High levels of intracellular cysteine promote oxidative DNA damage by driving the fenton reaction

Abstract

Escherichia coli is generally resistant to H(2)O(2), with >75% of cells surviving a 3-min challenge with 2.5 mM H(2)O(2). However, when cells were cultured with poor sulfur sources and then exposed to cystine, they transiently exhibited a greatly increased susceptibility to H(2)O(2), with <1% surviving the challenge. Cell death was due to an unusually rapid rate of DNA damage, as indicated by their filamentation, a high rate of mutation among the survivors, and DNA lesions by a direct assay. Cell-permeable iron chelators eliminated sensitivity, indicating that intracellular free iron mediated the conversion of H(2)O(2) into a hydroxyl radical, the direct effector of DNA damage. The cystine treatment caused a temporary loss of cysteine homeostasis, with intracellular pools increasing about eightfold. In vitro analysis demonstrated that cysteine reduces ferric iron with exceptional speed. This action permits free iron to redox cycle rapidly in the presence of H(2)O(2), thereby augmenting the rate at which hydroxyl radicals are formed. During routine growth, cells maintain small cysteine pools, and cysteine is not a major contributor to DNA damage. Thus, the homeostatic control of cysteine levels is important in conferring resistance to oxidants. More generally, this study provides a new example of a situation in which the vulnerability of cells to oxidative DNA damage is strongly affected by their physiological state.

FIG. 1.

A cystine pulse transiently sensitizes cells to H₂O₂. (A) E. coli AB1157 (wild-type) cells were grown to log phase in minimal medium containing sulfate (circles) or cystine (squares). Cystine (0.5 mM) was added (solid symbols) or not added (open symbols) to the cells, and 3 min later, 2.5 mM H₂O₂ was added. At intervals, cells were diluted, and viability was determined by colony formation. (B) AB1157 cells were grown to log phase in medium containing sulfate. At time zero, cystine (0.5 mM) was added. At time points, aliquots were removed and challenged for 3 min with 2.5 mM H₂O₂.

FIG. 2.

Cystine-H₂O₂ treatment generates abundant DNA damage. Log-phase E. coli AB1157 cells were treated with either 0.5 mM cystine, 2.5 mM H₂O₂, or cystine and H₂O₂, or left untreated as a control (con). Total genomic DNA was isolated, and qPCR was performed, using equivalent amounts of template DNA for each reaction mixture. PCR products were scanned, and the relative fluorescence was normalized to the untreated control. Values are the means and standard deviations (error bars) from three experiments.

FIG. 3.

Cystine-H₂O₂ treatment accelerates mutagenesis. E. coli AB1157 cells were grown to log phase and exposed for 3 min to 0.5 mM cystine, 0.1 mM H₂O₂, or both cystine and H₂O₂ or left untreated as a control (con). After 3 min, catalase was added to scavenge H₂O₂, and cells were spread on LB plates containing thymine or thymine and TMP. Values are the means and standard deviations (error bars) from three experiments. Note that the scale is exponential.

FIG. 4.

Conditions that confer H₂O₂ sensitivity. Different E. coli strains and treatments were studied and are grouped in sets of bars. For bars 1 to 3, AB1157 cells were grown to the log phase and treated with iron chelators for 5 min (none [bar 1], 1 mM dipyridyl [bar 2], and 20 mM desferrioxamine [bar 3]) before the addition of 0.5 mM cystine and 2.5 mM H₂O₂. For bars 4 to 11, AB1157 cells were treated with a 0.5 mM concentration of an alternative sulfur species (cysteine [bar 4], homocystine [bar 5], sulfide [bar 6], thiosulfate [bar 7], sulfite [bar 8], GSH [bar 9], GSSG [bar 10], and DTT [bar 11],) instead of cystine. For bars 12 and 13, AB1157 cells were grown in minimal medium containing cystine until the cells reached early log phase, and then the cells were washed twice and suspended in minimal medium containing sulfate for 1 h. For bar 13, 20 μg of chloramphenicol per μl was present during the period of growth on sulfate. Cystine was then added, and the cells were challenged with H₂O₂ for 3 min. For bars 14 and 15, BW25113 and SP53 (cysB) cells were grown in minimal medium containing djenkolic acid and tested for cystine-mediated H₂O₂ sensitivity. JTG10 (gshA) cells (bar 16) and WP840 (gor) cells (bar 17) were grown in minimal medium containing sulfate and tested for cystine-mediated H₂O₂ sensitivity. Values are the means and standard deviations (error bars) from five samples.

FIG. 5.

Evidence that H₂O₂ sensitivity requires cystine uptake. E. coli AB1157 cells were grown in minimal medium containing sulfate until they reached log phase, and then they were exposed to different concentrations of d- or l-cystine for 3 min before 2.5 mM H₂O₂ was added. Viability was determined after 3 min of H₂O₂ exposure.

FIG. 6.

GSH-deficient cells have less DNA damage than wild-type cells. E. coli JTG10 cells (gshA) were treated with both cystine and H₂O₂. Total genomic DNA was isolated, and qPCR was performed using four different amounts of template DNA. Data shown are for cells before (stippled bars) and after (gray bars) cystine-H₂O₂ treatment. Compare with the GSH⁺ parent (Fig. 2, cystine+H₂O₂ bar).

FIG. 7.

Cystine treatment increases cysteine content. E. coli AB1157 (wild-type) (A) and JTG10 (gshA) (B) cells were grown in minimal medium containing sulfate until they reached log phase, and the cells were treated with 0.5 mM cystine for 3 min (thick line) or not treated (thin line). Acid-soluble thiols were isolated, labeled with mBBr fluorescent dye, and separated by HPLC. Cysteine and GSH peaks are indicated.

FIG. 8.

Cysteine reduces ferric iron better than GSH does. Ferric chloride (10 μM) was mixed in anaerobic 20 mM Tris-HCl (pH 7.4) with 3 mM cysteine or GSH. At each time point, aliquots were removed and ferrous iron was assayed.

FIG. 9.

Cysteine efficiently drives Fenton-mediated DNA damage in vitro. In an anaerobic Coy chamber, 33 ng of pACYC184 plasmid was mixed in 3.5 mM NaHCO₃ buffer (pH 7.2) with 10 μM FeCl₃, 20 μM GSH or cysteine, and 50 μM H₂O₂. At each time point, the reaction was stopped by adding catalase. The reaction mixture was run in a 1% agarose gel. RF, relaxed form; SF, supercoiled form.

FIG. 10.

Mechanism of cystine-mediated H₂O₂ sensitivity. The CysB protein, which is activated during growth on relatively poor sulfur sources, stimulates the concerted import and reduction of cystine. Reduced GSH is necessary for this process. When saturating cystine is provided to erstwhile sulfur-poor cells, intracellular cysteine pools transiently rise to supranormal levels. Free iron catalyzes electron transfer from excess cysteine to H₂O₂, generating hydroxyl radicals.