Introduction of the activated N-ras oncogene into human fibroblasts by retroviral vector induces morphological transformation and tumorigenicity

The introduction of activated N-ras cDNA into normal diploid human skin fibroblast cell cultures using the retroviral vector pZIPneo results in a spectrum of morphologies ranging from near normal to, in rare instances, dense piled-up colonies of morphologically transformed cells. However, none of the clones isolated were transformed as assessed by growth on agar or tumorigenicity in nude mice. Introduction of both c-myc and N-ras oncogene cDNAs into normal skin fibroblasts failed to produce transformation as assessed by growth on agar and tumorigenicity in nude mice, although c-myc infection alone conferred immortality and the resultant doubly infected cell line was immortal. Using the same construct, activated N-ras cDNA was shown to transform immortalized human fibroblasts to tumorigenicity. However, immortalization per se was shown not to guarantee 'co-operation' with an activated N-ras gene to give malignant transformation. Although numerical and structural chromosome aberrations (clonal and non-clonal) were observed in some of the cell strains isolated after retroviral infection, these were not directly associated with viral infection, the presence of the oncogenes or with the morphologically transformed phenotype.     

Clustering of discrete cell properties essential for tumorigenicity and metastasis. I. Studies of Syrian hamster embryo fibroblasts spontaneously transformed in vitro

The expression of two discrete cell properties related to the host natural effector mechanisms, i.e., resistance to damage by H2O2, a cytotoxic product of activated macrophages, and the ability to secrete PGE, which inhibits NK-cell cytotoxicity, has been examined in parental Syrian hamster embryo cells spontaneously transformed in vitro (STHE strain) and in 18 in vivo selected sublines. In all cell variants, resistance to H2O2 and PGE-releasing activity were either both expressed, or not expressed at all. Parental STHE cells and 5 variants selected in vivo, which were equally highly susceptible to H2O2-induced damage, did not release any detectable amount of PGE upon contact with NK cells. In contrast, 13 other STHE variants selected in vivo and characterized by their resistance to H2O2, all released PGE upon contact with NK cells. Thus, these two biochemically unrelated cell phenotypic characteristics are likely to be either simultaneously selected in vivo, or united in cluster which pre-exist or appear in rare cell variants of the parental cell population in the conditions of in vivo natural selection pressure.     

Modulation of differentiation in PA-1 human teratocarcinoma cells after N-ras oncogene-induced tumorigenicity.

We have been studying the effect of oncogenes on differentiation using the human ovarian teratoma-derived cell line PA-1. From this study we have characterized variants representing four stages relevant to multistage carcinogenesis, two non-tumorigenic and two tumorigenic. The two non-tumorigenic cell variants differ in that one is resistant to transformation by ras oncogenes whereas the other can be transformed to tumorigenicity. When these non-tumorigenic PA-1 variants are treated with retinoic acid (RA), a morphogen, they stop dividing, begin to express homeobox genes, and change in morphology. Transfection of an activated N-ras oncogene into ras-resistant non-tumorigenic PA-1 cells does not alter the RA responsiveness of the cells, indicating that expression of the activated oncogene is not sufficient for blocking RA-induced differentiation. Spontaneous activation of an N-ras oncogene leading to tumorigenic transformants and gene transfer-induced N-ras transformants are resistant to these effects of RA. However, another spontaneous transformant of PA-1 cells that does not contain an activated N-ras is responsive to RA. We prepared somatic cell hybrids of the RA-non-responsive, N-ras-transformed and tumorigenic PA-1 cell and the RA-responsive, ras-resistant non-tumorigenic PA-1 cell; the hybrid cell lines continue to express the oncogene but are non-tumorigenic. These non-tumorigenic hybrids are responsive to RA with regard to morphological changes, growth arrest and induction of homeobox gene expression. Tumorigenic revertants of these hybrids arise as a result of the loss of some chromosomes; these hybrid cells express the oncogene but have lost RA responsiveness. These results indicate that tumorigenic transformation in general is not sufficient to induce RA resistance, and resistance to differentiation may be oncogene-specific. In addition, the expression of an activated N-ras oncogene alone is insufficient to induce resistance to RA and ras-induced tumorigenicity is necessary. Therefore, some feature of cellular metabolism that is altered by and discordantly segregates with tumorigenic transformation controls responsiveness to RA. This controlling element is presumably a tumor suppressor.     

Clustering of discrete cell properties essential for tumorigenicity and metastasis. II. Studies of Syrian hamster embryo fibroblasts transformed by Rous sarcoma virus

Tumorigenic and metastasizing activities (TGA; MA) and susceptibility, or resistance to H2O2 and PGE-releasing activity (H2O2R + PGEs+ phenotype) have been examined in 6 Syrian hamster embryo cell strains transformed in vitro with Rous sarcoma viruses (Schmidt-Ruppin and Prague strains). Early observations of extremely high level of TGA and even MA of RSV-SR-transformants never selected in vivo have been confirmed. The correspondence of these properties with a high level of expression of H2O2R + PGEs+ phenotype and its clustering character were demonstrated in 4 RSV-SR transformants, while significantly lower expression of all these characteristics, including TGA, was observed in 2 RSV-Prague transformants. High level of spontaneous MA was noticed in some RSV-SR transformants. A tumor cell line induced in vivo by RSV-SR did not differ from the cell strain transformed in vitro by RSV-SR. Inhibition of H2O2R + PGEs+ phenotype in one of RSV-SR transformants was obtained with non-toxic doses of BCNU and indomethacin, leading to a marked decrease of TGA.     

The role of wild-type and mutated N-ras in the malignant transformation of liver cells

In order to determine the role of N-ras overexpression and mutation in malignant liver cell transformation, wild-type and mutated N-ras were transfected into the rat liver epithelial cell line OC/CDE 22, and N-ras expression, growth kinetics, growth in soft agar, and tumorigenicity in vivo as well as the involvement of the mitogen-activated protein kinase (MAPK) signal transduction pathway in the expression of the malignant phenotype were analyzed. Although OC/CDE 22 cells transfected with wild-type N-ras showed a high expression of N-ras at the mRNA and protein levels, the cells did not grow in soft agar and were not tumorigenic in vivo. In contrast, OC/CDE 22 cells transfected with mutated N-ras showed anchorage-independent growth and were tumorigenic. When cultured in fetal bovine serum-supplemented medium, OC/CDE 22 cells expressing mutant N-ras showed a higher proliferation rate than nontransfected OC/CDE 22 cells or OC/CDE 22 cells transfected with wild-type N-ras. When held in serum-free medium, untreated OC/CDE 22 cells did not grow at all, while OC/CDE 22 cells transfected with wild-type or mutant N-ras proliferated at a similar rate, which can be explained by the high MAPK activity in these cells. Selective inhibition of the MAPK cascade abolished the growth of OC/CDE 22 cells carrying mutant N-ras in soft agar; furthermore, these cells ceased pile up and formed monolayers on Petri dishes. Thus, activation of the MAPK signaling pathway, though alone not sufficient to malignantly transform liver cells (as shown in liver cells overexpressing wild-type N-ras), is not only essential for growth control but also for the expression of the malignant phenotype (as demonstrated in liver cells transformed by mutated N-ras).     

Tumorigenicity of human HT1080 fibrosarcoma X normal fibroblast hybrids: chromosome dosage dependency

The tumorigenic capacity of hybrids formed by fusion of the highly tumorigenic HT1080 human fibrosarcoma cell line with nontumorigenic normal fibroblasts was examined. The HT1080 also contains an activated N-ras oncogene. Near-tetraploid hybrids which contained an approximately complete chromosomal complement from both parental cells were nontumorigenic when 1 X 10(7) cells were injected s.c. into athymic (nude) mice, whereas the parental HT1080 cells produced tumors in 100% of the animals with no latency period following injection of 2 X 10(6) cells. Tumorigenic variants were obtained from these hybrids which had lost only a few chromosomes compared to cells from the nontumorigenic mass cultures. In addition, several near-hexaploid hybrids were obtained which contained approximately a double chromosomal complement from the HT1080 parental line and a single chromosomal complement from the normal fibroblasts. All of these near-hexaploid hybrids produce tumors in 100% of nude mice with no latency period. Our results indicate that tumorigenicity of these particular human malignant cells of mesenchymal origin can be suppressed when fused with normal diploid fibroblasts. In addition, the results suggest that tumorigenicity in this system is chromosomal dosage dependent, since a diploid chromosomal complement from normal fibroblasts is capable of suppressing the tumorigenicity of a near-diploid but not a near-tetraploid chromosomal complement from the tumorigenic HT1080 parent. Finally, the loss of chromosome 1 (the chromosome to which the N-ras oncogene has been assigned) as well as chromosome 4 was correlated with the reappearance of tumorigenicity in the rare variant populations from otherwise nontumorigenic near-tetraploid hybrid cultures. Our results also suggest the possibility that tumorigenicity in these hybrids may be a gene dosage effect involving the number of activated N-ras genes in the hybrids compared to the gene(s) controlling the suppression of the activated N-ras genes.     

Sensitivity of melanoma cell lines to natural killer cells: a role for oncogene-modulated HLA class I expression?

N-ras and c-myc oncogenes were found to be activated in melanoma. High c-myc expression renders melanoma cell lines sensitive to lysis by natural killer (NK) cells. This effect is mediated by locus-specific downmodulation of HLA-B expression by c-myc. Cell lines with a mutation in the N-ras gene were relatively sensitive to NK cells irrespective of HLA class I expression. These findings indicate that NK cells can kill tumor cells with activated myc or ras oncogenes in various ways, thus providing potential mechanisms to eliminate cancer cells with an activation of these oncogenes.     

Murine natural anti-tumor antibodies. II. The contribution of natural antibodies to tumor surveillance

The objective of this study was to investigate whether natural anti-tumor antibodies (NAb) contributed to the surveillance of small inocula of syngeneic tumors. Experiments were designed to distinguish between NAb and NK-mediated host resistance. Four approaches were used: (I) the isolation of tumor variants differing in their tumorigenicity and susceptibility to these mechanisms, and to activated macrophages; (2) a comparison of the effects of adjuvants on the modification of both host resistance and the activity of these effectors; (3) the relationship between the ontogeny of the natural resistance mechanisms and tumorigenicity in aged mice; and (4) the use of a Winn-type assay. These studies provided support for a role in natural resistance correlated with both the ability of tumors to bind NAb, and the production of NAb in adjuvant-stimulated mice. Furthermore, the frequency of tumors observed after tumor challenge more closely correlated with the ontogeny of natural antibody than NK cells, and tumors coated with NAb were less tumorigenic than controls. The reduced tumorigenicity of an NK-sensitive tumor, when compared to an NK-resistant variant of the same line, provided evidence for NK-cell-mediated natural resistance in young adult mice. It was concluded that natural resistance to tumors is a complex phenomenon dependent on the tumor phenotype, as well as the activity of several effector mechanisms, and that natural anti-tumor antibody must be considered an important component of host resistance.     

Comparative analysis of natural killer cell and macrophage recognition of concanavalin A-resistant Chinese hamster ovary cells: role of membrane oligosaccharides

As was previously shown, a mutation rendering Chinese hamster ovary (CHO) cells concanavalin A (Con A) resistant (Con AR) resulted in enhanced binding and susceptibility to lysis by natural killer (NK) cells. In the present study altered cell surface carbohydrate expression on the Con AR CR-7 cell line was demonstrated. This could be detected by flow cytometry with the use of fluorescein isothiocyanate-conjugated lectins Con A, wheat-germ agglutinin, and peanut agglutinin, and by a radiolabeled monoclonal antibody (49H.8) that recognizes a phenyl-P-galactose determinant. Two other independently selected Con AR CHO lines (BCR-2 and ECR-1) also exhibited increased NK cell reactivity, which suggests that a similar and, likely, a single mutation has altered the phenotype of these cells. Tunicamycin, an inhibitor of N-linked glycosylation, was able to reverse the binding to the CHO lines with the use of conditions that did not affect protein or DNA synthesis. The mutation rendering the CR-7 line Con AR and more sensitive to NK cell lysis and binding resulted in reduced killing and adhesion by activated bone marrow-derived macrophages from DBA/2J and CBA/CaJ mouse strains. Inhibition of macrophage and NK cell lysis by simple monosaccharides was demonstrated; however, macrophage killing was inhibited only by D-mannose, whereas NK cell lysis was reduced by all sugars tested. Neither macrophage nor NK cell adherence to tumor monolayers was affected by any of the monosaccharides. These results suggest that complex N-linked oligosaccharides are important for NK cell and macrophage binding and lysis.     

Studies on clonal heterogeneity in two spontaneously metastasizing mammary carcinomas of recent origin

We have studied the clonal heterogeneity of 2 spontaneously metastasizing mammary carcinomas which recently arose spontaneously in C3H/He female retired breeders. Cells of early (2nd to 5th) transplant generations of these tumors were cloned by a combination of semi-solid agarose colony formation and limiting dilution techniques. Growth characteristics of the various clones in vitro and their tumorigenicity in vivo were evaluated. Subsequently, the role of host immunity and of interclonal interactions in regulating growth of the different clones in vivo were analyzed. We found that, whereas all 16 clones isolated from one tumor (T-58) grew rapidly in vivo and in vitro, 10 clones isolated from the second tumor (MT-2) showed a wide disparity in their growth rates in vivo. Taken together, these clones could generally be divided into 3 categories: (1) rapidly growing lines which grew in vivo at rates similar to or higher than those of the parental line; (2) slow-growing lines which grew more slowly than the parental line; and (3) non-growers which failed to produce tumors in vivo with doses of up to 5 X 10(6) cells injected either s.c. or i.v. but grew in vitro at rates comparable to the parental line. No correlation could be established between the various growth potentials exhibited by these tumor lines and tumor cell morphology in vitro and in vivo, as determined by light and electron microscopy. Sublethal irradiation (550-650 R) of young animals prior to tumor inoculation, or before inoculation of tumor cells into old, low NK syngeneic mice, failed to modify the growth of slow-or non-growing lines in vivo, indicating that host cellular defense mechanisms against the clones, if existent, were not mediated by NK or radiosensitive B or T cells. When clonal interactions were studied by the simultaneous injection of different clones in vivo at different s.c. sites, we found that a slow-growing line failed to modify the growth rate of a rapidly growing line, but accelerated the growth of a second slow-growing line injected simultaneously on the contralateral side, and that this enhancement of tumor growth was radioresistant. A mixture of these 2 lines also grew more rapidly than the individual lines alone. Our findings suggest that phenotypic variations in tumorigenicity can be found in clonal lines derived from spontaneous primary tumors and that these variations are not related to cell cycle properties as measured in vitro.(ABSTRACT TRUNCATED AT 400 WORDS)     

Analysis by somatic-cell fusion of the phenotypic properties of a mutant embryonal carcinoma cell-line

We have previously reported the isolation of an embryonal carcinoma (EC) cell mutant created via retroviral insertion (1). The mutant (NR1-6) is aberrant in respect to morphology, tumorigenicity and response to retinoic acid (RA). In order to determine the dominance/recessive hierarchy and genetic relationships of these phenotypes, we created hybrid lines from mutant and parental cells. Three hybrid clones (Hyb 1, 2a and 2b), which retain both mutant and parental loci, were analyzed. Our results indicate that in all parameters - morphology, adhesion, qualitative and quantitative response to RA and tumorigenicity - the hybrids behave similarly to parental cells. In no case did any of the hybrid lines exhibit a mutant phenotype. The data indicate that the phenotypes we have studied are genetically interrelated and that the parental characteristics are dominant.     

Malignant properties of P635 glioma are independent of glial fibrillary acidic protein expression

Summary To examine the relationship between the malignant behavior of rat glioma cells and expression of the differentiation antigen glial fibrillary acidic protein (GF), we assayed the tumorigenicity and metastatic ability of P635 and its GF + or GF – clones. We injected P635 (GF+) and clone 45 (GF–) cells intramuscularly in nude mice. Both lines formed local tumors which metastasized to lungs with equal efficiency. We then injected these lines intravenously in nude mice. Both produced experimental metastases in lungs and other organs with equal efficiency. Finally we injected P635 and several GF+ and GF– clones into the chorioallantoic membrane veins of naturally immune-deficient chick embryos and measured cell growth in embryonic liver. We again found that all clones survived and grew equally well. P635 cells are capable of extracranial growth and metastasis, two important features of the malignant phenotype. We conclude that GF in P635 is a neutral marker with respect to tumorigenicity and metastatic ability.     

Inhibition of tumor growth and enhancement of metastasis after transfection of the γ-interferon gene

Cells from the spontaneous metastatic TS/A mammary adenocarcinoma of a BALB/c mouse were transfected with the murine γ-interferon (IFN-γ) gene. Six clones (IFN-γ clones) releasing between 2 and 6,000 international units (IU) of IFN-γ/ml culture medium, were compared to TS/A parental cells (TS/A-pc) and to cells transfected with neomycin resistance gene only (NEO cells). Autocrine IFN-γ up-regulated membrane expression of H-2 class-1 and Ly-6 glycoproteins, but did not alter cellular proliferation in vitro. All IFN-γ clones gave rise to progressive tumors with a growth rate significantly slower than that of tumors induced by TS/A-pc and NEO cells, and inversely correlated with the amount of IFN-γ secreted. TS/A-pc and NEO tumors displayed a marginal reactive infiltrate, whereas those formed by IFN-γy clones were massively infiltrated mostly by macrophages. In T- and NK-deficient mice the growth of tumors formed by I FN-y clones was not enhanced. In vitro tests showed that IFN-γ clone cells were markedly more lysed by macrophages than TS/A-pc and NEO cells, while they remained poorly sensitive to NK and LAK cells. These data as a whole suggest that the development of solid tumors by IFN-γ clones is primarily hampered by macrophages and not by T-lymphocytes or NK cells. When spontaneous metastatic ability was compared, 2 IFN-γ clones releasing 2-4 IFN-γ lU/ml were significantly more metastatic, while most IFN-γ clones appeared to be as metastatic as NEO cells. By contrast, following intravenous challenge, all IFN-γ clones produced 5-10 times more experimental metastases than NEO cells. The higher metastatic ability of IFN-γ clones was attributed to increased resistance to NK cells since, in NK-depleted BALB/c mice, metastatic spread of IFN-γ clones was not enhanced, whereas a 50-fold increase in the number of metastases was found upon injection of NEO cells.     

Membrane dynamic alteration associated with the tumorigenicity of polyoma-transformed and revertant hamster cells.

Steady state fluorescence polarization studies were carried out on membranes of polyoma-virus-transformed, revertants, re-revertants, and normal golden hamster cells. Four clones of revertant cells which exhibit different levels of reversion were isolated. The degree of reversion in these revertant clones was characterized in vitro by their contact inhibition behavior and in vivo by their tumorigenicity to hamsters. A good correlation between these two criteria was observed. Re-revertant cells were obtained from that revertant clone which exhibited the highest degree of reversion (similar to normal cells) and which could still produce tumors in hamsters. Re-revertants resembled the polyomavirus-transformed cells in both contact inhibition and tumorigenicity. Fluorescence polarization values of 1.6-diphenyl-1,3,5-hexatriene-labelled cells correlated well with cell tumorigenicity: normal and revertant cells which were not malignant exhibited distinctly lower fluorescence polarization values than those of the tumorigenic transformed and re-revertant cells. Intermediary revertant clones exhibited fluorescence polarization values in between those of normal and transformed cells, yet indistinguishable within experimental uncertainty.     

p38 kinase is a key signaling molecule for H-Ras-induced cell motility and invasive phenotype in human breast epithelial cells

Ras expression has been suggested as a marker for tumor aggressiveness of breast cancer,including the degrees of invasion and tumor recurrence.We showed previously that H-ras, but not N-ras, up-regulates matrix metalloproteinase 2 expression and induces invasive phenotype in MCF10A human breast epithelial cells (A. Moon, et al. Int. J. Cancer, 85: 176-181, 2000). In this study, we show that H-ras also promotes cell motility more effectively than N-ras in MCF10A cells. We have investigated H-ras-specific signaling pathway(s) critical for H-ras-mediated cell motility and invasive phenotype. Whereas neither H-ras nor N-ras activated c-Jun NH(2)-terminal kinase 1, both H-ras and N-ras effectively activated extracellular signal-regulated protein kinase (ERK) -1,2. Importantly, prominent activation of p38 mitogen-activated protein kinase was shown only in H-ras-activated cells but not in N-ras-activated MCF10A cells. Functional significance of H-ras-activated p38 in invasiveness and cell motility was evidenced by studies using SB203580, a chemical inhibitor of p38, and a dominant-negative construct of p38. Whereas inhibition of c-Jun NH(2)-terminal kinase 1 activity had no effect on H-ras-induced MCF10A cell invasion and motility, the inhibition of the ERK pathway using a chemical inhibitor PD98059 or dominant-negative mutant of mitogen-activated protein/ERK kinase 1, an activator of ERKs, significantly reduced H-ras-induced invasion and migration. We also provide evidence that p38 and, to a lesser degree, ERKs, are critical for H-ras-mediated up-regulation of matrix metalloproteinase 2. Taken together, the present study shows that H-ras activation of both p38 and ERKs induces cell invasion and motility, whereas N-ras activation of ERKs alone is not sufficient. This study reveals the p38 kinase as a key signaling molecule differentially regulated by H-ras and N-ras, leading to H-ras-specific cell invasive and migrative phenotypes in human breast epithelial cells.     

A novel cell-based assay for the evaluation of anti-ras compounds.

In order to identify drugs active against mutated ras oncogenes we have developed an in vitro assay employing two clones of the human fibrosarcoma cell-line, HT1080 which carries an N-ras gene mutated at codon 61. Clone, HT1080scc2, retains the transformed phenotype of the parental line, whilst the other, HT1081c, is a morphologically flat, non-tumourigenic, revertant with under-representation of the chromosome carrying the transforming N-ras allele. The clear implication of mutant ras in maintaining the transformed nature of HT1080scc2 was confirmed when these cells were microinjected with the pan ras neutralising antibody Y13-259, which resulted in the morphological detransformation of these cells to a phenotype resembling that of the HT10801c clone. A number of known anti-cancer drugs with modes of action unrelated to ras function were found to be equipotent against both clones. However, when compounds chosen on the grounds of their potential selective cytotoxic or differentiating activity were tested some interesting results were obtained. Thus 8-bromo cAMP affected some morphological detransformation of HT1080scc2 cells and reduced their colony forming potential. The IMP-dehydrogenase inhibitors, tiazafurin and mycophenolic acid also flattened the morphology of the transformed clone. Fumagillin, an antibiotic reported to exhibit selective activity against ras transformed cells showed very marked and selective cytostatic effects against HT1080scc2 cells with IC50 values as low as 1 x 10(-11) M.     

Natural killer cells activated by interleukin 2 treatment in vivo respond to interleukin 2 primarily through the p75 receptor and maintain the p55 (TAC) negative phenotype.

Interleukin 2 (IL-2) induced activation of unstimulated resting natural killer (NK) cells or resting T-cells initially occurs following binding of IL-2 through the p75 receptor that is expressed primarily by these cells. However, this IL-2/p75 interaction induces TAC chain synthesis and formation of high affinity IL-2 receptor required for the proliferation of resting peripheral blood lymphocytes. In this study, we present data indicating that NK cells activated by in vivo IL-2 treatment, in contrast to resting NK cells, respond and proliferate to further IL-2 in vitro using primarily the p75 receptor with only a minor component of cells responding through the high affinity receptor. These in vivo activated NK cells minimally expressed the TAC chain and maintained this TAC negative phenotype while proliferating in response to IL-2. The primary involvement of the p75 receptor in the proliferative response of these cells to IL-2 was demonstrated by the need for concentrations of IL-2 higher than 44 pM to obtain a significant response and by the dramatic inhibition of this response by anti-p75 monoclonal antibody. Anti-TAC monoclonal antibody inhibited only the poor proliferation obtained at low doses of IL-2 suggesting a minor role for TAC and high affinity IL-2 receptors. This was in contrast to the partial inhibition of proliferation by anti-p75 or anti-TAC observed in unstimulated pretherapy peripheral blood lymphocytes suggesting that these cells respond to IL-2 through both high affinity receptors and intermediate affinity p75 receptors. The T-cells isolated from in vivo activated peripheral blood lymphocytes, despite expressing TAC, were not responsive to IL-2, suggesting that these cells express predominantly nonfunctional low affinity TAC receptors. NK cells activated by IL-2 in vivo represent a unique model system of IL-2 dependent cells that respond and proliferate to IL-2 essentially through the p75 IL-2 receptor.