New genes involved in cancer identified by retroviral tagging

Retroviral insertional mutagenesis in BXH2 and AKXD mice induces a high incidence of myeloid leukemia and B- and T-cell lymphoma, respectively. The retroviral integration sites (RISs) in these tumors thus provide powerful genetic tags for the discovery of genes involved in cancer. Here we report the first large-scale use of retroviral tagging for cancer gene discovery in the post-genome era. Using high throughput inverse PCR, we cloned and analyzed the sequences of 884 RISs from a tumor panel composed primarily of B-cell lymphomas. We then compared these sequences, and another 415 RIS sequences previously cloned from BXH2 myeloid leukemias and from a few AKXD lymphomas, against the recently assembled mouse genome sequence. These studies identified 152 loci that are targets of retroviral integration in more than one tumor (common retroviral integration sites, CISs) and therefore likely to encode a cancer gene. Thirty-six CISs encode genes that are known or predicted to be genes involved in human cancer or their homologs, whereas others encode candidate genes that have not yet been examined for a role in human cancer. Our studies demonstrate the power of retroviral tagging for cancer gene discovery in the post-genome era and indicate a largely unrecognized complexity in mouse and presumably human cancer.     

High-throughput retroviral tagging to identify components of specific signaling pathways in cancer

Genetic screens carried out in lower organisms such as yeast, Drosophila melanogaster and Caenorhabditis elegans have revealed many signaling pathways. For example, components of the RAS signaling cascade were identified using a mutant eye phenotype in D. melanogaster as a readout. Screening is usually based on enhancing or suppressing a phenotype by way of a known mutation in a particular signaling pathway. Such in vivo screens have been difficult to carry out in mammals, however, owing to their relatively long generation times and the limited number of animals that can be screened. Here we describe an in vivo mammalian genetic screen used to identify components of pathways contributing to oncogenic transformation. We applied retroviral insertional mutagenesis in Myc transgenic (E mu Myc) mice lacking expression of Pim1 and Pim2 to search for genes that can substitute for Pim1 and Pim2 in lymphomagenesis. We determined the chromosomal positions of 477 retroviral insertion sites (RISs) derived from 38 tumors from E mu Myc Pim1(-/-) Pim2(-/-) mice and 27 tumors from E mu Myc control mice using the Ensembl and Celera annotated mouse genome databases. There were 52 sites occupied by proviruses in more than one tumor. These common insertion sites (CISs) are likely to contain genes contributing to tumorigenesis. Comparison of the RISs in tumors of Pim-null mice with the RISs in tumors of E mu Myc control mice indicated that 10 of the 52 CISs belong to the Pim complementation group. In addition, we found that Pim3 is selectively activated in Pim-null tumor cells, which supports the validity of our approach.     

MMTV insertional mutagenesis identifies genes, gene families and pathways involved in mammary cancer

We performed a high-throughput retroviral insertional mutagenesis screen in mouse mammary tumor virus (MMTV)-induced mammary tumors and identified 33 common insertion sites, of which 17 genes were previously not known to be associated with mammary cancer and 13 had not previously been linked to cancer in general. Although members of the Wnt and fibroblast growth factors (Fgf) families were frequently tagged, our exhaustive screening for MMTV insertion sites uncovered a new repertoire of candidate breast cancer oncogenes. We validated one of these genes, Rspo3, as an oncogene by overexpression in a p53-deficient mammary epithelial cell line. The human orthologs of the candidate oncogenes were frequently deregulated in human breast cancers and associated with several tumor parameters. Computational analysis of all MMTV-tagged genes uncovered specific gene families not previously associated with cancer and showed a significant overrepresentation of protein domains and signaling pathways mainly associated with development and growth factor signaling. Comparison of all tagged genes in MMTV and Moloney murine leukemia virus-induced malignancies showed that both viruses target mostly different genes that act predominantly in distinct pathways.     

Inactivation of the Wip1 phosphatase inhibits mammary tumorigenesis through p38 MAPK-mediated activation of the p16(Ink4a)-p19(Arf) pathway

Modulation of tumor suppressor activities may provide new opportunities for cancer therapy. Here we show that disruption of the gene Ppm1d encoding Wip1 phosphatase activated the p53 and p16 (also called Ink4a)-p19 (also called ARF) pathways through p38 MAPK signaling and suppressed in vitro transformation of mouse embryo fibroblasts (MEFs) by oncogenes. Disruption of the gene Cdkn2a (encoding p16 and p19), but not of Trp53 (encoding p53), reconstituted cell transformation in Ppm1d-null MEFs. In vivo, deletion of Ppm1d in mice bearing mouse mammary tumor virus (MMTV) promoter-driven oncogenes Erbb2 (also called c-neu) or Hras1 impaired mammary carcinogenesis, whereas reduced expression of p16 and p19 by methylation-induced silencing or inactivation of p38 MAPK correlated with tumor appearance. We conclude that inactivation or depletion of the Wip1 phosphatase with resultant p38 MAPK activation suppresses tumor appearance by modulating the Cdkn2a tumor-suppressor locus.     

Insertional mutagenesis identifies multiple networks of cooperating genes driving intestinal tumorigenesis

The evolution of colorectal cancer suggests the involvement of many genes. To identify new drivers of intestinal cancer, we performed insertional mutagenesis using the Sleeping Beauty transposon system in mice carrying germline or somatic Apc mutations. By analyzing common insertion sites (CISs) isolated from 446 tumors, we identified many hundreds of candidate cancer drivers. Comparison to human data sets suggested that 234 CIS-targeted genes are also dysregulated in human colorectal cancers. In addition, we found 183 CIS-containing genes that are candidate Wnt targets and showed that 20 CISs-containing genes are newly discovered modifiers of canonical Wnt signaling. We also identified mutations associated with a subset of tumors containing an expanded number of Paneth cells, a hallmark of deregulated Wnt signaling, and genes associated with more severe dysplasia included those encoding members of the FGF signaling cascade. Some 70 genes had co-occurrence of CIS pairs, clustering into 38 sub-networks that may regulate tumor development.     

Insertional mutagenesis in zebrafish rapidly identifies genes essential for early vertebrate development

To rapidly identify genes required for early vertebrate development, we are carrying out a large-scale, insertional mutagenesis screen in zebrafish, using mouse retroviral vectors as the mutagen. We will obtain mutations in 450 to 500 different genes--roughly 20% of the genes that can be mutated to produce a visible embryonic phenotype in this species--and will clone the majority of the mutated alleles. So far, we have isolated more than 500 insertional mutants. Here we describe the first 75 insertional mutants for which the disrupted genes have been identified. In agreement with chemical mutagenesis screens, approximately one-third of the mutants have developmental defects that affect primarily one or a small number of organs, body shape or swimming behavior; the rest of the mutants show more widespread or pleiotropic abnormalities. Many of the genes we identified have not been previously assigned a biological role in vivo. Roughly 20% of the mutants result from lesions in genes for which the biochemical and cellular function of the proteins they encode cannot be deduced with confidence, if at all, from their predicted amino-acid sequences. All of the genes have either orthologs or clearly related genes in human. These results provide an unbiased view of the genetic construction kit for a vertebrate embryo, reveal the diversity of genes required for vertebrate development and suggest that hundreds of genes of unknown biochemical function essential for vertebrate development have yet to be identified.     

AAV serotype 2 vectors preferentially integrate into active genes in mice

Recombinant adeno-associated virus serotype 2 (rAAV2) is a promising vector for gene therapy because it can achieve long-term stable transgene expression in animals and human subjects after direct administration of vectors into various target tissues. In the liver, although stable transgene expression primarily results from extrachromosomal vector genomes, a series of experiments has shown that vector genomes integrate into host chromosomes in hepatocytes at a low frequency. Despite the low integration efficiency, recent reports of retroviral insertional mutagenesis in mice and two human subjects have raised concerns about the potential for rAAV2-mediated insertional mutagenesis. Here we characterize rAAV2-targeted chromosomal integration sites isolated from selected or non-selected hepatocytes in vector-injected mouse livers. We document frequent chromosomal deletions of up to 2 kb at integration sites (14 of 14 integrations, 100%; most of the deletions were <0.3 kb) and preferred integration into genes (21 of 29 integrations, 72%). In addition, all of the targeted genes analyzed (20 of 20 targeted genes, 100%) were expressed in the liver. This is the first report to our knowledge on host chromosomal effects of rAAV2 integration in animals, and it provides insights into the nature of rAAV2 vector integration into chromosomes in quiescent somatic cells in animals and human subjects.     

Leukaemia disease genes: large-scale cloning and pathway predictions

Retroviral insertional mutagenesis in BXH2 and AKXD recombinant inbred mice induces a high incidence of myeloid or B- and T-cell leukaemia and the proviral integration sites in the leukaemias provide powerful genetic tags for disease gene identification. Some of the disease genes identified by proviral tagging are also associated with human disease, validating this approach for human disease gene identification. Although many leukaemia disease genes have been identified over the years, many more remain to be cloned. Here we describe an inverse PCR (IPCR) method for proviral tagging that makes use of automated DNA sequencing and the genetic tools provided by the Mouse Genome Project, which increases the throughput for disease gene identification. We also use this IPCR method to clone and analyse more than 400 proviral integration sites from AKXD and BXH2 leukaemias and, in the process, identify more than 90 candidate disease genes. Some of these genes function in pathways already implicated in leukaemia, whereas others are likely to define new disease pathways. Our studies underscore the power of the mouse as a tool for gene discovery and functional genomics.     

Large-scale discovery and genotyping of single-nucleotide polymorphisms in the mouse

Single-nucleotide polymorphisms (SNPs) have been the focus of much attention in human genetics because they are extremely abundant and well-suited for automated large-scale genotyping. Human SNPs, however, are less informative than other types of genetic markers (such as simple-sequence length polymorphisms or microsatellites) and thus more loci are required for mapping traits. SNPs offer similar advantages for experimental genetic organisms such as the mouse, but they entail no loss of informativeness because bi-allelic markers are fully informative in analysing crosses between inbred strains. Here we report a large-scale analysis of SNPs in the mouse genome. We characterized the rate of nucleotide polymorphism in eight mouse strains and identified a collection of 2,848 SNPs located in 1,755 sequence-tagged sites (STSs) using high-density oligonucleotide arrays. Three-quarters of these SNPs have been mapped on the mouse genome, providing a first-generation SNP map of the mouse. We have also developed a multiplex genotyping procedure by which a genome scan can be performed with only six genotyping reactions per animal.     

A mouse model of human L1 retrotransposition

The L1 retrotransposon has had an immense impact on the size and structure of the human genome through a variety of mechanisms, including insertional mutagenesis. To study retrotransposition in a living organism, we created a mouse model of human L1 retrotransposition. Here we show that L1 elements can retrotranspose in male germ cells, and that expression of a human L1 element under the control of its endogenous promoter is restricted to testis and ovary. In the mouse line with the highest level of L1 expression, we found two de novo L1 insertions in 135 offspring. Both insertions were structurally indistinguishable from natural endogenous insertions. This suggests that an individual L1 element can have substantial mutagenic potential. In addition to providing a valuable in vivo model of retrotransposition in mammals, these mice are an important step in the development of a new random mutagenesis system.     

Toward simpler and faster genome-wide mutagenesis in mice

Here we describe a practical Cre-loxP and piggyBac transposon-based mutagenesis strategy to systematically mutate coding sequences and/or the vast noncoding regions of the mouse genome for large-scale functional genomic analysis. To illustrate this approach, we first created loxP-containing loss-of-function alleles in the protocadherin alpha, beta and gamma gene clusters (Pcdha, Pcdhb and Pcdhg). Using these alleles, we show that, under proper guidance, Cre-loxP site-specific recombination can mediate efficient trans-allelic recombination in vivo, facilitating the generation of large germline deletions and duplications including deletions of Pcdha, and Pcdha to Pcdhb, simply by breeding (that is, at frequencies of 5.5%-21.6%). The same breeding method can also generate designed germline translocations between nonhomologous chromosomes at unexpected frequencies of greater than 1%. By incorporating a piggyBac transposon to insert and to distribute loxP sites randomly throughout the mouse genome, we present a simple but comprehensive method for generating genome-wide deletions and duplications, in addition to insertional loss-of-function and conditional rescue alleles, again simply by breeding.     

A systematic, genome-wide, phenotype-driven mutagenesis programme for gene function studies in the mouse

As the human genome project approaches completion, the challenge for mammalian geneticists is to develop approaches for the systematic determination of mammalian gene function. Mouse mutagenesis will be a key element of studies of gene function. Phenotype-driven approaches using the chemical mutagen ethylnitrosourea (ENU) represent a potentially efficient route for the generation of large numbers of mutant mice that can be screened for novel phenotypes. The advantage of this approach is that, in assessing gene function, no a priori assumptions are made about the genes involved in any pathway. Phenotype-driven mutagenesis is thus an effective method for the identification of novel genes and pathways. We have undertaken a genome-wide, phenotype-driven screen for dominant mutations in the mouse. We generated and screened over 26,000 mice, and recovered some 500 new mouse mutants. Our work, along with the programme reported in the accompanying paper, has led to a substantial increase in the mouse mutant resource and represents a first step towards systematic studies of gene function in mammalian genetics.     

Pten and p27KIP1 cooperate in prostate cancer tumor suppression in the mouse

The genetic bases underlying prostate tumorigenesis are poorly understood. Inactivation of the tumor-suppressor gene PTEN and lack of p27(KIP1) expression have been detected in most advanced prostate cancers. But mice deficient for Cdkn1b (encoding p27(Kip1)) do not develop prostate cancer. PTEN activity leads to the induction of p27(KIP1) expression, which in turn can negatively regulate the transition through the cell cycle. Thus, the inactivation of p27(KIP1) may be epistatic to PTEN in the control of the cell cycle. Here we show that the concomitant inactivation of one Pten allele and one or both Cdkn1b alleles accelerates spontaneous neoplastic transformation and incidence of tumors of various histological origins. Cell proliferation, but not cell survival, is increased in Pten(+/-)/Cdkn1b(-/-) mice. Moreover, Pten(+/-)/Cdkn1b(-/-) mice develop prostate carcinoma at complete penetrance within three months from birth. These cancers recapitulate the natural history and pathological features of human prostate cancer. Our findings reveal the crucial relevance of the combined tumor-suppressive activity of Pten and p27(Kip1) through the control of cell-cycle progression.     

Engineering a mouse balancer chromosome.

Balancer chromosomes are genetic reagents that are used in Drosophila melanogaster for stock maintenance and mutagenesis screens. Despite their utility, balancer chromosomes are rarely used in mice because they are difficult to generate using conventional methods. Here we describe the engineering of a mouse balancer chromosome with the Cre-loxP recombination system. The chromosome features a 24-centiMorgan (cM) inversion between Trp53 (also known as p53) and Wnt3 on mouse chromosome 11 that is recessive lethal and dominantly marked with a K14-Agouti transgene. When allelic to a wild-type chromosome, the inversion suppresses crossing over in the inversion interval, accompanied by elevated recombination in the flanking regions. The inversion functions as a balancer chromosome because it can be used to maintain a lethal mutation in the inversion interval as a self-sustaining trans-heterozygous stock. This strategy can be used to generate similar genetic reagents throughout the mouse genome. Engineering of visibly marked inversions and deficiencies is an important step toward functional analyses of the mouse genome and will facilitate large-scale mutagenesis programs.     

Nested chromosomal deletions induced with retroviral vectors in mice

Chromosomal deletions, especially nested deletions, are major genetic tools in diploid organisms that facilitate the functional analysis of large chromosomal regions and allow the rapid localization of mutations to specific genetic intervals. In mice, well-characterized overlapping deletions are only available at a few chromosomal loci, partly due to drawbacks of existing methods. Here we exploit the random integration of a retrovirus to generate high-resolution sets of nested deletions around defined loci in embryonic stem (ES) cells, with sizes extending from a few kilobases to several megabases. This approach expands the application of Cre-loxP-based chromosome engineering because it not only allows the construction of hundreds of overlapping deletions, but also provides molecular entry points to regions based on the retroviral tags. Our approach can be extended to any region of the mouse genome.     

Candidate lung tumor susceptibility genes identified through whole-genome association analyses in inbred mice

We performed a whole-genome association analysis of lung tumor susceptibility using dense SNP maps ( approximately 1 SNP per 20 kb) in inbred mice. We reproduced the pulmonary adenoma susceptibility 1 (Pas1) locus identified in previous linkage studies and further narrowed this quantitative trait locus (QTL) to a region of less than 0.5 Mb in which at least two genes, Kras2 (Kirsten rat sarcoma oncogene 2) and Casc1 (cancer susceptibility candidate 1; also known as Las1), are strong candidates. Casc1 knockout mouse tumor bioassays showed that Casc1-deficient mice were susceptible to chemical induction of lung tumors. We also found three more genetic loci for lung adenoma development. Analysis of one of these candidate loci identified a previously uncharacterized gene Lasc1, bearing a nonsynonymous substitution (D102E). We found that the Lasc1 Glu102 allele preferentially promotes lung tumor cell growth. Our findings demonstrate the prospects for using dense SNP maps in laboratory mice to refine previous QTL regions and identify genetic determinants of complex traits.     

Genome-wide, large-scale production of mutant mice by ENU mutagenesis

In the post-genome era, the mouse will have a major role as a model system for functional genome analysis. This requires a large number of mutants similar to the collections available from other model organisms such as Drosophila melanogaster and Caenorhabditis elegans. Here we report on a systematic, genome-wide, mutagenesis screen in mice. As part of the German Human Genome Project, we have undertaken a large-scale ENU-mutagenesis screen for dominant mutations and a limited screen for recessive mutations. In screening over 14,000 mice for a large number of clinically relevant parameters, we recovered 182 mouse mutants for a variety of phenotypes. In addition, 247 variant mouse mutants are currently in genetic confirmation testing and will result in additional new mutant lines. This mutagenesis screen, along with the screen described in the accompanying paper, leads to a significant increase in the number of mouse models available to the scientific community. Our mutant lines are freely accessible to non-commercial users (for information, see http://www.gsf.de/ieg/groups/enu-mouse.html).     

A gene-driven approach to the identification of ENU mutants in the mouse

The construction of parallel archives of DNA and sperm from mice mutagenized with ethylnitrosurea (ENU) represents a potentially powerful and rapid approach for identifying point mutations in any gene in the mouse genome. We provide support for this approach and report the identification of mutations in the gene (Gjb2) encoding connexin 26, using archives established from the UK ENU mutagenesis program.