Effect of 3-aminobenzamide on DNA strand-break rejoining and cytotoxicity in CHO cells treated with hydrogen peroxide.

The influence of the nuclear ADP-ribosyltransferase inhibitor 3-aminobenzamide on the DNA strand-break rejoining kinetics and cytotoxicity in Chinese hamster ovary cells following H2O2 treatment was investigated. For the DNA damage studies, cells were treated on ice with H2O2 (0-20 microM) for 1 h in serum-free medium, after which the H2O2 was removed and the cells were allowed to repair their damage in complete medium at 37 degrees C in the presence or absence of 3-aminobenzamide (5 mM) for periods up to 2 h. The DNA strand breaks remaining as a function of time were then estimated by alkaline elution. A linear relationship between the H2O2 concentration and the initial level of DNA single-strand breaks (zero time allowed for repair) was observed. No double-strand breaks or DNA-protein cross-links were detected at these doses. The rejoining of single-strand breaks after H2O2 (20 microM) alone was characterized by a single exponential process with a t1/2 of approx. 5 min. However, in the presence of 3-aminobenzamide, rejoining was much slower and biphasic, with t1/2 of approx. 10 and 36 min. The inhibitory action of 3-aminobenzamide was concentration-dependent and completely reversible in that, when the 3-aminobenzamide was removed from the treated cultures, the strand-break rejoining kinetics rapidly returned to the t1/2 of 5 min typical of H2O2 alone. Considerably higher concentrations of H2O2 (up to 600 microM) were required for cell killing compared to the DNA damage studies. Cell killing by H2O2 alone was characterized by a shoulderless, exponential survival curve (D0 = 880 microM). The cytotoxicity was potentiated when the cells were treated with 3-aminobenzamide (5 mM) for 1 h after the H2O2 treatment; the survival curve with 3-aminobenzamide also assumed a biphasic character (D0 of 212 microM and 520 microM). These results are consistent with the theory that OH.-induced single-strand breaks do not normally represent lethal lesions to the cell because of their rapid, efficient repair. However, interference with these repair processes (in this case by 3-aminobenzamide) can alter this relationship, possibly allowing lesion fixation.

Evidence for separate mechanisms of cytotoxicity in mammalian cells treated with hydrogen peroxide in the absence or presence of L-histidine.

Treating Chinese hamster ovary cells with 1 mM L-histidine markedly increases their susceptibility to killing by H2O2. The mechanism of this effect has not been firmly established, although previous studies have shown that L-histidine in combination with H2O2, in contrast to H2O2 alone, generates DNA double-strand breaks (DSBs), albeit following supralethal concentrations of the oxidant. Using the highly sensitive pulsed field gel electrophoresis technique, we examined the ability of H2O2-L-histidine combinations to induce DSBs in cells over the same oxidant concentration range that causes cytotoxicity. Thus the correlation between DSB induction and cell killing could be investigated directly without the necessity for extrapolating effects across different concentration ranges. We used a number of treatment protocols that allowed the compartmentation of L-histidine inside or outside the cells, or both. Increased cytotoxicity was invariably associated with the appearance of DSBs, and both parameters were dependent on the intracellular fraction of the amino acid. A linear relationship was found between cytotoxicity and DSB formation when the cells were either treated with H2O2 (at > or = 20 microM) and L-histidine concurrently or were exposed to the oxidant following pre-loading with L-histidine. On the other hand, no DSBs were detected in cells treated with: (a) H2O2 alone; (b) L-histidine plus H2O2 at < or = 20 microM; or (c) H2O2 in association with both L-histidine and excess (20 mM) L-glutamine (which prevents L-histidine uptake). Thus, separate mechanisms appear to underlie the cytotoxic response in cells treated with H2O2 in the absence and presence of L-histidine, with the latter process being associated with the induction of DSBs and having a threshold at approximately 20 microM H2O2. The linear correlation between DSBs and cell killing observed in cells treated with H2O2-L-histidine at H2O2 concentrations > or = 20 microM was similar to (but not superimposable on) the correlation curve established for gamma-irradiated cells; DSBs produced by gamma-rays were associated with more cell killing than those generated by the H2O2-L-histidine combination.

Hydrogen peroxide-and fetal bovine serum-induced DNA synthesis in vascular smooth muscle cells: positive and negative regulation by protein kinase C isoforms.

Hydrogen peroxide and fetal bovine serum stimulate DNA synthesis in growth-arrested smooth muscle cells with remarkably similar kinetics and cell density dependence. However, while stimulation with fetal bovine serum results in cell proliferation, that by H2O2 is followed by cell death. Depletion of conventional and novel protein kinase C isoforms, resulting from a long treatment with phorbol-12-myristate-13-acetate, further increases H2O2-induced DNA synthesis. On the other hand, the specific protein kinase C inhibitor calphostin C abolished the increased DNA synthesis promoted by fetal bovine serum or H2O2. H2O2 increases protein kinase C activity in smooth muscle cells. This effect is markedly reduced, but not abolished, by down-regulation of the alpha, delta and epsilon protein kinase C isoforms. Thus, the zeta isoform of protein kinase C, which is not down-regulated, may be responsible for the residual H2O2 stimulation of protein kinase C. In conclusion, the results obtained show that H2O2 stimulates protein kinase C activity and DNA synthesis in growth-arrested smooth muscle cells: these events are not followed by cell proliferation but rather by cell death. This H2O2 stimulated DNA synthesis appears to be negatively controlled by alpha, delta and epsilon isoforms and positively controlled by the zeta isoform of protein kinase C.

Hydrogen peroxide insult in cultured mammalian cells: relationships between DNA single-strand breakage, poly(ADP-ribose) metabolism and cell killing.

We examined the effect of exposure to H2O2 at 37 degrees C on Chinese hamster ovary cell survival, DNA single-strand break (SSB) induction and rejoining, and activation of poly(ADP-ribose) (ADPR) polymerase. The effect of the ADPR polymerase inhibitor 3-aminobenzamide on each of these processes was also determined. SSB induction increased progressively with increasing H2O2 concentration. SSB levels were maximal after approx. 5 min of exposure to H2O2 (100 microM) and then decreased at longer times. This decrease, which paralleled the time-dependent depletion of H2O2, was due to the rejoining of SSBs. 3-Aminobenzamide enhanced the level of SSBs at each time point. H2O2 increased the level of both ADPR synthesis and NAD+ depletion (both measures of ADPR polymerase activity) in a concentration-dependent fashion, with the maximum effect being reached after approx. 20 min. After 100 microM H2O2, the effects on both ADPR and NAD+ were reversible. 3-Aminobenzamide completely blocked the effects of the oxidant on both NAD+ and ADPR levels. Thus, SSB induction by H2O2 at 37 degrees C was accompanied by a marked but reversible stimulation of ADPR polymerase. However, cell killing by H2O2 was only slightly enhanced in the presence of 3-aminobenzamide (5 mM), so the above-mentioned effects do not appear to be relevant to the cytotoxic effect of H2O2 under these conditions. Comparing these results with data obtained previously for cells treated with H2O2 at 4 degrees C suggests that the mechanisms of DNA strand breakage and cell killing may be quite different at the two temperatures, and that DNA damage at 37 degrees C may be indirectly mediated by temperature-dependent metabolic events.

Hydrogen peroxide generated at the level of mitochondria in response to peroxynitrite promotes U937 cell death via inhibition of the cytoprotective signalling mediated by cytosolic phospholipase A2.

We have studied the relationships existing between delayed formation of H2O2 and activation of cytosolic phospholipase A2 (cPLA2), events respectively promoting toxicity or survival in U937 cells exposed to peroxynitrite. The outcome of an array of different approaches using phospholipase A2 inhibitors, or cPLA2 antisense oligonucleotides, as well as specific respiratory chain inhibitors and respiration-deficient cells led to the demonstration that H2O2 does not mediate toxicity by producing direct molecular damage. Rather, the effects of H2O2 were found to be upstream to the arachidonic acid (AA)-mediated cytoprotective signalling and in fact causally linked to inhibition of cPLA2. Thus, it appears that U937 cells exposed to nontoxic concentrations of peroxynitrite are nevertheless committed to death, which however is normally prevented by the activation of parallel pathways resulting in cPLA2-dependent release of AA. A rapid necrotic response, however, takes place when high concentrations of peroxynitrite promote formation of H2O2 at levels impairing the cPLA2 cytoprotective signalling.

Peroxynitrite stimulates the activity of cytosolic phospholipase A2 in U937 cells: the extent of arachidonic acid formation regulates the balance between cell survival or death.

Peroxynitrite stimulates in U937 cells release of arachidonic acid (AA) sensitive to various phospholipase A(2) (PLA(2)) inhibitors, including arachidonyl trifluoromethyl ketone (AACOCF(3)), which specifically inhibits cytosolic PLA(2) (cPLA(2)). This response linearly increases using non toxic concentrations of the oxidant, and reaches a plateau at levels at which toxicity becomes apparent. Three separate lines of evidence are consistent with the notion that AA generated by cPLA(2) promotes survival in cells exposed to peroxynitrite. Firstly, toxicity was suppressed by nanomolar levels of exogenous AA, or by AA generated by the direct PLA(2) activator melittin. Secondly AACOCF(3), or other PLA(2) inhibitors, promoted cell death after exposure to otherwise non toxic concentrations of peroxynitrite; exogenous AA abolished the enhancing effects mediated by the PLA(2) inhibitors. Finally, U937 cells transfected with cPLA(2) antisense oligonucleotides were killed by concentrations of peroxynitrite that were non-toxic for cells transfected with nonsense oligonucleotides. This lethal response was insensitive to AACOCF(3) and prevented by exogenous AA.

The role of hydrogen peroxide and RRR-alpha-tocopherol in smooth muscle cell proliferation.

Oxidants can be considered early growth signals, since they have been shown to activate a number of pathways that are also stimulated by growth factors. In particular, H(2)O(2) activates the protein kinase C signal transduction pathway in smooth muscle cells. These events certainly play a role in the activation of the DNA synthesis machinery although it is still unclear whether they can also regulate the lethal response. Evidence exists of an oxidant-mediated increase in tyrosine protein phosphorylation as an early event in the signal transduction cascade of growth factor receptors, leading to augmentation of cell proliferation. Oxidants can also induce transcription of enzymes, such as ornithine decarboxylase and the phosphatase CL-100. CL-100 is the first example of a new class of protein phosphatases responsible for modulating the activation of MAP kinase following exposure of quiescent cells to growth factors and further implicates MAP kinase activation/deactivation in the cellular response to hydrogen peroxide. Moreover H(2)O(2) activates the MAP kinase cascade by stimulating the tyrosine kinase and protein kinase C pathways. JNK1, a relative of the MAP kinase group, is activated by dual phosphorylation at Thr and Tyr during the UV response. RRR-alpha-tocopherol and RRR-beta-tocopherol have different and competing effects on smooth muscle cell proliferation, indicating that they do not act as antioxidants. The earliest event brought by RRR-alpha-tocopherol in the signal transduction cascade contolling receptor mediated cell growth is the inhibition of the transcription factor AP-1, activated by phorbol esters. RRR-beta-tocopherol alone is without effect but in combination with RRR-alpha-tocopherol prevents the AP-1-inhibiting effect of the latter. Protein kinase C is inhibited by RRR-alpha-tocopherol and not by RRR-beta-tocopherol, which also in this case prevented the effect of RRR-alpha-tocopherol. The inhibition of RRR-alpha-tocopherol of protein kinase C is not the consequence of a direct interaction but is due to a diminution, produced by RRR-alpha-tocopherol of the kinase phosphorylation. A tocopherol binding protein appears to be at the basis of the RRR-alpha-tocopherol, that discriminates between RRR-alpha-tocopherol and RRR-beta-tocopherol and initiates a cascade of events at the level of cell signal transduction leading to cell proliferation inhibition.

Cyclic GMP-dependent inhibition of acid sphingomyelinase by nitric oxide: an early step in protection against apoptosis.

Activation of acid and neutral sphingomyelinases, and the ensuing generation of ceramide, contributes to the biological effects of tumour necrosis factor-alpha (TNF-alpha), one of which is apoptosis. While the mechanisms of activation of sphingomyelinases by the cytokine are being unravelled, less is known about regulation of their activity. Nitric oxide has previously been shown to exert a cyclic GMP-dependent inhibition of early apoptotic events triggered by TNF-alpha in the U937 monocytic cell line. We therefore investigated whether inhibition of sphingomyelinases by nitric oxide plays a role in regulating such early events. We found that activation of both acid and neutral sphingomyelinases, triggered in the first minutes after U937 cell stimulation with TNF-alpha, is regulated in an inhibitory fashion by nitric oxide, working through generation of cyclic GMP and activation of protein kinase G. Using a range of inhibitors selective for either sphingomyelinase we found that the acid sphingomyelinase contributes to activation of the initiator caspase-8 and early DNA fragmentation and that inhibition of the acid enzyme by nitric oxide accounts for cyclic GMP-dependent early protection from apoptosis. We also found that the protective effect by both cGMP and acid sphingomyelinase inhibitors progressively disappeared at later stages of the apoptotic process. Inhibition of sphingomyelinases represents a novel action of nitric oxide, which might be of physiological relevance in regulating initial phases of apoptosis as well as other biological actions of ceramide.

Osmotically driven radial diffusion of single-stranded DNA fragments on an agarose bed as a convenient measure of DNA strand scission.

The present study describes the development and characterization of a novel technique, the alkaline-halo assay, for the assessment of DNA single strand breakage in mammalian cells. This technique allows the measurement of DNA lesions at the single cell level and presents the additional advantages of being rapid, sensitive, virtually costless and environmentally friendly, because it does not require the use of isotopes. The alkaline halo assay involves a series of sequential steps in which the cells are first treated, then embedded in melted agarose and spread onto microscope slides that are incubated for 2 min at ice-bath temperature to allow complete geling. The slides are then incubated for 20 min in a high salt alkaline lysis solution, for an additional 15 min in a hypotonic alkaline solution and, finally, for 10 min in ethidium bromide. Under these conditions, single-stranded DNA fragments spread radially from the nuclear cage and generate a fluorescent image that resembles a halo concentric to the nucleus remnants. The area of the halos increased at increasing levels of DNA fragmentation and this process was associated with a progressive reduction of areas of the nuclear remnants. These events were conveniently monitored with a fluorescence microscope and quantified by image processing analysis. The sensitivity of the alkaline-halo assay, which is based on the osmotically driven radial diffusion of single-stranded DNA fragments through agarose pores, is remarkably similar to that of the widely used alkaline elution and comet assays.

Mitochondrial formation of hydrogen peroxide is causally linked to the antimycin A-mediated prevention of tert-butylhydroperoxide-induced U937 cell death.

Antimycin A and 2-heptyl-4-hydroxyquinoline N-oxide (HQNO), both of which bind to the same site of complex III, prevented U937 cell killing promoted by tert-butylhydroperoxide (tB-OOH). This cytoprotection was not directly caused by inhibition of electron transport or reduced formation of tB-OOH-derived toxic species, but rather appeared to be the consequence of a mechanism involving mitochondrial formation of hydrogen peroxide. Ubisemiquinone was most likely the electron donor allowing the formation of superoxides and, as a consequence, of hydrogen peroxide.

Direct excision of 50 kb pair DNA fragments from megabase-sized fragments produced during apoptotic cleavage of genomic DNA.

The DNA of U937 cells exposed to two different apoptotic stimuli, namely the cocktail H2O2/3-aminobenzamide (3AB) and etoposide, was analyzed using field inversion gel electrophoresis (FIGE) as well as programmable, autonomously controlled electrode electrophoresis (PACE). The results obtained indicate that FIGE is not appropriate for sizing apoptotic DNA fragments. PACE appears to be more accurate and reliable and the results obtained with this technique strongly suggest that the 50 kb DNA fragments are directly excised from Mb-sized DNA fragments without the intermediate cleavage of 200-300 kb products.

Rotenone and pyruvate prevent the tert-butylhydroperoxide-induced necrosis of U937 cells and allow them to proliferate.

Exposure of U937 cells to tert-butylhydroperoxide (tB-OOH) led to cyclosporin A-sensitive mitochondrial membrane permeability transition and necrosis. Pyruvate and rotenone, which increase mitochondrial NADH via different mechanisms, prevented these responses and the cells which received these treatments proliferated with kinetics similar to those observed in untreated cells. In contrast with these results, cells rescued by cyclosporin A were unable to proliferate. Thus, mitochondrial NADH plays a pivotal role in preventing upstream events which result in the onset of mitochondrial membrane permeability transition and death in cells exposed to tB-OOH. These events appear to be critical for recovery of the ability of the cells to proliferate.

Arachidonic acid induces calcium-dependent mitochondrial formation of species promoting strand scission of genomic DNA.

Both the phospholipase A(2) activator melittin and reagent arachidonic acid (AA) are poor inducers of DNA single strand breaks in U937 cells. These responses, however, were dramatically increased by the calcium-mobilizing agent caffeine (Cf) or by the respiratory substrate pyruvate via a mechanism that involved enforced mitochondrial Ca(2+) accumulation and that was sensitive to lipoxygenase inhibitors. In permeabilized cells, the DNA damage generated by AA in combination with either Cf, L-malate or CaCl(2) was blunted by catalase. AA generated DNA strand scission also in HeLa cells supplemented with pyruvate via a mechanism identical to that observed in U937 cells. This response was associated with an enforced formation of free radical species. These results demonstrate that mitochondria play a pivotal role in the DNA-damaging response evoked by AA and provide the bases for a calcium-dependent mechanism whereby the AA produced during inflammatory processes may affect various pathologic conditions, including carcinogenesis.

Chilling followed by incubation at 37 degrees C causes a reduction in NAD+ levels which can be prevented by the poly(ADP-ribose)transferase inhibitor 3-aminobenzamide.

The exposure of cells for 60 min to a serum free medium at ice temperature followed by a return to normal culture conditions (30 min at 37 degrees C) caused a dramatic decrease in NAD+ levels. This decrease in NAD+ was prevented by 3-aminobenzamide. Alkaline elution analysis of DNA from cultures that were sisters to the ones utilized for measuring cellular NAD+ content revealed an absence of DNA breakage. These data suggest that poly(ADP-ribose)transferase may be induced in conditions not involving DNA fragmentation. The induction of this enzyme could therefore represent a cellular emergency reaction and not just a response to DNA damage.

The L-histidine-mediated enhancement of hydrogen peroxide-induced cytotoxicity is a general response in cultured mammalian cell lines and is always associated with the formation of DNA double strand breaks.

Micromolar concentrations of L-histidine increase the cytotoxicity of hydrogen peroxide in a number of cell lines including CHO (hamster), EAHY, McCoy's, U937 and CCRF-CEM (human), Vero (monkey) and SC-1 (mouse). Importantly, these cell lines displayed different degrees of sensitivity to H2O2 alone and the extent of enhancement elicited by the amino acid was more pronounced in resistant cell lines. The increased cytotoxicity was invariably associated with the formation of DNA DSBs and a remarkable correlation was found by plotting the level of DNA DSBs against the cytotoxic response. These results strongly support the hypothesis that the mechanism whereby L-histidine increases the toxicity elicited by H2O2 involves the formation of DNA DSBs and are consistent with the possibility that the amino acid might participate in the regulation of the physio-pathological response to oxidative stress in mammals.

Quercetin prevents DNA single strand breakage and cytotoxicity caused by tert-butylhydroperoxide: free radical scavenging versus iron chelating mechanism.

Although the antioxidant properties of flavonoids are well documented, it is still unclear whether these effects are dependent on radical scavenging or iron chelating activities. By using an experimental approach based on the notion that iron chelators suppress DNA strand scission and cytotoxicity caused by tert-butylhydroperoxide, whereas radical scavenging antioxidants prevent only the latter response, we provide experimental evidence indicating that the most prominent activity of the flavonoid quercetin resides in its ability to chelate iron. This experimental approach can be utilized for the assessment of iron chelation in the biological activity of flavonoids or other antioxidants.

3-Aminobenzamide inhibition of protein kinase C at a cellular level.

3-Aminobenzamide, a known inhibitor of poly-(ADP-ribose)-polymerase has been found in the cell line U-937 to inhibit protein kinase C at the same concentration as poly-(ADP-ribose)-polymerase. 3-Aminobenzamide was not able, however, to inhibit the isolated enzyme. An indirect mechanism of protein kinase C inhibition is proposed.

The phosphorylation state of MAP-kinases modulates the cytotoxic response of smooth muscle cells to hydrogen peroxide.

Micromolar concentrations of hydrogen peroxide induced the phosphorylation of mitogen-activated protein (MAP) kinases and a lethal response in growth-arrested smooth muscle cells (A7r5). The H202-induced phosphorylation of MAP-kinases was markedly lower in the presence of protein tyrosine kinase (PTK) inhibitors or in protein kinase C (PKC) down-regulated cells. Similarly, the toxicity of H202 was diminished by concomitant addition of either PKC or PTK inhibitors and was also lower in PKC down-regulated cells. These results are consistent with the possibility that phosphorylation of MAP-kinases is a critical event in the toxic response of cultured smooth muscle cells to H202.

Prevention of necrosis and activation of apoptosis in oxidatively injured human myeloid leukemia U937 cells.

A 3 h exposure to 1 mM H2O2 followed by 6 h post-challenge growth in peroxide-free medium induces necrosis in U937 cells. Addition of the poly(ADP-ribose)polymerase inhibitor 3-aminobenzamide during recovery prevents necrosis and triggers apoptosis, as shown by the appearance of apoptotic bodies, extensive blebbing and formation of multimeric DNA fragments as well as 50 kb double stranded DNA fragments. Thus, the same initial damage can be a triggering event for both apoptotic and necrotic cell death. Furthermore, necrosis does not appear to be a passive response to overwhelming damage.

Products of the phospholipase A(2) pathway mediate the dihydrorhodamine fluorescence response evoked by endogenous and exogenous peroxynitrite in PC12 cells.

A short-term exposure of PC12 cells to tert-butylhydroperoxide promotes a rapid oxidation of dihydrorhodamine sensitive to nitric oxide synthase inhibitors and peroxynitrite scavengers. This response was not directly caused by peroxynitrite, but rather appeared to be mediated by peroxynitrite-dependent activation of phospholipase A(2). The following lines of evidence support this inference: (i) the peroxynitrite-dependent dihydrorhodamine fluorescence response was blunted by low concentrations of two structurally unrelated phospholipase A(2) inhibitors; (ii) under similar conditions, the phospholipase A(2) inhibitors prevented release of arachidonic acid; (iii) low levels of arachidonic acid restored the dihydrorhodamine fluorescence response in nitric oxide synthase- as well as phospholipase A(2)-inhibited cells; (iv) the dihydrorhodamine fluorescence response induced by authentic peroxynitrite was also blunted by phospholipase A(2) inhibitors and restored upon addition of reagent arachidonic acid. We conclude that endogenous, or exogenous, peroxynitrite does not directly oxidize dihydrorhodamine in intact cells. Rather, peroxynitrite appears to act as a signalling molecule promoting release of arachidonic acid, which in turn leads to formation of species causing the dihydrorhodamine fluorescence response.