The inhibition of ABCB1/MDR1 or ABCG2/BCRP enables doxorubicin to eliminate liver cancer stem cells

Two ATP-binding cassette transporters, ABCB1/MDR1 and ABCG2/BCRP, are considered the most critical determinants for chemoresistance in hepatocellular carcinoma. However, their roles in the chemoresistance in liver cancer stem cells remain elusive. Here we explored the role of inhibition of MDR1 or ABCG2 in sensitizing liver cancer stem cells to doxorubicin, the most frequently used chemotherapeutic agent in treating liver cancer. We show that the inhibition of MDR1 or ABCG2 in Huh7 and PLC/PRF/5 cells using either pharmacological inhibitors or RNAi resulted in the elevated level of intracellular concentration of doxorubicin and the accompanied increased apoptosis as determined by confocal microscopy, high-performance liquid chromatography, flow cytometry, and annexin V assay. Notably, the inhibition of MDR1 or ABCG2 led to the reversal of the chemoresistance, as evident from the enhanced death of the chemoresistant liver cancer stem cells in tumorsphere-forming assays. Thus, the elevation of effective intracellular concentration of doxorubicin via the inhibition of MDR1 or ABCG2 represents a promising future strategy that transforms doxorubicin from a traditional chemotherapy agent into a robust killer of liver cancer stem cells for patients undergoing transarterial chemoembolization.


ABCG2 function assay using Hoechst 33342 extrusion
The Hoechst 33342 accumulation assay is used to investigate the activity of the ABCG2 drug efflux pump [2]. The ABCG2 inhibitor ko143 was dissolved in DMSO. The highest final concentration of DMSO in the cell culture medium for the assays was not more than 0.1%. The Huh7, PLC/PRF/5 and A549 cells were plated into the black 96-well plates respectively at a density of 3 × 10 4 cells per well and incubated under 5% CO2 at 37 °C for 24 h. The cell culture media were refreshed the next day with 200 µL phenol-red-free media. Cells were then incubated with various concentrations of ko143 as indicated for 30 min before the addition of 1 µM Hoechst 33342. Fluorescence was measured immediately at constant intervals (60 seconds) for 120 minutes with an excitation of 355 nm and an emission wavelength of 460 nm at 37 °C using a plate reader. The % relative fluorescence (FL) in the cells was expressed as: %FL = [(FLtreatment -Fnontreatment)/FLnontreatment] × 100%.

siRNA transfection
The dicer-substrate double-strand short interfering RNAs (siRNAs) were derived from previously published 21-mer, which were confirmed to have a high knockdown efficiency [3,4], and were extended to 27-mer dicer-substrate siRNA for this study to enhance the silencing efficiency [5]. The scrambled siRNAs have the same nucleotide composition as the MDR1 or ABCG2 siRNA but lack significant sequence homology to the human genome. The scrambled siRNAs were designed using the InvivoGen online tool (http://www.invivogen.com/sirnawizard/scrambled.php) and verified using the BLAST software tool (https://blast.ncbi.nlm.nih.gov/Blast.cgi). The siRNAs were synthesized by Integrated DNA Technologies followed by HPLC purification. Supplementary Table ST2 summarizes the sequences of siRNAs and primers used in this study. 4 Twenty-four hours prior to siRNA transfection, cells were plated in a 6-well cell culture plate at a density of 3 × 10 5 cells/well in 1.5 mL/well DMEM medium without antibiotics. The MDR1 siRNA, ABCG2 siRNA, and their corresponding scrambled siRNAs at a concentration of 20 nM were gently diluted in 250 µL of Opti-MEM reduced serum medium, respectively (ThermoFisher). In parallel, 10 µL of Lipofectamine 2000 was diluted in 250 µL Opti-MEM medium. Following incubation of 5 min at room temperature, the diluted siRNA and Lipofectamine 2000 were mixed gently and incubated for 10 minutes at room temperature, followed by the addition of 500 µL of siRNA-Lipofectamine complexes to each well (containing cells and 1.5 mL full DMEM medium in a well).
The plate was mixed gently by rocking back and forth. The cells were incubated in a humidified atmosphere containing 5% CO2 at 37 °C for 48 h followed by the assessment of expression levels of MDR1, ABCG2 using quantitative RT-PCR and Western analysis.

Cellular uptake of doxorubicin (DOX)
The Huh7 and PLC/PRF/5 cells were seeded at 8 × 10 3 cells per well in an 8-chamber slide (Lab-Tek ® II, Nunc, US) for 24 h in preparation for confocal microscopy. The cells were treated with MDR1 inhibitor verapamil (10 µM) or valspodar (1 µM), or ABCG2 inhibitor ko143 (1 µM) at 37 °C for 1 h followed by incubation with DOX (400 nM) for 40 min and visualized using confocal microscopy. Semi-quantification of cellular accumulation of DOX was performed using ImageJ software (National Institutes of Health, US).
To quantify DOX accumulation, the cells were washed with PBS after trypsinization and lysed with 200 µL of cell lysate buffer (50 mM Tris, pH 7.5, 375 mM NaCl, 1 mM EDTA, pH 8.0, 8% glycerol, and protein inhibitor cocktail). Four hundred microliter of acetonitrile was added to each sample to precipitate protein. The samples were vortexed for 30 seconds and centrifuged at 20,000 × g for 10 min. The supernatant was injected into the HPLC system (Waters e2695, Milford, US) and 5 the fluorescence of DOX was determined using a fluorescence detector (Waters 2475, Milford, US).
The amount of DOX was derived with the aid of a standard curve.

Tumorsphere formation assay
The tumorsphere assay was conducted according to the previously reported protocol [6]. The cells were seeded in 6-well cell culture plates and treated with 1 µM of MDR1 inhibitor valspodar, 10 µM of MDR1 inhibitor verapamil, or 1 µM of ABCG2 inhibitor ko143, and DOX (200 nM for Huh7 and 100 nM for PLC/PRF/5 cells respectively) for 48 h. Alternatively, the cells were treated with 20 nM siRNA to MDR1 or ABCG2 for 6 h followed by a further 42 h culture in the fresh culture medium.
Cells were then plated into round-bottom 96-well ultra-low attachment plates (Corning, #CLS3474) at a density of 10, 20, and 50 cells per well respectively for tumorsphere formation at 37 °C. For cells that were previously treated with inhibitor plus DOX, the same reagents were added in the tumorsphere culture media for the rest of the whole culture period. For cells that were transfected with siRNA and then incubated with DOX, the same DOX concentration was used in the tumorsphere culture media for the rest of the whole culture period. The cells were incubated in tumorsphere culture media for 7 days. The formation of tumorspheres was visualized by light microscopy. The frequency of tumorspheres was calculated using the online tool http://bioinf.wehi.edu.au/software/elda/index.html.