Amplified expression of the HER2/ERBB2 oncogene induces resistance to tumor necrosis factor alpha in NIH 3T3 cells.

Functional characterization of oncogene products that induce cellular transformation has progressed rapidly in recent years. However, less is known about the mechanism(s) by which the transformed cells may escape destruction by host immune defenses and form tumors. A recently described oncogene that has an important association with aggressive human breast carcinoma is "HER2," for human epidermal growth factor receptor 2. The oncogene has also been called NGL and human c-erbB-2 (ERBB2). In this paper we show that amplification of HER2 oncogene expression can induce resistance of NIH 3T3 cells to the cytotoxic effects of recombinant tumor necrosis factor alpha (rTNF-alpha) or macrophages. Resistance is accompanied by an increased dissociation constant for rTNF-alpha binding to high-affinity receptors on the HER2-transformed NIH 3T3 cells. The resistance phenotype is independent of transformation since NIH 3T3 cells transformed by the activated human homologue of the Harvey-ras oncogene (HRAS) retain high-affinity binding sites for rTNF-alpha as well as sensitivity to its cytotoxic effects. These results suggest that HER2 may potentiate tumorigenesis by inducing tumor cell resistance to host defense mechanisms.     

Oncogenes in human tumor cell lines: molecular cloning of a transforming gene from human bladder carcinoma cells.

The presence of dominant transforming genes in human tumor cell lines has been investigated. High molecular weight DNAs isolated from cell lines established from carcinomas and sarcomas of various organs as well as from a glioblastoma and two melanomas were utilized to transfect NIH/3T3 mouse fibroblasts. The DNAs of T24 and A2182, two cell lines derived from a bladder and a lung carcinoma, respectively, and of HT-1080, a cell line established from a fibrosarcoma, were able to transform recipient NIH/3T3 cells. First-cycle transformants exhibited anchorage-independent growth and were tumorigenic in athymic and immunocompetent mice. Moreover, they contained human DNA sequences and were able to transmit their malignant phenotype in additional cycles of transfection. Southern blot analysis of T24-derived transformants showed that a single fragment of human DNA specifically cosegregated with the malignant phenotype, suggesting that it contained the T24 oncogene. Therefore, these human sequences were molecularly cloned with lambda Charon 9A as the cloning vector. The resulting recombinant DNA molecule, designated lambda T24-15A, was shown to contain a 15-kilobase-pair EcoRI insert of human cellular DNA. lambda T24-15A DNA (either intact or EcoRI digested) transformed NIH/3T3 fibroblasts with a specific activity of 20,000 focus-forming units per pmol of cloned DNA. Our results indicate that we have molecularly cloned a biologically active oncogene present in T24 human bladder carcinoma cells.     

Transformation of human cells by DNAs ineffective in transformation of NIH 3T3 cells.

Neonatal human foreskin fibroblasts can be transformed to anchorage-independent growth by transfection with DNAs inefficient in transforming NIH 3T3 cells. Human cells transfected with DNA from GM 1312, a multiple myeloma cell line, or MOLT-4, a permanent lymphoblast line, grow without anchorage at a much higher frequency than do the parental cells and their DNAs can transform human cell recipients to anchorage-independent growth; they have extended but not indefinite life spans and are nontumorigenic. Human fibroblasts are also transformed by DNAs from two multiple myeloma lines that also transform 3T3 cells; however, restriction analysis suggests that different transforming genes in this DNA are acting in the human and murine systems. These results indicate that the human cell transfection system allows detection of transforming genes not effective in the 3T3 system and points out the possibility of detection of additional transforming sequences even in DNAs that do transform murine cells.     

Expression of the metastatic phenotype in cells transfected with human metastatic tumor DNA.

NIH 3T3 mouse fibroblasts form nonmetastasizing fibrosarcomas upon transformation by the Ha-ras oncogene isolated from the EJ human bladder carcinoma cell line and subcutaneous inoculation into immunocompetent NFS/NCr mice. DNA from a human metastatic tumor was transfected into these Ha-ras transformants, and one of the resulting colonies yielded a lung metastasis after subcutaneous inoculation. DNA was isolated from this metastasis and subjected to a second round of transfer into Ha-ras-transformed NIH 3T3 cells. Inoculation of these transfected cultures into mice led once again to formation of metastases, this time at a higher frequency. Examination of four of the resulting metastases revealed discrete human DNA fragments that were common to all four. These findings demonstrate that the metastatic phenotype can be transferred via DNA from cell to cell and is associated with the presence of a discrete DNA segment. This segment is not identical to the myc oncogene or to any of the frequently detected ras tumor oncogenes.     

Oncogene activation in human benign tumors of the skin (keratoacanthomas): is HRAS involved in differentiation as well as proliferation?

In vitro DNA amplification followed by oligonucleotide mismatch hybridization was used to study the frequency of HRAS mutations in the benign self-regressing skin tumors keratoacanthomas and in squamous cell carcinomas. We used freshly obtained keratoacanthomas as well as Formalin-fixed paraffin-embedded tissues from both types of tumors. DNA from 50 samples of each tumor type was analyzed for activating mutations involving codons 12 and 61. A relatively high percentage (30%) of HRAS mutations was found in the keratoacanthomas compared with 13% in the squamous cell carcinomas. The most frequent mutation identified is the A-T-to-T.A transversion in the second position of codon 61. The present findings demonstrate the involvement of the HRAS oncogene in human benign tumors. Moreover, they indicate that an activated HRAS oncogene is not sufficient to maintain a neoplastic phenotype and argue against a role of HRAS in the progression of skin tumorigenesis.     

Transformation of rodent immortalized and embryo cells by oncogenes. I. Mouse (NIH 3T3) and rat (FR 3T3) immortalized fibroblasts

Immortalized mouse NIH 3T3 cells were transformed by gene transfer of DNA isolated from a human bladder tumor cell line and plasmids containing an activated human Ha-ras oncogene insert. For gene transfer the calcium-phosphate co-precipitation method was used. Transformation was evaluated by morphological focus formation, growth in soft agar and tumor development in nude mice. In addition, immortalized rat FR 3T3 cells were transformed by Ha-ras, too. The co-transfer of ras and myc oncogenes did not enhance focus formation in FR 3T3 cells.     

A position 12-activated H-ras oncogene in all HS578T mammary carcinosarcoma cells but not normal mammary cells of the same patient.

Among 21 human mammary tumors analyzed for transforming genes by transfection of NIH/3T3 cells, only DNA of a carcinosarcoma cell line, HS578T, registered as positive. A Harvey (H)-ras oncogene identified in this line was cloned in biologically active form and the activating lesion was identified as a single nucleotide substitution of adenine for guanine in the 12th codon. This results in substitution of aspartic acid for glycine at this position of the p21 coding sequence. Knowledge that this alteration creates a restriction site polymorphism for Msp I/Hpa II in the H-ras protooncogene made it possible to survey for the presence of the activated H-ras allele in normal cells as well as in clonally derived tumor cell lines of the same patient. We demonstrated the presence of unaltered H-ras alleles in normal HS578 cells. In contrast, every clonally derived HS578T tumor cell line analyzed contained the H-ras oncogene possessing the genetic alteration at position 12. These findings establish that activation of this oncogene was the result of a somatic event selected within all HS578T tumor cells.     

Transfection of BLV containing DNA into NIH 3T3 cells

Cell DNA isolated from bovine leukosis virus (BLV) productive cell clones was transfected into the NIH3T3 cells. DNA from some cell clones was able to transform NIH3T3 cells. The transformed cells were cloned, and in 4 cell clones out of 33 bovine leukosis virus specific sequences were detected by hybridization with labeled BLV probe. According to the restriction analysis the BLV sequences were incomplete, they were rearranged, deleted, or both. The DNA from NIH3T3 transformants with BLV sequences was able to transform in the second round transfection experiments NIH3T3 cells again, but in these transformants BLV specific sequences were not detected. Cell DNA from sheep tumors induced by BLV was able to transform the NIH3T3 cells too, but BLV specific sequences were not present in the transformants. It appears that BLV specific sequences are not required for NIH3T3 cell transformation.     

Mutations activating human c-Ha-ras1 protooncogene (HRAS1) induced by chemical carcinogens and depurination.

In vitro modification of plasmids containing the human c-Ha-ras1 protooncogene (HRAS1) with the ultimate carcinogens N-acetoxy-2-acetylaminofluorene and r-7, t-8-dihydroxy-t-9, 10-epoxy-7,8,9,10-tetrahydrobenzo[alpha]pyrene (anti-BPDE) generated a transforming oncogene when the modified DNA was transfected into NIH 3T3 cells. The protooncogene was also activated by heating the plasmid at 70 degrees C, pH 4, to generate apurinic/apyrimidinic sites in the DNA. DNA isolated from transformed foci was analyzed by hybridization with 20-mer oligonucleotides designed to detect single point mutations within two regions of the gene commonly found to be mutated in tumor DNA. Of 23 transformants studied, 7 contained a mutation in the region of the 12th codon, whereas the remaining 16 were mutated in the 61st codon. Of the codon-61 mutants, 6 were mutated at the first base position (C X G), 5 at the second (A X T), and 5 at the third (G X C). The point mutations induced by anti-BPDE were predominantly G X C----T X A and A X T----T X A base substitutions, whereas four N-acetoxy-2-acetylaminofluorene-induced mutations were all G X C----T X A, and a single depurination-induced activation that was analyzed contained an A X T----T X A transversion. Together, these methods provide a useful means of determining point mutations produced by DNA-damaging agents in mammalian cells.     

Analysis of RAS gene mutations in acute myeloid leukemia by polymerase chain reaction and oligonucleotide probes.

In vitro DNA amplification followed by oligonucleotide dot blot analysis were used to study RAS gene mutations in acute myeloid leukemia (AML). Fifty-two presentation AML DNAs were screened for mutations in codons 12, 13, and 61 of NRAS and in codons 12 and 61 of KRAS and HRAS. Fourteen (27%) contained mutations--all in NRAS and predominantly in codon 12. The most common amino acid substitution identified was of glycine by aspartic acid at codon 12 (7/18), with a G----A transition being the most common base change (11/18). No particular correlation was observed between disease subtype and the incidence or type of NRAS mutation. In DNA samples from four patients, 2 NRAS mutations were found to coexist. NIH 3T3 focus-formation assays revealed that in each case the mutations were present in different NRAS alleles. We also report the absence of a mutated RAS gene in relapse DNAs of four patients in which a RAS oncogene had been detected at presentation. These observations suggest that RAS mutations arise as part of the evolution of neoplastic transformation.     

The human transforming growth factor type alpha coding sequence is not a direct-acting oncogene when overexpressed in NIH 3T3 cells.

A peptide secreted by some tumor cells in vitro imparts anchorage-independent growth to normal rat kidney (NRK) cells and has been termed transforming growth factor type alpha (TGF-alpha). To directly investigate the transforming properties of this factor, the human sequence coding for TGF-alpha was placed under the control of either a metallothionein promoter or a retroviral long terminal repeat. These constructs failed to induce morphological transformation upon transfection of NIH 3T3 cells, whereas viral oncogenes encoding a truncated form of its cognate receptor, the EGF receptor, or another growth factor, sis/platelet-derived growth factor 2, efficiently induced transformed foci. When NIH 3T3 clonal sublines were selected by transfection of TGF-alpha expression vectors in the presence of a dominant selectable marker, they were shown to secrete large amounts of TGF-alpha into the medium, to have downregulated EGF receptors, and to be inhibited in growth by TGF-alpha monoclonal antibody. These results indicated that secreted TGF-alpha interacts with its receptor at a cell surface location. Single cell-derived TGF-alpha-expressing sublines grew to high saturation density in culture. However, when plated as single cells on contact-inhibited monolayers of NIH 3T3 cells, they failed to form colonies, whereas v-sis- and v-erbB-transfected cells formed transformed colonies under the same conditions. Moreover, TGF-alpha-expressing sublines were not tumorigenic in nude mice. These and other results imply that TGF-alpha exerts a growth-promoting effect on the entire NIH 3T3 cell population after secretion into the medium but little, if any, effect on the individual cell synthesizing this factor. It is concluded that the normal coding sequence for TGF-alpha is not a direct-acting oncogene when overexpressed in NIH 3T3 cells.     

Hamster cell line suitable for transfection assay of transforming genes.

We established a subclone, SHOK, from the GHE-L cell line, an immortal line derived from a primary culture of Syrian hamster embryo cells, as a recipient cell line useful for the detection of oncogenes by transfection. SHOK cells were almost as susceptible as NIH 3T3 cells to focus formation by many oncogenes, including v-raf, v-Ha-ras, v-Ki-ras, or activated c-Ha-ras. The susceptibility of SHOK to focus formation was higher than that of NIH 3T3 for v-mos but was lower for v-fps, v-fgr, v-src, v-sis, and v-abl. When DNAs extracted from 27 human and murine tumors were tested for focus formation, 5 DNAs were positive in NIH 3T3 cells, whereas 9 were positive in SHOK cells at the primary transfection. Using SHOK cells as recipients of tumor cellular DNA, we isolated another oncogene and a c-Ki-ras2 gene mutated at codon 146 that were difficult to detect in NIH 3T3 cells. SHOK cells have a low rate of spontaneous transformation, produce easily distinguishable foci, and maintain a stable karyotype in transformed cells. In addition to being useful for the screening of human tumor DNAs, SHOK cells will be useful for the isolation of oncogenes from murine tumors because of their hamster origin.     

Activated protooncogenes in human lung tumors from smokers.

Fourteen primary human lung tumor DNAs from smokers were analyzed for transforming activity by two DNA transfection assays. Activated protooncogenes were detected in 3 of 11 tumor DNAs by the NIH 3T3 focus assay, whereas activated protooncogenes were detected in 11 of 13 tumor DNAs by the NIH 3T3 cotransfection-nude mouse tumorigenicity assay. K- or NRAS genes activated by point mutation at codons 12 or 61 were detected in a large cell carcinoma, a squamous cell carcinoma, and 5 adenocarcinomas. An HRAS oncogene activated by a different mechanism was detected in an epidermoid carcinoma. One adenocarcinoma was found to contain an activated RAF gene. Two unidentified transforming genes were detected in a squamous cell carcinoma DNA and two adenocarcinoma DNAs. Eight of 10 lung adenocarcinomas that had formed metastases at the time of surgery were found to contain RAS oncogenes. No significant increase in metastasis was observed in the lung adenocarcinomas that contained one or more 6-kilobase EcoRI alleles of the LMYC gene. Overall, 12 of 14 (86%) of the lung tumor DNAs from smokers were found to contain activated protooncogenes. RAS oncogenes appear to play a role in the development of metastases in lung adenocarcinomas.     

Activated neu oncogene sequences in primary tumors of the peripheral nervous system induced in rats by transplacental exposure to ethylnitrosourea.

Neurogenic tumors were selectively induced in high incidence in F344 rats by a single transplacental exposure to the direct-acting alkylating agent N-ethyl-N-nitrosourea (EtNU). We prepared DNA for transfection of NIH 3T3 cells from primary glial tumors of the brain and from schwannomas of the cranial and spinal nerves that developed in the transplacentally exposed offspring between 20 and 40 weeks after birth. DNA preparations from 6 of 13 schwannomas, but not from normal liver, kidney, or intestine of tumor-bearing rats, transformed NIH 3T3 cells. NIH 3T3 clones transformed by schwannoma DNA contained rat repetitive DNA sequences, and all isolates contained rat neu oncogene sequences. One schwannoma yielded a transformant with rat-specific sequences for both neu and N-ras. A point mutation in the transmembrane region of the putative protein product of neu was identified in all six transformants and in the primary tumors from which they were derived as well as in 5 of 6 schwannomas tested that did not transform NIH 3T3 cells. Of 59 gliomas, only one yielded transforming DNA, and an activated N-ras oncogene was identified. The normal cellular neu sequence for the transmembrane region, but not the mutated sequence, was identified in DNA from all 11 gliomas surveyed by oligonucleotide hybridization. Activation of the neu oncogene, originally identified [Schechter, A.L., Stern, D.F., Vaidyanathan, L., Decker, S.J., Drebin, J.A., Greene, M.I. & Weinberg, R.A. (1984) Nature (London) 312, 513-516] in cultured cell lines derived from EtNU-induced neurogenic tumors that by biochemical but not histologic criteria were thought to originate in the central nervous system in BD-IX rats, appears specifically associated with tumors of the peripheral nervous system in the F344 inbred strain.     

Development of a highly efficient expression cDNA cloning system: application to oncogene isolation.

We developed an expression cDNA cloning system capable of generating high-complexity libraries with unidirectionally inserted cDNA fragments and allowing efficient plasmid rescue. As an application of this system, a cDNA library was constructed from an NIH 3T3 transformant induced by mouse hepatocellular carcinoma DNA. Transfection of NIH 3T3 cells by the library DNA led to the detection of several transformed foci from which identical plasmids with transforming ability could be rescued. Structure and sequence analysis of the cDNA clones revealed that the oncogene was created by recombinational events involving an unknown gene and the mouse homologue of the B-raf protooncogene. Detection of the same genetic rearrangement in independent primary transformants implied that generation of the oncogene occurred within the tumor rather than during DNA transfection or cDNA library construction. The high frequency at which clones were identified and the large sizes of some of the transforming cDNA inserts isolated suggest wide applicability of this mammalian expression cloning system for isolating cDNAs of biologic interest.     

High frequency of N-ras activation in acute myelogenous leukemia

Using the NIH/3T3 cell transfection assay, activated cellular oncogenes have been detected in around 10% to 20% of human tumors. From a series of DNA preparations from tissues infiltrated with acute myelogenous leukemia (AML), 50% (3/6) caused transformation of NIH/3T3 cells. Thus AML appears to be the human tumor with the highest frequency of oncogenes detected by DNA transfection. In each case the oncogene involved was N-ras, a member of the ras gene family. Biologic and clinical parameters of AML patients with and without N-ras oncogenes in their tumors are discussed.     

Tumorigenic and metastatic properties of "normal" and ras-transfected NIH/3T3 cells.

To investigate the role of oncogene activation in the pathogenesis of malignant tumors, we have studied the tumorigenic and metastatic properties of NIH/3T3 secondary transfectants (designated A51) containing an activated c-Ha-ras-1 gene derived from the human T24 bladder carcinoma cell line and compared them with untransfected NIH/3T3 cells. Whereas subcutaneous implantation of NIH/3T3 cells in the supraclavicular region produced palpable tumors that failed to metastasize, NIH/3T3 cells inoculated in the footpad gave rise to malignant tumors that metastasized to the lung. Under identical conditions and irrespective of the site of implantation, A51 cells formed rapidly growing primary tumors that produced pulmonary metastases. In an assay for experimental metastasis, intravenously injected NIH/3T3 cells gave rise to pulmonary nodules only at high cell inocula and in long-term survivors (90 days after injection). In contrast, A51 cells formed multiple lung tumor colonies detectable 14 days after injection. These results indicate that "normal" untransfected NIH/3T3 cultures contain subpopulations of cells that express malignant properties and that transfection of NIH/3T3 cells with activated c-Ha-ras-1 accelerates formation of metastases.     

Mouse NIH 3T3 cells expressing human colony-stimulating factor 1 (CSF-1) receptors overgrow in serum-free medium containing human CSF-1 as their only growth factor.

Mouse NIH 3T3 cells expressing the human c-fms protooncogene encoding the receptor for colony-stimulating factor 1 (CSF-1) are able to proliferate in serum-free medium containing platelet-derived growth factor (PDGF), insulin, transferrin, and albumin as the only exogenous proteins. When PDGF and insulin were replaced by purified human recombinant CSF-1, the cells became spindle shaped and refractile, were no longer contact inhibited, and proliferated to high densities. Thus, transduction of the human c-fms gene into mouse fibroblasts can not only reprogram their growth factor requirements but can also induce ligand-dependent features of cell transformation. NIH 3T3 cells stably transformed by the feline v-fms oncogene or by a mutated, oncogenic human c-fms gene were able to proliferate in the absence of exogenous growth factors. A monoclonal antibody that prevents signal transduction by the human CSF-1 receptor inhibited the growth of cells transformed by the activated c-fms oncogene, confirming that CSF-1 receptor function was required to abrogate growth factor requirements and to maintain the transformed state.