Paradoxical effects of platelet-derived growth factor-A overexpression in malignant mesothelioma. Antiproliferative effects in vitro and tumorigenic stimulation in vivo.

Malignant mesothelioma is associated with asbestos exposure and remains resistant to all therapeutic intervention. Previous studies have suggested an enhancing role for platelet-derived growth factor (PDGF) in mesothelial tumorigenicity, although the mechanism by which PDGF facilitates tumorigenicity is unknown. Here, we evaluate the contribution of PDGF-A expression to mesothelial tumorigenicity using ectopic modulation of PDGF-A expression. We find, in accordance with other reports, that the receptor for PDGF-A, although expressed at high levels in normal human mesothelial cells, is not easily detectable in mesothelioma. Further, we show that PDGF-A overexpression is responsible for autocrine downregulation of its receptor. Our data indicate, surprisingly, that for mesothelioma cells in vitro, high-level activation of a PDGF-A-PDGF receptor loop is antiproliferative whereas abrogation of PDGF-A expression stimulates growth. These data suggest that PDGF-A does not contribute to tumorigenicity by autocrine stimulation of proliferation. In contrast, increased PDGF-A expression in vivo increases tumor incidence and growth rate and decreases the latency period to tumor formation whereas abrogation of PDGF-A expression decreases tumor incidence and increases latency. Thus, the tumorigenic effect of PDGF-A must act through paracrine mechanisms relevant at early stages of tumor initiation.

Human mesothelioma samples overexpress both cyclooxygenase-2 (COX-2) and inducible nitric oxide synthase (NOS2): in vitro antiproliferative effects of a COX-2 inhibitor.

Accumulating data demonstrate overexpression of both inducible NO synthase (NOS2) and cyclooxygenase-2 (COX2) in many epithelial neoplasias. In addition, cyclooxygenase inhibitors have been shown to have antineoplastic and prophylactic efficacy against human colon cancer and in mouse models of this disease. Mesothelioma arises in a context of asbestos exposure and chronic inflammation, which would be expected to enhance the expression of these inducible enzymes. This study demonstrates that both inducible enzymes were expressed in 30 human mesothelioma tissues but were not detectable in nonreactive mesothelial tissues from the same individuals. In contrast, areas of reactive mesothelial cells stained positively for these enzymes. In vitro exposure of human mesothelioma cell lines to the COX2 inhibitor, NS398, revealed dose- and time-dependent antiproliferative activity, whereas the NOS2 inhibitor, 1400W, had no detectable inhibitory effect. Surprisingly, nonmalignant human mesothelial isolates expressed both NOS2 and COX2 in vitro at the same level as mesothelioma cell lines but were less sensitive to NS398 inhibition. This finding indicates that these nonmalignant isolates may retain properties of reactive mesothelial cells and suggests that targets in addition to COX2 may be involved in the antiproliferative response of mesothelioma cell lines. These results have clinical significance because of the selective activity of the drug coupled with the therapeutic resistance and poor prognosis of mesothelioma. The findings presented here suggest that further preclinical studies of these inhibitors in animal models of mesothelioma would be of great interest.

Dominance of ErbB-1 heterodimers in lung epithelial cells overexpressing ErbB-2. Both ErbB-1 and ErbB-2 contribute significantly to tumorigenicity.

This article examines differential expression and heterodimer formation of ErbB family members in tumorigenic and nontumorigenic human bronchial epithelial cells (HBECs). This cell system was developed previously as a model for lung adenocarcinoma by overexpression of c-erbB-2 in nontumorigenic, T antigen-immortalized HBECs. Earlier studies demonstrated that a tumorigenic clone from T antigen-immortalized nontumorigenic cells overexpressing ErbB-2 endogenously produced high levels of transforming growth factor (TGF)-alpha, and that reducing TGF-alpha by 93% eliminated tumorigenicity. In the present report, comparison of ErbB species between the tumorigenic cells (E6T) and their nontumorigenic derivatives (E6TA) demonstrated all four receptors in both cell types. However, in E6TA cells, ErbB-3 and -4 were present primarily in ErbB-1 heterodimers, suggesting that ErbB-1 is a preferred heterodimer partner within this cell system, expressing endogenous ErbB receptors and ligands and overexpressing ErbB-2. The ErbB-1/-2 species was present at high levels in E6T and absent in E6TA cells. Mitogen-activated protein kinase activity was elevated in E6T relative to E6TA. Elevated activity was eliminated by blocking surface expression of either ErbB-1 or ErbB-2. Endoplasmic reticulum trapping of ErbB-1 eliminated tumorigenicity, whereas ErbB-2 internalization was selected against during tumor formation. These data demonstrate the importance of TGF-alpha-mediated signaling through the ErbB-1/-2 heterodimer in development of the tumorigenic phenotype. This work further suggests that ErbB-3 and -4 species may also contribute to tumorigenic conversion and that their expression levels may be increased by signaling initiated by TGF-alpha.

ErbB-2 kinase is required for constitutive stat 3 activation in malignant human lung epithelial cells.

Overexpression of the growth factor receptor ErbB-2/Her2/Neu has been implicated in the development of non-small-cell lung cancer. We have reported that the transformation of human lung epithelial cells by c-erbB-2 also requires an active ErbB-1 (EGF receptor) and the autocrine production of its ligand, TGF-alpha. In this report, we demonstrate that STAT 3 is constitutively activated in these cells by the TGF-alpha-stimulated ErbB-1/-2 heterodimer complex. STAT 3 activation was confirmed by mobility shift assays and nuclear localization. ErbB-1 was required, but not sufficient for the TGF-alpha-induced activation of STATs. Inhibition of ErbB-2 kinase activity by tyrphostin AG825 prevented the constitutive activation of STAT 3 in the TGF-alpha-producing, ErbB-1 expressing cell line. Our results demonstrate a requirement for ErbB-2 kinase activity to establish constitutive STAT 3 activation resulting from an autocrine ErbB-1/ TGF-alpha loop. Int. J. Cancer 83:564-570, 1999. Published 1999 Wiley-Liss, Inc.

Production of epidermal growth factor related ligands in tumorigenic and benign human lung epithelial cells.

We recently demonstrated that human lung epithelial cells, overexpressing ErbB-2, formed tumors in nude mice only when high levels of transforming growth factor alpha (TGFalpha) were produced. Cells transfected with a TGFalpha antisense vector failed to form tumors in nude mice. In order to further evaluate the importance, for tumorigenicity, of TGFalpha and its stimulation of ErbB family signalling, the production of other EGF family growth factors by these human lung epithelial cells was studied. We demonstrate for the first time that both tumorigenic and non-tumorigenic human lung epithelial cells produced, in addition to TGFalpha, amphiregulin, betacellulin, heparin-binding EGF and heregulin. These data suggest that human lung epithelial cells have the potential for multifactorial modulation of ErbB receptor family signalling through control of ligand as well as receptor production. In this system, the probable importance of TGFalpha-stimulated signaling for tumorigenicity is supported by its 13-fold higher production in tumorigenic as compared with non-tumorigenic cells and the 2-fold or lower differences observed in production of the other epidermal growth factor (EGF) family ligands.

Asbestos induction of extended lifespan in normal human mesothelial cells: interindividual susceptibility and SV40 T antigen.

Normal human mesothelial cells from individual donors were studied for susceptibility to asbestos-induction of apoptosis and generation of an extended lifespan population. Such populations were generated after death of the majority of cells and arose from a subset of mesothelial cultures (4/16) whereas fibroblastic cells (5/5) did not develop extended lifespan populations after asbestos exposure. All mesothelial cultures were examined for the presence of SV40 T antigen to obtain information on (i) the presence of SV40 T antigen expression in normal human mesothelial cells and (ii) the relationship between generation of an extended lifespan population and expression of SV40 T antigen. Immunostaining for SV40 T antigen was positive in 2/38 normal human mesothelial cultures. These cultures also had elevated p53 expression. However, the two isolates expressing SV40 T antigen did not exhibit enhanced proliferative potential or develop an extended lifespan population. Asbestos-generated extended lifespan populations were specifically resistant to asbestos-mediated but not to alpha-Fas-induced apoptosis. Deletion of p16Ink4a was shown in 70% of tumor samples. All mesothelioma cell lines examined showed homozygous deletion of this locus which extended to exon 1beta. Extended lifespan cultures were examined for expression of p16Ink4a to establish whether deletion was an early response to asbestos exposure. During their rapid growth phase, extended lifespan cultures showed decreased expression of p16Ink4a relative to untreated cultures, but methylation was not observed, and p16Ink4a expression became elevated when cells entered culture crisis. These data extend the earlier observation that asbestos can generate extended lifespan populations, providing data on frequency and cell type specificity. In addition, this report shows that generation of such populations does not require expression of SV40 T antigen. Extended lifespan cells could represent a population expressing early changes critical for mesothelioma development. Further study of these populations could identify such changes.

Resistance of pleural mesothelioma cell lines to apoptosis: relation to expression of Bcl-2 and Bax.

A failure of normal apoptosis, often due to mutant p53, may contribute to the formation of a cancer and to its resistance to therapy. Mesothelioma, an asbestos-induced tumor, is highly resistant to therapy but generally expresses wild-type p53. We asked whether mesothelioma was resistant to apoptosis and whether resistance was associated with altered expression of the antiapoptotic protein Bcl-2 or proapoptotic protein Bax. We found that three mesothelioma cell lines (1 with wild-type p53) were highly resistant to apoptosis induced by oxidant stimuli (asbestos, H2O2) or nonoxidant stimuli (calcium ionophore) compared with primary cultured mesothelial cells. By immunostaining, one of these three lines expressed Bcl-2 but only during mitosis. By immunoblotting, 3 of 14 additional mesothelioma lines (9 of 14 with wild type p53) expressed Bcl-2 but all 14 of 14 expressed the proapoptotic Bax, giving a low ratio of Bcl-2 to Bax. We conclude that mesothelioma cell lines are resistant to apoptosis and that the failure in apoptosis is not explained by Bcl-2 but by other mechanisms that counteract the proapoptotic effect of Bax.

The role of transforming growth factor alpha production and ErbB-2 overexpression in induction of tumorigenicity of lung epithelial cells.

Over-expression of erbB-2 is associated with shortened survival of patients with lung adenocarcinomas. We demonstrated that human lung epithelial cells, overexpressing erbB-2, formed tumours in nude mice only when high levels of transforming growth factor (TGF)-alpha were produced (E6T cells). To define the role that TGF-alpha production played in induction of tumorigenicity, a non-tumorigenic TGF-alpha-negative clone of ErbB-2 overexpressing cells (E2 cells) was transfected with an expression vector for TGF-alpha (E2alpha cells). Transfected clones produced TGF-alpha at 11-25% of the level produced by the E6T cell line. Tumorigenic E6T cells transfected with a TGF-alpha antisense vector (E6TA cells) expressed only 6% of the TGF-alpha level of the parental cells. Clones of E6T, E6TA, E2 and E2alpha were inoculated into athymic nude mice to measure tumorigenic potential. E6T cells formed tumours with a 70% efficiency. E2, E6TA and E2alpha cells failed to form tumours. The levels of EGFR were similar in non-tumorigenic E2 and tumorigenic E6T cells but higher in E2alpha and E6TA cells, and ErbB-2 were greatly overexpressed in an E2alpha clone. In vitro, ErbB-2 co-immunoprecipitated with EGFR in lysates of unstimulated E6T and E2alpha TGF-alpha-producing cells, indicating that the lower TGF-alpha levels were sufficient to induce in vitro heterodimerization. These studies suggest that induction of the tumorigenic phenotype depends on achieving a threshold level of TGF-alpha sufficient to activate downstream signalling by ErbB-2 containing active heterodimers.

Oncogenes and tumor-suppressor genes in mesothelioma--a synopsis.

Invariably mesothelioma is diagnosed late in the development of the disease when treatment is no longer effective. Therefore, a key to reducing the mortality rate of this neoplasm is knowledge of the general sequence of genetic events between initiation of mesothelial cells and the emergence of the metastatic tumor cells. Unfortunately, relatively little is known about the early changes in the genesis of this disease. Of the known changes, the most frequent are in the tumor-suppressor genes p16INK4a and NF2 and possibly the SV40 virus large T-antigen oncogene. The molecular nature of the changes in these genes as well as other alterations are addressed in this overview.

TGF-beta1 resistance is not associated with alterations in TGF-beta type II receptors in immortalized human lung epithelial cells.

A frequent cause of TGF-beta resistance is decreased expression or mutation of TGF-beta receptor Type II (TGFbetaRII) protein. We previously isolated two isogeneic subclones of the human bronchial epithelial cell line BEAS-2B that are respectively resistant (R.1) or sensitive (S.6) to the growth inhibitory effects of TGF-beta1. In this study, we examined TGFbetaIIR expression, ability to bind TGF-beta1, and cDNA sequence in the resistant cell line. Immunofluorescent analyses with antibody to TGFbetaRII indicated that this protein was expressed at the surface of R.1 cells. Affinity binding studies showed that TGF-beta1 bound equally well to the resistant (R.1) and sensitive (S.6) cell lines. PCR cloning and sequencing of TGFbetaRII cDNA revealed no changes from wild type in the resistant cell line. We conclude that alterations in TGF-beta Type II receptor are not responsible for TGF-beta resistance in this cell line.

High levels of MDM2 are not correlated with the presence of wild-type p53 in human malignant mesothelioma cell lines.

Prior analysis of 20 human mesothelioma cell lines for p53 status revealed only two mutations and one p53 null cell line, although p53 expression was detected in most cell lines. In addition, mRNA and protein expression of the retinoblastoma gene product in human mesothelioma cell lines is similar to normal controls. We have tested for p53 induction after exposure to ionising radiation and demonstrate this induction and, to a lesser extent, p21(WAF1) induction, in both normal mesothelial cells and p53-positive mesothelioma cell lines. We postulated that high levels of MDM2 might alter p53 and retinoblastoma tumour-suppressor function in mesothelioma. However, Southern blot analysis for mdm2 indicated that no amplification had occurred in 18 mesothelioma cell lines tested. Steady-state mRNA and protein levels also did not indicate overexpression. These results indicate that high levels of MDM2 are not responsible for inactivating the functions of wild-type p53 or the retinoblastoma gene product during the pathogenesis of malignant mesothelioma.

SV40-induced immortalization and ras-transformation of human bronchial epithelial cells.

Non-tumorigenic SV40-immortalized human cells may be transformed to tumorigenicity by activated oncogenes, but the molecular genetics of this process are still poorly understood. We describe here 4SV40-transformed bronchial epithelial (BE) cell lines that became immortalized after a period of crisis, and then transfection of 6 BE lines or sub-lines with an activated c-Ha-ras (EJ-ras) oncogene. pSV2neo-transfected cells did not form any tumors in athymic nude mice. Even though each of the EJ-ras-transfected lines was shown to be expressing the mutant ras gene, only one cell line, BEAS-2B, and 2 of its sub-lines were tumorigenic after transfection. We conclude that immortalization is not sufficient for BE cells to be transformed by the EJ-ras oncogene. Thus there are at least 2 unknown genetic events in this in vitro model of carcinogenesis: escape from crisis (immortalization), and development of ability to cooperate with activated ras in tumorigenic transformation. We found no evidence that either immortalization or ability to complement ras is related to abnormalities of the SV40 T antigens, of p110RB or of p53.

Clonal variation of tumorigenic potential in v-Ha-ras-transformed human bronchial epithelial cells: relationship to ras oncogene expression and CAD gene amplification.

Infection of an SV40 large-T antigen-"immortalized" human bronchial epithelial cell line with a Zip-v-Ha-ras retroviral vector resulted in a mass culture that was tumorigenic in athymic nude mice. A tumor cell line derived from passage of the mass culture in vivo, however, exhibited increased tumorigenicity and v-Ha-ras expression. To examine and compare the molecular events involving the ras oncogene during cell transformation in vitro and subsequent tumor formation in vivo, clonal cell populations were isolated from the v-Ha-ras-transformed mass culture. While the clonal cell lines exhibited diverse tumorigenic profiles, these differences did not correlate with v-Ha-ras expression. However, the expression of the activated ras gene, while not necessary for growth in vitro, did appear to be associated with a selective growth advantage in vivo. In addition, the modulation of gene amplification ability in these cells was not associated with the induction of tumorigenicity or v-Ha-ras expression.

Induction of anchorage independent growth and serum resistance in immortalized human bronchial epithelial cells by alteration of the cytoskeleton.

Malignant transformation is frequently accompanied by obvious changes in cytoarchitecture, but the importance of these changes has been difficult to assess in view of the large number of other cellular changes that also occur. In this study, we transfected the SV40-immortalized human bronchial epithelial cell line, BEAS-2B, with human wild-type beta or gamma actin gene expression plasmids to induce cytoskeletal changes and to determine whether this was associated with altered cellular growth properties. Cells expressing the exogenous full-length actin genes underwent a fibroblastoid change in morphology which was reflected in changes in their pattern of actin cable organization, and acquired both the ability to grow under anchorage-independent conditions and resistance to the normal growth inhibitory effects of fetal bovine serum. These phenotypic changes correlated with changes in actin mRNA levels, but not with changes in actin protein levels. The phenotypically altered cells were not tumorigenic when injected subcutaneously in athymic nude mice, and they retained the ability to suppress the tumorigenic potential of a lung carcinoma cell line, HuT-292. Therefore, alteration of the cytoskeleton of immortalized human bronchial epithelial cells resulted in the acquisition of some properties commonly found in malignant cells, but did not result in tumorigenicity.

Biological consequences of overexpression of a transfected c-erbB-2 gene in immortalized human bronchial epithelial cells.

In order to examine the effects of the overexpression of c-erbB-2 (HER-2, neu) on human bronchial epithelial cells, a human c-erbB-2 expression vector was introduced into the simian virus 40 large T-antigen-immortalized human bronchial epithelial cell line BEAS-2B. Isolation of multiple clonal cell lines after selection revealed a wide range of expression of the gene product gp185erbB-2. While three of six clones tested expressed gp185erbB-2 at levels detectable by immunocytochemistry, only one, B2BE6, induced adenocarcinoma-like tumors in athymic nude mice. Both a nontumorigenic clone, B2BE2, and a tumorigenic clone, B2BE6, expressed comparable amounts of gp185erbB-2, which became phosphorylated on tyrosine in response to treatment with the c-erbB-2 ligands gp30 and p75. These data suggest that overexpression of c-erbB-2 in human bronchial epithelial cells can contribute to, but is not sufficient for, induction of tumorigenicity in this human model system.

Development of tumorigenicity in simian virus 40-immortalized human bronchial epithelial cell lines.

Of five SV40-transformed clonal human bronchial epithelial cell lines previously shown to be nontumorigenic at early passages (R. R. Reddel et al., Cancer Res., 48: 1904-1909, 1988), two lines (BES-1A1 and BEAS-2B) from different donors have become weakly tumorigenic with further passaging. BES-1A1 passage 26 cells formed tumors in 3 of 9 athymic nude mice given s.c. injections, whereas BEAS-2B cells of > or = 32 passages formed highly cystic tumors at 8 of 58 injection sites after long latency periods [17 +/- 7 (SD) weeks]. These tumors took a total of 36 +/- 8 weeks to reach a diameter of 1.0 cm. Tumor cell lines were established from four BEAS-2B tumors, and these are resistant to the growth-inhibitory effects of serum, an inducer of squamous differentiation in BEAS-2B and normal bronchial epithelial cells. This finding supports the hypothesis that development of resistance to inducers of terminal squamous differentiation may be a step in the process of bronchial carcinogenesis. One of these tumor cell lines, B39-TL, is significantly more tumorigenic than the others and has a deletion from the short arm of chromosome 3 as has been described previously for some naturally occurring human bronchial carcinomas. Thus, from the clonally derived BEAS-2B cell line, cell populations with various degrees of tumorigenicity have developed. Analysis of the changes in these cells may yield insights into the multiple events involved in acquisition of the tumorigenic phenotype.

Tumorigenic conversion of human mesothelial cells as a consequence of platelet-derived growth factor-A chain overexpression.

Overexpression of platelet-derived growth factor (PDGF)-A as well as PDGF-B chain mRNA has previously been reported in human mesothelioma cell lines. In this report, it has been established that the A but not the B chain protein is expressed at detectable levels in cell lysates and conditioned medium from these cell lines. In order to investigate the effect of overexpression of PDGF-A chain in a human mesothelial cell model system, a retroviral vector containing a human PDGF-A chain cDNA insert under the control of the Moloney murine leukemia virus (MoMLV) promoter was inserted into the SV-40 T-antigen immortalized human mesothelial cell line MeT-5A. Selected cells showed overexpression of PDGF-A chain relative to MeT-5A and induced tumors in athymic nude mice. PDGF-A chain overexpression was also found in the tumor specimens excised from the mice. PDGF-A mRNA and protein were expressed at a higher level in the tumor explant cell lines, suggesting a correlation of tumorigenicity with A chain production.

Human mesothelioma cells and asbestos-exposed mesothelial cells are selectively resistant to amosite toxicity: a possible mechanism for tumor promotion by asbestos.

To determine if asbestos exposure could contribute to mesothelial cell carcinogenesis by selection and/or expansion of an initiated cell population, we compared normal human pleural mesothelial cells to either human mesothelioma cell lines or mesothelial cells transfected with cancer-related genes for sensitivity to amosite fibers in vitro. Neither normal nor mesothelioma cells were directly stimulated to replicate or increase DNA synthesis by any of the asbestos exposure conditions tested. The potential selective effect of asbestos exposure was demonstrated by a differential sensitivity of normal mesothelial cells and mesothelioma cells to amosite: for example, up to 20-fold higher concentrations of amosite fibers were required to inhibit replication of mesothelioma cell lines than normal mesothelial cells. In addition, a significant resistance (4-fold) to amosite toxicity was observed for SV40 immortalized mesothelial cell lines that had previously been selected in vitro for resistance to asbestos. SV40 immortalized cells that have become tumorigenic after transfection with either Ha-ras or PDGF A-chain genes were not significantly more resistant to the cytotoxic effects of amosite than primary normal cells, and the primary cells were equally sensitive to amosite as mesothelial cells that were only immortalized by SV40. The sensitivity of normal mesothelial cells to asbestos does not appear to be simply a result of general fragility of the mesothelial cells, since similar levels of hydrogen peroxide and silica were cytotoxic for normal mesothelial cells and mesothelioma cell lines. Because mesothelioma cells have a greater resistance to asbestos cytotoxicity than normal mesothelial cells, we hypothesize that a differential resistance to cell killing by asbestos fibers in vivo may result in a selective expansion of an initiated or transformed cell population and thus contribute to the carcinogenesis process. Since tumorigenicity and asbestos resistance occur independently of one another in genetically altered mesothelial cell lines, genotypic and phenotypic alterations that lead to tumorigenic conversion may not be the same changes that provide resistance to cell killing by asbestos.