Immunohistochemistry (IHC) staining
For histopathological analysis formalin fixed and paraffin embedded (FFPE) samples were assessed and graded using standard hematoxylin and eosin (H&E) sections (3–4 µm). The astrocytic origin was confirmed by glial fibrillary acidic protein (GFAP) staining. Immunohistochemistry for detection of the IDH1 R132H mutation was performed using the monoclonal antibody anti-IDH1 R132H (clone H09, 1:50, Dianova, Hamburg, Germany). For detection of Alpha-thalassemia/mental retardation syndrome X-linked (ATRX) mutations, which result in the loss of nuclear expression, the antibody anti-ATRX (clone AX1, 1:200, Dianova, Hamburg, Germany) was used. In addition, the O6-Methylguanine DNA‐methyltransferase (MGMT) promoter methylation status was assessed using methylation-specific pyrosequencing. Histological classification, molecular genetic analysis and tumour grading was accomplished by an experienced neuropathologist. To recognize any differences in the expression of the selected stem cell markers between the paired glioma samples, immunohistochemically staining was performed as follows: FFPE sections from the primary and recurrent tumor were deparaffinized in 100% xylene and rehydrated using graded alcohol series (100, 96, 70% for 5 min each). For antigen retrieval the specimens were heat-treated with 20 mM citric acid buffer, pH 6.0, in a pressure cooker for 10 min and cooked in TRS, pH 9, for 5 min for CD133 staining, respectively. Afterwards, endogenous peroxidase activity was inactivated by incubating the sections in 3% H2O2 for 15 min. Then sections were blocked with 10% normal goat serum (Thermo Fisher Scientific, Waltham, MA, USA) for 20 min at room temperature and then incubated overnight at 4°C with primary antibodies. The antibodies were directed against the stem cell markers CD133/1 (1:100, Cat# 130-090-422; Miltenyi Biotec, Bergisch Gladbach, Germany), Musashi (1:250, Cat# D270-3; MBL International, Woburn, Massachusetts, USA), Nanog (1:400, Cat# 4903; Cell Signaling, Danvers, Massachusetts, USA), Nestin (1:2400, Cat# MAB1259; R&D-Systems Inc., Minneapolis, USA), Oct 3/4 (1:50, Cat# sc-5279; Santa Cruz Biotechnology Inc., Dallas, Texas, USA), OPN (1:400, Cat# sc-21742; Santa Cruz Biotechnology Inc., Dallas, Texas, USA) and Sox2 (1:300, Cat# MAB2018; R&D-Systems Inc., Minneapolis, USA).
Next day, detection was performed using a multi-link kit (Super Sensitive Link-Label IHC Detection System, BioGenex, Fremont, California, USA), consisting of a link antibody and a label antibody, followed by development in 3, 3’-diaminobenzidine tetrahydrochloride (DAB) solution (Dako, Glostrup, Denmark). The sections were counterstained with hematoxylin for 2 min, dehydrated by inverse graded alcohol series (70, 96 and 100%) and mounted permanently with Histokit II (Roth, Karlsruhe, Germany). The percentage of stained cells per tumor area (100 tumor cells) was determined by standard bright-field microscopy using a 20x objective. A total of five different areas were examined. Normal brain from autopsies was used as negative control. Appropriate positive control tissues for the primary antibodies were colon carcinoma (CD133), testis (Musashi), seminoma (Nanog and Oct4), kidney (Nestin) and ovarian cancer (OPN and Sox2).
Immunofluorescence staining and imaging
For detection of a possible co-localization of CD133 and Nestin we established an immunofluorescence approach. For immunofluorescence staining, sections were deparaffinized and rehydrated as described above. Antigen retrieval was performed for 25 min in 20 mM citric acid buffer (pH 6.0), in a pressure cooker. Sections were rinsed with dH20 and in Tris-buffered saline (TBS) and were treated with 10% horse serum, 0.3% Triton X100 in TBS, for 1 h at room temperature. As primary antibodies, anti-human CD133 (Sigma-Aldrich, clone 2F8, ZooMAb® rabbit monoclonal, 1:50) and Nestin (R&D Systems, MAB1259, mouse monoclonal, 1:2400) were used. Antibodies were diluted in antibody-dilution buffer (DCS - Innovative Diagnostik Systeme, Hamburg, Germany) and incubated overnight. Next day, sections were washed twice in TBS and treated for 3 h at room temperature with the corresponding secondary antibodies (Jackson ImmunoResearch Laboratories, Inc., Pennsylvania, USA; stock solutions with 0.55 mg / ml IgG). The following secondary antibodies were used: donkey anti-rabbit IgG affiniPure (H + L)-Cy3-550 (1:400) and goat anti-mouse IgG affiniPure (H + L)-Alexa488 (1:400). Sections were washed twice in 1×TBS and stained with 4′, 6-diamidino-2-phenylindole (DAPI) (2mg/ml stock solution, freshly diluted 1:5000 in 1×TBS) for 5 min at room temperature. Sections were washed twice in 1×TBS and were finally embedded in Aquapolymount (Polysciences). Kidney tissue was used as positive control for anti-Nestin. For negative controls, biopsy samples from healthy donors (from autopsies) were used. Cross-reactivity of secondary antibodies was tested by using secondary antibodies in the absence of corresponding primary antibodies.
Confocal Laser scanning microscopy and image processing
Confocal imaging was performed with an IX81 Olympus microscope, an Olympus FV1000 confocal laser scanning system with a FVD10 SPD spectral detector and diode lasers (405, 473, 559 and 635 nm laser lines). Images were acquired with an Olympus UAPO 20x (air, numerical aperture 0.75) objective. For confocal scanning, a pinhole setting representing one Airy disc was used. 12-bit z-stack images were processed by maximum intensity projection and were adjusted in brightness and contrast using Image J software (Rasband, W.S., ImageJ, U.S. National Institutes of Health, Bethesda, Maryland, USA, https://imagej.nih.gov/ij/). Images are shown as RGB images (8-bit per color channel). Fluorescence images were processed for final presentation using Adobe Photoshop CS5.