A Systems Genetics Approach to Low Dose Radiation

Brynn H. Voy
Oak Ridge National Laboratory

Abstract

Ongoing debate about the safety of airport security scanners and, very recently, the earthquakerelated radiation leaks in Japan have heightened the public’s interest in the risks of exposure to radiation at low doses. Epidemiological detection of health risks from low dose radiation exposure is challenged by a number of factors, including biochemical intersection between the effects of low dose radiation and many other environmental stressors, the differential effects of lifestyle variables that impact the response to stress, and the underlying genetic variation within a population. Nonetheless, a very heterogeneous population is the ultimate target in which low dose radiation risk must be understood. Genetically, radiation sensitivity is a complex trait, conferred by interactions between multiple loci across the genome. It is also an emergent trait, the physiological consequences of which depend upon crosstalk between a variety of tissues and cell types, as well as the environment and lifestyle.

Systems genetics is an emerging discipline that exploits the genetic variation of populations to both uncover genetic variants that mediate phenotypic spectrums and assemble mechanistic pathways that confer disease risk. Genetic reference populations, such as panels of recombinant inbred strains, expand the dimension of systems genetics by allowing traits collected over time and space to be integrated and mined as if they were collected from the same individuals. We address genetic susceptibility to low dose radiation exposure using mouse recombinant inbred strain panels and a systems genetics framework, allowing two simultaneous and complementary outcomes: 1.) de novo identification of genetic variants that mediate low dose sensitvity, and 2.) extraction of multiscale networks that span from genetic variation through coordinated gene expression networks to overlying cellular and tissue level adaptions and the resultant systemic effects. Physiologically, our emphasis is on LDR effects on the immune system due to both its known radiosensitivity and its impact on many disease processes, such as carcinogenesis and cardiovascular disorders.

We have begun by first establishing the genetic architecture that underlies immunophenotypes in the BXD (C57BL/6J X DBA/2J) RI panel, which was created from two strains with established differences in radiation sensitivity. We reasoned that the coexpression networks that control radiation sensitive cell types are also likely to exhibit radiation sensitivity, and that genetic variation in these networks may mediate genetic susceptibility to radiation outcomes. Significant QTLs, candidate genes and gene networks were identified for CD4+ and CD8+ T lymphocytes (among other cell types), two populations that are highly sensitive to radiation at low doses. Importantly, expression of each of the two main candidate genes (acid phosphatase 1 (Acp1) and protein tyrosine phosphatase, receptor type, K (Ptprk)) that together explain the majority of variation in CD4+:CD8+ was significantly altered by a single low dose (10 cGy) of ionizing radiation. Using graph algorithms developed by co-PI Langston, we found that Acp1 is part of a large coexpression clique that appears to be involved in cell cycle control. In the same mice we also observed significant strain-dependent variation in radiationinduced superoxide dismutase (Sod) activity and identified a QTL and candidate genes for the important radiation response trait. Functionally, we found that the same low dose exposure singificantly enhanced an index (phagocytosis) of the innate immune system. We are now performing follow-on studies directed to identification of the mechanisms through which LDR enhances neutrophil phagocytosis as well as expanding these studies to include effects of LDR on cell migration, chemotaxis and bacterial killing.

Microarray expression profiling of spleens from C57BL/6J and DBA/2J mice 24 hours after irradiation suggests that low dose exposure significantly alters maturation and abundance of immune cell populations, at least acutely, and in a manner that differs markedly between strains. DBA/2J expression profiles suggest a suppression of immune cell maturation, processes that were not reflected in profiles from C57BL/6J, a strain considered to be relatively radiation resistant. We are now profiling thymus and bone marrow from the same mice to expand the scope of radiation effects on immune cells in these two strains, including the creation of low dose-specific, multi-tissue gene networks using graph algorithms. A major focus of this effort is to determine the longer term and potentially sustatined impact of low dose on the immune system in these two strains and to map the genetic determinants of these outcomes. We are also working to determine the extent to which the early (within a few hours) differences in radiation response mediated through p53 signaling, which differs between C57BL/6J and DBA/2J, determine the delayed impact on the immune system.

The BXD panel is a valuable model system for radiation sensitivity due to the known differences in sensitivity between the two parental strains. However, the emerging Collaborative Cross (CC), an RI panel created from eight rather than two strains, will gratly expand the ability to map genetic determinants of radiation sensitivity. As a precursor to transitioning to the CC we are profiling expression patterns in spleen of irradiated mice from the eight parental strains. We expect that this population will expand both the spectrum of radiation sensitive phenotypes and our ability to identify mechanisms of radiation sensitivity in a manner that will be translatable to the human population. | PDF Version