Genomic Analysis of Mouse Tumorigenesis by Mandy Chi-Mun Tam B.Sc. Biochemistry University of Toronto, 2000 Submitted to the Department of Biology in Partial Fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology September 2006 ? Massachusetts Institute of Technology All rights reserved. Signature of author: Department of Biology August 15, 2006 /17 Certified by: Tyler Jacks Professor of Biology Thesis Supervisor Accepted by: MASSACHUSETTSNS OF TECHNOLOGY SEP 1 8 2006 LIBRARIES -I Stephen Bell Professor of Biology Chairman, Biology Graduate Committee ARCHIVES ~t/,b$ 74 IWI / 01 .- rl Genomic Analysis of Mouse Tumorigenesis by Mandy Chi-Mun Tam Submitted to the Department of Biology on August 21, 2006 in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy in Biology ABSTRACT The availability of the human and mouse genome sequences has spurred a growing interest in analyzing mouse models of human cancer using genomic techniques. Comparative genomic studies on mouse and human tumors can be valuable in two major ways: in validating mouse models and in identifying genes that are common to mouse and human tumorigenesis. Many analytic tools have emerged in recent years for human genome mining. Some of these tools have been translated to the murine versions. The work in this thesis described the application of two new whole-genome analytic techniques to study mouse tumorigensis: Representational Oligonucleotide Microarray Analysis (ROMA) for tumor DNA copy number asessment and single nucleotide polymorphism (SNP) genotyping using the SNaPshot'" system (Applied Biosystems) to detect loss of heterozygosity (LOH) in mouse tumors. The murine version of ROMA was tested on DNA from early-stage KrasG1 2 D-derived lung cancers and metastatic retinoblastoma in mice with retinal-specific Rb and p130 deletions. We were interested in identifying the additional genetic lesions that got positively selected during tumorigenesis of these mice. Several recurrent chromosomal copy number gains and losses were observed in the DNA of KrasG1 2 D-derived lung tumors. In addition, a focal amplification of the murine N-Myc locus was detected in the metastatic retinoblastomas, demonstrating the capability of ROMA to detect copy number changes at a single-gene resolution. For genome-wide allelotyping, a panel of 147 mouse SNPs were individually validated in 129S4/SvJae vs. C57BL/6J strains and were chosen as markers in the genotyping panel. We worked out a multiplex protocol to genotype the SNPs in an efficient manner. Through this protocol, we generated low-density global LOH maps of lung tumors from mice expressing KrasG 12 D. LOH that spanned entire chromosomes was seen in a subset of the tumors. A loss of the wild-type p53 allele was also observed in some cases. Thesis Supervisor: Tyler Jacks, Ph.D. Position: Professor of Biology ACKNOWLEDGEMENTS The biggest reward in graduate school was the opportunity to meet many driven, bright, and interesting people. It is from them I learned much about science and about life. Biology is a study of life after all. I'd like to thank my advisor, Tyler Jacks, who guided my journey with much patience and understanding. It was a privilege to be able to learn from someone who leads with scientific insights, wisdom, and compassion. I am thankful to my thesis committee members and co-workers for their scientific advices. I was also fortunate to have collaborators who made this thesis work possible. I especially want to thank Alejandro Sweet-Cordero, David MacPherosn, David Mu, David Ginzinger, and all the technicians at the UCSF cancer center genome core for their assistance and guidance. I am grateful to have a group of classmates, roommates and friends who pushed my limit, kept me grounded, and made me laugh. I truly believe every person I met in this journey has made an impact on me. In particular, I'd like to thank my two classmates and good friends Ishara Mills and Mark Rosenzweig for their support and help in putting this thesis together and throughout graduate school. Finally, I want to acknowledge my parents, past teachers, and college friends, who shaped or shared my values, and supported my decision to pursue graduate study. Table of Contents A bstract ............................................................................ ....................................... 2 A cknow ledgem ents........................................................................................................2 Chapter 1: Introduction--Analysis of Genetic Alterations in Cancer ......................... 6 Viewing cancer from an evolutionary perspective............................... ............... Mutations as underlying cause of cancer................................................................... 8 Early cancer gene discovery tools................................... ......................................... 10 Reflections from past studies ..................................................................... 10 Evolving technologies to analyze the cancer genome.................................................. 12 Comparative Genome Hybridization (CGH) ....................................... ........ 12 CGH platforms ............................................................ 12 A pplications in cancer research.............................................................................. 15 The case for genome complexity reduction ...................................... ........ 16 A pplication considerations.................................................................................. 18 Single nucleotide polymorphism (SNP) genotyping................................................... 19 Applications in cancer research............................................... 19 SNP genotyping approaches ..................................... ................ 22 D igital K aryotyping............................... ...................................................... ........ 33 Emerging Approaches to analyze the Sequenced Genome ..................................... 34 G enom e resequencing............................................. .............................................. 34 Comparative Cancer Genomics.............................................. .................... 35 M ouse cancer genom ics .................................................................................... 35 Dog cancer genomics........................................................41 O utlook .............................................................................................. ..... ......... 42 Cancer genome sequencing................................................... ....................... 42 T hesis Scope ................................................................................ ........................... 43 R eferences.................................................................................... ......................... ... 44 Chapter 2: Mouse Representational Oligonucleotide Microarray Analysis Detects Copy Number Alterations in Murine Tumors of the Lung and Retina .................... 68 A bstract.......... . ..................... ......................................................... ........................... 68 R esults ..................................................................................... ............................... 7 1 Detection of N-Myc amplification in mouse retinoblastomas .................................. 74 Detection of recurrent whole chromosomal changes in primary lung tumors.............. 78 Common deletion of Csmdl in mouse and human lung cancer cell lines................. 86 D iscussion................................................................................ ............................... 89 On the technical capabilities of mouse ROMA platform .................................... .89 On the genetics of retinoblastomas ..................................... ............. 91 On the genetics of lung adenocarcinomas ......................................... ......... 92 p53 and genetic changes in Kras-initiated lung cancer model ................................. 93 Whole chromosomal copy changes ..................................................................... 93 Sub-chromosomal copy changes ................................................................. 96 M aterials and M ethods................................ ............................................................. 97 A cknow ledgem ents ................................................................................................... 99 R eferences........................................................................................................ ........ 99 Chapter 3: A Multiplexed SNaPshot TM Protocol for Genome-wide Mouse SNP Genotyping and Its Applifcation to Detect Loss-of-heterozygosity in Murine Lung Tumors ...................... ........................................................................................... 104 A bstract....................................................................................................................... 105 R esults ........................................................................................................................ 108 Validation of SNPs in 129S4/SvJae vs. C57BL/6J strains ........................................ 108 Establishment of a protocol for genome-wide multiplex SNP typing in the mouse... 109 Implementation of SBE assay to identify LOH regions in mouse lung tumors.......... 118 Loss of wild-type p53 on chromosome 11 .................................. 128 Reduced level of wild-type Kras on chromosome 6 ................................................. 128 Comparison of LOH results to copy number data................................. 129 D iscussion ................................................................................................................... 135 Using SNPs as markers for genome-wide LOH screen in mouse cancer................ 135 Implications on lung cancer genetics ..................................... 136 Combining LOH and CGH data................................. 136 Chromosome 11 in mouse lung tumorigenesis ..................................... 138 Note on genetic background............................... 139 Concluding note ........................................ 140 Materials and Methods ........................................ 140 Acknowledgements ........................................ 143 R eferences................................................................................................................... 144 Supplemental Information ....................................... 150 Chapter 4: Final Discussion ..................................... 169 On Technology........................................................................................................ 170 O n B iology.............................................................................................................. 173 References ...................... ....................................................................................... 182 Chapter 1 Introduction: Analysis of Genetic Alterations in Cancer Viewingt cancer from an evolutionary perspective Beneath the diversity of life is the dynamic process of change that alters chances for living. Some changes are ephemeral while others, such as those in the form of genetic mutation, can last for generations. Over time, only heritable traits can endure the test of natural selection, in which mutation is the key engine. The result is variations among life forms: the different species and the different individuals within a species. Some will thrive and some will not. The disease cancer has diverse manifestations: in its onset, pathology, malignancy and therapeutic response. This variation can probably be explained most fittingly from a Darwinian perspective. Played out by evolutionary rules, mutation and selection operate in a micro-scale to govern cancer development in the body. Each cancer is a clone of mutated cells selected for the ability to multiply and grow in an environment that is short of nutrients, oxygen and other survival factors. As each clone proliferates, successive mutations continue to create subclones with enhanced growth advantages. Neoplastic cells progress through malignant stages in these waves of clonal expansion. It is the inherent randomness of mutation that gives cancer its heterogeneous and often unpredictable expressions. Given the intrinsic mutability of the genome, cancer is perhaps an inevitable baggage of species evolution. Retention and refinement of genes that regulate cell growth, cell fate, cell migration, and cell-cell interaction allow our species to exist and flourish (Hanahan and Weinberg 2000). The same genetic networks, however, when deregulated by the inexorable force of mutation, have the potential to destroy an individual in the form of cancer. Cancer is essentially an uninvited hitchhiker in the evolutionary probability game. Cancer research aims to decode the micro-evolutionary game rules and to develop strategies to play against the odds. The analysis of mutations in tumors is fundamental to our understanding of the elements of the disease. The past few decades mark important developments in our understanding of molecular biology and technical ability to perform genetic analyses. Numerous oncogenes and tumor suppressor genes have been identified. Some become critical targets for therapy. With the recent emergence of genomics, cancer genetics has moved from single gene analyses to whole-genome diagnoses of the multiple mutations that exist. This development has allowed researchers to more fully capture the variable genetic landscape that is characteristic of cancer. This thesis describes efforts to identify genetic alterations in mouse models of human cancer by implementing some of the latest techniques in genome analysis. The analysis focuses on lung cancer, the current leading cause of cancer deaths worldwide, and touches on retinoblastoma, a childhood eye cancer. This first chapter will introduce readers of this thesis to various cancer genome analytic tools. Chapters 2 and 3 will summarize results generated by two complementary genomic techniques in the study of mouse tumorigenesis. Finally, the implications of this research will be discussed in the concluding chapter. Mutations as underlying cause of cancer While the concept that cancer is caused by underlying mutations is well known today, it is interesting to look back at how this notion arrived. Mankind recognized and named the disease cancer over two thousand years ago. Prior to the emergence of experimental medicine, the cause of cancer was widely believed to be an imbalance of bodily humors. The premise was replaced in the 18 h century by the lymph theory, which assumed cancer grew out of abnormal lymph fluid. In 1890, David von Hansemann made the first report of abnormal mitoses in tumors, suggesting a genetic cause of cancer (Shimkin 1977). In 1914, Theodor Boveri postulated the somatic mutation theory, which identified chromosomal abnormalities as possible culprits that caused cells to adopt cancerous properties (Manchester 1995). The theory, however, remained a conjecture for a few more decades due to limits in cellular and molecular genetics techniques at the time. Other prevailing ideas from the same period proposed cancer originated from trauma, viruses, or environmental factors. The trauma theory got disproved in the late 1920's; subsequently, researchers showed that cancer viruses disrupted expression of genes, while radiation and numerous chemical carcinogens were found to act by inducing mutations in the genome (Shimkin 1977). In 1960, a major step in verifying Boveri's mutation hypothesis came when Nowell and Hungerfold discovered in patients with chronic myeloid leukemia (CML) the Philadelphia chromosome, the first recurrent chromosomal abnormality associated with a cancer (Nowell and Hungerford 1960). In other cancers, while the probabilistic nature of mutation has often made it difficult to isolate specific genetic aberrations, retrospective surveys of data have eventually revealed a non-random pattern of genetic changes in cancers of most organs. The notion that cancer evolves through a selection of particular mutated genetic elements has gotten accepted beyond doubt. Some of these genetic constituents were later experimentally identified and broadly classified as oncogenes, for genes that enhance proliferation, growth, and differentiation when activated, and tumor suppressor genes, for genes that lose their normal regulatory roles when inactivated in cancer. Early cancer gene discovery tools Methods to identify disease genes fall into three categories: cytogenetic mapping, physical mapping, and linkage mapping. In the early days of cancer research, cytogenetics was the most feasible way to study genomic alteration in cells under a microscope. Chromosome banding techniques have enabled the discovery of numerous cancer-related chromosomal aberrations (Mitelman 2000). Recurrent translocations have been particularly useful in uncovering numerous oncogenes such as Abl in BCR-Abl and c-Myc in IgG-cMyc fusions of CML and Burkitt's lymphoma. Functional genomic screens such as the in-vitro transformation assay also helped to identify genomic DNA fragments containing oncogenes (Shih et al. 1981). Early genetic maps with restriction fragment length polymorphism (RFLP) and simple sequence length polymorphism (SSLP) markers have allowed the discovery of tumor suppressor genes through linkage mapping in familiar cancer syndromes and loss of heterozygosity (LOH) mapping in tumors of individuals. Reflections from past studies A recent review has summarized the many cancer gene mutations known today (Futreal et al. 2004). In that summary, it is clear that cancer genetic alterations appear in virtually all types but chromosomal aberrations are the predominance. Among the recurrent chromosomal changes, the target genes involved still remain to be identified. The Mitelman database is a catalog of cancer chromosomal aberrations (Mitelman et al. 2006). As of March, 2006, chromosomal aberrations have been documented in over 49,000 cancer samples in the forms of monosomy, trisomy, balanced and unbalanced translocations, and amplifications and deletions in various scales. In the numerous reports without observed chromosomal alterations, mutations at the nucleotide level such as microsatellite and other repeat instabilities and point mutations are often seen. While carcinomas constitute over 80% of all human cancers, carcinoma-related genetic alterations only represent 20-30% of the cataloged cases in both databases (Futreal et al. 2004; Mitelman et al. 2006). It is appearing that the many classes of mutations in cancer may require complementary approaches for discovery, and that many more cancer genes, especially those involved in the carcinoma development, remain to be discovered. With the availability of the human genome sequence, various new analytical tools have been developed to characterize the cancer-related genetic changes at a genomic level. The tremendous amount of information within a genome is best mined with tools with high throughput and resolution, the two grounds on which new technologies are competing. Evolving technologies to analyze the cancer genome Comparative Genome Hybridization (CGH) CCH involves simultaneous hybridization of differentially labeled test and reference DNA to obtain relative copy number information along a chromosomal position coordinate. Copy number alterations can reflect aneuploidy, unbalanced translocations, gene amplifications or deletions. Prior to genome sequencing, chromosomal positions were cytogenetically traced to bands on normal metaphase spreads. Genome sequencing has allowed the use of arrayed DNA stretches with known positions as mapping coordinates, thus raising the cap of resolving power (Albertson and Pinkel 2003). This section will summarize the basics and utilities of CGH in cancer genomics research. CGHplatforms Arrays of large-insert clones Initial array CGH platforms utilized large-insert genomics clones as the probing elements. These include BAC (Bacterial Artificial Chromosomes), YAC (Yeast Artificial Chromosomes), PAC (P1 Artificial Chromosomes), and cosmids (Solinas-Toldo et al. 1997; Pinkel et al. 1998). The initial study by Pinkel et al. (1998) showed the ability of BAC microarrays to detect difference in X-chromosome number in sex mistached samples and known copy number increases on chromosome 20 in breast cancer cell lines. More importantly, they were able to detect a novel deletion on 20q chromosome arm, showcasing the potential of BAC arrays to find changes that were missed by older methods. Array spotted with ~2400 BAC and P1 clones with an average ~1Mb resolution for the human genome was reported in (Snijders et al. 2001). To achieve better resolution, a tiling array with 32433 overlapping BAC clones was constructed (Ishkanian et al. 2004). A sub-Mb resolution was reported, allowing detection of a 300kb validated amplicon on 13q12.2 in a colorectal cell line and a 240kb validated deletion in a breast cancer cell line. While the power of such tiling BAC array is clear, the manufacturing of such array requires large-scale management of clones, making the technology hard for most labs to adopt. PCR strategies including degenerate oligonucleotide PCR (DOP-PCR) and ligation-mediated PCR have been successfully employed to amplify BACs before spotting (Hodgson et al. 2001). Arrays of cDNA Microarrays originally used for genome-wide gene expression analysis have been used for CGH (Pollack et al. 1999). Because genomic DNA is more complex than RNA representations, CGH requires a platform with a higher minimum sensitivity than one for gene expression analysis. The feasibility of cDNA and EST-based CGH was proved by Pollack et al. in 1999. Using Turner syndrome patient and sex-mismatched samples, they were able to detect gene copy number differences on X and Y chromosomes. In addition, they confirmed the known ERBB2 and MYC amplifications and p53 deletion in established cell lines. At a genomic level, they observed 62% highly amplified genes also have elevated expression, backing the case for a gene dosage effect on gene expression. However, the signal:noise ratio was poor, making averaging of signals over several probes necessary, which essentially limited resolution. Arrays of oligonucleotides Oligonucleotides can also be used as arraying elements. This is made possible by methods that allow precise in-situ synthesis of oligonucleotides include photolithography (Affymetrix, Nimblegen) and inkjet printing (Agilent). Such printing technologies have also made the manufacturing of oligonucleotide arrays scaleable. Theoretically, oligonucleotide array probes can be designed for any sequence, enabling high-density coverage of any region of interest. In practice, resolution is a function of signal to noise. Because of their short lengths, hybridization signal on each oligonucleotide is generally lower than on a BAC probe. As in cDNA arrays, averaging of signals over a few adjacent probes is necessary to obtain reliable calls, which in effect decreases resolution. One of the first successful implementations of oligonucleotide arrays was reported by (Barrett et al. 2004). The investigators successfully performed genome-wide CGH on Agilent Technology's -~17k oligonucleotide array originally designed for gene expression analysis. In addition, using probes designed specifically for CGH on chr x, 18, and 17 they were able to detect single copy change as verified in known chromosome x number difference in various cell lines. Later, a whole-genome array made by Agilent was reported to achieve a -70kb average resolution (Brennan et al. 2004). Affymetrix also has developed oligonucleotide arrays for CGH. Using ASO arrays designed for SNP detection, LOH and copy number changes could be used detected simultaneously (Bignell et al. 2004; Zhao et al. 2004). While most other platforms involve co- hybridization of tumor vs. normal DNA on each array, the Affymetrix chip is a single- channel array, requiring hybridization of a set of normal samples separately as reference. Non-commercially made oligonucleotide CGH arrays have also been reported(Carvalho et al. 2004; van den Ijssel et al. 2005). Recently, two oligonucleotide arrays were made and tested by Nimblegen: 1) a genome-wide array with 6kb median spacing and 2) a tiling oligonucleotide array covering selected regions at a sub-kb interval (Selzer et al. 2005), further demonstrating the power of oligonucleotide CGH arrays. Arrays with other elements ssDNA has been reported as probes in exon array CGH (Dhami et al. 2005). As a proof of principle, Dhami et al. tested an array with ssDNA probes to 162 exons for 5 human genes. Compared to arrays made with other probing elements, the ssDNA array had an enhanced signal: noise ratio and thus have a sensitivity of -2x higher. It remains to be seen whether this method can be scaled to cover the whole genome. Applications in cancer research Microarray formats provide the high-throughput capabilities for genome-wide studies. Some of the pioneering cancer studies are discussed above. Genome-wide CGH analysis has become increasingly routine in cancer research. A recent study by Tonon et al. on lung cancer showcased the power of array CGH in detecting disease loci. Using oligonucleotide and cDNA arrays made by Agilent to characterize human lung cancer samples and cell lines, numerous recurrent focal copy number changes could be seen, some were below 0.5Mb (Tonon et al. 2005). The authors' comparison of the CGH data with a gene expression analysis enabled them to further pinpoint critical candidates, which included p63 in squamous cell carcinoma. Beyond the use of CGH data in candidate gene discovery, it is worth mentioning its application in tumor classification, as has been done to distinguish renal cancer (Wilhelm et al. 2002) and multiple myeloid subtypes (Carrasco et al. 2006). The case for genome complexity reduction The human genome is very complex: over 50% of the genome sequence is comprised of repeats whereas only ~5% is protein coding (Lander et al. 2001). The abundance of these repeats, plus low-level shared sequences such as those by gene family members, pose a challenge for specific hybridization. Many hybridization-based genetic analysis protocols employ a genome complexity reduction step. In the classic case of Southern blotting, genomic DNA is fragmented and separated by electrophoresis prior to hybridization. For genomic analysis, PCR based methods are used. One is to degenerate oligonucleotides to random prime and amplify the genome (Telenius et al. 1992; Kuukasjarvi et al. 1997). The alternative method is restriction-enzyme based, which is more reproducible (Lucito et al. 1998). In the latter case, the resulting complexity of the representation can be controlled by the choice of restriction enzyme; a less frequent cutter would generate a less complex representation. Theoretically, a low-complexity representation (LCR) would improve hybridization by enhancing signal: noise ratio and reducing hybridization times. A protocol of preparing LCR was described in (Baldocchi and Flaherty 1997). The method involves digestion of test and reference genomic DNA with BglII, a 6bp cutter, and linker-based PCR-amplification. As PCR selectively amplifies smaller restriction fragments, only a ~2.5% low-complexity representation of the genome is made by this process. The usefulness of LCRs in genome-wide copy number analysis was first demonstrated in subtractive hybridization, in which a complexity reduction step is essential for mammalian genome analyses; the method of LCR subtraction is called Representational Difference Analysis (RDA) (Lisitsyn and Wigler 1995). RDA of tumor and normal genomes have been successfully performed to identify copy number alterations, including PTEN loss in multiple cancers (Lisitsyn et al. 1995; Li et al. 1997; Hamaguchi et al. 2002; Mu et al. 2003). Based on these successes, the next conceivable step was to test the use of LCR in array-based hybridization. ROMA -an alternative way to perform genome-wide CGH analysis Representational Oligonucleotide Microarray (ROMA) involves hybridization of LCRs on an array for copy number comparison. In a pilot experiment, 1-2K oligonucleotide arrays for used to probe for BglII representations (Lucito et al. 2000). They were able to demonstrate the preservation of original gene ratios in the LCRs, and that results were reproducible and had good signal to noise ratio. In addition, in the making of LCRs, small amount of samples could be amplified and that polymorphism at restriction site could be reflected in parallel. In a later experiment, a denser microarray (Nimblegen) with 80K oligonucleotide probes was tested (Lucito et al. 2003). An average genome-wide resolution of 30kb was achieved, and in some area, the resolution was as high as 15kb (Lucito et al. 2003). ROMA thus provides a high-resolution alternative to perform CGH studies (Jobanputra et al. 2005). Application considerations In choosing an array platform for CGH study, several factors need to be considered: the quantity and quality of DNA available, the platform performance and the cost and accessibility of the platform. In CGH, there is no need for culturing cells to prepare karyotypes as required for cytogenetic analysis, suggesting even DNA from archival tissues can be analyzed. Quantity and quality of DNA will determine what platforms can be used. BAC arrays can be performed with ~300ng of DNA while cDNA and oligonucleotide arrays require a few micrograms of starting materials, although whole- genome amplification can be performed at stake of an increased experimental cost (Lage et al. 2003). In addition, suboptimal quality of DNA such as those obtained from archival tissues may be incompatible with some protocols such as ROMA. It is often debated which array CGH platform will dominate the future. Resolution is likely the most important determining factor. Other factors such as manufacturability will also affect dominance. On those two grounds, Ylstra et al. recently suggested the future will likely belong to oligonucleotide arrays (Ylstra et al. 2006). Whether their prediction is correct or not, the increasingly powerful array CGH technology overall will likely yield many exciting data in the cancer genomics field. Single nucleotide polymorphism (SNP) genotyping While CGH enables high-resolution physical mapping of copy number alterations in cancer, allelic mapping with genetic markers can also be informative. Genome sequencing has led to the discovery of a vast number of SNPs. To date, over 12 million SNPs in the human genome are referenced in the National Center for Biotechnology Information (NCBI) dsSNP database build 126 (http://www.ncbi.nlm.nih.gov/projects/SNP/). The dense coverage has made SNPs the molecular marker of choice in genetic mapping. The following section will discuss the use of SNP genotyping in cancer research, in genome-wide and regional candidate mapping studies. A description in the different genotyping methods will then follow. Applications in cancer research Genome-wide loss of heterozygosity (LOH) mapping LOH maps can be used to infer tumor suppressor gene locations. In humans, most LOH events are not to be associated with copy number changes and would be missed by CGH (Huang et al. 2004; Beroukhim et al. 2006). LOH can be identified by analyzing SNPs within the tumor and normal DNA of the same individual. Early efforts to use SNPs for high-density LOH mapping included studies on lung cancer, neurofibromatosis type 2, and esophageal adenocarcinomas (Lindblad-Toh et al. 2000a; Mei et al. 2000). Regions of allelic imbalance could be identified in each case. Recently, using SNP arrays as a detection platform (described in a later section) and algorithms to integrate signal intensities to allelic calls, copy number information can be viewed along the LOH data, allowing the distinction of copy neutral LOH, copy number loss associated LOH, or copy number gain associated allelic imbalances (Bignell et al. 2004; Zhao et al. 2004; Zhao et al. 2005). The power of bioinformatics has also allowed LOH regions to be inferred from tumor samples without paired normal DNA using high-density SNP genotyping data (Beroukhim et al. 2006); such would be impossible without a high-resolution mapping tool. Aside from identifying candidate tumor suppressor gene region, LOH data can also be used for tumor classification, as shown by the LOH-based clustering of non-small cell lung cancer vs. small cell lung cancer samples (Janne et al. 2004). Genome-wide mapping of cancer susceptibility loci in population studies Aside from LOH mapping, SNP genotyping is also a powerful tool in linkage or linkage disequilibrium studies that attempt to map disease genes using a family-based approach. Genome-wide mapping can be enhanced by the availability of high-throughput, high- density SNP genotyping techniques. A traditionally popular marker set for mapping is the ABI Prism Mapping Set (Applied Biosystems), which consists of microsatellite markers at 10cM apart. Genome-wide SNP mapping panel such as the Affymetrix GeneChip (discussed below) allows simultaneous anlaysis of 10OK SNPs, which means markers are spaced at 0.34cM on average. A linkage study on prostate cancer has compared the used of SNPs and microsatellite markers in mapping(Schaid et al. 2004). The denser SNP markers indeed allowed the authors to get a better linkage resolution and identify more linkage peaks on multiple chromosomes. Genome-wide SNP based linkage studies have also been performed in other familial cancer cases including hereditary mixed polyposis syndrome (HMPS) (Cao et al. 2006) and chronic lymphocytic leukemia (CLL) (Sellick et al. 2005). For non-familiar cases, studies done by genome-wide SNP mapping of disease susceptibility loci have been performed on breast cancer (Ellis et al. 2006), Bloom syndrome and hereditary nonpolyposis colorectal cancer (HNPCC) (Mitra et al. 2004). Candidate loci were identified in each case; in the Bloom syndrome study, a single locus TSC0754862 was pinpointed (Mitra el al. 2004). Narrowing down candidate regions with SNPs Aside from genome-wide mapping, the high density of SNPs makes them the ideal genetic marker for narrowing down candidate regions. One candidate region that remains to be delineated is chromosome 3p, where LOH in multiple areas have been implicated in various cancers including over 90% of all lung cancers (Zabarovsky et al. 2002). While most older studies have approached the delineation problem with microsatellite markers, attempts to identify candidate tumor suppressor genes on chromosome 3p by SNP-based LOH mapping have begun (Tai et al. 2006). A linkage study that aims to scale down candidate tumor susceptibility region on 3p has also been performed on prostate cancer patients (Larson et al. 2005). In addition to delineating large chromosomal regions, gene specific studies to associate particular SNPs with a functional role in cancer have also proved to be informative. One recent study of such kind was performed on CHK2 in order to identify SNPs that can affect breast cancer susceptibility (Einarsdottir et al. 2006). SNP genotyping approaches The potential of exploiting SNPs as markers has stimulated a multitude of imaginative approaches to genotype them (Syvanen 2001; Engle et al. 2006). The choice of a genotyping protocol depends largely on the need of the research and the resources available. As discussed above, various types SNP-based studies can be performed in cancer research. A genome-wide study will put high throughput as top priority while a regional study will benefit from a highly flexible assay in order to genotype the SNPs of choice. The following section is a discussion on the scientific principles behind current SNP genotyping methods. Accuracy and robustness of an assay depend largely on the underlying reaction biochemistry. Then, the format and readout of the assay will determine what instruments are necessary and thus affect the ease of use, throughput, and cost. My discussion will be therefore divided into these two parts. In practice, because of the immense value of SNP genotyping in biomedical research, many assays have been commercialized. Table 1 is a summary of the working principles behind some commercial assays. Table 1: Decoding commercial buzzwords in SNP eenotvping Assay tradename GeneChip Genorama GoldenGate Invader iPLEX LightCycler MassEXTEND PinPoint SNaPshot SNPlex SNPstream Company Affymetrix Asper Illumina Third Wave Technology Sequenom Roche Sequenom Applied Biosystems Applied Biosystems Applied Biosystems Orchid/Beckman Working Priniciples* ASO array hybridization APEX OLA+ASO+microbeads on microarray Invasive cleavage SBE+MALDI-TOF FRET primer extension+MALDI-TOF primer extension+MALDI-TOF SBE+capillary electrophoresis OLA+capillary electrophoresis SBE+microarray * ASO=allele-specific oligonucleotides APEX=array-based primer extension OLA=oligonucleotide ligation assay SBE=single-based-extension FRET=fluorescent resonance energy transfer MALDI-TOF=-matrix assisted laser desorption ionization time of flight Hybridization Allele-specific oligonucletide (ASO) In differentiating SNP alleles, two ASO probes are used, each with a different allele of the SNP that is usually in the central region. Probe binds stably to the match allele but less so to the mismatch (Saiki et al. 1988). Enzymatic approaches Restriction enzyme The sensitivity of restriction endonucleases to distinguish short and defined sequences can be exploited. Restriction site length polymorphism (RFLP) is the historical method for genotyping SNP. At where a SNP changes a restriction enzyme site, a different digestion pattern can be seen(Syvanen 2001). Allele-specific amplification (ASA) or Primer extension This is a DNA polymerase-based method using two probes with a discriminating base at or near the 3'ends. When primers match the target, Taq polymerase catalyzed extension can occur (Waterfall and Cobb 2001). Single-base extension (SBE) In SBE or mini-sequencing, a primer is designed to anneal immediately upstream to the base of the SNP. A polymerase reaction is performed to extend the primers by one base with didedoxynucleotides at the SNP site (Sokolov 1990; Syvanen et al. 1990). Combined hybrdization-based/enzvmatic approaches Oligonucleotide-ligation assay (OLA) This assay involves a pair of probes with an allele-discriminating base at one end (either 5' or 3') and another oligonucleotide ending at the base adjacent to the SNP. Ligase mediates joining of the oligonucleotide to the probe that matches the allele(Landegren et al. 1988). Invasive cleavage assay The assay employs the use of Flap endonuclease (FEN), an enzyme that recognizes and cleaves the 'flap' that results from the binding of two overlapping oligonucleotides to the same target DNA with perfect match. To exploit this property of FEN to discriminate SNP, three oligonucleotides are employed: one pair of probes containing an internal allele-discriminating base, and an 'invader' oligonucleotide that can bind to target sequence on the 3' side of the SNP. When there is a perfect match, the 'flap' of the probe will get cleaved (Olivier 2005). Comments on the different biochemistries The accuracy of hybridization approaches depends largely on binding specificity. As such, attempts have been made to use probes with special nucleotides that can bind to complementary DNA tighter. The increase in stability can improve the allele- discriminating ability of the probe. TaqMan MGB probes that bind to minor groove of DNA is one example (Kuimelis et al. 1997). Syntheic nucleotide analogs including peptide nucleic acid (PNA) (Ross et al. 1997) and locked nucleic acid (LNA) (Orum et al. 1999) have also been used. PNAs are analogs with uncharged polyamide backbones (Ross et al. 1997) while LNAs contain an extra 2'-0, 4'-C-methylene bridge on the ribose ring of the nucleotide (Orum et al. 1999). For the enzymatic or combined approaches, accuracy much depends on the fidelity of the enzyme. DNA polymerases for SBE are very accurate. SBE assays are already used widely with various platforms e.g.(Nikiforov et al. 1994; Shumaker et al. 1996; Syvanen 1999), while the newer approaches --OLA and invasive cleavage -- are gaining acceptances as well. Detection principles The basis of detection dictates what assay platforms will be required. The two in conjunction affect how sensitive and quantitative the assay is. Fluorescence Fluoresence allows quantification and differentiation of alleles. In SBE, different fluoroscein labeled nucleotides can be used for incorporationg. Alternatively, targets can be fluorsencently labeled and amount of binding is measured (such as in 2-D microarrays). On the flip side, probes fixed on different fluorescently labeled microbeads is an alternative arraying format. FRET When two fluorophores with overlapping excitation and emission spectra are in close proximity, FRET can occur. FRET is the principle behind TaqMan (Applied Biosystems) probes (Livak et al. 1995; Livak 1999) and Molecular Beacon probes, and in LightCycler assay (Roche). In a TaqMan assay, two probes with different fluorescent dyes at 5' and a 3' quencher are used. Each probe has a discriminating base for each SNP allele. During PCR, probe with a mismatch will be displaced without cleavage while the matched probe gets cleaved, giving out fluorescence signal that gets monitored in real time(Livak et al. 1995; Livak 1999). In a Molecular Beacon assay, two probes with different fluorescent dyes at 5' and a 3' quencher are also used. Molecular Beacons are hairpin probes that contain a sequence complementary to target DNA (Tyagi and Kramer 1996; Tyagi et al. 1998). When the probe binds to perfectly matched target, the hairpin opens up to give up fluorescence. The LightCycler (Roche) assay is similar to TaqMan, but instead of using a single probe, LightCycler uses two different labeled probes binding to adjacent DNA sequences and one contains an allele-specific base (von Ahsen et al. 1999). Fluorescence polarization (FP) A DNA binding dye such as SYBR Green can be included in the reaction to detect formation of the product. Fluoresence polarization (FP) can also be used. The method uses polarized light to excite a fluorophore. The direction of emission depends on mass of the molecule, making it able to monitor the change in product size (Germer et al. 2000). Mass spectrometry (MS) Mass changes can be measured by matrix assisted laser desorption/ionization time-of- flight (MALDI-TOF) spectrometry. MALDI-TOF MS is highly sensitive and can resolve the smallest nucleotide difference, which is that of 9Da between A and T (Edwards et al. 2005). Thus, MALDI-TOF MS can assay which ddNTP was added to an SBE primer. In addition, SBE primers of varying lengths can be resolved, enabling multiplexing of reactions (Ross et al. 1998; Bray et al. 2001). Chemilluminescence Pyrosequencing is one example of chemilluminescence-based method (Langaee and Ronaghi 2005). It is a way to perform SBE. It is a system that measures the released pyrophosphate during addition of nucleotide, added one by one in specific order. The pyrophosphate is converted to ATP by ATP sulfurylase, and light is generated from ATP by luciferase. Degradation of the added nucleotide by apyrase frees up the template for next nucleotide addition round. Chemogenic signal Hapten-labelled nucleotides can be used for SBE. Hapten can be recognized by antibodies conjugated by enzyme that catalyzes light product formation like in an ELISA (Friedhoff et al. 1993). Nanoparticles Quantum dots Quantum dots are microfabricated nanoparticles that can be synthesized to emit very bright fluroscence at any wavelength, as emission is dependent on their sizes (Waggoner 2006). In SNP detection, the proof-of-concept use of quantum dots in place of traditional fluorophore has been demonstrated (Han et al. 2001). Quantum dots were able to increase detection sensitivity and multiplexing capability of assay (Xu et al. 2003). Gold nanoparticles The intense scattering of absorbed light by gold nanoparticles due to surface plasmon resonance can also be a way to enhance detection signal(Taton et al. 2000). Similar to quantum dots, gold particles have the potential to replace fluorophores to improve SNP detection sensitivity (Taton et al. 2000). Assay platforms Detection principle dictates the choice of platforms. The choice of platform determines the fixed cost and affects sensitivity and accuracy of detection. The accessibility of specific platforms can limit the choice of a researcher. Solid phase: Dot blots and reverse dot blots They are the classic methods for hybridizing ASO (Saiki et blot, target DNA is bound on nitrocellulose or nylon radiolabeled allele-specific oligonucleotides. Reverse is Thermodynamics of binding between nucleotide strands al. 1988). In a traditional dot filters and is probed with done in reverse dot blots. is not only affected by the single-base difference but also the surrounding sequence. Hybridization conditions differ for different probe-target pairs. The use of tetramethylammonium chloride (TMAC) as the hybridization solution has greatly increased the thermal stability range of oligonucleotides, allowing a universal hybridization condition (current protocols of molecular biology). Microarrays Microarrays allow simultaneous analysis of a large number of markers . Classicaly these are 2-D glass slides containing bound DNA probes in the form of fixed arrays. Data are obtained from scanned images. Microarrays can be used as a physical platform for hybridization or as a reaction platform for in-situ biochemistry. The first usage essentially evolves from the reverse dot blotting principle(Southern et al. 1992). One subtype is to attach ASO probes to the glass slide and allow hybridization of fluorescently labeled DNA with the SNP targets (Ranade et al. 2001). To circumvent the thermodynamic differences between different probe-target pairs, more than one probe per SNP can be used. For instance, the Affymetrix GeneChip system uses tens to hundred of ASO probes for each SNP. The other subtype of hybridization-based microarrays is to use different bound probes to fish for the SNP products resulting from one of the enzymatic reactions described. A new array format using microelectrodes to immobilize DNA has been developed by Nanogen. By controlling the current, binding and washing of DNA can been expedited. Application of such chip in hybridization-based SNP detection has been proven to be robust (Gilles et al. 1999). Aside from being a hybridization scaffold, microarrays can also be used as a platform to carry out SBE and OLA. SBE can be performed as an arrayed primer extension (APEX) reaction, in which primers for extension are fixed on a glass slide (Shumaker et al. 1996); polymermase and labeled ddNTPs are added to allow SBE reaction to proceed. As for the detection of OLA products, microarrays can be used to immobilize one of the ligation probes and allow OLA reaction to happen on the slide (Gerry et al. 1999). Microbeads Microbeads are small particles that can be linked to a probe. They are essentially a flexible form of a microarray, allowing flexibility in the assay design. SBE (Chen et al. 2000)and OLA (lannone et al. 2000) can be performed in fashions analogous to mixed microarrays. Hybridization kinetics is believed to be better on microbeads than on traditional planar microarrays, as beads can freely move in solution. Individual bead identification depends on the bead properties. One class is differentially labeled with fluorophores, which can be sorted by FACs (Chen et al. 2000; lannone et al. 2000). Beads barcoded with bound oligonucleotides can be coupled to a planar array for detection (Shen et al. 2005). Solution phase: Electrophoresis To detect SBE products, capillary electrophoresis can be used to detect single base extension product from different fluorescently labeled dideoxynucleotides. Capillary electrophoresis can be performed using channel capillaries as in a standard sequencing machine or on a microplate. Gradient separation One form of gradient separation is dynamic allelic-specific hybridization (DASH), in which a double-stranded DNA binding dye is used to monitor hybridization dynamics over a sweeping gradient of temperature, allowing more robust genotyping (Prince et al. 2001). An analogous idea using an electric field has also been reported (Sosnowski et al. 1997). Specialized instruments are required to set up these gradients. Comments on SNP genotyping The utilities of SNP genotyping in cancer research are wide. The number of approaches to genotype them seems to be bound only by human imagination. As mentioned, the choice of genotyping methods depends on the need of the experimenter. A whole-genome study will require a high-throughput format such as a microarray-based assay. The Affymetrix GeneChip has been the choice of most current whole-genome studies. Another highly multiplexed assay ideal for whole-genome studies is Illumina's GoldenGate assay, which combines microbeads and microarray on an array of arrays format (Table 1). Currently, a 100K GeneChip is available from Affymetrix and a 4.7K SNP linkage panel is available from Illumina (Matsuzaki et al. 2004; Murray et al. 2004). Capital investment required is the constraint. Other options such as electrophoresis based detection assays may provide the more cost-effective alternatives for smaller-scale studies. Other considerations in choosing an assay include the quantity and quality of the available DNA. While genome amplification and complexity reduction methods have been developed for the current methods of SNP genotyping (Jordan et al. 2002; Barker et al. 2004), new detection means such as nanoparticles may make future assays sensitive enough to be performed on minute amount of samples without pre-processing. The rapidly evolving SNP genotyping technologies will hopefully lead to many cancer genomic discoveries in the years to come. Digital Karyotyping Aside from array CGH and SNP genotyping, another new genome analysis tool based on the genome sequence is digital karyotyping. The basic concept of digital karyotyping is similar to serial analysis of gene expression (SAGE) (Velculescu et al. 1995). As described by Wang et al. 2002, digital karyotyping involves isolation of 21bp short sequence tags from specific locations in the genome. Isolated tags are ligated into a concatenated form called 'ditags', which are then amplified by PCR en-masse and sequenced. Individual tags are digitally extracted from sequence data and matched on chromosomes. Tag densities along chromosomes are used to evaluate DNA content (Wang et al. 2002). Digital karyotyping is a powerful tool that has enabled the discovery of several specific cancer-associated gene amplifications, such as those of the homeobox gene OTX2 in medulloblastoma (Boon et al. 2005), Notch3 and the chromatin remodeling gene Rsf-1/HBXAP in ovarian cancer (Shih Ie et al. 2005; Park et al. 2006). The biggest limitation to perform digital karyotyping is cost. Emerging Approaches to analyze the Sequenced Genome In the previous section, a few emerging technologies: array CGH, SNP genotyping platforms, and digital karyotyping, have been described. These technologies provide means to analyze cancer genomes in a systematic fashion. Another set of opportunities involves approaching the available genome sequence in new ways. I will first describe the resequencing of certain genes in the human genome to identify cancer-related mutations and then the comparative genomic approach by studying animal cancers. Genome resequencing Mutations in signaling pathways involved in cell proliferation, cell death and cell differentiation are thought to be key in cancer. Resequencing of genes and gene families involved in these pathways has been pursued to systematically identify cancer-causing mutations. One of the first studies by Davis et al. coupled a heteroduplex-electrophoresis method with direct sequencing. They identified BRAF mutation in >60% of melanomas and at lower frequency in other cancers such as colorectoal cancer and non-small cell lung cancer (Davies et al. 2002). Subsequent exon resequencing experiments have focused on protein tyrosine phosphatases (PTPs), protein tyrosine kinases (PTKs) and phosphatidylinositol 3-kinases (PI3Ks), identifying numerous cancer-specific mutations including PTPRT, EGFR, ERBB2, and PIK3CA (Paez et al. 2004; Samuels et al. 2004; Stephens et al. 2004; Wang et al. 2004). Resequencing the genome in a systematic manner is proving to be a fruitful approach. Comparative Cancer Genomics A comparative approach to study cancer is not a new idea. Many carcinogens in humans can similarly induce cancers in animals ---the first demonstration was tumor induction on rabbit ears from coal tar. Animal models of cancer can serve many research purposes: testing carcinogens, studying tumor biology, testing therapeutics etc.. The complete genome sequences of many model organisms are now available. Comparative analyses between genomes of different species have become feasible. In cancer research, comparative genomic studies are valuable in two major ways: 1) in validating animal models through an assessment of their degree of genetic resemblance to human disease and 2) in identifying genes and/or gene sets that are common to the model organism and human tumorigenesis. Mouse cancer genomics The genetic tractability of the mouse has made it an important animal model to study cancer (Van Dyke and Jacks 2002). A good mouse model should share phenotypic and genetic similarities to the human cancer it mimics (Hann and Balmain 2001). Genomics has aided the comparison of mouse models at both levels. Phenotypically, the use of global gene-expression as a validation tool between mouse and human cancers has been demonstrated (Sweet-Cordero et al. 2005). In addition, Sweet-Cordero et al. have shown that genome-wide gene expression data from controlled mouse experiments can help to filter molecular data from human samples. Genetically, a similar comparison can be performed using genomics tools that assess global genetic alterations. As in human cancers, chromosomal abnormalities have long been observed in tumors developed in mice (Sasaki 1982; Liyanage et al. 1996). In the past few years, the sequencing of the mouse genome has demonstrated the high-degree of conservation between mouse and human genomes and spurred the development of high-resolution tools to characterize genetic lesions in mouse tumors. The mouse genome The first complete sequence of the mouse was published in 2002 (Waterston et al. 2002). Some observations by the authors are summarized below. The mouse genome is slightly smaller than the human genome (2.5Gb vs. 2.9Gb) but contains about the same number of genes (30,000), while 80% of the mouse genes have a single identifiable ortholog in human. At a gross chromosomal level, 75 million years of independent evolution has resulted in many large-scale rearrangements but local gene orders are mostly maintained. In fact, about 90% of the mouse and human genomes can be divided into regions of conserved of synteny (i.e. same thread), where local structure is intact. The total amount of ~350 conserved syntenic segments have been evolutionarily shuffled throughout the mouse and human genomes. In thinking about comparative cancer genomics, the conservation of genes at 1:1 ratio suggests similar sets of genes likely control the same cellular processes in the two species. In addition, the partitioning of syntenic regions on different chromosomes provides a framework to assess the relative importance of each region in the other species. Analyzing the mouse cancer genome As in the old days of human cancer research, cytogenetics was the most accessible means to perform genomic analyses. Unlike human chromosomes, mouse chromosomes are acrocentric and similar in size. Karyotypic analysis in mouse cells were difficult until the development of chromosome painting techniques such as spectral karyotyping (Liyanage et al. 1996). In more recent years, mouse genome sequence availability has led to new tools for analysis. For copy number study, multiple array CGH platforms have been developed and the competition for better resolution has been fierce: starting from a custom-made BAC array covering the genome at 2-20Mb resolution (Hodgson et al. 2001), followed by BAC arrays with 1K probes (Cai et al. 2002), 2K probes (Snijders et al. 2005), 3K probes (Chung et al. 2004), and 19K probes (Li et al. 2004). The 19K array has clones spaced -39kb throughout the genome (Li et al. 2004). Oligonucleotide CGH arrays have also been applied for mouse genome analysis: 20K array made by Agilent was reported to have a -50kb genomic resolution (Brennan et al. 2004) while a non- commercially made 20K array has also been reported (van den Ijssel et al. 2005). Oligonucleotide arrays with> 40K probes are now available in the market. As for LOH mapping, several genome-wide SNP genotyping methods have been developed for the mouse (Petkov et al. 2004a; Owens et al. 2005; Moran et al. 2006). However, report that uses SNP in genome-wide LOH mapping in the mouse has not been made. While new genomic tools for mouse tumor analysis are being refined or developed, a step back to summarize current data from mouse tumor genome studies would be appropriate. Genetically engineered mouse cancer models enable the study of tumorigenesis in a controlled and reproducible fashion. Cooperating genetic lesions that enhance the tumorigenicity of the initiating mutant cells can be analyzed by array CGH. Several generalizations can be made from studying these models. First, large-scale chromosomal lesions appear to be predominant in telomerase active mice but the occasionally observed focal lesions have been helpful in pinpointing genetic regions important for tumorigenesis (Hodgson et al. 2001; Hackett et al. 2003). On the other hand, mice with dysfunctional telomeres exhibit a wider range of chromosomal abnormalities including more focal lesions (O'Hagan et al. 2002). Secondly, genetic background affects the lesions that are present (Hager et al. 2004). Thirdly, expression timing of the initiating mutation in a genetically engineered model can also influence the genetic alterations appeared (Hager et al. 2004). Finally, aside from identifying cancer genes, array CGH data can be used to classify tumors (O'Hagan et al. 2003). Some details leading to these sweeping statements are discussed below in the light of genetically engineered mouse models of various types cancers, including: 1) pancreatic islet carcinoma; 2) melanoma; 3) neuroblastoma, and 4) carcinomas including breast, colon, and skin tumors. Pancreatic islet carcinoma model The RIP-Tag mice express SV40 T antigens (Tag) under the control of the rat insulin promoter (RIP). In the initial pancreatic islet carcinoma study in RIP-Tag mice by Hodgson et al., most copy number alterations were observed to span large chromosomal areas but a focal lesion as small as ~3Mb could be detected on chromosome 16. That had allowed the authors to narrow down a previously known LOH region in the area that is syntenic to chromosome 3q in human (Hodgson et al. 2001). In addition, new observations of other recurrent chromosomal copy number changes let the authors identify a few candidate oncogenes or tumor-suppressor genes (Hodgson et al. 2001), illustrating the potential value in performing genomic analysis on mouse tumors. A later CGH study on the islet tumors showed that RIP-Tag mice on FVB/N, C57B1/6, and C4Heb/Fe backgrounds develop tumors with different copy number change spectra, suggesting the influence of genetic background (Hager et al. 2004). This is analogous to the varying susceptibility to different cancers within the human population. Another interesting finding in the islet cell cancer model was the timing effect of T-antigen expression on the copy number alterations seen (Hager et al. 2004). The authors proposed changes in tumor microenvironment at different time points can impose a different set of selection criteria on tumor cells. Melanoma model Melanomas develop in a RAS-induced pl1 9 Af-/- mouse model spontaneously but are accelerated by UV irradiation (Kannan et al. 2003). A use of array CGH data is tumor classification, as demonstrated in the classification of UV-induced vs. non-UV induced melanomas in these mice (O'Hagan et al. 2003). Neuroblastoma model Amplification of MYCN is frequently observed in human neuroblastoma (Brodeur et al. 1984). Neuroblastomas can be induced in mice by expressing human MYCN under a rat tyrosine hydroxylase (TH) promoter (Weiss et al. 1997). By CGH, several recurrent whole chromosomal gains and losses appeared to cooperate with MYCN to drive tumorigenesis in this model (Hackett et al. 2003). More interestingly, Hackett et al. were able to identify a minimally gained region on mouse chromosome 11 by aligning recurrent focal gains that were observed. A syntenic comparison of the region allowed delineation of a frequently gained region on human chromosome 17q to 15Mb (Hackett et al. 2003). Other carcinomas Carcinomas are predominant in aging humans, but tumor spectrum in mice is skewed toward a high incidence of lymphomas and soft tissue sarcomas (Artandi et al. 2000). Mutating p53 and mTerc (telomerase RNA component) not only shifts the mouse tumor spectrum to more human like but also leads to tumors with genetic aberrations more like those seen in human cancers; frequent aneuploidy, unbalanced translocations, amplifications and deletions are seen in tumors of these mice in the breast, colon, and lung (Artandi et al. 2000; O'Hagan et al. 2002). Significantly, genomic analysis by CGH suggested some of the recurrent copy number changes are syntenic to changes frequently seen in human cancers (O'Hagan et al. 2002). What's now and what's next for mouse cancer genomics Mouse models of cancer have become an integral part of cancer research for studying tumor biology and testing therapeutics. Using genomic analysis tools such as array CGH, the models can be validated for their degree of genetic resemblance to humans and be used to pinpoint the critical changes in human cancers. In addition, the availability of whole mouse genome sequence has made whole-genome association studies feasible in mice, easing the discovery of disease loci. Dog cancer genomics Aside from the mouse, the value of the domestic dog in cancer research is becoming realized. To complete the discussion on comparative cancer genomics, I will briefly mention the role of dogs in cancer research. The availability of many inbred strains has greatly facilitated mouse genetics. For dog, the extensive breeding of dogs through history has created many purebred strains. Genetic studies can be performed with less genetic background noise. In addition, many of these pure breeds show different susceptibility to diseases including cancer (Lindblad- Toh et al. 2005). As in mouse and humans, genetic association studies can be performed in dogs to identify genetic susceptibility loci. The dog genome sequence is now complete (Lindblad-Toh et al. 2005). The dog and human genomes sequence are highly homologous; over 90% of the dog sequence lies in regions of conserved synteny with humans. As in the case between human and mouse, segments of synteny are distributed throughout each of the dog and human genomes during evolution, enabling a comparison of the relative role of individual syntenic segment in each species. Like in human and mouse, recurrent chromosome aberrations have been observed in canine tumors (Dunn et al. 2000; Thomas et al. 2003; Milne et al. 2004). Characterization of genetic abnormalities may aid the discovery of orthologous cancer genes in humans. Recently, a 2-Mb resolution BAC microarray has been developed for CGH analysis of dog tumors (Thomas et al. 2005). The authors reported an osteosarcoma case that exhibited a wide range of abnormalities. Ongoing studies by the authors on a range of canine cancers including lymphoma, leukemia, osteosarcoma, and brain tumors were noted. In the years to come, a three-way comparison of tumor genomics between human, mouse, and dogs might yield interesting clues about the tumorigenesis process. Outlook Cancer genome sequencing Genomic analysis is instrumental for the discovery of many underlying mutations in cancer. Some of the early discoveries of oncogenes have led to the development of a new class of target cancer drugs such as Gleevec and Herceptin. Evolving genomic technologies have provided the platforms further insights. It is worth mentioning that on December 13, 2005, the National Cancer Institute (NCI) and the National Human Genome Research Institute (NHGRI) launched a pilot project to build The Cancer Genome Atlas (TCGA). It is an initiative to apply genome analysis technologies, including large-scale genome sequencing, to study cancer. According to the mission statement, the goal of the pilot is to "assess the feasibility of a full-scale effort to systematically explore the entire spectrum of genomic changes involved in human cancer" (NCI 2005). Future years in cancer genetics will likely be exciting. Thesis Scope The work presented in this thesis took a comparative approach to study cancer. Genomic studies were performed on mouse tumor models using two evolving tools: ROMA in characterizing copy number changes and genome-wide SNP genotyping in uncovering LOH regions. While the human ROMA platform has shown promise in delivering high- resolution data, the mouse version of ROMA needed to be tested for comparative genomic analysis purposes. Likewise, although the abundance of SNPs has made them valuable markers for LOH mapping in human tumors, the same concept had to be tested in mice. In proving the concept, a new protocol of SNP genotyping in mice was worked out. Chapters 2 and 3 describe the application of these two techniques in analyzing mouse models that mimic human lung cancer and retinoblastoma. In both mouse models, tumorigenesis was driven by mutations engineered in specific genes, KRas in the case of lung cancer and Rb and p130 in the retinoblastoma model. Cooperating mutations that arose in the initiating mutant cells during tumorigenesis were examined in this study by the genomic tools described. Recurrent genetic changes could be detected in both models. References Albertson, D.G. and Pinkel, D. 2003. Genomic microarrays in human genetic disease and cancer. Hum Mol Genet 12 Spec No 2: R145-152. Artandi, S.E., Chang, S., Lee, S.L., Alson, S., Gottlieb, G.J., Chin, L., and DePinho, R.A. 2000. Telomere dysfunction promotes non-reciprocal translocations and epithelial cancers in mice. Nature 406(6796): 641-645. Balsara, B.R. and Testa, J.R. 2002. Chromosomal imbalances in human lung cancer. Oncogene 21(45): 6877-6883. Barker, D.L., Hansen, M.S., Faruqi, A.F., Giannola, D., Irsula, O.R., Lasken, R.S., Latterich, M., Makarov, V., Oliphant, A., Pinter, J.H., Shen, R., Sleptsova, I., Ziehler, W., and Lai, E. 2004. Two methods of whole-genome amplification enable accurate genotyping across a 2320-SNP linkage panel. Genome Res 14(5): 901-907. Barrett, M.T., Scheffer, A., Ben-Dor, A., Sampas, N., Lipson, D., Kincaid, R., Tsang, P., Curry, B., Baird, K., Meltzer, P.S., Yakhini, Z., Bruhn, L., and Laderman, S. 2004. Comparative genomic hybridization using oligonucleotide microarrays and total genomic DNA. Proc Natl Acad Sci USA 101(51): 17765-17770. Ben-Avi, L., Durst, R., Shpitzen, S., Leitersdorf, E., and Meiner, V. 2004. Apolipoprotein E genotyping: accurate, simple, high throughput method using ABI Prism SNaPshot Multiplex System. JAlzheimers Dis 6(5): 497-501. Benavides, F., Conti, C.J., LaCava, M., Flores, M., Glasscock, E., Sternik, G., Gimenez- Conti, I.B., Johnston, D.A., Dunsford, H.A., Goldstein, L.S., and Rodriguez, L.V. 2003. Loss of heterozygosity analysis of mouse pulmonary adenomas induced by coal tar. Environ Mol Mutagen 41(4): 300-308. Beroukhim, R., Lin, M., Park, Y., Hao, K., Zhao, X., Garraway, L.A., Fox, E.A., Hochberg, E.P., Mellinghoff, I.K., Hofer, M.D., Descazeaud, A., Rubin, M.A., Meyerson, M., Wong, W.H., Sellers, W.R., and Li, C. 2006. Inferring loss-of- heterozygosity from unpaired tumors using high-density oligonucleotide SNP arrays. PLoS Comput Biol 2(5): e41. Bignell, G.R., Huang, J., Greshock, J., Watt, S., Butler, A., West, S., Grigorova, M., Jones, K.W., Wei, W., Stratton, M.R., Futreal, P.A., Weber, B., Shapero, M.H., and Wooster, R. 2004. High-resolution analysis of DNA copy number using oligonucleotide microarrays. Genome Res 14(2): 287-295. Boon, K., Eberhart, C.G., and Riggins, G.J. 2005. Genomic amplification of orthodenticle homologue 2 in medulloblastomas. Cancer Res 65(3): 703-707. Bray, M.S., Boerwinkle, E., and Doris, P.A. 2001. High-throughput multiplex SNP genotyping with MALDI-TOF mass spectrometry: practice, problems and promise. Hum Mutat 17(4): 296-304. Brennan, C., Zhang, Y., Leo, C., Feng, B., Cauwels, C., Aguirre, A.J., Kim, M., Protopopov, A., and Chin, L. 2004. High-resolution global profiling of genomic alterations with long oligonucleotide microarray. Cancer Res 64(14): 4744-4748. Brodeur, G.M., Seeger, R.C., Schwab, M., Varmus, H.E., and Bishop, J.M. 1984. Amplification of N-myc in untreated human neuroblastomas correlates with advanced disease stage. Science 224(4653): 1121-1124. Cai, W.W., Mao, J.H., Chow, C.W., Damani, S., Balmain, A., and Bradley, A. 2002. Genome-wide detection of chromosomal imbalances in tumors using BAC microarrays. Nat Biotechnol 20(4): 393-396. Cao, X., Eu, K.W., Kumarasinghe, M.P., Li, H.H., Loi, C., and Cheah, P.Y. 2006. Mapping of hereditary mixed polyposis syndrome (HMPS) to chromosome 10q23 by genomewide high-density single nucleotide polymorphism (SNP) scan and identification of BMPR1A loss of function. JMed Genet 43(3): el3. Carrasco, D.R., Tonon, G., Huang, Y., Zhang, Y., Sinha, R., Feng, B., Stewart, J.P., Zhan, F., Khatry, D., Protopopova, M., Protopopov, A., Sukhdeo, K., Hanamura, I., Stephens, O., Barlogie, B., Anderson, K.C., Chin, L., Shaughnessy, J.D., Jr., Brennan, C., and Depinho, R.A. 2006. High-resolution genomic profiles define distinct clinico-pathogenetic subgroups of multiple myeloma patients. Cancer Cell 9(4): 313-325. Carvalho, B., Ouwerkerk, E., Meijer, G.A., and Ylstra, B. 2004. High resolution microarray comparative genomic hybridisation analysis using spotted oligonucleotides. J Clin Pathol 57(6): 644-646. Chao, R.C., Pyzel, U., Fridlyand, J., Kuo, Y.M., Teel, L., Haaga, J., Borowsky, A., Horvai, A., Kogan, S.C., Bonifas, J., Huey, B., Jacks, T.E., Albertson, D.G., and Shannon, K.M. 2005. Therapy-induced malignant neoplasms in Nfl mutant mice. Cancer Cell 8(4): 337-348. Chen, D., Gallie, B.L., and Squire, J.A. 2001. Minimal regions of chromosomal imbalance in retinoblastoma detected by comparative genomic hybridization. Cancer Genet Cytogenet 129(1): 57-63. Chen, J., lannone, M.A., Li, M.S., Taylor, J.D., Rivers, P., Nelsen, A.J., Slentz-Kesler, K.A., Roses, A., and Weiner, M.P. 2000. A microsphere-based assay for multiplexed single nucleotide polymorphism analysis using single base chain extension. Genome Res 10(4): 549-557. Chung, Y.J., Jonkers, J., Kitson, H., Fiegler, H., Humphray, S., Scott, C., Hunt, S., Yu, Y., Nishijima, I., Velds, A., Holstege, H., Carter, N., and Bradley, A. 2004. A whole-genome mouse BAC microarray with 1-Mb resolution for analysis of DNA copy number changes by array comparative genomic hybridization. Genome Res 14(1): 188-196. Davies, H., Bignell, G.R., Cox, C., Stephens, P., Edkins, S., Clegg, S., Teague, J., Woffendin, H., Garnett, M.J., Bottomley, W., Davis, N., Dicks, E., Ewing, R., Floyd, Y., Gray, K., Hall, S., Hawes, R., Hughes, J., Kosmidou, V., Menzies, A., Mould, C., Parker, A., Stevens, C., Watt, S., Hooper, S., Wilson, R., Jayatilake, H., Gusterson, B.A., Cooper, C., Shipley, J., Hargrave, D., Pritchard-Jones, K., Maitland, N., Chenevix-Trench, G., Riggins, G.J., Bigner, D.D., Palmieri, G., Cossu, A., Flanagan, A., Nicholson, A., Ho, J.W., Leung, S.Y., Yuen, S.T., Weber, B.L., Seigler, H.F., Darrow, T.L., Paterson, H., Marais, R., Marshall, C.J., Wooster, R., Stratton, M.R., and Futreal, P.A. 2002. Mutations of the BRAF gene in human cancer. Nature 417(6892): 949-954. Dhami, P., Coffey, A.J., Abbs, S., Vermeesch, J.R., Dumanski, J.P., Woodward, K.J., Andrews, R.M., Langford, C., and Vetrie, D. 2005. Exon array CGH: detection of copy-number changes at the resolution of individual exons in the human genome. Am JHum Genet 76(5): 750-762. Dietrich, W.F., Radany, E.H., Smith, J.S., Bishop, J.M., Hanahan, D., and Lander, E.S. 1994. Genome-wide search for loss of heterozygosity in transgenic mouse tumors reveals candidate tumor suppressor genes on chromosomes 9 and 16. Proc Natl Acad Sci USA 91(20): 9451-9455. Dunn, K.A., Thomas, R., Binns, M.M., and Breen, M. 2000. Comparative genomic hybridization (CGH) in dogs--application to the study of a canine glial tumour cell line. Vet J 160(1): 77-82. Edwards, J.R., Ruparel, H., and Ju, J. 2005. Mass-spectrometry DNA sequencing. Mutat Res 573(1-2): 3-12. Einarsdottir, K., Humphreys, K., Bonnard, C., Palmgren, J., Iles, M.M., Sjolander, A., Li, Y., Chia, K.S., Liu, E.T., Hall, P., Liu, J., and Wedren, S. 2006. Linkage Disequilibrium Mapping of CHEK2: Common Variation and Breast Cancer Risk. PLoS Med 3(6): e168. Ellis, N.A., Kirchhoff, T., Mitra, N., Ye, T.Z., Chuai, S., Huang, H., Nafa, K., Norton, L., Neuhausen, S., Gordon, D., Struewing, J.P., Narod, S., and Offit, K. 2006. Localization of breast cancer susceptibility loci by genome-wide SNP linkage disequilibrium mapping. Genet Epidemiol 30(1): 48-61. Engle, L.J., Simpson, C.L., and Landers, J.E. 2006. Using high-throughput SNP technologies to study cancer. Oncogene 25(11): 1594-1601. Finney, R.E. and Bishop, J.M. 1993. Predisposition to neoplastic transformation caused by gene replacement of H-rasl. Science 260(5113): 1524-1527. Frazer, K.A., Wade, C.M., Hinds, D.A., Patil, N., Cox, D.R., and Daly, M.J. 2004. Segmental phylogenetic relationships of inbred mouse strains revealed by fine- scale analysis of sequence variation across 4.6 mb of mouse genome. Genome Res 14(8): 1493-1500. Friedhoff, P., Hahn, M., Wolfes, H., and Pingoud, A. 1993. Quantitative polymerase chain reaction with oligodeoxynucleotide ligation assay/enzyme-linked immunosorbent assay detection. Anal Biochem 215(1): 9-16. Fujiwara, T., Bandi, M., Nitta, M., Ivanova, E.V., Bronson, R.T., and Pellman, D. 2005. Cytokinesis failure generating tetraploids promotes tumorigenesis in p53-null cells. Nature 437(7061): 1043-1047. Futreal, P.A., Coin, L., Marshall, M., Down, T., Hubbard, T., Wooster, R., Rahman, N., and Stratton, M.R. 2004. A census of human cancer genes. Nat Rev Cancer 4(3): 177-183. Garnis, C., Lockwood, W.W., Vucic, E., Ge, Y., Girard, L., Minna, J.D., Gazdar, A.F., Lam, S., Macaulay, C., and Lam, W.L. 2006. High resolution analysis of non- small cell lung cancer cell lines by whole genome tiling path array CGH. Int J Cancer 118(6): 1556-1564. Germer, S., Holland, M.J., and Higuchi, R. 2000. High-throughput SNP allele-frequency determination in pooled DNA samples by kinetic PCR. Genome Res 10(2): 258- 266. Gerry, N.P., Witowski, N.E., Day, J., Hammer, R.P., Barany, G., and Barany, F. 1999. Universal DNA microarray method for multiplex detection of low abundance point mutations. JMol Biol 292(2): 251-262. Gilles, P.N., Wu, D.J., Foster, C.B., Dillon, P.J., and Chanock, S.J. 1999. Single nucleotide polymorphic discrimination by an electronic dot blot assay on semiconductor microchips. Nat Biotechnol 17(4): 365-370. Grupe, A., Germer, S., Usuka, J., Aud, D., Belknap, J.K., Klein, R.F., Ahluwalia, M.K., Higuchi, R., and Peltz, G. 2001. In silico mapping of complex disease-related traits in mice. Science 292(5523): 1915-1918. Hackett, C.S., Hodgson, J.G., Law, M.E., Fridlyand, J., Osoegawa, K., de Jong, P.J., Nowak, N.J., Pinkel, D., Albertson, D.G., Jain, A., Jenkins, R., Gray, J.W., and Weiss, W.A. 2003. Genome-wide array CGH analysis of murine neuroblastoma reveals distinct genomic aberrations which parallel those in human tumors. Cancer Res 63(17): 5266-5273. Hager, J.H., Hodgson, J.G., Fridlyand, J., Hariono, S., Gray, J.W., and Hanahan, D. 2004. Oncogene expression and genetic background influence the frequency of DNA copy number abnormalities in mouse pancreatic islet cell carcinomas. Cancer Res 64(7): 2406-2410. Hamaguchi, M., Meth, J.L., von Klitzing, C., Wei, W., Esposito, D., Rodgers, L., Walsh, T., Welcsh, P., King, M.C., and Wigler, M.H. 2002. DBC2, a candidate for a tumor suppressor gene involved in breast cancer. Proc Natl Acad Sci USA 99(21): 13647-13652. Han, M., Gao, X., Su, J.Z., and Nie, S. 2001. Quantum-dot-tagged microbeads for multiplexed optical coding of biomolecules. Nat Biotechnol 19(7): 631-635. Hanahan, D. and Weinberg, R.A. 2000. The hallmarks of cancer. Cell 100(1): 57-70. Hann, B. and Balmain, A. 2001. Building 'validated' mouse models of human cancer. Curr Opin Cell Biol 13(6): 778-784. Hemann, M.T., Fridman, J.S., Zilfou, J.T., Hernando, E., Paddison, P.J., Cordon-Cardo, C., Hannon, G.J., and Lowe, S.W. 2003. An epi-allelic series of p53 hypomorphs created by stable RNAi produces distinct tumor phenotypes in vivo. Nat Genet 33(3): 396-400. Herzog, C.R., Devereux, T.R., Pittman, B., and You, M. 2002. Carcinogenic induction directs the selection of allelic losses in mouse lung tumorigenesis. Cancer Res 62(22): 6424-6429. Hingorani, S.R., Wang, L., Multani, A.S., Combs, C., Deramaudt, T.B., Hruban, R.H., Rustgi, A.K., Chang, S., and Tuveson, D.A. 2005. Trp53R172H and KrasG12D cooperate to promote chromosomal instability and widely metastatic pancreatic ductal adenocarcinoma in mice. Cancer Cell 7(5): 469-483. Hodgson, G., Hager, J.H., Volik, S., Hariono, S., Wernick, M., Moore, D., Nowak, N., Albertson, D.G., Pinkel, D., Collins, C., Hanahan, D., and Gray, J.W. 2001. Genome scanning with array CGH delineates regional alterations in mouse islet carcinomas. Nat Genet 29(4): 459-464. Hoglund, M., Gisselsson, D., Hansen, G.B., and Mitelman, F. 2004. Statistical dissection of cytogenetic patterns in lung cancer reveals multiple modes of karyotypic evolution independent of histological classification. Cancer Genet Cytogenet 154(2): 99-109. Huang, J., Wei, W., Zhang, J., Liu, G., Bignell, G.R., Stratton, M.R., Futreal, P.A., Wooster, R., Jones, K.W., and Shapero, M.H. 2004. Whole genome DNA copy number changes identified by high density oligonucleotide arrays. Hum Genomics 1(4): 287-299. Iannone, M.A., Taylor, J.D., Chen, J., Li, M.S., Rivers, P., Slentz-Kesler, K.A., and Weiner, M.P. 2000. Multiplexed single nucleotide polymorphism genotyping by oligonucleotide ligation and flow cytometry. Cytometry 39(2): 131-140. Ishkanian, A.S., Malloff, C.A., Watson, S.K., DeLeeuw, R.J., Chi, B., Coe, B.P., Snijders, A., Albertson, D.G., Pinkel, D., Marra, M.A., Ling, V., MacAulay, C., and Lam, W.L. 2004. A tiling resolution DNA microarray with complete coverage of the human genome. Nat Genet 36(3): 299-303. Jackson, E.L., Olive, K.P., Tuveson, D.A., Bronson, R., Crowley, D., Brown, M., and Jacks, T. 2005. The differential effects of mutant p53 alleles on advanced murine lung cancer. Cancer Res 65(22): 10280-10288. Janne, P.A., Li, C., Zhao, X., Girard, L., Chen, T.H., Minna, J., Christiani, D.C., Johnson, B.E., and Meyerson, M. 2004. High-resolution single-nucleotide polymorphism array and clustering analysis of loss of heterozygosity in human lung cancer cell lines. Oncogene 23(15): 2716-2726. Jobanputra, V., Sebat, J., Troge, J., Chung, W., Anyane-Yeboa, K., Wigler, M., and Warburton, D. 2005. Application of ROMA (representational oligonucleotide microarray analysis) to patients with cytogenetic rearrangements. Genet Med 7(2): 111-118. Johnson, L., Mercer, K., Greenbaum, D., Bronson, R.T., Crowley, D., Tuveson, D.A., and Jacks, T. 2001. Somatic activation of the K-ras oncogene causes early onset lung cancer in mice. Nature 410(6832): 1111-1116. Jordan, B., Charest, A., Dowd, J.F., Blumenstiel, J.P., Yeh Rf, R.F., Osman, A., Housman, D.E., and Landers, J.E. 2002. Genome complexity reduction for SNP genotyping analysis. Proc Natl Acad Sci USA 99(5): 2942-2947. Kallioniemi, A., Kallioniemi, O.P., Sudar, D., Rutovitz, D., Gray, J.W., Waldman, F., and Pinkel, D. 1992. Comparative genomic hybridization for molecular cytogenetic analysis of solid tumors. Science 258(5083): 818-821. Kannan, K., Sharpless, N.E., Xu, J., O'Hagan, R.C., Bosenberg, M., and Chin, L. 2003. Components of the Rb pathway are critical targets of UV mutagenesis in a murine melanoma model. Proc Natl Acad Sci USA 100(3): 1221-1225. Kim, T.M., Yim, S.H., Lee, J.S., Kwon, M.S., Ryu, J.W., Kang, H.M., Fiegler, H., Carter, N.P., and Chung, Y.J. 2005. Genome-wide screening of genomic alterations and their clinicopathologic implications in non-small cell lung cancers. Clin Cancer Res 11(23): 8235-8242. Knudson, A.G., Jr. 1971. Mutation and cancer: statistical study of retinoblastoma. Proc Natl Acad Sci USA 68(4): 820-823. Kuimelis, R.G., Livak, K.J., Mullah, B., and Andrus, A. 1997. Structural analogues of TaqMan probes for real-time quantitative PCR. Nucleic Acids Symp Ser(37): 255- 256. Kuukasjarvi, T., Tanner, M., Pennanen, S., Karhu, R., Visakorpi, T., and Isola, J. 1997. Optimizing DOP-PCR for universal amplification of small DNA samples in comparative genomic hybridization. Genes Chromosomes Cancer 18(2): 94-101. Lage, J.M., Leamon, J.H., Pejovic, T., Hamann, S., Lacey, M., Dillon, D., Segraves, R., Vossbrinck, B., Gonzalez, A., Pinkel, D., Albertson, D.G., Costa, J., and Lizardi, P.M. 2003. Whole genome analysis of genetic alterations in small DNA samples using hyperbranched strand displacement amplification and array-CGH. Genome Res 13(2): 294-307. Landegren, U., Kaiser, R., Sanders, J., and Hood, L. 1988. A ligase-mediated gene detection technique. Science 241(4869): 1077-1080. Lander, E.S. Linton, L.M. Birren, B. Nusbaum, C. Zody, M.C. Baldwin, J. Devon, K. Dewar, K. Doyle, M. FitzHugh, W. Funke, R. Gage, D. Harris, K. Heaford, A. Howland, J. Kann, L. Lehoczky, J. LeVine, R. McEwan, P. McKeman, K. Meldrim, J. Mesirov, J.P. Miranda, C. Morris, W. Naylor, J. Raymond, C. Rosetti, M. Santos, R. Sheridan, A. Sougnez, C. Stange-Thomann, N. Stojanovic, N. Subramanian, A. Wyman, D. Rogers, J. Sulston, J. Ainscough, R. Beck, S. Bentley, D. Burton, J. Clee, C. Carter, N. Coulson, A. Deadman, R. Deloukas, P. Dunham, A. Dunham, I. Durbin, R. French, L. Grafham, D. Gregory, S. Hubbard, T. Humphray, S. Hunt, A. Jones, M. Lloyd, C. McMurray, A. Matthews, L. Mercer, S. Milne, S. Mullikin, J.C. Mungall, A. Plumb, R. Ross, M. Shownkeen, R. Sims, S. Waterston, R.H. Wilson, R.K. Hillier, L.W. McPherson, J.D. Marra, M.A. Mardis, E.R. Fulton, L.A. Chinwalla, A.T. Pepin, K.H. Gish, W.R. Chissoe, S.L. Wendl, M.C. Delehaunty, K.D. Miner, T.L. Delehaunty, A. Kramer, J.B. Cook, L.L. Fulton, R.S. Johnson, D.L. Minx, P.J. Clifton, S.W. Hawkins, T. Branscomb, E. Predki, P. Richardson, P. Wenning, S. Slezak, T. Doggett, N. Cheng, J.F. Olsen, A. Lucas, S. Elkin, C. Uberbacher, E. Frazier, M. Gibbs, R.A. Muzny, D.M. Scherer, S.E. Bouck, J.B. Sodergren, E.J. Worley, K.C. Rives, C.M. Gorrell, J.H. Metzker, M.L. Naylor, S.L. Kucherlapati, R.S. Nelson, D.L. Weinstock, G.M. Sakaki, Y. Fujiyama, A. Hattori, M. Yada, T. Toyoda, A. Itoh, T. Kawagoe, C. Watanabe, H. Totoki, Y. Taylor, T. Weissenbach, J. Heilig, R. Saurin, W. Artiguenave, F. Brottier, P. Bruls, T. Pelletier, E. Robert, C. Wincker, P. Smith, D.R. Doucette-Stamm, L. Rubenfield, M. Weinstock, K. Lee, H.M. Dubois, J. Rosenthal, A. Platzer, M. Nyakatura, G. Taudien, S. Rump, A. Yang, H. Yu, J. Wang, J. Huang, G. Gu, J. Hood, L. Rowen, L. Madan, A. Qin, S. Davis, R.W. Federspiel, N.A. Abola, A.P. Proctor, M.J. Myers, R.M. Schmutz, J. Dickson, M. Grimwood, J. Cox, D.R. Olson, M.V. Kaul, R. Raymond, C. Shimizu, N. Kawasaki, K. Minoshima, S. Evans, G.A. Athanasiou, M. Schultz, R. Roe, B.A. Chen, F. Pan, H. Ramser, J. Lehrach, H. Reinhardt, R. McCombie, W.R. de la Bastide, M. Dedhia, N. Blocker, H. Homischer, K. Nordsiek, G. Agarwala, R. Aravind, L. Bailey, J.A. Bateman, A. Batzoglou, S. Birney, E. Bork, P. Brown, D.G. Burge, C.B. Cerutti, L. Chen, H.C. Church, D. Clamp, M. Copley, R.R. Doerks, T. Eddy, S.R. Eichler, E.E. Furey, T.S. Galagan, J. Gilbert, J.G. Harmon, C. Hayashizaki, Y. Haussler, D. Hermjakob, H. Hokamp, K. Jang, W. Johnson, L.S. Jones, T.A. Kasif, S. Kaspryzk, A. Kennedy, S. Kent, W.J. Kitts, P. Koonin, E.V. Korf, I. Kulp, D. Lancet, D. Lowe, T.M. McLysaght, A. Mikkelsen, T. Moran, J.V. Mulder, N. Pollara, V.J. Ponting, C.P. Schuler, G. Schultz, J. Slater, G. Smit, A.F. Stupka, E. Szustakowski, J. Thierry-Mieg, D. Thierry-Mieg, J. Wagner, L. Wallis, J. Wheeler, R. Williams, A. Wolf, Y.I. Wolfe, K.H. Yang, S.P. Yeh, R.F. Collins, F. Guyer, M.S. Peterson, J. Felsenfeld, A. Wetterstrand, K.A. Patrinos, A. Morgan, M.J. de Jong, P. Catanese, J.J. Osoegawa, K. Shizuya, H. Choi, S. and Chen, Y.J. 2001. Initial sequencing and analysis of the human genome. Nature 409(6822): 860-921. Langaee, T. and Ronaghi, M. 2005. Genetic variation analyses by Pyrosequencing. Mutat Res 573(1-2): 96-102. Lanni, J.S. and Jacks, T. 1998. Characterization of the p53-dependent postmitotic checkpoint following spindle disruption. Mol Cell Biol 18(2): 1055-1064. Larson, G.P., Ding, Y., Cheng, L.S., Lundberg, C., Gagalang, V., Rivas, G., Geller, L., Weitzel, J., MacDonald, D., Archambeau, J., Slater, J., Neuberg, D., Daly, M.B., Angel, I., Benson, A.B., 3rd, Smith, K., Kirkwood, J.M., O'Dwyer, P.J., Raskay, B., Sutphen, R., Drew, R., Stewart, J.A., Werndli, J., Johnson, D., Ruckdeschel, J.C., Elston, R.C., and Krontiris, T.G. 2005. Genetic linkage of prostate cancer risk to the chromosome 3 region bearing FHIT. Cancer Res 65(3): 805-814. Li, J., Jiang, T., Mao, J.H., Balmain, A., Peterson, L., Harris, C., Rao, P.H., Havlak, P., Gibbs, R., and Cai, W.W. 2004. Genomic segmental polymorphisms in inbred mouse strains. Nat Genet 36(9): 952-954. Li, J., Yen, C., Liaw, D., Podsypanina, K., Bose, S., Wang, S.I., Puc, J., Miliaresis, C., Rodgers, L., McCombie, R., Bigner, S.H., Giovanella, B.C., Ittmann, M., Tycko, B., Hibshoosh, H., Wigler, M.H., and Parsons, R. 1997. PTEN, a putative protein tyrosine phosphatase gene mutated in human brain, breast, and prostate cancer. Science 275(5308): 1943-1947. Lillington, D.M., Kingston, J.E., Coen, P.G., Price, E., Hungerford, J., Domizio, P., Young, B.D., and Onadim, Z. 2003. Comparative genomic hybridization of 49 primary retinoblastoma tumors identifies chromosomal regions associated with histopathology, progression, and patient outcome. Genes Chromosomes Cancer 36(2): 121-128. Lindblad-Toh, K., Tanenbaum, D.M., Daly, M.J., Winchester, E., Lui, W.O., Villapakkam, A., Stanton, S.E., Larsson, C., Hudson, T.J., Johnson, B.E., Lander, E.S., and Meyerson, M. 2000a. Loss-of-heterozygosity analysis of small-cell lung carcinomas using single-nucleotide polymorphism arrays. Nat Biotechnol 18(9): 1001-1005. Lindblad-Toh, K. Wade, C.M. Mikkelsen, T.S. Karlsson, E.K. Jaffe, D.B. Kamal, M. Clamp, M. Chang, J.L. Kulbokas, E.J., 3rd Zody, M.C. Mauceli, E. Xie, X. Breen, M. Wayne, R.K. Ostrander, E.A. Ponting, C.P. Galibert, F. Smith, D.R. DeJong, P.J. Kirkness, E. Alvarez, P. Biagi, T. Brockman, W. Butler, J. Chin, C.W. Cook, A. Cuff, J. Daly, M.J. DeCaprio, D. Gnerre, S. Grabherr, M. Kellis, M. Kleber, M. Bardeleben, C. Goodstadt, L. Heger, A. Hitte, C. Kim, L. Koepfli, K.P. Parker, H.G. Pollinger, J.P. Searle, S.M. Sutter, N.B. Thomas, R. Webber, C. Baldwin, J. Abebe, A. Abouelleil, A. Aftuck, L. Ait-Zahra, M. Aldredge, T. Allen, N. An, P. Anderson, S. Antoine, C. Arachchi, H. Aslam, A. Ayotte, L. Bachantsang, P. Barry, A. Bayul, T. Benamara, M. Berlin, A. Bessette, D. Blitshteyn, B. Bloom, T. Blye, J. Boguslavskiy, L. Bonnet, C. Boukhgalter, B. Brown, A. Cahill, P. Calixte, N. Camarata, J. Cheshatsang, Y. Chu, J. Citroen, M. Collymore, A. Cooke, P. Dawoe, T. Daza, R. Decktor, K. DeGray, S. Dhargay, N. Dooley, K. Dooley, K. Dorje, P. Dorjee, K. Dorris, L. Duffey, N. Dupes, A. Egbiremolen, O. Elong, R. Falk, J. Farina, A. Faro, S. Ferguson, D. Ferreira, P. Fisher, S. FitzGerald, M. Foley, K. Foley, C. Franke, A. Friedrich, D. Gage, D. Garber, M. Gearin, G. Giannoukos, G. Goode, T. Goyette, A. Graham, J. Grandbois, E. Gyaltsen, K. Hafez, N. Hagopian, D. Hagos, B. Hall, J. Healy, C. Hegarty, R. Honan, T. Horn, A. Houde, N. Hughes, L. Hunnicutt, L. Husby, M. Jester, B. Jones, C. Kamat, A. Kanga, B. Kells, C. Khazanovich, D. Kieu, A.C. Kisner, P. Kumar, M. Lance, K. Landers, T. Lara, M. Lee, W. Leger, J.P. Lennon, N. Leuper, L. LeVine, S. Liu, J. Liu, X. Lokyitsang, Y. Lokyitsang, T. Lui, A. Macdonald, J. Major, J. Marabella, R. Maru, K. Matthews, C. McDonough, S. Mehta, T. Meldrim, J. Melnikov, A. Meneus, L. Mihalev, A. Mihova, T. Miller, K. Mittelman, R. Mlenga, V. Mulrain, L. Munson, G. Navidi, A. Naylor, J. Nguyen, T. Nguyen, N. Nguyen, C. Nguyen, T. Nicol, R. Norbu, N. Norbu, C. Novod, N. Nyima, T. Olandt, P. O'Neill, B. O'Neill, K. Osman, S. Oyono, L. Patti, C. Perrin, D. Phunkhang, P. Pierre, F. Priest, M. Rachupka, A. Raghuraman, S. Rameau, R. Ray, V. Raymond, C. Rege, F. Rise, C. Rogers, J. Rogov, P. Sahalie, J. Settipalli, S. Sharpe, T. Shea, T. Sheehan, M. Sherpa, N. Shi, J. Shih, D. Sloan, J. Smith, C. Sparrow, T. Stalker, J. Stange-Thomann, N. Stavropoulos, S. Stone, C. Stone, S. Sykes, S. Tchuinga, P. Tenzing, P. Tesfaye, S. Thoulutsang, D. Thoulutsang, Y. Topham, K. Topping, I. Tsamla, T. Vassiliev, H. Venkataraman, V. Vo, A. Wangchuk, T. Wangdi, T. Weiand, M. Wilkinson, J. Wilson, A. Yadav, S. Yang, S. Yang, X. Young, G. Yu, Q. Zainoun, J. Zembek, L. Zimmer, A. and Lander, E.S. 2005. Genome sequence, comparative analysis and haplotype structure of the domestic dog. Nature 438(7069): 803-819. Lindblad-Toh, K., Winchester, E., Daly, M.J., Wang, D.G., Hirschhorn, J.N., Laviolette, J.P., Ardlie, K., Reich, D.E., Robinson, E., Sklar, P., Shah, N., Thomas, D., Fan, J.B., Gingeras, T., Warrington, J., Patil, N., Hudson, T.J., and Lander, E.S. 2000b. Large-scale discovery and genotyping of single-nucleotide polymorphisms in the mouse. Nat Genet 24(4): 381-386. Lisitsyn, N. and Wigler, M. 1995. Representational difference analysis in detection of genetic lesions in cancer. Methods Enzymol 254: 291-304. Lisitsyn, N.A., Lisitsina, N.M., Dalbagni, G., Barker, P., Sanchez, C.A., Gnarra, J., Linehan, W.M., Reid, B.J., and Wigler, M.H. 1995. Comparative genomic analysis of tumors: detection of DNA losses and amplification. Proc Natl Acad Sci USA 92(1): 151-155. Livak, K.J. 1999. Allelic discrimination using fluorogenic probes and the 5' nuclease assay. Genet Anal 14(5-6): 143-149. Livak, K.J., Flood, S.J., Marmaro, J., Giusti, W., and Deetz, K. 1995. Oligonucleotides with fluorescent dyes at opposite ends provide a quenched probe system useful for detecting PCR product and nucleic acid hybridization. PCR Methods Appl 4(6): 357-362. Liyanage, M., Coleman, A., du Manoir, S., Veldman, T., McCormack, S., Dickson, R.B., Barlow, C., Wynshaw-Boris, A., Janz, S., Wienberg, J., Ferguson-Smith, M.A., Schrock, E., and Ried, T. 1996. Multicolour spectral karyotyping of mouse chromosomes. Nat Genet 14(3): 312-315. Lucito, R., Healy, J., Alexander, J., Reiner, A., Esposito, D., Chi, M., Rodgers, L., Brady, A., Sebat, J., Troge, J., West, J.A., Rostan, S., Nguyen, K.C., Powers, S., Ye, K.Q., Olshen, A., Venkatraman, E., Norton, L., and Wigler, M. 2003. Representational oligonucleotide microarray analysis: a high-resolution method to detect genome copy number variation. Genome Res 13(10): 2291-2305. Lucito, R., Nakimura, M., West, J.A., Han, Y., Chin, K., Jensen, K., McCombie, R., Gray, J.W., and Wigler, M. 1998. Genetic analysis using genomic representations. Proc Natl Acad Sci USA 95(8): 4487-4492. Lucito, R., West, J., Reiner, A., Alexander, J., Esposito, D., Mishra, B., Powers, S., Norton, L., and Wigler, M. 2000. Detecting gene copy number fluctuations in tumor cells by microarray analysis of genomic representations. Genome Res 10(11): 1726-1736. MacPherson, D., Conkrite, K., Tam, M., Mukai, S., Mu, D., and Jacks, T. submitted. Murine bilateral retinoblastoma exhibiting rapid onset, metastatic progression and N-myc gene amplification. Mairal, A., Pinglier, E., Gilbert, E., Peter, M., Validire, P., Desjardins, L., Doz, F., Aurias, A., and Couturier, J. 2000. Detection of chromosome imbalances in retinoblastoma by parallel karyotype and CGH analyses. Genes Chromosomes Cancer 28(4): 370-379. Makridakis, N.M. and Reichardt, J.K. 2001. Multiplex automated primer extension analysis: simultaneous genotyping of several polymorphisms. Biotechniques 31(6): 1374-1380. Manchester, K.L. 1995. Theodor Boveri and the origin of malignant tumours. Trends Cell Biol 5(10): 384-387. Manenti, G., Galbiati, F., Gianni-Barrera, R., Pettinicchio, A., Acevedo, A., and Dragani, T.A. 2004. Haplotype sharing suggests that a genomic segment containing six genes accounts for the pulmonary adenoma susceptibility 1 (Pas ) locus activity in mice. Oncogene 23(25): 4495-4504. Matsuzaki, H., Dong, S., Loi, H., Di, X., Liu, G., Hubbell, E., Law, J., Berntsen, T., Chadha, M., Hui, H., Yang, G., Kennedy, G.C., Webster, T.A., Cawley, S., Walsh, P.S., Jones, K.W., Fodor, S.P., and Mei, R. 2004. Genotyping over 100,000 SNPs on a pair of oligonucleotide arrays. Nat Methods 1(2): 109-111. Mei, R., Galipeau, P.C., Prass, C., Berno, A., Ghandour, G., Patil, N., Wolff, R.K., Chee, M.S., Reid, B.J., and Lockhart, D.J. 2000. Genome-wide detection of allelic imbalance using human SNPs and high-density DNA arrays. Genome Res 10(8): 1126-1137. Meraldi, P., Honda, R., and Nigg, E.A. 2002. Aurora-A overexpression reveals tetraploidization as a major route to centrosome amplification in p53-/- cells. Embo J 21(4): 483-492. Milne, B.S., Hoather, T., O'Brien, P.C., Yang, F., Ferguson-Smith, M.A., Dobson, J., and Sargan, D. 2004. Karyotype of canine soft tissue sarcomas: a multi-colour, multi- species approach to canine chromosome painting. Chromosome Res 12(8): 825- 835. Mitelman, F. 2000. Recurrent chromosome aberrations in cancer. Mutat Res 462(2-3): 247-253. Mitelman, F., Johansson, B., and Mertens, F. 2006. Mitelman Database of Chromosome Aberrations in Cancer. In. http://cgap.nci.nih.gov/Chromosomes/Mitelman. Mitra, N., Ye, T.Z., Smith, A., Chuai, S., Kirchhoff, T., Peterlongo, P., Nafa, K., Phillips, M.S., Offit, K., and Ellis, N.A. 2004. Localization of cancer susceptibility genes by genome-wide single-nucleotide polymorphism linkage-disequilibrium mapping. Cancer Res 64(21): 8116-8125. Moran, J.L., Bolton, A.D., Tran, P.V., Brown, A., Dwyer, N.D., Manning, D.K., Bjork, B.C., Li, C., Montgomery, K., Siepka, S.M., Vitaterna, M.H., Takahashi, J.S., Wiltshire, T., Kwiatkowski, D.J., Kucherlapati, R., and Beier, D.R. 2006. Utilization of a whole genome SNP panel for efficient genetic mapping in the mouse. Genome Res 16(3): 436-440. Mu, D., Chen, L., Zhang, X., See, L.H., Koch, C.M., Yen, C., Tong, J.J., Spiegel, L., Nguyen, K.C., Servoss, A., Peng, Y., Pei, L., Marks, J.R., Lowe, S., Hoey, T., Jan, L.Y., McCombie, W.R., Wigler, M.H., and Powers, S. 2003. Genomic amplification and oncogenic properties of the KCNK9 potassium channel gene. Cancer Cell 3(3): 297-302. Murray, S.S., Oliphant, A., Shen, R., McBride, C., Steeke, R.J., Shannon, S.G., Rubano, T., Kermani, B.G., Fan, J.B., Chee, M.S., and Hansen, M.S. 2004. A highly informative SNP linkage panel for human genetic studies. Nat Methods 1(2): 113- 117. NCI. 2005. The Cancer Genome Atlas. http://cancergenomenihgov Nikiforov, T.T., Rendle, R.B., Goelet, P., Rogers, Y.H., Kotewicz, M.L., Anderson, S., Trainor, G.L., and Knapp, M.R. 1994. Genetic Bit Analysis: a solid phase method for typing single nucleotide polymorphisms. Nucleic Acids Res 22(20): 4167- 4175. Norton, N., Williams, N.M., Williams, H.J., Spurlock, G., Kirov, G., Morris, D.W., Hoogendoorn, B., Owen, M.J., and O'Donovan, M.C. 2002. Universal, robust, highly quantitative SNP allele frequency measurement in DNA pools. Hum Genet 110(5): 471-478. Nowell, P.C. 1976. The clonal evolution of tumor cell populations. Science 194(4260): 23-28. Nowell, P.C. and Hungerford, D.A. 1960. A minute chromosome in human chronic granulocytic leukemia. Science 132: 1497. O'Hagan, R.C., Brennan, C.W., Strahs, A., Zhang, X., Kannan, K., Donovan, M., Cauwels, C., Sharpless, N.E., Wong, W.H., and Chin, L. 2003. Array comparative genome hybridization for tumor classification and gene discovery in mouse models of malignant melanoma. Cancer Res 63(17): 5352-5356. O'Hagan, R.C., Chang, S., Maser, R.S., Mohan, R., Artandi, S.E., Chin, L., and DePinho, R.A. 2002. Telomere dysfunction provokes regional amplification and deletion in cancer genomes. Cancer Cell 2(2): 149-155. Olivier, M. 2005. The Invader assay for SNP genotyping. Mutat Res 573(1-2): 103-110. Orum, H., Jakobsen, M.H., Koch, T., Vuust, J., and Borre, M.B. 1999. Detection of the factor V Leiden mutation by direct allele-specific hybridization of PCR amplicons to photoimmobilized locked nucleic acids. Clin Chem 45(11): 1898-1905. Owens, S.E., Broman, K.W., Wiltshire, T., Elmore, J.B., Bradley, K.M., Smith, J.R., and Southard-Smith, E.M. 2005. Genome-wide linkage identifies novel modifier loci of aganglionosis in the Sox0ODom model of Hirschsprung disease. Hum Mol Genet 14(11): 1549-1558. Paez, J.G., Janne, P.A., Lee, J.C., Tracy, S., Greulich, H., Gabriel, S., Herman, P., Kaye, F.J., Lindeman, N., Boggon, T.J., Naoki, K., Sasaki, H., Fujii, Y., Eck, M.J., Sellers, W.R., Johnson, B.E., and Meyerson, M. 2004. EGFR mutations in lung cancer: correlation with clinical response to gefitinib therapy. Science 304(5676): 1497-1500. Park, J.T., Li, M., Nakayama, K., Mao, T.L., Davidson, B., Zhang, Z., Kurman, R.J., Eberhart, C.G., Shih Ie, M., and Wang, T.L. 2006. Notch3 gene amplification in ovarian cancer. Cancer Res 66(12): 6312-6318. Pei, J., Balsara, B.R., Li, W., Litwin, S., Gabrielson, E., Feder, M., Jen, J., and Testa, J.R. 2001. Genomic imbalances in human lung adenocarcinomas and squamous cell carcinomas. Genes Chromosomes Cancer 31(3): 282-287. Petkov, P.M., Cassell, M.A., Sargent, E.E., Donnelly, C.J., Robinson, P., Crew, V., Asquith, S., Haar, R.V., and Wiles, M.V. 2004a. Development of a SNP genotyping panel for genetic monitoring of the laboratory mouse. Genomics 83(5): 902-911. Petkov, P.M., Ding, Y., Cassell, M.A., Zhang, W., Wagner, G., Sargent, E.E., Asquith, S., Crew, V., Johnson, K.A., Robinson, P., Scott, V.E., and Wiles, M.V. 2004b. An efficient SNP system for mouse genome scanning and elucidating strain relationships. Genome Res 14(9): 1806-1811. Pinkel, D. and Albertson, D.G. 2005. Array comparative genomic hybridization and its applications in cancer. Nat Genet 37 Suppl: S 11-17. Pinkel, D., Segraves, R., Sudar, D., Clark, S., Poole, I., Kowbel, D., Collins, C., Kuo, W.L., Chen, C., Zhai, Y., Dairkee, S.H., Ljung, B.M., Gray, J.W., and Albertson, D.G. 1998. High resolution analysis of DNA copy number variation using comparative genomic hybridization to microarrays. Nat Genet 20(2): 207-211. Pletcher, M.T., McClurg, P., Batalov, S., Su, A.I., Barnes, S.W., Lagler, E., Korstanje, R., Wang, X., Nusskern, D., Bogue, M.A., Mural, R.J., Paigen, B., and Wiltshire, T. 2004. Use of a dense single nucleotide polymorphism map for in silico mapping in the mouse. PLoS Biol 2(12): e393. Pollack, J.R., Perou, C.M., Alizadeh, A.A., Eisen, M.B., Pergamenschikov, A., Williams, C.F., Jeffrey, S.S., Botstein, D., and Brown, P.O. 1999. Genome-wide analysis of DNA copy-number changes using cDNA microarrays. Nat Genet 23(1): 41-46. Prince, J.A., Feuk, L., Howell, W.M., Jobs, M., Emahazion, T., Blennow, K., and Brookes, A.J. 2001. Robust and accurate single nucleotide polymorphism genotyping by dynamic allele-specific hybridization (DASH): design criteria and assay validation. Genome Res 11(1): 152-162. Pulciani, S., Santos, E., Long, L.K., Sorrentino, V., and Barbacid, M. 1985. ras gene Amplification and malignant transformation. Mol Cell Biol 5(10): 2836-2841. Radany, E.H., Hong, K., Kesharvarzi, S., Lander, E.S., and Bishop, J.M. 1997. Mouse mammary tumor virus/v-Ha-ras transgene-induced mammary tumors exhibit strain-specific allelic loss on mouse chromosome 4. Proc Natl Acad Sci USA 94(16): 8664-8669. Ranade, K., Chang, M.S., Ting, C.T., Pei, D., Hsiao, C.F., Olivier, M., Pesich, R., Hebert, J., Chen, Y.D., Dzau, V.J., Curb, D., Olshen, R., Risch, N., Cox, D.R., and Botstein, D. 2001. High-throughput genotyping with single nucleotide polymorphisms. Genome Res 11(7): 1262-1268. Reilly, K.M., Tuskan, R.G., Christy, E., Loisel, D.A., Ledger, J., Bronson, R.T., Smith, C.D., Tsang, S., Munroe, D.J., and Jacks, T. 2004. Susceptibility to astrocytoma in mice mutant for Nfl and Trp53 is linked to chromosome 11 and subject to epigenetic effects. Proc Natl Acad Sci USA 101(35): 13008-13013. Ross, P., Hall, L., Smirnov, I., and Haff, L. 1998. High level multiplex genotyping by MALDI-TOF mass spectrometry. Nat Biotechnol 16(13): 1347-1351. Ross, P.L., Lee, K., and Belgrader, P. 1997. Discrimination of single-nucleotide polymorphisms in human DNA using peptide nucleic acid probes detected by MALDI-TOF mass spectrometry. Anal Chem 69(20): 4197-4202. Saiki, R.K., Chang, C.A., Levenson, C.H., Warren, T.C., Boehm, C.D., Kazazian, H.H., Jr., and Erlich, H.A. 1988. Diagnosis of sickle cell anemia and beta-thalassemia with enzymatically amplified DNA and nonradioactive allele-specific oligonucleotide probes. NEngl JMed 319(9): 537-541. Samuels, Y., Wang, Z., Bardelli, A., Silliman, N., Ptak, J., Szabo, S., Yan, H., Gazdar, A., Powell, S.M., Riggins, G.J., Willson, J.K., Markowitz, S., Kinzler, K.W., Vogelstein, B., and Velculescu, V.E. 2004. High frequency of mutations of the PIK3CA gene in human cancers. Science 304(5670): 554. Sasaki, M. 1982. Current status of cytogenetic studies in animal tumors with special reference to nonrandom chromosome changes. Cancer Genet Cytogenet 5(2): 153-172. Schaid, D.J., Guenther, J.C., Christensen, G.B., Hebbring, S., Rosenow, C., Hilker, C.A., McDonnell, S.K., Cunningham, J.M., Slager, S.L., Blute, M.L., and Thibodeau, S.N. 2004. Comparison of microsatellites versus single-nucleotide polymorphisms in a genome linkage screen for prostate cancer-susceptibility Loci. Am JHum Genet 75(6): 948-965. Scholnick, S.B. and Richter, T.M. 2003. The role of CSMD1 in head and neck carcinogenesis. Genes Chromosomes Cancer 38(3): 281-283. Sellick, G.S., Webb, E.L., Allinson, R., Matutes, E., Dyer, M.J., Jonsson, V., Langerak, A.W., Mauro, F.R., Fuller, S., Wiley, J., Lyttelton, M., Callea, V., Yuille, M., Catovsky, D., and Houlston, R.S. 2005. A high-density SNP genomewide linkage scan for chronic lymphocytic leukemia-susceptibility loci. Am JHum Genet 77(3): 420-429. Selzer, R.R., Richmond, T.A., Pofahl, N.J., Green, R.D., Eis, P.S., Nair, P., Brothman, A.R., and Stallings, R.L. 2005. Analysis of chromosome breakpoints in neuroblastoma at sub-kilobase resolution using fine-tiling oligonucleotide array CGH. Genes Chromosomes Cancer 44(3): 305-319. Shao, C., Stambrook, P.J., and Tischfield, J.A. 2001. Mitotic recombination is suppressed by chromosomal divergence in hybrids of distantly related mouse strains. Nat Genet 28(2): 169-172. Shen, R., Fan, J.B., Campbell, D., Chang, W., Chen, J., Doucet, D., Yeakley, J., Bibikova, M., Wickham Garcia, E., McBride, C., Steemers, F., Garcia, F., Kermani, B.G., Gunderson, K., and Oliphant, A. 2005. High-throughput SNP genotyping on universal bead arrays. Mutat Res 573(1-2): 70-82. Shih, C., Padhy, L.C., Murray, M., and Weinberg, R.A. 1981. Transforming genes of carcinomas and neuroblastomas introduced into mouse fibroblasts. Nature 290(5803): 261-264. Shih le, M., Sheu, J.J., Santillan, A., Nakayama, K., Yen, M.J., Bristow, R.E., Vang, R., Parmigiani, G., Kurman, R.J., Trope, C.G., Davidson, B., and Wang, T.L. 2005. Amplification of a chromatin remodeling gene, Rsf- /HBXAP, in ovarian carcinoma. Proc Natl Acad Sci USA 102(39): 14004-14009. Shimkin, M.B. 1977. Contrary to nature: being an illustrated commentary on some persons and events of historical importance in the development of knowledge concerning. cancer. U.S. Dept. of Health, Education, and Welfare, Public Health Service, National Institutes of Health: for sale by Supt. of Docs., U.S. Govt. Print. Off., [Washington]. Shumaker, J.M., Metspalu, A., and Caskey, C.T. 1996. Mutation detection by solid phase primer extension. Hum Mutat 7(4): 346-354. Snijders, A.M., Nowak, N., Segraves, R., Blackwood, S., Brown, N., Conroy, J., Hamilton, G., Hindle, A.K., Huey, B., Kimura, K., Law, S., Myambo, K., Palmer, J., Ylstra, B., Yue, J.P., Gray, J.W., Jain, A.N., Pinkel, D., and Albertson, D.G. 2001. Assembly of microarrays for genome-wide measurement of DNA copy number. Nat Genet 29(3): 263-264. Snijders, A.M., Nowak, N.J., Huey, B., Fridlyand, J., Law, S., Conroy, J., Tokuyasu, T., Demir, K., Chiu, R., Mao, J.H., Jain, A.N., Jones, S.J., Balmain, A., Pinkel, D., and Albertson, D.G. 2005. Mapping segmental and sequence variations among laboratory mice using BAC array CGH. Genome Res 15(2): 302-311. Sokolov, B.P. 1990. Primer extension technique for the detection of single nucleotide in genomic DNA. Nucleic Acids Res 18(12): 3671. Solinas-Toldo, S., Lampel, S., Stilgenbauer, S., Nickolenko, J., Benner, A., Dohner, H., Cremer, T., and Lichter, P. 1997. Matrix-based comparative genomic hybridization: biochips to screen for genomic imbalances. Genes Chromosomes Cancer 20(4): 399-407. Soriano, P. 1999. Generalized lacZ expression with the ROSA26 Cre reporter strain. Nat Genet 21(1): 70-71. Sorrentino, V., McKinney, M.D., Drozdoff, V., Hume, C.R., and Fleissner, E. 1988. Spontaneous or carcinogen-mediated amplification of a mutated ras gene promotes neoplastic transformation. Oncogene Res 2(2): 189-195. Sosnowski, R.G., Tu, E., Butler, W.F., O'Connell, J.P., and Heller, M.J. 1997. Rapid determination of single base mismatch mutations in DNA hybrids by direct electric field control. Proc Natl Acad Sci USA 94(4): 1119-1123. Southern, E.M., Maskos, U., and Elder, J.K. 1992. Analyzing and comparing nucleic acid sequences by hybridization to arrays of oligonucleotides: evaluation using experimental models. Genomics 13(4): 1008-1017. Stephens, P., Hunter, C., Bignell, G., Edkins, S., Davies, H., Teague, J., Stevens, C., O'Meara, S., Smith, R., Parker, A., Barthorpe, A., Blow, M., Brackenbury, L., Butler, A., Clarke, O., Cole, J., Dicks, E., Dike, A., Drozd, A., Edwards, K., Forbes, S., Foster, R., Gray, K., Greenman, C., Halliday, K., Hills, K., Kosmidou, V., Lugg, R., Menzies, A., Perry, J., Petty, R., Raine, K., Ratford, L., Shepherd, R., Small, A., Stephens, Y., Tofts, C., Varian, J., West, S., Widaa, S., Yates, A., Brasseur, F., Cooper, C.S., Flanagan, A.M., Knowles, M., Leung, S.Y., Louis, D.N., Looijenga, L.H., Malkowicz, B., Pierotti, M.A., Teh, B., Chenevix-Trench, G., Weber, B.L., Yuen, S.T., Harris, G., Goldstraw, P., Nicholson, A.G., Futreal, P.A., Wooster, R., and Stratton, M.R. 2004. Lung cancer: intragenic ERBB2 kinase mutations in tumours. Nature 431(7008): 525-526. Sweet-Cordero, A., Mukherjee, S., Subramanian, A., You, H., Roix, J.J., Ladd-Acosta, C., Mesirov, J., Golub, T.R., and Jacks, T. 2005. An oncogenic KRAS2 expression signature identified by cross-species gene-expression analysis. Nat Genet 37(1): 48-55. Sweet-Cordero, A., Tseng, G.C., You, H., Douglass, M., Huey, B., Albertson, D., and Jacks, T. 2006. Comparison of gene expression and DNA copy number changes in a murine model of lung cancer. Genes Chromosomes Cancer 45(4): 338-348. Syvanen, A.C. 1999. From gels to chips: "minisequencing" primer extension for analysis of point mutations and single nucleotide polymorphisms. Hum Mutat 13(1): 1-10. -. 2001. Accessing genetic variation: genotyping single nucleotide polymorphisms. Nat Rev Genet 2(12): 930-942. Syvanen, A.C., Aalto-Setala, K., Harju, L., Kontula, K., and Soderlund, H. 1990. A primer-guided nucleotide incorporation assay in the genotyping of apolipoprotein E. Genomics 8(4): 684-692. Tai, A.L., Mak, W., Ng, P.K., Chua, D.T., Ng, M.Y., Fu, L., Chu, K.K., Fang, Y., Qiang Song, Y., Chen, M., Zhang, M., Sham, P.C., and Guan, X.Y. 2006. High- throughput loss-of-heterozygosity study of chromosome 3p in lung cancer using single-nucleotide polymorphism markers. Cancer Res 66(8): 4133-4138. Taton, T.A., Mirkin, C.A., and Letsinger, R.L. 2000. Scanometric DNA array detection with nanoparticle probes. Science 289(5485): 1757-1760. Telenius, H., Carter, N.P., Bebb, C.E., Nordenskjold, M., Ponder, B.A., and Tunnacliffe, A. 1992. Degenerate oligonucleotide-primed PCR: general amplification of target DNA by a single degenerate primer. Genomics 13(3): 718-725. Testa, J.R., Siegfried, J.M., Liu, Z., Hunt, J.D., Feder, M.M., Litwin, S., Zhou, J.Y., Taguchi, T., and Keller, S.M. 1994. Cytogenetic analysis of 63 non-small cell lung carcinomas: recurrent chromosome alterations amid frequent and widespread genomic upheaval. Genes Chromosomes Cancer 11(3): 178-194. Thomas, R., Scott, A., Langford, C.F., Fosmire, S.P., Jubala, C.M., Lorentzen, T.D., Hitte, C., Karlsson, E.K., Kirkness, E., Ostrander, E.A., Galibert, F., Lindblad- Toh, K., Modiano, J.F., and Breen, M. 2005. Construction of a 2-Mb resolution BAC microarray for CGH analysis of canine tumors. Genome Res 15(12): 1831- 1837. Thomas, R., Smith, K.C., Ostrander, E.A., Galibert, F., and Breen, M. 2003. Chromosome aberrations in canine multicentric lymphomas detected with comparative genomic hybridisation and a panel of single locus probes. Br J Cancer 89(8): 1530-1537. Tonon, G., Wong, K.K., Maulik, G., Brennan, C., Feng, B., Zhang, Y., Khatry, D.B., Protopopov, A., You, M.J., Aguirre, A.J., Martin, E.S., Yang, Z., Ji, H., Chin, L., and Depinho, R.A. 2005. High-resolution genomic profiles of human lung cancer. Proc Natl Acad Sci USA 102(27): 9625-9630. Trkova, M., Foretova, L., Kodet, R., Hedvicakova, P., and Sedlacek, Z. 2003. A Li- Fraumeni syndrome family with retained heterozygosity for a germline TP53 mutation in two tumors. Cancer Genet Cytogenet 145(1): 60-64. Tyagi, S., Bratu, D.P., and Kramer, F.R. 1998. Multicolor molecular beacons for allele discrimination. Nat Biotechnol 16(1): 49-53. Tyagi, S. and Kramer, F.R. 1996. Molecular beacons: probes that fluoresce upon hybridization. Nat Biotechnol 14(3): 303-308. van den Ijssel, P., Tijssen, M., Chin, S.F., Eijk, P., Carvalho, B., Hopmans, E., Holstege, H., Bangarusamy, D.K., Jonkers, J., Meijer, G.A., Caldas, C., and Ylstra, B. 2005. Human and mouse oligonucleotide-based array CGH. Nucleic Acids Res 33(22): e192. Van Dyke, T. and Jacks, T. 2002. Cancer modeling in the modem era: progress and challenges. Cell 108(2): 135-144. Velculescu, V.E., Zhang, L., Vogelstein, B., and Kinzler, K.W. 1995. Serial analysis of gene expression. Science 270(5235): 484-487. Venkatachalam, S., Shi, Y.P., Jones, S.N., Vogel, H., Bradley, A., Pinkel, D., and Donehower, L.A. 1998. Retention of wild-type p53 in tumors from p53 heterozygous mice: reduction of p53 dosage can promote cancer formation. Embo J17(16): 4657-4667. von Ahsen, N., Oellerich, M., Armstrong, V.W., and Schutz, E. 1999. Application of a thermodynamic nearest-neighbor model to estimate nucleic acid stability and optimize probe design: prediction of melting points of multiple mutations of apolipoprotein B-3500 and factor V with a hybridization probe genotyping assay on the LightCycler. Clin Chem 45(12): 2094-2101. Wade, C.M., Kulbokas, E.J., 3rd, Kirby, A.W., Zody, M.C., Mullikin, J.C., Lander, E.S., Lindblad-Toh, K., and Daly, M.J. 2002. The mosaic structure of variation in the laboratory mouse genome. Nature 420(6915): 574-578. Waggoner, A. 2006. Fluorescent labels for proteomics and genomics. Curr Opin Chem Biol 10(1): 62-66. Wang, T.L., Maierhofer, C., Speicher, M.R., Lengauer, C., Vogelstein, B., Kinzler, K.W., and Velculescu, V.E. 2002. Digital karyotyping. Proc Natl Acad Sci USA 99(25): 16156-16161. Wang, Z., Shen, D., Parsons, D.W., Bardelli, A., Sager, J., Szabo, S., Ptak, J., Silliman, N., Peters, B.A., van der Heijden, M.S., Parmigiani, G., Yan, H., Wang, T.L., Riggins, G., Powell, S.M., Willson, J.K., Markowitz, S., Kinzler, K.W., Vogelstein, B., and Velculescu, V.E. 2004. Mutational analysis of the tyrosine phosphatome in colorectal cancers. Science 304(5674): 1164-1166. Waterfall, C.M. and Cobb, B.D. 2001. Single tube genotyping of sickle cell anaemia using PCR-based SNP analysis. Nucleic Acids Res 29(23): E119. Waterston, R.H. Lindblad-Toh, K. Birney, E. Rogers, J. Abril, J.F. Agarwal, P. Agarwala, R. Ainscough, R. Alexandersson, M. An, P. Antonarakis, S.E. Attwood, J. Baertsch, R. Bailey, J. Barlow, K. Beck, S. Berry, E. Birren, B. Bloom, T. Bork, P. Botcherby, M. Bray, N. Brent, M.R. Brown, D.G. Brown, S.D. Bult, C. Burton, J. Butler, J. Campbell, R.D. Carninci, P. Cawley, S. Chiaromonte, F. Chinwalla, A.T. Church, D.M. Clamp, M. Clee, C. Collins, F.S. Cook, L.L. Copley, R.R. Coulson, A. Couronne, O. Cuff, J. Curwen, V. Cutts, T. Daly, M. David, R. Davies, J. Delehaunty, K.D. Deri, J. Dermitzakis, E.T. Dewey, C. Dickens, N.J. Diekhans, M. Dodge, S. Dubchak, I. Dunn, D.M. Eddy, S.R. Elnitski, L. Emes, R.D. Eswara, P. Eyras, E. Felsenfeld, A. Fewell, G.A. Flicek, P. Foley, K. Frankel, W.N. Fulton, L.A. Fulton, R.S. Furey, T.S. Gage, D. Gibbs, R.A. Glusman, G. Gnerre, S. Goldman, N. Goodstadt, L. Grafham, D. Graves, T.A. Green, E.D. Gregory, S. Guigo, R. Guyer, M. Hardison, R.C. Haussler, D. Hayashizaki, Y. Hillier, L.W. Hinrichs, A. Hlavina, W. Holzer, T. Hsu, F. Hua, A. Hubbard, T. Hunt, A. Jackson, I. Jaffe, D.B. Johnson, L.S. Jones, M. Jones, T.A. Joy, A. Kamal, M. Karlsson, E.K. Karolchik, D. Kasprzyk, A. Kawai, J. Keibler, E. Kells, C. Kent, W.J. Kirby, A. Kolbe, D.L. Korf, I. Kucherlapati, R.S. Kulbokas, E.J. Kulp, D. Landers, T. Leger, J.P. Leonard, S. Letunic, I. Levine, R. Li, J. Li, M. Lloyd, C. Lucas, S. Ma, B. Maglott, D.R. Mardis, E.R. Matthews, L. Mauceli, E. Mayer, J.H. McCarthy, M. McCombie, W.R. McLaren, S. McLay, K. McPherson, J.D. Meldrim, J. Meredith, B. Mesirov, J.P. Miller, W. Miner, T.L. Mongin, E. Montgomery, K.T. Morgan, M. Mott, R. Mullikin, J.C. Muzny, D.M. Nash, W.E. Nelson, J.O. Nhan, M.N. Nicol, R. Ning, Z. Nusbaum, C. O'Connor, M.J. Okazaki, Y. Oliver, K. Overton-Larty, E. Pachter, L. Parra, G. Pepin, K.H. Peterson, J. Pevzner, P. Plumb, R. Pohl, C.S. Poliakov, A. Ponce, T.C. Ponting, C.P. Potter, S. Quail, M. Reymond, A. Roe, B.A. Roskin, K.M. Rubin, E.M. Rust, A.G. Santos, R. Sapojnikov, V. Schultz, B. Schultz, J. Schwartz, M.S. Schwartz, S. Scott, C. Seaman, S. Searle, S. Sharpe, T. Sheridan, A. Shownkeen, R. Sims, S. Singer, J.B. Slater, G. Smit, A. Smith, D.R. Spencer, B. Stabenau, A. Stange-Thomann, N. Sugnet, C. Suyama, M. Tesler, G. Thompson, J. Torrents, D. Trevaskis, E. Tromp, J. Ucla, C. Ureta-Vidal, A. Vinson, J.P. Von Niederhausern, A.C. Wade, C.M. Wall, M. Weber, R.J. Weiss, R.B. Wendl, M.C. West, A.P. Wetterstrand, K. Wheeler, R. Whelan, S. Wierzbowski, J. Willey, D. Williams, S. Wilson, R.K. Winter, E. Worley, K.C. Wyman, D. Yang, S. Yang, S.P. Zdobnov, E.M. Zody, M.C. and Lander, E.S. 2002. Initial sequencing and comparative analysis of the mouse genome. Nature 420(6915): 520-562. Weiss, W.A., Aldape, K., Mohapatra, G., Feuerstein, B.G., and Bishop, J.M. 1997. Targeted expression of MYCN causes neuroblastoma in transgenic mice. Embo J 16(11): 2985-2995. Wilhelm, M., Veltman, J.A., Olshen, A.B., Jain, A.N., Moore, D.H., Presti, J.C., Jr., Kovacs, G., and Waldman, F.M. 2002. Array-based comparative genomic hybridization for the differential diagnosis of renal cell cancer. Cancer Res 62(4): 957-960. Wiltshire, T., Pletcher, M.T., Batalov, S., Barnes, S.W., Tarantino, L.M., Cooke, M.P., Wu, H., Smylie, K., Santrosyan, A., Copeland, N.G., Jenkins, N.A., Kalush, F., Mural, R.J., Glynne, R.J., Kay, S.A., Adams, M.D., and Fletcher, C.F. 2003. Genome-wide single-nucleotide polymorphism analysis defines haplotype patterns in mouse. Proc Natl Acad Sci USA 100(6): 3380-3385. Wong, M.P., Fung, L.F., Wang, E., Chow, W.S., Chiu, S.W., Lam, W.K., Ho, K.K., Ma, E.S., Wan, T.S., and Chung, L.P. 2003. Chromosomal aberrations of primary lung adenocarcinomas in nonsmokers. Cancer 97(5): 1263-1270. Wu, Y., Renard, C.A., Apiou, F., Huerre, M., Tiollais, P., Dutrillaux, B., and Buendia, M.A. 2002. Recurrent allelic deletions at mouse chromosomes 4 and 14 in Myc- induced liver tumors. Oncogene 21(10): 1518-1526. Xu, H., Sha, M.Y., Wong, E.Y., Uphoff, J., Xu, Y., Treadway, J.A., Truong, A., O'Brien, E., Asquith, S., Stubbins, M., Spurr, N.K., Lai, E.H., and Mahoney, W. 2003. Multiplexed SNP genotyping using the Qbead system: a quantum dot-encoded microsphere-based assay. Nucleic Acids Res 31(8): e43. Ylstra, B., van den Ijssel, P., Carvalho, B., Brakenhoff, R.H., and Meijer, G.A. 2006. BAC to the future! or oligonucleotides: a perspective for micro array comparative genomic hybridization (array CGH). Nucleic Acids Res 34(2): 445-450. You, M.J., Castrillon, D.H., Bastian, B.C., O'Hagan, R.C., Bosenberg, M.W., Parsons, R., Chin, L., and DePinho, R.A. 2002. Genetic analysis of Pten and Ink4a/Arf interactions in the suppression of tumorigenesis in mice. Proc Natl Acad Sci U S A 99(3): 1455-1460. Zabarovsky, E.R., Lerman, M.I., and Minna, J.D. 2002. Tumor suppressor genes on chromosome 3p involved in the pathogenesis of lung and other cancers. Oncogene 21(45): 6915-6935. Zhang, J., Hunter, K.W., Gandolph, M., Rowe, W.L., Finney, R.P., Kelley, J.M., Edmonson, M., and Buetow, K.H. 2005. A high-resolution multistrain haplotype analysis of laboratory mouse genome reveals three distinctive genetic variation patterns. Genome Res 15(2): 241-249. Zhang, Z., Wang, Y., Vikis, H.G., Johnson, L., Liu, G., Li, J., Anderson, M.W., Sills, R.C., Hong, H.L., Devereux, T.R., Jacks, T., Guan, K.L., and You, M. 2001. Wildtype Kras2 can inhibit lung carcinogenesis in mice. Nat Genet 29(1): 25-33. Zhao, X., Li, C., Paez, J.G., Chin, K., Janne, P.A., Chen, T.H., Girard, L., Minna, J., Christiani, D., Leo, C., Gray, J.W., Sellers, W.R, and Meyerson, M. 2004. An integrated view of copy number and allelic alterations in the cancer genome using single nucleotide polymorphism arrays. Cancer Res 64(9): 3060-3071. Zhao, X., Weir, B.A., LaFramboise, T., Lin, M., Beroukhim, R., Garraway, L., Beheshti, J., Lee, J.C., Naoki, K., Richards, W.G., Sugarbaker, D., Chen, F., Rubin, M.A., Janne, P.A., Girard, L., Minna, J., Christiani, D., Li, C., Sellers, W.R., and Meyerson, M. 2005. Homozygous deletions and chromosome amplifications in human lung carcinomas revealed by single nucleotide polymorphism array analysis. Cancer Res 65(13): 5561-5570. Zielinski, B., Gratias, S., Toedt, G., Mendrzyk, F., Stange, D.E., Radlwimmer, B., Lohmann, D.R., and Lichter, P. 2005. Detection of chromosomal imbalances in retinoblastoma by matrix-based comparative genomic hybridization. Genes Chromosomes Cancer 43(3): 294-301. Chapter 2 Mouse Representational Oligonucleotide Microarray Analysis Detects Copy Number Alterations in Murine Tumors of the Lung and Retina Mandy Tam', David MacPherson', Alice Shaw l , David Mu 2 , and Tyler Jacks 1 ' 3 1 Massachusetts Institute of Technology, Department of Biology and Center for Cancer Research, 77 Massachusetts Avenue, Cambridge, MA 02139 USA 2 Cold Spring Harbor Laboratory, Cold Spring Harbor, NY 11724, USA 3 Howard Hughes Medical Institute, 4000 James Bridge Road, Chevy Chase, MD 20185, USA The author prepared lung tumor DNA samples for analysis and performed Southern blotting validation of the retinoblastoma data. Retinoblastoma DNA was prepared by D. Macpherson. ROMA hybridizations were performed by D.Mu. The Kras mouse colony was maintained by A. Shaw. Abstract A wide range of copy number alterations in primary mouse tumors has been previously documented, including single copy gain or loss of entire chromosomes, partial gain or loss of a chromosome, high-amplitude focal amplifications, to low-level small deletions. To cover this broad spectrum, a genome-wide high-resolution CGH tool would be invaluable for mouse cancer DNA anlayses. Representational Oligonucleotide Microarray Analysis (ROMA) employs an integrated genome complexity reduction step that can enhance signal to noise ratio during genome hybridization. Among the many CGH platforms for genome-wide copy number analysis studies in human, ROMA has one of the highest resolving power averaging at 30kb (Lucito et al. 2003). We tested the mouse version of ROMA on mouse retinoblasomas and lung adenomas. We were able to detect a focal high-amplitude (>4.6fold) N-Myc amplification in retinoblastomas of a Rb/p130 DKO model, as well as numerous whole-chromosomal gains and losses in the same retinoblastoma sample set and in the lung tumors driven by a Kras mutation. Introduction Comparative Genomic Hybridization (CGH) is a technique that measures changes in the amount of DNA throughout the whole genome. Chromosomal copy number imbalances are commonly observed in cancer, particularly in carcinomas and these imbalances are detectable by CGH. The format of CGH has evolved from using metaphase spreads to microarrays as hybridization for differentially labeled tumor and normal genomes (Kallioniemi et al. 1992). Several microarray formats have been developed using probes that range from large-insert clones such as BACs (bacterial artificial chromosomes), PACs (P1 artificial chromosomes), or YACs (yeast artificial chromosomes), single- stranded DNA, cDNAs, and oligonucleotides (Solinas-Toldo et al. 1997; Pinkel et al. 1998; Pollack et al. 1999; Barrett et al. 2004; Dhami et al. 2005). Over the past few years, BAC-based and oligonucleotide-based arrays predominate the race for better resolution. Representational Oligonucleotide MicroArray (ROMA) is among one of the competing technologies that has proven useful in detecting genetic lesions in cancer (Lucito et al. 2003). ROMA involves the generation of low-complexity representations of the genomes to reduce hybridization noise. A sub-Mb resolution can be achieved (Lucito et al. 2003). In many human cancers, large-scale chromosomal abnormalities are common but identification of the critical regions is often difficult. Comparative genomic studies between mouse and human may be one way to aid this search. Individual chromosomes in the mouse can be delimited into regions of synteny that are conserved in human on separate chromosomes. Given this structure and the high overall conservation between mouse and human, genomic studies in mice may provide interesting insights into cancer- associated genetic lesions in human. The high-resolution ROMA technology has been translated for use in the mouse. In the present study, we utilized this platform to characterize two different types of mouse cancer: 1) retinoblastomas in mice doubly deleted for Rb and p130 in the retina (MacPherson et al. submitted), and 2) non-small cell lung cancer (NSCLC) in mice conditionally expressing a KrasG ' 2D mutant allele in the lung (Jackson et al. 2005). We sought to characterize the genetic alterations that may cooperate with these initiating genetic lesions. In humans, recurrent large-scale chromosomal imbalances have been observed. In retinoblastomas, chromosomal gains include those in 6p, lq, 2p and loss is frequently seen in 16q (Mairal et al. 2000; Chen et al. 2001; Lillington et al. 2003; Zielinski et al. 2005). In NSCLC, chromosomal gains in lq, 3q, 5p, 8q, and losses in 3p, 8p, 9q, 13q, 17p have been identified (Balsara and Testa 2002). In this chapter, we report the usefulness of ROMA in identifying single gene amplification as well as whole chromosomal changes in tumor-derived DNA from these mouse models. Results Representational oligonucleotide microarray analysis (ROMA) is a tool to detect copy number changes in the genome (Lucito et al. 2003). The technique involves the hybridization of low-complexity representations of tumor vs. normal genomes to oligonucleotide probes on microarrays (Figure 1). The usefulness of the ROMA platform to detect DNA copy number changes has been demonstrated (Jobanputra et al. 2005). Figure 1: Schematic for Representational Oligonucleotide MicroArray (ROMA) analysis A) Generation of low complexity representations (LCR) of the genomes. Tumor and normal DNA was digested with BglI and amplified by linker-based PCR. B) Tumor and normal LCRs were differentially labeled with Cy-5 and Cy-3 fluorescent dyes. Hybridization of the tumor vs. normal LCRs was performed on 84K oligonucleotide arrays. Each oligonucleotide was designed to bind selected BglII fragments in the mouse genome. A Representations (Reps) of Genome m I Restriction Digestion - - Adapter ligation/PCR l ~ Reps Reps of tumor DNA (cy5 labeled) + Reps of normal DNA (cy3 labeled) $ Hybridized to microarray Oligonucleotide to a single element of genome representation Data Analysis Groups in Cold Spring Harbor Laboratory and Nimblegen Systems Inc. have since translated the ROMA technique for genomic analysis in the mouse. This study reports the implementation of ROMA to characterize copy number changes in tumor DNA from two different mouse cancer models. Detection of N-Myc amplification in mouse retinoblastomas Retinoblastomas can be induced in mice with a retina-specific Rb deletion combined with an inactivation of p130 (MacPherson et al., submitted). As described by the authors, in this Rb/p130 DKO model, tumors could be consistently observed with a latency of 128?.18 days (means?s.d.). Early tumors arose in the periphery of the retina by PND21 when retinal development was completed. The tumors continued to progress in the adult mice and filled the posterior and anterior chambers of the eyes. At the experimental end point, tumors cells could be seen infiltrating the optic nerve. In addition, metastases to lymph nodes were observed in 38% of the animals. To identify genetic lesions that cooperate with Rb and p130 deletions in this retinoblastoma model, ROMA was performed on 8 lymph node metastases. Observed regions of copy number gain and loss are summarized in Table 1. Recurrent changes included whole chromosome 1 and chromosome 12 gains, which was each found in 4/8 tumors. In addition, focal amplifications in 12qAl.1 was detected in DNA from three tumors: 9806, 4836 and Drbl3. The amplicon sizes in the respective order were 1.9MB, 3.3Mb and 451kb with a 136kb minimal overlap, which harbors the N-myc oncogene (Figure 2A and B). Amplification of N-Myc was verified by Southern blotting. 3/6 tumor DNA samples in the Southern analysis have been profiled by ROMA, including those of Table 1. Summary of chromosomal changes in 9 metastatic retinoblastomas from Rb/pl30DKO mice. tumor ID gain amplification loss 9806 1, 12 12qA1.1, 12qF2 7217 12 4834 10qA4qter 2, 12, 18, 9qA5.3qter, 4qB3qter 4726 1 4848 1, 12, 19 4827 12qF2, 12qCl 4836a* 1 12qA1.1, 3qf3, 12qF1-2 3qA3, 17qe2, 17qe1.1 drbl3* 12 12qA1.1 Amplicons at 12qAl.1 harboring N-myc gene are in bold # Samples selected for ROMA analysis based on presence of N-myc amplification detected by Southern blot *Tail DNA used for ROMA hybridization was not from the tumor-containing mouse, thus, polymorphisms could contribute to focal changes Figure 2: Detection of N-Myc amplification in metastatic retinoblastomas A) Whole-genome copy number ratio plot of Drbl3 tumor vs. normal DNA. The Y- axis depicts the moving median fluorescence ratios of Cy5 labeled tumor to Cy3 labeled tail DNA. The X-axis is an index of the probes genomic order based on the UCSC mouse May 2004 (mm5) annotated assembly. Whole chromosome 12 is increased in copy ratio (labeled). Within the chromosome, a focally amplified region is also observed (arrow) and is mapped to 12qAl.1. B) A minimally overlapping region of amplification at 12qA1.1 in 3 retinoblastomas (9806, 4836, and drbl3). The 136kb core amplicon is highlighted in the zoomed- in copy ratio plots of the 3 samples. The X-axis marks the nucleotide coordinates on chromosome 12 and the Y-axis is fluorescence ratio. N-Myc is the only known RefSeq gene residing in the area. C) Southern analysis of N-Myc locus on DNA from 6 tumors. 3/6 tumors (Drbl3, 4836, and 4827) have been profiled with ROMA. To control for sample loading, blot was re-probed for Rosa26 locus on chromosome 6, a chromosome that did not show any copy number changes in the retinoblastoma samples. Relative signal of N-Myc to Rosa26 probes was quantified on phosphorimages with the ImageQuant software. Using wild-type spleen DNA as normal, fold of Myc amplification is shown. Samples with high-magnitude amplification are marked by asterisks. 0 500 1000 1500 2000 2500 WHOLE.GENOME QI 0a H Myc sa26 Fold ation 4.6 1.3 15.7 1.03 0.63 * * 1 2000000 03 130003000 1500000000 1700000000C 1200000000 1400000000 1600000000 Drbl3 l2qAl.1 2qA1.1 0 0 +12 "I A 10 8 7 6 5 4 3 100 - 9 8 7 6 C N- Ro. Amplifica Drbl 3 10 o 9 8 7 6 Drbl3 and 4836. A 4.6 and 15.7-fold increase of N-Myc copy number was seen in Drbl3 and 4836 tumor DNA respectively, confirming amplification data from ROMA (Figure 2C). Detection of recurrent whole chromosomal changes in primary lung tumors We performed a CGH study to identify genetic lesions that collaborate with a Kras mutation during lung cancer development in the mouse. We also assessed if a germline p53 mutation or loss can lead to different mutation spectra in these tumors. An inducible KrasG 12 D mouse model was used for the study. To induce lung tumors, KrasLSLGl 2 D;p53 +/+, KrasLSLG12D;p53 +/- and KrasLSLG 12 D;p5 3 R 27 0H/ + mice were intranasally infected with adenovirus Cre. 28 primary lung tumors were obtained from animals between 22-24 weeks of age. The majority of the tumors were early-stage with uniform nuclei, graded 1-2 according to the scale described in Jackson et al. 2005. As shown in Table 2, only 3/28 tumors contained grade 3 characteristics and 1/28 tumor was graded 4. The histological distribution is comparable between tumors coming from mice of the 3 genotypes. ROMA was performed to assess copy number alterations in tumors from mice of the 3 different genotypes. Tumor and normal DNA pairs were subjected to ROMA. Copy number gains and/or losses were observed in 10 out of 28 tumors. These changes graded 2 or higher. Among the changes as summarized in Figure 3, whole chromosomal copy alterations composed the majority. Chromosome 6 gain was the most frequent, found in 8/10 tumors that contained any changes. The second most common copy number Table 2: Characteristics of lung tumors analyzed The table lists the genotype of the mice from which individual lung tumors were dissected out. Histological grading of the tumors was assigned using the criteria described previously (Jackson et al. 2005). Tumors that exhibited any copy number changes by ROMA are highlighted. Mouse genotype Histology Grade 0-1, some normal tissue attached Grade 0-1, some normal tissue attached Grade 1 Grade 1 Grade 1 Grade 2 Grade Grade Grade 1-2 2 2 I26J5 1 KraS42 , p5.0a 1 1265a KrasG 12 D, p53R 2 7 0H/+ Grade 1-2 1232c KrasG12D, p 53 R270H/+ Grade 1-2 1232a Kras , p53" ? +? Grade 2 1291b Kras G12D , p 5 3 R270H/+ Grade 2, infiltrating lymphocytes 1291a Kras "", p53"o Grade 2 1782a KrasG2D, p53R270H/+ Grade 2+-3 1782b Krasc, p53 + n/a LKR10 KrasmA * tumor derived cell line LKR13 Kras" 1 * tumor derived cell line *LKR10 and LKR13 cell lines were derived from the same KrasLAl mouse. Kras ( i z D KrasG 12 D Kras G1 2 D Kras G1 2 D Kras G 1 2 D Kras G12 D KraS~lD Tumor ID 1186al 1186b1 1302b 1327a 1327b 1 1d 4 9h iLZ I SI.l 1275a 1248a 1278a KrasG 12 D, p5 3/ Kras G1 2 D, p53 +/ KrasG 12 D , D53 / Histologygenotype Figure 3: Summary of genetic changes seen primary lung tumors and lung tumor- derived cell lines Genetic changes observed in 9 of the tested tumors and 2 cell lines are summarized in a grid. Each row is an individual sample. Genotypes of mice that gave rise to the tumors were indicated. The two independently maintained cell lines LKR10O and LKR13 from the same mouse had almost identical ROMA profiles. The cell lines likely have the same clonal origin, thus their data are presented in same row (*) of this grid. The columns of the grid represent different chromosomes. Grey boxes indicate silence; solid orange means whole chromosomal gain; solid blue is whole chromosomal loss. Sub- chromosomal gains and losses are illustrated with shaded orange and blue respectively, and the region(s) altered are labeled inside the corresponding sample/chromosome box. EU~ .0 EU EU U E U m C C f N N % r% n N V4 1-4'. 4 4 0 0A 04 M M In z N N Iq C N N N' N '"4 , '. 4 '. ... I. ..A ?sd -1+ Esd +/HOL/Z ESd Ca]ZTE)se~d-> (ajZT9)sedi-> (aZTE)sed1->s c sjownI Ajewu~d U 0 V4 '-4 '-4 rq V4 m (A, InI N '-4 Eo 0O U m N i-I o o 0 2 E 0C 0 .o IL C r o 0 E c o Wu Q) 01 0 L- E alteration was chromosome 12 gain, which was found in 5/10 tumors. Other recurrent changes in order of prevalence are chromosomes 19 gain (3/10), 3 gain (2/10), 16 gain (2/10), 9 loss (2/10), and 11 loss (2/10). Interestingly, among the 28 tumors analyzed, one tumor 1247d showed copy number changes in multiple focal regions. Representative data are shown in Figure 4. p53 loss or mutation has been shown to accelerate lung tumor progression in the Kras mouse model. When compared to the lungs of KrasG12D;p53+/- mice, Kras12D;p53 R27 0 H/ + lungs showed an increase in tumor number and an increased proportion of higher grade tumors (Jackson et al. 2005). Because p53 functions in multiple pathways that maintain genomic integrity, we questioned whether the differences in lung tumorigenesis kinetics in KrasGl 2 D;p53+/+, KrasGl 2 D;p53+/- and KrasG12D;p53 R2 7 0H / + mice can be explained by elevated genomic instability due to p53 loss or mutation. In our study, it appeared that histological grading best correlated with the presence or absence of genetic changes, and that germline p53 genotype of the mice did not affect the spectra of lesions significantly. Tumors that showed any changes were of grade 2 or higher, with one exception, NT a, which contained a mix of grade 1 and 2 characteristics. The genotypes of mice that gave rise to the tumors did not appear to affect this trend for the most part. The one note was while all grade 2-3 tumors in KrasG2D;p53+/+ and KrasG1 2 D;p53+/- mice exhibited ROMA changes, two lesions of comparable grade Kras G2D;p 5 3 R270H/+ xsmice (1291a, 1782a) were silent. It remains unclear whether this silence was due to the difference in genotype, or a result of varying Figure 4: Representative ROMA moving median plots of lung tumors A moving median plot shows Cy5 to Cy3 signal ratios of from labeled tumor vs. tail DNA. The Y-axis is the logl0 fluoresence ratio and the X-axis is an index of the probes genomic order, based on UCSC mouse May 2004 (mm5) annotated assembly. Data from same chromosome are labeled with same color. A) Moving median plot of 1247d tumor genome, representative of samples showing sub- chromosomal gains and losses. B) Moving median plot of NTla tumor genome, representative of samples showing whole chromosomal gains and losses. 00 S2- z 4a UN. -3 1I rQ IU S560 1000 1500 2000 2500 Whole Mouse Genome B NTla 2 I,,,, ,,CI ID -0 rr 00 -1 Ug Zo -3 0 500 1000 1500 2000 2500 Whole Mouse Genome a i i i i A a i i i i I.L41U o o a r i 1 0 i0 stromal contamination in tumors that might have muted ROMA signals. Human subjectivity in histological grading should also be considered. Detection of focal changes in mouse lung cancer cell lines In addition to the 28 primary lung tumors, we performed ROMA to analyze copy number changes in two mouse lung cancer cell lines, LKR10 and LKR13. The cell lines were derived from the tumor-bearing lungs of one LAl mouse, which carried a latent allele of Kras G12D that got spontaneously activated by recombination (Johnson et al. 2001). In DNA from both cell lines, ROMA detected increased copies of whole chromosomes 6 and 19, which also showed recurrent gains in the primary samples. In addition, ROMA revealed multiple focal changes in DNA copy number in both cell lines that were unseen in most of the primary tumors (Figure 3). The sizes of lesions range from 0.095 to 3.9Mb, each containing one or more gene or EST. The changes were nearly identical in the two cell lines, suggesting a common clonal origin. Apparent phenotypic differences of these two cell lines were likely due to smaller-size genetic changes missed by ROMA and/or epigenetic differences. Common deletion of Csmdl in mouse and human lung cancer cell lines We compared data from our mouse lung tumor set to ROMA data from a human lung cancer cell line H460 (David Mu, unpublished data). Similar to the mouse LKR10 and 13 cells lines, H460 contains multiple focal copy number alterations. Syntenic regions containing changes were compared. Intriguingly, the orthologs of CUB and sushi multiple domains 1 (Csmdl) gene was reduced in copy number in mouse LKR10/LKR13 cell lines (Figure 5B) and human H460 cell line (Figure 5B). As one of the biggest genes, Figure 5: Csmdl deletion in mouse and human lun2 cancer cell lines A) ROMA moving median plot of chromosome 8 in mouse lung cancer cell line LKR10O. A ~2Mb focal deletion was seen on chromosome 8: 15321241-17425932 (UCSC mouse May 04 assembly). Csmdl is the only known RefSeq gene present. B) ROMA moving median plot of chromosome 8 of human lung cancer cell line H460. A focal lesion was seen in chr8: 4743548-5613707 (UCSC human April 03 assembly), within an 8 p region that is conserved in synteny with mouse chromosome 8. Csmdl is also the only known RefSeq gene present. Mouse LKR10 lung cancer cell line -Chromosome 8 copy number plot SI I 0 20000000 40000000 60000000 80000000 10000000012000000014000000 GP-start - Chromosome 8 Human H460 lung cancer cell line -chromosome 8 copy number plot 10.0 9 8 7 6 5 4 3 2 1.0 9 8 7 6 5 0 55000000.00 30000000.00 8( Chr 105000000.00 )000000.00 omosome 8 155000000.00 130000000.00 A 10.0 9 8 7 CSMD1 MYC 1 N- * 4 CSMD1 5000000.0' 41, - Csmdl covers 1.64Mb in the mouse genome and 2.06Mb in human. Loss of human chromosome 8p, where Csmdl resides, is a common event observed in -15% of lung cancers and lung cancer cell lines (David Mu, unpublished data). The identity of the critical tumor suppressor gene(s) in the region is still unclear. Csmdl is a potential candidate, which has been found deleted or inactivated in cancer or tumor cell lines (Scholnick and Richter 2003). We attempted to compare Csmdl expression level in normal lung vs. lung tumors in datasets generated using gene expression microarrays (Alice Shaw, unpublished). However, the absence of transcript signal in normal lung made comparison difficult. Discussion In this study, we tested the application of the mouse ROMA platform to characterize genetic alterations in retinoblastomas and lung adenocarcinomas in different genetically engineered mouse models. We were able to detect various forms of copy number changes including amplifications, deletions, and chromosomal gains and losses. We can hereby compare our results to some of the common genetic changes in human cancers. Future analysis using larger number of tumors via this strategy can thus be very informative. On the technical capabilities of mouse ROMA platform Array CGH analysis compares relative DNA sequence copy number between genomes. The value of a platform depends largely on its spatial resolution. The detectable changes in this study range from 95kb to whole chromosomal gains and losses. Among the sub- Mb lesions detected and subsequently verified was an amplification of the N-Myc oncogene in metastatic retinoblastomas, proving the usefulness of the mouse ROMA platform. Several different kinds of array CGH platforms are available for genome-wide studies in both mouse and humans. Arrays using BACs as probe elements are highly sensitive but spatial resolution is limited by the size of BACs, which range from 150kb to 200kb (Pinkel and Albertson 2005). Short oligonucleotides can greatly reduce the limit, despite its lower sensitivity needs to be compensated by the use of a higher amount of DNA in hybridization and the averaging of signals from 3-5 adjacent probes to make a reliable call (Pinkel and Albertson 2005). ROMA starts with a digestion-amplification protocol from as little as 50ng of DNA to make representations, which reduce hybridization noise by lowering genomic complexity (Lucito et al. 2003). Thus, while ROMA requires little starting materials as needed for BAC arrays, it also takes advantage of a sub-Mb resolution with an oligonucleotide array platform. This ability of the mouse ROMA was demonstrated by the detection of N-Myc amplification in retinoblastomas. The availability of both human and mouse ROMAs has opened a new avenue for comparative genomic analyses. In many human cancers, large-scale chromosomal abnormalities are common but delimitation of critical genetic regions is often difficult. There are 300+ syntenic segments covering over 90% of the mouse and human genomes. These conserved regions have been evolutionarily rearranged within and between chromosomes. The shuffling has made the mouse genome useful in assessing the relative importance of various syntenic segments that correspond to a large human chromosomal area relevant in cancer. Indeed, CGH screens on mouse cancers have been performed to delimit regions of genetic importance in the human disease (Hodgson et al. 2001; Hackett et al. 2003). Continual development of mouse CGH platforms like ROMA in parallel to human ones would enable more comparative studies to be done. On the genetics of retinoblastomas The Rb/p130 DKO mouse model developed tumors that histologically resemble human retinoblastoma (MacPherson et al. submitted). ROMA was used to assess if the similarity is also present at the genetic level. CGH and cytogenetic studies have suggested the majority of human retinoblastomas contained chromosomal imbalance. Chromosomes 1 q and 6p gains are the most frequent, found in over 50% of all human retinoblastomas (Mairal et al. 2000; Chen et al. 2001; Lillington et al. 2003; Zielinski et al. 2005). In the ROMA analysis of eight mouse retinoblastomas from Rb/pl30 DKO mice, gain of chromosomes 1 and 12 were seen in half of the samples. Interestingly, mouse chromosome 1 has three syntenic blocks on human chromosome 1q: 1q23.2-32, 1q32.1, and 1q32.2-42.1. Extra copies of the same orthologous gene(s) in one or more of these regions might be selected for in both mouse and human retinoblastomas. Some studies suggested the minimal region of gain in human to be 1q31. Further experiments are needed to determine if this is the case in our mouse model. Mouse chromosome 12, also apparently gained in half of our tumor DNA samples, has a region syntenic to human chromosome 2p. Furthermore, within chromosome 12, a minimally overlapping region of 136kb was seen amplified with even higher magnitude. The only known gene residing in this region is N-Myc. A common childhood nervous system tumor, neuroblastoma, frequently has N-Myc amplification, which marks rapid tumor progression (Brodeur et al. 1984). N-Myc overexpression has also been implicated in other neuronal cancers including human retinoblastomas (Mairal et al. 2000). While the amplification is often associated with gains of other genes on human chromosome 2p (Mairal et al. 2000; Lillington et al. 2003; Zielinski et al. 2005), N-Myc appears to be the critical gene in our mouse model. Our sample set contains only 8 late-stage metastatic tumors, making an extended study necessary to establish the timing of the N-Myc amplification. On the genetics of lung adenocarcinomas Our lab has described mouse models of lung cancer based on an expression of an activated KrasG 12D allele from its endogenous locus (Jackson et al. 2005). Microarray- based gene expression analyses have been performed to assess the molecular similarity between human lung cancer and tumors from a Kras-initiated mouse model (Sweet- Cordero et al. 2005). At the genetic level, one CGH study was previously done with a 2K BAC array platform on a set of Kras-induced lung tumors (Sweet-Cordero et al. 2006). Among the 59 tumors analyzed in that experiment, recurrent whole-chromosomal changes were detected but no focal copy number gains or losses could be seen. The current study described in this chapter differed from the prior one in three major ways: (1) a new technical platform was employed. With 84K arrayed oligonucleotides and a protocol to enhance signal to noise ratios using genomic representations, ROMA has the potential to provide an enhanced resolution to reveal focal changes that might be present. (2) This study has encompassed tumors from mice with double Kras and p53 mutations in the sample set, in order to assess if germline p53 status affects level or spectrum of genetic alterations. (3) The current study analyzed tumors from inbred mice with a pure 129S4/svJae background, instead of mice from a C57BL6J x 129S4/SvJae Fl cross. It has been suggested that genetic background can affect types of lesions in tumors. Of particular note is that mitotic recombination can be suppressed in Fl hybrids from two different parental strains. Here, ROMA was used to analyze 28 Kras-induced mouse lung tumors and 2 tumor- derived cell lines. In summary, we observed recurrent gains of chromosomes 3, 6, 12, 19 and losses of 9 and 11 in tumors that were graded 2 or higher using criteria set by Jackson et al. 2005. Focal subchromosomal copy number changes were detected in one tumor and the two LKR cell lines. p53 and genetic changes in Kras-initiated lung cancer model Despite a p53 germline lesion leads to more total tumors and more histologically advanced tumors in the Kras-initiated lung cancer model, there was no striking difference in mutation spectra of histologically comparable tumors from these mice. p53 can play a role in inducing cell cycle arrest, senescence, and apoptosis in face of DNA damage. Its loss can promote chromosomal instability and tumor progression in various mouse models (e.g.(Hingorani et al. 2005)). In our study, a p53 mutation appears to mainly act by providing a more permissive environment for the outgrowth of Kras mutant cells, instead of altering the kinds of cooperative genetic elements Kras needs to drive tumor progression. Whole chromosomal copy changes Changes in whole chromosome copy number constitute the overwhelming majority of changes seen in the primary lung tumor samples. This suggests non-disjunction was a major driver to create secondary lesions for Kras-initiated lung tumors to progress beyond grade 2 in histology. The presence of chromosomal copy number changes in almost all grade 2+ tumors has implied a selection of other genetic changes in the tumor initiating Kras mutant cells. The recurrent whole chromosomal copy number changes suggested gain or loss of one or more genes on these chromosomes might be important in this model of lung cancer. Chromosome 6 gain was seen in 80% of tumors with alterations found by ROMA. Mouse chromosome 6 harbors the Kras gene, pulmonary adenoma susceptibility 1 (Pasl) locus, and contains a region in synteny to a human 3q segment, all of which have been implicated in lung cancer. It will be of particular interest to test whether the chromosome 6 gain consistently corresponds to a copy increase of the mutant Kras allele. The Kras gene resides at the distal arm of chromosome 6. In vitro transformation of rodent cells can be promoted by amplification of the mutant ras gene (Sorrentino et al. 1988). One particularly intriguing experiment involved a study of Rat-1 cells engineered to express an H-ras activating mutation from its endogenous locus. Cells heterozygous for the mutation underwent spontaneous transformation at a low frequency, and most transformed cells had the mutant allele amplified (Finney and Bishop 1993). Finally, Kras amplification in human lung carcinoma has also been observed (Pulciani et al. 1985). On the other hand, other loci on chromosome 6 might have been selected in our Kras-initiated tumors. For instance, Pasl is a quantitative trait locus that affects lung cancer predisposition in mice. While Kras is the primary candidate, polymophisms in other genes such as Las] and Lrmp have also been associated with lung cancer susceptibility(Manenti et al. 2004). In addition, human chromosome 3q has been reported to gain in copy in various subtypes of lung cancer including adenocarcinomas (Testa et al. 1994; Pei et al. 2001; Balsara and Testa 2002; Garnis et al. 2006). Orthologous genes within the syntenic regions might be important for tumorigenesis in both species. Chromosome 12 gain was the second most frequent alteration, observed in 50% of all analyzed samples containing any changes. Interestingly, both mouse chromosomes 12 and 6 have regions syntenic to human chromosome 7. Furthermore, the conserved areas corresponding to the two mouse chromosomes tend to cluster adjacent to each other in two major areas on chromosome 7. Polysomy of human chromosome 7, as well as regional amplifications in both 7p and 7q arms have been seen in cytogenetic and array CGH studies on lung tumors and tumor cell lines (Balsara and Testa 2002; Wong et al. 2003; Kim et al. 2005; Garnis et al. 2006). On 7p, one report has described 7p22.1-22.3 and 7pl 1.2-15.3 gains in over 80% of samples (Garnis et al. 2006). Coincidently, one cluster with syntenic conservation to chromosome 6 and 12 happens to be within 7pl4- 22, which contains a many known genes including the developmentally important HoxA gene cluster and beta-integrin 8. On the other arm 7q, areas with synteny include 7q22- 36.1, which harbors the T-cell recepter-beta gene cluster among others. Among the chromosomal losses, chromosomes 9 and 11 were each reduced in copy in two tumors. A distal part of chromosome 9 is syntenic to human chromosome 3p21-22. Loss of human chromosome 3p is the most common event observed in lung cancer (Zabarovsky et al. 2002). In particular 3p21 loss is observed as one of the earliest event, which can be detected in the pre-malignant epithelium of smokers. As for chromosome 11, its distal arm has syntenic conservation to the entire human chromosome 17, where p53 resides. Sub-chromosomal copy changes Despite being the minority, one tumor and the two LKR cell lines exhibited multiple focal copy number changes. Some of the detected lesions were below 1 Mb in sizes. The two LKR cell lines shared almost identical lesions, suggesting they have the same clonal origin. A cell line represents a subclone within a tumor mass that got selected to expand in tissue culture. Different sets of genetic criteria were likely required for during cell line establishment vs. clonal outgrowth in vivo, which may explain the difference in kinds of lesions observed. In addition, according to the clonal evolution model (Nowell 1976), each tumor mass is likely a composite of heterogeneous subclones, which have individual proliferation rates and fates under selection. The absence of focal ROMA signals in the primary tumors does not directly imply the absence of focal lesions within its subclones. Instead, alterations may mask each other inside a heterogeneous tumor. Data from primary tumors are essentially an average of all the differences. In addition, ROMA lacks the ability to detect general polyploidy or non-reciprocal translocations, both of which will appear as constant total tumor to normal DNA ratios in all types of CGH studies. Detectable focal changes can allow delineation of critical genetic elements. When we compared ROMA datasets from the mouse LKR lung cancer cell lines to a human lung cancer cell line, focal deletions involving the Csmdl mouse and human orthologs were observed. While the physical loss of this locus still remains to be validated, ROMA seemingly has the resolving power to reveal common orthologous changes down to a single-gene level in a cross-species study. Future comparative genomic studies using ROMA or other array CGH platforms will likely provide more molecular insights on tumorigenesis. Materials and Methods Lung tumor DNA isolation All mouse protocols were approved by the animal care committees at the Massachusetts Institute of Technology. KRas L SLG 1 2D , KRaS LSL G12D ; p53+/- and KRaSLSLG12D; p53 R270H/+ mice on a 129S4/SvJae background was infected with adenovirus Cre as described in Jackson et al. 2005. Lung tumors were dissected from the lungs of mice 22-24 weeks after infection. One portion of each tumor was fixed in formalin, sectioned in paraffin, and stained in hematoxylin and eosin. Histological grading of each tumor was assigned based on a 1-5 scale as described in Jackson et al.. Remaining tumor material was stored at -80 0 C prior to DNA isolation. Tail from each mouse was collected for use as normal control. DNA was extracted from thawed tissues using reagents and protocols in Puregene DNA isolation kit (Gentra Systems, Inc.). LKR10 and LKR13 mouse lung cancer cell lines were derived from K-Ras LAl mouse on a 129S4/SvJae background. Retinoblastoma tumor generation As described in (MacPherson et al. submitted)), Pax6 a-enhancer Cre mice were bred with Rb?ox/?ox; p130-/- animals. Mice were maintained on a mixed C57BL/6; 129SvJ; FvB/n background. Late stage metatastic tumors were collected from mice at time of sacrifice, 183?30 days of age. Samples were frozen at -80 0 C until DNA isolation. ROMA analysis Genomic DNA from tumors and tails of corresponding mice were paired for each experiment. DNA was digested with BglII to obtain low complexity representations (LCRs) of the genomes. LCRs from tumor and normal tissues were differentially labeled with Cy 5 and Cy3 respectively by random priming. The hybridizations to oligonucleotide microarrays were performed as described in (Lucito et al. 2003).The design of the mouse ROMA arrays is described elsewhere (Lucito et al., in preparation). Array images were acquired with an Axon GenePix 4000B scanner. The raw array data were globally normalized. A moving window of a 5 data-points was used to smoothen the raw data by assigning the median value of the moving window to each central data-point. Southern analysis Genomic DNA was digested with EcoRI. N-Myc probe was a 1.1kB cDNA fragment including sequence from exons 2 and 3 that was obtained by PCR of a mouse embryo brain cDNA library using the following primers: 5'gaggacagcgcagataaagg and 5' cctcactcctaatccggtc. Rosa26 probe hyrbidization to chromosome 6 was used as control and performed as described in (Soriano 1999). Southern blotting was performed using standard protocols. Blot was first probed with the N-Myc probe, stripped and rehybridized to the Rosa26 probe to control for loading discrepancies. Band intensities on phosphoimager scans were quantified using ImageQuant. Acknowledfements The authors would like to thank Denise Crowley and Mike Brown for histology preparation. We also thank Rob Bronson for pathology consultation. D. Mu is supported by a grant from Joan's Legacy Foundation. T.J. is a Howard Hughes Investigator. References Balsara, B.R. and Testa, J.R. 2002. Chromosomal imbalances in human lung cancer. Oncogene 21(45): 6877-6883. Barrett, M.T., Scheffer, A., Ben-Dor, A., Sampas, N., Lipson, D., Kincaid, R., Tsang, P., Curry, B., Baird, K., Meltzer, P.S., Yakhini, Z., Bruhn, L., and Laderman, S. 2004. Comparative genomic hybridization using oligonucleotide microarrays and total genomic DNA. Proc Natl Acad Sci U SA 101(51): 17765-17770. Brodeur, G.M., Seeger, R.C., Schwab, M., Varmus, H.E., and Bishop, J.M. 1984. Amplification of N-myc in untreated human neuroblastomas correlates with advanced disease stage. Science 224(4653): 1121-1124. Chen, D., Gallie, B.L., and Squire, J.A. 2001. Minimal regions of chromosomal imbalance in retinoblastoma detected by comparative genomic hybridization. Cancer Genet Cytogenet 129(1): 57-63. Dhami, P., Coffey, A.J., Abbs, S., Vermeesch, J.R., Dumanski, J.P., Woodward, K.J., Andrews, R.M., Langford, C., and Vetrie, D. 2005. Exon array CGH: detection of copy-number changes at the resolution of individual exons in the human genome. Am JHum Genet 76(5): 750-762. Finney, R.E. and Bishop, J.M. 1993. Predisposition to neoplastic transformation caused by gene replacement of H-rasl. Science 260(5113): 1524-1527. Garnis, C., Lockwood, W.W., Vucic, E., Ge, Y., Girard, L., Minna, J.D., Gazdar, A.F., Lam, S., Macaulay, C., and Lam, W.L. 2006. High resolution analysis of non- small cell lung cancer cell lines by whole genome tiling path array CGH. Int J Cancer 118(6): 1556-1564. Hackett, C.S., Hodgson, J.G., Law, M.E., Fridlyand, J., Osoegawa, K., de Jong, P.J., Nowak, N.J., Pinkel, D., Albertson, D.G., Jain, A., Jenkins, R., Gray, J.W., and Weiss, W.A. 2003. Genome-wide array CGH analysis of murine neuroblastoma reveals distinct genomic aberrations which parallel those in human tumors. Cancer Res 63(17): 5266-5273. Hingorani, S.R., Wang, L., Multani, A.S., Combs, C., Deramaudt, T.B., Hruban, R.H., Rustgi, A.K., Chang, S., and Tuveson, D.A. 2005. Trp53R172H and KrasG12D cooperate to promote chromosomal instability and widely metastatic pancreatic ductal adenocarcinoma in mice. Cancer Cell 7(5): 469-483. Hodgson, G., Hager, J.H., Volik, S., Hariono, S., Wernick, M., Moore, D., Nowak, N., Albertson, D.G., Pinkel, D., Collins, C., Hanahan, D., and Gray, J.W. 2001. Genome scanning with array CGH delineates regional alterations in mouse islet carcinomas. Nat Genet 29(4): 459-464. Jackson, E.L., Olive, K.P., Tuveson, D.A., Bronson, R., Crowley, D., Brown, M., and Jacks, T. 2005. The differential effects of mutant p53 alleles on advanced murine lung cancer. Cancer Res 65(22): 10280-10288. Jobanputra, V., Sebat, J., Troge, J., Chung, W., Anyane-Yeboa, K., Wigler, M., and Warburton, D. 2005. Application of ROMA (representational oligonucleotide microarray analysis) to patients with cytogenetic rearrangements. Genet Med 7(2): 111-118. Johnson, L., Mercer, K., Greenbaum, D., Bronson, R.T., Crowley, D., Tuveson, D.A., and Jacks, T. 2001. Somatic activation of the K-ras oncogene causes early onset lung cancer in mice. Nature 410(6832): 1111-1116. Kallioniemi, A., Kallioniemi, O.P., Sudar, D., Rutovitz, D., Gray, J.W., Waldman, F., and Pinkel, D. 1992. Comparative genomic hybridization for molecular cytogenetic analysis of solid tumors. Science 258(5083): 818-821. Kim, T.M., Yim, S.H., Lee, J.S., Kwon, M.S., Ryu, J.W., Kang, H.M., Fiegler, H., Carter, N.P., and Chung, Y.J. 2005. Genome-wide screening of genomic alterations and their clinicopathologic implications in non-small cell lung cancers. Clin Cancer Res 11(23): 8235-8242. 100 Lillington, D.M., Kingston, J.E., Coen, P.G., Price, E., Hungerford, J., Domizio, P., Young, B.D., and Onadim, Z. 2003. Comparative genomic hybridization of 49 primary retinoblastoma tumors identifies chromosomal regions associated with histopathology, progression, and patient outcome. Genes Chromosomes Cancer 36(2): 121-128. Lucito, R., Healy, J., Alexander, J., Reiner, A., Esposito, D., Chi, M., Rodgers, L., Brady, A., Sebat, J., Troge, J., West, J.A., Rostan, S., Nguyen, K.C., Powers, S., Ye, K.Q., Olshen, A., Venkatraman, E., Norton, L., and Wigler, M. 2003. Representational oligonucleotide microarray analysis: a high-resolution method to detect genome copy number variation. Genome Res 13(10): 2291-2305. MacPherson, D., Conkrite, K., Tam, M., Mukai, S., Mu, D., and Jacks, T. submitted. Murine bilateral retinoblastoma exhibiting rapid onset, metastatic progression and N-myc gene amplification. Mairal, A., Pinglier, E., Gilbert, E., Peter, M., Validire, P., Desjardins, L., Doz, F., Aurias, A., and Couturier, J. 2000. Detection of chromosome imbalances in retinoblastoma by parallel karyotype and CGH analyses. Genes Chromosomes Cancer 28(4): 370-379. Manenti, G., Galbiati, F., Gianni-Barrera, R., Pettinicchio, A., Acevedo, A., and Dragani, T.A. 2004. Haplotype sharing suggests that a genomic segment containing six genes accounts for the pulmonary adenoma susceptibility 1 (Pas ) locus activity in mice. Oncogene 23(25): 4495-4504. Nowell, P.C. 1976. The clonal evolution of tumor cell populations. Science 194(4260): 23-28. Pei, J., Balsara, B.R., Li, W., Litwin, S., Gabrielson, E., Feder, M., Jen, J., and Testa, J.R. 2001. Genomic imbalances in human lung adenocarcinomas and squamous cell carcinomas. Genes Chromosomes Cancer 31(3): 282-287. Pinkel, D. and Albertson, D.G. 2005. Array comparative genomic hybridization and its applications in cancer. Nat Genet 37 Suppl: S 11-17. Pinkel, D., Segraves, R., Sudar, D., Clark, S., Poole, I., Kowbel, D., Collins, C., Kuo, W.L., Chen, C., Zhai, Y., Dairkee, S.H., Ljung, B.M., Gray, J.W., and Albertson, D.G. 1998. High resolution analysis of DNA copy number variation using comparative genomic hybridization to microarrays. Nat Genet 20(2): 207-211. 101 Pollack, J.R., Perou, C.M., Alizadeh, A.A., Eisen, M.B., Pergamenschikov, A., Williams, C.F., Jeffrey, S.S., Botstein, D., and Brown, P.O. 1999. Genome-wide analysis of DNA copy-number changes using cDNA microarrays. Nat Genet 23(1): 41-46. Pulciani, S., Santos, E., Long, L.K., Sorrentino, V., and Barbacid, M. 1985. ras gene Amplification and malignant transformation. Mol Cell Biol 5(10): 2836-2841. Scholnick, S.B. and Richter, T.M. 2003. The role of CSMD1 in head and neck carcinogenesis. Genes Chromosomes Cancer 38(3): 281-283. Solinas-Toldo, S., Lampel, S., Stilgenbauer, S., Nickolenko, J., Benner, A., Dohner, H., Cremer, T., and Lichter, P. 1997. Matrix-based comparative genomic hybridization: biochips to screen for genomic imbalances. Genes Chromosomes Cancer 20(4): 399-407. Soriano, P. 1999. Generalized lacZ expression with the ROSA26 Cre reporter strain. Nat Genet 21(1): 70-71. Sorrentino, V., McKinney, M.D., Drozdoff, V., Hume, C.R., and Fleissner, E. 1988. Spontaneous or carcinogen-mediated amplification of a mutated ras gene promotes neoplastic transformation. Oncogene Res 2(2): 189-195. Sweet-Cordero, A., Mukherjee, S., Subramanian, A., You, H., Roix, J.J., Ladd-Acosta, C., Mesirov, J., Golub, T.R., and Jacks, T. 2005. An oncogenic KRAS2 expression signature identified by cross-species gene-expression analysis. Nat Genet 37(1): 48-55. Sweet-Cordero, A., Tseng, G.C., You, H., Douglass, M., Huey, B., Albertson, D., and Jacks, T. 2006. Comparison of gene expression and DNA copy number changes in a murine model of lung cancer. Genes Chromosomes Cancer 45(4): 338-348. Testa, J.R., Siegfried, J.M., Liu, Z., Hunt, J.D., Feder, M.M., Litwin, S., Zhou, J.Y., Taguchi, T., and Keller, S.M. 1994. Cytogenetic analysis of 63 non-small cell lung carcinomas: recurrent chromosome alterations amid frequent and widespread genomic upheaval. Genes Chromosomes Cancer 11(3): 178-194. Wong, M.P., Fung, L.F., Wang, E., Chow, W.S., Chiu, S.W., Lam, W.K., Ho, K.K., Ma, E.S., Wan, T.S., and Chung, L.P. 2003. Chromosomal aberrations of primary lung adenocarcinomas in nonsmokers. Cancer 97(5): 1263-1270. 102 Zabarovsky, E.R., Lerman, M.I., and Minna, J.D. 2002. Tumor suppressor genes on chromosome 3p involved in the pathogenesis of lung and other cancers. Oncogene 21(45): 6915-6935. Zielinski, B., Gratias, S., Toedt, G., Mendrzyk, F., Stange, D.E., Radlwimmer, B., Lohmann, D.R., and Lichter, P. 2005. Detection of chromosomal imbalances in retinoblastoma by matrix-based comparative genomic hybridization. Genes Chromosomes Cancer 43(3): 294-301. 103 Chapter 3 A multiplexed SNaPshot" protocol for genome-wide mouse SNP genotyping and its application to detect loss-of-heterozygosity in murine lung tumors Mandy Tam', Alejandro Sweet-Cordero', Brian Weiss 2 , David Ginzinger 3 , Julie Weng 3 , Annie Fung 3 , and Tyler Jacks",4 'Massachusetts Institute of Technology, Department of Biology and Center for Cancer Research, 77 Massachusetts Avenue, Cambridge, MA 02139 USA 2 Cincinnati Children's Hospital Medical Center, 3333 Burnet Avenue, Cincinnati, OH 45229, USA 3 University of California at San Francisco, Comprehensive Cancer Center, San Francisco, CA94143, USA 4 Howard Hughes Medical Institute, 4000 James Bridge Road, Chevy Chase, MD 20185, USA The author, A.S.C. and B.W shared the initial work to validate individual SNPs. The author designed the multiplexing protocol, obtained lung tumor DNA, performed all PCR and genotyping data anlaysis. SNaPshot'" reactions were performed by J.W. and A.F.. 104 Abstract While the choice of human SNP genotyping methodologies is broad, few have been applied for whole-genome analysis in the mouse. In this chapter, we presented a new and validated protocol. We identified 358 published SNPs individually for polymorphism in 129S4/svJae vs. C57BL/6J strains and worked out a protocol to genotype a panel of 147 markers in the mouse using the SNaPshot TM (Applied Biosystems) genotyping system. Our method uses a standard DNA sequencing machine to resolve single-base extension (SBE) products of genotyped SNPs, enabling easy adaptation by most standard laboratories. We tested the use of the assay as a mapping tool of LOH in mouse tumors. We analyzed 20 lung tumor DNA samples from a Kras-driven mouse model of lung cancer and detected several LOH markers including loss of wild-type p53. 105 Introduction Genetic alterations are the underlying cause of cancer. Cancer-causing genes can be generally dividing into two categories: 'proto-oncogenes', which when activated by mutation or overexpression can promote abnormal proliferation, growth, and differential, and 'tumor suppressor genes', which when inactivated lose their normal regulatory roles in these processes. In the classical 'two-hit' hypothesis, loss of heterozygosity (LOH) is believed to promote cancer by inactivating tumor suppressor gene function (Knudson 1971). Genome-wide LOH screens by allelotyping are useful tools to identify loci that may harbor tumor suppressors genes. (Sokolov 1990) Mouse tumor models are valuable experimental tools to study human cancer genetics. In many human cancers, large-scale chromosomal abnormalities are common but finding of the critical regions is often difficult. Through evolutionary shuffling, most individual chromosomes in one mammalian species can be delimited into regions of synteny that are conserved in other species on separate chromosomes. This is true between human and mouse. Given such syntenic structure and that >90% of the human genome is covered by conserved areas in the mouse (Waterston et al. 2002), LOH studies of mouse cancer may provide an invaluable perspective to delineate large human genetic lesions. Genome-wide LOH screens have been performed on several mouse models of human cancer (Dietrich et al. 1994; Radany et al. 1997; Herzog et al. 2002; Wu et al. 2002; Benavides et al. 2003). LOH screens are traditionally performed with simple sequence length polymorphism (SSLP) markers. Genotyping of SSLP using PCR can be performed 106 by most laboratories without extensive capital investment but is cumbersome. Single nucleotide polymorphism (SNP) is becoming the marker of choice for various purposes of genome-wide screening, due to its dense genome coverage and the availability of several high-throughput genotyping methods. In the mouse, genome-wide SNP screens have been performed to study haplotype structure (Wade et al. 2002; Wiltshire et al. 2003; Frazer et al. 2004; Zhang et al. 2005), delineate strain relationships (Petkov et al. 2004b; Pletcher et al. 2004), and facilitate genetic mapping (Grupe et al. 2001; Pletcher et al. 2004; Owens et al. 2005; Moran et al. 2006). In the screening of LOH regions in mouse cancer, however, no report using SNPs as markers has been made to date. Various methods of SNP genotyping have been recently developed (Syvanen 2001). However, the cost of specialized instruments such as mass spectrometry machines may set a barrier for performing SNP mapping studies. This chapter presents a multiplexed protocol for genome-wide mouse SNP genotyping in mice using SNaPshot TM (Applied Biosystems), which couples single-base extension (SBE) reactions to capillary electrophoresis and fluorescence detection. The use of only standard PCR and sequencing machines should make the assay adaptable for use by most laboratories. We tested the protocol and the principle of using SNPs to identify LOH regions with a mouse lung cancer model. Mice expressing a KrasG 12 D allele develops tumors that histologically resemble human non-small cell lung cancer (NSCLC) (Johnson et al. 2001) and share similar gene expression profiles (Sweet-Cordero et al. 2005). We sought to characterize genetic changes associated with these tumors. We observed several cases of whole- 107 chromosomal LOH, suggesting other genetic factors can cooperate with a Kras mutation in tumorigenesis in this model. Results Validation of SNPs in 129S4/SvJae vs. C57BL/6J strains 358 SNPs distinguishing C57BL/6J and 129/Sv or 129S1/SvImJ sub-strains were selected from databases originating from two independent large-scale SNP discovery studies (Germer et al. 2000; Lindblad-Toh et al. 2000b). We designed SBE assays to individually test the genotypes of these chosen SNPs in C57BL/6J (B6) vs.129S4/SvJae (129S4) DNA. Supplemental Table 1 lists the SNP genotyping results of DNA from B6 and 129S4 mice maintained in-house and those from B6x129S4 Fl hybrids. The B6 and 129S4 DNA used for this analysis did not come from true parental mice crossed to generate the tested F1 hybrids. By our assay, a total of 172 SNPs were confirmed to be polymorphic in B6 and129S4, as defined by observed heterozygosity in F1 and homozygosity for different alleles between the parental strains. Among the unconfirmed markers, 41 were found to be non-polymorphic, 31 appeared 'heterozygous' in 129S4, 31 looked 'heterozygous' in B6 (10 of which also 'heterozygous' in 129), 67 were inconsistent among the three tested genotypes, and 25 failed genotyping. Our testing of a different 129 sub-strain vs. B6 may explain why some markers are non-polymorphic. Alternatively, the non-polymorphic markers may be false positives from the large-scale discovery screens. The high number of 'heterozygous' markers in DNA from the inbred strains was surprising. Cross hybridization of an SBE primer to another genomic position 108 may produce two allelic peaks and a 'heterozygous' score. It is also possible that despite extensive inbreeding, certain loci on the parental strains remain polymorphic, which may result in the inconsistent genotypes between tested Fl hybrids and the non-biological parental B6 and 129S4 strains inbred in-house. Nevertheless, 172 markers across the mouse genome were individually validated to be polymorphic between the two tested strains. Establishment of a protocol for genome-wide multiplex SNP typing in the mouse 147 validated SNPs were chosen for genotyping in a multiplex fashion (Figure 1). The positional distribution of SNPs in the mouse genome is illustrated in Figure 2. The mean distance between markers was 14.5Mbp, and at least 2 SNPs were present in each of 19 autosomes (Table 1). A single-base extension method was used to genotype the SNP panel. Schematic of the procedure is shown in Figure 1. The protocol employed 93 PCRs to amplify the 147 SNP targets (Supplemental Table 2). Pooling of PCR products reduced the number of subsequent SBE reactions to 15. The amounts of individual PCR products in each pool were adjusted according to the strength of the PCRs. In each SBE reaction, 5-11 SNPs were genotyped with primers of varied lengths, resolvable by capillary electrophoresis (Table 2). SBE primer concentration was adjusted to correct for differential signal strength. A representative SBE reaction output is shown in Figure 3. The relative height of the two allelic peaks from each marker was quantified on the electrograph. 109 Figure 1: Schematic for SNP eenotvying A multiplex strategy was used to genotype 147 SNPs in the mouse genome. (A) 93 PCRs were used to amplify genomic DNA containing 1-2 SNP targets in each reaction. Each SNP is depicted as [allele 1/allele 2]. (B) PCR products were pooled as templates for each single-base extension (SBE) reaction, in which fluorescently labeled dideoxynucleotide triphosphates were incorporated into primers of different lengths. 15 SBE reactions each genotyping 5-11 SNPs were performed. (C) SBE products were resolved on an ABI 3700 capillary DNA sequencer. 110 - -(Gill- -(Cm-- 000000000000 oo 00000 o 0 ( 00 VVT T T -- [C/T]-,- [G/A]-- - [cC/-T] rn [G/A]J *0 *0 L- -J ddC I [G/C]- ddT -[A/T] ddA ddG -[G/C]? -* -* minutes --- ddC -- ddT ddG ddA ddC ddT ddG ddA --- [C/T]- 147 SNPs 93 PCRs B 0 0 Figure 2: Positional representation of 147 screening markers in mouse genome Positions of markers on chromosomes are based on UCSC mouse Feb 2006 (mm8) assembly 112 ~!**. l w & A & * ,. 4, . ,L 4. 1 p.'A, ,,* 2, *" + *" . , . rr, , A. d &. A , * V p.. . . . . . + p~A * . - IV . . p . . v ,i ", ", '" " V v v" "" v " ci* * ,. , , ,. , , . C4 *o V- p-0 o 0 0 00 0 0 0 0 0 0 d qt D QcO 0S CNO CO 0 (dq Wy) uO!1!sOd oP!IoGfloN I A? ALL Table 1: Chromosomal distribution of SNP markers in the 2enome-wide screen Chromosome 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 Whole Genome Marker count 11 13 10 10 10 5 9 9 4 7 10 6 7 12 7 6 5 2 4 Minimum distance (Mb) 0.50 0.00 0.47 0.56 0.33 14.26 0.26 0.04 0.29 3.65 0.06 4.37 12.61 0.54 8.73 2.26 8.54 29.59 9.77 Maximum distance (Mb) 30.80 37.85 38.10 31.77 26.04 82.94 34.48 32.49 72.25 24.98 24.41 32.01 21.65 20.18 13.39 40.36 21.19 29.59 20.52 147 Mean distance (Mb) 16.18 13.10 14.69 14.69 12.11 32.38 15.13 12.97 28.09 14.78 10.02 13.40 16.23 9.57 11.50 16.76 13.84 29.59 14.74 14.54 114 Table 2: List of multiplexed SBE assays for whole-genome SNP genotypin2 SBE primer sequence and genotype of each SNP in 129xB6 Fl hybrid are shown. SBE primers were designed and organized such that those typing alternative nucleotide pairs differ in at least 2 bases of length. 115 Multiplex Primer SBE group Assay ID RefSNP ID SBE primer sequence Length SNP 412 rs3089257 GCCATTCGTCCCAGGG 16 CT 439 rs3023183 TCTCTGTGCCCACAGCCA 18 GA 471 rs3022975 GCTCTGGGAATGTGCT-TTC 20 CT 883 rs3668662 AAAGGCAGTGGGTACACATCAT 22 GA 772 rs3704164 GCTGCCATATGAAGATCTCCTCTA 24 CT 1A 405 rs3022887 TTTTGGGTGTTTCTATGATAACGCTC 26 GA 779 rs3660209 GTCAATAGGTGAGAAAAATATCAGACTG 28 CT 786 rs3720966 I iiIIIITTCACCCACGTTGTGACTTAAGC 30 GA 830 rs3714631 I I I I I I I I I I I I GAGGTGTGTTGGTGTGCACA 32 CT 850 rs3090435 I I I I I I I I I I GGTACTATCCCTGGCTTTTCACA 34 GA 426 rs3090833 I I I I I I I I I GAAAAAATTGCTrGTACTGTAGTCT 36 CT 109 rs3023468 GCACATGGTTCTGCCACA 18 GC 452 rs3023449 TGACACAACGCTGTTCCAGG 20 TA 868 rs3710059 1TTGGAGCTGGCACATCCACT 22 GC 455 rs3023436 TITTCCAGGGAGTAAAACATCAGG 24 TA 1B 879 rs3695889 "TTITCAAGTTAGAAGCATTGCCCTC 26 GC 823 rs3713224 TTTTACCAGTITCATTTCCCATACTTCT 28 AT 490 rs3090645 ITTTIITGGCTACTACACTAGCAAATCCATAG 32 TA 922 unmapped I IllI I IIAGAGCCAGACATAGTAGCACACGGA 36 AT 893 rs3688361 IITTIITTCCTTAGCAGTTIAGGAATATTTAGATAGTTAA 40 AT 808 rs3669262 TTATCAGCACCCGTCCCA 18 GT 415 rs3022979 TGAGGGCATTGGTTTCTRTC 20 CA 821 rs3704980 GACTAGTCCACATTGGTGAGCA 22 GT 461 rs3023258 CCTCCCATTA 1TITTTTTCTGAG 24 CA 422 rs3023045 TTAACTTGGAATATCAGGCTTTCTT 26 GT 79 rs3089102 TTTTTTGAGTGTCACATGGATTTGCAC 28 CA 600 rs3022802 TTTTCAAAGGGTATGAGAATATGGACTGG 30 GT 840 rs3694308 TTTTT"TCAAGTATATGGACTTGGAAGACAATG 32 CA 453 rs3023456 11 I I I IATTGGCCTTCTTATGTCGACTiTAT 34 GT 895 rs3662097 I I I I I I Il I 1 IA'I -TGGACTCTATATTrTGTCTGGGCT 38 GT 761 rs3665023 TGGCTGAGGGACTTGTGC 18 CT 478 rs3022989 ACAAAACTGCCAGTTGCTTC 20 GA 33 rs3023057 TrITATCGAGGTGCCTT1TGTC 22 CT 489 rs3090908 CAGAAACATAATTTCAAAGTGCA 24 GA 1D 757 rs3682376 CCTTTCTCCTATCTTTACTrCCTAGC 26 CT 803 rs3683689 ITITrTCACTGTGGTCAGACAGAATGCA 28 GA 713 rs3684370 TI I TI II CAGCAAGGAGTGTGTATGCA 30 CT 81 rs3023379 Tr-ATATTTATCTCACTGTGAAGTCTGCCTA 32 GA 831 rs3676476 1I l I I I IIICACTCAGTAGGTCAGAGCAGGG 34 CT 892 rs3717068 I I I I I I I I I I I IAGTCTACTTTCAGTGCTGTCCCAT 36 GA 790 rs3685393 CAGTTCCAAGCACCCACA 18 CT 762 rs3679837 ACTGCCAGTTCATGACCTCC 20 GA 741 rs3696551 CTGTCCCCGTAGACTAGACCTT 22 CT 770 rs4137954 GCTCTGGTAAGTIAACACACTCC 24 GA 756 rs3688884 TTTIT1TGATGGGTGGGTGTGTCAGT 26 CT 1E 700 rs3659426 TT II ITTTGGAGTCAGACAGGAATGGAA 28 GA 51 rs3023194 IIIIIIIIIII I GCCTTTAGGTTTCATGC 30 CT 913 rs3677860 I I I I I I I ICTGGCTGTTCATTATrTGTGCA 32 GA 701 rs3681957 1TTrTTTCTCAGGTACTGAGTGAGTTCCTAGACT 34 CT 911 rs3697014 I I I I I I I I I CTAAGCTTGTGTATCAACTGCCT 36 GA 812 rs3657504 I1 I II I I I iICAAGACCACATCTCCTATTCCTICT 38 CT 797 rs3696966 TAGGAGGGAACGGAGGC 17 GA 746 rs3657668 GCGGCTTCATrCTCCATCT 19 CT 882 rs3674239 TCTTCCACTCTTrGGTCCTCC 21 GA 727 rs3681675 TCAGGTAATGGAAAATCACTCAG 23 CT 1F 839 rs3667625 CAGACATTCTTACTCCATCATCTCC 25 GA 768 rs3716232 TTFT ITACACATCACAAGGCCACCTA 27 CT 809 rs3724533 1TTTITGGTAGGCAGGTACCAGAATCTCA 29 GA 769 rs3694785 TITTIrATACGACTTAGCTACAGTCCCTGG 31 CT 857 rs3705482 I-ITT-ITAATATATATCCAGTGGAATTGAGTGGT 33 GA 898 rs3667466 I I I I I I I AGGGACACACACATATT'G CT 35 CT 912 rs3702150 CAGGCGGGCTAAATTCA 17 CT 763 rs3686956 ATGTGACAGGGAAGTTGGC 19 GA 742 rs3722968 TCTCTCCTAGCTCATCCCATG 21 CT 814 rs3708958 GCTGAGTCACGGTACATAAAGTTGT 25 CT iG 833 rs3726717 1T1TCAGAAGCTCAGAAAGCATCAAG 27 GA 844 rs3691937 1TT T i TGCCTCTGTTGTGGCTAATCA 29 CT 843 rs3657720 11TTT I I ITCCTTGTAATAAGCCACAGCATCT 31 GA 827 rs3090608 II I I I I I II IICCGAGTGCTGACTCTGGGTT 33 CT 817 rs3664582 TTTTTAAGAAGTGTCCCAAATCCTTTCTATATAGT 35 GA 834 rs3689513 1II I I I I I I IGI-ICTTTGGAGACATATTGTGTGGITA 37 CT 856 rs3697769 CCTCACATGTGGACTCAGGC 20 GT 766 rs3724779 AGAGGACCTAATTGAGGACTGC 22 CA 86 rs3089070 TTTCCTAGTAT1TCTGTGGTCTT 24 GT 75 rs3023347 ACATCCATATTTACAAGGTCATAGAA 26 CA 1H 715 rs3704392 GTTAT-1ATGCTCCAACAGTTATTGAAA 28 GT 788 rs3713871 ITTTTRTTCCCTTAGCTTTCAAGTCCTTGC 30 CA 436 rs3089474 TT1GTGITTAGAATGTTGCTCTTAAGGGTTT 32 GT 835 rs3669413 ITFI1TCAAACTTGTGGTCGTGATAGATATTAG 34 CA 465 rs3088800 TAACAACTTTAGTIAACTAGAAATACTAAGTCTTGA 36 GT 446 rs3090586 TAGCGCACAGTGCCAGAA 18 CT 4 rs3022839 GAGTAACATCACAGCCTTCG 20 GA 32 rs3023037 TTTATGGTGCCAGAAAATCAAC 22 CT 488 rs3090260 TTAAAGTCTCAACTCCATCTITCC 24 GA 721 rs3664018 ITTTrTTCACAGCCTCAGAAAGTCCC 26 CT 2A 218 rs3023026 TTT-TCCACACCTCCACTATTATAAAGC 28 GA 820 rs3708255 mI I I I IIll I ICCTGTCCCTTCACCAGGG 30 CT 859 rs3658370 I I I I I I I CAATGTGGGTAAAACTGGCAAT 32 GA 96 rs3023416 ITT1 ECCCTTCAGAAATGAAAATrAATCTACTA 34 CT 80 rs3023386 TTTTTTT IAAGTTCATCATTCCCTAGGATGTTATA 36 GA 445 rs3089912 I I I I I I I I I I I I GAAAAAGGCAGTGACAAAGTATG 38 CT 818 rs3722942 GCTGCCCATTCTCACCT 17 GA 722 rs3681847 AGGAITGGATCAGCCATCA 19 CT 858 rs3674616 ATGCCCAGAGAGTGATCTAGAAG 23 CT 764 rs3711535 TTITTTTCCAT1TGTTCCAGAGGCA 25 GA 2B 339 rs3023161 CACCACTTGATATAGGAATGTACAC1T 27 CT 811 rs3023409 TTGTGATATGTGGAAGTTATATAAGCTTC 29 GA 732 rs3706063 I IIllltt I CATGTGGCTGG1TACCTCTC 31 CT 118 rs3023175 I I I I I I I ITCACAAGTCAGCATCAACGCAT 33 GA 822 rs3685188 1 I I I t Cil i C'TCTGCTGCTGTACAGCTC 35 CT 724 rs3664805 AAGATCATCAGGGGCCTG 18 CT 20 rs3022960 TTGAAACATGGAGACAAGGC 20 GA 448 rs3023243 TT'CGGCAACTGACITTGGACA 22 CT 483 rs3088822 GTACCTGTGAGTATTCAGTCAGCA 24 GA 67 rs3023265 TTCC'TA1TTTGCCAGTCTCCTTACT 26 CT 2C 800 rs3671678 1TTTT GTGGTGGGGGAGCAATATG 28 GA 321 rs3021908 IIt I I I II I IGACGGTGTGTGCCTACAAT 30 CT 860 rs3701351 TTTTTT I ITGCCAAACAA'TGTAAGAATGTGTG 32 GA 826 rs3663534 I I II IllTl CCAAACTCCAAGTTCTTCAGCC 34 CT 201 rs3023256 tI I I I II II I CTGGGAGAGGTAACTGCTAACT 36 GA 890 rs3658201 11 I I I I I I I TGGTITTGATTCTFCAGTGTAGT-IGG 38 CT 787 rs3712403 CCTTCCTGTCTGTTCCAGC 19 AT 794 rs3654982 TCTAGTCCACCAGCAGCAGAAAC 23 AT 2D 845 rs3023067 TT'TTCTTGCTGTCCTITTGAGCTGAG 27 TA 854 rs3710192 TTCTCCATGTTAAGCATTACAATTATGACTA 31 AT 851 rs3723894 TTTnTTGGCTAGITCATAGAGTATCAGAAATGTGT 35 AT 767 rs3672332 ACAGTGAGGAGGGGCCA 17 CT 435 rs3090731 CAGGTCTCCAGCTGAAAGC 19 GA 726 rs3667376 GTTFGTGACCTGGTCTCTGTG 21 CT 750 rs3709317 CTGCAATGAACATCACAGAGC 21 GA 740 rs3666032 GTGAAGGACAGACAGACAGACAG 23 GA 2E 738 rs3711350 TGACAATTTCTCACATGGTATTAGATC 27 GA 731 rs3674631 ITT1t TT CAGGAACGCAGTGATTG"TC 29 CT 862 rs3700023 TTTT TGATTGCCCTATAGCCATTACCTG 31 GA 832 rs3669022 I II II I I l CTTAGCCCTCCCCAACTTACC 33 CT 606 rs3023117 CAATAGATAT1TAGATGTTGCTATTGTrTATCTAC 35 GA 829 rs3660910 JI II IT I I I AATITG CI-IATGTGGCTGTCTATC 37 CT 909 rs3721297 AGGCAGGTCCATGCAGG 17 CA 484 rs3089436 CACATGGGGACTGTCCAAA 19 GT 841 rs3707288 TCACCTGCTCGTATTCCTGGA 21 CA 801 rs3726430 GGGTAGGGGTAGGAAGTAGAGAG 23 GT 2F 486 rs3090719 TTTTGGTCCCACCTTGTrACAGGTC 25 CA 745 rs3713298 GGTAGCTGCTAAATAATCTTCAAGAG 27 GT 733 rs3672323 TTTCTTAACTGTGAAGAACTAAACTGCAG 29 CA 101 rs3090912 T1T1FTTAAGTACTGATGG CTTGAGTCTrA 31 GT 789 rs3713838 1I II I I I I GAGGGTCAGAGCACTTGCAGTA 33 CA 900 rs3716435 CCCACGTTCCAACACACA 18 CT 730 rs3666331 CCTCAGTGAACTGCACATCC 20 GA 714 rs3662163 GCTGTATAAACTATGCCCCCAA 22 CT 37 rs3023051 ITITGAGAATGAAATGAACACCAG 24 GA 463 rs3088501 ATGCTAGTAGGAAGACTCTGGAACTA 26 CT 2G 413 rs3090381 TTAAGTACTTGGGTATGAAGTTCTCAAA 28 GA 723 rs4137557 1 I III I I I I I I IGGACCATrCTCCGTGTTCT 30 CT 816 rs3726591 11I IIilIttI iGGGAGGTCGGTATTAGGAGAC 32 GA 748 rs3685067 1 I 1I I I I I I I AGATGCCTCATCTGATAACAGG 34 CT 828 rs3706262 II I I II I I IICTCTGACTCCAGTCCCTCTGGG 36 GA 799 rs3719410 il l I I IIITICAATGAGCTTGAATTCTGCTAATAA 38 CT Implementation of SBE assay to identify LOH regions in mouse lung tumors 20 lung tumor DNA samples from 129S4xB6 Fl mice were subjected to LOH screening using the genome-wide SNP panel. The tumors range from grade 1-3 using criteria described in Jackson et al. (Table 3). 3828 out of 4290 SNP data points (89%) could be scored with confidence. For each data point, allelic imbalance factor (AIF) was calculated as described in materials and methods. The cutoff for positive call was set at AIF of >3 or <-3. With the assumption that most loci in the normal tissue of each Fl mouse are heterozygous, normal DNA was not subjected to the whole-genome screening protocol. Instead, assays were repeated on tumor and the corresponding tail DNA on individual markers that exhibited LOH in the first-pass screen. Figure 3 illustrates the global LOH landscape observed in this first pass. Markers scored positively for LOH are listed in Table 4. The majority (>98%) of tested loci remained heterozygous in the lung tumors, although certain regions of LOH were suggested by 44 positive markers. B6 markers on chromosome 14 were lost in DNA from one grade 3 tumor (840a), which was obtained from a Kra IA 2 mouse (Figure 4, Table 4). Two other grade 3 tumors were present in the sample set, 866h and 870a from KraslA2; p53+/- mice. Both tumors appeared to have lost all 129S4 markers on chromosome 11 (Figure 4, Table 4). 870a also showed LOH along chromosomes 9 and 10. In the initial screen, 16/44 positively scored markers were focal, defined by the absence of LOH in adjacent markers (Table 4). However, when 8 of these markers were screened in the second round, none had a positive score. Remaining markers still need to be screened in second round. On the other hand, positive scoring was concordant for most markers on chromosome 14 in 840a tumor DNA (Table 4). 118 Figure 3: Representative assay for multiple SNPs in single SBE reaction 11 SNPs were genotyped simultaneously in the SBE group lA (Table 2). A) Electrographic data from genotyping of SBE group lA on 803tail DNA. The X- axis is electrophoretic shift, a function affected by SBE primer length and dye chemistry of labeled ddNTPs. The Y-axis is signal amplitude reflecting the allele quantity. The blue, black, red, and green peaks correspond to the nucleotides G, C, T, A respectively. The G and A peaks of the SNP marker 786 are highlighted. B) Electrographic output from genotyping of SBE group 1A on 840a tumor DNA. Signal ratio of G to A peaks of marker 786 is altered, suggesting an LOH event. C) Summary of genotyping results of all group lA markers for 803tail and 840a. Signal ratio of the two allelic peaks was quantified for each SNP. AIF was calculated after normalizing peak ratios of sample to averages from three normal controls (see materials and methods). 119 0 - CD 0 '- 0) (o CtC- ? -o co' ,-,h 0,d ' 0_ r'-- d': t'o N cio 4~ (30 wool ng1 m i o ?( 00OUCNO--D O LOOC oO C 0 wI-- -- I-- - - a 0C 0) CU.) toN 0 0 0N 0 P I- I - CD V 00 0 0 CD 0 )- 0 - C1 ) Nr M o -- 0 0 N ~CDC N- O- N. II C~C 6 4 NJ6C 04 ND' 0 lu -aq uq 1? AVCI).0 C M - t 1 0 0 t-d-o 00 C ONNQNN N N C4 0 c1 I I 3I cc m ccrS 0 m- 0 o 0 co 00 0 c - CD 0 N C- (D -l ) co 0 a) Io 10 0 - 0 0- 0'? CD 9 > M m N 8 CU CC) - w 1. - CD 00 r,3 _I I _ ~I ~I I a O~ 0) N C' - N CC C C CC) a) ccc C-C 0 I- C c C%, dd 001 -c M m m Table 3: Characteristics of primary lung tumors analyzed The SNP screen was passed on 20 lung tumors from 129xB6 Fl hybrid mice and control tail DNA from two parental strains. The parental strains were not the biological parents of the F1 mice. The table lists the genotype of the mice from which individual lung tumors were dissected out. Histological grading was assigned using a grade 1-5 scale as described previously (Jackson et al. 2005). 121 122 Sample ID 796 1492 800d 800e 800i 803g 803i 803j 803k 818g 840a 840b 849a 849b 849c 849e 849f 850d 866h 866i 870a 878c Mouse genotype WT (C57CL6J) Kras LA2 (129S4/SvJAe) Kras A2 Kras LA2 Kras L 2 Kras LA2 Kras L2 Kras LA2 Kras L 2 Kras 4A2 Kras L 2 Kras L4 2 Kras L 2 Kras L 2 Kras LA2 Kras LA2 Kras L4 2 Kras M2 Kras 2; p53+/- Kras L42 ; p53+/- Kras L42; p53+/- Kras LA2; p 5 3 27OH/+ Histology (tail DNA) (tail DNA) Grade 1 Grade 2+ Grade 1-2 Grade 2, Grade 1 Grade 1 Grade 2, Grade 2, Grade 2+ (25%), grade 3 (75%) Grade 2, Grade 1 Grade 1 Grade 1 Grade 1 Grade 1 Grade 2+ Grade 2 (10%), 90% grade 3-3+ (90%) Grade 2, bone in tumors Grade 3+ Grade 1 Figure 4: Genome-wide LOH screening of primary mouse lung adenocarcinomas Grid illustrating data from the first-pass SNP screen: columns represent different samples; rows are markers organized by the chromosomal positions from UCSC Feb 2006 (mm8) assembly. Markers with positive AIF score are colored as follows: orange=AIF>3 (loss of B6 allele), blue=AIF<0.3 (loss of 129S4 allele). Shaded boxes are non-informative markers. DNA from 20 primary tumors, 2 parental strains, and representative F1 tails were screened on a genome-wide scale. Samples labeled with the same numerical prefix in the IDs were collected from the same mouse. 123 Q I Table 4: Positively scored LOH markers from first-pass genomic screen Listed are markers scored positively (AIF>3 or <0.3) in tumors but not control. The positive call rate was 1.1% (44/3828). In the second pass, 19 of the positive markers were individually genotyped by SBE on tumor and normal DNA. AIF was calculated using the paired data for each marker. The concordant positive call rate between the two passes was 37% (7/19). 126 Concordant Tumor Assay ID RefSNP ID Chromosome Nucleotide AIF (1st pass)AIF (2ndpass) LOH call 746 rs3657668 5 66812086 0.29 0.95 800e 892 rs3717068 13 30844331 3.08 800i 892 rs3717068 13 30844331 3.25 803g 746 rs3657668 5 66812086 0.27 818g 750 rs3709317 6 49704741 0.00 786 rs3720966 10727081 5.34 6.11 * 809 rs3724533 27108284 5.21 1.24 * 789 rs3713838 27644295 780.66 3275.75 * 790 rs3685393 30183496 295.64 5.27 * 811 rs3023409 48748605 6.37 3.20 * 14 840a 321 rs3021908 53097465 10.70 7.56 * 86 rs3089070 67715311 5.18 5.15 * 857 rs3705482 82835317 5.33 1.00 858 rs3674616 98406482 480.62 1.00 812 rs3657504 118523326 7.31 5.81 452 rs3023449 17 42120599 0.26 0.60 849b 4 rs3022839 1 137257629 3.09 849c 86 rs3089070 14 67715311 0.00 3.90 750 rs3709317 6 49704741 0.00 4.66 850d 879 rs3695889 9 86124796 0.02 1.00 452 rs3023449 17 42120599 0.24 0.75 850 rs3090435 6 146590775 3.30 201 rs3023256 44299873 0.21 461 rs3023258 54018188 0.00 866h 67 rs3023265 11 57576910 0.00 463 rs3088501 94146845 0.00 911 rs3697014 112103534 0.19 830 rs3714631 2 151657817 668.48 0.67 866i 833 rs3726717 3 52860548 98.83 1.00 892 rs3717068 13 30844331 0.19 1.06 766 rs3724779 8 117005581 23.23 768 rs3716232 16092806 180.31 769 rs3694785 9 16382724 169.88 770 rs4137954 98214467 4.87 445 rs3089912 20686036 77.82 446 rs3090586 24286974 89.13 912 rs3702150 10 68557908 3.70 870a 818 rs3722942 93074375 7.43 779 rs3660209 107394384 5.20 201 rs3023256 44299873 0.25 461 rs3023258 54018188 0.01 67 rs3023265 11 57576910 0.20 463 rs3088501 94146845 0.01 911 rs3697014 112103534 0.25 Count: 44 19 7 Total informative markers: 3828 Loss of wild-type p53 on chromosome 11 Three of the analyzed tumors (866h, 866i, 870a) came from KrasLA 2 ; p53+/- mice. The p53 null allele in the F1 mice came from a p53+/- (B6) parent, generated from 129 ES cells in initial targeting and subsequent backcrossing to B6 for 20+ generations. Interestingly, in the normal tissue of these mice, the 4 markers (799, 797, 700, 701) within +0.34Mb and -1.82Mb around the p53 locus on chromosome 11 were homozygous for the 129S4 allele (Figure 4). It appears that despite many generations of backcrossing to B6 mice, markers linked to the p53 knock-out allele on chromosome 11 remain the 129S4 ancestry. Other SNPs on chromosome 11 were heterozygous in normal DNA of these KrasL 2 ; p53+/- Fl mice. In the DNA from two tumors: 866h and 870a, the 129S4 alleles of these markers were lost, suggesting a selective loss of the whole chromosome 11 that contained wild-type p53. To test, PCR was performed to genotype the p53 locus. As shown in Figure 5A, the ratio of the wild-type to mutant p53 allele was decreased by about half in the PCR products from 866h and 870a, confirming the LOH data. The incomplete loss of wild-type signal could be due to a heterogeneous loss ofp53 among the tumor cells and/or contamination by normal stromal tissues in tumors. Reduced level of wild-type Kras on chromosome 6 One sample, 866h, showed a loss of the B6 allele at marker rs3090435 (assay ID 850). on the distal arm of chromosome 6. Intriguingly, the marker is only +1.39Mb away from the Kras gene towards the telomere. We sought to examine if the B6 allele of Kras, which marked the wild-type copy of the gene, has also undergone concomitant allelic loss. In 128 creating the Kraas 2 allele, a novel HindIII restriction site was introduced in Kras exon 1 along with the point mutation. An SBE assay was designed to test for the new SNP at the engineered HindIII site (Figure 5B). As shown in Figure 5C, the wild-type allele (A) of Kras in 866h showed a reduction in relative signal when compared to tail control by ~2x, which was within LOH cutoff. The assay was also performed 849c, 840b, 850d, and 866i tumor DNA. Neither the SNP nor Kras exhibited a change in allelic ratios in these tumors (data not shown). Comparison of LOH results to copy number data Four lung tumor samples (840a, 870a, 849c, 866h) were chosen for copy number analysis using a ROMA platform. ROMA uses oligonucleotide microarrays to relatively quantify the low-complexity representations of tumor vs. normal genomes. Global chromosomal copy number changes were apparent in 840a, 870a, and 866h (Figure 6). These changes were summarized in Table 5 along with LOH data. Discrepancy was apparent: multiple chromosomes have reduced in copy number but no LOH was associated. These include chromosomes 4, 5, 11, and 15 in 840a, chromosome 19 in 870a, and chromosome 9 in 866h. Interestingly, chromosome 6 was increased in copy number in 866h (Figure 6), suggesting the changes in allelic ratio observed in Kras and rs3090435 reflect a gain of the mutant (129S4) copy instead of a loss of wild-type allele. The wild-type allele of Kras has been suggested to suppress tumor development (Zhang et al. 2001). 129 Figure 5: Genotvping results at p53 and Kras loci A) p53 PCR genotyping was performed on tumor DNA from KrasL42; p53+/- mice to assess the relative intensities of wild-type p53 to the knocked-out allele with an inserted Neo cassette. B) An SBE assay designed to assess the ratio of KraSLA 2 to wild-type alleles of Kras. The assay genotypes for the single-nucleotide difference on the KrasLA 2 allele at an engineered HindIII site closely linked to the G12D expressing point mutation. Sequence presented comes from the wild-type Kras gene. The two single-nucleotides changes introduced in making the Kras 42 allele are marked by asterisks. The genotyping SBE primer is highlighted in the sequence with strand direction marked by the arrow. C) Kras genotyping results of 866h tumor DNA and control. Ratio of A/G peak heights reflects the wild-type to Kras? 2 allelic ratio. 130 866 870 866h tail 870a tail intl B a . ,;u 0.7 1 0.62 1.1 ensities Hindll G12D G A transltinn *.tirt. b - * .-attgt I TGTlGGTGTTGGAGC TGC GTAGGCAAGAGCGCCTGACGATA AG CTAATT TCACTT TGATGAGTATGA CCCTACGATraG. C A Neo - WT- "T". Ikl,.., Figure 6: Copy number analysis of tumor DNA by ROMA Four genome-wide moving median plots showing fluorescence ratios of labeled tumor to labeled tail DNA. DNA from four tumors: 840a, 870a, 849c, and 866h were analyzed by ROMA. The Y-axis is the loglO fluoresence ratio and the X-axis is an index of the probes genomic order based on UCSC mouse May 2004 (mm5) annotated assembly. Data from each chromosome are labeled with same color. Chromosomes with deviated signals from either the positive or negative baselines are labeled. 132 840a 4 -5 -11 870a -9 u-ilW- 849c +5 866h *90 o-11 0 whole mouse genome "As-1 I Table 5: Comparison of chromosomal LOH and copy number changes as suggested by SNP genotyping and ROMA analysis The same tumor DNA samples obtained from mice with indicated genotype were analyzed by SNPs and ROMA. The chromosomal LOH or copy number gains (+) and losses (-) observed respectively are listed. Tumor ID Mouse genotype LOH Copy number changes 840a KrasLA 2 14 -4, -5, -11, -14, -15, +6 870a KrasLA 2 ; p53+/- 9, 10, 11 -9, -10, -11, -19, +12 866h KraSA 2 ; p53+/- 11 -9, -11, +6 134 Discussion Using SNPs as markers for genome-wide LOH screen in mouse cancer This is the first report of a genome-wide LOH screen with SNPs in mouse tumor models. We described a protocol for performing genome-wide SNP genotyping by coupling SBE and capillary electrophoresis using SNaPshot TM fluorescence chemistry (Applied Biosystems). The robustness and sensitivity of the method in simultaneously genotyping multiple markers have been demonstrated by various groups (Makridakis and Reichardt 2001; Norton et al. 2002; Ben-Avi et al. 2004). Implementation of the method for genome-wide genotyping has not been previously described. The current study extended the technique to a larger scale to analyze 147 SNPs throughout the mouse genome in 15 SBE reactions. The screening procedure positively identified LOH of chromosome 11 in two samples of mouse tumors, both of which were confirmed to involve a loss of the wild-type p53 allele. Our data show SNP genotyping in Fl hybrid mice is a viable method for LOH screening. The limiting factor to perform genome-wide SNP screening in most labs is cost. Most other established methods for mouse genomic SNP genotyping involve specialized instruments, such as MALDI-TOF mass spectrometry(Wiltshire et al. 2003; Pletcher et al. 2004; Moran et al. 2006). On the other hand, the fixed cost of our screening method is essentially just the cost of primers, as only standard PCR and routine sequencing machines are needed. The presented protocol can likely be adoptable by most laboratories and can be further optimized to improve each SBE reaction to the highest accuracy level capable by same type of assay (Makridakis and Reichardt 2001; Norton et al. 2002). Sensitivity of the method needs to be tested on more markers in a dilution 135 experiment with heterozygous vs. different homozygous DNA. Directions for future improvements include further multiplexing of PCR, testing the assay utility with DNA amplified paraffin embedded samples, which would allow archived tissues to be characterized, and validating the use of the SNP panel on other mouse strains. Developing a bioinformatics approach to account for experimental noise may also improve the robustness of LOH assignments. Benchmarking the accuracy, sensitivity, and cost of this protocol against an established assay (Wiltshire et al. 2003; Pletcher et al. 2004; Moran et al. 2006) will provide an objective comparison of the different mouse SNP genotyping methods. Implications on lung cancer genetics Within the technical limit of our assay, the overall LOH rate was low in the experimental cohort of mouse lung tumors. The majority of the tumors analyzed maintained both parental alleles in all loci. Only 3/20 tumors exhibited various degree of LOH; all were histologically graded 3 or higher. Among the three tumors, LOH of chromosomes 9, 10, 11, and 14 was detected. Combining LOH and CGH data To compare LOH results to alterations in copy number, ROMA was used to characterize these 3 tumors along with one other sample, 849c. LOH may result through deletion, which is associated with a physical loss of the wild-type gene copy, or through non- disjunction-led chromosomal duplication, which might not result in a copy number change. In fact, most LOH events are not associated with copy number changes at least in humans (Huang et al. 2004; Beroukhim et al. 2006). In our case, limited by the 136 technicalities of the screening platforms, the reverse was observed: copy number losses of certain chromosomes were not associated with LOH (Table 5). Sample 840a provided the clearest example with ROMA showing reduced copy signals for chromosomes 4, 5, 11, 14, and 15 at nearly same magnitude, but only chromosome 14 showed LOH in our screen. There might be a a technical and a biological basis for the discrepancy. Technically, this discrepancy could be due to a difference in sensitivity of the two assays. While the quantitativeness of our assay was tested on two SNPs, other markers may behave differently. Furthermore, the sensitivity limit of ROMA is unclear. Although we were able to verify high-amplitude ROMA signal such as that from the N-Myc amplicon (see Chapter 2), subtle copy number differences might exist but not be detected by ROMA. The technical capability of the assay can be tested by confirming our LOH results using an independent assay, such as another SNP screen (Wade et al. 2002; Pletcher et al. 2004; Owens et al. 2005; Moran et al. 2006) or SSLP genotyping (Dietrich et al. 1994). We also attempted our screen on breast tumor DNA that have been previously characterized by array CGH (Chao et al. 2005). The concordance between known copy number losses and our LOH screening data appeared higher (Supplemental Figure 1), suggesting the discordance in the lung tumor samples may be real. Several biological reasons may explain the discordant copy number and allelotyping results. Since a tumor is believed to be a heterogeneous group of clonally selected cells, analysis using total DNA from a tumor is an assessment of the averaged genetic changes in its composite cells. As such, a chromosomal copy number reduction in total DNA implies a selection for the hemizygous state of that chromosome has occurred, while a lack of allelic loss suggests the choice of which chromosome to lose was random. This may 137 result if the chromosome involved contains one or more haplo-insufficient tumor suppressor genes, which can confer tumorigenicity when dosage is reduced through hemizygosity. Although tumor suppressor is classically thought to act through a 'two-hit' inactivation process, increasing number of haplo-insufficient tumor suppressor genes have been described. This model is based on the assumption that the tumor cells are diploid for the chromosomes without apparent changes by ROMA. However, if the cells are mostly tetraploid, a state that is believed to precede widespread aneuploidy (Fujiwara et al. 2005), which was in fact observed, the baseline for ROMA would no longer be two chromosomes. A decrease in ROMA signal could mean 4-1 or 4-2 chromosomes. Either one or both alleles could remain, leading to discordant results. While ROMA cannot distinguish ploidy, LOH analysis may lack the power to distinguish subtle allelic changes. Future experiments to complement karyotyping or FACS-based ploidy analysis on cells from the same tumors may resolve some issues. Chromosome 11 in mouse lung tumorigenesis Chromosome 11 LOH was seen in 2 tumors that came from KrasLA 2 ;p53+/- mice. In both cases, the lost chromosome contained the wild-type p53 allele. This result is consistent to the observation by Jackson et al. that tumors induced by conditionally expressing Kras G1 2D and ap53 deletion allele in the lung have also lost the wild-type copy ofp53. In the present study, chromosome 11 LOH occurred in combination with LOH of chromosome 6 in one tumor and LOH of chromosomes 9 and 10 in the other. It was possible that loss of p53 on chromosome 11 has provided a permissive environment for genome-wide aneuploidy. Consistent to the hypothesis, ROMA-generated copy number data showed widespread aneuploidy in these two tumors. p53 has been implicated in 138 preventing tetraploid cells to proceed through the cell cycle, which frequently leads to aneuploidy (Lanni and Jacks 1998; Meraldi et al. 2002; Fujiwara et al. 2005). Interestingly, in tumor 840a from a KrasLA 2 ;p53+/+ mouse, chromosome 11 was among the chromosomes that showed copy number loss without LOH, which might imply haplo- insufficiency as discussed above. One candidate locus could again be p53. Although often lost biallelically, ~50% of mouse and human tumors with a p53 mutation in fact retain the wild-type copy of the gene (Venkatachalam et al. 1998; Trkova et al. 2003). In addition, differences in p53 dosage have been seen to affect tumor phenotype of a mouse model (Hemann et al. 2003). Finally, it is important to note that mouse chromosome 11 is very gene-rich, and its distal arm shares conserved synteny to the whole human chromosome 17, which is home to many disease-related genes included the tumor suppressors Brcal and Nfl. In lung cancer, 17p12-13 deletion is frequently observed (Balsara and Testa 2002). While it is tempting to relate our observations of chromosome 11 LOH and/or copy number loss to p53 function, other loci may also be critical. Note on genetic background We generated our tumor samples using Fl mice from a 129xB6 cross. Strain-specific phenotypic differences appear to exist. Kras LA mice on a B6 background develop lung tumors at higher multiplicity than on 129S4 (Michel Dupage, personal communication). Molecularly, it is unclear whether tumors from C57BL/6J, 129S4/SvJae, and Fl mice have different types and rates of genetic changes. LOH resulting from mitotic recombination can get suppressed in F1 hybrids of different mouse strains (Shao et al. 2001). Furthermore, tumor susceptibility gene(s) specific to B6 and 129 strains have been mapped to chromosome 11 in a different tumor type (Reilly et al. 2004). Identity and 139 function of the modifiers are unclear but may affect LOH frequency of chromosome 11 in tumors of other types as well. Concluding note Using a multiplexed SNaPshot T " genotyping protocol for genome-wide SNP detection, we have detected LOH along several chromosomes on mouse tumors. Further optimization will improve both the throughput and accuracy of the assay. Aside from LOH identification in tumors, a protocol for SNP genotyping method for mouse can also be useful for other purposes such as positional cloning of modifiers in cancer and in other diseases. Materials and Methods Tumor DNA isolation KRas LA2/+ mice on a 129S4/SvJae background were crossed to wild-type, p53+/- or p53 R270H/+ mice on C57BL6J background to obtain Fl progeny. Lung tumors were dissected from the lungs of Fl mice between 5-8 months of age. One portion of each tumor was fixed in formalin, sectioned in paraffin, and stained in hematoxylin and eosin. Histological grading of each tumor was assigned based on a 1-5 scale as described previously (Jackson et al. 2005). Remaining tumor material was stored at -80 0 C prior to DNA isolation. DNA was extracted from thawed tissues using reagents and protocols in Puregene DNA isolation kit (Gentra Systems, Inc.). SNP genotyping 140 SNP genotype was assessed using a single-base extension (SBE) method (Sokolov 1990). SNP targets were first amplified by PCR using 2ng genomic DNA, 0.3mM dNTPs, 0.4mM PCR primers, lx GeneAmp Gold buffer and 0.5U of AmpliTaq Gold polymerase (Applied Biosystems) under the following conditions: 95 0 C for 9mins, 35 cycles of 94?C for 30s, 55 0 C for 30s and 72oC for 45s, and a final extension for 5 min at 72 0 C. Unincorporated PCR primers and dNTPs were removed with 2U exonuclease I (Applied Biosystems) and 2U shrimp alkaline phosphatase (Applied Biosystems) at 37 0 C for 1 hour. 0.15-1.5pmol of pmol of SBE primer was added to the treated PCR templates with SNaPshot Multiplex Ready Reaction Mix (Applied Biosystems) and cycled 25 times at 96 0 C for 10s, 50 0 C for 5s and 60 0 C for 30s. Post-extension products were treated with 0.25U of shrimp alkaline phosphatase (Applied Biosystems) to remove unincorporated ddNTPs. Final products were mixed with 0.25ml of Liz-120 size standards (Applied Biosystems) and ran on the Applied Biosystems 3700 DNA Analyzer. To test markers for polymorphism in 129S4/SvJae vs. C57BL6J strains, SNPs were identified in public databases generated from two large-scale SNP identification efforts (Germer et al. 2000; Lindblad-Toh et al. 2000b) and tested individually using primers listed in Supplementary Table 1. To perform genome-wide analysis, 147 validated SNPs were chosen. PCR were performed in singlet or duplex on 96-well plates and pooled according to their SBE group as in Table 2. Pooled products were purified and concentrated using multi-well PCR purification kit (Qiagen) for SBE reactions as described above. 141 Positional information of each SNP is obtained through sequence blat search of the UCSC Mouse Feb 2006 (mm8) assembly. SNP Data Analysis First round data analysis was done with Applied Biosystems GeneMapper 3.7 software. By comparing electrographs of multiplex and individual SBE reactions, peaks corresponding to each SNP in a multiplex lane were manually identified. Marker boundaries were set as bins to guide the software to automatically assign genotypes for peaks that fall within. All computer-generated assignments were inspected manually before final analysis. The height ratio of the peaks associated with the two alleles of each SNP was calculated. Allelic imbalance factor (AIF) of each marker in a sample is determined as follows: AIF = (HI/H 2 )/(Hlref/H 2 ref), where H 1 = Sample peak height associated with 129S4 allele, H 2 = Sample peak height associated with C57BL6J allele, Hiref = Reference peak height associated with 129S4 allele, and H2ref = Reference peak height associated with C57BL6J allele. Reference peak heights were calculated by averaging data from three independent normal Fl DNA controls. A positive call for LOH was made when AIF is >3 or <0.3 LOH assessment of p53 and Kras loci 142 Ratio of wild-type to knock-out (with neomycin cassette) allele of p53 in tumors was assessed using the standard tail genotyping protocol with the following primers: p53x6.5: ACAGCGTGGTGGTACCTTAT, p53x7: TATACTCAGAGCCGGCCT, and Neol8.5: TCCTCGTGCTTTACGGTATC, yielding 375bp and 525bp products corresponding to the wild-type and knocked-out alleles. Number of PCR cycles was reduced to 22 in genotyping tumors. To assess the ratio of Kras LA2 to wild-type alleles, an SBE assay was used to genotype the single-nucleotide difference in the LA2 allele at the novel HindIII site near the G12D expressing point mutation. The following PCR primers were used: LA-F 1: GGTTACTCTGTACATCTGTAGTCACTG and LA-R1: AAGCGCACGCAGACTGTAG. SBE primer sequence was AAGGCCTGCTGAAAATGACTGAGTATAA. DNA copy number analysis ROMA was employed to assess genome-wide copy number changes and was performed as described in Chapter Two of this thesis. Acknowledgements The authors would like to thank Kevin Shannon for generous gifts of breast tumor and cell line DNA samples used in experiment described in the supplemental section. We also thank Charlie Whittaker and Ray Farhner for computing support, Denise Crowley and 143 Mike Brown for histology preparation, and Rob Bronson for pathology consultation. T.J. is a Howard Hughes Investigator. References Balsara, B.R. and Testa, J.R. 2002. Chromosomal imbalances in human lung cancer. Oncogene 21(45): 6877-6883. Ben-Avi, L., Durst, R., Shpitzen, S., Leitersdorf, E., and Meiner, V. 2004. Apolipoprotein E genotyping: accurate, simple, high throughput method using ABI Prism SNaPshot Multiplex System. JAlzheimers Dis 6(5): 497-501. Benavides, F., Conti, C.J., LaCava, M., Flores, M., Glasscock, E., Sternik, G., Gimenez- Conti, I.B., Johnston, D.A., Dunsford, H.A., Goldstein, L.S., and Rodriguez, L.V. 2003. Loss of heterozygosity analysis of mouse pulmonary adenomas induced by coal tar. Environ Mol Mutagen 41(4): 300-308. Beroukhim, R., Lin, M., Park, Y., Hao, K., Zhao, X., Garraway, L.A., Fox, E.A., Hochberg, E.P., Mellinghoff, I.K., Hofer, M.D., Descazeaud, A., Rubin, M.A., Meyerson, M., Wong, W.H., Sellers, W.R., and Li, C. 2006. Inferring loss-of- heterozygosity from unpaired tumors using high-density oligonucleotide SNP arrays. PLoS Comput Biol 2(5): e41. Chao, R.C., Pyzel, U., Fridlyand, J., Kuo, Y.M., Teel, L., Haaga, J., Borowsky, A., Horvai, A., Kogan, S.C., Bonifas, J., Huey, B., Jacks, T.E., Albertson, D.G., and Shannon, K.M. 2005. Therapy-induced malignant neoplasms in Nfl mutant mice. Cancer Cell 8(4): 337-348. Dietrich, W.F., Radany, E.H., Smith, J.S., Bishop, J.M., Hanahan, D., and Lander, E.S. 1994. Genome-wide search for loss of heterozygosity in transgenic mouse tumors reveals candidate tumor suppressor genes on chromosomes 9 and 16. Proc Natl Acad Sci USA 91(20): 9451-9455. Frazer, K.A., Wade, C.M., Hinds, D.A., Patil, N., Cox, D.R., and Daly, M.J. 2004. Segmental phylogenetic relationships of inbred mouse strains revealed by fine- scale analysis of sequence variation across 4.6 mb of mouse genome. Genome Res 14(8): 1493-1500. 144 Fujiwara, T., Bandi, M., Nitta, M., Ivanova, E.V., Bronson, R.T., and Pellman, D. 2005. Cytokinesis failure generating tetraploids promotes tumorigenesis in p53-null cells. Nature 437(7061): 1043-1047. Germer, S., Holland, M.J., and Higuchi, R. 2000. High-throughput SNP allele-frequency determination in pooled DNA samples by kinetic PCR. Genome Res 10(2): 258- 266. Grupe, A., Germer, S., Usuka, J., Aud, D., Belknap, J.K., Klein, R.F., Ahluwalia, M.K., Higuchi, R., and Peltz, G. 2001. In silico mapping of complex disease-related traits in mice. Science 292(5523): 1915-1918. Hemann, M.T., Fridman, J.S., Zilfou, J.T., Hernando, E., Paddison, P.J., Cordon-Cardo, C., Hannon, G.J., and Lowe, S.W. 2003. An epi-allelic series of p53 hypomorphs created by stable RNAi produces distinct tumor phenotypes in vivo. Nat Genet 33(3): 396-400. Herzog, C.R., Devereux, T.R., Pittman, B., and You, M. 2002. Carcinogenic induction directs the selection of allelic losses in mouse lung tumorigenesis. Cancer Res 62(22): 6424-6429. Huang, J., Wei, W., Zhang, J., Liu, G., Bignell, G.R., Stratton, M.R., Futreal, P.A., Wooster, R., Jones, K.W., and Shapero, M.H. 2004. Whole genome DNA copy number changes identified by high density oligonucleotide arrays. Hum Genomics 1(4): 287-299. Jackson, E.L., Olive, K.P., Tuveson, D.A., Bronson, R., Crowley, D., Brown, M., and Jacks, T. 2005. The differential effects of mutant p53 alleles on advanced murine lung cancer. Cancer Res 65(22): 10280-10288. Johnson, L., Mercer, K., Greenbaum, D., Bronson, R.T., Crowley, D., Tuveson, D.A., and Jacks, T. 2001. Somatic activation of the K-ras oncogene causes early onset lung cancer in mice. Nature 410(6832): 1111-1116. Knudson, A.G., Jr. 1971. Mutation and cancer: statistical study of retinoblastoma. Proc Natl Acad Sci USA 68(4): 820-823. Lanni, J.S. and Jacks, T. 1998. Characterization of the p53-dependent postmitotic checkpoint following spindle disruption. Mol Cell Biol 18(2): 1055-1064. 145 Lindblad-Toh, K., Winchester, E., Daly, M.J., Wang, D.G., Hirschhorn, J.N., Laviolette, J.P., Ardlie, K., Reich, D.E., Robinson, E., Sklar, P., Shah, N., Thomas, D., Fan, J.B., Gingeras, T., Warrington, J., Patil, N., Hudson, T.J., and Lander, E.S. 2000. Large-scale discovery and genotyping of single-nucleotide polymorphisms in the mouse. Nat Genet 24(4): 381-386. Makridakis, N.M. and Reichardt, J.K. 2001. Multiplex automated primer extension analysis: simultaneous genotyping of several polymorphisms. Biotechniques 31(6): 1374-1380. Meraldi, P., Honda, R., and Nigg, E.A. 2002. Aurora-A overexpression reveals tetraploidization as a major route to centrosome amplification in p53-/- cells. Embo J21(4): 483-492. Moran, J.L., Bolton, A.D., Tran, P.V., Brown, A., Dwyer, N.D., Manning, D.K., Bjork, B.C., Li, C., Montgomery, K., Siepka, S.M., Vitaterna, M.H., Takahashi, J.S., Wiltshire, T., Kwiatkowski, D.J., Kucherlapati, R., and Beier, D.R. 2006. Utilization of a whole genome SNP panel for efficient genetic mapping in the mouse. Genome Res 16(3): 436-440. Norton, N., Williams, N.M., Williams, H.J., Spurlock, G., Kirov, G., Morris, D.W., Hoogendoorn, B., Owen, M.J., and O'Donovan, M.C. 2002. Universal, robust, highly quantitative SNP allele frequency measurement in DNA pools. Hum Genet 110(5): 471-478. Owens, S.E., Broman, K.W., Wiltshire, T., Elmore, J.B., Bradley, K.M., Smith, J.R., and Southard-Smith, E.M. 2005. Genome-wide linkage identifies novel modifier loci of aganglionosis in the Sox10Dom model of Hirschsprung disease. Hum Mol Genet 14(11): 1549-1558. Petkov, P.M., Ding, Y., Cassell, M.A., Zhang, W., Wagner, G., Sargent, E.E., Asquith, S., Crew, V., Johnson, K.A., Robinson, P., Scott, V.E., and Wiles, M.V. 2004. An efficient SNP system for mouse genome scanning and elucidating strain relationships. Genome Res 14(9): 1806-1811. Pletcher, M.T., McClurg, P., Batalov, S., Su, A.I., Barnes, S.W., Lagler, E., Korstanje, R., Wang, X., Nusskern, D., Bogue, M.A., Mural, R.J., Paigen, B., and Wiltshire, T. 2004. Use of a dense single nucleotide polymorphism map for in silico mapping in the mouse. PLoSBiol 2(12): e393. 146 Radany, E.H., Hong, K., Kesharvarzi, S., Lander, E.S., and Bishop, J.M. 1997. Mouse mammary tumor virus/v-Ha-ras transgene-induced mammary tumors exhibit strain-specific allelic loss on mouse chromosome 4. Proc Natl Acad Sci USA 94(16): 8664-8669. Reilly, K.M., Tuskan, R.G., Christy, E., Loisel, D.A., Ledger, J., Bronson, R.T., Smith, C.D., Tsang, S., Munroe, D.J., and Jacks, T. 2004. Susceptibility to astrocytoma in mice mutant for Nfl and Trp53 is linked to chromosome 11 and subject to epigenetic effects. Proc Natl Acad Sci USA 101(35): 13008-13013. Shao, C., Stambrook, P.J., and Tischfield, J.A. 2001. Mitotic recombination is suppressed by chromosomal divergence in hybrids of distantly related mouse strains. Nat Genet 28(2): 169-172. Sokolov, B.P. 1990. Primer extension technique for the detection of single nucleotide in genomic DNA. Nucleic Acids Res 18(12): 3671. Sweet-Cordero, A., Mukherjee, S., Subramanian, A., You, H., Roix, J.J., Ladd-Acosta, C., Mesirov, J., Golub, T.R., and Jacks, T. 2005. An oncogenic KRAS2 expression signature identified by cross-species gene-expression analysis. Nat Genet 37(1): 48-55. Syvanen, A.C. 2001. Accessing genetic variation: genotyping single nucleotide polymorphisms. Nat Rev Genet 2(12): 930-942. Trkova, M., Foretova, L., Kodet, R., Hedvicakova, P., and Sedlacek, Z. 2003. A Li- Fraumeni syndrome family with retained heterozygosity for a germline TP53 mutation in two tumors. Cancer Genet Cytogenet 145(1): 60-64. Venkatachalam, S., Shi, Y.P., Jones, S.N., Vogel, H., Bradley, A., Pinkel, D., and Donehower, L.A. 1998. Retention of wild-type p53 in tumors from p53 heterozygous mice: reduction of p53 dosage can promote cancer formation. Embo J 17(16): 4657-4667. Wade, C.M., Kulbokas, E.J., 3rd, Kirby, A.W., Zody, M.C., Mullikin, J.C., Lander, E.S., Lindblad-Toh, K., and Daly, M.J. 2002. The mosaic structure of variation in the laboratory mouse genome. Nature 420(6915): 574-578. Waterston, R.H. Lindblad-Toh, K. Birney, E. Rogers, J. Abril, J.F. Agarwal, P. Agarwala, R. Ainscough, R. Alexandersson, M. An, P. Antonarakis, S.E. 147 Attwood, J. Baertsch, R. Bailey, J. Barlow, K. Beck, S. Berry, E. Birren, B. Bloom, T. Bork, P. Botcherby, M. Bray, N. Brent, M.R. Brown, D.G. Brown, S.D. Bult, C. Burton, J. Butler, J. Campbell, R.D. Carninci, P. Cawley, S. Chiaromonte, F. Chinwalla, A.T. Church, D.M. Clamp, M. Clee, C. Collins, F.S. Cook, L.L. Copley, R.R. Coulson, A. Couronne, O. Cuff, J. Curwen, V. Cutts, T. Daly, M. David, R. Davies, J. Delehaunty, K.D. Deri, J. Dermitzakis, E.T. Dewey, C. Dickens, N.J. Diekhans, M. Dodge, S. Dubchak, I. Dunn, D.M. Eddy, S.R. Elnitski, L. Emes, R.D. Eswara, P. Eyras, E. Felsenfeld, A. Fewell, G.A. Flicek, P. Foley, K. Frankel, W.N. Fulton, L.A. Fulton, R.S. Furey, T.S. Gage, D. Gibbs, R.A. Glusman, G. Gnerre, S. Goldman, N. Goodstadt, L. Grafham, D. Graves, T.A. Green, E.D. Gregory, S. Guigo, R. Guyer, M. Hardison, R.C. Haussler, D. Hayashizaki, Y. Hillier, L.W. Hinrichs, A. Hlavina, W. Holzer, T. Hsu, F. Hua, A. Hubbard, T. Hunt, A. Jackson, I. Jaffe, D.B. Johnson, L.S. Jones, M. Jones, T.A. Joy, A. Kamal, M. Karlsson, E.K. Karolchik, D. Kasprzyk, A. Kawai, J. Keibler, E. Kells, C. Kent, W.J. Kirby, A. Kolbe, D.L. Korf, I. Kucherlapati, R.S. Kulbokas, E.J. Kulp, D. Landers, T. Leger, J.P. Leonard, S. Letunic, I. Levine, R. Li, J. Li, M. Lloyd, C. Lucas, S. Ma, B. Maglott, D.R. Mardis, E.R. Matthews, L. Mauceli, E. Mayer, J.H. McCarthy, M. McCombie, W.R. McLaren, S. McLay, K. McPherson, J.D. Meldrim, J. Meredith, B. Mesirov, J.P. Miller, W. Miner, T.L. Mongin, E. Montgomery, K.T. Morgan, M. Mott, R. Mullikin, J.C. Muzny, D.M. Nash, W.E. Nelson, J.O. Nhan, M.N. Nicol, R. Ning, Z. Nusbaum, C. O'Connor, M.J. Okazaki, Y. Oliver, K. Overton-Larty, E. Pachter, L. Parra, G. Pepin, K.H. Peterson, J. Pevzner, P. Plumb, R. Pohl, C.S. Poliakov, A. Ponce, T.C. Ponting, C.P. Potter, S. Quail, M. Reymond, A. Roe, B.A. Roskin, K.M. Rubin, E.M. Rust, A.G. Santos, R. Sapojnikov, V. Schultz, B. Schultz, J. Schwartz, M.S. Schwartz, S. Scott, C. Seaman, S. Searle, S. Sharpe, T. Sheridan, A. Shownkeen, R. Sims, S. Singer, J.B. Slater, G. Smit, A. Smith, D.R. Spencer, B. Stabenau, A. Stange-Thomann, N. Sugnet, C. Suyama, M. Tesler, G. Thompson, J. Torrents, D. Trevaskis, E. Tromp, J. Ucla, C. Ureta-Vidal, A. Vinson, J.P. Von Niederhausern, A.C. Wade, C.M. Wall, M. Weber, R.J. Weiss, R.B. Wendl, M.C. West, A.P. Wetterstrand, K. Wheeler, R. Whelan, S. Wierzbowski, J. Willey, D. Williams, S. Wilson, R.K. Winter, E. Worley, K.C. Wyman, D. Yang, S. Yang, S.P. Zdobnov, E.M. Zody, M.C. and Lander, E.S. 2002. Initial sequencing and comparative analysis of the mouse genome. Nature 420(6915): 520-562. Wiltshire, T., Pletcher, M.T., Batalov, S., Barnes, S.W., Tarantino, L.M., Cooke, M.P., Wu, H., Smylie, K., Santrosyan, A., Copeland, N.G., Jenkins, N.A., Kalush, F., Mural, R.J., Glynne, R.J., Kay, S.A., Adams, M.D., and Fletcher, C.F. 2003. Genome-wide single-nucleotide polymorphism analysis defines haplotype patterns in mouse. Proc NatlAcad Sci USA 100(6): 3380-3385. Wu, Y., Renard, C.A., Apiou, F., Huerre, M., Tiollais, P., Dutrillaux, B., and Buendia, M.A. 2002. Recurrent allelic deletions at mouse chromosomes 4 and 14 in Myc- induced liver tumors. Oncogene 21(10): 1518-1526. 148 Zhang, J., Hunter, K.W., Gandolph, M., Rowe, W.L., Finney, R.P., Kelley, J.M., Edmonson, M., and Buetow, K.H. 2005. A high-resolution multistrain haplotype analysis of laboratory mouse genome reveals three distinctive genetic variation patterns. Genome Res 15(2): 241-249. Zhang, Z., Wang, Y., Vikis, H.G., Johnson, L., Liu, G., Li, J., Anderson, M.W., Sills, R.C., Hong, H.L., Devereux, T.R., Jacks, T., Guan, K.L., and You, M. 2001. Wildtype Kras2 can inhibit lung carcinogenesis in mice. Nat Genet 29(1): 25-33. 149 Supplemental Information Supplemental Figure 1: Genome-wide LOH screening of mouse breast tumors and tumor cell line Grid illustrating data from the SNP screen columns: columns represent different samples; rows are markers organized by the chromosomal positions from UCSC Feb 2006 (mm8) assembly. AIF scores are labeled in each box. n/a=non-informative markers. Markers with positive AIF score are colored as follows: orange=AIF>3 (loss of B6 allele), blue=AIF<0.3 (loss of 129S4 allele). Tumor DNA came from breast tumors (Bl, B2) or breast tumor cell lines (B1CL and C2CL) from Nfl+/- mice treated with radiation and cyclophosphamide (Chao et al. 2005). DNA was gift from Kevin Shannon. B1 DNA expressed copy number losses in whole or large regions of chromosomes 8, 11, 12, 14, 19 when profiled by array CGH. Most loci on these chromosomes scored positively for LOH. 150 E Sample ID E o 0 o _ Z N N o z u) mom 16141960 713 1.17 0.9 33560855 714 0.906 0.88 0.99 0.94 34060397 715 0.663 0.95 53107439 898 1.159 1.29 1.59 1.29 66668249 600 1.351 1 97464195 821 1.019 111247157 822 0.95 0.93 121777951 721 1.325 1.35 1.18 138360597 4 1.066 1.19 1.22 1.22 155719123 823 1.066 0.91 1.06 0.98 177919260 900 1.33 1.05 1.27 1.07 12070016 722 0.883 0.88 1.07 12070228 723 1.03 1.01 1.08 28621371 724 0.938 0.94 1.13 45954896 726 1.57 1.13 46472711 727 0.82 0.57 2 69373291 405 1.215 0.89 1.18 0.82 107222011 826 1.011 0.92 118395454 827 1.4 1.14 1.2 152709331 830 0.823 0.86 0.87 168774394 730 1.102 1 1.21 1.28 169277255 731 1.25 1.33 12531240 732 0.698 0.85 28779003 831 0.95 1.18 1 0.89 43382257 832 1.21 1.16 54179064 833 0.658 1.25 3 66900560 20 1.42 0.93 1.09 1 80194307 834 1.259 0.87 96226609 835 1.069 0.95 0.98 134326444 412 0.878 0.83 0.91 0.85 144714634 413 1.457 1.75 0.81 0.9 8073046 471 1.122 0.65 1.01 26596737 738 0.94 1.54 53874480 839 1.471 1.25 61518885 840 1.312 1 4 74855800 415 1.109 0.88 85139822 841 1.507 99041841 478 1.054 0.72 0.85 0.87 130811701 740 1.427 1.27 1.37 1.24 131376085 741 0.832 0.85 0.86 0.92 140240863 218 0.984 1.01 0.74 0.84 22413727 32 0.885 1.08 23730340 742 1.164 1.38 1.11 0.99 49773577 422 1.069 1.53 50105985 745 0.873 0.71 0.84 5 65816926 746 0.44 79989864 843 1.239 1.39 1.23 0.79 99985513 844 116250135 37 1.053 0.83 0.91 0.75 124509140 33 1.326 1 1.06 131433208 748 0.899 0.91 1.04 0.98 18655079 426 1.06 0.94 1.26 1.04 35969149 845 1.119 1.05 1.16 6 50227000 750 0.941 0.79 0.95 0.96 133171923 922 1.109 0.98 0.96 0.94 148155268 850 0.934 1.17 0.97 4961 820 28228061 606 0.77 43598764 851 0.748 0.83 78082981 854 1.09 1.06 1.27 1.37 7 88768207 435 0.85 1.09 1.22 108372753 436 0.956 0.85 117242899 756 1.392 1.05 1.09 117499219 757 0.759 0.94 0.93 1.02 121080866 339 0.656 0.67 0.82 0.85 12714613 25189627 33689228 56995287 57031272 89521985 115796380 116457023 16547432 16834053 89080475 100802904 21831681 25479871 58804746 71047972 96023689 110493528 21791444 44081495 53799952 57358532 67338072 68899165 69447433 69504211 93914069 7 06159111 8 0588674 52059690 94317217 100385178 114701971 18116231 30724973 44450246 60683731 93814910 115468718 7269139 22818444 23356191 25893857 41573852 45227636 60528292 75859013 92333517 112509253 20402168 32440466 41170909 52494598 62885156 76278665 89405702 5677800 46036961 64302751 87237069 89495624 12294400 20829561 42022515 53137039 67644893 33996227 63588982 3777 20523042 30292072 44235554 1.01 1.32 0.95 1.21 1.04 0.85 0.88 1.38 0.83 1.1 1.02 4451 0.961 446 1.061 0.661 0.76 448 0.9611 1 0.97 463 1.257 1.18 483 814 786 809 789 811 321 86 857 858 895 96 808 859 860 486 913 101 488 794 490 455 882 883 452 862 453 803 109 0.9 0.441 0.843 1.165 0.964 1.183 0.868 1.022 0.975 0.724 0.839 1.308 1.109 1.024 1.083 0.876 0.699 0.993 1.145 0.974 1.453 1.295 1.205 1.037 1.128 1.412 0.96 1.18 0.96 0.91 1.05 0.87 0.99 1.72 1.31 0.99 1.04 1.04 2.UO 1.53 0.76 1.16 0.95 2.93 1.21 0.78 2.29 n Q n E CA 2.34 1.38 0.99 1.97 0.89 1.02 1.04 1.32 0.98 1.87 0.79 0.73 0.97 1.4 0.49 1.02 1.6 0.92 0.78 1.41 0.69 0.98 1.04 1.12 0.52 1.13 0.92 0.9 0.85 1.14 0.99 0.99 0.66 1.31 8 9 15 16 17 18 0.81 1.52 1.19 0.7 1.03 0.96 2.57 0.95 1.05 2.07 n 9C 1.29 0.91 1.03 1.17 0.89 0.98 1.24 1.2 1.14 0.89 1.08 0.87 1.15 1.01 1.11 0.93 1.28 0.73 1 0.89 1.03 0.94 1.23 1.25 0.64 1.07 1.1 1.23 0.99 0.97 1.2 0.93 0.99 1.23 0.83 1.27 1.06 0.92 0.98 0.88 1.13 1 0.96 1.15 0.99 0.71 0.94 1.31 0.91 0.98 111951607 ---- 445 0.96 1 -11 ^ .- I . A& Supplemental Table 1: Validation of 358 SNP markers in 129S4/SvJae vs. C57BL6J mouse strains Genotyping results of 358 SNPs from public databases were obtained using a single-base extension method. Data were obtained from genotyping DNA from 129S4/SvJae, C57BL6J and 129S4/SvJae x C57BL6J Fl mice. The 147 highlighted SNPs that distinguish between the strains were chosen for the genome-wide screen. Chromosomal coordinates are based on UCSC mouse Feb 2006 (mm8) annotated assembly. Primer sequences used for PCR and SBE are listed in Supplemental Table 1. 153 Assay ID RedSNP ID Chromosome Nucleotide 129S4 B6 F1(129S4xB6) 712 rs3691476 1 15866460 A CA CA 713 rs3684370 1 15963812 T C CT 714 rs3662163 1 32822284 T C CT 715 rs3704392 1 33320247 T G GT 400 rs3090110 1 45907197 C T 898 rs3667466 1 51794155 C T CT 600 rs3022802 1 64979783 G T GT 899 rs3694327 1 79643870 C C CA 716 rs3716254 1 93320412 T A A 717 rs3657283 1 93320420 T C C 821 rs3704980 1 97461966 T G GT 822 rs3685188 1 110654171 T C CT 718 rs3713835 1 111883427 C T T 719 rs3662850 1 111887865 C T T 721 rs3664018 1 121061603 C T CT 720 rs3689749 1 121090273 G A A 603 rs3022832 1 127102172 C - G 601 rs3090765 1 127691908 T G G 4 rs3022839 1 137257629 G A GA 823 rs3713224 1 154052109 A T AT 402 rs3022851 1 171617363 T C 403 rs3022871 1 174794383 C C 900 rs3716435 1 175899264 C T CT 824 rs3704926 1 184806157 CT T CT 722 rs3681847 2 11984102 T C CT 723 rs4137557 2 11984314 T C CT 724 rs3664805 2 28174259 C T CT 10 rs3022883 2 37572959 G G G 726 rs3667376 2 45472806 C T CT 727 rs3681675 2 45990821 T C CT 404 rs3089489 2 51552294 A A 405 rs3022887 2 68883673 G A GA 901 rs4137272 2 79034808 G G GA 406 rs3022888 2 84403279 A A 825 rs3669855 2 93586287 G G GA 902 rs3670817 2 94071001 C C CT 728 rs3665286 2 94807655 G G G 729 rs3692288 2 95029801 C C T 826 rs3663534 2 106743641 T C CT 827 rs3090608 2 117816539 C T CT 18 rs3022895 2 119273525 G GA GA 828 rs3706262 2 119748062 A G GA 829 rs3660910 2 133743791 T C CT 408 rs3022910 2 144973062 C T 830 rs3714631 2 151657817 C T CT 409 rs3089031 2 155762997 C 410 rs3022939 2 159674008 C 730 rs3666331 2 167851555 A G GA 731 rs3674631 2 168354454 T C CT 732 rs3706063 3 12332500 T C CT 733 rs3672323 3 12812319 A C CA 831 rs3676476 3 28110853 T C CT 411 rs3022953 3 37599994 C 832 rs3669022 3 42396732 C T CT 833 rs3726717 3 52860548 A G GA 20 rs3022960 3 65418042 G A GA 834 rs3689513 3 78626985 C T CT 835 rs3669413 3 93426396 C A CA 836 rs3722681 3 109900215 T CT 903 rs3672398 3 112220863 T AT AT 212 rs3022965 3 114011713 G 904 rs3715748 3 119894667 A GA GA 837 rs3691246 3 122466065 G GA 734 rs3714750 3 126334813 T A A 412 rs3089257 3 130719076 T C CT 413 rs3090381 3 140722145 A G GA 736 rs3656469 3 143498147 CT C CT 737 rs3706436 3 143786333 GT TA GTA 471 rs3022975 4 8073146 C T CT 414 rs3091112 4 8073150 GA GA 738 rs3711350 4 26559488 A G GA 739 rs3684156 4 27375172 T C CTA 838 rs3665192 4 39271744 T G 213 rs3088670 4 47484118 T C CT 839 rs3667625 4 54215713 A G GA 840 rs3694308 4 62440427 C A CA 415 rs3022979 4 75874108 C A CA 841 rs3707288 4 85885441 C A CA 709 rs3713394 4 86423755 T T CT 707 rs4135993 4 88399014 G G GA 708 rs3714181 4 88415879 T T CT 705 rs3680265 4 88774811 C C G 706 rs3696308 4 88775106 T T T 711 rs3686204 4 89249504 C C CT 710 rs3659287 4 89742501 C C CT 416 rs3090804 4 92499971 C T 478 rs3022989 4 99679747 G A GA 417 rs3088455 4 99679889 C C 842 rs3658845 4 115459278 C T T 905 rs3678308 4 120241830 G A A 31 rs3089514 4 123686862 A G A 419 rs3091114 4 129017764 C T T 740 rs3666032 4 131409009 A G GA 741 rs3696551 4 131618149 C T CT 906 rs3706432 4 132738971 A G GA 420 rs3023011 4 133639515 G GA 218 rs3023026 4 140634063 G A GA 32 rs3023037 5 23743495 T C CT 742 rs3722968 5 25064528 C T CT 743 rs3670794 5 25068508 AT GTA 421 rs3023040 5 32943077 T CT 907 rs3714665 5 37385820 G GA GA 422 rs3023045 5 50863147 T G GT 744 rs3659745 5 51041036 AT G 745 rs3713298 5 51194142 T G GT 746 rs3657668 5 66812086 T C CT 747 rs3672190 5 66812453 G A A 843 rs3657720 5 80572730 G A GA 844 rs3691937 5 101074114 C T CT 37 rs3023051 5 117479189 G A GA 33 rs3023057 5 125963636 C T CT 748 rs3685067 5 132834587 T C CT 749 rs3664890 5 135190462 G G G 424 rs3023060 5 142063375 C T CTG 475 rs3023062 5 145124169 A A 425 rs3088741 5 145124441 A A 426 rs3090833 6 18727440 C T CT 845 rs3023067 6 35949474 A T TA 750 rs3709317 6 49704741 G A GA 751 rs3716528 6 49740743 A T A 752 rs3659328 6 63635493 T C CA 846 rs3706583 6 63641367 T T 753 rs3707041 6 63641402 A A AT 329 rs3090936 6 66965136 C 848 rs3152183 6 82717476 A A 791 rs3704682 6 91992606 C C CA 792 rs3690102 6 100835585 T T CT 793 rs4137475 6 104904630 G G GA 847 rs3023083 6 106639546 A A 430 rs3090025 6 109876793 A A A 754 rs3677586 6 110852971 G G GA 755 rs3707407 6 111404795 A A GA 849 rs3023092 6 127839602 G A A 40 rs3089737 6 145414611 T TA 850 rs3090435 6 146590775 A G GA 45 rs3023116 7 33342591 CT CT 606 rs3023117 7 34260777 G A GA 607 rs3023123 7 45323985 C C C 48 rs3023134 7 57286126 A A A 852 rs3711840 7 63737541 A CA 608 rs3023129 7 65460296 A A A 908 rs3668498 7 69445393 A G GA 853 rs3704354 7 74666526 GC GT 46 rs3090876 7 82744531 T C CT 854 rs3710192 7 85574011 T A AT 435 rs3090731 7 96549948 A G GA 480 rs3023154 7 113870672 T C CT 436 rs3089474 7 116219037 T G GT 756 rs3688884 7 124892578 C T CT 757 rs3682376 7 125149486 T C CT 438 rs3089174 7 128497838 C T CT 339 rs3023161 7 129084304 T C CT 758 rs3726791 7 139774847 A T A 759 rs3717254 7 140610603 C C C 820 rs3708255 7 143881399 T C CT 896 rs3023174 8 7812735 G A GA 118 rs3023175 8 11995469 G A GA 760 rs3700240 8 27809735 GA GA G 761 rs3665023 8 27811369 T C CT 439 rs3023183 8 35896398 A G GA 54 rs3088450 8 52979846 A CT 762 rs3679837 8 60174269 A G GA 763 rs3686956 8 60210119 G A GA 610 rs3089230 8 71012852 C G G 897 rs3089636 8 71659444 C T T 609 rs3090460 8 74676357 A G G 51 rs3023194 8 91510521 T C CT 764 rs3711535 8 100556883 G A GA 765 rs3672284 8 100765637 GC GC GC 766 rs3724779 8 117005581 A C CA 767 rs3672332 8 117653778 T C CT 819 rs3696893 8 130456895 A G GA 768 rs3716232 9 16092806 C T CT 769 rs3694785 9 16382724 T C CT 876 rs3654109 9 32471104 A A GA 58 rs3023205 9 33547610 GC GC 877 rs3672091 9 44035088 GA A GA 474 rs3023212 9 44086543 TA T GTA 878 rs3671494 9 59461009 GA A GA 611 rs3023215 9 65973933 A G 441 rs3023216 9 71094035 C T CT 442 rs3023225 9 78134045 A T A 879 rs3695889 9 86124796 C G GC 770 rs4137954 9 98214467 A G GA 57 rs3090474 9 98298858 T G T 771 rs3688878 9 101293845 CT C CT 473 rs3089531 9 101742329 CT C CT 302 rs3023227 9 101742532 C C 780 rs3707022 9 108156940 GA A A 781 rs3670181 9 111379601 T T AT 782 rs3682508 9 112519550 CA A CA 783 rs3657074 9 118619658 C C C 784 rs3700226 9 123845575 G C CG 445 rs3089912 10 20686036 C T CT 446 rs3090586 10 24286974 C T CT 772 rs3704164 10 40602386 C T CT 773 rs3705210 10 40612186 G G G 774 rs3696307 10 53431464 C G G 912 rs3702150 10 68557908 C T CT 777 rs3656551 10 81767829 A G GA 447 rs3090761 10 88548187 T T T 477 rs3090759 10 88548420 CT C CT 64 rs3089366 10 89110543 A A A 818 rs3722942 10 93074375 A G GA 314 rs3089906 10 97678673 G G 778 rs3706590 10 107344215 A G GA 779 rs3660209 10 107394384 T C CT 72 rs3023249 11 11070846 C CA 460 rs3023251 11 20853004 C T CT 909 rs3721297 11 21792473 A C CA 201 rs3023256 11 44299873 G A GA 461 rs3023258 11 54018188 C A CA 73 rs3088940 11 56359390 G A 67 rs3023265 11 57576910 T C CT 799 rs3719410 11 67582900 C T CT 798 rs3719895 11 68481453 G A A 703 rs3709439 11 69031064 G T CT 797 rs3696966 11 69143793 G A GA 700 rs3659426 11 69692061 G A GA 701 rs3681957 11 69748839 C T CT 704 rs3707772 11 70267930 C T CT 68 rs3023278 11 71984385 T C G 910 rs3665064 11 82679842 C T CT 69 rs3089065 11 92981099 A GC 463 rs3088501 11 94146845 C T CT 71 rs3023315 11 99324816 C G GT 911 rs3697014 11 112103534 G A GA 870 rs3725545 12 16183804 CA A CA 871 rs3686668 12 27493470 GA G GA 872 rs3724341 12 42468236 A A AT 483 rs3088822 12 53089104 A G GA 873 rs3690309 12 57606163 G G GA 75 rs3023347 12 57628517 C A CA 874 rs3700688 12 72297329 C C 855 rs3703108 12 74842221 A A 815 rs3662694 12 76915409 G G GA 875 rs3700106 12 77791482 G G GA 856 rs3697769 12 89115007 G T GT 465 rs3088800 12 99163402 T G GT 816 rs3726591 12 105250597 C T CT 305 rs3023378 12 113750831 T 466 rs3023377 12 113751107 G T G 81 rs3023379 13 18303792 G A GA 892 rs3717068 13 30844331 G A GA 79 rs3089102 13 44412430 C A CA 484 rs3089436 13 60912145 T G GT 893 rs3688361 13 80402371 G C GC 80 rs3023386 13 96657145 A G GA 813 rs3144879 13 101262056 C CT CT 485 rs3090063 13 104292074 A T AT 313 rs3023394 13 114216919 A A A 309 rs3023392 13 114837305 A A 814 rs3708958 13 117780654 T C CT 785 rs3723026 14 7734219 A T A 786 rs3720966 14 10727081 G A GA 787 rs3712403 14 13327515 T A AT 788 rs3713871 14 24399404 C A CA 809 rs3724533 14 27108284 A G GA 789 rs3713838 14 27644295 A C CA 790 rs3685393 14 30183496 C T CT 811 rs3023409 14 48748605 A G GA 89 rs3023408 14 50023498 TA CA 321 rs3021908 14 53097465 T C CT 86 rs3089070 14 67715311 T G GT 857 rs3705482 14 82835317 G A GA 87 rs3088599 14 84734491 T C 858 rs3674616 14 98406482 T C CT 88 rs3090773 14 105296240 G A GA 812 rs3657504 14 118523326 C T CT 91 rs3023415 15 10765192 A G G 894 rs3088491 15 12374865 T C CT 492 rs3088634 15 19069275 T C CT 895 rs3662097 15 20153137 T G GT 96 rs3023416 15 31860737 C T CT 808 rs3669262 15 40601825 T G GT 92 rs3088488 15 42420068 G A A 859 rs3658370 15 51501719 G A GA 860 rs3701351 15 61455192 G A GA 93 rs3088506 15 71350770 C A 486 rs3090719 15 74403438 C A CA 94 rs3088710 15 86619624 T T CT 913 rs3677860 15 87544783 A G GA 810 rs3717898 15 95702188 C CT 101 rs3090912 16 6099904 T G GT 806 rs3667072 16 16273453 A A A 332 rs3089488 16 18511345 A A 880 rs3695744 16 27296228 A A A 881 rs3718034 16 35665619 CT C CT 487 rs3089787 16 38850538 CT T CT 807 rs3663711 16 39824346 G G G 488 rs3090260 16 45609646 A G GA 448 rs3023243 16 58212994 T C CT 794 rs3654982 16 63555294 T A AT 795 rs3719654 16 70347393 CT C CT 489 rs3090908 16 72217393 A G GA 796 rs3695101 16 76223854 GA G GA 490 rs3090645 16 85879752 A T TA 455 rs3023436 16 88046993 T A TA 304 rs3023441 16 97552996 A A 882 rs3674239 17 12357534 A G GA 104 rs3090500 17 25071190 A CG CA 884 rs3708501 17 30648071 G CA GA 43 rs3023110 17 32989898 T C CT 491 rs3088914 17 36695497 A C A 105 rs3023454 17 36882223 C T T 452 rs3023449 17 42120599 A T TA 862 rs3700023 17 52981854 G A GA 453 rs3023456 17 67310979 G T GT 202 rs3022791 17 71914187 A A A 863 rs3712928 17 78656292 G G 805 rs3687592 17 81485883 A A GA 885 rs3710028 17 83573655 T - GT 312 rs3089323 17 83844781 A A 864 rs3668190 17 91677947 C C 605 rs3089544 18 5088109 G A GA 886 rs3696042 18 8912956 CT T CT 865 rs3655356 18 10241322 GA G 887 rs3695261 18 21869777 C T CT 450 rs3023463 18 29943155 GT GT GT 803 rs3683689 18 33935868 A G GA 316 rs3089327 18 41908447 C C 888 rs3657200 18 50109691 T C CT 109 rs3023468 18 63609445 G C GC 866 rs3657018 18 78455291 C CT 889 rs3668347 18 80932716 T GT GT 804 rs3718427 18 88768623 G G GA 867 rs3665935 18 89757219 C C 891 rs3669192 19 16913884 C T CT 800 rs3671678 19 17422426 G A GA 890 rs3658201 19 20407978 C T CT 111 rs3023481 19 20484928 GA G 868 rs3710059 19 30660820 G C GC 336 rs3090951 19 44127717 C C 801 rs3726430 19 44193090 G T GT 869 rs3713040 19 55813052 GA 449 rs3023498 19 60821445 G GA GA 802 rs3685993 19 60910395 C T T 602 rs3022803 multiple C A 7 rs3022821 multiple G GA GA 401 rs3022823 multiple G G 725 rs4139354 multiple GA G GA 418 rs3022994 multiple T A CA 432 rs3023096 multiple T C C 851 rs3723894 multiple A T AT 776 rs3673999 multiple G G G 817 rs3664582 multiple G A GA 883 rs3668662 multiple G A GA 914 unmapped unmapped A G GA 915 unmapped unmapped A A A 916 unmapped unmapped T T CT 917 unmapped unmapped A A GA 918 unmapped unmapped - T 919 unmapped unmapped C C C 920 unmapped unmapped T C CT 921 unmapped unmapped C - CA 922 unmapped unmapped T A AT 923 unmapped unmapped T C C 604 rs3088804 x 99074052 A GA GA Supplemental Table 2: List of PCR and SBE primers used for validation. 160 Supplemental Table 3: PCR multiplexing strategv in ienome-wide screen Primers are same as listed in Supplemental Table 1 167 SBE group PCR grp Assay ID RefSNP ID 412 rs3089257 la-i 405 rs3022887 772 rs3704164 426 rs3090833 471 rs3022975 1A 779 rs3660209 786 rs3720966 830 rs3714631 883 rs3668662 850 rs3090435 1a-6 439 rs3023183 452 rs3023449 823 rs3713224 109 rs3023468 868 rs3710059 1B lb-3 893 rs3688361 1b-4 455 rs3023436 lb-5 879 rs3695889 lb-6 490 rs3090645 lb-7 922 unmapped 461 rs3023258 453 rs3023456 821 rs3704980 600 rs3022802 808 rs3669262 895 rs3662097 415 rs3022979 840 rs3694308 ic-5 422 rs3023045 1c-6 79 rs3089102 ld-1 33 rs3023057 713 rs3684370 1d-2 761 rs3665023 489 rs3090908 803 rs3683689 831 rs3676476 81 rs3023379 892 rs3717068 id-5 757 rs3682376 id-6 478 rs3022989 741 rs3696551 700 rs3659426 le-2 770 rs4137954 756 rs3688884 790 rs3685393 1E 51 rs3023194 762 rs3679837 701 rs3681957 le-5 911 rs3697014 812 rs3657504 le-6 913 rs3677860 1f-1 797 rs3696966 839 rs3667625 f-2 727 rs3681675 857 rs3705482 1F 1i-3 746 rs3657668 809 rs3724533 f-4 882 rs3674239 768 rs3716232 if-5 769 rs3694785 if-6 898 rs3667466 833 rs3726717 843 rs3657720 844 rs3691937 19-2 834 rs3689513 827 rs3090608 817 rs3664582 jq-4 912 rs3702150 1q-5 763 rs3686956 1Q-6 742 rs3722968 1g-7 814 rs3708958 lh-1 766 rs3724779 465 rs3088800 788 rs3713871 436 rs3089474 1H lh-3 856 rs3697769 75 rs3023347 715 rs3704392 835 rs3669413 lh-5 86 rs3089070 group PCR grp Assay ID RefSNP ID 4 rs3022839 2a-I 96 rs3023416 446 rs3090586 488 rs3090260 32 rs3023037 2A 445 rs3089912 721 rs3664018 859 rs3658370 2a-5 218 rs3023026 2a-6 820 rs3708255 2a-7 80 rs3023386 818 rs3722942 822 rs3685188 722 rs3681847 732 rs3706063 2B 339 rs3023161 811 rs3023409 858 rs3674616 118 rs3023175 2b-5 764 rs3711535 2c-1 20 rs3022960 483 rs3088822 448 rs3023243 860 rs3701351 2c-3 67 rs3023265 2C 321 rs3021908 724 rs3664805 826 rs3663534 2c-5 800 rs3671678 2c-6 201 rs3023256 2c-7 890 rs3658201 854 rs3710192 851 rs3723894 2D 2d-2 787 rs3712403 2d-3 794 rs3654982 2d-4 845 rs3023067 726 rs3667376 731 rs3674631 767 rs3672332 435 rs3090731 2e-3 862 rs3700023 2E 832 rs3669022 2e-4 606 rs3023117 829 rs3660910 2e-5 750 rs3709317 2e-6 740 rs3666032 2e-7 738 rs3711350 2f-1 486 rs3090719 745 rs3713298 2f-2 801 rs3726430 101 rs3090912 2F 2f-3 484 rs3089436 733 rs3672323 2f-4 841 rs3707288 789 rs3713838 2f-5 909 rs3721297 2g-1 714 rs3662163 816 rs3726591 2g-2 463 rs3088501 723 rs4137557 2g-3 413 rs3090381 2G 799 rs3719410 2o-4 37 rs3023051 2Q-5 730 rs3666331 2Q-6 748 rs3685067 2a-7 828 rs3706262 2g-8 900 r3716435 Chapter 4 Final Discussion 169 The sequencing of the human and mouse genomes has spurred a growing interest in analyzing mouse models of human cancer using genomic techniques. Comparative genomic studies on mouse vs. human tumors can be valuable in two major ways: 1) in validating mouse models through an assessment of their degree of genetic resemblance to human disease and 2) in identifying genes and/or gene sets that are common to mouse and human tumorigenesis. As described in Chapter 1, many analytic tools have emerged in recent years for human genome mining. Some of these tools have been translated to murine versions. The work described in Chapters 2 and 3 involved application testing of two newly translated mouse whole-genome analytic techniques: ROMA and SNaPshot (Applied Biosystems) SNP genotyping. With ROMA, a high-resolution view of copy number alterations in the tumor genome was possible. By SNaPshot (Applied Biosystems), low-density SNP-based draft maps indicative of LOH were obtainable. The applications were tested mainly on a murine model of lung cancer. Several recurrent chromosomal copy number gains and losses, as well as chromosomal LOH, were observed in this Kras-driven lung cancer model. Discussed in this concluding chapter are my views on the technology applied and the biology implied by the body of work. On Technology A wide range of copy number alterations in primary mouse tumors has been previously documented, including single copy gain or loss of entire chromosomes, partial gain or loss of a chromosome, high-amplitude focal amplifications, to low-level small deletions (Hodgson et al. 2001; You et al. 2002; Hackett et al. 2003; O'Hagan et al. 2003). To cover this broad spectrum, a genome-wide high-resolution CGH tool would be invaluable 170 cover this broad spectrum, a genome-wide high-resolution CGH tool would be invaluable for mouse cancer DNA anlayses. Among the many CGH platforms for genome-wide copy number analysis studies in human, ROMA has one of the highest resolving power averaging at 30kb (Lucito et al. 2003). Tested in Chapter 2, the mouse version of ROMA also appeared to be a powerful tool, allowing us to detect a focal high-amplitude (>4.6fold, assessed by Southern) N-Myc amplification in retinoblastomas of a Rb/p130 DKO model, as well as numerous whole-chromosomal gains and losses in the same retinoblastoma sample set and in the lung tumors driven by a Kras mutation. The mouse ROMA platform would be a useful tool to characterize mouse cancer- associated genetic alterations. Genome-wide copy number data from mouse tumors may potentially be used in several ways: 1) clear focal changes can pinpoint individual candidates such as N-Myc in our case; 2) clustering of data can identify copy number alteration patterns that define particular tumor subtypes in the mouse (O'Hagan et al. 2003); 3) analogous to the use of gene-set enrichment analysis (GSEA) approach to extract human tumor gene-expression signature from mouse tumor gene-expression pattern (Sweet-Cordero et al. 2005), one can imagine being able to extract human copy number change signatures by comparing with mouse data that are confounded with less genetic noise. There is also an abundant choice of SNP genotyping methodologies, although few have been applied for whole-genome analysis in the mouse. We attempted to contribute the following ways: 1) validate 358 published SNPs individually for polymorphism in 171 129S4/svJae vs. C57BL/6J strains, 2) worked out a protocol to genotype a panel of 147 markers in the mouse using SNaPshot (Applied Biosystems) genotyping system. The method uses a standard DNA sequencing machine to resolve SBE products of SNPs, which should be easily adaptable for use by interested labs, 3) tested the application of the assay as a mapping tool of LOH in mouse tumors. Although genome-wide SNP mapping of LOH is not new in human cancer genomics, the same concept has not been performed in mice probably due to the lack of accessible SNP genotyping protocols. Despite needing further improvements on overall genotyping accuracy and efficiency, our protocol was usable in analyzing LOH patterns in mouse lung tumors. We correctly detected the loss of wild-type p53 allele in a subset of samples, suggesting the concept of performing SNP-based LOH detection in Fl mice is going to be viable. When other groups concentrated their efforts in the genotyping a few markers in multiplex with the same type of assay, they were able to drive the level accuracy higher (to almost 100%) (Makridakis and Reichardt 2001; Norton et al. 2002). This suggests there is much room for us to optimize each of our SBE reaction to improve on accuracy. In addition, sensitivity of the method needs to be tested on more markers in a dilution experiment using heterozygous vs. different homozygous DNA. Then, in the future, increasing the density of SNPs in the panel would allow higher resolution. To date, over 6 million mouse SNPs across different strains have been referenced in the NCBI dbSNP public database. The challenge in the future SNP-based LOH mapping clearly is not the lack of markers, but is finding an efficient and cost-effective method to 172 genotype SNPs at a density an experiment requires, which may change as the project proceeds. Flexibility in genotyping the particular SNPs of choice will also be required. SNaPshot is a multiplexable system, which we used to genotype 5-11 SNPs in each SBE reaction and capillary run. No investment in specialized instrument is required. However, I think the biggest charm of the system is the flexibility in choosing the markers used in genotyping. The assay can be performed either individually or through mixing-and- matching and adding-and-removing of a few markers in a multiplex fashion. On Biology The applications of ROMA and SNaPshot genotyping were tested mainly on murine models of lung cancer that were initiated by a Kras mutation. Overall, copy number alterations or LOH seem to start appearing in genomes of tumors that are histologically graded 2-3, when they started to exhibit pleomorphic nuclei, prominent nucleoli, and nuclear molding. The correlation of detectable genetic alterations with higher histological grades fits the clonal evolution model of tumor progression. However, lower grade tumors, because of their smaller sizes, are more difficult to dissect out cleanly. We attempted to minimize this confounding factor by choosing samples with little normal tissues, as judged histologically. Laser capture microdissection could be used in the future as another way to circumvent this contamination possibility. Based on histopathology, lung cancer is grouped into two broad categories: non-small- cell lung cancer (NSCLC) and small-cell lung cancer (SCLC). About 80% of all cases are NSCLC, which are further divided into the following subtypes: adenocarcinoma, squamous cell carcinoma, bronchioalveolar carcinoma, and large-cell carcinoma. The 173 NSCLC subtypes behave similarly as a distinct group to SCLC in its therapeutic response (Minna et al. 2002). Genetically, SCLC and NSCLC also form two distinct groups. Genetic alterations manifest as large-scale chromosomal gains and losses, focal amplifications and deletions, or nucleotide changes. Multiple mutations are observed in lung cancer samples. Epigenetic changes such as methylation also occur commonly. Among the oncogenes, Ras mutations are detected in 20-50% of NSCLC and <1 % of SCLC (Slebos et al. 1990). Kras mutations consist of 90% of all Ras mutations and are almost exclusively found in adenocarcinomas (Slebos et al. 1990). Myc amplification or overexpression is seen more frequently in SCLC (20-35%) than in NSCLC (5-20%) (Richardson and Johnson 1993). Bcl2 is overexpressed in 75-95% of SCLC and 10-30% ofNSCLC (Pezzella et al. 1993; Kaiser et al. 1996). EGFR mutation is found in 20% of NSCLC and is associated with non-smoking related adenocarcinomas (Zochbauer-Muller et al. 2002). Among the tumor suppressor genes, p53 is found deleted or mutated in >50% of NSCLC and SCLC (Takahashi et al. 1989; Toyooka et al. 2003). p16I NK4 a hypermethylation or deletion occurs frequently; -30-50% NSCLC does not express p16 (Minna et al. 2002). The alternative reading frame product in the same locus pl 4 A l F is also inactivated in ~20% of NSCLC and -65% of SCLC (Nicholson et al. 2001). In addition to gene-specific lesions, cytogenetic, CGH, and LOH studies have revealed numerous large-scale chromosomal aberrations, suggesting more oncogenes and tumor suppressors remain to be discovered. LOH in chromosome 3p is the most prominent event, found in almost 100% of SCLC and >90% of NSCLC (Wistuba et al. 2001; Zabarovsky et al. 2002). Other frequent changes 174 include gains of lq, 3q, 5p, 8q, 1 1q, 12q, 19q, and losses of 4q, 10q in NSCLC, and gains of 3q, 5p, 8q, 19q and losses of 4q, 5q, and 13q in SCLC (Balsara and Testa 2002). The tumors developed in the Kras model histologically resemble human NSCLC (Jackson et al. 2005). Recurrent chromosomal copy number changes include +6, +12, +19, + 3, + 16, -9, -11 in 10/28 lung tumors. Each of these chromosomes is comprised of multiple syntenic human chromosomal regions. Genetic alterations in many of these regions have been observed in human lung cancers as summarized in Table 1. Chromosome 6 gain was seen in 80% of tumors we tested. Mouse chromosome 6 harbors the Kras gene and contains a region in synteny to a human 3q segment that have been implicated in lung cancer. Also of interest are the recurrent losses of chromosomes 9 and 11, with each reduced in copy in two tumors. A distal part of chromosome 9 is syntenic to human chromosome 3p21-22. Loss of human chromosome 3p is the most common event observed in lung cancer (Zabarovsky et al. 2002). In particular 3p21 loss is observed as an early event, which can be detected in the pre-malignant epithelium of smokers (Zabarovsky et al. 2002). As for chromosome 11, its distal arm has syntenic conservation to the entire human chromosome 17, where p53 resides. It is reassuring that the lung tumors developed in our mouse model contain certain regions that are syntenic to the regions altered in some human lung cancers. However, the genetic lesions observed in our mouse model span entire chromosomes, thus the subchromosomal regions critical to tumorigenesis in our model are unclear. 175 Table 1: Summary of copy number changes observed in mouse mouse model and their corresponding syntenic regions in human 176 Ln 0 0 N 0 m 'U N Ol '-4 'U I- N 00 r4 ch Nh Eo o . Enw 'UO LA>- NNVNNMM %D N '-I r N 0 0 N 0 U) E a .j 0 0h '- En 0\ I- 0 CL. 0 '- .0 N o E 0 LA o o N r- LIO lO "0 .-4 00. '-4 a r4Oi C7 a r (7 0L Goc 't -4 '-4 N. 0C00. '-4N Iro 0 7-I N 0e O L oN 0S N. 8D 0*.o0.aa0 I ol m w w A -1 V1 1. No LA EU (N o N 0 a a 0m , NU3 a. O -- N .j. 0' 0F E,.4N LAn oO\ C-4 NI- N V- N LA 7- '-4 Our model is driven by a Kras point mutation. In human, Kras point mutations occur at a higher frequency in smokers than in nonsmokers; one study reported the numbers to be 30% vs. 7% (Westra et al. 1993). Smoking is the biggest risk factor for lung cancer. Differences in the spectra of genetic alterations in smoking vs. non-smoking related lung cancer have been observed (Hirao et al. 2001; Sanchez-Cespedes et al. 2001; Wong et al. 2002; Sy et al. 2003; Wong et al. 2003). Cigarette smoke consists of multiple carcinogens including benzo(a)pyrene (BAP) and 4-methylnitrosamino- 1-3-pyridal- 1-butanone (NNK), which result in a prevalence of G-to-T and G-A transversions in smoking-related lung cancer (DeMarini 2004). The spectra of mutations differ in smoking vs. non- smoking related lung cancers (Pan et al. 2005). Interestingly, lp, 3p, 5q, 1 lq, and 17q, which share homology with gained or lost in the DNA from our Kras-induced mouse model, have been reported in the cited studies. Despite a certain degree of syntenic conservation of lung cancer genetic changes could be inferred, one apparent difference is that the predominance of whole-chromosomal changes in the lung tumors from mice. Such observation seems to be common among spontaneous tumors of different genetically engineered mouse models. On the other hand, multiple number and kinds of mutations, including translocations and sub-chromosomal lesions, are frequently seen in human carcinomas. At the cellular level, mouse cells differ from human cells by having longer telomeres, which might have protected mouse chromosomes from breakage events through break-fusion-bridge cycles caused by damaged telomere (O'Hagan et al. 2002). Indeed, tumors in mTerc-/- mice with deficient telomerase had a larger and more human-like variety of genetic changes (O'Hagan et al. 178 2002). Perhaps making mice containing both mTerc deletion and Kras mutation in the lung will lead more human-like focal genetic lesions in the lung tumors. Another point of interest is The set of recurrent chromosomal copy number changes we observed (in order of prevelance: +6, +12, +19, + 3, + 16, -9, -11) in 10/28 lung tumors by ROMA were different from a previously reported set of 16/59 samples from a closely related lung cancer model (+6, +8, +16, +19, -4, -11, -17) analyzed using BAC array CGH (Sweet-Cordero et al. 2006). Chromosome 6 and 19 gains and 11 losses are the only commonalities. While chromosome 12 gain was seen in 5/10 tumors with changes in our study, it was observed in 1 clear case in the other analysis. The discrepancy between the 2 studies raises a few questions. The discrepancy could be simply due to small sampling sizes and the random nature of mutational process. Formally, a pilot study can be initially performed to generate a statistical estimate of the amount of samples required in the main study. However, in practice, this is not often done, largely due to cost constraints and in the case of a human study, also because of the difficulty in obtaining patient samples. Meta-analyses may help to make sense of studies done in different times or labs (Hoglund et al. 2004). Alternatively, this discrepancy may reflect a true difference due to the different activation timing/mechanism between the conditional KrasLSLG 12 D and latent KrasG 12 DLA 2 alleles. Our study was performed using tumors materials obtained from KrasLSLG1 2 D;p 5 3+/+, KrasLSLG12D; p53+/- and KrasLSLG12D; p53 R270H/+ mice while the other study used mice containing a KrasG1 2 DLA 2 allele. However, in our study, tumors from mice of all 3 genotypes that were comparable 179 in histology were also comparable in the types and amount of genetic changes. The two Kras G12 D alleles lead to lung tumors with indistinguishable histology, despite the different timing/mechanism of activation. This would be reminiscent of the RIP-Rag islet cell cancer model, in which the timing of T antigen activation changes the pattern of copy number alterations without affecting tumor histology/progression (Hager et al. 2004). In addition, genetic background may contribute to the discrepancy by affecting copy number changes. One difference in the two experimental setups was the background of mice used: our analysis was on an inbred 129S4 strain while the other was done on F1 B6x129S4 tumors. A difference in tumor incidence in B6 vs. 129S4 strains has been observed in KrasG 12DLA2 mice, suggesting there are strain-specific modifying factors for tumor multiplicity. Genetic background effect on copy number spectrum has been reported in another mouse cancer model (Hager et al. 2004). The work of this thesis began the use of two powerful techniques: ROMA and SNP genotyping, to study mouse tumors. Our analysis of whole tumors has revealed a few stable DNA changes that might harvest critical genes for tumorigenesis. In the case of retinoblastoma, the candidate region was narrowed down to a single gene --N-Myc. The functional role of N-Myc in retinoblastoma remains to be characterized. In the case of lung cancer, an analysis of higher grade tumors will be worthwhile to attempt. Among the many unanswered questions surrounding the genetic alterations in cancer, whether genomic instability occurs is one of them. Genomic instability is a highly debated concept that was proposed to explain the origin of the many genetic alterations often seen in a cancer. Genomic instability describes an increased rate of genetic alterations in 180 cancer cells. Most of the described experiments were performed on DNA extracted from primary tumors. Such analyses in fact reflect the stable genetic changes that have undergone selection in the particular tumor. Measurement of genomic instability requires knowing the amount of changes that occur over a known number of cell division. Metaphase analysis of dividing tumor cells is probably the best measurement that can be done in tumors samples and can be included in future studies. Genomic analysis of tumorigenesis provides insight into the evolution of tumor cells. With the many new technologies, the coming years in cancer genome analytics will likely be full of excitements. References Balsara, B.R. and Testa, J.R. 2002. Chromosomal imbalances in human lung cancer. Oncogene 21(45): 6877-6883. DeMarini, D.M. 2004. Genotoxicity of tobacco smoke and tobacco smoke condensate: a review. Mutat Res 567(2-3): 447-474. Hackett, C.S., Hodgson, J.G., Law, M.E., Fridlyand, J., Osoegawa, K., de Jong, P.J., Nowak, N.J., Pinkel, D., Albertson, D.G., Jain, A., Jenkins, R., Gray, J.W., and Weiss, W.A. 2003. Genome-wide array CGH analysis of murine neuroblastoma reveals distinct genomic aberrations which parallel those in human tumors. Cancer Res 63(17): 5266-5273. Hager, J.H., Hodgson, J.G., Fridlyand, J., Hariono, S., Gray, J.W., and Hanahan, D. 2004. Oncogene expression and genetic background influence the frequency of DNA copy number abnormalities in mouse pancreatic islet cell carcinomas. Cancer Res 64(7): 2406-2410. Hirao, T., Nelson, H.H., Ashok, T.D., Wain, J.C., Mark, E.J., Christiani, D.C., Wiencke, J.K., and Kelsey, K.T. 2001. Tobacco smoke-induced DNA damage and an early age of smoking initiation induce chromosome loss at 3p21 in lung cancer. Cancer Res 61(2): 612-615. 181 Hodgson, G., Hager, J.H., Volik, S., Hariono, S., Wernick, M., Moore, D., Nowak, N., Albertson, D.G., Pinkel, D., Collins, C., Hanahan, D., and Gray, J.W. 2001. Genome scanning with array CGH delineates regional alterations in mouse islet carcinomas. Nat Genet 29(4): 459-464. Hoglund, M., Gisselsson, D., Hansen, G.B., and Mitelman, F. 2004. Statistical dissection of cytogenetic patterns in lung cancer reveals multiple modes of karyotypic evolution independent of histological classification. Cancer Genet Cytogenet 154(2): 99-109. Jackson, E.L., Olive, K.P., Tuveson, D.A., Bronson, R., Crowley, D., Brown, M., and Jacks, T. 2005. The differential effects of mutant p53 alleles on advanced murine lung cancer. Cancer Res 65(22): 10280-10288. Kaiser, U., Schilli, M., Haag, U., Neumann, K., Kreipe, H., Kogan, E., and Havemann, K. 1996. Expression of bcl-2--protein in small cell lung cancer. Lung Cancer 15(1): 31-40. Lucito, R., Healy, J., Alexander, J., Reiner, A., Esposito, D., Chi, M., Rodgers, L., Brady, A., Sebat, J., Troge, J., West, J.A., Rostan, S., Nguyen, K.C., Powers, S., Ye, K.Q., Olshen, A., Venkatraman, E., Norton, L., and Wigler, M. 2003. Representational oligonucleotide microarray analysis: a high-resolution method to detect genome copy number variation. Genome Res 13(10): 2291-2305. Makridakis, N.M. and Reichardt, J.K. 2001. Multiplex automated primer extension analysis: simultaneous genotyping of several polymorphisms. Biotechniques 31(6): 1374-1380. Minna, J.D., Fong, K., Zochbauer-Muller, S., and Gazdar, A.F. 2002. Molecular pathogenesis of lung cancer and potential translational applications. Cancer J 8 Suppl 1: S41-46. Nicholson, S.A., Okby, N.T., Khan, M.A., Welsh, J.A., McMenamin, M.G., Travis, W.D., Jett, J.R., Tazelaar, H.D., Trastek, V., Pairolero, P.C., Corn, P.G., Herman, J.G., Liotta, L.A., Caporaso, N.E., and Harris, C.C. 2001. Alterations ofpl4ARF, p53, and p73 genes involved in the E2F-1-mediated apoptotic pathways in non- small cell lung carcinoma. Cancer Res 61(14): 5636-5643. Norton, N., Williams, N.M., Williams, H.J., Spurlock, G., Kirov, G., Morris, D.W., Hoogendoorn, B., Owen, M.J., and O'Donovan, M.C. 2002. Universal, robust, 182 highly quantitative SNP allele frequency measurement in DNA pools. Hum Genet 110(5): 471-478. O'Hagan, R.C., Brennan, C.W., Strahs, A., Zhang, X., Kannan, K., Donovan, M., Cauwels, C., Sharpless, N.E., Wong, W.H., and Chin, L. 2003. Array comparative genome hybridization for tumor classification and gene discovery in mouse models of malignant melanoma. Cancer Res 63(17): 5352-5356. O'Hagan, R.C., Chang, S., Maser, R.S., Mohan, R., Artandi, S.E., Chin, L., and DePinho, R.A. 2002. Telomere dysfunction provokes regional amplification and deletion in cancer genomes. Cancer Cell 2(2): 149-155. Pan, H., Califano, J., Ponte, J.F., Russo, A.L., Cheng, K.H., Thiagalingam, A., Nemani, P., Sidransky, D., and Thiagalingam, S. 2005. Loss of heterozygosity patterns provide fingerprints for genetic heterogeneity in multistep cancer progression of tobacco smoke-induced non-small cell lung cancer. Cancer Res 65(5): 1664-1669. Pezzella, F., Turley, H., Kuzu, I., Tungekar, M.F., Dunnill, M.S., Pierce, C.B., Harris, A., Gatter, K.C., and Mason, D.Y. 1993. bcl-2 protein in non-small-cell lung carcinoma. NEngl JMed 329(10): 690-694. Richardson, G.E. and Johnson, B.E. 1993. The biology of lung cancer. Semin Oncol 20(2): 105-127. Sanchez-Cespedes, M., Ahrendt, S.A., Piantadosi, S., Rosell, R., Monzo, M., Wu, L., Westra, W.H., Yang, S.C., Jen, J., and Sidransky, D. 2001. Chromosomal alterations in lung adenocarcinoma from smokers and nonsmokers. Cancer Res 61(4): 1309-1313. Slebos, R.J., Kibbelaar, R.E., Dalesio, O., Kooistra, A., Stam, J., Meijer, C.J., Wagenaar, S.S., Vanderschueren, R.G., van Zandwijk, N., Mooi, W.J., and et al. 1990. K-ras oncogene activation as a prognostic marker in adenocarcinoma of the lung. N Engl JMed 323(9): 561-565. Sweet-Cordero, A., Mukherjee, S., Subramanian, A., You, H., Roix, J.J., Ladd-Acosta, C., Mesirov, J., Golub, T.R., and Jacks, T. 2005. An oncogenic KRAS2 expression signature identified by cross-species gene-expression analysis. Nat Genet 37(1): 48-55. 183 Sweet-Cordero, A., Tseng, G.C., You, H., Douglass, M., Huey, B., Albertson, D., and Jacks, T. 2006. Comparison of gene expression and DNA copy number changes in a murine model of lung cancer. Genes Chromosomes Cancer 45(4): 338-348. Sy, S.M., Wong, N., Mok, T.S., Tsao, M.S., Lee, T.W., Tse, G., Blackhall, F.H., Johnson, P.J., and Yim, A.P. 2003. Genetic alterations of lung adenocarcinoma in relation to smoking and ethnicity. Lung Cancer 41(1): 91-99. Takahashi, T., Nau, M.M., Chiba, I., Birrer, M.J., Rosenberg, R.K., Vinocour, M., Levitt, M., Pass, H., Gazdar, A.F., and Minna, J.D. 1989. p53: a frequent target for genetic abnormalities in lung cancer. Science 246(4929): 491-494. Toyooka, S., Tsuda, T., and Gazdar, A.F. 2003. The TP53 gene, tobacco exposure, and lung cancer. Hum Mutat 21(3): 229-239. Westra, W.H., Slebos, R.J., Offerhaus, G.J., Goodman, S.N., Evers, S.G., Kensler, T.W., Askin, F.B., Rodenhuis, S., and Hruban, R.H. 1993. K-ras oncogene activation in lung adenocarcinomas from former smokers. Evidence that K-ras mutations are an early and irreversible event in the development of adenocarcinoma of the lung. Cancer 72(2): 432-438. Wistuba, II, Gazdar, A.F., and Minna, J.D. 2001. Molecular genetics of small cell lung carcinoma. Semin Oncol 28(2 Suppl 4): 3-13. Wong, M.P., Fung, L.F., Wang, E., Chow, W.S., Chiu, S.W., Lam, W.K., Ho, K.K., Ma, E.S., Wan, T.S., and Chung, L.P. 2003. Chromosomal aberrations of primary lung adenocarcinomas in nonsmokers. Cancer 97(5): 1263-1270. Wong, M.P., Lam, W.K., Wang, E., Chiu, S.W., Lam, C.L., and Chung, L.P. 2002. Primary adenocarcinomas of the lung in nonsmokers show a distinct pattern of allelic imbalance. Cancer Res 62(15): 4464-4468. You, M.J., Castrillon, D.H., Bastian, B.C., O'Hagan, R.C., Bosenberg, M.W., Parsons, R, Chin, L., and DePinho, R.A. 2002. Genetic analysis of Pten and Ink4a/Arf interactions in the suppression of tumorigenesis in mice. Proc Natl Acad Sci US A 99(3): 1455-1460. Zabarovsky, E.R., Lerman, M.I., and Minna, J.D. 2002. Tumor suppressor genes on chromosome 3p involved in the pathogenesis of lung and other cancers. Oncogene 21(45): 6915-6935. 184