Our Science – Reilly Website
Karlyne M. Reilly, Ph.D.
The Genetic Modifiers of Tumorigenesis Section is focused on studying tumor biology and the role of genetic background in tumor initiation and progression, using mouse models of nervous system tumors in the context of Nf1 mutation. Cancer is a complex disease that involves the interaction of many factors, such as diet, environmental mutagens, and genetic background. In patients it is difficult to dissect the relative contributions of different factors to cancer, because of the heterogeneity of the human population. This is particularly true in the case of rare cancers where there may not be enough patients available to establish the statistical significance of different risk factors or even whether susceptibility is heritable. Mouse models of cancer allow for the control of diet and environment variables to focus on the role of genetics in tumorigenesis. Mouse models of rare cancers can be used to generate large homogenous populations to test the genetic basis of these cancers. We are focused on two rare, but incurable nervous system tumors, astrocytoma in the central nervous system (CNS) and malignant peripheral nerve sheath tumors (MPNSTs) in the peripheral nervous system (PNS), to better understand the complex genetics underlying tumor susceptibility.
Astrocytic gliomas, including astrocytoma and glioblastoma multiforme (GBM), have an incidence of 1-3 in 100,000 person-years and a 5-year survival rate of 3-29% depending on the grade at diagnosis. Although rare, human gliomas have been the focus of considerable research, providing a wealth of correlative patient data for cross-species comparison and hypothesis testing in mouse models. Patient data have emphasized the different molecular subtypes of GBM and their differential response to therapy, demonstrating the importance of a personalized approach to glioma treatment and the need for an improved understanding on how an individual's unique genetic makeup may influence prevention and treatment approaches. Mutations in NF1 and TP53 (the focus of our studies) are among the most common in GBM and are characteristic of particular subtypes.
Neurofibromatosis type 1 (NF1) is the most common genetic disease affecting the nervous system and is caused by mutation in the NF1 gene. It affects 1 in 3500 people with an increased risk for cancer in both the CNS and PNS. The disease is 100% penetrant, but shows variable expressivity. Comparison of monozygotic twins with more distantly related family members has provided strong evidence for genetic modifiers in many aspects of the disease. Epidemiological studies have shown that benign CNS and PNS tumors cluster in different sub-populations of NF1 patients suggesting that different risk factors may contribute to tumor formation at these different sites. MPNSTs arise in NF1 patients through malignant transformation of plexiform neurofibromas that is associated with mutation of Tp53.
Our research has identified at least 6 regions of the genome important for susceptibility to both astrocytoma and MPNST in a genetically engineered mouse model of NF1, in which the Nf1 and Trp53 genes are mutated. We have found that the effect of these regions is dependent on the sex of the animal and how it inherits mutations from its mother or father. We have identified specific [sex X genotype], [sex X parental inheritance], and [genotype X parental inheritance] interacting effects. Using a combination of mouse models, cell culture, genomic, and bioinformatic approaches, we are identifying the modifier genes underlying these effects. Our work suggests that risk for either MPNST or astrocytoma is best thought of as a 'bar code' in which individual susceptibility loci change risk depending on the larger genetic/epigenetic context. These types of genetic risk factors can be detected in mouse models through careful stratification of datasets and control of inheritance patterns through breeding; however they are likely to be difficult or impossible to detect in human genome-wide association studies (GWAS) without knowledge of how to stratify the population. As we develop a better understanding of epistatic interactions in mouse, this stratification in human populations may be possible to confirm candidate modifiers identified in mice.
The long-term goal of our research is to develop a better understanding of glial tumor biology and cancer risk. The identification of modifier genes and understanding their mechanism of action will help us to begin to understand the epistatic interactions between modifiers. This will greatly aid in the identification of susceptibility loci in human population. In addition, these modifier genes will give us a better understanding of the mechanisms of cancer susceptibility, and may provide new approaches for prevention and treatment of cancer.
This page was last updated on 7/1/2014.