Twenty-four oligodendrogliomas+ and two relapses were retrieved from the files of the Department of Pathology at "Marques de Valdecilla Hospital", Santander (Spain), over a period of more than ten years, beginning in 1981. Formalin-fixed, paraffin-embedded tumour samples from our files were employed. The tumours were characterized according to age, sex, year of diagnosis and histological degree in accordance with the WHO classification and outcome. All tumours underwent routine histological examination, including fixation in buffered formalin (4%).
Immunofluorescence and immunohistochemical techniques were used to co-localize the expression of 379-LP1248 and GFAP (Ab1 – Glial Fibrillary Acidic Protein – Bionova MS-280-R7). Propidium iodide was used for counterstaining DNA. Antibody 379-LP1248, recognizing a 16 amino acid peptide located in the third cytoplasm loop of the alpha1a-AR, was purchased from Life Span Biosciences, Seattle (WA).
Titration of antibody 379-LP1248 without antigen retrieval was carried out. The best results were obtained at 1/750 – 1/1000 dilution. Against the 379-LP1854, a secondary antibody Alexa Fluor 488 goat anti rabbit IgG(H+L) conjugated (Molecular Probes – Lab Net A-11008) was used at 1/100 and Alexa Fluor 633 rabbit anti mouse IgG(H+L) conjugated (Molecular Probes – Lab Net A-21063) at 1/100. Propidium iodide was used for 5 minutes.
A light-microscopic immunohistochemical study in an Axioplan/Axiophot 2 (Zeiss) was assessed semi-quantitatively by strength of staining (0 no stain, 1 weak, 2 moderate, and 3 strong). In addition, a confocal immunofluorescence quantitative study was performed in a Laser Scanning Microscope LSM 510 META (Zeiss) with a laser line of argon (458/488/514 nm 25 mW) and two laser lines of helium-neon (543 nm 1 mW; 633 nm, 5 mW). For this second purpose, we captured images from each case with three different optical magnifications, objectives: ×20 plan-Apochromat numerical aperture 0.75; ×40 plan-Neofluor numerical aperture 1.3 oil and ×63 plan-Apochromat numerical aperture 1.4 oil. We got an average of 7 pictures per case and a total number of 167 images for our statistical analysis. Between these images we always carried out a tile scan of 3 × 3 or 4 × 4 with 20× or 40× according to each case. From the 167 images collected, we recorded the mean, standard deviation and sum of the fluorescent intensities, area and density of expression for each channel ignoring the intensities lower than 4%. Following this method, we got nine different numerical values related to the immunoreactivity for the two primary antibody studied (GFAP and alpha1a-AR), and for the propidium iodade.
Finally, nuclear area, perimeter, aspect, axes (major and minor), diameters (max, mean and min.), radius (max. and min.) margination, ratio of perimeter-area, roundness and sizes (length and width) were measured by means of system for automated image analysis (Image Pro Plus – Media Cybernetics) based on a single tiff picture (512 × 512 pixels) per case at magnification ×400. In each case around 150 and 300 nuclei from the most cellular area were measured.
The statistical analysis was carried out using R software for Linux. For the univariate statistical analysis, the t-test was performed (two-sample t). A level of significance of p < 0.05 was adopted.