br Isolation of CSCs from
Isolation of CSCs from the general GBM cell line (GBM 8401) was
Fig. 1. Diagram of the NTHU cell irradiation system.
Fig. 2. Procedure for isolating CSCs from GBM 8401 cells using magnetic microbeads.
performed using the Miltenyi Biotec CD133 Microbead Kit. The CSC isolation procedure is illustrated in Fig. 2. First, the cell samples were resuspended in buffer-based blocking reagents to avoid nonspecific antibody labeling. Next, a 20-μL CD133 microbead solution was added to the cell samples and gently mixed with a pipette. During the labeling procedure, the cell–microbead mixture was placed inside a 4 °C re-frigerator for 15 min to enhance identification efficiency. When CD133 labeling was completed, the cell samples were loaded into an LS se-paration column and mounted on a separation magnet to separate bound and unbound cells. During separation, CD133+ cells were trapped in the LS column through magnetic force, whereas CD133- cells were washed out by the buffer solution. When separation was finished, the column was detached from the magnet, and the CD133+ cells were eluted using an eluting solution.
2.4. Immunofluorescence staining
In this study, γ-H2AX immunofluorescence was performed using mouse monoclonal primary Fluxametamide (A11001; Invitrogen, Waltham, USA) and goat anti-mouse IgG secondary antibodies conjugated with Alexa Fluor 488 dye (JBW301; Millipore, Burlington, USA). Before staining, cell samples were immersed for fixation in 2% paraformaldehyde for 15 min. After fixation, the samples were washed three times in phos-phate-buffered saline (PBS) and permeabilized using 0.2% Triton X-100 (T8787; Sigma-Aldrich, St. Louis, USA) for 5 min. To avoid nonspecific binding of antibodies, the cell samples were blocked by immersion in 1% bovine serum album (A7906; Sigma-Aldrich, St. Louis, USA) with PBS. After blocking, the samples were incubated overnight with 400× diluted antiphosphorylated histone H2A.X ser139 monoclonal antibody (JBW301; Millipore, Burlington, USA) at 4 °C. When primary antibody labeling was completed, the cell samples were again washed three times in the 1% BSA–PBS solution and stained using 200× diluted DyLight 488
conjugated goat anti-mouse IgG heavy and light chain secondary antibody (A90-116D2; Bethyl Laboratories, Montgomery, USA) at room tempera-ture for 1 h. Prior to image acquisition, the cell samples were mounted using DAPI-added mounting solution (ProLong Gold Antifade Mountant with DAPI; Life Technologies, Waltham, USA) to stain the cell nucleus and avoid photobleaching during image acquisition. For CD133 immunofluorescence staining, rabbit IgG CD133 (#3663; Cell Signaling, Danvers, USA) was chosen as the primary an-tibody and goat anti-rabbit IgG conjugated with Alexa Fluor 488 dye as the secondary antibody. The CD133 immunofluorescence staining protocol was similar to that of γ-H2AX staining, except that the per-meabilization step was omitted in CD133 staining to avoid dissolving the CD133 in the Triton X-100 nonionic detergent .
Immunofluorescence images were obtained using a Nikon TE2000-S inverted microscope and analyzed using a homemade ImageJ foci count macro modified from the PZ foci EZ macro proposed by Znojek . This foci count macro could automatically count the number and re-cognize the region of cell nuclei from a DAPI image. After recognition, the macro created a mask for the γ-H2AX image and automatically calculated the foci number and average intensity cell by cell; over 300 cells were counted in each sample to minimize statistical error.
To examine the radioresistance of CSCs and cell survival curves, a clinical linac (Unique; Varian) was used to provide uniform X-ray beams for irradiation of GBM 8401 and GBM 8401 CD133+ cells. The irradiation doses were 0, 2, 4, 6, and 8 Gy. Following irradiation, each culture dish was seeded with 500 cells and stored in an incubator for 1 week. The colonies were fixed in 75% alcohol, stained with crystal violet, and enumerated using a stereomicroscope.
3. Results and discussion
To verify the capability and efficiency of CSC isolation using the magnetic microbead method, CD133 immunofluorescence staining was performed on the cells before and after isolation. Fig. 3 shows the im-munofluorescence staining results of CD133 surface protein on GBM 8401 and GBM 8401 CD133+ cells. Staining of GBM 8401 CD133+ cells (Fig. 3B) exhibited homogeneous distribution of CD133 trans-membrane protein on the cell membranes, whereas this was not ob-served for GBM 8401 cells. This indicated that the magnetic microbead method was able to isolate CSCs from a general GBM cell line. We also demonstrated that isolated CSCs can grow into an independent cell line, but CD133+ characteristics can only be retained for several genera-tions. In long-term cultures, CD133+ characteristics disappear through differentiation, even in cells cultured with a stem cell cultivation sup-plement.