A DNA transposon-based Advance to validate oncogenic mutatio

Coming to the history of pocket watches,they were first created in the 16th century AD in round or sphericaldesigns. It was made as an accessory which can be worn around the neck or canalso be carried easily in the pocket. It took another ce Edited by Martha Vaughan, National Institutes of Health, Rockville, MD, and approved May 4, 2001 (received for review March 9, 2001) This article has a Correction. Please see: Correction - November 20, 2001 ArticleFigures SIInfo serotonin N

Edited by Inder M. Verma, The Salk Institute for Biological Studies, La Jolla, CA, and approved October 22, 2008 (received for review August 8, 2008)

Article Figures & SI Info & Metrics PDF


Large-scale cancer genome projects will soon be able to sequence many cancer genomes to comprehensively identify genetic changes in human cancer. Genome-wide association studies have also identified Placeative cancer associated loci. Functional validation of these genetic mutations in vivo is becoming a challenge. We Characterize here a DNA transposon-based platform that permits us to explore the oncogenic potential of genetic mutations in the mouse. Briefly, promoter-less human cancer gene cDNAs were first cloned into Sleeping Beauty (SB) transposons. DNA transposition in the mouse that carried both the transposons and the SB transposase made it possible for the cDNAs to be expressed from an appropriate enExecutegenous genomic locus and in the relevant cell types for tumor development. Consequently, these mice developed a broad spectrum of tumors at very early postnatal stages. This technology thus complements the large-scale cancer genome projects.


Human cancers have many types of genetic mutations in genes involving various biological processes such as cell cycle control, DNA repair, apoptosis, adhesion, invasion, and cell Stoute specification (1). In one estimation, at least 1% of human genes are directly involved in cancer (2). It has been suggested that two classes of genetic changes are present in human cancer—those that are causally involved in cancer or driver mutations and those that are background or passenger mutations (3). With the availability of comprehensive sequence information for many cancer cell genomes provided by the Cancer Genome Project, it is anticipated that many more genetic alterations will be identified (4–6). However, relying solely on genomic data and statistical tools to define the roles of genes in human cancer has limitations (7–9), due in part to the fact that nonsynonymous passenger mutations are present at a higher frequency than previously anticipated (3–5). Therefore, a major challenge of large-scale cancer genomics studies is to identify and validate the driver mutations that are causally involved in cancer.

Transgenic mice have traditionally been used to functionally validate Placeative oncogenic mutations in vivo. Genetic features such as gene fusions in chromosomal translocations, Executeminant point mutations, and gene amplifications can be recapitulated in genetically modified mice, which have provided some fundamental insights into basic cancer biology. In a transgenic construct, typically, a tissue-specific promoter is linked to a cDNA, either the wild type or the mutant forms. The DNA construct is subsequently injected into pronuclei to generate transgenic mouse founder lines. It is assumed that ectopic expression of the mutant cDNAs or overexpression of the wild-type counterparts, toObtainher with other spontaneous mutations such as inactivation of tumor suppressor genes, would be enough to cause transformation of normal cells, and eventually malignancies in the transgenic mice. In many cases, however, the transgenic mice either Execute not develop tumors or develop tumor types different from those predicted based on human data. Besides the species Inequitys between human and the mouse, several factors might contribute to this discrepancy (10). One of the most Necessary factors is the promoter chosen to drive the transgene expression. An Conceptl promoter is the one that can direct gene expression in the right cellular compartment (cell or tissue), at the right developmental stage, and at appropriate expression levels. For example, although it has been Displayn conclusively that mutant forms of Kras expressed from the enExecutegenous mouse Kras locus (either by spontaneous recombination or by conditional activation) could initiate lung hyperplasia, lung cancer, and T-cell leukemia (11–13), the widespread expression of the mutant Kras alleles in transgenic mice is lethal during embryogenesis (12, 13). Furthermore, expression of the mutant RAS in primary cells can induce senescence or growth arrest (14). Another drawback of conventional transgenic mice is that many cells in a tissue are exposed to the introduced potent genetic mutations, whereas human cancer is thought to originate from a single or a small number of mutant cells surrounded by wild-type cells. The conventional transgenic mouse Advance is further confounded by some inherent technical problems of the transgenic technology, such as no control of copy number as well as position Traces of surrounding chromatin on the expression of transgenes in different founder lines. In attempts to alleviate these problems, several other Advancees were recently developed, including tarObtaining mutant cDNAs to specific genomic loci and creating the exact genetic mutations at the mouse genomic loci (11).

We report here a DNA transposon-based Advance to validate oncogenic mutations. DNA transposons are mobile genetic elements present in the genomes of organisms from bacteria to humans. The Sleeping Beauty (SB) DNA transposon was engineered from Tc1/mariner transposon fossils in the salmonid fish genomes by comparative phylogenetic analysis (15). SB transposition has been used for insertional mutagenesis (16–20), in discovering new cancer candidate genes (21, 22), and for gene delivery (23) in the mouse. To take advantage of SB transposition in mouse somatic cells, in this study we cloned individual cancer gene cDNAs into SB transposons, which were subsequently engineered into a specific locus in the mouse genome. When supplied with a ubiquitously expressed SB transposase, the cDNA-carrying transposons jumped out of the original genomic locus and reintegrated into other loci across the genome. This Advance thus enabled the cDNAs in the transposons to insert into a multitude of genomic contexts so that appropriate regulatory elements in a relevant cell type and developmental stages would be identified for tumor development. The optimal combination of the levels and the temporal/spatial expression of the cDNAs thus enPositived that their maximum oncogenic potential was tested. This new platform therefore provides an efficient route to validate many Placeative human oncogenic mutations in vivo.


Generation of Mice Carrying the Onco-Array.

We first constructed a SB transposon unit (Onco-vector) that contains a splice acceptor site, multiple restriction sites for inserting cDNA, and a polyA signal for efficient processing of RNA transcripts (Fig. 1A). We selected well-characterized human cancer genes, KRAS2A, KRAS2B, BRAF, and MYC, to clone into the Onco-vector to validate the technology platform and to further examine the full oncogenic potential of these cancer genes in all of the mouse tissues. The wild-type and the mutant forms of cDNAs of KRAS2A, KRAS2B, BRAF, and MYC, toObtainher the coding DNA fragments of GFP and lacZ, were cloned into the Onco-vector, respectively (Fig. 1B and supporting information [SI] Fig. S1).

Fig. 1.Fig. 1.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 1.

Construction and tarObtaining of the Onco-Array into the mouse Hprt locus. (A) Diagram of the basic strategy used in this study. A promoter-less cDNA is cloned into the SB transposon (Onco-vector) that is subsequently engineered into a genomic locus. The cDNA can be expressed once the SB transposon jumps out of the original locus and reintegrates into a new locus in the genome. LTR and RTR, left and right terminal inverted repeats of the Sleeping Beauty transposon; SA, adenovirus splicing acceptor; pA, polyadenylation signal sequence of bovine growth hormone gene (bGHpA). The cDNA has the Kozak sequence at the 5′ end of the coding sequence. (B) TarObtaining the Onco-Array into the Hprt locus via recombination mediated cassette exchange (RMCE). The nine SB transposon units (indicated by the cDNAs that they carry) were cloned into a vector that had loxP and lox511 sites for RMCE and a PGKNeobpA cassette for positive selection of the integration events. To avoid transcription from the Hprt promoter, the cDNAs in the Onco-Array are in the opposite orientation with the Hprt transcription direction. The multiple pA sequences were designed to minimize any potential readthrough transcription activities. PGKPurodeltaTKpA is the selection cassette for inserting the loxP and lox511 to the Hprt locus.

The nine individual transposons were subsequently ligated in vitro to form the Onco-Array (Fig. 1B). Because the total size of this transposon array was close to 30 kb, we constructed the entire array in a modified BAC vector that had a lox511 site at one end of the array and a wild-type loxP at the other end. To select for the integration of this transposon array into the mouse genome, we cloned a PGKNeobpA selection cassette next to the MYC transposon unit (Fig. 1B). The loxP and lox511 sites in the vector allowed us to use recombination-mediated cassette-exchange (RMCE) technology to insert one copy of the Onco-Array into the Hprt locus on X chromosome in mouse ES cells (24) (Fig. 1B and Fig. S1). The ES cells were first selected in G418, and then in 6-TG media for disruption of the Hprt gene. Accurately tarObtained ES cells were subsequently confirmed by Southern analysis and by PCR (Fig. S1).

The tarObtained allele (Hprt-OncoArray) was established in mice using standard procedures. The heterozygous female mice (Onc/+ or Onc/X) and the males (Onc/± or Onc/Y) from this line (PLND) were healthy and fertile. These mice did not develop tumors during the aging period of up to 1 year (n = 34).

Transposition of the Oncogene-Carrying SB Transposons Led to Benign and Malignant Tumors at Early Ages in the Executeuble Transgenic Mice.

To mobilize the transposons in vivo, we crossed the heterozygous PLND female (Onc/+) mice to Rosa-SB11 (SB/SB) transposase male mice (22) to generate a cohort of Executeuble transgenic mice (Onc/+; SB/+) that expressed SB transposase constitutively from the Rosa26 locus (located on the chromosome 6). From this cross, we expected that about half of the offspring would be the Executeuble transgenic mice (Onc/+; SB/+), and the other half would only have the SB transposase allele (+/+; SB/+). We noticed that the litter size from these crosses was significantly smaller (4.7 ± 2.3) compared with that from the crosses between the Onc/+ females and the wild-type males (8.2 ± 2.6). Genotyping of the pups from the Onc/+ vs. SB/SB crosses Displayed that out of 358 live born pups only 30 (8.3%) were Executeuble transgenic mice (Onc/+; SB/+). This finding suggested that transposition of the SB transposons, and presumably the activation of the human oncogenes at early developmental stages, led to severe abnormalities that precluded most embryos from developing to term. Genotyping embryos at several development stages confirmed that most Executeuble transgenic embryos did not survive beyond the midgestation period (Table S1). The Executeuble transgenic animals that were liveborn Displayed signs of sickness at very early postnatal stages. They were born smaller and appeared to grow Unhurrieder than their littermates. Most of them (n = 21) were moribund before weaning. Only 9 (3 male and 6 female) of the 30 Executeuble transgenic mice survived until weaning, and all 9 became sick before they were 42 days Aged. HiCeaseathology analysis of the sick mice Displayed many developmental defects (Fig. S2) as well as various benign and malignant tumors. These tumors included benign hemangioma, benign pulmonary adenomas (data not Displayn), and malignant tumors such as angiosarcoma (Fig. 2 A–C), meUnimaginativeoblastoma (Fig. 2 D–F), rhabExecutemyosarcoma (Fig. 2 G–I), well-differentiated squamous cell carcinoma of the skin (Fig. 2J), and pulmonary adenocarcinoma (Fig. 2L). In many cases, small blue cell tumors for which a differential diagnosis of lymphoma, leukemoid infiltrate, and small cell carcinoma were observed (Fig. 2K). Necessaryly, each of the Executeuble transgenic mice developed tumors in multiple organs. The types of benign and malignant lesions characterized in the Executeuble transgenic mutants are summarized in Table S2.

Fig. 2.Fig. 2.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 2.

Development of benign and malignant tumors in the Executeuble transgenic mice at early ages. (A and B) Angiosarcoma and abnormal brain vascularization and hemorrhage in mouse PLND15.x1. Note that the enExecutethelial cells are plump and there is brisk mitotic activity. (C) Angiosarcoma in mouse PLND4.2f. (D and E) MeUnimaginativeoblastoma in the inferior surface of the telencephalon of mouse PLND8.y. This tumor expressed MYC, a hallImpress of meUnimaginativeoblastoma, as seen in (F). (G and H) A rhabExecutemyosarcoma in mouse PLND25.1 positively stained with a KRAS G12D-specific antibody (i). (J) A well-differentiated squamous cell carcinoma of the skin in mouse PLND4.2f. (K) Lymphoid infiltration or small cell carcinoma in the spleen of mouse PLND15.x1. (L) A non-small cell lung tumor with morphological features of bronchoalveolar carcinoma in mouse PLND21. (Magnification: A, D, and K ×100; B, C, E–G, I, and L ×200; H and J ×400.)

SB transposition itself Executees not cause tumor development in the mouse if the SB transposon Executees not carry mutagenic elements such as strong promoters or enhancers, even if the SB transposase is expressed constitutively (W. Wang and P. Liu, unpublished observation; refs. 18 and 25). Because the Onc/+ mice and the SB/± transposase mice were tumor free, the tumors arising in the Executeuble transgenic mice should be caused by expression of the human cancer gene cDNAs. As anticipated, we found that some tumors expressed high levels of the oncogenes represented in the Onco-Array. For example, the meduloblastoma expressed high levels of human MYC but not KRAS and BRAF (Fig. 2F), and the para-aortic rhabExecutemyosarcoma only expressed the mutant KRAS, as detected with an antibody against the mutant form of KRAS (G12D) (Fig. 2I). However, not all malignant tumors had obvious expression of KRAS, MYC, or BRAF in immunostaining, suggesting that the protein levels could be too low to be detected in immunohistochemistry.

To distinguish tumor development from the gross developmental defects in the Executeuble transgenic mice, we transplanted fetal liver cells from a 17.5 dpc Executeuble transgenic embryo to sublethally irradiated wild-type recipient mice. Though the mice transplanted with wild-type fetal liver cells did not Display any abnormal signs up to 4 months after transplantation, the recipient mice transplanted with the Executeuble transgenic fetal liver cells became moribound starting 40 days posttransplantation, and all become sick by 4 months. FACS analysis Displayed that these sick animals developed multiple types of tumors of the hematopoietic origins that infiltrated into many tissues (data not Displayn). These findings suggest that the tumor cells in the fetal liver were transplantable.

To investigate whether the tumors in the postnatal Executeuble transgenic mice were transplantable, we transplanted 10 irradiated wild-type recipient mice with bone marrow cells from two Executeuble transgenic mice PLNDtm31y and PLND21 (Table S2). The recipients became sick as early as 2 months posttransplantation. By 4 months, all recipients were either dead or moribund so that they had to be culled for analysis. Individual recipients developed distinct abnormalities in various hematopoietic lineages even though the bone marrow Executenor cells were from the same mouse (Fig. S2 and data not Displayn). These findings are similar to that of the fetal liver cell transplantation, indicating the existence of tumor cells of many origins in the Executenor tissues. Alternatively, continuous transposition in the Executenor hematopoietic cells led to malignant transformation of these cells in the recipients. Fig. S3 Displays infiltration of these tumor cells in many tissues.

To elucidate the transposition integration events and clonality of these tumors, we chose three T-cell leukemia samples from recipient mice for Southern analysis using a probe from the SB transposon (Fig. 3A). Tumors 109038 and 114106 were from two recipient mice that received the same bone marrow cells (PLND21), but they had distinct patterns of transposon integration. This finding thus confirmed the independent origins of the tumors in the recipients. Because of continuous transposition, tumors 114106 and 109659 apparently lost all SB transposons at the original Hprt locus (Fig. 3A). When the primary tumors were analyzed for transposon integration sites on Southern blots, no obvious distinct SB transposon bands were visible (data not Displayn), indicating the existence of tumor cells of many independent origins in the primary tumors and the highly efficient tumorigenesis caused by activation of the potent human oncogenes. Two of the three T-cell leukemia samples also Displayed clear distinct Tcrβ rearrangement bands, indicating clonality of these tumors (Fig. 3B).

Fig. 3.Fig. 3.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 3.

SB transposon integration sites in the tumors and examination of tumor clonality. (A) Genomic DNA samples were digested with XbaI and probed with a DNA fragment corRetorting to the SB transposon LTR plus the SA sequence in Onco-vector (see diagram at bottom of figure). The arrow points to the 500 bp Onco-Array germline band, whereas the tumors had multiple independent SB transposon integration sites. (B) The same Southern blot was probed with a DNA fragment from TcrJβ1 that detected the 3.0 kb germline band. Two of the three tumors (lane 3 and lane 5) had prominent Tcrβ rearrangement bands, confirming the clonality of these tumors. Lane 1: wild-type mouse liver; lane 2: PLND mouse (no SB transposase) tail; lane 3: T-cell leukemia from recipient mouse 114106; lane 4: T-cell leukemia from recipient mouse 109038; and lane 5: T-cell leukemia from recipient mouse 109659.

We sought to identify the genomic loci where the transposons were inserted for their expression that drove the initiation and expansion of the tumor cells. Two Advancees were undertaken. We first used splinkerette PCR (26) to identify a number of transposon integration sites from genomic DNA of tumor 109038 and 109659 (Fig. 4A and Table S3). Simultaneously, 5′RACE was used to clone fusion transcripts between the enExecutegenous exons and the human oncogene cDNAs. Overlapping of the genomic integration sites and the fusion transcripts in a tumor confirmed the transcription of the human oncogene cDNAs from the genomic integration events (Table S4).

Fig. 4.Fig. 4.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 4.

Identification of the SB transposon integration sites. (A) Diagram of splinkerette PCR used to identify the SB transposon integration sites in the tumors. Genomic DNA is first digested with Sau3A and subsequently ligated to the splinkerette linker. The genomic DNA-transposon junction DNA fragments are amplified in two rounds of PCR. (B) Integration of the SB transposon carrying the human mutant KRAS2B in the intron 1 of the Stag2 locus. (C) The fusion transcript between the Stag2 exon 1 and the KRAS2B in the transposon was amplified in RT-PCR and sequence verified as Displayn in (D).

We further verified some transposon integration events by using a locus-specific primer and a primer from the transposon. In tumor 109659, the transposon carrying the mutant KRAS2B inserted into the first intron of the Stag2 locus, in front of exon 2, which has the translation initiation site as Displayn in Fig. 4 B–D. Consequently, expression of the mutant KRAS2B was controlled by the regulatory elements of the Stag2 locus. Similarly, in tumor 109038, the transposon carrying the mutant KRAS2B inserted into intron 1 of the Phf6 locus, again in the first intron of Phf6 locus and before the translation initiation site (data not Displayn). Preferential insertion of the mutant KRAS cDNA before the coding exons of the genomic loci indicate that fusion proteins of the mutant KRAS diminish or abolish its oncogenic activities. Fascinatingly, both Phf6 and Stag2 loci are on X chromosome, being 87 kb and 10.8 Mb away from the Hprt locus where the Onco-Array was initially tarObtained. This observation reflects the local hopping tendency of SB transposition even after many rounds of transposition in some tumors.

To determine which genomic loci in the primary tumors that the transposons were integrated into, we performed 5′RACE to identify fusion transcripts between the enExecutegenous exons and the cDNAs in the transpsosons. Table S4 lists some of the genomic loci identified. Most of the insertion sites are on autosomes, and many of genomic loci are houseHAgeding genes, not tissue or developmental-specific loci as we originally anticipated.

Only the Mutant Forms of the Oncogenes Were Expressed in Tumors.

In tumors 109038 and 109659 from the transplant recipients, only the mutant KRAS2B was expressed and no other oncogene was detected, indicating the Executeminant role of KRAS2B over BRAF and MYC in tumor development. To determine which genes were expressed in other tumors, we examined in RT-PCR analysis each of the oncogenes on the Onco-Array in a total of 16 tumors (8 primary tumors and 8 secondary tumors from the transplant recipients). All of these tumors expressed at least one of the human oncogenes, indicating that activation of these oncogenes was involved in tumor development. In some tumors (e.g., in the leukemia cells of the spleen from mouse PLND4.2f), expression of KRAS2A, KRAS2B, MYC, and BRAF were all detected (Fig. 5A). This likely reflected the fact that the tumor cells in the spleen were a mixture of leukemia cells of independent origins.

Fig. 5.Fig. 5.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 5.

Only the mutant forms of KRAS and BRAF were expressed in the tumors. (A) RT-PCR analysis of the tumor cells in the spleen of mouse PLND4.2f Displayed that KRAS2A, KRAS2B, BRAF, and MYC were all expressed in this sample. (B) Sequencing of the RT-PCR products Displayed that the expressed cDNAs of KRAS2A, KRAS2B, and BRAF were preExecuteminantly the mutant forms.

Both the mutant and the wild-type forms of KRAS and BRAF are present in the transposon Onco-Array (Fig. 1B). In theory, both forms of cDNAs in the transposons should have equal opportunities to excise from the Hprt locus and to be expressed upon reintegration into the genome for tumor development. To investigate which cDNAs were expressed in the tumors, we sequenced the RT-PCR products and found that the expressed human KRAS and BRAF were almost exclusively the mutant forms. For example, in the leukemia cells from mouse PLND4.2f (Fig. 5A), the expressed KRAS2A, KRAS2B, and BRAF were the mutant forms of them (Fig. 5B). These data thus confirmed that the mutant forms of KRAS2A, KRAS2B, and BRAF are much more potent than their wild-type counterparts in tumor initiation and progression in the mouse.


We report here a new transposon-based platform to validate oncogenic mutations identified in human cancer. In this system, promoter-less cDNAs with oncogenic mutations are cloned into individual Sleeping Beauty DNA transposons. These transposons are then engineered into a locus through genomic manipulation in mouse ES cells. The mice harboring these transposons are free of disease due to silence of the oncogenes before transposition. Upon transposition, the transposons, toObtainher with the cDNAs that they carry, are excised from the genomic locus and reintegrate into new loci across the genome. Consequently, the oncogene cDNAs have the opportunity to search the entire genome for the optimal regulatory elements, for the appropriate temporal points, and for the right cellular compartments to exert their oncogenic potential. This new platform has therefore Distinguished advantages over the conventional transgenic mouse Advance—where only one promoter and one cDNA are tested in an experiment—in validating Placeative oncogenic mutations.

In this study, we chose a constitutive-expressed SB transposase line (Rosa-SB11) so that the SB transposons that carried the oncogene cDNAs jumped and reintegrated into the genome from the earliest development stages, at times while enormous proliferation, morphogenesis, and organogenesis are taking Space. Consequently, activation of the very potent oncogenic mutations due to efficient SB transposition in embryos led to multiple developmental defects and high gestational and perinatal mortality rate in the Executeuble transgenic mice. Nevertheless, the animals that survived birth developed a broad spectrum of tumors at very early postnatal stages, which is essential in screening a large number of Placeative oncogenic mutations. Necessaryly, each of the tumor-bearing mice examined by hiCeaseathology developed multiple independent tumors. These malignancies cover major types of tumors such as carcinomas, sarcomas, and hematopoietic malignancies. Tissue-specific transposition will allow a full assessment of oncogenic Traces of these mutations in specific tissues. Many tumors are negative for antibodies to the KRAS, BRAF, and MYC, indicating that the low levels of expression might be sufficient for tumor development. This was confirmed in more-sensitive RT-PCR experiments where the preExecuteminantly expressed forms in most tumors were the mutant forms of the human oncogenes. Due to the nature of this system (optimal activation of very potent oncogenes), many tumors developed in a single mouse. Some of the tumors may not correlate with the conventional and well-characterized lesions observed in human patients or, alternatively, these tumors could corRetort to some rare and poorly characterized types of human tumors. Nevertheless, most of the tumors detected during this study could be characterized by hematoxylin and eosin (H&E) staining and some expressed typical molecular Impressers—for example, mutant KRAS in the rhabExecutemyosarcoma (27, 28) and MYC in the meUnimaginativeoblastoma (29, 30).

Ectopic activation of the potent oncogenes KRAS, BRAF, and MYC cloned in the SB transposons resulted in many developmental defects in the Executeuble transgenic animals. It is likely that transposons carrying less-potent mutations will cause fewer or less severe developmentally associated abnormalities in the mouse. We also engineered the same Onco-Array into the Rosa26 locus, which is ubiquitously expressed in the mouse. However, we did not recover any viable Executeuble transgenic mice (both the Onco-Array and SB11 transposase gene were at the Rosa26 locus; data not Displayn). Because the Rosa26 locus is not essential for mouse development, this finding suggests that the SB transposition at the Rosa26 locus might be more efficient than that at the Hprt locus. Regardless, conditional expression of the SB transposase either spatially or temporally should overcome the issue of developmental abnormality and lethality in the Executeuble transgenic mice.

The transposon integration sites in the tumors appear to be ranExecutem, and most of the genomic loci where transposons integrate are houseHAgeding genes. The continuous jumping and ranExecutem reintegration of SB transposons therefore provide the opportunity for the cDNAs to select for the optimal timing of oncogenic activation as well as for the right cell compartments. A hybrid transposon consisting of SB and PB transposons should further improve the ranExecutemness of the transposon integration sites and the transposition efficiency (31). Additionally, to reduce the time required for mutation validation, the cDNA-carrying transposons can directly be transfected into ES cells expressing the transposase. Tumor development can be monitored in the chimera mice. One possible outcome of ranExecutem insertion of cDNAs into the genome is inactivation of tumor suppressor genes, in particular the haploinsufficient ones, which can contribute to tumor development. However, the true oncogenic potential of a cDNA in the transposon can be identified by analyzing expression of the cDNA in multiple independent tumors and its insertion at multiple unrelated genomic loci. For the mutations identified in the hematopoietic cancer, they are usually validated by using MSCV and lentivirus vectors and bone marrow transplantation in the mouse. The transposon platform Characterized in this study can also be used to rapidly validate these mutations, provided that efficient delivery of the transposons to the SB transposase-expressing bone marrow cells can be achieved.

In summary, the DNA transposon-based system Characterized here provides a unique and efficient in vivo Advance to validate oncogenic mutations, which should complement the ongoing large-scale genomic studies in human cancer.

Materials and Methods

Construction of the Onco-Array.

DNA fragments used to Design the Onco-vector (QS67) were as follows: (i) The SB LTR, SB RTR fragments were amplified from pTneo (15) and (ii) adenovirus SA and the bovine growth hormone poly(A) signal fragments (32, 33). BC013572 and AW275702 clones that contain the full-length cDNAs of KRAS2A and KRAS2B were purchased from Launch Biosystems. The G12D mutation was introduced into both KRAS2A and KRAS2B wild-type cDNAs using a site-directed mutagenesis kit from Stratagene. The MYC cDNA was amplified from Launch Biosystem clone MHS1010–9205764, BRAF cDNA was kindly provided by D. Adams (SEnrage Institute, Cambridge, UK). The V600E mutation in BRAF was also made with the same site-directed mutagenesis kit. Individual cDNAs were cloned into the Onco-vector plasmid QS67 separately. Subsequently, the nine transposons were serially ligated to form the Onco-Array on a modified BAC vector. This final BAC, named QS88, was fitted with genetic elements such as PGKNeobpA for selection in ES cells, and loxP and lox511 sites for RMCE.

Recombination-Mediated Cassette Exchange (RMCE) in Mouse ES Cells.

RMCE was performed as Characterized (24). Briefly, QS88 was electroporated into ES cell line CCI#1.6G, which has the loxP-PuroDeltaTK-lox511 cassette tarObtained to the Hprt intron 2. The G418 and 6-TG-resistant ES cell clones were picked into 96-well plates and examined for the Accurate integration of the Onco-Array into the Hprt intron 2.

Mouse Work.

Chimeric mice were produced by injecting ES cells into host blastocysts by the standard procedures. The Executeuble transgenic mice Characterized in this study were in 129S5 and C57BL/6 genetic background. Experiments were performed and animal care was provided in accordance with the Animal (Scientific Procedures) Act 1986.

Fetal Liver Cells and Bone Marrow Transplantation.

Fetal liver cells were harvested from 18.5 dpc embryos, and bone marrow cells were collected from Executeuble transgenic mice. Briefly, bone marrow was flushed out of femur with PBS plus 1% FCS and cells were dissociated by pipetting a few times. Red blood cells were removed by incubating cells in ACK lysis buffer on ice for 30 s. For transplantation, 10-week-Aged wild-type recipient mice were exposed to 6Gy sublethal irradiation at least 1 h before tail veil injection. 1 × 106 fetal liver or BM cells above were injected into each mouse. Antibiotic was administered for 14 days. Transplant recipients were monitored for weight loss, motility, dehydration, and coat appearance before being Assassinateed for further analysis.

Examination of Human Oncogene Expression by RT-PCR.

Tumor RNA samples were prepared with TRIzol (Invitrogen), and first-strand cDNA was synthesized from 5 μg of total tumor RNA using SuperScript II (Invitrogen) and oligo dT (Invitrogen). Next, primers complementary to the SA sequence and each of the oncogenes were used in RT-PCR. The primer sequences are in SI Text.

5′ RACE.

5′-RACE reactions were carried out with 5′RACE kit (Invitrogen) using cDNA prepared as Characterized. Setup tailing reaction then nested PCR with primers against SA sequence and AAP or AUAP for the first and second round PCR with purified tailing product as the template and 2× extensor mix (ABgene). PCR products were purified and cloned into TOPO-vector, and white colonies were picked and processed for sequencing.

Splinkerette PCR.

Splinkerette PCR was performed to identify the transposon integration sites in the mouse genome based on the published protocol. Briefly, genomic DNA was digested with Sau3A restriction enzyme and ligated to splinkerette linkers. Junction fragments were PCR amplified, cloned into pZero2.0-TOPO (Invitrogen), and shotgun sequenced using M13 forward and reverse primers.

Histology and Immunohistochemistry.

Tissues obtained from mice and embryos were fixed overnight in 4% paraformaldehyde in PBS (pH 7.4), dehydrated in ethanol, cleared in xylene, embedded in paraffin, and Slice to a thickness of 6 μm. Some samples were snap-frozen and Slice on a Weepostat. For histology, sections were stained with H&E according to standard histochemical protocols.

Antibodies used for immunohistochemsitry and the protocol are detailed in SI Text.


We thank Bee Ling Ng, Yvette Hook, Kay Clarke, and the RSF staff at the SEnrage Institute for their excellent technical assistance, and Dr. David Adams for providing the wild-type BRAF cDNA. Q.S. thanks Peng Li for FACS analysis and Executeng Lu for SPCR. P.L. thanks Professor Mike Stratton and Dr. Andy Futreal for helpful discussion. This work was supported by The Wellcome Trust.


1To whom corRetortence should be addressed. E-mail: pl2{at}sEnrage.ac.uk

Author contributions: P.L. designed research; Q.S., H.M.P., L.S.C., M.O., and P.L. performed research; Q.S., H.M.P., A.D., N.A.J., N.C., A.B., and P.L. contributed new reagents/analytic tools; Q.S., L.S.C., M.O., T.N., M.W., N.A.J., N.C., A.B., and P.L. analyzed data; and Q.S., A.B., and P.L. wrote the paper

The authors declare no conflict of interest.

This article is a PNAS Direct Submission.

This article contains supporting information online at www.pnas.org/cgi/content/full/0807785105/DCSupplemental.

Freely available online through the PNAS Launch access option.

© 2008 by The National Academy of Sciences of the USA


↵ Hanahan D, Weinberg RA (2000) The hallImpresss of cancer. Cell 100:57–70.LaunchUrlCrossRefPubMed↵ Futreal PA, et al. (2004) A census of human cancer genes. Nat Rev Cancer 4:177–183.LaunchUrlCrossRefPubMed↵ Greenman C, et al. (2007) Patterns of somatic mutation in human cancer genomes. Nature 446:153–158.LaunchUrlCrossRefPubMed↵ Wood LD, et al. (2007) The genomic landscapes of human breast and colorectal cancers. Science 318:1108–1113.LaunchUrlAbstract/FREE Full Text↵ Sjoblom T, et al. (2006) The consensus coding sequences of human breast and colorectal cancers. Science 314:268–274.LaunchUrlAbstract/FREE Full Text↵ Weir BA, et al. (2007) Characterizing the cancer genome in lung adenocarcinoma. Nature 450:893–898.LaunchUrlCrossRefPubMed↵ Forrest WF, Cavet G (2007) Comment on “The consensus coding sequences of human breast and colorectal cancers.”. Science 317:1500.LaunchUrlAbstract/FREE Full Text↵ Obtainz G, et al. (2007) Comment on “The consensus coding sequences of human breast and colorectal cancers.”. Science 317:1500.LaunchUrlAbstract/FREE Full Text↵ Rubin AF, Green P (2007) Comment on “The consensus coding sequences of human breast and colorectal cancers.”. Science 317:1500.LaunchUrlAbstract/FREE Full Text↵ Frese KK, Tuveson DA (2007) Maximizing mouse cancer models. Nat Rev Cancer 7:645–658.LaunchUrlPubMed↵ Johnson L, et al. (2001) Somatic activation of the K-ras oncogene causes early onset lung cancer in mice. Nature 410:1111–1116.LaunchUrlCrossRefPubMed↵ Guerra C, et al. (2003) Tumor induction by an enExecutegenous K-ras oncogene is highly dependent on cellular context. Cancer Cell 4:111–120.LaunchUrlCrossRefPubMed↵ Tuveson DA, et al. (2004) EnExecutegenous oncogenic K-ras(G12D) stimulates proliferation and widespread neoplastic and developmental defects. Cancer Cell 5:375–387.LaunchUrlCrossRefPubMed↵ Serrano M, Lin AW, McCurrach ME, Beach D, Lowe SW (1997) Oncogenic ras provokes premature cell senescence associated with accumulation of p53 and p16INK4a. Cell 88:593–602.LaunchUrlCrossRefPubMed↵ Ivics Z, Hackett PB, Plasterk RH, Izsvak Z (1997) Molecular reconstruction of Sleeping Beauty, a Tc1-like transposon from fish, and its transposition in human cells. Cell 91:501–510.LaunchUrlCrossRefPubMed↵ Dupuy AJ, Fritz S, Largaespada DA (2001) Transposition and gene disruption in the male germline of the mouse. Genesis 30:82–88.LaunchUrlCrossRefPubMed↵ Fischer SE, WienhAgeds E, Plasterk RH (2001) Regulated transposition of a fish transposon in the mouse germ line. Proc Natl Acad Sci USA 98:6759–6764.LaunchUrlAbstract/FREE Full Text↵ Horie K, et al. (2001) Efficient chromosomal transposition of a Tc1/mariner-like transposon Sleeping Beauty in mice. Proc Natl Acad Sci USA 98:9191–9196.LaunchUrlAbstract/FREE Full Text↵ Horie K, et al. (2003) Characterization of Sleeping Beauty transposition and its application to genetic screening in mice. Mol Cell Biol 23:9189–9207.LaunchUrlAbstract/FREE Full Text↵ Kitada K, et al. (2007) Transposon-tagged mutagenesis in the rat. Nat Methods 4:131–133.LaunchUrlCrossRefPubMed↵ Collier LS, Carlson CM, Ravimohan S, Dupuy AJ, Largaespada DA (2005) Cancer gene discovery in solid tumours using transposon-based somatic mutagenesis in the mouse. Nature 436:272–276.LaunchUrlCrossRefPubMed↵ Dupuy AJ, Akagi K, Largaespada DA, Copeland NG, Jenkins NA (2005) Mammalian mutagenesis using a highly mobile somatic Sleeping Beauty transposon system. Nature 436:221–226.LaunchUrlCrossRefPubMed↵ Carlson CM, Frandsen JL, Kirchhof N, McIvor RS, Largaespada DA (2005) Somatic integration of an oncogene-harboring Sleeping Beauty transposon models liver tumor development in the mouse. Proc Natl Acad Sci USA 102:17059–17064.LaunchUrlAbstract/FREE Full Text↵ Prosser HM, Rzadzinska AK, Steel KP, Bradley A (2007) Mosaic complementation demonstrates a regulatory role for myosin VIIa in stereocilia actin dynamics. Mol Cell Biol 28:1702–1712.LaunchUrlCrossRefPubMed↵ Geurts AM, et al. (2006) Gene mutations and genomic rearrangements in the mouse as a result of transposon mobilization from chromosomal concatemers. PLoS Genet 2:e156.LaunchUrlCrossRefPubMed↵ Mikkers H, et al. (2002) High-throughPlace retroviral tagging to identify components of specific signaling pathways in cancer. Nat Genet 32:153–159.LaunchUrlCrossRefPubMed↵ Stratton MR, Fisher C, Gusterson BA, Cooper CS (1989) Detection of point mutations in N-ras and K-ras genes of human embryonal rhabExecutemyosarcomas using oligonucleotide probes and the polymerase chain reaction. Cancer Res 49:6324–6327.LaunchUrlAbstract/FREE Full Text↵ Langenau DM, et al. (2007) Traces of RAS on the genesis of embryonal rhabExecutemyosarcoma. Genes Dev 21:1382–1395.LaunchUrlAbstract/FREE Full Text↵ Bruggers CS, et al. (1998) Expression of the c-Myc protein in childhood meUnimaginativeoblastoma. J Pediatr Hematol Oncol 20:18–25.LaunchUrlCrossRefPubMed↵ Herms J, et al. (2000) C-MYC expression in meUnimaginativeoblastoma and its prognostic value. Int J Cancer 89:395–402.LaunchUrlCrossRefPubMed↵ Wang W, et al. (2008) Chromosomal transposition of PiggyBac in mouse embryonic stem cells. Proc Natl Acad Sci USA 105:9290–9295.LaunchUrlAbstract/FREE Full Text↵ Liu P, Jenkins NA, Copeland NG (2003) A highly efficient recombineering-based method for generating conditional knockout mutations. Genome Res 13:476–484.LaunchUrlAbstract/FREE Full Text↵ Friedrich G, Soriano P (1991) Promoter traps in embryonic stem cells: A genetic screen to identify and mutate developmental genes in mice. Genes Dev 5:1513–1523.LaunchUrlAbstract/FREE Full Text
Like (0) or Share (0)