Many human cancers fail to effectively respond to chemotherapy, and cancers that initially respond frequently acquire drug resistance and relapse. Our lab uses emerging RNAi screening technology combined with murine stem reconstitution and tumor transplantation systems to investigate the genetic basis for intrinsic and acquired chemotherapeutic resistance. Our aim is to use these tractable mouse models to identify novel cancer drug targets, as well as strategies for tailoring existing cancer therapies to target the vulnerabilities of specific malignancies.
In vivo fluorescent imaging of a mouse lymphoma following reconstitution with stem cells infected with a GFP-tagged p53 shRNA.
Modeling tumor development and drug response using in vivo RNAi.
Our lab uses adoptive transfer experiments in the mouse hematopoietic system to model tumor development. Initially, this work focused on gene overexpression studies involving deregulated oncogenes. Subsequently, we have expanded the utility of this adoptive transfer system using stable RNAi. Using retroviral infection of small hairpin RNAs (shRNAs) targeting the tumor suppressor p53 into Myc-overexpressing hematopoietic stem cells, we have successfully generated tumors which biochemically and phenotypically suppressed gene expression over extended periods in vivo. This technology has allowed us to accelerate and expand our analysis of the impact of defined lesions both on tumor onset and therapeutic response. Interestingly, different p53 shRNAs produced distinct phenotypes in vivo, ranging from benign lymphoid hyperplasia to highly disseminated lymphomas that paralleled the nullizygous setting. In all cases, the severity and type of disease correlated with the extent to which specific shRNAs inhibited p53 activity. Therefore, RNAi can stably suppress gene expression in stem cells and reconstituted organs derived from those cells. Moreover, intrinsic differences between individual shRNA expression vectors targeting the same gene could be used to create an ‘epi-allelic series’ for dissecting gene function and tumorigenesis in vivo. These experiments demonstrate the potential of RNAi as a tool to study gene function in vivo.
Following these initial proof of principle experiments, we have begun to use in vivo RNAi to investigate the role of putative tumor suppressors in the inhibition of Myc-induced lymphomagenesis. This includes the targeted analysis of candidate tumor-suppressive pathways, as well as the use of shRNA libraries to perform unbiased screens for novel tumor suppressors.
Using in vivo RNAi to investigate drug action and chemotherapeutic resistance.
We have generated several shRNA vectors that mediate resistance to conventional chemotherapeutics in vivo. We are currently expanding this approach, through the use of targeted RNAi libraries, to assess the role of thousands of cancer-relevant genes in the response of diverse tumor types to a wide array of chemotherapeutics. This strategy allows us to perform genetic screens for mediators of chemotherapeutic response in a relevant therapeutic setting.
Importantly, this approach has broad flexibility. First, distinct tumor types can be examined to determine the effect of tumor genotype on mechanisms of chemotherapeutic resistance. Second, the pattern of shRNAs conferring drug resistance in an individual tumor treated with established chemotherapeutic can serve as a “treatment fingerprint”, such that the mechanisms of action of novel chemotherapeutics may be deduced by comparison with these established patterns. Third, in addition to examining drug resistance, we can use representational approaches to identify shRNAs that promote sensitivity to specific drugs. In doing so, we hope to identify drug targets whose inactivation synergizes with existing anti-cancer therapies.
Examining the role of genetic instability on acquired and intrinsic drug resistance.
Genetic instability is a hallmark of advanced and chemoresistant human malignancy. However, using existing models, it is difficult to distinguish between the role of genetic instability in cell transformation versus chemotherapeutic response. Furthermore, it remains unclear whether genetic instability itself mediates chemoresistance or whether drug resistance arises as the consequence of an underlying lesion that simultaneously promotes both chemoresistance and instability. Our development of technology to acutely suppress genes in tractable tumor models provides a means of distinguishing between the genetics of cell transformation, acute chemoresistance and acquired chemoresistance.
- We have generated a set of shRNAs targeting genomic stability pathways including DNA double-strand break repair, mismatch repair, homologous recombination, telomere function and mitotic checkpoints. These shRNAs are being used in combination with in vivo tumor transplantation systems to determine:
- 1) the effect of gene suppression on overall tumor growth in vivo,
- 2) the acute effect of gene suppression on tumor cell survival following treatment with cytotoxic chemotherapy and
- 3) the effect of long-term gene suppression and the consequent accumulation of genomic alterations on tumor cell survival following treatment with cytotoxic chemotherapy.
Ideally these experiments will clarify whether genetic instability is inherently advantageous or deleterious to an established tumor and whether the acute loss of DNA repair mechanisms or the acquired loss of genetic stability alters tumor drug response.
Ross Dickins, Michael Hemann, Jack Zilfou, David Simpson, Ingrid Ibarra, Gregory Hannon and Scott Lowe. Probing tumor phenotypes using stable and regulated synthetic microRNA precursors. Nat Genet. (2005) 37 (11): 1289-1295.
Michael T. Hemann, Anka Bric, Julie Teruya-Feldstein, Andreas Herbst, Jonas Nilsson, Carlos Cordon-Cardo, John Cleveland, William Tansey, and Scott Lowe. Evasion of the p53 tumor surveillance network by tumor-derived MYC mutants. Nature. (2005) 436 (7052): 807-811.
Michael Hemann, Jack Zilfou, Zhen Zhao, Darren Burgess, Greg Hannon, Scott Lowe. Suppression of tumorigenesis by the p53 target PUMA. Proc Natl Acad Sci. (2004) 101(25): 9333-8.
Michael Hemann, Jordan Fridman, Jack Zilfou, Eva Hernando, Patrick Paddison, Carlos Cordon-Cardo, Gregory Hannon and Scott Lowe. An epi-allelic series of p53 hypomorphs created by stable RNAi produces distinct tumor phenotypes in vivo. Nat Genet. (2003) 33(3): 396-400.