Mitochondrial localization of ABC transporter ABCG2 and its function in 5-aminolevulinic acid-mediated protoporphyrin IX accumulation

Abstract

Accumulation of protoporphyrin IX (PpIX) in malignant cells is the basis of 5-aminolevulinic acid (ALA)-mediated photodynamic therapy. We studied the expression of proteins that possibly affect ALA-mediated PpIX accumulation, namely oligopeptide transporter-1 and -2, ferrochelatase and ATP-binding cassette transporter G2 (ABCG2), in several tumor cell lines. Among these proteins, only ABCG2 correlated negatively with ALA-mediated PpIX accumulation. Both a subcellular fractionation study and confocal laser microscopic analysis revealed that ABCG2 was distributed not only in the plasma membrane but also intracellular organelles, including mitochondria. In addition, mitochondrial ABCG2 regulated the content of ALA-mediated PpIX in mitochondria, and Ko143, a specific inhibitor of ABCG2, enhanced mitochondrial PpIX accumulation. To clarify the possible roles of mitochondrial ABCG2, we characterized stably transfected-HEK (ST-HEK) cells overexpressing ABCG2. In these ST-HEK cells, functionally active ABCG2 was detected in mitochondria, and treatment with Ko143 increased ALA-mediated mitochondrial PpIX accumulation. Moreover, the mitochondria isolated from ST-HEK cells exported doxorubicin probably through ABCG2, because the export of doxorubicin was inhibited by Ko143. The susceptibility of ABCG2 distributed in mitochondria to proteinase K, endoglycosidase H and peptide-N-glycosidase F suggested that ABCG2 in mitochondrial fraction is modified by N-glycans and trafficked through the endoplasmic reticulum and Golgi apparatus and finally localizes within the mitochondria. Thus, it was found that ABCG2 distributed in mitochondria is a functional transporter and that the mitochondrial ABCG2 regulates ALA-mediated PpIX level through PpIX export from mitochondria to the cytosol.

Figure 1. Expression of ABCG2, PEPT and FECH, and ALA-mediated PpIX accumulation in various cancer cells.

Human lung adenocarcinoma cells (A549), human renal cancer cells (Caki-1),human colorectal cancer cells (DLD-1),human bladder cancer cells (T24) and human histiocytic lymphoma cells (U937) were used in this experiment. A, Western blotting analysis of whole cell lysates of various cancer cell lines. ABCG2 was detected with a polyclonal anti-ABCG2 (Cat. #4477; Cell Signaling Technology). Actin was used as a protein loading control. A representative blot of multiple experiments is shown. B, Relative contents of each protein were determined by densitometric scanning and were normalized using actin. C, Levels of accumulated ALA-mediated PpIX in various cancer cell lines. Bar graphs show flow cytometric analysis results of cellular PpIX content. Mean intensities of fluorescence (autofluorescence values) obtained from control experiments of each cell line were normalized to 100%. Results represent the mean ± SD of three independent experiments.

Figure 2. Confocal immunofluorescence microscopic analysis of ABCG2 localization.

Live cells cultured on glass coverslips were incubated with 25 nM of Mito Tracker Red CMXRos to label mitochondria (red); then they were fixed and processed for ABCG2 protein immunostaining (green). Confocal images were obtained from cells immunostained with mouse monoclonal anti-ABCG2 antibody (5D3; R&D Systems) and Alexa Fluor 488-conjugated goat anti-mouse IgG antibody. Insets show a magnified area, indicated by white arrows.

Figure 3. Distribution of ABCG2 in the mitochondrial fraction.

A, Subcellular fractions were isolated by a centrifugation method as described in Materials and Methods. ABCG2 was detected by western blotting analysis using polyclonal anti-ABCG2 (Cat. #4477; Cell Signaling Technology). CoxIV, Actin, SEC23A, GM130, and Na, K-ATPase were used as organelle-specific markers of mitochondria, cytosol, endoplasmic reticulum, Golgi apparatus, and plasma membrane, respectively. A representative blot of multiple experiments is shown. B, Highly purified mitochondria were isolated by OptiPrep density gradient centrifugation as described in Materials and Methods and the western blotting analysis was performed using antibodies as described in A. Equal volume of samples from both whole cell lysate and mitochondrial fraction of A549 cells were subjected to SDS-PAGE. A representative blot of multiple experiments is shown.

Figure 4. Regulation of the intra-mitochondrial content of ALA-mediated PpIX by Ko143, a specific inhibitor of ABCG2.

U937 cells and A549 cells were incubated for 3 h with 1 mM ALA in complete medium in the presence or absence of 1 µM Ko143. After incubation, cells were washed with ALA-free medium and mitochondrial and cytosolic fractions were isolated as described in Materials and Methods. The fluorescence spectrum of each fraction was analyzed by a fluorophotometer. A representative histogram of three independent experiments is shown. The levels of PpIX were shown by black line (No treatment), red line (ALA without Ko143), green line (ALA with Ko143), respectively.

Figure 5. ABCG2-dependent regulation of ALA-mediated PpIX accumulation.

A, Western blot analysis of whole cell lysates of HEK cells and ST-HEK cells. A representative blot of multiple experiments is shown. Na, K-ATPase and actin were used as a plasma membrane marker and loading control, respectively. B, Flow cytometric histograms of intracellular ALA-mediated PpIX contents in HEK cells and ST-HEK cells. The porphyrin autofluorescence (black line) in HEK cells and ST-HEK cells was measured. Cells were incubated with 1 mM ALA in complete medium for 3 h in the presence (green line) or absence (red line) of Ko134, and then cellular contents of PpIX were measured using a flow cytometer. C, Bar graphs show the flow cytometric analysis of the cellular PpIX contents. Mean intensities of fluorescence (autofluorescence values) obtained from control experiments of each cell line were normalized to 100%. Results represent the mean ± SD of three independent experiments. **P<0.01 vs. auto. †P<0.01 vs. ALA-treated.

Figure 6. Expression and distribution of FLAG-ABCG2 in HEK cells and ST-HEK cells.

A, Confocal immunofluorescence analysis of HEK cells and ST-HEK cells. Mitochondria, stained with Mito Tracker Red CMXRos; ABCG2, stained with monoclonal anti-FLAG IgG and Alexa Fluor 488-conjugated goat anti-mouse IgG. B, Western blotting analysis of mitochondrial fraction of HEK cells and ST-HEK cells. A representative blot of multiple experiments is shown.

Figure 7. Mitochondrial ABCG2 regulates the level of ALA-mediated PpIX accumulation in mitochondria of ST-HEK cells.

HEK and ST-HEK cells were incubated for 3 h with ALA in the presence or absence of 1 µM Ko143. After incubation, cells were washed with ALA-free medium and subcellular fractions were isolated as described in Materials and Methods. The fluorescence spectrum of each fraction was analyzed by a fluorophotometer. A representative histogram of three independent experiments is shown. The levels of PpIX were shown by black line (No treatment), red line (ALA without Ko143), green line (ALA with Ko143), respectively.

Figure 8. Accumulation and efflux of doxorubicin in mitochondria isolated from HEK cells and ST-HEK cells.

Mitochondrial fractions were isolated as described in Materials and Methods. Mitochondrial preparations of HEK cells (A) and ST-HEK cells (B) were incubated with 5 µM doxorubicin for 10 min, then mitochondria were washed with doxorubicin-free buffer A with or without 1 µM Ko143 and incubated for 1 h. Left panel, contents of doxorubicin in mitochondria were analyzed using a flow cytometer. A representative histogram of three independent experiments is shown. Right panel, bar graphs show the flow cytometric analysis of the mitochondrial doxorubicin contents. Results represent the mean ± SD of three independent experiments. Black bar indicates autofluorescence (Auto); red bar, doxorubicin treatment (DXR); yellow bar, wash without Ko143 (Wash); green bar, wash with Ko143 (Wash+Ko143). **P<0.01 vs. DXR. †P<0.01 vs. Wash.

Figure 9. Identification of the ABCG2 distribution in mitochondrial sub-compartment.

Mitochondrial preparations of A549 cells (A) and ST-HEK cells (B) were treated with proteinase K on ice for 60 min in the presence or absence of 0.2% Triton X-100 (TX-100). Then the proteins were analyzed by Western blotting using a polyclonal anti-ABCG2 antibody. A representative blot of three independent experiments is shown.

Figure 10. Characterization of protein N-glycosylation occurring on mitochondrial ABCG2 of A549 cells and ST-HEK cells.

Mitochondrial preparations of A549 cells (A) and ST-HEK cells (B) were treated with PNGase F or Endo H, and then ABCG2 was detected by Western blotting using a polyclonal anti-ABCG2 antibody. A representative blot of three independent experiments is shown.

Figure 11. Effect of Ko143 on the cytotoxic damage of ALA-treated cancer cells by photodynamic treatment.

Cells were exposed to light (29 mW/cm²) for 10 min using a Na-Li lamp after 3 h treatment with 1 mM ALA in the presence or absence of 1 µM Ko143. Twenty-four h after irradiation, the MTT assay was performed. Results represent the mean ± SD of three independent experiments. **P<0.01 vs. control. †P<0.01 vs. ALA-treated.

Figure 12. Involvement of mitochondrial ABCG2 in the regulatory mechanism of ALA-mediated PpIX accumulation in malignant cells.

PEPT1/2, oligopeptide transporters; ABCB6, ABC transporter B6.