Elevated presence of retrotransposons at sites of DNA double strand break repair in mouse models of metabolic oxidative stress and MYC-induced lymphoma.

The chromosomally integrated shuttle vector pUR288 contains a lacZ reporter gene to study mutagenesis in vivo. We used pUR288 to compare patterns of genomic instability in two mouse models, lymphoma resulting from deregulated c-MYC expression (lambda-MYC), and endogenous oxidative stress caused by partial glucose 6-phosphate dehydrogenase (G6PD) deficiency. We found previously that spontaneous mutations in both models were predominantly genomic rearrangements of lacZ with mouse sequences, while most mutations in controls were point mutations. Here, we characterized the fine structure of 68 lacZ/mouse rearrangements from lambda-MYC lymphomas and G6PD deficient mice by sequencing breakpoint junctions and determining the origin of recombining mouse sequences. Fifty-eight of 68 (85%) recombination partners were identified. The structure of rearrangements from both lambda-MYC and G6PD deficient mice were remarkably alike. Intra-chromosomal deletions and inversions were common, occurring in 41% (24/58) of rearrangements, while 59% (34/58) were random translocations between lacZ and other chromosomes. Signatures of double strand break repair by nonhomologous end joining were observed at breakpoint junctions; 37% (25/68) contained 1-4 bp microhomologies, while the remaining breakpoints had no sequence homology. Long interspersed nuclear element-1 (LINE-1 or L1) retrotransposons, which constitute approximately 10% of the mouse genome, were present at 25% (17/68) of breakpoints, suggesting their participation in rearrangements. The similarity in the structure of rearrangements is consistent with the hypothesis that genetic rearrangements in lambda-MYC lymphomas and G6PD deficient mice result from the same mechanism, mutagenic repair of DNA double strand breaks arising from oxidative damage.

Paradoxical decrease in mutant frequencies and chromosomal rearrangements in a transgenic lacZ reporter gene in Ku80 null mice deficient in DNA double strand break repair.

Repair of DNA double strand breaks (DSB), either by homologous recombination (HR) or nonhomologous end-joining (NHEJ), is essential to maintain genomic stability. To examine the impact of NHEJ deficiency on genomic integrity in Ku80 null (Ku-) mice, the chromosomally integrated shuttle vector pUR288, which includes a lacZ reporter gene, was used to measure mutations in vivo. Unexpectedly, a significant decrease was found in mutant frequencies of Ku- liver (5.04x10(-5)) and brain (4.55x10(-5)) compared to tissues obtained from normal (Ku+) littermates (7.92x10(-5)and 7.30x10(-5), respectively). No significant difference was found in mutant frequencies in spleen from Ku- (7.21x10(-5)) and Ku+ mice (8.16x10(-5)). The determination of the mutant spectrum in lacZ revealed the almost complete absence of chromosomal rearrangements (R) in Ku- tissues (0.5%, 3/616), a notable distinction from Ku+ controls (16.7%, 104/621). These findings suggest that accurate repair of DSB by HR and elimination of cells with unrepaired DNA damage by apoptosis are capable of maintaining genomic stability of the lacZ reporter in Ku- mice.

Redox imbalance and mutagenesis in spleens of mice harboring a hypomorphic allele of Gpdx(a) encoding glucose 6-phosphate dehydrogenase.

Mice harboring the activity-attenuated Gpdx(a-m2Neu) allele and also harboring a chromosomally integrated lacZ reporter gene to study mutagenesis (pUR288) were used to demonstrate that moderate glucose 6-phosphate dehydrogenase (G6PD) deficiency causes elevated mutagenesis and endogenous oxidative stress in the spleen. G6PD-deficient spleens with a residual enzyme activity of 22% exhibited a dramatic shift in the mutational pattern of lacZ (4.6-fold increase in the prevalence of recombination mutations of lacZ) together with a 1.8-fold increase in mutant frequencies in lacZ. A concomitant 3-fold reduction in catalase activity (dependent upon NADPH) indicated that the in vivo supply of G6PD-generated NADPH was insufficient. An additional 3-fold increase in oxidized glutathione suggested that redox control was disturbed in G6PD-deficient spleens. These findings indicate that G6PD is required for limiting oxidative mutagenesis in the mouse spleen. Gpdx(a-m2Neu) is the first hypomorphic allele of a mouse housekeeping gene associated with elevated somatic mutagenesis in vivo.

Genomic instability in mouse Burkitt lymphoma is dominated by illegitimate genetic recombinations, not point mutations.

lambda-MYC-induced mouse Burkitt lymphoma (BL) harboring the shuttle vector pUR288, which includes a lacZ reporter gene to study mutagenesis, was employed to assess genomic instability associated with MYC deregulation. The frequency of lacZ mutations in lymphomas was elevated only 1.75-fold above that in normal tissue, indicating that mouse BL does not exhibit a phenotype of hypermutability. However, the nature of lacZ mutations was strikingly different in normal tissues and lymphomas. While point mutations comprised approximately 75% of the mutations found in normal tissues, apparent translocations, deletions and inversions constituted the majority of mutations ( approximately 65%) in lymphomas. Genomic instability in mouse BL thus seems characterized by a preponderance of illegitimate genetic rearrangements in the context of near-background mutant frequencies. SKY analyses of cell lines from primary BL tumors revealed substantial changes in chromosomal structure, confirming the lacZ studies. Bi-allelic deletions of the tumor suppressor p16(Ink4a) were detected in six out of 16 cell lines, illustrating cellular selection of advantageous mutations. Together, these approaches indicate that MYC may contribute to lymphomagenesis through the dominant mutator effect of inducing chromosomal instability. The results further suggest that a phenotype of hypermutability (elevated mutant frequency) may not always be required for oncogenesis to occur.

Moderate G6PD deficiency increases mutation rates in the brain of mice.

Mice that harbored the x-ray-induced low efficiency allele of the major X-linked isozyme of glucose-6-phospate dehydrogenase (G6PD), Gpdx(a-m2Neu), and, in addition, harbored the transgenic shuttle vector for the determination of mutagenesis in vivo, pUR288, were employed to further our understanding of the interdependence of general metabolism, oxidative stress control, and somatic mutagenesis. The Gpdx(a-m2Neu) mutation conferred moderate G6PD deficiency in hemizygous males (Gpdx(a-m2Neu/y)) displaying residual enzyme activities of 27% in red blood cells and 13% in brain (compared to wild-type controls, Gpdx(a/y) males). In spite of this mild phenotype, the brains of G6PD-deficient males exhibited a significant distortion of redox control ( approximately 3-fold decrease in the ratio of reduced glutathione to oxidized glutathione), a considerable accumulation of promutagenic etheno DNA adducts ( approximately 13-fold increase in ethenodeoxyadenosine and approximately 5-fold increase in ethenodeoxycytidine), and a substantial elevation of somatic mutation rates ( approximately 3-fold increase in mutant frequencies in lacZ, the target and reporter gene of mutagenesis in the shuttle vector, pUR288). The mutation pattern in the brain was dominated by illegitimate genetic recombinations, a presumed hallmark of oxidative mutagenesis. These findings suggested a critical function for G6PD in limiting oxidative mutagenesis in the mouse brain.