Lack of XPC leads to a shift between respiratory complexes I and II but sensitizes cells to mitochondrial stress AbstractGenomic instability drives tumorigenesis and DNA repair defects are associated with elevated cancer. Metabolic alterations are also observed during tumorigenesis, although a causal relationship between these has not been clearly established. Xeroderma pigmentosum (XP) is a DNA repair disease characterized by early cancer. Cells with reduced expression of the XPC protein display a metabolic shift from OXPHOS to glycolysis, which was linked to accumulation of nuclear DNA damage and oxidants generation via NOX-1. Using XP-C cells, we show that mitochondrial respiratory complex I (CI) is impaired in the absence of XPC, while complex II (CII) is upregulated in XP-C cells. The CI/CII metabolic shift was dependent on XPC, as XPC complementation reverted the phenotype. We demonstrate that mitochondria are the primary source of H2O2 and glutathione peroxidase activity is compromised. Moreover, mtDNA is irreversibly damaged and accumulates deletions. XP-C cells were more sensitive to the mitochondrial inhibitor antimycin A, an effect also prevented in XPC-corrected cells. Our results show that XPC deficiency leads to alterations in mitochondrial redox balance with a CI/CII shift as a possible adaptation to lower CI activity, but at the cost of sensitizing XP-C cells to mitochondrial oxidative stress. IntroductionAlthough it is well known that tumor formation depends on a multitude of molecular events, mutation accumulation is a bona fide basis for cellular transformation1. The direct relationship between genomic instability and cancer can be best appreciated in inherited diseases that predispose affected individuals to early emergence of neoplasia.Mutations in genes that encode for DNA repair proteins cause cancer-prone syndromes2. DNA repair diseases usually lead to onset of cancer within the first two decades of the patient鈥檚 life. Xeroderma pigmentosum (XP) is one of these inherited diseases, characterized by photosensitivity, hyperpigmentation, premature skin aging and a 10,000-fold increase in the incidence of skin malignancies3. Mutations in eight genes have been described to give rise to XP: XP-A to XP-G and a variant form, XP-V (POLH) that encodes for an error-prone translesion synthesis DNA polymerase4. All seven canonical XP proteins, from XPA to XPG, are involved in nucleotide excision repair (NER) 鈥?for an extensive review on NER see ref. 5.Although the elevated cancer predisposition in XP patients could be directly explained by the DNA repair defect, a more severe set of phenotypes related to a vast degeneration in the central nervous system raised questions about other functions of XP proteins. Some XP patients also present Cockayne syndrome (CS) features, including growth and developmental retardation, neurodegeneration and premature aging. Some mutations in XPD and XPG can give rise to a combined XP/CS phenotype, while mutations in CSA and CSB, two components of the transcription coupled NER sub-pathway (TCR), give rise only to CS6. It has been speculated that TCR deficiency was the molecular cause of the neurological phenotype, but this hypothesis was partially discredited by evidence that some individuals affected by UV-sensitive syndrome (UVSS) carried mutations in CSA and CSB genes without any discernible neurodegeneration7, 8. Thus, some authors argued that the neurodegeneration phenotype could be due to accumulation of oxidized damage, since cells from XP-G (with a XP/CS phenotype), CS-A and CS-B patients were sensitive to oxidative stress9. Nonetheless, cells from XP-C patients also show increased sensitivity to oxidants while these patients do not manifest neurological abnormalities10, 11. In the global genome NER sub-pathway (GGR), the XPC protein participates in the initial step of lesion recognition in association with its binding partners hRAD23B and centrin-26. Although oxidatively-induced DNA damage is repaired primarily by the BER pathway, a role for XPC in the repair of oxidized DNA lesions has been demonstrated. XP-C cells accumulate 8-oxoGua in nuclear DNA after treatment with oxidizing agents, and the XPC protein interacts physically and functionally with OGG1, stimulating its catalytic activity10.There is growing evidence that DNA repair defects lead to mitochondrial dysfunction. Mitochondrial dysfunction has been well documented in CS, as CS-A and CS-B cells show impaired mitochondrial DNA (mtDNA) repair9, 12, 13, redox imbalance14 and increased mitochondrial autophagy15. Likewise, in cells from ataxia telangectasia (AT) patients, with a mutated ATM protein, as well as in ATM knockout mice, mitochondrial bioenergetics16, 17 and mtDNA repair defects18 have also been demonstrated. CSA, CSB and ATM proteins have been localized in mitochondria, and a direct role for these in mtDNA stability has been demonstrated12, 13, 16. However, not all DNA repair disorders with neurodegeneration can be directly linked to mtDNA repair. De Sanctis-Cacchione patients bearing mutation in XPA gene manifest late neurological symptoms that has been linked to dysfunctional mitophagy. Since XPA is a downstream effector of DNA damage recognition in both GGR and TCR, incomplete DNA repair events keep PARP1 activated, depleting NAD+ and altering NADH/NAD+ ratio. Nutrient-sensitive SIRT1 also uses NAD+ to deacetylate target proteins, including transcription factors that stimulate expression of PGC-1伪, a master mitochondrial biogenesis regulator, which, therefore, is also downregulated. Because PGC-1伪 regulates UCP2 expression, mitochondria from XP-A cells show increased mitochondrial membrane potential leading to elevated ROS generation, due to blocked electron flow with increased reverse electron flow, and to decreased mitophagy19. In line with these findings, it is well known that mitochondrial dysfunction is also a common feature of aging and age-associated diseases, such as cancer and neurodegeneration20, conditions that have been causally linked to genomic instability21.Mitochondrial dysfunction was also demonstrated in human keratinocytes after XPC knockdown22, 23. These effects were linked to nuclear DNA damage accumulation and NOX-induced hydrogen peroxide generation, but no direct effect of XPC knockdown on mitochondrial bioenergetics was investigated. Thus, we investigated whether XPC is directly involved in mitochondrial function and mtDNA maintenance. Using transformed and primary fibroblasts from XP-C patients, we evaluated mitochondrial bioenergetic parameters, ROS production, mtDNA integrity and cytotoxicity of mitochondrial inhibitors. We report that lack of XPC leads to impaired complex I (CI 鈥?NADH dehydrogenase complex) activity and altered mitochondrial H2O2 production/clearance, with consequent mtDNA damage accumulation. XP-C cells overcome complex I dysfunction upregulating complex II (CII 鈥?succinate dehydrogenase complex) components, but at a cost of increased cellular sensitivity to mitochondrial stressors. These results suggest that absence of XPC impacts mitochondrial maintenance and that mitochondrial dysfunction may contribute to tumor formation in an XPC-deficient background.ResultsComplex I supported mitochondrial respiration is impaired in XP-CRecently, Rezvani and colleagues22, 23 showed that XPC silencing in keratinocytes leads to a NOX-dependent metabolic shift from oxidative phosphorylation (OXPHOS) to glycolytic energy production. To directly investigate whether mitochondrial respiration is impaired in XP-C cells we measured oxygen consumption rates (OCR) of the different mitochondrial electron transport chain (ETC) complexes in transformed fibroblasts from normal and XP-C individuals (WT and unXP-C, respectively; see Suppl. Fig.聽S1 for XPC expression levels). In non-permeabilized cells, respiring only on endogenous substrates, OCR was already significantly lower in unXP-C cells, showing a 40% decrease when compared with WT (Fig.聽1A and B, basal). After permeabilizing the plasma membrane with digitonin, to give access to added CI substrates, unXP-C cells still showed a significantly lower OCR, indicating that endogenous substrate availability was not limiting (Fig.聽1B, dig). Phosphorylating respiration, measured after ADP addition, was also significantly lower in unXP-C cells, again showing around 40% decrease, from 7.91鈥壜扁€?.39鈥塶mol O2/min/2鈥壝椻€?06 cells in WT to 4.73鈥壜扁€?.36 in unXP-C cells (Fig.聽1B, ADP). Maximum, uncoupled respiration, measured after CCCP addition, was also significantly diminished in unXP-C cells, again, close to 40% lower (Fig.聽1B, CCCP). Surprisingly, OCR was completely recovered in XPC-corrected (corrXP-C) cells, which showed OCR similar to WT levels under all the conditions evaluated here. In locus genetic correction of the XP4PA cell line was confirmed by RT-qPCR and Western blotting and we observed approximately 53% and 60% of XPC protein and mRNA levels, respectively, in comparison with MRC-5 cell line (Suppl. Fig.聽S1). CorrXP-C cells show similar OCR at basal levels and even an increased digitonin-permeabilized OCR (Fig.聽1B, dig: WT鈥?鈥?.89鈥壜扁€?.27 vs. corrXP-C鈥?鈥?.03鈥壜扁€?.38). Respiratory rates under phosphorylating conditions were similar in WT and corrXP-C cells, as well as maximum uncoupled respiration. Together, these results suggest that XPC absence leads to decreased CI function or that XPC is necessary for its normal function. On the other hand, OCR in presence of carboxyatractyloside (Fig.聽1B, CAT) or oligomycin (not shown) was similar in unXP-C and WT cells, indicating that proton leak-associated mitochondrial respiration was not affected by lack of XPC. However, in corrXP-C cells, OCR in presence of CAT was increased in comparison with both WT and unXP-C cells, an indicative of increased proton leak. Thus, respiratory control (RC), representing the ratio between phosphorylating and non-phosphorylating respiration, is significantly decreased in mitochondria from unXP-C cells. Even though CI-linked maximum respiratory capacity is reverted in corrXP-C cells, RC is also decreased compared to WT, likely reflecting the increased proton leak-associated respiration after XPC correction (Fig.聽1B).Figure 1Complex I-supported mitochondrial function is impaired in XP-C cells. (A) O2 consumption for MRC-5 (WT), XP4PA (unXP-C) and XP4PA expressing XPC (corrXP-C) was measured as described, using 4鈥壝椻€?06 cells in standard reaction medium containing 5鈥塵M CI substrate, with the subsequent additions of 30鈥壜礛 Digitonin, 100鈥壜礛 ADP, 4鈥壜礛 CAT and 1鈥壜礛 CCCP. The figure shows a representative trace OCR experiment. (B) The rates of O2 consumption and respiratory control (RC, ratio between CCCP/CAT) were calculated from data obtained in the experiments represented in (A). (C) CI鈥?鈥塁III activity was measured using 5鈥壜礸 of crude mitochondrial protein. Reduction of cyt c was monitored at 550鈥塶m at 37鈥壜癈 for 5鈥塵in. For the calculations, only the region where the slopes of the reactions were linear were considered. The data represent mean鈥壜扁€塖D of 4 independent experiments. *p鈥?lt;鈥?.05, **p鈥?lt;鈥?.01 and ***p鈥?lt;鈥?.001.Full size image To further elucidate if CI-linked decreased OCR was due to decreased CI activity, we directly measured CI鈥?鈥塁III activity in crude mitochondrial fractions (CMF). CMF obtained from WT cells reduced 278.6鈥壜扁€?5.2鈥塶moles of cytochrome c/min/mg of protein vs. 184.4鈥壜扁€?1.5 from unXP-C cells, while XPC expression in corrXP-C cells restored CI鈥?鈥塁III activity to WT levels (Fig.聽1C). Accordingly, unXP-C CI鈥?鈥塁III activity was 34 and 41% lower than WT and corrXP-C cells, respectively, a reduction very similar to that observed in the OCR experiments. However, mRNA levels of both mitochondrial- (MT-ND1 and MT-ND4L) and nuclear-encoded (NDUFS1) CI subunits were unchanged in XP-C cells (Suppl. Fig.聽2A). Similarly, protein levels of two CI nuclear-encoded subunits were either unchanged (NDUFS3) or upregulated (NDUFS4) in unXP-C cells compared to corrXP-C cells (Suppl. Fig.聽2B). Taken together, these results indicate that unXP-C cells show decreased CI-linked OCR due to reduced CI activity, despite normal expression levels of some of its protein subunits.Impaired CI function in XP-C does not lead to mitochondrial respiratory dysfunctionOCR measurements in intact cells are especially informative because they reflect the use of all carbon sources of the cell, and ATP turnover (synthesis and degradation) includes every cellular reaction that demands ATP, directly or indirectly24. In order to evaluate the consequences of CI-impaired respiration on mitochondrial function in XP-C cells we performed OCR experiments in intact cells using the XF24 Extracelular Flux Analyzer (SeaHorse Bioscience). Although absolute OCR was increased in unXP-C compared to WT cells (Fig.聽2A), when we corrected by discounting non-mitochondrial respiration, basal respiratory parameters were unchanged (Fig.聽2B). Unexpectedly, unXP-C cells presented an increased maximum respiratory capacity compared to WT (p鈥?lt;鈥?.05), but this did not result in an increased spare respiratory capacity (Fig.聽2B). UnXP-C cells also displayed increased non-mitochondrial respiration. This was not surprising since Rezvani and co-workers22, 23 already demonstrated increased NOX-1 activity in XPC silenced keratinocytes. Although XP-C cells present decreased NADH-linked respiration, the sum of energetic input/output in intact cells suggested that mitochondrial respiration is not altered in XP-C.Figure 2Oxygen consumption is not altered in intact XP-C fibroblasts. (A) O2 consumption was measured in MRC-5 (WT) and XP4PA (unXP-C) as described, using 4鈥壝椻€?04 cells/well in XF Assay Medium containing 10鈥塵M glucose, with the subsequent additions of 1鈥壜礛 oligomycin (ATP synthase inhibitor), 200鈥塶M CCCP (uncoupler) and 1鈥壜礛 rotenone plus AA (CI and CIII inhibitors). The trace represents the mean鈥壜扁€塖D of three independent experiments and a schematic illustration of respiratory parameters. (B) Calculated respiratory parameters: Basal (base line OCR minus rotenone plus AA OCR), ATP (ATP-linked OCR 鈥?base line OCR minus oligomycin OCR), MRC (maximum respiratory capacity 鈥?CCCP OCR minus rotenone plus AA OCR), SRC (spare respiratory capacity 鈥?CCCP PCR minus base line OCR), PL (proton leak 鈥?oligomycin OCR minus rotenone plus AA OCR), NMR (non-mitochondrial respiration 鈥?rotenone plus AA OCR). The data represent mean鈥壜扁€塖D of 3 independent experiments. **p鈥?lt;鈥?.01. Full size image Nuclear-encoded complex II activity is upregulated in XP-C cellsSince we demonstrated that CI activity was decreased in XP-C cells without significant mitochondrial respiratory impairment in intact cells, we measured OCR using CII-linked substrates to check whether CII could be compensating for impaired CI respiration. Mitochondria from XP-C cells energized with succinate, feeding electrons directly into CII, show a 1.7 fold higher OCR than WT mitochondria, from 9.65鈥壜扁€?.13鈥塶mol O2/min/2鈥壝椻€?06 cells in WT to 18.34鈥壜扁€?.59 in unXP-C cells (Fig.聽3A and B). Likewise, uncoupled respiration supported by succinate was also significantly higher in the unXP-C cells. In accordance with the observation that XPC correction reverts XP-C-induced CI downregulation, CII activity also returns to WT levels in corrXP-C cells (WT鈥?鈥?.65鈥壜扁€?.13 vs. corrXP-C鈥?鈥?0.95鈥壜扁€?.60). We also observed a trend toward an increase in CII鈥?鈥塁III activity in crude mitochondrial fractions from unXP-C compared to WT CMF, although it did not reach statistical significance (Fig.聽3C).Figure 3Respiratory complex II, but not complexes III and IV, is upregulated in XP-C cells. (A) OCR was measured using 4鈥壝椻€?06 cells in standard reaction medium. Cells were permeabilized with 30鈥壜礛 digitonin, with the following sequential additions: 5鈥塵M succinate, 0.25鈥壜礛 CCCP, 1鈥壜礛 AA and 200鈥壜礛 TMPD鈥?鈥?鈥塵M ascorbate. The figure shows a representative trace OCR experiment. (B) The rates of oxygen consumption were calculated from data obtained as shown in panel A. (C) CII鈥?鈥塁III activity was measured using 12鈥壜礸 of crude mitochondrial protein. Reduction of cyt c was monitored at 550鈥塶m at 37鈥壜癈 for 5鈥塵in. It was only considered the range where slopes of the reactions were linear. (D) Expression analysis of SDHA and SDHB genes by RT-qPCR. Fold-change in mRNA expression of unXP-C relative to the WT cell line. ACTB was set as reference gene for all genes analyzed. (E) Western blotting of CI subunit SDHB and ATPase subunit F1伪. SDHB protein expression level was normalized against F1伪 in unXP-C relative to corrXP-C cells. The data represent mean鈥壜扁€塖D of 3 independent experiments. *p鈥?lt;鈥?.05 and **p鈥?lt;鈥?.01.Full size image On the other hand, feeding electrons directly into complex IV (CIV) by providing TMPD/ascorbate after adding AA to block electron flow from complex III (CIII) resulted in similar oxygen consumption for all cell lines (Fig.聽3B), indicating that CIV activity is similar in cells containing or lacking functional XPC. Indeed, we could not detect any changes in CIV OCR in corrXP-C cells as well. This was confirmed by measuring cytochrome oxidase (COX) activity in cell extracts, where no differences in COX activity between WT and unXP-C were detected (Suppl. Fig.聽S3).Together, these data show that mitochondria from XP-C cells display an imbalanced electron transport chain, with a significant decrease in CI activity, an increase in CII activity and no change in CIV, suggesting that these cells have deregulated expression and/or assembly of the OXPHOS complexes. Accordingly, mRNA expression of CII subunits, SDHA and SDHB, were significantly higher in both transformed (unXP-C) and primary (XP17VI) cells when compared to their WT controls (MRC-5 and AS405, respectively) (Fig.聽3D). Moreover, SDHB protein level was also increased in unXP-C compared to corrXP-C (Fig.聽3E). It is noteworthy that CII is the only OXPHOS complex that is exclusively nuclear-encoded, while CI is the OXPHOS complex with the largest number of mtDNA-encoded subunits.Increased mitochondrial H2O2 production is associated with decreased GPx activity in XP-C cellsIn most aerobic cell types, mitochondria represent a quantitatively important ROS generation site25. Because we found impaired CI activity together with increased mtDNA damage and deletion, we measured O2 鈥⑩垝 and hydrogen peroxide (H2O2) production in WT, unXP-C and corrXP-C cells using fluorescent probes and assay conditions that allowed us to discriminate mitochondrial vs. non-mitochondria generation. Antimycin A-induced (AA) mitochondrial O2 鈥⑩垝 production, measured in presence of the NOX inhibitor diphenyliodonium (DPI), was not altered in any of the three cell lines (Fig.聽4A). As expected, mild-treatment with the mitochondrial uncoupler CCCP decreased MitoSOXTM signal in all cell lines (not shown). On the other hand, H2O2 generation was significantly higher in unXP-C cells, from 5.7鈥壜扁€?.9鈥塸mols/min in WT cells to 9.0鈥壜扁€?.1鈥塸mols/min in the unXP-C cells, even in the presence of DPI and AA (Fig.聽4B). The increased H2O2 generation was dependent on XPC deficiency, as corrXP-C cells produced similar levels of H2O2 as WT cells (Fig.聽4B).Figure 4Redox imbalance in XP-C cells due to mitochondrial ROS production. (A) Superoxide generation was measured by flow cytometry using 1鈥壝椻€?05 cells incubated with 5鈥壜礛 Antimycin A, 5鈥壜礛 DPI and 5鈥壜礛 MitoSOX庐. (B) H2O2 production was followed spectrofluorometrically with 5鈥壜礛 Amplex UltraRed庐, in presence of 5鈥壜礛 Antimycin A, 5鈥壜礛 DPI. Rate of H2O2 production calculated after calibration performed with known H2O2 concentrations. (C) SOD activity was measured using the Abcam kit, following the manufacturer鈥檚 instructions. (D) GPx activity was measured using 100鈥壜礸 of total cellular protein. Oxidation of NADPH to NADP+ was monitored at 340鈥塶m on a temperature-controlled spectrophotometer, at 25鈥壜癈. The results presented are the mean鈥壜扁€塖D of n鈥?鈥? for superoxide production, n鈥?鈥? for H2O2 production, n鈥?鈥? for SOD and n鈥?鈥? for GPx activity. *p鈥?lt;鈥?.05 and **p鈥?lt;鈥?.01.Full size image The increased H2O2 production in XP-C cells was not directly linked to a higher O2 鈥⑩垝 dismutation rate, as total SOD activity was similar in all cell lines (Fig.聽4C). However, GPx activity inversely correlated to H2O2 production, with a significant decrease in GPx in unXP-C cells, which is partially corrected after XPC expression in corrXP-C cells (Fig.聽4D). Since SOD activity was unchanged, the elevated H2O2 level is likely the result of a decreased rate of H2O2 removal by GPx and other detoxifying systems, contributing to a possible role of GPx for an unbalanced mitochondrial redox status (Fig.聽4D). Together, these results indicate that unXP-C cells are under redox imbalance, likely due to mitochondrially-generated H2O2.XP-C cells accumulate irreversible mtDNA damagemtDNA is physically associated with the inner mitochondrial membrane, and thus a primary target of mitochondrial-generated ROS26. Mitochondrial CI is the ETC complex with the largest number of subunits encoded by mtDNA, with 7 of the 13 mtDNA encoded polypeptides being components of CI (MT-ND1, 2, 3, 4L, 4, 5 and 6). Because CI is impaired in XP-C cells and XP-C mitochondria are under a redox imbalance, we asked whether mtDNA is preferentially damaged and/or deleted. Using a long-extension PCR assay, which measures the inhibition of amplification of a large fragment (16鈥塳b for the mtDNA and 10鈥塳b for nDNA) by polymerase blockage at DNA lesions in the template, we found that the mtDNA is significantly more damaged in unXP-C cells when compared to WT normal cells, displaying an efficiency of amplification of only 20% that of the normal cells (Fig.聽5A, left bars). On the other hand, in nuclear DNA, we detected only a small decrease in amplification efficiency in the unXP-C cells, which was not statistically significant (Fig.聽5A, right bars). Together, these results indicate that, in cells lacking XPC, even under unstressed conditions, mtDNA accumulates significant amounts of DNA damage, while nuclear DNA is only marginally damaged.Figure 5Accumulation of DNA damage in mtDNA in XP-C cells. DNA damage was measured by XL-PCR amplification of a 16.3鈥塳bp fragment of the mtDNA (A, left, and C) and a 10.4鈥塳bp fragment of the HPRT gene, for the nuclear genome (A, right). Data is presented as ratio of bands intensity of long fragment over small fragment loading control (MT-ND1 and HPRT to mt and nDNA, respectively) relative to WT. The data represent mean鈥壜扁€塖D in transformed and primary cells (n鈥?鈥?). (B) MtDNA copy number and MT-ND4L frequency was assessed as the relative amplification of MT-ND1 (mtDNA gene) and HPRT (nDNA gene), and of MT-ND4L gene (frequently deleted) vs. MT-ND1 gene (rarely deleted), as described. WT and AS405 gene expression fold change are implicit and were set to 0.0 (zero) as control reference. The data represent mean鈥壜扁€塖D in transformed (WT, unXP-C, corrXP-C and XP-A, n鈥?鈥?) and primary (AS405 and XP17VI, n鈥?鈥? and n鈥?鈥?, respectively) cells *p鈥?lt;鈥?.05, **p鈥?lt;鈥?.01 and ***p鈥?lt;鈥?.001.Full size image Moreover, mitochondria from unXP-C cells had more deleted mtDNA molecules than mitochondria from normal cells, as the ratio of amplification of a region of the mitochondrial genome which is frequently deleted in humans, at the MT-ND4L gene, to a region which is seldom deleted, the MT-ND1 gene27, was significantly lower in the unXP-C than in WT (Fig.聽5B, right bars). It is noteworthy that this observed alteration was not due to decreased mtDNA copy number since we could not detect changes in the relative amplification of mtDNA (MT-ND1) vs. nDNA (HPRT) between unXP-C relative to WT cells (Fig.聽5B, left bars).To make sure that the accumulation of mtDNA damage was indeed a result of absence of XPC, we used the unrelated cell line XP17VI, a XP-C derived primary fibroblast, and AS405, a normal primary fibroblast cell line. XP17VI cells also showed accumulation of mtDNA damage (Fig.聽5C) and deletions when compared to the normal individual, AS405, since the ratio of MT-ND4L/MT-ND1 amplification was significantly lower (Fig.聽5B, right bars). In this cell line, however, we detected a significant decrease in mtDNA copy numbers when compared to its control (AS405, Fig.聽5B left bars).However, to our surprise, mtDNA damage and MT-ND4L frequency were unchanged in corrXP-C cells compared to isogenic unXP-C cells (Fig.聽5A and B). These results indicate that mtDNA integrity is irreversibly compromised in the XPC-deficient cells and cannot be reversed by the expression of wild type XPC in this background. Furthermore, this phenomenon seems to be specifically related to XPC deficiency, since no alteration in MT-ND4L frequency in XP-A relative to WT cell line was detected (Fig.聽5B, right).XPC does not localize to mitochondria in unstressed cellsSince the XPC protein has been directly linked to repair of oxidized DNA bases via a physical interaction with OGG110, and we detected significant accumulation of mtDNA damage in the XP-C cells, we investigated whether XPC localized in mitochondria. For that, we isolated highly purified mitochondria from WT and unXP-C cells, as demonstrated by the absence of Lamin B2 in the mitochondrial extracts (Fig.聽6A). Those extracts were then used to check for XPC presence by western blots. As shown in Fig.聽6B, XPC was detected in nuclear but not in mitochondrial extracts from WT cells, while no XPC signal was detected in either nuclear or mitochondrial extracts of unXP-C, since the transcript is unstable in these cells28. The purity of the extracts was again confirmed by probing for mitofusin 1 (mitochondrial protein) and Lamin B2 (nuclear protein). The results demonstrate that, under normal growth conditions, the XPC protein is mainly localized in the nucleus and does not localize in mitochondria. These observations suggest that the mitochondrial dysfunction detected in unXP-C cells is due to a role of XPC that is independent of its localization inside the organelle.Figure 6Expression and localization of the XPC protein in the cell. (A) Purity of mitochondrial extracts was checked by western blotting, using 10鈥壜礸 of nuclear (N) or mitochondrial (M) fractions from MRC-5 (WT) or XP4PA (unXP-C) per lane. Lamin B2 was used as nuclear marker and ATP5B as mitochondrial. Mitochondrial localization of XPC was analyzed by western blotting. (B) Western blotting of mitochondrial and nuclear fractions used 40鈥壩糶 of protein and anti-XPC, anti-Mfn1 as mitochondrial marker and anti-Lamin B2 as nuclear marker.Full size image NRF1 and SIRT3 are upregulated in XP-C cellsBecause ETC subunits and other redox-related mitochondrial proteins are regulated through transcriptional programs and post-translational modifications, we checked expression levels of critical genes involved in mitochondrial biogenesis and modulation of mitochondrial metabolism. Two genes, nuclear respiratory factor 1 (NRF1) and sirtuin 3 (SIRT3), were found to be upregulated in transformed unXP-C and primary XP17VI comparing to their WT counterpart (Fig.聽7A for transformed, and 7B for primary cells). These results indicate that lack of XPC leads to upregulation of transcription factor NRF1 and mitochondrial deacetylase SIRT3 in both transformed and untransformed fibroblasts.Figure 7Expression analysis of mitochondrial genes by RT-qPCR. Fold-change in mRNA expression of XP4PA (unXP-C) relative to the wild type cell line (MRC-5). ACTB was set as reference gene for all genes analyzed. WT gene expression fold change and were set to 0.0 (zero) as control reference. The data represent mean鈥壜扁€塖D of 4 independent experiments. *p鈥?lt;鈥?.05, **p鈥?lt;鈥?.01 and ***p鈥?lt;鈥?.001.Full size image XP-C cells are more sensitive to mitochondrially-induced oxidative stressAlthough total mitochondrial respiration was not impaired in XP-C cells, we hypothesized that the unbalanced redox status and mitochondrial ETC adaptive shift could sensitize mitochondria from cells lacking XPC to stress. To test that, we used the CIII inhibitor AA, which efficiently traps electron flow through the ETC. Consequently, there is an increase in one electron reduction of O2 generating O2 鈥⑩垝 and increased mitochondrial oxidative stress29. We also challenged cells with MB-induced singlet oxygen (1O2) using a treatment protocol that excludes MB from the nucleus. Because 1O2 is able to generate 8-oxoGua via endoperoxide formation across C4-C5 and N7-C8 double bounds of guanine30, this treatment induces oxidative DNA damage accumulation exclusively in the mtDNA.As seen in Fig.聽8A, the unXP-C cell line was more sensitive than WT at the two highest concentrations of AA (WT鈥?鈥?4.0%鈥壜扁€?.4 vs. unXP-C鈥?鈥?2.7%鈥壜扁€?.6 and WT鈥?鈥?4.8%鈥壜扁€?.3 vs. unXP-C鈥?鈥?6.5%鈥壜扁€?.9 for AA 1.0 and 2.0鈥壩糓, respectively) (p鈥?lt;鈥?.05). Although not statistically significant, cytotoxicity at 0.25 and 0.50鈥壩糓 of AA was more pronounced in the unXP-C cells as well. These cells were also more sensitive to photoactivated MB at the highest concentration, 20鈥壩糓 (WT鈥?鈥?3.9%鈥壜扁€?.2; unXP-C鈥?鈥?0.2%鈥壜扁€?.2) (p鈥?lt;鈥?.01) (Fig.聽8B). The presence of XPC protein in corrXP-C cells was sufficient to protect cells from the toxic effects of mitochondrial oxidative stress induced by AA at all concentrations tested, suggesting that the ETC shift and redox imbalance observed in the absence of XP-C, which is reverted in the corrected cell line, may contribute to the increased sensitivity to AA. However, XPC expression in the corrXP-C cell line failed to revert the cytotoxicity induced by MB鈥?鈥塴ight (Fig.聽8B). Taken together, these results indicate that the lack of XPC protein sensitizes cells against mitochondrial respiratory chain blockage, but not singlet oxygen toxicity.Figure 8Absence of XPC sensitizes cells to AA and MB-induced cell death. Survival rates of cells treated with (A) antimycin A for 4鈥塰 or with (B) methylene blue subjected to photoactivation during 30鈥塵in. After treatments, individual cells were seeded at 500 cells/plate and allowed to grow to visible colonies. Survival was calculated relative to the plating efficiency of untreated cells. Values represent the mean鈥壜扁€塻.e.m. of 4 independent experiments. *p鈥?lt;鈥?.05 and **p鈥?lt;鈥?.01.Full size image DiscussionSeveral recent evidences have indicated that DNA repair deficiency is mechanistically linked to mitochondrial, and hence metabolic, dysfunction. Most notably, in XP-A19, AT16,17,18 and CS cells and animal models12, 13, 31, there is significant redox imbalance and increased metabolic rate due to deficient mitophagy. A link between defective DNA repair, tumorigenesis and mitochondrial dysfunction is also strengthened by results in an AT mouse model, where preventing mitochondrial dysfunction delayed tumor development17. In the CS and AT, however, a fraction of the proteins mutated in these syndromes, CSA and CSB in case of CS, and ATM in AT, localize to mitochondria and their absence partially explain mitochondrial symptoms32. These effects could be, at least in part, attributed to a direct role of the proteins in maintaining mtDNA integrity18. In the case of XP proteins (A-G), however, mitochondrial localization has been reported, so far, only for XPD. In fact, we show here that XPC cannot be detected in mitochondrial extracts, at least in unstressed cells (Fig.聽6). Nonetheless, some XP symptoms, notably the neurodegeneration phenotype, overlap with classical mitochondrial diseases33 and impaired oxidative DNA damage repair has been implicated in XP etiology21. Moreover, mitochondrial dysfunction and accumulation of mtDNA damage are strongly associated with tumor development21, a hallmark feature of XP.Recently, Rezvani and coworkers22, 23 demonstrated that XPC silencing in keratinocytes led to NOX-1 activation and cytosolic H2O2 production, suggesting that mitochondrial dysfunction is secondary to these events. However, we show here that mitochondria are a primary source of H2O2 production in XP-C cells, even in presence of a NOX-1 inhibitor (Fig.聽4). This increased mitochondrial H2O2 production is accompanied by decreased GPx activity, indicating an intracellular redox imbalance. In addition, the mtDNA is significantly more damaged than the nuclear DNA in these cells, even in cells not exposed to any DNA damaging agent (Fig.聽5). Using a long-extension PCR method, which detects the presence of any polymerase-blocking lesion, we detected a significant accumulation of lesions in mtDNA and only a small, non-significant level of lesions in the nDNA from XP-C cells. Because the mtDNA is physically associated with the mitochondrial inner membrane, this result also supports the hypothesis that mitochondria are a primary source of ROS in XP-C cells and that mtDNA is a primary target for damage accumulation. In contrast, the low levels of nDNA lesions suggest that, at least under unstressed conditions, the nDNA repair defect does not contribute significantly to the cellular phenotypes observed in the XP-C cells. Moreover, we detected higher levels of mtDNA deletions both in transformed and in primary XP-C cells. MtDNA deletions are suggested to result from incomplete repair events34; thus, this observation also supports the idea that the mtDNA is constantly being damaged by mitochondrially-generated H2O2, like hydroxyl radical. This is in agreement with the observation that XPC correction does not revert mtDNA damage, since the mtDNA pool in the XP-C cells has already been irreversibly damaged due to the exposure to H2O2 before the cells were corrected. Thus, we can speculate that lesions accumulated in mtDNA and mutations that arise from them could be used as biomarkers of cell exposure to chronic imbalanced redox status, even though the redox balance has been restored.Several authors propose that accumulation of mutations in mtDNA plays an important role in the tumorigenic process, including skin cancer associated with exposure to UV35, the leading type of cancer observed in XP-C patients. On the other hand, recent results from large-scale sequencing of individual mitochondrial genomes from colorectal cancer patients showed lower mtDNA mutational burdens in cancer samples compared to normal samples36. These authors suggest that a shift from aerobic to glycolytic metabolism during tumor progression decreases ROS generation, thus 鈥減rotecting鈥?the mtDNA from oxidative damage induced mutagenesis. However, they do not rule out the possibility that primary mtDNA mutations contribute to the metabolic changes observed during the process.Since XPC does not localize to mitochondria in non-stressed conditions, it is unlikely that mtDNA damage accumulation results from impaired repair, making it, more likely, an event secondary to increased mitochondrial ROS generation. The relevant question, thus, becomes how the lack of XPC leads to increased mitochondrial ROS generation? In this regard, we detected a pronounced decrease in CI-linked oxygen consumption and activity, associated with an increase in CII-linked respiration. This CI/CII shift cannot be directly explained by expression levels, since subunits of both CI and CII were found to be upregulated in XP-C cells (Figs聽S2A and 3D,E). Surprisingly, a recent study conducted Hosseini and coworkers found the opposite effect in Xpc null mouse keratinocytes. However, in that model downregulation of NADH:ubiquinone oxireductase, succinate dehydrogenase and cytochrome c oxidase was linked to premature skin aging, not to cellular transformation37. CI is believed to produce O2 鈥⑩垝 at two sites, the flavin mononucleotide (FMN) and the ubiquinone (UQ) binding sites, thence CI and III are known sources of ROS formation29. In isolated mitochondria, a reverse electron flow from reduced UQ to CI has been detected20. On the other hand, succinate dehydrogenase complex is more likely to prevent ROS formation due to the intrinsic arrangement of its redox centers38. The assembly of CI into the mature holoenzyme involves the coordinated expression of the two genomes, plus a large group of proteins known as 鈥渁ssembly factors鈥?sup>39. Mutations in genes of both structural and/or catalytic proteins, as in mounting factors, result in loss of activity of the complex. Even when the catalytic activity is preserved, as in the case of Leber鈥檚 hereditary optic neuropathy (LHON) caused by mutations in three mitochondrial genes of CI, there is an increase of sensitivity to inhibitors of CI respiration40, and associated H2O2 generation.Although ROS have been classically linked to cellular decay, recent attention has been given to their functions in cell signaling. Consistently, H2O2 lower reactivity (when compared to radical species, like 鈥?/sup>OH) added to its relatively high diffusion rate prompted this species to be considered the main ROS in cell signaling41. Accordingly, Piantadosi Suliman demonstrated that low levels of oxidative stress trigger nuclear respiratory factor 1 (NRF1) phosphorylation and cytosol to nucleus translocation42. NRF1 is a major transcription factor that activates the expression of some key metabolic genes, regulating cellular growth and the expression of nuclear genes required for respiration, heme biosynthesis, and mitochondrial DNA transcription and replication. The same group demonstrated that NRF1 is recruited to two promoters of CII subunits, SDHA and SDHD43. Moreover, ChIP-seq analysis revealed not only SHDA, B and D as NRF1 target genes, but also from all five ETC complexes and some DNA repair pathways, more specifically BER and non-homologous end joining44. Indeed NRF1, SDHA and SDHB expression levels are increased in unXP-C cells and unrelated XP17VI primary fibroblasts compared to their WT counterpart (Figs聽3D,E and 7).In keratinocytes, Han and coworkers demonstrated a role of NRF1 linking redox homeostasis through promotion of enzymes involved in glutathione synthesis and UVB-induced DNA repair via GGR, more specifically, XPC expression. The expression of this DNA damage recognizing protein was significantly reduced upon NRF1 knockdown, as well as GSH/GSSG levels. Interestingly, treatment with cell permeable glutathione (GSH) was sufficient to restore XPC expression even in the absence of NRF1, demonstrating that redox homeostasis in vital to XPC transcription in keratinocytes45.While CII is not classically linked to ROS generation, mutations in some subunits of this complex give rise to cancer-prone diseases46. CII dysfunction leads to succinate accumulation with subsequent prolyl hydroxylase inhibition allowing stabilization of HIF1伪, which may act as an oncogene47. Furthermore, cellular studies expressing mutant SDHC (a subunit of CII) show that CII dysfunction gives rise to sub-lethal doses of ROS, promoting genomic instability and cell proliferation that favors tumorigenesis48, 49. More recently, a role for CII in ROS sensing and generation in apoptosis has also been proposed. CII displays two distinct enzymatic activities i) succinate dehydrogenase (SDH) and ii) succinate ubiquinone oxireductase (SQR)50. During the course of apoptosis, SDHA and B, which encompass the SDH activity, are dislodged from the membrane-anchored complex. SDH activity continues to operate removing electrons from succinate to FAD and Fe-S clusters in released SDHA/B, but apart from SDHC/D, the membrane-bond ubiquinone oxireductase site, O2 undergoes one-electron reduction producing O2 鈥⑩垝, overflowing the cell with ROS culminating with apoptosis50,51,52. Thus, it is possible that under conditions in which CI and CII are unbalanced, like that seen in the XP-C cells, CII activity contributes to chronic H2O2 generation and genomic instability, and consequently to driving the tumorigenic process.Together, these results indicate that XPC deficiency leads to mitochondrial-derived ROS formation as result of impaired CI activity, the major ETC complex encoded by mtDNA. Increased H2O2 production may trigger NRF1 leading to CII upregulation, as indicated by the increased succinate-linked OCR and CII鈥?鈥塁III activity. It is important to note that the CI/CII shift is completely dependent on the lack of XPC, as expression of the wild type protein reverts the phenotypes. As a consequence of this adaptive shift between CI and CII fueling preference, XP-C cells may counteract mitochondrially-generated ROS reducing the impact of imbalanced redox status, but at the cost of sensitivity to mitochondrial oxidative stress, considering that CII might have an important role as a sensor and an effector for apoptosis.Material and MethodsCell lines and cultureThe human SV40-transformed fibroblast cell lines MRC-5, XP4PA, corrected XP4PA and XP12RO cell line (hereinafter WT, unXP-C, corrXP-C and XP-A), and the human primary fibroblasts cell lines AS405 (wilt type) and XP17VI (XP-C) were kindly provided by Prof. Dr. Carlos F. M. Menck, ICB, University of S茫o Paulo. Both XP-C cell lines are homozygous for the prevalent mutation c.1643-1644delTG (1744鈥?745) in exon 9. All cell lines were maintained in DMEM/high glucose supplemented with 10% FBS, penicillin 100 IU/mL and streptomycin 100鈥壩糶/mL, at 37鈥壜癈 in a humidified atmosphere with 5% CO2 (standard conditions). Cultures were routinely sub-cultured, by trypsinization, when reached up to 80% confluence.Determination of O2 consumptionO2 consumption rates (OCR) of WT, unXP-C and corrXP-C cells (2鈥壝椻€?06 cells/mL) were measured using a Hansatech electrode (Hansatech Instruments Ltd.) and the high-resolution respirometry (HRR) O2k-core (Oroboros). Cells were permeabilized with 30鈥壜礛 digitonin in 2鈥塵L of standard reaction medium (125鈥塵M sucrose, 65鈥塵M KCl, 10鈥塵M HEPES, 2鈥塵M K2HPO4, 1鈥塵M MgCl2, pH 7.2) containing 0.33鈥塵M EGTA, 5鈥塵M CI-linked substrate cocktail (3.4鈥塵M malate, 1.86鈥塵M 伪-ketoglutarate, 2.1鈥塵M pyruvate, 2.1鈥塵M glutamate). Sequential additions of 100鈥壜礛 ADP, 4鈥壜礛 CAT and 0, 5鈥壜礛 CCCP were made to analyze respiratory states 2, 4 and uncoupled respiration, respectively. CII, III and IV OCR were also measured using HRR O2k-core with the same 2鈥塵L standard reaction medium. Cells were permeabilized with 30鈥壜礛 digitonin, with the following sequential additions: 5鈥塵M succinate, 0.25鈥壜礛 CCCP, 1鈥壜礛 AA and 200鈥壜礛 TMPD鈥?鈥?鈥塵M ascorbate. All analyses were made at 37鈥壜癈.In order to evaluate more physiological respiratory parameters, OCR was measured in intact cells using XF24 Extracellular Flux Analyzer (Seahorse Bioscience). Approximately 4鈥壝椻€?04 cells per well were seeded in XF24 V7 Cell Culture Microplates at standard culture conditions. After 12鈥塰, DMEM was washed off with 1鈥塵L of XF Assay Medium supplemented with 10鈥塵M D-glucose and cells were incubated in humidified CO2-free incubator at 37鈥壜癈 for 1鈥塰.Respiratory parameters were analyzed based in the manufacturer鈥檚 protocol for the Seahorse XF Cell Mito Stress Test Kit with slight modifications. Sequential additions of 1鈥壜礛 oligomycin, 200鈥塶M CCCP and 1鈥壜礛 rotenone plus AA were made to analyze basal respiration, ATP production, proton leak, maximum respiratory capacity, spare respiratory capacity and non-mitochondrial respiration.DNA damage quantification by XL-PCRDNA damage in the mitochondrial or nuclear DNA was estimated measuring PCR amplification efficiency of a long fragment of each genome (over 10 kilobases)53. DNA samples were isolated using the DNeasy Blood Tissue Kit庐 (Qiagen), according to the manufacturer鈥檚 instructions. Long-extension PCR (XL-PCR) reactions were performed as described54, with minor modifications. Cycling parameters were modified to include an annealing step of 30鈥塻 at 64鈥壜癈, and an increase in the denaturing step to 30鈥塻.DNA amplicons were submitted to electrophoretical separation in 0.8% agarose gel at 5鈥塚/cm for 4鈥塰 in TAE buffer (40鈥塵M Tris-acetate pH 7.8, 1鈥塵M EDTA). Gels were stained in 2鈥塵g/L ethidium bromide for 30鈥塵in and distained in deionized water for 10鈥塵in. Images were captured and analyzed by ImageJ (NIH). Equal DNA template loading was corrected by 螖Ct of qPCR for mtDNA copy number, as described below. We chose to use qPCR instead gel electrophoresis densitometry of small mtDNA as control due to smaller variation between samples, although means were comparable (not shown).Determination of mtDNA copy number and deletion levelsDNA samples were isolated using DNeasy庐 Blood Tissue Kit (Qiagen) according to the manufacturer鈥檚 instructions. Real time quantitative PCR (qPCR) was performed using SYBR Green PCR Master Mix (Applied Biosystems) according to the manufacturer鈥檚 instructions. Data analysis was performed by 螖螖Ct method and adjusted to fold-change. The following genes were analyzed by qPCR: HPRT for nDNA, MT-ND1 for mtDNA copy number and MT-ND4L for mtDNA deletion. MT-ND4L was chosen as a deletion marker due to the high susceptibility of the genomic region in which this gene is located27, 55. The primer sequences are presented in Supplemental Table聽1A.Mitochondrial isolation from cell culturesMitochondria were isolated using differential centrifugation and Percoll gradient, as described56.Western blottingNuclear and mitochondrial fractions were analyzed by Western blotting using standard techniques. Incubations with primary antibodies were performed in 1% non-fat dry milk in TBS-T, either 1鈥?鈥塰 at room temperature or overnight at 4鈥壜癈, with the following antibodies: anti-actin, N-terminal (1:2000; rabbit A2103, Sigma), anti-Lamin B2 (X223) (1:1000; mouse sc-56147, Santa Cruz Biotechnology), anti-ATPB antibody [3D5] (1:1000; mouse ab-14739, Abcam), anti-Mfn1 (H-65) (1:500; rabbit sc-50330, Santa Cruz Biotechnology), anti-XPC (1:400; rabbit ab21078 or 1:250; [3.26] mouse ab6264, Abcam), anti-NDUFS4 (2C7CD4AG3) (1:X; mouse ab87399, Abcam) and Complex II Western Blot Antibody Cocktail (1:X; mouse MS202, MitoSciences). After washing with TBS-T, appropriate secondary antibodies linked to horseradish peroxidase were applied at 1:5000 (bovine anti-goat, goat anti-rabbit or goat anti-mouse, Santa Cruz Biotechnology) in 1% milk in TBS-T. Membranes were then washed repeatedly with TBS-T and visualized using Amersham ECL Plus (GE Healthcare).Mitochondrial Superoxide ProductionFor quantification of mitochondrial superoxide generation, approximately 3鈥壝椻€?05 cells were seeded 48鈥塰 prior to the experiment in 60鈥塩m2 dishes (TPP). WT, unXP-C and corrXP-C cells were washed in pre-warmed PBS, trypsinized and centrifuged for 5鈥塵in at 300鈥?i>g at 37鈥壜癈. The cell pellets were suspended in 0.22鈥壩糾 filtered Hank鈥檚 Balanced Salt Solution (HBSS) supplemented with D-glucose 1鈥塯/L in an approximate final concentration of 1鈥壝椻€?06 cells/mL. Cells were incubated for 30鈥塵in at 37鈥壜癈 with: (i) 5鈥壜礛 MitoSOX (Molecular Probes); (ii) 5鈥壜礛 MitoSOX, 5鈥壜礛 antimycin A (AA) (CIII inhibitor) and 5鈥壜礛 DPI (diphenylene iodonium 鈥?to inhibit NOX-generated superoxide formation); and (iii) 5鈥壜礛 MitoSOX and 500鈥塶M CCCP (for uncoupling). Fluorescence intensity was analyzed by flow cytometry, in the FL-2 channel as described56. Results shown are relative to treatment ii) with MitoSOX鈥?鈥堿A鈥?鈥塂PI. TMRM controls were also performed (data not shown).H2O2 ProductionHydrogen peroxide production in WT, unXP-C and corrXP-C cells was determined fluorimetrically following the oxidation of Amplex UltraRed (Molecular Probes) to the highly fluorescent compound, resorufin57. For that, cells were incubated in HBSS containing 5鈥壜礛 Amplex UltraRed and 0.1鈥塙/mL horseradish peroxidase (HRP). Fluorescence was monitored over time in SpectraMax 190 reader (Molecular Devices) in 96-well plate format using excitation and emission wavelengths of 530 and 590鈥塶m respectively, with low sensitivity. Under these conditions, a linear increase in fluorescence indicates an increase in H2O2 formation. Calibration curves were made with known concentrations of H2O2.Measurements of enzyme activitiesGlutathione peroxidase (GPx) activity was determined by measuring the rate of formation of oxidized glutathione form reduced GSH in the presence of H2O2, detected by the change in absorbance at 340鈥塶m due to NADPH oxidation, as previously described58.Superoxide dismutase (SOD) activity was measured using the SOD activity colorimetric assay kit庐 (Abcam) according to the manufacturer鈥檚 protocol, in a microplate at 37鈥壜癈 at 450鈥塶m.CI鈥?鈥塈II and CII鈥?鈥塈II activities was measured as previously described59, 60 with slight modifications. Reactions were adjusted to fit in 96-well microplate platform. NADH cytochrome c oxidoreductase and succinate cytochrome c reductase activities were acquired in SoftMax Pro (Molecular Devices) and expressed as the rate of reduced cytochrome c formation per min per mg of protein.Cytochrome c oxidase activity was measured using a protocol based on a commercial kit (Sigma-Aldrich), which follows the decrease in absorbance at 550鈥塶m due to the oxidation of reduced cytochrome c. Enzyme activity was calculated using the difference in reduced and oxidized cytochrome c molar absorptivity coefficients of 21.84鈥塵M鈭?鈥壝椻€塩m鈭?.Enzyme activities are expressed relative to the total protein content, determined by Bradford (1976), using bovine gamma globulin (Bio-Rad) as standard.Clonogenic assayClonogenic cell survival is regarded as gold standard test to detect cellular toxicity鈥?/sup>. Briefly, 1.0鈥壝椻€?05 cells were seeded in 6-well plate (TPP) and incubated for 24鈥塰, at standard conditions. For MB treatment, cells were rinsed with PBS and incubated for 1鈥塰 with MB concentrations of 5.0, 10.0 and 20.0鈥壩糓 diluted in DMEM without FBS. All further steps were carried out in a dark room aided with blue lamp. At the end of the incubation, cells were washed twice with PBS and incorporated MB was photoactivated with a 634鈥塶m LED lamp for 30鈥塵in. Cells were then trypsinized, counted and approximately one thousand cells were seeded in 60鈥塵m Petri dishes (Nunc, Thermo-Scientific) in triplicate. Cells were incubated in standard conditions for 8 days. For treatment with AA, 500 cells were plated in 6-well plate (TPP) in triplicate and incubated for 24鈥塰 in standard conditions. Cells were washed twice with PBS and incubated with AA at 0.25, 0.50, 1.0 and 2.0鈥壩糓 in DMEM without FBS for 4鈥塰. After that, AA was washed out with PBS and cells were incubated in complete medium in standard conditions for 6 to 7 days. Colonies were washed in ice-cold PBS, ethanol-fixed and stained with 1% (w/v) crystal violet solution. Survival rate was calculated as the ratio of number of colonies in treated over non-treated conditions.Analysis of gene expressionTotal RNA was extracted using RNeasy Micro Kit (Qiagen), following the manufacturer鈥檚 instructions. The expression levels of MT-ND1, MT-ND4L, NDUFS1, NRF1, NFE2L1 (NRF2), PPARA (PPAR伪), PRKAA (AMPK), SIRT1, SIRT3, SDHA and SHDB mRNA were measured by quantitative RT-qPCR assays, with primers targeting the mRNAs expressed by these genes, and 尾-actin (ACTB) as the loading control. Three other housekeeping genes were compared to ACTB expression (TUBB, TBP and HPRT) none of which showed significant difference (not shown). Total RNA extracted from WT, unXP-C, corrXP-C, AS405 and XP17VI cells was used to generate cDNA with the High Capacity cDNA Reverse Transcription kit (Applied Biosystems). Quantitative RT-PCR reactions were performed using the Power SYBR Green PCR Master Mix (Applied Biosystems), according to directions, using 100鈥塶g template and 5鈥塸mol of each primer. Cycling, reading and data analyses were performed as described above. The primer sequences are presented in Supplemental Table聽S2.Statistical analysesResults are shown as mean鈥壜扁€塻tandard deviation (SD) or standard error of the mean (s.e.m.), or as representative traces of at least 3 independent experiments. The number of repeats is stated in the figure legends. ANOVA with Tukey regression was used in WT, unXP-C and corrXP-C comparison, while Student鈥檚 t test was used to assess differences between two groups, considering significant p鈥?lt;鈥?i>0.05. References1.Hanahan, D., Weinberg, R. A. Francisco, S. The hallmarks of cancer. Cell 100, 57鈥?0 (2000).CAS聽 Article聽 PubMed聽Google Scholar聽 2.Cleaver, J. E. Defective Repair Replication of DNA in xeroderma pigmentosum. Nature 218, 652鈥?56 (1968).ADS聽 CAS聽 Article聽 PubMed聽Google Scholar聽 3.DiGiovanna, J. J. Kraemer, K. H. Shining a light on xeroderma pigmentosum. J. Invest. Dermatol. 132, 785鈥?6 (2012).CAS聽 Article聽 PubMed聽 PubMed Central聽Google Scholar聽 4.Niedernhofer, L. J., Bohr, V. A., Sander, M. Kraemer, K. H. Xeroderma pigmentosum and other diseases of human premature aging and DNA repair: Molecules to patients. Mech. Age. Develop. 132, 340鈥?47 (2011).CAS聽 Article聽Google Scholar聽 5.Sch盲rer, O. D. Nucleotide excision repair in eukaryotes. Cold Spring Harb. Perspect. Biol. 5, a012609 (2013).Article聽 PubMed聽 PubMed Central聽Google Scholar聽 6.Cleaver, J. E., Lam, E. T. Revet, I. Disorders of nucleotide excision repair: the genetic and molecular basis of heterogeneity. Nat. Rev. Genet. 10, 756鈥?8 (2009).CAS聽 Article聽 PubMed聽Google Scholar聽 7.Horibata, K. et al. Complete absence of Cockayne syndrome group B gene product gives rise to UV-sensitive syndrome but not Cockayne syndrome. Proc. Natl. Acad. Sci. USA 101, 15410鈥?5415 (2004).ADS聽 CAS聽 Article聽 PubMed聽 PubMed Central聽Google Scholar聽 8.Nardo, T. et al. A UV-sensitive syndrome patient with a specific CSA mutation reveals separable roles for CSA in response to UV and oxidative DNA damage. Proc. Natl. Acad. Sci. USA 106, 6209鈥?4 (2009).ADS聽 CAS聽 Article聽 PubMed聽 PubMed Central聽Google Scholar聽 9.Stevnsner, T. et al. Mitochondrial repair of 8-oxoguanine is deficient in Cockayne syndrome group B. Oncogene 21, 8675鈥?2 (2002).CAS聽 Article聽 PubMed聽Google Scholar聽 10.D鈥橢rrico, M. et al. New functions of XPC in the protection of human skin cells from oxidative damage. EMBO J. 25, 4305鈥?5 (2006).Article聽 PubMed聽 PubMed Central聽Google Scholar聽 11.Bernardes de Jesus, B. M., Bj酶r氓s, M., Coin, F. Egly, J. M. Dissection of the molecular defects caused by pathogenic mutations in the DNA repair factor XPC. Mol. Cell. Biol. 28, 7225鈥?5 (2008).CAS聽 Article聽 PubMed聽 PubMed Central聽Google Scholar聽 12.Aamann, M. D. et al. Cockayne syndrome group B protein promotes mitochondrial DNA stability by supporting the DNA repair association with the mitochondrial membrane. FASEB J. 24, 2334鈥?6 (2010).CAS聽 Article聽 PubMed聽 PubMed Central聽Google Scholar聽 13.Kamenisch, Y. et al. Proteins of nucleotide and base excision repair pathways interact in mitochondria to protect from loss of subcutaneous fat, a hallmark of aging. J. Exp. Med. 207, 379鈥?0 (2010).CAS聽 Article聽 PubMed聽 PubMed Central聽Google Scholar聽 14.Pascucci, B. et al. An altered redox balance mediates the hypersensitivity of Cockayne syndrome primary fibroblasts to oxidative stress. Aging Cell 11, 520鈥?29 (2012).CAS聽 Article聽 PubMed聽Google Scholar聽 15.Scheibye-Knudsen, M. et al. Cockayne syndrome group B protein prevents the accumulation of damaged mitochondria by promoting mitochondrial autophagy. J. Exp. Med. 209, 855鈥?9 (2012).CAS聽 Article聽 PubMed聽 PubMed Central聽Google Scholar聽 16.Ambrose, M., Goldstine, J. V. Gatti, R. A. Intrinsic mitochondrial dysfunction in ATM-deficient lymphoblastoid cells. Hum. Mol. Genet. 16, 2154鈥?164 (2007).CAS聽 Article聽 PubMed聽Google Scholar聽 17.Valentin-Vega, Y. A. Kastan, M. B. A new role for ATM: regulating mitochondrial function and mitophagy. Autophagy 8, 840鈥? (2012).CAS聽 Article聽 PubMed聽 PubMed Central聽Google Scholar聽 18.Sharma, N. K. et al. Intrinsic mitochondrial DNA repair defects in Ataxia Telangiectasia. DNA Repair (Amst). 13, 22鈥?1 (2014).CAS聽 Article聽 PubMed聽Google Scholar聽 19.Fang, E. F. et al. Defective mitophagy in XPA via PARP-1 hyperactivation and NAD(+)/SIRT1 reduction. Cell 157, 882鈥?6 (2014).CAS聽 Article聽 PubMed聽 PubMed Central聽Google Scholar聽 20.Figueira, T. R. et al. Mitochondria as a source of reactive oxygen and nitrogen species: from molecular mechanisms to human health. Antioxid. Redox Signal 18, 2029鈥?4 (2013).CAS聽 Article聽 PubMed聽Google Scholar聽 21.Wallace, D. C. Mitochondria and cancer: Warburg addressed. In Cold Spring Harbor Symposia on Quantitative Biology 70, 363鈥?74 (2005).CAS聽 Article聽Google Scholar聽 22.Rezvani, H. R. et al. XPC silencing in normal human keratinocytes triggers metabolic alterations that drive the formation of squamous cell carcinomas. J. Clin. Invest. 121, 195鈥?11 (2011).CAS聽 Article聽 PubMed聽Google Scholar聽 23.Rezvani, H. R. et al. XPC silencing in normal human keratinocytes triggers metabolic alterations through NOX-1 activation-mediated reactive oxygen species. Biochim. Biophys. Acta - Bioenerg. 1807, 609鈥?19 (2011).CAS聽 Article聽Google Scholar聽 24.Brand, M. D. Nicholls, D. G. Assessing mitochondrial dysfunction in cells. Biochem. J. 435, 297鈥?12 (2011).CAS聽 Article聽 PubMed聽 PubMed Central聽Google Scholar聽 25.Turrens, J. F. Mitochondrial formation of reactive oxygen species. J. Physiol. 552, 335鈥?4 (2003).CAS聽 Article聽 PubMed聽 PubMed Central聽Google Scholar聽 26.Muftuoglu, M., Mori, M. P. de Souza-Pinto, N. C. Formation and repair of oxidative damage in the mitochondrial DNA. Mitochondrion 17, 164鈥?1 (2014).CAS聽 Article聽 PubMed聽Google Scholar聽 27.Damas, J. et al. Mitochondrial DNA deletions are associated with non-B DNA conformations. Nucleic Acids Res. 40, 7606鈥?1 (2012).CAS聽 Article聽 PubMed聽 PubMed Central聽Google Scholar聽 28.Chavanne, F. et al. Mutations in the XPC gene in families with xeroderma pigmentosum and consequences at the cell, protein, and transcript levels. Cancer Res. 60, 1974鈥?982 (2000).CAS聽 PubMed聽Google Scholar聽 29.Kowaltowski, A. J., de Souza-Pinto, N. C., Castilho, R. F. Vercesi, A. E. Mitochondria and reactive oxygen species. Free Radic. Biol. Med. 47, 333鈥?3 (2009).CAS聽 Article聽 PubMed聽Google Scholar聽 30.Ravanat, J. L., Mascio, P. D., Martinez, G. R., Medeiros, M. H. G. Cadet, J. Singlet Oxygen Induces Oxidation of Cellular DNA. J. Biol. Chem. 276, 40601鈥? (2001).CAS聽 PubMed聽Google Scholar聽 31.Berquist, B. R., Canugovi, C., Sykora, P., Wilson, D. M. Bohr, V. A. Human Cockayne syndrome B protein reciprocally communicates with mitochondrial proteins and promotes transcriptional elongation. Nucleic Acids Res. 40, 8392鈥?05 (2012).CAS聽 Article聽 PubMed聽 PubMed Central聽Google Scholar聽 32.Iyama, T. Wilson, D. M. DNA repair mechanisms in dividing and non-dividing cells. DNA Repair (Amst). 12, 620鈥?6 (2013).CAS聽 Article聽 PubMed聽 PubMed Central聽Google Scholar聽 33.Cleaver, J. E. Revet, I. Clinical implications of the basic defects in Cockayne syndrome and xeroderma pigmentosum and the DNA lesions responsible for cancer, neurodegeneration and aging. Mech. Ageing Dev. 129, 492鈥? (2008).CAS聽 Article聽 PubMed聽 PubMed Central聽Google Scholar聽 34.Krishnan, K. J. et al. What causes mitochondrial DNA deletions in human cells? Nat. Genet. 40, 275鈥? (2008).CAS聽 Article聽 PubMed聽Google Scholar聽 35.Birch-Machin, M. A., Russell, E. V. Latimer, J. A. Mitochondrial DNA damage as a biomarker for ultraviolet radiation exposure and oxidative stress. Br. J. Dermatol. 169, 9鈥?4 (2013).CAS聽 Article聽 PubMed聽Google Scholar聽 36.Ericson, N. G. et al. Decreased mitochondrial DNA mutagenesis in human colorectal cancer. PLoS Genet. 8, e1002689 (2012).CAS聽 Article聽 PubMed聽 PubMed Central聽Google Scholar聽 37.Hosseini, M. et al. Premature skin aging features rescued by inhibition of NADPH oxidase activity in XPC-deficient mice. J Invest Dermato. 135, 1108鈥?118 (2015).CAS聽 Article聽Google Scholar聽 38.Yankovskaya, V. et al. Architecture of succinate dehydrogenase and reactive oxygen species generation. Science 299, 700鈥? (2003).ADS聽 CAS聽 Article聽 PubMed聽Google Scholar聽 39.Mckenzie, M. Ryan, M. T. Assembly factors of human mitochondrial complex I and their defects in disease. IUBMB Life 62, 497鈥?02 (2010).CAS聽 Article聽 PubMed聽Google Scholar聽 40.Lenaz, G. et al. Bioenergetics of mitochondrial diseases associated with mtDNA mutations. Biochim. Biophys. Acta 1658, 89鈥?4 (2004).CAS聽 Article聽 PubMed聽Google Scholar聽 41.Lindahl, M., Mata-Cabana, A. Kieselbach, T. The disulfide proteome and other reactive cysteine proteomes: analysis and functional significance. Antioxid. Redox Signal 14, 2581鈥?642 (2011).CAS聽 Article聽 PubMed聽Google Scholar聽 42.Piantadosi, C. A. Suliman, H. B. Mitochondrial transcription factor A induction by redox activation of nuclear respiratory factor 1. J. Biol. Chem. 281, 324鈥?3 (2006).CAS聽 Article聽 PubMed聽Google Scholar聽 43.Piantadosi, C. A. Suliman, H. B. Transcriptional regulation of SDHa flavoprotein by nuclear respiratory factor-1 prevents pseudo-hypoxia in aerobic cardiac cells. J. Biol. Chem. 283, 10967鈥?7 (2008).CAS聽 Article聽 PubMed聽 PubMed Central聽Google Scholar聽 44.Satoh, J., Kawana, N. Yamamoto, Y. Pathway analysis of ChIP-seq-based NRF1 target genes suggests a logical hypothesis of their involvement in the pathogenesis of neurodegenerative diseases. Gene Regul Syst Bio. 7, 139鈥?52 (2013).Article聽 PubMed聽 PubMed Central聽Google Scholar聽 45.Han, W. et al. Nrf1 CNC-bZIP protein promotes cell survival and nucleotide excision repair through maintaining glutathione homeostasis. J Biol Chem 287, 18788鈥?8795 (2012).CAS聽 Article聽 PubMed聽 PubMed Central聽Google Scholar聽 46.Gottlieb, E. Tomlinson, I. P. M. Mitochondrial tumour suppressors: a genetic and biochemical update. Nat. Rev. Cancer 5, 857鈥?6 (2005).CAS聽 Article聽 PubMed聽Google Scholar聽 47.Selak, M. A. et al. Succinate links TCA cycle dysfunction to oncogenesis by inhibiting HIF-1伪 prolyl hydroxylase. Cancer Cell 7, 77鈥?5 (2005).CAS聽 Article聽 PubMed聽Google Scholar聽 48.Ishii, T. et al. A mutation in the SDHC gene of complex II increases oxidative stress, resulting in apoptosis and tumorigenesis. Cancer Res. 65, 203鈥? (2005).CAS聽 PubMed聽Google Scholar聽 49.Slane, B. G. et al. Mutation of succinate dehydrogenase subunit C results in increased O2 鈥⑩垝, oxidative stress, and genomic instability. Cancer Res. 66, 7615鈥?0 (2006).CAS聽 Article聽 PubMed聽Google Scholar聽 50.Lemarie, A., Huc, L., Pazarentzos, E., Mahul-Mellier, A.-L. Grimm, S. Specific disintegration of complex II succinate: ubiquinone oxidoreductase links pH changes to oxidative stress for apoptosis induction. Cell Death Differ. 18, 338鈥?9 (2011).CAS聽 Article聽 PubMed聽Google Scholar聽 51.Albayrak, T. Grimm, S. A high-throughput screen for single gene activities: Isolation of apoptosis inducers. Biochem. Biophys. Res. Commun. 304, 772鈥? (2003).CAS聽 Article聽 PubMed聽Google Scholar聽 52.Hwang, M. S., Rohlena, J., Dong, L. F., Neuzil, J. Grimm, S. Powerhouse down: Complex II dissociation in the respiratory chain. Mitochondrion 19, 20鈥? (2014).CAS聽 Article聽 PubMed聽Google Scholar聽 53.Yakes, F. M. Van Houten, B. Mitochondrial DNA damage is more extensive and persists longer than nuclear DNA damage in human cells following oxidative stress. Proc. Natl. Acad. Sci. USA 94, 514鈥? (1997).ADS聽 CAS聽 Article聽 PubMed聽 PubMed Central聽Google Scholar聽 54.Santos, J. H. J., Meyer, J. N. J., Mandavilli, B. S. B. Van Houten, B. Quantitative PCR-based measurement of nuclear and mitochondrial DNA damage and repair in mammalian cells. Methods Mol. Biol. 314, 183鈥?9 (2006).CAS聽 Article聽 PubMed聽Google Scholar聽 55.Rygiel, K. A., Grady, J. P., Taylor, R. W., Tuppen, H. A. Turnbull, D. M. Triplex real-time PCR鈥揳n improved method to detect a wide spectrum of mitochondrial DNA deletions in single cells. Sci Rep. 5, 9906 (2015).ADS聽 CAS聽 Article聽 PubMed聽 PubMed Central聽Google Scholar聽 56.Maynard, S., de Souza-Pinto, N. C., Scheibye-Knudsen, M. Bohr, V. A. Mitochondrial base excision repair assays. Methods 51, 416鈥?5 (2010).CAS聽 Article聽 PubMed聽 PubMed Central聽Google Scholar聽 57.Zhou, H., Duncan, R. F., Robison, T. W., Gao, L. Forman, H. J. Ca2+-dependent p47phox translocation in hydroperoxide modulation of the alveolar macrophage respiratory burst. Am. J. Physiol. 273, L1042鈥? (1997).CAS聽 PubMed聽Google Scholar聽 58.Mantha, S. V., Prasad, M., Kalra, J. Prasad, K. Antioxidant enzymes in hypercholesterolemia and effects of vitamin E in rabbits. Atherosclerosis 101, 135鈥?4 (1993).CAS聽 Article聽 PubMed聽Google Scholar聽 59.Spinazzi, M., Casarin, A., Pertegato, V., Salviati, L. Angelini, C. Assessment of mitochondrial respiratory chain enzymatic activities on tissues and cultured cells. Nat. Protoc. 7, 1235鈥?6 (2012).CAS聽 Article聽 PubMed聽Google Scholar聽 60.Payne, C. M. et al. Deoxycholate induces mitochondrial oxidative stress and activates NF-kappaB through multiple mechanisms in HCT-116 colon epithelial cells. Carcinogenesis 28, 215鈥?2 (2007).CAS聽 Article聽 PubMed聽Google Scholar聽 Download referencesAcknowledgementsWe thank Dr. Roger F. Castilho (Department of Clinical Pathology, School of Medical Sciences, Universidade Estadual de Campinas) for discussions and suggestions; Paolo G. La Guardia (in memoriam) for designing the CII, III and IV OCR experiments; Edilene de Souza Siqueira Santos and Roberto C茅sar Stahl for technical assistance. This work was supported by Funda莽茫o de Amparo 脿 Pesquisa do Estado de S茫o Paulo (FAPESP) through the following grants: CEPID-Redoxoma (13/07937-8), INCT-Redoxoma (08/57721-3), Instituto Nacional de Obesidade e Diabetes (08/57952-5) and thematic projects 10/51906-1 to NCS-P and 06/59786-0 and 11/50400-0 to AEV. RAPC and MPM were supported by CAPES fellowships, and DTS by a FAPESP fellowship (2012/11889-6).Author informationAuthor notesMateus P. Mori and Rute A. P. Costa contributed equally to this work.AffiliationsDepartamento de Bioqu铆mica, Instituto de Qu铆mica, Universidade de S茫o Paulo (USP), S茫o Paulo, SP, BrazilMateus P. Mori,聽Daniela T. Soltys,聽Thiago de S. Freire,聽Ign谩cio Amigo,聽Alicia J. Kowaltowski聽 聽Nadja C. de Souza-PintoDepartment of Clinical Pathology, School of Medical Sciences, Universidade Estadual de Campinas (UNICAMP), Campinas, SP, BrazilRute A. P. Costa,聽Franco A. Rossato聽 聽An铆bal E. VercesiAuthorsMateus P. MoriView author publicationsYou can also search for this author in PubMed聽Google ScholarRute A. P. CostaView author publicationsYou can also search for this author in PubMed聽Google ScholarDaniela T. SoltysView author publicationsYou can also search for this author in PubMed聽Google ScholarThiago de S. FreireView author publicationsYou can also search for this author in PubMed聽Google ScholarFranco A. RossatoView author publicationsYou can also search for this author in PubMed聽Google ScholarIgn谩cio AmigoView author publicationsYou can also search for this author in PubMed聽Google ScholarAlicia J. KowaltowskiView author publicationsYou can also search for this author in PubMed聽Google ScholarAn铆bal E. VercesiView author publicationsYou can also search for this author in PubMed聽Google ScholarNadja C. de Souza-PintoView author publicationsYou can also search for this author in PubMed聽Google ScholarContributionsM.P.M. and R.A.P.C. analyzed the data and wrote the manuscript. F.A.R. and M.P.M. performed OCR in Oroboros; M.P.M. and R.A.P.C. and performed experiment of H2O2 production; M.P.M. performed XL-PCR, QPCR, mitochondria isolation, Western blotting, OCR in SeaHorse and clonogenic assay; T.S.F. performed Western blotting; F.A.R. and D.T.S. performed flow cytometry analysis; M.P.M. and I.A. performed CI鈥?鈥塁III and CII鈥?鈥塁III activities; R.A.P.C., M.P.M. and I.A. performed SOD and GPx activity; R.A.P.C. performed cytochrome oxidase c experiment; D.T.S. performed RT-qPCR. A.J.K. and A.E.V. revised the manuscript; N.C.S.P. designed the experiments, analyzed the data, wrote and revised the manuscript. All authors read and approved the final manuscript.Corresponding authorCorrespondence to Nadja C. de Souza-Pinto.Ethics declarations Competing Interests The authors declare that they have no competing interests. Additional information Publisher\'s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.Electronic supplementary material Supplementary InformationRights and permissions This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article鈥檚 Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/ Reprints and PermissionsAbout this articleCite this articleMori, M.P., Costa, R.A.P., Soltys, D.T. et al. Lack of XPC leads to a shift between respiratory complexes I and II but sensitizes cells to mitochondrial stress. Sci Rep 7, 155 (2017). https://doi.org/10.1038/s41598-017-00130-xDownload citationReceived: 01 July 2016Accepted: 08 February 2017Published: 13 March 2017DOI: https://doi.org/10.1038/s41598-017-00130-x CommentsBy submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate. Sign up for the Nature Briefing newsletter 鈥?what matters in science, free to your inbox daily.