Search

p53 immunostaining of histological sections shows inter- and intratumor variability in distribution and staining intensity which are usually scored semiquantitatively. In order to investigate the variation in p53 expression more accurately and its possible relation to other cellular parameters (e.g., DNA content), we have studied the possibility to measure p53 accumulation by multiparameter flow cytometry. To this end we have evaluated seven, commercially available, monoclonal antibodies (MAbs) against p53 (MAbs 1801, 240, 246, 421, 1620, Do1, and Do7) on five tumor cell lines with known p53 gene status: MCF-7 (wild-type p53 gene), COV362.cl4 and T47d (mutated p53 genes), and SAOS-2 and HL60 (no p53 mRNA). Localization of immunofluorescence was investigated with confocal laser scanning microscopy, immunofluorescence signal intensity with flow cytometry, and antibody specificity with Western blotting. Subsequently, single cell suspensions from two breast carcinomas were flow cytometrically analyzed after triple staining for p53, cytokeratin 8/18, and DNA, and compared to immunohistochemical staining. MAbs Do1 and Do7, and to a lesser extent MAb 421, accurately discriminated p53 positive from p53 negative cell lines. Even at high concentrations these MAbs yielded nuclear immunofluorescence, whereas with MAbs 1801, 240, and 246 strong cytoplasmic signals in both the p53 accumulating and p53 negative cell lines were seen. By using lower antibody concentrations the cytoplasmic immunofluorescence disappeared, but simultaneously the nuclear p53 immunostaining intensity in p53 accumulating cell lines decreased, resulting in false negative nuclei. With MAb 1620 only weak intranuclear spots were obtained in all cell lines tested. Western blotting yielded results with MAbs 1801, Do1, and Do7 in the 53 kD region of the p53 accumulating cell lines. The signal intensity obtained with MAb 1801 was much less compared to MAbs Do1 and Do7. Although all three MAbs are also described as wild-type p53 specific, only MAbs, Do1 and Do7 showed bands in the 53 kD region of cell line MCF-7. With MAb 1801 ascites and MAb 1801 supernatant an additional approximately 80 kD band was present in all cell lines tested, including SAOS-2, indicating cross reactivity of this MAb. Immunohistochemical staining of two clinical breast carcinomas confirmed the results obtained in the cell lines. Multiparameter flow cytometric analysis of these breast carcinomas with MAbs Do1 and Do7 showed intratumor heterogeneity for p53 accumulation, which was independent of DNA index heterogeneity. We conclude that MAbs Do1 and Do7 enable quantitative analysis of p53 accumulation in a multiparameter flow cytometric analysis.

We have developed a "one-tube" triple staining procedure that allows the identification of intratumor phenotypic subpopulations by FCM. Solid tumors were dissociated by a combined mechanical/ enzymatic method. Ovarian ascites tumor cell aggregates were enzymatically dissociated using trypsin. An antikeratin 8/18 MAb was used to label the epithelial fraction of these tumor samples. A second MAb directed against the leukocyte common antigen (LCA) was applied to identify nonneoplastic DNA-diploid cells. Other MAbs used as a second marker were directed against a tumor-associate surface, a cytoplasmic, or a nuclear antigen. Cells were stained using subclass-specific fluorescein-isothiocyanate (FITC) or R-phycoerythrin (PE)-conjugated antibodies. DNA was stained with propidium iodide (PI). Triply stained samples were measured on a standard bench-top flow cytometer (FACScan). Keratin 8/18-positive cells, LCA-positive cells, and DNA could be simultaneously detected in dissociated breast carcinomas, mixed Müllerian tumors, and ovarian ascites specimen for refining DNA index (DI) calculations and S phase fraction (SPF) determination. Coefficients of variation (CV) of the G0G1 peak of the DNA histograms obtained ranged from 2.55% to 4.64% and from 2.71% to 4.71% for the DNA-diploid and -aneuploid fractions, respectively. In DNA-diploid tumors, antigen expression (HER-2/Neu, proliferating cell nuclear antigen) could be analyzed without interference of fluorescence signals from nonneoplastic cells. Neoplastic tumor subpopulations were clearly identified based on both DNA-ploidy status and heterogeneity of antigen expression. The present method offers new possibilities for multiparameter DNA FCM on clinical samples and enables the identification of intratumor neoplastic subpopulations based on antigen expression and DNA-ploidy status.