Studies of families with neurofibromatosis type 1 (NF1) have demonstrated that the severity of this disease depends on modifier genes in the individuals. We are using a mouse model of the malignancies associated with NF1 in which the Nf1 and p53 gene are mutated on the same chromosome and screening for modifier genes in different strains of mice. Importantly, both Nf1 and p53 are mutated in sporadic forms of glioblastoma and sarcoma, thus the understanding of tumor mechanisms gained from this project may have relevance to sporadic nervous system tumors and well as NF1. Using a combination of mouse genetics and expression profiling in tumors, we have recently identified several regions responsible for resistance to developing malignant peripheral nerve sheath tumors (MPNSTs), and several regions responsible for susceptibility to astrocytoma. The astrocytoma modifier regions affect the location of astrocytoma and the differential susceptibility of males and females. We have identified a candidate modifier gene on mouse chromosome 11 affecting MPNST susceptibility and a candidate modifier gene on mouse chromosome 12 affecting astrocytoma specifically in males. We are working to identify additional candidate modifier genes on other chromosomes, using information gathered from both mouse and humans. These candidate modifiers will then be tested for their role in tumorigenesis in our mouse model and then in human cancers. During fiscal year 2013, we have studied the mechanism of how the imprinted gene Grb10 on mouse chromosome 11 acts to block tumor growth of MPNSTs. We have continued our work from FY2012 by developing additional molecular and cellular biology tools to confirm the interaction of Grb10 with Igf1r and Nedd4 in MPNST cells and alters the phosphorylation state of Akt3. We are extending these studies to understanding which cell biological processes is altered by these interactions using overexpression and knockdown of these genes in combination with changes in Grb10 levels. Because Grb10 is involved in negative feedback loops within the ras signaling pathway, understanding the details of its biochemical function could have important implications for targeting the ras pathway. To identify modifiers of astrocytoma, we are performing network analysis of genotyping data from 600 backcross mice in collaboration with a group at University of Wisconsin. Our analysis shows that although astrocytoma susceptibility is complex, context-dependent, and different for males and females, it may involve common pathways, with different points in the pathway being most vulnerable in different individuals. We are working to understand how to recognize these pathways signatures so that they can be applied to complex human populations and Genome Wide Association Study datasets, with the goal of identifying heritable risk "barcodes" for glioma susceptibility. We have identified a candidate gene, Cdca7l, for the astrocytoma resistance locus in males (Arlm1) using cross-species comparison with human REMBRANDT and TCGA datasets. Our studies on Cdca7l in cultured cells suggest that this gene acts as an oncogene in male astrocytoma and glioblastoma cells from both mice and humans, but does not show the same action (and may have the opposite effect) in female cells. In fiscal year 2013 we have focused on understanding the mechanism underlying the male-female difference. We have found differential expression of histone demethylases in male and female cells and our preliminary data suggests that these differences interact with Cdca7l levels to regulate gene expression differently in males and females. This research may help to explain why men are at a higher risk for developing astrocytoma and glioblastoma compared to women. In addition to the male-specific Arlm1 locus, we have recently identified a female-specific modifier for pheochromocytoma (Pheom1) in collaboration with researchers at the University of Wisconsin. Pheom1 overlaps with the Down's syndrome region on mouse chromosome 16. A group studying Down's syndrome at Johns Hopkins University (MD) has shown that this region can increase adrenal gland tumors in females of the Nf1/p53 mouse model we use, consistent with the presence of Pheom1 in the Down's syndrome region. Combining this information with microarray data from human pheochromocytoma in collaboration with researchers at NCI and NICHHD, we have identified a short list of candidate genes that we will be testing for their role in pheochromocytoma and the potential mechanism of sex-specificity. In fiscal year 2013, we identified Cbr1 as a candidate modifier with interesting expression differences in male and female pheochromocytoma.
|Amlin-Van Schaick, Jessica; Kim, Sungjin; Broman, Karl W et al. (2012) Scram1 is a modifier of spinal cord resistance for astrocytoma on mouse Chr 5. Mamm Genome 23:277-85|
|Reilly, Karlyne M (2009) Brain tumor susceptibility: the role of genetic factors and uses of mouse models to unravel risk. Brain Pathol 19:121-31|
|Reilly, Karlyne M; Rubin, Joshua B; Gilbertson, Richard J et al. (2008) Rethinking brain tumors: the fourth Mouse Models of Human Cancers Consortium nervous system tumors workshop. Cancer Res 68:5508-11|