Differences in the expression of endogenous efflux transporters in MDR1-transfected versus wildtype cell lines affect P-glycoprotein mediated drug transport

Abstract

Background and purpose: P-glycoprotein (Pgp) efflux assays are widely used to identify Pgp substrates. The kidney cell lines Madin-Darby canine kidney (MDCK)-II and LLC-PK1, transfected with human MDR1 (ABCB1) are used to provide recombinant models of drug transport. Endogenous transporters in these cells may contribute to the activities of recombinant transporters, so that drug transport in MDR1-transfected cells is often corrected for the transport obtained in parental (wildtype) cells. However, expression of endogenous transporters may vary between transfected and wildtype cells, so that this correction may cause erroneous data. Here, we have measured the expression of endogenous efflux transporters in transfected and wildtype MDCK-II or LLC cells and the consequences for Pgp-mediated drug transport.

Experimental approach: Using quantitative real-time RT-PCR, we determined the expression of endogenous Mdr1 mRNA and other efflux transporters in wildtype and MDR1-transfected MDCK-II and LLC cells. Transcellular transport was measured with the test substrate vinblastine.

Key results: In MDR1-transfected MDCK cells, expression of endogenous (canine) Mdr1 and Mrp2 (Abcc2) mRNA was markedly lower than in wildtype cells, whereas MDR1-transfected LLC cells exhibited comparable Mdr1 but strikingly higher Mrp2 mRNA levels than wildtype cells. As a consequence, transport of vinblastine by human Pgp in efflux experiments was markedly underestimated when transport in MDR1-transfected MDCK cells was corrected for transport obtained in wildtype cells. This problem did not occur in LLC cells.

Conclusions and implications: Differences in the expression of endogenous efflux transporters between transfected and wildtype MDCK cells provide a potential bias for in vitro studies on Pgp-mediated drug transport.

Figure 1

Expression of endogenous (canine) Mdr1 in parental (wildtype; WT) and MDR1-transfected MDCK-II cells. As shown in A, expression of canine Mdr1 is about seven times lower in MDR1-transfected versus wildtype cells. B demonstrates expression of human MDR1 mRNA in the transfected cells and the lack of any human MDR1 expression in wildtype cells. C illustrates the protein content of Pgp in MDCK wildtype and MDR1-transfected MDCK cells. Because it is not possible to differentiate canine and human Pgp by Western blots using the C219 antibody, total (canine and human) Pgp expression is shown. Data in (A), (B) and (C) are shown as mean ± SEM of four experiments. *Significant group differences; P < 0.001. Data in (D) illustrate a representative experiment in which qPCR was used to determine mRNAs of canine Mdr1, Mrp1, Mrp2 and Mrp5 in MDCK wildtype cells versus cells transfected with human MDR1. Note the lower expression of Mdr1 and Mrp2 mRNAs in the transfected cells. MDCK, Madin-Darby canine kidney; MDR1, multidrug resistance 1; Pgp, P-glycoprotein; qPCR, quantitative PCR.

Figure 2

Expression of endogenous (porcine) Mdr1 in parental (wildtype; WT) and MDR1-transfected LLC cells. As shown in (A), the expression of canine Mdr1 is comparable in MDR1-transfected versus parental (WT) LLC cells. (B) demonstrates expression of human MDR1 mRNA in the transfected cells and the lack of any human MDR1 expression in wildtype cells. (C) illustrates the protein content of Pgp in LLC wildtype and MDR1-transfected MDCK cells. Because it is not possible to differentiate porcine and human Pgp by Western blots using the C219 antibody, total (porcine and human) Pgp expression is shown. Data in (A), (B) and (C) are shown as mean ± SEM of five experiments. *Significant group differences; P < 0.001. Data in (D) illustrate an experiment in which qPCR was used to determine mRNAs of porcine Mrp1 and Mrp2 in LLC wildtype cells versus cells transfected with human MDR1. Data are shown as mean ± SEM of three experiments. Note the higher expression of Mrp2 mRNAs in the transfected cells. *P = 0.0203. MDR1, multidrug resistance 1; Pgp, P-glycoprotein; qPCR, quantitative PCR.

Figure 3

Comparison of Pgp expression in different cell lines. Data are from a representative experiment in which Pgp expression was determined by Western blot in MDCK-II wildtype and MDCK-II cells transfected with human MRP1, MRP2 or MRP5. For comparison, Pgp expression in LLC wildtype and MDR1-transfected LLC cells is shown. All data shown in (A) are from one gel, whereas data in B are from another gel with a lower amount of protein added, which led to disappearance of the Pgp band in LLC wildtype cells. For each cell line, two different samples were tested except for LLC cells, for which four different samples are shown in (A) and (B). Note that Pgp appeared as one band at around 160 kDa in Western blots from LLC cells (A), while two bands at about 155 and 140 kDa were labelled with the C219 Pgp antibody in MDCK cells. It has been suggested previously that the lower band Pgp isoform in MDCK cells is less glycosylated than the upper band (Tang et al., 2002a), which represents a mature form of Pgp. The upper band was therefore used for the calculations of Pgp expression in MDCK cells shown in C. In (C), Pgp expression was normalized on actin expression to allow comparison of Pgp expression between cell lines. Data are shown as means ± SEM. In contrast to MDCK wildtype cells, which expressed a relatively high amount of endogenous (canine) Pgp, very low Pgp expression was determined in LLC wildtype cells at the conditions of the assay. Transfection with human MDR1 led to a marked Pgp expression in LLC cells. However, a comparable Pgp expression was also observed in MRP5-transfected MDCK cells. This experiment was repeated several times on different gels (not illustrated), confirming that MRP5-transfected MDCK cells exhibit high expression of endogenous Pgp. MDCK, Madin-Darby canine kidney; MDR1, multidrug resistance 1; MRP, multidrug resistance protein; Pgp, P-glycoprotein.

Figure 4

Comparison of Mrp2 expression in different cell lines. Data are from a representative experiment in which Mrp2 expression was determined by Western blot in MDCK-II wildtype and MDCK-II cells transfected with human MRP1, MRP2 or MRP5. For each cell line, three different samples are shown. As shown in (A) (MDCK-II cells) endogenous (canine) Mrp2 and recombinant human MRP2 appeared as one band at around 190 kDa. For comparison, Mrp2 expression in LLC wildtype and MDR1-transfected LLC cells is shown in (B). Note the increased expression of endogenous (pig) Mrp2 in MDR1-transfected LLC cells. In (C), Mrp2 expression was normalized on actin expression to allow comparison of Mrp2 expression between cell lines. Data are shown as mean ± SEM of three samples per cell line except MDCK wildtype, which is mean ± SEM from nine samples. In contrast to MDCK wildtype cells, which expressed a relatively high amount of endogenous (canine) Mrp2, very low Mrp2 expression was determined in LLC wildtype cells at the conditions of the assay. Transfection with human MRP2 led to a marked MRP2 expression in MDCK-II cells. *P < 0.05; ***P < 0.001; significant differences between transfected cells and respective wildtype cells. MDCK, Madin-Darby canine kidney; MDR1, multidrug resistance 1; MRP, multidrug resistance protein.

Figure 5

Representative transcellular transport experiments with vinblastine (2 µM) in MDCK (A–D) and LLC (E–H) cells. Experiments were performed by the conventional bi-directional (concentration gradient) assay, in which the drug is added to either the apical (for apical to basolateral transport, a-B) or the basolateral (for basolateral to apical transport, b-A) side of the monolayer. Data are given as the percentage of the initial drug concentration in the donor chamber versus time. Vinblastine transport was either determined alone or in the presence of the Mrp inhibitor MK571 (50 µM) or the Pgp inhibitor tariquidar (0.2 µM) or a combination of both inhibitors. For each experiment, transport ratios (TR) are given (for calculation, see Methods). Corrected transport ratios (cTRs) were calculated by dividing the TR obtained in MDR1-transfected cells by the TR obtained in wildtype/parental cells. A cTR of ≥1.5 is considered as an indicator of active, asymmetrical transport (Schwab et al., 2003). All experiments were performed in triplicate and values are shown as mean ± SEM. When no error bar is visible, the deviation was within the size of the symbols. MDCK, Madin-Darby canine kidney; MDR1, multidrug resistance 1; Pgp, P-glycoprotein.