Dynamic vs static ABCG2 inhibitors to sensitize drug resistant cancer cells

Abstract

Human ABCG2, a member of the ATP-binding cassette transporter superfamily, plays a key role in multidrug resistance and protecting cancer stem cells. ABCG2-knockout had no apparent adverse effect on the development, biochemistry, and life of mice. Thus, ABCG2 is an interesting and promising target for development of chemo-sensitizing agents for better treatment of drug resistant cancers and for eliminating cancer stem cells. Previously, we reported a novel two mode-acting ABCG2 inhibitor, PZ-39, that induces ABCG2 degradation in addition to inhibiting its activity. In this manuscript, we report our recent progresses in identifying two different groups of ABCG2 inhibitors with one inhibiting only ABCG2 function (static) and the other induces ABCG2 degradation in lysosome in addition to inhibiting its function (dynamic). Thus, the inhibitor-induced ABCG2 degradation may be more common than we previously anticipated and further investigation of the dynamic inhibitors that induce ABCG2 degradation may provide a more effective way of sensitizing ABCG2-mediated MDR in cancer chemotherapy.

Figure 1. Structures of PZ-8, 16, 34 and 38 in comparison with PZ-39.

The chemical structures are shown for PZ-8, (12E)-N'-((5-(3,4-dihydro-4-oxo-3-phenylquinazolin-2-ylthio)furan-2-yl)methylene)-2-(4-ethylphenoxy)acetohydrazide; PZ-16, 2-(4-(4-nitrophenoxy)phenyl)-2-oxoethyl2-(2-(4- chloro benzamido)acetamido)acetate; PZ-34, (E)-2-(4-ethoxyphenyl)-N'-(1-(4-(furan-2-carboxamido) phenyl)ethylidene)quinoline-4-carbohydrazide; PZ-38, (N-(2,5-dimethoxyphenyl)-2-({4-[4-(dimethylamino)benzylidene]-5-oxo-1-phenyl-4,5-dihydro-1H-imidazol-2-yl}sulfanyl)acetamide); and PZ-39 (N-(4-chlorophenyl)-2-[(6-{[4,6-di(4- morpholinyl)-1,3,5- triazin-2-yl] amino}-1,3-benzothiazol-2-yl)sulfanyl]acetamide).

Figure 2. Effect of PZ compounds on mitoxantrone accumulation and ABCG2 expression.

A, mitoxantrone accumulation. HEK293/ABCG2 cells were incubated with mitoxantrone for 30 min in the presence of DMSO (thin line) or 10 µM PZ compounds (thick line) followed by FACS analysis of mitoxantrone level. B, ABCG2 expression. HEK293/ABCG2 cells were incubated with 3.3 µM PZ compounds or DMSO control for various times followed by collection of cells and Western blot analysis of ABCG2 probed with monoclonal antibody BXP-21.

Figure 3. Effect of NSC compounds on ABCG2 expression and mitoxantrone efflux.

A, ABCG2 expression. HEK293/ABCG2 cells were treated with 10 µM each of the known ABCG2 inhibitors NSC-168201, NSC-120668, or FTC for various times followed by Western blot analysis of ABCG2 expression probed with monoclonal antibody BXP-21. B, mitoxantrone accumulation. HEK293/ABCG2 cells were incubated with mitoxantrone for 30 min in the presence of DMSO (thin line) or 10 µM NSC-168201, NSC-120668, or FTC (thick line) followed by FACS analysis of intracellular mitoxantrone level.

Figure 4. PZ-34 and PZ-38 inhibition of ABCG2-mediated mitoxantrone efflux.

A, mitoxantrone accumulation. HEK293/Vec or HEK293/ABCG2 cells were incubated with mitoxantrone for 30 minute in the presence of DMSO control, or 3 µM of PZ-34, PZ-38 or FTC control. The data are means ± SD from three independent experiments. B, dose response of PZ-34, PZ-38, and FTC control in restoring mitoxantrone accumulation in HEK293/ABCG2 cells. The thick line shows the level of mitoxantrone accumulation in HEK293/Vec cells, serving as a maximum accumulation control. C, effect on ABCB1 and ABCC1-mediated Adriamycin efflux. HEK293 cells with ABCC1 over-expression (HEK293/ABCC1) and BC19 cells with ABCB1 over-expression were incubated with Adriamycin in the absence (gray area) or presence (dotted line) of 3.8 µM PZ-34 or PZ-38 followed by FACS analysis. The thick line indicates the maximum level of accumulation in cells transfected with vector control. D, effect on ABCB1 and ABCC1 expression. HEK293/ABCC1 and BC19 cells with ABCB1 over-expression were treated with 10 µM PZ-34 or PZ-38 for 3 days followed by Western blot analysis of ABCB1 using monoclonal antibody C219 and ABCC1 using monoclonal antibody MRPr1. GAPDH was used as a loading control.

Figure 5. Effect of PZ-34 and PZ-38 on sensitizing drug resistance.

A, potency of PZ-34 and PZ-38 in reversing mitoxantrone resistance. HEK293/ABCG2 cells were treated without or with 0.1 µM (IC₁₀) mitoxantrone in the absence or presence of different concentrations of PZ-34 or PZ-38 followed by SRB assay. B, sensitization index of PZ-34 and PZ-38 in HEK293/ABCG2 cells. HEK293/ABCG2 cells were treated with various concentrations of mitoxantrone in the absence or presence of different concentrations of PZ-34 or PZ-38 followed by SRB assay. C and D, sensitization index of PZ-34 and PZ-38 in drug-selected MCF7/AdVp3000 cells. MCF7/AdVp3000 cells were treated with various concentrations of mitoxantrone (C), or Adriamycin (D) in the presence of DMSO (vehicle) or 500 nM of PZ-34 and PZ-38 followed by MTT assay. Sensitization index was calculated using IC₅₀ of the anticancer drugs in the absence or presence of PZ-34, PZ-38, or FTC. The data shown are mean ± SD of three independent experiments.

Figure 6. Effect of PZ-34 and PZ-38 on ABCG2 stability.

A, Western blot analysis. HEK293/ABCG2 cells were first treated with 5 µg/ml cycloheximide (CHX) followed by addition of 3 µM of PZ-34, PZ-38, FTC, or DMSO control for various times. The cells were then harvested for Western blot analysis of ABCG2 probed with monoclonal antibody BXP-21. Actin was used as a loading control. B, half-life of ABCG2. ABCG2 levels on Western blot as shown in panel A were determined using Scion Image and plotted against time of treatment. Data shown are mean± S.D of three experiments.

Figure 7. PZ-34 and PZ-38-binding induced ABCG2 degradation and conformational change.

A, ABCG2 degradation in lysosomes. HEK293/ABCG2 cells were treated with 3 µM of PZ-34 or PZ-38 in the absence or presence of 10 nM Bafilomycin A₁ (BMA1) or 2 µM MG-132 for various times. The cells were then harvested for Western blot analysis of ABCG2 expression probed with monoclonal antibody BXP-21. Actin was used as a loading control. B, ABCG2 conformational changes. HEK293/ABCG2 and MCF7/AdVp3000 cells were treated with 10 µM of PZ-34, PZ-38, or DMSO vehicle control followed by staining with monoclonal antibody 5D3 and FACS analysis. Arrowhead indicate the PZ-34 and PZ-38 treated groups.