Abstract
DNA double-strand breaks (DSBs) are considered as one of the primary causes of cancer but their induction by hydrogen peroxide (H2O2) is still controversial. In this work, we studied whether the high levels of H2O2 produced in the thyroid to oxidize iodide could induce DNA modifications. Scores of DNA damage, in terms of strand breaks, were obtained by comet assay (alkaline condition for single-strand breaks (SSBs) and neutral condition for DSBs). We demonstrated that in a rat thyroid cell line (PCCl3), non-lethal concentrations of H2O2 (0.1–0.5 mmol/l) as well as irradiation (1–10 Gy) provoked a large number of SSBs (∼2–3 times control DNA damage values) but also high levels of DSBs (1.2–2.3 times control DNA damage values). We confirmed the generation of DSBs in this cell line and also in human thyroid in primary culture and in pig thyroid slices by measuring phosphorylation of histone H2AX. l-Buthionine-sulfoximine, an agent that depletes cells of glutathione, decreased the threshold to observe H2O2-induced DNA damage. Moreover, we showed that DNA breaks induced by H2O2 were more slowly repaired than those induced by irradiation. In conclusion, H2O2 causes SSBs and DSBs in thyroid cells. DSBs are produced in amounts comparable with those observed after irradiation but with a slower repair. These data support the hypothesis that the generation of H2O2 in thyroid could also play a role in mutagenesis particularly in the case of antioxidant defense deficiency.
Introduction
Thyroid nodules are common and constitute an important clinical problem. They may occur in up to 50% of a population above 60 years old (Ross 2002). Among these nodules, 5% are cancerous, mainly papillary cancers (PTC). While the prevalence of clinically significant PTC remains relatively low, the prevalence of papillary microcarcinoma is clearly higher and reaches 13% in some series of autopsies (Nasir et al. 2000). There is no clear explanation for this high frequency of thyroid tumors. Iodine deficiency increases the prevalence of hot nodules and multinodular goiters but does not modify the global incidence of thyroid cancers even though it raises the relative proportion of follicular carcinoma (Krohn & Paschke 2002). Irradiation is the only environmental risk factor clearly implicated in thyroid cancer pathogenesis (Ron et al. 1995).
The initial event in the majority of PTC consists of an activation of the RAS/RAF/MEK/MAP kinase pathway, either directly by mutation of BRAF and more rarely of RAS, or indirectly by constitutive activation of tyrosine kinase receptors resulting from chromosomal rearrangements (RET/PTC, TRK; Lacroix et al. 2005). After irradiation, carcinogenic processes are mainly attributed to the formation of DNA double-strand breaks (DSBs; Sarasin et al. 1999). This damage results from a direct effect of irradiation but also from the generation of reactive oxygen species (ROS) formed during water radiolysis (Mikkelsen & Wardman 2003). However, the precise molecular mechanisms of most radiation-induced cancers are largely unknown. Nevertheless, irradiation is certainly not responsible for the majority of thyroid tumors.
Therefore, since 1990, we hypothesized that the elevated frequency of thyroid tumors (benign or malignant) could be partially explained by the prominent mutagenic environment present in the thyroid, resulting from its metabolism producing large amounts of hydrogen peroxide (H2O2; Song et al. 2007). Indeed, in thyroid, H2O2 generated by the DUOX enzymes (Dupuy et al. 1999, De Deken et al. 2000) is required and is the limiting co-factor for iodide oxidation by thyroperoxidase (TPO) and thyroid hormone synthesis (Nunez & Pommier 1982). Quantitatively, a stimulated thyrocyte generates almost as much H2O2 as a stimulated leukocyte (Corvilain et al. 1994). However, while the leukocyte dies soon after activation, the thyrocyte life is much longer (maximum 7 divisions during adulthood; Coclet et al. 1989) allowing mutations to accumulate. H2O2 could facilitate a mutagenic process and lead to tumorigenesis by altering the DNA (oxidation of bases, DNA single-strand breaks (SSBs) and DSBs). H2O2 could also enhance cell proliferation through various mechanisms (Stone & Yang 2006). Arguments to support the involvement of H2O2 in mutagenesis and etiopathogenesis of thyroid nodules were recently reviewed by our group and others (Maier et al. 2006, Song et al. 2007). The existence of a more mutagenic environment in the thyroid than other organs are suggested: 1) the spontaneous mutation rate is around 10 times higher in the thyroid than in the liver, 2) the comet assay detects increased levels of oxidized pyrimidine and purine in thyroid compared with other organs, and 3) immunohistochemistry methods reveal higher levels of 8-oxoguanine in the thyroid (Maier et al. 2006). A gene expression signature reflecting the differences in cellular response to γ-radiation and H2O2 could distinguish radiation-induced (from the Chernobyl Tissue Bank) and spontaneous PTCs (from French patients with no history of radiation exposure; Detours et al. 2007). These data reinforce our hypothesis suggesting that in thyroid, H2O2 could provoke DNA damage and mutations. Nevertheless, harmful effects caused by H2O2 are tightly controlled in thyroid, thanks to the restricted apical localization of its production and the presence of various intracellular H2O2 detoxifying enzymes, like seleno-dependent glutathione peroxidase (GPx; Kohrle et al. 2005). In epidemiological studies, it has been shown that the modification of antioxidant capacities in the diet can modulate cancer risk (Bertram et al. 1987, Ip et al. 1994, Vogt et al. 2003). Antioxidant status of cells may also play a role in the mutagenic potency of irradiation and oxidative stress (Kim et al. 2000, Neumann et al. 2003, Wang et al. 2003).
While carcinogenic DNA DSBs are clearly induced by irradiation, their induction by H2O2 is still controversial. Therefore, we analyzed in this work the capacity of H2O2 to generate DNA SSBs and especially DSBs and compared this with the effects observed with a well-known carcinogenic factor (irradiation) to determine if H2O2, like irradiation, could therefore be implicated in the pathogenesis of thyroid tumors.
Materials and methods
Cell lines and culture conditions
PCCl3 cells, a rat thyroid cell line, were cultured as previously described (Rigutto et al. 2007). Non-transformed rat fibroblasts F208 were grown in DMEM with 10% v/v fetal bovine serum, 1% v/v sodium pyruvate, 100 U/ml penicillin, 100 μg/ml streptomycin, and 2.5 μg/ml fungizone. Pig thyroids were obtained from freshly killed animals in accordance to ethical guidelines. Thyroids were cut into thin slices (0.3 mm) of ∼50 mg wet weight. Slices were preincubated at 37 °C for 1 h in 2 ml Krebs-Ringer–Hepes (KRH) medium and then transferred to fresh KRH before treatment (Corvilain et al. 2000). Primary cultures of human thyroid cells were prepared and incubated as previously described; the experiments were performed, four days after seeding the thyroid follicles, on differentiated thyrocytes (Roger et al. 1988). Human thyroid tissues were obtained from patients undergoing thyroidectomy for multinodular goiter with approval from the institutional ethic committee.
Cell treatment
Irradiation
Cells were exposed to a single dose of γ-irradiation from a Cesium137 source at a dose rate of 2.03 Gy/min at room temperature. To determine the initial DNA damage, cells were placed on ice immediately after treatment. For the kinetic experiments cells were incubated at 37 °C for different periods of time after irradiation.
Hydrogen peroxide
H2O2 (Merck) dilutions were prepared in culture medium immediately before use. The H2O2 solutions were diluted in the medium of dishes containing cells in culture or thyroid slices.
H2O2 generating system
A relatively stable production of H2O2 was obtained in the cell culture medium using proline (Sigma–Aldrich) in conjunction with 5 mU/ml d-amino-oxidase (DAO from Sigma–Aldrich).
dl-Buthionine-[S,R]-sulfoximine treatment
Cells were preincubated overnight with 10 mmol/l l-buthionine-sulfoximine (BSO; Sigma–Aldrich) before H2O2 or irradiation treatment.
Cell survival test
The cytotoxic potential of the different treatments was evaluated by the 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (MTS) cell proliferation assay. This colorimetric method measures the formation of a soluble formazan product that is directly proportional to the number of living cells in culture. After different intervals of time after treatment, cells were incubated with a MTS/phenazine methosulfate solution for 1–3 h in the dark at 37 °C and in a 5% CO2 atmosphere as described in the Promega (Promega) technical bulletin no. 169. The soluble formazan product has an absorbance maximum at 490–500 nm and was recorded using an ELISA plate reader. Viability was calculated as a percentage of the control. Triton 0.5% v/v was used as a positive control for cell death.
H2O2 measurement
H2O2 was measured in the cell culture medium by a sensitive fluorimetric assay based on the H2O2-dependent oxidation of homovanillic acid (3-methoxy-4-hydroxyphenylacetic acid, HVA; Sigma–Aldrich) to a highly fluorescent dimer (2,2′-dihydroxy-3,3′-dimethoxydiphenyl-5,5′-diacetic acid) by HRP (Benard & Brault 1971).
Iodide organification (protein bound iodide)
Pig thyroid slices were preincubated for 30 min in KRH buffer containing 0.5 g/l BSA supplemented or not with 0.5 mmol/l H2O2. The slices were then incubated in fresh medium supplemented with KI (10−5 mol/l) and 125I (1 μCi/ml) for 30 min. Methimazole (1 mmol/l), that blocks iodide organification, added during preincubation and incubation, was used to estimate the background in the assay. Iodide organification was measured in basal conditions and in slices stimulated by TSH (10 mU/ml) and ionomycin (2 μmol/l). The slices were homogenized in a methimazole solution (1 mmol/l). Proteins were precipitated with 10% v/v trichloroacetic acid and counted. Iodide bound to proteins is expressed as picomoles of iodide organified per 100 mg wet weight tissue/30 min.
Comet assays
The comet assay was adapted from Singh et al. (1988) and Olive et al. (1990) as previously described (Chico Galdo et al. 2006). This assay is based on the separation from supercoiled DNA of DNA loops containing strand breaks (SSBs and DSBs) that become free to migrate out of the nucleus towards the anode during an alkaline electrophoresis in a solution of 0.3 M NaOH, 1 mM EDTA, pH>13. To detect DSBs more specifically, the comet assay was adapted to neutral electrophoresis in a buffer of 300 mM sodium acetate, 100 mM Tris–HCl, pH 8.3 as previously described (Wojewodzka et al. 2002).
DNA images were captured after staining with ethidium bromide (20 μg/ml) with a Zeiss Axioplan 2 imaging microscope with a 40× objective lens. For quantification, the comets were classified into different categories. We subdivided cell DNA damage into five stages (1–5) for the alkaline assay (Collins 2004) and four stages (2–5) for the neutral assay according to the length and the intensity of the comet tail as illustrated in Fig. 1A. Stage 1 (no tail) and stage 2 (halo around the nucleus) corresponded to cells without a significant number of DNA strand breaks. Stages 3–5 corresponded to a gradual increase in DNA damage. We measured the comet score in 200 randomly selected cells per slide. Results were, first, expressed as the percentage of each stage of comets per slide. In a second step, a comet score was calculated, following a modification of Collins' method, as the sum of the percentage of each comet stage n (from 1 to 5) multiplied by n−1 (Collins 2004). For this calculation, all the negative comets (stages 1 and 2) were considered as stage 2. The scores were expressed in arbitrary units on a scale from 100 (all the comets are in stage 2) to 400 (all the comets are in stage 5). While the scale of the comet assay was the same in alkaline (reflecting SSBs and DSBs) and neutral condition (more specific for DSBs), the observed score remained semi-quantitative and therefore the values could not be compared to extrapolate the number of SSBs.
(A) Classification of the comets. Comet categories are defined by the size of the head (nucleus) and the length and intensity of the tail. Five stages are defined in alkaline conditions (upper panel) reflecting global DNA damage (SSBs and DSBs) and four stages in neutral conditions (lower panel) reflecting essentially DSBs. Stage 1, normal nucleus; stage 2, halo around the nucleus; stage 3–5, gradual increase in the length and intensity of the comet tail evolving in parallel with a decrease in the nuclear DNA content. (B) Comet assays performed on PCCl3 cells after incubation with 1 mM H2O2. Alkaline assays showed comets in stage 5 after labeling with ethidium bromide and an Ab ssDNA. In neutral assays, the tail of the comet was not labeled by the Ab ssDNA, even though almost all comets were in stage 3 after staining with ethidium bromide. This means that in the neutral condition, the DNA content in the tail results mainly from DSBs. Ab ssDNA, antibody against SSBs; SSBs, single-strand breaks; and DSBs, double-strand breaks.
Citation: Endocrine-Related Cancer 16, 3; 10.1677/ERC-09-0020
Comet immunoassays were performed as previously described (Wojewodzka et al. 2002).
Western blotting and immunocytochemistry
Two or four μg of histone extracts obtained after lysis in buffer (150 mmol/l NaCl, 1 mmol/l EDTA, 20 mmol/l Tris–HCl (pH 8), 0.5% v/v NP40) were separated by 15% SDS/PAGE and transferred to nitrocellulose. Immune complexes were detected with HRP-coupled anti-rabbit or anti-mouse IgG antibodies according to the ECL method (NEN Life science product) as previously described (De Deken et al. 2000). Some nitrocellulose membranes were analyzed by the Odyssey infrared imaging system (LI-COR, Biosciences, Erembodegem, Belgium) using fluorescent secondary antibodies (IRDye 680 Goat Anti-Mouse and IRDye 800 Goat Anti-Rabbit from LI-COR) as previously described (Picariello et al. 2006).
For immunocytochemistry, treated cells were washed with cold Tris 0.05 mol/l, NaCl 0.15 mol/l pH 7.4 (TBS) and fixed in 95% v/v ethanol – 5% v/v acetic acid. Cells were treated with 3% v/v horse serum/TBS before incubation with primary antibodies against phosphorylated Ser139 of histone H2AX (γH2AX) and total histone H2AX (Upstate Cell Signaling Solutions) (Bioconnect, TE Huissen, The Netherlands) at a 1:500 dilution in blocking buffer for 1 h at room temperature. After washing with TBS, cells were incubated with 1:400 Cy3-conjugated donkey anti-mouse secondary antibody (Jackson Immuno Research, Suffolk, UK) for 1 h at room temperature in the dark. Cells were washed with TBS and counterstained with bis-benzimide. Observations were performed with a Zeiss Axioplan 2 imaging microscope with a 40× objective lens.
Statistical analysis
Non-parametric unpaired tests were performed with the use of GraphPad Prism Software (San Diego, CA, USA).
Results
Assessment of DNA damage after irradiation in a thyroid cell line
We analyzed DNA damage (SSBs and DSBs) by comet assay immediately after γ-irradiation. Irradiation from 1 to 10 Gy of PCCl3 cells induced a dose-dependent increase in the comet score in alkaline condition (SSBs and DSBs) as well as in neutral condition (DSBs). This damage reached a significant score after irradiation of 1–2 Gy (Table 1). No significant cell death was observed by the MTS assay after an irradiation of 10 Gy (Table 2). We confirmed the presence of DSBs by analysis of phosphorylation of histone variant H2AX on serine 139 by western blotting. Histone H2AX phosphorylation, that reflects the presence of DSBs, increased in a dose-dependent way from 1 to 10 Gy (Fig. 2A). These results were corroborated by an immunocytochemistry method that detects foci of phosphorylated histone H2AX using the same antibody as in western blotting. A similar relationship was observed between the dose of γ-irradiation and the number of foci (pink spots in the nucleus) per cell as shown in Fig. 2B.
Score of DNA damage measured by comet assay in PCCl3 cells immediately after irradiation (A) and 15 min after addition of various H2O2 concentrations (B). DNA damage was evaluated in alkaline conditions (SSBs+DSBs) and in neutral conditions (DSBs). Data are expressed as mean (arbitrary units on a scale from 100 to 400)±s.e.m.; (n), number of measurements. Statistical analyses were made by comparison of control values and values obtained after treatments
| Alkaline | Neutral | |
|---|---|---|
| (A) Irradiation | ||
| Ctl | 120.4±2.1 (32) | 113.2±1.9 (26) |
| 1–2 Gy | 165.3±15.4 (6)* | 201.0±6.1 (6)* |
| 4–5 Gy | 202.1±15.8 (10)* | 244.5±30.0 (2)* |
| 10 Gy | 283.0±6.4 (21)* | 253.5±8.6 (24)* |
| (B) H2O2 (mmol/l) | ||
| 0 | 120.1±2.2 (38) | 110.5±1.1 (33) |
| 0.01 | 125.7±5.7 (9) | 105.8±1.9 (6) |
| 0.05 | 128.1±7.2 (10) | 132.8±11.2 (4)* |
| 0.1 | 245.0±11.7 (19)* | 215.6±13.4 (14)* |
| 0.2 | 321.8±45.2 (6)* | 241.9±23.2 (6)* |
| 0.5 | 400.0±0.0 (4)* | 225.4±42.3 (4)* |
| 1 | 391.3±8.1 (15)* | 204.5±9.7 (19)* |
*P<0.001. SSBs, single-strand breaks; DSBs, double-strand breaks.
Cell survival was evaluated by (A) MTS assay in PCCl3 cells and (B) human thyroid in primary culture for 2, 24, and 48 h after addition of various H2O2 concentrations and after different doses of irradiation. Triton 0.5% v/v was used as a positive control for cell death. Viability was calculated as a percentage of the control. Data are expressed as mean±s.e.m.; (n), number of measurements
| 2 h | 24 h | 48 h | |
|---|---|---|---|
| (A) H2O2 (mmol/l) | |||
| 0 | 100.0±1.7 (9) | 100.0±1.0 (9) | 100.0±2.2 (9) |
| 0.1 | 101.4±1.5 (9) | 101.9±0.6 (9) | 99.5±2.2 (9) |
| 0.2 | 100.9±1.1 (9) | 100.2±0.8 (9) | 95.3±1.3 (9) |
| 0.5 | 101.1±1.1 (9) | 101.0±1.2 (9) | 101.9±1.9 (9) |
| 1 | 77.4±2.3 (9)† | 87.4±2.0 (9)† | 94.9±1.5 (9) |
| 10 | 14.2±2.2 (9)† | 0.9±0.2 (9)† | 0.8±0.2 (9) |
| Irradiation (Gy) | |||
| 1 | 116.7±1.8 (9) | 115.8±1.9 (9) | 108.0±1.6 (9) |
| 10 | 125.1±2.1 (9) | 112.7±1.7 (9) | 104.2±2.4 (9) |
| Triton 0.5% v/v | 0.8±0.1 (9)† | 0.7±0.1 (9)† | 0.5±0.2 (9)† |
| (B) H2O2 (mmol/l) | |||
| 0 | 100.0±1.0 (9) | 100.0±0.5 (9) | 100.0±1.2 (6) |
| 0.1 | 102.2±1.7 (9) | 102.2±1.4 (9) | 99.8±1.3 (6) |
| 0.25 | 100.3±0.9 (9) | 99.5±1.5 (9) | 101.0±2.7 (6) |
| 0.5 | 102.9±2.6 (9) | 101.6±2.5 (9) | 99.2±2.5 (6) |
| 1 | 104.2±2.4 (9) | 100.2±2.8 (9) | 101.5±2.1 (6) |
| 10 | 41.8±8.4 (9)† | 53.8±5.2 (9)† | 59.3±14.3 (6)* |
| 100 | 2.0±1.2 (6)† | 2.6±1.0 (6)† | 1.2±0.6 (6)† |
| Irradiation (Gy) | |||
| 1 | 106.4±1.0 (9) | 101.7±1.2 (9) | 102.3±1.7 (6) |
| 5 | 112.6±3.3 (9) | 105.7±3.1 (9) | 105.4±2.0 (6) |
| 10 | 104.9±2.3 (9) | 104.5±2.8 (9) | 104.9±1.1 (6) |
| 50 | 106.1±4.3 (9) | 101.1±2.5 (9) | 104.2±1.7 (6) |
| Triton 0.5% v/v | 1.2±0.3 (9)† | 1.2±0.6 (6)† | 0.9±0.4 (6)† |
*P<0.05; †P<0.001.
DNA DSBs evaluation using phosphorylation of histone H2AX in PCCl3 cells an hour after irradiation (A and B) or addition of H2O2 in the medium (C and D). Phosphorylated (γH2AX) and total histones H2AX were detected with specific antibodies. (A) Western blot showing a dose-dependent effect of irradiation on H2AX phosphorylation. (B) Immunocytochemistry detecting phosphorylated histone H2AX foci (pink spots). (C) Western blot showing dose-dependent effect of H2O2 on H2AX phosphorylation. (D) Immunocytochemistry detecting phosphorylated histone H2AX foci (pink spots). DSBs, double-strand breaks.
Citation: Endocrine-Related Cancer 16, 3; 10.1677/ERC-09-0020
Assessment of DNA damage after exposure to H2O2 in thyroid and non-thyroid cell lines
As H2O2 is rapidly degraded when added to the cells, we first estimated how long H2O2 was present in the incubation medium of the cells in the described experiments. After addition of H2O2, the remaining quantities in the medium were measured at different time intervals in comparison with the same concentration of H2O2 added to medium without cells. 0.1 mmol/l H2O2 added to PCCl3 cells disappeared rapidly; 23 and 1.6% remained after 15 min and 1 h respectively.
DNA damage in PCCl3 cells was quantified by comet assay 15 min after exposure to H2O2. In alkaline conditions, no DNA damage was observed up to 0.05 mmol/l H2O2 but the scores increased abruptly and significantly above the control values at 0.1 mmol/l H2O2 and reached a maximum value (400) between 0.2 and 0.5 mmol/l H2O2 (Table 1) corresponding to the upper limit of the assay. The scores obtained in neutral conditions were significantly increased in cells exposed to a concentration of 0.05 mmol/l H2O2 compared to non-treated cells (Table 1). These scores reached a plateau at a H2O2 concentration of 0.1 mmol/l. We also examined the phosphorylation of histone H2AX after H2O2 treatments (Fig. 2C). One hour after the treatment, we observed a concentration-dependent increase of histone H2AX phosphorylation from 0.1 up till 1 mmol/l H2O2. Immunocytochemistry confirmed the presence of a rising number of phosphorylated histone H2AX foci, reflecting the presence of DNA DSBs, in cells treated with 0.1 and 1 mmol/l of H2O2 (Fig. 2D). H2O2 concentrations equal to or below 0.5 mmol/l did not induce significant cell death in contrast to 10 mmol/l H2O2 that provoked a massive cell death (Table 2). A viability of 77.4% in comparison with the control cells was observed after 2 h treatment with 1 mmol/l H2O2 (Table 2).
Western blotting (γ-H2AX) was also used to evaluate DNA damage in a non-thyroid rat cell line, the F208 fibroblast cell line. We observed the presence of DSBs by the detection of significant levels of phosphorylated H2AX with 0.5–1 mmol/l H2O2 (data not shown).
Assessment of DNA damage after exposure to H2O2 in pig thyroid slices and in human thyroid primary culture cells
DSBs were detected 1 h after 1 and 10 Gy irradiation in pig thyroid slices by measuring H2AX phosphorylation levels; 3.9±1.0-fold (mean±s.e.m.; P=0.0063) and 32.0±5.0-fold (mean±s.e.m.; P<0.0001) in comparison with the control for 1 and 10 Gy irradiation respectively. We also evaluated the induction of DSBs after incubation with different concentrations of H2O2: 0.1, 0.5, 1, and 10 mmol/l. One hour after treatment, a significant induction of phosphorylated H2AX was observed from 0.5 to 10 mmol/l H2O2; 2.3±0.5-fold (mean±s.e.m.; P=0.01), 4.9±1.1-fold (mean±s.e.m.; P=0.0056) and 33.1±2.9-fold (mean±s.e.m.; P<0.0001) in comparison with the control for 0.5, 1 and 10 mmol/l H2O2 respectively. Cell survival was indirectly evaluated by the capacity of the pig thyroid slices to organify iodide after treatments with H2O2. In the basal condition, 0.5 mmol/l H2O2 did not change the capacity of pig thyroid slices to organify iodide (protein bound iodide (PB125I) measurements); 97.8±11.2% (mean±s.e.m.) of the control was measured after H2O2 treatment. In slices stimulated with 10 mU/ml TSH or 2 μM ionomycin (that stimulate H2O2 production), H2O2 did not modify extensively the iodide organifying function; 76.5±5.2% of the control and 92.0±11.4% of the control were measured respectively.
In human thyroid in primary culture, the phosphorylation of histone H2AX was increased following a dose-dependent curve between 1 and 10 Gy (data not shown). No significant cell death was observed by the MTS assay 24 and 48 h after an irradiation of 1 to 50 Gy (Table 2). Human thyroid primary culture cells displayed phosphorylation of histone H2AX one hour after treatment with H2O2 from 0.2 to 0.5 mmol/l in a dose-dependent way; 1.9±0.3 fold (mean±s.e.m.; P=0.0199) and 8.2±3.5 fold (mean±s.e.m.; P=0.0161) in comparison to the control for 0.2 and 0.5 mmol/l H2O2 respectively. One representative experiment is shown in Fig. 3. By MTS assay, we measured that H2O2 concentrations equal to or below 1 mmol/l did not induce significant cell death 24 and 48 h after treatment in contrast to 100 mmol/l H2O2 that provoked a massive cell death already 2 h after treatment (Table 2).
DNA DSBs evaluation by phosphorylation of histone H2AX in human thyrocytes in primary culture an hour after H2O2 addition or irradiation. Phosphorylated and total histones H2AX were detected by western blotting with specific antibodies. DSBs, double-strand breaks. This western blot is representative of six independent cell cultures.
Citation: Endocrine-Related Cancer 16, 3; 10.1677/ERC-09-0020
Effect of depletion of glutathione on DNA damage induced by H2O2 and irradiation in PCCl3 cells
BSO irreversibly inhibits γ-glutamylcysteine synthetase (γ-GCS) leading to a decrease of glutathione (GSH) concentrations in the cells. GSH is a cofactor for selenium (Se) dependent GPx involved in the detoxification of most cellular H2O2. Comet assays in alkaline conditions were performed on PCCl3 cells preincubated overnight or not with 10 mmol/l BSO before H2O2 or irradiation treatment (Fig. 4A). BSO by itself did not have any effect on DNA integrity. Preincubation with 10 mmol/l BSO rendered cells more sensitive to H2O2 in terms of DNA breaks: 0.05 mmol/l H2O2 provoked more DNA damage in BSO preincubated cells than in corresponding control cells (score of 188±6 in BSO treated cells compared with 126±6 in cells not preincubated with BSO; P<0.0001). BSO also increased the DNA damage observed at 0.1 mmol/l H2O2 (P<0.0001). Significant DNA damage was induced after 2 and 4 Gy irradiation but contrary to what we observed after H2O2 treatments, BSO did not increase the DNA damage of irradiated cells (Fig. 4A).
(A) Effect of H2O2 and irradiation on comet formation in PCCl3 cells with or without overnight preincubation with 10 mM BSO. Comet assays were performed in alkaline condition. Comparisons were made between scores obtained with and without BSO. Data are expressed as mean±s.e.m. of three experiments in duplicate. NS, non-significant; *P<0.05; **P<0.01; ***P<0.001; BSO, l-buthionine-sulfoximine. (B) Effect of a combined treatment by a H2O2-generating system and irradiation on comet formation in PCCl3 cells. The H2O2-generating system constituted DAO and various concentrations of proline. Cells were preincubated during 1 h with the H2O2-generating system before being irradiated. Induced DNA damage was measured immediately after irradiation by alkaline comet assay. Data are expressed as mean±s.e.m. of three experiments in duplicate. Statistical significance was calculated and illustrated after comparing 1) control score values and values obtained with various concentrations of proline and 2) score values obtained with 50 mM proline, 4 Gy irradiation and combined H2O2/irradiation treatment. DAO, d-amino-oxidase; *P<0.05; **P<0.01; ***P<0.001.
Citation: Endocrine-Related Cancer 16, 3; 10.1677/ERC-09-0020
H2O2 concentrations measured in the incubation medium increased linearly with the dose of irradiation (2–20 Gy) from 1.0 to 10 μmol/l (data not shown). These concentrations of H2O2 are largely below those needed to obtain significant DNA damage.
DNA damage in PCCl3 cells after a combined treatment with H2O2 and irradiation
To evaluate DNA damage induced by a combined H2O2 and irradiation treatment, we measured radiation-induced DNA damage on cells preincubated with a H2O2-generating system constituted by proline and DAO. This H2O2-generating system produced in 1 h from 0.06 to 0.1 mmol/l H2O2 with 10 to 100 mmol/l proline (data not shown).
H2O2 generated by using 50 and 100 mmol/l proline during 1 h induced DNA damage with a score evaluated in alkaline conditions at 282±18 and 356±7 respectively. When cells preincubated with DAO/proline at 50 mmol/l during 1 h were irradiated with 4 Gy, the damage score recorded immediately after irradiation nearly reached the sum of the score observed after DAO/proline (50 mmol/l) treatment alone and after irradiation alone (Fig. 4B). This additive effect was not observed with 100 mmol/l proline due to the saturation of the assay (Fig. 4B).
Repair of DNA damage in PCCl3 cells
Kinetics of DNA repair after irradiation
DNA damage repair at different time intervals after irradiation was analyzed by the comet assay (Fig. 5A and B). In alkaline conditions (Fig. 5A), cells submitted to 10 Gy showed a high score of DNA damage (268±18) immediately after the treatment, rapidly decreased with time to reach a score of 135±17 after 1 h, close to the basal score (116±4). In the neutral comet assay (Fig. 5B), repair of DNA DSBs was also completed 1 h after a 10 Gy treatment (score of 254±25 immediately after treatment compared with 117±7 and 147±12 for respectively control and 1 h repair).
Kinetics of DNA repair after irradiation (A and B) and H2O2 treatment (C and D) in PCCl3 cells. Scores of DNA damage induced by 10 Gy irradiation were evaluated by the comet assay immediately after irradiation and then after various time intervals. Scores of DNA damage induced by 0.2 mM H2O2 were evaluated 15 min after the addition of H2O2 and then after various time intervals. (A and C) SSBs+DSBs measured in alkaline condition. (B and D) DSBs measured in neutral condition. Data are expressed as mean±s.e.m.; (n), number of measurements for each time interval; *P<0.05; **P<0.01; ***P<0.001; SSBs, single-strand breaks; and DSBs, double-strand breaks.
Citation: Endocrine-Related Cancer 16, 3; 10.1677/ERC-09-0020
Kinetics of DNA repair after H2O2 treatment
The kinetics of DNA break repair after exposure to H2O2 was also studied by the comet assay performed at different time intervals (Fig. 5C and D). Scores of DNA damage were maximal 15 min after exposure to 0.2 mmol/l H2O2. In alkaline conditions, the assay was almost saturated (score around 400). Comet scores decreased by ∼50% after 4 h and reached nearly basal values 6 h after treatment (score 128±4; Fig. 5C). The repair of DSBs (comet in neutral conditions) was slower than in irradiated cells; comet scores above control values were still observed 24 h after H2O2 treatment (104±2 for the control and 124±1 after 24 h repair; Fig. 5D).
Discussion
The thyroid produces large amounts of H2O2 that constitute a potentially mutagenic environment (Bjorkman & Ekholm 1988). The aim of this study was to compare, in a rat thyroid cell line (PCCl3) and in more physiological models (human thyroid primary culture cells and pig thyroid slices), DNA damage induced by a well-known carcinogenic factor (irradiation) with that obtained by a putative carcinogenic agent (H2O2).
The physiological levels of H2O2 in cells vary from 0.001 μmol/l to a maximum of 0.7 μmol/l (Song et al. 2007) but no data are available for the thyrocyte. As the apparent Km of TPO for H2O2 is ∼300 μmol/l, it has been hypothesized that H2O2 reaches especially high concentrations but in a restricted place at the periphery of the thyrocyte, in the postulated thyroxisome (Song et al. 2007). In thyroid, H2O2 is produced outside the thyrocyte at its apical pole by DUOX and is used by TPO located in the vicinity of DUOX to oxidize iodide. Therefore, our experimental model in which H2O2 is added in the incubation medium mimics the in vivo extracellular production. The range of H2O2 concentrations used in this study is comparable to those probably needed to oxidize iodide. It is difficult to estimate H2O2 concentrations achieved in the limited space of the follicular lumen in vivo but with the generation of 10 nmol/100 mg per h they could easily reach the micromolar level (Corvilain et al. 2000).
In our experiments, extracellular H2O2 is rapidly reduced and has nearly disappeared after 15 min. At the concentrations used in this study, H2O2 was not lethal as confirmed by viability tests. Apoptosis has not been specifically evaluated in this work but previous studies performed on thyroid cells failed to detect apoptosis for similar doses of irradiation and showed apoptosis only in a very small number of cells when treated with similar concentrations of H2O2 (Yang et al. 1997, Riou et al. 1999). These previous observations, along with the absence of mortality in H2O2 or irradiation treated cells exclude that apoptosis may significantly contribute to the measured comet scores.
Comet assays were used to evaluate DNA strand breaks in individual cells. We studied the effects of H2O2 and irradiation on generation of SSBs and DSBs. Interestingly, thyroid cancers occur at doses as low as 0.1 Gy with a linear dose–response curve and the related risk increases by 10 times at 1 Gy (Ron et al. 1995). In our different thyroid models, we observed a significant number of SSBs and DSBs after an irradiation of 1 Gy or more. No differences in sensitivity to irradiation were observed between PCCl3 cells, pig thyroid slices and human thyroid primary culture cells. High levels of DSBs formation were confirmed by estimation of phosphorylation of histone H2AX by western blotting and by immunocytochemistry. Experiments on non-transformed rat fibroblasts (F208) showed a significant production of DSBs with 1 Gy. These results are in keeping with previous data showing that the formation of ∼35 DSBs per gray, per cell, and per cell cycle is a constant (Rogakou et al. 1998, Wojewodzka et al. 2002, Takahashi et al. 2005).
DSBs provoked by H2O2 are considered to be rare events: 1 DSBs for ∼2000 SSBs (Bradley et al. 1979, Takahashi et al. 2005). However, Bradley & Kohn (1979) showed that in mouse leukemia L1210 cells, H2O2 induced DSBs with a ratio of DSBs to SSBs comparable with that caused by X-rays. Takahashi & Ohnishi (2005) reviewed one study demonstrating formation of DSBsand histone H2AX phosphorylation by immunocytochemistry in normal human fibroblasts exposed to 0.1 mmol/l H2O2 for 2 h (Takahashi et al. 2005).
DSBs are considered to be more carcinogenic than SSBs. We demonstrated in PCCl3 cells that high but non-lethal concentrations of H2O2 provoke a large number of SSBs but also as many DSBs as irradiation. In the presence of large amounts of H2O2-induced SSBs, some apparent DSBs could be due to closely spaced SSBs (Bradley et al. 1979). Therefore, we confirmed the presence of real DSBs detected by neutral comet assays by highlighting phosphorylated histone H2AX. In this test, PCCl3 cells demonstrated the same apparent damage caused by H2O2 (0.05–0.1 mmol/l) as by irradiation. Data obtained on human thyroid primary culture cells and on pig thyroid slices showed a threshold of respectively 0.2 and 0.5 mmol/l H2O2 for the appearance of DSBs. We also observed a significant number of DSBs after a 0.5 mmol/l H2O2 treatment of a non-thyroid cell line (F208, a non-transformed rat fibroblast cell line) meaning that many if not all mammalian cells are sensitive to such concentrations of H2O2. Thus, our work is clearly demonstrated by two different methods an induction of DSBs by H2O2 in a thyroid cell line and, more akin to the in vivo situation, in human thyroid primary cultures and pig thyroid slices. The demonstration of H2O2-induced DNA damage does not necessarily imply a mutagenic role but it can be extrapolated. A difficulty for this extrapolation is that levels of DNA damage acutely achieved in vitro must be compared with lower levels accumulated over years. However, several arguments support such an extrapolation: 1) H2O2, as the well accepted mutagen X-ray, induces DNA DSBs, 2) thyroid in which oxidative DNA damage has been demonstrated in vivo displays a higher level of mutations than liver (Maier et al. 2006), and 3) low levels of Se in serum (i.e. presumably lower activity of Se dependent GPx) constitute a risk factor promoting thyroid cancer development (Duntas 2006).
Comparison of the apparent sensitivity of our different models to extra-cellular H2O2 is probably worthless as it may reflect differences in cell membrane permeability and in antioxidant capacities of the cells. When H2O2 is applied to the exterior of cultured cells, the intracellular concentrations are estimated to be ∼10-fold lower than the extra-cellular concentrations (Song et al. 2007). In thyroid, under physiological conditions, a part of the H2O2 not used for thyroglobulin iodination may diffuse into the cells where it is degraded by very efficient antioxidant enzymes like Se dependent GPx. BSO decreases intracellular GSH and therefore the activity of Se dependent GPx. In PCCl3 cells exposed to H2O2, the presence of BSO decreased the concentration of H2O2 needed to observe DNA strand breaks to 0.05 mmol/l without affecting the damage induced by irradiation. BSO alone in absence of externally added H2O2 did not increase the level of DNA strand breaks probably because basal H2O2 production in PCCl3 cells is not sufficient to induce DNA damage (De Deken et al. 2002). Thyroid destruction in myxoedematous endemic cretinism has been related to impaired H2O2 degradation in stimulated but Se deficient thyroids (Contempre et al. 2004). Therefore, we may extrapolate that in vivo the potential DNA damaging effect of H2O2 will increase in case of deficient antioxidant defense.
Because radiation increases 8-oxoguanine modifications, it was suggested that both radiation and endogenous oxidative stress could synergistically lead to the initiation of thyroid cancer (Riou et al. 1998). We looked therefore for a possible synergic effect of a combined treatment of irradiation and H2O2 on DNA damage. The observed effects of irradiation on cells preincubated with a H2O2-generating system were additive with no synergy whatever the conditions used. The very low H2O2 concentrations measured after irradiation of the culture medium as well as the absence of a potentiating effect of BSO suggest that in PCCl3 cells, H2O2 produced through the radiolysis of water is not the main mechanism involved in DNA damage following irradiation.
As it is well known that DNA repair deficiencies are strongly associated with high cancer risk in humans, we compared the kinetics of repair of DNA breaks induced by irradiation and H2O2. DNA damage induced by a 10 Gy irradiation and measured by the comet assay in PCCl3 cells was completely repaired after 1 h. The kinetics of repair was clearly made slow for a similar amount of DSBs induced by H2O2. This observation is probably related to different parameters: 1) the very high quantity of SSBs produced by H2O2 could saturate the repair systems. 2) H2O2 may induce DNA damage, but also have direct inhibitory effects on DNA repair. H2O2 at 0.1 mmol/l can inactivate the human DNA mismatch repair system (Chang et al. 2002) and inhibit the repair of certain types of DNA lesions through redox control of ADP ribosylation and unscheduled DNA synthesis (Pero et al. 1990). 3) In any case, the delay in H2O2-induced damage repair is not due to the persistence of H2O2 in the medium as we demonstrated that in our experimental conditions, H2O2 rapidly (15 min) disappeared from the medium.
In conclusion, H2O2 produces DNA damage in the thyroid. Concentrations of H2O2 that cause significant DNA damage are not lethal for the cells and do not modify cell functioning. These observations reinforce the hypothesis that H2O2 is a potential carcinogenic agent in the thyroid. H2O2 induces SSBs but also more mutagenic DSBs in amounts comparable with what is obtained with irradiation. The low repair efficiency of DNA DSBs induced by H2O2 strengthens the possible role of H2O2, generated in the thyroid to oxidize iodide, in thyroid tumorigenesis. Therefore, chronic endogenous exposure of thyroid cells to H2O2 could be a key to explain the high frequency of thyroid tumors and thyroid microcarcinoma, particularly in case of antioxidant defense deficiency as demonstrated by the increase of damage observed in the presence of BSO and suggested in epidemiological studies in case of Se deficiency (Duntas 2006, Kaprara & Krassas 2006).
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This work was supported by Fonds Erasme pour la Recherche Médicale, Télévie, Fonds de la Recherche Scientifique Médicale (FRSM), Actions de Recherches Concertées de la Communauté Française de Belgique (ARC), European Union Contract FP6-36495 (GENRISK-T) and the ICT Impulse program 2006, Brussels Capital Region, Belgium (In Silico Project). XDD: postdoctoral researcher at the Fonds National de la Recherche Scientifique (FRS-FNRS).
Acknowledgements
We thank Claude Massart, Bernadette Bournonville, and Chantal Degraef for their excellent technical assistance.
References
Benard B & Brault J 1971 Production of peroxide in the thyroid. L'Unión Médicale du Canada 100 701–705.
Bertram JS, Kolonel LN & Meyskens FL Jr 1987 Rationale and strategies for chemoprevention of cancer in humans. Cancer Research 47 3012–3031.
Bjorkman U & Ekholm R 1988 Accelerated exocytosis and H2O2 generation in isolated thyroid follicles enhance protein iodination. Endocrinology 122 488–494.
Bradley MO & Kohn KW 1979 X-ray induced DNA double strand break production and repair in mammalian cells as measured by neutral filter elution. Nucleic Acids Research 7 793–804.
Chang CL, Marra G, Chauhan DP, Ha HT, Chang DK, Ricciardiello L, Randolph A, Carethers JM & Boland CR 2002 Oxidative stress inactivates the human DNA mismatch repair system. American Journal of Physiology. Cell Physiology 283 C148–C154.
Chico Galdo V, Massart C, Jin L, Vanvooren V, Caillet-Fauquet P, Andry G, Lothaire P, Dequanter D, Friedman M & Van Sande J 2006 Acrylamide, an in vivo thyroid carcinogenic agent, induces DNA damage in rat thyroid cell lines and primary cultures. Molecular and Cellular Endocrinology 257–258 6–14.
Coclet J, Foureau F, Ketelbant P, Galand P & Dumont JE 1989 Cell population kinetics in dog and human adult thyroid. Clinical Endocrinology 31 655–665.
Collins AR 2004 The comet assay for DNA damage and repair: principles, applications, and limitations. Molecular Biotechnology 26 249–261.
Contempre B, de Escobar GM, Denef JF, Dumont JE & Many MC 2004 Thiocyanate induces cell necrosis and fibrosis in selenium- and iodine-deficient rat thyroids: a potential experimental model for myxedematous endemic cretinism in central Africa. Endocrinology 145 994–1002.
Corvilain B, Laurent E, Lecomte M, Vansande J & Dumont JE 1994 Role of the cyclic adenosine 3′,5′-monophosphate and the phosphatidylinositol-Ca2+ cascades in mediating the effects of thyrotropin and iodide on hormone synthesis and secretion in human thyroid slices. Journal of Clinical Endocrinology and Metabolism 79 152–159.
Corvilain B, Collyn L, van Sande J & Dumont JE 2000 Stimulation by iodide of H2O2 generation in thyroid slices from several species. American Journal of Physiology. Endocrinology and Metabolism 278 E692–E699.
De Deken X, Wang D, Many MC, Costagliola S, Libert F, Vassart G, Dumont JE & Miot F 2000 Cloning of two human thyroid cDNAs encoding new members of the NADPH oxidase family. Journal of Biological Chemistry 275 23227–23233.
De Deken X, Wang D, Dumont JE & Miot F 2002 Characterization of ThOX proteins as components of the thyroid H2O2-generating system. Experimental Cell Research 273 187–196.
Detours V, Delys L, Libert F, Weiss SD, Bogdanova T, Dumont JE, Franc B, Thomas G & Maenhaut C 2007 Genome-wide gene expression profiling suggests distinct radiation susceptibilities in sporadic and post-Chernobyl papillary thyroid cancers. British Journal of Cancer 97 818–825.
Duntas LH 2006 The role of selenium in thyroid autoimmunity and cancer. Thyroid 16 455–460.
Dupuy C, Ohayon R, Valent A, Noel-Hudson MS, Deme D & Virion A 1999 Purification of a novel flavoprotein involved in the thyroid NADPH oxidase. Cloning of the porcine and human cdnas. Journal of Biological Chemistry 274 37265–37269.
Ip C, Lisk DJ & Scimeca JA 1994 Potential of food modification in cancer prevention. Cancer Research 54 1957s–1959s.
Kaprara A & Krassas GE 2006 Selenium and thyroidal function; the role of immunoassays. Hellenic Journal of Nuclear Medicine 9 195–203.
Kim H, Lee TH, Park ES, Suh JM, Park SJ, Chung HK, Kwon OY, Kim YK, Ro HK & Shong M 2000 Role of peroxiredoxins in regulating intracellular hydrogen peroxide and hydrogen peroxide-induced apoptosis in thyroid cells. Journal of Biological Chemistry 275 18266–18270.
Kohrle J, Jakob F, Contempre B & Dumont JE 2005 Selenium, the thyroid, and the endocrine system. Endocrine Reviews 26 944–984.
Krohn K & Paschke R 2002 Somatic mutations in thyroid nodular disease. Molecular Genetics and Metabolism 75 202–208.
Lacroix L, Soria JC, Bidart JM & Schlumberger M 2005 Oncogenes and thyroid tumors. Bulletin du Cancer 92 37–43.
Maier J, van Steeg H, van Oostrom C, Karger S, Paschke R & Krohn K 2006 Deoxyribonucleic acid damage and spontaneous mutagenesis in the thyroid gland of rats and mice. Endocrinology 147 3391–3397.
Mikkelsen RB & Wardman P 2003 Biological chemistry of reactive oxygen and nitrogen and radiation-induced signal transduction mechanisms. Oncogene 22 5734–5754.
Nasir A, Chaudhry AZ, Gillespie J & Kaiser HE 2000 Papillary microcarcinoma of the thyroid: a clinico-pathologic and prognostic review. In Vivo 14 367–376.
Neumann CA, Krause DS, Carman CV, Das S, Dubey DP, Abraham JL, Bronson RT, Fujiwara Y, Orkin SH & Van Etten RA 2003 Essential role for the peroxiredoxin Prdx1 in erythrocyte antioxidant defence and tumour suppression. Nature 424 561–565.
Nunez J & Pommier J 1982 Formation of thyroid hormones. Vitamins and Hormones 39 175–229.
Olive PL, Banath JP & Durand RE 1990 Heterogeneity in radiation-induced DNA damage and repair in tumor and normal cells measured using the ‘comet’ assay. Radiation Research 122 86–94.
Pero RW, Anderson MW, Doyle GA, Anna CH, Romagna F, Markowitz M & Bryngelsson C 1990 Oxidative stress induces DNA damage and inhibits the repair of DNA lesions induced by N-acetoxy-2-acetylaminofluorene in human peripheral mononuclear leukocytes. Cancer Research 50 4619–4625.
Picariello L, Sala SC, Martineti V, Gozzini A, Aragona P, Tognarini I, Paglierani M, Nesi G, Brandi ML & Tonelli F 2006 A comparison of methods for the analysis of low abundance proteins in desmoid tumor cells. Analytical Biochemistry 354 205–212.
Rigutto S, Hoste C, Dumont JE, Corvilain B, Miot F & De Deken X 2007 Duox1 is the main source of hydrogen peroxide in the rat thyroid cell line PCCl3. Experimental Cell Research 313 3892–3901.
Riou C, Remy C, Rabilloud R, Rousset B & Fonlupt P 1998 H2O2 induces apoptosis of pig thyrocytes in culture. Journal of Endocrinology 156 315–322.
Riou C, Tonoli H, Bernier-Valentin F, Rabilloud R, Fonlupt P & Rousset B 1999 Susceptibility of differentiated thyrocytes in primary culture to undergo apoptosis after exposure to hydrogen peroxide: relation with the level of expression of apoptosis regulatory proteins, Bcl-2 and Bax. Endocrinology 140 1990–1997.
Rogakou EP, Pilch DR, Orr AH, Ivanova VS & Bonner WM 1998 DNA double-stranded breaks induce histone H2AX phosphorylation on serine 139. Journal of Biological Chemistry 273 5858–5868.
Roger P, Taton M, Van Sande J & Dumont JE 1988 Mitogenic effects of thyrotropin and adenosine 3′,5′-monophosphate in differentiated normal human thyroid cells in vitro. Journal of Clinical Endocrinology and Metabolism 66 1158–1165.
Ron E, Lubin JH, Shore RE, Mabuchi K, Modan B, Pottern LM, Schneider AB, Tucker MA & Boice JD Jr 1995 Thyroid cancer after exposure to external radiation: a pooled analysis of seven studies. Radiation Research 141 259–277.
Ross DS 2002 Nonpalpable thyroid nodules – managing an epidemic. Journal of Clinical Endocrinology and Metabolism 87 1938–1940.
Sarasin A, Bounacer A, Lepage F, Schlumberger M & Suarez HG 1999 Mechanisms of mutagenesis in mammalian cells. Application to human thyroid tumours. Comptes Rendus de l'Académie des Sciences. Série III 322 143–149.
Singh NP, McCoy MT, Tice RR & Schneider EL 1988 A simple technique for quantitation of low levels of DNA damage in individual cells. Experimental Cell Research 175 184–191.
Song Y, Driessens N, Costa M, De Deken X, Detours V, Corvilain B, Maenhaut C, Miot F, van Sande J & Many M-C et al. 2007 Roles of hydrogen peroxide in thyroid physiology and disease. Journal of Clinical Endocrinology and Metabolism 92 3764–3773.
Stone JR & Yang S 2006 Hydrogen peroxide: a signaling messenger. Antioxidants & Redox Signaling 8 243–270.
Takahashi A & Ohnishi T 2005 Does γH2AX foci formation depend on the presence of DNA double strand breaks? Cancer Letters 229 171–179.
Vogt TM, Ziegler RG, Graubard BI, Swanson CA, Greenberg RS, Schoenberg JB, Swanson GM, Hayes RB & Mayne ST 2003 Serum selenium and risk of prostate cancer in U.S. blacks and whites. International Journal of Cancer 103 664–670.
Wang X, Phelan SA, Forsman-Semb K, Taylor EF, Petros C, Brown A, Lerner CP & Paigen B 2003 Mice with targeted mutation of peroxiredoxin 6 develop normally but are susceptible to oxidative stress. Journal of Biological Chemistry 278 25179–25190.
Wojewodzka M, Buraczewska I & Kruszewski M 2002 A modified neutral comet assay: elimination of lysis at high temperature and validation of the assay with anti-single-stranded DNA antibody. Mutation Research 518 9–20.
Yang T, Namba H, Hara T, Takmura N, Nagayama Y, Fukata S, Ishikawa N, Kuma K, Ito K & Yamashita S 1997 p53 induced by ionizing radiation mediates DNA end-jointing activity, but not apoptosis of thyroid cells. Oncogene 14 1511–1519.
