Abstract
RNA interference is a powerful tool for controlling gene expression in mammalian systems; as such, it has become an effective alternative to conventional knockout approaches. RNAi has proven to be an efficient method to inhibit tumor suppressor gene function and yield insight into the important players in cancer biology. Still, the promise of regulatable RNAi transgenic mice has yet to be realized because the reproducible generation of these animals remains a significant technical limitation. By combining optimized fluorescence-coupled mir30-based shRNA technology with high efficiency ES cell targeting, I have developed a flexible and scalable pipeline for the rapid and reliable production of doxycycline-regulated shRNA transgenic mice. These RNAi mice contain single copy DOX-regulatable shRNAs downstream of the endogenous Collagen Type 1 locus, allowing for spatial, temporal and reversible gene expression in mice. Using this platform, I generated novel DOX-regulated shRNA transgenic lines targeting the bioluminescence reporter Luciferase and endogenous tumor suppressor genes, including Trp53, INK4a and ARF, each showing strong doxycycline-dependent knockdown of its target protein, without disrupting processing of endogenous miRNAs. To study the role of these TSGs in the maintenance of Kras<super>G12D</super> driven lung adenocarcinomas, I crossed these to produce mice bearing an shRNA, CCSP-rtTA (clara cell specific promoter. reverse tet-transactivator), LSL-Kras<super>G12D</super> and LSL-Luciferase alleles. However, owing to the slow rate and high expense of producing quadruple transgenic mice, I later devised a strategy for"speedy" mouse model production. This approach entailed re-derivation of embryonic stem cells harboring the relevant alleles and subsequent generation of"mosaic" models produced by blastocyst injection. Using these RNAi mouse models, I show that INK4a/ARF or Trp53 downregulation by RNAi cooperates with Kras<super>G12D</super> to accelerate lung tumorigenesis and recapitulate the phenotypes of knockout models. Additionally, I investigate whether INK4a/ARF or Trp53 knockdown is required for tumor maintenance. Together, this work built a platform that greatly accelerates the rate at which one can study genetic interactions and tumor maintenance genes and also identify and validate new drug targets in vivo. This approach can be applied to build many other complex cancer models and thus may have significant implications for guiding future therapies.
