Int. J. Cancer: 80, 857–862 (1999) r 1999 Wiley-Liss, Inc. Publication of the International Union Against Cancer Publication de l’Union Internationale Contre le Cancer SPECTRUM OF TRANSFORMING SEQUENCES DETECTED BY TUMORIGENICITY ASSAY IN A LARGE SERIES OF HUMAN NEOPLASMS Johannes W.G. JANSSEN1,2*, Jürgen BRAUNGER1,2, Karin BALLAS2, Michael FAUST2, Ute SIEBERS2, Ada C.M. STEENVOORDEN2 and Claus R. BARTRAM1,2 1Institut für Humangenetik, Ruprecht-Karls-Universität, Heidelberg, Germany 2Sektion Molekularbiologie, Abteilung Kinderheilkunde II, Klinikum der Universität, Ulm, Germany We here summarize the analysis of 126 DNA samples from patients with hematopoietic neoplasias and solid tumors and from various tumor cell lines that were screened in the tumorigenicity assay. Thirty-eight samples were able to induce tumors after transfection in NIH/3T3 cells and injection into nude mice. Southern-blot analysis with a panel of oncogene probes revealed human ras genes in the vast majority of cases (25 N-ras, 2 K-ras, 1 H-ras) but also activated FGF4, dbl, ret and mas genes respectively. DNA samples from the 6 remaining transfectants were cloned into EMBL-3 phages and screened with a human specific repetitive Alu probe. Direct hybridization of a transfectant cDNA library allowed cloning of the ufo oncogene. Application of the exon-trapping technique to alu-positive phage DNA from the other transfectants enabled us to isolate tre, cot, B-raf, p85␤/HUMORF8 and a novel oncogene. Int. J. Cancer 80:857–862, 1999. r 1999 Wiley-Liss, Inc. Transfection of NIH/3T3 cells with genomic DNA from human tumors or human tumor cell lines has allowed the detection and cloning of a large number of oncogenes. (Wigler et al., 1979). The various genes identified in this way play a pivotal role in cell differentiation or cell proliferation and code for growth factors, receptors and second messengers. In the initial phase of application of the NIH/3T3 transformation strategy, the respective oncogenes belonged mostly to the ras family (Barbacid, 1987). However, improvements of the classical focus formation assay, such as the tumorigenicity assay (Fasano et al., 1984), expression cDNA cloning (Miki et al., 1991) and expression cloning by retroviral transfer of cDNA libraries (Whitehead et al., 1995) led to the discovery of various other transforming genes. The classical NIH/3T3 transformation assay and its different variants have already detected the following oncogenes: neu, met, ret, trk, raf, dbl, FGF4 (formerly hst), mas, lca, B-raf, tre, vav, hhc, ufo/axl, ect, mos, cot/tpl-2, tim, TC21, ost, lbc, FGF-8, lfc, lsc and NET1 (references in Janssen et al., 1998). Some of these oncogenes were present in an activated state in the original tumor material, whereas others became activated in vitro during the DNA-mediated gene transfer process. Here we summarize our experience in DNA transfection studies aimed at detecting transforming genes in a large series of hematopoietic neoplasias, solid tumors and tumor cell lines. MATERIAL AND METHODS Patients and cell lines Cells were studied from bone marrow or peripheral blood of patients with overt acute non-lymphocytic leukemia, chronic myeloproliferative syndromes and pre-leukemias. The samples were generously provided by Dr. H. Heimpel (University of Ulm, Germany). Samples obtained from patients with neuroblastomas and stomach carcinomas were provided by participating investigators of the German Neuroblastoma Study Group and by Dr. K. Grünewald (Department of Internal Medicine, University of Innsbrück, Austria) respectively. The following human cell lines were used: U937 (monoblastic), ML-1 (myelomonoblastic), Rc2a (myelomonocytic), HEL (erythroblastic), THP-1 (monoblastic), KG-1 (myeloblastic), CTV-2 (monoblastic), T24 (bladder carcinoma) and T47D (breast carcinoma). Transfection assay The tumorigenicity assay is based on the co-transfection of mouse NIH/3T3 fibroblasts with human tumor DNA and a dominant drug-resistant selectable marker, pRSVneo, followed by G418 selection and injection of the resulting 3T3 colonies into nude mice, as described (Janssen et al., 1987a). Genomic cloning and exon-trapping analysis DNA isolated from third-cycle tumor-induced mouse tumors was partially digested with Sau3AI, ligated to EMBL-3 BamHI arms, packaged, plated and screened with a human Alu probe (Janssen et al., 1991). Genomic sub-fragments of alu-positive phages were sub-cloned in both orientations into the SmaI site of the exon-trapping vector pL53In and transiently transfected into COS-7 cells (Auch and Reth, 1990). Total RNA (5 µg) was transcribed into cDNA with random hexamers and superscript RT, as recommended by the manufacturer (GIBCO BRL, Eggenstein, Germany). An aliquot (2 of 20 µl) cDNA reaction mixture was used in a 100-µl PCR reaction consisting of 20 mM Tris/HCl (pH 8.4), 50 mM KCl, 2.5 mM MgCl2, 0.001% gelatin (w/v), 200 µM of each deoxynucleotide triphosphate, 40 pmoles primers and 1 U Taq polymerase (Amplitaq DNA Polymerase, Perkin Elmer Applied Biosystems, Weiterstadt, Germany). Amplification was performed in an automated PCR processor (BioMed, Theres, Germany) as follows: 35 cycles comprising de-naturation at 92°C for 30 sec, annealing at 56°C for 60 sec, primer extension at 72°C for 90 sec, with an initial de-naturating step at 92°C for 3 min and a final extension step at 72°C for 10 min. PCR products were electrophoresed in a 2.5% agarose gel and stained with ethidium bromide. Southern-blot analysis Cellular DNA was digested with restriction endonucleases (Boehringer, Mannheim, Germany), electrophoresed on a 0.6% agarose gel and blotted onto Nytran 13N membranes (Schleicher and Schuell, Dassel, Germany). Filters were hybridized in 3 ⫻ SSC (0.45 M NaCl, 0.045 M sodium citrate), 5 ⫻ Denhardt’s, 200 µg/ml de-natured salmon sperm DNA, 1% SDS and 10% dextran sulfate at 63°C for 16 hr with random-prime-labeled probes (Pharmacia, Freiburg, Germany) with 2 µg/ml pRSVneo DNA as competitor DNA to avoid cross-hybridization with co-transfected pRSVneo sequences. Subsequently the filters were extensively washed in 3 ⫻ SSC, 0.1% SDS at 63°C, followed by a wash at moderate or high stringency. Filters were exposed to Kodak X-Omat DS film at ⫺70°C with Ilford intensifier screens. The following probes were used for hybridization analyses: the 38 portion of FGF4 (formerly Grant sponsor: Deutsche Krebshilfe; Grant number: 10-1253-JA-I; Grant sponsor: Deutsche Forschungsgemeinschaft. *Correspondence to: Institut für Humangenetik, Ruprecht-KarlsUniversität Heidelberg, Im Neuenheimer Feld 328, D-69120 Heidelberg, Germany. Fax: (49)6221-565155. E-mail: firstname.lastname@example.org Received 24 September 1998; Revised 5 November 1998 JANSSEN ET AL. 858 HST1) pORF1 cDNA (position 588-872); a partial 1.0-kb dbl cDNA clone (p3-7); a genomic ret fragment (pRet 3.3 and pRet 1.2); a 0.74-kb EcoRI-XhoI cDNA fragment of human B-raf. Northern-blot analysis Total RNA was isolated and purified by acid-guanidiniumisothiocyanate-phenol-chloroform extraction. Northern-blot analysis and stripping were performed as described (Janssen et al., 1998). Briefly, 10 µg of total RNA or 2 micrograms of poly A⫹ RNA were loaded onto a de-naturating agarose gel, electrophoresed in the presence of formaldehyde and transferred to Nytran 13 N nylon membranes (Schleicher and Schuell). Hybridization and washing were performed as described for Southern blotting. RESULTS DNA-transfection analyses of 126 tumor-DNA samples (46 acute myeloid leukemias, 33 chronic myeloproliferative syndromes, 7 myelodysplastic syndromes, 25 neuroblastomas, 12 stomach carcinomas, 2 breast-tumor cell lines and 1 bladdercarcinoma cell line) revealed that 38 samples were able to induce tumors in nude mice. Hybridization analyses showed the presence of an activated N-ras or K-ras gene in 25 and 2 hematopoietic neoplasias, respectively, as well as the expected H-ras sequences in the T24 bladder-carcinoma cell line. The results concerning the ras-positive tumors have been reported (Bos et al., 1985; Janssen et al., 1985, 1987a,b,c, 1988). DNA of the remaining 10 independent ras-negative primary tumors was able to induce secondary and TABLE I – DETECTION OF ACTIVATED GENES OTHER THAN ras FAMILY MEMBERS BY THE TUMORIGENICITY ASSAY Designation of primary tumor Origin CTV-2-T1 So-T3 SW-T1 U937-T2 HEL-T1 THP-1-T3 Gi-T1 Si-T2 Fr-T1 Ma-T1 monocytic cell line myeloproliferative disorder myeloproliferative disorder monocytic cell line erythroblastic cell line monocytic cell line neuroblastoma chronic myelocytic leukemia neuroblastoma (IV-N) stomach carcinoma Transforming sequence Mode of detection Hybridization with oncogene probes Genomic cloning and hybridization Exon-trapping strategy mas ⫹ (Janssen et al., 1988) ufo ⫹ (Janssen et al., 1991) p85␤/HUMORF8 ⫹ (Janssen et al., 1998) ret ⫹ cot ⫹ tre ⫹ FGF4 (hst) ⫹ dbl ⫹ B-raf ⫹ novel ⫹ FIGURE 1 – (a) Northern-blot analysis of ret gene expression in a third-cycle U937-T2 mouse tumor (U937T2g2) and NIH/3T3 recipient cells. poly A⫹ RNA (2 µg) isolated from NIH/3T3 cells and U937T2g2 tumor cells were submitted to Northern transfer. Filters were hybridized to a 32P-labelled genomic ret insert. 28S and 18S ribosomal RNA were used as molecular-weight markers. (b) Northern-blot analysis of FGF4 gene expression in Gi-T1 second-cycle mouse tumors. Total RNA (10 µg) isolated from Gi-T1B and C tumors, murine NIH/3T3 cells and human CTV-2 cells were submitted to Northern transfer. Filters were hybridized to a 32P-labelled FGF4 cDNA insert. Human FGF4-specific transcripts are indicated by an open arrow. The lower panel shows control hybridization with a 32P-labelled rat glyceraldehyde 3-phosphate dehydrogenase probe. ONCOGENES DETECTED BY THE TUMORIGENICITY ASSAY tertiary tumors. DNA of these secondary and tertiary tumors contained human repetitive sequences, suggesting the involvement of an activated human oncogene. The observation that all secondary and tertiary tumors that were derived from the same primary tumor contained a common set of human repetitive sequences supported this conclusion (Janssen et al., 1987c; and data not shown). These tumors were screened with a panel of oncogene probes, i.e., v-raf, c-fms, v-rel, v-src, v-fes, v-mos, p53, c-myc, N-myc, rat erbB2 (formerly neu), met, mcf-2 (⫽ dbl), mcf-3, erbB, c-mvb, PDGFB (formerly c-sis), c-abl, c-src, c-fos, c-ets1, mas, dbl, lca, ret and FGF4 (formerly hst) (Janssen et al., 1987c). DNA from 4 tumors contained the following oncogenes: mas (CTV-2-T1 tumor, Janssen et al., 1988), ret (U937T2 tumor), FGF4 (formerly hst) (GiT1 tumor) and dbl (SiT2 tumor) (Table I). Northern-blot analysis using a human ret probe revealed the common pattern of ret transcripts in RNA from the tertiary U937T2g2 transfectant (Fig. 1a, lane 2) (Ishizaka et al., 1989). RNA from normal NIH/3T3 cells did not show any ret transcripts (lane 1). Sequence analysis of various ret cDNA clones isolated from an U937T2g2 cDNA library confirmed the existence of 4 kinds of ret transcripts. These were generated by alternative splicing and different polyadenylation and resulted in 2 types of ret oncogene protein products (Ishizaka et al., 1989). Unfortunately, we did not identify any cDNA clones containing the 58 end of the ret transcripts and were thus not able to determine the possible fusion partner. The Northern-blot data were supported by Southernblot analysis showing amplified human ret DNA exclusively in secondary and tertiary U937T2 transfectants (data not shown). In accordance with earlier studies, Northern-blot analysis of 2 secondary GiT1 transfectants, GiT1C and GiT1B (Fig. 1b, lanes 3 and 4) showed one major FGF4 transcript of approximately 3.0 kb. and 2 minor transcripts (Yoshida et al., 1988), while RNA from a human tumor cell line (CTV-2) and control murine NIH/3T3 cells revealed no FGF4 transcripts (Fig. 1b, lanes 1 and 2). Southern-blot analysis showed the presence of dbl sequences in one primary and one secondary SiT2 transfectant (Fig. 2, lanes 1 and 2). Its hybridization pattern was similar to normal peripheral-blood-cell DNA (Fig. 2, lane 4). The DNA of NIH/3T3 cells did not show any hybridization with the human dbl cDNA probe (Fig. 2, lane 3). DNA derived from the remaining transfectants was cloned into EMBL-3 phages and screened for human sequences with a human repetitive Alu probe. Alu-positive phages were isolated and phage DNA was used directly as a probe to hybridize RNA or a cDNA library of the respective transfectant. In one case we were able to isolate and characterize a tyrosine-kinase receptor, called ufo (Janssen et al., 1991), but failed to identify any transcripts or positive cDNA clones with alu-positive genomic phage clones of the other transfectants. To circumvent possible problems related to the direct hybridization of complete genomic phage DNA, we decided to take advantage of the exon-trapping strategy. Genomic sub-fragments of alu-positive phage clones derived from unrelated tertiary transfectants were cloned in both orientations into the polylinker site of the pL53In exon-trapping vector, residing in an intron between 2 insulin exons (Auch and Reth, 1990). The resulting constructs were transfected into COS-7 cells and transiently expressed. RNA was isolated from the transfected COS-7 cells, transcribed into cDNA, and PCR-amplified using 2 specific exon-trap insulin primers. Various exon-trap constructs revealed a larger PCR product than the intrinsic control insulin PCR product, notably in one orientation (Fig. 3). The PCR products were cloned into a T-vector and sequenced with primers derived from the exon-trap vector. Sequence comparison of the PCR product of the SWT1 transfectant, SWM4 (Fig. 3, lane 2), revealed highest homology with the bovine p85␤ sub-unit of phosphatidylinositol (PI) 3-kinase. As we have reported, fusion of the human p85␤ sub-unit of PI 3-kinase and HUMORF8, a putative de-ubiquitin- 859 FIGURE 2 – Southern-blot analysis of the dbl gene. DNA samples (10 µg) obtained from a second-cycle Si-T2 transfectant (Si-T2a), a primary-cycle Si-T2 transfectant (Si-T2), control murine NIH/3T3 cells and normal peripheral blood were digested with EcoRI, electrophoresed in a 0.6% agarose gel and blotted onto nylon filter. The filter was hybridized with a human dbl cDNA oncogene probe, washed and exposed. Lambda-HindIII DNA fragments served as molecular-weight markers. ating enzyme, was generated during the DNA transfection process (Janssen et al., 1998). Sequence comparison of the PCR product of the HELT1 transfectant, HEL8 (Fig. 3, lane 4), revealed 100% homology with exons 4, 5 and 6 of the human cot oncogene (nucleotides 496-1033; Acc. Z14138). This transforming gene had been isolated earlier by transfection analysis with DNA from a human thyroid-carcinoma cell line using the SHOK cell line as a recipient, as well as by expression cloning with DNA from a Ewing’s-sarcoma cell line using NIH/3T3 cells as recipient cell line (Miyoshi et al., 1991; Chan et al., 1993). Southern-blot analysis confirmed the exontrapping data by demonstrating that the HEL transfectants exhibited 2 amplified human genomic cot fragments of about 4.5 and 1.5 kb (Fig. 4, lanes 1, 2 and 4) in addition to murine genomic cot fragments (Fig. 4, lane 3). Sequence comparison of the PCR product of the THP-1-T3 transfectant, THP-1/56 (Fig. 3, lane 6), revealed 100% homology with the human tre17 gene (nucleotides 4897-5838, clone 213; Acc. X63547), an oncogene detected with genomic DNA of a Ewing’s-sarcoma cell line via the tumorigenicity assay (Nakamura et al., 1988). In line with our exon-trapping data, Southern-blot analysis confirmed the presence of amplified human tre oncogene sequences in THP-1-T3 transfectants (Fig. 5, lanes 1 and 2) but not in NIH/3T3 cells (lane 3). Sequence comparison of the PCR products of 3 different genomic subfragments of the Fr-T1 transfectant, Fr6A2, Fr6BE2 and Fr6BES2 (Fig. 3, lanes 10, 12 and 860 JANSSEN ET AL. FIGURE 3 – RT-PCR analysis of exon-trap constructs from several different independent transfectants. Genomic sub-fragments of alu-positive phages or cosmids isolated from independent tertiary transfectants were sub-cloned into the SmaI site of the pL53In exon-trapping vector in both orientations and transiently transfected into COS-7 cells. RT-PCR products of the transfected COS-7 cells were loaded onto a 2.5% agarose gel. The position of the insulin exon is marked by asterisks. Amplification products containing trapped exon sequences are indicated by arrows. FIGURE 4 – Southern-blot analysis of the cot gene. DNA samples obtained from one tertiary and one quarternary HEL-T1 transfectant (HELT1b2 and HELT1b1b), NIH/3T3 cells and peripheral blood (BC) were digested with EcoRI, electrophoresed, blotted and hybridized with the HEL8 RT-PCR product obtained by exon trapping. LambdaHindIII DNA fragments are indicated. FIGURE 5 – Southern-blot analysis of the tre gene. DNA samples obtained from 2 secondary THP-1-T1 transfectants (THP-1T3b and THP-1T3c), NIH/3T3 cells and normal blood (BC) were digested with EcoRI, electrophoresed, blotted and hybridized with the THP-1/56 RT-PCR product obtained by exon trapping. ONCOGENES DETECTED BY THE TUMORIGENICITY ASSAY 861 14), revealed identity with 3 independent successive exons of the human B-raf oncogene (6BES with nucleotides 1235-1371; 6BE with nucleotides 1372-1489; 6A with nucleotide 1490-1574 of the human B-raf oncogene, Acc. M95712, M95720 and X54072). Southern blotting confirmed the exon-trap data (not shown). Northern-blot analysis showed an aberrant human B-raf transcript in Fr-T1 transfectant RNA, indicating that the human B-raf oncogene may have been activated through recombination, a common mode of raf activation (Fig. 6, lane 3). Sequence comparison of the PCR product of the Ma-T1 transfectant, MA30 (Fig. 3, lane 8), did not reveal any homology with known genes in the public database, suggesting that a novel oncogene had been identified. Northern-blot analysis with RNA of Ma transfectants and NIH/3T3 cells applying this Ma-exon-trap PCR product identified Ma-specific transcripts (data not shown). Identification of this novel oncogene is currently under investigation. DISCUSSION The application of the NIH/3T3 transformation assay has allowed the detection of a large number of oncogenes exhibiting diverse functions. In particular, the development of several variants of the classical NIH/3T3 focus formation assay made major contributions to this field. Except for ras genes, most genes became activated during the transfection process, by (i) quantitative alterations (over-expression) as a consequence of gene amplification, loss of genomic negative regulatory sequences, insertion of promoter sequences from the dominant drug-resistant selectable marker plasmid or the promoter sequences of the cloning vehicle used or (ii) qualitative changes such as gene truncation, point mutations or gene fusion. In other cases the oncogenes were already activated in the original tumor prior to gene transfer, as, e.g., activated ras genes in various neoplasms, the met gene in the MNNG-HOS cell line (Park et al., 1986), the erbB2 (formerly neu) gene in ENU-induced rat neuroblastomas (Shih et al., 1981) and the trk oncogene in a human colon carcinoma (Martin-Zanca et al., 1986). We and others have reported on activation of the mas gene by loss of negative regulatory genomic sequences (Young et al., 1986; Janssen et al., 1988), the activation of ufo/axl through amplification resulting in over-expression of this tyrosine-kinase receptor (Janssen et al., 1991), and the fusion of the human p85␤ sub-unit of PI 3-K with HUMORF8 sequences (Janssen et al., 1998). Our Northern-blot data suggest that the FGF4 (formerly hst) gene was activated due to the loss of negative regulatory sequences, which resulted in over-expression of the gene, a process similar to that described by Sakamoto et al. (1986). Our Northern analysis showed an aberrant B-raf transcript suggesting that the B-raf gene may have been activated by gene truncation. Northern analysis revealed high tre expression in THP-1-T1 transfectants compared with a moderate level of endogenous mouse tre expression in other transfectants. This observation may imply that de-regulation and over-expression of the human tre gene has led to transformation of the NIH/3T3 cells. Since human and murine tre transcripts have similar molecular weight, the exact mode of activation of tre in THP-1 transfectants remains elusive. Our data clearly demonstrate the usefulness of the exon-trapping technique in identification of activated oncogenes. In the case of the cot oncogene in HEL-T1 transfectants, exon trapping enabled us to detect 3 consecutive cot exons on one genomic sub-fragment of HELT1 tertiary transfectants and 3 consecutive B-raf exons on 3 different genomic fragments from Fr-T1 tertiary transfectants. In addition, this strategy enabled us to identify (i) the known tre oncogene, (ii) an oncogenic fusion product between the phosphatidylinositol 3-kinase p85␤ sub-unit and HUMORF8, and (iii) a novel oncogene. FIGURE 6 – Northern-blot analysis of B-raf gene expression in Fr-T1 second-cycle mouse tumors. Poly A⫹ RNA (2 µg) isolated from a secondary Fr-T1 transfectant (FrT1a), murine NIH/3T3 cells and the human CTV-2 cell line were submitted to Northern transfer. Filters were hybridized to a 32P-labelled B-raf cDNA insert. Human B-rafspecific transcripts are marked with an open arrow at the left of the upper panel. The aberrant B-raf transcript is indicated with an open arrow at the right of the upper panel. Our data support the view that the tumorigenicity assay may have a bias for the detection of specific oncogenes, such as mas, tre, B-raf, ret, cot, dbl and FGF4. The mode of oncogene activation (e.g., over-expression, gene fusion or loss of specific coding sequences) combined with inherent properties of the transfection procedure (e.g., DNA breakage, ligation of DNA fragments and amplification of DNA), its normal function, such as cellular signaling and, last but not least, the ability of a particular gene to transform NIH/3T3 cells, are factors that influence the detection frequency of this assay. Variants of the classical NIH/3T3 transformation assay and methods for isolating exon sequences from genomic sub-fragments continue to make possible the identification of even more genes involved in the complex process of tumorigenesis. ACKNOWLEDGEMENTS The excellent technical assistance of Mrs. M. Schmidberger, Ms A. Wunderlich and Ms U. Spadinger is gratefully acknowledged. We thank Mrs. A. Jordan for secretarial help with the manuscript. We thank the following colleagues for providing us with oncogene probes: Dr. H. Sakamoto (hst), Dr. A. Eva (dbl), Dr. G.M. Cooper (ret), Dr. T. Yamamoto (B-raf ). We also thank Dr. M. Reth for providing us with the pL53In exon-trapping vector. This work was supported by grants from the Deutsche Forschungsgemeinschaft and Deutsche Krebshilfe (10-1253-Ja I) to C.R.B. and J.W.G.J. 862 JANSSEN ET AL. REFERENCES AUCH, D. and RETH, M., Exon trap cloning: using PCR to rapidly detect and clone exons from genomic DNA fragments. Nucleic Acids Res., 18, 6743–6744 (1990). BARBACID, M., ras genes. Ann. Rev. Biochem., 56, 779–827 (1987). BOS, J.L., TOKSOZ, D., MARSHALL, C.J., VERLAAN-DE VRIES, M., VEENEMAN, G.H., VAN DER EB, A.J., VAN BOOM, J.H., JANSSEN, J.W.G. and STEENVOORDEN, A.C.M., Amino-acid substitutions at codon 13 of the N-ras oncogene in human acute myeloid leukemia. Nature (Lond.), 315, 726–730 (1985). CHAN, A.M.-L., CHEDID, M., MCGOVERN, E.S., POPESCU, N.C., MIKI, T. and AARONSON, S.A., Expression cDNA cloning of a serine kinase transforming gene. Oncogene, 8, 1329–1333 (1993). FASANO, O., BIRNBAUM, D., EDLUND, L., FOGH, J. and WIGLER, M., New human transforming genes detected by a tumorigenicity assay. Mol. cell. Biol., 4, 1695–1705 (1984). ISHIZAKA, Y., OCHIAI, M., TAHIRA, T., SUGIMURA, T. and NAGAO, M., Activation of the ret-II oncogene without a sequence encoding a transmembrane domain and transforming activity of two ret-II oncogene products differing in carboxy-termini due to alternative splicing. Oncogene, 4, 789–794 (1989). JANSSEN, J.W.G., LYONS, J., STEENVOORDEN, A.C.M., SELIGER, H. and BARTRAM, C.R., Concurrent mutations in two different ras genes in acute myelocytic leukemias. Nucleic Acids Res., 15, 5669–5680 (1987a). JANSSEN, J.W.G., SCHLEITHOFF, L., BARTRAM, C.R. and SCHULZ, C.R., An oncogenic fusion product of the phosphatidylinositol 3-kinase p85␤ subunit and HUMORF8, a putative deubiquitinating enzyme. Oncogene, 16, 1767–1772 (1998). JANSSEN, J.W.G., SCHULZ, A.S., STEENVOORDEN, A.C.M., SCHMIDBERGER, M., STREHL, S., AMBROS, P.F. and BARTRAM, C.R., A novel putative tyrosine–kinase receptor with oncogenic potential. Oncogene, 6, 2113– 2120 (1991). JANSSEN, J.W.G., STEENVOORDEN, A.C.M., COLLARD, J.G. and NUSSE, R., Oncogene activation in human myeloid leukemia. Cancer Res., 45, 3262–3267 (1985). JANSSEN, J.W.G., STEENVOORDEN, A.C.M., LOSEKOOT, M. and BARTRAM, C.R., Novel transforming sequences in human acute myelocytic leukemia cell lines. Oncogene, 1, 175–179 (1987c). JANSSEN, J.W.G., STEENVOORDEN, A.C.M., LYONS, J., ANGER, B., BÖHLKE, J.U., BOS, J.L., SELIGER, H. and BARTRAM, C.R., ras gene mutations in acute and chronic myelocytic leukemias, chronic myeloproliferative disorders and myelodysplastic syndromes. Proc. nat. Acad. Sci. (Wash.), 84, 9228– 9232 (1987b). JANSSEN, J.W.G., STEENVOORDEN, A.C.M., SCHMIDBERGER, M. and BARTRAM, C.R., Activation of the mas oncogene during transfection of monoblastic cell line DNA. Leukemia, 2, 318–320 (1988). MARTIN-ZANCA, D., HUGHES, S. and BARBACID, M., A human oncogene formed by the fusion of truncated tropomyosin and protein tyrosine kinase sequences. Nature (Lond.), 319, 743–748 (1986). MIKI, T., FLEMING, T.P., CRESCENZI, M., MOLLOY, C.J., BLAM, S.B., REYNOLDS, S.H. and AARONSON, S.A., Development of a highly efficient expression cDNA cloning system: application to oncogene isolation. Proc. nat. Acad. Sci. (Wash.), 88, 5167–5171 (1991). MIYOSHI, J., HIGASHI, T., MUKAI, H., OHUCHI, T. and KAKUNAGA, T., Structure and transforming potential of the human cot oncogene encoding a putative protein kinase. Mol. cell. Biol., 11, 4088–4096 (1991). NAKAMURA, T., HILLOVA, J., MARIAGE-SAMSON, R. and HILL, M., Molecular cloning of a novel oncogene generated by DNA recombination during transfection. Oncogene Res., 2, 357–370 (1988). PARK, M., DEAN, M., COOPER, C.S., SCHMIDT, M., O’BRIEN, S.J., BLAIR, D.G. and VANDEWOUDE, G.F., Mechanism of met oncogene activation. Cell, 45, 895–904 (1986). SAKAMOTO, H., MORI, M., TAIRA, M., YOSHIDA, T., MATSUKAWA, S., SHIMIZU, K., SEKIGUCHI, M., TERADA, M. and SUGIMURA, T., Transforming gene from human stomach cancers and a noncancerous portion of stomach mucosa. Proc. nat. Acad. Sci. (Wash.), 83, 3997–4001 (1986). SHIH, C., PADHY, L.C., MURRAY, M. and WEINBERG, R.A., Transforming genes of carcinomas and neuroblastomas introduced into mouse fibroblasts. Nature (Lond.), 290, 261–264 (1981). WHITEHEAD, I., KIRK, H. and KAY, R., Expression cloning of oncogenes by retroviral transfer of cDNA libraries. Mol. cell. Biol., 15, 704–710 (1995). WIGLER, M., PELLICER, A., SILVERSTEIN, S., AXEL, R., URLAUB, G. and CHASIN, L., DNA-mediated transfer of the adenine phosphoribosyltransferase locus into mammalian cells. Proc. nat. Acad. Sci. (Wash.), 76, 1373–1376 (1979). YOSHIDA, M.C., WADA, M., SATOH, H., YOSHIDA, T., SAKAMOTO, H., MIYAGAWA, K., YOKOTA, J., KODA, T., KAKINUMA, M., SUGIMURA, T. and TERADA, M., Human hst1 (hstf1) gene maps to chromosome band 11q13 and coamplifies with the int2 gene in human cancer. Proc. nat. Acad. Sci. (Wash.), 85, 4861–4864 (1988). YOUNG, D., WAITCHES, G., BIRCHMEIER, C., FASANO, O. and WIGLER, M., Isolation and characterization of a new cellular oncogene encoding a protein with multiple potential transmembrane domains. Cell, 45, 711–719 (1986).