Figure 1.
Characterization of breast cancer stem cell-like (CSC-like) cells.
(A) Parental MDA-MB-453 cells were transfected with a plasmid encoding GFP under the control of the CMV promoter (non-stem) or Oct3/4 promoter (stem-like). After selection in G-418, GFP+ colonies were pooled. Phase-contrast images and fluorescence images of parental cells (Parent: a, d), CMV-GFP-transfected non-stem cells (Non-stem: b, e) or Oct3/4-GFP-transfected CSC-like cells (Stem-like: c, f) were visualized by light (Phase-contrast: a–c) or UV (Fluorescence: d–f) microscopy. (B) Flow cytometry characterization of parental, non-stem, or CSC-like cells was performed. CMV promoter-driven GFP cDNA (e; non-stem cells) or human Oct3/4 promoter-driven GFP cDNA (f; stem-like cells) transfected MDA-MB-453 cells were stained with surface marker antibodies (CD24, CD44) and evaluated by flow cytometry. (a–c) Unstained cells and (a, d) parental cells. (C) Stem cell-associated Oct-4 gene expression was examined in MDA-MB-453 and MDA-MB-231 parental (P), non-stem (N) and CSC-cell like (S) cells. Cells were harvested with lysis buffer. Lysates containing equal amounts of protein (20 µg/ml) were separated by SDS-PAGE, and immunoblotted with anti-Oct-4 antibody. Actin was shown as an internal standard. (D) Mammosphere formation was compared in MDA-MB-231 and MDA-MB-453 CSC-like and non-stem cells. For mammosphere formation, 1,000 cells from stem-like cells or non-stem cells were plated into ultra-low attachment plates. Phase-contrast images of mammospheres of non-stem (left panels) or CSC-like (right panels) cells were obtained 4 days or 9 days later. (E) Xenograft tumor formation was established with CSC-like and non-stem MDA-MB-231 cells. For tumor formation in NOD/SCID mice, 1×104 stem-like or non-stem cells were injected into the upper mammary fat pad. Tumor volumes were measured 30 days after injection.
Figure 2.
Survival curves for non-stem and CSC-like MDA-MB-453 and MDA-MB-231 cells after irradiation.
(A) Colonies were obtained with non-stem and stem-like MDA-MB-453 cells. Cells were unirradiated (0 Gy) or irradiated (6.25 Gy), trypsinized, counted and plated. Cells were grown for 1–3 weeks and stained with crystal violet. (B, C) Survival curves for non-stem and stem-like MDA-MB-453 (B) and MDA-MB-231 (C) cells were determined after irradiation. Cells were exposed to various doses (1.25 Gy-8.75 Gy) of γ-radiation. Cells were trypsinized, counted and plated. Colony formation was determined 1–3 weeks after irradiation. Error bars represent standard error from the mean for three separate experiments.
Figure 3.
Flow cytometry analysis of cell cycle in asynchronous and effect of aphidicolin treatment on phase distribution and radiosensitivity in non-stem and CSC-like cells.
Flow cytometry was performed on non-stem and stem-like MDA-MB-453 cells (A) or MDA-MB-231 cells (B). The percentages of cells in the G2/M, S, and G1 were analyzed and plotted. Non-stem and CSC-like MDA-MB-453 cells were treated with aphidicolin (5 µM) for 16 hr. After aphidicolin treatment, cell cycle was analyzed (C) and radiation sensitivity was determined by colony formation after irradiation (6.25 Gy) (D). Error bars represent standard error from the mean for three separate experiments.
Figure 4.
Role of anti-oxidant agents in radiosensitivity of non-stem cells and CSC-like cells.
(A) Non-stem (N) and stem-like (S) MDA-MB-453 and MDA-MB-231 cells were harvested. Lysates containing equal amounts of protein (20 µg/ml) were separated by SDS-PAGE, and immunoblotted with anti-MnSOD, anti-CuZnSOD, or anti-catalase antibody. Actin was shown as an internal standard. (B) Densitometry analysis of each band was performed. The area integration of optical density of each band in stem-like cells (S) was compared with that in non-stem cells (NS). Error bars represent standard error from the mean for three separate experiments. (C) MDA-MB-453 non-stem cells and stem-like cells were treated with 200 µM L-buthionine-sulfoximine (BSO) for 24 hr and GSH content was determined. Error bars represent standard error from the mean for three separate experiments. (D) BSO-treated/untreated non-stem and stem-like MDA-MB-453 and MDA-MB-231 cells were irradiated at 6.25 Gy and survival was determined. Error bars represent standard error from the mean for three separate experiments.
Figure 5.
Analysis of sublethal damage repair.
(A) Non-stem and stem-like MDA-MB-453 cells were exposed to two fractions of γ-radiation (5.0+2.5 Gy for non-stem and 3.75+2.5 Gy for stem-like) and incubated at 24°C for various time intervals between two exposures. Survival was compared to control group (irradiated without post-incubation) and plotted. Error bars represent standard error from the mean for three separate experiments. (B) Stem-like MDA-MB-453 and MDA-MB-231 were exposed to two fractions of γ-radiation (3.75+2.5 Gy) and incubated at 24°C for various time intervals between two exposures. Survival was compared to control group (irradiated without post-incubation) and plotted. Error bars represent standard error from the mean for three separate experiments.
Figure 6.
Ionizing radiation-induced phosphorylation of ATM and effect of ATM inhibitor CP466722 on radiosensitivity.
(A, B) Non-stem and CSC-like MDA-MB-435 and MDA-MB-231 cells were irradiated at 8.75 Gy and phosphorylation (activation) of ATM was determined various times (0.5–12 hr) after irradiation. Lysates containing equal amounts of protein (20 µg) were separated by SDS-PAGE and immunoblotted with anti-ATM or anti-phospho-ATM antibody. Actin was shown as an internal standard. (C) MDA-MB-453 non-stem cells were pretreated with/without 100 µM CP466722 for 30 min, irradiated at 6.25 Gy and incubated various times before immunoblot analysis as described above. (D) MDA-MB-453 non-stem cells were pretreated with/without 100 µM CP466722 for 30 min, irradiated at various doses (1.25 Gy–8.75 Gy) and incubated for 6 hr before colony formation analysis. Error bars represent standard error from the mean for three separate experiments.
Figure 7.
Determination of ATM gene expression and ATM protein stability in non-stem and CSC-like MDA-MB-435 and MDA-MB-231 cells.
(A, B) Total RNA was extracted and reverse transcribed into cDNA. Human ATM mRNA was amplified and analyzed electrophoretically, and then quantified by Un-Scan-It gel software. (C, D) Cells were treated with 30 µg cycloheximide (CHM: >95% protein synthesis inhibition) for various times (2–16 hr) and harvested. Lysates containing equal amounts of protein (20 µg/ml) were separated by SDS-PAGE, and immunoblotted with anti-ATM or anti-actin antibody. Actin was shown as an internal standard. Densitometry analysis of each band was performed as described in Figure 4.
Figure 8.
Role of ATM in radiosensitivity in non-stem MDA-MB-453 (A) and MDA-MB-231 (B) cells.
Non-stem cells were infected with control shRNA or ATM shRNA lentiviral particle (2.5×104–105 IFU) and stable clones were selected by treatment with 10–100 µg/ml puromycin. ATM knockdown level was assessed by immunoblot assay as described in Fig. 6 (upper panels) and survival was determined after irradiation at 6.25 Gy (lower panels). Error bars represent standard error from the mean for three separate experiments.
Table 1.
Percentage of cells stained γ-H2AX positive after irradiation at 2.5 Gy.
Table 2.
Percentage of cells stained γ-H2AX positive after irradiation at 8.75 Gy.
Figure 9.
Ionizing radiation-induced γ-H2AX foci formation.
Non-stem and CSC-like MDA-MB-453 cells were irradiated at 2.5 Gy. After 0.5 hr or 12 hr incubation, phosphorylated H2AX was detected by immunofluorescent staining with anti-phospho-H2AX antibody. Nuclei were stained with DAPI.
Figure 10.
Alkaline comet images and their quantitative analysis for non-stem and stem-like MDA-MB-231 cells after irradiation.
(A) Control (0 Gy) or irradiated (6.25 Gy) cells for both non-stem cells (upper row) and stem-like cells (bottom row) were subjected to alkaline comet assay (refer to the Methods) and the DNA was visualized by using a propidium iodide. (B) Distribution of comet tail moments for different treatments was plotted. Tail moments are the products of the distance of DNA migration (microns) and the amount of separated DNA (%).