Accumulation of oxidative DNA damage in brain mitochondria in mouse model of hereditary ferritinopathy

Abstract

Tissue iron content is strictly regulated to concomitantly satisfy specialized metabolic requirements and avoid toxicity. Ferritin, a multi-subunit iron storage protein, is central to maintenance of iron homeostasis in the brain. Mutations in the ferritin light chain (FTL)-encoding gene underlie the autosomal dominant, neurodegenerative disease, neuroferritinopathy/hereditary ferritinopathy (HF). HF is characterized by progressive accumulation of ferritin and iron. To gain insight into mechanisms by which FTL mutations promote neurodegeneration, a transgenic mouse, expressing human mutant form of FTL, was recently generated. The FTL mouse exhibits buildup of iron in the brain and presents manifestations of oxidative stress reminiscent of the human disease. Here, we asked whether oxidative DNA damage accumulates in the FTL mouse brain. Long-range PCR (L-PCR) amplification-mediated DNA damage detection assays revealed that the integrity of mitochondrial DNA (mtDNA) in the brain was significantly compromised in the 12- but not 6-month-old FTL mice. Furthermore, L-PCR employed in conjunction with DNA modifying enzymes, which target specific DNA adducts, revealed the types of oxidative adducts accumulating in mtDNA in the FTL brain. Consistently with DNA damage predicted to form under conditions of excessive oxidative stress, detected adducts include, oxidized guanines, abasic sites and strand breaks. Elevated mtDNA damage may impair mitochondrial function and brain energetics and in the long term contribute to neuronal loss and exacerbate neurodegeneration in HF.

Figure 1

Long range-PCR (L-PCR) reveals amplification-blocking adducts in mtDNA in the FTL-tg mouse brain. Representative images of L-PCR products generated by amplification of the entire mitochondrial genome (16 kb) of FTL-tg and wild type mice are shown in the upper panel (A). Nuclear DNA was assessed by amplification of the 12 kb beta-globin mouse locus (A, middle panel). Amplification reactions assembled with rat DNA only, show a 5-kb product, which serves as an internal control (right lane). Amplification yields of mitochondrial DNA (16 kb) from FTL-tg brains (n=6) were significantly reduced compared to amplification yields obtained with wild type DNA (bar graphs; *P<0.05). No significant difference in amplification was detected in the case of nuclear DNA (right panel). The relative levels of mtDNA in wild type and FTL brains were assessed by measuring amplification products generated on the respective mtDNA templates by standard PCR amplification assays, which yield 1.3 kb products. Amplification of rat DNA (5 kb) served as internal PCR control (A, lower panel). To ascertain linearity of amplification assays, template titration reactions were assembled with 20–100 ng DNA and the linear range of amplification was determined (B). A representative image of 16 kb amplification yields is shown (left). Amplification products were quantified; values were averaged and plotted as a function of template amount. Counts obtained with 50 ng of mouse DNA were assigned the value of 1 (right).

Figure 2

Detection of oxidation adducts in nuclear DNA from FTL-tg and wild-type mice brains. DNA templates were preincubated without/with the DNA modifying enzymes, Fpg or Endo lll, prior to amplification of the 12-kb beta globin locus. Representative images of L-PCR amplification with/out Fpg (upper panel) and Endo lll (lower panel) are shown: preincubation with Fpg resulted in significant reduction (*P<0.05) in amplification of both, the wild type and FTL-tg (n=6), revealing oxidized guanines in both templates. The reduction in amplification following Endo lll digest, did not reach statistical significance (bottom right).

Figure 3

Greater levels of oxidative damage revealed in mtDNA of FTL-tg compared to wild-type brains. DNA was preincubated without/with the DNA modifying enzymes, Fpg or Endo lll, prior to PCR amplification of the 16-kb mitochondrial genome. Representative images of L-PCR amplification yields following Fpg digest (upper panel) and Endo lll (lower panel) are shown: Preincubation with Fpg resulted in a significant reduction (*P<0.05) in amplification of both, wild type and FTL-tg mtDNA templates (n=6). The reduction in L-PCR amplification was significantly greater in the case of FTL-tg when compared to wild-type mtDNA (†P<0.05). Likewise, the reduction in amplification of FTL-tg mtDNA following Endo lll digest, was significantly greater (†P<0.05) than reduction measured for the wild-type brain mtDNA, which did not reach statistical significance (bottom right).