The localization of alkaline phosphatase in HeLa cells was examined by electron microscopic histochemistry and subcellular fractionation techniques. Two monophenotypic sublines of HeLa cells which respectively produced Regan and non-Regan isoenzymes of alkaline phosphatase were used for this study. The electron microscopic histochemical results showed that in both sublines the major location of alkaline phosphatase is in the plasma membrane. The enzyme reaction was occasionally observed in some of the dense body lysosomes. This result was supported by data obtained from a subcellular fractionation study which showed that the microsomal fraction rich in plasma membrane fragments had the highest activity of alkaline phosphatase. The distribution of this enzyme among the subcellular fractions closely paralleled that of the 5'-nucleotidase, a plasma membrane marker enzyme. Characterization of the alkaline phosphatase present in each subeellular fraction showed identical enzyme properties, which suggests that a single isoenzyme exists among fractions obtained from each cell line. The results, therefore, confirm the reports suggesting that plasma membrane is the major site of alkaline phosphatase localization in HeLa cells. The absence of any enzyme reaction in the perimitochondrial space in these cultured tumor cells also indicates that the mitochondrial localization of the Regan isoenzyme reported in ovarian cancer may not be a common phenomenon in Regan-producing cancer cells.

No abstract

We used an optimized in-situ hybridization technique employing a biotinylated Epstein-Barr (EB) virus sequence, BamH1V (3.1 kb), to detect this sequence in 2 EB virus-infected cell lines (B95-8 and Namalwa) and 8 CNS lymphomas. We obtained a good hybridization signal from cytospins of B95-8 (EB virus productively infected) and Namalwa (EB virus latently infected, 1 copy per cell) cell lines. We were able to detect signal from both cell lines after overnight fixation in 10% formalin and paraffin embedding, but development time in the detection chromogen required longer incubation and the signal intensity was lower than in cytospin cells. We then used the technique to examine formalin-fixed, paraffin-embedded primary CNS lymphoma tissue from 4 patients who were immunocompromised (1 renal transplant, 3 acquired immune deficiency syndrome) and 4 patients who were not. All 4 CNS lymphomas from immunocompromised patients hybridized well with BamH1V, exhibiting a pattern of staining similar to Namalwa cells and nonlytically infected B95-8 cells. There was no relationship between the intensity and degree of reaction and the patients' survival. None of the 4 CNS lymphomas in immunocompetent patients or uninvolved brain showed any reactivity with BamH1V. We suggest that low-abundance targets are detectable in paraffin-embedded tissue by in-situ hybridization using biotinylated probes. Detection of EB viral sequences in CNS lymphomas in immunocompromised patients suggests a role for the virus in the pathogenesis of this tumor.

No abstract