A biochemical analysis of thoracic neuroblastomas: a Pediatric Oncology Group study.

A biochemical analysis was performed on tumor tissues from 20 patients who presented with thoracic neuroblastomas. Nine patients were under 1 year of age at the time of diagnosis, and 12 patients had stage A disease. Eighteen of the 20 patients are disease free with a mean follow-up of 5 1/2 years. The ganglioside composition of the tumor tissue was investigated, since these cell membrane components have been proposed to play a role in cell to cell interaction and may be altered on cell transformation. In addition, the ganglioside composition of the central nervous system changes with maturation. Previous studies in children with neuroblastoma have shown that tumor tissue containing more complex gangliosides is associated with a better prognosis. Neuroblastomas from patients with thoracic primaries were found to contain more complex gangliosides of the b series (GD1b, GT1b) and fewer monosialogangliosides, suggesting a more differentiated cellular composition. Tissue from one of the thoracic patients who died lacked GT1b. The absence of this ganglioside has proven to be an indicator of a poor prognosis. Four specimens contained no detectable GD2, which is thought to be a specific marker for neuroblastomas. These data suggest that the improved prognosis seen with thoracic neuroblastomas is due to a basic biologic difference within these tumors, and this finding should be taken into consideration when planning therapy.

Cellular effects of a brain extract factor or factors inhibiting neuroblastoma cell growth in vitro.

The mechanism of inhibition of neuroblastoma cell proliferation in vitro by extract prepared from the brains of newborn mice was partially elucidated. Flow cytometric analysis of propidium iodide-stained cells showed that a significant increase occurred in the pre-G0/G1 cell population after exposure to the extract for 24 hr. Fluorescence microscopy indicated that the pre-G0/G1 cells had small "micronuclei" in association with their nucleus, whereas post-G2/M cells had enlarged propidium iodide-staining nuclei. 3H-thymidine incorporation was unchanged from that of control cells, in cells exposed to the extract for either 6 or 24 hr. No effect was seen on choline acetyl transferase activity or on morphology. These results suggest that the inhibitor blocks neuroblastoma cell proliferation either by inducing an uneven distribution of DNA at the time of cell division or by altering its packaging within the nucleus.

Association of endogenous substrate with solubilized bovine brain sialidase.

Nonidet P40 solubilized up to 90% of the sialidase, active towards added ganglioside substrate, that was associated with the total membrane fraction prepared from gray matter of bovine brains. Solubilized sialidase acted upon endogenous substrate (sialic acid containing compounds solubilized with the enzyme), hydrolyzing approximately 50% of the readily available sialosyl residues within 20 min. During a 2-hr reaction time 80% of the polysialylated gangliosides solubilized with the enzyme were acted upon. A 20-min lag was observed before sialidase acted upon added ganglioside substrate. The lag could be reduced to less than 2 min when the enzyme was allowed to act on endogenous substrate prior to exposure to exogenous substrate, suggesting that the solubilized enzyme acted preferentially on endogenous substrate. A protease inhibitor prevented much of the 86% loss of activity towards added substrate that was seen when the enzyme was stored at 4 degrees C for 6 days; activity towards endogenous substrate decreased only 34%.

Ganglioside composition of human neuroblastomas. Correlation with prognosis. A Pediatric Oncology Group Study.

The ganglioside composition of neuroblastoma samples from 53 patients was determined. Tumor sites included the adrenals (15), and the thoracic (20), abdominal (15), and pelvic (3) areas. Age of the patients at the time of surgery ranged from 1 day to over 10 years and the Stage of the tumors from A to D. Differences in ganglioside patterns were observed, with neuroblastomas from patients who were either disease positive or dead of disease tending to have more monosialogangliosides and fewer gangliotetraose gangliosides of the B series (formula; see text) than tumors from patients who were disease-free. More specifically, six of the seven patients lacking GT1b died of disease, suggesting that the absence of GT1b is indicative of a poor prognosis.

Partial characterization of a brain extract factor(s) inhibitory to transformed neural cells.

The factor(s) present in extracts prepared from the brains of newborn A/J or C57B1/6 mice, which inhibits S20Y neuroblastoma cell growth in vitro, was partially characterized. Twice as much inhibitory activity was extracted per gram wet weight of brain than torso, and inhibitor recovery was greatest in extracts prepared from brains of mice 1 week or less in age. The inhibitory factor(s) was water-soluble and was stable to heating at 100 degrees C, to freezing, and to lyophilization. It was susceptible to the action of pronase. The factor(s) behaved like a molecule of molecular weight approximately 700 upon passage through ultrafiltration membranes. Growth of rat hepatoma (H4), murine melanoma (B16), and transformed murine fibroblasts (WT19 and B6-HCMV) was not significantly inhibited by brain extract. Growth of rat glioma cells (C6) was significantly reduced but to a lesser degree than that of murine neuroblastoma cells (S20Y and N115) and glioma cells (G26-20). These results suggest that the inhibitor expresses a cell specificity.

Response of mature mice to challenge with neuroblastoma after inoculation with neuroblastoma cells as neonates.

Injection (s.c.) of 2 X 10(5) S20Y neuroblastoma cells into adult A/J strain mice regularly produced a higher incidence of progressive, lethal tumor growth. In contrast, successful tumor growth was significantly less frequent (P less than 0.001) when the same number of tumor cells were injected into neonatal A/J mice. A significant number (P less than 0.001) of matured mice that had rejected the initial S20Y cell inoculum as neonates were able to reject 2 subsequent challenges with S20Y cells.

Density-dependent changes in gangliosides and sialidase activity of murine neuroblastoma cells.

Density-dependent changes in ganglioside composition, Vibrio cholerae neuraminidase (VCN)-susceptible sialyl residues, and membrane-associated sialidase activity were determined for the cholinergic murine neuroblastoma cell line S20Y. A decrease in total ganglioside sialic acid and VCN-releasable sialic acid was observed with increasing cell density. GM3 was the major ganglioside component of preconfluent S20Y cells, whereas GD1a was predominant in postconfluent cells. Sialidase activity increased in confluent and postconfluent cells and may account for the reduction in total ganglioside sialic acid observed with increasing cell density. In contrast, while adrenergic N115 cells showed a decrease in VCN-susceptible sialic acid residues with increasing cell density, there was no significant change in ganglioside composition or ganglioside sialic acid levels.

Distribution in spleen subcellular organelles of sialidase active towards natural sialogylcolipid and sialoglycoprotein substrates.

A procedure was devised for the preparation of enriched populations of subcellular organelles from homogenized bovine spleen. The fractions obtained were characterized for arylsulfatase, succinate dehydrogenase, UDPgalactosyltransferase and 5'-nucleotidase activities. The distribution of sialidase (acylneuraminyl hydrolase, EC 3.2.1.18) activity directed towards either endogenous substrate or exogenous ganglioside substrate suggests that it is enriched in the plasma membrane/microsomal fractions. Sialidase activity towards exogenous sialoglycoproteins, isolated from erythrocyte membrane, was enriched in the least dense of the plasma membrane/microsomal-containing fractions. The endogenous sialidase substrates were primarily the sialoglycolipids, hematoside and disialogangliosides. At the pH optimum, 3.8, and 37 degrees C, release of endogenous sialic acid was linear with time for 3 h. At the end of this time, 85% or more of the available endogenous substrate was hydrolyzed.

Ecto-ganglioside-sialidase activity of herpes simplex virus-transformed hamster embryo fibroblasts.

Cellular location of ganglioside-sialidase activity was determined in confluent hamster embryo fibroblasts transformed with herpes simplex virus type 2. Approximately equal specific activities of ganglioside-sialidase activity were found to be associated with the crude lysosomal and crude plasma membrane fractions isolated from whole cell homogenates. Whole transformed cells hydrolyzed exogenous ganglioside substrate, suggesting a partial location of the cellular sialidase on the outer surface of the plasma membrane of these cells. Intact cells were treated with the diazonium salt of sulfanilic acid, a nonpenetrating reagent inhibitory to ecto-enzymes (DePierre, J.W., and M. L. Karnovsky. 1974. J. Biol. Chem. 249:7111-7120). Cytoplasmic lactate dehydrogenase activity was not inhibited by this treatment, and mitochondrial succinate dehydrogenase activity was inhibited only 10%, indicating that intracellular enzymes were not affected. 5'-Nucleotidase activity was diminished 90%, and sialidase very rapidly lost 40% of its exogenously directed activity. These results show that, in herpes simplex virus-transformed fibroblasts, ganglioside-sialidase is both a lysosomal and a plasma membrane enzyme. The plasma membrane sialidase is capable of acting on endogenous plasma membrane sialolipids and also functions in the cultured transformed cell as an ecto-enzyme which can attack exogenous substrates.