Fluorescent detection of homologous recombination reveals the impact of genetic, physiological, and environmental factors on genomic stability

Abstract

Unless repaired correctly, DNA double strand breaks (DSBs) can cause the loss of millions of base pairs of information and can induce cellular toxicity. DSBs are repaired via mitotic homologous recombination (HR), non-homologous end-joining (NHEJ) or microhomology-mediated end-joining (MMEJ). Here we use the Fluorescent Yellow Direct Repeat (FYDR) mouse to examine these pathways. Specifically, we crossed FYDR mice with mice lacking an essential NHEJ protein. Consistent with in vitro studies, we observed an increase in HR in the NHEJ deficient mice, indicating a shift from one pathway to another. Additionally, FYDR mice deficient in ERCC1, a protein involved in several pathways including nucleotide excision repair and MMEJ, showed an increase in HR. We describe a possible model for this observation. HR is presumed to be largely limited to replicating cells; however, little is known about differences in HR rates between tissues. Thus, we engineered the Rosa26 Direct Repeat-GFP (raDR-GFP) mouse that enables study of HR in many tissues in response to endogenous and exogenous factors. The raDR-GFP mouse harbors two truncated EGFP genes integrated at the ROSA26 locus. HR at the locus yields a full-length EGFP gene and a fluorescent cell. In adult raDR-GFP mice, differences in frequency of recombinant cells among tissues of challenged and unchallenged mice demonstrate the utility of raDR-GFP mice in measuring exposure-induced HR and the importance of multi-tissue studies. We also observed the progressive accumulation of recombinant cells in the pancreas, liver, and colon with age. These data are consistent with the finding that cancer is an age-related disease requiring time to accumulate tumorigenic mutations. To test the hypothesis that chronic inflammation promotes the induction of DSBs, we bred raDR-GFP mice deficient in an anti-inflammatory cytokine. These mice showed an increase in spontaneous HR in the pancreas. Interestingly, 10 week-infection of RAG2-/- raDR-GFP mice with H. hepaticus, and longer-term 20-week infection with H. trogontum did not have the same effect on HR in the pancreas, liver, or colon. Further studies of large-scale sequence rearrangements, point mutations, and small deletions in multiple tissues in response to environmentally-induced inflammation are planned.