in cell lines. After our previous publication of a MDCK cell line completely lacking cMDR1 expression, others
M AN U
have used similar CRISPR-Cas9 approaches to knockout hMDR1 in multi-drug resistant human cell lines31 and cMDR1 in MDCK32. Although functional studies indicate that hMDR1/cMDR1 has been knocked down/out in these studies, to our knowledge, MDR1 protein quantification has not been
TE D
performed to verify the knockouts.
In conclusion we here describe the development of an MDCK-hMDR1cMDR1-KO cell line. This cell line, and its matching control cell line MDCKcMDR1-KO, lack endogenous cMDR1 expression, as verified
EP
using genomic screening and proteomics analysis. Functional validation further showed no MDR1 activity in the control cell line MDCKcMDR1-KO, whereas high hMDR1 activity was retained in the
AC C
MDCK-hMDR1cMDR1-KO cell line.
Karlgren
A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an improved tool for drug efflux studies ACCEPTED MANUSCRIPT
11 (14)
Acknowledgements We are grateful to Maria Mastej and Elin Khan for assistance with clonal expansion of the cells and Maria Mastej for cell culturing. We acknowledge the staff at the BioVis platform, Science for Life Laboratory, Uppsala University, for assistance with FACS, microscopy and imaging. This work was
RI PT
supported by the Swedish Fund for Research without Animal Experiments, the Åke Wiberg foundation, the Magnus Bergvall foundation, the Swedish Research Council (approval number 5715), and the German Federal Ministry of Education and Research (BMBF, InnoProfile-Transfer, grant
AC C
EP
TE D
M AN U
SC
number 03IPT612X).
Karlgren
A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an improved tool for drug efflux studies ACCEPTED MANUSCRIPT
12 (14)
References
7.
8. 9.
10.
11.
12.
13.
14. 15. 16.
17. 18.
19.
RI PT
SC
6.
M AN U
5.
TE D
3. 4.
EP
2.
Juliano RL, Ling V. A surface glycoprotein modulating drug permeability in Chinese hamster ovary cell mutants. Biochim Biophys Acta. 1976;455(1):152-162. Guidance for Industry Drug Interaction Studies — Study Design, Data Analysis, Implications for Dosing, and Labeling Recommendations. U.S. Dep of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research; 2012. Guideline on the Investigation of Drug Interactions. European Medicines Agency; 2013. Giacomini KM, Huang SM, Tweedie DJ, Benet LZ, Brouwer KL, et al. Membrane transporters in drug development. Nature reviews Drug discovery. 2010;9(3):215-236. Tweedie D, Polli JW, Berglund EG, Huang SM, Zhang L, et al. Transporter studies in drug development: experience to date and follow-up on decision trees from the International Transporter Consortium. Clin Pharmacol Ther. 2013;94(1):113-125. Cui Y, Konig J, Keppler D. Vectorial transport by double-transfected cells expressing the human uptake transporter SLC21A8 and the apical export pump ABCC2. Mol Pharmacol. 2001;60(5):934-943. Sasaki M, Suzuki H, Ito K, Abe T, Sugiyama Y. Transcellular transport of organic anions across a double-transfected Madin-Darby canine kidney II cell monolayer expressing both human organic anion-transporting polypeptide (OATP2/SLC21A6) and Multidrug resistanceassociated protein 2 (MRP2/ABCC2). J Biol Chem. 2002;277(8):6497-6503. Gottesman MM, Fojo T, Bates SE. Multidrug resistance in cancer: role of ATP-dependent transporters. Nat Rev Cancer. 2002;2(1):48-58. Hammerle SP, Rothen-Rutishauser B, Kramer SD, Gunthert M, Wunderli-Allenspach H. PGlycoprotein in cell cultures: a combined approach to study expression, localisation, and functionality in the confocal microscope. Eur J Pharm Sci. 2000;12(1):69-77. Konig J, Zolk O, Singer K, Hoffmann C, Fromm MF. Double-transfected MDCK cells expressing human OCT1/MATE1 or OCT2/MATE1: determinants of uptake and transcellular translocation of organic cations. Br J Pharmacol. 2011;163(3):546-555. Schinkel AH, Mayer U, Wagenaar E, Mol CA, van Deemter L, et al. Normal viability and altered pharmacokinetics in mice lacking mdr1-type (drug-transporting) P-glycoproteins. Proc Natl Acad Sci U S A. 1997;94(8):4028-4033. Kalvass JC, Pollack GM. Kinetic considerations for the quantitative assessment of efflux activity and inhibition: implications for understanding and predicting the effects of efflux inhibition. Pharm Res. 2007;24(2):265-276. Kuteykin-Teplyakov K, Luna-Tortos C, Ambroziak K, Loscher W. Differences in the expression of endogenous efflux transporters in MDR1-transfected versus wildtype cell lines affect P-glycoprotein mediated drug transport. Br J Pharmacol. 2010;160(6):1453-1463. Celius T, Garberg P, Lundgren B. Stable suppression of MDR1 gene expression and function by RNAi in Caco-2 cells. Biochem Biophys Res Commun. 2004;324(1):365-371. Di L, Whitney-Pickett C, Umland JP, Zhang H, Zhang X, et al. Development of a new permeability assay using low-efflux MDCKII cells. J Pharm Sci. 2011;100(11):4974-4985. Gartzke D, Delzer J, Laplanche L, Uchida Y, Hoshi Y, et al. Genomic Knockout of Endogenous Canine P-Glycoprotein in Wild-Type, Human P-Glycoprotein and Human BCRP Transfected MDCKII Cell Lines by Zinc Finger Nucleases. Pharm Res. 2015;32(6):20602071. Gartzke D, Fricker G. Establishment of optimized MDCK cell lines for reliable efflux transport studies. J Pharm Sci. 2014;103(4):1298-1304. Simoff I, Karlgren M, Backlund M, Lindstrom AC, Gaugaz FZ, et al. Complete Knockout of Endogenous Mdr1 (Abcb1) in MDCK Cells by CRISPR-Cas9. J Pharm Sci. 2016;105(2):1017-1021. Siegmund W, Modess C, Scheuch E, Methling K, Keiser M, et al. Metabolic activation and analgesic effect of flupirtine in healthy subjects, influence of the polymorphic NAT2, UGT1A1 and GSTP1. Br J Clin Pharmacol. 2015;79(3):501-513.
AC C
1.
Karlgren
26.
27.
28.
29. 30.
31. 32.
RI PT
25.
SC
24.
M AN U
23.
TE D
22.
EP
21.
13 (14)
Pastan I, Gottesman MM, Ueda K, Lovelace E, Rutherford AV, et al. A retrovirus carrying an MDR1 cDNA confers multidrug resistance and polarized expression of P-glycoprotein in MDCK cells. Proc Natl Acad Sci U S A. 1988;85(12):4486-4490. Broccatelli F, Larregieu CA, Cruciani G, Oprea TI, Benet LZ. Improving the prediction of the brain disposition for orally administered drugs using BDDCS. Adv Drug Deliv Rev. 2012;64(1):95-109. Li J, Wang Y, Hidalgo IJ. Kinetic analysis of human and canine P-glycoprotein-mediated drug transport in MDR1-MDCK cell model: approaches to reduce false-negative substrate classification. J Pharm Sci. 2013;102(9):3436-3446. Dukes JD, Whitley P, Chalmers AD. The MDCK variety pack: choosing the right strain. BMC Cell Biol. 2011;12:43. Boulton DW, DeVane CL, Liston HL, Markowitz JS. In vitro P-glycoprotein affinity for atypical and conventional antipsychotics. Life Sci. 2002;71(2):163-169. Wang JS, Zhu HJ, Markowitz JS, Donovan JL, DeVane CL. Evaluation of antipsychotic drugs as inhibitors of multidrug resistance transporter P-glycoprotein. Psychopharmacology (Berl). 2006;187(4):415-423. Nagasaka Y, Oda K, Iwatsubo T, Kawamura A, Usui T. Effects of aripiprazole and its active metabolite dehydroaripiprazole on the activities of drug efflux transporters expressed both in the intestine and at the blood-brain barrier. Biopharm Drug Dispos. 2012;33(6):304-315. Wager TT, Chandrasekaran RY, Hou X, Troutman MD, Verhoest PR, et al. Defining desirable central nervous system drug space through the alignment of molecular properties, in vitro ADME, and safety attributes. ACS Chem Neurosci. 2010;1(6):420-434. Artursson P, Matsson P, Karlgren M. In vitro characterization of interactions with drug transporting proteins. In: Sugiyama T, Steffensen B, editors. Transporters in Drug Development. AAPS Advances in the Pharmaceutical Sciences Springer; 2014. Polli JW, Wring SA, Humphreys JE, Huang L, Morgan JB, et al. Rational use of in vitro Pglycoprotein assays in drug discovery. J Pharmacol Exp Ther. 2001;299(2):620-628. Wang Q, Rager JD, Weinstein K, Kardos PS, Dobson GL, et al. Evaluation of the MDRMDCK cell line as a permeability screen for the blood-brain barrier. Int J Pharm. 2005;288(2):349-359. Yang Y, Qiu JG, Li Y, Di JM, Zhang WJ, et al. Targeting ABCB1-mediated tumor multidrug resistance by CRISPR/Cas9-based genome editing. Am J Transl Res. 2016;8(9):3986-3994. Xiao Y, McCollum B, Pratt J, Hauser S, Bourner M, et al. Better Understand ADME Properties of Drugs Using MDR1 Knock In/Out in MDCKII Cells. Poster at the AAPS annual meeting and exposition. 2016.
AC C
20.
A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an improved tool for drug efflux studies ACCEPTED MANUSCRIPT
Karlgren
A CRISPR-Cas9 generated MDCK-hMDR1 cell line without endogenous cMDR1 (cABCB1) – an improved tool for drug efflux studies ACCEPTED MANUSCRIPT
14 (14)
Figure legends Fig 1. The CRISPR-Cas9 target region in exon 4 of the canine ABCB1 gene. Chromatograms for the parental MDCK-hMDR1 cell line and the CRISPR-Cas9 generated MDCK-hMDR1cMDR1-KO (a). The position of the four-nucleotide deletion which leads to a frameshift missense mutation and premature
RI PT
stop codons downstream in exon 4 is indicated with a box /line. Alignment of MDCK-hMDR1cMDR1-KO and corresponding sequences for the cMDR1 reference sequence (gene ID 403879) and the previously
SC
published MDCKcMDR1-KO 18 (b).
Fig 2. Efflux ratios for digoxin alone and in the presence of the MDR1 inhibitor elacridar in the
M AN U
MDCK wildtype (wt), MDCKcMDR1-KO, MDCK-hMDR1cMDR1-KO, and MDCK-hMDR1 cells (a). Efflux ratios for prototypic MDR1 substrates in the MDCKcMDR1-KO and in the MDCK-hMDR1cMDR1-KO cell lines (b). Efflux ratios for six compounds (as selected from Broccatelli et al.21 in the MDCKcMDR1-KO and in the MDCK-hMDR1cMDR1-KO cell lines (c). The dashed lines indicate an efflux ratio of 1. Data are presented as mean ± standard deviation for one representative experiment performed in triplicate.
TE D
*p < 0.05; using student’s t-test. hMDR1 substrate classification for the six compounds in Figure c using NIH or NKI MDCK-hMDR1 cells as compiled and reported by Broccatelli et al.21 as well as hMDR1 substrate classification using the MDCK-hMDR1cMDR1-KO cells described in this paper (d).
EP
Efflux ratio cutoff values are shown in parentheses. A compound classified as an MDR1 substrate is
AC C
indicated with + and a compound classified as a non-substrate is indicated with -.
ACCEPTED MANUSCRIPT
Table 1. Transporter expression levels as determined using surrogate peptides in the MDCK wildtype (wt), MDCKcMDR1-KO, MDCK-hMDR1cMDR1-KO, and MDCK-hMDR1 cells. MDCK wt
MDCKcMDR1-KO
MDCK cells from filtersa cMDR1 + hMDR1 hMDR1 cBCRP + hBCRP
AGAVAEEVLAAIR FYDPLAGK SSLLDVLAAR
1.4±0.02** nd nd
nd nd nd
MDCK-hMDR1cMDR1-KO
MDCK-hMDR1
LOQ
9.5±2.3** 4.7±1.2 nd
14±2.8** 7.1±2.7 nd
0.054 0.010 0.054
RI PT
Peptide
SC
Protein
AC C
EP
TE D
M AN U
MDCK cells from flasksb cMDR1 + hMDR1 AGAVAEEVLAAIR 0.43
ED
M AN
CC EP TE D
SC
M AN U
RI P