The Fluorescent Probe Bodipy-FL-Verapamil Is a Substrate for Both P-glycoprotein and Multidrug Resistance-related Protein (MRP)-1

Abstract

Several fluorescent probes have been used in functional studies to analyze drug transport in multidrug-resistant cells by fluorescent microscopy. Because many of these molecules have some drawbacks, such as toxicity, nonspecific background, or accumulation in mitochondria, new fluorescent compounds have been proposed as more useful tools. Among these substances, Bodipy-FL-Verapamil, a fluorescent conjugate of the drug efflux blocker verapamil, has been used to study P-glycoprotein activity in different cell types. In this study we tested by fluorescent microscopy the accumulation of Bodipy-FL- Verapamil in cell lines that overexpress either P-glycoprotein (P-gp) or multidrug resistance-related protein 1 (MRP1). Expression of P-gp and MRP1 was evaluated at the mRNA level by RT-PCR technique and at the protein level by flow cytometric analysis using C219 and MRP-m6 monoclonal antibodies. Results indicate that Bodipy-FL-Verapamil is actually a substrate for both proteins. As a consequence, any conclusion about P-gp activity obtained by the use of Bodipy-FL-Verapamil as fluorescent tracer should be interpreted with caution.

Multidrug resistance (MDR) is one of the most important mechanisms of chemioresistance and a major problem in cancer chemotherapy. Primary or secondary chemioresistance is often mediated by overexpression of P-glycoprotein (P-gp) and/or multidrug resistance-related protein (MRP), both of which belong to the ATP-binding cassette (ABC) superfamily of membrane transport proteins (Ambudkar et al. 1999; Borst et al. 2000).

The availability of fluorescent molecules that are substrates for either P-gp or MRP is, of course, of great relevance in the attempt to predict the efficacy of chemotherapeutic strategies in clinical and experimental oncology. A series of fluorescent compounds, such as anthracyclines, acridines, calcein-AM, and rodamine 123, has been classically used to test P-gp activity (Kessel et al. 1991; Kühnel et al. 1997). Because many of these substances have some drawbacks, including cytotoxicity, nonspecific background, or accumulation in mitochondria, chemical companies have proposed new more sensitive and selective compounds. One such fluorochrome is Bodipy-FL-Verapamil (BFLV). This is a conjugate of the the P-gp efflux blocker verapamil that possesses the Bodipy-FL fluorophore, which has spectral properties similar to those of fluorescein. BFLV is a substrate for the efflux pump but does not bind to P-gp. This fluorescent probe has been used in studies of P-gp kinetics and in microscopic investigations for assessment of P-gp activity (Lelong et al. 1991; Simmons et al. 1995; Tosi et al. 1995; Crivellato et al. 1999; Candussio et al. 2001). Although the manufacturer's information leaflet for BFLV indicates that this compound may also be a substrate for MRP, no systematic investigations have been reported in the literature to support such an assumption.

In this study we tested by fluorescent microscopy the accumulation of BFLV in cell lines that overexpress either P-gp or MRP1. Single-cell microscopic analysis was carried out in the presence or absence of specific P-gp or MRP1 inhibitors. Expression of P-gp and MRP was evaluated at the mRNA level by the reverse transcriptase-polymerase chain reaction (RT-PCR) technique and at the protein level by flow cytometric analysis (FCA) using specific anti-P-gp and anti-MRP1 monoclonal antibodies.

Three cell lines were used in our study. The doxorubicin-resistant counterpart of the murine lymphoid neoplasm P388 (P388/ADR) was obtained from the NCI Frederick Cancer Research Facility/DC Tumor Repositor (Bethesda, MD) and was grown in suspension in RPMI 1640 medium supplemented with 10% fetal calf serum. The doxorubicin-resistant counterpart (LLC-PK1/ADR) of the pig kidney epithelial cell line LLC-PK1 (American Type Culture Collection, Rockville, MD; ATCC CRL-1392), was obtained in our laboratory by selection with doxorubicin as described previously (Decorti et al. 2001). LLC-PK1/ADR cells were grown in plastic bottles in medium 199 supplemented with 10% fetal bovine serum. PANC-1 cells, derived from human pancreatic epithelioid carcinoma, obtained from the American Type Culture Collection (ATCC CRL-1572), were grown in plastic bottles in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum. All cell lines were grown in 5% CO₂-95% air atmosphere at 37C. Cell culture media and supplements were obtained from Sigma-Aldrich (Milano, Italy).

The expression of P-gp and MRP1 at the mRNA level was evaluated by RT-PCR. Total RNA was extracted by SV total RNA Isolation Kit (Promega; Milano, Italy). Five μg of total RNA were reverse transcribed using Superscript II RNAse H (Gibco BRL Life Technologies; San Giuliano Milanese, Italy) and the cDNA (0.5–1 μl) was used in the PCR reaction in a final volume of 10 μl. Synthetic oligonucleotide primers (Invitrogen Life Technologies; San Giuliano Milanese) were used to investigate the presence of MDR1, mdr1a, mdr1b, and MRP1 mRNA transcripts. The primer sequences are listed in the legend to Figure 1. The amplification by Taq polymerase was performed using a DNA thermal cycler PTC 100 MJ Research. After denaturing at 94C for 5 min, the reaction was conducted for 20–35 cycles at different annealing temperatures, as indicated in the legend to Figure 1. For all primers, the reaction ended with a final step at 72C for 10 min. PCR products were subjected to electrophoresis on a 1.5% agarose gel and were visualized by ethidium bromide staining. To normalize the PCR data, primers amplifying a fragment of β-actin were used. The DNA marker was from Sigma-Aldrich (100-bp ladder).

Figure 1

RT-PCR analysis of MDR1, mdr1a, mdr1b, and MRP1 expression in P388/ADR, LLC-PK1/ADR, and PANC-1 cells. The size of the amplified product was determined from the mobility of the DNA size markers indicated in the corresponding lane. The primer sequences, the number of cycles, and the annealing temperatures are listed. Pig: MDR1 sense, 5′-AAGCGCTCATCAACTGTG-3′, MDR1 antisense 5′-GGCACTTTATGCAAACATTC-3′ (U27704 sequence); 392 bp, 50C, 35 cycles. Human: MDR1 sense 5′-GATGGACAGGATATT-AGGACC-3′ MDR1, antisense 5′-ATGGCACAAAATACACCAACA-3′ (M14758 sequence), 800 bp, 60C, 30 cycles. Human, pig, and rodents: MRP1 sense 5′-TTCCCATTTCAACGAGACCTTGCTG-3′, MRP1 antisense2, 5′-TCCGCCCCACGATGCCGACCT-3′ (L05628 sequence), 532 bp, 60C, 30 cycles. Rodents: mdr1a sense 5′-ACTCGGGAGCA-GAAGTTTGA-3′ mdr1a antisense 5′-GGAGCCACTGGACATTGAGT-3′ (M33581 sequence), 600 bp, 60C, 30 cycles. mdr1b sense 5′-GACG-GACAGGACATCAGGACC-3′ mdr1b antisense 5′-GCAAACACTGGT-TGTATGCAC-3′ (M81855 sequence), 823 bp, 60C, 30 cycles. Human, pig, and rodents: β-actin sense 5′-CACCAAAGCTGAGAGGGAA-ATCGTGCGTGA-3′, antisense 5′-ATTTGCGGTGCACGATGGAGGGGC-CGGACT-3′ (M10277 sequence), 500 bp, 60C, 25 cycles.

The expression of P-gp and MRP1 at the protein level was studied by FCA using the murine C219 (CIS Bio International; Gif-sur-Yvette, France) and MRP-m6 (Chemicon International; Temecula, CA) monoclonal antibodies (MAbs), which are specific for detection of P-gp and MRP1, respectively. A total of 5 × 10⁶/ml cells were fixed in 0.2% (v/v) paraformaldehyde solution in 100% acetone for 30 sec (for detection of C219 immunoreactivity) or with 100% ethanol at —20C for 15 min (for detection of MRP-m6 immunoreactivity). After the fixation step, the cells were centrifuged at 400 × g for 10 min at 4C, washed with 0.01 M PBS containing 3% bovine serum albumin (BSA), and re-suspended in the same buffer. The primary antibodies C219 (5–10 μg/ml) and MRPm6 (1.25–2.5 μg/ml) were added to aliquots of 200 μl cells (about 1 × 10⁶ cells). To determine the nonspecific staining, the respective isotypic controls IgG_2α (5–10 μg/ml) and IgG1 (1.25–2.5 μg/ml) were also used (Sigma-Aldrich). After 1-hr incubation at room temperature (RT), the cells were washed twice with PBS-3% BSA and finally resuspended in PBS-3% BSA containing the fluorescein isothiocyanate (FITC)-conjugated anti-mouse IgG secondary antibody (Sigma-Aldrich; 1:100). Incubation continued for 30 min at RT in the dark. Cells were then washed in PBS, resuspended in a final volume of 500 μl PBS, and subjected to FCA. FCA was performed by an EPICS ELITE EPS Coulter equipped with an argon/ion laser excitation source emitting at 488 nm. FITC fluorescence was collected through a 525-nm bandpass filter. At least 20,000 cells were analyzed and gated on the basis of the forward and sideward scatter. The green fluorescence amount was measured using log mode amplification. For the measure of intensity of the staining, the fluorescence index (FI) was calculated, which represents the ratio between the mean fluorescence intensity of cells stained with the specific antibody and that of cells stained with the isotypic control antibody in immunofluorescence experiments.

Intracellular accumulation of BFLV (Molecular Probes; Eugene, OR) was studied by fluorescence microscopy. For single-cell microscopic analysis, LLC-PK1/ADR and PANC-1 cells were seeded in Leighton tubes at a cell density of 2 × 10⁴ cells/ml in 1 ml of complete culture medium, and the uptake of BFLV was observed on confluent cells on the fifth day after inoculation. P388/ADR cells at a cell density of 1 × 10⁶ cells/ml were also used. Cells were incubated for specified periods in the absence or presence of substances known to modulate P-gp, such as verapamil and vinblastine (both from Sigma-Aldrich), or MRP activity, such MK571 (Biomol Research Lab; Plymouth Meeting, PA). Cells were examined under a Zeiss Axioskop microscope equipped with a mercury lamp (Osram HBO, 100 W; Oberkochen, Germany) for epifluorescence examination. BFLV was observed under appropriate excitation/emission filter setting conditions (excitation at 480 nm, emission at 510 nm) with fluorescein optics.

As expected, P388/ADR cells, which overexpressed P-gp at both the mRNA and the protein level but did not express MRP1 products (Figures 1, 2A, and 2B), exhibited a marked increase of intracellular BFLV-related fluorescence when treated with P-gp inhibitors, such as verapamil or vinblastine, but were unresponsive to MK571 (Figures 3A–3C). Incubation of P-gpoverexpressing LLC-PK1/ADR cells with inhibitors of P-gp activity induced a bright punctate cytoplasmic fluorescence (Figures 3D and 3E). Under the same experimental conditions, PANC-1 cells exhibited a weak fluorescence intensity (Figure 3H), according to the lower mRNA and protein expression of P-gp found in this cell line (Figures 1 and 2B). Inhibition of MRP1 activity in LLC-PK1/ADR and in PANC-1 cells by the specific MRP blocker MK571 led to accumulation of BFLV in both cell lines (Figure 3F and 3I). However, the intensity of intracellular fluorescence was higher in PANC-1 cells than in LLC-PK1/ADR cells, in line with the overexpression of MRP1 found in the former cell type by both RT-PCR (Figure 1) and FCA (Figure 2B).

Figure 2

Immunofluorescence detection of P-gp and MRP1 with monoclonal antibodies C219 (A) and MRPm6 (B), in P388/ADR, LLC/PK1/ADR, and PANC-1. Each bar represents the FI calculated as the ratio between the fluorescence intensity obtained by the specific antibody staining and that obtained with the isotypic control antibody (mean ± SEM of at least three separate experiments).

In conclusion, this study demonstrates that the fluorescent compound BFLV should not be considered a specific substrate for P-gp. Experimental data indeed provide evidence that this molecule is also transported by MRP1. BFLV is undoubtedly a useful tool for screening and analytical studies of the MDR pheno-type in many tumor cell types. In fact, it is nontoxic at the currently used concentrations and exhibits a strong and stable intrinsic fluorescence without fluorescent background. However, because BFLV is also a substrate for P-gp and MRP1, any result on P-gp activity obtained with this compound should be interpreted with caution.

Figure 3

Fluorescence microscopy of P388/ADR (A-C), LLC-PK1/ADR (D-F), and PANC-1 cells (G-I) treated with 0.125 mM BFLV for 30 min in the absence or presence of specific inhibitors of P-gp or MRP1. (A, D,G) Cells were incubated with BFLV alone for 30 min; there is an overall faint fluorescent signal. (B, E,H) Exposure to BFLV was preceded by incubation with 50 mM verapamil for 30 min; fluorescence in P388/ADR cells and LLC-PK1/ADR cells is remarkably increased, whereas the fluorescent signal in PANC-1 cells remains low. (C, F,I) Cells were incubated with 50 mM MK571 for 60 min before BFLV treatment. PANC-1 cells show a strong cytoplasmic fluorescence. On the other hand, LLC-PK1 cells exhibit a low increase in fluorescence, whereas fluorescence intensity in P388/ADR cells remains unchanged. Bars: A-C = 30 μm; D-I = 20 μm.