In cells undergoing oxidative stress, DNA damage may result from attack by .OH radicals produced by the Fenton reaction, and/or by nucleases activated by nuclear calcium. In the present study, the participation of these two mechanisms was investigated in HeLa cells. Nuclear-targeted aequorin was used for selectively monitoring Ca2+ concentrations within the nuclei ([Ca2+]n), in conjunction with the cell-permeant calcium chelator bis-(o-aminophenoxy)ethane-N,N,N', N'-tetraacetic acid acetoxymethyl ester (BAPTA/AM), the lipid-soluble broad-spectrum metal chelator with low affinity for Ca2+ and Mg2+ N,N,N',N'-tetrakis(2-pyridylmethyl)ethylenediamine (TPEN), and the high-affinity iron/copper chelator 1, 10-phenanthroline (PHE). In Ca2+-containing medium, H2O2 induced extensive DNA strand breaks and an increase in [Ca2+]n that was almost identical to that observed in the cytosol ([Ca2+]c). In cells bathed in Ca2+-free/EGTA medium, in which the increases in [Ca2+]n and [Ca2+]c due to H2O2 were significantly reduced, similar levels of DNA fragmentation also occurred. In cells preloaded with BAPTA/AM or TPEN, the small increase of [Ca2+]n normally elicited by H2O2 in Ca2+-free medium was completely buffered, and DNA damage was largely prevented. On the other hand, pretreatment with PHE did not affect the calcium response in the nuclei, but completely prevented DNA strand breakage induced by H2O2. Re-addition of 100 microM CuSO4 and 100 microM FeSO4 to TPEN- and PHE-treated cells prior to H2O2 challenge reversed the effect of TPEN and PHE, whereas 1 mM was necessary to negate the effect of BAPTA/AM. The levels of DNA strand breakage observed, however, did not correlate with the amounts of 8-hydroxy 2'-deoxyguanosine (8-OHdG): H2O2 did not produce 8-OHdG, whereas PHE alone slightly increased 8-OHdG levels. CuSO4 and FeSO4 enhanced the effects of PHE, particularly in the presence of H2O2. Exposure of cells to a mixture of CuSO4/FeSO4 also resulted in a significant increase in 8-OHdG levels, which was prevented in cells preloaded with BAPTA/AM. Similar results were obtained in a cell-free system using isolated calf thymus DNA exposed to CuSO4/FeSO4, regardless of whether H2O2 was present or not. These results suggest that BAPTA/AM prevents H2O2-induced DNA damage by acting as an iron/copper chelator. These data also indicate that caution must be exercised in using Ca2+ chelating agents as evidence for a role in cellular Ca2+ levels in experimental conditions in which transition-metal-ion-mediated oxidant production is also occurring.

The effects of oxidative stress on DNA damage and associated reactions, increased polyadenosine diphosphate-ribose polymerase (PARP) activity and decreased nicotinamide adenine dinucleotide (NAD) and adenosine triphosphate (ATP) contents, have been tested in primary cultures of porcine aortic endothelial cells. The cells were treated with 50-500 microM H2O2 for 20 min or 100 microM paraquat for 3 days or were exposed to 95% O2 for 2 and 5 days. The administration of 250-500 microM H2O2 resulted in a marked increase in PARP activity and a profound depletion of ATP and NAD. Although hyperoxia had no effect on PARP activity and reduced only slightly the ATP and NAD stores, it markedly reduced the ability of endothelial cells to increase PARP activity upon exposure to DNase. Paraquat had a similar effect. Human dermal fibroblasts were also exposed to 50-500 microM H2O2 for 20 min or 95% O2 for 5 days. Their response to H2O2 differed from that of endothelial cells by their ability to maintain the ATP content at a normal level. Fibroblasts were also insensitive to the effect of hyperoxia. These results suggest that the oxidant-related DNA damage is a function of the type of oxidative stress used and may be cell-specific.

Human endothelial cells exposed to H 2 0 2 showed reduced CREP DNA binding activity, enhanced HSF activation, and no induction of NFKB binding activity. Interestingly, H 2 0 2 was able to induce NFKB subunit p65 translocation in the nucleus. In contrast, cells exposed to TNFa showed enhanced CREP binding activity, activation of NFKB and no induction of HSE-HSF complex. Addition of H 2 0 2 , diamide and iodoacetic acid to the binding reaction mixture markedly reduced the DNA binding ability of the three transcription factors. Thus free sulfhydryls were important in DNA binding activity of CREP, NFKB and HSF, and the lack of induction of NFKB by H 2 0 2 in intact cells was likely caused by oxidation on a thiol, and not by a deficiency in the activation pathway.

The respective roles of H2O2 and .OH radicals was assessed from the protective effects of catalase and the iron chelator o-phenanthroline on 1) the inhibition of protein synthesis, and 2) DNA damage and the related events (activation of the DNA repairing enzyme poly(ADP)ribose polymerase with the associated depletion of NAD and ATP stores) in cultured endothelial cells exposed to the enzyme reaction hypoxanthine-xanthine oxidase (HX-XO) or pure H2O2. Catalase added in the extracellular phase completely prevented all of these oxidant-induced changes. O-phenanthroline afforded a complete protective effect against DNA strand breakage and the associated activation of the enzyme poly(ADP)ribose polymerase. By contrast, iron chelation was only partially effective in maintaining the cellular NAD and ATP contents, as well as the protein synthetic activity. In addition, the ATP depletion following oxidant injury was much more profound than NAD depletion. These results indicate that: 1) .OH radical was most likely the ultimate O2 species responsible for DNA damage and activation of poly(ADP)ribose polymerase; 2) both H2O2 and .OH radicals were involved in the other cytotoxic effects (inhibition of protein synthesis and reduction of NAD and ATP stores); and 3) NAD and ATP depletion did not result solely from activation of poly(ADP)ribose polymerase, but other mechanisms are likely to be involved. These observations are also compatible with the existence of a compartmentalized intracellular iron pool.

Pretreatment with NAC prior to Ad infection enhances virus entry, while treatment with NAC post-infection increases transgene transcription. This strategy permits the use of lower adenoviral loads and thus might be helpful for gene therapy of vascular diseases.