The effect of 3T3 fibroblasts on the expression of anchorage independence and cornification of oral keratinocytes

This study examined the effect of 3T3 fibroblasts on the expression of anchorage independence and the degree of cornification in early cultures of three carcinoma-derived epithelial cell lines (R59, R63a, R63b) and in one cell line derived from non-malignant dysplastic epithelium where there was no evidence of invasion (R66a). The epithelial cell lines originated from the palatal (R63a, R66a) and the lingual (R59, R63b) mucosa of rats that had been painted with the carcinogen 4-nitroquinoline N-oxide. In the absence of 3T3 fibroblasts, progressive culture resulted in an increase in the colony forming efficiency (CFE) of R63a, R63b and R59 and a decrease in the percentage of cornified cells in all cell lines. 3T3 fibroblasts caused a decrease in the CFE and the degree of cornification in the 3T3-dependent cell line (R63a), particularly at the lower passages, but these parameters remained essentially unchanged by 3T3 fibroblasts in the 3T3-independent cell lines (R59, R63b). 3T3 fibroblasts did not influence the cornification of R66a and this cell line remained anchorage dependent throughout the study. The results suggest that in malignant cell lines characterised by being independent of 3T3 fibroblasts (R63b, R59) the CFE was inversely correlated to the degree of cornification. However, in the malignant cell line showing a greater dependence on support (R63a) the relationship between CFE and cornification was unclear because these parameters may have been modulated by the presence of 3T3 fibroblasts. The cell line from dysplastic non-invasive tissue (R66a) differed from its malignant counterparts in the fact that CFE and cornification were unaffected by 3T3 fibroblasts despite previous studies showing a dependence on mesenchymal support.     

Assessing the tumorigenic phenotype of VERO cells in adult and newborn nude mice.

VERO cell lines are important substrates for viral vaccine manufacture. The mechanism by which these cells became neoplastically transformed is unknown. During tissue-culture passage, VERO cells can develop the capacity to form tumors. Although at the passage levels (around p140) currently used for vaccine manufacture, VERO cells are non-tumorigenic, questions have been raised about safety issues that might be associated with this capacity to acquire a tumorigenic phenotype. To begin to address these issues, the tumorigenicity of VERO cell lines, derived at different passage levels under different growth conditions, were evaluated in 365-day assays in adult and newborn nude mice. High passage (p>200) VERO cell lines established by random passaging in tissue culture produced tumors in adult (10 out of 27) mice and newborn (21 out of 30) mice, respectively. In contrast, a high passage (p>250) cell line established by passage at sub-confluence produced tumors only in newborn mice (16 out of 30). Progressively growing tumors began forming at 36 days in newborns and at 69 days in adults. Higher tumor incidences and shorter tumor latencies suggest that newborn nude mice may be more sensitive than adults in detecting the expression of a tumorigenic phenotype by some VERO cell lines.     

Transfer of anchorage independence by isolated metaphase chromosomes in hamster cells

The cellular property of being able to grow on agar (aga⁺) or to show anchorage independence has been transferred by means of metaphase chromosomes from CHO cells to BHK and other permanent transformed hamster lines unable to grow on agar. As with other genetic markers, the transferents are unstable when grown under non-selective conditions. The aga⁺ transferents are tumorigenic, providing further evidence for the association between the ability to grow in agar or anchorage independence and tumorigenicity. Evidence has been obtained in these experiments for the existence of at least two discrete events in the transformation of normal into tumorigenic cells. The ability to transfer and select for the aga⁺ marker in recipient cells indicates that tumorigenicity behaves dominantly phenotypically.     

Correlation of tumorigenicity with resistance to growth inhibition by cis-hydroxyproline

cis-Hydroxyproline, an inhibitor of collagen deposition, was examined for its effect on the growth of a variety of tumorigenic and non-tumorigenic cells. Virally, chemically, and spontaneously transformed murine cell lines were found to be less sensitive to cell spreading and growth inhibition by cis-hydroxyproline (CHP) than were their non-tumorigenic counterparts. The non-tumorigenic lines exhibited a higher rate of collagen accumulation in culture than the tumorigenic cell lines. The rate of collagen accumulation in culture without CHP and the growth inhibition by CHP were directly related. These results suggest that normal but not tumorigenic cells may require synthesis of an extracellular substrate containing collagen to support spreading and growth.     

Tumorigenicity of Vero cells

One of the current criteria for evaluating the acceptability of cell lines for use in vaccine production is lack of tumorigenicity. Vero cells represent an example of a class of cells known as continuous cell lines. They were derived from African green monkey kidney, and their growth properties and culture characteristics have many advantages over other cell substrates for use in vaccine production. We have tested Vero cells for tumorigenicity in nude mice and in a human muscle organ culture system, and found a significant increase in their tumorigenic potential with increasing passage numbers. Cells at passage 232 and higher produced nodules in all nude mice inoculated. Histologically the nodules were well defined, anaplastic tumors, which exhibited some of the characteristics of renal adenocarcinomas. In about 6 to 8 days all of the nodules began to regress. Data were obtained that suggested an immune mechanism was the basis for the regression phenomenon.     

Evaluation of the chemopreventive potential of retinoids using a novel in vitro human prostate carcinogenesis model

The prevalence of prostatic intraepithelial neoplasia (PIN) and latent prostatic carcinoma, representing multiple steps in carcinogenesis and progression to invasive carcinoma, makes them relevant targets for prevention. A unique family of human prostate epithelial cell lines, which mimic steps in prostate carcinogenesis and progression, were used to evaluate the chemopreventive potential of all-trans-retinoic acid (RA) and N-(4-hydroxyphenyl)retinamide (4-HPR). The effects of RA and 4-HPR on anchorage-dependent growth of an immortalized, non-tumorigenic cell line RWPE-1 and two tumorigenic cell lines, WPE1-NB14 and WPE1-NB11, derived from RWPE-1 by exposure to N-methyl-N-nitrosourea (MNU), were examined. Both tumorigenic cell lines grow more rapidly than the parent RWPE-1 cell line in monolayer culture. Further, while RWPE-1 cells do not form colonies in agar, both tumorigenic cell lines do, with a colony forming efficiency (CFE) of 1.85 and 2.04% for WPE1-NB14 and WPE1-NB11 cells, respectively. Both RA and 4-HPR inhibited anchorage-dependent growth of all cell lines and anchorage-independent growth of WPE1-NB14 and WPE1-NB11 cells, in a dose-dependent manner, however, 10 times more RA than 4-HPR was required to produce the same effect. RWPE-1 cells are not invasive but WPE1-NB11 cells are significantly more invasive than WPE1-NB14 cells. Both RA and 4-HPR inhibited invasion in vitro by WPE1-NB11 and WPE1-NB14 cells where the more malignant WPE1-NB11 cells showed greater inhibition of invasion by 4-HPR than by RA. Overall, 4-HPR was more effective than RA in inhibiting growth and invasion but the response varied amongst the cell lines. These three cell lines mimic progressive steps in carcinogenesis and progression, from immortalized, non-tumorigenic RWPE-1 cells, to the less malignant WPE1-NB14 to the more malignant WPE1-NB11 cells, and provide powerful models for studies on secondary and tertiary prevention, i.e. promotion and progression stages, respectively, of prostate cancer.     

Genetic analysis of tumorigenesis: I. Expression of tumor-forming ability in hamster hybrid cell lines

Two new fibroblasts cell lines from a male Chinese hamster embryo are described: one tumorigenic in nude mice (CHEF 16-2) and the other not (CHEF 18-1). Both lines have a stable diploid mode and mean of 22 with 10 pairs of homologous autosomes. The cell lines differ unambiguously in cloning ability in methylcellulose, colony morphology, and tumorigenicity; the expression of these traits was examined in a set of 18-1 X 16-2 hybrid clones. The results show initial suppression of tumorigenicity and anchorage independence, followed by growth in the nude mouse of subsets of cells with chromosomes reduced to the diploid range. Further studies are in progress to establish whether ability to grow in methylcellulose and in the nude mouse segregate coordinately at the cellular level, and whether this segregation has an identifiable chromosomal basis.     

Expression of transformation characters in colchicine-resistant tumor cells

Malignancy and anchorage independence of Djungarian hamster tumor cell lines resistant to different doses (0.1-5.0 micrograms/ml) of colchicine were studied. The clones with low colchicine resistance (15-20-fold) did not differ in tumorigenicity from parental cells. The TD50 for highly colchicine-resistant cells (200-800-fold) was several orders of magnitude higher than that for wild-type cells. Colchicine resistance did not affect the expression of the cells anchorage independence. The cloning efficiency in a semi-solid medium was the same both for colchicine-resistant cell lines and wild-type cells.     

Epithelio-mesenchymal transition in a neoplastic ovarian epithelial hybrid cell line

A hybrid cell line, IOSE-Ov29, was created through fusion of cells from the human ovarian adenocarcinoma line OVCAR3 and the non-tumorigenic SV40 Tag-transfected human ovarian surface epithelial line IOSE-29. OVCAR3 cells exhibit a differentiated epithelial phenotype, whereas line IOSE-29 expresses mesenchymal characteristics that were acquired in culture by epithelio-mesenchymal transition. Microsatellite analysis, comparative genomic hybridization (CGH), and MFISH showed the genotype of the IOSE-Ov29 cells to contain components of both parent cell lines, but to be predominantly OVCAR3 derived. IOSE-Ov29 resembled OVCAR3 and differed from IOSE-29 as shown by its unlimited life span, tumorigenicity, epithelial morphology, keratin, occludin, E-cadherin and CA125 expression, increased expression of kinases of the PI3K pathway, and loss of cGMP-dependent protein kinase expression. IOSE-29-derived properties included SV40 Tag expression, growth inhibition by activin, collagen type III secretion, increased adhesion and spreading on tissue culture plastic, and increased growth rate. Proliferation of all three lines was stimulated by FSH and ATP and inhibited by GnRH I and GnRH II. Interestingly, IOSE-Ov29 was more anchorage independent than either parent line and was the only line that invaded Matrigel in Boyden chambers and formed invasive branches in collagen gels. The results indicate that IOSE-Ov29 is an IOSE-29/OVCAR3 hybrid, which differs from both parent lines genetically and phenotypically. Unexpectedly, fusion with the non-tumorigenic IOSE-29 cells enhanced malignancy-associated characteristics of OVCAR3, presumably as a result of the expression of IOSE-29-derived mesenchymal properties that are usually acquired by carcinoma cells through epithelio-mesenchymal transition during metastatic progression.     

Potential neoplastic evolution of Vero cells: in vivo and in vitro characterization

Vero cell lines are extensively employed in viral vaccine manufacturing. Similarly to all established cells, mutations can occur during Vero cells in vitro amplification which can result in adverse features compromising their biological safety. To evaluate the potential neoplastic evolution of these cells, in vitro transformation test, gene expression analysis and karyotyping were compared among low- (127 and 139 passages) and high-passage (passage 194) cell lines, as well as transformed colonies (TCs). In vivo tumorigenicity was also tested to confirm preliminary in vitro data obtained for low passage lines and TCs. Moreover, Vero cells cultivated in foetal bovine serum-free medium and derived from TCs were analysed to investigate the influence of cultivation methods on tumorigenic evolution. Low-passage Vero developed TCs in soft agar, without showing any tumorigenic evolution when inoculated in the animal model. Karyotyping showed a hypo-diploid modal chromosome number and rearrangements with no difference among Vero cell line passages and TCs. These abnormalities were reported also in serum-free cultivated Vero. Gene expression revealed that high-passage Vero cells had several under-expressed and a few over-expressed genes compared to low-passage ones. Gene ontology revealed no significant enrichment of pathways related to oncogenic risk. These findings suggest that in vitro high passage, and not culture conditions, induces Vero transformation correlated to karyotype and gene expression alterations. These data, together with previous investigations reporting tumour induction in high-passage Vero cells, suggest the use of low-passage Vero cells or cell lines other than Vero to increase the safety of vaccine manufacturing.     

Genetic analysis of tumorigenesis. VIII. Suppression of SV40 transformation in cell hybrids and cytoplasmic transferants

Intraspecies somatic cell hybrids of BALB/c mouse 3T3 and SV40-transformed embryonic fibroblast (SVT2) cells were analyzed for transformation-associated properties and their tumorigenic potential in nude mice. In confirmation of our earlier findings, hybrids expressing the viral T-antigen were not suppressed for the ability to clone in medium with 1% serum. In contrast, division rate in medium with 1% or 10% serum, anchorage independence, cytochalasin-sensitive growth control, and tumorigenicity were suppressed noncoordinately, and the extent of suppression varied from one hybrid to another. Suppression was not simply determined by the increased chromosome content of the hybrid cells, nor was suppression correlated with rearrangements of the integrated viral sequence (SAGER et al., 1981a, b). Similar results were found in cytoplasmic transferants expressing T-antigen. Four independent transferants and subclones derived from them varied in the extent of suppression of anchorage independence and tumorigenicity. In both hybrids and transferants, a low serum requirement for clonal growth apparently was determined solely by expression of SV40 T-antigen, but other transformation properties, as well as tumorigenicity, appeared to require multiple changes in the cellular genome for their expression. These changes must occur during or after viral integration, since they are not expressed in uninfected 3T3 cells.     

Cell properties after repeated transplantation of spontaneously and of SV40 virus transformed mouse cell lines. I. Growth in culture.

"Spontaneously" or SV40 virus transformed AL/N mouse cell lines were passed repeatedly through syngeneic mice. Cell lines were re-established in culture from minced pieces of tumors in the presence of concentrated fetal calf serum or from tumor cells dispersed by trypsin. The aim of this study was to compare the two cell lines in regard to the selection processes which operate during such procedures by characterization of the resulting cell lines. Measurements of growth in tissue culture on substratum showed no significant difference between any of the transformed cell lines. The SV40 transformed cells and its derivative cells had a low anchorage requirement for growth. The greatest anchorage requirement for growth was in the normal untransformed cells and in the derivative cells from the "spontaneously" transformed cells which were established from minced tumors. The spontaneously transformed cells and all derivative cells had high tumorigenicity (TD50 is less than 10-2). The SV40 transformed cells had no observable tumorigenicity (TD50 is greater than 10-8), except when injected into irradiated mice (TD50 = 1-5 X 10-5 in the immunocompetent mice, 5 X 10-4 in the irradiated mice). The SV40 transformed derivative cells maintained their SV40 specific T antigen and their susceptibility to lysis by specific antiserum.     

Conversion of premalignant human cells to tumorigenic cells by methylmethane sulfonate and methylnitronitrosoguanidine

Nine human tumor cell lines derived from both epithelial and mesenchymal tumors exhibited either an anchorage-independent growth non-tumorigenic phenotype or an anchorage-independent tumorigenic phenotype. Transformed epithelial cell lines with the non-tumorigenic phenotype could be converted to a progressively growing tumor phenotype following treatment with either methylmethane sulfonate (MMS) or N-methyl-N-nitro-N-nitrosoguanidine (MNNG). In contrast, sarcoma derived cell lines with a non-tumorigenic phenotype could be converted to a progressively growing tumor phenotype only with MNNG. SV40 immortalized HET-1A non-tumorigenic phenotype cells could be converted to a progressively growing tumorigenic phenotype, infrequently, when treated with MNNG, but not MMS. Progressively growing tumors produced by either MMS or MNNG treated non-tumorigenic phenotypes exhibited metastatic potential in nude mice. Chemically treated HET-1A cells acquired the ability to produce tumor in mice but the tumor did not exhibit metastatic potential. In contrast, populations of tumorigenic cells were not rendered more biologically aggressive after treatment with either MMS or MNNG; i.e., the latency period for tumor development was not accelerated and the tumors did not exhibit metastatic potential. These results suggest that the biological effects of MMS and MNNG on non-tumorigenic, tumorigenic and immortalized cell lines are phenotype specific.Abbreviations AIG anchorage-independent growth - DMSO dimethyl sulfoxide - FBS fetal bovine serum - GM growth medium - MEM Eagle's minimum essential medium - MMS methylmethane sulfonate - MNNG N-Methyl-N-Nitro-N-Nitrosoguanidine - PDL population doubling - SCC squamous cell carcinoma     

Tumorigenic transformation of murine keratinocytes by the E5 genes of bovine papillomavirus type 1 and human papillomavirus type 16.

To examine the biological properties of the bovine papillomavirus type 1 (BPV) and human papillomavirus type 16 (HPV16) E5 genes, each was cloned separately into a retroviral expression vector and helper-free recombinant viruses were generated in packaging cell lines. The BPV E5 retroviruses efficiently caused morphologic and tumorigenic transformation of cultured lines of murine fibroblasts, whereas the HPV16 E5 viruses were inactive in these assays. In contrast, infection of the p117 established line of murine epidermal keratinocytes with either the BPV or the HPV16 E5 retrovirus resulted in the generation of tumorigenic cells. Pam212 murine keratinocytes were also transformed to tumorigenicity by the HPV16 E5 gene but not by the gene carrying a frameshift mutation. These results establish that the HPV16 E5 gene is a transforming gene in cells related to its normal host epithelial cells.     

Possible role of genetic cell variants in the viral induction of mouse mammary tumors

Out of three attempts to induce neoplasia in normal C57Bl mammary epithelial cells with the mouse mammary tumor virus (MuMTV) only one presented signs of tumorigenicity. Immunofluorescence showed that virus synthesis took place in all three sublines but tumorigenicity as detected by cell aggregation viability (CAV) and transplantation into syngeneic mice failed to occur in two of them. By comparison, cells from a BALB/c spontaneous mammary tumor that do not express MuMTV were 100% tumorigenic, whereas cells from a BALB/cfC3H tumor with a 95% virus-producing cell population had a normal CAV and were tumorigenic only in 60% of the test animals. This lack of correlation suggested that many of the virus-producing cell were not neoplastic and that neoplasia might occur under virus stimulation only if a restricted population of genetic cell variants existed. Accelerated tissue culture passages of virus-free C57Bl and BALB/c normal mammary cells resulted in their spontaneous neoplasia at Passages 23 and 50, respectively; when duplicated cells cryopreserved in early passages were revived and cultivated in the same manner, neoplasia occurred at Passages 27 and 58. The similarity of the passage numbers appears to confirm the existence of genetic cell variants among the normal cell population.     

Cellular tumorigenicity in nude mice: correlation with cell growth in semi-solid medium

Cultured cells derived from either normal or malignant tissues of several species have been tested by injection into the immune-deficient nude mouse in order to determine the cellular properties which are associated with tumorigenicity in vivo. Results show that one in vitro property consistently correlated with neoplastic growth in nude mice is the ability of the cell to form spherical colonies in a semi-solid growth medium such as methyl cellulose suspension. Cellular tumorigenicity is not determined solely by the malignancy of the tissue of origin, since cells derived from nonmalignant tissues become tumorigenic when they are no longer anchorage dependent for growth. In addition, acquisition of infinite growth potential in heteropioid cell lines is not in itself sufficient to confer tumorigenic capacity on the cells. These results suggest that the degree of cell growth in methyl cellulose is a useful parameter in vitro for predicting tumorigenicity in the animal, and also demonstrate the potential usefulness of the nude mouse for analysis of cellular malignancy irrespective of the tissue or species of origin.     

Enhance anchorage independence and tumorigenicity of aneuploid Chinese hamster cells with nearly doubled chromosome complements.

The role of a cell's chromosome complement in its tumorigenic and anchorage-independent growth properties in vitro was investigated by injecting a Chinese hamster cell line and its subclones into immunodeficient nude mice and by plating the cells in a semisolid medium containing methylcellulose. The parental WOR-6 cell clone originally consisted of 89% 1s cells and 11% cells with a nearly double (2s) complement. Tumors that developed from WOR-6 were found to consis entirely or primarily of cells with near 2s chromosome complements. Subclones of WOR-6 that contained only 1s cells rarely produced tumors in nude mice, even at high inoculum doses, whereas clones containing a high fraction of 2s cells were consistently tumorigenition, serial passage of WOR-6 cells in semisolid medium resulted in selective enrichment for near 2s cells and, concomitantly, greatly enhanced tumorigenicity. Analyses of G-banded chromosomes revealed that the 1s cells of the WOR-6 parental clone, which has a modal chromosome number of 21 and a range of 18 to 23, is completely or partially monosomic for some chromosomes and trisomic for others. The 2s cells, selected both in vivo through growth as tumors in nude mice and in vitro in semisolid medium, appeared to have resulted from preferential duplication of certain chromosomes of the 1s cells. Our results therefore suggest that cells which develop multiple copies of selected genes, while remaining functionally hemizygous for other loci, acquire an enhanced anchorage-independent growth potential in vitro and increased tumorigenicity. This conclusion is consistent with the observation that cellular tumorigenicity is correlated with anchorage independence (Rreedman and Shin, 1974) and leads support to OHNO'S (1974) suggesting that aneuploidy is a possible means employed by cells to express recessive phenotypes and increase their tumorigenicity.     

Investigations of the mechanism of decreased tumorigenicity of cells grown in BrdU

Transplantable SV40-transformed hamster cells cultivated in the presence of low concentrations of BrdU for prolonged periods of time and cells made deficient in the enzyme thymidine kinase (dTK) by continued exposure to BrdU became less tumorigenic. In both instances, when grown in BrdU the cells contained analog substituted DNA. The tumorigenicity of dTK⁺ cells exposed to low concentrations of BrdU, but not the dTK^? cells, returned to control values when the cells were grown in medium devoid of BrdU. A tumorigenic mouse cell line made dTK deficient also had diminished oncogenicity. However, transformed hamster cells made deficient in another salvage pathway enzyme, hypoxanthineguanine phosphoribosyl-transferase by growth in eight azaguanine, retained their tumorigenicity. Two of five revertant cell lines, in which thymidine kinase activity was restored, transplanted more readily to hamsters than the dTK^? cells from which they were derived. It is concluded that there is a relative loss of tumorigenicity when BrdU is incorporated into the DNA of tumorigenic cell lines, or when there is a genetic modification of thymidine kinase activity.     

Increased incidence of CAD gene amplification in tumorigenic rat lines as an indicator of genomic instability of neoplastic cells

It has been hypothesized that genomic instability is an important component of tumorigenesis. In an attempt to establish this relationship, we determined the frequencies with which two nontumorigenic and four tumorigenic rat liver epithelial cell lines underwent a particular type of genetic instability, gene amplification. By exposing cells to N-(phosphonoacetyl)-L-aspartate (PALA), a drug which specifically inhibits the aspartate transcarbamylase activity of the multifunctional CAD enzyme and selects for amplification of the CAD gene, we observed a striking parallel between the ability of these cell lines to become resistant to this drug and the ability of these same cells to form tumors after injection into day-old syngeneic rats. Cells of one highly tumorigenic line became resistant to PALA greater than 70 times more often than those of a non-tumorigenic line. Molecular analyses of eight independent PALA-resistant subclones confirmed that, in each case, this resistance was due to amplification of the CAD gene. Thus, our results demonstrate the relationship between tumorigenicity and at least one measure of genomic instability, CAD gene amplification. The method developed in this study provides a quantitative, rapid indicator of tumorigenicity and should prove useful in trying to elucidate the underlying basis of genomic instability in neoplastic cells.