Isolation and partial purification of growth factors with TGF-like activity from human malignant gliomas

Development of a non-selecting, non-perturbing method to study human brain tumor cell invasion in murine brain

The infiltrative nature of glial and some meningeal neoplasms is responsible for the failure of surgical removal and high recurrence rate of these tumors. Modeling of this process in vitro and in vivo will lead to a better understanding of the pathophysiology of this process and identify targets for novel therapy directed towards this phenotype. We present the results of the development and refinement of two model systems of tumor invasion: one in vitro barrier assay using the basement membrane extract Matrigel, and one in vivo where molecular detection of tumor cells allows single cell discrimination by in situ hybridization histochemistry. These techniques have strong correlations which validate their utility as measures of nervous system tumor invasion.

TM-1 cells from an established human malignant glioma cell line produce PDGF, TGF-alpha, and TGF-beta which cooperatively play a stimulatory role for an autocrine growth promotion

We have previously established a human malignant glioma cell line, TM-1. TM-1 cells could proliferate in the serum-free medium. In the present study, immunochemical analysis demonstrated that platelet-derived growth factor (PDGF), transforming growth factor (TGF)-alpha, and TGF-beta are present in the serum-free medium conditioned by growing TM-1 cells. While the cells appeared to possess a single type of binding sites for epidermal growth factor (EGF) with properties comparable to those determined for other tumor cells, the conditioned medium did not contain EGF.PDGF, TGF-alpha, and EGF added exogenously to serum-free media stimulated thymidine incorporation into DNA of TM-1 cells. In addition, antibodies specific for PDGF and TGF-alpha suppressed this activity. These results indicate autocrine and stimulatory roles of PDGF and TGF-alpha for the proliferation of TM-1 cells. As observed for other tumor cells, TGF-beta by itself weakly suppressed thymidine incorporation by TM-1 cells. However, TGF-beta employed in combination with TGF-alpha or EGF appeared to stimulate thymidine incorporation, suggesting that a cooperative action of TGF-beta with different growth factors may be involved in the stimulatory growth regulation at least for TM-1 cells.

Partial characterization of glioma-derived growth factor 2: a novel mitogenic activity from human cell line D-54 MG

We have shown that several human malignant glioma cell lines are stimulated by bacterial lipopolysaccharide (E. coli 0111:B4, 1 microgram/ml) to produce a high molecular weight (> 200 kD) growth activity for BALB 3T3, clone A31 cells. This glioma-derived growth factor (GDGF-2) acts like a 'competence' factor. Malignant glioma cell line D-54 MG constitutively produced GDGF-2, which we have partially characterized from serum-free conditioned culture medium. GDGF-2 is resistant to heat (100 degrees C, 5 min), acidic (pH 2, 2 hr) or reducing (0.5 M 2 ME, 30 min) conditions as well as exposure to RNases; however, it is sensitive to > 4 freeze-thaw cycles, alkaline (pH 11, 2 hr) conditions or pre-treatment with proteolytic enzymes. GDGF-2 had a pl of 6.8 determined by preparative isoelectric focusing, bound to DEAE, with elution at 35 and 185 mM NaCl and at 43% acetonitrile from a C4 reversed phase column. GDGF-2 activity was not neutralized by antibodies to TGF alpha, TGF beta, PDGF, VEGF or TNF alpha indicating that it is not immunochemically related to these growth factors. However GDGF-2 co-chromatographed on Superose 12 HPLC (250 x 9 mm; 5% isopropanol, 6 mM CHAPS in PBS) with a substance that suppressed growth of mink lung epithelial cells (Mv1Lu), but not BALB 3T3 cells, and could be neutralized by anti-TGF beta antibodies. GDGF-2 activity eluted from heparin columns in 0.6 M NaCl; thus, it is not a heparin binding growth factor. D-54 MG cell line produced alpha 2-macroglobulin (alpha 2M), which is known to bind TGF beta; however, immunoprecipitation of alpha 2M did not deplete TGF beta or GDGF-2 activity. Further, neither GDGF-2 or TGF beta can be dissociated into lower molecular weight active components by chromatography in high salt (2 M NaCl) or 2-ME (0.5 M). GDGF-2 may be a novel autocrine or paracrine mitogen, stimulating mitotic division or interfering with normal cell growth regulation.

Oncogenes: cause or consequence in the development of glial tumors

Recent developments in the field of oncogenes and growth stimulatory factors have provided limited but essential models in neuro-oncology. The observation in gliomas of platelet growth factor (PDGF)-like immunoreactivity fits with the autocrine secretion model, rising the possibility for the growth factor independence of the cancer cells. The discovery of the tumor suppressor genes, for which loss of function mutations are oncogenic as in the RB gene of the retinoblastoma and p53 gene, has introduced a new concept of oncogenesis which could be useful even in the cure of the neoplasms. Several oncogenes are amplified and/or expressed in brain tumors, some associated with polymorphism leading to abnormal protein products. Therefore, corresponding functions, such as production of deficient epidermal growth factor receptor (EGFR) encoded by erb-B, are impaired. Abnormal chromosomal patterns have been recognized in brain tumors and found mainly in chromosomes 7 and 22 on which oncogenes erb-B and sis are located, respectively. Location of proto-oncogenes, which are normally expressed in the brain, indicate that they share common distribution patterns mainly involving the cerebellum, hippocampus and olfactory bulbs. These proto-oncogenes may be regulated by physiological and pathological events. The concept of oncogene involvement in brain tumors must be extended to include the other factors such as G-proteins, growth factor receptors, membrane-associated and cytoplasmic protein kinases, which are all responsible for the control of the cell growth and their response to external signals including chemotherapeutic drigs.

Interaction of acidic fibroblast growth factor and transforming growth factor-beta in normal and transformed glia in vitro

The mitogenic and morphological effects of acidic fibroblast growth factor (aFGF) and transforming growth factor-beta (TGF-beta) were assessed on cultured fetal rat astrocytes and C6 rat glioma cells in the presence and absence of serum. Astrocytes incubated with aFGF exhibited an increase in mitotic activity and characteristic morphological changes involving extensive process formation and rounding of cell bodies. Astrocytes incubated with TGF-beta underwent a slight decrease in mitotic activity and remained morphologically unchanged. Cells exposed to a combination of aFGF and TGF-beta demonstrated an attenuation of both the mitogenic and morphological changes observed in the presence of aFGF alone. The C6 glioma cells cultured in the presence of aFGF underwent a characteristic morphological change from a rounded piling cell mass to a more spindle-shaped bipolar cell layer, accompanied by an increase in mitotic activity. In contrast to the astrocyte cultures, increased growth and similar morphological effects were produced by TGF-beta. The combination of aFGF and TGF-beta did not result in attenuation of the mitogenic and morphological changes (as seen in astrocytic cells). These results suggest that, in normal fetal rat astrocytes, TGF-beta is capable of attenuating the mitogenic and morphological changes induced by aFGF in vitro. In the transformed C6 glioma cell line, aFGF and TGF-beta elicit similar mitogenic and morphological changes, without evidence of an antagonistic interaction as seen in normal astrocytes.

EGF receptor amplification and expression in human brain tumours

Human epidermal growth factor receptor (EGFr) gene amplification, rearrangements and expression were studied in tumours of the human nervous system. EGFr gene amplification was studied in 46 brain tumours. Gene expression was analysed by northern blot in 37 tumours and binding of its protein to EGF in 27 tumours. The EGFr gene was simultaneously amplified (with arrangements in 12.5% of gliomas) and overexpressed in 53% (9/17) of malignant gliomas, but never in meningiomas. In five high grade gliomas, amplification was always associated with a high level of receptors. However, since high amounts of EGF receptors found in one glioma were not the result of gene amplification, several systems of deregulation in EGFr production may exist and could be located at translational and/or post-translational levels.

Modulation of T-cell function by gliomas

Patients with primary intracranial tumors (gliomas) exhibit a profound decrease in immunity, the mechanism of which has, until recently, remained obscure. Here Thomas Roszman, Lucinda Elliott and William Brooks reveal that T cells obtained from these patients exhibit defects in interleukin 2 secretion and in expression of the high-affinity IL-2 receptor and they discuss the role played by immunosuppressive factors produced by gliomas in inducing these defects.

Inability of mitogen-activated lymphocytes obtained from patients with malignant primary intracranial tumors to express high affinity interleukin 2 receptors

Patients with primary malignant brain tumors manifest a variety of abnormalities in cell-mediated and humoral immunity. Diminished T cell reactivity has been shown in these patients to be linked to deficiencies in interleukin 2 (IL-2) production that cannot be overcome by exogenous IL-2. In this study, specific binding of radiolabeled IL-2 to PHA-stimulated lymphocytes from brain tumor patients demonstrates that the number of high affinity interleukin 2 receptors (IL-2R) is greatly reduced. FACS analysis indicates that the relative density of the p55 protein (Tac protein) is lower on the mitogen-activated lymphocytes obtained from patients than on comparably treated lymphocytes from normal individuals. These data indicate that mitogen-stimulated lymphocytes obtained from patients have fewer functional high affinity IL-2R principally because of the failure to express sufficient levels of the p55 protein for association with the p75 protein. Northern analysis of total RNA isolated from mitogen-stimulated T cells from patients demonstrates normal levels of steady state mRNA, which codes for the p55 protein. Moreover, there is no defect in the postranslational processing of the primary translation product of this mRNA suggesting that normal levels of the p55 protein are produced in activated T cells from patients.

Critical appraisal of experimental radiation modalities for malignant astrocytomas

The management of patients with supratentorial malignant astrocytomas has remained a major problem. Patients continue to die from a lack of local control in 90% of cases despite an improvement of median survival seen with the use of postoperative radiation therapy. Because of this, there has been considerable interest in exploring novel ways of possibly improving results. This paper reviews the rationale and clinical results with the use of altered fractionation schemes, brachytherapy, radiation sensitizers, hyperthermia, particle therapy, and radiosurgery in the treatment of these patients. Currently, there is no demonstrated advantage with the use of these experimental modalities in the initial management of patients. There would appear to be some benefit for selected patients who are treated with brachytherapy at recurrence, but its efficacy as part of initial management remains to be determined determined in ongoing randomized prospective trials.