Page 48 - Mouse Molecular Genetics

Full Abstracts
Program number is above title. Author in bold is the presenter.
Next generation RNAi mouse models for drug discovery and toxicology assessment. Prem Premsrirut
Christof Fellmann
Lukas Dow
Johannes Zuber
Gregory Hannon
Scott Lowe
. 1)
Research & Development, Mirimus Inc., Cold Spring Harbor,
NY; 2) Cancer Biology, Memorial Sloan Kettering Cancer Center, New York, NY; 3) Differentiation and Disease, Institute of
Molecular Pathology, Vienna, Austria; 4) Bioinformatics and Genomics, Cold Spring Harbor Laboratory, Cold Spring Harbor,
RNA interference is a powerful tool for studying gene function, however, the reproducible generation of transgenic RNAi mice
remains a significant limitation. One main hurdle is the identification of potent RNAi triggers, or short hairpin RNAs (shRNAs),
that will induce stable and regulated gene silencing. Due to the lack of understanding of the requirements for shRNA biogenesis
and target suppression, many predicted shRNAs fail to efficiently induce gene suppression. We have developed a Sensor
assay that enables the biological identification of effective shRNAs at large scale and show that our assay reliably identifies
potent shRNAs that are surprisingly rare and predominantly missed by existing algorithms. By combining our sensored miR30-
based shRNAs with high efficiency embryonic stem cell targeting, we developed a fast, scalable pipeline for the production of
transgenic RNAi mice. We show that RNAi can cause sufficient knockdown to recapitulate the phenotypes of knockout mice,
particularly in cancer models. More importantly, unlike traditional knockout models, RNAi has the powerful advantage of
reversibility, since the endogenous gene remains intact. Using this system, we generated a number of inducible RNAi transgenic
lines and demonstrate how this approach can identify predicted phenotypes and also unknown functions for well-studied genes.
In addition, through regulated gene silencing we validate several tumor suppressor genes as potential therapeutic targets in T cell
acute lymphoblastic leukemia/lymphoma and lung adenocarcinoma, respectively. This system provides a cost-effective and
scalable platform for the production of RNAi transgenic mice targeting any mammalian gene that will be valuable tools for
performing preclinical target identification, validation and toxicity assessment in vivo.
Constructing functional genetic networks in mammalian cells. Christopher Kemp
Russell Moser, Michael Kao, Chang Xu,
Carla Grandori, Eddie Mendez. Fred Hutchinson Cancer Research Center, Seattle, WA.
The majority of biomedical research using model organisms such as mice is performed on a uniform genetic background.
However, as phenotype is the product of myriad genetic interactions, this approach limits our view of the underlying genetic
basis of phenotype. For example, genome scale analysis of synthetic lethal interactions has not been possible in mammalian cells.
Traditional methods to study such interactions in mice have relied on time consuming breeding experiments and are inefficient at
identifying functional genetic networks. Using such an approach, we identified synthetic lethality between the DNA damage
kinases Atm and DNA-PK (Current Biology 11:191, 2001) and DNA damage-conditional synthetic lethality between DNA-PK
and the tumor suppressor p53 (EMBO Reports, 10:87, 2009). However the lack of functional genetic tools limited our ability to
further dissect these pathways. To overcome this, we have developed a method to identify synthetic lethal interactions on a
genome scale in mammalian cells that utilizes high throughput screening (HTS) with arrayed siRNAs. Using viability as an
endpoint, we performed kinome wide siRNA screens with a series of mouse cells that carry defined mutations in p53 pathway
genes (Atm, DNA-PK, p19Arf and p53) and identified novel synthetic lethal and synthetic sick interactions with each of these
genes. To establish proof of principle that this approach can identify meaningful drug targets for cancer therapy, we focused on
the cell cycle kinase WEE-1 that when knocked down was selectively lethal to both mouse and human p53 mutant cancer cells.
Relative to p53 wild type cells, p53 mutant cancer cells were ~50 fold more sensitive to the WEE-1 inhibitor MK-1775 providing
independent confirmation of our siRNA screen result. This sensitivity of p53 mutant cells to WEE1 inhibition likely results from
a greater dependence on the S/G2 checkpoint to repair DNA and avoid mitotic catastrophe. Oral gavage of MK-1775 inhibited
growth of p53 mutant xenografts and augmented cisplatin response indicating the therapeutic potential of MK-1775 in p53
mutant tumors. In summary, using high throughput screening with well-based siRNAs identified a genome scale menu of
potential drug targets for cancer cells that carry defined oncogenic mutations. More generally, this functional genetic approach
enables, for the first time, the study of genetic interactions on a genome scale in mammalian cells.
Hot or Not?: Leveraging mouse genome diversity to identify hotspots of copy number variants. Kathleen A. Hill
Elizabeth O Locke
Andrea E. Wishart
Susan T. Eitutis
Jenna Butler
Mark Daley
. 1)
Department of Biology and Computer
Science, The University of Western Ontario, London, ON, Canada; 2) Department of Computer Science, The University of
Western Ontario, London, ON, Canada; 3) Department of Biology, The University of Western Ontario, London, ON, Canada.
The Mouse Diversity Genotyping Array (MDGA) currently provides a high resolution means for detection of copy number
variants using both SNP and CNV probe sets. Also, publicly available MDGA data exist for 306 different mouse genetic
backgrounds encompassing feral mice and the broad history and derivation of inbred mouse strains. PennCNV is currently among
the best platforms for microarray data analysis not yet applied to a comprehensive analysis of mouse data. We report the first
application of PennCNV including genomic wave adjustment to publicly available MDGA data representing 351 individuals,
achieving the most comprehensive CNV analysis to date. The data were prepared and converted using the MouseDivGeno
package as well as in-house scripts. The resultant 2067 CNV calls at least 500 nucleotides in length were analyzed using our
developed and separately reported software, Hotspot Detector for Copy Number Variants (HD-CNV), to identify recurrent CNV
regions detected across multiple samples. These regions are either commonly inherited or hotspots of recurrent mutational events.
This dataset serves as a reference for comparison with independently reported mouse CNVs associated with different genotypes