Epidermal growth factor (EGF) at 3 nM maximally inhibits the proliferation of A431 epidermoid carcinoma cells. We show that at lower concentrations, in the range of 3-100 pM, EGF has a mitogenic effect on A431 cells. In Epidermal growth factor (EGF) promotes the growth of many cell types in vitro (1-3) and inhibits proliferation of several cell types-e.g., GH4 rat pituitary tumor cells (4), A431 epidermoid carcinoma cells (5, 6), and certain human breast cancer cells (7). EGF initially binds to receptors homogeneously distributed on the cell surface. Subsequent events have been described by various investigators and include receptor phosphorylation, aggregation, internalization, and degradation in lysosomes (1). The mechanism by which these events induce DNA synthesis and cytokinesis is unknown.It has been found that at least 6-8 hr of EGF exposure are required to stimulate DNA synthesis (8). Das and Fox have suggested that EGF-induced internalization and degradation of the EGF receptor are rate-limiting factors for EGF-induced mitogenesis (9, 10), perhaps through production of a second messenger. Recent studies showed enhancement of EGF stimulation of DNA synthesis by amine compounds, which inhibited clustering of receptors in coated pits (11), and by phorbol esters, which reduced both the affinity of EGF receptors for EGF and its subsequent degradation (12)(13)(14). These results suggest that EGF stimulation of cell growth might only require the presence of EGF-EGF receptor complexes at the cell surface in contradiction to the above hypothesis.Shechter et al., on the other hand, suggested that the stimulatory effect of EGF might be mediated by small amounts of high-affinity EGF receptors, which remain at the cell surface for more than 8 hr when occupied by EGF (15). King and Cuatrecasas also have suggested that the accumulation of stable intracellular complexes between high-affinity receptors and EGF are involved in growth stimulation, but the role of these highaffinity receptors in mitogenesis remains unclear (16).A431 cells lend themselves to the study of EGF interactions with receptors because of their extremely high number of EGF receptors (1-3 x 10' per cell) (1,17,18

No abstract

By using Southern blot analysis, we found that in two cases of human glioblastoma multiforme, cells carried amplified c-erbB genes which bore short deletion mutations within the ligand-binding domain of the epidermal growth factor (EGF) receptor. The products of these mutated c-erbB genes were about 30 kilodalton (kDa) smaller than the normal 170-kDa EGF receptor, and the tumor cell membrane fractions containing the 140-kDa abnormal EGF receptor showed a significant elevation of tyrosine kinase activity without its ligand. In view of the similarity to the activated viral and cellular erbB genes in the avian system, these mutated and overexpressed EGF receptors might play a role in the onset or development of human glioblastoma cells.The c-erbB gene is known to be a proto-oncogene which encodes the epidermal growth factor (EGF) receptor (4,25,30). The v-erbB gene product of avian erythroblastosis virus in chickens has a truncated form of EGF receptor and expresses a higher level of tyrosine kinase activity without its ligand (11). Thus, qualitative and quantitative alterations appear to be crucial for the activation of the c-erbB gene in chickens. In human tumors, amplification of the c-erbB gene has been reported in several squamous cell carcinomas and glioblastomas (3,10,13,14,17,27,29). However, it was not certain whether the c-erbB gene in those tumor cells was also structurally altered.Since the activation of oncogenes in human brain tumors has not yet been extensively studied, we screened several transplantable brain tumors for the presence of abnormal proto-oncogenes, using Southern blot analysis (24). Table 1 lists the tumors tested; six were glioblastomas, and two were ependymomas. The histological characteristics of these tumors have not drastically changed during serial passages in athymic nude mice.Among 19 onc probes employed (myc, N-myc, myb, K-ras, N-ras, sis, src, fpslfes, yes, mos, fos, ros, fms, fgr, abl, rel, rafimil, erbB, erbB-2/neu), no amplification or rearrangement was detected in the transplantable brain tumors except for the v-erbB probe (26); two glioblastomas, GL-3 and GL-5, carried an amplified c-erbB gene (data not shown). Recently Libermann et al. have reported that about one-third of glioblastomas in primary human brain tumors contain an amplified c-erbB gene (13). They also indicated that this gene amplification is associated with a possible DNA rearrangement, but the details were not examined due to the limited availability of primary tumors. The frequency of amplification of the c-erbB gene in transplantable glioblastomas examined here is similar to that of their report.To examine the fine structure of the amplified c-erbB gene in glioblastomas, we used a human EGF receptor cDNA as a probe (kindly provided by I. Pastan, National Cancer Institute, Bethesda, Md.) (16, 28). EcoRI-digested DNAs of GL-3 and GL-5 cells carry a high copy number of the c-erbB gene (Fig. la), and the degree of amplification was almost the same as in the A431 squamous carcinoma cell line, which is known t...

No abstract

Expression of the c-erbB-2 gene product and the epidermal growth factor receptor (EGF-R) was investigated in 54 cases of human bladder cancer immunohistologically and by Western blot analysis. For detection of the c-erbB-2 product, two specific antibodies, a rabbit polyclonal antibody directed to the intracellular domain and a murine monoclonal antibody recognizing an epitope in the extracellular domain, were used. Seventeen cases of bladder cancer were stained by the anti-c-erbB-2 polyclonal antibody, while 20 cases were stained by the monoclonal antibody, with good correlation on both stainings (p less than 0.01). There were four c-erbB-2 positive cases in 26 G1 tumors, four in 15 G2 tumors, and nine in 13 G3 tumors. There were also eight erbB-2 positive cases in nine muscle-invasive tumors, nine of 45 superficial tumors, four of five with lymph node metastasis, and seven of 14 without metastasis, as revealed by staining with the polyclonal antibody. Thus, the c-erbB-2 gene product was more frequently expressed in high grade tumors (p less than 0.01), in high stage tumors (p less than 0.01), and nodal metastatic tumors (N.S. by Chi-square test). Twenty-two of the 54 tumors were stained by an anti-EGF-R monoclonal antibody, 528 IgG. The expression of EGF-R was independent of histological grading, tumor stage, and nodal status, and no correlation was observed between expression of the c-erbB-2 product and EGF-R. The c-erbB-2 product may be applicable as a tumor marker for evaluation of malignant potential, invasiveness, and probably metastatic potential of human bladder cancer.

Since the original demonstration by Kohler and Milstein (1) of the feasibility of using cell fusion to produce antibody-secreting hybridomas, several improved mouse myeloma parent cell lines have been developed. In recent years NS-1-Ag4-1 (NS-1) (2), Sp2/0-Ag-14 (Sp2/0) (3), and X63-Ag8.653 (X63.653) (4) have been commonly used to produce hybridomas . All three of these cell lines were independently derived from the original myeloma parent line P3-X63-Ag-8 (X63) (1), and all three cell lines differ from X63 with respect to Ig gene expression . Sp2/0 cells were obtained by subcloning Sp2/3-3, an X63 hybridoma (1, 3), while NS-1 and X63.653 were subcloned in several steps from X63. An examination of the growth requirements of NS-1 cells in serum-free medium revealed that NS-1 had absolute requirement for low density lipoprotein (LDL), which was not shared by NS-1 hybridomas (5). The lipid requirement of NS-1 cells was subsequently found to be a requirement for cholesterol (6). As cholesterol auxotrophy is an exceedingly rare phenotype in cultured mammalian cells, it was of interest to determine whether NS-1 cells were unique among X63 and the other derivatives of the MOPC21 cell line P3 (7). We have found that Sp2/0 cells require neither LDL (6) nor cholesterol for growth, and here we demonstrate that X63 and X63.653 cells, like NS-1 cells, are cholesterol auxotrophs . In addition, we report that cholesterol-independent variants of X63 and X63.653 can be selected . Materials and Methods Brief Definitive ReportMaintenance of Cells and Medium Preparation. P3-X63-Ag8, X63-Ag8 .653, Sp2/0-Ag14, and U937 cells were obtained from the American Type Culture Collection (Rockville, MD). Cells were maintained in a humidified atmosphere of 5% C02 at 37°C in RD + 5F (factor) medium supplemented with 1% or 5% FCS (HyClone Laboratories, Logan, UT). RD + 5F medium consisted of a 1 :1 mixture (by volume) of RPMI 1640 (Gibco Laboratories, Grand Island, NY) and DME (high glucose; Gibco Laboratories) with 5 ug/ml crystalline bovine insulin (Sigma Chemical Co., St. Louis, MO), 5 'Ug/ml human transferrin (Fe"-free), 10 juM 2-ME, 10 AtM 2-aminoethanol, and 10 nM sodium selenite (all from Sigma Chemical Co.). The 5F supplements were made as sterile 200X concentrates and stored at 4°C. RD medium included 15 mM Hepes buffer, 2 MM Lglutamine, 0 .01% sodium pyruvate (all from Sigma Chemical Co.), 2.2 g/L sodium bicarbonate (J.T. Baker Chemical Co., Phillipsburg, NJ), 20 mg/L penicillin G (Sigma Chemical Co.), 45 mg/liter streptomycin sulfate and 40 mg/L ampicillin (both Sigma

A quantitative assay method for epidermal growth factor receptors (EGFRs) of human tumor tissues was established, based on enzyme‐labeled avidin‐biotin (LAB) interaction with anti‐human EGFR monoclonal antibody 52SIgG. A standard calibration curve for EGFR estimation in human tumor tissues was obtained with A431#8 cells cloned from A431 human epidermoid carcinoma cell line. The coefficient of variance for the standard curve was below 35% in the application to tumor tissues from nude mice implanted with human tumor cell lines. The minimum tissue amount required for the quantitative assay was around 0.1 g (wet weight). Using the LAB method, the correlation between the level of EGFR number and tumor malignancy was examined for 14 human squamous cell carcinomas (SCCs) from the oral region. Seven of the SCCs showed a more than two‐fold higher EGFR number compared to normal gingival tissues. Three highly aggressive carcinomas with poor prognosis possessed five to ten times higher levels of EGFR number than normal tissues. The elevated EGFR level in the SCCs seems to correlate to increasing tumor size and the stage of SCCs as clinically classified according to the 1987 UICC TNM system.

No abstract