Complex formation of CSA/CSB proteins with mtDNA as well as mtOgg1 and mtSSBP1 in oxidatively stressed mt. (A) GEMSA: Biotin-labeled mtDNA was incubated with mitochondrial extracts of oxidatively stressed normal or CSA-deficient human cells and with subsequent PAGE. Incubation with CSA antibody in WT mitochondrial extract showed a supershift (arrow). CoIP with biotin-labeled mtDNA. CoIP of CSA– (B) or CSB–mtDNA (C) complexes occurred only in WT mitochondrial extracts but vanished under DNase digestion. Complexes containing mtOgg1, mtSSBP1, COXIV, and ATP-Sβ were extracted by CoIP from mitochondrial extracts of CSA- and CSB-deficient and WT cells, exposed to 30 µM H₂O₂. The presence of CSA (D) and CSB (E) could be detected in complex with mtOgg1 and mtSSBP1 (arrows) by Western blot. Pictures are representatives of at least three independent experiments.

Age dependent accumulation of mtDNA mutations in subcutaneous fat tissue of Csb^m/m and Csa^−/− mice. (A) Subcutaneous fat tissue from WT mice and mice lacking a functional Csb or the Xpa gene were assessed for a frequent mouse-mtDNA deletion (D17). Three age groups (13, 5, and 130 wk, 4 mice per group) of each genotype were investigated. Aged Csb^m/m mice show a strong increase of D17 in subcutaneous fat compared with WT or Xpa^−/− controls (Student's t test, P < 0.01). Each PCR experiment was repeated independently two times. (B) Hematoxylin and eosin–stained cross sections with reduced subcutaneous fat tissue in old Csb^m/m mice aged for 130 wk. Bar, 200 µm. (C) Measurement of D17 deletion in subcutaneous fat of WT, Csa^−/− and Xpd^TTD/TTD mice similar to A. Each PCR experiment was repeated independently two times. (D) Restriction digest of PCR product of deletion D17 with EcoRV. (left) Undigested fragment of 850 bp. (right) EcoRV digest leading to fragments of 570 and 280 bp. Data are given as representative of at least three independent experiments.

Mitochondrial localization of CSA and CSB in oxidatively stressed cells. Confocal laser scanning microscopy with green fluorescence for nuclear staining, blue fluorescence for mitochondrial staining, and red fluorescence staining for either CSA (A) or CSB (B). Pink fluorescence results from colocalization (white circles) of red mitochondrial CSA/CSB and blue mitochondrial staining. Pictures are representatives of at least five separate experiments. Bar, 1 µm. Exposure of cells to 25 µM H₂O₂ for 12 h leads to signal increase for mitochondrial CSA (C) and CSB (D). Histograms show signal intensities along the red line in the picture with simultaneous signal increment. (E) CSA-deficient fibroblasts transfected with CSA WT protein. Pictures are representatives of at least three independent experiments. Bar, 1 µm. (F) Whole-cell extracts (WCE) and mitochondrial extracts (Mt) prepared from H₂O₂-treated cells. Presence of mtDNA (IS) and absence of nuclear DNA (GAPDH), shown by PCR. Immunoblotting of mitochondrial ATP-synthetase β, phospho-NF-κB, S6, CSA, and CSB. Pictures are representatives of at least three independent experiments. (G) Electron microscopic image of oxidatively stressed normal human fibroblasts stained with gold-labeled (circles) CSA (top) or CSB (bottom). Bars: (top CSA image) 1.1 µm; (top CSB image) 1.6 µm; (bottom) 165 nm. Electron micrographs are representatives of at least two separate experiments. (H) Top: mt (10 µg protein per lane) isolated from H₂O₂-treated WT cells were incubated where indicated with 100 µg/ml proteinase K (PK) in the presence or absence of Triton X-100 (Tx-100). SDS-PAGE of mitochondrial proteins with antibodies against CSA, Tom20, Tim23, and superoxide dismutase (SOD). Bottom: mt (25 µg per lane) isolated from H₂O₂-treated WT cells were incubated with 25 µg/ml proteinase K (PK) in the presence or absence of Triton X-100 (Tx-100). SDS-PAGE of mitochondrial proteins with antibodies against CSB and cytochrome C. Pictures are representatives of at least two independent experiments.

Accelerated induction of mt mutations in UVA-irradiated CSA- and CSB-deficient cells. Primary human fibroblasts of two and three patients suffering from CSA and CSB, respectively, one XP patient (XP-D), and three normal individuals were irradiated with UVA and the relative amount of common deletion was assessed by quantitative real-time PCR. Levels of the common deletion are given in 2^−ΔΔCt as mean ± SD of at least three separate experiments. Levels above one indicate a higher amount of common deletion in UVA-treated cells relative to untreated control cells. Premature UVA-mediated induction of common deletion in CSA cells can be attenuated by treatment with the radical scavenger vitamin E. Expression of HA-tagged CSB protein in CSB-deficient cells lead to normalization of mitochondrial mutagenesis.

Aging-associated loss of subcutaneous fat tissue is mediated on the cellular level by increased cellular turnover. Oxidatively stressed fibroblasts containing high levels of mtDNA deletions show increased apoptosis. (A) Ziehl-Neelsen staining for lipofuscin of subcutaneous fat tissue from 130-wk-old Csb^m/m mice containing high levels of mtDNA deletions (6 WT and 4 Csb^m/m mice were investigated). Increased level of macrophages containing granular lipofuscin (arrowheads) and no increase in normal fat cells (asterisks). Arrow points to hair follicle. Bar, 50 µm. (B) Quantification of histological sections of lipofuscin-positive macrophages from A with 17% (1 of 6 individuals) of histological sections with lipofuscin-positive cells (macrophages) in WT and 75% (3 of 4 individuals) in Csb^m/m mice. (C) FACS analysis for annexin V–positive and propidium iodide (PI) negative (apoptotic) cells with (black bars) or without (gray bars) 2 wk of UVA irradiation. Increase of apoptotic cells in CSA- and CSB-deficient cells compared with WT and nonirradiated controls. Data are representative of at least three separate experiments.