Redox Regulation of Human OGG1 Activity in Response to Cellular Oxidative Stress^▿

Cell culture conditions.

The human Boleth lymphoblastoid cell line, kindly obtained from A. Schmitz (CEA), was established by the French Polymorphism Center (CEPH, Paris, France) after Epstein-Barr virus transformation of lymphocytes from a healthy donor. Cells were grown in suspension in RPMI 1640 supplemented with 15% heat-inactivated fetal bovine serum, 20 mM HEPES, 2.5 mM sodium pyruvate, 4 mM glutamine, and antibiotics (penicillin-streptomycin) (all provided by Gibco, Carlsbad, CA). Cells were maintained in a humidified incubator at 5% CO₂ at 37°C and fed with fresh medium at intervals of 48 h.

Cadmium treatment.

Cells were treated at a density of 1 × 10⁶ cells/ml with cadmium chloride in phosphate-buffered saline (PBS) for 30 min. They were washed twice in PBS and returned to the incubator in fresh medium. At different times during and after treatment they were washed in PBS by centrifugation and stored as dried pellets in liquid nitrogen until protein extraction.

Protein extraction.

Cell extracts were obtained by sonication of cell pellets in 20 mM Tris-HCl, pH 8; 250 mM NaCl; and a cocktail of aprotinin, antipain, and leupeptin (0.8 μg/ml each). The homogenate was centrifuged at 20,000 × g for 30 min at 4°C, and the supernatant was aliquoted and stored at −80°C for biochemical assays. Protein content was measured using a Bio-Rad assay kit (Bio-Rad Laboratories, Richmond, CA) with bovine serum albumin as a standard.

Intracellular cadmium determination.

At different times during and after metal exposure cells were washed twice in PBS-EDTA (2 mM) and lysed in lysis buffer (Promega) at 4°C. Samples were treated with ultrapure 65% nitric acid (Normaton quality grade; VWR Prolabo) and diluted in ultrapure water. Cadmium concentration was measured by inductively coupled plasma mass spectrometry (ICP-MS) using an X7 series quadrupole Thermo Elemental apparatus and related to the protein content. The ICP-MS apparatus was calibrated with a SPEX CertiPrep standard (Jobin Yvon, Longjumeau, France). Yttrium was used as an internal standard (1 ppb).

Determination of glutathione content.

An aliquot of the 20,000 × g supernatant from the cell lysate was treated with an equal volume of 10% trichloroacetic acid. After centrifugation at 15,000 × g for 15 min at 4°C, total glutathione content was measured on the supernatant by the Tietze recycling assay (1). Oxidized glutathione (GSSG) was measured by the same method after derivatization of reduced glutathione using 2-vinylpyridine.

Purification of human OGG1.

Glutathione S-transferase-tagged recombinant hOGG1 was expressed in Escherichia coli BL21. After induction, cells from 3 liters of culture (A₆₀₀ of 2.3) were harvested by centrifugation, resuspended, and sonicated in 120 ml of ice-cold lysis buffer (25 mM phosphate sodium, 10% glycerol, 1 mM EDTA, 500 mM NaCl, pH 7.6) with antipain (2.5 μg/ml), aprotinin (2.5 μg/ml), leupeptin (2.5 μg/ml), and lysozyme (1 mg/ml). The lysate was centrifuged at 100,000 × g for 30 min at 4°C, and the supernatant was incubated with 7 ml of glutathione Sepharose 4B (Amersham Biosciences) for 1 h at room temperature. The mixture was loaded on a column, and the resin was washed with 35 ml of lysis buffer and equilibrated with 35 ml of G0 buffer (25 mM phosphate sodium, 150 mM NaCl, pH 7.6). Twenty-seven milligrams of glutathione S-transferase-OGG1 was eluted in the same buffer with 30 mM glutathione. The protein was dialyzed against G0 and incubated with 250 U of thrombin (Amersham Biosciences) overnight at room temperature. The digestion mix was diluted three times with 25 mM sodium phosphate, pH 7.6, and loaded on a 1-ml column of resource S (Amersham Biosciences). Cleaved hOGG1 was eluted with a linear NaCl gradient in phosphate buffer. Nine milligrams of recovered protein was diluted four times with 25 mM sodium phosphate, pH 7.6, to 50 mM NaCl and loaded on a 1-ml Hitrap heparin column (Amersham Biosciences). hOGG1 was eluted with a linear NaCl gradient and collected at 400 mM NaCl. A 6.5-mg quantity of protein was obtained and conserved in 50% glycerol at −20°C.

8-OxoG DNA glycosylase assay.

A 34-mer oligonucleotide containing an 8-oxoguanine at position 16 was labeled at the 5′ end using [γ-³²P]ATP (3,000 Ci/mmol; Amersham Biosciences) and T4 polynucleotide kinase (New England Biolabs). The ³²P-labeled strand was hybridized to its complementary oligonucleotide containing a cytosine (C) opposite the lesion yielding the 8-oxo-G:C duplex. In a standard reaction mixture various amount of protein extracts or purified OGG1 (6-μl final volume) were added to a 9-μl reaction mixture containing 25 fmol of the 8-oxo-G:C-labeled duplex in 20 mM Tris-HCl, 1 mM EDTA, pH 6.8. The reaction mixtures were incubated at 37°C for 25 min. NaOH (0.1 N final concentration) was added, and the mixtures were further incubated for 10 min at 37°C and stopped with formamide dye, followed by heating for 5 min at 95°C. The products of the reaction were resolved by denaturing 20% polyacrylamide gel electrophoresis (19:1 acrylamide:bisacrylamide). Gels were scanned, and band intensities were quantified using a Storm PhosphorImager (Amersham Bioscience).

Western blot analysis.

Aliquots from cell extracts (50 μg) were denatured by being heated for 5 min at 95°C in either Laemmli buffer containing 100 mM dithiothreitol (DTT) (reducing Western blot assay) or Laemmli buffer without DTT (nonreducing Western blot assay). After denaturation samples were electrophoresed on a 12.5% sodium dodecyl sulfate-polyacrylamide gel and transferred to nitrocellulose membranes. Membranes were blocked for 1 h with 5% nonfat dry milk in PBS-T (PBS with 0.1% Tween 20) and incubated with rabbit polyclonal anti-human OGG1 (2) in 1% blocking reagent (Roche Diagnostic) for 2 h at room temperature. After three 15-min washes with PBS-T, the membrane was incubated with a horseradish peroxidase-conjugated anti-rabbit immunoglobulin G antibody at a 1/20,000 dilution at room temperature. After three additional washing steps, the protein-antibody complexes were visualized by ECL (Amersham Biosciences).

Analysis of hOGG1 redox conformation.

Purified hOGG1 was incubated for 30 min at 37°C with diamide or cadmium in 20 mM Tris-HCl (pH 6.8), precipitated by addition of 20% trichloroacetic acid (7.5% final concentration) for 15 min on ice, and centrifuged at 4°C at 16,000 × g for 20 min. After being washed with ice-cold acetone, the protein was incubated for 90 min at 30°C in 10 mM AMS (4-acetamido-4′-maleimidylstilbene-2,2′-disulfonic acid), denatured in Laemmli buffer without DTT, electrophoresed on a 15% sodium dodecyl sulfate-polyacrylamide gel, and revealed using the Imperial Protein staining reagent (Pierce).

Reversible inhibition of cellular 8-oxoG DNA glycosylase activity by cadmium.

To analyze the consequences of an acute exposure to cadmium on human cells, lymphoblastoid cells were incubated in the presence of 350 μM CdCl₂ for 30 min (10). At different times during the treatment and after removal of the Cd solution, cell extracts were analyzed for their 8-oxoG DNA glycosylase activity. Cd exposure led to a rapid and nearly total inhibition of the cellular 8-oxoG DNA glycosylase activity during the treatment and up to 15 min after it. Surprisingly, a complete recovery of the activity was observed starting 15 min after the cells were washed and put back on medium (Fig. 1).

Cadmium directly and irreversibly inhibits the DNA glycosylase activity of hOGG1 in vitro.

To establish whether cadmium can also directly inhibit the human OGG1 activity, as was shown for the mouse enzyme (48), we tested its effect on purified hOGG1 protein and on crude extracts from untreated Boleth cells. Our results confirmed those reported for the rodent protein by showing that cadmium inhibited 8-oxoG DNA glycosylase activity of purified OGG1 (Fig. 2A). However, doses as low as 5 μM had a potent inhibitory effect whereas concentrations higher than 100 mM were required for the mouse enzyme inactivation. For crude cellular extracts, doses above 50 μM resulted in inhibition of the enzyme (Fig. 2B). In all cases, adding an excess of the metal chelator EDTA simultaneously with Cd addition prevented the inhibitory effect of the metal (Fig. 2C). From these results, it could be suggested that the inhibition observed in cells exposed to Cd is due to a direct effect of the heavy metal on the hOGG1 protein. To test this hypothesis, we tried to reverse the cadmium-induced inactivation of hOGG1 obtained in either Cd-treated cells, Cd-exposed cell extracts, or Cd-exposed purified protein using EDTA. In neither case was EDTA able to revert the cadmium effect (Fig. 2C), thus indicating that in cells the reversible inhibition of OGG1 by cadmium is not simply due to a direct reversible interaction with the metal.

Cadmium accumulation, oxidative stress, and hOGG1 inhibition in cells.

In order to investigate the mechanism of cellular hOGG1 inhibition by Cd, we determined the intracellular concentration of the metal during and after treatment. As shown in Fig. 3A, Cd rapidly accumulated inside cells but was very efficiently eliminated from them after washing. Paralleling the kinetics of metal accumulation in and excretion from the cells, a sharp increase in oxidized glutathione was also observed, an indication of an intense oxidative stress (Fig. 3B). Dose-response curves using the same protocol of exposure indicate that hOGG1 DNA glycosylase activity was slightly inhibited with cadmium doses between 10 and 200 μM whereas complete inhibition was obtained with doses above 200 μM (Fig. 4A). This response correlates well with the induction of oxidative stress. Indeed, the fraction of oxidized glutathione progressively increased with cadmium doses, reaching a plateau at cadmium concentrations inducing complete inhibition of hOGG1 activity (Fig. 4B). Taken together, these results suggest the possibility of an oxidation mechanism for the inactivation of hOGG1 in Cd-treated cells rather than direct inactivation of the enzyme by the metal.

Alteration of hOGG1 activity in cadmium-treated cells is associated with changes in its redox state.

In Boleth cells cadmium exposure leads to a strong oxidative stress through inhibition of the pentose phosphate cycle enzymes and of glutathione reductase with a consequent alteration of the oxidized/reduced glutathione balance (A. Bravard et al., unpublished data). We therefore looked for a possible change in hOGG1 redox status associated with its inhibition and subsequent reactivation. Western blot analysis under nonreducing conditions of extracts from control cells or from cells treated with a noninhibitory Cd concentration (Fig. 5A, lanes 1 and 2) reveals three bands for hOGG1. In extracts from cells treated with the inhibitory concentration of Cd the upper hOGG1 band disappears at the expense of the faster-migrating form of the protein (Fig. 5A, lane 3). When the same extracts were loaded into the gel with a reducing buffer (Fig. 5A, lanes 5 to 7), the upper band was recovered in all extracts including the one from cells treated with 350 mM CdCl₂ with the corresponding loss of the faster-migrating band. Analysis of the purified protein under reducing conditions (lane 4) confirms that the upper band observed in the extracts corresponds to the reduced form of hOGG1 and indicates that the lower band enriched in the Cd-treated cells represents an oxidized form of the protein. The middle band either is a nonspecific band or reflects a nonredox modification of hOGG1 present in the extracts.

Figure 5B shows the analysis of the redox status of hOGG1 at different times in control cells or cells exposed to the inhibitory concentration of CdCl₂. At 15 and 30 min, when the DNA glycosylase activity is completely inhibited in the treated cells, the reduced form of hOGG1 cannot be detected (lanes 2 and 4), while it is present again at the 60-min point. This point corresponds to 30 min after the washing of the cells, when their redox status is normalized (Fig. 3) and the DNA glycosylase activity of hOGG1 is completely recovered (Fig. 1). The inactivation of the protein is paralleled by the enrichment of the lower band (Fig. 5B, lower panel, lanes 2 and 4) corresponding to the oxidized form of hOGG1. The oxidized nature of the lower band was confirmed by the presence of a single band of the same intensity when the samples were analyzed under reducing conditions (Fig. 5B, upper panel), ruling out the possibility of the lower band being a degradation product of the protein. Interestingly, incubating cells in 2 mM diamide [1,1′-azobis(N,N-dimethylformamide)], a thiol-oxidizing agent, also leads to the inhibition of the 8-oxoG DNA glycosylase activity correlated with a similar change in the migration pattern for hOGG1 (Fig. 5C). Taken together, these results suggest that the reversible inhibition of the hOGG1 protein activity induced in cells exposed to an acute cadmium treatment is mediated by reversible oxidation of the protein. Moreover, this inactivation of hOGG1 is not specific to the Cd treatment but rather a response to oxidative cellular stress.

Cysteine modifications in hOGG1 modulate its activity.

The above hypothesis implies that the 8-oxoG DNA glycosylase activity of hOGG1 can be modulated by oxidative alterations of the protein in response to the redox environment of the cell. In particular, cysteine residues, of which eight are present in hOGG1, are potential targets for oxidative modifications. We therefore examined whether OGG1 activity could be modulated by cysteine-modifying agents. The cysteine-blocking reagent N-ethylmaleimide (NEM) strongly inhibited the 8-oxoG DNA glycosylase activity of either purified hOGG1 or whole untreated cell extracts in a dose-dependent manner (Fig. 6A). Moreover, while hydrogen peroxide had no effect, other oxidants such as nitric oxide donors (data not shown) and diamide (Fig. 6A) were also able to inhibit the enzymatic activity of hOGG1 both in purified material and in cell extracts. In the case of diamide the inhibition of the extract activity was stronger than that of the pure protein. This might reflect the amplification of the oxidative response by secondary oxidative chain reactions involving other cellular components such as lipids. Interestingly, the migration pattern of purified OGG1 protein after oxidation by diamide was similar to that found in cadmium-treated cells (compare lanes 6 and 7 from Fig. 6B with Fig. 5). To test for the redox status of the thiol groups, the purified protein treated with either diamide or Cd was incubated in the presence of AMS before being loaded in the gel. AMS modifies proteins by the addition of a bulky adduct on reduced cysteine residues, causing a shift towards higher-molecular-weight forms. Figure 6B shows that exposure to diamide previous to AMS treatment blocks adduct formation, resulting in a protein with a lower molecular weight (lanes 2 to 4) compared to the completely reduced protein (lane 1). This confirms the oxidation of cysteine moieties by diamide. More important, Cd treatment of purified hOGG1 did not block the modification of cysteines by AMS (lane 5), implying that the metal does not directly oxidize the protein.

Since the hOGG1 inhibition in Cd-treated cells is reversible, we tested whether the enzyme inhibited by oxidation with diamide could be reactivated by reduction of the protein. As shown in Fig. 6C, diamide-induced inactivation of hOGG1 can be completely reversed by the reducing agent DTT. In the case of the Cd-inactivated protein shown in Fig. 6B, lane 5, treatment with DTT did not allow recovery of the activity (data not shown), supporting the notion of a distinct inactivation mechanism by direct exposure of the protein to the metal.