Establishment of Optimized MDCK Cell Lines for Reliable Efflux Transport Studies

RESEARCH ARTICLE – Pharmacokinetics, Pharmacodynamics and Drug Transport and Metabolism

Establishment of Optimized MDCK Cell Lines for Reliable Efflux Transport Studies DOMINIK GARTZKE, GERT FRICKER Ruprecht-Karls-University, Institute of Pharmacy and Molecular Biotechnology, Heidelberg 69120, Germany Received 18 October 2013; revised 16 January 2014; accepted 22 January 2014 Published online 15 February 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/jps.23901 ABSTRACT: Madin-Darby canine kidney (MDCK) cells transfected with human MDR1 gene (MDCK-MDR1) encoding for P-glycoprotein (hPgp, ABCB1) are widely used for transport studies to identify drug candidates as substrates of this efflux protein. Therefore, it is necessary to rely on constant and comparable expression levels of Pgp to avoid false negative or positive results. We generated a cell line with homogenously high and stable expression of hPgp through sorting single clones from a MDCK-MDR1 cell pool using fluorescenceactivated cell sorting (FACS). To obtain control cell lines for evaluation of cross-interactions with endogenous canine Pgp (cPgp) wild-type cells were sorted with a low expression pattern of cPgp in comparison with the MDCK-MDR1. Expression of other transporters was also characterized in both cell lines by quantitative real-time PCR and Western blot. Pgp function was investigated applying the Calcein-AM assay as well as bidirectional transport assays using 3 H-Digoxin, 3 H-Vinblastine, and 3 H-Quinidine as substrates. Generated MDCK-MDR1 cell lines showed high expression of hPgp. Control MDCK-WT cells were optimized in showing a comparable expression level of cPgp in comparison with MDCK-MDR1 cell lines. Generated cell lines showed higher and more selective Pgp transport compared with parental C 2014 Wiley cells. Therefore, they provide a significant improvement in the performance of efflux studies yielding more reliable results.  Periodicals, Inc. and the American Pharmacists Association J Pharm Sci 103:1298–1304, 2014 Keywords: ABC transporter; P-glycoprotein; MDCK; Cell lines; homogenous population; PCR; bidirectional transport assays

INTRODUCTION Interactions of drug candidates with ATP-binding cassette (ABC) transport proteins are thoroughly investigated in pharmaceutical industry. For that purpose, it is necessary to have an adequate cell culture model in the screening process to avoid false negative or false positive results at an early stage of drug discovery. The Madin-Darby canine kidney cell line type II transfected with the human MDR1 gene (MDCK-MDR1) is an established model for the identification of human Pglycoprotein (Pgp) substrates and inhibitors and serves as an important tool for drug candidate selection.1–4 A major aspect is the approval of promising pharmaceutical agents by governmental institutions, for example, the United States Food and Drug Administration (US FDA). In the “Guidance for Industry” released by the US FDA, rules of action recommend that new drug candidates should be assessed as Pgp substrates, inhibitors, or inducers. Among others, the MDCK cell line is accredited by the US FDA for in vitro transport studies5 and is widely used in transport studies.6–8 However, because of its nonhuman origin, endogenous canine transport proteins are expressed in this cell line.9 This might lead to an interference with the transfected transporter of interest. P-glycoprotein is one of the best-studied transport proteins in drug research and development because of its extremely wide substrate spectrum and its potential impact in the treatment of diseases. It is one reason for multidrug resistance often observed in cancer therapy10 and it is expressed in many organs Abbreviations used: Pgp, P-glycoprotein; MDCK, Madin-Darby canine kidney cell line; ABC transporters, ATP-binding cassette transporters; ER, efflux ratio; NER, net efflux ratio. Correspondence to: Gert Fricker (Telephone: +49-6221-548336; Fax: +496221-545971; E-mail: [email protected]) Journal of Pharmaceutical Sciences, Vol. 103, 1298–1304 (2014)  C 2014 Wiley Periodicals, Inc. and the American Pharmacists Association

1298

and barrier tissues like the blood–brain barrier or gut, liver, and kidney.11–13 In MDCK cells, the endogenously expressed canine Pgp (cPgp) can have substrates, which might not necessarily be substrates of the human Pgp (hPgp) and vice versa.14 Because of these differences, it is possible to obtain false positive candidates in transport assays. Therefore, the US FDA stated in its guidance to determine the so called “net efflux ratios” (NER), which is the ratio of the efflux ratio (ER) of the MDCK-MDR1 cells normalized to the ER of the untransfected wild-type cells (MDCK-WT).5 By calculating this NER, effects caused by endogenous transporters such as cPgp should be cancelled out. The high disadvantage of this method is the assumption of a similar background of endogenous transport proteins. However, this premise is not the case between the two cell lines15 and, therefore, relatively high ER in MDCK-WT cells occur even with reference compounds like Digoxin. To overcome this drawback, the aim of the present study was to establish subpopulations of MDCK-WT with minimal expression of canine Pgp as well as a subpopulation of MDCKMDR1 expressing significantly higher human Pgp. Simultaneously, these subpopulations of MDCK-WT and MDCK-MDR1 cell lines should also show comparable expression levels of the endogenous cPgp to assure a higher accuracy and stronger comparability in compound profiling studies. With this newly generated cell lines the rate of false results in transport studies would be reduced.

MATERIALS AND METHODS Cell Lines and Cell Culture Madin-Darby canine kidney cells type II (MDCK) transfected with human Pgp (MDR1, ABCB1) and MDCK-WT were provided by Prof. Piet Borst (The Netherlands Cancer Institute,

Gartzke and Fricker, JOURNAL OF PHARMACEUTICAL SCIENCES 103:1298–1304, 2014

1299

RESEARCH ARTICLE – Pharmacokinetics, Pharmacodynamics and Drug Transport and Metabolism

Table 1.

List of Primers Used for qRT-PCR

Protein Name

GenBank Nr.a /Ensembl Versionb

Bsep (ABCB11)

NM 001143932.1a

144

Mate1 (SLC47A1)

ENSCAFT00000028907.3b

115

Mate2-K (SLC47A2)

XM 858954.2a

172

Mct1 (SLC16A1)

ENSCAFT00000021207.2b

116

Ntcp (SLC10A1)

ENSCAFT00000026282.3b

119

Oat1 (SLC22A6)

ENSCAFT00000024510.1b

178

Oat2 (SLC22A7)

ENSCAFG00000023502.2b

206

Oat3 (SLC22A8)

ENSCAFG00000024439.2b

139

Oat4 (SLC22A11)

ENSCAFT00000022845.3b

237

Oat4C1 (SLCO4C1)

ENSCAFT00000046054.1b

235

Octn1 (SLC22A4)

ENSCAFT00000001320.3b

132

Octn2 (SLC22A5)

ENSCAFT00000001324.3b

241

Pept1 (SLC15A1)

NM 001003036.1a

278

Pept2 (SLC15A2)

ENSCAFT00000018583.3b

229

Mrp4 (ABCC4)

NM 001197174.1a

111

Product Length (bp)

Sequence TGCAGAGTCAAGGCGAGCCG TGGAGCGTTGTCGGAGTGAAG AGTGGGCCCTGAGAACCGTGG GCCTGTGGGCTGGACGCAA GGGCGCTCTTCCTCAACACCG CATAGCTCCTTGCTGGGCCCCT TTGGAGGTCTCGGGCTTGCCT GAACACAGGGCTGCCTGCCA CACGGCCCCGCTCAACTTCA TGGTGCAACCCAGCGAGAGC GGGCTCCTGGCTCCAGAGAAGA CTTCCCGGAGTGGGTGGCTCA CACTGGCATGGCTCTGGCTGG ACAACAGGAGGCCCTTCAGGCA GCTGGACGTGGGACACTGCAC GCCCTTCAGGCAGAGCCTTGG AGTGGCAGCTCCTGGACCCC GCTTCCGCCCAAACCGGTCA GGTCAACTGTTGCTGGGAACTGGG CCCCCAGCCATCGTGGATCATCTT AGGAACATGGCTGTGGGGGTCA AATGAAGACAGTCAGGCTGCCCA CATGGGCGTGGGTGTCAGCTC CCTTGTGTGGCCTGGAGTTTGCC GTCCTCTCTCCCGGACTGCCC GCGATCAGTGCGCCGAGGAT GTGGCCTGGTACAGTGGGCTG AGCCAGGAACTGGGGTTTTGGAAG TTGGAGTGGCAGTGGCGGTG GTTTGTCATCCCTCGACGTTGCT

Forward/Reverse f r f r f r f r f r f r f r f r f r f r f r f r f r f r f r

Cultivation medium was removed from MDCK cells and cells were washed twice with phosphate-buffered saline (PBS). Afterward, the cells were incubated with 1 :M Calcein-AM for 30 min in serum-free medium. Then, the cells were trypsinized and counted; 1,000,000 cells were taken and centrifuged for 10 min at 100g. The pellet was resuspended in 1 mL PBS with 2% FBS and again centrifuged. This step was repeated three times to wash the cells. The preparative sorting was done with a FACSAriaTM flow cytometer (Becton Dickinson, Heidelberg, Germany). Single cells were sorted for low activity of Pgp in MDCK-WT cells and high Pgp activity in MDCK-MDR1 cells, respectively. Thereby, Calcein-AM did not discriminate between human and canine Pgp.

and purity were measured spectrophotometrically at 260 and 280 nm using a NanoDrop 2000 photometer (PeqLab, Erlangen, Germany). For quantitative real-time PCR (qRT-PCR), RNA was reversed transcribed using iscriptTM Reverse Transcription for qRT-PCR (Bio-Rad Laboratories GmbH, Munich, Germany) according to the instructions of the manufacturer. QuantiFast SYBR Green PCR Kit (Qiagen, Hilden Germany) was used for qRT-PCR reaction. According to the protocol, the following cycling conditions were used: 95◦ C for 5 min, 95◦ C for 10 s, and 60◦ C for 30 s (for 40 cycles). After every run, a melting curve analysis was performed to verify the specificity of the PCR products. The qRT-PCR analysis was performed on a LightCycler carrousel based system using LightCycler 3 software (Roche Applied Science, Mannheim, Germany). To quantify qRT-PCR gene results, the standard curve method according to manufacturer specifications of the software was utilized. Results were normalized to the housekeeping gene glyceraldehyde-3phosphate dehydrogenase (Gapdh). Additionally, a standard with defined dilutions of cDNA amplified with Gapdh were measured on every run to quantify the amount of the gene of interest for a better normalization. Table 1 shows a list of primers used for performing qRT-PCR. The primers for hPgp, cPgp, cMrp2, cMrp1, cMrp5, and cGapdh were used as published previously.15

Quantitative Real-Time PCR

Western Blot

Amsterdam, Netherlands). For the cultivation of the cells, Dulbecco’s modified Eagle medium with 3.7 g/L NaHCO3 , 4.5 g/L D-glucose and L-glutamine was used. The medium was supplemented with heat-inactivated fetal bovine serum (FBS) and penicillin (10,000 U/mL)–streptomycin 10,000 :g/mL. The medium and the supplements were obtained from Biochrom AG (Berlin, Germany). The cells were grown under standard conditions (37◦ C, 95% relative humidity, 5% CO2 ) and were used for >50 passages. Preparative Sorting

 R

Total RNA was isolated from the cells using RNeasy Mini Kit and QIAshredderTM columns (Qiagen, Hilden, Germany) according to the manufacturers protocol. RNA concentrations DOI 10.1002/jps.23901

R

R

Membrane proteins were isolated from cultured cells using the ProteoExtract Native Membrane Protein Extraction Kit (Calbiochem ; Merck Millipore, Darmstadt, Germany). R

R

Gartzke and Fricker, JOURNAL OF PHARMACEUTICAL SCIENCES 103:1298–1304, 2014

1300

RESEARCH ARTICLE – Pharmacokinetics, Pharmacodynamics and Drug Transport and Metabolism

Membrane proteins were separated on a NuPage 4%–12% BisTris Gel (Novex Life Technologies GmbH, Darmstadt, Germany) and transferred to a polyvinylidene fluoride membrane (Merck Millipore, Darmstadt, Germany). Primary antibodies for Pgp protein (C129) Na-K-ATPase (M7-PB-E9) were used in combination with antimouse IgG-HRP (Enzo, L¨orrach, Germany). Bands were detected on a BIO-RAD Universal Hood II (Bio-Rad Laboratories GmbH) using Western Lightning PlusECL (PerkinElmer , Waltham, Massachusetts). Band intensities were measured utilizing Quantity One 1-D Analysis software (Bio-Rad Laboratories GmbH).

Efflux ratio (ER) = Papp, B−A /Papp, A−B

(3)

Statistics Data are expressed as mean ± SD. Statistical analysis was performed as two tailed, nonparametric Mann–Whitney t-test.

R

Calcein-AM Assay The Calcein-AM assay was based on our previously developed method.16 The cells were seeded in 96-well plates at a density of 300,000 cells/cm2 . The assay was performed at day 6 after seeding. At first, cells were washed twice with Krebs-Ringer bicarbonate buffer (KRB). Afterward, the cells were incubated with 1 :M Calcein-AM for 30 min at 37◦ C. Calcein-AM was removed and the cells were again washed twice with KRB followed by an incubation with 1% Triton-X at 37◦ C for 20 min. The fluorescence intensity was measured at wavelength of 485/520 nm on Fluoroskan Ascent (Labsystems, Frankfurt, Germany). Bi-Directional Transport Studies MDCK cells were seeded onto 96-well transwell plates with a polycarbonate membrane and a pore size of 0.4 :m (Merck Millipore, Billerica, Ma, USA). On day 1, the cells were seeded at a density of 300,000 cells/cm2 and the assay was performed on day 5. At the beginning of the experiment, the culture medium was removed and the cells were washed with Hank’s Balanced Salt Solution adjusted to a pH of 7.4 (HBSS; Gibco Life Technologies GmbH, Darmstadt, Germany). HBSS was removed and the cells were preincubated with fresh HBSS for 30 min. The radioactive substrates H3 -Quinidine (60–90 Ci/mmol), H3 Vinblastine (5–25 Ci/mmol), and H3 -Digoxin (15–40 Ci/mmol) were used. All substrates were purchased from American Radiolabeled Chemicals (Saint Louis, Missouri). 10 :Ci of each desired substrate was added to a total reaction volume of 100 :L (apical) or 200 :L (basolateral) HBSS. Depending on the desired transport direction, radioactive substrate solution was added to the apical compartment (A–B direction) or to the basolateral compartment (B–A direction), respectively. Samples were taken at 0 and 60 min with a volume of 50 :L of each compartment. Samples were transferred to a 96-well plate and 150 :L of scintillation fluid was added. As scintillation fluid, UltimaGoldTM (PerkinElmer ) was used. Radioactivity of the samples was measured in a MicroBeta2TM scintillation counter (PerkinElmer ). Transport study data were analyzed as follows: R

R

Papp (nm/s) = (Q/t) ∗ [1/(A × C0 )]

(1)

Q is the amount of drug solute transported (:mole), t represents the incubation time (s), A is the monolayer surface area (cm2 ), and C0 is the mean of starting-end drug concentration in the donor well (:M) Efflux ratio (ER) = Papp,B−A /Papp,A−B Papp (nm/s) = (Q/t) ∗ [1/(A × C0 )]

(2)

Gartzke and Fricker, JOURNAL OF PHARMACEUTICAL SCIENCES 103:1298–1304, 2014

RESULTS For standardized analysis, it is very important to use defined and homogeneous cell culture populations, which keep their characteristics over a long period of time. Flow cytometry analysis of the MDCK-MDR1 cells (MDR1) and MDCK-WT cells (MDCK) regarding Pgp expression showed a heterogeneous distribution (data not shown). For this reason, we aimed to generate more homogeneous cell populations by preparative sorting of single cells via fluorescence-activated cell sorting. For the MDR1 cells, single clones were sorted for high Pgp expression. To obtain a homogeneous control cell population in MDCK cells, clones were sorted for low Pgp expression. These clones were named MDR1 high and MDCK low, respectively. In order to verify the assumption of a population, which keeps its desired expression pattern and functional characteristics over a longer period of time to obtain reliable data, further experiments were performed. We investigated defined populations (clones) by comparing with them unsorted cell groups for Pgp expression levels and functional transport properties. The expression of human Pgp (hPgp) in the sorted single cell clones showed a significantly higher expression level compared with the unsorted cell pool of MDCK cells transfected with hPgp (Fig. 1a). After approximately 40 passages, the expression of hPgp in the MDR1 high clones was still significantly higher and showed a stable high expression level compared with unsorted control cells. As described previously, the MDR1 cells display less expression of endogenous canine Pgp compared with the nontransfected parental MDCK cells.15 This discrepancy was statistically significant for all passages (not highlighted in Fig. 1). As depicted in Figure 1b, the sorted clones MDCK low expressed significantly less cPgp compared with unsorted MDCK cells. These sorted clones also showed constantly lower expression after 40 passages. One purpose of the study was having cells with comparable expression patterns in MDCK-MDR1 and the MDCK-WT cells. With the sorted cell lines, a better convergence could be achieved. In addition to the gene expression levels, the protein expression levels were investigated by Western blot analysis. Pgp protein amounts were determined in every cell line. In MDR1 cells, totally detected Pgp consisted of hPgp and cPgp, in contrast to only cPgp in MDCK-WT cells. The Pgp protein level of the MDR1 high cells was increased compared with the parental cell line (MDCK-MDR1). This higher level was stable over 40 passages. The MDCK low cells showed less Pgp protein level compared with the MDCK-WT cells. It became obvious by qRT-PCR and Western blot that the mRNA levels and the protein levels of the different cells correlate very well (Fig. 2). Because expression patterns are not sufficient for a comprehensive characterization of the transport protein status of the cells, additional transport studies were performed. One DOI 10.1002/jps.23901

RESEARCH ARTICLE – Pharmacokinetics, Pharmacodynamics and Drug Transport and Metabolism

1301

Figure 2. Correlation of mRNA and protein expression of Pgp. Data from Western blot and qRT-PCR from MDR1 cells and MDCK-WT cells with their respective clones are shown. For Western blotting, peak densities of the protein bands were detected. Normalization of the values was done referring to Na-K-ATPase as a housekeeping gene. P10 = Passage 10, P50 = Passage 50, n = 3.

Figure 1. mRNA expression level of Pgp in different cell populations. RNA was isolated from MDCK, MDR1, and cultivated clones. Results from quantitative real-time PCR (qRT-PCR) from human Pgp (hPGP) and canine Pgp (cPgp) are shown. The genes were normalized to the expression of endogeneous Gapdh mRNA. (a) Represents results of hPGP and (b) represents results for cPgp. Mean ± SD of at least four experiments. P10 = Passage 10, P50 = Passage 50. *p < 0.05.

assay to confirm expression pattern of Pgp is the Calcein-AM assay (Fig. 3), which measures the intracellular accumulation of fluorescent Calcein. Calcein-AM is a substrate of Pgp and emits no fluorescent signal itself. Intracellularly, the esterbond is cleaved by esterases and the free Calcein displays a fluorescent signal. Underlying this mechanism, higher fluorescence intensity indicates lower activity of Pgp and vice versa lower signal indicates higher activity of Pgp.17 The MDCK low cells showed significantly higher fluorescence intensities, implying a reduced activity of Pgp in comparison with MDCK-WT (Fig. 3). On the other hand, the MDR1 high cells exhibited decreased fluorescence intensities, which were significantly lower compared with its parental cell line, indicating elevated Pgp activity. To obtain more specific information, radioactive transport assays were performed. In Figure 4, the NERs of different radiolabeled substrates are shown. 3 H-Digoxin, 3 H-Vinblastine, and 3 H-Quinidine were used as substrates and ERs were determined. With the ERs of the MDCK-MDR1 cell line compared with the ratio of the WT cell line, the NERs were calculated. According to the “Guidance for Industry,” a NER above 2 identifies a compound as substrate of the specific human transporter.5,18 The ERs of MDR1 high were significantly higher than the values of unsorted cells. In contrast, MDCK low cells show decreased ERs (not shown). This leads to higher NERs of the DOI 10.1002/jps.23901

Figure 3. Calcein-AM assay displaying Pgp activity. Cells were treated with 10 :M Calcein-AM and the fluorescence intensity was measured after a treatment with 1% Triton-X at wavelengths 485/520 nm. Results describe the fluorescence intensity of MDR1-MDCK, MDCK-WT cells and their respective single cell clones. High fluorescence mirrors low Pgp activity and otherwise low fluorescence implies high Pgp activity. Mean ± SD; n = 3. ***p < 0.001.

sorted cells compared with the NERs of the parental cell lines. Along with the expression levels of Pgp, the clones showed still higher functional ratios after 40 passages, which remain very stable. Quinidine revealed an average NER of around 3 for the parental cell lines, whereas the sorted cell lines displayed a NER of 10. In addition, for Vinblastine, the unsorted cells showed NERs of 1.5 in contrast to a value of 4 for the sorted cells. The NER of Digoxin increased from a value of 2 to a value of 3.5. This led to unambiguous identification of the three reference hPGP substrates. Still, there are endogenous ER effects that could be connected to residual cPgp or other endogenous transporters. For this reason, we were interested in a more precise characterization. We investigated various other transporters and their expression pattern in the cell lines to ensure having an adequate correlation between both cell lines. Besides Pgp, several transporters are expressed in MDCK cell lines (Figs. 5c and 5d). As described previously, it is noticeable that multidrug resistance related protein (Mrp2) is decreased in the MDR1 cell line compared with the Gartzke and Fricker, JOURNAL OF PHARMACEUTICAL SCIENCES 103:1298–1304, 2014

1302

RESEARCH ARTICLE – Pharmacokinetics, Pharmacodynamics and Drug Transport and Metabolism

there are other transporter genes expressed, most notably Mct1 and Octn2. In our studies, the expression level of Mct1 was similar to the level of cPgp. The expression of these several transporters represented in Figures 5c and 5d is shown for the parental cell lines MDR1 and MDCK, however, there were no differences detectable in comparison with the sorted cell lines (data not shown).

DISCUSSION

Figure 4. Transport studies with different radiolabeled Pgp substrates. Data show NERs of (a) 3 H-Quinidine, (b) 3 H-Vinblastine, (c) 3 H-Digoxin showing NERs. Line marks threshold value of 2. n = 3, *p < 0.05, **p < 0.01.

MDCK-WT.15 Interestingly, the sorted cells both showed a decrease in Mrp2 expression compared with its parental cells and both subclones were on a similar level. There is some reduction for Mrp4 as well, but the expression is still higher in the nontransfected cell line. The expression of all other genes (Mrp1, Mrp5) remains unchanged (Figs. 5a and 5b). Bcrp could not be detected in the cell lines. But, next to the ABC transporters, Gartzke and Fricker, JOURNAL OF PHARMACEUTICAL SCIENCES 103:1298–1304, 2014

It is well known that the expression of endogenous transport proteins differs between MDCK-WT and a transfected MDCKMDR1 variant,15 making interpretation of drug–transport interactions difficult. These findings, in addition to inconsistent data obtained from internal compound characterization, were the reason for the present investigation. First results confirmed the different expression pattern in showing that cPgp as well as Mrp2 were markedly decreased in the MDCK-MDR1 cells. One potential approach to solve this issue was a preparative sorting of the MDCK cells with Calcein-AM as a Pgp substrate in flow cytometry.19 Using this method, we saw a similar distribution as Di et al.19 of Pgp within the MDCK-WT and MDCK-MDR1 cells (data not shown). This distribution indicated a heterogeneous expression pattern of Pgp in the cell culture population with some cells having higher Pgp expression and therefore higher activity than other cells. Single cells were sorted to establish a more homogeneous cell population, which might lead to a stable activity over a longer period of passages. Expression analysis of mRNA levels and protein levels confirmed our hypothesis. The MDR1 high cells exhibit a significant higher level of human Pgp while showing a stronger decrease of endogenous cPgp (Figs. 1a, 1b, and 2). Several reasons could be responsible for the difference in the expression of endogenous cPgp in MDCK-MDR1 and MDCK-WT cells. The most likely explanation is a negative regulatory feedback, down regulating the endogenous Pgp.20,21 Furthermore, clonal differences between the WT and the transfected cell line or passage variations are also a possible explanation.15 Our observation that endogenous Mrp2 (cMrp2) was reduced in addition to cPgp in the sorted MDR1 high cell line is supporting the hypothesis of the regulatory feedback (Figs. 5a and 1b). The major effort was the establishment of two cell lines with a low but comparable expression of cPgp. The results of qRT-PCR and Western blot analysis (Figs. 1 and 2) confirm that the cPgp is significantly reduced in sorted cell lines, both the MDR1 high and MDCK low cells, and also show better comparability in their expression pattern compared with the parental cell lines. There are no significant differences between the sorted cell lines. The other aspect was to obtain a cell line that can be used for a longer period of passages. The results of qRT-PCR and Western blot at an early stage of passage and about 40 passages later indicate that the expression of cPgp is decreasing in the course of passaging. But it is also obvious that cPgp is decreasing in MDCK low and MDR1 high both in a similar range (Fig. 1b). In Figure 1b, it is shown that the MDCK-MDR1 cells have a decrease in cPgp expression as well, but the MDCK-WT did not decrease in a same range. In this case, the discrepancy between these two cell lines becomes even more pronounced. These results lead to the implication that the MDR1 high and MDCK low cells, when cultivated in parallel, facilitate cell populations with a comparable expression pattern for more reliable DOI 10.1002/jps.23901

RESEARCH ARTICLE – Pharmacokinetics, Pharmacodynamics and Drug Transport and Metabolism

Figure 5. mRNA expression level of various transporters. (a) Showing expression patterns of several Mrps in parental MDR1 and MDCKWT cells. (b) Expression of Mrp genes in sorted cell lines. (c) and (d) depict several other transporters tested for their expression in the parental cell lines, MDR1 and MDCK. (c) Represents the results for the WT cells, whereas (d) demonstrates findings for the MDR1 cells. RNA was analyzed via quantitative real-time PCR (qRT-PCR) and gene expressions were normalized to Gapdh. Data are shown as mean ± SD of at least three experiments. n.d.: not detectable. DOI 10.1002/jps.23901

1303

results. In contrast, even parallel cultivation of the parental cell lines could not achieve this. To further characterize the sorted cell lines, the activity of Pgp was investigated in two different assays. The results of the Calcein-AM assay were supporting the previous expression results. The MDR1 high cells demonstrated a higher Pgp activity compared with the MDCK-MDR1 cells. This higher activity was observed in similar passages. On the other hand, MDCK low cells showed less activity in comparison with MDCK-WT cells. These first functional results demonstrated a successful establishment of optimized subpopulations of cells for more sensitive transport studies. To confirm this hypothesis, bidirectional transport studies with radiolabeled substances Quinidine, Vinblastine, and Digoxin as well-known substrates of Pgp were performed.22–25 For these compounds, a strong improvement of the calculated NERs was achieved. For example, Quinidine displayed a fivefold increase in the NER (Fig. 4a). Vinblastine showed a NER below 2 for the parental cell lines implicating that Vinblastine would not be a substrate of human Pgp but is considered as a hPGP substrate based on current regulatory guidance. The NER of the sorted cell lines in contrast was also about fourfold increased and demonstrated a ratio around 4 (Fig. 4b). This means, the new generated cell lines indentified Vinblastine as a substrate of human Pgp as previously reported. An increasing NER was also observed for Digoxin (Fig. 4c). This increase was stable over a similar number of passages, which underlines the improvement in the identification of compounds as substrates of human Pgp. In transport studies, however, inconsistencies may occur. Next to the already mentioned main players in transport studies, it is important to understand the overall expression pattern of transporter proteins as molecules can be substrates for more than one transporter. Therefore, we investigated our cell lines for expression of various other transporters beside Pgp. Along with the ATP-dependent efflux pumps, the solute carrier (SLC) superfamily is known to have influence in drug disposition.26,27 The International Transporter Consortium published the distribution of various transporter proteins in different organs in humans in a review article.28 On the basis of this article, different transporters were chosen for investigation. In Figures 5c and 5d, the results of this screening are shown for MDCK-WT and MDCK-MDR1 cells. Most notably, the monocarboxylic acid transporter 1 (MCT1) and the organic cation/carnitine transporter (OCTN2), both part of the SLC superfamily, and Mrp4 are expressed. Next to cPgp and Mrp2, Mrp4 showed the highest expression among the ABC transporter with efflux function. In a previous study, especially MCT1, recognizing short-chain monocarboxylates and small drugs with carboxylate groups like salicylate, was expressed in numerous cell lines like CaCo-2, HEK293, or HeLa.29 Having this differentiated knowledge of the used cell lines helps to evaluate obtained results from performed assays. We could clearly demonstrate an improvement with the new generated cell lines MDCK low and MDR1 high regarding reliability of transport study data due to a reduction of background expression of canine Pgp and an increase of hPGP in MDCKMDR1 cell line. The heterogeneous expression of endogenous and transfected transporters is an often-neglected problem, when using such cell lines for drug screening purposes. Therefore, based on our present findings, we recommend laboratories to develop more homogeneous cell lines or that even regulatory agencies specify cell lines and subclones thereof to use. Gartzke and Fricker, JOURNAL OF PHARMACEUTICAL SCIENCES 103:1298–1304, 2014

1304

RESEARCH ARTICLE – Pharmacokinetics, Pharmacodynamics and Drug Transport and Metabolism

REFERENCES 1. Horio M, Chin KV, Currier SJ, Goldenberg S, Williams C, Pastan I, Gottesman MM, Handler J. 1989. Transepithelial transport of drugs by the multidrug transporter in cultured Madin-Darby canine kidney cell epithelia. Journal of hypertension. Supplement : official journal of the International Society of Hypertension 264(25) 14880–4 [Pub Med] 2. Pastan I, Gottesman MM, Ueda K, Lovelace E, Rutherford AV, Willingham MC. 1988. A retrovirus carrying an MDR1 cDNA confers multidrug resistance and polarized expression of P-glycoprotein in MDCK cells. Proc Natl Acad Sci USA 85(12):4486–4490. 3. Polli JW, Wring SA, Humphreys JE, Huang L, Morgan JB, Webster LO, Serabjit-Singh CS. 2002. Rational use of in vitro P-glycoprotein assays in drug discovery. J Pharmacol Exp Ther 299(2):620–628. 4. Yamazaki M, Neway WE, Ohe T, Chen I, Rowe JF, Hochman JH, Chiba M, Lin JH. 2001. In vitro substrate identification studies for pglycoprotein-mediated transport: Species difference and predictability of in vivo results. J Pharmacol Exp Ther 296(3):723–735. 5. Food and Drug Administration (FDA), U. S. D. o. H. a. H. S. Guidance for industry. 2006. Drug interaction studies—Study design, data analysis, and implications for dosing and labelling. US Food and Drug Administration, MD, USA. 6. Cho MJ, Thompson DP, Cramer CT, Vidmar TJ, Scieszka JF. 1989. The Madin Darby canine kidney (MDCK) epithelial cell monolayer as a model cellular transport barrier. Pharm Res 6(1):71–77. 7. Irvine JD, Takahashi L, Lockhart K, Cheong J, Tolan JW, Selick HE, Grove JR. 1999. MDCK (Madin-Darby canine kidney) cells: A tool for membrane permeability screening. J Pharm Sci 88(1):28– 33. 8. Braun A, Hammerle S, Suda K, Rothen-Rutishauser B, Gunthert M, Kramer SD, Wunderli-Allenspach H. 2000. Cell cultures as tools in biopharmacy. Eur J Pharm Sci 11 Suppl 2:S51–S60. 9. Goh LB, Spears KJ, Yao D, Ayrton A, Morgan P, Roland Wolf C, Friedberg T. 2002. Endogenous drug transporters in in vitro and in vivo models for the prediction of drug disposition in man. Biochem Pharmacol 64(11):1569–1578. 10. Germann UA. 1996. P-glycoprotein—A mediator of multidrug resistance in tumour cells. Eur J Cancer 32A(6):927–944. 11. Thiebaut F, Tsuruo T, Hamada H, Gottesman MM, Pastan I, Willingham MC. 1987. Cellular localization of the multidrug-resistance gene product P-glycoprotein in normal human tissues. Proc Natl Acad Sci USA 84(21):7735–7738. 12. Cordon-Cardo C, O’Brien JP, Casals D, Rittman-Grauer L, Biedler JL, Melamed MR, Bertino JR. 1989. Multidrug-resistance gene (Pglycoprotein) is expressed by endothelial cells at blood–brain barrier sites. Proc Natl Acad Sci USA 86(2):695–698. 13. Schinkel AH, Smit JJ, van Tellingen O, Beijnen JH, Wagenaar E, van Deemter L, Mol CA, van der Valk MA, Robanus-Maandag EC, te Riele HP. 1994. Disruption of the mouse mdr1a P-glycoprotein gene leads to a deficiency in the blood-brain barrier and to increased sensitivity to drugs. Cell 77(4):491–502. 14. Hochman J, Mei Q, Yamazaki M, Tang C, Prueksaritanont T, Bock M, Ha S, Lin J. 2006. Role of mechanistic transport studiesin lead optimization. Biotechnol: Pharm Aspects (4):25–47.

Gartzke and Fricker, JOURNAL OF PHARMACEUTICAL SCIENCES 103:1298–1304, 2014

15. Kuteykin-Teplyakov K, Luna-Tortos C, Ambroziak K, Loscher W. 2010. Differences in the expression of endogenous efflux transporters in MDR1-transfected versus wildtype cell lines affect P-glycoprotein mediated drug transport. Br J Pharmacol 160(6):1453–1463. 16. Bubik M, Ott M, Mahringer A, Fricker G. 2006. Rapid assessment of p-glycoprotein–drug interactions at the blood-brain barrier. Anal Biochem 358(1):51–58. 17. Essodaigui M, Broxterman HJ, Garnier-Suillerot A. 1998. Kinetic analysis of calcein and calcein-acetoxymethylester efflux mediated by the multidrug resistance protein and P-glycoprotein. Biochemistry 37(8):2243–2250. 18. Feng B, Mills JB, Davidson RE, Mireles RJ, Janiszewski JS, Troutman MD, de Morais SM. 2008. In vitro P-glycoprotein assays to predict the in vivo interactions of P-glycoprotein with drugs in the central nervous system. Drug Metab Dispos 36(2):268–275. 19. Di L, Whitney-Pickett C, Umland JP, Zhang H, Zhang X, Gebhard DF, Lai Y, Federico JJ 3rd, Davidson RE, Smith R, Reyner EL, Lee C, Feng B, Rotter C, Varma MV, Kempshall S, Fenner K, El-Kattan AF, Liston TE, Troutman MD. 2010. Development of a new permeability assay using low-efflux MDCKII cells. J Pharm Sci 100(11):4974–4985. 20. Lloyd C, Schevzov G, Gunning P. 1992. Transfection of nonmuscle beta- and gamma-actin genes into myoblasts elicits different feedback regulatory responses from endogenous actin genes. J Cell Biol 117(4):787–797. 21. Oswald S, May K, Rosin J, Lutjohann D, Siegmund W. 2010. Synergistic influence of Abcb1 and Abcc2 on disposition and sterol lowering effects of ezetimibe in rats. J Pharm Sci 99(1):422–429. 22. Silverman JA. 1999. Multidrug-resistance transporters. Pharm Biotechnol 12:353–386. 23. Chang C, Bahadduri PM, Polli JE, Swaan PW, Ekins S. 2006. Rapid identification of P-glycoprotein substrates and inhibitors. Drug Metab Dispos 34(12):1976–1984. 24. Szakacs G, Paterson JK, Ludwig JA, Booth-Genthe C. Gottesman MM. 2006. Targeting multidrug resistance in cancer. Nat Rev Drug Discov 5(3):219–234. 25. Patil AG, D’Souza R, Dixit N, Damre A. 2011. Validation of quinidine as a probe substrate for the in vitro P-gp inhibition assay in Caco-2 cell monolayer. Eur J Drug Metab Pharmacokinet 36(3):115–119. 26. Schinkel AH, Jonker JW. 2003. Mammalian drug efflux transporters of the ATP binding cassette (ABC) family: An overview. Adv Drug Deliv Rev 55(1):3–29. 27. Hediger MA, Romero MF, Peng JB, Rolfs A, Takanaga H, Bruford EA. 2004. The ABCs of solute carriers: Physiological, pathological and therapeutic implications of human membrane transport proteinsIntroduction. Pflugers Arch 447(5):465–468. 28. Giacomini KM, Huang SM, Tweedie DJ, Benet LZ, Brouwer KL, Chu X, Dahlin A, Evers R, Fischer V, Hilgren KM, Hoffmaster KA, Ishikawa T, Keppler D, Kim RB, Lee CA, Niemi M, Polli JW, Sugiyama Y, Swaan PW, Ware JA, Wright SH, Yee SW, Zamek-Gliszczyinski MJ, Zhang L. 2010. Membrane transporters in drug development. Nat Rev Drug Discov 9(3):215–236. 29. Ahlin G, Hilgendorf C, Karlsson J, Szigyarto CA, Uhlen M, Artursson P. 2009. Endogenous gene and protein expression of drugtransporting proteins in cell lines routinely used in drug discovery programs. Drug Metab Dispos 37(12):2275–2283.

DOI 10.1002/jps.23901